Abstract

X-linked ichthyosis (XLI) is often associated with a recurrent microdeletion at Xp22.31 due to non-allelic homologous recombination between the CRI-S232 low-copy repeat regions flanking the STS gene. The clinical features of these patients may include mental retardation (MR) and the VCX-A gene has been proposed as the candidate MR gene. Analysis of DNA from four XLI patients with MR by array-comparative genomic hybridization (array-CGH) on a 150 kb resolution X chromosome-specific array revealed a 1.5 Mb interstitial microdeletion with breakpoints in the CRI-S232 repeat sequences, each of which harbors a VCX gene. We demonstrate that the recombination sites in all four cases are situated in the 1 kb repeat unit 2 region present at the 3′ ends of the VCX-A and VCX-B genes thereby deleting VCX-A and VCX-B1 but not VCX-B and VCX-C. Array-CGH with DNA of an XLI patient with MR and an inherited t(X;Y)(p22.31;q11.2) showed an Xpter deletion of 8.0 Mb resulting in the deletion of all four VCX genes and duplication of both VCY homologs. These data confirm the role of VCX-A in the occurrence of MR in XLI patients. Moreover, we propose a VCX/Y teamwork-dependent mechanism for the incidence of mental impairment in XLI patients.

INTRODUCTION

X-linked ichthyosis (XLI) (OMIM 308100), with an estimated prevalence of one in 6000, is a genetic disorder affecting the skin and is caused by a deficit in the steroid sulfatase (STS) enzyme (reviewed in 1). About 90% of patients with ichthyosis miss the entire STS gene but the extent of the deletion can vary (24). Most of these STS deletion patients have XLI as the only clinical feature and it is believed that patients with more complex disorders, including MR, are the result of contiguous gene deletions caused by interstitial or terminal deletions (5,6). In many cases, a recurrent 1.5 Mb interstitial deletion has been identified in XLI patients with and without mental retardation (MR). It was suggested that this deletion is caused by unequal crossing-over between the highly conserved CRI-S232 repeat regions flanking the STS gene, each of which contains a VCX (Variable Charge X) gene (5,7). However, until now, the exact recombination sites have not been investigated.

The causative MR gene in XLI patients is not yet identified. The MR locus was initially mapped to a 3 Mb genomic interval between DXS31 (4.21 Mb) and STS (7.05 Mb) (3). Deletion mapping studies in patients with and without cognitive deficits proposed the VCX-A gene as the candidate MR gene at Xp22.31 (4). These authors demonstrated that, in contrast to four XLI patients with MR, the VCX-A was retained in patients with normal intelligence, whereas no correlation could be found for the other VCX genes. However, in the patients with MR, a much larger deletion was present comprising also the neuroligin 4 (NLGN4) gene (5.80 Mb), which was recently associated with autism (8) and reduced cognitive development (9).

In the present study, we fine-mapped the exact recombination site in four unrelated XLI patients with MR. In all four patients, the 1 kb repeat unit 2 (RU2) regions of VCX-A and VCX-B were identified as the recombination sites thereby deleting the VCX-A gene but not VCX-B. These data support the VCX-A as the candidate MR gene hypothesis (4). However, Xp terminal deletions including VCX-A were reported in normal intelligent male patients with t(X;Y)(p22.31;q11.2) (10,11) and breakpoints near the VCX-B gene. We next defined the breakpoints in our XLI patient with MR and t(X;Y)(p22.31;q11.2) and demonstrated an Xp terminal deletion including VCX-A, VCX-B1, VCX-B and VCX-C. Therefore, we suggest a VCX/Y gene dosage dependent mechanism for normal cognitive development rather than a single VCX MR candidate gene hypothesis, as proposed by Fukami et al. (4).

RESULTS

Array-CGH on patient samples

Four patients with ichthyosis and MR were selected for analysis by array-comparative genomic hybridization (array-CGH) on our high-resolution X-array. In all eight hybridizations (including color-flip experiments), the same nine clones at Xp22.31 showed log2 ratios of −1.45±0.50 in case the patient's material was labeled with Cy5, and 1.20 ±0.44 when the patient's DNA was labeled with Cy3. Graphs are shown for case 2 (Fig. 1A and B) and case 4 (Fig. 1C and D). The standard deviation (SD) of the ratios of the duplicates of unaffected clones as well as the overall SD of the ratios of all spots was always <0.12. The mean log2 ratios across the different hybridizations for the clones within the deletion are given in Table 1 and demonstrate the clone-dependent variability of the ratios. The most distal deleted clone was RP11-359O20 (6.30 Mb) and the most proximal one was RP11-527B14 (7.86 Mb). Both clones contain a CRI-S232 sequence, suggesting recombination between these known low-copy number repeat (LCR) regions.

Array-CGH with DNA of the t(X;Y)(p22.3;q11.2) patient revealed an Xp terminal deletion also with RP11-527B14 as the last deleted clone (Fig. 2A). The average normalized log2 ratio of deleted clones was −1.31±0.55. In the color-flip hybridization, the average normalized log2 ratio was 1.23±0.55. Hybridization of the t(X;Y) probe on the full genome array revealed a duplication of the Yq arm (Fig. 2B), starting from clone RP11-235I1 (15.2 Mb). These clones demonstrated an average normalized log2 ratio of 0.79±0.18. These data map the Yq11.2 breakpoint between RP11-292P9 (arrow in Fig. 2B) and RP11-235I1. The VCY-D gene (14.53 Mb) is located at the very end of RP11-292P9. Although no genomic clones from the 31 Mb Yq terminal part are present on the 1 Mb array, duplication of this part of the Yq arm was deduced from the presence of three signals from the subtelomeric PAR2 FISH probe (C8.2/1) in this patient (12).

Defining the recombination sites in the four microdeletion patients

The homologous CRI-S232 sequences are located at either side of the STS gene (13) and recombination between these sequences is reported to account for the majority of STS deletions (5). Each CRI-S232 contains a copy of a VCX/Y gene (7) of which the genomic structure is illustrated in Figure 3. The position of the repeat unit 1 (RU1) and RU2 are shown. The exact recombination sites in the four patients were analyzed first with primer pairs at either side of the VCX-A and VCX-B genes, as we expected that these regions were involved in non-allelic homologous recombination (NAHR). In all four patients, PCR products were obtained for the primer pairs distal to VCX-A (VCXA-dis) and proximal to VCX-B (VCXB-prox) but not for the internal primer pairs VCXA-prox and VCXB-dis (Fig. 4A). This demonstrated that recombination took place between the CRI-S232 LCRs that harbor the VCX-A and VCX-B genes. Next, we constructed primers sets at either side of the RU2 regions of VCX/Y genes. The RU2 region consists of a low-complexity repetitive sequence element (13) (Fig. 3). First, amplification just downstream of the RU2 region was performed with the primer set RU2-rev. The PCR products were directly sequenced and discrimination between the different VCX genes was based on three gene-specific nucleotides (Fig. 4B). In all patients, these nucleotides were unique for the VCX-A and VCX-C genes. This demonstrated that VCX-B was not present at this particular location and hence, the region just downstream of RU2 is derived from VCX-A. The RU1-for primers, which flank RU1, were then used to define the origin of the upstream part of the RU2 region. As these primers amplify all four VCX but not the two VCY genes, PCR products from different lengths were expected. In our patients, only two fragments were obtained. These were 163 and 463 bp for cases 1 and 4, 163 and 373 bp for case 2 and 163 and 523 bp for case 3, respectively (Fig. 4C). PCR products were sequenced and revealed that the 163 bp fragment, present in all patients, was derived from the RU1 of VCX-B with two copies of the repeat motif. The other bands of 463, 373 and 523 bp in length match with the RU1 of VCX-C with 12, 9 and 14 motif copies, respectively. Although the sequence similarity of the repeat motifs of RU1 within and between VCX gene members is extremely high (7), discrimination of the longer PCR fragments (373–523 bp) was evident from specific nucleotides outside the RU1 region. This result was confirmed by amplification with primers VCX-ex2 (Fig. 4D), which also revealed VCX-B- and VCX-C-specific sequences (unpublished data). Taken together, our results demonstrate that recombination in all four patients took place between the primer sets RU2-rev of VCX-A and RU1-for of VCX-B (Fig. 4D). Therefore, the ∼1 kb RU2 region served as the recombination site in these four XLI patients with MR. As the RU2 regions of VCX-A and VCX-B show>95% identity, share the same genomic orientation and are situated 1.5 Mb apart, they are a likely hot-spot for NAHR with removal of the intermediate region resulting in the deletion of VCX-A, HDHD1A, STS and VCX-B1 but leaving the VCX-B gene intact.

As we identified different number of RU1 repeat motifs for the VCX-C gene in our patients, we checked its variability in the population. For this, 10 control males were analyzed with the primer set RU1-for (Fig. 5). In controls 1, 2 and 9, four bands were obtained whereas all other cases had three bands. In these latter cases, one of the ‘longer’ bands was always more intense reflecting the presence of two different VCX genes with the same number of motifs. The ‘longer’ bands (283–523 bp) represent VCX genes with six to 14 repeat motifs (VCX-A, -B1 and -C). The 163 bp band corresponds to the VCX-B gene and was present in all individuals indicating that this gene always contains two repeat motifs.

Defining the recombination sites in the translocation patient

The t(X;Y)(p22.31;q11.2) patient was described previously and the breakpoints were initially mapped between DXS278 (8.06 Mb) and DXS7470 (8.30 Mb) on the X chromosome and between DYS391 (12.54 Mb) and DYS390 (15.71 Mb) on chromosome Y (12). We further investigated the Xp terminal deletion in the boy with the primer sets at either side of VCX-B (VCXB-dis and VCXB-prox) but did not detect any products indicating that the breakpoint is proximal to VCXB-prox. Next, we designed primer pairs at either side of VCX-C (VCXC-dis1 and VCXC-prox1) but again no products were obtained demonstrating that the VCX-C gene was also absent in this patient (Fig. 6A). Indeed, we confirmed the deletion of all VCX genes with the RU1-for primer set. Sequence analysis of the weak product revealed this band as the VCY-derived product with only a single repeat motif. Marker DXS7470 was positive in this patient indicating that the Xp breakpoint did occur in a 50 kb region distal to DXS7470, which leaves KAL1 intact (Fig. 6A). The Yq11.2 breakpoint was studied in the carrier mother of this patient. On the basis of our 1 Mb full genome array-CGH data, PCR primer sets were designed proximal to VCY-D and analyzed on the mother's DNA (Fig. 6B). Sequence analysis confirmed their identities. The absence of a PCR product with the VCY-prox2 primer set, but not with those of VCY-prox1 and DYS246, pinpointed the recombination site in a 70 kb region proximal to DYS246. Taken together, our data demonstrate that recombination occurred proximal to the VCX-C on the X chromosome and proximal to the VCY-D gene on the Y chromosome. Upon recombination, the Xpter part will be replaced by Yq11.2-qter. When this der(X) chromosome is transmitted to a male, all four VCX genes will be absent and both VCY genes will be present twice.

DISCUSSION

High-resolution X-array-CGH allowed us to detect an apparent similar 1.5 Mb microdeletion at Xp22.31 in four XLI patients with MR. Until now, fine-mapping of the breakpoints in XLI patients was never performed. Large-scale marker analysis in 22 Israelite and 80 Mexican XLI patients revealed the distal breakpoint of the recurrent microdeletion between DXS996 and DXS1139 and the proximal one between DXS1132 and DXS1134 (14,15). A similar result was obtained in a monozygotic male twin with XLI and MR (16) (summarized in Fig. 7A). With our array, the deletion was found to span the genomic region from RP11-359O20 to RP11-527B14, which both harbor a CRI-S232 LCR sequence each containing a member of the VCX/Y gene family. This suggests that these LCRs are involved in the NAHR process as suggested previously (4,5,7). The VCX/Y gene family consists of six known genes, four genes on Xp22.31 (Xpter–VCX-AVCX-B1VCX-BVCX-C–Xcen) and two on Yq11.2 (Ycen–VCY-DVCY-E–Yqter) (Fig. 7A and B, respectively). Originally, three neighboring genes were identified on X named VCX-2R, VCX-8R and VCX-10R, with the numbers referring to the number repeat motifs present in each gene (RU1 region) (7). Later, these genes were renamed to VCX-B, VCX-A and VCX-B1, respectively, because of repeat number polymorphism in the population although VCX-B always has two repeat motifs in RU1. The VCX-C gene was described as the fourth member of this family on Xp22.31 (4) and also contains a variable number of motifs. Very recently, the names changed again to VCX3A (VCX-A), VCX (VCX-B1), VCX2 (VCX-B) and VCX3B (VCX-C) but for the sake of uniformity in the literature, we have used the A–C numbering in this study. Both VCY genes (VCY-D and VCY-E) are present at Yq11.2 within a 100 kb region. They have identical but oppositely oriented coding regions and always contain a single 30 bp repeat in their RU1 (7). PCR and sequence analysis of the RU1 regions in our patients revealed the presence of the VCX-B (two repeats) and VCX-C (9–14 repeats) genes only (Fig. 4C), whereas in control individuals the four VCX genes were amplified (Fig. 5).

On the basis of gene-specific nucleotides, we demonstrated that the RU2 regions located at the 3′ ends of the VCX-A and VCX-B genes served as recombination sites, thereby deleting VCX-A, HDHD1A, STS and VCX-B1 but not VCX-B or VCX-C (Fig. 7A). The RU2 region consists of a variable number of repeat blocks between genes, and possibly between individuals. As the recombination occurred in all four patients, this region seems to be a hot-spot within the 10 kb CRI-S232 LCR. At least some of the recombination sites of LCR-mediated genomic disorders seem to have a positional preference (17,18).

XLI as part of a contiguous gene deletion syndrome at Xp22.31 is often associated with MR, leading to the hypothesis that an MR gene should be present within this deletion. In a genotype–phenotype correlation study aimed to map the genes for different clinical entities at Xp22.31, Ballabio et al. (3) assigned the MRX gene between marker DXS31 and the STS gene. This interval was subsequently narrowed (3,4,19,20) and in 2000, the group of Rappold proposed the VCX-A gene as the candidate MR gene based on PCR and sequence analysis of a 15 kb overlapping region in two XLI patients with normal intelligence. This region corresponds to the CRI-S232A homologous sequence in which only the VCX-A gene is located (4). The breakpoints of the microdeletions at Xp22.31 in XLI patients are most often situated near the VCX-A and VCX-B genes (summarized in Fig. 7A), which indicate recombination between directly oriented LCRs. The exact recombination sites, however, have not been studied until now. Combined data from the report of Fukami et al. (4) and this study suggest that depending on whether the VCX proximal or distal homologous sequences within the LCRs are used for recombination, the VCX-A or VCX-B gene, respectively, will be retained whereas the other one will be deleted. In addition, our data on XLI patients with MR who carry the recurrent 1.5 Mb microdeletion confirm the hypothesis that VCX-A might be the MR gene at Xp22.31. The size of the STS deletion present in the MR patients reported by Fukami et al. was much bigger (4) and includes the known MRX gene, NLGN4 (9).

However, reports on XLI patients with a t(X;Y)(p22.3;q11.2) karyotype and normal intelligence argue against this hypothesis. Although Xp;Yq rearrangements are rare, in most patients with t(X;Y)(p22.3;q11.2), the Xp terminal region is replaced by the Yqter part resulting in an Xpter deletion and a Yqter duplication in males who inherit the derX chromosome. In two reports on t(X;Y) patients with ichthyosis and normal intelligence, the Xp breakpoint is situated proximal to the VCX-A gene, which is therefore deleted (10,11). Instead, the VCX-B and VCX-C genes seem to be preserved. As both patients do not suffer from mental disability 1) the VCX-A gene might not be the MR gene, 2) a high variability of the MR spectrum with VCX-A deletions might exist or 3) other members of the VCX/Y family might compensate for its absence. In our t(X;Y) XLI patient who suffers from MR, the breakpoint is situated proximal to VCX-C thereby deleting all four VCX loci. In addition, we could demonstrate the double copy number of almost the entire Yq arm including both VCY genes.

VCX/Y proteins share a very conserved N-terminal constant domain (103 amino acid), which is highly positively charged because of many basic residues. In contrast, the C-terminal variable domain, which harbors the RU1 region, contains many acidic amino acid residues and therefore, is highly negatively charged (7). This structure indicates that the acidic variable repeat unit domain might neutralize, to some extent, the first basic domain. Clearly, this extent of neutralization will depend on the number of repeat motifs within RU1 and will, for example, be higher in VCX-A than in VCX-B. So far, the function of the VCX/Y family is not known but because of their small size and high charge, they resemble chromatin-associated proteins, which is in agreement with its presumed nuclear localization (7). It is tempting to speculate that the degree of protein activity or the binding strength of the constant basic domain to its target molecules will depend on the number of neutralizing repeat motifs. Then, the physiological outcome is the result of a combined action of all VCX/Y proteins. Instead of one gene being responsible for the MR phenotype, a dosage effect might exist for this group of genes, and the absence of one or more of these members during development might lead to cognitive deficits. This hypothesis would be an extension of the teamwork model proposed by Lahn and Page (7), in which the co-existence of VCX and VCY genes on the X and Y chromosome, respectively, is explained by the fact that various members of the VCX/Y protein family can complement each other in a particular function in spermatogenesis. In regard of our hypothesis, the clinical outcome might depend on the specific combination of preserved VCX/Y proteins. As they only differ in the number of repeat motifs in RU1, it might well be that the neutralizing capacity for the constant domain plays an important role here. In our patients with the interstitial 1.5 Mb deletion, the VCX-B, VCX-C and both VCY genes are preserved. Only the VCX-C RU1 domain has relatively high neutralizing capabilities. Instead, in the study of Fukami et al., two proteins with several repeat motifs are retained (VCX-A and -C). In the X/Y translocation patients without MR (10,11), the presence of VCX-B, VCX-C and two copies of each VCY might be sufficient whereas in our t(X/Y) patient who suffers from MR, duplication of the VCY genes might not be sufficient to compensate for the loss of all four VCX genes. The complexity might even increase because of the polymorphic number of RU1 repeat motifs in VCX-A, -B1 and -C. In view of this hypothesis, the phenotypic variability in a recently described XLI family with the recurrent microdeletion at Xp22.31 can be explained (21). These authors describe a family with several individuals affected with ichthyosis but only one was diagnosed with psychomotor delay and epilepsy. Marker analysis suggested that the VCX-A gene is deleted in the XLI/MR patient as well as his affected brother who has normal cognitive abilities (21). Finally, expression of a particular VCX/Y combination might be required only at a specific time point in the developing brain. However, the expression of these genes seems to be restricted to male germ cells (7,22), which is difficult to align with a role in cognitive development. Detailed expression analysis in specific embryonic and adult brain structures is required. Our VCX/Y teamwork hypothesis has to be tested further by careful analysis of the recombination sites in STS deletion patients and by quantitation of the repeat motifs present in the RU1 regions of the remaining VCX/Y genes. Interestingly, this gene family arose quite late in evolution because it is only found in simian primates and not in mice or lower organisms (7), which might implicate a role in cognitive development.

In conclusion, high-resolution X-array-CGH was used to delineate the borders of the recurrent microdeletions in four unrelated XLI patients with MR. Fine-mapping of the recombination sites provided information on the underlying mechanism. Additional fine-mapping studies in other XLI patients will be required to allow useful genotype–phenotype correlation studies. Moreover, this will enable us to test the teamwork-dependent mechanism that might define the mental state in XLI microdeletion patients. Although this complex mechanism might compromise on an easy way to counsel families of XLI patients, absence of several VCX proteins with a high number of neutralizing repeat motifs can be regarded as a high risk factor for cognitive impairment.

MATERIALS AND METHODS

Clinical case description of STS deletion patients

Case 1 is a 1.5-year-old boy who presents with ichthyosis in the neck region and palmoplantar keratoderma since birth. This boy shows delayed psychomotor milestones. He is not able to walk yet and only speaks a few words. His mother, who also carries the deletion, is asymptomatic. Case 2 is a 29-year-old man presenting with ichthyosis since birth. He attended a special school because of MR and hyperactive behavior. On clinical examination, he shows a relative microcephaly (third centile) for normal height and a hypoplasia of the maxilla. His asymptomatic mother also carries the deletion. Case 3 is a 6-years-old boy of Peruvian origin who presented with ichthyosis from birth. He was referred because of hyperactive behavior and psychomotor retardation. Clinical examination is normal besides the ichthyosis, which is predominantly present on the legs and soles. His psychomotor development is moderately retarded with a total IQ of 54. The mother of the boy also carries the deletion but is asymptomatic. Case 4 was seen in an institution for persons with a mental handicap, at the age of 55 years. He presented with mild to moderate MR and ichthyosis on both legs. Further clinical examination was normal. In none of the four cases, other family members were available for analysis.

The clinical study and initial molecular analysis of case 5 and his mother were reported previously (12). This patient was diagnosed with ichthyosis, developmental delay and facial dysmorphism, whereas his mother had borderline intelligence without other clinical characteristics. Both carry a translocation (X;Y)(p22.31;q11.2) with the Xp breakpoint mapped between DXS278 (7.96 Mb) and DXS7470 (8.30 Mb), and the Yq11.2 breakpoint between DYS391 (12.54 Mb) and DYS390 (15.71 Mb) (12).

Array-CGH

The X chromosome-specific array consists of 1070 genomic (BAC, PAC, cosmid and fosmid) clones from the X chromosome and 60 clones from autosomal origin. Clones were obtained from the Children's Hospital Oakland Research Institute (CHORI; http://bacpac.chori.org/home.htm) and The Sanger Institute (http://www.sanger.ac.uk/cgi-bin/). From the set of X clones, 950 have a well-defined start and stop positions on X as can be found at http://www.ensembl.org/andhttp://genome.ucsc.edu/. For most of the remaining clones, a FISH-based position was available. For the determination of the breakpoint at the Y chromosome, we have used the 1 Mb full genome array that was prepared from the 1 Mb tiling path clone set obtained from The Sanger Institute (http://www.sanger.ac.uk/cgi-bin/teams/team38/CloneRequest/CloneRequest) (23) that was previously described by us (24). BAC and PAC DNA, isolated from 1 ml bacterial cultures was amplified by two rounds of DOP–PCR using an amino-linked primer in the second PCR (25), and purified on Multiscreen purification plates (Millipore, Billericia, MA, USA). Purified amino-linked PCR products were spotted in duplicate at a concentration of 250 ng/µl on 3-D CodeLink Bioarray System slides (Amersham Biosciences, Uppsala, Sweden) with a Lucidea spotter (Amersham Biosciences). Genomic DNA was labeled with the Bioprime DNA Labeling System (Invitrogen, Carlsbad, CA, USA) using Cy3- and Cy5-labeled dCTP's (Amersham Biosciences) as described by the manufacturers. Probe concentration and labeling efficiencies were measured with the Nanodrop ND-1000 spectrophotometer (Nanodrop Technologies, Rockland, DE, USA). For each hybridization, 200 pmol of Cy5 and Cy3 probe each with a Cy-dCTP incorporation efficiency of at least 1%, was mixed with 100 µg Cot-1 DNA and except for some small modifications, probe preparation and pre-blocking of the slide was performed as described by Fiegler et al. (25). The probe mixture was dissolved in 45 µl hybridization solution and applied to a pre-hybridized slide, covered with a coverslip and hybridized for 40 h in a humid chamber saturated with 20% formamide; 2×SSC. After post hybridization washes and drying, slides were scanned with the Agilent G2565BA MicroArray Scanner System (Agilent Inc., Palo Alto, CA, USA) and the images were analyzed using the ArrayVision software (Imaging Research Inc., Ontario, Canada). Spot intensities were corrected for local background and only those spots with signal intensities of Cy5 and Cy3 at least 2-fold above background were used for further analysis. For each clone, the Cy5 to Cy3 background-corrected signal intensity ratio was calculated. Data normalization was performed over the mean of the ratios of all spots because we only performed male/male hybridizations. Thresholds for genomic gains or losses are calculated as the mean±3SD of the ratios of all clones included in the analysis. Hybridizations of samples with known genomic aberrations have been used previously to validate these cut-off criteria. Color-flip hybridization experiments were always conducted. All clones with aberrant ratios were sequence verified.

Validation by PCR and sequence analysis

Regular PCR was used to validate the extent of the deleted genomic regions. Primers were developed on the basis of the sequences of the BAC/PAC clones (NCBI) with the primer analysis tool of VectorNTi (Informax, North Bethesda, MD, USA). Subsequently, primers were subjected to BLAST analysis (http://www.ncbi.nih.gov/BLAST/) to define their potential cross-hybridization with the highly homologous members of the VCX/Y gene family. Finally, the primer sets were tested on control gDNA and the specific PCR products were directly sequenced. Primer names and sequences are shown in Table 2. PCR was performed on 100 ng purified genomic DNA with 1.25 U Taq polymerase (Amersham Biosciences) in a final volume of 50 µl. PCR conditions were as follows: 5 min at 95°C followed by 38 cycles which consist of 30 s at 95°C, 1 min at 58°C and 45 s at 72°C. Finally, the reaction was kept at 72°C for 7 min. PCR products were analyzed by agarose gel electrophoresis and visualized with EtBr.

Sequencing was performed on a ABI-PRISM 3100 capillary sequencer with the Big dye terminator v3.1 cycle sequencing kit (Applied Biosystems, Foster City, CA, USA) according to their instructions. Sequence analysis was done with the ContigExpress software of VectorNTi (Informax).

ACKNOWLEDGEMENTS

We would like to thank all the patients and their families as well as the medical staff, especially Dr M. Morren (Department Dermatology, University Hospital Leuven, Belgium) for their cooperation. We want to acknowledge Tom Bogaert from the Microarray Facility of the VIB (Belgium) for his support on the in silico construction of the X-array. We would like to thank the Mapping Core and Map Finishing groups of the Wellcome Trust Sanger Institute for initial clone supply and verification. H.V.E. is a post-doctoral researcher of the Fund for Scientific Research-Flanders, Belgium (FWO-Vlaanderen). This work was supported by a Research grant G-0229-01 of the Fund for Scientific Research-Flanders (FWO-Vlaanderen), Belgium and the European Union RTD Project no. QLRT-2001-01810.

Conflict of Interest statement. None declared.

Figure 1. Array-CGH analysis with the X-array on DNA of patients with XLI and MR identifies the recurrent 1.5 Mb deletion at Xp22.31. Log2 normalized ratios are plotted against the position on the X chromosome (in Mb) for cases 2 (A and B) and 4 (C and D). The expected ratios for unaffected genomic regions are within the normal interval (−0.36 to 0.36) whereas those for deleted regions are higher than 0.36 for control/patient hybridizations (A and B) or lower than −0.36 for patient/control hybridizations (C and D). The ratios for clones within this deletion can vary extensively (Table 1). The ‘gap’ at 60 Mb represents the centromeric region for which no clones are available. A more detailed view of the Xp22.31 regions of cases 2 and 4 are shown in (B and D), respectively. The location of the aberration on the X chromosome is indicated by the arrow (E) (taken from Ensembl).

Figure 1. Array-CGH analysis with the X-array on DNA of patients with XLI and MR identifies the recurrent 1.5 Mb deletion at Xp22.31. Log2 normalized ratios are plotted against the position on the X chromosome (in Mb) for cases 2 (A and B) and 4 (C and D). The expected ratios for unaffected genomic regions are within the normal interval (−0.36 to 0.36) whereas those for deleted regions are higher than 0.36 for control/patient hybridizations (A and B) or lower than −0.36 for patient/control hybridizations (C and D). The ratios for clones within this deletion can vary extensively (Table 1). The ‘gap’ at 60 Mb represents the centromeric region for which no clones are available. A more detailed view of the Xp22.31 regions of cases 2 and 4 are shown in (B and D), respectively. The location of the aberration on the X chromosome is indicated by the arrow (E) (taken from Ensembl).

Figure 2. Array-CGH analysis on DNA of the t(X;Y)(p22.31; q11.2) male patient with XLI and MR on the X-array (A) and the full genome 1 Mb array (B). An 8.0 Mb Xp terminal deletion was demonstrated on the X-array (A) and a duplication of almost the entire Yq arm on the 1 Mb array (B). The clones within the pseudoautosomal region 1 (PAR1) on X (0–2.2 Mb) are clearly identified because they show less affected log2 ratios (−0.4 to −1.0). Clone RP11-292P12 (indicated by the arrow) which contains the VCY-D gene, has a ‘normal’ ratio suggesting that at least most of this clone is not duplicated.

Figure 2. Array-CGH analysis on DNA of the t(X;Y)(p22.31; q11.2) male patient with XLI and MR on the X-array (A) and the full genome 1 Mb array (B). An 8.0 Mb Xp terminal deletion was demonstrated on the X-array (A) and a duplication of almost the entire Yq arm on the 1 Mb array (B). The clones within the pseudoautosomal region 1 (PAR1) on X (0–2.2 Mb) are clearly identified because they show less affected log2 ratios (−0.4 to −1.0). Clone RP11-292P12 (indicated by the arrow) which contains the VCY-D gene, has a ‘normal’ ratio suggesting that at least most of this clone is not duplicated.

Figure 3. Schematic representation of the VCX/Y gene structure. The coding region is located in two exons (exons 2 and 3) and contains the repeat unit 1 (RU1) region, which consists of a variable number of repeat motifs between and/or within the different VCX/Y genes. The repeat unit 2 (RU2) region is located at the 3′ end of the gene and is ∼1 kb in size.

Figure 3. Schematic representation of the VCX/Y gene structure. The coding region is located in two exons (exons 2 and 3) and contains the repeat unit 1 (RU1) region, which consists of a variable number of repeat motifs between and/or within the different VCX/Y genes. The repeat unit 2 (RU2) region is located at the 3′ end of the gene and is ∼1 kb in size.

Figure 4. Determination of the exact recombination sites in the four XLI patients with MR. (A) PCR analysis demonstrating the deletion of the region in between VCX-A and VCX-B with preservation of the outer sequences. Primer pairs are localized ∼5 kb from the VCX gene. (B) Nucleotides at VCX gene-specific positions, based on the 144 bp RU2-rev-derived PCR product, identified by sequencing of these products that are located just distal to the RU2 regions of the VCX genes. Only nucleotides for VCX-A and -C were found here. (C) PCR with the RU1-for primers and sequence analysis of the RU1 regions revealed the presence of the VCX-B (163 bp product in all four) and VCX-C (463, 373, 523 and 463 bp products in cases 1, 2, 3 and 4, respectively). The exact lengths of the PCR products were based on their sequenced products. (D) Schematic overview of the common recombination site in all four patients.

Figure 4. Determination of the exact recombination sites in the four XLI patients with MR. (A) PCR analysis demonstrating the deletion of the region in between VCX-A and VCX-B with preservation of the outer sequences. Primer pairs are localized ∼5 kb from the VCX gene. (B) Nucleotides at VCX gene-specific positions, based on the 144 bp RU2-rev-derived PCR product, identified by sequencing of these products that are located just distal to the RU2 regions of the VCX genes. Only nucleotides for VCX-A and -C were found here. (C) PCR with the RU1-for primers and sequence analysis of the RU1 regions revealed the presence of the VCX-B (163 bp product in all four) and VCX-C (463, 373, 523 and 463 bp products in cases 1, 2, 3 and 4, respectively). The exact lengths of the PCR products were based on their sequenced products. (D) Schematic overview of the common recombination site in all four patients.

Figure 5. Amplification products of the different RU1 regions of the four VCX genes in 10 control males. In all individuals, the VCX-B-specific 163 bp PCR product is present. The other ‘longer’ products might differ in size between the different persons. Taking into account the intensity of each band, a total of three ‘longer’ bands are expected indicating that four different VCX genes are present in all controls.

Figure 5. Amplification products of the different RU1 regions of the four VCX genes in 10 control males. In all individuals, the VCX-B-specific 163 bp PCR product is present. The other ‘longer’ products might differ in size between the different persons. Taking into account the intensity of each band, a total of three ‘longer’ bands are expected indicating that four different VCX genes are present in all controls.

Figure 6. Breakpoint mapping by PCR with region-specific primers in the t(X;Y)(p22.31;q11.2) male patient and his carrier mother. (A) Schematic representation of part of the Xp22.31 region with the positions of the genes and primer sets indicated. PCR products, shown in (a1 and a2), were analyzed on agarose gel and visualized with EtBr staining. The breakpoint (BKPNT) is situated in a 50 kb region between VCX-C and KAL1. (B) Schematic representation of Yq11.2 with the positions of the genes and PCR primer sets indicated. The PCR products are shown in (b1). The breakpoint (BKPNT) is situated in a 70 kb interval proximal to VCY-D and within RP11-292P12.

Figure 6. Breakpoint mapping by PCR with region-specific primers in the t(X;Y)(p22.31;q11.2) male patient and his carrier mother. (A) Schematic representation of part of the Xp22.31 region with the positions of the genes and primer sets indicated. PCR products, shown in (a1 and a2), were analyzed on agarose gel and visualized with EtBr staining. The breakpoint (BKPNT) is situated in a 50 kb region between VCX-C and KAL1. (B) Schematic representation of Yq11.2 with the positions of the genes and PCR primer sets indicated. The PCR products are shown in (b1). The breakpoint (BKPNT) is situated in a 70 kb interval proximal to VCY-D and within RP11-292P12.

Figure 7. Schematic representation of the Xp22.31 (A) and Yq11.2 (B) genomic regions. The position and orientation of the VCX genes as well as the DXS markers are indicated (exact positions are shown in Mb, between brackets). (A) The extents of the most common deletions identified in XLI patients are indicated with gray bars, whereas the dotted lines represent the breakpoint regions. The percentage of patients investigated with the respective deletion as well as the mental status of these patients, if known, is indicated on the right. The MRX(S) gene NLGN4 is located just distal to VCX-A. (B) The oppositely oriented VCY genes at Yq11.2 are only 70 kb apart.

Figure 7. Schematic representation of the Xp22.31 (A) and Yq11.2 (B) genomic regions. The position and orientation of the VCX genes as well as the DXS markers are indicated (exact positions are shown in Mb, between brackets). (A) The extents of the most common deletions identified in XLI patients are indicated with gray bars, whereas the dotted lines represent the breakpoint regions. The percentage of patients investigated with the respective deletion as well as the mental status of these patients, if known, is indicated on the right. The MRX(S) gene NLGN4 is located just distal to VCX-A. (B) The oppositely oriented VCY genes at Yq11.2 are only 70 kb apart.

Table 1.

Mean log2 ratios for the clones within and around the recurrent 1.5 Mb deletion in the four XLI patients with MR

  Reference/patient Patient/reference 
Mb Clone Mean SD Mean SD 
5.99 RP11-44F2 0.03 0.04 0.05 0.03 
6.13 RP11-459A10 0.05 0.04 0.01 0.01 
6.30 RP11-359O20 1.22 0.34 −1.04 0.38 
6.39 RP11-294K6 1.14 0.38 −1.12 0.42 
6.42 RP11-1M18 1.48 0.38 −1.86 0.56 
6.81 RP11-483M24 1.27 0.29 −1.57 0.25 
6.81 RP11-483M24 1.47 0.05 −1.39 0.40 
7.31 RP11-323F16 1.08 0.20 −1.22 0.28 
7.31 RP11-323F16 1.36 0.35 −1.24 0.63 
7.48 RP11-143E20 0.38 0.07 −0.62 0.13 
7.86 RP11-527B14 0.61 0.06 −0.73 0.20 
8.14 RP13-318B19 0.13 0.16 0.08 0.10 
8.90 RP11-430A19 0.05 0.03 −0.02 0.04 
  Reference/patient Patient/reference 
Mb Clone Mean SD Mean SD 
5.99 RP11-44F2 0.03 0.04 0.05 0.03 
6.13 RP11-459A10 0.05 0.04 0.01 0.01 
6.30 RP11-359O20 1.22 0.34 −1.04 0.38 
6.39 RP11-294K6 1.14 0.38 −1.12 0.42 
6.42 RP11-1M18 1.48 0.38 −1.86 0.56 
6.81 RP11-483M24 1.27 0.29 −1.57 0.25 
6.81 RP11-483M24 1.47 0.05 −1.39 0.40 
7.31 RP11-323F16 1.08 0.20 −1.22 0.28 
7.31 RP11-323F16 1.36 0.35 −1.24 0.63 
7.48 RP11-143E20 0.38 0.07 −0.62 0.13 
7.86 RP11-527B14 0.61 0.06 −0.73 0.20 
8.14 RP13-318B19 0.13 0.16 0.08 0.10 
8.90 RP11-430A19 0.05 0.03 −0.02 0.04 

Mb refers to the position on the X chromosome. SD for the four data points in both hybridization sets (reference DNA labeled with Cy5 versus patient DNA labeled with Cy3, and vice versa) are given next to the mean value. The deleted 1.5 Mb region is in italics.

Table 2.

PCR primer sequences used in this study

Primer Forward (5′–3′) Reverse (5′–3′) Amplicon length (bp) 
VCXB-prox 
CAG CTC ATT TTT GAG GGTG GGA T
 
CCA TTA GCC ATC ACT TTC ACT C
 
352 
VCXA-dis 
CAA GAT GTG TCC AGA CAT TCC C
 
GCC TGG ATA CTC CCA AAA TTC TCA T
 
372 
VCXA-prox 
AAG AAA AGA TCC TCC CAG TTC AC
 
CGG TCC TCT TCA TAT AGC AAA GT
 
313 
VCXB-dis 
TTG CCA AGA TCA ATC ATT CAG C
 
GAG AGC TCA AGC AAC TCA GAA TA
 
403 
RU2-rev 
TTG TAG GGA GGG GAC GCT GAA AGG
 
GAT GCA ATC TTG CTA GCA CAT ACA C
 
144 
RU1-for 
GGA GGA GCC AGT GAG CGA G
 
CGG CTG GGC CTG AAC TTA GTC
 
Variable 
VCXC-dis1 
TTA CCA TAT CCA CAG TAC GTA GTA
 
ACA GTA CTA CTT CTT ATG TCT CT
 
187 
VCXC-prox1 
AGT ACA AGC CCA GAC TTC AGC AAT
 
CAT GGT GAC TTG ATA GTG CCC AG
 
213 
VCY-prox1 
CCA AAG GTA GCC CTG GTC ACA CA
 
CCC CAT GTG GAA TCA CCA GAT A
 
325 
VCY-prox2 
TCA CAC GTC CAT GTA TCT CAT TGT G
 
GGT GAG AAT GAG AAA CAG ATG TGG
 
236 
Primer Forward (5′–3′) Reverse (5′–3′) Amplicon length (bp) 
VCXB-prox 
CAG CTC ATT TTT GAG GGTG GGA T
 
CCA TTA GCC ATC ACT TTC ACT C
 
352 
VCXA-dis 
CAA GAT GTG TCC AGA CAT TCC C
 
GCC TGG ATA CTC CCA AAA TTC TCA T
 
372 
VCXA-prox 
AAG AAA AGA TCC TCC CAG TTC AC
 
CGG TCC TCT TCA TAT AGC AAA GT
 
313 
VCXB-dis 
TTG CCA AGA TCA ATC ATT CAG C
 
GAG AGC TCA AGC AAC TCA GAA TA
 
403 
RU2-rev 
TTG TAG GGA GGG GAC GCT GAA AGG
 
GAT GCA ATC TTG CTA GCA CAT ACA C
 
144 
RU1-for 
GGA GGA GCC AGT GAG CGA G
 
CGG CTG GGC CTG AAC TTA GTC
 
Variable 
VCXC-dis1 
TTA CCA TAT CCA CAG TAC GTA GTA
 
ACA GTA CTA CTT CTT ATG TCT CT
 
187 
VCXC-prox1 
AGT ACA AGC CCA GAC TTC AGC AAT
 
CAT GGT GAC TTG ATA GTG CCC AG
 
213 
VCY-prox1 
CCA AAG GTA GCC CTG GTC ACA CA
 
CCC CAT GTG GAA TCA CCA GAT A
 
325 
VCY-prox2 
TCA CAC GTC CAT GTA TCT CAT TGT G
 
GGT GAG AAT GAG AAA CAG ATG TGG
 
236 

References

1
Hernandez-Martin, A., Gonzalez-Sarmiento, R. and De Unamuno, P. (
1999
) X-linked ichthyosis: an update.
Br. J. Dermatol.
 ,
141
,
617
–627.
2
Newman, R.S., Affara, N.A., Yates, J.R., Mitchell, M. and Ferguson-Smith, M.A. (
1990
) Physical mapping of deletion breakpoints in patients with X-linked ichthyosis: evidence for clustering of distal and proximal breakpoints.
Proc. R. Soc. Lond. B Biol. Sci.
 ,
242
,
231
–239.
3
Ballabio, A., Bardoni, B., Carrozzo, R., Andria, G., Bick, D., Campbell, L., Hamel, B., Ferguson-Smith, M.A., Gimelli, G., Fraccaro, M. et al. (
1989
) Contiguous gene syndromes due to deletions in the distal short arm of the human X chromosome.
Proc. Natl Acad. Sci. USA
 ,
86
,
10001
–10005.
4
Fukami, M., Kirsch, S., Schiller, S., Richter, A., Benes, V., Franco, B., Muroya, K., Rao, E., Merker, S., Niesler, B. et al. (
2000
) A member of a gene family on Xp22.3, VCX-A, is deleted in patients with X-linked nonspecific mental retardation.
Am. J. Hum. Genet.
 ,
67
,
563
–573.
5
Yen, P.H., Li, X.M., Tsai, S.P., Johnson, C., Mohandas, T. and Shapiro, L.J. (
1990
) Frequent deletions of the human X chromosome distal short arm result from recombination between low copy repetitive elements.
Cell
 ,
61
,
603
–610.
6
Yen, P.H., Tsai, S.P., Wenger, S.L., Steele, M.W., Mohandas, T.K. and Shapiro, L.J. (
1991
) X/Y translocations resulting from recombination between homologous sequences on Xp and Yq.
Proc. Natl Acad. Sci. USA
 ,
88
,
8944
–8948.
7
Lahn, B.T. and Page, D.C. (
2000
) A human sex-chromosomal gene family expressed in male germ cells and encoding variably charged proteins.
Hum. Mol. Genet.
 ,
9
,
311
–319.
8
Jamain, S., Quach, H., Betancur, C., Rastam, M., Colineaux, C., Gillberg, I.C., Soderstrom, H., Giros, B., Leboyer, M., Gillberg, C. et al. (
2003
) Mutations of the X-linked genes encoding neuroligins NLGN3 and NLGN4 are associated with autism.
Nat. Genet.
 ,
34
,
27
–29.
9
Laumonnier, F., Bonnet-Brilhault, F., Gomot, M., Blanc, R., David, A., Moizard, M.P., Raynaud, M., Ronce, N., Lemonnier, E., Calvas, P. et al. (
2004
) X-linked mental retardation and autism are associated with a mutation in the NLGN4 gene, a member of the neuroligin family.
Am. J. Hum. Genet.
 ,
74
,
552
–557.
10
Tobias, E.S., Bryce, G., Farmer, G., Barton, J., Colgan, J., Morrison, N., Cooke, A. and Tolmie, J.L. (
2001
) Absence of learning difficulties in a hyperactive boy with a terminal Xp deletion encompassing the MRX49 locus.
J. Med. Genet.
 ,
38
,
466
–470.
11
Doherty, M.J., Glas, I.A., Bennet, C.L., Cotter, P.D., Watson, N.F., Mitchell, A.L., Bird, T.D. and Farrell, D.F. (
2003
) An Xp; Yq translocation causing a novel contiguous gene syndrome in brothers with generalized epilepsy, ichthyosis, and attention deficits.
Epilepsia
 ,
44
,
1529
–1535.
12
Frints, S.G., Fryns, J., Lagae, L., Syrrou, M., Marynen, P. and Devriendt, K. (
2001
) Xp22.3; Yq11.2 chromosome translocation and its clinical manifestations.
Ann. Genet.
 ,
44
,
71
–76.
13
Knowlton, R.G., Nelson, C.A., Brown, V.A., Page, D.C. and Donis-Keller, H. (
1989
) An extremely polymorphic locus on the short arm of the human X chromosome with homology to the long arm of the Y chromosome.
Nucleic Acids Res.
 ,
17
,
423
–437.
14
Aviram-Goldring, A., Goldman, B., Netanelov-Shapira, I., Chen-Shtoyerman, R., Zvulunov, A., Tal, O., Ilan, T. and Peleg, L. (
2000
) Deletion patterns of the STS gene and flanking sequences in Israeli X-linked ichthyosis patients and carriers: analysis by polymerase chain reaction and fluorescence in situ hybridization techniques.
Int. J. Dermatol.
 ,
39
,
182
–187.
15
Jimenez Vaca, A.L., Valdes-Flores, M., Rivera-Vega, M.R., Gonzalez-Huerta, L.M., Kofman-Alfaro, S.H. and Cuevas-Covarrubias, S.A. (
2001
) Deletion pattern of the STS gene in X-linked ichthyosis in a Mexican population.
Mol. Med.
 ,
7
,
845
–849.
16
Gohlke, B.C., Haug, K., Fukami, M., Friedl, W., Noeker, M., Rappold, G.A. and Haverkamp, F. (
2000
) Interstitial deletion in Xp22.3 is associated with X linked ichthyosis, mental retardation, and epilepsy.
J. Med. Genet.
 ,
37
,
600
–602.
17
Bi, W., Park, S.S., Shaw, C.J., Withers, M.A., Patel, P.I. and Lupski, J.R. (
2003
) Reciprocal crossovers and a positional preference for strand exchange in recombination events resulting in deletion or duplication of chromosome 17p11.2.
Am. J. Hum. Genet.
 ,
73
,
1302
–1315.
18
Bayes, M., Magano, L.F., Rivera, N., Flores, R. and Perez Jurado, L.A. (
2003
) Mutational mechanisms of Williams–Beuren syndrome deletions.
Am. J. Hum. Genet.
 ,
73
,
131
–151.
19
Muroya, K., Ogata, T., Matsuo, N., Nagai, T., Franco, B., Ballabio, A., Rappold, G., Sakura, N. and Fukushima, Y. (
1996
) Mental retardation in a boy with an interstitial deletion at Xp22.3 involving STS, KAL1, and OA1: implication for the MRX locus.
Am. J. Med. Genet.
 ,
64
,
583
–587.
20
Weissortel, R., Strom, T.M., Dorr, H.G., Rauch, A. and Meitinger, T. (
1998
) Analysis of an interstitial deletion in a patient with Kallmann syndrome, X-linked ichthyosis and mental retardation.
Clin. Genet.
 ,
54
,
45
–51.
21
Lesca, G., Sinilnikova, O., Theuil, G., Blanc, J., Edery, P. and Till, M. (
2005
) Xp22.3 microdeletion including VCX-A and VCX-B1 genes in an X-linked ichthyosis family: no difference in deletion size for patients with and without mental retardation.
Clin. Genet.
 ,
67
,
367
–368.
22
Zou, S.W., Zhang, J.C., Zhang, X.D., Miao, S.Y., Zong, S.D., Sheng, Q. and Wang, L.F. (
2003
) Expression and localization of VCX/Y proteins and their possible involvement in regulation of ribosome assembly during spermatogenesis.
Cell Res.
 ,
13
,
171
–177.
23
Shaw-Smith, C., Redon, R., Rickman, L., Rio, M., Willatt, L., Fiegler, H., Firth, H., Sanlaville, D., Winter, R., Colleaux, L. et al. (
2004
) Microarray based comparative genomic hybridization (array-CGH) detects submicroscopic chromosomal deletions and duplications in patients with learning disability/mental retardation and dysmorphic features.
J. Med. Genet.
 ,
41
,
241
–248.
24
Vermeesch, J.R., Melotte, C., Froyen, G., Van Vooren S., Dutta, B., Maas, N., Vermeulen, S., Menten, B., Speleman, F., De Moor B. et al. (
2005
) Molecular karyotyping: array CGH quality criteria for constitutional genetic diagnosis.
J. Histochem. Cytochem.
 ,
53
,
413
–422.
25
Fiegler, H., Carr, P., Douglas, E.J., Burford, D.C., Hunt, S., Scott, C.E., Smith, J., Vetrie, D., Gorman, P., Tomlinson, I.P. et al. (
2003
) DNA microarrays for comparative genomic hybridization based on DOP–PCR amplification of BAC and PAC clones.
Genes Chromosomes Cancer
 ,
36
,
361
–374.