Human cytogenetics and human chromosome abnormalities (Genetics)

1. Introduction

Modern human cytogenetics can be divided into three distinct eras, on the basis of the technologies that were available at the time. The first of these, the “golden era” extended from the late 1950s to about 1970. During this time, methodology was introduced, that made it possible to obtain karyotypes from a variety of tissues; most importantly, peripheral blood became amenable for study, making it possible to rapidly obtain cytogenetic information from virtually anyone. This quickly led to the identification of several common chromosomal syndromes, including Down syndrome, Klinefelter syndrome, and Turner syndrome.

These early studies were hindered by the inability to unequivocally identify each human chromosome, a limitation that was overcome by the introduction of chromosome banding techniques in the early 1970s. Thus, the second era of human cytogenetics, the “era of banding”, began. Banding techniques made it possible to detect small structural chromosome rearrangements, such as deletions, duplications, or translocations that previously would have gone undetected. As a result, several new syndromes associated with deletions or partial trisomies were identified in the 1970s. Further, the ability to band chromosomes resulted in the emergence of important clinical applications of chromosome testing, for example, prenatal cytogenetic analysis and cancer cytogenetics. Initially restricted to laboratories in a few academic institutions, these fields quickly grew and are now common practice in clinical cytogenetic laboratories throughout the world.


The third era, “the era of molecular cytogenetics”, began in the 1980s and continues today. It represents the fusion of conventional cytogenetic with molecular methodologies, and has led to the precise characterization of a number of new chromosomal disorders, including imprinting and other genomic disorders. The techniques that comprise molecular cytogenetics are discussed in detail in other chapters in this section (see Article 12, The visualization of chromosomes, Volume 1, Article 22, FISH, Volume 1, and Article 23, Comparative genomic hybridization, Volume 1). In the remainder of this chapter, we summarize the incidence and types of chromosome abnormalities that have been identified using these methodologies. We will focus on the first types of abnormalities that were identified (i.e., conventional numerical and structural abnormalities), as several of the more recently identified types of abnormalities (microdeletion syndromes, imprinting disorders) are discussed in later chapters (see Article 17, Microdeletions, Volume 1 and Article 19, Uniparental disomy, Volume 1).

2. The incidence and types of human chromosome abnormalities

The normal diploid number of chromosomes in humans is 46, consisting of 22 pairs of autosomes and one pair of sex chromosomes (XX for females, XY for males). Because even the smallest autosome contains at least 200-300 genes, and imbalance of even a small number of genes is known to have profound effects on normal development, it might be expected that any kind of chromosome abnormality is problematic. Indeed, with rare exceptions this is the case and, considered as a class, chromosome abnormalities are the leading known cause of pregnancy loss and congenital birth defects in our species. In the following summary, we discuss the incidence and types of chromosome abnormalities, first focusing on abnormalities detectable on conventional cytogenetic analysis and subsequently briefly considering those detectable with specialized methodology.

2.1. “Old” chromosome abnormalities: numerical and “conventional” structural abnormalities

Early cytogenetic studies made it clear that a surprisingly high proportion of human conceptions are chromosomally abnormal. The vast majority of these are numerical, that is, involving too many or too few chromosomes, while a smaller proportion are structural, that is, resulting from breakage and reunion of chromosome segments.

There are four common types of numerical abnormalities in humans: trisomy, monosomy, triploidy, and tetraploidy, which cumulatively represent the leading genetic cause of mental retardation, congenital births defects, and pregnancy loss in our species.

Trisomy (the presence of an additional chromosome, leading to a diploid number of 47 rather than 46) is clinically the most important of these categories. Approximately, 0.3% of all liveborn infants are trisomic, with the most common single abnormality being trisomy 21, the chromosome complement responsible for most cases of Down syndrome. Occurring in approximately 1/600-1/800 liveborn infants, Down syndrome is associated with moderate to severe mental retardation and with a constellation of characteristic phenotypic features; by itself, it is the leading genetic cause of mental retardation in humans.

Sex chromosome trisomies are the next most common trisomies, resulting in individuals with 47,XXX, 47,XYY, or 47,XXY (Klinefelter syndrome) chromosome complements. Of these abnormalities, Klinefelter syndrome is the most serious clinically, as it is associated with infertility and frequently, mild cognitive or behavioral abnormalities.

Other trisomies are rare, owing to the fact that the phenotypic abnormalities prevent survival to term. However, they are extremely common among clinically recognized spontaneous abortions (miscarriages), accounting for at least 25% of all such conceptions. Trisomies for most chromosomes have been identified in spontaneous abortions, although there is enormous variation in the frequency of the different trisomies. For example, trisomy 16, the most commonly identified trisomy, accounts for approximately one-third of all trisomies identified in abortions and, together with trisomies 21 and 22, for one-half of all trisomies. Other commonly observed trisomies include trisomies 2, 13, 14, and 15. In contrast, certain trisomies (e.g., 1, 5, 11, and 19) are virtually nonexistent, presumably because they house genes that are vital for embryonic/fetal development.

Because of their clinical importance, a great deal of attention has been given to the genesis of human trisomies – how they originate and factors that may influence their likelihood of occurrence. Using DNA polymorphisms to trace the inheritance of the extra chromosome, studies of the last decade have made it clear that most trisomies originate from meiotic nondisjunction – that is, errors in cell division that lead to the presence of an extra (or missing) chromosome in the gametes (eggs or sperm). While errors in both oogenesis and spermatogenesis have been identified, meiotic divisions in the development of the egg appear to be the most susceptible to nondisjunction. Indeed, over 90% of all human trisomies appear to be attributable to maternal meiotic nondisjunction, with errors at the first meiotic division (MI) being the predominant source of trisomies. While the reason for this propensity to error remains unknown, it is likely due to the unusual biology of MI in the egg: the division is initiated before birth in the fetal female ovary, but arrests until the time of ovulation some 10-15 to 40-50 years later, at which time it is finally reinitiated and completed.

While a great deal is now known about the parent and meiotic origin of human trisomies, much less is known about factors that influence their frequency. One factor stands out: the association between increasing maternal age and trisomy, which is arguably the most important etiologic factor in human genetic disease. Among women under the age of 25, approximately 2% of all clinically recognized pregnancies are trisomic; by the age of 36, however, this figure increases to 10% and by the age of 42, to over 33%. This association between maternal age and trisomy is exerted without respect to race, geography, or socioeconomic factors. Thus it appears to be “hard-wired” into our species, although the evolutionary benefit that this provides, and its underlying molecular basis, remain obscure.

Until recently, increasing maternal age nondisjunction was the only factor known to influence the frequency of nondisjunction. However, studies of the past decade have finally detected a second contributing factor: abnormal meiotic recombination. Recombination is a process that occurs during the MI, involving pairing, synap-sis, and exchange (crossing-over) between homologous pairs of chromosomes; it is discussed in more detail in another chapter (see Article 13, Meiosis and mei-otic errors, Volume 1). Recombination results in the exchange of genetic material between homologs, thus acting to increase genetic diversity within a species. However, it has a second, sometimes overlooked, function. Specifically, the physical exchange of material serves to tether the homologs together during prophase of meiosis I, thus facilitating their alignment on the metaphase I plate and increasing the chances of proper segregation at the first division.

From studies of model organisms, it had long been known that alterations in recombination increased the likelihood of nondisjunction. With the advent of DNA polymorphism analysis in the 1990s, it became possible to study the effects of recombination on human nondisjunction, and to ask whether, as in model organisms, human trisomies were associated with altered recombination. The results have been compelling. For all human trisomies yet studied, aberrant meiotic recombination is a major contributing factor. To date, three types of abnormal recombination patterns have been identified: (1) complete failure of recombination between homologs, leaving the homologs to wander at random at MI, (2) recombinational events in which the homologs are held together only at the tips of the chromosomes, presumably leading to “weak” connections that are unable to sufficiently bind the homologs together, and (3) recombinational events that occur close to the centromeres of the chromosomes, possibly leading to inter-homolog connections that are too “strong”, preventing them from separating from one another at the first division.

The demonstration of a link between abnormal meiotic recombination and tri-somies has provided the first molecular correlate for human nondisjunction. Intensive efforts are now underway to determine why such abnormal recombinational patterns occur in the first place, and whether these patterns may be responsible for some proportion of maternal age-dependent human trisomies.

Monosomy (the absence of a single chromosome, leading to a diploid number of 45 rather than the normal 46) is sometimes thought of as the “flip side” of trisomy. That is, a nondisjunctional event should lead to a gamete with a missing chromosome, as well as a gamete with an additional chromosome. Thus, it might be expected that monosomies and trisomies would occur in equal frequency. However, this turns out not to be the case, as monosomies are exceedingly rare. Indeed, only one monosomy, sex chromosome monosomy (45,X; the chromosome complement associated with Turner syndrome), is identified in any frequency. It is identified in approximately 1/20 000 newborns, who develop as phenotypic females, but with short stature, streak gonads, and infertility. For reasons that are not clear, the 45,X chromosome complement is much more frequent in miscarriages where it accounts for nearly 10% of all such conceptions.

Why are monosomies so rare? Likely, they occur at appreciable frequency but are eliminated very early in gestation because of the detrimental effects of haplosufficiency, that is, the presence of only one, rather than two, copies of a gene(s). Haploinsufficiency for a single gene is frequently associated with fetal demise or severe birth defects; therefore, it is not surprising that haploinsufficiency for an entire chromosome would be incompatible with embryonic development. Indeed, the fact that 45,X conceptions frequently survive long enough to be detected as clinically recognizable miscarriages or, much more rarely, as liveborn infants, is presumably attributable to X-chromosome inactivation (see Article 15, Human X chromosome inactivation, Volume 1). This is the process whereby one of the two X chromosomes in normal female cells is silenced, leaving only one functional, expressed copy for most X-linked genes. As no such process exists for autosomes, autosomal monosomies are much more lethal than is the 45,X condition.

Triploid conceptions have an additional whole haploid set of chromosomes, that is, 69 rather than 46 chromosomes. They are almost never identified in liveborn infants but are extremely common in miscarriages, accounting for nearly 10% of such conceptions. Triploidy can result from one of two processes: (1) failure of a meiotic division, resulting in a gamete with 46 chromosomes, for example, failure of MI in oogenesis, followed by fertilization by a normal sperm, leads to triploidy of maternal origin or (2) penetration of a normal egg by two sperm (“dispermy”), yielding triploidy of paternal origin.

Both types of triploidy have been identified but, intriguingly, the phenotypes are different: for example, paternal triploids are characterized by the overgrowth of the placenta, while maternal triploids are not. This “parent-of-origin” difference in phenotype is an example of an imprinting effect, which is discussed in more detail later (see Article 19, Uniparental disomy, Volume 1).

Tetraploid conceptions have two extra sets of chromosomes, that is, 92 rather than 46 chromosomes. These are identified only in miscarriage samples and apparently almost always result from failure of an early mitotic cleavage division.

2.1.1. Structural chromosome abnormalities

Structural rearrangements involve breakage and reunion of chromosomal material. Although much less common than numerical abnormalities, they have a special clinical significance since they can be inherited through generations. That is, for some types of structural rearrangements, clinically normal individuals may carry the rearrangement in “balanced” form but transmit an “unbalanced” form to their progeny, leading to phenotypic abnormalities. This is in contrast to numerical abnormalities, which arise de novo in meiosis or in an early mitotic division, and thus are not passed on from generation to generation.

Rearrangements may be interchromosomal (involving exchange of material between different chromosomes) or intrachromosomal (involving the reordering, loss or gain of genetic material on a single chromosome).

The most important classes of interchromosomal rearrangement are transloca-tions, which are categorized as reciprocal or Robertsonian. Reciprocal transloca-tions simply refer to situations in which material is exchanged between different chromosomes, resulting in “hybrid” chromosomes containing segments of each of the original nontranslocated chromosomes. By themselves, they do not cause problems, since the individual carrying them may have the same genetic material as a chromosomally normal individual. However, there are at least two ways in which translocations may result in phenotypic abnormalities: (1) the breaks that led to the translocation events may be complex, leading to loss of material surrounding the break points and consequently to genetic imbalance and (2) during meiosis in individuals heterozygous for translocations, complex structures link the translocated chromosomes with their structurally normal homologous partners, making it possible that the wrong combination of translocated and nontranslocated chromosomes will pass together into the gametes. This results in genetic imbalance for the translocated portions, which, depending on the length of the translocated segments, can have disastrous phenotypic consequences.

For the most part, reciprocal translocations occur randomly throughout the genome, that is, there are no “hot spots” for reciprocal exchanges. However, there is one notable exception to this rule: the 11;22 translocation. First described in 1980, the translocation involves the exchange of material between the long arm of chromosome 11 (at q23) and the long arm of chromosome 22 (at q11). Balanced carriers of the 11;22 translocation are themselves clinically normal, but if their progeny inherit the translocation chromosome 22 as an “extra” chromosome, they manifest the “der 22 syndrome”, with a distinctive set of abnormal phenotypic features (e.g., with abnormalities of the ear, heart defects, cleft palate, and mental retardation).

Unlike other reciprocal translocations, the points of exchange are identical among the several hundred different families in which the 11;22 translocation has been identified. Thus, the 11;22 translocation is a rare example of a recurrent reciprocal translocation, involving regions of chromosomes 11 and 22 that are susceptible to rearrange with one another. Recent molecular analyses of these regions has elucidated the reason for this: the breakpoints on each chromosome are located within similar AT-rich palindromic repeats, creating cruciform DNA structures that presumably facilitate the illegitimate transfer of material between the nonhomologous chromosomes during meiosis.

Robertsonian translocations are a special class of reciprocal translocation, in which the long arms of two acrocentric chromosomes (chromosomes 13, 14, 15, 21, and 22) are fused, creating a chromosome that contains virtually all of the genetic material of the original two chromosomes. If the Robertsonian translocation is present in unbalanced form, a monosomic or trisomic conception results. For example, approximately 3% of cases of Down syndrome are attributable to unbalanced Robertsonian translocations, most often involving chromosomes 14 and 21. In this instance, the affected individual has the normal number of chromosomes (46) but in addition to two structurally normal chromosomes 21, carries a translocation 14/21 chromosome. This results in triplication for genes on the long arm of chromosome 21, leading to phenotypic abnormalities identical to those associated with conventional trisomy 21. Similarly, unbalanced Robertsonian translocations involving chromosome 13 account for a small proportion of individuals with trisomy 13 syndrome.

In addition to interchromosomal rearrangements, there are several types of intrachromosome structural abnormalities, including inversions (in which a segment of the chromosome is “flipped”, so that the orientation of genes is in the opposite orientation from that on a structurally normal chromosome), isochromosomes (in which one part of the chromosome is a mirror image of another), rings (in which breaks at both ends of the chromosome are sealed together, forming a circular chromosome), duplications (in which a segment of the chromosome is present in two or more copies), and deletions (in which a segment is lost). Of these, deletions are clinically the most significant. Two of the best characterized deletions, involving loss of material on chromosome 4p and 5p, result in Wolf-Hirschhorn syndrome and cri-du-chat syndromes, respectively. In both instances, the deletion involves the loss of a relatively small chromosomal segment; nevertheless, both are associated with multiple congenital anomalies, profound mental retardation, and reduced life span.

2.2. “New” chromosome abnormalities: microdeletion and imprinting syndromes

In 1986, Roy Schmickel coined the term “contiguous gene syndromes” to describe genetic disorders that mimicked single-gene disorders but that resulted from the deletion of a small number of tightly clustered genes. It was proposed that many such deletions would be so small as to be undetectable cytogenetically, that is, they would be “microdeletions”. Over the past two decades, the increasing application of molecular techniques has led to the identification of over 20 of these microdeletion syndromes; many of these have parent-of-origin effects (“imprinting disorders”). Several of these are discussed in later chapters (see Article 17, Microdeletions, Volume 1 and Article 19, Uniparental disomy, Volume 1), and will not be considered in detail here. However, in the following discussion, we focus on two types of these disorders, emphasizing some of the general principles that have emerged from their analysis: a microdeletion syndrome (velocardiofacial syndrome, VCF) and two complementary disorders exhibiting parent-of-origin effects (Angelman and Prader-Willi syndromes).

Deletions of chromosome 22 (at 22q11) are among the most common microdele-tions yet studied, as they are detected in approximately 1/3000 newborns. The most common syndrome associated with these deletions, VCF syndrome, includes a number of phenotypic defects, for example, learning disabilities or mild mental retardation, palatal defects, and congenital heart defects. Occasionally, individuals with a microdeletion of 22q11 deletion are more severely affected and are diagnosed as having Disyndrome, which involves abnormalities in the development of the third and fourth branchial arches, leading to thymic hypopla-sia, parathyroid hypoplasia, and conotruncal heart defects. In approximately 30% of these cases, a deletion at 22q11 can be detected with high-resolution banding; by combing conventional cytogenetics, fluorescence in situ hybridization (FISH) (see Article 12, The visualization of chromosomes, Volume 1 and Article 22, FISH, Volume 1) and molecular detection techniques (i.e., Southern blotting or polymerase chain reaction analyses), these rates improve to over 90%. Additional studies have demonstrated a surprisingly high frequency of 22q11 deletions in individuals with nonsyndromic conotruncal defects. Approximately 10% of individuals with a 22q11 deletion inherited it from a parent with a similar deletion.

Prader-Willi syndrome (PWS) and Angelman syndrome (AS) both involve abnormalities of chromosome 15. In 1981, David Ledbetter described a series of patients with PWS, a proportion of whom had a cytogenetically detectable deletion between 15q11 and 15q13. Subsequently, a number of AS patients were observed to have a deletion in the same region. This was curious, because the syndromes are very dissimilar – PWS is characterized by obesity, hypogonadism, and mild to moderate mental retardation, while AS is associated with microcephaly, ataxic gait, seizures, inappropriate laughter, and severe mental retardation. The resolution to this question came some years later, when it was recognized that the parental origin of the deletion determines which phenotype ensues, that is, if the deletion is paternal, the result is PWS and if the deletion is maternal, the result is AS. However, an additional complication to this story arose when it was recognized that not all individuals with PWS or AS carry the deletion; in fact, in both syndromes, many of the patients have two normal chromosomes 15. For most such individuals, it turns out that the parental origin of the chromosomes 15 is again the important determinant. In PWS, nondeletion patients invariably have two maternal and no paternal chromosomes 15, while for most nondeletion AS patients the reverse is true. This indicates that at least some genes on chromosome 15 are differentially expressed, depending on which parent contributed the chromosome. Additionally, this means that normal fetal development requires the presence of one maternal and one paternal copy of chromosome 15. Errors resulting in a deviation from this situation (e.g., fertilization of an egg with a 24, X,+15 karyotype by a sperm with a 22, X,-15 karyotype) will result in an abnormal phenotype, either PWS or AS.

Chromosomes that behave in this manner are said to be imprinted, and it is likely that several – although certainly not all – human chromosomes are imprinted. Chromosome 11 must be one of these, since it is now known that a proportion of individuals with the overgrowth syndrome Beckwith-Wiedeman syndrome have two paternal, but no maternal, copies of this chromosome.

In addition to microdeletion and imprinting syndromes, there is now at least one well-described microduplication syndrome. This is Charcot-Marie-Tooth syndrome Type 1A (CMT1A), a nerve conduction disease previously thought to be inherited as a simple autosomal dominant disorder. Molecular studies conducted in the 1990s demonstrated that affected individuals are heterozygous for a duplication of a small region of chromosome 17 (17p11.2-12). This makes the reason for the inheritance pattern clear: one-half of the offspring of affected individuals would inherit the duplication-carrying chromosome.

3. Summary

Chromosome abnormalities occur with astonishing frequency in our species, with an estimated 20-30% of all fertilized eggs having too many or too few chromosomes, or carrying structurally abnormal chromosomes. In the following chapters, we discuss the techniques that have led to the identification of these disorders, and describe the phenotypic outcomes associated with different classes of chromosome abnormality. Additionally, for those abnormalities in which the information is available, the molecular basis for the abnormalities is described. However, it will quickly become apparent that for most abnormalities, this information is simply not there. One of the major challenges in human genetics is the identification of the molecular bases for these conditions, and the identification of etiological factors that influence their occurrence.

Next post:

Previous post: