Ubiquitin-induced RNF168 condensation promotes DNA double-strand break repair

Abstract

Rapid accumulation of repair factors at DNA double-strand breaks (DSBs) is essential for DSB repair. Several factors involved in DSB repair have been found undergoing liquid-liquid phase separation (LLPS) at DSB sites to facilitate DNA repair. RNF168, a RING-type E3 ubiquitin ligase, catalyzes H2A.X ubiquitination for recruiting DNA repair factors. Yet, whether RNF168 undergoes LLPS at DSB sites remains unclear. Here, we identified K63-linked polyubiquitin-triggered RNF168 condensation which further promoted RNF168-mediated DSB repair. RNF168 formed liquid-like condensates upon irradiation in the nucleus while purified RNF168 protein also condensed in vitro. An intrinsically disordered region containing amino acids 460-550 was identified as the essential domain for RNF168 condensation. Interestingly, LLPS of RNF168 was significantly enhanced by K63-linked polyubiquitin chains, and LLPS largely enhanced the RNF168-mediated H2A.X ubiquitination, suggesting a positive feedback loop to facilitate RNF168 rapid accumulation and its catalytic activity. Functionally, LLPS deficiency of RNF168 resulted in delayed recruitment of 53BP1 and BRCA1 and subsequent impairment in DSB repair. Taken together, our finding demonstrates the pivotal effect of LLPS in RNF168-mediated DSB repair.

Keywords: DNA double-strand break repair; RNF168; intrinsically disordered region; liquid–liquid phase separation; polyubiquitin.

Fig. 1.

RNF168 forms liquid-like condensates at DNA damage sites. (A) Disordered region analysis of RNF168 using PONDR (http://www.pondr.com). (B) Representative images of exogenous RNF168-mEGFP in HEK 293T cells transfected with low (0.3 μg) or high (1 μg) dose of plasmid for 24 h. (C) 3D-capturing images of RNF168-mEGFP condensates in HEK 293T cells transfected with plasmid. (D) The IF assay was performed to detect the endogenous RNF168 and γ-H2A.X signals after 1 h recovery from irradiation (3 Gy) in HeLa cells. RNF168 formed puncta colocalized with γ-H2A.X in the nucleus. (E) Exogenous RNF168-mEGFP formed spherical puncta in the microirradiated region of which fluorescence intensity increased over time in HeLa cells. (F) FRAP assay of exogenous RNF168-mEGFP puncta in HeLa cells. (G) FRAP assay of microirradiation induced RNF168-EGFP puncta in HeLa cells. (H) The FRAP assay was performed within an RNF168-EGFP condensate in cells. (I) TAMRA-dyed RNF168 protein formed condensates in buffers containing 20 mM Tris-HCl (pH 7.4), 150 mM NaCl, and 5% PEG8000 visualized by confocal microscopy. (J) The impacts of protein concentration, salt content, and pH on RNF168 condensation in vitro. The fluorescence intensities of condensates were presented as the area × mean intensity (A. × M.). Data were presented as mean ± SEM (Scale bar, 5 μm) (unless otherwise specified in the image).

Fig. 2.

Disordered region 460–550aa is essential for the LLPS of RNF168. (A) Schematic diagram of RNF168 consisting of a RING domain and two IDRs (IDR1 and IDR2). IDR1 contains an MIU1, while IDR2 contains an MIU2 and a LRM2 motif, respectively. (B) Schematic diagram and representative images of mEGFP-tagged RNF168, ΔRING, ΔIDR1, and ΔIDR2 mutants in HEK 293T cells transfected with indicated plasmids. Counts of RNF168 puncta per cell were shown. (C) In vitro condensation assay of purified recombinant GST-IDR1-mEGFP and GST-IDR2-mEGFP protein. 10 μM IDR2-mEGFP protein formed spherical condensates in buffers containing 20 mM Tris-HCl (pH 7.4), 150 mM NaCl, and 5% PEG8000. (D) 20 μM IDR2-mEGFP condensates showed round and smooth surface captured by AFM under the same buffer condition to (C). (E) The impacts of protein concentration and salt content on condensation of IDR2-mEGFP protein. The fluorescence intensities were presented as the area × mean intensity (A. × M.). (F) Exogenous IDR2-Cry2-mCherry formed puncta at stimulation of blue light (488 nm) in HEK 293T cells. (G) FRAP assay of the blue light-induced IDR2-Cry2-mCherry puncta in HEK 293T cells. (H) Schematic diagram and representative images of mEGFP-tagged RNF168, ΔIDR2, Δ(323–442aa), Δ(460–550aa), Δ(479–550aa), ΔLRM2, Δ(460–504aa), and Δ(505–550aa) mutants in HEK 293T cells. Counts of RNF168 puncta per cell were shown. Data were presented as mean ± SEM (Scale bar, 5 μm) (unless otherwise specified in the image).

Fig. 3.

K63-linked polyubiquitin chains enhance the LLPS of RNF168. (A) Coomassie blue staining of purified RNF168 and Δ(460–550aa) proteins. (B) In vitro condensation assay of TAMRA-dyed RNF168 and Δ(460–550aa) proteins under distinct concentrations in buffers containing 20 mM HEPES (pH 7.8) and 150 mM NaCl. The mixture reacted at 4 °C for 2 h before observation. (C) In vitro condensation assay of TAMRA-dyed RNF168 and Δ(460–550aa) proteins with or without K63-linked polyubiquitin chains. Note that TAMRA-dyed RNF168 protein hardly formed condensates in high salt condition (210 mM NaCl). Addition of 2 μM K63-linked polyubiquitin chains (linkage length n ≥ 10) significantly promoted RNF168 to form liquid-like condensates under high protein concentration (10 μM). The polyubiquitin-induced RNF168 condensates enlarged over time. Comparatively, polyubiquitin only slightly induced Δ(460–550aa) to form small puncta albeit with high protein concentration (10 μM). (D) The impacts of K63-linked ubiquitin chains with distinct linkage lengths on the condensation of 10 μM TAMRA-dyed RNF168 proteins in buffers containing 20 mM HEPES (pH 7.8) and 210 mM NaCl. The mixture was incubated at 4 °C for 2 h before observation (Scale bar, 50 μm) (unless otherwise specified in the image).

Fig. 4.

LLPS is required for the recruitment of RNF168 and it-mediated H2A.X ubiquitination. (A) The laser microirradiation assay was performed to explore the impacts of IDR deletion on RNF168 recruitment at DSB sites. Where indicated, HeLa cells were transfected with siNC or siRNF168 for 48 h before microirradiation. Note that exogenous RNF168-mEGFP formed spherical condensates in the microirradiated region, while Δ(460–550aa)-mEGFP displayed ground-glass assembly, independent of endogenous RNF168. (B) The IF assay was performed to explore the impacts of IDR deletion on RNF168 recruitment at DSB sites. IF of RNF168 and γ-H2A.X was detected at 0, 0.5, 5, and 24 h after 3 Gy irradiation, and the numbers of colocalized foci were counted. Representative images of cells at 5 h after irradiation were presented. (C) Western blotting showed the impacts of IDR deletion on RNF168-mediated H2A.X ubiquitination. The levels of H2A.X ubiquitination were detected by anti-γ-H2A.X antibody at 1 h after 10 Gy irradiation. For (B and C), RNF168-KO HeLa cells stably re-expressing mEGFP or sgRNA-resistant RNF168 variants were used. (D) The IF assay was performed to explore the impacts of IDR deletion on RNF168-mediated ubiquitinated conjugates at DSB sites. IF of ubiquitinated conjugates (detected by anti-FK2) and γ-H2A.X was detected at 1 h after 3 Gy irradiation, and the numbers of FK2 foci were counted. (E) HeLa cells transfected with RNF168-mEGFP plasmid for 48 h were continuously treated with 2.5% 1,6-hexanediol from 1 h before irradiation till 1 h after irradiation (1 Gy). Cells were then fixed and subjected to the IF assay. The numbers of exogenous RNF168 foci as well as the colocalized foci of RNF168 and FK2 were counted. (F and G) IF assays were performed to explore the impacts of LLPS deficiency of RNF168 on BRCA1 (F) and 53BP1 (G) recruitment at DSB sites. IF of BRCA1, 53BP1, and γ-H2A.X was detected at 1 h after 3 Gy irradiation. For (D, F, and G), RNF168-KO or RNF168-knockdown (KD) HeLa cells stably re-expressing sgRNA- or shRNA-resistant RNF168 variants were used. Data were presented as mean ± SEM (Scale bar, 5 μm) (unless otherwise specified in the image).

Fig. 5.

LLPS of RNF168 promotes DSB repair. (A) Schematic diagram and representative images of Δ(460–550aa) and Δ(460–550aa)-LRM2 in HEK 293T transfected with indicated plasmids. (B) Flag-tagged protein/nucleosomes pull-down assay to verify the nucleosome binding capacity of RNF168 variants. HEK 293T cells were transfected with Flag-tagged RNF168, Δ(460–550aa), or Δ(460–550aa)-LRM2 plasmids for 48 h before cell lysis. The lysates were incubated with cell-derived nucleosomes, and the pull-downs were analyzed by western blotting. Nucleosomes were indicated by anti-H3. (C and D) The IF assay was performed to explore the impacts on ubiquitination (C) and 53BP1 recruitment (D) at DSB sites of distinct mutants including ΔMIU2, ΔLRM2, Δ(460–550aa), and Δ(460–550aa)-LRM2. IF of FK2, 53BP1, and γ-H2A.X was detected at 1 h after 3 Gy irradiation. Numbers of colocalized foci were counted. (E and F) The IF assay was performed to analyze the impacts on the DNA damage response process of Δ(460–550aa) (E) and Δ(460-550aa)-LRM2 (F). IF of γ-H2A.X was detected at distinct time points after 3 Gy irradiation. (G) Representative images of mEGFP-tagged Δ(460–550aa)-LRM2 and Δ(460–550aa)-LRM2-(FUSN)₂ in HEK 293T cells transfected with indicated plasmids. Counts of RNF168 puncta per cell were shown. (H) The IF assay was performed to analyze whether regaining LLPS function by FUSN fusion rescued the DSB repair defect of the Δ(460–550aa)-LRM2 mutant. IF of γ-H2A.X was detected at distinct time points after 3 Gy irradiation. Data were presented as mean ± SEM (Scale bar, 5 μm) (unless otherwise specified in the image).

Fig. 6.

LLPS promotes RNF168 interacting with 53BP1 and BRCA1 at DSB sites. (A and B) The proximity ligation assay (PLA) was performed to verify the impacts of LLPS on the interactions between RNF168 and downstream factor 53BP1 (A) or BRCA1 (B) using the RNF168-KO HeLa cells stably re-expressing mEGFP-tagged wild-type or mutant RNF168. The PLA signals were detected by anti-GFP, anti-53BP1 and anti-BRCA1. (C) Schematic diagram for ubiquitin-induced RNF168 condensation promoting DNA DSB repair. In the initial stage of DSB repair, RNF168 assembles at DSB sites via binding to ubiquitinated H1 catalyzed by RNF8 and undergoes LLPS in a self-interacting IDR-driven manner. RNF168 subsequently triggers and amplifies the ubiquitination of H2A.X, which provides multivalent interactions and further enhances RNF168 condensation. This positive feedback axis drives the rapid accumulation of RNF168 at DSB sites and the subsequent recruitment of downstream repair factors (53BP1 and BRCA1) for DSB repair. Data were presented as mean ± SEM (Scale bar, 5 μm) (unless otherwise specified in the image).