DNA homology requirements for mitotic gap repair in Drosophila.

Abstract

We used P transposable-element mobilization to study the repair of double-strand DNA breaks in Drosophila melanogaster premeiotic germ cells. Distribution of conversion tracts was found to be largely unaffected by changes in the length of sequence homology between the broken ends and the template, suggesting that only a short match is required. However, the frequency of repair was highly sensitive to single-base mismatches within the homologous region, ranging from 19% reversion when there were no mismatches to 5% when 15 mismatches were present over a 3455-bp span.Efficient recombination between two stretches of DNA requires that the two sequences locate each other in the genome and form the stable recombinational intermediate complex. Some degree of homology between the two interacting sequences is needed for efficient recombination by most pathways (1).The questions of how long the stretch of homology must be and how closely the sequences must match within this stretch have been approached with various experimental systems in bacteria (2-5), yeast (6-8), and mammals (9-15). The general conclusion from these studies is that the recombination mechanisms are much more sensitive to degree of homology than would be predicted purely on the basis of DNA-duplex stability. In Escherichia coli, for example, a single-base mismatch in 53 bp resulted in a 4-fold reduction in recombination (4), and a 40-fold reduction was seen when the mismatch proportion was increased to 10% (3). The minimum length of homology required has been estimated at 20-50 bp for E. coli (2-4) and 63-89 bp for Saccharomyces cerevisiae (7). For mammalian cells, estimates range from 163 to 300 bp (9, 13), but some studies indicate efficient recombination with as little as 20 bp (9, 10). One suggestion (2, 5) is that the length and stringency of homology requirements reflect an evolutionary adaptation to prevent recombination between unrelated sites and are, thus, expected to depend on size and complexity of the genome.In this report we make use of the recently developed method of transposable-element-induced gap repair to address the question of homology requirements for recombinational DNA repair. P elements are DNA transposons that have been used for a variety oftechniques in Drosophila (16). Recent data (17-19) led to a model in which P element excision leaves behind a double-stranded DNA break. According to this interpretation, such breaks are often extended into various-sized gaps through exonuclease activity. Repair occurs when the ends of the gap invade a homologous DNA sequence from which to copy the missing genetic information. The result is unidirectional transfer of information from the template sequence to the excision site. The template can be the sister strand or an ectopic site located elsewhere in the genome or extrachromosomally (17,18,20).The system allows us to examine the length and stringency of homology requirements in Drosophila melanogaster. A further advantage of this approach is that recombination occurs between sequ...

Comparison of Targeted-Gene Replacement Frequencies in Drosophila melanogaster at the forked and white Loci

“…In an equivalent experiment, the reversion rate of w hd was found to range between 5 and 18%, depending on the degree of homology of the template (7,30). The frequency of 8.1% in the present experiment is at the lower range of the frequency variations (5 to 18%) found by Nassif and Engels (30).…”

Differential Usage of Alternative Pathways of Double-Strand Break Repair in Drosophila

“…Whereas I-SceI cleavage produces ends with complementary 4-base, 39 overhangs, the ends formed by P transposition contain 17-base single-stranded, 39 overhangs that are not complementary (Beall and Rio 1997). Yet the overall conclusions are similar for both types of ends: (1) SDSA is used much more frequently than DHJ as evidenced by low crossover rates (Engels et al 1990;Johnson-Schlitz and Engels 1993;Nassif and Engels 1993); (2) SSA can be very efficient (Preston et al 2002;Rong and Golic 2003); and (3) the majority of NHEJ products have simple deletions, insertions, or a combination of deletion and insertion (Rong and Golic 2003;McVey et al 2004b;Weinert et al 2005). With both P-and I-SceI-induced breaks, many of the simple deletion products appear to exploit microhomology to align the ends (Staveley et al 1995;Romeijn et al 2005).…”

“…All the experiments presented here examine events in the male germ line while some earlier experiments have looked at the female germ line ( Johnson-Schlitz and Engels 1993). Previous work has investigated other aspects of DSB repair that we have not tested with Rr3 such as the length of conversion tracts (Gloor et al 1991), the effects of polymorphisms (Nassif and Engels 1993;Dray and Gloor 1997;Coveny et al 2002), the repair efficiency of large gaps (Nassif et al 1994;McVey et al 2004a), and the preference hierarchy of template location (i.e., sister chromatid vs. allelic vs. ectopic sites in cis or in trans) (Johnson-Schlitz and Engels 1993; Engels et al 1994;Keeler et al 1996;Lankenau et al 1996;Lankenau and Gloor 1998).…”

Allele-specific gene targeting in Candida albicans results from heterology between alleles The GenBank accession numbers for the sequences reported in this paper are AF247189 and AF247190 for PHR1-1 and PHR1-2, respectively.

“…Recombination between like alleles was a homologous event while recombination between the diverged alleles was a homeologous event. It is well established that sequence heterology interferes with recombination in prokaryotic and eukaryotic species (Chen & Jinks-Robertson, 1998 ;Elliott et al, 1998 ;Nassif & Engels, 1993 ;Stambuk & Radman, 1998) and this includes targeted gene replacements (Leung et al, 1997 ;Negritto et al, 1996). In S. cerevisiae a single nucleotide difference within a 350 bp region can reduce recombination fourfold (Datta et al, 1997).…”

“…Homeologous recombination is less efficient than homologous recombination (Chen & Jinks-Robertson, 1998 ;Elliott et al, 1998 ;Nassif & Engels, 1993 ;Stambuk & Radman, 1998) and this could account for the bias in gene replacement.…”