POLR2A Drives Tumor Cell Aggressive in Both VHL and PBRM1 Mutation Renal-Cell Carcinoma

Background: Loss of VHL always results in the loss of PBRM1 and causes aggressive clear cell renal cell carcinoma. However, VHL mutation was not signicantly associated with worse survival, and PBRM1 modulate the tumor behavior is not clear. Thus, exploration of key molecules promoting the tumor aggressive is urgent in both VHL and PBRM1 RCC patient. Methods and results: POLR2A was screened out by analyzing The Cancer Genome Atlas mutation data. Gene Set Enrichment Analysis results showed that E2F, G2M, and mTOR1 pathways were all altered in response to POLR2A high expression. Furthermore, In vitro, knockdown of POLR2A in 769-P and 786-O cells resulted in cell growth arrest and cell cycle blockade compared to control cells, the mechanism though decreasing cyclin D1-CDK4 axis. In vivo results were conrmed 786-O cells in which POLR2A expression was silenced, exhibited tumor growth inhibited compared to control group. Conclusions: POLR2A was the key protein after VHL and PBRM1 mutations in RCC, inhibition of POLR2A crippled cell viability and proliferation in vivo and in vitro, We anticipate POLR2A represents a novel candidate for RCC treatment.


Background
Clear cell renal cell carcinoma (ccRCC) is a type of kidney cancer that displays a variety of clinical behaviors [1,2]. Approximately 85-90% of renal cell carcinoma is ccRCC, and approximately 403,262 new cases of RCC occurred, with an estimated 175,098 deaths worldwide in 2018 [1]. The prognosis of metastatic ccRCC is poor, and the 5-year survival rate is <10% after diagnosis [3].
3p loss occurs in 70% of ccRCC patients and leads to the loss of tumor suppressor genes Von Hippel-Lindau (VHL), Polybromo-1 (PBRM1), Set domain-containing 2 (SETD2), and BRCA1-associated protein-1 (BAP1) [4]. Subsequently, the loss of VHL always results in the loss of PBRM1 and causes aggressive ccRCC [5][6][7]. The VHL gene, which encodes a tumor suppressor, the VHL protein, is highly mutated in ccRCC, and its mutation rate is approximately 47% [8]. The function of the VHL protein is to participate in the degradation of hypoxia-inducible factor (HIF) protein [9]. VHL mutation leads to VHL protein loss of function and causes HIF protein accumulation [10]; furthermore, the HIF protein activates the mammalian target of rapamycin (mTOR) and vascular endothelial growth factor (VEGF) signaling pathways [11]. To date, sunitinib and pazopanib (inhibitors of the VEGF/VEGFR pathway) and everolimus and temsiromus (inhibitors of mTOR kinase) have been used as clinical treatments for ccRCC patients [3]. However, neither VHL mutation nor deletion is signi cantly associated with worse survival in ccRCC patients according to Young et al's report [9]. PBRM1, which is involved in a crucial step in ccRCC carcinogenesis, encodes the BRG1-associated factor 180 (BAF180) protein, a subunit of the nucleosome remodeling complex [12]. In ccRCC, PBRM1 has an up to 40% mutation rate and a highly methylated gene promoter [8]. Loos of BAF180 disrupts the nucleosome remodeling complex and causes aggressive RCC, prostate cancer, and non-small cell lung cancer [13][14][15][16]. In Kapur et al's. report, BAF180 is not an independent predictor of outcome in ccRCC. PBRM1 mutations have been associated with high angiogenetic gene expression and appear to be able to predict the clinical bene t of targeted therapy [17]. Meanwhile, there are studies about PBRM1's effect on immune inhibitor clinical therapy. However, other reports in the literature present a contrary conclusion about the function of PBRM1 in immune therapy, and its mechanisms are unclear [18][19][20]. VHL and PBRM1 are both highly mutated genes in ccRCC [3]; however, VHL and PBRM1 mutations in combination exhibit aggressive pathological features in ccRCC, not alone.
In the present study, we found that POLR2A was lost after VHL and PBRM1 mutations. POLR2A, a subunit of the RNA polymerase II complex, has polymerase activity in mRNA synthesis [21]. To the best of our knowledge, POLR2A has been reported to mutation in many human cancers, including human triplenegative breast cancer, hepatocellular carcinoma, and human colorectal cancers [22][23][24][25]. Notably, the loss of POLR2A leads to a decreased abundance of RNA polymerase II in TP53 − /POLR2A − tumor cells, and TP53 − /POLR2A − tumor cells are more sensitive to α-amanitin (POLR2A is speci cally inhibited by αamanitin) than unaltered normal cells [25]. However, the role of POLR2A in RCC is not well known. The role of POLR2A was analyzed in our study, and we found that POLR2A may be a potential target for the treatment of ccRCC growth.

Patients & clinical samples
Transcriptome (count and FPKM value) information and mutational data of 539 ccRCC patients were downloaded from The Cancer Genome Atlas (TCGA, https://www.cancer.gov/tcga) database. Variance stabilizing transformation (VST) was used to normalize the data sets, and then, low-value genes were removed using heterogeneity analysis. According to the VHL gene, the wild-type VHL (VHL WT ) and mutant VHL (VHL MT ) patient groups were set, and then, by using the edgeR and clusterPro ler packages, gene set enrichment analysis (GSEA) was performed to determine signi cantly altered pathways between POLR2A Low and POLR2A high groups. MTT assay A total of 4000 cells were plated in 96-well plates in triplicate for 48 h. Cell viability was quanti ed using 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2-H-tetrazolium bromide (MTT). The growth rate was calculated as follows: average OD value in the POLR2A knockdown cell group/average OD value in the control group ×100%.
Clone formation assay A total of 1000 cells were plated in 6-well plates in triplicate. After incubation for 7 days, cell colonies were visualized by crystal violet (0.5% m/v), and then, images were captured under a microscope.

Animal experiments
These experiment were approved by the institutional review board of the First A iated Hospital of Xi'an Jiaotong University. We randomly separated 8 BALB/c nude mice (4 weeks, male) into 2 groups. These nude mice were subcutaneously injected with 3×10 6 cells (786-O control or shPOLR2A) into the left or right shoulder. tumor size and body weight were measured every 5 days for 25 days. Tumor volume was calculated using the following equation: tumor volume=length×width×height×0.523. At the end of the experiment, the animals were all euthanized, and tumor tissues were surgically excised from the nude mice.

Statistical analysis
Each experiment was repeated three times. Differences between two groups (Student's t-test) were compared by GraphPad Prism software (Version 6.0 software, GraphPad, USA), and data are shown as the mean ± SD with error bars (SEM). P<0.05 was considered signi cant in our study.

POLR2A is the hub gene after VHL and PBRM1 mutations in ccRCC
To search for hub genes or representative biomarkers related to VHL and PBRM1 mutations, we analyzed ccRCC TCGA data. Three hundred ninety-seven VHL WT patients and 155 VHL MT patients, 400 PBRM1 WT patients and 135 PBRM1 MT patients were included. Samples with no mutation or RNA-seq data from TCGA were not included in our analysis. Comparing the VHL WT and VHL MT ccRCC groups, we identi ed the top 100 differentially expressed genes in TCGA ( Figure 1A, supplementary table I). The same analysis was performed between PBRM1 WT and PBRM1 MT patients ( Figure 1A, supplementary table I).
Next, the hub proteins among the top 200 differentially expressed genes were used to construct a PPI network using STRING ( Figure 1B). POLR2A and POLR2E proteins were the hub protein in PPI results identi ed (supplementary Table II). POLR2A and POLR2E both encode RNA II polymerase, and this polymerase is responsible for synthesizing messenger RNA in eukaryotes. To the best of our knowledge, POLR2A is reported to be mutated in most tumors. However, the role of POLR2A in RCC requires investigation, and we analyzed the role of the POLR2A gene in subsequent experiments.
Pathway enrichment analysis for POLR2A in ccRCC To investigate the potential biological processes of POLR2A involved in RCC, We found that 44/50 distribution curves to be "bumpy" in POLR2A expression high-group enrichment, whereas, 6 signaling pathways were enrichment in POLR2A low expression group (Figure 2). Furthermore, we found the gene sets of "HALLMARK_MITOTIC_SPINDLE (Figure 2A)," "HALLMARK_G2M_CHECKPOINT ( Figure 2B)," and "HALLMARK_E2F_TARGET ( Figure 2C)" were all responsible for the high expression POLR2A in progress biological behavior of ccRCC, three tumor cell cycle control pathways. "HALLMARK_TGF_BETA_SIGNALING ( Figure 2D)", "HALLMARK_MYC_TARGETS_V2 ( Figure 2E)" and "HALLMARK_PI3K_AKT_MTOR_SIGNALING ( Figure 2F)" were also found existed in high POLR2A expression group, three overactivated pathways in most tumors. Figure 2 results revealed that high POLR2A was associated with increased tumor cell proliferation in ccRCC malignant progression.

POLR2A knockdown reduced RCC tumor cell growth in vitro
To con rm the cell cycle control of POLR2A on tumor cells in RCC, the 786-O and 769-P RCC cell lines with POLR2A knockdown were constructed ( Figure 3A, 3B). Figure 3A shows POLR2A mRNA and protein expression after POLR2A knockdown for 48 h. The MTT assay revealed that knocking down POLR2A in 786-O and 769-P cells led to cell growth inhibition in the two cell lines compared with the negative control ( Figure 3C, D). Furthermore, colony formation assays were performed using 786-O and 769-P cells. Consistently, as shown in Figure 3E, knocking down POLR2A dramatically decreased the number of colonies in 786-O and 769-P cells, indicating that the colony formation capacity of ccRCC cells was inhibited by knocking down POLR2A in tumor cells.
Proliferating cell nuclear antigen (PCNA), a proliferating cell nuclear antigen, plays an important role in the initiation of cell proliferation and is an indicator of cell proliferation. As shown in Figure 3F, PCNA was reduced after knocking down POLR2A in 786-O and 769-P cells. Consistently, the cyclin D1 and CDK4 proteins were decreased after knocking down POLR2A in 786-O and 769-P cells. The main function of cyclin D1 is to promote cell proliferation. Cyclin D1 binds to and activates CDK4, a cyclin-dependent kinase that is unique to the G1 period. These results suggest that POLR2A may lead to reduced expression of cyclin D1 and G1 arrest in renal cancer cells.

POLR2A knockdown crippled tumor growth in vivo
The cell cycle control effect of POLR2A in vivo was validated in the 786-O xenograft model. 8 nude mice were subcutaneously injected with 3 × 10 6 cultured 786-O (4 nude mice with control and 4 nude mice with shPOLR2A) cells. Figure 4A represents the whole animal experiment design. Figure 4B showed tumor tissues were surgically excised from the nude mice. We weighted two groups tumor weight and analyzed, tumor weight was signi cantly decreased in 786-O shPOLR2A group ( Figure 4C, P<0.01). In the whole experiment, mice body weight in two groups were also increased slowly, and no difference in 786-O control group and 786-O shPOLR2A group ( Figure 4D). Furthermore, we calculated tumor volume change between two group, the tumor volume decreased signi cantly in 786-O shPOLR2A group comparing with 786-O control group (Figure 4E, P<0.001).

Page 7/15
The above results suggested that silenced POLR2A can inhibit RCC tumor cell growth in vivo.

Discussion
In this study, we focused on POLR2A, a well-known mutation gene in human cancers [21,22,24]. VHL and PBRM1 had mutation rates of up to approximately 85% and 40% in ccRCC [3], respectively. POLR2A was the most highly dysregulated gene, with a high betweenness value (PPI) and high closeness value (PPI), after VHL and PBRM1 mutations in the top 100 differentially expressed genes in our analyzed results (Figure 1). We further characterized the bio-function of POLR2A, and our results showed that silencing POLR2A inhibited tumor cell growth in vitro and in vivo (Figure 3 and 4).
To the best of our knowledge, the role of POLR2A in ccRCC is poorly understood. It has been reported that POLR2A is essential for cell survival. Clark et al found that POLR2A alterations led to the dysregulation of key identity genes in meningeal development [22]. Consistent with the function of POLR2A, according to our results, POLR2A led to the dysregulation of key identity genes after VHL or PBRM1 mutations in ccRCC (Figure 1), and our results showed that tumor cell related pathways: E2F, G2M, mTOR1, and MYC signaling activated pathways were all in response to POLR2A high expression ( Figure 2). VHL, PBRM1, BAP1, SETD2, KDM5C, PIK3CA, PTEN, MTOR and TP53 are all recurrent mutations in ccRCC [27], and Sabine et al. has reported that VHL, TP53 and Rb1 deletion causes ccRCC in mice and allows the progression of ccRCC in mice [28]. Importantly, POLR2A, a neighboring gene of TP53 (located within 200 kb), has been reported to effect TP53 gene in many human cancers [23,29,30]. According to the literature, suppression of POLR2A inhibits human colorectal carcinoma cells and is associated with the loss of TP53, proliferation and tumorigenicity in a p53-independent manner [21]. Meanwhile, colorectal cancer cells with hemizygous loss of POLR2A were more vulnerable to POLR2A inhibition than unaltered normal cells [21].
Cyclin D-CDK4/6 accumulation allows entry into the cell cycle, otherwise preventing cell cycle exit [31].
Cyclin D1, CDK4, and PCNA were detected in 786-O and 769-P cells, decreased cyclin D1, CDK4, and PCNA protein amount were observed in shPOLR2A RCC cells ( Figure 3E). Colony formation and xenograft animal results further support the inhibitory effect of POLR2A on the growth of RCC cells ( Figure 3C, D, and Figure 5).
Indeed, α-amanitin, which speci cally represses POLR2A, has been reported to selectively inhibit the proliferation, survival and tumorigenic potential of tumor cells [32]. Furthermore, the toxicity of amanitin can be overcome by the conjugation of α-amanitin with monoclonal antibodies (cell surface marker) or by coating it with nanoparticles, as reported by Yunhua. et al. and Jiangsheng. et al., respectively [23,24].
Due to cyclin POLR2A oncogenic functions in RCC, these ndings suggest that targeting POLR2A could regulated tumor cell cycle exist or entrance.

Conclusions
In summary, our study revealed that POLR2A was the key protein stimulating RCC malignancy after VHL and PBRM1 mutations, and demonstrated that silenced POLR2A crippled cell viability and proliferation in vivo and in vitro. We anticipate that speci c target inhibition of POLR2A will be a therapeutic approach for treating human cancers.

Availability of data
The data used to support the ndings of this study are included within the article and the supplementary information les (Table and Table ).
Ethics approval and consent to participate Not applicable.

Consent for publication
All the data generated or analyzed in this study are included in this article. Other data that are relevant to this article are available from the corresponding author upon reasonable request. Figure 1 POLR2A is a hub gene following VHL and PBRM1 mutations in ccRCC. A Heat map of the top 100 signi cantly changed genes in patients with VHL and PBRM1 mutations compared to wild-type patients. Red indicates upregulated genes, and blue indicates downregulated genes. B The coexpression network showing the top 100 signi cantly changed genes in patients with VHL and PBRM1 mutations compared to wild-type patients. The network was constructed by STRING to investigate the PPI network.   Knocking down POLR2A inhibits tumor growth in vivo. A Protocol for bearing BALB/c nude mice treatment in this study. B Tumor images from control and shPOLR2A groups. C Tumor weight of xenografted mice in control and shPOLR2A groups. D Body weight changes of the xenografted mice in control and shPOLR2A groups. E Tumor volume changes in the xenografted mice during control and shPOLR2A groups. Statistical analysis of the data was calculated using student's t test between two groups, error bars indicate±SD, n=4. **p<0.01 as compared with the control group, ***p<0.001 as compared with the control group, ns no sign icant.