Inhibition of USP1 enhances anticancer drugs-induced cancer cell death through downregulation of survivin and miR-216a-5p-mediated upregulation of DR5

Abstract

Ubiquitin-specific protease 1 (USP1) is a deubiquitinase involved in DNA damage repair by modulating the ubiquitination of major regulators, such as PCNA and FANCD2. Because USP1 is highly expressed in many cancers, dysregulation of USP1 contributes to cancer therapy. However, the role of USP1 and the mechanisms underlying chemotherapy remain unclear. In this study, we found high USP1 expression in tumor tissues and that it correlated with poor prognosis in RCC. Mechanistically, USP1 enhanced survivin stabilization by removing ubiquitin. Pharmacological inhibitors (ML23 and pimozide) and siRNA targeting USP1 induced downregulation of survivin expression. In addition, ML323 upregulated DR5 expression by decreasing miR-216a-5p expression at the post-transcriptional level, and miR-216a-5p mimics suppressed the upregulation of DR5 by ML323. Inhibition of USP1 sensitized cancer cells. Overexpression of survivin or knockdown of DR5 markedly prevented the co-treatment with ML323 and TRAIL-induced apoptosis. These results of in vitro were proved in a mouse xenograft model, in which combined treatment significantly reduced tumor size and induced survivin downregulation and DR5 upregulation. Furthermore, USP1 and survivin protein expression showed a positive correlation, whereas miR-216a-5p and DR5 were inversely correlated in RCC tumor tissues. Taken together, our results suggest two target substrates of USP1 and demonstrate the involvement of survivin and DR5 in USP1-targeted chemotherapy.

Fig. 1. USP1 is upregulated in human renal clear carcinoma (RCC) tissues.

A Examination of USP1 protein expression in 40 paired primary RCC tissues and corresponding normal adjacent tissues. B Analysis of prognostic significance of USP1 in patients with RCC based on The Cancer Genome Atlas (TCGA, n = 527, upper) and Therapeutically Applicable Research to Generate Effective Treatments (TARGET, n = 124, lower). Investigation of apoptosis-related protein expression in cancer cell lines treated to 10–30 μM ML323 (C) or 2–7.5 μM pimozide (D) for 24 h. E, F The cancer cells were transfected with control siRNA or USP1 siRNA for 24 h. Protein expression was measured using western blotting (A, C–F).

Fig. 2. USP1 directly interacts and induces deubiquitination of survivin.

A Caki-1 cells were treated with 30 μM ML323 for the indicated time periods, and survivin mRNA expression were analyzed using RT-PCR and qPCR. B Caki-1 cells were pretreated with 20 μg/mL cycloheximide (CHX) for 30 min and then treated with 30 μM ML323 for the indicated time periods. C Caki-1 cells were transfected with control siRNA or USP1 siRNA followed by treatment with 20 μg/mL CHX for the indicated time periods. D Caki-1 cells were pretreated with 0.25 μM MG132 for 30 min and then treated with 30 μM ML323 for 24 h. E Caki-1 cells were transfected with the HA-ubiquitin (HA-Ub) followed by combinations of 0.25 μM MG132 and 30 μM ML323 for 12 h. Ubiquitination assay of survivin was performed using an anti-survivin antibody. F Investigation of interaction between USP1 and survivin for IP assay using an anti-survivin antibody. G Caki-1 cells were transfected with vector, USP1 WT or USP1 C90S and HA-ubiquitin followed by 0.25 μM MG132 for 12 h. Protein expression was measured using western blotting (B–G). The band intensity of survivin protein was quantified by using ImageJ.

Fig. 3. ML323 upregulates DR5 expression at the post-transcription level.

A Investigation of DR5 surface expression in Caki-1 cells treated to 30 μM ML323 for 12 h. B Investigation of DR5 mRNA expression using RT-PCR and qPCR in Caki-1 cells treated to 30 μM ML323 for the indicated time periods. C Caki-1 cells were transiently transfected with DR5 (−605) or DR5 (SacI) promoter plasmids, treated with 30 μM ML323 for 12 h, and analyzed by luciferase activity assay. D Caki-1 cells were pretreated with 2 μg/mL actinomycin D (Act D) for 30 min and then treated with 30 μM ML323 for the indicated time periods. DR5 mRNA expression were analyzed using RT-PCR and qPCR. Values in the graphs (A–D) represent the mean ± SD of three independent experiments. *P < 0.01 compared to the control. #P < 0.01 compared to Act D.

Fig. 4. Decrease of miR-216a-5p is involved in ML323-mediated DR5 mRNA upregulation.

A Investigation of microRNAs in Caki-1 cells treated to 30 μM ML323 for 24 h. B Caki-1 cells were transfected with miR-negative control (NC), miR-7-5p mimic, miR-221-3p mimic, miR-216a-5p mimic, or miR-21-3p mimic followed by 30 μM ML323 for 24 h. C, D Caki-1 cells were transfected with miR-NC or miR-216a-5p mimic followed by 30 μM ML323 for 24 h. DR5 protein and mRNA expression were analyzed using western blotting (C) and qPCR (D). E Caki-1 cells were transfected with luciferase reporter gene for DR5 3′UTR WT or mutant followed by 30 μM ML323 for 24 h and analyzed using luciferase activity assay. F Caki-1 and A549 cells were transfected with miR-216 promoter plasmid followed by 30 μM ML323 for 12 h. Protein expression was measured by western blotting (B, C). Values in the graphs (A, D–F) represent the mean ± SD of three independent experiments. *P < 0.01 compared to the control. #P < 0.01 compared to ML323 treated miR-NC-transfected group. **P < 0.01 compared to miR-216a-5 mimic-treated DR5 3′UTR WT.

Fig. 5. ML323 increases sensitivity to anticancer drugs-mediated apoptosis in cancer cells.

A, B Caki-1 cells were treated with a combination of 1 μM doxorubicin, 3 μg/mL etoposide, 30 μM cisplatin, 200 nM carboplatin, 50 ng/ml TRAIL, and 500 ng/mL anti-Fas in the presence or absence of 30 μM ML323 for 24 h. Cell viability was analyzed using the 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxyanilide (XTT) assay kit (A). C–E Caki-1 cells were treated with 30 μM ML323, 50 ng/mL TRAIL or combination for 24 h. Nuclear condensation was determined using DAPI staining. scale bar: 20 μm (D). DEVDase (caspase-3) colorimetric assay using DEVDase substrate (E). F Caki-1 cells were pretreated with 20 μM zVAD for 30 min and then treated with combination of 30 μM ML323 and 50 ng/mL TRAIL for 24 h. G, H Cancer (G) or normal cell (H) lines were treated with 30 μM ML323, 50 ng/mL TRAIL or combination for 24 h. Cell morphology was assessed using a microscope. scale bar: 50 μm (H). The sub-G1 population and protein expression were determined using flow cytometry and western blotting, respectively (B, C, F–H). Values in the graphs (A–C, E–H) represent the mean ± SD of three independent experiments. *P < 0.01 compared to the control. #P < 0.01 compared to combination of ML323 and TRAIL.

Fig. 6. Downregulation of survivin and upregulation of DR5 contribute to ML323-mediated TRAIL sensitization.

Caki-1 and HCT116 cells were transfected with vector and flag-survivin (A) or control siRNA or DR5 siRNA (B) followed by 30 μM ML323, 50 ng/mL TRAIL, or combination for 24 h. The sub-G1 population and protein expression were determined using flow cytometry and western blotting, respectively (A, B). Values in the graphs (A, B) represent the mean ± SD of three independent experiments. *P < 0.01 compared to combination of ML323 and TRAIL in vector. #P < 0.01 compared to combination of ML323 and TRAIL-treated control siRNA-transfected group.

Fig. 7. Depletion of USP1 contributes TRAIL sensitization.

A Caki-1 and HCT116 cells were transfected with control siRNA or USP1 siRNA followed by 50 ng/mL TRAIL for 24 h. B Caki-1 cells were transfected with vector or pcDNA3β-USP1 followed by combination of 30 μM ML323 and 50 ng/mL TRAIL for 24 h. The sub-G1 population and protein expression were determined using flow cytometry and western blotting, respectively (A, B). Values in the graphs (A, B) represent the mean ± SD of three independent experiments. *P < 0.01 compared to TRAIL-treated control siRNA-transfected group. #P < 0.01 compared to combinations of ML323 and TRAIL-treated pcDNA3β-USP1-transfected group.

Fig. 8. Anti-tumor effect of combination of ML323 and TRAIL in xenograft model.

A–C Mice bearing HCT116 cells were treated with 10 mg/kg ML323, 3 mg/kg GST-TRAIL, combinations, or vehicle (i.p.) for 21 days. Tumor volume and size were measured (A). Terminal deoxynucleotidyl transferase (TdT) dUTP Nick-End Labeling (TUNEL) assays were performed to check apoptosis in vivo. scale bar: 50 μm (B). Protein expression was measured by western blot analysis (C). D Examination of survivin and DR5 protein expression in 40 paired primary RCC tissues and corresponding normal adjacent tissues. E Correlation analysis of protein expression of USP1/survivin. F Examination of DR5 (left panel) and miR-216a-5p (right panel) mRNA expression in 40 paired primary renal cancer cell (RCC) tissues and corresponding normal adjacent tissues. G Analysis of prognostic significance of survivin and DR5 in RCC patients based on TCGA. *P < 0.01 compared to the vehicle. i.p. intraperitoneal injection.

Fig. 9. Schematic representation of the mechanism of sensitization to anticancer drugs by USP1 depletion.

Two signaling pathways are involved in ML323-mediated TRAIL sensitization. (1) ML323 decreases miR-216a-5p expression at the transcription level, stabilizes DR5 mRNA, and increases DR5 expression on cancer cell surface. (2) USP1, highly expressed in renal cancer, interacts with survivin and induces deubiquitination of survivin. When USP1 is inhibited by ML323, survivin is ubiquitinated and degraded. Eventually, sensitivity of TRAIL is increased by upregulation of DR5 and downregulation of survivin in ML323 treatment.