PSMC6 regulation of ovarian cancer cisplatin resistance unravels a new mode for proteasome targeting

Abstract

Ovarian carcinoma has still a poor prognosis. CRISPR/Cas9 loss-of-function screen revealed a relationship between the PSMC6 proteasome subunit expression and survival of cisplatin-sensitive and -resistant ovarian carcinoma cells. Increased levels of PSMC6 were evidenced in multiple ovarian carcinoma cell lines versus normal cells. An association between PSMC6 levels and tumour stages as well as with a reduced progression-free survival was found. Since a PSMC6 interactome analysis evidenced limited knowledge on PSMC6 biology, mechanistic studies were carried out. PSMC6 knockdown indicated reduced cell growth and clonogenicity in cisplatin-sensitive IGROV-1 and -resistant IGROV-1/Pt1 cells, with a higher impact in resistant cells. This behaviour was accompanied by the accumulation of ubiquitinated proteins and down-regulation of ERK1/2 phosphorylation mediated by increased DUSP6. PSMC6 silencing increased sensitivity to cisplatin in IGROV-1/Pt1 cells as shown by clonogenic assay and 3D spheroids. Since PSMC6 knockdown did not change sensitivity to 20S and 19S proteasome inhibitors, we suggest a new mode of proteasome targeting by interference with a proteasome ATPase. Overall, a link between PSMC6 and ovarian carcinoma aggressiveness is envisioned, highlighting PSMC6 as a potential diagnostic and therapeutic target.

Keywords: DUSP6; PSMC6; cisplatin; drug resistance; ovarian carcinoma; proteasome.

Figure 1

CRISPR/Cas9 drop out screen and analysis of mRNA and protein levels in a panel of ovarian carcinoma cell lines. a) Average survival score (ratio) of IGROV-1 (wild-type) cells compared to platinum-resistant (Pt) cells following CRISPR/Cas9 deletion of proteasome subunits. b) Levels of PSMC6 mRNAs in a panel of ovarian carcinoma cell lines. Total RNA was extracted from tumor cells and quantified for PSMC6 using qRT-PCR. ΔCt values were obtained by normalizing the Ct values of the target gene to the endogenous control GAPDH. c) and d) Protein expression of PSMC6 in a panel of ovarian carcinoma cell lines. Protein extracts were analyzed by Western blot. Actin and β-tubulin were used as control for loading. e) Protein expression of PSMC6 in IGROV-1, IGROV-1/Pt1, OSE-SV40 and FT240 cells. Protein extracts were analyzed by Western blot. β-tubulin was used as control for loading. The histograms represent the Western blot quantification. The band intensity was quantified by using the ImageQuant software. The intensities of PSMC6 bands were normalized to the intensities of the bands of the corresponding control protein.

Figure 2

PSMC6 expression correlations with ovarian cancer. a), b) and c) PSMC6 relative expression distribution. Boxplot representing the expression of the PSMC6 (in terms of -ΔCt) according to diagnosis (a), tumor grade (b) and stage (c). Each box indicates the 25th and 75th centiles of the distribution. The horizontal line inside the box indicates the median and the whiskers indicate the extreme measured values. d) Progression-free survival of ovarian carcinoma patients from the TCGA case material. e) The PSMC6 interaction network built using STRING. The thickness of the lines connecting the hubs represents the edge confidence.

Figure 3

Effects of PSMC6 silencing on cell growth of IGROV-1, IGROV-1/Pt1, TOV112D and TOV21G. Measurement of cell density (cells x 10³/ cm²) after 48 h from transfection in IGROV-1, IGROV-1/Pt1, TOV112D and TOV21G cell lines. Each value is a mean of 3 values from a representative experiment. The statistical analysis was performed by One-way ANOVA test. p < 0.05 were considered significant.

Figure 4

Efficiency of transient PSMC6 knockdown at mRNA and protein levels in IGROV-1 and IGROV-1/Pt1 cells, and colony forming ability of IGROV-1/Pt1 cells in soft agar and plastic dishes. a) and b) Total RNA was extracted from IGROV-1/Pt1 and IGROV-1 cells and quantified for PSMC6 using qRT-PCR at the experimental times of 48 h, 72 h and 144 h after transfection start using 100 nM of siRNA A and B. ΔCt values were obtained by normalizing the Ct values of the target gene to the endogenous control GAPDH. RQ values are referred to a calibrator, the untransfected cells, whose RQ value was set to 1 at each time point. c) and d). The expression levels of PSMC6 upon PSMC6 silencing were evaluated using Western blot analysis in IGROV-1/Pt1 and IGROV-1 cells harvested 48 h, 72 h and 144 h after transfection start. The band intensity was quantified for PSMC6 by using the ImageQuant software. Western Blot quantification reported in Table S3. e) IGROV-1/Pt1 cells were harvested 48 h after siRNA transfection start and they were counted and seeded to evaluate their clonogenic ability in soft agar. The reported values represent the mean ± SD of 6 fields in triplicate from a representative experiment. Kruskal-Wallis test followed by Bonferroni's test (p<0.0001 for Neg siRNA vs siRNA A; p<0.0001 for Neg siRNA vs siRNA B). f) Representative images from a colony formation assay carried out in PSMC6-silenced IGROV-1/Pt1 cells. These images were taken 10 days after seeding in soft agar. g) Colony forming ability was evaluated in plastic dishes. Cells were harvested 48 h after siRNA transfection start and their colony forming ability was evaluated. The reported values represent the mean ± SD of 3 wells from a representative experiment. Kruskal-Wallis test (p=0.05 for Neg siRNA vs siRNA A; p=0.05 for Neg siRNA vs siRNA B). Untr, untransfected; Neg, negative.

Figure 5

IGROV-1/Pt1 silencing and its effect on sensitivity to cisplatin by colony forming and spheroid assays. a) Percentage of survival of IGROV-1/Pt1 cells forming colonies in plastics after treatment with cisplatin at different concentrations. The IC₅₀ values for each experimental condition is reported. b) Relative diameter of IGROV-1/Pt1 spheroids for the indicated experimental conditions after 72 h treatment with 100 µM cisplatin. The statistical analysis was performed by One-way ANOVA test. p < 0.05 were considered significant. c) Representative images of IGROV-1/Pt1 untreated and treated spheroids after 72 h exposure to 100µM cisplatin.

Figure 6

Western Blot analysis of ERK1/2, AKT and their activation and quantitative analysis of apoptosis in PSMC6-silenced IGROV-1/Pt1 and IGROV-1 cells. IGROV-1/Pt1 (a,b) and IGROV-1 (c,d) cells were harvested 48 h, 72 h and 144 h after transfection start for protein analysis. Cell lysates were analyzed by Western blotting. The band intensity was quantified for P-ERK1/2, ERK1/2, P-AKT and AKT by using the software ImageQuant. Western Blot quantification reported in Table S4. e), f) For apoptosis, cells were silenced and 72 h after transfection of siRNAs, they were harvested for the Annexin V-binding assay. The shown values are the mean ± SD from 3 replicates of a representative experiment. Kruskal-Wallis test was applied to compare all samples (IGROV-1/Pt1: p=0.05 for Neg siRNA vs siRNA A; p=0.05 for Neg siRNA vs siRNA B; IGROV-1: p=0.1 for Neg siRNA vs siRNA A; p=0.05 for Neg siRNA vs siRNA B).

Figure 7

Western Blot analysis of DUSP5, DUSP6, ubiquitinated protein levels and PSMC5 proteasome subunits in PSMC6-silenced IGROV-1/Pt1 and IGROV-1 cells. IGROV-1/Pt1 and IGROV-1 cells were harvested 48 h and 72 h after transfection start for protein analysis. a) Cell lysates were analyzed by Western blotting for DUSP5 and DUSP6. b) IGROV-1/Pt1 and IGROV-1 cells were harvested 48 h and 72 h after transfection start for ubiquitinated protein accumulation analysis. Cell lysates were analyzed by Western blotting. β-tubulin represents control for loading. c) Western blot analysis of PSMC5 in IGROV-1/Pt1 and IGROV-1 cells. The band intensity was quantified by using the ImageQuant software. Western Blot quantifications are reported in Table S5.

Figure 8

Model for the effects of targeting of PSMC6 in ovarian carcinoma. The silencing of PSMC6 is associated with marked down regulation of the activation of the canonical MAPK pathway and an increase of the specific ERK1/2 DUSP6 phosphatase. PSMC6 knockdown appears to lead to impaired protein trasfer to proeasome and accumulation of ubiquitinated proteins. The figure also reports the two proteasome inhibitors bAP15 and bortezomib. Created with BioRender.com.