Elevated NCOR1 disrupts a network of dietary-sensing nuclear receptors in bladder cancer cells

Abstract

Increasingly invasive bladder cancer cells lines displayed insensitivity toward a panel of dietary-derived ligands for members of the nuclear receptor superfamily. Insensitivity was defined through altered gene regulatory actions and cell proliferation and reflected both reduced receptor expression and elevated nuclear receptor corepressor 1 (NCOR1) expression. Stable overexpression of NCOR1 in sensitive cells (RT4) resulted in a panel of clones that recapitulated the resistant phenotype in terms of gene regulatory actions and proliferative responses toward ligand. Similarly, silencing RNA approaches to NCOR1 in resistant cells (EJ28) enhanced ligand gene regulatory and proliferation responses, including those mediated by peroxisome proliferator-activated receptor (PPAR) gamma and vitamin D receptor (VDR) receptors. Elevated NCOR1 levels generate an epigenetic lesion to target in resistant cells using the histone deacetylase inhibitor vorinostat, in combination with nuclear receptor ligands. Such treatments revealed strong-additive interactions toward the PPARgamma, VDR and Farnesoid X-activated receptors. Genome-wide microarray and microfluidic quantitative real-time, reverse transcription-polymerase chain reaction approaches, following the targeting of NCOR1 activity and expression, revealed the selective capacity of this corepressor to govern common transcriptional events of underlying networks. Combined these findings suggest that NCOR1 is a selective regulator of nuclear receptors, notably PPARgamma and VDR, and contributes to their loss of sensitivity. Combinations of epigenetic therapies that target NCOR1 may prove effective, even when receptor expression is reduced.

Fig. 1.

Expression of nuclear receptor and corepressors in bladder cancer cell lines. Panel (A): nuclear receptor mRNA levels measured by Q-RT–PCR in RT112, HT-1376 and EJ28 compared with the levels in RT4 cells, which were arbitrarily set to 1. Total mRNA was isolated from triplicate cultures in mid-exponential phase, reverse transcribed and the target genes amplified according to the Materials and Methods. Each data point represents the mean of three separate experiments amplified in triplicate wells ±SEM (*P < 0.01). Panel (B): cytoplasmic (C) and nuclear (N) proteins were isolated from mid-exponential cultures from RT4, RT112, HT-1376 and EJ28 cells and resolved by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and probed with antibody to VDR and PPARγ. Representative blots are shown with the position of the proteins indicated on the left. Blots were subsequently stripped and reprobed for nucleolin. Panel (C): corepressor mRNA was measured as in Panel (A). Panel (D): proteins were isolated from mid-exponential cultures from RT4, RT112, HT-1376 and EJ28 cells and resolved by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and probed with antibody to NCOR1 and SLIRP. Representative blots are shown with the position of the proteins indicated on the left. Blots were subsequently stripped and reprobed for either α-tubulin or nucleolin.

Fig. 2.

Individual sigmoidal proliferation response curves toward a panel of nuclear receptor ligands and vorinostat. Panels (A and B): RT4 and EJ28 bladder cancer cells were plated into 96-well plates and treated with DHA (RXR), eicosapentaenoic acid (PPARα), ETYA (PPARγ), 22-HC (LXRα,β), GW3965 (LXRα,β), CDCA (FXR) LCA (VDR/FXR) and the HDAC inhibitor vorinostat (Vorin.) over a range of concentrations. After 96 h, with a redose after 48 h, proliferation was measured according to Materials and Methods and expressed as a percentage of untreated controls. Each data point represents the mean value of three separate experiments each undertaken in triplicate wells.

Fig. 3.

Modulation of NCOR1 expression in bladder cancer cells. Panel (A): the expression of NCOR1 was measured in stably selected RT4 clones, according to the Materials and Methods, in nuclear and cytoplasmic fractions in RT4 pNCOR #7 cells compared with RT4 pcDNA #6. Representative blots are shown with the position of the proteins indicated on the left. Blots were subsequently stripped and reprobed for nucleolin. Panel (B): EJ28 cells were either mock transfected or treated with scrambled RNA (ScRNA) and siRNA toward NCOR1. Representative blots are shown with the position of the proteins indicated on the left. Blots were subsequently stripped and reprobed for nucleolin.

Fig. 4.

Modulation of responsiveness to a panel of nuclear receptor ligands by targeting NCOR1 expression and activity. Panel (A): antiproliferative responses of RT4 pNCOR #7 were compared with RT4 pcDNA #6 clones with the indicated treatments. After 96 h, with a redose after 48 h, proliferation was measured according to Materials and Methods and expressed as a percentage of untreated controls. Each data point represents the mean value of three separate experiments each undertaken in triplicate wells, and significant differences are indicated (**P < 0.01, ***P < 0.001). Panel (B): antiproliferative responses of EJ28 cells in either mock transfected, scrambled RNA (ScRNA) or siRNA with the indicated treatments. After transfection and 48 h exposure to treatment proliferation was measured according to Materials and Methods and expressed as a percentage of untreated controls. Each data point represents the mean value of three separate experiments each undertaken in triplicate wells, and significant differences are indicated (**P < 0.01, ***P < 0.001). RT112 cells (Panel C) and EJ28 cells (Panel D) were treated with minimally active doses of the indicated nuclear receptor ligands, the HDAC inhibitor vorinostat and the combination. The doses were calculated from the individual agent response curves to be the ED₂₅ values as follows: RT112—1α,25(OH)₂D₃ (10 nM), 9 cis retinoic acid (9cRA) (100 nM), ETYA (9 μM) CDCA (90 μM) LCA (20 μM), vorinostat (0.3 μM); EJ28—1α,25(OH)₂D₃ (100 nM), 9 cis retinoic acid (100 nM), ETYA (30 μM) CDCA (100 μM) LCA (100 μM), vorinostat (0.8 μM). Proliferation was measured after 96 h, with re-dosing after 48 h. Strong-additive interactions are indicated where the observed combined inhibition is significantly greater than the predicted combined value (*P < 0.05, **P < 0.01, ***P < 0.001). All experiments were carried out in three independent times with triplicate wells.

Fig. 5.

Regulation by LCA of target genes on an HNF4α-like network in EJ28 cells. Cells were mock transfected or treated with scrambled RNA (ScRNA) and siRNA toward NCOR1 and then exposed to LCA (100 μM, 4 h). Total mRNA was isolated from triplicate cultures in mid-exponential phase, reverse transcribed and the target genes amplified according to the Materials and Methods. Each data point represents the mean of three separate experiments amplified in triplicate wells ±SEM *P < 0.05, ***P < 0.001.