Mechanisms of recombination between diverged sequences in wild-type and BLM-deficient mouse and human cells

Abstract

Double-strand breaks (DSBs) are particularly deleterious DNA lesions for which cells have developed multiple mechanisms of repair. One major mechanism of DSB repair in mammalian cells is homologous recombination (HR), whereby a homologous donor sequence is used as a template for repair. For this reason, HR repair of DSBs is also being exploited for gene modification in possible therapeutic approaches. HR is sensitive to sequence divergence, such that the cell has developed ways to suppress recombination between diverged ("homeologous") sequences. In this report, we have examined several aspects of HR between homeologous sequences in mouse and human cells. We found that gene conversion tracts are similar for mouse and human cells and are generally < or =100 bp, even in Msh2(-)(/)(-) cells which fail to suppress homeologous recombination. Gene conversion tracts are mostly unidirectional, with no observed mutations. Additionally, no alterations were observed in the donor sequences. While both mouse and human cells suppress homeologous recombination, the suppression is substantially less in the transformed human cells, despite similarities in the gene conversion tracts. BLM-deficient mouse and human cells suppress homeologous recombination to a similar extent as wild-type cells, unlike Sgs1-deficient Saccharomyces cerevisiae.

FIG. 1.

Models for noncrossover gene conversion resulting from DSB repair. DSB repair is initiated by resection of the DNA ends (black; strand directionality is designated a 3′ “tail”). The resected 3′ overhang invades the homologous donor template (gray), forming hDNA at the site of invasion (i), which acts as a primer/template for repair synthesis (gray dotted line). (A) In the canonical DSB repair (DSBR) model, the second strand of the DSB is captured, resulting in another stretch of hDNA (ii) and repair synthesis, to form a double Holliday junction. Depending on how the double Holliday junction is cleaved (arrowheads), resolution can result in a crossover (data not shown) or a noncrossover, as shown. (B) In SDSA, the newly synthesized strand dissociates from the D-loop and anneals to the other DNA end to form another stretch of hDNA (iii). Repair synthesis and ligation result in a noncrossover product. While one-end invasion is illustrated for the SDSA model, it is possible for both DNA ends to invade, resulting in gene conversion on both sides of the DSB (data not shown). In both models, hDNA formed by the newly synthesized strands can be repaired by MMR, resulting in gene conversion of markers (data not shown).

FIG. 2.

Homologous and homeologous recombination in wild-type and Msh2^−/− mouse ES cells. (A) Intrachromosomal recombination substrates measure homologous (H-DR-WT) and homeologous recombination (H-DR-10mu). S2neo is nonfunctional due to an insertion of the I-SceI recognition sequence at the endogenous NcoI site. pneo-WT is nonfunctional as both the 5′ and 3′ ends are truncated; pneo-10mu is similar but contains 10 silent restriction site polymorphisms that increase the divergence with S2neo to ∼1.5%. After I-SceI cleavage, noncrossover gene conversion between the two repeats results in neo⁺ recombinants, with the I-SceI converted to the NcoI site (striped). The numbers of base pairs of homologous sequences on each side of the DSB are shown. For H-DR-10mu, gene conversion may also result in the incorporation of the polymorphisms (A, ApaI; L, ApaLI; P, PstI; B, BamHI; X, XbaI; Nr, NruI; Ns, NsiI; Bs, BspEI; Na, NaeI; Pm, PmlI). Primers 1 and 2 were used to sequence the neo⁺ gene, and primers 3 and 4 were used to sequence pneo-10mu. (B) Msh2^−/− cells fail to suppress recombination between diverged sequences. The H-DR substrates were targeted to the Hprt locus in wild-type and Msh2^−/− ES cells. After I-SceI expression, the HR rate was determined by dividing the total number of neo⁺ colonies by the number of cells surviving electroporation. Rates relative to H-DR-WT are graphically represented. Data represent the mean and standard error of the mean (SEM) of four independent experiments. (C) Classes of gene conversion tracts from murine ES cells. Each tract was grouped into one of five classes (represented graphically in descending order): conversion of only the NcoI site, conversion of NcoI and NsiI (8 bp to right of DSB), conversion to right of the break (>8 bp; unidirectional), conversion to left of the break (unidirectional), and conversion to both sides of the break (bidirectional).

FIG. 3.

Homologous and homeologous recombination in wild-type human cells. (A) Human fibroblasts suppress recombination between diverged sequences in plasmid substrates. Recombination was determined in human fibroblasts (GM00637) by cotransfecting the H-DR-WT or H-DR-10mu substrate with the I-SceI expression plasmid and scoring for neo⁺ recombinants. As controls, wild-type and Msh2^−/− murine ES cells were similarly analyzed. Rates relative to H-DR-WT are graphically represented. Data represent the mean and SEM of four independent experiments. (B) HR by DSB-induced gene targeting. S2neo was randomly integrated into the chromosome of human fibroblasts (GM00637), and three single-copy integrants were identified (see Fig. S1B in the supplemental material). pneo-WT or pneo-10mu plasmids were then cotransfected with pCBASce. S2neo receives the break and can repair off the pneo-WT or pneo-10mu plasmids, giving rise to noncrossover neo⁺ gene conversion recombinants. (C) Human fibroblasts suppress recombination between diverged sequences during DSB-induced gene targeting. Rates relative to pneo-WT are graphically represented. Data represent the mean and SEM of three independent experiments, each containing at least two of three clones. Individual clones had similar relative rates (see Table S1 for values of each clone). (D) Gene conversion tracts in human fibroblasts from DSB-induced gene targeting. Fifty-eight neo⁺ clones derived from cells cotransfected with pneo-10mu and pCBASce were analyzed to determine the minimal extent of gene conversion. The average minimum and maximum gene conversion tracts are indicated (± SEM), where the minimum tract length is calculated to the last polymorphism converted in both directions and where the maximum tract length includes the distance to the next polymorphism in both directions. The numbers at the top of the graph are the percentages of recombinants with conversion at each polymorphism. NcoI (bold) was converted in 100% of recombinants. The number of clones represented by each tract is indicated on the right. (E) Classes of gene conversion tracts from human fibroblasts. Each tract from panel D was placed into one of five classes, as described in Fig. 2C.

FIG. 4.

Homologous and homeologous recombination in BLM-deficient mammalian cells. (A) Fibroblasts from BS patients were analyzed for their ability to suppress homeologous recombination utilizing the H-DR plasmid assay (left) and the DSB-induced gene-targeting assay (right). Rates relative to H-DR-WT (left) or pneo-WT (right) are represented. Data represent the mean and SEM of four independent experiments (left) or three independent experiments, each containing two of three clones (right). Individual clones had similar relative rates (see Table S1 in the supplemental material for values of each clone). (B) Gene conversion tracts in BS cells from DSB-induced gene targeting. Minimal gene conversion tracts from 61 neo⁺ clones are shown. The average minimum and maximum gene conversion tracts are indicated (± SEM), as described in the legend of Fig. 3D. The numbers at the top of the graph are the percentages of recombinants with conversion at each polymorphism. NcoI (bold) was converted in 100% of recombinants. The number of clones represented by each tract is indicated on the right. (C) Classes of gene conversion from BS cells. Each tract from panel B was grouped into one of five classes, as described in the legend of Fig. 2C. (D) BLM expression in Blm^tet/tet S2neo murine ES cells is suppressed by doxycycline (1 μg/ml for 48 h, the time of electroporation) to levels undetectable by Western blotting. (E) DSB-induced gene targeting in Blm^tet/tet and wild-type murine cells in the presence or absence of doxycycline. Homologous recombination between diverged sequences is suppressed in the presence or absence of BLM. The overall HR rate, however, is decreased about 1.4-fold in the absence of BLM. HR rates were determined by dividing the total number of neo⁺ colonies by the number of cells surviving 24 h after electroporation; the small decrease in Blm^tet/tet clonogenic formation in the presence of doxycycline was also factored in. Rates relative to pneo-WT donor sequence are graphically represented. Data represent the mean and SEM from six independent experiments.