Genetic Disease

All of the 30,000 or more human genes can mutate, and many of the mutant alleles cause or contribute to the risk of genetic disease. In some instances, pathology occurs if only one of the two alleles at a particular locus is mutant. In other instances, both alleles must be defective. There are several sources of genetic heterogeneity in disease. Different alleles at a single locus may produce different degrees of pathology or even quite different disorders. In contrast, mutations at different loci can produce the same phenotypic changes. Mutations that arise in somatic cells can also produce disease, but they cannot be transmitted to offspring.

 1. Disease, Genes, and Mutations

 A discussion of genetic disease should begin with a definition of disease. Most persons use the term to indicate a departure from normal function or health. This suggests that a clear distinction exists between the healthy state and disease. Although this dichotomy may be useful to insurance companies, it is not very descriptive of biological reality. Each of us is uniquely programmed genetically to cope with greater or less success with what the environment presents to us. And ultimately we are programmed not to survive. This perception was expressed by the great British physician A.E. Garrod in his 1931 monograph The Inborn Factors in Disease, “It is permissible to suppose that, as in the production of those wider departures from the standard of health which receive the name of diseases, both internal and external factors are at work, so also in the causation of those lesser departures, which it is difficult to name or classify, and which may be spoken of as trifling ailments, ¼”

Diseases, then, may be thought of as deviations from normality that are sufficient to interfere with function of the organism. Many such deviations are primarily associated with specific genotypes. Others are combinations of genotypic and environmental variation. Although all are of interest, our knowledge is largely limited to those diseases that are well outside the range of normal function. In the case of inherited diseases, this means that we are largely limited to considering mutations at single loci that have a large impact on the phenotype. Results from the Human Genome Project, coupled with more sophisticated statistical techniques, are making possible the detection of smaller and more complex genetic contributions to disease risk. As of May 2001, some 9,200 loci had been identified in the human genome. Over 7,000 loci had been assigned to specific chromosomes. Many are associated with a specific inherited variation or disease. Others are defined by the existence of their protein product or messenger RNA transcript. In some instances, a gene first observed in a nonhuman species is found to have a homolog in humans. The total number of loci is estimated to be approximately 30,000. This number is derived from the complete sequence of human DNA and is much lower than earlier estimates.

 All genes have the potential to mutate. It is not known, however, what portion of the total number have the potential to mutate to a form that leads to recognizable disease. This is because many genes may cause early embryonic death if mutations occur in them. For example, any mutant form of a gene that is critical in early development may be eliminated without being observed.

 2. Genotypes and Disease

The ultimate role of genes is the production of proteins of the right kind and amount in the right places and times. The vast majority of genes code for protein structures, and associated with each gene are regulatory elements—stretches of DNA that interact with other proteins to determine the amount of transcription of the gene. Most of the understanding of inherited diseases is in terms of function or malfunction of specific proteins.

 Whether a mutant gene causes disease depends on several factors. Insertion or deletion of a single nucleotide in the coding region causes a shift in the reading frame during translation, leading to complete absence of the gene product. Substitution of one nucleotide for another may be equally detrimental if it causes substitution of the wrong amino acid at a functional site. Or it may have no effect at all if the amino acid substitution does not modify the function of the protein. Many such variations exist in the normal human population. Some cause small differences in function that we consider normal variation. Mutations in regulatory regions of genes may change the amount of gene product—too little or too much—but not the structure of the protein.

The phenotypic effect of these changes in gene structure depend on the role of the gene. The amounts of many enzymes normally produced are in excess of the amount needed to maintain normal metabolism. These loci are described as haplosufficient, meaning that a heterozygote that has one normal and one nonfunctional allele is phenotypically normal. Many enzyme deficiencies are inherited as recessive traits because of the need for both alleles to be nonfunctional for disease to occur. Haplosufficiency also characterizes other types of proteins, such as cell cycle regulators and many receptors.

 Haploinsufficiency is also observed. For example, one normal copy of a b-globin gene in hemoglobin is insufficient for manufacture of adequate amounts of b-globin, and anemia results. Again, in the case of low-density lipoprotein (LDL) receptors, one normal allele is insufficient to produce the number of functional receptors required to maintain normal levels of blood cholesterol. The transmission of these traits produces pedigrees characteristic of dominant inheritance.

One group of mutations is designated dominant negative. The name derives from the fact that the mutant allele interferes with functions of the normal allele. In a typical example, the proteins form complexes either involving one type of subunit (i.e., the product of one locus) or involving multiple subunits from different loci. The product of the mutant allele may enter into the complex but

interfere with function of the complex. In the simplest case of a homodimer formed from two identical subunits, only the dimer composed of two normal subunits is functional. Both the mixed dimer and the dimer composed of two mutant subunits are nonfunctional. Because heterozygotes have reduced function, the transmission pattern is that of a dominant trait.

A variation on the previous example occurs when the mutant allele produces no protein product at all. In this case, the only subunits are produced from the normal allele, and all dimers or other multimers would be functional. This is observed in some collagen disorders, where deletion of an allele causes no phenotypic effect, the normal allele being haplosufficient. Only homozygotes for the mutant allele would generate a mutant phenotype. However, if the mutation is a minor variation in the amino acid sequence, this may interfere with production of normal collagen. The trait would appear as dominant in pedigrees.

Not all mutations or inherited syndromes involve single genes. In contiguous gene syndromes, small chromosomal deletions remove multiple closely-linked genes. The person therefore has only one copy of the genes in that segment rather than the normal two. Any genes that are haploinsufficient will affect the phenotype. The WAGR syndrome involves deletion of a small region of chromosome 11 that includes genes for Wilms' tumor, Aniridia, Genital anomalies, and Retardation of growth. The WAGR syndrome is therefore the sum of the single gene defects.

 3. Genetic Heterogeneity

If one considers all the possible nucleotide substitutions, insertions, and deletions in a gene that is many kilobases in size, the number of potential alleles at any one locus is enormous. In practice, one would never expect to see all possible mutant alleles in a population of finite size. However, in the case of other loci that have been extensively investigated, such as the globins, phenylalanine hydroxylase, cystic fibrosis, and Duchenne muscular dystrophy (DMD), hundreds of different mutant alleles have been identified. For the most part, these are individually very rare. In the case of the DMD locus, mutations are quickly eliminated by natural selection, and most mutations are found on analysis to be recent and to differ from other known mutations. In the case of other loci, a particular one or two mutations will be common, reflecting an increase in the frequency through genetic drift, founder effect, or heterozygote advantage. In the case of cystic fibrosis, some 70% of the mutant alleles in persons of European ancestry involve the identical three-nucleotide deletion, one that must have occurred many thousands of years ago to be so widely distributed.

Such allelic heterogeneity often translates into phenotypic heterogeneity. This is well illustrated in mutations of the DMD gene. Many of the mutations are associated with a protein product ) dystrophin) that has some activity. This results in a milder form of the disease (Becker muscular dystrophy). Similar variations in severity of cystic fibrosis have been associated with different allelic combinations.

Because of the very large number of alleles at some loci, the term homozygous is often a misnomer. For rare recessive traits, it means that the two alleles are defective in their function but not that they are identical. A person with two different alleles that are functionally similar is described as a compound heterozygote. Unless the genes have been analyzed at the molecular level, one cannot be certain of true homozygosity.

 Locus heterogeneity also occurs. This is the situation in which mutations at any of several loci produce the same phenotype. For example, early onset familial Alzheimer disease (FAD) is transmitted as a dominant trait and can be caused by mutations of any of three loci on three different chromosomes. The primary defect is presumably different at the molecular level, but the phenotypic results are the same.

An uncommon form of genetic heterogeneity is the production of different pathological conditions depending on where mutations occur in a gene. An example is the androgen receptor gene (AR), located on the X chromosome. The androgen receptor, when bound to testosterone and dihydrotestosterone, acts as a transcription factor in the nucleus. A number of mutations are known that interfere with binding and prevent the androgens from acting on target organs. The resulting disorder is known as androgen resistance. Mutations at other sites in the AR gene cause spinobulbar muscular atrophy.

 Different populations also vary in their complements of mutant alleles. The mutations that cause cystic fibrosis occur in some 2% of persons of European origin but occur at much lower frequencies in other populations. Tay-Sachs occurs at much higher frequencies among Ashkenazi Jews. These high frequencies of detrimental genes probably arose through genetic drift and founder effect, although selective advantage may account for the high frequencies in some instances, as in sickle cell anemia in Africans. Even when the overall frequencies of mutant alleles are similar in populations, the specific alleles may differ.

4. Genetic Diseases of Somatic Cells

When inherited diseases are discussed, the traditional reference is to diseases that are transmitted from one generation to the next. Many show simple dominant or recessive inheritance according to Mendelian rules. Others are more complex, being influenced by variation at multiple loci and by environment also. The variations that can be followed in pedigrees occur in the germ lines and are therefore transmissible, even though the phenotype that we observe is based on somatic cell function.

 Somatic cells, of course, have the same array of genes that are in the germ line, and they are subject to mutation as well as epigenetic alterations. The great importance of somatic cell genetics became apparent from the demonstration that mutations in somatic cells are essential parts of carcinogenesis. In the simplest case of retinoblastoma, a malignant tumor occurs when both alleles of the RB locus become inactive through mutation in a retinoblast. The normal restraints on cell growth no longer occur, and that cell forms a clone of cells that constitutes the tumor. The genotype of the original germline has changed in the clone of somatic cells but not in the germ cells. Retinoblastoma per se cannot be transmitted to offspring.

It has long been established that the risk of retinoblastoma is often transmitted as a simple dominant trait. In this case, one of the two required mutant alleles is transmitted in the germ line, and only one additional mutation is required in a retinoblast in order for the tumor to arise. This pattern has been established for a number of cancers for which high risk is found in certain families.

Somatic mutation must occur frequently at many loci. In most cases, there is no way to recognize an individual cell that has mutated. It is possible in the case of cancer because of the expansion of a clonal population from the original mutant cell. There are other instances in which somatic mutations also appear to be an essential part of the disease. For example, polycystic kidney disease is a dominantly inherited condition in which many renal cysts form. Comparison of the genotypes of the cysts with those of normal cells from the same person indicates that genetic alterations have occurred in the cysts and that each is a clone. As in cancers, clonal expansion makes it possible to detect the mutation that has occurred in a single somatic cell. Apparently two mutations are necessary, and, as in retinoblastoma, one can be transmitted through the germ line to all cells of the developing embryo. The study of somatic cell genetics is likely to be an area of great future interest as analytical techniques become ever more sensitive.