Mammalian BTBD12/SLX4 assembles a Holliday junction resolvase and is required for DNA repair

Abstract

Structure-specific endonucleases mediate cleavage of DNA structures formed during repair of collapsed replication forks and double-strand breaks (DSBs). Here, we identify BTBD12 as the human ortholog of the budding yeast DNA repair factor Slx4p and D. melanogaster MUS312. Human SLX4 forms a multiprotein complex with the ERCC4(XPF)-ERCC1, MUS81-EME1, and SLX1 endonucleases and also associates with MSH2/MSH3 mismatch repair complex, telomere binding complex TERF2(TRF2)-TERF2IP(RAP1), the protein kinase PLK1 and the uncharacterized protein C20orf94. Depletion of SLX4 causes sensitivity to mitomycin C and camptothecin and reduces the efficiency of DSB repair in vivo. SLX4 complexes cleave 3' flap, 5' flap, and replication fork structures; yet unlike other endonucleases associated with SLX4, the SLX1-SLX4 module promotes symmetrical cleavage of static and migrating Holliday junctions (HJs), identifying SLX1-SLX4 as a HJ resolvase. Thus, SLX4 assembles a modular toolkit for repair of specific types of DNA lesions and is critical for cellular responses to replication fork failure.

Figure 1. Human SLX4 associates with structure-specific endonucleases, telomere recognition proteins, miss-match repair proteins, and PLK1

(A) Proteomic analysis of SLX4-HA purified from 293TREX cells was performed as described in Experimental Procedures. High confidence candidate interacting proteins (HCIPs) were determined using CompPASS to derive Z and D^N-Scores. Proteins marked by red symbols fall below the stringent threshold set for HCIPs based on a D^N-Score = 1. (B) Summary of D^N-Scores, Z-Scores, and total spectral counts (TSCs) for major SLX4 associated proteins. Alternative names are indicated. (C–E) Plots of Z-Score versus D^N-Scores for SLX4 interacting proteins. The indicated proteins were stably expressed in 293T cells, purified and analyzed by LC-MS/MS. (F) Interaction network for the human SLX4 complex. Solid lines; interactions observed by mass spectrometry. Dotted lines; interactions in BIOGRID (red), STRING (black), or MINT (green) databases. (G–K) SLX4 was immunoprecipitated (+/− antigenic peptide) from U2OS cells with or without SLX4 depletion and immunoblotted. * indicates heavy chain

Figure 2. Structural anatomy of the human SLX4 complex

(A) Summary of the interaction data showing the approximate location of binding between SLX4 and associated proteins. (B) D^N-Score versus Z-Score plots for HA-SLX4ΔN and HA-SLX4ΔC showing the identity of proteins identified by mass spectrometry (see Figure 1 legend for details). (C) Summary of proteomic data for HA-SLX4ΔN, HA-SLX4ΔC, and HA-SLX4^SBD. TSC, total spectral counts. The full list of interacting proteins is provided in Table S1. (D–F) Domain mapping of SLX4 and its associated proteins. The indicated MYC-tagged proteins were expressed in 293T cells, immunoprecpitated as indicated, and immunoblotted. α-CDK2; loading control. (G, H) Interaction of SLX4 with target proteins using the yeast two-hybrid system. The indicated proteins were cloned into activation domain (AD) or DNA-binding domain (DB) vectors and two-hybrid analysis performed as described under Experimental Procedures. (I) Sequence relationship between vertebrate, yeast, Drosophila and C. elegans SLX4 in the C-terminal region involved in binding SLX1.

Figure 3. Reduction in components of the SLX4 complex sensitizes cells to DNA damaging agents

(A–D) The Multi-color Competition Assay (MCA) was used to examine the effect of depletion of SLX4 on the sensitivity of U2OS and HeLa cells to DNA damaging agents using independent shRNAs and siRNAs (panels A and C). Corresponding knockdown is shown (panel B and D). *: p<0.05; **: p<0.005, students T-test. Error bars are STDEV across 3 technical replicates. (E–F) Effect of siRNA depletion of ERCC4, MUS81, SLX1, and GEN1 on the resistance of HeLa cells to DNA damage (panel E). The indicated HeLa cell extracts were examined for protein expression after siRNA transfection (panel F). (G) U2OS cells expressing GFP-SLX4 were subjected to laser micro-irradiation and after 60 min, cells were processed for γH2AX staining and/or imaging of GFP-SLX4. Nuclei were visualized with DAPI. Arrowheads indicate position of stripes.

Figure 4. SLX4 associates with structure-specific endonuclease activity

(A) SLX4-HA or GFP-HA (0.5mg, 1.5mg) complexes were immunoprecipitated from 293T cells and were incubated with ³²P-end-labeled DNA substrates (30 min) prior to electrophoresis on native gels. Labeled strand indicated by *. (B, C) Endogenous SLX4 complexes were immunoprecipitated from 293T cells (0.5, 1.5mg) and incubated with ³²P-end-labeled DNA substrates (30 min) prior to electrophoresis on native gels (panel B) or immunoblotting (panel C).

Figure 5. SLX4-dependent cleavage of HJs maps to the C-terminal MUS81-EME1 and SLX1 binding domain (SBD)

(A) Schematic of SLX4 deletion fragments and relevant interactions. (B) The indicated SLX4 complexes (from 1.5mg of cell extract) were incubated with radiolabeled substrates (30 min) labeled on strand 1 prior to native gel electrophoresis and autoradiography. (C) Purified endonucleases produced in bacteria. (D, E) SLX4^SBD stimulates static and migrating HJ cleavage by SLX1. The indicated recombinant GST-SLX1-His₆-SLX4^SBD complexes (1, 3, or 5 ng) were incubated with XO-BE or X26 HJs at 37°C for 30 min and reaction products analyzed on native gels (panel D) or by phosphoimaging (panel E). SLX1CD: SLX1^E82A.

Figure 6. Cleavage specificity of SLX4 and SLX1-SLX4^SBD complexes

(A) The indicated SLX4 complexes purified from 293T cells were incubated with radiolabeled substrates (30 min) prior to 12% polyacrylamide-urea gel electrophoresis and autoradiography. In the structural model (bottom) longer bars represent more efficient cleavage. (B) HJ cleavage activity of SLX4 maps to the C-terminus containing the MUS81-EME1 and SLX1 binding sites. The indicated SLX4 immune complexes derived from 1.5 mg of 293T cell extract were incubated with HJ (X0-BE) labeled on strand 1 (30 min) and the products separated on denaturing gels. (C) Symmetrical cleavage of HJ (X0-BE) by SLX4 complexes and GEN1-HA. The indicated immune complexes were incubated with HJ (X0-BE) radiolabeled on either strand 1 or strand 3 and products resolved on a denaturing gel prior to autoradiography. (D) Major sites of cleavage observed with SLX4 (arrow and bars) or GEN1 (arrowhead) complexes. (E–G) Symmetrical cleavage of HJ (X0-BE) by SLX1-SLX4^SBD complexes. The indicated SLX4 complexes purified from 293T cells (panel E) or bacteria (panel F) were incubated with HJ (X0-BE) radiolabeled on strand 1 or strand 3 and products resolved on a denaturing gel prior to autoradiography. In panel G, the major site of HJ (X0-BE) cleavage by SLX1-SLX4^SBD is indicated by the arrow.

Figure 7. SLX1-SLX4 complexes resolve HJs through an ordered pathway in vitro and are required for efficient homology-directed repair of DSBs in vivo

(A) The indicated SLX4 and GEN1 complexes were incubated with an X26 substrate +/−T4 DNA ligase prior to separation on a denaturing gel and autoradiography. (B) The indicated SLX4-HA immune complexes were immunoblotted for ERCC4 and MUS81. (C) Recombinant SLX1-SLX4^SBD complexes generate cleaved X26 structures that are ligatable. (D) Requirement of SLX1 and MUS81 for cleavage of static HJs. 293T/SLX4-HA cells were transfected with control siRNA or siRNAs targeting SLX1 (SLX1#4) or MUS81 (MUS81-17). After 72 h, SLX4-HA complexes were assayed in cleavage assays using HJ X0-BE substrates labeled on either strand 1 or strand 3. Products were resolved on denaturing gels prior to autoradiography. (E) SLX4-HA immune complexes corresponding to panel D were immunoblotted for ERCC4 and MUS81. (F) qPCR analysis of mRNA isolated from cells transfected with control siRNA or siRNA targeting SLX1 (#4). mRNA abundance is relative to GAPDH. (G–I) MUS81-EME1 complexes cleave nicked HJs that mimic those produced by SLX1-SLX4. Bacterial GST-MUS81-His₆-EME1 complexes were incubated (37°C, 30 min) with synthetic nicked HJs (XO1 strand nicked at nucleotide 32) labeled on either strand 1 or strand 3 and the products analyzed on either native (panel G) or denaturing (panel H) gels. The positions of cleavage of the nicked XO structure labeled on strand 3 are shown in panel I. (J) GST-MUS81-His₆-EME1 (12.5, 25, 50 ng) was incubated (37°C, 30 min) with XO-BE in the presence or absence of 3 ng of recombinant SLX1-SLX4^SBD, and the products analyzed on denaturing gels. (K) DR-GFP U2OS were infected with viral vectors expressing the indicated shRNAs followed by I-SceI expression 48h later. The extent of homologous recombination was determined by flow cytometry. *, p < 0.05; students T-test. Error bars are STDEV across 3 technical replicates. (L) Model depicting distinct roles of SLX1 and MUS81 within the SLX4 complex for HJ cleavage (see text for details).