Autosomal Dominant Disorders

Autosomal dominant disorders are manifested in the heterozygous state, so at least one parent of an index case is usually affected; both males and females are affected, and both can transmit the condition. When an affected person marries an unaffected one, every child has one chance in two of having the disease. In addition to these basic rules, autosomal dominant conditions are characterized by the following:

The biochemical mechanisms of autosomal dominant disorders are best considered in the context of the nature of the mutation and the type of protein affected. Most mutations lead to the reduced production of a gene product or give rise to an inactive protein. The effect of such loss-of-function mutations depends on the nature of the protein affected. If the mutation affects an enzyme protein the heterozygotes are usually normal. Because up to 50% loss of enzyme activity can be compensated for, mutation in genes that encode enzymes do not manifest an autosomal dominant pattern of inheritance. By contrast, two major categories of nonenzyme proteins are affected in autosomal dominant disorders:

Less common than loss-of-function mutations are gain-of-function mutations. As the name indicates, in this type of mutation the protein product of the mutant allele acquires new properties not normally associated with the wild-type protein. The transmission of disorders produced by gain-of-function mutations is almost always autosomal dominant, as illustrated by Huntington disease (Chapter 28). In this disease the trinucleotide-repeat mutation affecting the Huntington gene (see later) gives rise to an abnormal protein, called huntingtin, that is toxic to neurons, and hence even heterozygotes develop a neurologic deficit.

To summarize, two types of mutations and two categories of proteins are involved in the pathogenesis of autosomal dominant diseases. The more common loss-of-function mutations affect regulatory proteins and subunits of multimeric proteins, the latter acting through a dominant-negative effect. Gain-of-function mutations are less common; they often endow normal proteins with toxic properties, or more rarely increase a normal activity (e.g., activating mutation in the erythropoetin receptor associated with a pathologic increase in red cell production).

Table 5-1 lists common autosomal dominant disorders. Many are discussed more logically in other chapters. A few conditions not considered elsewhere are discussed later in this chapter to illustrate important principles.

TABLE 5-1 Autosomal Dominant Disorders

* Discussed in this chapter. Other disorders listed are discussed in appropriate chapters in the book.

Autosomal Recessive Disorders

Autosomal recessive traits make up the largest category of mendelian disorders. Because autosomal recessive disorders result only when both alleles at a given gene locus are mutated, such disorders are characterized by the following features: (1) The trait does not usually affect the parents of the affected individual, but siblings may show the disease; (2) siblings have one chance in four of having the trait (i.e., the recurrence risk is 25% for each birth); and (3) if the mutant gene occurs with a low frequency in the population, there is a strong likelihood that the affected individual (proband) is the product of a consanguineous marriage. The following features generally apply to most autosomal recessive disorders and distinguish them from autosomal dominant diseases:

Autosomal recessive disorders include almost all inborn errors of metabolism. The various consequences of enzyme deficiencies are discussed later. The more common of these conditions are listed in Table 5-2. Most are presented elsewhere; a few prototypes are discussed later in this chapter.

TABLE 5-2 Autosomal Recessive Disorders

* Discussed in this chapter. Many others are discussed elsewhere in the text.

X-Linked Disorders

All sex-linked disorders are X-linked, and almost all are recessive. Several genes are located in the “male-specific region of Y”; all of these are related to spermatogenesis.¹³ Males with mutations affecting the Y-linked genes are usually infertile, and hence there is no Y-linked inheritance. As discussed later, a few additional genes with homologues on the X chromosome have been mapped to the Y chromosome, but no disorders resulting from mutations in such genes have been described.

X-linked recessive inheritance accounts for a small number of well-defined clinical conditions. The Y chromosome, for the most part, is not homologous to the X, and so mutant genes on the X do not have corresponding alleles on the Y. Thus, the male is said to be hemizygous for X-linked mutant genes, so these disorders are expressed in the male. Other features that characterize these disorders are as follows:

TABLE 5-3 X-Linked Recessive Disorders

* Discussed in this chapter. Others are discussed in appropriate chapters in the text.

There are only a few X-linked dominant conditions. They are caused by dominant disease-associated alleles on the X chromosome. These disorders are transmitted by an affected heterozygous female to half her sons and half her daughters and by an affected male parent to all his daughters but none of his sons, if the female parent is unaffected. Vitamin D–resistant rickets is an example of this type of inheritance.

Enzyme Defects and Their Consequences

Mutations may result in the synthesis of a defective enzyme with reduced activity or in a reduced amount of a normal enzyme. In either case, the consequence is a metabolic block. Figure 5-6 provides an example of an enzyme reaction in which the substrate is converted by intracellular enzymes, denoted as 1, 2, and 3, into an end product through intermediates 1 and 2. In this model the final product exerts feedback control on enzyme 1. A minor pathway producing small quantities of M1 and M2 also exists. The biochemical consequences of an enzyme defect in such a reaction may lead to three major consequences:

FIGURE 5-6 A possible metabolic pathway in which a substrate is converted to an end product by a series of enzyme reactions. M1, M2, products of a minor pathway.

Defects in Receptors and Transport Systems

Many biologically active substances have to be actively transported across the cell membrane. This transport is generally achieved by one of two mechanisms—through receptormediated endocytosis or by a transport protein. A genetic defect in a receptor-mediated transport system is exemplified by familial hypercholesterolemia, in which reduced synthesis or function of LDL receptors leads to defective transport of LDL into the cells and secondarily to excessive cholesterol synthesis by complex intermediary mechanisms. In cystic fibrosis the transport system for chloride ions in exocrine glands, sweat ducts, lungs, and pancreas is defective. By mechanisms not fully understood, impaired chloride transport leads to serious injury to the lungs and pancreas (Chapter 10).

Alterations in Structure, Function, or Quantity of Nonenzyme Proteins

Genetic defects resulting in alterations of nonenzyme proteins often have widespread secondary effects, as exemplified by sickle cell disease. The hemoglobinopathies, sickle cell disease being one, all of which are characterized by defects in the structure of the globin molecule, best exemplify this category. In contrast to the hemoglobinopathies, the thalassemias result from mutations in globin genes that affect the amount of globin chains synthesized. Thalassemias are associated with reduced amounts of structurally normal α-globin or β-globin chains (Chapter 14). Other examples of genetically defective structural proteins include collagen, spectrin, and dystrophin, giving rise to osteogenesis imperfecta (Chapter 26), heredit-ary spherocytosis (Chapter 14), and muscular dystrophies (Chapter 27), respectively.

Genetically Determined Adverse Reactions to Drugs

Certain genetically determined enzyme deficiencies are unmasked only after exposure of the affected individual to certain drugs. This special area of genetics, called pharmacogenetics, is of considerable clinical importance.¹⁴ The classic example of drug-induced injury in the genetically susceptible individual is associated with a deficiency of the enzyme G6PD. Under normal conditions glucose-6 phosphatedehydrogenase (G6PD) deficiency does not result in disease, but on administration, for example, of the antimalarial drug primaquine, a severe hemolytic anemia results (Chapter 14). In recent years an increasing number of polymorphisms of genes encoding drug-metabolizing enzymes, transporters, and receptors are being identified. In some cases these genetic factors have major impact on drug sensitivity and adverse reactions. It is expected that advances in pharmacogenetics will lead to patient-tailored therapy, or “personalized medicine.”

With this overview of the biochemical basis of single-gene disorders, we now consider selected examples grouped according to the underlying defect.

Marfan Syndrome

Marfan syndrome is a disorder of connective tissues, manifested principally by changes in the skeleton, eyes, and cardiovascular system.¹⁵ Its prevalence is estimated to be 1 in 5000. Approximately 70% to 85% of cases are familial and transmitted by autosomal dominant inheritance. The remainder are sporadic and arise from new mutations.

Ehlers-Danlos Syndromes (EDS)

EDSs comprise a clinically and genetically heterogeneous group of disorders that result from some defect in the synthesis or structure of fibrillar collagen. Other disorders resulting from mutations affecting collagen synthesis include osteogenesis imperfecta (Chapter 26), Alport syndrome (Chapter 20), and epidermolysis bullosa (Chapter 25).

Biosynthesis of collagen is a complex process that can be disturbed by genetic errors that may affect any one of the numerous structural collagen genes or enzymes necessary for post-transcriptional modifications of collagen. Hence, the mode of inheritance of EDS encompasses all three mendelian patterns. On the basis of clinical and molecular characteristics, six variants of EDS have been recognized.¹⁷ These are listed in Table 5-5. It is beyond the scope of this book to discuss each variant individually; instead, we first summarize the important clinical features that are common to most variants and then correlate some of the clinical manifestations with the underlying molecular defects in collagen synthesis or structure.

TABLE 5-5 Classification of Ehlers-Danlos Syndromes

As might be expected, tissues rich in collagen, such as skin, ligaments, and joints, are frequently involved in most variants of EDS. Because the abnormal collagen fibers lack adequate tensile strength, skin is hyperextensible, and the joints are hypermobile. These features permit grotesque contortions, such as bending the thumb backward to touch the forearm and bending the knee forward to create almost a right angle. It is believed that most contortionists have one of the EDSs. A predisposition to joint dislocation, however, is one of the prices paid for this virtuosity. The skin is extraordinarily stretchable, extremely fragile, and vulnerable to trauma. Minor injuries produce gaping defects, and surgical repair or intervention is accomplished with great difficulty because of the lack of normal tensile strength. The basic defect in connective tissue may lead to serious internal complications. These include rupture of the colon and large arteries (vascular EDS), ocular fragility with rupture of cornea and retinal detachment (kyphoscoliosis EDS), and diaphragmatic hernia (classical EDS).

The biochemical and molecular bases of these abnormalities are known in several forms of EDS. These are described briefly, because they offer some insights into the perplexing clinical heterogeneity of EDS. Perhaps the best characterized is the kyphoscoliosis type, the most common autosomal recessive form of EDS. It results from mutations in the gene encoding lysyl hydroxylase, an enzyme necessary for hydroxylation of lysine residues during collagen synthesis.¹⁸ Affected patients have markedly reduced levels of this enzyme. Because hydroxylysine is essential for the cross-linking of collagen fibers, a deficiency of lysyl hydroxylase results in the synthesis of collagen that lacks normal structural stability.

The vascular type of EDS results from abnormalities of type III collagen.¹⁹ This form is genetically heterogeneous, because at least three distinct types of mutations affecting the COL3A1 gene encoding collagen type III can give rise to this variant. Some affect the rate of synthesis of pro α1 (III) chains, others affect the secretion of type III procollagen, and still others lead to the synthesis of structurally abnormal type III collagen. Some mutant alleles behave as dominant negatives (see discussion under “Autosomal Dominant Disorders”) and thus produce severe phenotypic effects. These molecular studies provide a rational basis for the pattern of transmission and clinical features that are characteristic of this variant. First, because vascular-type EDS results from mutations involving a structural protein (rather than an enzyme protein), an autosomal dominant pattern of inheritance would be expected. Second, because blood vessels and intestines are known to be rich in collagen type III, an abnormality of this collagen is consistent with severe defects (e.g., spontaneous rupture) in these organs.

In two forms of EDS—arthrochalasia type and dermatosparaxis type—the fundamental defect is in the conversion of type I procollagen to collagen. This step in collagen synthesis involves cleavage of noncollagen peptides at the N terminus and C terminus of the procollagen molecule. This is accomplished by N terminal–specific and C terminal–specific peptidases. The defect in the conversion of procollagen to collagen in the arthrochalasia type has been traced to mutations that affect one of the two type I collagen genes, COL1A1 and COL1A2. As a result, structurally abnormal pro α1 (I) or pro α2 (I) chains that resist cleavage of N-terminal peptides are formed. In patients with a single mutant allele, only 50% of the type I collagen chains are abnormal, but because these chains interfere with the formation of normal collagen helices, heterozygotes manifest the disease. By contrast, the related dermatosparaxis type is caused by mutations in the procollagen-N-peptidase genes, essential for the cleavage of collagens. In this case the enzyme deficiency leads to an autosomal recessive form of inheritance.

Finally, the classical type of EDS is worthy of brief mention, since molecular analysis of this variant suggests that genes other than collagen genes may be involved in the pathogenesis of EDS. In 30% to 50% of these cases, mutations in the genes for type V collagen (COL5A1 and COL5A2) have been detected.²⁰ Surprisingly, despite a phenotype typical of EDS, no other collagen gene abnormalities have been found in the remaining cases.

To summarize, the common thread in EDS is some abnormality of collagen. These disorders, however, are extremely heterogeneous. At the molecular level, a variety of defects, varying from mutations involving structural genes for collagen to those involving enzymes that are responsible for post-transcriptional modifications of mRNA, have been detected. Such molecular heterogeneity results in the expression of EDS as a clinically variable disorder with several patterns of inheritance.

Familial Hypercholesterolemia

Familial hypercholesterolemia is a “receptor disease” that is the consequence of a mutation in the gene encoding the receptor for LDL, which is involved in the transport and metabolism of cholesterol. As a consequence of receptor abnormalities there is a loss of feedback control and elevated levels of cholesterol that induce premature atherosclerosis, leading to a greatly increased risk of myocardial infarction.²¹

Familial hypercholesterolemia is one of the most frequently occurring mendelian disorders. Heterozygotes with one mutant gene, representing about 1 in 500 individuals, have from birth a two-fold to three-fold elevation of plasma cholesterol level, leading to tendinous xanthomas and premature atherosclerosis in adult life (Chapter 11). Homozygotes, having a double dose of the mutant gene, are much more severely affected, and may have five-fold to six-fold elevations in plasma cholesterol levels. These individuals develop skin xanthomas and coronary, cerebral, and peripheral vascular atherosclerosis at an early age. Myocardial infarction may develop before age 20. Large-scale studies have found that familial hypercholesterolemia is present in 3% to 6% of survivors of myocardial infarction.

An understanding of this disorder requires that we briefly review the normal process of cholesterol metabolism and transport. Approximately 7% of the body’s cholesterol circulates in the plasma, predominantly in the form of LDL. As might be expected, the amount of plasma cholesterol is influenced by its synthesis and catabolism, and the liver plays a crucial role in both these processes (Fig. 5-7). The first step in this complex sequence is the secretion of very-low-density lipoproteins (VLDLs) by the liver into the bloodstream. VLDL particles are rich in triglycerides, but they contain lesser amounts of cholesteryl esters. When a VLDL particle reaches the capillaries of adipose tissue or muscle, it is cleaved by lipoprotein lipase, a process that extracts most of the triglycerides. The resulting molecule, called intermediate-density lipoprotein (IDL), is reduced in triglyceride content and enriched in cholesteryl esters, but it retains two of the three apoproteins (B-100 and E) present in the parent VLDL particle (see Fig. 5-7). After release from the capillary endothelium, the IDL particles have one of two fates. Approximately 50% of newly formed IDL is rapidly taken up by the liver by receptor-mediated transport. The receptor responsible for the binding of IDL to the liver cell membrane recognizes both apoprotein B-100 and apoprotein E. It is called the LDL receptor, however, because it is also involved in the hepatic clearance of LDL, as described later. In the liver cells, IDL is recycled to generate VLDL. The IDL particles not taken up by the liver are subjected to further metabolic processing that removes most of the remaining triglycerides and apoprotein E, yielding cholesterol-rich LDL particles. It should be emphasized that IDL is the immediate and major source of plasma LDL. There seem to be two mechanisms for removal of LDL from plasma, one mediated by an LDL receptor and the other by a receptor for oxidized LDL (scavenger receptor), described later. Although many cell types, including fibroblasts, lymphocytes, smooth muscle cells, hepatocytes, and adrenocortical cells, possess high-affinity LDL receptors, approximately 70% of the plasma LDL seems to be cleared by the liver, using a quite sophisticated transport process (Fig. 5-8). The first step involves binding of LDL to cell surface receptors, which are clustered in specialized regions of the plasma membrane called coated pits. After binding, the coated pits containing the receptor-bound LDL are internalized by invagination to form coated vesicles, after which they migrate within the cell to fuse with the lysosomes. Here the LDL dissociates from the receptor, which is recycled to the surface. In the lysosomes the LDL molecule is enzymatically degraded; the apoprotein part is hydrolyzed to amino acids, whereas the cholesteryl esters are broken down to free cholesterol. This free cholesterol, in turn, crosses the lysosomal membrane to enter the cytoplasm, where it is used for membrane synthesis and as a regulator of cholesterol homeostasis. The exit of cholesterol from the lysosomes requires the action of two proteins called NPC1 and NPC2 (see “Niemann-Pick Disease, Type C”). Three separate processes are affected by the released intracellular cholesterol, as follows:

FIGURE 5-7 Low-density lipoprotein (LDL) metabolism and the role of the liver in its synthesis and clearance. Lipolysis of very-low-density lipoprotein (VLDL) by lipoprotein lipase in the capillaries releases triglycerides, which are then stored in fat cells and used as a source of energy in skeletal muscles. See text for explanation of abbreviations used.

FIGURE 5-8 The LDL receptor pathway and regulation of cholesterol metabolism.

As mentioned earlier, familial hypercholesterolemia results from mutations in the gene specifying the receptor for LDL. Heterozygotes with familial hypercholesterolemia possess only 50% of the normal number of high-affinity LDL receptors, because they have only one normal gene. As a result of this defect in transport, the catabolism of LDL by the receptor-dependent pathways is impaired, and the plasma level of LDL increases approximately two-fold. Homozygotes have virtually no normal LDL receptors in their cells and have much higher levels of circulating LDL. In addition to defective LDL clearance, both the homozygotes and heterozygotes have increased synthesis of LDL. The mechanism of increased synthesis that contributes to hypercholesterolemia also results from a lack of LDL receptors (see Fig. 5-7). Recall that IDL, the immediate precursor of plasma LDL, also uses hepatic LDL receptors (apoprotein B-100 and E receptors) for its transport into the liver. In familial hypercholesterolemia, impaired IDL transport into the liver secondarily diverts a greater proportion of plasma IDL into the precursor pool for plasma LDL.

The transport of LDL via the scavenger receptor seems to occur at least in part into the cells of the mononuclear phagocyte system. Monocytes and macrophages have receptors for chemically altered (e.g., acetylated or oxidized) LDL. Normally the amount of LDL transported along this scavenger receptor pathway is less than that mediated by the LDL receptor–dependent mechanisms. In the face of hypercholesterolemia, however, there is a marked increase in the scavenger receptor–mediated traffic of LDL cholesterol into the cells of the mononuclear phagocyte system and possibly the vascular walls (Chapter 11). This increase is responsible for the appearance of xanthomas and contributes to the pathogenesis of premature atherosclerosis.

The molecular genetics of familial hypercholesterolemia is extremely complex. More than 900 mutations, including insertions, deletions, and missense and nonsense mutations, involving the LDL receptor gene have been identified. These can be classified into five groups (Fig. 5-9): Class I mutations are relatively uncommon, and they lead to a complete failure of synthesis of the receptor protein (null allele). Class II mutations are fairly common; they encode receptor proteins that accumulate in the endoplasmic reticulum because their folding defects make it impossible for them to be transported to the Golgi complex. Class III mutations affect the LDL-binding domain of the receptor; the encoded proteins reach the cell surface but fail to bind LDL or do so poorly. Class IV mutations encode proteins that are synthesized and transported to the cell surface efficiently. They bind LDL normally, but they fail to localize in coated pits, and hence the bound LDL is not internalized. Class V mutations encode proteins that are expressed on the cell surface, can bind LDL, and can be internalized; however, the pH-dependent dissociation of the receptor and the bound LDL fails to occur. Such receptors are trapped in the endosome, where they are degraded, and hence they fail to recycle to the cell surface.

FIGURE 5-9 Classification of LDL receptor mutations based on abnormal function of the mutant protein. These mutations disrupt the receptor’s synthesis in the endoplasmic reticulum, transport to the Golgi complex, binding of apoprotein ligands, clustering in coated pits, and recycling in endosomes. Each class is heterogeneous at the DNA level.

(Modified with permission from Hobbs HH et al.: The LDL receptor locus in familial hypercholesterolemia: mutational analysis of a membrane protein. Annu Rev Genet 24:133–170, 1990. © 1990 by Annual Reviews.)

The discovery of the critical role of LDL receptors in cholesterol homeostasis has led to the rational design of drugs that lower plasma cholesterol by increasing the number of LDL receptors. One strategy is based on the ability of certain drugs (statins) to suppress intracellular cholesterol synthesis by inhibiting the enzyme HMG CoA reductase. This, in turn, allows greater synthesis of LDL receptors (see Fig. 5-8).

Lysosomal Storage Diseases

Lysosomes are key components of the “intracellular digestive tract.” They contain a battery of hydrolytic enzymes, which have two special properties. First, they function in the acidic milieu of the lysosomes. Second, these enzymes constitute a special category of secretory proteins that are destined not for the extracellular fluids but for intracellular organelles. This latter characteristic requires special processing within the Golgi apparatus, which is reviewed briefly. Similar to all other secretory proteins, lysosomal enzymes (or acid hydrolases, as they are sometimes called) are synthesized in the endoplasmic reticulum and transported to the Golgi apparatus. Within the Golgi complex they undergo a variety of post-translational modifications, of which one is worthy of special note. This modification involves the attachment of terminal mannose6-phosphate groups to some of the oligosaccharide side chains. The phosphorylated mannose residues serve as an “address label” that is recognized by specific receptors found on the inner surface of the Golgi membrane. Lysosomal enzymes bind these receptors and are thereby segregated from the numerous other secretory proteins within the Golgi. Subsequently, small transport vesicles containing the receptor-bound enzymes are pinched off from the Golgi and proceed to fuse with the lysosomes. Thus, the enzymes are targeted to their intracellular abode, and the vesicles are shuttled back to the Golgi (Fig. 5-10). As indicated later, genetically determined errors in this remarkable sorting mechanism may give rise to one form of lysosomal storage disease.²²

FIGURE 5-10 Synthesis and intracellular transport of lysosomal enzymes.

The lysosomal acid hydrolases catalyze the breakdown of a variety of complex macromolecules. These large molecules may be derived from the metabolic turnover of intracellular organelles (autophagy), or they may be acquired from outside the cells by phagocytosis (heterophagy). With an inherited deficiency of a functional lysosomal enzyme, catabolism of its substrate remains incomplete, leading to the accumulation of the partially degraded insoluble metabolite within the lysosomes. Stuffed with incompletely digested macromolecules, these organelles become large and numerous enough to interfere with normal cell functions, giving rise to the so-called lysosomal storage disorders (Fig. 5-11). In addition to “missing enzymes,” lysosomal storage diseases can result from lack of any protein essential for normal function of lysosomes. Examples are:

FIGURE 5-11 Pathogenesis of lysosomal storage diseases. In the example shown a complex substrate is normally degraded by a series of lysosomal enzymes (A, B, and C) into soluble end products. If there is a deficiency or malfunction of one of the enzymes (e.g., B), catabolism is incomplete and insoluble intermediates accumulate in the lysosomes.

There are three general approaches to the treatment of lysosomal storage diseases. The most obvious is enzyme replacement therapy, currently in use for several lysosomal storage diseases. Another approach, the “substrate reduction therapy,” is based on the premise that if the substrate to be degraded by the lysosomal enzyme can be reduced, the residual enzyme activity may be sufficient to catabolize it and prevent accumulation. A more recent strategy is based on the understanding of the molecular basis of enzyme deficiency. In many disorders, exemplified by Gaucher disease, the enzyme activity is low because the mutant proteins are unstable and prone to misfolding, and hence degraded in the endoplasmic reticulum. In such diseases an exogenous competitive inhibitor of the enzyme can, paradoxically, bind to the mutant enzyme and act as the “folding template” that assists proper folding of the enzyme and thus prevents its degradation. Such molecular chaperone therapy is under active investigation.²³

Several distinctive and separable conditions are included among the lysosomal storage diseases (Table 5-6). In general, the distribution of the stored material, and hence the organs affected, is determined by two interrelated factors: (1) the tissue where most of the material to be degraded is found and (2) the location where most of the degradation normally occurs. For example, brain is rich in gangliosides, and hence defective hydrolysis of gangliosides, as occurs in G_M1 and G_M2 gangliosidoses, results primarily in accumulation within neurons and consequent neurologic symptoms. Defects in degradation of mucopolysaccharides affect virtually every organ, because mucopolysaccharides are widely distributed in the body. Because cells of the mononuclear phagocyte system are especially rich in lysosomes and are involved in the degradation of a variety of substrates, organs rich in phagocytic cells, such as the spleen and liver, are frequently enlarged in several forms of lysosomal storage disorders. The ever-expanding number of lysosomal storage diseases can be divided into rational categories based on the biochemical nature of the accumulated metabolite, thus creating such subgroups as the glycogenoses, sphingolipidoses (lipidoses), mucopolysaccharidoses (MPSs), and mucolipidoses (see Table 5-6). Only the most common disorders among the remaining groups are considered here.

TABLE 5-6 Lysosomal Storage Diseases

Tay-Sachs Disease (G_M2 Gangliosidosis: Hexosaminidase α-Subunit Deficiency)

G_M2 gangliosidoses are a group of three lysosomal storage diseases caused by an inability to catabolize G_M2 gangliosides. Degradation of G_M2 gangliosides requires three polypeptides encoded by three distinct genes. The phenotypic effects of mutations affecting these genes are fairly similar, because they result from accumulation of G_M2 gangliosides.²⁴ The underlying enzyme defect, however, is different for each. Tay-Sachs disease, the most common form of G_M2 gangliosidosis, results from mutations in the α-subunit locus on chromosome 15 that cause a severe deficiency of hexosaminidase A. This disease is especially prevalent among Jews, particularly among those of Eastern European (Ashkenazic) origin, in whom a carrier rate of 1 in 30 has been reported.

Morphology. The hexosaminidase A is absent from virtually all the tissues, so G_M2 ganglioside accumulates in many tissues (e.g., heart, liver, spleen), but the involvement of neurons in the central and autonomic nervous systems and retina dominates the clinical picture. On histologic examination, the neurons are ballooned with cytoplasmic vacuoles, each representing a markedly distended lysosome filled with gangliosides (Fig. 5-12A). Stains for fat such as oil red O and Sudan black B are positive. With the electron microscope, several types of cytoplasmic inclusions can be visualized, the most prominent being whorled configurations within lysosomes composed of onion-skin layers of membranes (Fig. 5-12B). In time there is progressive destruction of neurons, proliferation of microglia, and accumulation of complex lipids in phagocytes within the brain substance. A similar process occurs in the cerebellum as well as in neurons throughout the basal ganglia, brain stem, spinal cord, and dorsal root ganglia and in the neurons of the autonomic nervous system. The ganglion cells in the retina are similarly swollen with G_M2 ganglioside, particularly at the margins of the macula. A cherry-red spot thus appears in the macula, representing accentuation of the normal color of the macular choroid contrasted with the pallor produced by the swollen ganglion cells in the remainder of the retina (Chapter 29). This finding is characteristic of Tay-Sachs disease and other storage disorders affecting the neurons.

FIGURE 5-12 Ganglion cells in Tay-Sachs disease. A, Under the light microscope a large neuron has obvious lipid vacuolation. B, A portion of a neuron under the electron microscope shows prominent lysosomes with whorled configurations. Part of the nucleus is shown above.

(A, courtesy of Dr. Arthur Weinberg, Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX; B, electron micrograph courtesy of Dr. Joe Rutledge, University of Texas Southwestern Medical Center, Dallas, TX.)

Clinical Features

The affected infants appear normal at birth but begin to manifest signs and symptoms at about age 6 months. There is relentless motor and mental deterioration, beginning with motor incoordination, mental obtundation leading to muscular flaccidity, blindness, and increasing dementia. Sometime during the early course of the disease, the characteristic, but not pathognomonic, cherry-red spot appears in the macula of the eye in almost all patients. Over the span of 1 or 2 years a complete vegetative state is reached, followed by death at age 2 to 3 years. More than 100 mutations have been described in the α-subunit gene; most affect protein folding. Such misfolded proteins trigger the “unfolded protein” response (Chapter 1) leading to apoptosis. These findings have given rise to the possibility of chaperone therapy of Tay-Sachs disease.

Antenatal diagnosis and carrier detection are possible by enzyme assays and DNA-based analysis. The clinical features of the two other forms of G_M2 gangliosidosis, Sandhoff disease, resulting from β-subunit defect, and G_M2 activator deficiency, are similar to those of Tay-Sachs disease.

Niemann-Pick Disease, Types A and B

Niemann-Pick disease types A and B refers to two related disorders that are characterized by lysosomal accumulation of sphingomyelin due to an inherited deficiency of sphingomyelinase.²⁵ Type A is a severe infantile form with extensive neurologic involvement, marked visceral accumulations of sphingomyelin, and progressive wasting and early death within the first 3 years of life. In contrast, type B disease patients have organomegaly but generally no central nervous system involvement. They usually survive into adulthood. As with Tay-Sachs disease, Niemann-Pick disease types A and B are common in Ashkenazi Jews. The acid sphingomyelinase gene maps to chromosome 11p15.4 and is one of the imprinted genes that is preferentially expressed from the maternal chromosome as a result of epigenetic silencing of the paternal gene (discussed later). More than 100 mutations have been found in the acid sphingomyelinase gene and there seems to be a correlation between the type of mutation, the severity of enzyme deficiency, and the phenotype.

Morphology. In the classic infantile type A variant, a missense mutation causes almost complete deficiency of sphingomyelinase. Sphingomyelin is a ubiquitous component of cellular (including organellar) membranes, and so the enzyme deficiency blocks degradation of the lipid, resulting in its progressive accumulation within lysosomes, particularly within cells of the mononuclear phagocyte system. Affected cells become enlarged, sometimes to 90 μm in diameter, due to the distention of lysosomes with sphingomyelin and cholesterol. Innumerable small vacuoles of relatively uniform size are created, imparting foaminess to the cytoplasm (Fig. 5-13). In frozen sections of fresh tissue, the vacuoles stain for fat. Electron microscopy confirms that the vacuoles are engorged secondary lysosomes that often contain membranous cytoplasmic bodies resembling concentric lamellated myelin figures, sometimes called “zebra” bodies.

FIGURE 5-13 Niemann-Pick disease in liver. The hepatocytes and Kupffer cells have a foamy, vacuolated appearance due to deposition of lipids.

(Courtesy of Dr. Arthur Weinberg, Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX.)

The lipid-laden phagocytic foam cells are widely distributed in the spleen, liver, lymph nodes, bone marrow, tonsils, gastrointestinal tract, and lungs. The involvement of the spleen generally produces massive enlargement, sometimes to ten times its normal weight, but the hepatomegaly is usually not quite so striking. The lymph nodes are generally moderately to markedly enlarged throughout the body.

Involvement of the brain and eye deserves special mention. In the brain the gyri are shrunken and the sulci widened. The neuronal involvement is diffuse, affecting all parts of the nervous system. Vacuolation and ballooning of neurons constitute the dominant histologic change, which in time leads to cell death and loss of brain substance. A retinal cherry-red spot similar to that seen in Tay-Sachs disease is present in about one third to one half of affected individuals.

Clinical manifestations in type A disease may be present at birth and almost invariably become evident by age 6 months. Infants typically have a protuberant abdomen because of the hepatosplenomegaly. Once the manifestations appear, they are followed by progressive failure to thrive, vomiting, fever, and generalized lymphadenopathy as well as progressive deterioration of psychomotor function. Death comes, usually within the first or second year of life.

The diagnosis is established by biochemical assays for sphingomyelinase activity in liver or bone marrow biopsy. Individuals affected with types A and B as well as carriers can be detected by DNA analysis.

Gaucher Disease

Gaucher disease refers to a cluster of autosomal recessive disorders resulting from mutations in the gene encoding glucocerebrosidase.²⁷ This disease is the most common lysosomal storage disorder. The affected gene encodes glucocerebrosidase, an enzyme that normally cleaves the glucose residue from ceramide. As a result of the enzyme defect, glucocerebroside accumulates principally in phagocytes but in some subtypes also in the central nervous system. Glucocerebrosides are continually formed from the catabolism of glycolipids derived mainly from the cell membranes of senescent leukocytes and erythrocytes. It is clear now that the pathologic changes in Gaucher disease are caused not just by the burden of storage material but also by activation of macrophages and the consequent secretion of cytokines such as IL-1, IL-6, and tumor necrosis factor (TNF). Three clinical subtypes of Gaucher disease have been distinguished. The most common, accounting for 99% of cases, is called type I, or the chronic non-neuronopathic form. In this type, storage of glucocerebrosides is limited to the mononuclear phagocytes throughout the body without involving the brain. Splenic and skeletal involvements dominate this pattern of the disease. It is found principally in Jews of European stock. Individuals with this disorder have reduced but detectable levels of glucocerebrosidase activity. Longevity is shortened but not markedly. Type II, or acute neuronopathic Gaucher disease, is the infantile acute cerebral pattern. This form has no predilection for Jews. In these patients there is virtually no detectable glucocerebrosidase activity in the tissues. Hepatosplenomegaly is also seen in this form of Gaucher disease, but the clinical picture is dominated by progressive central nervous system involvement, leading to death at an early age. A third pattern, type III, is intermediate between types I and II. These patients have the systemic involvement characteristic of type I but have progressive central nervous system disease that usually begins in adolescence or early adulthood.

Morphology. Glucocerebrosides accumulate in massive amounts within phagocytic cells throughout the body in all forms of Gaucher disease. The distended phagocytic cells, known as Gaucher cells, are found in the spleen, liver, bone marrow, lymph nodes, tonsils, thymus, and Peyer’s patches. Similar cells may be found in both the alveolar septa and the air spaces in the lung. In contrast to other lipid storage diseases, Gaucher cells rarely appear vacuolated but instead have a fibrillary type of cytoplasm likened to crumpled tissue paper (Fig. 5-14). Gaucher cells are often enlarged, sometimes up to 100 μm in diameter, and have one or more dark, eccentrically placed nuclei. Periodic acid–Schiff staining is usually intensely positive. With the electron microscope the fibrillary cytoplasm can be resolved as elongated, distended lysosomes, containing the stored lipid in stacks of bilayers.

FIGURE 5-14 Gaucher disease involving the bone marrow. Gaucher cells (A, H&E; B, Wright stain) are plump macrophages that characteristically have the appearance in the cytoplasm of crumpled tissue paper (B), due to accumulation of glucocerebroside.

(Courtesy of Dr. John Anastasi, Department of Pathology, University of Chicago, Chicago, IL.)

In type I disease, the spleen is enlarged, sometimes up to 10 kg. The lymphadenopathy is mild to moderate and is body-wide. The accumulation of Gaucher cells in the bone marrow occurs in 70% to 100% of cases of type I Gaucher disease. It produces areas of bone erosion that are sometimes small but in other cases sufficiently large to give rise to pathologic fractures. Bone destruction occurs due to the secretion of cytokines by activated macrophages. In patients with cerebral involvement, Gaucher cells are seen in the Virchow-Robin spaces, and arterioles are surrounded by swollen adventitial cells. There is no storage of lipids in the neurons, yet neurons appear shriveled and are progressively destroyed. It is suspected that the lipids that accumulate in the phagocytic cells around blood vessels secrete cytokines that damage nearby neurons.

Clinical Features

The clinical course of Gaucher disease depends on the clinical subtype. In type I, symptoms and signs first appear in adult life and are related to splenomegaly or bone involvement. Most commonly there is pancytopenia or thrombocytopenia secondary to hypersplenism. Pathologic fractures and bone pain occur if there has been extensive expansion of the marrow space. Although the disease is progressive in the adult, it is compatible with long life. In types II and III, central nervous system dysfunction, convulsions, and progressive mental deterioration dominate, although organs such as the liver, spleen, and lymph nodes are also affected.

The diagnosis of homozygotes can be made by measurement of glucocerebrosidase activity in peripheral blood leukocytes or in extracts of cultured skin fibroblasts. In principle, heterozygotes can be identified by detection of mutations. However, because more than 150 mutations in the glucocerebroside gene can cause Gaucher disease, it is not possible to use a single genetic test.

Replacement therapy with recombinant enzymes is the mainstay for treatment of Gaucher disease; it is effective, and those with type I disease can expect normal life expectancy with this form of treatment. However, such therapy is extremely expensive. Because the fundamental defect resides in mononuclear phagocytic cells originating from marrow stem cells, bone marrow transplantation has been attempted. Other work is directed toward correction of the enzyme defect by transfer of the normal glucocerebrosidase gene into the patient’s bone marrow cells. Substrate reduction therapy with inhibitors of glucosylceramide synthetase is also being evaluated.

Glycogen Storage Diseases (Glycogenoses)

The glycogen storage diseases result from a hereditary deficiency of one of the enzymes involved in the synthesis or sequential degradation of glycogen. Depending on the tissue or organ distribution of the specific enzyme in the normal state, glycogen storage in these disorders may be limited to a few tissues, may be more widespread while not affecting all tissues, or may be systemic in distribution.³⁰

The significance of a specific enzyme deficiency is best understood from the perspective of the normal metabolism of glycogen (Fig. 5-15). Glycogen is a storage form of glucose. Glycogen synthesis begins with the conversion of glucose to glucose-6-phosphate by the action of a hexokinase (glucokinase). A phosphoglucomutase then transforms the glucose-6-phosphate to glucose-1-phosphate, which, in turn, is converted to uridine diphosphoglucose. A highly branched, large polymer is then built (molecular weight as high as 100 million), containing as many as 10,000 glucose molecules linked together by α-1,4-glucoside bonds. The glycogen chain and branches continue to be elongated by the addition of glucose molecules mediated by glycogen synthetases. During degradation, distinct phosphorylases in the liver and muscle split glucose-1-phosphate from the glycogen until about four glucose residues remain on each branch, leaving a branched oligosaccharide called limit dextrin. This can be further degraded only by the debranching enzyme. In addition to these major pathways, glycogen is also degraded in the lysosomes by acid maltase. If the lysosomes are deficient in this enzyme, the glycogen contained within them is not accessible to degradation by cytoplasmic enzymes such as phosphorylases.

FIGURE 5-15 Pathways of glycogen metabolism. Asterisks mark the enzyme deficiencies associated with glycogen storage diseases. Roman numerals indicate the type of glycogen storage disease associated with the given enzyme deficiency. Types V and VI result from deficiencies of muscle and liver phosphorylases, respectively.

(Modified from Hers H et al.: Glycogen storage diseases. In Scriver CR et al. [eds]: The Metabolic Basis of Inherited Disease, 6th ed. New York, McGraw-Hill, 1989, p 425.)

On the basis of specific enzyme deficiencies and the resultant clinical pictures, glycogenoses have traditionally been divided into a dozen or so syndromes designated by roman numerals, and the list continues to grow.³¹ On the basis of pathophysiology glycogenoses can be divided into three major subgroups (Table 5-7).

TABLE 5-7 Principal Subgroups of Glycogenoses

FIGURE 5-16 A, Normal glycogen metabolism in the liver and skeletal muscles. B, Effects of an inherited deficiency of hepatic enzymes involved in glycogen metabolism. C, Consequences of a genetic deficiency in the enzymes that metabolize glycogen in skeletal muscles.

FIGURE 5-17 Pompe disease (glycogen storage disease type II). A, Normal myocardium with abundant eosinophilic cytoplasm. B, Patient with Pompe disease (same magnification) showing the myocardial fibers full of glycogen seen as clear spaces.

(Courtesy of Dr. Trace Worrell, Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX.)

Alkaptonuria (Ochronosis)

Alkaptonuria, the first human inborn error of metabolism to be discovered, is an autosomal recessive disorder in which there is lack of homogentisic oxidase, an enzyme that converts homogentisic acid to methylacetoacetic acid in the tyrosine degradation pathway.³⁵ Thus, homogentisic acid accumulates in the body. A large amount is excreted, imparting a black color to the urine if allowed to stand and undergo oxidation.

Trisomy 21 (Down Syndrome)

Down syndrome is the most common of the chromosomal disorders and is a major cause of mental retardation. In the United States the incidence in newborns is about 1 in 700. Approximately 95% of affected individuals have trisomy 21, so their chromosome count is 47. FISH with chromosome 21–specific probes reveals the extra copy of chromosome 21 in such cases (Fig. 5-20). Most others have normal chromosome numbers, but the extra chromosomal material is present as a translocation. As mentioned earlier, the most common cause of trisomy and therefore of Down syndrome is meiotic nondisjunction. The parents of such children have a normal karyotype and are normal in all respects.

FIGURE 5-20 FISH analysis of an interphase nucleus, using locus-specific probes to chromosome 13 (green) and chromosome 21 (red), revealing three red signals consistent with trisomy 21.

(Courtesy of Dr. Stuart Schwartz, Department of Pathology, University of Chicago, Chicago, IL.)

Maternal age has a strong influence on the incidence of trisomy 21. It occurs once in 1550 live births in women under age 20, in contrast to 1 in 25 live births for mothers over age 45. The correlation with maternal age suggests that in most cases the meiotic nondisjunction of chromosome 21 occurs in the ovum. Studies in which DNA polymorphisms were used to trace the parental origin of chromosome 21 have revealed that in 95% of the cases with trisomy 21 the extra chromosome is of maternal origin. Although many hypotheses have been advanced, the reason for the increased susceptibility of the ovum to nondisjunction remains unknown.

In about 4% of cases of Down syndrome, the extra chromosomal material derives from the presence of a robertsonian translocation of the long arm of chromosome 21 to another acrocentric chromosome (e.g., 22 or 14). Because the fertilized ovum already possesses two normal autosomes 21, the translocated material provides the same triple gene dosage as in trisomy 21. Such cases are frequently (but not always) familial, and the translocated chromosome is inherited from one of the parents (usually the mother), who is a carrier of a robertsonian translocation, for example, a mother with karyotype 45,XX,der(14;21)(q10;q10).

Approximately 1% of Down syndrome patients are mosaics, usually having a mixture of cells with 46 and 47 chromosomes. This mosaicism results from mitotic nondisjunction of chromosome 21 during an early stage of embryogenesis. Symptoms in such cases are variable and milder, depending on the proportion of abnormal cells. Clearly, in cases of translocation or mosaic Down syndrome, maternal age is of no importance.

The diagnostic clinical features of this condition—flat facial profile, oblique palpebral fissures, and epicanthic folds (Fig. 5-21)—are usually readily evident, even at birth.³⁷ Down syndrome is a leading cause of severe mental retardation; approximately 80% of those afflicted have an IQ of 25 to 50. Ironically, these severely disadvantaged children may have a gentle, shy manner and may be more easily directed than their more fortunate normal siblings. It should be pointed out that some mosaics with Down syndrome have mild phenotypic changes and often even have normal or near-normal intelligence. In addition to the phenotypic abnormalities and the mental retardation already noted, some other clinical features are worthy of note.

FIGURE 5-21 Clinical features and karyotypes of selected autosomal trisomies.

Despite all these problems, improved medical care has increased the longevity of individuals with trisomy 21. Currently the median age at death is 47 years (up from 25 years in 1983).

Although the karyotype and clinical features of trisomy 21 have been known for decades, little is known about the molecular basis of Down syndrome. Chromosome 21 contains approximately 430 genes. Interestingly, there are several gene clusters, each of which is predicted to participate in the same biologic pathway. For example, there are 16 genes that are involved in the mitochondrial energy pathway, several that are likely to influence central nervous system development, and a group that is involved in folate metabolism. It is not known how each of these groups of genes is related to Down syndrome. The gene dosage hypothesis assumes that the phenotypic features of the trisomy 21 are related to overexpression of genes. In reality only about 37% of the genes on chromosomes 21 are overexpressed by 150%, whereas others have variable degrees of changes in expression. Further complexity in defining the specific genes involved in the pathogenesis of Down syndrome is related to the presence of several miRNA genes on chromosome 21 that can shut down translation of genes that map elsewhere in the genome.³⁹ Thus, despite the availability of the gene map of chromosome 21, the progress in understanding the molecular basis of Down syndrome remains slow.⁴⁰

Other Trisomies

A variety of other trisomies, involving chromosomes 8, 9, 13, 18, and 22, have been described. Only trisomy 18 (Edwards syndrome) and trisomy 13 (Patau syndrome) are common enough to merit brief mention here. As noted in Figure 5-21, they share several karyotypic and clinical features with trisomy 21. Thus, most cases result from meiotic nondisjunction and therefore carry a complete extra copy of chromosome 18 or 13. As in Down syndrome, an association with increased maternal age is also noted. In contrast to trisomy 21, however, the malformations are much more severe and wide-ranging. As a result, only rarely do these infants survive beyond the first year of life. Most succumb within a few weeks to months.

Chromosome 22q11.2 Deletion Syndrome

Chromosome 22q11.2 deletion syndrome encompasses a spectrum of disorders that result from a small deletion of band q11.2 on the long arm of chromosome 22.⁴¹ The syndrome is fairly common, occurring in as many as 1 in 4000 births, but it is often missed because of variable clinical features. These include congenital heart defects, abnormalities of the palate, facial dysmorphism, developmental delay, and variable degrees of T-cell immunodeficiency and hypocalcemia. Previously, these clinical features were considered to represent two different disorders—DiGeorge syndrome and velocardiofacial syndrome. Patients with DiGeorge syndrome have thymic hypoplasia, with resultant T-cell immunodeficiency (Chapter 6), parathyroid hypoplasia giving rise to hypocalcemia, a variety of cardiac malformations affecting the outflow tract, and mild facial anomalies. The clinical features of the so-called velocardiofacial syndrome include facial dysmorphism (prominent nose, retrognathia), cleft palate, cardiovascular anomalies, and learning disabilities. Less frequently, these patients also have immunodeficiency. Until recently the overlapping clinical features of these two conditions (e.g., cardiac malformations, facial dysmorphology) were not appreciated; it was only after these two apparently unrelated syndromes were found to be associated with a similar cytogenetic abnormality that the clinical overlap came into focus. Recent studies indicate that in addition to the numerous structural malformations, individuals with the 22q11.2 deletion syndrome are at a particularly high risk for psychotic illnesses, such as schizophrenia and bipolar disorders.⁴² In fact, it is estimated that approximately 25% of adults with this syndrome develop schizophrenia. Conversely, deletions of the region can be found in 2% to 3% of individuals with childhood-onset schizophrenia. In addition, attention deficit hyperactivity disorder is seen in 30% to 35% of affected children.

The diagnosis of this condition may be suspected on clinical grounds but can be established only by detection of the deletion by FISH (Fig. 5-22). By this test, approximately 90% of those previously diagnosed as having DiGeorge syndrome and 80% of those with the velocardiofacial syndrome have a deletion of 22q11.2. Thirty percent of individuals with conotruncal cardiac defects but no other features of this syndrome also reveal deletions of the same chromosomal region.

FIGURE 5-22 FISH on both metaphase chromosomes and an interphase cell from a patient with Di George syndrome demonstrating the deletion of the TUPLE1 probe (official name HIRA) localized to chromosome 22q11.2. The TUPLE1 probe is in red, and the control probe, localized to 22q, is in green. The metaphase spread shows one chromosome 22 with both a green signal (control probe) and a red signal (from the TUPLE1 probe). The other chromosome 22 shows only hybridization with the control probe (green), but no red signal since there is a deletion on this chromosome. The interphase cell shows two areas of hybridization with the control probe (in green) but also only one area of hybridization with the TUPLE1 probe (in red), illustrating a deletion of chromosome 22q11.2.

(Courtesy of Dr. Stuart Schwartz, Department of Pathology, University of Chicago, Chicago, IL.)

The molecular basis of this syndrome is not fully understood. The deleted region is large (approximately 1.5 megabases) and includes many genes. The clinical heterogeneity, with predominant immunodeficiency in some cases (DiGeorge syndrome) and predominant dysmorphology and cardiac malformations in other cases, probably reflects the variable position and size of the deleted segment from this genetic region. Approximately 30 candidate genes have been mapped to the deleted region. Among these, TBX1, a T-box transcription factor is most closely associated with the phenotypic features of this syndrome.⁴¹ This gene is expressed in the pharyngeal mesenchyme and endodermal pouch from which facial structures, thymus, and parathyroid are derived. The targets of TBX1 include PAX9, a gene that controls the development of the palate, parathyroids, and thymus. Clearly there are other genes that contribute to the behavioral and psychiatric disorders that remain to be identified.

Klinefelter Syndrome

Klinefelter syndrome is best defined as male hypogonadism that occurs when there are two or more X chromosomes and one or more Y chromosomes.⁴⁸ It is one of the most frequent forms of genetic disease involving the sex chromosomes as well as one of the most common causes of hypogonadism in the male. The incidence of this condition is approximately 1 in 660 live male births.⁴⁹ It can rarely be diagnosed before puberty, particularly because the testicular abnormality does not develop before early puberty. Most patients have a distinctive body habitus with an increase in length between the soles and the pubic bone, which creates the appearance of an elongated body. Also characteristic are eunuchoid body habitus with abnormally long legs; small atrophic testes often associated with a small penis; and lack of such secondary male characteristics as deep voice, beard, and male distribution of pubic hair. Gynecomastia may be present. The mean IQ is somewhat lower than normal, but mental retardation is uncommon. There is increased incidence of type 2 diabetes and the metabolic syndrome; and curiously, mitral valve prolapse is seen in about 50% of adults with Klinefelter syndrome. It should be evident that the clinical features of this condition are variable, the only consistent finding being hypogonadism. Plasma gonadotropin concentrations, particularly follicle-stimulating hormone, are consistently elevated, whereas testosterone levels are variably reduced. Mean plasma estradiol levels are elevated by an as yet unknown mechanism. The ratio of estrogens and testosterone determines the degree of feminization in individual cases.

Klinefelter syndrome is an important genetic cause of reduced spermatogenesis and male infertility.⁵⁰ In some patients the testicular tubules are totally atrophied and replaced by pink, hyaline, collagenous ghosts. In others, apparently normal tubules are interspersed with atrophic tubules. In some patients all tubules are primitive and appear embryonic, consisting of cords of cells that never developed a lumen or progressed to mature spermatogenesis. Leydig cells appear prominent, as a result of the atrophy and crowding of tubules and elevation of gonadotropin concentrations.

Patients with Klinefelter syndrome have a higher risk for breast cancer (20 times more common than in normal males), extragonadal germ cell tumors, and autoimmune diseases such as systemic lupus erythematosus.

The classic pattern of Klinefelter syndrome is associated with a 47,XXY karyotype (90% of cases). This complement of chromosomes results from nondisjunction during the meiotic divisions in one of the parents. Maternal and paternal nondisjunction at the first meiotic division are roughly equally involved. There is no phenotypic difference between those who receive the extra X chromosome from their father and those who receive it from their mother. Maternal age is increased in the cases associated with errors in oogenesis. In addition to this classic karyotype, approximately 15% of patients with Klinefelter syndrome have been found to have a variety of mosaic patterns, most of them being 46,XY/47,XXY. Other patterns are 47,XXY/48,XXXY and variations on this theme. As is the case with normal females, all but one X undergoes inactivation in patients with Klinefelter syndrome. Why then, do the patients with this disorder have hypogonadism and associated features? The explanation of this lies in the pattern of X inactivation. The gene encoding the androgen receptor, through which testosterone mediates its effects, maps on the X chromosome. The androgen receptor gene contains highly polymorphic CAG (trinucleotide) repeats. The functional response to androgens is dictated, in part, by the number of CAG repeats. With shorter CAG repeats, the effect of androgens is more pronounced. In persons with Klinefelter syndrome the X chromosome, bearing androgen receptor with the shortest CAG repeat, is preferentially inactivated. Such nonrandom X inactivation leaves the allele with the longest CAG repeat active, thus accounting for hypogonadism.

Turner Syndrome

Turner syndrome results from complete or partial monosomy of the X chromosome and is characterized primarily by hypogonadism in phenotypic females.⁵¹ It is the most common sex chromosome abnormality in females, affecting about 1 in 2000 live-born females.

With routine cytogenetic methods, three types of karyotypic abnormalities are seen in individuals with Turner syndrome. Approximately 57% are missing an entire X chromosome, resulting in a 45,X karyotype. Of the remaining 43%, approximately one third (approximately 14%) have structural abnormalities of the X chromosomes, and two thirds (approximately 29%) are mosaics. The common feature of the structural abnormalities is to produce partial monosomy of the X chromosome. In order of frequency, the structural abnormalities of the X chromosome include (1) an isochromosome of the long arm, 46,X,i(X)(q10) resulting in the loss of the short arm; (2) deletion of portions of both long and short arms, resulting in the formation of a ring chromosome, 46,X,r(X); and (3) deletion of portions of the short or long arm, 46X,del(Xq) or 46X,del(Xp). The mosaic patients have a 45,X cell population along with one or more karyotypically normal or abnormal cell types. Examples of karyotypes that mosaic Turner females may have are the following: (1) 45,X/46,XX; (2) 45,X/46,XY; (3) 45,X/47,XXX; or (4) 45,X/46,X,i(X)(q10). Studies suggest that the prevalence of mosaicism in Turner syndrome may be much higher than the 30% detected by conventional cytogenetic studies. With the use of more sensitive techniques, including FISH (discussed later) and polymerase chain reaction (PCR), and analysis of more than one cell type (e.g., peripheral blood and fibroblasts), the prevalence of mosaic Turner syndrome increases to 75%. Because 99% of 45,X conceptuses are nonviable, many authorities believe that there are no truly nonmosaic Turner syndrome patients. While this issue remains controversial, it is important to appreciate the karyotypic heterogeneity associated with Turner syndrome, because it is responsible for significant variations in phenotype. In patients who are truly 45,X, or in whom the proportion of 45,X cells is high, the phenotypic changes are more severe than in those who have readily detectable mosaicism. The latter may have an almost normal appearance and may present only with primary amenorrhea. Similarly, those with a Y chromosome–containing population (e.g., 45,X/46,XY karyotype) may be at risk of developing a gonadal tumor (gonadoblastoma).

The most severely affected patients generally present during infancy with edema of the dorsum of the hand and foot due to lymph stasis, and sometimes swelling of the nape of the neck. The latter is related to markedly distended lymphatic channels, producing a so-called cystic hygroma (Chapter 10). As these infants develop, the swellings subside but often leave bilateral neck webbing and persistent looseness of skin on the back of the neck. Congenital heart disease is also common, affecting 25% to 50% of patients. Left-sided cardiovascular abnormalities, particularly preductal coarctation of the aorta and bicuspid aortic valve, are seen most frequently. Cardiovascular abnormalities are the most important cause of increased mortality in children with Turner syndrome.⁵²

The principal clinical features in the adolescent and adult are illustrated in Figure 5-23. At puberty there is failure to develop normal secondary sex characteristics. The genitalia remain infantile, breast development is inadequate, and there is little pubic hair. The mental status of these patients is usually normal, but subtle defects in nonverbal, visual-spatial information processing have been noted. Of particular importance in establishing the diagnosis in the adult is the shortness of stature (rarely exceeding 150 cm in height) and amenorrhea. Turner syndrome is the single most important cause of primary amenorrhea, accounting for approximately one third of the cases. For reasons not quite clear, approximately 50% of patients develop autoantibodies that react with the thyroid gland, and up to half of these develop clinically manifest hypothyroidism. Equally mysterious is the presence of glucose intolerance, obesity, and insulin resistance in a minority of patients. The last mentioned is significant, because therapy with growth hormone, commonly used in these patients, worsens insulin resistance.

FIGURE 5-23 Clinical features and karyotypes of Turner syndrome.

The molecular pathogenesis of Turner syndrome is not completely understood, but studies have begun to shed some light.⁵³ As mentioned earlier, both X chromosomes are active during oogenesis and are essential for normal development of the ovaries. During normal fetal development, ovaries contain as many as 7 million oocytes. The oocytes gradually disappear so that by menarche their numbers have dwindled to a mere 400,000, and when menopause occurs fewer than 10,000 remain. In Turner syndrome, fetal ovaries develop normally early in embryogenesis, but the absence of the second X chromosome leads to an accelerated loss of oocytes, which is complete by age 2 years. In a sense, therefore, “menopause occurs before menarche,” and the ovaries are reduced to atrophic fibrous strands, devoid of ova and follicles (streak ovaries). Because patients with Turner syndrome also have other (nongonadal) abnormalities, it follows that some genes for normal growth and development of somatic tissues must also reside on the X chromosome. Among the genes involved in the Turner phenotype is the short stature homeobox (SHOX) gene at Xp22.33. This is one of several genes that remain active in both X chromosomes and has an active homologue on the short arm of the Y chromosome. Thus, both normal males and females have two copies of this gene. Haploinsufficiency of SHOX gives rise to short stature. Indeed, deletions of the SHOX gene are noted in 2% to 5% of otherwise normal children with short stature. In keeping with its role as a critical regulator of growth, the SHOX gene is expressed during fetal life in the growth plates of several long bones including the radius, ulna, tibia, and fibula. It is also expressed in the first and second pharyngeal arches. Just as the loss of SHOX is always associated with short stature, excess copies of this gene are associated with tall stature. Whereas haploinsufficiency of SHOX can explain growth deficit in Turner syndrome, it cannot explain other important clinical features such as cardiac malformations and endocrine abnormalities. Clearly several other genes located on the X chromosome are also involved.

Hermaphroditism and Pseudohermaphroditism

The problem of sexual ambiguity is exceedingly complex, and only limited observations are possible here; for more details, reference should be made to specialized sources.⁵⁴ It will be no surprise to medical students that the sex of an individual can be defined on several levels. Genetic sex is determined by the presence or absence of a Y chromosome. No matter how many X chromosomes are present, a single Y chromosome dictates testicular development and the genetic male gender. The initially indifferent gonads of both the male and the female embryos have an inherent tendency to feminize, unless influenced by Y chromosome–dependent masculinizing factors. Gonadal sex is based on the histologic characteristics of the gonads. Ductal sex depends on the presence of derivatives of the müllerian or wolffian ducts. Phenotypic, or genital, sex is based on the appearance of the external genitalia. Sexual ambiguity is present whenever there is disagreement among these various criteria for determining sex.

The term true hermaphrodite implies the presence of both ovarian and testicular tissue. In contrast, a pseudohermaphrodite represents a disagreement between the phenotypic and gonadal sex (i.e., a female pseudohermaphrodite has ovaries but male external genitalia; a male pseudohermaphrodite has testicular tissue but female-type genitalia).

True hermaphroditism, implying the presence of both ovarian and testicular tissue, is an extremely rare condition. In some cases there is a testis on one side and an ovary on the other, whereas in other cases there may be combined ovarian and testicular tissue, referred to as ovotestes. The karyotype is 46,XX in 50% of patients; of the remaining, most are mosaics with a 46,XX/46,XY karyotype. Only rarely is the chromosomal constitution 46,XY. The presence of testes implies that those with the 46,XX karyotype might possess Y-chromosomal material, in particular, the SRY gene, which dictates testicular differentiation. Indeed, molecular analysis has revealed SRY gene expression in the ovotestis of 46,XX true hermaphrodites, indicating either cryptic chimerism localized to the gonads or possibly a Y-to-autosome translocation.⁵⁵

Female pseudohermaphroditism is much less complex. The genetic sex in all cases is XX, and the development of the gonads (ovaries) and internal genitalia is normal. Only the external genitalia are ambiguous or virilized. The basis of female pseudohermaphroditism is excessive and inappropriate exposure to androgenic steroids during the early part of gestation. Such steroids are most commonly derived from the fetal adrenal affected by congenital adrenal hyperplasia, which is transmitted as an autosomal recessive trait. Biosynthetic defects in the pathway of cortisol synthesis are present in these patients, which lead secondarily to excessive synthesis of androgenic steroids by the fetal adrenal cortex (Chapter 24).

Male pseudohermaphroditism represents the most complex of all disorders of sexual differentiation. These individuals possess a Y chromosome, and thus their gonads are exclusively testes, but the genital ducts or the external genitalia are incompletely differentiated along the male phenotype. Their external genitalia are either ambiguous or completely female. Male pseudohermaphroditism is extremely heterogeneous, with a multiplicity of causes. Common to all is defective virilization of the male embryo, which usually results from genetically determined defects in androgen synthesis or action or both. The most common form, called complete androgen insensitivity syndrome (testicular feminization), results from mutations in the gene encoding the androgen receptor.⁵⁶ This gene is located at Xq12, and hence this disorder is inherited as an X-linked recessive.

Fragile-X Syndrome

Fragile-X syndrome is the prototype of diseases in which the mutation is characterized by a long repeating sequence of three nucleotides. Although the specific nucleotide sequence that undergoes amplification differs in the 20 or so disorders included in this group, in most cases the affected sequences share the nucleotides guanine (G) and cytosine (C). In the ensuing discussion we consider the clinical features and inheritance pattern of the fragile-X syndrome, to be followed by the causative molecular lesion. The remaining disorders in this group are discussed later, in this chapter and elsewhere in the book.

With a frequency of 1 in 1550 for affected males and 1 in 8000 for affected females, fragile-X syndrome is the second most common genetic cause of mental retardation, after Down syndrome. It is an X-linked disorder characterized by an inducible cytogenetic abnormality in the X chromosome and an unusual mutation within the familial mental retardation-1 (FMR1) gene. The cytogenetic alteration is seen as a discontinuity of staining or as a constriction in the long arm of the X chromosome when cells are cultured in a folate-deficient medium. Because it appears that the chromosome is “broken” at this locale, it is referred to as a fragile site (Fig. 5-25). It should be noted that more than 100 “fragile sites” have been found in the human genome.⁶⁰ Many, like the one seen in fragile-X syndrome, are sensitive to lack of folate in the medium, whereas others require different culture conditions. The significance of most fragile sites is unknown, since many are present in normal individuals.

FIGURE 5-25 Fragile X, seen as discontinuity of staining.

(Courtesy of Dr. Patricia Howard-Peebles, University of Texas Southwestern Medical Center, Dallas, TX.)

In fragile-X syndrome, the affected males are mentally retarded, with an IQ in the range of 20 to 60. They express a characteristic physical phenotype that includes a long face with a large mandible, large everted ears, and large testicles (macro-orchidism). Hyperextensible joints, a high arched palate, and mitral valve prolapse noted in some patients mimic a connective tissue disorder. These and other physical abnormalities described in this condition, however, are not always present and, in some cases, are quite subtle. The most distinctive feature is macro-orchidism, which is observed in at least 90% of postpubertal males.

As with all X-linked diseases, fragile-X syndrome affects males. Analysis of several pedigrees, however, reveals some patterns of transmission not typically associated with other X-linked recessive disorders (Fig. 5-26). These include the following⁶¹:

FIGURE 5-26 Fragile-X pedigree. Note that in the first generation all sons are normal and all females are carriers. During oogenesis in the carrier female, premutation expands to full mutation; hence, in the next generation all males who inherit the X with full mutation are affected. However, only 50% of females who inherit the full mutation are affected, and only mildly.

(Courtesy of Dr. Nancy Schneider, Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX.)

These unusual patterns perplexed geneticists for years, but molecular studies have begun to unravel the complexities of this condition.^62,63 The first breakthrough came when linkage studies localized the mutation responsible for this disease to Xq27.3, within the cytogenetically abnormal region. Within this region lies the FMR1 gene, characterized by multiple tandem repeats of the nucleotide sequence CGG in its 5′ untranslated region. In the normal population, the number of CGG repeats is small, ranging from 6 to 55 (average, 29). The presence of clinical symptoms and a cytogenetically detectable fragile site is related to the amplification of the CGG repeats. Thus, normal transmitting males and carrier females carry 55 to 200 CGG repeats. Expansions of this size are called premutations. In contrast, affected individuals have an extremely large expansion of the repeat region (200–4000 repeats, or full mutations). Full mutations are believed to arise by further amplification of the CGG repeats seen in premutations. How this process takes place is quite peculiar. Carrier males transmit the repeats to their progeny with small changes in repeat number. When the premutation is passed on by a carrier female, however, there is a high probability of a dramatic amplification of the CGG repeats, leading to mental retardation in most male offspring and 50% of female offspring. Thus, it seems that during the process of oogenesis, but not spermatogenesis, premutations can be converted to mutations by triplet-repeat amplification. This explains the unusual inheritance pattern; that is, the likelihood of mental retardation is much higher in grandsons than in brothers of transmitting males because grandsons incur the risk of inheriting a premutation from their grandfather that is amplified to a “full mutation” in their mothers’ ova. By comparison, brothers of transmitting males, being “higher up” in the pedigree, are less likely to have a full mutation. These molecular details also provide a satisfactory explanation of anticipation—a phenomenon observed by clinical geneticists but not believed by molecular geneticists until triplet-repeat mutations were identified. Why only 50% of the females with the full mutation are clinically affected is not clear. Presumably in those that are clinically affected there is unfavorable lyonization (i.e., there is a higher frequency of cells in which the X chromosome carrying the mutation is active). Recent studies indicate that premutations are not so benign after all. Approximately 30% of females carrying the premutation have premature ovarian failure (before the age of 40 years), and about one third of premutation-carrying males exhibit a progressive neurodegenerative syndrome starting in their sixth decade. This syndrome, referred to as fragile X–associated tremor/ataxia, is characterized by intention tremors and cerebellar ataxia and may progress to parkinsonism. However, it is clear that the abnormalities in premutation carriers are milder and occur later in life.

The molecular basis of mental retardation and other somatic changes is related to a loss of function of the familial mental retardation protein (FMRP). As mentioned earlier, the normal FMR1 gene contains up to 46 CGG repeats in its 5′ untranslated region. When the trinucleotide repeats in the FMR1 gene exceed approximately 230, the DNA of the entire 5′ region of the gene becomes abnormally methylated. Methylation also extends upstream into the promoter region of the gene, resulting in transcriptional suppression of FMR1. The resulting absence of FMRP is believed to cause the phenotypic changes.

FMRP is a widely expressed cytoplasmic protein, most abundant in the brain and testis, the two organs most affected in this disease. The function of FMRP in the brain is beginning to be unraveled.⁶⁴ FMRP is an RNA-binding protein associated with polysomes. Unlike other cells, in neurons protein synthesis occurs both in the perinuclear cytoplasm and in dendritic spines. According to current understanding, FMRP is first transported from the cytoplasm to the nucleus, where it assembles into a complex containing specific mRNA transcripts. The assembled complex is then exported to the cytoplasm. From here the FMRP-mRNA complex is transported to the dendrites close to the synapse (Fig. 5-27). Not all species of mRNA are transported by FMRP to the dendrites. Only those that encode proteins that regulate synaptic function are so shuttled by FMRP. At the synaptic junctions, FMRP suppresses protein synthesis from the bound mRNAs in response to signaling through group I metabotropic glutamate receptors (mGlu-R). In fragile-X syndrome a reduction in FMRP results in increased translation of the bound mRNAs at the synaptic junctions. Such imbalance in turn causes permanent changes in synaptic activity and ultimately mental retardation.

FIGURE 5-27 A model for the action of familial mental retardation protein (FMRP) in neurons.

(Adapted from Hin P, Warren ST: New insights into fragile X syndrome: from molecules to neurobehavior. Trends Biochem Sci 28:152, 2003.)

Although demonstration of an abnormal karyotype led to the identification of this disorder, PCR-based detection of the repeats is now the method of choice for diagnosis. With Southern blot analysis, distinction between premutations and mutations can be made prenatally as well as postnatally. Hence, this technique is valuable not only for establishing the diagnosis, but also for guiding genetic counseling. These techniques are described later.

Prader-Willi Syndrome and Angelman Syndrome

Prader-Willi syndrome is characterized by mental retardation, short stature, hypotonia, profound hyperphagia, obesity, small hands and feet, and hypogonadism.⁷⁰ In 65% to 70% of cases, an interstitial deletion of band q12 in the long arm of chromosome 15, del(15)(q11.2q13), can be detected. In most cases the breakpoints are the same, causing a 5-Mb deletion. It is striking that in all cases the deletion affects the paternally derived chromosome 15. In contrast with the Prader-Willi syndrome, patients with the phenotypically distinct Angelman syndrome are born with a deletion of the same chromosomal region derived from their mothers. Patients with Angelman syndrome are also mentally retarded, but in addition they present with ataxic gait, seizures, and inappropriate laughter. Because of their laughter and ataxia they have been referred to as “happy puppets.”⁷¹ A comparison of these two syndromes clearly demonstrates the parent-of-origin effects on gene function.

The molecular basis of these two syndromes lies in the genomic imprinting (Fig. 5-29). It is known that a gene or set of genes on maternal chromosome 15q12 is imprinted (and hence silenced), and thus the only functional allele(s) are provided by the paternal chromosome. When these are lost as a result of a deletion, the person develops Prader-Willi syndrome. Conversely, a distinct gene that also maps to the same region of chromosome 15 is imprinted on the paternal chromosome. Only the maternally derived allele of this gene is normally active. Deletion of this maternal gene on chromosome 15 gives rise to the Angelman syndrome. Molecular studies of cytogenetically normal patients with the Prader-Willi syndrome (i.e., those without the deletion) have revealed that they have two maternal copies of chromosome 15. Inheritance of both chromosomes of a pair from one parent is called uniparental disomy. The net effect is the same (i.e., the person does not have a functional set of genes from the [nonimprinted] paternal chromosomes 15). Angelman syndrome, as might be expected, can also result from uniparental disomy of paternal chromosome 15.

FIGURE 5-29 Diagrammatic representation of Prader-Willi and Angelman syndromes.

The genetic basis of these two imprinting disorders is now being unraveled. In the Angelman syndrome, the affected gene is a ubiquitin ligase that is involved in catalyzing the transfer of activated ubiquitin to target protein substrates. The gene, called UBE3A, maps within the 15q12 region, is imprinted on the paternal chromosome, and is expressed from the maternal allele primarily in specific regions of the brain.⁷² The imprinting is tissue-specific in that UBE3A is expressed from both alleles in most tissues. In approximately 10% of cases, Angelman syndrome occurs not as a result of imprinting but of a point mutation in the maternal allele, thus establishing a firm link between the UBE3A gene and Angelman syndrome. In contrast to Angelman syndrome, no single gene has been implicated in Prader-Willi syndrome. Instead, a series of genes located in the 15q11.2–q13 interval (which are imprinted on the maternal chromosome and expressed from the paternal chromosome) are believed to be involved. These include a gene that encodes small nuclear riboprotein N, which controls gene splicing and is expressed highly in the brain and heart. Loss of small nuclear riboprotein N function is believed to contribute to Prader-Willi syndrome. Molecular diagnosis (see later) of these syndromes is based on assessment of methylation status of marker genes and FISH.

The importance of imprinting is not restricted to rare chromosomal disorders. Parent-of-origin effects have been identified in a variety of inherited diseases, such as Huntington disease and myotonic dystrophy and in tumorigenesis.

Direct Detection of DNA Sequence Alterations by DNA Sequencing

DNA can be sequenced to obtain a readout of the order of nucleotides, and by comparison with a normal (wild-type) sequence, mutations can be identified. The ready availability of Sanger di-deoxynucleotide sequencing and automated capillary electrophoresis allows thousands of base pairs of genomic DNA to be routinely sequenced in a matter of hours.⁷⁵ The genes mutated in hundreds of mendelian disorders have been identified, and definitive diagnosis is possible by direct sequencing in most of them. Some disorders, most with recessive inheritance, are associated with a limited number of recurrent mutations, such as cystic fibrosis. Many others, especially those with dominant inheritance, can have mutations throughout the gene-coding region. Challenges to sole use of gene sequencing for the diagnosis of such diseases include the difficulty and high cost of analyzing large genes. For example, the gene associated with Duchenne muscular dystrophy possesses 79 exons, and the FBN1 gene mutated in Marfan syndrome possesses 65 exons; the sequencing of these genes in their entirety can be prohibitively expensive with current methodologies. Among other difficulties, it is not uncommon to detect sequence alterations of unknown significance, which cannot be definitively determined to be pathogenic in the absence of any functional data.

However, this picture is changing with remarkable speed. Rapidly advancing technology will make large-scale germ line sequencing applications feasible and may lead in the not distant future to the routine sequencing of entire individual human genomes. One high-throughput technology uses gene chips (microarrays) to sequence genes or portions of genes.⁷⁶ Short sequences of DNA (oligonucleotides) that are complementary to the wild-type sequence and to known mutations are “tiled” adjacent to each other on the gene chip, and the DNA sample to be tested is hybridized to the array (Fig. 5-30). Before hybridization the sample is labeled with fluorescent dyes. The hybridization (and consequently, the fluorescent signal emitted) will be strongest at the oligonucleotide that is complementary to wild-type sequence if no mutations are present, while the presence of a mutation will cause hybridization to occur at the complementary mutant oligonucleotide. Computerized algorithms can then rapidly “decode” the DNA sequence for hundreds of thousands of base pairs of sequence from the fluorescent hybridization pattern on the chip, and identify potential mutations. Perhaps the most exciting recent advance is technology termed “next-generation” sequencing, which involves PCR performed in an oil emulsion that allows over one million individual PCR reactions at once.⁷⁷ While at present very costly, over one billion nucleotides (one third of the human genome!) can be sequenced per run. Bioinformatics challenges of handling and interpreting such massive amounts of data are currently staggering, and much effort is devoted to such analysis.

FIGURE 5-30 Microarray-based DNA sequencing. A, A low-power digitized scan of a “gene chip” that is no larger than a nickel in size but is capable of sequencing thousands of base pairs of DNA. High-throughput microarrays have been used for sequencing whole organisms (such as viruses), organelles (such as mitochondria), and entire human chromosomes. B, A high-resolution view of the gene chip illustrates hybridization patterns corresponding to a stretch of DNA sequence. Typically, a computerized algorithm is available that can convert the individual hybridization patterns across the entire chip into actual sequence data within a matter of minutes (“conventional” sequencing technologies would require days to weeks for such analysis). Here, the upper sequence is the reference (wild-type) sequence, while the lower one corresponds to the test sample sequence. As shown, the computerized algorithm has identified a C→G mutation in the test sample.

(Adapted from Maitra A et al.: The Human MitoChip: a high-throughput sequencing microarray for the mitochondrial mutation detection. Genome Res 14:812, 2004.)

Detection of DNA Mutations by Indirect Methods

There are a large number of molecular techniques that detect DNA mutations without direct sequencing. Their development is driven by lower costs and higher throughput.

FIGURE 5-31 Allele-specific PCR for mutation detection in a heterogeneous sample containing an admixture of normal and mutant DNA. Nucleotides complementary to the mutant and wild-type nucleotides at the queried base position are labeled with different fluorophores, such that incorporation into the resulting PCR product yields fluorescent signals of varying intensity based on the ratio of mutant to wild-type DNA present.

FIGURE 5-32 Diagnostic application of PCR and Southern blot analysis in fragile-X syndrome. With PCR the differences in the size of CGG repeats between normal and premutation give rise to products of different sizes and mobility. With a full mutation, the region between the primers is too large to be amplified by conventional PCR. In Southern blot analysis the DNA is cut by enzymes that flank the CGG repeat region, and is then probed with a complementary DNA that binds to the affected part of the gene. A single small band is seen in normal males, a band of higher molecular weight in males with premutation, and a very large (usually diffuse) band in those with the full mutation.

Polymorphisms and Genome-Wide Analyses

As described earlier, linkage analysis utilizing DNA of affected families has been used to detect the presence of genes with large effects and high penetrance, the kind that give rise to Mendelian diseases. Similar analyses of complex (multifactorial) disorders, however, have been unsuccessful since conventional linkage studies lack the statistical power to detect variants with small effects and low penetrance, which are typical of the genes that contribute to complex disorders.

These limitations appear to have been overcome through genome-wide association studies (GWAS), a powerful method of identifying genetic variants that are associated with an increased risk of developing a particular disease.⁷⁸ Such variants may themselves be causative, or may be in linkage disequilibrium with other genetic variants that are responsible for the increased risk. In GWAS large cohorts of patients with and without a disease (rather than families) are examined across the entire genome for genetic variants or polymorphisms that are over-represented in patients with the disease. This identifies regions of the genome that contain a variant gene or genes that confer disease susceptibility. The causal variant within the region is then provisionally identified using a “candidate gene” approach, in which genes are selected based on how tightly they are associated with the disease and whether their biologic function seems likely to be involved in the disease under study. For example, a variant in a gene whose product regulates vascular smooth muscle tone (e.g., angiotensinogen) is a strong candidate to influence the risk of hypertension. As might be imagined, however, some linked genes would not have been expected to be associated with particular diseases based on prior knowledge; such surprises are one of the benefits of the unbiased, systematic nature of GWAS.

GWAS have been enabled by two major technological breakthroughs. First is the completion of the so-called “HapMap” project, which has provided more complete linkage disequilibrium patterns in three major ethno-racial groups, based on genome-wide single nucleotide polymorphism (SNP) mapping. The entire human genome can now be divided into blocks known as “haplotypes,” which contain varying numbers of contiguous SNPs on the same chromosome that are in linkage disequilibrium and hence inherited together as a cluster. As a result, rather than querying every single SNP in the human genome, it is possible to glean comparable information about shared DNA by simply looking for shared haplotypes, using single or a small number of SNPs that “tag” or identify a specific haplotype. Second, it is now possible to simultaneously genotype hundreds of thousands to a million SNPs at one time, in a cost-effective way, using high-density SNP-chip technology. Figure 5-34 demonstrates how information from the publicly available “HapMap” is utilized to manufacture SNP chips that can query genome-wide haplotypes in an unbiased manner. Thereafter, DNA from a cohort of individuals with a defined trait (say, hypertension) is analyzed using SNP chips for haplotypes that are overrepresented compared to individuals without the trait (i.e., in controls). This is followed by the “candidate gene” approach described above to further localize the causal gene (and in some instances, the functional polymorphism within that gene), associated with the trait.

FIGURE 5-34 General scheme for conducting a genome wide association study (GWAS). Using the publicly available “HapMap” data, the human genome is divided into “haplotypes” or regions of contiguous DNA inherited as a block, each identified by one or a few “tag” SNPs that identify the haplotype. In the example shown, locus 1 contains three haplotypes defined by different combinations of SNPs, where white signifies the more common “normal” sequence and each color designates a different SNP; thus, these haplotypes can be distinguished by assaying for only the blue and purple “tag” SNPs. Thereafter, high density SNP chips are constructed that contain these “tag” SNPs, in order to enable an unbiased genome-wide assessment of shared haplotypes between disease and control populations. Of note, “disease” refers to any defined phenotype, and could pertain to an actual disease entity like hypertension, or simply a quantitative trait like hair or eye color. Next, DNA obtained from the two cohorts is analyzed for overrepresented SNPs in the disease population (“cases”) versus the control samples—this is known as a case-control study. The most significant shared genomic regions of interest are then examined for candidate genes of interest—an example shown here in a search for loci associated with hypertension is angiotensinogen, a gene on chromosome 1 whose product regulates vascular smooth muscle tone. The final step is to perform a second case control study, this time using SNPs located within the gene of interest in order to confirm or refute the association with the trait, often in an independent population from the one in which the initial GWAS was conducted. In this example, individual SNPs within angiotensinogen gene are denoted as red vertical bars, and these SNPs will be tested in the second round of case-control study.

(Modified from Mathew CG: New links to the pathogenesis of Crohn disease provided by genome-wide association scans. Nat Rev Genet 9(1):9–14, 2008.)

In addition to shedding light on some of the most frequent human ailments such as diabetes, hypertension, coronary artery disease, schizophrenia and other mental disorders, and asthma, GWAS have also led to the identification of genetic loci that modulate common quantitative traits in humans, such as height, body mass, hair and eye color, and bone density. An updated catalog of published GWAS is maintained by the National Human Genome Research Institute (www.genome.gov), with the list currently at >200 studies and growing. The power of GWAS is underscored by the fact that within a very short time, nearly a dozen genes that confer risk for type 2 diabetes have been identified, of which one in particular, TCF7L2, has emerged as a strong candidate gene (see Chapter 24 for a detailed discussion).

With the incrementally lowered costs for genotyping individual patients for SNPs that might render them “at-risk” for a variety of multifactorial disease over their lifetime, there is emerging concern in the biomedical community that such information would be utilized for discrimination in the workplace or by healthcare agencies. In the United States, a law was passed in 2008 that explicitly prohibits discrimination based on an individual’s genetic makeup.

Southern Blotting

Changes in the structure of specific loci can be detected by Southern blotting, which involves hybridization of radiolabeled sequence-specific probes to genomic DNA that has been first digested with a restriction enzyme and separated by gel electrophoresis. The probe usually detects one germ line band in normal individuals. Importantly, a normal DNA sample is required to compare the pattern of the DNA in question. With the advent of FISH and microarray technology, Southern blotting is rarely used but remains useful in the detection of large-trinucleotide-expansion diseases including the fragile-X syndrome (see Fig. 5-32), and in detection of clonal immunoglobulin gene rearrangements in the diagnosis of lymphoma. The latter is being replaced by PCR-based methods.

Fluorescence in Situ Hybridization

FISH uses DNA probes that recognize sequences specific to particular chromosomal regions. As part of the Human Genome Project, large libraries of bacterial artificial chromosomes that span the entire human genome were created. The human DNA insert in these clones is on the order of 100,000–200,000 base pairs, which defines the limit of resolution of FISH for identifying chromosomal changes. These DNA clones are labeled with fluorescent dyes and applied to metaphase spreads or interphase nuclei. The probe hybridizes to its homologous genomic sequence and thus labels a specific chromosomal region that can be visualized under a fluorescent microscope. The ability of FISH to circumvent the need for dividing cells is invaluable when a rapid diagnosis is warranted (e.g., when deciding to treat a patient with acute myeloid leukemia with retinoic acid, which is only effective in a particular subtype with a chromosomal translocation involving the retinoic acid receptor gene [Chapter 14]). FISH can be performed on prenatal samples (e.g., cells obtained by amniocentesis, chorionic villus biopsy, or umbilical cord blood), peripheral blood lymphocytes, touch preparations from cancer biopsies, and even archival tissue sections. FISH has been used for detection of numeric abnormalities of chromosomes (aneuploidy) (see Fig. 5-20); for the demonstration of subtle microdeletions (see Fig. 5-22) or complex translocations not detectable by routine karyotyping; for analysis of gene amplification (e.g., HER2/NEU in breast cancer or N-MYC amplification in neuroblastomas); and for mapping newly isolated genes of interest to their chromosomal loci. Chromosome painting is an extension of FISH, whereby probes are prepared for entire chromosomes. The number of chromosomes that can be detected simultaneously by chromosome painting is limited by the availability of fluorescent dyes that emit different wavelengths of visible light. This limitation has been overcome by the introduction of spectral karyotyping (also called multicolor FISH). By using a combination of five fluorochromes and appropriate computer-generated signals, the entire human genome can be visualized (Fig. 5-35). So powerful is spectral karyotyping that it might well be called “spectacular karyotyping.”

FIGURE 5-35 FISH studies using multicolor FISH in a child with an undetermined abnormality. This technique uses ratio-labeled probes labeled with 23 distinct mixtures of 5 fluorophores to create a unique “color” for each chromosome. This analysis revealed a derivative chromosome 9, with 9p containing additional material from 22q.

(Courtesy of Dr. Stuart Schwartz, Department of Pathology, University of Chicago, Chicago, IL.)

Array-Based Comparative Genomic Hybridization (Array CGH)

It is obvious from the preceding discussion that FISH requires prior knowledge of the one or few specific chromosomal regions suspected of being altered in the test sample. However, genomic abnormalities can also be detected without prior knowledge of what these aberrations may be, using a global strategy such as array CGH. In array CGH the test DNA and a reference (normal) DNA are labeled with two different fluorescent dyes (most commonly Cy5 and Cy3, which fluoresce red and green, respectively) (Fig. 5-36). The differentially labeled samples are then hybridized to a glass slide spotted with DNA probes that span the human genome at regularly spaced intervals, and usually cover all 22 autosomes and the X chromosome. If the contributions of both samples are equal for a given chromosomal region (i.e., the test sample is diploid), then all spots on the array will fluoresce yellow (the result of an equal admixture of green and red dyes). In contrast, if the test sample shows an excess of DNA at any given chromosomal region (such as resulting from an amplification), there will be a corresponding excess of signal from the dye with which this sample was labeled. The reverse will be true in the event of a deletion, with an excess of the signal used for labeling the reference sample. Amplifications and deletions in the test sample can now be significantly better localized, often down to a few thousand base pairs. Newer arrays provide even higher resolution with more than 100,000 probes per array, and are at present being used to uncover copy number abnormalities in a variety of diseases, from cancer to autism. Array CGH is regularly performed in cases of mental retardation–developmental delay of unknown etiology or in children with dysmorphic features with negative karyotypes.

FIGURE 5-36 A, Array CGH is performed by hybridization of fluorescently labeled “test” DNA and “control” DNA on a slide that contains thousands of probes corresponding to defined chromosomal regions across the human genome. The resolution of most currently available array CGH assays is in the order of about 200 to 500 kilobases. Higher power view of the array demonstrates copy number aberrations in the “test” sample (Cy5, red), including regions of amplification (spots with excess of red signal) and deletion (spots with excess of green signal); yellow spots correspond to regions of normal (diploid) copy number. B, The hybridization signals are digitized, resulting in a virtual karyotype of the genome of the “test” sample. In the illustrated example, array CGH of a cancer cell line identifies an amplification on the distal long arm of chromosome 8, which corresponds to increased number of the oncogenic MYC.

(A, From Snijders AM et al.: Assembly of microarrays for genome-wide measurement of DNA copy number. Nat Genet 29:263, 2001. Web Figure A, Copyright 2001. Reprinted by permission from Macmillan Publishers Ltd.)

As discussed earlier in the chapter, CNVs are a recently discovered source of genetic polymorphism, which was uncovered using array CGH technology. While intriguing in terms of understanding the marked differences between individual genomes, they can be problematic in the clinical interpretation of array CGH data.⁷⁹ There are usually many CNVs observed when comparing any two genomes encompassing millions of bases of DNA. Deciding whether a specific change is a benign polymorphism or a critical disease-causing duplication or deletion can be difficult. Databases of CNVs now exist that are very helpful guides in deciding on the relevance of questionable CNVs. Another limitation of existing array CGH platforms is that they cannot detect balanced translocations, since there is a rearrangement but no genetic material is gained or lost. Nonetheless, the vastly superior sensitivity of molecular approaches should make assays such as array CGH first-line genomic diagnostic tests that have the potential of replacing traditional karyotyping.