eMedicine - Inherited Metabolic Disorders : Article by Pieter R Kark

You are in: eMedicine Specialties > Neurology > Pediatric Neurology

Inherited Metabolic Disorders

Article Last Updated: Dec 8, 2006

The field of inherited disorders of the nervous system has undergone major revolutions in the past 150 years. The 19th century saw the first systemic approach to disease through the use of rational, consistent outlines for taking histories and doing physical examinations (Kark, 1983). Scientific methods were applied to pathology and clinical medicine because of discoveries in physics and chemistry (both organic and inorganic). The discoveries led to knowledge of the organic chemistry of dyes, tissue staining, and improved microscopy (Adams; Bacon and Garrod, 1909; Charcot, 1962). The blood pressure apparatus, thermometer, stethoscope, tuning fork, and later, the reflex hammer were added to the clinician's armamentarium (Kark, 1987). With these tools, physicians and pathologists – often the same person– were able to apply Sir Francis Bacon's dictum to medicine: "inquire carefully into the origin of things" (Clarke, 1996).

Garrod summarized the initial discoveries of the 19th century and the turn of the 20th in his book, Inborn Errors of Metabolism, some 80 years ago. By the mid 1960s, defects that led to the accumulation of metabolic products in the urine, blood, or neural tissues were identified. These defects were largely problems in the catabolism of lipids and amino acids or in the rapid breakdown of glycogen. The identification of the metabolites that accumulated in a disease made possible the identification of the enzyme whose activity was deficient.

Although direct identification of abnormal protein structure still was not possible, such defects could be inferred, indirectly, by demonstrating alterations in the kinetic properties of patients' enzymes (ie, rates of reaction as concentrations of cofactor or substrate were varied) or in the rate at which heating the enzyme altered its catalytic properties. Such changes allowed a strong inference of a change in the structure of the enzyme-protein. A change in structure was in those days thought to be due to a change in an amino acid somewhere in the peptide chain of the protein. In the case of some hemoglobin variants, demonstrating substitutions of one or another amino acid was actually possible.

Over the next 2 or 3 decades, errors in glycolysis, the Krebs cycle, and adjacent pathways were elucidated by such methods, although some of these errors are debated still. By the mid 1980s, techniques largely had switched from those of the biochemistry of intermediates and enzymes to the identification of mutations in genes. This was done by a large number of techniques that make use of DNA fragments (restriction fragment-length polymorphisms) so as to permit linkage mapping and gene sequencing. As a result, we now know many genetic defects responsible for neurological disease, but frequently we do not know much about the resulting protein product and therefore the pathophysiologic basis for the disease (Gomez, 2001).

The human genome at one time was estimated to have 70,000-100,000 genes. New data from the Human Genome Project suggest this number may be closer to 30,000. Many of these appear to code for proteins produced in the brain. Data from the Human Genome Project surely will be useful in identifying mutations in the thousands of genes that must underlie inherited diseases of the central and peripheral nervous system. Genetic data also will be useful in identifying mutations and polymorphisms that predispose to some of the acquired diseases of the nervous system, some of which are discussed in this article. One of the major principles of pathophysiology that has appeared in recent decades is that many "acquired" diseases have one or several genetic bases – predispositions. Another principle is that diseases often appear only after 2 or 3 things go wrong. Some examples of each are discussed in this article.

Until the genes and their mutations that underlie neurological disease are characterized, inherited disorders have to be defined the way clinicians have been classifying disease over the last 2 centuries. These classifications make use of clinical descriptions in the living patient correlated with pathologic changes found at autopsy; with chemical changes in excreta, blood, cerebrospinal fluid (CSF), and tissues, or sometimes with abnormalities found on images of the brain and other organs. The methods of genetic analysis first described by Mendel and the correlations with human disease first collected by Garrod were fundamental to the knowledge of inherited diseases that was acquired in the latter two thirds of the 20th century.

Until the end of the 1970s, only a few methods were available for investigating inherited disease. They included the following:

At that time, however, correlating disease states with amino acid substitutions in specific proteins or peptides was not possible, except for hemoglobin variants: these were well characterized.

Diseases are intellectual constructs, not reality. They are constructs whose definition can change depending on the speaker, the audience, and most importantly the historical era in which the disease is described. Names of diseases are useful ways to tie what is being said to the medical technology and assumptions of the era in which the statement about an illness is being made. However, the description and definition of a disease each can change as technology improves and understanding of the pathophysiological processes evolves.

Even the basic name of a disease can change as decades (and even centuries) pass. This is particularly true in the last century or two. What was "idiocy" in the mid-nineteenth century became subdivided into terms such as "amaurotic idiocy" which later became Tay Sachs' disease," followed by "Tay-Sachs disease," then "generalized gangliosidosis" then "generalized GM2 gangliosidosis,generalized GM2 gangliosidosis, infantile type with McKusick classification 230500," and finally the current disease name, "severe deficiency of hexosaminidase A with infantile onset of symptoms."

The reality of medicine is the sick patient. The definition of a specific disease has less correlation to the real world than the construct of species has in biology. This is true even for genetic diseases, although genetic diseases superficially appear to be linked closely to a specific cause, a specific mutation. That this is not always true is discussed later in the article, under Polymorphism With and Without Disease.

Another critical issue in the definition of disease is that having a genetic defect that may result in a disease is not the same as having the disease. Having the defect means only having a propensity or a risk of developing the disease. Whether the problem is an enzyme deficiency in carbohydrate metabolism or the excessive triplet repeats that characterize the mutation for Huntington chorea does not matter. The biochemical or genetic abnormality is no more a definition of a disease state than a positive result on a purified–protein-derivative (PPD) skin test is for tuberculosis. The issue is not an academic one. In view of the desire of insurance carriers to avoid "pre-existent illnesses," the issue of when someone gets a disease is currently a matter of intense ethical and legal concern. One gets a disease when one has symptoms – not merely signs – that point to the disease, and not before that time.

Watson and Crick's work, and the work of both their acknowledged and unacknowledged collaborators in the 1950s, led to the unraveling of the genetic code, the recognition that the code of both DNA and RNA had to be read from a specific end of the macromolecule, the understanding that both DNA and RNA sequences are decoded in triplets, and the recognition that a triplet codes for only one amino acid. These triplets are called "codons." (Later work showed that some codons do not code for an amino acid. Some mark the beginning or end of a peptide chain.) Molecular biologists do not know the function of some long sequences of codons called "introns." Once we understood clearly how DNA codes for a protein, we recognized that an amino acid substitution in a protein must result from a change in the mRNA triplet as a result of a mutation in the corresponding DNA – the gene.

At first, clinical and biochemical investigations led to recognition of the consequences of altered protein products and to an understanding of how these might produce disease. The new molecular techniques led to a "reverse" approach. Alterations in DNA were traced to RNA changes. Science is only now on the verge of explaining some of these as changes in proteins (Gomez, 2001). This approach from DNA back flourished in the 1980s and 1990s as various methods were developed that showed alterations in the structure of nucleic acid chains independent of other aspects of biochemistry or biology.

One technique, called restriction fragment length polymorphism (RFLP) analysis, includes cutting DNA into fragments by using one or more bacterial enzymes, each with a specific nucleic acid recognition site that directs where the DNA is to be cut. The lengths of the fragments then are measured on the basis of their migration within a gel when exposed to an electrical current. Another technique, polymerase chain reaction (PCR), converts traces of DNA into large amounts of identical material. Scientists can analyze the genetic material once it has been magnified this way. These techniques have totally changed research on inherited diseases. They are similar to the methods society currently uses to identify each of us as individuals (by DNA fingerprinting) and to infer how we and other species may have evolved (by analyses of nuclear and mitochondrial DNA in individuals, even individuals dead for millions of years, and in species and in populations of species).

Genetic mutations have been presumed to be the basis of inherited disease since the time of Mendel and Garrod. Even now, some diseases that have been recognized for 100-150 years have known genetic defects but the protein products are not well characterized. Chediak-Higashi syndrome is an example of such an inherited metabolic defect; Huntington chorea a much better-known example in inherited diseases of the CNS. Many genetic diseases of the nervous system can be diagnosed accurately by DNA analysis, and the pattern of inheritance can be demonstrated within families. The molecular tools sometimes allow us predict who is likely to develop the disease and who in the family can neither develop it nor pass it on to offspring. With recent advances in enzyme replacement therapy and gene therapy, we may someday be able to treat or perhaps even prevent some of these disorders even if we do not know how the genetic change gives rise to the disease.

Metabolism is the physiological and biochemical mechanisms by which foodstuffs are taken in by the body and converted from one form to another to provide energy for all the activities of the body. Metabolism includes the methods our cells use to build multitudes of specific molecules that the body uses for its myriad activities. Some of these are small hormones or neurotransmitters, others are large enzymes or constituents of cell structures that, with several long chains of lipid, make a key part of a membrane or, with long chains of several peptides, make up a single functional protein. Metabolism includes not only the mechanisms for building molecules, but also the degradation processes that enable cells to excrete waste products. In its broadest sense, metabolism encompasses virtually every biochemical pathway and biophysical mechanism in the body and the resultant physiologic activities.

In a more limited sense, inborn errors of metabolism can be defined as disorders of the mechanisms by which specific major foodstuffs are converted to energy or cellular and tissue building blocks and final products and the mechanisms by which foodstuffs and products are degraded to be excreted. These include mechanisms involving absorption and modification of vitamins and minerals; mechanisms for degrading molecules to provide energy or to be excreted; mechanisms for making acetyl-coenzyme A, nonessential amino acids, cholesterol, long-chain fatty acids, prostaglandins, and the complex lipids they lead to; mechanisms for making the proteins that are the structure of cells, inside and out, and that are the prime catalysts of cellular chemistry, enzymes; and mechanisms for neutralizing molecules that represent potential environmental toxins.

Biochemical and biophysical processes that are related closely but are not included in the term "metabolism" include the following:

The distinction between these processes and what traditionally has been considered metabolism is arbitrary since they are clearly interdependent cellular functions. Ingesting a macromolecule is not considered part of metabolism but breaking it down in the resultant lysosome is. The ever-expanding knowledge of the genetic basis of neurological disease will probably blur the distinction more. The distinction initially was defined on the basis of a limited understanding of inborn errors from a physiological rather than molecular perspective.

Inherited diseases affect virtually all parts of the nervous system. Many of these disorders also exist in sporadic forms that do not seem to have a primary genetic cause. In time, clinicians may decide to rename these conditions so that the genetic and nongenetic forms are differentiated clearly. Examples of conditions that come in genetically determined and "nongenetic" types include dementias of the subtypes Alzheimer, Pick, frontotemporal, Parkinson disease, amyotrophic lateral sclerosis, and peripheral neuropathies. For each of these conditions, a small percentage of affected patients exhibit a clearly genetically determined cause and have pedigrees that demonstrate a classical Mendelian pattern of inheritance. For each condition, however, the majority of patients appear to have an acquired defect, without evidence of a genetic predisposition. As time goes on, additional genetic factors may be found that predispose to development of disease in even apparent sporadic cases.

Scientists now speculate that, if such gene alterations exist, they may be expressed because of epigenetic factors, modifier genes, or environmental influences.

In the strictest sense, inherited disorders of metabolism encompass a narrow spectrum of conditions that have been defined on a biochemical basis. Broad categories include disorders of carbohydrate metabolism, disorders of amino acid metabolism, organic acidemias, lysosomal storage diseases, disorders of fatty acid metabolism, and mitochondrial disorders. Most but not all of these conditions are associated with some neurological sequelae.

Another useful way to categorize inborn errors of metabolism is by the neurological subsystems most prominently affected. These are listed in Table 1. In addition to those listed here, some metabolic disorders produce acute changes in behavior or in the function of the forebrain, while others give rise to specific clinical features such as frontal bossing (eg, the mucopolysaccharidoses) or specific skin lesions (eg, Fabry disease, Refsum disease, ataxia telangiectasia). A few metabolic disorders primarily affect the liver, spleen, or heart and may be detected early by changes in these organs. The nervous system is affected late.

These particular features are dealt with specifically in the relevant articles of eMedicine. Readers also may refer to various modern textbooks of child neurology and neurology, such as Clarke's useful monograph Clinical Features of the Inherited Metabolic Disorders of the Nervous System, Rosenberg et al's The Molecular and Genetic Basis of Neurological Disease, and Scriver et al's The Metabolic and Molecular Basis of Inherited Disease.

Table 1. Typical Neurological Syndromes Associated with Known Inherited Metabolic Disorders

Syndrome	Inherited Metabolic Defects
Psychosis, irritability, mood disorder, hyperactivity, agitation, or hallucinations	Absorption of vitamin B-12, lysosomal defects (eg, Sanfilippo and Hunter diseases, GM2 gangliosidosis of late onset, neuronal ceroid lipofuscinosis, metachromatic leukodystrophy of late onset, Krabbe disease), peroxisomal defects (ie, adrenoleukodystrophy), Lesch-Nyhan syndrome (ie, defect of purine metabolism), Wilson disease (especially young onset), acute intermittent porphyria and other hepatic porphyrias, defects of the urea cycle, homocystinuria, cerebrotendinous xanthomatosis
Global mental retardation/developmental delay with progressive neurological signs; dementias	Lysosomal defects (early onset causes mental retardation, late onset causes dementia); disorders of amino acids, organic acids, carbohydrates (especially of pyruvate metabolism), copper metabolism (eg, Menkes disease, some cases of Wilson disease), peroxisomes (eg, adrenoleukodystrophy, Zellweger syndrome)
Seizures	Those of the disorders already mentioned that affect gray matter, especially peroxisomal defects (other than Refsum disease) and some lysosomal defects, pyridoxine-dependent defects (eg, glutamate decarboxylase deficiency); biotin-related defects (eg, biotinidase deficiency, branched chain amino acid defects); molybdenum cofactor deficiency; defects of pyruvate metabolism; defects of mitochondrial electron transport
Loss of vision	Those listed with seizures above
Optic atrophy	Leber hereditary optic atrophy (any of the several mitochondrial DNA mutations)
Retinitis pigmentosa	Defects of pyruvate metabolism, defects of mitochondrial electron transport chain, Refsum disease, abetalipoproteinemia, Hallervorden-Spatz disease
Cherry red spot (in macula)	Tay-Sach disease, sialidosis II (ie, lipomucopolysaccharidosis)
Extrapyramidal disorders (especially with rigidity, tremor, or chorea)	Wilson disease, Lesch-Nyhan syndrome, Segawa syndrome, some organic acid defects, defects of RBC glycolysis (eg, triosephosphate isomerase)
Ataxias	Defects of pyruvate metabolism, defects of amino acid metabolism (eg, Hartnup disease, maple syrup urine disease), defects of organic acid metabolism, defects of mitochondrial electron transport, defects of the urea cycle, lipoprotein disorders (abetalipoproteinemia, alpha-lipoproteinemia), peroxisomal defects (eg, Refsum disease), lysosomal disorders with predominantly gray matter effects
Pyramidal and cerebellar syndrome with or without peripheral neuropathy (white matter disease)	Peroxisomal (eg, adrenoleukodystrophy and its late onset variants, Zellweger syndrome, late stages of Refsum disease), Canavan disease (ie, aspartoacylase deficiency), Alexander disease (similar to Canavan disease, but biochemical defect not known), lysosomal defects (especially late onset variants)
Peripheral neuropathy with or without dysautonomia	Acute intermittent porphyria, peroxisome assembly defect, some cases of Refsum disease, some cases of abetalipoproteinemia, familial amyloidosis (not a metabolic defect in the strict sense cited above)
Myopathy	-
Episodic cramps and myoglobinuria related to exercise	Defects of glycolysis and of glycogenolysis; defects of oxidation of fatty acids, defects of carnitine and its derivatives
Ragged red diseases (ie, limb-girdle neuromuscular diseases with ragged red fibers seen on trichrome stain of frozen biopsied specimens)	Defects of mitochondria, especially of the electron transport chain and of proteins coded by mitochondrial DNA

Most inborn errors of metabolism are inherited as autosomal recessive conditions. Some are due to mutations on the X chromosome and follow an X-linked recessive genetic pattern. Some mitochondrial disorders are due to proteins that are transported into mitochondria and function there, but that are coded for by ordinary nuclear DNA. These follow an autosomal recessive pattern. Many mitochondrial disorders have a unique form of inheritance with only maternal transmission. The mitochondrial DNA (which is circular, like that of a bacterium) all comes from the egg and hence from the mother. None of the mitochondria in the sperm is passed on to the zygote.

Autosomal conditions generally affect equal numbers of males and females. X-linked recessive conditions generally affect only males. These males may be related through unaffected carrier females. In some conditions the differential expression of an autosomal gene can lead to one gender having more severe symptoms than the other. For some X-linked conditions, carrier females may have symptoms and signs that are considerably milder than those in affected males.

Many inherited diseases were first described in one race or ethnic group. Examples are sickle cell disease among black Africans, Tay-Sachs, familial dysautonomia, and Canavan disease in Ashkenazi Jews, cystic fibrosis in northern Europeans, and the inherited predisposition to multiple sclerosis (MS) in descendants of the Vikings. Over time, however, most inborn errors of metabolism have been found to occur in almost all races and groups that have been studied. Many populations are now characterized by admixture of various gene pools. Spontaneous mutations can occur in any person within a population. The occurrence of a defect in a population can be influenced by "founder effect."

The association of some inherited disorders of metabolism with one race, especially with Ashkenazi Jews, has turned out in the past 30 years to be, in part, an artifact of medical history. Between 1945 and 1970, more research on these disorders was carried out in the northeastern United States than elsewhere in the world. With little effort needed to recover from World War II in the United States, medical research was highly regarded, encouraged, and funded. This research had a strong technical basis, using key new biochemical techniques. Many of the bright doctors doing the research were in the largest American city, New York. Many were Ashkenazi, and so were many of their patients.

Only when countries in the rest of the world became able to afford similar equipment and similar research did we learn that many inherited disorders of metabolism are very widespread, even though rare in each population. The known incidence of metabolic disorders is often higher in one population than another, but few diseases are confined to a single race.

Of the few inherited disorders of intermediary metabolism still known to occur exclusively in one race, pentosuria, the best example, appears to be unique to Ashkenazi Jews. It is not really a disease: it has no symptoms, and affected persons do not suffer. Instead, it is a biochemical curiosity with no known ill effects. The phenomenon is exceedingly rare, but it results from any one of 3 known mutations. (That many mutations in so rare and, seemingly, clinically insignificant a change suggest to the author that the disorder may give an evolutionary disadvantage to humans who carry the mutation.)

Some inherited metabolic disorders are fatal in the first weeks or months of postnatal life, for example, severe defects in the conversion of pyruvate to acetyl coenzyme A (CoA), some urea cycle defects, and severe defects in the processing of fructose. Others are compatible with a very long life, for example, nonneuronopathic Gaucher disease, McArdle disease, and phenylketonuria (if treated with dietary restriction of phenylalanine). Many inborn errors of metabolism may exist that are entirely incompatible with life and that never result in a live-born infant. As advances occur in enzyme replacement therapy and, ultimately, we hope, in gene therapy, some inborn errors that cannot be treated today may become treatable in the future.

In the 1960s, inherited defects of the Krebs cycle were believed to be so deleterious that they would not be compatible with postnatal life. However, John Blass identified inherited abnormalities of the pyruvate dehydrogenase multi-enzyme complex in infants and children with neurological problems. Since then, geneticists and other physicians have hesitated to presume that any particular metabolic defect would necessarily or invariably be fatal.

The biochemical methods of the 1920s through the 1980s led to discoveries of defects in most of the known metabolic pathways. Some of the defects lead to neurological disorders, others to diseases confined to RBCs, liver, kidneys, or other organs. Most inborn errors have multisystem effects. For example, defects of pyruvate oxidation can affect the retina, the heart, and bone as well as the nervous system. Sickle cell anemia leads secondarily to strokes, as do defects of metabolism of the amino acid homocysteine. Other articles in eMedicine discuss specific inherited defects of amino acids, lipids, mucopolysaccharides, carbohydrate metabolism, and the breakdown of large molecules within lysosomes and peroxisomes.

Garrod's work in the 1920s led to the notion that a single genetic disease was likely due to a single inherited defect. This hypothesis led to fruitful work through the 1960s. By the 1970s, however, the complexity of genetic disorders became an issue in research. The first defects of amino acid metabolism that were elucidated followed the Garrod rule. These included phenylketonuria, maple syrup urine disease, homocystinuria as it first presented, and the first disorders of lysosomal metabolism. GM2 gangliosidosis led to Tay-Sachs disease, and specific defects in the breakdown of glycosaminoglycans (ie, mucopolysaccharidoses III) caused Sanfilippo disease.

Defects of other lysosomal enzymes led to the clinically different pictures of Hurler syndrome, Hunter syndrome, Morquio syndrome, and others. Since the mucopolysaccharides are degraded through an entirely different pathway than the globosides, the Garrod rule still held, despite the fact that lysosomes were involved in each case. Refsum disease, caused by lack of the enzyme catalyzing alpha-oxidation of the fatty acid phytanic acid, was the first peroxisomal defect to be discovered. This disease was distinct clinically from the next peroxisomal defect identified, adrenoleukodystrophy.

The defects of the enzymes that convert sugars to energy seemed to follow suit. Phosphorylase deficiency was a disorder of the muscles. Deficiency of fructose 1, 6-diphosphate dehydrogenase led to a disease of the brain and the liver. A single amino acid substitution can come about easily if one DNA base is changed in one codon at a specific site within the gene for that enzyme. Such a mutation is called a point mutation. (The deletion of a single base is also a point mutation but would result in a shift of the triplet reading frame and likely the production of a truncated protein with little or no activity.) By the late 1960s, many of the mutations of hemoglobin were known to be point mutations, and indirect evidence existed that this also was true for those few enzymes whose mutations could be studied in fine detail.

In the period from the 1960s to the early 1980s, research on the molecular basis of inherited disorders of metabolism assumed that these were largely due to a point mutation in the peptide chain of the relevant enzyme. A substitution in or near the active center of the enzyme seemed a ready explanation for a change in the binding of substrate or product and thus to a change in the enzyme's catalytic ability. The center is the catalytic cleft or region in the natural state of the folded protein, the region to which substrate(s) bind for catalysis to take place.

With the discovery of defects of oxidation of pyruvate to acetyl CoA in the late 1960s and early 1970s, the Garrod rule seemed no longer to be universally true. A defect of a certain severity in a specific pathway might be associated with a specific disease entity with a well-defined clinical picture, whereas defects of lesser severity in the same pathway or even in the same enzyme could sometimes be associated with different, less severe clinical entities. Thus, a severe defect of the pyruvate dehydrogenase complex often is associated with overwhelming lactic acidosis, mental deficiency, and severe hypotonia in infants; a more moderate defect is associated with ataxia and episodes of transient lactic acidosis in young children; even milder defects have been associated clearly with the onset of ataxia, areflexia, sensory loss, and abnormalities of heart and bone in adolescents.

Even in the middle of the 20th century, the details of the autosomal dominant neurological diseases did not really fit the Garrod rule. For example, severity of both Huntington chorea and myotonic dystrophy seemed to vary depending on whether the carrier was the father or the mother. Earlier onset or juvenile Huntington disease, most commonly seen in adults of middle age, often was seen in the children who inherited the gene from their father, whereas the congenital form of myotonic dystrophy was seen mostly in children born to affected mothers. Influences in the womb alone—sometimes even called "miasmas"—were postulated to account for these discrepancies. Until the actual mutation was established, those who did not make the initial observation tended to disregard it. People also disregarded and even scoffed at the observation that these 2 conditions had a tendency to become worse and worse in successive generations. This tendency is the phenomenon of "anticipation."

Thus, a grandfather may have developed symptoms of Huntington disease in his 50s; his children may have shown clear signs in their late 30s; and the grandchildren may have become symptomatic in their 20s. When this phenomenon first was described, clinicians incorrectly tried to attribute the earlier diagnosis of disease to improving clinical skills or ascertainment bias. Once the genetic basis was known – increasing numbers of triplet repeats in the mutant gene from generation to generation—anticipation "became" an established clinical fact in these diseases and in a number of other neurogenetic diseases that are due to triplicate codon repeats at the terminal end of the gene.

The mutations that lead to Huntington disease, myotonic dystrophy, Machado-Joseph disease, and a large number of other dominantly inherited neurological diseases are not point mutations after all. The mutations are in the number of triplets of cytosine--adenosine--guanine (CAG) at one end of the gene. In the case of Huntington disease, the normal allele has 18-20 CAG triplets. If more than 30 CAG triplets are present, the individual is a carrier of Huntington disease and has an approximately 90% chance of getting the disease if he or she lives to the age of 60 years or more.

The longer the chain of triplet repeats, the earlier the person is likely to have clinical signs and the more severe and rapid the clinical course is likely to be. The longer the stretch of triplet repeats, the more unstable this region becomes during meiosis and the more likely this stretch will expand even further in the next generation. In the case of HD, the gene from the father is more likely to result in very long chains of CAG and so in juvenile onset of HD than that transmitted through the mother. For myotonic dystrophy, the converse is true: the maternal gene is more likely to acquire excessive lengths of CAG.

While we think of patients with autosomal recessive disorders like Tay-Sachs or defects of pyruvate dehydroxygenase as being homozygous for a mutant gene, many are really compound heterozygotes. Each parent may carry a mutant allele, but the mutations may not be the same in both parents. The effect in the offspring still may be insufficient activity of the relevant enzyme, resulting in typical disease. The offspring is a compound heterozygote, not a true homozygote. This becomes important when genetic markers are used that detect one form of mutant gene but not other, less common mutations. A less common allele may be missed, giving the impression that one parent is not a carrier, or that the (compound heterozygote) subject is merely a carrier, not affected with the disease.

Even in the 1960s, some amino acid substitutions that were clinically insignificant were recognized in hemoglobin molecules. These variant hemoglobin molecules migrated at an abnormal rate on the chromatographic papers (and in later years in gels) that were used to find such molecular changes. Instead of having clinically identifiable problems such as sickle cell disease, people with these harmless substitutions were healthy and lived healthy lives. Such innocent genetic changes were called polymorphisms. Similar harmless polymorphisms were discovered in other proteins as soon as analyses of their substructures became easy and work could be done with samples from various populations.

Diseases occurred only if the amino acid substitution in a particular protein or peptide was of a particular kind and at a particular place in the peptide chain. The polymorphism had to interfere with function and it had to interfere to such a degree that the body could not compensate readily. Only then would the gene change lead to disease.

The fact that the mutation leading to Wilson disease would be a harmless polymorphism on a planet with little or no copper was pointed out in the 1970s and 1980s. Refsum disease would not exist in a society that did not consume phytanic acid. Favism would not be recognized if fava beans did not exist. Hypokalemic periodic paralysis would not manifest itself if an affected individual never consumed a meal high in sugar. Individuals with mild forms of fructose intolerance would not have symptoms if they never ate foods containing fructose.

The relation between polymorphism and disease can be even more complex. In a number of instances, two or more things must go wrong for a disease to appear. Refsum disease often presents in children aged 7-12 years with acute onset of nerve deafness or blindness, with neuropathy or ataxia. A large excess of phytanic acid is present in tissues or body fluids. An exogenous disease or stressor precipitates episodes. Documented examples include appendicitis, a severe viral illness, surgery, or a fracture. Each episode is self-limited with partial resolution over weeks to months. However, the disease is progressive. The residual effects accumulate in adolescence and early adult life, until the disease takes on a progressively downhill course rather than the remitting and relapsing course of childhood.

The defect is present from conception: an inability to oxidize phytanic acid at the alpha position. (Phytanic acid is a breakdown product of phytol, a major component of chlorophyll.) The disease appears only after an exogenous illness. As such, Refsum disease is analogous to multiple sclerosis (MS). Blass and Steinberg pointed out a further analogy when they suggested that Refsum may represent an ideal scientific model for MS (D. Steinberg and J. P. Blass, personal communications, 1969-1971). The biochemical defect underlying Refsum disease appears to have been propagated by the western Vikings, just as the genetic defect predisposing to MS (in or near the genes for the human leukocyte antigen complex) was spread by both the eastern and the western Vikings to European populations. Often a marked improvement in the signs and symptoms of Refsum disease can be demonstrated when prevention of exogenous illness is combined with dietary restriction of phytol and phytanic acid. Not unexpectedly, eating a diet

Unfortunately, for MS no recognized dietary or preventive measures exist, and only certain drugs related to immune phenomena—such as the ABC drugs—can decrease the risk of a relapse of MS.

The signs and symptoms of acute intermittent porphyria probably result from the accumulation of 2 neurotoxic metabolites in the pathway of porphyrin synthesis, delta-amino levulinic acid and porphobilinogen (see Porphyrias). The clinical disease may be associated with attacks of sensory neuropathy, ataxia, psychosis, and even coma, and occurs when a patient with the polymorphism or genetic predisposition ingests a drug or eats food that precipitously increases the synthesis of porphyrins. These precipitants include phenytoin, strawberries, or any one of the long list of potentially harmful substances outlined in the article on porphyrias. With a sudden demand on this metabolic pathway and up-regulation of enzymes to synthesize porphyrins combined with the inherited partial deficiency of uroporphyrinogen I synthetase, the toxic metabolites accumulate and cause damage to neural tissues, resulting in clinical symptoms.

Mercury poisoning typically produces clinically disabling symptoms only in approximately 10% of a population uniformly exposed to low but "toxic" doses of the metal. Some patients with frank mercury poisoning have pes cavus or mild kyphoscoliosis associated with their neuropathies. This observation led Raymond Adams to ask whether an inherited defect might be present in these patients, a defect that, were it a little more severe, might lead to an inherited neuropathy or an inherited myelopathy (R. D. Adams, personal communication, 1969). In the end, remember what Sir Francis Bacon said: "Inquire carefully into the origin of things."

Inherited Metabolic Disorders

AUTHOR AND EDITOR INFORMATION

INTRODUCTION

CLINICAL FEATURES AND DIFFERENTIAL DIAGNOSIS

EPIDEMIOLOGY AND STATISTICS

MORBIDITY AND MORTALITY

AFFECTED ENZYMES AND PATHWAYS

STRAIGHTFORWARD DISEASES, ESPECIALLY RECESSIVELY INHERITED

COMPLEX DISEASES, ESPECIALLY AUTOSOMAL DOMINANT

POLYMORPHISM WITH AND WITHOUT DISEASE

REFERENCES