We characterised the molecular breakpoint of patients with deletion covering the region between exon 44 and exon 50 of dystrophin gene.
We analysed introns 44, 45, 47, 48 and 49 which sequence was completely available in public database and that measure 248, 36, 54, 37 and 17 kb respectively. These regions were characterised for repetitive elements using the RepeatMasker program (http://www.hgmp.mrc.ac.uk) and matrix attachment regions (MARS) using the MAR-WIZ program (http://www.futuresoft.org/MAR-Wiz).
After masking the repetitive regions we designed a set of PCR primers and performing multiple PCR amplification of non overlapping fragments we determined the sublocalization of the deletion breakpoint in all the DNA samples with at least one end lying in that region.
To PCR amplify and analyse in more detail the deletion junctions, we performed a long range PCR using primers as close as possible to the breakpoint extremities or utilised a PCR based genome walking technique (Clontech) starting from the last known region on one side of the breakpoint. The PCR fragments so amplify were then cloned and sequenced.
In this way we succeeded in characterising 12 genomic junctions. We detected no substantial homologies between the normal DNA sequences located across the breakpoints. These data suggest that some mechanisms other than homologous recombination operated in the studied cases. In the breakpoint sequences analysed, we identified some elements that could be involved in double-strand break events. Inverted repeats, able to form stem-loop structures and associated with a strong topoisomerase II cleavage site, are found at or near to six breakpoints.
The sequence TTTAAA, known to be able to induce a curvature in the DNA molecule (which may predispose it to recombination), is found in the 50 bp flanking the studied deletion breakpoints with a frequency that is higher than the average frequency of this sequence in the whole introns (1/560 versus 1/712), as previously found in dystrophin intron 49. With regard to the breakpoint distribution, the results obtained indicate that they are widely scattered in intron 48 and 49, as also observed in other regions of the dystrophin gene.
Although the number of breakpoints analysed is not high, in intron 47 three of 11 breakpoint are clustered in a region of 2,5kb next to the only matrix attachment regions (MARs) identified. The analysis of DNA sequences across the breakpoints revealed in three cases an insertion of a few nucleotides (1-5 bp), in five cases a short homology region (2-4 bp), and more interestingly, in 4 junctions a duplication of variable length (9-24 bp). This molecular configuration at junction ends can be explained with the processes underlying nonhomologous end joining. Repair of double-strand breaks by this mechanism is achieved after limited processing of the DNA ends, followed by joining and re-ligation, which requires little or no sequence homology at the ends. This process, however, may lead to loss of or short insertions of nucleotides at the join, hence the alternative names of illegitimate recombination or errorprone recombination.
We have compared the frequencies of repetitive sequences in introns 47 and 48 with that of the whole dystrophin gene. We observed that introns 47 and 48 are not only characterized by the highest recombination rate in the gene of healthy individuals and in BMD/DMD patients, but present a very high percentage of repetitive sequence as well. A similar correlation among recombination rate, deletion frequency, and percentage of repetitive elements was also observed in intron 7. Deletion breakpoints in introns 47 and 48 do not appear associated to any particular type of repetitive sequences, as also observed in other dystrophin introns (intron 7, intron 44).
However, repetitive sequences often contain microsatellites, palindromes, and Alu-associated regions that could form hairpin loops predisposing to double-strand DNA breaks. Nonhomologous end joining is sometimes associated with the insertion of novel fragments of DNA into the double-strand break in yeast, plant, and also mammalian chromosomes. The sequences captured at double-strand molecular breaks had various sources, including microsatellites, retrotransposable elements, reverse transcripts of spliced introns, and exogenous DNA sequences.
On the basis of these observations we hypothesize that the ancestral nucleotide sequence of large introns might have contained nucleotide sequence that predisposed to doublestrand molecular breaks, recombination, and insertion of DNA fragments. The inserted sequences might then have favoured further insertions and recombination increasing intron size and deletion frequencies that are characteristic of these intronic regions.
Toffolatti L, et al.
Investigating the mechanism of chromosomal deletion: characterization of 39 deletion breakpoints in introns 47 and 48 of the human dystrophin gene. Genomics. 2002; 80(5):523-30.