Drosophila MUS312 and the vertebrate ortholog BTBD12 interact with DNA structure-specific endonucleases in DNA repair and recombination

Abstract

DNA recombination and repair pathways require structure-specific endonucleases to process DNA structures that include forks, flaps, and Holliday junctions. Previously, we determined that the Drosophila MEI-9-ERCC1 endonuclease interacts with the MUS312 protein to produce meiotic crossovers, and that MUS312 has a MEI-9-independent role in interstrand crosslink (ICL) repair. The importance of MUS312 to pathways crucial for maintaining genomic stability in Drosophila prompted us to search for orthologs in other organisms. Based on sequence, expression pattern, conserved protein-protein interactions, and ICL repair function, we determined that the mammalian ortholog of MUS312 is BTBD12. Orthology between these proteins and S. cerevisiae Slx4 helped identify a conserved interaction with a second structure-specific endonuclease, SLX1. Genetic and biochemical evidence described here and in related papers suggest that MUS312 and BTBD12 direct Holliday junction resolution by at least two distinct endonucleases in different recombination and repair contexts.

Figure 1. MUS312 is orthologous to BTBD12 and Slx4

(A) Domain architecture of S. cerevisiae Slx4, C. neoformans BSP1, D. melanogaster MUS312, and H. sapiens BTBD12. Open boxes: conserved C-terminal domain; filled boxes: internal motif; hatched boxes: predicted coiled-coils; stippled boxes: BTB domain of BTBD12 and the regions on MUS312 and BSP1 that have sequence similarity (Fig. S2). (B) Alignments of C-termini. Two divergent representatives from vertebrates, arthropods, and fungi are shown. Hsap = H. sapiens; Ggal = Gallus gallus; Tcas = Tribolium castaneum; Dm = D. melanogaster; Cgat = C. gatti; Scer = S. cerevisiae. Predicted alpha helices are underlined. (C) Yeast two-hybrid interactions. Serial dilutions of cells expressing the indicated fusions to the Gal4 DNA binding domain (BD) or activating domain (AD) were plated on -leu -trp or -leu -trp -his dropout plates; growth on the former requires the presence of both the BD and the AD plasmid, and growth on the latter indicates a physical interaction. Top half: human proteins; bottom half: fly proteins.

Figure 2. Btbd12 expression is increased in cells undergoing meiotic recombination

(A) Relative expression levels across multiple tissues. Btbd12 mRNA is most highly expressed in mouse testis and oocytes (biogps.gnf.org). (B) Post-natal testis expression. Btbd12 expression increases during development, peaking at sexual maturity (Schultz et al., 2003). (C) Stage-specific testis expression. Btbd12 expression peaks during pachytene, which is when meiotic recombination occurs (Namekawa et al., 2006).

Figure 3. BTBD12 acts in ICL-repair

(A) Depletion of BTBD12 or SLX1 affects cell proliferation. XTT reduction was measured four days after transfection of HeLa cells with siBTBD12, siSLX1, siXPF, or siControl. siBTBD12 and siSLX1 caused identical decreases in XTT reduction, indicating slowed proliferation or cell death; siXPF and siControl had no effect. Bars indicate mean of at least five experiments, and error bars denote SEM. Asterisks indicate P<0.05 by paired t test. (B), (C) Transfection with siRNA to deplete BTBD12 or SLX1 causes hypersensitivity to HN2. Three days after transfection with the indicated siRNA, cells were exposed to the indicated concentration of HN2 or MMS for 24 hours. Relative cell respiration was measured with the XTT assay, normalizing to decreases caused by siRNA treatment alone (panel A). Each bar represents the mean from five separate experiments, with error bars indicating S.E.M. Sensitivity was not detected at the lowest dose, and the highest dose caused extensive cell death in both control and experimental; the difference was significant at the intermediate dose, however. Asterisks indicate P <0.05 by paired t test. (D) HN2 causes early S phase accumulation after knockdown of BTBD12 or SLX1. Cell number is plotted as a function of DNA content. Cells were transfected with siRNA, then three days later were mock treated (left) or treated with HN2 (right). Bars indicate cells with 2C (G1) and 4C (G2) DNA content.

Figure 4. Synthetic lethality between mus312 and mus309

(A) Cross scheme to detect synthetic lethality. This cross generates progeny doubly mutant for mus312 and mus309. No double mutant adults (top genotype) eclosed, though there were expected to be as frequent as the lower genotype (the two genotypes not listed are inviable). (B) mus312 mus309 double mutants have small brains. Brains were dissected from wandering L3 larvae and stained with DAPI. Brains from mus312 or mus309 single mutants are indistinguishable from wild-type brains (data not shown), but brains from double mutants are severely under-developed. (C) mus312 mus309 synthetic lethality is not suppressed by mutation of spn-A. This cross generates progeny triply mutant for mus312, mus309, and spn-A. No triple mutant adults (top genotype) eclosed, though there were expected to be half as frequent as the lower genotype (either triple mutant chromosome over the TM3 balancer; TM3 / TM3 is inviable).