Cullin3-KLHL15 ubiquitin ligase mediates CtIP protein turnover to fine-tune DNA-end resection

Abstract

Human CtIP is a decisive factor in DNA double-strand break repair pathway choice by enabling DNA-end resection, the first step that differentiates homologous recombination (HR) from non-homologous end-joining (NHEJ). To coordinate appropriate and timely execution of DNA-end resection, CtIP function is tightly controlled by multiple protein-protein interactions and post-translational modifications. Here, we identify the Cullin3 E3 ligase substrate adaptor Kelch-like protein 15 (KLHL15) as a new interaction partner of CtIP and show that KLHL15 promotes CtIP protein turnover via the ubiquitin-proteasome pathway. A tripeptide motif (FRY) conserved across vertebrate CtIP proteins is essential for KLHL15-binding; its mutation blocks KLHL15-dependent CtIP ubiquitination and degradation. Consequently, DNA-end resection is strongly attenuated in cells overexpressing KLHL15 but amplified in cells either expressing a CtIP-FRY mutant or lacking KLHL15, thus impacting the balance between HR and NHEJ. Collectively, our findings underline the key importance and high complexity of CtIP modulation for genome integrity.

Figure 1. CtIP interacts with the CUL3-KLHL15 complex.

(a) HEK293 cells inducibly expressing StrepHA-tagged CtIP were used for tandem affinity purification of protein complexes. The number of unique peptides and sequence coverage for the proteins identified by mass spectrometry are listed. ‘/' delimitates the data from two biological replicates. (b) HEK293T cells were transfected with either empty vector (EV) or the GFP-CtIP expression constructs. Forty-eight hours after transfection, cells were lysed and whole-cell extracts were subjected to IP using anti-GFP affinity resin. Inputs and recovered protein complexes were analysed by immunoblotting. (c) Schematic representation of the human KLHL15 protein indicating truncation and single-amino acid point mutants thereof used in d–f. ‘3-box' denotes CUL3-interacting box, whereas the six Kelch repeats are indicated in pink. (d–f) HEK293T cells were transfected with either EV or the indicated FLAG-KLHL15 expression constructs. Forty-eight hours after transfection, cells were lysed and whole-cell extracts were subjected to IP using anti-FLAG M2 affinity resin. Inputs and recovered protein complexes were analysed by immunoblotting. Asterisks indicate neddylated CUL3.

Figure 2. CUL3-KLHL15 E3 ligase promotes CtIP protein turnover via the ubiquitin-proteasome pathway.

(a) U2OS Flp-In T-REx cells stably expressing doxycycline (Dox)-inducible GFP-KLHL15-wt were cultivated in the absence (−) or presence (+) of Dox. Twenty-four hours post induction, cells were either mock-treated or treated with MG-132 (10 μM) for 6 h and whole-cell lysates were analysed by immunoblotting. (b) Twenty-four hours post induction, same cells as in a were either mock-treated or treated with MLN-4924 (100 nM) for 8 h and whole-cell lysates were analysed by immunoblotting. (c) Same cells as in a were transfected with either non-targeting (CNTL) or CUL3 siRNA oligos. 48 h later, GFP-KLHL15 expression was induced with Dox for 24 h and whole-cell lysates were analysed by immunoblotting. (d) U2OS cells were transfected with the indicated siRNA oligos for 48 h. Total cell extracts and chromatin-enriched fractions were analysed by immunoblotting using the indicated antibodies. (e) Same cells as in a were transfected with the indicated siRNA oligos. Twenty-four hours post transfection, cells were grown for 24 h in the absence (−) or presence (+) of Dox and whole-cell lysates were analysed by immunoblotting. (f) HEK293 Flp-In T-REx cells inducibly expressing GFP-CtIP were transfected with the indicated siRNAs. Forty-eight hours later, cells were transfected with His-Ub and the expression of CtIP was simultaneously induced with Dox. Eight hours post induction, cells were transfected with siRNAs for a second time. Three days after the first siRNA transfection, cells were treated with MG-132 (20 μM) for 6 h followed by lysis in buffer containing guanidium-HCl. Ubiquitin conjugates were pulled-down (PD) with Ni-NTA-agarose beads, eluted and analysed by immunoblotting with anti-GFP antibody. (g) Recombinant FLAG-CtIP was incubated with ATP, ubiquitin, E1, E2 enzymes and CUL3-N8/RBX1-KLHL15 (E3) for 30 min at 37 °C. Unmodified and ubiquitinated CtIP protein species were detected by immunoblotting with anti-FLAG antibody. Asterisks indicate neddylated CUL3.

Figure 3. CtIP protein turnover is impaired in KLHL15 knockout cells.

(a) Lysates from a stable HEK293^Cas9 wild-type (wt) cell clone or from six different HEK293^Cas9 cell clones generated with the same sgRNA targeting KLHL15 (see Supplementary Fig. 3a) were subjected to western blotting with the indicated antibodies. (b) Total (T) cell extracts and chromatin-enriched (Chr.) fractions of HEK293^Cas9/wt and HEK293^{Cas9/KLHL15Δ} cells were analysed by immunoblotting using the indicated antibodies. (c) Same cells as in b were either mock-treated, treated with CHX (100 μg ml⁻¹) or MG-132 (20 μM) for the indicated time points and lysates were subjected to western blotting with the indicated antibodies (upper panel). SE and LE; short- and long-exposure times of the same immunoblot. Relative CtIP protein levels were determined by quantification of CtIP band intensity (normalized to TFIIH) with the ImageJ software (lower panels). Data are represented as mean values of densitometric quantification±s.e.m. (n=3).

Figure 4. KLHL15 overexpression leads to camptothecin hypersensitivity and defective DNA-end resection.

(a) U2OS Flp-In T-REx cells inducibly expressing GFP-KLHL15-wt or GFP-KLHL15-Y552A were cultivated in the presence or absence of Dox. Twenty-four hours post induction, cells were treated with the indicated doses of camptothecin and survival was determined after 10 days by colony-formation assay. Data are presented as the mean±s.d. (n=3). (b) Same cells as in a were mock-treated or treated with camptothecin (CPT, 1 μM) for 1 h and lysates were analysed by immunoblotting using the indicated antibodies. Asterisks indicate hyperphosphorylated forms of CtIP and RPA2, respectively. (c) Same cells as in b were labelled with BrdU (30 μM) for 24 h before CPT treatment. Cells were harvested, permeabilized, fixed, immunostained with anti-RPA2 or anti-BrdU antibody and analysed by FACS. Dot plots representing the intensity of the signals for RPA2 or BrdU staining (y axis) against the DNA content (x axis). Quantification gates were established in untreated samples and the percentage of cells within the gates is indicated. (d) HEK293 DR-GFP cells were transfected with the I-SceI in combination with the indicated FLAG-KLHL15 expression plasmids and harvested after 48 h for flow cytometry and immunoblot analysis. Data are represented as mean±s.d. (n=3). Statistical analysis were carried out using unpaired, two-tailed t-tests. P values expressed as ** (P<0.005) were considered significant. A.u., arbitrary units; FACS, fluorescence-activated cell sorting.

Figure 5. Mapping of a KLHL15-interacting region in CtIP.

(a) Schematic representation of GFP-tagged CtIP wild-type (wt) and truncation mutants. A 15-mer peptide encompassing amino acid residues 830–844 in the CtIP CTD important for KLHL15 binding is marked by a red box. (b) MBP-KLHL15 was coupled to amylose beads and incubated with lysates of HEK293T cells transfected with the indicated GFP-CtIP expression constructs for 48 h. Inputs and pulled-down protein complexes were analysed by immunoblotting. (c) HEK293T cells were cotransfected with the indicated GFP-CtIP expression constructs and His-Ubiquitin. Twenty-four hours post transfection cells were treated with MG-132 (20 μM) for 4 h. Cells were then lysed in buffer containing guanidium-HCl and ubiquitin conjugates were pulled-down using Ni-NTA-agarose beads, eluted, and analysed by immunoblotting with anti-GFP antibody. (d) MBP-KLHL15 pull-down assays were performed as in b. (e) Three microgram of a biotinylated 15-mer CtIP peptide (aa 830–844) indicated in a was coupled to streptavidin beads and incubated with lysates of U2OS^{GFP-KLHL15-wt} cells cultivated in absence (−) or presence (+) of Dox for 24 h. Inputs and pulled-down protein complexes were analysed by immunoblotting. (f) Three microgram of the CtIP peptide coupled to streptavidin beads was incubated with lysates of HEK293^Cas9/wt and HEK293^{Cas9/KLHL15Δ} cells. Inputs and pulled-down protein complexes were analysed by immunoblotting. (g) Recombinant MBP-KLHL15 was coupled to amylose beads and incubated with lysates of HEK293T cells in the absence (−) or presence of the indicated amounts of the CtIP 15-mer peptide. Inputs and pulled-down protein complexes were analysed by immunoblotting.

Figure 6. Identification of a conserved FRY motif in vertebrate CtIP essential for KLHL15 interaction.

(a) Alignment of the KLHL15-interacting region in human CtIP (aa 830–844) with the corresponding region in CtIP/Ctp1/Com1 orthologs of other species. A part of the KLHL15 binding domain in human PP2A/B'-beta is shown below. The rectangular box represents the putative KLHL15-binding motif (FRY). The highly conserved tyrosine residue (Y842 in human CtIP) is highlighted in red. Other, highly conserved amino acid residues are marked in bold typeface. (b) Bacterially expressed GST fusion proteins of the CtIP CTD (aa 790–897), either wt or Y842A mutant, were coupled to glutathione sepharose beads and incubated with lysates from HEK293T cells transfected with the FLAG-KLHL15 expression plasmid for 48 h. Pulled-down protein complexes and the input were analysed by immunoblotting. (c) HEK293T cells were transfected with either empty vector (EV) or the GFP-CtIP expression constructs. 48 h after transfection, cells were lysed and whole-cell extracts were subjected to IP using anti-GFP affinity resin. Inputs and recovered protein complexes were analysed by immunoblotting. (d) Recombinant MBP-KLHL15 was coupled to amylose beads and incubated with lysates from HEK293T cells transfected with the indicated GFP-CtIP expression constructs for 48 h. Inputs and pulled-down protein complexes were analysed by immunoblotting. (e) HEK293T cells were cotransfected with the indicated GFP-CtIP constructs and His-Ubiquitin. Twenty-four hours post transfection cells were treated with MG-132 (20 μM) for 4 h. Cells were then lysed in buffer containing guanidium-HCl and ubiquitin conjugates were pulled-down using Ni-NTA-agarose beads, eluted and analysed by immunoblotting with anti-GFP antibody. (f) HEK293T cells were transfected with CtIP siRNA and 24 h later cotransfected with the indicated siRNA-resistant GFP-CtIP expression constructs and FLAG-KLHL15. Forty-eight hours post siRNA transfection cells were analysed by immunoblotting (left). The GFP-CtIP signal intensities were quantified using ImageJ and represented as EV/FLAG-KLHL15 ratios (right). Data are represented as mean values of densitometric quantification±s.e.m. (n≥3). Asterisks indicate neddylated CUL3.

Figure 7. KLHL15-dependent CtIP ubiquitination governs DNA-end resection and DSB repair pathway choice.

(a) U2OS cells stably expressing doxycycline (Dox)-inducible siRNA-resistant GFP-CtIP-wt or GFP-CtIP-Y842A were transfected with CtIP siRNA. 24 h later, GFP-CtIP expression was induced with Dox. Forty-eight hours post siRNA-transfection, cells were either mock-treated or treated with CPT (1 μM) for 1 h. Total cell extracts and chromatin-enriched fractions were analysed by immunoblotting. (b) Same cells as in a were transfected with the indicated siRNA oligos. One day later, GFP-CtIP expression was induced for 24 h and cells were treated with CPT for 1 h before western blot analysis. (c) Same cells as in b were simultaneously labelled with BrdU (30 μM). 48 h post siRNA-transfection, cells were treated with CPT for 1 h, immunostained with anti-RPA2 or anti-BrdU antibody and analysed by FACS. Dot plots represent the intensity of the signals for RPA2 or BrdU staining (y axis) against the DNA content (x axis). Quantification gates were established in untreated samples and the percentage of cells within the gates are indicated. (d) HEK293^Cas9/wt and HEK293^{Cas9/KLHL15Δ} cells were transfected with the indicated siRNA oligos. Forty-eight hours later, cells were treated with CPT for 1 h and analysed by immunoblotting. (e) Same cells as in d were processed for FACS analysis as in c. (f) Same cells as in d were irradiated at 10 Gy and 3 h later analysed by immunoblotting. SE and LE denote short and long exposures of the same membrane. (g) Same cells as in d were irradiated with increasing doses of IR and survival was determined by colony-formation assay. Data are presented as the mean±s.d. (n=4). (h,i) HEK293 EJ5-GFP or DR-GFP cells were transfected with the indicated siRNAs. Two days later, cells were transfected with the I-SceI expression plasmid and harvested after 48 h for flow cytometry and immunoblot analysis. Data in h and i are represented as mean±s.e.m. (n=3) and as mean±s.d. (n=4), respectively. Unpaired, two-tailed t-tests were performed and P values expressed as *(P<0.05) and **(P<0.005) were considered significant. Asterisks indicate hyperphosphorylated forms of CtIP and RPA2. A.u., arbitrary units; FACS, fluorescence-activated cell sorting.

Figure 8. PIN1 isomerase facilitates KLHL15-dependent degradation of CtIP.

(a) HEK293T cells were cotransfected with FLAG-KLHL15 and either GFP-CtIP-wt, -S276A/T315A (2A, defective in PIN1 interaction), or -K467A (KA, defective in Cdh1 interaction) expression constructs. Forty-eight hours post transfection, cells were analysed by immunoblotting (left). Relative CtIP protein levels were determined by quantification of CtIP band intensity (normalized to Actin) with the ImageJ software (right). Data are represented as mean values of densitometric quantification±s.e.m. (n≥3). Statistical analysis was carried out using unpaired, two-tailed t-tests. P values expressed as *(P<0.05) was considered significant. (b) HEK293T cells were transfected twice with the indicated siRNA oligos for two consecutive days. Forty-eight hours after the first siRNA transfection, cells were transfected with either empty vector (EV) or FLAG-KLHL15 expression constructs. 72 h after the first siRNA transfection, cells were lysed and whole-cell extracts were subjected to IP using anti-FLAG M2 affinity resin. Inputs and recovered protein complexes were analysed by immunoblotting. The asterisk indicates neddylated CUL3. (c) HEK293^Cas9/wt and HEK293^{Cas9/KLHL15Δ} cells were transfected with EV or HA-PIN1 expression construct. Forty-eight hours after transfection, cells were either mock-treated or treated with CPT (1 μM) for 1 h. Whole-cell lysates were analysed by immunoblotting using the indicated antibodies. SE and LE denote short and long exposures of the same membrane. Asterisks indicate hyperphosphorylated forms of CtIP and RPA2, respectively.

Figure 9. CtIP ubiquitination by CUL3-KLHL15 promotes CtIP proteasomal degradation to fine-tune DNA-end resection and DSB repair pathway choice.

Schematic model depicting the molecular architecture of the CUL3-KLHL15 E3 ubiquitin ligase complex and the functional consequences of KLHL15-mediated CtIP protein turnover in controlling the balance between NHEJ and HR in normal KLHL15-expressing cells (upper part) and under pathological conditions of altered KLHL15 expression (lower part). CtIP isomerization by PIN1 triggers CtIP proteasomal degradation at least in part by facilitating CtIP ubiquitination by the CUL3-KLHL15 E3 ligase complex. See discussion for more details.