Human RNF169 is a negative regulator of the ubiquitin-dependent response to DNA double-strand breaks

Abstract

Nonproteolytic ubiquitylation of chromatin surrounding deoxyribonucleic acid double-strand breaks (DSBs), mediated by the RNF8/RNF168 ubiquitin ligases, plays a key role in recruiting repair factors, including 53BP1 and BRCA1, to reestablish genome integrity. In this paper, we show that human RNF169, an uncharacterized E3 ubiquitin ligase paralogous to RNF168, accumulated in DSB repair foci through recognition of RNF168-catalyzed ubiquitylation products by its motif interacting with ubiquitin domain. Unexpectedly, RNF169 was dispensable for chromatin ubiquitylation and ubiquitin-dependent accumulation of repair factors at DSB sites. Instead, RNF169 functionally competed with 53BP1 and RAP80-BRCA1 for association with RNF168-modified chromatin independent of its catalytic activity, limiting the magnitude of their recruitment to DSB sites. By delaying accumulation of 53BP1 and RAP80 at damaged chromatin, RNF169 stimulated homologous recombination and restrained nonhomologous end joining, affecting cell survival after DSB infliction. Our results show that RNF169 functions in a noncanonical fashion to harness RNF168-mediated protein recruitment to DSB-containing chromatin, thereby contributing to regulation of DSB repair pathway utilization.

Figure 1.

RNF169 accumulates in DSB repair foci in an RNF8/RNF168-dependent manner. (A) Schematic depiction of human RNF168 and RNF169 proteins, showing the location of conserved domains (amino acid residues are bracketed). (B) U2OS cells transiently transfected with the HA-RNF169 plasmid for 24 h were subjected or not subjected to IR (4 Gy), fixed 1 h later, and coimmunostained with HA and MDC1 antibodies. (C) U2OS cells stably expressing GFP-RNF169 were subjected or not subjected to microlaser irradiation and fixed 1 h later. (D) HeLa cells transfected or not transfected with RNF169 siRNA for 72 h were fractionated and immunoblotted with the indicated antibodies. (E) U2OS cells transfected with siRNAs for 48 h and with a HA-RNF169 plasmid for an additional 24 h were subjected or not subjected to IR (4 Gy), fixed 1 h later, and immunostained with the HA antibody. siCTRL, control siRNA. Bars, 10 µm.

Figure 2.

RNF169 recognizes RNF168-catalyzed ubiquitin structures at DSB sites via its MIU domain. (A) Alignment of the potential MIU motif in human RNF169 with human RNF168 MIU-2 and the signature MIU motif (Penengo et al., 2006). The RNF169 mutation disrupting the functional integrity of its MIU domain is indicated in red. Bold letters show residues of the MIU domain consensus motif. #, acidic residue; ϕ, large hydrophobic residue. N, N terminus; C, C terminus. (B) U2OS cells were transfected with the indicated HA-RNF169 constructs for 24 h, fixed, and immunostained with HA and MDC1 antibodies. (C) Quantification of data in B. At least 200 cells were counted for each treatment. (D) Lysates from HEK293T cells transfected with the indicated RNF169 expression plasmids or empty vector (−) for 24 h were subjected to Strep-Tactin pull-down under denaturing conditions, washed, and incubated with ubiquitin chains. Bound complexes were immunoblotted with ubiquitin and FLAG antibodies. (E) U2OS cells transfected with RNF168 siRNA for 48 h were transfected with siRNA-resistant GFP-RNF168 expression constructs for an additional 24 h and then subjected to IR (4 Gy) and fixed 1 h later. Cells were immunostained with the HA antibody. (F) U2OS cells cotransfected with GFP-RNF169 and WT or catalytically inactive (CI) Myc-USP3 constructs for 24 h were subjected to IR (4 Gy), fixed 1 h later, and immunostained with the Myc antibody. (G) Quantification of data in F. At least 200 cells were counted for each treatment. Results depict the means (±SD) of three independent experiments. IB, immunoblot; MM, molecular mass. Bars, 10 µm.

Figure 3.

RNF169 is dispensable for ubiquitin-dependent assembly of repair factors at DSB-modified chromatin. (A) U2OS cells transfected with the indicated siRNAs for 72 h were subjected or not subjected to IR (4 Gy), fixed 1 h later, and coimmunostained with 53BP1 and MDC1 or γ-H2AX antibodies. Bars, 10 µm. (B) Immunoblot analysis of the experiment in A. The asterisk denotes a cross-reactive band. (C) HEK293T cells were transfected with the indicated combinations of plasmids for 24 h. To analyze histone H2A ubiquitylation, FLAG immunoprecipitates were immunoblotted with the Myc antibody. (D) Ubiquitylation reactions containing recombinant H2A, HA-ubiquitin, E1, and E2 (UbcH5a) were incubated in the presence or absence of purified His₆-RNF168 or His₆-RNF169 for 1 h and processed for immunoblotting with HA, H2A, and His₆ antibodies. (E) Reactions containing the indicated components were processed as in D and immunoblotted with the HA antibody to visualize RNF168 and RNF169 autoubiquitylation. CTRL, control; IP, immunoprecipitation; MM, molecular mass; Ub, ubiquitin; WCE, whole-cell extract; WB, Western blot.

Figure 4.

RNF169 negatively regulates 53BP1 and RAP80 recruitment to DSBs to affect repair pathway choice. (A) U2OS cells transfected with GFP-RNF169 expression plasmids for 24 h were exposed to IR (4 Gy), fixed 1 h later, and immunostained with the 53BP1 antibody. Arrows depict cells in which 53BP1 focus formation is suppressed. (B) Quantification of data in A. At least 200 cells were counted for each treatment. Results depict the means (±SD) of three independent experiments. (C) U2OS cells transfected as in A were microlaser irradiated, fixed 1 h later, and coimmunostained with 53BP1 and RAP80 antibodies. (D) U2OS cells transfected and exposed to IR as in A were fixed and immunostained with the ubiquitin conjugate (FK2) antibody. (E) U2OS cells transfected with control (CTRL) or RNF169 siRNAs (see also Fig. S2 C) for 72 h were exposed to IR (1 Gy), fixed 30 min later, and processed for 53BP1 immunostaining. Representative images acquired with identical microscope settings are shown. (F) U2OS/DR-GFP cells were transfected with the indicated siRNAs for 48 h and then cotransfected with plasmids encoding I-SceI and RFP for 48 h. Flow cytometry analysis of the GFP/RFP ratio was used to measure HR efficiency. Data represent the means (±SD) of three independent experiments. (G) U2OS/DR-GFP cells were transfected with empty vector or RNF169 plasmids for 24 h and processed as in F. (H) Clonogenic survival of U2OS/S-FLAG-Strep–RNF169 WT cells induced or not induced with doxycycline (DOX) for 24 h and exposed to the indicated doses of IR. Results depict the means (±SD) of three replicates from one representative experiment. (I) As in H, using U2OS/S-FLAG-Strep–RNF169 *MIU cells. Expression of S-FLAG-Strep–RNF169 in cell lines is shown in Fig. S3 D. Bars, 10 µm.