Abstract

The constitutional t(11;22) is the most frequently occurring non-Robertsonian translocation in humans. The breakpoint (BP) of the t(11;22) has been identified within palindromic AT-rich repeats (PATRRs) on chromosomes 11 and 22, suggesting that hairpin/cruciform structures mediate double-strand breaks leading to the translocation. To further characterize the mechanism of the translocation, identification of the precise location of the translocation BP is essential. Thus, the PATRRs from normal chromosomes 11 have been analyzed in detail. The majority of individuals have a PATRR that is 445 bp in length with a nearly symmetrical structure. The shorter, previously reported 204 bp PATRR has been shown to be a rare polymorphism. There are several nucleotide differences between the proximal and distal arms of the 445 bp palindrome (cis-morphisms) that correspond to five polymorphic sites within the PATRR. Using these data, the junction fragments of 40 unrelated t(11;22) families have been examined to determine the position of their 11q23 BPs. Sequence analysis demonstrates that BPs are located at the center of the longer PATRR in 39 of 40 cases. The data suggest that the center of the palindrome is susceptible to double-strand breaks leading to translocations that sustain small symmetrical deletions at the BP junction. The sequence of the larger, chromosome 22 PATRR deduced from junction fragments has three cis-morphisms, and the derivative chromosomes sustain symmetric deletions at the center of 22q11 PATRR. In one unusual case, the BPs on both chromosomes appear to correspond to these cis-morphic sites, suggesting that double-strand breaks at mismatched regions caused this variant translocation. De novo t(11;22) BPs have been analyzed using translocations detected in sperm samples from normal males. cis-Morphisms reveal no exclusive utilization of a particular allele in meiosis to produce the translocation. Our data lend support to the hypothesis that palindrome-mediated double-strand breaks in meiosis cause illegitimate recombination between 11q23 and 22q11 resulting in this recurrent translocation.

Received June 29, 2001; Revised and Accepted September 10, 2001.

INTRODUCTION

The t(11;22) is the only known recurrent, site-specific, non-Robertsonian constitutional translocation in humans. Whereas balanced carriers have no clinical symptoms, severely affected offspring often demonstrate the supernumerary-der(22)t(11;22) syndrome through 3:1 meiotic malsegregation of the translocated chromosomes (1). In our previous study, we cloned the breakpoint (BP) of a balanced carrier and demonstrated that the BPs both on chromosomes 11 and 22 are located within similar palindromic AT-rich repeats (PATRRs) (2). PATRRs are predicted to form hairpin or cruciform DNA structures at physiological temperatures. Thus, we proposed that hairpin or cruciform DNA structures give rise to genomic instability that leads to the recurrent translocation. We analyzed the BPs of 40 unrelated t(11;22) cases and showed that all of the translocation BPs are located within PATRRs both on chromosomes 11 and 22 (3). Others have confirmed these findings (4).

Detailed analysis of the t(11;22) BP has been difficult because normal sequence for the chromosome 11 and 22 PATRRs is not available. Although human chromosome 22 has been completely sequenced, there still remain some gaps (5). The t(11;22) BP region on 22q11 is located within one of these gaps. Multiple BAC, PAC, cosmid and fosmid libraries have been screened but have not yielded a 22q11.2 BP spanning clone (1,2,68). Further, the sequenced chromosome 11 BAC clone encompassing the t(11;22) BP sustained a deletion of the entire PATRR (2). Previously, the gap on chromosome 11 was filled and the PATRR sequenced using the PCR product from a single individual (2). It has subsequently been determined that this sequence represents a rare polymorphic variant of shorter length. In order to investigate the mechanism of the recurrent t(11;22), a normal 11q23 sequence is required. However, long palindromic DNA sequences in the form of inverted repeats are known to be unstable and unclonable in bacteria (9). The use of recombination-deficient bacterial strains (10) has not improved the ability to clone this palindromic DNA. Although a novel procedure using a combination of bacteriophage λ nuclease and exonuclease III for cloning and sequencing long palindromic DNA sequences has been recently developed (11), the AT richness of this PATRR has prevented its analysis even with such approaches.

Thus, to successfully obtain the complete sequence of the chromosome 11 PATRR, we modified a conventional approach to generate multiple deletion mutants in plasmids. We applied this technique to PCR product containing the PATRR, obtained a shorter version of it, and then successfully cloned it. Here we report the complete sequence of the chromosome 11 PATRR and the precise location of multiple t(11;22) BPs. Further, we have identified nucleotide alterations between the proximal and distal arms of the PATRR (cis-morphisms) as well as variations between individuals (polymorphisms). These nucleotide alterations provide evidence suggestive of the dynamic nature of the PATRR.

RESULTS

Complete sequence of the chromosome 11 PATRR

For the initial analysis of the t(11;22) BP region, we compared the sequence of t(11;22) junction fragments with that of the 204 bp PATRR obtained by PCR of chromosome 11 from a normal individual (GenBank accession no. AF391128) (Fig. 1A) (2). In the 204 bp PATRR, there is a 24 bp asymmetric region at the center. In addition, there is an 8 bp mismatch in the stem of the putative hairpin, corresponding to an 8 bp deletion in the distal arm. Since the 24 bp region appeared only in the der(22) sequence, we initially localized the BP at the proximal start of the asymmetric region (2). However, additional analysis of t(11;22) junction fragments demonstrates that the 24 bp region also resides on the der(11) in the opposite orientation (Fig. 1B). The der(11) and der(22) sequence share 98% identity with one another, and the 8 bp region appears both on the der(11) and the der(22). Based on these data it seemed likely that there might be a polymorphic allele of the chromosome 11 PATRR. This PATRR allele would be at least 24 bp longer than the 204 bp PATRR and contains the 8 bp region on both its proximal and distal arms. Further, since the chromosome 11 sequence of the der(11) and the der(22) appear to be identical, we speculated that the longer PATRR would have a symmetrical structure and that the translocation BPs might be located at the center of the palindrome.

Thus, size variation of the chromosome 11 PATRR was analyzed by Southern hybridization. Sixteen DNA samples from normal individuals were studied following HindIII digestion and analysis with the PCR generated c14g probe flanking the PATRR (Fig. 1A). The majority of chromosomes yield a 4.5 kb band, whereas only one chromosome of the 32 studied has a shorter, 4.3 kb HindIII fragment (Fig. 2A). This is the same sample with the previously described 204 bp PATRR (2). Thus, these data indicate that the previously reported sequence was derived from a rare, shorter allele of the PATRR. Further, these results suggest that the majority of individuals have a larger palindrome, ∼200 bp longer than the originally reported 204 bp PATRR.

PCR with primers flanking the PATRR, using normal genomic DNA as template, was carried out to amplify the longer chromosome 11 PATRR. Using these 16 DNA samples, only the sample with the shorter allele easily produced a PCR product. The product was 850 bp in length. A 24 bp spacer-like region and 8 bp mismatch between the proximal and distal arms of the PATRR appear to facilitate amplification of the 850 bp PCR product. Others have previously noted their inability to generate a PCR product from the PATRR using standard approaches and available genomic sequence (4). It is likely that the longer PATRR is resistant to PCR amplification because it has a more complete palindrome that forms a ‘snap-back’ of the palindromic sequence. Thus, PCR was carried out at higher temperature to facilitate amplification by opening the ‘snap-back’ structure. Utilization of longer, 30 bp primers and higher temperature allowed amplification of PCR products in all DNAs tested (Fig. 2B). A 1.1 kb PCR product was generated from all samples except the one with the shorter PATRR. These new PCR conditions and primers produced a PCR product 250 bp longer than that of the original report (2).

To obtain the sequence of the normal chromosome 11 PATRR, sequencing of the PCR product from a t(11;22) balanced carrier who has only one intact chromosome 11 was attempted. DNA with significant secondary structure is difficult to sequence. PCR-based direct sequencing failed, presumably as a result of intrastrand ‘snap-back’ of the palindromic sequence. When the 1.1 kb PCR product was cloned into a plasmid vector, the palindromic region was invariably deleted during bacterial culture even when recombination deficient Escherichia coli strains were used. Thus, the PCR product was treated with nuclease to modify the symmetrical PATRR into an asymmetric, shorter version (Materials and Methods). Stabilization was accomplished with a minimum of a 50 bp deletion from either end of the PATRR allowing sequencing to be achieved. The sequence obtained is shown in Figure 3A (GenBank accession no. AF391129). The PATRR is 445 bp in length, with an AT content of 93%. The shorter, 204 bp PATRR has an asymmetric region at the center of the PATRR (GenBank accession no. AF391128). Further, there are two mismatched regions between the proximal and distal arms of the shorter PATRR including a deletion of 8 bp in the distal arm (cis-morphism). In the context of these PATRRs we define nucleotide differences between the proximal and distal arms of the palindrome as cis-morphisms. In contrast, there is no asymmetric region at the center of the 445 bp PATRR. The longer PATRR does not have this 8 bp mismatch and comprises a nearly complete palindromic structure (98% identity between the proximal and distal arms) (Fig. 3B). Therefore, the longer PATRR is predicted to form a nearly perfect hairpin structure. BAC clones encompassing this region (442e11, Genbank accession no. AC007707) have a complete deletion of the PATRR (Figs 1A and 2B). Cloning PCR products encompassing the PATRR in bacterial hosts results in variable deletions of the PATRR, converting the palindrome into an asymmetric structure (data not shown). In contrast, genomic DNA from long-term lymphoblastoid cell lines and a monochromosomal somatic cell hybrid with human chromosome 11 in a Chinese hamster background all yield the 1.1 kb PCR product (data not shown). Thus, the short, naturally occurring 204 bp rare polymorphic PATRR allele appears to be an atypical polymorphism that is a by-product derived from partial deletion of the long 445 bp PATRR.

Analysis of the t(11;22) BPs

Because of the AT-rich nature of the region, a nested PCR method to amplify the junction fragments from both derivative chromosomes was designed (3). Comparison of PCR products for the der(11) and der(22) from 40 balanced carriers from unrelated t(11;22) families indicates that they vary only slightly in size. Based on the assumption that the t(11;22) progenitor chromosome 11 had the longer, more prevalent allele of the PATRR, sequence derived from the 40 BPs was compared with the 445 bp chromosome 11 PATRR sequence (Fig. 4A; Table 1). Size differences among the PCR junction fragments appear to be the result of slight differences in BP position. Five of the 40 BPs are located precisely at the center of the PATRR (cases 4, 14, 15, 18 and 24). In all but one of the remaining cases, there are deletions of <50 bp around the BP. The endpoints of the deletion localize within a 50 bp region at the center of the PATRR in the majority of cases. Although the extent of the deletion differs among the cases, the deletion is symmetrically located at the center of the PATRR. The sequence derived from the proximal and the distal arms shows loss of the same number of nucleotides, resulting in an identical sequence derived from the chromosome 11 PATRR on both the der(11) and the der(22) (Fig. 4A). These data suggest a symmetrical structure for the BPs that is different from that previously reported (4).

PCR products of a markedly different size, indicating a completely different BP location, were produced from only one sample (case 23). Both junction fragments are identical to the typical 11q23 PATRR sequence up to the 8 bp region that is mismatched in the 204 bp PATRR, and then the sequence diverges (Fig. 5A). The chromosome 11 BPs for both the der(11) and der(22) localize to the 8 bp region, and then a significant portion of the PATRR including the center is deleted. On the chromosome 22 side of the der(11) sequence, the sequence beyond the BP partially matches the chromosome 22 PATRR sequence deduced from the other junction fragments. However, a 190 bp piece of a LINE1 element has been inserted into the AT-rich region of the chromosome 22 PATRR. An 8 bp target site duplication (TATATGTA) can be identified at both ends of the LINE1 sequence, suggesting it is likely to have been introduced by an insertion-based mechanism rather than a gene conversion. Short LINE1 insertions usually contain the 3′ end of a LINE1 element, but this 190 bp LINE1 sequence corresponds to one of the protein-coding regions. When compared with the LINE1 consensus sequence, numerous mutations and deletions are seen (77% identity). The mutations would render the protein non-functional.

PCR was performed to determine if the insertion is independent of the translocation and represents a population-based, chromosome 22 polymorphism. A 24 bp PCR primer was designed within the LINE1 element where the mutations cluster. The primer includes three of the variant nucleotides that differ from the consensus sequence. The second primer was located on chromosome 22 downstream of the LINE1 element. Seventy normal DNA samples were tested to determine whether they have the insertion including DNA samples of similar ethnic origin (Welsh). Only this t(11;22) carrier (case 23) yields the specific PCR product (data not shown). Therefore, we were unable to determine whether the LINE1 insertion took place before or after the translocation event.

cis-Morphism/polymorphism in the PATRR

A total of six polymorphisms were observed within the chromosome 11 PATRR of the der(11) and the der(22) sequence (positions 82, 119, 137, 155, 211, 223 and the corresponding nucleotides from the distal end of the PATRR) (Table 1). Among these polymorphisms, four sites corespond to cis-morphic sites between the proximal and distal arm of the normal 445 bp PATRR sequence (positions 119, 137, 155 and 223) (Table 1; Fig. 3). Because of the difficulty of the PCR and cloning approach, we did not generate sequences from an extensive number of normal individuals. However, it is likely that the sequence residing on the der(11) and the der(22) reflects the original sequence of the normal chromosome 11 and 22 PATRRs. Hence, the polymorphisms observed in BP junction fragments must exist as polymorphisms in normal PATRRs. Interestingly, positions 137 and 155 which show A→T (or T→A) nucleotide alterations tend to exist as heterozygous cis-morphisms (heterozygosity: 32/39 at position 137; 28/39 at position 155). In contrast, the remaining three sites that show insertion/deletion type differences tend to be homozygous in cis (heterozygosity: 0/40 at position 82, 15/39 at position 119 and 2/27 at position 211). The 8 bp deletion at the distal arm of the 204 bp PATRR was never detected in these junction fragments that are apparently derived from the larger 445 bp PATRR.

The chromosome 22 PATRR

We could not generate a PCR product encompassing a normal chromosome 22 PATRR. Despite the lack of a normal chromosome 22 sequence, identification of the BPs on both the der(11) and the der(22) junction fragments allowed us to reconstitute the normal chromosome 22 PATRR structure. The BPs on chromosome 22 varied among the carriers, suggesting that variable deletions also occur within the chromosome 22 PATRR. The deletions are symmetrical and located at the center of the PATRR in the majority of cases (Fig. 4B). The size of the PATRR that can be deduced from sequence of the junction fragments is 106 bp. However, the chromosome 22 BP region has a non-AT-rich, palindromic, NF1-like sequence at either end. The size of this region is 194 bp per arm. Hence, the total size of the chromosome 22 palindrome is at least 600 bp in length. Case 35 has a particularly long chromosome 22 sequence on the der(11). This sequence is 108 bp longer than the others, despite a typical size for the der(22) junction fragment. On the assumption that the normal chromosome 22 PATRR contains this 108 bp and also has a symmetric palindromic structure, the total size of the chromosome 22 PATRR may be as large as 816 bp. This indicates that the majority of t(11;22) carriers may have a large deletion generated by the translocation at the chromosome 22 PATRR. The size of the deletion could be >200 bp. There are also three cis-morphisms between the sequences of the proximal and distal arms of the chromosome 22 PATRR (Table 1). These sites of cis-morphism also show nucleotide changes among the carriers, suggesting the existence of polymorphisms in the normal sequence.

Analysis of de novo t(11;22)s from sperm samples

We previously examined sperm samples from four normal individuals to determine the frequency of t(11;22)s generated de novo. Junction fragment PCR analysis revealed the presence of de novo translocations in all four samples (12). Over 100 de novo junction fragments have been identified and the size of the PCR products is similar in all cases analyzed. Sequence analysis of 50 of these PCR products demonstrates BPs similar to those observed in the 39 typical t(11;22) families described in Table 1. Chromosome 11 BPs are located within the 50 bp region at the center of the 11q23 PATRR. The frequency of de novo translocation differs between the samples from different sperm donors, suggesting that there might be differences in the relative predisposition for producing this translocation among individuals. We reasoned that an individual with a more complete palindrome might be predisposed to producing this translocation in gametogenesis based on enhanced susceptibility to secondary structure formation. Thus, the sequence of the BP regions of these de novo t(11;22)s has been examined for the cis-morphism/polymorphisms (Table 2). Sequence analysis of both the der(11) and der(22) junction fragments demonstrates that all of the donors have the 445 bp PATRR on chromosome 11. Three of the subjects are heterozygous for some of the polymorphisms in the 445 bp PATRR. Although the cis-morphic status cannot be determined, sequence analysis of the PCR products demonstrates no exclusive utilization of a specific chromosome 11 PATRR allele in at least three of the donors (SP1, SP3 and SP4). This is based on the presence of heterozygous alleles of the chromosome 11 PATRR on both the proximal and distal arms of their junction fragments (Table 2). Further, although the frequency of de novo translocation was three times higher in SP4 than in other samples, this individual is also a heterozygote for the cis-morphism and no allele preference was observed in his t(11;22) junction fragments. These findings suggest that sequence variations within the PATRR do not play a direct role in predisposition for the translocation.

The chromosome 22 side of the de novo junction fragments was also analyzed for the cis-morphism/polymorphisms deduced from the chromosome 22 portion of BP junction fragments obtained from t(11;22) carriers. There does not appear to be a preferential usage of cis-morphisms in the chromosome 22 PATRR either. In two individuals (SP2 and SP4), more than three types of chromosome 22 PATRR allele were identified on either the der(11) or the der(22). This is in contrast with the level of variability for the chromosome 11 PATRR (Table 2). These results suggest the possibility that there might be several PATRRs of similar sequence arranged in tandem at the 22q11 BP region.

DISCUSSION

Genomic instability of the chromosome 11 PATRR

In this study, we report the complete sequence of the PATRR at the t(11;22) BP in 11q23. It is likely that this PATRR forms single-stranded hairpin or double-stranded cruciform structures even at physiological temperatures. It is known that such secondary structures mediate rearrangements leading to loss of the palindromic sequence. This has presumably prevented the cloning of this region by traditional methods. Using nucleases to shorten the palindrome and render it asymmetric, we have successfully cloned the chromosome 11 PATRR. Whereas the AT content of this PATRR is as high as 93% in its 445 bp, Gs and Cs display a biased distribution within the first 30 bp at the base of the putative hairpin stem (local GC content 38%) (Fig. 3A). Clones generated with a deletion of this relatively GC-rich region from either arm did not delete the insert despite the symmetric structure of the remaining PATRR, suggesting that the GC-rich region at the base of the hairpin stem plays an important role in stabilization of the secondary structure. The chromosome 22 PATRR shows a similar structure, containing a non-AT-rich flanking region (NF1-like) that is much longer than that of the chromosome 11 PATRR. The region of 22q11 where the t(11;22) BPs cluster is suggested to be a ‘hotspot’ for translocations involving other chromosomes with 22, and the long chromosome 22 PATRR is likely to be responsible for such translocations (1315). Perhaps extreme susceptibility of the chromosome 22 PATRR to the formation of an unstable secondary structure leads to these other occasional translocations.

The chromosome 11 PATRR is 445 bp long in the majority of normal individuals. It consists of a nearly perfect palindromic sequence, forming a symmetric hairpin. In bacteria, a palindrome exclusion system seems to operate and a palindromic sequence is generally susceptible to being entirely or partially deleted. In bacterial hosts, the perfect symmetry of a palindrome is often converted into an asymmetric structure resulting in stabilization (9). DNA slippage on the lagging strand during replication is a mechanism proposed as being responsible for the deletion (9). However, since non-homologous translocation appears to be an outcome associated with such PATRRs in humans, perhaps there exists another mechanism for palindrome resolution in higher organisms. Perhaps the system permits palindrome-mediated rearrangement of the genome. In contrast to bacteria and yeast, deletion within the PATRR is less frequently seen in humans. Only one normal chromosome 11 out of 32 examined has been identified as having a shorter 204 bp PATRR, which appears to be a variant generated from the 445 bp PATRR by deletion. Since the short PATRR appears to be a rare polymorphism, the longer complete PATRR is stably transmitted even in meiotic divisions except for occasional translocation formation. Long-term cultured human cell lines do not delete the PATRR, suggesting that this PATRR is relatively stably maintained and transmitted without deletion during mitosis. Our data indicates that a somatic cell hybrid with a normal human chromosome 11 in a Chinese hamster background faithfully maintains the PATRR, suggesting stability in rodents. On the other hand, palindromic transgenes in mice demonstrate a significant number of deletions (16,17). All of the BAC clones that encompass the 11q23 BP region completely delete the chromosome 11 PATRR and YAC clones that include the PATRR retain only a portion of it (data not shown). Perhaps different mechanisms involving palindrome stability operate between prokaryotes and higher organisms, between mitosis and meiosis, and between integrated chromatin and episomal DNA (18).

The BP of the t(11;22) is located at the tip of the hairpin

We have isolated the junction fragments from both the der(11) and der(22) from 40 independent t(11;22) families and found that all of the translocation BPs are located within the PATRR (3). Others have indicated similar findings (4,19). In this study, we have further localized the BPs at the center of the PATRR. A similar mechanism has been proposed based on sequence of t(11;22) junction fragments in the absence of the complete normal chromosome 11 or 22 sequences (4). Since we have now generated the complete sequence of the normal chromosome 11 PATRR, we have been able to verify the precise location of the t(11;22) BP. It is believed that palindromic DNA forms hairpin/cruciform structures and our data indicate that the BPs are located at the tip of this putative hairpin. Since the tip of the hairpin is sensitive to nucleases (20), the initiating step of the translocation may be a double-strand break mediated by this hairpin-nicking activity.

The majority of cases have variable numbers of nucleotides deleted from the chromosome 11 PATRR. Even with these small deletions, the actual BP is likely to be initiated at the tip of the hairpin, because reconstruction of the normal sequences from the junction fragments generates a symmetric structure. Our analysis includes three of the same samples reported as asymmetric deletions by Edelmann et al. (4) (cases 32, 33 and 34). We interpret these three translocations as having sustained symmetric deletions (Fig. 4, case 32). DNA at the free end of the nick may be further deleted by nuclease activity prior to repair. Escherichia coli SbcCD protein forms a large complex and functions as an ATP-dependent double-strand DNA exonuclease and an ATP-independent single-strand endonuclease. It has been demonstrated that this protein complex cleaves the DNA of hairpin structures in vitro (20). The initial cleavage occurs at the tip of the hairpin and longer incubation deletes bases moving towards the stem. The mammalian homologs of the SbcC and SbcD may be Rad50 and Mre11, respectively (21) and a role for these enzymes during recombination has been proposed (22). Perhaps these proteins play a role in the generation of the t(11;22) translocation. Thus, in the future, studies to determine the involvement of these proteins in t(11;22) translocation formation are likely to assist in the elucidation of the mechanism.

The chromosome 22 side of the BP also demonstrates deletions, suggesting that the chromosome 22 PATRR might also be cleaved by such a hairpin-specific nuclease activity. Only when such a nick is generated on both strands, does a complete double-strand break of the chromosome occur. It has been demonstrated that two double-strand breaks, each on different chromosomes are sufficient to generate a reciprocal translocation (23). Thus, the PATRRs on chromosomes 11 and 22, both of which form hairpin/cruciform structures at physiological temperatures, appear to be independently cleaved at their tip by a hairpin-specific nuclease (Fig. 5B). This is likely followed by the introduction of subsequent nucleotide deletions by additional nucleases. When both double-strand breaks occur in a single cell, the translocation may be generated by a repair mechanism involving non-homologous end joining (18,22).

cis-Morphism/polymorphism in the PATRR and predisposition of t(11;22)

We identified sequence differences between the proximal and distal arms of the PATRR. Interestingly, the sites that show cis-morphisms also demonstrate polymorphism among individuals. This fact cannot easily be explained by the evolution of these mutations and their relation to the generation of the inverted repeat. Although mfold (http://BiBiServ.TechFak.Uni-Bielefeld.DE/mfold/) predicts a simple hairpin structure as a secondary structural model for the chromosome 11 PATRR, the palindrome may form kinks at certain positions resulting in different sensitivity to nuclease activity (http://www2.icgeb.trieste.it/~dna/bend_it.html) (24). Such additional secondary structure may make certain nucleotide positions within the PATRR susceptible to mutation, resulting in alterations of the same nucleotide as a cis-morphism and a polymorphism.

An interesting observation is that amongst the five variant sites in the chromosome 11 PATRR, three have a tendency to be homozygous in cis. These three are insertion/deletion type polymorphisms. In contrast, single nucleotide alterations tend to be heterozygous in the PATRR. This suggests that a mismatch repair mechanism may work on pseudo-double-stranded DNA in the hairpin structure. It is well known that defects of the mismatch repair system lead to dinucleotide repeat instability in humans. Insertion/deletion type mismatches rather than single nucleotide alterations in the PATRR may tend to be a substrate for mismatch repair enzymes.

Analysis of de novo translocations using sperm samples demonstrated differences in the frequency among individuals, implicating differential susceptibility to this translocation (12). Our data suggest that paternal age effects do not seem to affect the frequency. With regard to the palindromic nature of the BP region, it could be hypothesized that an individual with a more complete palindrome might be more predisposed to generating the translocation. However, sequence of the junction fragments suggests that polymorphism of the chromosome 11 PATRR does not appear to contribute to differences in translocation susceptibility. The chromosome 22 side cannot be analyzed in detail because of the unavailability of the entire PATRR sequence. One possibility is that there is no predisposition among individuals and a factor that might alter the frequency is related to temperature. Even at 42°C the PATRR on chromosome 11 denatures (unpublished data). Thus, body temperature of the individual or temperature at the time of sampling may affect the frequency. A higher body temperature could promote secondary structure, which is likely to contribute to generation of the translocation.

Chromosome 22 PATRR as a hotspot for translocation

The estimated size of the chromosome 22 PATRR deduced from junction fragment sequences is at least 600 bp, similar to that of the chromosome 11 PATRR. The size of the AT-rich sequence within the chromosome 22 PATRR is shorter than that of the chromosome 11 PATRR. Further, the chromosome 22 PATRR appears to be a site for translocations involving not only the chromosome 11 PATRR, but also other chromosomal regions (1315). However, the structure of this region is still an enigma. We could not generate a PCR product encompassing a normal chromosome 22 PATRR, even though we designed PCR primers located in the GC-rich region inside the palindrome that would amplify a product that should be smaller than that of the chromosome 11 PATRR. Palindromic structures may have such a self-annealing, ‘snap-back’ nature that primers are unable to anneal to the template DNA. Because of the absence of sequence data, we could only design one primer flanking the PATRR, resulting in PCR failure. This may also indicate that something is still missing from the chromosome 22 sequence deduced from junction fragments. Possible explanations for the failure of the PCR could be the size of the gap, the AT-richness of the sequence within the gap or the presence of significant secondary structure.

The sequence deduced from the de novo junction fragments of sperm samples from individual donors were analyzed with respect to the cis-morphism/polymorphisms. As many as three different sequence variants of the chromosome 22 portion of either the proximal or distal arms are seen amongst the junction fragments isolated from a single individual. It suggests that there are potentially more than two similar tandem PATRRs located at the BP region on 22q11. In addition to the hairpin stability, the presence of multiple tandem palindromes may provide another possible explanation for the numerous translocations that take place at 22q11.

An unusual mechanism for generation of the t(11;22)

One atypical case shows a different position of t(11;22) BP. The der(11) junction fragment of case 23 includes a LINE1 insertion in the chromosome 22 side. None of the 39 other der(11) junction fragments show the insertion. Since we could not identify this LINE1 insertion in 140 normal chromosomes, we cannot determine whether the insertion took place before or after the translocation event. However, since the only case with a different BP has this insertion, it is unlikely to be a coincidence. Capture of a transposable element during the translocation process (25) is not likely, since the location of the insertion is 66 bp from the translocation BP. Since numerous mutations and deletions were identified within the inserted LINE1 sequence, the insertion is likely to be an old event, probably occurring before the translocation took place and segregating as a rare polymorphism.

In the analysis of de novo t(11;22)s in sperm samples from normal individuals, we found that all four individuals examined have the 445 bp chromosome 11 PATRR. Although over 100 junction fragments from de novo translocations were obtained, all of the translocation BPs are similar and none demonstrated this type of atypical BP or LINE1-element-generated rearrangement. The implication is that the ancestral chromosome 11 in this t(11;22) family (case 23) may have had a different sequence in their PATRR, generating the translocation through a different mechanism. The chromosome 11 BP of this atypical case corresponds to the site where the short 8 bp region is deleted from the distal arm of 204 bp PATRR. The short 204 bp PATRR has an asymmetric region at the center as well as the 8 bp deletion on the distal arm, resulting in an asymmetric chromosome 11 structure. Since the LINE1 fragment is located in the stem of the putative chromosome 22 PATRR, the LINE1 insertion also renders the chromosome 22 PATRR asymmetric (Fig. 5B). This type of palindromic structure may not be cleaved at the tip because of its asymmetry. The BP of the chromosome 11 PATRR of this case indicates that the breakage might occur as a result of mismatch cleavage by a nuclease. Thus, mismatched regions in the PATRRs may also be susceptible to double-strand breaks leading to the translocation. Additional studies to determine the frequency of such variant mechanisms should assist in the elucidation of genomic instability mechanisms that lead to chromosomal rearrangements in general and the t(11;22) in particular.

The PATRR mediates site-specific double-strand breaks and repair as a mechanism of t(11;22) generation

In this paper, we demonstrate that t(11;22) BPs are located at the hairpin tip in the majority (39/40) of the t(11;22)s. It is remarkable that the BPs cluster within a 50 bp region at the center of chromosome 11 PATRR. Whereas constitutional translocations usually occur at random, the t(11;22) is the only known recurrent translocation. Cancer-related translocations occur in a non-random fashion, since the translocation leads to clonal expansion of the progenitor cell. Even in cancer-related translocations, the translocation BPs do not localize to such a confined region. Hence, we propose that the center of the PATRR forms the tip of a hairpin that is susceptible to breakage, leading to a double-strand break. In our analysis of sperm samples, t(11;22)-specific PCR products were detected only in meiotic and not in mitotic cells, even when the somatic cells were derived from individuals with chromosomal instability syndromes (12). Introduction of the PATRR into a meiotic cell type, yeast or the rodent germline, would represent future experiments to examine the proposed mechanism of t(11;22) generation. Further, it will be of particular interest to determine whether there are additional PATRRs in the human genome which are sites of recombination with chromosomes 22 and 11. Further, the t(11;22) may be generated through a physiological double-strand break and repair system during meiosis. Clarification of the mechanism of formation of the t(11;22) may lead to an understanding of double-strand break and repair pathways that operate in normal meiosis.

MATERIALS AND METHODS

Samples

All samples used in this study were obtained with informed consent. Some of the samples used in this study were derived from patients participating in a research study of the t(11;22) as reported previously(1). Genomic DNA was isolated from whole blood using the Puregene DNA isolation kit (Gentra Systems, Minneapolis, MN). GM03372, GM04403, GM06229, GM03193 (Welsh) and GM08228 (Welsh) were purchased from the Coriell Cell Repositories (Camden, NJ). Sperm samples were obtained from normal healthy volunteers after informed consent. Sperm DNA was extracted using similar methods.

Sequence analysis of the PATRR

The PCR primers were designed using the sequence of BAC 442e11 (GenBank accession no. AC007707). To amplify the PCR product containing the PATRR, longer primers with higher anealing temperatures were used to avoid ‘snap-back’ annealing of the palindromic region. 30mer primers were used for the amplification: H3INV reverse, 5′-GGAAGTTAGAGAAAACTGAGAATATACAGA-3′; BSTINV forward, 5′-AGACTCTCATTCATGGAACCCCAAACCATATG-3. PCR conditions were as follows: five cycles of 94°C for 30 s, 68°C for 30 s and 72°C for 1 min, followed by 35 cycles of 94°C for 30 s, 66°C for 30 s and 72°C for 1 min. PCR products were purified with the QIAquick gel purification kit (Qiagen Inc., Valencia, CA).

A series of deletion mutants were generated with the Exo-Mung Deletion Kit (Stratagene, La Jolla, CA). For protection of one end of the PATRR, the PCR product was cut with a restriction enzyme that produces a 5′ overhang and the resulting overhang was filled in with α-thio-dNTPs. The product was subjected to exonucleaseIII and mung bean nuclease to decrease its length. The deleted products were size fractionated by agarose-electrophoresis and the DNA was extracted from the gel fraction containing the expected fragments. Then the DNA was blunt-ended with T4 DNA polymerase, phosphorylated with T4 kinase and ligated into pBluescript (Stratagene). The construct was transfected into SURE cells (mcrA mcrBC recB sbcC recJ umuC uvrC) (Stratagene) by electroporation.

Southern analysis

The sequence flanking the t(11;22) BP was obtained from the t(11;22) BP spanning BAC, b442e11 (accession no. AC007707). The probe used for Southern analysis, c14g, was generated by PCR from b442e11. The primers used are: c14g forward, 5′-TGGGTCTTACCTCTTGAGCT-3′; c14g reverse, 5′-CTGAAAGTAGAGGTGACGGA-3′. PCR conditions are as described earlier (2). Probes were radio-labeled with [α-32P]dCTP using the random primer method. Southern hybridization was performed using standard methodology.

BP analysis

To obtain the junction fragments from the der(11) or the der(22), primers used for PCR amplification were as follows: c14i reverse, (5′-GGAAGTTAGAGAAAACTGAGAA-3′) and JF22.2 (5′-CCTCCAACGGATCCATACT-3′) for der(11), and c14h forward, (5′-AACACTCCCACTGACAGCTA-3′) and JF22.2 for der(22). PCR conditions were as follows: five cycles of 94°C for 30 s, 58°C for 30 s and 72°C for 1 min, followed by 35 cycles of 94°C for 30 s, 56°C for 30 s and 72°C for 1 min. Primers for the second PCR were JFN11 (5′-CAGAAAGGGAGAGCATGTAG-3′) and JFN22 (5′-CGTTGAAGGATGCAGGATGT-3′) for the der(11), and JFN11.2 (GGTTGAAGAATCTTGGCTGG-3′) and JFN22 for the der(22). The condition of the second PCR was 35 cycles of 94°C for 15 s, 58°C for 15 s and 72°C for 30 s. The JF22.2 primer corresponds to c appearing in Kehrer-Sawatzki et al. (14).

To amplify the chromosome 22 region with the LINE1 insertion, PCR was performed with OH1 primers (5′-GGAGAATGAGTTTGACGAATTAAC-3′) in conjunction with JF22.2 primer.

ACKNOWLEDGEMENTS

The authors wish to thank Tamim Shaikh for extremely valuable discussions and careful review of the manuscript. We would like to thank Stephanie St.Pierre and the t(11;22) Together Network (http://www.nt.net/~a815/index.html), the Kennedy/Hopkins NICHD Mental Retardation Core Grant, HD24061, and numerous clinicians for assistance in obtaining patient samples for this study. These studies were supported in part by CA39926, DC02027 and HD26979 from the NIH. The studies were also supported by funds provided by the Charles E.H.Upham Chair in Pediatrics to B.S.E.

+

To whom correspondence should be addressed at: The Children’s Hospital of Philadelphia, 1002 Abramson Research Center, 3516 Civic Center Boulevard, Philadelphia, PA 19104, USA. Tel: +1 215 590 3856; Fax: +1 215 590 3764; Email: beverly@mail.med.upenn.eduPresent address:Hiroki Kurahashi, Division of Functional Genomics, Department of Post-Genomics and Diseases, Osaka University Medical School, B9, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan

Figure 1. (A) Diagram of the chromosome 11 PATRR. The chromosome 11 genomic region in the vicinity of the PATRR is shown to-scale from proximal to distal; 445 bp PATRR (top), 204 bp PATRR (middle), and BAC442e11 (bottom). The PATRR is indicated by a black box. PCR primers used for amplification of the PATRR are indicated by arrows. The probe which was used for Southern hybridization, c14g, is indicated by the open box. H, HindIII. (B) Sequence comparison between the der(11) and the der(22) junction fragments derived from a single t(11;22) balanced carrier. Sequences are shown from the chromosome 11 side to the chromosome 22 side. Authentic BPs deduced from the longer 445 bp PATRR are indicated by arrows. Lower case letters indicate chromosome 11 sequence, and upper case letters indicate chromosome 22 sequence. Asterisks indicate nucleotides that match between the two sequences. Although the der(11) and the der(22) junction fragment sequences are derived from different derivative chromosomes, they show 98% identity. The sequence shown in bold in the der(22) is identical to the sequence of the asymmetric loop at the center of the short, polymorphic 204 bp PATRR (2). Triangles indicate BPs previously deduced from the shorter 204 bp PATRR. The sequence shown in bold appears in the der(11) sequence in the opposite direction (italics). The boxed 8 bp indicates the region deleted from the distal arm of the short 204 bp PATRR that creates a mismatch between the proximal and distal arms of the 204 bp PATRR. Although the distal arm of the 204 bp PATRR does not have this 8 bp region, it is observed in the majority of der(22) sequences derived from t(11;22) carriers.

Figure 1. (A) Diagram of the chromosome 11 PATRR. The chromosome 11 genomic region in the vicinity of the PATRR is shown to-scale from proximal to distal; 445 bp PATRR (top), 204 bp PATRR (middle), and BAC442e11 (bottom). The PATRR is indicated by a black box. PCR primers used for amplification of the PATRR are indicated by arrows. The probe which was used for Southern hybridization, c14g, is indicated by the open box. H, HindIII. (B) Sequence comparison between the der(11) and the der(22) junction fragments derived from a single t(11;22) balanced carrier. Sequences are shown from the chromosome 11 side to the chromosome 22 side. Authentic BPs deduced from the longer 445 bp PATRR are indicated by arrows. Lower case letters indicate chromosome 11 sequence, and upper case letters indicate chromosome 22 sequence. Asterisks indicate nucleotides that match between the two sequences. Although the der(11) and the der(22) junction fragment sequences are derived from different derivative chromosomes, they show 98% identity. The sequence shown in bold in the der(22) is identical to the sequence of the asymmetric loop at the center of the short, polymorphic 204 bp PATRR (2). Triangles indicate BPs previously deduced from the shorter 204 bp PATRR. The sequence shown in bold appears in the der(11) sequence in the opposite direction (italics). The boxed 8 bp indicates the region deleted from the distal arm of the short 204 bp PATRR that creates a mismatch between the proximal and distal arms of the 204 bp PATRR. Although the distal arm of the 204 bp PATRR does not have this 8 bp region, it is observed in the majority of der(22) sequences derived from t(11;22) carriers.

Figure 2. Size polymorphism of the chromosome 11 PATRR. (A) Southern analysis to demonstrate the size polymorphism of the chromosome 11 PATRR. The c14g probe flanking the chromosome 11 PATRR was generated by PCR and hybridized to HindIII-digested DNA from normal individuals. The majority yield a 4.5 kb band (arrow a). One individual (lane 1) is a heterozygote for 4.5 and 4.3 kb alleles (arrow b). (B) PCR analysis to demonstrate the size polymorphism of the chromosome 11 PATRR. M, 1 kb ladder plus (Gibco BRL, Rockville, MD) used as a size marker. Lane 1, BAC 442e11 which deleted the entire PATRR; lanes 2–5, DNA samples from normal individuals; lane 6, H2O control. The majority of samples yield a 1.1 kb PCR product (lanes 3–5). One individual yields a smaller 850 bp PCR product (lane 2). Although this individual is heterozygous for the polymorphism by Southern analysis, DNA from this individual yields only the shorter PCR product due to the difficulty of amplifying the long PCR product. The ladder is an artifact originating from ‘stutter bands’ on amplification of the AT-rich repeat.

Figure 2. Size polymorphism of the chromosome 11 PATRR. (A) Southern analysis to demonstrate the size polymorphism of the chromosome 11 PATRR. The c14g probe flanking the chromosome 11 PATRR was generated by PCR and hybridized to HindIII-digested DNA from normal individuals. The majority yield a 4.5 kb band (arrow a). One individual (lane 1) is a heterozygote for 4.5 and 4.3 kb alleles (arrow b). (B) PCR analysis to demonstrate the size polymorphism of the chromosome 11 PATRR. M, 1 kb ladder plus (Gibco BRL, Rockville, MD) used as a size marker. Lane 1, BAC 442e11 which deleted the entire PATRR; lanes 2–5, DNA samples from normal individuals; lane 6, H2O control. The majority of samples yield a 1.1 kb PCR product (lanes 3–5). One individual yields a smaller 850 bp PCR product (lane 2). Although this individual is heterozygous for the polymorphism by Southern analysis, DNA from this individual yields only the shorter PCR product due to the difficulty of amplifying the long PCR product. The ladder is an artifact originating from ‘stutter bands’ on amplification of the AT-rich repeat.

Figure 3. Entire sequence of the 445 bp PATRR on chromosome 11. (A) Double strand format of the 445 bp PATRR sequence. Numbers indicate the nucleotide positions from the starting point of the PATRR on chromosome 11 (shown proximal to distal). The sequence forms an inverted repeat structure. The distal half comprises a complementary sequence to the proximal half. The tip of the putative hairpin structure is indicated by an arrow. Relatively GC-rich regions within the PATRR are boxed. (B) Sequence comparisons between the proximal and distal (complementary) arms of the PATRR. The 8 bp region deleted from the short 204 bp PATRR is underlined. Six sites showing cis-morphism or polymorphism are indicated in bold (proximal positions 82, 119, 137, 155, 211 and 223). Heterozygosity for the cis-morphisms at positions 119, 137, 155 and 223 are observed in this sequence.

Figure 3. Entire sequence of the 445 bp PATRR on chromosome 11. (A) Double strand format of the 445 bp PATRR sequence. Numbers indicate the nucleotide positions from the starting point of the PATRR on chromosome 11 (shown proximal to distal). The sequence forms an inverted repeat structure. The distal half comprises a complementary sequence to the proximal half. The tip of the putative hairpin structure is indicated by an arrow. Relatively GC-rich regions within the PATRR are boxed. (B) Sequence comparisons between the proximal and distal (complementary) arms of the PATRR. The 8 bp region deleted from the short 204 bp PATRR is underlined. Six sites showing cis-morphism or polymorphism are indicated in bold (proximal positions 82, 119, 137, 155, 211 and 223). Heterozygosity for the cis-morphisms at positions 119, 137, 155 and 223 are observed in this sequence.

Figure 4. Sequence of BP regions of the PATRRs deduced from junction fragments of t(11;22) carriers. (A) Comparison of chromosome 11 PATRR sequences. Junction fragment sequences are shown from proximal on the der(11) to distal on the der(22). Sequence of a normal 445 bp PATRR is indicated at the top, followed by seven t(11;22) examples. The small deletions are symmetrically located at the center of the PATRR. (B) Comparison of chromosome 22 PATRR sequences. The junction fragment sequences are shown from proximal on the der(22) to distal on the der(11). The seven cases are the same as shown in (A). The deletions also show symmetry around the center of the PATRR.

Figure 4. Sequence of BP regions of the PATRRs deduced from junction fragments of t(11;22) carriers. (A) Comparison of chromosome 11 PATRR sequences. Junction fragment sequences are shown from proximal on the der(11) to distal on the der(22). Sequence of a normal 445 bp PATRR is indicated at the top, followed by seven t(11;22) examples. The small deletions are symmetrically located at the center of the PATRR. (B) Comparison of chromosome 22 PATRR sequences. The junction fragment sequences are shown from proximal on the der(22) to distal on the der(11). The seven cases are the same as shown in (A). The deletions also show symmetry around the center of the PATRR.

Figure 5. Variant BP of the t(11;22). (A) Sequence of the der(11) and der(22) junction fragments derived from the atypical t(11;22) (case 23). Sequences are shown from the chromosome 11 side to the chromosome 22 side. Upper case letters indicate the PATRR region and triangles indicate the BPs. The chromosome 22 side of the der(11) has a LINE1 insertion (bold) with an 8 bp target site duplication (underline). (B) BP location of der(11) and the der(22) junction fragments deduced from the normal 445 bp chromosome 11 PATRR is shown in the upper panels. Chromosome 11 is depicted by hatched boxes and chromosome 22 by shaded boxes. The BP locations in the putative hairpin structure of normal PATRRs on chromosome 11 and 22 are shown below. Variable numbers of nucleotides are deleted at the tip of the PATRR. (C) BP locations in the der(11) and the der(22) junction fragments deduced from case 23 are shown. The short 204 bp polymorphism of the PATRR on normal chromosome 11 forms an asymmetric structure. The variant chromosome 22 PATRR that includes a LINE1 insertion also forms asymmetry. Arrows indicate the deduced BPs for this rearrangement.

Figure 5. Variant BP of the t(11;22). (A) Sequence of the der(11) and der(22) junction fragments derived from the atypical t(11;22) (case 23). Sequences are shown from the chromosome 11 side to the chromosome 22 side. Upper case letters indicate the PATRR region and triangles indicate the BPs. The chromosome 22 side of the der(11) has a LINE1 insertion (bold) with an 8 bp target site duplication (underline). (B) BP location of der(11) and the der(22) junction fragments deduced from the normal 445 bp chromosome 11 PATRR is shown in the upper panels. Chromosome 11 is depicted by hatched boxes and chromosome 22 by shaded boxes. The BP locations in the putative hairpin structure of normal PATRRs on chromosome 11 and 22 are shown below. Variable numbers of nucleotides are deleted at the tip of the PATRR. (C) BP locations in the der(11) and the der(22) junction fragments deduced from case 23 are shown. The short 204 bp polymorphism of the PATRR on normal chromosome 11 forms an asymmetric structure. The variant chromosome 22 PATRR that includes a LINE1 insertion also forms asymmetry. Arrows indicate the deduced BPs for this rearrangement.

Table 1. Comparative analysis of cis-morphisms and polymorphisms within the PATRRs deduced from the sequence of junction fragments

ND, not determined; –, deleted; *, combination with other rare polymorphisms; **, 108 bp longer than that of case 1.

aCase numbers of this table correspond to those of Kurahashi et al. 2000 (3).

bThe deletion sizes were determined using the junction fragment sequence of case 1 as a standard.

cFor details see text.

Table 1. Comparative analysis of cis-morphisms and polymorphisms within the PATRRs deduced from the sequence of junction fragments

ND, not determined; –, deleted; *, combination with other rare polymorphisms; **, 108 bp longer than that of case 1.

aCase numbers of this table correspond to those of Kurahashi et al. 2000 (3).

bThe deletion sizes were determined using the junction fragment sequence of case 1 as a standard.

cFor details see text.

Table 2.

Analysis of cis-morphism/polymorphisms of the PATRR in de novo t(11;22)s in sperm from normal individualsa

Sample  Chromosome 11 PATRR Chromosome 22 PATRR  
  Position 82 Position 119 Position 137 Position 155 Position 211 Typeb Position 41 Position 72 Position 79 Typec  
SP1 der(11) AA ATT ATAAA TTTTT TA TAA  
  AA ATT ATAAA TTTTT TA TAA  
  AA ATT ATAAA TTTTT – TAA/ATT a, b Double 
  AA ATT ATAAA TTTTT TA ATT  
  AA ATT ATAAA TTTTT TA ATT  
  AA ATT AAAAA TTTAT – ATT  
 der(22) AA ATT AAAAA TTTAT – TAA  
  AA ATT AAAAA TTTAT – TAA  
  AA ATT AAAAA TTTAT TATA ATT  
  AA ATT AAAAA TTTAT – TAA  
  AA ATT AAAAA TTTAT TA ATT  
  AA ATT AAAAA TTTAT TA TAA  
SP2 der(11) AAA ATTATT AAAAA TTTAT TA TAA  
  AAA ATTATT AAAAA TTTAT – TAA/ATT G/T c, d Double 
  AAA ATTATT AAAAA TTTAT TA T/C TAA/ATT a, c or b, e Double 
  AAA ATTATT AAAAA TTTAT TA ATT  
  AAA ATTATT AAAAA TTTAT TA TAA  
  AAA ATTATT AAAAA TTTAT – TAA  
 der(22) AAA ATTATT AAAAA TTTAT – TAA  
  AAA ATTATT AAAAA TTTAT TA TAA  
  AAA ATTATT AAAAA TTTAT – ATT  
  AAA ATTATT AAAAA TTTAT – TAA  
  AAA ATTATT AAAAA TTTAT TA ATT  
  AAA ATTATT ND ND ND TAA/ATT a, b Double 
SP3 der(11) AAA ATTATT ATAAA TTTAT TA ATT  
  AAA ATTATT ATAAA TTTAT TATATTA ATT  
  AAA ATTATT ATAAA TTTAT TA – – ND  
  AAA ATTATT ATAAA TTTAT – ATT Double 
  AAA ATTATT ATAAA – – E or F – – ND  
  AAA ATTATT ATAAA TTTAT – E or F TAA  
 der(22) AAA ATTATT ATAAA TTTAT TA – – ND  
  AAA ND ND ND ND ND TAA Double 
  AAA ATTATT ATAAA TTTAT TA ATT  
  AAA ND ND ND ND ND ND ND ND Double 
  AAA ATTATT ATAAA TTTAT TATA TAA  
SP4 der(11) AAA ATTATT AAAAA TTTAT TA – – ND  
  AA ATTATT ATAAA TTTTT TA TAA  
  AAA ATTATT AAAAA TTTAT TA ATT  
  AAA ATTATT AAAAA TTTAT TA – – ND  
  AAA ATTATT AAAAA TTTAT TA ATT –  
  AA ATTATT ATAAA TTTTT – ATT –  
  AAA ATTATT AAAAA TTTAT TA ATT –  
 der(22) AA/AAA ATT/ATTATT AAAAA TTTAT TA TAA/ATT a, b Double 
  AA/AAA ND ND ND ND B, D T/C ND ND ND Double 
  AAA ATTATT AAAAA TTTAT TA TAA  
  AAA ATTATT AAAAA TTTAT – – – ND  
  AA ATT AAAAA TTTAT TA TAA  
  AA ATT AAAAA TTTAT – ATT  
  AAA ATTATT AAAAA TTTAT – ND  
  AA ATT AAAAA TTTAT TA – – ND  
Sample  Chromosome 11 PATRR Chromosome 22 PATRR  
  Position 82 Position 119 Position 137 Position 155 Position 211 Typeb Position 41 Position 72 Position 79 Typec  
SP1 der(11) AA ATT ATAAA TTTTT TA TAA  
  AA ATT ATAAA TTTTT TA TAA  
  AA ATT ATAAA TTTTT – TAA/ATT a, b Double 
  AA ATT ATAAA TTTTT TA ATT  
  AA ATT ATAAA TTTTT TA ATT  
  AA ATT AAAAA TTTAT – ATT  
 der(22) AA ATT AAAAA TTTAT – TAA  
  AA ATT AAAAA TTTAT – TAA  
  AA ATT AAAAA TTTAT TATA ATT  
  AA ATT AAAAA TTTAT – TAA  
  AA ATT AAAAA TTTAT TA ATT  
  AA ATT AAAAA TTTAT TA TAA  
SP2 der(11) AAA ATTATT AAAAA TTTAT TA TAA  
  AAA ATTATT AAAAA TTTAT – TAA/ATT G/T c, d Double 
  AAA ATTATT AAAAA TTTAT TA T/C TAA/ATT a, c or b, e Double 
  AAA ATTATT AAAAA TTTAT TA ATT  
  AAA ATTATT AAAAA TTTAT TA TAA  
  AAA ATTATT AAAAA TTTAT – TAA  
 der(22) AAA ATTATT AAAAA TTTAT – TAA  
  AAA ATTATT AAAAA TTTAT TA TAA  
  AAA ATTATT AAAAA TTTAT – ATT  
  AAA ATTATT AAAAA TTTAT – TAA  
  AAA ATTATT AAAAA TTTAT TA ATT  
  AAA ATTATT ND ND ND TAA/ATT a, b Double 
SP3 der(11) AAA ATTATT ATAAA TTTAT TA ATT  
  AAA ATTATT ATAAA TTTAT TATATTA ATT  
  AAA ATTATT ATAAA TTTAT TA – – ND  
  AAA ATTATT ATAAA TTTAT – ATT Double 
  AAA ATTATT ATAAA – – E or F – – ND  
  AAA ATTATT ATAAA TTTAT – E or F TAA  
 der(22) AAA ATTATT ATAAA TTTAT TA – – ND  
  AAA ND ND ND ND ND TAA Double 
  AAA ATTATT ATAAA TTTAT TA ATT  
  AAA ND ND ND ND ND ND ND ND Double 
  AAA ATTATT ATAAA TTTAT TATA TAA  
SP4 der(11) AAA ATTATT AAAAA TTTAT TA – – ND  
  AA ATTATT ATAAA TTTTT TA TAA  
  AAA ATTATT AAAAA TTTAT TA ATT  
  AAA ATTATT AAAAA TTTAT TA – – ND  
  AAA ATTATT AAAAA TTTAT TA ATT –  
  AA ATTATT ATAAA TTTTT – ATT –  
  AAA ATTATT AAAAA TTTAT TA ATT –  
 der(22) AA/AAA ATT/ATTATT AAAAA TTTAT TA TAA/ATT a, b Double 
  AA/AAA ND ND ND ND B, D T/C ND ND ND Double 
  AAA ATTATT AAAAA TTTAT TA TAA  
  AAA ATTATT AAAAA TTTAT – – – ND  
  AA ATT AAAAA TTTAT TA TAA  
  AA ATT AAAAA TTTAT – ATT  
  AAA ATTATT AAAAA TTTAT – ND  
  AA ATT AAAAA TTTAT TA – – ND  

Double, two or more different PCR products were generated in a single PCR reaction. Sequence could not be determined by direct sequencing. ND, not determined; –, deleted; *, other rare polymorphism.

aEach line represents analysis of a single PCR product.

bcis-Morphism/polymorphisms were arbitrarily assigned a letter (A–F) based on sequence type.

ccis-Morphism/polymorphisms were arbitrarily assigned a letter (a–d) to indicate sequence type.

References

1 Shaikh, T.H., Budarf, M.L., Celle, L., Zackai, E.H. and Emanuel, B.S. (
1999
) Clustered 11q23 and 22q11 breakpoints and 3:1 meiotic malsegregation in multiple unrelated t(11;22) families.
Am. J. Hum. Genet.
 ,
65
,
1595
–1607.
2 Kurahashi, H., Shaikh, T.H., Hu, P., Roe, B.A., Emanuel, B.S. and Budarf, M.L. (
2000
) Regions of genomic instability on 22q11 and 11q23 as the etiology for the recurrent constitutional t(11;22).
Hum. Mol. Genet.
 ,
9
,
1665
–1670.
3 Kurahashi, H., Shaikh, T.H., Zackai, E.H., Celle, L., Driscoll, D.A., Budarf, M.L. and Emanuel, B.S. (
2000
) Tightly clustered 11q23 and 22q11 breakpoints permit PCR-based detection of the recurrent constitutional t(11;22).
Am. J. Hum. Genet.
 ,
67
,
763
–768.
4 Edelmann, L., Spiteri, E., Koren, K., Pulijaal, V., Bialer, M.G., Shanske, A., Goldberg, R. and Morrow, B.E. (
2001
) AT-rich palindromes mediate the constitutional t(11;22) translocation.
Am. J. Hum. Genet.
 ,
68
,
1
–13.
5 Dunham, I., Shimizu, N., Roe, B.A., Chissoe, S., Hunt, A.R., Collins, J.E., Bruskiewich, R., Beare, D.M., Clamp, M., Smink, L.J. et al. (
1999
) The DNA sequence of human chromosome 22.
Nature
 ,
402
,
489
–495.
6 Shaikh, T.H., Kurahashi, H., Saitta, S.C., O’Hare, A.M., Hu, P., Roe, B.A., Driscoll, D.A., McDonald-McGinn, D.M., Zackai, E.H., Budarf, M.L. et al. (
2000
) Chromosome 22-specific low copy repeats and the 22q11.2 deletion syndrome: genomic organization and deletion endpoint analysis.
Hum. Mol. Genet.
 ,
9
,
489
–501.
7 Funke, B., Edelmann, L., McCain, N., Pandita, R.K., Ferreira, J., Merscher, S., Zohouri, M., Cannizzaro, L., Shanske, A. and Morrow, B.E. (
1999
) Der(22) syndrome and velo-cardio-facial syndrome/DiGeorge syndrome share a 1.5-Mb region of overlap on chromosome 22q11.
Am. J. Hum. Genet.
 ,
64
,
747
–758.
8 Shaikh, T.H., Kurahashi, H. and Emanuel, B.S. (
2001
) Evolutionarily conserved low copy repeats (LCRs) in 22q11 mediate deletions, duplications, translocations, and genomic instability: an update and literature review.
Genet. Med.
 ,
3
,
6
–13.
9 Leach, D.R.F. (
1994
) Long DNA palindromes, cruciform structures, genetic instability and secondary structure repair.
Bioessays
 ,
16
,
893
–900.
10 Doherty, J.P., Lindeman, R., Trent, R.J., Graham, M.W. and Woodcock, D.M. (
1993
) Escherichia coli host strains SURE and SRB fail to preserve a palindrome cloned in λ phage: improved alternate host strains.
Gene
 ,
124
,
29
–35.
11 Qin, Z. and Cohen, S.N. (
2000
) Long palindromes formed in Streptomyces by nonrecombinational intra-strand annealing.
Genes Dev.
 ,
14
,
1789
–1796.
12 Kurahashi, H. and Emanuel, B.S. (
2001
) Unexpectedly high rate of de novo constitutional t(11;22) translocations in sperm from normal males.
Nat. Genet.
 ,
29
,
139
–140.
13 Li, M., Budarf, M.L., Chien, P., Barnoski, B.L., Emanuel, B.S. and Driscoll, D.A. (
1995
) Clustering of DiGeorge/velocardiofacial-associated translocations suggestive of a translocation ‘hotspot’.
Am. J. Hum. Genet.
 ,
57 (suppl.)
,
A119
.
14 Kehrer-Sawatzki, H., Haussler, J., Krone, W., Bode, H., Jenne, D.E., Mehnert, K.U., Tummers, U. and Assum, G. (
1997
) The second case of a t(17;22) in a family with neurofibromatosis type 1: sequence analysis of the breakpoint regions.
Hum. Genet.
 ,
99
,
237
–247.
15 Rhodes, C.H., Call, K.M., Budarf, M.L., Barnoski, B.L., Bell, C.J., Emanuel, B.S., Bigner, S.H., Park, J.P. and Modandas, T.K. (
1997
) Molecular studies of an ependymoma-associated constitutional t(1;22)(p22;q11.2).
Cytogenet. Cell Genet.
 ,
78
,
247
–252.
16 Akgün, E., Zahn, J., Baumes, S., Brown, G., Liang, F., Romanienko, P.J., Lewis, S. and Jasin, M. (
1997
) Palindrome resolution and recombination in the mammalian germ line.
Mol. Cell. Biol.
 ,
17
,
5559
–5570.
17 Collick, A., Drew, J., Penberth, J., Bois, P., Luckett, J., Scaerou, F., Jeffreys, A. and Reik, W. (
1996
) Instability of long inverted repeats within mouse transgenes.
EMBO J.
 ,
15
,
1163
–1171.
18 Lewis, S.M. (
1999
) Palindromy is eliminated through a structure-specific recombination process in rodent cells.
Nucleic Acids Res.
 ,
27
,
2521
–2528.
19 Tapia-Paez, I., O’Brien, K.P., Kost-Alimova, M., Sahlen, S., Kedra, D., Bruder, C.E., Andersson, B., Roe, B.A., Hu, P., Imreh, S. et al. (
2000
) Fine mapping of the constitutional translocation t(11;22)(q23;q11).
Hum. Genet.
 ,
106
,
506
–516.
20 Connelly, J.C., Kirkham, L.A. and Leach, D.R. (
1998
) The SbcCD nuclease of Escherichia coli is a structural maintenance of chromosomes (SMC) family protein that cleaves hairpin DNA.
Proc. Natl Acad. Sci. USA
 ,
95
,
7969
–7974.
21 Sharples, G.J. and Leach, D.R. (
1995
) Structural and functional similarities between the SbcCD proteins of Escherichia coli and the RAD50 and MRE11 (RAD32) recombination and repair proteins of yeast.
Mol. Microbiol.
 ,
17
,
1215
–1217.
22 Khanna, K.K. and Jackson, S.P. (
2001
) DNA double-strand breaks: signaling, repair and the cancer connection.
Nat. Genet.
 ,
27
,
247
–254.
23 Richardson, C. and Jasin, M. (
2000
) Frequent chromosomal translocations induced by DNA double-strand breaks.
Nature
 ,
405
,
697
–700.
24 Brukner, I., Sanchez, R., Suck, D. and Pongor, S. (
1995
) Trinucleotide models for DNA bending propensity: comparison of models based on DNaseI digestion and nucleosome packaging data.
J. Biomol. Struct. Dyn.
 ,
13
,
309
–317.
25 Moore, J.K. and Haber, J.E. (
1996
) Capture of retrotransposon DNA at the sites of chromosomal double-strand breaks.
Nature
 ,
383
,
644
–646.