Fig 2: A possible mechanism of HELQ function in HR and SSA.HR, In HR, following DSB, 5' DNA ends are resected by nucleases, which generates RPA-coated 3' overhangs. RPA is displaced by RAD51 giving rise to presynaptic nucleoprotein filament formation. This RAD51-filament invades homologous DNA duplex through its 3' end. DNA invasion result in D-loop formation, which is extended by DNA synthesis. If the SDSA pathway is invoked, HELQ can disrupt the D-loop to displace and/or anneal the extended strand with the broken duplex via its helicase and strand annealing activities. Alternatively, in DSBR, the D-loop can be greatly extended and RPA-coated DNA strands in D-loop and second 3' overhang can be captured by HELQ. During this second-end capture step, HELQ strips RPA from ssDNA and anneals complementary DNA strands together to prime DNA synthesis, restoring the broken DNA and resulting in either crossover or non-crossover products. SSA, similar to HR, longer RPA-coated 3' overhangs can be captured by HELQ. Post capture, HELQ strips RPA from both ssDNAs and actively anneals the DNA strands together using base-pairing. The remaining flaps are cleaved, and nicks are ligated by specific nuclease and ligases completing the DSB repair reaction.

Fig 3: RAD51 forms a co-complex with HELQ and stimulates HELQ unwinding activity.a, Representative gel of the DNA unwinding assay with the indicated concentrations of HELQ with 3' overhang substrate. The asterisk indicates the position of fluorescein isothiocyanate (FITC) labelling at 5' end of the oligo. The products were resolved on a 10% native polyacrylamide gel. b, Quantification of the experiments such as shown in a and Extended Data Fig. 1b–d. HELQ concentrations of only 1–90 nM are shown. n = 4 independent experiments. Data are mean ± s.e.m. c, Representative gel of DNA unwinding of 3' overhang substrate with HELQ and the indicated concentrations of RAD51 or RecA. d, e, Quantification of the experiments shown in c and Extended Data Fig. 2c, d for RAD51 (d) and RecA (e). n = 3 (3' overhang), n = 4 (Y-structure), n = 3 (D-loop) and n = 3 (lagging strand fork) independent experiments. Data are mean ± s.e.m. f, Representative gel of the DNA unwinding assay of 3' overhang substrate with the indicated concentrations of HELQ in the absence and presence of RPA (20 nM). g, Quantification of the experiments shown in f. n = 3 independent experiments. Data are mean ± s.e.m. h, Schematics of the experimental set-up of the optical tweezer (C-Trap) system to observe DNA unwinding. These experiments were performed at room temperature. i–k, Bead centre displacement measured between the traps as a function of time with 25 nM RAD51 (i), 50 nM HELQ (j), and 25 nM RAD51 and 50 nM HELQ (k). The traces represent individual DNA molecules (n = 4). l, Representative kymographs of single Alx–RAD51 binding events on gapped DNA in the presence or absence of 50 nM HELQ or HELQ(K365M). Unidirectional movement of Alx–RAD51 indicates translocation of Alx–RAD51–HELQ complex. Scale bars, 60 s (horizontal), 10 µm (vertical, left), 5 µm (vertical, right).

Fig 4: HELQ specifically unwinds substrates with 3' overhang and D-loop.a, SDS-PAGE gel (4–12% polyacrylamide) showing purified recombinant human HELQ WT and HELQ(K365M) from insect cells. The gel was stained with Coomassie brilliant blue (CBB). We used two and single preparations of HELQ WT and HELQ(K365M) respectively in this study. b-d, Representative native gels (10% polyacrylamide) of DNA unwinding assay of D-loop, Y-structure and lagging strand fork with indicated concentrations of HELQ. The asterisk indicates the position of FITC labelling at 5' end of oligo. e, Quantification of experiments such as shown in b-d and Fig. 1a with HELQ concentration ranging from 1-270 nM. n= 4 independent experiments; mean ± S.E.M. f-g, Representative native gels of DNA unwinding assay of dsDNA and 5'-overhang with indicated concentrations of HELQ. h, Native gel showing the DNA unwinding assay of 3'-overhang by HELQ in the presence of ATP (2 mM) and ATP?S (2 mM), a poorly hydrolysable ATP analogue (n=2). i, Native gel showing DNA unwinding assay of 3'-overhang with indicated concentrations of HELQ(K365M) (n =2). j, Electrophoretic mobility shift assay (EMSA) with ssDNA and indicated concentrations of HELQ and HELQ(K365M). The final products were resolved with native 6% polyacrylamide gels. k, Quantification of experiments such as shown in j. n= 3 independent experiments; mean ± S.E.M. l, EMSA with dsDNA and indicated concentrations of HELQ and HELQ(K365M). m, Quantification of experiments such as shown in l. n= 3 independent experiments; mean ± S.E.M.

Fig 5: HELQ strips RPA from ssDNA.a. Native gel (longer 6% polyacrylamide) showing RPA stripping assay with HELQ (10 nM) alone and with indicated concentrations of RPA. b, Native gel (longer 6% polyacrylamide) showing RPA stripping assay with indicated concentrations of HELQ in the absence and presence of RPA (40 nM). c, Schematic of immobilized dual labelled (Cy3 and Cy5) DNA in the absence of RPA. d–e, Representative intensity trajectory (top) and corresponding FRET trajectory (bottom) of dual labelled DNA in the absence of RPA. f, Time-binned FRET histogram of DNA only, fit with gaussian. g, Schematic of immobilized dual labelled DNA in the presence of RPA. h–i, Representative intensity trajectory (top) and corresponding FRET trajectory (bottom) of dual labelled DNA in the presence of RPA. Stable high FRET is observed. j, Time-binned FRET histogram of DNA in the presence of RPA. Fit with gaussian. k, Schematic of the experimental set up of single-molecule FRET-based RPA striping assay. DNA dual-labelled with the FRET pair Cy3 and Cy5 is immobilized on the microscope slide. In the absence of RPA, a short 6 nt sequence of homology causes the DNA to fold into a high FRET state. Upon RPA binding, the DNA unfolds resulting in a low FRET state. Addition of HELQ results in cycling between the low (bound) and high (free) FRET states as RPA is bound and removed respectively. l, Example of single-molecule fluorescence trajectory (top, Cy3 in blue, Cy5 in red) and corresponding FRET (bottom) showing the transition from low FRET (bound) to high FRET (free). m, Representative FRET trajectory of DNA in the presence of 1 nM RPA and 200 nM HELQ, spikes of high FRET correspond to RPA removal events. n, Plot of dwell time of RPA bound (toff), low FRET, state with increasing HELQ concentration. o, Plot of dwell time of free state (ton), high FRET, on RPA removal with increasing HELQ concentration. p, Representative FRET trajectory of DNA in the presence of 1 nM RPA and 200 nM HELQ KM. q, Representative gel of DNA annealing assay with HELQ (3 nM), RPA (40 nM) and indicated concentrations of RAD51. r, Quantification of experiments such as shown in q. n = 3 independent experiments; mean ± S.E.M.