Dual-utility NLS drives RNF169-dependent DNA damage responses

Abstract

Loading of p53-binding protein 1 (53BP1) and receptor-associated protein 80 (RAP80) at DNA double-strand breaks (DSBs) drives cell cycle checkpoint activation but is counterproductive to high-fidelity DNA repair. ring finger protein 169 (RNF169) maintains the balance by limiting the deposition of DNA damage mediator proteins at the damaged chromatin. We report here that this attribute is accomplished, in part, by a predicted nuclear localization signal (NLS) that not only shuttles RNF169 into the nucleus but also promotes its stability by mediating a direct interaction with the ubiquitin-specific protease USP7. Guided by the crystal structure of USP7 in complex with the RNF169 NLS, we uncoupled USP7 binding from its nuclear import function and showed that perturbing the USP7-RNF169 complex destabilized RNF169, compromised high-fidelity DSB repair, and hypersensitized cells to poly (ADP-ribose) polymerase inhibition. Finally, expression of USP7 and RNF169 positively correlated in breast cancer specimens. Collectively, our findings uncover an NLS-mediated bipartite mechanism that supports the nuclear function of a DSB response protein.

Keywords: DNA damage; DNA repair; RNF169; USP7; deubiquitylation.

Fig. 1.

RNF169 interacts directly with USP7. (A) Affinity purification of RNF169 protein complexes identified a list of RNF169-interacting proteins. Names and numbers of distinct peptides from top hits are shown. (B) U2OS RNF169 SFB cells were induced or not induced with doxycycline (Dox) for 24 h. Cell lysates were subjected to IP with Streptavidin beads, and RNF169-associated immunoprecipitates were analyzed by Western blotting analysis using indicated antibodies. DYRK1A and CHK1 were used as positive and negative controls, respectively. (C) Cell lysates derived from HeLa cells were incubated with Protein A agarose beads conjugated with anti-RNF169 antibody or control. Immunoprecipitates were detected with anti-RNF169 or anti-USP7 antibodies. (D) Schematic illustration of full-length (FL) and deletion mutants of RNF169 used in this study. MIU, motif interacting with ubiquitin; RING, Really Interesting New Gene. (E) HEK293T cells were transiently transfected with plasmids encoding Myc-USP7 and HA-tagged RNF169 wild type and mutants (as depicted in D). Cell lysates were incubated with anti-HA affinity matrix, and immunoprecipitates were blotted with anti-HA and anti-Myc antibodies. (F) RNF169 interacts directly with USP7. Purified GST-USP7 fusion proteins were incubated with purified MBP-RNF169 immobilized on amylose resin. (G and H) HEK293T cells were transfected with plasmids encoding SFB-tagged USP7 FL or mutants. Cell lysates were prepared for pull-down experiments against purified MBP-RNF169. A schematic illustration of USP7 constructs is shown. CD, catalytic domain; HA, human influenza hemagglutinin; IP, immunoprecipitation; TRAF, TNF receptor-associated factor.

Fig. 2.

Crystal structure of USP7 UBL1–3-RNF169_620–632 peptide. (A) Alignment of KxxxK motif from RNF169 and other USP7 UBL domain-binding proteins. (B) Comparison of the dissociation constants of GMPS, ICP0, and RNF169 peptides with bacterially purified USP7 UBL1–3. (C, Left) Electrostatic surface representation of the USP7 UBL1–3 domain complexed with RNF169_620–632 peptide. Acidic regions are shown in red, and basic regions are shown in blue. (C, Right) Ribbon representation of the complex of RNF169_620–632 peptide and the USP7 UBL1–3 domain, with UBL1, UBL2, and UBL3 colored blue, gold, and gray, respectively. The RNF169 peptide is shown in stick representation; the missing loop (residues 668–673) connecting the UBL1 and UBL2 domains is represented using a dotted line. (D) Binding interface between RNF169_620–632 peptide and USP7. Key hydrogen bonds bridging the interactions are highlighted by brown dashed lines. (E) Two acidic pockets (A and B) on the surface of UBL1–2 are optimized for the recognition of the lysine residues (K623 and K627) of RNF169. The “sliced” side view is plotted to show engagement of the side chains of the lysine residues in the binding pockets.

Fig. S1.

ITC curves for USP7 UBL1–3 over GMPS316–328 (A), ICP0615–627 (B), and RNF169620–632 (C) (Fig. 2B).

Fig. 3.

Critical residues of the USP7- and RNF169-binding interface. (A) Mutational analysis of the interaction between RNF169_620–632 peptide and USP7 UBL1–3 mutants by ITC experiments. Mutations include the USP7 single-point mutants E759A, D762A, and D764A and the triple mutant E759A/D762A/D764A (M1). (B) HEK293T cells were cotransfected with HA-tagged USP7 (wild type and M1 mutant) and Flag-ICP0 or Flag-RNF169. Cell lysates were incubated with anti-HA agarose beads, and immunoprecipitates were analyzed using anti-HA and anti-Flag antibodies. (C) Sequence alignment of the USP7-binding region on RNF169 (608–655 aa) of different species. Key lysine residues are shown in red. (D) U2OS cells were transfected with HA-RNF169 (wild type and mutants) and their subcellular localization was examined by indirect immunofluorescence staining using anti-HA antibodies. Nuclei were visualized by DAPI staining. (Magnification: 60×.) (E) Mutational analysis of the interaction between USP7 UBL1–3 and the RNF169_620–632 peptide by ITC experiments. Mutations include RNF169 K623A, K627A, and K623A/K627A (Left) and K623R, K627R, and K623R/K627R (Right). (F) Cell lysates derived from HEK293T cells expressing indicated HA-tagged RNF169 proteins were incubated with anti-HA agarose beads. Immunoprecipitates were separated by SDS/PAGE, and Western blotting experiments were performed using anti-HA and anti-USP7 antibodies.

Fig. S2.

“Sliced” side view of the binding pockets reveals engagement of the side chains of the lysine residues (A and C) or the arginine residues (B and D) in the binding pockets.

Fig. S3.

Superpositions of USP7 UBL1–3 domains of the RNF169–USP7 complex (PDB ID code 5GG4) with the domains of the ICP0–USP7 (A, PDB ID code 4WPH) and UHRF1–USP7 (B, PDB ID code 5C6D) complexes. RNF169, ICP0, and UHRF1 peptides are shown in green, red, and yellow, respectively. The orientations of the structures in the enlarged panels are slightly adjusted to illustrate the side-chain engagement of the lysine residues better.

Fig. 4.

USP7 deubiquitylates and stabilizes RNF169. (A) HeLa cells pretreated with indicated siRNAs were harvested, and cell lysates were subjected to Western blotting analysis with indicated antibodies. *Nonspecific bands. (B) HeLa cells were treated with DMSO, P22077 (20 μM), or P5091 (15 μM) for 2 h before ionizing radiation (IR) treatment (10 Gy). Cells were harvested 4 h after IR treatment, and lysates were immunoblotted with indicated antibodies. (C) Lysates derived from USP7 KO HeLa cells (clones 4 and 7) and control HeLa cells were analyzed by Western blotting experiments using indicated antibodies. (D) USP7 KO HeLa cells and their parental cells were treated with 50 μg/mL cycloheximide (CHX), and cells were harvested at indicated time points for processing. Western blotting experiments were performed using indicated antibodies to evaluate the expression of RNF169. Quantification is shown, and data represent mean ± SEM from three independent experiments (*P < 0.01; **P < 0.001). WT, wild type. (E) HeLa cells and their USP7 KO derivatives were treated with MG132 (10 μM) for 4 h. Cell lysates were prepared, and IP experiments were performed under denaturing conditions using Protein A beads conjugated with anti-RNF169 antibodies. Western blotting experiments were performed using indicated antibodies. (F) Strep-Flag-USP7 proteins were incubated at 37 °C for 10 min with immobilized Flag-ubiquitylated proteins, including Myc-RNF169. Thereafter, reaction products were boiled in Laemmli buffer, separated by SDS/PAGE, and processed for Western blotting using anti-Myc antibodies to determine the ubiquitylation status of Myc-RNF169. (G) Cell lysates derived from control or USP7 KO HeLa cells and their reconstituted counterparts (WT and mutants) were subjected to Western blotting analysis. Relative RNF169 expression from three experiments is plotted (***P < 0.001 vs. HeLa). ns, not significant. (H) USP7 KO HeLa cells were cotransfected with Flag-RNF169, HA-tagged USP7 (WT and mutants), or HA vector. Cells were treated with 10 μM MG132 for 4 h and were subsequently processed for IP experiments using anti-Flag beads under denaturing conditions. Western blotting analyses were performed using indicated antibodies. (I) RNF169 KO HeLa cells reconstituted with WT RNF169 or its mutants (∆C6 or 2KR) were subjected to 50 μg/mL CHX, and cells were harvested at indicated time points. Proteins were separated by SDS/PAGE and were immunoblotted with indicated antibodies. Quantification is shown and represents data of three independent experiments (**P < 0.01 vs. WT). (J) RNF169 KO HeLa cells stably expressing Flag-tagged RNF169 (WT and 2KR) were treated with or without IR before the ubiquitylation status of RNF169 proteins was analyzed. Procedures are essentially as described in H. (K) USP7 deubiquitylated RNF169, but not its 2KR mutant, in vitro. The experiment was performed essentially as in F. (L) Representative histological photomicrographs demonstrating direct correlation between nuclear RNF169 and nuclear USP7 expression in breast cancer. Statistics analyses revealed a significant correlation (P < 0.05) between immunohistochemical expression of RNF169 and USP7 (Materials and Methods). (Magnification: 10× in L; scale bars, 20 μM in Insets.)

Fig. S4.

(A) HeLa cells [wild type (WT)] and their USP7-deficient derivatives (clones 4 and 7) were cotransfected with Flag-ubiquitin and Myc-RNF169 expression constructs. Twenty-four hours after transfection, cells were treated with MG132 (10 μM for 4 h) and were subsequently processed for IP experiments under denaturing conditions (buffer containing 1% SDS) using anti-Flag agarose beads. Western blotting analyses were performed using indicated antibodies. (B) USP7 KO HeLa cells (clone 4) were cotransfected with Flag-ubiquitin, Myc-RNF169, and HA-tagged USP7 (WT and mutants) or an empty HA vector. Cells were processed, and the ubiquitylated status of RNF169 was analyzed as described in A. (C) RNF169 KO HeLa cells were cotransfected with Flag-ubiquitin and Myc-RNF169 or Myc-RNF169 (2KR), and were treated with or without ionizing radiation (IR) before being processed to determine the ubiquitylation status of RNF169 essentially as described for A.

Fig. S5.

Cross-cancer alteration summary for USP7. Cutoff at altered samples = 5% (data from cBioPortal). adeno, adenocarcinoma; Broad, Broad Institute; BRCCRC, British Columbia Cancer Research Center; DLBC, lymphoid neoplasm diffuse large B-cell lymphoma; DLBCL, diffuse large C-cell lymphoma; FHCRC, Fred Hutchinson Cancer Research Center; METABRIC, Molecular Taxonomy of Breast Cancer International Consortium; MPNST, malignant peripheral nerve sheath tumor; MSKCC, Memorial Sloan Kettering Cancer Center; NEPC, neuroendocrine prostate cancer; PCNSL, primary central nervous system lymphoma; pub, publication; TCGA, The Cancer Genome Atlas.

Fig. 5.

USP7 promotes RNF169 loading onto DSBs. (A) U2OS cells stably expressing Flag-RNF169 were transfected with small interfering control (siCTR) or USP7-targeting siRNAs (siUSP7-1 or siUSP7-2). Forty-eight hours after transfection, cells were subjected to IR treatment (10 Gy) and were subsequently processed for immunostaining experiments using anti-Flag and anti–γ-H2AX antibodies. (B) Quantification is shown for results derived from three independent experiments. Data represent mean ± SEM (**P < 0.01 vs. siCTR). (C) Western blotting analysis was performed to assess RNAi-mediated knockdown efficiency in A. (D–F) RNF169 IRIF were assessed in HeLa cells following essentially the same procedures described in A–C (**P < 0.01 vs. siCTR). (G–I) U2OS cells stably expressing Flag-RNF169 were treated with DMSO, P22077 (20 μM), or P5091 (15 μM). Cells received IR treatment (10 Gy) and were processed for immunostaining or Western blotting experiments 4 h afterward. Quantification of results in G is shown, and the results are derived from three independent experiments (***P < 0.001). Western blotting analyses were performed using standard procedures with indicated antibodies. (J) Parental HeLa cells and USP7 KO and their derivatives (WT, C223S, M1) were subjected to IR treatment (10 Gy). Cells were preextracted before paraformaldehyde fixation 6 h afterward, and were processed for immunostaining experiments using anti-RNF169 and anti–γ-H2AX antibodies. (K) Quantification of results derived from three independent experiments is shown, and depicts the percentage of cells with RNF169 IRIF (***P < 0.001). ns, not significant. (Magnification: 60×.)

Fig. 6.

USP7-RNF169 axis promotes HR repair. (A) Schematic of gene conversion-based HR reporter is shown. GFP positivity reflects DSB repair via the HR pathway. (B) DR-U2OS cells pretreated with indicated siRNAs were transfected with plasmid encoding the I-Sce1 endonuclease. Flow cytometric analysis of the GFP-positive cell population was performed 48 h posttransfection. Data represent mean ± SEM from three independent experiments (**P < 0.01; ***P < 0.001). (C) Western blotting experiment was performed to assess RNAi-mediated knockdown efficiency in cells used in B. (D) RNF169 KO HeLa cells stably expressing vector only or RNF169 (WT and mutant) were electroporated with plasmids encoding I-Sce1 and DR-GFP. Cells were harvested and subjected to flow cytometric analysis 48 h postelectroporation. Data represent mean ± SEM from three independent experiments (*P < 0.05). (E) HeLa RNF169 KO cells stably expressing vector only and RNF169 (WT and mutants) were treated with indicated doses of the PARPi Olaparib. Cells were allowed to grow for 10 d before they were stained with Coomassie Blue. Data represent mean ± SEM from three independent experiments (**P < 0.01; ***P < 0.001). (F) Western blotting analysis of the protein level of RNF169 in cells used in D and E. (G and H) RNF169 KO HeLa cells were mock-transfected or transiently transfected with expression constructs encoding either HA-tagged RNF169 (WT) or its 2KR mutant in addition to the plasmid-based DR-GFP reporter and an Isce1 plasmid to evaluate HR repair efficiency. HA-RNF169 2KR plasmid was transfected at increasing molar ratios to cells transfected with WT RNF169 (1:1, 2:1, and 3:1 ratios). Cells were subjected to flow cytometric analysis (G) or Western blotting analysis (H) using indicated antibodies. Results represent mean ± SEM from three independent experiments (*P < 0.05). (I and J) USP7 KO HeLa cells reconstituted with empty vector (HA), WT USP7 (WT), its DUB mutant (C223S), or the RNF169-binding defective mutant (M1) were subjected to the gene conversion assay to evaluate HR repair efficiency. (I) Results represent mean ± SEM from three independent experiments (*P < 0.05; ***P < 0.001). (J) Western blotting analysis was performed to assess HA-USP7 and RNF169 expression. HA, human influenza hemagglutinin.

Fig. 7.

Proposed working model. In the presence of USP7, RNF169 is stabilized and productively assembles at DSBs to facilitate HR-driven DNA repair. In the absence of USP7, RNF169 undergoes active degradation, resulting in attenuated accumulation at DSBs and impaired HR repair.

Fig. S6.

(A) U2OS or HeLa cells were treated with nontargeting [small interfering control (siControl)] or RNF168-specific (siRNF168) siRNAs. Cell lysates were separated by SDS/PAGE and were subjected to Western blotting analyses using indicated antibodies. The asterisk denotes nonspecific bands. (B) RNF169 KO HeLa cells reconstituted with WT RNF169 or its Really Interesting New Gene (RING) domain deletion mutant (ΔRING) were subjected to 50 μg/mL cycloheximide (CHX). Cells were harvested at the indicated time point for processing. Proteins were separated by SDS/PAGE and were processed for immunoblotting analyses using indicated antibodies. (C) Quantification is shown and represents data (mean ± SEM) derived from three independent experiments.

Fig. S7.

HeLa cells pretreated with control siRNAs (siControl) or RNF169-targeting siRNAs (siRNA169) were lysed with NETN buffer supplemented with Bitnuclease. Cell lysates were processed for Western blotting analysis using indicated antibodies.