Two auxiliary factors promote Dmc1-driven DNA strand exchange via stepwise mechanisms

Abstract

Homologous recombination (HR) is a universal mechanism operating in somatic and germ-line cells, where it contributes to the maintenance of genome stability and ensures the faithful distribution of genetic material, respectively. The ability to identify and exchange the strands of two homologous DNA molecules lies at the heart of HR and is mediated by RecA-family recombinases. Dmc1 is a meiosis-specific RecA homolog in eukaryotes, playing a predominant role in meiotic HR. However, Dmc1 cannot function without its two major auxiliary factor complexes, Swi5-Sfr1 and Hop2-Mnd1. Through biochemical reconstitutions, we demonstrate that Swi5-Sfr1 and Hop2-Mnd1 make unique contributions to stimulate Dmc1-driven strand exchange in a synergistic manner. Mechanistically, Swi5-Sfr1 promotes establishment of the Dmc1 nucleoprotein filament, whereas Hop2-Mnd1 defines a critical, rate-limiting step in initiating strand exchange. Following execution of this function, we propose that Swi5-Sfr1 then promotes strand exchange with Hop2-Mnd1. Thus, our findings elucidate distinct yet complementary roles of two auxiliary factors in Dmc1-driven strand exchange, providing mechanistic insights into some of the most critical steps in meiotic HR.

Keywords: Dmc1; Rad51; double-stranded break; fission yeast; homologous recombination.

Fig. 1.

Hop2-Mnd1 stimulates Dmc1-mediated strand exchange, leading to robust accumulation of hyper JMs. (A) Schematic of the in vitro strand exchange assay using PhiX174 DNA as substrates. (B) css was first incubated with Dmc1, then mixed with Hop2-Mnd1, RPA, and finally with lds, followed by a further incubation for 2 h at 30 °C. The products were analyzed by agarose gel electrophoresis. The signal denoted with asterisks corresponds to “hyper JMs,” which are defined as JMs bigger than those normally formed in reactions containing Swi5-Sfr1. (C) Time course analysis of Dmc1-mediated strand exchange promoted by either Hop2-Mnd1 or Swi5-Sfr1. (D) Synergistic activation of Dmc1 by Hop2-Mnd1 and Swi5-Sfr1. Dmc1 was preincubated with css, then mixed with the indicated auxiliary factor(s) and lds, and the reaction was incubated for 1 h at 30 °C. In B–D, the following concentrations of substrates were used: css, 10 µM; Dmc1, 5 µM; lds, 10 µM; RPA, 1 µM. In C and D, Swi5-Sfr1 and Hop2-Mnd1 were used at concentrations of 0.5 µM and 0.25 µM, respectively. For the graphs in B–D, mean values ± SD from three independent experiments are shown.

Fig. 2.

Hop2-Mnd1 stimulates efficient strand exchange by bypassing the donor dsDNA end dependency. (A) dsDNA ends examined in the strand exchange reactions in B. (B) Strand exchange reactions were conducted with lds with various ends as indicated. css was first incubated with Dmc1 for 5 min, then supplemented with Swi5-Sfr1 or Hop2-Mnd1 and RPA, and incubated for a further 5 min. The reaction was then initiated through the addition of lds and incubated for 1 h at 30 °C. Reaction products were analyzed by agarose gel electrophoresis. css, 10 µM; Dmc1, 5 µM; Hop2-Mnd1, 0.25 µM; lds, 10 µM; RPA, 1 µM; Swi5-Sfr1, 0.5 µM. Mean values ± SD from three independent experiments are shown.

Fig. 3.

Hop2-Mnd1 is not an efficient stabilizer of the Dmc1 presynaptic filament. (A) Dmc1-driven strand exchange reaction with RPA-precoated ssDNA. css was precoated with RPA, then mixed with Dmc1, lds, and the indicated auxiliary factor. (B) Schematic of the RPA chase assay. (C) RPA chase assay. css annealed to a biotinylated oligo was initially precoated with Dmc1, then incubated in the presence of the indicated auxiliary factor, followed by the addition of RPA. css was precipitated with streptavidin-coated magnetic beads and associated proteins were examined (css-bound), along with proteins left in the supernatant (unbound), by SDS-PAGE. css, 10 µM; Dmc1, 5 µM; Hop2-Mnd1, 0.25 or 0.5 µM; RPA, 1 µM; Swi5-Sfr1, 0.5 or 1 µM. (D) Presynaptic filament stability assayed by fluorescence anisotropy. Dmc1 presynaptic filaments were formed by mixing Dmc1 with fluorescently labeled oligo-dT (72-mer), then the mixture was incubated with the indicated auxiliary factor. Stability of the formed filaments was examined by diluting the mixture (40-fold) and monitoring the change in fluorescence anisotropy in real time. (E) Synergistic activation of Dmc1-driven strand exchange by Hop2-Mnd1 and Swi5-Sfr1 with RPA-precoated ssDNA. css precoated with RPA was mixed with Dmc1, the indicated auxiliary factor(s), and lds prepared with StuI. The reaction was then incubated at 30 °C for 60 min. css, 10 µM; Dmc1, 5 µM; Hop2-Mnd1, 1 µM; lds, 10 µM; RPA, 2 µM; Swi5-Sfr1, 4 µM. For the graphs in A, C, and E, mean values ± SD from three independent experiments are shown. Representative kinetics of the reactions are shown in D.

Fig. 4.

Hop2-Mnd1 allows Dmc1 to exchange strands with homology embedded within nonhomologous DNA. (A) Schematic of the D-loop assay employing a linear ssDNA and either lds (Left) or nc (Right). (B) Gel images of the D-loop assay with naked lss. (C) Gel images of the D-loop assay with RPA-precoated lss. In B and C, lds (Left) or nc (Right) was used. In B, lss was preincubated with Dmc1, then mixed with auxiliary factor(s) and dsDNA. In C, lss was precoated with RPA, then mixed with Dmc1, auxiliary factor(s), and dsDNA. After the addition of dsDNA, the reaction was incubated at 30 °C for 1 h. Yield is expressed as the percentage of the signal for the products to the sum of the products and either lds or nc. Dmc1, 5 µM; Hop2-Mnd1, 0.25 µM; lds, 5 µM; lss, 5 µM; nc, 5 µM; RPA, 1 µM; Swi5-Sfr1, 0.5 µM in B and 4 µM in C. For the graphs in B and C, mean values ± SD from three independent experiments are shown.

Fig. 5.

The effect of heterology at lss ends on strand exchange. (A) Schematic of the D-loop assay using lss with heterology (70 bp) at its end(s). (B) lss was preincubated with Dmc1, then mixed with auxiliary factor(s), RPA, and lds. The reaction was then incubated at 30 °C for 1 h. Dmc1, 5 µM; Hop2-Mnd1, 0.25 µM; lds, 5 µM; lss, 5 µM; RPA, 1 µM; Swi5-Sfr1, 0.5 µM. (C) Quantification of results shown in B. Yield is expressed as the percentage of the signal for the products to the sum of the lds and products. Mean values ± SD from three independent experiments are shown. (D) Stepwise involvement of two auxiliary factors in the establishment of Dmc1-driven synapsis. (a) Following meiotic DSB formation and end-resection, RPA binds to the ssDNA exposed at a DSB end. (b) Swi5-Sfr1 functions as a canonical “mediator” and facilitates the replacement of RPA with Dmc1, which forms a nucleoprotein filament. (c) Hop2-Mnd1 executes its role as an initiator of strand exchange by allowing the Dmc1 filament to invade a donor dsDNA duplex, which constitutes initiation of the synaptic phase. (d) Swi5-Sfr1 and Hop2-Mnd1 may then promote further strand transfer within the synaptic phase, leading to heteroduplex extension. See text for more details.