Regulation of p53 Stability and Apoptosis by a ROR Agonist

Activation of p53 function leading to cell-cycle arrest and/or apoptosis is a promising strategy for development of anti-cancer therapeutic agents. Here, we describe a novel mechanism for stabilization of p53 protein expression via activation of the orphan nuclear receptor, RORα. We demonstrate that treatment of cancer cells with a newly described synthetic ROR agonist, SR1078, leads to p53 stabilization and induction of apoptosis. These data suggest that synthetic ROR agonists may hold utility in the treatment of cancer.


Introduction
In approximately 50% of human cancers the p53 gene is mutated, but in the remaining half of cancers activation of p53 function is considered to be a valuable strategy for development of anti-cancer therapeutics. p53 plays a critical role in limiting cell proliferation and inducing apoptosis in response to cellular stress/ damage and abnormal function of p53 is associated with cancers [1]. p53 function is tightly regulated by modulation of protein stability. Under most conditions, p53 protein is undetectable primarily due to interaction of p53 with the E3 ubiquitin ligase MDM2 and succeeding proteosomal degradation. A number of compounds that inhibit the MDM2-p53 interaction or the subsequent steps toward proteosomal degradation are under evaluation for their anti-cancer activity.
Epidemiological data indicates that disruption of circadian rhythmicity is associated with development of cancer [2,3,4,5]. Based on these data, the World Health Organization has classified shift-work associated with a disrupted circadian rhythm as a probable carcinogen [6]. Disruption of the circadian rhythm in rodents leads to increased tumor progresssion [7,8,9,10] and disturbances in the expression of critical clock genes has been noted in several breast and liver cancer cell lines [11,12,13,14].
The retinoic acid receptor-related orphan receptor a (RORa) is a member of the nuclear receptor superfamily that plays a critical role in regulation of the circadian clock. RORa expression oscillates in a circadian manner and plays an important role in modulation of expression of core clock components such as BMAL1, CLOCK and NPAS2 [15,16,17,18,19]. RORa expression is induced in response to a variety of cellular stresses [20,21] and is downregulated in several breast, prostate, and ovarian cancer cell lines [21]. Additionally, RORa is expressed at very low levels in many cancers [21] suggesting that low RORa expression may be one mechanism underlying tumorigenesis. Based on these reports we focused on identification of pathways where RORa may regulate cell proliferation.

Results
A chromatin immunoprecipitation (ChIP) -microarray screen was performed in the hepatocellular carcinoma cell line, HepG2, to identify RORa occupancy sites within the genome as we previously described [19]. We discovered RORa occupancy within the proximal promoter of the SOX4 gene (Fig. 1A), which we found particularly intriguing because of its role in the regulation of p53 stability and function [22]. The tumor suppressor p53 plays a critical role in limiting cell proliferation and inducing apoptosis in response to cellular stress/damage and abnormal function of p53 is associated with cancers [1]. SOX4 directly interacts with p53 limiting its ability to be ubiquitinated by MDM2 and thus increases its stability [22]. In fact, induction of SOX4 expression is required for p53 stabilization in response to DNA damage [22]. Bioinformatic analysis of the RORa occupancy site revealed a putative ROR response element that was conserved between humans, mice and xenopus (Fig. 1A). We confirmed occupancy of the SOX4 promoter by RORa using a ChIP assay as shown in Fig. 1B. The SOX4 promoter conveyed RORa-dependent regulation of a luciferase reporter gene in HEK293 cells as illustrated in Fig. 1C. This regulation was dependent on the RORE identified and shown in Fig. 1A since mutation of the RORE sequence rendered the construct unresponsive to RORa (Fig. 1D). Adenoviral overexpression of RORa in HepG2 cells resulted in an increase in SOX4 mRNA expression whereas knock-down of RORa expression reduced SOX4 mRNA expression in the same cell line (Fig. 1D). The limited effect on SOX4 mRNA after RORa knock-down may be due to compensatory actions of RORc, which is known to act in concert with RORa in HepG2 cells [23]. Based on previous observations that altering SOX4 expression modulates p53 stability, we hypothesized that RORa expression may correlate with p53 protein stability [22]. Indeed, we observed that overexpression of RORa in HepG2 cells was associated with an increase in SOX4 expression leading to increased p53 protein levels ( Fig. 1D). Consistent with this observation, decreasing RORa expression in these cells leads to decreased p53 protein levels (Fig. 1E). We directly measured the stability of p53 under conditions where RORa was overexpressed by treating cells with cychoheximide and noted that overexpression of RORa clearly stabilized p53 expression (Fig. 1F). Additionally, overexpression of RORa led to increased expression of p53 target genes that play a key role in cell cycle arrest (p21) and apoptosis (PUMA) ( Fig. 2A) [1]. Knock-down of p53 suppressed the ability of RORa overexpression to increase the expression of these genes ( Fig. 2A lower panels). Further, analysis of HepG2 cells overexpressing RORa revealed that the number of cells in sub-G1 increased substantially over control cells (34% vs 8%) while cells in S and G2/M phase were also substantially reduced consistent with induction of apoptosis (Figs. 2B & 2C). These data are consistent with the observed increase in p53 stability and increase in p21 and PUMA expression noted in Fig. 2B. Furthermore, the increase in sub-G1 cells induced by overexpression of RORa was blocked when p53 expression was knocked down demonstrating that RORa induction of apoptosis is p53-dependent (Fig. 2D). MCF-7 breast cancer cells show similar results when RORa is overexpressed; a significant increase in sub-G1 cells relative to control cells (15.5% vs 4.5%) (Fig. 2E).
Based on or results where overexpression of RORa leads to increased p53 protein stability, we examined the potential of a RORa agonist we recently identified to increase p53 stability. We recently characterized several synthetic ROR ligands including the first synthetic, selective ROR ligand, SR1078 (Fig. 3A) [24,25,26]. SR1078 functions as an agonist by activating RORa leading to an increase in transcription of RORa target genes [24]. Consistent with this activity, we noted that SR1078 induced the expression of SOX4 mRNA in HepG2 cells as well as a well-characterized RORa target gene, REV-ERBa (Fig. 3B). We observe that SR1078 treatment also results in an increase in p53 protein levels ( Fig. 3C) similar to the results we observed with overexpression of RORa. SR1078 treatment also led to a significant increase in the expression of p53 target genes p21 and PUMA (Fig. 3D). Knockdown of p53 suppressed the ability of SR1078 to increase the expression of these genes ( Fig. 3D lower panels). Consistent with the increase in p53 protein levels as well as the increase in the expression of p53 target genes, we found that SR1078 treatment led to increased apoptosis as indicated by the increase in HepG2 cells in sub-G 1 (0.9% control vs. 9.4% SR1078) (Fig. 3E). The increase in apoptosis induced by SR1078 was both RORa-and p53-dependent since siRNA-mediated knock down of either of these genes suppressed the ability of SR1078 to increase cells in sub-G1 (Figs. 3F &3G).

Discussion
The tumor suppressor protein p53 plays an essential role in regulation of key cellular processes including DNA repair, cell cycle, and apoptosis. In approximately half of all human cancers the p53 gene is deleted or mutated [1]. In many cancers with wild type p53, the activity of the tumor suppressor is inhibited by various effectors. One clear example of this is found in tumors where MDM2 is overexpressed due to an amplification of a chromosome segment that includes MDM2 [27]. This leads to abnormal degradation of p53 and thus a similar phenotype to tumors with a mutant or deleted p53 gene. Inhibition of abnormal degradation of p53 is a logical pharmacological target and, in fact, several small molecule inhibitors of the MDM2-p53 interaction including nutlin-3, RITA, spirooxindoles and quilinols are being investigated as anti-cancer agents due to their abilities to increase cellular p53 protein levels through inhibition of MDM2-directed proteosomal degradation of p53 [28].
Our data suggests that a small molecule synthetic RORa agonist can increase p53 protein stability leading to increased p53 function and subsequent apoptosis. MDM2 inhibitors have been the focus of significant efforts to develop anti-cancer agents that function via activation of p53 activity. Here, we demonstrate that RORa agonists may also be useful for activation of p53 activity and thus represent a novel target for development of anti-cancer therapeutics. Most nuclear receptors that have identified ligands are well-characterized targets for drugs used in the clinic and the nature of the nuclear receptor ligand binding domain typically allows for optimization of small molecule ligands for drug development. Thus, RORa clearly represents a unique target for stabilization of p53 that is quite distinct from the challenging effort to inhibit a protein-protein interaction such as the MDM2-p53 interaction. While this manuscript was under revision, Kim et al. also described a role for p53 in regulation of p53 stability and function [29]. Although they show also that RORa regulates p53 stability they demonstrate a distinct mechanism from that proposed in this manuscript for increasing p53 stability involving the enhancement of p53-HAUSP interaction [29]. Kim et al indicate that they could not rule out additional non-HAUSP mechanism for regulation of p53 stability by RORa [29], which is consistent with our observation that RORa directly regulates the expression of SOX4, a critical gene involved in MDM2-dependent regulation of p53 stability.

Plasmids and viruses
The SOX4 promoter (21121 to +90) was amplified from genomic DNA of HepG2 cells (ATCC, Manassas, VA) and cloned into pTAL-Luc luciferase report vector (Clontech, CA) to make the pTAL-SOX4 reporter construct. PGL4.73 reporter was from Promega (Madison, WI). pTrex-RORa and pTrex-RORc were from Phenex Pharmaceuticals AG. RORa was tagged with FLAG and subcloned into pAd/CMV/V5-DEST vector through Gateway TM technique (Invitrogen). The adenovirus with FLAG-RORa was produced according to the manufacturer's instructions.

Site-directed Mutagenesis
The SOX4 promoter mutant constructed by site directed mutagenesis as previously described [25,30]. The RORE (2178 to 2165) was mutated from GGAATGAGGTCAG to GGAAT-GAGGGGGG. The mutant primers targeting ROR binding site are: GCTCTGTAAATTGGAATGAGGGGGATTTGGAGC-TTCTC (forward) and GAGAAGCTCCAAATCCCCCTCA-TTCCAATTTACAGAGC (reverse). The mutant primers were used to amplify mutant plasmid from pTAL-SOX4 reporter using PfuUltra HF DNA polymerase. The PCR product were treated with Dpn I to select for mutation-containing synthesized DNA and then transformed into XL1-Blue supercompetent cells. Positive clones were picked up and grew overnight in LB media. The plasmid were isolated using QIAprep Spin Miniprep Kit (Qiagen). The mutant construct was verified by sequencing.
Cell culture and luciferase assay HEK293 cells (ATCC) were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, CA) at 37uC under 5% CO 2 . 24 h prior to transfection, HEK293 cells were plated in 96-well plates at a density of 15610 3 cells/well. Transfections were performed using Lipofectamine TM 2000 (Invitrogen). Each well was transfected with 20 ng pGL4.73, 50 ng ROR and 100 ng pTAL-Luc or mutant. Eight hours post-transfection, the cells were treated with vehicle or ligands. Twenty-four hours post-treatment, the luciferase activity was measured using the Dual-GloTM luciferase assay system (Promega). The value from experimental reporter was normalized to control reporter. The values indicated represent the means 6 S.E. from four independently transfected wells. The experiments were repeated at least three times.

Overexpression of RORa and siRNA Knockdown
The HepG2 cells (ATCC) were maintained in MEM supplemented with 10% fetal bovine serum at 37uC under 5% CO2. HepG2 cells were plated in 6-well plate one day before infection. The cells were infected with adenovirus for 24 hours and then switched to regular growth media. Twenty-four hours later, the cells were harvested to isolate total RNA. For knockdown assay, the control siRNA, human RORa siRNA, and human p53 siRNA (Thermo Scientific) were transfected with LipofectamineTM RNAiMAX (Invitrogen) by using reverse transfection. After 24 hours, cells were harvested to perform quantitative PCR assay or western blot.

Western Analysis
HepG2 cells were washed once with phosphate-buffered saline and then incubated for 10 min at 4uC in 100 ml of TNT lysis buffer (50 mM Tris-Cl, pH 7.5, 150 mM NaCl, and 1% Triton X-100) and a complete miniprotease inhibitor mixture (Roche Applied Science). Samples were then scraped and harvested into 1.5-ml microcentrifuge tubes, vortexed for 30 s, and then centrifuged (4256 g for 10 min). Protein levels in the supernatants were determined using a Coomassie protein assay kit (Bio-Rad), and 20 mg of protein from each sample was separated by SDSPAGE (BioRad -10%) and then transferred to a polyvinylidene difluoride membrane (Millipore, Milford, MA) and immunoblotted with primary antibodies: RORa (BioLegend), TP53 (Cell Signaling) or a-tubulin (Sigma) and horseradish peroxidase-conjugated secondary antibodies (Jackson Immunoresearch). Detection of the bound antibody by enhanced chemiluminescence was performed according to the manufacturer's instructions (Santa Cruz). For p53 half-life experiments, 10 mM cyclohexamide was added to the cells for 30, 60, or 90 minutes prior to harvesting the HepG2 cells for western analysis.

FACS analysis
HepG2 cells and MCF7 cells (ATCC) were plated in 24-well plates the day before infection. The cells were infected with adenovirus 24 hours later. On the day of analysis, cells were harvested, washed, and fixed with 70% ethanolat 220uC.

ChIP/chip screening
HepG2 cells were infected with adenovirus for 24 hours and then switched to regular growth media for another 24 h. The cells were harvested and sent to Genpathway for ChIP/chip assay as previously described [31,32,33]. The RORa ChIP/chip experiment has been previously described [19].

Statistical Analysis
The Student's t test was used to test for significant differences between groups.  HepG2 cells with SR1078 leads to an increase in SOX4 and REV-ERBa mRNA expression. C) Treatment of HepG2 cells with SR1078 leads to increased p53 protein levels. + indicates 1 mM and ++ indicates 5 mM SR1078. D) Treatment of HepG2 cells with SR1078 leads to increased expression of p53 target genes, p21 and PUMA. Experiments shown in the lower panels were performed identical to the upper panels with the exception of inclusion of siRNA treatments as indicated. E) Cell cycle analysis of control HepG2 cells treated with vehicle control or SR1078. Note the substantial increase in cells in sub-G 1 following SR1078 treatment, 0.9% vs 9.4% indicated on the graph. F) Cell cycle analysis of HepG2 cells treated with either control siRNA or RORa siRNA in the presence of vehicle control or SR1078. G) Cell cycle analysis of HepG2 cells treated with either control siRNA or p53 siRNA in the presence of vehicle control or SR1078. *, indicates p,0.05. doi:10.1371/journal.pone.0034921.g003