Inherited deficiency of DIAPH1 identifies a DNA double strand break repair pathway regulated by γ-actin

Abstract

DNA double strand break repair (DSBR) represents a fundamental process required to maintain genome stability and prevent the onset of disease. Whilst cell cycle phase and the chromatin context largely dictate which repair pathway is utilised to restore damaged DNA, it has been recently shown that nuclear actin filaments play a major role in clustering DNA breaks to facilitate DSBR by homologous recombination (HR). However, the mechanism with which nuclear actin and the different actin nucleating factors regulate HR is unclear. Interestingly, patients with biallelic mutations in the actin nucleating factor DIAPH1 exhibit a striking overlap of clinical features with the HR deficiency disorders, Nijmegen Breakage Syndrome (NBS) and Warsaw Breakage Syndrome (WABS). This suggests that DIAPH1 may play a role in regulating HR and that some of the clinical deficits associated with DIAPH1 mutations may be caused by an underlying DSBR defect. In keeping with this clinical similarity, we demonstrate that cells from DIAL (DIAPH1 Loss-of-function) Syndrome patients display an HR repair defect comparable to loss of NBS1. Moreover, we show that this DSBR defect is also observed in a subset of patients with Baraitser-Winter Cerebrofrontofacial (BWCFF) syndrome associated with mutations in ACTG1 (γ-actin) but not ACTB (β-actin). Lastly, we demonstrate that DIAPH1 and γ-actin promote HR-dependent repair by facilitating the relocalisation of the MRE11/RAD50/NBS1 complex to sites of DNA breaks to initiate end-resection. Taken together, these data provide a mechanistic explanation for the overlapping clinical symptoms exhibited by patients with DIAL syndrome, BWCFF syndrome and NBS.

Fig. 1. Loss of DIAPH1 is associated with microcephaly, seizures, recurrent respiratory infections and visual impairment.

a Table showing a subset of DIAPH1 gene variants identified within our cohort of patients and whether they exhibit the primary clinical features of DIAL syndrome: microcephaly, seizures, recurrent respiratory infections and visual impairment. b Diagrammatic representation of the DIAPH1 protein highlighting individual domains. Recessive mutations in DIAPH1 identified within our cohort of patients for which cell lines were available are shown in red. Previously reported autosomal recessive variants in DIAPH1 associated with SCBMS are shown in blue. Dominant mutations in DIAPH1 associated with hearing loss and/or thrombocytopenia are shown in black. FH1-2 Formin Homology domains 1-2, GBD GTPase binding domain, CC coiled-coil, DID DIAPH1 inhibitory domain, DAD DIAPH1 activation domain. c–e Western blots on cell extracts derived from DIAL syndrome patients indicating the level of DIAPH1 protein expression. c, d (Fibroblasts) WT1 and SW are fibroblasts from unrelated healthy controls. e (Lymphoblastoid cell lines - LCLs) SC and NBS are LCLs from a healthy control and a patient with Nijmegen Breakage Syndrome (NBS) carrying a homozygous c.657del5 mutation in NBN, respectively. KAP1 was used as a protein loading control. c–e n = 1 independent experiments. Source data are provided as a Source Data file.

Fig. 2. Cells lacking DIAPH1 display an inability to repair DSBs.

a WT (WT1) and DIAL syndrome fibroblasts (DIAPH1-P1) were mock treated or irradiated with 3 Gy of IR and harvested for Western blotting at time points post-irradiation. Cell extracts were separated by SDS-PAGE and Western blotting was carried out with the antibodies indicated. Representative of n = 2 independent experiments. b DIAPH1-P1 fibroblasts complemented with either an empty vector (P1 + vector) or HA-tagged WT DIAPH1 (P1 + DIAPH1) were treated as in (a). Representative of n = 2 independent experiments. c, d Quantification of γ-H2AX foci in complemented DIAPH1-P1 fibroblasts before and after exposure to 3 Gy IR. γ-H2AX foci were visualised by immunofluorescence microscopy and quantified in a minimum of 500 cells were counted per time point, per experiment. The mean of n = 3 independent experiments is shown with the SEM. Representative images of the cells at different time points before and post-irradiation are shown (c). The scale bars represent 10 μm. e, f Micronuclei were quantified from cells described in (c, d) before and after exposure to 3 Gy IR (24 h), 100 nM CPT (1 h exposure followed by 48 h recovery) and 10 μM ETOP (30 min exposure followed by 48 h recovery). The mean of n = 3 independent experiments is shown with the SEM. A minimum of 500 cells were counted per time point, per experiment. Representative images of the cells at different time points before and after treatment are shown (e). The scale bars represent 10 μm. g Quantification of chromosome aberrations in a WT lymphoblastoid cell line (LCL) (Normal - SC), DIAL syndrome LCL (DIAPH1-P5) and NBS LCL (NBS) before and 24 h after 1 Gy of IR, 0.5 nM CPT or 10 nM ETOP. Chromosome aberrations includes chromatid/chromosome gaps/breaks, chromatid/chromosome fragments and chromosome radials/exchanges. Representative images of each type of aberration are shown. The mean of n = 3 independent experiments is shown with the SEM. A minimum of 50 metaphases were counted per cell line in each experiment. Statistical significance was calculated using: (d) a two-way ANOVA with multiple comparisons ([untreated] p = 0.278, [24 h post-IR] p = 0.0116), (f) an unpaired Student’s t-test (two-sided, equal variance) and (g) an ordinary one-way ANOVA with multiple comparisons. Source data are provided as a Source Data file.

Fig. 3. Cells lacking DIAPH1 exhibit an inability to repair DNA DSBs by HR.

a Complemented DIAPH1-P1 fibroblasts were irradiated with 1 Gy of IR and then fixed/permeabilised at the time points indicated. γ-H2AX foci were quantified in G1-phase (mitosin negative) and G2-phase (mitosin positive) cells at 1 h and 8 h post-irradiation. The mean number of foci per cell of n = 3 independent experiments is shown (red line). Foci in a minimum of 100 cells were counted per time point, per experiment. b Quantification SCEs in WT, DIAL syndrome and NBS LCLs by CPT (0.5 nM) or ETOP (10 nM). The mean of n = 3 independent experiments is shown (red line). Approximately 30 metaphases for each cell line were scored per experiment. c siRNA transfected HeLa cells were transfected with the plasmids required to initiate HR at a Cas9-induced DSB located within the LMNA locus. HR (% RFP positive cells with mClover-LMNA staining) was quantified using immunofluorescence. The mean of n = 3 independent experiments is shown with the SEM. A minimum of 500 cells were counted per time point, per experiment. Representative images of transfected cells are shown. The scale bars represent 10 μm. d–f Complemented DIAPH1-P1 fibroblasts were treated with 100 nM CPT for 1 h and then fixed/permeabilised 1 h and 4 h post-treatment. Cells were stained with antibodies to (d) BRCA1, (e) RPA2 and (f) RAD51 in conjunction with CENPF/mitosin as a marker of S/G2 cells. BRCA1 and RAD51 foci and RPA2 fluorescence intensity were quantified in a minimum of 100 S/G2 cells per timepoint, per experiment. The mean of n = 3 independent experiments with the SEM is shown (red line). Statistical significance was calculated using: (a) an ordinary one-way ANOVA with multiple comparisons, (b) a Kruskal–Wallis test ([untreated] p = 0.7703, [24 h post-CPT] p = <0.0001, [24 h post-ETOP] p = <0.0001), (c) one-way ANOVA with multiple comparisons, (d) a two-way ANOVA with multiple comparisons and (e, f) a Kruskal–Wallis test ([e] p = <0.0001, [f] p = <0.0001). Source data are provided as a Source Data file.

Fig. 4. DIAPH1 binds to and promotes recruitment of the MRN complex to sites of DSBs.

a, b Complemented DIAPH1-P1 fibroblasts were treated with 100 nM CPT for 1 h, 30 min into treatment cells were exposed to 10 μM EdU for 30 min and then fixed/permeabilised 1 h and 4 h post-treatment. Cells were stained with the antibodies indicated and protein localisation visualised by STORM microscopy. Quantification of the average local density of MRE11, BRCA1 and RPA2 detected around individual EdU molecules within a region of interest (ROI) was carried out and normalised to the untreated controls. A minimum of 100 ROIs were quantified over three independent experiments. The median of n = 3 independent experiments is shown (red line). Representative images of the cells treated with CPT are shown. The scale bar represents 1500 nm. c U-2 OS-FokI cells were transfected with the indicated siRNAs and then fixed/immunostained with an antibody to MRE11 4 h after FokI induction. The fluorescence intensity of MRE11 per FokI focus was quantified from 100 cells per experiment. The median of n = 3 independent experiments is shown (red line). d Quantification of resection of an AsiS1-induced DSB on chromosome 1 at a position 335 bp and 1618 bp away from the DSB using qPCR in siRNA transfected U-2 OS cells. The mean of n = 3 independent experiments is shown with the SEM. e siRNA transfected HeLa cells were labelled with 25 μM BrdU for 24 h, treated with 10 Gy IR for 1 h and then permeablised/fixed 4 h post-irradiation. Cells were stained with an anti-BrdU antibody and CENPF (as a marker of S/G2 cells) and the intensity of nuclear BrdU staining was quantified in 100 S/G2 cells per experiment. Cells were treated with 50 μM mirin 1 h prior to irradiation where indicated. The mean of n = 3 independent experiments is shown (red line). Statistical significance was calculated using: (a, b, e) a Kruskal–Wallis test ([a] p = 0.0034, [b] p = <0.0001, [e] p = <0.0001) and (c, d) an unpaired Student’s t-test (two-sided, equal variance). Source data are provided as a Source Data file.

Fig. 5. DIAPH1 is recruited to DSBs.

a (Top) Representative images of the recruitment of MRE11 and DIAPH1 to FokI-induced DNA DSBs in U-2 OS cells. The scale bars represent 10 μm. (Bottom) Quantification of the percentage of cells with FokI foci that colocalise with MRE11 or DIAPH1 from n = 3 independent experiments. b Complemented DIAPH1-P1 fibroblasts were treated with 10 μM EdU for 30 min, exposed to 100 nM CPT for 1 h and then permeabilised/fixed. EdU positive cells were labelled with Alexa-fluor-488 using click chemistry. Cells were then subjected to a PLA reaction using antibodies to DIAPH1 and γH2AX. (Top) Representative images of the PLA reaction (red spots) in EdU positive (green nuclei) and negative (blue) nuclei stained with DAPI. (Bottom) PLA spots were quantified in at least 50 EdU-positive and 50 EdU-negative cells per experiment, per cell line, per condition. The mean number of PLA spots per cells from n = 3 independent experiments is shown. c GFP or GFP-DIAPH1 was purified from 293FT cell extracts using GFP-Trap and co-purified proteins were subjected to SDS-PAGE and Western blotting with the antibodies indicated. Representative of n = 2 independent experiments. d Quantification of the percentage of DIAPH1 positive FokI foci in U-2 OS cells depleted of either MDC1 or MRE11 or treated with either 5 μM ATM inhibitor (ATMi) or 50 μM mirin. The mean number of DIAPH1 positive FokI foci per cell from n = 3 independent experiments is shown. Statistical significance was calculated using: (a) an unpaired Student’s t-test (two-sided, equal variance), (b) a Kruskal–Wallis test (p = <0.0001) and (d) an ordinary one-way ANOVA. Source data are provided as a source data file.

Fig. 6. Mutations in ACTG1 and components of the Arp2/3 complex but not ACTB are associated with a DNA DSB repair defect.

a Sequence alignment of the first 70 amino acids of human β- and γ-actin proteins. b, c Quantification of γ-H2AX foci (b) and micronuclei (c) in patient-derived fibroblasts with de novo mutations in ACTB or ACTG1 before and after a 1 h exposure to 100 nM CPT. γ-H2AX foci and micronuclei were visualised by immunofluorescence microscopy and quantified in cells 24 h and 48 h following the removal of the CPT, respectively. The mean of n = 3 independent experiments is shown with the SEM. A minimum of 500 cells were counted per time point, per experiment. d, e Quantification of γ-H2AX foci (d) and micronuclei (e) in DIAL syndrome patient P1-derived fibroblasts complemented with either an empty vector, WT DIAPH1 or an actin nucleation defective mutant (I862A) before and after a 1 h exposure to 100 nM CPT. γ-H2AX foci and micronuclei were visualised by immunofluorescence microscopy and quantified in untreated cells and cells 1 h, 24 h and 48 h following the removal of the CPT. The mean of n = 3 independent experiments is shown with the SEM. A minimum of 500 cells were counted per time point, per experiment. f, g Quantification of γ-H2AX foci (f) and micronuclei (g) in patient-derived fibroblasts with mutations in ARPC1B, ARPC4 or ARPC5 before and after exposure to CPT (as in b, c). Fibroblasts derived from patients with Nijmegen Breakage Syndrome (NBS) and Ataxia-Telangiectasia (A-T) were used as controls for cells with a DNA DSB repair defect. The mean of n = 3 independent experiments is shown with the SEM. A minimum of 500 cells were counted per time point, per experiment. Statistical significance was calculated using: (b–g) an ordinary one-way ANOVA ([b] p = <0.0001, [c] p = <0.0001, [f] p = <0.0001, [g] p = <0.0001). Source data are provided as a Source Data file.

Fig. 7. DIAPH1, γ-actin and the Arp2/3 complex function to promote MRE11-dependent localisation to DSBs and DNA end-resection.

a GFP or GFP-DIAPH1 was purified from 293FT cell extracts using GFP-Trap and co-purified proteins were subjected to SDS-PAGE and Western blotting with the antibodies indicated. Representative of n = 2 independent experiments. b U-2 OS-FokI cells were transfected with the indicated siRNAs and then fixed/immunostained with an antibody to MRE11 4 h after FokI induction. The fluorescence intensity of MRE11 per FokI focus was quantified from 100 cells per experiment. The median of n = 3 independent experiments is shown (red line). c siRNA transfected HeLa cells were labelled with 25 μM BrdU for 24 h, treated with 10 Gy IR for 1 h and then permeablised/fixed 4 h post-irradiation. The intensity of nuclear BrdU staining was quantified in 100 S/G2 cells per experiment. Cells were treated with 50 μM mirin 1 h prior to irradiation where indicated. The mean of n = 3 independent experiments is shown (red line). d HeLa cells were transfected with the indicated siRNAs, exposed to 100 nM CPT for 1 h and then fixed/permeabilised 4 h post-treatment. RAD51 foci were quantified in a minimum of 100 S/G2 cells per experiment. The mean of n = 3 independent experiments with the SEM is shown (red line). e Quantification of CPT-induced RAD51 foci in DIAL syndrome patient P1-derived fibroblasts complemented with either an empty vector, WT DIAPH1 or the I862A DIAPH1 mutant. RAD51 foci were quantified in a minimum of 100 S/G2 cells per experiment. The mean of n = 3 independent experiments with the SEM is shown (red line). f siRNA transfected HeLa cells were transfected with the plasmids required to initiate HR at a Cas9-induced DSB located within the LMNA locus. HR (% RFP positive cells with mClover-LMNA staining) was quantified using immunofluorescence. The mean of n = 3 independent experiments is shown with the SEM (red line). A minimum of 500 cells were counted per time point, per experiment. Statistical significance was calculated using: (b–e) a Kruskal–Wallis test ([b] p = <0.0001, [c] p = <0.0001, [d] p = <0.0001, [untreated] p = 0.0044, [4 h post-CPT] p = <0.001) and (f) an ordinary one-way ANOVA (p = <0.0001). Source data are provided as a Source Data file.