PRMT5 promotes cell proliferation by inhibiting BTG2 expression via the ERK signaling pathway in hepatocellular carcinoma

Abstract Increasing evidence suggests that PRMT5, a protein arginine methyltransferase, has roles in cell growth regulation and cancer development. However, the role of PRMT5 in hepatocellular carcinoma (HCC) progression remains unclear. Here, we showed that PRMT5 expression was frequently upregulated in HCC tissues, and its expression was inversely correlated with overall survival in HCC patients. PRMT5 knockdown markedly inhibited in vitro HCC proliferation and in vivo tumorigenesis. We revealed that the mechanism of PRMT5‐induced proliferation was partially mediated by BTG downregulation, leading to cell cycle arrest during the G1 phase in HCC cells. Ectopic BTG2 overexpression decreased HCC growth, caused cell cycle arrest at the G1 phase, and downregulated Cyclin D1 and Cyclin E1 protein expression. Furthermore, we found that PRMT5‐induced ERK phosphorylation regulated BTG2 expression in HCC cells, whereas pretreatment with a selective ERK1/2 inhibitor (PD184352) significantly reversed the effect of PRMT5 on BTG2 expression. Our results indicated that PRMT5 promotes HCC proliferation by downregulating BTG2 expression via the ERK pathway.


Introduction
Hepatocellular carcinoma (HCC) is the third leading cause of cancer-related mortality worldwide, causing approximately 600,000 deaths each year [1]. In China, HCC is the third leading cause of cancer death and the fourth most frequently diagnosed cancer [2]. Radical surgical resection of tumors and liver transplantation are the curative ORIGINAL RESEARCH PRMT5 promotes cell proliferation by inhibiting BTG2 expression via the ERK signaling pathway in hepatocellular carcinoma treatments for liver cancer. However, because of its late symptom presentation, most HCC patients are not candidates for surgical resection or liver transplantation, and the therapeutic options for these advanced HCC patients are limited. Currently, sorafenib is the only FDA-approved drug for advanced liver cancer. However, the clinical benefits of sorafenib remain unacceptable [3], and new targets for HCC therapy are urgently needed.
Protein arginine methyltransferase 5 (PRMT5) is the major type II arginine methyltransferase and catalyzes the symmetric methylation of histone and nonhistone proteins. Recently, accumulating evidence has shown that PRMT5 modulates cell growth and transformation. Through its methyltransferase ability, PRMT5 regulates the expression of a wide spectrum of target genes [4]. PRMT5 methylates the histones H3R8 and H4R3, leading to cell cycle gene silencing (CDKN2A and CCNE) and tumor suppressor gene silencing (ST7, NM23, RBL2, and SMAD7) [4][5][6]. Additionally, PRMT5 methylates nonhistone proteins. For example, an experiment using MCF-7 breast cancer cells showed that inhibiting PRMT5 decreased p53 stability and the expression of p53 target genes [7]. PRMT5 knockdown inhibits proliferation and cell cycle transition from the G1 to S phase, and a large number of corroborating studies have shown that PRMT5 levels are elevated in various cancers, including gastric, colorectal, lung, lymphoma, leukemia, and liver cancers [5,[8][9][10][11]. However, the role of PRMT5 in HCC progression has not been fully elucidated. Identifying its downstream targets will aid in elucidating the mechanism of PRMT5 in liver cancer development.
B-cell translocation gene 2 (BTG2) was first identified in 1991 in rat PC12 cells induced by nerve growth factor and in mouse 3T3 fibroblasts stimulated with tetradecanoylphorbol acetate [12,13]. BTG2 belongs to a family of antiproliferation genes [14]. BTG2 overexpression is associated with cell cycle arrest from G1 to S phase by suppressing Cyclin B1, Cyclin D1, and Cyclin E1 [15,16]. Numerous studies have shown that as a tumor suppressor, BTG2 is inhibited in human cancers, such as skin, lung, breast, and gastric cancers, and in bladder carcinoma [17][18][19][20]. To date, mutations in BTG2 have not been detected, indicating that epigenetic modification might be the main mechanism by which BTG2 expression is modulated. Recently, the chromatin-modifying enzyme SETD1A was found to regulate BTG2 expression through an miRNA network. However, whether arginine methyltransferases regulate BTG2 expression in HCC is unknown.
In this study, we found that PRMT5 was overexpressed in HCC and that PRMT5 promoted in vitro and in vivo cancer cell growth. We detected an inverse correlation between PRMT5 and BTG2 expressions. Inhibition of PRMT5 elevated BTG2 expression by activating ERK signaling. Moreover, silencing BTG2 expression reversed the effects induced by PRMT5 knockdown in HCC cells.

Materials and Methods
Cell lines and reagents HEK 293T cells and the HCC cell lines LO2, Huh7, Bel7402, PLC/PRF/5, HepG2, MHCC97H, and SK-Hep1 were purchased from the Chinese Academy of Sciences Cell Bank of Type Culture Collection (CBTCCCAS). The cell lines were maintained in DMEM (Gibco, Carlsbad, CA) supplemented with 10% fetal bovine serum (Gibco) in an atmosphere of 5% CO 2 at 37°C. ERK1/2 inhibitor (PD184352) and the selective PRMT5 inhibitor GSK591/ EPZ015866 were purchased from Selleck (Houston, TX, USA).

Patient samples and tissue microarray
Hepatocellular carcinoma and corresponding adjacent normal liver tissue samples were collected after HCC tumor resection at Sun Yat-sen Memorial Hospital, Guangzhou. In total, 138 HCC patients who received curative resection between 2005 and 2012 at Sun Yat-sen Memorial Hospital were enrolled in this study. The study protocol was approved by the Institute Review Board of the University of Sun Yat-sen, and written informed consent was obtained from each patient. HCC patients with a distinctive pathologic diagnosis, no preoperative systemic or local treatment, and complete follow-up data were selected for inclusion in this study. The pTNM classification for HCC was defined according to The American Joint Committee on Cancer/International Union Against Cancer staging system (7th edition, 2010). To detect PRMT5 mRNA levels, 48 pairs of HCC samples and corresponding adjacent normal liver tissues were collected after surgical resection and stored in liquid nitrogen. A tissue microarray (TMA) consisting of 138 pairs of HCC tissues and corresponding adjacent normal liver tissues was constructed by Shanghai Biochip Company (Shanghai, China).

Immunohistochemistry
The expression of PRMT5 and BTG2 proteins was assessed with immunohistochemistry in a tissue microarray using rabbit polyclonal antibodies specifically against PRMT5 (Abcam, Cambridge, USA ab109451), BTG2 (Abcam, Cambridge, USA ab85051), and Ki67 (Abcam, Cambridge, USA ab15580). Immunohistochemistry staining was performed using the Dako Envision Plus System (Dako, Carpinteria, CA) according to the manufacturer's protocol.

Establishment of cell lines stably knocking down PRMT5 and overexpressing BTG2
PRMT5 (human) shRNA lentiviral vectors were purchased from Shanghai Genechem Co. Ltd. (Shanghai, China), and BTG2 (human) overexpression lentiviral vectors were purchased from Shanghai Generay Biotech Co. Ltd. Lentivirus was produced by cotransfection of 293T cells with viral vectors (pCMV-MDL, pCMV-REV, and pCMV-VSVG). Medium containing virus was collected and concentrated using PEG8000 (Sigma-Aldrich). Infected cells were treated with puromycin (final concentration: 1 μg/mL) for 2 weeks of selection.

MTT proliferation assay
Cell proliferation was analyzed using an MTT assay according to the manufacturer's protocol. In brief, 1000 cells in 100 μL/well were seeded in triplicate in 96-well plates for 1-72 h at 37°C in 5% CO 2 . Then, 20 μL of MTT solution was added to each well and incubated with the cells for 4 h. Then, 150 μL DMSO was added to each well and incubated for 15 min. Absorbance at 570 nm was measured with a microplate reader.

Colony formation assay
Hepatocellular carcinoma cells were seeded at 1000 cells per well into six-well plates. The number of colonies was assessed using crystal violet staining after 2 weeks. Three independent experiments were performed, and the results are expressed as the mean ± SD.

Cell cycle analysis
Post-treatment, cells were harvested, washed twice with PBS, fixed in 75% EtOH, and stained with PI. Stained cells were analyzed with a flow cytometer using Beckman Coulter (Brea, CA) and ModFit (Verity, ME) software.

Nuclear extracts
Nuclear extracts were obtained with an N-PER kit (Pierce, Waltham, USA) according to the manufacturer's instructions.

Xenograft tumors in nude mice
A xenograft tumor growth assay was performed by subcutaneously injecting Huh7 cells (5 × 106) stably transduced with lentiviruses encoding scrambled shRNA (shControl) or shRNA targeting PRMT5 (shPRMT5) in 0.2 mL of phosphate-buffered saline (PBS) into the right shoulder of 6-week-old BALB/c nude mice. All the mice were sacrificed 4 weeks after injection. Tumors were removed and weighed. All the animal experiments were performed according to the institutional standard guidelines at Sun Yat-sen University. The IRB number is IACUC-DD-17-0803, and the animal study approved number is SYXK (Yue) 2016-0112.

Statistical analysis
All the statistical analyses were performed with SPSS 19.0 or GraphPad Prism 6.0. Paired t-tests were used to

PRMT5 overexpression in HCC tissues and correlation with patient clinicopathological features and survival
To investigate PRMT5 expression in HCC, we performed immunohistochemical staining on tumor tissue microarrays (TMAs) containing pairs of HCC specimens and corresponding pericancerous liver tissues from an HCC cohort (n = 138). Immunostaining for PRMT5 was observed predominantly in the cytoplasm but was occasionally seen in nuclei (Fig. 1A1). Of the 138 HCC cases, 117 cases (84.7%) were positive for PRMT5 expression with a score>5; in contrast, 68 of 138 cases (49.2%) had positive scores in adjacent liver tissues (Fig. 1A2). Taking the staining intensity into account, the PRMT5 staining score in HCC was also higher than that in adjacent nontumor tissues (Fig. 1A3). We confirmed PRMT5 upregulation in an additional 48 paired HCC tissues using real-time PCR (Fig. 1B). Furthermore, PRMT5 protein levels were increased in HCC cell lines compared to those in a normal liver cell line (LO2) (Fig. 1C). Correlation analysis demonstrated that PRMT5 expression was positively correlated with tumor size (Table 1). Kaplan-Meier analysis showed that enhanced PRMT5 expression was correlated with reduced overall survival and higher recurrence rates for patients with HCC (Fig. 1D). Multivariate Cox regression analysis determined that PRMT5 overexpression is an independent and significant predictor of survival and recurrence in HCC after surgical resection ( Table 2).

PRMT5 knockdown inhibits in vitro and in vivo HCC cell proliferation
To explore the role of PRMT5 in HCC cell proliferation, we established two cell lines, Huh7-shPRMT5 and SK-shPRMT5, which were stably transduced with shPRMT5 lentivirus. Significant inhibition of endogenous PRMT5 expression in Huh7 and SK-Hep1 cells was confirmed by western blotting analysis ( Fig. 2A). MTT (Fig. 2B) and colony formation (Fig. 2C) assays demonstrated that stable silencing of PRMT5 significantly decreased HCC cell proliferation. Flow cytometry analysis showed that PRMT5 inhibition induced cell cycle arrest at the G1 phase in HCC cells (Fig. 2D).
To determine whether PRMT5 provided an in vivo growth advantage to HCC cells, xenograft studies were performed. Tumors in mice injected with cells stably knocking down PRMT5 were smaller and lighter than those in the control group ( Fig. 3A and B). In addition, xenograft tumor sections were stained for Ki67 to observe tumor proliferation status, which indicated that tumors derived from cells with stably silenced PRMT5 showed significantly reduced proliferation (Fig. 3C and D). Together, these results showed that inhibition of PRMT5 reduced cancer cell proliferation and tumor growth.

PRMT5 downregulates BTG2 expression in HCC cells
By analyzing the GEO database, we found that BTG2 levels were significantly upregulated in several cancers, including lung cancer (www.ncbi.nlm.nih.gov/geo/query/ acc.cgi?acc=GSE56757) and prostate cancer (www.ncbi. nlm.nih.gov/geo/query/acc.cgi?acc=GSE65965), after knockdown of PRMT5. However, the regulation of BTG2 by PRMT5 in HCC has not been verified. Therefore, we performed RT-qPCR and western blotting in HCC cells after PRMT5 knockdown. Specific knockdown of PRMT5 significantly increased BTG2 mRNA levels in Huh7-shPRMT5 cells and SK-shPRMT5 cells by three-to fourfold when compared with those in the scrambled shRNA-treated group (Fig. 4A). Likewise, BTG2 protein levels were significantly upregulated after PRMT5 inhibition in HCC cells (Fig. 4B). Furthermore, immunohistochemistry showed that PRMT5 expression was negatively correlated with BTG2 expression in HCC tissues ( Fig. 4C and D).

PRMT5 downregulates BTG2 expression through ERK signaling
BTG2 has been reported to be regulated by ERK signaling in oral squamous cancer cells. Thus, we examined this possibility in HCC cancer cells. The levels of BTG2,p-cRaf, and p-ERK were both increased after PRMT5 knockdown in HCC cells (Fig. 5A). After HCC cells were pretreated with an ERK phosphorylation inhibitor (PD184352) for 24 h to block ERK activation, the increased BTG2 protein levels induced by PRMT5 inhibition were significantly suppressed (Fig. 5B). A previous study indicated that PRMT5 activity was necessary for its function, prompting us to treat HCC cells with a potent and specific PRMT5 chemical probe, GSK591. After treatment with 500 nmol/L GSK591 for 4 days, a significant loss of PRMT5-catalyzed methylarginine on H4 (H4R3me2s), upregulation of BTG2, and phosphorylation of ERK were observed, whereas inhibiting ERK phosphorylation with PD184352 treatment reversed these effects (Fig. 5C). These data indicated that PRMT5 downregulates BTG2 expression through ERK signaling and demonstrates the requirement for PRMT5 activity in its HCC-promoting function.

BTG2 overexpression causes cell cycle arrest in the G1 phase and inhibits in vitro HCC proliferation
To investigate the role of BTG2 in HCC cells, we established two stable cell lines (Huh7-BTG2 and SK-BTG2) infected with BTG2-overexpressing lentivirus (LV-BTG2). BTG2 expression was confirmed by western blotting. BTG2 overexpression inhibited Cyclin D1 and Cyclin E1 expressions, as detected by immunoblotting analyses (Fig. 6A). The growth of Huh7-BTG2 and SK-BTG2 cells was significantly decreased compared to Huh7-Control and SK-Control cells as determined with an MTT assay (Fig. 6B). To assess the effect of BTG2 on tumorigenicity, a colony formation experiment was performed. Colony formation by Huh7-BTG2 and SK-BTG2 cells was significantly inhibited (Fig. 6C). Additionally, we analyzed changes in cell cycle distribution. In BTG2-overexpressing HCC cells (Huh7-BTG2 and SK-BTG2 cells), the percentage of cells in the G1 phase was higher than that in the Huh7-Control and SK-Control cells. The quantitative analysis showed that Huh7-BTG2 and SK-BTG2 cells exhibited 11.1% and 16.2% increases in the G1 phase population. In contrast, the cell percentage in S and G2/M phases decreased in BTG2-overexpressed cells (Fig. 6D1 and  D2).

PRMT5 promotes HCC proliferation in part by inhibiting BTG2 expression
Considering that inhibition of PRMT5 induced BTG2 expression in HCC cells and that BTG2 inhibited proliferation in HCC cells, we speculated that knockdown of PRMT5 inhibited HCC cell growth by stimulating BTG2 expression. To verify this hypothesis, Huh7-shPRMT5 cells and SK-shPRMT5 cells were transfected with siRNA (siBTG2), and Huh7-shControl cells and SK-shControl cells were transfected with scrambled RNA (siNT). The protein levels of PRMT5, BTG2, Cyclin D1, and Cyclin E1 were detected by western blotting (Fig. 7A). PRMT5 inhibition significantly enhanced BTG2 expression, decreased Cyclin D1 protein levels, and arrested the cell cycle at the G1 phase. Depletion of BTG2 by siRNAs partially reversed this effect (Fig. 7B).

Discussion
Accumulating evidence has demonstrated that HCC is a disease involving the accumulation of multiple epigenetic and genetic alterations [21]. These alterations lead to tumor suppressor inactivation and the activation of oncogenes [4]. PRMT5 belongs to the PRMT family of enzymes; these enzymes have been implicated in gene transcription, signal transduction, and DNA repair. PRMT5 is the major symmetric arginine methyltransferase and is ubiquitously expressed and regulates important cellular functions, including cell growth, proliferation, and differentiation [5,[22][23][24]. Dysregulation of PRMT5 has been implicated in carcinogenesis, including lung cancer, gastric cancer, and prostate cancer [25][26][27].
In this study, we explored PRMT5 oncogenicity in HCC. We found that PRMT5 was frequently upregulated in HCC tissues and HCC cell lines compared with corresponding adjacent noncancerous tissues and a normal liver cell line (LO2), respectively. PRMT5 overexpression was associated with tumor size (P = 0.003). In addition, patients positive for PRMT5 expression exhibited shorter overall survival time and recurrence time than patients negative for PRMT5 expression. Furthermore, multivariate analysis revealed that PRMT5 expression was an independent and significant risk factor for survival and recurrence following curative resection. These data suggested that PRMT5 might be an oncogene and a useful prognostic biomarker.
One of the procancer mechanisms of PRMT5 is the silencing of tumor suppressor genes. Previous studies have reported that PRMT5 regulates cell proliferation by silencing the tumor suppressor genes ST7, NM23, RBL2, and SMAD7 [23,28]. However, the mechanism by which PRMT5 regulates cancer cell proliferation remains undetermined. By analyzing the GEO database, we found that the level of BTG2 was significantly upregulated in several cancers, including lung cancer and prostate cancer, after PRMT5 knockdown. Notably, BTG2 has been implicated in the growth of gastric cancer and liver cancer cells [15,19]. Importantly, our study demonstrated that BTG2 was significantly upregulated in PRMT5-silenced HCC cells. In addition, immunohistochemistry showed that PRMT5 expression was inversely correlated with BTG2 protein expression in HCC tissues. These data demonstrate that BTG2 might be a PRMT5 target.
The MAPK pathway plays a crucial role in cell proliferation, differentiation, survival, and apoptosis [29][30][31][32]. A large number of reports have shown that ERK activation suppresses cell proliferation or causes cell death [33,34]. The amplitude and duration of the ERK signal play a key role determining the final biological outcome [34]. Some reports have suggested that a large and sustained increase in ERK activation suppresses cell proliferation, whereas rapid and transient activation of ERK induces cell growth [35,36]. Pedro et al. [37] reported that inhibition of PRMT5 activity or mutation of RAF affected ERK signal amplitude and duration, redirecting PC12 cells from proliferation to differentiation. Recently, Lee et al. [16] reported that epigallocatechin-3-gallate increased BTG2 levels via p38 and the ERK1/2 signaling pathways, which suppressed OSCC cell growth. Therefore, we hypothesized that PRMT5 regulated BTG2 expression via the ERK pathway. Our results showed that PRMT5 inhibition in HCC cells significantly induced p-Raf, p-ERK1/2, and BTG2 expressions. The BTG2 protein increase induced by PRMT5 inhibition was blocked when cells were pretreated with an ERK inhibitor (PD0325901). In addition, Chen et al. [27] reported that PRMT5 activity was necessary for its phenotypes. To provide additional evidence for this hypothesis, we used GSK591 to block PRMT5 activity. Our data indicated that blocking PRMT5 activity had the same effects as inhibiting PRMT5 expression. Collectively, our data showed that downregulation of PRMT5 or its activity enhanced BTG2 expression via the ERK1/2 MAPK signaling pathway.
Consistent with a previous report, our study showed that BTG2 upregulation significantly induced cell cycle arrest in the G1 phase and suppressed proliferation in HCC cells [15]. Because PRMT5 inhibition decreased HCC proliferation and induced BTG2 expression, we hypothesized that BTG2 functions downstream of PRMT5 in the proliferative regulation of HCC cells. Our further cell cycle and western blotting assays verified that PRMT5 inhibition induced G1/S arrest in HCC cells by inhibiting Cyclin D1 and Cyclin E1, whereas pretreatment of cells with siRNA targeting BTG2 in shPRMT5-HCC cells reversed this effect. These results indicated that the regulatory effect of PRMT5 on the cell cycle and proliferation in HCC was partially mediated by BTG2.
In conclusion, we found that PRMT5 was frequently overexpressed in HCC and that its enhanced expression was associated with a poor prognosis. PRMT5 inhibition induced cell cycle arrest in the G1 phase and inhibited in vitro and in vivo proliferation. PRMT5 inhibition increased BTG2 expression through ERK phosphorylation, thus inducing cell cycle arrest and suppressing proliferation in HCC cells. These results suggest that PRMT5 might be a potential new therapeutic target in HCC.