Fig 2: FUS forms a complex with PARP1, XRCC1, and LigIII. a IB of Flag-FUS co-Immunoprecipitation (co-IP) from HEK293 cells for oxidative DNA damage repair proteins with or without GO treatment (100 ng/ml for 1 h). The IP was performed with anti-Flag antibody. b IB of endogenous FUS co-IP from SH-SY5Y cells for PARP1, XRCC1, and LigIII. The IP was performed with anti-FUS antibody. c Scheme of human iPSC line differentiation to motor neurons. Representative phase-contrast images of iPSC and differentiated motor neurons are shown. Scale bar = 50 µm. d Immunofluorescence (IF) staining of motor neuron markers. Representative images of motor neurons that stained Isl-1, MAP2, or βIII-tubulin indicated ~80% differentiation efficiency. Scale bar = 50 µm. e IB of endogenous FUS co-IP from differentiated motor neurons for PARP1, XRCC1, and LigIII. f Quantitation of IB band intensity in Fig. 2b, e to show level change of PARP-1, XRCC1, and LigIII in FUS IP after GO treatment. The error bars are standard deviation of experiments performed in triplicate. g PLA of FUS vs PARP1, XRCC1, or LigIII in SH-SY5Y and iPSC-derived motor neurons with or without GO treatment. Nuclei stained with DAPI. Scale bar = 5 µm. h Quantitation of PLA foci from 25 motor neuronal cells (**p < 0.01, two-tailed unpaired Student’s t-test). The error bars are standard deviation

Fig 3: Mutant FUS induces DNA ligation defects to inhibit oxidized damage repair. A model showing the involvement of FUS for optimal DNA nick ligation in healthy neurons to facilitate efficient oxidative genome damage repair, and how loss of functional FUS in ALS leads to DNA nick ligation defects. FUS is required for PARP-1-dependent recruitment of XRCC1/LigIII complex at oxidative DNA damage sites. Two familial FUS mutations impair DNA nick ligation by distinct mechanisms. Substantial cytoplasmic localization of P525L FUS causes DNA ligation defect due to loss of functional FUS from nucleus. The R521H/C FUS fails to form the repair complex with XRCC1/LigIII and PARP-1 and is not recruited at damage site. Unrepaired DNA SSBs caused by impaired DNA nick ligation, together with secondary DSBs that may be generated from closely placed single-strand breaks, may substantially contribute to neurodegeneration in familial FUS-ALS patients

Fig 4: FUS facilitates PARP-1-dependent recruitment of XRCC1/LigIII. a, b Laser ablation microscopy for recruitment of GFP-XRCC1 after FUS knockdown. GFP-XRCC1 and FUS shRNA were co-transfected into HEK293 cells followed by laser treatment and microscopy 48 h after the transfection, as in Fig. 7a. Scale bar = 5 µm. Histogram shows the intensity of GFP-XRCC1 at laser track (see movie clip in Supplementary Movie 4 and 5 for live imaging). The error bars are standard deviation. c, d PLA of PARP1 vs XRCC1 and PARP-1 vs LigIII in WT or FUS mutant patient-derived motor neurons after GO treatment. Nuclei stained with DAPI. The average number of PLA foci from 25 cells were quantified. The error bars are standard deviation (*p < 0.05; **p < 0.01, two-tailed unpaired Student’s t-test). Scale bar = 5 µm. e, f XRCC1 co-IP (endogenous) from WT or FUS mutant patient-derived fibroblast lines and IB probed for PARP-1 and XRCC1. Histogram shows quantitation of IB band intensity. The error bars are standard deviation of experiments performed in triplicate (*p < 0.05, two-tailed unpaired Student’s t-test). g A model showing that PARP-1-dependent recruitment of XRCC1/LigIII at DNA SSBs is facilitated by FUS

Fig 5: Correction of FUS mutations rescue DNA ligation defects. a Scheme of correction of FUS mutation in familial ALS patient-derived iPSC lines by CRISPR knock-in to generate isogenic control lines and their differentiation to motor neurons. b IF motor neurons for FUS to analyze its nucleus vs. cytosolic distribution. Nuclei stained with DAPI. Zoomed images on right showing FUS distribution in arrow indicated nuclear. Scale bar = 10 µm. c, d In vitro DNA ligation assay. XRCC1 IP complex from nuclear extract of mutant and isogenic control motor neurons and quantified (**p < 0.01; ***p < 0.001, two-tailed unpaired Student’s t-test). e LA-PCR-based DNA damage repair kinetic analysis. Genome DNA extracted from iPSC-derived motor neurons at indicated time points after release from exposure to GO (100 ng/ml) for 1 h. Amplification products analyzed by pico green-based quantitation. All error bars are standard deviation of experiments performed in triplicate