EZH2 inhibition enhances the activity of Carboplatin in aggressive-variant prostate cancer cell lines

Abstract

Background: Aggressive Variant Prostate Cancers (AVPCs) are incurable malignancies. Platinum-based chemotherapies are used for the palliative treatment of AVPC. The Polycomb Repressive Complex 2 (PRC2) promotes prostate cancer progression via histone H3 Lysine 27 tri-methylation (H3K27me3). EZH2 encodes the catalytic subunit of PRC2. A recently developed nucleosome capture technology (Nu.Q^Ⓡ).measures H3K27me3 levels in biological fluids. EZH2 inhibitors (EZH2i) are being tested in clinical trials. We hypothesize that epigenetic reprogramming via EZH2i improves the efficacy of Carboplatin in AVPC and that EZH2i activity can be measured via both cellular- and cell-free nucleosomal H3K27me3 (cf-H3K27me3) levels.

Methods: We studied the expression of PRC2 genes in clinical prostate cancer cohorts (bioinformatics). We determined the effect of EZH2i on cellular- and cf-H3K27me3 levels. We measured dose-dependent effects of Carboplatin with/without EZH2i on AVPC cell viability (IC₅₀). We used RNA-Seq to study how EZH2i modulates gene expression in AVPC cells.

Results: PRC2 genes were significantly up-regulated in AVPC vs other prostate cancer types. EZH2i reduced both cellular and cf-H3K27me3 levels. EZH2i significantly reduced Carboplatin IC₅₀. EZH2i reduced the expression of DNA repair genes and increased the expression of p53-dependent pro-apoptotic factors.

Conclusions: EZH2i plus Carboplatin is a promising combination treatment for AVPC.

Keywords: Carboplatin; EZH2; Tazemetostat; aggressive variant prostate cancer; epigenetics; neuroendocrine prostate cancer; personalized epigenetic therapies.

Figure 1.

(a) Expression of EED, EZH2, and SUZ12 mRNA levels in n = 264 PCa samples (CbioPortal, accessed on 10/01/2023) with or without NEPC features. **** = p < 0.0001, two-tailed Student t-test Correlation between gene expression and AR activity score for genes (b) EED, (c) EZH2 and (d) SUZ12. Analises for a-d were conducted using the Metastatic Castration Resistant Prostate Cancer dataset on Cbioportal (e). Pathway analysis of genes significantly co-expressed with EZH2 in the Cbio-portal Neuroendocrine Prostate Cancer-Multi Institute dataset (cases were defined based on cell morphology and expression of NEPC makers). Genes with Pearson correlation coefficient > 0.85 (EZH2) were downloaded in Metascape for pathway analysis (https://metascape.Org/gp/index.Html#/main/step1).

Figure 2.

H3K27me3 levels in DU-145 (a and c) and OPT7714 cells (b and d) exposed to different EZH2 inhibitors (upper panel, representative image; lower panel statistical analysis of three quantified blots). Cells were treated with DMSO (vehicle), Tazemetostat 1, 5, and 10 µM, GSK-126 1, 5, and 10 µM, and CPI-1205 1, 5, and 10 µM for 72 hours; cellular protein extracts were analyzed by immunoblotting using anti-H3K27me3 and anti-GAPDH antibodies. Nu.Q Volition-measured H3K27me3 levels in the supernatant of DU-145 (c) and OPT7714 (d) cells in the same conditions described above. ***p < 0.001 and ****p < 0.0001 one-way ANOVA with Dunnett post-hoc test (c and d).

Figure 3.

Growth inhibitory effects of Tazemetostat, GSK-126, and CPI-1205 on DU-145 (a) and OPT7714 (b) cells. The cells were counted after exposure to different concentrations of EZH2 inhibitor (0.01, 1, 10, and 25 µm), or DMSO for 10 days. Cell viability of DU-145 (c) and OPT7714 (d) treated with combination of GSK-126 and Carboplatin vs Carboplatin alone. Cells were exposed to 10 µm GSK-126 for 72 hours, followed by GSK-126 (10 µm) and Carboplatin (different concentrations as indicated by the X axis) for the following 72 hours. Cells were counted at the end of the experiment. IC₅₀ values of DU-145 (e) and OPT7714 (f) treated with a combination of GSK-126 and Carboplatin vs Carboplatin alone. ***p<.05, ****p<.01, two-tailed Student’s t-test.

Figure 4.

Cell viability in Kucap-13 NEPC spheroids exposed to GSK-126 and platinum agents. (a,b) Cells were treated at fixed doses of GSK-126 and Carboplatin as described in the Methods. At the end of the experiments, cell viability was measured with the Luna system (a) or with the Caspase-Glo 3/7 assay (Promega-b) Statistical analysis: **p < 0.01; ***p < 0.001 (One-way ANOVA with Dunnett’s multiple comparisons test). C) Cell viability of Kucap-13 cells treated with combination of GSK-126 and Cisplatin vs Cisplatin alone, as indicated in Methods. IC₅₀ (Cisplatin plus GSK-126) = 5.78 µM; IC₅₀ (Cisplatin) = 31.99 µm.

Figure 5.

Transcriptome analysis of DU-145 cells exposed to EZH2 inhibitor. (a) Volcano plot of RNA Seq from cells exposed to GSK-126 vs control. Differentially expressed genes are colored in red if they are above the fold-change and -Log (p.Adjust-value) threshold; genes colored in green are only above the -Log (p.Adjust value) threshold. The names of some highly up- and down-regulated genes are shown in the plot. (b) a comparison of normalized enrichment scores (NES) that are significantly different in cells exposed to GSK-126 or control. A positive NES indicates up-regulation of the pathway in cells treated with GSK-126. A negative NES indicates down-regulation of the pathway in cells treated with GSK-126.