SECTION 12 Metabolic disorders

12.1 The inborn errors of metabolism General aspec

12.1 The inborn errors of metabolism: General aspects 1929

ESSENTIALS The inborn errors of metabolism are those inherited diseases in which the phenotype includes a characteristic constellation of bio- chemical abnormalities related to an alteration in the catalytic ac- tivity of a single specific enzyme, activator, or transport protein. Mechanism of diseases—​mutations in the proteins giving rise to the inborn errors of metabolism affect primary, secondary, tertiary, or quaternary structure. This can lead to an enormous variety of con- sequences, including (1)  abolishing, decreasing, or (occasionally) increasing protein activity; (2) affecting activator proteins, or binding of hormones and other ligands to cell surfaces or other structures; (3)  impeding intracellular trafficking and folding of proteins, as well as their post-​translational modification through, for example, glycosylation, phosphorylation, or prenylation; (4)  affecting the transport of metabolites across cellular membranes; and (5) affecting the formation, activation, or transport of key cofactors for enzymes and other proteins (e.g. vitamins or metal ions). Clinical presentation—​the manifestations of metabolic disease are protean and may seem nondescript, especially in adults, hence a high level of suspicion may be required to make a correct diagnosis. Inborn errors of metabolism usually come to light in the neonatal period or infancy, but they may arise even in mature adults for the first time, when their rate of progression may be indolent. Prevention and screening—​there is a strong case for mass popu- lation screening for some inborn errors of metabolism at the presymptomatic stage to allow early detection and introduction of proven treatment before irreversible damage occurs. Management—​definitive cure of the underlying abnormality is available for a few disorders, but precise characterization of the biochemical disturbance often permits rational treatment to be or- ganized and provides the basis for further therapeutic endeavours. General approaches include (1) restriction of a substrate that cannot be metabolized including molecules derived from the diet; (2) re- placement of a missing metabolic product; (3) removal of poisonous metabolites or rebalancing overproduction of toxic intermediates; (4) administering pharmacological doses of a cofactor, sometimes a vitamin, that may also stabilize a mutant enzyme; (5) replacement of a missing gene product, usually by enzymatic augmentation therapy or pharmacological chaperones, to prevent premature aggregation and denaturation; (6) repression of an overproduced protein or me- tabolite by stable RNA inhibition (e.g. in transthyretin-​induced amyl- oidosis or by supressing 5-​aminolaevulinic acid synthase in acute porphyria syndromes); (7) transplantation of cells (e.g. haematopoi- etic stem cells) or organs (e.g. liver) as a ‘gene replacement therapy’; and (8) activation of a poorly functioning protein (e.g. the successful use of an oral activator of a relatively uncommon mutant cystic fi- brosis transmembrane regulator protein such as the G551D missense variant in patients with the cognate disease). Introduction Definition and prevalence The inborn errors of metabolism are those inherited diseases in which the phenotype includes a characteristic constellation of biochemical abnormalities related to an alteration in the catalytic activity of a single specific enzyme, activator, or transport protein. About 1500 such disorders have been characterized with an estimated overall birth frequency of 1 in 1500 live births in nonconsanguineous popu- lations. While these are now recognized as belonging individually to the category of ‘rare diseases’, they can also be viewed as para- digmatic examples of the interplay between the constitutional and environmental aspects of disease as well as selective factors acting over the long course of human evolution. Historical perspective Inborn errors of metabolism were the brainchild of the physician Archibald Garrod in the early 20th century. In the century since their discovery, this brilliant concept has proved to have far-​reaching implications in biochemistry, genetics, evolutionary biology, and medical practice. William Bateson, a biologist and early champion of Mendel, showed Garrod that central to his idea of the ‘inborn’ was the concept of the gene. Garrod proposed that genetic deter- minants specify the activity of enzymes which catalyse particular metabolic reactions. Thus Garrod was the first physician and clin- ical biochemist who, with Bateson, applied Mendelian genetics to 12.1 The inborn errors of metabolism:
General aspects Timothy M. Cox and Richard W.E. Watts† † It is with great regret that we report that Richard W.E. Watts died on 11 February, 2018.

SECTION 12  Metabolic disorders 1930 humans: they came to understand the segregation of recessive traits in pedigrees affected by hereditary metabolic diseases and the fre- quent role of consanguinity in very rare disorders. Since Garrod’s original description, the term ‘inborn error’ has broadened to in- clude mutations affecting the function of other proteins, such as the structural proteins fibrillin and collagen as well as transport proteins in which mutations also cause disease. Garrod considered that some ‘simple’ biochemical traits inherited as Mendelian recessive characters, such as alkaptonuria, appeared to have little, if any, apparent effect on health (in fact, in this emblem- atic disorder, the focus of much of his early experimentation, the great man was mistaken: alkaptonuric subjects develop severe arth- ritis, renal and prostatic calculi, and often die prematurely from the consequences of cardiovascular disease (see Chapter 12.2)). In other conditions, such as albinism or porphyria, environmental factors (e.g. sunlight, barbiturates) cooperate with host determinants in the development of clinical manifestations. Thus Garrod promulgated the notion of ‘chemical individuality’ and genetic predisposition to disease; in so doing, he adduced a strong theoretical underpin- ning to the concept of ‘diathesis’—​a hitherto pervasive term of the 19th century, largely concealing prejudice and ignorance but long persistent in clinical thinking. Indeed, years after the publication of Garrod’s work, the great American geneticist Thomas Hunt Morgan stated in his Nobel lecture of 1934: I am aware, of course, of the ancient attempts to identify certain gross physical human types—​the bilious, the lymphatic, the ner- vous, and the sanguine dispositions, and of more modern attempts to classify human beings into the cerebral, respiratory, digestive and muscular, or, more briefly into asthenics and pycnics. Some of these are proposed to be more susceptible to certain ailments or diseases than are other types, which in turn have their own consti- tutional characteristics. These well-​intended efforts are, however, so far in advance of our genetic information that the geneticist may be excused if he refuses to discuss them seriously. In fact, by 1931 Garrod had developed his ideas in a prescient essay, Inborn Factors in Disease, which has prodigious implications for a modern synthesis of the concept of disease; he had advanced his logic from the inborn error to chemical individuality—​a universal quality of the whole species, as opposed to the single individual. It is clear that Garrod had in mind the operation of Darwinian prin- ciples: in the example of infectious disease, he refers to the individu- ality of the human individual and the microbe: In our fight against infective diseases we are not confronted with blind forces, acting at random, but with the disciplined offensive of highly trained foes. Whilst on the one hand the weapons of attack have been improved by evolution, there has been a corres- ponding evolution of protective mechanisms of great ingenuity, and of no small efficiency, for the defence of the individual attacked. To understand the diverse manifestations of disease, including those clearly due to infectious agents such as Mycobacterium tu- berculosis, and variable responses to drugs (many of which were metabolized by enzymes in the liver), Garrod further considered the idea of individual uniqueness and interactions with the envir- onment: he realized that an infinite multitude of responses to en- vironmental factors were determined by constitutional (genetic) variation in the individual and, in effect, that the operation of selection in human evolution is also played out within the micro- cosm of disease. While it has been pointed out that Garrod did not use the terms ‘multifactorial’ or ‘susceptibility’ and had a rudimentary under- standing of genetics, his ideas foresaw the concept of ‘complex’ dis- eases with their dynamic gene–​environment interactions. After the more accessible monogenic disorders, the immense technological power available for molecular analysis of the human genome is yielding extraordinary information of both therapeutic and diag- nostic significance. As to the complex diseases, it now seems that the discovery, characterization, and quantification of environmental factors and their interactions with human genetic variants is perhaps the greatest impediment to the mechanistic understanding of these disorders. To summarize: like many others before and after him, the inten- sive study of rare human phenotypes led Archibald Garrod to make observations of astonishing relevance to large fields of medicine. Perhaps the greatest, most penetrating—​and lasting—​insight to emerge from the concept of the ‘inborn’ has been the realization that disease can no longer be viewed solely in the context of the ‘broken machine’ metaphor, but rather is the consequence of interactions between individual uniqueness and an environment for which that individual is, at a given time, maladapted or ‘unfit’. Although the constraint of space prevents full consideration of this theme, the Darwinian perspective clearly has far-​reaching consequences for the teaching and practice of medicine, indeed a new field of evolu- tionary medicine emerges directly (see Chapter 2.2). Classifications of inherited diseases
of metabolism Almost all the inborn errors of metabolism arise from mutations in the nuclear genome and have Mendelian patterns of inheritance, but 13 genes are encoded by the mitochondrial genome, and when these are mutated the cognate diseases are maternally transmitted. Mutations in the proteins giving rise to the inborn errors of me- tabolism affect primary, secondary, tertiary, or quaternary structure. These can lead to an enormous variety of consequences, including (1) abolishing, decreasing, or (occasionally) increasing protein ac- tivity; (2) affecting activator proteins, or binding of hormones and other ligands to cell surfaces or other structures; (3) impeding intra- cellular trafficking and folding of proteins, as well as their post-​ translational modification through, for example, glycosylation, phosphorylation, or prenylation (a post-​translational modification in which a isoprenyl group is added to a cysteine residue—​a process which mediates protein interactions, especially protein–​membrane interactions); (4) affecting the transport of metabolites across cel- lular membranes; and (5)  affecting the formation, activation, or transport of key cofactors for enzymes and other proteins (e.g. vita- mins or metal ions). The complex genome The complexity of the human genome is becoming ever more ap- parent, with recent findings from a project called the Encyclopedia of DNA Elements (ENCODE) confirming that, contrary to previous supposition, most of the 3 billion base pairs that it contains have a function. Regulation of gene function is proving to be much more

12.1  The inborn errors of metabolism: General aspects 1931 intricate that at first thought. There are nearly 20 000 genes that en- code proteins, and DNA sequences also encode thousands of add- itional RNA molecules. Of the greater than 11 000 DNA sequences classified as pseudogenes, several are now known to be active in some cell types or individuals. We now know that genes can overlap and may have multiple start and termination points. ENCODE has uncovered 4 million short stretches of DNA that control gene ac- tivity, which can act combinatorially and in different cell types to give each a unique identity. Some of the RNA strands (transcribed as microRNAs) also influence gene expression. In a reference to the lexicon of human genetics (Mendelian Inheritance in Man), there are estimated to be 24  600 potential human phenotypes and 8670 known or suspected Mendelian dis- eases, the inheritance of which can be described as being auto- somal recessive, autosomal dominant, sex-​linked, or transmitted maternally through the mitochondrial genome (Online Mendelian Inheritance in Man, http://​www.ncbi.nlm.nih.gov/​Omim/​, http://​ www.omim.org/​statistics/​entry as of 27 June 2018). The inborn errors of metabolism are those inherited diseases in which the phenotype includes a characteristic constellation of chemical abnormalities related to an alteration in the catalytic activity of a single specific enzyme, activator, or transport pro- tein. There are unifactorially inherited diseases in which the cur- rent techniques are too insensitive for a chemical abnormality to be identified, so that the syndrome has to be defined in clinical, gross structural, and/​or pathological terms; further study is likely to demonstrate that many of these fall into the category of inborn errors of metabolism. Almost all the so-​called single-​gene diseases arise from muta- tions in the nuclear genome. A few mitochondrial proteins have their structures encoded in the mitochondrial DNA (mtDNA). This genetic information is transmitted only through the female line and maternal inheritance inborn errors of metabolism in- clude mitochondrial diseases. The nuclear and the maternally in- herited diseases stem from mutations of DNA which directs the synthesis of a specific polypeptide chain. The molecular changes in the enzyme protein may affect the primary, secondary, tertiary, or quaternary structure, decreasing, increasing, or abolishing its catalytic activity. Some mutations affect the function of an acti- vator protein, others reduce the binding of hormones and para- crine factors to cell surfaces and/​or subcellular structures, and some derange the migration of proteins within cells; another group impairs the transport of metabolites across cellular and subcellular membranes (Table 12.1.1). Most intracellular enzymes are lo- cated in the cytosol where they are correctly orientated in relation to one another, sometimes as macromolecular complexes, and to their substrates. Some are linked to cellular membranes and sev- eral are located in anatomically defined subcellular structures or organelles:  the Golgi apparatus, mitochondria, lysosomes, and peroxisomes. Mitochondrial diseases The mitochondrial genome is a circular double strand containing 16.5 kb of DNA. It encodes 13 of the polypeptide subunits of re- spiratory chain enzymes, the remainder of which (c.60) are encoded in the nuclear DNA. Hitherto, mutations in 26 genes in the mito- chondrial genome are associated with defined human phenotypes (see Chapter 24.19.5). Abnormal mitochondrial function impairs the supply of energy for biochemical work in all tissues and there- fore has wide-​ranging effects. Each mitochondrion also contains 24 RNA genes that participate in intramitochondrial protein synthesis. Transcription and translation of mtDNA are regulated by the nu- cleus through interactions with the noncoding D-​loop region of the mitochondrial genome. Human cells contain about 1000 copies of mtDNA, but the individual mitochondria in a cell may not all carry a given specific mutation and different cells carry different propor- tions of mutated mitochondria (heteroplasmy). The proportion of mutant mtDNA must exceed a critical level before the mitochon- drial respiratory chain disease declares itself. This variability, as well as tissue-​specific differences in dependence on oxidative metab- olism, explains, at least partially, why some tissues are preferentially affected in patients with mtDNA diseases. Table 12.1.1  Examples of diseases in which there is defective transport of an enzyme or metabolite within cells or across cell membranes Disease Metabolic abnormality Cystinuria Failure to transport cystine, lysine, ornithine, arginine, and homoarginine across the plasma membrane of the proximal renal tubular epithelium and the small intestinal mucosa I (inclusion body)-​cell disease (mucolipidosis II/​III) Failure to generate the key molecular recognition signal, mannose 6-​phosphate, on nascent lysosomal glycoproteins that allows them to bind and enter the organelle via mannose 6-​phosphate receptors. Multiple acid hydrolases are mistargeted and deficient in lysosomes, leading to build up of undegraded macromolecules (as inclusions). The disease is due to defective action of N-​acetylglucosamine-​1-​phosphotransferase—​a heteromeric complex of polypeptides that are the products of two genes Cystinosis (cystine storage disease, Lignac’s disease) Failure to transport cystine produced by intralysosomal proteolysis across the lysosomal membrane and into the cytosol Salla disease Failure to transport N-​acetylneuraminic acid (sialic acid) across the lysosomal membrane and into the cytosol Mucopolysaccharidoses Failure to degrade glycosaminoglycans (mucopolysaccharides), the undegraded mucopolysaccharides are neither transportable across lysosomal membranes nor capable of being removed from the lysosomes by exocytosis Tay–​Sachs disease Defective post-​translational processing of the α chain of β-​N-​acetylhexosaminidase (hexosaminidase A) in some mutants. This prevents the enzyme from migrating from the endoplasmic reticulum, where it is glycosylated, to the Golgi apparatus for phosphorylation of its mannosyl residues and hence to lysosomes and the exterior of the cell Primary hyperoxaluria type 1 (some cases) Mislocation of alanine: glyoxylate aminotransferase in mitochondria as opposed to its normal location in peroxisomes. This arises because a rare mutation (Gly 170 → Arg) is present simultaneously with the common polymorphism (Pro 11 → Leu). The mutation (Gly 170 → Arg) prevents dimerization of the molecule which, in turn, allows the weak mitochondrial targeting sequence generated by the polymorphism (Pro 11 → Leu) to direct the molecule to mitochondria instead of peroxisomes

SECTION 12  Metabolic disorders 1932 Postmitotic tissues (e.g. neurons, muscle, and endocrine tis- sues) have high levels of mutated mtDNA and are often clinically affected, whereas rapidly dividing tissues (e.g. bone marrow) tend to be less often clinically affected. Differences in the proportions of mutated and nonmutated mtDNA between and within family mem- bers also contribute to the wide phenotypic range encountered in the mitochondrial diseases. The spermatozoal cytoplasm, including its mitochondria, is entirely lost at fertilization and for this reason mitochondrial diseases are only transmitted through the female line. Clinically affected women rarely transmit a mtDNA deletion to their children. However, a woman with a heteroplasmic mtDNA point mutation or duplication may transmit a variable amount of mutated mtDNA to her progeny. The number of mtDNA molecules in each oocyte is reduced and then amplified to a total of about 105 during early development of the oocyte; this presumably random process contributes to the different amounts of mutated mtDNA in different children in the same family. Women whose gametes contain high concentrations of mtDNA are more likely to have clinically affected children than mothers with lower concentrations of mtDNA. The general clinical manifestations of the mitochondrial diseases are shown in Table 12.1.2 and specific examples of mitochondrial dis- eases are given in Box 12.1.1. Peroxisomal diseases Some enzymes and other proteins that are encoded in the nuclear DNA are specifically expressed in peroxisomes, to which they are imported soon after translation. Mutations in these genes result in the peroxisomal diseases listed in Box 12.1.2. Diseases due to de- fects in peroxisomal proteins are discussed in Chapters 12.9, 12.10, and 24.17. Lysosomal diseases Lysosomes are subcellular organelles containing hydrolases with low optimum pH values (‘acid hydrolases’) which catalyse the degrad- ation of cellular macromolecules. The macromolecules are either derived from the metabolic turnover of structural cellular compo- nents or have entered the cell by endocytosis. The products of this macromolecular degradation process leave the lysosomes by specific efflux processes. Table 12.1.2  The main clinical manifestations of diseases due to mitochondrial dysfunction Disease group Clinical manifestations Defects of fatty acid oxidation Hypoglycaemia Hepatic dysfunction Cardiac failure Myopathy Sudden infant death Respiratory chain disorders Lactic acidosis Encephalopathy Hypotonia Poor feeding Failure to thrive Convulsions Box 12.1.1  Some mitochondrial diseases Mitochondrial DNA defects Rearrangements (deletions and duplications) • Chronic progressive external ophthalmoplegia • Kearns–​Sayre syndrome (hypoparathyroidism with deafness) • Diabetes and deafness Point mutations in protein encoding genes • Leber’s hereditary optic neuropathy • Leber’s hereditary optic neuropathy/​dystonia • Neurogenic weakness, ataxia, and retinitis pigmentosa • Leigh’s syndromea Point mutations in tRNA genes • Mitochondrial encephalopathy with lactic acidosis and stroke-​like episodes • Myoclonic epilepsy with ragged-​red fibres • Myopathy • Cardiomyopathy • Diabetes and deafness • Encephalomyopathy • Leigh’s syndromea Point mutations in rRNA genes • Nonsyndromic sensorineural deafness • Aminoglycoside-​induced nonsyndromic deafness Nuclear DNA defects Nuclear genetic disorders with a mitochondrial basis • Friedreich’s ataxia (frataxin) • Autosomal recessive hereditary spastic paraplegia Nuclear genetic disorders of the mitochondrial respiratory chain • Leigh’s syndrome (complex I deficiency)a • Optic atrophy and ataxia • Leigh’s syndrome (complex IV deficiency)a Nuclear genetic disorders associated with multiple mtDNA deletions • Autosomal dominant external ophthalmoplegia • Mitochondrial neurogastrointestinal encephalomyopathy (thymidine phosphorylase deficiency) a An example of different mutations providing the same clinical syndrome (phenocopies). mtDNA, mitochondrial DNA; rRNA, ribosomal RNA; tRNA, transfer RNA. Data from Chinnery PF and Turnbull DM (1999) The Lancet 354 (supplement 1), S17–​S21. Box 12.1.2  Peroxisomal diseases • Zellweger’s syndrome (absent peroxisomal membranes) • Pseudo-​Zellweger’s syndrome • Adrenoleukodystrophy • Pseudo-​neonatal adrenoleukodystrophy • Acatalasia • Infantile Refsum’s disease • Refsum’s disease (classical form) • Hyperpipecolic acidaemia • X-​linked adrenoleukodystrophy • Chondrodysplasia punctatum rhizomelia • Primary hyperoxaluria type 1

12.1  The inborn errors of metabolism: General aspects 1933 In most of the lysosomal storage diseases, an inborn error of metabolism affects a specific lysosomal enzyme so that either undegraded or partially degraded macromolecules accumulate in the lysosomes (see Chapter 12.8). The engorged lysosomes distort the internal architecture of the cell, disturb its function, and in- hibit the activities of other lysosomal enzymes so that macromol- ecules other than those related to the primary enzyme deficiency also accumulate. Cystinosis (cystine storage disease) and Salla disease (N-​ ace­tylneuraminic (sialic) acid storage disease) are due to metabolic lesions involving the specific efflux processes by which these small molecules generated by the intralysosomal hydrolysis of complex substrates cystine and sialic acid, respectively, leave the lysosome (Table 12.1.1). Lysosomal enzymes are glycoproteins which are subject to exo- cytosis and reuptake by endocytosis. Their protein moieties are synthesized on the rough endoplasmic reticulum and the oligosac- charide side chains are added in the Golgi apparatus. The addition of a terminal mannose 6-​phosphate residue recognition marker is necessary if the enzyme molecule is to be correctly routed into the lysosomes, and if it is to be available for receptor-​mediated re- uptake from the interstitial fluids. The types of lysosomal storage diseases and the nature of their metabolic defects together with examples of each group are presented in Table 12.1.3. Heterogeneity in the inborn errors of metabolism The individual inborn errors of metabolism are defined on the basis of the phenotype, including the specific enzyme lesion, and by their pattern of inheritance. Close study of any particular in- born error of metabolism reveals unexpected heterogeneity and we are increasingly recognizing diverse patterns of inheritance due to a variety of mechanisms, including somatic mosaicism, dominant negative effects in complex multimeric pathways, as well as transcriptional silencing of imprinted genes. This may be due to: • multiple allelism • mutations at different gene loci affecting the structure of different polypeptide chains in a single enzyme protein • mutations at different gene loci affecting different proteins with similar catalytic functions • differences in the overall genetic background against which the single mutation acts • environmental factors. Table 12.1.3  Lysosomal storage diseases other than cystinosis and Salla disease; a more complete listing is given in Chapter 12.8
(Table 12.8.1) Name Defect Example Sphingolipidoses Failure to degrade compounds containing a sphingoid (sphingolipids, ceramides, sphingomyelins, and glycosphingolipids including the gangliosides (sialoglycosphingolipids)) Tay–​Sachs disease (GM2-​gangliosidosis) Gaucher disease (glucocerebrosidosis) Mucopolysaccharidoses Failure to degrade the glycosaminoglycans: dermatan, heparan, and keratan sulphates. Incompletely degraded glycosaminoglycan fragments accumulate in the lysosomes as well as extracellularly. This causes secondary deficiencies of other lysosomal enzymes and other undegraded macromolecules, particularly sphingolipids, accumulate Hurler’s disease Hunter’s disease Morquio disease (MPS 1, 2, 4) Glycoproteinoses A group of enzyme defects in the catabolism of glycoproteins in which characteristic abnormal macromolecules accumulate Fucisidosis Mannosidosis Acid lipase deficiency Two clinically distinct variants. Cholesteryl esters and triglycerides accumulate in most tissues due to deficiency of lysosomal acid lipase Wohlman’s disease Cholesteryl ester storage disease Glycogenosis II Lack of intralysosomal hydrolysis of glycogen Glycogenosis type II (only member of this group) Mucolipidoses Originally defined as being clinically intermediate between the sphingolipidoses and mucopolysaccharidoses but without mucopolysacchariduria (abnormal glycosaminoglycan excretion). Subsequently shown to include patients with (1) deficient neuraminidase activities with respect to either glycoprotein substrates (mucolipidosis I, also classified as a glycoproteinosis and termed sialidosis) or ganglioside substrates (mucolipidosis IV); (2) clinically mild and severe variants of uridine-​diphosphate-​N-​acetylglucosamine: lysosomal enzyme precursor N-​acetylglucosamine phosphate transferase (mucolipidoses II (I-​cell disease) and III (pseudo-​Hurler polydystrophy) respectively—​see Table 12.1.1) See text Miscellaneous Numerous defects affect the action or activation of lysosomal enzymes—​sulphatase-​modifying factor, which is critical for a post-​translational modification of a common cysteine residue in eight of 18 putative human sulphatases; the four sphingolipid activator proteins, A–​D, and GM2 activator protein, which act as enzymatic activators in multiple stages of lysosomal sphingolipid degradation, as well as lysosomal membrane digestion Deficiency of the lysosomal integral membrane protein, LIMP-​2, associated with selective deficiency of glucocerebrosidase in nonmacrophage lineage cells and action myoclonus–​renal failure syndrome. Mutations in the X-​chromosome-​linked LAMP-​2 lysosomal membrane protein cause Danon’s disease with defective clearance of autophagic debris, including glycogen, principally affecting heart, skeletal muscle, and the brain Multiple sulphatase deficiency GM2 gangliosidosis (AB variant) Action myoclonus–​renal failure Danon’s disease

SECTION 12  Metabolic disorders 1934 General approaches to the management of inborn errors of metabolism Clinical indicators of an inborn error of metabolism The manifestations of metabolic disease are protean and may seem nondescript, especially in adults. Some clinical settings suggest the presence of an inborn error of metabolism (Box 12.1.3), but in other circumstances a high level of suspicion may be required to make a correct diagnosis. Inborn errors of metabolism usually come to light in the neonatal period or infancy, but can occur at any time—​even in mature adults for the first time and in whom the rate of progression may be indolent. In an appropriate clinical context—​ for example, unexplained acute neonatal illness and/​or failure to thrive in early infancy, developmental slowing and arrest followed by retrogression, or unusual physiognomy—​the critical clue often comes from taking an appropriate family history, with specific in- quiries about affected siblings, possible parental consanguinity, pa- ternity, miscarriages, perinatal deaths, abortions, about the sexes of possibly affected relatives and their placement on the maternal or paternal side of the family, the ages at death of relatives, as well as the ethnic and geographical origins of the parents. Impaired func- tion of proteins that are localized to the mitochondria, lysosomes, and peroxisomes are associated with particular clinical and bio- chemical characteristics that reflect the compartmentalized func- tions of these organelles. A broad perspective of the physician’s role Inborn errors of metabolism encompass diseases that are disabling, disfiguring, and painfully life-​shortening, hence—​as in many areas of medicine—​the role of the physician is much broader than dis- pensing the perceived and much-​awaited ‘magic’ of a cure. From the clinical aspect, often the best achievable goals are preventing rapid deterioration and ensuring that emergency measures are in place so that acute metabolic decompensation can be quickly contained, with preservation of tissue integrity and organ function, especially of the brain. The time-​honoured adage: ‘to cure sometimes, to re- lieve often—​but comfort always’ is particularly telling in this field of practice. It is also important to emphasize the critical importance of prompt diagnosis—​even of what might prove to be an incurable disease—​together with the mandatory requirement for providing genetic advice. Principles of treatment While many inborn errors of metabolism have severe and poten- tially lethal effects, as a group, effective treatments are continually being introduced. Therapeutic advances are increasingly based on a scientific understanding of the inherited defect, to which bio­ chemical knowledge can be rationally applied. The treatments available for the individual disorders are diverse and often must be specially developed for individual patients. General approaches include (1)  restriction of a substrate that cannot be metabolized including molecules derived from the diet; (2)  replacement of a missing metabolic product; (3) removal of poisonous metabol- ites or rebalancing overproduction of toxic intermediates; (4) ad- ministering pharmacological doses of a cofactor, sometimes a vitamin, that may also stabilize a mutant enzyme; (5) replacement of a missing gene product, usually by enzymatic augmentation therapy or pharmacological chaperones, to prevent premature ­aggregation and denaturation; (6) repression of an overproduced protein or metabolite by stable RNA inhibition; (7) transplant- ation of cells (e.g. haematopoietic stem cells) or organs (e.g. liver) as a ‘gene replacement therapy’; and (8)  activation of a poorly ­functioning protein. The principles involved are summarized in Table 12.1.4. While dealing with the complexities of each individual disorder, the essence of clinical care for these chronic disorders requires maturity, vigilance, and the perspective of the generalist in paedi- atric or adult medicine to ensure provision of what is now termed holistic medicine. Palliative surgical and other measures may be needed to deal with specific complications (e.g. corneal grafting to restore vision in patients with corneal clouding due to one of the mucopolysaccharidoses). Consideration should also be given to meeting the educational and social needs of these patients as well as to optimizing their overall clinical state and correcting the biochem- ical parameters. Beyond the skills of other physicians, surgeons, biochemists, and geneticists, successful management of patients with inborn errors of metabolism requires multidisciplinary engagement by colleagues Box 12.1.3  Clinical presentations which, in the absence of acquired or other congenital causes, suggest an inborn error of metabolism • Unexplained acute neonatal illness and/​or failure to thrive in early in- fancy. (Marked muscle hypotonia, recurrent fits, comas, acidosis, and vomiting, especially if withholding milk feeds causes temporary im- provement, are especially suggestive) • Developmental slowing and arrest followed by retrogression • Developmental slowing and arrest leading to unexplained intellectual disability • Unusual physiognomy, multiple skeletal deformities with develop- mental delay and retrogression • Multiple skeletal deformities alone (dysostosis multiplex especially suggests a lysosomal storage disease) • Gross visceromegaly • Specific dietary intolerances • Haemolytic anaemia • Unusual body odoura • Urolithiasis • Cataracts in early lifeb • Dislocation of the optic lensc • Persistent jaundice and hepatic cirrhosis in infancy. • Abnormal cutaneous photosensitivity • Hypopigmentation • Abnormal drug sensitivity • A history of recurrent perinatal deaths and/​or stillbirths • Hydrops fetalis in the absence of blood group incompatibility between mother and fetus (red cell enzyme defects) a Examples are phenylketonuria (mousy, musty), branched chain ketoacidosis (maple syrup), methionine malabsorption (oast house, dry celery), isovaleric acidaemia (sweaty feet), trimethylaminuria (stale fish), multiple carboxylase deficiency (tom cat’s urine), and hawkinsinuria (swimming pool). b Examples are Fabry’s disease, galactosaemia, galactokinase deficiency, Lowe’s syndrome, mannosidosis, osteogenesis imperfecta, Refsum’s disease, and Wilson’s disease. c Examples are Ehlers–​Danlos syndrome, homocystinuria, hyperlysinuria, Marfan’s syndrome, and sulphite oxidase deficiency.

12.1  The inborn errors of metabolism: General aspects 1935 with special skills related to metabolic medicine—​including diet- itians, social workers, educationalists, and occupational therapists. It is particularly important to plan for the handover of specialist care from the paediatrician to the most appropriate adult physician when follow-​up in a paediatric department becomes inappropriate. The perfect outcome is to achieve a physically and mentally healthy and fulfilled adult who is capable of begetting healthy children. Unfortunately, the nature of many of the inborn errors militates against this ideal so treatment has to aim at optimizing the child’s potential in all its physical, mental, and social aspects. Treatment and support also have to be extended to the parents and siblings who, if not overtly affected themselves, may be carriers of the ab- normal gene concerned and require appropriate advice about the transmission of the disease to other offspring and other aspects of the condition. Particular treatments Protein replacement therapies The ability to clone human genes into bacteria and eukaryotic cells for ectopic expression which can then produce large amounts of the human gene product is opening the horizons for treatment by protein replacement or more accurately, augmentation. The devel- opment of macrophage-​targeted β-​glucosylceramidase enzyme re- placement therapy for Gaucher’s disease is a notable development in this field and is now regarded as the definitive treatment for the non-neuronopathic manifestations. Attempts to utilize transplanted fibroblasts and amniotic cells as a source for enzyme replacement therapy have not been successful. Bone marrow transplantation (‘haematopoietic stem cell transplant- ation’) has been used for the treatment of two groups of inherited Table 12.1.4  General approaches to the treatment of inborn errors of metabolism Method Examples General measures (directed to mitigate) Ultraviolet radiation (congenital erythropoietic and variegate porphyrias, and in albinism) Ionizing radiation in the DNA repair enzyme defects (xeroderma pigmentosum, ataxia telangiectasia) Infections (agammaglobinaemia). Restriction of a dietary substrate which cannot be metabolized Partial inhibition of formation of toxic metabolites (inhibitors of biosynthesis) Phenylalanine restriction in phenylketonuria Protein restriction in the hyperammonaemias Elimination of galactose in galactosaemia Restriction of dietary phytanic acid and congeners (Refsum’s disease) Medications (oestrogens, barbiturates, etc. in acute intermittent porphyria) statins (hypercholesterolaemia); haem arginate (acute porphyrias); nitisinone (hereditary tyrosinaemia type I); allopurinol (gout); eliglustat (type I Gaucher disease) Supplying a missing metabolic product Orotic aciduria: treatment with uridine triacetate which is metabolized to uridylic acid Hartnup disease: nicotinic acid to control pellagra Removal of toxic metabolite Haemodialysis and peritoneal dialysis as temporary treatment of an acute metabolic crisis due to a diffusible toxic metabolite, and to correct certain secondary biochemical abnormalities quickly Either specific chemical detoxication (e.g. penicillamine in Wilson’s disease) or solubilization (e.g. penicillamine in cystinuria); accelerated metabolic disposal (phenylbutyrate and sodium benzoate in hyperammonaemia due to urea cycle defects) Pharmacological doses of a cofactor (only some cases of each disease respond). Note: this stratagem has features of pharmacological chaperone therapy since natural cofactors such as vitamins also stabilize mutant enzymes Propionic acidaemia: biotin Ubidecarenone (respiratory chain disorders due to coenzyme Q10 deficiency) Homocystinuria: pyridoxine Primary hyperoxaluria (type I): pyridoxine Methylmalonic acidaemia: vitamin B12 and other cobalamins Replacement of a missing gene product Adenosine deaminase deficiency Gaucher disease: β-​glucosylceramidase (mannose-​terminated to confer selective uptake by tissue macrophages) Haemophilia: clotting factor VIII Activating a membrane transporter Cystic fibrosis (due to G551D mutation—​a relatively uncommon variant of the cystic fibrosis transmembrane regulator protein, CFTR, a chloride channel) Bone marrow transplantation Adenosine deaminase deficiency; infantile Krabbe’s disease; α-​mannosidosis Haematopoietic stem cell transplantation Adenosine deaminase deficiency Liver transplantation Hereditary tyrosinaemia (type I) Antitrypsin deficiency Primary hyperoxaluria (type I) Urea cycle disorders Crigler–​Najjar syndrome (type I) Gene replacement RNA interference therapy (siRNA) Adrenoleukodystrophy adenosine deaminase deficiency Patisiran for transthyretin-​related polyneuropathy (systemic amyloidosis) Note: the examples chosen are situations in which either the proposed treatment is established or in which it can be recommended as elective therapy even though the results of prolonged evaluation are still awaited.

SECTION 12  Metabolic disorders 1936 metabolic disorders: those in which it is desired to replace a par- ticular type of nonfunctioning bone marrow cell by its normally functioning counterpart, and those in which an attempt has been made to utilize the fact that the bone marrow produces leucocytes and that these cells exocytose (release) their lysosomal enzymes for endocytic uptake by enzyme-​deficient cells in the body tissues generally. Bone marrow transplantation has been more successful with the first group of diseases, which includes disorders of neutrophil func- tion (e.g. cyclic neutropenia), functional abnormalities of lympho- cytes, and osteopetrosis; the beneficial effect on the latter being due to the introduction of normal osteoclast precursors derived from granulocyte–​macrophage progenitors in the marrow. The results of bone marrow transplantation in the second group of diseases, namely those in which the white cell lineage derived from the transplanted bone marrow is used to supply normal en- zyme to enzyme-​deficient tissues, for example, Hurler’s disease (mucopolysaccharidosis type 1) and Krabbe’s disease, has been less effective. In the latter case, recent courageous studies have shown that transplantation of HLA-​matched umbilical cord blood in the first 10 days of life has been moderately successful in promoting neurological development in infants born to couples with previously affected offspring and detected by screening early for this otherwise rapidly progressive neurodegenerative disorder. Haematopoietic stem cells have been implanted into the fetus in utero to correct se- vere congenital immunodeficiency, but this has not, so far, been ap- plied to diseases without immunodeficiency. This procedure takes advantage of the immunological tolerance of the fetus. The possibility of using liposomes and resealed erythrocyte en- velopes as carriers of therapeutic enzymes is also being explored; linking purified or recombinant therapeutic enzymes such as adeno- sine deaminase to polymers such as polyethylene glycol may usefully prolong their survival in circulating plasma. Receptor-​targeted therapies Definitive enzymatic augmentation with receptor-​targeted therapies has attracted much attention. In Gaucher’s disease, this strategy has proved to be very effective and commercially successful: global sales of the mannose-​terminated glucocerebrosidase for about 6000 patients worldwide enabled the Genzyme corporation to rise to a leading pos- ition in the biotechnology industry. Other industrial competitors have followed suit with targeted enzyme preparations approved for this and other lysosomal diseases such as Fabry’s disease, Pompe’s disease, and mucopolysaccharidosis types I and II (see Chapter 12.8). Substrate-​reduction therapies Substrate-​reduction therapy with the use of specific inhibitors to regulate the flux through impaired degradative pathways, by partial blockade of the rate-​limiting step is useful in low-​density lipoprotein receptor deficiency (heterozygous familial hypercholesterolaemia as well as the very rare homozygous variant), and thus was born the pharmaceutical star of the statin drugs, which are in wide general use. An analogous approach involving inhibition of the first committed step in the biosynthesis of glycosphingolipids is already showing promise in the glycosphingolipid diseases such as Gaucher’s dis- ease. Small-​molecule inhibitors of this pathway that are safe, well tolerated, and that penetrate the blood–​brain barrier have the po- tential to improve the outcome for many patients with progressive neurological complications of the sphingolipidoses, who would otherwise be without hope. Currently two such agents are in active clinical development. In alkaptonuria, a disease in which Garrod maintained a lifelong interest, the use of substrate-​reduction therapy is also far advanced. Nitisinone, a triketone inhibitor of the precursor to homogentisic acid at the level of hydroxyphenylpyruvate dioxygenase in the tyro- sine degradation pathway, is a licensed agent for tyrosinaemia type 1. In very small doses, this agent has a striking effect on the forma- tion of toxic oxidative metabolites of homogentisic acid which lead to the life-​shortening manifestations of alkaptonuria, and it appears likely that at last a well-​tolerated and definitive treatment for this landmark disorder is within sight. Pharmacological chaperones The concept of pharmacological chaperones, based on the ability of small molecules to bind to mutant proteins to prevent their inactiva- tion by abnormal folding, intracellular aggregation, and mistargeting, is receiving much attention. It has yet to be adopted extensively in practice, although the chaperone approach is in late-​phase clinical de- velopment in Fabry’s disease and is being explored in Pompe disease. Organ transplantation Liver transplantation is used as a form of functional complementa- tion in some inborn errors of metabolism such as glycogen storage disease type I and severe recurrent acute porphyria where this organ is the specific site of the metabolic lesion. Liver transplantation has the advantage that the enzyme is introduced in the correct organ, in the correct cell with its correct subcellular location, and correctly orientated with respect to its substrate and other enzymes with which it must act in concert, for example, in the urea cycle disorders such as ornithine transcarbamylase deficiency. Liver transplantation can also be regarded as a form of gene re- placement therapy in that the donor liver contains the normal gene which will direct the synthesis of a normal enzyme protein. Prenatal transplantation of fetal liver stem cells has potential in the treatment of some inborn errors of metabolism. Successful engraftment at the 12th to 24th week after fertilization with partial correction of the metabolic defect has been demonstrated in β-​thalassaemia. Gene replacement and cell-​based therapies Gene therapy and cell-​based therapies including bone marrow transplantation are at various stages of clinical evaluation and devel- opment, assisted by the recent capacity to develop credible models of specific disorders in genetically modified animals. Haematopoietic stem cell gene therapy Gene replacement using retroviral vectors and gene constructs can be used to introduce the desired DNA sequence into the patient’s explanted haematopoietic stem cell genome. These genetically cor- rected cells are cultured and then returned to the patient’s circula- tion, where they may have therapeutic potential in diseases where expression of the metabolic lesion in the haematopoietic system determines the phenotype or in those situations where genetic- ally corrected migratory cells of haematopoietic origin can deliver normal enzyme to the enzyme-​deficient tissues. This approach has recently been reported in young patients with metachromatic leukodystrophy, with the phase I/​II clinical trial results indicating

12.1  The inborn errors of metabolism: General aspects 1937 convincing evidence of some neurological benefit or ‘rescue’ com- pared with historical and family control patients not so treated, and retrospectively with patients treated by transplantation of haem- atopoietic stem cells from healthy donors. Despite the difficulty in determining efficacy directly in such studies, the disease did not manifest or progress in the three patients 7 to 21 months beyond the age at which this would have been predicted. Somatic cell gene therapy Although somatic cell gene therapy using viral vectors and/​or gene constructs to introduce the desired DNA sequences into other cell types is currently being investigated extensively in in vitro model systems and in animal models of some human inborn errors of me- tabolism (e.g. using hepatocytes), few of these have reached appli- cation in clinical practice. However, this approach, using lentiviral vectors which have the advantage that they can be used to transduce by nuclear integration of viral sequences in mitotic cells, has had qualified therapeutic success in trials in children with combined immunodeficiency. Several patients in this Anglo-​French trial un- fortunately developed a late-​onset T-​cell lymphocytosis leading to leukaemia, later shown to be related to the integration of vector sequences at a genomic ‘hot spot’ leading to activation of a neigh- bouring endogenous proto-​oncogene. These patients responded to antileukaemic chemotherapy with satisfactory control of the complication—​ with continued amelioration of their disabling im- munodeficiency disease—​but safety considerations have retarded clinical development until improved vector systems can be utilized. Promising results of a gene therapy trial using a lentiviral vector to correct the enzymatic abnormality in leucocytes derived from haematopoietic precursors in a very rare immunodeficiency disease, adenosine deaminase deficiency, have also been reported—​so far with no mutagenic effects. Eight years after the procedure, eight of ten patients with severe combined immunodeficiency no longer re- quired enzyme-​replacement therapy and lived normally. Gene therapy with autologous CD34+ haematopoietic stem cells transduced with a third generation lentiviral vector reduced or elimin- ated the need for long-term red-cell transfusions in 22 patients with se- vere β-thalassemia without serious adverse events related to this vector. Given the challenges of donor availability and the added risks of allo- geneic haematopoietic stem-cell transplantation in this disease and the prior occurrence of neoplastic change or of preferential integration of vector at specific sites in the host genome, early regulatory approval of this stratagem, is encouraging for numerous inborn metabolic diseases. The possibility of using adeno-​associated viral vectors as a means of introducing corrected genes for into nonmitotic cells of the nervous system is being explored in human patients; these vectors are main- tained as episomal elements which do not integrate readily into the host genome (with the attendant risk of mutagenesis) but persistently express the corrective protein. Adeno-​associated vectors have been successfully used in early gene therapy trials of the retinal disease, Leber’s congenital amaurosis, with direct intraorbital gene delivery. Recombinant adeno-associated viral vector serotype 9, which supplies the deficient Smn1 protein, was approved by the FDA as Zolgensma after a phase 3 clinical trial in infants less than 2 years of age with spinal muscular atrophy—including the pre-symptomatic phase. The agent prolongs event-free survival and increases motor function. Salutary outcomes in rapidly progressive genetic diseases open up the field for further exploration of gene therapy in the human brain. The unique capacity for complementation of soluble lysosomal proteins to be secreted by cells and taken up at a distance by others (‘secretion–​recapture’) renders those lysosomal diseases in which neurological manifestations are prominent as excellent targets for clinical exploration of gene therapy (see Chapter 12.8). Here, the principle of allowing a proportion of neural cells to be stably trans- duced by vector, thus to serve as a source of a given corrective pro- tein that can be taken up into the lysosomes that lack the enzyme in nearby neurons, is an attractive strategy for therapeutic exploration. Although there are some prospects of correcting some enzyme defects in the somatic cell genome, the correction of defects in the germline seems remote, although the development of advanced in vitro fertilization techniques, preimplantation DNA analysis, gene transfer, insertion or conversion, and embryo implantation pro- cedures may render this possible. However, the prospect of human germline modification will arouse many complex ethical issues, and these may hold up research and clinical application. Screening for inborn errors of metabolism The realization that very early diagnosis is essential in order to achieve good results in the treatment of many inborn errors of metabolism, such as phenylketonuria and galactosaemia, has stimulated interest in the possibility of examining either whole populations or selected groups of predisposed individuals for the biochemical differences which characterize particular inherited metabolic diseases. Diagnosis is needed at a stage which is not only presymptomatic but which pre- cedes the onset of self-​perpetuating secondary pathological changes. Screening for inborn errors of metabolism may be either non- selective (whole population) or selective. The latter, which includes carrier detection studies, aims to cover a part of the population. This may be defined on clinical, genetic, ethnic, or geographical grounds. Phenylketonuria and congenital hypothyroidism are the only metabolic disorders for which neonatal whole-​population screening is generally practised, although medium chain fatty acid dehydro- genase deficiency has been included recently in the United Kingdom and many other countries. There is wide international variation and galactosaemia, cystic fibrosis, and congenital adrenal hyper- plasia (21-​hydroxylase deficiency) have been proposed. Until recently it has been held that whole-​population screening should only be established for treatable or preventable diseases, and the con- sistency of the association of the proposed biochemical or other marker and the serious clinical phenotype must have been proved beyond any doubt. There must be a reliable and robust analytical method suitable for use with a sample of blood or urine which can be obtained without distressing either the parents or the baby. The possibility that metabolic screening will bring to light previously unrecognized variants, which are either mild and do not require treatment, or which by virtue of a fundamentally different bio- chemical lesion will resist the currently established therapies, has to be borne in mind. Phenylketonuria illustrates these difficulties. Here, beside classical phenylketonuria, whole-​population screening has identified both the clinically unimportant essential (mild) hyperphenylalaninaemia, and the devastatingly serious, but treat- able, inborn errors of tetrahydrobiopterin synthesis which produce the ‘malignant’ hyperphenylalaninaemia syndrome. In a subset of patients with classical phenylketonuria, tetrahydrobiopterin also

SECTION 12  Metabolic disorders 1938 improves blood phenylalanine control, and may, in the long term, allow the burden of stringent dietary treatment to be relaxed. It is also possible that in some cases immediate postnatal screening and treatment may be too late to prevent minor manifestations of the disease (e.g. in congenital hypothyroidism). The incidence of disease which merits whole-​population screening should be at least similar to that of phenylketonuria in white Europeans (between 1 in 6000 and 1 in 14 000). Cystic fibrosis has a birth frequency of approximately 1 in 2500 (heterozygous car- rier frequency l in 25) in white persons of European ancestry and would merit neonatal whole-​population screening on this basis. Molecular genetic approaches are potentially useful. If the dis- ease is not too genetically heterogeneous, and when the full range of possible causative mutations is known, the specific mutation could be sought directly. Some individuals classified as being homozy- gotes on the basis of classical genetic analysis prove to be compound (‘double’) heterozygotes, that is, they carry two different mutations, each affecting either the maternal or paternal allele of the same gene. The number of inborn metabolic errors in which the affected indi- viduals and the heterozygous carriers can be identified by molecular analysis of genomic DNA is increasing rapidly. The conditions which are identified by the application of DNA analytical methods include such numerically important diseases as sickle cell anaemia, β-​thalassaemia, haemophilia, Duchenne muscular dystrophy, cystic fibrosis, medium-​chain acyl-​CoA dehydrogenase deficiency, and phenylketonuria, as well as rarer but devastating conditions such as the Lesch–​Nyhan syndrome. Prenatal diagnosis The procedures used in prenatal diagnosis are: • direct examination of the fetus by ultrasonography and fetoscopy • chemical analysis of amniotic fluid • biochemical and cytological analysis of cultured amniotic cells (amniocytes) obtained by amniocentesis at weeks 15 to 16 of pregnancy • DNA analysis on uncultured amniocytes • karyotypic enzymological and DNA analysis of chorionic villi obtained by biopsy at weeks 10 to 12 of pregnancy • biochemical studies on tissue obtained by fetal biopsy in utero • sequencing of circulating cell-​free DNA (cfDNA) during pregnancy. Carrier state diagnosis Carriers are either individuals carrying the gene for a recessive dis- order, which does not express itself in the heterozygous state (e.g. phenylketonuria), or those who carry the gene for a dominant dis- order, that is, one which does express itself in the heterozygous state, but in which symptoms occur in later life (e.g. Huntington’s disease). The general approaches to carrier state diagnosis are: • detection of minor clinical, radiological, and clinicopathological abnormalities • demonstration of levels of enzyme activity in tissue (e.g. leuco- cytes or cultured fibroblasts) which are intermediate between those observed in individuals homozygous for the abnormal and the normal forms of the enzyme respectively (the observed level of activity may not be exactly 50% of the normal value) • demonstration of intermediate levels of a characteristic metabolite in an accessible body fluid • demonstration of mosaicism with respect to the product of the mutant gene on the X chromosome in the case of sex-​linked reces- sive disorders • direct gene analysis using either a specific gene probe or a linked restriction fragment length polymorphism. The ability to recognize asymptomatic carriers of serious recessive diseases and presymptomatic individuals in the case of dominant disorders raises major ethical and social issues with respect to the psychological impact that this information will have on the affected individuals and their families. This is especially so with the clinically normal carriers of a crippling, lethal, and untreatable disease such as Huntington’s disease. Applications of genome sequencing Recent years have seen astonishing improvements in the accuracy, extent, and rate at which genetic information can be accessed by sequencing of human DNA. In the four decades since robust methods for determining the nucleotide base order in natural DNA molecules were first reported, the growth of sequence information has been astronomical as a result of ‘next-generation sequencing’ methods. Meanwhile, the desire to generate and store these data has also burgeoned. The simultaneous explosion in bioinformatics and statistical genetics combined with the need for clinical prognostica- tion is driving a revolution. Analysis dependent on massive parallel sequencing has intensified knowledge of the pathological anatomy of the human genome; research is ongoing to conduct simulations and molecular modelling and incorporate empirical structural biology to realize a key scientific ambition with the aim of improving the capacity to predict the clinical consequences of genomic variants. DNA sequencing now extends far beyond the domain of laboratory science and evolutionary studies: in biology it has proliferated cat- egorically and dominantly for clinical use. Rapid, cheap, and increas- ingly reliable, DNA sequencing applications in medicine are enabling predictive testing in certain circumstances, for example, highly pene- trant Mendelian disorders caused by recurrent variants, but this is the exception rather than the rule for most genetic disorders. Now DNA analysis is beginning to support natural appetites for clinical prognos- tication: for diagnosis, decision-​making, therapeutic discovery, and drug targeting, but knowledge of the clinical implications of genomic variation is currently very incomplete, as emphasized by the fact that, of approximately 20 000 genes in the human genome, only about 30% have a known role in human disease. Each human genome contains 4 to 5 million variants (positions where the genetic code in the indi- vidual differs from the human reference sequence), and distinguishing disease-​causing variants from rare background variants can be very challenging. We are only beginning to learn how to integrate DNA sequencing information into clinical practice, and there can be con- siderable dangers in assuming that a particular DNA sequence has (or will have) a particular clinical impact in a particular patient. In clinical contexts where there is limited experience of sequencing methodology, there needs to be caution in the diagnostic field and it has been suggested that the following stratagems be adopted: (1) obtain detailed clinical phenotyping to create a carefully thought out and justifiable differential diagnosis—​if it seems biologically unlikely that a particular gene is involved in the generation of a particular

12.1  The inborn errors of metabolism: General aspects 1939 phenotype, then it is likely to be wrong to assume that a mutation in that gene is causal; (2) scrutinize the family history and mode of inheritance, and compare (when possible) gene sequence in the proband with that in other family members; and (3) confirm DNA sequence data by manual inspection of the sequencing reads for genes that are of interest, and check these by independent methods approved for clinical use. It is also appropriate to point out that mas- sive parallel sequencing does not detect all genetic abnormalities: it may be necessary to deploy alternative sequencing such as Sanger sequencing and deletion/​duplication testing to detect single nucleo- tide and copy number deletion and in-​frame deletions/​duplications that can readily be missed. In the field of rare diseases, of which the huge catalogue of inborn errors of metabolism occupies a central place, DNA sequencing is of arresting interest for practitioners. The perennial need for prompt diagnosis carries with it the prospect of comfort and utility when faced with diagnostic challenges and countless choice, and also has potential implications for mass screening as much as for the stricken individual. Sequencing circulating cell-​free DNA in pregnancy The most widespread clinical application of DNA testing is pre- natal testing for fetal chromosomal abnormalities, such as trisomy 21 in Down’s syndrome. As of 2017, up to 6 million women have undergone sequence analysis of cfDNA for fetal aneuploidy (see Chapter 3.9). cfDNA is obtained at about 10 weeks of pregnancy from small samples of maternal blood. Described accurately as the ‘fastest growing genetic test in medical history’, this technology, introduced as a plasma analyte after 20 years of research by Dennis Lo and col- leagues, offers great hope for applications of DNA sequencing in clin- ical medicine—​for example, in cancer diagnosis and monitoring and for diagnosis and screening in single-​gene disorders. In relation to screening for inborn errors and rare diseases gener- ally, ‘liquid’ biopsy of fetal DNA circulating in maternal blood offers potentially easy access to the genomic DNA of the fetus at risk in a noninvasive manner. However, while testing for fetal aneuploidy and some cancers has little need for nucleotide-​level accuracy (chromo- somes can be counted without checking for sequence variation), the application to noninvasive diagnosis of suspected genetic diseases as well as de novo screening for single-​gene disorders has more fas- tidious requirements. Given the potential value and need for such testing, however, rapid progress has been made: there are reliable means to detect fetal rhesus D genotypes in rhesus D-​negative women, and some skeletal dysplasias can be identified when the ultrasonographic findings prompt suspicion. Most inborn errors of metabolism are autosomal recessive dis- orders and here it is first necessary to quantify the abundance of maternal alleles or haplotypes in the fetus relative to the wild-​type counterpart in the maternal DNA component. Relative dosages of maternal DNA and fetal in plasma have been used successfully to predict transmission of sickle cell anaemia, ß-​thalassaemia and haemophilia-​ care being taken to obtain sequence from fetal cfDNA components to assemble the structure of the haplotype flanking the putative mutant locus. At the time of writing, inborn errors such as congenital adrenal hyperplasia due to 21-​hydroxlase de- ficiency and the X-​linked lysosomal disease, Hunter syndrome (mucopolysaccharidosis type II), have been successfully diagnosed by analysis of cfDNA in maternal plasma. In vitro fertilization and the inborn errors of metabolism The human embryo produced by in vitro fertilization can be biopsied at a very early stage of development (i.e. at the eight-​cell stage). A single cell is removed and examined for the DNA mutation responsible for the disease which the parents are known to be carrying or for parental haplotypes tightly linked to the parental mutations). This technique, known as preimplantation genetic diagnosis, enables only embryos which do not carry the disease-​causing genotype to be implanted. Animal genetic models of inborn errors of metabolism in humans Animal models of the inborn errors of metabolism occur spontan- eously and have been used in therapeutic research for many years, but the capacity to generate models of genetic disease by transgenic techniques has greatly advanced this avenue of exploration. Not only do such models offer the hope of shedding important light on the mechanisms of disease, they have much to offer in the develop- ment of innovative treatments before attempting to transfer these to patients—​now referred to as translational medical research. The discovery of embryonic stem cells in the mouse and the ability to manipulate the mammalian genome by targeted homologous recom- bination have been instrumental in generating ‘knock-​out’ models of human genetic diseases. Once the cognate nuclear gene of the mouse has been disrupted in embryonic stem cells, these cells are injected into the inner cell mass of individual host blastocysts. In some of the resultant chimeric embryos, the embryonic stem cells harbouring the mutant locus contribute to the development of the gonads in the adult progeny; ultimately, when this is the case, offspring can be bred to homozygosity for the disrupted locus and studied. Refinements of this technology based on the use of regulatory sequences and tissue-​specific promoter elements permit the target locus to be manipulated at will in the whole animal at a predefined stage of development by the administration of small molecules that bind to control elements (inducible ‘knock-​out’ and/​or ‘knock-​in’ models models) or allow the genetic locus of interest to be deleted in particular tissues (conditional knock-​out model). Murine and other living experimental models of human diseases are valuable in medicinal research but limitations to the method- ology remain when cognitive and behavioural abnormalities are critical features of the clinical phenotype in patients; even with the constraints of recruitment in individually rare diseases, the experi- mental system by which innovative treatments are best tested for use in human patients remains the clinical trial. The future Astonishing progress has been made in the understanding, diag- nosis, and treatment of inborn errors of metabolism and it is clear that the future holds immense promise for continued advancement and introduction of credible therapies for conditions that until re- cently were beyond hope—​either for clinical control, rescue, or even reversal. Prodigious research efforts to accelerate and perfect understanding of many rare inborn disorders are now bearing fruit. Molecular genetics, cell biology, and biochemistry have been pro- ductively collated into clinical practice as a result of a contemporary explosion of knowledge about the genome, the associated lexicon of

SECTION 12  Metabolic disorders 1940 inherited disease, and the derived information about protein struc- ture and function. For these reasons, it seems that the future highlights of this rare and formerly neglected clinical field will be dominated by spin-​out discoveries that ultimately relate to the study of DNA. Genetic sequence data Incremental improvements in the application of genomic sequen- cing will be used routinely to investigate infants and children with undiagnosed conditions, especially those in whom a strong gen- etic basis is suspected. Increasingly in private health facilities and services available in resource-​rich societies, this technology will be extended to older patients. Depending on the analytical strata- gems chosen and whether whole-​exome or genomic sequencing is adopted, even now some providers claim that disease-​causing muta- tions are identified in up to one-​third of patients. It is almost inevitable that economies of scale resulting from cen- tralized national provision, and refinements in the bioinformatic interrogation and hierarchical stratagems used to analyse the data, will enhance its value for earlier diagnosis, and (one hopes) better coordinated access to counselling and specialized care. With ex- perience and improved information about human genetic variation in different populations, greater clarity will emerge on the clinical value of this approach. With introduction of large population-​scale studies such as the Biobank resource in the United Kingdom and the All of Us research programme in the United States of America, data interpretation and family studies will become better aligned to the experience and prac- tices of local diagnostic teams. In many cases, the validated diag- noses obtained by next-generation sequencing lift the health of the patient and ultimately enhance clinical care with appropriate service provision. Greater experience with these methods will overcome the numerous pitfalls of whole-​exome and whole genome sequencing, where overzealous betrayal of time-​honoured clinical principles can lead to diagnostic errors and chaotic management. Careful clinical testing and the importance of phenotype-​guided molecular testing to drive diagnosis, together with confirmation of diagnosis by in- dependent methods (e.g. biochemical analysis), should, with time, minimize these disturbing errors due to lack of experience and expertise. Liquid biopsy and maternal plasma cfDNA sequencing will in- creasingly translate to clinical care of genetic diseases. As the discov- erers of this technology are aware, the application of more precise screening represents a critical advance in medicine, with reduced need for invasive procedures will attract significant policy debate as inevitable ethical challenges present themselves. Specific therapies The benefit of specific therapies that target specific molecular defects in which cause-​and-​effect pathogenesis has been established is an area of intense aspiration and, now, productive endeavour for fu- ture application. The interventions already include augmenting gene therapies (e.g. recombinant adeno-​associated viral vectors systems that transduce nonmitotic cells and third-​generation integrating lentiviral vectors that transduce actively dividing cells) and gene suppression therapies that diminish RNA expression and thus re- duce biosynthesis of harmful mutant or unregulated wild-​type pro- tein (e.g. antisense methods and use of small interfering RNAs) have also recently been approved for clinical use (transthyretin-​related amyloidosis and polyneuropathy). Other molecular therapies offering generalizable principles for expanded application include the protein channel potentiator drugs for class III mutations of the cystic fibrosis transmembrane regulator protein, CFTR. Mutation-​specific stratification of CFTR therapy with the singe agent, ivacaftor, facilitates increased chloride trans- port by potentiating the channel-​open probability (or gating) of the G551D-​CFTR variant protein. Progress in access to orphan therapies With a burgeoning interest in so-​called personalized medicine, in part driven by the pharmaceutical industry, beyond distinct cancer variants, inborn errors of metabolism are increasingly seen as the exemplary case for therapeutic exploration. The advantages of the orphan drug legislation (including 7–​10 years of marketing exclu- sivity in the United States of America and Europe, respectively) pro- vide strong incentives for the development of innovative agents for such rare diseases with identified genetic targets. On an individual basis, many orphan agents are exceptionally costly for each patient and thus have attracted considerable political attention for reim- bursement. More challenging will be the commercial development of treatments involving one-​off interventions such as gene therapy, because here the ‘cost models’ allowing credible investment by bio- pharmaceutical organizations and shareholders have yet to be put in place. While there is not space here to discuss these matters (see ‘Further reading’), they will undoubtedly come to increasing polit- ical importance as new and effective medicines are introduced for inborn errors of metabolism and cognate rare disorders. FURTHER READING Accurso FJ, et al. (2010). Effect of VX-​770 in persons with cystic fibrosis and the G551D-​CFTR mutation. N Engl J Med, 363, 1991–​2003. Alison MR, Islam S, Lim SM (2009). Cell therapy for liver disease. Curr Opin Mol Ther, 11, 364–​74. Altshuler D, Daly MJ, Lander ES (2008). Genetic mapping in human disease. Science, 322, 881–​8. Auricchio A, Smith AJ, Ali RR (2017). The future looks brighter after 25 years of retinal gene therapy. Hum Gene Ther, 28, 982–​7. Bainbridge JWB, et al. (2008). Effect of gene therapy on visual function in Leber’s congenital amaurosis. N Engl J Med, 358, 2231–​9. Barry PJ, Donaldson AL, Jones AM (2018). Ivacaftor for cystic fibrosis. BMJ, 361, 287–​9. Bearn AG (1993). Archibald Garrod and the individuality of man. Oxford University Press, Oxford. Bianchi DW, Chiu RWK (2018). Sequencing of circulating cell-​free DNA during pregnancy. N Engl J Med, 379, 464–​73. Biffi A (2017). Hematopoietic gene therapies for metabolic and neuro- logic diseases. Hematol Oncol Clin North Am, 31, 869–​81. Buckley B (2008). Clinical trials of orphan medicines. Lancet, 371, 2051–​5. Cartier N, et al. (2009). Haemopoietic stem cell therapy with a lentiviral vector in X-​linked leukodystrophy. Science, 326, 818–​23. Childs B (2004). A logic of disease. In Scriver CR, et al. (eds) Metabolic and molecular bases of inherited disease, 8th edition. McGraw-​Hill, New York. http://​www.ommbid.com. Chinnery PF (2015). Mitochondrial disease in adults: what’s old and what’s new? EMBO Mol Med, 7, 1503–​12.

12.1  The inborn errors of metabolism: General aspects 1941 Cox TM (2012). Alkaptonuria: leading to the treasure in exceptions. JIMD Rep, 5, 49–​57. Cox TM, Cachón-​González MB (2012). The cellular pathology of lyso- somal diseases. J Pathol, 226, 241–​54. Cox TM, et  al. (2017). Eliglustat maintains long-​term clinical sta- bility in patients with Gaucher disease type 1 stabilized on enzyme therapy. Blood, 129, 2375–​83. D’Costa J, Mansfield SG, Humeau LM (2009). Lentiviral vectors in clinical trials: current status. Curr Opin Mol Ther, 11, 554–​64. Deverman BE, et al. (2018). Gene therapy for neurological disorders: progress and prospects. Nat Rev Drug Discov, 17, 641–​59. Endo A (2008). A gift from nature: the birth of the statins. Nat Med, 14, 1050–​2. Enquist IB, et al. (2006). Effective cell and gene therapy in a murine model of Gaucher disease. Proc Nat Acad Sci U S A, 103, 112–​19. Fan JQ (2003). A contradictory treatment for lysosomal storage disorders: inhibitors enhance mutant enzyme activity. Trends Pharmacol Sci, 24, 355–​60. Gahl WA (2012). The battlefield of rare diseases: where uncommon insights are common. Sci Transl Med, 4, 154ed7. Gahl WA, Balog JZ, Kleta R (2007). Nephropathic cystinosis in adults: natural history and effects of oral cysteamine therapy. Ann Intern Med, 147, 242–​50. Garrod AE (1909). Inborn errors of metabolism. Oxford University Press, Oxford. Garrod AE (1931). The inborn factors in disease: an essay. Clarendon Press, Oxford. Gooptu B, Lomas DA (2009). Conformational pathology of the serpins:  themes, variations, and therapeutic strategies. Annu Rev Biochem, 78, 147–​76. Grabowski GA (2008). Phenotype, diagnosis, and treatment of Gaucher’s disease. Lancet, 372, 1263–​71. Green ED, Rubin EM, Olson MV (2017). The future of DNA sequen- cing. Nature, 550, 179–​81. Grieger JC, Samulski RJ (2005). Adeno-​associated virus as a gene therapy vector: vector development, production and clinical appli- cations. Adv Biochem Eng Biotechnol, 99, 119–​45. Haffner ME (2006). Adopting orphan drugs—​two dozen years of treating rare diseases. N Engl J Med, 354, 445–​7. Haffner ME (2016). History of orphan drug regulation –​ United States and beyond. Clin Pharmacol Ther, 100, 342–​3. High KA, Aubourg P (2011). rAAV human trial experience. Methods in Molecular Biology, 807, 429–​57. High KA, Roncarolo MG (2019). Gene Therapy. N Engl J Med, 381, 455–64. Khanna A, et al. (1999). Liver transplantation for metabolic liver dis- ease. Surg Clin North Am, 79, 153–​62. Lo YM, et al. (1997). Presence of fetal DNA in maternal plasma and serum. Lancet, 350, 485–​7. MacFarland R, Turnbull DM (2009). Batteries not included:  diag- nosis and management of mitochondrial disease. J Intern Med, 265, 210–​28. Maguire AM, et  al. (2008). Safety and efficacy of gene transfer in Leber’s congenital amaurosis. N Engl J Med, 358, 2240–​8. Miyamoto BE, Kakkis ED (2011). The potential investment impact of improved access to accelerated approval on the development of treat- ments for low prevalence rare diseases. Orphanet J Rare Dis, 6, 49. Nathwani AC, et  al. (2011). Adenovirus-​associated virus vector-​ mediated gene transfer in hemophilia B. N Engl J Med, 365, 2357–​65 Pena LDM, et al. (2018). Looking beyond the exome: a phenotype-​first approach to molecular diagnostic resolution in rare and undiag- nosed diseases. Genet Med, 20, 464–​9. Phornphutkul C, et al. (2002). Natural history of alkaptonuria. N Engl J Med, 347, 2111–​21. Rangarajan S, et al. (2017). AAV5-​factor viii gene transfer in severe hemophilia A. N Engl J Med, 377, 2519–​30. Ranganath LR, et al. (2018). Nitisinone arrests ochronosis and ­decreases rate of progression of Alkaptonuria: Evaluation of the effect of nitisinone in the United Kingdom National Alkaptonuria Centre. Mol Genet Metab, 125, 127–34. Reeve AK, Krishnan KJ, Turnbull DM (2008). Mitochondrial DNA mutations in disease, aging and neurodegeneration. Ann N Y Acad Sci, 1147, 21–​9. Sessa M, et al. (2016). Lentiviral haemopoietic stem-​cell gene therapy in early-​onset metachromatic leukodystrophy: an ad-​hoc analysis of a non-​randomised, open-​label, phase 1/​2 trial. Lancet, 388, 476–​87. Shashi V, et al. (2019). A comprehensive iterative approach is highly effective in diagnosing individuals who are exome negative. Genet Med, 21, 161–​72. Suwannarat P, et  al. (2005). Use of nitisinone in patients with alkaptonuria. Metabolism, 54, 719–​28. Tambuyzer E (2010). Rare diseases, orphan drugs and their regula- tion: questions and misconceptions. Nat Rev Drug Discov, 9, 921–​9 Tardieu M, et al. (2017). Intracerebral gene therapy in children with mucopolysaccharidosis type IIIB syndrome: an uncontrolled phase 1/​2 clinical trial. Lancet Neurol, 16, 712–​20. Testa F, et al. (2013). Three-​year follow-​up after unilateral subretinal delivery of adeno-​associated virus in patients with Leber congenital amaurosis type 2. Ophthalmology, 120, 1283–​91. Thompson AA, et al. (2018). Gene Therapy in Patients with Transfusion-Dependent β-Thalassemia. N Engl J Med, 378, 1479–93. Vermeersch JR, Voet T, Devriendt K (2016). Prenatal and pre-​ implantation genetic diagnosis. Nat Rev Genet, 17, 643–​56. Wang J, et al. (2017). MARRVEL: integration of human and model or- ganism genetic resources to facilitate functional annotation of the human genome. Am J Hum Genet, 100, 843–​53. Worgall S, et al. (2008). Treatment of late infantile neuronal ceroid lipofuscinosis by CNS administration of a serotype 2 adeno-​ associated virus expressing CLN2 cDNA. Hum Gene Ther, 19, 463–​74. Zhang KY, Tung BY, Kowdley KV (2007). Liver transplantation for metabolic liver diseases. Clin Liver Dis, 11, 265–​81. Zschocke J (2008). Dominant versus recessive: molecular mechanisms in metabolic disease. J Inher Metab Dis, 31, 599–​618.

12.10 Hereditary disorders of oxalate metabolism T

12.10 Hereditary disorders of oxalate metabolism: The primary hyperoxalurias 2174

ESSENTIALS Primary hyperoxalurias (PHs) are rare inherited disorders character- ized by an increased endogenous synthesis of oxalate caused by a deficiency in one of several liver enzymes involved in glyoxylate metabolism. The excess oxalate is eliminated from the body by the kidneys. High concentrations of oxalate in the urine increase the risk of calcium oxalate deposition in the kidney (resulting in nephrocalcinosis) and in the urinary tract (leading to urinary stones). Primary hyperoxaluria is characterized by recurring calcium oxalate stones, presenting from early childhood to late adult life. Over time, deposition of calcium oxalate crystals in kidney tissue leads to kidney damage with progressive loss of kidney function. Primary hyperoxaluria type 1 (PH1; alanine–​glyoxylate aminotransferase deficiency) is the most severe form with a me- dian age at end-​stage renal failure reached during young adult- hood. Patients with PH type 2 (PH2; glyoxylate/​hydroxypyruvate reductase deficiency) and PH type 3 (PH3; 4-​hydroxy-​2-​oxo-​ glutarate aldolase deficiency) may show preservation of kidney function well into adulthood. Systemic deposition of calcium ox- alate (oxalosis) can follow kidney failure and increased plasma oxalate levels. Diagnosis is made by DNA analysis of peripheral blood samples, or more rarely by enzyme assay of liver biopsy tissue. Prenatal diag- nosis can be accomplished in the first trimester by DNA analysis of chorionic villus samples. Treatment relies on high fluid intake, inhibitors of calcium ox- alate crystallization, and, when required, urological procedures for stone removal. Some patients with PH1 respond to vitamin B6 treatment. Management of end-​stage renal failure is difficult as dialysis, whether haemo-​ or peritoneal, cannot match oxalate production. Isolated kidney transplantation places patients at risk of recurring oxalate deposition in the graft in PH1 patients not re- sponsive to vitamin B6. Liver transplantation, usually combined with kidney transplantation, is a curative treatment for PH1 but car- ries significant risks. Introduction Oxalate, hyperoxaluria, and oxalosis Oxalate is an end product of metabolism with no known useful bio- logical function in humans, indeed oxalate can be distinctly det- rimental to complex life forms because of the low solubility of its calcium salt. The solubility product of calcium oxalate is readily ex- ceeded in urine, resulting in its crystallization and aggregation into calculi. Under physiological conditions oxalate, especially calcium oxalate, is only poorly absorbed from the gut, so a limited amount of the body’s oxalate is supplied directly by the diet. Most is de- rived by endogenous synthesis from dietary precursors, or collagen turnover. Most oxalate in the body is removed by urinary excretion. Little appears to be excreted into the gut, but the physiological importance of intestinal elimination, especially in the presence of normal renal function, is unclear. The predominant role of the kidney in oxalate removal makes it the prime target for calcium oxalate deposition (see ‘Renal and urinary manifestations’). The reference range for plasma oxalate in health adults is 1 to 3 µmol/​litre and urinary excretion is less than 450 µmol/​24 h.  In healthy children, the 24-​h oxalate excretion and random urine oxalate/​creatinine ratios vary according to age. However, when normalized for body surface area, urinary excretion rates for chil- dren 2 years or older are similar to those of adults (i.e. <450 µmol/​ 1.73 m2 per 24 h). Genetic causes of hyperoxaluria are rare, but can be severe. The number of different genetic causes of hyperoxaluria is un- known. Three monogenic causes of hyperoxaluria have been well-​ characterized. These are primary hyperoxaluria (PH) types 1, 2, and 3 (PH1, PH2, and PH3). Historical perspectives The condition now recognized as PH was first identified by Lepoutre in 1925. However, it was another quarter of a century be- fore it was described in detail, and it was not until 1957 that it was recognized as a metabolic disorder. The next great leap forward 12.10 Hereditary disorders of oxalate metabolism: The primary hyperoxalurias Sonia Fargue, Dawn S. Milliner, and Christopher J. Danpure

12.10  Hereditary disorders of oxalate metabolism: The primary hyperoxalurias 2175 came in 1968 when Williams and Smith realized that PH was at least two disorders, now known as PH1 and PH2. The enzyme ­defect in PH2 was recognized at the time, but the defect in PH1 did not emerge until 1986. Since then, advances in understanding of the PHs have been rapid, offending genes having been cloned, and numerous mutations identified. The gene defect associated with PH3 was identified in 2010. Treatments have evolved in parallel with our increased under- standing of the aetiology and pathophysiology of the condition. Until 20 years ago, the outlook for patients with PH was bleak. However, in the past two decades, life expectancy for most patients has improved markedly following the introduction of more rational medical and surgical treatments, of which enzyme replacement therapy by liver transplantation stands out. Many PH patients are alive today who would not be were it not for liver transplantation. Our increased understanding, especially of enzyme genotype–​phenotype relation- ships, has led to the exciting prospect of new, possibly mutation-​ specific, pharmacological treatments in the not too distant future. Aetiology and pathogenesis The primary hyperoxalurias are a group of rare hereditary disorders of which only three, PH1 (OMIM 259900), PH2 (OMIM 260000), and PH3 (OMIM 613616) are well characterized. The three types are simple autosomal recessive disorders of glyoxylate metabolism that result in marked increases in the metabolic production of ox- alate and the resulting deposition of insoluble calcium oxalate in the kidney and urinary tract. Despite these apparent similarities, the molecular bases of these disorders are completely different. PH1 is caused by a de- ficiency of the liver-​specific peroxisomal enzyme alanine–​ glyoxylate aminotransferase (AGT, Enzyme Commission (EC) number 2.6.1.44). PH2 is caused by a deficiency of the more widely distributed cytosolic and mitochondrial enzyme glyoxylate/​ hydroxypyruvate reductase (GRHPR, EC 1.1.1.26/​79). PH3 is caused by a deficiency of 4-​hydroxy-​2-​oxoglutarate aldolase (HOGA, EC 4.1.3.16), a mitochondrial enzyme, highly expressed in liver and kidney tissues. Biochemical abnormalities The outcome of AGT, GRHPR, or HOGA deficiency is increased synthesis and urinary excretion of oxalate. In common with all aminotransferases, AGT requires a metabolite of vitamin B6, pyri- doxal phosphate, as cofactor. GRHPR is dependent on reduced nicotinamide-​adenine dinucleotide phosphate (NADH) but HOGA does not require a cofactor. AGT normally catalyses the conver- sion of the intermediary metabolite glyoxylate to glycine, but its absence in PH1 allows glyoxylate to be oxidized to oxalate and re- duced to glycolate instead (Fig. 12.10.1). GRHPR normally catalyses the reduction of glyoxylate to glycolate as well as the reduction of hydroxypyruvate to d-​glycerate. However, its deficiency in PH2 al- lows glyoxylate to be oxidized to oxalate and hydroxypyruvate to be reduced to l-​glycerate (Fig. 12.10.1). HOGA catalyses the conver- sion of 4-​hydroxy-​2-​oxoglutarate (HOG) to glyoxylate and pyruvate. The deficiency of HOGA results in accumulation of HOG. The exact mechanism by which this leads to increased oxalate production is still unclear and may involve inhibition of GRHPR by HOG and/​ or conversion of HOG to glyoxylate or oxalate by as yet unknown enzyme(s). Oxalate cannot be further metabolized and can only be removed from the body by renal and, to a lesser degree, gastrointestinal excre- tion. Although glycolate, l-​glycerate, and HOG can be further me- tabolized, their increased rate of synthesis in the PHs exceeds their ability to be removed metabolically; hence, large amounts of these metabolites are also removed by renal excretion. In most, but not all, patients, hyperoxaluria is accompanied by hyperglycolicaciduria (in PH1), hyper-​l-​glyceric aciduria (in   X X X X X alanine pyruvate glyoxylate glycine AGT DAO glyoxylate glycolate GO glycolate oxalate hydroxyproline hydroxypyruvate D-glycerate L-glycerate glycoladehyde HOG glyoxylate + pyruvate glycolate elevated in PH1, 2, 3 elevated in PH1 elevated in PH2 DHG elevated in PH3 GRHPR GRHPR HOGA GRHPR Hepatocyte Urine LDH cytosol LDH p er o xi so m e m it oc h o n d ri o n Fig. 12.10.1  Main pathways of glyoxylate metabolism in human liver cells. The ‘X’ indicates the location of the enzyme defects in primary hyperoxaluria type 1, 2, or 3. The membranes are likely to be permeable to most or all of the metabolites shown (dotted lines). AGT, alanine–​glyoxylate aminotransferase; DAO, d-​amino acid oxidase; DHG, dihydroxyglutarate; GO, glycolate oxidase; GRHPR, glyoxylate/​hydroxypyruvate reductase; HOG, 4-​hydroxy-​2-​oxoglutarate; HOGA, 4-​ hydroxy-​2-​oxoglutarate aldolase; LDH, lactate dehydrogenase; PH1, primary hyperoxaluria type 1; PH2, primary hyperoxaluria type 2, PH3, primary hyperoxaluria type 3.

section 12  Metabolic disorders 2176 PH2), or elevated concentrations of HOG and dihydroxyglutarate (DHG) in PH3. Concomitant hyperoxaluria and hyperglycolic aciduria used to be considered pathognomonic of PH1, and concomitant hyperoxaluria and hyper-​l-​glyceric aciduria path- ognomonic of PH2. However, up to one-​quarter of PH1 pa- tients do not exhibit hyperglycolicaciduria, and some PH2 patients do not have hyper-​l-​glyceric aciduria. In PH3 elevated concentrations of HOG and DHG, a metabolite of HOG, are found in the urine. Although glycolate and l-​glycerate are useful in the differential diagnosis of PH1 and PH2, they themselves appear to cause no ill effects. All the pathological sequelae of the PHs are associated with the increased synthesis and excretion of oxalate. Molecular genetics The phenotype of PH1 is heterogeneous both at a clinical and at a molecular level. Three major enzymic categories are recognized: (1) absence of both AGT catalytic activity and AGT immunoreactive protein, (2)  absence of AGT catalytic activity but the presence of AGT immunoreactive protein, and (3)  presence of both AGT catalytic activity and AGT immunoreactive protein. Surprisingly for a recessive disease, many patients in the last category can have AGT activity similar to that found in asymptomatic heterozygotes. In most of the latter patients, disease is caused by a protein traf- ficking defect in which AGT is mistargeted from its normal loca- tion in the peroxisomes to the mitochondria. Although mistargeted AGT is still enzymically active, it is unable to fulfil its metabolic function (i.e. glyoxylate transamination) properly when located in the mitochondria. AGT is encoded by the AGXT gene, which contains 11 exons, spanning approximately 10 kb on chromosome 2q37.3. More than 200 mutations and polymorphisms have been identified at the AGXT locus, the three most common of which are described in Table 12.10.1. Many mutations in AGXT, including some of the most common, segregate and functionally interact with a very common polymorphism that results in a Pro11Leu amino acid replacement. GRHPR is encoded by the GRHPR gene, which contains nine exons and spans approximately 9 kb in the pericentromeric re- gion of chromosome 9. HOGA is encoded by the HOGA1 gene, lo- cated on chromosome 10q24.2, containing seven exons, spanning 27 kb. Rather fewer mutations have been found at the GRHPR and HOGA1 loci. Structural biology The X-​ray crystal structures of AGT (Fig. 12.10.2), GRHPR, and HOGA have been determined, enabling the effects of many of the missense mutations and, in the case of AGT, their interactions with the Pro11Leu polymorphism to be rationalized. The most common muta- tion found in PH1, with an allelic frequency of 30 to 40% in Europeans, leads to a Gly170Arg amino acid replacement. This mutation, to- gether with the common Pro11Leu polymorphism, is responsible for most cases of AGT peroxisome-​to-​mitochondrion mistargeting. The Pro11Leu polymorphism generates a functionally weak N-​ terminal mitochondrial targeting sequence, the efficiency of which Table 12.10.1  The most common mutations and polymorphisms found in alanine–​glyoxylate aminotransferase, glyoxylate reductase, and 4-​hydroxy-​2-​oxo glutarate aldolase in European and North American populations Polymorphism/​ mutation Description Allelic frequencya PH1 patients (%) Normal populationa (%) AGXT polymorphisms Pro11Leub, c Substitution of proline by leucine at residue 11 c.50 15–​20 Intron 1 duplicationc A 74-​bp partial duplication of intron 1 c.50 15–​20 Ile340Metc Substitution of isoleucine by methionine at residue 340 c.50 15–​20 AGXT mutations Gly170Argb, d Substitution of glycine by arginine at residue 170 30–​50 33–​34dupC Insertion of a single base (C) leading to a frameshift c.12 Ile244Thrd, e Substitution of isoleucine by threonine at residue 244 6–​9 PH2 patients (%) GRHPR mutations c.103delG Frameshift, premature stop codon c. 35 c.403_​404+2del Mis-​splicing 18 PH3 patients (%) HOGA1 mutations Glu315del In frame deletion of glutamine residue 31 c.700+5G>T Mis-​splicing 49 PH1, primary hyperoxaluria type 1; PH2, primary hyperoxaluria type 2; PH3, primary hyperoxaluria type 3. a In European and North American populations. b Pro11Leu and Gly170Arg synergistically interact to misdirect AGT from its normal location in hepatocyte peroxisomes to mitochondria. c These three polymorphic variations together define the minor AGXT allele. d Mutation segregates with the minor allele of AGXT. e Ile244Thr has a much higher frequency in some North African and Spanish populations.

12.10  Hereditary disorders of oxalate metabolism: The primary hyperoxalurias 2177 is enhanced by the additional presence of the p.Gly170Arg muta- tion. Interestingly, the Gly170Arg mutation on its own is predicted to be without any untoward consequences, at least in vitro. Other AGT mutations have been shown to be able to result in mitochondrial mistargeting in in vitro systems, although only the p.Phe152Ile muta- tion is known to do so in PH1. Other forms of primary hyperoxaluria There are other forms of PH in addition to PH1, PH2, and PH3. These are indeterminate in number and poorly characterized. Case studies have been published of individuals with elevated urinary oxalate of presumed metabolic origin, but who have normal AGT and GRHPR activities, or absence of mutation in any the genes responsible for PH1, PH2, and PH3. Potential ex- planations of the basic defects in these non-​PH1, non-​PH2, or non-​PH3 patients have included dysfunction of other metabolic enzymes involved indirectly in oxalate synthesis, abnormalities in enteric oxalate absorption, and defects in renal oxalate excre- tion, but no conclusive proof has been forthcoming for any of these possibilities. Epidemiology The PHs are rare disorders which account for 1 to 10% of cases of paediatric end-​stage renal disease depending on the country. In Europe, PH1 has an estimated prevalence of 1.0 to 2.9 per mil- lion people and an incidence of 0.12 to 0.15 per million per year. However, both prevalence and incidence are likely to be greater in populations with a high frequency of consanguinity. There are limited data available regarding the prevalence or inci- dence of PH2 or PH3. However, within the population of patients with PH, PH1 represents 70 to 80% of patients, with PH2 and PH3 representing approximately 10% each. A recent analysis of carrier frequency in the general population suggests that PH is under-​ diagnosed and/​or incompletely penetrant, especially for PH3. Specific mutations frequencies vary between ethnic groups in all three PH types. Clinical features Renal and urinary manifestations PH presents with symptoms related to urolithiasis, usually in child- hood but sometimes in adult life. Recurring stone formation is characteristic, as is progressive kidney damage, especially in PH1 patients who advance to end-​stage kidney failure at a median of 25 to 35 years of age in European and North American populations. There is better preservation of kidney function overall in PH2 and PH3. With progressive loss of kidney function, a rising plasma ox- alate concentration leads to deposition of calcium oxalate in many organs (systemic oxalosis) (see ‘Systemic oxalosis’). Variability of clinical expression is marked, with some patients reaching end-​stage renal failure in early childhood, while others retain renal function into late adulthood. High concentrations of oxalate in the urine result in the forma- tion of calcium oxalate crystals that attach to the renal tubule epi- thelium. The crystals are then endocytosed by renal tubule epithelial cells and migrate to the renal interstitium. There, the crystals incite an inflammatory, giant-​cell reaction that results in renal injury, ul- timately leading to interstitial fibrosis. Widespread calcium oxalate deposition in the renal parenchyma is termed nephrocalcinosis and is usually visible on renal imaging studies. Aggregation of cal- cium oxalate crystals in the urinary space leads to stone formation (nephrolithiasis or urolithiasis). For reasons that remain poorly understood, infants and young children appear more likely to de- velop nephrocalcinosis, although it can occur at any age. Stones in the absence of nephrocalcinosis are more characteristic in older children and adults. Although marked hyperoxaluria is present from early infancy, the age at which symptoms develop is highly variable, ranging from a few months to late adulthood. In most patients, symptoms or find- ings related to urolithiasis (pain, haematuria, stone passage) are evi- dent in early childhood. Recurring stone formation is characteristic, often requiring multiple stone removal procedures. Over time, the damaging effects of calcium oxalate deposition in the kidney, epi- sodes of transient obstruction due to stones, and injury related to stone-​removal procedures or infection result in irreversible loss of renal function. End-​stage kidney failure can occur at any age, from infancy to the sixth decade of life, with a median of approximately 30 years in PH1 (Fig. 12.10.3). End-​stage kidney failure is also pos- sible in patients with PH2 or PH3, despite being considered milder forms of PH. In a few patients, the first clinical manifestation of PH is kidney failure, with symptoms of uraemia prompting medical attention. On evaluation, nephrocalcinosis and/​or bilateral renal stones are usually found. Occasionally, the diagnosis is made on renal biopsy in a pa- tient in whom PH was not considered on clinical grounds. A severe infantile form of PH1 results in irreversible kidney failure during the first year or two of life, presenting as failure to thrive. Systemic oxalosis When kidney function falls below a GFR of about 30 to 35 ml/​ min per 1.73 m2, the kidney is unable to excrete the excess oxalate ­produced by the liver and the plasma oxalate concentration begins to rise abruptly. When the calcium oxalate product in plasma ex- ceeds saturation, calcium oxalate crystals are deposited in many Fig. 12.10.2  Crystal structure of human alanine–​glyoxylate aminotransferase. PLP, pyridoxal phosphate in orange. Proline residues in position 11 are shown on the N-​terminal arms. Residues with known pathological mutations in red (G41R, F152I, G170R, I244T).

section 12  Metabolic disorders 2178 organs and tissues (systemic oxalosis), resulting in progressively severe multisystem disease. Painful, nonhealing ulcers of the skin, fracturing osteodystrophy, refractory anaemia, complete heart block, and heart failure due to oxalate cardiomyopathy are features of systemic oxalosis. Without prompt and definitive management, death ensues. Differential features There are no clinical features that can reliably differentiate among the three PH types in an individual patient, but PH2 is characterized by slightly lower oxalate excretion rates, fewer stone episodes, and better preservation of renal function than PH1. PH3 patients have an earlier onset of symptoms but lower oxalate excretion rates and slower progression of disease. Though symptomatic stones are not infrequently encountered in infants and young children with PH3, the frequency of stone events may improve during follow-​up. There are rare reports of kidney failure in PH3, which may be ­related to complications of urolithiasis. Differential diagnosis Hyperoxaluria is a well-​recognized risk factor in the common condition of idiopathic calcium oxalate kidney stone disease. Although its causes in such patients remain unclear, they are almost certainly multifactorial in nature, with both environ- mental and genetic components (see Chapter 21.14). Anything that increases the body’s burden of oxalate, or elevates the con- centration of oxalate in the urine, increases the risk of calcium oxalate deposition in the kidney and/​or urinary tract resulting in nephrocalcinosis and/​or urinary stones. Environmental causes of hyperoxaluria include excessive dietary intake of oxalate (par- ticularly when combined with low calcium intake) and extended periods of dehydration. Intake of oxalate precursors, such as intravenous ascorbic acid in patients receiving parenteral nutri- tion, or accidental ingestion of ethylene glycol, is occasionally responsible. Enhanced gut absorption of oxalate is often encoun- tered in patients with gastrointestinal disease or after small bowel resection. Malabsorptive gastric bypass procedures, performed for management of obesity, are emerging as an increasingly frequent cause of enteric hyperoxaluria. Hyperoxaluria is also encountered in patients receiving medications that alter fat absorption, such as tetrahydrolipstatin (orlistat). Clinical investigations PH should be considered in any child with urinary tract stones or nephrocalcinosis and in adults with recurrent calcium oxalate stones, especially if the clinical history extends back into childhood. Impaired renal function in a patient with calcium urolithiasis or nephrocalcinosis, or in a sibling, should also suggest the diagnosis. A presumptive diagnosis of PH can often be made on the basis of urinary oxalate, glycolate, l-​glycerate, and HOG/​DHG excre- tion. Due to highly age-​dependent normal ranges in young children, random urine oxalate/​creatinine ratios are best regarded as an initial screen. If the ratio appears elevated, a timed (12–​24-​h) urine collec- tion should be obtained for more reliable diagnostic information. Repeating the measure on at least three occasions can be useful to help with false positives (as in the case of small children) and false negatives. It should be kept in mind that urinary oxalate excretion can be misleadingly low in patients with advanced renal failure, and concomitant hyperglycolic aciduria (in PH1) or hyper-​l-​glyceric aciduria (in PH2) is not always present. Urinary HOG and DHG are elevated in patients with PH3, though their diagnostic sensitivity and specificity remain to be established. Plasma levels of oxalate, glycolate, and glycerate are rarely of diagnostic benefit in patients whose renal function is well maintained, though they can be valu- able in those with renal failure. Plasma oxalate concentrations are often increased in patients with end-​stage renal disease from causes unrelated to PH, although the degree of elevation in this setting is typically modest. Definitive diagnosis requires confirmation of homozygosity or compound heterozygosity for known mutations of AGT, GRHPR or HOGA, or the determination of either AGT (PH1) or GRHPR (PH2) enzyme activity on a percutaneous needle biopsy of the liver. The identification of more than 200 mutations in PH1 allows the possi- bility of diagnosis by DNA analysis in suitable families. It has been estimated that, even in the absence of family history, screening pos- sible European or North American PH1 patients for the three most common mutations (Table 12.10.1) would be able to diagnose PH1 with an efficiency of 34%. When coupled with family and linkage analysis studies the success rate is greatly improved. Increasingly, more comprehensive gene sequencing is being used, leading to im- proved diagnostic efficiency, with mutations in AGXT, GRHPR, or HOGA1 found in approximately 90% of patients with PH symptoms. The urinary profile of PH-​related metabolites (glycolate, l-​glycerate, HOG, DHG) can be used to target the initial genetic testing. In pedigrees with a known mutation, screening of family members is straightforward. The determination of AGT (for PH1) or GRHPR (for PH2) activity on liver biopsy, is now mostly reserved for diagnosis confirmation 100 80 60 40 p<0.0001 PH1 PH2 PH3 NMD 76% (129) 43% (49) 60 40 20 Age (y) Renal Survival (%) 100% (18) 82% (11) 96% (12) 96% (8) 100% (14) 100% (12) 12% (7) PH2 PH3 NMD PH1 66% (4) 96% (4) 100% (7) 20 0 Fig. 12.10.3  Renal survival in primary hyperoxaluria. Kaplan–​Meier renal survival plot of patients with PH1 (blue), PH2 (red), PH3 (green), and PH patients with no mutation detected (NMD, black). The lower table shows renal survival estimates with number of patients at risk in parentheses. Source data from Hopp K, et al. (2015) Phenotype-​genotype correlations and estimated carrier frequencies of primary hyperoxaluria. JASN. 2015 Oct;26(10):
2559–70. Copyright © 2018 American Society of Nephrology.

12.10  Hereditary disorders of oxalate metabolism: The primary hyperoxalurias 2179 or exclusion in the absence of mutations of any of the three genes involved. Prenatal diagnosis relies on DNA (mutation or linkage) ana- lysis of material obtained from chorionic villus samples in the first trimester. Treatment There are no international guidelines for the specific treatment of PH and recommendations are based on expert opinions. The man- agement initially involves maintenance of high fluid intake; medi- cations to inhibit calcium oxalate crystallization and decrease oxalate production, with use of pharmacological doses of pyridoxine (vitamin B6) for some PH1 patients; and urological procedures and support for end-​stage renal failure as required. Treatments that target reduction of calcium oxalate crystal or stone formation, either in the urine or in the blood and body tis- sues (in patients with end-​stage kidney disease) are suitable for all types of PH, whereas those addressing enzyme dysfunction are more likely to be disease specific (Fig. 12.10.4). Diet and fluids Decreasing dietary oxalate in PH is of limited use since the main source of oxalate is endogenous. Calcium intake should remain normal as it binds intestinal oxalate. Excessive vitamin C intake should be avoided, especially in end-​stage renal disease, as ascorbic acid can be broken down to oxalate. Patients with PH who have adequate renal function should main- tain a high oral fluid intake in order to keep oxalate in the urine as dilute as possible. A suitable target level is 2 to 3 litres/​m2 body sur- face area, distributed throughout the day, using the lower range for most adults. In infants and young children, placement of a feeding or gastrostomy tube may be needed to assure sufficient intake and in situations of high fluid loss or limited oral intake, intravenous fluids may be required to maintain hydration in PH patients. Pharmacological treatments Reduction in calcium oxalate crystal formation can be accomplished by lowering the urine oxalate concentration and by the use of medi- cation. Urine alkalinization with citrates, either as sodium citrate (0.1–​0.15 g/​kg per day) or equivalent doses of either sodium or po- tassium citrate, reduce the degree of calcium oxalate saturation in the urine. Other inhibitors of crystallization may be used such as neutral phosphates (providing 20–​30 mg/​kg per day of elemental phosphorus in divided doses) to increase the excretion of pyrophos- phate ions, which inhibit heterogeneous calcium oxalate crystal nucleation, seeded growth, and aggregation. Magnesium supple- ments (e.g. magnesium oxide 200 mg/​day in adult patients) may also inhibit crystal growth and aggregation. The doses used should be sufficient to produce a material increase in the urinary excretion of either phosphate or magnesium. Phosphate and magnesium should be avoided if there is renal insufficiency. In about one-​third of PH1 patients, pharmacological doses (5–​ 8 mg/​kg per day, <20 mg/​kg per day) of pyridoxine (vitamin B6) cause a significant (>30%) reduction in urinary oxalate levels and improvement in clinical condition. Pyridoxal phosphate, a metab- olite of pyridoxine, is the cofactor for AGT, and acts both to in- crease the enzymatic activity of AGT and as a chaperone to help AGT folding and targeting. Patients carrying the p.Gly170Arg or p.Phe152Ile mutations have been shown to be able to respond to pyridoxine treatment, although to varying degrees. Some who are homozygous for p.Gly170ARg demonstrate normalization or near normalization of urine oxalate excretion while receiving pyridoxine. PH1 AGXT (Vit B6) Gene mutation Protein folding targeting Enzyme functional deficiency AGT GR HOGA Gene therapy Chemical chaperones Enzyme replacement therapy: liver transplantation Treatments aimed at causes of PH Symptomatic treatments Substrate depletion (GO, HypDH) Reducing Oxalate synthesis (LDH) Oxalate degradation Hydration Crystallization inhibitors Lithotripsy Surgery Dialysis Kidney transplantation Increased glyoxylate Increased oxalate synthesized Elevated oxalate excretion Ca Ox crystals Ca Ox stones Renal failure GRHPR HOGA1 PH2 PH3 Fig. 12.10.4  Current and future approaches to the treatment of primary hyperoxaluria types 1 (PH1), 2 (PH2), and 3 (PH3). Current and potential treatments are indicated below the respective molecular and pathophysiological step targeted for PH1, PH2, and PH3. Treatments aimed at the pathways on the left tend to be directed at the causes of disease and are usually specific for each type of primary hyperoxaluria (vitamin B6 is specific to PH1). The treatments for the pathway on the right are aimed at the clinically observable symptoms and are likely to be common to all three types. AGT, alanine–​glyoxylate aminotransferase; CaOx, calcium oxalate; GO, glycolate oxidase; GRHPR, glyoxylate/​hydroxypyruvate reductase; HOGA, 4-​hydroxy-​2-​oxoglutarate aldolase, HypDH, hydroxyproline dehydrogenase. HOGA is expressed in both liver and kidney so that liver transplantation would not necessarily be the only enzyme replacement strategy.

section 12  Metabolic disorders 2180 There are indications that other mutations may also benefit from pyridoxine treatment. A formal testing of pyridoxine responsiveness during a 3-​month trial with urinary oxalate measurements before and after initiation of pyridoxine is recommended for all PH1 pa- tients. Pyridoxal phosphate is not required for the activity of GRHPR or HOGA, hence pyridoxine is ineffective in PH2 and PH3 patients. Radiological and surgical interventions Obstructive uropathy requires prompt stent placement or per- cutaneous nephrostomy to relieve the obstruction. For PH pa- tients, minimally invasive methods are preferred in dealing with stones. Endoscopic procedures methods including semi-​rigid ureterorenoscopy (URS), flexible ureterorenoscopy (RIRS), and per- cutaneous nephrolithotomy (PCNL) are techniques used with suc- cess, with endoscopic lithotripsy using ultrasonic, electrohydraulic, and laser techniques. Extracorporeal shock wave lithotripsy (ESWL) is also used. Open lithotomy for large calculi has become exceptional. Stone debris may require either external drainage via a nephrostomy or internal drainage via a stent, although stents and other foreign bodies in the urinary tract may rapidly become en- crusted with calcium oxalate deposits. Close follow-​up is essential with regular radiological and/​or ultrasonographic assessment, the aim being to keep the kidneys as free from stones as possible while minimizing repeated ESWL or invasive procedures. Renal replacement therapy In most patients, haemodialysis and peritoneal dialysis are not cap- able of preventing progression of systemic oxalosis. Combined liver and kidney transplantation—​the treatment of choice in patients with PH1 who do not respond well to pyridoxine and are approaching end-​stage renal failure—​entails its own significant risks. Management of end-​stage renal failure is difficult. The high rate of oxalate synthesis most often exceeds achievable rates of its removal, even with intensive haemodialysis regimens or combined haemo-​ and peritoneal dialysis. The condition of patients with renal failure progressively worsens as calcium oxalate is deposited throughout the body (systemic oxalosis). Kidney transplantation can resolve the uraemic consequences of kidney failure and reduce plasma oxalate concentrations to levels that fall below the supersaturation threshold for calcium oxalate. However, kidney transplantation alone is problematic in PH1: pa- tients who respond fully to pyridoxine (with normalization or near normalization of urine oxalate) can do well, as can those with PH2, but otherwise the new kidney is at significant risk from oxalate deposition and rapid failure, particularly if there is delayed graft function. Liver transplantation The rationale for liver transplantation relies on the fact that AGT is more or less liver specific and, although GRHPR is more widely dis- tributed, its activity in the liver greatly exceeds that in other tissues. Liver replacement thus has the potential to replace all, or almost all, the body’s requirement for AGT and, to a lesser extent, GRHPR. Several hundred liver transplantations, often combined with kidney transplantation, have been carried out worldwide for PH1 resulting in a metabolic cure, although it may take many years for the urinary excretion of oxalate to be normalized. This is especially the case if patients have spent many years with poor renal function or on haemodialysis, during which time the corporeal load of calcium ox- alate has built up, particularly in the bones. Pre-​emptive liver transplantation before the GFR has decreased to 30 ml/​min per 1.73 m2 is an option to be considered if PH1 is diag- nosed early and is following an aggressive course. The risks of the transplant procedure, the added years of immunosuppression, and the difficulty in accurate prediction of rate of loss of renal function must be balanced against the benefit. Heterotopic auxiliary liver transplantation is theoretically unsound since the remaining native liver continues to make large amounts of oxalate. Liver transplant- ation has been shown to reduce oxalate excretion to normal in a single PH2 patient who had progressed to end stage kidney disease (Dhondup et al. 2018). Further experience will be needed to evaluate its role in management of PH2. Timing of renal replacement therapy/​transplantation Initiation of maintenance dialysis or transplantation should be ac- complished as soon as the plasma oxalate concentration begins to exceed the solubility threshold for calcium oxalate. This occurs in most patients at a GFR of 20 to 25 ml/​min per 1.73 m2, though can occur earlier in some cases. The purpose of early initiation of renal replacement therapy is to minimize systemic oxalosis and reduce the risk of calcium oxalate deposits in any subsequently grafted kidney. Indeed, it has been suggested that PH1 patients who are either un- responsive or only partially responsive to pyridoxine should be managed with pre-​emptive (before dialysis is required) combined liver and kidney transplantation. By contrast, PH1 patients who re- spond fully to pyridoxine, with normalization or near normalization of urine oxalate while on treatment, and patients with PH2, can do well with kidney transplantation alone. Any time from initiation of dialysis to transplantation should be kept as short as possible to min- imize systemic oxalate accumulation. Vigorous dialysis, required daily in most patients, is needed. The plasma oxalate concentration and urine oxalate excretion rate should be followed sequentially before and after transplantation until normal. Elimination of tissue oxalate stores can take up to 3 years or more following successful transplantation. Careful man- agement of hyperoxaluria throughout this time is essential to avoid damage to the renal allograft. Future developments A recent avenue of research is based on substrate reduction, which has the potential to be applicable to more PH types and mutations. The aims are either to reduce the amount of glyoxylate produced, since it is the precursor to oxalate, or decrease its oxidation to oxalate. The inhibition of the enzyme glycolate oxidase is targeting the peroxi- somal source of glyoxylate. Inhibition of the enzyme hydroxyproline oxidase targets the mitochondrial source of glyoxylate. Inhibition of the enzyme lactate dehydrogenase targets the oxidation of glyoxylate to oxalate, the last step, common to all PH types. These new thera- peutic strategies rely on small interfering RNA (siRNA) administra- tion or use the CRISPR/Cas9 technology. Other more conventional strategies aim at identifying drugs capable of such enzyme inhib- itions. Recent work on calcium oxalate mediated kidney inflamma- tion suggest a potential adjunct role for targeting the inflammatory reaction to preserve renal function. Following promising work in

12.10  Hereditary disorders of oxalate metabolism: The primary hyperoxalurias 2181 animal models of PH1, clinical trials of siRNA therapeutics are cur- rently underway (phase II/III) and show great promise. Just as the discovery of AGT deficiency in PH1 heralded the introduction of enzyme replacement therapy by liver transplantation 30 years ago, so recent discoveries on the functional relationships between mu- tations and enzyme dysfunction will lead to the design of pharma- cological countermeasures. Mutation-​specific chemical chaperones have potential as future treatments for patients with PH1 or PH2 who have missense mutations in AGT or GRHPR. Several screening procedures have been developed to identify chemical chaperones in panels of repurposed pharmaceutical drugs. Gene therapy for PH was forecast more than 15  years ago. Hepatocyte transplantation in an AGT knockout mouse model has been attempted, but this technique requires suppression of the host hepatocytes. Adeno-​associated virus gene transfer has also been attempted in mice but requires further improvements. A re- cent study has identified a drug that can decrease the mitochon- drial mistargeting of AGT seen with certain mutations. Continued research in that direction may yield more candidates. Probiotics such as the oxalate-​degrading bacteria Oxalobacter for- migenes have been the subject of interest for a few years, but have not proven their efficacy in human clinical studies for PH so far, despite their potential to increase oxalate gut secretion. Multiple novel avenues for treatment of PH are explored and offer new hope for patients with this rare disease. FURTHER READING Anders HJ, et al. (2018). The macrophage phenotype and inflammasome component NLRP3 contributes to nephrocalcinosis-related chronic kidney disease independent from IL-1-mediated tissue injury. Kidney Int, 93(3), 656–69. Belostotsky R, et al. (2010). Mutations in DHDPSL are responsible for primary hyperoxaluria type III. Am J Hum Genet, 87, 392–​9. Cochat P, et al. (2012). Primary hyperoxaluria type 1: indications for screening and guidance for diagnosis and treatment. Nephrol Dial Transplant, 27, 1729–​36. Cramer SD, et al. (1999). The gene encoding hydroxypyruvate reduc- tase (GRHPR) is mutated in patients with primary hyperoxaluria type II. Hum Mol Genet, 8, 2063–​9. Danpure CJ, Jennings PR (1986). Peroxisomal alanine:glyoxylate aminotransferase deficiency in primary hyperoxaluria type I. FEBS Lett, 201, 20–​4. Dhondup T, et al. (2018). Combined liver-kidney transplantation for primary hyperoxaluria type 2: A case report. American Journal of Transplantation, 18(1), 253–7. Dutta C, et  al. (2016). Inhibition of glycolate oxidase with dicer-​ substrate siRNA reduces calcium oxalate deposition in a mouse model of primary hyperoxaluria type 1. Mol Ther, 24, 770–​8. Hopp K, et al. (2015). Phenotype-​genotype correlations and estimated carrier frequencies of primary hyperoxaluria. JASN, 26, 2559–​70. Hoppe B, et al. (2009). The primary hyperoxalurias. Kidney Int, 75, 1264–​71. Liebow A, et al. (2017). An investigational RNAi therapeutic targeting glycolate oxidase reduces oxalate production in models of primary hyperoxaluria. J Am Soc Nephrol, 28(2), 494–503. Mandrile G, et al. (2014). Outcome of primary hyperoxaluria type 1 correlates with AGXT mutation type: data from a large European study. Kidney Int, 86, 1197–​204. Martin-​Higueras C, Luis-​Lima S, Salido E (2016). Glycolate oxidase is a safe and efficient target for substrate reduction therapy in a mouse model of primary hyperoxaluria type I. Mol Ther, 24, 719–​25. Monico CG, Milliner DS (2001). Combined liver-​kidney and kidney-​ alone transplantation in primary hyperoxaluria. Liver Transpl, 7, 954–​63. Monico CG, et  al. (2005). Pyridoxine effect in type I  primary hyperoxaluria is associated with the most common mutant allele. Kidney Int, 67, 1704–​9. National Center for Biotechnology Information. Online Mendelian Inheritance in Man (OMIM): http://​www.ncbi.nlm.nih.gov/​entrez/​ query.fcgi?d6MIM; Primary hyperoxaluria type 1:  http://​omim. org/​entry/​259900; Primary hyperoxaluria type 2:  http://​omim. org/​entry/​260000; Primary hyperoxaluria type 3: http://​omim.org/​ entry/​613616. Purdue PE, et al. (1990). Identification of mutations associated with peroxisome-​to-​mitochondrion mistargeting of alanine:glyoxylate aminotransferase in PH1. J Cell Biol, 111, 2341–​51. Rumsby G (2015). Molecular basis of primary hyperoxaluria and strat- egies for diagnosis. Expert Opin Orphan Drugs, 3, 663–​73. Salido EC, et al. (2006). Alanine-​glyoxylate aminotransferase-​deficient mice, a model for primary hyperoxaluria that responds to adeno- viral gene transfer. Proc Natl Acad Sci USA, 103, 18249–​54. Watts RW, et al. (1987). Successful treatment of primary hyperoxaluria type 1 by combined hepatic and renal transplantation. Lancet, 2, 474–​5. Zabaleta N, et al. (2018). CRISPR/Cas9-mediated glycolate oxi- dase disruption is an efficacious and safe treatment for primary hyperoxaluria type I. Nat Commun, 9(1), 5454.

12.11 A physiological approach to acid– base disor

12.11 A physiological approach to acid– base disorders: The roles of ion transport and body fluid compartments 2182

ESSENTIALS The normal pH of human extracellular fluid is maintained within the range of 7.35 to 7.45. The four main types of acid–​base disorders can be defined by the relationship between the three variables, pH, Pco2, and HCO3 –​. Respiratory disturbances begin with an increase or decrease in pulmonary carbon dioxide clearance which—​through a shift in the equilibrium between CO2, H2O, and HCO3 –​—​favours a
decreased hydrogen ion concentration (respiratory alkalosis) or an increased hydrogen ion concentration (respiratory acidosis) respectively. Metabolic acidosis may result when hydrogen ions are added with a nonbicarbonate anion, A−, in the form of HA, in which case bicarbonate is consumed, or when bicarbonate is removed as the sodium or potassium salt, increasing hydrogen ion concentration. Metabolic alkalosis is caused by removal of hydrogen ions or add- ition of bicarbonate. Laboratory tests usually performed in pursuit of diagnosis, aside from arterial blood gas analysis, include a basic metabolic profile with electrolytes (sodium, potassium, chloride, bicarbonate), blood urea nitrogen, and creatinine. Calculation of the serum anion gap, which is determined by subtracting the sum of chloride and bicar- bonate from the serum sodium concentration, is useful. The normal value is 10 to 12 mEq/​litre. An elevated value is diagnostic of meta- bolic acidosis, helpful in the differential diagnosis of the specific metabolic acidosis, and useful in determining the presence of a mixed metabolic disturbance. Acid–​base disorders can be associated with (1) transport pro- cesses across epithelial cells lining transcellular spaces in the kidney (e.g. renal tubular acidosis), gastrointestinal tract (e.g. vomiting), and skin (e.g. cystic fibrosis); (2) transport of acid anions from intracel- lular to extracellular spaces—​anion gap acidosis (e.g. diabetic keto- acidosis, lactic acidosis); and (3)  intake (e.g. infusions with high chloride content). Introduction The normal pH of human extracellular fluid (ECF) is maintained within the range of 7.35 to 7.45. Though intracellular pH regula- tion may be more critical to processes such as protein synthesis, cell growth, and reproduction, it is generally assumed that cell pH, which is more difficult to measure, is reflected in the readily accessible ECF. The usual approach taken to understand how acid–​base disturb- ances are generated starts with the principle of LeChâtellier:  the tendency for chemical reactions thrown out of equilibrium to move in the direction that restores the equilibrium state. This is apparent in the overall equation relating concentrations of carbon dioxide to hydrogen ion and bicarbonate ion in the following expression CO +H O H CO H +HCO 2 2 2 3 + 3 CO ↔ ↔ ←↑ − P 2

(Equation 1) The four main types of acid–​base disorders can be defined by the relationship between the three variables, pH, Pco2, and HCO3–​ (Table 12.11.1). Respiratory disturbances begin with an increase or decrease in pulmonary carbon dioxide clearance which through a shift in the equilibrium favours a decreased hydrogen ion concentration (respiratory alkalosis) or an increased hydrogen ion concentration 12.11 A physiological approach to acid–​base disorders: The roles of ion transport and body fluid compartments Julian Seifter Table 12.11.1  Features of the four main types of primary acid–​base disorder pH Pco2/​HCO3 − Primary disorder Acidaemia Low HCO3 − Metabolic acidosis High Pco2 Respiratory acidosis Alkalaemia High HCO3 − Metabolic alkalosis Low Pco2 Respiratory alkalosis

12.11  A physiological approach to acid–base disorders 2183 (respiratory acidosis) respectively. Metabolic acidosis may result when hydrogen ions are added with a nonbicarbonate anion, A−, in the form of HA, in which case bicarbonate is consumed as the re- action shifts left. Alternatively, bicarbonate may be removed as the sodium or potassium salt, shifting the reaction to the right, increasing hydrogen ion concentration. Metabolic alkalosis is caused by the removal of hydrogen ion or the addition of bicarbonate. In the case of adding acids such as HCl, acid phosphates, and sulphates, or organic acids such as lactic or ketoacids, there would be a decrease in bicarbonate proportional to the increase in A−. The bicarbonate lost due to HA appearance is renewed by net renal ex- cretion of hydrogen and the anion in order to restore acid–​base balance. Since the kidney initially filters large quantities of bicar- bonate from the ECF, all of the filtered bicarbonate must first be reabsorbed just to stay even. Once that process is accomplished, nonbicarbonate buffers, usually ammonium and acid phosphate, remove hydrogen in the urine accompanied by A− and in the pro- cess generate ‘new’ bicarbonate. The metabolic acids that require urinary excretion are usually in the range of 1 to 2 mEq of H+ per kg of body weight. By comparison, the volatile carbon dioxide produc- tion per day is about 20 moles. The diagnostic approach to identifying acid–​base disorders The key questions to ask when attempting to diagnose the cause of an acid–​base disorder are shown in Box 12.11.1. The usual diagnostic approach to an acid–​base disorder begins with a complete history and physical examination. When taking a history, it is important to understand the quantity, contents, and source of fluid losses or gains from the body; ingested substances; and certain diseases known to be associated with acid–​base disorders. Examples include vomiting (metabolic alkalosis), diarrhoea (metabolic acidosis), chronic ob- structive lung disease (respiratory acidosis), pneumonia (respira- tory alkalosis), and so on. Laboratory tests usually performed, aside from arterial blood gas analysis, include a basic metabolic profile with electrolytes (so- dium, potassium, chloride, bicarbonate), blood urea nitrogen, and creatinine. Disturbance of the bicarbonate concentration alone does not prove a metabolic disturbance because there are two other variables in equilibrium with bicarbonate: carbon dioxide and the hydrogen ion concentration. Characterization of the type of acid–​base disturbance A low bicarbonate is consistent with either metabolic acidosis or re- spiratory alkalosis. If blood gases reveal low pH (acidaemia) and low bicarbonate (metabolic acidaemia), the dominant process is a meta- bolic acidosis. The response to metabolic acidosis is through stimu- lation of carotid body and central nervous system chemosensors, which stimulate an increase in alveolar ventilation, so the Pco2 is expected to drop. Normal compensation for metabolic acidosis is predicted based on an expected range of Pco2 established by ob- servation of subjects with simple metabolic acidosis. Should the actual Pco2 be lower than the predicted value, the diagnosis of re- spiratory alkalosis as a second primary disturbance can be made. If the actual Pco2 is higher than predicted, then a simultaneous re- spiratory acidosis is present. It is apparent that metabolic acidosis and its hyperventilatory compensation cause HCO3− and Pco2 to fall. The ratio of HCO3−/​Pco2 is proportional to the pH, as dem- onstrated in the Henderson–​Hasselbalch relationship expressing equation 1 in logarithmic terms, where the pK is 6.1 and the con- centration of carbon dioxide in water is 0.03 mmol/​litre per mmHg of equilibrated gas. pH = p

(Equation 2) The finding of a metabolic acidosis does not rule out multiple processes simultaneously present. There are many findings that help diagnose mixed disturbances. In each of the primary disturb- ances, once the dominant process is identified, it is necessary to infer the degree of compensation by looking at empirical data (see Box 12.11.2 for a range of normal compensations). Determining whether the respiratory response to a metabolic acidosis is as expected for a mixed disorder allows the recognition or exclusion of an additional primary respiratory disorder, but a mixed metabolic disturbance is still possible. For example, Winters’ equation—​which gives the expected value for the patient’s Pco2 if there is adequate respiratory compensation for a metabolic acidosis (Box 12.11.2)—​is still valid when metabolic acidosis and alkalosis coexist if the predominant process is acidosis. In the case where a high-​chloride acidosis and a low-​chloride alkalosis coexist, it is not possible to differentiate this double disturbance from a simple meta- bolic acidosis. Box 12.11.1  The key questions for diagnosis of the cause of an acid–​base disorder Approach to acid–​base disorders 1 Is there acidaemia (pH <7.35) or alkalaemia (pH >7.45)? 2 What is the primary process (metabolic or respiratory, acidosis or alkalosis)? 3 Is there an appropriate compensatory response? 4 If this is an anion gap acidosis, are there other clues to a second pri- mary process? Box 12.11.2  Compensations for acid–​base disturbances Pco2 is measured in mmHg and HCO3 − in mEq/​litre. The direction of compensatory response in Pco2 or HCO3 − is the same as the ini- tial change in HCO3 − or Pco2, but compensation is almost never complete Expected values if simple disturbance Metabolic acidosis (Winters’ formula): Pco2 = (1.5 × HCO3 −) + 8 ± 2 Metabolic alkalosis: Pco2 = 0.7[HCO3 −] + 20 ± 5 Acute respiratory acidosis: [HCO3 −] = 24 + [(Pco2 − 40)/​10] Chronic respiratory acidosis: [HCO3 −] = 24 + 4[(Pco2 − 40)/​10] Acute respiratory alkalosis: [HCO3 −] = 24 –​ 2[(Pco2 − 40)/​10] Chronic respiratory alkalosis: [HCO3 −] = 24 − 5[(Pco2 − 40)/​10]

section 12  Metabolic disorders 2184 The serum anion gap It is useful to calculate the serum anion gap, which is determined by subtracting the sum of chloride and bicarbonate from the serum sodium concentration (the measured ions). The normal value is approximately 10 to 12 mEq per litre, the amount of charge asso- ciated with a normal albumin concentration. The relationship is as follows: Na (Cl + HCO ) = (albumin ) + unmeasured A = anion gap + 3 − − − − − (Equation 3) An elevated anion gap is diagnostic for a metabolic acidosis. Not only is the presence of an anion gap helpful in the differential diagnosis of the specific metabolic acidosis (Box 12.11.3), but it is also useful in determining the presence of a mixed metabolic disturbance. A  calculation of the increment in anion gap (ob- served anion gap minus normal anion gap of 10 mEq per litre) can be compared to a calculation of the decrease in bicarbonate concentration (normal bicarbonate of 25 mEq per litre minus the observed bicarbonate concentration). If this relationship is ap- proximately 1:1, then it is likely that the acidotic disorder is due to an unmeasured acid anion. However, if the rise in anion gap is greater than the fall in bicarbonate, a process raising the bicar- bonate concentration such as a metabolic alkalosis coexists. This combination might be seen in a patient who is vomiting and has diabetic ketoacidosis. If the fall in bicarbonate concentration ex- ceeds the rise in anion gap, then the second process is most likely a hyperchloraemic acidosis. Metabolic acidosis In diagnosing the cause of metabolic acidosis, calculation of the serum anion gap can distinguish hyperchloraemic acidosis (Box 12.11.4) from common organic anion acidosis (Box 12.11.3), some overproduced in the body, others caused by ingestion of toxic substances. These conditions are discussed later in this chapter. Metabolic alkalosis Metabolic alkalosis is generally divided into two categories based on its responsiveness to chloride. Chloride-​responsive metabolic alkal- osis is associated with ECF and chloride depletion and is seen in cases of gastric fluid loss and diuretic use. As is the case of hyperchloraemia versus anion gap acidosis, a diagnostic clue in metabolic alkalosis comes from the serum electrolytes. Bicarbonate is increased with a corresponding fall in serum chloride (hypochloraemic alkalosis, Box 12.11.5). Chloride-​unresponsive metabolic alkalosis is seen in patients with ECF expansion in conditions such as primary aldos- teronism and hypokalaemia. These conditions are discussed later in this chapter. Acid–​base disorders as disturbances of chemistry of the extracellular fluid That the normal H+ concentration is in the range of 40 nM, while the bicarbonate concentration is in the 25 mM range, is an indication that protons are involved in many reactions within the body, including with water and many buffers such as phosphate, haemoglobin, and the amino groups on many proteins (carbamino compounds). Use Box 12.11.3  Causes of high anion gap metabolic acidosis • Lactic acidosis:

—​ Severe illness

—​ Sepsis

—​ Shock

—​ Seizures

—​ Malignancy

—​ Drugs

—​ Metformin

—​ Nucleoside reverse transcriptase inhibitors • Uraemia • Rhabdomyolysis • Ketoacidosis:

—​ Diabetes

—​ Ethanol (alcohol)

—​ Starvation • Poisoning:

—​ Methanol

—​ Ethylene glycol

—​ Propylene glycol

—​ Toluene (glue sniffer) hippurates • Drugs:

—​ Salicylates

—​ Iron

—​ Isoniazid •​ Pyroglutamic acid (5-​oxoprolinuria)

—​ Acetaminophen (paracetamol) •​ d-​Lactic acid Box 12.11.4  Causes of hyperchloraemic acidosis and their diagnosis Chloride poor losses Renal losses • Diuretics with high cation content:

—​ Acetazolamide

—​ Potassium-​sparing • Renal tubular acidosis:

—​ Early only in proximal

—​ Chronic in distal • Nonbicarbonate anion:

—​ Ketoacidosis

—​ Hippurate anion Gastrointestinal losses • Small bowel losses • Most diarrhoea Chloride rich gains • 0.9%, 0.45% saline • Ringer’s lactate Urinary electrolytes • Suggest renal cause of acidosis:

—​ UNa + UK –​ UCl is positive

—​ Low urinary NH4Cl • Suggest nonrenal cause of acidosis:

—​ UNa + UK –​ UCl is negative

—​ High urinary NH4Cl

12.11  A physiological approach to acid–base disorders 2185 of the standard bicarbonate approach relying on the HCO3−/​Pco2 system to describe acid–​base phenomena depends on the isohydric principle: that a single H+ concentration is in equilibrium with mul- tiple other buffer pairs. Alternative approaches involve consider- ation of electrostatic forces requiring electroneutrality between all cations and anions in the ECF giving dependency of HCO3− and H+ to concentrations of strong electrolytes such as sodium, potassium, and chloride. Pertinent to the approach taken in this chapter, one could apply either theory because all that is required is disturbance of ionic species within the ECF. Since clinical acid–​base disorders are defined in terms of blood chemistry, and because plasma volume is part of the total ECF volume, the approach taken here is to describe the ways by which electrolytes including hydrogen and bicarbonate ions can be intro- duced into or removed from the ECF. As shown in Fig. 12.11.1, the boundaries of the ECF are cell membranes at the interface of intracellular water and the interstitial fluid compartment of the ECF or specialized epithelial cells that line compartments known as transcellular spaces. The transcellular spaces are outside cell water, so fall into a special category of ECF, and their contents reflect solute exchanges with the interstitial space. This chapter will approach acid–​base disorders through understanding these cell membrane boundaries and their transport functions. We start with mechanisms by which acid–​base disturbances ori- ginate through transfer of electrolytes such as sodium, potassium, chloride, hydrogen, and bicarbonate across epithelial cell mem- branes that line transcellular spaces within the ECF. Transcellular fluids include the lumina of the entire gastrointestinal tract, the kidney tubules, and sweat gland ducts. These particular epithe- lial lined spaces have a common feature in having an outlet to the external world. Other transcellular spaces are the pleural and peri- toneal space and cerebrospinal fluid. These spaces reflect systemic acid–​base conditions through equilibration rather than cause them because they are not in continuity with the external environment. A second mechanism for introduction of acids or bases into or out of the ECF involves movement of ions between cells and the extracellular space. Organic acid anions (HA) such as ketoacids and lactic acids produced in the liver, muscle, or other tissues, and metabolites of toxins absorbed from the gastrointestinal tract, undergo such internal transfers across cell plasma membranes as they enter the ECF. These examples constitute the anion gap acidoses because the anion (A−) entering the ECF is not chloride. When the same anions, such as lactate and citrate, are ingested as the sodium or potassium salt, taken up by liver cells, and oxi- dized, it is bicarbonate that is transported to the ECF, which may cause metabolic alkalosis if production exceeds clearance. It should be noted that respiratory disorders also involve trans- port of carbon dioxide into or out of the ECF. Carbon dioxide generated by cell respiration diffuses into venous plasma before entering red blood cells en route to the lungs, where the process is reversed as carbon dioxide diffuses from red cell to plasma to lung interstitium to cells lining the airspaces prior to alveolar ventilation. Acid–​base disorders associated with transport across epithelial cells lining transcellular spaces The following discussion emphasizes ion transport processes of the plasma membranes of epithelia in various organs. The first to be examined are renal tubular epithelial cells. The reason be- hind starting with the kidney is that the same cell transporters that regulate acid–​base in normal physiology may initiate acid–​base disturbances and compensate for disturbances that are generated elsewhere in the body. Primary renal disorders as well as the com- pensations for respiratory and metabolic disturbances can be con- sidered together. Box 12.11.5  Causes of hypochloraemic alkalosis and their diagnosis Renal causes of hypochloraemia Diuretics and renal channelopathies • NKCC:

—​ Furosemide, bumetanide

—​ Bartter’s syndrome • NaCl:

—​ Thiazides

—​ Gitelman’s syndrome Nonrenal causes of hypochloraemia • Gastrointestinal:

—​ Gastric

—​ Congenital chloridorrhoea

—​ Infection • Skin:

—​ Cystic fibrosis Urinary electrolytes • Suggest renal cause of alkalosis:

—​ Urinary electrolytes reveal Cl− loss

—​ UNa + UK –​ UCl is low or negative • Suggest nonrenal cause of alkalosis:

—​ Urinary electrolytes reveal low Cl−

—​ UNa + UK –​ UCl is high Transcellular fluids Extracellular fluid Interstitial fluid Components of total body water Intracellular fluid P l a s m a Fig. 12.11.1  Body fluid compartments.

section 12  Metabolic disorders 2186 Acid–​base disorders and the kidney Glomerular filtration utilizes mechanical energy of the cardiac-​ generated blood pressure to form a glomerular ultrafiltrate that requires modification by the renal tubules before being excreted as urine. The role of the tubules is primarily to reclaim necessary fluids, electrolytes, and solutes while allowing the elimination of ap- propriate quantities of waste substances. The process accounts for as much as 15% of the body’s energy expenditure in the form of high-​ energy phosphates. Oxygen consumption is high, particularly in the renal cortex, and is related to these transport functions. In normal circumstances, glomerular filtration rate (GFR) is in the order of 180 litres/​day and the amount of freely filtered solutes can be calculated as the product of GFR and the concentration of the solute in the arterial plasma. For bicarbonate, the quantity might be more than 4500 mEq/​day. The uncontrolled excretion of even a small quantity of this filtrate could prove fatal. The proximal tubule The first tubular segment to confront this large filtrate is the prox- imal tubule (Fig. 12.11.2). The early proximal tubule has abundant brush-​border membranes on the apical surface allowing for a large surface area for reabsorption. As shown in Fig. 12.11.2a, the luminal fluid remains isosmotic to plasma and accounts for approximately 50% of sodium and water reabsorption, but 85% of filtered bicar- bonate is reclaimed by the proximal tubule. The normal process of fluid resorption involves many steps, as shown in Fig. 12.11.2b. There must be adequate ATP production by proximal tubule mitochondria to fuel the sodium potassium ATPase which pumps sodium into the extracellular space in exchange for potassium. The consequence is a low intracellular sodium con- centration and a cell-​negative electrical potential difference across the plasma membranes. A  sodium electrochemical gradient thus formed favours entry of sodium from the lumen into the cell. Many transporters utilize the energy of the sodium gradient to cotransport important solutes such as glucose and amino acids back into the cell. The luminal secretion of protons from the cell occurs by exchange with sodium entering the cell (sodium–​hydrogen ex- change). Once the proton is in the lumen of the proximal tubule it can combine with a filtered bicarbonate ion, forming carbonic acid. The dehydration of carbonic acid to water and carbon dioxide is kinetically favoured by carbonic anhydrase IV in the brush-​border membrane. Carbon dioxide can then re-​enter the cell where it com- bines with hydroxyl ions left in excess in the cell with the secretion of hydrogen into the lumen. This reaction of hydroxyl and carbon dioxide to form bicarbonate is catalysed by intracellular carbonic anhydrase II. The bicarbonate thus formed, representing filtered bicarbonate, can then pass across the basolateral membrane to the extracellular space via a sodium bicarbonate cotransporter (NBC) with a 1:3 stoichiometry. This electrogenic ratio of sodium to bicar- bonate provides enough driving force for completing proximal bi- carbonate reabsorption. Another way of viewing NBC function is that it protects against cellular alkalinization. Bicarbonate reabsorp- tion would be decreased as interstitial fluid bicarbonate concentra- tion is elevated when bicarbonate is ingested and the gradient for bicarbonate to exit the cell diminished. This is one mechanism by which bicarbonaturia is achieved when serum bicarbonate levels go up, even minimally. The net result of proximal bicarbonate reabsorption is a luminal pH decrease by the end of the proximal tubule to approximately 6.5. The reduced delivery of bicarbonate to the more distal segments of the tubule, which have lower capacity for bicarbonate absorption, al- lows completion of the process of reabsorption in regulated fashion. Proximal tubular reabsorption of sodium and HCO3− is regulated by hormones such as angiotensin II and catecholamines (which in- crease) and by parathyroid hormone (which decreases) the exchange of sodium and hydrogen. Other factors that regulate the activity of sodium–​hydrogen exchange and sodium bicarbonate cotransport include hypokalaemia and hypercapnia, which increase the process. Hypocapnia has the opposite effect, decreasing proximal sodium bi- carbonate reabsorption. Another important function of the proximal tubule in acid–​base balance is the mitochondrial production of ammonia from glu- tamine and glutamate. The ammonia formed can get into the urine by diffusion into the lumen or by counter-​transport with sodium via the sodium–​hydrogen exchanger. Basolateral glutamine uptake and ammonia production by glutaminase activity is increased in both respiratory and metabolic acidosis and will increase urinary net acid excretion as ammonium chloride. While ammonia is pro- duced in the proximal tubule, regulation of how much is ‘trapped’ and excreted as ammonium is a function of the distal nephron where urine pH falls to a value as low as 5. Hypokalaemia also increases ammonia production. Role of the proximal tubule in metabolic acidosis Considering the many steps involved in proximal HCO3− reabsorp- tion, it is understandable that an abnormality at any one of the steps could result in delivering an amount of bicarbonate exceeding the capacity for distal nephron reabsorption. Bicarbonaturia and an alkaline urine would then result and metabolic acidosis develop. Because the excretion of bicarbonate is with sodium and potassium and less chloride, the ECF will express hyperchloraemic metabolic acidosis. These syndromes are known collectively as proximal renal tubular acidoses. Proximal renal tubular acidosis (type 2)  Proximal renal tubular acidosis (type 2)  is characterized by a decreased threshold for Fig. 12.11.2  The early proximal tubule: (a) luminal content; (b) cellular transport mechanism. CA, carbonic anhydrase.

12.11  A physiological approach to acid–base disorders 2187 bicarbonate reabsorption. HCO3–​ wasting and concomitant urinary losses of potassium occur until a lower level of serum bicarbonate re- duces the filtered HCO3–​ to a level at which the combined remaining function of the abnormal proximal tubule and low capacity distal nephron can completely reabsorb filtered bicarbonate. At that point, the urine becomes acid (pH <5.3) and net acid production equals net acid excretion, with a steady-​state low plasma HCO3−. Attempts to raise the plasma bicarbonate to normal may be difficult because the added bicarbonate will promptly enter the urine (high fractional excretion of bicarbonate), unnecessary because once the low bicar- bonate is achieved, a balance of acid produced and excreted occurs, and risky because the more bicarbonate excreted in the urine, the greater is potassium loss. However, correcting acidosis with bicar- bonate replacement is especially necessary in growing children. Isolated proximal renal tubular acidosis may result from muta- tions of specific transporters of the proximal tubule, such as the sodium–​hydrogen exchanger 3 (NHE3) or NBC, or from hereditary deficiency of carbonic anhydrase isoforms or carbonic anhydrase inhibitors. More commonly, proximal renal tubular acidosis is as- sociated with Fanconi’s syndrome or generalized proximal tubule dysfunction. Causes include genetic diseases such as glucose-​6-​ phosphatase deficiency, cystinosis, hereditary fructose intolerance, and Wilson’s disease. Acquired conditions such as multiple myeloma and Sjögren’s syndrome should be considered in an adult patient. For further discussion see Chapter 21.15. Primary hyperparathyroidism results in proximal renal tubular acidosis and hypophosphataemia secondary to inhibition of Na/​H exchange (NHE3) and sodium phosphate cotransport in the prox- imal tubule by parathyroid hormone through cyclic AMP. The Cl−/​ phosphate ratio in plasma may be elevated. Hyperparathyroidism is one of the few causes of metabolic acidosis and hypercalcaemia. Drug toxicity with aminoglycosides, cisplatin, and ifosfamide may cause proximal tubule dysfunction. The antiretroviral drug tenofovir, a nucleotide analogue reverse transcriptase inhibitor used in the treatment of human immunodeficiency virus-​1 (HIV-​1) and hepatitis B, is a cause of Fanconi’s syndrome. The syndrome also may be seen after kidney transplantation. As shown in Fig. 12.11.3, the late proximal tubule receives fluid that is low in bicarbonate and high in chloride due to preferential sodium bicarbonate reabsorption earlier. Chloride may then enter the proximal tubule cell in exchange for base, or chloride—​favoured by its high luminal to interstitial fluid concentration gradient—​ may pass through the paracellular route creating a positive luminal voltage that would increase paracellular sodium absorption. When filtered chloride is high at the outset (filtrate of plasma and ECF), and bicarbonate low, no such chloride gradient develops. A clinical corollary is that carbonic anhydrase inhibitors and hyperchloraemic acidoses per se decrease the lumen to interstitial fluid chloride gra- dient and therefore result in diuresis of sodium chloride. The metabolic acidosis that develops in chronic kidney disease is related to the failure to produce ammonia, thereby limiting the amount of net acid that can be excreted in the urine. If that amount is smaller than acid production within the body, then metabolic acidosis will develop. Many organic and inorganic anions, such as phosphate and sulphates, are retained at GFRs of less than 25 ml/​ min and constitute an increased anion gap in association with the metabolic acidosis. Patients with chronic kidney disease will not normally compensate for respiratory acidosis or nonrenal metabolic acidosis because they do not have the ability to further increase am- monia production. Role of the proximal tubule in metabolic alkalosis An increase in proximal sodium bicarbonate reabsorption allows less bicarbonate to flow distally and will maintain the existing level of plasma bicarbonate. In the case of volume (sodium) depletion with increased angiotensin II, or during hypokalaemia, it becomes diffi- cult to excrete an increased bicarbonate load. In order to eliminate any excess of filtered bicarbonate, there must be normal sodium and potassium balance. Metabolic alkalosis, if present, is therefore main- tained until replacement of potassium and extracellular volume is achieved (discussed later in this chapter). Role of the proximal tubule in compensation for respiratory disorders In respiratory acidosis, increased ammonium excretion allows for a rise in plasma bicarbonate because elimination of hydrogen ions that does not result in bicarbonate reabsorption instead results in ‘new’ bicarbonate. Though the final formation of ‘new’ bicarbonate is a distal function involving hydrogen secretion, the ammonia ne- cessary for the process is made in the proximal tubule and stimu- lated by hypercapnia. The role of a high Pco2 in respiratory acidosis to increase sodium bicarbonate reabsorption in the proximal tubule is the basis of the maintenance of high bicarbonate concentrations in compensated respiratory acidosis. In patients with compensated chronic respiratory acidosis who are acutely ventilated to normal Pco2, high bicarbonate levels remain until chloride is replaced (posthypercapnic alkalosis). Taken together the compensation for respiratory acidosis re- quires the generation of new bicarbonate and increased proximal reabsorption of filtered sodium bicarbonate. By contrast, the renal compensatory mechanism in respiratory alkalosis is decreased so- dium bicarbonate reabsorption in the proximal tubule causing alka- line urine and a low plasma bicarbonate concentration. Thick ascending limb of Henle’s loop As the tubular fluid leaves the proximal tubule and enters the loop of Henle, the process of bicarbonate reabsorption continues in the Late proximal tubule Fig. 12.11.3  Late proximal tubule. (a) Luminal content; (b) cellular transport mechanism.

section 12  Metabolic disorders 2188 thick ascending limb. Luminal sodium hydrogen exchangers and basolateral sodium bicarbonate cotransporters reabsorb approxi- mately 5% of the bicarbonate filtered. As shown in Fig. 12.11.4, the thick ascending limb has a major role in sodium chloride reabsorp- tion for maintenance of extracellular volume, osmoregulation (both concentrating and diluting mechanisms), and divalent cation re- absorption (calcium and magnesium). Role of the thick ascending limb in hypochloraemic alkalosis The sodium, potassium, two-​chloride cotransporter, NKCC, can be inhibited by drugs such as furosemide and bumetanide and by hypercalcaemia through a unique mechanism of the basolateral calcium receptor. Patients with hypercalcaemia lose salt in the urine. The effect of hypercalcaemia is to inhibit the NKCC trans- porter, much like furosemide. This provides a mechanism for the frequently observed association between hypercalcaemia and al- kalosis. Inhibition of chloride reabsorption in this segment leads to hypochloraemic metabolic alkalosis. Bartter’s syndrome is an autosomal recessive disorder associated with extracellular volume depletion and excessive urinary chloride loss that results in hypokalaemia and hypochloraemic metabolic alkalosis. Secondary increases of plasma renin and aldosterone occur, as does renal juxtaglomerular cell hyperplasia. The syndrome resembles the effects of furosemide on the thick ascending limb of Henle. Gene mutations in the NKCC cotransporter, the renal outer medullary K+ channel (ROMK), and Cl–​ channels (Bartin) have been described. Because calcium reabsorption occurs in the thick ascending limb of Henle, Bartter’s syndrome, like furosemide, causes hypercalciuria and nephrocalcinosis, as well as polyuria due to decreased urinary concentrating ability. For further discussion see Chapters 21.2.2 and 21.16. Distal tubule As shown in Fig. 12.11.5, the distal tubule receives input from the loop of Henle and plays an important role in continued salt re- absorption to maintain extracellular volume and in urinary dilution by separating salt from water in the absence of antidiuretic hormone. Furthermore, the segment is responsive to parathyroid hormone which increases calcium absorption in the distal tubule. Calcium reabsorption is increased when the sodium chloride cotransporter (NCC) is inhibited. NCC is often referred to as the thiazide-​sensitive sodium chloride transporter because of its inhibition by thiazide diuretics. Role of the distal tubule in hypochloraemic alkalosis Like furosemide, thiazide diuretics cause hypochloraemic alkal- osis due to urinary chloride loss. Gitelman’s syndrome is an auto- somal recessive cause of extracellular volume depletion, urinary chloride wasting, and hypokalaemic metabolic alkalosis. It is due to inactivating mutations in the SLC12A3 gene encoding the thiazide-​ sensitive NCC cotransporter of the renal distal tubule. Urinary con- centrating ability is preserved because the cortical distal tubule has no effect on the interstitial concentrating gradient of the medulla, and patients are hypocalciuric because decreased sodium chloride reabsorption in the distal tubule is associated with a decrease in calcium excretion (hence its usefulness in hypercalciuric states). Hypomagnesaemia may also be severe. Patients who present with hypokalaemic metabolic alkalosis with normal or low blood pressure and have urinary chloride concen- trations greater than 25 mEq/​litre may be taking diuretics such as furosemide or thiazides surreptitiously; a diuretic screen can docu- ment the presence of the drug. If the screen is negative, Bartter’s or Gitelman’s syndrome should be considered. Bartter’s syndrome is less common, usually more severe, and presents in young patients. The presence of hypercalciuria favours Bartter’s syndrome, whereas hypocalciuria and hypomagnesaemia suggest Gitelman’s syndrome. The hypokalaemia in both syndromes is due to increased sodium delivery to more distal collecting tubule segments where K+ se- cretion occurs and depends on delivered sodium, and secondary hyperaldosteronism. For further discussion, see Chapter 21.2.2. Role of the distal tubule in hyperchloraemic acidosis In Gordon’s syndrome (pseudohypoaldosteronism type 1), in- creases in Na+ and Cl–​ reabsorption through increased activity of the distal thiazide-​sensitive NCC transporter leads to hypertension and hyperkalaemic, hyperchloraemic acidosis, volume expansion, and (consequently) low renin and aldosterone. The hyperkalaemia is due to decreased sodium delivery to more distal collecting tubule segments where K+ secretion occurs and depends on delivered so- dium, and aldosterone. For further discussion see Chapter 21.15. Thick ascending limb of Henle Fig. 12.11.4  Thick ascending limb of Henle. (a) Luminal content; (b) cellular transport mechanism. CaR, calcium receptor Distal convoluted tubule Fig. 12.11.5  Distal tubule. (a) Luminal content; (b) cellular transport mechanism.

12.11  A physiological approach to acid–base disorders 2189 Collecting duct The final 10% of filtered bicarbonate reabsorption occurs in the renal collecting duct. The mechanism by which this occurs involves the coordinated effort of two types of cells: the principal cell and the intercalated cell (Figs. 12.11.6 and 12.11.7). As in other nephron segments, principal cell basolateral sodium–​potassium ATPase lowers intracellular sodium and provides a gradient for inwardly directed sodium movement. In this cell type, sodium enters through the epithelial sodium channel known as ENaC. The process of so- dium reabsorption renders the lumen of the collecting duct elec- tronegative. If no other ion transport occurred, the lumen-​negative charge would increase to a degree that would shut down further so- dium reabsorption. However, if either a cation enters the lumen or an anion leaves the lumen, the negative charge would be neutralized or ‘compensated’ allowing more sodium to be reabsorbed. Charge compensation for the negative lumen occurs by potassium secretion through luminal potassium channels or by selective anion reabsorp- tion. Another mechanism for charge compensation is proton secre- tion by the electrogenic proton ATPase of adjacent α-​intercalated cells shown in Fig. 12.11.7. The secreted protons can combine with the small amount of re- maining filtered bicarbonate delivered from proximal sites to re- form luminal carbon dioxide which then enters the intercalated cell where—​in the presence of intracellular carbonic anhydrase—​it re- forms bicarbonate for return to the ECF by a chloride–​bicarbonate exchanger. To the extent that the secreted proton combines instead with ammonia or phosphate in the lumen, the hydroxyl left over within the cell (as the proton is secreted), combines with ambient carbon dioxide to recreate bicarbonate which returns to the ECF as ‘new’ bicarbonate to replace the extracellular bicarbonate initially lost by acid produced within the body. In other circumstances, particularly on an alkaline ash diet (con- sisting mainly of fruit, vegetables, and milk) where bicarbonate needs to be eliminated, the β-​intercalated cell (with reversed polarity of proton pump and chloride–​bicarbonate exchange) predominates, allowing for bicarbonate secretion into the lumen in exchange for chloride (Fig. 12.11.8). Coordinated function of the principal and intercalated cells involves sodium reabsorption by ENaC, lumen negativity, and increased distal acidification of the urine. ENaC is regulated by aldosterone through the intracellular mineralocorticoid re- ceptor (MR) in the principal cell. That receptor can be activated by either aldosterone or cortisol. Cortisol is prevented from nor- mally activating the MR by virtue of 11 β-​hydroxysteroid de- hydrogenase type 2, which degrades cortisol to inactive cortisone. Aldosterone production is increased independently by adrenal se- cretion due to angiotensin II and hyperkalaemia. The dual mech- anism for aldosterone production allows for potassium secretion in volume expanded states and potassium conservation in hypo- volaemia. The so-​called aldosterone paradox, which refers to the ability of the kidney to retain sodium and chloride with minimal potassium secretion in the presence of volume depletion and yet to maximize potassium excretion without sodium retention in hyperkalaemia, may be mediated by decreased activity of potas- sium channels in the collecting duct when angiotensin II is pre- sent and increased activity when angiotensin II is low. When AII is high, more sodium is reabsorbed upstream from the potassium-​ secreting principal cells, decreasing net sodium for potassium exchange. Cortical collecting duct Fig. 12.11.6  Collecting duct principal cell. (a) Luminal content; (b) cellular transport mechanism. 11β OHSD, 11β hydroxysteroid dehydrogenase; AldoR, aldosterone receptor; ENaC, epithelial
sodium channel. Cortical and medullary collecting duct: α-intercalated cell Fig. 12.11.7  Collecting duct α-​intercalated cell. (a) Luminal content; (b) cellular transport mechanism. CA, carbonic anhydrase. Cortical collecting duct: β-intercalated cell Fig. 12.11.8  Collecting duct β-​intercalated cell. (a) Luminal content; (b) cellular transport mechanism. CA, carbonic anhydrase.

section 12  Metabolic disorders 2190 From this analysis, it is apparent that renal acidification can be- come abnormal at many different steps within either the principal or intercalated cells of the collecting duct. If the disease results in decreased hydrogen excretion, the abnormalities are collect- ively known as distal renal tubular acidoses. If the disease instead leads to increased acid excretion, then metabolic alkalosis ensues. Disorders of the principal cell typically cause hyperkalaemia with acidosis or hypokalaemia with alkalosis because of the similar direction of potassium and hydrogen ion secretion. However, in- creased ENaC activity syndromes can cause extracellular volume expansion or be the appropriate response to extracellular volume depletion. Similarly, decreased ENaC activity can be the cause of extracellular volume depletion or the appropriate response to extra- cellular volume expansion. Role of the collecting duct principal cell in hyperkalaemic, hyperchloraemic acidosis (distal/​type 4 renal tubular acidosis) This combination of abnormalities suggests dysfunction of the cor- tical collecting duct, where acidification of urine and disorders in potassium secretion may occur. Some patients with high blood po- tassium and hyperchloraemic acidosis can lower urinary pH below 5.3, whereas others appear to have defects in both potassium balance and urinary acidification. Hyperkalaemia itself may worsen meta- bolic acidosis since the high potassium outcompetes ammonia reabsorption at the K-​site of NKCC in the thick limb, thereby decreasing NH3 accumulation by countercurrent multiplication in the medullary interstitium. Causes include hyporeninaemic hypoaldosteronism, as seen in diabetic renal disease; other tubulointerstitial diseases, usually with some renal impairment; sickle cell anaemia; or the use of drugs such as renin inhibitors, β-​blockers, and nonsteroidal anti-​inflammatory drugs, where cyclooxygenase-​2 inhibition of the cells of the macula densa result in decreased renin secretion. Low renin and aldosterone levels can be found in cases of volume expansion with hypertension. Ciclosporin and tacrolimus may lead to decreased electrical driving forces for K+ and H+ secretion. Hyperkalaemic acidosis with elevated renin and low aldosterone is found in adrenal insufficiency, isolated hypoaldosteronism, and the use of angiotensin-​converting enzyme inhibitors, and angio- tensin II receptor blockers. High renin and aldosterone levels are anticipated when the renal collecting duct cell is insensitive to aldos- terone, as with urinary tract obstruction, sickle cell anaemia, amyl- oidosis, and systemic lupus erythematosus. Inhibition of aldosterone action with spironolactone, eplerenone, or new nonsteroidal MR receptor antagonists may cause hyperkalaemic acidosis, as does ENaC inhibition by amiloride, triamterene, tri- methoprim, and lithium. Pseudohypoaldosteronism type 1 is due to autosomal recessive, inactivating mutations of the Na+ channel ENaC, whereas autosomal dominant pseudohypoaldosteronism type 1 is due to mutations of the MR. Both cause hypovolaemia, metabolic acidosis, and hyperkalaemia with secondary increases in renin and aldosterone. Hyperchloraemic metabolic acidoses with a normal or ele- vated potassium concentration can develop as a result of the addition of chloride salts such as NaCl, KCl, CaCl2, NH4Cl, ar- ginine and lysine hydrochlorides, or HCl itself. If the quantity of Cl–​ introduced exceeds the ability of the kidney to eliminate NH4Cl in urine, hyperchloraemia will develop. Electroneutrality is maintained by a decrease in the serum HCO3–​ concentration, and a hyperchloraemic acidosis ensues. Renal production of NH3 increases in an attempt to improve HCl (NH4 Cl) excretion. Hyperkalaemia can occur because the acidaemia favours the exit of K+ from cells within the body. Acidaemia also inhibits K+ secretion in the renal collecting duct. Role of the collecting duct principal cell in hypokalaemic, hypochloraemic alkalosis with extracellular volume
expansion (hypertension) The renal conditions that cause metabolic alkalosis and volume expansion are due to a proportionately greater increase in Na+ re- absorption above that required to maintain a steady state of Na+ balance, rather than primary loss of the Cl–​ anion. As Na+ is reab- sorbed, electroneutrality is maintained by an increase in plasma HCO3–​. The extracellular volume and plasma Na+ concentration may be increased, Cl–​ appears in urine, and hypochloraemia is not present. In the kidney, the loss of net acid as NH4Cl in excess of the acid produced generates a metabolic alkalosis in which the ‘new’ bicarbonate is due to proton secretion by the distal nephron H+-​ ATPases. The H+ combines with NH3 to form NH4+ in urine. The increased plasma HCO3–​ will be filtered, but in the absence of a stimulus to increase proximal HCO3–​ reabsorption, the HCO3–​ will flow distally to be reabsorbed by the increased H+ secretion of the collecting duct. At first, the alkalosis is mild, but increased cor- tical collecting duct Na+ reabsorption leads to increased K+ secretion and hypokalaemia. Hypokalaemia increases the capacity for prox- imal HCO3–​ reabsorption, thereby opposing the effect of volume expansion, so that distal delivery of HCO3–​ decreases. The higher than normal distal H+ secretion titrates urinary buffers so further ‘new’ HCO3–​ is formed and the alkalosis worsens. Metabolic alkal- oses of this type are sustained by hypokalaemia and the alkalosis can be treated by potassium replacement. Specific causes of renal alkalosis with hypokalaemia, volume ex- pansion, and hypertension can be classified according to levels of renin and aldosterone. Primary increases in renin with secondary increases in aldosterone can be seen in patients with unilateral renal artery stenosis, renin-​secreting tumours of the kidney, and malig- nant hypertension. Low renin and elevated aldosterone levels are characteristic of primary hyperaldosteronism from adrenal adenoma or hyperplasia. If the aldosterone secretion is autonomous, a high salt intake will worsen the hypokalaemia and the alkalosis because more sodium will be delivered distally because of the volume expansion, and al- dosterone, usually shut off in such circumstances, remains active causing increased potassium secretion. Low renin and low aldosterone are seen when volume expan- sion is due to a high cortisol level in Cushing’s syndrome and the hypercortisolism of adrenocorticotropic hormone-​secreting tu- mours. Inhibition of the intracellular enzyme 11β-​hydroxysteroid dehydrogenase, which normally inactivates cortisol to form cor- tisone in the principal cell, will also result in low renin levels and low aldosterone levels, as endogenous cortisol generates the hypokalaemic alkalosis. Both genetic mutations (the ap- parent mineralocorticoid excess syndrome) and an excess con- sumption of glycyrrhizic acid (found in liquorice or anisette) are causes of this enzyme block. Another cause of hypertension with hypokalaemic alkalosis but with low renin and aldosterone

12.11  A physiological approach to acid–base disorders 2191 levels is Liddle’s syndrome, in which an activating mutation in the cortical collecting duct Na+ channel (ENaC) leads to increased Na+ reabsorption. Hypokalaemic metabolic alkalosis may also develop without volume expansion when a nonreabsorbable anion is presented to the cortical collecting duct lumen. Nitrates, sulphates, and certain anti- biotics such as nafcillin, carbenicillin, and ticarcillin may obligate K+ and H+ secretion as Na+ is reabsorbed. Topical administration of silver nitrate to burn victims may result in alkalosis. Secondary hyperaldosteronism associated with high levels of renin contributes to the hypochloraemic alkalosis associated with chloride losses from diuretics or other states of chloride and extra- cellular volume depletion. When the secondary hyperaldosteronism is due to volume depletion in a setting of metabolic acidosis (such as proximal tubular acidosis), hypokalaemia but not metabolic alkal- osis is observed. Role of the collecting duct intercalated cell in hypokalaemic, hyperchloraemic acidosis (distal/​type 1 renal tubular acidosis) In distal renal tubular acidosis (type 1), failure to excrete NH4Cl leads to an inability to excrete adequate net acid, thereby leading to continuous retention of acid in the body. The degree of acidaemia is often severe, with pH reaching values as low as 7.2, whereas urine pH usually exceeds 5.3. Kindreds have been described in which mutations in genes for the distal vacuolar H+-​ATPase cause an autosomal recessive distal renal tubular acidosis with deafness. Mutations resulting in defective Cl/​ HCO3 exchange protein (AE1) have been linked to an autosomal dominant form of distal renal tubular acidosis. Distal renal tubular acidosis is also associated with autoimmune disorders, including systemic lupus erythematosus and Sjögren’s syndrome, and genetic diseases, including sickle cell anaemia, Wilson’s disease, Fabry’s disease, cystic kidney diseases, and her- editary elliptocytosis. Hypercalciuria and hyperoxaluria may cause distal renal tubular acidosis; nephrocalcinosis and nephrolithiasis may be present. Increased proximal tubular citrate reabsorption as a consequence of the chronic acidosis leads to hypocitraturia, a risk factor for calcium nephrolithiasis. A chronically alkaline urine is a risk for pure CaHPO4 stones, and when the latter are found distal renal tubular acidosis should be suspected. Amyloidosis may be manifested as severe acidaemia and other tubular dysfunction, including nephrogenic diabetes insipidus. Chronic tubulointerstitial diseases of the kidney, including reflux nephropathy and urinary ob- struction, may result in renal tubular acidosis with hypokalaemia or hyperkalaemia. Acute tubulointerstitial nephritis may also result in renal tubular acidosis. Drugs such as amphotericin B can cause hypokalaemic distal renal tubular acidosis. Topiramax, used for migraines, is a carbonic anhydrase inhibitor that may cause mixed proximal and distal renal tubular acidosis. Distinguishing proximal and distal renal tubular acidosis  In contrast to proximal renal tubular acidosis, distal renal tubular acidosis (type 1) is generally a more severe metabolic disorder that may be accompanied by hypercalciuria, nephrocalcinosis, calcium phosphate kidney stones, and bone disease that includes rickets in children and osteomalacia in adults. It is necessary to treat distal acidosis because of relentless acid retention, and—​compared to proximal renal tubular acidosis—​easier to treat with enough bicar- bonate to cover the usual production rate of acids, and safer to treat because potassium improves with return to normal pH. Proximal and distal renal tubular acidoses usually can be dis- tinguished by clinical evaluation (Table 12.11.2). Helpful find- ings include the presence of a urine pH greater than 5.3 in distal but not proximal renal tubular acidosis during acidaemia; a frac- tional excretion of bicarbonate as high as 10 to 15% during bicar- bonate loading in proximal renal tubular acidosis; and the lowering of serum potassium upon correction of proximal but not distal tubular acidosis. In order to know what the kidney is doing with respect to elec- trolyte excretion, it is necessary to evaluate the urine chemistry. Considering sodium, potassium, and chloride in the urine, sodium and potassium concentrations relative to chloride that are dispro- portionate to that which exists in the ECF can predict whether the loss of that urine will have an acidifying or alkalizing effect on the ECF. For example, if the sum of the sodium and potassium concen- trations greatly exceeds the chloride concentration in urine, there will be a tendency to acidify the body fluids. If there is a meta- bolic acidosis in the blood, then such an excretion pattern by the kidney suggests that the kidney is the culprit in the generation of the acidosis. The diagnosis would then be called renal tubular acidosis and the unmeasured anion accompanying sodium and Table 12.11.2  Comparison of renal tubular acidoses Proximal (type 2) Classic distal (type 1) Hyporeninaemic hypoaldosteronism (type 4) Common causes Ifosfamide NRTI (tenofovir, adefovir, cidofovir) Myeloma Sjögren’s syndrome SLE Amphotericin CKD plus: • DM, amyloid • obstruction • sickle cell • SLE NSAIDs Treatment Bicarbonate (large dose) Bicarbonate (1 mEq/​kg per day) K+-​lowering treatment: • Diuretics • Kayexalate • Low K diet • Mineralocorticoid CKD, chronic kidney disease; DM, diabetes mellitus; SLE, systemic lupus erythematosus; NSAIDs, nonsteroidal anti-​inflammatory drugs; NRTI, nucleoside reverse transcriptase inhibitors.

section 12  Metabolic disorders 2192 potassium might be bicarbonate. If, by contrast, chloride was ex- creted in concentrations disproportionately high to the sum of so- dium and potassium concentrations, then an alkalinizing effect on plasma would be predicted. If a systemic acidaemia were present, then this urinary pattern would be considered appropriate and compensatory, favouring a search for an extrarenal cause of meta- bolic acidosis such as diarrhoea or respiratory acidaemia. From the urinary point of view, the extra chloride in the urine would need be accompanied by an unmeasured cation, which we know would be ammonium. Role of the collecting duct in compensation for nonrenal metabolic and respiratory disorders Although the proximal tubule plays a major role in compensation for these disorders, intracellular acidification of intercalated cells by both metabolic and respiratory acidosis leads to increased exo- cytotic expression of proton ATPase on the luminal membrane, increasing urinary acidification. A maximally acid urine will trap more ammonia in the form of ammonium. Alkalinization of cells in metabolic and respiratory alkalosis has the opposite effect. Renal compensations for respiratory disturbances can also be in- ferred from urinary chemistry. For example, the renal compensation for respiratory acidosis should be the loss of urinary electrolytes in the pattern that would alkalinize the extracellular space, that is, loss of high chloride concentration relative to sodium and potassium. As pre- viously discussed, the cation accompanying chloride is ammonium. Acid–​base disorders and the gastrointestinal tract The organs of the gastrointestinal tract produce secretions that en- able the absorption of fluid, electrolytes, and organic solutes derived from the metabolism of protein, carbohydrate, and fat. Excessive production or loss of these secretions will disturb the economy of electrolytes and acid–​base equivalents in the ECF because the source of these secretions is the ECF. Acid–​base disturbances also develop when ingested quantities of chloride or bicarbonate salts exceed the ability of the kidney or in some cases liver to clear them from the circulation. Stomach In Fig. 12.11.9a, the gastric lumen is shown to contain HCl as well as sodium and potassium. The volume of gastric fluid produced per day is usually about 2 litres, but in disease states, particularly with bowel obstruction, the volume can increase markedly. Normally the fluid flows distally into the upper small intestine where hydrogen and pan- creatic bicarbonate combine to form carbon dioxide and water, while the sodium and chloride are absorbed along the small intestine. In that way, there is internal balance of acid–​base. However, a transient postprandial sequestration of HCl in the gastrointestinal tract leads to a phenomenon known as the alkaline tide, a transient alkalinization of the body fluids and the urine. Shown in Fig. 12.11.7b, the acid-​ secreting parietal cell of the stomach has potassium-​hydrogen ATPase on the apical side, which is responsible for the secretion of hydrogen ion into the stomach against its chemical gradient. Stomach pH may reach values as low as 1 to 2. Chloride is secreted by chloride channels. Role of the stomach in hypochloraemic metabolic alkalosis If gastric fluid is removed by a nasogastric tube or by vomiting, then the loss of chloride and the gain of bicarbonate within the extracellular space would result in hypochloraemia and metabolic alkalosis. With vomiting, the initiating or generating event is loss of HCl. As shown in Fig. 12.11.9b, secretion of HCl into the stomach lumen by the parietal cell is coupled to the absorption of HCO3–​ in exchange for chloride at the basolateral membrane. With vomiting, initial increases in serum HCO3–​ are filtered by the renal glomeruli and excreted in urine accompanied by Na+ and K+; volume depletion begins to develop as sodium is lost in the urine and chloride in the vomitus. Thus, extracellular volume depletion is not accompanied by low sodium excretion in metabolic alkalosis. As vomiting continues, extracellular volume depletion worsens, glom- erular filtration falls, and HCO3–​ filtration is limited. The volume depletion activates the renin–​angiotensin II–​aldosterone system and proximal tubule fluid and NaHCO3 reabsorption increase as a consequence. Distal nephron Na+ reabsorption increases under the influence of aldosterone, and that results in greater H+ secretion, thereby enhancing HCO3–​ reabsorption. These effects reduce renal Na+ loss but at the expense of maintaining the metabolic alkalosis. Significant K+ losses, which occur as a result of the bicarbonaturia and hyperaldosteronism, lead to hypokalaemia, which is due to renal, not gastrointestinal, losses as a consequence of attempts to maintain extracellular volume. Similar to initiating events in prox- imal renal tubular acidosis, the hypokalaemia during generation of gastric alkalosis is of distal nephron aetiology as bicarbonate leaves the proximal tubule in a volume-​depleted state of high aldoster- onism, thereby enhancing potassium secretion. The hypokalaemia further increases proximal NaHCO3 reabsorp- tion, distal H+ secretion, and K+ reabsorption via the proton-​potassium ATPase of the intercalated cell of the collecting duct, all at the ex- pense of further reabsorption of HCO3–​. At the new steady state after vomiting or nasogastric suctioning ceases, the paradoxical aciduria of metabolic alkalosis develops as HCO3–​ reabsorption is complete and the urine contains low levels of Na+, K+, and Cl–​. The patient may be hypovolaemic, hypokalaemic, and alkalaemic, but because Na+, K+, and acid–​base balance are intrinsically linked, life-​threatening volume depletion, K+ depletion, and alkalaemia are usually avoided. Small intestine As shown in Fig. 12.11.10a, the fluid delivered from the upper small bowel is alkaline and contains iso-​osmotic sodium, chloride, and Stomach Fig. 12.11.9  Stomach and gastric parietal cell. (a) Luminal content; (b) cellular transport mechanism. CA, carbonic anhydrase.

12.11  A physiological approach to acid–base disorders 2193 bicarbonate. Most ingested nutrients such as glucose and amino acids and much of the bicarbonate are absorbed in more proximal bowel, such as duodenum and jejunum, and water is absorbed os- motically. The bicarbonate comes predominately from pancreatic secretions, with some contribution of biliary secretion, while the chloride is primarily dietary and that remaining from gastric acid secretion. In Fig. 12.11.11a, the luminal contents of fluid in the ileum and proximal colon are shown. An important role of these late small in- testinal and early colonic segments is to absorb sodium, chloride, bicarbonate, and water. In Fig. 12.11.11b, a cell is shown dem- onstrating the mechanism by which this absorption takes place. The luminal membrane contains both sodium–​hydrogen exchan- gers and chloride–​bicarbonate exchangers oriented in such a way to allow entry of sodium and chloride into the cell in exchange for secreted hydrogen and bicarbonate: thus the mechanism for sodium chloride absorption is by ‘double exchange’. In jejunal cells that favour sodium bicarbonate absorption, the sodium–​ hydrogen exchanger functions without a chloride–​bicarbonate antiporter, while in some villous cells of ileum and large intestine, chloride–​bicarbonate exchange without accompanying sodium–​ hydrogen exchange is the mechanism for chloride absorption and bicarbonate secretion. Role of the small intestine in hyperchloraemic acidosis In patients who have small bowel malabsorption, as in inflamma- tory bowel disease, pancreatitis, or with an infectious gastroenteritis, large volumes of small intestinal losses result in high bicarbonate-​ containing diarrhoea. Patients with an ileostomy can lose large quantities of pancreatic secretions rich in bicarbonate resulting in hyperchloraemic metabolic acidosis because the large amounts of bi- carbonate present in the lumen being delivered from upstream seg- ments are greater than the chloride anion in a relative comparison to ECF sodium and chloride. In some situations where bowel motility is poor, large quantities of pancreatic secretions accumulate in the intestine creating a hyperchloraemic metabolic acidosis. On occa- sion, acidaemia will result from vomiting small bowel contents or by duodenal or jejunal drainage. Ileal segments are used as urinary diversion conduits to replace bladder function. In such a situation, particularly when there is obstruction to flow, the ileal segment maintaining its ‘double exchange’ properties can lead to the reabsorption of excessive quantities of chloride by the loop, resulting in hyperchloraemic acidosis. Diabetic patients receiving pancreas and renal transplantation in which the exocrine pancreas drains into the urinary bladder can develop severe metabolic acidosis with hyperchloraemia, and this is one of the reasons why bowel drainage of the exocrine pancreas is now the preferred surgical technique in pancreatic transplantation. Secretagogues such as vasoactive intestinal peptide (VIP), which is associated with neoplasms of the pancreas or sympathetic chain, cause large losses of HCO3–​ in stool, with a resulting hypokalaemic, hyperchloraemic acidosis. Role of the small intestine in hypochloraemic alkalosis From the previous discussion, one could predict the following: if double exchange is dysfunctional as would be the case in a mutation of the chloride–​bicarbonate exchanger, then the luminal fluid would contain large quantities of chloride. If that chloride was unable to be absorbed by the downstream large intestine, then stool chloride would be increased. In Zollinger–​Ellison syndrome, excessive gastrin-​induced gastric acid secretion may result in large volumes of acidic stool with high chloride content resulting in hypochloraemic alkalosis. Congenital chloridorrhoea (chloride-​losing diarrhoea) is an autosomal reces- sive disorder of defective intestinal, apical Cl/​HCO3 exchange asso- ciated with the downregulated adenoma (DRA) gene, so-​named for its decreased expression in some patients with villous adenomas or adenocarcinomas. Intestinal crypt cells secrete chloride across the apical mem- brane in association with the cystic fibrosis transmembrane con- ductance regulator (CFTR) and this process may be activated by secretogogues including neurohumoral substances, cyclic AMP, and certain toxins in infectious diarrhoea. Most diarrhoeal illnesses, including the secretory diarrhoeas, result in metabolic acidosis ra- ther than alkalosis, but the acid–​base disorder is determined by the electrolyte content of the stool. Diarrhoea does not cause metabolic Duodenum and proximal jejunum Fig. 12.11.10  Upper small intestine. (a) Luminal content; (b) cellular transport mechanism. Ileum and proximal colon Lumen Fig. 12.11.11  Lower small intestine and early colon. (a) Luminal content; (b) cellular transport mechanism. CA, carbonic anhydrase; ClC, chloride channel.

section 12  Metabolic disorders 2194 alkalosis unless the stool electrolyte relationship [Na+ + K+− Cl–​] is less than plasma HCO3−. Colon Fig.  12.11.12a represents the diminishing amounts of water and electrolytes in the distal colon. In Fig. 12.11.12b a colonic cell is shown in which, as in the renal collecting duct, aldosterone increases sodium absorption via the epithelial sodium channel and potassium is secreted into the lumen. Role of colonic diarrhoea and hyperchloraemic acidosis Diarrhoeal disorders affecting these areas usually cause hyper­ chloraemic acidosis as unabsorbed sodium and potassium are lost with organic anions of bacterial origin. Chloride and bicarbonate concentrations may be low. Hypokalaemic, hyperchloraemic acid- osis results from loss of body fluids low in Cl–​ relative to Na+ and K+ when compared with the ratio of Cl–​ to Na+ in ECF. It is not always the case that metabolic acidosis caused by diarrhoea is due to bicar- bonate loss. Stool losses of Na+, K+, and HCO3–​ characterize most small bowel diarrhoea, while organic acid anions such as butyrate and acetate of bacterial origin are often the lost anions in colonic diarrhoea. Role of colonic diarrhoea in hypochloraemic alkalosis Rarely, villous adenomas or adenocarcinomas of the rectosigmoid secrete excessive quantities of sodium, potassium, and chloride, re- sulting in severe hypokalaemic, hypochloraemic metabolic alkal- osis. Some infectious diarrhoeas may result in alkalosis. From this discussion of the gastrointestinal tract, it should be noted that functions of the early small intestine, where bicarbonate from pancreatic and biliary secretions is absorbed along with glu- cose, amino acids, and water of dietary source, are analogous with the situation in the proximal tubule, where bicarbonate, glucose, amino acids, and water are reabsorbed from a filtrate of plasma. Likewise, the large intestine absorbs sodium and water and se- cretes potassium in an aldosterone regulated fashion, much like the collecting duct of the kidney. As emphasized in this discus- sion, the abnormal function of intestinal epithelial cells can result in volume depletion in association with either acidosis or alkal- osis, depending on the balance of losses of sodium and potassium compared to chloride. The term contraction alkalosis is therefore misleading. Acid–​base disorders and the skin Sweat gland ducts The sweat gland ducts (Fig. 12.11.13), like the principal cells in the kidney and the cells of the colon, contain aldosterone-​sensitive ENaC. Na+ absorption from the glandular duct renders the lumen electronegative. Normally the negative lumen drives chloride absorption through the CFTR, leading to limited volumes of hypotonic sweat. Cystic fibrosis and metabolic alkalosis Patients with cystic fibrosis may develop hypochloraemic alkalosis as a consequence of excessive sweat chloride content related to their CFTR gene mutation. When Cl− absorption is decreased in cystic fibrosis, the lumen becomes more negative, decreasing Na+, Cl−, and fluid absorption, leading to ‘salty’ sweat; the proportionally large Cl− loss may generate hypochloraemic metabolic alkalosis. Acid–​base disorders associated with transport of acid anions from intracellular to extracellular spaces: anion gap acidosis Several organic acid anions are produced metabolically in one cell type and carried by the blood to another cell type where they can be used as a fuel for further metabolism. For example, the Cori cycle involves the production of lactate in skeletal muscle, where it en- ters the interstitial fluid through monocarboxylic acid transporters (MCTs) on the plasma membrane, driven by solute gradients (Fig. 12.11.14). The lactate is then transported by similar MCTs into hepatic cells where it enters gluconeogenic pathways to form glucose. Thus, when considering the plasma as part of the ECF, the steady-​state lactate concentration is equal to the ratio of production to clearance. Excessive levels may occur with overproduction or de- creased clearance. The reason that plasma does not accumulate pyruvate in the same way is that the MCT has greater specificity for lactate. Pyruvate ­remains in the cell as a potential energy source through acetyl co- enzyme A. The ketoacids acetoacetate and β-​hydroxy butyrate are made in liver mitochondria and transported out of the liver cells by Distal colon Fig. 12.11.12  Distal colon. (a) Luminal content; (b) cellular transport mechanism. AldoR, aldosterone receptor; ENaC, epithelial sodium channel. Sweat gland duct Fig. 12.11.13  Sweat gland duct. (a) Luminal content; (b) cellular transport mechanism. AldoR, aldosterone receptor; CFTR, cystic fibrosis transmembrane conductance regulator.

12.11  A physiological approach to acid–base disorders 2195 MCTs, circulate, and are then cleared by brain and heart tissue (via MCTs) to be used as fuel when glucose is low, as in starvation. As with lactate, the plasma level of ketoacids represents the production to clearance ratio. Because these anions are filtered by the kidney, they can enter the urine (by renal clearance), obligating cations sodium and potassium for electroneutrality. The urinary excretion of sodium with a nonchloride anion leaves behind a relatively greater amount of extracellular chloride than sodium (hyperchloraemic acidosis). Thus, with lactic acidosis and ketoacidosis there are two mechanisms for acidosis; one an anion gap due to anion production greater than total clearance, and a hyperchloraemic component that involves the production of HA entering plasma and loss of NaA in the urine. In such a case, the decrease in HCO3–​ will exceed the increased anion gap, especially if the GFR and the filtered load of the anion are high. Box 12.11.3 lists causes of high anion gap metabolic acidosis, some of which are now discussed further. Diabetic ketoacidosis Diabetic ketoacidosis is defined as hyperglycaemia with meta- bolic acidosis resulting from generation of the acid anions β-​hydroxybutyrate and acetoacetate in response to insulin defi- ciency and elevated counter-​regulatory hormones such as epineph- rine and glucagon. Most commonly seen in cases of type 1 diabetes mellitus, severe stress can occasionally bring on ketoacidosis in type 2 diabetes mellitus diabetics. The lack of insulin increases lipolysis in adipose tissue; free fatty acids are transported to the liver, where hepatic mitochondria pro- duce ketone bodies, including acetoacetate, from acetyl coenzyme A. In the presence of a high NADH/​NAD ratio, the more reduced form of β-​hydroxybutyrate is produced. Ketoacids produced within hepatocytes are secreted into the ECF by the monocarboxylate-​ proton cotransporter, thereby causing acidaemia (Fig. 12.11.15). Ketoacidosis is also seen in cases of starvation, in which it is gen- erally mild and not associated with hyperglycaemia. The urinary dipstick nitroprusside test for ketones may underestimate the degree of ketosis because it does not detect β-​hydroxybutyrate, and the ketone test may become more positive as treatment helps metabolize β-​hydroxybutyrate to acetoacetate. This problem can be addressed by direct measurement of serum β-​hydroxybutyrate. Treatment of diabetic ketoacidosis consists of volume repletion, insulin administration (with dextrose if necessary to avoid hypogly- caemia), and potassium replacement. Bicarbonate administration should be considered only if ketoacidosis is accompanied by shock in conjunction with arterial pH of less than 7.0. Alcoholic ketoacidosis Alcoholic ketosis occurs in a patient who has been drinking very heavily without eating. The pathophysiology is based on the over- production of β-​hydroxybutyrate and (to a lesser extent) acetoacetate because of an increased production of free fatty acids from adipose tissue. Alcohol inhibits the conversion of lactate to glucose in the liver, favouring hypoglycaemia with fasting. The oxidation of ethanol in- creases the ratio of NADH to NAD+ and favours the production of β-​hydroxybutyrate from acetoacetate. Damage to mitochondria by al- cohol can further elevate the ratio of β-​hydroxybutyrate to acetoacetate by preventing reoxidation of NADH to NAD. The oxidative metab- olism of ethanol favours the reaction of dehydrogenase enzymes to form β-​hydroxybutyrate and lactate (opposing glucose production). Alcoholic ketoacidosis usually follows binge drinking and may be associated with withdrawal symptoms and the associ- ated hyperadrenergic state. It is associated with abdominal pain, vomiting, starvation, and volume depletion. In contrast to diabetic ketoacidosis, coma is rare. The blood glucose level is generally low or normal, and the insulin level is frequently low. The blood alcohol level may be unrecordable (absent) or elevated on initial evaluation. The osmolar gap, if secondary to ethanol, should be equal to the ethanol concentration in milligrams per decilitre divided by 4.6. If this calculation does not yield the expected gap based on the ethanol concentration, ingestion of another alcohol such as methanol, iso- propanol, or ethylene glycol should be suspected. Acetone, the product of acetoacetate metabolism, is seen during recovery and may register as an unmeasured osmol. By contrast, isopropanol me- tabolizes directly to acetone and causes ketosis without acidosis. Liver Fat Lipogenesis Lactate Heart, brain Skeletal muscle Monocarboxylic acid transporters (MCTs) Ketoacids Ketoacids Oxidation Glucose Lactate Lactate Glycolysis Fig. 12.11.14  Monocarboxylic acid transporters (MCTs) in lactate and ketoacid transport. During exercise glucose is broken down to lactate in skeletal muscle. Outwardly directed lactate gradients drive lactate into the extracellular fluid via an MCT. The liver takes up that lactate, also via an MCT, and converts it to glucose, which is available for export to the muscle cells for completion of the Cori cycle. Ketoacids are produced in the liver from fatty acids. They are then transported via an MCT, particularly in diabetes and hypoglycemic conditions, and the ketoacids are then transported via MCTS into brain and heart muscle cells as a source of energy through oxidation. Monocarboxylate transporters (MCTs) Hepatocyte Lactate Acetoacetate β-hydroxybutyrate Interstitial fluid (SLC)16 solute carrier family Fig. 12.11.15  An example of an MCT transporter in a hepatic cell shows coupling to hydrogen ion as well as the substrates that are transported through the (SLC) solute carrier family.

section 12  Metabolic disorders 2196 Treatment of alcoholic acidosis consists of volume repletion with 0.9% saline, administration of thiamine (50 to 100 mg intravenously), and enough glucose to treat hypoglycaemia, and the correction of any hypophosphataemia, hypokalaemia, and hypomagnesaemia that may be present. The acid–​base disturbance usually resolves after several hours. Both hypophosphataemia and thiamine deficiency, which may not be apparent until 12 to 24 h after the initiation of treatment in an undernourished patient, are exacerbated by glucose administration and may contribute to an associated lactic acidosis. Lactic acidosis Lactate, the final product in the anaerobic pathway of glucose me- tabolism, is produced from pyruvate by the following reaction cata- lysed by lactate dehydrogenase: NADH +pyruvate +H lactate +NAD

(Equation 4) Lactic acidosis is caused by an imbalance in the rates of lactate pro- duction and clearance, primarily in the liver. Lactic acidosis, which increases the anion gap, is most often due to impaired lactate clear- ance due to circulatory failure, hypoxia, and mitochondrial dysfunc- tion that increase anaerobic glycolysis and the rate of reduction of pyruvate to lactate. Lactate, once formed, results in acidaemia after transport into the ECF by the organic acid anion transporter, MCT1. Other causes of lactic acidosis are thiamine deficiency, hypo­ phosphataemia, isoniazid toxicity, and hypoglycaemic states. Metformin may cause lactic acidosis, particularly in elderly patients with cardiac, hepatic, or renal dysfunction. Nucleoside antivirals, including zidovudine, may cause lactic acidosis and abnormal liver function as a result of toxic mitochondrial effects. Abnormal mito- chondrial function is also a feature of aspirin overdose. The anti- biotic linezolid is another cause of lactic acidosis. Many tumours utilize glycolysis for energy and produce large quantities of lactate. Treatment of lactic acidosis is aimed at correcting the underlying cause. Tissue perfusion and ventilation need to be restored. If the arterial pH is 7.0 or less, or when shock or cardiac failure has de- veloped, sodium bicarbonate therapy should be considered. This is usually given as an isotonic infusion (1.26% sodium bicarbonate, or 5% dextrose with added bicarbonate). Treatment carries risks of pul- monary oedema and hypernatraemia. In patients with intestinal bacterial overgrowth, disorientation, and ataxia, an anion gap metabolic acidosis may develop after a carbohydrate meal because of lactobacilli production of d-​lactate. This isomer of the mammalian l-​lactate can be measured only by a specific d-​lactate assay. The condition is treated with oral antibiotics and appropriate diet. Ethylene glycol Ethylene glycol is a constituent of antifreeze and also used as an in- dustrial solvent. It has a sweet taste and patients occasionally ingest it as a substitute for ethanol. Although ethylene glycol itself is not par- ticularly damaging, its highly toxic metabolites include glyoxylate, glycolate, oxalic acid, and ketoaldehydes. These acidic products are formed by metabolism within cells catalysed by alcohol and alde- hyde dehydrogenase and then transported into the ECF by MCTs. Oxalate is transported across cells by anion exchangers in the SLC26 gene family. Glycolic acid appears to be primarily responsible for the metabolic acidosis observed. Intoxication is characterized by profound central nervous system symptoms, including diplopia, seizures and coma, severe metabolic acidosis, and cardiac, pulmonary, and renal failure. Patients are often dehydrated and hypernatraemic because of os- motic diuresis from the renal clearance of the alcohol. Calcium oxalate crystals in the urine may cause intratubular obstruction and acute kidney injury. Patients typically have a high osmolal gap, initially defined as the difference between the measured and the calculated serum osmolality: S 2(Na )+ glucose [mg/dl] 18

osm + − ÷ ÷ 8     (Equation 5) Table 12.11.3 shows the differential diagnosis of patients with anion gap metabolic acidosis and high osmolar gaps. The serum osmolality should be measured by a freezing point depression technique and compared with the calculated osmo- lality. If possible, ethanol, ethylene glycol, propylene glycol, and methanol levels should be measured directly; each is associated with a metabolic acidosis. An increased anion gap is attributable to ethylene glycol metab- olites. A high osmolar gap will also be present because of the un- charged alcohol, but an osmolar gap may not be present if all of the alcohol has been converted to the toxic anionic forms. Treatment is aimed at rehydration with 0.9% saline and correc- tion of acidosis with NaHCO3 based on an estimate of the bicar- bonate deficit. When there is an osmolal gap, competitive inhibition of alcohol dehydrogenase should be initiated with fomepizole at a loading dose of 15 to 20 mg/​kg intravenously in 100 ml 0.9% sa- line over 30 min to 1 h, followed by a maintenance dose of 10 mg/​ kg every 12 h. An alternative is to use ethanol itself, in which case a solution of 10% ethanol in 5% dextrose can be given as a loading dose of 0.6 g/​kg intravenously, followed by a maintenance dose of 150 mg/​kg per h in alcoholic patients, or 65 mg/​kg per h in non- alcoholic patients. The blood ethanol level should be maintained at 100 to 200 mg/​dl. The goal of therapy is to prevent metabolism of the uncharged glycol to acidic products. Haemodialysis is required in severe cases. Some automotive fluids now contain the less toxic propylene glycol. Propylene glycol, a 3-​carbon glycol, is used as a diluent in some intravenous medications such as lorazepam. It metabolizes to lac- tate. Treatment consists of early recognition, fluid replacement (es- pecially if associated with an osmotic diuresis), and withdrawal of the offending agent. Table 12.11.3  Anion and osmolal gap in diagnosis of intoxications Anion gap acidosis Osmolal gap Diagnosis + Normal Salicylates + High Ethanol Ethylene glycol Propylene glycol Methanol − High Isopropanol

12.11  A physiological approach to acid–base disorders 2197 Methanol Methanol (wood alcohol) is a component of shellac and windshield wiper fluid and is highly toxic to the central nervous system after metabolism by alcohol and aldehyde dehydrogenase to formalde- hyde and formic acid. Optic papillitis may cause blindness. Treatment consists of competitive inhibitors for alcohol dehydro- genase, including ethanol or fomepizole, in similar amounts as for ethylene glycol poisoning, to reduce the formation of acid anions and the anion gap while maintaining a higher level of methanol in the blood. Haemodialysis may be necessary to increase elimination. 5-​Oxoprolinuria 5-​Oxoprolinuria is detected in debilitated patients with depleted intracellular glutathione (GSH) who are taking paracetamol (acet- aminophen). The accumulation of 5-​oxoproline (pyroglutamic acid) is caused by further drug-​induced depletion of GSH through interference with the γ-​glutamyl transpeptidase pathway respon- sible for creating GSH for shuttling amino acids into the cytosol. Normal glutathione levels are necessary for feedback inhibition of γ-​glutamylcysteine synthase, which regulates the activity of the cycle and is metabolized to 5-​oxoproline. As shown in Fig. 12.11.16, 5-​oxoproline leaves cells through plasma membrane H+-​coupled SLC16A1/​MCT1 transporter. Salicylate intoxication Salicylate intoxication can be caused by accidental overdose, thera- peutic overdose, or in a parasuicide or suicide attempt. Salicylates may cross cell membranes through nonionic diffusion, anion ex- change, and organic anion transporters (OAT1). Salicylate functions as an uncoupler of oxidative phosphorylation and consequently results in increased oxygen consumption and CO2 production. However, the increase in alveolar ventilation resulting from stimu- lation of central chemoreceptors overcomes this increase in CO2. The most common clinical manifestation is a combined anion gap metabolic acidosis and respiratory alkalosis, although the condition also can be manifested as either one or the other only. Children are often seen with metabolic acidosis, whereas adults often have pre- dominant respiratory alkalosis. Hypoglycaemia, ketoacidosis, and lactic acidosis may result. Other manifestations of intoxication include haemorrhage, fever, nausea and vomiting, hyperventilation, diaphoresis, tinnitus, and occasionally polyuria followed by oliguria. Severe cases may lead to seizures, respiratory depression, and coma. Noncardiogenic pul- monary oedema is sometimes seen in adults. Respiratory alkalosis is the result of a direct stimulatory effect of salicylate on the medullary respiratory control centre. Salicylate in- toxication also increases the metabolic rate. Diagnosis is suspected by the clinical presentation and confirmed by the salicylate level. Treatment is aimed at correcting the meta- bolic acidosis and removing salicylate. Bicarbonate as a sodium salt should be administered according to an estimated calculation of the deficit if metabolic acidosis predominates. Salicylates are removed by alkaline diuresis because the less reabsorbable salicylate anion will predominate when the urine pH increases. In severe poisoning, or when renal failure is present, dialysis may be required. Acid–​base disorders associated with intake The ECF contains approximately 140 mM sodium and 100 mM chloride. Sodium concentration is critically important in osmotic regulation, hence if sodium salts are ingested, then water will be taken in and conserved in order to maintain a constant plasma so- dium concentration. By contrast, plasma chloride concentration is determined by fluid balance and acid–​base requirements. The in- take of sodium and chloride in equivalent concentrations therefore requires a greater clearance of the chloride ion than of the sodium ion, and this clearance is primarily urinary. In order to eliminate chloride without sodium, and without necessitating potassium ex- cretion, there must be a cation to accompany chloride to achieve electroneutrality. Ammonium serves that purpose, and yet we think of ammonium excretion as a mechanism for acid elimination in the urine. That the intake of sodium chloride requires the loss of ammo- nium chloride (having an alkalinizing effect on the ECF) suggests that sodium chloride is an acidifying salt. If the intake of sodium chloride required the production of ammonium that exceeded the maximal ability of the kidney to do so, then the amount of acid ex- creted would be limited and hyperchloraemic metabolic acidosis should follow. Intravenous fluids may cause metabolic disorders by direct access to the ECF. Acidosis may be caused by infusions of high chloride salts including saline. Alkalosis may result from the addition of intravenous alkali salts of metabolizable organic anions. The normal response to NaHCO3 is rapid urinary alkalinization because of an unaltered threshold for HCO3–​ reabsorption. However, a marked ex- cess of HCO3–​, as may be administered in an attempt to alkalinize a patient’s urine, expands volume and causes an alkalaemia, especially in the presence of volume depletion or low glomerular filtration. Milk-​alkali syndrome, usually seen when patients in chronic kidney disease ingest milk or calcium antacids, is associated with hypercalcaemia, alkalaemia, and normal Cl–​. Other situations in which intake of alkali salts results in metabolic alkalosis include infusion of large quantities of sodium salts of acetate, citrate, lac- tate, or bicarbonate; hyperalimentation with acetate salts; peritoneal dialysis with acetate or lactate dialysate; or excessive transfusions or plasmapheresis in which large quantities of citrate, used as an anti- coagulant, are delivered. Entry of hydrogen ions into cells can also lead to metabolic al- kalosis in patients with hypokalaemia. If an alkalotic patient is not hypochloraemic, electroneutrality must be maintained either by depletion of an alternative anion or by an excessive concentration of a cation. An example of a metabolic alkalosis associated with MCT1 transports 5-oxoproline, as does SLC5A8 5-Oxoproline H+ Fig. 12.11.16  MCT1 transport of 5-​oxoproline (pyroglutamate).

section 12  Metabolic disorders 2198 depletion of a nonchloride anion is hypoproteinaemic alkalosis, with hypoalbuminaemia and a small anion gap. Chloride balance is normal and chloride appears in urine. Symptoms of acid–​base disorders Mild metabolic alkalosis up to a pH of 7.50 is usually asymptomatic. When the pH exceeds 7.55, however, the alkalosis itself and the com- pensatory hypoventilation are frequently associated with metabolic encephalopathy. Symptoms include confusion, obtundation, de- lirium, and coma. The seizure threshold is lowered, and tetany, par- aesthesias, muscular cramping, and other symptoms of low ionized calcium are seen. In patients with hypocalcaemia, these signs may be seen at pH values above 7.45, hence extreme caution should be taken if a decision is made to alkalinize a hypocalcaemic patient with acidaemia and this should only be done if there is a very pressing need for rapid correction of acidosis. Other findings include cardiac tachyarrhythmias and hypotension. Lactate production increases as a result of increased anaerobic glycolysis. Healthy, trained athletes may develop severe acidaemia (pH <7.0), but in acutely ill patients, blood pH as low as 7.2 may cause shock, and cardiac arrhythmia. Symptoms of chronic metabolic acidosis include nausea, vomiting, anorexia, and dyspnoea on exertion. Acutely, patients often exhibit Kussmaul respirations and volume depletion. Neurological symptoms include fatigue and lethargy with depression of the sensorium. Treatment of acid–​base disorders Metabolic acidosis Aside from treatment of the underlying condition, patients whose pH is less than 7.2 are typically treated with infusions of sodium bicarbonate, guided by the estimated bicarbonate deficit, calculated using the serum HCO3–​ concentration in mEq per litre: Amount of HCO = (25 [HCO ]) wt(kg)/2 3 3 − − − ×   (Equation 6) In general, the correction of metabolic acidaemia should be based on a calculated amount, with not more than 50% of the estimate given before re-​measurement of electrolyte and bicarbonate concen- trations and recalculation. It should be noted that this equation is used for deficit correction only; the ongoing losses of 1 to 2 mEq/​ kg per day, equivalent to the daily acid load, should be replaced in distal renal tubular acidosis with NaHCO3, KHCO3, or citrate salts in divided doses. Citrate should be avoided as an alkalinizing salt in patients with low GFR. Metabolic alkalosis The treatment of metabolic alkalosis rarely depends on giving back HCl. In chloride responsive alkalosis, replacement with 0.9% saline is indicated, but can be complicated by worsening of hypokalaemia if the bicarbonate is promptly excreted, mandating simultaneous treatment with potassium chloride. Expansion in patients with hypovolaemic hyponatraemia may correct the low sodium concen- tration faster than a safe rate, in which case the clinician must be prepared to replace with hypotonic fluids or add an antidiuretic hor- mone analogue to control water loss. The derivation of equation 6 is important in that it suggests the volume of distribution of bicarbonate to be 50% body weight. In fact, the bicarbonate volume of distribution is closer to 20% body weight, or the ECF volume. At pH 7.40, the fraction of total body buffer capacity for the bicarbonate system approximates 0.4, thus 0.2/​0.4 is the derivation of 0.5 × body weight. Due to the isohydric relation- ship being nonlinear, at acid pH the denominator of 0.2 decreases, meaning that bicarbonate becomes a less important buffer and con- sequently the bicarbonate amount calculated as replacement using equation 6 will be an underestimate. Further, in an acidaemic con- dition the change in chloride added to the ECF may differ from the amount of bicarbonate leaving the ECF, the latter being more. That the change in chloride equals the change in bicarbonate concentra- tion is evidence that the bicarbonate concentration is dependent on the balance of strong ions like sodium and chloride. FURTHER READING Adrogué HJ, et al. (2009). Assessing acid-​base disorders. Kidney Int, 76, 1239–​47. Batlle DC, et al. (1988). The use of the urinary anion gap in the diagnosis of hyperchloremic metabolic acidosis. N Engl J Med, 318, 594–​9. Chadha V, Alon US (2009). Hereditary renal tubular disorders. Semin Nephrol, 29, 399–​411. De Backer D (2003). Lactic acidosis. Intensive Care Med, 29, 699–​702 Emmett M, Narins RG (1977). Clinical use of the anion gap. Medicine (Baltimore), 56, 38–​54. Fordtran JS (1971). Organic anions in fecal contents. N Engl J Med, 284, 329–​30. Gennari FJ, Weise WJ (2008). Acid-​base disturbances in gastrointes- tinal disease. Clin J Am Soc Nephrol, 3, 1861–​8. Gennari FJ (2011). Pathophysiology of metabolic alkalosis: a new clas- sification based on the centrality of stimulated collecting duct ion transport. Am J Kidney Dis, 58, 626–​36. Halestrap AP, Wilson MC (2012). The monocarboxylate transporter family—​role and regulation. IUBMB Life, 64, 109–​19. Koeppen BM (2009). The kidney and acid-​base regulation. Adv Physiol Educ, 33, 275–​81. Kraut JA, Madias NE (2007). Serum anion gap: its uses and limitations in clinical medicine. Clin J Am Soc Nephrol, 2, 162. Luke RG, Galla JH (2012). It is chloride depletion alkalosis, not con- traction alkalosis. J Am Soc Nephrol, 23, 204–​7. Oh MS, Carroll HJ (1977). The anion gap. N Engl J Med, 297, 814–​17. Seldin DW, Rector FC Jr (1972). Symposium on acid-​basis homeo- stasis:  the generation and maintenance of metabolic alkalosis. Kidney Int, 1, 306–​21.

12.12 The acute phase response, hereditary periodi

12.12 The acute phase response, hereditary periodic fever syndromes, and amyloidosis 2199

12.12.1 The acute phase response and C- reactive p

12.12.1 The acute phase response and C- reactive protein 2199

12.12 The acute phase response, hereditary periodic fever syndromes, and amyloidosis CONTENTS 12.12.1 The acute phase response and C-​reactive protein  2199 Mark B. Pepys 12.12.2 Hereditary periodic fever syndromes  2207 Helen J. Lachmann, Stefan Berg, and Philip N. Hawkins 12.12.3 Amyloidosis  2218 Mark B. Pepys and Philip N. Hawkins 12.12.1  The acute phase response
and C-​reactive protein Mark B. Pepys ESSENTIALS The acute phase response—​trauma, tissue necrosis, infection, inflam- mation, and malignant neoplasia induce a complex series of non- specific systemic, physiological, and metabolic responses including fever, leucocytosis, catabolism of muscle proteins, greatly increased de novo synthesis and secretion of a number of ‘acute phase’ plasma proteins, and decreased synthesis of albumin, transthyretin, and high-​ and low-​density lipoproteins. The altered plasma protein concentration profile is called the acute phase response. All endo- thermic animals mount a similar response, suggesting that it may have survival value, and increased availability of proteinase inhibi- tors, complement, clotting, and transport proteins presumably enhances host resistance, minimizes tissue injury, and promotes re- generation and repair. Acute phase proteins—​these are mostly synthesized by hepato- cytes, in which transcription is controlled by cytokines including interleukin 1, interleukin 6, and tumour necrosis factor. The circu- lating concentrations of complement proteins and clotting factors increase by up to 50 to 100%; some of the proteinase inhibitors and α1-​acid glycoprotein can increase three-​ to fivefold; but C-​reactive protein (CRP) and serum amyloid A protein (an apolipoprotein of high-​density lipoprotein particles) are unique in that their concen- trations can change by more than 1000-​fold. C-​reactive protein—​this consists of five identical, nonglycosylated, noncovalently associated polypeptide subunits. It binds to au- tologous and extrinsic materials which contain phosphocholine, including bacteria and their products. Ligand-​bound CRP activates the classical complement pathway and triggers the inflammatory and opsonizing activities of the complement system, thereby con- tributing to innate host resistance to pneumococci and probably to recognition and safe ‘scavenging’ of cellular debris. Clinical features—​(1) determination of CRP in serum or plasma is the most useful marker of the acute phase response in most inflam- matory and tissue damaging conditions. It is a stable analyte, easy to measure, and has proven value in monitoring therapeutic responses. (2) Acute phase proteins may be harmful in some circumstances. Sustained increased production of serum amyloid A protein can lead to the deposition of AA-​type, reactive systemic amyloid, a serious and if untreated, usually fatal condition that can complicate chronic infective and inflammatory diseases. CRP, through its capacity to acti- vate complement, can exacerbate ischaemic (and possibly also other forms) of tissue damage. Introduction The principal plasma proteins that change in concentration in the acute phase response are listed in Table 12.12.1.1. C-​reactive protein (CRP) has particular clinical utility as a robust and easily measured systemic marker for monitoring the extent, activity, and response to therapy in many inflammatory, infective, and tissue-​damaging conditions. C-​reactive protein CRP was the first protein to be discovered that behaves as an acute phase reactant, and was named for its calcium-​dependent inter- action with the somatic C-​polysaccharide of pneumococci, in which CRP recognizes phosphocholine residues. CRP also binds to other

section 12  Metabolic disorders 2200 substances that contain phosphocholine, including phospho- lipids, some plasma lipoproteins, and the plasma membranes of damaged cells. In addition, CRP binds specifically to small nu- clear ribonucleoprotein particles when these are exposed in dead or damaged cells. The CRP molecule consists of five identical, nonglycosylated, noncovalently associated polypeptide subunits, each of mass 23027 Da and containing 206 amino acid residues. The subunits have a flattened β-​sheet jellyroll fold with a single intrachain disulphide bond, and are arranged in an annular config- uration with cyclic pentameric symmetry. There is a single calcium-​ dependent ligand-​binding site on the medial aspect of each subunit, all located on the same planar face of the molecule. A distinct but closely related plasma protein, serum amyloid P component, which is not an acute phase protein in humans, has a very similar molecular structure with the same fold, characteristic of the lectin-​fold super- family. CRP and serum amyloid P component belong to the phylo- genetically conserved pentraxin family. Although many properties of human CRP have been reported in experimental systems, no structural polymorphism of human CRP has been observed nor has any human CRP deficiency been described, so the actual functions of human CRP in humans are not yet known. Ligand-​bound CRP activates the classical complement pathway via C1, and can trigger the inflammatory and opsonizing activ- ities of the complement system. A significant biological function of CRP may thus be to recognize and scavenge cellular debris, promoting its safe clearance and helping to maintain tolerance to potential autoantigens. CRP may also contribute to innate resist- ance against infection with bacteria that express phosphocholine, and experiments in CRP knockout mice show that CRP is essential for innate host defence against pneumococci. On the other hand, complement activation by CRP binding to damaged tissue exacer- bates ischaemic and possibly other forms of tissue injury. However, in healthy subjects, infusion of pure CRP has no proinflammatory, proatherogenic, or any other adverse effects. Serum concentration of CRP Circulating CRP is synthesized by the hepatocytes under tran- scriptional regulation by the proinflammatory cytokines, especially IL-​6. CRP is a trace protein in apparently normal healthy individ- uals, the median value in adults being 0.8 mg/​litre, with an inter- quartile range of 0.3 to 1.7 mg/​litre. Among apparently healthy subjects, 90% of CRP values are less than 3 mg/​litre and 99% less than 10 mg/​litre. Serum CRP concentrations are lower in healthy newborns, but reach adult values within a few days. Normal values in the indigenous Japanese population are substantially lower than in white Caucasians. Serial studies of normal subjects and of monozygotic and dizygotic twins show that each individual’s base- line CRP value is rather constant and is substantially genetically determined. Baseline CRP is strongly correlated with body mass index, especially abdominal obesity, and is also higher in smokers, hypertensive subjects, diabetics, those who take little or no exercise, and individuals from the lower socioeconomic classes. Occasional higher values of CRP seen in ostensibly healthy people almost cer- tainly reflect intercurrent subclinical pathology. In large surveys of the unscreened general population, there is a trend towards higher values with increasing age, with the median value rising to about 2 mg/​litre, and this likely reflects the higher incidence of many dif- ferent pathological processes with age. Serum CRP concentration rises rapidly in the acute phase re- sponse and can exceed 300 mg/​litre by 48 h after a severe stimulus such as acute systemic bacterial infection, major trauma or surgery, or acute myocardial infarction. With uncomplicated resolution of injury or effective treatment of infection, the circulating CRP con- centration generally falls equally rapidly. The speed of change and incremental range of CRP concentra- tions are exceptional among all the acute phase proteins, apart from serum amyloid A protein, which behaves in a similar fashion. The half-​life of CRP in the circulation is 19 h and is constant in all conditions, regardless of the presence of an acute phase re- sponse or its cause. In contrast to other acute phase proteins, such as clotting factors, complement proteins, transport proteins, and proteinase inhibitors, CRP does not undergo major local seques- tration or consumption, fragmentation, or complex formation. This means that, unlike most of the other acute phase reactants, the single major determinant of the circulating concentration of CRP is its rate of synthesis. Since this in turn is dependent on the intensity of the acute phase stimulus, the serum CRP level usually closely reflects the extent and activity of disease. These properties underlie the value in clinical practice of precise measurement of the serum CRP concentration. Drug or other treatments do not affect CRP production unless they also affect the disease process that is responsible for the induction of CRP synthesis. The only ex- ception is combined ciclosporin and steroid treatment given after renal transplantation. This suppresses the CRP response to renal allograft rejection, though not that provoked by infection. The only physical condition that seriously interferes with the capacity to in- terpret CRP levels is severe hepatocellular impairment, since CRP is made exclusively in the liver. Table 12.12.1.1  Plasma protein concentrations in the acute phase response Protein Increased Decreased Proteinase inhibitors α1-​antitrypsin Inter α-​antitrypsin α1-​antichymotrypsin Coagulation proteins Fibrinogen Prothrombin Factor VIII Plasminogen Complement proteins Cls Properdin C2, B C3, C4, C5 C56 C1INH Transport proteins Haptoglobin Haemopexin Caeruloplasmin Miscellaneous C-​reactive protein Albumin Serum amyloid A protein Transthyretin (prealbumin) Fibronectin High-​density lipoprotein α1-​acid glycoprotein Low-​density lipoprotein Gc globulin

12.12.1  The acute phase response and C-reactive protein 2201 Conditions associated with marked increases in serum CRP concentration Most tissue-​damaging processes, infections, inflammatory diseases of unknown aetiology, and malignant neoplasms are associated with a major acute phase response of CRP. CRP production is exquisitely sensitive to all these pathologies and is thus a nonspecific response to disease. It can never, on its own, be used as a diagnostic test. However, if the CRP result is interpreted in the light of full clinical information about the patient, it can provide exceptionally useful in- formation for clinical management. Thus in nearly all the conditions listed in Box 12.12.1.1 the CRP concentration reflects quite precisely the extent and activity of disease. With deterioration, the CRP value rises, whereas with spontaneous or therapeutically induced remis- sion, the CRP value falls, and it thereby supplies an objective index of progress that is rarely available in any other way. Infection Most forms of systemic microbial infection are associated with high serum CRP concentrations and, although the peak values attained in different patients cover a wide range, serial assays in individual subjects usually show an excellent correlation between the CRP value and the severity of disease and its response to treatment. Acute sys- temic Gram-​positive and Gram-​negative bacterial infections are among the most potent stimuli for CRP production. Systemic fungal infections occurring in immunodeficient hosts are also associated with high CRP values, whereas the concentrations in chronic bac- terial infections such as tuberculosis and leprosy are usually rather lower, though nevertheless still markedly raised. Uncomplicated viral infections, particularly meningitis, may induce only a very modest response or none at all. Clinical rhinovirus infection (common cold) and influenza are associated with minor increases in CRP concentration in a proportion of individuals, though this may reflect secondary bacterial infection. However, severe influenza virus, sys- temic cytomegalovirus or herpes simplex infections of immunosup- pressed patients cause a major CRP response. Little is known about the CRP response to metazoan parasitic infestation in otherwise healthy subjects but malaria, especially Plasmodium falciparum in- fection, is associated with high CRP values, as are Pneumocystis spp. and Toxoplasma spp. infections in immunodeficient patients. Minor or localized low-​grade infection may not stimulate CRP production greatly, but the major CRP response in acute, serious bacterial infection is almost invariable and is present at all ages from premature neonates to older people. It also occurs in patients who are immunosuppressed or immunocompromised, whether by a pri- mary disease such as leukaemia, lymphoma, or other malignancy, by AIDS, or by treatment with cytotoxic drugs, corticosteroids, or ir- radiation. This is of particular importance in the very young, in older people, in compromised hosts, and in any other patient in whom the usual clinical signs and symptoms of infection, including fever and neutrophil leucocytosis, may be masked or lacking (Figs. 12.12.1.1 and 12.12.1.2). Furthermore, at the onset of bacterial infection, es- pecially in patients who are otherwise well following elective surgery or myocardial infarction, the CRP response frequently precedes clinical symptoms, including fever, by up to 24 to 48 h. Once infection is diagnosed or suspected and antimicrobial treatment has been commenced, frequent monitoring of the serum CRP concentration provides an objective means of assessing the Box 12.12.1.1  Conditions associated with major increases in serum CRP concentration • Infection and immunological complications of infection • Inflammatory disease:

—​ Rheumatoid arthritis

—​ Juvenile chronic (rheumatoid) arthritis

—​ Ankylosing spondylitis

—​ Psoriatic arthritis

—​ Systemic vasculitis

—​ Polymyalgia rheumatica

—​ Reiter’s disease

—​ Crohn’s disease

—​ Familial Mediterranean fever, CAPS • Necrosis:

—​ Myocardial infarction

—​ Tumour embolization

—​ Acute pancreatitis • Trauma:

—​ Surgery

—​ Burns

—​ Fractures • Malignant neoplasia:

—​ Lymphoma

—​ Hodgkin’s disease

—​ Carcinoma, sarcoma Fig. 12.12.1.1  A 69-​year-​old diabetic man was admitted with a 3-​day history of confusion, cough, and incontinence of urine. There was clinical and radiological evidence of a left-​sided pneumonia and although both the temperature and white cell count remained normal, the CRP value was high (119 mg/​litre), confirming the suspicion of infection. Following treatment with amoxicillin, 250 mg three times daily, the CRP concentration fell rapidly, in a characteristic exponential manner, and he made a speedy recovery with return of continence and improved mental state.

section 12  Metabolic disorders 2202 response, which is often not otherwise available. Effective therapy is associated with a rapid, exponential fall in CRP value, with a half-​ life of about 24 h, and occurrence of this pattern is an encouraging prognostic sign (Fig. 12.12.1.2). Normalization of the CRP usually corresponds to clinical cure of the infection and may thus be used to determine the necessary duration of antimicrobial therapy. On the other hand, especially in neutropenic or immunodeficient pa- tients, persistent elevation of CRP at the end of a course of anti- biotics often presages relapse or recurrence of infection. When bacterial infection is complicated by abscess formation or for any other reason is less readily eradicated by antimicrobial drugs, the serum CRP concentration may remain elevated or may fall lin- early rather than exponentially during treatment. Such a pattern should raise questions regarding the dosage of the drugs, the sen- sitivity of the organism, and/​or stimulate a diagnostic search both for localized pus and for other underlying, noninfective pathologies such as malignancy. Indeed, in the absence of one of the chronic idio- pathic inflammatory conditions which are known to be associated with high CRP concentrations (see later subsections), the persist- ence of a raised serum CRP concentration is usually a grave prog- nostic sign indicating the presence of either uncontrolled infection and/​or other serious pathology likely to cause death. However, with alteration in the antimicrobial drug regimen or the evacuation of pus or elimination of other pathology, the rapid fall in CRP that may then be observed is an encouraging objective sign of clinical improvement. These considerations apply at all ages and regardless of intercur- rent pathology, with the exception of severe hepatocellular impair- ment. In view of the very small amount of serum required for the assay and the speed and precision of automated CRP immunoassays, it is apparent that routine monitoring of serum CRP makes a valu- able contribution to the recognition and management of infectious diseases. Situations in which these applications have been well docu- mented are listed in Box 12.12.1.2. Meningitis is of particular interest in view of its potential severity and the importance of rapid diagnosis and appropriate treatment. Bacterial meningitis is associated with much higher serum CRP levels at presentation than cases of aseptic or proven viral meningitis. The latter frequently have CRP concentrations within the normal range or which are only very slightly raised, unless they develop sec- ondary bacterial infective complications; patients with tuberculous meningitis have intermediate values. Appropriate therapy for either bacterial or tuberculous meningitis causes the CRP ­concentration to fall, and this can be used to monitor objectively the response to treatment. Baseline CRP values are much lower at birth and for the first few days than in older children or adults. Also, neonatal infections pro- gress much more rapidly and can have a fatal outcome before the CRP response has produced concentrations detectable in routine assays. It is therefore essential to use highly sensitive methods cap- able of detecting and precisely measuring CRP in the range 0.05 to 5.0 mg/​litre, otherwise the critical initial acute phase response to in- fection will be missed. Inflammatory disease Most of the chronic inflammatory diseases of unknown aetiology (Box 12.12.1.1), with some notable exceptions described later in this chapter, are associated with high CRP values when they are ac- tive. Serial measurements of CRP in individuals with any of these Fig. 12.12.1.2  An 86-​year-​old woman had been refusing food and drink for 6 weeks. She was dehydrated, but rehydration in hospital failed to improve her mental state. She was paranoid and refused nursing and medical care. Paraphrenia was diagnosed and deterioration continued. A CRP concentration of 130 mg/​litre and a white cell count of 13.5 × 109/​litre were then found. Chest radiography, normal on admission, now showed a cavitating lesion from which 150 ml of pus were aspirated. Intravenous ampicillin reduced neither the CRP nor the white cell count, prompting a change of therapy to gentamicin and metronidazole. Streptococcus equinus was finally identified in the pus and treatment was changed to benzyl penicillin alone. The CRP value then fell exponentially but rather slowly. The patient’s clinical and mental state gradually improved and she was eventually discharged. Box 12.12.1.2  Applications of serum CRP measurement in infectious disease • Pyogenic bacterial infections, including:

—​ bacteraemia and septicaemia in children and adults

—​ bacteraemia and septicaemia in neonates

—​ bacterial and other infections in immunosuppressed patients

—​ bacterial infections after major elective surgery or other invasive procedures

—​ infective relapse after abdominal surgery for sepsis

—​ peritonitis in patients on chronic ambulatory peritoneal dialysis

—​ acute appendicitis (differential diagnosis)

—​ evaluation of antibiotic therapy for female pelvic infection

—​ laryngotracheitis/​pharyngitis/​epiglottitis in children

—​ chorioamnionitis after premature rupture of membranes

—​ disseminated versus localized gonococcal infection

—​ infection precipitating sickle cell crisis • Meningitis (viral < tuberculosis < bacterial) • Deep fungal infection

12.12.1  The acute phase response and C-reactive protein 2203 diseases generally reflect the extent and activity of their condition as determined by clinical examination and by other laboratory tests. Rheumatoid arthritis is the most common and important disease in this group and the correlation between CRP values in individual patients and the extent and activity of arthritis is very well estab- lished. Importantly, there are appreciable differences between the CRP concentrations attained in different subjects with apparently similar severity of arthritis, but in each case the CRP value always reflects current disease activity. Furthermore, CRP values precisely predict future progression of bone erosion and joint damage. Left unchecked, high CRP values are inevitably followed by progressive erosive disease, whereas treatments that reduce the CRP concentra- tion retard or arrest this process. In some of the inflammatory disorders, for example, systemic vas- culitis or Crohn’s disease (Fig. 12.12.1.3), unlike rheumatoid arth- ritis, the pathology is relatively inaccessible to direct examination, and serum CRP measurement provides the best available, objective index of disease activity. Furthermore, the presence or absence of a CRP response can distinguish between symptoms or organ dysfunc- tion that are due to currently active inflammation and those that are the consequence of fibrosis and scarring from previous episodes. This can be very important when treatments include steroids and other powerful and potentially hazardous immunosuppressive, anti-​ inflammatory, and cytotoxic drugs. It permits precise titration of dosages and may help to avoid excessive or unnecessary use. Induction of clinical remission and control of the underlying dis- ease process is associated with prompt normalization of the CRP. However, CRP also becomes abnormal with intercurrent infection, a common complication of some of these disorders and their treat- ments, and this serves to focus diagnostic attention often before the infection has become too severe or even before it is clinically evident. Monitoring the CRP response to antimicrobial therapy can then help to confirm the diagnosis and the efficacy of therapy. Persistent elevation of the CRP after eradication of infection may indicate relapse of the underlying inflammatory disease, requiring additional anti-​inflammatory treatment. Necrosis Untreated acute myocardial infarction is invariably associated with a major CRP response, as is elective embolization leading to necrosis of tumours in the liver and elsewhere. The peak concentration of CRP occurs about 50 h after the onset of pain in acute myocardial infarc- tion patients who do not undergo revascularization, and is earlier and smaller following effective early revascularization. CRP produc- tion usually correlates in magnitude, though not in timing, with the peak serum concentration of the specific myocardial markers, cre- atine kinase MB and troponin. In patients who recover uneventfully, the CRP value falls rapidly towards normal in the usual exponential fashion. However, complications such as persistent cardiac dysfunc- tion, further infarction, aneurysm formation, intercurrent infection, thromboembolism, or postinfarction syndrome are associated with either persistently raised CRP values or a secondary increase after the initial decrease. Myocardial rupture is seen only in patients with high peak CRP values (>200 mg/​litre) and the peak CRP concentration after acute myocardial infarction strongly predicts overall outcome, including survival, in the short, medium, and long term. Stable angina and coronary arteriography investigations do not stimulate CRP production, whereas other relevant causes of chest pain, including pulmonary embolism, pleurisy, or pericarditis, pro- duce raised CRP values. Routine assays of CRP after infarction or in patients with chest pain may thus assist in diagnosis and in the recognition and management of complications, including iatrogenic infection associated with invasive cardiovascular monitoring. Serum CRP concentrations closely reflect the severity and progress of acute pancreatitis, providing a better guide to intra-​abdominal events than other markers such as leucocyte counts, erythrocyte sedimentation rate (ESR), temperature, and the plasma concentra- tions of antiproteinases. A CRP concentration greater than 100 mg/​ litre at the end of the first week of illness is associated with a more prolonged subsequent course and a higher risk of the development of a pancreatic collection. Serial CRP measurements can therefore guide the use of appropriate imaging techniques and help to confirm resolution before discharge from hospital. Trauma and surgery The CRP concentration always rises after significant trauma, sur- gery, or burns, peaking after about 2 days and then falling towards normal with recovery and healing. Infections or other tissue-​ damaging complications alter this normal pattern of CRP response Severe Moderate Mild Improving 40 0 (mg/ day) (183) Prednisolone C-reactive protein (mg/l) ESR (mm in 1h) 100 0 Stool frequency 16 12 8 4 0 July 1979 (days) Hospital admission 1979 (months) 1980 59 60 61 62 63 64 65 66 67 Body weight (kg) Improving 16 14 2 1 2 1 0 1 0 1 2 8 8 6 Fig. 12.12.1.3  A 26-​year-​old man with pancolonic Crohn’s disease.
He was admitted with severe exacerbation; temperature 38°C; pulse
110 beats/​min; 16 stools per day; haematocrit 41.5%, leucocytes
13.8 × 109/​litre. Rectal mucosa severely inflamed with histiocytic granulomas on biopsy. Rapid improvement occurred with oral and rectal prednisolone, ampicillin, and metronidazole, with complete clinical and histological remission on day 11. A relapse 5 months later responded promptly to a short course of oral and rectal prednisolone. CRP and ESR values were both high during the initial exacerbation. The rapid response to treatment was paralleled by a prompt fall in CRP concentration, whereas the ESR responded more slowly. Despite clinical remission and a normal ESR, the CRP remained slightly elevated, suggesting persistent low-​grade inflammatory activity, and it rose further during a subsequent relapse when the ESR did not change. Reproduced from Fagan A, et al. (1982). Serum levels of C-​reactive protein in Crohn’s disease and ulcerative colitis. Eur J Clin Invest, 12, 351–​9, with permission.

section 12  Metabolic disorders 2204 and the failure of the CRP to continue falling, or the appearance of a second peak, may precede clinical evidence of intercurrent infection by 1 to 2 days. Serial prospective CRP measurement is therefore ad- visable in such patients. Malignancy Most malignant tumours, especially when they are extensive and metastatic, induce an acute phase response. This is particularly so with those neoplasms that cause systemic symptoms such as fever and weight loss, for example, Hodgkin’s disease (stage B) and renal carcinoma, but raised CRP values are seen with many others. In some studies, notably of prostatic carcinoma and bladder car- cinoma, the CRP concentration at presentation has been found to correlate with the overall tumour load and also with the prognosis, being higher for a given mass of tumour in those patients who sub- sequently fare worse. The CRP value may also correlate better with progress and regression of tumour than other, more specific tumour markers. However, given the nonspecific nature of the acute phase response and the limited number of adequate studies performed a definite role for CRP measurement in the management of cancer patients, other than in cases of intercurrent infection, has not been established. Allograft rejection In the era before routine immunosuppression with combined ciclosporin and steroid treatment, rejection episodes following renal allografting were generally associated with increased production of CRP. However, this treatment almost completely suppresses the CRP response to allograft rejection. In contrast, the acute phase response of serum amyloid A protein is unaffected and, importantly, intercur- rent infection still stimulates greatly increased production of both CRP and serum amyloid A protein. Conditions associated with minor increase in serum CRP concentration Despite unequivocal evidence of active inflammation and/​or tissue damage, the conditions listed in Box 12.12.1.3 are usually associ- ated with only minor increases in the serum CRP concentration, and in many cases it may even remain normal in the face of severe disease. The contrasts of systemic lupus erythematosus (SLE) with rheumatoid and other arthritic conditions shown in Box 12.12.1.1, and between ulcerative colitis and Crohn’s disease, are very striking. However, intercurrent microbial infection provokes a major CRP response in all the conditions shown in Box 12.12.1.3, and this is of great value in diagnosis and management, especially in SLE and leukaemia. The basis of the apparently selective failure of the acute phase response of CRP (which is also shown by serum amyloid A protein) is not known, but presumably involves defect(s) in the pathways that mediate the acute phase response to autologous in- flammation and tissue damage. Pyrexia is common in SLE and may be caused by microbial in- fection or by activity of the lupus itself. Both SLE and its treatment predispose to infection, and steroids and immunosuppressive drugs can mask the usual symptoms and signs of infection. Furthermore, infection can trigger exacerbations of SLE. This is a serious clinical situation and infection remains one of the most common causes of death in patients with SLE. CRP values of 60 mg/​litre or more are very rare in SLE in the absence of infection, whereas values below 60 mg/​litre are seen in patients with documented infection only when it is rather mild and often localized (e.g. to the skin or lower urinary tract). Differential diagnosis and management of fever in SLE are thus considerably improved by the measurement of serum CRP concentration (Fig. 12.12.1.4). Box 12.12.1.3  Conditions associated with minor increases in serum CRP concentration • Systemic lupus erythematosus • Scleroderma • Dermatomyositis • Ulcerative colitis • Leukaemia • Graft-​versus-​host disease 41 37 °C 150 100 50 0 November ESR C-reactive protein ESR (mm in first hour) (mg/l) Fever Rash Arthritis Cefuroxime Gentamicin Abdominal pain Diarrhoea E. coli septicaemia Pulse doses methylprednisolone Prednisolone 30 mg/day Jan Dec 28 22 16 4 Sept Jun 0 1 ra M Fig. 12.12.1.4  A 12-​year-​old girl with a 3-​year history of SLE; recurrent febrile episodes, polyarthritis, cutaneous vasculitis, and episodes of asymptomatic bacteriuria. Intermittent treatment was with prednisolone, azathioprine, and plasma exchange. Serum CRP concentration was only marginally increased throughout but erythrocyte sedimentation rate was persistently high. Fever recurred with diarrhoea and abdominal pain. All microbial cultures were negative except for growth of Escherichia coli from the urine. Despite oral cefalexin and prednisolone her condition deteriorated, with severe neutropenia, probably due to azathioprine. The CRP value rose from 36 to 101 mg/​litre and then 137 mg/​litre, and at this stage her blood culture grew Escherichia coli. Intravenous antibiotics were given and the serum CRP concentration fell rapidly, but there was little clinical improvement. Active SLE appeared then to be the sole cause of the fever and this was confirmed by the development of a diffuse vasculitic rash and polyarthritis. Three pulse doses of methylprednisolone were given intravenously on successive days and produced a dramatic improvement in her clinical state with resolution of the fever. This case illustrates (1) the differential response of CRP to fever resulting from activity of SLE alone and fever due to bacterial infection; (2) the rapid response of CRP both to the onset and to the effective treatment of serious bacterial infection; and (3) the failure of ESR measurements to provide any useful information in this complex and rapidly evolving clinical situation. Reproduced from Pepys MB, Langham JG, de Beer FC (1982). C-​reactive protein in systemic lupus erythematosus. Clin Rheum Dis, 8, 91–​103, with permission.

12.12.1  The acute phase response and C-reactive protein 2205 The reason why leukaemia patients fail to mount more than a modest CRP response, even during induction therapy when there is massive death of leukaemia cells, is not known. However, the CRP response to infection is normal. Since all febrile episodes in leu- kaemia must initially be treated as infective, the main value of CRP monitoring is to determine the response to therapy and assist in de- cisions about its duration. Acute or chronic graft-​vs-​host disease after bone marrow transplantation is usually associated with only a modest CRP response, if any. However, the immunosuppressive treatments used to prevent bone marrow rejection and to control graft-​vs-​host disease render the patients susceptible to intercurrent infections, often with unusual microorganisms, and these are always associated with high CRP values. CRP monitoring is therefore very useful in post-​transplant management. Interpretation of clinical serum CRP measurements The CRP response is not specific and CRP measurements on their own can therefore never be diagnostic of any particular condition, nor should they be used in isolation for any other clinical purpose. The CRP value can only be interpreted in the light of all other avail- able clinical and laboratory information. Provided this is done, it can make a most useful contribution to overall assessment of the patient and determination of the best management. The applications fall into three main categories: • Screening for organic disease • Monitoring of extent and activity of disease—​infection, inflam- mation, malignancy, and necrosis • Detection and management of intercurrent infection Screening for organic disease CRP production is a very sensitive response to organic disease. A normal CRP value therefore eliminates many possible types of path- ology and is a reassuring finding. Those serious conditions that stimu- late CRP production only weakly, if at all, for example, SLE, ulcerative colitis, and leukaemia, are all readily recognized by clinical examin- ation and simple routine investigations. The presence of a raised CRP value is unequivocal evidence of active pathology, though this may not necessarily be the cause of the complaint for which the patient pre- sented. Such a finding, in the absence of other obvious abnormality, warrants a repeat CRP assay after a few days when a trivial cause such as an upper respiratory tract infection will have resolved. Further in- vestigation of a persistently raised CRP concentration will then de- pend on the severity of the complaint and other clinical findings. Monitoring extent and activity of disease Once the diagnosis is established, in those diseases which cause major CRP production, serial measurements reflect activity and re- sponse to treatment and can be used for monitoring. However, they can only be interpreted provided other possible intercurrent causes of an acute phase response, particularly infections, are excluded. Detection and management of intercurrent infection Production of CRP is a very sensitive response to most forms of infection and a raised concentration is thus a useful guide to the possible presence of infection in otherwise normal subjects or in- dividuals with a primary condition that predisposes to infection. In disorders that themselves increase the CRP concentration, the decision as to whether infection is present or not must depend on clinical examination and other laboratory tests; the role of CRP testing is then to demonstrate rapidly and objectively whether there is a response to whatever treatment is used. Effective antimicrobial therapy of infection is always associated with a prompt fall in the CRP, whereas persistent CRP production indicates continuing infec- tion and/​or activity of the underlying disease. There is no other ob- jective test that yields this sort of information so accurately. Changes in the results of clinical examinations and tests of organ function usually lag hours or days behind the CRP response. Other considerations CRP and body temperature The acute phase response, which is best measured clinically by quan- tification of the serum CRP concentration, is part of the systemic re- sponse to disease. Monitoring of this same response by measurement of body temperature is an integral part of routine clinical manage- ment. CRP production is triggered by the same cytokines that cause fever, and the serum CRP concentration is therefore in part a bio- chemical surrogate for the body temperature. However, the CRP re- sponse is not susceptible to the many vagaries of thermoregulation and of routine clinical measurement of body temperature. The precise numerical value of the CRP concentration and its changes over time reflect much more accurately than the temperature the intensity of the underlying stimulus. Furthermore, there is often a CRP response in the absence of fever, especially in neonates and older people, and also in many chronic inflammatory conditions at any age. There is thus a good case for charting serial serum CRP concentrations along- side the standard temperature chart in appropriate patients. CRP or ESR? The only other comparable nonspecific index of disease that is rou- tinely measured is the ESR. The ESR reflects, in part, the intensity of the acute phase response, especially that of fibrinogen and the α-​globulins, but is also largely determined by the concentration of immunoglobulins, which are not acute phase reactants. These pro- teins all have half-​lives of days to weeks. ESR thus changes very much more slowly than the CRP concentration, and it rarely reflects precisely the clinical status of the patient at the time of testing. ESR is also dependent on the number and morphology of the red cells, which bear no relationship to the acute phase response. In addition, there is a significant diurnal variation in ESR, depending on food intake, which is not seen in the CRP concentration. Finally, the dy- namic range of the ESR is much less than that of CRP concentration and the precision and reproducibility of ESR measurement is poor compared to the robust clinical chemistry immunoassays available for CRP. The ESR is therefore of limited use as an objective index of disease activity on which management decisions can be based. In all clinical situations that have been carefully evaluated, ran- ging from acute bacterial infections to the chronic remittent in- flammatory diseases, such as Crohn’s disease, rheumatoid arthritis, and other inflammatory arthropathies, and systemic vasculitis in its various forms, frequent prospective measurements of CRP reflect disease activity very much more closely than measurements of the ESR. However, the ESR remains a useful screening test for the detec- tion of paraproteinaemias, especially multiple myeloma, in which an acute phase response is often absent.

section 12  Metabolic disorders 2206 CRP and cardiovascular disease The 1994 report of a prognostic association between increased CRP and serum amyloid A protein values and outcome in severe un- stable angina, and the discovery of a significant predictive associ- ation between baseline CRP values in the general population and future coronary events, triggered an avalanche of work in this field. These observations became increasingly controversial, but the re- cent publications of very large-​scale observational and genetic epidemiological studies have eventually resolved the major issues. The key questions are whether the measurement of baseline CRP concentration provides information useful for the assessment and management of cardiovascular disease risk, and whether CRP itself contributes to the pathogenesis of atherosclerosis, atherothrombosis, and/​or ischaemic tissue injury. The possible involvement of CRP in atherogenesis was first sug- gested by the binding of CRP to low-​density lipoprotein and the pres- ence of CRP in atherosclerotic lesions. In recent years, a wide range of proinflammatory and cell-​activating effects have been claimed for CRP, based on in vitro studies with various cell types. Unfortunately, nearly all this work was done with commercial CRP preparations produced in recombinant bacteria, and none of the positive obser- vations have been reproducible with authentic pure human CRP isolated from human source material. The amazing range of potent proinflammatory and cell-​activating properties ascribed to CRP is not consistent with the fact that neither the administration of large amounts of pure human CRP in normal healthy animals nor the 1000-​fold natural acute phase response of CRP in patients are as- sociated with any such effects. Furthermore, intravenous infusion of authentic, pure, pharmaceutical grade human CRP into healthy adult human volunteers had no proinflammatory, proatherogenic, or any other adverse effects. In experimental animal models of ath- erosclerosis, CRP either has no effect on atherogenesis in vivo or is atheroprotective. Human epidemiological studies of atherosclerosis burden have had varying results, but overall provide no compelling evidence for an association with CRP values. Baseline CRP values are significantly associated with all the known risk factors and pathogenetic mechanisms for coronary heart disease events, and about 70% of the variance in baseline CRP is ascribable to these factors. Although CRP concentration is thus not independently associated with cardiovascular disease risk, a statistically significant association remains even after maximal ad- justment. However, the level of association is markedly less than was originally reported and is comparable with the association with cardiovascular disease risk of other nonspecific systemic markers of inflammation, such as low plasma albumin, raised white cell count, ESR, and serum amyloid A  protein. The Emerging Risk Factors Collaboration meta-​analysis of 52 major epidemiological studies of baseline CRP values and cardiovascular disease prediction, showed that CRP measurement adds almost no useful information to risk assessment, potentially helping to prevent only one additional event over a period of 10 years for every 400–​500 people screened. Also, since statins lower risk when administered at any level of low-​density cholesterol and in all subgroups of the population, re- gardless of intercurrent disease or additional risk factors, there is no justification for the use of CRP measurement to select patients for statin treatment. Indeed, measuring the exquisitely nonspecific CRP in this context, without comprehensive review of a patient with a raised value, risks missing other important pathologies. The unfortunate conflation of association with causality triggered much speculation about whether CRP is a pathogenetic factor for cardiovascular disease events. However, Mendelian randomization genetic epidemiological studies looking at hereditary polymorphisms associated with higher or lower baseline CRP values all show no ­association with cardiovascular disease risk. This negative outcome is entirely consistent with the negative in vivo animal studies of CRP and atherogenesis. In contrast, experimental animal studies robustly show that human CRP can exacerbate pre-​existing ischaemic injury via a complement dependent mechanism and that this can be blocked by experimental drugs that inhibit CRP function. Preliminary clinical studies of extra-corporeal absorption of CRP, to lower the circulating CRP concentration, are in progress in patients with acute myocardial infarction. Future testing of more effective novel CRP blocking drugs should demonstrate whether this mechanism is clinically relevant. Serum amyloid A protein Serum amyloid A  protein, an apolipoprotein of high-​density lipo- protein particles, is a marked acute phase reactant, its concentration rising from normal levels of about 2 mg/​litre by as much as 1000-​ fold. It is essential to monitor and control serum amyloid A protein levels in patients with reactive systemic, AA type amyloidosis (see Chapter 12.12.3). It is a critical marker of disease control in patients with hereditary periodic fever syndromes, to ensure that they do not develop systemic AA amyloidosis. Serum amyloid A protein concen- tration is also the most sensitive marker of rejection episodes in renal allograft recipients and is useful in routine monitoring of these patients. FURTHER READING Boralessa H, et al. (1986). C-​reactive protein in patients undergoing cardiac surgery. Anaesthesia, 41, 11–​15. Casas JP, et al. (2008). C-​reactive protein and coronary heart disease: a critical review. J Intern Med, 264, 295–​314. C-​reactive Protein Coronary Heart Disease Genetics Collaboration (CCGC) (2011). Association between C reactive protein and cor- onary heart disease: mendelian randomisation analysis based on in- dividual participant data. BMJ, 342, d548. Elliott P (2009). Genetic loci associated with C-​reactive protein levels and risk of coronary heart disease. JAMA, 302, 37–​48. Emerging Risk Factors Collaboration (2009). C-​reactive protein concentration and risk of coronary heart disease, stroke and mor- tality: an individual participant meta analysis. Lancet, 375, 132–​40. Emerging Risk Factors Collaboration (2009). C-​reactive protein, fi- brinogen and cardiovascular disease prediction. N Engl J Med, 367, 1310–​20. Fagan EA, et al. (1982). Serum levels of C-​reactive protein in Crohn’s disease and ulcerative colitis. Eur J Clin Invest, 12, 351–​60. Griselli M, et al. (1999). C-​reactive protein and complement are im- portant mediators of tissue damage in acute myocardial infarction. J Exp Med, 190, 1733–​9. Hartmann A, et al. (1997). Serum amyloid A protein is a clinically useful indicator of acute renal allograft rejection. Nephrol Dial Transplant, 12, 161–​6. Kushner I, Rzewnicki D, Samols D (2006). What does minor elevation of C-​reactive protein signify? Am J Med, 119, 166.e117–​28. Lane T, et al. (2014). Infusion of pharmaceutical-​grade natural human C-​reactive protein is not pro-​inflammatory in healthy adult human volunteers. Circ Res, 114, 672–​6.

12.12.2 Hereditary periodic fever syndromes 2207

12.12.2 Hereditary periodic fever syndromes 2207

12.12.2  Hereditary periodic fever syndromes 2207 Liuzzo G, et al. (1994). The prognostic value of C-​reactive protein and serum amyloid A protein in severe unstable angina. N Engl J Med, 331, 417–​24. Lowe GDO, Pepys MB (2006). C-​reactive protein and cardiovascular disease: weighing the evidence. Curr Atheroscler Rep, 8, 421–​8. Pepys MB (2005). CRP or not CRP? That is the question. Arterioscler Thromb Vasc Biol, 25, 1091–​4. Pepys MB, Hirschfield GM (2003). C-​reactive protein: a critical up- date. J Clin Invest, 111, 1805–​12. Pepys MB, Lanham JG, de Beer FC (1982). C-​reactive protein in sys- temic lupus erythematosus. Clin Rheum Dis, 8, 91–​103. Pepys MB, et al. (2005). Proinflammatory effects of bacterial recombinant human C-​reactive protein are caused by contamination with bacterial products, not by C-​reactive protein itself. Circ Res, 97, e97–​103. Pepys MB, et al. (2006). Targeting C-​reactive protein for the treatment of cardiovascular disease. Nature, 440, 1217–​21. Pepys MB (2018).The Pentraxins 1975–2018: Serendipity, Diagnostics and Drugs. Front Immunol, 9, 2382. Ridker PM, et al. (1997). Inflammation, aspirin, and the risk of cardio- vascular disease in apparently healthy men. N Engl J Med, 336, 973–​9. Riess W, et al. (2018). First in man: case report of selective C-reactive pro- tein apheresis in a patient with acute ST segment elevation myocardial infarction. https://www.hindawi.com/journals/cric/2018/4767105/ Simons JP, et al. (2014). C-​reactive protein is essential for innate resist- ance to pneumococcal infection. Immunology, 142, 414–​20. Starke ID, et al. (1984). Serum C-​reactive protein levels in the man- agement of infection in acute leukaemia. Eur J Cancer, 20, 319–​25. van Leeuwen MA, et al. (1997). Individual relationship between pro- gression of radiological damage and the acute phase response in early rheumatoid arthritis. Towards development of a decision sup- port system. J Rheumatol, 24, 20–​7. Wasunna A, et al. (1990). C-​reactive protein and bacterial infection in preterm infants. Eur J Pediatr, 149, 424–​7. 12.12.2  Hereditary periodic fever syndromes Helen J. Lachmann, Stefan Berg,
and Philip N. Hawkins ESSENTIALS The hereditary periodic fever syndromes or hereditary auto­ inflammatory diseases are disorders of innate immunity that mostly present in childhood and are characterized by recurrent, self-​ limiting, seemingly unprovoked episodes of fever and systemic in- flammation that occur in the absence of autoantibody production or identifiable infection. Disorders include (1) familial Mediterranean fever (FMF), due to mutations in the gene encoding pyrin; (2) tumour necrosis factor (TNF) receptor-​associated periodic syndrome (TRAPS), due to mu- tations in a gene for a TNF receptor; (3) mevalonate kinase defi- ciency (MKD), caused by mutations in the mevalonate kinase gene; and (4) the cryopyrin-​associated periodic syndromes (CAPS), which include (a) familial cold urticarial syndrome, (b) Muckle–​Wells syn- drome, and (c)  chronic infantile neurological, cutaneous, and ar- ticular syndrome. With advances in genetics, further syndromes are continually being recognized. These are all extremely rare and in the majority are only known to affect a handful of kindred or individuals. Understanding of the molecular pathogenesis of these disorders provides unique insights into the regulation of innate immunity and inflammation. Diagnosis relies on recognition of suggestive clinical features (e.g. fever with peritonitis and/​or pleurisy, arthralgia/​arthritis) that are al- most always accompanied by a substantial acute phase response, and is supported by genetic testing. With the exception of FMF, which is a common disease in certain geographic areas, hereditary periodic fever syndromes are rare and easily overlooked in the differ- ential diagnosis of recurrent fevers. Clinical features and management—​attacks can be mild to debili- tating and short to prolonged, while their most feared complication is AA amyloidosis. Effective therapies are available for some syndromes, for example: (1) FMF—​daily prophylactic colchicine prevents clinical attacks and susceptibility to AA amyloidosis, (2)  CAPS—​treatment with anti-​IL-​1 agents produces rapid and often complete clinical and serological remission, and (3) TRAPS—​anti-​IL therapies are extremely effective. Introduction The hereditary periodic fever syndromes are a group of multisystem disorders characterized by fluctuating or irregularly recurring epi- sodes of fever and systemic inflammation, notably affecting the joints, eyes, skin, and serosal surfaces. More than 30 syndromes are now recognized; many of these are extremely rare and only af- fect a handful of individuals. Diseases affecting more than 1 per million, and therefore likely to be encountered in specialist prac- tice, include familial Mediterranean fever (FMF), tumour necrosis factor (TNF) receptor-​associated periodic syndrome (TRAPS), mevalonate kinase deficiency (MKD, previously known as the hyperimmunoglobulin D and periodic fever syndrome (HIDS)), and the cryopyrin-​associated periodic syndromes (CAPS). The latter are a spectrum of three hitherto apparently distinct dis- orders of increasing severity:  familial cold urticarial syndrome (now known as familial cold autoinflammatory syndrome; FCAS), Muckle–​Wells syndrome (MWS), and chronic infantile neuro- logical, cutaneous, and articular syndrome (CINCA). The latter is also known in the United States of America as neonatal-​onset multisystem inflammatory disorder (NOMID). Although many symptoms of these disorders are shared, there are clear distinctions in the pattern of their inheritance, the dur- ation and frequency of attacks, and their overall clinical picture (Table 12.12.2.1). With a few exceptions, these diseases are usu- ally compatible with normal life expectancy, though with the ever-​looming grave threat of AA amyloidosis. Recent insights into their molecular pathogenesis have led to enhanced diagnosis through DNA analysis, the development of rational therapies, and have shed important new light on regulation of the innate ­immune system.

section 12  Metabolic disorders 2208 Table 12.12.2.1  Features of inherited periodic fever syndromes Periodic fever syndrome Gene, chromosome OMIM Mode of inheritance Predominant population Usual age at onset Potential precipitants of attacks Distinctive clinical features Typical duration of attacks Typical frequency of attacks Characteristic laboratory abnormalities Treatment FMF MEFV, chromosome 16 249100 608107 Autosomal recessive (dominant in rare families) Eastern Mediterranean Childhood/​ early adulthood Usually none Occasionally menstruation, fasting, stress, or trauma Short severe attacks of peritonitic and/​or pleuritic pain, colchicine-​ responsive, erysipelas-​like erythema 1–​3 days Variable Marked acute phase response during attacks Colchicine TRAPS TNFRSF1A, chromosome 12 142680 191190 Autosomal dominant (can be de novo) Northern European, but reported in many ethnic groups Childhood/​ early adulthood Usually none Prolonged symptoms More than a week (may be very prolonged) Variable (may be continuous) Marked acute phase response during attacks Low levels of soluble TNFR1 when well Anti-​IL-​ 1therapies High-​dose corticosteroids MKD MVK, chromosome 12 260920 251170 Autosomal recessive Northern European, but reported in many ethnic groups Infancy Immunizations Diarrhoea and lymphadenopathy. 3–​7 days 1–​2 monthly Acute phase response, and mevalonate aciduria during attacks, elevated IgD and IgA Anti-​IL-​1 therapies, anti-​ TNF therapies, anti-​IL-​6 therapies FCAS NLRP3, chromosome 1 120100 606416 Autosomal dominant Northern but reported in many ethnic groups European, Neonatal/​ infancy Exposure to cold environment Cold-​induced fever, arthralgia, rash, and red eye 24–​48 h Depends on environmental factors Acute phase response during attacks; to a lesser extent when well Cold avoidance, anti-​IL-​1 therapies MWS NLRP3, chromosome 1 191900 606416 Autosomal dominant, Northern European, but reported in many ethnic groups Neonatal/​ infancy Marked diurnal variation Cold environment, but less marked than in FCAS Urticarial rash, red eye, sensorineural deafness Continuous (often worse in the evenings) Often daily Varying but marked acute phase response most of the time Anti-​IL-​1 therapies CINCA/​ NOMID NLRP3, chromosome 1 607115 606416 Sporadic Northern European, but reported in many ethnic groups Neonatal/​ infancy None Urticarial rash, aseptic meningitis, deforming arthropathy, sensorineural deafness, developmental damage Continuous Continuous Varying but marked acute phase response most of the time Anti-​IL-​1 therapies PAPA PSTPIP1 (CD2BP1), chromosome 15 604416 606347 Autosomal dominant Northern European, but reported in many ethnic groups Childhood None Pyogenic arthritis, pyoderma gangrenosum, and cystic acne Intermittent attacks with migratory arthritis Variable (may be continuous) Acute phase response during attacks Anti-​TNF therapies

12.12.2  Hereditary periodic fever syndromes 2209 (continued) Blau’s syndrome NOD2 (CARD15), chromosome 16 605956 186580 Autosomal dominant None Age <5 years None Granulomatous polyarthritis, iritis, and dermatitis Continuous Continuous Sustained modest acute phase response Corticosteroids, some reports of benefit from anti-​TNF agents FCAS2 NLRP12, chromosome 19 611762 609648 Autosomal dominant South/​Central American, but reported in other ethnic groups Neonatal Sometimes exposure to cold Urticarial rash, fever, arthralgia Intermittent Variable Acute phase response during attacks None know DIRA IL1RN, chromosome 2 612852 147679 Autosomal recessive Northern Europe and Central America Infancy None Pustular rash, sterile osteomyelitis of long bones Continuous Continuous Sustained acute phase response Anakinra (recombinant IL-​1 receptor antagonist) Majeed’s syndrome LIPN2, chromosome 18 609628 605519 Autosomal recessive Middle East Infancy None Recurrent multifocal osteomyelitis, congenital dyserythropoietic anaemia, neutrophilic dermatosis 2-​7 days Variable Acute phase response during attacks NSAIDs and corticosteroids. Possible benefit from IL-​1 blocking agents DITRA IL36RN, chromosome 2 614204 605507 Autosomal recessive None Variable—​ from infancy onwards Drug initiation or withdrawal, menstrual cycle, pregnancy Recurrent fever, generalized rash, and disseminated pustules often but not always on a background of psoriasis Intermittent Variable Acute phase response during attacks, typical spongiform pustules on skin biopsy Possible benefit from IL-​1 blocking agents CANDLE/​JMP PSMB8, chromosome 6 256040 177046 Autosomal recessive None Infancy None Lipodystrophy, characteristic rash, swollen lips, violaceous eyelids Continuous Continuous Acute phase response, typical appearances on skin biopsy None established? JAK inhibitors Early-​onset inflammatory bowel disease/​IL-​10R deficiency IL10RA, chromosome 11, IL10RB, chromosome 21, or IL-​10, chromosome1 613148 612567 612381 Autosomal recessive None Infancy None Very early-​ onset severe inflammatory bowel disease with perianal fistulae Continuous Continuous Acute phase response, colitis with granulomas Stem cell transplantation APLAID PLCG2, chromosome 16 614878 600220 Autosomal dominant Single family Infancy Rash worse with heat or sun exposure Recurrent blistering rash and mild humoral immune deficiency with sinopulmonary infections Continuous Continuous Acute phase response, dense inflammatory infiltrate in skin, decreased IgM and IgA, no autoantibodies Partial response to IL-​1 blockade and corticosteroids

section 12  Metabolic disorders 2210 Periodic fever syndrome Gene, chromosome OMIM Mode of inheritance Predominant population Usual age at onset Potential precipitants of attacks Distinctive clinical features Typical duration of attacks Typical frequency of attacks Characteristic laboratory abnormalities Treatment HOIL-​1 deficiency RBCK1, chromosome 20 615895 Autosomal recessive Two European families Infancy None Polyglucosan myopathy and cardiomyopathy with immunodeficiency and bacterial infection and autoinflammation Intermittent Intermittent Acute phase response, intracellular glycogen inclusions, on muscle biopsy All 4 cases died in childhood—​ symptomatic improvement with corticosteroids, allogenic stem cell transplantation attempted DADA2 CECR1, chromosome 22 615688 Autosomal recessive None Childhood None Systemic and local polyarteritis nodosa, livedo racemose, and early-​onset stroke Intermittent Intermittent Acute phase response during attacks Anti-​TNF SAVI TMEM173, chromosome 5 615934 Autosomal dominant None Infancy None Rash on cheeks, ears, nose, and digits with cartilage damage, failure to thrive, interstitial lung disease Continuous Continuous Capillary inflammation, acute phase response None established? JAK inhibitors APLAID, autoinflammation and PLCγ2-​assocaited antibody deficiency and immune dysregulation, CINCA/​NOMID, chronic infantile neurological, cutaneous, and articular syndrome/​neonatal-​onset multisystem inflammatory disorder; DADA2, deficiency of adenosine deaminase 2; DIRA, deficiency of the IL-​1 receptor antagonist; DITRA, deficiency of the IL-​36 receptor antagonist; FCAS, familial cold autoinflammatory syndrome; FMF, familial Mediterranean fever; HOIL-​1, haem-​oxidized IRP2 ubiquitin ligase 1; IL-​1, interleukin 1; MKD, mevalonate kinase deficiency; MWS, Muckle–​Wells syndrome; PAPA, pyogenic arthritis, pyoderma gangrenosum, and acne syndrome; SAVI, STING-​associated vasculopathy with onset in infancy; TNF, tumour necrosis factor; TNFR1, tumour necrosis factor receptor 1; TRAPS, tumour necrosis factor receptor-​associated periodic syndrome. Table 12.12.2.1  Continued

12.12.2  Hereditary periodic fever syndromes 2211 Historical perspective Although the hereditary periodic fever syndromes have only been identified as such during the last few decades, there are various ancient references to them, particularly FMF. Perhaps the earliest extant clinical description is found in William Heberden’s 1802 Commentaries on History and Care of Disease: ‘Pains which are regu- larly intermittent, the fits of which return periodically as those of an ague; such as I have known in the bowels, stomach, breast, loins, arms and hips, though it be but seldom that such parts suffer in such a manner.’ Familial Mediterranean fever Genetics FMF is predominantly inherited in a recessive manner. The gene as- sociated with FMF, MEFV, which encodes a protein called pyrin, was identified through positional cloning in 1997. MEFV is ex- pressed in neutrophils, monocytes, dendritic cells, and fibroblasts. Expression is up-​regulated in response to inflammatory activators such as interferon-​γ and TNFα. The more than 40 MEFV mutations that are associated with FMF encode either single amino acid substi- tutions or small deletions. The mutations that cause FMF are mostly in exon 10, but also occur elsewhere, particularly in exons 1, 3, 5, and 9. Mutations in each of the two MEFV alleles can be identified in most patients with FMF. Most individuals with a single mutated allele re- main healthy carriers but heterozygote FMF is well recognized and may account for up to 20% of cases. While it is inherently likely that different mutations will affect the function of a protein to differing extents, several findings suggest that the methionine residue at pos- ition 694 is especially important. Five different pathogenic exon 10 mutations involving Met694 have been identified, and individuals homozygous for Met694Val have particularly severe disease. Simple heterozygous deletion of this residue is associated with autosomal dominant FMF of variable penetrance in the British population. More extensive disruption of a single MEFV allele by one or more mutations may account for other rare reports of dominant FMF. Much more commonly, FMF affecting more than one generation of a family is pseudodominantly inherited, reflecting consanguinity or a high local prevalence of the heterozygous carrier state. One particular pyrin variant, E148Q encoded in exon 2, is ex- tremely frequent in Asian populations, with an allele frequency of 10 to 20%, and occurs in other populations at a much lower fre- quency. Although pyrin E148Q can cause typical FMF when coupled with various exon 10 mutations, homozygosity for E148Q alone is thought not to be associated with disease in the vast majority of cases. There is, however, a suggestion that the presence of pyrin E148Q might intensify non-​FMF types of inflammation. Pathology Unusually, given the recessive inheritance, the mutations underlying FMF appear to result in gain of function. Pyrin is a key component of an inflammasome which activates caspase-​1 by autoproteolysis. Maturation of interleukin (IL)-​1β and IL-​18 requires cleavage by caspase 1 which also induces macrophage pyroptosis. Assembly of inflammasomes is triggered by recognition of intracellular pathogen or damage-​associated molecular patterns. Recent work suggests that pyrin is an indirect sensor of a wide variety of bacterial toxins; pyrin interacts, via its N-​terminal death domain, with cytoskeleton microtubules and appears to detect pathological perturbation of actin polymerization dynamics caused by bacterial modification or inactivation of Rho GTPases. More than 20 other proteins that have homology with pyrin’s N-​terminal sequence are now classified generically to have a pyrin domain and are members of the death domain superfamily. They play important roles in the assembly and activation of apoptotic and inflammatory complexes through homotypic protein–​protein inter- actions. Proteins with pyrin domains are involved in inflammation, apoptosis, and nuclear factor (NF)-​κB signalling and have been im- plicated in pathways in CAPS as well. Epidemiology FMF occurs worldwide, though predominantly in populations arising from the eastern Mediterranean basin, particularly non-​ Ashkenazi Jews, Armenians, Turks, and Levantine Arabs. The prevalence of FMF has been estimated to be 1 in 250 to 1 in 500 among non-​Ashkenazi Jews and 1 in 1000 in the Turkish popula- tion. The carrier frequency is as high as one in five among Armenian, Turkish, and North African Jewish populations, fuelling speculation that the FMF trait may have conferred survival benefit, most likely through enhanced resistance to microbial infection mediated by up-​ regulation of the innate inflammatory response. Males and females are equally affected. FMF usually presents in childhood, in 60% be- fore the age of 10 years and in 90% by 20 years. Clinical features Attacks of FMF occur irregularly at variable frequencies and may be precipitated by physical and emotional stress, menstruation, and diet. The onset is rapid and symptoms resolve spontaneously within 6 to 72 h. Fever with peritonitis and/​or pleurisy are the hallmark features, but occur with widely varying intensity from very mild to severely incapacitating. The clinical picture may mimic an acute surgical abdomen with ileus and vomiting, and 40% of patients undergo laparoscopy before the diagnosis is made. Pleuritic attacks occur in 40% of patients, characteristically unilaterally, either alone or in association with peritonitis. Pericarditis is rare and cardiac tamponade extremely rare. Headache with meningism is reported, but generally the nervous system is not involved in attacks. Orchitis occurs in less than 5% of males, most commonly in early childhood, when it can be confused with torsion of the testis. Transient arth- ralgia in lower-​limb joints is not infrequent in acute attacks, and usually subsides within 2 to 4 days; dramatic acute oligoarthritis of the large joints of the lower limb can occur but chronic destructive inflammation is rare. There is a rare but genuine association be- tween FMF associated with MEFV M694V and chronic sacroiliitis. A characteristic erysipelas-​like erythematous rash occurs in 20% of patients (Fig. 12.12.2.1), usually on the lower extremities. Myalgia can be part of the constitutional upset during acute attacks, but up to one-​fifth of patients complain of persistent muscle pain on exer- tion, usually affecting the calves. The rare but distinct syndrome of protracted febrile myalgia presents as severe pain, mainly affecting the lower limbs or abdominal musculature; symptoms may persist for weeks and be accompanied by a vasculitic rash, but usually re- spond to corticosteroids.

section 12  Metabolic disorders 2212 Clinical investigation Acute attacks are accompanied by a number of laboratory abnor- malities including neutrophilia, raised ESR, and a dramatic acute phase response. Investigations may be required to exclude other potential causes of symptoms but, in general, imaging by radiog- raphy, ultrasonography, or echocardiography during acute attacks is unrewarding. The results of genetic testing must be interpreted with care, since some individuals with paired pathogenic MEFV mutations remain completely healthy, and others with apparent carrier status develop classical FMF. Furthermore, MEFV spans 10 exons, and most diag- nostic laboratories offer only limited screening. However, in classical populations, the absence of a mutation in exon 10 makes a diagnosis of FMF unlikely. Treatment Supportive treatment, including analgesia, may be required in acute attacks, but the mainstay of therapy in FMF is prophylaxis with colchicine, a serendipitous discovery made by Goldfinger in 1972. Continuous treatment with colchicine 1 to 2 mg daily prevents or substantially reduces the symptoms of FMF in at least 95% of cases. Recent data also suggest that colchicine acts by a number of mech- anisms including competition with pyrin for cleavage by caspase 1 thereby reducing the effect of N-​terminal cleavage pyrin in enhan- cing NF-​κB activation in FMF. Long-​term use of colchicine is advisable in every patient with FMF and mandatory in those with AA amyloidosis. Although colchicine is extremely toxic in large overdose, the small regular doses required for the treatment of FMF are very well tolerated. Particularly at dose initiation colchicine may cause diarrhoea which often responds to a lactose-​free diet. Despite theoretical concerns about antimitotic po- tential, large cohort studies are very reassuring about the use of col- chicine throughout conception and pregnancy. The concentration of colchicine in breast milk is sufficiently low to permit breastfeeding. Acute initiation or increase of the dose of colchicine does not usually ameliorate acute attacks. In the few patients who are genuinely resistant to colchicine, there have been reports of benefit in such patients following treat- ment with IL-​1 blocking agents and a number of small trials have confirmed this. Tumour necrosis factor receptor-​associated periodic syndrome Genetics TRAPS is an autosomal dominant disease associated with mutations in the 10-​exon TNF receptor superfamily 1A gene (TNFRSF1A) on chromosome 12p13. A disproportionate number of the 50 or so as- sociated mutations disrupt the coding of cysteine residues in the first and second extracellular domains. Pathology TNF is a key mediator in the inflammatory response, with several activities including increased expression of adhesion molecules, induction of cytokine secretion, and activation of leucocytes. TNF receptor 1 (TNFR1) is a member of the death domain superfamily and comprises an extracellular region containing four cysteine-​rich domains, a transmembrane domain, and an intracellular death do- main. Binding of TNF results in trimerization of the receptor and activation of NF-​κB, with downstream induction of inflammation or apoptosis. Under normal circumstances, TNF signalling is ter- minated by cleavage of the extracellular domain; this results in the release of soluble TNFR1 into the plasma, competitively inhibiting the binding of circulating TNF to cell surface receptors. The mechanisms by which heterozygous TRFRSF1A mutations cause TRAPS are still unclear, and may well vary according to the specific mutation. Postulated explanations have included defective TNFR1 shedding from the cell surface, TNF-​induced apoptosis, NF-​ κB activation, aberrant activation of c-​Jun N-​terminal kinase (JNK) and p38 mitogen-​activated protein kinases (MAPK), and mitochon- drial reactive oxygen species inducing proinflammatory cytokine production. Recent work has focused on aberrant trafficking of the variant protein inducing an unfolded protein response and endo- plasmic reticulum stress with production of reactive oxygen species. These intracellular stress responses appear to be mediated by inositol-​ requiring enzyme 1α (IRE1α), protein kinase-​like endoplasmic re- ticulum kinase (PERK), and activating transcription factor 6 (ATF6). Epidemiology TRAPS was first described in 1982, somewhat tongue-​in-​cheek, as ‘familial Hibernian fever’, reflecting a preponderance of pa- tients from Ireland and Scotland in early reports. It is now clear that TRAPS occurs in diverse populations, including white, Jewish, Arab, and Central American populations, but the disease is rare with an estimated prevalence of 1 to 2 per million. Males and females are affected equally and the median age at presentation is 4 years. Most mutations are associated with high penetrance, but one variant that can be associated with TRAPS (R92Q) is present in approximately 2% of healthy chromosomes, and is thus variously regarded as a low-​ penetrance mutation or polymorphism. Clinical features Attacks of TRAPS are far less distinct than in FMF, sometimes lasting many weeks, and almost one-​third of patients have fairly continuous symptoms. Approximately one-​third of patients with pathogenic mutations report no family history. This is the case in more than 80% of patients carrying R92Q which tends to be associated with milder disease and older age at presentation. Symptoms are rather Fig. 12.12.2.1  Typical erythematous erysipelas-​like rash in a man with familial Mediterranean fever.

12.12.2  Hereditary periodic fever syndromes 2213 variable: more than 95% of patients experience fever and 80% com- plain of arthralgia or myalgia; abdominal pain occurs in 75%, and a rash (often overlying areas of myalgia) occurs in 50%. Other fea- tures include pleuritic pain, lymphadenopathy, conjunctivitis, and periorbital oedema. There are also reports of central nervous system manifestations resembling multiple sclerosis and TNFRSF1A R92Q confers a weak but significant genetic predisposition to multiple sclerosis. Clinical investigation Symptoms are almost universally accompanied by a very marked acute phase response. Genetic testing is pivotal in establishing the diagnosis. Interpretation of the significance of R92Q remains diffi- cult and depends heavily on the clinical picture. Treatment Despite high hopes for anti-​TNF biological agents, their effect in the treatment of TRAPS has proved disappointing in many patients. Acute attacks can be suppressed with corticosteroids, but pro- longed treatment may be required at potentially harmful doses. IL-​1 blockade is the most effective current option in severe TRAPS and in the majority of cases induces a complete disease response. Mevalonate kinase deficiency Genetics MKD is an autosomal recessive disease caused by mutations in the mevalonate kinase gene (MVK) on the long arm of chromosome 12. To date, 160 MKD-​associated mutations have been described, spanning exons 2 to 11, the most common of which encode MVK variants V377I and I268T. The carriage frequency in the population of the Netherlands, in which MKD is most prevalent, is estimated to be 1 in 65. Pathology The MVK mutations associated with MKD result in 85 to 95% de- ficiency in mevalonate kinase activity. This enzyme is involved in cholesterol, farnesyl, and isoprenoid biosynthesis. It is not yet known how mevalonate kinase deficiency causes recurrent inflammation, although there is speculation that reduced protein isoprenylation causes defective lymphocyte apoptosis. Other mutations in MVK result in even greater reduction in enzyme activity, and cause the much more severe disease known as mevalonic aciduria. Epidemiology MKD is most prevalent in the Netherlands, where it was first de- scribed as HIDS in 1984. It has subsequently been reported in other populations, including Arabs and South-​East Asians, though is least rare among northern European white populations. The disease usually presents in the first year of life and occurs equally in both sexes. There are about 200 patients with MKD on the Dutch disease registry, and only some dozens recognized in the United Kingdom. Clinical features Symptoms are episodic and are often well circumscribed. Attacks are irregular, usually last 4 to 6 days, and can be provoked by vaccination, minor trauma, surgery, or stress. Fever, cervical lymphadenopathy, and abdominal pain with vomiting and diarrhoea are typical. Other common symptoms include headache, arthralgia, large joint arth- ritis, erythematous macules and papules, and aphthous ulcers. Rare complications include intellectual impairment, epilepsy, and retin- itis pigmentosa. The disease sometimes ameliorates in early adult life and older patients may remain free of symptoms for years. Clinical investigation Diagnosis is supported by the presence of mevalonic acid in the urine during clinical attacks when there is an elevated acute phase response. Serum immunoglobulin (Ig)-​D and IgA concentration are persistently elevated in 80% of patients although, particularly in the very young, this is nondiscriminative with respect to other autoinflammatory diseases. A mutation in both alleles of MVK can be identified in most patients, more than 80% of whom have the most common V337I variant. Treatment Treatment for milder disease remains largely supportive, including nonsteroidal anti-​inflammatory drugs (NSAIDs) and on-​demand corticosteroids. Colchicine and thalidomide have no convin- cing benefit. There is no evidence of benefit from treatment with simvastatin, an inhibitor of 3-​hydroxy-​3-​methylglutaryl coenzyme A (HMG-​CoA) reductase (the enzyme before MKD in the chol- esterol biosynthetic pathway). There are reports of responses to etanercept, tocilizumab, and anti-​IL-​1 agents; the latter are currently the most widely used biologics. In severe early-​onset disease, bone marrow transplantation can be curative. Cryopyrin-​associated periodic syndromes Genetics CAPS comprises a spectrum of disease associated with mutations in the gene NLRP3/​CIAS1 on chromosome 1q44 that encodes the death domain protein variously known as NLRP3 (previously NALP3) and cryopyrin. Dominant inheritance is evident in about 75% of patients with MWS and FCAS, whereas CINCA, the most severe phenotype, is usually due to a de novo mutation. More than 140 single amino acid substitutions have been reported; the majority in exon 3. Somatic mosaicism is increasing recognized in both early-​ and later-​onset disease. The relationship between mutation and clin- ical phenotype can differ between individuals, even within a family, although some mutations are associated with a greater risk of neuro- logical damage. Pathology NLRP3 is expressed in peripheral blood leucocytes and chondro- cytes, and encodes a death domain protein that contains a pyrin do- main, a nucleotide-​binding site domain, and a leucine-​rich repeat motif. Following recognition and binding, via its leucine-​rich repeat, of an intracellular pathogen-​associated molecular pattern, NLRP3 associates with other members of the death domain superfamily to form a multimeric cytosolic assembly, the inflammasome. This re- sults in activation of caspase 1, which processes pro-​IL-​1 to produce the active cytokine; it also up-​regulates NF-​κB expression, resulting

section 12  Metabolic disorders 2214 in increased IL-​1 gene expression. IL-​1 is a major proinflammatory cytokine involved in mediating local and systemic responses to in- fection and tissue injury. The remarkable response to IL-​1 receptor blockade in CAPS confirms that the clinical features are substan- tially mediated by IL-​1. Epidemiology So far, most reported patients have European ancestry, but CAPS occurs worldwide. Onset of disease is in early infancy, and there is no sex bias. Clinical features FCAS was first described in 1940 as recurrent episodes of cold-​ induced fever, arthralgia, red eye (with inflammation at multiple levels), and rash. MWS was described in 1962 as a much more per- sistent urticaria-​like rash, conjunctivitis, arthralgia, and fever, com- plicated by progressive sensorineural deafness and a high risk of AA amyloidosis (Fig. 12.12.2.2). CINCA was described as a sporadic se- vere inflammatory disorder that presents in the neonatal period with involvement of many organs including the skin, skeletal system, and central nervous system. Bony overgrowth and premature ossification may occur, particularly in the skull and knees, and chronic aseptic meningitis can result in severe developmental delay, optic atrophy, and deafness. It is now evident that FCAS, MWS (Fig. 12.12.2.3), and CINCA represent a spectrum of a single disease entity. Clinical investigation There is usually an acute phase response, and often leucocytosis and thrombocytosis that can vary substantially between measurements, and may not at times be present in some patients at the milder end of the disease spectrum (i.e. with FCAS and mild MWS). Audiometric evidence of sensorineural hearing loss should be sought, and charac- teristic bony abnormalities may be radiologically evident in CINCA. Features consistent with chronic meningitis may be evident on lumbar puncture, fundoscopy, and MRI. A mutation in NLRP3 can be readily identified in most patients with FCAS and MWS, though about half of those with CINCA are somatic mosaics. Treatment Treatment with anti-​IL-​1 agents produces rapid and often complete clinical and serological remission of CAPS. Three IL-​1 inhibitors, anakinra, canakinumab and rilonacept, are licensed for the treat- ment of CAPS. There is hope that early treatment may prevent neurological and skeletal abnormalities. Other hereditary periodic fever syndromes Pyogenic arthritis, pyoderma gangrenosum,
and acne (PAPA) syndrome This autosomal dominant disease is caused by mutations in CD2BP1/​PTSTPIP1, the gene encoding CD2 binding protein 1. The underlying pathogenesis remains poorly understood, although there is evidence that CD2 binding protein 1 interacts with the pyrin pathway. It is characterized by early-​onset recurrent sterile arthritis typically occurring after minor trauma, severe cystic acne, and pyo- derma gangrenosum. Penetrance seems variable and asymptomatic gene carriage is recognized. There is no proven treatment, responses to steroids are usually partial, and early reports suggest variable re- sponses to anti-​TNF or anti-​IL-​1 agents. In general, skin disease is more treatment refractory than the arthritis. Blau’s syndrome This was first described in 1985 as an autosomal dominant syn- drome of sarcoid-​like granulomatous infiltration causing a triad of arthritis, dermatitis, and uveitis. Sporadic disease in the absence of a family history is sometimes called early-​onset sarcoidosis. The joint involvement is usually synovitis or tenosynovitis, sometimes causing camptodactyly (fixed flexion deformity of the proximal interphalangeal joint, usually most marked in the little finger). The rash is characteristically tan coloured and can be ichthyotic. Approximately a third of patients develop visual impairment from panuveitis and its complications. Other recognized features in- clude fever, erythema nodosum, granulomatous hepatitis, and large vessel vasculitis. Fig. 12.12.2.2  Typical rash of mevalonate kinase deficiency. These lesions appear during febrile attacks. Fig. 12.12.2.3  Urticarial rash of Muckle–​Wells syndrome. These lesions, accompanied by conjunctivitis, arthralgia, and fever, appeared daily in the early evenings.

12.12.2  Hereditary periodic fever syndromes 2215 Presentation is usually before the age of 5 years and is caused by missense mutations in NOD2/​CARD15, another member of the death domain superfamily. NOD2 mutations have also been im- plicated in familial Crohn’s disease but the gain-​of-​function mu- tations in Blau’s syndrome occur in the NACTH domain whereas in Crohn’s disease, there are loss-​of-​function mutations in the LRR region. The NOD2 mutations associated with Blau’s syn- drome have not been found in adults with sarcoidosis. Most cases are steroid responsive but steroid-​sparing agents and increas- ingly anti-​TNF therapy may be required in more severe disease (Fig. 12.12.2.4). NALP12-​associated periodic fever syndrome This autosomal dominant syndrome was described in 2008 in fam- ilies from the Caribbean. Patients presented in infancy or the neo- natal period with a syndrome with some features of cold induction, fever, arthralgia and myalgia, urticarial rash, and sensorineural deaf- ness. The nonsense and splice site mutations identified in NALP12 appear to reduce its inhibitory effect on NF-​κB signalling. There is no established treatment. Deficiency of the IL-​1 receptor antagonist (DIRA) This autosomal recessive disease was first described in 2009. It is due to mutations in the IL1RN gene that result in total deficiency of IL-​1 receptor antagonist. The disease has been reported in only a handful of families of various ethnicities including Northern Europe and Central America. The disease presents in the immediate neonatal period with a pustular rash, joint swelling, multifocal osteitis of the ribs and long bones, heterotopic ossification, and periarticular soft tissue swelling. Treatment is replacement of the missing protein with recombinant IL-​1Ra (anakinra), and good responses have been re- ported in all patients treated so far. Majeed’s syndrome This autosomal recessive syndrome characterized by chronic recurrent multifocal osteomyelitis (CRMO), congenital dyserythropoietic anaemia, and inflammatory dermatosis was first described in 1989 in consanguineous Arab kindred. Onset is usu- ally in the neonatal period and attacks consist of several days of fever, severe pain, and the appearance of periarticular soft tissue swelling. Long-​term complications of growth retardation and flexion contractures are well recognized. The disease was found to be due to mutations in LPIN2, a widely expressed gene thought to play a role in lipid metabolism, in 2005. NSAIDs and corticoster- oids can provide symptomatic relief. Recent case reports suggest IL-​1 blockade may be effective although the long-​term effect on dyserythropoiesis is not yet known Deficiency of interleukin-​36 receptor
antagonist (DITRA) This rare disease is a recessively inherited autoinflammatory disease due to mutations in IL36RN resulting in unregulated signalling of IL-​36α, -​β, and -​γ. It is characterized by recurrent episodes of a gen- eralized sterile pustular psoriasis accompanied with neutrophilia, a marked acute phase response, and fever. Age at onset varies from childhood to the sixth decade. Episodes may be precipitated by stress, pregnancy, or initiation or withdrawal of drugs and can be life-​threatening. There is no established treatment but case reports suggest anakinra has been beneficial. Chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature
(CANDLE syndrome) or proteasome-​associated autoinflammatory syndrome (PRAAS), or joint contractures, muscular atrophy, microcytic anaemia, and panniculitis-​induced lipodystrophy (JMP) This recessive disease is characterized by a neonatal onset of inter- mittent fevers, swollen violaceous eyelids, thicken lips, erythema- tous rash with a mixed myeloid, neutrophilic, and histiocytic infiltrate on skin biopsy, arthralgia/​arthritis, and progressive lipodystrophy with later-​onset joint contractures and raised acute phase response. Patients have been found to carry PSM8 mutations. This encodes an inducible subunit of the immune proteosome and deficiency is postulated to result in failure of proteolysis resulting in accumulation of damaged protein and intracellular stress. This appears to result in dysregulation of the interferon signalling pathway and early reports suggest treatment with Janus kinase in- hibitors may be useful. IL-​10 receptor deficiency This is also known as early-​onset inflammatory bowel disease and presents with enterocolitis with perianal fistulae, recurrent fever, arthritis, and folliculitis in the first 3 months of life. It is a reces- sive disease caused by mutations in IL10RA, IL10RB, or IL10 re- sulting in decreased IL-​10 signalling and macrophage activation. Conventional treatment of inflammatory bowel disease is ineffective and bone marrow transplantation has been curative. Autoinflammation and PLCγ2-​associated antibody deficiency and immune dysregulation (APLAID) Only a single kindred is known with this autosomal dominant disease caused by gain-​of-​function mutations in PLCG2 which encodes phospholipase Cγ2 (PLCγ2). This enzyme is involved in several immunological pathways and in APLAID it may acti- vate the NLRP3 inflammasome. Patients have low IgA and IgM and immunodeficiency with recurrent sinopulmonary infections. Autoinflammatory features include blistering rash, interstitial pneumonitis, ocular inflammation, and arthralgia. Deletions in Fig. 12.12.2.4  Camptodactyly due to Blau’s syndrome.

section 12  Metabolic disorders 2216 PLCG2 are known to cause a different syndrome, PLCγ2-​associated antibody deficiency and immune dysregulation (PLAID) with an autoimmune phenotype. Haem-​oxidized IRP2 ubiquitin ligase 1 deficiency
(HOIL-​1 deficiency) This fatal autosomal recessive disease has been reported in two fam- ilies and is due to loss-​of-​function mutations in HOIL1 also known as RBCK. This encodes a component of the linear ubiquitin chain assembly complex, which adds polyubiquitin chains to substrate proteins and plays a role in NF-​κB induction. The phenotype is a combination of severe immunodeficiency and autoimmunity with recurrent fever, systemic inflammation, hepatosplenomegaly, and lymphadenopathy. The patients also developed amylopectinosis resulting in skeletal and cardiac myopathy. All four affected indi- viduals died in childhood, cortisteroids were reported to be symp- tomatically helpful. Deficiency of ADA2 (DADA2) This was first described in 2014 as a monogenetic form of polyarteritis nodosa (PAN). It is a recessive disease caused by loss-​of-​function mutations in CECR1 encoding adenosine deaminase 2 (ADA2). ADA2 is thought to be a growth factor for endothelial cells as well as leucocytes and deficiency appears to induce an inflammatory vasculopathy. Clinical manifestations include childhood systemic and local PAN, recurrent fever, mild immunodeficiency, livedo ra- cemose, and early-​onset stroke. Treatment with anti-​TNF agents has been reported to be effective and bone marrow transplantation has been reported to induce clinical benefit and improve ADA2 levels. A small study of fresh frozen plasma as a source of replacement ADA2 in acute attacks was not supportive. STING-​associated vasculopathy with onset in infancy (SAVI) This autosomal dominant disease due to gain-​of-​function muta- tions in TMEM173 (encoding the stimulator of interferon genes (STING)) was first described in 2014 and results in increased inter- feron 1 signalling. Fewer than 20 patients are known and the dis- ease is characterized by very early-​onset facial and digital rash with scarring, cartilage destruction, and capillaritis on biopsy. Other fea- tures include failure to thrive, lymphadenopathy, fever, and intersti- tial lung disease. There is no proven treatment but treatment with Janus kinase inhibitors seems rational based on the apparent role of upregulated interferon signalling. Autoinflammatory diseases of unknown aetiology Periodic fever, aphthous stomatitis, pharyngitis, and adenitis (PFAPA) First described in 1987, the diagnosis is suggested by a recurrent fever of early onset and one of the following associated symptoms: oral aphthous ulcers, cervical lymphadenopathy, or pharyngitis, in the absence of recurrent upper respiratory tract infections or cyclic neutropenia. Characteristically the children are entirely well be- tween attacks. In a large case series, median age at presentation was 2.5 years and 83% of children presented before their fifth birthday with a slight male preponderance of 62%. The ‘acronym symptoms’ of aphthous oral ulcers, pharyngitis, and cervical lymphadenopathy are frequently not all present during a single attack. During attacks, the acute phase response is often strikingly elevated. In general the prognosis is good and most children will outgrow their symptoms within a decade. For many clinicians, the strongest diagnostic pointers are the extreme regularity of attacks (although attacks may be missed particularly in the summer) and the response to a small dose of cor- ticosteroids. Before diagnosis, recurrent infections and cyclic neu- tropenia must be excluded. The first-​line treatment of PFAPA is a single dose of corticosteroid given at the start of the attack. Padeh et al. suggested that the dra- matic response to a single oral dose of corticosteroids is sufficiently unique to PFAPA syndrome that it could be used as a diagnostic criterion. The H2 receptor antagonist cimetidine and colchicine have been tried in PFAPA with variable reports of success. Tonsillectomy is the only treatment for which there is supportive evidence from clinical studies. Tonsillectomy can be curative in PFAPA; in gen- eral, more than 50% of children appear to have excellent long-​term results. However, these data may be biased since many centres se- lectively refer children with persistently enlarged tonsils for surgery, and it is possible that responses may occur preferentially in this subgroup. Schnitzler’s syndrome This was first reported in 1974 and is characterized by a chronic urticarial-​like rash, a monoclonal IgM (IgM kappa in 85%) gammopathy, and systemic inflammation usually presenting as fever. The median age at onset is 51 years and there is a slight male preponderance. The monoclonal protein appears central to the pathogenesis although the mechanism remains unclear. About a fifth of patients eventually progress to overt plasma cell malignancy. Chemotherapy has been used in the past but does not appear to relieve the syndrome and should only be used for conventional haematological indications. The treatment of choice for Schnitzler’s syndrome is IL-​1 blockade which is highly effective. Differential diagnosis of the hereditary periodic fever syndromes These diseases have a broad differential diagnosis, particularly at first presentation, which is influenced by age and encompasses a vast spectrum of infectious, immune, and neoplastic disorders (Table 12.12.2.2). Conversely, symptoms such as fever, arthralgia, and rashes in a patient known to have a hereditary periodic fever syndrome may also result from an alternative intercurrent disorder. Prognosis and complications Although CINCA/​NOMID can be sufficiently severe to cause death within the first few decades, life expectancy among most patients

12.12.2  Hereditary periodic fever syndromes 2217 with hereditary recurrent fever syndromes is relatively good, and ex- cellent in those for whom there is now effective therapy. The most serious and life-​threatening complication of these diseases generally is AA amyloidosis. AA amyloidosis This usually presents with proteinuric kidney dysfunction. AA amyloid fibrils are derived from the circulating acute phase protein serum amyloid A protein (SAA), which is synthesized by hepato- cytes under the transcriptional regulation of IL-​1, IL-​6, and TNFα. The terminology is either serum amyloid A protein or SAA. The circulating concentration of SAA in health is less than approxi- mately 3 mg/​litre, but this can rise by up to 1000-​fold in the pres- ence of inflammation. In chronic inflammatory diseases generally, AA amyloidosis occurs in up to 5% of patients after a median dur- ation of about 20 years, but is much more frequent among patients with untreated inherited periodic fever syndromes. This may reflect their lifelong nature and their capacity to stimulate remarkably high plasma concentrations of SAA, even when they seem clinically qui- escent. Before the widespread introduction of colchicine prophy- laxis, up to 60% of patients with FMF died of amyloidosis, and even recently it was reported in 13% of a large Turkish series. The inci- dence of AA amyloidosis in TRAPS and MWS is approximately 25%, but is less than 5% in MDK, perhaps because the disease often ameli- orates in early adulthood. The natural course of AA amyloidosis is renal failure and early death, but this can be prevented by treatment of the underlying inflammatory disorder that substantially suppresses the produc- tion of serum amyloid A.  Indeed, treatment such as colchicine in FMF and anakinra in CAPS can halt further deposition of AA amyloid, facilitate gradual regression of existing deposits, and lead to preservation or even improvement in renal function. Regular long-​term measurement of serum amyloid A is vital in patients with AA amyloidosis. Likely future developments The recent elucidation of the pathogenesis of these diseases has led to major advances in their treatment, most notably in CAPS. It is likely that continued studies will shed further light on aspects of the innate immune system and inflammation generally, and on strategies for the treatment of TRAPS and MKD in particular. The clinical significance of low-​penetrance mutations and polymorphisms in the genes asso- ciated with inherited periodic fever syndromes will be sought. CAPS provides a powerful model of IL-​1-​driven disease, and a uniquely in- formative test bed for the early-​phase development of novel IL-​1 in- hibitors that may have applications in many common inflammatory disorders, ranging from gout and rheumatoid arthritis to sepsis. FURTHER READING Ben-​Chetrit E, Levy M (2003). Reproductive system in familial Mediterranean fever: an overview. Ann Rheum Dis, 62, 916–​19. Canna SW, Goldbach-​Mansky R (2015). New monogenic auto­ inflammatory diseases—​a clinical overview. Semin Immunol, 37, 387–​94. Cowen EW, Goldbach-​Mansky R (2012). DIRA, DITRA, and new in- sights into pathways of skin inflammation: what’s in a name? Arch Dermatol, 148, 381–​4. Goldbach-​Mansky R, et  al. (2006). Neonatal-​onset multisystem in- flammatory disease responsive to interleukin-​1beta inhibition.
N Engl J Med, 355, 581–​92. Holzinger D, et al. (2015). From bench to bedside and back again: trans- lational research in autoinflammation. Nat Rev Rheumatol, 10, 573–​85. Lachmann HJ, et al. (2007). Natural history and outcome in systemic AA amyloidosis. N Engl J Med, 356, 2361–​71. Lachmann HJ, et al. (2014). The phenotype of TNF receptor-​associated autoinflammatory syndrome (TRAPS) at presentation: a series of 158 cases from the Eurofever/​EUROTRAPS international registry. Arthritis Rheum, 73, 2160–​7. Table 12.12.2.2  Differential diagnosis of inherited periodic fever syndromes Abdominal pain and fever Thoracic pain and fever Arthritis and fever Fever, rash, and myalgia Nonhereditary periodic fever syndromes Acute surgical abdomen Myocardial infarction Septic arthritis Viral illness PFAPA (periodic fever, aphthous stomatitis, pharyngitis, and adenopathy) Acute cholecystitis Pneumonia/​pleurisy Juvenile idiopathic arthritis Systemic lupus erythematosus Schnitzler’s syndrome Pyelonephritis Pericarditis Rheumatic fever Cellulitis/​erysipelas Pelvic inflammatory disease Pulmonary embolism Lyme disease Behçet’s disease Endometriosis Palindromic arthritis Cyclic neutropenia Mesenteric adenitis Crystalline arthritis (gout) and calcium pyrophosphate dihydrate crystal deposition disease Malignancy Systemic vasculitis Adult-​onset Stills disease and systemic-​onset juvenile idiopathic arthritis Dermatomyositis Hereditary or acquired angio-​oedema (not associated with fever) Adult-​onset Stills disease and systemic-​onset juvenile idiopathic arthritis

12.12.3 Amyloidosis 2218

12.12.3 Amyloidosis 2218

section 12  Metabolic disorders 2218 Levy R, et  al. (2015). Phenotypic and genotypic characteristics of cryopyrin-​associated periodic syndrome:  a series of 136 patients from the Eurofever Registry. Ann Rheum Dis, 7411, 2043–​9. Liu Y, et al. (2012). Mutations in proteasome subunit β type 8 cause chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature with evidence of genetic and phenotypic het- erogeneity. Arthritis Rheum, 64, 895–​907. Niel E, Scherrmann JM (2006). Colchicine today. Joint Bone Spine, 73, 672–​8. Schneiders MS, et al. (2005). Manipulation of isoprenoid biosynthesis as a possible therapeutic option in mevalonate kinase deficiency. Arthritis Rheum, 54, 2306–​13. Sfriso P, et  al. (2012). Blau syndrome, clinical and genetic aspects. Autoimmun Rev, 12, 44–​51. Simon A, et al. (2013). Schnitzler’s syndrome: diagnosis, treatment, and follow-​up. Allergy, 68, 562–​8. Smith EJ, et al. (2010). Clinical, molecular, and genetic characteristics of PAPA syndrome: a review. Curr Genomics, 11, 519–​27. Sönmez HE, et al. (2016). Familial Mediterranean fever: current per- spectives. J Inflamm Res, 17, 13–​20. Twig G, et al. (2014). Mortality risk factors associated with familial Mediterranean fever among a cohort of 1.25 million adolescents. Ann Rheum Dis, 73, 704–​9. Zhao Y, Shao F (2016). Diverse mechanisms for inflammasome sensing of cytosolic bacteria and bacterial virulence. Curr Opin Microbiol, 29, 37–​42. 12.12.3  Amyloidosis Mark B. Pepys and Philip N. Hawkins ESSENTIALS Amyloidosis is the clinical condition caused by extracellular de- position of amyloid in the tissues. Amyloid deposits are composed of amyloid fibrils, abnormal insoluble protein fibres formed by misfolding of their normally soluble precursors. About 30 different proteins can form clinically or pathologically significant amyloid fi- brils in vivo as a result of either acquired or hereditary abnormalities. Small, focal, clinically silent amyloid deposits in the brain, heart, sem- inal vesicles, and joints are a universal accompaniment of ageing. Clinically important amyloid deposits usually accumulate progres- sively, disrupting the structure and function of affected tissues and lead inexorably to organ failure and death. There is no licensed treat- ment which can specifically clear amyloid deposits, but intervention which reduces the availability of the amyloid fibril precursor proteins can arrest amyloid accumulation and may lead to amyloid regression with clinical benefit. Pathology—​amyloid fibrils of all types are similar: straight, rigid, and nonbranching; of indeterminate length and 10–​15 nm in diameter; and with their subunit proteins arranged in a stack of twisted anti- parallel β-​pleated sheets. The fibrils bind Congo red dye producing pathognomonic green birefringence when viewed in high-​intensity cross-​polarized light, and the protein type can be identified by immunostaining or proteomic analysis. Amyloid deposits always contain a nonfibrillar plasma glycoprotein, serum amyloid P com- ponent, the universal presence of which is the basis for use of radio- isotope-​labelled serum amyloid P component as a diagnostic tracer. Clinicopathological correlation—​amyloid may be deposited in any tissue of the body, including blood vessels walls and connective tissue matrix; clinical manifestations are correspondingly diverse. Although there are some typical clinical presentations related to fibril type, there are many forms of amyloidosis in which there is little or no concordance between the fibril protein, or the geno- type of its precursor, and the clinical phenotype. Identification of the amyloid fibril protein is always essential for appropriate clinical management. Specific types of amyloidosis Monoclonal immunoglobulin light chain (AL) amyloidosis—the most prevalent type of systemic amyloidosis currently diagnosed. The fi- brils consists of all or part of the variable (VL) domain of monoclonal immunoglobulin light chains. May complicate any B-cell dyscrasia but most cases are associated with otherwise ‘benign’ ‘monoclonal gammopathy of undetermined significance’ (MGUS). Highly variable idiotypic disease but characteristic presentations include involve- ment of the heart (restrictive cardiomyopathy), kidneys (protein- uria, renal failure), gut (motility disorders, malabsorption), tongue (macroglossia), and nerves (painful sensory polyneuropathy; auto- nomic neuropathy). Treatment is with cytotoxic chemotherapy, similar to that used in myeloma, aimed at suppression of the causa- tive B-cell clone. Transthyretin (ATTR) amyloidosis—caused by wild type trans­thyretin (TTR) forming amyloid predominantly in the heart, mostly in eld- erly men. Previously diagnosed extremely rarely but shown by new imaging methods to be much more common with prevalence lately approaching that of AL amyloidosis, and potentially exceeding it in future. Hereditary systemic amyloidoses—include hereditary ATTR amyl- oidosis caused by mutations in the TTR gene. Presents as progres- sive peripheral and autonomic neuropathy and varying degrees of visceral involvement (formerly known as familial amyloidotic poly- neuropathy), and sometimes with predominant cardiomyopathy. Reactive systemic (AA) amyloidosis—fibrils composed of AA pro- tein derived from the acute phase protein, serum amyloid A protein (SAA). It can be a complication of any chronic inflammatory disorder (e.g. rheumatoid arthritis, Crohn’s disease) or chronic infections in which SAA concentrations are persistently increased. Has become very rare in recent years due to improved management of chronic inflammation and infection. Usually presents with proteinuria and/or organomegaly (e.g. hepatosplenomegaly); nephrotic syndrome may develop before progression to end-stage renal failure. Introduction Amyloidosis and amyloid deposits Amyloid deposits can be systemic, that is present anywhere in the body. The predominant locations and most damaging clinical

12.12.3  Amyloidosis 2219 effects can vary widely in different types of amyloidosis and in dif- ferent individuals. In systemic amyloidosis the deposits can be in the extracellular space of any tissue and any organ except the brain, because the amyloid fibrils are derived from circulating, normally soluble, globular plasma proteins. There are also localised forms of amyloidosis in which the deposits are confined to a single tissue or organ system, such as the cerebral blood vessels walls in cerebral amyloid angiopathy or localised amyloid of the respiratory system or urogenital tract. Crucially, in all forms of amyloidosis, tissue damage and thus disease are directly caused by the amyloid de- posits that disrupt normal tissue architecture and thus function. In contrast, the histopathology in some other diseases includes local amyloid deposits, for example, the Aβ amyloid plaques in the brain in Alzheimer’s disease and islet amyloid in the islets of Langerhans in the pancreas of patients with type 2 diabetes, but there is no direct evidence that these deposits cause disease. They should therefore not be included as types of amyloidosis. Similarly, microscopic amyloid deposits are frequently present in the elderly, in the seminal vesicles and the walls of large arteries, without evidence that they necessarily cause disease. Systemic amyloidosis causes about 1 in 1500 of all deaths in developed countries but this figure will likely rise as the true prevalence of wild type ATTR amyloidosis is recognized and is a serious and important disease because it is often difficult to diag- nose, it is usually fatal, and its management is complex and costly. Most systemic amyloidosis is a complication of other underlying primary conditions, which include monoclonal gammopathies of all types, chronic inflammatory disorders, and dialysis for end-​ stage renal failure. Hereditary amyloidosis is very rare, except in a few geographic foci, but its diversity is remarkable. It is im- portant because of its poor prognosis, the complexity of clinical management, the difficult genetic issues involved, and its con- siderable value as a model for understanding the pathogenesis of amyloid deposition. Clinicopathological correlation Although there are some correlations between fibril protein type and clinical manifestations, there are also many forms of acquired and hereditary amyloidosis in which there is little or no concord- ance between the fibril protein, or the genotype of its precursor, and the clinical phenotype (Tables 12.12.3.1 and 12.12.3.2). There are evidently genetic and/​or environmental factors, distinct from the amyloid fibril protein itself, that determine whether, when, and where clinically significant amyloid deposits form. The nature of these important determinants of amyloidogenesis is obscure. Furthermore, the mechanisms by which amyloid deposition causes disease are poorly understood. While a heavy amyloid load is in- variably a bad sign, there may be a poor correlation between the local amount of amyloid and the level of organ dysfunction. Active deposition of new amyloid is often associated with accelerated de- terioration compared with stable, long-​standing deposits. Nascent or newly formed amyloid fibrils generated in vitro may also be cyto- toxic to cultured cells, whereas aged or ex vivo fibrils are generally inert, but it is not known if or how this relates to effects in vivo. In most forms of systemic amyloidosis there is overwhelming evi- dence that tissue damage and resultant disease are caused by the physical presence and accumulation of amyloid deposits, and not by cytotoxicity of the amyloidogenic proteins or their prefibrillar aggregates. Clinical types of amyloidosis Reactive systemic (AA) amyloidosis Associated conditions AA amyloidosis occurs in association with chronic inflammatory disorders, chronic local or systemic microbial infections, and, oc- casionally, malignant neoplasms. In western Europe and the United Table 12.12.3.1  Acquired amyloidosis syndromes Clinical syndrome Fibril protein Systemic AL amyloidosis, associated with immunocyte dyscrasia, myeloma, monoclonal gammopathy, occult dyscrasia AL derived from monoclonal immunoglobulin light chains Local nodular AL amyloidosis (skin, respiratory tract, urogenital tract, etc.) associated with focal immunocyte dyscrasia AL derived from monoclonal immunoglobulin light chains Reactive systemic AA amyloidosis, associated with chronic active diseases AA derived from SAA Non-hereditary wild-type transthyretin amyloidosis Transthyretin derived from plasma transthyretin Sporadic cerebral amyloid angiopathy Aβ derived from APP Haemodialysis-​associated amyloidosis; localized to osteoarticular tissues or systemic β2-​microglobulin derived from high plasma levels Primary localized cutaneous amyloid (macular, papular) ? Keratin-​derived Primary localized cutaneous amyloid (macular, papular) ? Keratin-​derived Systemic ALECT2 amyloidosis causing renal dysfunction Leucocyte chemotactic factor 2 Ocular amyloid (cornea, conjunctiva) Not known Orbital amyloid AL or AH derived from monoclonal Ig AA, amyloid A; AH, monoclonal immunoglobulin heavy chain fragment; AL, monoclonal immunoglobulin light chain; ALECT2, leucocyte chemotactic factor 2 amyloid; APP, amyloid precursor protein; Ig, immunoglobulin; SAA, serum amyloid A protein.

section 12  Metabolic disorders 2220 States of America, the most frequent predisposing conditions are idiopathic rheumatic diseases, notably including rheumatoid arth- ritis and juvenile idiopathic arthritis (Box 12.12.3.1). AA amyloid- osis has become increasingly rare, reflecting improved treatment of chronic inflammatory disorders, and for reasons that are not clear, the incidence is lower in the United States of America than in Europe. Amyloidosis is exceptionally rare in systemic lupus erythematosus, related connective tissue diseases, and in ulcerative colitis in which there is a blunted acute phase response of serum amyloid A protein, the precursor of AA amyloid fibrils. Tuberculosis and leprosy are important causes of AA amyloidosis where these infections remain endemic. Chronic osteomyelitis, bronchiectasis, chronically infected burns, and decubitus ulcers, as well as the chronic pyelonephritis of paraplegic patients, are other well-​recognized associations (Box 12.12.3.1). Hodgkin’s disease and renal carcinoma, which often cause fever, other systemic symptoms, and a major acute phase response, are the malignancies most commonly associated with systemic AA amyloidosis. Intriguingly, about 10% of patients with AA amyloidosis do not have a clinically obvious chronic inflammatory disease, and may erroneously be assumed to have AL amyloidosis. The commonest identifiable pathologies in such cases in our own experience have been previously undiagnosed inherited periodic fever syndromes and cyto- kine-​secreting Castleman’s disease tumours of the solitary plasma cell type, located in either the mediastinum or the gut mesentery. Clinical features AA amyloid involves the viscera, but may be widely distributed without causing clinical symptoms. More than 90% of patients present with nonselective proteinuria resulting from glomerular deposition, and nephrotic syndrome may develop before progression to end-​ stage renal failure. Haematuria, isolated tubular defects, nephrogenic diabetes insipidus, and diffuse renal calcification rarely occur. Kidney size is usually normal, but may be enlarged, or, in advanced cases, reduced. End-​stage chronic renal failure is the cause of death in 40 to 60% of cases, but acute kidney injury may be precipitated by hypo- tension and/​or salt and water depletion following surgery, excessive use of diuretics, or intercurrent infection. The second most common presentation is with organ enlargement, such as hepatosplenomegaly or thyroid goitre, with or without overt renal abnormality, but in any case, amyloid deposits are almost always widespread at the time of presentation. Involvement of the heart and gastrointestinal tract is frequent, although the former rarely causes functional impairment. Table 12.12.3.2  Hereditary amyloidosis syndromes Clinical syndrome Fibril protein Predominant peripheral nerve involvement, hereditary transthyretin amyloidosis associated polyneuropathy (familial amyloid polyneuropathy). Autosomal dominant Transthyretin variants (commonly Met30, >120 others described) Apolipoprotein A-​I N-​terminal fragment of Arg26 variant Predominant cranial nerve involvement with lattice corneal dystrophy. Autosomal dominant Gelsolin, fragment of variants Asn187 or Tyr187 Oculoleptomeningeal amyloidosis. Autosomal dominant Transthyretin variants Non-​neuropathic, prominent visceral involvement. Autosomal dominant Apolipoprotein A-​I N-​terminal fragment of variants Lysozyme variants Fibrinogen α-​chain variants β2-​microglobulin variant Asn76 Predominant cardiac involvement, no clinical neuropathy, hereditary transthyretin amyloidosis associated cardiomyopathy. Autosomal dominant Transthyretin variants Hereditary cerebral haemorrhage with amyloidosis (cerebral amyloid angiopathy). Autosomal dominant:   Icelandic type (major asymptomatic systemic amyloid also present) Cystatin C, fragment of variant Glu68   Dutch type Aβ derived from APP variant Gln693 Cutaneous deposits (bullous, papular, pustulodermal) Not known Amino acids: Arg, arginine; Asn, asparagine; Gln, glutamine; Glu, glutamic acid; Met, methionine; Tyr, tyrosine. Aβ, amyloid β; APP, amyloid precursor protein; SAA, serum amyloid A protein. Box 12.12.3.1  Conditions associated with reactive systemic amyloid A amyloidosis Chronic inflammatory disorders • Rheumatoid arthritis • Juvenile idiopathic arthritis • Ankylosing spondylitis • Psoriasis and psoriatic arthropathy • Reiter’s syndrome • Adult Still’s disease • Behçet’s syndrome • Crohn’s disease • Hereditary periodic fever syndromes Chronic microbial infections • Leprosy • Tuberculosis • Bronchiectasis • Decubitus ulcers • Chronic pyelonephritis in paraplegics • Osteomyelitis • Whipple’s disease Neoplasms • Castleman’s disease • Hodgkin’s disease • Renal carcinoma • Carcinomas of gut, lung, and urogenital tract • Basal cell carcinoma • Hairy cell leukaemia

12.12.3  Amyloidosis 2221 AA amyloidosis may become clinically evident early in the course of associated disease, but the incidence increases with the duration of the primary condition. The mean duration of chronic rheumatic diseases, such as rheumatoid arthritis, ankylosing spondylitis, or juvenile rheumatoid arthritis, before amyloid is diagnosed is 12 to 14 years, although they can present much sooner. For most patients, the prognosis is closely related to the degree of renal involvement and the effectiveness of treatment for the underlying inflammatory con- dition. In the presence of persistent uncontrolled inflammation, 50% of patients with AA amyloidosis die within 5 years of diagnosis; how- ever, if the acute phase response can be consistently suppressed, pro- teinuria can cease, renal function can be retained, and the prognosis is much better. The availability of chronic haemodialysis and trans- plantation prevents early death from uraemia per se, but amyloid de- position in extrarenal tissues may be responsible for a less favourable prognosis than for some other causes of end-​stage renal failure. Amyloidosis associated with immunocyte
dyscrasia: monoclonal immunoglobulin
light chain (AL) amyloidosis Associated conditions AL amyloidosis may complicate almost any dyscrasia of cells of the B-​lymphocyte lineage, including multiple myeloma, malignant lymphomas, and macroglobulinaemia, but most cases develop on a background of monoclonal gammopathy of undetermined signifi- cance. Amyloidosis occurs in up to 10% of cases of myeloma, in a lower proportion of other malignant B-​cell disorders, and about 2% of patients with monoclonal gammopathy of undetermined signifi- cance, which are, of course, much more common than myeloma. A monoclonal paraprotein or abnormal serum free light chain concentration can be detected in the serum of most patients with AL amyloidosis, confirming a monoclonal gammopathy. Subnormal levels of some or all serum immunoglobulins, or increased numbers of marrow plasma cells may provide less direct clues to the under- lying aetiology. Until recently it has been the practice to diagnose ap- parently primary cases of amyloidosis, with no previous predisposing inflammatory condition or family history of amyloidosis, as AL type by exclusion. However, it has now been recognized that autosomal dominant, hereditary non-​neuropathic amyloidosis, particularly that caused by variant fibrinogen α-​chain, may be poorly penetrant and of late onset, so there may be no family history. The coincident oc- currence of a monoclonal gammopathy, which occurs in more than 10% of the healthy older population, may then be gravely misleading, and it is essential to exclude by genotyping all known amyloidogenic mutations, and to seek positive immunohistochemical or proteomic identification of the amyloid fibril protein in all cases. Clinical features AL amyloidosis occurs equally in men and women, usually over the age of 50 years, but occasionally in young adults. It has a lifetime incidence (and is the cause of death) of between 0.5 and 1 in 1000 individuals in the United Kingdom. The clinical manifestations are protean, as virtually any tissue other than the brain may be directly involved. Uraemia, heart failure, or other effects of amyloidosis usu- ally cause death within 2 years of diagnosis, unless the underlying B-​cell clone is effectively suppressed. The heart is affected in 90% of patients with AL amyloidosis. In 30% of these, restrictive cardiomyopathy is the presenting feature and in up to 50% of these patients it is fatal. Other cardiac presenta- tions include arrhythmias and angina. Measurement of circulating brain natriuretic peptide (BNP) or its precursor N-​terminal-​ proBNP provide a sensitive index of cardiac dysfunction in cardiac AL amyloidosis, and often show rapid improvement when produc- tion of the aberrant clonal light chain is substantially suppressed by cytotoxic chemotherapy. This suggests that amyloidogenic light chains, perhaps in an aggregated prefibrillary form, may them- selves have intrinsic cardiotoxicity, in addition to their deposition as amyloid fibrils. Renal AL amyloidosis has the same manifest- ations as renal AA amyloidosis, but the prognosis is worse. Gut in- volvement may cause disturbances of motility (often secondary to autonomic neuropathy), malabsorption, perforation, haemorrhage, or obstruction. Macroglossia occurs rarely, but is almost pathogno- monic. Hyposplenism sometimes occurs in both AA and AL amyl- oidosis. Painful sensory polyneuropathy, with early loss of pain and temperature sensation, followed later by motor deficits, is seen in 10 to 20% of cases, and carpal tunnel syndrome in 20%. Autonomic neuropathy leading to orthostatic hypotension, impotence, and gastrointestinal disturbances may occur alone or together with the peripheral neuropathy, and has a very poor prognosis. Skin involve- ment takes the form of papules, nodules, and plaques, usually on the face and upper trunk, and involvement of dermal blood vessels results in purpura, occurring either spontaneously or after minimal trauma, and is very common. Articular amyloid is rare, but may mimic acute polyarticular rheumatoid arthritis, or it may present as asymmetrical arthritis affecting the hip or shoulder. Infiltration of the glenohumeral joint and surrounding soft tissues occasionally produces the characteristic ‘shoulder pad’ sign. A rare but serious manifestation of AL amyloidosis is an acquired bleeding diathesis that may be associated with deficiency of factor X, and sometimes also factor IX, or with increased fibrinolysis. It does not occur in AA amyloidosis, although in both AL and AA disease there may be serious bleeding in the absence of any identifiable coagulation factor deficiency due to widespread vascular amyloid deposits. Senile amyloidosis Some amyloid is seen in all autopsies on individuals over 80 years of age, but it is not known whether this contributes to the ageing process or whether it is an epiphenomenon that becomes clinically important only when it is extensive. Wild-​type transthyretin (cardiac) amyloidosis Up to 25% of older people have systemic deposits of wild-​type transthyretin amyloid involving the walls of the heart and blood ves- sels, smooth and striated muscle, fat tissue, renal papillae, and al- veolar walls. In contrast to most other forms of systemic amyloidosis (including hereditary transthyretin amyloid caused by point mu- tations in the transthyretin gene), clinical manifestations are usu- ally restricted to the heart and carpal tunnel syndrome. The brain is not involved. The typical clinical picture of progressive restrictive cardiomyopathy is usually fatal within 5 years. The isoleucine 122 variant of transthyretin, which occurs in about 4% of black individ- uals of African ethnicity, is associated with a greatly increased risk of cardiac transthyretin amyloidosis, the phenotype of which is in- distinguishable from the wild-​type form. It is likely that senile car- diac amyloidosis is substantially underdiagnosed and may account for many cases of cardiac failure associated with preserved ejection fraction in the elderly.

section 12  Metabolic disorders 2222 Senile focal amyloidosis Microscopic and clinically silent amyloid deposits of different fibril types, localized to particular tissues, are very commonly found in older people. Deposits of β-​protein (Aβ, see ‘Alzheimer’s disease’) as amyloid in cerebral blood vessels and intracerebral plaques seen in normal older brains may or may not have been the harbinger of Alzheimer’s disease had the patient survived long enough. Amyloid deposits composed of apolipoprotein A-​I are present in most osteoarthritic joints at surgery or autopsy, usually in close association with calcium pyrophosphate deposits, and affect the articular cartilage and joint capsule. However, the significance of this age-​associated articular amyloid, the amount of which is correlated with neither the presence nor clinical severity of osteoarthritis, is not known. The corpora amylacea of the prostate are composed of β2-​microglobulin amyloid fibrils. Amyloid in the seminal vesicles is derived from semenogelin I, an exocrine secretory product of the vesicle cells. Isolated deposits of cardiac atrial amyloid consist of atrial natriuretic peptide. The focal amyloid deposits commonly pre- sent in atheromatous plaques of older subjects are of two types: con- taining fibrils either composed of medin, a fragment of lactadherin, or the N-​terminal fragment of apolipoprotein A-​I. Cerebral amyloid The brain is a very common and important site of amyloid depos- ition (Box 12.12.3.2), although, possibly because of the blood–​brain barrier, there are never any deposits in the cerebral parenchyma it- self in any form of acquired systemic visceral amyloidosis. However, cerebrovascular transthyretin amyloid may occur in familial amyloid polyneuropathy, although the serious clinical entity of oculoleptomeningeal amyloidosis is very rare and associated with certain uncommon transthyretin variants. The major forms of brain amyloid are confined to the brain and cerebral blood vessels, with the single exception of cystatin C amyloid in hereditary cerebral haemorrhage with amyloidosis, Icelandic type, in which there are major, though clinically silent, systemic deposits. Alzheimer’s disease By far the most frequent and important type of amyloid in the brain is that related to Alzheimer’s disease, which is the most common cause of dementia and affects more than 3 million individuals in the United States of America and a corresponding proportion of other Western populations. It is generally a disease of older people, and its prevalence is therefore increasing. The clinical differential diagnosis of senile de- mentia and the positive identification of Alzheimer’s disease are diffi- cult and often of limited precision in life. However, intracerebral and cerebrovascular amyloid deposits are hallmarks of the neuropatho- logical diagnosis and are detectable by PET imaging with radionuclide-​ labelled ligands specific for amyloid fibrils composed of β-​protein (Aβ). The Alzheimer’s disease Aβ protein is a 39-​ to 43-​residue cleavage product of the large amyloid precursor protein. The vast majority of cases of Alzheimer’s disease are sporadic, but there are also families with an autosomal dominant pattern of inheritance and usually early onset. In about 20 families there are causative mutations in the APP gene for amyloid precursor protein on chromosome 21, and most other kindreds have mutations in the genes for presenilin 1 (chromo- some 14) and presenilin 2 (chromosome 1). All these mutations are associated with increased production from amyloid precursor pro- tein of Aβ1–​42, the most amyloidogenic form of Aβ. Since all in- dividuals with Down’s syndrome (trisomy 21) develop Alzheimer’s disease if they survive into their 40s, there is evidently a close link between amyloid precursor protein, Aβ overproduction, Aβ amyl- oidosis, and the pathogenesis of Alzheimer’s disease. However, it re- mains unclear whether or how Aβ per se, or the amyloid fibrils that it forms, contribute to the neuronal loss that underlies the dementia. Synthetic Aβ fibrils formed in vitro are markedly cytotoxic, and cause the death of cultured cells by apoptosis and necrosis. Although it is not clear to what extent these findings reflect phenomena that may be responsible for neurodegeneration in vivo, there is increasing evidence, from both transgenic mouse models of Alzheimer’s dis- ease and in vivo intracerebral injection of different molecular con- formations of Aβ, that small oligomeric prefibrillar aggregates of Aβ are associated with and cause cognitive dysfunction. There is controversy about the correlation between the severity of dementia in Alzheimer’s disease and the extent of amyloid angiopathy and plaques. Nevertheless, the fact that patients with Alzheimer’s dis- ease caused by amyloid precursor protein and presenilin mutations have exactly the same neuropathology as sporadic cases, including tangles, argues strongly that the amyloid precursor protein and Aβ pathway can be of primary pathogenetic significance. In addition to the Aβ deposits in the brains of patients with Alzheimer’s disease and Down’s syndrome, there are also exten- sive ‘amorphous’ deposits throughout the brain. These do not stain with Congo red, and are detectable only by immunohistochemical staining. Their significance is unknown. They apparently precede the appearance of histochemically identifiable amyloid, but are not necessarily the precursor of it because they are present in areas such as the cerebellum in which Aβ is never seen. The nonfibrillar, nonamyloid protein apolipoprotein E is demonstrable in many amyloid deposits, including those of Alzheimer’s disease. The APOE4 gene (chromosome 19), encoding one of the three isoforms of this apolipoprotein, is strongly associated with a predisposition to develop Alzheimer’s disease and with increased amounts of amyloid in the brain, but the underlying mechanisms are unknown. Another neuropathological feature of Alzheimer’s disease, and some other neurodegenerative conditions, is the neurofibrillary tangle located intracellularly within neuronal cell bodies and pro- cesses. These tangles have a characteristic ultrastructural morph- ology of paired helical filaments, and are composed of an abnormally Box 12.12.3.2  Cerebral amyloidosis • Age-​related amyloid angiopathy with or without intracerebral deposits (cerebral amyloid angiopathy) • Hereditary amyloid angiopathy of meningeal and cortical vessels asso- ciated with cerebral haemorrhage:

—​ Icelandic type (variant cystatin C)

—​ Dutch type (variant Aβ) • Hereditary amyloid angiopathy affecting the entire central nervous system • Alzheimer’s disease:  sporadic, familial, or associated with Down’s syndrome • Cerebral amyloid associated with prion disease:

—​ Sporadic spongiform encephalopathy, Creutzfeldt–​Jacob disease, variant Creutzfeldt–​Jakob disease

—​ Familial prion disease, familial Creutzfeldt–​Jacob disease, Gerstmann–​Sträussler–​Scheinker syndrome atypical familial prion disease • Familial oculoleptomeningeal amyloidosis

12.12.3  Amyloidosis 2223 phosphorylated form of the normal neurofilament protein, tau. They bind Congo red and then give the pathognomonic green birefrin- gence of amyloid when viewed in polarized light. Although their electron microscopic ultrastructure is distinctly different from that of amyloid fibrils, the most recent review of amyloid nomenclature includes them as amyloid. All extracellular and most intracellular tangles are coated with serum amyloid P component. Sporadic cerebral Aβ amyloidosis and cerebral amyloid angiopathy At autopsy, up to 60% of the brains of nondemented older individuals contain Aβ amyloid in the cerebral blood vessels, and there may also be focal intracerebral Aβ plaques. These deposits may or may not have been harbingers of Alzheimer’s disease, had the patients sur- vived long enough, but are increasingly recognized as an important cause of cerebral haemorrhage and stroke, to be distinguished from atherosclerotic cerebrovascular disease. Progressive deposition of Aβ amyloid occurs in the walls of arteries of up to about 2 mm in diameter, arterioles, and capillaries in the cerebral cortex and lepto- meninges. Deposits may less commonly occur in the cerebellum and brainstem. Cerebral amyloid angiopathy is a common cause of intracerebral haemorrhage beyond the age of 60 years, which has a very high mortality approaching 50%. Recognized consequences of the deposits also include progressive cognitive impairment, which may be rapid, and transient neurological symptoms. Aβ amyloid deposition interferes with vascular structure resulting in loss of smooth muscle cells, thickening of vessel walls, and lu- minal narrowing, endothelial dysfunction and fragility that the ren- ders vessels susceptible to microaneurysm formation and bleeding. Advancing age and APOE genotype are the only known risk fac- tors for sporadic cerebral amyloid angiopathy. Polymorphisms in the APOE gene encoding the ε2 and ε4 isoforms are associated with greater risk of developing the disease and its severity, and individ- uals with both these alleles have the earliest onset of disease and greatest probability of recurrent stroke. The mechanism by which apolipoprotein E contributes to disease are not clear, but may in- clude promoting Aβ amyloid deposition and inducing structural changes in involved vessels. Hereditary cerebral haemorrhage with amyloidosis: hereditary cerebral amyloid angiopathy Icelandic type (OMIM 604312) Cerebrovascular amyloid deposits composed of a fragment of a gen- etic variant of cystatin C are responsible for recurrent major cere- bral haemorrhages starting in early adult life in members of families originating in western Iceland. There is autosomal dominant in- heritance and appreciable, but clinically silent, amyloid deposits are present in the spleen, lymph nodes, and skin. There is no extra- vascular amyloid in the brain, and the neurological deficits, often including dementia, of surviving patients are compatible with their cerebrovascular pathology. Dutch type (OMIM 605714) In families originating from a small region on the coast of the Netherlands the autosomal dominant inheritance of a genetic variant of Aβ, which is deposited as cerebrovascular amyloid, results in re- current normotensive cerebral haemorrhages starting in middle age. There are also amorphous Aβ deposits in the brain and early senile plaques, without congophilic amyloid cores. Multi-​infarct dementia occurs in survivors, but some patients become demented in the ab- sence of stroke. Amyloid outside the brain has not been reported. Cerebral amyloid associated with prion disease The neuropathology of a group of progressive, invariably fatal spongiform encephalopathies sometimes, but certainly not always, includes intracerebral amyloid plaques. These diseases are trans- missible and in some cases hereditary. The sporadic and familial Creutzfeldt–​Jacob disease, the familial Gerstmann–​Sträussler–​ Scheinker syndrome, and kuru are caused by prions (PrPSc), con- formational isoforms of the normal physiological cellular prion protein (PrPC). The human diseases are closely related to the animal diseases scrapie of sheep and goats, transmissible encephalopathy of mink, elk, and male deer, and bovine spongiform encephalopathy. Variant Creutzfeldt–​Jacob disease is apparently the result of trans- mission of bovine spongiform encephalopathy to humans. The significance of amyloid per se in these disorders is not clear because it is not always histologically detectable, and in some dis- orders is not seen, for example, fatal familial insomnia and bovine spongiform encephalopathy (which is apparently a result of the transmission of ovine scrapie to cattle). When scrapie or its human counterparts are transmitted to experimental animals by inoculation of affected brain tissue, the development of intracerebral amyloid depends on the strain of infectious agent and the genetic back- ground of the recipient. Even when amyloid is present in the brain it is not seen elsewhere (e.g. in the spleen), although the latter is a rich source of the infective agent. However, when the infective agent is exhaustively and highly purified from brain or spleen it forms typ- ical congophilic amyloid fibrils composed of the proteinase-​resistant subunit PrPSc, and when amyloid deposits are present in affected brains they immunostain with antiprion antibodies. The amyloid fibril protein is thus directly related to the cause of the encephalopathy, but histologically demonstrable amyloid de- position is evidently not necessary for the expression of disease. Indeed, recent work in transgenic and knockout mouse strains clearly demonstrates both that prion amyloid deposition is not a necessary condition for the development of transmissible spongi- form encephalopathy, and that expression of the normal cellular isoform, PrPC, is absolutely required. Neuronal damage may per- haps be caused by a cytotoxic interaction between prefibrillar PrPSc aggregates and the normal PrPC, or indeed by other mechanisms entirely. This is a different situation from the extracerebral amyl- oidosis, and from cystatin C and nonhereditary cerebral amyloid angiopathies, in which amyloid deposition is invariably present when there is clinical disease, and is unequivocally the cause of tissue damage. Hereditary systemic amyloidosis Hereditary transthyretin amyloidosis (OMIM 176300) Hereditary transthyretin amyloidosis is an autosomal dominant ­syndrome with onset at any time from the second decade onwards. It is characterized by progressive peripheral and autonomic neur- opathy and varying degrees of visceral involvement especially af- fecting the heart, the h vitreous of the eye, kidneys, thyroid, and adrenals. There are usually amyloid deposits throughout the body involving the walls of blood vessels and the connective tissue matrix; the pathology is due to these deposits. Major foci exist in Portugal,

section 12  Metabolic disorders 2224 Japan, and Sweden but familial amyloid polyneuropathy has been reported in most populations throughout the world. There is con- siderable variation in the age of onset, rate of progression, and in- volvement of different systems, although within families the pattern tends to be more consistent. There is remorseless progression over 5 to 15 years and the disorder is invariably fatal. Death results from the effects and complications of peripheral and/​or autonomic neur- opathy, or from cardiac or renal failure. Hereditary transthyretin amyloidosis is caused by mutations in the gene for the plasma protein transthyretin (formerly known as prealbumin). The most frequent of these worldwide causes a me- thionine for valine substitution at position 30 in the mature protein, but more than 100 amyloidogenic mutations have been described; substitution of alanine for threonine at position 60 is the most fre- quent cause in the United Kingdom. There is often little correlation between the underlying mutation and the clinical phenotype, which is evidently determined by other genetic and possibly also environ- mental factors, although in a few cases certain mutations are uniquely associated with particularly aggressive or relatively organ-​limited dis- ease. The amyloidogenic transthyretin mutations are not always pene- trant, and asymptomatic methionine-​30 homozygotes over the age of 60 years have been reported. Rare kindreds with the apolipoprotein A-​ I arginine 26 variant, which usually causes non-​neuropathic amyloid- osis, may present with prominent peripheral neuropathy resembling transthyretin familial amyloid polyneuropathy (OMIM 107680). Familial amyloid polyneuropathy with predominant cranial neuropathy (OMIM 137350) Originally described in Finland, but now reported in other popula- tions, this very rare, autosomal dominant, hereditary amyloidosis presents in adult life with cranial neuropathy, lattice corneal dys- trophy, and distal peripheral neuropathy. There may be skin, renal, and cardiac manifestations and microscopic amyloid deposits are widely distributed in connective tissue and blood vessel walls; life expectancy approaches normal. The amyloid fibrils are derived from variants of the actin-​modulating protein gelsolin, encoded by point mutations. Non-​neuropathic systemic amyloidosis (OMIM 105200) In this rare autosomal dominant syndrome of major systemic amyl- oidosis without clinical evidence of neuropathy, the patterns of organ involvement and overall clinical phenotype vary between families. The kidneys are often the most severely affected organ, leading to hypertension and renal failure, but the heart, spleen, liver, bowel, connective tissue, and exocrine glands may all be involved. Following clinical presentation there is inexorable progression to death or organ failure requiring transplantation. Clinical presen- tation is typically around the sixth decade, but can occur in early adulthood. The amyloid proteins so far identified are genetic vari- ants of apolipoprotein A-​I and A-​II, lysozyme, and least rarely the fibrinogen α-​chain. A single kindred has lately been identified with a genetic variant of β2-​microglobulin. Familial amyloid cardiomyopathy Cardiac amyloidosis without overt involvement of other viscera or neuropathy, progressing inexorably to death, is associated with cer- tain transthyretin gene mutations and is inherited in an autosomal dominant manner with variable penetrance (Table 12.12.3.2). By far the most common variant is transthyretin isoleucine 122, which occurs in 4% of African Americans and causes cardiac amyloidosis from the sixth decade onwards. Familial Mediterranean fever (OMIM 249100) Familial Mediterranean fever is an autosomal recessive autoinflammatory disorder caused by mutations in the gene MEFV on chromosome 16 that encodes a neutrophil-​specific protein of unknown function called pyrin (see Chapter 12.12.2). The disease is characterized by recurrent episodes of fever, abdominal pain, pleurisy, or arthritis, and predominantly occurs in non-​Ashkenazi Jews, Armenians, Anatolian Turks, and Levantine Arabs. Among Sephardi Jews of North African origin, and in the other populations (except Armenians and, to a lesser extent, Ashkenazi Jews), untreated familial Mediterranean fever is eventually complicated in a high pro- portion of cases by typical systemic AA amyloidosis. Furthermore, some patients with familial Mediterranean fever present with AA amyloidosis before they have experienced any symptoms, and this is consistent with the recent finding that a substantial acute phase plasma protein response is frequently present, even in asymptom- atic individuals. The variable incidence of amyloidosis in patients with familial Mediterranean fever from different populations is not wholly explained by their specific pyrin gene mutations, and is an- other illustration of the unknown genetic determinants of clinical amyloidosis. Heterozygosity for deletion of methionine at position 694 of pyrin is associated with dominantly inherited but otherwise typical familial Mediterranean fever in the British population. Haemodialysis-​associated amyloidosis In the past, almost all patients with end-​stage renal failure who were maintained on haemodialysis for more than 5 years developed amyloid deposits composed of β2-​microglobulin. These deposits were predominantly osteoarticular and were associated with carpal tunnel syndrome, large-​joint pain and stiffness, soft tissue masses, bone cysts, and pathological fractures. Renal tubular amyloid con- cretions could also form. The serious clinical problems associated with such amyloidosis were a major cause of morbidity in patients on long-​term dialysis. Furthermore, in some patients more exten- sive deposition occurred, most commonly in the spleen but also in other organs, and a few cases of death associated with systemic β2-​ microglobulin amyloid were reported. The β2-​microglobulin was derived from the high plasma con- centration that developed in renal insufficiency, which was not adequately cleared by dialysis. This condition also occurred (with lower incidence) in patients on continuous ambulatory peritoneal dialysis and has rarely been reported in patients with chronic kidney disease who have never been dialysed. Improved clearance of β2-​ microglobulin by current dialysis membranes and procedures has much reduced the incidence and severity of this form of amyloidosis. Leucocyte chemotactic factor 2 amyloidosis A recently identified and still obscure form of systemic amyloidosis is associated with amyloid deposits derived from leucocyte chemotactic factor 2 (LECT2). The aetiology of this apparently sporadic disease is unknown. Most patients present in their sixth to seventh decades with slowly progressive renal impairment; proteinuria tends to be low grade and hypertension is a frequent accompaniment. Although splenic and adrenal amyloid deposits are frequently evident on serum

12.12.3  Amyloidosis 2225 amyloid P component scintigraphy, the clinical manifestations are predominantly renal. The disease is apparently most prevalent in Mexican Americans and South Asians, and renal biopsy data from the United States of America suggests that ALECT2 amyloidosis may account for up to 2.5% of amyloidotic renal biopsy specimens. The LECT2 gene, located on chromosome 5q31.1–​q32, consists of four exons that encode 151 amino acids, of which 133 are present in the secreted product. In addition to having chemotactic activity for neutrophils, LECT2 may have other physiological functions related to cell growth and repair after damage. Although synthesized pre- dominately in the liver, it also is expressed in other tissues, including testis, vascular and endothelial smooth muscle cells, and kidney. The pathogenesis underlying ALECT2 amyloid formation is un- known, but the fibrils have been demonstrated to be derived from full-​length, 133-​residue LECT2 protein. Endocrine amyloid Many tumours of APUD (amine precursor uptake and decarboxyl- ation) cells that produce peptide hormones have amyloid deposits in their stroma. These are probably composed of the hormone peptides; in the case of medullary carcinoma of the thyroid, the fibril sub- units are derived from procalcitonin. In insulinomas, the amyloid fibril protein is a novel peptide first identified in that site and subse- quently shown to be the fibril protein in the amyloid of the islets of Langerhans seen in type 2 (maturity-​onset) diabetes. This peptide is called islet amyloid polypeptide, or amylin, and shows appreciable homology with calcitonin gene-​related peptide. Amyloid of this type is an almost universal feature of the pancreatic islets in type 2 diabetes, and becomes more extensive with increasing duration and severity of the disease. Although the amyloid itself is probably not initially responsible for the metabolic defect in this form of dia- betes, it is likely that progressive amyloid deposition, leading to islet destruction, subsequently does contribute to the pathogenesis. The possible hormonal or other role of islet amyloid polypeptide itself, which is produced by the islet β-​cells, is also not yet clear. Rare localized amyloidosis syndromes Amyloid deposits localized to the skin occur in both acquired and hereditary forms of amyloidosis. Primary localized cutaneous amyl- oidosis presents in adult life as macular or papular lesions, the fibrils of which may be derived from keratin. Hereditary cutaneous amyloid lesions are rare, of unknown fibril type, and are sometimes associated with other, nonamyloid, multisystem disorders. Amyloid deposits in the eye cause local problems in the cornea (corneal lattice dystrophy) or conjunctiva, while orbital amyloid presents as mass lesions that can disrupt eye movement and the structure of the orbit. In one such case, the fibril protein has been identified as a fragment of IgG heavy chain. Lactoferrin and keratoepithelin have been identified as the amyloid fibril proteins in different cases of corneal amyloidosis. Localized foci of AL amyloid can occur anywhere in the body in the absence of systemic AL amyloidosis, the most common sites being the skin, upper airways and respiratory tract, and the uro- genital tract. They may be associated with a local plasmacytoma or B-​ cell lymphoma producing a monoclonal immunoglobulin, but often the cells, which must be present to produce the amyloidogenic pro- tein, are scattered inconspicuously in the affected tissue. The clinical problems caused by these space-​occupying amyloidomas can often be ameliorated by surgical resection, but this is not always possible. Amyloid fibrils Regardless of their very diverse protein subunits, amyloid fibrils of different types are remarkably similar: straight, rigid, nonbranching, of indeterminate length, and 10 to 15 nm in diameter. They are insol- uble in physiological solutions, relatively resistant to proteolysis, and bind Congo red dye, producing pathognomonic green birefringence when viewed in intense cross-​polarized light. Electron microscopy reveals that each fibril consists of two or more protofilaments, the precise number varying with the fibril type. The X-​ray diffraction patterns of all the different ex vivo amyloid fibrils, and of synthetic fibrils formed in vitro, that have been studied demonstrate the pres- ence of a common core structure within the filaments: the subunit proteins are arranged in a stack of twisted antiparallel β-​pleated sheets lying with their long axes perpendicular to the long axis of the fibril. Many different proteins, including molecules totally unrelated to amyloidosis in vivo, can be refolded after denaturation in vitro to form typical, stable, congophilic cross-​β fibrils. Although it is not clear why only the 26 known amyloidogenic proteins adopt the amyloid fold and persist as fibrils in vivo, a major unifying theme is that, in all cases studied, the precursors are relatively destabilized. Even under physiological or other conditions they may encounter in vivo, they adopt partly unfolded states that involve the loss of tertiary or higher-​order structure. These misfolded intermediates readily aggregate, with retention of β-​sheet secondary structure, into protofilaments and fibrils. Once the process has started, seeding may also play an important facilitating role, so that amyloid deposition may progress exponentially as expansion of the amyloid template captures further precursor molecules. Amyloid fibril proteins and their precursors Immunoglobulin light chain AL proteins are derived from the N-​terminal region of monoclonal immunoglobulin light chains and consist of all or part of the vari- able domain. Intact light chains may occasionally be found, and the molecular weight therefore varies between about 8 and 30 kDa. The light chain of the monoclonal paraprotein is either identical to, or clearly the precursor of, AL isolated from the amyloid deposits. AL fibrils are more commonly derived from λ chains than from κ chains, despite the fact that κ chains predominate among both normal immunoglobulins and the paraprotein products of immunocyte dyscrasias. A  particular λ-​chain subgroup, λVI, was initially identified as an AL protein in two cases of immunocyte dyscrasia-​associated amyloidosis, before it had been recognized in any other form, and it is present in many more cases of AL amyl- oidosis. Furthermore, sequence analysis of Bence Jones proteins of both κ and λ types from patients with AL amyloidosis, and of AL proteins themselves, shows that these polypeptides contain unique amino acid replacements or insertions compared with nonamyloid monoclonal light chains. In some cases, the changes involve the re- placement of hydrophilic framework residues by hydrophobic res- idues, changes likely to promote aggregation and insolubilization; in others, the monoclonal light chains from amyloid patients have been directly demonstrated to have decreased solubility and a greater propensity for precipitation than control nonamyloid proteins. The inherent amyloidogenicity of particular monoclonal light chains

section 12  Metabolic disorders 2226 has been elegantly confirmed in an in vivo model in which isolated Bence Jones proteins were injected into mice. Animals receiving light chains from AL amyloid patients developed typical amyloid de- posits composed of the human protein, whereas animals receiving light chains from myeloma patients without amyloid did not. Amyloid A (AA) The AA protein is a single nonglycosylated polypeptide chain usually of mass 8000 Da and containing 76 residues corresponding to the N-​ terminal portion of the 104-​residue serum amyloid A protein (SAA). Smaller and larger AA fragments, even some whole SAA molecules, have also been reported in AA fibrils. SAA is an apolipoprotein of high-​density lipoprotein particles, and is the polymorphic product of a set of genes located on the short arm of chromosome 11. It is highly conserved in evolution and is a major acute phase reactant. Most of the SAA in plasma is produced by hepatocytes, in which the synthesis is under transcriptional regulation by cytokines (especially inter- leukin 1, interleukin 6, and TNF) acting via NF-​κB-​like transcription factors, and possibly others. After secretion it rapidly associates with high-​density lipoproteins, from which it displaces apolipoprotein A-​I. The circulating concentration can rise from normal levels of up to 3 mg/​litre to over 1000 mg/​litre within 24 to 48 h of an acute stimulus, and with ongoing chronic inflammation the level may re- main persistently high. Certain isoforms of SAA, the products of dif- ferent genes, are predominantly synthesized elsewhere in the body by macrophages, adipocytes, and certain other cells. Although they also associate with high-​density lipoproteins, their acute phase synthesis is stimulated differently and they presumably have different func- tions. There is also a closely related family of high-​density lipoprotein trace apoproteins that are not acute phase reactants; they have been designated constitutive SAAs, although they do not form amyloid. The precursor of amyloid fibril AA protein is circulating SAA, from which amyloid fibril AA protein is derived by proteolytic cleavage. Such cleavage can be produced by macrophages and by a variety of proteinases, but since further cleavage of AA is readily demon- strable in vitro it is not clear why the AA peptide persists in amyloid. Furthermore, it is not known whether the process of AA fibril gen- eration involves cleavage of SAA before and/​or after aggregation of monomers. Persistent overproduction of SAA causing sustained high circulating levels is a necessary condition for deposition of AA amyloid, but it is not known why only some individuals in this state develop amyloidosis. In mice, only one of the three major isoforms of SAA is the precursor of AA in amyloid fibrils. Human SAA isoforms are more complex, but homozygosity for particular types seems to fa- vour amyloidogenesis, although there may also be ethnic differences. The normal functions of SAA are not known, although modu- lating effects on reverse cholesterol transport and on lipid function in the microenvironment of inflammatory foci have been proposed. A protein, homologous with SAA, produced by rabbit fibroblasts has been reported to act as an autocrine stimulator of collagenase pro- duction in vitro. Other reports of potent cell-​regulatory functions of isolated denatured delipidated SAA have yet to be confirmed in physiological preparations of SAA-​rich high-​density lipoproteins. Regardless of its physiological role, the behaviour of SAA as an ex- quisitely sensitive acute phase protein with an enormous dynamic range makes it an extremely valuable empirical clinical marker. It can be used objectively to monitor the extent and activity of infective, in- flammatory, necrotic, and neoplastic disease. Furthermore, routine monitoring of SAA should be an integral part of the management of all patients with AA amyloidosis or disorders predisposing to it, as control of the primary inflammatory process in order to reduce SAA production is essential if amyloidosis is to be halted, enabled to regress, or prevented. Automated immunoassay systems for SAA are available that are calibrated on the World Health Organization international reference standard. Transthyretin Transthyretin, formerly known as prealbumin, is a normal nonglycosylated plasma protein with a relative molecular mass of 55 044 Da. It is composed of four identical noncovalently associ- ated subunits, each of 127 amino acids. It is produced by hepatocytes and the choroid plexus, and is a significant negative acute phase protein. Each tetrameric molecule is able to bind a single thyroxine or triiodothyronine molecule and up to 15% of circulating thyroid hormone is transported in this way. Transthyretin also forms a 1:1 molecular complex with the vitamin A transporter, retinol-​binding protein, and is essential for keeping it in the circulation. Transthyretin is encoded by a single-​copy gene, but is appreciably polymorphic and more than 120 different point mutations encoding single residue substitutions have been identified. Normal wild-​type transthyretin is an inherently amyloidogenic protein that forms the fibrils in wild-​type transthyretin (also known as senile systemic) amyloidosis. Almost all variant forms of transthyretin have been as- sociated with hereditary amyloidosis, and show decreased stability in vitro compared with the wild type. Individuals heterozygous for transthyretin mutations have a mixture of wild-​type and variant transthyretin monomers in their circulating transthyretin, and if they develop amyloidosis both forms are often present, although the variant may predominate in the amyloid fibrils. In addition to intact protomers, ATTR amyloid fibrils also usually con- tain the C-terminal fragment, residue 49-127, of transthyretin. This is produced by a recently elucidated mechano-enzymatic mechanism, in which physiological scale shear forces critically facilitate cleavage of transthyretin at residue 48 by plasmin. This specific cleavage catalyses, and is necessary for, transthyretin amyloid fibril formation. Plasmin, the pivotal essential mediator of physiological fibrinolysis, is thus also a key pathogenic factor in transthyretin amyloidosis. Amyloid beta (Aβ) The fibril protein in the intracerebral and cerebrovascular amyloid of Alzheimer’s disease, Down’s syndrome, and hereditary amyloid angiopathy of the Dutch type is a 39-​ to 43-​residue sequence de- rived by proteolysis from a precursor protein of high molecular weight, the amyloid precursor protein (APP), encoded on the long arm of chromosome 21. Several isoforms of APP are generated by alternative splicing of transcripts from the 19-​exon gene, yielding three major forms: APP695, APP751, and APP770. These are each single-​chain, multidomain glycoproteins with the 47 residues of the C-​terminal lying within the cytoplasm, a 25-​residue membrane-​ spanning region, and the rest of the molecule lying extracellularly. APP751 and APP770 contain a 56-​residue Kunitz-​type serine pro- teinase inhibitor domain encoded by exon 7. Following glycosylation and membrane insertion, APPs are cleaved extracellularly, close to the transmembrane sequence, by so-​ called APP secretase activity. This releases, in the case of APP751 and APP770, a molecule known as proteinase nexin II, which avidly

12.12.3  Amyloidosis 2227 binds factor XIa, trypsin, and chymotrypsin, as well as epidermal growth factor-​binding protein and the γ subunit of nerve growth factor. The predominant species of mRNA found in the brain encodes APP695, which lacks the proteinase inhibitor domain, while mRNA for APP751 is the most abundant in other tissues. Despite this, 85% of secreted APP in the brain is proteinase nexin II. Interestingly, APP secreted by a glial cell line is substantially glycosylated with chon- droitin sulphate glycosaminoglycan chains. APP also undergoes high-​affinity interactions with heparan sulphate. These observations suggest that APP may have important functions in cell adhesion, cell migration, and modulation of growth-​factor activities. APP pro- teinase nexin II is present in and released by platelets, and probably functions in the clotting cascade. The amyloidogenic peptide Aβ, encoded by parts of exons 16 and 17, corresponds to the part of the APP sequence that extends from within the cell membrane into the extracellular space. Secretase cleavage of APP to release the soluble form cannot therefore gen- erate intact Aβ itself, or larger fragments containing it. However, there is an alternative processing pathway for APP, in which it is taken up whole by lysosomes and cleaved to yield fragments that do contain the whole Aβ sequence. Furthermore, APP cleaved at the N-​ terminus of Aβ, and soluble Aβ itself, are normally produced by cell lines and by mixed brain cells in culture, and are present in the cere- brospinal fluid. However, the source of the Aβ in the intracerebral amorphous deposits, and that which aggregates as amyloid fibrils in the brain and cerebral blood vessels, is still not known. The 42-​ residue form of Aβ (Aβ1–​42) is markedly the most amyloidogenic, and all the mutations in the APP and presenilin genes that are associ- ated with hereditary Alzheimer’s disease result in increased produc- tion of this fragment. Increased availability of the precursor is thus responsible for amyloidogenesis, but the pathogenesis of neuronal damage and dementia remain unclear. Cystatin C Cystatin C (formerly called γ-​trace) is an inhibitor of cysteine pro- teinases, including cathepsin B, H, and L. It is encoded by a gene on chromosome 20 and consists of a single nonglycosylated polypep- tide chain of 120 residues. It is present in all major human biological fluids at concentrations compatible with a significant physiological role in proteinase inhibition. The normal concentration in cere- brospinal fluid is 6.5 mg/​litre (range 2.7–​13.7), but is much lower (2.7 mg/​litre, range 1.0–​4.7) in patients with the Icelandic type of hereditary cerebral amyloid angiopathy, in whom fragments of the glutamine-​68 genetic variant of cystatin C form the amyloid fibrils. This reduced concentration is diagnostically useful, and is evident even in presymptomatic carriers of the cystatin C gene mutation. The point mutation that causes the disease encodes a glutamine for leucine substitution in the mature protein, and the amyloid fibril protein consists of the C-​terminal 110 residues of the variant. This N-​terminally truncated form is not detectable in the cerebrospinal fluid of affected patients, suggesting that cleavage takes place either in close proximity to fibril deposition or after the fibrils have formed. The variant cystatin C is less stable than the wild type and readily forms fibrils in vitro. It is not known whether cerebral haemorrhage in cystatin C amyloidosis is caused simply by the damaging effects of vascular amyloid deposition, or whether deficiency in inhibitory capacity for cysteine proteinases also plays a part. Gelsolin Gelsolin is a widely distributed 90-​kDa cytoplasmic protein that binds actin monomers, nucleates actin filament growth, and severs actin filaments. Alternative transcriptional initiation and message processing from a single gene on chromosome 9 are responsible for the synthesis of a secreted form of gelsolin (93 kDa), which circu- lates in the plasma at a concentration of about 200 mg/​litre. Its func- tion in the blood is not known, but may be related to the clearance of actin filaments released by dying cells. In the Finnish type of heredi- tary amyloidosis, the amyloid fibril protein is a 71-​residue fragment of variant gelsolin, with asparagine substituted for aspartic acid at position 15 (corresponding to residue 187 of the mature molecule), and the same mutation has been discovered in affected kindreds from different ethnic backgrounds. In one Danish family with the same phenotype there is a different mutation at the same nucleo- tide, predicting a tyrosine for aspartic acid substitution at residue 187. Synthetic and recombinant peptides that include the asparagine for aspartic acid substitution at residue 187 are less soluble than the wild-​type sequence and readily form amyloid fibrils in vitro. Apolipoprotein A-​I and A-​II Apolipoprotein A-​I is the most abundant apolipoprotein among the high-​density lipoprotein particles, and participates in their central function of reverse cholesterol transport from the periphery to the liver. Apolipoprotein A-​I variants are extremely rare, and may be phenotypically silent or may affect lipid metabolism. However, at least 15 different variants of apolipoprotein A-​I, including single-​ and multiple-​residue substitutions and deletions, have been associ- ated with amyloidosis. These are inherited in an autosomal dominant manner and are usually highly penetrant, but there are marked vari- ations in the age and manner of presentation, even within the same family and in different kindreds with the same mutation. The amyloid fibril protein consists, in all cases studied, of the first 90 or so N-​terminal residues, even when the causative variant residue(s) are more distal. Wild-​type apolipoprotein A-​I is also amyloidogenic, forming the deposits associated with atheromatous plaques in older people. The various amyloidogenic mutations pre- sumably encode sequence changes that render apolipoprotein A-​I less stable and/​or more liable to cleavage that yields the fibrillogenic N-​terminal fragment. Predominantly renal amyloidosis has also been described in a handful of families in association with several different mutations in a normal stop codon in the gene for apolipoprotein A-​II, which results in a peptide extension from residue 78. Lysozyme Lysozyme is the classic bacteriolytic enzyme of external secretions, discovered by Fleming in 1922. It is also present at high concentra- tion within articular cartilage and in the granules of polymorphs, and is the major secreted product of macrophages. Lysozymes are present in most organisms in which they have been sought, although their physiological role is not always clear. The complete structures of hen egg-​white and human lysozymes are known to atomic reso- lution, and their catalytic mechanism, epitopes, folding, and other aspects of their structure–​function relationship have been analysed exhaustively. Wild-​type lysozyme, unlike transthyretin and β2-​ microglobulin, is not inherently amyloidogenic and, coupled with

section 12  Metabolic disorders 2228 the extensive knowledge of its structure and folding, it is therefore a valuable model for the investigation of amyloid fibrillogenesis. There is only one copy of the lysozyme gene in the human genome, and no disease is associated with lysozyme other than amyloidosis. The first lysozyme mutations identified to cause amyloidosis were the sub- stitution of threonine for isoleucine at residue 56 in one family, and histidine for aspartic acid at residue 67 in another. These dramatic changes in residues that are extremely conserved throughout the lysozyme and related α-​lactalbumin protein families destabilize the native fold, so that the variants readily adopt partly unfolded states, even under physiological conditions, and spontaneously aggregate in vitro, and evidently also in vivo, into amyloid fibrils. Several fur- ther amyloidogenic variants of lysozyme have subsequently been described. Islet amyloid polypeptide Islet amyloid polypeptide (IAPP; amylin) is a 37-​residue molecule encoded by a gene on chromosome 12 and with 46% sequence hom- ology to the neuropeptide calcitonin gene-​related peptide. Islet amyloid polypeptide is produced in the β-​cells of the pancreatic is- lets of Langerhans, and is stored in and released from their secretory granules together with insulin. It has been reported to modulate in- sulin release and to induce peripheral insulin resistance, vasodilata- tion, and lowering of plasma calcium, but neither its physiological role nor its contribution to diabetes are known. The amyloidogenicity of islet amyloid polypeptide depends on the amino acid sequence between residues 20 and 29, as shown by in vitro fibrillogenesis with synthetic peptides. The synthetic deca- peptide IAPP20–​29, and even the hexapeptide IAPP25–​29, form amyloid-​like fibrils in vitro, whereas other fragments do not. There is also a correlation between conservation of this sequence and de- position of the amyloid in the islets of diabetic animals of different species. However, the role of the amyloid in diabetogenesis remains to be established. In the degu, a South American rodent, spontan- eous diabetes is associated with islet amyloid composed of insulin, and xenogeneic insulin can also form amyloid in humans at the site of repeated therapeutic insulin injections. β2-​microglobulin β2-​microglobulin is a nonglycosylated, nonpolymorphic, single-​ chain protein of 99 residues, with a single intrachain disulphide bridge and a relative molecular mass of 11 815, encoded by a single gene on chromosome 15. It becomes noncovalently associated with the heavy chain of major histocompatibility class I antigens, and is required for transport and expression of the major histocompati- bility complex (MHC) at the cell surface. Amino acid sequence homology places β2-​microglobulin in the superfamily that includes immunoglobulins, T-​cell receptor α-​ and β-​chains, Thy-​1 (CD90), MHC class I and II molecules, secretory component, etc. Its three-​ dimensional structure is a typical β-​barrel with two antiparallel pleated sheets comprising three and four strands, respectively, and closely resembles an immunoglobulin domain. β2-​microglobulin is produced by lymphoid and a variety of other cells, in which it stabilizes the structure and function of MHC class I antigens at the cell surface. When these complexes are shed by cleavage of the heavy chain at the cell surface, free β2-​microglobulin is released. The circulating concentration of β2-​microglobulin is 1 to 2 mg/​litre and the protein is rapidly cleared by glomerular filtra- tion and then catabolized in the proximal renal tubule. Impairment of renal function is associated with retention of β2-​microglobulin and increased circulating concentrations because there is no other site for its catabolism. Daily production of β2-​microglobulin is about 200 mg, and in patients in end-​stage renal failure on haemodialysis, plasma β2-​microglobulin levels rise to and remain at levels of about 40 to 70 mg/​litre. Isolated unaltered β2-​microglobulin can form amyloid-​like fibrils itself in vitro, and most studies of ex vivo β2-​ microglobulin fibrils show the whole intact molecule to be the major subunit, although fragments and altered forms of β2-​microglobulin have also been reported. Intriguingly, a highly amyloidogenic variant of β2-​microglobulin has been identified in a French kindred with hereditary systemic amyloidosis manifesting with slowly progressive gastrointestinal symptoms and autonomic neuropathy. Renal function and cir- culating concentration of β2-​microglobulin were normal. The Asp76Asn β2-​microglobulin variant was thermodynamically un- stable and uniquely fibrillogenic in vitro under physiological conditions. Glycosaminoglycans Amyloidotic organs contain more glycosaminoglycans than normal tissues, and at least some of this is a tightly bound integral part of the amyloid fibrils. These fibril-​associated glycosaminoglycans are heparan sulphate and dermatan sulphate in all forms of amyloid that have been investigated. Fibrils isolated by water extraction and separated from other tissue components contain 1 to 2% by weight glycosaminoglycan, none of which is covalently associated with the fibril protein. Interestingly, in systemic AA and AL amyloid- osis, the only forms in which this has been studied so far, there is markedly restricted heterogeneity of the glycosaminoglycan chains, suggesting that particular subclasses of heparan and dermatan sulphates are involved. Immunohistochemical studies demonstrate the presence of proteoglycan core proteins in all amyloid deposits, and that these are closely related to fibrils at the ultrastructural level. However, in isolated fibril preparations much of the glyco- saminoglycan material is free carbohydrate chains, and it is not yet clear whether this represents aberrant glycosaminoglycan metab- olism related to amyloidosis or is just an artefact of postmortem degradation of core protein. The significance of glycosaminoglycans in amyloid remains un- clear, but their universal presence, intimate relationship with the fibrils, and restricted heterogeneity all suggest that they may be im- portant. Glycosaminoglycans are known to participate in the organ- ization of some normal structural proteins into fibrils and they may have comparable fibrillogenic effects on certain amyloid fibril pre- cursor proteins. Furthermore, the glycosaminoglycans on amyloid fibrils may be ligands to which serum amyloid P component, an- other universal constituent of amyloid deposits, binds. Amyloid P component and serum amyloid P component Amyloid deposits in all different forms of the disease, both in hu- mans and in animals, contain the nonfibrillar glycoprotein, amyloid P component. Amyloid P component is identical to and derived from the normal circulating plasma protein, serum amyloid P com- ponent, a member of the pentraxin protein family that includes

12.12.3  Amyloidosis 2229 C-​reactive protein. Human serum amyloid P component is secreted only by hepatocytes, is a trace constituent of plasma (women: mean 24 mg/​litre, range 8–​55, men: mean 32 mg/​litre, range 12–​50), and is not an acute phase reactant. Nevertheless, apart from the fibrils themselves, amyloid P component is always by far the most abun- dant protein in all amyloid deposits. Serum amyloid P component consists of five identical noncovalently associated subunits, each with a molecular mass of 25 462 Da, arranged in a pentameric disc-​like ring. The tertiary fold of the subunit is dominated by antiparallel β-​sheets, forming a flattened β-​barrel with jellyroll topology and a core of hydro- phobic side chains. This is the lectin fold, shared with a variety of other animal, plant, and bacterial carbohydrate-​binding proteins (lectins). Serum amyloid P component is a calcium-​dependent ligand-​binding protein; its best-​defined specificity is for the 4,6-​ cyclic pyruvate acetal of β-​d-​galactose, but it also binds avidly and specifically to DNA, chromatin, glycosaminoglycans (particu- larly heparan and dermatan sulphates), and to all known types of amyloid fibrils. The latter interaction is responsible for the unique specific accumulation of serum amyloid P component in amyloid deposits. Aggregated, but not native, serum amyloid P compo- nent also binds specifically to C4-​binding protein and fibronectin from plasma, although serum amyloid P component is not com- plexed with any other protein in the circulation. In addition to being a plasma protein, human serum amyloid P component is also a normal constituent of certain extracellular matrix structures. It is covalently associated with collagen and/​or other matrix compo- nents in the lamina rara interna of the human glomerular basement membrane, and is present on the microfibrillar mantle of elastin fibres throughout the body. Although no deficiency of serum amyloid P component has been described, and it has been stably conserved in evolution, its physio- logical function remains unclear. There is a single copy of its gene on chromosome 1, no polymorphism of the amino acid sequence, and the single biantennary oligosaccharide chain attached to as- paragine at residue 32 is the most invariant glycan of any known glycoprotein. Studies of serum amyloid P component knockout mice have shown that serum amyloid P component is involved in host resistance to some infections, and contributes to the pathogen- esis of others, but these animals are otherwise healthy and have a normal lifespan. The serum amyloid P component molecule is highly resistant to proteolysis and, although not itself a proteinase inhibitor, its binding to amyloid fibrils in vitro protects them against proteolysis. Once bound to amyloid fibrils in vivo, serum amyloid P component per- sists for very prolonged periods and is not catabolized at all, by con- trast with its rapid clearance from the plasma (half-​life 24 h) and prompt catabolism in the liver. These observations suggest that serum amyloid P component may contribute to the persistence of amyloid deposits in vivo; serum amyloid P component knockout mice show retarded and reduced induction of experimental AA amyloidosis, confirming that serum amyloid P component is signifi- cantly involved in the pathogenesis of amyloidosis. Other proteins in amyloid deposits A number of plasma proteins, other than the fibril proteins them- selves and the serum amyloid P component, have been detected immunohistochemically in some amyloid deposits. These in- clude α1-​antichymotrypsin, some complement components, apolipoprotein E, and various proteins of the extracellular matrix and basement membrane. None of these is as universal, abundant, or selective as serum amyloid P component, and their role, if any, in the pathogenesis or effects of amyloid deposition is not known. Diagnosis and monitoring of amyloidosis The gold standard for diagnosis of amyloid is Congo red staining of histological sections under strictly defined conditions, followed by viewing under intense cross-​polarized light to detect the path- ognomonic green birefringence. Although inherently not complex or challenging, the essential practical procedures are often not per- formed adequately, leading to a substantial proportion of false-​ negative and false-​positive histopathological diagnoses which can be devastating for individual patients. The relative rarity of amyl- oidosis and its highly variable and protean clinical presentation also mean that clinicians often do not think of it and thus do not request its histological identification. The variable standard of histological assessment compounds the problem and together these factors ex- plain why the diagnosis is very challenging in routine practice. The result is that amyloidosis is frequently diagnosed very late and pa- tients already have advanced organ damage before they come to spe- cialist attention. Even when properly stained and viewed, biopsies provide extremely small samples and therefore can never provide information on the extent, localization, progression, or regression of amyloid deposits, either locally or in the whole patient. The devel- opment of radiolabelled serum amyloid P component as a specific tracer for amyloid was therefore a major advance in clinical amyl- oidosis and provided a wealth of new information on the natural history of many different forms of amyloid and their response to treatment. Histochemical diagnosis of amyloidosis Biopsy In many patients with systemic amyloidosis, amyloid is an unex- pected finding on biopsy of the kidneys, liver, heart, bowel, per- ipheral nerves, lymph nodes, skin, thyroid, or bone marrow during investigation of undiagnosed clinical problems. When amyloidosis is suspected clinically, biopsy of the rectum or subcutaneous fat is the least invasive. Amyloid is present at these sites in 50 to 70% of cases of systemic AA or AL amyloidosis. Alternatively, a clinically affected tissue may be biopsied directly. Congo red and other histochemical stains Many cotton dyes, fluorochromes, and metachromatic stains have been used to stain amyloid in tissue sections but Congo red staining, specifically using an alcoholic alkaline solution, and the re- sultant green birefringence when viewed with high-​intensity cross-​ polarized light, is the pathognomonic histochemical test. The stain is unstable and must be freshly prepared every 2 months or less and it is essential that the batch of Congo red has been validated. It is crit- ical to have a section thickness of 5 to 10 µm and to include in every staining run a positive control tissue containing modest amounts of amyloid. Reliable reporting on the microscopic appearances

section 12  Metabolic disorders 2230 absolutely depends on the use of a microscope of suitable quality equipped to provide high-​intensity cross-​polarized light. It is impos- sible to rigorously recognize or exclude amyloid unless these condi- tions are all fulfilled. Immunohistochemistry Although many amyloid fibril proteins can be identified immunohistochemically, the demonstration of amyloidogenic proteins in tissues does not, on its own, establish the presence of amyloid. Congo red staining and green birefringence are always required, and immunostaining may then enable the amyloid to be classified. Antibodies to serum amyloid A  protein are commer- cially available and always stain AA deposits, similarly with anti-​β2-​ microglobulin antisera and haemodialysis-​associated amyloid. In AL amyloid, the deposits are stainable with standard antisera to κ or λ immunoglobulin light chains in only about one-​half of cases, probably because the light-​chain fragment in the fibrils is usually the N-​terminal variable domain, which is largely unique for each mono- clonal protein. Immunohistochemical staining of transthyretin, Aβ, and prion protein amyloid may require pretreatment of sections with formic acid or alkaline guanidine, or deglycosylation. Electron microscopy Amyloid fibrils cannot always be convincingly identified ultra­ structurally, and electron microscopy alone is not sufficient to con- firm the diagnosis of amyloidosis. Problems of histological diagnosis The tissue sample must be adequate (e.g. the inclusion of sub- mucosal vessels in a rectal biopsy specimen), and failure to find amyloid does not exclude the diagnosis. The unavoidable sampling problem means that biopsy cannot reveal the extent or distribu- tion of amyloid. Experience with Congo red staining and viewing is required if clinically important false-​negative and false-​positive results are to be avoided. Immunohistochemical staining requires positive and negative controls, including demonstration of the spe- cificity of staining by absorption of positive antisera with isolated pure antigens. Proteomic analysis Laser capture microdissection of amyloid deposits from histological sections, followed by proteolytic digestion of the excised amyloid material and mass spectrometric identification of the fragments can also be used for typing. Suitable for use only in major centres of expertise, this highly specialized technique is an adjunct to immunohistochemistry for identification of amyloid fibril proteins, especially in the case of previously unknown fibrils, but appropriate interpretation of the results can be challenging. Nonhistological investigations Structural imaging Two-​dimensional echocardiography showing small, concentrically hypertrophied ventricles, generally impaired contraction, dilated atria, homogeneously echogenic valves, and ‘sparkling’ echodensity of ventricular walls is strongly suggestive of cardiac amyloidosis. However, clinically significant restrictive diastolic impairment may be difficult to detect, even by comprehensive Doppler echocardiog- raphy and other functional studies. Myocardial strain imaging has emerged as a useful additional echocardiographic technique in car- diac amyloidosis, with the ‘relative apical sparing’ pattern of longitu- dinal strain being a strong pointer to the diagnosis. Cardiac MRI provides functional and morphological informa- tion on cardiac amyloid, and yields more accurate measurements of volume, mass, and wall thickness than echocardiography, which may be of particular benefit in sequential assessments. An advan- tage of cardiac MRI is myocardial tissue characterization, which provides information on the presence and extent of amyloid de- posits. Characteristic patterns of global subendocardial and trans- mural late gadolinium enhancement have very high diagnostic sensitivity for cardiac amyloidosis and correlate with prognosis. T1 mapping studies performed after administration of gadolinium contrast can quantify the extracellular amyloid load and myocar- dial cell mass. Scintigraphy Scintigraphy following injection of technetium 99m-​labelled 3,3-​ diphosphono-​1,2 propanodicarboxylic acid (99mTc-​DPD), which was developed as a bone tracer, has lately been repurposed as a remarkably sensitive method for detecting the presence of cardiac transthyretin amyloid deposits. Localization of this tracer to the heart to a greater in- tensity than to the bones in the absence of a monoclonal gammopathy is pathognomonic for cardiac transthyretin amyloidosis. Cardiac localization of this tracer can also occur in some patients with AL, apolipoprotein A-​I, and other types of amyloidosis, but usually to only a minor degree. Genetic and biochemical tests In cases of known or suspected hereditary amyloidosis, the gene defect must be characterized. If amyloidotic tissue is available, the fibril protein may be known and the corresponding gene can then be studied, but if no tissue containing amyloid is available, screening of the genes for known amyloidogenic proteins must be undertaken. There are biochemical and immunochemical tests for screening the plasma for amyloidogenic variant protein products of mutant genes (e.g. for transthyretin and apolipoprotein A-​I variants), but molecular genetic analysis of DNA is easier to perform and is the most direct approach. However, regardless of the DNA results, it is desirable, if possible, to directly identify the respective protein in the amyloid. Serum amyloid P component as a specific tracer in amyloidosis The universal presence in amyloid deposits of amyloid P com- ponent, derived from circulating serum amyloid P component, is the basis for the use of radioisotope-​labelled serum amyloid P component as a diagnostic tracer in amyloidosis. No localization or retention of labelled serum amyloid P component occurs in healthy subjects or in patients with diseases other than amyloidosis (Fig. 12.12.3.1a). Radioiodinated serum amyloid P component has a short half-​life (24 h) in the plasma and is rapidly catabolized, with complete excretion of the iodinated breakdown products in the urine. However, in patients with systemic or localized extracerebral

12.12.3  Amyloidosis 2231 amyloidosis, the tracer rapidly and specifically localizes to the de- posits, in proportion to the quantity of amyloid present, and per- sists there without breakdown or modification (Figs. 12.12.3.1b and 12.12.3c). Highly purified serum amyloid P component, iso- lated from donor plasma according to pharmaceutical current good manufacturing practice, is oxidatively iodinated under conditions that preserve its function intact. The medium-​energy, short half-​ life, pure γ-​emitter 123I is used for scintigraphic imaging, and the long half-​life isotope 125I is used for metabolic studies. The dose of radioactivity administered (<4 mSv) is well within accepted safety limits and more than 30 000 studies have been completed without any adverse effects. In addition to high-​resolution scintigraphs, the uptake of tracer into various organs can be precisely quantified and, together with highly reproducible metabolic data on the plasma clearance and whole-​body retention of activity, the progression or regression of amyloid can be monitored serially and quantitatively. Dual modality three-​dimensional single-​photon emission com- puted tomography coupled with CT enables precise localization of 123I-​SAP to be determined when there is tracer uptake in unusual anatomical sites. Important observations regarding amyloid include the fol- lowing: the different distribution of amyloid in different forms of the disease; amyloid in anatomical sites not available for biopsy (adrenals, spleen); major systemic deposits of forms of amyloid previously thought to be organ-​limited; a poor correlation between the quantity of amyloid present in a given organ and the level of organ dysfunction; a nonhomogeneous distribution of amyloid within individual organs; and evidence for rapid progression and sometimes regression of amyloid deposits with different rates in different organs (Fig. 12.12.3.2). Examples of major regression of amyloidosis, when it has been possible to reduce or eliminate the supply of fibril precursor, are very encouraging. Studies with labelled serum amyloid P component thus make a valuable con- tribution to the diagnosis and management of patients with sys- temic amyloidosis, and in the United Kingdom these are routinely available for all known or suspected cases of amyloidosis in the National Health Service National Amyloidosis Centre at the Royal Free Hospital, London. Fig. 12.12.3.2  Serial posterior whole-​body 123I-​serum amyloid P component scintigraphs of a man with AL amyloidosis complicating benign monoclonal gammopathy. At presentation (scan 1) there was uptake in the spleen, liver, and bone marrow, obscuring any possible renal signal. Chemotherapy was given before scan 2, which shows increased spleen uptake, reduced liver uptake, and some renal uptake, but no change in total amyloid load determined by measurements of the clearance and retention of the tracer (not shown). Subsequently, he had recurrent splenic infarctions and splenectomy was performed. Thereafter (scan 3) there was increased tracer uptake in the liver, although a notably lower total amyloid load. Six months later (scan 4) liver and kidney uptake, plasma clearance, and whole-​body retention of tracer were all reduced, indicating regression of amyloid. He was clinically much improved and remained well. (a) (b) (c) Fig. 12.12.3.1  Whole-​body scintigraphs 24 h after intravenous injection of 123I-​labelled human serum amyloid P component. (a) Anterior view of a normal control subject showing the distribution of residual tracer in the blood pool and radioactive breakdown products in urine in the bladder; note the absence of localization or retention of tracer anywhere in the body. (b) Posterior (left) and anterior (right) views of a patient with juvenile chronic arthritis complicated by AA amyloidosis. There is uptake of tracer in the spleen, kidneys, and adrenal glands, a typical distribution of AA amyloid in which the spleen is involved in 100% of cases, kidneys in 75%, and adrenals in 40%. Note the reduced blood pool and bladder signal compared with (a). This patient, whose amyloid was diagnosed by renal biopsy 15 years before when nephrotic syndrome developed, and who was then treated with chlorambucil, had been in complete remission for 10 years during which there had been no acute phase response. At the time of this scan there was no biochemical abnormality in blood or urine, despite the very appreciable amyloid deposits, illustrating the discordance between the presence of amyloid and clinical effects. (c) Posterior (left) and anterior (right) views of a patient with monoclonal gammopathy complicated by extensive AL amyloidosis. There is uptake and retention of tracer in the liver, spleen, kidneys, bone marrow, and soft tissues around the shoulder. This scintigraphic pattern of amyloid distribution is pathognomonic for AL amyloidosis; bone-​marrow uptake has not been seen in any other type. Note the complete absence of blood pool or bladder signal resulting from complete uptake of the tracer dose into the substantial amyloid deposits.

section 12  Metabolic disorders 2232 Management of amyloidosis There have been substantial recent advances in the management of systemic amyloidosis, in particular active measures to support failing organ function while attempts are made to reduce the supply of the amyloid fibril precursor protein. Serial serum amyloid P com- ponent scintigraphy in more than 8000 patients with various forms of amyloidosis has shown that control of the primary disease pro- cess, or removal of the source of the amyloidogenic precursor, often results in regression of existing deposits and recovery or preserva- tion of organ function. This strongly supports aggressive interven- tion, and relatively toxic drug regimens or other radical approaches can be justified by the poor prognosis. Such an approach, leading to reduced morbidity and improved survival, was the basis for the es- tablishment of the National Health Service National Amyloidosis Centre. However, clinical improvement in amyloidosis is often ­delayed long after the underlying disorder has remitted, reflecting the very gradual regression of the deposits that is now recognized to occur in most patients. Continuing production of the amyloid precursor protein should be monitored as closely as possible in the long term, to determine the requirement for and intensity of treatment for the underlying primary condition. In AA amyloid- osis this involves frequent estimation of the plasma SAA level, and in AL amyloidosis it requires monitoring of the serum free light-​ chain concentration or other markers of the underlying monoclonal plasma cell proliferation. Specific conventional therapies AA amyloidosis The treatment of AA amyloidosis ranges from potent anti-​ inflammatory and immunosuppressive biological drugs in patients with rheumatoid arthritis, to lifelong prophylactic colchicine in fa- milial Mediterranean fever, and surgery in conditions such as re- fractory osteomyelitis and the tumours of Castleman’s disease. Most patients with AA amyloidosis complicating inflammatory arthritis can now be treated effectively with one or other of the many bio- logical agents now available (i.e. anticytokine (TNF, IL-​1, IL-​6) and anti-​CD20 antibodies etc.); IL-​1 inhibition with biological agents has revolutionized the management and prognosis of most patients with inherited periodic fever syndromes. Median SAA concentration has been shown to be a strong pre- dictor of both survival and renal outcome; sustained remission of inflammation in association with normal SAA values is associated with an almost 18-​fold lower risk of death than median SAA levels of greater than 155 mg/​L. Median survival of 79 to 137 months has been reported in a very large series. Approximately 40% of patients will eventually require renal replacement therapy, with a median time to dialysis from diagnosis of 78 months. AL amyloidosis Treatment of AL amyloidosis is based on that for myeloma, although the plasma cell dyscrasias in AL amyloidosis are often very subtle and tolerance of chemotherapy is frequently limited by amyloidotic organ dysfunction. Chemotherapy for AL amyloidosis therefore needs to be tailored to individual patients based on age, performance status, organ involvement, cardiac staging, and the known toxicities of individual drugs, balancing the ideal for complete remission of the clonal disease with the need to minimize treatment-​related mor- bidity and mortality. Quantitative measurements of serum free light chains using the robust, sensitive, Freelite immunoassay are usually the most ef- fective means for evaluating the early effects of chemotherapy as well as the requirement for ongoing treatment. Although partial haem- atological responses, that is, greater than 50% reduction in aberrant light chain concentration, can be beneficial, more complete clonal responses, that is, greater than 90% reduction, are strongly desir- able and are associated with the best clinical outcomes. Serial meas- urements of the cardiac biomarker serum NT-​proBNP are highly informative with respect to the response of cardiac amyloidosis to chemotherapy. Many different chemotherapy regimens are in current use, re- flecting the paucity of clinical trials and international consensus, ranging from relatively low-​intensity oral melphalan and dexametha- sone through various intermediate-​dose drug combinations to high-​ dose chemotherapy with autologous peripheral stem cell rescue in younger and fitter patients. In our current practice, most patients with AL amyloidosis receive a cyclic combination of the alkylating agent cyclophosphamide, the proteasome inhibitor bortezomib (Velcade) and dexamethasone, delivered 3-​ to 4-​weekly over 3 to 6 months. More recently developed proteasome inhibitors including carfilzomib and ixazomib are undergoing evaluation in AL amyl- oidosis. The immunomodulatory drugs, thalidomide, lenalidomide, and pomalidomide, are also widely used in combination with dexa- methasone and sometimes with proteasome inhibitor in addition. A multidisciplinary specialist approach is required to formulate and supervise optimal clinical management. Specific innovative therapies Transthyretin amyloidosis Several therapies have been undergoing advanced clinical testing in transthyretin amyloidosis and three have been approved by regu- latory agencies. While hepatic transplantation has been shown to reduce neurological disease progression in patients with early-​stage familial amyloid polyneuropathy associated with the methionine-​ 30 substitution, in whom cardiac amyloid deposits are absent, this approach is limited by donor liver availability, associated morbidity and mortality, and continued deposition of wild-​type transthyretin amyloid in the heart when cardiac amyloid is present at the time of surgery. Small-​molecule drugs New pharmacological therapies include the small-​molecule drugs diflunisal and tafamidis, which are bound in the plasma by transthyretin and postulated to inhibit its conversion into amyloid fibrils. The pivotal trial of tafamidis involved 441 pa- tients with wild-​type as well as the hereditary form of disease. In 264 patients who received the active drug, all-​cause mortality and cardiovascular-​related hospitalizations were reduced as com- pared with placebo (30% reduction in death over a period of 2½

12.12.3  Amyloidosis 2233 years); decline in functional capacity and quality of life was also decreased. Tafamidis (dose 20 mg daily) is approved for familial amyloid polyneuropathy by the European Medicines Agency and authorized by NHS England for treatment of transthyretin familial amyloid polyneuropathy in patients with stage 1 symptomatic polyneuropathy to delay peripheral neurological impairment and before liver transplantation. RNA interference therapy In 2018, two therapies—​independently developed by competing biopharmaceutical companies (Alnylam and Akcea)—​received marketing approval. These respectively use either an intravenously administered small interfering double-​stranded RNA or a sub- cutaneous antisense oligonucleotide that complements the mes- senger transthyretin RNA. By selectively blocking the cognate sense mRNA transcript of the TTR gene, hepatic synthesis of the unstable transthyretin protein, which misfolds to give rise to amyloid de- posits, is markedly inhibited. The Food and Drug Administration in the United States of America and the European Medicines Agency have approved both patisiran (Onpattro) and inotersen (Tegsedi), respectively, for treatment of the neuropathy caused by hereditary transthyretin amyloidosis. In the larger of the two phase III clinical trials, involving 255 patients, patisiran was shown to reverse neur- opathy in most; it also had salutary effects on a composite measure of life quality, reducing autonomic manifestations and improving activities of daily living. These studies are encouraging for patients with transthyretin amyloidosis but also serve as a landmark in the field of diseases caused by the proteins misfolding. Patisiran and inotersen need to be repeatedly administered, and while this mandates an at- tractive business model for companies developing high-​cost therapies for rare diseases, it is not clear that the high annual charges proposed and health gains achieved will be reimbursable worldwide. Access to the drugs may thus be limited for many af- fected patients. General measures The disabling arthralgia of β2-​microglobulin amyloidosis may partially respond to nonsteroidal anti-​inflammatory drugs or cor- ticosteroids, but even the most severe symptoms usually rapidly vanish following renal transplantation. The basis for this remark- able clinical response is unclear, since although transplantation rapidly restores normal β2-​microglobulin metabolism, regression of β2-​microglobulin amyloid may not be evident for many years. Supportive therapy remains critical in systemic amyloidosis, with the potential for delaying target organ failure, maintaining quality of life, and prolonging survival while the underlying pro- cess can be treated. Rigorous control of hypertension is vital in renal amyloidosis. Surgical resection of amyloidotic tissue is oc- casionally beneficial but, in general, a conservative approach to surgery, anaesthesia, and other invasive procedures is advisable. Should any such procedure be undertaken, meticulous attention to blood pressure and fluid balance is essential. Amyloidotic tis- sues may heal poorly and are liable to bleed. Diuretics and vaso- active drugs should be used cautiously in cardiac amyloidosis because they can reduce cardiac output substantially in the pa- tient with stiff ventricles. Dysrhythmias may respond to conven- tional pharmacological therapy or to insertion of pacemakers and implantable defibrillators. Replacement of vital organ function, notably by dialysis, may be necessary, and cardiac, renal, and liver transplantation procedures have a role in selected cases. Excellent results of renal transplantation in end-​stage renal failure have been obtained in patients with apolipoprotein A-​I amyloidosis, and in selected patients with systemic AA and AL amyloidosis; it probably also remains the treatment of choice for patients with hereditary fibrinogen α-​chain amyloidosis, although recurrent amyloid deposition of this type usually affects the transplant within 5 to 10 years. Future treatment possibilities As noted at the beginning of this chapter, most of the pathogen- esis of disease in amyloidosis is caused by the accumulation of the extracellular amyloid deposits which physically disrupt tissue architecture and thus function. The deposition of more amyloid is invariabl y associated with clinical deterioration but sufficient reduction of fibril precursor protein abundance halts amyloid ac- cumulation with slow regression of amyloid deposits in some pa- tients, stabilization of clinical status and potential improvement. However, therapeutic interventions to directly target amyloid deposits and promote their clearance from the tissues would be highly desirable. Proof of concept, with unprecedented, clinically beneficial, removal of visceral amyloid, has been demonstrated by antibody targeting of the serum amyloid P component present in all human amyloid deposits. FURTHER READING Adams D, et al. (2018). Patisiran, an RNAi therapeutic, for hereditary transthyretin amyloidosis. N Engl J Med, 379, 11–​21. Benson MD, et al. (2018). Inotersen treatment for patients with heredi- tary transthyretin amyloidosis. N Engl J Med, 379, 22–​31. Booth DR, et  al. (1997). Instability, unfolding and aggregation of human lysozyme variants underlying amyloid fibrillogenesis. Nature, 385, 787–​93. Gatt ME, Palladini G (2013). Light chain amyloidosis 2012: a new era. Br J Haematol, 160, 582–​98. Hawkins PN (2002). Serum amyloid P component scintigraphy for diagnosis and monitoring amyloidosis. Curr Opin Nephrol Hypertens, 11, 649–​55. Hawkins PN, Lavender JP, Pepys MB (1990). Evaluation of systemic amyloidosis by scintigraphy with 123I-​labeled serum amyloid P com- ponent. N Engl J Med, 323, 508–​13. Kumar S, et al. (2012). Revised prognostic staging system for light chain amyloidosis incorporating cardiac biomarkers and serum free light chain measurements. J Clin Oncol, 30, 989–​95. Kyle RA, Gertz MA (1995). Primary systemic amyloidosis: clinical and laboratory features in 474 cases. Semin Hematol, 32, 45–​9. Lachmann HJ, et al. (2002). Misdiagnosis of hereditary amyloidosis as AL (primary) amyloidosis. N Engl J Med, 346, 1786–​91. Lachmann HJ, et al. (2007). Natural history and outcome in systemic AA amyloidosis. N Engl J Med, 356, 2361–​71.

section 12  Metabolic disorders 2234 Larsen CP, et  al. (2010). Prevalence and morphology of leukocyte chemotactic factor 2-​associated amyloid in renal biopsies. Kidney Int, 77, 816–​19. Mangione, PP, et al. (2018). Plasminogen activation triggers transthyretin amyloidogenesis in vitro. J Biol Chem, 293(37), 14192–​9. Marcoux, J. et al. (2015). A novel mechano-enzymatic cleavage mech- anism underlies transthyretin amyloidogenesis. EMBO Mol Med,
7, 1337–49. Maurer MS, et  al. (2018). Tafamidis treatment for patients with transthyretin amyloid cardiomyopathy. N Engl J Med, 379, 1007–​16. Papa R, Lachmann HJ (2018). Secondary, AA, amyloidosis. Rheum Dis Clin North Am, 44, 585–​603. Pepys MB (2006). Amyloidosis. Annu Rev Med, 57, 223–​41. Richards DB, et  al. (2015). Therapeutic clearance of amyloid by antibodies to serum amyloid P component. N Engl J Med, 373, 1106–​14. Richards DB, et al. (2018). Repeat doses of antibody to serum amyloid P component clear amyloid deposits in patients with systemic amyl- oidosis. Sci Transl Med, 10(422), eaan3128. Sipe JD, et  al. (2014). Nomenclature 2014:  amyloid fibril pro- teins and clinical classification of the amyloidosis. Amyloid, 21,
221–​4. Valleix S, et  al. (2013). Hereditary systemic amyloidosis due to Asp76Asn variant β2-​microglobulin. N Engl J Med, 366, 2276–​83.

12.13 a1- Antitrypsin deficiency and the serpinopa

12.13 a1- Antitrypsin deficiency and the serpinopathies 2235

ESSENTIALS α1-​Antitrypsin is an acute phase glycoprotein synthesized by the liver that functions as an inhibitor of a range of proteolytic en- zymes, most importantly neutrophil elastase in the lung. Ninety-​five per cent of severe plasma deficiency of α1-​antitrypsin results from homozygosity for the Z allele (Glu342Lys), which causes the protein to undergo a conformational transition and form ordered polymers that are retained within hepatocytes as periodic acid–​Schiff-​positive, diastase-​resistant inclusions. Clinical features—​all adults homozygous for the Z allele of
α1-​antitrypsin have a minor degree of portal fibrosis that is often sub- clinical, but up to 50% have clinically evident cirrhosis and occasionally hepatocellular carcinoma. They also develop panlobular emphysema that typically affects the lung bases and is greatly exacerbated by smoking. Cor pulmonale and polycythaemia are late features. Diagnosis and management—​severe genetic deficiency of α1-​ antitrypsin is readily diagnosed by low plasma levels and the virtual absence of the α1-​band on protein electrophoresis. Patients should abstain from smoking and avoid agents that cause hepatic injury, such as excessive alcohol and obesity. Emphysema is treated along conventional lines. α1-​Antitrypsin replacement therapy is widely used in North America to slow the progression of the lung disease and has recently been licensed by the European Medicines Agency, but its clinical efficacy remains contentious and it has no effect on liver dis- ease. Clinical trials are underway to ‘knock down’ the expression of mutant Z α1-​antitrypsin within hepatocytes to try to prevent cirrhosis. Other serpinopathies—​the polymerization that underlies α1-​ antitrypsin deficiency is found in other members of the serine pro- tease inhibitor (or serpin) superfamily to cause diseases as diverse as thrombosis (antithrombin), angio-​oedema (C1 inhibitor), and de- mentia (neuroserpin). Introduction α1-​Antitrypsin deficiency was first described by Carl-​Bertil Laurell and Sten Eriksson in 1963 when they reported five individuals in whom there was a deficiency of the α1 band on serum protein electrophoresis. Three of the individuals had emphysema and one had a family history of emphysema. α1-​Antitrypsin is a 394-​amino acid, 52-​kDa acute phase glycoprotein synthesized by the liver, the lung and gut epithelial cells, neutrophils, and alveolar macro- phages. It is present in the plasma at a concentration of between 0.9 and 1.8 g/​litre and functions as an inhibitor of a range of proteolytic enzymes of which the most important is neutrophil elastase. α1-​ Antitrypsin deficiency results from point mutations that cause the protein to misfold and be retained within hepatocytes which in turn causes liver disease. The lack of circulating α1-​antitrypsin causes un- controlled tissue digestion within the lung and hence emphysema. Genetics and pathogenesis of disease Genetics α1-​Antitrypsin is subject to genetic variation resulting from mu- tations in the 12.2-​kb, 7-​exon SERPINA1 gene on the long arm of chromosome 14 (14q32.1) (OMIM 107400). Over 100 allelic vari- ants have been reported and classified using the PI (protease in- hibitor) nomenclature that assesses α1-​antitrypsin mobility in isoelectric focusing analysis. Normal α1-​antitrypsin migrates in the middle (M) and variants are designated A (anodal) to L if they migrate faster than M, and N to Z if they migrate more slowly. Many of these variants have been sequenced at the DNA level and shown to result from point mutations in the α1-​antitrypsin gene (Table 12.13.1). For example, the Z allele results from the substitu- tion of a positively charged lysine for a negative glutamic acid at pos- ition 342. The S allele results from the substitution of a neutral valine for a glutamic acid at position 264. Point mutations are inherited as a simple Mendelian trait; the normal genotype is designated PI MM or PI M, a heterozygote for the Z gene is PI MZ, and a homozygote is PI ZZ or PI Z. α1-​Antitrypsin alleles are codominantly expressed, with each allele contributing to the plasma level of protein. Therefore each of the deficiency alleles results in a characteristic decrease in the plasma concentration of α1-​antitrypsin; the S variant forms 60% of the normal M concentration and the Z variant 10 to 15%. Thus com- binations of alleles have predictable effects: the MZ heterozygote has an α1-​antitrypsin plasma level of 60% (50% from the normal M allele 12.13 α1-​Antitrypsin deficiency and the serpinopathies David A. Lomas

section 12  Metabolic disorders 2236 Table 12.13.1  Pathogenic alleles that cause α1-​antitrypsin deficiency. The letter represents the migration on isoelectric focusing and the name is typically the origin of the first reported individual Variant Mutation Molecular basis of disease Clinical features Epidemiology Deficiency alleles I Arg39Cys Protein misfolding; able to form heteropolymers with
Z α1-antitrypsin Reduced serum protein No clear disease association Disease only reported in compound heterozygotes King’s Hisp334Asp Rapid polymerization in hepatocyte endoplasmic reticulum, delayed secretion Neonatal jaundice. Presumed high risk of emphysema in homozygote/​ compound heterozygote Case report Mheerlen Pro369Leu Retained in the endoplasmic reticulum, none secreted High risk of emphysema in homozygotes/​compound heterozygotes. Unknown liver disease risk Case report Mmalton Δ52Phe (M2 variant) Intracellular degradation and polymerization; low serum concentration Well-​established association with liver disease and emphysema in homozygotes Most common rare deficiency allele in Sardinia; seen sporadically in the UK and Canada Mmineral springs Gly67Glu Abnormal post-​translational biosynthesis but no polymerization; low serum concentration Emphysema in homozygotes Unusual as described in a
Afro-​Caribbean individual in the United States Mnichinan Δ52Phe and Gly148Arg Intracellular polymerization in hepatocytes and plasma deficiency Risk of liver disease and emphysema Case report (Japanese family with consanguineous origin) Mpalermo Δ51Phe Serum deficiency High risk of emphysema in homozygotes Case report Mprocida Leu41Pro Unstable protein structure leading to intracellular degradation; reduced inhibitory activity of circulating protein High risk of emphysema in homozygotes Case report Mvall d’hebron (=Mwurzburg) Pro369Ser Retained in the endoplasmic reticulum, none secreted Presumed risk of emphysema in homozygotes/​compound heterozygotes; 50% normal serum
α1-​antitrypsin level in M/​vall d’hebron
(/​wurzburg) heterozygotes Case reports from Spain and Germany Mvarallo Δ41–​51, replaced with 22 bp sequence creating stop codon at 70–​71 Unknown intracellular defect Presumed risk of emphysema in homozygotes/​compound heterozygotes; 50% normal serum α1-​antitrypsin level in M/​Mvarallo heterozygote Case report Pittsburgh Met358 Arg Function altered to an antithrombin Fatal bleeding disorder Case report Plowell (=QO Cardiff) and Pduarte Asp256Val (M1 and M4 alleles respectively) Intracellular degradation and plasma deficiency Increased risk of emphysema in Z/​QO compound heterozygotes Case report S Glu264Val Protein misfolding and reduced secretion; able to form heteropolymers with Z α1-​antitrypsin Emphysema seen in SZ heterozygotes but less severe than in ZZ homozygotes. Cirrhosis reported in SZ heterozygotes Most common deficiency variant. Carrier frequency: 1:5 Northern Europe 1:30 USA 1:23 Australian Caucasian 1:26 New Zealand Caucasian Rare/​non-​existent in Asia, Africa and Australian Aboriginals Siiyama Ser53Phe Intracellular degradation and polymerization; low serum concentration Liver disease and emphysema in homozygotes Rare, but most common deficiency allele in Japan Wbethesda Ala336Thr Intracellular degradation, serum levels 50% normal Risk of liver disease and emphysema in compound heterozygotes Case report Ybarcelona Asp256Val and Pro391His Unknown intracellular defect; very low serum protein Severe emphysema reported in homozygote Case report Z Glu342 Lys Intracellular degradation and polymerization; low serum concentration Homozygotes: well-​established association with liver disease and emphysema. MZ heterozygotes may be more susceptible to airflow obstruction and chronic liver disease Commonest severe deficiency variant. Carrier frequency: 1:27 Northern Europe 1:83 USA 1:75 Australian Caucasian 1:46 New Zealand Caucasian Not seen in China, Japan, Korea, Malaysia, Northern and Western Africa

2237 12.13  α1-​Antitrypsin deficiency and the serpinopathies Variant Mutation Molecular basis of disease Clinical features Epidemiology Zausburg (=Ztun) Glu342 Lys (M2 variant) Intracellular degradation and polymerization; low serum concentration Liver disease and emphysema in homozygotes/​compound heterozygotes Case report Zwrexham Ser−19Leu and Glu342 Lys (Z mutation) Poor expression, low serum concentration Emphysema reported in compound Z/​ Zwrexham compound heterozygotes. Unclear whether Ser-​19 Leu would cause disease in absence of Z mutation Case report Null (QO) alleles QO Bellingham Lys217 stop codon No detectable α1-​antitrypsin mRNA High risk of emphysema in homozygotes/​compound heterozygotes Case report QO Bolton Δ1bpPro362 causing stop codon at 373 Truncated protein; intracellular degradation and no secreted protein High risk of emphysema in homozygotes/​compound heterozygotes Case report QO Cairo Lys259 stop codon Unknown intracellular defect High risk of emphysema in homozygotes/​compound heterozygotes Case report QO Clayton Pro362 insC causing stop codon at 376 Truncated protein; intracellular degradation and no secreted protein High risk of emphysema in homozygotes/​compound heterozygotes Case report QO Devon (=QO Newport) Gly115Ser and Glu342 Lys (Z mutation) Intracellular degradation and polymerization; reduced serum concentration Risk of emphysema and liver disease in compound heterozygotes. Unclear whether Gly115Ser would cause disease in absence of Z mutation Case report QO Granite Falls Δ1bpTyr160 causing stop codon No detectable α1-​antitrypsin mRNA Severe emphysema reported in Z compound heterozygote Case report QO Hong Kong Δ2bpLeu318 causing stop codon at 334 Truncated protein; intracellular aggregation (no polymerization), degradation and no secreted protein High risk of emphysema in homozygotes/​compound heterozygotes Case reports (individuals of Chinese descent) QO Isola di Procida Δ17 Kb inc. exons II–​V No detectable α1-​antitrypsin mRNA Emphysema reported in Mprocida compound heterozygote Case report QO Lisbon Thr68Ile Truncated protein; not secreted High risk of emphysema in homozygotes. 50% normal serum α1-​antitrypsin in M/​QO Lisbon heterozygotes Case report QO Ludwisghafen Ile92 Asn Disruption of tertiary structure; intracellular degradation and no detectable serum protein High risk of emphysema in homozygotes/​compound heterozygotes Case report QO Mattawa (M1allele)/​QO Ourém (M3 allele) Leu353Phe causing stop codon at 376 Truncated protein; misfolding and reduced serum levels Emphysema reported in homozygotes Case reports QO Riedenburg Whole gene deletion No gene expression High risk of emphysema in homozygotes/​compound heterozygotes Case report QO Saarbueken 1158dupC causing stop codon at 376 Truncated protein; not secreted High risk of emphysema in homozygotes. 50% normal serum α1-​antitrypsin in M/​QO Saarbueken heterozygotes Case report QO Trastevere Try194 stop codon Reduced mRNA, degradation of truncated protein; not secreted Emphysema reported in compound heterozygote Case report QO West ΔGly164 Lys191 Aberrant mRNA splicing, intracellular degradation and no detectable serum protein Emphysema reported in compound heterozygote Case report Reproduced from Dickens, J.A. and Lomas D.A. (2011) Why has it been so difficult to prove the efficacy of alpha-​1-​antitrypsin replacement therapy? Insights from the study of disease pathogenesis. Drug Des Devel Ther, 5, 391–​405 with permission. Table 12.13.1  Continued

section 12  Metabolic disorders 2238 and 10% from the Z allele), the MS heterozygote 80%, and the SZ heterozygote 40%. Rarely, point mutations can result in null alleles that express no functional α1-​antitrypsin and there is a case report of a dysfunctional α1-​antitrypsin that no longer inhibits neutrophil elastase but which can inhibit other serine proteases. The Pittsburgh mutant (Met358Arg) converted α1-​antitrypsin into an inhibitor of thrombin, thereby causing a fatal bleeding diathesis. The molecular basis of α1-​antitrypsin deficiency Liver disease α1-​Antitrypsin functions by presenting its reactive-​centre methio- nine residue on an exposed loop of the molecule such that it forms an ideal substrate for the enzyme neutrophil elastase (Fig. 12.13.1). The conformational transition that ensues results in the formation of a stable complex that inhibits the enzyme and allows it to be elim- inated from sites of inflammation. The Z mutation (Glu342Lys) results in normal translation of the gene, but 85% of the Z α1-​ antitrypsin is retained within the endoplasmic reticulum with only 10 to 15% entering the circulation. The Z mutation distorts the re- lationship between the loop and the A β-​pleated sheet that forms the major feature of the molecule. The consequent perturbation in structure allows the reactive-​centre loop of one α1-​molecule to lock into the A sheet of a second to form a dimer which then extends to form chains of loop-​sheet polymers (Fig. 12.13.1). The forma- tion of these polymers is temperature and concentration dependent and is localized to the endoplasmic reticulum of the hepatocyte (Fig. 12.13.2). These chains of polymers become interwoven to P D M* M Z Fig. 12.13.1  Mutant Z α1-​antitrypsin is retained within hepatocytes as polymers. The structure of α1-​antitrypsin is centred on β-​sheet A (green) and the mobile reactive-​centre loop (red). Polymer formation results from the Z variant of α1-​antitrypsin (Glu342Lys at P17; arrowed) or mutations in the shutter domain (blue circle) that open β-​sheet A to favour partial loop insertion and the formation of an unstable intermediate (M*). The patent β-​sheet A can accept the loop of another molecule to form a dimer (D) which then extends into polymers. From Gooptu, B., Hazes, B., Chang, W.-​S.W., Dafforn, T.R., Carrell, R.W., Read, R. & Lomas, D.A. (2000). Inactive conformation of the serpin
a1-​antichymotrypsin indicates two stage insertion of the reactive loop; implications for inhibitory function and conformational disease.
Proc. Natl. Acad. Sci (USA), 97, 67–​72, with permission. (a) (b) (c) Fig. 12.13.2  Z α1-​antitrypsin is retained within hepatocytes as intracellular inclusions. These inclusions are PAS positive and diastase resistant (a) and are associated with neonatal hepatitis and hepatocellular carcinoma. (b) Electron micrograph of a hepatocyte from the liver of a patient with Z α1-​antitrypsin deficiency shows the accumulation of α1-​antitrypsin within the rough endoplasmic reticulum (arrow). These inclusions are composed of chains of α1-​antitrypsin polymers (c). (b) and (c) Reproduced from (i) Lomas, D.A., Evans, D.L., Finch, J.T. & Carrell, R.W. (1992). The mechanism of Z α1-​antitrypsin accumulation in the liver. Nature, 357, 605–​ 607. Copyright © 1992, Springer Nature. (ii) Lomas, D.A., Finch, J.T., Seyama, K., Nukiwa, T. & Carrell, R.W. (1993). α1-​antitrypsin Siiyama (Ser53→Phe); further evidence for intracellular loop-​sheet polymerisation. J. Biol. Chem., 268, 15333–​15335, with permission.

2239 12.13  α1-​Antitrypsin deficiency and the serpinopathies form the insoluble intracellular aggregates that are the hallmark of α1-​antitrypsin liver disease. The process of intrahepatic poly- merization also underlies the severe plasma deficiency of the rare Siiyama (Ser53Phe), Mmalton (deletion of residue 52) and King’s (His334Asp) deficiency alleles and the mild plasma deficiency of the S (Glu264Val) and I (Arg39Cys) variants (Table 12.13.1). There is a strong genotype–​phenotype correlation that can be explained by the molecular instability caused by the mutation and in par- ticular the rate at which the mutant forms polymers. Those mutants that cause the most rapid polymerization cause the most retention of α1-​antitrypsin within the liver. This in turn correlates with the greatest risk of liver damage and cirrhosis, and the most severe plasma deficiency. Misfolded Z α1-​antitrypsin within hepatocytes is cleared by the proteosome but the ordered polymers are not de- tected by the unfolded protein response and are handled by less well understood pathways including autophagy. Lung disease The development of emphysema associated with α1-​antitrypsin deficiency is greatly accelerated by tobacco smoking. Emphysema results from uncontrolled enzymatic activity within the lung with those individuals with plasma α1-​antitrypsin levels of less than 40% of normal being most at risk. This is compounded by a fivefold reduc- tion in association rate kinetics with neutrophil elastase caused by the Z mutation and the polymerization of secreted Z α1-​antitrypsin within the airways and alveoli. The formation of polymers inacti- vates α1-​antitrypsin (thereby further reducing the protein avail- able to inhibit neutrophil elastase) and the polymers themselves are chemotactic for neutrophils and so may drive some of the excessive inflammation that characterizes the lung disease associated with α1-​ antitrypsin deficiency. Epidemiology Two point mutations have been shown to explain the vast ma- jority of cases of α1-​antitrypsin deficiency. The Z allele causes the most severe plasma deficiency and is most prevalent in southern Scandinavia and the north-​western European seaboard where 4% of the population are MZ heterozygotes and 1 in 1700 are PI Z homozygotes. The gene frequency of the Z allele reduces towards the south and east of Europe. In contrast, the S allele causes only mild plasma deficiency and is most common in southern Europe where up to 28% of the population are MS heterozygotes. The S allele becomes less frequent as one moves north-​east. The frequencies of the Z allele in the United States of America are similar to the lowest frequencies in Europe but the S allele is more common than in nor- thern Europeans. α1-​Antitrypsin deficiency is infrequent in Asian, African, and Middle Eastern populations. It is also rare in Japan, but when present it is usually due to the Siiyama mutation (Ser53Phe). In the genetically isolated island of Sardinia, the commonest cause of severe α1-​antitrypsin deficiency is the Mmalton mutation (deletion of residue 52). The Z allele is believed to have arisen from a single origin 66 generations or 2000  years ago. The high frequency in southern Scandinavia suggests that the mutation arose in the Viking popula- tion. The date of origin implies that the allele arose when the Vikings populated mid/​northern Europe and before their migration to Scandinavia. It is likely that the Z allele of α1-​antitrypsin was then distributed across northern Europe by the Viking raiders between 800 and 1100 ad, and then to the United States of America and the rest of the world during migration over the past 200 years. The S al- lele appears to have arisen in the north of the Iberian peninsula, but the date of origin is uncertain. This mutation was similarly intro- duced into North America by mass migration. Clinical features α1-​Antitrypsin deficiency and liver disease The accumulation of abnormal protein starts in utero and is char- acterized by periodic acid–​Schiff (PAS)-​positive, diastase-​resistant inclusions of α1-​antitrypsin in the periportal cells (Fig. 12.13.2). Seventy-​three per cent of Z α1-​antitrypsin homozygote infants have a raised serum alanine aminotransferase in the first year of life but in only 15% of people is it still abnormal by 12 years of age. Similarly serum bilirubin is raised in 11% of PI Z infants in the first 2–​4 months but falls to normal by 6 months of age. One in ten in- fants develops cholestatic jaundice and 6% develop clinical evidence of liver disease without jaundice. These symptoms usually resolve by the second year of life but approximately 15% of patients with cholestatic jaundice progress to juvenile cirrhosis. The overall risk of death from liver disease in PI Z children during childhood is 2 to 3%, with boys being at more risk than girls. All adults with the Z allele of α1-​antitrypsin have slowly progressive hepatic damage that is often subclinical and only evident as a minor degree of portal fi- brosis. However, up to 50% of Z α1-​antitrypsin homozygotes present with clinically evident cirrhosis and occasionally with hepatocellular carcinoma. α1-​Antitrypsin deficiency and emphysema Patients with emphysema related to α1-​antitrypsin deficiency usu- ally present with increasing dyspnoea with cor pulmonale and poly- cythaemia occurring late in the course of the disease. Emphysema associated with Z α1-​antitrypsin deficiency differs from ‘usual chronic obstructive pulmonary disease (COPD)’ with normal levels of M α1-​antitrypsin in that it affects predominantly the bases rather than the apices of the lungs, it is associated with panlobular rather than centrilobular disease, and it results from the expression of different genes when assessed by microarray analysis. However, in many cases the distribution of disease is indistinguishable from ‘usual COPD’. High-​resolution CT scans are the most accurate method of assessing the distribution of panlobular emphysema and for monitoring the progress of the pulmonary disease, although this currently has little value outside clinical trials. Lung function tests are typical for emphysema with a reduced FEV1/​FVC ratio (forced expiratory volume in 1 s/​forced vital capacity) and FEV1, gas trap- ping (raised residual volume/​total lung capacity ratio), and a low gas-​transfer factor. Partial reversibility of airflow obstruction (as defined by an increase of 12% and 200 ml in FEV1) is common in individuals with COPD secondary to α1-​antitrypsin deficiency as it is in many individuals with COPD. The most important factor in the development and progression of emphysema in α1-​antitrypsin deficiency is tobacco smoking.

section 12  Metabolic disorders 2240 Other conditions associated with α1-​antitrypsin deficiency α1-​Antitrypsin deficiency is associated with an increased prevalence of asthma, panniculitis, granulomatosis with polyangiitis, and pos- sibly pancreatitis, gallstones, bronchiectasis, and intracranial and intra-​abdominal aneurysms. There appears to be a reduced risk of cardiovascular disease. Clinical investigation The severe genetic deficiency of α1-​antitrypsin is readily diagnosed by low plasma levels and the virtual absence of the α1 band on protein electrophoresis. As α1-​antitrypsin is an acute phase protein, most laboratories will report levels with another acute phase reactant, such as α1-​antitchymotrypsin or C-​reactive protein, which allows an assessment of the likelihood of deficiency in the context of the inflammatory response. The acute phase response raises the plasma level of α1-​antitrypsin, but the plasma level of the PI Z homozygote can never reach the normal range. The deficiency variant is then as- signed a PI phenotype according to the migration of the protein on an isoelectric focusing gel. The mutation underlying the deficiency can be determined by sequencing the SERPINA1 gene. Commercial kits permit detection of the Z and S alleles but will not detect null or other rare alleles. Treatment The treatment of α1-​antitrypsin deficiency depends largely on the avoidance of stimuli causing repeated pulmonary inflammation—​ primarily smoking. Patients with α1-​antitrypsin deficiency-​related emphysema should receive conventional therapy with trials of bron- chodilators and inhaled corticosteroids, pulmonary rehabilitation, and, where appropriate, assessment for long-​term oxygen therapy and lung transplantation. The role of lung volume-​reduction surgery in this group is less clear as the disease is basal rather than apical and resections of this region are technically more difficult. Any benefits are shorter lasting than in individuals with COPD and normal levels of α1-​antitrypsin. The lung disease results from a deficiency in the anti-​elastase screen. This may be rectified biochemically by intravenous infu- sions of α1-​antitrypsin. Registry data suggest that individuals with α1-​antitrypsin deficiency and an FEV1 of 35 to 49% predicted may derive benefit from replacement therapy. One controlled trial has shown reduced progression of CT markers of emphysema in in- dividuals receiving intravenous α1-​antitrypsin, but none has been powered to detect an effect on patient-​reported outcomes or mor- tality. α1-​Antitrypsin replacement therapy is widely used in North America and has recently been licensed by the European Medicines Agency but its clinical efficacy remains contentious. All Z homozygotes have some liver damage and, as such, would be wise to avoid alcohol abuse and obesity. PI Z homozygotes should be monitored for the persistence of hyperbilirubinaemia as this, along with deteriorating results of coagulation studies, indicates the need for liver transplantation. Clinical trials are underway to ‘knock down’ the expression of mutant Z α1-​antitrypsin within hepatocytes to prevent the protein overload that causes cirrhosis. Parents with a child with severe Z α1-​antitrypsin liver disease may require genetic counselling. The likelihood of similar severe liver damage in a subsequent Z homozygote sibling is approximately 20%. The uncommon α1-​antitrypsin deficiency-​associated panniculitis usually responds to dapsone, 100 to 150 mg daily, for 2 to 4 weeks, but occasionally it necessitates the administration of intravenous α1-​ antitrypsin replacement therapy. Prognosis Estimates of the annual rate of decline in FEV1 range from 41 to 109 ml in individuals with α1-​antitrypsin deficiency although one study reported a rate of decline of 316 ml/​year in current smokers. The fastest rate of decline is in current smokers (and, to a lesser extent, ex-​smokers), men, individuals aged 30–​44 years, those with FEV1 values between 35 and 79% predicted, and those with a broncho- dilator response. Respiratory failure accounts for 50 to 72% of deaths in individuals with α1-​antitrypsin deficiency with the second most common cause of death being liver cirrhosis (10–​13%). Most chil- dren avoid significant liver damage in childhood but are still at risk of disease in adult life. The factors that predict progressive liver dis- ease are unclear but males and the obese appear to be most at risk. The only significant cohort study has followed 184 individuals with α1-​antitrypsin deficiency (127 PI Z, 2 PI Znull, 54 PI SZ, 1 PI Snull) from birth to 34 years of age. One PI SZ and five PI Z children died in early childhood (two of liver disease and two of other causes but were found to have histological signs of cirrhosis or fibrosis at post- mortem) and 12% and 6% of PI Z subjects had abnormal liver func- tion tests at 18 and 26 years respectively but no clinical evidence of liver disease. All the 34-​year-​olds had normal liver and lung function (including the 14% of individuals who were current or ex-​smokers) but smoking frequency was significantly lower among individuals with α1-​antitrypsin deficiency than in the controls. A logical follow-​on from the association of α1-​antitrypsin defi- ciency with emphysema is an assessment of the risk of COPD in heterozygotes who carry an abnormal Z allele and a normal M allele. These individuals have plasma α1-​antitrypsin levels that are approxi- mately 60% of normal. A population-​based study demonstrated that PI MZ α1-​antitrypsin heterozygotes do not have a clearly increased risk of lung damage. However, if groups of patients are collected who already have COPD, then the prevalence of PI MZ individuals ap- pears to be elevated. In addition, a longitudinal study has demon- strated that among COPD patients, the PI MZ heterozygotes have a more rapid decline in lung function. These data suggest that either all PI MZ α1-​antitrypsin individuals are at slightly increased risk for the development of COPD, or that a subset of the PI MZ α1-​ antitrypsin subjects are at substantially increased risk of pulmonary damage if they smoke. Other ‘serpinopathies’ α1-​Antitrypsin is the archetypal member of a superfamily of proteins termed the serine protease inhibitors, or serpins, that have closely related structures and functions. These inhibitors control various in- flammatory cascades, including coagulation (antithrombin), com- plement activation (C1-​inhibitor), and fibrinolysis (α2-​antiplasmin).

2241 12.13  α1-​Antitrypsin deficiency and the serpinopathies Pathological processes that underlie the deficiency of one member may account for deficiency of others. Indeed, the process of polymer formation has also been reported in deficiency mutants of antithrombin, C1-​inhibitor, α1-​antichymotrypsin, and heparin co- factor II. These polymers are inactive as proteinase inhibitors and so predispose the individual to thrombosis (antithrombin) and angio-​ oedema (C1-​inhibitor). The plasma deficiency that results from the polymerization of mutants of α1-​antichymotrypsin has been asso- ciated with COPD in some (but not all) association studies, but the plasma deficiency of heparin cofactor II has yet to be associated with a clinical phenotype. Perhaps the most striking serpinopathy results from the polymerization of mutants of a neuron-​specific serpin, neuroserpin, to cause the novel inclusion-​body dementia known as familial encephalopathy with neuroserpin inclusion bodies (FENIB; OMIM 604218). This is inherited as an autosomal dominant trait with the inclusions of neuroserpin in the brain being PAS positive and diastase resistant, identical to those of Z α1-​antitrypsin in the liver. The six mutations that have been described show a striking in- verse correlation between the rate that the protein forms polymers and the age of onset/​severity of the dementia. New and emerging treatments Other treatments at earlier stages of development include gene and stem cell therapy and chemical chaperones. Vectors carrying the α1-​antitrypsin gene have been targeted to liver, lung, and muscle in animals. There is good expression of α1-​antitrypsin but further data are required to assess whether this can be achieved in humans. In particular, it is important to determine the length of time of protein expression and whether the levels of α1-​antitrypsin in the epithelial lining fluid of the lung are sufficient to prevent ongoing proteolytic damage. Genomic correction of fibroblast-​derived, induced pluri- potent stem cells provides a novel strategy to generate ‘corrected hepatocytes’ from individuals with α1-​antitrypsin deficiency, but further development is required before these can be used as ‘hep- atocyte replacement therapy’ in humans. The antiepileptic drug carbamazepine increases autophagy and so promotes the degrad- ation of Z α1-​antitrypsin in cell lines and mouse models of disease. Clinical trials are underway to evaluate the efficacy of this agent in α1-​antitrypsin deficiency-​related liver disease in humans. The long-​ term aim is to exploit our understanding of the pathogenesis of α1-​antitrypsin deficiency to develop small molecules to block poly- merization and so treat the associated liver and lung disease. FURTHER READING Chapman KR, et al. (2015). Intravenous augmentation treatment and lung density in severe α1-​antitrypsin deficiency (RAPID):  a ran- domised, double-​blind, placebo-​controlled trial. Lancet, 386, 360–​8. Davis RL, et al. (2002). Association between conformational muta- tions in neuroserpin and onset and severity of dementia. Lancet, 359, 2242–​7. Eriksson S, Carlson J, Velez R (1986). Risk of cirrhosis and primary liver cancer in alpha1-​antitrypsin deficiency. N Engl J Med, 314, 736–​9. Gooptu B, Lomas DA (2009). Conformational pathology of the serpins—​themes, variations and therapeutic strategies. Ann Rev Biochem, 78, 147–​76. Hidvegi T, et  al. (2010). An autophagy-​enhancing drug promotes degradation of mutant alpha1-​antitrypsin Z and reduces hepatic fi- brosis. Science, 329, 229–​32. Laurell C-​B, Eriksson S (1963). The electrophoretic α1-​globulin pat- tern of serum in α1-​antitrypsin deficiency. Scand J Clin Lab Invest, 15, 132–​40. Lomas DA (2006). The selective advantage of α1-​antitrypsin deficiency. Am J Resp Crit Care Med, 173, 1072–​7. Mahadeva R, et al. (2005). Polymers of Z α1-​antitrypsin co-​localise with neutrophils in emphysematous alveoli and are chemotactic in vivo. Am J Pathol, 166, 377–​86. Owen MC, et al. (1983). Mutation of antitrypsin to antithrombin.α1-​ antitrypsin Pittsburgh (358 Met to Arg), a fatal bleeding disorder. N Engl J Med, 309, 694–​8. Tanash HA, et  al. (2015). The Swedish α1-​antitrypsin Screening Study: health status and lung and liver function at age 34. Ann Am Thorac Soc, 12, 807–​12. Sveger T (1976). Liver disease in alpha1-​antitrypsin deficiency de- tected by screening of 200,000 infants. N Engl J Med, 294, 1316–​21. Yusa K (2011). Targeted gene correction of α1-​antitrypsin deficiency in induced pluripotent stem cells. Nature, 478, 391–​4.

SECTION 13 Endocrine disorders Section editor: Mark Gurnell 13.1 Principles of hormone action  2245 Rob Fowkes, V. Krishna Chatterjee, and Mark Gurnell 13.2 Pituitary disorders  2258 13.2.1 Disorders of the anterior pituitary gland  2258 Niki Karavitaki and John A.H. Wass 13.2.2 Disorders of the posterior pituitary gland  2277 Niki Karavitaki, Shahzada K. Ahmed,
and John A.H. Wass 13.3 Thyroid disorders  2284 13.3.1 The thyroid gland and disorders of thyroid function  2284 Anthony P. Weetman and Kristien Boelaert 13.3.2 Thyroid cancer  2302 Kristien Boelaert and Anthony P. Weetman 13.4 Parathyroid disorders and diseases altering calcium metabolism  2313 R.V. Thakker 13.5 Adrenal disorders  2331 13.5.1 Disorders of the adrenal cortex  2331 Mark Sherlock and Mark Gurnell 13.5.2 Congenital adrenal hyperplasia  2360 Nils P. Krone and Ieuan A. Hughes 13.6 Reproductive disorders  2374 13.6.1 Ovarian disorders  2374 Stephen Franks, Kate Hardy, and Lisa J. Webber 13.6.2 Disorders of male reproduction and male hypogonadism  2386 P.-​M.G. Bouloux 13.6.3 Benign breast disease  2406 Gael M. MacLean 13.6.4 Sexual dysfunction  2408 Ian Eardley 13.7 Disorders of growth and development  2416 13.7.1 Normal growth and its disorders  2416 Gary Butler 13.7.2 Normal puberty and its disorders  2428 Fiona Ryan and Sejal Patel 13.7.3 Normal and abnormal sexual
differentiation  2435 S. Faisal Ahmed and Angela K. Lucas-​Herald 13.8 Pancreatic endocrine disorders and multiple endocrine neoplasia  2449 B. Khoo, T.M. Tan, and S.R. Bloom 13.9 Diabetes and hypoglycaemia  2464 13.9.1 Diabetes  2464 Colin Dayan and Julia Platts 13.9.2 Hypoglycaemia  2531 Mark Evans and Ben Challis 13.10 Hormonal manifestations of non​endocrine disease  2541 Thomas M. Barber and John A.H. Wass 13.11 The pineal gland and melatonin  2553 J. Arendt and Timothy M. Cox

12.2 Protein- dependent inborn errors of metabolis

12.2 Protein- dependent inborn errors of metabolism 1942

ESSENTIALS Protein-​dependent inborn errors of metabolism are caused by in- herited enzyme defects of catabolic pathways or intracellular trans- port of amino acids. Most result in an accumulation of metabolites upstream of the defective enzyme (amino acids and/​or ammonia), causing intoxication. Protein-​dependent metabolic diseases usually have a low preva- lence except for some high-​risk communities with high consan- guinity rates. However, the cumulative prevalence of these disorders is considerable (i.e. at least >1:2000 newborns) and represents an important challenge for all public health systems. Types of protein-​dependent inborn errors of metabolism Amino acid disorders—​enzyme deficiencies in the proximal part of amino acid catabolism result in accumulation of precursor amino acids which are detectable by ninhydrin (a chemical used to detect ammonia or primary and secondary amines) and thus are called amino acid disorders. Phenylketonuria (PKU) is the most frequent such condition in white people. Organic acid disorders—​distal enzyme defects of amino acid deg- radation result in pathological accumulation of organic acids but not the precursor amino acid. These disorders became detectable after the introduction of gas chromatography–​mass spectrometry and are called organic acid disorders. Urea cycle defects—​breakdown of amino acids results in the re- lease of ammonia that is detoxified by the urea cycle, which is com- posed of five catalytic enzymes, a cofactor producer, and at least two transport proteins. The biochemical hallmark of urea cycle defects is hyperammonaemia. Understanding of the protein-​dependent inborn errors is based on the observation that some pathological metabolites impair key intracellular functions, such as energy metabolism, and thus when elevated may become toxic. These metabolites are excreted by urine or following conjugation to L-​carnitine or L-​glycine. However, in some diseases, such as disorders of tetrahydrobiopterin metabolism, clinical symptoms result from inadequate production of essential metabolites, such as the monoaminergic neurotransmitters. Clinical presentation Children with inherited disorders of amino acid, organic acid, or the urea cycle are usually born at term after an uneventful pregnancy and are initially asymptomatic. The onset of the first symptoms is varied, ranging from neonatal metabolic decompensation to onset of symp- toms during adulthood. Irreversible organ damage and/​or early death often follow if the diagnosis is delayed or missed. Metabolic decompensations in childhood are triggered by excess intake of pro- tein and—​most importantly—​secondary to breakdown of body pro- tein during episodes that induce catabolism. Family history—​if carefully taken, this may reveal important clues to the diagnosis of protein-​dependent inborn metabolic errors. Most disorders are inherited as autosomal recessive traits, which may be suspected if the parents are consanguineous or the family has a confined ethnic or geographic background. Carriers for particular disorders and affected children may be more frequent in certain communities (e.g. Amish), ethnic groups (e.g. Ashkenazi Jews, Arabic tribes), or countries that have seen little immigration over many cen- turies (e.g. Finland). Specialist investigations are often started only after a second affected child is born into a family: older siblings may be found to suffer from a similar disorder as the index patient or have died from an acute unexplained disease. Disease spectrum—​this is broad, but follows a distinct pattern in specific disorders, for instance: (1) untreated patients with classical PKU and cerebral organic acid disorders characteristically present with neurological symptoms. (2) Acute life-​threatening decompen- sation is common in classical organic acid, urea cycle defects, and maple syrup urine disorder; the young infant vomits or refuses to feed and then deteriorates rapidly. (3) Asymptomatic protein-​dependent inborn metabolic errors are rare, but there are a few known enzyme defects, such as histidinaemia, which do not produce disease. Investigation and management Every infant presenting with symptoms of unexplained metabolic crisis, intoxication, or encephalopathy requires urgent evaluation of metabolic parameters, including analyses of arterial blood gases, serum glucose and lactate, plasma ammonia and amino acids, acylcarnitine profiling in dried blood spots, and organic acid analysis in urine. Acute emergency therapy—​basic principles are to (1)  suppress muscle and liver protein catabolism and ensure a glucose supply above the basal metabolic demand; (2) treat any precipitating illness; (3) reduce increased production of toxic metabolites by reduction or omission of natural protein; (4) enhance detoxifying mechanisms 12.2 Protein-​dependent inborn errors of metabolism Georg F. Hoffmann and Stefan Kölker

12.2  Protein-dependent inborn errors of metabolism 1943 and urinary excretion of pathological metabolites; (5)  aggressively treat dehydration and acidosis; (6) prevent secondary carnitine de- pletion; and (7) provide alternative routes of ammonia disposal in hyperammonaemia. Long-​term treatment—​this aims principally to mitigate the meta- bolic consequences of enzyme deficiencies by compensating for them, including: (1) reduction of toxic metabolites by dietary restric- tion of precursor amino acids, prevention of catabolism, stimulation of residual enzyme activity (e.g. with cofactors), and detoxification strategies; and (2)  substitution with depleted substrates, such as biotin, cobalamin, or L-​dopa. However, efficacy is often low in pa- tients in whom diagnosis is made after the onset of symptoms, hence newborn screening programmes have been introduced in many countries, the criteria for implementation of which include: (1) reli- able presymptomatic disease detection, (2) treatability of the disease, and (3) starting of treatment in presymptomatic children. Successful treatment of affected individuals is often difficult to achieve. Careful supervision in metabolic centres involving an ex- perienced multidisciplinary team is invaluable for the best outcome. Treatment is time-​ and cost-​intensive, often lifelong, and mostly per- formed at home, hence regular training and support of patients and their families is essential to prevent irreversible complications. All patients should carry an emergency card that gives details of their condition and relevant contact numbers. Parent and patient organ- izations can offer useful support. Detailed description of individual disorders is to be found in the text of this chapter, and further information on diagnosis, genetic testing, treatment and follow-​up is available from several online databases (see ‘Further reading’). Introduction Humans depend on dietary protein as a source of amino acids; they are the metabolic basis of all functional and structural proteins in the body. Some amino acids—​termed essential—​cannot be synthesized by the human body, such l-​isoleucine and l-​phenylalanine. Renal conservation of amino acids is extremely effective, with clearance values mostly less than 1%. Stool nitrogen losses are about 1 g/​day and are mostly of bacterial origin. In contrast to glucose and fatty acids, amino acids taken in excess of requirement cannot be stored but are used for energy. The initial step of degradation is the removal of the amino group. Ammonia enters the urea cycle for conversion to urea. The remaining carbon skeletons are degraded via multistep individual pathways to central metabolic intermediates such as acetyl coenzyme A (CoA) or tricarb- oxylic acid cycle intermediates. Some enzymes require coenzymes, and inherited disease may be due to defects of the apoenzymes or their vitamin coenzymes, for example, biotin, pyridoxine (vitamin B6), or cobalamin (vitamin B12). Amino acids can be specifically detected by the ninhydrin re- action, which became available in the late 1940s, resulting in the identification of disorders such as phenylketonuria (PKU) or maple syrup urine disease. Breakdown of many amino acids occurs mostly intramitochondrially through degradation of CoA-​activated car- bonic acids, the so-​called acyl-​CoA compounds. These nonamino organic acids are not detectable by amino acid analysis. Since defects of the latter phases of amino acid degradation induce accumulation of organic acids but not amino acid precursors, these disorders be- came detectable after the introduction of gas chromatography, es- pecially gas chromatography–​mass spectrometry (GC/​MS) in the 1960s and 1970s and have been termed organic acid disorders. Thus the terminology amino acid and organic acid disorders is not based on pathophysiological differences but simply on the different analyt- ical approaches. In this chapter, amino acid disorders, urea cycle defects, and or- ganic acid disorders are described; defects in mitochondrial me- tabolism and amino acid transport in the kidney tubule and small intestine are not considered. Historical perspective In 1902, Archibald Garrod introduced the term ‘inborn errors of metabolism’. An extraordinary scientist and paediatrician, he used consanguinity and distribution of cases in families to introduce the hypothesis that autosomal recessive inheritance according to Mendel’s rediscovered laws would explain the occurrence of the alkaptonuria phenotype, a defect in tyrosine degradation. Soon afterwards he also recognized albinism, cystinuria, and pentosuria as inborn errors. Metabolic medicine is closely linked with advances in laboratory techniques. The use of paper chromatography by Bickel and Dent and of automated column chromatography by Moore and Stein opened the field of amino acid disorders. In the late 1960s, Tanaka discovered isovaleric aciduria by GC/​MS, and this was followed by the identification of numerous organic acid disorders. More re- cently, the rise of molecular biology has revolutionized the field, and now tandem mass spectroscopy and next-​generation sequencing are proving powerful tools in screening and diagnosis. Monogenic defects have been identified for almost every known enzymatic step of protein metabolism. Often it was the discovery of patients with enzyme defects which unravelled individual steps in human metabolism. Until the early 1950s, no treatment of any genetic disorder ex- isted; destiny would take its course, and genetic counselling about recurrence risks was all that could be offered. That changed when, in 1953, Bickel showed that PKU can be successfully treated and that early diagnosis and dietary treatment change the outcome from severe learning difficulties to normal psychosocial develop- ment. Subsequently, many other metabolic diseases became man- ageable in a similar way using the substrate deprivation strategy. Pharmacological doses of vitamins proved useful in defects of cobalamin and biotin metabolism, homocystinuria, and others. Simultaneously with the perception that identification of chil- dren before the onset of clinical symptoms is indispensable to im- prove the outcome, reliable and cheap screening methods have been developed. In the United States of America, Guthrie set the cornerstone for newborn screening by developing a bacterial in- hibition assay to detect PKU. Despite early disagreement and re- sistance by the medical profession, newborn screening has proven its worth over the years and the test is still called the ‘Guthrie test’ worldwide. In 1999, the World Health Organization announced orphan dis- eases as a major future health challenge. Among these diseases,

SECTION 12  Metabolic disorders 1944 disorders of amino acid and organic acid metabolism are especially important because of their cumulative prevalence (>1:2000 new- borns) and because successful therapy is available for most of them. Inborn metabolic diseases have become a significant challenge for healthcare systems, particularly in countries where infectious dis- eases and other perinatal problems are receding in importance. Aetiology, genetics, pathogenesis, and pathology The clinical manifestations of most protein-​dependent inborn errors are thought to result from toxicity of the accumulating key metabolites to specific organs inducing selective or multiple organ failure. This ‘toxic metabolite hypothesis’ has influenced research and allowed the development of effective treatment. Despite increasing knowledge of pathophysiology, the most rele- vant concepts are derived from clinical research. For example, de- fects of all six enzymes in the degradation pathway of phenylalanine and tyrosine are known by now (see also ‘Defects of phenylalanine and tyrosine metabolism’). Defects in the first enzyme, phenyl- alanine hydroxylase, cause PKU (learning difficulties, seizures, ataxia, paresis, behavioural problems) and deficiency of tyrosine aminotransferase, the next enzyme, induces tyrosinaemia type II (corneal erosions, painful hyperkeratotic lesions, behavioural prob- lems). A  defect of 4-​hydroxyphenylpyruvate dioxygenase is the cause of tyrosinaemia type III which is possibly a nondisease, only a few patients develop neurological manifestations. A block in the next step of the pathway, 4-​hydroxyphenylpyruvate dioxygenase, results in alkaptonuria (ochronosis, arthritis, heart disease), whereas deficiency of the last enzyme, fumarylacetoacetase, pro- duces a disease deadly in early childhood, tyrosinaemia type I, presenting with failure to thrive, liver failure, hepatosplenomegaly, hepatocarcinoma, and porphyria-​like crises. The distinct syndromes resulting from defective breakdown of aromatic amino acids could never have been inferred simply by biochemical exploration of the metabolic pathway. Epidemiology As a group, protein-​dependent disorders are by far the most common, acutely life-​threatening inborn errors of metabolism (esti- mated prevalence >1:2000 newborns). However, reliable epidemio- logical data are scarce as all reports suggest a significant portion of patients who evade diagnosis and are considered to have neonatal sepsis or sudden infant death syndrome. All disorders cannot be re- liably ascertained clinically, and until recently population neonatal screening has only been implemented for PKU. Most epidemio- logical data are available from European countries, Japan, and the United States of America, highlighting variations based on ethnic background, migrations, and/​or genetic isolation. In a few com- munities, the prevalence of individual disorders may increase up to five times the cumulative prevalence of amino acid and organic acid disorders in European countries, Japan, and the United States of America. For example, glutaric aciduria type I is found in up to 1 in 300 newborns in the Amish Community (United States of America) and the Oji-​Cree First Nations (Canada), and in Qatar the preva- lence of classic homocystinuria is 1 in 600 newborns. Prevention With the first successful treatment of a young girl with PKU, the need for timely diagnosis and implementation of treatment became imperative. In most inborn errors, affected neonates are completely asymptomatic and onset of irreversible symptoms during infancy and childhood can often be prevented if treatment is started while the child is asymptomatic. Since inborn errors of metabolism are rare, only neonatal mass screening can guarantee timely detection. However, which diseases are the most appropriate for screening re- mains debatable. The criteria of Wilson and Jungner (1968) for an implementation to newborn screening include: (1) reliable disease detection in a presymptomatic state of the disease, (2) treatability of the disease, and (3) the start of treatment in the presymptomatic chil- dren. In the 1960s, these criteria were achieved for PKU screening, which developed into one of the most important programmes of preventive medicine. Additional inborn errors such as maple syrup urine disease, galactosaemia, congenital hypothyroidism, and biotinidase deficiency were incorporated into newborn screening programmes of some countries. In the 1990s, a revolutionary technology, tandem mass spectros- copy (MS/​MS), was adopted for newborn screening. The possi- bilities of multianalyte detection by MS/​MS led to a change in the screening paradigm, that is, one test for many diseases (instead of one test for one disease). MS/​MS improved screening for diseases from the conventional screening panels and opened the chance for inclusion of many other inborn errors of metabolism. However, each novel candidate disease has to be evaluated with respect to whether this disease fulfils the criteria for a disease to be screened (see Chapters 2.12 and 12.1), taking into consideration differences in national healthcare systems. As a consequence, the number of screened inborn errors of metabolism varies considerably ranging from two disorders (United Kingdom, Switzerland) up to more than 50 disorders (some parts of the United States of America). Notably, the United States screening panel also includes conditions that can be regarded as nondiseases or have at least a doubtful pathological meaning, such as the 3-​methylcrotonyl CoA carboxylase deficiency. It should be appreciated that a liberal expansion of the screening panel burdens the healthcare system, the affected individuals, and the increasing number of false-​positive individuals and their fam- ilies. Given these difficulties, it is to be hoped that screening politics will become harmonized in a joint international effort. Clinical considerations and diagnostic work-​up History A careful family history may reveal important clues to the diagnosis of protein-​dependent inborn metabolic diseases. Most disorders are inherited as autosomal recessive traits which may be suspected if the parents are consanguineous or the family has a confined ethnic or geographic background. Carriers for particular disorders and affected children may be more frequent in certain communities (e.g. Amish), ethnic groups (e.g. Ashkenazi Jews, Arabic tribes), or countries that have seen little immigration over many centuries (e.g. Finland). Often specialist investigations are started only after a second affected child is born into a family. Older siblings may be

12.2  Protein-dependent inborn errors of metabolism 1945 found to have a similar disorder to the index patient, or to have died from an acute unexplained disease classified as ‘sepsis with uniden- tified pathogen’, ‘encephalopathy’, or ‘sudden infant death syndrome’. Notably, the disease course of the same disorder may vary consid- erably even within families depending on genotype–​phenotype correlation (if any), varying X-​inactivation in female carriers (e.g. ornithine transcarbamylase deficiency), and dominant disorders with variable penetration (e.g. Segawa’s disease). As a result of the successful treatment of inborn errors of metab- olism, an increasing number of affected women are reaching repro- ductive age. If they become pregnant, there may be a risk for their fetuses to be harmed by toxic metabolites from the mother. Especially important is maternal PKU, which is likely to become a major health problem. Other maternal conditions may cause ‘metabolic’ disease in the neonate or infant postnatally, for example, methylmalonic aciduria and hyperhomocystinaemia, in fully breastfed children of mothers who have pernicious anaemia or who are on a vegan diet, which fosters nutritional vitamin B12 deficiency. Clinical spectrum The range of clinical and biochemical manifestations of the protein-​ dependent metabolic errors is wide. Here we focus on the clinical manifestation and differential diagnosis of disorders presenting with acute metabolic decompensations (Boxes 12.2.1 and 12.2.2). There is only a limited repertoire of pathophysiological sequences in the response to metabolic intoxication and, consequently, a limited number of therapeutic measures. Timely and correct intervention during the initial episode is a critical prognostic factor. Many protein-​dependent metabolic errors already manifest in the first days of life with progressive irritability or drowsiness. Most typically, a young infant may vomit or refuse to feed and then rap- idly deteriorates. The initial erroneous diagnoses are usually neo- natal sepsis or intracranial haemorrhage: a presumptive diagnosis of a protein-​dependent inborn error should be considered with equal priority. Children with milder forms may be repeatedly admitted, for example, with unusual metabolic acidosis, hypoglycaemia, or neutropenia in the course of common infections especially gastro- enteritis, before an inborn disorder of metabolism is considered, and routine clinical chemistry may be normal in between crises. A substantial number of patients with protein-​dependent inborn errors of metabolism may present differently with acute encephalop- athy or chronic and fluctuating progressive neurological disease. The so-​called cerebral organic acidaemias (e.g. glutaric aciduria type I) characteristically present with (progressive) neurological symptoms such as ataxia, myoclonus, extrapyramidal symptoms, and metabolic stroke. Routine clinical chemistry is often unrevealing. Important diag- nostic clues such as progressive disturbances of myelination, cere- bellar atrophy, frontotemporal atrophy, signal abnormalities, and/​or infarcts of the basal ganglia can be derived from MRI of the brain. Chronic subdural effusions, haematomas, and retinal haemorrhages in infants and toddlers are characteristic findings in glutaric aciduria type I, although they are more commonly due to child abuse. Laboratory investigations The early consideration of metabolic diseases is of the utmost im- portance. Basic evaluation of metabolic parameters including ana- lyses of blood gases, serum glucose and lactate, plasma ammonia and amino acids, acylcarnitine profiling in dried blood spots (MS/​MS), and organic acid analysis in urine (GC/​MS) should be performed on an emergency basis in every patient presenting with symptoms of unexplained metabolic crisis, intoxication, or encephalopathy. Routine laboratory parameters Diagnostic clues can be obtained from routine laboratory inves- tigations such as electrolytes (also required for the calculation of the anion gap), urinary ketones, serum transaminases, and cre- atine kinase. Any child admitted to an intensive care unit with life-​ threatening nonsurgical illness should be tested for these parameters. Box 12.2.1  Presentation of organic acidurias Intoxication • Kussmaul tachypnoea/​acidotic breathing • Peculiar smell • Refusal of/​adverse reaction to feeding • Protracted episodic vomiting • Erroneous diagnosis of pyloric stenosis (with acidosis) • Reye’s syndrome presentation • Hepatomegaly/​liver failure • Rhabdomyolysis • Sudden infant death syndrome (SIDS) or ‘near miss’ SIDS Acute encephalopathy • Coma • Seizures (myoclonic, intractable) • Acute profound dyskinesia • Pseudotumour cerebri • Cerebral/​intraventricular haemorrhage in full-​term babies • Stroke-​like episodes Chronic encephalo(myelo)pathy • Progressive psychomotor deterioration • Macrocephaly • Ataxia (progressive) • Hypotonia • Dystonia, athetosis • Myoclonus • Seizures (myoclonic, intractable) • Peripheral neuropathy • Pyramidal signs—​‘cerebral palsy’ • Pronounced deficiency of speech • Congenital cerebral malformations Box 12.2.2  Clinical chemical indices of organic acidurias • Metabolic acidosis • Increased anion gap • Hyperglycaemia • Ketosis and ketonuria (especially suggestive in newborns) • Lactic acidosis • Hyperammonaemia • Hyperuricaemia • Hypertriglyceridaemia • Increase of transaminases • Granulocytopenia, thrombocytopenia, and anaemia • Hypoketotic hypoglycaemia (fatty acid oxidation defects) • Increased creatine kinase (fatty acid oxidation defects) • Myoglobinuria (fatty acid oxidation defects)

SECTION 12  Metabolic disorders 1946 Amino acid analysis Many metabolic parameters show considerable diurnal fluctu- ations. For example, plasma amino acid concentrations are highly dependent on the metabolic status, and standard samples should be obtained at least 4 h postprandially. Many amino acids can be reliably quantified in dried blood spots by MS/​MS (e.g. for PKU). Homocysteine and tryptophan require specific methods (usu- ally high-​performance liquid chromatography (HPLC)) for exact quantification. Regular amino acid analyses are required in pa- tients on specific dietary treatments to adjust intake of amino acids and to recognize a deficiency of essential amino acids and micronutrients. For optimal results, it is important to separate plasma as soon as possible and to ship samples frozen on dry ice. Haemolysis or shipment of whole blood results in useless values for some amino acids. Some potential problems are summarized in Box 12.2.3. Quantitative urinary amino acid analysis is indicated only if a renal tubular reabsorption defect such as cystinuria is sus- pected, and (in addition to plasma analysis) in hyperammonaemia when increased urinary excretion of specific metabolites (e.g. argininosuccinate) may be diagnostic. Organic acid analysis Organic acid analysis is best performed on early morning urine specimens. Complete information of the clinical status and recent management of the patient is indispensable for correct interpret- ation, which is based on key diagnostic metabolites or character- istic biochemical patterns. Repeated analyses may be necessary, preferably during exacerbation of metabolic decompensation, since analyses may be intermittently normal. Characteristic me- tabolites may, however, also become masked in severe metabolic decompensation and ketosis. Some patients with organic acid dis- orders may exhibit only slight elevations of diagnostic metabol- ites that may be underestimated by conventional analysis, such as in 4-​hydroxybutyric aciduria, glutaric aciduria type I, and N-​acetylaspartic aciduria. In these disorders, quantification by stable isotope dilution assays is preferred. This is also the method of choice for biochemical prenatal diagnosis of organic acid dis- orders in amniotic fluid, if the causative mutations are not avail- able, providing more rapid diagnosis than enzyme analysis of cultured amniocytes. Acylcarnitine analysis A complementary and rapid diagnostic technique for some or- ganic acid disorders is the analysis of acylcarnitine by MS/​ MS—​by analogy to newborn mass screening—​since accumu- lating acyl-​CoA esters are in equilibrium with corresponding acylcarnitines. Principles of treatment General aspects Protein-​dependent metabolic disorders are chronic conditions that involve various organ systems and thus require a multidisciplinary approach to care and treatment. Patients with genetic diseases that are prone to acute decompensations should carry an emergency card. Vaccinations should be carried out as recommended and should also include vaccinations against varicella, hepatitis A, pneumococcus, and influenza. Special precautions must be taken before, during, and after surgery/​anaesthesia. Dietary treatment In many protein-​dependent errors of metabolism, therapy is based on reduced intake of precursors in deficient pathways, prevention of catabolism, and an intensification of therapy during intercurrent illnesses. This aims to diminish the supply of toxic metabolites and restore energy supply. Dietary treatment must meet the general, age-​ dependent, and individual requirements for energy and essential nutrients to ensure normal growth and development (Table 12.2.1). Protein deficiency induces catabolism, failure to thrive, and growth retardation, and secondary depletion of essential amino acids and micronutrients may induce life-​threatening complications such as lactic acidosis (thiamine or biotin depletion) or pellagra (niacin depletion). Supplementation of precursor-​free mixtures of amino acids and semisynthetic supplements of minerals and trace elements minimizes the risk for malnutrition. Pharmacotherapy Carnitine at daily doses of 50–​200 mg/​kg body weight is essential for the elimination of accumulating toxic acyl-​CoA compounds and for the restoration of intramitochondrial free CoA-​SH in most organic acid disorders. In cofactor-​responsive disorders, enzyme activity may be restored by specific vitamins, for example, in biotinidase deficiency, cobalamin-​responsive methylmalonic acidurias, and riboflavin-​responsive multiple acyl-​CoA dehydrogenase deficiency. Box 12.2.3  Some pitfalls of amino acid analysis • Shipping/​storage without adequate cooling: ↓ glutamine, asparagine, cysteine, homocysteine; ↑ glutamate, aspartate • Haemolysis:  ↓ arginine, glutamine; ↑ aspartate, glutamate, glycine, ornithine • Postprandial changes: all amino acids Table 12.2.1  Protein requirements Age Revised safe values (g/​kg per day) 0–​1 months 2.69 1–​2 months 2.04 2–​3 months 1.53 3–​4 months 1.37 4–​5 months 1.25 5–​6 months 1.19 6–​9 months 1.09 9–​12 months 1.02 1–​3 years 1.0–​0.92 4–​10 years 0.88–​0.86 11–​18 years 0.86–​0.77 Source data from Dewey KG, et al. (1996). Protein requirements of infants and children. Eur J Clin Nutr, 50 Suppl 1, S119–​47.

12.2  Protein-dependent inborn errors of metabolism 1947 The accumulation of toxic metabolites derived from gut bacteria, such as propionic acid, can be reduced by intestinal antibiotics (e.g. metronidazole). Emergency treatment Treatment of intercurrent illness at home Protein-​dependent inborn errors of metabolism often present with acute life-​threatening decompensation requiring prompt decisions and measures. A limited number of therapeutic measures have to be taken immediately (Box 12.2.4, Table 12.2.2). It is imperative to decrease catabolism at an early stage of de- compensation. As this usually happens at home, it is essential to educate the family adequately. Home treatment should in- clude adequate control of fever and vomiting, moderate protein restriction, and ample calories, glucose, and fluid (Box 12.2.4). Intake of natural protein can be completely eliminated for the first 24 h of illness, especially if the patient receives precursor-​ free supplements of amino acids. After 24 h, stepwise reintroduc- tion of natural protein is necessary to prevent protein catabolism. Immediate hospital admission and intravenous treatment is in- dicated when vomiting persists, fluid and dextrose intake remain poor, the clinical condition deteriorates, or the disease course is prolonged. On admission to hospital, these patients must be as- sessed and treated without delay. If emergency management is carried out in peripheral hospitals, this should ideally be super- vised in consultation with a knowledgeable and experienced physician or paediatrician. Emergency treatment in hospital Provision of ample quantities and control of fluid and electrolytes is indispensable and must be continued before any laboratory results are available. Glucose infusions must be adapted to age to provide an adequate energy supply. For example, in neonates glucose infusion is usually started at 10 mg/​kg per min (i.e. 14.4 g glucose/​kg body weight per day). An insulin drip may be necessary to prevent hyper- glycaemia and to induce an anabolic state. Overhydration is rarely a problem in metabolic crises as they are mostly accompanied by dehydration. Electrolytes, glucose, lactate, and acid–​base balance should be checked at least every 6 h and serum sodium should be maintained at no less than 138 mmol/​litre. If lactate is constantly increasing while the glucose supply is increasing, one should con- sider a primary defect or secondary inhibition or energy metab- olism, such as in classic organic acid disorders. Antibiotics should be started if there is evidence for an infectious cause. Antipyretics should be administered liberally since they help to reduce the add- itional bioenergetic costs of fever. Carnitine is essential for the elimination of toxic acyl-​CoA es- ters in organic acidaemias, to prevent secondary carnitine deple- tion, and to replenish the intracellular CoA pool. Carnitine should be administered intravenously at 100 to 200 mg/​kg per day. In hyperammonaemia, nitrogen-​disposing drugs are used: • Sodium benzoate, 250 mg/​kg as bolus initially over 1 to 2 h, then 250 (to 500) mg/​kg per 24 h. Box 12.2.4  Basic principles for acute emergency therapy 1 Suppress muscle and liver protein catabolism and ensure a glucose supply above the basal metabolic demand 2 Treat the precipitating illness 3 Reduce increased production of toxic metabolites by reduction or omission of natural protein 4 Enhance detoxifying mechanisms and urinary excretion of patho- logical metabolites 5 Aggressively treat dehydration and acidosis 6 Prevent secondary carnitine depletion 7 Provide alternative routes of ammonia disposal in hyperammonaemia Table 12.2.2  Home and outpatient emergency treatment Age (years) % kcal/​100 ml Daily amount A. Glucose polymer/​maltodextrin solutiona 0–​1 10   40 150–​200 ml/​kg 1–​2 15   60 95 ml/​kg 2–​10 20   80 1200–​2000 ml/​day

10 25 100 2000 ml/​day B. Protein intake Natural protein Stop (if amino acid supplements are administered) or reduce to 50% of maintenance therapy (if no amino acid supplements are administered). Reintroduce and increase within 1–​2 days Amino acid mixtures If tolerated, amino acid supplements should be administered according to maintenance therapy, e.g. 0.8–​1.0 g/​kg body weight/​dayb C. Pharmacotherapy l-​Carnitine Double carnitine intake: 200 mg/​kg body weight/​day orally (if tolerated) Antipyreticsc If temperature >38.5°C, e.g. ibuprofen (10–​15 mg/​kg body weight per dose, 3–​4 doses daily) a Maltodextran/​dextrose solutions should be administered every 2 h day and night. If neonates and infants already receive a specific dietary treatment, protein-​free food can be continued but should be fortified by maltodextran. Patients should be reassessed every 2 h. b All calculations should be based on the expected weight, not the actual weight. c Paracetamol administration may be dangerous in acute metabolic decompensation (risk for glutathione depletion).

SECTION 12  Metabolic disorders 1948 • Sodium phenylacetate, 250 mg/​kg as bolus initially over 1 to 2 h, then 250 (to 600)  mg/​kg per 24 h; alternatively, sodium phenylbutyrate is administered at the same concentration orally. • Arginine hydrochloride, 420 mg/​kg (i.e. 2 mmol/​kg) as bolus ini- tially over 1 to 2 h, then 420 mg/​kg per 24 h. If the response to emergency treatment is poor, the patient de- teriorates, or the ammonia concentration exceeds 400 to 500 µmol/​ litre (neonate, infant), haemofiltration or haemodialysis should be urgently considered. Since intracranial pressure due to cerebral oe- dema appears earlier in older children, adolescents, and adults than in newborns, infants, and younger children, extracorporeal detoxi- fication should be considered if ammonia concentration exceeds 200 µmol/​litre or even as first-​line treatment. If persisting lactic acid- osis is present, thiamine (100–​500 mg/​day) and biotin (10–​20 mg) should be given empirically. Monitoring of treatment Dietary treatment without adequate monitoring is dangerous since disease-​specific complications, therapy-​specific adverse events (e.g. malnutrition), and developmental delay might be overlooked. Anthropometric parameters such as weight, height, and head cir- cumference should be recorded at each visit. Psychomotor devel- opment must be regularly assessed with appropriate tests. Weight loss or insufficient weight gain in affected children is often caused by inadequate dietary treatment and may herald impending metabolic decompensation. The major aim of biochemical monitoring is to ensure that nu- trition is not compromised. Biochemical evaluation includes blood count, serum electrolytes, calcium, phosphate, magnesium, ferritin level, liver and kidney function tests, alkaline phosphatase, total protein, albumin, prealbumin, transferrin, cholesterol, triglycer- ides, zinc, copper, retinol (plasma), carnitine, ammonia, lactate, and plasma amino acids. Although analyses of specific metabolic parameters are required to confirm the diagnosis of an inborn error of metabolism, these parameters are often not informative for bio- chemical follow-​up monitoring since the relationship between the metabolic parameters and outcome is unclear for most disorders. However, regular monitoring of some metabolic parameters is ne- cessary since they are directly related to the outcome. For example, plasma phenylalanine is monitored in PKU, plasma leucine in maple sugar urine disease, plasma glutamine and arginine in urea cycle defects, and plasma homocysteine in trans-​sulphuration and remethylation defects. Likely future developments The scientific and technological advances described in the previous sections have offered much benefit to patients with inborn errors of metabolism. To implement and utilize them properly, much remains to be done. Initially, metabolic physicians and scientists need to combine their efforts and concentrate on well-​conducted international studies and development of evidence-​based guide- lines. Significant differences still exist in the diagnostic procedures, treatment, and monitoring of many diseases, resulting in a wide variation in outcome. Even for PKU, the disease with the greatest and longest experience in successful therapy, current guidelines recommend different cut-​offs for the indication of treatment ran- ging from 400 µmol/​litre in the United Kingdom to 360  µmol/​litre in the United States of America and 600 µmol/​litre in France and Germany. The knowledge of the academic community must be combined and structured, transferred to the physicians and other medical staff, and implemented in healthcare systems. Nowadays, this process has become much easier by means of numerous re- commendations, information, and even projects available on the Internet, permanent professional email round tables, Internet editions of book and journals, and open-​access databases. In the necessary implementation process, regional differences such as availability of funds, local pathology, and religious and geographic factors must be taken into account. Accordingly, specialized na- tional metabolic centres and appropriate metabolic networks should be established and properly maintained. Unfortunately, novel diagnostic and therapeutic possibilities (Box 12.2.5), such as newborn screening or enzyme replacement therapy, are relatively expensive and are still unrealistic for many countries where there are no screening programmes and perhaps no well-​organized healthcare system. Individual disorders A summary of protein-​dependent inborn errors of metabolism including the enzyme defect, incidence, gene locus, and Online Mendelian Inheritance in Man (OMIM) number is given in Table 12.2.3. Urea cycle defects Aetiology/​pathophysiology The major source of ammonia is catabolism of protein, which is detoxified to urea in the liver (Fig. 12.2.1). The efficiency of hep- atic ammonia detoxification is enhanced through the action of glu- tamine synthase. Hyperammonaemia (plasma ammonia >80 µmol/​ litre in newborns; >50 µmol/​litre after the newborn period) is caused by increased production (e.g. by intestinal urease-​producing bac- teria) or decreased detoxification of ammonia. Decreased detoxifi- cation results from inherited or acquired deficiency of key enzymes and transporters of the urea cycle, or bypassing of the liver (e.g. open hepatic duct). Secondary impairment of ammonia detoxification re- sults from conditions where glutamate or acetyl-​CoA are decreased, Box 12.2.5  New treatment strategies in inborn errors of metabolism • Supplementation with end products • Anaplerotic therapy • Enzyme replacement • Chemical chaperones • Specific blockade of biosynthetic pathways • Specific blockade of degradation pathways • Specific blockade of pathophysiological signalling • (Stem) cell therapy • Gene therapy

12.2  Protein-dependent inborn errors of metabolism 1949 Table 12.2.3  Summary of protein-​dependent inborn errors of metabolism Disease Enzyme defect Incidencea Gene map locus Gene name OMIM (phenotype number) Defects of the urea cycle Argininaemia Arginase 1 1:100 000 6q23 ARG1 207800 Argininosuccinic aciduria Argininosuccinate lyase 1:50 000 7cen–​q11.2 ASS1 207900 Citrullinaemia type I Argininosuccinate synthetase 1 1:50 000 9q34 ASL 215700 Deficiency of Citrin <1:200 000 7q21.3 SLC25A13 605814 (neonatal onset) 603471 (adult onset) Deficiency of N-​Acetylglutamate synthase <1:200 000 17q21.3 NAGS 237310 Deficiency of Carbamoylphosphate synthetase 1 1:50 000 2q35 CPS1 237300 Deficiency of Ornithine carbamoyltransferase 1:30 000 Xp21.1 OTC 311250 Dibasic amino aciduria II, lysinuric protein intolerance <1:200 000 14q11.2 SLC7A7 222700 Hyperornithinaemia–​hyperammonaemia–​ homocitrullinuria syndrome Ornithine transporter <1:200 000 13q14 SLC25A15 238970 Carbonic anhydrase VA deficiency Mitochondrial carbonic anhydrase VA Unknown 16q24.2 CA5A 114761 Defects of branched-​chain amino acid metabolism Isovaleric aciduria Isovaleryl-​CoA dehydrogenase 1:80 000 15q14–​q15 IVD 243500 Maple syrup urine disease Branched-​chain keto acid dehydrogenase (lipoamide) 1:200 000 248600   Type Ia E1 component α-​chain 19q13.1–​q13.2 BCKDHA   Type Ib component β-​chain 6p21–​p22 BCKDHB   Type II dihydrolipoamide branched-​chain
  transacylase (E2 component) 1p31 DBT 3-​Methylcrotonylglycinuria 3-​Methylcrotonyl-​CoA carboxylase 1:60 000 210200   α-​subunit 3q25–​q27 MCCC1   β-​subunit 5q12–​q13 MCCC2 3-​Methylglutaconyl-​CoA hydratase deficiency
(3-​methylcrotonyl aciduria type I) 3-​Methylglutaconyl-​CoA hydratase <1:200 000 9q22.31 AUH 250950 TAZ defect or Barth syndrome (3-​methylglutaconic aciduria type II) Tafazzin <1:200 000 Xq28 TAZ 302060 OPA3 defect or Costeff’s syndrome
(3-​methylglutaconic aciduria type III) OPA3A and OPAB protein <1:200 000 19q13.2–​q13.3 OPA3 258501 3-​Methylglutaconic aciduria type IV (i.e. MEGDEL syndrome, TMEM70 defect, or not otherwise specified) E.g. polymerase-​γ, transmembrane protein 70, succinate-​CoA ligase, serine active site-​ containing protein 1 or not yet classified <1:200 000 e.g. 15q26.1, 8q21.11, 13q14.2, 6q25.3, or ? e.g. POLG1, TMEM70, SUCLA2, SERAC1, or? 250951 (if not otherwise specified) DNAJ19 defect or DCMA syndrome
(3-​methylglutaconic aciduria type V) Translocase of the inner mitochondrial membrane 14 Unknown 3q26.33 DNAJC19 610198 (continued)

SECTION 12  Metabolic disorders 1950 Disease Enzyme defect Incidencea Gene map locus Gene name OMIM (phenotype number) 2-​Methyl-​3-​hydroxybutyryl-​CoA deficiency 2-​Methyl-​3-​hydroxybutyryl-​CoA dehydrogenase <1:200 000 Xp11.2 HSD17B10 300438 Methylmalonic aciduria (mut0/​mut− defects) Methylmalonyl-​CoA mutase 1:100 000 6p12.3 MUT 251000 Propionic aciduria Propionyl-​CoA carboxylase 1:200 000   α-​chain 13q32 PCCA 232000   β-​chain 3q21–​q22 PCCB 232050 3-​Hydroxyisobutyryl-​CoA hydrolase deficiency 3-​Hydroxyisobutyryl-​CoA hydrolase Unknown 2q32.2 HIBCH 250620 Short-​chain enoyl-​CoA hydratase deficiency Mitochondrial short-​chain enoyl-​CoA hydratase 1 Unknown 10q26.3 ECHS1 616277 Defects of lysine, hydroxylysine, and tryptophan metabolism 2-​Aminoadipic and oxoadipic aciduria Dehydrogenase E1 and transketolase domains-​ containing protein 1 <1:200 000 10p14 DHTKD1 204750 2-​Oxoadipic aciduria Dehydrogenase E1 and transketolase domains-​ containing protein 1 <1:200 000 10p14 DHTKD1 245130 Glutaric aciduria type I Glutaryl-​CoA dehydrogenase 1:100 000 19p13.2 GCDH 231670 Gyrate atrophy of choroid and retina Ornithine-​oxoacid/​ ornithine aminotransferase <1:200 000 10q26 OAT 258870 Hyperlysinaemia Saccharopine dehydrogenase/​lysine:α-​ ketoglutarate reductase <1:200 000 7q31.32 AASS 238700 Saccharopinuria Saccharopine dehydrogenase/​lysine:α-​ ketoglutarate reductase <1:200 000 7q31.32 AASS 268700 Multiple carboxylase deficiency Biotinidase deficiency Biotinidase 1:80 000 3p25 BTD 253260 Holocarboxylase synthetase deficiency Holocarboxylase synthetase <1:200 000 21q22.1 HLCS 253270 Other organic acidurias N-​Acetylaspartic aciduria (Canavan’s disease) Aspartoacylase; aminoacylase 2 <1:200 000 17pter–​p13 ASPA 271900 Ethylmalonic encephalopathy Mitochondrial matrix protein <1:200 000 19q13.2 ETHE1 602473 D-​2-​Hydroxyglutaric aciduria Type I: D-​2-​hydroxyglutaric acid dehydrogenase <1:200 000 2p25.3 D2HGDH 600721 Type II: isocitrate dehydrogenase 2 (mitochondrial) <1:200 000 15q26.1 IDH2 613657 L-​2-​Hydroxyglutaric aciduria FAD-​dependent L-​2-​hydroxyglutarate dehydrogenase <1:200 000 14q22.1 L2HGDH 236792 Combined D-​2-​ and L-​2-​hydroxyglutaric aciduria Mitochondrial citrate transporter <1:200 000 22q11.21 SLC25A1 615182 Defects of phenylalanine and tyrosine metabolism Alkaptonuria Homogentisate 1,2-​dioxygenase <1:200 000 3q21–​q23 HGD 203500 BH4 deficiency, dopa-​responsive dystonia (dominant) Guanosine-​5-​triphosphate cyclohydrolase 1:100 000 14q22.1–​q22.2 GCH1 128230 Table 12.2.3  Continued

12.2  Protein-dependent inborn errors of metabolism 1951 BH4 deficiency Deficiency of Dihydropteridine reductase <1:200 000 4p15.32 QHPR 261630 Deficiency of Guanosine-​5-​triphosphate cyclohydrolase <1:200 000 14q22.1–​q22.2 GCH 233910 Deficiency of 6-​Pyruvoyltetrahydropterin synthase <1:200 000 11q22.3–​q23.3 PTS 261640 Deficiency of Sepiapterin reductase <1:200 000 2p13.2 SPR 612716 Deficiency of Pterin-​4α-​carbaminoline dehydratase Unknown 10q22.1 PCBD1 264070 Phenylketonuria (PKU) Phenylalanine hydroxylase 1:10 000 12q24.1 PAH 261600   Type I (Classical PKU = Phe >1200 µmol/​litre) c.50%   Type II (Mild PKU = 360–​600 µmol/​litre ≤ Phe ≤ 1200 µmol/​litre) c.30%   Type III (Non-​PKU HPA/​MHP = Phe <360–​600 µmol/​ litre) c.20%   Types II+III (BH4-​PAH = Phe <1200 µmol/​litre + BH4-​ responsive) c.35% Hyperphenylalaninaemia with primapterinuria Pterin-​4α-​carbinolamine <1:200 000 10q22.1 PCBD 264070 Tyrosinaemia type I Fumarylacetoacetase 1:100 000 15q23.1 FAH 276700 Tyrosinaemia type II Tyrosine aminotransferase <1:200 000 16q22.2 TAT 276600 Tyrosinaemia type III 4-​Hydroxyphenylpyruvate dioxygenase <1:200 000 12q24.31 HPD 276710 Neurotransmitter diseases and related disorders Deficiency of Aromatic L-​amino acid decarboxylase <1:200 000 7p12.1 DDC 608643 Deficiency of Dopamine β-​hydroxylase <1:200 000 9q34.2 DBH 223360 Deficiency of GABA transaminase <1:200 000 16p13.2 ABAT 613163 Deficiency of 3-​Phosphoglycerate dehydrogenase <1:200 000 1p12 PHGDH 601815 Deficiency of Tyrosine hydroxylase <1:200 000 11p15.5 TH 605407 Folinic acid-​responsive epilepsy (see ‘Pyridoxine-​ dependent epilepsy’) α-​Aminoadipic semialdehyde dehydrogenase (antiquitin) <1:200 000 5q23.2 ALDH7A1 266100 4-​Hydroxybutyric aciduria Succinic semialdehyde dehydrogenase <1:200 000 6p22.3 ALDH5A1 271980 Hyperprolinaemia type II L-​Δ1-​pyrroline-​5-​carboxylate dehydrogenase <1:200 000 1p36 ALDH4A1 239510 Nonketotic hyperglycinaemia (glycine encephalopathy) 1:60 000 605899 H-​protein deficiency 16q22.24 GCSH P-​protein deficiency 9p24.1 GLDC T-​protein deficiency 3p21.31 AMT Other: transient 605899 Pyridoxal phosphate-​dependent epilepsy Pyridox(am)ine 5′-​phosphate oxidase <1:200 000 17q21.32 PNPO 610090 Pyridoxine-​dependent epilepsy α-​Aminoadipic semialdehyde dehydrogenase (antiquitin) <1:200 000 5q23.2 ALDH7A1 266100 (continued)

SECTION 12  Metabolic disorders 1952 Disease Enzyme defect Incidencea Gene map locus Gene name OMIM (phenotype number) Defects of trans-​sulphuration and remethylation Deficiency of S-​adenosyl-​homocysteine hydrolase <1:200 000 20q11.22 AHCY 613752 Deficiency of Adenosine kinase <1:200 000 10q22.2 ADK 614300 Deficiency of Cysthationine γ-​lyase < 1:70 000 1p31.1 CTH 219500 Deficiency of Glycine N-​methyltransferase <1:200 000 6p21.1 GNMT 606664 Deficiency of Methionine adenosyltransferase 1 <1:200 000 10q23.1 MAT1A 250850 Deficiency of Methionine synthase reductase (cobalamin E) <1:200 000 5p15.31 MTRR 236270 Deficiency of Methionine synthase (cobalamin G) <1:200 000 1q43 MTR 250940 Deficiency of 5,10-​Methylene-​tetrahydrofolatreductase <1:200 000 1p36.22 MTHFR 236250 Homocystinuria Cystathionine-​β-​synthase 1:100 000 21q22.3 CBS 236200 a Incidences as estimated in the white population; they vary between populations of different ethnic background. <1:200 000 indicates incidence very low but uncertain because not specifically determined. Of some disorders, only two or three families are as yet known worldwide. BH4-​PAH, BH4-​responsive phenylalanine hydroxylase deficiency; Phe, phenylalanine; PKU, phenylketonuria. Table 12.2.3  Continued

12.2  Protein-dependent inborn errors of metabolism 1953 such as in organic acid defects, mitochondrial β-​oxidation de- fects, carnitine depletion, or valproate therapy, or where toxic acyl-​CoAs are increased, such as propionyl-​CoA in propionic and methylmalonic aciduria or isovaleryl-​CoA in isovaleric aciduria. Hyperammonaemia is neurotoxic, resulting in brain oedema, convulsions, and coma. Neuropathological evaluation reveals an alteration of astrocyte morphology including cell swelling (acute hyperammonaemia) and Alzheimer type II astrocytosis (chronic hyperammonaemia). The brain relies on energy-​dependent glu- tamine synthesis by astrocytic glutamine synthetase for the removal of excess ammonia. As a consequence, increased brain ammonia is considered to amplify glutamatergic signalling and cause redistri- bution of cerebral blood flow and metabolism, impairment of brain energy metabolism affecting the glutamate/​glutamine cycle, and in- creased serotonin secretion. Hyperammonaemia exerts reversible (mostly serotoninergic) and irreversible effects. Peak plasma am- monia concentrations exceeding 500 µmol/​litre or a coma lasting more than 2 to 3 days appears to be associated with irreversible de- fects which worsen with the duration of the coma. All inherited urea cycle defects follow an autosomal recessive trait except for ornithine transcarbamylase deficiency which is X-​linked. Clinical presentation Urea cycle defects are among the most common inborn errors of metabolism (cumulative incidence is approximately 1 in 40 000 newborns). Six inherited urea cycle defects are well described, that is, deficiencies of N-​acetylglutamate synthetase, carbamoyl-​ phosphate synthase 1, ornithine transcarbamylase, argininosuccinate synthetase and lyase, and arginase 1 (Fig. 12.2.1). Deficiency of glu- tamine synthetase has also been identified but is not described here. Five urea cycle defects share a common but variable clinical presen- tation due to hyperammonaemia. Arginase 1 deficiency and defects of cellular transport including transporter proteins for the dibasic amino acids ornithine (hyperornithinaemia–​hyperammonaemia–​ homocitrullinuria syndrome) and aspartate (citrullinaemia II) result in a more subtle disease with predominantly neurological symptoms. Onset of symptoms may occur at any age; however, it is par- ticularly frequent during the neonatal period, late infancy, and puberty, and is precipitated by excess protein or episodes that in- duce catabolism such as infectious diseases, trauma, or cortisone therapy. In general, symptoms are less severe with increasing age at onset. Neonatal presentation starts after a short asymptomatic interval with poor feeding, vomiting, lethargy, tachypnoea, and/​or irritability which cannot be distinguished clinically from neonatal sepsis. Untreated, acute encephalopathy rapidly progresses to death. In infancy, the symptoms are less acute and more variable than in the neonatal period including anorexia, vomiting, developmental delay, and behavioural problems. In X-​linked ornithine transcarbamylase deficiency, female carriers may also be affected due to variable in- activation of the X chromosome (the Lyon hypothesis). Clinical presentation ranges from acute liver failure, cognitive disability, and behavioural problems to psychiatric disease. In arginase 1 defi- ciency, patients usually present with progressive spasticity which is often mistaken for cerebral palsy, seizures, and learning difficulties. Dystonia and ataxia may develop. Acute decompensation occurs rarely. The phenotypic variation of patients with urea cycle disorders as well as evidence-​based recommendations for diagnosis, treat- ment, and follow-​up have recently been reported by an international consortium of experts. Diagnosis Emergency analysis of ammonia must be part of the basic investiga- tions in all patients at all ages with unclear encephalopathy or acute hepatic failure. Among the inherited hyperammonaemias, two-​thirds are due to urea cycle defects and one-​third to organic acid and other inborn errors. Blood gas analyses and anion gap determinations may show alkalosis and normal anion gap in urea cycle defects and acidosis and increased anion gap in organic acid disorders. Characteristic biochemical changes (glutamine, alanine, citrulline, ornithine, ar- ginine, argininosuccinic acid, orotic acid, uracil) can be identified by plasma amino acid analysis, GC/​MS analysis of urinary organic acids, or HPLC analysis of orotic acids and orotidine. The diagnosis can be confirmed by enzyme analysis in liver tissue (all urea cycle defects except for N-​acetylglutamate synthase deficiency), fibro- blasts (argininosuccinate synthase 1 and lyase), or molecular genetic studies. Prenatal diagnosis is possible. Autosomal recessive inherited urea cycle disorders can be identified by molecular genetic studies on chorionic villus biopsy. Enzyme analysis can be performed for deficiencies of argininosuccinate lyase and synthase. Arginase defi- ciency can also be diagnosed biochemically by fetal blood analysis. Therapy and outcome The aim of treatment is to correct the biochemical disorder (glutamine in plasma <800–​1000 µmol/​litre, ammonia <80 µmol/​litre, arginine 80–​150 µmol/​litre) and to ensure that the patient grows normally and thrives. The major metabolic strategies are (1) reduction of natural protein to decrease ammonia production, (2) supplementation with essential amino acids to prevent malnutrition and to reutilize ni- trogen for the synthesis of nonessential amino acids, (3) replacement of arginine or citrulline which become essential amino acids in all urea cycle disorders except for arginase 1 deficiency, and (4) utiliza- tion of alternative pathways for nitrogen excretion. This last strategy ­includes application of sodium benzoate (250–​500 mg/​kg per day) and Fumarate Mitochondrion = Ornithine transporter Cytosol Ammonia HCO3− CPS1 Carbamyl phosphate Orotic acid Orotidine Uracil Glutamate N-acetylglutamate NAGS Citrulline Aspartate Argininosuccinate Arginine Urea Cycle Ornithine OTC ASS ASL Arginase 1 Urea T T ⊕ Fig. 12.2.1  The urea cycle. ASL, argininosuccinate lyase; ASS, argininosuccinate synthase; CPS1, carbamyl phosphate synthase 1; NAGS, N-​acetylglutamate synthase; OTC, ornithine transcarbamylase; T, ornithine transporter. Source data from Zschocke J, Hoffmann GF (2011). Vademecum metabolicum. Manual of metabolic paediatrics, 3rd edition. Schattauer, Stuttgart.

SECTION 12  Metabolic disorders 1954 sodium phenylbutyrate or phenylacetate (250–​600 mg/​kg per day) to conjugate glycine or glutamine, resulting in urinary excretion of waste nitrogen in alternative compounds (hippurate, phenyl­ acetylglutamine). In N-​acetylglutamate synthase ­deficiency, N-​ carbamylglutamate can be used as an alternative allosteric activator of carbamoyl ​phosphate synthase. All patients with urea cycle defects are at risk of acute metabolic decompensation precipitated by metabolic stress such as protein load, infection, anaesthesia, or surgery. To prevent or reverse meta- bolic crises, a stepwise implementation of an intensified emergency treatment is required (see also ‘Emergency treatment’). If diet and pharmacotherapy is insufficient to improve hyperammonaemia sig- nificantly and rapidly, haemofiltration or haemodialysis should be considered. Main factors that determine outcome are duration and severity of hyperammonaemia the specific disease, and age at disease onset are considered as most important. In general, a beneficial outcome critically relies on rapid diagnosis and immediate start of treatment after the onset of first symptoms. Carbonic anhydrase VA deficiency Aetiology/​pathophysiology Bicarbonate cannot enter the mitochondria and thus is generated within the mitochondria by two carbonic anhydrases:  VA and VB. Carbonic anhydrase VA provides bicarbonate as a substrate to carbamyl phosphate synthase 1, the first enzymatic step of the urea cycle, as well as to three mitochondrial carboxylases, pyruvate carboxylase, propionyl-​CoA carboxylase, and 3-​methylcrotonyl-​ CoA carboxylase which are involved in energy metabolism and the catabolic pathways of branched-​chain amino acids, respectively. Combined dysfunction of these four mitochondrial enzymes due to limited availability of their substrate bicarbonate causes a biochem- ical derangement including hyperammonaemia, impaired energy metabolism (affecting gluconeogenesis and tricarboxylic cycle) with lactic acidosis, as well as organic acidurias resembling propionic aciduria and 3-​methylcrotonyl-​CoA carboxylase deficiency (see ‘3-​Methylcrotonylglycinuria’). Presentation Recently, four patients with this disease have been reported. Three of them presented with lethargy, tachypnoea, hypoglycaemia, hyperammonaemia, hyperlactatemia with hyperalaninaemia, and respiratory alkalosis during the first days of life; in one of them, the initial metabolic crisis occurred at age 13 months. During the follow-​ up, episodes of acute encephalopathy were precipitated by catabolic stress. Motor and mental development was within the normal range in one child, whereas delayed motor development due to ataxia and mild axial hypotonia, psychomotor retardation, or learning difficul- ties was found in the other children. Diagnosis Metabolic tests reveal a unique pattern of elevated lactate and hyperammonaemia with elevated glutamine and alanine, but low citrulline and arginine in plasma in combination with increased urinary excretion of lactate, ketone bodies, propionate metabolites, methylcrotonylglycine, and 3-​hydroxyisovaleric acid. The meta- bolic pattern can be identified by analysis of plasma amino acids and organic acids in urine. Notably, newborn screening profiles, specifically C3 and C5OH levels, were unremarkable in all index pa- tients. The diagnosis can be confirmed by molecular genetic testing. Carbamyl phosphate synthase 1 and N-​acetylglutamate synthase de- ficiency are the most relevant differential diagnosis. In children with negative molecular genetic test results in CPS1 and NAGS genes, car- bonic anhydrase VA deficiency should be considered. Treatment and outcome Treatment with preventive sick-​day management using high-​caloric, lipid-​rich and low-​protein formula to enhance anabolism and to re- duce the formation of toxic metabolites as well as carglumic acid to enhance the activity of carbamyl phosphate synthase 1 can be ad- ministered. Although carbonic anhydrase VA deficiency should be considered as treatable condition, treatment strategies have not yet been studied systematically. The long-​term outcome of this disease is unknown. Defects of branched-​chain amino acid metabolism Maple syrup urine disease Maple syrup urine disease was first reported in 1954 by Menkes, Hurst, and Craig, who noticed an unusual odour reminiscent of maple syrup in the urines of four infants who died from a rapidly pro- gressive neurological disease. In newborn screening programmes, a prevalence of approximately 1 in 200 000 newborns is encountered but in the Mennonites in Pennsylvania, the prevalence is as high as 1 in 200. Maple syrup urine disease is frequent in other ethnic groups and isolates such as persons of French Canadian origin. In maple syrup urine disease, the branched-​chain amino acids leucine, isoleucine, and valine, their corresponding α-​keto acids and hydroxy acid derivatives, as well as l-​alloisoleucine are increased in physiological fluids. These amino acids and their metabolites accumulate due to inherited deficiency of the thiamine-​dependent branched-​chain α-​keto acid dehydrogenase complex, consisting of subunits E1α, β, E2, and E3 (Fig. 12.2.2). l-​Alloisoleucine results from racemization of the 3-​carbon of l-​isoleucine during transamination. Its elevation is pathogno- monic for maple syrup urine disease. Presentation Several clinical presentations have been delineated but there is con- siderable overlap. Most frequently the condition comes to light in the first few days of life with lethargy, irritability, poor feeding, and neurological deterioration. Later-​onset forms of maple syrup urine disease are slower with failure to thrive, developmental delay, and sometimes seizures; episodic ataxia and stupor sometimes pro- gressing to coma may be precipitated by high protein intake or intercurrent illness. In patients showing a response to thiamine, the condition tends to resemble later-​onset maple syrup urine disease. A very rare related disease results from deficiency of lipoamide de- hydrogenase presenting after the neonatal period with lactic acid- osis, hypotonia, developmental delay, abnormal movement, and progressive neurological deterioration. Most patients with maple syrup urine disease have the classic form. If untreated, these neonates quickly deteriorate, developing lethargy, hypotonia alternating with muscular rigidity, opisthotonic posturing, and seizures (Fig. 12.2.3). Despite giving its name to

12.2  Protein-dependent inborn errors of metabolism 1955 the disease, the characteristic odour may be absent. Neuroimaging shows localized or diffuse generalized cerebral oedema. Convulsions appear regularly and electroencephalography reveals abnormalities with comb-​like rhythms (5–​9 Hz) of spindle-​like sharp waves over the central regions and multiple shifting spikes and sharp waves with suppression bursts. Untreated patients succumb within a few days. Prominent neuropathological signs of untreated maple syrup urine disease are cerebral atrophy, including neuron loss in pontine nu- clei and the thalamus and myelin deficiency; spongy degeneration and astrocytic hyperplasia occur. Hypodensities may be present in globus pallidus and thalamus. In a few patients, mostly with inter- mittent or intermediate variants, the metabolic defect can be cor- rected by thiamine (‘thiamine-​responsive’ variant). Effective doses vary from 10 mg up to 300 mg per day. Diagnosis Maple syrup urine disease is strongly suggested when an odour of maple syrup is present (most noticeably in the ear wax). Immediate confirmation by positive 2,4-​dinitrophenylhydrazine testing is suf- ficient justification to initiate treatment in families at high risk. Diagnosis is confirmed by detection of increased plasma con- centrations of leucine, isoleucine, and valine and/​or by increased urinary excretion of α-​keto and hydroxy acids. The detection of l-​alloisoleucine is diagnostic. Reduced enzyme activity of the branched-​chain α-​keto acid dehydrogenase complex in leuco- cytes, lymphoblasts, cultured fibroblasts, or amniocytes confirms the diagnosis. Except for the common Mennonite mutation, the 2-Oxoisocaproate 2-OH-isocaproate 3-OH-isovalerate 3-Methylcrotonyl- glycine 3-OH-isovalerate Isovalerylglycine Isoleucine 2-Oxo- 3-methylvalerate 2-Methylbutyryl-CoA Tiglyl-CoA Aminotransferase BCKDH MBD Hydratase Valine 2-Oxo- isovalerate Isobutyryl-CoA Methylacrylyl-CoA Aminotransferase BCKDH Hydratase Leucine 2-Oxo- isocaproate Isovaleryl-CoA 3-Methyl- crotonyl-CoA Aminotransferase BCKDH IVD MCC MHBD Deacylase Hydratase 3-Oxothiolase DH HMG-CoA lyase 2-Methyl- 3-OH-butyryl-CoA 3-OH-isobutyryl-CoA 3-Methyl- glutaconyl-CoA 2-Methyl- acetoacetyl-CoA 3-OH-isobutyrate 3-OH-3-methyl- glutaryl-CoA Methylmalonate semialdehyde IBD Propionyl-CoA Acetyl-CoA Acetoacetate Tiglyl- glycine 3-OH-propionate Methylcitrate Methylmalonyl-CoA Succinyl-CoA Krebs cycle Mutase 2-OH- isovalerate 3-Methylglutarate Carboxylase Alloisoleucine DH Fig. 12.2.2  Metabolism of branched-​chain amino acids. BCKDH, branched chain α-​keto acid dehydrogenase (deficient in MSUD); DH, dehydrogenase; hydratase, 3-​methylglutaconyl-​CoA hydratase (deficient in 3-​methylglutaconic aciduria type I); IVD, isovaleryl-​CoA dehydrogenase (deficient in isovaleric academia); MCC, 3-​methylcrotonyl-​CoA carboxylase (deficient in methylcrotonylglycinuria); MCM, methylmalonyl CoA mutase (deficient in methylmalonic aciduria); MHBD, 2-​methyl-​3-​hydroxybutyryl-​CoA dehydrogenase (deficient in 2-​methyl-​3-​hydroxybutyryl-​CoA dehydrogenase deficiency); PCC, propionyl-​ CoA carboxylase (deficient in propionic aciduria). Accumulating pathologic metabolites are shown in italics. Source data from Zschocke J, Hoffmann GF (2011). Vademecum metabolicum. Manual of metabolic paediatrics, 3rd edition. Schattauer, Stuttgart. Fig. 12.2.3  Opisthotonic hypertonic comatose infant with maple syrup urine disease.

SECTION 12  Metabolic disorders 1956 molecular genetics of maple syrup urine disease are too complex for diagnostic use. Prenatal testing is available by enzymatic ana- lysis of amniotic cells. Treatment and outcome Emergency treatment aims to reduce branched-​chain amino acids, particularly leucine. To induce anabolism, high calorie intake is re- quired. Most importantly, glucose stimulates endogenous insulin secretion activating protein synthesis. If required, insulin should be started early. In parallel, supplements free of branched-​chain amino acids should be administered by nasogastric drip feeding. Since low plasma concentrations of isoleucine and valine limit protein syn- thesis, cautious supplementation to decrease leucine concentrations is mostly required. Extracorporeal detoxification (haemodialysis, haemofiltration) may be required if leucine exceeds 20 mg/​dl (1500 µmol/​litre). Liver transplantation may be considered a reasonable treatment option for patients with classic maple syrup urine disease. The decision of medical treatment versus transplantation, however, is very complex and must be reached for each patient individually. Long-​term treatment of maple syrup urine disease is based on dietary restriction of branched-​chain amino acids and supplemen- tation of thiamine, if proven beneficial. Management requires close and lifelong regulation of diet. Children with the classic form of maple syrup urine disease have a satisfactory prognosis only if they are diagnosed and treated before symptom onset; for this reason MS/​MS-​based newborn screening has been introduced in some countries. Isovaleric aciduria Aetiology/​pathophysiology Isovaleric aciduria was described by Tanaka in 1966. It is caused by deficiency of isovaleryl-​CoA dehydrogenase, an enzyme located proximally in the catabolic pathway of the essential branched-​chain amino acid leucine (Fig. 12.2.2). The encoding IVD gene is local- ized on 15q14–​q15. Due to the metabolic block, isovaleryl-​CoA accumulates, and the pathognomonic metabolite isovalerylglycine is formed by conjugation of isovaleryl-​CoA to the amino group of glycine through the activity of the mitochondrial enzyme glycine- ​N-​acylase. It is suggested that accumulating acyl-​CoA esters sequester CoA, thereby disturbing energy metabolism. Specifically, isovaleryl-​ CoA inhibits pyruvate dehydrogenase and N-​acetylglutamate syn- thase causing lactic acidosis and hyperammonaemia. Furthermore, isovaleric acid inhibits granulopoiesis and occurs during metabolic decompensations. Clinical presentation Half of the patients with isovaleric aciduria present in the neonatal period with severe metabolic crises that may lead to coma and death, whereas the remainder experience chronic intermittent disease with episodes of metabolic acidosis and psychomotor retardation. Both phenotypes can occur within the same family suggesting a modi- fying role of environmental and epigenetic factors. A mild, poten- tially asymptomatic phenotype exists due to a common mutation (c.932C>T; p.A282V). This mutation was detected in one-​half of mutant alleles in patients identified by newborn screening and also in older, healthy siblings. During metabolic crises, patients present with the typical features of classic organic acid disorders, that is, acidosis, ketosis, vomiting, progressive alteration of consciousness, and, finally, overwhelming illness, deep coma, and death if not given appropriate therapy. Clinical abnormalities often develop within the first days of life. A pathognomonic foul odour reminiscent of sweaty feet, caused by isovaleric acid, occurs. Abnormalities of the haematopoietic system such as thrombocytopenia, neutropenia, or pancytopenia develop; hyperammonaemia is usually mild. In the chronic intermittent form, children slide into recur- rent metabolic crises because of a high intake of protein or minor infections inducing a catabolic state. Cytopenias develop as described earlier, and hyperglycaemia may develop, most likely due to stress-​induced counter-​regulatory hormonal ef- fects. Pancreatitis may be a complication of isovaleric aciduria. Older patients may have normal psychomotor development or mild to severe learning difficulties, depending on the frequency of decompensation and the age of diagnosis and institution of treatment. Diagnosis The clinical symptoms of isovaleric aciduria resemble other or- ganic acidaemias; even the suggestive odour may be shared by similar disorders (Boxes 12.2.1 and 12.2.2). The combination of ketoacidosis, dehydration, and hyperglycaemia has led to erro- neous diagnosis of diabetic ketoacidosis, and persistent vomiting in infancy to the wrong suggestion of hypertrophic pyloric sten- osis and unnecessary surgery. A reliable way to accomplish the diagnosis is quantitative analysis of urinary organic acids and acylglycines by GC/​MS or the analysis of acylcarnitine profiles by MS/​MS. During metabolic decompensation, the urinary organic acid profile reveals high excretion of isovalerylglycine which remains elevated. 3-​Hydroxyisovaleric acid only increases during metabolic decom- pensation. Isovalerylcarnitine is the characteristic acylcarnitine of this disease and its urinary excretion increases following supple- mentation with l-​carnitine. The diagnosis of isovaleric aciduria can be confirmed by enzyme analysis in fibroblasts or mutation analysis in specialized laboratories. Several methods have been successfully used for prenatal diagnosis including stable isotope dilution ana- lysis of isovalerylglycine, MS/​MS detection of isovalerylcarnitine in amniotic fluid, or macromolecular labelling from (1-​14C)-​isovaleric acid in cultured amniocytes. Molecular diagnosis is only available in a research setting. Treatment and outcome Total natural protein intake is restricted according to the patient’s leucine tolerance and is adjusted to age-​specific requirements. To provide a complementary source of the other amino acids, a leucine-​ free formula is available. Beyond childhood, a protein-​restricted diet allowing a moderate restriction of leucine intake is usually suf- ficient. In addition, urinary excretion of isovaleryl-​CoA as nontoxic carnitine conjugates is activated by supplementation with carnitine (50–​100 mg/​kg per day). During acute decompensation, isovaleric aciduria is treated fol- lowing the general principles for other organic acid disorders (see ‘Emergency treatment’).

12.2  Protein-dependent inborn errors of metabolism 1957 Aspirin is contraindicated in patients with isovaleric aciduria be- cause salicylic acid is a competing substrate for glycine-​N-​acylase, interfering with isovalerylglycine synthesis. Most children will survive the first life-​threatening episode if correct treatment is set in place early. If effective treatment can be installed before any severe metabolic decompensation, it will sig- nificantly improve outcome. Therefore, in some countries isovaleric aciduria is screened for in newborns using MS/​MS. 3-​Methylcrotonylglycinuria 3-​Methylcrotonylglycinuria is an inborn error of leucine catab- olism due to deficiency of 3-​α-​methylcrotonyl-​CoA carboxylase (Fig. 12.2.2). It appears to be the most frequent inborn organic acid disorder, with a frequency of 1 in 50 000 newborns. The 3-​methylcrotonylglycinuria enzyme requires biotin as a cofactor, and the isolated enzymatic defect must be differentiated from pri- mary deficiencies in the biotin pathway (see ‘Biotinidase deficiency’ and ‘Holocarboxylase synthetase deficiency’). As a consequence of 3-​methylcrotonylglycinuria deficiency, 3-​hydroxyisovaleric acid, 3-​hydroxyisovalerylcarnitine, 3-​methylcrotonylcarnitine, and 3-​methylcrotonylglycine accumulate. Clinical presentation From the follow-​up of individuals identified by newborn screening it has become evident that deficiency of 3-​methylcrotonylglycinuria is a genetic condition with low clinical expressivity and penetrance, representing largely (c.90%) a nondisease. Less than 10% of affected individuals may develop mostly mild neurological symptoms which are often not clearly attributed to 3-​methylcrotonylglycinuria de- ficiency. However, a few patients may develop acute metabolic de- compensation (ketoacidosis, hypoglycaemia, hyperammonaemia, Reye-​like syndrome) precipitated by febrile illness during infancy; this may be fatal if untreated. Diagnosis The diagnosis is confirmed biochemically by identification of 3-​hydroxyisovaleric acid and 3-​methylcrotonylglycine in urine (GC/​MS) or 3-​hydroxyisovalerylcarnitine in dried blood spots or plasma (MS/​MS whereas in patients with additionally increased 3-​hydroxypropionic, methylcitric, or lactic acids multiple carb- oxylase deficiency or biotinidase deficiency should be considered. In particular, 3-​hydroxyisovalerylcarnitine concentrations which spontaneously decrease to normal values in follow-​up inves- tigations of any neonate should prompt the investigation of 3-​ methylcrotonylglycinuria deficiency in the mother. Significantly reduced enzyme activity in fibroblasts or leucocytes or mutation analysis confirms the diagnosis. It is important to ex- clude multiple carboxylase deficiency by demonstrating normal en- zyme activities of propionyl-​CoA carboxylase, pyruvate carboxylase, as well as biotinidase. Prenatal diagnosis is possible by stable isotope dilution analysis of amniotic fluid or by enzymatic and molecular analyses in cultivated amniocytes or chorionic villi. Treatment and outcome Most affected individuals do not require specific treatment, with the exception of carnitine supplementation if secondary carnitine depletion is found. However, moderate protein restriction and ad- ministration of leucine-​free amino acid supplements has been tried. 3-​Methylcrotonylglycinuria is usually unresponsive to biotin whereas, in those with the p.R385S mutation, biotin responsiveness has been reported. If acute metabolic decompensation occurs, af- fected patients are treated as with other organic acid disorders (see ‘Emergency treatment’). Most affected individuals remain asymp- tomatic without specific treatment and thus the benefit of newborn screening remains to be elucidated. 3-​Methylglutaconic acidurias Increased urinary excretion of 3-​methylglutaconic acid is the biochemical hallmark of a heterogeneous group of inborn errors termed 3-​methylglutaconic acidurias types including a primary defect in leucine catabolism, primary mitochondrial disorders, for example, Pearson’s syndrome and ATP synthase deficiency, and patients with Smith–​Lemli–​Opitz syndrome, a cholesterol biosyn- thesis disorder. Whereas in 3-​methylglutaconyl-​CoA hydratase de- ficiency elevated 3-​methylglutaconic is caused by a primary defect in leucine degradation, in all other diseases with 3-​methylglutaconic aciduria the increase of this metabolite is thought to be secondary to mitochondrial membrane biosynthesis, maintenance, and phospholipid remodelling or disturbed cholesterol biosynthesis. Interestingly, leucine degradation is linked to cholesterol biosyn- thesis via the Popjak shunt and the 3-​hydroxy-​3-​methylglutaryl-​ CoA salvage pathway. With recognition of an increasing number of underlying defects in recent years, the initial nomenclature of 3-​ methylglutaconic acidurias has been revised. In the following, both old (type I–​V) and new nomenclature (specifying the syndrome and affected gene) are given. Primary 3-​methylglutaconic aciduria 3-​Methylglutaconic aciduria type I Aetiology/​pathophysiology  3-​Methylglutaconic aciduria type I  is caused by deficiency of 3-​methylglutaconyl-​CoA hydratase (Fig. 12.2.2) required for the conversion of 3-​methylglutaconyl-​ CoA to 3-​hydroxy-​3-​methylglutaryl-​CoA in leucine catabolism. The hydratase is identical to an RNA-​binding protein (designated AUH) possessing enoyl-​CoA hydratase activity. The defect leads to an accumulation of 3-​methylglutaconic, 3-​methylglutaric, and 3-​hydroxyisovaleric acids. Clinical presentation  The clinical phenotype of affected individ- uals is variable and also includes an asymptomatic disease course. Patients present with neurological symptoms including delayed speech and motor development. Metabolic decompensation with hypoglycaemia and metabolic acidosis is rare but can occur fol- lowing a catabolic state. The recent discovery of the disorder in adult-​onset patients with slowly progressive ataxia, dementia, and leukoencephalopathy may point to the long-​term nature and mani- festations of this disease. Diagnosis  Urinary excretion of large amounts of 3-​methylglutaconic, 3-​methylglutaric, and 3-​hydroxyisovaleric acids but normal excretion of 3-​hydroxy-​3-​methylglutaric acid points to hydratase deficiency. Increased 3-​hydroxyisovalerylcarnitine is a hint for either type of 3-​methylglutaconic aciduria. The definitive diagnosis is made by en- zyme analysis in fibroblasts or by mutation analysis. Treatment and outcome  The need for treatment has not been es- tablished, especially for dietary treatment. The outcome appears

SECTION 12  Metabolic disorders 1958 favourable as a significant number of untreated patients have never developed symptoms. Secondary 3-​methylglutaconic acidurias TAZ defect or Barth’s syndrome (formerly,
3-​methylglutaconic aciduria type II) Aetiology/​pathophysiology  The molecular basis of Barth’s syn- drome is deficiency of tafazzin which is localized in the inner mitochondrial membrane affecting phospholipid metabolism, in particular cardiolipin. The origin of elevated levels of 3-​ methylglutaconic and 3-​methylglutaric acids in Barth’s syndrome is unknown. The identification of the causative gene allowed the retrospective classification of different families labelled in the past as X-​linked endocardial fibrosis, severe X-​linked cardiomyopathy, or Barth’s syndrome. All these entities have been shown to share the same mo- lecular pathology. Clinical presentation  In 1983, Barth and colleagues described an X-​linked neuromuscular disease characterized by dilated cardio- myopathy, skeletal myopathy, retarded growth, and neutropenia. Patients may present at birth or during the first weeks of life, usually with congestive cardiac failure. With long-​standing cardiac disease, endocardial fibroelastosis may develop. Delayed gross motor mile- stones, myopathic facies, a waddling gait, and a positive Gower’s sign are common. Occasionally patients may show moderate lactic acidosis. Postnatal growth retardation may be severe, and beyond 2 years of age patients are usually very stunted but with normal head circumferences. Diagnosis  Barth’s syndrome should be considered in any male presenting with dilated cardiomyopathy. If neutropenia, idiopathic myopathy, and growth retardation are also present, the diagnosis of Barth’s syndrome is almost certain. Biochemically, increased 3-​ methylglutaconic acid is usually found in urine but is not a constant feature. 2-​Ethylhydracrylic acid may be also elevated. Muscle dis- ease and lactic acidaemia may initiate a work-​up for mitochondrial disorders. Muscle biopsy may reveal involvement of deficient re- spiratory chain complexes I and IV. The diagnosis is confirmed by cardiolipin analysis in thrombocytes or mutation analysis. Mutation analysis makes prenatal diagnosis now available. Treatment and outcome  Children affected by Barth’s syndrome need to be carefully managed mainly by expert cardiologists; im- munologists and neurologists should also be involved. Cardiac ar- rhythmias carry a poor prognosis and may require implantation of an internal cardiac defibrillator. Successful heart transplantation has been carried out. Due to increased susceptibility to severe bacterial infections, infectious diseases need to be treated promptly and ag- gressively. Protein restriction and carnitine supplementation has been employed with unclear benefit. About 25% of patients with Barth’s syndrome succumb during infancy and early childhood due to cardiac complications or overwhelming bacterial infections. OPA3 defect or Costeff’s syndrome (formerly,
3-​methylglutaconic aciduria type III) Aetiology/​pathophysiology  Costeff’s syndrome is caused by mu- tations in the OPA3 gene resulting in a defect of a putative mito- chondrial protein with yet unknown function. The origin of elevated levels of 3-​methylglutaconic and 3-​methylglutaric acids is also un- known. So far the disorder has only been reported in Iraqi Jews. Clinical presentation  The determining clinical presentation is early-​onset optic atrophy, which may be accompanied by nystagmus. In later childhood or adolescence, patients may develop extrapyr- amidal signs and moderate cognitive impairment. In about one-​half of the patients, spasticity develops and progresses over years. Diagnosis  Costeff’s syndrome should be suspected in patients pre- senting with early-​onset optic atrophy if additional neurological symptoms develop.  3-​Methylglutaconic aciduria is a biochemical indicator of Costeff’s syndrome, which may now be proven by mo- lecular analysis. Treatment and outcome  Effective treatment has not been reported. Treatment is symptomatic and focuses on the prevention of disabil- ities due to progressive spasticity. The disease appears stationary but the long-​term outcome is unknown. Specified and not otherwise specified 3-​methylglutaconic aciduria type IV Aetiology/​pathophysiology  3-​Methylglutaconic aciduria type IV is undoubtedly the most heterogeneous and increasing group of 3-​methylglutaconic acidurias. As unexplained 3-​methylglutaconic aciduria (i.e. type IV) was also found incidentally in asymptomatic adults, it appears doubtful that this biochemical feature by itself is of pathophysiological relevance. Diagnosis  Patients are identified by elevated urinary con- centrations of 3-​methylglutaconic and 3-​methylglutaric acids. Classification of type IV methylglutaconic aciduria is made by ex- clusion of known causes of 3-​methylglutaconic aciduria (see other subsections), primary mitochondrial disorders (e.g. Pearson’s syn- drome), and Smith–​Lemli–​Opitz syndrome. Four clinical pheno- typic groups have been delineated including patients with an encephalopathic, hepatocerebral, cardiomyopathic, and myopathic disease form. Genetic testing in these patients has elucidated a high number of disease-​causing mutations in the POLG1, SUCLA2, TMEM70, and RYR1 genes which have been associated with known diseases. Furthermore, MEGDEL syndrome due to mutations in the SERAC1 gene was recently described. Based on this, the mo- lecular cause of 3-​methylglutaconic aciduria type IV can be iden- tified in many patients. In patients with a known molecular basis of 3-​methylglutaconic aciduria, 3-​methylglutaconic aciduria type IV shall be replaced by a more specific terminology, for example, SERAC1 defect or MEGDEL syndrome, or TMEM70 defect. Treatment and outcome  No effective treatment has been re- ported. Treatment is symptomatic and focuses on the prevention of neurological deterioration. The identification of type IV 3-​ methylglutaconic aciduria is not yet of prognostic relevance. DNAJC19 defect or dilated cardiomyopathy with
ataxia (DCMA) syndrome (formerly, 3-​methylglutaconic aciduria type V) Aetiology/​pathophysiology  Another type of 3-​methylglutaconic aciduria has recently been elucidated in 18 patients of the Canadian Dariusleut Hutterite population. It is an autosomal recessive con- dition caused by a mutated DNAJC19 gene. Proteins of the DNAJ

12.2  Protein-dependent inborn errors of metabolism 1959 domain are involved in molecular chaperone systems, DNAJ19 having been localized to the inner mitochondrial membrane. Clinical presentation  Inherited DNAJ19 deficiency leads to clin- ical presentation which initially resembles Barth’s syndrome (3-​ methylglutaconic aciduria type II) with early-​onset severe dilated (or noncompaction) cardiomyopathy with conduction defects. However, it also leads to with nonprogressive cerebellar ataxia, tes- ticular dysgenesis, and growth failure. Diagnosis  The diagnosis can be made by genetic testing in patients with a suggestive clinical presentation and 3-​methylglutaonic aciduria. Therapy and outcome  No effective treatment has been reported. Treatment is symptomatic and focuses on the prevention of cardiac deterioration. 2-​Methyl-​3-​hydroxybutyryl-​CoA dehydrogenase deficiency Aetiology/​pathophysiology 2-​Methyl-​3-​hydroxybutyryl-​CoA dehydrogenase deficiency is a rare cerebral organic acid disorder. This mitochondrial enzyme is involved in the catabolism of isoleucine and branched-​chain fatty acids (Fig. 12.2.2). Retrospectively, patients were misdiagnosed as having 3-​oxothiolase deficiency until Zschocke and colleagues (2000) recognized the separate distinct clinical and biochemical presentation. Inheritance is X-​chromosomal semidominant (fe- males may be symptomatic). Disease-​causing mutations were iden- tified in the HSD17B10 gene. The pathophysiology of this disease is unknown. The enzyme is identical to an amyloid β-​peptide-​binding protein which is implicated in Alzheimer’s disease. Clinical presentation 2-​Methyl-​3-​hydroxybutyryl-​CoA dehydrogenase deficiency mostly results in a progressive neurodegenerative disease. Regression usu- ally becomes obvious in late infancy or early childhood but is vari- able. Affected boys usually develop truncal hypotonia with spasticity of the limbs, dyskinesia and athetosis, a horizontal nystagmus, and retinal blindness. Motor and mental skills are completely lost, as are sensory modalities. Epilepsy is frequently found and is usually dif- ficult to treat. When hypertrophic cardiomyopathy was diagnosed, deterioration was rapid with death due to progressive heart failure. Neuroimaging documents progressive generalized atrophy, basal ganglia injury, periventricular white matter abnormalities, and oc- cipital infarctions in individual cases. Heterozygous female patients may be asymptomatic or may have variable stationary psychomotor retardation with impaired hearing. Diagnosis The disease should be considered in children presenting with early-​onset progressive encephalopathy, especially if X-​linked in- heritance is suggested. The biochemical hallmark of this disease is increased urinary excretion of 2-​methyl-​3-​hydroxybutyric acid and tiglylglycine. Elevations of 2-​ethylhydracrylic acid and 3-​ hydroxyisobutyric acid in urine may also be found. These abnor- malities may be subtle. Treatment and outcome No effective rational treatment is known. Care of patients with this disease should repeatedly entail (1) assessment of muscle and cardiac function, (2)  neurological examination including elec- troencephalography and MRI, and (3) assessment of visual and hearing system. The prognosis is mostly poor, with death in early childhood. Propionic aciduria Aetiology/​pathophysiology In 1961, Childs and coworkers described the index patient with pro- pionic aciduria. Since ketosis and hyperglycinaemia were the bio- chemical hallmarks recognized, the disorder was lumped together with methylmalonic acidurias as ‘ketotic hyperglycinaemia’ to dis- tinguish it from nonketotic hyperglycinaemia. Implementation of GC/​MS analysis to metabolic diagnostic work-​up allowed the differentiation of these disorders in the 1970s. Propionic aciduria is caused by an autosomal recessive inherited deficiency of biotin-​ dependent duodecameric propionyl-​CoA carboxylase, the first step in propionate metabolism, in which propionyl-​CoA is converted to methylmalonyl-​CoA (Fig. 12.2.2). Over 100 disease-​causing muta- tions have been identified in the PCCA gene (13q32) and the PCCB gene (3q21–​22). Propionyl-​CoA is formed from the catabolism of isoleucine, threonine, methionine, valine, odd-​numbered fatty acids, and the side chain of cholesterol, and from gut bacteria. Deficiency of propionyl-​CoA carboxylase gives rise to accumulation of propionyl-​CoA and metabolites of alternative propionate oxidation such as 2-​methylcitric acid, 3-​hydroxypropionic acid, tiglic acid, propionylcarnitine, and propionylglycine. All of these can be de- tected and quantified by GC/​MS (urine, plasma) or MS/​MS (dried blood spots, plasma). Elevated propionyl-​CoA and its pathological derivatives inter- fere with a variety of metabolic pathways including inhibition of (1)  the glycine cleavage enzyme resulting in hyperglycinaemia, (2)  N-​acetylglutamate synthase resulting in hyperammonaemia, and (3) pyruvate dehydrogenase complex as well as several enzymes of the tricarboxylic acid cycle resulting in lactic acidaemia and hyperketosis, and severe impairment of energy metabolism. Clinical presentation Propionic aciduria usually presents with severe neonatal meta- bolic decompensation characterized clinically by multiorgan failure and biochemically by hyperammonaemia, metabolic acidosis, hyperketosis, lactic acidaemia, hyperglycinaemia, and hyperalaninaemia. Propionic aciduria may be misinterpreted as sepsis or ventricular haemorrhage. Acute metabolic decompen- sation and long-​term complications usually involve organs with a high energy demand, including the brain, heart and skeletal muscle, liver, and bone marrow. Frequent signs and symptoms are failure to thrive, microcephaly, mild to severe motor disabilities and learning difficulties, truncal hypotonia, extrapyramidal symptoms (dystonia, chorea), seizures, cardiomyopathy, myopathy, hepatomegaly, acute or chronic pancreatitis, leucopenia, thrombocytopenia, anaemia, or pancytopenia, whereas renal complications are uncommon. Metabolic decompensations in infancy or childhood are similar to those in the neonatal period. The first symptom is often vomiting; this has led to erroneous diagnosis of pyloric stenosis or duodenal obstruction, resulting in a number of pyloromyotomies or other explorations. Basal ganglia injury, mostly affecting the putamen,

SECTION 12  Metabolic disorders 1960 occurs (Fig. 12.2.4); generalized cerebral atrophy and white matter disease is common. A small subgroup of patients exhibit almost exclusively encephal- opathy and progressive neurological disease, resembling a lysosomal storage disorder. A milder form of propionic aciduria reported in Japan manifests from childhood with mild learning difficulties or extrapyramidal symptoms, and only occasionally with metabolic acidosis. Finally, some individuals remain asymptomatic until teenage years and are identified during family studies. Diagnosis The method of diagnosis is GC/​MS analysis of organic acids (urine) or MS/​MS analysis of acylcarnitines (dried blood spots, plasma, urine). Characteristic metabolites are 2-​methylcitric acid, 3-​hydroxypropionic acid, tiglic acid, propionylglycine, and propionylcarnitine. The absence of methylmalonic acid excludes methylmalonic acidurias, and the absence of β-​hydroxyisovaleric acid and β-​methylcrotonylglycine rules out multiple carboxylase de- ficiency. In plasma and urine, increased concentrations of glycine and ketone bodies may be present. Confirmation of diagnosis is made by enzyme analysis in leucocytes or fibroblasts, or by mutation analysis. Prenatal diagnosis can be made by mutation analysis, enzyme ana- lysis, or quantitative GC/​MS analysis of 2-​methylcitric acid. Treatment and outcome Prevention of metabolic decompensation is the most important de- terminant of outcome. During acute decompensation, propionic aciduria is treated like other organic acid disorders (see ‘Emergency treatment’). Long-​term treatment is based on lifelong dietary restric- tion of the precursors isoleucine, valine, methionine, and threonine, as well as by supplementation with l-​carnitine. As significant propi- onate production occurs in the gut, intermittent decontamination (10–​14 days/​month) with metronidazole or colistin as well as meas- ures preventing constipation are often used. Some patients exhibit recurrent or almost chronic hyperammonaemia, especially during infancy. This may necessitate additional supplementation with ar- ginine or citrulline and/​or administration of sodium benzoate or phenylbutyrate. However, benzoate treatment may aggravate the depletion of free carnitine and CoA. Biotin responsiveness in propi- onic aciduria is very rare, if present at all. More than 20 children with propionic aciduria have undergone orthotopic liver transplantation, but the outcome is mixed. Auxiliary as well as living-​related liver transplantations have been successfully performed, but liver trans- plantation in propionic aciduria seems to be more complicated than in patients with urea cycle defects. Patients with neonatal onset of symptoms still have a poor out- come. Patients with late onset of symptoms reach adulthood but often have physical and mental disabilities; nonetheless, some patients can survive to adulthood with normal intellects. The phenotypic vari- ation of patients with propionic aciduria as well as evidence-​based recommendations for diagnosis, treatment, and follow-​up have re- cently been reported by an international consortium of experts. Methylmalonic aciduria Aetiology/​pathophysiology Methylmalonic aciduria is the biochemical hallmark of a heteroge- neous group of inborn metabolic errors with a cumulative prevalence of at least 1 in 100 000 newborns in Europe. Index patients were first described in 1967 by Oberholzer and Stokke. This section focuses on isolated methylmalonic aciduria caused by mutations in the MUT gene localized on 6p21 encoding the apoenzyme methylmalonyl-​ CoA mutase. Methylmalonyl-​CoA mutase can alternatively be im- paired by defects in the biosynthesis of 5′-​deoxyadenosylcobalamin, deficient cobalamin transport, or by acquired cobalamin deficiency as in pernicious anaemia. In infancy, severe progressive disease may develop in breastfed infants of mothers who have (undiagnosed) pernicious anaemia or adhere to a strict vegan diet. Methylmalonic acid is a more reliable index of body stores of cobalamin than cobalamin levels in blood. d-​Methylmalonyl-​CoA is formed in propionate metabolism by carboxylation of propionyl-​CoA. l-​Methylmalonyl-​CoA is formed from d-​methylmalonyl-​CoA by d-​methylmalonyl-​CoA racemase and, subsequently, is converted to succinyl-​CoA by the dimeric 5′-​deoxyadenosylcobalamin-​dependent mitochondrial enzyme methylmalonyl-​CoA mutase (Fig. 12.2.2). As with propionic aciduria (see ‘Propionic aciduria’), impairment of energy metabolism by propionyl-​CoA and 2-​methylcitric acid plays a key role in the pathophysiology of methylmalonic acidurias, resulting in multiorgan failure. In addition, methylmalonic acid may exert additional toxic effects. Clinical presentation Patients with severe methylmalonyl-​CoA mutase deficiency (mut0) usually present with neonatal metabolic crises which are clinically Fig. 12.2.4  Transverse MRI image of a 7-​year-​old girl, who had been diagnosed with propionic aciduria in infancy and had been successfully treated since then. While in good metabolic control, she suddenly became comatose. Massive infarction of the basal ganglia had occurred, and the child died a few days later. Spin echo technique. Courtesy of Drs R. Haas and W.L. Nyhan, Department of Pediatrics, University of California, San Diego, USA.

12.2  Protein-dependent inborn errors of metabolism 1961 and biochemically (except for methylmalonic acid) indistinguish- able from those of patients with propionic aciduria. In patients with residual methylmalonyl-​CoA mutase activity (mut−), the onset of symptoms is more variable. Neonatal onset of symptoms is found as is a chronic intermittent form, that is, precipitation of recurrent metabolic crises in infancy and children following a high intake of protein or a catabolic state. Long-​term complications are frequent, in particular in mut0 patients. These include failure to thrive, chronic neurological symptoms such as extrapyramidal movement disorder, motor disabilities, learning difficulties, and epilepsy, cardiomyop- athy, myopathy, and pancreatitis. Neuroradiological studies demon- strate lesions of globus pallidus, generalized cerebral atrophy, and white matter disease. The development of chronic renal failure in a large proportion of patients appears inevitable. Diagnosis A reliable way to make the diagnosis is GC/​MS analysis of urinary organic acids or MS/​MS analysis of acylcarnitines showing ele- vated concentrations of methylmalonic acid as well as of metab- olites of alternative propionate oxidation (e.g. propionylglycine, 3-​hydroxypropionic acid, 2-​methylcitric acid, propionylglycine, and propionylcarnitine; as in propionic aciduria). These biochemical abnormalities have a considerable interday and intraday variation and are influenced by responsiveness to cobalamin and metabolic state. Differential diagnosis of methylmalonic aciduria is acquired cobalamin depletion or inherited cobalamin deficiencies, transient mild methylmalonic acidurias of unknown origin in infants, and methylmalonic encephalopathy due to deficiency of succinyl-​CoA synthase. Concomitant megaloblastic anaemia and an increase of plasma homocysteine indicates disturbed cobalamin metabolism as the cause of methylmalonic aciduria. Standardized criteria to define responsiveness to hydroxocobalamin are not established. The determination of methylmalonyl-​CoA mutase activity in fibroblast extracts, mutation analysis or the in- vestigation of labelled propionate incorporation following transfec- tion by a vector containing cloned mutase cDNA in intact patients’ fibroblasts may be required to differentiate primary defects of methylmalonyl-​CoA mutase (mut0, mut−) from primary defects of 5′-​ deoxyadenosylcobalamin (cblA and cblB defects). Prenatal diagnosis is available by enzyme or mutation analyses as well as by quantitative stable isotope dilution assay of 2-​methylcitric acid. Treatment and outcome Metabolic maintenance and emergency treatment follows the treat- ment principles for organic acid disorders in general and propionic aciduria in particular (see ‘Propionic aciduria’). In addition, substi- tution with cobalamin may be beneficial, since partial or complete response to cobalamin has been demonstrated (except for mut0 pa- tients). In neonates and infants, intramuscular hydroxocobalamin is required; children and adults may be treated with oral cyano- cobalamin. Chronic renal failure may progress, necessitating haemodialysis or peritoneal dialysis. Kidney transplantation has been performed in these patients. Liver transplantation can provide enzyme activity to ameliorate the metabolic defect and the idea of combined liver–​kidney or isolated liver transplantation has emerged. The benefit remains doubtful, however, as mortality is significant; in addition, liver transplantation does not reliably protect against se- vere neurological and renal complications. The phenotypic variation of patients with methylmalonic aciduria as well as evidence-​based recommendations for diagnosis, treatment, and follow-​up have re- cently been reported by an international consortium of experts. 3-​Hydroxyisobutyryl-​CoA hydrolase deficiency Aetiology/​pathophysiology 3-​Hydroxyisobutyryl-​CoA hydrolase (HIBCH) catalyses the fifth step of valine catabolism converting 3-​hydroxyisobutyryl-​ CoA to 3-​hydroxyisobutyrate and is due to biallelic mutations of the HIBCH gene which is located on 2q32.2. HIBCH de- ficiency is biochemically characterized by accumulation of 3-​hydroxyisobutyrylcarnitine deriving from 3-​hydroxyisobutyryl-​ CoA and S-​2-​carboxypropyl-​L-​cysteine and -​cysteamine deriving from methyacrylyl-​CoA. Whereas 3-​hydroxyisobutyrylcarnitine can be eliminated via urinary excretion, methylacrylyl-​CoA is a highly reactive compound which readily undergoes addition reac- tion with sulphhydryl groups. Inactivation of sulhydryl-​containing enzymes such as respiratory chain complexes and cofactors is con- sidered as the major pathomechanism. Clinical presentation So far this disease has rarely been described. Patients presented with a (Leigh-​like) mitochondrial encephalopathy starting in infancy, delayed global development, muscular hypotonia, poor feeding, and multiple malformations (dysmorphic facial features, vertebral anomalies, tetralogy of Fallot, agenesis of cingulate gyrus and corpus callosum) in one patient. Neurological symptoms are progressive. Diagnosis The diagnosis is based on metabolic tests demonstrating elevated 3-​hydroxyisobutyrylcarnitine by tandem mass spectrometry and 2-​methyl-​2,3-​dihydroxybutyric acid and 2-​hydroxyisovaleric acid by organic acid analysis. S-​2-​carboxypropyl-​L-​cysteine and -​ cysteamine can be determined by specific HPLC analysis. Enzymatic testing of the deficient enzyme in fibroblast and molecular genetic testing confirms the diagnosis. Analysis of respiratory chain en- zymes in muscle biopsy often show decreased activity of pyruvate dehydrogenase complex and/​or multiple deficiencies of respiratory chain complexes. Treatment and outcome Since this disease is thought to be caused by accumulation of toxic metabolites of the valine catabolic pathway, a low-​valine diet should be considered as a treatment option. Carnitine supplementation prevents secondary carnitine depletion. However, the efficacy of this therapeutic approach has not yet been systematically studied. Short-​chain enoyl-​CoA hydratase deficiency Aetiology/​pathophysiology Mitochondrial short-​chain enoyl-​CoA hydratase (ECHS1) is a multispecific enzyme that catalyses the hydration of chain-​shortened α,β-​unsaturated enoyl-​CoA thioesters in the β-​oxidation spiral of fatty acids as well as in the catabolic pathways of valine, isoleucine, tryptophan, and lysine. Deficiency of ECHS1, which is coded by the ECHS1 gene located on 10q26.3, induces a very similar biochem- ical phenotype as in HIBCH deficiency. In contrast to HIBCH defi- ciency, however, tiglylglycine but not 3-​hydroxyisobutyrylcarnitine

SECTION 12  Metabolic disorders 1962 is elevated. Inconsistently, there is evidence of mildly impaired mitochondrial oxidation of short-​chain fatty acids. The biochemical derangement highlights that in analogy to HIBCH deficiency, im- pairment of valine catabolism and thus accumulation of the toxic metabolite methacrylyl-​CoA is most important. Clinical presentation Patients present with (Leigh-​like) mitochondrial encephalopathy, dystonia, epilepsy, optic nerve atrophy and cardiomyopathy. Onset of symptoms is usually found in the newborn period or during in- fancy. The disease course is progressive and in its severest form it might be fatal in infancy or childhood. Diagnosis Diagnosis of this disease should be considered in patients with a combination of mitochondrial encephalopathy and cardiomyop- athy. Urinary excretion of 2-​methyl-​2,3-​dihydroxybutyric acid is determined by organic acid analysis, S-​2-​carboxypropyl-​L-​cysteine and -​cysteamine can be analysed using specific HPLC methods. Enzymatic testing of the deficient enzyme in fibroblast and mo- lecular genetic testing confirm the diagnosis. Analysis of respira- tory chain enzymes in muscle biopsy often show decreased activity of pyruvate dehydrogenase complex and/​or multiple deficiencies of respiratory chain complexes. Treatment and outcome In analogy to HIBCH deficiency, a low-​valine diet should be con- sidered a treatment option. It remains to be elucidated whether this therapeutic intervention is able to improve the disease course. Malonic aciduria Aetiology/​pathophysiology First described in 1984, very few patients with malonic aciduria have been delineated until now. Malonic aciduria is caused by malonyl-​ CoA decarboxylase deficiency leading to a disturbed fatty acid me- tabolism. Malonyl-​CoA is the first committed intermediate of fatty acid synthesis. In addition, it regulates carnitine acyltransferases among other enzymes steering fatty acid metabolism. The cytocolic enzyme is found most often in the liver, brain, heart, and skeletal muscle. Clinical presentation The clinical presentation is variable but mostly involves acute meta- bolic episodes with progressive lethargy, hypotonia, and hepato- megaly associated with metabolic acidosis. Hypoglycaemia, lactic acidosis, and/​ or mild hyperammonaemia can also develop. Cardiac involvement is present in about 40% of patients with cardiomy- opathy which can progress to cardiac failure. Other patients were identified with less specific symptoms such as developmental delay, hypotonia, seizures, and short stature. Diagnosis Urinary organic acids identify increased malonic acid, sometimes in combination with fumaric acid, malic acid, and aethylmalonic acid. During metabolic decompensations ketosis develops with elevated dicarboxylic acids. Total and free carnitine levels are reduced due to the formation of malonylcarnitine, which may allow population newborn screening. Treatment and outcome A clearly effective therapeutic regimen has not been established. Carnitine supplementation as well as a diet high in carbohydrates and low in long-​chain triglycerides improved the clinical symp- toms as well as metabolic disturbances in patients. In some patients, medium-​chain triglycerides supplements appeared helpful. Little is known about the long-​term prospects of this disorder. Patients were stable on the different treatment options at least until adolescence. Defects of lysine, hydroxylysine, and tryptophan metabolism The common catabolic pathway of lysine, hydroxylysine, and trypto- phan is summarized in Fig. 12.2.5. Hyperlysinaemia I/​hyperlysinaemia II or saccharopinuria Hyperlysinaemia/​saccharopinuria is caused by a recessive deficiency of the bifunctional protein 2-​aminoadipic semialdehyde synthase. As hyperlysinaemia/​saccharopinuria is considered a nondisease, af- fected individuals do not require specific treatment. 2-​Amino-​/​2-​oxoadipic aciduria Disease-​causing mutations in the DHTKD1 gene cause autosomal recessive 2-​amino-​/​2-​oxoadipic aciduria. This gene encodes for the E1 subunit of a 2-​oxoglutarate dehydrogenase complex-​like pro- tein in the lysine degradative pathway. Most affected individuals re- main asymptomatic, whereas others may present with variable mild neurological symptoms. Glutaric aciduria type I Aetiology/​pathophysiology Glutaric aciduria type I  was described in 1975. It occurs with an estimated frequency of 1 in 100 000 newborns, but which is Acetyl-CoA Kynurenine 2-Aminoadipic semialdehyde 2-Aminoadipic acid 2-Oxoadipic acid Glutaryl-CoA Lysine Tryptophan Saccharopine Phospho- hydroxylysine 3-OH-kynurenine 2-Oxoadipic acid 2-Amino- adipic acid Reductase Dehydrogenase Aminotransferase Dehydrogenase GCDH Crotonyl-CoA SCHAD Hydroxylysine Kinase Dioxygenase Kynureninase Cytosol Mito- chondrion Transport in/out mitochondria Fig. 12.2.5  Catabolic pathway of lysine, tryptophan, and hydroxylysine. 2-​Aminoadipic semialdehyde synthase (deficient in hyperlysinaemia/​ saccharopinuria); 2-​aminoadipate aminotransferase (deficiency has not yet been reported), 2-​oxoglutarate dehydrogenase-​like complex (deficient in in 2-​amino-​/​2-​oxoadipic aciduria); glutaryl-​CoA dehydrogenase (GCDH; deficient in glutaric aciduria type I). Source data from Zschocke J, Hoffmann GF (2011). Vademecum metabolicum. Manual of metabolic paediatrics, 3rd edition. Schattauer, Stuttgart.

12.2  Protein-dependent inborn errors of metabolism 1963 considerably higher (up to 1 in 300) in some communities (e.g. the Amish in Pennsylvania, United States of America, and the Oji-​Cree First Nations in Canada). Glutaric aciduria type I is caused by de- ficiency of flavin adenine dinucleotide-​dependent glutaryl-​CoA dehydrogenase, a mitochondrial enzyme in the catabolic pathway common to tryptophan, lysine, and hydroxylysine (Fig. 12.2.5). Glutaryl-​CoA dehydrogenase is encoded by the GCDH gene lo- calized on 19p13.2. More than 200 disease-​causing mutations have been described. There is no genotype–​phenotype correlation. As a consequence of glutaryl-​CoA dehydrogenase deficiency, glutaric, 3-​hydroxyglutaric, and (inconsistently) glutaconic acids as well as glutarylcarnitine accumulate. The limited permeability of the blood–​brain barrier to dicarboxylic acids (such as glutaric acid) leads to their accumulation in the brain (trapping hypothesis). Some of these metabolites are neurotoxins. Candidate mechanisms are stimulation of excitotoxic cell damage via activation of N-​methyl-​d-​ aspartate receptors, and inhibition of 2-​oxoglutarate dehydrogenase and the dicarboxylate shuttle between astrocytes and neurons. Clinical presentation Newborns are often asymptomatic but may present with transient and subtle neurological symptoms such as truncal hypotonia or asymmetric posturing. (Progressive) macrocephaly occurs in 75% of patients. Neuroimaging in infancy often reveals hypoplasia of the temporal pole, subependymal pseudocysts, and delayed myelin- ation; subdural fluid collections may be found which may be mis- taken as nonaccidental trauma. The prognostically relevant event of glutaric aciduria type I is the onset of an acute encephalopathic crisis which is usually precipi- tated by a catabolic state (e.g. febrile illness) during infancy and early childhood. Encephalopathic crises characteristically result in acute striatal injury and, subsequently, dystonia. Approximately 15% of patients with glutaric aciduria type I follow a chronic disease course and develop the same neurological symptoms as the acutely injured children over the first 2 years of life without overt crisis (insidious-​ onset variant) or during adolescence/​adulthood presenting with leukoencephalopathy (late-​onset variant). Asymptomatic indi- viduals occur occasionally. Neuroradiological abnormalities are frequently found, including widening of the sylvian fissure due to re- duced opercularization (Fig. 12.2.6a), ventriculomegaly and striatal lesions which develop after the encephalopathic crisis (Fig. 12.2.6b), and leukoencephalopathy which is mostly periventricular but may also affect subcortical U fibres (Fig. 12.2.6c). Diagnosis Glutaric aciduria type I should be suspected in patients with macro- cephaly and an extrapyramidal movement disorder starting in in- fancy or childhood. The diagnostic process can be guided by further clinical features. Diagnosis is ascertained by GC/​MS detection of glutaric and 3-​hydroxyglutaric acids in organic acid analysis (urine, plasma, or cerebrospinal fluid) or by MS/​MS detection of elevated glutarylcarnitine (dried blood spots, plasma, urine). Confirmation by enzymatic analysis in leucocytes or fibroblasts or demonstration of two pathogenic mutations is advisable. A subgroup of patients presents with a mild biochemical phenotype (low excretors) and thus may be missed if diagnostic work-​up does not include quan- titative methods (e.g. stable isotope dilution assay). Examination of the carnitine status usually reveals low total and free carnitine. Prenatal diagnosis is possible by determining glutaric acid with stable isotope dilution techniques and by enzymatic and/​or mo- lecular testing. Treatment and outcome The principal aim of treatment is the prevention of encephalopathic crises and neurological deterioration. Strict adherence to the emer- gency protocol is especially important (see ‘Emergency treatment’). During the vulnerable period (i.e. until age 6 years), lysine-​restricted dietary treatment (including lysine-​free amino acid supplements) and carnitine supplementation is recommended. Riboflavin is widely used but is of doubtful benefit. Treatment efficacy of movement dis- orders is still poor. Baclofen, benzodiazepines, and trihexyphenidyl are widely used to treat dystonia. Botulinum toxin and intrathecal baclofen are valid additions. If patients are diagnosed while they are asymptomatic, treatment prevents brain degeneration in the majority of patients. Notably, best outcome results (≥90% remain healthy) were achieved for patients following international guide- line recommendations including a low-​lysine diet and carnitine supplementation for maintenance treatment and immediate emer- gency treatment during any putatively threatening episode such as intercurrent infectious diseases. Deviation from this combined metabolic treatment increases the risk of motor disability such as in untreated patients. More than 90% of untreated patients are thought to develop neurological disabilities. Life expectancy is markedly re- duced following the manifestation of dystonia. Hyperornithinaemia (ornithine-​5-​aminotransferase): gyrate atrophy Autosomal recessive hyperornithinaemia associated with gyrate atrophy of the choroid and retina is caused by deficiency of ornithine-​5-​aminotransferase. Clinical presentation Progressive myopia is the first clinical symptom, followed by pro- gressive chorioretinal degeneration with night blindness starting late in the first decade. Loss of peripheral vision proceeds to tunnel vision and eventually blindness by the third or fourth decade. The principal abnormality is an atrophy of choroid and retina. Cataracts also develop but optic discs, cornea, and iris remain normal. A few patients develop mild proximal muscle weakness. Diagnosis Severe isolated hyperornithinaemia is usually discovered by amino acid analysis with plasma ornithine concentrations ranging from 400 to 1400 µmol/​litre (normal <200 µmol/​litre). The disease can be confirmed enzymatically by decreased activity of ornithine-​ 5-​aminotransferase in fibroblasts as well as by identification of disease-​causing mutations in the OAT gene, but the diagnosis is usu- ally evident. Treatment and prognosis Permanent reduction of plasma ornithine into the normal range (<200 µmol/​litre) is required to stop or at least slow chorioretinal degeneration. Only a small proportion of patients respond to pharmacological doses of the ornithine-​5-​aminotransferase co- factor pyridoxine. Additional therapeutic approaches to reduce or- nithine are the augmentation of renal losses by administration of

SECTION 12  Metabolic disorders 1964 pharmacological doses of l-​lysine or α-​aminoisobutyric acid (which is not metabolized), or substrate deprivation by dietary arginine re- striction. Combined treatment appears to be necessary since no single therapy is unequivocally effective. Multiple carboxylase deficiency The water-​soluble vitamin biotin is a cofactor of four important carboxylases that take part in gluconeogenesis, fatty acid synthesis, and the catabolism of several amino acids and odd-​chain fatty acids (Fig. 12.2.7). The covalent binding of biotin with apocarboxylases forming the active holocarboxylases is catalysed by biotin holocarboxylase synthetase. In the biotin cycle, biotin is recycled after proteolytic degradation of holocarboxylases (Fig. 12.2.8). Biotin in small amounts is widely present in natural foods. Within the body, biotin bound to holocarboxylases represents the major source. In dietary and in endogenous sources, biotin is protein-​bound as (a) (b) (c) Fig. 12.2.6  (a) Axial T2-​weighted MRI spin echo image of a 2½-​year-​old boy with glutaryl-​CoA dehydrogenase deficiency. He was diagnosed neonatally, never suffered an encephalopathic crisis, and developed no major neurological deficit. Extension of sylvian fissures which was mild during early infancy had slowly regressed. He did not develop characteristic frontotemporal atrophy and showed a normal myelination. (b) Axial T2-​weighted spin echo image of a 15-​month-​old boy with glutaryl-​CoA dehydrogenase deficiency 2 weeks after acute encephalopathic crisis. In addition to extension of sylvian fissures, hyperintensity of putamen, caudate, and pallidum are obvious. (c) T2-​weighted axial and coronal MRIs of a 66-​year-​old man with glutaryl-​CoA dehydrogenase deficiency demonstrating confluent white matter changes, wide temporopolar and insular cerebrospinal fluid spaces, and cortical atrophy, but normal signal of basal ganglia. The previously healthy man presented from the age of 50 with slowly progressive neurological disease, including seizures, dementia, and speech problems. Aggressive behaviour as well as acoustic and visual hallucinations led to the suggestion of psychiatric disease. (c) Reproduced with permission from Külkens et al. 2005.

12.2  Protein-dependent inborn errors of metabolism 1965 biocytin or short biotinyl peptides. Liberation of biotin from its pro- tein conjugates is catalysed by biotinidase. Biotinidase deficiency Aetiology/​pathophysiology Biotinidase regenerates biotin from endogenous sources and liber- ates protein-​bound biotin, which derives from natural foodstuffs and the holocarboxylases. Free biotin is recycled and used for the reformation of holocarboxylases by the action of holocarboxylase synthetase through the biotin cycle (Fig. 12.2.8). The primary bio- chemical defect in most patients with late-​onset multiple carb- oxylase deficiency was shown in 1983 to be a profound deficiency of serum biotinidase encoded by the BTD gene (3p25). The metabolic abnormalities caused by deficiency of the respective biotin-​dependent carboxylases are as follows:  lactic acidosis due to pyruvate carb- oxylase deficiency; hyperammonaemia and accumulation of me- tabolites of alternative propionate metabolism (see also ‘Propionic aciduria’) due to propionyl-​CoA carboxylase deficiency; and eleva- tion of 3-​hydroxyisovaleric acid, 3-​methylcrotonylglycine, and 3-​ hydroxyisovalerylcarnitine (see also ‘3-​Methylcrotonylglycinuria’) due to methylcrotonyl-​CoA carboxylase deficiency. Clinical presentation Onset of first symptoms is variable, ranging from 1 week to 10 years of age. The mean age of presentation is between 3 and 6 months. Provision of biotin by the mother in utero delays symptoms and bio- chemical abnormalities in newborns with biotinidase deficiency. The most frequent symptoms are lethargy, hypotonia, seizures, and ataxia often in combination with stridor, episodes of hyperventi- lation, and apnoea. If undiagnosed and untreated, progression of the disease can be potentially fatal (Fig. 12.2.9). In older children, progressive neurological disease is often the leading presentation, including ataxia, (myoclonic) epileptic encephalopathy, and devel- opmental delay. Neurosensory hearing loss and ophthalmic dis- orders, such as optic atrophy, develop in most untreated patients. Skin rash and/​or alopecia are hallmarks of the disease. Diagnosis Urinary organic acid analysis is useful for differentiating isolated carboxylase deficiencies from the multiple carboxylase deficien- cies that occur in biotinidase deficiency and holocarboxylase syn- thase deficiency. However, metabolic abnormalities are highly variable and are absent at birth when the patient is not biotin de- pleted. Whereas accumulation of abnormal organic acid metabol- ites may show characteristic metabolites of propionic aciduria (see also ‘Propionic aciduria’), pyruvate carboxylase deficiency, and 3-​methylcrotonylglycinuria (see also ‘3-​Methylcrotonylglycinuria’) (Fig. 12.2.2), only 3-​hydroxyisovaleric acid may be found ele- vated, especially in the early stages of the disease. Notably, 3-​hydroxyisovaleric acid is also the most commonly elevated urinary metabolite in holocarboxylase synthetase deficiency, 3-​methylcrotonyl-​CoA carboxylase deficiency, and acquired biotin deficiency. Biotin is decreased in plasma and urine and biocytin is increased in urine. Diagnosis is made by analysis of serum biotinidase activity. Enzymatic activity less than 10% is classified as profound biotinidase deficiency and activity between 10 and 30% as partial biotinidase deficiency. Furthermore, few patients with decreased affinity of biotinidase for biocytin (Km variants) exist. They may show erro- neously high residual activity on in vitro testing. Prenatal diagnosis is feasible by measurement of biotinidase activity but may not be necessary because of effective treatment and favourable clinical out- come. Newborn screening for biotinidase deficiency is now estab- lished in many countries. Treatment and outcome Biotinidase deficiency is effectively treated by daily oral adminis- tration of pharmacological doses of biotin. Restriction of protein is not necessary. Administration of 5 to 10 mg of oral biotin per day promptly reverses or prevents all clinical and biochemical abnor- malities. Biotin treatment has to be maintained lifelong and has no Succinate Oxaloacetate 3* 4* 2* 1* Citrate Acetyl-CoA Pyruvate Methylmalonyl-CoA Propionyl-CoA Malonyl-CoA Fatty acids 3-Methylglutaconyl-CoA 3-Methylcrotonyl-CoA Leucine Glucose Valine Isoleucine Threonine Odd-chain fatty acids Fig. 12.2.7  Important carboxylases in amino acid metabolism. Asterisked enzymes are 1, 3-​methylcrotonyl coenzyme A carboxylase; 2, propionyl coenzyme A carboxylase; 3, pyruvate carboxylase; and 4, acetyl coenzyme A carboxylase. Biotin Apocarboxylases (PCC, MCC, PC, ACC) Holocarboxylases Biocytin, short biotinylpeptides Holocarboxylase synthetase Proteolysis Biotinidase Protein-bound biotin (diet) Lysine Lysylpeptides Biotinidase Fig. 12.2.8  The biotin cycle. Biotin is cleaved from biocytin (biotinyl-​ lysine) or small peptides by biotinidase. Activation of the apoenzymes resulting in functioning carboxylases (3-​methylcrotonyl-​CoA, propionyl-​CoA, acetyl-​CoA, and pyruvate carboxylases) is carried out by holocarboxylase synthetase. ACC, acetyl-​CoA carboxylase; MCC, 3-​methylcrotonyl-​CoA carboxylase; PC, pyruvate carboxylase; PCC, propionyl-​CoA carboxylase. Source data from Zschocke J, Hoffmann GF (2011). Vademecum metabolicum. Manual of metabolic paediatrics, 3rd edition. Schattauer, Stuttgart.

SECTION 12  Metabolic disorders 1966 side effects. Most patients with biotinidase deficiency known today were detected by newborn screening. Patients with Km variants have an increased risk of becoming biotin deficient and thus must also be treated with biotin. After early detection and consequent treatment, the outcome of biotinidase deficiency is excellent. Holocarboxylase synthetase deficiency Aetiology/​pathophysiology Holocarboxylase synthetase deficiency is a rare, autosomal reces- sive disease. Several disease-​causing mutations have been identified in the HLCS gene (21q22.1). Only about 40 patients have been re- ported. Residual activity has been observed in all affected individuals suggesting that complete enzyme deficiency may be lethal in utero. The coenzyme biotin is attached to the various apocarboxylases by the enzyme holocarboxylase synthetase. The carboxyl group of biotin is linked by an amide bond to an ε-​amino group of a specific lysine residue of the apoenzymes. Deficiency of holocarboxylase synthetase leads to failure of synthesis of all carboxylases, causing biochemical and clinical abnormalities attributable to the dysfunc- tion of each respective carboxylase. Clinical presentation Although holocarboxylase synthetase deficiency was initially termed early-​onset multiple carboxylase deficiency, the age of onset of symptoms varies widely, from a few hours after birth to 6 years of age. Nevertheless, about one-​half of patients present acutely in the first days of life with severe metabolic decompensation, leth- argy, hypotonia, vomiting, seizures, and hypothermia. Patients with early-​onset presentation exhibit severe metabolic acidosis with lactic acidaemia, ketosis, and hyperammonaemia in analogy to biotinidase deficiency (see also ‘Biotinidase deficiency’). The meta- bolic derangement may quickly progress from lethargy to coma and early death. Skin rashes, feeding difficulties, vomiting, muscular hypotonia and hypertonia, seizures, and the odour of male cat urine are other symptoms. Ataxia, tremor, hyporeflexia, or hyperreflexia are neurological manifestations of the disease. Diagnosis Biochemical abnormalities of holocarboxylase synthetase deficiency are analogous to those described for patients with biotinidase de- ficiency (see also ‘Biotinidase deficiency’). Importantly, plasma biotin is normal in holocarboxylase synthetase deficiency as is serum biotinidase activity. Holocarboxylase synthetase is charac- terized by deficient activities of carboxylases in peripheral blood leucocytes prior to biotin administration; the activities of these enzymes increase to near-​normal or normal values after biotin treatment. Indirect confirmation of holocarboxylase synthetase de- ficiency and differentiation from biotinidase deficiency is feasible by measurement of activities of the mitochondrial carboxylases in skin fibroblasts showing residual activity of 0 to 30% when incu- bated in low-​biotin (10−10 mol/​litre) medium and an increase, some- times to normal values in biotin-​supplemented medium (10−6–​10−5 mol/​litre). In biotinidase deficiency, the activity of mitochondrial carboxylases in fibroblasts is normalized even under low-​biotin conditions. Definite diagnosis of holocarboxylase synthetase de- ficiency is not routinely available. Prenatal diagnosis is feasible ei- ther by demonstrating decreased carboxylase activities in cultured amniocytes or by demonstration of elevated 3-​hydroxyisovaleric acid and/​or methylcitrate in amniotic fluid. Prenatal molecular diagnosis can be offered in families with previously known disease-​ causing mutations in the HLCS gene. Treatment and outcome Holocarboxylase synthetase deficiency can be treated effectively with pharmacological doses of biotin. The required dose of biotin is dependent on the severity of the enzyme defect and has to be (a) (b) Fig. 12.2.9  Two T2-​weighted images of a 7-​month-​old boy with biotinidase deficiency. (a) The image displays absence of normal myelin signal in the cerebellum as well as hyperintense signal in both pyramidal tracts. (b) The image shows absence of normal myelin signal, cerebral atrophy, and symmetrical hyperintense lesions of both thalami. Courtesy of Dr. T. Bast, Department of Pediatric Neurology, University of Heidelberg, Heidelberg, Germany.

12.2  Protein-dependent inborn errors of metabolism 1967 assessed individually. In most patients, 10 to 20 mg of biotin per day is sufficient, but some need higher doses, that is, 40 to 100 mg/​ day. In spite of apparently complete recovery, biochemical and clin- ical abnormalities persist in some patients owing to the high Km for biotin in the defective holocarboxylase synthetase. In case of acute decompensation, treatment according to the emergency protocol in organic acidurias (see ‘Emergency treatment’) has to start without delay. It is unclear whether prenatal treatment with biotin is bene- ficial. The prognosis is good if treatment is initiated immediately, except for affected individuals with Km variants. Other organic acidurias d-​2-​Hydroxyglutaric aciduria type I and II Aetiology/​pathophysiology d-​2-​Hydroxyglutaric aciduria is an aetiologically heterogeneous cerebral organic acid disorder first described by Chalmers and col- leagues in 1980. d-​2-​Hydroxyglutaric aciduria type I is caused by deficiency of d-​2-​hydroxyglutarate dehydrogenase, a mitochon- drial enzyme converting d-​2-​hydroxyglutarate to 2-​oxoglutarate. Pathogenic mutations have been identified in the D2HGDH gene on 2p25.3. Recently, autosomal dominant germline mutations of the IDH2 gene located on 15q26.1 causing increased conversion of 2-​oxoglutaric acid to d-​2-​HG using NADPH by isocitrate dehydro- genase 2 were identified as molecular cause for d-​2-​hydroxyglutaric aciduria type II.  Neurodegeneration in d-​2-​hydroxyglutaric aciduria is explained by activation of N-​methyl-​d-​aspartate recep- tors and inhibition of respiratory chain complexes (cytochrome c oxidase, ATP synthase) by d-​2-​hydroxyglutaric acid. Clinical presentation Patients with d-​2-​hydroxyglutaric aciduria exhibit variable pheno- types. They have been divided into two subgroups based on clin- ical, neuroradiological, and molecular findings. Patients with d-​2-​hydroxyglutaric aciduria type I are moderately affected and usu- ally follow a mild clinical course with variable symptoms including learning difficulties, muscular hypotonia, and macrocephaly. Rarely individuals remain almost asymptomatic, that is, presenting only with well-​treatable oligoepilepsy or even with no neurological symp- toms. The clinical presentation of patients with d-​2-​hydroxyglutaric aciduria type II is usually more severe than in patients with type I.  Patients present with encephalopathy of early infantile onset, demonstrating a combination of catastrophic epilepsy, muscular hypotonia, cerebral visual failure, and severe psychomotor retard- ation. Facial dysmorphism, macrocephaly, and cardiomyopathy may also be present. Neuroimaging findings in these patients show ventriculomegaly, enlarged subarachnoid spaces, subdural effusions, subependymal cysts, and delayed cerebral maturation (Fig. 12.2.10). Recently, agenesis of the corpus callosum, bilateral involvement of the striatum, and cerebral artery infarctions were added to the spectrum. Diagnosis The biochemical hallmark of this disease is the accumulation of d-​2-​hydroxyglutaric acid in all body fluids. Type I patients excrete lower concentrations of d-​2-​hydroxyglutarate than type II patients. Demonstration of elevated levels of 2-​hydroxyglutaric acid must be followed up by differential quantitation of the two isomers l-​ and d-​ 2-​hydroxyglutaric acid. 2-​Oxoglutaric acid and other tricarboxylic acid cycle intermediates are usually also elevated in urine. γ-​ Aminobutyric acid (GABA) and total protein concentrations may be elevated in cerebrospinal fluid. d-​2-​Hydroxyglutaric acid can also be elevated in multiple acyl-​CoA dehydrogenase deficiency, succinic semialdehyde dehydrogenase deficiency, and following bacterial overgrowth of the urine specimen. However, due to characteristic additional parameters these differential diagnoses are usually easy to exclude. Prenatal diagnosis can be performed either through genetic testing or by metabolite determination in amniotic fluid by stable isotope dilution GC/​MS assay. Treatment and outcome No specific therapy exists to date. Long-​term care of patients should entail regular evaluation of cardiomyopathy and the progression of neurological disease. The prognosis of d-​2-​hydroxyglutaric aciduria is extremely variable. Severely affected children may die in infancy, while moderately affected patients have a better prognosis up to an unimpaired life. l-​2-​Hydroxyglutaric aciduria Aetiology/​pathophysiology l-​2-​Hydroxyglutaric aciduria is a rare, autosomal recessively in- herited cerebral disorder. The disease is caused by deficiency of the flavin adenine dinucleotide-​dependent mitochondrial enzyme l-​2-​ hydroxyglutarate dehydrogenase converting l-​2-​hydroxyglutarate to 2-​oxoglutarate. This enzyme is encoded by the L2HGDH gene on 14q22.1. The pathophysiology of this disease is unknown. Clinical presentation l-​2-​Hydroxyglutaric aciduria was first described by Duran and coworkers in 1980. It is characterized by progressive loss of Fig. 12.2.10  Axial T1-​weighted spin echo image of a 2-​month-​old girl with d-​2-​hydroxyglutaric aciduria type II. The lateral ventricles are highly dilated, occipital more than frontal, the cerebral maturation is delayed. Reproduced with permission from Kölker et al. 2002.

SECTION 12  Metabolic disorders 1968 myelinated arcuate fibres and a spongiform encephalopathy. In the first 2 years of life, mental and psychomotor development may be normal or slightly delayed. Febrile seizures, nonspecific develop- mental delay, and muscular hypotonia are the presenting symptoms. Progressive ataxia, variable extrapyramidal and pyramidal signs, epilepsy, and progressive learning difficulties eventually develop. By adolescence, patients are usually bedridden and severely mentally disabled (IQ 40–​50). Two patients have developed cerebral tumours. Two patients presented at birth with depressed vital signs, severe epileptic encephalopathy, and an abnormal CT scan showing cere- bellar involvement; however, the disease course is usually slowly progressive without metabolic decompensation. The neuroimaging findings in l-​2-​hydroxyglutaric aciduria are unique and mostly uniform comprising a progressive loss of arcuate fibres combined with progressive cerebellar atrophy and signal changes in globus pallidus and the dentate nuclei (Fig. 12.2.11). Diagnosis l-​2-​hydroxyglutaric aciduria results in a rather homogeneous clin- ical picture and characteristic abnormalities on neuroimaging. Clinical or neuroradiological suspicion should prompt GC/​MS analysis of urinary organic acids followed by differentiation of l-​ 2-​ and d-​2-​stereoisomers. Lysine is often increased both in plasma and cerebrospinal fluid. Prenatal diagnosis is based on the analysis of l-​2-​hydroxyglutaric acid in amniotic fluid samples or molecular analysis. Treatment and outcome No specific therapy exists to date. Epilepsy can generally be controlled by antiepileptic medications. Patients with l-​2-​hydroxyglutaric aciduria can be expected to reach adult life. The oldest known pa- tients are close to 40 years of age, bedridden, and severely disabled. Combined d-​2-​ and l-​2-​hydroxyglutaric aciduria Aetiology/​pathophysiology The molecular basis of combined d-​2-​ and l-​2-​hydroxyglutaric aciduria has recently been unravelled. The disease is caused by homozygous or compound heterozygous mutations in the SLC25A1 gene (gene locus 22q11.21) resulting in a dysfunction of the mito- chondrial citrate carrier and thus in impaired mitochondrial citrate efflux. Clinical presentation In a similar manner to patients with d-​2-​hydroxyglutaric aciduria type II, patients with the combined d-​2-​ and l-​2-​hydroxyglutaric aciduria usually present with a severe clinical manifestation in the newborn period. This includes epileptic encephalopathy, mus- cular hypotonia, respiratory insufficiency, extrapyramidal move- ment disorders, cortical visual failure, microcephaly, and severe developmental delay. Agenesis of corpus callosum and optic nerve hypoplasia may be present. Otherwise, brain MRI may be similar to patients with d-​2-​hydroxyglutaric aciduria type II. Diagnosis Clinical and neuroradiological suspicion should prompt GC/​MS analysis of urinary organic acids and differentiation of l-​2-​ and d-​2-​ stereoisomers. The diagnosis can be confirmed by molecular genetic analysis. Treatment and outcome Treatment is symptomatic. Patients with a severe onset and intract- able epileptic seizures have a poor prognosis: eight of twelve recently reported cases died between 1 month and 5 years of age. N-​Acetylaspartic aciduria (Canavan’s disease) Aetiology/​pathophysiology N-​Acetylaspartic aciduria is a devastating infantile neuro­ degenerative disorder. In 1931, a child with spongy matter (a) (b) Fig. 12.2.11  (a) Axial T2-​weighted spin echo image of an 8½-​year-​ old boy with l-​2-​hydroxyglutaric aciduria. Subcortical white matter is severely deficient with much less involvement of the internal capsule and the periventricular white matter. Please note signal changes in the putamen. (b) Axial T2-​weighted spin echo image of an 8½-​year-​old boy with l-​2-​hydroxyglutaric aciduria. Please note hyperintense lesions in both dentate nuclei. (a) Reproduced with permission from Kölker et al. 2002.

12.2  Protein-dependent inborn errors of metabolism 1969 degeneration was described by Canavan. In 1986, it was rec- ognized that N-​acetylaspartic aciduria was caused by deficient aspartoacylase in a child with a similar clinical presentation. In 1988, aspartoacylase deficiency was definitely linked to Canavan’s disease. Canavan’s disease is found in all ethnic populations but re- veals a much higher frequency in Ashkenazi Jews (1 in 5000 to 1 in 14 000 newborns). The frequent missense mutation p.E285A in the aspartoacylase gene, localized on 17p13-​pter, accounts for more than 80% of alleles in Ashkenazi Jews and for 60% of alleles in pa- tients of non-​Jewish origin. In healthy individuals, high concen- trations of N-​acetylaspartic acid (8 mmol/​g tissue) are exclusively found in brain tissue. Aspartoacylase is localized in oligodendrocytes catalysing the deacetylation of N-​acetylaspartic acid to produce acetate, a substrate for the synthesis of myelin lipids including cholesterol. It has been proposed that N-​acetyl-​l-​aspartate may function as a molecular water pump in myelinated neurons, transporting water against its gradient from neurons to oligodendrocytes. Thus aspartoacylase de- ficiency may cause both accumulation of metabolic water causing spongiform white matter changes, and deficiency of acetyl groups needed for cholesterol biosynthesis, causing demyelination; both are characteristic of Canavan’s disease. Clinical presentation Canavan’s disease mostly manifests at age 2 to 4 months with delayed development. Hypotonia with prominent head lag, epilepsy, loss of previously acquired skills, as well as progressive megalencephaly are regularly found. Seizures and optic nerve atrophy develop during the second year of life. As the disease progresses, affected children develop pyramidal signs, and finally decerebration. Neuroimaging reveals characteristic symmetrical leukodystrophic changes with loss of arcuate fibres; histology demonstrates spongi- form degeneration, in particular of the cortex and subcortical white matter (Fig. 12.2.12) with less involvement in the cerebellum and brainstem. In infancy, changes may be subtle and misinterpreted as delayed myelination or periventricular leukomalacia. Variant Canavan’s disease has been described and partially been proven to be caused by the same metabolic defect. Diagnosis Muscular hypotonia, head lag, and progressive megalencephaly in infancy are the classic clinical triad of Canavan’s disease. The identification of the accumulating N-​acetylaspartic acid by GC/​MS analysis and confirmation of the suspected diagnosis by en- zyme analysis (skin fibroblasts) or mutation analysis has obviated the need for brain biopsy for the diagnosis of Canavan’s disease. Prenatal diagnosis is possible by quantitative GC/​MS analysis of N-​acetylaspartic acid in amniotic fluid or by mutation analysis. In contrast, enzyme activity is unsuitable for reliable prenatal diagnosis. Treatment and outcome Management is symptomatic (antiepileptics) and palliative. Special care is needed to prevent recurrent aspirations. Many patients need tube or gastrostomy feeding. Dietary therapies have not been shown to be beneficial and are potentially harmful. A promising protocol for gene therapy was published in 2002 involving the transfer of human aspartoacylase cDNA intraventricularly; however, the clinical changes were not pronounced and were relatively tran- sient. The prognosis of infantile Canavan’s disease is rapidly fatal, whereas milder disease has been described with survival beyond the teenage years. Ethylmalonic encephalopathy Aetiology/​pathophysiology Ethylmalonic encephalopathy is a devastating, infantile, autosomal recessive neurometabolic disorder affecting the brain, gastrointes- tinal tract, and peripheral veins. The underlying metabolic defect was identified in a β-​lactamase-​like, iron-​coordinating metalloprotein of the mitochondrial matrix encoded by the ETHE1 gene. Only re- cently, it was elucidated using Ethe1-​deficient mice that the deficient protein is a mitochondrial sulphur dioxygenase which is involved in the catabolism of sulphide in ethylmalonic encephalopathy. As a consequence, toxic levels of sulphide and thiosulphide are found causing powerful inhibition of cytochrome c oxidase, short-​chain fatty acid oxidation, and exerting vasoactive and vasotoxic effects. This explains deficient mitochondrial energy metabolism, the ab- normal accumulation short-​chain organic acids, acylglycines and acylcarnitines, as well as microangiopathy. Clinical presentation Ethylmalonic encephalopathy is characterized biochemically by ethylmalonic aciduria and methylsuccinic aciduria, lactic acidaemia, and clinically by severe psychomotor retardation, acrocyanosis, pe- techiae, and chronic diarrhoea. Newborns present with muscular hypotonia followed by pro- gressive neurological deterioration, especially pyramidal dys- function, learning difficulties, orthostatic acrocyanosis with distal Fig. 12.2.12  Axial fast spin echo image of a 6½-​year-​old girl with aspartoacylase deficiency. Note the marked discrepancy between the severely affected subcortical white matter and the relatively spared central white matter, at least frontally. Reproduced with permission from Kölker et al. 2002.

SECTION 12  Metabolic disorders 1970 swelling, chronic diarrhoea, and recurrent petechiae (Fig. 12.2.13). Haematuria is often present. MRI scans show signal changes in cere- bellar white matter and lesions in the basal ganglia, the latter ap- pearing suddenly. Diagnosis The biochemical hallmark is increased urinary excretion of ethylmalonic and methylsuccinic acids associated with abnormal excretion of C4-​ and C5-​ (n-​butyryl-​, isobutyryl-​, isovaleryl-​, and 2-​ methylbutyryl-​) acylglycines and acylcarnitines as well as intermit- tent lactic acidosis. Since primary mitochondrial disorders are an important differential diagnosis, enzymatic analyses of respiratory chain enzymes in muscle biopsy specimen have been performed in some patients revealing secondary cytochrome c oxidase deficiency. Mutation analysis of the ETHE1 gene provides the definitive diag- nosis including prenatal diagnosis. Increased ethylmalonate in urine is also found in multiple-​ and short-​chain acyl-​CoA dehydrogenase deficiencies, primary respiratory chain deficiencies, and Jamaican vomiting sickness. Treatment and outcome No effective treatment is known. The prognosis is poor and ethylmalonic encephalopathy is usually lethal in early childhood. Defects of phenylalanine and tyrosine metabolism Phenylketonuria The hyperphenylalaninaemias are a group of disorders characterized by defective hydroxylation of phenylalanine to tyrosine resulting in plasma phenylalanine values above the normal fasting range of 40 to 80 µmol/​litre. PKU was first identified by the Norwegian Asbjørn Følling in 1934 in several severely disabled individuals. Følling de- termined the urinary excretion of phenylpyruvic acid which led to the previously used term ‘phenylpyruvic oligophrenia’. In 1947, Jervis localized the metabolic error as an inability to oxidize phenyl- alanine to tyrosine. In 1953, Bickel and colleagues demonstrated that a phenylalanine-​restricted diet was beneficial, and was thus the first successful treatment of an inborn error of metabolism and one which led the way to early diagnosis by newborn screening and treat- ment. The worldwide overall incidence of PKU is approximately 1 in 10 000, with a large national and ethnic variability. Aetiology/​pathophysiology PKU is an autosomal recessive disorder caused by a severe defect of phenylalanine hydroxylase which converts phenylalanine into tyrosine (Fig. 12.2.14). Tetrahydrobiopterin is required as a cofactor and thus hyperphenylalaninaemia may also be caused by inappro- priate generation of tetrahydrobiopterin. Through mechanisms still not completely understood, the excess phenylalanine is toxic to the central nervous system. Phenylalanine competes with the transport of large neutral amino acids through the blood–​brain barrier using the sodium-​independent system L and induces cerebral depletion of these amino acids and, subsequently, reduced synthesis of pro- teins and neurotransmitters (large neutral amino acid hypothesis of PKU). In addition, phenylalanine competes with glycine and glu- tamate at their binding sites in N-​methyl-​d-​aspartate and α-​amino-​ 3-​hydroxy-​5-​methyl-​4-​isoxazolepropionic acid receptors, thus impairing glutamate signalling and, subsequently, synapse forma- tion and cognitive function. Furthermore, phenylalanine inhibits the rate-​limiting enzyme of cholesterol biosynthesis, 3-​hydroxy-​3-​ methylglutaryl-​CoA reductase, and switches forebrain oligodendro- cytes to a nonmyelinating state. Clinical presentation Untreated, PKU almost invariably causes severe learning difficulties. Newborns with PKU are asymptomatic since fetal phenylalanine is metabolized by the mother’s liver. On regular intake of natural pro- tein, phenylalanine levels quickly rise. Constitutional abnormal- ities (80–​100% of patients) such as hypopigmentation of the skin and hair (fair) and iris (blue) develop rapidly because synthesis of Fig. 12.2.13  Patient with ethylmalonic encephalopathy. Phenylalanine Phenyl pyruvate BH4 BH2 4* 3* BH4 BH2 3* 5* 6* BH4 BH2 3* 5OH Tryptophan 5-Hydroxytryptamine 7* Tryptophan 1* 2* Tyrosine Dihydroxyphenylalanine Dopamine Noradrenaline Adrenaline Phenylacetate Phenylacetylglutamine p-Hydroxyphenyl pyruvate Homogentisate Maleylacetoacetate Fumarylacetoacetate Fumarate + acetoacetate Fig. 12.2.14  The metabolism of phenylalanine and tyrosine and the role of tetrahydrobiopterin. The asterisked enzymes are 1, phenylalanine hydroxylase; 2, tyrosine hydroxylase; 3, dihydrobiopterin reductase; 4, tyrosine aminotransferase; 5, homogentisic acid oxidase; 6, fumaryl acetoacetate hydrolyase; and 7, tryptophan hydroxylase.

12.2  Protein-dependent inborn errors of metabolism 1971 melanin from tyrosine is impaired. Elevated phenylacetate excretion gives the urine an odour reminiscent of mice and can cause an ec- zematous skin eruption. Delayed psychomotor development may become evident from the third month of life. It has been estimated that one IQ point is lost for each week of delay in diagnosis and treatment. Cognitive function is severely compromised in untreated children (IQ <40). Microcephaly and movement disorders are frequent, as are hyperexcitability as well as hypoexcitability and seizures; some pa- tients develop autistic behaviour or aggressiveness. Most patients with untreated PKU cannot be managed by their families and re- quire institutional care. Diagnosis In many countries, newborns are screened for increased phenyl- alanine levels in dried blood spots during the first days of life (new- born screening). Originally, newborn screening of phenylalanine was performed by a bacterial inhibition assay (Guthrie test). The implementation of MS/​MS techniques has, however, significantly improved the early identification of affected individuals by new- born screening. Confirmation of a positive screening result is per- formed by quantitative amino acid analysis and mutation analysis. Liver biopsy and subsequent determination of the hepatic activity of phenylalanine hydroxylase is not indicated. Defects in the metabolism of tetrahydrobiopterin (BH4), the cofactor of phenylalanine hydroxylase, have to be differentiated from classic PKU by urinary pterin analysis and enzyme ana- lysis of dihydropteridine reductase in dried blood spots. In many centres, an oral dose of 20 mg/​kg BH4 is administered. To per- form this test accurately, the initial plasma phenylalanine con- centration should be greater than 400 µmol/​litre (6.7 mg/​dl). Following BH4 administration, plasma samples are collected for phenylalanine and tyrosine analysis at defined time points as well as urine samples for pterin analysis. Notably, BH4 normalizes phenylalanine concentrations in patients with a primary disorder of BH4 (see ‘Defects of biopterin metabolism’). This test has the advantage that it may also identify BH4-​responsive individuals with PKU. Treatment and outcome The most important therapeutic intervention in PKU is phenylalanine-​restricted dietary treatment. Regular phenylalanine determinations are used for monitoring. Unfortunately, recom- mendations for PKU treatment differ considerably with regard to cut-​off levels to begin dietary treatment, age-​dependent recom- mendations for phenylalanine concentrations, frequency of clinical examinations, and phenylalanine monitoring (Table 12.2.4). There is no rational explanation for this. The concept of dietary treatment has four components: (1) com- plete avoidance of food containing abundant phenylalanine (e.g. meat, fish, milk, etc.); (2) calculated intake of natural food with a low phenylalanine/​protein ratio (e.g. vegetables and fruit) and low-​ protein products; (3)  adequate intake of energy substrates; and (4) calculated intake of phenylalanine-​free amino acid supplements, vitamins, minerals, and trace elements. During catabolic states phenylalanine concentrations may increase, which is counteracted by dietary reduction of phenylalanine intake. In contrast, during growth spurts in childhood and adolescence the requirement for phenylalanine may transiently increase. When a very strict diet is begun early and is well maintained, affected children can expect normal development. Regression of IQ and development of neurological symptoms when diets were stopped in later childhood have led to continuation of dietary treat- ment into the teenage years and adulthood. Patients generally have not suffered when the diet was stopped at or after 15 or 16 years of age. However, there is no follow-​up with respect to IQ change of a substantial number who have been off diet for 20 years or more. Most recommendations and centres have adopted a philosophy of ‘diet for life’. However, the urgent need for more detailed informa- tion remains. Maternal PKU In 1980, Lenke and Levy reported the severe effects of maternal hyperphenylalaninaemia in the fetus (Table 12.2.5). The clinical features are similar to the fetal alcohol syndrome, and the severity of manifestations depends on the maternal phenylalanine level. In addition to learning difficulties and behavioural disorders, the ad- verse effects include malformations such as cardiac defects (usu- ally conotruncal), microcephaly, dysmorphic features, intrauterine growth retardation, neuronal migration disorders, and agenesis of the corpus callosum. Treatment and outcome Because of active placental transport, the ratio of fetal to maternal phenylalanine plasma levels is 1.5 to 1.7. Maternal phenylalanine values should be between 120 and 360 µmol/​litre, which requires a strict diet and very careful monitoring twice weekly. Microcephaly and congenital heart disease in the offspring of mothers returning to diet at the seventh or eighth week emphasizes the need for pre- conception diet and training. Lowering maternal plasma phenyl- alanine concentrations during pregnancy to a level between 120 and 360 µmol/​litre results in a favourable outcome in virtually all cases. Defects of biopterin metabolism In the hydroxylation of phenylalanine, the cofactor BH4 is con- sumed and must be regenerated. BH4 is formed in a three-​step pathway from guanosine triphosphate. The first and rate-​limiting reaction is catalysed by guanosine triphosphate cyclohydrolase and leads to the production of dihydroneopterin triphosphate. A defi- ciency of BH4 does not only impair phenylalanine hydroxylase in the liver, resulting in hyperphenylalaninaemia, but also tyrosine hydroxylase, tryptophan hydroxylase, as well as nitric oxide syn- thases (Fig. 12.2.15). Tyrosine hydroxylation is needed for the synthesis of noradrenaline and dopamine, and tryptophan hydrox- ylation for the production of serotonin. BH4 is therefore crucial to the production of neurotransmitters. The supply of this coenzyme is impaired in five recessively inherited enzyme defects. Most produce hyperphenylalaninaemia, which may not be marked. All but pterin-​ 4α-​carbinolamine dehydratase deficiency cause progressive neuro- logical disease. In less than 1% of newborns a raised phenylalanine value detected by newborn screening is due to a defect of biopterin metabolism. The enzyme defects lead to reduced levels of BH4 within the cen- tral nervous system without significantly affecting phenylalanine metabolism in the liver (normal plasma phenylalanine). However,

SECTION 12  Metabolic disorders 1972 Table 12.2.4  Guidelines for treatment and monitoring of PKU: international comparison Germany 1999 UK 1993 USA 2014 Indication for dietary treatment

600 µmol/​litre 400 µmol/​litre 360 µmol/​litre Start of dietary treatment As soon as possible ≤ day 20 of life ≤ day 7 of life Recommendations for phenylalanine levels and frequency of phenylalanine monitoring Germany 1999 UK 1993 USA 2014 Age Germany 1999 UK 1993 USA 2014 40–​240 µmol/​litre (0.7–​4 mg/​dl) 120–​360 µmol/​litre (2–​6 mg/​dl) 120–​360 µmol/​litre (2–​6 mg/​dl) 0 2–​4×/​month 4×/​month 4×/​month 1 1–​2×/​month 2×/​month 2 3 4 5 2×/​month School age: 6 120–​480 µmol/​litre (2–​8 mg/​dl) 7 8 9 40–​900 µmol/​litre (0.7–​15 mg/​dl) 10 1×/​month 1×/​month 11 Adolescence and adulthood Adolescence and adulthood 12 120–​700 µmol/​litre (2–​11.7 mg/​dl) 120–​360 µmol/​litre (2–​6 mg/​dl) 13 1×/​month 14 15 40–​1200 µmol/​litre (0.7–​20 mg/​dl) 16 4–​6 ×/​year 17 120–​360 µmol/​litre (2–​6 mg/​dl) 18+ Recommendations for clinical monitoring Germany 1999 Germany 2004 UK 1993 USA 2000 Dietary training Amino acid profile Nutrition No details Anthropometric data Blood count Growth Health status Minerals, trace elements General health status Neurological status Calcium and phosphorus metabolism Psychological development Enzymes: AP, GOT, GPT Vitamins and serum lipid status

12.2  Protein-dependent inborn errors of metabolism 1973 turnover of serotonin and the catecholamines in the brain can still become severely compromised. Fasting plasma phenylalanine levels are always normal in the dominantly inherited guanosine triphos- phate cyclohydrolase deficiency (Segawa’s disease) and the auto- somal recessive sepiapterin reductase deficiency. Clinical presentation Except for pterin-​4α-​carbinolamine dehydratase (PCBD1) defi- ciency, autosomal recessive defects of biopterin metabolism re- sult in severe encephalopathies. Common but variable symptoms are progressive learning difficulties, dystonia, chorea, oculogyric crises, convulsions, tremor, spasticity, microcephaly, growth re- tardation, swallowing difficulties, and depressive and aggressive be- haviour. Diurnal variation is often present. Onset of symptoms is in the first months of life with hypotonia; sometimes affected new- borns have difficulties in postnatal adaptation. Signs of autonomic dysfunction include hypersalivation, temperature instability, leth- argy, hypersomnolence, and episodes of sweating and pallor. Less frequently reported are ‘bulbar’ signs (drooling, dysarthria, ab- normal tongue movements), ‘ataxia’, probably not cerebellar ataxia or sensory ataxia but dystonic gait, and Gower’s sign. PCBD1 is a bifunctional protein that acts as an enzyme in the regeneration of BH4 and as a dimerization cofactor of the transcription factors HNF1A and HNF1B, which are important in liver, pancreas, and kidney development and function. Mutations in PCBD1 have re- cently been reported to cause early-​onset nonautoimmune diabetes mellitus highlighting that PCBD1 activity is required for early pan- creatic development. In addition, adult patients with PCBD1 defi- ciency and hypomagnesaemia due to renal magnesium wasting have been identified demonstrating that PCBD1 also plays an important role in the kidney, in particular in the distal convoluted tubule. In later infancy and childhood, defects in the metabolism of the biogenic monoamines may be suspected in patients with (fluc- tuating) extrapyramidal disorders, in particular parkinsonism dystonia or more general ‘athetoid cerebral palsy’, and vegetative disturbances. A  severe epileptic encephalopathy and progressive learning difficulties may be present. Diagnosis Every infant with hyperphenylalaninaemia detected in a population newborn screening programme or in the course of other diagnos- tics later in life must be carefully investigated for possible defects of biopterin metabolism (see also ‘Phenylketonuria’). Differential diag- nosis requires the analysis of pterins in urine or from Guthrie cards as well as the determination of enzyme activity of dihydropteridine reductase in dried blood spots. If the initial plasma phenylalanine concentration is above 400 µmol/​litre (6.7 mg/​dl), oral loading with BH4 (20 mg/​kg) will result in normalization of phenylalanine values within 4 to 8 h. Urinary biopterin and neopterin values are low in the guanosine triphosphate cyclohydrolase deficiency, whereas 6-​ pyruvoyltetrahydrobiopterin synthase deficiency has high neopterin values and low biopterin values. In patients with dihydropteridine reductase deficiency, neopterin is normal or slightly elevated and biopterin very high. After the biochemical diagnosis, all defects should be ascertained enzymatically and, if available, by mutation analysis. Following a diagnosis of a defect of biopterin metabolism, a lumbar puncture becomes necessary for analysis of the neurotrans- mitter metabolites 5-​hydroxyindoleacetic acid and homovanillic acid as well as neopterin, biopterin, and 5-​methyltetrahydrofolic acid. This allows differentiation between severe and mild forms of BH4 deficiencies and sets the indication for treatment with the neurotransmitter precursors l-​dopa and 5-​hydroxytryptophan. In patients with suggestive encephalopathies and normal phenyl- alanine values, analysis of neurotransmitters in cerebrospinal fluid is the only way of diagnosis. Treatment and outcome Blood phenylalanine concentrations should be more rigidly con- trolled than in classic PKU patients. In patients with guan­osine triphos- phate cyclohydrolase deficiency and 6-​pyruvoyl­tetrahydrobiopterin deficiency, administration of BH4 appears to be the most efficient therapy in controlling blood phenylalanine levels. Patients with dihydropteridine reductase deficiency need a low-​phenylalanine diet as in PKU. Deficiency of neurotransmitters requires treatment with the neurotransmitter precursors l-​dopa (3–​15 mg/​kg per day) and 5-​hydroxytryptophan (2–​9 mg/​kg per day) in combination with carbidopa (10 or 25% of l-​dopa). Lumbar punctures must be repeated regularly to adjust doses. In patients revealing l-​dopa-​induced peak-​ dose dyskinesia slow-​release forms of drugs can be used, and reaching the upper therapeutic limits of l-​dopa may be an indication for the use of monoamine oxidase and/​or catechol-​O-​methyltransferase Table 12.2.5  Incidences (%) of abnormalities in the offspring of mothers affected with classical PKU Congenital abnormalities Maternal PKU Unaffected mothers Mental disability 92 5 Microcephaly 73 4.8 Intrauterine retardation 40 9.6 Congenital heart defects 12 0.8 Source data from Lenke R, Levy HL (1980). Maternal PKU and hyperphenylalaninemia: an international study of treated and untreated pregnancies. N Engl J Med, 303, 1202–​8. GTP BH4 PTPS Neopterin SR BH2 Biopterin PAH, TYH, TPH, NOS PCD DHPR GTPCH Fig. 12.2.15  Biopterin metabolism. BH4 is synthesized and regenerated by five enzymes. BH4 is consumed as a cofactor in the hydroxylation of tyrosine and tryptophan as well as phenylalanine (see also PKU) and nitric oxide synthase (NOS). BH2, dihydrobiopterin. Relevant enzyme defects: DHPR, dihydropteridine reductase; GTPCH, GTP cyclohydrolase; PCD, pterin carbinolamine dehydratase; PTPS, 6-​pyruvoyl-​ tetrahydropterin synthase; SR, sepiapterin reductase. Source data from Zschocke J, Hoffmann GF (2011). Vademecum metabolicum. Manual of metabolic paediatrics, 3rd edition. Schattauer, Stuttgart.

SECTION 12  Metabolic disorders 1974 inhibitors. Patients with dihydropteridine reductase deficiency, in addition, need administration of folinic acid to restore normal cere- brospinal fluid folate concentrations. Normal long-​term psychomotor development can be achieved but outcome strongly depends on the age when the diagnosis is made and how rigidly therapy is followed, especially in early life. Dominantly inherited guanosine triphosphate cyclohydrolase deficiency Clinical presentation Dominantly inherited guanosine triphosphate cyclohydrolase deficiency, often called Segawa’s disease, is an eminently treat- able condition. Early recognition is therefore of crucial import- ance. Presentation in children usually occurs within the first decade of life with a mean age of onset of symptoms being about 7 years (range 16 months to 13 years). The first symptom is usu- ally postural dystonia of one leg with progression to all limbs followed by action dystonia and hand tremor within the next 10 to 15 years, during which time cognition remains intact. Occasionally, in older children, the first signs may start in the arms with torticollis or writer’s cramp (focal dystonia). The dys- tonia is frequently asymmetrical and accompanied by reduced facial expression or slowing of fine finger movements. Diurnal fluctuation is often present, with symptoms improving after night-​ time sleep or bed rest. The variation in presenting symptoms is large. Penetrance is reduced and many carriers of a mutant gene are asymptomatic. Diagnosis In classic cases with prominent dystonia of the lower limbs, marked diurnal variation, as well as worsening of the symptoms after ex- ercise, the clinical diagnosis of the deficiency is easily made, in particular in the presence of dramatic and sustained response to l-​dopa. However, the diagnosis can be a real challenge in atypical cases, in which it can be ascertained by determining BH4, and de- creased levels of neopterin and homovanillic acid in cerebrospinal fluid. Confirmation of the diagnosis can be achieved by enzyme ana- lysis in cultured skin fibroblasts or by mutation analysis. Treatment and outcome Treatment relies on l-​dopa in combination with 10 to 25% carbidopa. Amounts administered have varied between 3 and 10 mg/​kg per day divided into one to four doses with the effectiveness of treatment being monitored by the clinical outcome. The long-​term prognosis is usually excellent. Tyrosinaemias The steps in tyrosine metabolism starting with the rate-​limiting step—​the conversion to p-​hydroxyphenylpyruvic acid by tyrosine aminotransferase—​are outlined in Fig. 12.2.14. Intermediates of this tyrosine metabolism are used for production of catecholamines, dopamine, and the principal pigment of hair and skin, melanin. Tyrosinaemia type I (fumarylacetoacetase deficiency) Clinical presentation  Tyrosinaemia type I is also known as hepatorenal tyrosinosis. About one-​third of patients present acutely in the early weeks of life with failure to thrive, vomiting, hepatomegaly, fever, oedema, and epistaxis; by the end of the first year of life 90% have developed symptoms. The disease can progress rapidly and death from hepatic failure often occurs in infancy. A milder more chronic presentation is compatible with survival for several years with chronic liver disease, a renal tubular Fanconi’s syndrome with hypophosphataemic rickets, and episodic abdom- inal pain and neuropathy suggestive of acute porphyria. The most serious complication is hepatocellular carcinoma which develops in early childhood in one-​third of untreated patients. Diagnosis  Raised plasma tyrosine (often together with methio- nine), succinylacetone, and 5-​aminolaevulinic acid excretion as well as renal Fanconi’s syndrome are the biochemical markers of tyrosinaemia type I caused by a deficiency of fumarylacetoacetate hydrolyase, the last enzyme in the pathway of tyrosine degradation (Fig. 12.2.14). Serum α-​fetoprotein is usually strikingly elevated. Succinylacetone, formed from fumarylacetoacetate, is the most spe- cific diagnostic metabolite. Plasma tyrosine values may be normal, resulting in insufficient specificity of this parameter for newborn screening. Fumarylacetoacetate hydrolyase can be assayed in lymphocytes or fibroblasts. It is nonspecifically depressed in the liver in a var- iety of liver diseases. The measurement of succinylacetone in amni- otic fluid and activity of fumarylacetoacetate hydrolyase in cultured amniocytes or chorionic villus samples forms the basis of prenatal diagnosis, if informative mutations are not available. Treatment and outcome  Restricted intake of tyrosine and phenyl- alanine may reduce the excretion of succinylacetone and produce regression of the Fanconi tubular defects, but does not cure the liver disease. The risk of hepatocellular carcinoma remains and early liver transplantation was the treatment of choice until nitisinone (2-​(2-​nitro-​4-​trifluoromethylbenzoyl)1–​3-​cyclohexanedione) was introduced by Lindstedt and colleagues in 1991. Nitisinone almost completely blocks 4-​hydroxyphenylpyruvate dioxygenase thus turning tyrosinaemia type I into tyrosinaemia type III and redu- cing the production of toxic metabolites. Treatment with nitisinone should start as soon as the diagnosis is made with a dose of 1 mg/​ kg per day. In most patients there is a rapid improvement in liver and renal function; succinylacetone should disappear from the urine within 1 week of treatment. Patients need to be treated with a diet low in phenylalanine and tyrosine at the same time as introducing nitisinone. Plasma levels of tyrosine should be kept between 250 and 500 µmol/​litre. The long-​term results of nitisinone treatment are encouraging with greatly reduced incidence of liver damage and hepatic car- cinoma. Liver transplantation remains the treatment of choice for a few patients who do not respond to nitisinone and if there is any suggestion of malignant change. Tyrosinaemia type II (tyrosine aminotransferase deficiency) Clinical presentation  Corneal erosions and dendritic ulcers may form within a few months of birth with later scarring, nystagmus, and glaucoma. Skin lesions may begin after the eye lesions with blis- tering, painful palms and soles, and hyperkeratosis. Tongue changes have been described. Learning difficulties are an inconstant feature in about 50% of patients, but language defects may be more common with possible impaired coordination and self-​mutilation.

12.2  Protein-dependent inborn errors of metabolism 1975 Diagnosis  Tyrosine aminotransferase, which is deficient, cata- lyses the formation of p-​hydroxyphenylpyruvic acid (Fig. 12.2.14). Plasma tyrosine values reach 20 times normal (normal 40–​100 µmol/​ litre) in younger patients and 10 times normal in others. There is increased excretion of tyrosine, N-​acetyltyrosine, tyramine, and of phenolic acids; there is no Fanconi’s syndrome and no increase in succinylacetone. The clinical features and amino acid analyses are usually sufficient for diagnosis, which may be confirmed either by measuring the en- zyme activity in liver or by molecular genetic studies. Treatment and outcome  A low-​tyrosine, low-​phenylalanine diet has been used to produce rapid improvement of skin and eye mani- festations. Corneal transplants can be valuable. The neurological symptoms appear to improve less. The degree of dietary control needed to sustain clinical improvement is uncertain. Plasma tyro- sine concentrations less than 500 μmol/​litre are considered desirable. Tyrosinaemia type III (4-​hydroxyphenylpyruvate
dioxygenase deficiency) 4-​Hydroxyphenylpyruvate dioxygenase deficiency (Fig. 12.2.14) ap- pears to be very rare and possibly without clinical pathology, that is, a nondisease. It may be associated with learning difficulties and pos- sibly other neurological complications. The biochemical findings are similar to those in tyrosinaemia type II, but the plasma values of tyrosine are usually less than 1200 µmol/​litre. Enzyme and mo- lecular genetic studies can prove the diagnosis. Most patients are treated with a low-​tyrosine, low-​phenylalanine diet. Alkaptonuria Clinical presentation In 1902, alkaptonuria was the first disorder to be recognized as an inborn error of metabolism by Garrod. It is caused by a deficiency of homogentisate dioxygenase resulting in the accumulation of homogentisic acid and its oxidized derivative benzoquinone acetic acid. The latter can then be polymerized to form a dark pigment which is deposited in connective tissue. The disorder is extremely rare in most populations but occurs with greatly increased fre- quency in the Dominican Republic and in Slovakia. Presentation in infancy occurs only if discoloration of the urine is noticed. It is usu- ally normal when passed but darkens on standing (more rapidly at alkaline pH) to deep brown or almost black. Back pain begins in the second and third decade with increasing stiffness due to interverte- bral disc degeneration. Involvement of the hips, knees, and shoul- ders follows. Greyish discoloration of cartilage is seen in the pinna, and pigment is deposited in the sclera. Abnormal pigmentation is seen in the heart valves and joint cartilages, and pigmented stones are common in the prostate. Valvular calcification is prominent, es- pecially in the coronary arteries. Recent studies of the natural course of alkaptonuria indicate that it is associated with premature heart disease and premature death with long-​standing impairment of quality of life. Pigment deposition with involvement of the fibrolipid components of atherosclerotic plaques cause calcific stenosis of the aortic valve. In 58 patients studied by Phornphutkul and colleagues (2002), life-​table analysis showed that joint replacement occurred at a mean age of 55 years, renal stones at 64 years, and cardiac valve involvement at 54 years; coronary calcification occurred at a mean age of 59 years. Diagnosis Homogentisic acid can be demonstrated by urinary organic acid analysis. Enzymatic as well as molecular confirmation is possible. Plasma tyrosine concentrations are normal. Treatment and outcome So far no treatment has been shown to prevent the long-​term com- plications. The prognosis for the joints is poor. By the fifth decade, the lumbar spine is likely to be rigid and other joints will be ser- iously affected. Patients often require large amounts of analgesic and risk the complications of long-​term consumption of nonsteroidal anti-​inflammatory agents, which may exacerbate incipient coronary heart disease. Homogentisic acid can be decreased by a low-​protein diet. It is very probable that specifically designed low-​phenylalanine and low-​ tyrosine diets would lower the production still further. Nitisinone, the triketone inhibitor of 4-​hydroxyphenylpyruvate dioxygenase introduced by Lindstedt in 1991, greatly reduces overproduction of homogentisic acid in alkaptonuria. Early studies from Gahl’s group at the National Institutes of Health, United States of America, showed that in adults of both sexes with alkaptonuria, an oral dose of 1.05 mg twice daily reduced urinary homogentisic acid excretion from a mean of 4 g to 0.2 g per day. More than 220 patients with her- editary tyrosinaemia type I have received the drug at daily doses of 0.5 to 2.0 mg/​kg body weight and even at these doses it is generally well tolerated, apart from mild blood cytopenias. In alkaptonuria nitisinone, as predicted, may elevate the plasma tyrosine concentra- tions (in the early trial from c.70 to 760 μmol/​litre) and there is thus a theoretical risk of lens opacities, which can be avoided by careful slit-​lamp monitoring, plasma amino acid measurement, and dietary adjustment. In alkaptonuria, the outcome of nitisinone treatment will take many years to evaluate fully, but comprehensive therapeutic study is justified by the clear relationship between overproduction of a single metabolite and life-​shortening tissue manifestations with disabling joint disease. Neurotransmitter diseases and related disorders Monogenic defects of neurotransmission have become recog- nized as a cause of early-​onset, severe, progressive, and often treat- able encephalopathies. The diagnosis is based on the quantitative determination of the neurotransmitters or their metabolites in cerebrospinal fluid, that is, glycine, serine, and GABA, the acidic me- tabolites of the biogenic monoamines, and individual pterin species (Box 12.2.6). Determinations of metabolites in blood or urine are neither sensitive nor specific. In contrast to inborn errors in cata- bolic pathways, neurotransmitter defects are determined by the Box 12.2.6  Cerebrospinal fluid: investigation for neurotransmitter disorders • Cells, protein, immunoglobulin classes, and glucose (plus plasma glu- cose and evaluation of blood–​brain barrier) • Lactate and pyruvate • Amino acids (plus plasma obtained simultaneously) • Biogenic monoamine metabolites • Individual pterin species • 5-​Methyltetrahydrofolate

SECTION 12  Metabolic disorders 1976 interplay of biosynthesis, degradation, and receptor status. Even borderline abnormalities can be diagnostic and their recognition requires a strictly standardized sampling protocol and adequate age-​ related reference values. Disorders of monoamine metabolism Defects in the metabolism of the biogenic monoamines affect sero- tonin and/​or catecholamine (dopamine and noradrenaline) metab- olism (Fig. 12.2.16). They present from infancy or childhood with (fluctuating) extrapyramidal disorders, in particular parkinsonian dystonia or more general ‘athetoid cerebral palsy’, and vegetative disturbances, most noticeably hypoglycaemia. A severe epileptic en- cephalopathy and progressive learning difficulties may be present. Tyrosine hydroxylase deficiency Tyrosine hydroxylase catalyses the hydroxylation of l-​tyrosine to l-​dopa, the rate-​limiting step in the biosynthesis of the catechol- amines dopamine, noradrenaline, and adrenaline (Fig. 12.2.16). The iron-​containing mixed function oxidase requires molecular oxygen and the cofactor BH4. Tyrosine hydroxylase is expressed only in catecholaminergic neurons and the adrenal medulla. Tyrosine hydroxylase deficiency has become incorporated into concepts and classifications of dystonias as the cause of reces- sive l-​dopa-​responsive dystonia, but can also present as l-​dopa-​ nonresponsive dystonia or progressive early-​onset encephalopathy. Clinical presentation  Clinical symptoms often develop between 3 and 7 months of age. Most patients show a substantial clin- ical improvement already on low doses of l-​dopa together with the decarboxylase inhibitor carbidopa, although in contrast to l-​dopa-​responsive dystonia due to haploinsufficiency of guanosine triphosphate cyclohydrolase I, often neither the neurological status nor the catecholamine levels in cerebrospinal fluid can be com- pletely normalized in most patients. At the severe end of the spectrum, virtually no movements are observed, not even dystonic movements. Some patients are more severely affected and present with a progressive neurometabolic disorder from early infancy with a progressive infantile enceph- alopathy characterized by abnormal extrapyramidal movements and affecting several cerebral and possibly cerebellar systems. It is important to stress that such patients also show symptoms of sig- nificant catecholamine deficiency, such as hypoglycaemia and in- adequate stress responses. There is an obvious tendency to preterm birth with troublesome cardiorespiratory perinatal adaptation. Most infants with tyrosine hydroxylase deficiency develop sur- prisingly normally until an arrest of motor development with a characteristic combination of neurological symptoms later in infancy. Hypokinesia, marked truncal hypotonia, a mask face, oculogyric crises, myoclonic jerks, and an extrapyramidal tremor can progressively develop. The last three symptoms can be mis- taken as epileptic phenomena. Oculogyric crises are present but, as with the miosis, may go undiagnosed because of prominent ptosis. Contractures, failure to thrive, and immobilization may develop. It appears likely that life expectancy is significantly reduced; (dys- tonic) cerebral palsy is a likely descriptive (mis-​)diagnosis. Some patients did not develop extrapyramidal symptoms in the first year of life, were able to walk independently, and followed a clin- ical course best summarized as spastic paraplegia. Their symptoms fully resolved following l-​dopa supplementation, and they are now healthy adults living independently. 5-HIAA noradrenaline DβH adrenaline PNM 3-O-methyldopa vanillyllactic acid COMT DOPAC MAO COMT 3MT COMT COMT MAO HVA NM M VMA MHP G ALD MAO + aldd. MAO + alcd. COMT COMT COMT MAO tryptophan tyrosine 5-OH-tryptophan serotonin dopamine L -DOPA AADC + B6 TR + BH4 TH + BH4 AADC + B6 Fig. 12.2.16  Metabolism of biogenic monoamines. 5-​ HIAA, 5-​hydroxyindolacetic acid; AADC, aromatic l-​aminoacid decarboxylase; alcd., alcohol dehydrogenase; ALD, intermediate aldehyde (3-​methoxy-​4-​hydroxyphenyl-​ hydroxyacetaldehyde); aldd., aldehyde dehydrogenase; BH4, tetrahydrobiopterin; COMT, catechol-​ortho-​methyltransferase; DbH, dopamine-​b-​hydroxylase; DOPAC, 3,4-​dihydroxyphenylacetic acid; HVA, homovanillic acid; M, metanephrine; MAO, monoaminooxidase; MHPG, 3-​methoxy-​4-​hydroxy-​phenylglycol; MT, 3-​methoxytyramine; NM, normetanephrine; PNM, phenylethanolamine-​N-​methyltransferase; TH, tyrosin hydroxylase; TR, tryptophan hydroxylase; VMA, vanillylmandelic acid; . . ., several steps involved.

12.2  Protein-dependent inborn errors of metabolism 1977 Diagnosis  The diagnosis of tyrosine hydroxylase deficiency can only be made via cerebrospinal fluid analysis following a stand- ardized lumbar puncture protocol. A  characteristic metabolite constellation is found: low concentrations of metabolites of dopa- minergic neurotransmission homovanillic acid and 3-​methoxy-​ 4-​hydroxyphenylethyleneglycol in the presence of normal concentrations of metabolites belonging to the serotonin neurotrans- mission system such as 5-​hydroxyindoleacetic acid (Fig. 12.2.16). Urinary determinations of catecholamines and homovanillic acid turned out to be inconclusive in several affected individuals. Enzyme analysis is not possible in tyrosine hydroxylase deficiency because tissues expressing enzyme activity—​brain and adrenal medulla—​are difficult to obtain. Thus mutation analysis is the only way to confirm the diagnosis. Treatment and outcome  Therapeutic interventions with l-​dopa together with the decarboxylase inhibitor carbidopa and selegiline were able to improve and/​or even normalize the clinical picture in most patients but not all. Despite all therapeutic interventions, the disease course can be lethal. Treatment with l-​dopa has to be started slowly and carefully, with doses as low as 0.5 mg/​kg per day in two to six divided doses to avoid dyskinesias due to hypersensitivity and up-​regulation of dopamine receptors in dopamine-​deficient patients. In such patients, l-​dopa can only be increased very slowly, sometimes over several years. Slow-​release preparations may be useful to ensure constant l-​dopa levels. In general, incremental steps of l-​dopa/​carbidopa should not be more than 1 mg/​kg per day. Aromatic l-​amino acid decarboxylase deficiency Aromatic l-​amino acid decarboxylase deficiency is caused by autosomal recessively inherited mutations in the DDC gene. The enzyme is required for the synthesis of both serotonin and the catecholamines. Clinical presentation  Clinical symptoms are indistinguishable from those of patients with tyrosine hydroxylase deficiency. The se- verity seems to fall into two groups. About half of the patients pre- sent with feeding difficulties, autonomic dysfunction, and hypotonia in the neonatal period. In the first few months of life dystonia or intermittent limb spasticity, axial and truncal hypotonia, extreme irritability, oculogyric crises, and psychomotor retardation be- come obvious. More mildly affected patients may initially develop unremarkably or only slightly delayed and present with motor re- tardation, hypokinesia, rigidity, and truncal hypotonia from early childhood. Diagnosis  The enzyme deficiency leads to accumulation of 3-​ O-​methyldopa, 5-​hydroxytryptophan, and l-​dopa together with low concentrations of the end products homovanillic acid and 5-​ hydroxyindoleacetic acid (Fig. 12.2.16). 3-​O-​Methyldopa is formed by methylation of l-​dopa. Confirmation of the diagnosis is by en- zyme assay in plasma and finally by mutation analysis. Treatment and outcome  Different approaches using dopamine agonists (pergolide, pramipexole, bromocriptine, and ropinirole) and/​or nonselective monoamine oxidase inhibitors (tranyl­ cypromine, phenelzine) have been attempted. Response to treat- ment is variable but outcome appears to be better in more mildly affected and later-​presenting patients. The overall prognosis is guarded. About half of the patients improve on individual treatment regimens and acquire different degrees of motor and psychosocial skills. Others do not show any improvements. Dopamine β-​hydroxylase deficiency Clinical presentation  Recessively inherited mutations in the dopa- mine β-​hydroxylase gene lead to lowered levels of noradrenaline within central and autonomic noradrenergic neurons (Fig. 12.2.16). The disorder is characterized by sympathetic noradrenergic de- nervation and adrenomedullary failure. The central consequences appear minimal. Syndromes become obvious in adolescence with noradrenergic failure, severe orthostatic hypotension, and ptosis of the eyelids. During childhood fatigue, episodes of fainting, syn- copes, and exercise intolerance are generally present. Physical and cognitive function is normal. In males, autonomic neuropathy leads to retrograde ejaculation. Diagnosis  Dopamine β-​hydroxylase deficiency is classified as a pri- mary autonomic neuropathy. Conditions that lead to chronic failure of the autonomic nervous system are, therefore, the primary differ- ential diagnosis. Biochemically, dopamine β-​hydroxylase deficiency is different from other conditions with orthostatic hypotension or autonomic dysfunction. Failure to produce noradrenaline and the consequent lack of end-​product inhibition of tyrosine hydroxylase leads to a noradrenaline/​dopamine ratio of less than 0.1, and such a finding is pathognomonic for the disease. An increase in blood pressure and correction of the orthostatic hypotension in response to dihydroxyphenylserine is also diagnostic. Some 3 to 4% of the normal adult population have near zero levels of the enzyme in plasma, therefore plasma enzyme determination alone cannot be used to confirm the diagnosis, it requires mutation analysis. Treatment and outcome  Dopamine β-​hydroxylase deficiency is treated with dihydroxyphenylserine. This compound is decarb- oxylated by l-​amino acid decarboxylase to form noradrenaline. Administration of 250 to 500 mg twice daily results in an increase in blood pressure and sustained relief of the orthostatic symptoms. Without appropriate treatment postural hypotension can lead to sig- nificant injuries or even death. Disorders of pyridoxine metabolism In 1954, Hunt and colleagues described a patient with a seizure dis- order that was successfully treated solely by administration of pyri- doxine (vitamin B6) and coined the term ‘pyridoxine dependency’. It became good clinical practice to test for pyridoxine responsive- ness in every child with ‘difficult-​to-​treat’ seizures starting before 2 years of age. Later, a similar therapeutic response was described in the same clinical constellation for folinic acid. Finally, the enzymatic defect has been pinpointed to the α-​aminoadipic semialdehyde de- hydrogenase located in the lysine degradation pathway in the brain, which results in the accumulation of the intermediate piperideine-​6-​ carboxylate scavenging pyridoxal phosphate. A similar pathogenic mechanism again leading to intractable seizures is responsible for pyridoxal deficiency in hyperprolinaemia type II and during treat- ment with the tuberculostatic drug isoniazid. Another monogenic defect in humans is directly located within the synthesis of pyridoxal 5′-​phosphate: pyridox(am)ine 5′-​phosphate oxidase deficiency resulting in pyridoxal phosphate-​responsive seiz- ures (Fig. 12.2.17).

SECTION 12  Metabolic disorders 1978 Each newborn with severe neonatal/​infantile epileptic enceph- alopathy should have a lumbar puncture and then immediately receive consecutive therapeutic trials with vitamin B6, pyridoxal 5′-​phosphate, and folinic acid. Aetiology/​pathophysiology Pyridoxine-​dependent epilepsy and folinic acid-​responsive seiz- ures are treatable causes of neonatal epileptic encephalopathy. The genetic base of both conditions is autosomal recessive inher- itance of pathogenic mutations in the ALDH7A1 (antiquitin) gene causing deficiency of the enzyme α-​aminoadipic semialdehyde dehydrogenase located in the pipecolic acid pathway, the major route of cerebral lysine oxidation. As a consequence of accumu- lating α-​aminoadipic semialdehyde and the cyclic compound ∆1-​ piperideine 6-​carboxylate, which spontaneously forms an adduct with pyridoxal phosphate via a Knoevenagel reaction, pyridoxal phosphate is inactivated resulting in cerebral depletion of pyridoxal phosphate. Pyridoxal phosphate-​dependent enzymes such as glu- tamate dehydrogenase, GABA transaminase and aromatic l-​amino acid dehydrogenase are inactivated by pyridoxal phosphate deple- tion causing significant disturbance in the metabolism of the neuro- transmitters dopamine, serotonin, glutamate and GABA and thus a severe epileptic encephalopathy. The conversion of pyridoxine, pyri- doxal, and pyridoxamine to pyridoxal phosphate however remains unaffected. Clinical presentation Pyridoxine-​dependent epilepsy can be heterogeneous in its presen- tation, and sometimes idiopathic epilepsies respond to treatment with high-​dose pyridoxine. Classical patients with pyridoxine-​ dependent epilepsy present with an intractable seizure disorder within the first 2 days of life, and at the latest within 28 days. In some patients intrauterine convulsions are reported. There is no consistent electrographic pattern. Continuous and discontinuous backgrounds, suppression burst-​like patterns, and hypsarrhythmia have all been observed. There are additional atypical presentations: (1) late onset, that is, later than 28 days; (2) neonatal onset but with an initial response to conventional anticonvulsant therapy; (3) neonatal onset with initially negative, but a later sustained positive response to pyridoxine. Folinic acid-​sensitive seizures have been an enigmatic clinical and biochemical entity until it has been elucidated recently that they are alleic to pyridoxine-​dependent epilepsy. Patients present with myo- clonic or clonic seizures, apnoea, and irritability within 5 days after birth. The electroencephalogram shows a discontinuous background pattern with multifocal spikes and sharp waves. Without specific treatment seizures will only be partially controlled. Psychomotor development will become severely impaired. It is therefore recom- mended that all patients with ‘difficult-​to-​treat’ seizures starting be- fore 2 years should have a trial of pyridoxine and folinic acid (usually given orally in this circumstance). Diagnosis The diagnosis of pyridoxine-​dependent epilepsy and folinic acid-​ responsive seizures should be suspected clinically in patients with neonatal epileptic encephalopathy or ‘difficult-​to-​treat’ seizures starting before 2  years of age who respond to pyridoxine and/​ or folinic acid. Because it is a treatable condition, a high index of suspicion is warranted. Both pyridoxine and pyridoxal phosphate may cause apnoea and prolonged cerebral depression after the ini- tial dose, and resuscitation equipment and intensive care facilities should be available. The suspected diagnosis can be confirmed by measurement of α-​ aminoadipic semialdehyde in body fluids. Elevated CSF and plasma pipecolic acid is also used as a biomarker. Furthermore, CSF analysis may reveal a monoamine pattern similar to l-​amino acid dehydro- genase deficiency, elevated glutamate, and decreased GABA concen- trations. Enzyme assay and mutation analysis of the ALDH7A1 gene is the most definitive proof of diagnosis. Treatment and outcome Treatment requires 5 to 30 mg/​kg body weight per day of pyridoxine in one dose. Successful treatment with folinic acid can be achieved with 3 to 5 mg/​kg body weight per day of folinic acid given in three doses. Doses need to be increased and adjusted to body weight during growth. Breakthrough seizures are an obvious criterion for increasing the dose. There is evidence that lower doses of pyridoxine and folinic acid, while controlling seizures, may still not prevent the development of cognitive impairment. High doses of pyridoxine carry the risk of developing skin photosensitivity as well of a per- ipheral sensory neuropathy. Doses up to 1 g/​day can be regarded as safe in older children. Serial cognitive assessment is therefore recommended. If there is a positive family history of pyridoxine-​ dependent seizures, maternal treatment in utero is indicated. Since pyridoxine-​dependent epilepsy and folinic acid-​sensitive seizures appear to be genetically and biochemically identical, this new understanding requires a re-​evaluation of optimal strategies such as the combined use of pyridoxine and folinic acid as well as of a low-​lysine diet aiming to reduce the accumulation of α-​aminoadipic semialdehyde and ∆1-​piperideine 6-​carboxylate. Hyperprolinaemia type II: l-​Δ1-​pyrrolines-​5-​carboxylate dehydrogenase deficiency Clinical presentation  For a long time hyperprolinaemia type I, which has no clinically relevant phenotype, was not separ- ated from hyperprolinaemia type II. Also, as individuals with PK PNPO PNPO PK PK Membrane-associated phosphatases Cellular uptake PK Pyridoxamine Pyridoxine Pyridoxamine-P Pyridoxine-P Pyridoxal-P Pyridoxal Intracellular pyridoxal-phosphate Fig. 12.2.17  Pyridoxine metabolism. Pyridoxal phosphate (PALP; vitamin B6) is cofactor of transamination and decarboxylation reactions in various pathways including serotonin and dopamine biosynthesis. It is synthesized from dietary pyridoxal, pyridoxamine, and pyridoxine; enzymes involved include pyridoxal kinase (PK) and pyridox(am)ine
5-​phosphate oxidase (PNPO). Source data from Zschocke J, Hoffmann GF (2011). Vademecum metabolicum. Manual of metabolic paediatrics, 3rd edition. Schattauer, Stuttgart.

12.2  Protein-dependent inborn errors of metabolism 1979 hyperprolinaemia type II often have no clinical manifestations, hyperprolinaemia was considered a nondisease. However, on inves- tigation of larger cohorts of affected individuals it became obvious that hyperprolinaemia type II can lead to epilepsy in more than 50% of patients. The epilepsy usually disappears in adulthood. Diagnosis  Plasma concentrations of proline are highly elevated, ex- ceeding 1500 μmol/​litre. Whereas proline is the only amino acid ele- vated in plasma and cerebrospinal fluid, glycine and hydroxyproline are also found elevated in urine as these three amino acids share a common renal tubular transport system. Hyperprolinaemia type II must be distinguished from hyperprolinaemia type I by demon- stration of elevated levels of l-​Δ1-​pyrroline-​5-​carboxylate and/​or by enzyme assay or molecular analysis. Treatment and outcome  Unless a seizure disorder is present, no specific treatment is required. In a child with a seizure disorder, treatment with 5 to 30 mg/​kg body weight per day of pyridoxine in one dose should be started. There are usually no adverse sequelae. Pyridoxal phosphate-​responsive seizures: pyridox(am)ine
5′-​phosphate oxidase deficiency Clinical presentation  Pyridoxal phosphate responsive seizures due to pyridox(am)ine 5′-​phosphate oxidase deficiency (Fig. 12.2.17) result in a most severe early neonatal encephalopathy with con- vulsions, myoclonus, rotatory eye movements, and sudden clonic contractions. Seizures are resistant to conventional anticonvulsant therapy. Many patients are born prematurely, and fetal distress is common, including ‘signs of asphyxia’ and low Apgar scores. Early (lactic) acidosis and hypoglycaemia may be observed. Thus pyri- doxal phosphate-​responsive seizures must enter the differential diagnosis of hypoxic–​ischaemic encephalopathy in prematurely born infants. Diagnosis  The deficiency of pyridox(am)ine 5′-​phosphate oxidase results in combined deficiencies of l-​amino acid decarboxylase, threonine dehydratase, ornithine δ-​aminotransferase, and the gly- cine cleavage enzyme with the concomitant biochemical findings. In addition, some patients display variable lactic acidaemia as well as a tendency to hypoglycaemia. However, no biochemical abnor- mality is 100% specific or sensitive, and a positive response to the drug remains the most reliable indication of pyridoxal phosphate-​ responsive seizures. The diagnosis is confirmed by mutation analysis. Treatment and outcome  Pyridoxal 5′-​phosphate given by nasogas- tric tube is dramatically effective in stopping seizures and improving the appearances of the electroencephalogram. Long-​term treat- ment requires 30 to 60 mg/​kg body weight per day of pyridoxal 5′-​ phosphate in four doses. Doses need to be increased and adjusted to body weight during growth. Patients probably require lifelong supplementation. Breakthrough seizures are an obvious criterion for increasing the dose. So far many questions remain open with regards to prognosis. Serial cognitive assessment is recommended. Defects of glycine and serine metabolism Nonketotic hyperglycinaemia Nonketotic hyperglycinaemia is the second most common dis- order of amino acid metabolism, second to PKU, with an overall worldwide frequency estimated at 1 in 60 000 births. It is caused by deficient activity of the glycine cleavage system which represents the main catabolic route of glycine (Fig. 12.2.18) and is present at high levels in liver, brain, and placenta. In brain, it keeps glycine levels very low, resulting in a typically low cerebrospinal fluid to plasma glycine ratio. Glycine is connected to multiple biochemical pathways. Most im- portant is the generation of methylenetetrahydrofolate. The glycine cleavage system is made up of four mitochondrial proteins, P, H, T, and L. The P protein is a decarboxylase requiring pyridoxal phos- phate. The heat-​resistant H protein contains lipoic acid and carries the aminomethyl moiety. Both proteins are needed to generate CO2 from the carbon-​1 of glycine. The T protein requires tetrahydrofolate and produces methylenetetrahydrofolate from carbon-​2 of glycine. The fourth protein (L protein) is needed to transfer hydrogen from the lipoic acid moiety of the H protein to nicotinamide adenine diphosphate. Clinical presentation  Symptoms of nonketotic hyperglycinaemia are exclusively neurological. Pregnancy and delivery are generally uneventful. Hiccupping in utero maybe recognized retrospectively. Lethargy, convulsions, anorexia, poor feeding, and vomiting pro- gress to coma and unresponsiveness 24 to 48 h after birth. Patients are severely hypotonic. Seizures with hiccupping and myoclonic spasms are prominent, and there is a burst suppression pattern on electroencephalography. Apnoea worsens during the third day of life, mostly requiring ventilation. The mortality rate at this stage is high, especially, if the children are not ventilated. After 2 to 3 weeks the patients improve slightly and no longer require intensive care. However, intellectual development does not occur in survivors, seiz- ures persist, and tendon reflexes are increased. Microcephaly, poor head control, profound retardation, and a picture of spastic cerebral palsy develop. Up to 15% of patients with neonatal presentation have a better recovery after the neonatal period. They have a milder seizure dis- order, usually controlled by benzoate therapy or by a single anticon- vulsant. Most of these patients make some developmental progress, but they are still mentally disabled with developmental quotients varying between 10 and 60. Variant forms of nonketotic hyperglycinaemia present in later in- fancy or childhood with severe seizures, spastic paraparesis, clonus, and extensor plantar responses with modestly raised plasma and cerebrospinal fluid glycine values. Optic atrophy with cerebellar signs has also been described. The outcome is similar to that of pa- tients with the severe form of neonatal nonketotic hyperglycinaemia. Glycine Serine 3-Phosphoserine 3-Phosphohydroxypyruvate 3-Phosphoglycerate Methylene tetrahydrofolate Glycine + H2O Tetrahydrofolate CO2 + H2O Fig. 12.2.18  Reversible glycine cleavage to carbon dioxide and water is illustrated together with reversible interconversion of serine and glycine. These reactions also serve to generate one-​carbon units.
3-​phosphoglycerate (glycolysis) is the ultimate source.

SECTION 12  Metabolic disorders 1980 Diagnosis  Confirmation of diagnosis by enzyme assay and/​or mo- lecular analysis is highly advisable and should be pursued to facili- tate future prenatal diagnosis. Biochemically, nonketotic hyperglycinaemia is characterized by elevated glycine in plasma and in cerebrospinal fluid, with glycine being more elevated in cerebrospinal fluid than in plasma. Plasma glycine is elevated to values of 600 to 1200 µmol/​litre but may vary throughout the day, and can be normal at times. Normal values for cerebrospinal fluid levels of glycine are around 4 to 5 µmol/​litre, the normal cerebrospinal fluid to plasma ratio being less than 0.04. In nonketotic hyperglycinaemia patients, the cerebrospinal fluid to plasma glycine ratio is between 0.07 and 0.30. Great care must be taken to obtain simultaneous plasma and cere- brospinal fluid samples. Diagnostic pitfalls can arise due to postpran- dial blood sampling, blood contamination of the cerebrospinal fluid, profound liver dysfunction, and treatment with valproate. Urine or- ganic acids must be determined to exclude propionic aciduria and methylmalonic aciduria, as well as glyceric aciduria. Activity of the glycine cleavage system can only be reliably measured on liver biop- sies and in direct uncultured chorionic villi for prenatal diagnosis. So far the molecular structures of the P protein, the T protein, and the H protein have been elucidated, allowing molecular diagnosis of defects of these three proteins. Molecular studies have demonstrated a defect of the P protein in about 50 to 60% of patients and in the T protein in about 30% of patients; a few patients were found to have mutations in the GLDC gene leaving approximately 15% of patients with no mutations found after all three genes had been analysed. Treatment and outcome  Therapeutic interventions are unsatisfac- tory. Some damage to the central nervous system may be prenatal. Withdrawal of artificial ventilation and intensive care support should be discussed with the parents of neonates in the apnoeic phase. Once breathing resumes, most patients survive for many years. Plasma glycine can be lowered by exchange transfusion or peri- toneal dialysis but without clinical improvement. Low-​protein diets have only a limited effect on decreasing plasma glycine concentra- tions. Supplying one-​carbon units in the form of methionine or N-​formyltetrahydrofolate has not helped. The combination of so- dium benzoate to increase glycine excretion and diazepines, which compete for inhibitory glycine receptors in the central nervous system, has lowered plasma and cerebrospinal fluid levels of glycine and reduced seizures. Doses up to 600 to 750 mg/​kg per day may be required to lower glycine sufficiently to values between 120 and 280 µmol/​litre. At such high, potentially toxic doses monitoring of benzoate levels is advised, nevertheless gastric irritation is very fre- quent and gastric protection with H2-​antihistamine or proton pump inhibitors is preventively recommended. Most patients need gastric tube feeding or gastrostomy. Gastro-​ oesophageal reflux develops frequently, and many patients benefit from a Nissen fundoplication. Recurrent bronchitis is a major problem and bronchopneumonia is frequently the cause of death. For patients with mild nonketotic hyperglycinaemia, management of the hyperactivity can be a major challenge. 3-​Phosphoglycerate dehydrogenase deficiency Serine is synthesized from the glycolytic intermediate 3-​phosphoglycerate by 3-​phosphoglycerate dehydrogenase yielding 3-​phosphohydroxypyruvate (Fig. 12.2.18). Deficiency of this en- zyme leads to serine deficiency. Clinical presentation  Patients with serine deficiency due to 3-​ phosphoglycerate dehydrogenase deficiency have congenital micro- cephaly. They develop severe psychomotor retardation with spastic tetraparesis and severe microcephaly. Seizures usually start in in- fancy as West’s syndrome with hypsarrhythmia. The MRI scan is characterized by striking delayed or absent myelination, with subse- quent cortical and subcortical atrophy. Variable symptoms include cataract, hypogonadism, megaloblastic anaemia, and nystagmus. Diagnosis  Serine deficiency in 3-​phosphoglycerate dehydrogenase deficiency is most reliably diagnosed in cerebrospinal fluid with values less than 14 µmol/​litre (normal cerebrospinal fluid serine 42–​86 µmol/​litre in infancy). Serine values in fasting plasma are also reduced (28–​64 µmol/​litre, controls 70–​187 µmol/​litre). However, nonfasting plasma levels can be normal. Treatment and outcome  l-​Serine should be administered orally until normalized (300–​500 mg/​kg per day). If seizures persist, gly- cine should be added up to 300 mg/​kg per day. A very satisfactory outcome was achieved by antenatal treatment in one patient. Defects of γ-​aminobutyric acid metabolism GABA is formed from glutamate in the brain by the cytosolic en- zyme glutamate decarboxylase, which requires pyridoxal phos- phate (Fig. 12.2.19). Glutamate can be regenerated from GABA by transamination with ketoglutarate (GABA transaminase), which is also pyridoxal phosphate dependent. The other product is succinic semialdehyde, which is dehydrogenated to succinate and enters the citric acid cycle. Deficiency of succinic semialdehyde dehydrogenase leads to formation and excretion of 4-​hydroxybutyric acid. GABA transaminase deficiency Some patients with GABA transaminase deficiency presented with a fatal neonatal encephalopathy, characterized by seizures, hypotonia, hyperreflexia, a high-​pitched cry (cat cry), and accelerated growth. The diagnosis can be suspected from significantly elevated levels of GABA (both free and total), as well as β-​alanine and homocarnosine Fig. 12.2.19  Synthesis and catabolism of 4-​aminobutyric acid (GABA). The enzymes recognized for known monogenic disorders in humans are shown in boxes: GAD, glutamic acid decarboxylase deficiency, GT, GABA transaminase deficiency, SSADH, succinic semialdehyde dehydrogenase deficiency. The cofactor vit. B6 (vitamin B6) is underlined. Source data from Zschocke J, Hoffmann GF (2011). Vademecum metabolicum. Manual of metabolic paediatrics, 3rd edition. Schattauer, Stuttgart.

12.2  Protein-dependent inborn errors of metabolism 1981 in cerebrospinal fluid. Plasma levels of these amino acids are also increased, but not as significantly. The diagnosis must be confirmed by enzyme assay and possibly mutation analysis, both of which can also be used for prenatal diagnosis. Unfortunately, there is no rational treatment available. Recently, in a family suffering from encephalomyopathic mitochondrial DNA depletion syndrome, the underlying molecular defect was detected in GABA transaminase encoded by 4-​aminobutyrate aminotransferase (ABAT). Apparently, GABA transaminase connects GABA and nucleoside metabolism resulting in a neurometabolic disorder including mitochondrial DNA depletion. Succinic semialdehyde dehydrogenase deficiency
(4-​hydroxybutyric aciduria) Clinical presentation  The clinical presentation of succinic semialdehyde dehydrogenase deficiency is highly heterogeneous, even within sibships. The cardinal manifestations are complex and rather nonspecific: hypotonia and delay of motor, mental, and fine motor skills and language. Ataxia and/​or seizures occur in about half of the patients. Hyperkinesis and aggressive and autistic behaviour are additional features. MRI studies show bilateral globus pallidus abnormalities but again not constantly. Diagnosis  Diagnosis is usually suspected by demonstrating in- creased levels of γ-​hydroxybutyrate by organic acid analysis. It is confirmed by enzyme assay and preferentially additional mutation analysis. Treatment and outcome  A common treatment for succinic semialdehyde dehydrogenase deficiency is the antiepileptic drug vigabatrin. The results have been encouraging in some patients, but of little value or even detrimental in others. Seizures respond to con- ventional anticonvulsants. A ketogenic diet shows promise. Succinic semialdehyde dehydrogenase deficiency is a slowly progressive en- cephalopathy in childhood; it eventually stabilizes in most patients. Defects of trans-​sulphuration and remethylation The trans-​sulphuration pathway transfers the sulphur of methio- nine to serine, thus generating cysteine (Fig. 12.2.20). Methionine adenosyltransferase, with widely distributed isoenzyme forms, produces S-​adenosylmethionine, the donor in a variety of methy- lation reactions. S-​Adenosylhomocysteine is cleaved to homocyst- eine, the sulphydryl compound that exists in reversible equilibrium with its disulphide homocystine. Half of the homocysteine formed goes through the trans-​sulphuration pathway and the other half takes a methyl group from betaine (betaine methyltransferase) or 5-​methyltetrahydrofolic acid (methionine synthase). The latter is a cobalamin-​dependent enzyme which is functionally impaired in de- fects of vitamin B12 metabolism. In addition, methionine synthase reductase is necessary to keep the methionine synthase-​bound co- balamin in a functional state. The remethylation of homocysteine is also impaired if the activity of the reductase that generates 5-​ methyltetrahydrofolate is inadequate. When accumulation of homo- cysteine and its products homocystine and the also formed mixed disulphide results from defects of homocysteine remethylation, plasma methionine concentrations are low. They are high when homocystine accumulates from impaired activity of cystathionine β-​synthase. Classic homocystinuria: cystathionine β-​synthase deficiency Clinical presentation Untreated classic homocystinuria is a slowly progressive, devastating multiorgan disorder. First symptoms in childhood are a rapidly pro- gressive myopia and lens dislocation. Lens dislocation usually oc- curs in preschool years, but later dislocation is well recognized in pyridoxine-​responsive patients, and a few have not developed it even in adult life. Monocular and binocular blindness has been rela- tively frequent due to secondary glaucoma, staphyloma formation, buphthalmos, and retinal detachment. In the older child, skeletal abnormalities and learning difficul- ties become obvious. Genu valgum and pes cavus are usually the first signs of skeletal changes, which include osteoporosis and spon- taneous crush vertebral fractures. The common abnormalities seen in Marfan’s syndrome—​high-​arched palate, pectus excavatum or carinatum, genu valgum, pes cavus or planus, scoliosis—​are all well recognized in homocystinuria. Arachnodactyly is less common and the fingers not infrequently (and elbows occasionally) show mild flexion contractures. Skeletal disproportion with a crown–​ pubis length less than the pubis–​heel length is usual (Fig. 12.2.21). Learning difficulties affect two-​thirds of patients. Patients responsive to pyridoxine (vitamin B6) (see following ‘Treatment and outcome’ subsection) have generally higher IQ values than nonresponsive patients. Seizures affect about one-​fifth of patients and a few show extrapyramidal features, sometimes with severe involuntary move- ments. Psychiatric disturbances have also been described. Thromboembolism is a major cause of morbidity and the main cause of high premature mortality. Thromboses have been described in a wide variety of arteries and veins: cerebral, coronary, mesen- teric, renal, and peripheral. –CH3 –CH3 3* Tetrahydrofolate Glycine SO4 Cysteine Cystathionine Serine Homocysteine 1 S-adenosyl homocysteine S-adenosyl methionine Methionine 4 Dimethyl glycine Methyltetrahydrofolate Methylene tetrahydrofolate Betaine –CH3 CH2 2 Fig. 12.2.20  The trans-​sulphuration pathway from methionine to cysteine is shown on the right and the remethylation of homocysteine on the left. Asterisked enzymes are: 1, cystathionine β-synthase; 2, methylene tetrahydrofolate reductase, 3, methionine synthase; and 4, betaine methyltransferase.

SECTION 12  Metabolic disorders 1982 Diagnosis Elevated plasma methionine values between 100 and 500 µmol/​ litre (sometimes higher) are seen with plasma total homocysteine values of 50 to 200 µmol/​litre. A mixed disulphide (half homocyst- eine, half cysteine) is always present at concentrations somewhat below those of homocystine. Diagnosis requires the determination of fasting quantitative plasma amino acids, as well as plasma total homocysteine. Total homocysteine measured by HPLC includes both homocysteine moieties of homocystine, the homocysteine moiety of the mixed disulphide, and the homocysteine bound to plasma proteins. The urine gives a positive nitroprusside test (it is also positive in cystinuria). However, this test can be falsely negative. Unfortunately, methionine elevation is unreliable in the early days of life, hampering the possibility of newborn screening. This can be reliably performed by screening for homocystinuria but still exclu- sively detects the more severely pyridoxine nonresponsive patients. Confirmation of the diagnosis can be performed by enzyme assay using cultured skin fibroblasts and/​or mutation analysis, which al- lows prenatal diagnosis. Treatment and outcome Optimal outcome of treatment depends on its earliest possible intro- duction. Treatment is focused on correcting homocysteine levels; lifelong monitoring is essential. In about one-​half of the patients, oral pyridoxine rapidly reduces methionine and homocysteine to near-​normal values. The first treatment should be to try using doses from as low as 50 mg in infants to 1000 mg/​day in older children or adults and reducing the dose if a response is achieved; 5 to 10 mg/​ day of folic acid should also be given. Very large sustained doses (1000 mg/​day or more) in adults can cause peripheral neuropathy. Those responding only partially or not at all to pyridoxine require a very low-​protein diet supplemented with a methionine-​free amino acid supplement, minerals, and vitamins. Biochemical control may only be achieved in older children and adults on natural protein in- takes of 5 to 10 g/​day. Plasma cystine should be maintained in the normal range and supplementation should be considered. Both folic acid (5–​10 mg/​day) and betaine (up to 12 g/​day) can further re- duce plasma homocysteine levels but may produce large elevations of plasma methionine. Low red-​cell folate values occur and even megaloblastic anaemia. Low serum vitamin B12 values also occur and should be corrected. Treatment started early can prevent or re- duce the clinical sequels and lower the incidence of vascular events throughout life; many patients have a normal life expectancy. Methylene tetrahydrofolate reductase deficiency Clinical presentation Neurological features predominate with psychomotor retardation, seizures, abnormalities of gait, and psychiatric disturbance. The age of symptom development varies widely from infancy with a pro- gressive encephalopathy with apnoea, seizures, and microcephaly to adulthood with ataxia, motor abnormalities, psychiatric symptoms, subacute degeneration of spinal cord, and cerebrovascular events. Demyelination occurs and the changes may resemble the classic findings of subacute combined degeneration seen in vitamin B12 de- ficiency. The risk of vascular disease is high. Diagnosis Plasma methionine concentrations are below normal and plasma homocysteine concentrations are in the range 20 to 200 µmol/​litre with an elevated excretion of 15 to 600 µmol/​day. As homocysteine is easily missed on amino acid analysis, quantitative determination of total homocysteine by HPLC is the most important clue to diag- nosis. There is no megaloblastic anaemia. The enzyme can be as- sayed in liver, leucocytes, lymphocytes, or fibroblasts also allowing prenatal diagnosis. Treatment and outcome Betaine in large doses (20–​150 mg/​kg per day) effectively lowers plasma homocysteine and raises plasma methionine. Other treat- ments tried alone or in combination include folinic acid, vitamin B12, pyridoxine, and methionine. Some have suggested a cocktail of all these treatments. It is difficult to be sure of clinical success. Deficiencies of methionine synthase reductase (cobalamin E defect) and methionine synthase (cobalamin G defect) Clinical and biochemical findings of methionine synthase reduc- tase (cobalamin E defect) and methionine synthase (cobalamin G defect) deficiencies are virtually identical. Characteristic find- ings are developmental delay and megaloblastic anaemia, but the onset may be in later in childhood with dementia and spasticity. Retinal degeneration, cardiac defects, and haemolysis have been described. Megaloblastic anaemia occurs in almost all patients. Biochemical findings include low plasma methionine and raised homocysteine as well as homocystine in plasma and urine. Methylmalonic acid should be measured in urine to exclude other cobalamin defects (see ‘Methylmalonic aciduria’). Methionine synthase can be as- sayed in liver or fibroblasts and antenatal diagnosis has been carried out on cultured amniocytes. Cells with the cobalamin E defect re- quire specific reducing conditions to demonstrate the deficient en- zyme activity. Molecular diagnosis is possible for both conditions. Treatment involves large doses of hydroxocobalamin with betaine Fig. 12.2.21  A patient with cystathionine synthase deficiency. Note the kyphosis and disproportionate short trunk.

12.2  Protein-dependent inborn errors of metabolism 1983 and possibly folinic acid. Success of therapy and outcome is variable and often unfavourable. Other defects of sulphur amino acid metabolism Among several additional defects known, cystathioninuria due to cystathionase deficiency is probably clinically harmless. Cystathionine in excess of 1 g/​day may be excreted at clearance values close to the glomerular filtration rate. Methionine adenosyltransferase deficiency causes raised plasma methionine levels (up to 1200 µmol/​litre; normal 15–​30 µmol/​litre) and appears to be harmless in most patients. The enzyme defect is partial. Severe deficiency of methionine adenosyltransferase I/​III may be associated with demyelination and neurological features. In such patients, treatment with S-​adenosylmethionine (400 mg of the toluene sulphonate, twice daily) is an option. Glycine N-​methyltransferase deficiency is very rare and was dem- onstrated in children with mild liver disease. Biochemical findings included elevated plasma methionine and S-​adenosylmethionine levels. Similarly rare appear to be patients affected with S-​ adenosylhomocysteine hydrolase. Pathology and clinical findings are significant in liver, muscle, and the nervous system. Biochemical findings are complex, with elevated plasma methionine, S-​ adenosylhomocysteine, and S-​adenosylmethionine levels. Total homocysteine and cystathionine may also be slightly elevated. FURTHER READING Ando T, et  al. (1971). Propionic acidemia in patients with ketotic hyperglycinemia. J Pediatr, 78, 827–​32. Besse A, et al. (2015) The GABA transaminase, ABAT, is essential for mitochondrial nucleoside metabolism. Cell Metab, 21, 417–​27. Bickel H, Gerrard J, Hickmans E (1953). Influence of phenylalanine on PKU. Lancet, 2, 812–​13. Baumgartner MR, et  al. (2014). Proposed guidelines for the diag- nosis and management of methylmalonic and propionic acidemia. Orphanet J Rare Dis, 9, 130. Blau N, et al. (eds) (2014). Physician’s guide to the diagnosis, treatment, and follow-​up of inherited metabolic disease, 2nd edition. Springer, Heidelberg. Brown GK, et al. (1984). Malonyl coenzyme A decarboxylase defi- ciency. J Inherit Metab Dis, 7, 21–​6. Bursell MK, et al. (2011). Adenosine kinase deficiency disrupts the methionine cycle and causes hypermethioninemia, encephalopathy, and abnormal liver function. Am J Hum Genet, 78, 507–​15. Canavan MM (1931). Schilder’s encephalitis perioxalis diffusa. Arch Neurol Psychiatry, 25, 299. Danhauser K, et al. (2012). DHTKD1 mutations cause 2-​aminoadipic and 2-​oxoadipic aciduria. Am J Hum Genet, 91, 1082–​7. Dewey KG, et al. (1996). Protein requirements of infants and children. Eur J Clin Nutr, 50 Suppl 1, S119–​47. Dixon MA, Leonard JV (1992). Intercurrent illness in inborn errors of intermediary metabolism. Arch Dis Child, 67, 1387–​91. Edvardson S, et al. (2013). Agenesis of corpus callosum and optic nerve hypoplasia due to mutations in SLC25A1 encoding the mitochon- drial citrate transporter. Am J Hum Genet, 50, 240–​5. Ensenauer R, et al. (2004). A common mutation is associated with a mild, potentially asymptomatic phenotype in patients with isovaleric acidemia diagnosed by newborn screening. Am J Hum Genet, 75, 1136–​42. Ferdinandusse S, et al. (2013). HIBCH mutations can cause Leigh-​like disease with combined deficiency of multiple mitochondrial respira- tory chain enzymes and pyruvate dehydrogenase. Orphanet J Rare Dis, 8, 188. Fernandes J, et al. (eds) (2006). Inborn metabolic diseases, 4th edition. Springer, Heidelberg. Ferre S, et al. (2014). Mutations in PCBD1 cause hypomagnesemia and renal magnesium wasting. J Am Soc Nephrol, 25, 574–​86. Garrod AE (1902). The incidence of alkaptonuria. A study in chemical individuality. Lancet, 2, 1616–​20. Garrod AE (1909). Inborn errors of metabolism. Oxford University Press, Oxford. Goodman SI, et al. (1975). Glutaric aciduria: a ‘new’ disorder of amino acid metabolism. Biochem Med, 12, 12–​21. Guthrie R, Susi A (1963). A simple phenylalanine method for detecting PKU in large populations of newborn infants. Pediatrics, 32, 338–​43. Haack TB, et al. (2015). Deficiency of ECHS1 causes mitochondrial encephalopathy with cardiac involvement. Ann Clin Translat Neurol, 2, 492–​509. Hoffmann B, et al. (2006). Impact of longitudinal plasma leucine levels on the intellectual outcome in patients with classic MSUD. Pediatr Res, 59, 17–​20. Häberle J, et al. (2012). Suggested guidelines for the diagnosis and management of urea cycle disorders. Orphanet J Rare Dis, 7, 32. Hoffmann GF (1994). Selective screening for inborn errors of metabolism—​past, present and future. Eur J Pediatr, 153 Suppl 1, S2–​8. Hoffmann GF, Blau N (eds) (2014). Congenital neurotransmitter disor- ders. Nova Science Publishers, New York. Hoffmann GF, Surtees RA, Wevers RA (1998). Cerebrospinal fluid investigations for neurometabolic disorders. Neuropediatrics, 29, 59–​71. Hoffmann GF, et al. (1994). Neurological manifestations of organic acid disorders. Eur J Pediatr, 153 Suppl 1, S94–​100. Hörster F, et  al. (2007). Long-​term outcome in methylmalonic acidurias is influenced by the underlying defect (mut0, mut−, cblA, cblB). Pediatr Res, 62, 225–​30. Hunt AD Jr, et al. (1954). Pyridoxine dependency: report of a case of intractable convulsions in an infant controlled by pyridoxine. Pediatrics, 13, 140–​5. Jakobs C, ten Brink H, Stellaard F (1990). Prenatal diagnosis of in- herited metabolic disorders by quantitation of characteristic metab- olites in amniotic fluid: facts and future. Prenat Diagn, 10, 265–​71. Karnebeek CD, et  al. (2014). Mitochondrial carbonic anhydrase VA deficiency resulting from CA5A alterations presents with hyperammonemia in early childhood. Am J Hum Genet, 94, 453–​61. Koch R, et al. (2003). The maternal PKU international study: 1984–​ 2002. Pediatrics, 112, 1523–​9. Kölker S, et al. (2006). Natural history, outcome, and treatment efficacy in children and adults with glutaryl-​CoA dehydrogenase deficiency. Pediatr Res, 59, 840–​7. Kölker S, et al. (2015). The phenotypic spectrum of organic acidurias and urea cycle disorders. Part 1: the initial presentation. J Inherit Metab Dis, 38, 1041–​57. Kölker S, et al. (2015). The phenotypic spectrum of organic acidurias and urea cycle disorders. Part 2: the evolving clinical phenotype. J Inherit Metab Dis, 38, 1059–​74.

SECTION 12  Metabolic disorders 1984 Lenke R, Levy HL (1980). Maternal PKU and hyperphenylal­ aninemia: an international study of treated and untreated pregnan- cies. N Engl J Med, 303, 1202–​8. Ly TB, et  al. (2003). Mutations in the AUH gene cause 3-​ methylglutaconic aciduria type I. Hum Mutat, 21, 410–​17. Menkes JH, Hurst PL, Craig JM (1954). New syndrome: progressive familial infantile cerebral dysfunction associated with an unusual urinary substance. Pediatrics, 14, 462–​7. Millington DS, et  al. (1990). Tandem mass spectrometry:  a new method for acylcarnitine profiling with potential for neonatal screening for inborn errors of metabolism. J Inherit Metab Dis, 13, 321–​4. Nota B, et al. (2013). Deficiency in SLC25A1, encoding the mitochon- drial citrate carrier, causes combined d-​2-​ and l-​2-​hydroxyglutaric aciduria. Am J Human Genet, 92, 627–​31. Nyhan WL, Barshop BA, Al-​Aqueel A (2012). Atlas of metabolic dis- eases. Hodder Headline, London. Oberholzer VG, et al. (1967). Methylmalonic aciduria: an inborn error of metabolism leading to chronic metabolic acidosis. Arch Dis Child, 42, 492–​504. Peters H, et al. (2014). ECHS1 mutations in Leigh disease: a new in- born error of metabolism affecting valine metabolism. Brain, 137, 2903–​8. Phornphutkul C, et al. (2002). Natural history of alkaptonuria. N Engl J Med, 347, 2111–​21. Posset R, et al. (2019). Impact of Diagnosis and Therapy on Cognitive Function in Urea Cycle Disorders. Ann Neurol, 86, 116–28. Prietsch V, et al. (2002). Emergency management of inherited meta- bolic disease. J Inherit Metab Dis, 25, 531–​46. Salomons GS, et al. (2007) Clinical, enzymatic and molecular charac- terization of nine new patients with malonyl-​coenzyme A decarb- oxylase deficiency. J Inherit Metab Dis, 30, 23–​8. Schaefer F, et al. (1999). Dialysis in neonates with inborn errors of me- tabolism. Nephrol Dial Transplant, 14, 910–​18. Schulze A, et al. (2003). Expanded newborn screening for inborn errors of metabolism by electrospray ionization-​tandem mass spectrom- etry: results, outcome, and implications. Pediatrics, 111, 1399–​406. Scriver CR, et al. (eds) (2001). The metabolic and molecular bases of inherited disease, 8th edition. McGraw-​Hill, New York. Strauss KA, et al. (2006). Elective liver transplantation for the treat- ment of classical maple syrup urine disease. Am J Transplant, 6, 557–​64. Suwannarat P, et  al. (2005). Use of nitisinone in patients with alkaptonuria. Metabolism, 54, 719–​28. Tanaka K, et al. (1966). Isovaleric acidemia: a new genetic defect of leucine metabolism. Proc Natl Acad Sci, 56, 236–​42. Unsinn C, et al. (2016). Clinical course of 63 patients with neonatal onset urea cycle disorders in the years 2001–2013. Orphanet J Rare Dis, 11(1), 116. Wilson JMG, Jungner G (1968). Principles and practice of screening for disease. Public Health Papers No. 34. World Health Organization, Geneva. Wolf B, et al. (1983). Biotinidase deficiency: the enzymatic defect in late-​ onset multiple carboxylase deficiency. Clin Chim Acta, 13, 273–​81. Wolf B, et al. (1983). Deficient biotinidase activity in late-​onset mul- tiple carboxylase deficiency. N Engl J Med, 308, 161. Wortmann S, et al. (2012). Mutations in the phospholipid remodel- ling gene SERAC1 impair mitochondrial function and intracellular cholesterol trafficking and cause dystonia and deafness. Nat Genet, 44, 797–​802. Wortmann S, et al. (2013). 3-​Methylglutaconic aciduria—​lessons from 50 genes and 977 patients. J Inherit Metab Dis, 36, 913–​21. Zschocke J, Hoffmann GF (2011). Vademecum metabolicum. Manual of metabolic paediatrics, 3rd edition. Schattauer, Stuttgart. Zschocke J, et  al. (2000). Progressive infantile neurodegeneration caused by 2-​methyl-​3-​hydroxybutyryl-​CoA dehydrogenase defi- ciency: a novel inborn error of branched chain fatty acid and isoleu- cine metabolism. Pediatr Res, 48, 852–​5.

12.3 Disorders of carbohydrate metabolism 1985

12.3 Disorders of carbohydrate metabolism 1985

12.3.1 Glycogen storage diseases 1985 Robin H. Lac

12.3.1 Glycogen storage diseases 1985 Robin H. Lachmann and Timothy M. Cox

CONTENTS 12.3.1 Glycogen storage diseases  1985 Robin H. Lachmann and Timothy M. Cox 12.3.2 Inborn errors of fructose metabolism  1993 Timothy M. Cox 12.3.3 Disorders of galactose, pentose, and
pyruvate metabolism  2003 Timothy M. Cox 12.3.1  Glycogen storage diseases Robin H. Lachmann and Timothy M. Cox ESSENTIALS Glycogen is a highly branched glucose polymer with a compact structure found predominantly in liver and muscle. Liver glycogen is important in the maintenance of euglycaemia during fasting; muscle glycogen is an immediate source of glucose for energy production during exercise. Genetic disorders affecting proteins that regulate glycogen synthesis and breakdown cause marked accumulation of glycogen in diverse tissues, and pathological glycogen often has an abnormal macromolecular structure. Depending on the enzyme system involved, diseases of glycogen metabolism principally affect liver and muscle. Clinical features are related to accumulation of glycogen in tissues and/​or failure to release glucose. Glycogen storage is associated with organomegaly and tissue injury: this usually affects the liver and/​or muscles, including the heart, but in severe cases other organs may be involved. Fasting hypoglycaemia occurs where hepatic breakdown of glycogen is impaired. Hyperlipidaemia, hyperlactataemia and hyperuricaemia leading to gout occur in those disorders with a major liver component, and poor metabolic control is associated with de- velopment of hepatic adenomas and frank liver cancers. Glycogen diseases that affect muscle usually present with rhabdomyolysis, ex- ercise intolerance, and muscle pain or weakness. Recently, several inherited multisystem disorders with neurodegeneration, such as polyglucosan body disease and Lafora’s disease, have been shown to result from abnormal glycogen structures in diverse cell types, including neurons. Formerly, diseases of glycogen metabolism were diagnosed by showing excess storage of glycogen in the tissue of interest, accom- panied by reduced activity of particular glycogen-​metabolizing en- zymes. Currently, where available, molecular analysis of genomic DNA is the preferred method for providing a definitive diagnosis. The mainstay of treatment of glycogen diseases affecting the liver is dietary, including pre-​emptive management of hypoglycaemia that is readily provoked by fasting. In infants and children, continuous provision of carbohydrate by the nasogastric route may be required to maintain euglycaemia. Adults can usually be managed by a com- bination of frequent sugary snacks and the use of uncooked corn- starch as a slow-​release source of glucose. Dietary interventions may also ameliorate some of the glycogen diseases that affect muscle, and weakness and pain after exertion can be improved by graduated exercise programmes in some patients. Introduction Maintenance of blood glucose is an essential homeostatic func- tion:  profound hypoglycaemia causes encephalopathy and car- diac arrhythmias and is rapidly fatal if not treated promptly. Two processes are involved in maintaining normal blood glu- cose during periods of fasting:  de novo synthesis of glucose (gluconeogenesis) and the release of glucose from carbohydrate stores (glycogenolysis). The body stores carbohydrate in the form of glycogen, a branched polymer of glucose. Glycogen stores in the liver are used to maintain normoglycaemia, but muscle also stores glycogen for its own use as an energy source during exer- cise. In this chapter, we will discuss the metabolism of glycogen and the inherited metabolic disorders which affect its synthesis and breakdown. Glycogen Glycogen allows for the compact storage of glucose in a form that has a minimal osmotic effect but which is readily accessible and meta- bolically active. The core of a glycogen molecule is a small protein, 12.3 Disorders of carbohydrate metabolism

SECTION 12  Metabolic disorders 1986 glycogenin. Branched chains of polymerized α-​d-​glucose units are covalently attached to this at their reducing termini (Fig. 12.3.1.1). The glucose molecules in glycogen chains are linked to each other by α-​1,4 glycosidic bonds with α-​1,6 bonds at the branch points (Fig.  12.3.1.1b). Glycogen molecules can contain up to 60 000 glucose molecules, have a molecular weight of several mil- lion daltons, and are visible to the electron microscope. The liver and muscle contain between 200 and 300 g of glycogen. The arbor- ization of the molecule, with large numbers of long outer chains that terminate in nonreducing glucose residues means that the en- zymes of glycogen degradation can rapidly release large quantities of glucose. Glycogen storage diseases (GSDs) can be caused by defects in glycogen synthesis or glycogen breakdown (Fig. 12.3.1.2). The stored glycogen may have a normal or an aberrant structure. Depending on the enzymatic defect, glycogen metabolism in the liver, muscle, or both tissues may be affected. Glycogen synthesis A glycogen molecule starts life with the autoglycosylation of a glycogenin molecule at a specific tyrosine residue. This primer molecule is then acted on by glycogen synthase which uses uridine diphosphoglucose molecules to form the α-​1,4 linkages of the nas- cent sugar chain. The α-​1,6 branch points required for the complex structure of glycogen are formed by ‘branching enzyme’ (amylo-​ (1,4 → 1,6) transglucosidase). It transfers a minimum of six α-​1,4-​linked glucose units from the distal ends of glycogen chains to a 1,6 pos- ition on the same or a neighbouring chain. Glycogen synthase is a highly regulated enzyme complex that exists in distinct isoforms in muscle and liver. Glycogen synthase is subject to phosphorylation control that inhibits its activity: the phos- phorylation of at least nine serine residues is brought about by pro- tein kinases and reversed by protein phosphatase I. This inhibition can be overcome by the allosteric activator, glucose 6-​phosphate. (a) CH2OH CH2OH CH2OH CH2O CH2OH CH2OH O O O O O OH OH OH OH OH O O OH OH O O OH OH O O OH OH O O Glycogenin OH OH OH ...O O O OH OH ...O CH2OH CH2OH (b) Fig. 12.3.1.1  (a) A cross-​sectional view of glycogen, showing the core glycogenin protein surrounded by branches of glucose units, up to 60 000 of which can be contained within a glycogen granule. (b) Linear chains of glucose are formed by α-​1,4 glycosidic bonds, with α-​1,6 bonds at the branch points.

12.3.1  Glycogen storage diseases 1987 Regulation of glycogen synthase is important in maintaining blood glucose. Glucagon and adrenaline indirectly inhibit glycogen synthase by maintaining protein phosphatase I in its inactive con- figuration and promoting phosphorylation of glycogen synthase. Insulin stimulates glycogen synthase by activating protein phos- phatase I and promoting its dephosphorylation. Glycogen breakdown Two enzymes are involved in the breakdown of glycogen in the cyto- plasm: phosphorylase and debranching enzyme. Other enzymes are required to then produce free glucose. Phosphorylase sequentially removes glucose 1-​phosphate units from the α-​1,4-​linked chains of glycogen. Debranching enzyme pos- sesses transferase and α-​1,6-​glucosidase activities. When phosphor- ylase has degraded glycogen chains to within four α-​1,4-​glucosyl units of an α-​1,6 linkage, three glucose residues are transferred to the end of another chain by the glycosyltransferase activity. Debranching enzyme then hydrolyses the remaining α-​1,6 bond to release free glucose using its amylo-​1,6-​glucosidase activity. Debranching en- zyme also cleaves the unique glucosyl–​tyrosine linkage that anchors the terminal reducing glucose unit to glycogenin. The main product of glycogen breakdown in muscle and liver is glucose 1-​phosphate. Glucose 1-​phosphate is a key intermediate of glycolysis, gluconeogenesis, glycogenolysis, and the pentose phos- phate pathway, but cannot be transferred outside the cell. However, after conversion to glucose 6-​phosphate by phosphoglucomutase, free glucose is formed by the action of glucose 6-​phosphatase. Glucose 6-​phosphatase exists as a multicomponent complex in the endoplasmic reticulum of hepatocytes and, to a lesser extent, in renal tubular cells—​it is not found in muscle. The complex contains glu- cose 6-​phosphatase, several proteins that facilitate the transport of glucose, glucose 6-​phosphate, and phosphate, as well as other stabil- izing and regulatory units. Hepatic activity of glucose 6-​phosphatase is the predominant metabolic source of blood glucose. In muscle, glucose 6-​phosphate obtained from the breakdown of glycogen is used directly as an energy source via glycolysis. Glycolytic defects can also affect glycogen metabolism in muscle (e.g. phosphofructokinase-​1 deficiency). Phosphorylase in liver and skeletal muscle is activated by phos- phorylation in response to hormonal or neural stimulation—​a com- plex process that is mediated by the hepatic and muscle isoforms of phosphorylase kinases. Phosphorylase kinase is in turn regulated by protein kinase A (cAMP-​dependent protein kinase), calcium and kinase activation of calmodulin, and protein phosphatases 1 and 2A. Another enzyme, acid α-​1,4-​glucosidase (otherwise known as acid maltase), also has an important role in the metabolism of glycogen, but in the lysosome not the cytoplasm. This lysosomal hydrolase is present in all cells except erythrocytes. It has no role in Glycogen synthase Branching enzyme UDP-glucose pyrophosphorylase Acid α-1,4 glucosidase (lysosomes) (cytosol) amylo 1,6-glucosidase Phosphorylase Debranching enzyme Phosphoglucomutase Phosphohexose isomerase Phosphofructokinase (glycolysis) Glucose 6-phosphatase Hexokinase/ glucokinase Glucose 1-phosphate Glucose 6-phosphate Fructose 6-phosphate Glucose 1-phosphate Glucose 6-phosphate Fructose 6-phosphate Fructose 1,6-diphosphate Fructose 1,6-diphosphate Phosphoglucomutase Phosphohexose isomerase Fructose diphosphatase (gluconeogenesis) Glycogenin primer Uridine diphosphoglucose GLYCOGEN GLYCOGEN SYNTHESIS BREAKDOWN GLUCOSE Fig. 12.3.1.2  The synthesis and degradation of glycogen.

SECTION 12  Metabolic disorders 1988 glycolysis, but hydrolyses the glycogen which is constantly entering the lysosome via autophagy. This pathway seems to be particularly important in muscle. Discovery and classification of glycogen storage diseases The study of patients with GSDs has played an essential part in elucidating the biochemical pathways described in the previous sections. In 1929, von Gierke described ‘hepatonephromegalia glykogenica’ with glycogen storage in the liver and kidney. Twenty years later, the husband and wife team of CF and GT Cori showed that this disease was due to deficiency of glucose-​ 6-​phosphatase activity (GSD Ia). However, some patients with glycogen storage in the liver had normal glucose-​6-​phosphatase activity, and were later shown to have glucose-​6-​phosphate trans- porter defects (GSD Ib). Other patients were described who stored abnormal forms of glycogen (GSD IV), or who accumulated glycogen in muscle as well as (GSD III), or instead of (GSD V), liver, or where the primary site of glycogen storage was the lysosome rather than the cytoplasm (GSD II). More recently, the recognition of polyglucosan bodies has led to the description of new diseases which involve glycogen me- tabolism, as well as expanding the phenotype of classical GSD IV. A summary of GSDs, their enzymology, and principal features is given in Table 12.3.1.1. Although GSDs have traditionally been split into those that affect the liver and those that affect muscle, many are in reality multisystem disorders. The most important clinical features, however, remain fasting hypoglycaemia and Table 12.3.1.1  The glycogen storage disorders: genetic and enzymatic defects and principal clinical features GSD designation Gene Locus Common term/​implicated protein Supplementary Clinical features 0 (L) 0 (M) GYS1 GYS2 12p12.1 19q13.33 Glycogen synthase Glycogen synthase Liver isozyme Muscle isozyme Hepatomegaly, hypoglycaemia Cardiomyopathy Ia Ib G6P6 SLC37A4 17q21.31 11q23.3 Von Gierke’s disease Glucose 6-​phosphatase Glucose 6-​phosphatase translocase Hypoglycaemia, hyperlacticaemia, hyperuricaemia, hypertriglyceridaemia Hepatomegaly, hepatic adenomas Renal failure GSD Ib also has neutropenia and colitis II GAA 17q25.3 Pompe’s disease Lysosomal (α-​glucosidase) Cardiomyopathy (infantile form) Proximal myopathy, hypoventilation III AGL 1p21.2 Cori–​Forbes disease (Limit dextrinosis) Debranching enzyme Hypoglycaemia, hepatomegaly, cardiomyopathy, myopathy IV GBE1 3p12.2 Andersen’s disease (Amylopectinosis) Branching enzyme Infantile liver failure. Cardiomyopathy, myopathy Adult polyglucosan body disease (neurogenic bladder, spastic paraparesis and peripheral neuropathy) V PYGM 11q12–​q13.3 McCardle’s disease Glycogen phosphorylase (muscle) Exercise intolerance, rhabdomyolysis VI PYGL 14q22.1 Hers’ disease Glycogen phosphorylase (liver) Hypoglycaemia, hepatomegaly (very rare—Mennonite) VII PFKM 12q13.3 Tarui’s disease Phosphofructokinase (muscle) Exercise intolerance, rhabdomyolysis Haemolytic anaemia [VIII] N/​A N/​A [See Hers’ disease] N/​A (obsolete) IX a1 IX a2 IX b IX c PHKA1 PHKA2 PHKB PHKG2 Xq12–​q13.1 Xp22.13 16q12.1 16p12.2–​11.2 Phosphorylase kinase α1 subunit Phosphorylase kinase α2 subunit Phosphorylase kinase β subunit Phosphorylase kinase γ subunit Muscle (regulatory) Liver (regulatory) Liver and muscle (regulatory) Liver (catalytic subunit) Exercise intolerance, rhabdomyolysis Hypoglycaemia, hepatomegaly Liver and muscle involvement Liver and testis X PGAM2 7p13 Phosphoglycerate mutase-​2 Muscle isozyme XI LDHA 11p15.1 Lactate dehydrogenase A (M subunit) Muscle isozyme, desquamation XII ALDOA 16p11.2 Aldolase A (ubiquitous and muscle) Associated with haemolysis XIII ENO3 17p13.2 Enolase-​β Extremely rare—​muscle XIV PGM1 1p31.3 Phosphoglucomutase 1 Also disordered protein glycosylation (CDG1T) XV GLYG1 GLYG1 3q24 3q24 Glycogenin (absent glycogen) Glycogenin (abnormal glycogen -​ polyglucosan) Reduced or absent glycogen Polyglucosan body myopathy-​2 (proximal) Cardiac and skeletal myopathy

12.3.1  Glycogen storage diseases 1989 myopathy and it can still be useful to distinguish between hepatic and muscle GSDs. The overall incidence of GSDs has been estimated at 1 in 20 000 with the commonest being GSD IX, followed by GSD I, II, and III. The clinical features of the commoner disorders are described in the following sections. Glycogen storage disease type I  (von Gierke’s disease) Biochemistry GSD I is due to glucose-​6-​phosphatase deficiency. GSD Ia is caused by defects in subunits of the endoplasmic reticular enzyme com- plex that enable production of glucose from glucose 6-​phosphate. In GSD Ib, the endoplasmic reticular transmembrane protein glucose-​6-​phosphate translocase is deficient. In both forms, the production of glucose from both glycogenolytic and gluconeogenic pathways is blocked, resulting in profound fasting hypoglycaemia. Accompanying this, there is a build-​up of glucose 6-​phosphate. This is then metabolized by the pentose phosphate shunt, or transferred back into glycogen which is stored in the liver and, to a lesser extent, the kidney. The products of glucose-​6-​phosphate metabolism have an important role to play in the other metabolic consequences of GSD I: hyperlactataemia, hyperuricaemia, and hypertriglyceridaemia. The hypoglycaemia is somewhat mitigated by the fact that small quantities of free glucose can be liberated by the α-​1,6-​glucosidase activity of the secondary action of debranching enzyme. Residual production of glucose probably also occurs by lysosomal hydrolysis of glycogen. Lactic acidaemia results from stimulation of glycolysis at the level of phosphofructokinase by high concentrations of glucose 6-​ phosphate (and hence fructose 6-​phosphate); lactate cannot be re- cycled in the liver to form new glucose and lactic acidosis results. Failure to dephosphorylate glucose 6-​phosphate stimulates substrate cycling and increases the activity of the pentose phos- phate pathway, with enhanced production of the reduced form of nicotinamide-​adenine dinucleotide phosphate (NADPH), ribose 5-​phosphate, and purines. Degradation of purine nucleotides by AMP-​deaminase and the coordinated action of xanthine oxidase on inosine phosphate and hypoxanthine leads to overproduction of uric acid in the liver. The deaminase is activated when the concentra- tion of free phosphate falls as a result of sequestration in sugar phos- phate esters. Lactate competes with urate for excretory pathways in the kidney and this also contributes to the hyperuricaemia. Enhanced flux through glycolysis and underutilization of gluconeogenic precursors leads to increased production of the re- duced form of nicotinamide-​adenine dinucleotide (NADH) and NADPH, glycerol, and acetyl coenzyme A, and these in turn in- duce hypertriglyceridaemia. Malonyl coenzyme A, derived from acetyl coenzyme A, inhibits the carnitine acyltransferase system and blocks the oxidation of fatty acids; thus marked ketosis does not usu- ally develop. Clinical presentation Patients typically present in infancy with symptomatic hypogly- caemia and failure to thrive, accompanied by a swollen abdomen due to hepatomegaly. Hypoglycaemic encephalopathy is often ac- companied by seizures and can be fatal:  recurrent episodes lead to permanent neurodisability. Children have impaired growth and increased subcutaneous fat deposition leading to a ‘doll’s face’ appearance. With aggressive dietary management (see following subsection on ‘Management’), the immediate life-​threatening complications can be avoided. With improved survival, the chronic, multisystem complications of GSD I have emerged (Fig. 12.3.1.3). There is persistent hepatomegaly, with glycogen storage accom- panied by gross infiltration with fat. Cirrhosis and portal hyper- tension are, however, rare. Short stature, often combined with obesity, is common. The kidneys are enlarged by glycogen depos- ition. Progressive focal glomerulosclerosis and proximal tubular injury with a secondary Fanconi’s syndrome may also occur. Short periods of fasting, or other metabolic stressors such as infection, provoke hypoglycaemia and lactic acidosis. In the longer term, poor metabolic control causes growth arrest; hyperuricaemia and gout; marked hypertriglyceridaemia (which can lead to acute pancreatitis) and hypercholesterolaemia with raised very low-​density lipoprotein and normal low-​density lipoprotein cholesterol concentrations in the plasma; and prolonged bleeding time related to an acquired von Willebrand-​like defect affecting the platelet. Hepatic adenomas are seen in adults. These can regress with improved metabolic control, but there is a risk of transformation to hepatocellular carcinoma, and all patients need to be carefully monitored with regular liver MRI. Patients with defects of the glucose-​6-​phosphate translocase system (type 1B) also have a neutropenia with impaired neutrophil migration and chemotaxis and are prone to recurrent bacterial in- fections. These patients are also at risk of developing granulomatous colitis with clinical features similar to ulcerative colitis. Partial deficiencies of the glucose-​6-​phosphatase system lead to variable clinical expression: in Japan, a milder form of GSD Ia oc- curs due to the common G727T mutation that is prevalent in that country. GSD I should be considered in patients presenting with glucagon-​unresponsive hypoglycaemia with or without liver en- largement in adult life. Management Historically, GSD I (and the other GSDs presenting with hypogly- caemia in infancy) were associated with a very poor outcome. This has been transformed by the introduction of aggressive dietary man- agement aimed at maintaining a constant exogenous supply of glu- cose to meet basal requirements. Regular oral carbohydrate during the day and continuous overnight pump feeding with glucose, de- livered either by nasogastric or gastrostomy tube, clearly improves clinical and biochemical parameters. Subsequently, fasting tolerance has been improved with the use of uncooked cornstarch (obtained from the supermarket and suspended in water): this acts as a ‘slow-​ release’ form of glucose and, particularly in older patients, allows for more time between meals during the day and for some patients to discontinue overnight feeds. Modified cornstarches have now been produced with the aim of increasing fasting tolerance further, al- though it is not yet clear that they offer a significant benefit over shop-​bought cornflour. Maintaining normoglycaemia requires a diet with about 65% of dietary energy as carbohydrate. Continuous glucose monitoring can be useful in adjusting doses of uncooked cornstarch and concentrations of overnight feeds. Regular dietetic

SECTION 12  Metabolic disorders 1990 review is important to minimize excessive weight gain and insulin resistance, and ensure the diet is nutritionally complete. Intercurrent illness can rapidly provoke hypoglycaemia and pa- tients with GSD I are given an ‘emergency regimen’ to use in times of metabolic stress. This consists of frequent oral glucose polymer. If for any reason patients can’t tolerate oral intake, 10% dextrose should be given intravenously at a rate of 2 ml/kg per hour. Hyperlipidaemia and hyperuricaemia need to be treated. Hyperfiltration or albuminuria indicates renal involvement and angiotensin-​converting enzyme inhibitors or angiotensin receptor blockers should be introduced. Hypocitraturia may contribute to the increased incidence of nephrolithiasis and citrate supplementation may be useful. Iron supplementation is often needed. Osteopenia is common and calcium and vitamin D supplementation should be considered. Surveillance for hepatic adenomas is important. MRI with the use of intravenous contrast is the preferred method. About 70 to 80% of adult patients have been reported to have at least one lesion, and these progress in size or number in 50% of cases. The occurrence of adenomas seems to be related to metabolic control and in some cases improving biochemical parameters can lead to adenoma re- gression. Spontaneous regression is also seen (Fig. 12.3.1.4). The occurrence of hepatic adenomas is concerning because they can progress to hepatocellular carcinoma:  predicting this pro- gression is difficult. Blood markers such as α-​fetoprotein have not proved useful. A  rapid increase in size or number of adenomas, changes in vascularization, and bleeding should lead to a multidis- ciplinary team review to discuss surgical intervention, including liver transplantation. Human granulocyte colony-​stimulating factor is often required in patients with GSD Ib to increase the neutrophil count and control mouth ulcers, recurrent infections, and inflammatory bowel disease. Long-​term use of granulocyte colony-​stimulating factor is associ- ated with a number of complications and should be supervised by a haematologist. Due to the dangers of fasting and the bleeding tendency associ- ated with GSD I, surgery must be managed carefully. Patients should be admitted the day before so that fasting can be covered with intra- venous glucose. Platelets should be available in case of postoperative haemorrhage. Epistaxis Acute pancreatitis Insulin resistance Splenomegaly (GSD Ib) Hypersplenism Neurology dependent on duration and severity of hypoglycaemia and acidosis Acute—coma Low IQ Seizures White matter parencyhmal loss Gross hepatomegaly in infancy —improves with metabolic control Hepatic steatosis Hepatic adenoma Hepatocellular carcinoma (rare) Pulmonary hypertension (rare) Diarrhoea Inflammatory bowel disease (GSD Ib) Chronic anaemia Platelet dysfunction Neutropenia Neutrophil dysfunction Mouth ulcers (GSD Ib) (GSD Ib) Gout Osteopenia Other Growth failure / delay Myopathy Polycystic ovaries Delayed puberty Nephromegaly Hyperfiltration in childhood Hypofiltration in adult life Tubulointerstitial disease Proteinuria Renal failure Renal calculi Fig. 12.3.1.3  Complications of GSD I.

12.3.1  Glycogen storage diseases 1991 Pregnancy in women with GSD I is now relatively routine. With careful planning, close attention to glycaemic control, and increased carbohydrate requirements, especially in the second half of preg- nancy, and a well-​managed labour, outcomes are good. With optimal medical management, patients with GSD I now lead relatively normal lives, but for some patients, good metabolic con- trol is never obtained. For these patients, liver transplantation offers a long-​term ‘cure’ for many features of the disease. Where there is also end-​stage renal failure, combined liver and renal transplant- ation can be performed. Glycogen storage disease type II
(Pompe’s disease) GSD II causes hypertrophic cardiomyopathy in infants and a pro- gressive skeletal myopathy in older patients. It is primarily classified as a lysosomal storage disorder and is discussed in Chapter 12.8. Glycogen storage disease type III
(Forbes–​Cori disease) Biochemistry GSD III is due to deficiency of debrancher enzyme. This results in the storage of structurally abnormal glycogen, with short outer chains, called limit dextrin, in both liver and muscle. Although glycogenolysis is blocked, gluconeogenesis is unaffected and fasting hypoglycaemia is milder than that seen in GSD I  and accompanied by ketosis rather than lactic acidosis. The sec- ondary metabolic consequences are mostly confined to a mild hyperlipidaemia. Clinical presentation GSD III affects both liver and muscle. Hypoglycaemia and the hep- atic consequences of storage dominate the clinical picture in chil- dren, with fasting hypoglycaemia and poor growth. The condition is less severe than GSD I and even in historic cohorts, most patients survived to adulthood. In adults, fasting tolerance improves and on the whole hypoglycaemia can be prevented with dietary management. Hepatic adenomas have only rarely been reported, although pa- tients can occasionally develop cirrhosis, and the kidneys are not affected. Patients do, however, develop muscle symptoms and complain of exercise intolerance, although rhabdomyolysis is not a recognized feature. Some patients develop a progressive, disabling myopathy with pronounced distal weakness and myopathic facies. Cardiac muscle is also involved and hypertrophic cardiomyopathy can result in arrhythmias or heart failure (Fig. 12.3.1.5). Management The management of hypoglycaemia in childhood is as in GSD I. In adult patients, it is important not to overtreat: with home glucose monitoring it is often possible to reduce the dietary content of com- plex carbohydrate. It has been suggested that the skeletal myopathy and cardiomy- opathy seen in GSD III is not solely due to glycogen storage and that energy deficit may also have a role to play. In theory, this might be addressed by providing alternative sources of energy. Ketone bodies can be provided directly as d,l-​3-​OH butyrate or by use of a ketogenic diet. A high-​protein diet should enhance gluconeogenesis. To date, there have been isolated case reports of improvements in cardiomyopathy and skeletal myopathy but no systematic studies of these approaches have been done. Although left ventricular hypertrophy occurs in many patients, its clinical significance is not clear. To date, there are very few case reports of heart failure or significant arrhythmia in adults. This may change as patients age and periodic echocardiography and ECG monitoring is probably prudent. The incidence of clinically significant hepatic fibrosis and cir- rhosis may also increase with age and liver imaging can be used to monitor this as well as the occurrence of hepatic adenomas. Fig. 12.3.1.4  Hepatic adenoma (white arrow) in left lobe of liver of a young woman with GSD Ia. Fig. 12.3.1.5  A 42-​year-​old woman with GSD III and myopathy. She has myopathic facies and a scoliosis.

SECTION 12  Metabolic disorders 1992 Polyglucosan body disease (glycogen storage diseases types IV, VII, XV, and 0) Biochemistry Polyglucosan body disease (PBD) is characterized by the storage of aggregates of abnormal polysaccharides which are less branched than normal glycogen. Polyglucosan has a fibrillar structure and, un- like glycogen is at least partially resistant to digestion with amylase. Polyglucosan is seen in the heart and parts of the brain as a product of normal ageing, but in PBD, aggregates occur at an earlier age and in a wide variety of different tissues. PBD is not a single genetic entity: more than seven different mo- lecular causes of polyglucosan body formation have been recognized to date. Some of these are known GSDs (glycogenin deficiency (GSD XV), branching enzyme deficiency (GSD IV), glycogen synthase de- ficiency (GSD 0), and phosphofructokinase deficiency (GSD VII)) but other involved proteins do not seem to have a direct role in glycogen metabolism (i.e. RBCK1, a ubiquitin ligase which regulates the NF-​κB pathway and AMP-​activated protein kinase (AMPK)). The biogenesis of polyglucosan bodies is not fully understood, but experimental work suggests that, at least in some cases, an imbal- ance between the activities of glycogen synthase and debranching enzyme may be important. Clinical presentation Branching enzyme deficiency (GSD IV) is the best characterized of the PBDs. The classical form of GSD IV presents with progres- sive liver failure in the first years of life. If these children are given liver transplants they go on to develop myopathy. Some patients present with isolated skeletal or cardiomyopathy. This can be of early onset, in which case it can progress quickly to respiratory failure and death, but other patients present as adults with slowly progressive disease. The term adult PBD refers to a form of branching-​enzyme defi- ciency which presents between the ages of 40 and 60 with a com- bination of neurogenic bladder, spastic paraparesis, and peripheral neuropathy. Imaging shows leukodystrophy. Polyglucosan bodies are found throughout the central and peripheral nervous system. The condition is progressive and patients usually die within 20 years of diagnosis. The other causes of PBD are rarer and generally present as hyper- trophic cardiomyopathy with or without skeletal myopathy in chil- dren or adolescents. Glycogen storage disease type V
(McArdle disease) Biochemistry McArdle described a patient who suffered exercise-​induced my- algia in whom lactate fell during ischaemic exercise rather than rising, suggesting a defect in glycogenolysis. Enzymology sub- sequently showed a deficiency of muscle phosphorylase activity. Patients are asymptomatic during low intensity, aerobic exer- cise, when muscle depends on fatty acid oxidation for energy, but during anaerobic exercise patients rely on glycolysis and develop symptoms. Clinical presentation Typically patients develop painful muscle cramps soon after the start of exercise. Continued high-​intensity exercise leads to rhabdo- myolysis and acute kidney injury (which is normally fully revers- ible). However, if patients continue to exercise at lower intensities, symptoms resolve and they are able to continue. This ‘second wind’ phenomenon is useful diagnostically and is due to the switch from glycolysis to alternative energy sources in aerobic exercise. Management No drug or dietary treatment has been shown to be effective in GSD V. There is some evidence that aerobic physical training is safe, and may improve exercise tolerance. This is probably due to an increased capacity for fatty acid oxidation. Treatment of rhabdomyolysis-​induced acute kidney injury is the same as for other more common causes of rhabdomyolysis (see Chapter 21.5). Glycogen storage disease type IX Biochemistry GSD IX is due to deficiency of phosphorylase kinase. Phosphorylase kinase consists of four subunits, two of which have tissue-​specific isoforms. The commonest form of GSD IX, and the commonest GSD, is X-​linked and due to mutations in PHKA2. Clinical presentation GSD IXa is a hepatic GSD presenting early in life with hepatomegaly and fasting hypoglycaemia and ketosis. It is milder than GSD I and symptoms generally resolve in adulthood. Liver fibrosis has, how- ever, been reported as a long-​term complication. The other forms of GSD IX are much rarer and can lead to muscle as well as liver disease. Management Management of hypoglycaemia is as for GSD I, but adult patients have normal fasting tolerance and do not need uncooked cornstarch. Diagnosis of glycogen storage disease Most patients with hepatic GSDs present with hypoglycaemia in early life. Historically, definitive diagnosis relied on demonstrating glycogen storage and assaying enzyme activity in the affected tissue: many adults with GSD I still bear the scars of liver biopsies performed in infancy. This was invasive and technically difficult, and has to a large extent been superseded by new techniques. In infants with suggestive symptoms, biochemical profiling, with measurements of glucose, lactate, and ketones, can suggest the correct diagnosis. In some cases (e.g. GSD III), this can be con- firmed by enzymology using leucocytes, but in GSD I, molecular genetic analysis is required as the enzymes are only expressed in liver (Table 12.3.1.2). Some patients present with hepatomegaly without biochemical features suggesting GSD. In these cases, histological examination re- veals glycogen storage. If frozen tissue has been kept, enzymology

12.3.2 Inborn errors of fructose metabolism 1993 T

12.3.2 Inborn errors of fructose metabolism 1993 Timothy M. Cox

12.3.2  Inborn errors of fructose metabolism 1993 may then confirm the diagnosis of a GSD. Electron microscopy can also be helpful if structurally abnormal glycogen is present. A diag- nostic fast, with measurement of glucose, lactate, and ketones can also provide useful information. Hepatomegaly with glycogen storage is not, however, always due to a GSD: hepatic glycogenosis is a well-​recognized complication of poorly controlled diabetes mellitus. Muscle GSDs either present as exercise-​induced muscle pain or rhabdomyolysis, or progressive skeletal myopathy or cardiomyop- athy. For patients who have exercise intolerance, and in whom de- fects of energy metabolism are suspected, the ischaemic forearm exercise test has traditionally been the diagnostic test of choice, demonstrating a rise in ammonia but not lactate in GSD V. This test is always unpleasant for the patient, and can precipitate rhabdo- myolysis, and is now seldom performed. A 12-​min walking test, monitoring speed and heart rate, can be used to demonstrate the second wind effect in patients with GSD V. A number of the en- zymes involved can be assayed in leucocytes, but for others muscle is needed (Table 12.3.1.2). Testing for individual enzymes is laborious and expensive and, increasingly, molecular genetics is becoming the diagnostic test of choice. Next-​generation sequencing allows large arrays of genes to be sequenced at the same time, and genetic panels to screen for all causes of rhabdomyolysis, or all known GSDs, are now available. FURTHER READING Boers SJB, et al. (2014). Liver transplantation in glycogen storage dis- ease type I. Orphanet J Rare Dis, 9, 47. Brambilla A, et al. (2014). Improvement of cardiomyopathy after high-​ fat diet in two siblings with glycogen storage disease type III. JIMD Rep, 17, 91–​5. Dagli A, Sentner CP, Weinstein DA (2010). Glycogen storage ­ disease type III. In:  Pagon RA, et  al. (eds) GeneReviews®. University of Washington, Seattle. http://​www.ncbi.nlm.nih.gov/​ books/​NBK26372/​ Derks TGJ, Smit GPA (2015). Dietary management in glycogen storage disease type III: what is the evidence? J Inherit Metab Dis, 38, 545–​50. Hedberg-​Oldfors C, Oldfors A (2015). Polyglucosan storage myop- athies. Mol Aspects Med, 46, 85–​100. Julián MT, et  al. (2015). Hepatic glycogenosis:  an underdiagnosed complication of diabetes mellitus? World J Diabetes, 6, 321–​5. Kishnani PS, et al. (2010). Glycogen storage disease type III diagnosis and management guidelines. Genet Med, 12, 446–​63. Kishnani PS, et al. (2014). Diagnosis and management of glycogen storage disease type I: a practice guideline of the American College of Medical Genetics and Genomics. Genet Med, 16, e1. Lomako J, Lomako WM, Whelan WJ (2004). Glycogenin: the primer for mammalian and yeast glycogen synthesis. Biochim Biophys Acta, 1673, 45–​55. Mayorandan S, et al. (2014). Glycogen storage disease type III: modi- fied Atkins diet improves myopathy. Orphanet J Rare Dis, 9, 196. Mochel F, et al. (2012). Adult polyglucosan body disease: natural his- tory and key magnetic resonance imaging findings. Ann Neurol, 72, 433–​41. Olpin SE, et al. (2015). The investigation and management of meta- bolic myopathies. J Clin Pathol, 68, 410–​17. Preisler N, et al. (2015). Skeletal muscle metabolism is impaired during exercise in glycogen storage disease type III. Neurology, 84, 1767–​71. Quinlivan R, et al. (2010). McArdle disease: a clinical review. J Neurol Neurosurg Psychiatry, 81, 1182–​8. Quinlivan R, Martinuzzi A, Schoser B (2014). Pharmacological and nutritional treatment for McArdle disease (glycogen storage disease type V). Cochrane Database Syst Rev, 11, CD003458. Quinlivan R, et  al. (2011). Physical training for McArdle disease. Cochrane Database Syst Rev, 12, CD007931. Raben N, et al. (2012). Autophagy and mitochondria in Pompe dis- ease: nothing is so new as what has long been forgotten. Am J Med Genet C Semin Med Genet, 160C, 13–​21. Valayannopoulos V, et  al. (2011). Successful treatment of severe cardiomyopathy in glycogen storage disease type III with D,L-​3-​ hydroxybutyrate, ketogenic and high-​protein diet. Pediatr Res, 70, 638–​41. 12.3.2  Inborn errors of fructose metabolism Timothy M. Cox ESSENTIALS Most people in developed countries ingest 50 to 100 g fructose equivalents daily in their diet, arising from fructose itself, sucrose, and sorbitol. After rapid carrier-​mediated absorption across the in- testinal epithelium, fructose is metabolized (mainly in the liver) by the enzymes ketohexokinase (fructokinase), aldolase B, and triokinase, eventually being converted into glucose or glycogen. Dietary sugars—​burgeoning constituents in food and drinks worldwide—​ have undesirable effects on those with limited capacity to metabolize fructose, including severe illness or death in young patients. ‘Fructose malabsorption’ describes incomplete absorption of fructose that is associated with abdominal symptoms and diarrhoea reminiscent of intestinal disaccharidase deficiency. Symptoms occur after ingestion of fructose-​ or sorbitol-​rich foods and drinks such as apple juice, but as yet a convincing genetic cause for this condition Table 12.3.1.2  GSDs where enzyme activity can be assayed and tissues needed for assay Disease Enzyme Tissue GSD I Glucose 6-​phosphatase Liver GSD II α-​1,4-​glucosidase Whole blood or dried blood spot GSD III Glycogen debrancher Whole blood, muscle or liver GSD IV Glycogen brancher Whole blood, muscle or liver GSD V Phosphorylase Muscle GSD VI Phosphorylase Whole blood or liver GSD VII Phosphofructokinase Muscle GSD IX Phosphorylase b kinase Whole blood or liver GSD IX Fructose 1,6-​bisphosphatase Whole blood or liver

SECTION 12  Metabolic disorders 1994 has not been found. Symptoms improve when the offending sugars are avoided. Three inborn errors of fructose metabolism are recognized and these disorders are vivid examples of gene–​environment interactions:

  1. Essential or benign fructosuria due to fructokinase deficiency—​a very rare disorder with apparently no ill effects.
  2. Hereditary fructose intolerance (fructosaemia)—​an autosomal recessive disease caused by deficiency of aldolase B.  Typically presents at weaning when most fatalities occur. May come to light at any age with postprandial abdominal pain and vomiting, symptomatic hypoglycaemia (which may induce seizures), hypophosphataemia, acidosis, and other metabolic disturbances after consumption of offending foods and drinks. Unrecognized disease causes failure to thrive/​growth retardation, a Fanconi-​like renal syndrome with nephrocalcinosis, and jaundice with lethal liver injury. Parenteral infusion of fructose or its congeners may cause death from acute hepatorenal injury. Diagnosis formerly depended on a controlled intravenous fructose challenge test or demonstration of deficient aldolase B isozyme activity in liver or small intestinal biopsy material; currently molecular analysis of the aldolase B gene is preferred and usually decisive. Treatment requires institution of a strict sugar-​exclusion diet supplemented by folic acid and vitamin C. Early diagnosis and dietary modifica- tion are critical for well-​being and normal development.
  3. Fructose-​1,6-​diphosphatase deficiency—​a very rare disease of infancy and childhood associated with failure of hepatic gluconeogenesis causing bouts of severe hypoglycaemia, ke- tosis, and lactic acidosis provoked by infection and starvation. Metabolic decompensation is provoked by dietary fructose, related sugars, and/​or ketogenic fat. Diagnosis depends on en- zymatic assay of fructose 1,6-​diphosphatase in fresh liver biopsy samples or molecular analysis of the cognate gene. Treatment requires a fructose-​exclusion diet containing abundant carbohy- drate energy, restricted fat, and protein. Acute episodes of acidosis or hypoglycaemia are controlled by intravenous glucose infu- sions, with bicarbonate if required. Introduction Fructose is an important and burgeoning component of the modern diet; it occurs as a free monosaccharide in fruit, nuts, honey, and some vegetables but is now abundant and practically ubiquitous in popular, mass-​produced drinks. Free fructose is released from the disaccharide sucrose in the gut lumen by the sucrase–​isomaltase complex at the brush-​border membrane of the mucosal epithelium; in modern times, sucrose is the principal source of fructose for most persons. Finally, the sugar alcohol, sorbitol (a constituent of medi- cines and tablets, as well as diabetics), is converted quantitatively to fructose in the liver and intestine. Inborn errors of fructose metabolism provide unique ex- amples of gene–​environment interactions in recent economic his- tory. Emergence of these conditions reflects revolutionary dietary changes resulting from the mass industrialization of sugar farming and manufacture. Were it not for the sugar industry, it is unlikely that these disorders would have come to light, even though several mutations in ALDOB (encoding the specific aldolase B isozyme) which cause hereditary fructose intolerance are widespread and shared in different populations, indicating ancient origins. Given the primacy of fructose and sucrose in the phenotypic ex- pression of these disorders, an understanding of the driving forces responsible for the disease is of central importance for holistic clin- ical management. Certainly, the relatively rare conditions that are due to disturbance of fructose metabolism share features with en- vironmentally driven syndromes of obesity, nonalcoholic hepatic steatosis, and hyperlipidaemia. In the background, a succession of regrettable historical events led to the introduction of sugar pri- marily into the modern European diet. Insatiable demands for sugar obtained in plantations by foreign slave labour in New World col- onies has left many painful consequences, for example, the wide geo- graphic dissemination of sickle cell disease and other effects such as economic deprivation and social injustice. Consumption of fructose and sucrose The sugar trade expanded during the 16th to 19th centuries ce as the mass capture and trading of slaves took hold. Partially modelled on earlier Arabic trading in Africa, the use of slave labour in the sugar plantations in New World and other colonies has a haunting legacy. Nowadays, however, this history has an ironic payback in the consequential effects of sugar on global health. The mass industri- alization of sugar production declares the nature of the human ap- petite for sweet flavours, and exogenous sugar is added to food and drinks in modern societies across the world. Disorders of fructose metabolism and its assimilation mirror the ubiquity of sucrose and fructose in the diet. Consumption of free sugars varies greatly by age and country. Most people in developed countries ingest 50 to 100 g fructose equivalents daily in the diet. The top cane and beet sugar (sucrose)-​ consuming countries are shown in Table 12.3.2.1. Global production of sugar (sucrose from cane and beet) continues to rise, but over the last three decades, novel manufacture of fructose as a high-​fructose corn syrup, enzymatically derived from starch in maize (corn), has also burgeoned. This intensely sweet sugar, first enzymatically pro- duced commercially from excess maize in the early 1970s, mainly from United States mid-​Western agriculture, has been popular in the United States of America, Mexico, and China, but is now manu- factured and used in the European Union and Australia. Sugar manufacture—​a challenge for global health World manufacture of sugar and sweeteners is growing; the global 2017 market was about $97 billion, representing an estimated 184 million metric tons of sugar as sucrose. Until recently, all large-​ scale sugar production came from sugar beet and sugar cane, which in modern times yield products that are indistinguishable in taste and composition. Cane sugar, principally from tropical countries, represents about 80% of the global market and is generally cheaper to manufacture than beet sugar, which has enjoyed political and eco- nomic support from local farming systems, generally in nontropical countries. Direct market competition of beet with cane sugar had occurred in postcolonial times because beet sugar-​producing coun- tries mitigated losses on exports by high revenue in domestic mar- kets, thus allowing expansion of domestic beet cultivation.

12.3.2  Inborn errors of fructose metabolism 1995 High-​fructose corn syrup is now used extensively as a sweetener in drinks and processed foods. In 2017, the global market for this preparation alone was $4.5 billion dollars (c.5% of cane and beet sugar) and this continues to rise by more than 2% annually. In 1970, the average per capita daily consumption in the United States of America of 105 g calorific sweetener contained only 0.5 g from high-​ fructose corn syrup, but by 2004 this had risen to 52.4 g of the 124.8 g of the sweetener ingested on average each day. As the World Health Organization emphasizes—​and as patients who are genuinely intolerant realize—​many of the sugars consumed today are ‘hidden’ in processed foods and drinks; they are neither overtly present nor declared as sweeteners. An average portion of ketchup sauce contains about 4 g of free sugar; a single can of soda contains up to 40 g sucrose and/​or fructose. Depending on the re- gional manufacturing and local distribution, the most popular international carbonated drinks obtained from one long-​established company based in the United States of America contain 100 to 140 g/​litre sugar, mostly as fructose. Food and drink labels are often deceptive:  a popular mint-​ flavoured confection in the United Kingdom is described as con- taining several grams of carbohydrate but ‘sugar-​free’. In fact, all the carbohydrate is present as sorbitol—​a direct source of fructose. These mints, peppery rather than sweet to the taste, caused seizures and nephrocalcinosis in a strictly controlled and food-​conscious mathematician with hereditary fructose intolerance (whose affected brother had died in infancy). While a popular carbonated mixer drink flavoured with quinine contains 80 g/​litre sugar, its ‘naturally light’ alternative contains nearly 30 g/​litre. In Europe, sugar intake in adults ranges from about 8% of total energy intake in Hungary and Norway, to more than twice this pro- portion in Spain and the United Kingdom. Intake is relatively greater among children, ranging from about 12% in Denmark, Slovenia, and Sweden, to nearly 25% in Portugal. Generally, consumption of additional sugar is greater in urban compared with rural commu- nities, but this may not hold in countries where production of cane sugar is still a dominant industry, such as Brazil. Changes in the sugar economy and mitigating factors Production of sugar from cane, beet, and increasingly from cereal starch is a massive global industry, but more than 70% of world sugar production is never traded on the open market. Brazil, one of the first since its early Portuguese colonization, and still the largest pro- ducer, controls half the global market but pays subsidies measured in billions annually to its sugar industry. Until very recently, a complex tariff rate system providing direct support of domestic sugar pro- duction was used in the United States of America, which maintained the price there up to 90% higher than the world market price at an annual charge of 3.7 billion dollars to American consumers. As of August 2018, a ‘zero-​for-​zero’ policy is poised for legislation to end domestic sugar subsidies after other major international producers such as Brazil and India agree reciprocally to cease sugar industry subsidies. Strong subsidies also operated in the European Union, but with a ruling from the World Health Organization some quotas were abolished in 2015. Perhaps reflecting former colonial respon- sibilities, in 2009 the European Union granted Least Developed Countries zero-​tariff access status to the European market as part of its ‘Everything but Arms’ initiative. Control of sugar consumption The World Health Organization and American Heart Association recommend that women consume less than six teaspoons of sugar per day, which amounts to 25 g; men are recommended to ingest no more than nine teaspoons of sugar per day (38 g). The American Academy of Pediatrics recommends that children between 2 and 18 years take in no more than six teaspoons (25 g) per day. It is note- worthy that none of these recommendations apply to sugar that oc- curs naturally in foods, such as the fructose present in fruits, nuts, and honey. Given the increasing awareness of the growth of sugar consump- tion, the food and beverage industry is replacing sugar or corn syrup with non-​nutritive sweeteners in a range of products that tradition- ally contained sugar. Aspartame has been a popular artificial sweet- ener in the US food industry, and its price dropped after expiry of the relevant patent in 1992. However, since 2008, sucralose has become the most popular nonsugar sweetener, replacing aspartame to artifi- cially sweeten foods and beverages. Hidden practices also operate in the United States of America, where it has until recently been possible to mislabel foods and greatly minimize their apparent sugar content. From 2018, in a con- tested mandate, the Food and Drug Administration required all food manufacturers to identify and explicitly list all ‘added sugars’ in their Nutrition Facts labels. This new food labelling regime clearly regards sugar as a major public enemy. As in many other countries, previ- ously these sugars were concealed under the ‘Total Carbohydrates’ section of the label, and only naturally occurring sugars were em- phasized. Given the strong influence of food labels on the public, the new regulation is likely to have a salutary effect, and particularly for patients with disorders of fructose metabolism. Some consider that sugar sales and promotion will soon be on the wane, but while this seems to be a premature judgement, sugar is starting to rival tobacco in perception as a major public ill in some quarters. Biochemistry of fructose metabolism The pathways of fructose metabolism are summarized in Fig. 12.3.2.1. Phosphorylated forms of fructose are critical intermediates in the glycolytic and gluconeogenic metabolic pathways in all cells. Table 12.3.2.1  Daily sugar consumption (grams per capita) 1 United States 126.4 2 Germany 102.9 3 Netherlands 102.5 4 Ireland 96.7 5 Australia 95.6 6 Belgium 95.0 7 United Kingdom 93.2 8 Mexico 92.5 9 Finland 91.5 10 Canada 89.1 Data from the World Agricultural Outlook Board—​United States Department of Agriculture USDA https://​public.govdelivery.com/​accounts/​USDAFAS/​subscriber/​new and Statista: https://​www.statista.com/​statistics/​496002/​sugar-​consumption-​worldwide/​ (accessed 26 August 2018).

SECTION 12  Metabolic disorders 1996 Fructose is absorbed by a carrier mechanism that facilitates trans- port across the intestinal epithelium; this process is mediated by the glucose transporter isoforms GLUT5 and GLUT2, the latter prob- ably contributing to efflux across the basolateral membrane of the enterocyte. Uptake of fructose is rapid and notable for its lack of dependence on insulin. Fructose is then conveyed via the portal bloodstream to the liver, where it is assimilated. The jejunal mucosa and proximal tu- bule of the kidney are subsidiary sites of fructose metabolism. Assimilation of fructose depends on the concerted activities of the enzymes ketohexokinase (fructokinase), aldolase B, and triokinase, which are expressed specifically in these tissues. Uptake of fructose occurs independently of insulin and its incorporation into inter- mediary metabolism bypasses the regulation of glycolysis at the level of phosphofructokinase-​1. For these reasons, solutions of fruc- tose or sorbitol were advocated and, in the past, extensively used for parenteral nutrition, particularly in German-​speaking countries. However, the occurrence of lactic acidosis, hyperuricaemia, and other serious consequences has led to their withdrawal from hyper- alimentation regimens in most, if not all, regions. Fructokinase rapidly phosphorylates fructose at the 1-​carbon position. This enzyme has a high affinity for its substrates and the intestinal mucosa and liver rapidly convert fructose to fructose 1-​phosphate; in other tissues, the capacity of hexokinase to phos- phorylate fructose at the 6-​carbon position is limited. Similarly, the fate of fructose 1-​phosphate in the fructose-​metabolizing tissues is dependent on a specific isozyme of aldolase, aldolase B. This has greater activity towards fructose 1-​phosphate than does its ubi- quitous counterpart aldolase A, the natural substrate of which is fructose 1,6-​diphosphate. Cleavage of fructose 1-​phosphate gen- erates glyceraldehyde and dihydroxyacetone phosphate. These tri- oses enter the intermediary pools of carbohydrate metabolism, and, as a result of triokinase activity, glyceraldehyde is phosphorylated so that the two triose phosphates may be condensed by aldolase A to form the glycolytic and gluconeogenic intermediate fructose 1,6-​diphosphate. Fructose malabsorption The occurrence of abdominal symptoms and diarrhoea, reminiscent of intestinal disaccharidase deficiency, in response to ingested fruc- tose is well recognized by gastroenterologists and often attributed to incomplete absorption of fructose: it is therefore called ‘fructose malabsorption’. The symptoms occur in adults and children after in- gestion of fructose-​rich or sorbitol-​rich foods and drinks such as apple juice, and usually recede when the sugars are excluded from the diet. Many such individuals, as well as a high proportion of healthy control subjects, have findings suggestive of fructose malab- sorption based on hydrogen breath tests, but definitive evidence of true malabsorption is usually lacking. The molecular basis of this syndrome and of the wide variation of tolerance to dietary fructose and its congeners is not known. Moreover, in several patients complaining of fructose-​related intestinal symptoms, molecular analysis of the human GLUT5 gene, which encodes a major intestinal fructose transporter, has so far failed to identify causal mutations. Recently, mice lacking the action of a critical glucose-​activated transcription factor—​ carbohydrate-​responsive element-​binding protein (ChREBP)—​ that regulates glucose and lipid metabolism develop diarrhoea and weight loss after administration of diets high in sucrose or fructose, but not high-​glucose diets. These effects are associated with poor induction of the fructose carrier, GLUT5, in the small intestine. Given these findings, it is conceivable that genetic vari- ation in the activity of this nutritional regulatory pathway in the intestine accounts for dietary idiosyncrasy to fructose and sucrose in humans. While no information about ChREBP in patients com- plaining of fructose malabsorption is yet available, the field is one of great interest for large-​scale investigations that extend to obesity and the metabolic syndrome. Other studies have suggested that the distal small intestine and colon of patients who experience abdominal flatulence and diar- rhoea after ingesting fructose-​containing foods contain a bacterial population with enhanced uptake and anaerobic metabolism of fructose. No conclusive evidence has yet been provided to support these observations and more investigative studies are needed in those patients who experience symptoms attributed to malabsorp- tion of this sugar, including measurement of intestinal fructose ab- sorption, metabolism, and transport. Essential (benign) fructosuria (OMIM 229800) This is a rare disorder (estimated birth frequency 1 in 130 000) of little clinical consequence. The abnormality is transmitted as an autosomal recessive condition and is demonstrated by the pres- ence of a reducing sugar in the blood and urine, especially after meals rich in fructose. The abnormality is caused by the defi- ciency of fructokinase activity in the liver and intestine, signifi- cantly reducing the capacity to assimilate this sugar. Mutations in Sucrose Fructokinase Fructose 1-phosphate ALDOLASE B Glyceraldehyde FRUCTOSE triokinase triose phosphate isomerase dihydroxyacetone phosphate Krebs’ cycle pyruvate glyceraldehyde 3-phosphate ALDOLASE A Fructose 1,6-diphosphate Fructose diphosphatase Fructose 6-phosphate Glucose 6-phosphate Glucose 1-phosphate GLYCOGEN GLUCOSE Sorbitol Fig. 12.3.2.1  Fructose metabolism. Gluconeogenesis from triose phosphates, lactate, glycerol, amino acids, and Krebs cycle intermediates such as oxaloacetate, requires reversal of the committed reactions of glycolysis. It is the enzyme fructose 1,6-​diphosphatase that releases the glucose precursor fructose 6-​phosphate from fructose 1,6-​diphosphate. Thus, when the remaining reactions of glycolysis are reversed, exogenous fructose provides a source of glucose or glycogen. Fructose 1,6-​ diphosphatase is active in the liver, kidney, and intestine, and is a key enzyme of gluconeogenesis.

12.3.2  Inborn errors of fructose metabolism 1997 the human ketohexokinase gene on chromosome 2p23.3–​p23.2 have been identified in patients with essential fructosuria, thus confirming the suspected molecular defect in this condition. Fructose metabolism occurs slowly in essential fructosuria as a re- sult of conversion to fructose 6-​phosphate by hexokinase in adi- pose tissue and muscle, but, while plasma concentrations remain high postprandially, large amounts of fructose appear in the urine. Essential fructosuria may be confused with diabetes mellitus if the nature of the mellituria is not defined; with the use of glucose oxi- dase strips in preference to the older chemical methods for urin- alysis, such confusion is now unlikely in the routine clinical testing of urine worldwide. No treatment beyond recognition and explan- ation appears to be necessary. Hereditary fructose intolerance (fructosaemia) (OMIM 229600) This disorder, first recognized in 1956, is the most common in- herited defect of fructose metabolism with an estimated frequency of about 1 in 20 000 births in Europe. Determination of aldolase B mutation frequency in DNA obtained from neonatal blood spots in- dicated a frequency of 1 in 18 000 live births in the United Kingdom. The disease has been reported in populations throughout the world, including China and Israel. Hereditary fructose intolerance is transmitted as an autosomal re- cessive trait and, although it manifests itself first in early infancy, the effects of clinical disease may not be recognized until late childhood or even adult life. Provided the diagnosis is made before visceral damage occurs, hereditary fructose intolerance responds completely to a strict exclusion diet and patients can survive to old age. The first patient ever reported (as a young adult) was healthy and an active grandparent at well over 80 years of age. Clinical features The cardinal features of this illness are vomiting, diarrhoea, upper abdominal pain, and hypoglycaemia that are induced by the con- sumption of foods, drinks, or medicines containing fructose, or the related sugars, sucrose or sorbitol. The infant is first exposed to the offending sugars at weaning or on first transfer from breast milk to artificial feeds, and—​with continued exposure to the harmful foods—​a generalized metabolic disturbance with lactic acidosis, hyperuricaemia, and hypophosphataemia develops. Hypoglycaemia causes trembling, irritability, and cognitive im- pairment. Attacks are associated with pallor, sweating, and, when severe, result in loss of consciousness, sometimes accompanied by generalized seizures. These episodes typically occur within 30 min of meals that contain large quantities of fructose or sucrose. Continued ingestion of noxious sugars is associated with renal tubular disease, liver injury with jaundice, and impaired blood co- agulation. In children and infants, there is marked failure to thrive; growth retardation becomes apparent and the child becomes listless and miserable. Persistent exposure to fructose and the related in- jurious sugars in feeds and drinks given to infants leads to structural liver injury with cirrhosis, aminoaciduria, coagulopathy, and coma leading to death. Survival is dependent on recognition of the effects of fruit and sugar by the mother or, especially in older infants, by vomiting or forcible rejection of food by the patient. Infants who survive the stormy period of weaning develop a strong aversion to sweet-​tasting foods, vegetables, and fruits. This usually affords protection against the worst effects of fructose and sucrose, but abdominal symptoms with bouts of tremulousness, irritability, and altered consciousness due to hypoglycaemia usually continue. It has become clear that many cases escape diagnosis in infancy and childhood, but the risk of illness related to dietary indiscretion remains throughout life. Characteristically, children and adults with hereditary fructose intolerance show a striking reduction in, or ab- sence of, dental caries. They usually have notable preferences for foods and drinks, with striking dietary peculiarities. As explained in relation to the growth of the sugar industry, nutritional behaviours are changing worldwide, and increasingly it is found that hidden or undeclared sugars are the culprit in patients with fructose intoler- ance who remain unwell despite attempts to modify food and drink intake. A syndrome of chronic sugar intoxication has been identi- fied in older children and adolescents with hereditary fructose in- tolerance and may persist in the adult. General lack of vigour and developmental retardation are prominent features. Hypoglycaemia, though obvious after heavy fructose loading, may be insignificant after chronic low-​level exposure in older children. Similarly, tests of hepatic and renal function may be only mildly abnormal. The intermittent presence of reducing sugar in the urine may indicate fructosuria; amino aciduria may also be present. Persistent ingestion of fructose and sucrose is toxic to the kidney and liver, so that renal tubular acidosis (occasionally with calculi) as well as hepatosplenomegaly occur in younger patients. Severe growth retardation may be accompanied by rachitic bone disease that complicates the Fanconi-​like syndrome of proximal renal tubular disturbance with bicarbonate wasting. Growth retardation responds to dietary treatment and is usually accompanied by regres- sion of the other disease manifestations. Metabolic defect Hereditary fructose intolerance is caused by a deficiency of aldolase B in the liver, small intestine, and proximal renal tubule. These tis- sues experience injury as a result of persistent exposure to fructose in patients affected by the disorder. In the absence of the fructose-​ 1-​phosphate-​splitting activity of aldolase B, the intracellular pool of inorganic phosphate is depleted. Studies in vivo by 31P magnetic res- onance spectroscopy show that 80% of hepatic free phosphate is se- questrated as sugar phosphates after the infusion of small quantities of fructose (250 mg/​kg body weight). The secondary metabolic dis- turbances are initiated by the accumulation of fructose 1-​phosphate in a milieu where free inorganic phosphate is reduced:  there is competitive inhibition of aldolase A and inhibition of phosphor- ylase activity so that glycogenolysis and gluconeogenesis are im- paired. Thus, challenge with fructose leads to hypophosphataemia and hypoglycaemia that is refractory to glucagon or the infusion of gluconeogenic metabolites such as glycerol or dihydroxyacetone. During challenge with fructose, high concentrations of fructose 1-​phosphate cause feedback inhibition of fructokinase, thereby limiting the incorporation of fructose in the liver. As a result, fructosaemia occurs and, when the blood concentration exceeds about 2 mmol/​litre, fructosuria is apparent. Although the assimila- tion of fructose by the specialized pathway is blocked, only a small fraction of the fructose load is recovered in the urine. Studies show that 80 to 90% of the fructose is taken up under these circumstances

SECTION 12  Metabolic disorders 1998 by adipose tissue and muscle, where it can serve as an alternative substrate for hexokinase with conversion to fructose 6-​phosphate. Electrolytic disturbances occur during challenge with fructose. Hypokalaemia results from acute renal impairment with defective urinary acidification. There is a defect of proximal tubule function with bicarbonate wasting and acidosis. Occasionally, acute flaccid weakness due to hypokalaemia accompanies the other effects of fructose exposure. In patients with hereditary fructose intolerance, the administration of fructose reproducibly increases serum magne- sium concentrations. This is probably explained by the breakdown of magnesium–​ATP complexes, releasing intracellular magnesium ions as a result of nucleotide degradation by adenosine deaminase. Significant ingestion of fructose is thus also accompanied by marked hyperuricaemia in patients with hereditary fructose intolerance. In the absence of acute exposure to fructose, only minor ab- normalities of blood analytes are detectable and the blood glucose concentration is normal, even after prolonged fasting. Often trivial elevation of serum transaminase activities occur; red cell folate and white cell ascorbate concentrations may be reduced as a result of re- strictive dietary habits. Pathology and molecular genetics Persistent ingestion of fructose and related sugars in hereditary fruc- tose intolerance causes hepatic injury; there is diffuse fatty change and increased glycogen deposition. Hepatocyte necrosis with intralobular and periportal fibrosis occurs and fully developed cir- rhosis results from continued exposure to fructose. After acute ex- perimental challenge, electron microscopy has shown irregular electron-​dense material surrounded by membranous structures, sug- gesting a florid lysosomal reaction to intracellular deposits of fruc- tose 1-​phosphate. Parenteral administration of fructose or sorbitol may induce the abrupt onset of hepatorenal failure associated with bleeding. Histological examination shows hepatic necrosis in these cases (Fig. 12.3.2.2). Loss of cellular functions (e.g. in the proximal renal tubule) is probably caused by depletion of ATP resulting from the arrested metabolism of fructose by the specialized pathway. The source of the severe abdominal pain that follows ingestion of fructose is unknown, but stimulation of visceral afferent nerves by the local release of purine nucleotides or lactate may be responsible. The genetic basis of aldolase B deficiency has been studied inten- sively and numerous mutations responsible for hereditary fructose intolerance have been identified. The human aldolase B gene maps to chromosome 9q22.3. Several point mutations affecting the func- tion of the enzyme are sufficiently widespread in patients of European origin to merit focused diagnostic investigation. One particular mu- tation, originally termed Ala149→Pro, which disrupts residues in a substrate-​binding domain of aldolase B, is prevalent in populations of European descent. This mutation accounts for most alleles ­responsible for fructose intolerance, but others, including those originally named Ala174→Asp, Asn334→Lys, and a four-​base deletion in exon 4, are suf- ficiently frequent and widespread to merit initial examination in a specialized molecular diagnostic laboratory (see following ‘Molecular diagnosis’ subsection). The intragenic deletion has also been reported from China. Biochemical and structural studies of the expressed mu- tant enzymes reveal two main classes of aldolase B in hereditary fruc- tose intolerance, active tetrameric variants which are unstable and readily lose their quaternary structure and mutant aldolases that re- tain their normal tetrameric structure but are catalytically impaired. Differential diagnosis In infancy and childhood, presentation of persistent vomiting, with failure to thrive, acidosis, hypoglycaemia, and/​or jaundice suggest a wide differential diagnosis, but fructose intolerance may be indi- cated by the nutritional history and feeding difficulties. Possibilities include surgical diseases such as pyloric stenosis and even biliary atresia, but particularly inborn errors of metabolism, including galactosaemia, Reye’s syndrome, hepatitis, renal tubular disease, Wilson’s disease, mitochondrial DNA depletion syndrome, con- genital defects of glycosylation, hereditary tyrosinaemia type 1, long-​chain 3-​hydroxyacyl-​CoA dehydrogenase deficiency, classic methylmalonic aciduria, and citrullinaemia type 1. A carbohydrate-​deficient glycoprotein syndrome may be sus- pected on the basis of biochemical screening tests carried out in paediatric investigations, since untreated patients with hereditary fructose intolerance almost invariably show a type I  pattern of carbohydrate-​deficient serum transferrin on isoelectric focusing; this is corrected within a few weeks of fructose exclusion and is due to transient inhibition of phosphomannose isomerase implicated in glycoprotein processing and biosynthesis. In older infants and chil- dren, poisoning might have occurred, and this will mainly reflect inadvertent exposure to toxins, of which paracetamol in paediatric elixir preparations and suspensions is a prominent accidental risk. If fructose intolerance is considered, then sucrose, sorbitol, and fructose should be excluded completely and immediately before de- finitive diagnosis. Striking improvement, suggestive of hereditary fructose intolerance, may be seen on institution of the appropriate exclusion diet (including fluids) within a few days, and this can be life-​saving in infants and children. Diagnosis Formerly, diagnosis of fructose intolerance required the demon- stration of fructose-​1-​phosphate aldolase deficiency in tissue (liver or small intestinal mucosa), but increasingly demonstration of the presence of two causal mutant alleles of ALDOB is employed. Molecular analysis of the ALDOB gene in genomic DNA from an oral swab or blood sample can be carried out as soon as the diagnosis is suspected. Fig. 12.3.2.2  Post-mortem needle liver aspirate (Mallory’s trichrome stain). Reproduced from Ali M, Rosien U, Cox TM (1993). DNA diagnosis of fatal fructose intolerance from archival tissue. Q J Med 86, 25–30 with permission from Oxford University Press.

12.3.2  Inborn errors of fructose metabolism 1999 Molecular diagnosis Direct genetic diagnosis of hereditary fructose intolerance is now possible and is the preferred method, particularly (but not exclu- sively) for patients of European ancestry, from whom most of the widespread causal mutant alleles of ALDOB hitherto have been re- ported. The nomenclature currently adds one to the mutated residue that occurs in the aldolase B protein (A149P is currently pAla150Pro or more conveniently A150P). Widespread founder mutations have been reported from regions elsewhere, including North India. While molecular analysis of aldolase B genes for the presence of common mutations responsible for the disease can be carried out by specialized laboratories equipped for genetic testing, the rapid spread of DNA diagnostic methods worldwide is greatly improving recognition of the disease and its severity. Some specialized diag- nostic facilities, usually based in hospital laboratories offering paediatric services, have reported useful practical protocols for hier- archical ALDOB mutation screening. Failure to identify two of the more frequent mutant alleles in patients with suspected hereditary fructose intolerance should encourage a systematic approach to mo- lecular diagnosis, if necessary to include definitive sequencing of the entire human aldolase B gene (ALDOB). It is obvious that this stratagem obviates invasive or hazardous investigations using tissue biopsy procedures, or cumbersome parenteral challenge with sugar solutions, and ultimately is likely to reduce overall healthcare costs. The diagnosis has important consequences for relatives of the propositus and will provide information critical for the introduction of a rigorous and life-​long exclusion diet. Enzymatic analysis Aldolase B deficiency may be demonstrated definitively by enzym- atic analysis of biopsy samples obtained from the liver or small intes- tinal mucosa. Biochemical assay of fructaldolases characteristically demonstrates markedly reduced or absent fructose 1-​phosphate cleavage activity with a partial deficiency of fructose 1,6-​diphosphate aldolase. Since fructaldolase deficiency may accompany other par- enchymal disease of the liver, and because liver biopsy for biochem- ical analysis is invasive, these assays are of limited value in the acutely ill or jaundiced patient. Intravenous fructose tolerance test (see Fig.12.3.2.3) The intravenous fructose tolerance test was previously useful for diagnosis, particularly in adults; however, preparations of fructose suitable for intravenous use are now difficult to obtain and direct diagnosis by molecular analysis of ALDOB is preferred. In any event, failure to obtain fructose solutions suitable for parenteral use should not encourage the administration of fructose or sucrose orally, since administration by this route may induce catastrophic effects with severe pain, acidosis, and even shock. A child or adult is unlikely to return for further care after such a casual toxic exposure to a highly offensive sugar load given orally, usually against their will. Even in recent times, critical illness polyneuropathy has occurred in at least one affected child after misconceived diagnostic oral challenge with fructose (hereditary fructose intolerance was formally diagnosed by retrospective molecular analysis of the aldolase B gene). If no other method for investigating the patient is available, then the intravenous tolerance test should be carried out under controlled conditions with medical personnel at hand. It requires the infusion of 0.25 g/​kg (0.2 g/​kg in infants) of d(+)-​fructose as a 20% solution over a few minutes; blood samples for potassium ions, magnesium ions, phosphate ions, and glucose are taken before the administration and at regular intervals over a 2-​h period. In fructose intolerance, epigastric and loin pain usually accompany the infusion, and hypo- glycaemic coma may occur; hypophosphataemia is characteristic. Characteristically, since gluconeogenesis is blocked during exposure to fructose or its congeners, the acute hypoglycaemia fails to respond to glucagon and therefore glucose for parenteral injection must be available. Responses differ between individuals, and hypoglycaemia is usually milder in adults; typical responses in hereditary fructose in- tolerance and a control subject are shown in Fig. 12.3.2.3. The toler- ance test should not be carried out in patients with overt signs of liver disease where it may occasionally yield misleading results, and in a patient with hereditary fructose intolerance the challenge will dan- gerously aggravate the disease—​particularly in infants and children. The ability to identify disease alleles by analysing genomic DNA obtained from very small samples of blood or tissue may not only be beneficial for the investigation of infants with this disorder but also for neonatal testing before dietary exposure occurs. There is a strong case for trials in which the utility of mass population screening for fructose intolerance, a preventable nutritional disease, is investigated but despite much effort this apparently justifiable course has yet to be adopted. (a) 5.0 4.0 3.0 2.0 1.0 2.0 1.6 1.2 0.8 0.4 –10 Time (min) (b) Blood glucose (mmol/l) Serum phosphate (mmol/l) 4.0 3.0 2.0 1.0 1.6 1.2 0.8 0.4 90 80 70 60 50 40 30 20 10 0 Fig. 12.3.2.3  (a) Intravenous fructose tolerance tests in a 39-​year-​old woman with hereditary fructose intolerance proved by fructaldolase assay and DNA analysis. (b) An age-​matched and sex-​matched control subject with alcohol-​related episodic hypoglycaemia.

SECTION 12  Metabolic disorders 2000 Treatment Provided that organ failure and serious tissue injury do not super- vene, patients with hereditary fructose intolerance recover rapidly when the toxic sugars are withdrawn. Children who survive by ac- quiring a protective pattern of eating behaviour avoid foods which provoke abdominal symptoms. The aversion extends to most sweet-​ tasting items of food and drink as well as fruits and vegetables; it re- mains lifelong and consumption of fructose (and sucrose) is usually reduced to less than 5 g daily. It has been shown that normal growth and development can be assured in growing children and adoles- cents if less than 40 mg/​kg fructose equivalents are ingested daily. Dietary treatment of fructose intolerance mitigates the disorder but requires the almost complete exclusion of sucrose, fructose, and sorb- itol. A changing and notable feature of the disease is the increasing con- tribution of added sweeteners and additives to the diet and present in drinks. The daily consumption of sugar should be reduced to less than 40 mg fructose equivalents per kilogram of body weight (i.e. 2–​3 g for an adult) in order to reverse the disease manifestations and establish normal development in affected infants and children. The ubiquity of fructose and its congeners in the Western diet presents serious diffi- culties. Not only are fructose and its congeners present in unexpected foods, such as certain types of potato, but the sugars are added to foods unexpectedly and deceptively. Adult patients have usually restricted their consumption of fructose to less than 20 g daily and the source of the residual sugar may be difficult to establish. For this reason, the ad- vice of an experienced dietitian should be sought (Box 12.3.2.1). Particular care needs to be taken with sugar-​coated pills and espe- cially with liquid medications for paediatric use, as large amounts of fructose, sucrose, and sorbitol are frequently present. Children and adults with hereditary fructose intolerance may tol- erate the taste of confectionery that contains large quantities of nox- ious sugars but in which the sweetness is masked by other flavours, such as peppermint, which they enjoy. This behaviour may lead to unexplained hypoglycaemic symptoms and other signs of sugar toxicity. Occasionally, patients are unable to tolerate certain foods that are permitted on their diet sheets; in doubtful cases it is advis- able to avoid the offending item or to have it analysed. Patients with hereditary fructose intolerance may lack folic acid and vitamin C. Supplements of these vitamins in particular are re- commended, especially during pregnancy, but, as with other medi- cines, care has to be taken to avoid harmful sugars contained in the preparation. Although the use of fructose-​containing or sorbitol-​ containing preparations for intravenous nutritional supplementation has now been stopped, in the era before fructose and sorbitol infu- sions were banned in Europe, at least 16 patients developed toxicity and died (Fig. 12.3.2.2). Some medicines that are given parenterally are still reconstituted in solutions containing harmful quantities of sorbitol or fructose. Hepatorenal failure has recently been reported after the administration of amiodarone in a polysorbate solution to a patient with hereditary fructose intolerance, with dire consequences. Prognosis Untreated hereditary fructose intolerance is a potentially fatal disease in infants and young children in whom it ultimately causes irreversible liver disease, renal tubular impairment, and episodic, life-​threatening hypoglycaemia. The proportion of infants that die of unrecognized fruc- tose toxicity is unknown but given the discrepancy between the preva- lence of mutant aldolase B alleles and the apparent rarity of the disease, the author estimates this to be at least one half of all those born with the condition. Occasionally, adolescents and adult patients may succumb to the inadvertent use of parenteral fructose or sorbitol, but this practice, which until recently was popular in German-​speaking countries, is now obsolete. With the introduction of a strict exclusion diet, the disorder is compatible with a normal quality and duration of life. Fructose diphosphatase deficiency
(OMIM 229700) Clinical features This very rare, recessively inherited disorder presents with hypo- glycaemia, ketosis, and lactic acidosis in early infancy. Fewer than 100 cases have been reported since its original description in 1970. Severe, sometimes fatal, acidosis is associated with infection and starvation, and most cases present within the first few days of life or in the neonatal period. Onset during the first year of life is the rule. In newborn infants, the severe metabolic disturbance shows it- self by acidotic hyperventilation, which may be accompanied by ir- ritability, disturbed consciousness, seizures, or coma. The unusual combination of ketonaemia, lactic acidaemia, and hypoglycaemia is induced by fasting, the administration of fructose, sorbitol, and glycerol, and by ingestion of a diet rich in fat. Episodes in the neo- natal period respond well to infusions of glucose and bicarbonate but, after an interval, further attacks occur, often provoked by inter- current infection. Lethargy accompanied by hyperventilation is fol- lowed abruptly by prostration, coma, and seizures. Investigations reveal hypoglycaemia, ketosis, and profound lactic acidosis; there is also hyperuricaemia, aminoaciduria, and ketonuria. If the infant survives, hepatomegaly due to fatty infiltration may be detected but overt clinical disturbances of hepatic or renal tubular function are not seen. The untreated disease is associated with growth retardation. Box 12.3.2.1  Food items not allowed for patients with hereditary fructose intolerance and fructose diphosphatase deficiencya • Table sugar • Fruit sugar, all fruit and fruit products, including tomatoes • Sorbitol—​often used in confectionery (especially), as an excipient or stabilizer in medication or diabetic foods • Honey, syrup, treacle, and molasses • Diabetic foods • Chocolate and sherbet • Preserves, jams, and marmalade • Frankfurters, honey-​roast ham, and sweet-​cured ham • Processed cheese spreads • Cream and cottage cheese with chives, pineapple, etc. • Flavoured milks and yoghurts • Wheatgerm, brown rice, and bran • Breakfast cereals • Coffee essence and powdered milk • Carbonated sweet drinks • Allspice, nuts, coconut, carob, and peanut butter • Mayonnaise, pickles, salad dressings, and sauces • Some potatoes (especially stored, new potatoes) • Most legumes a  More information is provided in ‘Further reading’.

12.3.2  Inborn errors of fructose metabolism 2001 The first infant to be affected by fructose diphosphatase deficiency in a given family may succumb before the diagnosis is established and in any case fares worse than siblings for whom the appropriate diet and prompt control of the condition are instituted. The response to treatment is favourable, however, and fructose diphosphatase deficiency is ultimately compatible with a benign course and with normal growth and development. Metabolic defect Deficiency of fructose 1,6-​diphosphatase causes failure of gluconeogenesis in the liver, although the abnormality may be de- tected in intestinal mucosa, kidney, and in cultured mononuclear cells from peripheral blood. The muscle isozyme of fructose 1,6-​ diphosphatase is not affected. Between meals, blood glucose is maintained by glycogenolysis and hence the onset of disturbed metabolism in fructose diphosphatase deficiency depends on the availability of hepatic glycogen. Since fe- brile illnesses accelerate the consumption of liver glycogen, the accom- panying anorexia with or without vomiting may deplete glycogen stores critically. Acidosis results from the accumulation of gluconeogenic precursors including lactate, pyruvate, and alanine as well as ketone bodies, which cannot be utilized. Hypoglycaemia that is unresponsive to glucagon and associated with exhaustion of glycogen stores occurs; it does not respond to normal gluconeogenic substrates (e.g. glycerol, amino acid solutions, dihydroxyacetone, sorbitol, or fructose); indeed, administration of these aggravates the metabolic disturbance. The pathogenesis of hypoglycaemia and accompanying disturbances in fructose diphosphatase deficiency is complex and not completely explained by exhaustion of hepatic glycogen stores. Well-​fed patients have a normal response to glucagon but are intolerant of high-​fat diets, as well as fructose, sorbitol, alanine, glycerol, and dihydroxyacetone administration. Challenge with these nutrients induces hypogly- caemia, hyperuricaemia, and hypophosphataemia, accompanied by an exaggerated rise in blood lactate levels. The hypoglycaemia is then unresponsive to glucagon, indicating a secondary inhibition of phosphorylase activity in the liver, which results from the build-​up of phosphorylated sugar intermediates that cannot be further metab- olized in the context of reduced intracellular free inorganic phosphate. Adenosine deaminase is activated primarily because of reduced phos- phate concentrations, so that purine nucleotides are broken down to uric acid. Failure to utilize glucogenic amino acids and metabolites such as dihydroxyacetone and glycerol appears to stimulate trigly- ceride formation in the liver, which induces steatosis. Unlike heredi- tary fructose intolerance (discussed previously), high concentrations of hepatic fructose 1-​phosphate do not occur, and profound disturb- ances of blood coagulation or hepatic or renal tubule function with progressive structural damage are absent in fructose diphosphatase deficiency. Similarly, aversion to foods that aggravate the disorder does not develop in affected infants and children; this may be explained by the absence of pain and abdominal symptoms in the condition. Diagnosis The importance of establishing the diagnosis of fructose diphosphatase deficiency cannot be overemphasized. Proper dietary control and protocols for the institution of appropriate therapy depend on recog- nizing the complex disturbance that underlies this disease. Fructose diphosphatase deficiency should be considered in other- wise normal infants who develop unexplained severe acidosis or hypoglycaemia associated with episodes of infection. The combination of ketosis and lactic acidosis with hypoglycaemia is highly suggestive of a disorder affecting the gluconeogenic pathway, including deficiency of glucose 6-​phosphatase, pyruvate carboxylase, pyruvate dehydrogenase, and phosphoenolpyruvate carboxykinase. The absence of abdominal distress, haemolysis, jaundice, coagulopathy, and disturbances of the proximal renal tubule differentiates the condition from hereditary fruc- tose intolerance, tyrosinosis, and Wilson’s disease. Confusion may arise with disorders associated with secondary defects in gluconeogenesis, es- pecially the Reye’s-​like syndrome caused by deficiencies of long-​chain, medium-​chain, and short-​chain acyl coenzyme A dehydrogenase activ- ities, as well as defects of carnitine metabolism. Organic acidaemias are also readily distinguished by biochemical screening methods. Provocative tests using food deprivation and the administration of infusions of fructose, sorbitol, or glycerol should be avoided in the acutely ill infant or child with suspected deficiency of fructose 1,6-​ diphosphatase (or fructose intolerance). The definitive diagnosis depends on the demonstration of selectively decreased fructose diphosphatase activity in tissue samples. Most frequently, the enzym- atic defect will be identified by biochemical assay of a freshly obtained liver biopsy specimen, which allows other metabolic disorders and gluconeogenic defects to be confidently excluded. The defect may also be demonstrated in biopsy samples of jejunal mucosa and in cultured monocyte-​derived macrophages obtained from peripheral blood. However, the presence of fructose 1,6-​diphosphatase in these tissues is metabolically inconsequential and, although useful for confirm- ation of the diagnosis where it is strongly suspected, in practice, de- cisive identification of this disorder normally depends on a systematic biochemical analysis of liver tissue in an experienced laboratory. The human fructose-​1,6-​diphosphatase (FBP1) gene maps to chromo- some 9q22.2–​q22.3, and inactivating mutations have been identified in the disease. Unlike fructose intolerance, however, these mutations tend to be private and thus individually of less diagnostic significance for routine laboratory use in this disorder since mutational heterogen- eity appears to be the rule. However, a minor exception to this occurs in the Japanese population, where one mutation (960–​961insG) ap- pears to account for almost one-​half of mutant FBPI alleles. Treatment Dietary control and avoidance of starvation with rapid relief of fe- brile illnesses are the mainstays of management. Minor infections and injuries require prompt attention, and intravenous glucose therapy should be instituted early in acute episodes to avoid hypo- glycaemia and acidosis. Fasting should be avoided as far as possible, while night-​time feeding may be needed in infants during recovery from injuries or infections, and after strenuous exercise in older chil- dren. The habit of taking meals at regular 4-​h intervals is best incul- cated when the patient is young. The diet should exclude excess fat; sorbitol, sucrose, and fructose must be strictly avoided. Breast milk is rich in lactose, which is readily assimilated, but difficulties arise on transfer to artificial feeds during weaning. In addition, medications and syrups containing fructose, sucrose, or sorbitol present a special danger to patients with fructose diphosphatase deficiency. A diet excluding these sugars but containing 56% calories as carbohydrate, with 32% calories as fat and 12% as protein, has produced normal growth and development. Acute episodes of acidosis or hypogly- caemia are controlled rapidly by intravenous administration of glu- cose with or without bicarbonate as required.

SECTION 12  Metabolic disorders 2002 FURTHER READING Ali M, Rellos P, Cox TM (1998). Hereditary fructose intolerance. J Med Genet, 35, 353–​65. Ananth N, et al. (2003). Two cases of hereditary fructose intolerance. Indian J Clin Biochem, 18, 87–​92. Baerlocher K, et al. (1978). Hereditary fructose intolerance in early childhood: a major diagnostic challenge. Survey of 20 symptomatic cases. Helv Paediatr Acta, 132, 605–​8. Baker L, Wingrad AI (1970). Fasting hypoglycaemia and metabolic acidosis associated with deficiency of fructose-​1,6-​diphosphatase deficiency. Lancet, ii, 13–​6. Bell L, Sherwood WG (1987). Current practices and improved recom- mendations for treating hereditary fructose intolerance. J Am Diet Assoc, 87, 721–​8. Bijarnia-​Mahay S, et al (2015). Molecular diagnosis of hereditary fruc- tose intolerance:  founder mutation in a community from India. JIMD Rep, 19, 85–​93. Bonthron DT, et al. (1994). Molecular basis of essential fructosuria: mo- lecular cloning and mutational analysis of human ketohexokinase (fructokinase). Hum Mol Genet, 3, 1627–​31. Bouteldja N, Timson DJ (2010). The biochemical basis of hereditary fructose intolerance. J Inherit Metab Dis, 33, 105–​12. Burlina AB, et al. (1990). Clinical and biochemical observations on three cases of fructose-​l,6-​diphosphatase deficiency. J Inherit Metab Dis, 13, 263–​6. Chambers RA, Pratt RTC (1956). Idiosyncrasy to fructose. Lancet, ii, 340. Choi HW, et al. (2012). A novel frameshift mutation of the ALDOB gene in a Korean girl presenting with recurrent hepatitis diagnosed as hereditary fructose intolerance. Gut Liver, 6, 126–8. Cox TM (1993). Iatrogenic deaths in hereditary fructose intolerance. Arch Dis Child, 69, 413–​15. Cox TM (2002). The genetic consequences of our sweet tooth. Nat Rev Genet, 3, 481–​7. Cox TM (2009). Hereditary fructose intolerance (fructosaemia). In: Lifton R, et al. (eds) Genetic diseases of the kidney, pp. 619–​43. Elsevier, New York. Cross NC, Tolan DR, Cox TM (1988). Catalytic deficiency of human aldolase B in hereditary fructose intolerance caused by a common missense mutation. Cell, 53, 881–​5. Cross NC, et al. (1990). Molecular analysis of aldolase B genes in her- editary fructose intolerance. Lancet, 335, 306–​9. Curran BJ, Havill JH (2002). Hepatic and renal failure associated with amiodarone infusion in a patient with hereditary fructose intoler- ance. Crit Care Resusc, 4, 112–​15. Davit-​Spraul A, et  al. (2008) Hereditary fructose intolerance:  fre- quency and spectrum mutations of the aldolase B gene in a large patients cohort from France—​identification of eight new mutations. Mol Genet Metab, 94, 443–​7. Debray FG, et al. (2018). Are heterozygous carriers for hereditary fructose intolerance predisposed to metabolic disturbance when ex- posed to fructose? Am J Clin Nutr, 108, 292–99. Dursun A, et al. (2001). Mutation analysis in Turkish patients with her- editary fructose intolerance. J Inherit Metab Dis, 24, 523–​6. Elpeleg ON, Hurvitz H, Branski D (1989). Fructose-​1,6-​diphosphatase deficiency: a 20-​year follow-​up. Am J Dis Child, 143, 140–​2. Gibson PR, et al. (2007). Review article: fructose malabsorption and the bigger picture. Aliment Pharmacol Ther, 25, 349–​63. Greenwood J (1989). Sugar content of liquid prescription medicines. Pharm J, 243, 553–​7. Gruchota J, et  al. (2006). Aldolase B mutations and prevalence of hereditary fructose intolerance in a Polish population. Mol Genet Metab, 87, 376–​8. Hunter RW, Hughey CC, Lantier L, et al. (2018). Metformin re- duces liver glucose production by inhibition of fructose-1-6- bisphosphatase. Nat Med, 24, 1395–1406. James CJ, et al. (1996). Neonatal screening for hereditary fructose in- tolerance: frequency of the most common mutant aldolase B allele (A149P) in the British population. J Med Genet, 33, 837–​41. Kelishadi R, Mansourian M, Heidari-Beni M (2014). Association of fruc- tose consumption and components of metabolic syndrome in human studies: a systematic review and meta-analysis. Nutrition, 30, 503–10. Kikawa Y, et al. (2002). Diagnosis of fructose 1,6-​bisphosphatase defi- ciency using cultured lymphocyte fraction: a secure and noninvasive alternative to liver biopsy. J Inherit Metab Dis, 25, 41–​6. Krishnamurthy V, et al. (2007). Three successful pregnancies through dietary management of fructose-​1,6-​bisphosphatase deficiency.
J Inherit Metab Dis, 30, 819. Lanaspa MA, Andres-Hernando A, Orlicky DJ, et al. (2018). Ketohexokinase C blockade ameliorates fructose-induced metabolic dysfunction in fructose-sensitive mice. J Clin Invest, 128, 2226–38. Lee HJ, Cha JY (2018). Recent insights into the role of ChREBP in ­intestinal fructose absorption and metabolism. BMB Reports, 51, 429–36. Li H, et al. (2018). Acute liver failure in neonates with undiagnosed hereditary fructose intolerance due to exposure from widely avail- able infant formulas. Mol Genet Metab, 123, 428–​32. Mock DM, et al. (1983). Chronic fructose intoxication after infancy in children with hereditary fructose intolerance: a cause of growth retardation. N Engl J Med, 309, 764–​70. Moses SW, et  al. (1991). Fructose-​1,6-​diphosphatase deficiency in Israel. Isr J Med Sci, 27, 1–​4. Odièvre M, et al. (1978). Hereditary fructose intolerance in childhood. Diagnosis, management and course in 55 patients. Am J Dis Child, 132, 605–​8. Pagliara AS, et  al. (1972). Hepatic fructose-​1,6-​diphosphatase defi- ciency. A  cause of lactic acidosis and hypoglycaemia in infancy.
J Clin Invest, 51, 2115–​23. Pronicka E, et al. (2007). Elevated carbohydrate-​deficient transferrin (CDT) and its normalization on dietary treatment as a useful bio- chemical test for hereditary fructose intolerance and galactosemia. Pediatr Res, 62, 101–​5. Reimers A, Spigset O (2003). Declaration of fructose and fructose-​ related adverse effects in commercial drug preparations in European countries. Drug Saf, 26, 1057–​9. Sachs B, Sternfeld L, Kraus G (1942). Essential fructosuria: its patho- physiology. Am J Dis Child, 63, 252. Santer R, et al. (2005). The spectrum of aldolase B (ALDOB) mutations and the prevalence of hereditary fructose intolerance in Central Europe. Hum Mutat, 25, 594. Shepherd SJ, Gibson PR (2006). Fructose malabsorption and symp- toms of irritable bowel syndrome: guidelines for effective dietary management. J Am Diet Assoc, 106, 1631–​9 Simons N, et al. (2019). Patients with aldolase B deficiency are char- acterized by an increased intrahepatic triglyceride content. J Clin Endocrinol Metab, pii: jc.2018-02795. Steinmann B, Gitzelmann R, Van den Berghe G (2001). Disorders of fructose metabolism. In: Scriver CR, et al. (eds) The metabolic and molecular bases of inherited disease, 8th edition, vol. II, pp. 1489–​520. McGraw-​Hill, New York. http://​www.ommbid.com. Softic S, et al. (2017). Divergent effects of glucose and fructose on hep- atic lipogenesis and insulin signaling. J Clin Invest, 27, 4059–​74. Thabet F, et al. (2002). Severe Reye syndrome: report of 14 cases managed in a pediatric intensive care unit over 11 years. Arch Pediatr, 9, 581–​6. Valadares ER, et al. (2015). Hereditary fructose intolerance in Brazilian patients. Mol Genet Metab Rep, 15, 35–​8.

12.3.3 Disorders of galactose, pentose, and pyruva

12.3.3 Disorders of galactose, pentose, and pyruvate metabolism 2003

12.3.3  Disorders of galactose, pentose, and pyruvate metabolism 2003 Yang TY, et al. (2000). Hereditary fructose intolerance presenting as Reye’s-​like syndrome: report of one case. Acta Paediatr Taiwan, 41, 218–​20. Wasserman D, et al. (1996). Molecular analysis of the fructose trans- porter gene (GLUT5) in isolated fructose malabsorption. J Clin Invest, 98, 2398–​402. World Health Organization (WHO) (2015). Sugars Intake for Adults and Children. WHO/​NMH/​NHD/​15.2 (Executive summary). WHO, Geneva. 12.3.3  Disorders of galactose, pentose, and pyruvate metabolism Timothy M. Cox ESSENTIALS Inborn errors of galactose metabolism Galactose is principally found as free lactose in dairy products. Three inborn errors of galactose metabolism are recognized: Galactokinase deficiency (‘galactose diabetes’)—​a very rare con- dition which impairs the assimilation of dietary galactose such that the free sugar and its metabolites appear in plasma and urine. Conversion of galactose to osmotically active galactitol in tissues causes premature bilateral cataracts and (occasionally) pseudotumour cerebri in infants, which are prevented by early in- stitution of a galactose-​ and lactose-​free diet. Classical galactosaemia (galactose-​1-​phosphate uridylyltransferase deficiency)—​the commonest (1/​47  000 births) and most important
disorder. High concentrations of galactose in the plasma and tissues
lead to aberrant glycosylation of glycoproteins and other glycoconjugates, including lipids. The principal manifestations are a bac- tericidal defect associated with neonatal Escherichia coli sepsis; failure to thrive; and—​in older patients—​growth retardation, mental retardation, renal Fanconi’s syndrome, jaundice, and hepatosplenomegaly: without exclusion of lactose and galactose, death with cirrhosis is the rule. Diagnosis is made by plasma galactose, galactose 1-​phosphate, and red cell transferase determinations in blood spots obtained after birth and refined by molecular analysis of the GALT gene. Prompt institution of an appropriate diet allows survival into late adult life, but disabling cognitive and language defects and other neurological manifestations persist. Attenuated nonacute galactosaemia presents in adult life with cataracts and progressive neurological disease; premature ovarian failure is the rule in affected women. Uridine diphosphate galactose-​4´-​epimerase deficiency—​a rare but often harmless disorder which may be identified by neonatal screening. Rarely, cataract, sensorineural deafness, and impaired psychomotor development with hepatorenal features of classical galactosaemia occur, with favourable responses to the galactose exclusion diet. Pentosuria Essential pentosuria is an asymptomatic, autosomal recessive trait affecting glucuronate metabolism, principally found in Ashkenazi Jews. Disorders of pyruvate metabolism Deficiency of the pyruvate dehydrogenase complex is the most common inherited disorder with lactic acidaemia, most often due to deficiency of the E1α subunit inherited as a dominant X-​linked character. Presentation is with overwhelming neonatal acidosis; moderate lactic acidosis with progressive neurological features; or—​ in male children and young adults—​an indolent neurological course without overt acidosis but with episodes of cerebellar ataxia induced by carbohydrate administration. Pyruvate carboxylase deficiency causes lactate/​pyruvate acidosis with a necrotizing encephalopathy resembling Wernicke’s encephalopathy. Hypoglycaemia may complicate intercurrent infections and starvation. Disorders of galactose metabolism Metabolism of galactose Galactose is derived from exogenous sources but constitutively de- rived de novo by metabolic interconversion from endogenous glu- cose. The chief exogenous source is the disaccharide, lactose, which is present in milk and dairy products, following the action of mu- cosal lactase in the small intestine. The concentration of lactose in human breast milk is approximately 70 g/​litre (200 mM); this explains the relative sweetness of human breast milk, compared with cows’ milk. Newborn infants obtain about one-​fifth of their dietary energy supply in the form of galactose, which is obtained by digestion of lactose to galactose and glucose in equimolar amounts. Galactose occurs as a free sugar in certain fruits such as tomatoes and avocados, as well as legumes, brassicas, and other vegetables. It is also complexed with other molecules present in food and is a component of membrane glycoproteins, glycosamino- glycans, and glycosphingolipids abundant in nervous tissue. Assimilation of galactose from dietary sources and the de novo biosynthesis of metabolically active galactose nucleotides share reactions involving the interconversion of galactose and glucose common to the Leloir pathway. This pathway utilizes nucleoside (uri- dine) diphosphate sugar intermediates that interact with galactose 1-​phosphate which directly enters mainstream carbohydrate me- tabolism (Fig. 12.3.3.1). High-​energy uridine diphosphate (UDP)-​ sugar intermediates, especially UDP-​glucose, UDP-​galactose, and their nitrogen-​containing derivatives such as UDP-​galactosamine, are critical building blocks in the formation of glycoproteins and glycolipids, including glycosphingolipids. Intracellular concentra- tions of these metabolites reflect the activity of the Leloir pathway. The Leloir pathway employs four enzymes that catalyse essen- tial reactions for the metabolic incorporation of galactose derived from dietary sources and by de novo biosynthesis. By their actions, hexose units derived from galactose enter glycolysis, are incorp- orated into glycogen, and used for key biosynthetic processes. The enzymes are (1) galactose mutarotase—​which facilitates intercon- version of the α-​ and β-​d-​galactose anomers to maintain the source of galactose in the alpha conformation necessary for its further assimilation. (2)  Galactokinase—​which catalyses the rapid phos- phorylation of β-​d-​galactose to form galactose 1-​phosphate in the liver and renal proximal renal tubule. (3)  Galactose 1-​phosphate uridyltransferase—​responsible for the conversion of galactose 1-​ phosphate and UDP-​glucose to glucose 1-​phosphate and UDP-​gal- actose. (4) UDP-​galactose-​4´-​epimerase—​which reversibly catalyses

SECTION 12  Metabolic disorders 2004 regeneration of UDP-​glucose and allows formation of UDP-​ galactose from UDP-​glucose. UDP-​galactose is a critical building block for the biosynthesis of essential glycoconjugates, includ­ ing glycosaminoglyans, glycoproteins, and glycosphingolipids. Inherited defects in the interconversion of these metabolites in- crease blood and tissue concentrations of galactose—​especially when the diet contains milk or dairy products. Exogenous pathway Free galactose is released from lactose (β-​d-​galactopyranosyl-​ (1→4)-​d-​glucose) in the diet by the action of lactase at the intestinal brush-​border membrane. Lactase preserves in the reaction products, galactose and glucose, the anomeric β-​configuration of the galactose moiety in the substrate. After intestinal uptake of these monosac- charides by the sodium-​dependent hexose transporter (SGLUT1), galactose enters the portal bloodstream by facilitated diffusion across the basolateral membrane via GLUT2, and this low-​affinity, high-​ capacity transporter also brings about the uptake by hepatocytes. Metabolic incorporation of galactose occurs only as its α-​d-​ galactose anomeric form, which is supplied by the action of the ubiquitous bidirectional enzyme, mutarotase, that maintains the conformational equilibrium of the α-​ and β-​anomers of d-​galactose. Intracellular incorporation of exogenous d-​galactose that is se- lective for the α-​anomer involves an endergonic reaction catalysed by galactokinase; phosphate-​bond energy from ATP drives the reaction thermodynamically. After rapid phosphorylation of free galactose at the 1-​carbon position to form galactose 1-​phosphate, principally in the liver, the high-​energy sugar nucleotide, UDP-​galactose, is generated by the action of galactose-​1-​phosphate uridyltransferase. UDP-​glucose can be regenerated by the action of UDP-​galactose-​4´-​ epimerase, which promotes flux through the Leloir pathway. De novo synthesis UDP-​galactose-​4´-​epimerase enables the galactose moiety to be generated from glucose for the synthesis of complex glycoconjugates, providing a de novo supply of UDP-​galactose that is independent of exogenous galactose from milk and other dietary components. The pathway is most active in the growing fetus. While the Leloir pathway is the main route of galactose metabolism in humans, minor pathways also operate: reduction of galactose to galactitol and oxi- dation to galactonic acid can occur, but these only partially mitigate the metabolic block at the level of the transferase that causes classic galactosaemia. A major source of endogenous galactose is the recyc- ling of galactose-​containing glycosaminoglycans, glycoproteins, and glycosphingolipids in the lysosomal compartment. Galactitol appears at high concentrations in the blood and urine in classic galactosaemia, galactokinase deficiency, and epimerase defi- ciency. If the transferase is markedly deficient or absent, galactose, gal- actose 1-​phosphate, galactitol, and gluconate accumulate in the tissues. Galactokinase deficiency: ‘galactose diabetes’ Failure to phosphorylate galactose in the liver and other tissues impairs its clearance from the blood so that the free sugar mainly derived from lactose in the intestine, as well as its metabolites, galactonic acid and galactitol, appear in the urine. Genetics The human gene for galactokinase maps to chromosome 17q24, with a putative second locus on chromosome 15. Numerous mutations respon- sible for galactokinase deficiency have been identified in the GALK1 gene at its chromosome 17 locus. Many of these are private, but the so-​ called Osaka variant, a missense mutation (p.A198V), was first identi- fied through mass neonatal screening and has a prevalence of 4.1% in Japanese individuals and 2.8% in Koreans; it is uncommon among indi- viduals of Taiwanese and Chinese ancestry. The Osaka GALK1 variant has been reported to occur in 7.8% of Japanese adults with bilateral cata- racts, in whom it may represent a true population risk factor. Homozygous deficiency of galactokinase is exceptionally rare, occurring with an approximate frequency of 1 in 1  000  000 live births. However, galactokinase deficiency occurs in East central Europe with a high incidence among the Roma (‘Gypsies’), in whom a founder Romani GALK1 mutation, p.P28T, has been found in Spain, Bulgaria, Bosnia and Hungary. The carrier frequency in these groups is about 1 in 50, predicting a birth frequency of about 1 in 10 000 and justifying institution of a screening programme that is culturally acceptable to and actionable within this population group. Clinical features Precocious formation of cataracts in infants and children is charac- teristic, and some heterozygotes develop cataracts before the age of 40 years. When blood concentrations are high, galactose is taken up by the lens and converted to the end product, the sugar alcohol, galactitol, by the action of aldose reductase: subsequent toxic or osmotic effects lead to swelling and irreversible damage to lens fibres. Several infants have presented with benign intracranial hypertension (pseudotumour cerebri), possibly as a result of comparable osmotic effects of galactitol in the brain. Patients with galactokinase deficiency persistently excrete reducing sugar in their urine, but, apart from possible confusion with diabetes mellitus, this has no apparent significance. Dietary lactose Galactose Glucose Galactose 1-phosphate UDP-glucose UDP-galactose Glucose 1-phosphate Galactose 1-phosphate uridyltransferase Galactokinase UDP 4-epimerase Glycogen Glucose 6-phosphate Intestinal lactase Fig. 12.3.3.1  Galactose metabolism.

12.3.3  Disorders of galactose, pentose, and pyruvate metabolism 2005 Diagnosis and treatment Galactokinase deficiency should be suspected in infants or chil- dren with cataracts, and ideally reducing sugar (which will not react with glucose oxidase test strips) should be sought in the urine. Definitive diagnosis is by enzymatic assay of galactokinase in erythrocytes or cultured fibroblasts, which differentiates the dis- order from classic galactosaemia and hypergalactosaemia due to vascular disease in the liver. In populations with newborn surveillance for high blood gal- actose concentration, the deficiency may be detected as a result of finding an abnormal blood galactose concentration with normal transferase and epimerase activities. Definitive enzymatic meas- urements can be conducted on amniocytes and cultured skin fibro- blasts. Neonatal screening that depends on tests for galactose in the blood will not detect galactokinase deficiency. Lifelong treatment with a lactose-​ and galactose-​exclusion diet prevents cataract formation, and early cataract formation in infants may even be reversed; otherwise surgical removal may be required. Urinary galactitol concentrations, which have been reported to ex- ceed 2500 mmol/​mol creatinine, fall to within the reference range for healthy subjects (<3 mmol/​mol creatinine) after some weeks of dietary treatment. Although there are numerous reports of bilateral cataracts in heterozygotes for galactokinase deficiency, it remains unclear whether any propensity to cataract formation in later life is prevented by dietary restriction; some authors have suggested that cataracts are more fre- quent in otherwise healthy individuals who consume abundant dairy products, but have no galactokinase deficiency. In the face of this con- troversy, it appears to be most prudent to recommend modest restric- tion of lactose intake in heterozygotes for galactokinase deficiency. Galactose-​1-​phosphate uridyltransferase deficiency: galactosaemia Unlike individuals in whom galactokinase is deficient, when patients who lack galactose-​1-​phosphate uridyltransferase ingest lactose, there is a significant rise in intracellular galactose 1-​phosphate as well as blood galactose concentrations. The severe consequences of classic galactosaemia are attributed to the toxic effects of galactose 1-​ phosphate, principally in the liver, proximal renal tubule, and brain. Three main forms of galactosaemia are readily recognized: they may be termed classical, adult (variant), and mild—​most pa- tients present in the neonatal period with the classical variant. Increasingly, patients with strongly predictive galactose-​1-​phos- phate uridyltransferase deficiency, enzymology and/​or GALT geno- type for this acute disease are identified by neonatal screening. Genetics Galactosaemia is transmitted as an autosomal recessive trait with an overall estimated frequency of 1 in 47 000 in liveborn infants. It is more frequent in some isolated groups, most notably in the modern Traveller population of Ireland, where there is a birth fre- quency of 1 in 480 compared with 1 in 30 000 in the non-​Traveller Irish population. In African American patients from the United States of America, a relatively mild disorder has been reported that is probably due to an unstable enzyme variant; uridylyltransferase activity is absent from the red cells of these patients but amounts to some 10% of normal in samples of liver and small intestinal tissue. Patients with the so-​called Duarte variant possess about one-​half of the normal enzyme activity in erythrocytes but remain asymptom- atic; premature ovarian failure does not occur in women harbouring this variant. Galactosaemia is rare in Japan. The human galactosyl-​1-​phosphate uridylyltransferase gene, GALT, maps to human chromosome 9p13 and encodes a protein of molecular weight 43 kDa, which exists as a functional homodimer. Molecular analysis indicates that most patients with classic galactosaemia har- bour missense-​type mutations and are compound heterozygotes. Numerous variant transferase enzymes are known, and more than 330 disease-​associated mutations are reported. There are several wide- spread mutations: p.Q188R accounts for about two-​thirds of mutant alleles in white Europeans, with a strong North-​West dominance, and greater than 90% of mutant alleles in patients from Ireland. In Eastern and Central Europe, the mutant GALT p.K285N accounts for about for 30% of galactosaemic alleles and has a strong association popula- tions of Slavic origin. The so-​called Duarte transferase mutation that is most frequent in persons of African ancestry has been identified as p.N314D. Molecular analysis of the transferase gene, GALT, now renders prenatal diagnosis of at-​risk pregnancies possible. Pathogenesis Although the exact mechanism of toxicity is unknown, the accumu- lation of galactose 1-​phosphate in a milieu with depleted inorganic phosphate probably inhibits other enzymatic reactions involving phos- phorylated intermediates and may cause purine nucleotide depletion. Aldose reductase is responsible for the direct reduction of gal- actose to galactitol, which is not metabolized further in the polyol pathway and accumulates in tissues where it contributes to the pathophysiology of galactosaemia, resembling sorbitol in its ability to cause rapid-​onset cataract formation, and with potential effects in the induction of cerebral oedema and pseudotumour cerebri. Sustained excess of d-​galactose as well as relative deficiency or distribution of high-​energy sugar nucleotides that are required for key glycosylation reactions is likely to have effects on the brain lipids and countless glycoproteins, including circulating hor- mones. It appears plausible that the deficiency of UDP-​galactose will affect the biosynthesis of key galactosphingolipids by UDP-​ galactosylceramide transferase in neural cells. A toxic effect on the fetal ovary due to maternal hypergalactosaemia has been postulated to account for the hypergonadotropic hypo- gonadism in affected women and girls, but abnormal glycosylation of follicle-​stimulating hormone and Müllerian factor have also been suggested. Clinical features Classical galactosaemia Classical galactosaemia is associated with absent or near absent galactose-​1-​phosphate uridyltransferase activity, and typically with the GALT p.Q188R/​Q188R genotype accompanied by severe clin- ical features in the neonatal period. Affected infants nearly always appear normal at birth, but vomiting or diarrhoea, jaundice, and hepatomegaly usually occur in the first few weeks. There is failure to gain weight, spontaneous bruising, and progressive enlargement of the liver. Cataracts may be apparent at 1 month of age, by which time abdominal distension with ascites has developed. Pseudotumour cerebri, often presenting with prom- inent anterior fontanelle presumably related to the osmotic effects of galactitol, may be apparent shortly after birth.

SECTION 12  Metabolic disorders 2006 Learning difficulties do not become apparent until later in the first year of life and vary greatly in severity. Many patients with galactosaemia develop severe infections with Escherichia coli during the neonatal period: Gram-​negative bacterial sepsis may be the first indication of this disorder in young infants. A bactericidal defect in circulating leucocytes has been postulated. In adult patients after reversal of the acute galactose toxicity syndrome, the most obvious sequelae are growth failure, neurological deficit, and, in women, pri- mary ovarian failure with infertility. A few patients with galactosaemia remain asymptomatic while ingesting milk, but eventually fail to gain weight. Such patients may come to light during childhood or even adult life with varying degrees of learning difficulties and cataracts. Hepatomegaly and intermittent galactosuria are usually present, and often there is a his- tory of feeding difficulties on institution of modified formula feeds during the neonatal period. The neurological manifestations of classic galactosaemia are highly variable but, despite prompt institution of dietary therapy, a degree of intellectual disability is common in affected children and adults. Characteristic learning difficulties in mathematics and spa- tial relationships with behavioural deficits have been observed, and children with galactosaemia have a particularly high risk for lan- guage impairment. Neurological manifestations, despite induction of a galactose-​free diet, can include seizures, apraxia, extrapyram- idal disorders with tremor and dystonia, and cerebellar deficits. Histological examination of the brain shows nonspecific signs of in- jury with gliosis and Purkinje cell loss in the cerebellum. Serum tests of liver function are nonspecifically deranged but histological examination of the liver shows lobular fibrosis, fatty change, bile ductular proliferation, and progression to frank cir- rhosis. Involvement of the proximal renal tubule is shown by gen- eralized aminoaciduria and occasionally a full-​blown Fanconi’s syndrome with vacuolation of tubular epithelial cells. Follow-​up studies of female patients with galactosaemia have shown a high incidence of gonadal failure with ovarian atrophy. Although this complication appears to be more common in patients in whom dietary therapy was delayed, no clear cause-​and-​effect relationship has been established. Men with galactosaemia have been reported to have a higher than expected prevalence of cryptorchidism and low semen volumes, but the specificity of these findings in patients with this chronic metabolic disease is unclear. In the clear-​cut case of women and adolescent females, hypogonadism and premature ovarian failure has more obvious effects on health, well-​being, and potential parental fulfilment. However, pregnancies occur in a few women with classic galactosaemia and usually result in the birth of healthy infants with no evidence of teratogenic effects. Lactation and breastfeeding pro- ceed normally. Since most females develop premature ovarian failure with hypergonadotropic hypogonadism, the hypogonadism and the imposed nutritional factors for metabolic control contribute to osteo- porosis, which occurs at high frequency, even in young adults. Adult or variant galactosaemia Adult or variant galactosaemia is a relatively indolent but important disease that often defies prompt conventional diagnosis. Residual galactose-​1-​phosphate uridyltransferase activity (c.10% of healthy reference mean) is detectable and the homozygous GALT p.S135L/​ S135L missense mutation is typically present. The condition presents with neurological disease and early cataracts; hepatomegaly is not prominent and decompensated liver disease does not occur. About 85% of women with this condition have premature ovarian failure reflected in delayed menarche, amenorrhoea, oligomenorrhoea, and/​or secondary amenorrhoea/​premature menopause. Mild galactosaemia Mild, or biochemical, galactosaemia arises from biochemical screening in at-​risk or targeted populations: the subjects harbour GALT mutations with modest effects on galactose-​1-​phosphate uridyltransferase activity and/​or stability (c.25%), of which the com- pound heterozygous genotype, p.N314D/​Q188R, the so-​called Duarte 2 variant, is characteristic. These individuals are asymptomatic. Diagnosis Recognition of hereditary galactosaemia in early infancy is of paramount importance since the acute effects of galactose poi- soning may be reversed by the institution of a lactose-​exclusion diet. Nearly all infants with classic galactosaemia or clinical variant galactosaemia can be identified in newborn screening that includes testing for galactosaemia. However, clinical variant galactosaemia may be missed if total blood galactose is analysed without deter- mining red cell transferase activity. Definitive diagnosis relies on the determination of galactose-​ 1-​phosphate uridylyltransferase activity and other galactose-​ metabolizing enzymes in red cells, skin fibroblasts, or leucocytes by means of a specific enzymatic assay. This procedure is required to confirm the results of initial screening tests conducted on dried blood spots (Beutler assay). The transferase activity in patients with classic galactosaemia is generally less than 1% of the normal reference range. In classical galactosaemia, red cell galactose-​1-​ phosphate is usually greater than 0.4 mmol/​litre (10 mg/​dL) and red cell transferase activity is absent or just detectable. It is important to realize that in many patients with attenuated (‘mild’ or asymptomatic) clinical variants of galactosaemia, trans- ferase activity is much higher in brain and intestinal tissue (c.10% of healthy reference values), even though it may be absent or barely detectable in the red cell assay. Individuals with attenuated, clin- ical variant galactosaemia may have erythrocyte activity close to or above 1% of reference values, but very rarely greater than 15%. Reliable enzymatic or genetic testing for heterozygotes can be conducted in the parents of a child who died before the diagnosis was confirmed. In particular populations, neonatal screening for elevated blood galactose and galactose-​1-​phosphate concentra- tions is carried out routinely. Molecular analysis of the GALT gene encoding galactose-​1-​phosphate uridylyltransferase in at-​risk preg- nancies is useful and can usually be requested for advising affected families (see following ‘Pregnancy’ subsection). Differential diagnosis Hereditary fructose intolerance and hereditary tyrosinaemia type 1 are credible differential diagnoses in infants and young children with the acute, classical form of galactosaemia. Transient galactosaemia with mixed glycosuria also occurs in the Fanconi–​Bickel syn- drome, now known to be due to biallelic mutations in the GLUT2 glucose–​galactose carrier, which is the facilitative glucose trans- porter in hepatocytes, pancreatic β-​cells, enterocytes, and proximal renal tubular cells. Studies in infants have shown that persistent hypergalactosaemia can also be explained by portosystemic venous

12.3.3  Disorders of galactose, pentose, and pyruvate metabolism 2007 shunts that are often associated with patent ductus venosus or other congenital vascular abnormalities in the liver. Doppler ultrasonog- raphy is a convenient noninvasive investigation to search for such shunts. Treatment Without strict dietary treatment, most patients with classical galactosaemia die in early infancy, although some may survive with liver disease and learning difficulties beyond childhood. The course of galactosaemia is strikingly altered on withdrawal of lactose (and galactose), although the outcome of neurological disease is often disappointing and it appears that the galactose-​free diet fails to confer benefit on mental development when instituted beyond the age of 2 years. An international guideline was set out in 2017 with practical recommendations for the management and follow-​up of patients with this disease. Dietary exclusion Lactose is present in many nondairy foods, hence advice from an experienced dietician, as well as meticulous attention to detail, is required to eliminate it satisfactorily. Free galactose is found in fruits and vegetables, especially avocados, peas and beans, as well as other legumes. In infants, soybean milks or commercial casein hydrolysates are used as milk substitutes, and therapy is monitored by periodic assay of red cell galactose-​1-​phosphate concentrations. Soya milk contains galactose equivalents complexed to other mol- ecules (about 15 mg/​litre) and there is a trend to adopt a completely galactose-​free artificial formula in the treatment of affected infants, but there is no proof that this biochemically successful strategy in- duces better long-​term outcomes. To avoid other potential vegetable sources of bound galactose in complex oligosaccharides is challenging and of uncertain value. However, even with high fruit and vegetable consumption, daily ex- posure to galactose from such sources is unlikely to exceed 70 mg (<0.4 mmol). This compares with the endogenous daily de novo syn- thesis of galactose in healthy adults, which approaches 6 mmol (1 g). A comprehensive retrospective study that examined more than 230 children and adults with classical galactosaemia found no as- sociation between rigorous nondairy galactose restriction in early childhood and five key long-​term outcomes. It thus appears un- necessary to take an extreme view of dietary measures and forbid fruit and vegetable intake once recovery from metabolic decompen- sation has occurred. Lifelong adherence to the exclusion of lactose and foods con- taining free galactose should be advocated. However, after challenge studies conducted in the Netherlands, the United Kingdom, and elsewhere, avoidance of fruits and vegetables is no longer standard practice. Monitoring metabolic control In the untreated state, the concentration of red cell galactose 1-​ phosphate is above 5 mmol/​litre, but with close adherence to the diet it falls within a few months to less than 0.25 mmol/​litre. Although biochemical monitoring has not been shown closely to predict out- comes, as a rule, expert centres recommend that the desired long-​ term target for blood galactose-​1-​phosphate concentrations should be about 0.15 mmol/​litre of erythrocytes; monitoring two to three times per year in the first decade is usually practised. Management of complications Hypogonadism is a considerable problem for young girls and women with galactosaemia: it affects self-​perception, physical devel- opment, life quality, self-​confidence—​and where relevant—​family life and expectations. Most women with galactosaemia develop oligomenorrhoea and secondary amenorrhea within a few years of their first period. In one series, only 5 out of 17 women older than age 22 years had normal menstruation. Women and adolescent girls with galactosaemia benefit from the ability to discuss matters related to their reproductive health and fertility, this enable them to obtain the necessary referrals for gynaecological and endocrinological treatment with safe hormone supplementation. Review allows for regular metabolic monitoring and serial evalu- ation of bone mineralization density and vitamin D status, with opportunities to intervene where appropriate to avoid the risk of fragility fractures, particularly in women with this disease who are a great risk of osteoporosis. Pregnancy In pregnancies of heterozygous mothers who have had affected chil- dren, there is evidence that premature cataracts can be avoided in the fetus if the maternal intake of lactose is restricted. In late preg- nancy, lactosaemia and lactosuria are common findings and result from the physiological induction of lactose biosynthesis in mam- mary tissue. In rare cases, there is a potential risk of self-​intoxication when women with homozygous deficiency of the transferase be- come pregnant and breastfeed, so that scrupulous dietary precau- tions are needed to maintain metabolic control during lactation. Organization of care Maintaining appropriate lifelong care for patients with galactosaemia in specialist clinics shows benefits in the provision of dietary man- agement with expert advice as well as developmental monitoring and assessment of cognitive function that is matched to educa- tional needs. Regular review in paediatric, transitional, and then adult metabolic specialist centres is critical for many patients who have overt or hidden difficulties with speech or cognition, to which should be added apraxia and the compounding effects of sensineural deafness. Prognosis The acute manifestations of galactosaemia and growth failure re- spond quickly to dietary therapy and cataract formation is prevented; in the early phases in the neonatal period, prompt intervention can lead to complete regression of cataracts. Unfortunately, some pa- tients have significant neurological deficits despite prompt and con- scientious treatment. An international survey reported the long-​term outcome in 350 patients receiving dietary therapy. The presence of ovarian failure and elevated galactose-​1-​phosphate concentrations in patients ap- parently ingesting no lactose or galactose emphasized the import- ance of the endogenous pathway and may also explain the emergence of neurological disease in treated patients. Several pregnancies have been reported in women with classic galactosaemia, including subjects homozygous for the p.Q188R mu- tation. In such pregnancies, high concentrations of galactitol were found in amniotic fluid, but cord blood values were within the range found in galactosaemic patients receiving strict dietary therapy.

SECTION 12  Metabolic disorders 2008 Thus, although galactitol present in maternal plasma can traverse the placenta, it probably does not harm the heterozygous fetus. Prevention Genetic counselling As a recessive condition, the diagnosis of galactosaemia has conse- quences for members of the affected pedigree. Appropriate genetic counselling is required. Screening Several retrospective studies indicate that neonatal screening pre- vents early death; in one survey, 80% of patients who underwent newborn screening were diagnosed by 14 days of age, compared with only 35% of patients who were not tested but who had manifest dis- ease. A Cochrane review in 2017 concluded that there were no ran- domized controlled studies or controlled clinical studies, published or unpublished, comparing the use of any newborn screening test to diagnose infants with galactosaemia and presenting a comparison between a screened population compared with a nonscreened popu- lation. No studies of newborn screening for galactosaemia were found. However, neonatal screening for galactosaemia is available in the United States of America and in Europe, but only a small per- centage of newborns in the United Kingdom are tested. Future prospects Galactokinase inhibitors—​restriction of exogenous substrate One approach to advance this field would be to embrace the toxicity hypothesis and target the overproduction and raised intracellular concentrations of galactose 1-​phosphate in classical galactosaemia. The immediate reaction target is galactokinase because this en- zyme catalyses the first committed step for the metabolic incorp- oration of α-​d-​galactose, upstream of the transferase in the Leloir pathway. At least one ‘lead’ candidate small-​molecule inhibitor has been identified for this potential therapeutic development, but con- cerns include (a) it would in effect generate systemic galactokinase deficiency, a recognized metabolic disease; (b) this would block all acquisition of exogenous galactose and, in combination with the severe transferase deficiency, may well have unforeseen effects on the endogenous galactose pathway; (c) it would fail to address the neurological and other effects of the disease that almost certainly depend on de novo synthesis of galactose and essential biosynthetic pathways; and (d) as a kinase, galactokinase will have numerous homologies with hundreds of critical kinases affecting metabolic regulation, hence securing adequate safety and proven selectivity of any inhibitor will be a formidable challenge. Enhancement of residual galactose-​1-​phosphate transferase activity Several promising small molecules that serve as potential pharma- cological chaperones have been identified, but the clinical target ex- tends beyond the liver to the brain, and the need to deliver the drug at sufficiently high concentrations to secure safe long-​term efficacy after traversing the blood–​brain barrier remains challenging. Gene therapy At the time of writing, investigators at Nationwide Children’s Hospital and the University of Utah have announced a stratagem based on delivering a codon-​optimized version of the human galactose-​1-​phosphate uridylyltransferase gene to express func- tional GALT protein. Using recombinant adeno-​associated viral vectors, abundant synthesis of the wild-​type transferase has been obtained in cells from patients with galactosaemia. It seems likely that the clinical development of this project will involve attempts to deliver a vector with suitable tissue tropisms to allow adequate trans- duction of the liver as well as the brain, with sustained expression of the therapeutic protein sufficient to alleviate the disease. Uridine diphosphate-​4´-​epimerase deficiency The gene for human UDP-​galactose-​4´-​epimerase has been mapped to chromosome 1p36–​p35, and numerous mutant alleles have been identified. Epimerase deficiency is a very rare autosomal reces- sive condition that may be identified during screening for classic galactosaemia. In most cases there are no symptoms attributable to galactosaemia, and follow-​up studies have confirmed the usu- ally benign nature of this anomaly. However, a few cases of more marked deficiency of UDP-​4´-​epimerase have been discovered in patients otherwise manifesting the classic features of galactosaemia. The autosomal recessive nature of this inherited disorder has been confirmed by demonstrating a partial epimerase deficiency in the healthy parents of an affected infant. The condition may be con- trasted with the transferase deficiency that allows the formation of small amounts of endogenous galactose in the presence of an intact epimerase. In the absence of epimerase activity, the individual is dependent on exogenous sources of galactose, since this cannot be derived from glucose. As a complete deficiency of the epimerase would lead to an absolute lack of UDP-​galactose for galactosphingolipid syn- thesis, the ingestion of very small quantities of galactose has been recommended so that brain development and biosynthesis of es- sential galactosides can proceed. Because of the dual activity of the epimerase towards UDP-​acetyl glucosamine as well as UDP-​glucose, it has been suggested that small supplements of the aminoacetyl gal- actosamine should also be provided. Pentosuria Pentosuria is caused by the excessive renal excretion of l-​xylulose. This has no clinical significance except that it may lead to the incor- rect diagnosis of diabetes mellitus should tests for reducing sugar be carried out on the urine. Xylulose does not react with urinary test strips based on the glucose oxidase method. The disorder has historical significance as one of the five original ‘inborn errors of metabolism’ investigated by Archibald Garrod in his seminal work. Although pentosuria is a rare autosomal reces- sive trait, its frequency in Ashkenazi Jews may be as high as 0.05%. It is caused by deficiency of l-​xylulose reductase, a nicotinamide adenine dinucleotide phosphate-​dependent enzyme in the oxida- tive pathway of glucuronate metabolism, resulting in the daily ap- pearance of 1 to 4 g xylulose and l-​arabitol in the urine. Output is continuous throughout life but greatly enhanced by the ingestion of glucuronic acid or drugs that are excreted as glucuronides. l-​xylulose reductase is present in many cells including red cells and hepatocytes. Several reactions remove the carboxyl carbon atom of d-​glucuronic acid to generate the pentose l-​xylulose, which is

12.3.3  Disorders of galactose, pentose, and pyruvate metabolism 2009 converted to its stereoisomer, d-​xylulose. d-​Xylulose is phosphor- ylated to d-​xylulose 5-​phosphate, which can be converted to hexose phosphates in the reactions of the pentose phosphate shunt. The diagnosis is made definitively by confirming the enzymatic defect in erythrocytes, but pentosuria is most readily confirmed by paper chromatographic analysis of urine using n-​butanol, ethanol, and water (50:10:40) as the partitioning solvent and orcinol-​ trichloroacetic acid as a detection agent; the sugar has a high mo- bility (RF 0.26) and is identified by its red colour on development. Contemporary methods of mass spectrometric analysis of urine are likely to detect the copious quantities of pentose directly and could give rise to temporary confusion in a child under investigation for unrelated illness. Long-​term monitoring of 40 individuals with pentosuria over more than 16 years showed no decrease in life expectancy. Inborn errors of pyruvate metabolism The organic acids, pyruvate and lactate, are key interconvertible intermediates in energy metabolism. Pyruvate is mainly generated from glucose, but also by oxidative deamination of alanine, from the other 3-​carbon amino acids, cysteine and serine, and indirectly from other amino acids. Breakdown of pyruvate proceeds by oxidation, first by pyruvate dehydrogenase, then the Krebs cycle, and finally the respiratory chain; anabolic assimilation of pyruvate is mediated by pyruvate carboxylase. Pyruvate is at the cross roads of energy metabolism: after entering the mitochondrion it generates acetyl coenzyme A and enters the Krebs cycle, by which it is oxidized. Pyruvate contributes the back- bone in the formation of amino acids including alanine, and con- tributes critically to gluconeogenesis after the action of pyruvate carboxylase and phosphoenolpyruvate carboxykinase. Pyruvate as a source of coenzyme A is used in lipogenesis. Lactate is the product of anaerobic glycolysis and is generated en- tirely from reduction of pyruvate by lactate dehydrogenase, and it is disposed of by the reversal of this reaction. Defective metabolism of pyruvate therefore readily leads to the accumulation of lactate, the development of lactic acidaemia and the build-​up of alanine. An important function of the Krebs (tricarboxylic acid) cycle is the generation of reduced nicotinamide adenine dinucleotide (NAD), which is used to generate chemical energy in the form of ATP by the action of the electron transport chain. When the metab- olism of pyruvate or the related α-​ketoacid dehydrogenases in the mitochondria is disrupted, it is not surprising that the high energy-​ requiring tissues of nervous system are almost invariably implicated. Pyruvate dehydrogenase deficiency Deficiency of pyruvate dehydrogenase, a mitochondrial enzyme complex which generates acetyl coenzyme A from pyruvate, is the most common cause of lactic acidosis in newborn infants and chil- dren, but it is also associated with neurodegenerative syndromes in adults. Coenzyme A is one of the critical substrates for the formation of citrate and without the feed-​forward delivery of pyruvate into the Krebs cycle, the cycle would be arrested and mitochondrial oxida- tive phosphorylation coupled to energy production would be mark- edly depressed. Pyruvate dehydrogenase is a multienzyme complex of five enzyme proteins: three (E1, E2, and E3) are catalytic, and two (pyruvate de- hydrogenase phosphatase and pyruvate dehydrogenase kinase) are regulatory. The combined molecular mass of the pyruvate dehydro- genase complex is about 8.5 million daltons and the complex com- prises 30 units of E1, 60 units of E2, and 6 units each of E3 and X (which is required to anchor E3 to E2). The whole complex of poly- peptides is the product of 10 distinct genes and requires three cofac- tors (thiamine pyrophosphate, lipoic acid, and coenzyme A) as well as the binding protein for one of the catalytic subunits (E3BP). The E1 catalytic protein is a heterotetramer of two α subunits and two β subunits. It is activated by dephosphorylation through the ac- tion of pyruvate dehydrogenase phosphatase and catalyses the rate-​ limiting reaction of pyruvate oxidation, for which it requires the activated form of thiamine (vitamin B1), thiamine pyrophosphate. Phosphorylation of the E1 complex is down-​regulated by pyruvate dehydrogenase kinase, which orchestrates reciprocal allosteric con- trol of pyruvate oxidation. The second and third catalytic proteins, E2 (dihydrolipoamide S-​ acetyltransferase) and E3 (dihydrolipoamide dehydrogenase), are linked by co-​binding to E3-​binding protein and have shared func- tions and combine functionally with the Krebs cycle dehydrogenases, namely the α-​ketoglutarate dehydrogenase and branched-​chain α-​ ketoacid dehydrogenase complex. Genetics E1α-​subunit mutations While all the genes that encode components of the pyruvate de- hydrogenase complex map to the nuclear genome, the most common cause of pyruvate dehydrogenase deficiency is due to mu- tations in the E1α subunit, a protein encoded on the short arm of the X chromosome (Xp22.12). Although the disease is characteristically more severe in males, manifestations in the heterozygous female are unusually frequent for an X-​linked disease and probably reflect the low functional reserve of the enzyme complex in the brain and the adverse cell-​intrinsic effects of lyonization in the mosaic situation. Only one-​quarter of the mothers of male patients harbour a causal mutation, thus most patients arise by new germline mutations and recurrence in further offspring in the same pedigree is uncommon. Mutations in other subunits Pyruvate dehydrogenase deficiency can be caused by mutations in the E1β subunit, E1 phosphatase deficiency, E2 (dihydrolipoamide acetyltransferase deficiency), E3 (dihyrolipoamide dehydrogenase deficiency), and E3BP (E3 binding protein) Biochemical defect The pyruvate dehydrogenase complex catalyses the conversion of pyruvate to acetyl CoA within mitochondria and is rate limiting for aerobic metabolism of glucose in the brain. Daily glucose consump- tion is 125 g in the adult brain, hence the pyruvate dehydrogenase complex is critical for brain metabolism since this is normally en- tirely dependent on the oxidative breakdown of glucose. Where the activity of the complex is impaired, accumulated pyruvate may either be reduced to lactate or transaminated to alanine, so that hyperalaninaemia and varying degrees of lactic acidaemia occur. Very rare defects in dihydrolipoyl dehydrogenase are associated

SECTION 12  Metabolic disorders 2010 with deficiency of branched-​chain ketoacid dehydrogenase, pre- sumably because of the shared molecular function. Failure to carry out oxidative reactions in regions of the cortex and midbrain causes neuronal death; deficiency of four-​carbon intermediates may critically impair synthesis of neurotransmitter molecules and lead to a Parkinsonian phenotype. There are three main activities associated in the complex: (1) pyru- vate dehydrogenase, a thiamine pyrophosphate-​dependent complex (E1); (2) dihydrolipoyl transacetylase (E2); and (3) dihydrolipoyl dehydrogenase, a flavoprotein (E3). Also associated are a pyruvate dehydrogenase-​specific kinase and phosphatase (both involved in overall metabolic regulation of the complex) as well as an es- sential lipoate-​containing protein other than dihydrolipoamide transacetylase in the pyruvate dehydrogenase complex (X-​lipoate), which possesses an acyl transfer function. Clinical features and prognosis The extent of clinical expression of the enzymatic defect is highly variable, but three principal patterns of pyruvate dehydrogenase complex defects are recognizable: (1) neonatal lactic acidosis, fre- quently associated with agenesis or dysgenesis of the corpus cal- losum; (2) Leigh’s encephalopathy in infants and children up to the age of 5 years; and (3) intermittent ataxia in adults. In females, mu- tations in the E1α subunit cause more homogeneous but severe dis- ease with dysmorphism, microcephaly, spastic paraplegia, and mild/​ moderate cognitive impairment. There may be fulminant disease in the newborn infant:  intra- uterine development is impaired, marked acidosis (blood lactate

10  mmol/​litre) is present at birth, and the condition is rapidly fatal. In other cases, lactic acidaemia may not be apparent and the disease comes to light because of intrauterine growth failure, neo- natal hypotonia/​asphyxia and feeding difficulty, and the principal abnormality is progressive psychomotor retardation often accom- panied by brainstem injury and disease of the basal ganglia. There is dysgenesis with structural abnormalities of the olivopontocerebellar tract and periventricular grey matter. Cortical atrophy and agenesis of the corpus callosum have also been reported in association with spastic quadriplegia, especially in patients presenting with neonatal acidosis. In patients who present with severe acidosis at birth, sub- acute necrotizing encephalomyelopathy of the Leigh type has been confirmed at necropsy with cystic appearances principally in the cerebral cortex, basal ganglia, and brainstem. Without intensive treatment, death usually occurs in infancy; however, should feeding by gavage be instituted, there is a protracted course with failure of neurological development, microcephaly, quadriplegia, seizures, and blindness due to the development of optic atrophy. Intermittent cerebellar ataxia or torsion dystonia have been recorded, and choreoathetoid movements occur. Peripheral neuropathy with onset in infancy has been observed. Involuntary eye movements in children are associated with a progressively deteriorating course. E1α subunit mutations A milder form of the disorder occurs in defects of the X-​linked E1α gene (PDHA1), but because pyruvate dehydrogenase deficiency is of key importance in brain metabolism, expression of disease is ob- served in females and affected males, hence this form of pyruvate de- hydrogenase complex deficiency is an X-​linked dominant disorder. In boys, episodic cerebellar ataxia may be induced by feeding carbohydrate-​rich foods or medicinal glucose. In these patients, some of whom are otherwise unimpaired and have normal intelli- gence, blood lactate concentrations may only be trivially elevated, and they do not normally exceed 10 mmol/​litre in this condition. In contrast, in other patients a progressive brainstem disorder oc- curs, characteristic of Leigh’s disease, with haemorrhagic necrosis and symmetrical spongiform appearances in the periventricular grey matter, thalami, midbrain, pons, medulla, and spinal cord; the mammillary bodies are spared. There are fascinating similarities between pyruvate dehydro- genase complex deficiency and diseases related to thiamine defi- ciency, with or without induction by exposure to alcohol (ethanol). About one-​third of patients with pyruvate dehydrogenase complex deficiency have facial appearances reminiscent of the fetal syndrome due to maternal consumption of excess alcohol. This dysmorphism is characterized by a narrow head, retroussé nose, flared nostrils, and an elongated philtrum; there is frontal bossing of the skull and a broad nasal bridge. In the acquired syndrome, acetaldehyde from the maternal circulation is believed to inhibit pyruvate dehydro- genase in the fetus, and Robinson and colleagues have suggested that low endogenous activity of the pyruvate dehydrogenase com- plex due to genetic deficiency in the fetus is responsible for the de- velopmental abnormalities. A striking connection between agenesis of the corpus callosum, usually in patients with neonatal pyruvate dehydrogenase defi- ciency, has been made with the Marchiafava–​Bignami syndrome, a condition characterized by degeneration of the corpus callosum and associated with longstanding abuse of alcohol. Finally, in Wernicke’s encephalopathy, the effects of thiamine de- ficiency and deficiency of the pyruvate dehydrogenase complex on the brain occur principally in the regions of the greatest metabolic activity, especially in the brainstem and basal ganglia. Diminished activity of the pyruvate dehydrogenase complex is caused by thia- mine pyrophosphate deficiency, possibly combined with inhibition by the ethanol metabolite, acetaldehyde, as a plausible common factor in neuropathogenesis. Hereditary spinocerebellar degeneration appearing in early adult life has been attributed to deficiency of pyruvate dehydrogenase, but there is no direct relationship to Friedreich’s ataxia. Mutations in other subunits E1β subunit  Very rare patients with mutations in the E1β subunit have early-​onset disease, delayed development, but moderate pro- gression of neurological disease, later with involvement of the basal ganglia and brainstem nuclei; patients with features of Leigh’s syn- drome has been reported. E1 phosphatase regulatory protein  Mutations in the E1 phos- phatase regulatory protein have been shown to cause hypotonia and feeding difficulties with psychomotor retardation, and at least one case with an acute neurological disease with lethal lactic acidosis in infancy has been described. Two mildly affected adult brothers of Turkish origin have been reported to be living in their twenties after treatment with a ketogenic diet. E2 (dihydrolipoamide acetyltransferase) deficiency  E2 (dihydrolipoamide acetyltransferase) deficiency is an exception- ally rare disease, the principal manifestations of which are dystonic

12.3.3  Disorders of galactose, pentose, and pyruvate metabolism 2011 episodes together with less prominent typical features of pyruvate de- hydrogenase deficiency, such as hypotonia and ataxia. Radiological examination reveals discrete lesions restricted to the globus pallidus. E3 (dihydrolipoamide dehydrogenase) deficiency  E3 (dihydrolipoamide dehydrogenase) deficiency is an autosomal re- cessive condition due to mutations in E3 dihydrolipoyl dehydro- genase, common to the action of the pyruvate dehydrogenase complex and the α-​ketodehydrogenase complex and the branched-​ chain α-​ketoacid dehydrogenase complexes. Clinical manifest- ations range from an acute disease in infants with fatal metabolic crises and decompensation associated with early-​onset neurological manifestations, to isolated liver disease appearing first in adoles- cence or adult life. Hepatic decompensation, heralded by nausea and vomiting, progresses rapidly to portosystemic encephalopathy which is accompanied by coagulation failure, hypoglycaemia, and bleeding. Liver failure can result in death, even in those with late-​ onset disease. The biochemical defect in dihydrolipoyl dehydrogenase defi- ciency interferes with the Krebs cycle as well as the decarboxylation of pyruvate. The diagnosis is confirmed by the presence of patho- genic mutations in the DLD gene, but recently it has been proposed that the appearance of citrullinaemia is an important biomarker. A variant presentation has recently been described with a late-​ onset mitochondrial myopathy and limited evidence of liver dis- ease. One well-​documented 19-​year-​old patient had lactic acidosis, a complex amino-​ and organic aciduria, and progressive exertional fatigue. Muscle biopsy showed mitochondrial proliferation and lack of cross-​reacting dihydrolipoamide dehydrogenase. Empirical ribo- flavin supplementation induced a complete resolution of exercise intolerance with the partial restoration of the protein and resolution of pathological mitochondrial proliferation in the muscle. Oral ad- ministration of lipoic acid has been reported to correct the organic acidaemia with clinical improvement in other patients. These find- ings prompt more systematic use of all the relevant vitamin cofactors in the pyruvate dehydrogenase complex, here illustrating a classical chaperone effect elsewhere familiar in adult metabolic practice (e.g. in attenuated pyridoxine-​responsive homocystinuria). E3BP (E3 binding protein)  More than 20 patients with E3BP de- ficiency have been reported. It appears generally to be less severe than pyruvate dehydrogenase deficiency due to mutations in the E1α subunit. A few have shown neonatal onset with lactic acidosis, the survivors of which had a clinical course similar to the late-​onset patients who suffer psychomotor retardation and encephalopathy—​ pyramidal spasticity, in some cases with microcephaly. Neonatal lactic acidosis is more frequent in males. For prenatal testing, diagnosis by molecular analysis of DNA is essential—​and is also valuable for the identification of the other defects such as E1β-​ subunit deficiency. An auxiliary gene for the E1α subunit is localized as a result of retroposition from the X chromosome to the long arm of chromo- some 4, but is expressed only during spermatogenesis; its presence, however, indicates the critical need for activity of the complex in nearly all tissues. Causal mutations in the PDHA1 gene on the X chromosome have been described; most appear to be short deletions or duplications and, at present, are not generally applicable for diag- nosis. However, analysis of X-​chromosome inactivation patterns, by determination of methylation status, has proved useful for the evaluation of enzymatic assays of fibroblasts obtained from obligate carriers or female patients in whom the diagnosis is suspected. Investigation and diagnosis The diagnosis is suspected from the presence of severe acidosis at birth. It may also emerge during the investigation of neurological deficits, especially where they are associated with intrauterine growth failure. Measurement of glucose, lactate, pyruvate, 3-​hydroxybutyrate, and acetoacetate in whole blood, as well as plasma amino acid concentrations, should be carried out. Hyperammonaemia with citrullinaemia, hyperlysinaemia, and hyperalaninaemia may be found. The NADH/​NAD+ ratio is informative when within the healthy reference range because use of NADH is unimpaired in pyruvate dehydrogenase deficiency, whereas in respiratory electron chain defects with defective complexes I, III and IV there is an ele- vated lactate/​pyruvate ratio and NADH is typically elevated. Routine screening of urine samples for organic acids may identify excessive pyruvate and lactate. Urine organic acid analysis requires the assistance of a specialized laboratory equipped for gas chro- matography, and mass spectrometry is increasingly used in major centres. Determination of lactate and pyruvate concentrations in cerebro- spinal fluid are of critical value and require special conditions for collection, transport, and storage before assay. In patients without clinically evident acidosis, cerebral disease is accompanied by striking elevations of lactate and pyruvate in the cerebrospinal fluid. Muscle biopsy for mitochondrial studies and determination of the redox state in cultured skin fibroblasts using the lactate:pyruvate ratio may also be valuable, but further specialized studies will re- quire advice from a biochemical and genetics service with experi- ence in the diagnosis of inborn errors of metabolism. Given that these measurements can show wide fluctuations in acutely ill pa- tients, several samples should be examined as recovery occurs so that the steady-​state abnormalities are reflected. Glucose challenges are not critical for diagnosis, but pyruvate rises markedly in pyruvate dehydrogenase deficiency. Neuroradiological imaging reveals ventricular dilatation and cerebral atrophy. In several infant girls with pyruvate dehydrogenase deficiency, MRI showed hypoplasia of the corpus callosum as well as loss of normal white matter signal intensity. Proton magnetic resonance spectroscopy revealed high-​abundance signals for brain lactate with decreased intensity of N-​acetylaspartate, while phos- phorus magnetic resonance spectroscopy of skeletal muscle showed abnormally low muscle phosphorylation potentials, in keeping with the predicted biochemical disturbance. Pathological examination of previously affected siblings shows shrinkage of gyri, with involve- ment of the medulla shown by loss or hypoplasia of the pyramids. The pathological features of Wernicke’s encephalopathy may be pre- sent. The corpus callosum may be absent. Definitive diagnosis depends on genetic and enzymatic studies in skin fibroblasts or blood leucocyte samples. This should in- clude indirect measurement of the activity of the whole complex by determining release of 14CO2 from [1-​14C pyruvate] from cul- tured cells in the presence or absence of high thiamine pyrophos- phate to explore vitamin responsiveness, and in the presence or absence of dichloroacetate which activates the enzyme in intact cells by inhibiting the regulatory E1 kinase. Mutation analysis of

SECTION 12  Metabolic disorders 2012 the X-​linked PDHA1 gene or other PDH-​related genes permits de- cisive diagnosis. While chorionic villus biopsy tissue or cultured amniocytes can be used for prenatal diagnosis, prior studies of ma- terial obtained from previously affected probands is often invaluable. Treatment Institution of a high-​fat, low-​carbohydrate, ketogenic diet may ameliorate the biochemical abnormalities, but—​given the degree of neurological impairment that is normally present at diagnosis—​ only very modest clinical improvement can be expected in those pa- tients with established disease. Therapeutic responses to the administration of high-​dose thia- mine (500 mg daily) have been reported in patients with partial enzymatic deficiency, notably where ataxia and abnormal eye move- ments reminiscent of Wernicke’s encephalopathy or features indica- tive of Leigh’s disease are conspicuous. Dichloroacetate, which is a structural analogue of pyruvate, is an inhibitor of the regulatory E1α-​subunit kinase and has been used for the treatment of primary lactic acidaemia, particularly in pa- tients with pyruvate dehydrogenase deficiency. Clinical trials in- dicate that correction of the biochemical abnormality depends on the molecular defect, and heterogeneity in patient selection may ex- plain the equivocal clinical responses observed in long-​term studies. Nonetheless, dichloroacetate appears to be well tolerated and de- serves consideration in patients who fail to respond to other meas- ures, including the recommended ketogenic diets with high-​dose thiamine supplementation. In patients with mutations in the multifunctional flavoprotein E3 dihydrolipoyl dehydrogenase, oral administration of lipoic acid has been reported to correct the organic acidaemia with clin- ical improvement. More striking has been the report of high-​dose riboflavin supplementation in a young adult with a mitochondrial myopathy and lactic acidosis producing salutary metabolic, histo- logical, and functional reversal, plausibly due to a chaperone effect of this critical enzyme cofactor. Given the severe nature of these diseases, and in the absence of clinical trial data, there may be justification for empirical clinical use of the vitamin cofactors in selected cases. In patients with seizures, the use of sodium valproate cannot be recommended by the author: the agent is an inhibitor of mitochon- drial metabolism and has been implicated in unmasking and aggra- vating several mitochondrial diseases. Other anticonvulsants affect mitochondrial metabolism, including carbamazepine, phenytoin, oxcarbazepine, ethosuximide, zonisamide, topiramate, gabapentin, and vigabatrin. Where possible in disorders of pyruvate-​driven oxidative phosphorylation, it would seem prudent to avoid these agents, but valproate probably should be avoided altogether in pa- tients with defective activity of the mitochondrial pyruvate dehydro- genase complex. Pyruvate carboxylase deficiency Inborn defects in pyruvate carboxylase, a biotin-​dependent gluconeogenic enzyme, cause hypoglycaemia or profound meta- bolic acidosis with neurodegenerative features. The neuronal loss is prominent, although the enzyme is principally expressed in astro- cytes and other non-​neuronal cells, suggesting impairment of the supply of nutrients derived from metabolic activity in astroglia that are essential for neuronal survival. The manifestations closely resemble those caused by deficiencies of pyruvate dehydrogenase activity and appear to be determined by the degree of residual pyru- vate carboxylase activity. Genetics This disorder is transmitted as an autosomal recessive trait. In se- verely affected patients with hyperammonaemia, pyruvate carb- oxylase protein and its mRNA are absent in the liver. A partially inactive variant enzyme is detectable in other patients. Biochemical defect Pyruvate decarboxylase is a biotin-​dependent enzyme of the mito- chondrial matrix which catalyses the first step in the formation of oxaloacetate from pyruvate and carbon dioxide and is activated allo- sterically by acetyl coenzyme A. It is critical enzyme for the produc- tion of glucose by gluconeogenesis: this is achieved by carboxylation of pyruvate to form oxaloacetate, which is shuttled to the cytosol where it is acted upon by phosphoenol pyruvate carboxykinase to generate the glucose precursor phophoenolpyruvate, which is the first committed step in de novo glucose formation. Thus, in pyru- vate carboxylase deficiency, hypoglycaemia would be expected after glycogen stores are depleted. Pyruvate carboxylase as a source of lipids is explained by its intramitochondrial proximity: acetyl coenzyme A condenses with pyruvate to generate citrate. Impaired synthesis of lipids explains the often widely distributed loss of white matter in pyruvate carboxylase deficiency. Krebs cycle intermediates may become depleted so that syn- thesis of neurotransmitters is impaired. There may also be a reduced supply of aspartate for the arginosuccinate synthase reaction of the urea cycle, hence the association with hyperammonaemia. Clinical features Three broad clinical types of pyruvate carboxylase have been recognized. Type A (infantile form) The North American form of the disease is associated the onset of vomiting, metabolic acidosis (lactate is 2–​10 mmol/​litre), and collapse in infants aged 2 to 6 months and associated with intercur- rent infection. The patients develop ataxia, pyramidal tract signs, and nystagmus:  severe mental retardation and seizures develop rapidly. An enlarged liver is present and neuroradiological imaging shows subdural fluid, lesions resembling antenatal ischaemia-​like brain lesions, and periventricular haemorrhagic cysts accom- panied by cortical atrophy. Myelination is retarded and the patient relentlessly deteriorates to die, almost always in infancy or early childhood. Type B (severe, neonatal form) The so-​called French form, with severe prostration within the first 48 h of life. There is vomiting, hypotonia, lethargy, hypothermia, and rapid neurological deterioration with tremor, rigidity, poor move- ment, and abnormal ocular movements. The disease is rapidly fatal in most cases; those who survive the early days are unresponsive and die from respiratory infection before the age of 6 months. There is a marked lactate acidosis with concentrations of 10 to 20 mmol/​litre (normal is <2.2 mmol/​litre).

12.3.3  Disorders of galactose, pentose, and pyruvate metabolism 2013 Type C (intermittent/​benign form) A rare third subtype is compatible with survival to adult life with episodic lactic acidosis and ketosis (<10  mmol/​litre), these epi- sodes resolve with supportive measures and parenteral fluids. While subcortical leukodystrophy has developed in some patients, many develop normally and have cognitive ability and motor-​skill devel- opment that is within the healthy reference range. Diagnosis The condition is suspected when acidosis and neurological dis- ease occur in infants, especially in the presence of hypoglycaemia. Specific diagnosis requires enzymatic assay in fibroblasts, which can also be used for carrier detection. The residual activities appear to correspond to the clinical phenotype approximately. The diagnosis can be confirmed by molecular analysis of genomic DNA for the po- tential use in case of the need for assisted reproduction when the PC gene data are hypothecated to the mother. Treatment Episodes of acidosis are treated with intravenous sodium bicar- bonate, and glucose may be required for hypoglycaemia. There is evidence that ketogenic diets containing 50% fat and 20% carbohy- drate ameliorate the biochemical disturbance and delay the onset of neurological disease. The administration of high-​dose glutamate and aspartate, which may act as a source of oxaloacetate, appear to have been beneficial in some patients, at least on the composition of the plasma amino acids. Use of the anaplerotic seven-​carbon triheptanoin has been de- scribed, with reports of benefit in some cases, but not all. Although biotin therapy has been disappointing in pyruvate carboxylase defi- ciency, occasional responses to high-​dose lipoic acid and thiamine treatment, which may stimulate pyruvate metabolism by the de- hydrogenase complex, have been recorded. However, a collabora- tive effort to investigate their effects in selected patient groups under stratified conditions is needed. Experimental hepatic allotransplantation has been carried out in patients with pyruvate carboxylase deficiency, with salutary effects on plasma amino acids apart from glutamine, but the effect on brain function was not easy to determine in the first patient treated. FURTHER READING Inborn errors of galactose metabolism Berry GT (2000). Classic galactosemia and clinical variant galactosemia. In: Adam MP, et al. (eds). GeneReviews®. University of Washington, Seattle. https://​www.ncbi.nlm.nih.gov/​books/​NBK1518/​ Berry GT, Walter JH (2012). Disorders of galactose metabolism. In: Saudubray J-​M, van den Berghe G, Walter JH (eds) Inborn meta- bolic diseases, 5th edition, pp. 141–​50. Springer-​Verlag, Berlin. Bosch AM, et al. (2004). Living with classical galactosemia: health-​ related quality of life consequences. Pediatrics, 113, e423–​8. Coelho AI, et  al. (2017). Sweet and sour:  an update on classic galactosemia. J Inherit Metab Dis, 40, 325–​42. Coman DJ et al. (2009). Galactosemia, a single gene disorder with epi- genetic consequences. Pediatr Res, 67, 286–​92. Cornblath M, Schwartz R (1991). Disorders of galactose: metabolism. In: Cornblath M, Schwartz R (eds) Disorders of carbohydrate me- tabolism in infancy, 3rd edition, pp. 295–​324. Blackwell Scientific, Boston. Demirbas D, et al. (2018). Hereditary galactosemia. Metabolism, 83, 188–​96. Frederick AB, Cutler DJ, Fridovich-​Keil JL (2017). Rigor of non-​dairy galactose restriction in early childhood, measured by retrospective survey, does not associate with severity of five long-​term outcomes quantified in 231 children and adults with classic galactosemia.
J Inherit Metab Dis, 40, 813–​21. Frey P (1996). The Leloir pathway: a mechanistic imperative for three enzymes to change the stereochemical configuration of a single carbon in galactose FASEB J, 10, 461–​70. Fridovich-​Keil JH, Walter JH (2001). Galactosemia. In: Scriver CR, et  al. (eds) Metabolic and molecular bases of inherited disease, 8th edition, p. 1553–​87. McGraw-​Hill, New  York. http://​www. ommbid.com. Holton JB, et al. (1981). Galactosaemia. A new severe variant due to uridine diphosphate galactose-​4-​epimerase deficiency. Arch Dis Child, 56, 885–​7. Lak R, Yazdizadeh B, Davari M, et al. (2017). Newborn screening for galactosaemia. Cochrane Database Syst Rev, 12, CD012272. Murphy M, et  al. (1999). Genetic basis of transferase-​deficient galactosaemia in Ireland and the population history of Irish Travellers. Eur J Hum Genet, 7, 549–​54. Ridel KR, Leslie ND, Gilbert DL (2005). An updated review of the long-​term neurological effects of galactosemia. Pediatr Neurol, 33, 153–​61. Robinson BH, et al. (1996). Disorders of pyruvate carboxylase and pyruvate dehydrogenase complex. J Inherit Metab Dis, 19, 452–​62. Robinson BH (2001). Lactic acidemia:  disorders of pyruvate carb- oxylase and pyruvate dehydrogenase. In:  Scriver CR, et  al.
(eds) Metabolic and molecular bases of inherited disease, 8th edition, pp. 2275–​84. McGraw-​Hill, New York. http://​www.ommbid.com. Rubio-​Agusti I, et al. (2013) Movement disorders in adult patients with classical galactosemia. Mov Disord, 28, 804–​10. Rubio-​Gozalbo ME, et al. (2006). The endocrine system in treated
patients with classical galactosemia. Mol Genet Metab, 89, 316–​22. Schweitzer S, et al. (1993). Long-​term outcome in 134 patients with galactosaemia. Eur J Paediatr, 152, 36–​43. Tyfield L (2000). Galactosaemia and allelic variation at the galactose-​1-​ phosphate uridyltransferase gene. A complex relationship between genotype and phenotype. Eur J Pediatr, 159, S204–​7. Tyfield L, et al. (1999). Classical galactosemia and mutations at the galactose-​1-​phosphate uridyl transferase (GALT) gene. Hum Mutat, 13, 417–​30. Van Calcar SC, et al. (2014). A re-​evaluation of life-​long severe galactose restriction for the nutrition management of classic galactosemia. Mol Genet Metab, 112, 191–​7. Waggoner DD, Buist NRM, Donnell GN (1990). Long-​term prognosis in galactosaemia: results of a survey of 350 cases. J Inherit Metab Dis, 13, 802–​18. Waisbren SE, et al. (2012). The adult galactosemic phenotype. J Inherit Metab Dis, 35, 279–​86. Welling L, et al. (2017). International clinical guideline for the manage- ment of classical galactosemia: diagnosis, treatment, and follow-​up. J Inherit Metab Dis, 40, 171–​6.

SECTION 12  Metabolic disorders 2014 Pentosuria Hiatt HH (2001). Pentosuria. In: Scriver CR, et al. (eds) Metabolic and molecular bases of inherited disease, 8th edition, pp. 1590–​9. McGraw-​ Hill, New York. http://​www.ommbid.com. Inborn errors of pyruvate metabolism Brassier A, et  al. (2013). Dihydrolipoamide dehydrogenase defi- ciency: a still overlooked cause of recurrent acute liver failure and Reye-​like syndrome. Mol Genet Metab, 109, 28–​32. Brown GK, et al. (1994). Pyruvate dehydrogenase deficiency. J Med Genet, 31, 875–​9. Brown RM, et al. (2006). Pyruvate dehydrogenase E3 binding protein (protein X) deficiency. Dev Med Child Neurol, 48, 756–​60. Carrozzo R, et al. (2014). Riboflavin responsive mitochondrial myopathy is a new phenotype of dihydrolipoamide dehydrogenase deficiency. The chaperon-​like effect of vitamin B2. Mitochondrion, 18, 49–​57. Dahl H-​M, et al. (1992). X-​linked pyruvate dehydrogenase E1-​alpha subunit deficiency in heterozygous females: variable manifestation of the same. J Inherit Metab Dis, 15, 835–​47. DeBrosse SD, et al. (2012). Spectrum of neurological and survival out- comes in pyruvate dehydrogenase complex (PDC) deficiency: lack of correlation with genotype. Mol Genet Metab, 107, 394–​402. Head RA, et al. (2005). Clinical and genetic spectrum of pyruvate de- hydrogenase deficiency:  dihydrolipoamide acetyltransferase (E2) deficiency. Ann Neurol, 58, 234–​41. Hinman LM, et al. (1989). Deficiency of pyruvate dehydrogenase com- plex in Leigh’s disease fibroblasts: an abnormality in lipoamide de- hydrogenase affecting PDHC activation. Neurology, 39, 70–​5. Liu YM, et al. (2003). A prospective study of growth and nutritional status in children treated with the ketogenic diet. J Am Diet Assoc, 103, 707–​12. Lissens W, et  al. (2000). Mutations in the X-​linked pyruvate ­dehydrogenase (E1) alpha subunit gene (PDHA1) in patients with a pyruvate dehydrogenase complex deficiency. Hum Mutat,
15, 209–​19. McWilliam CA, et al. (2010). Pyruvate dehydrogenase E2 deficiency: a potentially treatable cause of episodic dystonia. Eur J Paediatr Neurol, 14, 349–​53. Mellick G, Price L, Boyle R (2004). Late-​onset presentation of pyruvate dehydrogenase deficiency. Mov Disord, 19, 727–​9. Quinonez SC, et al. (2014). Newborn screening for dihydrolipoamide dehydrogenase deficiency: citrulline as a useful analyte. Mol Genet Metab Rep, 1, 345–​49. Robinson BH (2001). Lactic acidemia:  disorders of pyruvate ­carboxylase and pyruvate dehydrogenase. In:  Scriver C, et  al.
(eds) The metabolic and molecular bases of inherited disease,
8th edition, pp. 2275–​84. McGraw-​Hill, New York. http://​www. ommbid.com. Robinson BH (2006). Lactic acidemia and mitochondrial disease. Mol Genet Metab, 89, 3–​13. Shevell MI, et al. (1994). Cerebral dysgenesis and lactic acidemia: an MRI/​MRS phenotype associated with pyruvate dehydrogenase defi- ciency. Pediatr Neurol, 11, 224–​9. Stacpoole PW, et al. (2008). Evaluation of long-​term treatment of chil- dren with congenital lactic acidosis with dichloroacetate. Pediatrics, 121, e1223–​8. Wang D, De Vivo D. (2009). Pyruvate carboxylase deficiency. In: Adam MP, et  al. (eds) GeneReviews®. University of Washington, Seattle. https://​www.ncbi.nlm.nih.gov/​pubmed/​20301764 Wexler ID, et al. (1997). Outcome of pyruvate dehydrogenase defi- ciency treated with ketogenic diets. Studies in patients with identical mutations. Neurology, 49, 1655–​61.

12.4 Disorders of purine and pyrimidine metabolism

12.4 Disorders of purine and pyrimidine metabolism 2015

ESSENTIALS These disorders are due to abnormalities in the biosynthesis, inter- conversion, and degradation of the purines—​adenine and guanine—​ and of the pyrimidines—​cytosine, thymine, and uracil. All are heterocyclic bases which exist in tri-​, di-​, and monophosphorylated forms, and as either deoxyribosylated or ribosylated derivatives (de- oxyribose and ribose are pentose carbohydrates). The phosphor- ylated deoxyribosylated and ribosylated derivatives are termed ‘nucleotides’, and the purely ribosylated derivatives, which lack the phosphate group, are ‘nucleosides’. The purine nucleotides, their cyclic derivatives (cAMP and cGMP), and their more highly phosphorylated derivatives have functions in many aspects of intermediary metabolism. Purine compounds also function as signal transducers, neurotransmitters, vasodilators, and mediators of platelet aggregation. The polynucleotide deoxyribonucleic acid (DNA) contains equimolar amounts of adenosine monophosphate (adenylic acid, AMP), guanosine monophosphate (guanylic acid, GMP), thymidine monophosphate (thymidylic acid, TMP), and cytidine monophosphate (cytidylic acid, CMP). Uridine monophosphate (uridylic acid, UMP) re- places TMP in the polynucleotide ribonucleic acid (RNA). Disorders of purine metabolism The end point of purine metabolism in humans is uric acid. When uric acid levels become supersaturated in body fluids, uric acid and sodium urate monohydrate crystallize, causing gout. This re- sults from either overproduction or underexcretion of urate, or from a combination of these defects. Decreased net tubular urate secretion is most often due to genetic polymorphism in uric acid transporters and is the most common cause of primary (‘idiopathic’) gout. Gout may be secondary to a wide variety of renal disorders, ranging from simple reduction in glomerular filtration rate (chronic kidney disease) to specific defects, for example, the autosomal dom- inant tubulointerstitial kidney diseases caused by mutations in the genes encoding uromodulin (UMOD), hepatocyte nuclear factor-​1b (HNF1B), renin (REN), and mucin-​1 (MUC1). Gout is also a consequence of enzymatic defects that accel- erate de novo purine synthesis. X-​linked hypoxanthine-​guanine phosphoribosyltransferase deficiency results in a clinical spectrum extending from hyperuricaemia alone to hyperuricaemia with profound neurological and behavioural dysfunction (Lesch–​Nyhan syndrome). Phosphoribosyl pyrophosphate synthetase superactivity presents with uric acid lithiasis or gouty arthritis in childhood or early adult life. Acute attacks of gout are treated with nonsteroidal anti-​ inflammatory drugs, colchicine, or steroids. First-​line treatment to prevent acute attacks or manage chronic tophaceous gout is with the xanthine oxidase inhibitor, allopurinol. Hypouricaemia may be caused by inherited disorders of uric acid biosynthesis (e.g. xanthine oxidase deficiency) or may be due to in- herited or acquired renal tubule transport defects. Other diseases of purine metabolism cause diverse abnormalities and are generally the result of single gene defects, for example, ad- enosine deaminase and purine nucleoside phosphorylase catalyse sequential steps in the metabolism of purine ribonucleosides and deoxyribonucleosides and are highly expressed in lymphoid cells; their deficiency causes lymphotoxic metabolites to accumulate and leads to lymphopenia and severe combined immunodeficiency. Disorders of pyrimidine metabolism The de novo synthesis of pyrimidine nucleotides involves a series of six reactions beginning with the formation of carbamyl phosphate and concluding with orotidine monophosphate, which then under- goes a series of interconversion and salvage reactions. The inherited disorders of pyrimidine metabolism (e.g. orotic aciduria), which have diverse presentations, are much less common and/​or much less easily recognized than disorders of purine metabolism. Disorders of purine metabolism Purine metabolism Biosynthesis, interconversion, degradation, and salvage The purine nucleotides are built up in a stepwise manner (de novo synthesis) and undergo a series of interconversion and salvage re- actions and a final degradative process to yield uric acid, as shown in Fig. 12.4.1. The dietary intake of nucleoproteins contributes to uric acid formation. Ingested adenine and guanine nucleotides are degraded to free purine bases and, hence, to uric acid by enzymes 12.4 Disorders of purine and pyrimidine metabolism Anthony M. Marinaki, Lynette D. Fairbanks, and Richard W.E. Watts† † It is with great regret that we report that Richard W.E. Watts died on 11 February, 2018.

SECTION 12  Metabolic disorders 2016 in the intestinal fluids and in the mucosa of the small intestine, so that the products of their metabolism do not mix with the cor- responding endogenous metabolic pools except at the final uric acid stage. De novo synthesis contributes about 300 to 600 mg (1.8–​3.6 mmol/​ day) and dietary purines about 600 to 700 mg (3.6–​4.2 mmol/​day) to the dynamic urate metabolic pool of about 1200 mg (7.2 mmol) expressed as uric acid. Each day about two-​thirds of the uric acid is excreted in the urine and about one-​third is excreted via the gut where it is destroyed mainly by bacterial uricolysis. Renal handling of urate The urate anion is freely filterable at the renal glomerulus, only 5 to 10% being very loosely bound to the plasma proteins (α1–​2-​ globulin fraction). The physiologically important pKa value of uric acid is 5.75, so that it exists mainly as the monovalent urate anion in plasma (pH 7.4) and assumes more of the free acid form when it passes into regions of the renal tubule, the contents of which are at lower pH values. The kidney handles urate by: • glomerular filtration (virtually no hindrance to passage through the glomerular filtration barrier) • proximal tubular reabsorption by urate anion exchangers, pre- dominantly URAT1, OAT4, and OAT10 in the endothelial brush border (99% of the filtered load) • reabsorption into the circulation at the basolateral membrane, predominantly by the fructose–​urate transporter GLUT9 (SLC2A9) • urate excretion at the apical membrane by MRP4, NPT1, NPT4, and BCRP/​ABCG2. The net renal clearance of uric acid is approximately 10% of the filtered load and is in the range of 6 to 11 ml/​min per 1.73m2 (1.73 m2 = average body surface area of an adult). Genome-​wide association studies have led to considerable ad- vances in the understanding of this complex process by identifying genes in uric acid transport and genetic variants predisposing to gout (Fig. 12.4.2). PRPS ADSL DNA dATP dADP dAMP AMP sAMP GAR ATP HCO3 H2O Aspartate ATP Glutamine ATP Fumarate Fumarate FGAR FGAM AIR CAIR SAICAR AICAR FAICAR IMP ADP ATP RNA Polyamine metabolism Adenine Deoxyadenosine Adenosine Deoxyinosine Inosine Deoxyguanosine Guanosine GTP GDP GMP XMP dGMP dGDP dGTP DNA RNA 8-OH Adenine 2-8-OH Adenine Uric acid Xanthine Hypoxanthine Guanine ADSL NT APRT ADA PNP PNP HPRT NT ATIC XO XO Guanase Ribose-5-Phosphate +ATP Phospho-D-ribosyl -1- pyrophosphate + glutamine 5-Phosphoribosyl-1-amine + Glycine-ATP Fig. 12.4.1  Pathways of purine metabolism in humans. ADA, adenosine deaminase; ADSL, adenylosuccinate lyase; APRT adenine phosphoribosyltransferase; ATIC, AICAR transformylase/​IMP cyclohydrolase; HPRT, hypoxanthine-​guanine phosphoribosyltransferase; NT, 5′-​nucleotidase; PNP, purine nucleoside phosphorylase; PRPS, phosphoribosylpyrophosphate synthetase; XO, xanthine oxidase. De novo synthesis intermediates: AICAR, 5-​phosphoribosyl-​5-​amino-​4-​imidazolecarboxamide; AIR, 5-​phosphoribosyl-​aminoimidazole; CAIR,
5-​phosphoribosyl-​5-​amino-​4-​imidazolecarboxylate; FAICAR, 5-​phosphoribosyl-​5-​formamido-​4-​imidazolecarboxamide; FGAM,
5-​phosphoribosyl-​N-​formylglycinamidine; FGAR, 5-​phosphoribosyl-​N-​formylglycinamide; GAR, 5-​phosphoribosyl-​glycinamide; SAICAR, 5-​phosphoribosyl-​5-​amino-​4-​imidazole-​succinocarboxamide.

12.4  Disorders of purine and pyrimidine metabolism 2017 URAT1, encoded by the gene SLC22A12, was the first uric acid transporter to be identified and is involved in urate reabsorption in the proximal tubule. It is expressed apically in the brush border epithelium and is an anion exchanger stimulated by an outwardly directed chloride gradient. Lactate, pyrazinoate, and nicotinate are substrates for the antiporter activity of URAT1 thereby increasing urate reabsorption. The uricosuric agents benzbromarone, pro- benecid, and losartan are inhibitors. OAT4 (SLC22A11) and OAT10 (SLC22A13) are also expressed in epithelial cells of the proximal tu- bule apical membrane and play a role in the reabsorption of uric acid from the proximal tubule in exchange for dicarboxylates. The multidrug resistance protein MRP4 and the transporter ABCG2 contribute to uric acid efflux into the tubular lumen. ABCG2 is also expressed in the intestine and plays an important role in the extrarenal excretion of uric acid through the gut. An amino acid substitution p.Q141K in ABCG2 is associated with significantly increased plasma uric acid levels and an increased risk for gout in multiple ethnic backgrounds. The data suggests that at least 10% of gout cases in individuals of European descent are attributable to this variant. The p.Q141K variant is unexpectedly also associated with increased urinary uric acid output because of the decreased intestinal excretion associated with the variant. More recently, the ABCG2 p.Q141K variant has been associated with a poor response to allopurinol therapy, although the mechanism for this is unclear. The urate fructose–​glucose transporter GLUT9 (SLC2A9) is a voltage-​dependent uric acid transporter. Alternative splicing leads to two transcripts, GLUT9a encoded by 12 exons with 540 amino acids and GLUT9b, a shorter protein of 512 amino acids encoded by 13 exons. GLUT9b expression is restricted to liver and kidney, whereas GLUT9a has a broad tissue distribution including liver, kidney, intestine, leucocytes, and interestingly, chondrocytes. It is suggested that the functions of this urate anion transporter is to transport urate formed intracellularly by purine catabolism and so maintain intracellular urate concentrations below the solubility limit and prevent intracellular crystallization. In the kidney, GLUT9 may be the principal pathway of basolateral urate transport from the proximal tubule cell. Plasma urate concentration Reference range The currently quoted overall reference range for plasma uric acid in adults is 3.5 to 8.1 mg/​dl (210–​480 µmol/​litre) for men and 2.5 to 6.5 mg/​dl (150–​390 µmol/​litre) for women. The corresponding value for children is 1.0 to 4.0 mg/​dl (60–​240 µmol/​litre), with the lowest values in infancy. It rises to adult values at puberty with values being lower in women than in men. Extrinsic factors, particularly diet, plumbism, the prevalence of a high ethanol intake in the com- munity, and the prevalence of diseases such as malaria and thalas- saemia, which lead indirectly to either increased purine biosynthesis or decreased excretion (Table 12.4.1), affect the plasma urate distri- bution in different populations. The plasma urate concentration decreases during pregnancy, the reference range being 1.7 to 4.5 mg/​dl (100–​270 µmol/​litre). Hyperuricaemia is a characteristic and often an early feature of pre-​eclampsia, preceding proteinuria and hypertension, and is a Basolateral membrane Tubular cell GLUT9 Urate Urate Urate Urate Urate Urate URA11 OAT4 ABCG2 MRP4 NPT1 and NPT4 OAT10 OAT1 OAT3 Apical membrane (brush border) Approximately 90% of filtered uric acid is reabsorbed Circulation Approximately 10% of filtered uric acid is excreted Tubular lumen Fig. 12.4.2  Proximal tubule uric acid anion transport. OAT1 and OAT3 mediate urate uptake from the basolateral membrane. Secretion at the apical membrane into the tubule is via ABCG2, MRP4, NPT1, and NPT4. Tubular reabsorption is via URAT1, OAT4, and OAT10. GLUT9 promotes reabsorption of urate back into the circulation. Approximately 90% of urate filtered in the glomerulus is reabsorbed in the proximal tubule. This increases to approximately 95% in patients with hyperuricaemia due to uric acid underexcretion.

SECTION 12  Metabolic disorders 2018 diagnostically valuable parameter. It results from a reduced renal urate clearance and tends to be associated with hypocalciuria. Epidemiological studies show significant variations in plasma urate concentrations between different ethnic groups, for example, Maoris and Polynesians have higher values than Western Europeans and Americans. This illustrates the genetic, presumably polygenic, aspects in the control of serum uric acid. Other epidemiological studies em- phasize the importance of the environmental factors of purine, pro- tein, and alcohol intake. For example, Gresser and Zöllner showed that the cumulative frequency of plasma urate, expressed as uric acid, rose from approximately 6.2 mg/​dl (370 µmol/​litre) to about 9.0 mg/​ dl (536 µmol/​litre) between 1962 and 1971 in association with the im- proved nutritional state of the Bavarian population from the near star- vation conditions following the Second World War (Fig. 12.4.3). This effect was not apparent in the female population. Similarly, the plasma urate levels of immigrant communities with low values in their home- lands rise towards the values prevailing in the host country as they adopt the lifestyle and dietary habits of that country (e.g. Filipinos migrating to the United States of America). Migrants with genetically determined high urate levels become even more hyperuricaemic. The frequency distribution of plasma urate values based on asymptomatic populations is only approximately Gaussian, with an excess of higher values due to the inclusion of some asymptom- atic hyperuricaemic subjects. Although plasma is saturated with monosodium urate at a concentration of 7.0 mg/​dl (420 µmol/​litre), higher concentrations of urate can remain in a stable supersaturated solution in plasma without producing any symptoms. Ignoring the slight asymmetry of the frequency distribution and defining nor- mality as the mean value ±2 standard deviations about the mean, normal values of 7.0 mg/​dl (420 µmol/​litre) for men and 6.0 mg/​dl (360 µmol/​litre) for women have been widely adopted and this has led to considerable overtreatment of patients with quite innocuous plasma urate concentrations. Asymptomatic hyperuricaemia: to treat or not to treat? Routine biochemical screening frequently identifies patients with hyperuricaemia. The treatment of asymptomatic hyperuricaemia with urate-​lowering therapies carries a significant risk of toxicity and in the absence of evidence for clear clinical benefit, recommenda- tions are that asymptomatic hyperuricaemia should not be treated. The evidence for a causal association between hyperuricaemia and hypertension, cardiovascular disease, metabolic syndrome, and an increased risk of renal failure is contradictory. If indeed causal, any association is likely to have a small effect of limited clinical significance compared to other comorbidities. A few studies have compared measures of renal function in patients with asymptom- atic hyperuricaemia treated with allopurinol to those in patients not treated for hyperuricaemia, and although there is some evidence for an improvement in estimated GFR with treatment, these small studies have had a duration of a few months rather than years. There is also little evidence that asymptomatic hyperuricaemia increases the risk of gouty arthritis. However, although there is currently insufficient evidence to suggest that treatment of asymptomatic hyperuricaemia is beneficial, lifestyle advice on diet and exercise should be given to the patient and may be of benefit in lowering uric acid levels as well as improving other comorbidities. Gout (For further discussion see Chapter 19.10.) The incidence of gout has been estimated at about 0.2 to 0.35 per 1000. The incidence in- creases with age and is higher in men than in women, although the incidence in women rises with age. In men the first attack has usually occurred by 50 years of age and in women by 70 years of age. Table 12.4.1  Causes of hyperuricaemia and gout Increased uric acid synthesis Decreased renal clearance Increased uric acid production due to HPRT deficiency or PRPS synthase superactivity Genetic polymorphism in urate transporters URAT1, GLUT9,
and ABCG2 Dietary sources including fructose Chronic kidney disease Chronic haemolytic anaemia of
any cause Autosomal dominant tubulointerstitial kidney diseases Myeloproliferative disorders Hyperparathyroidism Malignancies Bartter’s syndrome Psoriasis Hypothyroidism Obesity Obesity Down’s syndrome Diabetes Glycogen storage diseases Lead poisoning Fructose intolerance Lactic acidosis Gaucher’s disease Drug administration (e.g. diuretics, pyrazinamide, and ethambutol) Cumulated frequency (%) 100 80 60 40 20 Serum uric acid (mg/dl) Men Cumulated frequency (%) 100 80 60 40 20 Serum uric acid (mg/dl) Women 1962 1989 1971 1962 1989 1971 12 0 2 4 6 8 10 12 0 2 4 6 8 10 Fig. 12.4.3  Differences in the cumulative frequencies in urate levels in female and male blood donors in Bavaria between 1962 and 1989. Urate deposition in man and its clinical consequences. Reproduced from Gresser U, Zöllner N (1991). Urate deposition in man and its clinical consequences. With kind permission of Springer Science + Business Media.

12.4  Disorders of purine and pyrimidine metabolism 2019 Gout is a classic example of a multifactorial disease in which there is an interplay of genetic and environmental factors. The overall ef- fects of this interplay are wide, extending from cases where there is a clear-​cut family history with autosomal dominant inheritance (Fig. 12.4.4) to those where environmental factors may be major de- terminants, although often against a genetic background that may be either unifactorial or multifactorial. Gout per se does not shorten life, although some of its complications may do so in the absence of treatment. Gout is defined as the syndrome brought about by the crystalliza- tion of monosodium urate monohydrate in vivo from body fluids supersaturated with this salt. This results from either overproduc- tion or underexcretion of urate, or from a combination of these de- fects (Table 12.4.1). The underlying causes of hyperuricaemia and gout are as follows: • Decreased net tubular urate secretion: this occurs in those cases of gout previously described as being idiopathic (or primary), and the hereditary predisposition is often compounded by environ- mental factors (e.g. high dietary purine and alcohol intake). • Identifiable enzymatic defects that accelerate de novo urate syn- thesis: X-​linked hypoxanthine-​guanine phosphoribosyltransferase (HPRT) deficiency, if complete or virtually complete, causes Lesch–​Nyhan syndrome. Lesser degrees of deficiency cause X-​ linked recessive hyperuricaemia, gout, and uric acid stones with minor neurological abnormalities in some cases. • Phosphoribosylpyrophosphate (PRPP) synthetase superactivity: this also presents as X-​linked recessive hyperuricaemia, gout, and uric acid stones and, in some cases, neurological manifestations (e.g. deafness). Gout due to urate under-​excretion is characterized by a reduced fractional excretion of urate defined as the ratio of urate clearance to the GFR. In the presence of normal overall renal function, this can be measured on a timed urine sample with a simultaneous plasma sample. The equation simplifies to: Fractional clearance of urate = U [urate] × P [creatinine]/​P [urate]
× U [creatinine] where U and P represent urate and plasma concentrations. The fractional clearance can be used to assess the role of renal tubular dysfunction in the production of hyperuricaemia provided that the overall renal function is normal. An elevated urinary urate to cre- atinine ratio is indicative of purine overproduction. Associations of hyperuricaemia and gout are shown in Box 12.4.1. Hyperuricaemia may be a marker of coincident cardiac disease. Elevated plasma uric acid concentrations are observed in patients with ischaemic cardiac disease, however there is no evidence that uric acid is directly toxic to the myocardium. Increased urate levels could arise from up-​regulated vascular adenosine synthesis associ- ated with ischaemia and subsequent degradation of adenosine to uric acid. Plasma uric acid accounts for 60% of the free-​radical scavenging activity in human plasma, for example, it interacts with peroxynitrile to form a stable nitric oxide donor, so promoting vasodilatation and reducing the potential for peroxynitrile-​induced oxidative damage. Conversely, it could have an adverse effect on endothelial function by promoting leucocyte adhesion to the endothelium. Clinical features Acute gouty arthritis Acute gout is a sodium urate monohydrate-​induced crystal in- flammation of joints, bursae, and tendon sheaths. Clinically the affected structures—​classically the first metatarsophalangeal joint is the first joint affected—​become acutely inflamed, exquisitely tender, warm to the touch, and the overlying skin becomes red, shiny, and itchy and may desquamate as the inflammation subsides spontaneously over the course of 5 to 15 days in the absence of treatment (Fig. 12.4.5). Inflammation is usually maximal within 24 h of onset and is accompanied by pyrexia and malaise. The American College of Rheumatology criteria for the clinical diagnosis of acute gout are shown in Box 12.4.2. The presence of 6 I II III IV V VI 3 2 1 7 3 2 1 1 2 1 2 3 4 5 4 5 6 4 5 6 6 7 8 9 8 9 2 1 * 3 * Fig. 12.4.4  Pedigree chart of a family showing autosomal dominant inheritance of gout complicated in some cases by renal failure (autosomal dominant tubulointerstitial kidney disease). ■, ●, male and female subjects, respectively, with hyperuricaemia and renal failure; ■, ●, male and female subjects not known to be affected; ■, ●, deceased male and female subjects; ↗, propositus; ↘, subjects whose rates of mononuclear cell de novo purine synthesis were measured and shown to be normal; *, babies who were examined clinically but not further investigated. Reproduced with permission from McDermott, et al. (1984). Clin Sci, 67, 249–​58.
© Biochemical Society and Medical Research Society (http://​www.clinsci.org). Box 12.4.1  Associations of hyperuricaemia and gout The following abnormalities (features of the ‘metabolic syndrome’) are commonly associated with, but not causally related to, hyperuricaemia and gout: • Obesity • Dyslipidaemia (usually type 4) with raised very low-​density lipopro- teins and normal cholesterol levels, and sometimes hypercholester- olaemia with elevated low-​density lipoprotein cholesterol and low high-​density lipoprotein cholesterol • Hypertension • Insulin resistance with hyperinsulinaemia and impaired glucose tolerance • Ischaemic heart disease Thus, these patients may display the features of the ‘metabolic syndrome X’.

SECTION 12  Metabolic disorders 2020 of the 11 criteria has a 95% specificity in differentiating gout from pseudogout (calcium pyrophosphate gout) and an overall sensi- tivity of 85%. Although acute gouty arthritis is typically a monoarthritis, some patients have short, recurrent, mild attacks of discomfort and swelling of other affected joints. Some 10% of attacks affect more than one joint, and typical attacks may provoke migratory attacks in other joints. Multiple, simultaneous attacks are rare. Some attacks are triggered by trauma, intercurrent illness, surgery, alcohol, dietary ex- cess, diuretics, and other medications (Box 12.4.3). An acute septic arthritis is the most important differential diagnosis of acute gouty arthritis. The joint fluid contains negatively birefringent sodium urate monohydrate (Fig. 12.4.6) as opposed to the positively birefringent crystals of calcium pyrophosphate in pseudogout and is diagnostic. Attacks of acute gouty arthritis usually occur when the plasma urate is rising or falling. The cells in the joint fluid are a mixture of mono- cytes, macrophages, and polymorphonuclear leucocytes. The spontaneous resolution of an attack of acute gouty arthritis depends on the differentiation of monocytes to macrophages that ef- ficiently phagocytose crystals. This conclusion is based on studies of the changing pattern of proinflammatory cytokines tumour necrosis factor (TNF), interleukin-​1 (IL-​1), and interleukin-​6 (IL-​6) secreted by monocyte/​macrophage cells at different degrees of differentiation and their ability to phagocytose monosodium urate crystals in vitro. The crystals are removed by mature phagocytes. It is proposed that this mechanism prevents the development of acute gouty arthritis in stable asymptomatic hyperuricaemic patients. TNFα, IL-​1β, and IL-​6 secretion promote E-​selectin expression and secondary neutro- phil capture. Differentiation over 3 to 5 days leads to development of a noninflammatory phenotype with lack of proinflammatory cyto- kine secretion, lack of endothelial cell activation, and lack of sec- ondary neutrophil recruitment. Acquisition of the noninflammatory phenotype correlates with expression of macrophage antigens but not with dendritic cell marker or activation marker. Monocytes and macrophages are similarly phagocytic and control particle, zymosan-​elicited secretion of all the cytokines in both cell types. Coincubation with monosodium urate suppressed zymosan-​ induced TNFα secretion from macrophages but not monocytes. In summary, differentiated macrophages provide the mechanism for removal of sodium monohydrate crystals. Fig. 12.4.5  Gout presenting in the metatarsophalangeal joint of the big toe. Note the slight redness of the skin overlying the joint (arrow). Image by James Heilman, MD ((CC BY-SA 3.0 (http://creativecommons.org/licenses/ by-sa/3.0)). Box 12.4.2  American College of Rheumatology criteria for the diagnosis of acute gouty arthritis • More than one attack of acute arthritis • Maximum inflammation developing within 1 day • Monoarthritis • Redness over the affected joint • The first metatarsophalangeal joint is painful and swollen • Unilateral first metatarsophalangeal joint involved • Unilateral tarsal joint attack • Tophus (proven or suspected) • Hyperuricaemia • Asymmetrical swelling of a joint on radiography • Subcortical cysts with an erosion on radiography • Joint fluid culture negative for microorganisms during an attack The patient must have at least six of the listed criteria or have either proven sodium urate monohydrate crystals in the joint fluid or a proven tophus. Reproduced with permission from Hochberg MC (2001). Gout. In: Silman AJ, Hochberg MC (eds) Epidemiology of the rheumatic diseases, 2nd edition, pp. 230–​42. Oxford University Press. Box 12.4.3  Drugs and dietary supplements causing hyperuricaemia and gout • Thiazide diuretics (including bendroflumethiazide, chlortalidone, cyclopenthiazide, indapamide, metolazone, and xipamide) • Loop diuretics (including furosemide, bumetanide, and torasemide) • Pyrazinamide and ethambutol • Low-​dose aspirin • Nicotinic acid Reproduced with permission from Gibson TJ (2013), Hypertension, its treat- ment, hyperuricaemia and gout. Curr Opin Rheumatol, 25, 217-22. Copyright © 2013 Lippincott Williams. Fig. 12.4.6  Negatively birefringent uric acid crystals in synovial fluid. Under polarized light, horizontal crystals appear as yellow. Reproduced from Dalbeth, N. Laboratory testing in gout diagnosis and management. In (Ed.), Gout (Oxford Rheumatology Library). Oxford, UK: with permission from Oxford University Press.

12.4  Disorders of purine and pyrimidine metabolism 2021 Chronic tophaceous gout Large deposits (tophi) containing monosodium urate monohydrate crystals produce firm nodules over affected joints on the extensor surfaces of the fingers (Box 12.4.3, Fig. 12.4.7), hands, olecranon bursas (commonly bilateral), extensor surfaces of the forearm, Achilles tendon, the helix of the ear, and in the renal parenchyma. Tophi may discharge white chalky material, containing sodium urate monohydrate. They cause the bone erosions and joint destruction with secondary degenerative arthritis that is seen on radiographs. Tophus formation can be regarded as an attempted, but disordered, healing process in response to the presence of sodium urate monohy- drate crystals in tissues. Saturnine gout The link between alcohol consumption, lead poisoning, and gout should be considered in socioeconomic backgrounds where the il- licit brewing of alcohol is commonplace. Lead solder and piping in the equipment used for the illicit brewing and distillation of alcohol may be a source of lead contamination in the beverages produced. Lead toxicity and saturnine gout may also derive from occupational exposure, lead-​based paints (particularly in childhood), ceramic glazes, additives to petrol, drinking water systems, and even nat- ural remedies. The association between lead and hyperuricaemia remains poorly understood but is most likely to be a consequence of chronic kidney disease. In addition to the symptoms of gouty arthritis, symptoms of lead poisoning are usually present. The condition occurs equally in both males and females, and often at a relatively young age of 40 to 60 years, even younger in those where there has been excessive child- hood exposure to lead. These symptoms and signs include protein- uria and chronic kidney disease, as well as anaemia characterized by red cell basophilic stippling due to the inhibition of pyrimidine-​5ʹ-​ nucleotidase. Burton’s lines, a bluish line on the gums, may be present. Lead excretion in urine is elevated, and serum lead levels are ele- vated in cases with current lead exposure. Treatment includes limi- tation of continued lead exposure, lead chelation therapies such as EDTA, dimercaprol, and succimer, as well as xanthine oxidase in- hibitors to lower blood urate levels. Gout due to autosomal dominant tubulointerstitial kidney disease Rare cases of familial gout, presenting in early adulthood or before, have been identified for a long time in families with a history of renal failure. It is now known that familial (juvenile) hyperuricaemic nephropathy is associated with at least four gene defects, and these disorders are more accurately known as the autosomal dominant tubulointerstitial kidney diseases (ADTKD). ADTKD-​UMOD is due to defects in the gene encoding uromodulin (UMOD) and a specific syndrome ‘medullary cystic kidney disease:  familial juvenile hyperuricaemic nephropathy’ (MCKD/​FJHN) is now recognized. These patients show impaired urine concentrating ability, hyperuricaemia and hypouricosuria due to a reduced net tubular excretion of uric acid, cysts (specifically at the corticomedullary junction), interstitial fibrosis, and ultimately renal failure. These cases are associated with mutations in the gene directing the synthesis of the glycoprotein uromodulin, also known as Tamm–​Horsfall protein, which occurs in the cells of the thick, as- cending segment of Henle’s loop, and in renal collecting tubule cells. Uromodulin excretion is diminished. Heterozygous missense gene mutations or small insertion/​deletion events in the gene encoding uromodulin have been demonstrated, and it is proposed that these produce changes in the tertiary structure with conformational changes in the protein due to reduction in the intramolecular disul- phide bonding, which alters the folding pattern of the protein and glycosylation. Uromodulin is an 85-​kDa glycoprotein which also has a role in preventing renal stone formation, the modulation of im- mune responses, and urothelial cytoprotection. Management Acute attack Full doses of any of the nonsteroidal anti-​inflammatory drugs (NSAIDs) are effective in terminating attacks of acute gout. Indomethacin is particularly favoured by some clinicians. Colchicine remains a very effective remedy. The American College of Rheumatology guidelines recommendation is that acute gout can be treated with a loading dose of 1.2 mg of colchicine followed by 0.6 mg 1 h later. This regimen can then be followed by prophylaxis dosing of 0.6 mg once or twice daily 12 h later, until the gout at- tack resolves. For countries where 0.5 mg tablets only are available, the advice is 1.0 mg colchicine as the loading dose, followed by 0.5 mg 1 h later, and then followed, as needed, after 12 h, by continued colchicine (up to 0.5 mg three times daily) until the acute attack re- solves. Lower total doses (1.5–​1.8 mg) have comparable efficacy to higher doses in acute gout but appear to be without the unwanted gastrointestinal toxicity. High doses of colchicine can cause gastro- intestinal haemorrhage and favour the development of other severe side effects, including profuse diarrhoea, rashes, renal and hep- atic damage, and (more rarely) peripheral neuropathy, myopathy, and alopecia in the long term. Intravenous colchicine is no longer recommended. An attack of acute gout can be effectively terminated by the adrenocorticotropin analogue, tetracosactrin, by a single intra- venous dose of hydrocortisone, with a short course of oral prednis- olone (typically 30 mg daily), or (for some joints) by steroid injection into the affected joint. Rebound attacks of acute gout tend to occur unless the situation is covered by either colchicine or an NSAID. Fig. 12.4.7  Thiazide or loop diuretic treatment for hypertension are frequent causes of hyperuricaemia and gout. Tophus overlying a Heberden’s node in an elderly patient taking bendroflumethiazide for hypertension. Reproduced with permission from Gibson TJ (2013), Hypertension, its treatment, hyperuricaemia and gout. Curr Opin Rheumatol, 25, 217–​22. Copyright © 2013 Lippincott Williams.

SECTION 12  Metabolic disorders 2022 Pharmacological doses of colchicine disrupt the microtubular function in inflammatory cells. This mode of action gives it the po- tential to do more widespread damage. Short intensive courses of colchicine should not be repeated at less than 3-​day intervals, al- though lower doses (0.5–​2 mg/​day) can be used for longer periods, as in the treatment of familial Mediterranean fever. Rasburicase (dosage 20 mg/​kg per day, treatment for more than 5 days is not recommended) is a recombinant urate oxidase derived from Saccharomyces cerevisiae. It catalyses the oxidation of urate to allantoin which is five times more soluble than uric acid at urinary pH values and is the purine metabolite excreted by nonprimate spe- cies. Acute hypersensitivity reactions have been reported in 5% of patients who do not have a history of allergy. It should not be used in pregnancy or in glucose-​6-​phosphate dehydrogenase deficiency. It can be used to terminate an attack of acute gouty arthritis, but this seems unnecessary with the availability of well-​established methods. However, it may have a place in the treatment of acute uric acid neph- ropathy in tumour lysis syndrome and in patients who are allergic to allopurinol and the other drugs used to treat hyperuricaemia and gout. More recently, pegloticase a polyethylene glycol conju- gate of a recombinant porcine uricase which is less immunogenic than rasburicase and therefore more suited for longer-​term therapy, has benefited patients with a severe gout disease burden and re- fractoriness to oral urate-​lowering therapies, but has not been ap- proved for this purpose by the National Institute for Health and Care Excellence in the United Kingdom. The agent appears to have powerful hypouricaemic effects with debulking of tophi in severely affected patients resistant to other therapies. Infusion reactions are frequent, although frank anaphylaxis appears to be uncommon. Rapid breakdown of plasma urate by uricase has been associated with a high frequency of acute exacerbations of gout in the early weeks after its introduction. Moreover, since all putative treatments of hyperuricaemia based on the action of uricases have the potential to generate abundant hydrogen peroxide and other oxidants, their introduction for long-​term use carries with it an appreciable risk of tissue injury (see contraindication for use of rasburicase, in ‘Acute uric acid nephropathy’). Although pegylated and other preparations of uricases from various sources demonstrate clear efficacy in vivo and remain attractive for therapeutic research, at the time of writing, this approach does not yet have an established place for the treat- ment of severe chronic gout. Interval treatment Asymptomatic hyperuricaemia should not be treated with urate-​ lowering drugs unless the patient experiences more than one acute attack of gout per year (Box 12.4.4). Allopurinol, a xanthine oxi- dase inhibitor, is effective in preventing acute gout by reducing the serum urate concentration to a value below the solubility of sodium urate monohydrate in plasma so that tophaceous deposits are mo- bilized and healing occurs. This applies to the tophi in bones as well as elsewhere. The drug should be introduced at a low level (e.g. 100–​ 200 mg daily) and increased under cover of either colchicine or an NSAID, which should be continued until the serum urate concen- tration has stabilized at a normal level. Allopurinol is then continued indefinitely. Initiating allopurinol without cover may cause attacks of acute gout as the serum urate concentration falls. Moderately severe gout may require as much as 300 to 600 mg allopurinol daily and occasionally as much as 700 to 900 mg/​day given in divided doses. Between 10 and 20 mg/​kg body weight per day is an appropriate dosage for children. The incidence of adverse reactions to allopurinol is low but they can be severe and occasionally fatal. Reactions include erythema multiforme progressing to Stevens–​Johnson syndrome and toxic epidermal necrolysis (associated with the HLA B*5801 allele), ex- foliative dermatitis, vasculitis, interstitial nephritis, eosinophilia, hepatocellular damage, polyneuropathy, bone marrow suppres- sion, disturbances of vision and taste, as well as gastroenteropathy. Allopurinol potentiates the effect of coumarin anticoagulants (e.g. warfarin), azathioprine, and 6-​mercaptopurine, and predisposes to an ampicillin or amoxicillin rash. At high dosage and in the presence of greatly increased purine synthesis, it may cause radiotranslucent xanthine and oxypurinol urinary stones. There is also an increased risk of toxicity with captopril (especially in the presence of renal failure) and with ciclosporin. Much of the overall toxicity of allopurinol is due to the metabolite oxypurinol, which has a much longer half-​life in vivo than the parent compound. Special care is necessary in the presence of advanced chronic kidney disease and a dose of 100 to 150 mg is usually suffi- cient in this circumstance. Patients with hyperuricaemia due to renal failure rarely develop gout, possibly due to their immunoparesis. Treatment when  allopurinol produces adverse reactions  The specific xanthine oxidase inhibitor, febuxostat, has been ap- proved by the European Commission, the National Institute for Health and Care Excellence in the UK and by the Food and Drug Administration in the United States of America, for the treatment of chronic gout in which it rapidly decreases serum urate concen- trations. Febuxostat is appropriate for patients hypersensitive to or intolerant of allopurinol, those in whom allopurinol has failed to control symptomatic hyperuricaemia, and in patients with chronic kidney disease where uricosuric therapy is contraindicated. As with allopurinol, suitable prophylaxis against exacerbation of acute gout is indicated (e.g. with colchicine) when treatment with febuxostat is started. The drug is approved in European countries at 80 and 120 mg daily. In the United States of America, the label is for a daily dose of 40 mg, increasing to 80 mg after at least 2 weeks if the serum urate concentration remains elevated. Febuxostat at a dose of 40 mg/​day is associated with a higher likelihood of achieving a target serum uric acid level of 6 mg/​dl (0.36 mmol/​L) than allopurinol given at Box 12.4.4  Hyperuricaemia detected on routine biochemical screening • Search for an identifiable cause (e.g. dietary factors, myeloproliferative disease, medications) • Check renal function • Imaging to detect the presence of uric acid urinary calculi • Measure uric acid excretion after eliminating dietary and medication factors • Treat if:

—​ more than one attack of acute gouty arthritis per year

—​ chronic joint damage attributed to gout

—​ tophi

—​ hyperuricaemic nephropathy

—​ uric acid urolithiasis.

12.4  Disorders of purine and pyrimidine metabolism 2023 the commonly used doses of 300 mg/​day. Nevertheless, a meta-​ analysis dating from 2013 concluded that there was no evidence that febuxostat is superior to allopurinol for clinically relevant out- comes. Given the higher cost of febuxostat, the evidence suggests that febuxostat should not be routinely used as a first-​line treat- ment for chronic gout. Since it is an inhibitor of xanthine oxidase, febuxostat, like allopurinol, has the potential for highly toxic drug interactions with azathioprine, 6-​mercaptopurine, and theophylline and its derivatives. Patients for whom the treatment of hyperuricaemia and gout is essential and in whom therapy with xanthine oxidase inhibi- tors have been ineffective present a special problem, especially if they have impaired overall renal function. The uricosuric drugs sulphinpyrazone, probenecid, and benzbromarone, together with a sufficiently high fluid intake to provide a measured urine output of at least 3 litres/​24 h and alkalization of the urine with sodium or potassium bicarbonate or sodium or potassium citrate, are an approach to this problem, but may be inappropriate in the overall clinical context, for example, in patients with cardiac or renal failure. Only sulphinpyrazone is readily available in the United Kingdom. Uricosuric drugs may be inefficient in the presence of renal failure and are contraindicated in the presence of uric acid urinary stones. The uricosuric agent benzbromarone is sometimes effective in pa- tients with renal failure when other uricosuric agents have lost their efficacy. The use of oxypurinol (at low dosage) has also been pro- posed. Protocols are also available for the desensitization of patients who have experienced adverse reactions to allopurinol and in whom the risk of uric acid stone formation, with the potential for further reduction of renal function, presents a problem. Other hyperuricaemic conditions Acute uric acid nephropathy This complicates the treatment of widespread malignant disease, particularly chemotherapy and/​or radiotherapy of leukaemias and lymphomas. The nephropathy is of multifactorial origin and may form part of the acute tumour lysis syndrome with accompanying tubular necrosis. These patients are usually underhydrated, acid- otic, and have high rates of uric acid production from nucleoprotein degradation in the apoptotic tumours. Acute uric acid nephropathy has occasionally been reported after extremely severe muscular ex- ercise, after severe epileptic seizures, and in patients with gout due to grossly increased rates of de novo purine synthesis. The renal lesion is the intratubular precipitation of uric acid crys- tals. In addition, the renal pelvis and ureters may also be blocked by crystal aggregates and/​or uric acid stones. Acute uric acid neph- ropathy can be avoided by giving allopurinol for several days be- fore starting the chemotherapy or radiotherapy. The condition presents as acute oliguric renal failure. Imaging techniques should be used to exclude the presence of bilateral ureteric obstruction by radiotranslucent uric acid stones. Treatment is by: • induction of an alkaline diuresis • haemodialysis, peritoneal dialysis, or haemofiltration • percutaneous nephrostomy and/​or ureteric catheterization may be needed if there is an element of postrenal obstruction due to impacted aggregates of sodium urate crystals or uric acid stones • disruption or removal of impacted stones. Rasburicase has been licensed for use as a single-​course therapy for hyperuricaemia in the acute paediatric and adult tumour lysis syndrome. The enzyme has a plasma half-​life of 18 to 24 h and is markedly antigenic, therefore having little application as an off-​ label agent in severe tophaceous gout and certainly not sustainable for more than a few months. On account of its capacity to induce oxidant injury and thus haemolysis in susceptible individuals, rasburicase is contraindicated in patients with glucose-​6-​phosphate dehydrogenase deficiency. Ethanol-​induced hyperuricaemia Ethanol is oxidized to acetaldehyde by the liver. This raises the ratio of reduced nicotinamide adenine dinucleotide to nicotina- mide adenine dinucleotide, which in turn promotes the reduc- tion of pyruvate to lactate in the hepatocytes. Lactate competes with urate in the renal tubular excretory mechanisms and thereby promotes urate retention. There is often an element of starva- tion ketoacidosis in chronic alcoholics, with acetoacetate and β-​hydroxybutyrate also competing for the renal tubular excretory mechanisms which subserve urate tubular secretion. In addition, there is increased urate production associated with ethanol intake, first due to the high purine content of some alcoholic beverages (e.g. beer) and second because the metabolism of alcohol involves increased dephosphorylation and degradation of adenine nucleo- tides in the liver. Uric acid urolithiasis Pure uric acid stones account for 5% of all urinary stones in patients in the United Kingdom, but there is a much higher incidence else- where (e.g. in the Middle East). In Israel, about 40% of urinary cal- culi are composed of uric acid and 75% of patients with primary gout develop renal calculus disease. Overall, uric acid urolithiasis occurs in about 10% of patients with gout, more often in secondary gout than in primary gout, and sometimes associated with an im- paired ability to alkalinize the urine. Ileostomy predisposes to uric acid urolithiasis because of (1) chronic bicarbonate loss, which leads to a persistent acidification of the urine, and (2) a concentrated urine due to excessive water loss. Urinary uric acid concentrations close to or more than those at which spontaneous crystallization begins are frequent in these circumstances. The genetic causes of uric acid urolithiasis are rare: (1) HPRT deficiency, (2)  PRPP superactivity, and (3)  inherited renal hyporuricaemia (congenital failure of the renal tubular reabsorp- tion of urate). The urinary uric acid concentration is the main determinant of uric acid stone formation. The concentration depends on the state of hydration, the rate of de novo purine synthesis, the rate of metabolic turnover of purine compounds, the dietary intake of purines and alcohol, and the action of uricosuric drugs (e.g. sulphinpyrazone). Calcium oxalate stone formation is increased 30-​fold in patients with gout, and hyperuricosuria is common in nongouty stone-​ formers. Uric acid microcrystals may act as epitaxial nucleation sites for calcium oxalate crystallization. It is also possible that colloidal uric acid adsorbs urinary glycosaminoglycan inhibitors of crystal- lization and crystal growth. Uric acid stone disease is treated by hydration to maintain a urine volume of at least 3 litres/​24 h, alkalization of the urine, and allopur- inol if there is hyperuricosuria. The use of sodium and potassium

SECTION 12  Metabolic disorders 2024 salts for alkalization has to be carefully reviewed in the light of con- current diseases, particularly impaired renal and cardiac function. The standard imaging techniques (particularly ultrasonography) are required for the diagnosis of these radiotranslucent stones. Stones can be fragmented or removed by standard procedures. For further discussion of urinary stones, see Chapter 21.14. Hereditary renal hypouricaemia and uric acid stones The causes of hypouricaemia are summarized in Box 12.4.5. Renal hypouricaemia may be due to renal tubular damage by gen- etic diseases or by toxic damage (Box 12.4.5), and this may be as- sociated with other features of Fanconi’s syndrome. Reduced net tubular reabsorption of urate occurs as an isolated renal tubular reabsorption defect due to loss of function mutations in the genes directing the synthesis of the urate carriers. Type 1 is due to loss of function mutations in URAT1 (SLC22A12) and type 2 due to mutations in GLUT9 (SLC2A9). Inheritance is autosomal reces- sive. Hyperuricosuria is a feature and may amount to 1000 mg (5.9 mmol) per 24 h in homozygous patients, with a lesser degree of hyperuricosuria in heterozygotes. Uric acid urolithiasis occurs in about 25% of the homozygotes, most commonly in patients with combined hyperuricosuria and hypercalciuria. Treatment with allopurinol has, counter-intuitively, been used to prevent the recurrence of renal stones in patients who have experi- enced acute renal injury after exercise. The rationale is to decrease the generation of uric acid thereby decreasing the filtered uric acid load and lowering the risk of precipitation in the renal tubules. Reduced tubular urate reabsorption can occur in other inherited or acquired renal tubule transport defects (Box 12.4.5). Hypoxanthine-​guanine phosphoribosyltransferase deficiency: Lesch–​Nyhan syndrome and its variants Lesch–​Nyhan syndrome results from mutations in the gene encoding HPRT, an enzyme which catalyses the salvage of hypo- xanthine and guanine to inosine monophosphate (IMP) and guano- sine monophosphate (GMP), respectively, as shown in Fig. 12.4.1. In affected male hemizygotes, the lack of HPRT results in increased levels of PRPP due to failure to salvage hypoxanthine or guanine. Elevated PRPP levels then act as a driver of de novo purine syn- thesis, resulting in purine overproduction. The clinical spectrum extends from hyperuricaemia alone to hyperuricaemia with pro- found neurological and behavioural dysfunction. The biochemistry and molecular genetics of this disorder have been studied exten- sively. Functional assays of HPRT on cultured fibroblasts or intact red cells, rather than erythrocyte lysates, give a better correlation between the degree of residual enzyme activity and clinical pheno- types. Mutation analysis is a valuable tool for genetic counselling, the identification of carriers, and prenatal diagnosis. Pathophysiology Both the de novo purine synthesis and the HPRT-​catalysed purine salvage pathways are present in all parts of the normal brain. HPRT activity is absent or defective but the de novo synthesis pathway remains active in patients with the Lesch–​Nyhan syndrome. The present view is that the neurological manifestations are brought about by a neurotransmitter imbalance, probably mainly in the basal ganglia. This imbalance is possibly due to a deficient supply of metabolic energy resulting from the nonsalvage of hypoxan- thine and guanine causing a deficiency of adenine nucleotides that provide energy for short bursts of neurotransmitter syn- thesis. However, the positron emission tomography evidence of dopamine receptor deficiency is the main concrete evidence for a neurotransmitter defect, either directly or indirectly because of guanosine triphosphate deficiency underlying the Lesch–​Nyhan syndrome. There is increased excretion of the serotonin metabolite 5-​hydroxyindoleacetic acid and decreased levels of homovanillic acid, a major metabolite of dopamine, in the cerebrospinal fluid. Deficiency of basal ganglia dopamine systems emerging during the first 2 months of life has been demonstrated in a mouse model of Lesch–​Nyhan syndrome. Failure of pubertal development and testicular atrophy in HPRT deficiency are attributed to an inadequate supply of purine nucleo- tides to meet the increased metabolic energy requirement in the Box 12.4.5  Causes of hypouricaemia Inherited disorders of uric acid biosynthesis • Genetic defects in the molybdoflavoprotein enzymes:

—​ Xanthinuria type I (isolated xanthine oxidase deficiency)

—​ Xanthinuria type II (combined xanthine oxidase and aldehyde oxi- dase deficiencies)

—​ Molybdenum cofactor deficiency (xanthine oxidase, aldehyde oxi- dase, and sulphite oxidase deficiency) • Purine nucleoside phosphorylase deficiency Secondary reduction in uric acid biosynthesis • Allopurinol and oxypurinol medication • Hepatic failure • Acute intermittent porphyria Inherited renal hypouricaemia (isolated renal tubule reabsorption defect) • Loss of function mutations in urate transporters URAT1 (SLC22A12) and GLUT9 (SLC2A9) • Inherited causes of Fanconi’s syndrome and its variants (the syndrome of multiple renal tubule reabsorption defects) Acquired causes of Fanconi’s syndrome and its variants • Metal poisoning (Cd, Zn, Cu, Pb, Hg, Ur) • Multiple myelomatosis • Nephrotic syndrome • Malignant disease (paraneoplastic syndrome) • Autoimmune disease (i.e. Sjögren’s syndrome) • Thermal burns • Primary hyperparathyroidism • Acute renal tubular necrosis • Renal transplant rejection Drugs • Drugs used either as uricosuric agents or to block other aspects of renal tubule excretion (sulphinpyrazone, probenecid, benzbromarone) • NSAIDs with uricosuric properties • Phenylbutazone • Azapropazone • Aspirin dosage greater than 4 g/​day • Coumarin anticoagulants (e.g. warfarin) • Outdated tetracycline (5α-​6-​anhydro-​4-​epitetracycline) Nutritional deficiencies • Vitamins B12, C, and D • Kwashiorkor

12.4  Disorders of purine and pyrimidine metabolism 2025 testis at this time. A similar inability to meet energy requirements may underlie the neurological manifestations. A partial defect in adrenocortical 11β-​hydroxylation of steroids is demonstrable in patients with the Lesch–​Nyhan syndrome after ACTH stimulation and is thought to be linked with a failure to modulate mitochon- drial function for this hydroxylation due to a deficiency of purine nucleotides. Clinical features The clinical features of the most severely affected patients who are correctly referred to as having classic Lesch–​Nyhan syndrome or as having ‘complete or virtually complete HPRT deficiency’ are sum- marized in Box 12.4.6. In some cases, the enzyme has altered kin- etics or is unstable but has 1 to 5% residual activity. Patients with partial enzyme defects of 0 to 5% HPRT activity in red cell lysates but more than 8% activity in the fibroblast assays have gout and renal complications but no neurological manifestations. The disease fre- quency is about 1 in 380 000 births. Infants affected by HPRT deficiency have a lower than average birth weight, indicating some degree of intrauterine growth retard- ation. The first clinical sign may be the presence of red grit (uric acid crystals with absorbed urinary pigments) on the nappy. Affected in- fants are hypotonic from birth, although this is frequently not re- marked on before poor head control becomes apparent at the age of about 3 months. Postnatal growth, which becomes more marked after the second year of life, is also subnormal (Fig. 12.4.8) as indicated by sequential measurement of body weight, accurate assessment of body length being impossible due to the dystonic posturing. The overall pattern of weight growth follows centile lines for the first 2 years of life and thereafter slows to about 1 kg/​year, or about half normal; a pubertal growth spurt is not observed. Head growth and bone development are less affected than weight. The poor weight gain cannot be attrib- uted to either renal failure or malnutrition. Torsion dystonia, with its two components of abnormal pos- turing and episodic rigidity, is superimposed on the basic hypotonia that is present between the dystonic episodes. Severe dysarthria is associated with dyskinesia of the face, mouth, pharynx, and the larynx, which greatly limits communication and even the ability to point accurately, leading to great frustration. The self-​injurious be- haviour and dyskinesia are eliminated or much reduced when the child is concentrating on a self-​selected activity, such as watching an interesting television programme. Self-​injury and dyskinesia are exacerbated by excitement, such as the arrival of a visitor, fear, frus- tration, and unsuccessful attempts at volitional motor activity. The children also appear to be aware of the value of this behaviour as an attention-​seeking manoeuvre, and sometimes appear to use it in a manipulative manner. Although learning difficulties have been stressed as a feature of the Lesch–​Nyhan syndrome, they are of inconstant severity and are neither marked nor specific. The apparent degree of intellectual dis- ability may be affected by the extensive disorder of expressive motor functions that exceeds the comprehension defect, by the lack of basic social and educational opportunities, and by the lack of intel- ligence tests for older children who have lacked these opportunities. However, for whatever combination of reasons, there does appear to be a decline of intellect from the age of 8 to 10 years. Self-​injurious behaviour usually begins at about 2 years of age. Its severity and the ingenuity with which the patients exploit new ways of self-​injury exceed that encountered in any other clinical situation. It is not a constant feature and some patients never show it; in most its severity waxes and wanes. Self-​injury can produce very severe damage, such as complete destruction of the lower lip or trau- matic amputation of a fingertip. The patients feel pain normally and are aware of their compulsion; they are afraid of it but are unable to control it. Nyhan and his colleagues consider it to be the clinical hallmark of complete HPRT deficiency, as opposed to those patients with some residual enzyme activity (which may or may not be meas- urable in erythrocyte lysates). The severe dystonic spasms with violent extension of the neck can produce damage to the cervical spinal cord and produce motor pyramidal tract signs in the legs. The phenotypes associated with appreciable residual HPRT activity vary from the neurological def- icit described for the complete Lesch–​Nyhan syndrome but without self-​mutilation (Lesch–​Nyhan variant), to patients with only X-​ linked gout and/​or urolithiasis and only very subtle, if any, neuro- logical features. Box 12.4.6  Clinical manifestations of the Lesch–​Nyhan syndrome (complete or virtually complete absence of HPRT deficiency) • X-​linked recessive inheritance • Failure of overall growth • Muscle hypotonia • Delayed motor development • Torsion dystonia • Aggressive behaviour • Dysarthria • Variable degree of intellectual deterioration in later childhood • Megaloblastic anaemia (in some cases only) • Hyperuricaemia and hyperuricaciduria with gout and tophus develop- ment after puberty and urolithiasis occasionally during the first decade of life • Failure of pubertal development and testicular atrophy at the age when puberty would be expected to occur 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Age (years) 0 10 20 30 40 Weight (kg) Fig. 12.4.8  Patterns of growth in the weight of 13 boys with the Lesch–​ Nyhan syndrome: each patient is shown by a different symbol. The 50th and 3rd centiles are shown. With kind permission from Springer Science+Business Media: J Inherit Metab Dis, Lesch-​Nyhan syndrome: Growth delay, testicular atrophy and a partial failure of the 11β-​hydroxylation of steroids, 10, 3, 1987, 210–​223, R. W. E. Watts. Copyright © 1987, SSIEM and MTP Press Limited.

SECTION 12  Metabolic disorders 2026 Investigation There are no structural or ultrastructural changes in the brain as judged by light and electron microscopy or on electroencephalog- raphy. Reductions in the size of the caudate nucleus, the putamen, and total cerebral volume have now been demonstrated by refined MRI. Dopamine transporter reduction and hence dopamine defi- ciency has been demonstrated by positron emission tomography. Imaging studies have supported the concept that HPRT defi- ciency constrains brain development with particular emphasis on the basal ganglia, and defective function of dopaminergic neurons is specifically involved. A more recent study of adult patients with classic Lesch–​Nyhan syndrome showed larger reductions of white (26%) than grey (17%) matter volume relative to healthy controls. These reductions were less marked in patients classified as Lesch–​ Nyhan variant with reductions of white (14%) and grey (15%) matter volume. Both patient groups demonstrated reduced volume in medial inferior white matter regions. Compared with the variant group, classic Lesch–​Nyhan syndrome patients showed larger re- ductions in inferior frontal white matter adjoining limbic and tem- poral regions and the motor cortex. These regions likely include such long association fibres as the superior longitudinal and un- cinate fasciculus. Patients with Lesch–​Nyhan syndrome whose hyperuricaemia has been controlled and who have not had renal damage, often live be- yond their third decade, but may succumb to sudden death, usually due to respiratory problems. Management Control of hyperuricaemia Allopurinol should be administered to reduce the plasma urate and urine uric acid concentration in order to prevent gouty arth- ritis, urate nephropathy, and renal calculi. As HPRT deficiency leads to increased de novo purine synthesis and purine overpro- duction, allopurinol may lead to dramatically increased concentra- tions of xanthine in urine and consequent xanthine nephropathy. Allopurinol treatment should be titrated against plasma uric acid as well as urinary uric acid and xanthine concentrations. The pa- tient should also be kept well hydrated to minimize the risk of xanthine stone formation. Allopurinol treatment from birth does not prevent the behavioural phenotype. All therapeutic attempts at neuropharmacological manipulation have been unsuccessful. There are however anecdotal reports that intrathecal baclofen ameliorated the motor symptoms and also improved behaviours. Oral S-​adenosylmethionine at doses of 20 to 25 mg/​kg per day has been reported to dramatic reduce self-​injurious and aggressive be- haviour, as well as a milder reduction of dystonia in five Malaysian patients. Behavioural manifestations Following on from the discussion of pathophysiology, some aspects of the Lesch–​Nyhan phenotype may be related to dysfunction of the small central, but widely projecting, aminergic pathways involved in learning. It has been suggested that the self-​injurious behaviour in the Lesch–​Nyhan syndrome is due to an imbalance between the ac- tivities of catecholaminergic neurons and 5-​hydroxytryptaminergic neurons, the former being largely concerned with learning by re- ward and the latter with learning by punishment. Patients with the Lesch–​Nyhan syndrome are insensitive to punishing stimuli and do not learn when such stimuli are used to reinforce the desired behaviour, which in this case is to refrain from self-​injury, and their ability to learn from rewarding stimuli is also impaired. Psychotherapeutic techniques that are effective in elim- inating self-​injurious behaviour in other situations fail in patients with the Lesch–​Nyhan syndrome. They could be modified by a pro- gramme of positive reinforcement of abstaining from self-​injury and ‘time out’, but this has proved difficult to achieve in the long term. The reinforcement strategy was found to be unsuitable for use at home because it involved apparently ignoring the self-​injury and only paying attention to the child in the absence of self-​injurious behaviour, which was misinterpreted by friends and relations as un- kindness or indifference. Managing self-​injurious behaviour therefore requires teaching good communication skills, relaxation techniques, consistent hand- ling from all those involved in care, as well as physical restraints and specially designed equipment. Dental extraction, physical restraints with splints and bandages, and strapping the patient into a specially designed padded wheelchair fitted with a firm padded head support to prevent cervical spine injury during violent opisthotonic spasms, are usually needed to limit the effects of compulsive self-​mutilation. Children whose restraints have been temporarily released ask or indicate their wish for the bandages, straps, etc. to be replaced so that they are less able to damage themselves. Every effort should be made to exploit the intellect of these patients and to keep them in a stimulating environment. Clinical genetic aspects The Lesch–​Nyhan syndrome and its variants are inherited in a sex-​linked recessive manner. Clinical manifestations in the female carriers are rare, but subtle alterations in purine metabolism, with small increases in the rates of de novo purine synthesis and in- creased uric acid excretion and occasionally mild asymptomatic hyperuricaemia, have been reported. There are extremely rare re- ports of female Lesch–​Nyhan cases which have been attributed to nonrandom inactivation of the X chromosome. Genomic analysis is preferred for the identification of carrier females. The demonstra- tion of mosaicism by HPRT+ and HPRT− hair roots due to random inactivation of the X chromosome is error prone and is not widely available. Similarly, autoradiographic techniques to demonstrate two cell populations (HPRT+ and HPRT−) in fibroblast cultures are rarely used. Prenatal diagnosis is possible using chorionic villus samples obtained during the ninth week of pregnancy; this permits elective termination of an affected fetus before the end of the first trimester of pregnancy. In vitro fertilization with enzymatic assay or geno- typing on a cell removed at the four-​cell stage to ensure that only unaffected embryos are implanted is possible. Other disorders of purine metabolism PRPP synthetase superactivity and PRPP synthetase deficiency The enzyme PRPP synthetase (PRPS) catalyses the production of PRPP, which is required for the first specific and rate-​limiting re- action in the de novo pathway of purine synthesis, is encoded by the PRPS1 gene. It is subject to feedback inhibition by purine nu- cleotides. Mutations associated with diminished sensitivity to this

12.4  Disorders of purine and pyrimidine metabolism 2027 feedback inhibition, lead to increased PRPP production, which in turn acts as a driver for increased de novo purine synthesis, leading to hyperuricaemia, hyperuricosuria, and gout. The condition is in- herited in an X-​linked recessive fashion. Affected males develop uric acid lithiasis or gouty arthritis in childhood or early adult life. Hyperuricaemia is often severe and in the range 0.5 to 1 mmol/​litre, with uric acid excretion of 5 to 15 mmol/​24 h. Heterozygous females are usually asymptomatic, al- though some degree of increased purine synthesis de novo and occa- sionally gout, may occur with nonrandom X-​inactivation. In some families, the disorder presents in childhood with associ- ated neurological features such as motor retardation and learning difficulties, ataxia, deafness, hypotonia, disturbed speech, and the development of polyneuropathy, intracerebral calcifications, and dysmorphic facial features. Carrier females may have mild hyperuricaemia and some degree of hearing impairment. The con- stellation of associated disorders varies in different families. Carriers can be identified by mutation analysis and by studies in cultured skin fibroblasts. Amniocentesis, prenatal diagnosis, and preventive termination of pregnancy are not justified in this condi- tion, unless one of the unusually severe phenotypes is known to be segregating in the family. The hyperuricaemia, primary purine over- production, and uricosuria can be well controlled with allopurinol. Mutations in PRPS1 can also lead to PRPP synthetase defi- ciency of varying degree and a broad spectrum of phenotypes including syndromic and nonsyndromic hearing loss comprising Charcot–​Marie–​Tooth, X-​linked recessive disease type 5 (CMTX5), nonsyndromic sensorineural deafness (DFN2), and Arts’ syndrome. Hearing loss in affected males is bilateral and ranges from moderate to profound. Arts’ syndrome is characterized by profound sensori- neural hearing impairment, early-​onset hypotonia, delayed motor development, mild to moderate mental retardation, ataxia, and optic atrophy. Oral supplementation with S-​adenosylmethionine (30 mg/​ kg per day) has provided significant clinical benefit to two brothers diagnosed with Arts’ syndrome. This suggests that other patients with PRPS synthetase deficiency, including mildly affected carrier females, may benefit from S-​adenosylmethionine supplementation. Uric acid concentrations in patients with PRPS deficiency may be low or normal and levels cannot be used as a diagnostic marker. Adenine phosphoribosyltransferase deficiency and 2,8-​dihydroxyadeninuria These patients lack adenine phosphoribosyltransferase activity; adenine accumulates behind the metabolic block and is oxi- dized by xanthine oxidase to the very insoluble compounds 2-​ hydroxyadenine and 2,8-​dihydroxyadenine. These compounds are excreted in the urine along with adenine itself, where it forms radiotranslucent stones that are white or pale fawn in colour. These rough and friable calculi have, in the past, been widely misdiagnosed as uric acid stones because 2,8-​dihydroxyadenine reacts as if it were uric acid in colorimetric assays. The use of enzymatic uric acid as- says has obviated this confusion. Adenine phosphoribosyltransferase deficiency has an autosomal recessive pattern of inheritance and is clinically silent in hetero- zygotes. There are two subtypes (I and II). Type I patients have no detectable enzyme activity, being homozygotes or compound het- erozygotes for null alleles. Type II patients have between 5 and 25% residual enzyme activity. Whereas type I patients are encountered in many racial groups, the type II subtype has so far only been identi- fied in the Japanese population. This condition often presents in early life because of the extremely low solubility of 2,8-​dihydroxyadenine in renal tubule fluid and urine. Severe obstructive uropathy and renal failure may occur in infancy. Treatment is by hydration and xanthine oxidase inhibition with allopurinol, and with standard measures to disrupt or remove the stones and to manage urinary infections and renal failure. Xanthinuria Xanthine stones occur in patients with xanthinuria due to deficiency of xanthine oxidase/​reductase deficiency and occasionally, in those who are being treated with the xanthine oxidase inhibitor, allopur- inol. The latter is particularly likely in patients with accelerated de novo purine synthesis, as in patients with the Lesch–​Nyhan syn- drome. Xanthinuria is inherited in an autosomal recessive manner, and hypoxanthine and xanthine accumulate behind the metabolic block. The plasma urate concentration and urine uric acid excretion are typically less than 0.06 mmol/​litre (1.0 mg/​dl) and 0.30 mmol/​ 24 h (50 mg/​24 h), respectively, when the patient is taking an un- restricted diet. It is a rare, perhaps under-​recognized condition. Concentrations of urine oxypurines (hypoxanthine plus xanthine) are characteristically elevated. Normal subjects have plasma levels between 0.00 and 0.15 mmol/​litre (0.00–​0.25 mg/​dl) and urine levels of 0.07 to 0.13 mmol/​24 h (11–​22 mg/​24 h); patients with xanthinuria typically have plasma levels between 0.03 and 0.05 mmol/​litre (0.00–​ 0.90 mg/​dl) and urine levels of 0.60 and 3.5 mmol/​24 h (100–​600 mg/​24 h). Xanthine accounts for 60 to 90% of the total xanthine plus hypoxanthine excreted, presumably reflecting the more active meta- bolic turnover of hypoxanthine and its efficient salvage by hypo- xanthine phosphoribosyltransferase. Hypoxanthine and xanthine are mainly derived from adenine and guanine nucleotides, respect- ively (Fig. 12.4.1). Hypoxanthine has a relatively high solubility and causes no problems. At any age, about one-​third of cases present with radiotranslucent xanthine stones. These stones are usually smooth, soft, and yellow-​ brown. Xanthinuric myopathy is a rare complication. Type 1 xanthinuria is due to an isolated defect in xanthine oxi- dase/​reductase, while type 2 xanthinuria is due to a defect in mo- lybdenum sulphurase which catalyses the terminal step in the synthesis of the molybdopterin cofactor necessary for the activity of both aldehyde oxidase and xanthine oxidase. These patients present with xanthine stones and are detected by their inability to convert allopurinol to oxypurinol, a reaction normally catalysed by aldehyde oxidase. Xanthine stones also occur when there is a combined deficiency of the three molybdoflavoprotein enzymes, xanthine oxidase, sulphite oxidase, and aldehyde oxidase, because of defective molybdopterin cofactor synthesis. The clinical picture in these patients is overshadowed by the sulphite oxidase defi- ciency that produces severe brain damage and dislocation of the ocular lenses. Adenylosuccinase deficiency and ATIC deficiency Adenylosuccinase (adenylate succinate lyase (ADSL)) catalyses the eighth step on the 10-​step de novo purine synthesis pathway and the second step in one of the purine nucleotide interconversion path- ways, the formation of ATP from inosine monophosphate.

SECTION 12  Metabolic disorders 2028 The patients present in infancy with severe psychomotor dis- abilities, autism, and axial hypotonia with normal tendon reflexes. Self-​mutilation has been recorded in some cases and cerebellar hypoplasia is present on CT scans. The presence of succinyl adenosine and succinyl aminoimidazole carboxamide riboside in plasma, cerebrospinal fluid, and urine confirms the diagnosis. There is gross purine overproduction with high levels of purine ribosides in urine. Urine and plasma uric acid levels are normal. Partial enzyme deficiencies have been demonstrated in liver, kidney, muscle, lymphocytes, and fibro- blasts, with mutations identified in more than 70 patients. Clinical severity is variable and correlates with the ratio of succinyl ad- enosine to succinyl aminoimidazole carboxamide riboside in the cerebrospinal fluid. The bifunctional enzyme 5-​aminoimidazole-​4-​carboxamide ribonucleotide formyltransferase/​IMP cyclohydrolase (ATIC) cata- lyses the last two reactions of de novo purine synthesis. A deficiency of ATIC leads to the accumulation of the dephosphorylated sub- strate of ATIC, aminoimidazol carboxamide riboside, as well as the accumulation of succinylaminoimidazole carboxamide riboside and succinyladenosine in body fluids. Both deficiencies are inherited as autosomal recessive conditions. For ADSL deficiency, the growth retardation has been improved by adenine (10 mg/​kg) and allopurinol. The latter promotes purine conservation by blocking hypoxanthine oxidation to xanthine and uric acid, and prevents the oxidation of administered adenine to 2,8-​dihydroxyadenine. There is currently no effective treatment for either enzyme deficiency. Oral S-​adenosylmethionine has been suggested as a purine replacement therapy in ADSL deficiency, but showed no clinical benefit in a patient treated for 9 months. Myoadenylate deaminase deficiency Myoadenylate deaminase is the muscle-​specific isoenzyme of ad- enylate deaminase which catalyses the deamination of adenosine monophosphate nucleotide (AMP) to inosine monophosphate (IMP) during muscle contraction. This reaction is necessary for normal muscle function. Myoadenylate deaminase deficiency may be congenital, due to a mutation in the gene directing the syn- thesis of the protein, or associated with a wide range of muscle diseases including the muscular dystrophies, polymyositis, and dermatomyositis. Patients with congenital myoadenylate deaminase deficiency pre- sent at any age including early childhood with a syndrome of muscle weakness and muscle cramps during and after exertion. There is some decrease in muscle mass, some hypotonia, and a little muscle weak- ness. There may be a modest rise in plasma creatine phosphokinase levels and nonspecific electromyographic changes. The lack of am- monia and IMP occurs normally in the venous outflow from the af- fected muscles during exercise, and the enzyme deficiency can be demonstrated histochemically. The pattern of inheritance is auto- somal recessive, not all of the homozygotes have clinical symptoms, and the heterozygous carriers are clinically silent. The nonsense variant c.34C>T, (p.Q34X) in exon 2 is polymorphic in Caucasian populations. Exon 2 is alternatively spliced and transcripts lacking the c.34C>T, (p.Q34X) variant encode a functionally active protein. The acquired disorder may be due to the coincidental disease arising in a patient whose inherited myoadenylate deaminase deficiency would otherwise be silent. Myoglobinuria following strenuous exercise has been reported in a few cases and hence the risk of rhabdomyolysis has led some authors to recommend the avoidance of vigorous exercise. Such ad- vice is only appropriate if exertion-​related myoglobinuria has oc- curred or been suspected. Oral ribose (2–​60 g/​day, or taking a dose before vigorous exercise) has been reported to produce symptomatic improvement. Inborn errors of purine metabolism and immunodeficiency Adenosine deaminase and purine nucleoside phosphorylase catalyse sequential steps in the metabolism of purine ribonucleosides and deoxyribonucleosides. These enzymes are highly expressed in the lymphoid cells and their deficiency, which causes the lymphotoxic substrates 2′-​deoxyadenosine or 2′-​deoxyguanosine to accumulate, leads to lymphopenia and immunodeficiency. Most patients with adenosine deaminase deficiency lack both cell-​mediated (T-​cell) and humoral-​mediated (B-​cell) immunity resulting in severe combined immunodeficiency disease. Although purine nucleoside phosphorylase deficiency causes defective T-​ cell-​mediated immunity, these patients may possess either normal, hyperactive, or reduced humoral immunity. Most patients with these enzyme deficiencies present in infancy or early childhood, with se- vere infections caused by pathogens or opportunistic organisms. Adenosine deaminase deficiency About 85% of patients with adenosine deaminase deficiency present as infants with severe combined immunodeficiency disease. Among all severe combined immunodeficiency disease patients, adenosine deaminase deficiency accounts for 10 to 15% of cases. Although ad- enosine deaminase deficiency classically presents in infancy (early onset), a minority of patients have a clinically less severe variant with delayed onset presenting with severe combined deficiency between the ages of 1 and 10 years. Very rarely, patients may presents in the second to fourth decade. The prevalence of adenosine deaminase deficiency has been estimated at between less than 1 in 106 and 1 in 2 × 105 live births. Adenosine deaminase deficiency is inherited in an auto- somal recessive fashion. The diagnosis is made by meas- uring adenosine deaminase activity in erythrocytes and the presence of deoxyadenosine nucleotides in red cell nucleotide pro- files. Heterozygote detection and prenatal diagnosis are best done by genetic characterization. In addition to immunoparesis, clinical feature include growth failure, absent tonsils and lymph nodes, and absence of a thymus shadow on radiography. Characteristic skeletal abnormalities in- clude anterior rib cupping, scapular spurring, and are present at diagnosis in about half of individuals. Sensorineural deafness and other neurological symptoms have been reported, but may be sec- ondary to infections, autoimmunity, or transplantation. The prognosis in untreated severe adenosine deaminase defi- ciency is very poor with death due to multiple recurrent infections during the first year of life. Adenosine and 2′-​deoxyadenosine, derived from the breakdown of DNA due to cell death, accumulate proximal to the metabolic block; 2′-​deoxyadenosine is the primary lymphotoxic precursor in adeno- sine deaminase deficiency and elevated levels are present in urine. Erythrocytes contain markedly raised levels of deoxyadenosine triphosphate and reduced activity of S-​adenosylhomocysteine

12.4  Disorders of purine and pyrimidine metabolism 2029 hydrolase due to inactivation by 2′-​deoxyadenosine. Erythrocyte ATP is reduced. The level of deoxyadenosine triphosphate in erythrocytes correlates with clinical expression and with the level of residual adenosine deaminase activity. There are several mechanisms by which adenosine deaminase deficiency can impair immune function. Accumulation of deoxyadenosine triphosphate can induce apoptosis in lymphoid cells. This may be related to deoxyadenosine triphosphate-​induced inhibition of ribonucleotide reductase blocking DNA replication in dividing cells and to deoxyadenosine triphosphate-​induced DNA strand breaks in nondividing lymphocytes. Deoxyadenosine triphosphate also activates the protease (caspase 9)  involved in apoptosis. S-​adenosylhomocysteine hydrolase blocks S-​ adenosylmethionine-​mediated transmethylation reactions. The for- mation of deoxyadenosine triphosphate from 2′-​deoxyadenosine activates inosine monophosphate dephosphorylation thereby leading to depletion of cellular ATP. It has also been suggested that lymphocyte function may be impaired by aberrant signal trans- duction mediated by deoxyadenosine acting through G-​protein-​ associated receptors or from an altered costimulatory function of T-​cell-​associated adenosine deaminase complexing protein CD26/​ dipeptidyl peptidase IV. Treatment is based on enzyme replacement. Enzyme replace- ment therapy with pegylated bovine adenosine deaminase provides a source of the enzyme to remove the toxic metabolites in the short term until the patient can be transplanted. Allogeneic haemato- poietic stem cell transplant is done if a fully HLA-​matched sibling or family donor is available. Transplants from unrelated or haplo-​ identical donors have been less successful. Haematopoietic stem cell gene therapy to insert a functional adenosine deaminase copy into the genetic material of the patient has recently become available and has shown curative potential. Measurement of deoxyadenosine ­triphosphate levels in red cells is useful for monitoring therapy. Purine nucleoside phosphorylase deficiency Purine nucleoside phosphorylase deficiency occurs less frequently than adenosine deaminase deficiency. In addition to the clinical results of immunoparesis, more than 50% of these patients have neurological abnormalities including disorders of muscle tone, de- layed motor and intellectual development, ataxias, tremors, spastic tetraparesis, and behavioural difficulties. Autoimmune haemolytic anaemia and megaloblastic bone marrow changes have been occa- sional associations. There appears to be a particular susceptibility to virus infection such as varicella, vaccinia, and cytomegalovirus. The tonsils and the thymus are small or absent and the lymph nodes are deficient in the thymus-​dependent areas. Circulating lymphocyte counts are usually very low with a low percentage of T lymphocytes and de- pressed or absent responsiveness to mitogen-​induced transform- ation. Serum immunoglobulin levels and antibody responses to pneumococcal polysaccharide and keyhole limpet haemocyanin are typically increased in these children with purine nucleoside phosphorylase deficiency, and the occasional finding of mono- clonal IgM paraprotein strongly suggests that B-​cell hyperactivity and changes in antibody production are secondary to T-​cell dysregulation. Purine nucleoside phosphorylase deficiency is associated with the accumulation and excretion of 2′-​deoxyguanosine and deoxyinosine as well as guanosine and inosine. Paradoxically there is massive purine overproduction and excretion. Plasma uric acid may be low, but not in all patients. Erythrocyte concentrations of deoxyguanosine triphosphate are markedly raised in purine nucleoside phosphorylase-​deficient cells. T cells but not B cells appear to be particularly susceptible to 2′-​deoxyguanosine toxicity, probably as a result of accumula­ tion of deoxyguanosine triphosphate, inhibition of ribonucleotide reductase, impairment of DNA synthesis, and eventually cell death. There are few reports of successful bone marrow or stem trans- plantation in purine nucleoside phosphorylase-​deficient patients, possibly reflecting an avoidance of high-​risk procedures in children with a later-​onset presentation than seen in adenosine deaminase deficiency. There is, however, increasing evidence that early inter- vention may be beneficial and may also prevent further neurological deterioration. Other conditions Adenosine kinase deficiency has been reported in patients from three families presenting with severe developmental delay and liver dysfunction. Biochemical finding suggested a block in the methio- nine cycle, with plasma methionine, S-​adenosylmethionine, and S-​adenosylhomocysteine all elevated. A defect in deoxyguanosine kinase is associated with mitochon- drial DNA depletion with a predominantly hepatocerebral pheno- type. Liver transplantation may be beneficial in patients in whom neurological disease is absent or mild. Mutations in inosine monophosphate dehydrogenase type 1 (IMPDH1), which together with IMPDH2 catalyse the first step in the conversion of inosine monophosphate to guanosine monophosphate, are associated with autosomal dominant retinitis pigmentosa (RP10 form). Inactivating mutations in the gene NT5C2 encoding cyto- solic purine 5’-​nucleotidase (cytosolic nucleotidase II), AMPD2 encoding adenosine monophosphate deaminase-​2, and ENTPD1 (ectonucleoside triphosphate diphosphohydrolase 1)  are associ- ated with hereditary spastic paraplegias, neurodegenerative motor neuron diseases characterized by progressive age-​dependent loss of corticospinal motor tract function. Disorders of pyrimidine metabolism The pathways of pyrimidine biosynthesis interconversion and deg- radation are shown in Fig. 12.4.9. The de novo synthesis of pyr- imidine nucleotides involves a series of six reactions beginning with the formation of carbamyl phosphate and concluding with orotidine monophosphate, which then undergoes a series of inter- conversion and salvage reactions. The first three steps in the de novo synthesis pathway are encoded in a gene directing the synthesis of the multifunctional protein that encompasses carbamyl phosphate synthetase, aspartate transaminase, and dihydro-​orotase. The fourth step is catalysed by dihydro-​orotate dehydrogenase which is encoded in a single gene. The fifth and sixth steps are catalysed by the gene directing the synthesis of the bifunctional protein encoding orotate phosphoribosyltransferase and orotidine 5′-​monophosphate de- carboxylase, which reside in separate regions of the protein uridine

SECTION 12  Metabolic disorders 2030 monophosphate synthetase (UMPS). The pyrimidines are degraded to β-​alanine and β-​aminoisobutyrate. The inherited disorders of pyrimidine metabolism are much less common and/​or possibly much less easily recognized than disorders of purine metabolism. Dihydro-​orotate dehydrogenase deficiency
(Miller’s syndrome) Dihydro-​orotate dehydrogenase is located on the inner membrane of mitochondria and catalyses the conversion of dihydro-​orotate to orotate, the fourth step in the de novo synthesis of pyrimidines. A deficiency of dihydro-​orotate dehydrogenase causes Miller’s syn- drome or postaxial acrofacial dysostosis syndrome which is charac- terized by micrognathia, orofacial clefts, malar hypoplasia, aplasia of the medial lower lid eyelashes, coloboma of the lower eyelid, and cup-​shaped ears, combined with postaxial limb deformities. Miller’s syndrome was the first Mendelian disorder whose genetic basis was identified by whole-​exome sequencing. Dihydro-​orotate levels are not reported to be elevated in urine. Hereditary orotic aciduria Orotic aciduria is due to point mutations in the gene UMPS directing the synthesis of the bifunctional protein UMP synthase which cata- lyses the last two steps on the pyrimidine biosynthetic pathway. There is massive overproduction of orotic acid due to loss of feed- back inhibition of carbamyl phosphate synthase, which is the first and rate-​limiting step on the metabolic pathway. Orotic aciduria presents during infancy with severe megaloblastic hypochromatic anaemia, orotic acid crystalluria, and occasionally, radiotranslucent orotic acid urinary stones. Cardiac malformations, mild intellectual impairment, and strabismus have been reported. Orotic aciduria is inherited as an autosomal recessive defect. Patients heterozygous for a UMPS mutation may have mildly elevated orotic acid in urine and this may explain some cases of benign hereditary orotic aciduria found on screening. Enzyme assays on erythrocyte lysates may show low levels of orotate phosphoribosyltransferase or orotidine 5′-​monophosphate decarboxylase activity, or both, depending on the location and nature of the underlying genetic defect. Administration of uridine (100–​ 150 mg/​kg per day), which is converted to uridine monophosphate (Fig. 12.4.9), produces a prompt haematological response in those patients with a megaloblastic anaemia. Treatment needs to be started as soon as the diagnosis is made during infancy in order to minimize the possibility of persistent neurological deficits. Orotic aciduria has been found in urea cycle defects, lysinuric protein intolerance, purine nucleoside phosphorylase deficiency, normal pregnancy, and during allopurinol administration. Pyrimidine 5′-​nucleotidase deficiency (uridine 5’-​ monophosphate hydrolase) This autosomal recessive disorder leads to a life-​long nonspherocytic mild to moderate chronic haemolytic anaemia. Uridine tri- phosphate and cytidine triphosphate accumulate in the red cells which show basophilic stippling and reticulocytosis. There is hepatosplenomegaly. The enzyme is assayed in erythrocytes and activities between 0 and 30% of normal have been reported. Developmental delay and learning difficulties have been reported in some patients. Treatment is supportive. Transfusions are rarely required and splenectomy is not indicated. Lead poisoning can also be associated with acquired erythrocyte pyrimidine 5′-​nucleotidase deficiency. Deficiency of dihydropyrimidine dehydrogenase This autosomal recessive disorder presents with a variable pheno- type ranging from microcephaly, hypertonia, epilepsy, learning dif- ficulties, and autism, to mild behavioural abnormalities, even within the same family. Some cases have only been diagnosed during adult life when they have developed life-​threatening or fatal toxicity ­following cancer chemotherapy with 5-​fluorouracil. Uracil and thy- mine are elevated in the urine of completely deficient patients, but not in carriers. Absent enzyme activities have been demonstrated in leucocytes, liver, and fibroblasts. There is no effective treatment for this condition and the prognosis for life is very variable. Partial dihydropyrimidine dehydrogenase deficiency due to a heterozygous variant genotype occurs in approximately 6% of Caucasian popu- lations, is asymptomatic but is associated with severe toxicity to fluoropyrimidine-​based chemotherapy. Deficiency of dihydropyrimidinase Dihydropyrimidinase catalyses the second step of the pyrimi- dine degradation pathway. Patients present with neurological and gastrointestinal abnormalities and markedly elevated levels of 5,6-​ dihydrouracil and 5,6-​dihydrothymine in body fluids. There is con- siderable phenotypic variation, even within families. Patients with dihydropyrimidinase deficiency are also at risk of severe toxicity to fluoropyrimidine chemotherapy. Fig. 12.4.9  Pathways of pyrimidine metabolism in humans. 5′-​NT, pyrimidine 5′-​nucleotidase; CPSH, carbamyl phosphate synthetase II; DHPD, dihydropyrimidine dehydrogenase; NP, pyrimidine nucleoside phosphorylase; ODC, orotidine decarboxylase (OPRT

12.4  Disorders of purine and pyrimidine metabolism 2031 ß-​Ureidopropionase deficiency Ureidopropionase catalyses the conversion of N-​carbamyl-​ß-​ alanine and N-​carbamyl-​ß-​aminoisobutyric acid to ß-​alanine and ß-​aminoisobutyric acid, ammonia, and CO2. Patients pre- sent mainly with neurological abnormalities (intellectual dis- abilities, seizures, abnormal tonus regulation, microcephaly). N-​carbamyl-​ß-​alanine and N-​carbamyl-​ß-​aminoisobutyric acid are markedly elevated in urine and blood. As with the other de- fects of pyrimidine degradation, phenotypic variability of ß-​ ureidopropionase deficiency was demonstrated in one family in which the index patient was clinically affected whereas the same genotype did not lead to overt symptoms in the mother of the patient. Thymidine phosphorylase deficiency Thymidine phosphorylase catalyses the conversion of thy- midine and deoxyuridine to thymine and uracil respectively. A deficiency of the enzyme results in the accumulation of both thymidine and deoxyuridine in blood and urine and results in the mitochondrial depletion syndrome mitochondrial neurogastrointestinal encephalomyopathy (MNGIE). The intra- cellular accumulation of these nucleosides leads to imbalances of mitochondrial deoxynucleotide pools impairing mitochon- drial DNA replication and leading to depletion of mitochondrial DNA. MNGIE is a multisystemic autosomal recessive disorder with onset typically before the age of 30  years, with most pa- tients presenting as children, some within 5  months of birth. Symptoms include ptosis, progressive external ophthalmoplegia, gastrointestinal dysmotility, cachexia, peripheral neuropathy, and leukoencephalopathy. Allogeneic haematopoietic stem cell transplantation is reported to restore enzyme function and improve clinical manifestations in the long term. Thymidine kinase 2 deficiency Mutations in TK2 encoding thymidine kinase 2, which catalyses the phosphorylation of deoxypyrimidine nucleosides in the mito- chondrial matrix, are mainly associated with the myopathic form of mitochondrial depletion syndromes. The phenotypic spectrum ranges from infantile myopathy with motor regression and early death to milder forms of myopathy with prolonged survival, my- opathy with liver involvement, or chronic progressive external oph- thalmoplegia in adults. FURTHER READING Cameron JS, Simmonds HA (2005). Hereditary hyperuricaemia and renal disease. Semin Nephrol, 25, 9–​18. Dalvi SR, Pillinger MH (2013). Saturnine gout, redux: a review. Am J Med, 126, 450.e1–​8. de Brouwer AP, et  al. (2010). PRPS1 mutations:  four distinct syn- dromes and potential treatment. J Hum Genet, 86, 506–​18. Eckardt KU, et  al. (2015). Autosomal dominant tubulointerstitial kidney disease:  diagnosis, classification, and management—​a KDIGO consensus report. Kidney Int, 88, 676–​83. Fam AG (2001). Difficult gout and new approaches for control of hyperuricaemia in the allopurinol-​allergic patient. Curr Rheumatol Rep, 3, 29–​35. Ferraro PM, et al. (2015). A London experience 1995-​2012: demo- graphic, dietary and biochemical characteristics of a large adult co- hort of patients with renal stone disease. QJM, 108, 561–​8. Frampton JE (2015). Febuxostat: a review of its use in the treatment of hyperuricaemia in patients with gout. Drugs, 75, 427–​38. Harkness, et  al. (1988). Lesch-​Nyhan syndrome and its pathogen- esis:  purine concentrations in plasma and urine with metabolite profiles in CSF. J Inherit Metab Dis, 11, 239–​52. Hart TC, et  al. (2002). Mutations of the UMOD gene are respon- sible for medullary cystic kidney disease 2 and familial juvenile hyperuricaemic nephropathy. J Med Genet, 39, 882–​92. Johnson RJ, et al. (2013). Uric acid and chronic kidney disease: which is chasing which? Nephrol Dial Transplant, 28, 2221–​8. Khanna D, et al. (2012). 2012 American College of Rheumatology guide- lines for management of gout. Part 1: systematic nonpharmacologic and pharmacologic therapeutic approaches to hyperuricemia. Arthritis Care Res (Hoboken), 64, 1431–​46. Khanna D, et  al. (2012). 2012 American College of Rheumatology guidelines for management of gout. Part 2: therapy and antiinflam- matory prophylaxis of acute gouty arthritis. Arthritis Care Res (Hoboken), 64, 1447–​61. Merriman TR (2015). An update on the genetic architecture of hyperuricemia and gout. Arthritis Res Ther, 17, 98. Mount DB (2013). The kidney in hyperuricemia and gout. Curr Opin Nephrol Hypertens, 22, 216–​23. Rampoldi L, et  al. (2003). Allelism of MCKD, FJHN, and GCKD caused by impairment of uromodulin export dynamics. Hum Mol Genet, 12, 3369–​84. Rannou F, et al. (2015). Diagnostic algorithm for glycogenoses and myoadenylate deaminase deficiency based on exercise testing parameters: a prospective study. PLoS One, 10, e0132972. Scriver CA, et al. (eds) (2001). The metabolic and molecular basis of inherited disease, 8th edition. McGraw–​Hill, New York. Sperling O (2006). Hereditary renal hypouricaemia. Mol Genet Metab, 89, 14–​18. Torres RJ, Puig JG (2007). Hypoxanthine-​guanine phosopho­ ribosyltransferase (HPRT) deficiency:  Lesch-​Nyhan syndrome. Orphanet J Rare Dis, 2, 48. van Kuilenburg AB, et  al. (2010). Dihydropyrimidinase defi- ciency: phenotype, genotype and structural consequences in 17 pa- tients. Biochim Biophys Acta, 1802, 639–​48. Van Kullenburg AB, et al. (2004). Beta-​ureidopropionase deficiency: an inborn error of pyrimidine degradation associated with neuro- logical abnormalities. Hum Mol Genet, 13, 2793–​801. Watts RWE, et al. (1987). Lesch-​Nyhan syndrome; growth delay, tes- ticular atrophy and a partial failure of 11β-​hydroxylation of steroids. J Inherit Metab Dis, 10, 210–​23. Woodward OM (2015). ABCG2: the molecular mechanisms of urate secretion and gout. Am J Physiol Renal Physiol, 309, 485–​8. Zanella A, et al. (2006). Hereditary pyrimidine 5′-​nucleotidase defi- ciency from genetics to clinical manifestations. Br J Haematol, 133, 113–​23. Zhang W, et  al. (2006). EULAR evidence based recommendations for gout. Part II: Management Report of a task force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT). Ann Rheum Dis, 65, 1312–​24.

12.5 The porphyrias 2032 Timothy M. Cox

12.5 The porphyrias 2032 Timothy M. Cox

ESSENTIALS The porphyrias are a remarkable family of metabolic disorders char- acterized biochemically by overproduction of haem precursors, principally in the liver and bone marrow. The acute porphyrias are inborn errors of varying penetrance that affect enzymatic steps in a tightly regulated biosynthetic pathway for haem; nonacute acquired forms also occur in genetically predisposed individuals. Haem formation for erythropoiesis, while adaptable to physio- logical and pathological changes, is generally constant, but hepatic synthesis of haem undergoes extreme and rapid oscillations to meet dynamic changes in various pathophysiological states, and this in- fluences the clinical expression of porphyria, which is often latent. Acute porphyrias Clinical presentation—​life-​threatening neurovisceral attacks occur in four of the porphyrias:  acute intermittent porphyria, varie- gate porphyria, hereditary coproporphyria, and Doss’ porphyria
(5-​aminolaevulinate dehydratase deficiency). These present with ab- dominal pain, psychiatric symptoms, and signs of sympathetic and hypothalamic autonomic overactivity, sometimes accompanied by convulsions and motor and sensory deficits. They typically de- velop on exposure to environmental or endogenous factors that place a demand for hepatic haem biosynthesis, the most frequent being changes in reproductive steroid hormones either due to nat- ural hormone cycles or the administration of exogenous gonadal steroids, starvation, intercurrent infection, alcohol, and drugs. Acute porphyrias may also be associated with overproduction of photo- active metabolites and thus long-​term photosensitivity, which is ag- gravated during acute attacks. Diagnosis—​this is key to survival of an acute attack of porphyria, which can be suspected on the basis of the past history, in par- ticular of photosensitivity or the intermittent discoloration of urine, and family history, and is confirmed by finding excess water-​soluble haem precursors in urine. Enzymatic studies can later be used to verify the exact type of suspected porphyria, with molecular analysis of genes encoding relevant haem synthetic enzymes used to identify at-​risk individuals in affected pedigrees. Management—​treatment of an acute porphyric attack mandates immediate withdrawal of inappropriate drugs and other precipitating factors; infusions of haem arginate or other licensed preparations of haem shorten life-​threatening episodes and may be effective prophy- laxis for recurrent porphyria in women with periodic attacks. Genetic counselling informed by molecular analysis of cognate genes, and lifetime management advice should be offered to first-​degree rela- tives of patients in whom acute porphyria has been diagnosed. Beyond comprehensive genetic characterization of the acute porphyrias, promising developments in molecular therapy are coming to fruition, including RNA interference-​mediated silencing of the rate-​limiting biosynthetic enzyme, hepatic aminolaevulinate synthase 1. Nonacute porphyrias The nonacute porphyrias are photosensitivity syndromes caused by excess photoactive macrocyclic porphyrins triggered especially by visible light in the blue–​violet range (380–​420 nm). In the most severe form, manifestations are of severe blistering lesions on sun-​exposed skin, particularly of the hands and face, with the formation of vesicles and bullae that may become infected. Healing may lead to loss of digits, scarring of the eyelids, nose, lips, and scalp, and occasionally blindness due to corneal scarring. Protoporphyria, either autosomal due to impaired incorporation of iron or X-​linked due to overexpression of erythroid aminolaevulinate synthase, causes burning pain and erythema with oedema; blistering is absent. Diagnosis is based on finding excess formed porphyrins in blood and excreta. Sunlight exposure should be minimized until the porphyrin abnormality is corrected, for example, by phlebotomy to deplete iron excess that usually aggravates porphyria cutanea tarda, or by liver or haematopoietic stem cell transplantation in some (very rare) cases. The approved synthetic analogue of α-​melanocyte-​ stimulating hormone, afamelanotide, increases skin pigmentation by increasing melanin production and significantly improves sunlight tolerance and quality of life in patients with at least one cutaneous form of (nonacute) porphyria, namely protoporphyria. Introduction The haem biosynthetic pathway holds great fascination for biochem- ists, who marvel at the evolution of ancient enzymes which interact to bring about the formation of the pigments of life, haemoglobin, the cytochromes, chlorophyll, and the cobalamins (vitamin B12). It is 12.5 The porphyrias Timothy M. Cox

12.5  The porphyrias 2033 unfortunate that, because of complexities in their chemical structure and nomenclature, the important diseases associated with their dis- turbed haem metabolism are perceived as obscure. These consider- ations apply particularly to the acute porphyrias, which are rare but distressing syndromes that mimic other acute illnesses but for which recognition may be critical for the patient’s survival; too often their inherited nature is overlooked and the diagnosis is not established until permanent disability (or even death) supervenes. The porphyrias are rare diseases caused by disturbances in the multistep pathway for the formation of haem, a pigment essential for oxygen transfer and the energy-​yielding reactions of electron trans- port. The formation of haem is tightly regulated so that acquired or hereditary defects of any of its component reactions lead to the over- production of haem precursors. Potentially photoactive macrocyclic compounds and toxic precursors of pyrroles thus accumulate. Most of the human porphyria syndromes result from uncommon genetically determined deficiencies of unitary enzymes of the haem biosynthetic pathway, but certain toxins including lead, iron, and hydrocarbons influence the pathway and cause porphyria in sus- ceptible individuals. Similarly, the metabolism of endogenous molecules, including steroid hormones, and xenobiotics including alcohol and many therapeutic drugs, disturb the delicate biochem- ical equilibrium in asymptomatic patients with latent porphyria. Catastrophic gene–​environment interactions in previously fit indi- viduals may precipitate sporadic attacks of acute porphyria. Haem metabolism General perspective In healthy adults, daily formation of haem requires the incorpor- ation of about 25 mg of iron into protoporphyrin IX to generate ap- proximately 275 to 300 mg of haem de novo. About 15% of haem biosynthesis occurs in the liver and 80% in the erythroid marrow, with the main bulk of haem being biochemically coordinated in haemoglobin and, to a lesser extent, as myoglobin in the muscles. The remaining haem is mostly present in enzyme proteins, including catalase, tryptophan pyrrolase, nitric oxide synthase, guanylate cyclase, prostaglandin synthase, nucleotide phosphodiesterases, and cytochromes (in the endoplasmic reticulum cytochrome P450 system as well as the mitochondrial electron transport chain). Hepatic synthesis of haem is subject to rapid and wide fluctu- ations but haem formation in the erythropoietic bone marrow is, under most circumstances, constitutive and stable. However, haem synthesis may be increased, either as the erythron expands and proliferates to meet the demands of blood loss or haemolysis, or in response to intrinsic disease of the bone marrow—​or conditions where there is tissue hypoxia or inappropriate erythropoietic drive (see Chapters 22.6.2 and 22.6.4). Very little free haem is present in the cytosol or plasma compart- ment, in the latter of which it is avidly bound to the glycoprotein, haemopexin. Haem proteins undergo appreciable turnover and the haem component is rapidly degraded by distinct haem oxygenases in many tissues, especially those populated by macrophages. Haem oxygenases break open the porphyrin ring to liberate iron, thereby generating linear pyrroles that are metabolized further to biliverdin and bilirubin pigments with the stoichiometric release of carbon monoxide. Many tissues use this one-​way destructive route to dis- pose of haem, formation of which in health is balanced by fastidious control of de novo biosynthesis. The routes for the formation and deg- radation of haem are distinct (Fig. 12.5.1). Unlike the biosynthesis of other essential building blocks (e.g. cholesterol and other sterols, purines and pyrimidines, sphingolipids, and glycosphingolipids), once formed there is no means to recycle haem or recover its com- plex macromolecular precursors through salvage reactions. Functions and formation of haem Haem serves as a key prosthetic group in haem proteins, including cytochromes, myoglobin, and haemoglobin, by which it fulfils its essential biological roles as a transporter of oxygen and electrons in the respiratory chain. It is also essential for transcription fac- tors that regulate circadian activities, binding reversibly to a nuclear receptor (Rev-​erbα) which is a crucial regulator of the core clock functions in biological rhythms. Haem suppresses the action of Rev-​erbα on expression of proteins involved in maintenance of glu- cose homeostasis and gluconeogenesis. A molecular understanding of these recently discovered interactions holds much promise for elucidating how acute attacks of porphyria are induced by starva- tion, sepsis, endogenous hormonal factors, and xenobiotics. Haem biosynthesis requires eight coordinated reactions that are catalysed by cytoplasmic and mitochondrial enzymes which ex- change substrate intermediates between intracellular compartments (Fig. 12.5.1). 5-​aminolaevulinate synthetase The first committed, essential, and rate-​limiting precursor, 5-​amino­laevulinate, is generated by condensation of glycine and the Krebs cycle intermediate, succinyl coenzyme A. This re- action is brought about by one or other of the two isozymes of 5-​aminolaevulinate synthetase (ALAS): the gene for the ‘hepatic’ enzyme, ALAS1, maps to the autosome chromosome 3 and is ubi- quitously expressed; the gene for the distinct erythroid isoform, ALAS2, is found on the X chromosome and undergoes dosage compensation in females by lyonization. These enzymes them- selves are localized in the mitochondrial matrix. Expression of the Fig. 12.5.1  Main pathways for haem biosynthesis and degradation in humans.

section 12  Metabolic disorders 2034 gene encoding ALAS1 is rate limiting for the formation of haem and is increased in response to a key regulator of mitochondrial biogenesis that stimulates activity of the Krebs cycle, the transcrip- tion factor peroxisome proliferator-​activated receptor coactivator 1α (PGC-​1α). To preclude the build-​up of toxic porphyrin inter- mediates when the supply of iron is restricted, transcription of aminolaevulinate synthase is regulated by the presence of iron-​ binding elements. Terminal differentiation of erythroid cells depends upon tran- scriptional activation of ALAS2 by the erythroid factors, especially GATA1, and activation of ALAS2 appears to be a prerequisite for in- duction of the other genes that encode the enzymes of haem biosyn- thesis. ALAS2 mRNA harbours an iron-​response element that binds iron-​regulatory proteins (IRP1 and -​2), thereby blocking translation of the ALAS2 mRNA; these iron-​sulphur binding proteins are de- stabilized when iron is present and translation can proceed. These elaborate molecular control processes serve to coordinate haem bio- synthesis with iron availability in erythroid cells and in this way the risk of toxic injury from the convergence of highly reactive biochem- ical intermediates generated from different corners of the cellular machinery can be minimized. Pyridoxal 5-​phosphate (derived from vitamin B6) is an essential cofactor for ALAS isozymes. Deficiency of pyridoxine or interfer- ence with its metabolism leads to sideroblastic anaemia. 5-​Aminolaevulinate dehydratase After its formation, 5-​aminolaevulinate is exported to the cyto- plasm where the abundant cysteine-​rich (heavy metal-​sensitive) enzyme, 5-​aminolaevulinate dehydratase, catalyses the formation of hydroxymethylbilane (alternatively known as preuroporphyrinogen), which spontaneously forms the monopyrrole porphobilinogen. 5-​Aminolaevulinate dehydratase is a multimeric enzyme with re- active sulphydryl groups that are particularly sensitive to the toxic effects of heavy metals, especially lead, so that activity of this enzyme is a sensitive measure of environmental and industrial toxicity. The enzyme is inhibited competitively by the metabolite succinylacetone, concentrations of which rise to inhibitory levels in patients who have the defect of aromatic amino acid degradation tyrosinaemia type I. Patients with tyrosinaemia type I and lead poisoning have neurovisceral manifestations that resemble the acute porphyrias, and it appears likely that overproduction of aminolaevulinate, as a result of arrest at the 5-​aminolaevulinate dehydratase reaction, con- tributes to this effect. Porphobilinogen deaminase and uroporphyrinogen III synthetase Four molecules of porphobilinogen are enzymatically condensed to yield the macrocyclic tetrapyrrole uroporphyrinogen III in a com- plex reaction brought about by porphobilinogen deaminase and uroporphyrinogen III synthetase. These enzymes act coordinately to reverse the orientation of one porphobilinogen molecule to yield uroporphyrinogen III, an isoform member of the III series of por- phyrins that are the sole precursors of haem in biological systems. Uroporphyrinogen III decarboxylase, coproporphyrinogen III oxidase, protoporphyrinogen oxidase, and ferrochelatase The cytoplasmic enzyme uroporphyrinogen III decarboxylase decarboxylates the four acetate substituent side chains of uro­porphyrinogen to yield coproporphyrinogen III, which is then reimported into the mitochondrion for further oxidative decarb- oxylation. Coproporphyrinogen III oxidase modifies the two propi- onate side chains to vinyl groups yielding protoporphyrinogen IX, the penultimate precursor of haem. Protoporphyrinogen oxidase removes six hydrogen atoms to yield protoporphyrin IX, which is the substrate for the final step in haem biosynthesis. The insertion of ferrous ions into the porphyrin macrocycle to form ferroprotohaem (haem) is catalysed by the mitochondrial enzyme ferrochelatase. Control of haem synthesis The highly regulated control mechanism of haem biosynthesis en- sures that the free concentrations of the toxic intermediates involved in the pathway are kept low unless there is a metabolic arrest at one of the biosynthetic reactions; under these circumstances, an over- production of the intermediate compounds occurs which can be used for diagnosis. This overproduction predisposes to the develop- ment of the particular clinical porphyric syndrome. A knowledge of the enzymatic steps and of the differential solu- bility of the haem precursors facilitates appropriate diagnostic testing for the precise identification of suspected porphyria. In general, over- production of the early precursors such as 5-​aminolaevulinic acid is a common feature of those syndromes associated with neurovisceral manifestations or acute attacks of porphyria. Aminolaevulinate, in par- ticular, represents a common biochemical marker of such attacks and those syndromes that mimic the acute porphyrias, such as hereditary tyrosinaemia type I and lead poisoning. In patients with cutaneous photosensitivity, overproduction of the formed porphyrin macro- cycles can also be detected in plasma, urine, and faeces in which they are distributed according to their aqueous solubility (Table 12.5.1). The profile of molecules that are overproduced in a given syn- drome may be predicted from the level at which the enzymatic ar- rest occurs as flux through the pathway is stimulated by diminished negative feedback. In those porphyrias where the principal site of production appears to be in the liver, including the acute porphyrias and porphyria cutanea tarda, fluctuations through the biosynthetic pathway as a result of regulatory effects from environmental or en- dogenous factors can occur very rapidly; indeed minute-​to-​minute oscillations in biosynthetic haem fluxes have been recorded in the liver. Thus, in starvation and on challenge with xenobiotic reagents (which place a demand for the production of haem to meet the needs for new cytochrome formation), as well as with endogenous hor- monal changes, enhanced flux through the pathway leads to toxic overproduction of 5-​aminolaevulinic acid. By the same token, rapid repression of the haem biosynthetic pathway in the liver can be in- duced by the administration of exogenous haem, a useful agent in the control of acute attacks and which rapidly corrects the disturbed metabolism (see next paragraph). Table 12.5.1  Solubility and routes of excretion of haem precursors Plasma Urine Faeces 5-​Aminolaevulinate ++ +++ − Porphobilinogen ++ +++ − Uroporphyrins I and III + ++ + Coproporphyrins I and III + + +++ Protoporphyrin IX + − +++

12.5  The porphyrias 2035 Haem formation in the erythron is more rapid than that in the liver but is not subject to sudden oscillations in synthetic rates. Nonetheless, in patients with erythropoietic porphyrias such as congenital porphyria, enhanced rates of red cell destruction when hypersplenism supervenes or in response to light exposure greatly exacerbate the overproduction of porphyrin intermediates and aggravate photosensitivity due to increased porphyrin release. Short-​term experiments indicate that exogenous haem may par- tially repress the endogenous haem biosynthetic pathway in eryth- roid tissue, but this has not proved to be useful for long-​term relief in the erythropoietic porphyrias. Blood transfusion to suppress erythropoiesis or definitive replacement of bone marrow by trans- plantation has, however, proved to be successful in controlling the devastating manifestations of congenital erythropoietic porphyria. Classification and epidemiology of the porphyrias The porphyrias have been classified into hepatic and erythropoietic types depending on the main site of overproduction of haem pre- cursors. For clinical purposes, however, a useful operational defin- ition of the porphyric syndromes distinguishes the acute from the nonacute porphyrias. Classification Acute porphyrias cause life-​threatening neurovisceral manifestations which are typically precipitated by sporadic environmental factors. All but one of these disorders is inherited as a dominant condition, but a striking feature is their clinical heterogeneity with great variation in expressivity and penetrance—​the latter often termed ‘latency’. Nonacute porphyrias are characterized by photosensitivity syn- dromes due to overproduction of macrocyclic porphyrins which cause light-​induced skin injury. Several of the acute porphyrias may also cause overproduction of porphyrin intermediates that are either intrinsically fluorescent or readily oxidized to become fluorescent. These porphyrias may at times be accompanied by marked photo- sensitivity and blistering skin reactions, which are usually exacer- bated during the acute attacks. Epidemiology The frequency and epidemiology of the porphyrias are areas of con- tinued uncertainty: estimated prevalence of all porphyrias is from 1 in 300 to 1 in 200 000 in different populations. Difficulties arise because there are many types of porphyria and, being rare, often episodic, and of low penetrance, they frequently escape formal diagnosis. It is also regrettable that biochemical complexity and scientific nomenclature intimidates many practitioners. These con- siderations also impede proper management, especially since costly technological sophistication and specialist referral is often needed for definitive diagnosis. Patients referred with skin manifestations to dermatologists are more readily identified, but the associated or subsequent neurovisceral manifestations in underlying porphyrias such as variegate or coproporphyria are not always recognized in the nonacute clinical context. Acute porphyrias There are regional variations in reported frequency of acute inter- mittent porphyria. A prospective study of newly diagnosed patients with genetic porphyrias from 11 European countries suggested an annual incidence for symptomatic acute porphyria of 0.2 per mil- lion. The incidence of symptomatic acute intermittent porphyria was similar in all countries (0.13 per million per year; 95% confi- dence interval 0.10–​0.14), excepting Sweden (0.51; 95% confidence interval 0.28–​0.86). Prevalence of overt acute intermittent porphyria was 5.4 cases per million. Higher figures have been reported in other studies: one in Norway gave the prevalence of acute intermittent porphyria to be approximately 4 in 100 000, with an overall annual incidence of 0.5 to 1 in 100 000, and the condition occurs in about 1 in 1000 of the Lapp population that is shared between Sweden, Finland, and Norway. Acute variegate porphyria is most common in the Afrikaner population of South Africa:  traced to a Dutch settler in the 17th century, it has a prevalence of about 3 in 1000 persons. The condition has spread to all ethnic groups within the South African population, molecular analysis of which confirms the presence of a single dominant missense mutant allele of the protoporphyrinogen IX oxidase gene (p.R59W). There are iso- lated reports of high prevalence among the potters in the Bikaner district of Rajasthan, India. Nonacute porphyrias The most common condition in this category is porphyria cutanea tarda, with an estimated prevalence of approximately 1 in 10 000 in Norway. Protoporphyria, the second most frequent of the nonacute porphyrias, in which systemic features with cholestatic liver disease are rare (5%), is estimated to have a frequency of 1 in 75 000 to 1 in 200 000 births. Tables 12.5.2 to 12.5.4 set out the individual defects that charac- terize the clinical porphyrias and summarize the clinical features of these hereditary syndromes. Table 12.5.2  The porphyria syndromes Hereditary porphyria Acute porphyrias: • Acute intermittent porphyria • Variegate porphyriaa • Hereditary coproporphyriaa • Doss’ porphyria—​aminolaevulinate dehydratase deficiency Nonacute porphyrias: • Congenital erythropoietic porphyria—​Gunther’s disease • Protoporphyria • X-​linked protoporphyria • Porphyria cutanea tardab—​sporadic or familial • Hepatoerythropoietic porphyriac Acquired porphyria • Hexachlorobenzene porphyria • Lead poisoning (plumboporphyria) • Hereditary tyrosinaemia type 1 a Acute syndromes also accompanied by long-​term skin photosensitivity. b Porphyria cutanea tarda is not a simple monogenic disorder; it is almost always provoked by environmental agents such as hepatitis C, oestrogens, iron excess, or alcohol. c Homozygous uroporphyrinogen III decarboxylase deficiency.

section 12  Metabolic disorders 2036 Table 12.5.3  Main biochemical abnormalities in the porphyrias Disorder Enzyme defect Biochemical abnormality Acute intermittent porphyria Porphobilinogen deaminase Increased urinary porphobilinogen and 5-​aminolaevulinate Variegate porphyria Protoporphyrinogen IX oxidase Increased urine 5-​aminolaevulinate and porphobilinogen (especially acute attacks) Increased stool coproporphyrin III and protoporphyrin Hereditary coproporphyria Coproporphyrinogen III oxidase Increased urine 5-​aminolaevulinate and porphobilinogen (especially acute attacks) Increased stool coproporphyrin III more than protoporphyrin Doss’ porphyria Aminolaevulinate dehydratase Increased urinary 5-​aminolaevulinate Faecal porphyrins normal Porphyria cutanea tarda Uroporphyrinogen III decarboxylasea Increased urine uroporphyrin I and III Increased faecal hepatacarboxylic porphyrin and isocoproporphyrin Congenital erythropoietic porphyria Uroporphyrinogen III synthase Increased urine, plasma, and red cell uroporphyrin I and coproporphyrin I Normal 5-​aminolaevulinate and porphobilinogen Increased faecal coproporphyrin I Protoporphyria Ferrochelatase Increased protoporphyrin in stool and red cells (metal free) Caseinolytic mitochondrial matrix peptidase chaperone subunit (ClpXP) Increased protoporphyrin in stool and red cells (metal free) X-​linked protoporphyria Aminolaevulinate synthase 2 Increased protoporphyrin in stool and red cells (heterozygous females mosaic) Hexachlorabenzene porphyria Uroporphyrinogen III decarboxylase Increased urinary uroporphyrin I and III and hepatocarboxylic and other acetic acid substituents Hereditary tyrosinaemia I Aminolaevulinate dehydrataseb (acquired deficiency) Increased urinary 5-​aminolaevulinate and succinylacetone (toxic metabolite) Lead poisoning Aminolaevulinate dehydratase, ferrochelatase ± impaired iron delivery from transferrin Increased urinary 5-​aminolaevulinate, raised red cell protoporphyrin and zinc protoporphyrin a Homozygous deficiency also responsible for hepatoerythropoietic porphyria. b Inborn deficiency of fumarylacetoacetate hydrolase leads to excess formation of the 5-​aminolaevulinate hydratase inhibitor, succinyl acetone (4,6-​dioxoheptanoate). Reference ranges: urine—​total porphyrins 20 to 320 nmol/​litre; 5-​aminolaevulinate <52 μmol/​litre (urine:creatinine porphobilinogen ratio <1.5); porphobilinogen <10.7 μmol/​litre. Faeces—​total porphyrins 10 to 200 nmol/​g dry weight. Red cell—​total porphyrins 0.4 to 1.7 μmol/​litre. Laboratory ranges supplied by Porphyria Service, Department of Medical Biochemistry, University Hospital of Wales NHS Trust, Heath Park, Cardiff CF4 4XW (Professor G.H. Elder). Table 12.5.4  Principal manifestations of the porphyrias Acute intermittent porphyria Acute neurovisceral attacks Variegate porphyria Acute neurovisceral attacks Skin photosensitivity with fragility, scarring, hairiness, and pigment changes Hereditary coproporphyria Acute neurovisceral attacks, blistering skin lesions, photosensitivity Doss’ porphyria Acute neurovisceral attacks, susceptibility to lead exposure. (Only six confirmed cases reported) Porphyria cutanea tarda Blistering skin lesions on light exposure, pigment changes, atrophy, and scarring—​ also may be associated with manifestations of iron storage disease, also hepatitis C Congenital erythropoietic porphyria Haemolytic anaemia, hypersplenism, porphyrinuria, extreme photosensitivity with skin ulceration and injury; adult or late onset reported Hepatoerythropoietic porphyria Resembles congenital erythropoietic porphyria: blisters, photosensitive skin with scar formation, haemolysis, red urine Protoporphyria and X-​linked protoporphyria Photosensitivity; early-​onset, characterized by burning pain, oedema—​scarring rare X-​linked disease Occasional cholestatic liver disease, protoporphyrin gallstones—​fulminant or subfulminant hepatic failure complicated by neurovisceral syndrome, especially in perioperative state Hexachlorobenzene porphyria Resembles sporadic porphyria cutanea tarda Lead poisoning Neurovisceral manifestations with signs of disordered red cell haemoglobinization Hereditary tyrosinaemia I Toxic neurovisceral disease

12.5  The porphyrias 2037 Pathogenesis Acute neurovisceral attacks These attacks occur in four of the porphyrias indicated in Tables 12.5.2 to 12.5.4. In all but one, the very rare Doss’ porphyria (aminolaevulinate dehydratase deficiency, a recessive disease), the inheritance is as an autosomal dominant trait. Clinical expression is characterized by recurrent acute, life-​ threatening attacks of neuropathy that include abdominal pain, psy- chiatric symptoms including anxiety, and signs of sympathetic and hypothalamic autonomic overactivity with tachycardia and systemic hypertension. The illness is sometimes accompanied by convulsions and principally motor rather than sensory deficits; the neurological features may be confused with Guillain–​Barré syndrome. Acute decompensation is characteristically precipitated by drugs which induce hepatic haem formation and are metabolized by the hepatic cytochrome P450 system located in the endoplasmic re- ticulum. Neuropathological examination shows axonal degener- ation and central chromatolysis in anterior horn cells and in the brain. Electromyography may reveal denervation compatible with a primary axonal neuropathy of peripheral nerves. Although psy- chiatric symptoms often accompany the attacks, there is conflicting evidence that patients with acute porphyria have a greater risk of chronic psychiatric illness than other patients with long-​standing episodic illnesses that affect the quality of life and are associated with physically disabling manifestations. Role of 5-​aminolaevulinic acid The porphyric crises of 5-​aminolaevulinate dehydratase deficiency are associated with lone overproduction of 5-​aminolaevulinic acid, which is common to all porphyrias associated with acute manifest- ations. It has been suggested that toxic effects of other factors may contribute, but evidence supports the view that 5-​aminolaevulinic itself, or a direct metabolite of this critical precursor of porphyrin synthesis, has a causal role in the acute porphyric attack. 1. Mutations in aminolaevulinate synthase 1 (ALAS1) have not been causally associated with human diseases. Mutations in the erythroid isozyme gene, ALAS2, cause two diseases: inactivating or low-​activity mutations, as with deficiency of the cofactor pyridoxal 5-​phophate, cause sideroblastic anaemia; mutations at the C-​terminus of the gene are responsible for X-​linked nonacute protoporphyria, in which there is a common four-​nucleotide base deletion. 2. Acute, ‘nonsurgical’ abdominal pain, encephalopathy, and per- ipheral motor mononeuropathy are typical features of toxic lead exposure, which is associated with specific acquired bio- chemical abnormalities including elevated 5-​aminolaevulinic excretion together with elevated free protoporphyrin—​changes predicted by the effects on aminolaevulinate synthase and ferrochelatase, respectively (both sulphydryl-​rich enzymes that are readily susceptible to inhibition by lead). As an addendum to these observations, it has been shown that individuals who are otherwise asymptomatic heterozygotes, identified from studies of pedigrees with 5-​aminolaevulinate dehydratase deficiency (‘Doss’ porphyria’), are themselves exquisitely sensitive to en- vironmental lead exposure, thereby developing symptomatic plumboporphyria. 3. Hereditary tyrosinaemia type 1 is an inborn error of tyrosine metabolism due to the deficiency of fumarylacetoacetate hydro- lase:  there is liver disease (with a high risk of hepatocellular carcinoma), renal tubular disease, and neurological manifest- ations. Acute neurological crises, with abdominal symptoms, painful extremities, and hypertension with hyponatraemia, occur episodically and may lead to fatal respiratory failure—​ widely referred to as porphyria-​like syndrome. When man- aged by dietary restriction of tyrosine and phenylalanine, these episodes persist, but the illness is characterized by ele- vation of the secondary metabolite succinylacetone (5,7-​ dioxoheptanoate) and 5-​aminolaevulinate concentrations in plasma and urine. Succinylacetone is a potent noncompeti- tive inhibitor of 5-​aminolaevulinic dehydratase (Ki 30 nM), thus explaining the accumulation of 5-​aminolaevulinate. The neurological crises in tyrosinaemic patients resolve rapidly on introduction of nitisinone, an inhibitor of 4-​ hydroxyphenylpyruvate dioxygenase which catalyses conver- sion of 4-​hydroxyphenylpyruvate to homogentisic acid, thus blocking the proximal tyrosinaemia pathway and correcting the pathological generation of succinylacetone. If started early, the clinical manifestations are ameliorated but the long-​term neurocognitive outcomes are often impaired. The structure of aminolaevulinate is analogous to the inhibi- tory neurotransmitters γ-​aminobutyric acid and l-​glutamate. It seems likely that 5-​aminolaevulinate may interfere with the ac- tion of the γ-​aminobutyric acidergic system, the best evidence for which appears to be its ability to inhibit melatonin production in the rat pineal gland in vivo, as has been described in patients with recurrent acute porphyric attacks. It has been further postulated that under the conditions of the acute attack there may be a de- ficiency of essential haem proteins, such as the cytochrome P450 isozymes in the liver, with further disturbances in secondary me- tabolism; other possibilities include a decrease in the activity of hepatic tryptophan dioxygenase, leading to increased formation of 5-​hydroxytryptamine (serotonin). At present there is no clear resolution between combined or in- dividual effects of acute porphyria on the production of neurotoxic pseudotransmitters (aminolaevulinate) or secondary local deficiency of haem. However, the beneficial results of liver transplantation in pa- tients with disabling recurrent attacks of acute intermittent porphyria indicate that the principal cause of the acute syndrome is the hepatic over- production of toxic haem precursors. In any event, there is convincing evidence of abnormal neurotransmitter function and increased sero- tonin production, as well as direct interference of γ-​aminobutyric acid receptors by toxic concentrations of 5-​aminolaevulinate. Supplying ex- ogenous haem during the acute attack, however, would be expected to correct both arms of this disturbed metabolism, which may account for the beneficial biochemical and clinical effects observed with its use. The recent development of a mouse model of porphyrinogen deaminase deficiency showing sensitivity to barbiturates serves as an authentic model of the biochemical and neuropathological mani- festations of acute porphyria and may clarify much about the patho- genesis. Finally, decisive evidence for causal role of aminolaevuinate or a related metabolite has emerged from the spectacular clinical and biochemical success of short-interfering RNA molecules (siRNAs) that are able to modulate synthesis and abundance of ALAS1.

section 12  Metabolic disorders 2038 The findings in mice that model acute intermittent porphyria have been decisively translated into a human RNA interference agent (givosiran) that specifically targets hepatic ALAS1 gene expression. Not only was this drug able to prevent the biochemical abnormal- ities, it markedly reduced the paralysis associated with barbiturate challenge in affected animals and in humans. Photosensitivity In living cells, most of the macrocyclic precursors of the haem biosyn- thetic pathway are present as their reduced porphyrinogen precursors which are not photoreactive. However, when these tetrapyrroles (uroporphyrinogen, coproporphyrinogen, and proto­porphyrinogen) are produced in excess, they diffuse into plasma and tissues where they react with ambient oxygen to form their parent porphyrins, which are spectacularly fluorescent. Porphyrins absorb light maximally in the Soret region (400–​420 nm) and between 500 and 600 nm (both within the visible light range of 380–​700 nm); they re-​emit this light energy at lower wavelengths to give pink, orange, or red fluorescence. The double-​bond resonance structure of these macrocyclic compounds promotes the formation of singlet oxygen by the transfer of absorbed energy to ground-​state oxygen through light activation, and it appears that generation of singlet oxygen brings about the photodermatoses associated with the porphyrias. Porphyrias associated with overpro- duction of formed macrocyclic haem precursors are thus associated with photosensitivity. The particular skin reactions that develop differ between the particular enzyme defects, which may be explained prin- cipally by the degree of hydrophobicity of the overproduced porphy- rins and their solubility in cellular membranes. The first tetrapyrrole that serves as an immediate precursor to haem is uroporphyrinogen III, formation of which requires coordinated action of the two cytoplasmic enzymes uroporphyrinogen I synthase (porphobilinogen deaminase) and uroporphyrinogen III cosynthase. In the absence of adequate cosynthase activity there is a marked overproduction of porphyrins of the I series, which do not form bio- logically active ferroprotohaem. Deficiency of uroporphyrinogen III cosynthase leads to the very rare but disabling syndrome of Gunther’s disease (congenital erythropoietic porphyria), characterized by ex- treme photosensitivity, haemolysis, and the passage of pink urine containing abundant porphyrins of the I isoform. Persistently high concentrations of these toxic molecules in body fluids lead to staining of the teeth and bones and extreme photosensitive damage, often with cruel and painful skin disfigurement and hair loss. Porphyria cutanea tarda is caused by deficiency of uro­ porphyrinogen decarboxylase, defects of which involve complex interactions between heredity and environmental factors. The en- zyme activity is markedly decreased in the presence of excess tissue iron and, although rare familial cases of porphyria cutanea tarda occur, most patients have a sporadic disease that is provoked by exposure to environmental toxins such as alcohol, oestrogens, hy- drocarbons, iron (often associated with mutations in the haemo- chromatosis gene HFE), and hepatitis C.  At the time of writing, the pathogenic relationship between these external factors and the manifestations of uroporphyrinogen decarboxylase deficiency is unclear. Skin biopsies show subepidermal bullas and electron mi- croscopy reveals vacuoles in the cells of the superficial dermal epi- thelium. In this disease, as in protoporphyria, the endothelium of the dermal capillary is thickened and the vessels are surrounded by complement and mucopolysaccharide deposits. The final step in the haem biosynthetic pathway involves insertion of ferrous iron into the protoporphyrin nucleus generated enzymatically from protoporphyrinogen IX by protoporphyrinogen IX oxidase. This last step occurs in the mitochondrion. Ferrochelatase depends on the iron–​transferrin cycle for the delivery of iron from plasma transferrin. In the bone marrow, when the iron supply is deficient, freely available zinc may be preferentially converted to zinc protoporphyrin ra- ther than ferroprotohaem, thus offering a convenient means to monitor iron-​deficient erythropoiesis. Similarly, industrial lead exposure, which inhibits both iron delivery and the activity of the sulphydryl enzyme ferrochelatase, causes accumulation of zinc protoporphyrin and free protoporphyrin in erythroid precursors and reticulocytes. Deficiency of ferrochelatase leads to the accu- mulation of free protoporphyrin in liver tissue, plasma, and the skin where it induces marked photosensitivity. The accumula- tion of excess protoporphyrin in red cell precursors leads to the characteristic fluorocytes (young red cells containing excess free protoporphyrin) that are the easily recognized hallmark of patients with burning photosensitivity caused by protoporphyria. In protoporphyria, an adequate oxygen supply has been shown to be critical for the development of experimental phototoxicity in vivo. Singlet oxygen and other radicals may lead to lipid peroxidation and cross-​linking of membrane proteins with acti- vation of late complement components. In the more severe dis- ease, congenital erythropoietic porphyria, egress of uroporphyrin I from circulating erythrocytes, which may be destroyed within capillaries, leads to gross accumulation of porphyrin in dermal tissue and juxtaposed epithelium. Exposure to light is known to promote photohaemolysis, indicating that light of the visible wavelength can penetrate the skin sufficiently to induce porphyrin photoactivation in situ. Induction of acute porphyric attacks Acute attacks of porphyria may be life-​threatening illnesses that occur in genetically predisposed individuals who usually remain asymp- tomatic. The acute episodes develop on exposure to environmental or endogenous factors that place a demand for hepatic haem biosyn- thesis; this leads to the overproduction of porphyrin intermediates and pyrrole precursors. The most frequent precipitating factors are changes in reproductive steroid hormones either due to natural hor- mone cycles or the administration of exogenous gonadal steroids. Starvation, including that associated with surgical procedures and anaesthesia, intercurrent infections, and many xenobiotics, including alcohol as well as prescription drugs, over-​the-​counter agents, and chemicals present in health foods can precipitate acute porphyria. Boxes 12.5.1 and 12.5.2 list drugs that have been classified as unsafe in patients with porphyria, either because they have been shown to be porphyrinogenic in animals or in vitro studies, or have been associated with acute attacks in patients with porphyria. These tables are taken from the British National Formulary published by the British Medical Association and the Royal Pharmaceutical Society of Great Britain. It is noteworthy that slight changes in the chemical structure can lead to marked differences in the ability of the drug to induce attacks of porphyria. Inspection of contemporary national and international websites provides up-​to-​date information of immediate relevance and value, especially in relation to local pre- scribing practice (see ‘Further reading’). A clinically more useful list

12.5  The porphyrias 2039 is of safe drugs is provided in Box 12.5.3. For further information see the ‘Complete 2018 List of Safe Drugs in Acute Porphyrias’ (http://​ www.drugs-​porphyria.org/​), which provides an updated current list with informative notes. Tolerance of alcohol varies greatly in patients with porphyria, many of whom appear to tolerate it in modest amounts. Alcohol is, however, best avoided, although at the same time it is wise not to implicate al- cohol in an acute attack unless other causes have been excluded. There is emerging evidence that cigarette smoking, which induces enzymes of the haem-​rich cytochrome P450 system, is prevalent in patients who have frequent acute attacks of porphyria. The author recommends that persuasive advice to stop smoking is given when- ever possible. Acute attacks of porphyria occur in the four conditions known as the hepatic porphyrias and particularly occur for the first time in latent carriers who are aged between 15 and 40 years. Attacks have been recorded in children before puberty but are very rare and usu- ally occur during febrile illnesses or are precipitated by the use of porphyrinogenic drugs and over-​the-​counter remedies. Although the porphyrias occur in a latent state in men with a frequency that is equal to that in women, women who have acute porphyria out- number men by at least two to one. The recent description of Rev-​erbα, a haem sensor involved in the coordination of metabolic pathways and circadian rhythms, as well as PGC-​1α, a transcriptional coactivator involved in the regulation of ALAS1 expression, offers the hope that a better understanding of the mechanism by which environmental influences trigger acute porphyria in susceptible individuals will be forthcoming. Genetic variation in these pathways, particularly the cytochrome P450 system also may go some way to explain the immense variation that individuals show in their susceptibility to the attacks. Clinical features of acute porphyria The clinical manifestations of an acute attack are very diverse and the condition may be indistinguishable from many other disorders. The common neurovisceral symptoms of acute porphyric attacks are listed in Box 12.5.4 and, of these, abdominal pain is the most common presenting symptom. The pain itself may be difficult to identify since it is usually constant but poorly localized and usually not associated with tenderness. Beyond severe myalgic pains, some- times affecting anterior abdominal muscles, there may be an associ- ated colicky component and later ileus with abdominal distension, which may mimic a surgical emergency. Constipation is a charac- teristic symptom but diarrhoea with increased borborygmi can also occur. The patient is usually markedly distressed with anxiety and tachycardia. Marked arterial hypertension (not always previously noted), sometimes paroxysmal, is the rule Development of pain in the limbs is a frequent feature, particu- larly in the upper thighs and also in other somatic muscles of the chest, lumbar region, shoulders, and neck. Ultimately, muscle weak- ness and respiratory paralysis may occur. The patient, almost in- variably very anxious, becomes restless, frankly disturbed, or even deluded as in a toxic confusional state. Prominent mental disturb- ances are reported in patients with the posterior encephalopathy syndrome and cerebral vasculopathy that accompanies the parox- ysmal and severe systemic hypertension of untreated attacks. The inability of attending medical personnel to identify the cause of the pain and the distress associated with it often leads to alienation and an exaggeration of the complaints, which may be difficult to diag- nose. Should a suggestion of psychiatric illness (typically some type of somatic symptom disorder) be made by attending staff, this in- variably compounds the distress experienced by the patient. From every aspect, distressed patients with suspected porphyria should be treated attentively and every effort should be made to accelerate the definitive diagnosis and treat accordingly It is a general rule that the mental features of porphyria subside rapidly as the biochemical disturbance resolves. Hypertension, sweating, and tremor together with tachycardia indicate marked sympathetic overactivity and cardiac arrhythmias may ensue. In about 10% of severe attacks, grand mal seizures de- velop, treatment of which may prolong the attack since many anti- convulsants are highly porphyrinogenic. With sustained attacks, there may be signs of a peripheral neuropathy that is related to axonal degeneration, principally affecting motor nerves. Peripheral neuropathy in its early stages may not affect the limb and tendon re- flexes, but with time these will be decreased or absent. In prolonged porphyric attacks, an ascending muscle weakness rapidly affecting the respiratory muscles and diaphragm, and with bulbar paralysis, may lead to ventilatory failure and death if life-​saving cardiorespira- tory resuscitation and intensive care measures are delayed. In a full-​blown attack, mental symptoms including anxiety, sleep- lessness, and depression are often prominent; the terrifying nature of the illness only aggravates distress in the patient. If the porphyric attack is sustained as a result of failed diagnosis or inadequate man- agement, progressive alienation, visual and auditory hallucinations, and frank paranoia with homicidal outbursts may occur. Such dis- turbances are difficult to contain within the environment of the busy acute hospital. Box 12.5.1  Classes of drug that are unsafe in acute porphyrias • Anabolic steroids • Antidepressants, monoamine oxidase inhibitors (contact the United Kingdom Porphyria Medicines Information Service (UKPMIS) for advice) • Antidepressants, tricyclic and related (contact UKPMIS for advice) • Barbiturates (includes primidone and thiopental) • Contraceptives, hormonal (for detailed advice contact UKPMIS or a porphyria specialist) • Hormone replacement therapy (for detailed advice contact UKPMIS or a porphyria specialist) • Imidazole antifungals (applies to oral and intravenous use; topical antifungals are thought to be safe due to low systemic exposure) • Non-​nucleoside reverse transcriptase inhibitors (contact UKPMIS for advice) • Progestogens (for detailed advice contact UKPMIS or a porphyria specialist) • Protease inhibitors (contact UKPMIS for advice) • Sulphonamides (includes co-​trimoxazole and sulfasalazine) • Sulphonylureas (glipizide and glimepiride are thought to be safe) • Taxanes (contact UKPMIS for advice) • Triazole antifungals (applies to oral and intravenous use; topical antifungals are thought to be safe due to low systemic exposure) For further information see UK Porphyria Welsh Medicines Information Service (2017): https://​www.wmic.wales.nhs.uk/​specialist-​services/​drugs-​in-​porphyria.

section 12  Metabolic disorders 2040 Although seizures may be a presenting sign of the acute attack, they often occur in association with fulminant hyponatraemia re- sulting from the inappropriate secretion of antidiuretic hormone. Treatment of hyponatraemia in the acute attack poses special difficulties (see ‘Hyponatraemia and seizures’). Inappropriate use of hypotonic dextrose will aggravate hyponatraemia and seiz- ures and may induce fatal cerebral herniation due to severe brain oedema. Box 12.5.3  Drug classes thought to be safe in acute porphyrias • Antihistamines: cetirizine, chlorpheniramine, and cyclizine • Diuretics:  acetazolamide, amiloride, bumetanide, cyclopenthiazide, and triamterene • Ergot derivatives: oxytocin is probably safe • Sulphonylureas: glipizide • Analgesics:  morphine, diamorphine, codeine, dihydrocodeine, fen- tanyl, and pethidine. • Tranquillizers: chlorpromazine, and haloperidol • Local anaesthetics:  bupivacaine and lignocaine can be used with caution • Antimicrobials: rifamycins have been used without ill effect in some patients Box 12.5.4  Clinical manifestations of acute porphyria • Abdominal pain • Vomiting • Constipation • Limb, head, neck, and chest pain • Muscle weakness • Sensory loss • Hypertension • Tachycardia • Convulsions • Respiratory paralysis • Fever • Psychiatric symptoms Box 12.5.2  Individual drugs unsafe in acute porphyrias • Aceclofenac • Alcohol • Amiodarone • Aprepitant • Artemether with lumefantrine • Bexarotene • Bosentan • Busulfan • Carbamazepine • Chloral hydrate (although evidence of hazard is uncertain, manufacturer advises avoid) • Chloramphenicol • Chloroform (small amounts in medicines probably safe) • Clemastine • Clindamycin • Cocaine • Danazol • Dapsone • Diltiazem • Disopyramide • Disulfiram • Ergometrine • Ergotamine • Erythromycin • Etamsylate • Ethosuximide • Etomidate • Flutamide • Fosaprepitant • Fosphenytoin • Griseofulvin • Hydralazine • Ifosfamide • Indapamide • Isometheptene mucate • Isoniazid (safety uncertain, contact UKPMIS for advice) • Ketamine • Mefenamic acid (safety uncertain, contact UKPMIS for advice) • Meprobamate • Methyldopa • Metolazone • Metyrapone • Mifepristone • Minoxidil (safety uncertain, contact UKPMIS for advice) • Mitotane • Nalidixic acid • Nitrazepam • Nitrofurantoin • Orphenadrine • Oxcarbazepine • Oxybutynin • Pentazocine • Pentoxifylline • Pergolide • Phenoxybenzamine • Phenytoin • Pivmecillinam • Pizotifen • Porfimer • Raloxifene • Rifabutin (safety uncertain, contact UKPMIS for advice) • Rifampicin • Riluzole • Risperidone • Spironolactone • Sulfinpyrazone • Tamoxifen • Temoporfin • Thiotepa • Tiagabine • Tibolone • Topiramate • Toremifene • Trimethoprim • Valproate • Verapamil • Xipamide

12.5  The porphyrias 2041 Acute attacks of porphyria appear to be more common in women as a result of changes in sex steroids: many women who have peri- odic attacks do so in the 1 or 2 days before the onset of menstrual bleeding, but sometimes attacks lasting a day or two may have their onset in the mid-​menstrual phase soon after ovulation. The pattern may worsen as the menopause approaches, but severe attacks usually cease with the onset of oligomenorrhoea or amenorrhoea. Although it appears that progestogens are principally respon- sible for cyclical or periodic attacks in women and are more porphyrinogenic than oestrogens, pregnancy itself is not invari- ably associated with adverse outcomes in women at risk from acute attacks. Seizures and hypertension due to acute porphyria may be attributed erroneously to eclampsia. Drugs that provoke attacks, such as metoclopramide, may be used mistakenly to control gastro- intestinal symptoms in pregnancy and thus place the woman and her unborn infant at risk. Many mild attacks of porphyria resolve spontaneously within a few days, either as a result of withdrawal of the precipitating factor or because of natural hormonal rhythms. Prolonged attacks are usu- ally the consequence of multiple factors and delays in the institution of definitive therapy. The ensuing neurological injury, accompanied in severe attacks by bulbar and respiratory paralysis, may lead to prolonged or permanent disability. Experience shows that in many such cases inappropriate drugs have been given to counter the early manifestations of the condition (e.g. analgesics, psychotropic drugs and anticonvulsants), hence the initiating medical interventions ultimately prove to be critical determinants of outcome where the diagnosis is not suspected or, if known, has been perilously ignored. Diagnosis of acute attacks Diagnosis of the acute attack is often suspected from scrutiny of the past history, including episodes of photosensitivity with blistering lesions or intermittent discolouration of urine. The passage of frank wine-​coloured or permanganate-​coloured urine is unusual, but if present indicates a full-​blown established attack. The family history is often informative, with a history of abdominal pain in first-​degree family members, with or without photosensitivity skin eruptions. In all instances, it is the overproduction of haem precursors that characterizes the condition biochemically and this is the principal means by which a diagnosis can be made during an acute attack, confirmation of which requires the demonstration of increased por- phyrin precursors in the urine. Most commonly, increased excretion of the monopyrrole, porphobilinogen, is accompanied by increased excretion of urinary 5-​aminolaevulinate, but porphobilinogen ex- cretion is not increased in the extremely rare aminolaevulinate dehydratase deficiency or in the pseudoporphyria of lead poisoning. Without genetic characterization in a first-​degree relative, ‘cold’ diagnosis can be very challenging when the biochemical abnor- malities have corrected themselves. Future developments in gen- etic diagnosis may well improve the interval diagnosis of acute porphyrias, but more work is needed before information from fo- cused sequencing of the responsible genes can be used appropriately to facilitate clinical diagnosis of these important conditions. Prognosis An early series showed that during the first acute attack of porphyria half the patients died. However, perhaps as a result of better hospital facilities to deal with severe or adverse outcomes, the mortality and effects of the disease in patients with acute attacks have improved. Reports from a single centre reported that about three-​quarters of patients with acute intermittent porphyria or variegate porphyria were able to lead normal lives after an acute attack. Recurrent attacks of pain occurred only in a minority during a period of prolonged follow-​up; these recurrent attacks were most likely to occur in the first 3 years. The development of national centres for the treatment of por- phyria, the early detection of genetic predisposition in at-​risk first-​ degree relatives, and the dramatic reduction in prescriptions of porphyrinogenic drugs, such as barbiturates and sulphonamides, together with better treatment of acute attacks can undoubtedly contribute to improved outcome. Nonetheless, acute porphyria re- mains potentially life-​threatening, and deaths or marked disability due to prolonged, mismanaged, or undiagnosed attacks are all too frequent. Complications Acute encephalopathy—​posterior reversible encephalopathy syndrome (reversible posterior leucoencephalopathy syndrome) This syndrome has been reported in exceptionally ill patients with acute porphyria. Clinically, the patient is confused and complains of headache, visual phenomena, and field loss, often accompanied by generalized seizures. Ischaemia of the occipital cortex during acute attacks with severe hypertension has been associated, in several in- stances, with failed recognition of colours or of human faces (proso- pagnosia) and cortical blindness. Magnetic resonance imaging typically reveals bilateral symmet- rical and gyriform lesions in the cerebrum and cerebellar hemi- spheres due to oedema. The lesions are initially hyperintense on T2-​weighted images and prominent in the posterior regions of the occiput and pons as well as parietal lobes. Should the patient make a neurological recovery with stabilization of blood pressure, the radiological changes are potentially reversible over a matter of 1 to 2 weeks. However, in some cases permanent neurological impair- ment including visual changes and seizures is the result. Though un- common, death may occur with progressive swelling due to cerebral oedema, compression of the brainstem, increased intracranial pres- sure, or intracerebral haemorrhage. Posterior reversible encephalopathy syndrome may recur in about 5 to 10% of cases. This happens most commonly in cases accom- panied by hypertension, and where sustained or paroxysmal sys- temic hypertension is difficult to control or escapes clinical attention. Other neurological complications Rapidly recurrent attacks of porphyria may be associated with se- vere motor neuropathy and sustained hypertension; postural hypo- tension may result from autonomic neuropathy. Cranial nerve palsies can occur in severe cases, typically affecting the facial and vagus nerves. Chronic kidney disease Chronic kidney disease is frequently associated with hypertension and a common long-​term comorbidity of the acute porphyrias, but es- pecially acute intermittent porphyria. Nearly all patients with recur- rent acute attacks are or will be affected, as revealed in a large cohort

section 12  Metabolic disorders 2042 of 415 patients with confirmed deficiency of hydroxymethylbilane synthase and acute intermittent porphyria. Chronic kidney dis- ease arose over 10 years in nearly 60% of the symptomatic patients, in whom an annual decline in the glomerular filtration rate of ap- proximately 1 ml/​min per 1.73 m2 was documented; proteinuria was rarely observed. Renal histology showed a chronic tubulointerstitial nephropathy associated with a fibrous intimal hyperplasia and focal cortical atrophy. The effects of quasi-​pathological amounts of por- phyrin precursors were examined and the investigators suggested that endoplasmic reticulum stress, apoptosis, and epithelial changes reflected injury to proximal tubular cells. Chronic kidney disease associated with acute porphyria should be considered in the presence of chronic tubulointerstitial nephropathy or focal cortical atrophy with severe proliferative arteriosclerosis. Hepatic carcinoma There are multiple reports of primary liver carcinoma in the three most frequent acute porphyrias: acute intermittent porphyria, varie- gate porphyria, and hereditary coproporphyria. Histological reports indicate that most of these tumours are hepatocellular carcinomas, but cholangiocarcinomas are also found. Annual incidences of 0.16 to 0.35% are reported in cohorts of patients followed. Tumours are associated with age over 50 years and undoubtedly more frequent in patient cohorts from Sweden and Norway, suggesting the presence of additional genetic factors due to at least one high-​prevalence, cosegregating modifying factor. No consistent environmental co- factor (drug, hepatitis virus, or other features) has been identified. Specificity in relation to the acute porphyrias may not be ab- solute: the author has recently cared for an unusual patient, a 62-​ year-​old man with congenital erythropoietic porphyria and with no other frank toxicity, iron storage, or hepatitis infection: a cryptic but ultimately fatal hepatocellular carcinoma developed at the time of splenectomy. Regular radiological surveillance of the liver, typically by ultra- sonography, is appropriate for older patients suffering from acute porphyria. Individual acute porphyrias These are, in a descending order of frequency:  acute intermit- tent porphyria, variegate porphyria, hereditary coproporphyria, and Doss’ porphyria (5-​aminolaevulinate dehydratase deficiency). The first three of these disorders occur in at-​risk heterozygotes for a single mutant allele in the cognate gene as autosomal dominant traits; 5-​aminolaevulinate dehydratase deficiency is inherited as a very rare autosomal recessive trait. Acute intermittent porphyria This, the most frequent of the acute porphyrias, is caused by muta- tions in the porphobilinogen deaminase gene that maps to human chromosome 11q23 in which well over 200 mutations have been identified. Several widespread mutations have been identified in cer- tain populations, but most are reported in only one or two pedigrees. Two isozymes of the human porphobilinogen deaminase enzyme occur in the tissues: an erythroid mRNA variant and a nonerythroid transcript that encodes 17 additional amino acid residues in its N-​ terminus, leading to synthesis of a housekeeping ubiquitous isozyme and an erythroid-​specific isozyme. Most mutations cause a decrease in the abundance as well as the activity of the porphobilinogen deaminase enzyme in all tissues. A few mutations associated with lack of the detectable protein product from the mutant allele are as- sociated with reduction of the housekeeping isozyme but normal enzymatic activity of the erythroid-​specific isozyme. Thus, in such patients, hepatic porphobilinogen deaminase activity may be re- duced to approximately half normal values while the activity of the easily accessed red cell enzyme is within the normal range. A few mutations lead to the synthesis of a catalytically impaired but stable porphobilinogen deaminase protein from the cognate mutant allele, but these are a minority. Molecular analysis of the porphobilinogen deaminase gene in patients with acute intermit- tent porphyria has been very valuable in establishing diagnosis of latent heterozygotes at risk in the affected family, for the provision of appropriate counselling, and for the introduction of preventative strategies (see ‘Latency or genetic penetrance’). Acute intermittent porphyria is characterized solely by acute porphyric attacks and cutaneous photosensitivity does not occur. In most instances, the patients do not notice any change in their urine, but when the urine is allowed to stand, the increased excretion of pyrroles leads to the formation of coloured oxidation products of porphobilinogen (loosely called porphobilin), which may lead to obvious discoloration (Fig. 12.5.2). During the increased excretion of porphyrin precursors, water-​soluble porphyrins, formed as a re- sult of nonenzymatic photochemical reactions, induce a pink discol- ouration. During severe acute attacks, copious excretion of pyrrole precursors, including porphobilin, may occasionally give the urine a striking appearance resembling blackcurrant juice or strong solu- tions of potassium permanganate. Latency or genetic penetrance The incidence and severity of acute attacks in acute intermittent por- phyria and variegate porphyria are generally greater than in heredi- tary coproporphyria. Various estimates indicate between 1 in 10 to 1 in 5 of heterozygotes experience acute attacks of porphyria during their lifetime; a study in France indicated penetrance of 23%. Recent large-​scale studies using genomic/​exomic sequencing data in white Fig. 12.5.2  Urine from a patient with acute intermittent porphyria around the time of an acute attack (left); control urine (right). A positive reaction with Ehrlich’s diazo reagent is shown in the patient’s urine following the addition of 50 µl urine to 1 ml of 2% acidic dimethyl benzaldehyde. Subsequent tests showed that the pink diazo adduct was insoluble in chloroform and other organic solvents indicating the presence of excess porphobilinogen. (Urobilinogen in excess may give a positive reaction with the diazo reagent but the product is readily extracted into organic solvents.)

12.5  The porphyrias 2043 persons identified nonsynonymous and two consensus splice-​site pathogenic variants for a combined prevalence of approximately 0.00056. Since the estimated prevalence of acute attacks is approxi- mately 0.000005, and the estimated frequency of clinical pathogenic variants is approximately 0.00056, the penetrance of acute intermit- tent porphyria appears to be as low as 1% of all heterozygotes. Since the disease is monogenic, very low penetrance points to the exist- ence of modifying factors (environmental or genetic) that either pre- dispose heterozygotes to the acute attack or suppress expression of the disease. More recent research into acute intermittent porphyria from France has reconsidered the concept of autosomal dominant in- heritance with incomplete penetrance. This work examined the prevalence, penetrance, and heritability in affected pedigrees with 602 overt patients, 1968 first-​degree relatives, and control samples from the general population. The pathogenicity of the 42 missense variants identified was assessed in silico, and also systematically in vitro by measuring residual enzyme activity of recombinant mutant hydroxymethylbilane synthases. The minimal estimated prevalence of acute intermittent porphyria in the general population was 1 in 1299. Thus, 50 000 subjects would be expected to carry the gen- etic trait in France, allowing penetrance to be estimated at 22.9% in affected families but in only 0.5 to 1% of the general population. Intrafamily studies showed correlations to be strong overall and modulated by kinship and the area in which the person was living, immediately pointing to a combination of interacting and strong influences of genetic and environmental modifiers on the expres- sivity of the trait in pedigrees. Null alleles were associated with a more severe phenotype and a higher penetrance than for other mu- tant alleles. The authors concludes that the striking difference in the penetrance of the mutant hydroxymethylbilane synthases between the general population and the French families affected by porphyria indicates that the inheritance does not follow the classical autosomal dominant model: expression of disease is modulated by strong en- vironmental and genetic factors that are independent of the ‘causal’ HMBS gene. In practice, increasing use of molecular diagnostic methods for screening at-​risk families, institution of appropriate avoidance, and the careful dissemination of information to family members and their medical advisers will further reduce the likelihood of dis- ease in latent gene carriers. Latent carriers of acute intermittent porphyria have a high frequency of hypertension and, although this should be treated, the potential for inducing attacks is in- creased by the uninformed prescription of antihypertensive drugs. Some patients appear to have depression and other chronic mental symptoms, and at least one survey has reported an increased prevalence of acute intermittent porphyria in patients attending long-​stay psychiatric facilities, which again puts them at risk from the ill-​considered use of porphyrinogenic neuroleptic and other psychoactive drugs. Variegate porphyria Variegate porphyria is particularly frequent among white South African people and other ethnic groups within that country. The condition is associated with typical acute attacks of porphyria as well as skin manifestations (the van Rooten skin). Acute attacks of porphyria occur very much as in acute intermittent porphyria. In a series of patients, more than one-​half presented with skin lesions alone, one-​fifth had acute neurovisceral disease, and a similar pro- portion had acute attacks as well as cutaneous disease. Cutaneous photosensitivity resembles that seen in porphyria cutanea tarda and hereditary coproporphyria (see ‘Hereditary coproporphyria’) with fragility, milia, hyperpigmentation, and hairi- ness of light-​exposed skin. During acute sunlight exposure, vesicles and even large bullas may form. Microscopic examination of the affected skin shows deposits of immunoglobulin and hyaline ma- terial (that stains positively with periodic acid–​Schiff reagent) in the dermal capillaries with proliferation of the basal lamina. As with porphyria cutanea tarda, ingestion of reproductive steroids (e.g. the oral contraceptive pill) may induce the cutaneous manifestations of variegate porphyria in otherwise latent heterozygotes. A few severely affected patients with variegate porphyria have inherited mutations of the protoporphyrinogen oxidase gene (that maps to chromosome 1q22–​q23) from each parent, leading to homozygous ‘dominant’ variegate porphyria. These individuals pre- sent in childhood with a severe phenotype associated with marked photosensitivity and a neurological syndrome as described briefly in ‘Protoporphyrinogen oxidase’. Hereditary coproporphyria This condition is an infrequent and often mild form of acute por- phyria which may be associated with cutaneous manifestations. It is due to mutations in the coproporphyrinogen III oxidase gene that maps to chromosome 3q12 and is transmitted as an autosomal dom- inant trait of low penetrance. The condition usually presents with acute attacks of abdominal pain, as with the other acute porphyrias, and about 30% of patients develop cutaneous photosensitivity. As with some other porphyrias, several children presenting with marked photosensitivity in childhood have been shown to have in- herited a mutant allele of the coproporphyrinogen III oxidase gene from each parent giving rise to so-​called homozygous dominant her- editary coproporphyria. Particular mutations in the gene are usually restricted to individually infected pedigrees. As with the other acute porphyrias, molecular analysis of the coproporphyrinogen III oxi- dase gene may be of value in identifying at-​risk heterozygotes for genetic counselling and provision of appropriate advice about the prevention and management of symptomatic disease. 5-​Aminolaevulinate dehydratase deficiency
(Doss’ porphyria) Only a few affected homozygotes for this recessive condition have been identified. Molecular analysis of the cognate gene has revealed the presence of compound heterozygosity and homozygosity for point mutations in the gene which map to chromosome 9q34. As with the porphobilinogen deaminase gene, there are two promoter regions and alternative noncoding exons that allow for the synthesis of housekeeping and erythroid-​specific transcripts. Less than 10 cases of this porphyria have been reported, but it seems likely from the individual case histories of those identified that the disease will be under-​recognized as the cause of acute abdominal crises, usually presenting shortly after puberty and associated with neurological symptoms, including respiratory paralysis. The condition resembles acute lead poisoning. The urine contains an excess of 5-​aminolaevulinate but the excretion of porphobilinogen and tetrapyrrolic haem precursors is normal. Heterozygotes for aminolaevulinate dehydratase deficiency have been reported in at

section 12  Metabolic disorders 2044 least one lead worker in whom peripheral neuropathy was ascribed to simple lead poisoning, but it may have resulted from the suscep- tibility of the residual 5-​aminolaevulinate dehydratase to inhibition by environmental lead. Unusual genetic variants of the acute porphyrias (homozygotes or compound heterozygotes) In the last 20 years, very rare homozygous forms of porphyria have been recognized where the presence of two mutant alleles of the causative gene are responsible for severe clinical disease. Most in- dividuals affected prove to be compound heterozygotes for two mu- tant alleles of the cognate gene; true homozygotes are most likely to occur only in consanguineous pedigrees. Porphobilinogen deaminase (hydroxymethylbilane synthase) Bialleic mutations in the third enzyme in the haem biosynthetic pathway led, in one well-​studied pedigree, to an unusual neurological syndrome with impaired cerebellar function and slowly progressive white matter disease with hypomyelination. There was also truncal muscle weakness and wasting, notably associated with ptosis, a clin- ical sign compatible with mitochondrial disease. Biochemical tests revealed excessive urinary porphobilinogen, 5-​aminolaevulinate, porphobilinogen, and uroporphyrin. Furthermore hepta-​, hexa-​, penta-​, and coproporphyrins I were highly increased in urine and typical of patients with acute porphyria during a metabolic crisis. The author suggests that these features indeed indicate a mitochon- drial defect, probably due to impaired oxidative phosphorylation related to the haem deficiency in critical cytochrome components of the electron transport chain, such as cytochrome c. Supportive evidence is provided by defective mitochondrial energetics in the brain and skeletal muscle of mice generated as homozygous mutants for the porphobilinogen deaminase locus, and by a patient having consistently elevated plasma lactate concentrations and metabolic acidosis. Intermittent hypoglycaemia was found in one patient who died at the age of 20 years due to cardiorespiratory failure; their sibling had died suddenly at 9 years of age of unknown causes. Despite greatly reduced hydroxymethylbilane synthase activity in these two sib- lings due to compound heterozygosity for adjacent base transitions in the same codon in exon 10 of the PBG deaminase gene, neither had any evidence of metabolic decompensation or the neurovisceral manifestations of an acute porphyric attack; neither of the biological parents had ever suffered acute porphyria. Coproporphyrinogen oxidase This is a very rare syndrome principally dominated by haemato- logical (erythroid) features of haemolytic anaemia accompanied by hepatosplenomegaly and marked photosensitivity. The patients thereby resemble those suffering from congenital erythropoietic porphyria (Gunther’s disease) due to biallelic mutations that induce marked deficiency of uroporphyrinogen III cosynthase. This presen- tation is associated with expression of a missense mutation, p.K404E, in either homozygous or compound heterozygous form with a dis- abling splice-​site mutation in the COX gene. The hallmark of this disorder is the incomplete decarboxylation of the tetracarboxylic coproporphyrinogen III to yield excess tricarboxylic intermediate harderoporphyrin rather than the principal product of wild type coproporphyrinogen III oxidase, the dicarboxylic isomer and immediate haem precursor, protoporphyrinogen IX. These patients do not develop acute neurovisceral symptoms of acute porphyria. Protoporphyrinogen oxidase Patients harbouring biallelic mutations of protoporphyrinogen oxidase present in infancy or childhood with a severe phenotype associated with disfiguring photosensitivity accompanied by soft tissue mutilation, evidence of digital bone loss (osteolysis), and impaired cognitive development. Peripheral skeletal abnormal- ities include medially deviated and shortened fifth digits, known as clinodactyly, and slight growth retardation. A mild distal sensory neuropathy has been reported. There is later development of gen- eralized seizures, usually during adolescence or early adulthood. Cranial magnetic resonance imaging reveals a leucodystrophy with hypomyelination of long tracts. Biochemical investigation reveals striking elevation of red cell and plasma protoporphyrin which is predominantly zinc-​chelated, a feature of homozygous hepatic porphyrias. None of the patients so far reported has devel- oped attacks of acute porphyria. Ferrochelatase Although homozygous ferrochelatase deficiency (protoporphyria) is not strictly related to an acute porphyria, patients who are com- pound homozygotes or true homozygotes have a severe photosensi- tivity syndrome that is more frequently complicated by accelerated cholestatic liver disease and pigment cirrhosis. However, the pos- ition is etymologically complex, since generally only those who in- herit a mutation on one FECH allele and a polymorphism in trans on the other allele, which is common in general European popula- tion and diminishes expression of the ferrochelatase, develop clin- ical manifestations. Such patients are not conventional compound heterozygotes since the polymorphism is common in the apparently healthy population (c.11% of Europeans). However, rare individuals with biallelic inheritance of disabling FECH mutations appear to be especially predisposed to the development of cholestatic liver dis- ease with cirrhosis that aggravates the photosensitivity and leads to a potentially fatal outcome unless liver transplantation is undertaken. Individual cutaneous porphyrias Congenital erythropoietic porphyria Congenital erythropoietic porphyria (Gunther’s disease) is a classic but very rare syndrome now known to have an astonishing range of presentation from severe haemolytic anaemia in utero or severe photosensitivity presenting soon after birth (with excess porphyrins staining the teeth and urine) to mild late-​onset forms presenting with skin lesions in adult life. Most patients have a mild to severe haemolysis with increased reticulocytosis, circulating normoblasts, decreased serum haptoglobin, and increased unconjugated bilirubin concentrations. Inclusion bodies are often seen in marrow, erythroid cells, and circulating normoblasts. Splenomegaly develops in child- hood, thereby causing pancytopenia as a result of hypersplenism; this accelerates the haemolysis and leads to compensatory erythro- poiesis in the bone marrow. Under these circumstances, splenec- tomy may help to control the condition. The classic skin manifestations are of severe blistering lesions on sunlight-​exposed skin, particularly of the hands and face, with the

12.5  The porphyrias 2045 formation of vesicles and bullae that may become infected. There are pigmentary changes with greatly increased skin fragility. Healing of the lesions with or without consequential infection often leads to cutaneous deformities with loss of digits, scarring of the eyelids, nose, lips, ears, and scalp, and occasionally blindness due to corneal scarring (Fig. 12.5.3). Examination of the teeth shows erythrodontia and deformities, and exposure to ultraviolet light may reveal striking dental fluorescence. The condition is associated with osteoporosis (a) (b) Fig. 12.5.3  Congenital erythroid porphyria (Gunther’s disease). (a) Pinnae and hand showing porphyrin deposits and tissue destruction due to photonecrosis. (b) Splenomegaly. (c) Successive urine samples before and after splenectomy showing progressive postoperative reduction in porphyrinuria (uroporphyrin I and coproporphyrin I).

section 12  Metabolic disorders 2046 and resorption of long bones as a result of gross expansion of the erythroid bone marrow. Mutations in the uroporphyrinogen III synthase gene that maps to chromosome 10q25.3–​q26.3 have been shown to be respon- sible for this disease and thus may assist in the prenatal diagnosis of mothers who have previously given birth to an affected infant and are harbouring an at-​risk pregnancy. Constitutive activation of the haem biosynthetic pathway in erythroid cells leads to persistent overproduction of uroporphyrinogen I  and coproporphyrinogen I as by-​products of the defective synthesis of uroporphyrinogen III, the sole precursor of protoporphyrin IX and haem. These reduced and colourless metabolites become oxidized to the fluorescent tissue and urinary porphyrins associated with the passage of pink urine that characterizes this often devastating disease. Several infants and children with congenital erythropoietic por- phyria have been successfully treated by haematopoietic stem cell transplantation (HSCT) and this remains a convincing option for treatment of this very severe and otherwise life-​shortening in- born error of haem metabolism. Splenectomy is often required in early childhood to reduce transfusion requirements and improve cytopenias (Fig. 12.5.3). It has recently been identified that an ap- proved antifungal agent, ciclopirox, binds allosterically to and stabilizes several naturally occurring uroporphyrinogen III syn- thase human mutants and restores their activity. This may lead to an innovative treatment for this severe and destructive disease, which currently lacks molecular therapy. Porphyria cutanea tarda This disease is the most common of the cutaneous porphyrias and, unlike other hepatic porphyrias, is never associated with acute porphyric crises. It is characterized by skin blistering which is re- lated to sunlight exposure. Aetiology Toxic cutaneous porphyria may result from environmental exposure to dioxin or to hexachlorobenzene, particularly after industrial accidents such as that which occurred in Turkey in the 1960s. Occasional cases have been reported after exposure to other halo- genated phenols, but under these circumstances it appears simply to be an environmental toxic syndrome which is separate from the sporadic porphyria cutanea tarda that is precipitated by other spe- cific environmental factors including increased hepatic storage iron, excess ethanol consumption, administration of oestrogens, hepatitis C virus infection, human immunodeficiency virus infection, and (possibly) nutritional deficiencies including antioxidants such as vitamin C. Most individuals who develop sporadic porphyria cutanea tarda prove to have increased iron stores in association with the presence of one or more mutant alleles for the HFE gene that also predispose to the development of hereditary adult haemochromatosis. Many patients also consume excess alcohol and smoke. There is a clear as- sociation with renal impairment in which the development of the disease can be explained by the presence of iron overload (as a result of defective iron utilization with or without routine iron supplemen- tation, particularly in patients on haemodialysis) and failure to ex- crete excess plasma porphyrins that do not readily diffuse through the peritoneal cavity or haemodialysis membranes. In sporadic porphyria cutanea tarda, there is a partial defi- ciency of uroporphyrinogen III decarboxylase activity in the liver and no family history of the condition. The sequencing of the human uroporphyrinogen decarboxylase gene that maps to human chromosome 1p34 has not provided any evidence of mutations to account for the tissue-​specific enzyme deficiency, and no isoforms of the enzyme have yet been identified, hence the molecular patho- genesis of sporadic porphyria cutanea tarda remains unknown, but it is clear that iron and other environmental influences inactivate the hepatic enzyme. The relationship between regulators of iron homeo- stasis and the demand for haem biosynthesis in the hepatocytes of affected individuals is not understood, but it appears likely from studies in experimental animals that genetic variation in the ex- pression and activity of cytochrome isozymes such as P450 IA2 may be critical for disease expression. Irreversible inhibition of hepatic (c) Pre-operative Day of surgery Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 1 2 3 4 5 6 7 8 9 10 11 Fig. 12.5.3  Continued

12.5  The porphyrias 2047 Fig. 12.5.4  Porphyria cutanea tarda in a 60-​year-​old heterozygote for the HFE p.C282Y mutation. This man, a taxi driver, had noticed irritation after exposure of his hands to light transmitted through the windscreen. He had noticed fragility and blistering combined with pigmentary changes typical of this disorder. After treatment by controlled phlebotomy his skin complaint has regressed. uroporphyrinogen decarboxylase may also explain the occurrence of toxic porphyria cutanea tarda after exposure to halogenated hy- drocarbons, metabolites of which cause experimental uroporphyria in animals. Less than one-​quarter of patients who have porphyria cutanea tarda show a familial susceptibility to the condition, when muta- tions in one allele of the human uroporphyrinogen decarboxylase gene lead to catalytic deficiency of the enzyme in all cells, including erythrocytes. In most instances, the genetic defect leads to par- tial reduction of the enzyme protein encoded by the mutant allele. Studies of pedigrees affected by familial porphyria cutanea tarda indicate that expressivity of the trait is very low; less than 10% of heterozygotes develop clinical disease. Conversely, a very few pa- tients present with a syndrome that closely resembles congenital erythropoietic porphyria with marked blistering skin lesions, excess hair growth, and cutaneous scarring in association with the excre- tion of pink or red urine. These individuals represent a homozy- gous form of uroporphyrinogen decarboxylase deficiency, termed hepatoerythropoietic porphyria, associated with a variety of muta- tions in the uroporphyrinogen III decarboxylase gene. In hepatoerythropoietic porphyria, the activity of uro­ porphyrinogen decarboxylase is markedly deficient, although re- sidual activity remains to preserve essential haem biosynthesis in the erythron and liver. Most patients ultimately develop spleno- megaly with accelerated haemolysis closely resembling con- genital erythropoietic porphyria. Molecular analysis of the human uroporphyrinogen decarboxylase gene may assist the prenatal diag- nosis of at-​risk pregnancies in women who have already given birth to an affected infant. Clinical features The clinical features of porphyria cutanea tarda of whatever form are very characteristic and are confined to light-​exposed skin (Fig. 12.5.4). Usually, the only signs are of erosions resulting from minor trauma in skin, with increased fragility as a result of light ex- posure, typically on the dorsum of the hands. Other changes include the development of large subepidermal bullae after exposure to light, which may burst leaving ulcerated lesions that are slow to heal. Increased pigmentation, often accompanied by areas of decreased pigmentation, is a common feature combined with increased hair growth, particularly on the face. Patients with porphyria cutanea tarda do not always notice the photosensitivity and rarely experience marked pain unless exposed to brilliant sunlight. Occasionally, there is evidence of dermal injury and loss of nails, damage to the conjunctivae, and hair loss. Careful examination of the affected areas shows small depigmented cuta- neous scars and the formation of milia. If bacterial infection occurs and there is repeated exposure to sunlight, then severe and per- manent scarring may result. Typically, porphyria cutanea tarda occurs in middle-​aged men with a history of alcohol use and in women after institution of oestrogen replacement therapy; in young persons, infection with hepatitis C or the immunodeficiency virus may precipitate dis- ease expression. Frank signs of hepatomegaly or iron overload are rare in porphyria cutanea tarda but have been noted; as with adult haemochromatosis, there is a significantly increased frequency of hepatocellular carcinoma. Occasionally patients with porphyria cutanea tarda may notice an increase in urine excretion of formed porphyrins which, especially after concentration overnight, may resemble the colour of tea or cola. The stool and urine contain large quantities of coproporphyrins and uroporphyrins that fluoresce intensely on exposure to long-​ wavelength ultraviolet light when placed in a suitable vessel for its transmission (namely silica rather than standard glass). Similarly, examination of liver biopsy specimens under ultraviolet light reveals bright red/​orange fluorescence; microscopic examination may also show coincidental hepatitis with or without excess deposits of stain- able tissue iron reflecting the increased iron storage of this disease. In sporadic porphyria cutanea tarda, increased storage iron is re- flected by the modest elevations of serum ferritin that often occur in association with the presence of one or more copies of the C282Y allele of the HFE gene that maps to human chromosome 6 and which is associated with adult haemochromatosis. Treatment Sunlight exposure should be avoided as much as possible and sun- block creams used until the porphyrin abnormality is corrected. Care is needed to protect fragile skin from mechanical injury and from infection. Patients should be advised to moderate or stop their intake of alcohol and avoid the use of iron tonics and sex hormones, especially oestrogens. Screening should be undertaken for chronic infection with human immunodeficiency virus and hepatitis viruses, especially hepatitis C. Management should also include imaging or biopsy of the liver if serum liver-​related tests are abnormal, as well as measurement of serum α-​fetoprotein since there is a risk of hepatocellular carcinoma in this disease. Most patients with porphyria cutanea tarda respond to iron de- pletion by phlebotomy and initial iron status should be determined by measuring serum ferritin concentrations. Weekly or fortnightly removal of 500 ml of blood will usually correct the abnormal urine and plasma porphyrin profile within a few months, but maintenance phlebotomy will be required, usually amounting to the removal of 2 to 4 units of blood at intervals each year. Successful therapy re- duces the urinary excretion of porphyrins to normal. Patients with

section 12  Metabolic disorders 2048 porphyria complicating renal failure should be treated with recom- binant human erythropoietin and depleted of iron by gentle phle- botomy or parenteral desferrioxamine if necessary. The cutaneous manifestations of porphyria cutanea tarda respond rapidly to low-​dose chloroquine treatment, which should be con- sidered in patients with persistent symptoms or at the outset before iron storage has been fully corrected. This action of chloroquine was discovered empirically, but the agent forms complexes with uroporphyrin deposits and promotes their external cellular disposal, promoting excretion of uroporphyrin from the liver and inducing marked but transient porphyrinuria. Although chloroquine usually provides rapid relief from the cutaneous disease and photosensi- tivity, it does not correct the underlying metabolic defect in the liver and its long-​term use is not recommended unless the other pro- vocative factors in porphyria cutanea tarda have been removed. The usual effective dose of chloroquine is 100 to 200 mg given once or twice weekly; larger doses are associated with marked hepatic tox- icity in porphyria cutanea tarda. The drug is reported to have no therapeutic effect on other photosensitive porphyrias. (Erythropoietic) protoporphyria and X-​linked protoporphyria Protoporphyria is caused by the overproduction of the immediate precursor of haem, protoporphyrin IX, principally in the bone marrow. It causes an unusual cutaneous photosensitivity syndrome that presents in infancy and is a neglected cause of fatal hepatobiliary disease in about 5% of those affected. Genetics Protoporphyria is inherited as a recessive condition but often generation-​to-​generation transmission of the disease has been ob- served. An original ingenious postulate of an autosomal three-​allele mode of transmission has been further refined by the remarkable identification of causal mutations at three distinct chromosomal loci—​the ferrochelatase gene (FECH on chromosome 18q21.31), the erythroid 5-​aminolaevuliate synthase gene (ALAS2, which maps to Xp11.21), and the caseinolytic mitochondrial matrix peptidase chaperone subunit gene (CLPX, localized on chromo- some 15q21.32). Inheritance of mutations in the coding region of the ferrochelatase gene that partially inactivate the enzyme are coinherited in the trans isomer with a low-​expression allele (FECH IVS3–​48T>C) that occurs at polymorphic frequency (c.10%) in the population, which gives rise to apparent dominant transmission in some families. Parent-​to-​offspring transmission of protoporphyria occurs in less than 10% of cases, but in all instances of the disease there is a marked deficiency of the enzyme ferrochelatase (substan- tially <50% of control values). The asymptomatic carrier parent only shows mild ferrochelatase deficiency with occasional fluorescent red cells that are even visible on examining the unstained blood film by conventional ultraviolet light transmission microscopy. A few patients, usually with clinically severe protoporphyria, have biallelic mutations in ferrochelatase—​true recessive protoporphyria—​and these may be predisposed to the severe cholestatic liver disease described in several pedigrees. Since ferrochelatase, an iron-​sulphur cluster protein located on the inner mitochondrial membrane, is responsible for the final step of haem biosynthesis in the inner mitochondrial membrane, where it catalyses the insertion of ferrous iron into the protoporphyrin IX macrocycle, partial deficiency of this enzyme will give rise to the compensatory increase in protoporphyrin abundance and hence the consequential photodynamic effects of protoporphyria. These findings are reflected in a large series of 226 patients with a clinical and biochemical diagnosis of protoporphyria from the United States of America. This demonstrated an equal sex distri- bution and mean age of 37 years. A ferrochelatase mutation and the common low-​expression mutation was detected (presumed in trans) in 186 (82.3%) and only one patient had two FECH mutations. Twenty-​two patients had X-​linked protoporphyria (9.7%; 10 male and 12 female). Of note, nine patients (4.0%) had symptomatic and biochemical evidence of protoporphyria but no detectable mutation in the FECH or ALAS2 genes (see following subsections). CLPX gene-​related protoporphyria Several patients with clinical and biochemical protoporphyria but lacking FECH or ALAS2 mutations have been shown to harbour heterozygous mutations in an unusual AAA+ (ATPases associated with diverse cellular activities) protease, caseinolytic mitochondrial matrix peptidase chaperone subunit (ClpXP), of the Clp protease complex that regulates ALAS2 activity among other mitochondrial proteins. The gene CLPX encodes a chaperone subunit that confers ATP-​dependent specificity of the Clp protease complex, which acts as an unfolding enzyme that regulates ALAS1 and ALAS2 proteins. Another action of the protein is to facilitate incorporation of the essential pyridoxal 5´-​phosphate cofactor into 5-​aminolaevulinate synthase. Detailed studies in several experimental systems of the first mu- tant ClpXP protein in a family with protoporphyria but no muta- tions in either FECH or ALAS2 genes convincingly showed that the mutation would disrupt the ATP binding domain of the protein but leave the activating and stabilizing chaperone function towards the ALAS pyridoxal cofactor intact. The predicted net effect would be activation of the ALAS2 enzyme in erythroid cells and in effect protoporphyrin substrate overdrive at the next rate-​limiting step in the pathway, ferrochelatase. These are early findings and more com- prehensive genetic studies are needed fully to explore the role of ClpX mutations in erythropoietic protoporphyria. X-​linked protoporphyria X-​linked protoporphyria is due to inheritance of gain-​of-​function variants of ALAS2 enzyme in which erythroid precursor cells over- produce 5-​aminolaevulinate and accounts for 5 to 10% of patients with a diagnosis of protoporphyria. While this disorder also demon- strates generation-​to-​generation transmission and may erroneously be considered as a dominant trait of variable penetrance, the ALAS2 gene on the short arm of the X chromosome undergoes gene-​dosage compensation by lyonization. Females with the trait may be asymp- tomatic and—​as somatic mosaics—​show greater variability of clin- ical expression than affected males within their pedigree. Diagnosis The detection of markedly increased free erythrocyte protoporphyrin and zinc-​chelated erythrocyte protoporphyrin is the most sen- sitive biochemical diagnostic test for this disease. Identification of pathogenic gain-​of-​function variants affecting the last exon of ALAS2, the gene encoding erythroid-​specific 5-​aminolaevulinate synthase 2, confirms the diagnosis. These mutations have proved

12.5  The porphyrias 2049 to be sequence variants in exon 11 and in other coding and spli- cing regions. Thorough expression studies of the cognate isozyme show consistently enhanced catalytic activity for the formation of aminolaevulinate, which in the pathological milieu of erythroid precursor cells in protoporphyria presumably leads to a relative, secondary bottleneck at the level of ferrochelatase with the conse- quential constitutive overproduction of protoporphyrin IX. Clinical features and pathology Skin disease Protoporphyria characteristically presents with severe burning pain and cutaneous irritation on exposure to visible light and is usually obvious in infancy or early childhood. Erythema and diffuse oe- dema may follow marked light exposure, but vesicles, blistering, and altered skin fragility are most unusual. After several years, increased pigmentation and thickening of the skin (lichenification) occur, es- pecially over the knuckles. A typical feature is of shallow scarring in the malar regions of the cheeks and at the angle of the lips, where scarring is termed ragades. Overt scarring is unusual. There are no changes in urine colour. Protoporphyria is often the subject of de- layed diagnosis because of the marked disparity between the severity of the symptoms and the development of physical signs in the skin. The cutaneous pathology results from photoactivation of red cell-​ and plasma-​derived protoporphyrin IX in skin capillaries (Figs. 12.5.4 and 12.5.5). Protoporphyrin IX is a hydrophobic mol- ecule that dissolves in cell membranes; it has a photoactivation spec- trum in the Soret region with subsidiary activation by green and yellow light. Photoinjury is associated with complement activation and release of vasoactive factors; there is intracellular epidermal oedema accompanied by acute inflammatory changes and extrava- sated red cells. Deposits of hyaline material are found in superficial capillaries with thickening of the basement membranes. Haematological and liver disease Mild hypochromic microcytic changes with mild anaemia are usually the only manifestations of disturbed haem biosynthesis and iron metabolism in the bone marrow, although examin- ation of the marrow may reveal occasional sideroblasts with intramitochondrial iron deposits. Haemolysis is usually clinically insignificant until severe cholestatic hepatic disease occurs, when splenomegaly and hypersplenism aggravate haemolysis. The photosensitivity worsens under these circumstances and there is upper abdominal pain with splenic enlargement, jaundice, and ex- treme photosensitivity as concentrations of free protoporphyrin in the plasma rise (Fig. 12.5.6). A vicious cycle of decompensation is established with either fulminant hepatic failure associated with cholestasis due to protoporphyrin deposits within biliary radicals, or the development of cirrhosis. Without treatment (hepatic trans- plantation) the prognosis is dismal. Protoporphyria is normally associated with trivial abnormalities of serum liver-​related tests but in a few patients micronodular cirrhosis with pigment deposition occurs. Examination of the liver under po- larized light shows birefringent crystals with a characteristic Maltese-​ cross appearance, and examination under long-​wave ultraviolet light reveals bright fluorescence. Gallstones containing precipitated protoporphyrin occur frequently, but cholestasis results principally from intracellular and canalicular precipitation of protoporphyrin. Deteriorating hepatic disease is heralded by generalized abdominal pain, splenic enlargement, worsening jaundice, and haemolysis. Treatment Interruption of the enterohepatic circulation of protoporphyrin with charcoal or polymeric cationic resins, such as cholestyramine, may arrest the early downhill course by binding protoporphyrin or promoting hepatic bile acid secretion. However, once established, hepatic decompensation and accelerating photosensitivity is rapid. Haematopoietic stem cell transplantation or bone marrow transplantation Studies in mice and in a few human patients with protoporphyria confirm that the disease can be arrested, and in practical terms Fig. 12.5.5  Fluorescence microscopy of an unstained blood film from a patient with erythropoietic protoporphyria. Note the fluorescence of increased free protoporphyrin within individual young erythrocytes and reticulocytes. Fig. 12.5.6  Examination of human plasma under long-​wave ultraviolet light. Plasma on the left was obtained from a patient with protoporphyrin hepatopathy and greatly increased photosensitivity. On the right is plasma obtained from a healthy subject. Maximum fluorescence was obtained by exposure to visible light in the violet and green–​ yellow spectral regions, corresponding to the absorbance bands of protoporphyrin. Note the bright red fluorescence due to the presence of high concentrations of free protoporphyrin.

section 12  Metabolic disorders 2050 completely corrected by HSCT or bone marrow transplantation. Carried out sufficiently early, HSCT allows at least partial recovery of the damaged or failing protoporphyric liver and is the ideal pro- cedure in patients who are at high risk of fatal liver disease: por- phyrin biochemistry is corrected and photosensitivity is no longer present, liver-​related tests and imaging can be restored to healthy values. Theoretically, the definitive therapy of protoporphyria will re- quire restoration of erythroid cell ferrochelatase activity in bone marrow. There is a single report of successful marrow transplant- ation in protoporphyria with coincidental myeloid leukaemia. This procedure cured the symptomatic protoporphyria. In future, either bone marrow transplantation or erythroid progenitor gene therapy will be used to correct this disease in patients with life-​threatening liver sequelae. Ancillary treatment by blood transfusion or red cell exchange transfusion will reduce the immediate source of plasma and red cell protoporphyrin, and in the immediate preoperative period plasmapheresis may also reduce phototoxicity. Liver transplantation Established severe protoporphyrin hepatotoxicity remains an in- dication for liver transplantation, but even successful treatment is likely to be complicated by recurrence of the disease in the engrafted liver, with the pace at which the liver deteriorates being difficult to predict. For these reasons, some patients will be considered for serial hepatic and marrow transplantation/​HSCT. There is evidence that splenectomy may reduce the haemolytic component of end-​ stage protoporphyria, hence consideration should be given to the simultaneous removal of the enlarged spleen at the time of the liver transplantation. In some patients with end-​stage liver disease due to proto­ porphyria, a bizarre neurological syndrome has been identified. Axonal neuropathies requiring mechanical ventilation and cranial nerve palsies have been reported in the perioperative period. Under these circumstances, coproporphyrin and uroporphyrins appear in the urine and may account for a blistering photosensitivity in end-​ stage protoporphyric liver disease. Operative treatment in patients with protoporphyria can be very dangerous as a result of phototoxic injury to visceral tissues and mucous membranes exposed to brilliant vertical lighting in the operating theatre. Surgical lights are best attenuated by the use of filters that reduce spectral power output below 530 nm; such precau- tions should be used throughout the perioperative period to reduce overall phototoxicity in the clinical environment. Management of photosensitivity Photosensitivity is managed by avoiding excessive light exposure, remembering that visible light of exciting green and violet wave- lengths traverses ordinary window glass. Effective sunscreen prepar- ations may assist management, especially in young children at risk. For many years, β-​carotene has been given to patients with protoporphyria: it may absorb light energy at the appropriate wave- lengths and also serve as a free-​radical quenching agent. The prep- aration Lumitene at a dosage of 120 to 180 mg/​day is normally used. This causes orange staining of the skin due to carotenaemia, but is otherwise well tolerated and may improve tolerance to sunlight when plasma carotene concentrations between 10 and 15 µmol/​ litre are achieved. A recent review of 20 studies of concluded that β-​carotene had only small or marginal objective benefit for the light-​ induced symptoms of protoporphyria. Melanin, in the form of eumelanin, quenches ultraviolet light and scavenges free radicals; it also acts as a neutral density filter that re- duces all wavelengths of light. Moreover, melanogenesis may pro- vide a major antioxidant drive. Increasing melanin formation in the epidermis by narrow-​band phototherapy has been shown to im- prove phototoxic symptoms in five patients with protoporphyria, whose tolerance of a standardized source of high-​radiance xenon light improved progressively over 120 days. While the narrow-​band phototherapy requires careful exposure protocols and is reserved for specialist centres, the benefit gained provides proof of concept for the exploration of other means to stimulate melanin synthesis. Administration of a depot preparation of afamelanotide, an α-​ melanocyte-​stimulating hormone analogue, has been explored in clinical trials in patients with protoporphyria in Europe and the United States of America. Compared to placebo, subcutaneous implants of the drug given every 60  days increased the patients’ hours of direct exposure to sunlight, greatly reduced the number of phototoxic reactions, and the patients’ quality of life improved. Adverse events were mostly mild, serious adverse events were not considered to be related to the afamelanotide, and it has received marketing approval as Scenesse for this indication in the European Union since 2014. Treatment of an acute porphyric attack It is essential to establish that the symptoms complained of are caused by an acute attack of porphyria. Of key importance is the careful la- boratory analysis of urine and blood early in the course of the illness. This demonstrates elevated concentrations of porphyrins and haem precursors typified by elevated urinary 5-​aminolaevulinate and porphobilinogen, which should be high in an attack of acute por- phyria. The urine sample should be freshly taken from the patient and protected from light before analysis to avoid nonenzymatic conversion of the porphyrin precursors to porphyrins and hence misdiagnosis. Immediate management An immediate and fastidious review of avoidable factors that would precipitate or aggravate an attack is mandatory. The precipitating factors are usually drugs, alcohol, exogenous or endogenous hor- monal changes, fasting (including that due to dieting), or recent surgical procedures. More than 100 drugs may induce attacks of porphyria (Boxes 12.5.1 and 12.5.2). Particular care should be taken to exclude agents that are obtained over the counter as tonics or herbal remedies. Any agent that might be implicated should be stopped immediately. Abdominal pain and distress, together with anxiety, require prompt treatment; opiates which are safe in porphyria may be useful, although they often exacerbate constipation. Opiates may be com- bined with phenothiazine tranquillizers such as chlorpromazine, which may usefully potentiate their action. Since starvation induces attacks of porphyria and haem biosyn- thesis may be suppressed by the ingestion of carbohydrate, it is advised that patients with minor attacks should eat regular meals containing carbohydrate in a complex form such as starch for its

12.5  The porphyrias 2051 slow release. One-​half to two-​thirds of the energy intake should be derived from ingested carbohydrate. The management of an acute attack should involve repeated moni- toring for the development of hyponatraemia, which may be very se- vere as a result of inappropriate secretion of antidiuretic hormone. In the past, intravenous glucose or fructose solutions have been ad- vocated as a means to suppress haem biosynthesis in the liver. Great caution is needed in the use of these agents, either as 5 or 20% solu- tions, since they exacerbate hyponatraemia and may cause fatal cere- bral oedema. In the author’s view, if the patient is sufficiently unwell not to be able to control the attack with oral carbohydrate-​rich food, parenteral preparations of haem such as haem arginate, rather than glucose or other sugar solutions, should be administered. Haem therapy Haem arginate is administered by a short intravenous infusion in porphyric crises of sufficient severity to merit hospital admission or those associated with limiting pain or metabolic disturbance. Haem arginate supplied by Orphan Europe (see ‘Sources of information) is provided as a stable 25 mg/​ml concentrate and should be adminis- tered at a dosage of 3 mg/​kg body weight (to a maximum dose of 250 mg) once daily for up to 4 days. It should be given in 100 ml physio- logical saline infused into a large vein over at least 30 min. Haem arginate, like all preparations of haem, tends to polymerize and is unstable, hence the administration should be completed within 1 h after diluting the concentrate, the shelf-​life of which is about 2 years. In the United States of America, haemin appears to be a compar- able preparation for suppressing hepatic haem synthesis and cor- recting the metabolic disturbance of the acute attack. Haem arginate and a preparation of haem albumin are apparently somewhat more stable than haemin, which tends to produce phlebitis or interfere with the action of coagulant proteins. Recovery from an acute attack depends on the degree of damage to the nervous system and may occur within 1 or 2 days if haem therapy is introduced at the outset. Certain proof of clinical benefit of haem treatment is lacking, but there is sufficient evidence of its benefit for it to be approved in 19 countries, including the United Kingdom. Haem arginate has a rapid effect on the excretion of aminolaevulinate and porphobilinogen in acute porphyria, and retrospective studies suggest that outcomes are better than that in patients previously documented before the use of the agent. Moreover, the results of a double-​blind study com- paring placebo and haem therapy showed a trend in favour of haem arginate in terms of duration of hospital stay and the requirement for pain relief, although the differences did not quite reach statistical significance in the limited study of 12 patients. On the balance of probabilities, however, the evidence for a beneficial effect of haem arginate therapy, particularly at the onset of a porphyric attack, is compelling. Haem therapy should be used in any patient with significant hyponatraemia, incipient neuropathy, seizures, or bulbar paralysis, and in any patient with severe symptoms, particularly of abdominal pain. It must be recognized that patients with established neuropathy may take many months or even years to recover from an attack and, if it is to be effective, haem therapy should be introduced early. Occasional patients, usually women, are seen in whom repeated acute attacks occur irrespective of the use of one or two courses of haem arginate. The reason for this is unknown, but it is pos- sible that haem arginate therapy induces tachyphylaxis as a result of exaggerated oscillation of haem catabolism by the induction of haem oxygenase in the liver. Tin protoporphyrin, an inhibitor of haem oxygenase, has been considered in this circumstance. This agent is only available in specialist centres and, because it contains toxic heavy metal and itself may induce photosensitivity, is currently not recommended for routine use. Hypersensitivity reactions to haem arginate are rare and the drug has been used during attacks in pregnant women without injury to either the mother or the child. Haem contains 10% by weight of iron and the maximum daily dose of haem arginate would contain only 23 mg of elemental iron; the development of iron storage disease is unlikely, except in very rare instances where an acutely ill patient receives numerous infu- sions of haematin over prolonged periods. However, patients who have received regular haem arginate infusions over more than 10 years have developed iron accumulation, with increased serum ferritin concentrations and magnetic resonance signals indicating iron accumulation in the liver, spleen, and bone marrow. Such iron accumulation has been associated with histological evidence of hep- atic fibrosis and emphasizes the need for careful evaluation of the relative harm and benefits of treating refractory disease by haem arginate or liver transplantation. Carbohydrate loading Where haem therapy is not available, parenteral carbohydrate loading is the only alternative treatment for an acute attack; 2 litres of a 20% weight per volume glucose solution is recommended over a 24-​h period, administered through a central venous catheter. There are risks from giving such therapy as outlined previously and in the author’s opinion the treatment has been superseded by the introduc- tion of stable preparations of haem. Management of complications of acute porphyric attacks Hypertension Hypertension is frequent in porphyric attacks and may be very se- vere as a result of sympathetic overactivity; during the attack, sinus tachycardia is frequent. β-​Blockers are effective in the control of the hypertension and labetalol and propranolol are safe; they also relieve the sinus tachycardia. Hyponatraemia and seizures Hyponatraemia may be very severe and in acute porphyria pro- gresses on a daily basis during the course of the acute attack in most patients. The rapid onset of severe hyponatraemia clearly contributes to confusion and other mental symptoms associated with a porphyric attack. Prompt treatment by fluid restriction and appropriate careful use of hypertonic saline is needed (see Chapter 22.2.1). The place- ment of a patient with porphyric abdominal pain on a surgical ward, with the almost inevitable administration of an intravenous infusion of 5% dextrose, may contribute to death as a result of cerebral oe- dema or the complications of rapid-​onset hyponatraemia. Grand mal seizures in acute porphyric attacks pose a par- ticular problem for management; they are often precipitated by the hyponatraemia that frequently complicates an acute attack. Clearly, from every aspect, prevention is the optimal course of action and appropriate management of the electrolytic abnormality is an

section 12  Metabolic disorders 2052 essential element of treatment. Even in an era of greater awareness, fits and status epilepticus occur and pose special difficulties. Seizures have been treated successfully with parenteral diazepam or the re- lated benzodiazepine, temazepam. Carbamazepine, lorazepam, and midazolam are probably (but not definitely) safe in acute por- phyria. Clonazepam and valproate have each been used for seizure prevention; the generally outmoded therapy of bromide may also have a role. Acetazolamide, which has been used as a minor agent in seizure prophylaxis, has been used safely in acute porphyria, but many first-​line drugs such as carbamazepine, sodium valproate, phenytoin, and chloral hydrate have been classified as unsafe or are frankly porphyrinogenic. Primidone and phenobarbitone are abso- lutely forbidden. Further problems arise in the management of acutely disturbed patients who are not responsive to the safe but outdated phenothia- zine, chlorpromazine. Thioridazine is classed as unsafe, but paren- teral haloperidol was used with good effect in a few patients with uncontrollable or life-​threatening manic aggression and paranoid disturbance during their acute attack. In all instances, prescription of any agent to a patient who has had or is having an acute porphyric crisis must involve consultation with a reliable pharmacopoeia with individual drugs categorized for safety (see Boxes 12.5.1–12.5.3). Cimetidine The ability of most drugs to initiate attacks of porphyria appears in many instances to be related to their effects on the induction of haem biosynthesis in the liver and specifically for the formation of the relevant P450 xenobiotic-​metabolizing isoforms. One key isoform involved in the induction of porphyria is inhibited, at least in vitro, by the H2 antagonist cimetidine. It has been reported that cimetidine at 400 to 800 mg daily is sufficient to inhibit induction of this P450 isozyme in adult humans. Cimetidine has been ad- ministered with occasional success as a means to inhibit or control spontaneous porphyric crises and as a last resort it might be con- sidered in patients with life-​threatening and otherwise uncontrol- lable disease. Prevention and management of recurrent acute porphyric attacks Hormonal interventions Young women with cyclical porphyric attacks may benefit tem- porarily from hormonal intervention by the use of gonadotropin-​ releasing hormone analogues such as goserelin or buserelin. These agents inhibit androgen, oestrogen, and progestogen production and as a result they induce menopausal-​like symptoms and depres- sion, as well as rapid decreases in trabecular bone density. Doses sufficient to suppress luteinizing and follicle-​stimulating hormone concentrations in serum are required. Prolonged use for more than a few months is not recommended, but buserelin may be used intranasally and may be more convenient. To avoid the worst aspects of hypogonadism in women, low-​dose oestrogen therapy under ap- propriate gynaecological supervision may be coadministered once cyclical porphyric attacks have come under control. Given the risk of accelerated skeletal demineralization and osteoporosis, it is prudent, especially in women, to monitor bone mineralization density in pa- tients receiving gonadotrophin-​releasing hormone analogues and to use appropriate bone conserving therapy as advised. Acute perimenstrual attacks can be controlled by the prompt ad- ministration of haem arginate for 1 to 2 days at the predicted time of susceptibility. Liver transplantation The combination of recurrent life-​threatening porphyric attacks and poor venous access for administration of therapeutic haem prepar- ations, or unresponsiveness to haem treatment, has led to the use of liver transplantation in a few young women with this disease. This approach can be successful, although it should be reserved for those who are able to cooperate with the peri-​ and postoperative surgical regimens. Scrupulous attention to removing all definable risk fac- tors, including smoking, is clearly necessary before such measures are considered. The first successful transplant was carried out in 2002 in a young woman with severe frequent neurovisceral attacks which were cured by the procedure. A subsequent retrospective review of 10 patients (9 women) aged 18 to 50 years reported that clinical and biochem- ical remission occurred in all patients, with urinary porphyrin precursors returning to the healthy reference range within 72 h of surgery. In a few cases, patients with renal failure requiring haemo- dialysis and associated hypertension have undergone simultaneous liver and kidney transplantation. Gene therapies Complementing, liver-​directed gene therapy Gene delivery of porphobilinogen deaminase (hydroxymethylbilane synthase) to hepatocytes using a recombinant adeno-​associated virus vector serotype 5 (rAAV2/​5-​PBG) in a murine model of acute intermittent porphyria prevented acute porphyric attacks after chal- lenge of the animals with barbiturate. In a subsequent human phase I, open label, dose-​escalation, multicentre clinical trial, administra- tion of rAAV2/​5-​PBGD to patients with severe acute intermittent porphyria was safe, but metabolic correction was not achieved at the doses tested although two out of eight patients were able to stop haemin infusions. However, securing long-​term correction of the liver would require permanent transduction of the entire organ and this challenge has yet to be overcome with the recombinant adeno-​ associated viral system. RNA interference Short-​interfering RNA molecules (siRNAs) are able to modulate gene expression. Experiments conducted in mice that model acute inter- mittent porphyria have explored an investigational RNA interference agent (givosiran) that specifically targets hepatic ALAS1 gene expres- sion: it was able to prevent and curtail the biochemical abnormalities and paralysis associated with barbiturate challenge in affected animals. Givosiran is currently undergoing late-​phase clinical development for the treatment of acute hepatic porphyria: monthly subcutaneous ad- ministration suppresses pathologically induced liver ALAS1 activity in a sustained manner, thereby decreasing aminolaevulinate and porphobilinogen to near normal concentrations in plasma and urine. At the time of writing, givosiran has been granted ‘Breakthrough Therapy’ designation by the Food and Drug Administration in the United States of America and PRIME designation by the European Medicines Agency; it has also been granted orphan drug designations in both the United States of America and the European Union for the

12.5  The porphyrias 2053 treatment of acute porphyria. Its safety and efficacy are under investi- gation in the ENVISION phase III clinical trial and an ongoing phase I/​II study. The outcomes of these studies have yet to be formally evalu- ated by the regulatory authorities, but interim updates indicate that annualized rates of acute porphyric attacks and of haemin use are both reduced by over 90%, with no significant safety concerns excepting an anaphylactic reaction in a single patient. Given the striking efficacy and relatively sustained and salutary effects of givosiran in patients with severe recurrent acute porphyria under- going clinical trials, it seems likely that those who are able to gain access to this innovative RNA interference therapy will no longer be looking to liver transplantation or repeated use of haem infusions to control their disease. At the time of writing (2019), long-term safety data and reim- bursement stratagems for this are incompletely worked out; however, it seems very likely that the recommended management of all but the most mild or occasional disease will include siRNA therapy. Sources of information Drugs, drug interactions, and safe prescribing British National Formulary, British Medical Association, Tavistock Square, London WC1H 9JP. United Kingdom and Royal Pharmaceutical Society of Great Britain, 1 Lambeth High Street, London SE1 7JN. The Drug Database for Acute Porphyria:  http://​www.drugs-​ porphyria.org/​. The United Kingdom Drug Information Pharmacists Group: http://​www.ukdipg.org.uk. The Welsh Medicines Information Centre (WMIC) provides ad- vice on drug safety in acute porphyria. The Cardiff safe drug list is available via their website. Contact is by telephone: +44 (0)29 2074 3877 or fax: +44 (0)29 2074 3879. Haem arginate (Normosang) is supplied in the United Kingdom by Orphan Europe (UK) Limited: 200 Brook Drive, Green Park, Reading, Berkshire, RG2 6UB, UK. http://​www.orphan-​europe. com. Telephone: +44 (0)1491 414 333. Medical Information e-​mail: infoUK@orphan-​europe.com; stock availability:  krobinson@ orphan-​europe.com Advice about management of acute porphyria National Acute Porphyria Service—​Cardiff & Vale University Health Board and Kings College Hospital London are designated by the Advisory Group for National Specialised Services, NHS England, to provide a national service for patients with active acute porphyria (acute intermittent porphyria, variegate porphyria, her- editary coproporphyria). Two further Regional Porphyria Centres provide services (in Salford and Leeds) for patients who have recently had a new acute attack or with recurrent acute attacks. They provide clinical advice and support to healthcare professionals within the patient’s own hospital. 24-​h emergency telephone: 029 2074 7747. Patient associations The British Porphyria Association, 136, Devonshire Road, Durham City, DH1 2BL, UK. http://​www.porphyria.org.uk; email: helpline@ porphyria.org.uk The American Porphyria and Canadian Porphyria Foundations may be accessed via the Internet websites. Warning jewellery:  it is often valuable in patients with acute porphyrias for them to have a wrist bracelet or neck pendant that provides information about diagnosis in medical emergencies. Details in the United Kingdom can be obtained from The MedicAlert Foundation, 12 Bridge Wharf, 156 Caledonian Road, London N1 9UU. Telephone: +44 (0)207 833 3034. FURTHER READING Anderson KE, et al. (2004). Disorders of heme biosynthesis: X-​linked sideroblastic anemia and the porphyrias. In: Scriver CR, et al. (eds) The metabolic and molecular bases of inherited disease, 8th edi- tion, vol. 2, pp. 2991–​3062. McGraw-​Hill, New York. http://​www. ommbid.com. Balwani M, et al. (2017). Clinical, biochemical, and genetic characteriza- tion of North American patients with erythropoietic protoporphyria and X-​linked protoporphyria. JAMA Dermatol, 153, 789–​96. Balwani M, et  al. (2017). Acute hepatic porphyrias:  recommenda- tions for evaluation and long-​term management. Hepatology, 66, 1314–​22. Bylesjö I, Wikberg A, Andersson C (2009). Clinical aspects of acute intermittent porphyria in northern Sweden:  a population-​based study. Scand J Clin Lab Invest, 69, 612–​18. Chiabrando D, Mercurio S, Tolosano E (2014). Heme and erythropoi- esis: more than a structural role. Haematologica, 99, 973–​83. Collins P, Ferguson J (1995). Narrow-​band UVB (TL-​01) photo- therapy: an effective preventative treatment for the photdermatoses. Br J Dermatol, 132, 956–​63. Cox TM (2007). The porphyrias. In: Lomas D (ed) Horizons in medi- cine, vol. 19, pp. 67–​83. Royal College of Physicians, London. Elder G, et al. (2013). The incidence of inherited porphyrias in Europe. J Inherit Metab Disease, 36, 849–​57. Elder GH, Smith SG, Smyth SJ (1990). Laboratory investigation of the porphyrias. Ann Clin Biochem, 27, 395–​412. Fontanellas A, Ávila MA, Berraondo P (2016). Emerging therapies for acute intermittent porphyria. Expert Rev Mol Med, 18, e17. Gorchein A (1997). Drug treatment in acute porphyrias. Br J Clin Pharmacol, 44, 427–​34. Handshin C, et al. (2005). Nutritional regulation of hepatic heme syn- thesis and porphyria through PGC-​1α. Cell, 122, 505–​15. Holme SA, et  al. (2006). Erythropoietic protoporphyria in the U.K.: clinical features and effect on quality of life. Br J Dermatol, 155, 574–​81. Innala E, et al. (2010). Evaluation of gonadotropin-​releasing hormone agonist treatment for prevention of menstrual-​related attacks in acute porphyria. Acta Obstetrica Gynecol Scandanvica, 89, 95–​100. Kauppinen R, Mustajoki P (1992). Prognosis of acute porphyrias: oc- currence of acute attacks, precipitating factors, and associated dis- eases. Medicine (Baltimore), 71, 1–​13. Kauppinen R (2005). Porphyrias. Lancet, 365, 241–​52. Langendonk JG, et  al. (2015). Afamelanotide for erythropoietic protoporphyria. N Engl J Med, 373, 48–​59. Lenglet H, et al. (2018). From a dominant to an oligogenic model of inheritance with environmental modifiers in acute intermittent por- phyria. Hum Mol Genet, 27, 1164–​73. Marsden JT, et al. (2015). Audit of the use of regular haem arginate in- fusions in patients with acute porphyria to prevent recurrent symp- toms. JIMD Rep, 22, 57–​65.

section 12  Metabolic disorders 2054 Millward LM, et  al. (2001). Self-​rated psychosocial consequences
and quality of life in the acute porphyrias. J Inherit Metab Dis, 24, 733–​47. Mustajoki P, Nordmann Y (1993). Early administration of heme arginate for acute porphyric attacks. Arch Int Med, 153, 2004–​8. Neeleman RA, et al. (2018). Medical and financial burden of acute intermittent porphyria. J Inherit Metab Dis, 41, 809–​17. Peoc’h K, et  al. (2018). Hepatocellular carcinoma in acute hepatic porphyrias: a Damocles sword. Mol Genet Metab, Oct 9. doi: 10.1016/​ j.ymgme.2018.10.001. Pischik E, Kauppinen R (2015). An update of clinical management of acute intermittent porphyria. Appl Clin Genet, 8, 201–​14. Poh-​Fitzpatrick MB (1985). Porphyrin-​sensitized cutaneous photo- sensitivity: pathogenesis and treatment. Clin Dermatol, 3, 41–​82. Puy H, Gouya L, Deybach JC (2010). Porphyrias. Lancet, 375, 924–​37. Sardh E, Harper P, Balwani M, et al. (2019). Phase 1 trial of an RNA interference therapy for acute intermittent porphyria. New Engl J Med, 380, 549–58. Schmid R (1998). The porphyrias. Semin Liver Dis, 18, 1–​101. Shaw PH, et al. (2001). Treatment of congenital erythropoietic por- phyria in children by allogeneic stem cell transplantation: a case re- port and review of the literature. Bone Marrow Transplant, 27, 101–​5. Singal AK, et al. (2014). Liver transplantation in the management of porphyria. Hepatology, 60, 1082–​9. Stein PE, Badminton MN, Rees DC (2017). Update review of the acute porphyrias. Br J Haematol, 176, 527–​38 Sylantiev C, et al. (2005). Acute neuropathy mimicking porphyria in- duced by aminolevulinic acid during photodynamic therapy. Muscle Nerve, 31, 390–​3. Urquiza P, et al. (2018). Repurposing ciclopirox as a pharmacological chaperone in a model of congenital erythropoietic porphyria. Sci Transl Med, 10, eaat7467. Whitman JC, Paw BH, Chung J (2018). The role of ClpX in erythropoi- etic protoporphyria. Hematol Transfus Cell Ther, 40, 182–​8. Windon AL, et al. (2018). Erythropoietic protoporphyria in an adult with sequential liver and hematopoietic stem cell transplantation Am J Transplant, 18, 745–​9. Yin L, et al. (2007). Rev-​erbα, a heme sensor that coordinates meta- bolic and circadian pathways. Science, 318, 1786–​9.

12.6 Lipid disorders 2055

12.6 Lipid disorders 2055

ESSENTIALS High blood cholesterol and high blood triglycerides are causal risk factors for atherosclerotic cardiovascular disease, which remains the leading cause of death in the developed world. Lipid and lipoprotein metabolism Cholesterol, triglycerides, and fat-​soluble vitamins are transported with specific proteins in the blood as multimeric complexes called lipoproteins. Lipid and lipoprotein metabolism are effected by three principal physiological processes: (1) intestinal absorption of dietary lipid and transport in the blood of dietary lipid and lipids, principally derived from the liver (as triglyceride-​rich lipoproteins) to peripheral tissues for catabolism by skeletal and cardiac muscle or storage in adipose tissue; (2) return of triglyceride-​rich lipoprotein remnants to the liver, hepatic synthesis of low-​density lipoprotein (LDL), and the transport of cholesterol between peripheral tissues and the liver; and (3) reverse cholesterol transport by high-​density lipoprotein (HDL) between peripheral tissues and the liver. Dyslipidaemias are disorders of lipoprotein metabolism in which there is elevation of total cholesterol and/​or triglycerides, often ac- companied by reduced levels of HDL cholesterol. They are caused by a combination of genetic (primary) and acquired (secondary) factors (lifestyle, metabolic conditions, and drugs). Causes of dyslipidaemia Particular lipid disorders—​(1) polygenic hypercholesterolaemia—​in people with predominantly hypercholesterolaemia who do not show Mendelian inheritance, and do not have the clinical features of primary hypercholesterolaemia syndromes, polygenic hyperchol- esterolaemia is likely. (2)  Familial hypercholesterolaemia (FH)—​an autosomal codominant disorder most commonly caused by dele- terious mutation of the LDLR gene which encodes the LDL receptor. Premature atherosclerotic cardiovascular disease is very common. (3)  Combined hypercholesterolaemia and hypertriglyceridaemia—​ mixed (combined) dyslipidaemia is common and caused by a combination of genetic (primary) and acquired (secondary) fac- tors. Elevated fasting triglycerides, increased total cholesterol and low HDL concentration are commonly associated with athero- sclerotic vascular disease. (4) Familial combined hyperlipidaemia—​ polygenic and non-​Mendelian. (5) Familial dysbetalipoproteinaemia (also called type 3 hyperlipoproteinaemia)—​a Mendelian reces- sive disorder, which becomes manifest when an acquired cause of dyslipidaemia also occurs. (6) Severe hypertriglyceridaemia—​can be due to overproduction of very LDL, defective peripheral lipolysis, and/​or reduced triglyceride uptake. Can be associated with recurrent pancreatitis. Secondary or aggravating factors—​these include excess alcohol, diabetes mellitus, obesity and insulin resistance, hypothyroidism, chronic kidney disease, nephrotic syndrome, Cushing’s syndrome, certain drugs (such as β-​adrenergic blocking agents or thiazide diur- etics), and liver disease. Management of dyslipidaemia The key questions are: (1) what classes of lipoproteins and lipids are increased or decreased in the patient’s plasma? (2) Does the pa- tient has a primary (genetic) or secondary (acquired) dyslipidaemia (often contributions from both influences)? (3)  Is the patient at risk of atherosclerotic cardiovascular disease or acute pancrea- titis? (4)  What other risk factors (e.g. hypertension or diabetes) are present? (5) What treatments might be used to address these abnormalities? Hypercholesterolaemia In those with a 10% or greater risk of a cardiovascular event in the next 10 years, according to United Kingdom guidelines, it is usually recommended that cholesterol should be reduced. In the United States of America, a threshold of 7.5% for treatment is used, with the option for treatment at 5% risk. The National Institute for Health and Care Excellence in the United Kingdom currently recommends target reductions greater than 40% of non-​ HDL cholesterol (an accurate predictor of cardiovascular risk), but such targets are no longer recommended in the United States of America. The principal objectives of treatment are primary and secondary prevention of atherosclerotic cardiovascular disease and its complications. Lifestyle—​an atheroprotective lifestyle should be instituted, with encouragement of weight loss in overweight and obese individuals. Statins—​HMG-​CoA reductase inhibitors are the first-​line drugs for the treatment of hypercholesterolaemia, with much evidence sup- porting their use in reducing the risk of atherosclerotic cardiovascular disease. 12.6 Lipid disorders Jaimini Cegla and James Scott

section 12  Metabolic disorders 2056 Other drugs—​these include ezetimibe (blocks the action of the NPC1L1 protein and intestinal cholesterol absorption, bile acid sequestrants (resins), nicotinic acid, and PCSK9 inhibitors (fully humananized monoclonal antibodies that block the protease, PCSK9, at the surface of hepatocytes thereby reducing the physio- logical degradation of the LDL receptor). LDL apheresis—​generally restricted to homozygous FH patients and heterozygotes with rapidly progressive cardiovascular disease. Raised lipoprotein(a)—​associated with seriously increased risk of atherosclerotic cardiovascular disease. Specific treatment is chal- lenging as treatment of acquired factors does not affect concentra- tions of lipoprotein(a). Nicotinic acid (no longer readily available in Europe and the United Kingdom) and PCSK9 inhibitors have some effect. An antisense lipoprotein(a) mRNA inhibitor has proved re- markably efficacious in early clinical trials and is in late-​phase clinical development. Hypertriglyceridaemia Triglyceride concentrations greater than 10 mmol/​litre (900 mg/​dl) cause an increased risk of acute pancreatitis and require prompt treatment. A secondary treatment goal is to reduce the risk of ath- erosclerotic cardiovascular disease. In addition to diabetes mellitus, other causes including dietary in- discretion with high fructose intake, obesity, insulin resistance, excess alcohol consumption, reproductive hormone deficiency, or medical use of steroid hormones should be considered. Lifestyle changes—​these often reduce plasma triglyceride concen- trations markedly. A reasonable dietary goal is to restrict total fat in- take to around 20–​30 g daily and avoid refined carbohydrate. Drug treatment—​includes fibrates (the drugs of first choice), omega-​ 3 fatty acids, and statins (can reduce modest hypertriglyceridaemia but have no value in severe hypertriglyceridaemia). Introduction Despite a more than 50% reduced prevalence over the past 50 years (Fig. 12.6.1), atherosclerotic cardiovascular disease remains the leading cause of death in the developed world. Atherosclerotic car- diovascular disease accounts for about a third of all deaths, and 60% of people will suffer major life-​threatening cardiovascular events. By contrast with the developed world, the occurrence of atherosclerotic cardiovascular disease is increasing in the developing world where it is becoming a leading course of mortality. Atherosclerotic cardiovascular disease is a disease of large-​ and medium-​sized arteries. It develops slowly over many years as a direct consequence of several major risks factors. Evidence from epidemiology, meta-​analyses of cholesterol-​lowering end-​point clinical trials, human and animal genetics, and pathology over- whelmingly demonstrate the prominence of elevated levels of high blood cholesterol as a causal risk factor for atherosclerotic cardio- vascular disease. Cholesterol is termed the ‘agent provocateur’ of atherosclerosis, because its accumulation as oxidized low-​density lipoprotein (LDL) cholesterol within macrophages in the artery wall has a direct patho- genetic role in atherosclerosis. Cholesterol-​laden macrophages have a characteristic foamy appearance. ‘Foam cells’ are a hallmark feature of atherosclerosis (Fig. 12.6.2) (see Chapter 16.13.1). High blood triglycerides (TGs) are also causally linked to atherosclerotic cardiovascular disease. Unhealthy diet, lack of exercise, hyperten- sion, diabetes, and smoking are other major risk factors associated with atherosclerotic cardiovascular disease. Improvements in lifestyle and pharmaceutical treatment to re- duce blood cholesterol and other risk factors for atherosclerotic car- diovascular disease in part account (Fig. 12.6.1) for the decrease in disease prevalence in the developed world. In the countries of the developing world, the risk factors for atherosclerotic cardiovascular disease are becoming more prevalent as their populations espouse a Western lifestyle, and this is likely to be the cause for the increase in disease. Cholesterol, TGs, and fat-​soluble vitamins are transported with specific proteins in the blood as multimeric complexes called lipoproteins. Here the normal physiology of lipoproteins will be described, together with their pathophysiology in relation to ath- erosclerotic cardiovascular disease and other diseases, particularly acute pancreatitis. The importance of genetic, dietary, and acquired factors which influence lipoprotein metabolism, together with diagnostic and treatment approaches to disease prevention will be discussed. Lipoproteins, lipids, and apolipoproteins Lipoproteins Lipoproteins are large, multimeric complexes of lipid and specific proteins called apolipoproteins (see ‘Apolipoproteins’). They trans- port lipid in plasma and other bodily fluids (lymph, interstitial fluid) between metabolically active tissues (Table 12.6.1, Fig. 12.6.3) (see ‘Lipid and lipoprotein metabolism’). The outer shell of lipoproteins consists of amphipathic (having hydrophilic and hydrophobic parts) lipids (phospholipid and free cholesterol) and apolipoproteins surrounding a core of water-​ insoluble, hydrophobic lipids (TG and cholesteryl ester). Lipoproteins are grouped into five classes (Table 12.6.1, Fig. 12.6.3) according to their density (assessed by ultracentrifugation), which determines their size (assessed by MRI and nondenaturing electro- phoresis). The amount of lipid in a lipoprotein determines its density, because lipids are not as dense as water. The lipid and apolipoprotein composition of lipoproteins differs (Table 12.6.1, Fig. 12.6.4). Chylomicrons contain the most lipid and are the least dense; they are amongst the largest entities se- creted from eukaryotic cells. Chylomicron remnants, the product of peripheral lipolysis of chylomicron TG, remain very large. Very low-​density lipoproteins (VLDLs), VLDL remnant intermediate-​ density lipoproteins (IDLs), LDLs, and high-​density lipoproteins (HDLs) are increasingly dense and less buoyant and have lower lipid content. LDL and HDL also vary in size and lipid compos- itions and this affects their atherogenicity (see ‘Lipid and lipopro- tein metabolism’). Lipoprotein(a) Lipoprotein(a) (Lp(a)) is a LDL particle with a single molecule of apolipoprotein(a) (apo(a)) linked to apoB100 by a disulphide bridge (Fig. 12.6.5). Its physiological function is uncertain. Lp(a) is not

12.6  Lipid disorders 2057 fully formed until the apo(a) protein is for the main part conju- gated with LDL in the extrahepatic space, though some assembly may occur with intracellular LDL-​sized particles. Apo(a) is similar to plasminogen; the genes reside together in the genome, but apo(a) has no catalytic activity. In addition to the plasminogen domain, it contains repeated domains called kringles, after a Belgian cake, which is similar in shape to apo(a). Apo(a) proteins vary due to a size polymorphism, and a variable number of kringle IV repeats (each of 114 amino acids) in the gene. This results in apo(a) proteins with 10 to 50 or more kringle IV re- peats. There is an inverse correlation between the size of the apo(a) isoform and the Lp(a) plasma concentration. The larger the isoform, the slower the rate of production, which limits the plasma concen- tration. Particle number is the now favoured measure as this reflects atherogenicity (Fig. 12.6.6). Lp(a) is cleared mainly by the liver but the receptor has not been determined. Some may be cleared by the kidney because Lp(a) plasma levels increase in chronic renal failure. 1600 (a) 1200 800 400 0 Age-standardized cardiovascular disease death rate per 100,000 Men Women Year 1950 1960 1970 1980 1990 2000 2010 1950 1960 1970 1980 1990 2000 2010 Hungary Japan USA France Greece Australia Finland Norway Congenital abnormalities 73 deaths Suicide 952 males 249 females Coronary heart disease 7,987 males Breast cancer 3,665 females Coronary heart disease 15,466 males 7,394 females Coronary heart disease 17,894 males 21,405 females 25000 (b) 15000 20000 1000 5000 0 Number of deaths 1–4 5–34 35–64 Age (years) 65–79 80+ Men Women Fig. 12.6.1  Cardiovascular death rates in the developed world. (a) Trends in death rates from cardiovascular diseases in adults over 30 years of age in selected countries with vital registration and medical certification of the underlying cause of death. Death rates are age-​standardized to the World Health Organization standard population and smoothed using a
5-​year moving average. (b) Death rates in the United Kingdom by age. Source data from (Panel (a)) Ezzati M and Riboli E. (2012). Can noncommunicable diseases be prevented? Lessons from studies of populations and individuals. Science, 21, 337(6101), 1482–​7. Copyright © 2012, American Association for the Advancement of Science. (Panel (b)) Government data, Office for National Statistics, UK 20 October 2011.

section 12  Metabolic disorders 2058 Lipids The major lipoprotein lipids are TG, phospholipid, free cholesterol, cholesteryl ester, and fat-​soluble vitamins (Fig. 12.6.7). Triglycerides TGs are the essential energy transfer and storage lipids. They com- prise three fatty acids, which can be either saturated or unsaturated fatty acids, esterified to a glycerol backbone, and are water-​insoluble. The main dietary sources of TGs are fatty red meat, poultry skin, lard, high-​fat dairy products, shellfish, and shrimps. Vegetable oils contain TGs. TGs are also the product of endogenous synthesis in the liver from excess dietary carbohydrate. Excess carbohydrate in the liver signals the activation of a carbohydrate responsive tran- scription factor MLXIPL/​CHREBP, and of LXR alpha and sterol regulatory element-​binding protein (SREBP), which together acti- vate the genes encoding the enzymes required for the biosynthesis of fatty acids and, in turn, TGs (Fig. 12.6.8). Hepatic TG is normally secreted as VLDL (see ‘Hepatic lipid transport’). Excess bodily TG is stored as fat depots in adipose tissue, and in liver and other tissues in times of gross nutritional excess—​ overweight and obesity; or burned as a source of energy by skeletal and cardiac muscle. Phospholipids Phospholipids are a major structural component of most bio- logical membranes. They also have an important role in cell sig- nalling through acetyl choline and prostaglandin synthesis, and phosphatidylinositol signalling pathways. They are amphipathic: the head is hydrophilic and the tail hydrophobic. This property enables them to form the lipid bilayer of biological membranes, and the monolayer outer shell of lipoproteins. Like TGs, they comprise a hydrophobic glycerol backbone, but with only two fatty acids bonded to the glycerol (usually one satur- ated and one unsaturated). The third side-​group of glycerol is occu- pied by a hydrophilic phosphate group. Fig. 12.6.2  Foam cells, a hallmark feature of atherosclerotic plaque. Atherosclerotic plaque (AHA stage 4, i.e. with lipid necrotic core and fibrous cap). Brown, immunoperoxidase/​di-​aminobenzidine for CD68 (a macrophage lysosome protein). Blue, haematoxylin counterstain. A series of macrophages are shown, including foam cells in increasing stages of lipid accumulation and foam cell formation. The brown-​ staining extracellular matrix is not background or nonspecific staining but represents residual epitopes on cellular debris derived from dead macrophages. For reference, typically, a leucocyte would be expected to be approximately 10 mm in diameter. The foam cell shown is therefore in the region of 10-​fold the diameter of a typical leucocyte. Photo courtesy of Joseph Boyle, Imperial College London. Table 12.6.1  Major types of lipoproteins in circulating blood HDL LDL VLDL Chylomicrons Physical property Particle size (nm) 4.5–​12 18–​25 30–​75

75 Density (g/​litre) 1064–​1210 1019–​1063 1006 <950 Components Triglycerides 3% 10% 59% 80–​95% Cholesterol (free and esters) 2% 45% 15% 3–​7% Phospholipids 27% 22% 5% 3–​6% Proteins 48% 20–​25% 10% 1–​2% Apolipoproteins A1/​2, C1/​2/​3, E B100,C2/​3, E, (a) B100, C2/​3, E B48, C2/​3, E Other components LCAT, CETP paroxonase Vitamin E Vitamin E Retinyl ester Vitamin E 0.95 1.006 Particle density (g/mL) Particle size, diameter (nm) 1.02 1.06 1.10 1.2 5 VLDL remnants VLDL Chylomicron remnants Chylomicron IDL HDL3 HDL2 sdLDL Lp(a) 18–25 Buoyant LDL 10 20 40 60 80 1000 Fig. 12.6.3  Lipoprotein subclasses classed by particle size and density.

12.6  Lipid disorders 2059 The phosphate groups can be modified with simple organic mol- ecules such as choline to form lecithin or phosphatidylcholine, or serine, ethanolamine, or inositol to form phosphatidylserine, phosphatidylethanolamine, or phosphatidylinositol respectively. Plasmalogens and sphingomyelin are also phospholipids, and constitute more than 20% of total phospholipids in cell membranes. Eggs, organ and lean meats, fish, shellfish, cereal grains, and oil- seeds are the main sources of phospholipids. Their synthesis in ani- mals occurs in the cytosol through endoplasmic reticulum (ER) membrane-​located enzymes. Cholesterol Free cholesterol is an essential structural component of all animal plasma membranes. It is required to maintain membrane structural integrity and fluidity by reason of its packing with phospholipids. The properties of cholesterol are conferred by a single hydroxyl (OH) group, which renders cholesterol polar, but only slightly sol- uble in water. These properties enable cholesterol to function in intracellular transport by caveolae and clathrin-​coated pits and cell signalling through receptor clustering in lipid rafts. Cholesterol serves as a precursor for the biosynthesis of steroid hormones, bile acids, and vitamin D. LDL receptor binding Cysteine rich Lipoprotein assembly Triglyceride cholesteryl-ester Fat-soluble vitamins Other apolipoproteins ApoC5, ApoE Phospholipid Free cholesterol COOH ApoB100 75–1000 nM ApoB100, B48 LDL receptor binding Cysteine rich Triglyceride rich lipoprotein LDL Phospholipid Free cholesterol COOH 20–30 nM ApoB100 NH2 NH2 Fig. 12.6.4  Comparison of a triglyceride rich lipoprotein (e.g. VLDL and chylomicron) and low-​density lipoprotein. LDL receptor binding Cysteine rich Phospholipid Free cholesterol S–S (ApoB100–Lp(a)) COOH 200–300A ApoB100 KIV-3 to 10 Kringle IV type 2 repeats (variable number) (KIV-2) KV Protease (plasminogen homology) NH 2 Fig. 12.6.5  The structure of Lp(a). Lp(a) is an LDL particle with a single molecule of apo(a) linked to apoB100 by a disulfide bridge. Apo(a) contains an inactive protease domain, one kringle V, and kringle IV which is composed of 10 different types based on their amino acid sequences, referred to as KIV types 1 to 10. KIV types 1 and 3 to 10 are present as single copies, whereas KIV type 2 (KIV-​2) is present as multiple copies, varying in number from 3 to more than 40 copies, explaining the isoform size heterogeneity between patients. Lp(a)-P(nmol/L) Lp(a)-Mass(mg/dL) 50 0 100 150 200 250 900 800 700 600 500 400 300 200 100 0 MW = 700–800 MW = 600–700 MW = 300–600 Fig. 12.6.6  Discordance between Lp(a) mass and particle number assays. Comparison of Lp(a) particle number in nmol/​litre (Lp(a)-​P) with Lp(a) mass in mg/​dl for samples of large (blue, n = 51), intermediate (green, n = 25), and small (red, n = 38) Lp(a) isoform sizes. Isoform sizes determined by Western blot analysis. The difference in slopes observed here indicates that the mass assay is influenced by apo(a) isoform size. Source data from Guadagno PA, et al. (2015). Validation of a lipoprotein(a) particle concentration assay by quantitative lipoprotein immunofixation electrophoresis. Clinica Chimica Acta, 439, 219–​24.

section 12  Metabolic disorders 2060 Fig. 12.6.7  The chemical structure of key lipids molecules. Glucose Triglycerides MLXIPL L-PK Citrate Acetyl-CoA Malonyl-CoA FFA FAS LCE SCD ACC ACL Pyruvate Glycolysis Lipogenesis Mitochondria Krebs cycle ApoB100 ApoC2/3/A5 VLDL FFA LPL Adipose tissue Fig. 12.6.8  De novo biosynthesis of fatty acids from carbohydrate. Excess carbohydrate in the liver signals the activation of a carbohydrate responsive transcription factor, MLX-​interacting protein-​like (MLXIPL) (also known as carbohydrate-​responsive element-​binding protein (CHREBP)), which activates the genes encoding the enzymes required for the biosynthesis of fatty acids and in turn triglycerides. ACC, acetyl-​ CoA carboxylase; ACL, ATP-​citrate lyase; FAS, fatty acid synthase; FFA, free fatty acids; LCE, long-​chain fatty acyl elongase; L-​PK, liver pyruvate kinase; LPL, lipoprotein lipase; SCD, stearoyl-​CoA desaturase.

12.6  Lipid disorders 2061 The main dietary sources of cholesterol are the same as TGs. Egg yolks are also rich in cholesterol. Plants make cholesterol in very small amounts. Rather, plants make phytosterols, which are chem- ically similar to cholesterol and can compete with cholesterol for ab- sorption in the intestinal tract, thus potentially reducing cholesterol absorption (see ‘Plant sterols’). Cholesterol is essential for animal life therefore most cells synthe- size cholesterol from acetyl-​coenzyme A (CoA) through a complex multistep pathway. Humans synthesize about 1 g of cholesterol daily, and compensate for any excess absorption by reducing cholesterol synthesis. Thus 12 h after ingestion, cholesterol will show little effect on total body cholesterol content or concentrations of cholesterol in the blood. In the 7 h after ingestion of cholesterol, however, the levels in plasma increase, and this is moderated by genetic factors (see ‘Lipid and lipoprotein metabolism’). Cholesterol is susceptible to oxidation and easily forms oxygen- ated derivatives known as oxysterols. Oxysterols exert inhibitory actions on cholesterol biosynthesis. Oxidized LDL is associated with the pathogenesis of atherosclerotic cardiovascular disease (see ‘Introduction’). Cholesterol is also oxidized by the liver into bile acids, which may be conjugated with glycine, taurine, glucuronic acid, or sulphate (Fig. 12.6.9). A mixture of conjugated and nonconjugated bile acids (c.80% of biliary constituents), with free cholesterol and phospho- lipids (c.20% of bile constituents), is excreted from the liver into the bile, and stored in the gallbladder. The gallbladder empties in response to food in the intestine. In the intestine, bile salts form miscelles, which solubilize lipids and facilitate their absorption. Approximately 95% of the bile acids are reabsorbed from the ileum and the remainder are lost in the faeces. The excretion and reabsorption of bile acids forms the basis of the enterohepatic circulation, which is essential for the digestion and absorption of dietary lipids. The liver also excretes free cholesterol in bile into the duodenum. Typically, about 50% of the excreted choles- terol along with dietary cholesterol is reabsorbed by the small bowel, but this varies between people. The biosynthesis of cholesterol is directly regulated by the choles- terol levels in the cell (Fig. 12.6.10). Thus high dietary intake reduces endogenous production, whereas lower intake does the reverse. The principal regulatory mechanism is the sensing of intracellular chol- esterol in the ER by the proteins SREBP1 and SREBP2. In the pres- ence of cholesterol, SREBP is bound to two other proteins: SREBP cleavage-​activating protein (SCAP) and insulin-​induced gene 1 (INSIG1). At low cholesterol levels, INSIG1 dissociates from the SREBP–​ SCAP complex, which then moves to the Golgi apparatus. In the Golgi apparatus, SREBP is cleaved by site-​1 protease and site-​2 protease (S1P and S2P), two enzymes that are activated by SCAP at low cholesterol levels. After cleavage SREBP migrates to the nu- cleus, where it binds to the sterol regulatory element (SRE), and pro- motes transcription of multiple genes that control lipid formation, HO 7 HO H HO Cholesterol Chenodeoxycholic acid 7α-hydroxylase (CYP7A1) ‘Classic pathway’ HO 7 HO HO H HO HO 7-hydroxycholesterol Cholic acid Several steps Sterol 12α-hydroxylase (CYP8B1) HSDB37 HSDB37 Fig. 12.6.9  Bile acid synthesis. Cholesterol is oxidized by the liver into bile acids. A hydroxyl optionally added in the liver determines the formation of chenodeoxycholic acid versus cholic acid. In the liver, bile acids may be conjugated with glycine, taurine, glucuronic acid, or sulphate. A mixture of conjugated and nonconjugated bile acids, with free cholesterol and phospholipids, is excreted from the liver into the bile, and stored in the gallbladder. In the gut, a hydroxyl group may be removed by gut bacteria to create deoxycholic acid and lithocholic acids from cholic acid and chenodeoxycholic acid respectively.

section 12  Metabolic disorders 2062 metabolism and energy supply, including the LDL receptor (LDLR) and 3-​hydroxy-​3-​methyl-​glutaryl-​CoA reductase (HMGCoAR). The LDLR removes LDL from plasma. HMGCoAR is the rate-​ limiting enzyme of cholesterol biosynthesis. The turnover of HMGCoAR by protein degradation is sensitive to cholesterol levels in the cell (Fig. 12.6.11). The activity of the LDLR is also regulated by protein convertase subtilisin/​kexin type 9 (PCSK9), an extracellular protein which binds to the epidermal growth factor-​like repeat A (EGF-​A) domain of the LDLR, inducing its degradation. After internalization, the LDLR usually disassociates in the acid environment of the endosome and recycles to the cell surface. With PCSK9 bound, the LDLR is targeted to the lysosome and degraded rather than recycled. Cholesteryl esters are formed between the carboxylate group of a fatty acid and the hydroxyl group of cholesterol. Cholesteryl esters have a lower solubility in water due to their increased hydrophobicity. They are the intracellular storage form and intravascular transport form of cholesterol. Plant sterols Phytosterols derived from plants account for 25% of dietary sterols. The most commonly occurring phytosterols in the human diet are β-​sitosterol, campesterol and stigmasterol, which account for about 65%, 30%, and 3% of diet contents, respectively, but are not normally well absorbed by humans. Stanols are saturated sterols, having no double bonds in the sterol ring structure, and normally form only a tiny component of the diet, but are synthesized as a gut microbial byproduct of cholesterol metabolism. Absorbed plant sterols are re-​excreted in bile or by the enterocyte. Plant sources are vegetable oils, nuts, peanuts, and avocados. Plant sterols compete with cholesterol for absorption, with potential beneficial effect on plasma cholesterol levels. SCAP SCAP WD SREBP HLH WD SREBP HLH SREBP HLH INSIG SRE – ACS – FAS – GPAT – Squalene synthase – HMGCoAR – LDLR – etc. High cholesterol Low cholesterol ER membrane Cytoplasm Lumen Cytoplasm Lumen Nucleus S2P S1P Golgi apparatus SREBP HLH Fig. 12.6.10  Regulation of cholesterol biosynthesis. The principal regulatory mechanism is the sensing of intracellular cholesterol in the ER by the proteins, sterol regulatory element-​binding protein (SREBP) 1 and 2. Starting out wrapped in the ER membrane, their activation as transcription factors requires a maturation process tightly controlled by the levels of cholesterol present in the membrane. In the presence of cholesterol, SREBP is bound to two other proteins: SREBP cleavage-​activating protein (SCAP) and insulin-​induced gene 1 (INSIG1). SCAP acts as a sensor of the content of cholesterol in the ER membrane. In the presence of high levels of cholesterol, SCAP remains anchored in the ER membrane due to its interaction with the INSIG proteins. At low cholesterol levels, INSIG1 dissociates from the SREBP–​SCAP complex, which then moves to the Golgi apparatus. SREBP is then cleaved by site-​1 protease and site-​2 protease (S1P and S2P), two enzymes that are activated by SCAP at low cholesterol levels. This fragment contains a basic helix–​loop–​helix (HLH) leucine zipper domain, which functions as a transcription factor. After cleavage, SREBP migrates to the nucleus, where it binds to the sterol regulatory element (SRE), and promotes transcription of multiple genes that control lipid formation, metabolism, and energy supply, including the LDL receptor (LDLR) and 3-​hydroxy-​3-​methyl-​glutaryl-​CoA reductase (HMGCoAR). WD, WD40 domain. Source data from Desvergne B, et al. (2006). Transcriptional regulation of metabolism. Physiological Reviews, 86(2), 465–​514.

12.6  Lipid disorders 2063 Trans fatty acids Artificial trans fats are created in an industrial process that adds hydrogen to liquid vegetable oils to make them more solid. The primary dietary source for trans fats in processed food is ‘partially hydrogenated oils’. They are used by food manufacturers because they have a long shelf life and are able to withstand repeated heating without breaking down. Doughnuts, biscuits, crackers, muffins, pies, and cakes are examples of foods that may contain trans fat. Trans fats may also be produced when ordinary vegetable oils are heated to fry foods at very high temperatures and this is one reason why take- away foods can sometimes be high in trans fats. While many old-​ fashioned margarines contain a high proportion of trans fats, newer brands should contain low amounts. Fat-​soluble vitamins Vitamin E is one of the most abundant lipid-​soluble antioxidants found in the plasma and somatic cells of higher mammals. It comes mainly from plant rather than animal sources. The main dietary forms of preformed vitamin A are carotenoids in fruits and veget- ables and long-​chain fatty acids esters of retinol in foods of animal origin. Apolipoproteins Apolipoproteins function as structural components of lipopro- tein particles (Table 12.6.2). They also act as cofactors for en- zymes of lipid metabolism and as ligands for cell surface receptors. There are two types of apolipoproteins. The first are the huge LDL receptors LDL Protein ER Lysosome Cholesteryl linoleate 1. HMG CoA reductase 2. ACAT 3. LDL receptors Cholesterol Cholesteryl oleate Amino acids Regulatory actions Lysosomal hydrolysis Internalization LDL binding Fig. 12.6.11  Intracellular uptake of LDL by the LDL receptor. LDL is internalized through receptor-​mediated endocytosis. The cholesterol derived from lysosomal hydrolysis exerts feedback to protect the cell from overaccumulation of cholesterol by (1) suppressing activity of 3-​hydroxy-​3-​methylglutaryl-​coenzyme A reductase (HMG-​CoA reductase), the rate-​controlling enzyme of cholesterol biosynthesis; (2) activating acyl-​CoA:cholesterol acyltransferase (ACAT), a cholesterol-​esterifying enzyme so that excess cholesterol can be stored as cholesteryl ester droplets; and (3) suppressing synthesis of new LDL receptors thus preventing further cholesterol intake into the cell. Source data from Goldstein J and Brown M (2009). The LDL receptor. Arterioscler Thromb Vasc Biol, 29, 431–​8. Table 12.6.2  Major apolipoproteins and their function Apolipoprotein Primary source Lipoprotein association Function ApoA1 Intestine, liver HDL, chylomicrons Structural protein HDL Activates LCAT ApoA2 Liver HDL, chylomicrons Structural protein HDL ApoA4 Intestine, liver HDL, chylomicrons Unknown ApoA5 Liver VLDL, chylomicrons Promotes LPL-​mediated TG lipolysis Apo(a) Liver Lp(a) Unknown ApoB48 Intestine Chylomicrons, chylomicron remnants Structural protein for chylomicrons ApoB100 Liver VLDL, IDL, LDL, Lp(a) Structural protein for VLDL, LDL, IDL, Lp(a) Ligand for binding LDL receptor ApoC1 Liver Chylomicrons, VLDL, HDL Unknown ApoC2 Liver Chylomicrons, VLDL, HDL Cofactor for LPL ApoC3 Liver, intestine Chylomicrons, VLDL, HDL Inhibits LPL activity and lipoprotein binding to receptors ApoE Liver Chylomicron remnants, IDL, HDL Ligand for binding to LDL receptor and other receptors

section 12  Metabolic disorders 2064 apoB-​containing lipoproteins, which associate with lipid droplets irreversibly (they do not exchange between lipoprotein particles) from the time of assembly of lipoproteins within the ER and se- cretion from the cell, until they are cleared from the circulation by the LDLR. Lipid binding through apoB is mainly by amphipathic β strands, with less of the amphipathic α helices that characterize other apolipoproteins. The two forms of apoB, apoB100 (514 kDa) and apoB48 (247 kDa), are the products of the same gene, and in humans are synthesized in the liver and intestine respectively. ApoB48 is generated by editing of the APOB100 mRNA by a site-​specific RNA editing cytosine deaminase designated apolipoprotein B mRNA editing enzyme, catalytic polypeptide 1 (apoBEC1), which terminates the transla- tional reading frame, to create the smaller apoB48. ApoB48 is required for TG-​rich chylomicron assembly and se- cretion from the intestine, and for the delivery of TG to peripheral tissues. ApoB48 lacks the C-​terminal LDLR binding domain found in apoB100. The clearance from the circulation of chylomicron remnants after depletion of TGs is mediated by apoE through the LDLR. ApoB100 is required for TG-​rich VLDL assembly and secretion from the liver. After secretion, VLDL goes on to form remnants or IDL and subsequently LDL, which is cleared from the circulation by direct interaction of apoB100 with the LDLR. All other apolipoproteins are very much smaller and make im- portant exchanges between lipoprotein particles in the course of lipoprotein metabolism and remodelling in the circulation. The smaller exchangeable apolipoproteins, the apoAs, apoCs, and apoE form a multigene family, and have similar intron/​exon organiza- tion in their genes. They bind lipid through repeated amphipathic α helices in their structure. They are the products of duplication of an ancestral gene. ApoA1 and apoA2 are the core structural proteins of HDL. ApoA4 is thought to act primarily in intestinal lipid absorption. ApoA5 pos- sibly acts by increasing VLDL production in the liver, stimulation of proteoglycan-​bound lipoprotein lipase (LPL) at the endothelium of capillaries in peripheral organs, or enhancing the clearance of TG-​ rich lipoproteins via lipoprotein receptors in the liver. The function of apoC1 and apoC4 is uncertain. ApoA1 and apoC2 are cofactors for lecithin–​cholesterol acyltransferase (LCAT) and LPL respectively. ApoC3 is an inhibitor of LPL. ApoE is a ligand for the LDLR and mediates the clearance of chylomicron and VLDL remnants or IDLs by the liver LDLRs. Lipid and lipoprotein metabolism Lipid and lipoprotein metabolism can be grouped into a variety of physiological processes: (1) the intestinal absorption of dietary lipid (long-​chain fatty acids, cholesterol, fat-​soluble vitamins) and transport in the blood of dietary lipid and hepatically derived lipids, principally TGs as TG-​rich lipoproteins, to peripheral tis- sues for catabolism by skeletal and cardiac muscle or storage in adipose tissue; (2) the transfer of the TG-​rich lipoproteins rem- nants back to the liver, and the formation of LDL and the transport of cholesterol between peripheral tissues and the liver; and (3) re- verse cholesterol transport by HDL between peripheral tissues and the liver. Intestinal lipid absorption and transport
as chylomicrons Intestinal lipid sources TG is the main lipid in the diet, contributing 90 to 95% of the lipid-​ derived energy. Dietary lipids also include phospholipids, cholesterol and plant sterols, and fat-​soluble vitamins. The main phospholipid in the small intestinal lumen is phosphatidylcholine, mostly derived endogenously from bile, 10 to 20 g per day in humans, with 1 to 2 g contributed by the diet. The predominant dietary sterols are choles- terol, of animal origin, and sitosterol, the major plant sterol. The Western diet provides 300 to 500 mg of cholesterol daily. Bile contributes 800 to 1200 mg daily and intestinal mucosal turnover provides around 300 mg. Approximately 50% of the cholesterol in the intestine is absorbed, but this varies between individuals; the re- mainder is excreted in faeces. In high absorbers, there is more likely to be an increase in plasma cholesterol levels, and better response to a low-​cholesterol diet. The essential fat-​soluble vitamins, vitamins E and A, are derived mainly from plants, or secondary animal sources. Emulsification, digestion, and micelle formation Lipid digestion begins in the mouth and stomach through the action of lingual lipase and gastric enzymes. Gastric peristalsis breaks up dietary fat globules into much smaller emulsion droplets (emulsifi- cation). Fine emulsion droplets enter the duodenum and mix with amphipathic biliary phospholipid, free cholesterol, bile acids, pan- creatic enzymes, and colipase (an amphipathic protein that anchors lipase at the surface of the emulsion droplet). Emulsification greatly increases the surface area where water-​soluble enzymes act to digest water-​insoluble lipids. In the jejunum, TG is digested primarily by pancreatic lipase to yield free fatty acids and glycerol; phospholipid digestion is car- ried out by pancreatic phospholipase A2 and lysophospholipase; and cholesteryl ester. About 10 to 15% of dietary cholesterol is hydrolysed by cholesterol esterase to release free cholesterol. After digestion, monoglycerides, free fatty acids, free cholesterol, and fat-​ soluble vitamins associate with bile salts and phospholipids to form micelles. Micelles are about 200 times smaller than emulsion drop- lets, and are small enough to enter between the microvilli and be absorbed (Fig. 12.6.12). Absorption Free fatty acids are taken up from the intestinal lumen into the entero- cytes and used for the biosynthesis of neutral fats (TG, cholesteryl ester) (Fig. 12.6.12). Fatty acid transport proteins (FATPs) such as FATP4 and FAT/​CD36 facilitate the uptake of fatty acids by the enterocytes. Cholesterol and plant sterol absorption is controlled by Niemann–​ Pick C1-​like 1 (NPC1L1) and ATP-​binding cassette (ABC) proteins ABCG5 and ABCG8, which act as cholesterol uptake transporters and as plant sterol efflux transporters respectively. ABCA1 transfers enterocyte cholesterol to apoA1 for intestinal HDL formation and secretion into lymph, and accounts for 30% of all HDL. Vitamin E absorption requires micelle formation, and scav- enger receptor class B member 1 (SRB1) for absorption. Dietary vitamin A is in the form of carotenoids and fatty acid esters of ret- inol. Carotenoids are cleaved to generate retinol or absorbed intact.

12.6  Lipid disorders 2065 Retinyl esters must be hydrolysed by lipase to release free retinol before it can be taken up by enterocytes from micelles. Chylomicrons Nascent chylomicrons are huge, spherical, TG-​rich lipoproteins. ApoB48 is the principal structural component of chylomicrons. It is very large, hydrophobic, and nonexchangeable. The lipoprotein core contains TGs (85% of chylomicron lipid), cholesteryl esters, and fat-​ soluble vitamins. The surface contains a monolayer of phospholipids (mainly phosphatidylcholine) and free cholesterol. The surface is populated by apoA1, apoA4, and the apoCs. Free fatty acids (>12 carbons), monoglycerides, and vitamin E are transferred from the enterocyte microvillus membrane to the ER for re-​esterification and lipoprotein assembly by fatty acid binding Efflux Micelles MTP CM Golgi ER ApoA1 Nascent HDL CM Blood Lymph NPC1L1 ABCG5/8 ApoB-independent ApoB48-dependent CD36 MAG Retinol SR-B1 C C FA FA FA RE 2 3 4 5 6 1 ACAT FABP FATP ABCA1 DGAT LRAT CRBP A4 HDL B48 R VitE CE Fig. 12.6.12  Lipid absorption. Hydrolysed lipids are solubilized in micelles and presented to the apical membranes of enterocytes. Here, transport proteins facilitate the uptake of various lipid entities: Niemann–​Pick C1-​like 1 (NPC1L1) is involved in cholesterol uptake, CD36 and fatty acid (FA) transport protein (FATP) facilitates FA transport and scavenger receptor class B type I (SR-​BI) is involved in vitamin E (Vit E) uptake. In the cytosol, FA-​binding protein (FABP) and cellular retinol-​binding protein (CRBP) transport FAs and retinol (R) respectively. In the endoplasmic reticulum (ER) membrane, acyl-​CoA:cholesterol acyltransferase (ACAT), diacylglycerol acyltransferase (DGAT), and lecithin:retinol acyltransferase (LRAT) facilitate the esterification of cholesterol, monoacylglycerols (MAG), and retinol respectively. These esterified products are then incorporated into apoB48-​ containing chylomicrons; this process is mediated by microsomal triglyceride (TG) transport protein (MTP). The newly synthesized prechylomicrons are transported in specialized vesicles to the Golgi apparatus for further processing and secretion. Enterocytes also express ATP-​binding cassette (ABC) transporter A1 on the basolateral membrane to promote the efflux of cholesterol. A3, apoA3A4, apoA4; C, free cholesterol; CE, cholesteryl ester; RE, retinyl ester. Source data from Iqbal J, et al. (2009). Intestinal lipid absorption. Am J Physiol Endocrinol Metab, 296, 1183–​94.

section 12  Metabolic disorders 2066 proteins (FABPs). Re-​esterification of free fatty acids and cholesterol is mediated by diacylglycerol acyltransferase (DGAT1) and acetyl-​ CoA acetyltransferase (ACAT2). Retinol is transferred to the ER by cellular retinol-​binding proteins (CRBP). Chylomicrons nucleate in the ER around a single molecule of apoB48. TGs, phospholipids, free cholesterol, and cholesteryl esters along with fat-​soluble vitamins are loaded onto apoB48 during chylo- micron assembly in the ER by a microsomal triglyceride transfer protein (MTTP). Chylomicrons are further lipidated in the secretary pathways, before secretion from the enterocyte basolateral mem- brane into the lymphatic system, and delivery by the thoracic duct to the systemic circulation. In the circulation, chylomicrons undergo extensive modification (Fig. 12.6.13). They transfer phospholipid and cholesteryl ester to HDL, which in turn donates apoC2 and apoE to the nascent chylo- micron, converting it to a mature chylomicron. In the peripheral capillaries of adipose tissue, skeletal muscle and heart chylomicrons associate with LPL, which is anchored to glycosylphosphatidylinositol-​anchored protein (GPIHBP1). Lipase maturation factor 1 (LMF1) resides in the ER, and is involved in the maturation and transport of LPL, hepatic lipase (HL), and pancre- atic lipase through the secretory pathway. ApoC2 is the cofactor for LPL activity. Chylomicron TG is hydrolysed by LPL releasing free fatty acids, which are stored in fat as TG or burnt to create energy in muscle. Some free fatty acids associate with albumin and are transported to other tissues, mainly the liver. The chylomicron progressively de- creases in size as TG is hydrolysed. Once TG is depleted, the chylomicron remnant returns apoC2 to the HDL but retains apoE. The remnant of this metabolism is rapidly cleared from the circulation by the interaction of apoE with hepatic LDLRs. In the normal postprandial state, few or no chylomicrons or chylomicron remnants are present in blood after a prolonged fast, except in those with dyslipidaemia, in whom they may persist. Hepatic lipid transport as VLDL, IDL, and LDL VLDL Lipids from the liver are transported to the periphery on VLDLs (Fig. 12.6.13). VLDLs, like chylomicrons, are TG-​rich lipoproteins. They also carry free cholesterol and cholesteryl ester, with a TG-​to-​ cholesterol ratio of approximately 5:1, and vitamin E. VLDLs, like chylomicrons, nucleate in the ER around a single molecule of apoB, in this case apoB100 rather than apoB48. ApoB100 is twice the size of the apoB48 on the centile system. VLDL TGs are derived predominantly from the esterification of the long-​chain fatty acids in the liver. These come from chylomicron remnants and through de novo synthesis from excess dietary carbo- hydrate. As with chylomicrons, TG is loaded on to apoB100 in the ER through the agency of MTTP. After secretion into the blood, VLDL acquires several molecules of apoE and of the apoC apolipoproteins from HDL. As with chylo- microns, the VLDL TGs are hydrolysed by LPL in peripheral adi- pose tissue, muscle, and heart blood capillaries. IDL and LDL The remnants of peripheral lipolysis, IDLs, disassociate from LPL. After TG hydrolysis, IDLs contain approximately the same amount of cholesterol and TG. About half of IDLs are removed by liver LDL receptors through interaction with its ligand apoE. The remaining IDL is refashioned by HL with removal of further TG and phospho- lipid to form the cholesterol-​rich LDL. In the process of LDL formation, apoE and the apoCs are trans- ferred to other lipoprotein particles. LDL comprises one molecule of apoB100, a shell of phospholipid and free cholesterol, and a core of cholesteryl ester. LDL can vary in size according to its neural lipid content; particles with more TG and less cholesteryl ester are smaller and denser (sdLDL). The size of LDL has implications for its atherogenicity (Table 12.6.1, Fig. 12.6.4). Most LDL is removed ApoB100 Adipose tissue, muscle, heart LPL, ApoA5, C2,C3 Adipose tissue, muscle, heart LPL, ApoA5, C2,3 ApoE LDL VLDL ApoC IDL LDLR Exogenous Endogenous Chylomicron ApoB48 ApoC ApoE Gut Liver Other sites Remnant Fig. 12.6.13  The exogenous and endogenous lipid metabolism pathways. Chylomicrons from the gut transport dietary triglycerides to tissue where they are removed by lipoprotein lipase (LPL). The remnants are taken up by the liver via remnant and LDL receptors and catabolized. VLDL is synthesized in the liver and transports endogenous triglycerides to tissue where they are removed by LPL, resulting in IDL. Some IDL is taken up directly by hepatocytes but for the majority of IDL, further triglyceride is removed by hepatic lipase and thereby IDL is converted to LDL. LDL is removed by the liver via the LDL receptor (LDLR).

12.6  Lipid disorders 2067 from the blood by the liver through the binding to the LDLR by a C-​terminal LDL receptor binding domain in apoB100. Reverse cholesterol transport by HDL While most cells make cholesterol, it cannot be degraded and only the liver and intestine excrete it, either by secretion into bile or into the intestinal lumen by the enterocyte. Biliary cholesterol is either in the form of free cholesterol or bile acids, after the conversion of liver cholesterol to bile acids. HDL mediates the ‘reverse cholesterol trans- port’ of cholesterol derived from peripheral cell plasma membranes back to the liver and gut for recycling or excretion (Fig. 12.6.14). Freshly secreted apoA1 acquires phospholipids and free choles- terol from liver and intestinal cell plasma membranes, where their efflux is mediated by ABCA1. The nascent HDL gains additional free cholesterol from peripheral cells or circulating lipoproteins. The free cholesterol in HDL is esterified by the enzyme LCAT, which is as- sociated with HDL, and the cholesteryl ester transferred to the core of the lipoprotein particle. ApoA1 is a necessary cofactor for LCAT. The acquisition of further cholesteryl ester, phospholipid, and add- itional apoCs and apoE transferred from chylomicrons and VLDL during lipolysis creates the mature HDL particle. HDL particles are extensively modified in the blood. Cholesteryl ester and phospholipid are transferred by cholesteryl ester transfer protein (CETP) and phospholipid transfer protein (PLTP) to HDL from other lipoproteins or between various ‘classes’ of HDL lipo- proteins (Table 12.6.1, Fig. 12.6.14). CETP and PLTP generate a TG-​rich HDL, which is substrate for HL. HL hydrolyses TG and phospholipid to generate small HDL particles. Another lipase, endo- thelial lipase, hydrolyses HDL phospholipid and generates even smaller HDL, which is rapidly catabolized. HDL cholesterol (HDL-​C) is cleared by the liver. Cholesterol from HDL can also be taken up by hepatic SRB1 cell surface receptors that mediate the selective uptake of lipids into cells. Cholesteryl ester from HDL is also transferred to apoB-​containing IDL and LDL lipoproteins in exchange for TGs by CETP. These apoB-​containing particles are then removed by LDLR-​mediated endocytosis. Some apoA1 is catabolized by the kidneys. Dyslipidaemia affecting cholesterol and triglyceride plasma levels The term dyslipidaemia is used to describe disorders of lipoprotein metabolism in which there is elevation of TC and/​or TGs, often ac- companied by reduced levels of HDL-​C. It is very common. Dyslipidaemia is also used here to encompass high levels of apoB and LDL particle number without elevation of LDL chol- esterol (LDL-​C), which indicates the presence of sdLDL; high levels of Lp(a); low levels of apoB-​containing lipoproteins; low HDL-​C; and high levels of HDL-​C. The terms hyperlipidaemia or hyperlipoproteinaemia refer to raised lipids, and do not encompass HDL. The term dyslipidaemia is used preferentially in this chapter. Cholesterol Cholesteryl ester CETP and PLTP, transfer of cholesteryl
ester and phospholipids SR-B1 Bile Excretion Intestine Liver Macrophage Peripheral cell Circulation Tissue CM VLDL Mature HDL ApoA1 ABCG1 ABCA1 ABCG1 ABCA1 Nascent discoid HDL HDL Fig. 12.6.14  Reverse cholesterol transport. Nascent HDL acquires free cholesterol from extrahepatic cells. The free cholesterol in HDL is esterified by lecithin cholesterol acyltransferase (LCAT). ApoA1 is a necessary co-​factor for LCAT. Further cholesteryl esters (CE), phospholipids, and apoC and apoE are transferred during lipolysis from chylomicrons and VLDL to create mature HDL particle. Cholesteryl ester and phospholipid are transferred by cholesteryl ester transfer protein (CETP) and phospholipid transfer protein (PLTP) respectively. CETP and PLTP generate a TG-​rich HDL, which is then hydrolysed by hepatic lipase to generate small HDL particles. HDL-​C is cleared by the liver. Additionally, cholesterol from HDL can be taken up by hepatic SRB1 cell surface receptors.

section 12  Metabolic disorders 2068 The units used here are in mmol/​litre with mg/​dl in brackets. The conversion factor for cholesterol from mmol/​litre to mg/​dl is times 38.7, and for TGs is times 88.6. In clinical practice, the lipid profile of total cholesterol (TC), TG, and HDL levels is routinely measured, and LDL-​C and/​or non-​HDL cholesterol (NHDL-​C) usually calculated. The use of calculated LDL-​C versus NHDL-​C is discussed later in this chapter. The measurement of lipid and lipoprotein levels in clinical prac- tice is important because the presence of raised plasma levels of cholesterol is causally associated with atherosclerotic cardiovascular disease, and intervention to lower cholesterol levels will decrease the risk of atherosclerotic cardiovascular disease events. Raised plasma TG is also causally associated with increased atherosclerotic cardio- vascular disease risk, and lowering TGs decreases atherosclerotic cardiovascular disease events. By contrast, low plasma HDL levels are a marker of increased atherosclerotic cardiovascular disease risk, but this relationship is not causal. Less commonly there can be severe hypertriglyceridaemia, which should be recognized as it can cause potentially fatal, acute pancreatitis, the risk of which can be reduced by treatment of hypertriglyceridaemia. Dyslipidaemia is caused by a combination of genetic (primary) and acquired (secondary) factors (lifestyle, medical or metabolic conditions, and drugs). The genetic architecture of dyslipidaemia is made up of both common and low-​frequency polymorphic variants, which affect lipid and lipoprotein metabolism and other metabolic traits such as obesity, insulin resistance, and hypertension. Rare al- leles make a small population, but large individual contribution to genetic risk. Genetic variation confers around 50% to the total vari- ation in LDL-​C, TG, and HDL-​C levels. Traditionally, familial hyperlipidaemias have been typed ac- cording to the Fredrickson (World Health Organization) classifica- tion (Table 12.6.3), which is based on the pattern of lipoproteins on electrophoresis or ultracentrifugation. This is a descriptive typing of apoB-​containing lipoproteins, and does not include HDL. The Fredrickson classification does not distinguish among the gene defects that are wholly or partially responsible for the dyslipidaemia. Here, reflecting advances in mechanistic cell biology and molecular genetics, a classification-​based mechanism is pre- ferred, and where the genetic basis is known this is used, rather than the Fredrickson classification. The Fredrickson classification is of value in severe hypertriglyceridaemia (see the ‘fridge test’, described in ‘Familial chylomicronaemia (syndrome)). The evidence base The evidence concerning the relationship the between plasma lipid and lipoprotein levels and atherosclerotic cardiovascular disease comes from epidemiology, genetics, and the meta-​analysis of clin- ical trials to treat dyslipidaemia. This data is supported by experi- mental medicine and pathology. Epidemiology demonstrates a very strong relationship between blood cholesterol at all blood cholesterol levels and atherosclerotic cardiovascular disease events; TG levels throughout the normal range and with mild to moderate elevations are similarly associated (Figs. 12.6.15 and 12.6.16). Genome-​wide association studies and Mendelian randomization (Fig. 12.6.17) studies show that this relationship between LDL-​C and TG levels and atherosclerotic cardiovascular disease is causal. Mendelian causes of severe hypercholesterolaemia are strongly linked to premature atherosclerotic cardiovascular disease. Severe hypertriglyceridaemia also possibly increases the risk of atheroscler- otic cardiovascular disease. An important distinction between TGs and cholesterol is that TGs can be broken down by most cells, but cholesterol can be de- graded by none. This property of cholesterol and the small size of LDL, which enables it to enter the vascular intima, account for the direct atherogenicity of LDL. In the intima it is oxidized to form oxidized LDL, which along with native LDL, is taken up by activated macrophages leading to a cascade of events that cause atheroscler- otic cardiovascular disease. The role of TGs in atherogenesis is less clear. TG-​rich lipoproteins and remnants both appear to be involved. TG-​rich chylomicron and VLDL are, however, too large to enter the vascular intima, where they could have a direct pathogenetic role in atherosclerotic car- diovascular disease. Rather, free fatty acids and monoacylglycerol released by TG lipolysis from TG-​rich lipoproteins can produce low-​grade intimal inflammation. Remnants are small enough to pass into the vascular intima, where remnant cholesterol is the likely agent provocateur of atherosclerotic cardiovascular disease. Foam cells accumulate cholesterol not TGs. The meta-​analyses of large clinical trials with statins overwhelm- ingly and conclusively demonstrate that LDL-​C lowering at all Table 12.6.3  Fredrickson/​World Health Organization classification of primary hyperlipidaemias Type Overnight serum High lipoprotein particles Clinical disorders Serum TC Serum TG 1 Creamy top layer Chylomicrons Lipoprotein lipase deficiency, apoC2, apoA5, GPIHBPI, LMF1, GDP1 deficiency N ++ 2a Clear LDL Familial hypercholesterolaemia, polygenic hypercholesterolaemia, nephrotic syndrome, hypothyroidism, familial combined hyperlipidaemia ++ N 2b Clear LDL, VLDL Familial combined hyperlipidaemia ++ + 3 Turbid IDL Dysbetalipoproteinaemia +/​++ +/​++ 4 Turbid VLDL Familial hypertriglyceridaemia, familial combined hyperlipidaemia, sporadic hypertriglyceridemia, diabetes N+ ++ 5 Creamy top,
turbid bottom Chylomicrons, VLDL Diabetes, lipoprotein lipase deficiency, apoC2, apoA5, GPIHBP1, LMF1, GPD1 deficiency + ++ +, increased; ++, greatly increased; N, normal; N+, normal or increased; TC, total cholesterol.

12.6  Lipid disorders 2069 plasma concentrations reduces atherosclerotic cardiovascular dis- ease events. This is the case for LDL-​C levels to as low as 1.3 mmol/​ litre (50 mg/​dl). The remarkable success of statins is without harm, apart from a modest increase in the risk of diabetes. Even this low level of LDL-​C is double the true physiological con- centration seen in animals in the wild and human neonates, so that the level to which LDL-​C can be reduced to prevent atherosclerosis remains to ascertained. No increase in the risk of haemorrhagic stroke has been observed at very low cholesterol levels. While large clinical trials have not formally evaluated the effect of TG reduction on atherosclerotic cardiovascular disease events, sec- ondary analysis of trials of fibrates and nicotinic acid on TG reduc- tion indicate a comparable effect to that achieved with statins. This needs to be validated in appropriately designed clinical trials. Contrary to previously held views, low HDL levels are not as- sociated with increased atherosclerotic cardiovascular disease risk in genome-​wide association studies and Mendelian randomiza- tion studies, and the meta-​analysis of large clinical trials show no benefit from increasing HDL levels. Rather, HDL levels appear to be a biomarker of risk conferred by associated factors such as obesity, insulin resistance, and diabetes. Whether improvement in the functionality of HDL, as, for example, in reverse cholesterol trans- port improves atherosclerotic cardiovascular disease risk is not at present known. Intestinal lipid absorption and dyslipidaemia Intestinal cholesterol absorption varies between 30 and 80% in indi- viduals, and on average is around 50%. These differences are partly genetic and likely to be conferred by common and low-​frequency variants for the main part, but strongly influenced by secondary (acquired) factors. 40 30 20 10 5 4 3 2 1 0 4.0 5.0 6.0 7.0 8.0 Total cholesterol (mmol/L) Shanghai study CVD mortality (MRFIT) Bulk of CVD Total mortality (MRFIT) Age-adjusted 6-year death rate per 1,000 men (n=361,662) Fig. 12.6.15  Population cholesterol levels and cardiovascular risk. MRFIT study: this prospective study of 356 222 American men, aged 35 to 57 years in 1973 to 1975, demonstrated that the relationship between serum cholesterol and risk of cardiovascular disease death was continuous, graded (dose related), and strong over the entire range of the distribution of cholesterol levels. The Shanghai Study: Chen et al. followed 9021 men and women aged 35 to 64 from urban Shanghai, China, for 8–​13 years and found that even for cholesterol levels significantly lower than those in the MRFIT study, there was a significant correlation between cholesterol levels and atherosclerotic cardiovascular disease death, with the risk increasing 4.5 times from the lower values to the higher and no apparent threshold. MRFIT study figure source data from: Stamler J, et al. (1986). Is relationship between serum cholesterol and risk of premature death from coronary heart disease continuous and graded? Findings in 356,222 primary screenees of the Multiple Risk Factor Intervention Trial (MRFIT). JAMA, 256(20), 2823–​8. Shanghai Study figure source data from: Chen Z, et al. (1991). Serum cholesterol concentration and coronary heart disease in population with low cholesterol concentrations. BMJ, 303(6797), 276–​82. Non-fasting triglycerides (mmol/L) Mainly fasting triglycerides (mmol/L) 0 1 2 3 4 5 Hazard ratio (95% CIs) for ASCVH N = 93410 (cardiac events = 7183) Median follow-up 6 years N = 302430 (cardiac events = 12785) Median follow-up 8 years Copenhagen City Heart Study and Copenhagen General Population Study Emerging Risk Factors Collaboration 0 1 1 2 3 1 2 3 4 5 6 7 2 3 4 5 Hazard ratio (95% CIs) for ASCVD 4 Fig. 12.6.16  Population triglyceride levels and cardiovascular disease risk. Observational associations between raised concentrations of triglycerides and cardiovascular disease in the Copenhagen City Heart Study and Copenhagen General Population Study combined (left) and in the Emerging Risk Factors Collaboration (right). Hazard ratios were estimated by Cox proportional hazard regression models, and were adjusted for age, sex, and trial group. From Nordestgaard BG and Varbo A (2014) Triglycerides and cardiovascular disease. Lancet, 384 (9943):626–​35.

section 12  Metabolic disorders 2070 Primary factors affecting intestinal lipid absorption Polymorphic genetic variation at the APOE locus accounts for about a quarter of variability in the absorption of cholesterol. APOE2 car- riers have the lowest absorption and APOE4 carriers the highest absorption. Those with an APOE4 allele also show a greater eleva- tion in plasma cholesterol levels in response to dietary cholesterol. Common variants of the cholesterol 7 α-​hydroxylase gene (CYP7A1) also affect intestinal bile acids and cholesterol absorption. Rare mu- tations of the NPC1L1 gene decrease cholesterol absorption and are associated with a decreased risk of atherosclerotic cardiovascular disease. Sitosterolaemia Sitosterolaemia also known as ‘phytosterolaemia’ is a rare autosomal recessive disease. LDL-​C levels can be normal to severely elevated in the familial hypercholesterolaemia (FH) range (see ‘Familial hyper- cholesterolaemia’). It results from a deleterious mutation in either of two related ABC half-​transporter genes, ABCG5 and ABCG8, which are expressed in enterocytes and liver cells. ABCG5 and ABCG8 proteins heterodimerize to form a transporter for plant sterols such as sitosterol and campesterol, and cholesterol from the liver into bile and from the enterocytes into the gut. Normally approximately 5% of plant sterols are absorbed by the jejunum, and these are disposed of by secretion into bile after cycling through the liver. Consequently, systemic levels of plant sterols remain very low. Patients with sitosterolaemia have high jejunal absorption of plant sterols and cholesterol, but biliary and faecal excretion is decreased causing high blood and tissue con- centrations of plant sterols and cholesterols. In the liver, the high level of sterols suppresses the expression of the LDLR gene, with low LDL-​C clearance and high plasma cholesterol levels. In con- sequence, there is a failure to respond well to statins as the LDLR is not induced. Patients can have prominent tendon xanthomas, and these may be disproportionate to the cholesterol levels, and occurrence of prema- ture atherosclerotic cardiovascular disease. They also have abnormal haematological and liver function test results due to the presence of plant sterols in cell membranes. They can get attacks of haemolysis, and may have splenomegaly. These distinct clinical characteristics suggest the diagnosis: tendon xanthomas, perhaps in disproportion to the cholesterol levels, and high cholesterol in the absence of a family history of atherosclerotic cardiovascular disease. A further clue to the clinical diagnosis is a poor response to statins. The diagnosis is made by the laboratory finding of high blood levels of sitosterol and campesterol, and gene sequencing. The diagnosis is important because a low plant sterol diet, cholesterol absorption inhibitors, and bile acid sequestrants are highly effective treatments, while statins are less so. LDL and sitosterol blood levels require monitoring as elevated plant sterols are themselves atherogenic. Acquired (secondary) factors affecting intestinal lipid absorption Secondary factors such as dietary fibre and plant sterol consump- tion affect cholesterol absorption. The microbiome is increasingly recognized as important. Intestinal bacteria dehydroxylate and deconjugate bile acids, and this affects cholesterol absorption. They also break down nonabsorbable dietary constituents rendering them absorbable and a source of nutrients. Stanols are synthesized as a gut microbial byproduct of cholesterol metabolism. Despite the primary and secondary factors affecting cholesterol absorption, a 1-​g per day difference in dietary intake of cholesterol on average only results in a 5% change in plasma cholesterol. This is the main reason why restriction in dietary cholesterol intake is no longer recommended in the United States of America (see ‘Treatment of dyslipidaemia’ and ‘Lifestyle’). HDL-C allele score Restricted (19 SNPs) 16 (11826,53514) LogTG allele score Restricted (27 SNPs) .. .. .. LDL-C allele score Restricted (19 SNPs) .. .. .. 0.91 (0.42, 1.98) 0.817 1.61 (1.00, 2.59) 0.05 1.92 (1.68, 2.19) 4.6x10–22 0.5 1 2 3 Odds ratio Allele score Studies (cases, total) Odds ratio per 1 unit increase in lipid (95% Cl) P-value Fig. 12.6.17  Mendelian randomization of HDL, TG, and LDL for atherosclerotic cardiovascular disease risk. This meta-​analysis demonstrates the effect of a 1-​unit increase in blood lipid traits on atherosclerotic cardiovascular disease risk. The genetic findings support a causal effect of triglycerides and LDL on atherosclerotic cardiovascular disease risk, but a causal role for HDL-​C is not clear. Estimates were derived incorporating data on the association between the allele scores and blood lipid traits from prospective cohorts (in which most individuals were free from disease when lipid traits were measured) and applying this estimate to all studies with data on the association between the scores and coronary heart disease. Adapted from: Holmes MV, et al. (2015) Mendelian randomization of blood lipids for coronary heart disease. Eur Heart J. 36(9):539–​50.

12.6  Lipid disorders 2071 Interindividual variation in cholesterol absorption and response in plasma cholesterol levels does, however, have therapeutic impli- cations. High absorbers respond well to dietary restriction of choles- terol and in turn to ezetimibe treatment, and low absorbers respond poorly. The measurement of noncholesterol sterols and stanol concen- trations provides a measure of cholesterol absorption status—​high absorbers have high plant sterol plasma levels—​but these tests are not generally available, nor are they fully validated. ApoE4 carriers also have higher levels of cholesterol absorption, and tend to show a poor response to statins, perhaps partly due to the effect of apoE4 on cholesterol absorption. Empirical use of a low-​cholesterol diet in those with hyperchol- esterolaemia is well worthwhile as high absorbers may show a dra- matic effect. Patients with a poor response to ezetimibe are likely to be low absorbers. Chronic kidney disease (CKD) patients respond well to ezetimibe with reduction in atherosclerotic cardiovascular disease events, when used with statins, potentially implicating increased cholesterol absorption in chronic renal disease. Reduced production of chylomicrons and VLDL by
the intestine and liver Primary factors affecting production
of chylomicrons and VLDL Abetalipoproteinaemia The production of apoB48-​containing chylomicrons and apoB100-​ containing VLDL in the gut and liver respectively requires MTTP to load the nascent lipoproteins with TGs, phospholipids, and fat-​ soluble vitamins (as discussed previously). Abetalipoproteinaemia is a rare Mendelian recessive disorder caused by deleterious mutations of the MTTP gene. Blood levels of cholesterol and TG are exceptionally low, and chylomicrons, VLDL, LDLs, and apoB are absent from blood. Parents display normal lipid and apoB-​containing lipoprotein levels. Affected infants may ap- pear normal at birth, but by the first month of life fail to thrive; they develop steatorrhea, abdominal distention, and growth retardation due to fat and fat-​soluble vitamin malabsorption. Neurological and retinal abnormalities occur due to fat-​soluble vitamin deficiency, particularly of vitamins E and A, which are nor- mally carried to the liver by chylomicrons and chylomicron rem- nants. Vitamin E is then transported on VLDL out of the liver to other tissues. Infants develop neurological abnormalities with loss of reflexes and decreased vibration and joint position sense. Later in life, in the third and fourth decade, ataxia and a spastic gait develop. A pigmented ret- inopathy is also characteristic, with reduced night-​time and colour vision and later eventually blindness. This constellation of clinical fea- tures is similar to Friedreich’s ataxia, and can lead to incorrect diag- nosis. In addition, patients have acanthocytosis of red blood cells. The diagnosis is made by intestinal biopsy and special fat staining as with oil red O, which shows fat accumulation in enterocytes, and confirmed by DNA sequencing of the MTTP gene. Diagnosis and management is best achieved in specialized centres. Abetalipoproteinaemia patients are treated with a low-​fat, high-​ calorie diet, and large parenteral fat-​soluble vitamin supplements especially of vitamin E.  Early treatment is essential to prevent neurological complications, which can still develop even with ap- propriate parenteral vitamin supplements. Medium-​chain TGs can be helpful, with the caveat that their long-​term use may be hepato- toxic. Without treatment death is usually by the third decade. In later life, due to failure to export lipid there is chronic fatty liver, often with raised transaminases, and there is increased fibrosis, pro- gressing sometimes to cirrhosis. Hypobetalipoproteinaemia Hypobetalipoproteinaemia is an autosomal codominant disorder caused by mutation of the APOB gene, with low TC, LDL-​C, and apoB levels. Often mutations lead to the formation of truncated apoB, with defective chylomicron and VLDL and the formation of small lipoproteins that are rapidly removed from the blood. The diagnosis is based on the lipid levels and inheritance pattern. DNA sequencing will confirm the diagnosis, but is not usually done as the gene is large. Heterozygotes often have TCs of approximately 2.5 mmol/​litre (100 mg/​dl), TGs of approximately 0.5 mmol/​litre (45 mg/​dl), and LDL-​C levels below 1.25 mmol/​litre (50 mg/​dl). ApoB levels are approximately 20 mmol/​litre (60–​140 mg/​dl, 5th to 95th percentile). Vitamin E levels tend to be mildly reduced or low normal. Fatty liver with raised transaminases is common, though not as marked as in abetalipoproteinaemia. Inflammation with pro- gression to fibrosis and cirrhosis is not frequent. Hypobetalipoproteinaemia patients tend to be long-​lived, pre- sumably due to the low levels of LDL-​C and reduced atherosclerotic cardiovascular disease. Hypobetalipoproteinaemia homozygotes are very rare. They resemble abetalipoproteinaemia, but the parents are typical hypobetalipoproteinaemia heterozygotes, and the fat-​soluble vitamin deficiency and neurological sequelae less problematic, because of the ability to transport some fat. Again, diagnosis can be made by intestinal biopsy showing fat accumulation in the enterocytes. PCSK9 deficiency PCSK9 deficiency also causes very low remnant and LDL plasma levels, and is discussed further later in this chapter. Acquired (secondary) factors affecting production of chylomicrons and VLDL Very low levels of apoB-​containing lipoproteins with LDL-​C below 1.5 mmol/​litre (60 mg/​dl) can be a feature of serious chronic illness such as cirrhosis, cancer cachexia, malnutrition, or malabsorption. In the generally well person, however, it may reflect genetic defi- ciency of apoB-​containing lipoproteins (see ‘Abetalipoproteinaemia’ and ‘Hypobetalipoproteinaemia’). Overproduction of VLDL by the liver and dyslipidaemia Mixed (combined) dyslipidaemia with elevated fasting levels of TGs, increased TC, and low levels of HDL is often seen in clinical practice. Excess production of the VLDL by the liver is a common cause of mixed dyslipidaemia. It is usually caused by a combination of gen- etic (primary) and acquired (secondary) factors. Mixed dyslipidaemia is often associated with the metabolic syn- drome, which includes obesity, insulin resistance, hypertension, diabetes, renal disease, and pre-​eclampsia (Table 12.6.4). Often there is sdLDL. Hypertriglyceridaemia and excess sdLDL with low

section 12  Metabolic disorders 2072 levels of HDL is called ‘atherogenic dyslipidaemia’, because of its as- sociation with atherosclerotic cardiovascular disease (Table 12.6.1, Fig. 12.6.17). SdLDL is produced by vascular remodelling of lipoproteins as a result of metabolic disturbance. The key predisposing factor is hypertriglyceridaemia with large VLDL, which leads to the forma- tion of slowly metabolized LDL particles that are subject to exchange processes with removal of cholesteryl ester from the particle core and replacement with TG. This altered LDL is a substrate for HL, and lipolysis generates sdLDL. SdLDL shows defective clearance by the LDLR, increased vascular proteoglycan binding, and suscepti- bility to oxidation, rendering it more atherogenic. The metabolic factors associated with the formation of sdLDL also probably con- tribute to atherosclerotic cardiovascular disease. Mixed dyslipidaemia is also associated with a carbohydrate-​rich diet, excessive alcohol consumption, and drugs such as oestrogen (Table 12.6.4). Primary causes of VLDL overproduction Common and low-​frequency variants of a number of genes that increase the risk of atherosclerotic cardiovascular disease are asso- ciated with VLDL overproduction. Mendelian causes are listed in Table 12.6.5. Familial combined hyperlipidaemia (FCHL) FCHL features increased levels of apoB-​containing lipoprotein pro- duction from the liver giving rise to elevated levels of fasting VLDL TG, LDL-​C, and often low levels of HDL. TG levels vary between 3 and 7 mmol/​litre (265 and 620 mg/​dl), TC levels between 5 and 10 mmol/​litre (200 and 400 mg/​dl), and HDL-​C below 1 mmol/​ litre (40 mg/​dl) in males and 1.25 mmol/​litre (50 mg/​dl) in females. Patients tend to have sdLDL and plasma apoB levels are increased. Sometimes LDL-​C levels can be normal, but LDL particle number and apoB levels are increased, indicative of sdLDL. This condition has been called hyperapoB. FCHL affects about 1% of people. About a quarter of patients with FCHL develop premature atherosclerotic cardiovascular dis- ease (men <55 years and women <65 years). The lipid phenotype (high TC or TG or both) in an individual can vary from time to time depending on metabolic factors such as diet, exercise, overweight, insulin resistance, and diabetes. FCHL often does not emerge until early or even middle adult life probably due to the accumulation of the metabolic factors, which contribute to the phenotype. It is rarely seen in children. Although the FCHL clusters in families, the lipid phenotype varies between individuals. While inheritance studies have suggested major gene effects, the mode of inheritance is non-​Mendelian, and no single gene defect has been identified to cause this disorder. Rather, it is likely to be polygenic due to the clustering of several genes with significant gen- etic variation. The diagnosis is indicated by family history, and the finding of mixed dyslipidaemia, or isolated raised TC or TGs in family mem- bers, in whom the lipid phenotype varies, and low HDL. ApoB levels are always high. Early treatment should be vigorous because of the high risk of premature atherosclerotic cardiovascular disease. Dietary interven- tion should include decreased intake of sugar and starch, which are turned into TG by the liver and secreted as VLDL (Table 12.6.6), in conjunction with weight reduction and physical exercise. Insulin resistance should be recognized and treated. Diabetes should be aggressively managed. High-​intensity lipid lowering with statins is indicated, often with omega-​3 fatty acids derived from fish. Fibrates are indicated if the TGs are not reduced by statin and omega-​3 fatty acid treat- ment (Tables 12.6.7 and 12.6.8), and the benefits outweigh the risk Table 12.6.4  Secondary causes of dyslipidaemia VLDL elevated LDL elevated LDL reduced Obesity Type 2 diabetes Glycogen storage disease Nephrotic syndrome Hepatitis Alcohol Renal failure Sepsis Cushing’s syndrome Pregnancy Acromegaly Lipodystrophy Drugs: oestrogen, β-​blockers, glucocorticoids, bile acid
  binding resins, retinoic acid Hypothyroidism Nephrotic syndrome Cholestasis Acute intermittent porphyria Anorexia nervosa Hepatoma Drugs: thiazides, ciclosporin, carbamazepine Severe liver disease Malabsorption Malnutrition Gaucher’s disease Chronic infectious disease Hyperthyroidism Drugs: nicotinic acid IDL elevated Chylomicrons elevated HDL elevated HDL reduced Lp(a) elevated Multiple myeloma Monoclonal gammopathy Autoimmune disease Hypothyroidism Autoimmune disease Type 2 diabetes Alcohol Exercise Exposure to chlorinated hydrocarbons Drugs: oestrogen Obesity Type 2 diabetes Smoking Gaucher’s disease Acute glomerular nephritis Chronic kidney disease Nephrotic syndrome Inflammation Menopause Orchidectomy Hypothyroidism Acromegaly Drugs: growth hormone, isotretinoin

12.6  Lipid disorders 2073 of rhabdomyolysis with combined statin and fibrate use (see ‘Drug treatment of hypertriglyceridaemia’). Lipodystrophy In lipodystrophy, there is a marked reduction in adipose tissue, which may affect all or just some adipose depots (partial lipodystrophy). There is dyslipidaemia with raised TGs, TC, and low HDL. VLDL is overproduced and there is decreased lip- olysis of TG-​rich chylomicrons and VLDL. Complete absence of fat is sometimes congenital. It is associated with absence of leptin. A variety of causal gene defects have been described. It is very rare. Table 12.6.5  Genetic defects affecting apoB-​containing lipoprotein metabolism Genetic disorder Gene defect Lipoproteins elevated Clinical findings Hypercholesterolaemia Familial hypercholesterolaemia (AD) LDL receptor (LDLR) LDL Tendon xanthomas, CHD Familial defective apoB100 (AD) ApoB100 (APOB) LDL Tendon xanthomas, CHD Autosomal dominant hypercholesterolaemia, type 3 (AD) PCSK9 (PCSK9) LDL Tendon xanthomas, CHD Autosomal recessive hypercholesterolaemia (AR) ARH (LDLRAP) LDL Tendon xanthomas, CHD Sitosterolaemia (AR) ABCG5 or ABCG8 LDL Tendon xanthomas, CHD Hypertriglyceridaemia Lipoprotein lipase deficiency (AR) LPL (LPL) CM, VLDL Eruptive xanthomas, hepatosplenomegaly, pancreatitis Familial apoC2 deficiency (AR) ApoC2 (APOC2) CM, VLDL Eruptive xanthomas, hepatosplenomegaly, pancreatitis ApoA5 deficiency (AR) ApoA5 (APOA5) CM, VLDL Eruptive xanthomas, hepatosplenomegaly, pancreatitis GPIHBPI deficiency (AR) GPIHBP1 CM Eruptive xanthomas, pancreatitis LMFI deficiency Eruptive xanthomas, pancreatitis Combined hyperlipidaemia Familial hepatic lipase deficiency (AR) Hepatic lipase (LIPC) VLDL remnants, HDL Pancreatitis, CHD Familial dysbetalipoproteinaemia (AR) ApoE (APOE) CM remnants VLDL remnants Palmer and tuberoeruptive xanthomas AD, autosomal dominant; AR, autosomal recessive; CHD, coronary heart disease; CM, chylomicron. Table 12.6.6  Cardioprotective lifestyle measures Diet • Total fat intake 30% or less of total energy intake, with saturated fat 7% or less of energy, and minimal trans fat. Saturated fat should be replaced by mono-​unsaturated and poly-​unsaturated fats (in moderation). Cholesterol intake should be limited to 300 mg daily • Replace fat from animal sources with mono-​unsaturated fat from olive oil or rape seed oil, or spreads made from these oils • Try to eat red meat no more than 3–​4 times weekly • Instead of frying foods—​which adds unnecessary fats and calories—​use cooking methods that add little or no fat, like stir-​frying, grilling, baking, poaching, and steaming • Avoid overheating cooking oils to their smoke-​point as this causes oils to lose their beneficial nutrients and forms trans fats • Avoid prolonged storage of oils, especially in the light, as this causes oils to become rancid through the formation of compounds like butyrate which, though not harmful, taste unpleasant • Choose wholegrain varieties of starchy food • Reduce sugar intake from food and drink, and this will reduce both glucose and fructose • Eat at least five portions of fruit and vegetable daily, but beware starchy vegetables, and very sweet fruit in excess, raw is better than cooked • Eat two portions of fish per week, including one of oily fish • Eat at least four to five portions of unsalted nuts, seeds, and legumes per week Exercise • 3000–​4000 metabolic equivalent of tasks (MET) minutes a week.a This can be a combination of many different activities • For example, to reach the total number, one could do all of these every day: • Climbing stairs for 10 min • Vacuuming for 15 min • Gardening for 20 min • Running for 20 min • Walking or cycling for 25 min • Or do all of these every day: • Biking for an hour • Walking the dog for 30 min at a leisurely stroll • Cooking and washing dishes for 30 min • Or do this every day: • Running at a vigorous pace for an hour • Weight loss, avoidance of excessive alcohol consumption, and smoking cessation are to be strongly encouraged Notes: recent guidelines from the United States of America no longer restrict cholesterol consumption as, on average, this does not greatly impact plasma cholesterol levels. Cholesterol consumption is best limited in those with hypercholesterolaemia, as individuals vary greatly in the amount they absorb and the effect of this on plasma levels of cholesterol. Two large meta-​analyses have suggested that reduction in saturated fat consumption and increased unsaturated fat consumption does not decrease atherosclerotic cardiovascular disease risk. The interpretation of these conclusions is unclear as other studies indicate benefit from reduced saturated fat consumption. a Kyu H, Bachman V, Alexander L, et al. (2016) Physical activity and risks of breast cancer, colon cancer, diabetes, ischemic heart disease, and ischemic stroke events: systematic review and dose-​response meta-​analysis for the Global Burden of Disease Study 2013. BMJ, 354, i3857.

section 12  Metabolic disorders 2074 Primary partial lipodystrophy is a codominant disorder as- sociated with defects of lamin A  and other genes. Acquired partial lipodystrophy is described in membranoproliferative glom- erulonephritis and autoimmune diseases such as systemic lupus erythematosus, often with complement C3 consumption and the presence of C3 nephritic factor. Partial lipodystrophy is associated with decreased fat on the torso and limbs, though there may also be increased fat in other areas such as the face. Insulin resistance is very common, and this may be profound. There is usually hepatic steatosis, and this can be exacerbated by cer- tain drugs that block VLDL secretion such as nicotinic acid. Diabetes may develop. Acute pancreatitis can be precipitated by various drugs (Table 12.6.4). Patients with partial lipodystrophy are at increased risk of athero- sclerotic cardiovascular disease. Treatment can be difficult, especially in complete lipodystrophy. The treatment of the dyslipidaemia is similar to that in FCHL, focusing on the insulin resistance and diabetes. GPD1 deficiency This is a very rare autosomal disorder caused by mutations in the GPD1 (glycerol-​3-​phosphate dehydrogenase) protein, which cata- lyses the reversible redox reaction of dihydroxyacetone phosphate and NADH to glycerol-​3-​phosphate and NAD. Presentation is with massive hepatomegaly, steatosis, and marked, albeit transient, hypertriglyceridemia in infancy. VLDL production is increased. The hypertriglyceridaemia corrects with age. Acquired (secondary) causes of increased VLDL production. Acquired causes of VLDL overproduction often accompany primary causes, particularly as this condition is polygenic in origin. The sec- ondary causes of dyslipidaemia are listed in Table 12.6.4. Diet and alcohol Excess dietary sugar and starch are rapidly turned into fatty acids and then TGs by the liver leading to increased VLDL production. Dietary carbohydrate excess is very common in Western society. Table 12.6.7  Statin characteristics Drug Reduction in TC (%) Reduction in LDL-​C (%) Reduction in TG (%) Increase in HDL-​C (%) Metabolism Half-​life (h) Hydrophilic Pravastatin 16–​25 22–​34 15–​24 2–​12 Sulphation 2–​3 Hydrophilic Fluvastatin 16–​27 22–​36 12–​25 3–​11 CYP2C9 0.5–​3.0 Lipophilic Simvastatin 19-​36 26–​47 12–​34 8–​16 CYP3A4 1–​3 Lipophilic Atorvastatin 25–​45 26–​60 17–​53 5–​13 CYP3A4 13–​30 Lipophilic Rosuvastatin 33–​46 45–​63 10–​35 8–​14 CYP2C9 19 Hydrophilic Pitavastatin 20–​45 20–​48 6–​24 14–​25 CYP2C9 11 Lipophilic Table 12.6.8  Nonstatin lipid-​lowering drugs Major indications Starting dose Maximal dose Mechanism of action Common side effects Cholesterol absorption blockers Ezetimibe Elevated LDL-​C 10 mg daily 10 mg daily ↓Cholesterol absorption, ↑LDL receptors Headache, dyspepsia Bile acid sequestrants Colesevelam Cholestyramine Colestipol Elevated LDL-​C 3.75 g daily 4 g daily 5 g daily 4.375 g daily 32 g daily 40 g daily ↑Bile acid excretion and ↑LDL receptors Constipation, worsening hypertriglyceridaemia MTP inhibitor Lomitapide HoFH 5 mg daily 60 mg daily ↓VLDL production Nausea, steatorrhoea, increased hepatic fat ApoB inhibitor Mipomersen HoFH 200 mg SC weekly 200 mg SC weekly ↓VLDL production Injection site reactions, flu-​like symptoms, increased hepatic fat Fibric acid derivativesa Gemfibrozil Fenofibrate Bezofibrate Elevated TG 300 mg twice daily 67 mg once daily 400 mg once daily 600 mg twice daily 267 mg once daily 400 mg once daily ↑LPL ↓VLDL production Dyspepsia, myalgia, gallstones, elevated transaminases Omega-​3 fatty acids Omega-​3 fatty acid
ethyl esters Elevated TG 2 g daily 4 g daily ↓VLDL synthesis ↑TG catabolism Dyspepsia HoFH, homozygous FH. a With concomitant statin, there is an increased risk of myopathy and rhabdomyolysis; use reduced dose. Risk highest with gemfibrozil.

12.6  Lipid disorders 2075 Too much, frequent alcohol consumption impairs fatty acid β-​ oxidation leading to increased TGs and VLDL production, often in the presence of raised HDL-​C caused by an increased transport rate of apoA1 and apoA2. Elevated TGs and HDL-​C found together sug- gests excessive alcohol consumption. Obesity, insulin resistance, and type 2 diabetes (See Chapters 11.5, 11.6, and 13.9.1.) Dyslipidaemia is a common finding in obesity, insulin resistance with hyperinsulinaemia, and diabetes. The dyslipidaemia features raised TGs and low HDL-​C; often with raised LDL-​C and sdLDL—​atherogenic dyslipidaemia. Overweight and obesity are associated with inflammation in fat tissue leading to insulin resistance, hyperinsulinaemia, and per- turbed lipid metabolism—​the metabolic syndrome. A prominent characteristic is VLDL overproduction from the liver, which is driven by free fatty acids released from fat tissue and transported to the liver principally on albumin. In the liver, free fatty acid is converted to TG and phospholipid and assembled into VLDL for discharge into the blood, and this is favoured by hyperinsulinaemia. Accompanying reduction in LPL function, with impaired lipolysis of chylomicrons and VLDL TG, worsens the ele- vation of TGs. The metabolic syndrome is associated with a serious excess risk of atherosclerotic cardiovascular disease. The treatment is weight loss and good diabetes control. Lipid lowering medication may be required, and should be considered in diabetes. Cushing’s syndrome (See Chapter 13.5.1.) Glucocorticosteroid excess due to Cushing’s syndrome or glucocorticoid treatment causes increased fatty acid and TG biosynthesis, VLDL overproduction, and elevated blood TGs. Cushing’s syndrome sufferers often have raised TGs and low HDL-​C, and sometimes high LDL-​C. Glucocorticoid-​induced in- sulin resistance may contribute to VLDL overproduction. The management is that of Cushing’s syndrome, or modulation of the therapeutic doses of glucocorticoids. Nephrotic syndrome (See Chapters 21.3 and 21.6.) Increased VLDL secretion is common in the nephrotic syndrome. This is considered to be caused by the low oncotic pressure in blood due to hypoalbuminaemia leading to increased biosynthesis of liver protein, but the mechanism for this is not understood. Treatment of the kidney disease is im- portant to decrease the loss of albumin in the urine and correct the hypoalbuminaemia. Unrelieved nephrotic syndrome is likely to need lipid-​lowering drug treatment. Liver disease The liver affects lipoprotein metabolism through several pro- cesses because it is both a major organ of lipoprotein synthesis and of clearance. VLDL production is increased with modest hypertriglyceridaemia in hepatitis caused by viruses, alcohol, and medications. Failure of VLDL synthesis with marked reduction in plasma chol- esterol and TGs occurs in severe hepatitis and hepatic failure. Bile is an important channel for excretion of cholesterol, either after conversion to bile acids or directly as cholesterol. Markedly raised plasma cholesterol levels can occur in cholestatic disease, which impairs biliary cholesterol excretion. Free cholesterol and phospholipids are secreted into the blood as a lamellar particle des- ignated lipoprotein-​X (LP-​X). LP-​X can form xanthomas in skin folds similar to those found in FDBL (xanthomata striata palmaris), as well as planar and eruptive xanthomas. Statins appear to be safe and effective in severe chronic cholestasis such as in primary biliary cirrhosis. Increased liver transaminases and statin treatment are dis- cussed later in this chapter (see ‘Drug treatment of hypercholester- olaemia’ and ‘Statins’). Defective peripheral lipolysis and dyslipidaemia The main goal in severe hypertriglyceridaemia is to prevent acute pancreatitis (see ‘Drug treatment of hypertriglyceridaemia’). Raised plasma TGs are also associated with an increased risk of atheroscler- otic cardiovascular disease. Decreased lipolysis of TG-​rich lipoproteins is commonly associ- ated with dyslipidaemia. LPL is the principal enzyme responsible for the peripheral lipolysis of chylomicron and VLDL TGs. Decreased LPL activity gives rise to fasting hypertriglyceridaemia with low levels of HDL-​C, mostly in the absence of raised LDL-​C or apoB. LPL is synthesized by and secreted from adipocytes, and skeletal and heart muscle cells, where it becomes bound to vascular endothelial cells through GPIHBP1. ApoC2 is a necessary cofactor for activation of LPL. ApoA5 and LMF1 activate and mature LPL respectively. LPL may be reduced for genetic or acquired reasons. Primary causes of defective lipolysis of triglyceride-​rich lipoproteins Fifty per cent of the variance of TG levels has a genetic basis, and is likely to be caused by a combination of common and low-​frequency genetic variants, and rare alleles. Variants are described in multiple genes that cause increased plasma TG levels, and several of these affect lipolysis. Numerous variants are also described in LPL and genes directly associated with LPL activity, and some of these are associated with Mendelian recessive forms of hypertriglyceridaemia (Table 12.6.5). Familial chylomicronaemia (syndrome) Primary deficiency of LPL or of its cofactor apoC2 causes marked hypertriglyceridaemia (type 1 and 5 hyperlipoproteinaemia) (Table 12.6.3). Fasting TG levels are invariably greater than 10 mmol/​litre (900 mg/​dl), and can be 20 to 40 times normal with levels of 30 to 65 mmol/​litre (2500–​5700 mg/​dl). In the nonfasting state, particularly when there are intercurrent factors such as poorly controlled diabetes, plasma TG levels can reach 100 times the normal value with levels of 170 mmol/​litre (15 000 mg/​dl). In type 5 hyperlipidaemia, where increased VLDL production is prominent, fasting cholesterol levels can be five times normal. Primary LPL deficiency is an autosomal recessive disorder with a frequency of around 1 per million people. It is best characterized for mutations of the LPL and APOC2 genes, but mutants of APOA5, GPIHBP1, and LMF1 (see ‘Chylomicronaemia due to other gene de- fects’) can all cause chylomicronaemia with a type 1 or 5 ‘hyperlipid- aemia pattern’. Many causal mutations of LPL and APOC2 have been described. LPL heterozygotes often have mild to moderate eleva- tions of TG. APOC2 defective heterozygotes do not. Type 1 ‘hyper- lipidaemia’ is more likely to be caused by null alleles, and type 5 by less deleterious alleles.

section 12  Metabolic disorders 2076 Nearly all patients have recurrent episodes of severe abdom- inal pain, with or without overt acute pancreatitis, that interfere with normal life and result in frequent hospitalizations. These episodes can result in chronic pancreatitis and symptoms of exo- crine or endocrine pancreatic insufficiency, including diabetes and even fatal events. Other symptoms include arthralgia and neuro- logical symptoms such as loss of feeling in the feet or legs, and memory loss. On physical examination, patients may display small yellowish papules (eruptive xanthomas), which are often grouped on the ex- tensor surfaces of the arms and legs, back, and buttocks. They are painless, but may itch. Visualization of the fundus reveals milky-​ white discoloration of retinal blood vessels—​lipaemia retinalis. Hepatosplenomegaly may be caused by chylomicron uptake by the mononuclear phagocyte system. Not all patients develop acute pan- creatitis or the cutaneous features. There is an increased risk of ath- erosclerotic cardiovascular disease, but premature atherosclerotic cardiovascular disease is not a major manifestation. A simple diagnostic test is to place a fasting plasma sample in the refrigerator at 4°C overnight. Colloquially this is called the ‘fridge test’ or refrigerator test. Chylomicrons are of low density and float to the top, where they form a narrow, creamy band with a clear or mildly cloudy infranatant (type 1), or the infranatant may be more markedly turbid (type 5); this infranatant is VLDL and remnants. The diagnosis is best confirmed by showing absence of lipase ac- tivity for TGs in postheparin plasma, which releases endothelium-​ bound LPL. LPL activity is very low in LPL and apoC2 defective patients. In apoC2-​ or apoA5-​deficient patients, however, it is cor- rected by supplementing apoC2 or apoA5 from normal plasma. DNA sequencing of the LPL, APOC2, APOA5, GPIHBP1, and LMF1, and possibly other hypertriglyceridaemic genes, will help es- tablish the diagnosis. In the short term, the management of very severe hyper­ triglyceridaemia to prevent acute pancreatitis is as described in ‘Drug treatment of hypertriglyceridaemia’. For the long-​term treat- ment of chylomicronaemia, dietary fat restriction to less than 20 to 30 g daily with fat-​soluble vitamin supplementation is neces- sary. Calories can be supplemented with medium-​chain TGs which provide energy, as they are absorbed directly into the portal vein, though there is a possibility of liver damage with long-​term treat- ment. Exercise can reduce TG levels. Secondary factors, commonly diabetes, should be vigorously treated, and those drugs that increase TGs avoided or substituted. Factors that increase VLDL production will overwhelm residual LPL function and greatly worsen hypertriglyceridaemia. Fibrates are the drugs of first choice in the management of severe hypertriglyceridaemia (Table 12.6.8). Omega-​3 fatty acids (fish oils) act by decreasing VLDL secretion so that they are not useful in pure LPL deficiency, where they can increase hypertriglyceridaemia (see ‘Omega-​3 fatty acids’). In type 5 ‘hyperlipoproteinaemia’, they are an important treatment and can be used at doses well in excess of doses normally used in patients with mild forms of hypertriglyceridaemia, particularly for short periods. Gene therapy with LPL has also been trialled using an adeno-​ associated viral vector containing the LPL gain-​of-​function variant (Alipogene Tiparvovec) given by intramuscular injection leading to myocyte expression of LPL. A second-​generation modified anti- sense APOC3 mRNA inhibitor (2ʹ-​O-​(2-​methoxyethyl)-​modified antisense oligonucleotide) has also proved remarkably efficacious as a treatment, confirming the role of apoC3 in the LPL-​mediated metabolism of TG-​rich lipoproteins. Small-​molecule inhibitors or DGAT1 and MTTP are efficacious in reducing TG absorption, but are not licensed for this use. The management of primary hypertriglyceridaemia is particu- larly troublesome in acute hypertriglyceridaemic pancreatitis and during pregnancy, which are discussed in ‘Drug treatment of hypertriglyceridaemia’. Chylomicronaemia due to other gene defects Deleterious mutations of APOA5, GPIHBP1, and LMF1 can all cause chylomicronaemia syndrome with a type 1 or 5 ‘hyperlipidaemia pattern’. Affected patients can be homozygotes or even be heterozy- gous for defects of two of these genes acting together to reduce LPL activity, and produce the type 1 or 5 ‘hyperlipidaemia pattern’, par- ticularly type 5. LMF1 mutations cause combined lipase deficiency. DNA sequencing of these genes will disclose mutations in more than 80% of cases of type 1 and 60% of type 5 ‘hyperlipidaemia’. Others genes, which have yet to be characterized and secondary factors con- tribute to this phenotype. The treatment is as previously described. Familial hypertriglyceridaemia (FHTG) FHTG features raised fasting TGs, in the range 2.3 to 10 mmol/​litre (200 to 900 mg/​dl) without another primary or secondary reason, often low levels of HDL-​C, and familial clustering. LDL-​C levels may be normal or low due to reduced formation from TG-​rich lipo- proteins in the blood. ApoB levels unlike in FCHL are normal, and the risk of atherosclerotic cardiovascular disease is not usually very high. There is generally reduced lipolysis of TG-​rich lipoproteins, often with overproduction of VLDL by the liver. The blood is likely to display a type 4  ‘hyperlipoproteinaemia picture’ without chylomicrons rather than type 1 or 5.  Acquired factors can worsen the hypertriglyceridaemia and give risk to chylomicronaemia and acute pancreatitis (see ‘Secondary causes of impaired lipolysis’). No single gene has been identified, rather the condition appears to be caused by a combination of gene variants leading to polygenic hypertriglyceridaemia. It is important to identify and treat associ- ated acquired causes. Diet and exercise are valuable. Lipid-​lowering medication, with statins (if plasma TGs are <5 mmol/​litre (450 mg/​ dl)), fibrates, and fish oils may be required to treat higher TG levels. The risks of the combined use of statins and fibrates are shown in Table 12.6.8. Acquired (secondary) causes of impaired lipolysis of triglyceride-​rich lipoproteins The secondary causes of dyslipidaemia are listed in Table 12.6.4. Obesity, insulin resistance, and type 2 diabetes (See Chapters  11.5, 11.6, and 13.9.1.) Obesity, insulin resistance, and type 2 diabetes are associated with increased VLDL secretion, but can also decrease LPL activity. There may be decreased expres- sion of the LPL gene in peripheral tissues, and increased expres- sion of the gene for the LPL inhibitor apoC3 in the liver leading to hypertriglyceridaemia. Clinical management is by weight loss and good diabetes control. Lipid-​lowering medication may be required.

12.6  Lipid disorders 2077 Alcohol As well as increasing VLDL secretion, excessive alcohol significantly decreases LPL activity. Defective hepatic clearance of remnants and LDL as a cause of dyslipidaemia Defective clearance of LDL and chylomicron and VLDL lipoprotein remnants by the hepatic LDLR is a frequent cause of dyslipidaemia. It is caused by genetic and acquired factors. Variation in a number of genes interferes with lipoprotein clearance by the LDLR and these may give rise to polygenic disorders with elevated cholesterol. Mendelian disorders, which give rise to defective hepatic clear- ance of apoB-​containing lipoproteins, can cause marked elevation of LDL-​C with premature atherosclerotic cardiovascular disease (Table 12.6.5). The clinical features of polygenic and Mendelian hypercholesterolaemia overlap. A diet high in saturated and trans fats decreases LDLR activity and increases plasma cholesterol levels, which can worsen genetic causes of hypercholesterolaemia. Hypothyroidism, CKD, and oes- trogen deficiency after the menopause are associated with reduced LDLR activity. Primary (genetic) causes of impaired liver uptake of lipoproteins Multiple common gene variants have been identified that affect clearance of LDL and remnant lipoproteins by the LDLR. Indeed, more than 50% of the variation in LDL-​C levels is determined by genetic factors. In people with predominantly hypercholesterolaemia (type 2a and b ‘hyperlipidaemia’; Table 12.6.3) who do not show Mendelian inheritance, and do not have the features of one of the very rare primary hypercholesterolaemia syndromes de- scribed in the following subsections, in whom a DNA diagnosis is not made, polygenic hypercholesterolaemia is likely. Plasma cholesterol levels often overlap with those found in FH, but the inheritance is not dominant, perhaps affecting around 10% of first-​degree relatives. Inheritance of the number of variants to- gether with elevated cholesterol coupled with diet is generally the cause of this condition. In the evaluation of hypercholesterolaemia, however, single-​gene Mendelian disorders are relatively common and should be con- sidered in the differential diagnosis (Table 12.6.5). Common genetic variants also exacerbate FH. Familial hypercholesterolaemia FH is an autosomal codominant disorder caused by deleterious mu- tation of the LDLR gene. Patients have asymptomatic high TC, due to elevation of remnants and LDL-​C, without raised TGs, usually in the range of 7.5 to 10 mmol/​litre (290–​390 mg/​dl). Premature ath- erosclerotic cardiovascular disease is very common. The frequency of heterozygous FH is as high as 1 in 200 of the population according to recent estimates. In certain populations, the frequency can be even higher due to a founder effect. These include Afrikaners in South Africa, French Canadians, Christian Lebanese, and the Finns. It is important to recognize and treat FH early so as to prevent ath- erosclerotic cardiovascular disease in middle life. Secondary causes of marked hypercholesterolaemia, including hypothyroidism, neph- rotic syndrome, and obstructive hepatic disease, must be excluded in the differential diagnosis. Numerous LDLR gene mutations have identified by DNA sequen- cing (Fig. 12.6.18). Some are recurrent due to small founder effects Class I: LDLR is not synthesized at all Class II: LDLR is not properly transported from the endoplasmic reticulum (receptor negative) to the Golgi apparatus for expression on the cell surface. Class III: LDLR does not properly bind LDL on the cell surface because of a defect in either apolipoprotein B100 (R3500Q) or in LDL-R. Class IV: LDLR bound to LDL does not properly cluster in clathrin-coated pits for receptor-mediated endocytosis. Class V: LDLR is not recycled back to the cell surface. Extent of LDLR deficiency will depend on deleteriousness of amino acid change. 1 Signal sequence 21 a.a. Ligand binding 292 a.a. EGF precursor homology 400 a.a. O-linked sugars 58 a.a. Membrane- spanning 22 a.a. Cytoplasmic 50 a.a. mRNA Binding Transport Recycling Transport PCSK9 binding No effect Secreted Internalization ARH binding Exon number 15 18 mRNA = 5.3 kb 3’ 5’ 45 kb 17 16 14 13 12 11 10 9 8 7 6 5 4 3 2 Fig. 12.6.18  LDL receptor gene (LDLR) structure and familial hypercholesterolaemia mutations.

section 12  Metabolic disorders 2078 or possibly recurrent mutation at susceptible sites such as CpG dinucleotides. Mutations causing complete loss of LDLR activity (receptor nega- tive) compared to those with low activity (receptor defective) lead to higher levels of cholesterol. LDLR defects decrease hepatic clearance of LDL from the blood; removal of IDL is also decreased with in- creased production of LDL from IDL. Individuals with two mutated receptor alleles (FH homozygotes) with the same or different mu- tations (compound heterozygotes) have very much higher levels of LDL-​C than those with one mutant allele (FH heterozygotes). Hypercholesterolaemia in FH patients is present from birth or be- fore. The early lesions of atherosclerosis, fatty streaks, can be seen in the fetus. Hypercholesterolaemia can be detected in the neonate if family history indicates the need for cholesterol measurement. Detection of FH is either due to finding hypercholesterolaemia on health assessment screening, suspicion due to adverse family history, the appearance of tendon xanthomas (Achilles tendon, dorsum of the hand or feet), cutaneous xanthelasmas, often with corneal arcus, or development of premature atherosclerotic cardiovascular disease. Clinical stigmata are present in many but not all patients with het- erozygous FH, and this may depend on the level of cholesterol. Dominant inheritance causes an affected parent to transmit the disease to half of all children, so that usually there is a strong family history of premature atherosclerotic cardiovascular disease from the one half of the family harbouring the mutant LDLR gene. Polygenic factors from both sides of the family may confound this pattern. Untreated FH heterozygotes have a very high risk of premature atherosclerotic cardiovascular disease. Males have an approximately 50% likelihood of myocardial infarction before the age of 60 and females have an approximately 30% chance. The manifestations of atherosclerotic cardiovascular disease vary greatly in their age of ap- pearance. This may depend on the level of cholesterol determined by whether the LDLR mutation causes deficient (10% of mutations) or defective LDLR protein function (Fig. 12.6.18), and the presence of other atherosclerotic cardiovascular disease risk factors such as hypertension, smoking, obesity, diabetes, and elevation of Lp(a). DNA sequencing is the only definitive diagnostic test, and with advances in this technology, this becomes increasingly available in routine practice through specialized medical centres. Four genes are routinely analysed: LDLR, PCSK9, and LDLRAP by exon sequen- cing, and APOB by typing the glutamine for arginine at amino acid 3500 mutation (see ‘Familial defective apoB100 (FDB)’). A DNA diagnosis will be found in about 70% of patients with ‘definite’ FH (see ‘Simon Broome criteria’, Table 12.6.9). In those with possible FH on Simon Broome criteria, a DNA diagnosis will be found in about 30% because many of these patients have polygenic hypercholester- olaemia. DNA diagnosis facilitates family cascade screening. No other laboratory test is available for diagnosis, though in re- search laboratories LDLR binding studies for LDL in cultured cells has been valuable in characterizing the function of different dele- terious mutations. Clinical criteria have therefore been established for diagnosis. In the United Kingdom, these are the Simon Broome criteria (Table 12.6.9), and in other parts of Europe such as the Netherlands, similar criteria are used. FH heterozygotes should be very vigorously treated to reduce LDL-​C levels, and other risk factors minimized, preferably starting in childhood. In children, a low-​cholesterol diet with low saturated and trans fat is valuable. Bile acid sequestrants can be used before puberty, with other lipid-​lowering drugs being introduced after puberty. The rationale for this delay is the importance of cholesterol for growth and sex hormone biosynthesis, but this is not evidence based. Increasingly, some centres are commencing statin therapy in the first decade of life, depending on the child’s LDL levels and the family history of premature atherosclerotic cardiovascular disease. The powerful, long-​acting statins are the most efficacious drugs and should be used at high intensity in adults (Tables 12.6.7 and 12.6.10). It is reasonable to supplement statins with cholesterol absorption inhibitors or bile acid sequestrants, and often these drugs are used in conjunction to achieve greater than 50% LDL-​ C lowering. Other newer drugs, known as PCSK9 inhibitors, have recently become available for treatment of conditions such as FH, but their use is currently highly restricted (see ‘Drug treatment of hypercholesterolaemia’). In those planning pregnancy, statins should be stopped, as animal data suggests that statins are teratogenic although there is no human data to support this. When stopping statins prior to conception, a 3-​month washout period is often recommended, but this is not evi- dence based, and shorter periods are probably acceptable. Targets for LDL-​C/​NHDL-​C lowering should be to greater than 50% of the starting highest level. Greater reduction to the levels ad- vised for secondary prevention is reasonable, LDL-​C to less than 2 mmol/​litre (80 mg/​dl), to minimize the risk of atherosclerotic car- diovascular disease, and this is mandatory in established disease. The role of imaging by carotid ultrasonography of CT coronary cal- cium scoring in detecting premature atherosclerotic cardiovascular disease is discussed in ‘Diagnosis of dyslipidaemia’. Some FH heterozygotes will not achieve greater than 50% LDL-​C lowering with available medication. These patients are candidates for use of PCSK9 inhibitors (see ‘Drug treatment of Table 12.6.9  Simon Broome criteria for the diagnosis of familial hypercholesterolaemia Definite familial hypercholesterolaemia • Total cholesterol >6.7 mmol/​litre or LDL cholesterol >4.0 mmol/​litre in a child <16 years • Total cholesterol >7.5 mmol/​litre or LDL cholesterol >4.9 mmol/​litre in an adult. (Levels either pretreatment or highest on treatment) • Plus tendon xanthomas in patient, or in first-​degree relative (parent, sibling, child), or in second-​degree relative (grandparent, uncle, aunt) • Or DNA-​based evidence of an LDL receptor mutation or familial defective apoB100 Possible familial hypercholesterolaemia • Total cholesterol >6.7 mmol/​litre or LDL cholesterol >4.0 mmol/​litre in a child <16 years • Total cholesterol >7.5 mmol/​litre or LDL cholesterol >4.9 mmol/​litre in an adult • Plus a family history of myocardial infarction: below age of 50 in second-​degree relative or below age 60 in first-​degree relative • Or a family history of raised total cholesterol >7.5 mmol/​litre in adult first-​ or second-​degree relative or >6.7 mmol/​litre in child or sibling <16 years

12.6  Lipid disorders 2079 hypercholesterolaemia’ and ‘PCSK9 inhibitors’). FH patients whose LDL-​C remains markedly elevated, greater than 5 mmol/​litre (200 mg/​dl) with cardiovascular disease or 7.5 mmol/​litre (300 mg/​dl) without cardiovascular disease, despite maximum tolerated drug treatment, especially those with high Lp(a) levels, can be probably be improved with LDL apheresis (see ‘Drug treatment of hyperchol- esterolaemia’ and ‘LDL apheresis’). Homozygous FH is caused by mutations of the LDLR gene passed on from each parent. The frequency is around 1 in a mil- lion people. Those with complete absence of LDLR activity (re- ceptor negative) compared to those with low, but present activity (receptor defective) have higher LDL-​C levels. LDL-​C levels in homozygous FH patients range from about 12 (480) to in excess of 25 mmol/​litre (1000 mg/​dl). Those with higher LDL-​C levels develop tendon xanthomas on the hands, wrists, elbows, knees, Achilles tendons, or buttocks. Homozygotes develop atherosclerotic cardiovascular disease in childhood or adolescence. Lesions are common in the aortic root and aortic valve with subvalvular stenosis, often reaching into the coronary ostia, which become narrowed. The carotids and periph- eral blood vessels become narrowed and can be problematic. Sudden cardiac death is relatively common. Symptoms reflect decreased car- diac output and cardiac ischaemia. Without treatment, death is in the second to third decade, reflecting disease severity. Apheresis is the mainstay for treating homozygous FH. With apheresis and lipid-​ lowering drugs, this grim prognosis is much improved. Homozygous FH is suggested by the very high TC (>12 mmol/​ L, 480 mg/​dl) and LDL-​C levels (>10 mmol/​L, 400 mg/​dl) in the absence of an acquired reason. Skin xanthomas are virtually pathog- nomonic. Atherosclerotic cardiovascular disease in a young patient and TC greater than 7.5 mmol/​litre (300 mg/​dl) in both parents are both highly suggestive of homozygous FH. DNA sequencing will generally identify LDLR mutations. Receptor-​negative homozygotes are most commonly due to consanguinity, but are fortunately rare. Homozygous FH patients should be treated early and most vigor- ously to retard the relentless onset and progression of atheroscler- osis. While maximal drug treatment may be adequate to treat some LDLR patients most will require apheresis as well. High-​intensity statin treatment, ezetimibe, and sequestrants are reasonably given together. Receptor-​negative patients do not respond to these drugs. Liver transplantation has been used to treat homozygous FH, because the liver is the main site of LDL clearance, but its use is limited by problems associated with immunosuppression and graft rejection. Three new drugs have potential value. PCSK9 inhibitors which increase LDLR activity apparently have value in receptor-​defective patients and are now licensed for use. A MTTP inhibitor, which prevents chylomicron and VLDL synthesis and subsequent conver- sion into remnants and LDL is effective in lowering LDL. Antisense oligonucleotide treatment to block apoB synthesis is used in the United States of America, but not in Europe. These drugs are dis- cussed in later subsections. Familial defective apoB100 (FDB) FDB is an autosomal dominant hypercholesterolaemia, which is clinically like heterozygous FH. Mutations of the APOB gene cause FDB, and these usually affect the region of apoB100 that contains the LDLR-​binding domain, which is lacking in apoB48. It is mostly usually caused by a predominant mutation of glutamine for arginine at amino acid 3500, which reduces the binding and clearance of LDL by the LDLR. It is seen in patients of central European origin and in derived populations in the United States of America, where the frequency is 1 in 500 to 1 in 1000, and this reflects a founder effect mutation. It is common in the Pennsylvania Amish but is found worldwide. Recurrent mutation occurs as Arg3500 is encoded by a CpG di- nucleotide, which is a hotspot for mutation. Mutations elsewhere in the gene have also been described. FDB is less common than FH, likely because the target for muta- tion is mainly restricted to the LDLR binding domain of apoB100. Homozygotes are extremely rare. Individuals with both LDLR and FDB mutations have been described, and they behave as homozygotes. In FDB there is elevated plasma LDL-​C, but not of remnants be- cause clearance through the interaction of apoE with the LDLR is normal. Thus TC levels tend to be lower than in FH. TGs are normal. Tendon xanthomas are less prominent than in FH. Premature ath- erosclerotic cardiovascular disease is common. FDB cannot be distinguished clinically from FH. Homozygotes tend to be less severely afflicted than FH homozygotes due to the lower cholesterol levels, because remnants are cleared normally through the interaction of apoE with the LDLR. The definitive diag- nosis is by DNA genotyping, which is usually focused on the amino acid 3500 mutation, or by DNA sequencing. The treatment is similar Table 12.6.10  High-​, moderate-​, and low-​intensity statin therapy High intensity Moderate intensity Low intensity Daily dose lowers LDL-​C on average by 50% Daily dose lowers LDL-​C on average by 30% to <50% Daily dose lowers LDL-​C on average by <30% Atorvastatin 40–​80 mg Rosuvastatin 20 (40) mga Atorvastatin 10–​20 mg Rosuvastatin 10 mg Simvastatin 20 mgb Pravastatin 40 mg Fluvastatin XL 80 mg Fluvastatin 20 mg twice daily Pitavastatin 2 mg Simvastatin 10 mg Pravastatin 10–​20 mg Fluvastatin 20–​40 mg Pitavastatin 1 mg Mechanism of action is by lowering cholesterol synthesis and increasing liver LDLR activity. Common side effects are myalgia, myopathy, elevated transaminase, and dyspepsia. Specific statins and doses are noted in bold that were evaluated in randomized clinical trials, which showed a reduction in major cardiovascular events. Statins and doses not tested in clinical trials are in italics. Individual responses to statin therapy vary and there might be a biological basis for a poor response. a Starting dose 5 mg if hypothyroid, >65 years, Asian. The 40-​mg dose is contraindicated in Asians. b Simvastatin 80 mg is not recommended due to the increased risk of myopathy, including rhabdomyolysis.

section 12  Metabolic disorders 2080 to FH for heterozygotes. The very rare homozygotes may require apheresis as well as drugs. Autosomal dominant hypercholesterolaemia due
to gain-​of-​function mutations in PCSK9 PCSK9 gene gain-​of-​function mutations are a very rare autosomal dominant form of FH. The function of PCSK9 was described earlier in this chapter. These mutations are much more frequent in individ- uals of African descent. Loss-​of-​function mutations in PCSK9 can cause low levels of LDL-​C (see ‘Enhanced hepatic clearance of rem- nants and LDL as a cause of low cholesterol levels’). Gain-​of-​function missense mutations that increase the activity of PCSK9 cause hypercholesterolaemia because the number of LDLRs is reduced. Clinically, patients are similar to those with FH, but they are highly responsive to PCSK9 inhibitors (see ‘Drug treatment in hypercholesterolaemia’ and ‘PCSK9 inhibitors’). Autosomal recessive hypercholesterolaemia (ARH) ARH is a very rare Mendelian recessive disorder. Clinically, pa- tients have plasma LDL-​C levels between those of heterozygous and homozygous FH, and the TC levels are more variable. ARH results from mutation of the gene encoding the LDLR-​ adapter protein, (LDLRAP), which is needed for LDLR-​mediated endocytosis particularly in the liver. LDLRAP binds to the in- ternal domain of the LDLR, linking it to the endocytic machinery (Fig. 12.6.11). With ARH, LDL binds the LDLR, but is not endocytosed, and accumulates at the liver cell surface. Individuals are mostly of Middle Eastern, Turkish, or Sardinian origin, but cases of ARH occur worldwide. Tendon xanthomas are found. Atherosclerotic cardiovascular disease develops by the third decade. ARH is more responsive to lipid-​lowering therapy than homozygous FH and may not require apheresis. Cholesteryl ester storage disease Cholesteryl ester storage disease is a rare Mendelian recessive dis- order. It features raised TC and LDL-​C, often with increased TGs and low HDL-​C. Its worst manifestation, Wolman’s disease, is fatal in infancy. Cholesteryl ester storage disease and Wolman’s disease are caused by deleterious mutation of the gene encoding lysosomal acid lipase (LAL). Patients with no enzyme activity are likely to have the childhood severe presentation and those with partial deficiency the adult presentation. LAL hydrolyses TGs and cholesteryl esters from remnants and LDL taken up by the liver by the LDLR. This leads to the accumulation of large amounts of neutral lipid in liver cells and hepatosplenomegaly, with steatosis, fibrosis, and ultimately cirrhosis. The high LDL may be the result of both increased produc- tion of apoB-​containing lipoproteins and decreased clearance. The diagnosis is suggested in patients with elevated LDL-​C, low HDL-​C, and fatty liver without the metabolic syndrome. Diagnosis is made by measuring LAL activity on a dried blood spot, and veri- fied by DNA sequencing. It is important to establish the diagnosis and evaluate the liver as cirrhosis may occur. Heterozygotes may have mild to moderately disturbed lipids. In 2015, the Food and Drug Administration in the United States of America and European Medicines Agency recommended granting a marketing authorization for Kanuma (sebelipase alfa), recom- binant acid lipase, for the treatment of LAL deficiency. The recom- mendation was based on four studies which provided evidence on the safety and efficacy in infants (<6 months of age), children, and adults. A total of 106 patients (including 14 infants) with LAL defi- ciency received treatment with sebelipase alfa. Significant improve- ments were observed for a number of disease parameters, including improvement in survival of infants with LAL deficiency for which no treatment was available up until now. The long-​term efficacy of this treatment is continuing through an ongoing study in infants with LAL deficiency. Sitosterolaemia This rare autosomal recessive cause of increased absorption of chol- esterol and suppressed LDLR expression has already been described in the Sitosterolaemia’ subsection in ‘Primary factors affecting intes- tinal lipid absorption’. Familial dysbetalipoproteinaemia (FDBL) FDBL is also called type 3 hyperlipoproteinaemia (Table 12.6.3). It is a Mendelian recessive disorder, which becomes manifest when an acquired cause of dyslipidaemia also occurs. It features combined dyslipidaemia with similarly raised levels of both cholesterol and TGs due to accumulation of remnant particles (chylomicron and VLDL remnants). Remnant clearance by the LDLR is mediated by apoE on these lipoproteins. ApoE has three common alleles. ApoE3 is most fre- quent; apoE2 and apoE4 each differ from apoE3 by one amino acid. Homozygosity for the apoE2 allele (frequency of about 1/​200) underlies FDLB. Overall the APOE gene locus confers approximately 1.5 to polygenic dyslipidaemia and to risk of atherosclerotic cardio- vascular disease. The apoE4 allele causes a modestly higher LDL-​C level and increases cholesterol absorption (see ‘Primary factors af- fecting intestinal lipid absorption’), with increased atherosclerotic cardiovascular disease risk, but does not cause FDBL. The presence of one or two APOE4 gene copies does, however, increase the risk of Alzheimer’s disease. ApoE2 has much reduced binding affinity to the LDLR, so that it slows the rate of remnant clearance; homozygosity for apoE2 is present in most patients with FDBL; heterozygotes are at moderate increased risk of dyslipidaemia. Very rare dominant mu- tants of apoE can cause FDBL. The acquired (Table 12.6.4) precipitating factors for FDBL in- clude high fat and carbohydrate diets, obesity, insulin resistance and diabetes, hypothyroidism, kidney disease, HIV, alcohol, and some drugs. In women, it is rare prior to the menopause, after which oes- trogen deficiency plays a role. FDBL patients present as adults with mixed dyslipidaemia, with similarly raised TC and TGs carried in IDL, and normal HDL-​C. Typically, the TC is 8 to 12 mmol/​litre (320–​480 mg/​dl) and TGs 5 to 20 mmol/​litre (420–​1800 mg/​dl). LDL-​C is low due to reduced formation from VLDL and normal clearance by the LDLR. There may be premature atherosclerotic cardiovascular disease, affecting the peripheral blood vessels as well as the coronaries. There are dis- tinctive, pathognomonic xanthomas; tuberoeruptive xanthomas form small groups of small papules on the elbows, knees, or but- tocks, but can expand to be thumb nail-​sized; palmar xanthomas (called xanthoma striae palmaris) are orangey, yellow discolorations in the palmar and wrist creases. The diagnosis is best made by demonstrating apoE2 homozygosity in conjunction with high levels of remnant lipoproteins. Methods of phenotyping apoE include ultracentrifugation (β-​quantification),

12.6  Lipid disorders 2081 lipoprotein electrophoresis (broad β-​band), and MRI, but these are not routinely available. Polymerase chain reaction can be used to type the common alleles, but will miss rare disease-​causing variants. As a rule, cases of FDBL have a TC/​apoB ratio greater than 6.0 and a TG/​apoB ratio of less than 10.0, which are highly predictive, whereas in type 4 hyperlipidaemia, the TC/​apoB ratio is below 5.0 and in type 5 hyperlipidaemia, the TG/​apoB ratio is much greater than 10.0. FDBL needs to be vigorously treated because of the high risk of premature atherosclerotic cardiovascular disease. The acquired metabolic factors that have precipitated FDBL need to be managed. FDBL patients improve with weight reduction, atheroprotective life- style (Table 12.6.6), and reduction of alcohol consumption. Fibrates are the drug of first choice as they have a spectacular ef- fect in lowering TGs and cholesterol in FDBL through VLDL and IDL reduction. Often both LDL-​C and HDL-​C increase. A statin may also be required, with the caveat concerning the safety of the combined use of these drugs. Fish oils can reasonably be used if TGs remain raised. Hepatic lipase deficiency HL hydrolyses TGs and phospholipids in remnants and HDL, which favours liver uptake by apoE, and conversion of remnants into LDL. HL deficiency is a very rare autosomal recessive disease caused by mutations of the HL gene, a relative of LPL. Deficiency features mixed dyslipidaemia due to remnant accumulation and raised HDL-​C, and in this respect resembles alcohol overconsumption. To make the diagnosis, HL activity is measured in postheparin plasma or the DNA of the gene sequenced. Statin therapy is appropriate to reduce potential atherosclerotic cardiovascular disease risk from remnant accumulation. Acquired (secondary) causes of impaired liver uptake of lipoproteins The secondary causes of dyslipidaemia are listed in Table 12.6.4. Diet Diet plays a major role in determining LDL-​C levels. Western na- tions such as the United States of America and European nations have overall higher cholesterol levels than nations such as China and Japan. Even subjects with FH in China have much lower chol- esterol levels than their Western counterparts. Animals in the wild and human neonates have very low cholesterol levels (LDL-​Cs of 0.65 mmol/​litre (25 mg/​dl)), consistent with the view that the human diet plays a major role in determining plasma cholesterol levels. Saturated fat raises LDL-​C more than most other dietary compo- nents. Saturated fat is found in red meats, dairy products, chocolate, baked goods, deep-​fried food, and processed food—​all common in the Western diet. Among saturated fatty acids, lauric, myristic, and palmitic acids are considered to be more hypercholesterolaemic than stearic acids. Lauric and myristic acids are found in coconut. Trans fatty acids (trans fats) also raise LDL-​C and lower HDL-​ C. Trans fats are made when hydrogen is added to vegetable oil to harden it. They are used in processed food to prolong its shelf life. Trans fats are also generated by high-​temperature frying. Saturated and trans fats increase plasma LDL-​C by increasing the formation of LDL in the plasma compartment by decreasing LDL turnover and by decreasing the activity of the LDLR through the activity of SREBP. Because dietary cholesterol intake is not found to correlate well with serum cholesterol levels, dietary restriction is not now recom- mended in the United States of America, but reduction of saturated fat intake will perforce decrease cholesterol consumption. Recent meta-​analyses have not found an association between saturated fat consumption and atherosclerotic cardiovascular disease, while other meta-​analyses have suggested increased risk (Table 12.6.6). Saturated fat should be replaced by unsaturated fat to lower LDL-​ C levels (Table 12.6.6). Low-​fat diets which replace saturated fat with carbohydrates lower LDL-​C, but also lower HDL levels. Hypothyroidism Hypercholesterolaemia is common and often severe in hypothy- roidism. It is due to reduction of liver LDLR levels and reduced LDL and remnant clearance. The expression of the LDLR gene is decreased in hypothyroidism. Mild hypertriglyceridaemia may coexist. Screening of all patients with high LDL-​C for hypothy- roidism is mandatory as hypothyroidism is easy to overlook. Hormone replacement corrects the hypercholesterolaemia, un- less there is another underlying dyslipidaemia. Statin treatment in hypothyroidism can be dangerous because of the risk of severe muscle toxicity. Oestrogen and progesterone Endogenous oestrogens are important regulators of lipid metab- olism and inhibit atherosclerotic cardiovascular disease develop- ment in premenopausal women. Oestrogen reduces LDL-​C levels and increases HDL levels by increasing LDLR activity and decreasing LDL production and apoA1 and ABCA1 protein expression. It also potently reduces the oxidation of LDL. Administration of oestrogen to postmenopausal women in clinical trials using ‘conjugated’ (horse) oestrogens and synthetic progestins (medroxyprogesterone acetate) as hormone therapy have shown in- creased atherosclerotic cardiovascular disease risk despite beneficial effects on the lipid profile. More recent trials using natural oestrogen or the selective oestrogen receptor modulators such as lasofoxifene suggest therapeutic potential for the prevention of atherosclerotic cardiovascular disease, warranting further clinical trials. But this is mitigated by increase breast and ovarian cancer risk. Oral contraceptives and oral oestrogens (as hormone re- placement therapy) are contraindicated in patients with severe hypertriglyceridaemia (type 1, 3, 4, and 5 hyperlipidaemias) as they are reported to precipitate acute pancreatitis. Oestrogen decreases activity of HL, the actions of which include hydrolysis of VLDL, pos- sibly decreased LPL activity, and increased synthesis of TGs in the liver and secretion of VLDL. Oestrogen patches do not affect TG levels. Progesterone-​only pills reduce HDL. Enhanced hepatic clearance of remnants and LDL as a cause of low cholesterol levels PCSK9 deficiency Inherited deleterious alleles of PCSK9 cause increased LDLR ac- tivity and decreased plasma LDL-​C levels. Deleterious mutations of PCSK9 are best described in people of African origin, but are also described in Europeans. Heterozygotes have a 30 to 40% decrease in plasma levels of LDL-​C with much reduced occurrence of ath- erosclerotic cardiovascular disease most likely as a consequence of the low LDL-​C. Homozygotes for such mutations have LDL-​C levels

section 12  Metabolic disorders 2082 below 0.5 mmol/​litre and are well, indicating that lifelong low chol- esterol levels are not deleterious to human health. HDL cholesterol and dyslipidaemia In clinical practice, TC, TGs, HDL-​C, LDL-​C, and/​or NHDL-​C plasma levels (the standard lipid profile) are routinely determined and the ratio of TC to HDL-​C calculated. The TC-​to-​HDL-​C ratio is one of the best predictors of atherosclerotic cardiovascular disease risk, and it is the plasma lipid metric used in standardized athero- sclerotic cardiovascular disease risk calculators (see ‘Screening of plasma lipid and lipoprotein levels’). Low levels of blood HDL-​C are very common in atheroscler- otic cardiovascular disease sufferers. Yet low HDL-​C is a not causal factor in atherosclerotic cardiovascular disease, rather than a marker of association with other risk factors. Low HDL-​ C is not independent despite its important role in reverse chol- esterol transport, and may reflect the clustering of primary and secondary atherosclerotic cardiovascular disease risk factors, which lead to the strong inverse correlation of low HDL-​C with atherosclerotic cardiovascular disease risk. HDL-​C levels are much affected by other atherosclerotic cardiovascular disease risk fac- tors including Toll-​like receptors, obesity, insulin resistance, and systemic inflammation. Primary causes of low HDL-​C (hypoalphalipoproteinaemia). As with the other components of the standard lipid profile, HDL-​C levels are determined by genetic and acquired factors. Genetic fac- tors determine approximately 50% of the total phenotypic variance of HDL-​C, and these are accounted for by the clustering of multiple common polymorphisms, low frequency, and rare variants (as pre- viously described). Increased risk of atherosclerotic cardiovascular disease in people with low HDL may occur due to the presence of other genes affecting pathways of lipid metabolism that increase risk. Common genetic variants that affect the components of meta- bolic syndrome also affect HDL levels. A number of the key genes involved in HDL metabolism have been found to have deleterious mutations, which affect HDL biosynthesis and catabolism and can result in dramatic reductions in plasma levels of HDL-​C (Table 12.6.11). Quite unlike the genes that confer high levels of LDL-​C, which greatly increase the risk of atherosclerotic cardiovascular disease, these genetic forms of hypoalphalipoproteinaemia are not defini- tively linked to increased risk of atherosclerotic cardiovascular disease. APOA1 gene cluster deletions The genes encoding apoA1, apoA5, apoC3, and apoA4 are grouped together on chromosome 11. Patients with deletion of the entire APOA1 gene, or of the whole gene cluster, have almost no HDL. In these patients, the absence of LCAT activation leads to increased free cholesterol levels in the blood and tissues such as the eyes and skin, which can form substantial deposits in the cornea and skin re- sulting in corneal opacities and palmar xanthomas. Despite having very low levels of apoA1, deficient patients are not at increased risk of atherosclerotic cardiovascular disease. APOA1 gene mutations Deleterious mutations of the APOA1 gene are a rare cause of low HDL-​C (often <0.5 mmol/​litre (20 mg/​dl)). A number of variants have been described and these are usually named after the place in which they were identified, such as apoA1 Milano and Marburg. They do not apparently cause premature atherosclerotic cardiovas- cular disease. This lack of association with atherosclerotic cardio- vascular disease is despite the finding that many apoA1 variants produce very low levels of plasma HDL-​C levels due to defective LCAT activation by apoA1 and enhanced removal of the abnormal HDL lipoprotein. ApoA1 Milano, which leads to dimerization of apoA1 with itself or apoA2, appears to decrease the risk of atherosclerotic cardiovas- cular disease, and has been used as a potential therapeutic agent, and this has led to development of infusible HDL mimetics that rapidly remove cholesterol from arteries and stabilize unstable plaque. In addition some coding sequence variants of APOA1 and APOA2 can aggregate and cause systemic amyloidosis. Table 12.6.11  Genetic defects affecting HDL metabolism Variant Molecular defect Inheritance Metabolic defect Lipoprotein abnormality Clinical features Premature atherosclerosis Familial apoA1 deficiency ApoA1,3,4,5 gene cluster Autosomal codominant Absent apoA1 biosynthesis HDL <0.5 mmol/​L, TGs increased when cluster is deleted Planar xanthomas, corneal opacities No Familial apoA1 structural mutations Abnormal apoA1 Autosomal dominant Rapid apoA1 catabolism and abnormal HDL HDL 0.4–​0.8 mmol/​L, TGs normal Often none, sometimes corneal opacities No Familial LCAT LCAT deficiency (complete) Autosomal recessive Rapid HDL catabolism HDL <0.5 mmol/​L, TGs increased Corneal opacities, anaemia, proteinuria, renal insufficiency No Fish-​eye disease LCAT deficiency (partial) Autosomal recessive Rapid HDL catabolism HDL <0.5 mmol/​L, TGs increased Corneal opacities No Tangier disease ABCA1 deficiency Autosomal codominant Very rapid HDL catabolism HDL <0.5 mmol/​L, TGs usually increased Corneal opacities, enlarged orange tonsils, hepatosplenomegaly, peripheral neuropathy No to yes Familial hypoalphalipoproteinaemia Unknown Autosomal dominant Usually rapid HDL catabolism HDL <1.0 mmol/​L, TGs normal Often none, sometimes corneal opacities No to yes

12.6  Lipid disorders 2083 Tangier disease (ABCA1 deficiency) Tangier disease is caused by rare Mendelian codominant defects of the ABCA1 gene. ABCA1 mediates the cellular efflux of unesterified cholesterol and phospholipids for capture by apoA1 from the liver, small intestine, and peripheral tissues. Without ABCA1, the poorly lipidated apoA1 is rapidly removed from the blood. Tangier disease patients have very low blood HDL-​C and apoA1 levels. Heterozygotes have low HDL-​C levels of approximately 0.5 mmol/​ litre (20 mg/​dl). Cholesterol collects in the mononuclear phagocyte system causing hepatosplenomegaly and pathognomonic enlarged greyish, yellow, or orange tonsils. Patchy peripheral neuropathy and rarely a syringomyelia-​like condition can occur. Nearly all children affected by Tangier disease are identified on the basis of large, yellow-​orange tonsils, but it can be undetectable or overlooked in adults because tonsils have often been removed. Foam cell formation from lipid storage in cells can be detected by endoscopic examination of the rectal mucosa. In many patients, proctoscopy reveals a pale mucosa studded with 1-​ to 2-​mm dis- crete orange-​brown spots. Other signs of Tangier disease are thrombocytopenia, anaemia, gastrointestinal disorders, and cor- neal opacities. The diagnosis can be confirmed by DNA sequencing. There is no clear evidence whether Tangier disease patients have an increased risk of atherosclerotic cardiovascular disease. LCAT deficiency Deleterious mutations of the LCAT gene cause a rare Mendelian recessive disease. LCAT secretion from the liver is decreased, and LCAT is reduced or absent from circulating lipoproteins. LCAT es- terifies free cholesterol in lipoproteins to form cholesteryl esters. LCAT deficiency greatly increases the proportion of free cholesterol in lipoproteins (from 25 to 70% of total plasma cholesterol). HDL maturation is defective and there is rapid clearance of apoA1—​ LCAT’s activating cofactor. Complete LCAT deficiency with complete absence of protein ac- tivity contrasts with partial deficiency of enzyme activity. In com- plete deficiency and partial deficiency (called fish-​eye disease), there is progressive corneal opacification caused by accumulation of free cholesterol in the cornea. Very low circulating HDL-​C and often hypertriglyceridaemia are features of both disorders. Complete LCAT deficiency is also associated with haemolytic an- aemia and progressive renal failure. In partial deficiency, there are no such clinical features. Premature atherosclerotic cardiovascular disease has been described, but is not a usual accompaniment to ei- ther form of the disease. The diagnosis is suggested by the corneal opacification and the combination of haematological abnormality, renal dysfunction with proteinuria, very low HDL-​C, and high TG levels. The LDL is cholesteryl ester poor and small and sometimes apoB is raised; these later features may be atherogenic. The diagnosis is made by measuring LCAT activity in plasma, and confirmed by DNA sequencing. Lipid-​lowering treatment should be considered. Familial hypoalphalipoproteinaemia (isolated low HDL) The familial clustering of low plasma HDL-​C, well below 1.0 mmol/​ litre (40 mg/​dl), with normal TGs and LDL particle size, with Mendelian (possibly codominant) transmission, without secondary causes, is called familial or primary hypoalphalipoproteinaemia. Some patients can have heterozygous defects of ABCA1 and strictly speaking have Tangier disease. LCAT deficiency can usu- ally be excluded clinically, but some patients may have APOA1 gene defects. Polygenic clustering of low HDL gene variants is the likely cause as no major gene has been discovered. The diagnosis is made in patients with no known genetic or sec- ondary causation. Mechanistically, there is often rapid catabolism of HDL, and of apoA1 and apoA2. It can be associated with athero- sclerotic cardiovascular disease, but the extent to which this is the case may depend on the underlying gene defects. Acquired (secondary) causes of low HDL-​C The secondary causes of dyslipidaemia are listed in Table 12.6.4. Obesity, insulin resistance, and diabetes By far the most significant among the acquired factors associated with low HDL-​C is the tide of obesity afflicting modern society, which leads to the metabolic syndrome of insulin resistance and type 2 diabetes. This clustering of metabolic abnormalities markedly increases the risk of atherosclerotic cardiovascular disease. HDL levels are decreased through complex effects on lipid and lipoprotein metabolism. There is increased lipolysis in adipose tissue; particularly intra-​abdominal fat, which is metabolically very active. Free fatty acids are released into the portal circulation. The liver converts free fatty acids into TGs. This and an increased supply of glucose, which leads to fatty acid and further TG biosynthesis and overproduction of VLDL, raises the concentration of circulating TG-​rich lipoproteins. There is reciprocal exchange of lipids between lipoprotein par- ticles. Cholesteryl esters are transferred to VLDL and chylomicron remnants, while TGs are transferred to LDL and HDL particles to form highly atherogenic, cholesterol-​poor sdLDL, and HDL (Table 12.6.1, Fig. 12.6.13). There is also rapid catabolism of HDL-​C and its core protein apoA1 with further reduction of HDL-​C. Atherogenic dyslipidaemia—​the simultaneous presence of raised TGs and apoB concentration, and an increased proportion of sdLDL with low HDL is associated with a considerable increase in the risk for atherosclerotic cardiovascular disease. The management of these lifestyle and metabolic factors and the treatment of raised cholesterol are the most efficacious approaches to manage low HDL (Table 12.6.6). Lipid lowering is reasonable. Renal disease Very rarely, patients with glomerulonephritis with massive protein- uria can develop very low levels of HDL-​C, which is self-​limiting with the resolution of the proteinuria. Urinary loss of apoA1 con- tributes to the low HDL-​C. Inherited causes of very high levels of HDL-​C Very high levels of HDL-​C with large, fluffy HDL-​C, and reduced particle number appear to be atherogenic. CETP deficiency Deleterious mutation of both copies of the CETP gene lead to a marked increase in HDL-​C levels (>4 mmol/​litre (160 mg/​dl)) and produce large, fluffy HDL without increased particle number, due to

section 12  Metabolic disorders 2084 the high cholesteryl ester content. Heterozygotes have only modestly raised HDL-​C levels. Normally, CETP transfers cholesteryl esters from HDL to apoB-​containing lipoproteins and TG in reciprocity (Fig. 12.6.14). In its absence, there is increased cholesteryl ester in HDL; plasma levels of LDL-​C are reduced. The large cholesterol-​rich HDL particles are cleared slowly. CETP deficiency is rare outside Japan, where it was first diagnosed. It is uncertain whether CETP deficiency causes or prevents atherosclerotic cardiovascular dis- ease, but low CETP activity is not associated with increased lon- gevity. Clinical trials do not yet support the use of CETP inhibitors to reduce the risk of atherosclerotic cardiovascular disease by raising HDL. Importantly, the finding of unduly advanced disease on vas- cular imaging should indicate the need for lipid lowering. Lipoprotein(a) production by the liver Lipoprotein(a) Lp(a) is synthesized exclusively in the liver. Its plasma levels differ greatly among people, and 75% of this reflects genetic variation in the LPA gene. Secondary factors such as renal disease, oestrogen de- pletion, and severe hypothyroidism can increase Lp(a) to a modest extent. As described earlier and in Fig. 12.6.5, a common copy-​number variation within the LPA gene determines the number of kringle IV repeats and hence the isoform size of apo(a). An inverse relation- ship exists between the number of repeats and Lp(a) plasma levels. A small number of common variant alleles account for much of the genetic variance conferred by the LPA gene locus. Variants particu- larly associated with increased atherosclerotic cardiovascular dis- ease tend to be small and be associated with an increased particle number. As a consequence of genetic variation, Lp(a) plasma levels can vary 100-​fold, where borderline risk is defined as greater than 30 mg/​dl (75 nmol/​litre), high risk is 50 mg/​dl (125 nmol/​litre), and very high risk is greater than 50 mg/​dl (>125 nmol/​litre), rarely up to a massive 300 mg/​dl (750 nmol/​litre). Lp(a) levels should be checked in any patient with premature car- diovascular disease, FH, family history of premature cardiovascular disease, family history of elevated Lp(a), recurrent cardiovascular disease despite statin treatment, and at least a 3% 10-​year risk of fatal cardiovascular disease according to the European guidelines. An increased level of Lp(a) is associated with a seriously increased risk of atherosclerotic cardiovascular disease, including both cor- onary disease and stroke. It also increases the risk of calcific aortic stenosis. The risk of atherosclerotic cardiovascular disease is greater with very high LDL-​C, particularly in FH. The mechanism by which an increased level of Lp(a) lipoprotein increases the risk of disease is not well understood. This may involve LDL-​C and atherogenesis, inhibition of conversion of plasminogen to plasmin, activation of tissue factor and thrombogenesis, or the carriage of proinflammatory oxidized phospholipids. Treatment is problematical (see ‘Drug treatment of hypercholes- terolaemia’ and ‘Lipoprotein(a)’). The treatment of acquired factors does not impact Lp(a) levels. Nicotinic acid and the new PCSK9 in- hibitors can each reduce Lp(a) by approximately 30%. Nicotinic acid is not available in the United Kingdom. CETP inhibitors also po- tentially lower Lp(a). An antisense Lp(a) mRNA inhibitor has also proved remarkably efficacious as a treatment in early clinical trials, and is in development. In one large clinical trial (AIM-​HIGH), however, after LDL-​C lowering to 1 to 2 mmol/​litre (40 to 80 mg/​dl) with statins no further benefit was accrued from the addition of nico- tinic acid to lower the Lp(a), suggesting that dramatic lipid lowering alone is a reasonable treatment. Aspirin or another antiplatelet drug should be given to suppress the thrombogenicity of Lp(a). Other secondary causes of dyslipidaemia Chronic kidney disease Modest elevation of TGs is frequently seen in CKD due to impaired lipolysis through alteration in the composition of circulating TGs, which become enriched with the LPL inhibitor apoC3. There is reduced LPL and HL activity. Lp(a) is increased due to decreased clearance. Atherosclerotic cardiovascular disease is common in se- vere CKD so that dyslipidaemia needs vigorous treatment with the combination of statins and cholesterol absorption inhibitors as this will decrease cardiovascular events, and mitigates the need for high-​ dose statins and potential toxicity. The reduction of LDL-​C/​NHDL-​ C is reasonable at all levels to reduce cardiovascular events. Transplant patients, due to the immunosuppressive drugs they re- quire, often have dyslipidaemia, which again needs careful, but vig- orous treatment with statins and cholesterol absorption inhibitors and sequestrants to help avoid muscle side effects from statins and possible nephrotoxicity. Anorexia nervosa Anorexia affects mainly teenagers and young adults, and is the third most common long-​term illness among teenagers. Fifty per cent of suffers develop dyslipidaemia, often with high total, LDL, and HDL cholesterol. TGs are not raised. The pathogenesis is not clear, and is certainly not simply due to malnutrition, which reduces plasma cholesterol levels. It is argued that lipid lowering is not necessary be- cause HDL is raised as well as LDL. TC levels can, however, be over 9 mmol/​litre (360 mg/​dl), so that careful review on a case-​by-​case basis is required. Alcohol and other substance abuse need to be con- sidered as exacerbating factors. Effective treatment of the primary disorder is the best way forward. Drugs Numerous medicinal drugs affect lipid and lipoprotein metabolism through various mechanisms and cause dyslipidaemia. Many drugs also interact with lipid-​lowering drugs, so that coadministration should be approached with caution (Tables 12.6.4 and 12.6.8). See also Table 12.6.4 for other secondary causes of dyslipidaemia. The patient with dyslipidaemia Cholesterol reduction with statins is highly effective in decreasing vascular events. TG reduction appears to be similarly effective, but the data are less robust than for cholesterol. Increasing HDL-​C is not effective in reducing events—​rather, HDL is a biomarker for other risk factors. The diagnosis of dyslipidaemia and its effective treat- ment is thus of major clinical importance. This section is based on the up-​to-​date evidence-​based guide- lines on lipids from the United States of America and Europe, with key differences highlighted (Table 12.6.12). It is supported by the latest thinking in the field. A particular focus of these guidelines is

12.6  Lipid disorders 2085 a patient-​centred approach to treatment options and lifestyle with emphasis on individual global risk factor assessment through the use of risk calculator tools. Screening of plasma lipid and lipoprotein levels The measurement of standard lipid and lipoprotein levels should be a routine part of clinical practice. Screening is best done by the general practitioner or as part of a health screen. According to guidelines from the United States of America, all adults above 21 years of age should have TC, TGs, HDL-​C, LDL-​C, and/​or NHDL-​C screened, and screening should be repeated approximately every 5 years. In the United Kingdom, screening is recommended in those over 40 years (Box 12.6.1). Screening is particularly strongly indicated in patients with established atherosclerotic cardiovascular disease (secondary prevention) with a view to establishing or maintaining treatment goals. It is also strongly recommended in those with a family history of atherosclerotic cardiovascular disease, particularly if premature, those with other atherosclerotic cardiovascular disease risk factors, all those with acute pancreatitis, and to monitor those on lipid-​lowering medication. Most clinical chemistry laboratories measure TC and TG enzymatically. The HDL-​C is usually measured after precipitation of apoB-​containing lipoproteins. The LDL-​C is then most commonly estimated using the Friedewald formula: [LDL-​C] = [TC] minus [TG divided by 2.2] minus [HDL-​C] (where all concentrations are given in mmol/​litre). For this, the VLDL cholesterol is estimated by dividing the plasma TG by 2.2, the usual TG to cholesterol in VLDL particles. Table 12.6.12  Comparison of international lipid guidelines: NICE (2014), ESC/​EAS (2011), and ACC/​AHA (2013) NICE guidelines ESC/​EAS guidelines ACC/​AHA guidelines Year of publication 2014 2011 2013 Use of evidence Comprehensive literature review Comprehensive literature review Randomized controlled trials Risk assessment tool QRISK2 SCORE Pooled cohort equations End points CHD death, CHD (MI or angina), stroke and transient ischaemic attack CHD death or fatal stroke CHD death, nonfatal MI, fatal or nonfatal stroke Derivation sample British population, updated annually 12 European countries (pooled data) 4 cohort studies (pooled data) Risk factors selected in the multivariable model Age, sex, total cholesterol, HDL-​C, systolic blood pressure, hypertension treatment status, diabetes mellitus, smoking status, ethnicity, family history of CHD, body mass index, socioeconomic deprivation, rheumatoid arthritis, CKD, and atrial fibrillation Age, sex, total cholesterol, systolic blood pressure, and smoking status (NB: separate models for high-​ and low-​ risk countries) Age, sex, total cholesterol, HDL-​C, systolic blood pressure, antihypertensive treatment status, diabetes mellitus, and smoking status (NB: separate models created for white patients and black patients) Cholesterol treatment targets endorsed 40% reduction in non-​HDL from pretreatment level LDL-​C (see below) Consider apoB or non-​ HDL-​C as alternative target No Statin therapy for primary prevention in those without diabetes mellitus 10-​year risk ≥10% or CKD LDL-​C ≥4.9 mmol/​litre (190 mg/​dl) LDL-​C <4.9 mmol/​litre (190 mg/​dl), and: • 10-​year risk ≥10% • Moderate to severe CKD LDL-​C ≥2.5 mmol/​litre (100 mg/​dl) and • 10-​year risk 5–​9.9% • Severe risk factors: LDL-​C ≥3.0 mmol/​ litre (115 mg/​dl) and • 10-​year risk 1–​4.9% LDL-​C ≥4.9 mmol/​litre (190 mg/​dl) LDL-​C 1.8–​4.8 mmol/​litre
(70–​189 mg/​dl) and: • 10-​year risk ≥7.5% • 10-​year risk <7.5% after consideration of other factors Statin therapy for primary prevention in those with diabetes mellitus Type 2 diabetes mellitus and 10-​year risk ≥10% Type 1 diabetes mellitus and age

40 years, duration of disease 10 years, nephropathy, or CVD risk factors Type 2 diabetes mellitus and LDL-​C ≥2.5 mmol/​litre (100 mg/​dl) High-​risk type 2 diabetes mellitusa and LDL-​C ≥1.8 mmol/​litre (70 mg/​dl) Type 1 diabetes mellitus and target organ damage LDL-​C ≥70 mg/​dl CKD considered a high-​risk feature Yes Yes No Recommendations for the elderly QRISK2 is validated to age ≤84 years SCORE validated for ages 40–​65 years Clinician judgement in elderly Pooled cohort risk equations not validated for age >79 years Consider lower-​intensity statin Additional considerations for risk assessment Non-​LDL-​C targets Non-​LDL-​C targets Lifetime risk ACC/​AHA American College of Cardiology/​American Heart Association; apoB, apolipoprotein B; CHD, coronary heart disease; CKD, chronic kidney disease; CVD, cardiovascular disease; ESC/​EAS, European Society of Cardiology/​European Atherosclerosis Society; HDL-​C, high-​density lipoprotein cholesterol; LDL-​C, low-​density lipoprotein cholesterol; MI, myocardial infarction. NICE, National Institute for Health and Care Excellence; SCORE, Systemic Coronary Risk Evaluation. a High-​risk type 2 diabetes mellitus is defined as diabetes mellitus plus one of the following risk factors: established atherosclerotic cardiovascular disease, CKD, age over 40 years, and one or more cardiovascular risk factor or target organ damage. Source data from with permission from Brown MS, Goldstein JL. Receptor-mediated endocytosis: Insights from the lipoprotein receptor system. Proc Natl Acad Sci U S A. 1979;76: 3330–3337.

section 12  Metabolic disorders 2086 While the Friedewald equation is an adequate method, it has shortcomings. Calculated LDL-​C is not accurate in patients who are nonfasting, have TGs greater than 5.0 mmol/​litre (450 mg/​dl), or have type 3 hyperlipoproteinaemia. The equation is particularly inaccurate when LDL-​C is below 1.8 mmol/​litre (70 mg/​dl). For ex- ample, calculated LDL-​C underperforms in obese diabetics, and de- viates significantly from LDL-​C directly measured at concentrations below 1.8 mmol/​litre (70 mg/​dl). LDL-​C can be measured directly by a variety of methods. As with calculated LDL-​C, directly measured LDL-​C has shortcom- ings. Directly measured LDL-​C can be discordant with other LDL-​ related measures and may not reflect atherosclerotic cardiovascular disease risk. Recent guidelines endorse the use of NHDL-​C (i.e. TC minus HDL-​C). NHDL-​C accurately predicts atherosclerotic cardiovas- cular disease risk. It does not suffer from the discrepancies with LDL-​C measurement. A  fasting sample is not generally needed, which is more convenient for patients. NHDL-​C is on average 1.0 mmol/​litre (40 mg/​dl) higher than LDL-​C. Nonfasting TGs are also better predictors than fasting TGs of atherosclerotic cardiovas- cular disease events. The main disadvantage of moving to NHDL-​C is that national guidelines have previously used (and in the United States of America still do use) LDL-​C. Furthermore, clinical trials and other clinical studies have often been performed using LDL-​C. It will take effort, but should improve patient care. The use of risk assessment tools such as QRISK2 (Box 12.6.2), Joint British Societies’ consensus recommendations for the pre- vention of cardiovascular disease (JBS2/​3), or the American heart risk calculator is recommended to assess absolute atherosclerotic cardiovascular disease risk for the primary prevention in people up an age between 75 to 84 years. Risk assessment using calcu- lators is based on large amounts of observational data from the general population in whom the risk predictions are valid. The lipid metric most calculators use is the TC-​to-​HDL ratio, which is the best predictive lipid parameter that we have. In addition, these tools evaluate the contribution of ethnicity, deprivation, dia- betes, kidney disease, and rheumatic conditions such as systemic lupus erythematosus to the risk of atherosclerotic cardiovascular disease. Global risk, however, has never been used as a selection criterion for statin trials. While risk is driven largely by older age, in the ab- sence of vascular risk factors, it merits treating by cholesterol reduc- tion as evidence suggests this will reduce risk. There is little data on HIV-​positive and solid organ transplant patients, but lipid lowering may be considered. Risk assessment tools that aid clinical decisions about lifestyle modification, and whether to use lipid-​ and blood pressure-​lowering medication are valuable, but should not replace clinical judgement. Statin treatment is recommended in those with a 10-​year risk above 7.5% in the United States of America and it is suggested that treatment be considered above 5% risk at 10 years. In the United Kingdom, the recommended threshold is 10%. Caveats to the use of risk calculators are given in Box 12.6.2. Risk calculators should not be used in those people with suspected genetic dyslipidaemia. Guides for the treatment of dyslipidaemia are given in Tables 12.6.7, 12.6.8, and 12.6.10 and Fig. 12.6.19. In a nonfasting individual, a NHDL-​C concentration greater than 5.5  mmol/​litre (220 mg/​dl) may indicate genetic hypercholester- olaemia that requires further investigation. If nonfasting TGs are greater than 5.0 mmol/​litre (450 mg/​dl), a fasting lipid panel should be performed; persistent elevation will suggest a genetic cause. Box 12.6.1  United Kingdom criteria for screening lipids 1 People with diagnosed coronary heart disease (CHD) or other oc- clusive arterial disease (cerebrovascular accident, peripheral vas- cular disease) not yet on cholesterol-​lowering therapy for secondary prevention. 2 People with diagnosed CHD or other occlusive arterial disease taking cholesterol-​lowering therapy for secondary prevention—​to check that target lipid concentrations are being achieved. 3 People without diagnosed CHD or other occlusive arterial disease not on cholesterol-​lowering therapy—​when CHD risk is to be estimated (e.g. in people known to have CHD risk factors, especially those with a family history of premature CHD1). (Risk assessment tool: http://​www. qrisk.org/​.) 4 People without diagnosed cardiovascular or other occlusive arterial disease taking cholesterol-​lowering therapy for primary prevention—​ to check that target lipid concentrations are being achieved. 5 People with CVD risk equivalents—​patients with diabetes mellitus, hypertension, or familial hypercholesterolaemia. 6 Patients admitted with acute pancreatitis. Box 12.6.2  QRISK risk assessment tool for atherosclerotic cardiovascular disease and its exclusion criteria • The QRISK2 risk assessment tool can be used to assess atherosclerotic cardiovascular disease (ASCVD) risk for primary prevention in people up to 84 years. • Do not use QRISK2 in people with an estimated GFR of less than 60 ml/​min per 1.73 m2 and/​or albuminuria as these people are at in- creased risk of ASCVD. • Do not use QRISK2 in people with type 1 diabetes as these people are at increased risk of ASCVD • Do not use QRISK2 for people with pre-​existing ASCVD • Do not use QRISK2 for people who are at high risk of developing CVD because of familial hypercholesterolaemia or other inherited disorders of lipid metabolism. • Use QRISK2 to assess CVD risk in people with type 2 diabetes. • Note that standard ASCVD risk scores will underestimate risk in people who have additional risk because of underlying medical conditions or treatments including:

—​ people treated for HIV

—​ people with serious mental health problems

—​ people taking medicines that can cause dyslipidaemia (e.g. anti- psychotic medication, corticosteroids, or immunosuppressant drugs)

—​ people with autoimmune disorders, e.g. systemic lupus erythematosus. • Note that ASCVD risk will be underestimated in people who are al- ready taking antihypertensive or lipid modification therapy, or who have recently stopped smoking. Use clinical judgement to decide on further treatment of risk factors in people who are below the ASCVD risk threshold for treatment. • Note that severe obesity (body mass index >40  kg/​m2) increases ASCVD risk. • Consider people aged 85 or older to be at increased risk of ASCVD because of age alone, particularly people who smoke or have raised blood pressure.

12.6  Lipid disorders 2087 Referral to a specialist lipid clinic is indicated in patients with (1) possible familial dyslipidaemia, (2) those who fail to respond adequately to diet and first-​line drug therapy, (3) those with severe hypertriglyceridaemia who are at risk of acute pancreatitis, (4) those in whom statin intolerance is severe, and (5) those for whom there is any uncertainty about diagnosis. In the lipid specialist clinic, estimate of lipids are still gener- ally done on a fasting blood specimen, taken 12–​14 h after an overnight fast with complete dietary restriction (with the excep- tion of water and medication). This is of value because postpran- dial TGs remain elevated for several hours, particularly in those with certain forms of dyslipidaemia, and to make an accurate Statin Benefit Groups Cardioprotective life-style is the basis of ASCVD prevention. In those not receiving statin, recalculate estimated 10-y ASCVD risk every 4–6 y in individuals aged 40–84 y without clinical ASCVD or diabetes and with LDL–C 1.8–4.74 mmol/L (70–189 mg/dL) Clinical ASCVD LDL–C ≥4.75 mmol/L (190 mg/dL) Age <84 y High-intensity statin (Moderate-intensity statin if not suitable for high-intensity statin) Age >84 y OR if not candidate for high-intensity statin Moderate-intensity statin High-intensity statin (Moderate-intensity statin if not candidate for high-intensity statin) Moderate-intensity statin Estimated 10-y ASCVD risk ≥10%* High-intensity statin Moderate-to-high intensity statin Adults age >21 y and a candidate for statin therapy No No No No Yes Yes Yes Yes Yes Yes Yes Definitions of high- and moderate-intensity statin therapy High Daily dose lowers LDL–C by appox. ≥50% Moderate Daily dose lowers LDL–C by appox. 30% to <50% Diabetes* Type 1 or 2 Age 40–84 y Estimate 10-y ASCVD risk with risk calculation tools ≥10% estimated 10-y ASCVD risk and age 40–84 y Benefit of statin in ASCVD prevention may be less clear in others Consider additional factors influencing ASCVD risk and potential ASCVD risk benefits and adverse effects, drug interactions, and patient attitude to statin treatment *In diabetes less than 40 years statins are optional and depend on clinical judgement Fig. 12.6.19  Statin therapy for atherosclerotic cardiovascular disease (ASCVD): primary and secondary prevention. Incorporating aspects of both the 2013 American College of Cardiology/​American Heart Association Blood Cholesterol Guideline for statin initiation and the National Institute for Health and Care Excellence guideline (CG181) ‘Cardiovascular disease: risk assessment and reduction, including lipid modification’. Note: the American guidance considers statin initiation at a 10-​year atherosclerotic cardiovascular disease risk of 5%/​7.5% whereas the United Kingdom threshold is 10%. The Americans also consider atherosclerotic cardiovascular disease risk at age 21 years in comparison to the United Kingdom where atherosclerotic cardiovascular disease risk is assessed at age 40. Reproduced from Stone NJ, et al. (2013). ACC/​AHA guideline on the treatment of blood cholesterol to reduce atherosclerotic cardiovascular risk in adults: a report of the American College of Cardiology/​American Heart Association Task Force on Practice Guidelines. Circulation. 24; 129 (25 Suppl 2):S1–​45 with permission from Wolters Kluwer and Rabar S, et al. (2014). Lipid modification and cardiovascular risk assessment for the primary and secondary prevention of cardiovascular disease: summary of updated NICE guidance. BMJ, 349, g4356 with permission from BMJ Group.

section 12  Metabolic disorders 2088 diagnosis of the nature of the dyslipidaemia, fasting TGs are preferable. Diagnosis of dyslipidaemia Once screening has established that a patient has a dyslipidaemia a full clinical, diagnostic evaluation is needed. The key questions are: (1) what classes of lipoproteins and lipids are increased or de- creased in the patient’s plasma? (2) Does the patient has a primary (genetic) or secondary (acquired) dyslipidaemia, or as is often the case contributions from both? (3) Is the patient at risk of atheroscler- otic cardiovascular disease or acute pancreatitis? (4)  What other atherosclerotic cardiovascular disease or pancreatitis risk factors are present? (5) What are the treatment options? The evaluation should include a full medical, family, social, and lifestyle history, and physical examination. The physical exam- ination should be thorough, but specifically focus on cutaneous, tendon and ocular manifestations of dyslipidaemia, and evaluation of the cardiovascular system. Blood tests should include standard lipids, urea and electrolytes with estimated GFR, liver function tests, fasting blood glucose, thyroid function tests including thyroid-​stimulating hormone and Lp(a) measurement (Figs. 12.6.5 and 12.6.6), and apoB and apoA1 as measures of apoB-​containing lipoproteins and HDL particles. ApoE phenotyping is indicated if TC and TGs are both elevated. In statin-​intolerant patients, SLCO1B1 genotyping may be indicated (see ‘Muscle’). Urine analysis should assess the presence of protein- uria. A resting ECG should be performed to alert to overt CVD. The clinical evaluation and tests described just described will de- termine the presence of secondary causes of dyslipidaemia including CKD and nephrotic syndrome, hepatitis and cholestasis, the meta- bolic syndrome and diabetes, and hypothyroidism. In the absence, or with only a minor contribution from a sec- ondary cause of dyslipidaemia, then a primary (genetic) cause is likely. However, patients will often display components of both. In primary dyslipidaemia, a full family history often with lipid studies in the family and occasionally specific tests will indicate the diagnosis. Imaging of the arteries by carotid ultrasonography or CT cor- onary artery calcium can be of value, as risk factors do not accur- ately predict the extent of atherosclerotic cardiovascular disease. The carotid ultrasonographic and coronary artery calcium scans are ap- proximately 70% correlated, and both detect the presence of athero- sclerotic plaque. The age at which coronary calcium can be detected is 40 years in men and 50 in women. A coronary artery calcium score greater than the 75th percentile is indicative of an increased risk of myocardial infarction or stroke. In younger patients with suspected FH, due to more ‘cholesterol-​ exposure years’ the finding of carotid plaques or coronary artery cal- cium will support the diagnosis as longstanding dyslipidaemia in FH leads to plaques at an earlier age. In polygenic hypercholesterolaemia in which the dyslipidaemia is considered to develop later than in FH, due to less ‘cholesterol-​ exposure years’ the coronary artery calcium is likely to be less advanced, and imaging may help the differential diagnosis from FH, if clinical criteria or a DNA diagnosis have not already established this. Risk assessment tools have recently been combined with coronary artery calcium scores and appear to be better predictors of risk than risk assessment tools alone, but require better validation, and may not help in younger people aged less than 40 years. More esoteric lipoprotein biomarkers are sometimes measured and these include LDL and HDL particle number, sdLDL measures, and HDL fractions, as well as markers of general inflammation such as high-​sensitivity C-​reactive protein and fibrinogen, and markers of arterial inflammation such as MPO and LpPLA2. Increased LDL particle number and the presence of sdLDL are directly pro-​atherogenic; low HDL particle number is a predictor of risk. Inflammatory markers indicate active disease. The rela- tionship of inflammatory markers to the incidence of vascular events is uncertain. However, for monitoring treatment, per- sistent inflammation may indicate the need for more aggressive treatment. Hypercholesterolaemia In the United Kingdom, the mean TC is approximately 5.2 mmol/​ litre (210 mg/​dl) for men and women. Recommended levels are TC 5  mmol/​litre (200 mg/​dl), NHDL-​C 4  mmol/​litre (160 mg/​ dl), and LDL-​C 3 mmol/​litre (120 mg/​dl) or less in healthy adults. Nonetheless, approximately 40% of myocardial infarctions occur in apparently healthy people with levels TC below these levels. After measuring lipid levels global risk assessment is, therefore, recommended. In those with a 10% or greater risk of an event in the next 10 years, cholesterol levels should be reduced according to United Kingdom guidelines. In the United States of America, a threshold of 7.5% for treatment is used, with the option for treat- ment at 5% risk. In younger people, in whom the likelihood of an event in the next 10 years is low, a lifetime assessment is reasonable, though not well validated by population data. TC levels in adults above 6 mmol/​litre (240 mg/​dl), that is, the 95th percentile, are high and abnormal and suggest the presence of genetic factors. In those with suspected primary hypercholesterol- aemia, risk calculation should not be used as the tools grossly under- estimate risk. Secondary causes also need to be identified. Often primary and secondary causes coexist. The differential diagnosis of primary from secondary factors is greatly helped by an accurate family history. A TC level above 7.5 mmol/​litre (300 mg/​dl) and a dominant pat- tern of inheritance (50% of siblings and children) with premature atherosclerotic cardiovascular disease will suggest the diagnosis of FH, and is diagnostic if accompanied by xanthomas (see ‘Familial hypercholesterolaemia’; Table 12.6.9). Severe hypercholesterolaemia without a family history is rare, but might suggest a recessive disorder and diagnosis of ARH, sitosterolaemia, or cholesteryl ester storage disease (see ‘Cholesteryl ester storage disease’). In suspected FH patients, DNA sequencing is likely to reveal a causal mutation in the LDLR or one of the other three genes asso- ciated with FH in 70% of patients with ‘definite’ FH (Table 12.6.9, Fig. 12.6.18). In patients with TC levels below or approximately 7.5 mmol/​litre (300 mg/​dl), with ‘possible’ FH, or in whom there is a family history, but not clearly dominant, perhaps coming from both sides of the family, a clear DNA diagnosis is less likely. It is likely to be achieved in approximately 30% of individuals. In patients, who do not receive a DNA diagnosis by sequencing, the clustering of polymorphic hypercholesterolaemia alleles and a

12.6  Lipid disorders 2089 polygenic mode of inheritance is likely, and may affect about 10% of family members. There is often overlap between FH and poly- genic hypercholesterolaemia, and polygenic factors can worsen the FH phenotype. Hypertriglyceridaemia Clinically, hypertriglyceridaemia can be classified as mild to mod- erate (fasting plasma levels >1.7–​2.3 mmol/​litre (150–​200 mg/​dl) and <5.0  mmol/​litre (450 mg/​dl)), severe (fasting 5–​10  mmol/​L, 450–​900 mg/​dl)), and very severe (fasting >10 mmol/​litre (900 mg/​dl). A fasting TG level elevated above 5 mmol/​litre (450 mg/​dl) sug- gests a primary cause, but secondary factors are often also present, of which undiagnosed or poorly controlled diabetes is common (Table 12.6.4). If a nonfasting level is above 5 mmol/​litre (450 mg/​dl), a fasting level should be obtained. Critically, those patients with TG levels above 10  mmol/​litre (900 mg/​dl) have an increased risk of acute pancreatitis and this becomes more likely with levels between 15 and 20  mmol/​litre (1350 and 1800 mg/​dl). These patients should be treated to lower their TG levels and the risk of pancreatitis. The ‘fridge test’ may help in making the initial diagnosis. If chylomicrons are observed (type 1 and 5 hyperlipidaemia), then familial chylomicronaemia or a related disorder must be considered. The measurement of LPL activity or DNA sequencing can help in making the diagnosis. The finding of chylomicrons is an indicator of an increased risk of acute pancreatitis. Many people with chylomicronaemia also have raised plasma VLDL (type 5). These people may not have a Mendelian problem, but a genetic predisposition plus acquired factors. Excessive VLDL secretion can swamp residual LPL function and grossly worsen hypertriglyceridaemia. In addition to diabetes, other secondary causes such as dietary indiscretion, obesity, insulin resistance, ex- cess alcohol consumption, and reproductive hormone deficiency or treatment should be considered (Tables 12.6.4 and 12.6.6). Mixed (combined) dyslipidaemia The diagnosis of patients with combined hyperlipidaemia may be tricky as a variety of genetic and acquired factors can be involved. High plasma levels of TC and TGs are seen in patients with in- creased plasma levels of VLDL, remnants lipoproteins, and LDL-​C. Measurement of plasma apoB levels should be performed. A careful family history helps make the diagnosis. If the TC is greater than 7.0 mmol/​litre (280 mg/​dl), FH must be considered, as approximately 30% of people have FH particularly if there is a Mendelian autosomal dominant family history. A DNA diagnosis can be ascertained by sequencing. In such patients, other polygenic factors or a secondary cause is likely to be responsible for the high TGs. If the TC and TGs are raised to around the same level, this sug- gests a defect of remnant clearance (remnants contain equimolar amounts to TG and cholesterol), and FDBL or type 3 hyperlipid- aemia must be considered (see ‘Familial dysbetalipoproteinaemia’) (Table 12.6.3). In cases of FDBL a TC/​ApoB ratio greater than 6.0 and a TG/​ApoB ratio of less than 10.0 are highly predictive, whereas in type 4 hyperlipidaemia, the TC/​apoB ratio is below 5.0 and in type 5 the TG/​apoB ratio is much greater than 10.0. In all such patients, apoE typing should be performed as the presence of apoE2 homo- zygosity will confirm the diagnosis. The family history of FDBL is of an autosomal recessive disorder. The secondary trigger factor should be discovered. An assay of plasma apoB levels also helps diagnose patients with FCHL. In FCHL, a family history is to be expected, but this is likely to be polygenic (see ‘Primary causes of VLDL production’). Those with increased apoB are at high risk of atherosclerotic cardiovas- cular disease. In FHTG, the apoB levels are not generally raised. HL deficiency is a very rare cause of mixed dyslipidaemia. Obesity, the metabolic syndrome of insulin resistance, and dia- betes are common secondary causes of mixed dyslipidaemia, often with atherogenic dyslipidaemia. There is frequent exacerbation by genetic factors. In mixed dyslipidaemia, ascertainment of the nature of the lipo- protein abnormality will greatly affect treatment options and efficacy. Treatment of dyslipidaemia The major treatment objectives are primary and secondary preven- tion of atherosclerotic cardiovascular disease and its complications, and prevention of hypertriglyceridaemic, acute pancreatitis. Cholesterol reduction to prevent atherosclerotic cardiovascular disease There is overwhelmingly robust data that the reduction in LDL-​C with statins at all ages less than 75 years, at all cholesterol levels, with and without other risk factors such as diabetes and smoking, greatly decreases the risk of atherosclerotic cardiovascular disease, and its complications such as angina, heart attack, heart failure, stroke, and overall mortality. Even at older ages evidence is reasonable that treatment reduces risk. Patients with high cholesterol must, there- fore, be assessed for atherosclerotic cardiovascular disease risk and appropriate treatment. This is particularly the case at the higher TC levels as seen in FH. All patients with atherosclerotic cardiovascular disease should be treated irrespective of initial cholesterol values. The treatment pathway for the patient with hypercholesterolaemia is illustrated in Fig. 12.6.19 and Table 12.6.12. In the United States of America, particular emphasis is given to patients with (1) clinical atherosclerotic cardiovascular disease; (2) primary TC levels above 4.75 mmol/​litre (190 mg/​dl); (3) diabetics above 40 years of age, with LDL-​C levels between 1.8 and 4.75 mmol/​litre (70 to 190 mg/​dl); and (4) estimated 10-​year risk greater than 7.5%. The recommended target reduction to be achieved by the treat- ment of dyslipidaemia is a greater than 40% reduction in NHDL-​ C by the National Institute for Health and Care Excellence in the United Kingdom; targets are no longer recommended in the United States of America. Previously, LDL-​C targets were used, but these have not been tested in clinical trials. Rather, now a high-​, moderate-​, and low-​ intensity statin treatment approach based on risk is adopted (Fig. 12.6.19). The ‘lower is better’ approach to achieve values of at least less than 2.5 mmol/​litre for NHDL-​C (equivalent to <1.8 mmol/​ litre for LDL-​C) is adopted by the JBS3, but not by guidelines from the United Kingdom or United States of America as it has not been tested (Fig. 12.6.20). In patients with other serious risk factors such as diabetes, CKD, and high Lp(a), it is reasonable to reduce LDL-​C levels to 1 mmol/​ litre (40 mg/​dl). The association between atherosclerotic cardiovascular disease and LDL-​C levels will probably bottom out at levels of 0.65 mmol/​

section 12  Metabolic disorders 2090 litre (25 mg/​dl), the ‘physiological level’ in animals and human neo- nates, so that potentially there is much scope for further risk reduc- tion, and no indication so far that we are overtreating to hazardously low LDL-​C levels. In addition, the concept that more ‘cholesterol-​exposure years’ as in FH, compared to polygenic and secondary hypercholesterolaemia patients and healthy people, is associated with increased risk sug- gests that reduction of cholesterol in early life could result in long-​ term benefit. Evidence to better support these important concepts is still to be established. Lifestyle Modification of lifestyle is important in the management of ath- erosclerotic cardiovascular disease risk. An atheroprotective life- style should be instituted (Table 12.6.6). In overweight and obese individuals, weight loss should be encouraged in line with re- cent guidelines (http://​www.nhs.uk/​livewell/​loseweight/​Pages/​ Loseweighthome.aspx in the United Kingdom and http://​www. cdc.gov/​healthyweight/​losing_​weight in the United States of America). This is particularly the case in patients with the meta- bolic syndrome or diabetes, where the risk of atherosclerotic cardiovascular disease is high. The patient should have expert nu- tritional counselling. They should expect on average a 5 to 10% reduction in chol- esterol levels, and this is coloured by genetic factors so that this can vary considerably. Aerobic exercise has only a small effect in lowering cholesterol levels, but it has general health and circula- tory health benefits in addition to improving lipids. Exercise can be dramatic in reducing TG levels. Other risk factors should be minimized. Drug treatment of hypercholesterolaemia The decision to implement cholesterol-​lowering drug treatment is determined by the level of LDL-​C/​NHDL-​C, the presence of other atherosclerotic cardiovascular disease risk factors, and overall ath- erosclerotic cardiovascular disease risk. A  risk assessment tool should be used to assess the 10-​year risk in patients more than 40 years of age. Exclusions to the use of risk assessment tools are shown in Box 12.6.2. This should aid clinical decisions about lifestyle and whether to use lipid and blood pressure-​lowering medication, but should not replace clinical judgement. National guidelines are of great help in making this decision (https://​www.nice.org.uk/​guidance/​cg181 for the United Kingdom and https://​www.guideline.gov/​summaries/​ summary/​48337 for the United States of America). Present guide- lines suggest a 10-​year risk of atherosclerotic cardiovascular dis- ease of more than 7.5% and more than 10% should be considered for statin treatment in the United States of America and United Kingdom respectively. FH or other genetic disorders of lipid metabolism are important exclusions to the use of risk assessment tools as they grossly under- estimate risk. In younger people, in whom the likelihood of an event in the next 10 years is low, a lifetime assessment is reasonable. The treatment of FH in children is discussed in previous sections. Statins Statins are the first-​line drug for the treatment of hypercholester- olaemia. The evidence base for their use to prevent atherosclerotic cardiovascular disease is very strong. Statins are orally active inhibitors of the cholesterol biosynthesis enzyme HMG-​CoA reductase. HMG-​CoA reductase is rate limiting in the multistep cholesterol biosynthesis pathway. Inhibition of chol- esterol biosynthesis reduces intrahepatic cholesterol levels, leading to activation of the LDLR gene and increased hepatic LDLR activity (see ‘Lipids’ and ‘Cholesterol’). The resultant increased uptake of LDL by the liver reduces plasma according to statin dose. Apart from lipid lowering, evidence does not support other mechanisms to re- duce atherosclerotic cardiovascular disease. Statins vary considerably in their structure, hydrophobicity and hydrophilicity, potency, and duration of action (Table 12.6.7). The efficacy of statins in reducing LDL-​C/​NHDL-​C is patient de- pendent, but beyond the initial lowest dose, doubling the statin dose incrementally reduces the LDL-​C by approximately 6%, that is, ap- proximately 20% overall. With high-​intensity treatment, reductions of LDL-​C/​NHDL-​C are routinely greater than 50%, and even with low-​intensity statin treatment, a 30% reduction can be anticipated. If TGs are raised, but to less than approximately 5.0 mmol/​litre (450 mg/​dl), statins also reduce TGs in the same proportion as LDL-​ C, and this is not statin-​type dependent. Statins are administered once daily orally and without adverse re- actions in most people. With the longer-​acting statins this can be at any time of day, but with the shorter-​acting statins, administration is best in the evening (Table 12.6.7). The guidelines for statin use are summarized in Fig. 12.6.19 and Table 12.6.12. Statin side effects One in ten people will experience mild to moderate side effects, which are fully reversible on drug withdrawal. The most common Fig. 12.6.20  Results of clinical trials and effect of LDL lowering on atheroma volume. The relationship between low-​density lipoprotein cholesterol levels and change in per cent atheroma volume for several intravascular ultrasonography trials. There is a close correlation between these two variables (r2 = 0.97). Note that in the Asteroid trial, treatment to LDL levels below currently accepted guidelines can actually regress atherosclerosis in coronary disease patients. A-​Plus, Asteroid, A Study to Evaluate the Effect of Rosuvastatin on Intravascular Ultrasound-​Derived Coronary Atheroma Burden; Avasimibe and Progression of Lesions on Ultrasound; Camelot, Comparison of Amlodipine vs Enalapril to Limit Occurrences of Thrombosis; Reversal, Reversal of Atherosclerosis With Aggressive Lipid-​Lowering. Source data from Nissen SE, et al. (2006). Effect of very high-​intensity statin therapy on regression of coronary atherosclerosis: the ASTEROID trial. JAMA, 295(13), 1556–​65.

12.6  Lipid disorders 2091 side effect is muscle pain, due to interference with muscle mitochon- drial electron transport function. Indigestion, headache, fatigue, and joint pain also occur. Muscle The risk of statin-​induced myopathy is increased by age, renal im- pairment, concomitant administration of drugs, and dietary com- ponents (grapefruit juice in large quantities increases statin side effects by increasing blood levels) that interfere with the oxidation of statins, mainly through using the same cytochrome P450 (CYP) pathway (Table 12.6.7), and those with a previous history of statin-​ associated muscle pain. The risk of statin myalgia and myopathy is statin and dose related. It depends in part on a pharmacogenetic variation of SLCO1B1 (OATP2), which transports statins into the liver, and the CYPs that metabolize statins. Polymorphic variants of the SLCO1B1 gene are present in 15% of the population, with a homozygous frequency of 1 in 200. SLCO1B1 variants increase blood statin levels, and are associated with myalgia, and myopathy with raised creatine kinase (CK) levels, as well as the other side effects. A single copy of one of the SLCO1B1 variants increases the risk of muscle problems from high-​dose simvastatin treatment ap- proximately fivefold, and two copies by approximately 20-​fold. By contrast, rosuvastatin treatment is less associated with muscle com- plaints, and this is regardless of SLCO1B1 genotype, and despite the greater potency of rosuvastatin. Atorvastatin appears to be less likely to cause muscle and other toxicities than simvastatin. It is possible that the increased circulating levels of the more hydro- philic statins (pravastatin and rosuvastatin) produced by SLCO1B1 variants are less toxic to muscle than the more hydrophobic statins such as simvastatin, even though the more hydrophilic statins such as rosuvastatin are more dependent on the SLCO1B1 protein for transport into the liver. In consequence of its greater toxicity, simvastatin is less favoured for use. The differences between statins are in part due to alternative pathways of statin metabolism, as well as genetic variation of the CYP450s. Myopathic complaints are apparently more likely with statins oxidized by CYP3A4, simvastatin, and atorvastatin rather than those not oxidized by CYP3A4, pravastatin, and rosuvastatin. Although there are insufficient data to come to firm conclu- sions about the value of SLCO1B1 genotyping in predicting statin side effects, available data do suggest that those with the decreased transport variants are at increased risk of statin side effects. These genotyping assays are available and may be of value in those patients in whom high-​intensity statin therapy is indicated, particularly if their history indicates previous side effects on statins. In patients in whom muscle problems or other side effects are serious, genotyping of SLCO1B1 variants is now common practice as this may inform the choice of statin and dose. SLCO1B1 homozygotes are at greater risk of serious myopathy and the fortunately rare rhabdomyolysis. Serious myopathy is best avoided by starting statins, particularly in those at risk, at modest doses rather than straightaway at the top dose. The exception to this ‘slowly, slowly’ approach is in patients with acute coronary syndromes, when the top dose of the more powerful statins is given. Patients who start statins should visit their doctor at once if they get unexpected muscle pain. In patients who get muscle pain, the plasma CK should be measured to differentiate myalgia from my- opathy. In the event of raised CK greater than five times the upper limit of the reference range, the statin should be stopped as there is a risk of rhabdomyolysis. This is a serious condition likely to require hospitalization, as it can cause acute tubular necrosis in the kidneys. CK levels in patients on statins are not routinely measured as ele- vated CK without muscle pain is not a feature of myopathy, and does not mean the statin needs to be stopped. Low vitamin D levels can exacerbate symptoms, so that vitamin D should be measured, and replenished if low. There are no arguments for the use of CoQ10 to reduce statin muscle side effects. Raised CK levels have other causes, the most common of which is physical training. Primary muscle disease is also potentially a problem and may militate against statin use and needs specialist as- sessment. MacroCK, a CK-​IgG antibody complex often associated with an underlying autoimmune myositis, or oligomeric mitochon- drial CK often seen in patients with malignancy or hepatic disease are not uncommon, and are diagnosed by simple laboratory tests. A benign inherited hyperCKaemia due to defective caveolae, can also confound interpretation. The management of statin muscle problems is to (1)  reduce dose and increase incrementally until the threshold for side ef- fects is reached or (2)  change the statin or resort to nonstatin lipid-​lowering medications. Atorvastatin or rosuvastatin given on a weekly, biweekly, or alternate-​day basis are efficacious because of their long half-​life and potency (Table 12.6.7). Referral to a spe- cialist centre is advised for severe problems. CK levels should be monitored. Liver Liver enzymes should be measured prior to starting statin therapy, at 3 and 12  months according to United Kingdom guidelines. Measurement is no longer recommended in the United States of America unless there are signs or symptoms of liver disease—​ fatigue, weakness, loss of appetite, abdominal pain, or icterus. Statins can be started with safely in patients with abnormal liver enzymes, even over three times normal, but if after starting statins liver transaminases (alanine transaminase and aspartate aminotransferase) become further raised, the statin should be stopped while the situation is assessed. Although data are not available on the toxicity of hydrophilic (Table 12.6.7; actively transported by SLCO1B1) versus hydro- phobic (passive diffusion and high first-​pass uptake) statins, a dif- ferent statin or a lower dose can be introduced to determine the effect on liver function. Abnormal transaminases are also common in the fatty liver that commonly accompanies obesity, insulin resistance and diabetes, and excess alcohol consumption, and can be confused as being due to statins. The statin-​associated increase in transaminases resolves upon discontinuation of the medication, whereas that due to other causes does not. Serious statin-​induced hepatitis is very rare, so that there is a reduced tendency to monitor liver function as indicated by guidelines from the United States of America. Too many physicians stop statin medication unnecessarily and forget the benefits of statins in reducing cardiovascular morbidity and mortality by approximately 50% because they think a small rise in the liver function tests means that there is ongoing damage to the liver.

section 12  Metabolic disorders 2092 Statins can be used safely in patients with chronic liver disease and well-​treated cirrhosis, but the physician may need to follow the pa- tient more closely than would occur in a normal healthy patient on a statin, and the same applies to approximately threefold increases in transaminases. Increased transaminases must first be established to be due to statins. If so, the dose can be reduced or the statin changed. Persistently raised transaminases can be tolerated but requires careful monitoring for risk of fibrosis or cirrhosis. Brain Hydrophobic statins enter the brain, whereas hydrophilic statins do not (Table 12.6.7). Not infrequently hydrophobic statins cause cerebral symptoms including sleep disturbance, vivid dreams, anx- iety, and memory disturbance. These side effects are fully reversible on stopping the hydrophobic statin and substituting a hydrophilic statin. There is no evidence that the statin memory disturbance con- tributes to dementia. Other causes of these symptoms such as other medications, neuropsychiatric conditions, or organic brain disease must be excluded. Diabetes and other side effects A serious consequence of statin therapy is an increased risk of dia- betes with an estimated risk in excess of 10%, particularly with higher doses of the potent statins. In most people with high ath- erosclerotic cardiovascular disease risk, the reduced risk outweighs the risk of diabetes. Statins are also the main treatment to prevent macrovascular disease in diabetics. If the glycated haemoglobin or fasting glucose levels increase or indeed diabetes develops, changing statin therapy is not recommended. Lifestyle modification may miti- gate the risk or the diabetes. Very rarely, neuropathy, autoimmune myopathy—​distinct from toxic myopathy, an autoimmune lupus-​like syndrome, autoimmune hepatitis, and pancreatitis are ascribed to statin use. Haemorrhagic stroke has not been a problem even when the LDL-​C is very low. The value of statin in the treatment of those with advanced heart failure or on haemodialysis is uncertain. Otherwise, in very large meta-​analyses, statins have been shown to be effective and safe at all cholesterol levels and at all adult ages. There is no propensity to develop cancer. They are the most used lipid-​lowering drug. Ezetimibe Ezetimibe is a small-​molecule drug that specifically blocks NPC1L1 and intestinal cholesterol absorption (see ‘Intestinal lipid absorption and transport as chylomicrons’). Intestinal cholesterol is two-​thirds derived from bile and about one-​third dietary. A single ezetimibe 10 mg dose reduces cholesterol absorption by more than 50% so that that there is signification reduction of delivery of intestinal choles- terol to the liver and consequent induction of LDLR activity. Plasma LDL-​C is reduced by about 20%, and this adds to the ef- fect of statins when the two drugs are used together. Occasionally a larger dose of 20 mg has been used safely, but this is not gener- ally efficacious. Ezetimibe has no effect on TGs or HDL. It is only used alone in those intolerant of statins. For sitosterolaemia, it is the drug of choice. It has been shown to be efficacious in redu- cing atherosclerotic cardiovascular disease events in clinical trials when used with statins. Deficiency of the NPC1L1 gene is associated with reduced atherosclerotic cardiovascular disease risk, again sup- porting its therapeutic role. Ezetimibe is safe and well tolerated. The main side effects are headache and gastrointestinal in about 1% of people; infrequent side effects are myalgia, abnormal liver function tests, and rarely hyper- sensitivity reactions (rash, angio-​oedema) or myopathy may occur. Cases of rhabdomyolysis have been reported, as has pancreatitis. Ezetimibe is used as an adjunct to maximum statin dosing and when statin side effects occur. Bile acid sequestrants (resins) These time-​tested drugs bind bile acids (Table 12.6.8) in the intes- tine, and promote their excretion instead of their normal reabsorp- tion by the ileum. In turn, the liver synthesizes more bile acids from cholesterol, which reduces the liver cholesterol pool and induces the LDLR with increased LDL-​C clearance. Bile acid sequestrants are traditionally available in resin form as cholestyramine and colestipol, which must be suspended in water; colesevelam is a tablet. Sequestrants need to be taken twice a day with the higher dose in the morning with breakfast and some- what less in the afternoon. Colesevelam is taken as up to four tab- lets with breakfast and up to three tablets in the afternoon. Taking sequestrants with breakfast is desirable as it coincides with emptying of the gallbladder, which fills overnight, and facilitates maximum bile acid binding. The effect of sequestrants is dose dependent. At maximum dosing with colesevelam, around a 30% reduction in LDL-​C levels can be achieved. Resins should be given with caution in patients with hyper­ triglyceridaemia as they can worsen this. Interruption of the entero- hepatic recirculation of bile acids has important effects on hepatic lipoprotein metabolism. Activation of phosphatidic acid phos- phatase promotes hepatic TG synthesis and induces VLDL secre- tion, and consequently, increases plasma TG levels. Sequestrants are not systemically absorbed so that other side ef- fects are restricted to intestinal bloating and constipation. In conse- quence, they are very safe and can be used in pregnancy and during lactation. They are efficacious when given with statins or ezetimibe, but need to be taken several hours apart from other drugs and vitamins, which they bind to and prevent their absorption. They can be taken with statins and ezetimibe for the management of severe hyperchol- esterolaemia. In statin intolerance, they can be used alone or with ezetimibe. Resins have not been tested in clinical trials. Their use needs to be determined by clinical experience. Recalcitrant heterozygous FH, and homozygous FH, including those with high Lp(a), are indica- tions, particularly with symptomatic atherosclerotic cardiovascular disease. They are also indicated in those with serious statin side ef- fects, and in pregnancy and lactation. Nicotinic acid Nicotinic acid, also called niacin or vitamin B3, has been used to lower lipids over many years, particularly in the United States of America, but is no longer recommend for use in Europe (see next paragraph). In adipocytes, it decreases lipolysis by activation of the nicotinic acid receptor (NIACR1, a high-​affinity G protein-​coupled receptor) which reduces the levels of intracellular cAMP thereby inhibiting lipolysis. In the liver, nicotinic acid suppresses APOC3

12.6  Lipid disorders 2093 gene expression, thereby enhancing LPL activity. It suppresses VLDL and LDL-​C by about 30% at optimum doses and raises HDL-​ C comparably. Clinical trial data do not support its use to raise HDL for prevention of atherosclerotic cardiovascular disease. Clinical trials, however, have shown that combination use of statins with drugs containing nicotinic acid did not lead to add- itional benefits in reducing the risk of major vascular events such as heart attack and stroke, but did result in a higher frequency of non- fatal but serious gastrointestinal events and infection. As a result, nicotinic acid is not available for use in Europe. It is also, with PCSK9 and possibly CETP inhibitors, one of the few drugs that reduce Lp(a). A 30% reduction in Lp(a) can be anticipated at optimum doses of these drugs, but this is of uncertain value (see ‘Lipoprotein(a)’). Cutaneous flushing is a troubling side effect of nicotinic acid. It is mediated through NIACR1 and prostaglandin in the skin. Nicotinic acid therapy is therefore usually started at a low dose and slowly increased to higher doses, under the cover of aspirin to re- duce prostaglandin activity and flushing. Nicotinic acid can cause dyspepsia, mild increases in transaminases, and plasma uric acid. It can precipitate gout in susceptible people. Acanthosis nigricans and maculopathy are rare side effects. There are no strong arguments for the use of nicotinic acid in lipid lowering. PCSK9 inhibitors Inhibitors of PCSK9 are fully human monoclonal antibodies that block PCSK9 at the liver surface and reduce LDLR degradation. This concept emerged from the discovery that loss-​of-​function muta- tions in the PCSK9 gene reduce plasma LDL-​C, along with the risk of atherosclerotic cardiovascular disease. They have proved highly efficacious in meta-​analysis of clinical trials in reducing LDL-​C and TGs by approximately 50% and increasing HDL, but end-​point trials have not been reported as yet. They also reduce Lp(a) by 30%. These new and exciting drugs have recently received regulatory approval by the FDA and European Medicines Agency, but the criteria for general use are restricted to severe genetic hypercholesterolaemia with progressive atherosclerotic cardiovascular disease and failure of adequate response to other lipid-​lowering agents. Their initial target is genetic hypercholesterolaemia, where cholesterol lowering is not adequate, or when there are severe statin side effects. With very very low LDL-​C levels, the risk of haemorrhagic stroke has not materialized. Dietary supplements Supplements to the diet with plant sterols or stanols (such as FloraProActive or Benecol respectively) which compete for chol- esterol absorption thus reducing plasma cholesterol levels can be used as an adjunct to lifestyle measures. A daily intake of 1.5 to 2.4 g sterols or stanol ester can lower the plasma cholesterol by 7 to 10% in 2 to 3 weeks as part of a healthy diet and lifestyle. Drugs for homozygous FH Two orphan drugs are available to treat homozygous FH. Lomitapide is an inhibitor of MTTP, which mimics abetalipoproteinaemia in its mechanism of action, and suppresses VLDL secretion. It is approved for use in homozygous FH. It can achieve a greater than 30% reduc- tion of LDL-​C when used alone or in conjunction with apheresis. It is effective in receptor-​negative patients, because its action is inde- pendent of LDLR activity. Its side effects are mechanism of action based, with intestinal upsets due to fat malabsorption and fatty liver. Progressive hepatic fibrosis has not been problematic thus far. Another drug is mipomersen a second-​generation 2ʹ-​O-​ methoxyethyl chimeric antisense oligonucleotide, which inhibits the synthesis of apoB. It is approved in the United States of America for treatment of homozygous FH. It achieves a similar LDL-​C reduc- tion to lomitapide. As with lomitapide, mipomersen causes intes- tinal upsets due to fat malabsorption and fatty liver. It has not been approved in Europe due to a 50 to 70% rate of side effects, mainly injection site reactions, flu-​like symptoms, liver enzyme elevations, and proteinuria. PCSK9 inhibitors also reduce LDL-​C by greater than 30% in homozygous FH patients, who are receptor deficient, but have no effect in receptor-​negative patients. LDL apheresis Apheresis (ἀφαίρεσις ‘a taking away’) is a physical approach to re- moving LDL from the blood, analogous to haemodialysis. The blood of the patient is passed through a separator, which removes LDL by specific binding to columns and spares other lipoprotein fractions including HDL and returns the blood to the patient. Most homozygous FH patients do not achieve satisfactory reduc- tion of LDL-​C levels on maximum drug treatment alone and disease will progress. Those with atherosclerotic cardiovascular disease and LDL-​C levels above 5 mmol//​litre (200 mg/​dl), or without athero- sclerotic cardiovascular disease and LDL-​C levels above 7 mmol/​ litre (280 mg/​dl) are candidates for apheresis. Apheresis is performed weekly or twice-​monthly depending on the degree of lipid lowering achieved. It is well tolerated and achieves a good reduction in LDL-​C levels, though there is rebound at the end of the treatment cycle to high levels of LDL-​C, due to increased synthesis and defective LDLR function. Atherosclerotic cardiovas- cular disease still progresses, but at a reduced rate. The main disadvantage is long-​term access to the circulation, which is best achieved by an arteriovenous fistula, but venous access or a central line can be used. The advent of effective apheresis and effective drug treatment with conventional drugs has greatly improved the prognosis for homozy- gous FH. Further improvement is hoped for and anticipated with the drugs described previously described. Lipoprotein(a) The treatment of elevated Lp(a) is a problem. It is a serious ath- erosclerotic cardiovascular disease risk factor and a risk factor for calcific aortic stenosis. There is no truly effective treatment. Both nicotinic acid and PCSK9 inhibitors reduce Lp(a) by approximately 30%, but the efficacy of this is uncertain compared to dramatic lipid lowering. Nicotinic acid is not available in the United Kingdom. The role of CETP inhibitors in lowering Lp(a) has still to be established. A  second-​generation antisense Lp(a) mRNA inhibitor has also proved remarkably efficacious as a treatment in early clinical trials, and is in development. In the large AIM-​HIGH clinical trial, after LDL-​C lowering to 1 to 2 mmol/​litre (40 to 80 mg/​dl) with statins no further benefit was accrued from the addition of nicotinic acid to lower the Lp(a). Profound lipid lowering should be aimed at and this may be

section 12  Metabolic disorders 2094 sufficient. Whether this applies with very high Lp(a) levels (>100 mg/​dl (250 nmol/​litre) is not known. Aspirin or another antiplatelet drug should be given to suppress the thrombogenicity of Lp(a). In patients with high Lp(a) and symptomatic atherosclerotic car- diovascular disease, apheresis is effective in reducing Lp(a) levels and has potential to reduce disease progression, but its use is un- likely to be commonplace. Triglyceride reduction to prevent acute pancreatitis and atherosclerotic cardiovascular disease The primary treatment goal in severe hypertriglyceridaemia is to lower TGs rapidly to reduce the risk of acute pancreatitis. Elevated plasma TGs are also a risk factor for atherosclerotic cardiovascular disease, as previously discussed. A secondary goal is therefore to re- duce the risk of atherosclerotic cardiovascular disease. TG levels in patients are best measured on fasting samples, be- cause in the nonfasting state diet can have a profound effect on TG levels. Nonfasting TGs are better predictors of atherosclerotic car- diovascular disease risk than fasting TG, probably because they better reflect our usual status. Patients with fasting TGs levels above 10 mmol/​litre (900 mg/​dl) are at increased risk of acute pancreatitis. Although in practice acute pancreatitis is rare with levels below 15 mmol/​litre (1300 mg/​dl), fat consumption can readily achieve this level in the predisposed. Those patients that display chylomicronaemia (type 1 and 5 hyper- lipidaemia) as ascertained by the ‘fridge test’ are at particular risk of acute pancreatitis. TGs can vary markedly and rapidly and a patient with only mod- erately elevated TGs may develop acute pancreatitis following a short period of dietary indiscretion, which leads to much higher TG levels. There are also patients, however, with persistent marked hypertriglyceridaemia who never develop pancreatitis. Pancreatitis is therefore an unpredictable complication of hypertriglyceridaemia and usually strikes unexpectedly. It is thus generally accepted that patients with very high TG levels should be treated to reduce TGs and the risk of acute pancreatitis. No clinical trial has been per- formed to validate this view. Lifestyle Moderately severe hypertriglyceridaemia (<15  mmol/​litre (1300 mg/​dl)) in the absence of chylomicrons (type 4 hyperlipidaemia) can be managed in the outpatient setting. Lifestyle changes will usually significantly reduce plasma TG levels. A reasonable dietary goal is to restrict total fat intake to around 20 to 30 g daily. This is not always easy to achieve, because normal consumption is approximately 70 g daily, and requires dedication from the patient in understanding their dietary fat consumption. Excessive intake of starch and sugar should be discouraged because they drive TG production in the liver. A formal dietary consultation and regular review with a dietitian with specific experience in the management of severe hypertriglyceridaemia is desirable. Dietary fat restriction needs constant reinforcement. Spiking TG values on follow-​up are often related to dietary indiscretions. Alcohol con- sumption should be limited or stopped. Regular physical activity, particularly strenuous activity, is valu- able to reduce TG levels, and may have dramatic results. Weight reduction by diet and exercise should help decrease TG. In obese and overweight individuals, weight loss should be encouraged. ‘The rescue diet’ In more severe hypertriglyceridaemia (>15 mmol/​litre (1300 mg/​ dl)), especially with chylomicronaemia, and out-​of-​control diabetes, the patient is often best managed in the hospital setting to achieve rapid control. An extremely low-​fat diet, less than 10 g of fat daily, for about 3 days is necessary (Table 12.6.13). This diet is called the ‘rescue diet’ as it rapidly lowers TGs. This strict low-​fat diet is not easy to maintain and not nutritionally adequate in the long term. Secondary factors Other factors contributing to hypertriglyceridaemia need to be ac- tively sought and treated. In clinical practice, the most common problem is either undiagnosed or uncontrolled diabetes. In suscep- tible individuals, certain drugs, such as oestrogen, steroids, retin- oids, and protease inhibitors, can also trigger hypertriglyceridaemia (Table 12.6.4). If drugs are contributing significantly to hyper­ triglyceridaemia, treatment should be switched or discontinued if the patient’s clinical condition allows and there are effective alterna- tive treatment options. Further information on the secondary causes of hypertriglyceridaemia is found in Table 12.6.4. Drug treatment of hypertriglyceridaemia Severe hypertriglyceridaemia with TGs above 5  mmol/​litre (450 mg/​dl) despite adequate lifestyle management is likely to need drug treatment. Fibrates and omega-​3 fatty acids derived from fish are the only drugs available to treat hypertriglyceridaemia in the United Kingdom. Nicotinic acid is not used in Europe, but is in the United States of America (see ‘Nicotinic acid’). Statins can reduce TG, when levels are below 5 mmol/​litre (450 mg/​dl), but have no value in severe hypertriglyceridaemia. Statins may be necessary with fibrates if LDL-​C/​NHDL-​C remains high after TGs have been controlled (Table 12.6.8, with caveats for the combined use of statins and fibrates). Ezetimibe does not lower TGs significantly but can be combined with fibrates if additional LDL-​C lowering is required and statins are not tolerated. Cholestyramine can raise TGs and should be avoided in moderate to severe hypertriglyceridaemia. Fibrates Fibrates are central to the management of severe hypertrigly­ ceridaemia (TGs >5 mmol/​litre (450 mg/​dl)) and are the drugs of first choice. They lower TGs, increase HDL-​C, and may either lower or in some cases increase LDL-​C. Fibrates are particularly effica- cious in FDBL. They do not very much decrease atherosclerotic cardiovascular disease events due to hypercholesterolaemia, but are efficacious in hypertriglyceridaemia. Fibrates regulate lipid metabolism by their agonist effect on the nuclear receptor PPARα. PPARα stimulates LPL and apoA5 expres- sion and inhibits apoC3 expression. The increase in plasma HDL-​ C depends partly on an overexpression of apoA1 and apoA2. They also increase fatty acid β-​oxidation by mitochondria. An increase in LDL-​C arises in hypertriglyceridaemic subjects when more efficient lipolytic processing brought about by fibrates results in increased LDL-​C production.

12.6  Lipid disorders 2095 They are safe and have few side effects, but can increase the likeli- hood of gallstones. In the presence of existing gallstones they should be used with caution. Fibrates (particularly gemfibrozil) are associ- ated with toxic myopathy especially when combined with statins or nicotinic acid (Table 12.6.8). Care and appropriate monitoring is needed in patients on anti- coagulants and some diabetic blood glucose-​reducing drugs as fibrates interact with these classes of drug. Fibrates are excreted renally and doses need to be adjusted to renal function. They raise creatinine by about 10%, but this is not due to a lowered GFR and reverses on discontinuation. In the light of recent reanalysis of clinical trials of fibrates, their use in treating mild to moderate hypertriglyceridaemia and preventing atherosclerotic cardiovascular disease needs reconsideration. Omega-​3 fatty acids Omega-​3 polyunsaturated fatty acids or fish oils are present in high concentration in oily fish. They come from a variety of plants sources, but omega-​3 fatty acids of plant origin are less well studied, and are not a recommended substitute for fish oils. Eicosapentaenoic acid and docosahexaenoic acid are the main active ingredients in fish oil. Fish oils are given in capsules as 4 g daily in divided doses, and effect- ively reduce TG levels in moderately severe hypertriglyceridaemia, and may lower TGs by up to 40% in some patients. They work in part mechanistically by increasing the turnover of MLXIPL mRNA (see ‘Lipids’ and ‘Triglycerides’), which inhibits the de novo biosynthesis of fatty acids from carbohydrate. Fish oils are effective for the treatment of moderate hyper­ triglyceridaemia with levels of approximately 5 mmol/​litre (450 mg/​ dl). With more severe hypertriglyceridaemia, they are good in com- bination with fibrates. Higher doses of up to 12 g have been used with apparent safety and efficacy in very severe hypertriglyceridaemia, but should not be used in pure type 1 hyperlipidaemia, where they may exacerbate the phenotype. High doses can also be used with ap- parent safety in hypertriglyceridaemia during pregnancy. They can cause a modest increase in LDL-​C. The main side effect is dyspepsia. They may increase the bleeding time. It is important that the fish oil is purified to remove mercury, dioxins, polychlorinated biphenyls, and other toxins that contam- inate fish, particularly if prolonged use is anticipated or in pregnancy. There are active clinical trials to see if they reduce atherosclerotic cardiovascular disease risk due to high TGs. Table 12.6.13  Hypertriglyceridaemia rescue diet Daily menu Grams of fat Grams of fat Breakfast (1.7 g) 125 ml orange juice 0.3 1 banana 0.4 3/​4 cup Rice Krispies 0.0 250 ml skimmed milk 0.5 1 slice white bread 0.5 15 ml honey 0.0 Lunch (1.6 or 2.4 g) 2 medium potatoes (2 slices bread) 0.2 (1.0) 60 g fat-​free cottage Salad (lettuce, cucumber, tomato ...) 0.5 cheese 0.9 Supper (2.4 or 3.4) 375 ml white rice 0.6 (1.6) 125 ml tomato/​onion mix 0.4 125 ml lentils 0.4 Vegetables (carrot, broccoli) 0.4 Fruit (3 slices of pineapple) 0.6 Snacks (1.3 g) Apple, morning 0.6 Pear, afternoon, morning 0.7 Other supplements No diabetes Diabetes Beverages Carbonated drinks including colas Lucozade Fruit juice, including orange, apricot, apple, grape Dietetic cold drinks Low-​calorie Lecol, Oros Sweets Boiled sweets Jelly babies, wine gums, marshmallows Peppermints, vitamin C sweets Artificially sweetened Spreads Sugar, syrup, honey, molasses Jam, marmalade Dietetic jams Desserts Jelly, canned fruit, custard made with skimmed milk (0.4 g fat/​250 ml) Meringues without cream Dried fruit Artificially sweetened jelly Low-​calorie canned fruit Source data from Blom DJ, et al. (2010). Hypertriglyceridaemia: Aetiology, Complications and Management. JEMDSA, 15(1), 11–​17.

section 12  Metabolic disorders 2096 Acute pancreatitis Severe hypertriglyceridaemia is a well-​established trigger for acute pancreatitis. The pathogenesis of hypertriglyceridaemic acute pancreatitis remains ill-​understood. A  likely precipitating factor is sludging of the very large chylomicron particles in the microvasculature of the pancreas leading to leakage of pancre- atic enzymes into the circulation. This may lead to intravascular TG hydrolysis by lipase with subsequent bulk release of ‘toxic’ proinflammatory free fatty acids into the circulation; and activation of the proteolytic enzyme trypsin in the circulation may lead to pan- creatic autodigestion. Accurate measurement of serum amylase is challenging in the presence of lipidaemia and pancreatitis may be falsely ruled out when the amylase is not apparently elevated. In many patients, TGs are only measured several days after the onset of pancrea- titis and a prolonged period of nil per mouth. In such situations, hypertriglyceridaemia may have improved markedly and may then be erroneously excluded as a possible cause of pancreatitis. The treatment of hypertriglyceridaemic pancreatitis does not differ greatly from that of pancreatitis of any other cause. Metabolic disturbances should be sought and controlled. Should total paren- teral nutrition be necessary, it is important to avoid excess fat supply (e.g. Intralipid or Lipovenoes). Other therapeutic measures in order to correct the hypertriglyceridemia include the use of low molecular weight heparin and insulin. Apheresis or plasma exchange will rapidly, but transiently, lower plasma TGs, and may have a role if the high TGs are intransigent to other treatment approaches. In the early stages of recurrent acute pancreatitis and in pregnancy it may have value. There is, however, no evidence that patients treated with apheresis recover more rap- idly or have fewer pancreatitis-​associated complications, or have reduced mortality. Subsequently, severe restriction of dietary fat in- take is necessary. Pregnancy Hypertriglyceridaemia can be particularly troublesome in preg- nancy. It is usually worse in the third term, when physiological hormonal changes normally increase VLDL production. This will potentially markedly exacerbate an underlying genetic defect in peripheral lipolysis (see ‘Primary causes of defective lipolysis of triglyceride-​rich lipoproteins’). Gestational diabetes may exacerbate this, and needs appropriate control if necessary with metformin and insulin. A  low-​fat diet needs careful monitoring to avoid reduced nutrition. Omega-​3 fatty acids are highly efficacious as they lower VLDL secretion. Doses of fish oils well above the usually recommended daily maximum of 4 g can be given with apparent efficacy and safety; up to 12 g appears efficacious. Fibrates are also helpful in the third term, when teratogenicity is not a problem, and they are apparently safe and effective. In patients at risk of or with acute pancreatitis, apheresis may be indicated. An additional risk of severe hypertriglyceridaemia in pregnancy is still birth. Bariatric surgery Weight loss surgery is highly efficacious in the treatment of severe obesity, and is effective in the management of type 2 diabetes. If obesity and diabetes are present in a patient with hypertriglyceridaemia then it can be highly effective in reducing TG levels. In the decision as to whether to offer a patient bariatric surgery, the presence of dia- betes and hypertriglyceridaemia are important considerations. Treatment protocols and new drugs Treatment protocols, existing drugs, and new drugs for the treat- ment of dyslipidaemia are given in Tables 12.6.7, 12.6.8, 12.6.10, and 12.6.12 and Fig. 12.6.19. FURTHER READING Abifadel M, et al. (2012). Identification and characterization of new gain-​of-​function mutations in the PCSK9 gene responsible for autosomal dominant hypercholesterolemia. Atherosclerosis, 223, 394–​400. Adhyaru BB, et al. (2015). New cholesterol guidelines for the manage- ment of atherosclerotic cardiovascular disease risk: a comparison of the 2013 American College of Cardiology/​American Heart Association Cholesterol Guidelines with the 2014 National Lipid Association recommendations for patient-​centered management of dyslipidemia. Cardiol Clin, 33, 181–​96. Alphonse PAS, et al. (2016). Revisiting human cholesterol synthesis and absorption:  the reciprocity paradigm and its key regulators. Lipids, 51, 519–​36. Blom DJ, et  al. (2010). Hypertriglyceridaemia:  aetiology, complica- tions and management. JEMDSA, 15, 11–​17. Brunham LR, et al. (2006). Intestinal ABCA1 directly contributes to HDL biogenesis in vivo. J Clin Invest, 116, 1052–​62. Chowdhury R, et al. (2014). Association of dietary, circulating, and supplement fatty acids with coronary risk: a systematic review and meta-​analysis. Ann Intern Med, 160, 398–​406. Clarke R, et al. (2009). Genetic variants associated with Lp(a) lipopro- tein level and coronary disease. N Engl J Med, 361, 2518–​28. Cohen J, et al. (2005). Low LDL cholesterol in individuals of African descent resulting from frequent nonsense mutations in PCSK9. Nat Genet, 37, 161–​5. Cuchel M, et al. (2013). Efficacy and safety of a microsomal triglyceride transfer protein inhibitor in patients with homozygous familial hypercholesterolaemia:  a single-​arm, open-​label, phase 3 study. Lancet, 381, 40–​6. Danik JS, et  al. (2013). Lack of association between SLCO1B1 polymorphisms and clinical myalgia following rosuvastatin therapy. Am Heart J, 165, 1008–​14. Dayspring TD, et al. (2015). Biomarkers of cholesterol homeostasis in a clinical laboratory database sample comprising 667,718 patients. J Clin Lipidol, 9, 807–​16. De Oliveira e Silva ER, et al. (2000). Alcohol consumption raises HDL cholesterol levels by increasing the transport rate of apolipoproteins A-​I and A-​II. Circulation, 102, 2347–​52. De Souza RJ, et al. (2015). Intake of saturated and trans unsaturated fatty acids and risk of all cause mortality, cardiovascular disease, and type 2 diabetes:  systematic review and meta-​analysis of observa- tional studies. BMJ, 351, h3978. Fernandez ML, et al. (2005). Mechanisms by which dietary fatty acids modulate plasma lipids. J Nutr, 135, 2075–​8. France M, et al. (2014). Treatment of homozygous familial hyperchol- esterolemia. Clin Lipidol, 9, 101–​18. Gaudet D, et  al. (2014). Targeting APOC3 in the familial chylomicronemia syndrome. N Engl J Med, 371, 2200–​6.

12.6  Lipid disorders 2097 Geary RS, et  al. (2015). Clinical and preclinical pharmacokinetics and pharmacodynamics of Mipomersen (Kynamro):  a second-​ generation antisense oligonucleotide inhibitor of apolipoprotein B. Clin Pharmacokinet, 54, 133–​46. Guadagno PA, et al. (2015). Validation of a lipoprotein(a) particle con- centration assay by quantitative lipoprotein immunofixation elec- trophoresis. Clin Chim Acta, 439, 219–​24. Hirano K, et al. (1997). Cholesteryl ester transfer protein deficiency is extremely frequent in the Omagari Area of Japan. Marked hyperalphalipoproteinemia caused by CETP gene mutation is not associated with longevity. Arterioscler Thromb Vasc Biol, 17, 1053–​9. Holmes MV, et al. (2015). Mendelian randomization of blood lipids for coronary heart disease. Eur Heart J, 36, 539–​50. Iqbal J, et al. (2009). Intestinal lipid absorption. Am J Physiol Endocrinol Metab, 296, 1183–​94. JBS3 Board (2014). Joint British Societies’ consensus recommenda- tions for the prevention of cardiovascular disease (JBS3). Heart, 100, ii1–​67. Jones ML, et al. (2014). The human microbiome and bile acid metab- olism: dysbiosis, dysmetabolism, disease and intervention. Expert Opin Biol Ther, 14, 467–​82. Keene D, et al. (2014). Effect on cardiovascular risk of high density lipoprotein targeted drug treatments niacin, fibrates, and CETP in- hibitors:  meta-​analysis of randomised controlled trials including 117 411 patients. BMJ, 349, g4379. Kesäniemi YA, et  al. (1987). Intestinal cholesterol absorption effi- ciency in man is related to apoprotein E phenotype. J Clin Invest, 219, 578–​81. Lee J, et al. (2008). Hypertriglyceridemia-​induced pancreatitis created by oral estrogen and in vitro fertilization ovulation induction. J Clin Lipidol, 2, 63–​6. McClelland RL (2015). 10-​Year coronary heart disease risk prediction using coronary artery calcium and traditional risk factors. J Am Coll Cardiol, 66, 1643–​53. Meeusen JW (2015). Reliability of calculated low-​density lipoprotein cholesterol. Am J Cardiol, 116, 538–​40. Mora S, et al. (2009). Comparison of LDL cholesterol concentrations by Friedewald calculation and direct measurement in relation to cardiovascular events in 27331 women. Clin Chem, 55, 888–​94. Mora S, et al. (2014). Discordance of low-​density lipoprotein (LDL) cholesterol with alternative LDL-​related measures and future cor- onary events. Circulation, 129, 553–​61. Mozaffarian D, et  al. (2010). Effects on coronary heart disease of increasing polyunsaturated fat in place of saturated fat: a system- atic review and meta-​analysis of randomized controlled trials. PLoS Med, 7, e1000252. Navarese EP, et al. (2015). Effects of proprotein convertase Subtilisin/​ Kexin type 9 antibodies in adults with hypercholesterolemia:
a systematic review and meta-​analysis. Ann Intern Med, 163, 40–​51. Nayor M, et al. (2016). Recent update to the US Cholesterol Treatment Guidelines: a comparison with international guidelines. Circulation, 133, 1795–​806. Nilssona SK, et al. (2011). Apolipoprotein A-​V; a potent triglyceride reducer. Atherosclerosis, 219, 15–​21. Nissen SE, et al. (2006). Effect of very high-​intensity statin therapy on regression of coronary atherosclerosis. JAMA, 295, 1556–​65. Ohwada R, et al. (2006). Etiology of hypercholesterolemia in patients with anorexia nervosa. Int J Eat Disord, 39, 598–​601. Puri R, et al. (2015). Impact of statins on serial coronary calcification during atheroma progression and regression. J Am Coll Cardiol, 65, 1273–​82. Raal FJ, et al. (2015). Inhibition of PCSK9 with evolocumab in homo- zygous familial hypercholesterolaemia (TESLA Part B): a random- ised, double-​blind, placebo-​controlled trial. Lancet, 385, 341–​50. Rabar S, et al. (2014). Lipid modification and cardiovascular risk as- sessment for the primary and secondary prevention of cardiovas- cular disease: summary of updated NICE guidance. BMJ, 349, g4356. Reiner Z, et al. (2011). ESC/​EAS guidelines for the management of dyslipidaemias. Eur Heart J, 32, 1769–​818. Reiner Z, et al. (2014). Lysosomal acid lipase deficiency: an under-​ recognized cause of dyslipidaemia and liver dysfunction. Atherosclerosis, 235, 21–​30. Sinderman A, et al. (2002). Hypertriglyceridemic hyperapoB in type 2 diabetes. Diabetes Care, 25, 579–​82. Staels B, et al. (1998). Mechanism of action of fibrates on lipid and lipo- protein metabolism. Circulation, 98, 2088–​93. Stefanutti C, et al. (2013). Severe hypertriglyceridemia-​related acute pancreatitis. Ther Apher Dial, 17, 130–​7. Stefanutti C, et al. (2015). The lipid-​lowering effects of lomitapide are unaffected by adjunctive apheresis in patients with homozygous familial hypercholesterolaemia:  a post-​hoc analysis of a Phase 3, single-​arm, open-​label trial. Atherosclerosis, 240, 408–​14. Stone NJ, et al. (2014). Treatment of blood cholesterol to reduce athero- sclerotic cardiovascular disease risk in adults: synopsis of the 2013 ACC/​AHA cholesterol guideline. J Am Coll Cardiol, 63, 2889–​934. Surakka I, et al. (2015). The impact of low-​frequency and rare variants on lipid levels. Nat Genet, 47, 589–​97. Surendran RP, et  al. (2012). Mutations in LPL, APOC2, APOA5, GPIHBP1 and LMF1 in patients with severe hypertriglyceridaemia. J Intern Med, 272, 185–​96. Talmud PJ, et al. (2013). Use of low-​density lipoprotein cholesterol gene score to distinguish patients with polygenic and monogenic familial hypercholesterolaemia: a case-​control study. Lancet, 381, 1293–​301. Teslovich TM, et al. (2010). Biological, clinical and population rele- vance of 95 loci for blood lipids. Nature, 466, 707–​13. The Emerging Risk Factors Collaboration (2009). Major lipids, apolipoproteins, and risk of vascular disease. JAMA, 302, 1993–​2000. The TG and HDL Working Group of the Exome Sequencing Project (2014). Loss-​of-​function mutations in APOC3, triglycerides, and coronary disease. N Engl J Med, 371, 22–​31. U.S. Department of Health and Human Services and U.S. Department of Agriculture (2015). 2015–​2020 dietary guidelines for Americans, 8th edition. http://​health.gov/​dietaryguidelines/​2015/​guidelines/​. Voight BF, et al. (2006). Plasma HDL cholesterol and risk of myocardial infarction: a mendelian randomisation study. Lancet, 380, 572–​80. Wolff E, et al. (2011). Cholesterol-​absorber status modifies the LDL cholesterol–​lowering effect of a Mediterranean-​type diet in adults with moderate cardiovascular risk factors. J Nutr, 141, 1791–​8.

12.7 Trace metal disorders 2098

12.7 Trace metal disorders 2098

12.7.1 Hereditary haemochromatosis 2098 William J.

12.7.1 Hereditary haemochromatosis 2098 William J.H. Griffiths and Timothy M. Cox

CONTENTS 12.7.1 Hereditary haemochromatosis  2098 William J.H. Griffiths and Timothy M. Cox 12.7.2 Inherited diseases of copper metabolism:
Wilson’s disease and Menkes’ disease  2115 Michael L. Schilsky and Pramod K. Mistry 12.7.1  Hereditary haemochromatosis William J.H. Griffiths and Timothy M. Cox ESSENTIALS Hereditary haemochromatosis syndromes are inherited disorders whereby inappropriate absorption of iron by the small intestine leads to iron deposition in the viscera, endocrine organs, and other sites, causing structural injury and impaired function. The most common form is classical adult (HFE-​related) haemochromatosis, but other forms are recognized. Extended genetic platforms are increasingly used for specific diagnosis and noninvasive methods are increas- ingly used to evaluate hepatic damage. The mainstay of treatment is venesection although iron chelation therapy is an emerging oral alternative. Unravelling the molecular genetics of haemochroma- tosis is underpinning promising new therapies for disorders of iron homeostasis. Classical adult (HFE-​related) haemochromatosis Aetiology and pathogenesis—​inherited as a recessive trait and due to mutations in the major histocompatibility complex class I-​related HFE gene that appear to reduce liver production of hepcidin. The principal mutant allele of HFE, designated C282Y, is carried by ap- proximately 1 in 10 individuals of European ancestry, hence around 1 in 200 are homozygotes, usually with biochemical abnormalities of iron storage that may lead to full-​blown clinical haemochromatosis. Clinical features—​expression of disease may range from slight abnormalities of blood parameters that reflect iron metabolism to the established clinical syndrome of cutaneous pigmentation, car- diomyopathy, endocrine failure (especially diabetes mellitus and hypogonadism), arthritis, and pigment cirrhosis. Typical symptoms include fatigue, arthralgia, abdominal pain, and loss of libido. Diagnosis—​usually established by demonstrating abnormalities of iron metabolism, with fasting serum transferrin iron saturation above 55% in males and 45% in females along with elevated serum ferritin concentration. Molecular analysis of the HFE gene, in particular for homozygosity for the C282Y allele, is confirmatory. Management and prognosis—​this is directed to the removal of iron by phlebotomy, typically 500 ml of blood each week, until the serum ferritin concentration is reduced to within the low normal range, after which the frequency of phlebotomy is reduced. The oral iron che- lator deferasirox has shown promise as an alternative to venesection in phase II trials. Specific treatment may be required for estab- lished end-​organ failure. Patients with advanced fibrosis or cirrhosis should undergo 6-​monthly surveillance by ultrasonography and serum α-​fetoprotein estimation for early detection of hepatocellular carcinoma. The main causes of death in untreated patients are hepatocellular failure and carcinoma of the liver. Intervention prior to the onset of cirrhosis or diabetes avoids the deterioration in sur- vival associated with late presentation. Family members—​first-​degree relatives should be offered screening. Asymptomatic subjects in whom molecular analysis of the HFE or non-​HFE iron overload genes indicates a genetic predispos- ition to the disease require re-​evaluation by clinical and biochemical testing at intervals of between 2 and 5 years. Symptomatic subjects or those with significant iron indices (i.e. ferritin >750 µg/​litre) should be considered for immediate treatment. Introduction Pathological storage of iron The body contains about 4 g of iron, 3 g of which is complexed with haem to form haemoglobin, myoglobin, and the cytochromes. The nonhaem storage compartment, which consists of ferritin and its proteolytic degradation product haemosiderin, represents up to 0.5 g of elemental iron in adult women and slightly more than 1 g in adult men. Excess storage of body iron (iron overload) is asso- ciated with an increase in hepatic iron concentrations and of the surrogate biomarker, serum ferritin. Minimal iron storage occurs when more than 1.5 g of total body iron is present. This is reflected in a hepatic iron concentration of approximately 30 µmol/​g of tissue 12.7 Trace metal disorders

12.7.1  Hereditary haemochromatosis 2099 with a serum ferritin level of usually less than 250 µg/​litre. Moderate iron storage disease is reflected by a serum ferritin of approximately 500 µg/​litre. Under these circumstances, the hepatic iron concentra- tion rises to 100 µmol/​g. Severe iron storage disease (>5 g of storage iron) is shown by a hepatic iron concentration of over 200 µmol/​g liver tissue, with a serum ferritin level of at least 750 µg/​litre. Under these circumstances, tissue injury with impaired function is almost invariably present. Clinical subtypes of haemochromatosis Adult haemochromatosis The familiar form of haemochromatosis is the classical adult type, which typically presents in middle age and is usually expressed in men. The disorder is inherited as a recessive trait and is due to mu- tations in a gene, HFE, that maps to the short arm of chromosome 6 in close apposition to the HLA class I loci of the human major histo- compatibility complex (MHC). Expression of iron storage disease in individuals carrying mutations in the HFE gene is very variable and is influenced by several environmental and sexual factors, as well as emerging genetic modifiers. Mutant alleles of the HFE gene that predispose to adult-​type haemochromatosis are widespread and fre- quent in populations of northern European origin. There is evidence from haplotype analysis that a single mutation arose on an ances- tral chromosome 6 and spread throughout this population, prob- ably as a result of the migration of the Vikings from Scandinavia. The disease occurs throughout the world as a result of intermarriage but is at its highest frequency in France, Germany, Great Britain, Ireland, Northern Italy, Scandinavia, Spain, and Eastern Europe as far as European Russia. Colonization has led to its appearance in all populations of the United States of America and in Australasia, and for the same reason hereditary adult-​type haemochromatosis also occurs in South America. Classical adult-​type haemochromatosis (HFE-​related haemo- chromatosis) is a slowly progressive disease affecting the liver, endocrine system, heart, and joints; it is often only diagnosed when irreversible tissue injury has occurred. The condition predisposes to the development of primary carcinomas of the liver. A rare genetic form of adult haemochromatosis occurs in patients homozygous for mutations in the transferrin receptor 2 (TFR2). This form has been described mainly in southern Europeans and is termed type 3 haemochromatosis using the Online Mendelian Inheritance in Man (OMIM) classification (Table 12.7.1.1). The phenotype resem- bles HFE-​related haemochromatosis (type 1) although it is gener- ally more severe and presents at a younger age. The TFR2 protein is mainly expressed in the liver and has a lower affinity for iron uptake than the ubiquitous transferrin receptor. Latterly, a role for TFR2 in erythrocyte production has been suggested whereby it forms a component of the erythropoietin receptor complex and may act as an iron sensor. Identification of the protein responsible for iron transport across the basolateral surface of enterocytes provided a candidate for a re- cently recognized atypical form of haemochromatosis. The iron ex- porter in question has been termed ferroportin and appears to also control iron release from hepatocytes and, importantly, macrophages. Single missense mutations in the SLC40A1 mutations which en- code ferroportin are associated with a specific dominantly inherited phenotype. Haemochromatosis due to heterozygous SLC40A1 muta- tions has been coined ferroportin disease and also referred to as type 4 haemochromatosis. The disorder is typified by a raised ferritin level with normal or low transferrin saturation and a tendency for anaemia with poor venesection tolerance. Not restricted to white people, the condition is recognized in Asians and a unique and common poly- morphism (p.Q248H) in Southern African populations may con- tribute to the indigenous iron overload observed. Iron loading in type 4 haemochromatosis occurs predominantly within the reticuloendothelial system with splenic uptake visible on MRI (Fig. 12.7.1.1). On liver microscopy, Kupffer cells are iron-​ laden with relative sparing of hepatocytes. SLC40A1 mutations re- sult in iron trapping within macrophages and it has been proposed that the reduced availability of plasma iron either directly or via an ensuing anaemia drives increased intestinal absorption. As well as the phenotype described earlier, where spillover of iron into hepato- cytes is minimal and disease course benign, a second ‘nonclassical’ phenotype, less commonly observed, is characterized by an elevated transferrin saturation, hepatic parenchymal iron deposition, and liver disease. Juvenile haemochromatosis Since the identification of adult iron storage disease by several European physicians during the 19th century, a similar disease has Table 12.7.1.1  Inherited disorders of iron storage Disorder OMIM number Locus Gene/​protein Atransferrinaemia 209300 3q21 Transferrin Acaeruloplasminaemia 604290 3q23–​q25 Caeruloplasmin Haemochromatosis (type 1) (adult) 235200 6p21.3 HFE Haemochromatosis (type 2A) (juvenile) 608374 1q HJV/​haemojuvelin Haemochromatosis (type 2B) (juvenile) 606464 19q13.12 HAMP/​hepcidin Haemochromatosis (type 3) (adult) 604250 7q22 Transferrin receptor 2 Haemochromatosis (type 4) (adult, dominant) 606069 2q32 SLC40A1/ferroportin Haemochromatosis (type 5) (adult, dominant) 615517 11q12.3 FTH1 Haemochromatosis (neonatal) 231100 Unknown Unknowna a Autosomal recessively inherited in only a few families.

section 12  Metabolic disorders 2100 been recognized in children and young adults who may develop iron storage disease of a more severe character, now designated juvenile haemochromatosis (Fig. 12.7.1.2). This is defined as iron storage disease occurring before the age of 35 years. It evolves rapidly, typ- ically affects the heart and endocrine system, and causes infantilism and hypogonadism, as well as life-​threatening cardiac arrhythmias. Juvenile haemochromatosis is inherited as a very rare recessive trait in which there is an increased frequency of consanguinity among the parents of affected subjects. Juvenile haemochromatosis resem- bles the severe iron storage disease associated with the iron-​loading anaemias, such as β-​thalassaemia. Juvenile haemochromatosis af- fects males and females equally—​an observation that reflects the overwhelming nature of the iron homeostatic defect. Iron overload develops before the modifying effects of menstruation and dietary factors supervene. The genetic basis of juvenile haemochromatosis has been elu- cidated and revealed key proteins involved in iron metabolism. Most cases have been associated with mutations in the HJV gene on chromosome 1q, encoding the protein haemojuvelin; the homozygous mutation G320V accounts for approximately 50% of HJV-​associated or type 2A haemochromatosis. Haemojuvelin is ex- pressed predominantly by hepatocytes but also in cardiac and endo- crine tissues. The common mutations abrogate expression at the cell surface of hepatocytes where haemojuvelin may act as a coreceptor for bone morphogenetic protein as part of an intracellular signalling mechanism for synthesis of the peptide hepcidin. A smaller number of cases (type 2B haemochromatosis) are explained by mutations in the HAMP gene on chromosome 19 which codes directly for hepcidin. In mice with disruption of murine HAMP, or its promoter sequence, hepatic iron loading occurs. Conversely, overexpression of murine HAMP results in anaemia in keeping with hepcidin sup- pressing intestinal iron absorption. Indeed, HAMP overexpression overrides the effect of the C282Y mutation on dietary iron up- take and prevents haemochromatosis in HFE-​deficient mice. This finding supports the more severe phenotype observed in this form of juvenile disease compared with HFE-​related haemochromatosis. Hepcidin is thought to play a central role in iron homeostasis and current models are premised on hepcidin acting as a putative iron-​ regulatory hormone with effects on end organs, including the intes- tine and the monocyte/​macrophage system. Neonatal haemochromatosis Neonatal haemochromatosis is a newly identified syndrome of un- certain cause, characterized by congenital cirrhosis or fulminant hepatitis associated with the widespread deposition of iron in hep- atic and extrahepatic tissues. Approximately 100 cases of neonatal haemochromatosis have been reported. Neonatal haemochroma- tosis occurs in the context of maternal disease (including viral in- fection) and in the presence of maternal antinuclear factor, as a complication of metabolic disease in the fetus, and sporadically or recurrently, without overt cause, in siblings, including maternal half-​siblings. This latter observation indicates that conception by Fig. 12.7.1.1  Magnetic resonance imaging (T2 weighted) demonstrating iron overload in the liver and spleen of a patient with classical ferroportin (FPN) disease (top panel). This 48-​year-​old male presented with a serum ferritin of 3000 µg/​litre, transferrin saturation of 21%, and was heterozygous for the W158C mutation of the SLC40A1 gene. For comparison is a normal control film (bottom left panel) and a patient with HFE-​related haemochromatosis (bottom right panel) demonstrating low signal in the liver only as iron loading here is predominantly in hepatocytes rather than the reticuloendothelial system.

12.7.1  Hereditary haemochromatosis 2101 the use of sperm donors in women who have had a previously af- fected infant should not be recommended. Although infants with neonatal haemochromatosis die of liver disease shortly after birth, there are many instances where survival is associated with a com- plete recovery and thereafter normal growth and development with no signs of abnormal iron metabolism. Recently, it has been shown that the outcome of pregnancies at risk for neonatal haemo- chromatosis is improved by treatment of the mother with high-​dose intravenous infusions of pooled human immunoglobulin, thereby suggesting the operation of a humoral factor and a significant overlap with gestational alloimmune liver disease. However, the in- volvement of genetic determinants, possibly of paternal origin, in an alloimmune response has not been excluded. In other pedigrees, although neonatal haemochromatosis appears to have a clear her- editary basis, no predictive genetic test is yet available to inform the outcome of at-​risk pregnancies for this devastating disease. Prevalence and epidemiology Juvenile and neonatal haemochromatosis are rare disorders that occur sporadically, but hereditary adult haemochromatosis is widely disseminated and of global importance. Removal of toxic iron by re- peated venesection improves the outcome for adult haemochroma- tosis. If this treatment is instituted before irreversible tissue injury occurs, venesection may restore health and a normal life expectancy. (a) (b) (c) (d) Fig. 12.7.1.2  Juvenile cardiac haemochromatosis. Histological appearances of explanted heart from a 27-year-old woman (patient 1) who underwent orthotopic cardiac transplantation for refractory heart failure; hypogonadotrophic hypogonadism had been present for 8 years. Whole-mount sections of left ventricular wall stained with haematoxylin and eosin (a) or (b), Perls’s ferrocyanide reagent, show iron deposits principally affecting sub-epicardium and peripheral one-third of myocardium. High-powered microscopic views (×400: haemotoxylin and eosin, (c); Perls’s reagent, (d)) show widespread punctate aggregates of stored iron in cardiac myocytes and interstitial cells. Note the absence of inflammatory changes. Although iron was most abundant in the sub-epicardial zone, microscopic views show that degenerating myocytes containing iron deposits, some with hyperchromatic nuclei, were distributed throughout the myocardium. Reproduced from Kelly AL et al. (1998). Hereditary juvenile haemochromatosis: a genetically heterogeneous life-threatening iron-storage disease. QJM, 91, 607–618 with permission from Oxford University Press.

section 12  Metabolic disorders 2102 For these reasons, there has been much discussion about the early recognition of iron storage disease by the introduction of population-​based screening programmes, using genetic testing or phenotypic biochemical screening methods, that can be applied to communities at risk. In European populations, around 1 in 10 individuals carries one copy of an allele of the HFE gene that predisposes to iron storage disease, and between 1 in 100 and 1 in 400 people in these popula- tions are homozygotes or compound heterozygotes with biochem- ical abnormalities of iron storage that may lead to full-​blown clinical haemochromatosis. Thus, the mutant allele, designated C282Y of HFE, which is the principal determinant of iron storage disease, occurs at polymorphic frequency and is one of the most common genetic abnormalities leading to an autosomal recessive disease in populations of northern European origin. In European patients with iron storage disease due to hereditary haemochromatosis, the fre- quency of homozygosity for the C282Y HFE allele ranges from about 35% in southern Italy to more than 90% in the British Isles, including Ireland. In Australia, homozygosity for C282Y occurs in almost 100% of patients with hereditary haemochromatosis. However, as discussed later, although useful for diagnosis, homozygosity for the C282Y mu- tation of HFE is not tantamount to a diagnosis of established iron storage disease nor, therefore, of clinical haemochromatosis. Clinical expression of haemochromatosis is highly dependent on age and it is very rare for there to be detectable disease in adults below the age of 20 years. As clinical disease is much more common in men than women, it is likely to reflect environmental factors and the modification of disease expression due to blood loss associated with menstruation and the investment in pregnancies, as well as the comparatively reduced dietary complement of iron in women. Other environmental factors, particularly the consumption of alcohol, ap- pear to interact with predisposing genetic factors to induce the clin- ical expression of iron storage disease in C282Y homozygotes. Most patients with the disease develop symptoms at, or above, the age of 40 years. However, studies of iron metabolism by biochemical meas- urements or tissue biopsy may reveal early evidence of iron storage in the long presymptomatic phase of this condition. With greater awareness of the diverse clinical manifestations of adult type heredi- tary haemochromatosis, detection on the basis of early symptoms, for example arthralgia and fatigue, may be possible. Thus, there is a marked disparity in populations in which C282Y homozygosity is prevalent and the frequency with which symptomatic haemo- chromatosis is diagnosed. Phenotypic expression of disease For epidemiological purposes, since there is no internationally agreed case definition of haemochromatosis, caution is needed in interpreting claims that haemochromatosis is the most common inherited disorder affecting European peoples. Phenotypic expres- sion of the disease may range from the established clinical syn- drome (which includes cutaneous pigmentation, cardiomyopathy, endocrine failure—​especially diabetes mellitus and hypogonadism, arthritis, and pigment cirrhosis) to a slight abnormality of blood parameters that reflect iron loading—​elevated serum transferrin iron saturation and serum ferritin measurements. Such studies that are available to determine the penetrance and expressivity of the haemochromatosis gene have provided widely varying re- sults in different populations. In Australia, where the mean intake of iron in the diet appears to be much greater than in the average European population today, most middle-​aged male C282Y homo- zygotes appear to express at least one clinical manifestation of iron storage disease. Similarly, a study of homozygous relatives (princi- pally siblings) within pedigrees known to have haemochromatosis suggest that about one-​half of the men over 40 years of age, and about one in six of the women over 50 years of age, have at least one haemochromatosis-​related clinical disorder. This latter survey, conducted in the United States of America, suggests that an im- portant proportion of homozygous relatives of patients with estab- lished haemochromatosis, especially men, have conditions such as cirrhosis and arthropathy, as well as abnormalities of serum liver-​ related tests that are not detected by spontaneous clinical referral. Many reports of disease expression in haemochromatosis may, however, be questioned because of the prevalence of cosegregating genes within affected pedigrees, as well as early household environ- mental factors common to siblings that may predispose to disease expression. Studies in mice support this explanation, since it has been shown that several independent genetic determinants control the extent of iron loading observed in mouse models of iron storage disease generated by targeted disruption of the murine homologue of the HFE gene. In contrast, surveys conducted in outbred popu- lations, for example, in Jersey, show a great disparity between the predicted frequency of homozygosity for C282Y and the number of recorded cases with the disease attending local hospitals. These latter studies may reflect the underdiagnosis of haemochromatosis, and an inability to bring together the unitary clinical manifestations of the disease into a unifying diagnostic category. However, widely differing degrees of disease penetrance almost certainly account for the apparent shortfall of diagnosed cases in populations at risk. At present, no clear data in large unbiased population surveys are available to assess disease penetrance and the modifying effects of lifestyle factors such as alcohol, nutrition, and diet, as well as preg- nancy and menstruation, that are likely to influence the effects and rate of iron storage in human C282Y homozygotes. Mortality figures show that death is rarely attributed to hereditary haemochroma- tosis in populations at risk. This fact contrasts starkly with the well-​ established known complications of the full clinical syndrome, in which early death results from cirrhosis of the liver, hepatocellular carcinoma, endocrine failure, or cardiac complications. In a North American study of more than 41 000 individuals at- tending a health appraisal clinic, no evidence of an increased fre- quency of symptoms was identified in those genetically predisposed to iron storage disease. The only significant clinical history iden- tified in the at-​risk group was that of hepatitis or prior liver com- plaints. Only one of the 152 identified C282Y homozygotes had signs and symptoms of adult haemochromatosis. This provoca- tive report, indicating a very low clinical penetrance (<1%) of the haemochromatosis genotype in an unusual group of adults over the age of 26 years, raises important questions about the introduc- tion of mass population screening for this potentially treatable iron storage disease by genetic or even biochemical methods. However, the high prevalence of impotence, joint symptoms, chronic fatigue, and other complaints such as cardiac arrhythmias in the study group as a whole, raises disturbing questions about the valid application of this report to other populations. It is perhaps not surprising that in a group where, on average, more than 40% complained of a general limitation of their health and/​or joint symptoms, and in which more

12.7.1  Hereditary haemochromatosis 2103 than 35% of the male participants scored positively on symptom en- quiry about impotence, a significant contribution from predisposing haemochromatosis alleles could not be identified. Nonetheless, this large study raises key questions about the utility of screening for adult haemochromatosis as a genetic disease. To provide evidence for screening in haemochromatosis, other population surveys which address the morbidity and mortality of individuals harbouring disease alleles, have been attempted. For ex- ample, the effects of iron storage in C282Y homozygotes were re- ported in a comprehensive study of about 30 000 individuals aged between 40 and 69 years from Melbourne, Australia. Of 203 subjects found to be homozygous for the C282Y allele, ‘iron overload-​ related disease’ occurred in 28% of the men and 1.2% of the women. Longitudinal studies have shown that iron overload in C282Y homo- zygotes is not always progressive and indeed may recede in some cases. In the Melbourne study, follow-​up for an average of 11.4 years showed that the hazard ratio for death from any cause was 1.04 (con- fidence limits 0.67–​1.62) in C282Y homozygotes compared with subjects who did not harbour any copy of this mutant HFE allele. Not all individuals with mild iron loading require treatment and this clearly has implications for the introduction of mass population screening programmes for HFE-​related haemochromatosis. Aetiology, pathophysiology, and pathology Young patients with haemochromatosis absorb an increased amount of dietary iron in their upper intestine compared with normal con- trol subjects. In established iron storage disease, iron absorption continues at a rate that is inappropriate for the level of iron stores, as reflected by serum ferritin and tissue iron determinations. In the absence of an effective excretory pathway, the increased ab- sorption of iron by the intestine leads to a progressive accumulation of the metal in the parenchymal cells of the liver, heart, endocrine glands, and specialized type B synoviocytes. Excess iron accumu- lates in the pancreas where it is found in both acinar and endocrine cells of the islet, although there is a particular predisposition in the early phases of iron loading to the islet β-​cell. Iron also accumulates to toxic levels in the gonadotrophs of the anterior pituitary gland, leading to hypogonadotropic hypogonadism. Iron may accumulate in the adrenal gland, where it is concentrated particularly in those cells that secrete aldosterone, in the zona glomerulosa. Iron accumu- lates in the cardiac myocytes and conducting tissue of the heart, in the chief cells of the parathyroid, and in parenchymal cells throughout the body. The consequences of toxic iron storage include diabetes mellitus, cirrhosis of the liver, cardiomyopathy with or without con- duction defects, hypogonadism, arthritis with chondrocalcinosis, adrenocortical deficiency, and, rarely, hypoparathyroidism. Evidence for the intrinsic toxicity of iron in haemochromatosis is provided by the regression of the pathological changes following measures taken to reduce iron, for example, the use of iron chelators and removal of body iron by venesection. Venesection stimulates the mobilization and removal of iron from the storage compartment by increasing the demand for red cell production in the bone marrow. Mechanism of iron toxicity High concentrations of iron salts are toxic to cultured cells. The ad- ministration of iron chelates to experimental animals has induced diabetes with iron loading in the liver and pancreas, as well as the generation of (renal) carcinomas. Injections of iron salts induce local sarcomas in experimental animals, with evidence of species sus- ceptibility. In humans, sarcomas or carcinomas have arisen, albeit rarely, at sites of therapeutic injections of iron, and it is possible that the complications of silicosis and asbestos exposure result from the complement of iron associated with these particulates. A wealth of indirect but corroborative evidence indicates that the primary effect of excess free iron is to promote the formation of oxygen free rad- icals, which mediate the damage to cells and tissues that is observed in iron storage disease. In established haemochromatosis, the iron-​ binding capacity of plasma transferrin may be exceeded, so that a proportion of the iron present in the blood remains reactive as a low molecular weight species only loosely attached to plasma proteins. Nontransferrin iron in human plasma stimulates the peroxidation of unsaturated lipids and can form reactive complexes that react with DNA, thus suggesting a mechanism for genome toxicity and car- cinogenesis related to iron overload. Iron is highly electroreactive, and coupling of the Fenton and Haber–​Weiss reactions leads to the formation of hydroxyl radicals as a result of the catalytic interactions between superoxide and ferric ions. Tissues with significant iron storage show peroxidative injury in membrane lipid fractions. The lysosomal compartment appears to be particularly suscep- tible to iron-​mediated damage, since iron in the form of ferritin and its degradation product haemosiderin accumulates within lysosomes to form the particulate ferrugineous granules known as siderosomes. In haemochromatosis, there is an increased activity of lysosomal enzymes with biochemical evidence of increased lyso- somal fragility indicating disruption of the integrity of the lysosomal membrane by iron. These changes revert to normal when the tissue iron is removed by venesection or by the use of specific iron chela- tors. It seems likely that the electrochemical reactivity of iron, and its particular propensity to accelerate the formation of oxygen free radicals, mediate its injurious effects on cell membranes, and on the nuclear genome, leading to cancerous change. However, des- pite great advances in the understanding of free-​radical chemistry, the cause-​and-​effect relationship between iron storage and tissue injury is difficult to prove unequivocally. Nonetheless, much ex- perimental evidence points to the development of iron-​mediated peroxidative injury of cellular membranes including the lysosome, as well as iron-​mediated genotoxicity. Whatever their physiochem- ical basis might be, common mechanisms of iron toxicity clearly exist, since the pathological and clinical manifestations of all iron storage syndromes, including secondary haemochromatosis associ- ated with blood transfusion and the iron-​loading anaemias, are al- most identical. Iron absorption In established haemochromatosis, where the burden of iron may increase body iron stores by at least 10-​fold, measurements usu- ally show that iron absorption is within the normal range. Studies in young patients with rapidly progressive disease show a markedly increased absorption of iron. After depletion therapy, the rate of re- covery of iron stores is greatly enhanced for many years in patients with haemochromatosis, reflecting a persistent homeostatic abnor- mality in the retention of dietary iron. The daily absorption of be- tween 2 and 4 mg of iron over a period of 30 to 40 years accounts for the degree of iron loading that occurs at presentation in patients

section 12  Metabolic disorders 2104 with haemochromatosis, and compares with the normal absorption of 0.8 to 1.0 mg in men and in women, up to 2 mg daily. In effect, the abnormal absorption of iron represents a disturbed regulation of the final common pathway for the acquisition of iron from the environ- ment by the small intestinal mucosa. A report, describing the transplantation of intestine and liver from an HFE C282Y homozygote into a recipient without haemochroma- tosis, supports the small intestine as a key site of expression of the hereditary defect in adult haemochromatosis. The transplantation was associated with early iron overloading in the recipient, together with raised serum transferrin iron saturations—​a phenomenon not observed in recipients of hepatic allografts obtained from donors later found to be homozygous for C282Y. In addition to recent evi- dence for a principal role of the key hepatic iron regulator hepcidin, prior studies in vitro and in vivo have suggested that there is a quali- tative abnormality of the uptake and transfer of iron from the intes- tinal lumen in patients with hereditary haemochromatosis that may represent a local HFE-​dependent mechanism. Previous studies of mutant strains of mice with abnormalities of iron metabolism shed light on the iron-​absorption mechanism. The identification of a single gene encoding the divalent metal transporter protein, DMT1, which is expressed in the upper small intestine and cells of the erythron, provided a molecular understanding of the iron deficiency and the microcytic anaemia that occurs in the mk/​mk mouse strain. A single point mutation in the DMT1 gene interferes with the uptake of ferrous iron, since it disrupts the cognate transmembrane carrier protein mainly expressed in the mucosa of the proximal small in- testine, at the site of iron absorption, and in the erythroid precursor cells. Since in vitro studies of the expressed protein DMT1 show that it serves only as a carrier of divalent cations, and that interference with this pathway is sufficient to induce iron deficiency in a mam- malian species, ferrous iron uptake is probably the main pathway by which inorganic iron is acquired by the intestine. A variable, but often substantial, component of dietary iron is pre- sent in the organic form as haem. A full molecular understanding of the uptake and transfer pathways for the absorption of iron com- plexes to the porphyrias is also needed. Whole-​body studies show that the absorption of the radiolabelled iron moiety of haemoglobin is enhanced in patients with adult-​type haemochromatosis. Early studies in dogs have shown that, in the presence of proteolytic di- gestion products of globin, the haem complex is taken up intact by mucosal epithelial cells; free iron is then released by the action of intracellular haem oxygenases. The contribution of haemoglobin, myoglobin, and cytochromes to the iron overload in patients with haemochromatosis has not been quantified, but iron complexed to haem may well represent an important component of the total burden of body iron in symptomatic haemochromatosis. Recent identification of a putative transporter of haem iron on the brush border of mammalian duodenum is a key advance. Haem carrier protein 1 (HCP1) is up-​regulated in response to iron deficiency and hypoxia, but its contribution to the dysregulated absorption of iron in hereditary haemochromatosis is unclear. The discovery of DMT1 immediately indicated a possible role for this important protein in human haemochromatosis. Overexpression of DMT1 mRNA had been identified in the intestinal mucosa of patients homozygous for the C282Y mutation with her- editary haemochromatosis, as well as in mice with iron storage dis- ease due to targeted disruption of the HFE gene. Contemporaneous studies in experimental animals identified a cytochrome-​containing ferrireductase that is also localized to the intestinal brush-​border membrane; this reductase was cloned from murine intestine and its human homologue subsequently identified. Expression of mu- cosal ferrireductase is specific to the apical microvillous membrane of mammalian intestinal mucosa and appears to be induced in re- sponse to nutritional iron deficiency. Mucosal ferrireductase re- duces ferric irons derived from the diet in the lumen for delivery to the DMT1 carrier protein, the final divalent pathway for inorganic iron uptake by intestinal mucosa. The mRNA species encoding murine DMT1 exist in two isoforms, one of which contains an iron-​response element in its 3′ region, which would allow for the post-​transcriptional regulation of protein expression controlled by intracellular iron status. A similar translational control of transferrin receptor expression has been described with the 3′ iron-​response element in the mRNA encoding the human transferrin receptor. Since the isoform of DMT1 containing the iron-​response element is preferentially expressed in the duodenum, it seems likely that changes in intracellular iron status regulate the expression of this carrier protein in iron deficiency and haemochromatosis. Studies in Hfe knockout mice indicate that the functional expression of the DMT1 protein is enhanced in the murine model of haemochroma- tosis, leading to increased iron uptake across the brush-​border membrane of iron presented in the ferrous form. The action of rate-​ limiting ferrireductases at the brush-​border membrane functionally coupled to DMT1 activity appears to explain the enhanced isotopic uptake of ferric iron in this model of haemochromatosis. Delivery of iron from enterocytes to the systemic circulation is mediated by the basolaterally expressed membrane protein ferroportin. Ferroxidases including hephaestin, encoded on the X chromosome and mutated in the sex-​linked anaemic mouse sla, me- diate the transfer of iron across the intestinal mucosa in conjunction with ferroportin. It seems likely that, in hereditary haemochroma- tosis and physiological iron deficiency, post-​transcriptional control of carrier proteins responsible for the uptake and transfer of iron occurs in the absorptive epithelium on the tips of the intestinal villi. Thus, homeostatic mechanisms in the proximal intestine operate to bring about the coordinated transfer of iron presented in the in- testinal lumen specifically to meet body requirements. Although functional interactions of HFE molecules with the identified com- ponents of the absorptive pathway have yet to be clarified, the HFE protein probably influences iron status in intestinal stem cells within the crypt. By these means, the expression of key absorptive proteins such as DMT1 may be imprinted, thus influencing their subsequent functional activity during ascent up the villus. How the antimicro- bial peptide hepcidin interacts with this machinery is uncertain. It is proposed that hepcidin is a negative stimulator of intestinal iron absorption and that hepatic synthesis increases in response to iron overload and inflammation but decreases in the iron-​deficient state; hepcidin acts as a ligand for ferroportin whereby binding is thought to abrogate iron export into the circulation. Genetics and molecular biology The principal determinant of adult haemochromatosis has long been known to be tightly linked to the human MHC loci on the short arm of chromosome 6. In 1996, mutations in the HLA class I-​ linked haemochromatosis gene, HFE, were shown to predispose to the adult form of the disease. The most common mutation in

12.7.1  Hereditary haemochromatosis 2105 the nonclassical MHC class I HFE protein affects a key cysteine residue, which contributes to the formation of the conserved α-​ 3 helix that interacts cotranslationally with the β2-​microglobulin protein. This association is required for the cell surface expression of all class I MHC molecules. Most patients with haemochroma- tosis are thus homozygous for a cysteine to tyrosine mutation at codon 282 (C282Y) of the nascent HFE protein. An increased fre- quency of this mutation, in association with the more common H63D missense mutation, also occurs in adult haemochromatosis (Fig. 12.7.1.5). A minor variant, affecting the same region in the α1 helix, S65C, is also occasionally associated with the C282Y al- lele in compound heterozygotes with adult iron storage disease and indeed a number of uncommon pathogenic variants in HFE have since been reported. Apart from reducing cell surface expression of the mutant C282Y polypeptide, and thus the abundance of this protein within a population of cytoplasmic vesicles, a functional explanation for the qualitative abnormality of iron metabolism that characterizes haemochromatosis remains putative. Structural studies have pro- vided a molecular basis for an interaction between HFE and trans- ferrin receptor proteins. HFE may bind transferrin receptors, alter the affinity for the receptor to transferrin, and in turn affect the de- livery of transferrin-​bound iron into cells. This interaction may be more specific in the case of TFR2, expression of which is restricted to locations such as the hepatocyte membrane; here, an iron-​sensing role is postulated. The recently identified peptide hepcidin has become the focus of attention as the potential circulatory signal for body iron status with a key role in the disturbed iron homeostasis of hereditary haemo- chromatosis. Rather than the expected compensatory increase, hepcidin expression is paradoxically decreased and unresponsive in haemochromatosis caused by mutations in HFE, TFR2, and HJV. The proteins encoded by these genes are predominantly synthesized within the liver. Given that these types of haemochromatosis are qualitatively similar and differ mainly in severity, it has been ar- gued that these observations point to a common pathway in hep- atocytes which signals hepcidin production downstream. In support of this argument, the distribution of tissue iron in HFE-​related haemochromatosis can be altered after inducing experimental overexpression of hepcidin. Many recent studies suggest that hepcidin production in hep- atocytes is linked to a haemojuvelin-​dependent signalling pathway which is responsive to plasma iron saturations. Recent studies also suggest an effector function of hepcidin, whereby direct interaction with ferroportin at the cell surface results in its internalization and degradation with consequent reduced export of iron from macro- phages and enterocytes. SLC40A1 mutations appear either to abro- gate iron export function or interfere with the ability of membranous ferroportin to bind hepcidin; differential effects on cellular iron ex- port correlate with what appear to be the two discrete phenotypes observed in type 4 haemochromatosis. Serum hepcidin concentra- tions appear to be increased in ferroportin iron overload as opposed to other forms of haemochromatosis. In summary, a reduction in circulating hepcidin as a consequence of haemochromatosis gene defects, or directly as a result of HAMP gene mutations, appears to enhance plasma iron and subsequent tissue loading; it has been proposed that attenuated inhibition of macrophage and enterocyte ferroportin activity is responsible for this effect. An action of hepcidin as an effector molecule after hepatic signalling is central to the current favoured model of iron homeostasis (Fig. 12.7.1.6), but how the crypt programming hy- pothesis aligns with this model is not readily explained. Recent experiments are unravelling the detailed pathways that influence hepcidin synthesis within hepatocytes. A  number of cell surface interactions result in downstream signalling of hepcidin synthesis via SMAD complexes (Fig. 12.7.1.7). Experimental work is focusing on possible novel molecular therapies—​for example, interfering RNAs targeting TMPRSS6 have been shown to ameliorate iron over- load in mouse models of haemochromatosis by increasing hepcidin expression. Pathology of iron storage Heavy deposits of iron in the tissues are associated with fibrosis and cell death. Simple inspection reveals an overt rust-​like discol- ouration of the liver, spleen, pancreas, heart, and lymph nodes. The liver is usually enlarged and haemosiderin is found in all cell types with the formation of fibrous septa and hyperplastic nodules. These nodules, which may be the forerunners of hepatocellular carcin- omas, contain little stainable iron, unlike the adjacent parenchyma. The dominant site of iron deposition during the early phases is within hepatocytes, but soon iron loading may be observed in all cell types, including the lining cells of biliary canaliculi, Kupffer cells, and stellate cells (Figs. 12.7.1.3 and 12.7.1.4). Similarly, in the pancreas there is fibrosis and iron deposition in the acini, ducts, and islets of Langerhans. Staining with Perls’ reagent reveals marked haemosiderin deposition in the exocrine and endo- crine glands, including many cell types in the testes. Haemosiderin is also markedly increased in the chief cells of the parathyroid, the an- terior pituitary, the zona glomerulosa of the adrenal, and the thyroid. In the joints, there is loss of the intra-​articular space with chondrocalcinosis and deposits of haemosiderin in the synovium. Electron microscopy shows selective deposits of ferritin and haemosiderin within type B synoviocytes. Radiological examin- ation of the joints shows collapse of articular surfaces, subchondral cyst formation, and prominent formation of periarticular osteo- phytes. In the heart, pericardial constriction with fibrosis may Fig. 12.7.1.3  Low-​power needle-​biopsy appearance of liver specimen stained with haematoxylin and eosin from a 67-​year-​old man with adult haemochromatosis due to homozygosity for the HFE C282Y mutation. Note the large hyperplastic nodules and fibrosis.

section 12  Metabolic disorders 2106 occasionally be observed, but the principal abnormality is seen in the myocardium with degeneration and vacuolation of cardiac myocytes and intermyocyte fibrosis that involves conducting tissue in the septa. Surviving myocytes show eosinophilic degeneration and evidence of hypertrophy. Microscopical examination shows that, in established cases of haemochromatosis, all tissues except the choroid plexus are affected by the iron storage process. In the past, it was considered that transfusional and other types of sec- ondary iron storage disease predominantly affected the cells of the mononuclear macrophage system, such as the Kupffer cells of the liver, rather than the parenchymal cells. Iron deposits in the Tf LIVER MACROPHAGES GUT Fe Hepcidin HFE Haemojuvelin TfR2 Fpn Fpn Fig. 12.7.1.6  Molecular regulation of iron homeostasis. This is maintained by hepcidin, a peptide released by hepatocytes into the circulation under stimulatory control of a common pathway involving HFE, haemojuvelin, and transferrin receptor 2. Hepcidin normally inhibits iron export from enterocytes and macrophages via its interaction with ferroportin. Mutations in haemochromatosis genes reduce hepcidin expression and allow excess iron to enter the plasma compartment and bind to transferrin with consequent tissue iron loading. Fig. 12.7.1.4  High-​power micrograph of the liver biopsy specimen shown in Fig. 12.7.1.3 stained with Perls’ reagent. Note the extensive deposits of ferric iron in all cell types including Kupffer cells, cells lining small biliary radicles, and in a punctate distribution within parenchymal hepatocytes. Liver cells are hyperplastic. C282Y Plasma membrane Cytosol HOOC S S S S S S H63D NH2 2 3 1 2-Microglobulin ( 2m) Fig. 12.7.1.5  Diagram of nonclassical MHC class I-​like HFE molecule shown in juxtaposition with the β2-​microglobulin. The location of the two frequent amino acid substitutions (C282Y and H63D) that predispose to the development of adult haemochromatosis is indicated by the arrows. BMP-R SMAD 1/5/8 Haemojuvelin BMP6 Neogenin TFR2 HFE HAMP (hepcidin) transcription IL-6 TMPRSS6 sHJV TWSG1 NUCLEUS HEPATOCYTE MEMBRANE GDF15 IRON TFR1 ? Fig. 12.7.1.7  Molecular signalling of hepcidin synthesis in hepatocytes. Iron stimulates the binding of bone morphogenetic protein 6 (BMP6) to the membrane-​bound BMP receptor for which haemojuvelin is a coreceptor. This in turn activates the phosphorylation of SMAD proteins which translocate to the nucleus to promote transcription of the HAMP gene and synthesis of pro-​hepcidin. The BMP pathway is also activated by neogenin but inhibited by the TMPRSS6 gene product matriptase-​2, soluble haemojuvelin (sHJV), and twisted gastrulation 1 (TWSG1). Growth and differentiation factor 15 (GDF15) inhibits SMAD phosphorylation. HFE and transferrin receptor 2 (TFR2) act as an iron sensors and stimulate the BMP/​SMAD pathway through an as yet unknown mechanism, although recent studies suggest an interaction with the BMP type 1 receptor ALK3. IL-​6 stimulates hepcidin production during inflammation, independently of iron status. Figure adapted from Medicine, Vol 39/​10, W. J. H. Griffiths, Haemochromatosis, pp. 597–​601, Copyright (2011), with permission from Elsevier.

12.7.1  Hereditary haemochromatosis 2107 macrophage system may be less damaging than in other cell types, but it is difficult at present to relate evidence of iron-​mediated in- jury to its cellular distribution. Progressive tissue injury follows the long-​term cumulative toxicity of iron storage and its consequen- tial effects on organ structure and cellular function. A striking, but unexplained, feature of iron storage disease in the liver and other tissues is the absence of overt necrosis. In cancer cell biology, the novel term ‘ferroptosis’ describes iron-​dependent death distinct from apoptosis, necrosis, and autophagy. Quantitative aspects of iron storage disease Chemical determination of tissue iron content yields useful infor- mation about the severity of iron loading in haemochromatosis. In normal individuals, the total concentrations of liver iron do not ex- ceed 0.15% by dry weight, but in established haemochromatosis, the value is usually 1% or more. In severely affected patients with un- treated hereditary haemochromatosis or secondary haemochroma- tosis, the amount of iron may exceed 5% of the dry weight of tissue. The overall burden of body iron in patients with haemochromatosis is usually in excess of 5 g in hereditary disease, a figure that rises with age. Estimates indicate that the total burden in patients with advanced haemochromatosis can be as much as 40 to 60 g, most of this accumulating in the liver. The pancreas and other organs such as the lymph, thyroid, pituitary, and salivary glands typically show an increase of more than 10 times the normal iron content. Other methods of iron quantification, other than the crude estima- tion offered by serum ferritin concentrations, include histochemical iron grading using Perls’ reagent, MRI techniques, and quantitative phlebotomy. Clinical features Adult haemochromatosis The clinical features of adult haemochromatosis include skin pig- mentation. The pigment may be manifest as a generalized slate-​grey coloration, due principally to melanin, or localized bronzed pig- mentation particularly of the lower limbs, associated with iron de- posits in adnexal dermal structures, as well as melanin. Histological examination of the skin reveals increased melanocyte activity in conjunction with iron deposits, particularly in cutaneous sweat and apocrine glands. Increased skin pigmentation is a common, but not invariable, manifestation of haemochromatosis. It increases as the disease progresses and may be a late manifestation of the condition. Absence of pigmentation should consequently never be regarded as a contraindication to the diagnosis of iron storage disease. Iron storage disease invariably affects the liver, which is usually enlarged and may be cirrhotic, but portal hypertension and spleno- megaly are rare end-​stage features of haemochromatosis. The enlarged liver, even in the absence of cirrhosis, may contain single or multifocal hepatocellular carcinomas. Hypogonadism is often present and is typ- ically preceded by a long history of fatigue, sexual asthenia, and impo- tence, as well as premature menopause and loss of libido in women. In men, there is gynaecomastia, circumoral vertical skin wrinkling, and loss of body hair; the genitalia show premature atrophy. Many patients with haemochromatosis suffer from arthritis at an early phase in the illness and this may indeed be the sole manifestation of the condition for many years. The arthritis typic- ally affects the second and third metacarpophalangeal joints of the hands and feet (Fig. 12.7.1.8). These joints show painful swelling without obvious inflammatory changes. Distal interphalangeal joint disease is also recorded and is usually considered to be typical of osteoarthritis. Many joints, including the wrist, elbow, shoulder, and knee, may be affected and the changes in these joints are typ- ically associated with chondrocalcinosis that is detected radio- logically. The affected joints show loss of joint space, subchondral cysts, and, especially in the digits, prominent osteophyte formation (Fig. 12.7.1.9). Recent studies show that premature and disabling arthritis in the hip and other large joints is a characteristic feature of haemochromatosis. The symptoms of haemochromatosis are notoriously nonspecific and slow in their progression. Fatigue is often reported and may be a manifestation of hypogonadism and the onset of diabetes mellitus. Fig. 12.7.1.8  Arthropathy in a man with adult haemochromatosis forced to stop manual work because of painful arthritis, especially in the second and third metacarpophalangeal joints. Note the increased skin pigmentation. Fig. 12.7.1.9  Radiograph of hands in a 51-​year-​old woman with haemochromatotic arthropathy of the hands for many years. Note the loss of joint space, especially in metacarpophalangeal joints with subchondral cyst formation and osteophyte growth. Chondrocalcinosis is present in the ulnar fibrocartilage at the wrist.

section 12  Metabolic disorders 2108 Atrial fibrillation may be an early manifestation of cardiomyop- athy. Later, paroxysmal arrhythmias and cardiac failure supervene, leading to shortness of breath and fatigue. Occasional patients with haemochromatosis present with isolated features, such as abnormal liver-​related tests detected during a routine examination for health insurance, or with arthralgia and signs of arthropathy in association with diabetes, impaired libido, or sexual failure. Cardiomyopathy with heart failure or isolated arrhythmias is an unusual lone presen- tation of the disease. The differential diagnosis of haemochromatosis is very wide, but the presence of diabetes with abnormal liver function or hepatomegaly, or an association with endocrine failure or arthropathy, should prompt consideration of iron storage dis- ease. Likewise, the presence of seronegative polyarthropathy with pigmentation, hepatomegaly, or any of the associated endocrino- logical changes should initiate immediate testing for evidence of haemochromatosis. In young patients with hypogonadism or cardiomyopathy, iron storage disease should be considered. Juvenile haemochromatosis is often neglected by endocrinologists investigating young patients for infantilism or hypogonadotropic hypogonadism. The condi- tion may be responsible for cases of undiagnosed seronegative polyarthropathy. Haemochromatosis should be considered in any patient with signs and symptoms of chronic liver disease, including those with sustained mild elevation of serum transaminase activ- ities, particularly since the liver is affected early in the course of the iron overload. In fully established cases, skin pigmentation which may be either of a grey colour, as a result of increased melanin, or, especially on the shins, a yellow-​brown bronze colour. Pigmentation in association with diabetes with or without arthropathy and hepatomegaly almost always signifies established iron storage disease. Diagnosis It is critically important to establish a diagnosis of haemochroma- tosis at the earliest opportunity. There is strong evidence that if treat- ment to remove iron before established structural injury occurs, then tissue function and symptoms improve. Several studies indicate that removal of iron from patients diagnosed in the precirrhotic phase of adult haemochromatosis is associated with a normal or near-​normal life expectancy. Laboratory investigations In adult haemochromatosis, the diagnosis can be usually established by demonstrating abnormalities of iron metabolism (fasting serum transferrin saturation with iron >55% in males and 45% in females) together with a measurement of serum ferritin concentration that provides evidence of increased iron stores. Molecular analysis of the HFE gene for homozygosity for the common (C282Y) predisposing allele to the development of adult haemochromatosis may be very useful in patients of European ancestry. There is an increased fre- quency of compound heterozygotes for the C282Y/​H63D or, more rarely, C282Y/​S65C genotypes in patients with evidence of iron storage disease. For patients with elevated ferritin but normal or low transferrin saturation, ferroportin disease should be considered. Given the genetic variants that are now recognized as causes of haemochromatosis, it is clear that if any doubt exists as to the diag- nosis, or molecular analysis of the HFE gene or of non-​HFE iron overload genes fails to identify known pathogenic mutations, then tissue diagnosis is indicated. This is usually carried out by liver bi- opsy with histochemical determination, and preferably chemical quantification, of tissue iron content. Although a liver biopsy is asso- ciated with small but definable risks, it does offer a key opportunity for the evaluation of liver structure and of the injury consequent upon iron deposition. The finding of cirrhotic change carries with it a worse prognosis. Cirrhotic change is also a major predictor of the occurrence of hepatocellular carcinoma, which occurs rarely in noncirrhotic subjects with iron storage disease (Fig. 12.7.1.10). For C282Y homozygotes, liver biopsy may be reserved for those at risk of significant liver fibrosis. When serum aminotransferase values are normal, hepatomegaly is absent and the serum ferritin is below 1000 µg/​litre, the risk of significant fibrosis is negligible. This validated tool applies also to asymptomatic individuals identified through family screening or routine blood testing. More recently, transient elastography has been shown to reduce the requirement for biopsy in at-​risk patients and can accurately classify severe fi- brosis in around 60% of homozygotes with ferritin above 1000 µg/​ litre and raised transaminases. Serum iron-​saturation determinations, and particularly serum ferritin concentrations, may signify conditions other than iron storage disease. Serum ferritin is elevated in inflammatory states, in certain malignancies such as Hodgkin’s disease and in any condi- tion associated with significant necrosis of parenchymal liver cells. Under these circumstances liver biopsy is recommended, since it is most likely to provide a definitive diagnosis of iron storage dis- ease. Sometimes, however, liver biopsy is not possible, either be- cause the patient will not consent to it, or because of the presence of ascites and a bleeding disorder, especially thrombocytopenia. Under these circumstances, MRI of the liver can demonstrate iron storage if moderate or severe. A reduced signal on T2-​weighted Fig. 12.7.1.10  Adult haemochromatosis. Section of liver lobe after surgical resection to remove a primary hepatocellular carcinoma arising in an iron-​loaded but, unusually, noncirrhotic liver in this disorder. The patient, aged 62 years, had been partially treated by venesection but recently noticed increasing lethargy. A raised serum α-​fetoprotein concentration led to the diagnosis. Moderate histochemical evidence of iron storage was found in the nonmalignant tissue excised at surgery.

12.7.1  Hereditary haemochromatosis 2109 imaging correlates with significant iron deposition and a crude as- sessment of hepatic iron concentration is possible with dedicated data manipulation. If a liver biopsy is not possible and MRI of the liver does not reveal increased ferromagnetic signals indicative of iron storage, there are two further options: measurement of urinary iron excre- tion after parenteral administration of desferrioxamine, and, where the patient will tolerate it, quantitative phlebotomy. Injection of 500 mg of desferrioxamine intramuscularly in a patient with iron overload will usually induce the daily excretion of more than 2 mg of iron as the ferrioxamine complex in the urine. Ferrioxamine excretion may be increased in patients with haemolytic anaemia but, when elevated, is generally indicative of iron storage disease. Weekly phlebotomy of 500 ml will remove approximately 225 mg of iron, and thus provides a means of estimating the amount of iron removed from the storage compartment when undertaken to in- duce a mild hypochromic anaemia of approximately 10.5 to 11.0 g of haemoglobin/​dl or a serum ferritin concentration of less than 30 µg/​litre. Iron overload exists when the estimated iron removed by this method exceeds 1.5 g. Diagnosis in family members The diagnosis of haemochromatosis, whether it be of the adult or juvenile form, has immediate implications for that individual’s first-​ degree relatives. All forms of haemochromatosis have a strong her- editary basis and even some forms of neonatal haemochromatosis may, in some families, be inherited as an autosomal recessive trait. A dominant transmission pattern has been established in the case of type 4 haemochromatosis. Although the penetrance and expressivity of homozygosity for the various alleles that predispose to haemochromatosis is not yet established, the risks of the disease in first-​degree family members is sufficiently high to warrant systematic study. Clearly, the impli- cations for asymptomatic or undiagnosed relatives of the index case are potentially very large. Hence, considerable care and sensi- tivity are needed in the means of informing them about the condi- tion through the identified index case. In large families there may be formidable difficulties, so that the help of genetic counselling services, as well as formal assistance from physicians practised in medical genetics, may be needed. There can be little doubt, however, that at-​risk relatives should be offered the opportunity for further diagnostic and clinical evaluation in relation to iron storage disease. The condition is readily susceptible to iron depletion therapy in its early stages. Moreover, there may be additional considerations for patients who wish to make reproductive choices and who will need to be reassured that appropriate testing can be carried out on their future offspring. In HFE haemochromatosis, molecular analysis of the HFE gene may assist in assessing the risk of disease, particularly in asymptom- atic siblings. Phenotypic screening, however, is useful at the level of clinical evaluation for evidence of liver disease, hypogonadism, arthritis, pigmentation, and diabetes. Determining the biochemical phenotype first involves assay of the serum parameters of disordered iron metabolism. Since the serum parameters may be abnormal be- fore iron-​mediated tissue injury has occurred, liver biopsy should be considered particularly if the serum ferritin is greater than 1000 µg/​ ml or the serum transaminases are raised. In first-​degree relatives, in whom molecular analysis of the HFE or non-​HFE iron overload genes indicates a genetic predisposition to the disease, periodic re-​evaluation is needed by clinical and bio- chemical testing at intervals of not more than 5 years. In members of families affected by haemochromatosis due to mutations in the HFE or non-​HFE iron overload gene who were not found to carry the predisposing mutations and whose ferritin and iron parameters are normal, liver biopsy is not mandatory and the risk of the devel- opment of significant iron storage disease in less than 5 or 10 years is extremely low. In patients with no known pregenetic disposition and normal tissue biopsy findings, further follow-​up screening is not indicated. From the foregoing it can be seen that there is an urgent need to characterize the genotype–​phenotype relationship in both HFE and non-​HFE haemochromatosis. Unfortunately, no genetic locus has yet been identified for neonatal haemochromatosis, although this is a subject of continuing research. In at-​risk pregnancies, neo- natal haemochromatosis may be occasionally recognized by MRI during the third trimester, which may show increased iron signals in the fetal liver. After birth, biopsy of the oral mucosa on the gums or inner lip may reveal histological evidence of iron storage in minor salivary glands of affected infants. Environmental cofactors and disease expression Many patients with adult haemochromatosis give a history of exces- sive current or prior alcohol consumption. In the past, physicians have been tempted to attribute evidence of excess tissue iron in these individuals solely to the consumption of alcohol. In practice, how- ever, it appears that those individuals who have biopsy-​proven evi- dence of hepatic iron storage usually prove to carry two predisposing alleles of the HFE gene and therefore have true haemochromatosis. Although no clear predictors for the expression of disease in first-​ degree relatives at risk are available, disease expression is reduced in women of reproductive age. Most practising clinicians consider that age and alcohol consumption are the main identifiable environ- mental factors that contribute to disease expression in predisposed homozygotes. Other comorbid factors, including heritable factors, that may influence the expression of HFE mutations in homozy- gous subjects, include the presence of adult coeliac disease. There are few data that define the relationship between haemochromatosis and coeliac disease, but subclinical coeliac disease may ameliorate the long-​standing effects of iron loading in C282Y homozygotes. Cosegregation of haemochromatosis and coeliac disease has re- cently been reported in a large Swedish study. Discriminatory polymorphisms in genes which modulate oxidative stress and other fibrogenic cytokine responses have been postulated as influencing expression of haemochromatosis in C282Y homozygotes. Recently, discrete genome-​wide association studies in HFE haemochroma- tosis have identified the transferrin gene as a significant modifier of iron status and proprotein convertase subtilisin/​kexin type 7 (PCSK7) as a host risk factor for liver cirrhosis. The A736V TMPRSS6 polymorphism may also influence the clinical expression of HFE haemochromatosis as may a variant in the glyceronephosphate O-​ acyltransferase (GNPAT) gene. The identification, since HFE, of several genes associated with haemochromatosis has provided some insight into the pheno- typic variation of primary iron overload. The phenotype of type 4

section 12  Metabolic disorders 2110 haemochromatosis certainly appears distinct. It is now apparent, how- ever, that individuals with juvenile mutations may present later and with milder disease than described historically for juvenile haemo- chromatosis patients. Those with TFR2 mutations have occasionally presented young with a severe iron overload phenotype more rem- iniscent of juvenile haemochromatosis. Iron overload with varying severity has been accounted for by compound heterozygous forms of HFE and juvenile mutations, termed digenic inheritance. Moreover, C282Y homozygotes with the most severe iron overload may carry an additional juvenile mutation to account for increased disease ex- pression. The classical haemochromatosis phenotypes overlap and combinations of genetic alteration contribute to a spectrum of disease. Management Since it is the toxicity of iron that is responsible for the manifest- ations of all forms of haemochromatosis, treatment is directed to the removal of iron at the earliest possible stage. Venesection In adult and juvenile haemochromatosis, the preferred method of treatment is iron depletion by means of phlebotomy. This is best instituted by the removal of approximately 500 ml of venous blood each week by needle puncture of peripheral veins in the antecubital fossa. In young patients, it may be possible to increase the frequency of venesection to twice per week after several once-​weekly proced- ures. In elderly patients and those with hypoalbuminaemia as well as end-​organ failure and heart disease, the frequency of venesection should be commuted to within the rate tolerated. Coincidental in- flammatory disease may impede the erythropoietin-​mediated drive to haemopoiesis, and, particularly in the early phases of treatment, mild haemorrhagic anaemia may ensure. Thus, adjustments need to be made according to the early responses to venesection therapy, and regular monitoring of the haemoglobin concentration or haem- atocrit is advisable. Difficulties may arise in delivering this deceptively simple treat- ment as a result of poor organization of health service provision and of unavailability of suitable healthcare personnel to carry out the venesection procedure. Every practical effort should be made to ensure that the procedure is convenient for the patient, who is often a young or middle-​aged person in full-​time employment, and who may find regular access to the treatment centre problematic. In cold weather, or in patients with poor circulation or inconspicuous superficial venous access, the use of topical local anaesthetics such as lidocaine cream or even local diffusable preparations of glyceryl tri- nitrate, applied 30 to 60 min before the venesection procedure, may greatly improve venous access. Likewise, the simple technique of immersing the arm in warm water to improve peripheral blood flow may be critical for establishing confidence in treatment staff. Since patients with haemochromatosis usually harbour a large burden of iron, requiring repeated phlebotomy over a period of several years, every effort should be made to preserve the integrity of their per- ipheral veins. In the authors’ view, the use of a local anaesthetic is usually unwarranted since it involves further tissue invasion in the region of the antecubital fossa with needles. Moreover, repeated in- jections of the irritant fluid often lead to sclerosis around the venous access site. Where blood transfusion services can assume some, if not all, responsibility for the phlebotomy of haemochromatosis pa- tients, the inconvenience of hospital-​based services can be circum- vented and blood supplies can be enhanced safely. Not all patients require immediate treatment but it is vital to inter- vene if ferritin concentrations are above or approaching 1000 µg/​ litre. Some patients will be at lower risk of morbidity and mortality, typically those presenting with lower iron indices, in older age or those who are female. Such patients could be observed initially or indeed recommended to undergo blood donation with monitoring if otherwise eligible. Of note, since October 2012 the National Blood Service in England and Wales has accepted selected haemochroma- tosis patients for donation at up to 6-​week intervals. An Australian study in homozygotes with ferritin concentrations of between 300 and 1000 µg/​litre comparing iron reduction with sham treatment is currently underway. Duration of venesection therapy One 500 ml unit of peripheral blood contains approximately 225 mg of elemental iron. Thus most patients with established haemochromatosis will require weekly phlebotomy for a period of 1 to 2 years. The objective of this treatment is to restore serum ferritin concentrations to within the low normal range and, if pos- sible, to induce a mild iron deficiency anaemia of approximately 11.5 g haemoglobin/​dl. Having thus achieved a satisfactory deple- tion of body iron stores, interval maintenance phlebotomy, carried out according to ferritin measurements, four to six times per year is usually sufficient to maintain normal iron stores with a serum ferritin concentration less than 100 µg/​litre. Some authorities sug- gest that serum ferritin values below 30 µg/​litre should ideally be achieved. In patients with juvenile haemochromatosis, who have a higher than normal intestinal iron absorption, more frequent phlebotomy may be needed to maintain a healthy iron balance. Proton pump inhibitors appear to significantly reduce the need for venesection, by reducing iron absorption through alkalization of the stomach, although currently are not recommended for this sole purpose. Iron chelation therapy Alternative methods of iron removal are needed for patients with severe clinical manifestations of haemochromatosis, such as life-​ threatening cardiac arrhythmias and those with severe liver dis- ease and hypoalbuminaemia, who are incapable of withstanding frequent phlebotomy. The preferred alternative involves chelation therapy with the parenteral agent desferrioxamine. As indicated in Chapter 22.6.4, the subcutaneous administration of desferrioxamine brings about the removal of a maximum of 20 to 25 mg of iron daily and is thus generally less efficient than vigorous weekly phlebotomy. However, desferrioxamine may gain access to cellular pools of iron that are important in the pathogenesis of tissue injury in established iron storage disease, and therefore may offer particular benefit in pa- tients critically ill with arrhythmias due to haemochromatotic car- diomyopathy. Although the nature of this so-​called chelatable iron pool is unknown, there is strong circumstantial evidence that its depletion by means of intravenous desferrioxamine treatment may reverse the life-​threatening consequences of terminal iron storage disease in patients with haemochromatosis. Moreover, the removal

12.7.1  Hereditary haemochromatosis 2111 of 140 mg of chelatable iron per week represents about two-​thirds of the amount that can be removed by weekly phlebotomy. A biological advantage may also be gained by therapeutic access to a reactive, low molecular weight, chelatable fraction responsible for the injurious effects of cellular iron overload. Parenteral desferrioxamine may be given intravenously for life-​ threatening cardiac disease, as described in Chapter 22.6.4, or, in the nonemergent situation, by subcutaneous infusion using portable infusion pumps for 12 to 14 h, five or six times per week. It must be stressed, however, that chelation therapy is not the preferred option for the treatment of established haemochromatosis and should be re- stricted to those patients unable to tolerate phlebotomy as a result of anaemia or hypoalbuminaemia, or in whom life-​threatening cardio- myopathy or liver disease is present. Newer oral iron chelators with promising safety profiles are becoming established for secondary iron overload. One such chelator, deferasirox, has been shown to be efficacious in an early phase study in genetic haemochromatosis, re- peated in a further phase II study published very recently. General measures Attention should be given in patients with haemochromatosis to the diagnosis and treatment of end-​organ failure. This particu- larly applies to the management of diabetes mellitus by diet and insulin where necessary, as well as hormone replacement therapy for hypogonadism (see Chapter  13.6.2). In men, intramuscular depot injections of testosterone enantate (250 mg every 2–​3 weeks) are recommended to improve libido and inhibit the development of premature osteoporosis. Similarly, conventional sex hormone replacement therapy should be used in women with premature gonadal failure as a result of haemochromatosis. Cardiac failure in patients with haemochromatosis due to cardiomyopathy and hepatic failure consequential upon pigmentary cirrhosis should be treated by standard methods. Organ transplantation may be used successfully, but correction of systemic iron overload should be undertaken as soon as practicable to restore normal function in all organ systems. Rarely, end-​organ hormone deficiencies re- sult from thyroid infiltration and parathyroid and adrenocortical disease. These deficiencies should be vigorously sought for in the clinical evaluation of the patient at presentation. The appearance of lethargy, faintness due to postural hypotension, or symptomatic hypocalcaemia demands immediate investigation and institution of appropriate replacement therapy. Patients with cirrhosis should undergo 6-​monthly surveillance by ultrasonographic examination and α-​fetoprotein estimation for early detection of hepatocellular carcinoma. Prognosis The main causes of death in untreated patients with haemo- chromatosis are hepatocellular failure, primary carcinoma of the liver (including hepatocellular carcinoma), and, rarely, cholangiocarcinomas. Cardiac failure due to haemochromatotic cardiomyopathy and untreated diabetes also contribute to death. Although not categorically proven, evidence from retrospective surveys suggest that life expectancy is improved by removing iron from patients with haemochromatosis of whatever cause and the subsequent maintenance of normal iron homeostasis. Most patients experience an improvement in well-​being on iron depletion therapy and, during its early phases, there is evidence that hypogonadotropic hypogonadism may improve with this therapy. Similarly, the mani- festations of cardiomyopathy with intractable cardiac failure or tachyarrhythmias can improve after the removal of iron. The cirrhosis of haemochromatosis appears not to be reversible, although the earlier precirrhotic manifestations of hepatic disease improve greatly on the removal of iron with an apparent restoration of normal life expectancy. Indeed, there is mounting evidence that hepatic fibrosis, short of cirrhosis, can reverse following iron deple- tion. In all patients, there is at least a twofold increase in the survival rate at 5 years from the point of diagnosis with the introduction of phlebotomy. In patients studied during the 1950s and 1960s, the 5-​year survival rate improved from 18% to more than 65% in all haemochromatosis subjects treated. In patients diagnosed with haemochromatosis but without cir- rhosis, iron depletion therapy is associated with a near normal or normal life expectancy compared with a sex-​ and age-​matched con- trol cohort derived from the same population. It is notable, however, that the indolent nature of this storage disorder and the long-​term survival of patients who are affected by it has, so far, rendered long-​ term controlled studies of the effects of phlebotomy on eventual out- come almost impossible to achieve. However, a wealth of evidence, based on the understanding of the pathogenesis and documented responses to iron depletion in individual patient cohorts, indicates that early removal of iron is highly desirable—​indeed, it may be de- cisive in determining a good outcome from all forms of human iron storage disease, including all subtypes of hereditary haemochroma- tosis so far established. Hepatocellular carcinoma occurs mostly in patients with iron storage disease who have established cirrhosis and the risk appears to persist despite removal of iron. Although hepatocellular carcinoma and cholangiocarcinoma have been reported in noncirrhotic pa- tients with haemochromatosis, these are rare phenomena. Systematic ultrasonographic surveillance is vital if liver cancer is to be detected at a stage where potentially curative treatment can be offered. Since all the evidence suggests that patients with haemochromatosis are more likely to have diabetes mellitus and other manifestations of the disease, every encouragement should be given to the prompt diag- nosis of the condition and early institution of iron depletion therapy. Increasingly, it has been recognized that the arthropathy of haemo- chromatosis can be disabling, whether or not it is associated simply with joint pain (arthralgia) or progressive and noninflammatory joint destruction. The disease is associated with a loss of cartilage and, in many large joints, chondrocalcinosis. Although the re- sponse of the arthropathy to iron depletion therapy is controver- sial, the weight of observation indicates that, once established, the arthropathy of haemochromatosis progresses independently of body iron status and of iron depletion treatment. It seems intrin- sically likely that effective removal of excess body iron stores before the development of joint symptoms will prevent their onset and pro- gression. However, at present only cross-​sectional data are available to support this contention. In summary, observations in adult haemochromatosis suggest that once the disease is established in association with cirrhosis or diabetes mellitus, it diminishes life expectancy. In fact, a more recent

section 12  Metabolic disorders 2112 study demonstrated that among treated C282Y homozygotes those with serum ferritin levels greater than 1000 µg/​litre at diagnosis had a fivefold relative risk of death. The prognosis for cardiomyopathy in juvenile haemochromatosis is very poor but it may be improved by early diagnosis and the early institution of vigorous iron deple- tion therapy. In several cases, the outcome has been improved by allogeneic cardiac transplantation. In adult patients with established pigment cirrhosis, hepatic transplantation has been undertaken and, provided the other systemic manifestations of haemochroma- tosis have been adequately treated, the procedure is associated with a good overall prognosis. Prevention and control The importance of early recognition and the institution of iron de- pletion therapy in all forms of haemochromatosis cannot be over- emphasized. Molecular analysis of the HFE gene, together with biochemical characterization using serum transferrin iron satur- ation estimations and serum ferritin concentrations, has the power to assist greatly in the detection of presymptomatic first-​degree rela- tives of patients with haemochromatosis. In relation to whole populations in which mutations in the HFE gene are frequent, the health implications based on mass screening remain contentious. Superficially, adult hereditary haemochroma- tosis due to mutations in the HFE gene appears to be an ideal con- dition for DNA-​based mass population screening. The condition is attributable to a single gene, and a single mutation of diagnostic significance is prevalent (gene frequency 5–​10%). Disease-​related mutations in HFE (especially C282Y) are easily tested for by means of techniques based on the polymerase chain reaction. At the same time, HFE-​mediated haemochromatosis has a long incubation period without symptoms, and all the evidence suggests that the in- stitution of treatment for presymptomatic disease is cheap, simple, and effective. On the other hand, genetic identification of at-​risk individuals is associated with problems of stigmatization, increased anxiety, and potential life insurance weighting, all of which are familiar aspects in well-​rehearsed debates about genetic testing in the gen- eral population. These aspects must be considered, together with the age-​related penetrance of the homozygous state for HFE C282Y variants and, as yet, unknown combined genetic and environmental influences on disease expression. Uncertainty as to the significance of these factors has held back the introduction of mass population screening by DNA-​based methods. In light of the present state of knowledge, it is clear that homozygosity for the C282Y allele of HFE cannot be considered to be tantamount to a diagnosis of hereditary haemochromatosis. More information is needed from outbred populations, rather than from homozygotes identified as a result of screening family members of index cases having full-​blown clinical disease. Family studies provide a false measure of disease expressivity, presumably as a result of shared environments and of the cosegregation of po- tential disease-​modifying genes within defined pedigrees. Finally, it must be emphasized that difficulties also occur for the evaluation of the burden of haemochromatosis in the population at large. Although there are definitions of iron storage disease that reflect the abnormal biochemical genotype, the manifestations of the clinical disease are variable and protean. Moreover, as pointed out earlier, no internationally agreed case definition of haemochromatosis exists, which creates additional difficulties for the introduction of public health measures and appropriate policy review of nationwide screening procedures. Future directions Although startling progress has been made in the discovery of many components that serve to regulate iron homeostasis in hu- mans, more information is needed before a full molecular under- standing of the mechanisms of iron homeostasis can be achieved. The genetic basis of some neonatal and further variant forms of adult haemochromatosis has yet to be fully explored. The inter- actions between iron regulatory molecules on the hepatocyte membrane are not fully resolved but this interplay and the asso- ciated downstream signalling pathway for hepcidin appear key to body iron homeostasis—​thus a promising target for future therapeutic intervention. A challenging task will be the detailed understanding of how environmental cofactors determine the ex- pression of iron storage disease in genetically predisposed indi- viduals. Alcohol is a long-​standing candidate, but the mechanism by which it leads to increased delivery of toxic iron to the tissues is, at present, poorly understood. Recognizing genetic modifiers of disease expression may, in future, inform natural history and treatment decisions in asymptomatic individuals at risk from iron storage disease. Greater understanding of these issues and of pene- trance in particular populations will determine local screening practices for disease prevention. Newly identified iron storage diseases By general agreement, the term haemochromatosis is used to de- scribe systemic syndromes of pathological iron storage that affect many tissues and disturb the function of diverse organ systems. Conversely, several distinct clinical syndromes of local iron tox- icity have been identified, especially in the eye and brain. Although these syndromes are individually rare, they are important because they are potentially accessible to measures that reduce cellular free iron (e.g. metal chelation, mentioned earlier), and because they demonstrate the central importance of metabolic iron in selected tissues. A fuller understanding of these disorders, and the cognate cell metabolic pathways they affect, may well shed light on ill-​ understood aspects of tissue iron physiology. Additional informa- tion is available by reference to the OMIM website at http://​www. ncbi.nlm.nih.gov/​omim. Hereditary hyperferritinaemia cataract syndrome (OMIM 600886) The sole clinical manifestation of this condition is of congenital bilateral ferrugineous nuclear cataracts due to the disposition of excess ferritin light chain polypeptide in the ocular lenses. The serum ferritin concentrations are moderately elevated but no evi- dence of systemic iron storage is found. The disorder is caused

12.7.1  Hereditary haemochromatosis 2113 by mutations in the 5′ noncoding iron-​response element of the ferritin light-​chain (FTL) gene that leads to unregulated transla- tional overexpression of ferritin light chains. These polypeptides accumulate in the lenses and disturb their tissue organization and refractile properties. The hyperferritinaemia cataract syndrome is, as expected for an overexpression disease, inherited as a dom- inant trait. Measurement of serum ferritin concentrations may identify at-​risk family members. The gene encoding ferritin light chain polypeptide maps to chromosome 19q3.3-​qter. Latterly, a syndrome without cataracts has been identified due to mutations in exon 1 of the FTL gene—​this forms an important differential in individuals with unexplained hyperferritinaemia with normal transferrin saturation. Adult-​onset basal ganglia disease (OMIM 606159) A single pedigree has been identified with a dominantly inherited disorder showing features of late-​onset extrapyramidal dysfunction resembling parkinsonism or Huntington’s disease. Imaging and aut- opsy studies revealed cavitation of the basal ganglia with deposition of iron and ferritin protein in adjacent tissue, especially in the pu- tamen and the globus pallidus. The macroscopic appearances showed widespread reddish discolouration of affected tissues. This disorder was mapped to chromosome 19q13.3 and a single mutation, a point insertion of a single adenine at nucleotide 461, was identified in exon 4 of the FTL gene. The mutation is predicted to disrupt the C-​ terminal sequence of the ferritin light-​chain molecule and disturb the iron-​binding core of the hetero-​ or homomeric protein. Serum ferritin concentrations were found to be abnormally low in affected heterozygotes. Although this disorder has so far only been identified in a single large pedigree, it further illustrates the importance of fer- ritin in tissue iron metabolism and, especially, in selective regions of the brain. This disorder has been termed a ‘neuroferritinopathy’ and may be the first of several diseases affecting cellular iron pathways in iron-​rich brain tissue. Acaeruloplasminaemia with iron deposition (haemosiderosis) in basal ganglia (OMIM 277900) This disorder is associated with mild systemic iron deposition and deficiency of the plasma copper-​binding protein, caeruloplasmin. Caeruloplasmin has long been known to possess ferroxidase activity and the ability to enhance the mobilization and delivery of iron to and from macrophages and hepatocytes. It promotes iron loading of intact ferritin micelles. Acaeruloplasminaemia, due to mutations in the gene encoding caeruloplasmin on chromosome 3q21–​24, is an autosomal recessive trait. The deficiency is associated with diabetes mellitus, dementia, and extrapyramidal features including parkin- sonism, with choreoathetosis as well as cerebellar ataxia. MRI shows altered signals in the basal ganglia, and retinal degeneration may be apparent by fundoscopy. Excess systemic iron is demonstrable by examination of liver tissue and the serum ferritin concentra- tion is moderately elevated; however, low serum iron transferrin saturations with hypochromic microcytic anaemia, reminiscent of copper deficiency, are usually present. Infusions of plasma or purified caeruloplasmin may correct the systemic abnormalities of iron metabolism, but probably do not influence the dementia or the other neurological deficits, at least once these are established. The role of caeruloplasmin replacement or indeed parenteral chelation therapy with desferrioxamine or trientine, especially in the early evolution of the neurological syn- drome, has not yet been established. The interplay between copper and iron metabolism is well illustrated by this severely disabling illness. Acaeruloplasminaemia illustrates the particular sensitivity of the basal ganglia to disturbances of iron metabolism. In this con- text, it is notable that caeruloplasmin expression is abundant in glia in the brain microvasculature juxtaposed to the pigment-​containing dopaminergic neurones of the substantia nigra and inner layer of the retina. Hallervorden–​Spatz disease: pantothenate kinase-​ associated neurodegeneration (OMIM 234200) This disease has been familiar to neurologists and neuropatholo- gists since its original description by two, now discredited, German neuroscientists of the Nazi period. The clinical features indicate basal ganglia disease and dementia with retinal degeneration leading to optic atrophy. The disorder often presents with club foot deformity in children and adolescents; extrapyramidal rigidity pre- ceded by choreoathetosis usually follows rapidly. Dementia, optic atrophy, and generalized seizures occur in the later stages, and death usually ensues by the age of 30 years. Although late-​onset forms of the disease are known, a striking feature is the presence of iron pigment in the basal ganglia and substantia nigra, now easily recognized by MRI. The heredofamilial nature of this syndrome has been known since its first description. Hallervorden–​Spatz dis- ease is now known to be an autosomal recessive trait due to mu- tations in the pantothenate kinase 2 gene (PANK2) that maps to chromosome 20p13. Pantothenate kinase 2 is abundant in the retina and target regions of the brain and regulates the formation of coenzyme A. Deficiency of pantothenate kinase 2 would deplete sensitive neural tissues with a high metabolic rate of coenzyme A; the defect may also lead to a consequential accumulation of cysteine, which normally condenses with the enzyme product, phosphopantothenate. In the presence of high concentrations of free iron, excess cysteine may accelerate the formation of cytotoxic oxygen free radicals. For some years, cysteine accumulation has been independently observed in the iron-​rich nigrostriatal regions of the brain affected by this disorder. Identification of PANK2 mutations offers the hope of improved diag- nosis of this neurodegenerative disorder, and, more importantly, the prospect of specific therapy using supplementation to enhance local coenzyme A  activity and phosphopantothenate concentra- tions in affected neural tissue. Latterly, the term ‘neurodegeneration with brain iron accumulation’ (NBIA) has been coined to encom- pass several inherited neurological disorders with basal ganglia involvement—​nine genes have thus far been implicated including PANK2 and FTL. The accelerated interest in this field is leading to trials of iron chelation in Parkinson’s disease. Further practical information Many patients’ associations and societies exist to serve the needs of patients in their respective countries. In the United Kingdom, useful information can be obtained from The Haemochromatosis Society, Haemochromatosis UK, PO Box 6356, Rugby, CV21 9PA.

section 12  Metabolic disorders 2114 Haemochromatosis UK is a working name of The Haemo­ chromatosis Society (a registered charity). Office: office@huk.org.uk; Helpline:helpline@huk.org.uk; Telephone: +44 (0)3030401102. The society’s website (http://​www.haemochromatosis.org.uk) in- cludes links to similar societies in other parts of the world. FURTHER READING Adams P, et al. (2009). Screening for iron overload: lessons from the HEmochromatosis and IRon Overload Screening (HEIRS) Study. Can J Gastroenterol, 23, 769–​72. Adams PC, Speechley M, Kertesz AE (1991). Long-​term survival analysis in hereditary haemochromatosis. Gastroenterology, 101, 368–​72. Allen KJ, et al. (2008). Iron-​overload-​related disease in HFE hereditary hemochromatosis. N Engl J Med, 358, 221–​30. Andersen RV, et al. (2004). Hemochromatosis mutations in the general population: iron overload progression rate. Blood, 103, 2914–​19. Barton JC, et al. (2012). Increased risk of death from iron overload among 422 treated probands with HFE hemochromatosis and serum levels of ferritin greater than 1000 μg/​L at diagnosis. Clin Gastroenterol Hepatol, 10, 412–​16. Bulaj ZJ, et al. (2000). Disease-​related conditions in relatives of patients with hemochromatosis. N Engl J Med, 343, 1529–​35. Camaschella C, et al. (2000). The gene TfR2 is mutated in a new type of haemochromatosis mapping to 7q22. Nat Genet, 25, 14–​15. Cancado R, et al. (2015). Deferasirox in patients with iron overload secondary to hereditary hemochromatosis: results of a 1-​yr phase 2 study. Eur J Haematol, 95, 545–​50. Cullis JO, et  al. (2018). Investigation and management of a raised serum ferritin. Br J Haematol, 181, 331–​40. De Gobbi M, et al. (2002). Natural history of juvenile haemochroma- tosis. Br J Haematol, 117, 973–​99. De Tayrac M, et al. (2015). Genome-​wide association study identifies TF as a significant modifier gene of iron metabolism in HFE hemo- chromatosis. J Hepatol, 62, 664–​72. Dusek P, Schneider SA (2012). Neurodegeneration with brain iron ac- cumulation. Curr Opin Neurol, 25, 499–​506. European Association for the Study of the Liver (2010). EASL clinical practice guidelines for HFE hemochromatosis. J Hepatol, 53, 3–​22. Falize L, et al. (2006). Reversibility of hepatic fibrosis in treated genetic hemochromatosis: a study of 36 cases. Hepatology, 44, 472–​7. Fargion S, et al. (1992). Survival and prognostic factors in 212 Italian patients with genetic haemochromatosis. Hepatology, 15, 655–​9. Finch SC, Finch CA (1955). Idiopathic hemochromatosis, an iron storage disease. Iron metabolism in hemochromatosis. Medicine (Baltimore), 34, 381–​430. Fleming ME, et al. (1999). Mechanism of increased iron absorption in murine model of hereditary haemochromatosis: increased duodenal expression of the iron transporter, DMT-​1. Proc Natl Acad Sci USA, 96, 3143–​8. Gao J, et al. (2009). Interaction of the hereditary hemochromatosis protein HFE with transferrin receptor 2 is required for transferrin-​ induced hepcidin expression. Cell Metabolism, 9, 217–​27. Kellerher T, et al. (2004). Increased DMT1 but not IREG1 or HFE mRNA following iron depletion therapy in hereditary haemo- chromatosis. Gut, 53, 1174–​9. Kelly AL, et al. (1998). Hereditary juvenile haemochromatosis: a gen- etically heterogeneous life-threatening iron-storage disease. QJM, 91, 607–18. Kelly AL, et al. (2001). Classification and genetic features of neonatal haemochromatosis: a study of twenty-​seven affected pedigrees and molecular analysis of genes implicated in iron metabolism. J Med Genet, 38, 599–​10. Le Gac G, et al. (2004). The recently identified type 2A juvenile haemo- chromatosis gene (HJV), a second candidate modifier of the C282Y homozygous phenotype. Hum Mol Genet, 13, 1913–​18. Legros L, et al. (2015). Non-​invasive assessment of liver fibrosis in C282Y homozygous HFE hemochromatosis. Liver Int, 35, 1731–​8. McCance RA, Widdowson EM (1937). Absorption and excretion of iron. Lancet, 233, 680–​4. McKie AT, et al. (2000). A novel duodenal iron-​regulated transporter, IREG1, implicated in baso-​lateral transfer of iron to the circulation. Mol Cell, 5, 299–​309. McKie AT, et al. (2001). An iron-​regulated ferric reductase associated with the absorption of dietary iron. Science, 291, 1755–​9. McLaren CE, et al. (2015). Exome sequencing in HFE C282Y homo- zygous men with extreme phenotypes identifies a GNPAT variant associated with iron overload. Hepatology, 62, 429–​39. Merryweather-​Clarke AT, et al. (2003). Digenic inheritance of muta- tions in HAMP and HFE results in different types of haemochroma- tosis. Hum Mol Genet, 12, 2241–​7. Meynard D, et al. (2009). Lack of the bone morphogenetic protein BMP6 induces massive iron overload. Nat Genet, 41, 478–​81. Montosi G, et al. (2001). Autosomal-​dominant hemochromatosis is as- sociated with a mutation in the ferroportin (SLC11A3) gene. J Clin Invest, 108, 619–​23. Nai A, et  al. (2015). The second transferrin receptor regulates red blood cell production in mice. Blood, 125, 1170–​9. Nemeth E, et  al. (2004). Hepcidin regulates cellular iron efflux by binding to ferroportin and inducing its internalization. Science, 306, 2090–​3. Nicolas G, et  al. (2003). Constitutive hepcidin expression prevents iron overload in a mouse model of hemochromatosis. Nat Genet, 34, 97–​101. Niederau C, et al. (1996). Long-​term survival in patients with heredi- tary haemochromatosis. Gastroenterology, 110, 1107–​19. Olynyk JK, et  al. (2004). Evolution of untreated hereditary hemo- chromatosis in the Busselton population: a 17-​year study. Mayo Clin Proc, 79, 309–​13. Papanikolaou G, et al. (2004) Mutations in HFE2 cause iron overload in chromosome 1q-​linked juvenile hemochromatosis. Nat Genet, 36, 77–​82. Roetto A, et  al. (2003). Mutant antimicrobial peptide hepcidin is associated with severe juvenile hemochromatosis. Nat Genet, 33, 21–​2. Sheldon JH (1935). Haemochromatosis. Oxford University Press, London. Simon M, Bourel M, Genetet B (1977). Idiopathic hemochroma- tosis: demonstration of recessive transmission and early detection by family HLA typing. N Engl J Med, 297, 1017–​21. Stickel F, et al. (2014). Evaluation of genome-​wide loci of iron metab- olism in hereditary hemochromatosis identifies PCSK7 as a host risk factor of liver cirrhosis. Hum Mol Genet, 23, 3883–​90. Schmidt PJ et al. (2013). An RNAi therapeutic targeting Tmprss6 de- creases iron overload in Hfe(-​/​-​) mice and ameliorates anemia and iron overload in murine β-​thalassemia intermedia. Blood, 121, 1200–​8. Taylor SA, et al. (2018).The Effects of Gestational Alloimmune Liver Disease on Fetal and Infant Morbidity and Mortality. J Pediatr, 196, 123–128.e1.

12.7.2 Inherited diseases of copper metabolism Wil

12.7.2 Inherited diseases of copper metabolism: Wilson’s disease and Menkes’ disease 2115

12.7.2  Inherited diseases of copper metabolism: Wilson’s disease and Menkes’ disease 2115 12.7.2  Inherited diseases of copper metabolism: Wilson’​s disease and Menkes’​ disease Michael L. Schilsky and Pramod K. Mistry ESSENTIALS Copper is an essential metal that is an important cofactor for many proteins and enzymes. Two related genetic defects in copper trans- port have been described, each with distinct phenotypes. Wilson’s disease An uncommon disorder (1 in 30  000) caused by autosomal re- cessive loss-​of-​function mutations in a metal-​transporting P-​type ATPase (ATP7B) that result in defective copper excretion into bile and hence copper toxicity. Typical presentation is in the second and third decade of life with liver disease (ranging from asymptomatic to acute fulminant hepatic failure or chronic end-​stage liver dis- ease) or neurological or psychiatric disorder (dystonia, dysarthria, parkinsonian tremor, movement disorder, a spectrum of psychiatric ailments). While no single biochemical test or clinical finding is suf- ficient for establishing the diagnosis, typical findings include low serum caeruloplasmin, high urinary copper excretion, and elevated liver copper content. Corneal Kayser–​Fleischer rings may be seen. Treatment is with copper chelating agents and zinc. Liver transplant- ation is required for fulminant hepatic failure and decompensated liver disease unresponsive to medical therapy. Menkes’ disease A rare disorder (1 in 300 000) caused by X-​linked loss-​of-​function mutations in a P-​type ATPase homologous to ATP7B (ATP7A) that result in defective copper transport across intestine, placenta, and brain and hence cellular copper deficiency. Clinical presentation is in infancy with facial dimorphism, connective tissue disorder, hypo- pigmentation, abnormal hair, seizures, and failure to thrive, usually followed by death by age 3 years (although some variants with a milder phenotype result from milder mutations, e.g. occipital horn syndrome). Treatment, which is only effective when presymptomatic diagnosis is made in a sibling after florid presentation in a previous affected sibling, is with intravenous copper histidine. Introduction Copper is an essential metal that is an important cofactor for many proteins and copper-​containing enzymes involved in cellular res- piration, antioxidant defence, pigment production, neurotrans- mitter formation, connective tissue synthesis, and iron homeostasis. Therefore, in states of impaired copper homeostasis resulting in copper excess or copper deficiency, tissue injury and organ dysfunc- tion ensue. The average diet provides substantial amounts of copper, typically between 2 and 5 mg/​day, most of which is eventually ex- creted in the bile and then in the stool. Copper is absorbed by en- terocytes mainly in the duodenum and proximal small intestine and transported in the portal circulation bound to albumin and histidine to the liver where it is avidly removed from the circulation. The liver utilizes some copper for metabolic needs, synthesizes and secretes the copper containing the protein caeruloplasmin, and excretes ex- cess copper into the bile (Fig. 12.7.2.1). Wilson’s disease (OMIM 277900) Introduction and historical perspective Wilson’s disease (hepatolenticular degeneration) was first described in detail in 1912 by Kinnear Wilson as progressive lenticular de- generation, a familial, lethal neurological disease accompanied by chronic liver disease leading to cirrhosis. Over the next few decades, the role of copper in pathogenesis was established, and the pattern of inheritance was determined to be autosomal recessive. Wilson’s disease was uniformly fatal until treatments were devel- oped just over half a century ago, when it became one of the first liver diseases for which effective pharmacological treatment was identified. The first chelating agent, introduced in 1951 for the treat- ment of Wilson’s disease, was the intramuscularly administered compound British Anti-​Lewisite (BAL or dimercaptopropanol). The identification and testing of an orally administered chelator, d-​penicillamine, by John Walsh in 1956 revolutionized treatment. Other treatment modalities have since been introduced, including the use of zinc salts to block enteral copper absorption, trientine and tetrathiomolybdate to chelate copper, and liver transplantation, which may be life-​saving and curative because the liver is the pri- mary site of the metabolic disorder. Genetic defect In 1993, the gene defect in Wilson’s disease was identified. The ATP7B gene encodes a metal-​transporting P-​type ATPase, which is expressed mainly in hepatocytes and functions in the transmembrane transport of copper. The absent or reduced function of ATP7B protein leads to im- paired excretion of excess copper into the bile, leading to copper accu- mulation and toxicity. Eventually copper is released into the bloodstream and deposited in extrahepatic tissues. Failure to incorporate copper into caeruloplasmin is an additional consequence of the loss of functional ATP7B protein. Caeruloplasmin devoid of copper, apocaeruloplasmin, has a short half-​life, which causes decreased plasma levels found in most patients with Wilson’s disease (Fig. 12.7.2.1). Wilson’s disease occurs worldwide with an average incidence of approximately 30 per million. The carrier frequency is approxi- mately 1 in 90. However, a recent study of the frequency of ATP7B mutations in the United Kingdom suggested that this established figure may underestimate the frequency of affected individuals and carriers, and estimated a disease incidence of 1:8000. Clinical features There is protean phenotypic presentation comprising various com- binations of liver disease, progressive neurological disorder, and psychiatric disorder. Presentation with liver disease occurs more frequently in children and younger adult patients than in older adults. Overt liver disease is the most common presenting feature in childhood with the most common age of presentation between 10 and 13 years, but manifestations may be present as early as age 3 to 5 years in rare individuals. In contrast, neurological disease oc- curs as the initial presenting feature in adults, usually in the third

section 12  Metabolic disorders 2116 or fourth decade of life. This sequence reflects the natural history of primary hepatic involvement followed by neurological and other extrahepatic organ dysfunction. Symptoms at any age can be non- specific and there is considerable overlap between distinct hepatic and neurological presentations frequently cited in the literature. A patient presenting with liver disease aged between 5 and 40  years with decreased serum caeruloplasmin and detectable Kayser–​Fleischer rings represents the classic form of Wilson’s dis- ease. However, about one-​half of patients presenting with liver dis- ease do not possess two of these three criteria and pose a challenge in trying to establish the diagnosis. Moreover, as with other liver dis- eases, patients may not come to medical attention when their clin- ical disease is comparatively mild. Even when presymptomatic siblings are excluded, the age at which Wilson’s disease may present is both younger and older than generally appreciated, though most present between the ages of 5 and 35 years. Wilson’s disease is increasingly diagnosed in children younger than 5 years old, with atypical findings in children under 2 years old that include cirrhosis in a 3-​year-​old and fulminant hepatic failure in a 5-​year-​old. The oldest patients diagnosed with Wilson’s disease were in their early 70s. With new molecular testing capabilities, testing for Wilson’s disease can be performed even in utero or in newborns, and cases are now being diagnosed earlier than ever before. Liver presentations The diversity of liver disease encountered in patients with Wilson’s disease is summarized in Table 12.7.2.1. Wilson’s disease should be considered in the differential diag- nosis of patients with unexplained liver disease and when neuro- logical and/​or psychiatric symptoms occur concurrently with liver disease. Liver involvement can range from asymptomatic, with only biochemical abnormalities or an isolated clinical finding of hepatomegaly, to acute fulminant hepatic failure or chronic end-​ stage liver disease. Children may be entirely asymptomatic, with hepatomegaly or abnormal serum aminotransferases found only in- cidentally. Some patients may have a brief clinical illness resembling an acute viral hepatitis or mononucleosis, and others may present with features indistinguishable from autoimmune hepatitis. Some may present with only biochemical abnormalities or histological findings of steatosis on liver biopsy and many others with signs of chronic liver disease with advancing fibrosis and inflammation and evidence of compensated or decompensated cirrhosis. Patients may present with isolated splenomegaly due to clinically unapparent cir- rhosis and portal hypertension. Wilson’s disease may also present as acute fulminant hepatic failure with an associated Coombs’-​negative haemolytic anaemia and acute kidney injury. Some patients have transient episodes of jaundice, due to haemolysis. Low-​grade haemolysis may be associ- ated with Wilson’s disease when liver disease is not clinically evident. Neurological presentations Neurological manifestations of Wilson’s disease typically present later than the liver disease, most often in the third decade of life. However, earlier subtle findings may appear in paediatric patients, including changes in behaviour, deterioration in school work or the inability to perform activities requiring good hand–​eye coordination. Patient may exhibit small handwriting as in Parkinson’s disease (micrographia). Other common findings in those presenting with neurological dis- ease include tremor, lack of motor coordination, drooling, dysarthria, dystonia, and spasticity or athetosis. Because of pseudobulbar palsy, transfer dysphagia may also occur, with a risk of aspiration if severe. Dysautonomia may be present, but usually in concert with other neurological findings. Migraine headaches and insomnia may be re- ported, but seizures are infrequent. Along with behavioural changes, Cu CPN Golgi +Cu Lysosome −Cu Cu Cu Cu ATP7B ATP7B Hepatocyte canalicular membrane hCTR basolateral membrane caeruloplasmin Cu-albumin hCTR Cu Cu-GSH bile bile cMOAT Trans Golgi Endosome +Cu ATOX1 Cu Golgi ATP7A Enterocyte basolateral membrane hCTR apical membrane hCTR Cu Trans Golgi ATOX1 Cu Cu Cu +Cu −Cu Fig. 12.7.2.1  Cellular copper trafficking in the hepatocyte and enterocyte depicting the contrasting metabolic defects in Wilson’s disease and Menkes’ disease. ATP7B is the major copper transporter in the hepatocyte, and ATP7A fulfils this role in the enterocyte. Copper gains access to both cell types via copper transporter 1, hCTR, and is delivered by the copper chaperone ATOX1 to ATP7B and APT7A, respectively, residing in the trans-​ Golgi network. Increasing cell copper content is associated with trafficking of ATP7B towards the apical canalicular membrane and copper excretion in bile in the hepatocyte. In contrast, in the enterocyte, increasing dietary copper leads to net absorption of copper via the basolateral surface. While significant amounts of ATP7B expression are relatively restricted to hepatocytes and a few other cell types, ATP7A expression is more ubiquitous except that it is minimally expressed in the liver. In the other cells that express ATP7A, the basolateral presence of ATP7A when copper is abundant, and this protein’s copper transport activity, result in the cellular excretion of excess copper. In the kidney, ATP7A and ATP7B are active in copper reabsorption or excretion from the body.

12.7.2  Inherited diseases of copper metabolism: Wilson’s disease and Menkes’ disease 2117 other psychiatric manifestations include depression, anxiety, and even frank psychosis. Many but not all individuals with neurological or psychiatric manifestations may have cirrhosis, but frequently they are not symptomatic from their liver disease. Eye manifestations Kayser–​Fleischer rings represent deposition of copper in the Descemet’s membrane of the cornea (Fig. 12.7.2.2). When they are visible by direct inspection, they appear as a band of golden-​ brownish pigment near the limbus. A slit-​lamp examination by an experienced observer is required to identify Kayser–​Fleischer rings in most patients. Rarely, they may be found in patients with chronic cholestatic diseases and in children with neonatal cholestasis; how- ever, these disorders can usually be distinguished from Wilson’s dis- ease on clinical grounds or on histology. Kayser–​Fleischer rings are present in approximately 95% of patients with a neurological presen- tation but in only approximately 40 to 50% of patients with predom- inant hepatic disease at the time of diagnosis. Sunflower cataracts, also found by slit lamp examination but with a lower frequency, represent deposits of copper in the lens. These typic- ally do not obstruct vision, and—​along with Kayser–​Fleischer rings—​ will gradually disappear with effective medical treatment or following liver transplant. Reappearance of either of these ophthalmological findings in a medically treated patient in whom these had previously disappeared suggests noncompliance with therapy. Diagnostic testing A diagnosis of Wilson’s disease should be considered in any patient with unexplained liver disease, especially if associated with neurological and psychiatric disease; patients with acute fulminant liver failure, especially if haemolysis is present; and first-​degree relatives of affected patients. No single biochemical test or clinical finding is sufficient to establish the diagnosis. A combination of clinical and biochemical evaluation is necessary to make the diagnosis, and this has led to the use of a scoring system that uses these findings and tests to determine the probability of establishing a diagnosis of Wilson’s disease (Leipzig criteria). Biochemical liver tests Serum aminotransferase activities are generally abnormal in Wilson’s disease except at a very early age. In many patients, the degree of ele- vation of aminotransferase activity may be mild compared to other liver injuries and does not reflect the severity of the liver disease. Measures of liver synthetic function such as coagulation factors and proteins such as albumin may be reduced significantly in such cases. Caeruloplasmin Caeruloplasmin is a 132-​kDa protein synthesized mainly in the liver and is also an acute phase reactant. The vast majority of caeruloplasmin is secreted into the circulation from hepatocytes as a copper-​carrying protein containing six copper atoms per mol- ecule (holocaeruloplasmin), and the remainder as the protein lacking copper (apocaeruloplasmin). Caeruloplasmin is the major carrier for copper in the blood, accounting for 90% of the circulating copper in normal individuals. It is also a ferroxidase and a nitric oxide oxidase, so it influences nitric oxide homeostasis. Levels of serum caeruloplasmin may be measured enzymatically by their copper-​dependent oxidase activity towards these substrates, or by antibody-​dependent assays. Results generally are regarded as equivalent, but it should be noted that immunological assays routinely in clinical use may overestimate caeruloplasmin concentrations since they do not discriminate be- tween apocaeruloplasmin and holocaeruloplasmin. A serum caeruloplasmin level of less than 200 mg/​L (<20 mg/​dl, although the lower level of the normal range can vary between la- boratories) is considered consistent with Wilson’s disease. Low caeruloplasmin levels are found in approximately 95% of cases. However, serum caeruloplasmin alone as a screening test for Wilson’s disease in patients referred with liver disease has a low positive Table 12.7.2.1  Clinical patterns of hepatic, neurological, and psychiatric disease in patients with Wilson’s disease Hepatic Asymptomatic hepatomegaly Isolated splenomegaly Persistently elevated serum aminotransferase activity (AST, ALT) Fatty liver Acute hepatitis Resembling autoimmune hepatitis Cirrhosis—​compensated or decompensated Fulminant hepatic failure Neurological Movement disorders (tremor, involuntary movements) Drooling, dysarthria Rigid dystonia Pseudobulbar palsy Dysautonomia Migraine headaches Insomnia Seizures Psychiatric Depression Neurotic behaviours Personality changes Psychosis Other systems Renal abnormalities: aminoaciduria and nephrolithiasis Skeletal abnormalities: premature osteoporosis and arthritis Cardiomyopathy, dysrhythmias Pancreatitis Hypoparathyroidism Menstrual irregularities; infertility, repeated miscarriages Fig. 12.7.2.2  Florid Kayser–​Fleischer ring in a patient with Wilson’s disease. Courtesy of Dr Susan Hall Forster, Yale School of Medicine.

section 12  Metabolic disorders 2118 predictive value (approximately 6%), although the very lowest values have a slightly higher predictive value than levels near or at the lowest limit of normal. Moreover, a low caeruloplasmin level is found in 20% of healthy heterozygote carriers of Wilson’s disease, and in other disorders including protein-​losing states (gut and kidney), poor hepatocellular synthetic function, and in another genetic disorder where the caeruloplasmin gene is affected, acaeruloplasminaemia. Serum copper Although a disease of copper overload, the total serum copper in Wilson’s disease is usually decreased in proportion to the decreased caeruloplasmin in the circulation. In patients with severe liver injury, serum copper may be within the normal range or even elevated des- pite a decreased serum caeruloplasmin level. In the setting of acute fulminant hepatic failure due to Wilson’s disease, levels of serum copper may be markedly elevated due to the sudden release of the metal from tissue stores. Normal or elevated serum copper levels in the face of decreased levels of caeruloplasmin indicate an increase in circulating ‘free’ or noncaeruloplasmin-​bound copper. Urinary copper excretion The amount of copper excreted in the urine in a 24-​h period, which reflects the amount of noncaeruloplasmin copper in circulation, may be helpful for diagnosing Wilson’s disease and for monitoring treatment. Basal measurements can provide useful diagnostic infor- mation so long as copper does not contaminate the collection ap- paratus (this is less problematic with current plastic disposables) and the urine collection is complete. Basal 24-​h urinary excretion of copper in Wilson’s disease is typically more than 100 µg (1.6 µmoles) in symptomatic patients, but a level above 40 µg (>0.6 µmoles) may indicate Wilson’s disease and requires further investigation. Liver copper concentration Liver copper content of more than 250 µg/​g dry weight provides crit- ical diagnostic information and should be obtained in cases where the diagnosis is not straightforward and in younger patients. In untreated patients, normal hepatic copper content (<40–​50 µg/​g dry weight) almost always excludes a diagnosis of Wilson’s disease. Further diag- nostic testing is indicated for patients with intermediate copper con- centrations (70–​250 µg/​g dry weight), especially if there is active liver disease or other symptoms of Wilson’s disease. The major problem with attempting diagnosis based on hepatic parenchymal copper concentra- tion is that in the later stages of Wilson’s disease distribution of copper within the liver is often inhomogeneous. Furthermore, copper con- centrations may seem falsely low in those with significant amounts of fibrosis in the biopsy, and—​since this is a weight-​based measurement—​ the error is potentially greater in smaller specimens. Liver biopsy findings The earliest histological abnormalities in the liver include microvesicular and macrovesicular steatosis, glycogenated nu- clei, and focal hepatocellular necrosis. The liver biopsy may show classic histological features of autoimmune hepatitis. With pro- gressive parenchymal damage, fibrosis and subsequently cirrhosis develops and is frequently found in most patients by the second decade of life. Cirrhosis is usually macronodular, although occa- sionally micronodular. In the setting of fulminant hepatic failure, there is marked hepatocellular degeneration, hepatocytes apoptosis, and parenchymal collapse, typically on the background of cirrhosis. Detection of copper in hepatocytes by orcein or rhodanine staining is highly variable. In extreme cases, nodules lacking histochemically detectable copper are found next to cirrhotic nodules with abundant copper, and negative histochemistry for copper does not exclude the diagnosis. Electron microscopy reveals characteristic mitochondrial abnormalities, dilated cristae and crystalline deposits, in hepatocytes in the early phase of the disease (when steatosis is evident). Neurological findings and radiological imaging of the brain Neurological disease may manifest with parkinsonian features of dystonia, hypertonia, and rigidity, with tremors and dysarthria. Muscle spasms, which can lead to contractures, dysarthria, dys- phonia, and dysphagia can be incapacitating. Movement disorder may be present as well. At this stage of the disease, MRI or CT scan- ning of the brain may detect structural abnormalities in the basal ganglia. Most frequently found are increased density on CT and hyperintensity on T2-​weighted MRI in the region of the basal gan- glia, but other regions of the brain may be involved. Genetic studies ATP7B mutation analysis can be difficult because of the multiplicity of mutations, the occurrence of mutations in noncoding sequences, and the large size of the gene that spans around 80 kb. Pedigree ana- lysis using haplotypes of polymorphisms flanking the Wilson’s dis- ease gene can be used but it is mostly replaced by methods relying on high-​throughput sequencing. Most patients are compound heterozy- gotes. Currently, over 500 mutations of ATP7B have been identified. Mutation analysis is an especially valuable diagnostic strategy for certain well-​defined populations harbouring prevalent ATP7B mutations. Populations with a single predominant mutation in- clude those of Iceland, Japan, Korea, Sardinia, Spain and the Canary Islands, and Taiwan. Certain populations in Eastern Europe also show predominance of the H1069Q mutation, accounting for nearly 40% of disease alleles. Genotype–​phenotype correlation is imperfect as in most other inherited metabolic diseases of the liver, indicating an important role for modifier genes and environmental factors in the determination of phenotypic characteristics. However, a large multinational study and a meta-​analysis suggest that homozygosity for the H1069Q mutation is associated with neurological presenta- tion in adults. The H1060Q Wilson ATPase resides in a highly con- served sequence in the cytoplasmic loop, SEHPL, and appears to result in defective trafficking of the mutant protein. Diagnostic considerations in specific target populations Liver diseases which mimic Wilson’s disease Patients with Wilson’s disease, especially younger ones, may have clin- ical features and histological findings on liver biopsy indistinguishable from autoimmune hepatitis. All children with apparent autoimmune hepatitis and any adult patient with the presumptive diagnosis of autoimmune hepatitis failing to respond appropriately to cortico- steroids must be evaluated for Wilson’s disease. Hepatic steatosis in Wilson’s disease is rarely as severe as in nonalcoholic fatty liver dis- ease. Nevertheless, occasional patients with Wilson’s disease closely resemble nonalcoholic fatty liver disease or have both diseases. Acute fulminant liver failure A high level of clinical suspicion is essential because transplant re- ferral is required given the poor prognosis for medical management

12.7.2  Inherited diseases of copper metabolism: Wilson’s disease and Menkes’ disease 2119 of these patients, and there are disease-​specific management and family screening that should follow. Most patients with the ful- minant hepatic failure presentation of Wilson’s disease have a char- acteristic pattern of clinical findings: • Coombs’-​negative haemolytic anaemia with features of acute intravascular haemolysis • Coagulopathy unresponsive to parenteral vitamin K administration • Rapid progression to renal failure • Relatively modest rises in serum aminotransferases (typically <2000 IU/​litre) from the beginning of clinical illness • Normal or markedly subnormal serum alkaline phosphatase (typ- ically <40 IU/​litre) with alkaline phosphatase to bilirubin ratio of less than 4:1. • AST-​to-​ALT ratio of greater than 2:1. • Female-​to-​male ratio of greater than 2:1. Serum caeruloplasmin is usually decreased, but the predictive value of this test in the setting of acute liver failure is poor. Serum copper and 24-​h urinary excretion of copper are greatly elevated. The serum copper is usually greater than 200 µg/​dl (31.5 µmol/​litre). Kayser–​Fleischer rings may be identified to support the diagnosis of Wilson’s disease but are absent in approximately 50% of these pa- tients. Expeditious diagnosis is critically important since, without timely liver transplantation, death is almost inevitable. Management of these patients awaiting transplantation includes measures to sup- port their liver injury and specifically is geared to reduce the removal of the excess circulating copper to prevent further liver and renal injury, reduce haemolysis, and help stabilize the patient. Family screening First-​degree relatives of any patient newly diagnosed with Wilson’s disease must be screened. Assessment should include serum copper, caeruloplasmin, liver function tests, slit-​lamp examination of the eyes for Kayser–​Fleischer rings, and basal 24-​h urinary copper. Individuals without Kayser–​Fleischer rings who have subnormal caeruloplasmin and abnormal liver tests undergo liver biopsy to confirm the diag- nosis. Molecular analysis of the ATP7B gene is increasingly available and should be used as primary screening tool, especially for sibling screening once the proband is identified. Treatment should be initi- ated for all individuals over 3 years old identified as patients by family screening, and individualized for younger patients. Perspective on diagnosis Over the years, diagnostic advances have enabled a more system- atic evaluation of individuals suspected to have Wilson’s disease, and unlike the early description by Wilson, many patients are now diag- nosed before they develop neurological symptoms. These include the recognition of corneal Kayser–​Fleischer rings, 50% of patients with hepatic presentations, and 98% of those with neurological or psychiatric presentation of disease. Molecular diagnostic studies have made it feasible to identify presymptomatic and symptomatic individuals by analysing for disease-​specific mutations of the ATP7B gene. However, since de novo genetic diagnosis is currently expen- sive, not universally available, and (most importantly) sometimes inconclusive, a combination of clinical findings and biochemical testing is still necessary to establish the diagnosis of Wilson’s disease. Treatment In general, the approach to treatment is dependent upon whether there is active disease or symptoms, whether neurological, psychiatric, or hepatic, or whether the patient is identified prior to the onset of clinical symptoms. This distinction helps in determining the choice of therapy and the dosages of medications utilized. The recommended initial treatment of symptomatic patients or those with active disease is with chelating agents. The largest treatment experience worldwide is with penicillamine; however, there is more frequent consideration of trientine for primary therapy. Combination therapy, in which zinc is utilized in conjunction with a chelating agent (temporally separated), has a theoretical basis in both blocking copper uptake and eliminating excess copper. Past studies of the use of tetrathiomolybdate as an alter- native chelating agent for the initial treatment of neurological Wilson’s disease suggest that this drug may be useful as initial therapy for pa- tients presenting with neurological symptoms, and newer studies are underway to revisit this treatment (Table 12.7.2.2). Table 12.7.2.2  Pharmacological treatments of Wilson’s disease Drug/​dose in adults Mode of action Neurological deterioration Side effects Comments Penicillamine 750–​1500 mg in 2 or 3 divided doses; requires supplemental pyridoxine General chelator Induces cupruria 10–​20% during the initial phase of treatment Fever, rash, proteinuria, lupus-​like reaction Aplastic anaemia Leucopenia Thrombocytopenia Nephrotic syndrome Degenerative changes in skin; Elastosis perforans serpiginosa Serous retinitis Hepatotoxicity Colitis (rare) Reduce dose for surgery to promote wound healing and during pregnancy to reduce teratogenicity Trientine 750–​1500 mg in 2 or 3 divided doses General chelator Induces cupruria 10–​15% during the initial phase of treatment Gastritis; aplastic anaemia rare Sideroblastic anaemia Colitis (rare) Reduce dose for surgery to promote wound healing and during pregnancy Zinc 75–​150 mg in 3 divided doses Metallothionein inducer Blocks intestinal absorption of copper Can occur during the initial phase of treatment; infrequent hepatic decompensation Gastritis; biochemical pancreatitis; zinc accumulation; possible changes in immune function No dosage reduction for surgery or pregnancy Tetrathio-​molybdate 120 mg in 6 divided doses (with meals and apart from meals) Chelator Blocks copper absorption Reports of rare neurological deterioration during the initial treatment Anaemia; neutropenia Hepatotoxicity Experimental in the United States of America and Europe

section 12  Metabolic disorders 2120 Once the disease symptoms or biochemical abnormalities have stabilized, typically in 2 to 6 months after the initiation of therapy, a reduced dosage of chelators or zinc therapy can be used for main- tenance treatment. Patients presenting without symptoms may be treated with either maintenance dosages of a chelating agent or with zinc from the outset. Failure to comply with lifelong therapy has led to recurrent symptoms and liver failure, the latter requiring liver transplant for survival. Monitoring of therapy includes monitoring for compliance as well as for potential treatment-​induced side effects. Liver transplantation Liver transplantation is the only effective option for those with Wilson’s disease who present with acute fulminant hepatic failure and is indicated for all Wilson’s disease patients with decompensated liver disease unresponsive to medical therapy. In acute fulminant liver failure due to Wilson’s disease, interventions to rapidly reduce elevated free circulating copper may reduce secondary organ injury while the patient awaits a suitable organ donor. Liver transplantation corrects the hepatic metabolic defects of Wilson’s disease and over time reverses extrahepatic copper disposition. Living donor liver transplantation has been successfully performed for Wilson’s dis- ease, including the use of donor livers from heterozygote carriers for Wilson’s disease. One-​year survival following liver transplantation ranges from 79 to 87%, and those who survive this early period con- tinue to survive long term. A liver transplant is not recommended as the primary treatment for neurological Wilson’s disease since the liver disease is stabilized by medical therapy in most of these individuals and outcomes with a liver transplant in the setting of advanced neurological disease are not always beneficial. Menkes’ disease (OMIM 309400) Menkes’ disease is an X-​linked recessive neurodegenerative disorder presenting in infancy due to mutations in the ATP7A gene, which encodes a P-​type ATPase homologous to ATP7B (Fig. 12.7.2.1). The pathology and disease manifestations reflect decreased activities of enzymes that require copper as a cofactor, such as dopamine-​β-​ hydroxylase, cytochrome c oxidase, and lysyl oxidase. Affected infants appear healthy at birth but by the age of approxi- mately 2 months develop hypotonia, seizures, skin and joint laxity, hair twisting (pili torti), and failure to thrive, usually followed by death by 3 years of age from end-​stage neurodegenerative disease. Treatment with daily injections of copper histidine may im- prove the outcome if started presymptomatically soon after birth. However, newborn screening is not routinely available and early detection is difficult because clinical abnormalities in affected new- borns are absent or subtle, hence this type of pre-​emptive treatment is only possible when presymptomatic diagnosis is made in a sibling after florid presentation in a previous affected child. The usual biochemical markers, low serum copper and caerulo­ plasmin, are unreliable in the neonatal period. Molecular diagnosis is the preferred option when available, but this is rendered difficult by the large size of ATP7A (150 kb) and diversity of mutation types, including large deletions and chromosomal rearrangements. A useful test for neonatal diagnosis of Menkes’ disease has been developed involving the measurement of serum neurotransmitter levels. Dopamine-​β-​hydroxylase converts dopamine to noradren- aline and these transmitters in turn can be further metabolized to dihydroxyphenylacetic acid to dihydroxyphenylglycol, respectively. In Menkes’ disease, deficiency of dopamine-​β-​hydroxylase (an en- zyme that depends on copper for its activity) leads to a high ratio of dopamine to noradrenaline as well as of dihydroxyphenylacetic acid to dihydroxyphenylglycol. These characteristic abnormal- ities can be used to identify presymptomatic disease, allowing pre-​ emptive therapy with copper histidine and resulting in delay in the development of typical neurodegeneration and other changes, but most often not the arrest of the disease. Trials of gene therapy are in development. Occipital horn syndrome is a milder allelic variant of Menkes’ dis- ease that may present later in life without the same neurodegenerative changes observed in the typical early presentation described preiously. Clinical features include skeletal deformity (including exostoses of the occipital bone—​the ‘horns’), looseness of the skin, and joint laxity, features which previously led the condition to be considered a variant of Ehlers–​Danlos syndrome. FURTHER READING Bandmann O, Weiss KH, Kaler SG (2015). Wilson’s disease and other neurological copper disorders. Lancet Neurol, 14, 103–​13. Beinhardt S, et al. (2014). Long-​term outcomes of patients with Wilson disease in a large Austrian cohort. Clin Gastroenterol Hepatol, 12, 683–​9. Coffey AJ, et  al. (2013). A genetic study of Wilson’s disease in the United Kingdom. Brain, 136, 1476–​87. Ferenci P. (2014). Phenotype-​genotype correlations in patients with Wilson’s disease. Ann N Y Acad Sci, 1315, 1–​5. Ferenci P, et al. (2012). EASL Clinical Practice Guidelines: Wilson’s disease. European Association for Study of Liver. J Hepatol, 56, 671–​85. Guillaud O, et al. (2014). Long term results of liver transplantation for Wilson’s disease: experience in France. J Hepatol, 60, 579–​89. Kaler SG (2011). ATP7A-​related copper transport diseases-​emerging concepts and future trends. Nat Rev Neurol, 7, 15–​29. Kaler SG, et al. (2008). Neonatal diagnosis and treatment of Menkes disease. N Engl J Med, 358, 605–​14. Lorincz MT (2018). Wilson disease and related copper disorders. Handb Clin Neurol, 147, 279–​92. Lutsenko S (2014). Modifying factors and phenotypic diversity in Wilson’s disease. Ann N Y Acad Sci, 1315, 56–​63. Roberts EA, Schilsky ML, American Association for Study of Liver Diseases (2008). Diagnosis and treatment of Wilson disease: an up- date. Hepatology, 47, 2089–​111. Schilsky ML (2017). Wilson disease: diagnosis, treatment and follow-​ up. Clin Liver Dis, 21, 755–​67. Zimbrean PC, Schilsky ML (2014). Psychiatric aspects of Wilson dis- ease: a review. Gen Hosp Psychiatry, 36, 53–​62

12.8 Lysosomal disease 2121

12.8 Lysosomal disease 2121

ESSENTIALS Lysosomal function and classification of diseases The lysosome is a ubiquitous, single membrane-​bound intracellular organelle which continuously recycles biological macromolecules: it not only breaks down cell components but has a dynamic role in nu- trient and energy sensing that, through regulatory signalling, is critical for homeostasis and metabolic economy of the cell. More than 80 lysosomal diseases caused by single gene defects are known. These are classified according to the nature of the pri- mary storage molecules (biochemical classification) or according to the defective molecular cell physiology (functional classification), or (more usefully) a combination of these classification systems that in- corporates genetic information. Biochemical classification identifies (1)  sphingolipidoses; (2) mucopolysaccharidoses; (3) glycoproteinoses; (4) glycogenosis, with or without lysosomal debris derived from subcellular organ- elles due to impaired autophagy; and (5)  miscellaneous condi- tions with multiple classes of storage material such as the neuronal ceroid lipofuscinoses. Functional classification describes deficiency of (1) a specific acid hydrolase activity, (2) an activator protein,
(3) a lysosomal membrane protein or transporter, or (4) abnormal post-​translational modification of lysosomal proteins, and (5) ab- normal biogenesis of lysosomes. A unified classification will emerge from genetic characterization integrated with clinicopathological manifestations of the individual disorders. Clinical features and diagnosis About one in 5000 live-​born infants have a lysosomal disorder. Clinically diverse, the lysosomal diseases can appear at any age but are very rarely present at birth. Diagnosis is usually suspected on the basis of key clinical presentations of progressive neurodegenerative disease, often combined with visceral enlargement (especially splenomegaly), connective tissue and skeletal disease, or particular syndromic appearances. As with all disorders with strong hereditary determinants, diagnosis is crucial for prognosis and genetic counselling, because specific ther- apies may have disease-​modifying effects, and to prevent inappro- priate interventions (e.g. removal of an enlarged spleen). A detailed family history, including careful analysis of the extended pedigree, is essential. Diagnosis is confirmed by biochemical methods including examination of urine metabolites or specific enzymatic assays on leucocytes or cultured fibroblasts, histochemical stains of existing bi- opsy material, and/​or next-​generation DNA sequencing studies. Particular lysosomal diseases Fabry’s and Gaucher’s diseases (glycosphingolipidoses) are prob- ably the most frequent in the general population, but certain lyso- somal diseases are over-​represented in particular groups where consanguinity or endogamy is high (e.g. the high frequency of non-​neuronopathic Gaucher’s disease, infantile and late-​onset Tay–​ Sachs disease (GM2 gangliosidosis), and Niemann–​Pick disease type A (neuronopathic) in Ashkenazi Jews). Fabry’s disease—​an X-​linked disorder caused by deficiency of α-​galactosidase A,  which leads to the accumulation of globo­ triaosylceramide (Gb3), typically manifests in early childhood with lancinating pain and background burning sensations in the extrem- ities. Other features include diarrhoea, lack of peripheral sweating, impotence, high-​tone deafness, angiokeratomas, chronic kidney disease, hypertrophic cardiomyopathy, and stroke. Enzyme therapy with recombinant human α-​galactosidase A is very costly but im- proves neuropathic pain and cardiac hypertrophy. Gaucher’s disease is ​an autosomal recessive trait caused by func- tional deficiency of acid glucocerebrosidase. Characteristic manifest- ations of the most frequent form—​‘adult non-​neuronopathic’ (type 1)—​include pancytopenia, splenic enlargement, and bone pain with osteoporosis and episodic osteonecrosis. Parenteral administration of enzyme therapy with imiglucerase (Cerezyme), velaglucerase alfa (VPRIV), taliglucerase alfa (Elelyso) or oral substrate reduction therapy with miglustat (Zavesca) or eliglustat (Cerdelga) is extremely expensive but clinically effective. Other diseases discussed in this chapter include (1)  cystinosis, (2)  the mucopolysaccharidoses, (3)  Pompe’s disease (glycogen storage disease type II), (4) Niemann–​Pick diseases, (5) lysosomal acid lipase deficiency, (6) Danon’s disease, and (7) diseases more recently attributed to primary defects in lysosomes and related organelles. Treatments Despite some striking therapeutic advances in several lysosomal conditions such as Gaucher’s disease, there is no specific or curative 12.8 Lysosomal disease Patrick B. Deegan and Timothy M. Cox

section 12  Metabolic disorders 2122 treatment for most lysosomal disorders. Supportive and palliative measures are nonetheless of great benefit. Advanced cell and molecular therapies can have striking benefits in several lysosomal diseases: these include bone marrow (haem- atopoietic stem cell) transplantation, specific augmentation with receptor-​targeted recombinant human lysosomal enzymes, sub- strate biosynthesis inhibitors, pharmacological chaperones, and sub- strate dissolution stratagems. Gene therapy is in a promising phase of clinical development for this group of disorders. Lysosomal function Since their discovery more than 65  years ago by the late Nobel prize winner Christian de Duve, lysosomes and their associated endosomal structures have been at the centre of research into mo- lecular cell biology and membrane dynamics. Lysosomes are an in- tegral part of the intracellular digestive system (see Chapter 3.1) and acquire complex macromolecules for breakdown and recycling by three main pathways:  (1) receptor-​mediated endocytosis, (2)  en- gulfment and fusion (phagocytosis), and (3)  autophagy. Greater understanding of lysosomal function has arisen from biochemical definition of cellular macromolecules that accumulate when the or- ganelle is affected by hereditary diseases. Many other molecular components that bring about trafficking and membrane-​fusion events, for example, Rab proteins and SNARE proteins, are localized to, and interact with, the lysosome. These proteins form complexes that bring about the ebb and flow of substrates and digestive products as they move between com- partments in the greater lysosomal network within the cell. Live-​ cell imaging techniques reveal a highly dynamic constellation of particles that are continuously moving and fusing as a result of the release of phosphate-​bond energy through the action of GTPases. This chemical bond energy is used to drive the continuous traf- ficking of membrane constituents involved in the innumerable tran- sient interactions of the endosomal, lysosomal, and autophagosome compartments. Endocytosis and membrane flow Receptor-​mediated endocytosis occurs by means of clathrin-​coated pits, a process by which molecules are delivered after internaliza- tion to a peripheral, and later to a perinuclear endosomal compart- ment, ‘the endolysosome’. The endolysosome undergoes maturation to form a lysosome after the loss of certain membrane components and further acidification. Some molecules acquired by receptor-​mediated endocytosis (e.g. apolipoprotein B in low-​density lipoproteins) are specifically retrieved and returned ultimately to the cell surface, having des- patched their cargo to the lysosome. Other molecules that are not retrieved are ultimately degraded by fusion with mature lysosomes and enzymatic hydrolysis (e.g. the epidermal growth factor receptor system). The endosomal system also mediates the traffic of nascent acid hydrolases from the trans-​Golgi network to the lysosome, employing a specific mannose 6-​phosphate receptor targeting system. Plasma membrane proteins bound for degradation in the lysosomal system are incorporated into membrane-​bound vesicles within the endosomal lumen and thus enter the lysosome upon fusion. In contrast, structural or transporter proteins bound for incorporation into the lysosomal membrane remain within the limiting membrane of the endosome and form part of the lyso- somal membrane on fusion. Two multicomponent macromolecular complexes, CORVET (core subunit vacuolar/​endosome tethering complex) and HOPS (homotypic fusion and protein sorting), bring about vacuolar fu- sion with the endosomal, lysosomal, and autophagosome com- partments. Recently, a few patients with destabilizing mutations in HOPS components have been investigated. While the cell biology is incompletely understood, these patients will be instructive since the complex phenotype of Turkic patients with mutations in the vacuolar protein-​sorting associated protein, VP33A, clearly has pathological lysosomal storage with features resembling a mucopolysaccharidosis (MPS) despite a full complement of lyso- somal acid hydrolases. Further investigations of such extraordinary patients are likely to reveal much about the functional dynamics of lysosome–​endosome pathways in cell biology and their relevance to medicine. Phagocytosis Lysosomes are also involved in a specialized process for the degrad- ation of exogenous particulates and proteins, including microbes and effete cells such as erythrocytes and neutrophils. Although this engulfment and fusion process involving phagolysosomes is a feature of many cells, it is particularly active in macrophages and dendritic cells. In macrophages, cell surface components on bacteria and yeast, as well as exogenous cells, are recognized and bound by specific receptors on the plasma membrane. The phagocytes engulf for- eign material to form large vesicles in which acidification and proteolysis, as well as the secretion of degradative molecules (including reactive oxygen and nitrogen species), is initiated. The phagolysosome fuses with lysosomes and further acidification oc- curs, so that the acid hydrolases are activated to effect destruction of the ingested material. A specialized phagolysosome variant occurs in osteoclasts that are derived from myeloid cells of mononuclear phagocyte origin. The osteoclastic resorptive vacuole serves as a large exteriorized lyso- somal compartment which is independently acidified for resorption of bone. Autophagy Autophagy occurs within cells:  microautophagy describes the degradation of cytosolic components trapped during invagin- ation of endosomes and lysosomes, while macroautophagy de- scribes the engulfment of relatively large volumes, including organelles. In a constant process of membrane fusion and flow, the endo- plasmic reticulum (reticulophagy), ribosomes, mitochondria (mitophagy), peroxisomes and other lysosomes, and particulate ma- terial such as macromolecular complexes of glycogen are engulfed by autophagic vacuoles. Formation of these vacuoles is initiated when a flattened cisterna composed of membrane, the phagophore, encir- cles cytoplasm to form a double-​layered vesicle, the autophagosome: acidified late endosomes and lysosomes fuse with the nascent vacu- oles to form an autolysosome.

12.8  Lysosomal disease 2123 With progressive acidification, the complement of lysosomal hydrolases effects breakdown of the inner membrane and digestion of the vacuolar contents. After digestion is completed, the autolysosome acquires an electron-​dense—​and often autofluorescent—​core known as a residual body. When lysosomal function is impeded, the breakdown of endogenous macromolecules is impaired; this, together with a failure to break down exogenous macromolecular substrates, results in a characteristic pattern of pathological storage of the biological residue. Disturbed autophagy is a spectacular fea- ture of cystinosis, Pompe’s disease, and Danon’s disease, but prob- ably drives much of the pathogenesis of many, if not all, lysosomal diseases, including those involving cells where autophagy is particu- larly active, as in neurons. Recent exploration of the central signalling pathway that controls tissue and cell growth, the mechanistic target of rapamycin com- plex 1 (mTORC1), has shown that under nutrient-​rich conditions this complex is recruited to the cytoplasmic surface of the lyso- some where it interacts with a large assembly of proteins including a small guanine triphosphatase, Rheb, which is itself regulated by the prevailing energy charge, oxygen availability, and growth factor signalling. Under conditions in which amino acids are not abun- dant, mTORC1 dissociates from the membrane and is inactive. The energy-​requiring acidifying molecule, vacuolar (V)-​ATPase, which pumps protons into the lysosome, is also part of the assembly and may be influenced by amino acid abundance. The condition of X-​ linked myopathy with excess autophagy illustrates the role of the nutrient-​sensing feedback loop that controls autophagy; defects in a chaperone, VMA21, that assembles the V-​ATPase give rise to inadequate lysosomal acidification and reduced release of nutrient amino acids, thus leading to upregulated autophagic processes. Autophagy retrieves the basic building blocks of cellular compo- nents and proceeds hand-​in-​hand with de novo synthesis and the re- newal of intracellular compartments throughout life; as summarized earlier, the process is stimulated under conditions of starvation and disuse—​for example, in immobilized muscles or during involution of the anterior pituitary or mammary gland after pregnancy and lac- tation. When starvation is prolonged, macroautophagy slows down in favour of the lysosomal uptake of a class of large cellular proteins harbouring particular amino acid sequences which are recognized by receptors that mediate import into the organelle. The intrinsic and highly glycosylated lysosomal membrane protein, LAMP2, is implicated in the uptake of such cytosolic proteins. LAMP2 is mu- tated in the X-​linked disorder Danon’s disease in which liver as well as cardiac and skeletal muscle show prominent vacuoles engorged with glycogen and other debris, including remnants of cellular organelles. Common regulation of autophagy, lysosomal biogenesis, and lysosomal function A common regulatory system based on a transcription factor, TFEB, upregulates many of the processes required for integrated function of autophagy: formation of vesicles, production of factors necessary for cargo recognition and fusion of the autophagosome and lyso- some, synthesis of lysosomal enzymes, membrane proteins, and pH pumps. Phosphorylation of TFEB prevents its entry into the nucleus and hence its actions on the transcription of genes harbouring the responsive binding sequences in the CLEAR element. The system is activated by dephosphorylation of TFEB, indirectly after inhib- ition of the master-​regulator mTOR, a kinase that is controlled in part by the effects of low cytoplasmic concentrations of leu- cine and arginine as well as low-​energy charge during periods of starvation. Deficiency of the activity of one lysosomal enzyme usually leads to a coordinated upregulation of many proteins including those en- coded by genes containing CLEAR sequences that coordinate lyso- somal function and autophagy. Degradation and recycling of complex macromolecules Of the 400 or so proteins found in the lysosome, including membrane proteins, there are at least 50 lysosomal hydrolases. These include proteases, glycosidases, sulphatases, phosphatases, nucleases, lipases, and phospholipases. These enzymes require an acidic pH for optimal function, which in the lysosome is maintained at a pH of 4 to 5.5 by the ATP-​dependent proton transporter V-​ATPase. The lysosomal mem- brane and the acidic pH optimum of the hydrolases protect other cell components, at neutral pH, from indiscriminate autodigestion. Lysosomes display a range of pH values, properties that are linked to their position in the living cell (e.g. perinuclear or peripheral). Live-​cell microscopic imaging reveals highly dynamic structures that move rapidly in three dimensions with oscillating acidities that may reflect functional transients which change according to the flux of different substrates. Intralysosomal pH values could influence the rate of substrate hydrolysis according to the differential pH op- timum of particular reactions and thus the exchange of macromol- ecular substrates between distinct organelle populations during the digestive sequence. Targeting of hydrolases About 60 known soluble lysosomal enzymes and activator proteins continually digest and recycle cellular macromolecules. Most of these components are targeted to the lysosome by the mannose 6-​ phosphate cotranslational pathway. Immediately after biosynthesis, N-​linked oligosaccharides on lysosomal proteins are modified spe- cifically to generate the principal mannose 6-​phosphate recognition marker with the addition of a phosphate group on the sixth carbon of mannose residues. The appropriately phosphorylated proteins bind to mannose 6-​phosphate receptors in the trans-​Golgi network and the ligand–​receptor complexes enter clathrin-coated transport vesicles that traffic to an acidic prelysosomal compartment where the ligands dissociate. Soluble lysosomal proteins enter the lyso- somal matrix and most of the receptors recycle back to the trans-​ Golgi network to repeat the intracellular delivery process. About 20% of the mannose 6-​phosphate receptors traffic to the cell surface, where their complement of protein ligands is released but where fresh binding and hence recapture is stochastically possible. This creates the potential for glycoproteins harbouring the appro- priate hexosyl phosphate recognition signal that are discharged into the fluid phase to be distributed and recaptured by cell surface man- nose 6-​phosphate receptors. While this process may have a physio- logical role, for example, in regulating insulin-like growth factor (IGF) concentrations in plasma, because the cation-independent M6P receptor also functions as the IGF2 receptor its greatest signifi- cance is to provide the foundation for functional complementation of lysosomal diseases due to deficiency of soluble proteins (in this example, iduronate sulphatase).

section 12  Metabolic disorders 2124 Activator proteins Activator proteins are required for several lysosomal en- zymes: saposin C is required for the in vivo catalytic function of glucocerebrosidase, deficiency of which causes Gaucher’s disease. Indeed, a Gaucher’s disease-​like phenotype has been observed in the rare individuals with saposin C deficiency. The small protein, GM2 activator, is absolutely required for the digestion of GM2 ganglioside by ß-​hexosaminidase A and deficiency of the GM2 activator causes a disease with clinical features similar to Tay- Sachs disease. Another vivid example of a crucial activator for many lysosomal enzymes is sulphatase-​modifying factor 1 (SUMF1), an enzyme ex- pressed in the endoplasmic reticulum that catalyses the oxidation of a conserved cysteine residue within a protein domain of all known sulphatases, many of which are involved in the breakdown of glyco- saminoglycans and sulphated glycosphingolipids in the lysosome. Homozygous or compound heterozygous mutations in the human SUMF1 gene lead to multiple sulphatase deficiency (Austin’s disease) with features of MPS and metachromatic leukodystrophy as well as very dry skin, ichthyosis, due to defective activity of cutaneous steroid sulphatase. Transporter proteins Transporter proteins of the lysosomal membrane are less well characterized than the acid hydrolases. Maintaining a proton gra- dient with a concentration of protons approximately 1000-​fold greater than that in the ambient cytosolic pH depends on the pres- ence of a V-​type H+ ATPase in the lysosomal membrane. Other membrane transporters exist for the crucial counter-​exchange of ions, including the chloride channel, CLCN7. This channel is ex- pressed in the endosome–​lysosomal membranes of all cells but is abundant on the ruffled border of the osteoclast. Mutations in the CLCN7 gene are responsible for dominantly inherited and se- vere autosomal recessive forms of osteopetrosis due to a failure to acidify the subosteoclastic vacuole, an externalized lysosomal as- sembly of enzymes critical for modelling bone matrix during os- sification and maintenance of skeletal integrity throughout life. In the recessive variants, mutations of CLCN7 are also responsible for neurodegeneration. Other channels include those that transport Ca2+ (LAAT1), as well as the egress of small molecules generated by hydrolytic digestion of the macromolecular substrates in the lysosome—​ for example, cystinosin (cystine), CblF (cobalamin), and sialin (sialic acid) which are implicated respectively in cystinosis, methylmalonic acidaemia due to defective intracellular vitamin B12 transport, and Salla disease. Transporters are required to ex- port the products of lysosomal digestion, such as amino acids, monosaccharides, nucleosides, and ions, for reuse in cellular me- tabolism. Recent research has identified mucolipin as a channel for monovalent cations; impaired function of which causes mucolipidosis IV. The protein CLN3, deficiency of which results in a juvenile type of neuronal ceroid lipofuscinosis type 3 (‘Batten’s disease’), has been implicated in the egress of arginine. Antigen presentation Proteases of lysosomal origin, particularly the cysteine proteinases or cathepsins, are responsible for the cleavage of endocytosed protein antigens to generate peptide fragments. In antigen-​presenting cells, where abundant expression of several of these proteases is necessary, peptide fragments of the cognate antigens are presented in associ- ation with major histocompatibility complex (MHC) class II mol- ecules as a key step in the pathway that orchestrates the adaptive immune response. In mouse models of autoimmunity, deficiency of acid proteases—​such as cathepsins B and S, generated experimen- tally by gene disruption technology—​can ameliorate many of the disease manifestations. Definition and classification of lysosomal diseases Definition Lysosomal diseases result from inherited defects in lysosomal hydrolases and the mechanisms for delivering them to the organelle, lysosomal enzyme activators and cofactors, lysosomal membrane proteins, and carrier systems for the transport of the substrates and products of lysosomal digestion between the organelle and the cyto- plasm. Most of the enzymatic defects are restricted to the activity of a single hydrolase, but defects of activators and cofactors, as well as proteins involved in the processing of nascent lysosomal enzymes for organellar delivery, can lead to generalized defects of lysosome function. Classification Lysosomal disorders have been classified according to the nature of the primary storage compounds (biochemical classification) or ac- cording to the nature of the physiological defect (functional classifi- cation). A combination of two classification systems, incorporating genetic characterization, may allow a clearer description of the pathological basis of the condition. As the clinical manifestations of the 80 or more diseases associ- ated with inborn errors of lysosomal function are very diverse, the reader is referred to specialized literature for further information (see ‘Further reading’). Biochemical classification Sphingolipidoses Sphingolipids are amphiphilic compounds with a lipophilic moiety based on the amino-​alcohol sphingosine (usually linked to a long-​chain fatty acid to form ceramide) and a polar hydrophilic mono-​ or oligosaccharide chain; in sphingomyelins, which are the most abundant sphingolipids, the charged head group is either phosphorylcholine or phosphorylethanolamine. Sphingolipids are found in all plasma membranes and concentrated in large aggre- gates: the lipophilic moiety is often anchored by structural proteins such as tubulin in the lipid bilayer and the carbohydrate element or head group extends into the extracellular space. Sphingolipids mediate diverse cellular functions and serve as spe- cific receptors and cell recognition markers. They are continuously delivered to the lysosomal compartment in the course of mem- brane turnover in endocytosis, phagocytosis, and by autophagy. Deacylated forms of the sphingolipids (β-​d-​glucosylsphingosine, a ‘psychosine’, is the deacylated form of glucosylceramide), are water-​soluble and hence diffusible. Exploration of the roles of

12.8  Lysosomal disease 2125 such lysolipids and other lipid molecules, such as sphingosine 1-​ phosphate and ceramide, in signal transduction and other cellular processes is an expanding field of research and holds promise for a more comprehensive understanding of the molecular pathogenesis of sphingolipid diseases. Mucopolysaccharidoses Mucopolysaccharides, or glycosaminoglycans, are complex linear polysaccharides, composed of repeating units of polar disaccharides which are strongly negatively charged under physiological condi- tions; glycosaminoglycans contain amino sugars substituents and are often sulphated. When associated with a linear core protein, the glycosaminoglycans form even larger three-​dimensional complexes, known as proteoglycans. By virtue of their negative charge and extended structure, proteo- glycans attract water molecules and have important gel-​like proper- ties. Proteoglycans are essential components of ground substance in intercellular spaces and connective tissue, including cartilage, vit- reous humour, and synovial fluid. Lysosomal degradation of the carbohydrate moieties requires the participation of several glycosidases orchestrated in series. Lysosomal diseases associated with the failure to digest heparan sul- phate in particular, such as Hunter’s disease and Sanfilippo’s diseases (MPS II and MPS III subtypes), are all associated with prominent neurological disease. Glycoproteinoses Glycoproteins are proteins to which one or more oligosaccharide chains are attached covalently. The carbohydrate moiety is often branched and complex, mediating specific recognition by cell sur- face receptors. As with the glycosaminoglycans, lysosomal degrad- ation of the carbohydrate element requires the ordered participation of several glycosidases operating in sequence. Deficiency of one glycosidase, in effect, blocks the subsequent release of sugars in the reaction series, thus causing accumulation of oligosaccharides and other complex glycan molecules. Glycogenosis (Pompe’s disease) Deficiency of α1,4-​glucosidase, acid maltase, causes intra-​ and extralysosomal accumulation of glycogen in muscle and other tis- sues. This storage molecule is common to conditions of impaired autophagy (e.g. Danon’s disease) but is prominent in Pompe’s disease and confirms the role of the lysosome in constitutive and continuous remodelling of glycogen macromolecules. Multiple classes of storage material Several lysosomal hydrolases are not specific to one substrate: β-​ galactosidase deficiency, for example, leads to accumulation of a glycosphingolipid (GM1 ganglioside) and glycosaminoglycans (keratan sulphate and β-​galactosyl oligosaccharides) with greatly divergent clinical manifestations. This enzyme deficiency is respon- sible for a range of phenotypes, extending from a predominantly neurodegenerative disease, GM1 gangliosidosis, to the skeletal dis- order of Morquio B (MPS IVB) in which neurological impairment is typically absent. All eukaryotic sulphatases, including the eight sulphatases destined for the mammalian lysosome—​where they catalyse the removal of sulphate moieties from glycosaminoglycans, glycolipids, and glycopeptides—​require the conversion of a conserved cysteine residue to formylglycine for their activation. Genetic defects in the enzyme, sulphatase-​modifying factor (SUMF1), which mediates this conversion in the endoplasmic reticulum, leads to Austin’s dis- ease (multiple sulphatase deficiency)—​a condition which usually resembles late-​infantile metachromatic leukodystrophy but with prominent clinical and biochemical features of a complex MPS as well as ichthyosis, due to an accompanying deficiency of steroid sulphatase. Deficiency in the precursor of the small sphingolipid acti- vator proteins leads to loss of activity of the cognate hydrolases, whose activity depends on interaction with the activators saposins A to D. Depending on the selectivity of the defect, a distinct dis- ease complex, characterized by accumulation of a broad panel of glycosphingolipids, occurs when these proteins are deficient. A genetically and biochemically distinct activator, GM2 acti- vator protein, interacts crucially to form a ternary complex with the two subunits of hexosaminidase A and its specific natural sub- strate in the lysosome, GM2 ganglioside. Deficiency of this activator protein gives rise to a phenocopy of the severe neurodegenerative disorder Tay–​Sachs disease, but characteristically the activity of hexosaminidase A,  when determined with the usual fluorogenic substrates, is unimpaired. A similar phenomenon occurs in the dis- orders due to saposin deficiencies and may render correct diagnosis difficult for the unwary. Most acid hydrolases are glycoproteins that are specifically tar- geted to the lysosomal system through interaction between a mannose-​phosphate moiety and membrane mannose-​phosphate receptors. Failure to generate the mannose-​phosphate signal gives rise to widespread mistargeting of hydrolases and intralysosomal deficiency of the respective activities with consequent accumu- lation of many substrates (I-​cell disease); characteristically, body fluids, including plasma, have increased activities of many lysosomal hydrolases and although the disease shares many features of a MPS, the urine is usually free of glycosaminoglycans. Miscellaneous Please see Table 12.8.1 for details. Functional classification Classification based on the nature of the defect in cell biological terms is particularly relevant in relation to potential therapeutic ap- proaches. Examples of the type of defects listed here are given in the ‘Biochemical classification’ section, in Table 12.8.1, and throughout the text: • Deficiency of a specific acid hydrolase activity • Deficiency of an activator protein • Deficiency of a lysosomal membrane protein or transporter • Abnormal post-​translational modification of lysosomal proteins • Abnormal lysosomal biogenesis Deficiency of an acid hydrolase may come about either as a result of a mutation that reduces catalytic activity, or as a result of a muta- tion that impedes correct folding of the protein and delivery of the mutant enzyme to the lysosome. This has implications for pathogen- esis and potential therapeutic approaches.

section 12  Metabolic disorders 2126 Table 12.8.1  Representative classification of lysosomal diseases Disease Synonym OMIM Locus, gene Gene product and functional classification Storage material Sphingolipidoses Farber Lipogranulomatosis 228000 8p22 ASAH Acid ceramidase Acid hydrolase Cer Fabry Anderson–​Fabry 301500 Xq22 GLA α-​Galactosidase A Acid hydrolase Gb3 Gaucher Glucosylceramidosis 606463 230900 231000 230800 1q21 GBA Glucocerebrosidase Acid hydrolase GlcCer GM1 gangliosidosis 230500 230600 3p21 GLB1 β-​Galactosidase Acid hydrolase GM1 Tay–​Sachs GM2-​gangliosidosis B 272800 15q23 HEXA β-​Hexosaminidase α-​subunit Acid hydrolase GM2 Sandhoff GM2-​gangliosidosis O 268800 5q13 HEXB β-​Hexosaminidase β-​subunit Acid hydrolase GM2 Tay–​Sachs AB variant GM2 gangliosidosis AB 272750 5q32 GM2A GM2 activator protein Activator protein GM2 Krabbe Globoid cell leukodystrophy 245200 14q31 GALC β-​Galactosylceramidase Acid hydrolase GalCer Metachromatic leukodystrophy Arylsulphatase A deficiency 250100 22q13 ARSA Arylsulphatase A Acid hydrolase Sulphatide Prosaposin deficiency 176801 10q22 PSAP Prosaposin Activator protein Multiple lipids Saposin B deficiency Metachromatic leukodystrophy variant 249900 10q22 PSAP Saposin B Activator protein Sulphatide Saposin C deficiency Gaucher variant 610539 10q22 PSAP Saposin C Activator protein GlcCer Niemann–​Pick types A and B 257200 607616 11p15 SPMPD1 Acid sphingomyelinase Acid hydrolase SM Other lipidoses Niemann–​Pick type C1 257220 18q11 NPC1 NPC1 Probable transmembrane transporter Cholesterol, GSL Niemann–​Pick type C2 607625 14q24 NPC2 NPC2 Soluble transporter Cholesterol, GSL Wolman Cholesteryl ester storage disease 278000 10q23.2 LIPA Acid lipase Acid hydrolase Cholesteryl esters Mucopolysaccharidoses (MPS) MPS I Hurler Hurler/​Scheie, Scheie 607015 (MPS IH) 607015 (MPS IHS) 607016 (MPS IS) 4p16 IDUA α-​Iduronidase Acid hydrolase DS, HS MPS II Hunter 309900 Xq28 IDS Iduronate sulphatase Acid hydrolase DS, HS MPS IIIA Sanfilippo A 52900 17q25 SGS Heparan N-​sulphatase Acid hydrolase HS MPS IIIB Sanfilippo B 252910 17q21 NAGLU N-​acetyl glucosaminidase Acid hydrolase HS MPS IIIC Sanfilippo C 252930 8p11 TMEM76 HGSNAT α-​Glucosaminide acetyl-​CoA transferase Acid hydrolase HS MPS IIID Sanfilippo D 252940 12q14 GNS N-​acetylglucosamine 6-​sulphatase Acid hydrolase HS MPS IVA Morquio A 253000 16q24 GALNS Galactosamine 6-​sulphatase Acid hydrolase KS,CS MPS IVB Morquio B 253010 3p21 GLB1 Acid β-​galactosidase Acid hydrolase KS

12.8  Lysosomal disease 2127 Disease Synonym OMIM Locus, gene Gene product and functional classification Storage material MPS VI Maroteaux–​Lamy 253200 5q12 ARSB Arylsulphatase B N-​Acetyl galactosamine 4-​sulphatase Acid hydrolase DS MPS VII Sly 253220 7q21 GUSB Glucuronidase Acid hydrolase DS, HS, CS MPS IX Hyaluronidase deficiency 601492 3p21 HYL1 Hyaluronidase 1 Acid hydrolase HA Glycogen storage disease Pompe Glycogen storage disease type II 232300 17q25 GAA α-​Glucosidase Acid hydrolase Glycogen Multiple substrate storage due to single gene defects Multiple sulphatase deficiency Austin disease 272200 3p26 SUMF1 Formyl-​glycine generating enzyme Post-​translational modification Sulphatide, mucopolysaccharides Galactosialidosis 256540 20q13 PPCA Protective protein Cathepsin A Acid hydrolase GSL, polysaccharides Mucolipidosis type II I-​cell disease 252500 12q23 GNPTAB UDP-​GlcNac Phospho transferase α/​β unit Post-​translational modification Multiple lipids and oligosaccharides Mucolipidosis type IIIa Classic pseudo-​Hurler’s polydystrophy 252600 12q23 GNPTAB UDP-​G1cNac Phosphotransferase α/​β unit Post-​translational modification Multiple lipids and oligosaccharides Mucolipidosis type IIIb Pseudo-​Hurler’s polydystrophy 352605 16p GNPTG UDP-​G1cNac Phosphotransferase γ-​subunit Post-​translational modification Multiple lipids and oligosaccharides Mucolipidosis type IV 252650 19p13 MCOLN1 Mucolipin-​1 cation channel Transporter Multiple lipids and oligosaccharides VP33A deficiency Mucopolysaccharidosis plus syndrome Turks and Yakut 610034 12q24.31 VP33A is a common component of the heterohexameric tethering complexes involved in endocytic-​lysosomal fusion: HOPS (homotypic fusion and vacuole protein sorting) and the Rab GTPase-​dependent CORVET (class C core vacuole/​ endosome tethering) Glycosaminoglycans, especially heparan sulphate; increased acidification Glycoproteinoses Aspartylglucosaminuria 208400 4q32 AGA Glycosyl-​asparaginase Acid hydrolase Aspartyl-​glucosamine Fucosidosis 230000 1p34 FUCA α-​Fucosidase Acid hydrolase Oligosaccharides α-​Mannosidosis 248500 19q12 MAN2B1 α-​Mannosidase Acid hydrolase Oligosaccharides β-​Mannosidosis 248510 4q22 MANBA β-​Mannosidase Acid hydrolase Oligosaccharides Sialidosis Sialidase deficiency Mucolipidosis type 1 256550 6p21 NEU1 α-​Sialidase (neuraminidase) Acid hydrolase Oligosaccharides Schindler NAGA deficiency Kanzaki’s disease 609242 609241 22q13 NAGA α-​N-​acetyl galactosaminidase Acid hydrolase Oligosaccharides Lysosomal transport defects Cystinosis Abderhalden–​Kaufmann–​ Lignac disease 219800 219900 219750 17p13 CTNS Cystinosin, a transmembrane protein Cystine efflux Transporter Cystine Methylmalonic aciduria Vitamin B12 lysosomal release defect 277380 Unknown Vitamin B12 carrier (Cb1F) Transporter Vitamin B12 Table 12.8.1  Continued (continued)

section 12  Metabolic disorders 2128 Disease Synonym OMIM Locus, gene Gene product and functional classification Storage material Salla Sialuria 604322 6q14 SLC17A5 Sialin Transporter Sialic acid Lysosomal protease defect Pycnodysostosis 265800 1q21 CTSK Cathepsin K Acid hydrolase Collagen fibrils (osteoclasts) Autophagy defects (with glycogenosis) Danon Pseudoglycogenosis II 300257 Xq24 LAMP2 LAMP2 Membrane protein Vacuoles, macromolecular debris, glycogen, and membrane RNASET2 Leukoencephalopathy, cystic without megalencephaly; RNase T2-​deficient leukoencephalopathy 612951 6q27 Ribonuclease T2 deficiency (glycoprotein with endoribonuclease activity and acid optimum) Unknown in humans: but implicated in reticulophagy (endoplasmic reticulum autophagy)—​authentic zebrafish model shows accumulated rRNA in neurons X-​linked myopathy with excessive autophagy 310440 Xq28 XMEA VMA21 Membrane protein Vacuole Vacuolar myopathy Muscular dystrophy with vacuoles 601846 19p13 MDRV MDRV Unknown function Vacuoles Autophagy defects X-​linked vacuolar myopathy with excessive autophagy (XMEA) MEAX 310440 Xq28 VMA 21, chaperone for the lysosomal V-​ATPase. Vacuoles Neuronal ceroid lipofuscinosis (NCL) CLN1 (infantile, late-​ infantile, juvenile, adults) Haltia–​Santavuori 256730 1p32 CLN1/​PPT1 PPT1 palmitoyl protein thioesterase 1 Acid hydrolase SAPs (sphingolipid activator proteins) CLN2 (late infantile/​juvenile) Jansky–​Bielchowsky 204500 11p15 CLN2/​TPP1 TPP1 tripeptidyl peptidase 1 Acid hydrolase SCMAS (subunit c mitochondrial ATP synthase) CLN3 (juvenile) Spielmeyer–​Sjögren Batten 204200 16p12 CLN3 CLN3 (Battenin) transmembrane protein—​ endosomes, lysosomes as well as synaptic vesicles SCMAS CLN4A (adult) Kufs type A—​adult (autosomal recessive) 204300 15q23 CLN6 CLN6 Membrane transporter SCMAS CLN4B Kufs, Parry disease (autosomal dominant) 162350 20q13.33 (DNAJC5) Soluble, cytoplasmic cysteine string protein alpha—​ palmitoylation reduced SAPs CLN5 Finnish (late infantile, juvenile, adult) 256731 13q22.3 CLN5 CLN5—​a complex but soluble glycoprotein SCMAS CLN6 Lake–​Cavanagh Late infantile, adult—​Kufs type A variant 601780 15q23 CLN6 CLN6 Transmembrane protein possible ER transporter SCMAS CLN7 vLINCL (Turkish and others) 610951 4q28.2 MSFD8 CLN7/​MFSD8 Transmembrane protein SCMAS CLN8 Northern epilepsy 600143 8p23 CLN8 CLN8 Transmembrane protein transfers lysosomal enzymes from Golgi to endoplasmic reticulum SCMAS [CLN9] Obsolete Original Serbian-​German pedigree reclassified as CLN5 after genetic studies 609055 (obsolete) N/​A N/​A N/​A CLN10 Congenital with microcephaly, neonatal and late infantile CLN (very rare juvenile and adult variants) 610127 11p15 CTSD Cathepsin D Acid hydrolase SAPs Table 12.8.1  Continued

12.8  Lysosomal disease 2129 Disease Synonym OMIM Locus, gene Gene product and functional classification Storage material CLN11 Autosomal recessive (note: heterozygotes with GRN mutations develop autosomal dominant frontotemporal dementia with TDP43-​inclusions—​ OMIM 607485) 614706 17q21.31 GRN Progranulin Intraneuronal ubiquitin-​positive autofluorescent lipofuscin and rectilinear inclusions on electron microscopy CLN12 Juvenile (note: recessive mutations in ATP13A2 also cause Kufor–​Rakeb syndrome, PARK9, juvenile-​ onset atypical Parkinson’s disease with supranuclear gaze palsy, spasticity and dementia (OMIM 606693)) 610513 1p36.13 CLN12/​ATP13A2 ATPase type 13A2 Lysosomal membrane protein CLN13 Adult CLN (sometimes known as Kufs B—​autosomal recessive) 615362 11q13.2 CTSF Cathepsin F Soluble acid hydrolase Autofluorescent material CLN14 Infantile—​progressive myoclonic epilepsy 611725 7q11.21 CLN14 (KCTD7) Potassium channel tetramerization domain containing protein 7—​a putative cytoplasmic regulator of potassium conductance Fingerprint or rectilinear as well as osmiophilic deposits by electron microscopy Disorders in extended lysosomal apparatus (melanosomes, lamellar bodies) Chédiak–​Higashi 214500 1q42 LYST LYST Biogenesis defect Enlarged vacuoles Melanosomes MYOV Griscelli type 1 214450 15q21 MYO5A Myosin 5A Biogenesis defect Melanin granules RAB27A Griscelli type 2 603868 15q21 RAB27A RAB27A Biogenesis defect Melanin granules Melanophilin Griscelli type 3 609227 2q37 MLPH Melanophilin Biogenesis defect Melanin granules HPS-​1 Hermansky–​Pudlak type 1 604982 10q23 HPS1 HPS-​1 Biogenesis defect Multiple vacuoles HPS-​2 Hermansky–​Pudlak type 2 608233 5q14 AP3B1 AP3 β-​subunit Biogenesis defect Multiple vacuoles HPS-​3 Hermansky–​Pudlak type 3 606118 3q24 HPS3 HPS-​3 Biogenesis defect Multiple vacuoles HPS-​4 Hermansky–​Pudlak type 4 606682 22q11 HPS4 HPS-​4 Biogenesis defect Multiple vacuoles HPS-​5 Hermansky–​Pudlak type 5 607521 11p15 HPS5 HPS-​5 Biogenesis defect Multiple vacuoles Biogenesis defect Multiple vacuoles HPS-​6 Hermansky–​Pudlak type 6 607522 10q24 HPS6 HPS 6 Biogenesis defect HPS-​7 Hermansky–​Pudlak type 7 607145 6p22 DTNB1 Dysbindin Biogenesis defect Multiple vacuoles HPS-​8 Hermansky–​Pudlak type 8 609762 19q13 BLOC1S3 BLOC1S3 Biogenesis defect Multiple vacuoles HPS-​9 Hermansky–​Pudlak type 9 614171 15q1.21 BLOC1S6 Pallidin Biogenesis defect HPS-​10 Hermansky–​Pudlak type 10 617050 19p3.3 AP3D1 Biogenesis defect Surfactant metabolism dysfunction-​4 SMPD3 610921 16p13 ABCA3 ABCA3 transporter Alveolar proteins Congenital and lamellar ichthyosis Harlequin fetus 242500 601277 2q34 ABCA12 ABCA12 transporter Abnormal keratin Cer, ceramide; CS, chondroitin sulphate; DS, dermatan sulphate; Gb3, globotriaosylceramide; GlcCer, glucosylceramide; HA, hyaluronan; HS, heparan sulphate; KS, keratan sulphate; SM, sphingomyelin. Table 12.8.1  Continued

section 12  Metabolic disorders 2130 Pathophysiology Lysosomal storage of primary substrates In the initial period of discovery of lysosomal diseases, limited avail- ability of investigative tools influenced how the conditions were viewed. Pathological examination and microscopy showed not only organ enlargement but intracellular ‘storage’ bodies of remarkable appearance. Biochemical characterization of storage compounds guided research into discovery of enzyme activities and gene prod- ucts. This led to a somewhat simplistic view of the pathogenesis of lysosomal storage diseases as being related to the expansion of cells and organs containing relatively inert ‘storage’ material. In most cases, the contribution of storage material to organ enlargement is quantitatively very small and the widespread effects of lysosomal diseases on cell–​cell interactions with paracrine, inflammatory, and immunological consequences, remain unsolved. Latterly, the patho- genesis of lysosomal diseases in relation to so-​called secondary metabolites and their specific roles in signalling and cell biology is receiving attention; investigations are also underway into the effects of storage on membrane flow and trafficking within the cell. Although the amount of storage material that accumulates within lysosomes in the lysosomal diseases may be several hundred-​ or thousand-​fold greater than normal, the absolute amount of material may amount to only a few grams, even in an enlarged viscus such as the spleen, which may exceed 5 kg in some disorders. For example, the presence of a few grams of the sphingolipid, sphingomyelin, in Niemann–​Pick disease types A and B, is associated with massive vis- ceral enlargement with accompanying inflammatory, ischaemic, and other destructive changes due to the presence of storage cells. Similarly, marked pathological injury occurs: in the viscera and bone marrow spaces of patients with Gaucher’s disease; in the heart and skeletal muscles of patients with α-​glucosidase deficiency (with glycogen ac- cumulation in the sarcoplasm of striated and cardiac myocytes); in the kidney and heart of patients with Fabry’s disease; and affecting neurons throughout the nervous system of patients with ceroid neur- onal lipofuscinosis, Tay–​Sachs disease, and GM1 gangliosidosis. Secondary metabolites and their effects Although there often appears to be an anatomical relationship be- tween the extent of lysosomal storage and the development of overt disease in a particular organ, at present there is little mechanistic understanding of this relationship in molecular terms. Sphingolipids participate in cell recognition events and receptor biology; sphingo- lipid metabolites (the deacylated lysosphingolipids) also function as signalling molecules in apoptotic and proliferative responses. However, in two striking instances (Gaucher’s and Krabbe’s dis- eases, due to acid β-​glucocerebrosidase and β-​galactocerebrosidase deficiencies, respectively), multinucleated macrophages resem- bling the pathognomonic Gaucher’s or globoid cell can be induced in vitro by the water-​soluble molecules glucosylsphingosine and galactosylsphingosine (psychosine), which are overproduced in these diseases. At pathophysiological concentrations in culture, psychosines and related glycolipids inhibit cytokinesis and for this reason are im- plicated in the molecular pathways responsible for the presence of large macrophage-​like cells with several nuclei in the brain of patients with Krabbe’s disease (‘globoid cells’) and in the marrow, spleen, and other viscera in Gaucher’s disease (‘Gaucher cells’). Psychosines (glucosylsphingosine and galactosylsphingosine) ap- pear to interact with receptors on the plasma membrane of human monocytic-​lineage cells. Latterly, multiple myeloma, which occurs at a greatly increased frequency in Gaucher’s disease, has been reported to result from a T-​follicular helper type 2 (TFH2)-​mediated B-​cell prolif- erative response directed against the pathological glycosphingolipids present in serum (especially ß-​glucosylsphingosine), with clonal para- protein antibody reactivity directed against this circulating lipid mol- ecule in patients with Gaucher’s disease. Up to one-​third of patients with spontaneous monoclonal gammopathy apparently unrelated to Gaucher’s disease have been reported to have specific antibody reactivity to ß-​glucosylsphingosine or other sphingolipids such as lysophosphorylcholine. Another deacylated glycosphingolipid, lyso-​globotriaosylceramide (lyso-​GB3), has been identified in the plasma of patients with Fabry’s disease. This induces smooth muscle proliferation and is thus impli- cated in the severe systemic and cerebrovascular manifestations that characterize the condition. These biochemical findings may signify new approaches to the understanding of several lysosomal disorders associated with cell loss due to apoptosis and fibroinflammatory responses; since lysolipids diffuse readily from their site of formation, different ap- proaches to their treatment other than targeted enzyme replacement may be appropriate. Cellular effects The cellular reaction associated with lysosomal storage is often re- stricted and stereotypical. In neural tissue, several pathological hall- marks such as meganeurite formation and ectopic dendritogenesis accompany the accumulation of a wide assortment of storage com- pounds. It appears that lysosomal storage gives rise to a general- ized defect in the complex traffic flow mediated by the endosomal system with effects on autophagy, signal propagation at the synapse, axonal transport, myelin formation, and arborization of dendrites. Lipid rafts, detergent-​resistant islands within the plasma membrane, contain high local concentrations of gangliosides and play an im- portant role in many cell signalling events. Disordered lateral move- ment and recycling of raft components as part of a general or specific endosomal ‘traffic jam’ may have profound effects on cell signalling, as well as recycling processes mediated by autophagy. As an example of such cellular cascades, Niemann–​Pick C1 dis- ease has been linked to a sequence of events in which sphingosine accumulation causes depletion of acidic compartment calcium stores, leading to accumulation of various lysosomal lipids including cholesterol. Disorders in which autophagy is markedly disturbed, particularly those affecting the nervous system, can lead to mito- chondrial dysfunction as a result of impaired clearance of effete and damaged organelles (defective mitophagy). The causative link between Gaucher’s disease and Parkinson’s disease remains to be established, but one area of active research implicates the overproduction of mutant protein which then fails to fold correctly, accumulating within the endoplasmic reticulum, activating the unfolded protein response, the final effects of which may include apoptosis. Individuals who are heterozygous for pathological GBA1 mutations not only develop parkinsonian fea- tures but also Lewy body dementia: in such patients, the Lewy bodies costain for α-​synuclein, the mutant ß-​glucosylceramidase and ubiquitin.

12.8  Lysosomal disease 2131 Tissue and organ malfunction In a scientific era that offers powerful analytical techniques to ex- plore complex functional networks which lead to tissue pathology, the lysosomal diseases represent a promising field for investiga- tion using large-​scale, high-​throughput methods to investigate al- tered protein and gene expression in the context of cell signalling responses. An early application of this work has been the use of authentic experimental models of some of the more severe storage diseases generated by gene knockout technology, which facilitate re- search on otherwise inaccessible tissues such as the brain during the development of the storage phenotype. Gene expression profiling experiments conducted during periods of neuronal cell death have shown upregulation of genes related to the inflammatory process in the nervous system of mice that serve as a model of GM2 gangliosidosis. The activation of local microglia is shown by the signature of upregulated macrophage expression markers and lymphocyte chemoattractants, as well as genes encoding antigen-​presenting MHC class II molecules. Since Krabbe’s disease modelled in mice is mitigated by bone marrow transplantation, which supplies a population of genetically competent immune cells (and which is accompanied by the use of powerful immunosuppres- sion), it seems probable that the altered immunity accompanying bone marrow transplantation may itself modify the clinical expres- sion of lysosomal diseases affecting the brain, and such an effect may be independent of the storage material. Several indirect studies have indicated the release of inflamma- tory cytokines in at least one lysosomal storage disease (Gaucher’s disease), which may explain the metabolic and plasma protein ab- normalities associated with a sustained inflammatory response that characterizes the clinical syndrome. Hypertrophy and fibrosis char- acterize the organ responses in many lysosomal diseases. Whether the response is mediated at a cellular, paracrine, or endocrine level remains unclear. In Fabry’s disease, lyso-​Gb3 is a promising candi- date for an elusive, endocrine-​like factor in the cardiac and vascular aspects of this condition. The clinical presentation of malfunctioning organs in lysosomal storage disorders generally resembles the pathological outcomes of hypertrophy, fibrosis, and organ failure observed in other chronic conditions. Neurological syndromes vary with the anatomical site of greatest injury and with the relative involvement of grey or white matter (neuronopathic or myelination defects). Clinical presentation Natural course and severity range All lysosomal diseases disturb the catabolism of complex molecules in numerous tissues and their manifestations are usually progres- sive and permanent. They show no relationship to food intake and are generally independent of intercurrent illness. The rate of de- terioration depends in part on the degree of residual activity of the deficient enzyme or process: subtotal deficiencies present early in childhood with rapid evolution of disease; partial deficiencies emerge more slowly and often present in later childhood or adult life. The disease may be insidious, as in the indolent splenomegaly of adults with Gaucher’s disease, the renal impairment of Fabry’s dis- ease, or the muscle weakness of adult-​onset Pompe’s disease. Acute episodes may punctuate this process, giving rise to a step-​wise im- pairment of function, such as occurs with the osteonecrosis that typically affects the epiphyses of the long bones in Niemann–​Pick disease type B or Gaucher’s disease. Organomegaly and disturbed visceral function Those disorders that affect metabolically active organs such as the liver and kidney often cause functional impairment, including the manifestations of liver failure, portal hypertension, and—​in the case of the kidney—​rickets and metabolic acidosis, for example, as a con- sequence of Fanconi’s syndrome in cystinosis. Cardiac involvement leads to hypertrophy, diastolic dysfunction, conduction and rhythm disturbances, as well as thickening of the valves. Respiratory mani- festations of the mucopolysaccharidoses are usually first evident as a result of narrowing of large airways, but restricted ventilation due to skeletal disease often supervenes. Splenomegaly, complicated by marked functional hypersplenism, is characteristic of untreated Gaucher’s disease. Skeletal manifestations Skeletal effects predominate and are particularly cruel in several of the mucopolysaccharidoses. Severe growth retardation, joint stiffness, and atlantoaxial instability impair the quality and dur- ation of life. In Gaucher’s disease, diverse osseous manifestations include marrow infiltration, osteoporosis, osteonecrosis, lytic ­lesions, pathological fractures, and occasional plasmacytoma or frank myelomatosis (Fig. 12.8.1). Neurological features Lysosomal diseases are a prominent cause of progressive neuro- logical and mental deterioration in patients whose disease starts during adolescence up to mature adult life, and they should always be considered in the differential diagnosis of such presentations. Ataxia is a feature of GM1 and GM2 gangliosidoses, and a flaccid paraparesis in young children might suggest meta- chromatic leukodystrophy. Widespread white-​matter disease in association with frontal dementia and spastic paraparesis is a char- acteristic presentation of juvenile and adult forms of metachrom- atic leukodystrophy and Krabbe’s disease. Polyneuropathy and pyramidal signs are superimposed in both disorders. Early-​onset leukodystrophy is caused by metachromatic leukodystrophy, mul- tiple sulphatase deficiency, and Krabbe’s disease. The latter is a rare but important diagnostic entity in this group since the disease may be partially ameliorated by allogeneic marrow transplantation in very early life. Lysosomal diseases with prominent neurological manifestations are often associated with progressive mental deterioration, with or without the onset of spasticity, myoclonic seizures, and optic atrophy. Extrapyramidal signs including parkinsonism, athetoid movements, and dystonia are also frequent. Corneal opacities suggest cystinosis, I-​cell disease, mucopoly­ saccharidoses, mannosidosis, Fabry’s disease, and galactosialidosis, as well as one form of Gaucher’s disease with neuronopathic features (the D409H type IIIc variant). Perifoveal pallor with the appearance of pigmentation in the macula (the ‘cherry-​red spot’ in white per- sons) is a hallmark of Tay–​Sachs disease and other gangliosidoses affecting infants and young adults. Specific syndromes are described in later sections.

section 12  Metabolic disorders 2132 Diagnosis Clinical suspicion and family history Even in the critically ill, it is essential (whenever possible) to estab- lish the definitive diagnosis where a lysosomal disease is suspected, for several reasons: (1) specific treatment may be possible—​enzyme replacement therapy, bone marrow transplantation, or even oral substrate reduction and enzyme enhancement therapies may be available; (2) these disorders are inherited either as X-​linked or as autosomal recessive traits, which has important consequences for reproductive choice in other family members; and (3) the diagnosis may clarify unexplained symptoms in at-​risk relatives. The key to making the diagnosis of a rare lysosomal disease is often informed suspicion combined with dogged persistence. In most circumstances, once suspected, the lysosomal disease can be identified with relative ease by referral to a specialized regional ref- erence laboratory for the diagnosis of metabolic disorders: senior laboratory staff will usually advise about the handling of appro- priate tissue material for diagnostic studies, including the means for securing a genetic diagnostic by molecular analysis of genomic DNA (see ‘Molecular diagnosis’). In the first instance, simple histochemical stains of existing biopsy material and examination of urine metabolites, including lipids and oligosaccharides, may narrow down the diagnosis. More commonly, specific enzymatic assays are used. These are generally carried out on leucocytes isolated from fresh heparinized blood samples, or on fibroblasts cultured from small biopsy specimens of skin; the latter are particularly valuable since, once established, fibroblast cultures can be stored indefinitely for repeated and definitive study. Fabry’s disease, Niemann–​Pick disease types B and C, as well as Gaucher’s disease have often come to light in young or adult patients with particular syndromic presentations. Apart from paediatricians, general physicians, haematologists, nephrologists, neurologists, gastroenterologists and hepatologists, dermatologists, and even orthopaedic surgeons may be the first to evaluate the patient—​all of whom should be able to identify the condition by following diag- nostic pathways appropriate to their specialty. In any event, the diag- nosis of lysosomal storage diseases is rarely difficult, provided the expertise of trusted laboratory services is available for the conduct of biochemical assays, diagnostic DNA studies, and wide-​ranging histopathological examination. The value of good communication between laboratory staff and clinical investigators, to whom these patients are referred, cannot be overestimated. Pattern of inheritance A detailed family history, taking care to investigate the extended pedigree, is of critical importance for these high-​penetrance mono- genic diseases. Most lysosomal diseases are inherited as autosomal recessive traits and patients have biallelic mutations at the locus pri- marily implicated. The differential diagnosis narrows substantially if there is evidence of X-​linked inheritance characteristic of Fabry’s disease, MPS II (Hunter’s disease), or Danon’s disease. At-​risk family members will be identified, often living with presymptomatic or undiagnosed disease. Tracing family members is best carried out sensitively in cooperation with professional genetic counselling services and with the general oversight of patient representative organizations. Particular populations While the possibility of frank consanguinity should be explored (usu- ally from pairings between cousins), it is helpful to be aware of rare diseases that are over-​represented in small populations with high endogamy. Examples include the high frequency of Hermansky–​ Pudlak syndromes 1 and 3 among Puerto Ricans and certain isolated Swiss communities. Lysosomal diseases are also more frequent than (a) (b) Fig. 12.8.1  (a) A 35-​year-​old male with α-​mannosidosis. (b) A 20-​year-​old male with α-​ mannosidosis and coincident glutaric aciduria type 1. Facial appearance includes prominent brow, depressed nasal bridge, and prognathism. Note hearing aids.

12.8  Lysosomal disease 2133 expected in Ashkenazi Jews, including Hermansky–​Pudlak type 3, Niemann–​Pick disease (type A), and type 1 (non-​neuronopathic) Gaucher’s disease. Type 3 (chronic neuronopathic) Gaucher’s dis- ease is frequent in parts of arctic Sweden and in one caste group in Pakistan, among which several hundred patients have been identified. The high frequency of several mutant alleles of the HEXA gene re- sponsible for Tay–​Sachs disease in Ashkenazi Jews has led to greatly enhanced awareness of the disorder in this population, with suc- cessful international programmes for carrier detection. The birth of infants soon to be affected by Tay–​Sachs disease is now exceptional in the Ashkenazim, but infantile as well as attenuated, late-​onset forms of the disease occur with a high frequency in isolated popu- lations such as Moroccan Jews, French settlers in Quebec (Canada), the Cajun people of Louisiana (United States of America), and the Canadian Metis Indian population. Sandhoff’s disease, due to mutations in the HEXB gene and the other major form of GM2 gangliosidosis, is not over-​represented in Jews but occurs at high frequency in the aboriginal population in the district of Cordoba (Argentina) and in the Arabic followers of the Syriac Maronite Church in Lebanon, Syria, and Northern Cyprus. Infantile Krabbe’s disease occurs with a very high carrier fre- quency among the Druze people of Northern Israel and Lebanon and appears to be more frequent in Scandinavian countries and other parts of Northern Europe but an adult variant arising from a single missense mutation in the GALC gene is occurs at high fre- quency Catania, Italy. Radiology Ultrasonography, MRI, and CT may reveal visceral enlargement and infiltration, for example Niemann–​Pick disease, mucopoly­ saccharidoses, and Gaucher’s disease. Skeletal radiographs may reveal bone expansion in vertebrae and in the phalangeal and long bones, sometimes associated with infarction and collapse, par- ticularly in Niemann–​Pick disease type B and Gaucher’s disease. Echocardiography may reveal thickening and calcification of the cardiac valves (particularly of the aortic ring), infiltration of cardiac muscle causing ventricular hypertrophy in Pompe’s disease, Fabry’s disease, mucopolysaccharidoses I, IV, and VI, and, often strikingly, in Danon’s disease. Neuroradiology is informative, particularly in patients with mucopolysaccharidoses and in Morquio’s syndrome as well as MPS syndromes I, II, and VI where instability of the atlantoaxial joint may cause fatal subluxation as a consequence of connective tissue disease abutting the dens. MRI of the cervical spine in MPS is critical for assessing when to carry out posterior fusion to stabilize the joint. Similarly, investigations of the lower spine may determine the cause of progressive spinal deformity due to lumbar kyphosis and assist in the evaluation of the need for surgery. MRI of the brain is invaluable in the assessment of dementing illnesses: cortical and/​or white matter disease may be delineated. Imaging plays a key part in diagnosis of the striking white matter changes of Krabbe’s disease, multiple sulphatase deficiency, and metachromatic leukodystrophy (Fig. 12.8.2). Extensive white matter lesions and eventual cerebral atrophy also characterize the advanced stage of the neurological aspects of Fabry’s disease (Fig. 12.8.3). Pathology Although lysosomal defects occur in all tissues, the principal focus of each disease is manifest in those tissues with the most rapid turn- over of the parent macromolecule of which degradation is impaired. For example, in Gaucher’s disease the turnover of parent glycolipids appears to be greatest in the mononuclear phagocytes. Here the ac- cumulation of glycolipids derived from the breakdown of complex sphingolipids present in white cell and red cell membranes present in the formed blood elements occurs. With mild or moderate im- pairment of the responsible enzyme, glucocerebrosidase (acid ß-​ glucosylceramidase), the pathology is restricted principally to the macrophage-​containing tissues of the liver, spleen, bone marrow, and (occasionally) the lung. When inherited defects further impair the activity of the enzyme, the nervous system becomes a site of dis- ease: here the main source of accumulating glucosylceramide and glucosylsphingosine is derived from the recycling of the endogenous cellular sphingolipids, particularly gangliosides present in neuronal membranes. Microscopic pathology shows storage within dilated vesicular spaces, which represent diseased lysosomes. Sphingolipids, being amphipathic molecules, tend to accumulate in whorls known as ‘membranous cytoplasmic bodies’ where they assume a lamellar structure within lysosomal spaces. Paracrystalline and crystal- line material in distended lysosomes may also be seen under elec- tron microscopy, for example, in the accumulation of the charged glycolipid, sulphatide, that occurs in metachromatic leukodystrophy Fig. 12.8.2  T2-​weighted MRI of the brain of a young woman with adult-​onset metachromatic leukodystrophy—​psycho-​cognitive variant. Notice the high signal intensity, especially in the frontal white matter and periventricular regions. This patient presented with bizarre behaviour due to a frontal-​type dementia; there were no neurological signs or symptoms. Near total loss of short-​term memory with lack of planning and higher executive functions were prominent features of her illness.

section 12  Metabolic disorders 2134 (arylsulphatase A deficiency). With more water-​soluble substrates, granular material accumulates within the vesicular spaces. These spaces represent distended and often fused lysosomes, filled, for ex- ample, with undegraded glycogen macromolecular complexes in acid maltase deficiency (Pompe’s disease). As emphasized earlier, the pathological manifestations of the lysosomal diseases are diverse. They may range from enlargement of viscera with infiltration by abnormal macrophages containing storage material (foam cells of Niemann–​Pick disease or Gaucher cells) to bone infarction, neuronophagia, vacuolation of renal tubular cells, and diverse tissue infiltrates. Inclusion bodies may be observed in metachromatic-​stained cells of the urine deposit or in circulating neutrophils and lymphocytes (Maroteaux–​Lamy disease, MPS VI); staining with a periodic acid–​Schiff reagent may reveal diastase-​resistant glycolipid storage in the kidney and other organs in Fabry’s disease and other glycosphingolipidoses. The presence of metachromatic storage material in nervous tissue, including periph- eral nerves, is characteristic of the sphingolipidosis, metachromatic leukodystrophy (Fig. 12.8.4). The secondary effects of lysosomal ex- pansion related to upregulation of lysosomal proteins through the TFEB/​CLEAR transcriptional pathway include increased staining for tartrate-​resistant (type 5)  acid phosphatase, hexosamindases and other lysosomal markers, and intrinsic membrane proteins, for example, LAMP1. Ultrastructural examination is often diagnostic for lysosomal dis- eases: membrane-​bound vesicles containing storage material that (a) (c) (b) Fig. 12.8.3  Macroscopic and microscopic appearances of the brain of a patient with Fabry disease. The patient died 15 years after a successful renal transplant, aged 62 years and with a cardiac pacemaker, of a dementing illness having suffered multiple stroke-​like events. (a) CT examination of the brain demonstrating extensive white matter lesions. Cerebral atrophy characterizes the advanced stage of the neurological aspects of Fabry disease; ectopic calcification within the basal ganglia, cerebral cortex, and cerebellum is thought to locate to the media of small penetrating arteries. (b) Postmortem examination demonstrating cortical atrophy, ventricular dilatation, and white matter focal cavitation. (c) Histological examination of the brain demonstrating striking calcification of hypertrophic media of penetrating arteries, with associated leukoencephalopathy. Courtesy of Dr G. Alistair Lammy, Cardiff University.

12.8  Lysosomal disease 2135 may show a crystalline or concentric appearance, or—​in the case of glycogen in Pompe’s disease—​vacuoles with a granular appearance. The appearance of concentric arrays of material strongly suggests a sphingolipidosis. Complete absence of platelet ​dense granules by electron microscopy is characteristic in Hermansky–​Pudlak syndromes. Blood film examination Blood film examination is an often neglected but simple diag- nostic screening procedure that may indicate the diagnosis of a lysosomal disease. Amorphous material accumulates within the lysosomal vacuoles in the ceroid neuronal lipofuscinoses, mucopolysaccharidoses, and glycoproteinoses. In Chédiak–​Higashi disease, giant granules are readily visible in neutrophils, eosinophils, and granulocytes. Smaller pathological bodies may be evident in circulating white blood cells and are typical of several mucopolysaccharide diseases, particularly MPS I (Hurler–​Scheie), MPS VI (Maroteaux–​Lamy), and MPS VII (Sly’s syndrome) diseases in which granular deep lilac-​staining inclusions (the Alder–​Reilly abnormality) are readily detected in all leucocyte subtypes after staining by the Leishman method. Pathological vacuolation is prominent in peripheral blood lymphocytes in ceroid neuronal lipofuscinosis type 3 and Pompe’s disease (the latter in which the vacuoles stain with periodic acid–​Schiff reagent before, but not after, exposure to diastase). Lymphocyte vacuolation is reported in other lysosomal diseases (GM1 gangliosidosis, galactosialidosis, Salla disease, neuramin- idase deficiency, alpha mannosidosis, fucosidosis, I cell disease, and Niemann–​Pick disease type A). Diagnostic biochemistry For most lysosomal storage diseases, the suspected diagnosis can be confirmed by biochemical studies. Storage compounds can, as in the case of the glycoproteinoses and mucopolysaccharidoses, be detected in the urine. Initial colorimetric screening methods may confirm the presence of elevated concentrations of glycosaminogly- cans but are not specific; chromatographic separation of individual glycoconjugates will assist further. More often, the diagnosis is established by confirming reduced or absent activity of the cognate lysosomal acid hydrolase. Specialized laboratories carry out panels of these assays depending on the clinical details provided by the clinician. Most assay systems are based on the cleavage of synthetic fluorescent analogues of the natural substrate in question. Accurate clinical information greatly assists the laboratory in deciding which enzyme activity, of many, to assay. The usual sample is a peripheral blood leucocyte preparation made from whole blood, although fibroblast cultures obtained from skin biopsies or biopsy specimens of other tissues may be required. For some conditions, including Gaucher’s disease, Fabry’s disease Pompe’s disease and lysosomal acid lipase deficiency, accredited la- boratories offer diagnostic assays based on dried blood spots on card. These developments, partly initiated and funded by the pharma- ceutical industry, have the advantage of convenience for transport to the diagnostic lab in a stable form, but the diagnostic performance and reliability compared with conventional fluid-​phase assays car- ried out on fibroblast suspensions or blood leucocytes has yet to be completely established. There is increasing interest in development of newborn screening for lysosomal diseases, but so far this has been limited to a few specific geographic regions, such as Krabbe’s disease in Illinois, Kentucky, Missouri, New York, Ohio, Pennsylvania, and Tennessee (all United States of America) and Taiwan (particularly for Pompe’s disease). The biochemistry laboratory has a further role in determining the presence and concentration of markers of disease activity. Such biomarkers play an increasing role in clinical management, pharmaceutical development, and research. Markers in clinical practice include chitinase, chitotriosidase, and the chemokine CCL18/​PARC as markers of the presence and extent of tissue in- filtration by the eponymous cell in Gaucher’s disease. Particular lipid substrates (including lysolipids), in addition to their puta- tive mechanistic role, are used as potential biomarkers for diag- nosis and monitoring therapeutic effects: the unacylated, partly water-​soluble congeners of the primary storage glycosphingolipid, lysoglobotriasolceramide, and ß-​glucosylsphingosine are in- creasingly used in the clinical monitoring of Fabry’s disease and Gaucher’s disease, respectively. Molecular diagnosis Molecular analysis of genes encoding lysosomal enzymes may often support the enzymatic diagnosis, and may, on occasion, provide a rough prediction about the behaviour of the disease. DNA-​based studies are of particular value for future prenatal diagnosis in a par- ticular pedigree, and for the diagnosis of carrier status in at-​risk fe- males for heterozygosity in the X-​linked diseases such as Hunter’s, Danon’s, and Fabry’s diseases. In the last 20 years, there has been a strong and justified trend in favour of specific enzymatic and genetic diagnoses, rather than those based on the examination of biopsy material by light micros- copy with or without the additional use of special histochemical stains. Ultrastructural examination of biopsy material may be of particular value in recognizing the type of disorder but is rarely cru- cial for a specific diagnosis. Hitherto, histochemical and histopatho- logical methods have led to diagnostic inaccuracies, but it must be admitted that many cases of lysosomal disease—​particularly as they affect adults—​have in the past come to light as a result of bone marrow examinations, liver and muscle biopsies, and other Fig. 12.8.4  Sural nerve biopsy stained with toluidine blue from a patient with metachromatic leukodystrophy. Note the brown-​staining granular material within Schwann cells and perineurial macrophages typical of this disorder due to the deposition of the glycolipid sulphatide. Courtesy of Dr J. Xuereb, Addenbrooke’s Hospital.

section 12  Metabolic disorders 2136 procedures carried out in an attempt to arrive at a diagnosis in an otherwise puzzling condition. High-​throughput diagnostic DNA sequencing is now in use in several laboratories, either in the form of single gene sequencing to address a unitary suspected diagnosis or ‘panels’ of genes associated with a clinical presentation of interest. Research initiatives such as the ‘100,000 Genome Project’ in the United Kingdom aim to enu- merate sequence variants and predict the causative mutations in rare or unexplained metabolic conditions. Whole-​exome sequencing is also gaining traction in the practice of metabolic diseases, especially among paediatricians, but attribution of causation to innumerable sequence variants that emerge from such broad searches remains a considerable logistical challenge and many ‘diagnoses’ cannot be validated. Treatment Supportive and palliative therapy No specific or curative treatment is available for most of the lyso- somal disorders and as a consequence, the psychological and social burdens are pervasive. As discussed earlier, the organ response to the metabolic defect is often stereotypical and similar to that seen in other diseases, with treatment limited to those supportive and pal- liative measures shared with other chronic diseases. Occasionally, organ transplantation is required to deal with heart, liver, or kidney failure. Orthopaedic surgical techniques, such as joint replacement surgery and stabilization of kyphosis using Harrington rods, are frequently required and beneficial. Patients with obstructive hydro- cephalus benefit from the placement of shunts for cerebrospinal fluid. Middle ear effusions and glue ear are also treated convention- ally with grommets. Physiotherapy for restricted joint movement and muscle weakness is valuable. Mobility aids and ventilatory sup- port add to the range of expensive and invasive measures required in the absence of definitive treatment. The search for specific treatments Lysosomal diseases have been the focus of several prominent therapeutic discoveries. The cooperation of informed patient groups, applied medical research funded by government organiza- tions, and the commercial interest of medium-​sized pharmaceut- ical companies has been promoted by the introduction of Orphan Drug legislation. First enacted in the United States of America in 1983, and adopted in principle in Europe in 2001, the legisla- tion has facilitated the early exclusive licensing of products for rare diseases and has greatly enhanced corporate pharmaceutical investment. Orphan diseases are variously defined as those affecting fewer than 1 in 2000 of the population (Europe) or fewer than 200 000 individuals (United States of America); each lysosomal disease is, in effect, an ultra-​orphan disorder, that is, a disease affecting fewer than 1 in 50 000 individuals. Despite attracting great attention as a result of the high individual costs of treatment, the total national burden of treatments for these diseases in countries with devel- oped healthcare systems is low (in England, the costs of specific treatments for lysosomal diseases amounts to about 0.1% of the health budget). Orphan drug development in lysosomal diseases: fortunes and misfortunes Encouraged by the success of several ‘blockbuster’ products such as recombinant human growth factors and erythropoietin, the orphan drug industry has grown rapidly. An early adopter of these oppor- tunities, Genzyme Therapeutics, the corporation that first jointly de- veloped macrophage-​targeted enzyme therapies with the National Institutes of Health in the United States of America, and with sub- stantial financial support from the National Gaucher Foundation, introduced highly effective treatment for this disease within the aegis of the Orphan Drug Act. By 2009, alglucerase (Ceredase) followed by imiglucerase (Cerezyme) for Gaucher’s disease, supplied 5000 patients in more than 90 countries. The company, then the third lar- gest biotechnology corporation, reported revenues of about $4 bil- lion, of which about one-​third were due to Cerezyme; revenue from agalsidase alfa (Fabrazyme) for Fabry’s disease was $424 million. Commercial success on such a scale supported continuing invest- ment in even more challenging disorders, including Pompe’s dis- ease. Delivery of the therapeutic protein Myozyme (recombinant human acid α-​glucosidase (maltase)) to a large bulk of diseased skel- etal muscle, to which it is targeted by surface expression of mannose 6-​phosphate residues, was a major challenge. The therapeutic de- livery required administration of gram quantities of the remodelled recombinant glycoprotein at each infusion. The scale of the manu- facturing resources required for the developmental Pompe’s disease clinical trial programme led to a lack of secure reserve stocks of the corporation’s leading products. By mid 2009, a vesiviral infection impaired the viability of the re- combinant Chinese hamster cells in the bioreactors that synthesize Cerezyme and Fabrazyme. Manufacture was rapidly shut down with the immediate consequence of a ‘global supply restraint’ in which treatment for Gaucher’s disease and Fabry’s disease became critically limited. Many patients were without treatment for nearly 2  years. Fortunately, licensed products already in development were accelerated through the industrial scale-​up and regulatory approval processes, and with expanded compassionate use and ac- cess programmes the manufacturing void was gradually filled. Of note, the Genzyme corporation was purchased by Sanofi and full-​ scale manufacturing and its global supply capacity was restored by late 2012. This biopharmaceutical shut-​off of therapeutic supply was un- precedented in scale and totally unexpected. The episode revealed an inevitable but concealed risk associated with orphan drug legis- lation. While the orphan drug initiative serves as a powerful and successful incentive for drug development in neglected diseases with clear unmet needs, the rewards for market authorization of a first-​in-​class orphan agent with demonstrable efficacy and justifiable safety are essentially anticompetitive. Beyond marketing credibility for any given company and the bio- pharmaceutical industry overall, there are deeper consequences of the events described: (1) the episode should instruct all stakeholders of the need for any manufacturer to support and maintain adequate reserve stock, with shared costs as a result; (2) the value of sustained competition in therapeutic development, even in the realm of ultra-​ rare; and (3) the intellectual domination of the global community as a consequence of ‘life-​changing’ enzyme therapy in Gaucher’s dis- ease opened up and exposed a vulnerable global patient community

12.8  Lysosomal disease 2137 to the risk of corporate collapse and treatment withdrawal, although this nightmare scenario was avoided in this instance. Current therapeutic landscape At present, about 20 recombinant human enzyme preparations are in use or in late clinical investigation. Several companies are ex- panding interest in this rarefied field, with additional recombinant proteins (including biosimilar, modified, and semisynthetic mol- ecules), small-​molecule products, and even gene therapy in robust competitive development (Table 12.8.2). The magnification of interest that has accompanied successful medical research into this area has generally been a model of utility and progress. It continues to provide for many patients and their families the hope that definitive relief might be forthcoming. Nevertheless, given the high cost per patient, decision-​making bodies such as the United Kingdom National Institute for Health and Care Excellence (NICE) continue to look closely at the health eco- nomic benefits of individual treatments for rare diseases, operating as they do in healthcare systems that are financially constrained. Specific treatments and their mechanisms Augmentation of deficient activity Early experiments by Elizabeth Neufeld and colleagues using fibroblasts in which glycosaminoglycans accumulate due to mucopolysaccharidoses such as Hurler’s disease (MPS I, autosomal re- cessive) and Hunter’s syndrome (MPS II, X-​linked), showed that the rate of degradation—​rather than the rates of synthesis or secretion—​of 35S sulphate-​labelled substrate is severely disrupted. When (initially as a result of a laboratory error) fibroblasts obtained from these genetically distinct storage disorders were co-​cultured, the pathological accumu- lation of glycosaminoglycans in lysosomes was prevented. The biosyn- thetic labelling technique was also used to show that degradation of the substrates was restored to normal in these co-​culture experiments. Further investigation of this phenomenon by the Neufeld group demonstrated that each of the fibroblast cultures elaborated and de- livered a specific corrective factor to the medium, which ultimately proved to be a high molecular weight form of the hydrolases that were specifically lacking in fibroblasts from the corresponding disease. These corrective factors were identified in several comparable experi- ments using fibroblasts derived from other mucopolysaccharidoses and also several different classes of lysosomal disease; when taken up from the medium, the factors restore the impaired intracellular degradation of cognate substrates. Functional correction of the biochemical defects thus permitted an early classification of distinct complementation groups among the MPS syndromes, often before the individual enzymatic defects had been characterized. Specific receptor pathways for the biosynthesis and uptake of nascent lysosomal proteins during the course of organelle biogen- esis have been identified: the process is usually brought about by the so-​called recognition marker, mannose 6-​phosphate. This ter- minal sugar is generated by a specific mechanism involving two post-​translational modifying enzymes during the biosynthesis of soluble glycoproteins destined for the lysosomal matrix. Receptors, serving as intracellular lectins for mannose 6-​phosphate ligands are densely expressed on prelysosomal membranes and mediate uptake of suitably labelled nascent proteins into the developing or- ganelle. However, this trafficking process is not foolproof and 10 to 20% of newly formed soluble lysosomal proteins are misdirected to the plasma membrane from which they are released; by the same token, an appreciable population of cation-​independent mannose 6-​ phosphate receptors is expressed on the plasmalemma, serving the function of regulatory uptake of IGF. The ‘leakiness’ of this targeting system represents a default pathway for lysosomal protein secretion and recapture—​as well as mutual complementation between dif- ferent cells and tissues. Functional complementation of lysosomal storage disorders by supplying particular molecular isoforms of the enzymes that are deficient in individual diseases provides a scientific justification for enzyme replacement treatment. Successful application of enzyme-​ replacement is dependent on an understanding of glycoprotein chemistry, receptor-​mediated endocytosis, and the molecular cell biology of lysosomal biogenesis: identification of the secretion and recapture mechanism has provided further practical underpinning. The mannose 6-​phosphate pathway is not the only mechanism for delivering proteins to the lysosome; indeed, the first successful en- zyme replacement therapy for Gaucher’s disease employed human glucocerebrosidase that was modified specifically to reveal ter- minal unphosphorylated mannose residues that greatly enhanced delivery of the therapeutic protein to cells of the macrophage lin- eage that are the principal focus of the disease. Characterization of lysosomal recognition markers occurred at a time when other cell surface glycoprotein recognition systems were being identified: the asialoglycoprotein receptor, the first mammalian lectin identi- fied by Ashwell and Morell, can mediate the uptake of modified plasma proteins by parenchymal liver cells in vivo. Recent studies show that in some cells, for example, in the inner ear and brain, as well as lymphocytes (but not fibroblasts or macrophages), delivery of nascent acid glucocerebrosidase to lysosomes is dependent on a unique tissue-​specific chaperone function supplied by a lysosomal membrane protein (LIMP2). Mutations in the human LIMP2 gene appear to account for some atypical cases of Gaucher’s disease with neurological manifestations (including myoclonic epilepsy), kidney disease, and perplexing enzymology when examined in periph- eral blood cells and cultured skin fibroblasts; in these patients, acid glucocerebrosidase deficiency occurs in fibroblasts but not leuco- cytes or tissue macrophages. Haematopoietic stem cell transplantation Cellular complementation, by providing a source of wild-​type enzyme delivered from allogeneic bone marrow transplantation, has also had spectacular successes in several lysosomal disorders. In Gaucher’s Table 12.8.2  Lysosomal diseases with approved enzyme therapies Disease with approved therapy Disease with trial completed but drug not approved Acid lipase deficiency Anderson–​Fabry disease Gaucher disease types I and III α-​Mannosidosis MPS type I MPS type II MPS type IVA MPS type VI MPS type VII Neuronal ceroid lipofuscinosis type II Pompe disease Krabbe disease (peripheral) Metachromatic leukodystrophy (intrathecal) MPS type IIIA (intrathecal)

section 12  Metabolic disorders 2138 disease, where the pathogenic cell is of haematopoietic origin, bone marrow transplantation was effective in the past. Successful engraft- ment led to full correction of the biochemical defect and reversal of most of the visceral and haematological effects of the condition that had not already progressed irrevocably. Now that a safer treatment in the form of enzyme replacement is available, bone marrow trans- plantation, with its attendant risks, is very rarely indicated. In diseases due to deficiency of soluble hydrolases, donor cells that repopulate the microglia (the brain equivalent of tissue macro- phages) may participate in the secretion-​recapture mechanism, and in this form of cell replacement therapy would be expected to pro- vide a source of enzyme to vicinal cells. Haematopoietic stem cell transplantation has shown efficacy at an early stage of disease in several of the neurodegenerative lysosomal disorders, such as Hurler’s disease (MPS I); very young infants, par- ticularly in the immediate neonatal period, with Krabbe’s disease and, as described previously, in chronic neuronopathic Gaucher’s dis- ease. As a result of improved outcomes of the intervention in general, treating physicians are re-​evaluating the potential of haematopoietic stem cell transplantation in conditions where its role had previously been questioned, such as Hunter’s syndrome (MPS II). Gene therapy Gene therapy has long been discussed in relation to lysosomal dis- eases since the capacity to transduce a focus of cells using vectors expressing the deficient enzyme or protein within a tissue is an at- tractive possibility for sustained functional complementation. This approach has shown spectacular benefit in several different spontan- eous and transgenic animal models that are genetically and clinically coherent with their human counterparts. At present, two principal stratagems—​both dependent on viral vectors—​are being explored in lysosomal diseases; mainly those with life-​threatening features including neurological disease. Third-​generation lentiviral vectors are able to transduce dividing cells such as haematopoietic stem cells and integrate into the host-​ cell nuclear genome. Adeno-​associated viral vectors do not integrate into the nuclear DNA but remain episomal and direct biosynthesis of therapeutic transgene products in nonmitotic cells, including neural cells. In the case of haematopoietic stem cells, the genetically cor- rected autologous cells of haematopoietic origin can be reinfused into the donor after transduction ex vivo. Here after engraftment they are able to deliver the corrective protein function (wild-​type enzyme) to the tissues in circumstances where these migratory cells of haemato- poietic origin slowly populate the sites of disease, such as the brain and spinal cord. It is believed that they give rise to macrophages and lymphoid cells with the potential to secrete corrective factors for up- take and functional complementation of local disease. Reconstitution of bone marrow-​derived cells, engineered by lentiviral gene transfer to overexpress the wild-​type enzyme (in- tended to deliver an abundance of soluble enzyme), are being ex- plored and efficacy has been shown in late-​infantile and juvenile metachromatic leukodystrophy. Retroviral vectors and gene con- structs were used to introduce the desired DNA sequence encoding arylsulphatase A  into autologous explanted haematopoietic stem cells of young, presymptomatic subjects with metachromatic leukodystrophy, and the genetically corrected cells were expanded in culture and then returned to the patient’s circulation. Phase I/​II clinical trial results reported by Dr Alessandra Biffi and colleagues in Milan provide convincing evidence of some neurological benefit or ‘rescue’ compared with historical and sibling control patients with early-​onset disease not so treated. It is not certain whether transplantation of haematopoietic stem cells from healthy matched related donors give comparable or inferior results. Despite the dif- ficulty in determining efficacy directly in the gene therapy studies, remarkably the disease did not manifest or progress in the first eight of nine patients who underwent the autologous gene therapy pro- cedure. What is clear, however, from worldwide experience not only of haematopoietic stem cell transplantation but genetically modified autologous stem cell therapy, is that compelling clinically signifi- cant benefit is almost solely restricted to recipients who undergo the intervention in the presymptomatic phase of this disease. The other stratagem under active development for clinical applica- tion in the lysosomal diseases employs vectors based on the use of re- combinant nonpathogenic ‘passenger’ adeno-​associated picornavirus of several serotypes with preferential ability to transduce certain cell types, known as ‘tropisms’. These tropisms can be harnessed to facilitate delivery and expression of the cognate therapeutic transgene DNA to particular tissues such as the liver or neural cells. Each vector system has potential advantages and shortcomings, which are discussed in the suggested ‘Further reading’ material at the end of this chapter. At the time of writing, several trials using direct injection of re- combinant adeno-​associated viral vectors expressing the cognate corrective human proteins into the brains of children with two other neurological lysosomal diseases have been safely completed. Early safety and efficacy outcomes of the Sanfilippo’s disease type A (MPS IIIA) trial using rAAV rh10 have been reported: the safety criterion was met with an indication of stabilization in three of the four patients and possible improvement in one. Following the recent licensing of the vector, a more definitive phase III trial is planned. Encouraging outcomes of a phase I/​II trial using intracranial rAAV vectors serotype 5 in four children aged 20 to 53 months with Sanfilippo’s disease type B (MPS IIIB) have been reported. Not only were the safety requirements met, but over 24 months neurocognitive progression was improved in all patients compared to that expected. N-​acetylglucosamine activity was detected in lumbar cerebrospinal fluid and was 15 to 20% of that in unaffected children. Other phase I/​II clinical trials are in active de- velopment for other lysosomal diseases. Enzyme replacement therapy Discovery of the mechanism by which lysosomal proteins are delivered to the nascent organelle have gone hand-​in-​hand with the hope of treat- ment based on the targeting of therapeutic enzymes to diseased tissues. The first commercial preparation of glucocerebrosidase (alglucerase, Ceredase) was not licensed until 1991 and 1994 by the Food and Drug Administration (FDA) in the United States of America and by the European Medicines Agency, respectively, after decades of painstaking research. This was purified from placentae and its glycan structure modified enzymatically to reveal terminal mannose groups that bind the mannose receptor on cells of macrophage origin. It mitigated many features of Gaucher’s disease when given parenterally. The therapeutic and commercial success of alglucerase, along with potential difficulties in maintaining the supply of suitable placentae, stimulated the demand for a recombinant preparation (imiglucerase, Cerezyme). Expansion of the approach to include diseases that would be targeted using mannose 6-​phosphate, the more familiar lysosomal recognition marker, followed.

12.8  Lysosomal disease 2139 Since 2001, recombinant protein therapies have become avail- able for Fabry’s disease (two products), Hurler–​Scheie disease (MPS I), Hunter’s disease (MPS II), Maroteaux–​Lamy disease (MPS VI), Pompe’s disease (glycogen storage disease type II) and Morquio’s dis- ease type A (MPS IVA), and more recently lysosomal acid lipase de- ficiency (Wolman’s disease, the infantile form, and cholesteryl ester storage disease, the later-​onset form). Velmanase alfa (Lamzede) has recieved marketing authorisation by the European Medicines Agengy in 2018 for the treatment of Alpha-Mannosidosis. Trials of intra- thecal enzyme therapy are in progress for several lysosomal disorders that cause neurodegeneration, including MPS II and metachromatic leukodystrophy. Universal availability of these treatments is limited by their very high cost (licensed doses cost upwards of $200 000 per annum for an average adult) and by the requirement for a sophis- ticated healthcare infrastructure to support the delivery and moni- toring of the therapy. Thus, even imiglucerase (Cerezyme), which has been available for over 15 years and whose efficacy is clear, is available to fewer than 20% of patients globally for whom it is indicated. Analogous biological products have now been approved for treat- ment of Gaucher’s disease. Velaglucerase alfa (VPRIV) is produced by specific upregulation of the endogenous gene sequence in a fibro- sarcoma cell line cultured in the presence of an inhibitor of post-​ translational glycosylation to express human glucocerebrosidase decorated by high-​mannose glycans. After an extensive clinical trial programme and aided by the supply limitation of imiglucerase, it received marketing approval in 2010. Taliglucerase alfa is produced using recombinant technology in cultured carrot cells and has re- ceived authorization as Elelyso by the FDA in the United States of America (but not by the European Union). Two enzyme products are approved by the European Medicines Agency in Fabry’s disease. Agalsidase alfa (Replagal) is generated by targeted overexpression of the human AGAL gene in a human fibro- sarcoma cell line and is marketed at a dose of 0.2 mg/​kg every other week. Agalsidase beta (Fabrazyme) is a traditional recombinant product generated in Chinese hamster ovary cells and marketed at a dose of 1 mg/​kg every other week. Agalsidase alfa did not secure marketing approval in the United States of America. Development continues of an alglucosidase modified by the chemical addition of mannose phosphate ligands: avalglucosidase alfa is in phase III trials in Pompe disease. A clinical trial programme in Fabry’s disease using an agalsidase modified by addition of poly- ethylene glycol adducts—​pegylation—​has been initiated, with evi- dence of different biodistribution and cell uptake characteristics. The therapeutic position of newer enzyme products for other con- ditions will evolve, but it must be recognized that in general the con- ditions for which they are licensed are heterogeneous and, it appears, more intractable than the visceral and haematopoietic features of Gaucher’s disease. It seems likely that internalization of glycoproteins decorated by mannose 6-​phosphate signals is less rapid and effective in vivo than the uptake and delivery of those displaying terminal man- nose residues recognized by the mannose receptor on macrophages. Pharmacological chaperone therapy This stratagem is based on the ability of small molecules to bind to key regions of mutant proteins that are misfolded and thus prema- turely degraded in the endoplasmic reticulum and Golgi network. Aberrant protein folding is increasingly recognized as a molecular mechanism in inherited diseases since, as a result of disrupted cotranslational processing, it leads to an operational deficiency of protein function at the point of action. Pharmacological chaperones are molecules which bind to mutant proteins in a stable complex and thus assist delivery to the site of action in the correct cellular compartment. In the case of lysosomal enzymes, the candidate chaperone molecule is usually a competitive inhibitor of the nascent enzyme at neutral pH and designed to dis- sociate from the mature lysosomal enzyme on arrival at the acidic environment of the organelle. Pyrimethamine, a licensed oral antiprotozoal folate antagonist, has chaperone-​like effects in cells harbouring some HEXA muta- tions from some patients with attenuated forms of Tay–​Sachs and Sandhoff’s diseases. This drug, which traverses the blood–​brain barrier, has undergone small phase I/​II clinical trials in which its efficacy has mainly been disappointing. However, the authors have seen one patient with juvenile Sandhoff’s disease in whom pyri- methamine induced a large increase in enzyme activity in cultured skin fibroblasts, with more mature hexosaminidase protein in the lysosomal compartment together with partial clearance of the cog- nate substrate (GM2 ganglioside); improved cognitive power and neuropsychological test scores were seen on oral exposure to the drug (with calcium folinate supplements) over a 5-​year period. Several iminosugars correct the misfolding of mutant lysosomal glucocerebrosidases in experimental cell systems, including cul- tured fibroblasts obtained from Gaucher’s patients:  one of these, isofagomine, underwent clinical evaluation but failed to demonstrate sufficient efficacy. Equally, clinical trials of single-agent chaperone-​ based therapy in Pompe’s disease failed to show benefit. Clinical trials have completed with another iminosugar (1-​deoxygalactonojirimycin, migalastat) in patients with Fabry’s disease in whom in vitro studies indicate the potential for functional enhancement of several mutant α-​galactosidase variants. One trial in treatment-​naïve subjects dem- onstrated a modest reduction in storage vacuoles on kidney biopsy. In a separate switch study in patients already receiving enzyme therapy, subjects’ kidney function remained stable. In both studies, interesting and potentially salutary effects were noted on left ventricular mass by echocardiography. On the basis of these findings, migalastat received marketing authorization in the European Union in 2016 and in the United Kingdom has also received approval by NICE for reimburse- ment. Phase three trials are in progress to examine a combination of a chaperone and a second-generation enzyme therapy for Pompe dis- ease. In this case the aim of the chaperone is to act extracellularly to improved the stability and delivery of the therapeutic enzyme. Although the use of pharmacological chaperones is an attractive concept for the oral treatment of lysosomal diseases, hitherto—​ outside the use of neopterin in phenylketonuria and other vitamin cofactors such as pyridoxine for homocystinuria—​no small mol- ecule has shown true clinical efficacy in a putative misfolding disease. Restriction (or rebalancing) of substrate flux (‘substrate reduction therapy’) For many years, the accumulation of storage material within lyso- somes has been considered to be the precipitating factor for the development of tissue injury and the inflammatory response that accompanies the lysosomal storage disorders. By analogy with the development of atherosclerosis due to impaired catabolism of chol- esterol bound to low-​density lipoproteins, it is principally a failure

section 12  Metabolic disorders 2140 of degradation or export from the lysosome that leads to the patho- logical storage. Thus, like the statins which inhibit the first committed step in the biosynthesis of cholesterol, the concept of depleting the supply of macromolecular substrate to prevent the accumulation of injurious material has been developed experimentally and brought to clinical practice in the sphingolipid disorders. Two classes of inhibitor are prominent in therapeutic studies:  iminosugars derived from naturally occurring compounds (acting principally as sugar mimetics) and synthetic pyrrolidino com- pounds (acting as analogues of the ceramide moiety of sphingolipids). Iminosugars Some iminosugars related to deoxynojirimycin are inhibitors of the ceramide-​specific UDP-​glucosyltransferase reaction as the first committed step in the biosynthesis of glycosphingolipids. Following experimental studies in cultured cells with pathological storage of glycolipids in lysosomes and in murine models of debilitating human glycosphingolipidoses, clinical trials of N-​butyldeoxynojirimycin (miglustat, Zavesca, a particular analogue of iminosugars) were conducted in in Gaucher’s disease. Evidence of disease regression was obtained in an open-​labelled trial with reduction in visceral en- largement, enzymatic markers of Gaucher’s disease activity (plasma chitotriosidase activity), and a slow improvement in haematological parameters. The drug gained marketing authorization in 2002 as an orally active second-​line treatment for type 1 Gaucher’s disease in adults unable or unwilling to receive enzyme therapy. Although ap- proved as a first-​in-​class inhibitor of substrate biosynthesis, further research now suggests that among the off-​target effects of miglustat, its action as a more potent inhibitor of a neutral glucocerebrosidase involved in sphingolipid recycling contributes to its therapeutic ef- fects in non-​neuronopathic Gaucher’s disease. Since the iminosugars are small molecules with the potential to penetrate the blood–​brain barrier, the possibility of their use (either as a monotherapy or as a synergistic treatment with enzyme therapy) for neuronopathic Gaucher’s disease has been raised, as well as for the treatment of the otherwise intractable glycosphingolipidoses that cause severe neurological disease. However, miglustat was found not to be effective in young patients suffering from the neurological ef- fects of type 3 (chronic neuronopathic) Gaucher’s disease who were receiving imiglucerase, but it appeared to improve pulmonary mani- festations which usually fail to respond to enzyme therapy alone. Miglustat has also received marketing authorization in Europe for treatment of Niemann–​Pick disease type C, having shown delay in progression of the neurological features in a randomized, open-​label clinical trial. Synthetic pyrrolidino compounds Eliglustat, a pyrrolidino compound that is a ceramide analogue with a highly selective and potent inhibitory action on glucosylceramide biosynthesis, has demonstrated strong therapeutic effects in non-​ neuronopathic Gaucher’s disease. Encouraging 4-​year phase III and 8-​year phase II clinical studies and 1400 patient-​years of clinical trial exposure support its position for most adult patients with type 1 (non-​neuronopathic) Gaucher’s disease as an alternative first-​line agent to enzyme therapy. The oral activity in once-​ or twice-​daily dosing, represents an important advantage over intravenous infu- sions of enzyme therapy given every 2 weeks. Prescription requires specialist monitoring, with dosing guided principally by cytochrome P450 (CYP) genotyping since the agent is extensively metabolized by the CYP2D6 and to some extent CYP3A4 systems. Eliglustat (as Celderga) has received marketing authorization by the FDA in the United States of America as an oral first-​line agent for adults with type 1 (non-​neuronopathic) Gaucher’s disease, is authorized by the European Medicines Agency, and is approved by NICE for reim- bursement by the National Health Service (as in many other regions). Eliglustat is a substrate for the P-​glycoprotein MDR1 efflux transport system, and thus does not distribute effectively to the brain. With the recognition that there remains a large need for small molecule drugs that traverse the blood–​brain barrier to exert therapeutic effects in those lysosomal diseases with neurological effects, and that the biosynthesis of glucosylceramide represents a common target for several disabling sphingolipid disorders, the Sanofi Genzyme company has identified an additional inhibitory molecule, venglustat, for clinical conditions requiring systemic and/​or neurological targeting. A  phase I/​II clinical trial of this agent has been completed in Fabry’s disease, and similar trials are underway in adults with type 3 (chronic neuronopathic) Gaucher’s disease and in patients with Parkinson’s disease who are heterozy- gous for pathological mutation in the GBA1 gene. Given the bio- chemical relationships of glucosylceramide with the gangliosides that accumulate in Tay–​Sachs and Sandhoff’s diseases, as well as GM1 gangliosidosis, chronic forms of these neurological diseases are potential therapeutic targets for this drug class. Examples of lysosomal disorders Gaucher’s disease This disorder may occur at any age and has been regarded as the most frequent of the lysosomal storage diseases, although recent evidence suggests that Fabry’s disease, particularly in its attenu- ated forms, is substantially more frequent. The condition is usu- ally due to a catalytic deficiency of glucocerebrosidase, although rare cases of deficiency of its cognate sphingolipid activator protein (SAP-​C) may cause a severe disorder usually similar to the subacute neuronopathic form of true Gaucher’s disease. Numerous mutations responsible for the enzymatic deficiency have been identified in the human glucocerebrosidase gene and the reader is referred to the specialist literature for those genotype/​phenotype correlations that broadly apply to this protean disorder. The mature protein contains 497 amino-acid residues; a 39 amino-acid lead peptide is cleaved from the initially-translated polypeptide that contains 536 initial amino acids. A recent change in nomenclature now considers the position of amino-acids relative to the beginning of the 536 amino- acid peptide, rather than the 497 amino-acid mature protein. Thus the most frequent mutations, known previously as N370S and L444P, are now referred to as p.Asn409Ser and p.Leu483Pro respectively. Gaucher’s disease type 2 and type 3 Rarely, infants are born with an almost complete lack of gluco­ cerebrosidase activity: they die within a few days of birth or are still- born due to skeletal deformities and/​or dehydration as a result of loss of skin integrity (collodion babies). Infantile Gaucher’s disease (classified as acute neuronopathic or type 2 disease) is a rare neuronopathic disease with bulbar palsy, opisthotonus, and minor visceral enlargement. It is invariably fatal

12.8  Lysosomal disease 2141 in the first 2 years of life and does not respond to either systemic or intrathecal enzyme replacement therapy. While neurological disease may occur in children, adolescents, and young adults with Gaucher’s disease, it is less severe than in the infantile variant and is termed subacute neuronopathic or type 3 dis- ease. In such patients the disease is associated with supranuclear gaze palsies, ataxia, nerve deafness, myoclonus, and (occasionally) seizures. In type 3 disease the neurological condition usually deterior- ates slowly but is exacerbated if splenectomy is performed for the accompanying splenomegaly and associated pancytopenia. Where possible, and with vigorous enzyme therapy, splenectomy is best avoided, although partial splenectomy may be carried out to ameli- orate pressure effects and life-​threatening thrombocytopenia. Subacute neuronopathic disease is not always fatal and often improves with bone marrow transplantation and enzyme replace- ment therapy, although the effects of the latter are restricted to the systemic, non-​neurological aspects. Affected children may show striking visceromegaly, with the associated gaze palsies often playing a small part in the clinical presentation. Although juvenile subacute neuronopathic Gaucher’s disease (type 3) occurs sporadically in all populations, there is a small iso- late in Northern Sweden where all individuals are homozygous for a single point mutation in the glucocerebrosidase gene (L444P) that has arisen by descent from a common ancestor. Gaucher’s disease type 1 The most frequent form of Gaucher’s disease is the so-​called adult non-​neuronopathic form (type 1). This is found in all populations but is over-​represented in Jews of Ashkenazi origin. Although the condition does not commonly affect the nervous system, visceral and skeletal manifestations are prominent. Clinical features Characteristically, Gaucher’s disease presents with pancytopenia, with bleeding due to thrombocytopenia and splenic enlargement. Acutely painful episodes also occur in the bones, particularly during growth, and these episodes are followed by the evolving MRI appearances of osteonecrosis with consequential effects on the integrity of large joints, including the hip, knee, and shoulder (Fig. 12.8.5). The increased frequency of infarction events is an important aspect of Gaucher’s disease that, as yet, has not been ex- plained, and bone necrosis remains an aspect of the condition that often persists despite enzyme therapy. In the era before enzyme replacement therapy, splenectomy was often carried out during childhood to relieve the pressure effects of the en- larged organ and to ameliorate the effects of accompanying cytopenias. Although there appears to be a striking temporal association between splenectomy and the development of severe bone disease, it is unclear as to whether this is directly due to the effects of the splenectomy or the consequential manifestations of disease severity. Nonetheless, splen- ectomy is best avoided where at all possible. Splenectomy in Gaucher’s disease carries a greatly enhanced risk of overwhelming infection; this includes infection with protozoa, such as babesia and malaria, as well as capsulated bacteria, for example, Streptococcus pneumoniae, Haemophilus influenzae, and Neisseria meningitidis. In addition to the effects of osteonecrosis, the osseous manifest- ations of Gaucher’s disease are very diverse and include the presence of expanded bone lesions (Fig. 12.8.6) with surrounding cortical Fig. 12.8.5  T2-​weighted MRIs obtained from the lower femur and upper tibia of a 30-​year-​old woman with non-​neuronopathic Gaucher’s disease experiencing pain due to acute avascular necrosis of bone. Note the geographical areas of increased signal intensity on the T2-​weighted image due to increased tissue water representing oedema surrounding the necrotic tissue. Courtesy of Professor D. Lomas, Addenbrooke’s Hospital. Fig. 12.8.6  Expanded lytic lesion at the distal end of the femur in a 44-​year-​old woman with severe Gaucher’s disease complicated by osteoporosis, osteonecrosis, and, as shown, expanded lytic lesions in long bones leading to local infiltration of the marrow space by Gaucher tissue.

section 12  Metabolic disorders 2142 thinning related to Gaucher cell infiltrates within the bone marrow (‘Gauchomas’). Diffuse osteoporosis accompanied by pathological fractures may also compound the skeletal manifestations. Kyphosis due to crush fractures of vertebrae are common in untreated adults, particularly in postmenopausal women. Gaucher’s disease may rarely be associated with pulmonary infil- trates, including reticulonodular opacities, restrictive lung defects, and various abnormalities of the pulmonary circulation, causing pulmonary hypertension. The hepatopulmonary syndrome, ac- companied by platypnoea and associated with severe scarring liver disease or cirrhosis and portal venous hypertension, has also been reported in severely affected adults. In its untreated state, Gaucher’s disease is a miserable condi- tion leading to progressive skeletal deformity, pancytopenia, and visceral enlargement with failing organ function punctuated by painful visceral bone crises. The mean age of death in a single large series reported from Pittsburgh, Pennsylvania, was 60 years during the pretreatment era, but this does not take into account the poor quality of life of most affected individuals. Some homo- zygotes for ‘mild’ missense mutations in the glucocerebrosidase gene (especially the widespread mutation, N370S) may escape detection and remain asymptomatic throughout a long adult life. Detailed investigation reveals only a mild thrombocytopenia and trivial splenomegaly in some cases. However, monoclonal gammopathy is frequently present after the age of 45 years. It is uncertain as to what extent the presence of such mutations in the population at large (homozygosity for N370S occurs in about one in 960 Ashkenazi Jews) contributes to the development of β-​cell lymphoproliferative disorders, such as B-​cell lymphoma and mye- loma, in this at-​risk group. Other clinical aspects Parkinsonism and Gaucher’s disease—​a complex relation- ship  Approximately 5% of patients develop extrapyramidal disease resembling parkinsonism in middle life. The response to dopaminergic agents is often less clear than in classical idio- pathic Parkinson’s disease and the disorder may progress more rapidly. This complication may reflect the emerging strong but ill-​ understood relationship between mutant glucocerebrosidase al- leles and Parkinson’s disease and especially Lewy body-​associated diseases and α-​synucleinopathies in several populations: heterozy- gous mutations in the gene encoding glucocerebrosidase represent the commonest single genetic association with Parkinson’s disease in all populations studied. It appears that there is at most a small gene-​dosage effect and Parkinson’s disease appears to be little more frequent in patients with Gaucher’s disease than in their heterozy- gous parents. An international study by Sidransky and colleagues involved the search for two frequent missense GBA1 mutations, L444P and N370S, in patients with Parkinson’s disease attending 16 centres. A total of 5691 patients with Parkinson’s disease (780 Ashkenazi Jews) and 4898 control subjects (387 Ashkenazi Jews) were geno- typed. Among Ashkenazi Jewish subjects, either mutation was found in 15% of patients but only 3% of control subjects; among non-​Ashkenazi Jewish subjects, either mutation was found in 3% of patients and less than 1% of controls. GBA1 was fully sequenced in 1983 non-​Ashkenazi Jewish patients, and mutations were iden- tified in 7%, showing that limited mutation screening can miss half the mutant alleles. The odds ratio for any glucocerebrosidase mu- tation in patients with Parkinson’s disease compared with controls was 5.43. When compared with patients with Parkinson’s disease who did not carry a GBA mutation, those with a mutation had an earlier presentation with the disease, were more likely to have affected relatives, and were more likely to have atypical clinical manifestations. Relationship to  B-​cell malignancy  Gaucher’s disease abnor- malities include a polyclonal immunoglobulin response that may progress to monoclonal gammopathy, amyloidosis, or even frank myeloma and B-​cell lymphoma, with an estimated 20-​ to 40-​fold in- creased risk compared with heathy subjects without Gaucher’s dis- ease. These malignant complications are now an important cause of death in adult patients with type 1 Gaucher’s disease. Their cause appears to be related to the clonal expansion of B cells that specif- ically secrete antibodies directed against the pathological complex lipids and that are driven, at least initially, by a subclass of follicular B helper T cells (TFH2) recognizing the glycosphingolipids presented by the CD1d molecule. Other plasma and metabolic abnormalities  Low-​density lipopro- tein and high-​density lipoprotein cholesterol fractions are abnormal in the plasma. Basal metabolic rate is increased. Some lysosomal enzymes are elevated, including tartrate-​resistant acid phos- phatase, hexosaminidase, and a human chitinase, chitotriosidase. Chitotriosidase may reflect the severity of the disease and has proved to be very useful for monitoring Gaucher’s disease activity in response to treatment. The enzyme is elevated, sometimes several hundredfold above normal, in the untreated condition. Pathology The pathognomonic abnormality is the presence of large storage cells, which are activated macrophages (Gaucher cells), typically found in the splenic sinusoids. The Gaucher cells (Figs. 12.8.7 and 12.8.8) replace the Kupffer cells of the liver, alveolar macrophages of the lung, and in the bone marrow. Fig. 12.8.7  Light micrograph of a Leishman-​stained bone marrow biopsy obtained from a 23-​year-​old man with type 1 Gaucher’s disease. Note that the large, pale-​blue staining Gaucher cells with striated cytoplasm replace the Kupffer cells of the liver, alveolar macrophages of the lung, and of the bone marrow.

12.8  Lysosomal disease 2143 Diagnosis The diagnosis of Gaucher’s disease is based on white-​cell acid β-​glucosidase activity, which may be accompanied by the elevation of one or more related marker enzymes such as chitotriosidase or tartrate-​resistant acid phosphatase in the serum. Spleen tissue, liver biopsy material, or bone marrow aspirates may show the character- istic oligonucleate storage cells demonstrating striated cytoplasm on Leishman staining (Fig. 12.8.7), but which appear as pink sheets in tissue sections stained with haematoxylin and eosin. Molecular ana- lysis of the glucocerebrosidase gene may identify widespread mutant glucocerebrosidase alleles and may assist in the diagnosis and inves- tigation of family members at risk for this recessive disorder. Treatment Until recently, the treatment for Gaucher’s disease was palliative. Bone marrow transplantation has been undertaken in a few infants and children with rapidly progressive disease, including those with the subacute neuronopathic form type 3. When successful, this may correct most of the systemic manifestations of the condition and restore growth, and some observers believe that it may arrest fur- ther neurological deterioration. However, bone marrow transplant- ation is no longer in routine use because of the accompanying severe risk resulting from the procedures and constraints in the supply of donors, especially MHC-​matched, first-​degree relatives. Steady im- provement in the outcome of stem cell transplantation in general raises the possibility that this stratagem could be re-​evaluated in patients with chronic neuronopathic Gaucher’s disease in circum- stances where health services resources will not sustain the long-​ term costs of high-​cost molecular therapies (whether enzyme therapy or other potential chronic therapies should, for example, venglustat prove to be safe and effective). Enzyme augmentation therapy Enzyme augmentation therapy was introduced during the early 1990s in the form of a natural product extracted from the human placenta, alglucerase (Ceredase). The recombinant glycoform, imiglucerase (Cerezyme), is supplied as a lyophilized powder which is reconstituted for intravenous infusion, given most commonly every 2 weeks. After a few weeks of enzyme administration, most patients show an improvement in the blood parameters of disease activity and a reduction of the chronic inflammatory response that accompanies the condition. The platelet count rises and there is correction of the hypersplenic blood picture, with reduction in hepatosplenomegaly. There is improvement in asthenia and quality-​ of-​life measures. Similar salutary effects are noted with the use of the more recently licensed enzyme therapies. Since most patients express the protein antigen endogenously, hypersensitivity and immune reactions are very rare. Apart from the inconvenience of periodic intravenous infusions, treatment is well tolerated and many patients in Europe and the United Kingdom choose to take their treatment as self-​administered infusions at home. Controversy remains as to the appropriate dose of enzyme therapy, whether in the form of imiglucerase or the newer products velaglucerase alfa or taliglucerase, but most authorities agree that administration of the enzyme should be lifelong. There are several schools of thought as to whether enzyme therapy should be admin- istered at a high dose to start with, perhaps then reducing as evidence of disease regression becomes clear, or whether a more variable but lower dose be given and altered according to response. Disease activity is assessed by objective parameters, including visceral en- largement, and by determination of surrogate biomarkers such as chitotriosidase and blood counts. The application of simple defined therapeutic goals with close monitoring of individual patients has much to recommend it. Achievement of key goals and amelior- ation of disease-​associated parameters is more rapid when high-​ dose enzyme therapy is administered. In patients with the subacute neuronopathic form of the condition (type 3), international guide- lines suggest that a dose of at least 60 units of enzyme/​kg bodyweight per month is necessary to secure disease regression. This is very ex- pensive, costing as much as £200 000 per year for an adult. Substrate reduction therapies Miglustat  When taken for several months, miglustat (Zavesca; N-​butyldeoxynojirimycin) appears to reduce the content of gan- gliosides in circulating white cells and has salutary effects on key laboratory and clinical parameters of Gaucher’s disease activity. It is licensed in the United States of America and Europe for use in mild to moderate type 1 Gaucher’s disease, albeit with certain restric- tions. Short-​duration unwanted effects (including diarrhoea due to inhibition of intestinal disaccharidases) are frequent, although they usually respond well to dietary adjustments. The occurrence of per- ipheral neuropathy after long-​term administration may restrict the indications for its use. A trial to determine whether or not Zavesca is inferior to maintenance therapy with Cerezyme in patients with type 1 Gaucher’s disease after stable control of their disease who then switch to the oral agent has shown that for some patients the disease remains stable, but many discontinued the medication either as a re- sult of adverse effects or re-​emergent features of disease. Eliglustat  In the ENGAGE trial, 40 patients were randomized to eliglustat or placebo for 9 months. Salutary changes in the key parameters of spleen and liver volume, haemoglobin, and platelet count were observed, to the degree expected of enzyme therapy. In the ENCORE study, 160 patients were randomized to receive Fig. 12.8.8  Electron micrograph showing the cytoplasm of a Gaucher cell in the spleen of a 56-​year-​old man removed because of life-​ threatening thrombocytopenia and pain due to a recent splenic infarct. Note the vesicular spaces filled with fibrillary glycolipid storage material.

section 12  Metabolic disorders 2144 eliglustat (n = 106) or continued enzyme therapy (n = 54) after at least 3 years of enzyme therapy. The trial met its primary endpoint demonstrating noninferiority of eliglustat in composite haemato- logical and visceral parameters over 12 months, and 4-​year follow-​ up data showed safe stabilization in most patients. Use is restricted to individuals who are not ultra-​rapid metabolizers of the drug by the CYP2D6 enzyme. Complex drug–​drug interactions require careful prescribing practice, but safety and tolerability have been ac- ceptable in the clinical trial programme, with no sustained or major safety concerns hitherto. Other aspects of treatment Treatment for Gaucher’s disease should include appropriate immun- ization and antimicrobial prophylaxis in the fortunately diminishing number of patients who have undergone splenectomy. Osteoporosis may be an indication for bisphosphonate drugs. Patients may require joint replacement surgery to ameliorate the effects of bone infarction crises and, in rare instances, liver transplantation for end-​stage liver disease. All surgical procedures carry a risk of haemorrhage in the face of thrombocytopenia, platelet dysfunction, or blood coagula- tion factor abnormalities. It is thus critically important to engage expert assistance from a haematologist in planning surgical inter- ventions. Bone marrow transplantation probably does not have a role today, except in rare circumstances. Evidence of metabolic bone disease complicating the disorder should be always sought and osteoporosis should be treated promptly with enzyme replacement therapy, with the additional consideration of orally active or parenteral bisphosphonates. Where present, a defi- ciency of 25-​hydroxyvitamin D should probably be treated with ap- propriate supplements. Some patients develop deficiency of vitamin B12 and this should be sought for and treated promptly. On account of the increased risk of infection due to intrinsic chemotactic and phagocytic defects as well as splenectomy, patients with Gaucher’s disease undergoing surgery or with systemic infec- tion should be promptly treated, preferably with parenteral anti- microbial agents. Fabry’s disease Fabry’s disease is an X-​linked disorder, unlike many of the lyso- somal diseases, apart from Danon’s and Hunter’s disease (MPS II). Deficiency of α-​galactosidase A  causes the accumulation of ceramide trihexoside (otherwise known as globotriaosylceramide) and related compounds including the deacylated equivalent lyso-​ GB3, which principally derives from the breakdown of lipids present in senescent red cells. A notable feature is the presence of clinical signs and symptoms in most heterozygous female carriers of the condition. Although these manifestations are usually less severe and of later onset than in affected hemizygous males, florid and life-​ shortening clinical disease has often been observed (and ignored) in affected women. Clinical features and prognosis The most characteristic symptoms of the ‘classical’ or severe form of the disease, usually indicative of absent or very low enzyme activity, are the onset in early childhood of lancinating pain with background burning sensations in the extremities that are made worse by exercise and exposure to extremes of temperature. These attacks can be very disabling and represent neuropathic pain, which is difficult to control. The acroparaesthesias are often attributed to Raynaud’s phenomenon, which is indeed associated with Fabry’s disease, but this relationship is unclear. Nonetheless, many of the symptoms of Fabry’s disease can be explained by neuropathy affecting autonomic nervous tone. Patients with Fabry’s disease have disturbing gastrointestinal symptoms, characterized by diarrhoea shortly after eating; attacks of abdominal pain associated with unexplained fever also occur. These abdominal symptoms may also be related to autonomic neuropathy. Most men with established ‘classical’ disease notice a striking ab- sence of peripheral sweating, and often suffer erectile dysfunction. They often have a characteristic facial appearance (Fig. 12.8.9). High-​tone loss of hearing is also a common feature of Fabry’s dis- ease, which appears to reflect selective injury to cochlear neurons. Affected male hemizygotes have small, raised, red vascular skin le- sions (angiokeratomas) particularly around the buttocks and genital region (Fig. 12.8.10). These lesions are often detected in limited areas of affected heterozygous females and reflect X-​chromosome inactivation patterns in the skin. With increasing age, progressive tubular, interstitial, and glom- erular disease leads to proteinuria and renal failure. Many patients require renal support, including haemodialysis, peritoneal dialysis, or kidney transplantation. Cardiac hypertrophy, especially of the left ventricle, occurs with conduction defects leading to a shortened PR interval and a pro- longed QRS complex, later accompanied by tachyarrhythmias and complete heart block. Left ventricular hypertrophy may be associ- ated with functional limitation due to diastolic dysfunction. The use of MRI has drawn attention to typical patterns of fibrosis in the mid-​ wall of the myocardium in particular regions of the left ventricle. This fibrosis may progress in the absence of cardiac hypertrophy, particularly in women. It is associated with risk of arrhythmia and reduced response to specific therapy. There is an increasing recognition of variant forms of Fabry’s disease, which appear to be predominantly manifested by cardiomyopathy—​without the ‘classical’ acroparaesthesia, anhidrosis, and angiokeratomas—​in older patients with appreciable residual α-​galactosidase activity. Disease of capillaries and medium-​sized vessels in the brain is associated with unusual microvascular changes, particularly in the posterior cerebral circulation, and also causes stroke. Disease expression in many carrier females, who may rarely de- velop renal failure, is often accompanied by angiokeratomas that are seen to be restricted to certain dermatomes on careful examination, and asymptomatic corneal opacification with whorl-​like cataracts on slit-​lamp examination. Sudden cardiac arrhythmias, stroke, and renal failure are the most common causes of death in patients with Fabry’s disease. In men with the classical form of the condition and in the absence of specific or supportive treatment, death occurs at a median age of 48 to 49 years, with a greatly reduced quality of life during the antecedent symp- tomatic period. Life expectancy in affected heterozygous women is also shortened. Sometimes the lancinating acroparaesthesias are sufficient to cause severe depression and even suicide. Diagnosis Diagnosis is made by demonstrating the abnormal glycolipid in urine or plasma, as well as by assay of α-​galactosidase A in tears, plasma, white cells, dry blood spots, or other tissue material.

12.8  Lysosomal disease 2145 (a) (b) (c) (d) (g) (e) (f) (h) (i) Fig. 12.8.9  Facial images of nine men with the classical form of Fabry disease; although subtle, the facial appearances include periorbital oedema, depressed nasal bridge, prominent brow ridge, and full lips.

section 12  Metabolic disorders 2146 Molecular analysis of the α-​galactosidase A gene on the long arm of the X chromosome is worthwhile because it allows the unambiguous detection of female heterozygotes and may thus be useful during the reproductive period, particularly for antenatal diagnosis. Despite the presence of active disease, ceramide trihexoside concentrations and α-​galactosidase A assays are often within normal limits in af- fected female heterozygotes. Treatment Supportive care Hitherto, the treatment for Fabry’s disease has been palliative, involving the use of anticonvulsants (including gabapentin) for the acroparaesthesias and neuropathic pain. Gastrointestinal symptoms sometimes respond to antimotility agents or to pancreatic enzyme supplements, but these agents have not been subjected to controlled trials. Renal failure is managed by dialysis or by renal transplant- ation; occasionally, cardiac transplantation has been required for cardiomyopathy; pacemakers and antiarrhythmic drugs may also be needed. Specific therapies To date, two preparations of recombinant human α-​galactosidase A have been licensed: agalsidase-​alfa (Replagal—​not approved in the United States of America) and agalsidase-​beta (Fabrazyme). These may differ slightly in their post-​translational glycosylation status for delivery to endothelial, epithelial, and other cells that represent the pathological focus of this disease. Administration of these prepar- ations to male hemizygotes has improved lipid accumulation in the plasma and in renal biopsy samples. Both products have also been shown in double-​blind, placebo-​controlled trials to improve clinical endpoints of the disease, including neuropathic pain, stabilization of renal function, and ventricular mass, as well as conduction defects that represent infiltrative cardiomyopathy, but substantial reversal of established organ malfunction has not been achieved. Unlike Gaucher’s disease, targeting to the affected cells and tis- sues in Fabry’s disease probably results from receptor-​mediated uptake of protein molecules harbouring the common lysosomal recognition marker, mannose 6-​phosphate, a less efficient and less specifically targeted system. In one remarkable instance, therapy with galactose infusions appears to have mitigated this condition by stabilizing the nascent mutant enzyme, thereby enhancing residual α-​galactosidase A activity with slow clearance of cardiac glycolipid storage. Clinical trials of the pharmacological chaperone1-​ deoxyglactonojirimycin, migalastat, which is predicted to stabilize certain residual α-​galactosidase A variants in Fabry’s disease and, by preventing misfolding, increase their delivery to the lysosome, have been reported. In patients naïve to therapy, who were found to have mutations amenable to the chaperone effect, statistically significant albeit modest reductions in the storage material were seen on histological analysis of kidney biopsy samples. In patients already on enzyme therapy, a further trial demonstrated that no additional decline in renal filtration function took place in the group randomized to migalastat. In both trials, reductions in left ventricular mass index were observed, although the full therapeutic meaning and clin- ical impact of this finding needs to be established. Migalastat has received marketing approval in the United States of America and Europe. Because of its distinct mechanism of action, which requires the binding of this inhibitory molecule to the active site of the en- zyme to achieve better folding, the drug is given in an alternate-​day regimen to permit disengagement of the inhibitor from the enzyme once it has reached the lysosome. (a) (b) Fig. 12.8.10  Two patients with Fabry disease showing (a) diffuse telangiectatic lesions over flank and abdomen, and (b) hemispheric papules in suprapubic area. Reproduced with permission from Mulliken J. Capillary Malformations, Hyperkeratotic Stains, Telangiectasias, and Miscellaneous Vascular Blots. From: Mulliken and Young's Vascular Anomalies: Hemangiomas and Malformations, 2nd edition. Ed. John B. Mulliken, Patricia E. Burrows, and Steven J. Fishman. 2013. Courtesy of Dr Harley A. Haynes.

12.8  Lysosomal disease 2147 Mucopolysaccharidoses These disorders are caused by a deficiency of lysosomal hydrolases that catalyse the cleavage of complex glycosaminoglycans, which are macromolecular components of connective tissues including joints, bones, heart, and major arteries. Clinical manifestations of each of these disorders reflect an individual enzymatic deficiency and the resulting accumulation of mucopolysaccharide derivatives, of which dermatan-, keratan-, chondroitin-, and heparan sulphates are the principal components. In general, the accumulation of the complex substrates that are normally linked to proteins to form proteoglycans is associated with visceral enlargement, heart valve disease—as well as bony abnormalities, joint stiffness, corneal clouding and short stature. The accumulation of heparan sulphate may particularly be associated with the development of brain disease, including thick- ening of the leptomeninges, hence hydrocephalus is an often neg- lected factor in cerebral impairment that may also be attributed to lysosomal storage affecting neurons of the brain and peripheral gan- glia as well as the retina. Clinical features and pathology Typically, these disorders are associated with coarse facial features (Figs. 12.8.11 and 12.8.12), bone shortening, and skeletal abnor- malities, as well as disturbances of dentition, the gums, and middle ear. Abnormalities of the tracheobronchial cartilages and upper air- ways may be associated with respiratory infections and obstructive lung disease. The coronary arteries and cardiac valves may be infil- trated by glycosaminoglycans, leading to coronary artery occlusion and aortic and mitral valve malfunction. Similar changes may occur in peripheral arteries, particularly those supplying the viscera. In the eye, the basal layers of the cornea show swelling, cytoplasmic vacuolization, and storage granules leading to opacification. Scleral thickening may impinge upon the optic nerve. Excess urinary excretion of glycosaminoglycan products, including dermatan sulphate and heparan sulphate, character- istically occur in the mucopolysaccharidoses. This abnormality should immediately prompt further investigations by enzymatic and genetic studies in blood leucocytes and/​or fibroblasts obtained from cultured skin biopsy samples. The inheritance pattern of the mucopolysaccharidoses is typical of autosomal recessive traits with the exception of Hunter’s disease (MPS II, which is due to iduronate sulphatase deficiency) that maps to the X chromosome and is ex- pressed predominantly in boys and men. Female heterozygotes for Hunter’s disease only very rarely show evidence of neurological im- pairment or connective tissue abnormalities. Treatment Palliative treatment is a very important aspect of the management of these diseases and should include the provision of multidisciplinary support for children and young adults with the accompanying devel- opmental disabilities. Sustained provision for the long-​term man- agement of the condition in affected families is desirable. Surgical procedures Corneal transplantation may be required to improve vision where retinal degeneration is not dominant. Carpel tunnel syndrome with compression neuropathy of the median nerve is very common and, when indicated, surgical treatment is often beneficial. Particular care is required in patients with mucopolysaccharidoses such as Hurler’s syndrome when surgical procedures under general anaesthetic are required for relief of hydrocephalus, myringotomy, hernia repair, relief of airways obstruction due to laryngeal disease, and corrective spinal or joint surgery. Infiltration of the soft tissues of the upper and lower airways, as well as the heart and cervical spine (which may include subluxation of the atlanto-​occipital joint), is associated with high perioperative mortality. Tracheostomy may be required to avoid life-​threatening complications of intubation, Fig. 12.8.11  Facial image of a male patient aged 20 years with MPS I (Hurler’s syndrome) who underwent a bone marrow transplant at the age of 1 year; some coarsening of the facial appearance persists. Fig. 12.8.12  Twenty-​year-​old female patient with Morquio A syndrome (MPS IV) with marked skeletal deformity and growth retardation.

section 12  Metabolic disorders 2148 but complications may arise with general anaesthesia beyond that of difficulties with endotracheal intubation. Extensive preoperative assessment should therefore be conducted whenever an anaesthetic is required for any procedure, particularly to assess the stability of the atlantoaxial joint, the airway, and the presence of coronary artery disease (that may predispose to perioperative myocardial infarc- tion). Where possible, an anaesthetist experienced in the manage- ment of patients with MPS disorders should be consulted. Specific treatments Bone marrow transplantation  Bone marrow transplantation using HLA-​identical sibling and HLA-​matched nonsibling donors has been investigated extensively in the mucopolysaccharidoses. Long-​term clinical trials have confirmed the beneficial effects of successful transplantation with reversal of hepatosplenomegaly and obstructive airways disease. In some cases there is improved longevity, with a possible reduction also in the incidence of sec- ondary hydrocephalus. However, at present, transplantation does not cure the condition and is unable to reverse established brain injury and most of the crippling skeletal manifestations. If it is to be considered, bone marrow transplantation should therefore be carried out early in the course of these diseases. The therapeutic position of bone marrow and cord blood-​derived stem cell therapy is most clearly established for Hurler’s disease (the more severe variant of MPS I). Enzyme replacement therapies  Enzyme replacement therapy has long been under investigation in MPS I (Hurler’s syndrome, Hurler–​ Scheie syndrome, and Scheie’s syndrome), which was one of the first of such disorders to be subjected to intensive laboratory study. In clinical trials, recombinant human α-​l-​iduronidase, now licensed as laronidase (Aldurazyme) given by weekly infusion intravenously, after 1 year clearly showed a reduction in lysosomal storage: liver volume decreased and there was an improved rate of growth as well as improvement in the range of joint movements at sites character- istic of connective tissue infiltration in this condition. With a reduc- tion in the storage material in the upper airways, there was also an improvement in episodes of hypoventilation during sleep. After a few weeks of enzyme treatment, urinary glycosaminoglycans abnor- malities were corrected. Although many patients developed serum antibodies, only transient immune reactions, including urticaria, occurred during the infusions. Enzyme replacement therapies have received market authoriza- tion for patients suffering from MPS II (Hunter’s syndrome with iduronate sulphatase deficiency) and MPS VI (Maroteaux–​Lamy disease due to arylsulphatase B deficiency) following successful clinical trials. Enzyme therapy with elosulfase alfa for MPS IVA (Morquio A) has been evaluated in clinical trials: benefits are re- ported in clinical measures of endurance and lung function and the therapy has received marketing approval in the United States of America and Europe, although the complexities of funding delayed patient access in the United Kingdom. Favourable responses to en- zyme replacement therapy have also been reported in animal models of related disorders, including the cone-​head mouse that represents a faithful model of MPS VII (Sly’s disease), due to deficiency of acid β-​glucuronidase. Following results of a clinical trial in the exception- ally rare MPSVII, vestronidase alfa (MEPSEVII) received marketing authorisation from the U.S. Food and Drug Administration in 2017. Questions still arise of how clinical benefits and an improved quality of life can be best assessed. However, encouraging results showing an improved quality of life, mobility, nutrition, and educa- tional achievements have already been documented in several MPS disorders in response to enzyme therapy, even where pre-​existing developmental effects and mental retardation are established. Pompe’s disease Glycogen storage disease type II, due to acid maltase deficiency, otherwise known as Pompe’s disease, is an autosomal recessive dis- order of glycogen metabolism caused by deficient activity of lyso- somal acid maltase—​α-​glucosidase. Acid maltase deficiency was the first of the lysosomal storage diseases to be so characterized by H.-​G. Hers, a colleague of de Duve. The disease occurs in many countries and ethnic groups. The prevalence of this disorder is one in about 150 000 and males and females are affected equally. Clinical features and pathology Pompe first reported infants with massive cardiac hypertrophy and skeletal weakness with hypotonia, enlargement of the tongue and liver, and a uniformly fatal outcome. Acid maltase releases glucose units from the carbohydrate storage macromolecule, glycogen, as well as from the disaccharide, maltose. The enzyme is profoundly deficient in infants with Pompe’s disease and partial deficiencies in the enzyme, detectable in all cells, are responsible for later-​onset forms in children, adolescents, and adults. No clear correlation be- tween the degree of enzyme deficiency and the severity of disease is possible. Pathological accumulation of glycogen within vacuolar lyso- somal spaces occurs in skeletal muscles and (on occasion) other tissues, but it is noteworthy that in certain muscles, microscopic examination may be normal or show only trivial abnormalities, especially in patients with late-​onset disease. Hence, the diagnosis of late-​onset Pompe’s disease may be difficult and routine muscle biopsies may not identify all those affected. The combined use of muscle biopsy with biochemical assays and molecular analysis of the acid glucosidase gene (at least for the common IVS1 mutation) should be considered in patients with unclassified myopathy. Acid maltase is normally responsible for constitutive autophagy and mo- lecular remodelling of intracytoplasmic glycogen; when deficient, abnormal glycogen accumulates within the lysosomal vacuole and elsewhere in the cell. In pathways not yet completely understood, this pathological accumulation is associated with tissue injury, but large cytoplasmic collections of autophagic debris appear to disrupt the contractile apparatus. Since glycogen is a storage molecule abun- dant in muscle cells, it is these cells that are the principal focus of acid maltase deficiency. Patients with infantile onset of symptoms have predomin- antly skeletal muscle disease, hypertrophic cardiomyopathy, or macroglossia; hepatomegaly is not a feature of Pompe’s disease in the absence of cardiac failure. Onset of disease in children and adults with weakness and poor athletic performance is associated with de- layed achievement of developmental motor milestones. Ultimately the clinical appearance is dominated by proximal muscle weakness with lordosis of the spine; patients adopt the Gower manoeuvre in rising from the squatting position. Late-​onset forms observed in adults usually present as a progres- sive proximal myopathy with the variable addition of diaphragmatic

12.8  Lysosomal disease 2149 and respiratory muscle paralysis leading to respiratory failure, but the rate of progression is unpredictable. The onset of symptoms varies between the age of 10 and 60 years. In most patients there is a history of longstanding proximal weakness with involvement of the truncal muscles and weakness in the hips in advance of the upper limb girdle. Poor physical strength and failure in gymnastic activities may be the clue. In children and adolescents, the condition may be misdiagnosed as a late-​onset muscular dystrophy or even polymyositis, leading to inappropriate treatment. Ultimately the progressive proximal weakness is apparent and associated with re- spiratory failure: the latter is presaged by fatigue, breathlessness on exertion, and sleepiness due to marked ventilatory failure—​carbon dioxide retention causes morning headaches. Occasionally, dys- phagia for solids may result from weakness of voluntary pharyngeal muscles that initiate swallowing. Treatment Alglucosidase alfa (Myozyme) has been developed as a mannose 6-​phosphate-​containing recombinant human acid α-​glucosidase (rhGAA) for the treatment of patients of any age with Pompe’s dis- ease (GSD II). Enzyme replacement therapy is administered to re- store enzymatic activity, deplete accumulated glycogen, and prevent its further accumulation to allow repair of damaged myocytes. In the very severe infantile form of the condition, where survival beyond 1 year of age is unusual, treatment with Myozyme has been associated with prolonged survival. In a trial of rhGAA in infants aged 6 months or younger, all were alive at 18 months whereas only 2% of the historical cohort group survived to this age. Most patients treated with rhGAA had normal growth and significant motor de- velopment during the treatment period. In another report, two se- verely affected (wheelchair-​ and ventilator-​dependent) patients remained stable during an 8-​year period of enzyme therapy, and in a third, moderately affected patient, muscle strength improved mark- edly and the ability to walk was regained. In some instances, however, the outcome has been disappointing and, in general, better outcomes are seen with early treatment and in patients who do not develop high-titre antibody responses to the re- combinant protein. Recently, in infants predicted to form high-titre, neutralising antibody responses on the basis of mutation analysis, immune-​modifying treatments have been given at the outset of en- zyme therapy in an attempt to tolerize the immune system. Taken as a whole, the efficacy of enzyme replacement therapy for acid maltase deficiency emphasizes the need for prompt clinical recognition and diagnosis, especially in infants and young children. Several studies have confirmed the therapeutic efficacy of rhGAA in patients suffering from attenuated forms of Pompe’s disease. In the present authors’ experience with adult patients suffering from acid maltase deficiency, improvements in skeletal and respiratory muscle function are seen in the first year of treatment, with stability or a slower rate of decline maintained thereafter. It is unclear if, once lost, diaphragmatic function can be regained; we contend that re- stricting treatment to those with severely weak and wasted limbs and respiratory failure due to diaphragmatic paralysis will greatly underestimate its capacity to improve life quality or restore the func- tion of injured muscles. A second-​generation enzyme preparation is undergoing clinical trial evaluation with the aim of improving the delivery of enzyme to the lysosomal compartment of skeletal muscle. The Genzyme (Sanofi) agent is a form of α-​glucosidase modified chemically to display many times more mannose phosphate than the parent com- pound, Myozyme. Even with specific treatment, the role of physical therapy, respira- tory assessment and support, nutritional care, and measures aimed at general rehabilitation remain crucial for functional outcome and improved quality of life. Niemann–​Pick diseases Niemann–​Pick disease types A and B Niemann–​Pick disease types A and B are, respectively, neuronopathic and non-​neuronopathic variants of acid sphingomyelinase de- ficiency, a sphingolipid disorder leading to the accumulation of sphingomyelin. The condition resembles many of the manifest- ations of Gaucher’s disease, with a characteristic secondary storage cell which is also a macrophage. The Niemann–​Pick cell has a foamy appearance rather than the characteristic striated cytoplasm of the Gaucher cell: there is prominent infiltration of the lungs as well as the marrow cavity, liver and spleen. Niemann–​Pick disease type A  is associated with disabling neuronopathic features and dementia in infants and young chil- dren. Niemann–​Pick disease type B may occur in adults who have only trivial splenomegaly and minor pulmonary infiltrates that are only exacerbated at times of intercurrent chest infection; they are at risk from osseous disease related to marrow infiltration, as with Gaucher’s disease. At present, no specific treatments are available apart from the prompt treatment of pulmonary infection and the management of the consequences of skeletal infiltrates and episodes of avascular ne- crosis. Some patients, including those previously misdiagnosed as having Gaucher’s disease, may have undergone splenectomy to relieve pressure symptoms or the haematological effects of hypersplenism. Since this disease is primarily a disorder of macrophages, it should be susceptible to enzymatic complementation using the mannose receptor. At the time of writing, clinical research to develop recom- binant human acid sphingomyelinase is well advanced. Clinical trials were delayed to address safety concerns from studies in animal models in which high-​dose therapy led to a fatal inflammatory reac- tion, believed to be due to release of bioactive ceramide, the product of the catalytic reaction. After the completion of a gradual dose escalation study to min- imize release of ceramide product from accumulated substrate, a trial in adult patients with Niemann Pick type B disease is now continuing as an open-​label phase II/​III clinical trial to evaluate the safety and efficacy of different doses of recombinant acid sphingomyelinase when administered once every 2 weeks. In children up to the age of 18 years with this condition, a one-​year phase I and II multicentre, open-​label clinical trial to evaluate the safety and tolerability of recombinant human acid sphingomyelinase administered paren- terally once every 2 weeks is recruiting patients. It is intended that this agent will be started at 0.3 mg/​kg dose, gradually increasing to a maximum of 3 mg/​kg. The outcome of these long-​awaited studies to advance the understanding of investigational enzyme therapy will be received with great interest in this very rare but severe dis- ease. Unfortunately, since the biosynthesis of sphingomyelin is not regulated by the uridine diphosphate-​glucosylceramide synthase

section 12  Metabolic disorders 2150 reaction, and so far no clinical inhibitors of this biosynthetic step are available, any exploration of substrate reduction therapy for the severe neuronopathic manifestations of Niemann–​Pick disease type A will be long in coming. Niemann–​Pick disease type C Niemann–​Pick disease type C is a distinct disease that may pre- sent with jaundice in infants or children; the initial hepatitic illness usually resolves but may lead to fatal liver failure with cholestatic features. Intractable and progressive neurological disease occurs in childhood and early adult life, with ataxia, seizures, (vertical) supra- nuclear gaze palsy, and progressive diffuse cortical injury. Death usually occurs in the third or fourth decade. Niemann–​Pick disease type C is not due to a primary defect of acid sphingomyelinase but to mutations in two distinct lysosomal proteins, NPC1 and NPC2, that when mutated produce subtypes of the disease. The physiological role of the NPC1 transmembrane pro- tein remains unclear, as is the pathological cellular cascade that leads to NPC disease. Some investigators propose a lipid transport role for NPC1, others a role in lipid-​sensing, while some evidence points to a complex involvement in endosomal calcium flux and other sec- ondary and downstream effects. Niemann–​Pick disease type C is also associated with the appear- ance of foam cells in the macrophages; the Kupffer cells of the liver may be enlarged and a cholesterol trafficking defect is apparent in most cells. A rare complication is inflammatory bowel disease which has many features in common with Crohn’s disease and a prominent infiltrate of storage macrophages in the inflamed tissue. The molecular defect in this disease, though not manifest in the skin, may be detected in skin-​derived fibroblasts after culture and ex- posure to low-​density lipoprotein cholesterol: in Niemann–​Pick dis- ease type C, cholesterol is taken up and accumulates in intracellular droplets that stain positively with the fluorescent dye filipin. Within the brain, Niemann–​Pick disease type C causes neuronophagia and the accumulation of gangliosides and other complex sphingolipid storage products that may induce neuronal injury. The use of statins and other agents that interfere with cholesterol metabolism has not been effective in arresting the course of this cruel illness. Clinical trials using N-​butyldeoxynojirimycin (miglustat, Zavesca) have followed the delayed onset and increased survival of mice homozygous for a spontaneous mutation in the NPC1 gene that serves as an authentic model, recapitulating many features of the human disease. A randomized controlled trial and several co- hort studies have reported improvements in or stabilization of sac- cadic eye movements during 1 to 5 years of therapy. Swallowing was also shown to improve or remain stable during the randomized trial (up to 2 years). These findings were supported by long-​term obser- vational cohorts (up to 6 years). A meta-​analysis of dysphagia—​a clinically important therapeutic endpoint for the disease since as- piration pneumonia is a frequent cause of hospitalization and death—​demonstrated a clear survival benefit with miglustat that was accompanied by improved swallowing. Serial studies showed decrease in calbindin in cerebrospinal fluid during treatment, sug- gesting reduced cerebellar Purkinje cell loss, and MRI studies dem- onstrated a protective effect on cerebellar and subcortical structure that correlated with clinical symptom severity. This research led to marketing approval of the drug by the European Medicines Agency, but it has yet to be approved for use in Niemann–​Pick disease type C by the FDA in the United States of America. Treatment of Niemann–​ Pick C disease is currently the principal use of this agent, rather than type 1 Gaucher’s disease, the original indication. Recently, two candidate products for preclinical development in Niemann–​Pick type C have been identified. One is the recom- binant human heat-​shock protein HSP70, and the other is an orally available small molecule, arimoclomol, that serves to induce heat-​ shock proteins, including HSP70. These candidates have produced encouraging biochemical and disease-​modifying effects in the Niemann–​Pick type C mouse model through their actions on lyso- somal integrity and on scrambled or denatured endogenous mol- ecules. The compounds have completed toxicity studies and clinical trial results are awaited. Cyclodextrins are complex ring structures that can solubilize lipids and are widely used domestic chemicals. When given system- ically and into the central nervous system, cyclodextrin was associ- ated with slowing and/​or prevention of neurodegeneration in both a mouse and feline model of the disease. A multicentre phase III clinical trial of the intrathecal use (via lumbar puncture) of VTS-​ 270 (2-​hydroxypropyl-​β-​cyclodextrin) is underway, as is a phase I/​ IIa trial of its use in patients with neonatal hepatitis. Cyclodextrin infusions, especially via intrathecal administration, are a laborious and challenging intervention and it remains unclear how paren- teral cyclodextrin will find its therapeutic position in the long-​term management of this disease, but orally active analogues are being explored for later application in this disease. Cholesteryl ester storage disease and Wolman’s disease These are late-​onset (cholesteryl ester storage disease) and infantile (Wolman’s disease) forms of lysosomal acid lipase deficiency, which causes the accumulation of cholesteryl esters in the lysosome. Wolman’s disease is a devastating, fatal illness in which the infant fails to thrive, has massive hepatomegaly, adrenal calcification, and intestinal malabsorption. Death is almost inevitable within the first year of life. Cholesteryl ester storage disease, by contrast, manifests as a more indolent liver disease, with hepatic steatosis, progressing in many cases to fibrosis and cirrhosis. Patients have accelerated and often severe atherosclerosis with dyslipidaemia characterized by low plasma high-​density lipoprotein cholesterol and variably high low-​ density lipoprotein cholesterol concentrations. It can be difficult to distinguish cholesterol ester storage disease from other commoner causes of fatty liver disease, except that the usual risk factors for the latter (overweight, diabetes, and other features of the metabolic syndrome) are normally absent and on histological analysis of liver biopsy specimens the lipid droplets are small, indicating lysosomal rather than cytosolic location: microvesicular steatosis is thus a hall- mark of cholesterol ester storage disease and macrovesicular stea- tosis reflects the spectrum of nonalcoholic liver disease. An enzyme preparation, sebelipase alfa, has been subject to in- tensive clinical trials for these disorders. The phase I clinical study demonstrated clear and early pharmacodynamic effects in patients with cholesteryl ester storage disease. Favourable outcomes of an international, randomized, double-​blind, placebo-​controlled phase III trial of sebelipase alfa in children and adults with lysosomal acid lipase deficiency, and the phase II/​III trial of sebelipase alfa in infants

12.8  Lysosomal disease 2151 with Wolman’s disease were reported in 2014. In summary, in 66 chil- dren and adults with lysosomal acid lipase deficiency administration of the enzyme met the primary endpoint of restoring of serum alanine aminotransferase (used as a biomarker of liver injury) to the healthy reference range. The agent was administered parenterally on alternate weeks at 1 mg/​kg for the double-​blind treatment period of 20 weeks. The median age of patients enrolled in the trial was 13 years of age (range 4–​58) and fibrosis or cirrhosis was documented in all 32 pa- tients who had had baseline liver biopsy samples. In a continuing open-​ label follow-​up study, relative to placebo, markers of dyslipidaemia and liver fat content improved, with sustained reduction in markers of liver injury and further improvements in low-​density lipoprotein cholesterol. Worldwide marketing approval has been granted on the basis of these results. At the time of writing, funding for this treat- ment is available in some countries and is under negotiation in others (including the United Kingdom). There is evidence that sustained use of sebelipase may lead to a substantial reversal of fibrosis in this condi- tion and in the authors’ view, the life-​saving effects in children make a compelling case for authorized reimbursement. Danon’s disease In 1981, two cases of cardiomyopathy in male infants with skeletal myopathy and mental retardation were reported by Danon and col- leagues. The skeletal pathology suggested type II glycogenosis but no deficiency of acid maltase activity was present. Mutations in the gene encoding LAMP2, located on the X-​chromosome, have been iden- tified. LAMP2 is a highly glycosylated integral membrane protein of the lysosome with a role in mediating fusion of the autophagic vacuole with the lysosome. Deficiency leads to accumulation of vacuoles containing autophagic debris, including mitochondria and granular deposits of glycogen. In affected males, the clinical features include a dramatic hyper- trophic cardiomyopathy, a mild skeletal myopathy, and mild to mod- erate learning difficulties. The cardiomyopathy is particularly prone to give rise to malignant ventricular arrhythmias. Before the intro- duction of implanted defibrillation devices, the median age of death of classically affected hemizygotes was about 20  years. A  milder phenotype, apparently restricted to the heart, is seen in heterozy- gous women. Apart from supportive measures, no specific treatment is currently available, although a few hemizygous male patients have been successfully treated by cardiac transplantation. Mutations in the LAMP2 gene have been found at high frequency (6%) in men with unexplained severe hypertrophic cardiomyopathy. Diseases recently attributed to lysosomal dysfunction The characterization of lysosomal defects in several ill-​understood disorders with diverse clinical manifestations continues to reveal much about the role of the lysosome in cellular functions of sig- nificance in medicine and molecular physiology. Several recently studied lysosomal diseases in this category are briefly described here. Neuronal ceroid lipofuscinoses Clinical features, genetic basis, and pathology The neuronal ceroid lipofuscinoses are the most common group of progressive brain diseases that usually affect children and young adults; 13 independent genetic groups have so far been identified with an estimated incidence of 1 in 12,500 live births. Ceroid lipofuscinosis, neuronal type 1 (CLN1) is due to mutations in a gene encoding palmitoyl:protein thioesterase 1, an enzyme in- volved in lysosomal degradation of acetylated proteins. CLN2 is due to defects in the gene encoding the acid hydrolase, tripeptidyl-​ peptidase. CLN3 is the most frequent form and particularly common in Nordic countries; it was the first lysosomal disease ever to have been reported in the literature (in a Norwegian family) in 1826, and is due to deficiency of a lysosomal transmembrane protein that may serve as a transporter molecule. Childhood forms of these disorders are almost invariably inherited as recessive traits and result in a progressive dementia combined with epilepsy (sometimes myoclonic), blindness, and early death. The family history may, however, suggest dominant transmission of CLN11, a puzzling recessive disease caused by mutations in the GRN gene encoding progranulin, with confusion arising because hetero- zygotes develop frontotemporal lobar degeneration with ubiquitin-​ positive inclusions (Online Mendelian Inheritance in Man (OMIM) 607485). In only one of these conditions, Kufs adult-​onset neuronal lipofuscinosis, is inheritance of a single copy of the mutant CLN4/​ DNAJC5 gene both necessary and sufficient to cause disease, which is always transmitted as an autosomal dominant. In several instances, neuronal ceroid lipofuscinoses represents defects in elements of intralysosomal protein catabolism, indi- cating that the turnover of the cognate proteins is very high in cor- tical neurons. Realization that the neuronal ceroid lipofuscinoses represent inherited disorders of lysosomal protein metabolism is very recent, but the discovery clearly has important consequences for better understanding the pathology of this family of cruel neurodegenerative disorders and for developing better diagnostic tools (especially for prenatal application) as well as innovative treatments The most familiar form of these diseases has, in Anglophone coun- tries, been widely termed ‘Batten’s disease’. In 1915, Dr F.E. Batten had, at a time when these disorders fell into the descriptive category of ‘familial amaurotic idiocy’, correctly differentiated infantile neur- onal ceroid lipofuscinosis from Tay–​Sachs disease. However, with identification of the biochemical causation and responsible genetic loci, this terminology and many other eponymous terms from the medical literature of the 19th and 20th centuries now has little prac- tical value in the enlarging canon of lysosomal disorders. The striking pleiotropic effects of mutations at loci responsible for the neuronal ceroid lipofuscinoses reflect the severe impairments of multiple lysosomal functions which cause this class of exclusively neurodegenerative diseases. Several of the genes encode lysosomal proteins, including acid hydrolases (CLN1, CLN2, CLN10, CLN13); a soluble lysosomal protein in CLN11; a protein, progranulin, that functions in the secretory pathway; two cytoplasmic proteins that interact with lysosomal membranes (CLN4, CLN14); and many transmembrane proteins with diverse subcellular locations (CLN3, CLN6, CLN7, CLN8—​and the lysosomal ATPase, type 13A2 in CLN12). Pathological studies show the characteristic accumulation of autofluorescent storage debris (lipofuscin) within neurons and lysosomes in other cells; this material consists of several oxidized and ubiquitinated proteins and often includes soluble cytochrome C derived from the mitochondrial F1ATPase complex and saposin fragments. The storage of this material occurs preferentially in

section 12  Metabolic disorders 2152 lysosomes of the nervous system and is associated with progressive neuronal death leading to a marked atrophy of the brain; cerebral atrophy is particularly obvious in the early-​onset forms of the neur- onal lipofuscinoses. Diagnosis Diagnosis of these diseases requires clinical persistence, which is driven by the need for clarity and to provide genetic and prognostic advice to family members and carers. The electroencephalogram is usually informative with early development of occipital spike poten- tials after photic stimulation. MRI shows atrophy, characteristically first in the cerebellum and vermis but which progresses to general- ized cerebral atrophy, ultimately with profound shrinkage (unlike GM2 gangliosidosis from which the CLN syndromes need to be distinguished). The presence on a blood smear of vacuoles in lymphocytes in a juvenile disorder would be typical of CLN3 disease. Enzymatic tests conducted on white cell pellets can readily define suspected CLN1 or CLN2 disease. Ultrastructural studies by electron microscopy in blood cells and fibroblasts may demonstrate the characteristic storage deposits, usually in lysosomal structures, in blood cells or tissue specimens. Advances in molecular diagnostics allow the iden- tification of defective genes and their protein products in several dis- tinct clinical phenotypes. Supportive and symptomatic management This clinically heterogeneous family of relentless neurodegenerative diseases poses great challenges: the provision of continuing care for what in most cases is a chronic, cruel, and fatal disease affecting chil- dren and young adults. As described, neuronal ceroid lipofuscinoses are characterized by dementia (one of the most frequent causes in young persons), epilepsy, motor deterioration, and visual loss, and for most there is currently no specific therapy and little prospect of therapy. Discovery of the genetic basis of 14 clinical variants permits prenatal and postnatal diagnosis in affected pedigrees by molecular analysis of genomic DNA that is of key importance for provision of genetic counselling. Beyond an experimental therapy, and one approved molecular therapy, palliative measures are employed for symptom relief. Much distress accompanies loss of the ability to swallow and vocalize and of independent movement. While the use of feeding gastrostomies is critical for maintaining hydration and giving regular medication to treat epilepsy and muscle spasms, death is usually inevitable in CLN2 by the early-​mid teenage years, often from aspiration pneumonia. The response to anticonvulsant medication is best judged by the level of symptomatic relief and the minimum effective dose for rea- sonable clinical control should be used. Complex drug regimens, especially those that use more than two drugs, often compound the unwanted effects and are counterproductive. Sodium valproate and lamotrigine are preferred, and phenytoin, carbamazepine, and vigabatrin (and topiramate in CLN2 disease) are probably best avoided in patients with neuronal ceroid lipofuscinoses. In generalized persistent seizures, diazepam and/​or lorazepam are used in the short term to gain control. Under these circum- stances there may be a place for introducing phenobarbitone for frequent severe attacks, and myoclonic seizures may also respond to ethosuximide. Another valuable measure, when frequent severe attacks occur, is the introduction of a ketogenic diet, which is often of decisive benefit in patients whose seizures are otherwise very challenging. Myoclonus may be exacerbated by carbamazepine and gabapentin (and pregabalin) as well lamotrigine, hence careful reduction of the anticonvulsant regimen may help to control this often distressing manifestation. Zonisamide has been reported to control myoclonus, and levetiracetam and piracetam may also have therapeutic effects. Spasticity requires prompt and enthusiastic use of physiotherapy and sometimes splinting to prevent or mitigate the tendency for development of fixed flexural deformities with painful spasms and emergence of pressure sores. Local use of botulinum toxin may as- sist in severe cases. The spasmolytic agents baclofen and tizanidine can be effective, and there are preliminary reports that tetrahydro- cannabinol may be useful in some patients. Benzodiazepines are no longer popular since their unwanted effects (excessive drowsiness and dribbling) are so unwelcome. Experimental and specific therapies Cysteamine  In vitro studies suggested that the use of the lysosomotropic thiol agent, cysteamine, may activate residual palmitoyl-​protein thioesterase activity in patients with CLN1 or solubilize the intralysosomal ceroid in this disease. Cysteamine bitartrate (Cystagon), used in the treatment of cystinosis, was ex- plored in a 7-​year, open-​label, substrate-​reduction therapy trial in four patients with atypical, juvenile-​onset CLN1. Five untreated pa- tients with the same CLN1 mutations, three of whom were siblings, were included as controls. The treatment substantially decreased the storage material in peripheral lymphocytes in a dose-​dependent manner, and a minor slowing of the disease progression compared to the controls was observed in three out of the four treated patients. A long-​term pilot trial of oral cysteamine bitartrate and N-​ acetylcysteine was conducted in 10 children below 3 years of age with any two of the seven most lethal CLN1 mutations. Outcomes in nine patients after follow-​up for 8 to 75 months were compared with the reported natural history of the disease and that of affected older siblings. While no trial participant acquired new skills and retinal function decreased progressively, the average time to an isoelectric EEG (52 months) was longer than in historical controls (36 months), and parents and physicians reported less irritability, improved alert- ness, or both in seven patients. It seems unlikely that this treatment will be used widely. Enzymatic augmentation and gene therapy in CLN2  Recombinant human tripeptidyl peptidase 1 was investigated as a potential treat- ment for CLN2 disease in children aged 3 to 16 in an open-​label, multicentre clinical trial. The protein, termed cerliponase-​alfa, at 30, 100, and 300 mg was initially infused at 2-​weekly intervals into cerebral ventricles through an in-​dwelling device, later maintained at 300 mg over at least 96 weeks. Outcome was determined by de- cline in a motor-​language disease score, with data compared with a study conducted on the course of the disease in historical controls. The mean (± standard deviation) unadjusted rate of decline in the motor-​language score per 48-​week period was 0.27 ± 0.35 points in treated patients and 2.12 ± 0.98 points in 42 historical controls (mean difference, 1.85). Cerliponase alfa received global marketing approval as Brienura from the FDA (for children >3  years) and European Medicines Agency in 2017 for patients at any age. At the time of writing in the United Kingdom, NICE has given provisional

12.8  Lysosomal disease 2153 approval for Brienura with limited reimbursement from the National Health Service in England under a managed access programme. Anecdotal reports and post-​marketing information provides some reassurance that this intensive and laborious intervention provides clinically useful benefit and stabilization of disease in otherwise stricken children. It is not yet known whether the therapy will delay the onset of blindness. Two small-​scale early-​phase clinical trials of gene therapy have been conducted in infants with CLN2 using recombinant adeno-​associated vectors delivered intracranially. Two distinct vector serotypes with potentially different cellular tropisms have been used. At the time of writing no beneficial outcomes have been reported from either trial. Papillon–​Lefèvre syndrome This is an unusual syndrome, inherited as an autosomal reces- sive trait, resulting in periodontal disease with tooth loss and palmoplantar keratosis that is associated with a selective deficiency of cathepsin C activity within the azurophil granules of neutrophilic polymorphonuclear leucocytes. Several mutations have been identi- fied within the gene encoding cathepsin C, which is an exo-​cysteine protease, also known as dipeptidyl peptidase I, that serves as a multi- faceted scaffold on which numerous chymotrypsin-​like proteases are activated during neutrophil maturation. These include granule serine peptidases such as elastase, cathepsin G and proteinase 3 in neutrophils, and chymase and tryptase in mast cells; a partial role in the activation of granzyme B, a key effector system of natural killer cells, has been suspected from animal studies but only described in one affected human pedigree. Deficiency of antimicrobial pep- tides released during the normal inflammatory process has also been shown. A more severe allelic variant known as Haim–​Munk syndrome, originally reported from Cochin in Southern India, is associated with onychogryphosis, pes planus, arachnodactyly, and osteolysis involving the distal phalanges (acro-​osteolysis). It appears that the enzyme deficiency leads to the failure of bac- terial clearance in the gums, thereby causing destructive periodon- titis and tooth loss. The corresponding role of cathepsin C within the dermal epithelium is not known, but a failure of cathepsin C activity reproducibly leads to epithelial abnormalities and thickening of the skin, particularly on the soles of the feet. Some patients with disabling skin manifestations have obtained benefit by the use of retinoids, with or without antimicrobial therapy. These agents are, however, unlikely to improve early-​onset destructive periodontal disease which leads to loss of primary and secondary dentition. Recurrent oral infection with Aggregatibacter actinomycetemcomitans infection has been reported. The importance of the Papillon–​Lefèvre syndrome rests not only on the identification of lysosomal cathepsin C as an important com- ponent of immune defences against bacteria that preferentially in- vade the privileged periodontal site, but also on the involvement of this enzyme in the normal turnover of keratinized skin as well as defence against microbial invasion. Spondyloenchondrodysplasia with immune dysregulation This autosomal recessive skeletal dysplasia with intracranial calcifi- cation had been long recognized, but association with deficiency of the lysosomal tartrate-​resistant iron-​containing purple (type 5) acid phosphatase and diverse clinical manifestations of autoimmunity, including lupus erythematosus, has been recent. The type 5 acid phosphatase is a readily measured lysosomal en- zyme expressed in osteoclasts and pathological macrophages. In healthy persons, the enzyme is also abundant in Langerhans and dendritic cells. Apart from its ability to degrade skeletal phospho- proteins, including osteopontin, it probably modulates the effector pathways of phagocytic activation or antigen presentation. There is little or no evidence of ‘storage’: the disease illustrates the extraordinary diversity of lysosomal functions in the whole animal and how genetic disturbances can induce wide-​ranging clinical ef- fects. Enchondromatous lesions are seen in long bones with sclerosis and irregularity of the metaphyseal plate. Lateral spine radiographs reveal platyspondyly and irregularity of the vertebral endplates. There is intracranial calcification in the basal ganglia, thalami and deep cerebral gyri. The manifestations of autoimmunity are characterized by elevated antinuclear antibody and anti-​double-​stranded DNA antibody titres with hypocomplementaemia. The clinical course in patients is varied but generally florid, with hypothyroidism, vitiligo, thrombocytopenia requiring splenectomy, autoimmune haemolytic anaemia, hepatosplenomegaly, nonerosive arthropathy, and vascu- litic skin eruptions. Defects of organelle assembly: Chédiak–​Higashi, Griscelli’s, and Hermansky–​Pudlak syndromes Inherited defects of protein complexes that participate in the bio- genesis of lysosomes and their related secretory organelles such as melanosomes are increasingly being recognized. The organ- elles with specialized functions that closely resemble lysosomes are termed lysosome-​related organelles and include δ-​granules in platelets; Weibel–​Palade bodies of endothelial cells; lytic granules and vesicles implicated in the immune ‘synapse’ in lymphocytes; basophil and azurophil granules in polymorphonuclear leucocytes; lamellar bodies in type 2 pneumocytes; neuromelanin granules in the catecholaminergic neurones of the nigro-​strial pathway, and the melanosomes of the iris, choroid, and skin. Most of these organelles maintain an acidic intra-​organellar milieu, but while they often em- ploy the lysosomal recognition marker, mannose 6-​phosphate, they do not necessarily have a full complement of lysosomal membrane proteins such as LAMP1 and LAMP2 and other characteristics. Chédiak–​Higashi, Griscelli’s, and Hermansky–​Pudlak syndromes are rare conditions inherited as autosomal recessive traits. All cause oculocutaneous albinism, often in association with abnormal platelet granules and melanosomes in the skin and eyes: partial al- binism is frequent. Chédiak–​Higashi and Griscelli’s syndromes Chédiak–​Higashi syndrome is caused by mutations in the lyso- somal trafficking regulator gene located on chromosome 1q44. It predisposes to microbial infection and there are giant lysosomal granules in peripheral blood granulocytes; ceroid storage occurs in the nervous system and lungs. The clinical phenotype results from a complex set of immune defects affecting natural killer cells and neutrophilic leucocytes. Natural killer cell cytotoxicity is absent. Neutrophils, melanocytes, neurons, muscle cells, and Schwann cells show giant inclusion bodies. Recurrent cutaneous and systemic pyogenic infections occur with defective neutrophil and monocyte migration. Neurodegeneration is a prominent feature in young adults, but death often results from a rapidly progressive lymphoproliferative disorder.

section 12  Metabolic disorders 2154 The Griscelli’s syndrome(s) are three unusual variants: one (type III) is a simple form of albinism, and the others combine albinism with defective immunity (type II) or neurological deficits (type I). Griscelli’s syndrome type II with immunological defects is caused by mutations in Rab27a, a soluble GTPase, which regulates the flow of melanosomes in melanocytes and regulates exocytosis of lytic gran- ules at the point of the ‘immune synapse’ in cytotoxic T lympho- cytes. Deficient Rab27a thus causes dysfunctional T lymphocytes and pigmentary abnormalities. The Griscelli’s syndrome type I, which also has neurological symptoms, is caused by mutations in the motor protein, myosin Va, which may cooperate with Rab27a to transport melanosomes along actin filaments but apparently does not participate in the exocytosis of lytic T cell granules. Hermansky–​Pudlak syndrome Nine genetically distinct Hermansky–​Pudlak disorders are known. Hermansky–​Pudlak syndrome type 2 is caused by mutations in the β-​3A adaptin gene which is associated with altered trafficking of lysosomal proteins in melanosomes, lysosomes, and platelet-​ dense granules leading to storage pool deficiency. The gene maps to chromosome 10q. Inheritance is autosomal recessive. Although very rare, one of the Hermansky–​Pudlak syndromes occurs at a high frequency in the Swiss Alps and the Puerto-​Rican population where it is the most common single-​gene defect. Clinical features include a bleeding tendency due to abnormal platelets; diminished pigmentation of the skin, iris, and hair; and diverse inflammatory complications including granulomatous col- itis, cardiomyopathy, and severe pulmonary fibrosis. The pulmonary disease appears to be related to defective release of surfactant by type 2 pneumocytes. Hermansky–​Pudlak syndrome type 2 causes a mild bleeding di- athesis and platelet dense bodies are absent; the patients are suscep- tible to bacterial infection due to congenital neutropenia. There are clear similarities between Hermansky–​Pudlak and Chédiak–​Higashi syndromes, and further functional studies of their respective cognate proteins should refine our knowledge about the regulation of synthesis and coordinated assembly of lysosomes and related organelles. Treatment of biogenesis defects These disorders are often very severe. For example, patients with Chédiak–​Higashi disease and Griscelli’s syndrome type 2 occasion- ally develop a life-​threatening syndrome with fever, jaundice, and pancytopenia—​haemophagocytic lymphohistiocytosis—​which is related to impaired natural killer and cytotoxic T-​cell function and failure to resolve of lymphocyte and macrophage activation. These cells proliferate, releasing inflammatory cytokines that in- duce fever. Jaundice, hepatosplenomegaly, and pancytopenia are present in a hyperacute illness which is usually triggered by infec- tion with Epstein–​Barr virus or other viruses. High-​dose cortico- steroids and immunomodulatory agents are needed to suppress the inflammation. Rituximab and ciclosporin have been successful where the response to corticosteroids, ciclosporin, and etoposide was inadequate. When this accelerated lymphohistiocytic phase of the illness is resolved, transplantation with haematopoietic stem cells which replace the defective components of immune system with normal effector cells reduces the risk that this potentially fatal syndrome will recur. Oculocutaneous albinism may require aids for poor vision due to retinal photoinjury, especially at school, and protection against high-​intensity ultraviolet and visible light-​induced damage, with skin carcinoma, should be offered. Haemorrhagic manifestations may require platelet transfusions and parenteral desmopressin (DDAVP—​1-​desamino-​8-​d-​arginine vasopressin) to improve platelet function in the short term. Aspirin and other nonsteroidal drugs should be avoided if possible. Serious microbial infections with bacteria are common in affected children and fungal infections due to Candida or Aspergillus also occur. Immunization for common viral and bacterial infections, including influenza, Haemophilus influenzae, and pneumococci, should be given, and appropriate antimicrobial drugs used promptly where infection is likely. Established pulmonary fibrosis proceeds rapidly and may require treatment with domiciliary supplemental oxygen in the home, and lung transplantation may be successful in selected cases. A smoke-​ free environment is likely to be advantageous. FURTHER READING The study of lysosomal diseases is burgeoning: critical cellular func- tions carried out in the lysosomal compartment place the study of this organelle at the heart of contemporary molecular cell biology. The sheer pace of discovery and involvement of the lysosome in many pathological conditions and processes, with or without an overt gen- etic basis, means that comprehensive sources of up-​to-​date infor- mation are hard to find. Here we principally identify references of immediate application in the clinical field. Books Alberts B, et  al. (2007). Molecular biology of the cell, 5th edition. Garland Science (Taylor & Francis Group), New York. Barranger JA, Cabrera-​Salazar MA (eds) (2007). Lysosomal storage dis- orders. Springer Science, New York. Mehta A, Winchester B (eds) (2012). Lysosomal storage diseases: a prac- tical guide. Wiley-​Blackwell, London. Mole S, Williams R, Goebel H (eds) (2011). The neuronal ceroid lipo- fuscinoses (Batten disease), 2nd edition. Oxford University Press, Oxford. Nyhan WL, Barshop BA, Ozand PT (eds) (2005). Atlas of metabolic diseases, 2nd edition. Hodder Education, London. Saftig P (ed) (2005). Lysosomes. Springer Science, New York. Reviews of diagnosis and treatment of lysosomal diseases Anderson G, et  al. (2005). Blood film examination for vacuolated lymphocytes in the diagnosis of metabolic disorders; retrospective experience of more than 2,500 cases from a single centre. J Clin Pathol, 58, 1305–​10. Baldo BA (2015). Enzymes approved for human therapy: indications, mechanisms and adverse effects. BioDrugs, 29, 31–​55. Cox TM (2016). Lysosomal diseases. In: Bond JD (ed) Encyclopedia of Cell Biology, Vol. I, pp. 763–​88. Elsevier, Waltham. De Duve C (1964). From cytases to lysosomes. Fed Proc, 23, 1045–​9. Hocquemiller M, et  al. (2016). Adeno-​associated virus-​based gene therapy for CNS diseases. Hum Gene Ther, 27, 478–​96. Kornfeld S, Mellman I (1989). The biogenesis of lysosomes. Annu Rev Cell Biol, 5, 483–​525. Mindell JA (2012). Lysosomal acidification mechanisms. Annu Rev Physiol, 74, 69–​86.

12.8  Lysosomal disease 2155 Neufeld EF (2011). From serendipity to therapy. Annu Rev Biochem, 80, 1–​15. Pineda M, Walterfang M, Patterson MC (2018). Miglustat in Niemann–​ Pick disease type C patients: a review. Orphanet J Rare Dis, 13, 140. Piper RC, Luzio JP (2001). Late endosomes: sorting and partitioning in multivesicular bodies. Traffic, 2, 612–​21. Platt FM, d’Azzo A, Davidson BL, Neufeld EF, Tifft CJ (2018). Lysosomal storage diseases, Nat Rev Dis Primers, 4, 27. Popovic D, et  al. (2012). Rab GTPase-​activating proteins in autophagy:  regulation of endocytic and autophagy pathways by direct binding to human ATG8 modifiers. Mol Cell Biol, 32, 1733–​44. Ravikumar B, et al. (2010). Regulation of mammalian autophagy in physiology and pathophysiology. Physiol Rev, 90, 1383–​435. Saftig P, Klumperman J (2009). Lysosome biogenesis and lysosomal membrane proteins: trafficking meets function. Nat Rev Mol Cell Biol, 10, 623–​35. Schulz A, et al. (2013). NCL diseases—​clinical perspectives. Biochim Biophys Acta, 1832, 1801–​6. Sengupta S, Peterson TR, Sabatini DM (2010). Regulation of the mTOR complex 1 pathway by nutrients, growth factors, and stress. Mol Cell, 40, 310–​22. Von Figura K (1991). Molecular recognition and targeting of lyso- somal proteins Curr Opin Cell Biol, 3, 642–​6. Walkely SU, Xu H, Ren D (2015). Lysosomal physiology. Annu Rev Physiol, 77, 57–​80. Yang Z, Klionsky DI (2010). Eaten alive: a history of macroautophagy. Nat Cell Biol, 12, 814–​22. Journal articles Aerts JM, et al. (2008). Biomarkers for lysosomal storage disorders: identification and application as exemplified by chitotriosidase in Gaucher disease. Acta Paediatr, 97 Suppl 457, 7–​14. Baldo G, Giugliani R, Matte U (2014). Gene delivery strategies for the treat- ment of mucopolysaccharidoses. Expert Opin Drug Deliv, 11, 449–​59. Barton NW, et al. (1991). Replacement therapy for inherited enzyme deficiency macrophage-​targeted glucocerebrosidase for Gaucher’s disease. N Engl J Med, 324, 1464–​70. Ben Turkia H, et al. (2013). Velaglucerase alfa enzyme replacement therapy compared with imiglucerase in patients with Gaucher dis- ease. Am J Hematol, 88, 179–​84. Biffi A, et al. (2013). Lentiviral hematopoietic stem cell gene therapy benefits metachromatic leukodystrophy. Science, 341, 1233158. Briggs TA, et al. (2011). Tartrate-​resistant acid phosphatase deficiency causes a bone dysplasia with autoimmunity and a type I interferon expression signature. Nat Genet, 43, 127–​31. Burton BK, et al. (2015). A phase 3 trial of sebelipase alfa in lysosomal acid lipase deficiency. N Engl J Med, 373, 1010–​20. Cartier N, Aubourg P (2008). Hematopoietic stem cell gene therapy in Hurler syndrome, globoid cell leukodystrophy, metachromatic leukodystrophy and X-​adrenoleukodystrophy. Curr Opin Mol Ther, 10, 471–​8. Cherqui S, Courtoy PJ (2017). The renal Fanconi syndrome in cystinosis:  pathogenic insights and therapeutic perspectives. Nat Rev Nephrol, 13, 115–​31. Chtarto A, et al. (2013). A next step in adeno-​associated virus-​mediated gene therapy for neurological diseases: regulation and targeting. Br J Clin Pharmacol, 76, 217–​32. Cheng SH (2014). Gene therapy for the neurological manifestations in lysosomal storage disorders. J. Lipid Res, 55, 1827–​38. Chien YH, et al. (2013). Long-​term efficacy of miglustat in paediatric patients with Niemann–​Pick disease type C. J Inherit Metab Dis, 36, 129–​37. Cox TM, Schofield JP (1997). Gaucher’s disease: clinical features and natural history. Ballières Clin Haematol, 10, 657–​89. Cox TM, et al. (2008). Management of non-​neuronopathic Gaucher disease with special reference to pregnancy, splenectomy, bisphosphonate therapy, use of biomarkers and bone disease moni- toring. J Inherit Metab Dis, 31, 319–​36. Cox TM, et  al. (2012). Evaluation of miglustat as maintenance therapy after enzyme therapy in adults with stable type 1 Gaucher disease: a prospective, open-​label non-​inferiority study. Orphanet J Rare Dis, 7, 102. Cox TM et  al. (2017). Eliglustat maintains long-​term clinical sta- bility in patients with Gaucher disease type 1 stabilized on enzyme therapy. Blood, 129, 2375–​83. Deegan PB, Cox TM (2012). Imiglucerase in the treatment of Gaucher disease: a history and perspective. Drug Des Devel Ther, 6, 81–​106. Eng CM, et  al. (2001). Safety and efficacy of recombinant human alpha-​galactosidase A-​replacement therapy in Fabry’s disease. N Engl J Med, 345, 9–​16. Eng CM, et al. (2007). Fabry disease: baseline medical characteristics of a cohort of 1765 males and females in the Fabry Registry. J Inherit Metab Dis, 30, 184–​92. Escolar ML, et al. (2005). Transplantation of umbilical-​cord blood in babies with infantile Krabbe’s disease. N Engl J Med, 352,
2069–​81. Fratantoni JC, et  al. (1969). The defect in Hurler and Hunter syn- dromes. II. Deficiency of specific factors involved in mucopolysac- charide degradation. Proc Natl Acad Sci USA, 6, 360–​6. Huizing M, et al. (2008). Disorders of lysosome-​related organelle bio- genesis: clinical and molecular genetics. Annu Rev Genomics Hum Genet, 9, 359–​86. Kamath RS, et  al. (2014). Skeletal improvement in patients with Gaucher disease type 1: a phase 2 trial of oral eliglustat. Skeletal Radiol, 43, 1353–​60. Kaplan P (2014). Clinical potential of eliglustat tartrate in the
treatment of type 1 Gaucher disease. Res Rep Endocrine Disord, 4, 1–​8. Kishnani PS, et al. (2007). Recombinant human acid-​glucosidase: major clinical benefits in infantile-​onset Pompe’s disease. Neurology, 68, 99–​109. Kohlschutter A, Schulz A, Denecke J (2014). Epilepsy in neuronal ceroid lipofuscinoses. J Pediatr Epilepsy, 3, 199–​206. Lukina E, et al. (2014). Eliglustat, an investigational oral therapy for Gaucher disease type 1: phase 2 trial results after 4 years of treat- ment. Blood Cells Mol Dis, 53, 274–​6. Maegawa GH, et al. (2006). The natural history of juvenile or subacute GM2 gangliosidosis: 21 new cases and literature review of 134 previ- ously reported. Pediatrics, 118, e1550–​62. Maegawa GH, et al. (2007). Pyrimethamine as a potential pharmaco- logical chaperone for late-​onset forms of GM2 gangliosidosis. J Biol Chem, 282, 9150–​61. McEachern KA, et  al. (2007). A specific and potent inhibitor of glucosylceramide synthase for substrate inhibition therapy of Gaucher disease. Mol Genet Metab, 91, 259–​67. Meikle PJ, et  al. (1999). Prevalence of lysosomal storage disorders. JAMA, 281, 249–​54. Mindell JA (2012). Lysosomal acidification mechanisms. Annu Rev Physiol, 74, 69–​86. Mistry PK, et al. (2015). Effect of oral eliglustat on splenomegaly in pa- tients with Gaucher disease type 1: the ENGAGE randomized clin- ical trial. JAMA, 313, 695–​706. Mole SE (2014). Development of new treatments for Batten disease. Lancet Neurol, 13, 749–​51.

section 12  Metabolic disorders 2156 Musolino PL, et al. (2014). Hematopoietic stem cell transplantation in the leukodystrophies:  a systematic review of the literature. Neuropediatrics, 45, 169–​74. Nair S, et al. (2016). Clonal immunoglobulin against lysolipids in the origin of myeloma. N Engl J Med, 374, 555–​61. Nair S, et al. (2018). Antigen-mediated regulation in monoclonal gammopathies and myeloma. JCI Insight, 3(8), pii: 98259. Neudorfer GM, et al. (2005). Late-​onset Tay–​Sachs disease: pheno- typic characterization and genotypic correlations in 21 affected pa- tients. Genet Med, 7, 119–​23. Pavlova EV, et al. (2019). The lysosomal disease caused by mutant VPS33A. Hum Mol Genet, 28, 2514–30. Pellegrini N, et al. (2005). Respiratory and limb muscle weakness in adults with Pompe disease. Eur Respir J, 26, 1024–​31. Porto AF (2014). Lysosomal acid lipase deficiency:  diagnosis and treatment of Wolman and cholesteryl ester storage diseases. Pediatr Endocrinol Rev, 12 Suppl 1, 125–​32. Robertson PL, Maas M, Goldblatt J (2007). Semiquantitative assessment of skeletal response to enzyme replacement therapy for Gaucher’s disease using the bone marrow burden score. Am J Roentgenol, 188, 1521–​8. Rosenbloom BE, et al. (2005). Gaucher disease and cancer incidence: a study from the Gaucher Registry. Blood, 105, 4569–​72. Saftig P, et al. (2001). Disease model: LAMP-​2 enlightens Danon dis- ease. Trends Mol Med, 7, 37–​9. Sardiello M, et al. (2009). A gene network regulating lysosomal biogen- esis and function. Science, 325, 473–​7. Schiffmann R, et al. (2001). Enzyme replacement therapy in Fabry dis- ease: a randomized controlled trial. JAMA, 285, 2743–​9. Schiffmann R, et al. (2008). Randomized, controlled trial of miglustat in Gaucher’s disease type 3. Ann Neurol, 64, 514–​22. Schulz A, et al. (2018). Study of intraventricular cerliponase alfa for CLN2 disease. N Engl J Med, 378, 1898–​907. Settembre C, et al. (2011). TFEB links autophagy to lysosomal biogen- esis. Science, 332, 1429–​33. Shayman, JA (2013). The design and clinical development of inhibitors of glycosphingolipid synthesis: will invention be the mother of ne- cessity? Trans Am Clin Climatol Assoc, 124, 46–​60. Sidransky E (2012). Gaucher disease: insights from a rare Mendelian disorder. Discov Med, 14, 273–​81. Sidransky E, et al. (2009). Multicenter analysis of glucocerebrosidase mutations in Parkinson’s disease. N Engl J Med, 361, 1651–​61. Tardieu M, et  al. (2014). Intracerebral administration of adeno-​ associated viral vector serotype rh.10 carrying human SGSH and SUMF1 cDNAs in children with mucopolysaccharidosis
type IIIA disease: results of a phase I/​II trial. Hum Gene Ther, 25, 506–​16. Tardieu M, et al. (2017). Intracerebral gene therapy in children with mucopolysaccharidosis type IIIB syndrome: an uncontrolled phase 1/​2 clinical trial. Lancet Neurol, 16, 712–​20. Van Capelle CI, et al. (2008). Eight years experience with enzyme re- placement therapy in two children and one adult with Pompe dis- ease. Neuromuscul Disord, 18, 447–​52. Van den Hout HM, et  al. (2003). The natural course of infantile Pompe’s disease: 20 original cases compared with 133 cases from the Literature. Pediatrics, 112, 332–​40. Weinreb NJ, et  al. (2013). Long-​term clinical outcomes in type 1 Gaucher disease following 10  years of imiglucerase treatment. J Inherit Metab Dis, 36, 543–​53. Winchester B (2012). Classification of lysosomal storage diseases. In: Mehta A, Winchester B (eds) Lysosomal storage disorders: a practical guide, pp. 37–​48. Wiley-​Blackwell, Hoboken, New Jersey. Websites Online Mendelian Inheritance in Man: http://​www.ncbi.nlm.nih.gov. Scriver CR, et al. (2004). Metabolic and Molecular Bases of Inherited Disease, 8th edition. Part 16: Lysosomal disorders, Chapters 134–​54. McGraw-​Hill, New York. http://​www.ommbid.com.

12.9 Disorders of peroxisomal metabolism in adults

12.9 Disorders of peroxisomal metabolism in adults 2157

ESSENTIALS The peroxisome is a specialized organelle which employs molecular oxygen in the oxidation of complex organic molecules including lipids. Enzymatic pathways for the metabolism of fatty acids, including very long-​chain fatty acids (VLCFA), enable this organelle to carry out
β-​oxidation in partnership with mitochondria. A  peroxisomal pathway for isoprenoid lipids derived from chlorophyll, such as phytanic acid, utilizes α-​oxidation, but a default mechanism involving ω-​oxidation may also metabolize phytanic acid and its derivatives. The biochemical manifestations, molecular pathology, and di- verse clinical features of many peroxisomal disorders have now been clarified, offering the promise of prompt diagnosis, better manage- ment, and useful means to provide appropriate genetic counselling for affected families. At the same time, specific treatments including rigorous dietary interventions and plasmapheresis to remove undegraded toxic metabolites offer credible hope of improvement and prevention of disease in affected individuals. Inborn errors of peroxisomal metabolism usually present in in- fancy and childhood, but some disorders typically become manifest later in life and in adults, in whom the progress is often slow. Particular adult peroxisomal disorders X-​linked adrenoleukodystrophy (X-​ALD)—​due to mutation in the gene for an ATP-​binding cassette (ABC) protein of unknown function and characterized by accumulation of unbranched saturated VLCFAs, particularly hexacosanoate (C26), in the cholesterol esters of brain white matter, adrenal cortex, and certain sphingolipids of the brain. The disease has multiple phenotypes: it may present in adolescence with slowly progressive stiffness, clumsiness, weakness, weight loss, and skin pigmentation typical of Addison’s disease; it may present in adults with primarily psychiatric manifestations. Most cases de- velop increasing handicap; management is palliative and supportive in most instances. Adult Refsum’s disease—​due in most cases to mutations in the gene for phytanoyl-​CoA hydroxylase (PHYH) such that patients are un- able to detoxify phytanic acid by α-​oxidation and have greatly ele- vated levels of this in their plasma. Usually presents in late childhood with progressive deterioration of night vision, the occurrence of progressive retinitis pigmentosa, and anosmia; late features include deafness, ataxia, polyneuropathy, ichthyosis, and cardiac arrhyth- mias. Treatment is by restriction of dietary phytanic acid, with or without its elimination by plasmapheresis or apheresis. Neuropsychiatric adult peroxisomal disorders Historical perspective The likely first description of X-​linked adrenoleukodystrophy (X-​ ALD; OMIM 300100) was in 1910 when a 6-​year-​old child devel- oped abnormal eye movements, apathy, and mental deterioration. His gait then deteriorated and skin darkening was noted prior to his death a few months later. Examination of the brain by Schilder showed central demyelination, perivascular lymphocytes, foam cells, and gliosis which he termed encephalitis periaxalis diffusa. Other cases he later described are likely due to other leukodystrophies. Adrenoleukodystrophy was defined in 1970, with its characteristic adrenal changes of excess very long-​chain fatty acids (VLCFAs) and cholesterol esters present in cell inclusion bodies. These VLCFAs were later recognized as pathognomonic and identifiable in plasma samples and the primary defect was identified as an inability to me- tabolize them. The gene was mapped to Xq18 and identified as a member of the ATP-​binding cassette (ABC) transporter family. X-​ ALD was localized to the peroxisome. Subsequently, mouse models have been developed which show some clinical features of human disease such as adrenomyeloneuropathy but typically lack the cere- bral changes seen in man. Aetiology Adrenoleukodystrophy is characterized by the accumulation of un- branched saturated VLCFAs with a chain length of 24 to 30 carbons, particularly hexacosanoate (C26), in the cholesterol esters of brain white matter, in the adrenal cortex, and in certain sphingolipids of the brain. The disorder shows X-​linked inheritance with expression in female heterozygotes. The disruptive effects of the accumulation of VLCFAs, especially hexacosanoic acid (C26:0), on cell membrane structure and function may explain the neurological manifestations seen in adrenoleukodystrophy patients. VLCFAs cause alterations in membrane fluidity and affect cortisol secretion from cultured cells 12.9 Disorders of peroxisomal metabolism in adults Anthony S. Wierzbicki

section 12  Metabolic disorders 2158 of adrenal cortical origin. In addition, albumin has only one C26 binding site compared with more than six for shorter fatty acids, so limiting its efficacy as a reverse transport protein for excess VLCFAs. Clinical features X-​ALD is heterogeneous: seven phenotypes occur in males and five are recognized in females. Childhood cerebral adrenoleukodystrophy presents between the ages of 5 and 10 years with emotional lability, hyperactivity/​withdrawal, and mental deterioration, mimicking at- tention deficit disorder which evolves to parietal lobe dysfunction with apraxia, astereognosis, and later dementia. MRI shows a char- acteristic pattern of symmetric involvement of the posterior parieto-​ occipital white matter in 85% of patients, frontal involvement in 10%, and an asymmetric pattern in the rest. The clinical phenotype of X-​ALD shows a variable progression which may be interrupted by periods of arrest on MRI sometimes lasting 5 to 10 years. The adolescent form is adrenomyeloneuropathy which presents with slowly progressive stiffness, clumsiness, weakness, weight loss, and skin pigmentation typical of Addison’s disease. Autonomic function including micturition and erectile function are affected later. Somatosensory, auditory, and brainstem evoked potential are abnormal with some cases of abnormal visual and peripheral nerve conduction abnormalities. Brain MRI scans are abnormal in 50% of men and 80% of women, usually affecting corticospinal tracts with later parenchymal changes. Depression and emotional lability are common. Adult cerebral adrenoleukodystrophy is a variant of adrenomyeloneuropathy occurring after age 20 without spinal cord symptoms. The primary signs are psychiatric with a presentation of psychotic mania and may include schizophrenia or dementia. Some cases show a pure initial Addisonian picture with no neurological involvement; all are autoantibody negative. The onset of Addison’s disease is usually in childhood but the neurological changes follow in 20 to 30 years. Subtle hyper-​reflexia or impaired vibration sense and subtle MRI or neurophysiological signs may be detected earlier in these cases. It had been considered that neurological changes were mild or absent in carriers of X-​ALD, but up to 20% are symptom- atic. Women who are X-​ALD heterozygotes usually present with adrenomyeloneuropathy at age 30 to 40. Subtle signs are often de- tected prior to presentation but eventually the full picture occurs, with late-​onset dementia. A  recent prospective study of 46 fe- male carriers found an age-​dependent phenotype of myelopathy occurring in 63% and faecal incontinence in 28% independent of X-​inactivation status. These were associated with abnormalities in plasma VLCFAs and decreased fibroblast β-​oxidation of VLCFAs. In female adrenoleukodystrophy heterozygotes, adrenal cortical insufficiency rarely develops, although isolated min- eralocorticoid insufficiency may occur but may be difficult to diagnose. Furthermore, adrenoleukodystrophy heterozygotes are predisposed to hypoaldosteronism related to the use of nonsteroidal anti-​inflammatory drugs (NSAIDs). A subclinical decrease in gluco- corticoid reserve, as measured by synthetic ovine corticotropin-​ releasing hormone testing, may be present in most of these women. Aldosterone levels should be included in ACTH stimulation testing done to detect adrenal insufficiency in affected women. NSAIDs should be considered a risk factor for the develop- ment of hypoaldosteronism in women who are heterozygous for adrenoleukodystrophy. Rare presentations include olivopontocerebellar atrophy which has been described as X-​ALD ataxia in Japanese. Other uncommon presentations include unilateral masses which can mimic brain tumours and cases of spontaneous remission of neurological symptoms. A clinical syndrome that mimics features of ALD is acyl-​coenzyme A (CoA) oxidase deficiency but most cases present with severe neo- natal disease. Neuropathology There are two distinct forms of neuropathology associated with X-​ ALD. Pure adrenomyeloneuropathy is a distal axonic neuropathy while the cerebral forms are associated with inflammation. In cere- bral X-​ALD, brain pathology is often grossly normal though with signs of cerebral atherosclerosis. Grey matter is unaffected but white matter disease occurs in a rostrocaudal direction with demyelination prominent in the parieto-​occipital cortex and the cerebellum. The detailed pathology shows oligodendroglial cell loss, astrocytosis, and a perivascular inflammatory infiltrate. In the noncerebral form, de- myelination is seen in the corticospinal tracts with no obvious in- flammation and only mild gliosis and occasional macrophages. In the adrenal cortex, cells are filled with lamellar deposits of cholesterol es- ters with primary cortical atrophy and no evidence of inflammation or antibodies, with milder changes in the adrenomyeloneuropathy form. In men with X-​ALD, the testes show Leydig cell alterations, again with lamellar deposits. It has been estimated that at least 10% of males with Addison’s disease (adrenocortical failure) have X-​linked adrenomyeloneuropathy or unrecognized X-​ALD. Metabolism of VLCFAs VLCFAs are derived from the diet and endogenous synthesis, with between 20 and 80% derived from synthesis depending on the study. The synthetic pathway occurs in brain microsomes with re- peated additions of malonyl-​CoA units to palmitic (C16:0) or stearic (C18:0) acid precursors. There are probably separate pathways for C20:0 and C22:0 (behenic) fatty acids with the C22:0 pathway also elongating C22:1 (erucic) acid. Synthesis of VLCFAs starts with use of a specific activator pro- tein SLC27A4—​the fatty acid transport protein 4 (FATP4). The synthesis of saturated VLCFA, monounsaturated VLCFA (MUFA), and polyunsaturated fatty acids (PUFAs) occurs in the endoplasmic reticulum by four distinct enzymes; elongation of very long-​chain fatty acids ligase (ELOVL), 3-​ketoacyl-​CoA reductase (HSD17B12), 3-​hydroxyacyl dehydratase (HACD3), and trans-​2,3,-​enoyl-​CoA reductase (TECR). The initial condensation reaction catalysed by ELOVL is usually rate limiting. Mammals have seven different ELOVL enzymes (ELOVL1–​7) but only a single enzyme has been identified so far for each synthetic pathway. ELOVL1, ELOVL3, ELOVL4, and ELOVL6 are involved in the synthesis of saturated and monounsaturated fatty acids and ELOVL2, ELOVL4, ELOVL5, and ELOVL7 are essential for PUFA metabolism. The synthesis of C24:0 and C26:0 VLCFAs is carried out by the concerted action of ELOVL6 (C18:0–​C22:0) and ELOVL1 (C24:0–​C26:0) of which the latter is the key rate-​limiting step. Degradation of VLCFAs occurs by β-​oxidation within peroxi- somes after activation by specific acyl-​CoA ligases which are chain-​ length specific. Again, FATP4 plays a key role in activating the fatty acid.

12.9  Disorders of peroxisomal metabolism in adults 2159 Molecular genetics: the X-​ALD protein and its homologues The X-​ALD gene was mapped to a region of the X-​chromosome close to the glucose-​6-​phosphate dehydrogenase gene. The gene was established to code for an ABC protein of still unknown function but likely to involve the translocation of a variety of substrates across extra-​ and intracellular membranes, including lipids, sterols, and drugs. The ABCD1 protein (adrenoleukodystrophy protein) maps to Xq28 and is mutated in X-​ALD. ABCD1 is a member of the ABC transporter superfamily. It ex- presses a half transporter which is located in the peroxisome. The gene has an open reading frame of 2235 bases which encodes a 745-​ amino acid protein with 38.5% amino acid identity and 78.9% simi- larity to another peroxisomal protein (ABCD3). Mutations in ABCD1 result in X-​ALD in animal models, with elevated VLCFAs. ABCD1 is one of four related peroxisomal trans- porters that are found in the human genome, the others being ABCD2 (adrenoleukodystrophy related protein) (OMIM 601081), ABCD3 (peroxisomal membrane protein 70) (OMIM 170995), and ABCD4 (P70R/​PMP69) (OMIM 603214). The adrenoleukodystrophy pro- tein and the adrenoleukodystrophy-​related protein are expressed on oligodendroglia, while the adrenoleukodystrophy-​related protein and peroxisomal membrane protein 70 are found in neurons of the central nervous system. These genes are highly conserved in evolu- tion, and two homologous genes are present in the yeast genome, PXA1 and PXA2, which also transport long-​chain fatty acids. The 80-​kDa protein encoded by this gene is absent in patients with X-​ ALD, in whom X-​ALD mRNA was undetectable. Most of the ABCD1 mutations (>450) in X-​ALD are point mutations, but large deletions have been described. There is no correlation between genotype and phenotype. In 15 to 20% of obligate female heterozygotes, false-​ negative results occur for plasma VLCFAs. Mutation analysis is the only reliable method for the identification of heterozygotes. Overexpression of the adrenoleukodystrophy protein and its homologue, the adrenoleukodystrophy-​related protein (ABCD2), can restore the impaired peroxisomal β-​oxidation in the fibro- blasts of adrenoleukodystrophy patients. However, it seems that functional replacement of the adrenoleukodystrophy protein by adrenoleukodystrophy-​related protein is not due to stabiliza- tion of the mutated adrenoleukodystrophy protein. Similarly, the adrenoleukodystrophy-​related protein and peroxisomal membrane protein 70 could restore the peroxisomal β-​oxidation defect in the liver of adrenoleukodystrophy protein-​deficient mice by stimulating Aldr and Pmp70 gene expression through a dietary treatment with the peroxisome proliferator fenofibrate. These results suggested that a correction of the biochemical defect in adrenoleukodystrophy might be possible by drug-​induced overexpression or ectopic expression of the adrenoleukodystrophy-​related gene. The adrenoleukodystrophy protein transporter may facilitate the interaction between per- oxisomes and mitochondria, the two sites within the cells where β-​oxidation of VLCFAs occurs. The phenotype of X-​ALD was thought to be based on microglial activation for cerebral effects, while inflammation is less involved in adrenomyeloneuropathy but transcriptome studies show that a combination of effects of the defi- ciency on oxidative phosphorylation and adipocytokine and insulin signalling are responsible for the phenotypes. Many papers have described mutations in the ABCD1 gene in X-​ ALD patients, indeed more than 600 different mutations have now been described (http://​www.x-​ald.nl) of which 51% are missense, 28% frame-​shift, and 12% nonsense mutations; 6% are small in- sertions/​deletions and 13% are exon deletions. About 75% of all nonrecurrent ABCD1 mutations result in the absence of ABCD1 protein (http://​www.x-​ald.nl). Nonsense and frame-​shift mutations as well as large deletions lead to a truncated protein. Many missense mutations result in unstable protein whose detection is likely to be dependent on the specificity and sensitivity of the method used. The lack of anti-​ABCD1 immunofluorescence (IF) using microscopy in cultured fibroblasts is commonly used to assess ABCD1 pro- tein expression. However, fibroblasts express relatively low levels of ABCD1 and so this method may miss ABCD1 expression detectable by the more sensitive western blot methods. Re-​investigation of ‘IF negative’ cell lines using improved techniques has confirmed this supposition, hence some previously suspected cases may require re-​analysis. Disease-​associated missense mutations are not equally distributed over the ABCD1 protein. Analysis of 300 missense mu- tations showed clustering in two major regions (Fig. 12.9.1). Epidemiology Screening and diagnostic records suggest that the prevalence is a minimum of 1 in 22 500 to 1 in 62 000. In contrast, the use of the Hardy–​Weinberg approach and genetic frequency data suggests a combined male to female frequency of 1 in 18 000 similar to phenyl- ketonuria (1 in 12 000). Differential diagnosis The differential diagnosis of neuropsychiatric abnormalities is shown in Table 12.9.1. X-​ALD can mimic attention deficit disorder, multiple sclerosis, organic dementias, and psychoses among neuro- logical diseases, and Addison’s disease and hypogonadism among endocrine disorders (Table 12.9.2). The critical clinical differential element is the finding of abnormal ACTH concentrations and skin pigmentation with neurological signs, however subtle. Clinical investigation Clinical biochemistry The primary abnormality in X-​ALD is an accumulation of VLCFAs (>C22) which occur in myelin. C26:0 can account for up to 5% of brain cerebrosides and sulphatides. In X-​ALD, both saturated and unsaturated forms of C26:0 (cerotic) and C24:0 (lignoceric) acids ac- cumulate with reductions in C24:1(n-​9) (nervonic) acid. Normally, shorter fatty acids accumulate in brain cholesterol esters, but in X-​ALD, by contrast, these are mostly C26:0 and are enriched in myelin and in areas of demyelination. Similarly, C26:0 accumu- lates in white matter phosphatidylcholine phospholipids, C24:0 and C24:1 in gangliosides. Erythrocytes, plasma, and cultured fibroblasts all contain a 2-​ to 10-​fold excess of VLCFAs. The diagnostic test re- lies on measurement of C26:0 levels and the ratios of C26 to C22:0 (docosahexaenoic acid) and C26:0 to C24:0 (tetracosanoic acid). Some neonatal paediatric screening programmes have begun to implement screening for C26:0 phosphatidylcholine as a marker of X-​ALD in their dried blood spot analysis programmes. Results can be confirmed by fibroblast studies or by the use of sequencing techniques. Highly elevated VLCFA levels are also found in peroxisomal biogenesis disorders but these show a different clinical presentation to X-​ALD or transiently with ketogenic diets for seizures. False-​negative results may occur in patients consuming excess C22:1;

section 12  Metabolic disorders 2160 Table 12.9.1  Psychiatric signs and inborn errors of metabolism in adolescents and adults Disorder Confusion Mental retardation Behavioural disturbance Catatonia Visual hallucination Psychosis Depression Urea cycle defect + +/​–​ + + + + + Homocysteine disorders + + + + + +/​–​ + Porphyria + + + +/​–​ +/​–​ Wilson’s disease +/​–​ + +/​–​ + CTX + + + + MLD + + GM2 gangliosidosis + + + + + Mannosidoses + + + + + X-​ALD + + + Acyl-​CoA oxidase (pseudoneonatal adrenoleukodystrophy) + + + Nonketotic hyperglycinaemia + + Monoamine oxidase A deficiency + + Creatine transporter deficiency + + Succinic semi-​aldehyde dehydrogenase deficiency + + Niemann–​Pick C + + + + + CTX, cerebrotendinous xanthomatosis; MLD, metachromatic leukodystrophy; X-​ALD, X-​linked adrenoleukodystrophy. Reproduced from Sedel F et al. (2007a). Psychiatric manifestations revealing inborn errors of metabolism in adolescents and adults. J Inherit Metab Dis, 30, 631–​41, with permission. 10 8 6 4 2 0 50 100 N mPTS Transmembrane domain Number of disease-causing mutations per block of 10 amino acids Evolutionary conservation (% AA identity per block of 10 amino acids) Walker A Walker B C 0 20 40 60 80 100 150 200 250 300 350 400 450 500 550 600 650 700 750 Fig. 12.9.1  The degree of interspecies conservation and location of human mutations in the ABCD1 gene. The first region of conservation/​disease-​causing mutations is located in the transmembrane domain region (amino acids 83–​344) and the second is located in the ATP-​ binding domain (amino acids 500 and 668). The N-​terminal 73 amino acids and the C-​terminal 50 amino acids are mostly spared, hence caution is warranted when interpreting sequencing data suggesting missense mutations outside these key regions. Reproduced with permission from Wiersinger C, Eichler FS, Berger J. The genetic landscape of X-​linked adrenoleukodystrophy: inheritance, mutations, modifier genes, and diagnosis. Appl Clin Genet. 2015; 8: 109–​ 121. Copyright © 2015 Wiesinger et al.

12.9  Disorders of peroxisomal metabolism in adults 2161 ω-​9 (erucic acid; Lorenzo’s oil) which is found in mustard and rapeseed oils. A few affected males (0.1%) have borderline normal C26:0 levels and 15% of obligate female carriers have normal results. Effective mu- tation detection in these families is therefore fundamental to the un- ambiguous determination of genetic status. Of particular concern are female members of kinships with segregating X-​ALD mutations, be- cause normal levels of VLCFA do not guarantee a lack of carrier status. Prenatal diagnosis is possible from cultured amniocytes or chorionic villus cells. Abnormal liver function tests are a common finding in adrenoluekodystrophies and occur secondary to disturbances in di-​ and trihydroxycholestanoic acid (DHCA and THCA) metabolism. Radiology A MRI scan often reveals biochemical changes before the develop- ment of clinical symptoms. Eighty per cent of childhood cerebral adrenoleukodystrophy patients have symmetric periventricular white matter changes in the posterior parietal and occipital lobes with a dorsocaudal progression with time (Fig. 12.9.2a). Patients with adrenomyeloneuropathy typically have abnormalities in the pyramidal tracts (Fig 12.9.2b). Contrast studies show up areas of active demye- lination, inflammation with breakdown of the blood–​brain barrier, and gliosis. The Loes score (34-​point X-​ALD severity score) based on the five patterns of disease visible on MRI is used to determine severity and prognosis and is used as a decision aid prior to bone marrow transplantation. The presence of demyelination and gadolinium en- hancement are used to differentiate stable from likely progressive in- flammatory changes on MRI scanning (Fig. 12.9.2c). Proton magnetic resonance spectroscopy shows only mild reduction in N-​acetyl aspar- tate, normal choline and myo-​inositol, and normal lactate in patients Table 12.9.2  Differential diagnosis of X-​ALD Presentation Differential diagnosis Childhood neurological with normal endocrinology Hyperactivity, attention deficit disorder Epilepsy/​seizures Brain tumour Metachromatic/​globoid leukodystrophy Postencephalitic syndromes, e.g. subacute sclerosing panencephalitis Myelinoclastic diffuse sclerosis Childhood neurological with hypoadrenalism Addison’s disease with post-​hypoglycaemic damage X-​linked glycerol kinase deficiency Central pontine myelinolysis Glucocorticoid deficiency with achalasia Hypoadrenalism Secondary causes of hypoadrenalism Adrenomyeloneuropathy Multiple sclerosis Familial or other spastic parapareses Spinocerebellar/​olivopontocerebellar degeneration Cervical spondylosis Spinal cord tumour, e.g. ependymoma Adult cerebral Schizophrenia Depression Epilepsy/​organic psychosis Alzheimer’s disease or other dementias Brain tumour Heterozygote with symptoms Multiple sclerosis Chronic spinal disease Spinal cord tumour Cervical spondylosis (A) (a) (B) (C) (D) Fig. 12.9.2  (a) MRI of the brain in a case of childhood cerebral adrenoleukodystrophy (ALD) showing characteristic extensive white matter changes in the parieto-​occipital region and internal capsules on FLAIR sequences (A). This area is initially affected in about 80% of cases of cerebral ALD. The rim enhances after administration of gadolinium on T1 sequences (B). In about 20% of cases the site of initial involvement in cerebral ALD is the frontal white matter as shown on this FLAIR image of a different patient with cerebral ALD (C), with prominent rim enhancement after administration of gadolinium on a T1-​weighted image (D). (b) MRI of the brain in a patient with adrenomyeloneuropathy showing increased signal in the pyramidal tracts on T2-​weighed coronal (A) and axial (B) images indicative of Wallerian degeneration. (c) MRI of the brain (T2 (A) and FLAIR (C) images; T1 with gadolinium (B, D)) of a patient with adrenomyeloneuropathy who rapidly deteriorated clinically with new symptoms of cognitive decline. On MRI, extensive white matter changes were seen in the parieto-​occipital white matter and corpus callosum (A), but no enhancement of the lesion after administration of gadolinium (B). A follow-​up MRI about 3 months later shows progression of the white matter lesion (C) and there is now faint enhancement of the rim of the lesion after gadolinium administration (D). Reproduced with permission from Engelen M, Kemp S, de Visser M, van Geel BM, Wanders RJ, Aubourg P, Poll-​The BT. X-​linked adrenoleukodystrophy (X-​ALD): clinical presentation and guidelines for diagnosis, follow-​up and management. Orphanet J Rare Dis. 2012 Aug 13;7:51. doi: 10.1186/​1750-​1172-​7-​51.
Copyright © 2012 Engelen et al.; licensee BioMed Central Ltd.

section 12  Metabolic disorders 2162 with arrested as compared with progressing disease where N-​acetyl aspartate levels are significantly reduced while choline compounds, myo-​inositol, and lactate are raised. 18Fluorodeoxyglucose positron emission tomography shows increased glucose uptake in the frontal lobes with decreased activity in the temporal lobes and cerebellum in patients with X-​ALD. The increase in frontal activity correlated with scores from psychological evaluations. Proton spectroscopy using N-​acetyl aspartate shows up neuronal loss, while choline compound studies assaying phosphocholine and glycerophosphocholine indicate membrane turnover and demyelin- ation, and myo-​inositol compounds seem to be indices of gliosis. The presence of lactate indicates the anaerobic metabolism of the inflammatory cell infiltrate. In the adrenomyeloneuropathy brain, MRIs may be normal in 50% of men and 80% of women but diffuse spinal cord atrophy is present. Endocrinology Overt hypoadrenalism occurs in 40% of patients with child- hood cerebral adrenoleukodystrophy and 80% have a deficient cortisol response on Synacthen testing. In childhood disease, 80% show abnormal adrenal stimulation test results, while in adrenomyeloneuropathy, between 30 and 50% show normal re- sponses. Clinical Addison’s disease is found in 1% of female hetero- zygotes. In adrenoleukodystrophy heterozygotes, adrenal cortical insufficiency rarely develops, although hypoaldosteronism may occur, especially if NSAIDs are being used. ACTH levels are in- creased in male patients. Levels of follicle-​stimulating hormone or luteinizing hormone are increased in 50 to 70% of patients with adrenomyeloneuropathy, while testosterone levels are reduced in 20% with low normal levels of dehydroepiandrosterone sulphate. Neurophysiology Hearing is normal but brainstem auditory evoked potentials are abnormal in 95% of adrenomyeloneuropathy patients and 42% of heterozygote patients. Abnormalities in visual evoked potentials are also found as latencies and are increased in 20% of men with adrenomyeloneuropathy but in more than 70% with childhood cere- bral disease. Electroretinograms are normal. Subtle demyelination and axonal loss patterns of nerve conduction are found in 90% of men and 67% of women with adrenomyeloneuropathy, usually af- fecting the legs more than the arms. Neuropsychological tests can show up deficits in parieto-​occipital function affecting visuospatial parameters and auditory processing, while frontal lobe lesions affect executive functions, emotions, problem solving, and anticipatory processing. Treatment The progressive nature of X-​ALD means that comprehensive family and professional management support services are required. Leukodystrophies are associated with progressive learning diffi- culties, psychiatric disturbance, and increasing disability. Painful muscle spasms are common and should be managed with diazepam, baclofen, or gabapentin. Bulbar muscle function may be lost with disease progression, thus requiring special attention to feeding to reduce the risk of aspiration pneumonia. The routine management of patients with X-​ALD includes regular clinical reviews allied with MRI scanning at approximately 3–​6-​month intervals depending on the rate of progression. Endocrine assessment is performed at baseline and repeated if the clinical syndrome includes features of hypoadrenalism. Dietary therapy was based on the restriction of the intake of C26:0 to less than 15% of normal intake, but early trials showed no effect of this on levels of VLCFA levels. Addition of oleic acid normalized VLCFA levels in fibroblasts and oral glyceryl trioleate reduced VLCFA levels by 50% with an improvement in nerve con- duction measures. A  4:1 combination of glyceryl trioleate and trierucate (Lorenzo’s oil) normalized VLCFA levels within 1 month and prompted mass use of this intervention. No evidence of a clin- ically relevant benefit from dietary treatment with Lorenzo’s oil has been seen in many studies of patients with neurological involvement and X-​ALD, and asymptomatic thrombocytopenia was noted in 30% of patients. The fatty acid composition of the plasma and liver, but not that of the brain, improves with this therapy, suggesting that (A) (B) (b) (A) (c) (B) (C) (D) Fig. 12.9.2  Continued

12.9  Disorders of peroxisomal metabolism in adults 2163 little erucic acid crossed the blood–​brain barrier. Thus, dietary sup- plementation with Lorenzo’s oil is of limited value in correcting the accumulation of saturated VLCFAs in the brain of patients with es- tablished neurological adrenoleukodystrophy. In a study of 89 asymptomatic boys with X-​ALD who had normal MRI scans, Lorenzo’s oil and moderate fat restriction were pre- scribed for 6.9 ± 2.7 years. Plasma fatty acids and clinical status were followed as measures of outcome. Twenty-​four per cent devel- oped MRI abnormalities and 11% developed neurological and MRI abnormalities. The trial concluded that the reduction of C26:0 by Lorenzo’s oil was associated with a reduced risk of developing MRI abnormalities. Lorenzo’s oil therapy is indicated in asymptomatic boys with X-​ALD who have normal brain MRI scans. Experience with other adrenoleukodystrophy patients indicated that total fat in- take in excess of 30 to 35% of total calories may counteract or nullify the C26:0-​reducing effect of Lorenzo’s oil. Patients who develop progressive MRI abnormalities should be considered for haematopoietic stem cell transplantation, but the 5-​year mortality is 38% and survival is increased by 8 months on average. Results in 283 boys with X-​ALD who received haematopoietic cell bone marrow transplantation showed that the estimated 5-​ year survival was 66%. The leading cause of death was disease pro- gression. Donor-​derived engraftment occurred in 86% of patients. Demyelination involved parietal–​occipital lobes in 90%, leading to visual and auditory processing deficits in many boys. Bone marrow transplantation must be considered very early, even in a child without symptoms but with signs of demyelination on MRI, if a suit- able donor is available. There are few data on the usefulness of bone marrow transplantation in adrenomyeloneuropathy. Adrenal function must be monitored since 80% of asymptomatic patients with adrenoleukodystrophy develop evidence of adrenal insufficiency and adrenal hormone replacement therapy should be provided when indicated by laboratory findings. Given the inflammation associated with X-​ALD, a number of immunosuppressive regimens have been investigated. Studies of cyclophosphamide, immunoglobulin, and interferon-​β have been unsuccessful. Prognosis The prognosis in X-​ALD depends on the presentation. As yet, there are no methods of determining which type of disease will result from a given mutation as genotype–​phenotype correlation is poor. Once leukodystrophy begins, the prognosis is poor as progression is inevitable. Data from inherited error bone marrow transplant regis- tries shows prolongations in life with transplantation in X-​ALD but do not record improvements in quality of life. Future developments Other potential therapeutic approaches to X-​ALD include the use of lipid-​lowering drugs. Lowering cholesterol activates human ABCD2 in cultured cells. In mice, a sterol regulatory element exists in the Abcd2 promoter and overlaps sites for liver X receptor/​retinoid X receptor heterodimers. Adipose Abcd2 is induced by SREBP1c, whereas hepatic Abcd2 expression is down-​regulated by concur- rent activation of liver X receptor-​α and SREBP1c. Hepatic Abcd2 expression in liver X receptor-​α/​β mice is inducible to levels vastly exceeding wild type. Statins (3-​HMG-​CoA reductase inhibitors) are capable of nor- malizing VLCFA levels in primary skin fibroblasts derived from X-​ALD patients. They block the induction of proinflammatory cyto- kines through effects on rho kinase. Twelve patients with X-​ALD were treated with lovastatin for up to 12 months. Levels of C26:0 declined from pretreatment values and stabilized at various levels during a period of observation of up to 12 months, which does not correlate with the type of adrenoleukodystrophy gene mutation. In six patients, erythrocyte C26:0 levels fell by 50%. All patients with adrenomyeloneuropathy remained neurologically stable. However, follow-​up trials have been unsuccessful. The PPAR-​α agonist-​mediated induction of ABCD2 expression seems to be indirect and possibly mediated by the sterol-​responsive element-​binding protein 2 in mice. In addition PPAR-​α is involved in the regulation of ELOVL1, a key step in VLCFA synthesis. In vitro CoA esters of both bezafibrate and gemfibrozil inhibit ELOVL1 and could form starting points for novel drug development. However, a study of the pan-​PPAR agonist bezafibrate in 10 male patients failed to show any effects of plasma or erythrocyte VCLFA concentrations despite reducing plasma triglyceride levels as predicted. Studies in animal models have suggested that the PPAR-​γ (with some PPAR-​α activity) agonist pioglitazone may reduce axonal degeneration but there have been no studies in humans. Sodium 4-​phenylbutyrate reduces VLCFA levels through its ef- fects on peroxisomal function and increases adrenoleukodystrophy-​ related protein levels. However, human studies have failed to show consistent beneficial effects. ω-​Oxidation is an alternative oxidation route for VLCFAs. These fatty acids are substrates for the ω-​oxidation system in human liver microsomes and are converted into ω-​hydroxy fatty acids and further oxidized to dicarboxylic acids via cytochrome P450 (CYP)-​mediated reactions. The high sensitivity towards the specific CYP inhibitor 17-​octadecynoic acid suggested that ω-​hydroxylation of VLCFAs is catalysed by the CYP4A/​F subfamilies, particularly CYP4F2 and CYP4F3B, and that therapies capable of increasing ω-​oxidation may have the potential to reduce the progression of the disease. Previously gene therapy has been attempted for X-​ALD using lenti- virus transformation of white cells and 9 to 14% of cells showed re- constitution of ABCD1 expression over 24 months. A more modern gene therapy approach using an adeno-​associated virus construct AAV-​9/​ABCD1 shows appropriate neurological tropism following intracerebroventricular or intravenous injection and reduces plasma and brain VLCFA levels in Abcd1-​deficient mice. Given the extensive oxidative stress associated with demyelin- ation in X-​ALD there has been interest in antioxidant therapies in the treatment of X-​ALD. A combination of the antioxidants α-​ tocopherol, N-​acetyl-​cysteine, and α-​lipoic acid reduced demyelin- ation in Abcd1-​deficient mice. There are no studies of this approach in humans. There has been an explosion of interest in novel therapeutic strat- egies for inherited errors of metabolism. Classical enzyme replace- ment therapy for X-​ALD is impossible given the need to replace a peroxisomal transporter molecule, but other strategies utilizing technologies such as stabilized mRNA technology combined with liposome or other shielded delivery technologies allied with various methods of delivering tissue specificity may be more successful. These show promise in cell culture models but no studies have yet been performed in human X-​ALD. None of these technologies

section 12  Metabolic disorders 2164 have reached animal models let alone human trials in peroxisomal diseases. Neuro-​ophthalmic adult peroxisomal disorders Introduction Though survival is improving for peroxisomal biogenesis disorders and more subtle defects are now diagnosed, most still present in the neonatal period or in infancy. This is also true for most single en- zyme peroxisomal deficiencies. Only one group of disorders presents later, with the onset of symptoms often in early teenage years but, due to delays in diagnosis, many are not identified until they reach adulthood. In contrast to the neuropsychiatric or endocrine pres- entation associated with adrenoleukodystrophy, these peroxisomal disorders present as central and peripheral neuropathies—​a neuro-​ ophthalmic picture. They are often termed Refsum’s disease though, given the multiple underlying genetic defects, it would be better to refer to them as Refsum’s syndrome. The syndrome comprises three genetic disorders:  phytanoyl-​CoA hydroxylase deficiency (clas- sical adult Refsum’s disease), atypical rhizomelic chondrodysplasia punctata type 1, and the newly described α-​methylacyl-​CoA racemase deficiency. Historical perspective Adult Refsum’s disease (OMIM 266510), also called heredopathia atactica polyneuritiformis, is a hereditary sensory motor neur- opathy type IV. It was first described in 1947, but only recognized as a syndrome by Refsum in 1962. He described a constellation of signs comprised of retinitis pigmentosa, anosmia, deafness, ataxia, and polyneuropathy allied with raised levels of protein in the cere- brospinal fluid. The biochemical defect was identified in 1963 when phytanic acid was noted in the plasma of affected patients and de- fective α-​oxidation was later suggested as the cause of adult Refsum’s disease. This disease was thought to be unifactorial with admittedly some rare aberrant complementation studies until 1995 when, after the localization of the gene for phytanoyl-​CoA hydroxylase, up to 50% of cases in one series were shown not to be linked to chromo- some 10 but to chromosome 6. Eventually, the novel defect was iden- tified as a variant of rhizomelic chondrodysplasia punctata type 1 and caused by mutations in peroxin 7. In parallel with this discovery, three patients were described in 1997 with a phenotype of sensory neuropathy and a subtle bile acid disorder but whose families in- cluded siblings with a Refsum’s-​like syndrome which was identified as due to a deficiency in α-​methylacyl-​CoA racemase. A clinical phenocopy associated with polyneuropathy, hearing loss, ataxia, ret- initis pigmentosa, and cataract (PHARC) (OMIM 612674) has re- cently been described. In contrast to other Refsum-​like syndromes, phytanic acid levels are normal in this condition. Clinical features In contrast to Zellweger’s syndrome (OMIM 214100), neonatal adrenoleukodystrophy (OMIM 202370), infantile Refsum’s dis- ease (OMIM 266500), and rhizomelic chondrodysplasia (OMIM 601757), adult Refsum’s disease usually presents in late childhood with progressive deterioration of night vision, the occurrence of progressive retinitis pigmentosa, and anosmia (Table 12.9.3). Anosmia, contrary to early reports, is a constant feature of adult Refsum’s disease. After 10 to 15 years, deafness, ataxia, polyneur- opathy, ichthyosis, and cardiac arrhythmias can occur. Short meta- carpals or metatarsals are found in about one-​third of patients. Rare findings include psychiatric disturbance and proteinuria. Premature death may result from cardiac arrhythmias. α-​Methylacyl-​CoA racemase (OMIM 604489)  presents with adult-​onset sensorimotor neuropathy (Table 12.9.3). It may be ac- companied by retinitis pigmentosa, visual field restriction and loss of acuity, axonal sensorimotor neuropathy, and myopathy-​like adult Refsum’s disease. Other features described have included primary hypogonadism, hypothyroidism, spastic paraparesis, epileptic seiz- ures, and mild developmental delay. More severe childhood-​onset cases have shown a phenotype of defects in bile acid synthesis al- lied with fat-​soluble vitamin deficiencies, coagulopathy, and cholestatic liver disease and a resemblance to a Niemann–​Pick type C phenotype. Table 12.9.3  Comparison of clinical features of underlying metabolic defects associated with adult Refsum’s disease Adult Refsum’s disease
(n c. 300) Rhizomelic chondrodysplasia (n c.5) α-​Methylacyl-​CoA racemase (n c.6) PHARC (n c.30) Retinitis pigmentosa Age >12 Age >12 Age >20 Age >30 Cataract Age >30 Age >30 ? Age >5 Anosmia All All ? Absent Sensorineural deafness Age >40 Age >40 ? Age >5 Sensory neuropathy Age >20 PA dependent Age >20 PA dependent Age >30 Axonal/​ demyelinating Age >30 Variable progressive Ataxia Age >20 PA dependent Age >20 PA dependent Variable Progressive Cholestasis No No Yes No Biochemistry: Phytanic acid Pristanic acid (μmol/​L)

300 <0.2 100 <0.2 <200 10 Normal (<10) Normal (0.5–​3) PA, phytanic acid; PHARC, polyneuropathy, hearing loss, ataxia, retinitis pigmentosa, and cataract.

12.9  Disorders of peroxisomal metabolism in adults 2165 PHARC shares many clinical features of Refsum’s disease but lacks the anosmia and possibly the osteological changes (Table 12.9.3). Some mitochondrial disorders in the Leigh’s syndrome spectrum, including neurogenic muscle weakness, ataxia, and retinitis pig- mentosa (NARP), caused in many cases by mutations in mitochon- drial MT-​ATP6, may also share some clinical features with adult Refsum’s disease. Aetiology Phytanic acid (3R,S,7R,11R,15-​tetramethylhexadecanoic acid) is an isoprenoid lipid derived from the phytol side chain of chlorophylls by bacterial degradation in ruminants, invertebrates, or pelagic fish (see Fig. 12.9.3). Phytol can be oxidized to an unsaturated fatty acid, phytenic acid, and this is saturated to phytanic acid by a pathway involving fatty aldehyde dehydrogenase 10 (FALDH-​10) in micro- somes. The significance of this pathway in humans is unclear though high phytanic acid levels have been described in some patients defi- cient in FALDH-​10 with Sjögren–​Larsson syndrome. Most phytanic acid is ingested from the adipose tissue and muscle of herbivores or pelagic fish. The average human daily dietary intake of phytanic acid in Western societies is between 50 and 100 mg, of which about 50% is absorbed and metabolized. Phytanic acid is transported in plasma bound to very low-​density lipoprotein and later low-​density lipoprotein, with its elimination allied to reverse cholesterol transport (high-​density lipoprotein). Phytanic acid is preferentially taken up by the liver and may account for up to 50% of the free fatty acid pool in hepatocytes. This pool is labile and can be acutely mobilized by stress, infection, or starva- tion, resulting in rapid phytanic acid release. Plasma phytanic acid concentrations are less than 10% of the levels found in adipose tissue and neurons, which accumulate phytanic acid because of its hydro- phobicity. The elimination half-​life of total body phytanic acid is usually between 1 and 2 years. Most fatty acids are metabolized by the β-​oxidation path- ways in peroxisomes and mitochondria. Phytanic acid cannot be metabolized by this route due to the presence of a β-​methyl group. Instead, phytanic acid is metabolized either by α-​oxidation to pristanic acid, or by ω-​oxidation from the other end of the molecule. Using radiolabelled [14C]-​phytanic acid as a substrate, an enzyme activity responsible for the α-​oxidation of phytanic acid in cell lys- ates was described in 1967. This activity was eventually localized within peroxisomes and, after 30 years, the pathway responsible for α-​oxidation has been clarified. α-​Oxidation of phytanic acid Most phytanic acid metabolism occurs in the liver and kidney by α-​oxidation, though skin fibroblasts are used for clinical diagnostic purposes. Phytanic acid from plasma enters the peroxisome in asso- ciation with the sterol carrier protein-​2 (SCP-​2) and is metabolized by a four-​step initial α-​oxidation pathway (Fig. 12.9.4). Unusually, it appears this pathway can metabolize two stereoiso- mers of its substrate equally well. One carbon atom is then removed from the latter in a lyase reaction to give pristanal and formyl-​CoA. Pristanal is then oxidized to pristanic acid which is thio-​esterified using CoA to give a racemic mixture. The action of α-​methylacyl-​ CoA racemase converts the (2R)-​epimer to the (2S)-​epimer. Further degradation of (2S)-​pristanic acid by the stereospecific β-​oxidation pathway then occurs, with the release of propionyl and acetyl-​CoA units. Further β-​oxidation reactions (including epimerization) are required to generate the dimethylundecanoic and dimethylnonanoic and methyl-​heptanoic acid derivatives, which are finally exported for mitochondrial β-​oxidation. Disordered ω-​oxidation of phytanic acid Patients with adult Refsum’s disease are unable to detoxify phytanic acid by α-​oxidation, and so the ω-​oxidation pathway is the only metabolic pathway available for its degradation (Fig. 12.9.5). This pathway produces 3-​methyladipic acid as the final metab- olite, which is excreted in the urine. Thus, 3-​methyladipic acid concentrations can be used as an index of the molar activity of the 0 0 10 20 30 Age (years) 40 50 60 RP Ichthyosis Ataxia Deafness Neuropathy Anosmia 2 4 6 8 10 Number of patients 12 14 16 Fig. 12.9.3  Cumulative incidence of clinical features on presentation of 15 patients with Refsum’s disease. RP, retinitis pigmentosa. Reproduced from Wierzbicki AS, et al. (2002). Refsum’s disease: a peroxisomal disorder affecting phytanic acid alpha-​ oxidation. J Neurochem, 80, 727–​35, with permission.

section 12  Metabolic disorders 2166 ω-​oxidation pathway. After ingestion of a test load of phytanic acid, 3-​methyladipic acid is detected in the urine of healthy controls and adult Refsum’s disease heterozygotes showing that ω-​oxidation plays a significant role in postprandial metabolism of phytanic acid in humans. The activity of the ω-​oxidation pathway is approximately doubled in patients with adult Refsum’s disease, but this microsomal pathway has considerable reserve capacity. The balance of intake of phytanic acid and its ω-​oxidation is likely to determine long-​term concentra- tions of the lipid. Patients with adult Refsum’s disease often clinically relapse during episodes of illness or drastic weight loss. Fasting in- duces ketosis and lipolysis and acute mobilization of phytanic acid in hepatocyte and adipocyte fatty acid pools. This process can in- duce a release of 5000 mg (c.15 mmol) per day of phytanic acid (50 times normal). In experimental ketosis, following acute starvation, phytanic acid doubled in 29 h in patients with adult Refsum’s dis- ease and an 80% rise was seen in urinary 3-​methyladipic acid levels, indicating that ω-​oxidation was buffering part of this rise. Phytanic acid concentrations can exceed the capacity of the residual α-​ and ω-​oxidation pathways. Excess phytanic acid is excreted by low-​ affinity pathways. Phytanic acid can be glucuronidated and it can also be lost nonspecifically in the urine as nephropathy is a feature of adult Refsum’s disease. The enzymology of the ω-​oxidation pathway in adult Refsum’s disease has been clarified and occurs through the microsomal CYP4A system as well as the peroxisome. The capacity of the ω-​ oxidation pathway has been measured by the excretion of 2,6-​ dimethyloctanedioic acid (the C10 ω-​2-​methyl thioester derivative of phytanic acid) at 30 mg phytanic acid (89 µmol) per day. However, other studies measuring 3-​methyladipic acid excretion showed a far lower capacity of 6.9 mg (20.4 µmol) per day. These differences in activity may reflect the metabolic fates of the respective markers. Both 2,6-​dimethyloctanedioic acid and 3-​hexanedioic acid are products of the initial steps of ω-​oxidation and may be dependent on carnitine ester formation for activation and further metabolism. The initial steps of ω-​oxidation appear to have a greater capacity than that of the whole pathway when measured by the final product 3-​methyladipic acid. Molecular toxicology of Refsum’s syndrome The exact mechanism of the toxicity of phytanic acid to neuronal, cardiac, and bone tissue is gradually being clarified. Structural Fig. 12.9.4  Metabolic pathway for α-​oxidation of phytanic acid and β-​oxidation of pristanic acid producing 4,8 dimethynonanoyl-​ CoA, propionyl CoA, and acetyl-​CoA as end-​products. This pathway is integrated with PXMP-​2 porin responsible for transport of small molecules, PMP34 for ATP, and a transporter for phytanoyl-​CoA. Reproduced with permission from Wanders RJA, Komen J, Ferdinandusse S. Phytanic acid metabolism in health and disease. Biochimica et Biophysica Acta (BBA) -​ Molecular and Cell Biology of Lipids. Copyright © 2011 Elsevier B.V. Fig. 12.9.5  ω-​Oxidation mediated by a microsomal CYP4A enzyme yields phytane 1,16 dioic acid which can be metabolized to a wide spectrum of metabolites by β-​oxidation. Reproduced with permission from Wanders RJA, Komen J, Ferdinandusse S. Phytanic acid metabolism in health and disease. Biochimica et Biophysica Acta (BBA) -​ Molecular and Cell Biology of Lipids. Copyright © 2011 Elsevier B.V.

12.9  Disorders of peroxisomal metabolism in adults 2167 homology between phytanic acid and vitamin A, vitamin E, geranyl pyrophosphate, and farnesyl pyrophosphate has been noted and it has been suggested that phytanic acid may have a role in the regu- lation of isoprenoid metabolism and protein prenylation. Recent studies have identified that phytanic acid and also pristanic acid are direct toxins to mitochondria and it has been found that phytanic acid has a rotenone-​like action in uncoupling complex I in the oxida- tive phosphorylation chain in the mitochondrial inner membrane, with subsequent likely production of reactive oxygen species, causes secondary calcium-​driven changes through GPR40, and induces apoptosis in neuronal cells through a histone deacetylase-​mediated mechanism. This metabolic toxicity may explain why neuronal or allied retinal pigment tissues rich in mitochondria are the prime tis- sues affected in adult Refsum’s disease. The molecular toxicology of pristanic acid is unknown, although it is likely that the mild ophthalmic features seen in some cases may relate to phytanic acid toxicity as for phytanoyl-​CoA hydroxylase deficiency. Although both di-​ and trihydroxycholestanoic acids levels are elevated in α-​methylacyl-​CoA racemase, there is no phenotype of itching associated with this disorder. The cause of the sensory neuropathy in α-​methylacyl-​CoA racemase still remains to be determined. Molecular genetics The defect in adult Refsum’s disease was soon identified as being due to the lack of an α-​oxidase. It took 30 years for the enzyme responsible, phytanoyl-​CoA hydroxylase, to be identified. Two groups identified the gene for phytanoyl-​CoA hydroxylase simul- taneously in 1997. The phytanoyl-​CoA hydroxylase gene includes nine exons and codes for a 338-​amino acid protein including the 30-​amino acid signal domain, which is cleaved on entry into the peroxisome. Like all the peroxisomal targeting sequence type 2 proteins, phytanoyl-​CoA hydroxylase is transported into the peroxisomes by the protein transporter peroxin 7. Deficiency in this transporter is responsible for rhizomelic chondrodysplasia punctata (RCDP) type 1. Phytanoyl-​CoA hydroxylase is an iron (II) and 2-​oxoglutarate-​dependent oxygenase, with little overall sequence similarity to other human oxygenases. Numerous point and splice mutations in phytanoyl-​CoA hydroxylase have now been described in adult Refsum’s disease patients, many of which affect 2-​oxoglutarate conversion. Significantly, all cause complete inactivation of the protein; no partial function mutations have yet been identified. Genetic mapping studies have shown that in most cases, but not all, classical adult Refsum’s disease maps to chromosome 10. The locus for the second form of adult Refsum’s disease, comprising about 10% of cases, was localized to chromosome 6q22–​q24 and biochemical studies of fibroblasts from patients with adult Refsum’s disease established that these patients have subtle deficiencies of per- oxisomal targeting sequence type 2-​dependent enzyme functions (plasmalogen synthesis) consistent with mild variants of RCDP, though they lack any clinical signs specific to childhood-​onset form of RCDP where significant deficiencies in plasmalogen syn- thesis result in intellectual impairment and other neurological signs. Ironically, one of the original patients described with adult Refsum’s disease turned out to have the RCDP variant. A limited number of mutations have been described that cause Refsum’s–​RCDP and it is unclear why these mutations should preferentially lead to specific mislocalization of phytanoyl-​CoA hydroxylase in contrast to the other peroxin 7 imported proteins. Epidemiology Neuropathic adult peroxisomal disorders are rare, with a prevalence of 1 in 106 in Europe and, for unexplained reasons, 10-​fold less in the United States of America. As with all recessive conditions, they are more common in cultures or localities with strong founder ef- fects where consanguineous marriages are frequent. The classical Refsum’s phenotype is usually found in genetic ophthalmic services where it may represent 1% of retinitis pigmentosa cases. No sur- veys have been performed on the incidence of α-​methylacyl-​CoA racemase among patients with neuropathy. Differential diagnosis The differential diagnoses of the neuropathic disorders and relevant signs and investigations are shown in Tables 12.9.4 and 12.9.5. With classical adult Refsum’s disease, the differential diagnosis includes the various genetic retinitis pigmentosa syndromes if neurological signs are subtle and other rare neurological disorders (Table 12.9.6). Clinical investigation The key investigations in the case of suspected neuropathic adult Refsum’s disease are the measurement of phytanic acid (for adult Refsum’s disease) and pristanic acid (for suspected α-​methylacyl-​ CoA racemase). These are diagnostic. For clinical staging purposes, electroretinograms are often performed but often show flat responses characteristic of well-​ established retinitis pigmentosa. Visual fields should be assessed regularly as functional diplopia is a long-​term complication of adult Refsum’s disease. Slit-​lamp examination for cataracts is also indi- cated, as these can be treated. Ideally, retinal photography should be performed so that the extent of retinitis pigmentosa and its pro- gression can be monitored on a long-​term basis. Anosmia can be detected by screening using the standard four-​bottle smell test, but is best quantified by more extensive profiles (e.g. the University of Pennsylvania smell identification test). Auditory function should be assessed by auditory evoked potentials and hearing tests and moni- tored every 5 years. Peripheral neuropathy should be investigated by peripheral nerve conduction studies for somatosensory poten- tials and electromyography. A  nonspecific demyelination pattern is typical of adult Refsum’s disease. Osteo-​ or chondrodysplasia is best identified by a radiological survey of hands and feet for short metatarsals and knee radiology for signs of current or previous chondrodysplasia. Subtler signs that may accompany these definitive tests include an electrolyte profile showing mild hypokalaemia and a Fanconi-​ like aminoaciduria which can occur in adult Refsum’s disease. Liver function tests should be performed. If bilirubin is raised or α-​methylacyl-​CoA racemase is suspected, a detailed bile acid profile should be performed by mass spectrometric methods. As the differ- ential diagnoses include vitamin deficiencies, vitamin A and E levels should be measured to exclude retinol-​deficiency retinopathy and tocopherol-​deficient ataxia. Vitamin B12 and folate determinates are used to exclude cobalamin/​folate deficient neuropathy. To differentiate phytanoyl-​CoA hydroxylase from peroxin 7 adult Refsum’s disease, it is necessary to measure plasma VLCFAs and plasmalogens. However, often the deficiencies are subtle and these

section 12  Metabolic disorders 2168 investigations may appear normal. For a definitive diagnosis, a skin biopsy should be taken, fibroblasts grown, and detailed enzyme and immunofluorescence profiles examined in a specialist peroxisomal laboratory. Criteria for diagnosis The pathognomonic finding in adult Refsum’s disease is greatly ele- vated phytanic acid concentrations in the plasma (>200 µmol/​litre; normal <30 µmol/​litre), in contrast to other peroxisomal disorders where levels are usually lower and other metabolic abnormalities are also present. Unlike in rhizomelic chondrodysplasia punctata or the peroxisomal biogenesis disorders, no intellectual defects are seen, bone abnormalities are mild (if present at all), and there is no de- fect in plasmalogen synthesis. In infantile Refsum’s disease, which is a mild clinical variant of the peroxisomal biogenesis disorder encompassing Zellweger’s disease as its most severe form, numerous Table 12.9.4  Differential diagnosis of treatable adult neuropathies caused by inborn errors of metabolism Disease Onset Neurology Signs Chemistry Treatment Screening Fabry’s disease 10–​20 Small fibre, sensory Stroke; cardiomyopathy; renal Low
α-​galactocerebrosidase ERT WBC
α-​galactocerebrosidase Serine deficiency 10–​20 Axonal Growth delay; ichthyosis Low CSF/​plasma serine Serine Plasma amino acids Cerebrotendinous xanthomatosis 10–​40 Axonal, demyelination, sensorimotor Mental retardation; ataxia, spastic paraparesis. Tendon xanthomata Cholestanol Chenodeoxycholate Cholestanol Adult Refsum’s disease/​syndrome 10–​50 Demyelination, sensorimotor Retinitis pigmentosa, ataxia, anosmia Phytanic acid Low phytanic acid diet Phytanic acid Porphyrias 10–​50 All Neuropsychiatric Dermatological PBG and δALA Various PBG and δALA Wilson’s disease 15–​50 Axonal, demyelination, sensorimotor Movement disorder Copper/​caeruloplasmin Chelation Copper/​ caeruloplasmin CSF, cerebrospinal fluid; δALA, δ-​aminolaevulinic acid; PA, phytanic acid; PBG, porphobilinogen. Reproduced from Sedel F et al (2007) Peripheral neuropathy and inborn errors of metabolism in adults. J Inherit Metab Dis, 30, 642–​53, with permission. Table 12.9.5  Differential diagnosis of other adult neuropathies caused by inborn errors of metabolism Disease Age of onset Neuropathy Signs Chemistry Treatment Screening Mitochondrial myopathy 15–​50 All Retinitis pigmentosa, epilepsy, ataxia CSF/​plasma lactate None Lactate, muscle biopsy Metachromatic leukodystrophy 15–​50 Demyelination, sensorimotor Psychiatry, ataxia Aryl-​sulphatase A None/​bone marrow transplant Aryl-​sulphatase A Krabbe’s disease 15–​50 Demyelination, sensorimotor Spastic paraparesis WBC galactocerebrosidase None/​Bone marrow transplant WBC galactocerebrosidase GM2 gangliosidosis 15–​50 All Psychiatry; ataxia, Hexosaminidase None WBC hexosaminidase AMACR 10–​50 Demyelination, sensorimotor Retinitis pigmentosa, ataxia, anosmia, IQ Pristanic acid, D/​ THCA Low PA diet Pristanic acid Abetalipoproteinaemia 5–​20 Axonal, sensory, sensorimotor Ataxia, movement disorder, retinitis pigmentosa, acanthocytes Low cholesterol, low apolipoprotein B, vitamins A and E Vitamins A and E Apolipoprotein B Vitamin E deficiency 10–​20 Axonal, demyelination, sensorimotor Ataxia, movement disorder, retinitis pigmentosa, acanthocytes Vitamin E Vitamin E Vitamin E Homocysteine metabolism (CblC) 15–​50 Axonal, motor neuron disease, sensorimotor Psychiatric, stroke, leukoencephalopathy; Macrocytosis Homocysteine; methylmalonic acid Folate, vitamins B12 and B6, betaine Homocysteine X-​ALD 15–​50 Axonal, demyelination, sensorimotor Neuropsychiatric leukoencephalopathy; adrenal failure Very long-​chain fatty acids ?Lorenzo’s oil Very long-​chain fatty acids NARP 5–​20 Demyelination, sensorimotor Retinitis pigmentosa ataxia Mitochondrial—​MT-​ ATP6 None; supportive Mitochondrial DNA PHARC 10–​30 Demyelination, sensorimotor Retinitis pigmentosa ataxia, cataract Endocannabinoid(?) None None AMACR, α-​methylacyl-​CoA racemase; CSF, cerebrospinal fluid; D/​THCA, di-​/​trihydroxycholestanoic acid; PA, phytanic acid; WBC, white blood cell. Adapted from Sedel F, et al. (2007). Peripheral neuropathy and inborn errors of metabolism in adults. J Inherit Metab Dis, 30, 642–​53, with permission.

12.9  Disorders of peroxisomal metabolism in adults 2169 subtle peroxisomal defects are present and the condition presents from birth. In α-​methylacyl-​CoA racemase neuropathy, the pathognomonic findings are raised levels of pristanic acid (>100 µmol/​litre) allied with increases in di-​ and trihydroxycholestanoic acids. A secondary elevation of phytanic acid may be seen, but levels are usually be- tween 50 and 100 µmol/​litre. Treatment Long-​term prospects for the treatment of adult Refsum’s disease (or at least for some forms) are good as it is one of the few inherited disorders of metabolism with an exogenous precipitating cause. The disease is treated symptomatically by restriction of phytanic acid in- take in the diet or its elimination by plasmapheresis or apheresis. These regimens reduce plasma phytanic acid levels by between 50 and 70%, to values typically around 100 to 300 µmol/​litre, and can eliminate phytanic acid completely from fat stores in some patients. There is long-​term efficacy and safety data for a phytanic acid re- strictive diet and to a lesser extent for plasmapheresis/​apheresis, which are used in a few centres. Treatment is generally most effective in resolving symptoms of ichthyosis, less so sensory neuropathy and least so ataxia, and it has uncertain effects on the progression of ret- initis pigmentosa, anosmia or deafness, although it seems to sta- bilize these signs. Prognosis The prognosis in adult Refsum’s disease depends on the degree to which phytanic acid concentrations are decreased. In untreated disease, presentation is with progressive weakness and neuropathy usually following an acute infective illness which leads to anorexia and acute hepatic phytanic acid release exacerbating the condition. Concentrations of phytanic acid in the plasma usually exceed 1000 µmol/​litre. Left untreated, cardiomyopathy and sudden death can occur. If phytanic acid levels are reduced by plasmapheresis and by adequate parenteral nutrition, and then a low phytanic acid diet is followed, prognosis is good. Any myopathy usually resolves within 2 to 3 weeks, though acute visual and auditory deterioration may be irrecoverable. In long-​term cases, patients are blind, deaf, and anosmic, have extensive peripheral myopathy, and are often wheelchair bound. In acute adult Refsum’s disease, once phytanic acid levels fall to less than 500 µmol/​litre, ichthyosis resolves followed by improvement in Table 12.9.6  Differential diagnosis of retinitis with neurological signs (apart from adult Refsum’s disease) Presentation OMIM Neurological and other signs Abetalipoproteinaemia 200100 Ataxia, movement disorder, retinitis pigmentosa, acanthocytes Vitamin E deficiency 600415 Ornithine aminotransferase deficiency 258870 Gyeate atrophy Myopathy Usher’s syndrome Ia 276900 Congenital deafness, ataxia Usher’s syndromes II 276901 Moderate progressive deafness Bardet–​Biedl–​Moon syndrome 209900 Polydactyly Truncal obesity Hypogonadism Short stature Mental retardation Kearns–​Sayre syndrome 530000 Ophthalmoplegia Cardiomyopathy Ceroid lipofuscinosis (Batten’s disease) 204300 Seizures Dyskinesia Dementia X-​linked macular degeneration 304020 Ataxia Myoclonic encephalopathy NARP 551500 Neurogenic proximal muscle weakness Ataxia Dementia Seizures PHARC 612674 Polyneuropathy Hearing loss Ataxia Cataract OMIM, Online Mendelian Inheritance in Man.

section 12  Metabolic disorders 2170 myopathy and neuropathy. If phytanic acid levels can be restored to normal values, then it is likely that ophthalmological changes will be minimal or slow, but sudden step-​like deteriorations can occur. The principal long-​term disability is increasing loss of visual field with subsequent diplopia and progressive cataract formation. Auditory function generally remains good unless phytanic acid levels are sub- stantially raised, in which case audiological deterioration occurs with the need for cochlear implants. Although acute myopathy re- solves, patients may have muscle spasms or contractures which may be either related to the adult Refsum’s disease or secondary to the osteodystrophy. Splints and the surgical correction of osteopathy may be required. Other issues Adult Refsum’s disease is a potentially treatable cause of retinitis pig- mentosa or neuropathy. The average delay to diagnosis is 12 years and a simple biochemical screening test exists for these disorders. Given that earlier implementation of dietary restriction of phytanic acid would likely arrest the disease process before retinitis is estab- lished, screening for phytanic and pristanic acidaemias should be considered as an important investigation in retinitis pigmentosa or peripheral neuropathy. Future developments The causes of neuropathic adult peroxisomal disorders are in- completely delineated. Cases of pristanic acidaemia with an adult Refsum’s disease phenotype exist for which no cause has yet been found. Similarly, all cases of adult Refsum’s disease currently de- scribed are null-​function variants, so the phenotype associated with low partial function has not been identified. It may be entirely normal, but it is possible that some cases of retinal dystrophy or peripheral neuropathy may actually be caused by mild phytanoyl-​ CoA hydroxylase mutations. No cases of deficiency of phytanic acid lyase have been described, although the condition should show signs of an adult Refsum’s disease phenotype with some overlap into the α-​oxidation pathway-​associated features of α-​methylacyl-​ CoA racemase and possibly non-​neurological effects of 2-​hydroxy-​ fatty acid deficiency. A number of lyase enzymes with peroxisomal targeting signal motifs remain to be placed on the α-​oxidation pathway and these may be associated with neuropathy or retinitis pigmentosa syndromes. Reduction of dietary phytanic acid is already successful in ameli- orating most nonophthalmic symptoms in long-​term studies with diet and to a lesser extent apheresis. Orlistat has been shown to re- duce phytanic acid in some patients. Newer, more efficacious, ther- apies are still required to fully reverse the progression of this disease. The signalling pathways which regulate α-​oxidation in humans are unclear. In rodents, the retinoid X receptor β and peroxisomal proliferator-​activating receptor α pathways control α-​oxidation and thus fibrate (PPAR-​α agonist) therapy increases activity, but this does not seem to be true in humans. As ω-​oxidation is capable of large increases in activity and is principally mediated through CYP enzymes, it forms a good candidate for therapeutic interventions to induce enzyme activity and reduce phytanic acid levels in Refsum’s disease. However, at the present time, no drug therapy trials of com- pounds capable of modulating either the α-​ or the ω-​oxidation path- ways have been conducted in humans. PHARC (polyneuropathy, hearing loss, ataxia, retinitis pigmentosa, and cataract) syndrome A phenocopy sharing many features of adult Refsum’s disease has recently been described (Table 12.9.3). This disorder forms part of a newly designated group of phospholipid, sphingolipid, and fatty acid synthesis disorders. The synthesis of phospholipids involves multiple steps from dihydroxyacetone phosphate (DHAP) and gly- cerol (Fig. 12.9.6). Clinical features PHARC is a progressive syndrome and has the key clinical features of polyneuropathy, hearing loss, ataxia, retinitis pigmentosa, and cata- ract (MIM 612674). It is marked by early-​onset cataract and hearing loss, retinitis pigmentosa, and involvement of both the central and peripheral nervous systems, including demyelinating sensorimotor polyneuropathy and cerebellar ataxia. Approximately 30 cases have been described. Most seem to have an adult Refsum’s disease-​type phenotype but lack the key feature of anosmia which is universally present in adult Refsum’s disease. PHARC seems to be a progressive disorder, unlike adult Refsum’s disease, and seems not to show tem- poral or environmental variability in symptoms. Clinically, the onset of retinitis pigmentosa seems to be later in PHARC than in adult Refsum’s disease (approximately age 35–​40 vs 15–​20 years), cataracts occur earlier (age 20–​30 vs 30–​50 years), and early-​onset deafness is common (age <15 years vs 50 years in adult Refsum’s disease). Patients have a demyelinating neuropathy and develop chronic ataxia at varying ages. Cerebellar atrophy is more common on MRI (25%) than in adult Refsum’s disease, but milder phenotypes have retinitis pigmentosa associated with only mild sensorineural hearing loss and cataract and no other signs or symptoms. Diagnosis The biochemical profile differs from adult Refsum’s disease or other α-​oxidation disorders in showing no abnormality in phytanic or pristanic acid concentrations or in any other peroxisomal metabolites. PHARC was originally mapped to chromosome 20 and is caused by mutations in α/​β hydrolase-​12 (ABHD12), an enzyme involved in hydrolysing the endocannabinoid 2-​arachidonyl-​glycerol and possibly other structurally related substrates including very long chain fatty acids. ABDH12 can be inhibited by triterpenoids (e.g. betulonic acid). Phytol and phytanic acid are diterpenes. The me- tabolism of 2-​arachidonyl-​glycerol in the brain is complex as 85% is metabolized by mono-​acylglycerol lipase (MAGL) or ABDH6, while ABDH12 has a far greater role in microglial 2-​arachidonyl-​glycerol metabolism suggesting that microglial dysfunction is the patho- genic mechanism behind PHARC. Both ABDH12 and ABDH16 are involved in cellular lyso-​phosphatidylserine metabolism and modu- late cytokine production. How these biochemical findings relate to causing the full clinical phenotype is unclear. Treatment No specific treatment has been defined for PHARC. Symptomatic treatment is recommended for individual deficits. The effective- ness of a dietary restriction of phytanic acid in this condition is

12.9  Disorders of peroxisomal metabolism in adults 2171 Dihydroxyacetone 3-phosphate Glycerol-3-phosphate Glycerol Monoacylglycerol ADP ATP CTP PPi 14 18 15 16 17 O P O O O CO CH2 CH2 CH2 CH2 CO CH O CH2 CH2 CO Cardiolipin CO O O O C O H HO P CH O 19 4 3 5 7 8 6 1 1 2 4 Lysophosphatidic Acid Phosphatidic Acid ADP ATP CDP Diacylglycerol Diacylglycerol Diacylglycerol Triacylglycerol Phosphatidyl Serine Phosphatidyl Inositol Phosphatidyl Choline Phospho- choline choline ADP ATP CMP CDP- choline Phosphatidyl Ethanolamine Myoinositol CMP Phosphoglycero- phosphate Phosphatidyl- Glycerol Phospho- glycerol Lyso- phospholipids COOH Prostaglandins Leukotriens Arachidonic Acid 2-Arachidonoyl Glycerol Glycerol H2O H2O H2O Fatty acid Fatty acid 10 9 9 13 12 11 + Acyl CoA CoA-SH Acyl CoA CoA-SH Fig. 12.9.6  Major reactions involved in phospholipids biosynthesis: CDP, cytidyldiphosphate; CMP, cytidylmonophosphate; CoA, coenzyme A; CTP, cytidyltriphosphate. Phospholipids are synthesized de novo from glycerol or from glyceraldehyde 3-​phosphate and dihydroxyacetone phosphate (DHAP). Glycerol-​ 3-​phosphate is esterified twice with acyl-​CoA by acyl-​CoA transferase (1) to form lysophosphatidic (LPA) acid and then phosphatidic acid (PA). In adipose, muscular tissues, and skin, LPA is converted into PA by α/​β-​hydrolase-​5 (ABHD5) (2). In mitochondria, PA and LPA are obtained from monoacylglycerol (MAG) and diacylglycerol (DAG), respectively, a reaction catalysed by acylglycerol kinase (4). PA can be converted either into DAG by phosphatidic acid phosphatase (3), or into CDP-​DAG by phosphatidic acid cytidyltransferase (14). DAG is an essential inter-​mediate for the synthesis of triglycerides and of various phosphoglycerides: phosphatidyl serine (PS) by PS-​synthase (5), which is transformed to phosphatidyl ethanolamine (PE) by PS-​carboxylase (6), phosphatidyl inositol (PI) by PI-​synthase (7) and phosphatidyl choline (PC) by PC-​synthase (8). On the other side, CDP diacylgly-​cerol is converted, by the sequential action of phosphatidic acid glycerol phosphate synthase (15) and phosphatase (16), into phosphatidylglycerophosphate—​a precursor of cardiolipin, an important phospholipid of the mitochondrial membrane. This is catalysed by cardiolipin synthase (17) and enriched by linoleic acid by the remodelling enzyme, monolysocardiolipin acyl transferase (also called tafazzin). The alternative pathway for DAG and fatty acids synthesis is their release from membrane phospholipids by specific phospholipases (PLA2) (9) and lysophospholipases (Neuropathy target esterase) (10). Phospholipase C (11) leads to the membrane release of DAG and its further conversion by diacylglycerol lipase (12) into 2-​arachidonoyl glycerol, which can be hydrolysed into arachidonic acid by α/​β-​hydrolase-​12 (ABDH12) (13). Arachidonic acid is the starting molecule of many complex fatty acids like prostaglandins and leukotrienes. Reproduced with permission from Lamari, F., Mochel, F., Sedel, F. et al. J Inherit Metab Dis (2013) 36: 411. https://​doi.org/​ 10.1007/​s10545-​012-​9509-​7. Copyright © 2012 SSIEM and Springer.

section 12  Metabolic disorders 2172 unknown, but unlikely to be completely efficacious given the dif- ferent aetiology. FURTHER READING Ahmed RM, et al. (2014). A practical approach to diagnosing adult onset leukodystrophies. J Neurol Neurosurg Psychiatry, 85, 770–​81. Aubourg P, et al. (1993). A two-​year trial of oleic and erucic acids (‘Lorenzo’s oil’) as treatment for adrenomyeloneuropathy. N Engl J Med, 329, 745–​52. Baldwin EJ, et  al. (2010). The effectiveness of long-​term dietary therapy in the treatment of adult Refsum disease. J Neurol Neurosurg Psychiatry, 81, 954–​7. Berger J, Forss-​Petter S, Eichler FS (2014). Pathophysiology of X-​ linked adrenoleukodystrophy. Biochimie, 98, 135–​42. Brown PJ, et al. (1993). Diet and Refsum’s disease. The determination of phytanic acid and phytol in certain foods and application of this knowledge to the choice of suitable convenience foods for patients with Refsum’s disease. J Hum Nutr Diet, 6, 295–​305. Budka H, Sluga E, Heiss WD (1976). Spastic paraplegia associated with Addison’s disease: adult variant of adrenoleukodystrophy. J Neurol, 213, 237–​50. Cartier N, et al. (2009). Hematopoietic stem cell gene therapy with a lentiviral vector in X-​linked adrenoleukodystrophy. Science, 326, 818–​23. de Beer M, et al. (2014). Frequent occurrence of cerebral demyelin- ation in adrenomyeloneuropathy. Neurology, 83, 2227–​31. Engelen M, et al. (2014). X-​linked adrenoleukodystrophy in women: a cross-​sectional cohort study. Brain, 137, 693–​706. Engelen M, Kemp S, Poll-​The BT (2014). X-​linked adrenoleukodystrophy: pathogenesis and treatment. Curr Neurol Neurosci Rep, 14, 486. Engelen M, et al. (2015). Bezafibrate for X-​linked adrenoleukodystrophy. Biochim Biophys Acta, 1851, 231–​7. Ferdinandusse S, et al. (2000). Mutations in the gene encoding peroxi- somal alpha-​methylacyl-​CoA racemase cause adult-​onset sensory motor neuropathy. Nat Genet, 24, 188–​91. Ferdinandusse S, et  al. (2010). Adult peroxisomal acyl-​coenzyme A oxidase deficiency with cerebellar and brainstem atrophy. J Neurol Neurosurg Psychiatry, 81, 310–​2. Fiskerstrand T, et al. (2010). Mutations in ABHD12 cause the neuro­ degenerative disease PHARC:  an inborn error of endocannabinoid metabolism. Am J Hum Genet, 87, 410–​17. Goldman JM, et al. (1985). Screening of patients with retinitis pig- mentosa for heredopathia atactica polyneuritiformis (Refsum’s dis- ease). Br Med J (Clin Res Ed), 290, 1109–​10. Gong Y, et al. (2015). Adeno-​associated virus serotype 9-​mediated gene therapy for X-​linked adrenoleukodystrophy. Mol Ther, 23, 824–​34. Griffin JW, et al. (1977). Adrenomyeloneuropathy: a probable variant of adrenoleukodystrophy. Neurology, 27, 1107–​3. Haberfeld W, Spieler F (1910). Zur diffusen Hirn-​Rueckmarksclerose im Kindesalter. Dtch Z Nervenh, 40, 436–​63. Huffnagel IC, et al. (2019). The Natural History of Adrenal Insufficiency in X-Linked Adrenoleukodystrophy: An International Collaboration. J Clin Endocrinol Metab, 104, 118–26. Jangouk P, et al. (2012). Adrenoleukodystrophy in female heterozy- gotes: underrecognized and undertreated. Mol Genet Metab, 105, 180–​5. Jansen GA, Waterham HR, Wanders RJ (2004). Molecular basis of Refsum disease: sequence variations in phytanoyl-​CoA hydroxylase (PHYH) and the PTS2 receptor (PEX7). Hum Mutat, 23, 209–​18. Jansen GA, et al. (1997). Refsum disease is caused by mutations in the phytanoyl-​CoA hydroxylase gene. Nat Genet, 17, 190–​3. Kamat SS, et al. (2015). Immunomodulatory lysophosphatidylserines are regulated by ABHD16A and ABHD12 interplay. Nat Chem Biol, 11, 164–​71. Kemp S, et  al. (2012). X-​linked adrenoleukodystrophy:  clinical, metabolic, genetic and pathophysiological aspects. Biochim Biophys Acta, 1822, 1465–​74. Lamari F, et al. (2013). Disorders of phospholipids, sphingolipids and fatty acids biosynthesis: toward a new category of inherited meta- bolic diseases. J Inherit Metab Dis, 36, 411–​25. Loes DJ, et al. (2003). Analysis of MRI patterns aids prediction of pro- gression in X-​linked adrenoleukodystrophy. Neurology, 61, 369–​74. Mahmood A, et al. (2007). Survival analysis of haematopoietic cell trans- plantation for childhood cerebral X-​linked adrenoleukodystrophy: a comparison study. Lancet Neurol, 6, 687–​92. Mihalik SJ, et al. (1997). Identification of PAHX, a Refsum’s disease gene. Nat Genet, 17, 185–​9. Morita M, Imanaka T (2012). Peroxisomal ABC transporters: structure, function and role in disease. Biochim Biophys Acta, 1822, 1387–​96. Moser AB, et al. (1999). Plasma very long chain fatty acids in 3000 peroxisome disease patients and 29,000 controls. Ann Neurol, 45, 100–​10. Moser HW, et al. (1991). Clinical aspects of adrenoleukodystrophy and adrenomyeloneuropathy. Dev Neurosci, 13, 254–​61. Moser HW, et al. (2005). Follow-​up of 89 asymptomatic patients with adrenoleukodystrophy treated with Lorenzo’s oil. Arch Neurol, 62, 1073–​80. Nishiguchi KM, et al. (2014). Exome sequencing extends the pheno- typic spectrum for ABHD12 mutations: from syndromic to non-​ syndromic retinal degeneration. Ophthalmology, 121, 1620–​7. Odone A, Odone M (1989). Lorenzo’s oil. A  new treatment for adrenoleukodystrophy. J Pediatr Neurosci, 5, 55–​60. Parikh S, et al. (2015). A clinical approach to the diagnosis of patients with leukodystrophies and genetic leukoencephelopathies. Mol Genet Metab, 114, 501–​15. Purdue PE, et al. (1999). Rhizomelic chondrodysplasia punctata, a per- oxisomal biogenesis disorder caused by defects in Pex7p, a peroxisomal protein import receptor: a mini-​review. Neurochem Res, 24, 581–​6. Refsum’s S (1946). Heredopathia atactica polyneuritiformis. Acta Psychiatr Scand, 38 Suppl, 9–​15. Schilder P (1924). Die Encephalitis periaxalis diffusa. Arch Psychiatr Nervenkr, 71, 327–​35. Sedel F, et  al. (2007). Psychiatric manifestations revealing inborn errors of metabolism in adolescents and adults. J Inherit Metab Dis, 30, 631–​41. Sedel F, et al. (2007). Peripheral neuropathy and inborn errors of me- tabolism in adults. J Inherit Metab Dis, 30, 642–​53. Steinberg D, et al. (1967). Refsum’s disease: nature of the enzyme de- fect. Science, 156, 1740–​2. van den Brink DM, et al. (2003). Identification of PEX7 as the second gene involved in Refsum’s disease. Am J Hum Genet, 72, 471–​7. Wanders RJ (2011). Phytanic acid metabolism in health and disease. Biochim Biophys Acta, 1811, 498–​507. Wanders RJ, Ferdinandusse S (2012). Peroxisomes, peroxisomal dis- eases, and the hepatotoxicity induced by peroxisomal metabolites. Curr Drug Metab, 13, 1401–​11. Wanders RJ (2018). Peroxisomal disorders: Improved laboratory diag- nosis, new defects and the complicated route to treatment. Mol Cell Probes, 40, 60–​9.

12.9  Disorders of peroxisomal metabolism in adults 2173 Wierzbicki AS, et al. (2002). Refsum’s disease: a peroxisomal disorder affecting phytanic acid alpha-​oxidation. J Neurochem, 80, 727–​35. Wierzbicki AS, et  al. (2003). Metabolism of phytanic acid and
3-​methyl-​adipic acid excretion in patients with adult Refsum’s dis- ease. J Lipid Res, 44, 1481–​88. Websites Adult Refsum’s Disease website (information for patients, carers and clinicians): http://​refsumdisease.org. United Leukodystrophy Foundation website: http://​www.ulf.org. X-​linked Adrenoleukodystrophy database: http://​www.X-​ald.nl.