Fig 2: Concomitant loss of RAD51AP1 and RAD54L results in increased replication stress and genome instability. (A) Schematic of the experimental protocol for the DNA fiber assay. (B) Median IdU tract length (green) after HU in HeLa, AP1 KO, R54L KO-1, and R54L/AP1 DKO-1 cells. Data points are from 100 to 150 fibers of three independent experiments each, with medians in red. (C) Median IdU tract length (green) under unperturbed conditions (NT). Data points are from 100 to 150 fibers of three independent experiments each, with medians in red. (D) Median CldU tract length (red) after HU in HeLa, AP1 KO, R54L KO-1, and R54L/AP1 DKO-1 cells. Data points are from 100 to 150 fibers of three independent experiments each, with medians in red. (E) Median CdU tract length (green) under unperturbed conditions (NT). Data points are from 100 to 150 fibers of three independent experiments each, with medians in red. *, p < 0.05; **, p < 0.01; ****, p < 0.0001; ns, non-significant; Kruskal-Wallis test followed by Dunn’s multiple comparisons test. (F) Schematic of the experimental protocol to induce chromosomal aberrations. (G) Representative micrographs of chromosomal aberrations after HU; radial (black arrow head), chromatid break (red arrow head) and chromatid gaps (blue arrow heads). (H) Aberrations per metaphase after HU in HeLa, AP1 KO, R54L KO-1, and R54L/AP1 DKO-1 cells. Data points are from 100 metaphases of two independent experiments each. Means (grey lines) ± SD (red lines) are shown. **, p < 0.01; ****, p < 0.0001; ns, non-significant; one-way ANOVA followed by Tukey’s multiple comparisons test.

Fig 3: Proposed model of RAD54 activity at ALT telomeresRAD54 functions downstream of pre-synaptic events to regulate the processing of D-loop structures (partial-X or PX junction), promoting telomere synthesis via BIR at ALT telomeres. RAD54 branch migration activity may function coordinately with BLM to promote BTR-dependent dissolution of recombination intermediates. Specifically, RAD54 may promote branch migration and ultimately dissolution through regions of heterology either in the D-loop or that arise during error-prone DNA extension events. Alternatively, RAD54 may function to dissolve heterologous recombination intermediates to promote formation of more homologous intermediates that will lead to higher fidelity telomere extension events. In the absence of RAD54, like in the absence of BLM, ALT cells rely on nucleolytic cleavage of recombination intermediates by the SMX resolvase complex.

Fig 4: RAD51 binds to the RAD51AP1/NCP complex, RAD51AP1 stimulates duplex capture with the NCP and is recruited to the chromatin fraction in human cells after DNA damage. A, schematic of the duplex capture assay with His6/FLAG-tagged RAD51AP1 and either nucleosome-free DNA (i.e., 147 bp dsDNA) or the NCP. B, qualitative analysis of captured DNA (beads) and DNA in supernatant by agarose gel electrophoresis. C, quantitative analysis: Symbols are the results from independent experiments. Bars are the means from three independent experiments ±1 SD; *, p < 0.05; **, p < 0.01; two-way ANOVA. D, schematic of the D-loop reaction with chromatinized pBluescript II SK (-) plasmid DNA. E, addition of RAD51AP1 (50, 100 and 200 nM; lanes 2–4) or RAD54 (100 and 200 nM; lanes 5–6) promotes the RAD51-mediated D-loop reaction on chromatinized DNA. F, quantification of the results. Symbols are the results from independent experiments. Bars are the means from two to four independent experiments ± 1SD. *, p < 0.05; **, p < 0.01; ***, p < 0.001; multiple t test analysis. G, agarose gel to show that wild-type RAD51AP1 (100 and 200 nM; lanes 2–3) promotes the RAD51-mediated D-loop reaction on chromatinized DNA, but RAD51AP1-K3WA (100 and 200 nM; lanes 4–5) and RAD51AP1-K7WA (100 and 200 nM; lanes 6–7) are unable to do so. H, western blots of fractionated extracts of the nuclei from HT1080 and U2OS cells without and after exposure to mitomycin C (MMC). The signals for ß-Actin and histone H3 serve as loading and fractionation control, respectively. RAD54 is shown for comparison purposes.

Fig 5: RAD54 localizes to ALT telomeres in response to DNA damage Combined IF and DNA FISH analysis of RAD54 (IF) and telomeres (FISH) in ALT and non-ALT cell lines. White arrows indicate RAD54 foci that colocalize with telomeres. Scale bars = 10 µm.Quantification of data in A. A cell was counted positive if it contained 1 or more colocalization event between RAD54 and the telomere. At least 100 cells were counted per cell line per repeat. For SaOS2, NOS, SJSA1, HeLa LT n = 3. For Cal72, U2OS, HuO9 n = 4. Values shown are mean ± SD.Combined IF and DNA FISH analysis of RAD54 (IF), PML (IF), and telomeres (FISH) in ALT cell lines SaOS2 and HuO9. Scale bars = 10 µm.Quantification of HuO9 data from C. RAD54 foci were detected using particle analysis in Fiji. Total RAD54 foci were counted. RAD54 foci were considered in APBs if they colocalized with both PML and telomere. At least 100 cells counted per repeat, n = 3. Values shown are mean ± SD.Western blot showing expression of TRF1-FOK1-mCherry fusion protein following doxycycline induction, both WT and D450A (nuclease dead). Cells were treated with 40 ng/ml doxycycline for 16 h and then 1 µM Shield-1 and 1 µM 4-OHT for 4 h.Combined IF and DNA FISH for RAD54 (IF) and telomeres (FISH) in U2OS-TRF1-FOK1-WT or U2OS-TRF1-FOK1-D450A (nuclease dead) cells. Cells were treated as in E immediately prior to staining. Scale bars = 10 µm.Quantification of data from F. RAD54 foci were detected using particle analysis in Fiji. At least a total of 300 cells were contained from 3 repeated experiments. Values shown are mean ± SEM. Values were compared using Kruskal–Wallis test followed by Dunn's multiple comparison test.Quantification of data from F. RAD54 foci were detected using particle analysis in Fiji. Cells were counted as positive if they counted 8 or more RAD54 foci that colocalized with telomeres. Values shown are mean ± SEM of n = 3 biological replicates. Values were compared using standard two-way ANOVA followed by Sidak's multiple comparison test.