The Kaplan-Meier Plotter (http://kmplot.com/analysis/) was used to analyze the association between USP14 expression and progression-free survival (PFS) of patients with ovarian cancer. The following datasets were used for the analysis: GSE14764 (30), GSE15622 (31), GSE18520 (32), GSE19829 (33), GSE23554 (34), GSE26193 (35,36), GSE26712 (37,38), GSE27651 (39), GSE30161 (40), GSE3149 (41), GSE51373 (42), GSE63885 (43), GSE65986 (44), GSE9891 (45) and TCGA. ‘USP14’ was input in Affy id/Gene symbol and chose ‘Affy ID: 201671_x_at’. The median expression level of USP14 mRNA was regarded as the dividing line and 1435 patients were divided into low expression group and high expression group. Other parameters remain unchanged by default and draw Kaplan-Meier plot. P-values <0.05 were considered to indicate a statistically significant result.

Cisplatin (cat. no. P4394) was obtained from Sigma-Aldrich; Merck KGaA. IU1 [1-[1-(4-fluorophenyl)-2,5-dimethylpyrrol-3-yl]-2-pyrrolidin-1-ylethanone] (cat. no. S7134) (a specific small-molecule inhibitor of USP14) was purchased from Selleck Chemicals (Houston, TX, USA).

The mouse monoclonal anti-Cx32 (cat. no. 59948) antibody was purchased from Santa Cruz Biotechnology, Inc. (Dallas, Texas, USA). The mouse monoclonal anti-β-tubulin (cat. no. 86298), mouse monoclonal anti-β-actin (cat. no. 3700), rabbit monoclonal anti-USP14 (cat. no. 11931), rabbit monoclonal anti-caspase-3 (cat. no. 9662), rabbit monoclonal anti-cleaved caspase-3 (cat. no. 9664), rabbit anti-caspase-9 (cat. no. 9502), rabbit anti-Na,K-ATPase (cat. no. 3010) and mouse monoclonal IgG1 (cat. no. 5415S) antibodies were purchased from Cell Signaling Technology, Inc. (Danvers, MA, USA). Peroxidase-AffiniPure goat anti-mouse IgG (H+L) (cat. no. 115-035-003) and Peroxidase-AffiniPure goat anti-rabbit IgG (H+L) (cat. no. 111-035-003) were obtained from Jackson (West Grove, PA, USA). The rabbit monoclonal anti-ubiquitin (cat. no. EPR8830) was purchased from Abcam (Cambridge, UK). The secondary antibody IPKine HRP, mouse anti-rabbit IgG LCS (cat. no. A25002) was purchased from Abbkine Scientific Co., Ltd. (Wuhan, China).

A2780 cisplatin-sensitive cells (A2780) were obtained from the American Type Culture Collection (ATCC; Manassas, VA, USA). The A2780 cisplatin-resistant cells (A2780-CDDP) were established by a previously described method (11). Both A2780 and A2780-CDDP cell lines were cultured in Dulbecco's modified Eagle's medium (DMEM) (cat. no. 12800017; Gibco; Thermo Fisher Scientific, Inc.) containing 10% fetal bovine serum (FBS) (Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin/streptomycin at 37°C in an atmosphere containing 5% CO₂.

After growing A2780 cells to 30–50% confluence, 50 nM of non-specific siRNA (#siN0000001-1-5) and siRNAs targeting the human USP14 gene were transfected into cells with Lipofectamine 3000 transfection reagent (cat. no. L3000015; Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. The sequences of the synthesized siRNAs targeting USP14 (Guangzhou RiboBio, Co., Ltd., Guangzhou, China) were as follows: siUSP14-1, 5′-CTGGCATATCGCTTACGTT-3′; siUSP14-2, 5′-TCCAGTATTCCACCTATTA-3′; and siUSP14-3: 5′-TTGCCGAGAAAGGTGAACA-3′.

To generate stably transfected cells, A2780 cells were transfected with lentiviral plasmids containing the shCx32 sequence or with the negative control vector (pLVX-shRNA-tdTomato-Puro), which were constructed by Landbiology, Co., Ltd. (Guangzhou, Guangdong, China). Lentiviral particles were prepared by transfecting 293T cells with the lentiviral plasmids, PAX2 and pMD2.G at a defined ratio. A2780 cells were incubated with medium containing the virus and Polybrene for 48 h. After infection, cells were selected by culturing with puromycin (2 µg/ml) for 2 weeks. The sequences of the short hairpin RNA targeting Cx32 (shCx32) were as follows: 5′-GCTGCAACAGCGTTTGCTACTCGAGTAGCAAACGCTGTTGCAGCTTTTTTT-3′. shRNA with non-specific sequences (5′-TTCTCCGAACGTGTCACGTTTCTCGAGAAACGTGACACGTTCGGAGAA-3′) was used as the control scrambled RNA.

Cell Counting Kit-8 (CCK-8) (Dojindo Molecular Technologies Inc., Kumamoto, Japan) was used to examine cell viability according to the manufacturer's instructions. Cells were seeded in 96-well plates at a suitable density of 5,000 cells/well. Then, siUSP14-1 was added for 24 h and the cells were treated with cisplatin for 48 h or the cells were cotreated with IU1 and cisplatin for 48 h. CCK-8 solution diluted with FBS-free DMEM at a ratio of 1:9 was added in a total volume of 100 µl. The 96-well plates were incubated at 37°C for 1–2 h. The optical density (OD) was determined at 450 nm using an Epoch microplate spectrophotometer (BioTek; Winooski, VT, USA). The concentration of cisplatin resulting in 50% growth inhibition (IC₅₀) was calculated using GraphPad Prism 6.0 software (GraphPad Software, Inc.).

Cells were lysed at 4°C in buffer [1 mM β-glycerophosphate, 2.5 mM sodium pyrophosphate, 20 mM Tris-HCl (pH 7.4), 1% Triton X-100, 150 mM NaCl, 1 mM EGTA, 1 mM EDTA and 1 mM Na₃VO₄] supplemented with protease inhibitor cocktail (cat. no. P8340; Merck KGaA) (1:1,000). After ultrasonication, cell lysates were cleared at 12,000 × g for 30 min at 4°C. The protein concentration was determined with a BCA protein assay kit (Thermo Fisher Scientific, Inc.). A total of 20 µg of protein from each sample was separated using SDS-PAGE and transferred to PVDF membranes. Membranes were blocked with 5% (w/v) skim milk in wash buffer TSBT (TBS and 0.05% Tween-20) for 1 h and were then incubated with specific antibodies against Cx32 (dilution 1:2,000), USP14 (dilution 1:1,000), cleaved caspase-3 (dilution 1:1,000), caspase-3 (dilution 1:1,000), caspase-9 (dilution 1:1,000), Na,K-ATPase (dilution 1:1,000), β-actin (dilution 1:10,000) and β-tubulin (dilution 1:10,000) overnight at 4°C. After being washed for three times with TBST, membranes were incubated with HRP-conjugated secondary antibodies (dilution 1:10,000) for 2 h at room temperature. The membranes were then washed with TBST before being visualized using a Chemiluminescent HRP Substrate Kit (Millipore; Billerica, MA, USA) and scanned using an ImageQuant LAS 4000™ (GE Healthcare). β-tubulin and β-actin were used as loading control, and band densities were quantified using ImageJ software (NIH, National Institutes of Health, Bethesda, MD, USA).

DUB activity can be inhibited irreversibly using ubiquitin-vinylsulfone (Ub-VS), which forms an adduct with the active site cysteine in DUB of the thiol protease class (22). Cells were lysed via mild sonication in ice-cold buffer containing 50 mM Tris (pH 7.4), 5 mM MgCl₂, 250 mM sucrose, 1 mM DTT, 2 mM ATP and 1 mM PMSF. Lysates were cleared by centrifugation at 12,000 × g for 30 min at 4°C, and 30 µg of protein extract was incubated for 15 min at 37°C with 1 µM HA-Ub-VS (cat. no. U-212; Boston Biochem, Cambridge, MA, USA). After boiling in loading buffer, labeled cell lysates were subjected to immunoblot analysis. After transferring to PVDF membranes, HA-Ub-VS labeled USP14 was immunodetected using the USP14 antibody. The probe group was a blank control and β-actin was used as a loading control. The band densities of USP14-Ub-VS were quantification for the activity of USP14 (46).

A2780 cells were seeded in 6-well plates and incubated with IU1 (50 µM), cisplatin (8 µg/ml) or cotreated with IU1 (50 µM) and cisplatin (8 µg/ml) for 48 h. Afterward, the cells were trypsinized and washed twice with cold PBS. After the cells were re-suspended in binding buffer, Annexin V-FITC and propidium iodide (PI) (cat. no. KGA105-KGA108; Keygen; Nanjing, Jiangsu, China) were used to stain cells for 15 min away from light at room temperature. Subsequently, the cells were immediately analyzed with FACScan (Beckman Instruments, Fullerton, CA, USA) and cell apoptosis was analyzed using FlowJo 7.6 software (FlowJo LLC, Ashland, OR, USA).

Approximately 1.5×10⁵ A2780 cells were seeded into 6-well plates. After adherence, the cells were incubated with IU1 or siUSP14-1 for 48 h. Total RNA was extracted using TRIzol (cat. no. A33252; Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. The concentration of RNA was measured by NanoDrop 2000 Spectrophotometer (Thermo Fisher Scientific, Inc.). Then 1 µg total RNA was reverse transcribed using a Transcriptor cDNA Synthesis kit (cat. no. 4896866001; Roche; Basel, Switzerland) by C1000 Thermal Cycler (Bio-Rad Laboratories, Inc.; Hercules, CA, USA). The resulting cDNA was subjected to real-time qPCR in a final reaction volume of 20 µl using FastStart Universal SYBR Green Master (Rox) (cat. no. 04913914001; Roche) by the ABI Applied Biosystems StepOne Quantitative Real-Time PCR System (Thermo Fisher Scientific, Inc.). The thermocycling conditions were as follows: 10 min at 95°C, followed by denaturation at 95°C for 15 sec; annealing and extension at 60°C for 60 sec. Primers were as follows: USP14 (forward, 5′-GGCTTCAGCGCAGTATATTA-3′ and reverse, 5′-CAGATGAGGAGTCTGTCTCT-3′); Cx32 (forward, 5′-ACACCTTGCTCAGTGGCGTGA-3′ and reverse, 5′-GGGACCACAGCCGCACATGG-3′); and GAPDH (forward, 5′-GGAGCGAGATCCCTCCAAAAT-3′ and reverse, 5′-GGCTGTTGTCATACTTCTCATGG-3′). The analysis for USP14 and normalization control gene GAPDH was performed according to the 2^−ΔΔq method (47).

A2780 cells were treated with IU1 (50 µM for 48 h) and then incubated with 10 µM MG132 (cat. no. S2619; Selleck Chemicals) for 4 h prior to harvesting in buffer [1 mM β-glycerophosphate, 2.5 mM sodium pyrophosphate, 20 mM Tris-HCl (pH 7.4), 1% Triton X-100, 150 mM NaCl, 1 mM EGTA, 1 mM EDTA and 1 mM Na₃VO₄] supplemented with protease inhibitor cocktail (1:1,000). The samples were then centrifuged at 12,000 × g for 30 min, and the supernatants were used for immunoprecipitation. Briefly, 1–2 µl of mouse monoclonal anti-Cx32 was added to 500–1,000 µg of protein and incubated for 3 h at 4°C. Non-specific antibodies were used as controls. Then, 20 µl of Protein G plus/Protein A agarose suspension (cat. no. IP05; Merck KGaA) was added to the mixture. Following incubation for more than 12 h at 4°C, the samples were centrifuged at 12,000 × g for 3 min, and the protein G-Sepharose precipitate was washed 5 times in an appropriate wash buffer (500 mM NaCl, 50 mM Tris-HCl, 6 mM EDTA and 1% Triton X-100, pH 8.3), resuspended in 4X loading buffer, and denatured at 100°C for 5 min. The input represented 10% of the total amount of protein in the lysates before immunoprecipitation. All samples were subjected to immunoblot analysis. After separation by SDS-PAGE, the proteins samples were transferred to PVDF membrane. Membranes were blocked with 5% (w/v) skim milk in wash buffer (TBS and 0.05% Tween-20) for 1 h and were then incubated with specific antibodies against ubiquitin (dilution 1:1,000). The secondary antibody IPKine HRP, mouse anti-rabbit IgG LCS (dilution 1:5,000) was used to incubated the membrane for eliminating heavy chain interference.

GJIC was evaluated by a ‘parachute’ dye coupling assay, as described by Goldberg et al (48) and Koreen et al (49). Calcein-AM (cat. no. C3100MP) and CM-DiI (cat. no. C3100MP) were purchased from Invitrogen (Thermo Fisher Scientific, Inc.). Cells were seeded in 12-well plates at an appropriate density of 50,000 cells/well. After confluence, the donor cells were labeled with both Calcein-AM (green fluorescence, GJ permeable) and CM-DiI (red fluorescence, non-permeable) at 37°C for 30 min. After being rinsed and trypsinized, 500 donor cells were seeded onto the receiver cells per well and incubated at 37°C for 4–6 h. Signal intensity was observed using a fluorescence microscope (Olympus IX71; Tokyo, Japan). The average number of receiver cells (green fluorescence) around every donor cell (both green and red fluorescence) was recorded as an index of GJIC.

Approximately 2–3×10⁶ A2780 cells were seeded in a 10-cm culture dish. IU1 or siUSP14-1 was administered for 48 h when the confluence was 30–40%. Membrane-bound and cytoplasmic proteins were extracted by using a ProteoExtract Transmembrane Protein Extraction Kit (cat. no. 71772; Merck KGaA) according to the manufacturer's instructions. Cells were collected in PBS and centrifuged at 1,000 × g for 5 min at 4°C. Cells were resuspended in Extraction Buffer 1 containing protease inhibitor cocktail and incubated for 10 min at 4°C. The cytosolic (soluble) protein fraction was harvested after centrifugation at 1,000 × g for 5 min at 4°C. The pellet was resuspended in Extraction Buffer 2 containing TM-PEK reagent B and protease inhibitor cocktail. The mixture was incubated for 45 min at 4°C with gentle agitation, and membrane proteins were obtained in the supernatant.

Every in vitro experiment was performed with a minimum of three independent cell cultures. The data were statistically analyzed by Student's t-test (2 groups), one-way ANOVA (>2 groups), followed by Tukey's multiple comparison test or two-way ANOVA (>2 groups), followed by Bonferroni post test with GraphPad Prism 6.0 software. Histograms or scatter diagrams were constructed with the GraphPad Prism 6.0 software (GraphPad Software, Inc.). The results are expressed as the mean ± SE, and P<0.05 was considered indicative of statistical significance.

A cohort of 1435 cases of ovarian cancer of following datasets were used for Kaplan-Meier analysis: GSE14764 (30), GSE15622 (31), GSE18520 (32), GSE19829 (33), GSE23554 (34), GSE26193 (35,36), GSE26712 (37,38), GSE27651 (39), GSE30161 (40), GSE3149 (41), GSE51373 (42), GSE63885 (43), GSE65986 (44), GSE9891 (45) and TCGA. The results showed that decreased USP14 expression is associated with poorer progression-free survival (PFS) of patients with ovarian cancer (P<0.001; Fig. 1A). Then we performed experiments to explore the role of USP14 in cisplatin resistance of ovarian cancer.

Activity and expression of USP14 were decreased in A2780-CDDP cells. The A2780-CDDP cells were established by a previously described method (11). The IC₅₀ of the A2780 cells used in present study was 5.946±0.61 µg/ml, and the IC₅₀ of the A2780-CDDP cells was 32.45±2.52 µg/ml. The resistance index (RI) was 5.46. A2780-CDDP cells were moderately resistant to cisplatin and qualified for subsequent experiments (Fig. 1B). As showed in Fig. 1C, the expression of Cx32 was significantly higher while the expression of USP14 was significantly lower in the A2780-CDDP cells when compared to the expression levels in A2780 cells. Furthermore, the results of DUB trap assay showed that USP14 activity was significantly decreased in the A2780-CDDP cells (Fig. 1D).

Next we sought to determine whether USP14 is involved in cisplatin resistance. It has been reported that IU1 specifically inhibits USP14 by preventing its docking on the proteasome (22). Therefore IU1 was used as a USP14 inhibitor. A2780 cells were incubated with gradient concentrations (5, 10, 25, 50 and 100 µM) of IU1 for 48 h. The results of CCK-8 assay indicated that 100 µM IU1 significantly inhibited A2780 cell proliferation while 0–50 µM IU1 had no effect (Fig. 2A). Hence, we chose 50 µM IU1 for subsequent studies. As expected, 50 µM IU1 clearly inhibited the activity of USP14 in A2780 cells without changing its protein level (Fig. 2B). The results of CCK-8 assay revealed that IU1 markedly counteracted cisplatin cytotoxicity (Fig. 2C). To make the experiments more convincing, siRNAs were used to inhibit USP14 expression. siUSP14-1 showed the greatest efficiency in silencing USP14 expression (Fig. 2D). Moreover, siUSP14-1 decreased both the protein level and activity of USP14 (Fig. 2E). Then A2780 cells were transfected with siUSP14-1 for 24 h, followed by cisplatin administration for 48 h. Similarly, USP14 knockdown decreased cisplatin cytotoxicity in A2780 cells (Fig. 2F).

Additionally, the apoptosis rate of the cells was examined by flow cytometry, and the levels of cleaved caspase-9 and cleaved caspase-3, two well-known protein markers of apoptosis, were assessed by western blotting. As showed in Fig. 2G-I, significant apoptosis of A2780 cells was induced by cisplatin (8 µg/ml, 48 h) but was significantly suppressed by cotreatment with the USP14 inhibitor IU1 (50 µM, 48 h; Fig. 2G and H) or siUSP14-1 (Fig. 2I). Taken together, these results indicated that USP14 inhibition decreased cisplatin-induced apoptosis.

Our previous studies reported that the anti-apoptotic effect of USP14 inhibition was closely related to the upregulation of Cx32 expression in cervical and ovarian cancer (11,13). The present results showed that cisplatin cytotoxicity was also reduced by USP14 inhibition in A2780 cells. Then, we aimed to ascertain whether USP14 inhibition could affect the expression of Cx32. As showed in Fig. 3A and B, IU1 (50 µM, 48 h) and siUSP14-1 alone or in combination with cisplatin upregulated Cx32 expression. To further explore this mechanism, qPCR and immunoprecipitation were performed. The results demonstrated that the mRNA and ubiquitination level of Cx32 remained unchanged. Moreover, the ubiquitination level of Cx32 in A2780 cells was consistent with that in A2780-CDDP cells (Fig. 3C and D). Taken together, USP14 inhibition increased the expression of Cx32 without changing its mRNA and ubiquitination levels.

The above results showed that IU1 and siUSP14-1 alone or in combination with cisplatin upregulated Cx32 expression. To determine the role of Cx32 in the anti-apoptotic effect of USP14 inhibition, Cx32 expression was knocked down with shRNA in A2780 cells (Fig. 4A). As shown in Fig. 4B, cytotoxicity was increased in the Cx32-knockdown cells compared with that in A2780 cells after administration of 2 or 4 µg/ml cisplatin. Similarly, cytotoxicity was increased in the Cx32-knockdown cells compared with that in A2780 cells after co-administration. Although the anti-apoptotic effect of cotreatment remained after Cx32 knockdown, the extent was indeed lessened compared to that in A2780 cells when cisplatin was administered at 2, 4, 8 or 16 µg/ml. Consistent with the above results, Cx32 knockdown in A2780 cells partially abrogated the cisplatin resistance induced by USP14 silencing (Fig. 4C). These findings suggest that Cx32 plays an important role in USP14 inhibition-induced cisplatin resistance.

The upregulation of Cx32 induced by USP14 inhibition was found to contribute to cisplatin insensitivity. We aimed to ascertain whether GJIC and Cx32 localization are affected by USP14 inhibition. As depicted in Fig. 1A, USP14 suppression caused a marked reduction in GJIC (Fig. 5A). In addition, membrane and cytoplasmic proteins were extracted after administration of IU1 or siUSP14-1. As expected, the results demonstrated that Cx32 was predominantly distributed in the cytoplasm and the Cx32 that localized on the membrane was decreased (Fig. 5B and C). In summary, Cx32 internalization caused by USP14 inhibition caused the cisplatin insensitivity in a non-gap junction-mediated manner.