Distinct pathways of homologous recombination controlled by the SWS1–SWSAP1–SPIDR complex

Homology-directed repair (HDR), a critical DNA repair pathway in mammalian cells, is complex, leading to multiple outcomes with different impacts on genomic integrity. However, the factors that control these different outcomes are often not well understood. Here we show that SWS1–SWSAP1-SPIDR controls distinct types of HDR. Despite their requirement for stable assembly of RAD51 recombinase at DNA damage sites, these proteins are not essential for intra-chromosomal HDR, providing insight into why patients and mice with mutations are viable. However, SWS1–SWSAP1-SPIDR is critical for inter-homolog HDR, the first mitotic factor identified specifically for this function. Furthermore, SWS1–SWSAP1-SPIDR drives the high level of sister-chromatid exchange, promotes long-range loss of heterozygosity often involved with cancer initiation, and impels the poor growth of BLM helicase-deficient cells. The relevance of these genetic interactions is evident as SWSAP1 loss prolongs Blm-mutant embryo survival, suggesting a possible druggable target for the treatment of Bloom syndrome.

a. SWS1-SWSAP1 interaction with BLM. MBP-BLM was pulled down with anti-MBP beads and interacting proteins were identified by western blotting with the indicated antibodies. All of the proteins are human. n=3, where n is the number of independent experiments. b. FRET analysis shows that mouse SWS1 and SWSAP1 form a complex in RPE cells. An interaction with SWS1 is only observed when SWSAP1 is tagged at the C terminus. Clover and mRuby2-Clover fusion, n=5; mRuby2+Clover, SWS1-mRuby2 fusion+SWSAP1-Clover fusion, mRuby2-SWS1 fusion+SWSAP1-Clover fusion, SWS1-mRuby2 fusion+Clover-SWSAP1 fusion, mRuby2-SWS1 fusion+Clover-SWSAP1 fusion, n=4, where n is the number of independent experiments. c-f. Interactions are observed for mouse SWSAP1 with SWS1 (c) and with SPIDR (d) and also between the three proteins with RAD51 (e,f) in co-immunoprecipitation experiments from HEK293 cells. FLAG-tagged proteins were pulled down with anti-FLAG beads and interacting proteins were identified by western blotting with the indicated antibodies. n=3, where n is the number of independent experiments. g,h. RAD51 interactions are also observed with human SWS1-SWSAP1 (g) and SWS1-SWSAP1-SPIDR (h) using purified proteins. MBP-tagged proteins were pulled down with anti-MBP beads and interacting proteins were identified by western blotting with the indicated antibodies. n=3, where n is the number of independent experiments. i,j. Interactions are observed for mouse SWS1-SWSAP1 (i) and SPIDR (j) with DMC1 in coimmunoprecipitation experiments from HEK293T cells. FLAG-tagged proteins were pulled down with anti-FLAG beads and interacting proteins were identified by western blotting with the indicated antibodies. n=3, where n is the number of independent experiments.Error bars in b, mean ± s.d. ***P ≤ 0.001; ****P ≤ 0.0001; unpaired t test, two-tailed.
All source data are provided in the Source Data file. and Swsap1 Spidr 129/B6 Blm tet/tet (c) ES cells. Genomic structure and sequence around the gRNA (blue) and PAM (red) sequences are shown. The number for each mutant clone is shown on the left, together with the detected mutated allele(s). The truncated protein product predicted from each mutated allele is diagrammed on the right. 129/B6 Blm tet/tet were used in c to be able to regulate BLM expression (-Dox, BLM is expressed; +Dox, BLM is not expressed). Error bars in b,c,e,f, mean ± s.d. **P ≤ 0.01; ****P ≤ 0.0001; unpaired t test, two-tailed.
All source data are provided in the Source Data file.

Supplementary Fig. 7: IH-HR analysis in Sws1 -/-, Swsap1 -/-, and Spidr -/-ES cells.
a. IH-HR in individual clones for data presented in Fig. 2b. Relative to WT cells, colony counts expressed relative to wild-type cells transfected with an empty vector within each experiment.
b. In Sws1 -/cells, RAD51 K133R (KR) and BRC3 peptide expression reduces HDR in DR-GFP to a similar extent as in wild-type and Swsap1 -/cells. (Compare with Fig. 2d).
c. Transient overexpression of RAD51 I287T (IT) reduces HDR in wild-type cells using the DR-GFP reporter, although to a lesser extent than RAD51 K133R.
where n is the number of independent experiments. This graph uses the same data as in Fig. 2f but here the data are presented relative to wild-type cells that are transfected with an empty vector (-).
All source data are provided in the Source Data file. Supplementary Fig. 9: Loss of SWS1-SWSAP1-SPIDR reduces SCEs and LOH, and enhances cell proliferation in the absence of BLM.
a. SCE analysis in Blm tet/tet cells after Dox exposure to deplete BLM. Representative metaphase chromosome spreads are shown along with quantification from individual clones for data presented in where n is the number of independent experiments.  Genotypes for Swsap1 are shown first followed by genotypes for Blm. g. Timed matings showing that Swsap1 -/-Blm -/embryos are obtained at nearly normal Mendelian ratios at E15.5, whereas Blm -/single mutants are highly underrepresented. Partial data from this table are presented in Fig. 4a.
All source data are provided in the Source Data file.