Chromosome Abnormalities and Genetic Counseling , 3rd Edition

5.Sex Chromosome Translocations

The sex chromosomes (gonosomes) are different, and sex chromosome translocations need to be considered separately from translocations between autosomes. A sex chromosome can engage in translocation with an autosome, with the other sex chromosome, or even with its homolog. The unique qualities of the sex chromosomes have unique implications in terms of the genetic functioning of gonosome–autosome translocations. Unlike any other chromosome, the X is capable of undergoing “transcriptional silencing.” This fact has crucial consequences for the carriers of X-autosome translocations, both the balanced and the unbalanced. And unlike any other chromosome, the Y is composed of chromatin which is, in large part, permanently inert. Some translocations of this inert material can thus be of no clinical significance.

BIOLOGY

The X-Autosome Translocation

Both females and males can be heterozygous (or hemizygous) for an X-autosome translocation, in balanced or unbalanced state. But the implications for the two sexes are quite different, thus we need to treat the two cases separately. With respect to the female, a fundamental requirement for the counselor is to have an outline understanding of X chromosome inactivation.

THE X CHROMOSOME AND THE X-AUTOSOME TRANSLOCATION IN THE FEMALE

The normal female has two X chromosomes, and yet the possession of only a single X is sufficient to produce normality in the 46,XY male. Are the sexes really so genetically different? Does the female really need a second X? The answer is a qualified no. The second X is largely surplus and is subject to transcriptional silencing very early in the life of the female. At around the second week following conception one of the X chromosomes in every cell of the female conceptus is randomly genetically inactivated, a process called (after Dr. Mary Lyon) lyonization. In all descendant progeny cells following this event, the same X chromosomes remain inactive or active, respectively. This dosage compensation allows for a functional monosomy of most of the X chromosome. Transcriptional silencing is initiated at an X inactivation center (XIC) in Xq13 (Fig. 5-1) and spreads in both directions along the chromosome. Within the XIC is a gene, XIST, which is cis-acting (that is, it can influence only the chromosome that it is actually on) and is transcribed only from the inactivated X. This transcript, named “XIST” for X (inactive) specific transcript, is not translated into protein but functions as an RNA molecule. The XIST RNA “coats” the X chromatin, although how this might actually cause the coated genes to become silenced remains speculative (White et al., 1998b). Indeed, Lyon herself (1998) plainly comments that “the mechanism of X-chromo-some inactivation remains largely a mystery.” Possible factors include degrees of methylation and acetylation of the DNA, and the extent of chromatin folding. Normal women can have quite skewed ratios of active Xm and inactive Xp chromosomes1 and vice versa, even more than 90:10, and there can be differences in ratios between different tissues in a woman (Sharp et al., 2000). The inactive X replicates late during the cell cycle; the active X replicates early, along with the autosomes.

Figure 5-1. Notable regions of the sex chromosomes. AZF, azoospermia factor regions a–d. Dots show specific loci: DAZ, deleted in azoospermia locus; MLS, microophthalmia with linear skin lesion gene; SHOX, short stature homeobox gene (X chromosome); SHOXY, short stature homeobox gene (Y chromosome); SRY, testis-determining locus.

But this is not to say that the female's second X chromosome is unnecessary (a rather obvious statement, considering the difference between 46,XX and 45,X individuals). Not all genes on the X chromosome are inactivated and thus some loci are functionally disomic in the normal female. There is a block to the spread of inactivation into the primary pseudoautosomal region (PAR), which comprises the terminal 2.6 megabases of Xp in band p22.3; this segment has a homologous region on distal Yp (Fig. 5-1). There is a secondary PAR that extends over 320 kb within distal Xq, having homology with distal Yq (Kvaløy et al., 1994). An obligate recombination event occurs in the primary PAR of the X and the Y chromosome at male meiosis; recombination between the secondary PAR, if it occurs at all, is infrequent. Certain other loci elsewhere on the X than in the PAR (some of which have homologs on the Y) are not subject to inactivation, and disomic expression of these genes in the female (and, for some, in the male) is normal (Disteche, 1995).

Patterns of Inactivation, and Phenotypes, in the Balanced Translocation Carrier

The balanced X-autosome translocation carrier has two translocation chromosomes, the der(X) and the der(autosome). The X segment in one of these, most commonly the der(X), contains the XIC, and the X segment in the other, usually on the der(autosome), lacks the XIC. The latter segment, having no XIC of its own and being beyond the influence of the XIC on the other derivative, is always active. The only way, then, for the karyotypically balanced female X-autosome heterozygote to achieve a functionally balanced genome is to use, as her active X complement, the two parts of the X in the two translocation chromosomes: together, they add up to an equivalent whole, and functioning, X chromosome. The other chromosome, the normal X, is inactive. The cartoon karyotype in the 46,X,t(X;12) carrier mother in Fig. 5-2 shows the normal X as inactive (dotted outline), and the X-segments of the der(X) and der(12) as active (solid outline).

Probably, the mechanism to bring about this asymmetric inactivation is as follows. Inactivation is initiated at random in each cell, at either of the XICs. Some cells will be functionally balanced, with the intact X inactive, as described above. Others, in which the intact X is active, will have a functional X disomy, due to the X segment on the der(autosome) being outside the reach of transcriptional silencing. According to this theory, cell selection then eliminates the functionally disomic X lines (Fig. 5-3, sequence a → b → c). This mechanism is successful in a fraction of heterozygotes, and aside from a possible gonadal effect (see below), such individuals are phenotypically normal.

This mechanism, as it would seem, may not infrequently fail, and phenotypic abnormality is the consequence. “Not infrequently” may translate as about 25%, with reference to the literature study of Schmidt and Du Sart (1992), but a more valid figure would come from unbiased data. If some functionally disomic cells survive and come to comprise part of the soma (Fig. 5-3, sequence a → b → d), this would presumably have some deleterious effect. The natural prediction is that only cells with small disomies would be capable of survival. Thus, we might more commonly expect to observe in these affected carrier females translocation breakpoints in distal Xp or distal Xq (Xp22 and Xq28), which would impart disomy for only a very small segment of either distal X short arm or distal X long arm. This was indeed the observation in the reviews of Schmidt and Du Sart (1992) and Du Sart et al. (1992). To the contrary, however, the X breakpoints in the data assembled from a survey conducted by the U.K. Association of Clinical Cytogeneticistsshowed a wider distribution, and the X and der(X) did not necessarily display a skewed inactivation (Waters et al., 2001; A. J. M. Crocker, pers. comm., 2001). There was no particular predilection for site of breakpoint, other than Xq22 being overrepresented. These workers believe that their data may be more representative, having avoided a reporting bias. Given their observations, they propose that gene disruption may be similarly important as a factor causing phenotypic abnormality. Part of the difference between the two studies may lie in the nature of the presenting phenotypes, whether these be malformation/cognitive compromise or merely compromise of ovarian function.

Figure 5-2. Inactivation patterns. Heterozygous mother with a balanced X-autosome translocation, showing patterns of inactivation in herself and in her two chromosomally unbalanced children with partial Turner and partial Klinefelter syndromes, respectively. The dashed outline indicates inactivated chromosome. The inactivation pattern of a theoretical third child with a partial X trisomy is shown at right. Note that the balanced carrier inactivates her normal X chromosome, while it is the abnormal X which is inactivated in the unbalanced offspring (and, in the third child, one of the additional normal X chromosomes as well). Based on family in Figure 5-4.

Figure 5-3. Skewing or non-skewing of X chromosome inactivation, as a theoretical explanation for the X-autosome carrier being of either normal or abnormal phenotype (see text). (a) Before X inactivation occurs, both the normal and the der(X) are active in all cells (shown in light gray). (b) X inactivation occurs as a random, cell-autonomous process. Cells shown in white have the der(X) as the active chromosome, and thus the genetic activity of these cells is balanced with respect to X chromosomal output. The cells shown in dark gray have the normal X active, and consequently their X chromosomal activity is imbalanced, because of the additional output from the X-segment of the der(autosome). Subsequently in embryonic development one of two events occurs. (c) The cells with the normal X active (dark gray) die out, due to their functional genetic imbalance, leaving only the cells with the der(X) active (white). These latter cells functionally are genetically balanced, and the phenotype is normal. The individual has a skewed X-inactivation pattern. Or (d) the dark gray cells persist, in spite of their functional genetic imbalance (the defect is not severe enough to be lethal), and the individual is a mosaic of functionally balanced tissue (white cells) and imbalanced tissue (dark gray cells). Consequently, the phenotype is abnormal. (Adapted from Lanasa and Hogge, 2000.)

Measuring Inactivation Status. Inactivation status can be assessed cytogenetically and molecularly. Replication banding enables, in principle, distinction of the early replicating (active) and the late replicating (inactivated) X chromosomes (e.g., Fig. 5-11). Only a small number of cells, a hundred or so, can realistically be studied by this technique, and only from the tissue represented by the sample taken (usually blood, occasionally amniotic fluid cells). Molecular methodology enables large numbers of cells to be analyzed, which allows a more precise estimate of the ratio of normal-Xactive to translocation-Xactive cells.2In this respect, the androgen receptor locus, at Xq13 (quite close to the XIC), is often used. The observation of a complete skew of translo-cation-Xactive and normal-Xactive, in the representative tissue analyzed, would indicate that the same 100:0 proportion applied elsewhere in the soma, at least in the phenotypically normal heterozygote. Since it is impossible to test the entire soma (and in particular the brain), it would have to remain an open question in a phenotypically abnormal but structurally balanced X-autosome heterozygote that a more random skewing pattern might apply in some tissues, notwithstanding a complete skew in the peripheral tissue(s) tested. Abnormal individuals may show incomplete intertissue concordance of inactivation status, with sometimes quite different ratios in different tissues—e.g., 80:20 in blood and 30:70 in skin (Schmidt and Du Sart, 1992).

Patterns of Inactivation in the Unbalanced Female Offspring

As a rule (but one that can be broken), the pattern of inactivation that occurs will be the one that allows the least amount of functional imbalance. This is typically arrived at in the karyotypically unbalanced daughter by inactivation of the abnormal chromosome, always supposing that the choice exists (and the choice can exist only if the abnormal chromosome contains an XIC). If the abnormal chromosome is a der(X) from a single-segment exchange, containing no autosomal material other than a telomeric tip, it comprises, essentially, a deleted Xp or Xq chromosome. In a girl with the 46,X,der(X) karyotype, preferential inactivation of this deleted X leads simply to a phenotype of partial Turner syndrome. Consider the family segregating a t(X;12) shown in Figs. 5-2 and 5-4. The segregation shown in Fig. 5-2 (daughter from adjacent-1) and Fig. 5-4a (daughter) illustrates the case for an Xq deletion. Here, the normal X is active (shown as solid outline in Fig. 5-2), and the der(X) inactivated (dotted outline). Leichtman et al. (1978) provide an example of the Xp deletion circumstance in a three-generation family with seven persons having an Xp Turner syndrome variant on the basis of a segregating t(X;1).

Figure 5-4. (a) Mother with balanced X;12 translocation, showing two different segregant outcomes. Her daughter presented with clinical Turner syndrome, in whom the karyotype was initially interpreted as del(X)(q22). Her son was subsequently studied, and he had a partial Klinefelter syndrome. (Case of J. A. Sullivan.) (b) The presumed pachytene configuration during gametogenesis in the mother (X chromatin is open, no. 12 and Y chromatin is cross-hatched; dot indicates X inactivation center). Light arrows indicate movements of chromosomes to daughter cells in adjacent-1 segregation, as observed in the daughter with partial Turner syndrome. Heavy arrows show the tertiary trisomy combination seen in the son with partial Klinefelter syndrome. These two segregations are represented in b and c in Figure 5-6.

If the der(X) carries a large translocated autosomal segment, conferring, therefore, a partial autosomal trisomy in the 46,X,der(X) subject, the effects of this imbalance may be mitigated by selective inactivation of the abnormal chromosome. Transcriptional silencing can spread into the autosomal chromatin on the der(X), converting, at least partially, a structural autosomal trisomy into a functional autosomal disomy (Kulharya et al., 1995; Garcia-Heras et al., 1997; Garcia-Heras et al. 1998; Orellana et al., 2001; Sharp et al., 2002). The de novo cases described in Garcia-Heras et al. (1997) and Orellana et al. (2001) are notable, with abnormal children being trisomic for large lengths of chromatin, much of 10q and most of 14q respectively, that would otherwise surely have been lethal in utero. An example of a familial case is that of Garcia-Heras et al. 1998, who report on the terminated pregnancy of an X;15 carrier mother. The 19-week fetus was trisomic for almost all of 15q, and while trisomy 15 typically causes first trimester abortion, here only rather mild abnormalities of fetal morphology were to be noted. Presumably, the “improvement” is achieved by the autosomal material being coated with XIST, flowing over from the X component of the der(X). The extent of spread of this sort of “protective” inactivation is variable, and its benefit (or not) unpredictable, and this may relate to variable loss of the XIST autosomal coating (Hall et al., 2002). Figure 5-8 shows an example of blocked spread of inactivation into the autosomal (16p) segment: observe the der(X) in the lower row, with pale (inactivated) long arm and dark (active) short arm. Autosomal chromatin, or certain parts thereof, may be resistant to the inactivating influences to which the X chromatin is responsive (Sharp et al., 2001).

A good illustration of this scenario is given in Petit et al. (1994). These authors describe the case of a normal grandmother with a balanced t(X;4)(p22.1;p14) who had a der(X)t(X;4) daughter in whom the additional 4p material on the X;4 translocation chromosome was inactivated in all cells studied (except for the most distal band, 4p16). The daughter's only significant problems were social and personality difficulties; the physical phenotype was unremarkable. Usually, of course, partial trisomy 4p would produce a severe phenotype. Further proving the point, this woman herself had a daughter with the same unbalanced karyotype, the same inactivation pattern, and phenotypic abnormalities essentially confined to personality. White et al. (1998b) report a similar circumstance, in this case in a phenotypically normal woman with a menstrual irregularity who proved to have the karyotype 46,X, der(X)t(X;4)(q22;q24). The two structural imbalances conferred by this translocation derivative are a deletion of about half of Xq and a duplication of a little over half of distal 4q. These researchers measured directly the activity of 20 genes and expressed sequence tags (ESTs) in the translocated autosomal segment, and showed that 14 of them were not expressed and 6 were. The expressed genes/ESTs were located at various points in the 4q segment, indicating that the inactivation process, as it progressed along the chromosome, “jumped across” some regions. This 4q segment accounted for 2.8% of the haploid autosomal length—a very large imbalance, and one that otherwise would surely have caused embryonic lethality. It would rather seem that all, or practically all, of the dosage-sensitive genes in this substantial segment of chromatin had been inactivated, and the minority of genes that yet remained active produced no ill effect due to an overall 150% activity on their part. Yet another case to solidify the argument is that of Sharp et al. (2001), who studied an aunt and niece with 46,X,der(X),t(X;10)(q26.3q23.3), the mothers being 46,X,t(X;10) carriers. Four genes studied on the translocated segment of 10q origin on the der(X) were inactive, and one (the most telomeric) active. The above three examples notwithstanding, the probability of abnormality in most cases will be high.

The converse, whereby the process of spreading autosomal inactivation may be detrimental to the phenotype, by converting a disomy (or near disomy) into a functional monosomy, is rarely observed. The family illustrated in Figure 5-10 provides a possible example. At first, one might have expected only a Turner syndrome phenotype in the daughter with a 45,X,der(X),-22 karyotype, since the essential defect appeared to comprise an Xp deletion, with her total complement of 22q material being intact or nearly so. However, a more severe clinical picture evolved, and this may have reflected, speculatively, a transcriptional silencing of some crucial 22q loci, despite the apparent block to inactivation at the breakpoint on cytogenetic study. This case is mentioned further below.

If the derivative chromosome has no XIC, the karyotypic imbalance cannot be modified. This circumstance typically applies in the girl with a 46,XX,der(autosome) karyotype: the translocated segment of X origin on the der(au-tosome) cannot be inactivated, and a functional partial X disomy is the consequence (Sivak et al., 1994). Figure 5-5 demonstrates a functional disomy3 for a part of Xp (Xp22.31–pter) in a chromosomally unbalanced daughter; in this instance, since the autosomal breakpoint is at the telomere, we assume there to be little or no effect from a 10q monosomy. Gustashaw et al. (1994) describe a similar case in which the Xp imbalance was surely the sole cause of the abnormal phenotype, since the autosomal breakpoint was in 13p and the loss of one acrocentric short arm has, of itself, no effect.

Figure 5-5. Functional X disomy. (a) Mother with balanced X;10 translocation (upper) and her daughter with a 46,XX,der(10) karyotype from adjacent1 segregation (lower). The translocation is t(X;10)(p22.31;q26.3). Dashed box on cartoon karyotype indicates preferentially inactivated chromosome; dot indicates X inactivation center. The der(10) contains Xp material in the translocated segment, which cannot be inactivated, and so the daughter has functional X disomy. Since the 10q breakpoint is in the terminal band, we may regard this as an effectively single-segment exchange, with the phenotype of severe mental deficit and minor dysmorphism due entirely to disomy for the small Xp22.31-pter segment. (Case of A. Ma and H. R. Slater.) (b) The presumed pachytene configuration during gametogenesis in the mother (X chromatin is open, no. 10 chromatin is crosshatched; dot indicates X inactivation center). Arrows indicate movements of chromosomes to daughter cells in adja-cent-1 segregation; heavy arrows show the combination observed in this family. This is essentially the segregation a shown in Figure 5-6.

Ovarian Function and the X Critical Regions

Breakpoints at certain locations in the X may affect ovarian function. A breakage and reunion within two “critical regions” is characteristically associated with gonadal dysgenesis. These regions are Xq13–q22 and Xq22–q27, separated by a narrow region within Xq22 that is not critical (Fig. 5-1) (Therman et al., 1990). About one-third of all carriers in the U.K. collaborative study presented with a disorder of ovarian functioning, and in these women the breakpoint was typically in a critical region (Waters et al., 2001). The ovarian disorder ranged from an early age of menopause, generally before 40 years (this being defined as premature ovarian failure), through to failure of onset of menstruation at the normal age of puberty (primary amenorrhea). Premature ovarian failure is particularly associated with breaks in Xq26.1–q27 and Xq13.3–q21.1, and there may be gene clusters in these regions that determine ovarian activity, although actual disruption of such genes is not likely to be a common mechanism (Marozzi et al., 2000; Prueitt et al., 2002). One possible mechanism is the impairment of X chromosome activity in meiosis or mitosis, leading to an accelerated rate of oocyte atresia (Layman, 2002). However, the U.K. study noted that some women with X-autosome translocations having a breakpoint in these regions were still fertile, as the data in Table 5-1 also attest. In one series of 30 women presenting with premature ovarian failure in whom the cytogenetic findings were reviewed, Devi and Benn (1999) recorded one to be an X-autosome translocation heterozygote, thus, an infrequent cause of this problem. The affected relatives of this woman showed quite some variation in their ages of menopause.

COMPROMISE OF GAMETOGENESIS IN THE MALE

Almost invariably, the male hemizygote is infertile, due to spermatogenic arrest (Kalz-Füller et al., 1999). In two men subject to testicular biopsy, Quack et al. (1988) showed germ cell maturation arrest mostly at the pachytene stage of meiosis I, although a few cells managed to make the first and some even the second meiotic metaphase. Disruption of the sex vesicle (see below, Y-Autosome Translocations) is the presumed mechanism (Schmidt and Du Sart, 1992).

Table 5.1. Occurrence of Gonadal Dysgenesis (Primary and Secondary) in t(X-autosome) Women according to X Chromosome Breakpoint

Breakpoint

Gonadal dysgenesis

Normal gonadal function

Xpter–q12

5

37

q13

4

8

q13–q22

20

1

q22

11

6

q22–q25

7

1

q26

3

5

q27–qter

1

9

Source: From Therman et al. (1990).

MEIOSIS

In oogenesis, a quadrivalent presumably forms, just as in the two-way translocation between autosomes. Figures 5–6 and 5–7 set out certain segregant outcomes that may be viable in gametes produced by the female heterozygote, for various categories of single-segment and double-segment translocation, as discussed below. Given the greater survivability of X imbalances due to inactivation and likewise a possible lessened effect of autosomal imbalance, a greater number of conceptuses are potentially viable than from the autosome–autosome translocation. The “rules” of segregation (p. 68) may not apply; for example, a viable adjacent2 malsegregation can occur with a derivative chromosome having a large centric segment. The coexistence of tertiary monosomy and adjacent-2 aneuploidy in the family described in Figure 5-10, two otherwise very uncommon segregations, reflects the unique characteristics of the X-autosome translocation.

In the male, spermatogenesis is completely disrupted, as noted above, with infertility being the consequence. It may thus be rather academic, although not without interest, to note that a quadrivalent configuration may indeed form, comprising the der(X), the der(autosome), the normal autosome, and the Y (Quack et al., 1988). The question might become less academic if such a man were to be considered for assisted reproduction.

Categories of Translocation and Modes of Segregation

We consider here various chromosomal scenarios, which ought to cover the majority of clinical circumstances. As for terminology related to the size of translocated segments, if one of the translocation breakpoints is at the telomeric tip of either the autosome or the X chromosome, and thus only one of the translocated segments (X or autosomal) comprises an important amount of chromatin, this may be considered an effective single-segment exchange. If both translocated segments are of significant size, this is a double-segment exchange.

Single-Segment Exchange, X Translocated Segment

The first two columns in Figure 5-6 and the first column in Figure 5-7, segregations ac and segregation a, respectively, depict the general form of a translocation in which the single important exchanged segment comprises X chromatin. A particular example is shown in Figure 5-4, in which the derivative X chromosome is deleted for a large segment of Xq, and has only the telomeric tip of 12p in exchange. A child receiving this abnormal “Xq-” in place of a normal X, or as an additional chromosome, could present with a partial form of a gonosomal aneuploidy syndrome. Thus a daughter with 46,X,der(X) from adjacent-1 malsegregation (b in Fig. 5-6) would have a variant form of Turner syndrome. From tertiary trisomy (c in Fig. 5-6), a son with 47,XY,der(X) would have incomplete Klinefelter syndrome; and a 47,XX,+der(X) daughter might show the 47,XXX phenotype to a diminished degree.

Figure 5-6. Major categories of adja-cent-1 and 3:1 malsegregation in the X-autosome female carrier. The top row shows quadrivalents at maternal meiosis, and the following rows various combinations of segregant products. X chromatin is shown open, autosomal chromatin is cross-hatched. The dot indicates X inactivation center. Single-segment and double-segment are defined in the text. X exchanges can occur in either Xp or Xq; only Xq exchanges shown here. Circled letters provide reference points for text comments. *Effect of autosomal duplication may be lessened by spreading of transcriptional silencing into the autosomal segment of the der(X). **Blocking of spread of inactivation into the autosomal segment of the der(X) may avoid further functional autosomal monosomic effect.

More severe consequences follow the countertype adjacent-1 segregation, a in Figure 5-6. In the family in Figure 5-4, conceptions with 46,der(12) from adjacent-1 segregation would be functionally disomic for a large and unsurvivable amount of Xq and would abort. However, if the translocated X segment is small, the functionally disomic X state may be viable. This is shown in Figure 5-5, in which the mentally retarded and dysmorphic daughter has a 46,XX,der(10) karyotype and is functionally disomic for the small amount of Xp22.31–pter.

As for adjacent-2 segregation (Fig. 5-7, a), such a gamete would, in theory, have viability only if it is the der(autosome) which is transmitted, along with the intact autosome, and if the X segment of the der(autosome) includes the XIC. In that case, inactivation could spread through the autosomal material, converting, at least partially, a structural autosomal trisomy into a functional autosomal disomy. Of course there would be a partial X monosomy as well. This scenario is discussed in more detail in the section on double-segment exchange, below.

A truly single-segment X-autosome translocation, the translocated segment comprising X material, is recorded in de Vries et al. (1999). A mother had a submicroscopic segment of the PAR in distal Xp (p22.31–pter) translocated across to the short arm of a no. 14, but, as far as could be seen, there was no reciprocal movement back to the X of any 14p material. She transmitted the der(X) to a son, who presented signs interpreted as consistent with nullisomy for certain genes in the distal PAR: the SHOXMRXCDPX and STS genes, their absence being responsible respectively and collectively for short stature, developmental delay, short limbs, and ichthyosis.

Figure 5-7. Three categories of adjacent-2 malsegregation in the X-auto-some female carrier. The top row shows quadrivalents at maternal meiosis, and the next row various combinations of adjacent-2 segregant products. Note that these potentially viable outcomes occur only in the setting of the transmitted derivative chromosome, be it the der(X) or the der(autosome), having an X inactivation center (XIC). In the first two columns, the der(auto-some) has the XIC; here, the X breakpoint must be in proximal Xq, above the XIC, as depicted. In the third column, in which the der(X) has the XIC, X exchanges can occur either in Xp, or in Xq distal to the XIC; only an Xp exchange is shown here. X chromatin is shown open, autosomal chromatin is cross-hatched. The dot indicates XIC; der(A), der(autosome). Circled letters provide reference points for text comments. *Effect of autosomal duplication may be lessened by spreading of transcriptional silencing into the autosomal segment of the der(A). **Blocking of spread of inactivation into the autosomal segment of the der(X) may avoid further functional autosomal mono somic effect.

Single-Segment Exchange, Autosomal Translocated Segment

A single segment of autosomal origin, with only the telomeric tip of Xp or Xq translocated in exchange, is shown in the middle column of Figure 5-6, segregations dg. The imbalanced conceptions from 2:2 adjacent-1 malsegregation would be partially monosomic or partially trisomic for the autosomal segment— 46,der(autosome) and 46,der(X), respectively (segregations d and e in Fig. 5-6). In the 46,X,der(X) female the partial autosomal trisomic state may have an attenuated phenotype due to spreading of inactivation from the XIC of the der(X) into the autosomal segment. The 46,Y,der(X) male conceptus, in which no X inactivation occurs, would show the undiluted effect of the partial autosomal trisomy. The partially monosomic state, 46,XX,der(autosome) or 46,XY,der(autosome), would be no different if the other chromosome participating in the translocation had been an autosome instead of an X, and the typical clinical picture associated with that autosomal deletion would be expected.

Double-Segment Exchange, Adjacent-1

In a double-segment exchange with adjacent1 segregation (right column, Fig. 5-6, segregations hi), in the unbalanced conceptus, there may be effects of a combined X functional disomy and autosomal monosomy or of X monosomy (or nullisomy) and autosomal trisomy. Such combinations would often be lethal in utero. But in the 46,X,der(X) female (segregation i), the effects may be very considerably modified by spreading of inactivation.

Consider the t(X;16) illustrated in Figure 5-8. The 46,X,der(X) daughter has both a monosomy for most of Xp, giving a Turner-like phenotype, and a structural trisomy for most of 16p. Following spread of inactivation in the der(X) into its autosomal segment in a fraction of cells, the 16p trisomy has been converted in these cells into a functional 16p disomy. In 76% of cells, however (and in the cell illustrated), the inactivation has not extended into the 16p segment. Thus she has effectively a functional mosaic 16p trisomy/16p disomy. This same combination with a Y replacing the X as the intact sex chromosome, 46,Y,der(X), with nullisomy Xp/trisomy 16p, would not be viable. The other adjacent-1 conceptions with 46,XX,der(16) and 46,XY,der(16) (light arrows in Fig. 5-8; h in Fig. 5-6) would not be similarly modifiable and would have a very large functional imbalance, and they would also be expected to abort early in the pregnancy.

Figure 5-8. Spread of inactivation into autosomal segment. (a) Mother with balanced X;16 translocation (upper) and her daughter with a 46,X,der(X) karyotype from adja-cent-1 segregation (lower). The translocation is t(X;16) (p11;p12). Replication banding shows active (darker-staining) and inactive (lighter-staining) chromosome segments. The normal X is inactivated in all cells analyzed in the mother (dashed box on cartoon karyotype; dot indicates X inactivation center). The daughter's abnormal X lacks Xp and contains distal 16p material. This chromosome is preferentially inactivated (dashed outline of box), but in 76% of cells analyzed (lymphocytes) the inactivation has not continued through the translocated 16p segment (dotted outline of box). The phenotype is the combined result of the Xp monosomy and a “partial” 16p trisomy. The child is short and has a developmental age of about 21/2 at a chronological age of 4 years. (Case of C. E. Vaux.) One other daughter had the same balanced translocation as the mother, and showed consistent inactivation of the normal X chromosome in blood lymphocytes, but suffered intellectual deficit. (b) The presumed pachytene configuration during gametogenesis in the mother (X chromatin is open, no. 16 chromatin is cross-hatched; dot indicates X inactivation center). Arrows indicate movements of chromosomes to daughter cells in adjacent-1 segregation; heavy arrow shows the combination observed in this family. This is essentially the segregation i in Figure 5-6, with an Xp breakpoint in this case.

Double-Segment Exchange, Adjacent-2

Adjacent-2 segregation typically produces trisomy for much of one chromosome along with monosomy for much of the other, and in the usual autosome–autosome translocation, this is not remotely viable (e.g., segregation (5) in Fig. 4-4). But such an enormous degree of structural imbalance can be accommodated in some X-autosome translocations in a female conceptus. First, consider the case of the intact autosome and the derivative autosome being transmitted together: 46,X,X,+der(autosome). Provided the X segment of the der(autosome) includes the XIC (segregation b in Fig. 5-7), inactivation can spread from the XIC in both directions and into the autosomal segment, counteracting the effect of the autosomal duplication, at least partially. The concomitant partial X monosomy is, of itself, a viable state. The child would be expected to display a partial Turner phenotype, upon which the effect of a variably inactivated partial autosomal trisomy would be added. This is illustrated in Leisti et al. (1975), who record a mother carrying a t(X;9)(q11;q32) and her daughter being 46,X,+X,+der(9). In the daughter, inactivation spread through much of the autosomal segment, which very substantially, although not completely, neutralized the effect of the partial trisomy 9: she had a Turner phenotype with superadded microcephaly and mental defect. The case in Williams and Dear (1987) is similar, with a retarded and dysmorphic child having the karyotype 46,-X,+X,der(10), t(X;10)(q11;q25)mat, but in this instance inactivation into the autosomal segment was apparently blocked at the centromere of the der(10). This left the child with an effective functional duplication of 10p, along with the X deletion (Fig. 5-9). For a male conception in this setting, an adjacent-2 conceptus could not survive.

Second, viability is also possible in one very rare circumstance of an intact X and the der(X) being transmitted together, with the adjacent2 karyotype 46,XX,-(autosome),+der(X), segregation c in Figure 5-7. The der(X) must contain an XIC; its autosomal segment must comprise a very substantial amount of the chromatin of that autosome; and there must be little or no spread of inactivation beyond the X segment of the translocation chromosome into the autosomal segment. In this way, the autosomal component can maintain sufficient disomic genetic activity to produce a viable phenotype. Only autosomes with genetically small short arms could enable these criteria to be met. An example from a maternal t(X;22)(p21.3;q11.21) is shown in Figure 5-10b (bottom row). The der(X) comprises most of an X and all, or almost all, of 22q. If its 22q segment were blocked from inactivation, there would be, in effect, a near-normal functional disomy 22, along with a partial XXX syndrome. In fact, this woman had a mild intellectual disability and attended a special school; the relative contributions to her phenotype of the two components of the cytogenetic abnormality are open to speculation.

Double-Segment Exchange, 3:1 Segregation with Tertiary Monosomy

The same criteria noted above may also hold in the rare situation of tertiary monosomy being viable (in essence, this is segregation k in Fig. 5-6). The t(X;22) in Figure 5-10 again provides an example. In the index case in this family with the tertiary monosomy state 45,X,-X,+der(X),22, (middle row, Fig. 5-10b), the der(X) chromosome is preferentially inactivated, but inactivation has not (at least on blood lymphocytes) spread through to the 22q component of the der(X). Thus a functional 22 disomy is maintained, or nearly so. The important structural imbalance, one might have predicted, could have been limited to the Xp21.3–pter deletion (loss of 22p being without effect), with the phenotype confined to a Turner-like picture. In the event, however, there were minor congenital anomalies, and the child was assessed as being intellectually disabled. This suggests that the pattern of inactivation elsewhere in the soma may have differed from that observed on peripheral blood, and there might be a degree of functional 22q monosomy in other tissues, including brain.

Figure 5-9. Adjacent-2 segregation. (a) Mother with balanced X;10translocation (upper) and her daughter with a 46,-X,X,+der(10) karyotype (below) on G-banding. The translocation is t(X;10)(q11;q25). Replication banding showed the normal X to be inactivated in all 30 lymphocytes analyzed in the mother (dashed box on cartoon karyotype; dot indicates X inactivation center). The daughter's der(10) was preferentially inactivated (dashed outline of box) in 50/50 cells, but the inactivation did not continue through to the 10p segment (dotted outline of box). The phenotype is the combined result of the 10p duplication and Xp monosomy. (Case of J. Williams; Williams and Dear, 1987.) (b) The presumed pachytene configuration during gametogenesis in the mother (X chromatin is open, no. 10 chromatin is cross-hatched; dot indicates X inactivation center). Arrows indicate movements of chromosomes to daughter cells in adjacent-2 segregation; heavy arrows show the combination observed in this family. This is segregation b in Figure 5-7.

3:1 Segregation, Interchange Trisomy/Monosomy

From each of the categories of single and double segment exchange, 3:1 interchange trisomy could theoretically produce Klinefelter syndrome or XXX syndrome along with the balanced translocation; and interchange monosomy could produce standard 45,X Turner syndrome. We are aware of only one such outcome, an infertile woman with 47,XX,t(X;12) from a 46,X,t(X;12)(q22;p12) mother, the imbalance being equivalent to 47,XXX (Madan et al., 1981).

4:0 Segregation

With a trisomically viable autosome, e.g., chromosome 21, a 48,XX,der(X),der(21) karyotype might be equivalent to the 48,XXX,21 state, a combined Down syndrome plus XXX syndrome. We know of no such report.

Origin of the X-Autosome Translocation

All de novo balanced X-autosomal translocations studied thus far have been of paternal origin, which may reflect the availability in male meiosis of the X chromosome for exchange with other chromosomes; the X is only paired at one end (the Xp tip), and the rest of the chromosome “dangles free,” so to speak (Powell et al., 1994). In one well-analyzed example, Gialcone and Francke (1992) did a molecular dissection on a de novo t(X;4)(p21.2;q31.22) in a girl with Duchenne muscular dystrophy and proposed a format in which two GAAT sequences 5 kb apart in Xp and one GAAT in 4q came together during meiosis in spermatogenesis, deleted the 5 kb length in Xp (which comprised a small part of the dystrophin gene), and reformed as a der(X) and a der(4). Similar mechanisms likely underlie the formation of many X-autosome translocations. Once a balanced translocation is established in a family, and if the heterozygous state is associated with phenotypic normality, male infertility dictates that transmission thereafter will be matrilineal. We comment upon the de novo X-autosome translocation identified at prenatal diagnosis on p. 420.

Figure 5-10. 3:1 tertiary monosomy and adjacent-2 segregation both occurring in the same family (see text). (a) Pedigree of family segregating at(X;22)(p21.3;q11.21). Filled symbol, imbalanced state; half-filled symbols, heterozygote/hemizygote; N, 46,XX. (b) Partial karyotypes of heterozy-gotes (upper), and of the two unbalanced states (lower). On replication banding, the normal X is inactivated in all cells analyzed in the heterozygotes, whereas the der(X) is inactivated in the two affected persons (dashed box on cartoon karyotype; dot indicates X inactivation center). In the affected child in generation III with a 45,X,-X,+der(X),-22 karyotype (middle karyotype), the der(X) was positive for the probe sc11.1, which recognizes a sequence in the DiGeorge critical region. In 50/50 cells, the der(X) chromosome showed apparently no inactivation going through to its 22 component (dotted outline of box), but the clinical picture might suggest otherwise (see text). The affected woman II:1 has the adjacent-2 karyotype 46,XX,+der(X),-22 (bottom karyotype). (Case of T. Burgess.) (c) The presumed pachytene configuration during gametogenesis in the heterozygote (X chromatin is open, no. 22 chromatin is cross-hatched). Heavy arrow indicates movement of the der(X) chromosome to one daughter cell in 3:1 segregation (essentially segregation k, Fig. 5-6). Dashed arrows show the movement of chromosomes in the adjacent-2 combination (segregation c, Fig. 5-7).

Translocations that Disrupt X-borne Loci

Certain breakpoints are on record as being associated with particular X-linked Mendelian disorders (Schlessinger et al., 1993). These are seen in females with de novo X-autosome translocations in which the actual locus has been disrupted. With preferential inactivation of the normal X, there remains no functional copy of the normal allele—an effectively “nulliallelic” state. The classic historical example of this is the female Xp21-autosome translocation heterozygote who has Duchenne/Becker muscular dystrophy, as mentioned above and an example of which is illustrated in Fig. 5-11. Quite a few X-borne loci have been mapped by study of rare or unique female patients with the particular Mendelian condition and an X-autosome translocation. A girl with lissencephaly (see also p. 285) and a de novo X;2 translocation pointed the way to the XLIS gene at Xq22.3 (Dobyns et al., 1992). The disruption of an X-linked neuronal gene, oligophrenin-1, caused isolated mental defect in a female with an X-autosome translocation 46,X,t(X;12) (q11;q15). The breakpoint was in the second intron, and thus the first two exons of the gene were on the der(12) and the remaining 23 exons on the der(X). No transcript could be produced, because of this disruption of the allele, and the other allele on the normal X having been inactivated (Billuart et al., 1998). A notable example is that of a retarded woman with autism and multiple skeletal exostosis who had a presumed de novo t(X;8)(p22.13;q22.1). The authors of this study suggest that these two phenotypic effects were due, respectively, to disruption of a “brain gene” on the X, and to a position effect on a “bone gene” on the no. 8 chromosome. In this instance, the brain gene (GRPR, expressed in the limbic system of the brain) was among those X loci not subject to inactivation, and so the net result of the translocation was a reduction from diallelic to monoallelic expression (Ishikawa-Brush et al., 1997). In those cases in which no gene disruption can be identified at the molecular level, explanations for an abnormal phenotype remain speculative, but position effect is a good possibility (Nothwang et al., 2000).

Y-Autosome Translocations

Y-autosome translocations fall into two major Yq-breakpoint categories, one of which has important clinical implications, and the other of which does not. Certain other rare forms exist (Hsu, 1994). First, some brief comments on the nature of the Y chromosome are in order.

THE Y CHROMOSOME AND Y-AUTOSOME TRANSLOCATIONS

The particular raison d'être of the Y chromosome is to bring about male development. The testis-determining gene, SRY, lies in the euchromatic region on the short arm, just 5 kb proximal to the pseudoautosomal boundary. As noted above, the primary pseudoautosomal regions (PAR1) of the Y and X short arm contain homologous loci, and certain other loci elsewhere in the Y have homologs on the X (Disteche, 1995). The secondary pseudoautosomal region (PAR2) is located at distal Yq and distal Xq; its loss from Yq seems to be without phenotypic consequence (Kühl et al., 2001). From the point of view of reproductive health, three azoospermia factor regions on Yq are of importance, named AZFa, b, and c, and these are discussed in more detail in Chapter 21 (p. 351). Otherwise, about half the Y—theamount is variable—comprises the genetically inert heterochromatic region of the long arm (Yq12), which contains highly repetitive DNA sequences.

Figure 5-11. A de novo X-autosome translocation 46,X,t(X;4)(p21;p16) in which the dystrophin locus at the Xp21 breakpoint is presumed to be disrupted, in a 7-year-old girl. In consequence very little dystrophin is produced, and the girl has a Becker-like muscular dystrophy. The approximate position of the dystrophin locus is indicated (arrowhead) on the intact X. The intact X is preferentially inactivated, as shown here with replication banding and indicated in dashed outline on the cartoon karyotype. Early replicating (active) chromatin and the late replicating (inactivated) chromatin stain dark and light, respectively. (Case of J. A. Sullivan.)

During normal meiosis in the male, the X and Y chromosomes recombine, synapsing at the pseudoautosomal regions at the tips of Xp to Yp. The two sex chromosomes joined together in this way comprise the sex vesicle. A properly formed sex vesicle is necessary for normal meiosis and spermatogenesis, and anything that interferes with its normal formation, such as the presence of translocation chromosomes, will compromise the process of sperm development. We have seen above that an X-autosome translocation in the male practically always causes spermatogenic arrest. Fairly infrequently, some autosomal translocations, and more especially those involving an acrocentric chromosome, can cause interference with the sex vesicle, with consequential infertility (p. 84). Similar considerations apply to the balanced Y-autosome translocation (other than the Y-acrocentric translocation) (Maraschio et al., 1994). The autosomal components of the quadrivalent, dragged into the sex vesicle, as it were, can have a sabotaging effect, disrupting the process of meiosis. Delobel et al. (1998) illustrate this circumstance in their study of a phenotypically normal infertile man with a translocation t(Y;6)(q12;p11.1). They analyzed meiotic preparations from a testicular biopsy, noting the configuration of the quadrivalent, this comprising the X, der(Y), der(6), and normal chromosome 6. The autosomal elements of the quadrivalent were seen to have been drawn into the sex vesicle and to be hyper-condensed. The result was spermatogenic arrest at the pachytene stage, with subsequent degeneration of spermatocytes. Yet, and in contradistinction to the X-autosome case, the Y-autosome carrier may occasionally retain fertility. Otherwise, fertility may be “rescueable” by means of assisted reproduction, as discussed further below.

Reciprocal Y Long-Arm and Autosome (Other Than Acrocentric Short Arm) Translocation

Reciprocal exchange between the Y long arm and an autosome produces a balanced Y-autosome translocation. In the form being considered here, the Y breakpoint is usually given as q11.2 or q12, and the autosomal breakpoint is anywhere on the autosomal karyotype, other than an acrocentric short arm. There are associated phenotypic abnormalities in a few individuals, and this may be due to a disruptive effect at the breakpoints or a deletion of autosomal material distal to the breakpoint (Erickson et al., 1995). In most, the rearrangement may be truly balanced, with the physical and intellectual phenotype being normal, and infertility is the usual presenting factor. Given this latter fact, it follows logically enough that the translocation would typically arise as a de novo event, and this is indeed the observation. The infertility may be a result of disruption of the sex vesicle, as discussed above.

The man with the 46,X,t(Y;18)(q11.2q21) translocation shown in Figure 5-12 had been karyotyped in the course of investigation for infertility with severe oligospermia, he being an otherwise normal person. With the availability of intractyoplasmic sperm injection (ICSI), biological paternity becomes possible for such men. In this Y;18 case, of the 16 possible embryos, more than half, including one of the 4:0 segregants, would in theory be viable; the reader may wish to work out which ones these might be. Only one sperm, the 23,X, is capable of producing a phenotypically and karyotypically normal child; the other gamete from alternate segregation, 23,t(Y;18), would produce a son who might well have similar infertility. Applying preimplantation chromosome analysis, the chromosomally unbalanced embryos could be discarded. With a small number of eggs retrieved on each stimulation cycle, a normal combination might well not happen, given that there are 14 unbalanced possibilities, if each outcome were equally likely. But in fact the observations of Sklower Brooks et al. (1998) and Giltay et al. (1999) (below) provide some encouragement that the odds for the Y-autosome carrier (in other words, the meiotic predisposition) may be tipped in favor of the normal and balanced forms. As it turned out in this Y;18 case, one embryo was indeed 46,XX, and this was successfully implanted.

Figure 5-12. Y-autosome translocation and infertility. This t(Y;18)(q11.2q21) was identified in a man presenting with oligospermia during investigation for infertility. The fact that some sperm are still being produced allowed the option of IVF with ICSI. A considerable number of these unbalanced gametes could, in theory, be viable. Only a 46,XX daughter (which he did in fact have) could be both karyotypically and phenotypically normal. (Case of L. Harris and L. Wilton.)

A few familial cases have been recorded. Teyssier et al. (1993) document fertility in father and infertility in son: a man with severe oligoasthenospermia had a t(Y;1)(q11;q11), and his father proved to carry the same translocation. Intact fertility is well illustrated in the family described by Sklower Brooks et al. (1998), depicted in Figure 5-13. One son in a sibship of five males and two females, a university graduate, presented for genetic counseling when his wife had a third miscarriage (they also had a normal daughter). The deceased father must have carried a t(Y;8)(q12; p21.3), with three sons showing the balanced state and two sons having inherited an unbalanced complement, 46,X,der(Y). The unbalanced state conferred a partial trisomy for 8p22–pter, which was associated with a mild learning difficulty. This family also illustrates the point mentioned above that normal/balanced segregations may, in some instances, be favored, despite the seeming odds against them.

Figure 5-13. A Y-autosome translocation, not involving an acrocentric short arm. In this particular example, and somewhat unusually, fertility is apparently normal. The autosomal translocated segment is of small size, structurally and functionally, and the aneuploid state with a dup(8p) is not only viable but associated with only a mildly abnormal intellectual phenotype and an essentially normal physical appearance. (a) Family tree. Filled symbol, unbalanced karyotype; half-filled symbol, balanced carrier. The deceased grandfather is presumed to have been a translocation heterozygote. (b) Partial karyotype of a translocation heterozygote (upper), showing the Y;8 translocation, and one of the individuals with the unbalanced complement (lower). (c) The presumed pachytene configuration during gametogenesis in the heterozygote (no. 8 chromatin is open, Y and X chromatin are cross-hatched). Arrows indicate movements of chromosomes to daughter cells in “adjacent-1” segregation; heavy arrows show the combination observed in this family. (Case of S. Sklower Brooks; from SklowerBrooks, S., Genovese, M., Gu, H., Duncan, C. J., Shanske, A. and Jenkins, E. C., Normal adaptive function with learning disability in duplication 8p including band p22, Am. J. Med. Genet., © 1998 Am. J. Med. Genet., and with the permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc.)

This question is more directly answered in the study by Giltay et al. (1999), who undertook a sperm analysis in a man with a t(Y;16) (q11.21;q24). Sperm were present but few in number, with many abnormal forms (oligoasthenoteratozoospermia). Although alternate segregation accounts for only 2 out of the 16 segregation possibilities, in this case half of all morphologically normal spermatozoa were normal or balanced, with about 40% showing adjacent segregation and about 10% with 3:1. But the fractions were less favorable if abnormal sperm were included. With reference to assisted conception and using ICSI, Giltay et al. speak of in vitro and in vivo selection (the former artificial, the latter natural) combining to effect a considerable reduction in the risk for an unbalanced offspring. In fact, the man in their study had had three children by ICSI, two normal daughters and a carrier son.

Y Long-Arm and Acrocentric Short-Arm Translocations

Yqh Material on Acrocentric Short Arm, 46-Chromosome Count

In some 70% of Y-autosome translocations the other chromosome is one of the acrocentrics. There is no loss or gain of euchromatin; the result is that one acrocentric carries some phenotypically irrelevant Y heterochromatin, looking rather like (and sometimes mistaken for) a very long short arm (Neumann et al., 1992). A normal acrocentric (e.g., chromosome 15) is replaced by a der(15). The breakpoints are sited in the acrocentric short arm (p11–p13) and in the heterochromatin of the Y long arm (Yq12) (Fig. 5-14). In about half, the acrocentric is a no. 15 and next most commonly, a no. 22; close homology between heterochromatic blocks in 15p and Yq may be the reason that Y;15 rearrangement is favored (Cohen et al., 1981; Metzler-Guillemain et al., 1999). Categorization can be according to the D (13–15) or G (21–22) group acrocentric chromosome involved; thus Y;Dp and Y;Gp translocations. Males and females can equally be carriers. For the record, phenotypic normality for the der(22)t(Yq;22p) has been reported in the homozygous state (Leschot et al., 1986). Hsu (1994) reviewed 25 papers on the topic of t(Yq;Dp) and t(Yq;Gp), concluding that, in the familial case, neither phenotypic abnormality nor infertility is likely to be associated. But she does point out the advisability of checking for Yp material on the translocation chromosome if there has not otherwise been the observation of normal females in the family with the translocation (who would by that fact demonstrate an absence of the male-determining region). Rare disquieting reports need to be viewed cautiously, such as that of Rajcan-Separovic et al. (2001), who raise the possibility of a secondary chromosomal abnormality in documenting the remarkable instance of a woman with the karyotype 46,XX,der(15) t(Y;15)(q12;p13)mat and of older childbearing age who had had two trisomic 15 pregnancies.

Rare Forms

Yqh Material on Nonacrocentric Chromosome, 46-Chromosome Count

If Yq heterochromatin is translocated to the tip of an autosome, other than to an acrocentric short arm, there may or may not be reproductive implications. A der(1)t(Y;1)(q12;p36) in a French family could be traced back to a common couple married in 1773, with self-evident fertility, male and female, for more than two centuries (Morel et al., 2002a). Some other similar translocations have been associated with spermatogenic failure.

Figure 5-14. An example of the Y-autosome translocation involving an acrocentric autosome, the der(15) depicted here, with the breakpoint in the acrocentric short arm. Normal chromosome 15 and normal Y shown alongside for comparison (no. 15 chromatin is cross-hatched, Y heterochromatin is filled, Y euchromatin is open). The translocation chromosome can be carried equally by males and females. The karyotype appears unbalanced, but the phenotype is normal.

Y Material plus Acrocentric Short Arm Tip, 46-Chromosome Count

This is essentially the countertype of the common Y;Dp and Y;Gp described above, with the other reciprocal product, the der(Y), replacing a normal Y. Hoshi et al. (1998) identified a perfectly normal man, the father of three, who had a 46,X,Y,t(Y;15)(q12;p13) karyotype. The der(Y) contained the necessary male-deter-mining and fertility regions. He was only investigated because his sister had a gonadal tumor of testicular origin and proved to have the mosaic karyotype 46,X,Y,t(Y;15)/45,X.

Y Material to Autosome Translocation, 45-Chromosome Count, including “45,X Male”

The testis-determining region of the Y, containing the SRY gene, can be translocated onto an autosome, usually an acrocentric (Farah et al., 1994). The individual, phenotypically male, has 45 chromosomes, including the Yautosome fusion product. The translocated Y segment may be beyond the level of cytogenetic resolution, and the karyotype can appear as 45,X (“45,X male”) until FISH and molecular studies cast further light. The translocation may be of no phenotypic or reproductive effect, and represent a “new” sex-determining mechanism, as Callen et al. (1987) document in a family identified quite by chance in the course of a research study in which a man and two sons had the karyotype 45,X,dic(Y;22) (q11.23;p11.2). Similarly, White et al. (1998a) discovered a 45,X,dic(Y;15)(q11.23;p11.1) at prenatal diagnosis, and showed the same chromosome in the father.

More often, the reproductive and sometimes the physical phenotype is affected. Azoospermia is a frequent finding, as documented in a review of 15 cases of “45,X male” (Copelli et al., 2000). In their own azoospermic patient, these authors recorded a somewhat Turner-like somatotype, but with normal male external genitalia and secondary sexual characteristics and normal endocrinological indices. He was SRY positive. The azoospermia in these men may be due to perturbation of the process of meiosis rather than loss or compromise of Yq spermatogenesis genes (Buonadonna et al., 2002). Gimelli et al. (1996) studied an abnormal boy whose initially “45,X” karyotype they were able with more accuracy to show as being 45,X,der(18)t(Y;18)(q11;p11.2). His phenotype could thus be seen essentially to reflect a del(18p) picture. The Y component may be translocated insertionally, as Yenamandra et al. (1997) demonstrated in a phenotypically abnormal “45,X” boy, in one of whose no. 4 chromosomes, at 4p15.3, the SRY-bearing segment was accommodated. Rather more obvious cytogenetically is the de novo dicentric Y;13 translocation, 45,X,dic(Y;13)(p11.3;p11.2), described in Shanske et al. (1999a): the translocation comprised almost a complete 13Y composite. Their patient was a very short and otherwise normal 10-year-old boy, in whom the SHOX growth control gene, normally located in the primary pseudoautosomal region, was absent.

Dicentric p-to-p Y;Acrocentric Translocation, 45-Chromosome Count

A p arm–to–p arm fusion produces a translocation chromosome that contains completely, or almost completely, the genetic material of the Y chromosome and an acrocentric. In one case, even Yp arm telomeric sequences were present at the fusion point in a child investigated for minor physical anomalies and learning difficulty and who had normal genital anatomy (Boutouil et al., 1996). Alves et al. (2002a) found a Yp;13p fusion chromosome in an otherwise normal man presenting with oligospermia. About half of premeiotic spermatocytes were unbalanced, but most sperm that actually made it to maturity (85%) had a balanced complement.

Yp;15q Translocation and Prader-Willi Syndrome

A very few cases of Prader-Willi syndrome have been due to a fusion between a Y and a no. 15, with breakpoints in Yp and at 15q13, the karyotype being 45,X,t(Y;15) (see p. 324). These patients are male.

X-Y Translocations The Classical X-Y Translocation

Of the major types of X-Y translocation, the classical form is the most frequently seen (Fig. 5-15a, b). The X and Y breakpoints are constant, at the cytogenetic level, involving Xp22.3 in the distal X short arm and Yq11 in the proximal Y long arm (Bernstein, 1985; Ballabio and Andria, 1992). It is certainly readily recognized cytogenetically and has the karyotypic notation 46,X or Y,der(X),t(X;Y)(p22.3;q11). The important genotypic defect is deletion of the distal Xp segment, with the loss including the PAR1. At the molecular level, there is variation in the amount of Xp deleted, and the phenotype depends at least in part on which of the following distal Xp genes may be lost: ARSE (arylsulfatase E, for chondrodysplasia punctata), SHOX (short stature homeobox, for Leri-Weill dyschondrosteosis), STS (steroid sulfatase, for ichthyosis), KAL (Kallmann syndrome), MRX (mental retardation), and OA1 (ocular albinism). The person who is 46,X,der(X)t(X;Y), typically a female, has a partial monosomy for this Xp segment; and the 46,Y,der(X)t(X;Y) individual, always a male, is partially nullisomic.

Figure 5-15. Four ways in which XY translocations are seen. (a) The classical t(X;Y)(p22.3;q11) together with a normal X (in a female). (b) The classical t(X;Y) together with a normal Y (in a male). (c) The cryptic t(Xp;Yp), with the Yp segment containing the SRY gene, in a “46,XX male.” (d) The cryptic t(Xp;Yp) as the sole gonosome, in a “45,X male.” For Y chromatin, white indicates Y euchromatin, black indicates that part of distal Yp euchromatin encompassing the pseudoautosomal region and the SRY locus, cross-hatching indicates Yq heterochromatin. Note that gonadal sex accords with the absence (a) or presence (b–d) of the SRY gene.

The female t(X;Y) heterozygote is characteristically fertile and of normal intelligence. If SHOX is deleted, the monosomic (haploinsufficient) state for this gene determines a particular form of short stature and wrist deformity (Leri-Weill dyschondrosteosis) (Calabrese et al., 1999). The male t(X;Y) hemizygote is typically the son of a t(X;Y) mother (Hsu, 1994). Some may be cognitively normal, in those in whom the breakpoint is more distal. Should a more proximal molecular breakpoint expose a Mendelian “brain gene,” such as MRX mentioned above, mental impairment results, and apparently this is more usually the case. If the male has Leri-Weill dyschondrosteosis, it is no more marked, and indeed frequently less so, than in the female, reflecting the fact that the SHOXlocus is in the PAR1, and each sex still retains one copy of the gene, on their normal X or Y, respectively. Infertility is almost invariable, due to  spermatogenic arrest (Gabriel-Robez et al., 1990). Sperm production has, however, been documented in one case, albeit at a very low level (125,000/milliliter), in a man with the typical 46,Y,der(X),t(X;Y)(p22.3;q11) karyotype (Morel et al., 2001). He was of normal intelligence, height 165 cm (5 feet 5 inches), with normal external genitalia and normal endocrine indices, and he had presented with infertility. There were equal numbers of 23,der(X) and normal 23,Y sperm, but about 20% of sperm were otherwise abnormal, the most common defect being 24,Y,der(X).

The majority of cases are familial. Presumably, the X-Y chromosome arose following a reciprocal exchange between the X and Y during spermatogenesis in the individual fathering the originating (female) translocation carrier in the family. This event is facilitated by the apposition of X and Y segments having a high degree of homology—for example, a crossover between the Kallmann locus on the X chromosome and a Kallmann-like nonfunctional pseudogene on the Y chromosome long arm (Guioli et al., 1992). In the female, the pattern of X-inactivation tends toward inactivation of the der(X)t(X;Y), but this is variable and unpredictable (Gabriel-Robez et al., 1990).

Some X-Y translocations cytogenetically apparently identical to the classical type are associated with the rare syndrome of microophthalmia with linear skin defects (MLS) in the 46,X,der(X)t(X;Y) female (Kotzot et al., 2002). The same syndrome is reported in the cryptic Xp-Yp translocation (Anguiano et al., 2003) described in the following section. The phenotype may result from the effect of disruption, loss, or aberrant inactivation of the MLS gene, but this remains to be proven (Ogata et al., 1998).

Other Variant Forms

Other forms of t(X;Y) typically arise de novo, and are associated with infertility, and in some with intellectual deficit. The dicentric X-Y translocation comprises an almost complete Y attached at a distal Yp breakpoint to an X chromosome, at either an Xp or an Xq breakpoint. For example, Baralle et al. (2000) describe a girl with a Yp to Xp rearrangement, the karyotype 46,X,dic(X;Y)(p22.3;p11.2), who presented with Leri-Weill dyschondrosteosis. Her pubertal development was regarded as being normal, although she had yet (at age 14 years) to undergo menarche. Her femaleness was due to absence of the SRY gene, the breakpoint on Yp being proximal to its locus. Other rare types (listed and illustrated in Hsu, 1994) include der(X) chromosomes from translocations of varying lengths of Yq to a breakpoint at various levels on Xp or Xq, and der(Y) chromosomes from translocations of varying small lengths of Xp to a breakpoint at various levels on Yq.

The Cryptic Xp-Yp Translocation (“XX Male” and “45,X Male”)

This form of the X-Y translocation is usually not visible (or barely visible) to the cytogeneticist without the use of FISH with Yp sequences as probe. Again, the X breakpoint is within Xp22.3; but the Y breakpoint is in the short arm, proximal to the testis-determining gene (SRY). The genotypic consequences are loss of the distal region of the X chromosome and the transfer of the SRY gene onto an almost intact X chromosome. Thus, the person is male. The karyotype would initially appear to the cytogeneticist either as 46,XX or as 45,X. In truth, it is 46,X,der(X)t(X;Y)(p22.3;p11) or 45,der(X)t(X;Y)(p22.3;p11). This translocation accounts for most supposed XX males, and some 45,X males. If there is loss of one copy of the SHOX gene, Leri-Weill dyschondrosteosis (see above) is the expected consequence (Stuppia et al., 1999). The translocation arises from an abnormal X-Y recombination during paternal meiosis, and is always sporadic (Weil et al., 1994). The affected males are invariably infertile.

46,X,del(Yq) with Cryptic Xq-Yq Interchange

A third category is the X-Y translocation arising de novo from an exchange during paternal spermatogenesis between Yq and distal Xq, producing an apparent del(Yq) chromosome that actually contains a very small segment of distal Xq (Lahn et al., 1994). The karyotype could be written 46,X,der(Y)t(X;Y) (q28;q11.21). The functional distal Xq disomy produces a severe phenotype (Gastier et al., 2002). In the female, the reciprocal 46,X,der(X) karyotype would hypothetically imply a mild phenotype, essentially reflecting a very small distal del(Xq), and assuming that the abnormal X would be preferentially inactivated.

X-X Translocations

The general karyotype is written 46,X,der(X), t(Xp;Xq), and the resultant imbalance is a dup/del of Xp/Xq, or vice versa. The translocation could have arisen following unequal recombination between the two X chromosomes in the oocyte. Or, the rearrangement could have occurred within the one X chromosome, folding in on itself, in which case the origin is more likely paternal (Giglio et al., 2000). Xp11.23 and Xq21.3 are favored as breakpoints, and the translocated del/dup segments may therefore be large. Other variations of X-rearrangement have been labeled as an X-X translocation, for example, the isodicentric qter–p22.1::p22.1–qter reported in Tarkan et al. (1995), but the process of formation in some of these may not truly have been an actual translocating of a segment of chromatin from one site to another. Indeed, the validity of the term could also be argued for the intrachromosomal del/dup rearrangement noted above, in which a U-type exchange may have been the basis.

Pubertal and/or menstrual abnormality is the usual presentation, and infertility is the rule (Letterie, 1995). Maternal transmission is, however, recorded in Reinehr et al. (2001), concerning a mother and daughter with short stature having a t(X;X)(p22.1;q26) derivative chromosome. These breakpoints are distal, and thus the del/dup segments in this case are small. The monosomic Xp segment included the SHOX gene, and this presumably was the cause of the short stature. The mother had a normal menstrual history, and had two other healthy children of normal heights.

Y-Y Translocation

For the sake of completeness, the existence of the Y-Y translocation is noted. Hsu (1994) lists three cases in her review of structural Y abnormalities.

GENETIC COUNSELING

The X-Autosome Translocation

Fertility is affected in the X-autosome heterozygote and hemizygote. Approximately half of the female carriers, and practically all males, are likely to be infertile. If fertile, the female heterozygote has a substantial risk of having abnormal offspring due to an imbalanced chromosomal constitution. At one end of the scale, the abnormality might be mild (e.g., partial Klinefelter syndrome) or barely discernible (e.g., partial X trisomy). At the other end, it could be severe (e.g., partial X disomy or autosomal aneuploidy, not modified by inactivation). The counselor should determine the theoretical gametic combinations from the particular category of translocation, with reference to the examples described in the Biology section. Adjacent-1 and 3:1 tertiary trisomy are the major malsegregation modes to be considered. Figures 5-6 and 5-7 provide a guide; but each translocation needs to be assessed on its own merits. General comments follow.

1. A single-segment translocation with an X segment of large size would imply risks for partial Turner, partial Klinefelter, and partial XXX syndromes (Fig. 5-2; Fig. 5-6, segregations b and c). A single-segment translocation with an X segment of small size would imply a risk not only for one of these three partial gonosomal aneuploidies but also for functional disomy for a small distal Xp or Xq segment, which would have a severe outcome (Fig. 5-6, segregation a).

2. A single-segment translocation with an autosomal translocated segment of “viable size” (Fig. 5-6, segregations de) implies a risk for partial autosomal monosomy or trisomy from adjacent-1 segregations. In the female conceptus, the trisomy may be modified by spreading of inactivation, but this is unpredictable.

3. Any 2:2 unbalanced segregant from a double-segment translocation (Fig. 5-6, segregations hi) has a combined dupli-cation/deficiency, and spontaneous abortion is probable. But spreading of inactivation in a female conception may attenuate a partial autosomal trisomy and allow for survival, albeit with phenotypic defect.

4. Adjacent-2 possibilities need individual assessment (Fig. 5-7).

The Level of Risk

The risk for most female heterozygotes who are fertile will be substantial. An otherwise nonviable unbalanced conception may survive because inactivation tempers the imbalance; and some conceptions with the structurally balanced complement may be functionally unbalanced because of aberrant inactivation patterns. The risks of having a liveborn child with a structural and/or functional aneuploidy may be in the range 20%–40%. As we discuss above, the components making up the total risk may comprise a very mild abnormality through to severe mental and physical defect. Only with the 46,XX and 46,XY karyotype can one be confident of normality, other things being equal.

The Balanced Inherited X-Autosome Detected Prenatally

A balanced X-autosome karyotype identified at prenatal diagnosis, the fetus being female, is not necessarily associated with the same phenotype as that of the mother. A normal mother would have, presumably, a “perfect” 100:0 concordance of activation status, with respect to the translocation X and the intact X (see above). But the protective mechanism could fail in a daughter of hers, and some tissues could express a functional partial X disomy, with consequential phenotypic defect, as the case shown in Figure 5-8 may illustrate. A possible relevance of X inactivation status at prenatal diagnosis has insufficient precision to be of much use, although experience is being accumulated (Feldman et al., 1999).

Too little information exists concerning the phenotype of the male hemizygote born to a female X-autosome heterozygote for any firm advice to be offered. Normality has been recorded in this setting, but so has major genital defect (Buckton et al., 1981; Kleczkowska et al., 1985), which in one case was the consequence of compromised function of the androgen receptor gene (Callen and Sutherland, 1986). Fetal ultrasonography may be useful to check for normal male genital development. This approach was offered to the mother whose karyotype appears in Figure 5-10 and who had a 46,Y,t(X;22) result at amniocentesis in her second pregnancy. A normal baby boy was subsequently born, whose infant development has been completely normal. Otherwise normal male carriers would almost certainly be infertile.

Y-Autosome Translocations

The Apparently Balanced Y-Autosome Translocation

It is notable that the same balanced Y-autosome translocation can behave differently in different male members of a family in terms of fertility; this is presumably a further example of the importance of the background genetic contribution to the control of the mechanics of spermatogenesis (Teyssier et al., 1993; Rumpler, 2001). For those who are fertile, risk data are too few to form a secure base for genetic counseling. From first principles, unbalanced forms are probable, several of which will often be viable (according to the autosome in question and the site of the autosomal breakpoint), and the option of prenatal diagnosis is appropriately offered.

As discussed in the Biology section, in spite of there being several more imbalanced than balanced possibilities, there are tentative grounds for supposing that alternate segregations (normal and balanced forms) may be favored. The t(Y;8) family of Sklower Brooks et al. (1998) noted above and shown in Figure 5-13demonstrated three of the four predicted alternate and “adjacent-1” karyotypic outcomes: 46,XX, the 46,X,t(Y;8) balanced carrier, and 46,X,der(Y), the former two outnumbering the latter. The 46,X,der(Y) karyotype produced sons with an 8p duplication; the other unbalanced karyotype, 46,XX,der(8), would have produced a daughter with an 8p deletion. Manifestly, the carrier male, while he could have a normal daughter, could never conceive a 46,XY child. Sperm karyotyping, if available, may be a helpful investigation. In the man with the rare 13p;Yp fusion mentioned above (Alves et al., 2002a), having demonstrated that most sperm had a balanced complement, reassurance could be offered, in this particular case, that if pregnancy were achievable there would be a good chance of producing a normal child.

For the infertile man, assisted reproduction may offer the possibility of paternity. A sperm count way below the level needed for natural conception may yet allow retrieval of sperm for ICSI. Testicular aspiration may provide sperm even when they are completely absent in the ejaculate. With the need for in vitro fertilization (IVF), preimplantation diagnosis (PGD) becomes attractive because of the probable substantial genetic risk, in most cases, for unbalanced forms, and because the embryo is nicely accessible. Taking the example of the oligospermic man with a 46,X,t(Y;18) (q12;q11.2) karyotype, shown in Figure 5-12, he could, in theory, and through IVF, have a 46,XX daughter and a 46,X,t(Y;18) son like himself. The substantial fraction of unbalanced forms that could be viable in this case, out of the 16 total possible conceptions, becomes a relevant matter at PGD. These issues of IVF and PGD are discussed in more detail in Chapter 24.

The Yqh-Acrocentric Translocation

Probably, these translocations can be regarded as being no more than interesting variant chromosomes and of no clinical significance. In the case of the t(Yq;15p), a theoretical risk for trisomy 15 with correction to uniparental disomy (White et al., 1998a; Rajcan-Separovic et al., 2001) is not to be overstated.

The “45,X” Yp-Acrocentric Translocation

These chromosomes are probably stable, and do not (if fertility is achievable) imply a risk for phenotypically abnormal offspring (Callen et al., 1987).

The Classical X-Y Translocation

The female with an X-Y translocation is usually fertile and of normal intelligence. She has a 50% risk of having a child, whether a son or daughter, who would have the translocation. An X-Y translocation son may be abnormal according to the extent of distal Xp nullisomy and the loci involved (Seidel et al., 2001). If the mother is short, an X-Y translocation daughter would also be short. As with Turner syndrome, growth hormone treatment may be appropriate for such a child. Like her mother, she would probably be fertile. A child receiving the mother's normal X would, of course, be normal, 46,XX or 46,XY; and thus prenatal diagnosis is appropriately offered.

The male X-Y translocation carrier is almost invariably infertile. A sperm chromosome study has been undertaken in only one der(X)t(X;Y) man, referred to in the Biology section above (Morel et al., 2001). He had severe oligozoospermia, and notably sex chromosome aneuploidy was recorded in 20% of sperm. Otherwise, 40% of sperm were normal 23,Y, and 40% had the t(X;Y). Conception in such a case could only ever be achieved via IVF. If preimplantation diagnosis were to be attempted, the choice of a 46,XY embryo (the only normal gonosomal possibility) would avoid the genetic risk for the next generation.

X-X Translocations

Infertility is the expectation, and a theoretical question of genetic risk will usually be academic. In a small imbalance, fertility may be retained, as in the example of Reinehr et al. (2001) discussed above. The genetic risk would, in essence, be the same as for the 46,X,abn(X) heterozygote (p. 208). A daughter receiving the X-X translocation would be expected to have a phenotype similar to that of her mother. A male pregnancy would be very likely to miscarry, due to an X nullisomy/dis-omy. If the del/dup segments were very small, viability might be possible, but with probable major phenotypic defect. Children receiving the mother's normal X chromosome would, of course, be normal, other things being equal.

Notes

1. The chromosome from the woman's mother is denoted Xm, and the one from her father, Xp.

2. The correlation between molecular and late-replica-tion analysis may not necessarily be perfect (Sharp et al., 2001).

3. Presumably, there is also a functional trisomy for the pseudoautosomal region within this segment.