MED12 and CDK8/19 Modulate Androgen Receptor Activity and Enzalutamide Response in Prostate Cancer

Abstract

Prostate cancer progression is driven by androgen receptor (AR) activity, which is a target for therapeutic approaches. Enzalutamide is an AR inhibitor that prolongs the survival of patients with advanced prostate cancer. However, resistance mechanisms arise and impair its efficacy. One of these mechanisms is the expression of AR-V7, a constitutively active AR splice variant. The Mediator complex is a multisubunit protein that modulates gene expression on a genome-wide scale. MED12 and cyclin-dependent kinase (CDK)8, or its paralog CDK19, are components of the kinase module that regulates the proliferation of prostate cancer cells. In this study, we investigated how MED12 and CDK8/19 influence cancer-driven processes in prostate cancer cell lines, focusing on AR activity and the enzalutamide response. We inhibited MED12 expression and CDK8/19 activity in LNCaP (AR+, enzalutamide-sensitive), 22Rv1 (AR-V7+, enzalutamide-resistant), and PC3 (AR-, enzalutamide-insensitive) cells. Both MED12 and CDK8/19 inhibition reduced cell proliferation in all cell lines, and MED12 inhibition reduced proliferation in the respective 3D spheroids. MED12 knockdown significantly inhibited c-Myc protein expression and signaling pathways. In 22Rv1 cells, it consistently inhibited the AR response, prostate-specific antigen (PSA) secretion, AR target genes, and AR-V7 expression. Combined with enzalutamide, MED12 inhibition additively decreased the AR activity in both LNCaP and 22Rv1 cells. CDK8/19 inhibition significantly decreased PSA secretion in LNCaP and 22Rv1 cells and, when combined with enzalutamide, additively reduced proliferation in 22Rv1 cells. Our study revealed that MED12 and CDK8/19 regulate AR activity and that their inhibition may modulate response to enzalutamide in prostate cancer.

Keywords: Mediator complex; androgen receptor; enzalutamide; prostate cancer.

Figure 1.

MED12 knockdown inhibits prostate cancer cell proliferation. (A) MED12 gene copy number in cohorts of primary prostate cancer (TCGA-PRAD) and castration-resistant prostate cancer (SU2C-PRAD) tissues. (B) MED12 protein quantification in a benign prostatic hyperplasia cell line (BPH1) and prostate cancer cell lines (LNCaP, 22Rv1, VCaP, DuCaP, Du145, and PC3). A representative Western blot is shown for each cell line. (C) RNA interference and CRISPR screens evaluating prostate cancer cell dependency on MED12 expression for their growth. Lower scores correspond to a higher dependency: a score of 0 is equivalent to nonessential genes and a score of −1 is equivalent to the median of essential genes. (D) MED12 protein quantification in LNCaP, 22Rv1, and PC3 cells transfected with siCtrl and siMED12. A representative Western blot is shown for each cell line. (E) Fluorescent-based cell number quantification of LNCaP, 22Rv1, and PC3 cells transfected with siCtrl and siMED12. (F) Fluorescent-based quantification of the total cell number in LNCaP, 22Rv1, and PC3 spheroids formed after pretransfection of siMED12 and siCtrl in 2D cells (total transfection time: 7 days). Data are represented as mean ± SD. *P < .05, **P < .01, ***P < .001.

Figure 2.

MED12 knockdown downregulates c-Myc signaling and protein expression. (A) Top MSigDB hallmarks related to cell proliferation and significantly altered in LNCaP, 22Rv1, and PC3 cells after MED12 knockdown (false discovery rate < 0.05; transfection time: 3 days). (B) Barcode plot of Myc targets V1 and V2 in LNCaP, 22Rv1, and PC3 cells after MED12 knockdown. (C) c-Myc protein expression in control and siMED12-transfected LNCaP, 22Rv1, and PC3 cells (transfection time: 3 days). A representative Western blot is shown for each cell line. Data are presented as mean ± SD. Statistical significance: *P < .05, **P < .01, ***P < .001. (D) MED12 and c-MYC gene expression correlation in cohorts of primary prostate cancer (TCGA_PRAD, GSE62872, GSE21034, GSE46691) and castration-resistant prostate cancer tissues (SUC2_PRAD).

Figure 3.

MED12 knockdown decreases androgen receptor activity. (A) Barcode plot and batch-corrected heatmap of androgen response in 22Rv1 cells after MED12 knockdown (transfection time: 3 days). Statistical significance: false discovery rate < 0.05. (B) Quantitative reverse transcription polymerase chain reaction quantification of KLK3, FKBP5, and TMPRSS2 mRNA expression in LNCaP and 22Rv1 3D models (transfection time: 7 days). All data are presented as mean ± SD. (C) Quantification of prostate-specific antigen protein secreted by 22Rv1 in cell medium following MED12 knockdown (chemiluminescent-based assay). Statistical significance: *P < .05, **P < .01, ***P < .001.

Figure 4.

MED12 knockdown inhibits AR-V7 protein expression and increases enzalutamide efficacy. (A) AR-FL and AR-V7 protein expression in LNCaP and 22Rv1 cells upon MED12 knockdown [transfection time in 2D models: 6 days (22Rv1), 3 days (LNCaP); transfection time in 3D models: 7 days]. A representative Western blot is shown for each cell line. (B) Effect of combined siMED12 and 10 µM enzalutamide treatment on prostate-specific antigen secretion medium (chemiluminescent-based assay), KLK3 mRNA (quantitative reverse transcription polymerase chain reaction), and cell proliferation (fluorescent-based nuclei counting) in 22Rv1 and LNCaP cells. All data are presented as mean ± SD. Statistical significance: *P < .05, **P < .01, ***P < .001.

Figure 5.

CDK8/19 inhibition and enzalutamide additively decrease cell proliferation in 22Rv1. (A) MED12 and CDK8 gene expression correlation in cohorts of primary (TCGA_PRAD, GSE62872, GSE21034, GSE46691) and castration-resistant prostate cancer tissues (SUC2_PRAD). (B) Relapse-free survival of treatment-naïve prostate cancer patients (TCGA database) with high (media = above) and low (media = below) CDK8 expression. (C) Dose-response curves of SEL-120-34A (CDK8/19 inhibitor) in LNCaP, 22Rv1, and PC3 (fluorescent-based assay, treatment time: 3 days). The IC50 for cell proliferation was reported. (D) MSigDB cancer hallmarks significantly altered in LNCaP, 22Rv1, and PC3 cells upon CDK8/19 inhibition (treatment time: 3 days). Statistical significance: false discovery rate < 0.05. (E) Quantification of prostate-specific antigen secreted by LNCaP and 22Rv1 treated for 3 days with CDK8/19 inhibitor (SEL-120-34A) or Ctrl (dimethylsulfoxide) (chemiluminescent-based assay). (F) Quantification of AR-FL and AR-V7 protein expression in 22Rv1 cells treated for 3 days with CDK8/19 inhibitor (SEL-120-34A) or Ctrl (dimethylsulfoxide). A representative Western blot is shown for each cell line. Data are presented as mean ±SD. (G) Cumulative population doubling level of cells treated with: Ctrl (dimethylsulfoxide); 1.5 µM CDK8/19; 10 µM enzalutamide; 1.5 µM CDK8/19 and 10 µM enzalutamide (number of replicates: 1). All data are presented as mean ±SD. Statistical significance: *P < .05, **P < .01, ***P < .001. Abbreviation: CDK, cyclin-dependent kinase.