The MMS22L-TONSL complex mediates recovery from replication stress and homologous recombination

Abstract

Genome integrity is jeopardized each time DNA replication forks stall or collapse. Here we report the identification of a complex composed of MMS22L (C6ORF167) and TONSL (NFKBIL2) that participates in the recovery from replication stress. MMS22L and TONSL are homologous to yeast Mms22 and plant Tonsoku/Brushy1, respectively. MMS22L-TONSL accumulates at regions of ssDNA associated with distressed replication forks or at processed DNA breaks, and its depletion results in high levels of endogenous DNA double-strand breaks caused by an inability to complete DNA synthesis after replication fork collapse. Moreover, cells depleted of MMS22L are highly sensitive to camptothecin, a topoisomerase I poison that impairs DNA replication progression. Finally, MMS22L and TONSL are necessary for the efficient formation of RAD51 foci after DNA damage, and their depletion impairs homologous recombination. These results indicate that MMS22L and TONSL are genome caretakers that stimulate the recombination-dependent repair of stalled or collapsed replication forks.

Figure 1. MMS22L prevents spontaneous DNA damage during S phase

(A) Representative immunoblotting of HeLa cells transfected with either control (siCTRL) or MMS22L SMARTpool (siMMS22L). Arrow indicates specific MMS22L band. Asterisk indicates a nonspecific immunoreactive protein that serves as an internal loading control. M, molecular size standards. (B) Spontaneous 53BP1 focus formation correlates with efficiency of MMS22L knockdown. HeLa cells were transfected with either control or MMS22L siRNAs. 48 h post-transfection cells were fixed and processed for anti-53BP1 immunofluorescence (red) and DAPI staining (blue). An immunoblot documenting the extent of MMS22L depletion is found in Figure S2A. (C) Quantitation of (B). At least 100 cells per condition were counted. Data are represented as the mean ± SD (n=3). (D) U2OS cells were transfected either with control, MMS22L or CHK1 siRNAs. 48 h post-transfection, cells were treated either with DMSO or with 5 μM PHA767491 for 6 h, fixed and processed for 53BP1 immunofluorescence. Foci-positive cells were defined as cells with more than 5 53BP1 foci. At least 100 cells per condition were counted. Data are represented as the mean ± SD (n=3). (E) FACS profiles of hTERT RPE-1 cells transfected with control or MMS22L siRNAs. The proportion of cells in G1, S and G2 phases of the cell cycle are indicated. PI, propidium iodide. (F) HeLa cells were transfected with either control or MMS22L siRNAs. 48 h post-transfection cells were fixed and processed for anti-phospho-histone H3 immunofluorescence to identify mitotic cells. At least 100 cells per condition were counted. Data are represented as the mean ± SD (n=3). (G) Depletion of MMS22L results in checkpoint activation. Whole cell extracts of HeLa cells transfected with either control or MMS22L siRNAs were analyzed by immunoblotting with antibodies to CHK1 phosphoserine 317 (pS317), total CHK1 (CHK1), MMS22L or tubulin (loading control).

Figure 2. MMS22L interacts with TONSL/NFKBIL2

(A) MMS22L interactions detected by IP-MS (Table S2) grouped according to known biological associations. The strength of interaction is indicated by the boldness of the arrow. Arrows pointing between the different groups indicate a reported biological interaction. (B) Domain structure of NFKBIL2/TONSL. TPR, tetratricopeptide repeat; ANK, ankyrin repeat; UBL, ubiquitin-like domain; LRR, leucine-rich repeat; aa, amino acid. (C) Representative immunoblotting of TONSL knockdown. U2OS whole cell extracts were analyzed by immunoblotting with antibodies to TONSL or actin (loading control). Arrow indicates specific TONSL band. M, molecular size standards. (D) Endogenous MMS22L and TONSL co-immunopreciptiate (IP) each other. A control immunoprecipitation with a non-specific rabbit IgG fraction (IgG) was performed in parallel. Immunoprecipitated proteins were subjected to immunoblot analysis with the indicated antibodies. (E) TONSL interaction network (Table S2) as described in (A). (F) Whole cell lysates (WCE) of 293T cells transiently expressing GFP-TONSL and/or MMS22L-mFlag were subjected to immunoprecipitation with anti-GFP antibodies followed by immunoblot analysis with the indicated antibodies. Arrows and numbers at the right indicate the migration of different MCM proteins. M, molecular size standards.

Figure 3. MMS22L and TONSL are required for recovery from replication stress

(A) FACS profiles of RPE-1 hTERT cells 48 h post-transfection with control (siCTRL) or TONSL (siTONSL) siRNAs. (B) TONSL depletion in HeLa cells results in spontaneous 53BP1 foci. 48 h post-transfection with the indicated siRNAs, cells were fixed and processed for anti-53BP1 immunofluorescence and DAPI staining. Quantitation of 53BP1 foci are shown for each siRNA tested in the accompanying histogram. At least 100 cells per condition were counted. Data are represented as the mean ± SD (n=3). (C) Immunoblot analysis of transfections shown in (B). Blots of whole cell extracts were probed with antibodies to TONSL and tubulin (loading control). (D) Clonogenic survival assays in MMS22L-depleted cells following X-irradiation (IR) or treatment with camptothecin (CPT). Cells were then permitted to grow for 14 d before fixation and staining. Colonies with >50 cells were counted. The surviving fraction is represented as the mean ± SD (n=3). (E) Cell cycle profiles of HeLa cells recovering from a 30 min incubation with 0.1 μM CPT. Cells were transfected with control or MMS22L siRNAs 48 h prior to DMSO or CPT treatment. (F) HeLa cells were transfected with control or MMS22L siRNAs. 48 h post-transfection, cells were treated with DMSO (-) or 0.1 μM CPT for 30 min and then released into fresh medium for the indicated times. Whole cell extracts were analyzed by immunoblotting with antibodies to MMS22L, CHK1 phosphoserine 317 (pS317), total CHK1 (CHK1) or actin (loading control).

Figure 4. MMS22L promotes DNA replication on damaged templates

(A) Above is a schematic of the fork velocity experiment. Below are representative images of single DNA fibers of HeLa cells transfected with control (siCTRL) or MMS22L (siMMS22L) siRNAs. Arrows indicate direction of DNA synthesis. Green, CldU; red, IdU; blue, DNA. Scale bar: 50 kb. (B) Box plot representation of IdU tract lengths in cells transfected with control and MMS22L siRNAs. The number within each box indicates the median of the result. Each box plot consists of data from three independent experiments and represents a total of 160 measurements. p-values between the indicated samples were calculated using a non-parametric Mann-Whitney test. (C) Fork velocities derived from the experiment depicted in A-B. Data are represented as the mean ± SD. (D) Above is a schematic of the CPT recovery experiment. Below are representative images of single DNA fibers from the fork recovery experiment. Arrows indicate direction of DNA synthesis. Colors are as in (A). (E) Box plot representation of IdU tract lengths in cells transfected with control and MMS22L siRNAs. The number within each box indicates the median result. Each box plot consists of data from 2 independent experiments and represents a total of 200 measurements. p-values were calculated as in (B).

Figure 5. MMS22L and TONSL accumulate at sites of DNA damage

(A-B) U2OS cells expressing MMS22L-mFlag (A) or unaltered U2OS cells (B) were microirradiated with a UVA laser and processed for RPA32 and Flag (MMS22L; A) or TONSL (B) immunofluorescence. Dashed lines indicate nuclei as determined by DAPI staining (not shown). Scale bar: 10 μm. (C) Quantitation of the percentage of cells with a γ-H2AX stripe that also contained a MMS22L or TONSL stripe after transfection with the indicated siRNAs. The experiments were done with U2OS cells stably expressing MMS22L-mFlag (for the Flag staining) or unaltered U2OS cells for the endogenous (endo) TONSL and MMS22L staining. At least 80 cells per condition were counted. See Figure S6A-C for representative micrographs. (D) U2OS cells stably expressing MMS22L-mFLag (top) or unaltered U2OS cells (bottom) were transfected either with control (siCTRL), MMS22L (siMMS22L) or TONSL (siTONSL) siRNAs. 48 h after transfection, cells were treated with 1 μM CPT for 3 h before being fixed and processed for myc (MMS22L; top) or TONSL (bottom) immunofluorescence. Dashed lines indicate nuclei as in (A). Scale bar: 10 μm. (E) U2OS cells stably expressing MMS22L-mFlag (top) or unaltered U2OS (bottom) were treated with 1 μM CPT for 3 h before being fixed and processed for RPA32 and Flag (MMS22L; top) or TONSL (bottom) immunofluorescence. Dashed lines indicate nuclei as in (A). Scale bar: 10 μm.

Figure 6. MMS22L and TONSL colocalize at stalled replication forks

(A) U2OS cells were treated with DMSO or the indicated DNA damaging agents (1 μM CPT for 3 h, 3mM HU for 2 h, 0.02% MMS for 2 h, 50 J/m² UV-irradiation followed by 2 h recovery). Cells were then fixed and processed for TONSL and RPA32 immunofluorescence. Dashed lines indicate nuclei as determined by DAPI staining (not shown). Scale bar: 10 μm. (B) U2OS cells stably expressing Flag-tagged RAD51 were transfected either with control (siCTRL), TONSL (siTONSL) or MMS22L (siMMS22L) siRNAs. 48 h after transfection, cells were treated with 1 μM CPT for 3 h before being fixed and processed for Flag immunofluorescence. Dashed lines indicate nuclei as in (A). Scale bar: 10 μm. (C) Quantitation of Flag-RAD51 foci from the experiment described in (A) and from an identical experiment done with 10 Gy ionizing radiation instead of CPT. Foci-positive cells were defined as cells with more than 10 Flag-RAD51 foci. At least 100 cells per condition were counted. Data are represented as the mean ± SD (n=3). (D) Quantitation of RPA foci in U2OS cells treated with 0.5 μM CPT for 30 min (t=0) and then released from the treatment and collected at the indicated time points. Representative images are shown in Figure S7B. (E) U2OS cells were transfected with control, MMS22L or TONSL siRNAs. 48 h after transfection, cells were irradiated with 10 Gy and 4 h later were fixed and processed for BRCA2 and γ-H2AX immunofluorescence. Insets are magnifications of the indicated fields. Dashed lines indicate nuclei as in (A). Scale bar: 10 μm. (F) Quantitation of BRCA2 foci from the experiment described in (E). Foci-positive cells were defined as cells with more than 5 BRCA2 foci. At least 75 cells per condition were counted. Data are represented as the mean ± SD (n=3).

Figure 7. The MMS22L-TONSL complex promotes homologous recombination

(A) Effect of MMS22L or TONSL depletion on homologous recombination. HeLa DR-GFP cells were transfected with the indicated siRNAs, and 48 h later were transfected with or without an I-SceI expression plasmid. 48 h after transfection, cells were harvested and assayed for GFP expression by FACS analysis. Data are represented as the mean ± SD (n=3). (B) Effect of MMS22L depletion on sister chromatid exchange in BLM deficient cells (PSNG13; BLM^-/-) and isogenic complemented cells expressing BLM cDNA (PSNF5; BLM⁺) transfected with control (siCTRL) or MMS22L siRNAs. At least 40 metaphase cells were counted for each condition. Statistical significance was calculated with a Mann-Whitney test. (C) Bloom syndrome cells (BLM^−/−) and isogenic complemented cells expressing BLM cDNA (BLM⁺) were first transduced with a lentivirus expressing either GFP or RFP. The resulting fluorescent protein-marked BLM^−/− and BLM⁺ lines were plated at a 1:1 ratio and were transfected with control, MMS22L, or TONSL siRNAs. 96 h post-transfection cells were fixed and imaged with a high-content microscope to determine the BLM^−/− to BLM⁺ ratio. Data was normalized to the control siRNA condition to eliminate proliferation and transfection differences between cell lines. The data represents the mean +/− SEM (n=2). (D) The TONSL LRR and TPR repeats are required for RAD51 focus formation. Top: Schematic representation of the domain architecture and physical interactions of TONSL. U2OS cell lines containing stable integrations of the indicated TONSL cDNA were transfected with a TONSL siRNA (#1) and were either left uninduced (−) or were induced (+) by tetracycline (Tet) addition for 24 h prior to a 10 Gy IR dose. 3h later, the cells were fixed and processed for RAD51 immunofluorescence. The number of cells with more than 5 RAD51 foci is presented as the percentage deviation from the wild type, induced condition.