USP5 Binds and Stabilizes EphA2 to Increase Nasopharyngeal Carcinoma Radioresistance

Abstract

Radioresistance poses a major challenge in nasopharyngeal carcinoma (NPC) treatment. However, the underlying mechanism of NPC radioresistance remains poorly understood, and the promising radiosensitizer for NPC radiotherapy is also lacked. Overexpression of USP5 and EphA2 has been linked to various cancers, and both the proteins have attracted considerable attention for the development of new anti-cancer drugs. Here, we report that USP5 interacts with EphA2, and increases EphA2 protein stability and expression by ubiquitin proteasome pathway in the NPC cells. Mebendazole (MBZ), a broad-spectrum anthelmintic drug, transcriptionally inhibits USP5 expression, and then promotes EphA2 ubiquitination degradation in the NPC cells. Functionally, USP5 enhances in vitro and in vivo NPC cell radioresistance via stabilizing EphA2, and MBZ decreases in vitro and in vivo NPC cell radioresistance via targeting USP5/EphA2 axis. Moreover, the levels of USP5 and EphA2 are significantly higher in the radioresistant NPCs than those in the radiosensitive NPCs, and both proteins for predicting patient prognosis are superior to individual protein. These findings suggest that USP5 binds and stabilizes EphA2 by ubiquitin proteasome pathway to promote NPC radioresistance, and MBZ increases NPC radiosensitivity by targeting USP5/EphA2 axis, and is a potential radiosensitizer in NPC and perhaps in other cancers.

Keywords: Drug repurposing; EphA2; Mebendazole; Nasopharyngeal carcinoma; Radioresistance; USP5.

Figure 1

Identification and validation of USP5 as a protein interacted with EphA2 in NPC cells. (A) Identification of USP5 as a protein interacted with EphA2 by mass spectrometry (MS). The amino acid sequence of a doubly charged peptide with m/z 589.77740 was identified as SSENPNEVFR, and Mascot search showing the peptide matched with USP5. (B-E) Validation of USP5 as a protein interacted with EphA2. (B) Co-IP showing interaction of endogenous USP5 and EphA2 in the HK1 and 5-8F NPC cells. (C) Co-IP showing interaction of exogenous USP5 and EphA2 in the HEK293 cells co-transfected with Flag-USP5 and Myc-EphA2 expression plasmids. (D) GST pull-down showing direct interaction of USP5 and EphA2. Purified His tagged USP5 protein was mixed with purified GST or GST tagged EphA2 protein immobilized on glutathione beads. Samples were electrophoresed and immunoblotted with antibodies against GST or Histidine. (E) Immunofluorescent staining showing colocalization of EphA2 and USP5 in the HK1 and 5-8F NPC cells. Scale bar, 10 μm. IP, Immunoprecipitation; IB, Immunoblotting.

Figure 2

USP5 stabilizes EphA2 by ubiquitin proteasome pathway in NPC cells. (A) Western blot showing the expression levels of USP5 and EphA2 in the HK1 and 5-8F cells stably transfected with shRNAs against USP5 CDS or 3'UTR, and their shNC control cells. (B) QRT-PCR showing the expression levels of USP5 and EphA2 mRNA in the HK1 and 5-8F cells with stable knockdown of USP5. (C) Western blot showing the effect of USP5 knockdown on EphA2 protein stability in HK1 and 5-8F NPC cells treated with 20µg/mL cycloheximide (CHX) for indicated times. (D) Western blot showing reversion of EphA2 protein levels by proteasome inhibitor MG132 in the HK1 and 5-8F NPC cells with stable knockdown of USP5. Cancer cells were treated with 10 µM MG132 for indicated times. (E) USP5 knockdown increases EphA2 polyubiquitination in the HK1 and 5-8F NPC cells. Cancer cells were treated with 10µM MG132 for 12 hours, and subjected to immunoprecipitation analysis with anti-EphA2 antibody followed by immunoblotting with anti-polyubiquitin antibody. (F) The effect of wild type USP5 (USP5-WT) and catalytically inactive mutant USP5 (USP5-C335A) on EphA2 polyubiquitination in the HEK293 cells. HEK293 cells were co-transfected with indicated plasmids for 48 hours and treated with 10 µM MG132 for another 12 hours, and subjected to immunoprecipitation analysis with anti-EphA2 antibody followed by immunoblotting with anti-polyubiquitin antibody. (G) Western blot showing the effect of USP5-WT and USP5-C335A on EphA2 protein levels in the HEK293 cells co-transfected with indicated plasmids. (H) Type of USP5-decreased EphA2 polyubiquitination. HEK293 cells were transfected with indicated plasmids for 48 hours, treated with 10µM MG132 for 12 hours, and subjected to immunoprecipitation analysis with anti-EphA2 antibody followed by immunoblotting with anti-polyubiquitin antibody. Data represent means ± SD. ***, P < 0.001; ns, no significance. shUSP5, knockdown of USP5 by shRNA; shNC, scramble shRNA negative control; IP, immunoprecipitation; IB, immunoblotting.

Figure 3

Mebendazole degrades EphA2 via inhibiting USP5-mediated EphA2 deubiquitination in NPC cells. Western blot (A) and qRT-PCR (B) showing the protein and mRNA levels of USP5 and EphA2 in the HK1 and 5-8F NPC cells treated with indicated concentrations of MBZ for 48 hours. (C) Dual-luciferase reporter gene assay showing the effect of MBZ on USP5 promoter activity in the HK1 and 5-8F NPC cells. The luciferase reporter plasmid driven by USP5 promoter or control plasmid and pRL-TK plasmid were co-transfected into cancer cells for 12 hours. 36 hours after treated with indicated concentrations of MBZ, USP5 reporter activity was assessed by using the Dual-luciferase reporter assay system. (D) MBZ decreases EphA2 protein stability in the HK1 and 5-8F NPC cells. Cancer cells were treated with 2.4 µM (HK1 cells) or 5.8 µM (5-8F) for 36 hours, followed by treatment with 20µg/mL CHX for indicated times, and subjected to western blot analysis with anti-EphA2 antibody. (E) Western blot showing reversion of EphA2 protein levels by proteasome inhibitor MG132 in the HK1 and 5-8F NPC cells with MBZ. Cancer cells were treated with 2.4 µM (HK1 cells) or 5.8 µM (5-8F) for 36 hours, followed by treatment with 10 µM MG132 for indicated times. (F) USP5 overexpression reverses EphA2 polyubiquitination levels in the NPC cells treated with MBZ. HK1, 5-8F NPC cells and their respective USP5 overexpression cells were treated with IC50 of MBZ, and subjected to immunoprecipitation analysis with anti-EphA2 antibody followed by immunoblotting with anti-polyubiquitin antibody. (G) USP5 overexpression reverses EphA2 protein levels in the NPC cells treated with MBZ. HK1, 5-8F NPC cells and their respective USP5 overexpression cells were treated with IC50 of MBZ, and subjected to western blot analysis with anti-EphA2 antibody. Data represent means ± SD. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, no significance.

Figure 4

USP5 increases in vitro and in vivo NPC cell radioresistance through EphA2 mediation. (A) Clonogenic survival assay showing the radiosensitivity of HK1 and 5-8F NPC cells with USP5 knockdown or USP5 knockdown and EphA2 overexpression (OE). Cancer cells were exposed to 4 Gy ionizing radiation (IR), and colonies that formed after incubation of 12 days were stained with crystal violet. Representative images are shown on the left, and quantitative data are presented on the right. Cells without irradiation served as control. (B) Flow cytometric analysis showing the irradiation-induced apoptotic rate of HK1 and 5-8F NPC cells with USP5 knockdown or USP5 knockdown and EphA2 OE. 48 hours after 4Gy IR, cell apoptosis was detected. Representative results are shown on the left, and quantitative data are presented on the right. Cells without irradiation served as control. (C) Immunofluorescent staining showing the irradiation-induced γH2AX focus number of HK1 and 5-8F NPC cells with USP5 knockdown or USP5 knockdown and EphA2 OE. 4 hours after 4Gy IR, γH2AX foci were detected. Representative images are shown on the left, and quantitative data are presented on the right. (D) Xenografts showing the in vivo radiosensitivity of 5-8F NPC cells with USP5 knockdown, 5-8F NPC cells with USP5 knockdown and EphA2 overexpression and shNC control cells. The photographs (left), growth curves (middle) and weight (right) of xenograft tumors 12 days after initial radiotherapy. 7 days after the inoculation, tumor-bearing mice received a total of 12Gy IR, and radioresponse was estimated after initial treatment. Xenografts without irradiation served as control. shUSP5, knockdown of USP5 by shRNA; EphA2-OE, EphA2 overexpression. **, P < 0.01; ****, P < 0.0001; ns, no significance.

Figure 5

MBZ increases in vitro NPC cell radiosensitivity. (A) Clonogenic survival assay showing the effects of MBZ on the radiosensitivity of HK1 and 5-8F NPC cells. Cells were treated with indicated concentration of MBZ for 48 hours, followed by IR (0-8Gy). Colonies that formed after incubation of 12 days were stained with crystal violet and photographed (left), and sensitive enhancement ratio (SER) of MBZ was calculated (right). (B) Flow cytometric analysis showing the effects of MBZ on irradiation-induced apoptosis of HK1 and 5-8F NPC cells. Cancer cells were treated with indicated concentration of MBZ for 48 hours, followed by 4Gy IR. 48 hours after IR, cell apoptosis was detected. Representative results are shown on the left, and quantitative data are presented on the right. Cells without irradiation served as control. (C) Immunofluorescent staining showing the effects of MBZ on irradiation-induced γH2AX focus number of HK1 and 5-8F NPC cells. Cancer cells were treated with indicated concentration of MBZ for 48 hours, followed by 4Gy IR. 4 hours after IR, γH2AX foci were detected. Representative images are shown on the left, and quantitative data are presented on the right. *, P < 0.05; **, P < 0.01; ****, P < 0.0001; ns, no significance.

Figure 6

Mebendazole increases NPC cell radiosensitivity via targeting USP5/EphA2 axis. (A) USP5 or EphA2 overexpression (OE) reverses the effect of MBZ on colonic formation ability in HK1 and 5-8F NPC cells treated with IR. Cancer cells were treated with IC50 of MBZ for 48 hours, followed by 4Gy IR, and subjected to clonogenic survival assay. Colonies that formed after incubation of 12 days were stained with crystal violet. Representative images are shown on the left, and quantitative data are presented on the right. (B) USP5 or EphA2 OE reverses the effect of MBZ on irradiation-induced apoptosis of HK1 and 5-8F NPC cells treated with IR. Cancer cells were treated with IC50 of MBZ for 48 hours, followed by 4Gy IR. 48 hours after IR, cell apoptosis was detected by flow cytometry. (C) USP5 or EphA2 OE reverses the effect of MBZ on γH2AX focus number of HK1 and 5-8F NPC cells treated with IR. Cancer cells were treated with MBZ for 48 hours, followed by 4Gy IR. 4 hours after IR, γH2AX foci were detected by immunofluorescent staining. (D-E) MBZ enhances in vivo NPC cell radiosensitivity via inhibiting USP5/EphA2 axis. A schematic diagram illustrating the treatment plan of MBZ or/and IR in the nude mice with xenografts generated from 5-8F NPC cells (D). The photographs of xenograft tumors 12 days after initial treatment (E). The growth curves of xenograft tumors after initial treatment (F). Summary of weight data of xenograft tumors 12 days after initial treatment (G). **, P < 0.01; ****, P < 0.0001; ns, no significance.

Figure 7

Expression levels of USP5 and EphA2 correlate with NPC radiosensitivity and patient prognosis. (A) Positive correlation between alterations for USP5 and EphA2 expressions in NPC tissues. Representative IHC images of low and high expression of USP5 and EphA2 in NPC tissues are shown on the left, and statistical analysis of the correlation between USP5 and EphA2 expressions is presented on the right (P<0.001, Spearman correlation test). (B) Representative IHC images of USP5 and EphA2 expression in the radiosensitive and radioresistant NPC tissues. (C) Histograms showing the proportion of radiosensitive and radioresistant NPCs based on the expression levels of USP5, EphA2, or both proteins. Statistical differences were determined by χ2 test. (D) Survival analysis of the NPC patients. Kaplan-Meier survival analysis of disease-free survival (top) and overall survival (bottom) for 119 NPC patients based on the expression levels of USP5, EphA2 or both proteins. Log-rank test was used to calculate P value. (E) A model for radiosensitization of Mebendazole by targeting USP5/EphA2 axis in NPC. In NPC cells, USP5 binds and stabilize EphA2 by inhibiting its ubiquitinaion degradation, promoting NPC cell radioresistance (left). Mebendazole increases NPC cell radiosensitivity by transcriptionally inhibiting USP5 expression and then targeting EphA2 degradation (right). Scale bars = 50 μm. **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.