The American insect cytogeneticist W. R. B. Robertson first described translocations of chromosomes resulting from the fusion of two acrocentrics in his study of insect speciation in 1916, and this type of translocation is named Robertsonian (abbreviation, rob) in his honor. There are five human acrocentric autosomes— nos. 13, 14, 15, 21, and 22 (the 13, 14, and 15 are the D group chromosomes, and the 21 and 22 comprise the G group), and all are capable of participating in this type of translocation. The composite chromosome produced includes the complete long arm chromatin of the two fusing chromosomes, although it lacks at least some of the short arm chromatin. Robertsonian translocations are among the most common balanced structural rearrangements seen in the general population with a frequency in newborn surveys of about 1 in 1000 (Blouin et al., 1994). Historically, the most important Robertsonian translocations are the D;21 and G;21, which are the basis of most familial translocation Down syndrome. In recent years, the role of uniparental disomy has come to be recognized; in the Robertsonian context, this is of relevance with the two imprintable acrocentric chromosomes, nos. 14 and 15.
In this chapter we consider the case of the phenotypically normal person who carries, in balanced form, a Robertsonian translocation. We generally use a short cytogenetic description for the carrier state, thus, 45,XX,rob (14q21q) or simply rob(14q21q). The correct ICSN designation for a short arm to short arm fusion Robertsonian translocation is, for example, 45,XX,der(14;21)(p10;p10) or 45,XX, rob(14;21)(p10;p10).
BIOLOGY
The great majority of balanced Robertsonian translocations involve two different chromosomes (a heterologous or nonhomologous translocation); those involving the fusion of homologs (homologous translocation) are very rare. Heterologous translocations can be transmitted through many generations of phenotypically normal heterozygotes, whereas the homologous translocation is almost always seen as a de novo event in the consultand. As Table 6-1 shows, the rob(13q14q) and the rob(14q21q) are predominant. If we exclude the rob(21q21q), most of which are actually isochromosomes for 21q, the rob(13q14q) accounts for around 75% of all Robertsonian translocations in unbiased studies, and indeed is the most common single chromosome rearrangement in the human race. Since 1 in 1000 persons is a rob heterozygote, the prevalence of the rob(13q14q) carrier is about 1 in 1300. Karyotypes of the 13q14q and 14q21q carrier states and of the unbalanced 14q21q state leading to translocation Down syndrome are shown in Figures 6-1, 6-2 and 6-3. Balanced carriers for any of the five homologous translocations seem to be of about equal rarity.
|
Table 6.1. Frequency of Robertsonian Translocationsa |
|||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||
Formation of the Translocation
There are three possible mechanisms of formation of the heterologous translocation: fusion at the centromere (centric fusion), union following breakage in one short arm and one long arm (essentially, a whole-arm reciprocal translocation), and union following breakages in both short arms (Guichaoua et al., 1986) (Fig. 6-4). The first two, which may occur very rarely (if ever, in the case of centric fusion), produce a translocation chromosome with one centromere (monocentric), and the latter results in a chromosome with two centromeres (dicentric). The common rob(13q14q) and rob(14q21q) translocations are practically always dicentric, and are formed predominantly during female meiosis, with consistent breakpoints at the molecular level (Bandyopadhyay et al., 2002). In some dicentrics, one centromere is “suppressed,” and the chromosome appears monocentric. This heterogeneity of formation is not of any clinical significance that can presently be discerned. In the reciprocal type, the other product may rarely survive as a stable small bisatellited extra structurally abnormal chromosome (ESAC) (Schmutz and Pinno, 1986). The propensity to recombine may be the consequence of recombination between similar sequences shared by acrocentric chromosomes. The predominance of the rob(13q14q) and the rob(14q21q) may be due to specific homologous but inverted segments in these pairs of chromosomes that encourage crossover, while the variable breakpoints in the uncommon translocations point to a more random process (Bandyopadhyay et al., 2001a, 2001b, 2002). Rare cases are due to postzygotic joining together, a point that can be proven when the two component chromosomes can be shown to have come one from one parent and one from the other (Bandyopadhyay et al., 2003). Just as a Robertsonian translocation can form de novo from the fusion of chromosomes, it can also (very rarely) revert to two separate chromosomes by a “back-mutational” fission (Pflueger et al., 1991).
|
|
|
Figure 6-1. The balanced rob(13q14q) in a phenotypically normal male. |
|
|
|
Figure 6-2. The balanced rob(14q21q) in a phenotypically normal male. |
|
|
|
Figure 6-3. The unbalanced rob(14q21q) in a girl with translocation Down syndrome. |
|
|
|
Figure 6-4. Mechanisms of formation of Robertsonian translocations. (a) Centric fusion, giving a monocentric chromosome; (b) breakage in one short arm and one long arm, giving a monocentric; and (c) breakage in both short arms, giving a dicentric or, after suppression of one centromere, a monocentric. |
The homologous Robertsonian chromosome may arise from fusion in the zygote of the paternal and maternal homologs, in which case it is a true translocation. Alternatively, it may be an isochromosome, with the stage having been set in meiosis: a nullisomic egg due to a maternal nondisjunction leads to a monosomic conceptus, which is then “rescued” by reduplication of the paternal homolog as an isochromosome and thus with uniparental disomy for the chromosome in question (discussed below). The site of formation may be at the first mitosis, a conclusion we drew from studying a woman with 45,XX,rob(13q13q), who showed no mosaicism on biopsy samples from a number of different tissues taken during surgery for tubal ligation (Gardner et al., 1974). Berend et al. (1999)showed a de novo 45,i(13q), upd(13)pat in a normal infant to have complete isozygosity for chromosome 13 markers, indicative of the scenario of postzygotic monosomy rescue. In another instance, they were able to show a paternal meiotic origin of the i(13q) in a normal adolescent with 45,i(13q), upd(13)pat. This individual would have had trisomy 13, had it not been for gametic complementation: the mother contributed a nullisomic 13 ovum (she was actually a 13q14q heterozygote). These two cases came to attention only through fortuitous discovery at prenatal diagnosis.
Nucleolar Organizing Regions and the Robertsonian Translocation
The nuclear organizing regions (NORs) are located in the “stalks” of the short arms of the acrocentric chromosomes and comprise multiple copies of genes coding for ribosomal RNA (see p. 235). When a Robertsonian translocation forms, the NORs of two of the fusing chromosomes are lost, at least with the rob(13q14q) and rob (14q21q). Thus, an individual with a Robertsonian translocation has 8 NORs instead of the usual 10. Not all NORs are active: as judged by silver (Ag-NOR) staining, most individuals have four to seven per cell that are functioning (Varley, 1977). Presumably, there is a minimum requirement for normal cellular function.
The Heterologous Robertsonian Translocation
MEIOSIS
This Robertsonian translocation chromosome comprises the long arm elements of two different acrocentric chromosomes. At meiosis in the heterozygote, the translocation chromosome and the two normal acrocentric homologs synapse as a trivalent. Following 2:1 segregation, six types of gamete are produced (Fig. 6-5). “Alternate” segregation leads to the production of normal and balanced gametes; and adjacent segregation produces two types of disomic and two types of nullisomic gamete. 3:0 segregation is very rare (Honda et al., 2000; Durban et al., 2001).
In obvious contrast to what happens with the reciprocal translocation, the chromosomally abnormal conceptuses essentially have a complete aneuploidy. Only unbalanced conceptuses that are effectively trisomic for chromosome 13 or 21 can survive substantially through the course of the pregnancy (whether to fetal death, stillbirth, or live birth). Fetal trisomies 14, 15, and 22 are expected to end in miscarriage in the first trimester.
|
|
|
Figure 6-5. Meiotic behavior of the Robertsonian translocation. (a) Trivalent at synapsis. (b) Normal and (c) carrier gametes from “alternate” segregation. (d) Disomic and (e) nullisomic gametes from adjacent segregation. Note that there are six possible combinations (ignoring 3:0 segregation), of which two are normal/balanced, and four are unbalanced. |
Of these six possible outcomes, are some more likely to occur than others? Is there a “meiotic drive” favoring the production of a particular category of gamete? What is the relationship to the sex of the heterozygous parent? Undoubtedly, there is a differential survival of unbalanced forms, but could there be a difference between balanced and normal conceptions? Is an unbalanced translocation form more or less likely to survive than the same imbalance in simple form (translocation trisomy 21 versus standard trisomy 21, for example)? Answers to these questions are emerging, some consistent, others somewhat conflicting. Starting with the gamete, sperm and oocyte studies show considerable fractions of unbalanced forms. About 10%–15% of sperm of male Robertsonian heterozygotes show an unbalanced form (Honda et al., 2000; Frydman et al., 2001). On oocyte analysis, using the ingenious approach of FISH analysis on polar bodies, Munné et al. (2000a) determined in four 45,XX,rob(14q21q) carriers an average 42% of unbalanced forms, and seven 45,XX,rob(13q14q) carriers with an average 32%, and Durban et al. (2001) provide similar data (Table 6-2). Naturally, most if not all of the individuals proceeding to gamete testing in these reported studies will have experienced reproductive difficulty, and thus the data from their gametes may not necessarily be applicable to the larger number of carriers with apparently normal fertility.1
Alternate segregation is, apparently, favored, at least in most carriers (but see Infertility, below). In the male heterozygote, translocation Down syndrome and translocation trisomy 13 are scarcely ever seen in the offspring (Daniel et al., 1989; Pellestor, 1990). The characteristic cis-configuration of the trivalent at meiosis, with the two intact chromosomes lined up together opposite the matching q arm components of the translocation, is the probable reason that symmetric (alternate) segregation usually occurs, with a minority of disomic sperm, averaging about 10%, being produced from asymmetric segregation (Table 6-2) (Honda et al., 2000). In a correlation of sperm and preimplantation genetic diagnosis (PGD) karyotypes in two 45,XY,rob(13q14q) carriers, the frequencies of disomic/nullisomic sperm on the one hand and trisomic/monosomic embryos on the other were fairly similar, suggesting that these aneuploid sperm have unimpaired fertilizing ability (Escudero et al., 2000a). Selective pressures come into play subsequently, with abortion of chromosome 13 and 14 aneuploidies. Alves et al. (2002b) similarly showed about three-quarters of embryos at PGD from three male 13q14q heterozygotes to be normal/balanced, with around 15% reflecting adjacent segregation, and the remainder complex mosaic. The numbers of embryos from a 15q21q and a 14q21q carrier were too small to draw a firm conclusion, but it was interesting that none of 5 embryos of the 14q21q carrier showed adjacent segregation.
In the main, alternate segregation is favored also in the female, albeit less strongly so than in the male; however, unbalanced forms may actually be in the majority in some women. On inference from polar body analysis in a small number of cases, about one third to one half of oocytes display unbalanced segregation (Table 6-2). For the female rob(13q14q) heterozygote, translocation trisomy 13 is an uncommon observation in a term pregnancy (Table 6-3). Although one-tenth or thereabouts of oocytes may be disomic 13, apparently the great majority of resulting trisomic conceptions abort. In a PGD study that included four female rob(13q14q) carriers, about one-third of embryos were unbalanced due to adjacent segregation, and about 10% due to 3:0 segregation (Emiliani et al., 2002). These workers showed that selection begins to operate with progression through the morula and blastocyst stages, and the fraction of unbalanced embryos reduces. Selection presumably continues to operate as development proceeds, and the great majority of unbalanced pregnancies abort. In contrast, for the female rob(Dq21q) and rob(21q22q) carrier the risk for a translocation Down syndrome child is substantial. Preliminary findings from gamete research indicate that a substantial fraction of oocytes, up to about 20%, may be disomic 21, and evidently around half of these may survive through to term as a trisomy 21 pregnancy (Tables 6-2 and 6-3).
|
Table 6.2. Segregation Patterns in Gamete Studies of 14 Male and 15 female Robertsonian Heterozygotes. Each Male Case is Shown Separately; the Females are Grouped by Translocation Type and Citationa,b (given the small numbers of gametes) |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Meiotic Drive. Meiotic drive is an influence whereby one of the alternate products at meiosis may be favored and have a better than even chance of coming to be in the successful gamete. The Robertsonian translocation provides an apparent example. There had been conflicting reports in the literature, but in a stringent review, de Villena and Sapienza (2001) were able to substantiate the case. Female carriers of rob translocations have, in their offspring, a ratio close to 60:40 for the balanced rob compared to normal karyotypes. No such effect could be confirmed for the male rob carrier. Daniel (2002) has confirmed these interpretations in a retrospective analysis of prenatal diagnosis data, with rigorous attention to the need to avoid bias, showing a 116:81 ratio in favor of balanced carrier offspring compared to normal karyotypes where the mother is the carrier parent, versus a 42:41 ratio for carrier fathers (the fewer number of fathers reflect a male infertility with the rob; see below). Intriguingly, these observations are mostly not reflected in the findings at the direct examination of gametes (Table 6-2), for which no very satisfactory explanation can presently be offered.
|
Table 6.3. Estimates of Risks of Having a Child with Aneuploidy or with a Uniparental Disomy Syndrome for the Heterologous Rob Carrier |
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Postzygotic “Correction” and Uniparental Disomy
Trisomic Correction
An initial translocation trisomy may be “corrected” by mitotic loss of one of the free homologs and lead to uniparental disomy (UPD) in the embryo. For example, a presumed mechanism whereby UPD 15 could arise from a rob(13q15q) parent is outlined in Figure 6-6. Essentially, adjacent segregation produces a trisomic 15 conception, and then loss of the chromosome 15 contributed from the other parent2 at an early postzygotic stage “corrects” the karyotype. UPD has no abnormal effect if the chromosome is not subject to imprinting (except for the question of isozygosity for a recessive gene; see below), and chromosomes 13, 21, and 22 are in this category. If there is UPD for an imprintable chromosome—in this context, chromosome 14 or 15—a UPD syndrome would result. Engel and Antonarakis (2002) list just four recorded cases from a familial heterologous rob: one of upd(14)mat, one upd(14)pat, and two of upd(15)mat. James et al. (1994) made a specific search in a group of 14 phenotypically abnormal rob carriers, in most of whom the translocation was inherited, and identified only one case of UPD. A similar retrospective study by Berend et al. (2002a) which included 30 phenotypically abnormal heterozygotes with a familial heterologous translocation revealed two patients with UPD, both having a rob(14q15q), one with upd(14)mat, and the other upd(15)pat. Thus, UPD due to a parental rob is extremely rare, with the worldwide total of reported cases yet to be numbered in double figures.3
Residual Low-Level Trisomy. If the correcting mitosis occurs too late to include every cell that will contribute to the embryo proper, a translocation trisomic cell line may persist. The only example in the survey of Berend et al. (2000a) was that of a child with upd(13)mat from a rob(13q14q) mother, in whom a low level (4%) of trisomy 13 was shown on prenatal diagnosis, 45,XX,rob(13q14q)[48]/46,XX, 13,rob(13q14q)[2]. Bruyere et al. (2001) record in an abstract three such cases detected in a series of 281 prenatal diagnoses.
Monosomic Correction
Correction may also go the other way—that is, the conversion of a monosomic conceptus into a disomic one. Typically, the monosomy would be the consequence of a nondisjunction in a carrier mother having produced an ovum lacking one homolog (say, a nullisomic 14 ovum). The rescue is achieved by the replication of the homolog contributed in the gamete of the other parent. A replicate free chromosome may be produced, in which case the karyotype will appear normal. Or, the homolog could replicate as an isochromosome, producing the intriguing circumstance of a de novo Robertsonian-like chromosome in the setting of a true Robertsonian parent. Either event would take place at an early postzygotic stage, and would necessarily lead to uniparental isodisomy (Berend et al., 2000a). Since the initial malsegregation would most likely have been in maternal gametogenesis, with the replicating homolog therefore of paternal origin, the UPD in this setting will typically be a paternal isodisomy. Should there happen to be a recessive gene on this paternal chromosome, the child would be homozygous (isozygous) at this locus, and be affected with the particular Mendelian condition (p. 318). This phenomenon is apparently very rare.
|
|
|
Figure 6-6. Uniparental disomy 15 from a rob(13q15q) parent, due to “trisomy rescue.” The heterozygous parent produces a malsegregant gamete with the translocation, and with a free no. 15. The conception thus has trisomy 15. Subsequently, as a postzygotic event, the chromosome 15 from the other parent is lost. Since most malsegregations will have been of maternal origin, the UPD in this setting will usually be a maternal heterodisomy. (No. 13 elements are open, no. 15 elements are crosshatched. The two no. 15 elements from the carrier parent are asterisked.) |
Association with Infertility
Recurrent Miscarriage
Infertility includes the inability to achieve conception and the inability to sustain a pregnancy through to live birth (the latter also called “infecundity”). There is an approximately sevenfold excess of Robertsonian heterozygotes among couples who are infertile (Tharapel et al., 1985). A minority of rob carriers may have an individual predisposition, not necessarily shared by their heterozygous relatives, for a high frequency of unbalanced segregations, an insight that has been afforded by in vitro fertilization (IVF) studies. In Munné et al.'s (2000a) report noted above, three women had greater than half of their oocytes with aneuploidy. Alternatively, or additionally, there may be a proneness for a mitotic error in the immediate postzygotic cell divisions (Delhanty et al., 1997). Conn et al. (1998) treated two couples with a Robertsonian translocation who had been unable to achieve a normal pregnancy: a man with 45,rob(13q14q) and a woman with 45,rob(13q21q). They were able to karyotype a total of 33 3-day embryos from the two couples. A considerable majority of embryos, almost 90%, were chromosomally abnormal. Of these, 40% were trisomic or monosomic for 13, 14, or 21 (some mosaic, and some double monosomic), and this might have been expected. Notably, 60% had a “chaotic karyotype,” in which the chromosome constitution varied randomly from cell to cell, and indeed the karyotype of the original zygote could not usually be determined. Thus, the rob translocation may, of itself, compromise the fidelity of the first few mitoses in some infertile couples (Emiliani et al., 2003).
Male Infertility
A 10-fold excess of Robertsonian heterozygosity is found in men presenting with infertility due to oligospermia, especially in severe oligospermia (Chandley, 1988; Guichaoua et al., 1990; De Braekeleer and Dao, 1991). A case has been made that, in this setting, synapsis is incomplete in the trivalent, and the heterochromatic regions of the short arms remain unpaired; these “exposed” regions then interfere with pairing in the X-Y bivalent, so that spermatogenesis is blocked from further progression. Guichaoua et al. (1990) have directly observed the asynapsed short arms of the trivalent associating with the X-Y bivalent in testicular tissue from an oligospermic man heterozygous for a rob(14q22q), and Navarro et al. (1991) have similarly studied a rob(13q14q) man. Electronmicroscopic sperm analysis in a rob(14q22q) man with oligoasthenospermia showed marked ultrastructural defects in the great majority of spermatozoa, and attempted IVF was unsuccessful (Baccetti et al., 2002). Mice with several Robertsonian translocations show spermatogenic arrest if the translocations form a chain and associate with the sex chromosomes (Johannisson and Winking, 1994). The high frequency of male infertility in the general population (Skakkebæk et al., 1994) precludes straightforward interpretation in the individual infertile man with 45,XY,rob(13;14); it is also notable that men in the same family with the same translocation can be fertile (Rosenmann et al., 1985). Since most infertile men with a rob produce some sperm, assisted reproduction using IVF and intracytoplasmic sperm injection (ICSI) may in some be successful.
Rare Complexities
Mosaicism with Coexisting Robertsonian and Isochromosome Rearrangements
Gross et al. (1996) describe a woman who had had two children with DS, due to an isochromosome of 21q. She herself had two cell lines: one with a rob(21q22q) and the other with the i(21q), her karyotype thus 45,XX,rob (21q22q)/46,XX,i(21q). In blood only one cell was 46,i(21q), and in skin fibroblasts none was. However, on ovarian biopsy, about a third of the cells had the isochromosome, thus explaining the two DS children. In addition, one single blood cell was 45,XX,rob(14q21q). Gross et al. suggest that the woman's chromosomal complement was initially (at conception or very early embryogenesis) 46,i(21q), with the rob(21q22q) subsequently forming from one of the 21q arms of the isochromosome and one no. 22. The rob(14q21q) cell may represent a further such event with a no. 14 chromosome. In their literature review, Gross et al. compiled 14 other similar cases.
A different construction is put forward by Berend et al. (1998), who studied a child with DS and a mosaic karyotype with two different de novo acrocentric rearrangements. One cell line (from peripheral blood lymphocytes) was 45,rob(14q21q), and the other was 46,i(21q),21. At first, a sequential mechanism seemed the most probable, with one rearrangement setting up the circumstance for the other to arise. But the most ingenious and intellectually satisfying explanation is not necessarily the correct one! The rob(14q21q) comprised a no. 14 of maternal and a no. 21 of paternal origin, while the i(21q) was a duplicate of the other maternal chromosome 21. Thus, it appears that the two abnormal chromosomes, the Robertsonian translocation and the isochromosome, may have been generated at separate postzygotic events. Another case studied by this same group (Bandyopadhyay et al., 2003), a child with 45,rob(14q21q)/46,i(21q), may have had a simpler explanation, since the same paternal no. 21 was involved in both the rob and the isochromosome. Two contemporaneous events, each due to some sort of “instability” of the paternal no. 21, may have engendered the two cell lines right back at the time of the very first mitosis.
Heterozygous Couple
An interesting curiosity is the extremely rare case of a union between Robertsonian heterozygotes. For example, Martinez-Castro et al. (1984) describe two parents both with a 45,rob(13q14q) karyotype whose three phenotypically normal children had a diploid number of 44, with their no. 13s and their no. 14s existing as a pair of rob(13q14q) translocations. Two rob(13q14q) × rob(13q14q) couples, being first cousin pairs and all four having the same rob(13q14q) by descent, each presented with three first-trimester abortions in Bahçe et al. (1996). A couple both carrying a rob(14q21q) are recorded as having had a child with DS, with the unique karyotype 45,XY,rob(14q21q)pat,rob(14q21q)mat,21 mat (Rajangam et al., 1997). Similarly, Mori et al. (1985) reported a couple both of whom were 45,rob(13q15q), and who had had a child with translocation trisomy 13. Because of a founder effect, this otherwise rare Robertsonian translocation was rather common in their small village in the province of Cuidad Real in Spain, and this couple were presumably distantly related, even though they were unaware of any link. The reader may care to construct a hypothetical balanced karyotype with 2n = 41 and five Robertsonian translocation chromosomes (but theoretically nonviable due to absence of NORs).
The Homologous Robertsonian Translocation (or Acrocentric-Derived Isochromosome)
This Robertsonian translocation chromosome comprises the long arm elements of two acrocentric chromosomes that are the same. The site of formation is typically postmeiotic (Robinson et al,. 1994). If the translocation forms from the fusion of the two parental homologs, then manifestly there is biparental inheritance (Abrams et al., 2001). If, however, the rearrangement is actually an isochromosome, each long arm is an exact copy of the other, and there will be uniparental isodisomy. Such an isochromosome may have arisen as a “correction” of monosomy in the zygote.
MEIOSIS
Only two segregant outcomes are possible at meiosis in this 45,rob heterozygote. Either the gamete will receive the translocation chromosome and be effectively disomic, or it will not and be nullisomic. Essentially, this is 1:0 segregation (or “1+1”:0 segregation). No balanced gamete is possible. Thus, if the other gamete is normal, only trisomic or monosomic conceptions are possible. Occasionally, conceptuses with translocation trisomy 13 are viable, and translocation trisomy 21 not infrequently survives to term. None of the other unbalanced possibilities (trisomies 14, 15, and 22, nor any of the monosomies) is viable.
Postzygotic trisomic correction is a mechanism that, rarely, could enable the carrier to have a phenotypically normal child. If, say, in the case of an unbalanced 46,-22,rob(22q22q) conception, the free chromosome 22 were lost at a very early mitosis, genetic balance in this cell line would be restored, with a 45,rob(22q22q) karyotype. Provided the unbalanced cell line contributed negligibly or not at all to the embryo, and provided there were no effect due to uniparental disomy (and in the case of chromosome 22, there is not), the child would be normal. Engel and Antonarakis (2002) list only five such recorded circumstances, two involving upd(13)mat, one upd(13)pat, and two upd(22)mat. Monosomic rescue is another theoretical mechanism, whereby the homolog from the other parent could be duplicated postzygotically, as two separate homologs or as an isochromosome, to produce a pregnancy with either a normal karyotype or 45,iso. Finally, for completeness (but hardly ever in reality), gametic complementation is to be mentioned, whereby the non-rob parent contributes a gamete that happens to lack the homolog for which the rob parent's gamete is disomic (Berend et al., 1999). For the rob(14q14q) and rob(15q15q) carrier, even if one of these rescuing mechanisms did happen, the child would in any event be abnormal, since these UPDs lead to an abnormal phenotype.
Rare cases of mosaicism for a “Robertsonian isochromosome” offer insights into causative mechanisms. Bartsch et al. (1993) note some recorded cases of parental mosaicism for 47,i+(21q), and describe their own unique case of a woman with 47,+i(21p)/47,i+(21q)— some hundreds of cells from blood, gonad, marrow, and skin were 47,i+(21p), and one single blood cell was 47,i+(21q)—who had had two children with Down syndrome due to the karyotype 46,i(21q). In herself, apparently, the isochromosomes arose as a postzygotic event from a 47,+21 conception, with classic centromere misdivision at the pre-embryo stage. The i(21p) line came to be the predominant in most tissues, but the i(21q) line had at least some representation in gonad and blood.
GENETIC COUNSELING
The Heterologous Robertsonian Translocation Carrier
Infertility and Miscarriage
The Robertsonian translocation involving non-homologs is occasionally associated with repeated spontaneous abortion and male infertility. It is unclear, in an individual case, whether the association might be causal or fortuitous (see above, Biology). We can theorize that, in some miscarrying couples, there may have been a majority of zygotes with nonviable adjacent segregants; and in some infertile males, the translocation may disrupt spermatogenesis. Cytogenetic analysis of products of conception and of testicular tissue, respectively, may cast some light. It remains possible that some other cause could underlie the problem. The infertile male usually produces some sperm and may thus be a candidate for IVF using ICSI and possibly PGD (Lee and Munné, 2000; see Table 6-4 and Chapter 24).
RISKS OF HAVING ABNORMAL OFFSPRING FROM INDIVIDUAL TRANSLOCATIONS
Figures for the risks of having an abnormal child or for the probability of detecting an unbalanced form at prenatal diagnosis are taken (making a few assumptions about extrapolating to the rare translocations) from data of a number of North American and European collaborative studies (Harris et al., 1979; Ferguson-Smith, 1983; Daniel et al., 1989) and are set out in Table 6-3. These data relate essentially to the risk for a full trisomy. Risks for UPD are drawn from the review of Silverstein et al. (2002), again allowing for figures from the more common translocations being applicable to the rarer ones. Detailed comments on each individual translocation follow, with general comments thereafter on the theoretical risks of uniparental disomy, “isozygosity” for a recessive gene, and interchromosomal effect.
The More Common Translocations
rob(13q14q)
The karyotype of the balanced rob(13q14q) is shown in Figure 6-1. Translocation trisomy 13 can result from adjacent-1 segregation, with a typical Patau syndrome phenotype. The risk for this is very small. Almost all instances are index cases in families, not secondary cases. A review of several pedigrees in Harris et al. (1979), well subjected to statistical rigour, identified no increased risk for a malformed infant (they noted that a risk of up to 2% might have been missed, due to the sample size). In a European collaborative study, none of 230 prenatal diagnoses had an unbalanced karyotype (Boué and Gallano, 1984), suggesting a risk of less than 0.4%. An incidence in Daniel et al.'s (1989) North American data of 3/204 (1.5%) may have been influenced by ascertainment bias, but in any event combining the two data sets gives a figure of only 0.7%. A compromise figure of up to 0.5% may be a fair one to give. If there is male infertility, needing IVF with ICSI to achieve pregnancy, the additional exercise of PGD would be reasonable to improve the chances of producing a normal/balanced conception. PGD may also be a reasonable choice for some female heterozygotes (Frydman et al., 2001; Scriven et al., 2001). The facts relating to five couples assessed in Scriven et al., in whom this question arose, are set out in Table 6-4, and their cases illustrate various aspects. Emiliani et al. (2002) point out the risk for associated mosaicism of the embryo, and advise sampling of two blastomeres, rather than the usual single cell. The matter of UPD 14 is noted below.
rob(14q21q)
The rob(14q21q) is the most important Robertsonian translocation in terms of its frequency and genetic risk, and shows a marked difference according to the sex of the parent. Most familial translocation DS is due to the rob(14q21q) (Fig. 6-2). Adjacent segregation may lead to the conception of translocation trisomy 21 (Fig. 6-3). At amniocentesis, the female heterozygote has a risk for translocation trisomy 21 of about 15% (Ferguson-Smith, 1983; Boué and Gallano, 1984; Stene and Stengel-Rutkowski, 1988; Daniel et al., 1989).
The risk of having a liveborn child with translocation DS is a little less (around 10%). This likely reflects the loss, through spontaneous abortion, of a fraction of DS fetuses after the time during gestation when prenatal diagnosis is done.
|
Table 6.4. Reasons for Considering Preimplantation Genetic Diagnosis, and the Outcomes, in Five Robertsonian Couples |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The risk for the male heterozygote is very different. Very few examples of DS due to paternal transmission of a rob are known, and the risk figures that have been derived are all low single-digit percentages. In one FISH study on a rob(14;21) carrier, only 2.5% of more than 16,000 sperm analyzed were disomic 21, which is the karyotype that could give rise to translocation DS (Honda et al., 2000). If this figure were the same for all male carriers, and if a fraction of all trisomic 21 conceptions are lost (and see p. 364), a risk figure of about 1% is appropriate both on extrapolation from this sperm study and from empiric observations from clinical studies. The matter of UPD 14 is noted below.
The Rare Translocations
rob(13q15q)
Few data are available concerning genetic risks to the carrier (Mori et al., 1985; Daniel et al., 1989). We would expect these individuals are no more likely to produce adjacent segregants than the rob(13q14q) carrier, and a similar risk of 1% for translocation trisomy 13 may therefore apply. The risk for UPD 15 is noted below.
rob(13q21q)
In Boué and Gallano's (1984) study, the risk for translocation DS, in terms of the likelihood of detection at amniocentesis, was 10% for the female; in Daniel et al.'s (1989) study, the figure was 17%. This 10%–17% range suggests there is no real difference from the 15% that applies to the common rob(14q21q). The risk for the male heterozygote is low, and probably similar to the 1% proposed for the male rob(14q21q) carrier. A 0.5% or less risk for translocation trisomy 13 at term may apply for either sex. UPD is not a concern.
rob(13q22q)
We presume the risk for translocation trisomy 13 would be small, and perhaps similar to that for the rob(13q14q). In Boué and Gallano's (1984) study of 262 Robertsonian prenatal diagnoses not involving chromosome 21, there were only 3 rob(13q22q) cases, and in fact one of these showed trisomy 13; no unbalanced karyotypes were diagnosed in Daniel et al.'s (1989) 7 cases. UPD is not an issue.
rob(14q15q)
Adjacent segregants (translocation trisomy 14, translocation trisomy 15) are invariably lethal in utero. UPD 14 or UPD 15 are possible outcomes, as noted below.
rob(14q22q) and rob (15q22q)
The potentially trisomic states from these translocations (trisomy 14, 15, or 22) would all be anticipated to abort spontaneously. Neu et al. (1975) record the segregation of a rob(14q22q) chromosome in a large family in which some carriers had an increased miscarriage rate. We comment below on UPD.
rob(15q21q)
From Boué and Gallano's (1984) small series of nine carrier mothers, one (11%) had translocation trisomy 21 detected at amniocentesis; and in Daniel et al.'s (1989) data, the fraction was 0/9. These figures derive from too small a body of data to be sure that the risk is truly different from the more solidly based 15% that applies to the rob(14q21q) female carrier. Again, we suppose a low risk (about 1%) for the male carrier in terms of DS. The possibility of UPD is noted below.
rob(21q22q)
For a rob(21q22q) carrier parent, the risks for translocation trisomy 21 are about the same as for the rob(14q21q), according to the sex of the parent. UPD need not be a concern.
Uniparental disomy
Uniparental disomy in a setting of parental Robertsonian heterozygosity is extremely rare, as noted above. The four syndromes that can theoretically arise are UPD 14, maternal and paternal, and UPD 15, maternal and paternal. The UPD 14 syndromes are described in Engel and Antonarakis (2002), and maternal and paternal UPD 15 are better known as Prader-Willi and Angelman syndromes, respectively. In the review of Silverstein et al. (2002), combining prenatal diagnostic data from four groups, 2 instances of UPD were identified out of 315 prenatal diagnoses (mostly familial, some de novo), for a point estimate of 0.65%. In one of the two UPD pregnancies, with upd(13)mat from a mother with 45,XX,rob(13q14q), a low-grade residual trisomy 13 was also discovered. The other prenatal diagnosis was of a de novo rob(13q14q) with upd(14)mat
Should UPD testing be proposed when a balanced familial rob is identified at prenatal diagnosis, or even when the karyotype is normal, bearing in mind isodisomic UPD due to monosomic correction (Silverstein et al., 2002)? Berend et al. (2000a) suggest the question needs to be raised when chromosome 14 or 15 is involved. Gualandi et al. (2000) similarly propose that “it is reasonable to offer parent-of-origin studies in all situations at increased risk, after informed consent has been obtained in the course of a genetic counseling session.” Jay et al. (2001) are more sanguine, commenting that “routine UPD testing is not warranted where there is no chromosome 15 involvement.” At least all are in agreement concerning chromosome 15 translocations, and methylation-based testing with respect to the 5 SNRPN locus is appropriately offered (Glenn et al., 2000). Since the UPD 14 syndromes are coming to be accepted as true entities (Engel and Antonarakis, 2002), a case might conceivably be made to include chromosome 14 translocations as well. On the other hand, given the extreme rarity of (recognized) cases of UPD 14 (only a single-digit number of UPD 14 cases from a 14-containing rob being reported), and mindful of the contrasting figure of a few million4 rob(13q14q) heterozygotes in the world, not doing UPD testing can quite readily be justified. As for maternal and paternal UPD 13, 21, and 22, these are apparently without phenotypic effect, and need not be a cause for concern.
“Isozygosity” for a Recessive Gene. Monosomic rescue, whether producing an isochromosome or a 46,N karyotype, has the potential to cause an autosomal recessive disorder, should the non-rob parent happen to be heterozygous for a Mendelian condition whose locus was on the chromosome in question. But the risk is likely to be very low. Barring knowledge of such a condition (e.g., Bloom syndrome, locus on chromosome 15) elsewhere in the family, molecular testing is not practicable. Of the more common recessive genes that might in some jurisdictions be suitable for population screening (cystic fibrosis, thalassemia, Tay-Sachs, sickle cell), none has its locus on an acrocentric chromosome.
Interchromosomal Effect
The concept of an interchromosomal effect has been invoked in the setting of the balanced Robertsonian heterozygote. Could a translocation somehow influence the distribution of another chromosome not involved in the rearrangement, with the production of a gamete aneuploid for a chromosome not involved in the translocation? Anecdotal reports of DS children born to 14q22q and 13q14q rob carriers (Farag et al., 1987; Sikkema-Raddatz et al., 1997b) seemed to support this notion. However, formal segregation studies in large numbers of families with a rob(13q14q) or with trisomy 21 showed no excess of trisomic offspring or of parental Robertsonian translocations, respectively (Harris et al., 1979; Lindenbaum et al., 1985). Therman et al. (1989) ascertained no Robertsonian translocation through a trisomic child other than one that included the trisomic chromosome. Sperm karyotypes of male heterozygotes show no excess of disomy for other chromosomes (Pellestor, 1990; Syme and Martin, 1992). These pieces of evidence amass a rather strong case that the Robertsonian translocation typically influences the segregation of no chromosomes other than those of which it is comprised. Against these observations, however, are those of Gianaroli et al. (2002), who examined the embryos produced at IVF from a number of rob heterozygotes, eleven male and four female, showing that imbalance due to common aneuploidies (mostly chromosomes 13, 16, 18, 21, 22) actually outnumbered those due to the translocation per se. Their privileged window of observation may have offered a clearer view, although, as indeed these authors acknowledge, the fact of infertility in the couples, and the need in many for ICSI to have been used, may have led to bias. Perhaps an interchromosomal effect, if real, might apply only in the rob heterozygote whose gonadal function is compromised. In any event, their proposition that these other chromosomes be included in the PGD test panel for the rob carrier is to be noted.
The Homologous Robertsonian Translocation Carrier
We refer to these rearrangements as “rob,” recognizing, as discussed above, that some such cases do actually involve an acrocentric-derived isochromosome (“rob-iso”). Virtually all conceptions of the heterozygote result in either trisomy or monosomy. Monosomy results in occult abortion. Trisomies 14 and 15 always, and trisomy 22 virtually always, miscarry. Most trisomic 13 pregnancies miscarry, although some last until the third trimester, while many trisomic 21 pregnancies will proceed through to the birth of a child with Down syndrome. Practically speaking, no normal child could be produced (the scenario of postzygotic correction, discussed above, can scarcely be raised as a realistic hope). Appropriate advice for these carriers is to consider sterilization. Alternatively, the use of donor gametes may allow the couple to have a normal child.
Specific comments relating to the risk for abnormal offspring in each type of rob follow.
rob(13q13q)
The carrier parent can produce only monosomic or trisomic 13 conceptions, and these would either miscarry or, in the case of trisomy, produce a very abnormal child (Patau syndrome). Three recorded exceptions to this statement are given in Slater et al. (1994, 1995) and Stallard et al. (1995) of a normal parent having a normal child with rob(13q13q). The translocations were probably dicentric 13q isochromosomes, arising from postzygotic correction, and thus the children had uniparental isodisomy.5
rob(14q14q), rob(15q15q)
Trisomies and monosomies for chromosomes 14 and 15 are not viable, and thus all pregnancies of these heterozygotes would be expected to terminate in occult abortion or miscarriage. Even if postzygotic correction did happen, the child would have a UPD syndrome, according to the translocation and the sex of the transmitting parent. Thus it is, in theory and in reality, impossible to have a normal child from any gamete of the heterozygote. Cheung et al. (1997) made a prenatal diagnosis of a de novo rob(15q15q) in the balanced state and advised the parents that, in the fullness of time, their child “would not be able to bear offspring without clinical assistance.”
rob(21q21q)
Although the rob(21q21q) is extremely rare, every counselor knows about this famous translocation. It is a classic example of a genetic risk of (practically) 100%. All pregnancies continuing to term can be expected to produce a child with DS. Sudha and Gopinath (1990), for example, report a couple who had 13 pregnancies, with 4 children proven or presumed to have had DS, and 9 miscarriages. The mother was 45,rob(21q21q). No case of postzygotic correction for this translocation has ever been reported.
rob(22q22q)
All conceptions would be monosomic or trisomic 22, other things being equal. For example, one carrier woman had 24 miscarriages, but no normal child (Farah et al., 1975). Two cases are mentioned above of postzygotic correction, with the birth of a normal child, but this is not a realistic hope to offer in the individual case.
Prenatal Diagnosis of the De Novo Homologous Robertsonian Translocation. The de novo homologous Robertsonian translocation (or isochromosome) has a high risk for UPD; this entity is commented on in Chapter 25.
Notes
1. Differences between studies may reflect different criteria of reproductive abnormality, such as infertility due to oligospermia or normospermia with repeated miscarriage (Honda et al., 2000). Frydman et al. (2001) suggest that their slightly lower figures for unbalanced forms (9%–13% for 13;14, and 7%–9% for 14;21) may be the more accurate, having accounted for misinterpretation of nullisomy by using normal control spermatozoa.
2. Note that with one or the other no. 15 being the candidate to be lost, the risk for UPD to be generated is 50%. This is in contrast with correction in standard trisomy (p. 314), in which, with three candidate chromosomes, the chances are 1 in 3 for the “wrong” one to be lost.
3. Even more rare is UPD due to gametic complementation, whereby the rob parent contributes a 14- or 15containing rob, and the other parent a nullisomic 14 or 15 gamete (Berend et al., 2002a).
4. From 1/1000 (frequency of all rob) × 74% (proportion of rob with no. 14) × 6 billion (world population) = 4,440,000.
5. This makes the incidental point that UPD 13, maternal or paternal, is without phenotypic effect (Slater et al., 1995).
