Prostate cancer is a prevalent public health problem, especially because noncutaneous advanced malignant forms significantly affect the lifespan and quality of life of men worldwide. New therapeutic targets and approaches are urgently needed. The current study reports elevated expression of R1 (CDCA7L/RAM2/JPO2), a c-Myc–interacting protein and transcription factor, in human prostate cancer tissue specimens. In a clinical cohort, high R1 expression is associated with disease recurrence and decreased patient survival. Overexpression and knockdown of R1 in human prostate cancer cells indicate that R1 induces cell proliferation and colony formation. Moreover, silencing R1 dramatically reduces the growth of prostate tumor xenografts in mice. Mechanistically, R1 increases c-Myc protein stability by inhibiting ubiquitination and proteolysis through transcriptional suppression of HUWE1, a c-Myc–targeting E3 ligase, via direct interaction with a binding element in the promoter. Moreover, transcriptional repression is supported by a negative coexpression correlation between R1 and HUWE1 in a prostate cancer clinical dataset. Collectively, these findings, for the first time, characterize the contribution of R1 to prostate cancer pathogenesis.
These findings provide evidence that R1 is a novel regulator of prostate tumor growth by stabilizing c-Myc protein, meriting further investigation of its therapeutic and prognostic potential.
Prostate cancer is the second most frequently diagnosed cancer and fifth leading cause of cancer-related death in men worldwide with the highest incidence rates found in Western countries (1). Despite decreasing death rates for prostate cancer in recent years, attributed mainly to improved early detection and treatment, aggressive castration-resistant and metastatic prostate cancer are still incurable and lethal (2–4). A greater understanding of the mechanisms governing prostate cancer initiation and progression could help develop effective strategies for treating advanced prostate cancer.
The pathogenesis of prostate cancer is a multistep process involving a number of genetic alterations and epigenetic dysfunctions by which benign prostatic epithelial cells transition to high-grade prostatic intraepithelial neoplasia (PIN), invasive adenocarcinoma, and distant metastasis (3, 5). Activation of the proto-oncogene c-Myc, known to regulate the transcription of numerous genes and pathways, is a molecular event found constantly in different stages of disease progression (6, 7). In addition to the frequent overexpression of c-Myc in PIN, with a stepwise increase from normal to low-grade PIN to high-grade PIN, nuclear c-Myc protein overexpression was observed in both localized prostate cancer and metastatic disease (6). Enforced expression of c-Myc in mouse prostate further revealed the role of c-Myc in driving the onset of PIN, which progressed to invasive adenocarcinoma, albeit at different rates (8, 9). In addition, c-Myc may also contribute to disease progression to castration resistance by conferring androgen-independent prostate cancer cell growth (10). In spite of accumulated c-Myc studies in prostate cancer research over the past 30 years, the precise regulation of c-Myc still remains largely unclear, and merits continued exploration for a better understanding of prostate cancer pathogenesis and progression.
R1 (RAM2/CDCA7L/JPO2), a c-Myc–interacting protein essential for cellular transformation, was first identified in medulloblastoma cells (11). We demonstrated that R1 is a novel transcription repressor, which directly interacts with the promoters of monoamine oxidase genes by competing with Sp family transcription factors to suppress gene expression (12, 13). The role of R1 as a transcription factor was further supported by its chromatin-binding dynamics upon interaction with transcriptional coactivator LEDGF/p75 (14, 15). However, the role of R1 in prostate cancer and particularly the regulatory relationship between R1 and c-Myc has not been explored. In this study, we investigated for the first time the functional and mechanistic roles of R1 in prostate cancer using both cell line and tumor xenograft models, and also evaluated R1 expression in human prostate cancer samples and its correlation with disease prognosis and survival in clinical datasets.
Materials and Methods
Prostate cancer tissue microarrays including a total of 93 primary adenocarcinomas and 19 normal prostate tissues were purchased from US Biomax. Specimens were stained with antibodies specific for R1 (Sigma-Aldrich) following our published protocol (16). Each sample was scored on the basis of staining intensity (I) and the proportion of tumor cells stained by quantity (q) to obtain a final score defined as the product of I × q. The scoring system for I was: 0 = negative, 1 = low, 2 = moderate, and 3 = intense immunostaining. The scoring system for q was: 0 = negative, 1 = 1%–25% positive, 2 = 26%–50% positive, 3 = 51%–75% positive, and 4 = 76%–100% positive cells. All scoring was performed by a pathologist.
Cells and reagents
Human prostate cancer LNCaP, DU145, PC-3, and human embryonic kidney 293T cell lines were obtained from the ATCC. Human prostate cancer C4-2B (17), ARCaPE and ARCaPM (18, 19) cell lines, and two pairs of patient-derived human prostate normal epithelial (PNE) and prostate cancer epithelial (PCE) cell lines (20) were established as described previously. All cell lines were cultured in RPMI1640 medium, T-medium, or DMEM (Thermo Fisher Scientific) supplemented with 10% FBS (Atlanta Biologicals) and 1% penicillin/streptomycin (Thermo Fisher Scientific). Human prostate epithelial cells (PrEC) were purchased from Lonza and cultured following the manufacturer's instructions. Human R1 expression construct was generated by insertion of human R1 coding region at EcoRI-BglII sites in p3×FLAG-CMV vector (Sigma-Aldrich) containing a neomycin-resistant gene. Human c-Myc-pcDNA3 and HA-ubiquitin-pcDNA3 expression constructs were obtained from Addgene. The pM4-min-tk-luc, containing 4 c-Myc–binding E-box sites, and the parental promoterless pmin-tk-luc reporter constructs were kindly provided by Dr. Bernhard Luscher (Hannover Medical School, Hannover, Germany; ref. 21). Human HUWE1 promoter luciferase reporter construct was purchased from GeneCopoeia. pRL-TK plasmid-expressing Renilla luciferase was purchased from Promega. Human R1, c-Myc, and nontargeting control shRNA lentiviral particles that carry a puromycin- or neomycin-resistant gene were purchased from Sigma-Aldrich. Human R1 siRNAs were purchased from Santa Cruz Biotechnology. Cycloheximide and MG132 were purchased from Sigma-Aldrich.
Generation of stable overexpression and knockdown cells
R1 overexpression was achieved by transfection of FLAG-tagged human R1 expression construct into cells using Lipofectamine 2000 reagent (Thermo Fisher Scientific) following the manufacturer's instructions, followed by 3-week G418 selection (500 μg/mL) for stable clones. Stable shRNA–mediated R1 knockdown (KD) was achieved by infecting cells with lentiviral particles expressing a R1 shRNA TRCN0000364645 (shR1 #1, mainly used in this study and usually dubbed as “shR1”) or TRCN0000369291 (shR1 #2), followed by 2-week puromycin selection (2 μg/mL) for establishing stable cell lines. A pCMV empty vector and scrambled control shRNA lentiviral particles were used as controls in stable overexpression and KD cells, respectively. Overexpression and KD of gene(s) in stable cells were validated by immunoblotting analysis. Established stable cells were maintained in culture medium supplemented with either G418 or puromycin at the same doses used for selection.
Total RNA was isolated using the RNeasy Mini Kit (Qiagen) and reverse-transcribed to cDNA by M-MLV reverse transcriptase (Promega) as described previously (22). For immunoblots, cells were extracted with RIPA buffer in the presence of a protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific), and blots were performed as described previously (22, 23) using primary antibodies against R1 (Sigma-Aldrich), FLAG (M2, Sigma-Aldrich), c-Myc (D84C12, Cell Signaling Technology), HUWE1 (Bethyl Laboratories), MIZ-1 (R&D Systems), p53 (DO-1, NeoMarkers), or β-Actin (AC-15, Santa Cruz Biotechnology). Immunoblots were subjected to morphometric analysis by ImageJ software (NIH, Bethesda, MD) for scanned films or Image Lab software (Bio-Rad) for digital images. To use ImageJ to quantify protein band intensity, protein bands on scanned Western blot films were selected and defined in a rectangular box as region of interest. The mean histogram value of each selected box was measured, which after subtracting the background histogram value indicates the intensity of each band. The band intensity of target protein was normalized to that of loading control for comparisons between groups.
Cell proliferation assays
To determine the effect of R1 on cell proliferation, control, and R1-overexpressing/KD cells were seeded on 6-well plates (2 ×104 cells/well). To determine the effect of R1 on contact inhibition during cell proliferation, control and R1-overexpressing cells were seeded on 48-well plates with 3.5 × 105 cells per well, allowing confluence to be reached prior to cell counting. Cell numbers from triplicate wells were counted by a hemocytometer.
Colony formation assays
Cells were suspended in culture medium containing 0.3% agarose (FMC BioProducts) and placed on top of solidified 0.6% agarose in 6-well plates. The developed colonies were counted and recorded under a microscope after a 2-week incubation.
Migration and invasion assays
Assays were performed using 6.5-mm transwell inserts (8-μm pore size) coated with either collagen I or Growth Factor Reduced Matrigel (BD Biosciences) for migration and invasion assays, respectively. Cells were serum-starved overnight before seeding to eliminate the interference of any proliferative effect with cell migration or invasion. Cells were seeded inside transwell inserts containing culture medium without serum. After 16–48 hours, cells that translocated to the bottom surface of filters were fixed in 4% formaldehyde. The fixed membranes were stained using 1% crystal violet. Assays were quantified by counting the number of stained nuclei in five independent 200× fields in each transwell.
Tumor xenograft studies
Male 4-week-old athymic nude mice were purchased from Taconic, housed in the animal research facility at University of Southern California, and fed a normal chow diet. 1 × 106 PC-3 (shCon and shR1) cells were mixed with Matrigel (BD Biosciences) and injected subcutaneously into mice. Each mouse was injected on the right flank. Five mice were used for each group. Tumor size was measured every 3 days by caliper from the time of formation of palpable tumors and tumors were dissected and weighed after 5 weeks. Tumor volume was calculated by the formula of length × width2 × 0.52 (24). Tumors were fixed in 4% formaldehyde and embedded in paraffin for IHC analysis.
IHC analyses of clinical specimens and tumor xenograft samples were performed using antibodies against R1 (Sigma-Aldrich), Ki-67 (SP6, Abcam), c-Myc (D84C12, Cell Signaling Technology), or HUWE1 (Bethyl Laboratories) following our published protocol (25) with minor modifications. Briefly, formalin-fixed paraffin-embedded sections (4 μm) were deparaffinized, rehydrated, and subjected to antigen retrieval. After incubation in Dual Endogenous Enzyme Block solution (Dako) for 10 minutes, sections were treated with primary antibody diluted by different folds with Antibody Diluent solution (Dako) at 4°C overnight. The section was then washed three times in PBST (PBS containing 0.2% Tween-20) for 5 minutes per washing. To detect specific staining, each section was treated for 30 minutes with EnVision + Dual Link System-HRP (Dako), which contained HRP-conjugated goat antibodies against mouse and rabbit IgG. Sections were washed three times for 5 minutes each and specific stains were developed with 3′3-diaminobenzidine (Dako). Image acquisition was performed using a Nikon camera and software. Magnification was ×400 (scale bars: 20 μm).
Flow cytometric analysis
To determine the effect of R1 on cell-cycle progression, control and R1-KD cells were fixed, stained with propidium iodide (PI, 25 μg/mL), and analyzed by FACScan flow cytometer (BD Biosciences) on the basis of 2N and 4N DNA content to determine the distribution of different cell-cycle phases. To determine the effect of R1 on cell apoptosis, cells subjected to transient or stable KD of R1 were fixed, costained with PI and FITC Annexin V using the FITC Annexin V Apoptosis Detection Kit II (BD Biosciences) following the manufacturer's instructions, and analyzed by BD Accuri C6 (BD Biosciences) to determine the apoptotic cell population characterized by PI−/Annexin V+ cells. Quantitative analysis was conducted by FlowJo software (FlowJo).
Luciferase reporter assays
To determine the direct effect of R1 on c-Myc transcriptional activity, control and R1-KD PC-3 cells were transfected with the pM4-min-tk-luc construct, which contains 4 c-Myc–binding E-box sites located upstream of the Firefly luciferase gene (21) and the pRL-TK Renilla luciferase reporter construct, with the latter used to normalize transfection efficiency. The parental pmin-tk-luc construct without any c-Myc–binding sites was used as a negative control. To examine the direct effect of R1 on HUWE1 promoter, control and R1-overexpressing PC-3 cells were transfected with the desired HUWE-1 promoter Gaussia luciferase reporter plasmids and the pRL-TK plasmid. Transfections were performed with Lipofectamine 2000 reagent. Relative light units were calculated as the ratio of Firefly or Gaussia luciferase activity to Renilla luciferase activity. Renilla luciferase activity was determined in harvested cell lysates by the Dual-Luciferase Reporter 1000 Assay System (Promega). Gaussia luciferase activity was determined in conditioned media by the Secrete-Pair Gaussia Luciferase Assay Kit (GeneCopoeia).
Quantitative real-time PCR
qPCR was conducted using SYBR Green PCR Master Mix and run with the Applied Biosystems 7500 Fast Real-Time PCR System (Applied Biosystems). PCR conditions included an initial denaturation step of 3 minutes at 95°C, followed by 40 cycles of PCR consisting of 30 seconds at 95°C, 30 seconds at 60°C, and 30 seconds at 72°C. The PCR data were analyzed by the 2−ΔΔCt method (26). All primer sequences used were: R1 forward 5′-CGAGGAAGAGGAAGATGAAGAA-3′ and reverse 5′-GAAAACAACTGCTCGGAAGAAC-3′; c-Myc forward 5′-TGCTGCCAAGAGGGTCAAGT-3′ and reverse 5′-GTGTGTTCGCCTCTTGACATTC-3′; HUWE1 forward 5′-GCACTCTGCAATCCTCACAA-3′ and reverse 5′-CCTCTGGCTCTACAGGCATC-3′; FBW7 forward 5′-CGTTGCAGGGGCATACTAAT-3′ and reverse 5′-ATGCAATTCCCTGTCTCCAC-3′; SKP2 forward 5′-TGCTAAGCAGCTGTTCCAGA-3′ and reverse 5′-AAGATTCAGCTGGGTGATGG-3′; CHIP forward 5′-CAATCTGCAGCGAGCTTACA-3′ and reverse 5′-CTGTTCCAGCGCTTCTTCTT-3′; TRUSS forward 5′-AAGGACAGCACCTGCCTAGA-3′ and reverse 5′-GTCCATGTAGGGCTCCTCAA-3′; and β-Actin forward 5′-TTGTTACAGGAAGTCCCTTGCC-3′ and reverse 5′-ATGCTATCACCTCCCCTGTGTG-3′.
Site-directed mutational analysis of R1 and HUWE1 promoters
Site-directed mutagenesis was used to delete the leucine zipper domain in R1 expression construct and to mutate the putative R1-binding site identified in 1.3-kb HUWE1 promoter. Wild-type promoter-luciferase reporter construct was used as a template. Mutagenesis was carried out using QuickChange II XL Site-Directed Mutagenesis Kit (Agilent Technologies) following the manufacturer's instructions. The sequences of primers used for creating R1 deletion promoter were forward 5′-GAGAGTTCAGATGCTAACTCGATGCCAGATTTCTTCCCAGTACG-3′ and reverse 5′-ATCTGGCATCGAGTTAGCATCTGAACTCTCCTGGCTCTCATCC-3′. The sequences of primers used for mutagenesis of HUWE1 promoter were forward 5′-GTCTTGTGCAACATAGGGATGTGTTACAGTTTGATCTGGAGAGATTC-3′ (mutated nucleotides underlined). Deletion and mutated nucleotides were verified by DNA sequencing.
In vivo ubiquitination assays
293T cells were transiently transfected with 2 μg of HA-tagged ubiquitin, 1 μg of c-Myc and/or 1 μg of R1 expression construct in 6-well plates, where pcDNA3.1 was used to keep the amount of transfected DNA equal between groups. Transfections were performed with Lipofectamine 2000 reagent. The cells were split 1:4 into 10-cm dishes 24 hours after transfection to allow growth for another 48 hours. Cells were treated with 10 μmol/L MG-132 for 4 hours prior to harvest and lysed in RIPA buffer supplemented with a protease and phosphatase inhibitor cocktail as described above. One milligram of cell lysates was subjected to immunoprecipitation with anti-c-Myc IgG resin (9E10, Thermo Fisher Scientific) overnight at 4°C using the Pierce Co-Immunoprecipitation Kit (Thermo Fisher Scientific) following the manufacturer's instructions, followed by immunoblotting with anti-HA antibody (12CA5, Sigma-Aldrich) to detect c-Myc protein–ubiquitin conjugates.
Chromatin immunoprecipitation analysis and qPCR
Chromatin immunoprecipitation (ChIP) assays were used to determine the association of R1 protein with HUWE1 promoter in FLAG-tagged R1-overexpressing PC-3 cells by a SimpleChIP Enzymatic Chromatin IP Kit (Cell Signaling Technology) following the manufacturer's instructions. Briefly, chromatin was cross-linked with nuclear proteins, enzymatically digested with micrococcal nuclease followed by sonication, and immunoprecipitated with anti-FLAG antibody (M2, Sigma-Aldrich). After being pelleted with agarose beads and purified, immunoprecipitates were subjected to qPCR with a pair of primers specifically targeting the HUWE1 promoter region that encompasses the R1-binding site. IgG included in this kit was used as a negative control for IP. All primer sequences used were: HUWE1 promoter forward 5′-AACTGCAGGGCTATGGTGAA-3′ and reverse 5′-CACTGGAGTTATGATGCTATGC-3′; MAOA core promoter (serving as positive control) forward 5′-GTGCCTGACACTCCGCGGGGTT-3′ and reverse 5′-TCCTGGGTCGTAGGCACAGGAG-3′; and HUWE1 intron 2 (serving as negative control) forward 5′CCCAGGTGGATTGTTGGGGG-3′ and reverse 5′-GGAACCCTTAAGCTCACACAGCA-3′. Ten percent of chromatin prior to the IP step was saved as input and data were presented as the percent of input from three separate experiments.
Data were presented as the mean ± SEM in figure legends. Correlations were determined by Pearson correlations. All other comparisons were analyzed by unpaired two-tailed Student t test. A P value less than 0.05 was considered statistically significant.
R1 is highly expressed in prostate cancer and associated with poor prognosis/survival in patients with prostate cancer
To determine R1 expression in human prostate cancer, we analyzed R1 protein levels by IHC in a tissue microarray assembling both human normal prostatic (n = 19) and tumor (n = 93) samples. As shown in Fig. 1A, the intensity of widespread R1 staining in both nuclei and cytoplasm was higher in cancer cells than normal prostatic epithelial cells with statistical significance demonstrated by quantitative analysis (Fig. 1B). We also found increased R1 mRNA expression associated with biochemical recurrence and poor survival in patients with prostate cancer in a prostate cancer clinical dataset (Fig. 1C and D; ref. 28). Next, we assayed R1 protein expression in a spectrum of human prostate cancer cell lines with different levels of aggressiveness and invasiveness, including LNCaP, C4-2B, DU145, PC-3, ARCaPE, and ARCaPM, as well as normal human prostatic epithelial cells (PrEC). The R1 protein levels were elevated in all cancer cell lines compared with PrEC cells (Fig. 1E). We also analyzed R1 protein expression in 2 pairs of human prostate normal epithelial (PNE) and prostate cancer epithelial (PCE) cells derived from clinical prostate tumors and adjacent normal prostates. Similarly, these showed increased R1 protein expression in PCE1 and PCE2 cells relative to their corresponding controls (Fig. 1F). In addition, R1 mRNA expression was upregulated in all prostate cancer cell lines compared with PrEC cells (Fig. 1G), which was consistent with our observations of R1 protein changes. From these results in both clinical samples and cell line models, we concluded that R1 is highly expressed in prostate cancer and has potential prognostic value for distinguishing aggressive from indolent prostate cancer.
R1 promotes prostate cancer cell proliferation, colony formation, and prostate tumor growth
To study the potential functional role of R1 in prostate cancer, we stably overexpressed FLAG-tagged R1 in PC-3 cells. Enforced expression of R1 as confirmed by Western blot analysis (Fig. 2A) increased PC-3 cell proliferation by upto 40% in a 6-day observation course compared with control cells where an empty vector was stably expressed (Fig. 2B). Both control and R1-overexpressing cells exhibited similar proliferation rates after confluence was reached, suggesting that contact inhibition is not involved in R1-induced cell proliferation (Supplementary Fig. S1). We also knocked down R1 expression in 2 prostate cancer cell lines, PC-3 and C4-2B, using 2 shRNAs targeting separate nonoverlapping R1 coding regions followed by antibiotic selection to establish stable cell clones with successful KD confirmed by Western blot analysis (Fig. 2C). Stable silencing of R1 significantly suppressed cell growth in both lines with upto 64% and 61% decreases for PC-3 and C4-2B cells, respectively, over a 7-day time course (Fig. 2D). Interestingly, R1 KD did not affect cell apoptosis in PC-3 and C4-2B cells under both steady-state and dynamic conditions, where stable shRNA–mediated R1 KD or a siRNA-mediated stepwise decrease of R1 expression over time was achieved, respectively (Supplementary Fig. S2). Moreover, we showed that overexpression of R1 increased the colony-forming ability of PC-3 cells with an increase of 63% in the number of colonies compared with control cells (Fig. 2E), whereas R1 KD reduced the ability of PC-3 cells to form colonies relative to control cells by an up to 55% drop in colony number (Fig. 2F). In addition, overexpression of R1 also led to a significant increase in PC-3 cell migration and invasion. In contrast, R1 KD reduced the ability of both PC-3 and C4-2B cells to migrate and invade (Supplementary Fig. S3). Taken together, these results obtained from different cell line models strongly support the idea that R1 promotes prostate cancer cell aggressiveness and invasiveness.
To determine whether the R1 effects observed in cell lines could be recapitulated in vivo, we established PC-3 prostate tumor xenograft mouse models. After being implanted subcutaneously into male nude mice, R1-KD PC-3 cells showed slower tumor growth rates in comparison with control cells expressing a control shRNA that targets no known mammalian genes (Fig. 3A). Moreover, R1-KD cells formed tumors that were smaller, with an average tumor weight of 107 ± 63 mg, compared with large tumors with an average weight of 201 ± 39 mg in controls (Fig. 3B). Ki-67 staining of tumor specimens harvested at the experimental endpoint further revealed a 43% decrease of Ki-67+ cells in R1-KD group (Fig. 3C and D). In addition, R1 protein staining showed reduced intensity in R1-KD tumor samples, suggesting that shRNA–mediated silencing of R1 expression is highly effective and sustainable under in vivo conditions (Fig. 3C). These results in aggregate indicate the in vivo tumor-promoting function of R1.
R1 increases c-Myc expression by enhancing c-Myc protein stability
To assess the effect of R1 on cell-cycle progression, which may underlie its proliferation-enhancing effect in prostate cancer cells, we conducted cell-cycle analysis to determine the distribution of different cell-cycle phases in R1-KD cells. As demonstrated in Fig. 4A, R1 KD effectively reduced the percentage of cells in S phase by 23% and 19% in PC-3 and C4-2B cells, respectively, which was accompanied by slight increases in the G0/G1 phase. Because R1 was first identified as a partner protein of c-Myc, a direct regulator of cell-cycle machinery (11, 29), we analyzed c-Myc expression profiles in R1-manipulated cells. Enforced expression of R1 increased c-Myc protein expression in PC-3 cells, whereas shRNA-mediated KD of R1 resulted in a decrease of c-Myc protein levels in both PC-3 and C4-2B cells (Fig. 4B). Using a luciferase reporter construct under the control of 4 c-Myc–binding E-box sites as an indicator of c-Myc transcriptional activity (21), we found decreased c-Myc activity as reflected by lower luciferase reporter activity in R1-KD cells (Fig. 4C) in line with reduced c-Myc protein expression in these cells, suggesting that R1 is able to affect both c-Myc protein expression and activity. We also demonstrated dramatic loss of nuclear c-Myc protein staining in R1-KD PC-3 tumor samples compared with controls (Fig. 4D). However, neither overexpression nor KD of R1 significantly affected c-Myc mRNA expression in these cells (Fig. 4E). This suggests that R1-mediated upregulation of c-Myc may be due to increased protein stability. To determine whether this mechanism is responsible for R1 upregulation of c-Myc, we performed an in vivo ubiquitination analysis of c-Myc protein with or without R1 expression. We found that R1 markedly inhibited both endogenous and exogenous c-Myc ubiquitination (Fig. 4F), with an average 60% (n = 5) inhibition of exogenous c-Myc ubiquitination achieved (Fig. 4G). To determine whether R1/c-Myc physical interaction is a possible mechanism underlying R1 upregulation of c-Myc, we examined the effect of a R1 deletion construct deficient in the leucine zipper domain required for R1 binding to c-Myc on c-Myc ubiquitination (11). As shown in Supplementary Fig. S4, both wild-type and deleted R1 suppressed c-Myc ubiquitination to a similar extent, suggesting that R1/c-Myc interaction is not involved in R1 upregulation of c-Myc. To examine the effect of R1 on c-Myc proteolysis, we subjected R1-manipulated cells to cycloheximide treatment to inhibit protein synthesis. Consistent with the inhibition of ubiquitination, transient overexpression of R1 significantly reduced the proteolysis of c-Myc protein in cells accompanied by a nearly 2-fold increase in c-Myc half-life (Fig. 4H and I, t1/2 = 51 ± 8 minutes and t1/2 = 85.7 ± 2.3 minutes in control and R1-overexpressing cells, respectively, n = 3 for both), which is parallel to decreased c-Myc protein stability along with a 50% drop of c-Myc half-life in R1-KD cells (Fig. 4J and K, t1/2 = 48 ± 5 minutes and t1/2 = 24 ± 2 minutes in control and R1-KD cells, respectively, n = 3 for both). In addition, we showed that R1 KD failed to suppress the growth and colony formation of PC-3 cells that received prior shRNA-mediated KD of c-Myc (Fig. 4L and M), indicating that c-Myc is a functional mediator of R1 effects.
R1 transcriptionally suppresses the E3 ligase HUWE1 that mediates the ubiquitination and degradation of c-Myc protein
The stability of c-Myc protein has been reported to be controlled by a number of E3 ligases that target c-Myc for proteasome-mediated degradation (30, 31), which led us to speculate on possible alterations of these E3 ligases in response to R1 elevation in prostate cancer cells. By qPCR screening the expression levels of 5 E3 ligases, FBW7, SKP2, ChIP, TRUSS, and HUWE1, known to directly regulate c-Myc (32–36), we demonstrated that R1 KD significantly increased HUWE1 expression in both PC-3 and C4-2B cells with the others remaining unaffected by R1 (Fig. 5A). We also showed a 43% decrease of HUWE1 mRNA levels in R1-overexpressing cells compared with control cells (Fig. 5B). Moreover, enforced expression of R1 downregulated HUWE1 protein levels in PC-3 cells (Fig. 5C). HUWE1 is mainly expressed in the cytoplasm but also partially in the nucleus to mediate ubiquitin-dependent degradation of target proteins in different subcellular locations (37). The subcellular localization of HUWE1 may be regulated by the modulated exposure of the nuclear localization signal of HUWE1 located in the middle of the protein distal to the WWE domain (38). Examining the expression patterns of HUWE1 in PC-3 tumor samples, we further observed dramatic induction of widespread cytoplasmic and partial nuclear HUWE1 protein staining when R1 was knocked down (Fig. 5D).
Considering the innate nature of R1 as a transcription repressor competing with the Sp family of transcription factors to downregulate target gene expression, as first demonstrated in transcriptional regulation of monoamine oxidase genes (12, 13), we next determined whether R1 directly regulates HUWE1 at the transcription level. To explore this idea, we analyzed 1.3-kb HUWE1 promoter sequences and identified a region (−224/−218) that exhibits strong sequence similarity to the canonical GC-rich R1-binding site (Fig. 5E). Using a 1.3-kb DNA segment located upstream of the transcription start site of HUWE1 as a template, we generated a mutant HUWE1 promoter reporter construct harboring 3 point mutations in the center of the putative R1-binding element. Compared with the wild-type HUWE1 promoter reporter showing a R1-induced 26% decrease in promoter activity, the mutated HUWE1 promoter was no longer repressed by overexpression of R1 in PC-3 cells (Fig. 5F). To confirm direct occupancy of R1 with the sequences in the HUWE1 promoter in vivo, we performed ChIP assays. We isolated chromatin-nuclear protein complexes immunoprecipitated with anti-FLAG antibody from FLAG-tagged R1-overexpressing PC-3 cells, and analyzed these by qPCR using primers that specifically encompass the putative R1-binding site in HUWE1 promoter. As shown in Fig. 5G, we were able to detect a physical association of R1 with the HUWE1 promoter sequences, which is parallel to the expected binding of R1 with the MAOA core promoter serving as a positive control. Moreover, limited signals were seen from the negative controls, where either nonspecific IgG antibody was used in the immunoprecipitation step or HUWE1 intron 2 was probed to confirm the specificity of R1 binding to the HUWE1 promoter sequences. In addition, we demonstrated a negative coexpression correlation between R1 and HUWE1 in a previously reported prostate cancer clinical dataset (Fig. 5H, P = 0.0372 by Pearson correlation; ref. 27), which is consistent with our findings in cell lines. These results collectively demonstrate that R1 suppresses the transcription of the c-Myc–targeting E3 ligase HUWE1 by directly interacting with its promoter to upregulate c-Myc protein expression in prostate cancer cells.
In summary, our data suggest that the increased intrinsic R1 expression in prostate cancer activates cell proliferation and cell-cycle progression by a mechanism that involves direct transcriptional suppression of the E3 ligase HUWE1 to prevent c-Myc protein degradation, thereby activating c-Myc protein expression (Fig. 6). Our data further showed the key features of this pathway in prostate tumor xenograft samples and in clinical datasets, supporting the essential role of R1 in prostate tumor growth and progression.
In this study, we demonstrated for the first time the elevated expression of R1 in human prostate cancer tissue samples compared with normal counterparts. We also showed higher R1 expression at both protein and mRNA levels in a spectrum of human prostate cancer cell lines which exhibit different characteristics and behaviors compared with normal human prostatic cells. Overexpression and KD of R1 in human prostate cancer cell lines revealed that R1 induces cell proliferation, colony formation, and migration/invasion. Moreover, silencing R1 reduced the growth of PC-3 tumor xenografts in mice. Although R1 was first identified as a c-Myc oncoprotein interactor, little progress has been made so far in understanding its role in cancer in general and in prostate cancer specifically. Previous studies showed that R1 enhances medulloblastoma transformation with induced aggressive phenotypes (11, 39) and also hepatocellular carcinoma progression (40). Our results obtained from prostate cancer are consistent with those observations, further supporting a tumor-promoting effect of R1 in cancers.
R1 was shown to upregulate c-Myc protein expression and transcriptional activity by enhancing protein stability. R1 has been found to physically interact with the c-Myc NH2-terminal domain essential for modulating c-Myc oncogenic properties, as demonstrated in different cellular settings including medulloblastoma and neuroblastoma cells (11, 41). This interaction results in two possible mechanisms where R1 either acts directly as an E3 ligase or regulates other factor(s) to indirectly modulate c-Myc protein stability. The lack of both the E6AP carboxyl terminus (HECT) and the really interesting new gene (RING) domains in R1 protein sequences, two domains mediating the direct transfer of ubiquitin from E2 to substrate, disqualifies R1 from being an E3 ligase candidate (13, 42). Rather, we demonstrated that R1 acts as a repressor to transcriptionally suppress HUWE1 expression, a c-Myc–targeting E3 ligase, to indirectly stabilize c-Myc protein in the present model systems. Compelling evidence has indicated the critical role of HUWE1 as an E3 ligase regulating c-Myc protein stability, c-Myc–activated genes, and c-Myc–driven neoplasia and oncogenesis (refs. 34 and 43–46). As reported in several independent studies by different groups, manipulation of HUWE1 levels alone was sufficient to produce dramatic changes in c-Myc ubiquitination (34, 47), which to some extent supports our proposed model where R1 exerts a significant impact on c-Myc protein stability by modulating a single c-Myc–regulating E3 ligase. In line with the previous observations of enhanced c-Myc–transforming activity by R1/c-Myc interaction in medulloblastoma (11), increased R1 expression in prostate cancer also exacerbated c-Myc function in prostate cancer cells by promotion of cell-cycle progression. HUWE1 has been previously reported to stabilize MIZ-1, another binding partner of c-Myc, and affect ARF–p53-mediated apoptosis (44, 46), but we demonstrated marginal effects of R1 on MIZ-1 and p53 in prostate cancer cells (Supplementary Fig. S5), suggesting that the downstream effect of R1-HUWE1 axis might be cell-context and/or cell-type dependent. In addition, we showed that c-Myc is a key downstream mediator of R1′s effect in prostate cancer cells as evidenced by the abolition of R1 shRNA-induced suppression of proliferation and colony formation of PCa cells by c-Myc silencing. Although the current study emphasizes c-Myc protein stability as a likely major mechanism underlying R1 regulation of c-Myc in prostate cancer cells, other possible c-Myc–centric mechanisms warrant further investigations to more comprehensively advance our understanding of how R1 contributes to the oncogenic activity of c-Myc and how the R1/c-Myc protein complex regulates prostate cancer as well as other malignancies.
Activation of the proto-oncogene c-Myc is an important step not only in the early phases of prostate cancer such as PIN but also throughout the entire progression of prostate cancer including advanced recurrent and metastatic stages (7). The significant influences of c-Myc in prostate cancer are mainly attributed to its transcriptional regulation of numerous downstream target genes. Several independent microarray gene expression analyses from human prostate cancer and prostate-specific Hi-MYC mice have thus far identified a distinct c-Myc–driven expression signature (7, 8). On the other hand, c-Myc is also able to cooperate with other oncogenic signaling such as AKT to promote prostate tumorigenesis and alter sensitivity to therapy (48). These accumulated gene and pathway profiles lay the mechanistic foundations for identifying potential R1-interacting factors, given the intimate regulatory relationship between R1 and c-Myc demonstrated by other groups and by us. This idea has been coincidentally supported by recent findings of AKT activation by R1 in medulloblastoma (39). Future studies exploring the gene and signaling networks governed by R1 are merited to uncover potential unknown functions of R1 in prostate cancer.
One notable finding of our study was the discovery that R1 directly interacts with an element in HUWE1 promoter to suppress gene transcription, which was supported by clinical evidence of a negative coexpression correlation between R1 and HUWE1. According to the literature, R1 functions in cancers largely through its role as a transcription factor. R1 forms a ternary complex with transcription coactivator LEDGF/p75 and chromatin in transcriptional regulation, which may contribute to the development of MLL fusion–driven acute leukemia (14, 49). In medulloblastoma, the cooperation of R1 with LEDGF/p75 also promotes cell migration and metastasis by activation of AKT signaling (39). However, the direct suppressing effect of R1 by serving as a transcription repressor on MAOA gene expression, which was first demonstrated in neuroblastoma cells, turns out to be marginal in prostate cancer, leading to activation of MAOA (13, 50). These contrasting studies suggest a context-dependent role of R1 in the transcriptional regulation of target genes in different types of cancers. A cancer type–specific delineation of molecular states associated with R1 may advance our understanding of R1-mediated transcriptional machinery in diverse cancers.
We provided clinical evidence that higher R1 expression is associated with recurrence and decreased survival in patients with prostate cancer, which indicates the potential of R1 for prediction of disease prognosis in prostate cancer, particularly in advanced phenotypes. These clinical analyses are consistent with in vitro observations of an increasing trend of R1 protein expression from weak/low-metastatic to aggressive/high-metastatic cells.
In summary, we present the first study showing increased intrinsic R1 expression in prostate cancer and its association with poor clinical outcomes in patients. We also uncovered the underlying molecular mechanisms contributing to R1-induced prostate cancer growth and progression. Our findings highlight the novel role of R1 in prostate cancer and reveal R1 as a new potential prognostic marker and therapeutic target for prostate cancer.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Conception and design: H.E. Zhau, L.W.K. Chung, B.J. Wu, J.C. Shih
Development of methodology: T.-P. Lin, B.J. Wu, J.C. Shih
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J. Li, Q. Li, X. Li, N. Zeng, J.-M. Huang, C.-Y. Chu,B.J. Wu, J.C. Shih
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): T.-P. Lin, J. Li, X. Li, N. Zeng, B.J. Wu, J.C. Shih
Writing, review, and/or revision of the manuscript: T.-P. Lin, B.J. Wu, J.C. Shih
Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): B.J. Wu, J.C. Shih, C. Liu, Q. Li, C.-H. Lin
Study supervision: B.J. Wu, J.C. Shih
The authors thank Bernhard Luscher (Hannover Medical School, Hannover, Germany) for providing pM4-min-tk-luc and pmin-tk-luc plasmids, Bin Qian (Department of Pharmacology and Pharmaceutical Sciences, University of Southern California, Los Angeles, CA) for providing technical assistance, and Gary Mawyer for editorial assistance.
This work was supported by the Department of Defense Prostate Cancer Research Program grants W81XWH-12-1-0282 (to J.C. Shih and H.E. Zhau) and W81XWH-15-1-0493 (to J.B. Wu and H.E. Zhau); the Daniel Tsai Family Fund and the Boyd and Elsie Welin Professorship (to J.C. Shih); and NIH/NCI grant 2P01CA098912, the Board of Governors Cancer Research Chair, and the Steven Spielberg Fund in Prostate Cancer Research (to L.W. Chung).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.