Clinicians at one time concerned themselves only with what they could discover by bedside evaluation and laboratory investigation. In the parlance of genetics, the patient's symptoms and signs constitute his or her phenotype. Now the means are at hand for defining a person's genotype, the actual information content inscribed in the 2 m of coiled DNA present in each cell of the body—or half that amount in every mature ovum or sperm. Most phenotypic characteristics—and this includes diseases as well as human traits such as personality, adult height, and intelligence—are to some extent determined by the genes. The importance of the genetic contribution varies widely among human phenotypes, and methods are only now being developed to identify the genes involved in complex traits and most common diseases. Moreover, the importance of interactions between environment and genotype in producing phenotypes cannot be overstated despite the obscurity of the actual mechanisms.

The billions of nucleotides in the nucleus of a cell are organized linearly along the DNA double helix in functional units called genes, and each of the 20,000–23,000 human genes is accompanied by various regulatory elements that control when it is active in producing messenger RNA (mRNA) by a process called transcription. In most situations, mRNA is transported from the nucleus to the cytoplasm, where its genetic information is translated into proteins, which perform the functions that ultimately determine phenotype. For example, proteins serve as enzymes that facilitate metabolism and cell synthesis; as DNA binding elements that regulate transcription of other genes; as structural elements of cells and the extracellular matrix; and as receptor molecules for intracellular and intercellular communication. DNA also encodes many small RNA molecules that serve functions still being defined, including regulating gene transcription and interfering with the translational capacity of some mRNAs.

Chromosomes are the vehicles in which the genes are carried from generation to generation. Each chromosome is a complex of protein and nucleic acid in which an unbroken double helix of DNA is coiled and supercoiled into a space many orders of magnitude less than the extended length of the DNA. Within the chromosome there occur highly complicated and integrated processes, including DNA replication, recombination, and transcription. In the nucleus of each somatic cell, humans normally have 46 chromosomes, which are arranged in 23 pairs. One of these pairs, the sex chromosomes X and Y, determines the sex of the individual; females have the pair XX and males the pair XY. The remaining 22 pairs are called autosomes (Figure e2–1). In addition to these nuclear chromosomes, each mitochondrion—found in varying numbers in the cytoplasm of all cells—contains multiple copies of a small chromosome. This mitochondrial chromosome encodes a few of the proteins for oxidative metabolism and all of the transfer RNAs used in translation of proteins within this organelle. Mitochondrial chromosomes are inherited almost entirely from the cytoplasm of the fertilized ovum and are therefore maternal in origin.

View Large Figure e2–1 Add to 'My Saved Images'  Normal karyotype of a human male. Prepared from cultured amniotic cells and stained with Giemsa stain. About 400 bands are detectable per haploid set of chromosomes.

In all somatic cells, the 44 autosomes and one of the X chromosomes are transcriptionally active. In males, the active X is the only X; portions of the Y chromosome are also active. In females, the requirement for dosage compensation (to be equivalent to the situation in males) is satisfied by inactivation of most of one X chromosome early in embryogenesis. This process of X chromosomal inactivation, although incompletely understood, is known to be random, so that on average, in 50% of a female's cells, one of the X chromosomes will be active, and in the other 50% the homologous member of the pair will be active. The phenotype of the cell is determined by which genes on the chromosomes are active in producing mRNA at any given time.

Transcriptional control is highly complex. Among other processes, an allele of some genes is prevented from transcribing mRNA through the process of imprinting. Imprinting occurs in the gamete, typically by adding a methyl group to cytosine nucleotides in the regulatory region of the allele. Methylation results in downregulation of the allele and is gamete-specific. Thus, the allele of some genes inherited from the father is permanently turned off, while the alleles of other genes inherited from the mother are similarly inhibited. Other processes can affect expression of specific alleles, including biochemical modification of certain histones. Most strikingly, these effects can persist across generations and be influenced by the environment. One example of the latter is the reduction in imprinting seen in times of famine, when methyl groups are deficient in the diet. This field of non-Mendelian inheritance is termed epigenetics.

Genes & Chromosomes

In all genes, information is contained in parcels called exons, which are interspersed with stretches of DNA called introns that do not encode any information about the protein sequence. However, introns may contain genetic regulatory sequences, and some introns are so large that they encode an entirely distinct gene.

The exact location of a gene on a chromosome is its locus, and the array of loci constitutes the human gene map. A variation of this map, identifying selected loci known to be involved in human disease, is shown in Figure e2–2. The difference in the higher resolution of the ordering of genes achievable by molecular techniques (such as linkage analysis) compared to cytogenetic techniques (such as visualization of small defects) is substantial, though the gap is narrowing. The chromosomes in the "standard" karyotype shown in Figure e2–1 have about 450 visible bands; under the best of cytologic and microscopic conditions, a total of about 1600 bands can be seen. But even in this extended configuration, each band contains dozens—sometimes hundreds—of individual genes. Thus, loss (deletion) of a small band will involve loss of many coding sequences and will have diverse effects on the phenotype. Routine cytogenetic techniques of visualizing chromosomes are largely being supplanted by array-based approaches.

View Large Figure e2–2 Add to 'My Saved Images'  A partial "morbid map" of the human genome. Shown next to the ideogram of the human X and Y chromosomes are representative mendelian disorders caused by mutations at that locus. Over 460 phenotypes have been mapped to the X chromosome and 8 to the Y chromosome. (Courtesy of V. McKusick and J. Strayer.)

The number and arrangement of genes on homologous chromosomes are identical even though the actual coding sequences of homologous genes, or the number of copies of those genes, may not be. Homologous copies of a gene are termed alleles. In comparing alleles, it must be specified at what level of analysis the comparison is being made. When alleles are truly identical—in that their coding sequences and the number of copies are invariant—the individual is homozygous at that locus. At a coarser level, the alleles may be functionally identical despite subtle variations in nucleotide sequence—with the result either that the proteins produced from the two alleles are identical or that whatever differences there may be in amino acid sequence will have no bearing on the function of the protein. If the individual is being analyzed at the level of the protein phenotype, allelic homozygosity would again be an apt descriptor. However, if the analysis were at the level of the DNA—as occurs in nucleotide sequencing—then, despite functional identity, the alleles would be viewed as different and the individual would be heterozygous for that locus. Heterozygosity based on differences in the protein products of alleles has been detectable for decades and was the first hard evidence concerning the high degree of human biologic variability. In the past decade, analysis of DNA sequences has shown genetic variability to be much more common—differences in nucleotide sequence between individuals occur about once every 1100 nucleotides. Much longer sequences can also be present in varying number of copies, or they can be entirely deficient, often with no obvious effect on the phenotype. The identification of such copy number variants has been one of the most exciting and useful discoveries in human genetics in the past several years. Because of evolutionary selection against deleterious sequence changes, DNA variation in coding regions of genes occurs once every 2000 nucleotides, and less than half of those variants cause a change in an amino acid. However, the level of sequence variation among humans, regardless of their ethnic origins, is much less common (3- to 10-fold) than in our primate ancestors.

Mutation

Allelic heterozygosity most often results when different alleles are inherited from the egg and the sperm, but it also occurs as a consequence of spontaneous alteration in nucleotide sequence (mutation). Genetic change occurring during formation of an egg or a sperm is called a germinal mutation. When the change occurs after conception—from the earliest stages of embryogenesis to dividing cells in the body of the oldest adult—it is termed a somatic mutation. As is discussed below, the role of somatic mutation in the etiology of human disease is now increasingly recognized.

The coarsest type of mutation is alteration in the number or physical structure of chromosomes. For example, nondisjunction (failure of chromosome pairs to separate) during meiosis—the reduction division that leads to production of mature ova and sperms—causes the embryo to have too many or too few chromosomes, a situation called aneuploidy. Rearrangement of chromosome arms, such as occurs in translocation or inversion, is a mutation even if breakage and reunion do not disrupt any coding sequence. Thus, the phenotypic effect of gross chromosomal mutations can range from profound (as in aneuploidy) to nil.

A bit less coarse, but still detectable cytologically, are deletions of part of a chromosome. Such mutations almost always alter phenotype, because a number of genes are lost; however, a deletion may involve only a single nucleotide, whereas about 1–2 million nucleotides (1–2 megabases) must be lost before the defect can be visualized by the most sensitive cytogenetic methods short of in situ hybridization or array analysis. Deletions and duplications of substantial regions of nucleotide sequence are remarkably common among humans. Many are apparently harmless and are passed from parent to child in an autosomal dominant pattern of inheritance. Others involve one or more genes that can have subtle or profound clinical consequences. The latter are often de novo, meaning that neither parent has the copy number variation, which must have arisen during meiosis of one of the gametes.

Mutations of one or a few nucleotides in exons have several potential consequences. Changes in one nucleotide can alter which amino acid is encoded; if the amino acid is in a critical region of the protein, function might in this way be severely disturbed (eg, sickle cell disease). On the other hand, some amino acid substitutions have no detectable effect on function, and the phenotype is therefore unaltered by the mutation. Similarly, because the genetic code is degenerate (two or more different three-nucleotide sequences called codons encode the same amino acids), nucleotide substitution does not necessarily alter the amino acid sequence of the protein. Three specific codons signal termination of translation; thus, a nucleotide substitution in an exon that generates one of the stop codons usually causes a truncated protein, which is nearly always dysfunctional. Other nucleotide substitutions can disrupt the signals that direct splicing of the mRNA molecule and grossly alter the protein product. Finally, insertions and deletions of one or more nucleotides can have dramatic effects—any change that is not a multiple of three nucleotides disrupts the reading frame of the remainder of the exon—or potentially minimal effects (if the protein can tolerate the insertion or loss of an amino acid).

Mutations in introns may disrupt mRNA splicing signals or regulatory elements, or may be entirely silent with respect to the phenotype. A great deal of variation in nucleotide sequences among individuals (averaging one difference every few hundred nucleotides) resides within introns. Mutations in the DNA between adjacent genes may also be silent or may have a profound effect on phenotype if regulatory sequences are disrupted; these regulatory sequences may affect genes that are millions of bases away, presumably because of the intricate folding of the DNA molecule within chromatin.

A novel mechanism for mutation, which also helps explain clinical variation among relatives, has been discovered in myotonic dystrophy, Huntington disease, fragile X mental retardation syndrome, Friedreich ataxia, and other disorders. A region of repeated trinucleotide sequences close to or within a gene can be unstable in some families; expansion of the number of repeated units within this segment beyond a critical threshold is associated with disease, either by downregulation of that allele or the production of a defective protein. The longer the trinucleotide repeat, the more severe the phenotype, a phenomenon termed anticipation, because the condition worsens from one generation to the next.

Mutations may occur spontaneously or may be induced by environmental factors such as radiation, medication, or viral infections. Both advanced maternal and paternal age favor mutation, but of different types. In women, meiosis is completed only when an egg ovulates, and chromosomal nondisjunction is increasingly common as the egg becomes older. The risk that an aneuploid egg will result increases exponentially and becomes a major clinical concern for women older than their early 30s. In men, mutations of a subtler sort—affecting nucleotide sequences—increase with age. Offspring of men over age 40 years are at an increased risk for having mendelian conditions, primarily autosomal dominant ones.

Genes in Individuals

For some quantitative traits such as adult height or serum glucose concentration in normal individuals, it is difficult to distinguish the contributions of individual genes; this is because in general, phenotypes are the products of multiple genes acting in concert, and obviously influenced by the environment and chance. However, if one of the genes in the system is aberrant, a major departure from the "normal" or expected phenotype might arise. Whether the aberrant phenotype is serious (ie, a disease) or even recognized depends on the nature of the defective gene product and how resilient the system is to disruption. The latter point emphasizes the importance of homeostasis in both physiology and development—many mutations go unrecognized because the system can cope, even though tolerances for further perturbation might be narrowed.

In other words, most human characteristics and common diseases are polygenic, whereas many of the disordered phenotypes traditionally thought of as "genetic" are monogenic but still influenced by other loci in a person's genome.

Phenotypes due to alterations at a single gene are also characterized as mendelian, after Gregor Mendel, the monk and part-time biologist who studied the reproducibility and recurrence of variation in garden peas. He showed that some traits were dominant to other traits, which he called recessive. The dominant traits required only one copy of a "factor" to be expressed, regardless of what the other copy was, whereas the recessive traits required two copies before expression occurred. In modern terms, the mendelian factors are genes, and the alternative copies of the gene are alleles. Let A be the common (normal) allele and let a be a mutant allele at a locus: If the same phenotype is present whether the genotype is A/a or a/a, the phenotype is dominant, whereas if the phenotype is present only when the genotype is a/a, it is recessive.

In medicine, it is important to keep two considerations in mind: First, dominance and recessiveness are attributes of the phenotype, not the gene; and second, the concepts of dominance and recessiveness depend on how the phenotype is defined. To illustrate both points, consider sickle cell disease. This condition occurs when a person inherits two alleles for ^S-globin, in which the normal glutamate at position 6 of the protein has been replaced by valine; the genotype for the -globin locus is HbS/HbS, compared to the normal HbA/HbA. When the genotype is HbS/HbA, the individual does not have sickle cell disease, so this condition satisfies the criteria for being a recessive phenotype. But now consider the phenotype of sickled erythrocytes. Red cells with the genotype HbS/HbS clearly sickle—but, if the oxygen tension is reduced, so do cells with the genotype HbS/HbA. Therefore, sickling is a dominant trait.

A mendelian phenotype is characterized not only in terms of dominance and recessiveness but also according to whether the determining gene is on the X chromosome or on one of the 22 pairs of autosomes. Traits or diseases are therefore called autosomal dominant, autosomal recessive, X-linked recessive, and X-linked dominant.

Genes in Families

Since the first decade of the twentieth century, the patterns of recurrence of specific human phenotypes have been explained in terms of principles first described by Mendel in the garden pea plant. Mendel's second principle—usually referred to as his first¹—is called the law of segregation and states that a pair of factors (alleles) that determines some trait separates (segregates) during formation of gametes. In simple terms, a heterozygous (A/a) person will produce two types of gametes with respect to this locus—one containing only A and one containing only a, in equal proportions. Offspring of this person will have a 50–50 chance of inheriting the A allele and a similar chance of inheriting the a allele.

¹Mendel's first law stated that—from the perspective of the phenotype—it mattered not from which parent a particular mutant allele was inherited. For years this principle was thought to be too obvious to be codified as anybody's "law" and was therefore ignored. In fact, however, recent evidence from studies of human disorders suggests that certain genes are "processed" (imprinted) as they move through the gonad and that processing in the testis is different from that in the ovary. Thus, not only is this first mendelian principle important, it was incorrect as originally formulated from observations in peas.

The concepts of genes in individuals and in families can be combined to specify how mendelian traits will be inherited.

Autosomal Dominant Inheritance

The characteristics of autosomal dominant inheritance in humans can be summarized as follows: (1) A horizontal pattern occurs in the pedigree, with multiple generations affected (Figure e2–3). (2) Heterozygotes for the mutant allele show an abnormal phenotype. (3) Males and females are affected with equal frequency and severity. (4) Only one parent must be affected for an offspring to be at risk for developing the phenotype. (5) When an affected person mates with an unaffected one, each offspring has a 50% chance of inheriting the affected phenotype. This is true regardless of the sex of the affected parent—specifically, male-to-male transmission occurs. (6) The frequency of sporadic cases is positively associated with the severity of the phenotype. More precisely, the greater the reproductive fitness of affected persons, the less likely it is that any given case resulted from a new mutation. (7) The average age of fathers is advanced for isolated (sporadic or new mutation) cases.

View Large Figure e2–3 Add to 'My Saved Images'  A pedigree illustrating autosomal dominant inheritance. Square symbols indicate males and circles females; open symbols indicate that the person is phenotypically unaffected, and filled symbols indicate that the phenotype is present to some extent.

Autosomal dominant phenotypes are often age-dependent, less severe than autosomal recessive ones, and associated with malformations or other physical features. They are pleiotropic in that multiple, even seemingly unrelated clinical manifestations derive from the same mutation, and variable in that expression of the same mutation among people will differ.

Penetrance is a concept associated with mendelian conditions—especially dominant ones—and the term is often misused. It should be defined as an expression of the frequency of appearance of a phenotype (dominant or recessive) when one or more mutant alleles are present. For individuals, penetrance is an all-or-none phenomenon—the phenotype is either present (penetrant) or not (nonpenetrant). The term variability—not "incomplete penetrance"—should be used to denote differences in expression of an allele.

The most frequent cause of apparent nonpenetrance is insensitivity of the methods for detecting the phenotype. If an apparently normal parent of a child with a dominant condition was in fact heterozygous for the mutation, the parent would have a 50% chance at each subsequent conception of having another affected child. A common cause of nonpenetrance in adult-onset mendelian diseases is death of the affected person before the phenotype becomes evident but after transmission of the mutant allele to offspring. Thus, accurate genetic counseling demands careful attention to the family medical history and high-resolution scrutiny of both parents of a child with a condition known to be a mendelian dominant trait.

When both alleles are expressed in the heterozygote, as in blood group AB, in sickle trait (HbS/HbA), in the major histocompatibility antigens (eg, A2B5/A3B17), or in sickle-C disease (HbS/HbC), the phenotype is called codominant.

In human dominant phenotypes, the mutant allele in homozygotes is almost always more severe than in heterozygotes.

Autosomal Recessive Inheritance

The characteristics of autosomal recessive inheritance in humans can be summarized as follows: (1) A horizontal pattern occurs in the pedigree, with a single generation affected (Figure e2–4). (2) Males and females are affected with equal frequency and severity. (3) Inheritance is from both parents, each a heterozygote (carrier) and each usually clinically unaffected. (4) Each offspring of two carriers has a 25% chance of being affected, a 50% chance of being a carrier, and a 25% chance of inheriting neither mutant allele. Thus, two-thirds of all clinically unaffected offspring are carriers. (5) In matings between individuals, each with the same recessive phenotype, all offspring will be affected. (6) Affected individuals who mate with unaffected individuals who are not carriers have only unaffected offspring. (7) The rarer the recessive phenotype, the more likely it is that the parents are consanguineous (related).

View Large Figure e2–4 Add to 'My Saved Images'  A pedigree illustrating autosomal recessive inheritance. (Symbols as in Figure e2–3.)

Autosomal recessive phenotypes are often associated with deficient activity of enzymes and are thus termed inborn errors of metabolism. Such disorders include phenylketonuria, Tay-Sachs disease, and the various glycogen storage diseases and tend to be more severe, less variable, and less age-dependent than dominant conditions.

When an autosomal recessive condition is quite rare, the chance that the parents of affected offspring are consanguineous is increased. As a result, the prevalence of rare recessive conditions is high among inbred groups such as the Old Order Amish. On the other hand, when the autosomal recessive condition is common, the chance of consanguinity between parents of cases is no higher than in the general population (about 0.5%).

Two different mutant alleles at the same locus, as in HbS/HbC, form a genetic compound (compound heterozygote). The phenotype usually lies between those produced by either allele present in the homozygous state. Because of the large number of mutations possible in a given gene, many autosomal recessive phenotypes are probably due to genetic compounds. Sickle cell disease is an exception. Consanguinity is strong presumptive evidence for true homozygosity of mutant alleles and against a genetic compound.

X-Linked Inheritance

The general characteristics of X-linked inheritance in humans can be summarized as follows: (1) There is no male-to-male transmission of the phenotype (Figure e2–5). (2) Unaffected males do not transmit the phenotype. (3) All of the daughters of an affected male are heterozygous carriers. (4) Males are usually more severely affected than females. (5) Whether a heterozygous female is counted as affected—and whether the phenotype is called "recessive" or "dominant"—depends often on the sensitivity of the assay or examination. (6) Some mothers of affected males will not themselves be heterozygotes (ie, they will be homozygous normal) but will have a germinal mutation. The proportion of heterozygous (carrier) mothers is negatively associated with the severity of the condition. (7) Heterozygous women transmit the mutant gene to 50% of their sons, who are affected, and to 50% of their daughters, who are heterozygotes. (8) If an affected male mates with a heterozygous female, 50% of the male offspring will be affected, giving the false impression of male-to-male transmission. Of the female offspring of such matings, 50% will be affected as severely as the average hemizygous male; in small pedigrees, this pattern may simulate autosomal dominant inheritance.

View Large Figure e2–5 Add to 'My Saved Images'  A pedigree illustrating X-linked inheritance. (Symbols as in Figure e2–3.)

The characteristics of X-linked inheritance depend on phenotypic severity. For some disorders, affected males do not survive to reproduce. In such cases, about two-thirds of affected males have a carrier mother; in the remaining third, the disorder arises by new germinal mutation in an X chromosome of the mother. When the disorder is nearly always manifest in heterozygous females (X-linked dominant inheritance), females tend to be affected about twice as often as males; and on average an affected female transmits the phenotype to 50% of her sons and 50% of her daughters.

X-linked phenotypes are often clinically variable—particularly in heterozygous females—and suspected of being autosomal dominant with nonpenetrance. For example, Fabry disease (-galactosidase A deficiency) may be clinically silent in carrier women or may cause stroke, renal failure, or myocardial infarction by middle age.

Germinal mosaicism occurs in mothers of boys with X-linked conditions. The chance of such a mother having a second affected son or a heterozygous daughter depends on the fraction of her oocytes that carries the mutation. Currently, this fraction is impossible to determine. However, the presence of germinal mosaicism can be detected in some conditions (eg, Duchenne muscular dystrophy) in a family by analysis of DNA, and this knowledge becomes crucial for genetic counseling.

Mitochondrial Inheritance

Mutations in the genes encoded by the mitochondrial chromosome cause a variety of diseases that affect (in particular) organs highly dependent on oxidative metabolism, such as the retina, brain, kidneys, and heart. Because a person's mitochondria derive almost entirely from the ovum, the inheritance pattern is distinct from that of mendelian disorders and is termed "maternal" or, more appropriately, "mitochondrial." An affected woman can pass the defective mitochondrial chromosome to all of her offspring, whereas an affected man has little risk of passing his mutation to a child (Figure e2–6). Because each cell and the ovum contain many mitochondria and because each mitochondrion contains many chromosomes, two situations are possible: If every chromosome in every mitochondrion carries the same mutation, the person is said to be homoplasmic for the mutation. On the other hand, if only some of the mitochondrial chromosomes carry the mutation, the person is heteroplasmic. In the latter case, an offspring may inherit relatively few mitochondria bearing the mutation and have mild disease or no disease.

View Large Figure e2–6 Add to 'My Saved Images'  Mitochondrial ("maternal") inheritance. A mitochondrial genetic mutation, indicated by darkened symbols, is passed by the female (circle) to all of her offspring, including males (squares). Of subsequent offspring, males do not transmit the mutation, but females continue to transmit the mutation to all of their offspring because mitochondria are passed through ova, not sperm. For simplicity, although both parents are shown for the first generation, subsequent generations do not show the genetic partners, who are assumed to lack the mutation. Note: All or only some of the mitochondria may carry the mutation, a variable that affects the clinical expression of the mutation. (See text regarding homoplasmic and heteroplasmic individuals.)

Over 16,000 human genes have been identified or implied through their phenotypes and inheritance patterns in families. This total represents 60–70% of all genes thought to be encoded by the 22 autosomes, two sex chromosomes, and the mitochondrial chromosome. Victor McKusick and colleagues began an international effort to catalogue human mendelian variation. This persists as Online Mendelian Inheritance in Man (OMIM).

Disorders of Multifactorial Causation

Many disorders cluster in families but are not associated with evident chromosomal aberrations or mendelian inheritance patterns. Examples include congenital malformations such as cleft lip, pyloric stenosis, and spina bifida; coronary artery disease; type 2 diabetes mellitus; and various forms of neoplasia. They are often characterized by varying frequencies in different racial or ethnic groups, disparity in sexual predilection, and greater frequency (but less than full concordance) in monozygotic than in dizygotic twins. This inheritance pattern is called "multifactorial" to signify that multiple genes interact with various environmental agents to produce the phenotype. The familial clustering is assumed to be due to sharing of both alleles and environment.

For most multifactorial conditions, there is little understanding of which particular genes are involved, how they and their products interact, and in what way different nongenetic factors contribute to the phenotype. For some disorders, biochemical and genetic studies have identified mendelian conditions within the coarse phenotype: Defects of the low-density lipoprotein receptor account for a small fraction of cases of ischemic heart disease (a larger fraction if only patients under age 50 years are considered); familial polyposis of the colon predisposes to adenocarcinoma; and some patients with emphysema have inherited deficiency of ₁-proteinase (₁-antitrypsin) inhibitor. Despite these notable examples, this reductionistic preoccupation with mendelian phenotypes is unlikely to explain the great majority of human disease; but even so, in the last analysis, much of human pathology will prove to be associated with genetic factors in cause, pathogenesis, or both.

Our ignorance about fundamental genetic mechanisms in development and physiology has not completely restricted practical approaches to the genetics of multifactorial disorders. For example, recurrence risks are based on empiric data derived from observation of many families. The risk of recurrence of multifactorial disorders is increased in several instances: (1) in close relatives (siblings, offspring, and parents) of an affected individual; (2) when two or more members of a family have the same condition; (3) when the first case in a family is in the less commonly affected sex (eg, pyloric stenosis is five times more common in boys; an affected woman has a threefold to fourfold greater risk of having a child with pyloric stenosis); and (4) in ethnic groups in which there is a high incidence of a particular condition (eg, spina bifida is 40 times more common in whites—and even more frequent among the Irish—than in Asians).

For many apparently multifactorial disorders, too few families have been examined to have established empiric risk data. A useful approximation of recurrence risk in close relatives is the square root of the incidence. For example, many common congenital malformations have an incidence of 1:2500 to 1:400 live births; the calculated recurrence risks are thus in the 2–5% range—values that correspond closely to experience.

Much effort has been directed at identifying the genetic contributors to multifactorial disease. Genome-wide association studies (GWAS) have included large cohorts of patients with a given disease. These patients have hundreds of thousands or a million single nucleotide polymorphisms that are documented and compared with a similarly large and ethnically age-matched cohort of individuals with no evidence of the disease. In this way, hundreds of common disorders have been associated with even more specific markers in the human genome. Despite these impressive results, the reality is that a given nucleotide change only alters the relative risk for a given disease slightly (eg, relative risk 1.2–1.4) with very few exceptions. More importantly, very few of the nucleotide polymorphisms occur within genes, or if they do, they do not alter the function of the gene in any discernable way. Thus, much work remains to define the genetic contributors to common diseases, and the family history remains an important clinical tool in predicting risk.

Chromosomal Aberrations

Any deviation from the structure and number of chromosomes as displayed in Figure e2–1 is, technically, a chromosomal aberration. Not all aberrations cause problems in the affected individual, but some that do not may lead to problems in offspring. About 1:200 live-born infants has a chromosomal aberration that is detected because of some effect on phenotype. This frequency increases markedly the earlier in fetal life the chromosomes are examined. By the end of the first trimester of gestation, most fetuses with abnormal numbers of chromosomes have been lost through spontaneous abortion. For example, Turner syndrome—due to the absence of one sex chromosome and the presence of a single X chromosome—is a relatively common condition, but it is estimated that only 2% of fetuses with this form of aneuploidy survive to term. Even more striking in live-born children is the complete absence of most autosomal trisomies and monosomies despite their frequent occurrence in young fetuses.

Types of Chromosomal Abnormalities

Major structural changes occur in either balanced or unbalanced form. In the latter, there is a gain or loss of genetic material; in the former, there is no change in the amount of genetic material but only a rearrangement of it. At the sites of breaks and new attachments of chromosome fragments, there may be permanent structural or functional damage to one gene or to only a few genes. Despite no visible loss of material, the aberration may nonetheless be recognized as unbalanced through an abnormal phenotype and the chromosomal defect confirmed by molecular analysis of the DNA.

Aneuploidy results from nondisjunction—the failure of a chromatid pair to separate in a dividing cell. Nondisjunction in either the first or second division of meiosis results in gametes with abnormal chromosomal constitutions. In aneuploidy, more or fewer than 46 chromosomes are present (Table e2–1). The following are all forms of aneuploidy: (1) monosomy, in which only one member of a pair of chromosomes is present; (2) trisomy, in which three chromosomes are present instead of two; and (3) polysomy, in which one chromosome is represented four or more times.

Table e2–1. Clinical phenotypes resulting from aneuploidy.

If nondisjunction occurs in mitosis, mosaic patterns occur in somatic tissue, with some cells having one karyotype and other cells of the same organism another karyotype. Patients with a mosaic genetic constitution often have manifestations of each of the genetic syndromes associated with the various abnormal karyotypes.

Translocation results from an exchange of parts of two chromosomes.

Deletion is loss of chromosomal material.

Duplication is the presence of two or more copies of the same region of a given chromosome. The redundancy may occur in the same chromosome or in a nonhomologous chromosome. In the latter case, a translocation will also have occurred.

An isochromosome is a chromosome in which the arms on either side of the centromere have the same genetic material in the same order—ie, the chromosome has at some time divided in such a way that it has a double dose of one arm and absence of the other.

In an inversion, a chromosomal region becomes reoriented 180 degrees out of the ordinary phase. The same genetic material is present, but in a different order.