Chromosomal stabilisation by a subtelomeric rearrangement involving two closely related Alu elements
Chromosomal stabilisation by a subtelomeric rearrangement involving two closely related Alu elementsJ. Flint, J. Rochette1, C. F. Craddock, C. Dodé2, B. Vignes3, S. W. Horsley, L. Kearney, V. J. Buckle, H. Ayyub and D. R. Higgs*
MRC Molecular Haematology Unit, Institute of Molecular Medicine, John Radcliffe Hospital, Headington, Oxford, OX3 9DU, UK, 1Pediatrie I et Laboratoire de Genetique, Chu Amiens - 80054, France, 2CNRS URA 1968, Institut Pasteur, Paris, France and 3Hopital de Montmorency, 95160, France
Received April 16, 1996;Revised and Accepted May 15, 1996
We have characterised a subtelomeric rearrangement involving the short arm of chromosome 16 that gives rise to [alpha]-thalassaemia by deleting the major, remote regulatory element controlling [alpha]-globin expression. The chromosomal breakpoint lies in an Alu family repeat located only ~105 kb from the 16p subtelomeric region. The broken chromosome has been stabilised with a newly positioned telomere acquired by recombination between this 16p Alu element and a closely related subtelomeric Alu element of the Sx subfamily. It seems most likely that this abnormal chromosome has been rescued by the mechanism of telomere capture which may reflect a more general process by which subtelomeric sequences are normally dispersed between chromosomal ends.
Telomeres are the specialised protein/DNA structures that `cap' the ends of linear eukaryotic chromosomes, preventing end to end fusions and exonucleolytic degradation. In addition they are required for complete replication of DNA at the end of the chromosome and are thought to play an important role in cell division. Human telomeric DNA comprises 2-20 kb of simple repeat sequence (TTAGGG)n orientated 5' -> 3' towards the end of the chromosome. Immediately subtelomeric lie complex families of repetitive DNA which may extend for several hundred kilobases (kb) before merging with the unique, chromosome-specific sequences. Currently all known telomere functions can be attributed to the simple telomeric repeats and their associated proteins; the role, if any, of subtelomeric repeats is not known (reviewed in 1 -3 ).
Variable amounts of a telomere (including telomere and subtelomeric repeats) may be lost as a result of chromosome breakage. Such broken chromosomes are unstable unless they acquire a pre-existing telomere by recombination [so-called telomere capture (4 )] or are repaired by the direct addition of simple telomeric repeats to the free end of the chromosome (so-called chromosomal healing; reviewed in 5 ). While these processes have been well documented in a wide variety of species, relatively little is known about such events in higher eukaryotes, including man, in which telomere loss due to chromosome breakage may be a relatively common mechanism underlying genetic disease.
Rearrangements involving the tip of the short arm of chromosome 16 (16p13.3) frequently delete the duplicated [alpha]-globin genes ([alpha]2 and [alpha]1) which lie a short but variable distance (170-430 kb) from the telomere (6 ,7 ). Occasionally such rearrangements leave the [alpha] genes intact but down regulate their expression by removing a critical cis-acting regulatory element (HS-40) which lies 40 kb upstream of the embryonic [zeta] gene (Fig. 1 ) (8 ,9 ). Therefore attention may be drawn to subtle, telomeric rearrangements involving 16p13.3 via the associated haematological phenotype of [alpha]-thalassaemia (6 ).
Normal individuals have two [alpha] genes on each chromosome 16 with the genotype [alpha][alpha]/[alpha][alpha]. Alpha-thalassaemia, which is common throughout all tropical and subtropical regions, most frequently results from the deletion of one (-[alpha]) or both (- -)[alpha] genes from chromosome 16. Individuals who inherit three (-[alpha]/[alpha][alpha]) or two (- -/[alpha][alpha] or -[alpha]/-[alpha]) [alpha] genes have [alpha]-thalassaemia trait. Those who inherit only one functional [alpha] gene (- -/-[alpha]) have HbH disease, a moderately severe form of haemolytic anaemia associated with tetramers of [beta] chains ([beta]4 called HbH) (reviewed in 6 ).
During a survey to determine the molecular basis for [alpha]-thalassaemia we identified an affected family from Algeria (Table 1 ). No family member has any phenotypic abnormalities other than [alpha]-thalassaemia. Of five siblings, one (MB) has HbH disease. Mapping the genomic DNA showed that rather than having one [alpha] gene, as expected, she had inherited a single copy of the [alpha] gene on each chromosome 16. This suggested the presence of an additional mutation down regulating [alpha]-globin expression on one of these chromosomes.
Family studies implicated the maternal chromosome; although both parents (MY and MM) have the genotype -[alpha]/[alpha][alpha], the more severely reduced haematological indices of the mother (MM) are consistent with the presence of two rather than three functional genes (6 ). Consequently, the maternal chromosome (subsequently denoted -[alpha]MB) inherited by MB and MF was examined in further detail.
Genomic restriction mapping using a variety of digests and previously described probes (data not shown) showed that the [alpha]-globin gene map derived from this chromosome was typical of the common, well characterised -[alpha]3.7 type 1 allele in which the remaining [alpha] gene is usually fully functional. DNA sequence analysis of the [alpha] gene from -202 to +883 nt with respect to the mRNA cap site was identical to the previously published sequence (12 ; revised in 13 ) demonstrating that any secondary mutation did not lie in the [alpha] gene itself.
Analysis of 5'HVR (14 ), a polymorphic region located 70 kb upstream of the [alpha] complex (Fig. 1 ), usually demonstrates two alleles in normal individuals. MM has a single copy of this region indicating that her genotype is CC or C- (Table 2 ) and data not shown. Since neither MB (-[alpha]MB/-[alpha]) nor MF (-[alpha]MB/[alpha][alpha]) inherited a C allele, this region must be deleted from the -[alpha]MB chromosome. To investigate this further, metaphase chromosome spreads derived from an EBV-transformed B-lymphocyte cell line from MB were analysed by fluorescence in situ hybridisation (FISH) using the cosmid probes cGG1 (15 ), cGG4 (15 ) and cRA36 (16 ) (Figs 1 and 2 ). With cGG1, telomeric signals corresponding to both chromatids from each copy of chromosome 16 were seen (Fig. 2 a). With cGG4 and cRA36 (Fig. 2 b), signals were only detected from one copy of chromosome 16 suggesting that DNA spanning co-ordinates (kb) -29 to -89 (with respect of the [zeta] mRNA cap site) of the 16p subtelomeric region, including the [alpha]-globin regulatory element (HS-40), was deleted from the -[alpha]MB chromosome (see Fig. 1 ).
To identify the breakpoint precisely, genomic DNA from MB was analysed with a series of probes lying between the [alpha]-globin cluster (cGG1) and the more telomeric region contained within cGG4, which was known to be missing. Using probes centromeric to L1.1 (Fig. 1 ) in conventional Southern blot analysis no differences were seen between MB and normal DNA but, using L1.1, breakpoint fragments were seen in MB. Restriction mapping placed the chromosomal breakpoint between an Asp718 site at co-ordinate -15 and an EcoRI site at co-ordinate -17 (Fig. 4 ).
A series of experiments were undertaken to determine the chromosomal origin of the novel sequence lying beyond the -[alpha]MB breakpoint. First, by `reverse blotting', a 400 bp HindII fragment containing the telomeric end of the plasmid pMel10 was shown to contain low repetitive or unique sequence DNA. When used as a probe in Southern blots, with competitor DNA it identified the same BamHI-breakpoint fragment as L1.1 but also identified novel, single EcoRI, HindIII and SacI fragments that were not seen with L1.1: these fragments were the same size in both patient and control DNA (data not shown).
To identify the chromosomal origin of the novel sequence lying beyond the -[alpha]MB breakpoint we used a PCR based strategy to screen DNA from a panel of somatic cell hybrids containing different human chromosome complements. Having sequenced the 400 bp HindII fragment (in pMel 11), oligonucleotides were synthesised and conditions established to amplify a 220 bp fragment in genomic DNA. Using these primers, genomic DNA from cell lines containing only chromosomes 10, 18 or 22 gave a PCR product (Table 3 ).
Initially, these results suggested that the sequence lying beyond the -[alpha]MB breakpoint did not originate from chromosome 16. However, it has been previously established that there are at least four allelic variants (A, B, C or D) of the 16p telomere: sequence divergence between these alleles begins around co-ordinate -119 (see Fig. 1 ). It was therefore possible that the pMel 11 clone derived from a 16p allele not present in the hybrid panel that was initially analysed. This hypothesis was tested by screening interspecific hybrids containing each of the four known 16p alleles and demonstrated a PCR product in DNA from the B, C and D alleles but not allele A. The novel sequence must originate from the subtelomeric sequence since this is the only region to differ between each copy of chromosome 16 in these interspecific hybrids.
These experiments suggested the following explanations for the origin of the abnormal -[alpha]MB chromosome: (i) it is the product of a terminal translocation between 16p and chromosome 10, 18 or 22; (ii) it is the product of a terminal translocation involving at least one of the longer 16p alleles (B, C or D); (iii) it is the result of a subtelomeric, interstitial deletion from one of these long 16p alleles. We attempted to distinguish between these possibilities by sequencing the PCR products from each chromosome and the three 16p alleles (Fig. 6 ). No differences were found between the sequences of the B, C or D 16p alleles and the pMel HII clone, but a number of single base pair differences distinguish chromosomes 16, 10, 18 and 22, demonstrating that the novel sequence originates from one of the long 16p alleles B, C or D. Based on the size of the novel MluI fragment (300 kb) and the minimal estimated size of the deletion (~150 kb) it appears most likely that the broken chromosome recombined with a C-type allele. This hypothesis is strengthened by the observation that on PFGE, using DNA from an interspecific hybrid containing the -[alpha]MB chromosome, the 300 kb breakpoint fragment was identified by the subtelomeric probe TelBam3.4, characteristic of the A and C 16p telomeric regions (see legend to Fig. 1 for further explanation of 16p telomere polymorphism). Alleles D or B joined to the breakpoint of the chromosome by their subtelomeric regions would have produced a smaller MluI fragment that hybridised to the subtelomeric probe TelBam11 (7 ).
We have identified a novel telomeric rearrangement involving the most distal portion of the short arm of chromosome 16 by its deleterious effect on expression of the [alpha]-globin genes. The newly acquired, moderately repetitive sequence lying beyond the breakpoint of the abnormal chromosome was shown to be present in three non-homologous chromosomes (10, 18 and 22) and in the subtelomeric region of the three previously described 16p alleles (B, C and D) derived from normal individuals. The abnormal chromosome may therefore have arisen by either a subtelomeric interstitial deletion or translocation; the sequence data could not distinguish between these possibilities but demonstrated that the recombination event involved one of the 16p alleles rather than the other non-homologous chromosomes. The size and pattern of hybridisation (to TelBam3.4) of the novel telomeric fragment suggests that this was a C-type allele. Whatever the mechanism, the acquisition of a newly positioned telomere successfully stabilised this chromosome which appears to have been transmitted normally within family M.
Haematological studies and haemoglobin analyses were carried out as previously described (20 ). Globin chain synthesis was performed as described by Weatherall and Clegg (21 ) although the globin chains were separated by reverse phase high performance liquid chromatography using a semipreparative large pore Vydac C4 column.
Genomic DNA was purified by phenol chloroform extraction or by preparation in 0.5% agarose plugs (10 ,22 ). Hybrids representing 16p alleles A-28/4, B-Hy145.M, C-3-BH1E and D-PK10-1 J545 were used for PFGE analysis. Southern blotting using the probes NFG400, RA2.2, 5'HVR, RA1.4, L4, L2, L1, L1.1, L0, IZHVR, [alpha] globin (HBA) and 3'HVR was as previously described (see 10 and references therein). Pulsed field gel electrophoresis was performed using the following conditions in a Biorad Chef gel apparatus; initial pulse time 20 s, final pulse time 40 s at 200 V for 24 h.
Following electrophoresis gels were Southern blotted as described in Sambrook et al. (22 ).
Genomic DNA prepared from peripheral blood lymphocytes was digested to completion with BamHI, phenol extracted, precipitated and resuspended in water. Approximately 1 [mu]g of the DNA was ligated to a phage vector (EMBL3) that had been BamHI digested and dephosphorylated (purchased from Promega). After packaging of the ligation mix and transduction, approximately 100 000 plaques were screened with the probe L1.1. One positive was identified. This bacteriophage was amplified and DNA extracted by standard protocols (22 ).
Digestion of the bacteriophage with BamHI released the insert which was gel purified and subcloned into BamHI digested and dephosphorylated plasmid vector PUC18. Mapping of the breakpoint was carried out in this plasmid (designated pMel10). Sequence across the breakpoint was obtained by digesting the plasmid with SstI, re-ligating and using a universal primer and dideoxy sequencing reagents (USB). A similar strategy was used to obtain the normal sequence from the plasmid pBEL14. Sequence across the 400 bp HindII fragment was obtained after subcloning the HindII fragment of pMel10 into SmaI cut and dephosphorylated PUC18. Forward and reverse universal primers were used for dideoxy sequencing this plasmid.
PCR amplification of the 220 bp fragment that was used to compare the subtelomeric alleles was performed with the following primers (HII, 160F) 5'-GCA GTG CCA CAA TCC TAG CTC TTC-3' end (HII 378R) 5'-GAG TAT TTA TGC CTG ATT CAT GGC-3' in a buffer (KCI Boehringer buffer) containing 1 mM MgCl2 using the following conditions: denature 95oC for 1 min, anneal 55oC for 1 min, extend 72oC for 1 min (30 cycles).
This work was supported by the Medical Research Council and the Wellcome Trust. We are grateful to Drs W. G. Wood and R. J. Gibbons for their comments on the manuscript; we thank Liz Rose for help in preparing the manuscript. We are grateful to Professor Sir D. J. Weatherall for his continued support and encouragement.
5 Brown, W.R.A. (1992) Current Biology 2, 127-129.
6 Higgs, D.R. (1993) In D. R. Higgs and D. J. Weatherall (eds.) Bailliere's Clinical Haematology. International Practice and Research: The Haemoglobinopathies, Bailliere Tindall, London.
10 Flint, J., Craddock, C.F., Villegas, A., Bentley, D.P., Williams, H.J., Galanello, R., Cao, A., Wood, W.G., Ayyub, H. and Higgs, D.R. (1994) Am. J. Hum. Genet. 55, 505-512.MEDLINE Abstract
20 Weatherall, D.J. and Clegg, J.B. (1981) The Thalassaemia Syndromes, Blackwell Scientific Publications, Oxford.
21 Weatherall, D.J., Clegg, J.B. and Naughton, M.A. (1965) Nature 208, 1061-1065.MEDLINE Abstract
22 Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor.
23 Buckle, V.J. and Rack, K. (1993) In K. E. Davies (eds.) Human Genetic Diseases, IRL Press, Oxford.
*To whom correspondence should be addressed
This page is maintained by OUP admin. Last updated Thu Oct 31 15:26:30 GMT 1996. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996
