Thompson & Thompson Genetics in Medicine, 8th Edition

CHAPTER 6. The Chromosomal and Genomic Basis of Disease

Disorders of the Autosomes and Sex Chromosomes

In this chapter, we present several of the most common and best understood chromosomal and genomic disorders encountered in clinical practice, building on the general principles of clinical cytogenetics and genome analysis introduced in the previous chapter. Each of the disorders presented here illustrates the principles of dosage balance and imbalance at the level of chromosomes and subchromosomal regions of the genome. Because a wide range of phenotypes seen in clinical medicine involve chromosome and subchromosomal mutations, we include in this chapter the spectrum of disorders that are characterized by intellectual disability or by abnormal or ambiguous sexual development. Although many such disorders can be determined by single genes, the clinical approach to evaluation of such phenotypes frequently includes detailed chromosome and genome analysis.

Mechanisms of Abnormalities

In this section, we consider abnormalities that illustrate the major chromosomal and genomic mechanisms that underlie genetic imbalance of entire chromosomes or chromosomal regions. Overall, we distinguish five different categories of such abnormalities, each of which can lead to disorders of clinical significance:

• Disorders due to abnormal chromosome segregation (nondisjunction)

• Disorders due to recurrent chromosomal syndromes, involving deletions or duplications at genomic hot spots

• Disorders due to idiopathic chromosomal abnormalities, typically de novo

• Disorders due to unbalanced familial chromosomal abnormalities

• Disorders due to chromosomal and genomic events that reveal regions of genomic imprinting

The distinguishing features of the underlying mechanisms are summarized in Table 6-1. Although the categories of defects that result from these mechanisms can involve any chromosomes, we introduce them here in the context of autosomal abnormalities.


Mechanisms of Chromosome Abnormalities and Genomic Imbalance


Underlying Mechanism


Abnormal chromosome segregation


Aneuploidy (Down syndrome, Klinefelter syndrome)

Uniparental disomy

Recurrent chromosomal syndromes

Recombination at segmental duplications

Duplication/deletion syndromes

Copy number variation

Idiopathic chromosome abnormalities

Sporadic, variable breakpoints

Deletion syndromes (cri du chat syndrome, 1p36 deletion syndrome)


De novo balanced translocations

Gene disruption

Unbalanced familial abnormalities

Unbalanced segregation

Offspring of balanced translocations

Offspring of pericentric inversions

Syndromes involving genomic imprinting

Any event that reveals imprinted gene(s)

Prader-Willi/Angelman syndromes


The most common mutation in our species involves errors in chromosome segregation, typically leading to production of an abnormal gamete that has two copies or no copies of the chromosome involved in the nondisjunction event. Notwithstanding the high frequency of such errors in meiosis and, to a lesser extent, in mitosis, there are only three well-defined nonmosaic chromosome disorders compatible with postnatal survival in which there is an abnormal dose of an entire autosome: trisomy 21 (Down syndrome), trisomy 18, and trisomy 13. It is surely no coincidence that these chromosomes are the ones with the smallest number of genes among all autosomes (see Fig. 2-7). Imbalance for more gene-rich chromosomes is presumably incompatible with long-term survival, and aneuploidy for some of these is frequently associated with pregnancy loss (see Table 5-2).

Each of these autosomal trisomies is associated with growth retardation, intellectual disability, and multiple congenital anomalies (Table 6-2). Nevertheless, each has a fairly distinctive phenotype that is immediately recognizable to an astute clinician in the newborn nursery. Trisomy 18 and trisomy 13 are both less common than trisomy 21; survival beyond the first year is rare, in contrast to Down syndrome, in which average life expectancy is over 50 years of age.


Features of Autosomal Trisomies Compatible with Postnatal Survival


CNS, Central nervous system.

The developmental abnormalities characteristic of any one trisomic state must be determined by the extra dosage of the particular genes on the additional chromosome. Knowledge of the specific relationship between the extra chromosome and the consequent developmental abnormality has been limited to date. Current research, however, is beginning to localize specific genes on the extra chromosome that are responsible for specific aspects of the abnormal phenotype, through direct or indirect modulation of patterning events during early development (see Chapter 14). The principles of gene dosage and the likely role of imbalance for individual genes that underlie specific developmental aspects of the phenotype apply to all aneuploid conditions; these are illustrated here in the context of Down syndrome, whereas the other conditions are summarized in Table 6-2.

Down Syndrome

Down syndrome is by far the most common and best known of the chromosome disorders and is the single most common genetic cause of moderate intellectual disability. Approximately 1 child in 850 is born with Down syndrome (see Table 5-2), and among liveborn children or fetuses of mothers 35 years of age or older, the incidence of trisomy 21 is far higher (Fig. 6-1).


FIGURE 6-1 Maternal age dependence on incidence of trisomy 21 at birth and at time of amniocentesis. SeeSources & Acknowledgments.

Down syndrome can usually be diagnosed at birth or shortly thereafter by its dysmorphic features, which vary among patients but nevertheless produce a distinctive phenotype (Fig. 6-2). Hypotonia may be the first abnormality noticed in the newborn. In addition to characteristic dysmorphic facial features (see Fig. 6-2), the patients are short in stature and have brachycephaly with a flat occiput. The neck is short, with loose skin on the nape. The hands are short and broad, often with a single transverse palmar crease (“simian crease”) and incurved fifth digits (termed clinodactyly).


FIGURE 6-2 Phenotype of Down syndrome. A, Young infant. The nasal bridge is flat; ears are low-set and have a characteristic folded appearance; the eyes show characteristic epicanthal folds and upslanting palpebral fissures; the mouth is open, showing a protruding tongue. B, Brushfield spots around the margin of the iris (arrow). SeeSources & Acknowledgments.

A major cause for concern in Down syndrome is intellectual disability. Even though in early infancy the child may not seem delayed in development, the delay is usually obvious by the end of the first year. Although the extent of intellectual disability varies among patients from moderate to mild, many children with Down syndrome develop into interactive and even self-reliant persons, and many attend local schools.

There is a high degree of variability in the phenotype of Down syndrome individuals; specific abnormalities are detected in almost all patients, but others are seen only in a subset of cases. Congenital heart disease is present in at least one third of all liveborn Down syndrome infants. Certain malformations, such as duodenal atresia and tracheoesophageal fistula, are much more common in Down syndrome than in other disorders.

Only approximately 20% to 25% of trisomy 21 conceptuses survive to birth (see Table 5-2). Among Down syndrome conceptuses, those least likely to survive are those with congenital heart disease; approximately one fourth of the liveborn infants with heart defects die before their first birthday. There is a fifteen fold increase in the risk for leukemia among Down syndrome patients who survive the neonatal period. Premature dementia, associated with the neuropathological findings characteristic of Alzheimer disease (cortical atrophy, ventricular dilatation, and neurofibrillar tangles), affects nearly all Down syndrome patients, several decades earlier than the typical age at onset of Alzheimer disease in the general population.

As a general principle, it is important to think of this constellation of clinical findings, their variation, and likely outcomes in terms of gene imbalance—the relative overabundance of specific gene products; their impact on various critical pathways in particular tissues and cell types, both early in development and throughout life; and the particular alleles present in a particular patient's genome, both for genes on the trisomic chromosome and for the many other genes inherited from his or her parents.

The Chromosomes in Down Syndrome

The clinical diagnosis of Down syndrome usually presents no particular difficulty. Nevertheless, karyotyping is necessary for confirmation and to provide a basis for genetic counseling. Although the specific abnormal karyotype responsible for Down syndrome usually has little effect on the phenotype of the patient, it is essential for determining the recurrence risk.

Trisomy 21.

In at least 95% of all patients, the Down syndrome karyotype has 47 chromosomes, with an extra copy of chromosome 21 (see Fig. 5-9). This trisomy results from meiotic nondisjunction of the chromosome 21 pair. As noted earlier, the risk for having a child with trisomy 21 increases with maternal age, especially after the age of 30 years (see Fig. 6-1). The meiotic error responsible for the trisomy usually occurs during maternal meiosis (approximately 90% of cases), predominantly in meiosis I, but approximately 10% of cases occur in paternal meiosis, often in meiosis II. Typical trisomy 21 is a sporadic event, and thus recurrences are infrequent, as will be discussed further later.

Approximately 2% of Down syndrome patients are mosaic for two cell populations—one with a normal karyotype and one with a trisomy 21 karyotype. The phenotype may be milder than that of typical trisomy 21, but there is wide variability in phenotypes among mosaic patients, presumably reflecting the variable proportion of trisomy 21 cells in the embryo during early development.

Robertsonian Translocation.

Approximately 4% of Down syndrome patients have 46 chromosomes, one of which is a Robertsonian translocation between chromosome 21q and the long arm of one of the other acrocentric chromosomes (usually chromosome 14 or 22) (see Fig. 5-11). The translocation chromosome replaces one of the normal acrocentric chromosomes, and the karyotype of a Down syndrome patient with a Robertsonian translocation between chromosomes 14 and 21 is therefore 46,XX or XY,rob(14;21)(q10;q10),+21 (see Table 5-1 for nomenclature). Despite having 46 chromosomes, patients with a Robertsonian translocation involving chromosome 21 are trisomic for genes on the entirety of 21q.

A carrier of a Robertsonian translocation, involving, for example, chromosomes 14 and 21, has only 45 chromosomes; one chromosome 14 and one chromosome 21 are missing and are replaced by the translocation chromosome. The gametes that can be formed by such a carrier are shown in Figure 6-3, and such carriers are at risk for having a child with translocation Down syndrome.


FIGURE 6-3 Chromosomes of gametes that theoretically can be produced by a carrier of a Robertsonian translocation, rob(14;21). A, Normal and balanced complements. B, Unbalanced, one product with both the translocation chromosome and the normal chromosome 21, and the reciprocal product with chromosome 14 only. C, Unbalanced, one product with both the translocation chromosome and chromosome 14, and the reciprocal product with chromosome 21 only. Theoretically, there are six possible types of gamete, but three of these appear unable to lead to viable offspring. Only the three shaded gametes at the left can lead to viable offspring. Theoretically, the three types of gametes will be produced in equal numbers, and thus the theoretical risk for a Down syndrome child should be 1 in 3. However, extensive population studies have shown that unbalanced chromosome complements appear in only approximately 10% to 15% of the progeny of carrier mothers and in only a few percent of the progeny of carrier fathers who have translocations involving chromosome 21.

Unlike standard trisomy 21, translocation Down syndrome shows no relation to maternal age but has a relatively high recurrence risk in families when a parent, especially the mother, is a carrier of the translocation. For this reason, karyotyping of the parents and possibly other relatives is essential before accurate genetic counseling can be provided.

21q21q Translocation.

A 21q21q translocation chromosome is seen in a few percent of Down syndrome patients and is thought to originate as an isochromosome. It is particularly important to evaluate if a parent is a carrier because all gametes of a carrier of such a chromosome must either contain the 21q21q chromosome, with its double dose of chromosome 21 genetic material, or lack it and have no chromosome 21 representative at all. The potential progeny, therefore, inevitably have either Down syndrome or monosomy 21, which is rarely viable. Mosaic carriers are at an increased risk for recurrence, and thus prenatal diagnosis should be considered in any subsequent pregnancy.

Partial Trisomy 21.

Very rarely, Down syndrome is diagnosed in a patient in whom only a part of the long arm of chromosome 21 is present in triplicate. These patients are of particular significance because they can show what region of chromosome 21 is likely to be responsible for specific components of the Down syndrome phenotype and what regions can be triplicated without causing that aspect of the phenotype. The most notable success has been identification of a less than 2-Mb region that is critical for the heart defects seen in approximately 40% of Down syndrome patients. Sorting out the specific genes crucial to the expression of the Down syndrome phenotype from those that merely happen to be syntenic with them on chromosome 21 is critical for determining the pathogenesis of the various clinical findings.

Risk for Down Syndrome

A frequent problem in genetic counseling is to assess the risk for the birth of a Down syndrome child. The risk depends chiefly on the mother's age but also on both parents' karyotypes, as discussed previously. Down syndrome can be detected prenatally by karyotyping, by chromosomal microarray analysis, or by genome-wide sequencing of chorionic villus or amniotic fluid cells (see Fig. 5-9). Screening for Down syndrome is also possible now by noninvasive prenatal screening (NIPS) of cell-free fetal DNA in maternal plasma. As will be discussed in more detail in Chapter 17, although all pregnancies should be offered prenatal diagnosis, a decision to undergo invasive methods of prenatal testing balances the risk that a fetus has Down syndrome and the risk that the procedure of amniocentesis or chorionic villus sampling used to obtain fetal tissue for chromosome analysis will lead to fetal loss. However, with NIPS emerging as a screening test for Down syndrome and other relatively common aneuploid conditions, this paradigm and counseling considerations are likely to change in the years ahead (see Chapter 17).

The population incidence of Down syndrome in live births is currently estimated to be approximately 1 in 850, reflecting the maternal age distribution for all births and the proportion of older mothers who make use of prenatal diagnosis and selective termination. At approximately the age of 30 years, the risk begins to rise sharply, approaching 1 in 10 births in the oldest maternal age-group (see Fig. 6-1). Even though younger mothers have a much lower risk, their birth rate is much higher, and therefore more than half of the mothers of all Down syndrome babies are younger than 35 years. The risk for Down syndrome due to translocation or partial trisomy is unrelated to maternal age. The paternal age appears to have no influence on the risk.

Recurrence Risk

The recurrence risk for trisomy 21 or any other autosomal trisomy, after one such child has been born in a family, is approximately 1% overall. The risk is approximately 1.4% for mothers younger than 30 years, and it is the same as the age-related risk for older mothers; that is, there is a slight but significant increase in risk for the younger mothers but not for the older mothers, whose risk is already elevated. The reason for the increased risk for the younger mothers is not known. A history of trisomy 21 elsewhere in the family, although often a cause of maternal anxiety, does not appear to significantly increase the risk for having a Down syndrome child.

The recurrence risk for Down syndrome due to a translocation is much higher, as described previously.

Uniparental Disomy

Chromosome nondisjunction most commonly results in trisomy or monosomy for the particular chromosome involved in the segregation error. However, less commonly, it can also lead to a disomic state in which both copies of a chromosome derive from the same parent, rather than one copy being inherited from the mother and the other from the father. This situation, called uniparental disomy, is defined as the presence of a disomic cell line containing two chromosomes, or portions thereof, that are inherited from only one parent (see Table 6-1). If the two chromosomes are derived from identical sister chromatids, the situation is described as isodisomy; if both homologues from one parent are present, the situation is heterodisomy.

The most common explanation for uniparental disomy is trisomy “rescue” due to chromosome nondisjunction in cells of a trisomic conceptus to restore a disomic state. The cause of the originating trisomy is typical meiotic nondisjunction in one of the parental germlines; the rescue results from a second nondisjunction event, this one occurring mitotically at an early postzygotic stage, thus “rescuing” a fetus that otherwise would most likely be aborted spontaneously (the most common fate for any trisomic fetus; see Table 5-2). Depending on the stage and parent of the original nondisjunction event (i.e., maternal or paternal meiosis I or II), the location of meiotic recombination events, and which chromosome is subsequently lost in the postzygotic mitotic nondisjunction event, the resulting fetus or liveborn can have complete or partial isodisomy or heterodisomy for the relevant chromosome.

Although it is not known how common uniparental disomy is overall, it has been documented for most chromosomes in the karyotype by demonstrating uniparental inheritance of polymorphisms in a family. Clinical abnormalities, however, have been demonstrated for only some of these, typically in cases when an imprinted region is present in two copies from one parent (see the section on genomic imprinting later in this chapter) or when a typically recessive condition (which would ordinarily imply that both parents are obligate carriers; see Chapter 7) is observed in a patient who has only one documented carrier parent. It is important to stress that, although such conditions frequently come to clinical attention because of mutations in individual genes or in imprinted regions, the underlying pathogenomic mechanism in cases of uniparental disomy is abnormal chromosome segregation.

Genomic Disorders: Microdeletion and Duplication Syndromes

Dozens of syndromes characterized by developmental delay, intellectual disability, and a specific constellation of dysmorphic features and birth defects are known to be associated with recurrent subchromosomal or regional abnormalities (see Table 6-1). These small but sometimes cytogenetically visible deletions and/or duplications lead to a form of genetic imbalance referred to as segmental aneusomy. These deletions (and, in some cases, their reciprocal duplications) are typically detected by chromosomal microarrays. The term contiguous gene syndrome has been applied to many of these conditions, because the phenotype is often attributable to extra or deficient copies of multiple, contiguous genes within the deleted or duplicated region. For other such disorders, however, the phenotype is apparently due to deletion or duplication of only a single gene within the region, despite being associated typically with a chromosomal abnormality that encompasses several genes.

For many of these syndromes, although the clinical phenotype in different patients can be quite variable, the nature of the underlying genomic abnormality is highly similar. Indeed, for the syndromes listed in Table 6-3, high-resolution genomic studies have demonstrated that the centromeric and telomeric breakpoints cluster among different patients, suggesting the existence of genomic sequences that predispose to the rearrangements. Fine mapping in a number of these disorders has shown that the breakpoints localize to low-copy repeated sequences in the genome termed segmental duplications (see Chapter 4). Aberrant recombination between nearby copies of the repeats causes the deletions and/or duplications, which typically span several hundred to several thousand kilobase pairs. Extensive analysis of over 30,000 patients worldwide has now implicated this general sequence-dependent mechanism in 50 to 100 syndromes involving contiguous gene rearrangements, which collectively are sometimes referred to as genomic disorders.


Examples of Genomic Disorders Involving Recombination between Segmental Duplications


Based on Lupski JR, Stankiewicz P: Genomic disorders: the genomic basis of disease, Totowa, NJ, 2006, Humana Press; Cooper GM, Coe BP, Girirajan S, et al: A copy number variation morbidity map of developmental delay. Nat Genet 43:838-846, 2011; and Weischenfeldt J, Symmns O, Spitz F, et al: Phenotypic impact of genomic structural variation: insights from and for human disease. Nat Rev Genet 14:125-138, 2013.

It is this mechanistic association with segmental duplications that distinguishes this subgroup of deletion and duplication syndromes from others whose breakpoints are highly variable and are not associated with any identifiable genomic feature(s), and whose mechanistic basis appears idiopathic (see Table 6-1). Here we focus on syndromes involving chromosome 22 to illustrate underlying genomic features of this class of disorders.

Deletions and Duplications Involving Chromosome 22q11.2

Several deletions and duplications mediated by unequal recombination between segmental duplications have been documented within the proximal long arm of chromosome 22 and illustrate the general concept of genomic disorders (Fig. 6-4). A particularly common microdeletion involves chromosome 22q11.2 and is associated with diagnoses of DiGeorge syndrome, velocardiofacial syndrome, and conotruncal anomaly face syndrome. All three clinical syndromes are autosomal conditions with variable clinical expression, caused by a deletion of approximately 3 Mb within 22q11.2 on one copy of chromosome 22. The microdeletion and other rearrangements of this region shown in Figure 6-5 are each mediated by homologous recombination between segmental duplications in the region. The deletions are detected in approximately 1 in 4000 live births, making this one of the most common genomic rearrangements associated with important clinical phenotypes.


FIGURE 6-4 Model of rearrangements underlying genomic disorders. Unequal crossing over between misaligned sister chromatids or homologous chromosomes containing highly homologous copies of segmentally duplicated sequences can lead to deletion or duplication products, which differ in the number of copies of genes normally located between the repeats. The copy number of any gene or genes (e.g., A, B, and C) that lie between the copies of the repeat will change as a result of these genome rearrangements. For examples of genomic disorders, segmental duplications, and the size of the deleted or duplicated region, see Table 6-3.


FIGURE 6-5 Chromosomal deletions, duplications, and rearrangements in 22q11.2 mediated by homologous recombination between segmental duplications. A, Normal karyotypes show two copies of 22q11.2, each containing multiple copies of a family of related segmental duplications within the region (dark blue). In DiGeorge syndrome (DGS) or velocardiofacial syndrome (VCFS), a 3-Mb region is deleted from one homologue, removing approximately 30 genes; in approximately 10% of patients, a smaller 1.5-Mb deletion (nested within the larger segment) is deleted. The reciprocal duplication is seen in patients with dup(22)(q11.2q11.2). Tetrasomy for 22q11.2 is seen in patients with cat eye syndrome. Note that the duplicated region in the cat eye syndrome chromosome is in an inverted orientation relative to the duplication seen in dup(22) patients, indicating a more complex genomic rearrangement involving these segmental duplications. B, Expanded view of the 22q11.2 genomic region, indicating the common DGS/VCFS deletions (red) and more distal deletions (also mediated by recombination involving segmental duplications) that are seen in patients with other phenotypes (orange). Genes in the region (from browser) are indicated above the region. C, Two-color fluorescence in situ hybridization analysis of proband with DGS, demonstrating deletion of 22q11.2 on one homologue. Green signal is hybridization to a control region in distal 22q. Red signal shows hybridization to a region in proximal 22q that is present on one copy of the chromosome but deleted from the other (arrow). SeeSources & Acknowledgments.

Patients show characteristic craniofacial anomalies, intellectual disability, immunodeficiency, and heart defects, likely reflecting haploinsufficiency for one or more of the several dozen genes that are normally found in this region. Deletion of the TBX1 gene in 22q11.2 deletion syndrome is thought to play a role in as many as 5% of all congenital heart defects and is a particularly frequent cause of left-sided outflow tract abnormalities.

Compared to the relatively common deletion of 22q11.2, the reciprocal duplication of 22q11.2 is much rarer and leads to a series of distinct dysmorphic malformations and birth defects called the 22q11.2 duplication syndrome (see Fig. 6-5). Diagnosis of this duplication generally requires analysis by fluorescence in situ hybridization (FISH) on interphase cells or by chromosomal microarray.

The general concepts illustrated for disorders associated with 22q11.2 also apply to many other chromosomal and genomic disorders, some of the most common or more significant of which are summarized in Table 6-3. Together, these recurrent syndromes emphasize several important principles in human and medical genetics (see Box).

Lessons From Genomic Disorders

Genomic disorders collectively illustrate a number of concepts of general importance for considering the causes and consequences of chromosomal or genomic imbalance.

• First, with few exceptions, altered gene dosage for any extensive chromosomal or genomic region is likely to result in a clinical abnormality, the phenotype of which will, in principle, reflect haploinsufficiencyfor or overexpression of one or more genes encoded within the region. In some cases, the clinical presentation appears to be accounted for by dosage imbalance for just a single gene; in other syndromes, however, the phenotype appears to reflect imbalance for multiple genes across the region.

• Second, the distribution of these duplication/deletion disorders around the genome appears not to be random, because the location of families of segmental duplications, especially in pericentromeric and subtelomeric regions, predisposes particular regions to the unequal recombination events that underlie these syndromes.

• And third, even patients carrying what appears to be the same chromosomal deletion or duplication can present with a range of variable phenotypes. Although the precise basis for this variability is unknown, it could be due to nongenetic causes, to underlying genetic variation in the region on the non-deleted chromosome, or to differences elsewhere in the genome among unrelated individuals.

Idiopathic Chromosome Abnormalities

Whereas the abnormalities just described are mediated by the landscape of specific genomic features in particular chromosomal regions, many other chromosome abnormalities are due to deletions or rearrangements that have no definitive mechanistic basis (see Table 6-1). There are many reports of cytogenetically detectable abnormalities in dysmorphic patients involving events such as deletions, duplications, or translocations of one or more chromosomes in the karyotype (see Fig. 5-11). Overall, cytogenetically visible autosomal deletions occur with an estimated incidence of 1 in 7000 live births. Most of these have been seen in only a few patients and are not associated with recognized clinical syndromes. Others, however, are sufficiently common to allow delineation of clearly recognizable syndromes in which a series of patients have similar abnormalities.

The defining mechanistic feature of this class of abnormalities is that the underlying chromosomal event is idiopathic (see Table 6-1); most of them occur de novo and have highly variable breakpoints in the particular chromosomal region, thus distinguishing them as a class from those discussed in the previous section.

Autosomal Deletion Syndromes

One long-recognized syndrome is the cri du chat syndrome, in which there is either a terminal or interstitial deletion of part of the short arm of chromosome 5. This deletion syndrome was given its common name because crying infants with this disorder sound like a mewing cat. The facial appearance, shown in Figure 6-6, is distinctive and includes microcephaly, hypertelorism, epicanthal folds, low-set ears, sometimes with preauricular tags, and micrognathia. The overall incidence of the deletion is estimated to be as high as 1 in 15,000 live births.


FIGURE 6-6 Idiopathic deletion syndromes. A-C, Three different children with cri du chat syndrome, which results from deletion of part of chromosome 5p. Note, even among unrelated individuals, the characteristic facies with hypertelorism, epicanthus, and retrognathia. D, Phenotype-karyotype map of chromosome 5p, based on chromosomal microarray analysis of a series of del(5p) patients. E, Chromosomal microarray analysis of approximately 5-Mb deletion in band 1p36.3 (red), which is undetectable by conventional karyotyping. SeeSources & Acknowledgments.

Most cases of cri du chat syndrome are sporadic; only 10% to 15% of the patients are the offspring of translocation carriers. The breakpoints and extent of the deleted segment of chromosome 5p is highly variable among different patients, but the critical region missing in all patients with the phenotype has been identified as band 5p15. Many of the clinical findings have been attributed to haploinsufficiency for a gene or genes within specific regions; the degree of intellectual impairment usually correlates with the size of the deletion, although genomic studies suggest that haploinsufficiency for particular regions within 5p14-p15 may contribute disproportionately to severe intellectual disability (see Fig. 6-6).

Although many large deletions can be appreciated by routine karyotyping, detection of other idiopathic deletions requires more detailed analysis by microarrays; this is particularly true for abnormalities involving subtelomeric bands of many chromosomes, which can be difficult to visualize well by routine chromosome banding. For example, one of the most common idiopathic abnormalities, the chromosome 1p36 deletion syndrome, has a population incidence of 1 in 5000 and involves a wide range of different breakpoints, all within the terminal 10 Mb of chromosome 1p. Approximately 95% of cases are de novo, and many (e.g., the case illustrated in Fig. 6-6) are not detectable by routine chromosome analysis.

Detailed genomic analysis of various autosomal deletion syndromes underscores the idiopathic nature of these abnormalities. Typically, and in contrast to the genomic disorders presented in Table 6-3, the breakpoints are highly variable and reflect a range of different mechanisms, including terminal deletion of the chromosome arm with telomere healing, interstitial deletion of a subtelomeric segment, or recombination between copies of repetitive elements, such as Alu or L1 elements (see Chapter 2).

Balanced Translocations with Developmental Phenotypes

Reciprocal translocations are relatively common (see Chapter 5). Most are balanced and involve the precise exchange of chromosomal material between nonhomologous chromosomes; as such, they usually do not have an obvious phenotypic effect. However, among the approximately 1 in 2000 newborns who has a de novo balanced translocation, the risk for a congenital abnormality is empirically elevated several-fold, leading to the suggestion that some balanced translocations involve direct disruption of a gene or genes by one or both of the translocation breakpoints.

Detailed analysis of a number of such cases by FISH, microarrays, and targeted or whole-genome sequencing has identified defects in protein-coding or noncoding RNA genes in patients with various phenotypes, ranging from developmental delay to congenital heart defects to autism spectrum disorders. Although the clinical abnormalities in these cases can be ascribed to mutations in individual genes located at the site of the translocations, the underlying mechanism in each case is the chromosomal rearrangement itself (see Table 6-1).

Segregation of Familial Abnormalities

Although most of the idiopathic abnormalities just described are sporadic, other clinical presentations can occur because of unbalanced segregation of familial chromosome abnormalities. In these cases, the underlying mechanism for the clinical phenotype is not the chromosomal abnormality itself, but rather its transmission in an unbalanced state from a parent who is a balanced carrier to the subsequent generation (see Table 6-1).

The mechanism of pathogenesis here is distinguished from the mechanism of nondisjunction described earlier in this chapter. In contrast to aneuploidy or uniparental disomy, it is not the process of segregation that is abnormal in these cases; rather, it is the random nature of events during segregation that leads to unbalanced karyotypes and thus to offspring with abnormal phenotypes.

In the case of balanced translocations, for example, because the chromosomes involved form a quadrivalent in meiosis, the particular combination of chromosomes transmitted to a given gamete can lead to genomic imbalance (see Fig. 5-12), even though the segregation is itself normal.

Another type of familial structural abnormality that illustrates this mechanism involves inversion chromosomes. In this case, segregation of the inverted chromosome and its normal homologue during meiosis is typically uneventful; however, unbalanced gametes can be produced as a result of the process of recombination occurring within the inverted segment, in particular for pericentric inversions (see Fig. 5-13). Different inversion chromosomes carry different risks for abnormal offspring, presumably reflecting both the likelihood that a recombination event will occur within the inverted segment and the likelihood that an unbalanced gamete can lead to viable offspring. This overall risk must be determined empirically for use in genetic counseling. Several well-described inversions illustrate this point.

A pericentric inversion of chromosome 3 is one of the few for which sufficient data have been obtained to allow an estimate of the transmission of the inversion chromosome to the offspring of carriers. The inv(3)(p25q21) originated in a couple from Newfoundland in the early 1800s and has since been reported in a number of families whose ancestors can be traced to the Atlantic provinces of Canada. Carriers of the inv(3) chromosome are normal, but some of their offspring have a characteristic abnormal phenotype associated with the presence of a recombinant chromosome 3, in which there is duplication of the segment distal to 3q21 and deficiency of the segment distal to 3p25. The other predicted unbalanced gamete, with a duplication of distal 3p and deficiency of distal 3q, does not lead to viable offspring. The empirical risk for an abnormal pregnancy outcome in inv(3) carriers is greater than 40% and indicates the importance of family chromosome studies to identify carriers and to offer genetic counseling and prenatal diagnosis.

Not all pericentric inversions have a risk for abnormal offspring, however. One of the most common inversions seen in human chromosomes is a small pericentric inversion of chromosome 9, which is present in up to 1% of all individuals. The inv(9)(p11q12) has no known deleterious effect on carriers and does not appear to be associated with a significant risk for miscarriage or unbalanced offspring; the empirical risk is not different from that of the population at large, and it is therefore generally considered a normal variant.

Disorders Associated with Genomic Imprinting

For some disorders, the expression of the disease phenotype depends on whether the mutant allele or abnormal chromosome has been inherited from the father or from the mother. As we introduced in Chapter 3, such parent-of-origin effects are the result of genomic imprinting.

The effect of genomic imprinting on inheritance patterns in pedigrees will be discussed in Chapter 7. Here, we focus on the relevance of imprinting to clinical cytogenetics, as many imprinting effects come to light because of chromosome abnormalities. Evidence of genomic imprinting has been obtained for a number of chromosomes or chromosomal regions throughout the genome, as revealed by comparing phenotypes of individuals carrying the same cytogenetic abnormality affecting either the maternal or paternal homologue. Although estimates vary, it is likely that as many as several hundred genes in the human genome show imprinting effects. Some regions contain a single imprinted gene; others contain clusters of multiple imprinted genes, spanning in some cases well over 1 Mb along a chromosome.

The hallmark of imprinted genes that distinguishes them from other autosomal loci is that only one allele, either maternal or paternal, is expressed in the relevant tissue. The effect of such mechanisms on the clinical phenotype will necessarily depend on whether a mutational event occurred on the maternal or paternal homologue. Among the best-studied examples of the role of genomic imprinting in human disease are Prader-Willi syndrome (Case 38) and Angelman syndrome, and we discuss these next to illustrate the genetic and genomic features of imprinting conditions. An additional example, Beckwith-Wiedemann syndrome, is presented in Case 6.

Prader-Willi and Angelman Syndromes

Prader-Willi syndrome is a relatively common dysmorphic syndrome characterized by neonatal hypotonia followed by obesity, excessive and indiscriminate eating habits, small hands and feet, short stature, hypogonadism, and intellectual disability (Fig. 6-7). Prader-Willi syndrome results from the absence of a paternally expressed imprinted gene or genes. In approximately 70% of cases of the syndrome, there is a cytogenetic deletion of the proximal long arm of chromosome 15 (15q11.2-q13); the deletion is mediated by recombination involving segmental duplications that flank a region of approximately 5 to 6 Mb and in that sense is mechanistically similar to other genomic disorders described earlier (see Table 6-3). However, within this region lies a smaller interval that contains a number of monoallelically expressed genes, some of which are normally expressed only from the paternal copy and others of which are expressed only from the maternal copy (see Fig. 6-7). In Prader-Willi syndrome, the deletion is found only on the chromosome 15 inherited from the patient's father (Table 6-4). Thus the genomes of these patients have genomic information in 15q11.2-q13 that derives only from their mothers, and the syndrome results from the loss of expression of one or more of the normally paternally expressed genes in the region.


FIGURE 6-7 Prader-Willi syndrome (PWS) and Angelman syndrome (AS). A, PWS in a image-year-old boy with obesity, hypogonadism, and small hands and feet who also has short stature and developmental delay. B, Angelman syndrome in a 4-year-old girl. Note wide stance and position of arms. C,Chromosomal microarray detection of approximately 5-Mb deletion in 15q11.2-q13.1 (red). D, Schematic of the 15q11.2-q13 region. The PWS region (shaded in blue) contains a series of imprinted genes (blue) that are expressed only from the paternal copy. The AS region (shaded in pink) contains two imprinted genes that are expressed only from the maternal copy, including the UBE3Agene, which is imprinted in the central nervous system and mutations in which can cause AS. The region is flanked by nonimprinted genes (purple) that are expressed from both maternal and paternal copies. Common deletions of the PWS/AS region, caused by recombination between pairs of segmental duplications, are shown in green at the bottom. Smaller deletions of the imprinting center (IC; orange) and of a subset of genes in the small nucleolar RNA (snoRNA) gene cluster can also lead to PWS. cen, Centromere; tel, telomere. SeeSources & Acknowledgments.


Genomic Mechanisms Causing Prader-Willi and Angelman Syndromes


Prader-Willi Syndrome

Angelman Syndrome

15q11.2-q13 deletion

≈70% (paternal)

≈70% (maternal)

Uniparental disomy

≈20-30% (maternal)

≈7% (paternal)

Imprinting center mutation



Gene mutations

Rare (small deletions within snoRNA gene cluster)

≈10% (UBE3A mutations)




snoRNA, Small nucleolar RNA.

Data from Cassidy SB, Schwartz S, Miller JL, et al: Prader-Willi syndrome. Genet Med 14:10-26, 2012; Dagli AI, Williams CA: Angelman syndrome. In Pagon RA, Adam MP, Bird TD, et al, editors: GeneReviews [Internet], Seattle, 1993-2013, University of Washington, Seattle,

Notably, the low-copy repeats that flank the Prader-Willi and Angelman syndrome regions have also been implicated in other disorders, including duplication or triplication of the region or inverted duplication of chromosome 15. This underscores that although imprinting is responsible for the inheritance and specific clinical findings in Prader-Willi and Angelman syndromes, the underlying pathogenomic mechanism of all these disorders involves unequal recombination of the segmental duplications in the region.

In contrast, in most patients with the rare Angelman syndrome, which is characterized by unusual facial appearance, short stature, severe intellectual disability, spasticity, and seizures (see Fig. 6-7), there is a deletion of the same chromosomal region, but, now on the chromosome 15 inherited from the mother. Patients with Angelman syndrome therefore have genetic information in 15q11.2-q13 derived only from their fathers. This unusual circumstance demonstrates strikingly that the parental origin of genetic material (in this case, in a segment of chromosome 15) can have a profound effect on the clinical expression of a defect.

Some patients with Prader-Willi syndrome do not have cytogenetically detectable deletions; instead, they have two cytogenetically normal chromosome 15s, both of which were inherited from the mother (see Table 6-4). This situation illustrates uniparental disomy, introduced previously in this chapter in the section on abnormal chromosome segregation. A smaller percentage of patients with Angelman syndrome also have uniparental disomy, but in their case with two intact chromosome 15s of paternal origin (see Table 6-4). These patients add further emphasis that, although genomic imprinting is responsible for bringing such cases to clinical attention, the underlying defect in a proportion of cases is one of chromosome segregation, not one of imprinting per se, which is completely normal in these cases.

Primary defects in the imprinting process are seen, however, in a few patients with Prader-Willi syndrome and Angelman syndrome, who have abnormalities in the imprinting center itself (see Fig. 6-7). As a result, the switch from female to male imprinting during spermatogenesis or from male to female imprinting during oogenesis (see Fig. 3-12) fails to occur. Fertilization by a sperm carrying an abnormally persistent female imprint would produce a child with Prader-Willi syndrome; fertilization of an egg that bears an inappropriately persistent male imprint would result in Angelman syndrome (see Table 6-4).

Finally, there is evidence that the major features of the Prader-Willi and Angelman syndrome phenotypes can be accounted for by defects at particular genes within the imprinted region. Mutations in the maternal copy of a single gene, the ubiquitin-protein ligase E3A gene (UBE3A), have been found to cause Angelman syndrome (see Table 6-4). The UBE3A gene is located within the 15q11.2-q13 imprinted region and is normally expressed only from the maternal allele in the central nervous system. Maternally inherited single-gene mutations in UBE3A account for approximately 10% of Angelman syndrome cases. In Prader-Willi syndrome, several patients have been described with deletions of a much smaller region on the paternally inherited chromosome 15, specifically implicating the noncoding small nucleolar RNA (snoRNA)116 gene cluster in the etiology of the syndrome (see Fig. 6-7).

Other Disorders due to Uniparental Disomy of Imprinted Regions

Although it is unclear how common uniparental disomy is, it may provide an explanation for a disease when an imprinted region is present in two copies from one parent. Thus physicians and genetic counselors must keep imprinting in mind as a possible cause of genetic disorders.

For example, a few patients with cystic fibrosis and short stature have been described with two identical copies of most or the entirety of their maternal chromosome 7. In these cases, the mother happened to be a carrier for cystic fibrosis (Case 12), and because the child received two maternal copies of the mutant cystic fibrosis gene and no paternal copy of the normal allele at this locus, the child developed the disease. The growth failure was unexplained but might be related to loss of unidentified paternally imprinted genes on chromosome 7.

The Sex Chromosomes and Their Abnormalities

The X and Y chromosomes have long attracted interest because they differ between the sexes, because they have their own specific patterns of inheritance, and because they are involved in primary sex determination. They are structurally distinct and subject to different forms of genetic regulation, yet they pair in male meiosis. For all these reasons, they require special attention. In this section, we review the common sex chromosome abnormalities and their clinical consequences, the current state of knowledge concerning the control of sex determination, and abnormalities of sex development.

The Chromosomal Basis of Sex Determination

The different sex chromosome constitution of normal human male and female cells has been appreciated for more than 50 years. Soon after cytogenetic analysis became feasible, the fundamental basis of the XX/XY system of sex determination became apparent. Males with Klinefelter syndrome have 47 chromosomes with two X chromosomes as well as a Y chromosome (karyotype 47,XXY), whereas most Turner syndrome females have only 45 chromosomes with a single X chromosome (karyotype 45,X). These findings unambiguously establish the crucial role of the Y chromosome in normal male development. Furthermore, compared with the dramatic consequences of autosomal aneuploidy, these karyotypes underscore the relatively modest effect of varying the number of X chromosomes in either males or females. The basis for both observations can be explained in terms of the unique biology of the Y and X chromosomes.

The process of sex determination can be thought of as occurring in distinct but interrelated steps (Fig. 6-8):

• Establishment of chromosomal sex (i.e., XY or XX) at the time of fertilization

• Initiation of alternate pathways to differentiation of one or the other gonadal sex, as determined normally by the presence or absence of the testis-determining gene on the Y chromosome

• Continuation of sex-specific differentiation of internal and external sexual organs

• Especially after puberty, development of distinctive secondary sexual characteristics to create the corresponding phenotypic sex, as a male or female


FIGURE 6-8 The process of sex determination and development: establishment of chromosomal sex at fertilization; commitment to the male or female pathway of gonadal differentiation; sex-specific differentiation of internal and external genitalia and development of secondary sexual characteristics (phenotypic sex). Whereas the sex chromosomes play a determining role in specifying chromosomal sex, many genes located on both the sex chromosomes and the autosomes are involved in sex determination and subsequent sexual differentiation (see Table 6-8).

Whereas the sex chromosomes play a determining role in specifying chromosomal and gonadal sex, a number of genes located on both the sex chromosomes and the autosomes are involved in sex determination and subsequent sexual differentiation. In most instances, the role of these genes has come to light as a result of patients with various conditions known as disorders of sex development, and many of these are discussed later in this chapter.

The Y Chromosome

The structure of the Y chromosome and its role in sex development has been determined at both the molecular and genomic levels (Fig. 6-9). In male meiosis, the X and Y chromosomes normally pair by segments at the ends of their short arms (see Chapter 2) and undergo recombination in that region. The pairing segment includes the pseudoautosomal region of the X and Y chromosomes, so called because the X- and Y-linked copies of this region are essentially identical to one another and undergo homologous recombination in meiosis I, like pairs of autosomes. (A second, smaller pseudoautosomal segment is located at the distal ends of Xq and Yq.) By comparison with autosomes and the X chromosome, the Y chromosome is relatively gene poor (see Fig. 2-7) and contains fewer than 100 genes (some of which belong to multigene families), specifying only approximately two dozen distinct proteins. Notably, the functions of a high proportion of these genes are restricted to gonadal and genital development.


FIGURE 6-9 The Y chromosome in sex determination and in disorders of sex development (DSDs). Individual genes and regions implicated in sex determination, DSDs, and defects of spermatogenesis are indicated, as discussed in the text.

Embryology of the Reproductive System

The effect of the Y chromosome on the embryological development of the male and female reproductive systems is summarized in Figure 6-10. By the sixth week of development in both sexes, the primordial germ cells have migrated from their earlier extraembryonic location to the paired genital ridges, where they are surrounded by the sex cords to form a pair of primitive gonads. Up to this time, the developing gonad is ambipotent, regardless of whether it is chromosomally XX or XY.


FIGURE 6-10 Scheme of developmental events in sex determination and differentiation of the male and female gonads from the ambipotent gonad. See text for discussion.

Development into an ovary or a testis is determined by the coordinated action of a sequence of genes in finely balanced pathways that lead to ovarian development when no Y chromosome is present but tip to the side of testicular development when a Y is present. Under normal circumstances, the ovarian pathway is followed unless a particular Y-linked gene, originally designated testis-determining factor (TDF),diverts development into the male pathway.

If no Y chromosome is present, the gonad begins to differentiate to form an ovary, beginning as early as the eighth week of gestation and continuing for several weeks; the cortex develops, the medulla regresses, and oogonia begin to develop within follicles (see Fig. 6-10). Beginning at approximately the third month, the oogonia enter meiosis I, but (as described in Chapter 2) this process is arrested at dictyotene until ovulation occurs many years later.

In the presence of a normal Y chromosome (with the TDF gene), however, the medullary tissue forms typical testes with seminiferous tubules and Leydig cells that, under the stimulation of chorionic gonadotropin from the placenta, become capable of androgen secretion (see Fig. 6-10). Spermatogonia, derived from the primordial germ cells by successive mitoses, line the walls of the seminiferous tubules, where they reside together with supporting Sertoli cells, awaiting the onset of puberty to begin spermatogenesis.

While the primordial germ cells are migrating to the genital ridges, thickenings in the ridges indicate the developing genital ducts, the mesonephric (also called wolffian) and paramesonephric (also called müllerian) ducts, under the influence of hormones produced by specific cell types in the developing gonad. Duct formation is usually completed by the third month of gestation.

In the early embryo, the external genitalia consist of a genital tubercle, paired labioscrotal swellings, and paired urethral folds. From this undifferentiated state, male external genitalia develop under the influence of androgens, beginning at around 12 weeks of gestation. In the absence of a testis (or, more specifically, in the absence of androgens), female external genitalia are formed regardless of whether an ovary is present.

SRY is the Major Testis-Determining Gene

The earliest cytogenetic studies established the male-determining function of the Y chromosome. In the ensuing three decades, chromosomal and genomic analysis of individuals with different submicroscopic abnormalities of the Y chromosome and well-studied disorders of sex development allowed identification of the primary testis-determining region on Yp.

Whereas the X and Y chromosomes normally exchange in meiosis I within the Xp/Yp pseudoautosomal region, in rare instances, genetic recombination occurs outside of the pseudoautosomal region (Fig. 6-11). This leads to two rare but highly informative abnormalities—males with a 46,XX karyotype and females with a 46,XY karyotype—that involve an inconsistency between chromosomal sex and gonadal sex, as we will explore in greater detail later in this chapter.


FIGURE 6-11 Etiological factors of phenotypic males with a 46,XX karyotype or phenotypic females with a 46,XY karyotype by aberrant exchange between X- and Y-linked sequences. X and Y chromosomes normally recombine within the Xp/Yp pseudoautosomal segment in male meiosis. If recombination occurs below the pseudoautosomal boundary, between the X-specific and Y-specific portions of the chromosomes, sequences responsible for male gonadal sex determination (including the SRY gene) may be translocated from the Y to the X. Fertilization by a sperm containing such an X chromosome leads to a phenotypic male with XX testicular DSD. In contrast, fertilization by a sperm containing a Y chromosome that has lost SRY will lead to a phenotypic female with XY complete gonadal dysgenesis.

The SRY gene (sex-determining region on the Y) lies near the pseudoautosomal boundary on the Y chromosome. It is present in many males with an otherwise normal 46,XX karyotype (Case 41) and is deleted or mutated in a proportion of females with an otherwise normal 46,XY karyotype, thus strongly implicating SRY in normal male sex determination (see Fig. 6-11). SRY is expressed only briefly early in development in cells of the germinal ridge just before differentiation of the testis. SRY encodes a DNA-binding protein that is likely to be a transcription factor, which up-regulates a key autosomal gene, SOX9,in the ambipotent gonad, leading ultimately to testes differentiation. Thus, by all available genetic and developmental criteria, SRY is equivalent to the TDF gene on the Y chromosome. If SRY is absent or not functioning properly, then the female sex differentiation pathway ensues (see Fig. 6-10).

Although there is clear evidence demonstrating the critical role of SRY in normal male sexual development, the presence or absence of SRY/TDF does not explain all cases of abnormal sex determination. Other genes are involved in the sex determination pathway and are discussed later in this chapter.

Y-Linked Genes in Spermatogenesis

The prevalence of Y chromosome deletions and microdeletions in the general male population is reported to be approximately 1 in 2000 to 3000 males. However, microdeletions in the male-specific portion of Yq are found in a significant proportion of men with low sperm count, ranging from cases of nonobstructive azoospermia (no sperm detectable in semen) to severe oligospermia (<5 million/mL; normal range, 20 to 40 million/mL). These findings suggest that one or more genes, termed azoospermia factors (AZF), are located on the Y chromosome, and three such regions on Yq (AZFa, AZFb, and AZFc) have been defined (see Fig. 6-9).

Genomic analysis of these microdeletions led to identification of a series of genes that appear to be important in spermatogenesis. For example, the 3.5-Mb-long AZFc deletion region contains seven different families of genes that are expressed only in the testis, including four copies of the DAZ genes (deleted in azoospermia) that encode nearly identical RNA-binding proteins expressed only in the premeiotic germ cells of the testis. De novo deletions of AZFc arise in approximately 1 in 4000 males and account for approximately 12% of azoospermic males and approximately 6% of males with severe oligospermia. Deletion of only two of the four DAZ genes has been associated with milder oligospermia. Similar to the other genomic disorders described earlier in this chapter, they are mediated by recombination between segmentally duplicated sequences (see Table 6-3). AZFa and AZFb deletions, although less common, also involve recombination. The Yq microdeletions are not syndromic, however; they are responsible only for a defect in spermatogenesis in otherwise normal males. The explanation is that all of the genes involved in the AZF deletions are expressed only in the testis and have no functions in other tissues or cell types.

Overall, approximately 2% of otherwise healthy males are infertile because of severe defects in sperm production, and it appears likely that de novo deletions or mutations of genes on Yq account for a significant proportion of these. Thus men with idiopathic infertility should be karyotyped, and Y chromosome molecular testing and genetic counseling may be appropriate before the initiation of assisted reproduction by intracytoplasmic sperm injection for such couples, mostly because of the risk for passing a Yq microdeletion responsible for infertility to the infertile couple's sons.

The X Chromosome

Aneuploidy for the X chromosome is among the most common of cytogenetic abnormalities. The relative tolerance of human development for X chromosome abnormalities can be explained in terms of X chromosome inactivation, the process by which most genes on one of the two X chromosomes in females are silenced epigenetically, introduced in Chapter 3. X inactivation and its consequences in relation to the inheritance of X-linked disorders are discussed in Chapter 7. Here we discuss the chromosomal and genomic mechanisms of X inactivation and their implications for human and medical genetics (see Box at the end of this section).

X Chromosome Inactivation

The principle of X inactivation is that in somatic cells in normal females (but not in normal males), one X chromosome is inactivated early in development, thus equalizing the expression of X-linked genes in the two sexes. In normal female development, because the choice of which X chromosome is to be inactivated is a random one that is then maintained clonally, females are mosaic with respect to X-linked gene expression (see Fig. 3-13).

There are many epigenetic features that distinguish the active and inactive X chromosomes in somatic cells (Table 6-5). These features can be useful diagnostically for identifying the inactive X chromosome(s) in clinical material. In patients with extra X chromosomes (whether male or female), any X chromosome in excess of one is inactivated (Fig. 6-12). Thus all diploid somatic cells in both males and females have a single active X chromosome, regardless of the total number of X or Y chromosomes present.


Epigenetic and Chromosomal Features of X Chromosome Inactivation in Somatic Cells


Active X

Inactive X

Gene expression

Yes; similar to male X

Most genes silenced; ≈15% expressed to some degree

Chromatin state


Facultative heterochromatin; Barr body

Noncoding RNA

XIST gene silenced

XIST RNA expressed from Xi only; associates with Barr body

DNA replication timing

Synchronous with autosomes

Late-replicating in S phase

Histone variants

Similar to autosomes and male X

Enriched for macroH2A

Histone modifications

Similar to autosomes and male X

Enriched for heterochromatin marks; deficient in euchromatin marks

Xi, Inactive X.


FIGURE 6-12 Sex chromosome constitution and X chromosome inactivation. Top, In individuals with extra X chromosomes, any X in excess of one is inactivated, regardless of sex and regardless of the number of Y chromosomes present. Thus the number of inactive X chromosomes in diploid cells is always one less than the total number of X chromosomes. Bottom, Detection of inactive X chromosomes (Xi) in interphase nuclei from females with 46,XX, 47,XXX, 48,XXXX, and 49,XXXXX karyotypes. Regions of bright fluorescence indicate the presence of the histone variant macroH2A associated with inactive X chromosomes (see Table 6-5).

The X chromosome contains approximately 1000 genes, but not all of these are subject to inactivation. Notably, the genes that continue to be expressed, at least to some degree, from the inactive X are not distributed randomly along the X chromosome; many more genes “escape” inactivation on distal Xp (as many as 50%) than on Xq (just a few percent). This finding has important implications for genetic counseling in cases of partial X chromosome aneuploidy, because imbalance for genes on Xp may have greater clinical significance than imbalance for genes on Xq, where the effect is largely mitigated by X inactivation.

Patterns of X Inactivation.

X inactivation is normally random in female somatic cells and leads to mosaicism for two cell populations expressing alleles from one or the other X (Fig. 6-13). Where examined, most females have approximately equal proportions of cells expressing alleles from the maternal or paternal X (i.e., approximately 50:50), and approximately 90% of phenotypically normal females fall within a distribution that extends from approximately 25:25 to approximately 75:25 (see Fig. 6-13). Such a distribution presumably reflects the expected range of outcomes for a random event (i.e., the choice of which X will be the inactive X) involving a relatively small number of cells during early embryogenesis. For individuals who are carriers for X-linked single-gene disorders (see Chapter 7), this X inactivation ratio can influence the clinical phenotype, depending on what proportion of cells in relevant tissues or cell types express the deleterious allele on the active X.


FIGURE 6-13 X chromosome inactivation in karyotypes with normal or abnormal X chromosomes or X;autosome translocations. A, Normal female cells (46,XX) undergo random X inactivation, resulting in a mosaic of two cell populations (left) in which either the paternal or maternal X is the inactive X (Xi, indicated by shaded box). In phenotypically normal females, the ratio of the two cell populations has a mode at 50:50, but with variation observed in the population (right), some with an excess of cells expressing alleles from the paternal X and others with an excess of cells expressing alleles from the maternal X. B, Individuals carrying a structurally abnormal X (abn X) or X;autosome translocation in a balanced or unbalanced state show nonrandom X inactivation in which virtually all cells have the same X inactive. The other cell population is inviable or at a growth disadvantage because of genetic imbalance and is thus underrepresented or absent. der(X) and der(A) represent the two derivatives of the X;autosome translocation. SeeSources & Acknowledgments.

However, there are exceptions to the distribution expected for random X inactivation when the karyotype involves a structurally abnormal X chromosome. For example, in nearly all patients with unbalanced structural abnormalities of an X chromosome (including deletions, duplications, and isochromosomes), the structurally abnormal chromosome is always the inactive X. Because the initial inactivation event early in embryonic development is likely random, the patterns observed after birth probably reflect secondary selection against genetically unbalanced cells that are inviable (see Fig. 6-13). Because of this preferential inactivation of the abnormal X, such X chromosome anomalies have less of an impact on phenotype than unbalanced abnormalities of similar size or gene content involving autosomes.

Nonrandom inactivation is also observed in most cases of X;autosome translocations (see Fig. 6-13). If such a translocation is balanced, the normal X chromosome is preferentially inactivated, and the two parts of the translocated chromosome remain active, again likely reflecting selection against cells in which critical autosomal genes have been inactivated. In the unbalanced offspring of a balanced carrier, however, only the translocation product carrying the X inactivation center is present, and this chromosome is invariably inactivated; the normal X is always active. These nonrandom patterns of inactivation have the general effect of minimizing, but not always eliminating, the clinical consequences of the particular chromosomal defect. Because patterns of X inactivation are strongly correlated with clinical outcome, determination of an individual's X inactivation pattern by cytological or molecular analysis (see Table 6-5) is indicated in all cases involving X;autosome translocations.

The X Inactivation Center.

Inactivation of an X chromosome depends on the presence of the X inactivation center region (XIC) on that chromosome, whether it is a normal X chromosome or a structurally abnormal X. Detailed analysis of structurally abnormal, inactivated X chromosomes led to the identification of the XIC within an approximately 800-kb candidate region in proximal Xq, in band Xq13.2 (Fig. 6-14), which coordinates many, if not all, of the critical steps necessary to initiate and promulgate the silenced chromatin state along the near-entirety of the X chosen to become the inactive X. As introduced in Chapter 3, this complex series of events requires a noncoding RNA gene, XIST, that appears to be a key master regulatory locus for the onset of X inactivation. It is one of a suite of noncoding RNA genes in the interval, others of which may operate in the regulation of XIST expression and in other early events in the X inactivation process.

Significance of X Inactivation in Medical Genetics

Many of the underlying details of X inactivation are mechanistically similar to other, more localized epigenetic silencing systems (see Table 3-2). Nonetheless, there are a number of features of X inactivation that are of central importance to human and medical genetics:

• Its chromosomal nature reduces the impact of segmental- or whole-chromosome genetic imbalance, such that many numerical and structural abnormalities of the X chromosome are relatively less deleterious than comparable abnormalities of the autosomes.

• Its random nature and the resulting clonal mosaicism greatly influence the clinical phenotype of females who carry X-linked single-gene mutations on one of their X chromosomes (see Chapter 7).

• Its dependence on the XIC is required for normal XX female development, because even very small fragments of the X chromosome separated from the XIC can lead to severe phenotypic anomalies as a result of their expression from both copies of genes contained on the X fragment (see Fig. 6-14).


FIGURE 6-14 X chromosome inactivation and dependence on X inactivation center (XIC). A, On normal X chromosomes, XIC lies within an approximately 800-kb candidate region in Xq13.2 that contains a number of noncoding RNA (ncRNA) genes, including XIST, the master X inactivation control gene. In early development in XX embryos, the XIST RNA spreads along the length of one X, which will become the inactive X (Xi), with epigenetic silencing of most genes on that X chromosome, resulting in monoallelic expression of most, but not all X-linked genes. B, On structurally abnormal X chromosomes that lack the XIC, X inactivation cannot occur and genes present on the abnormal X are expressed biallelically. Although a fairly large abnormal X is shown here for illustrative purposes, in fact only very small such fragments are observed in female patients, who invariably display significant congenital anomalies, suggesting that biallelic expression of larger numbers of X-linked genes is inconsistent with normal development and is likely inviable.

Cytogenetic Abnormalities of the Sex Chromosomes

Sex chromosome abnormalities are among the most common of all human genetic disorders, with an overall incidence of approximately 1 in 400 live births. Like abnormalities of the autosomes, they can be either numerical or structural and can be present in all cells or in mosaic form. As a group, disorders of the sex chromosomes tend to occur as isolated events without apparent predisposing factors, except for an effect of late maternal age in the cases that originate from errors of maternal meiosis I. There are a number of clinical indications that raise the possibility of a sex chromosome abnormality and thus the need for chromosomal or genomic studies. These indications include delay in onset of puberty, primary or secondary amenorrhea, infertility, and ambiguous genitalia.

The most common sex chromosome abnormalities involve aneuploidy for the X and/or Y chromosomes. The phenotypes associated with these chromosomal defects are, in general, less severe than those associated with comparable autosomal disorders because, as discussed earlier, X chromosome inactivation, as well as the low gene content of the Y, minimize the clinical consequences of sex chromosome imbalance. By far the most common sex chromosome defects in liveborn infants and in fetuses are the trisomic types (XXY, XXX, and XYY), but all three are rare in spontaneous abortions. In contrast, monosomy for the X (Turner syndrome) is less frequent in liveborn infants but is the most common chromosome anomaly reported in spontaneous abortions (see Table 5-2).

Sex Chromosome Aneuploidy

The incidence and major features of the four conditions associated with sex chromosome aneuploidy are compared in Tables 6-6 and 6-7. These well-defined syndromes are important causes of infertility, abnormal development, or both, and thus warrant a more detailed description. The effects of these chromosome abnormalities on development have been studied in long-term multicenter studies of hundreds of affected individuals, some of whom have been monitored for more than 40 years. As a group, those with sex chromosome aneuploidy show reduced levels of psychosocial adaptation, educational achievement, occupational performance, and economic independence, and on average they score slightly lower on intelligence (IQ) tests than their peers. However, each group shows high variability, making it impossible to generalize to specific cases. In fact, the overall impression is a high degree of normalcy, particularly in adulthood, which is remarkable among those with major chromosomal anomalies. Because almost all patients with sex chromosome abnormalities have only mild developmental abnormalities, a parental decision regarding potential termination of a pregnancy in which the fetus is found to have this type of defect can be a very difficult and even controversial one.


Incidence of Sex Chromosome Abnormalities


DSD, Disorder of sex development.

Data updated from Robinson A, Linden MG, Bender BG: Prenatal diagnosis of sex chromosome abnormalities. In Milunsky A, editor: Genetic disorders of the fetus, ed 4, Baltimore, 1998, Johns Hopkins University Press, pp 249-285.


Features of Sex Chromosome Aneuploidy Conditions


Summarized from Ross JL, Roeltgen DP, Kushner H, et al: Behavioral and social phenotypes in boys with 47,XYY syndrome or 47,XXY Klinefelter syndrome. Pediatrics 129:769-778, 2012; Pinsker JE: Turner syndrome: updating the paradigm of clinical care. J Clin Endocrinol Metab97:E994-E1003, 2012; and AXYS, http:

Here, we use Klinefelter syndrome to illustrate the major principles of sex chromosome aneuploidy. A more detailed presentation of Turner syndrome (45,X and its variants) can be found in the Cases (Case 47).

Klinefelter Syndrome (47,XXY).

The phenotype of typical patients with Klinefelter syndrome is shown in Figure 6-15. Klinefelter patients are almost always infertile because of the failure of germ cell development, and patients are often identified clinically for the first time because of infertility; as such, Klinefelter syndrome is classified among disorders of sex development, as we shall see in the next section. Klinefelter syndrome is relatively common among infertile males (approximately 4%) or males with oligospermia or azoospermia (approximately 10%). In adulthood, persistent androgen deficiency may result in decreased muscle tone, a loss of libido, and decreased bone mineral density.


FIGURE 6-15 Phenotype of males with 47,XXY Klinefelter syndrome. The patients are tall and thin and have relatively long legs. They appear physically normal until puberty, when signs of hypogonadism become obvious. Puberty occurs at a normal age, but the testes remain small, and secondary sexual characteristics remain underdeveloped. Note narrow shoulders and chest. Gynecomastia is a feature of some Klinefelter males and is visible in the 16-year-old patient in A. SeeSources & Acknowledgments.

The incidence of Klinefelter syndrome is estimated to be as high as 1 in 600 male births. Approximately half the cases result from nondisjunction in paternal meiosis I because of a failure of normal Xp/Yp recombination in the pseudoautosomal region. Among cases of maternal origin, most result from errors in maternal meiosis I; maternal age is increased in such cases. Approximately 15% of Klinefelter patients have mosaic karyotypes, most commonly 46,XY/47,XXY. As a group, such mosaic patients have variable phenotypes, and some may have normal testicular development.

Although there is wide phenotypic variation among patients with this and other sex chromosome aneuploidies, some consistent phenotypic differences have been identified between patients with Klinefelter syndrome and chromosomally normal males (see Table 6-7). Verbal comprehension and ability are below those of 46,XY males. Patients with Klinefelter syndrome have a several-fold increased risk for learning difficulties, especially in reading, that may require educational intervention. Language difficulties may lead to shyness, unassertiveness, apparent immaturity, and an increased risk for depression. Although most Klinefelter males form normal adult relationships, many of the affected boys have relatively poor psychosocial adjustment. Because of the relatively mild yet variable phenotype, many cases are presumed to go undetected.

Disorders of Sex Development

Earlier in this chapter, we discussed the primary sex-determining role of the Y chromosome and the SRY gene. In this section, we examine the role of various genes in ovarian and testicular development and in the development of male and female external genitalia. Disorders of gonadal and sexual development can arise from errors at any of the major steps of normal sex determination outlined earlier (see Fig. 6-8). These conditions, ranging from gonadal abnormalities to complete incompatibility between chromosomal and phenotypic sex, are now collectively termed disorders of sex development (DSD). They are among the most common birth defects; worldwide, 1 in 4500 babies are born with significant ambiguous genitalia, and DSDs are estimated to account for over 7% of all birth defects.

Although the chromosomal sex of an embryo is established at the time of fertilization, for some newborn infants, assignment of sex is difficult or impossible because the genitalia are ambiguous, with anomalies that tend to make them resemble in part those of the opposite chromosomal sex. Such anomalies may vary from mild hypospadias in males (a developmental anomaly in which the urethra opens on the underside of the penis or on the perineum) to an enlarged clitoris in females. In some patients, as we discuss later, both ovarian and testicular tissue is present. Abnormalities of either external or internal genitalia do not necessarily indicate a cytogenetic abnormality of the sex chromosomes but may reflect chromosomal changes elsewhere in the karyotype, single-gene defects, or nongenetic causes. Nonetheless, determination of the child's karyotype, frequently accompanied by chromosomal microarray, is an essential part of the investigation of such patients and can help guide both surgical and psychosocial management, as well as genetic counseling.

The detection of cytogenetic abnormalities, especially when seen in multiple patients, can also provide important clues about the location and nature of genes involved in sex determination and sex differentiation, some of which are listed in Table 6-8. DSDs can be classified into several major phenotypic and mechanistic groups, examples of which are discussed in the following sections. We focus on a few examples to illustrate the critical balance among various genes and their products that is necessary for normal gonadal and genital development in both males and females (see Box). These examples also reinforce the wide range of cytogenetic and genomic approaches—from standard karyotypes to FISH to microarrays to direct mutation analysis—needed for diagnosis, clinical and psychosocial management, and genetic counseling in these conditions.

Gene Balance and Disorders of Sex Development

The discovery of different Y-linked, X-linked, and autosomal, chromosomal, genomic, and single-gene abnormalities in different patients underscores the finely tuned nature of the network of dosage-sensitive genes that control gonadal development. The right genes and their products have to be expressed in the right amounts at precisely the right time and in the right place in the developing embryo.

Imbalance in the expression of major genes in the sex development pathways can override the signals typical of the chromosomal sex, leading to testis formation, even in the absence of a Y chromosome, or to ovarian development, even in the presence of the Y. Mutations and/or dosage imbalance (duplications or deletions) of critical genes in these pathways can overcome chromosomal sex and lead to a mismatch between chromosomal and gonadal sex or between gonadal and phenotypic sex (Fig. 6-16).


Examples of Genes Involved in Disorders of Sex Development


DSD, Disorder of sex development.

Updated from Achermann JC, Hughes IA: Disorders of sex development. In Melmed S, Polonsky KS, Larsen PR, Kronenberg HM, editors: Williams textbook of endocrinology, ed 12, Philadelphia, 2011, WB Saunders, pp 886-934.


FIGURE 6-16 Disorders of sex development (DSDs), across the spectrum of developmental events in sex determination and gonadal differentiation (see Fig. 6-10). Selected DSDs are shown, along with particular gene mutations and genomic alterations that interfere with the primary effect of chromosomal sex (XX or XY) in sex development and shift—all or in part—sex development toward the opposite sex. These mutations, duplications, and deletions illustrate the role of gene balance and imbalance on development of gonadal sex, sex-specific differentiation, and phenotypic sex. See text and Tables 6-8 and 6-9. CAH, Congenital adrenal hyperplasia; CAIS, complete androgen insensitivity syndrome; PAIS, partial androgen insensitivity syndrome.

Disorders of Gonadal Development

Gonadal dysgenesis refers to a progressive loss of germ cells, typically leading to underdeveloped and dysfunctional (“streak”) gonads, with consequent failure to develop mature secondary sex characteristics. Gonadal dysgenesis is typically categorized according to the karyotype of a patient. Complete gonadal dysgenesis (CGD)—as in the case of XX males (now formally designated 46,XX testicular DSD) or XY females (now formally designated 46,XY CGD)—is characterized by normal-appearing external genitalia of the opposite chromosomal sex. Cases with ambiguous external genitalia are said to have partial gonadal dysgenesis. Gonadal dysgenesis can also be associated with sex chromosome DSDs; it is a consistent feature of Turner syndrome (see Table 6-7), and patients with a 45,X/46,XY karyotype have mixed gonadal dysgenesis.

Various types of gonadal dysgenesis, their clinical phenotypes, and genetic causes are summarized in Table 6-9 and illustrated schematically in Figure 6-16.


Disorders of Sex Development and their Characteristics


DSD, Disorder of sex development.

Summarized from Achermann JC, Hughes IA: Disorders of sex development. In Melmed S, Polonsky KS, Larsen PR, Kronenberg HM, editors: Williams textbook of endocrinology, ed 12, Philadelphia, 2011, WB Saunders, pp 886-934; and Pagon RA, Adam MP, Bird TD, et al, editors: GeneReviews [Internet]. Seattle, 1993-2013, University of Washington, Seattle,

Disorders Associated with a 46,XY Karyotype

We begin with DSDs associated with a 46,XY karyotype. The overall incidence of these conditions is approximately 1 in 20,000 live births. Although a number of cytogenetic or single-gene defects have been demonstrated, many such cases remain unexplained. Approximately 15% of patients with 46,XY CGD have deletions or mutations in the SRY gene that interfere with the normal male pathway. However, most females with a 46,XY karyotype have an apparently normal SRY gene.

The DAX1 gene in Xp21.3 encodes a transcription factor that plays a dosage-sensitive role in determination of gonadal sex, implying a tightly regulated interaction between DAX1 and SRY. Although production of SRY at a critical point in early development normally leads to testis formation, an excess of DAX1 resulting from duplication of the gene can apparently suppress the normal male-determining function of SRY,leading to ovarian development (see Fig. 6-16).

A key master gene in gonadal development and the target of SRY signaling is the SOX9 gene on chromosome 17. SOX9 is normally expressed early in development in the genital ridge and is required for normal testis formation. Mutations in one copy of the SOX9 gene, typically associated with a skeletal malformation disorder called camptomelic dysplasia, lead to complete gonadal dysgenesis in approximately 75% of 46,XY cases (see Table 6-8). In the absence of one copy of the SOX9 gene, testes fail to form, and the ovarian pathway is followed instead. The phenotype of these patients suggests that the critical step for the male pathway is sufficient SOX9 expression to drive the formation of testes, normally after up-regulation by the SRY gene. In 46,XY CGD, with either a mutation in SRY or a mutation in SOX9, the levels of SOX9expression remain too low for testis differentiation, allowing ovarian differentiation to ensue.

As many as 10% of patients with a range of 46,XY DSD phenotypes carry mutations in the NR5A1 gene, which encodes a transcriptional regulator of a number of genes, including SOX9 and DAX1. These mutations are associated with inadequate androgenization of external genitalia, leading to ambiguous genitalia, partial gonadal dysgenesis, and absent or rudimentary müllerian structures.

Disorders Associated with a 46,XX Karyotype

A series of phenotypes known as the 46,XX testicular DSDs (previously termed XX sex reversal) are characterized by the presence of male external genitalia in individuals with an apparently normal 46,XX karyotype. The overall incidence is approximately 1 in 20,000.

Most patients have a normal male appearance at birth and are not diagnosed until puberty because of small testes, gynecomastia, and infertility, despite otherwise normal-appearing male genitalia and pubic hair (see Table 6-9). As described previously in the section on the Y chromosome, most of these patients are found to have a copy of a normal SRY gene translocated to an X chromosome as a result of aberrant recombination (see Fig. 6-11(Case 41).

Those 46,XX males who lack an SRY gene, however, are a clinically more heterogeneous group. Approximately 15% to 20% of such patients are identifiable at birth because of ambiguous genitalia, including penoscrotal hypospadias and cryptorchidism (undescended testes); there are no identifiable müllerian structures, and their gender identity is male. A somewhat smaller percentage of patients, however, have bothtesticular and ovarian tissue, either as an ovotestis or as a separate ovary and testis, a condition known as 46,XX ovotesticular DSD (formerly called true hermaphroditism).

Patients with either testicular DSD or ovotesticular DSD who lack a translocated SRY gene have been the subject of intense investigation to identify the responsible genetic defect(s). Duplications of at least two genes have been described, suggesting that increased levels of transcriptional regulators can overcome the absence of SRY and initiate the testis-specific pathway (see Table 6-8 and Fig. 6-16). Both gene duplications and regulatory mutations can increase the level of SOX9 expression to bypass the requirement for SRY. Similarly, duplications of the X-linked SOX3 gene, which is very closely related in sequence to the SRY gene, can stimulate increased SOX9 expression, replacing the usual need for SRY.

Ovarian Development and Maintenance

A number of genes have been implicated in normal ovarian development through the study of DSDs (see Table 6-8). Thus ovarian development may not be the “default” pathway, as it is frequently described, but rather the result of balanced interactions among various genes, some of which normally stimulate the ovarian pathway and others of which normally inhibit factors involved in the opposing male pathway.

Ovarian maintenance typically lasts for up to five decades in normal females. Loss of normal ovarian function before the age of 40, as seen in approximately 1% of women, is considered premature ovarian failure(or premature ovarian insufficiency). It has long been thought that two X chromosomes are necessary for ovarian maintenance, because 45,X females, despite normal initiation of ovarian development in utero, are characterized by germ cell loss, oocyte degeneration, and ovarian dysgenesis. Further, patients with 47,XXX or with cytogenetic abnormalities involving Xq, as well as carriers of fragile X syndrome (Case 17), frequently show premature ovarian failure. Because many nonoverlapping deletions on Xq show the same effect, this finding may reflect a need for two structurally normal X chromosomes in oogenesis or simply a requirement for multiple X-linked genes.

Nearly a dozen specific genes have been implicated in familial cases of premature ovarian failure and in various forms of 46,XX gonadal dysgenesis.

Disorders of Sex Development Involving Phenotypic Sex

Patients described earlier illustrate a mismatch between their chromosomal sex and their gonadal sex, frequently leading to gonadal dysgenesis (see Fig. 6-16). In contrast, individuals with 46,XX or 46,XY DSD have gonadal tissue that matches their chromosomal sex. However, their mismatch lies in the establishment of phenotypic sex: here, their internal and/or external genitalia show features that are contrary to those expected normally for those of the given chromosomal and gonadal sex (see Fig. 6-16). Thus patients with 46,XX DSD have a 46,XX karyotype with normal ovarian tissue but with ambiguous or male genitalia. And those with 46,XY DSD have a 46,XY karyotype and testicular tissue but with incompletely masculinized or female external genitalia. On this basis, patients of both types were thus previously described as having “pseudohermaphroditism,” a term no longer in use.

Virilization of 46,XX Infants: Congenital Adrenal Hyperplasia

These patients include those who have 46,XX karyotypes with a normal uterus and ovaries but with ambiguous or male external genitalia due to excessive virilization. The majority of such patients have congenital adrenal hyperplasia (CAH), an inherited disorder arising from specific defects in enzymes of the adrenal cortex required for cortisol biosynthesis and resulting in virilization of 46,XX infants. In addition to being a frequent cause of female virilization, CAH accounts for approximately half of all cases presenting with ambiguous external genitalia. Ovarian development is normal, but excessive production of androgens causes masculinization of the external genitalia, with clitoral enlargement and labial fusion to form a scrotum-like structure (Fig. 6-17).


FIGURE 6-17 Masculinized external genitalia of a 46,XX infant caused by congenital adrenal hyperplasia (virilizing form). See text for discussion. SeeSources & Acknowledgments.

Although any one of several enzymatic steps may be defective in CAH, by far the most common defect is deficiency of 21-hydroxylase, which has an incidence of approximately 1 in 12,500 births. Deficiency of 21-hydroxylase blocks the normal biosynthetic pathway of glucocorticoids and mineralocorticoids. This leads to overproduction of the precursors, which are then shunted into the pathway of androgen biosynthesis, causing abnormally high androgen levels in both XX and XY embryos. Whereas 46,XX infants with 21-hydroxylase deficiency are born with ambiguous genitalia, affected 46,XY infants have normal external genitalia and may go unrecognized in early infancy. Of patients with classic 21-hydroxylase deficiency, 25% have the simple virilizing type, and 75% have a salt-losing type due to a mineralocorticoid deficiency that is clinically more severe and may lead to neonatal death. A screening test developed to identify the condition in newborns is now in use in many countries (see Chapter 16). Prompt medical, surgical, and psychosocial management of 46,XX CAH patients is associated with improved fertility rates and normal female gender identity.

Incomplete Masculinization of 46,XY Infants: Androgen Insensitivity Syndrome

In addition to disorders of testis formation during embryological development, causes of DSD in 46,XY individuals include abnormalities of gonadotropins, inherited disorders of testosterone biosynthesis and metabolism, and abnormalities of androgen target cells. These disorders are heterogeneous both genetically and clinically, and in some cases they may correspond to milder manifestations of the same cause underlying ovotesticular DSD. Whereas the gonads are exclusively testes in 46,XY DSD, the genital ducts or external genitalia are incompletely masculinized (see Fig. 6-16).

There are several forms of androgen insensitivity that result in incomplete masculinization of 46,XY individuals. Here we illustrate the essential principles by considering the X-linked syndrome known as androgen insensitivity syndrome (once known as testicular feminization). As the original name indicates, testes are present either within the abdomen or in the inguinal canal, where they are sometimes mistaken for hernias in infants who otherwise appear to be normal females. Although the testes in these patients secrete androgen normally, end-organ unresponsiveness to androgens results from absence of androgen receptors in the appropriate target cells. The receptor protein, specified by the normal allele at the X-linked androgen receptor (AR) locus, has the role of forming a complex with testosterone and dihydrotestosterone. If the complex fails to form, the hormone fails to stimulate the transcription of target genes required for differentiation in the male direction. The molecular defect has been determined in many hundreds of cases and ranges from a complete deletion of the AR gene to point mutations in the androgen-binding or DNA-binding domains of the androgen receptor protein.

Affected individuals are chromosomal males (karyotype 46,XY) who have apparently normal female external genitalia but have a blind vagina and no uterus or uterine tubes. The incidence of androgen insensitivity is approximately 1 in 10,000 to 20,000 live births, and both complete and partial forms are known, depending on the severity of the genetic defect. In the complete form (Fig. 6-18), axillary and pubic hair are sparse or absent, and breast development occurs at the appropriate age, but without menses; primary amenorrhoea is frequently the presenting clinical finding that leads to a diagnosis. Gender assignment is typically not an issue, and psychosexual development and sexual function (except for fertility) are that of a typical 46,XX female.


FIGURE 6-18 Phenotype of a 46,XY individual with complete androgen insensitivity syndrome. Note female body contours, breast development, absence of axillary hair, and sparse pubic hair. SeeSources & Acknowledgments.

Neurodevelopmental Disorders and Intellectual Disability

Lastly, we consider another class of disorders that like the disorders of sex development just discussed, frequently require a wide range of chromosomal and genomic approaches for diagnosis, management, and genetic counseling. Neurodevelopmental disorders are highly heterogeneous, encompassing impairments in cognition, communication, behavior, and motor functioning. Broadly considered, the category of neurodevelopmental disorders includes overlapping diagnoses such as intellectual disability (defined as impairment of cognitive and adaptive functions in childhood), autism spectrum disorder (ASD) (Case 5), and attention deficit hyperactivity disorder (ADHD). This category can also include various neuropsychiatric conditions such as schizophrenia and bipolar disorder, complex traits of the type that are considered later in Chapter 8.

The overall incidence of intellectual disability and developmental delay is estimated to be at least 2% to 3%, whereas ASD affects as many as 1%. Determining the genetic cause of intellectual disability in most patients is a particular challenge, especially in the absence of other clinical clues or information about the specific gene or region of the genome responsible. Especially in sporadic cases without an obvious family history, a precise diagnosis can be helpful for clinical management and genetic counseling. Thus the full range of screening methods must be considered, including karyotyping and chromosomal microarrays, as well as whole-exome and whole-genome sequencing.

Genomic Imbalance in Neurodevelopmental Disorders

In large studies comparing diagnostic yield in this patient population, chromosomal microarray analysis detects pathogenic genomic imbalances in approximately 12% to 16% of cases, approximately fivefold more than G-banded karyotyping alone; on this basis, chromosomal microarrays are increasingly considered a first-tier clinical test to identify genomic imbalance in patients with unexplained intellectual disability or ASD. Although an increase in the presence of multiple rare copy number variants (CNVs) is true both for intellectual disability and for ASD, the CNVs in patients with intellectual disability tend to be larger and to encompass more genes and are more likely of de novo origin than those detected in ASD patients. Many hundreds of genes have been implicated to date, with estimates as high as a thousand or more genes in the genome that, when present in too few or too many copies, can lead to neurodevelopmental disorders.

Although screening for genomic imbalance due to CNVs is increasingly accepted as a diagnostic tool, identifying individual genes and their pathogenic mutations remains a significant challenge because of clinical and genetic heterogeneity. Some genes appear to be recurrent targets of mutation, accounting for up to several percent of cases; exome sequencing can identify de novo coding variants with likely or proven pathogenicity in approximately 15% of patients with severe, sporadic nonsyndromic intellectual disability and in cohorts of patients with the diagnosis of ASD. Whole-genome sequencing has also identified likely pathogenic mutations, either de novo or inherited, in ASD and in intellectual disability.

Although these approaches are valuable for gene discovery, large-scale exome or genome sequencing as a strategy for routine clinical testing will likely require substantial further reductions in cost, as well as improvements in the ability to distinguish a pathogenic mutation from the great excess of variants of unknown significance that are found in any single genome.

X-Linked Intellectual Disability

A long-appreciated aspect of intellectual disability is the excess of males in the affected population, and a large number of mutations, microdeletions, or duplications causing X-linked intellectual disability have been documented. The collective incidence of such X-linked defects has been estimated to be as high as 1 in 500 to 1000 live births.

The most common cause of X-linked intellectual disability is mutation in the FMR1 gene in males with the fragile X syndrome (Case 17). However, nearly 100 other X-linked genes have been implicated in X-linked intellectual disability, mostly on the basis of large family studies. Chromosomal microarray analysis has identified presumptive causal CNVs and insertion-deletions in a further 10% of such families. In addition, exome-sequencing efforts summarized in the preceding section to identify de novo changes in patients with intellectual disability have revealed an excess of such mutations on the X chromosome.

Clinical Heterogeneity and Diagnostic Overlap

A particular challenge for understanding neurodevelopmental disorders, their etiology, and their clinical course is the extraordinary degree of clinical heterogeneity, co-occurrence of symptoms, and diagnostic overlap among them. For cases due either to CNVs or to single-gene mutations, the same defect can lead to different clinical diagnoses in different cases and even in different family members—some with intellectual disability, some with ASD, and some with diagnosed psychiatric conditions. This heterogeneity and overlap, even when categorized by genetic/genomic diagnosis rather than clinical diagnosis, suggests the need for further study of genotype/phenotype correlations to meaningfully capture the broad range of phenotypes that might emerge among individuals with the same genetic disorder. As illustrated in Figure 6-19, one important factor is to analyze the effect of the CNV or mutation by comparing affected individuals to their unaffected family members (rather than to unrelated individuals in the general population), thus minimizing confounding effects of the wide range of cognitive and behavioral phenotypes observed even in the general population.


FIGURE 6-19 Model for impact of genetic or genomic changes on an individual's cognitive, neurobehavioral, and motor development. Here, the observed profile of abilities in probands (solid boxes) shows the deleterious effect of a copy number variant (CNV) on the predicted profile expected from familial background (gray boxes). The phenotypic effect of a particular CNV varies among the three elements of neurodevelopment. The purple dotted line represents the diagnostic threshold (2 SD below the mean). A, In this family, the deleterious effect of the CNV on quantitative cognitive traits (e.g., IQ) results in a diagnosis of intellectual disability, whereas neurobehavioral and motor features do not fall within the clinically impaired range. B, In contrast, in a different family, because of different familial norms, the deleterious effect of the same CNV leads to a diagnosis of a neurobehavioral disorder (e.g., schizophrenia), but without intellectual disability or motor impairment. SeeSources & Acknowledgments.

General References

Achermann JC, Hughes IA. Disorders of sex development. Melmed S, Polonsky KS, Larsen PR, Kronenberg HM. Williams textbook of endocrinology. ed 12. WB Saunders: Philadelphia; 2011:886–934.

Gardner RJM, Sutherland GR, Shaffer LG. Chromosome abnormalities and genetic counseling. ed 4. Oxford University Press: Oxford, England; 2012.

Moore KL, Persaud TVN, Torchia MG. The developing human: clinically oriented embryology. ed 9. W.B. Saunders: Philadelphia; 2013.

References for Specific Topics

Bartolomei MS, Ferguson-Smith AC. Mammalian genomic imprinting. Cold Spring Harb Perspect Biol. 2011;3:a002592.

Baxter R, Vilain R. Translational genetics for diagnosis of human disorders of sex development. Annu Rev Genomics Hum Genet. 2013;14:371–392.

Cassidy SB, Schwartz S, Miller JL, et al. Prader-Willi syndrome. Genet Med. 2012;14:10–26.

Cooper GM, Coe BP, Girirajan S, et al. A copy number variation morbidity map of developmental delay. Nat Genet. 2011;43:838–846.

de Ligt J, Willemsen H, van Bon BWM, et al. Diagnostic exome sequencing in persons with severe intellectual disability. N Engl J Med. 2012;367:1921–1929.

Ellison JW, Rosenfeld JA, Shaffer LG. Genetic basis of intellectual disability. Ann Rev Med. 2013;64:441–450.

Gajecka M, MacKay KL, Shaffer LG. Monosomy 1p36 deletion syndrome. Am J Med Genet Part C Semin Med Genet. 2007;145C:346–356.

Higgins AW, Alkuraya FS, Bosco AF, et al. Characterization of apparently balanced chromosomal rearrangements from the Developmental Genome Anatomy Project. Am J Hum Genet. 2008;82:712–722.

Hughes IA, Davies JD, Bunch TI, et al. Androgen insensitivity syndrome. Lancet. 2012;380:1419–1428.

Hughes IA, Houk C, Ahmed SF, et al. Consensus statement on management of intersex disorders. Arch Dis Child. 2006;91:554–563.

Huguet G, Ey E, Bourgeron T. The genetic landscapes of autism spectrum disorders. Ann Rev Genomics Hum Genet. 2013;14:191–213.

Jiang Y, Yuen RKC, Jin X, et al. Detection of clinically relevant genetic variants in autism spectrum disorder by whole-genome sequencing. Am J Hum Genet. 2013;93:1–15.

Kaminsky EB, Kaul V, Paschall J, et al. An evidence-based approach to establish the functional and clinical significance of copy number variants in intellectual and developmental disabilities. Genet Med. 2011;13:777–784.

Korbel JO, Tirosh-Wagner T, Urban AE, et al. The genetic architecture of Down syndrome phenotypes revealed by high-resolution analysis of human segmental trisomies. Proc Natl Acad Sci USA. 2009;106:12031–12036.

Leggett V, Jacobs P, Nation K, et al. Neurocognitive outcomes of individuals with a sex chromosome trisomy: XXX, XYY, or XXY: a systematic review. Devel Med Child Neurol. 2010;52:119–129.

Mabb AM, Judson MC, Zylka MJ, et al. Angelman syndrome: insights into genomic imprinting and neurodevelopmental phenotypes. Trends Neurosci. 2011;34:293–303.

Moreno-De-Luca A, Myers SM, Challman TD, et al. Developmental brain dysfunction: revival and expansion of old concepts based on new genetic evidence. Lancet Neurol. 2013;12:406–414.

Morris JK, Alberman E, Mutton D, et al. Cytogenetic and epidemiological findings in Down syndrome: England and Wales 1989-2009. Am J Med Genet A. 2012;158A:1151–1157.

Najmabadi H, Hu H, Garshasbi M, et al. Deep sequencing reveals 50 novel genes for recessive cognitive disorders. Nature. 2011;478:57–63.

Silber SJ. The Y chromosome in the era of intracytoplasmic sperm injection. Fertil Steril. 2011;95:2439–2448.

Talkowski ME, Maussion G, Crapper L, et al. Disruption of a large intergenic noncoding RNA in subjects with neurodevelopmental disabilities. Am J Hum Genet. 2012;91:1128–1134.

Talkowski ME, Rosenfeld JA, Blumenthal I, et al. Sequencing chromosomal abnormalities reveals neurodevelopmental loci that confer risk across diagnostic boundaries. Cell. 2012;149:525–537.

Weischenfeldt J, Symmns O, Spitz F, et al. Phenotypic impact of genomic structural variation: insights from and for human disease. Nat Rev Genet. 2013;14:125–138.

Zufferey F, Sherr EH, Beckmann ND, et al. A 600 kb deletion syndrome at 16p11.2 leads to energy imbalance and neuropsychiatric disorders. J Med Genet. 2013;49:660–668.


1. In a woman with a 47,XXX karyotype, what types of gametes would theoretically be formed and in what proportions? What are the theoretical karyotypes and phenotypes of her progeny? What are the actualkaryotypes and phenotypes of her progeny?

2. Individuals carrying a copy of the inv(9) described in the text are clinically normal. Provide two possible explanations.

3. The birth incidence rates of 47,XXY and 47,XYY males are approximately equal. Is this what you would expect on the basis of the possible origins of the two abnormal karyotypes? Explain.

4. How can a person with an XX karyotype differentiate as a phenotypically normal male?

5. A small centric ring X chromosome that lacks the X inactivation center is observed in a patient with short stature, gonadal dysgenesis, and intellectual disability. Because intellectual disability is not a typical feature of Turner syndrome, explain the presence of mental retardation with or without other associated physical anomalies in individuals with a 46,X,r(X) karyotype. In a prenatal diagnosis involving a different family, a somewhat larger ring that contains the X inactivation center is detected. What phenotype would you predict for the fetus in this pregnancy?

6. A baby girl with ambiguous genitalia is found to have 21-hydroxylase deficiency of the salt-wasting type. What karyotype would you expect to find? What is the disorder? What genetic counseling would you offer to the parents?

7. What are the expected clinical consequences of the following deletions? If the same amount of DNA is deleted in each case, why might the severity of each be different?

a. 46,XX,del(13)(pter→p11.1:)

b. 46,XY,del(Y)(pter→q12:)

c. 46,XX,del(5)(p15)

d. 46,XX,del(X)(q23q26)

8. Provide possible explanations for the fact that persons with X chromosome aneuploidy are clinically not completely normal.

9. In genetics clinic, you are counseling five pregnant women who inquire about their risk for having a Down syndrome fetus. What are their risks and why?

a. a 23-year-old mother of a previous trisomy 21 child

b. a 41-year-old mother of a previous trisomy 21 child

c. a 27-year-old woman whose niece has Down syndrome

d. a carrier of a 14;21 Robertsonian translocation

e. a woman whose husband is a carrier of a 14;21 Robertsonian translocation

10. A young girl with Down syndrome is karyotyped and found to carry a 21q21q translocation. With use of standard cytogenetic nomenclature, what is her karyotype?

11. Paracentric inversions generally do not raise the problem of imbalance in offspring. Why not?