The expression and role of TRPV2 in esophageal squamous cell carcinoma

Background: Transient receptor potential vanilloid 2 (TRPV2) was recently shown to be involved in migrant potentials. The present study aimed to investigate its role in esophageal squamous cell carcinoma (ESCC). Methods: Knockdown experiments were conducted using TRPV2 siRNA in human ESCC cell lines, and anti-tumor effects were analyzed. The gene expression profiles of cells were analyzed using a microarray method. An immunohistochemical staining was performed on 62 primary tumor samples. Results: TRPV2 overexpression was observed in TE15 and KYSE170 cells. TRPV2 depletion suppressed proliferation, cell cycle progression, and invasion/migration ability, and induced apoptosis. A pathway analysis of microarray data showed that TRPV2 depletion down-regulated WNT/β-catenin signaling-related genes and basal cell carcinoma signaling-related genes. The suppression of tumor functions, such as proliferation, invasion, and angiogenesis, was predicted in the ontology analysis. Immunohistochemical analysis revealed a correlation between strong TRPV2 expression and a poor prognosis in ESCC patients. Conclusion: The present results suggest that TRPV2 regulates cancer progression by affecting WNT/β-catenin or basal cell carcinoma signaling, and that TRPV2 strong expression is associated with a worse prognosis in ESCC patients. These results provide an insight into the role of TRPV2 as a novel therapeutic target or biomarker for ESCC.

Transient receptor potential vanilloid (TRPV) channels have been identified as one of the molecules activated by capsaicin, a vanilloid-like molecule. TRPV1 was initially reported as a membrane protein triggered by heat or pain stimuli, VR1 in 1997 1 . Five additional molecules of the subfamily were subsequently cloned and named TRPV2, TRPV3, TRPV4, TRPV5, and TRPV6 [2][3][4][5][6] . Although all members of the TRPV subfamily were initially presumed to have similar functions, such as heat sensors, extensive physiological studies later revealed that TRPV2-6 did not all respond to temperature stimuli. Moreover, TRPV1-4 channels have non-selective cation-conducting pores, while the pores of TRPV5 and TRPV6 are highly calcium selective 7 . Among members of the TRPV subfamily, TRPV2 is expressed in many organs, such as the digestive tract, pancreas, and liver; furthermore, TRPV2 has been implicated in bowel inflammation, intestinal peristalsis, and the autocrine effects of insulin on pancreatic β-cells [8][9][10] . Therefore, TRPV2 may be a therapeutic target of digestive diseases.
Previous studies reported that TRPV2 was involved in cancer progression, migration, and invasion as well as in the therapeutic effects of anticancer drugs. Gambade et al. showed that the activation of TPRV2 regulated cancer migration via calcium entry 11 . Similarly, the involvement of TRPV2 in migration and invasion was demonstrated in bladder and prostate cancers 12,13 . TRPV2 was identified as a poor prognostic marker in gastric cancer and, in contrast, a good prognostic marker in breast cancer 14,15 . Liberati et al. interpreted this discrepancy as TRPV2 playing a major role in initial events that parallel the transformation of a single normal stem/ Figure 1. TRPV2 controls the proliferation and cell cycle progression of ESCC cells, (A) TRPV2 mRNA expression was analyzed in 7 ESCC cell lines. Quantitative RT-PCR showed that TRPV2 was strongly expressed in TE15, and more strongly expressed in KYSE170 than in the other ESCC cell lines. (B) TRPV2 protein expression of ESCC cell lines was evaluated by western blotting. Upper panel showed full-length gels of TRPV2, and lower panel indicated the cropped image of same gels for evaluating β-actin expression (full-length image: Supplementary Fig. 3). TRPV2 expression in TE15 and KYSE170 was similar in Western blotting. (C) TRPV2 expression of ESCC cell lines transfected with control and TRPV2 siRNA was evaluated by quantitative RT-PCR and western blotting. The blotting figure showed the cropped image of TRPV2 or β-Actin bands on same gels (full-length image: Supplementary Fig. 4). TRPV2 siRNA effectively reduced TRPV2 mRNA and protein levels in TE15 and KYSE170 cells. Mean ± SEM. n = 3. *p < 0.05 (significantly different from control siRNA). (D) The down-regulation of TRPV2 inhibited the proliferation of TE15 and KYSE170 cells. The number of cells was counted 48 and 72 h after siRNA transfection. Mean ± SEM. n = 4. *p < 0.05 (significantly different from control siRNA). (E) The down-regulation of TRPV2 partially reduced cell cycle progression from the G1 to S phase in TE15 and KYSE170 cells. The cell population of sub-G1 increased in TPRV2-depleted TE15 and KYSE170 cells. Cells transfected with control or NIS siRNA were stained with propidium iodide (PI) and analyzed by flow cytometry. Mean ± SEM. n = 3. *p < 0.05 (significantly different from control siRNA).

Analysis of cell migration and invasion.
The migration assay was conducted using a Cell Culture Insert with a pore size of 8 μm (BD Biosciences, Bedford, MA, USA) by our previously reported protocols 20 . Biocoat Matrigel (BD Biosciences) was used to evaluate cell invasion potential. Cells (TE15: 3.25 × 10 5 cells per well, KYSE170: 0.6 × 10 5 cells per well) were seeded in the upper chamber in serum-free medium 24 h after siRNA transfection. The lower chamber contained medium with 10% FBS. The chambers were incubated at 37 °C for 48 h in 5% CO 2 , and non-migrating or non-invading cells were then removed from the upper side of the membrane by scrubbing with cotton swabs. Migrating or invading cells were fixed on the membrane and stained with Diff-Quick staining reagents (Sysmex, Kobe, Japan). The migrating or invading cells on the lower side of the membrane were counted in four independent fields of view at ×100 magnification for each insert. Each assay was performed in triplicate.
Microarray sample preparation and hybridization. Microarray analysis was conducted using our previously reported protocols 20 . Briefly, total RNA was extracted using a RNeasy kit (Qiagen). RNA quality was monitored with an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA). Cyanine-3 (Cy3)-labeled cRNA was prepared from 0.1 μg of total RNA using a Low Input Quick Amp Labeling Kit (Agilent) in accordance with the manufacturer's instructions. Samples were purified using RNeasy columns (Qiagen). A total of 0.60 μg of Cy3-labeled RNA was fragmented and hybridized to an Agilent SurePrintG3 Human Gene Expression 8 × 60 K ver3.0 Microarray for 17 h. Slides were washed and scanned immediately using an Agilent DNA Microarray Scanner (G2565CA) in the one color scan setting for 8 × 60 K array slides.
Processing of microarray data. Processing of microarray data was conducted using our previously reported protocols 20 . Scanned images were evaluated with Feature Extraction Software 10.10 (Agilent) using default parameters to acquire background-subtracted and spatially detrended processed signal intensities 20 . Signal transduction networks were subsequently analyzed using an Ingenuity Pathway Analysis (IPA) (Ingenuity Systems, Qiagen, Redwood City, CA) 20 .
Patients and primary tissue samples. ESCC human samples were obtained from 62 patients with histologically confirmed primary ESCC undergoing esophagectomy at Kyoto Prefectural University of Medicine between 1999 and 2009. All of them were embedded in paraffin after 12 h of formalin fixation just after the surgery. We excluded patients with non-curative resected tumors, synchronous or metachronous cancers (in addition to ESCC), preoperative chemotherapy, or radiation therapy. All eligible patients provided written informed consent. The data of relevant clinicopathological and survival were obtained from the hospital database. Staging was principally diagnosed based on the International Union Against Cancer/Tumor Node metastasis Classification of Malignant Tumors (7 th edition) 21 .

Diseases or Functions
p value(−log) Predicted activation Activation z-score Number of molecules www.nature.com/scientificreports www.nature.com/scientificreports/ Immunohistochemistry. TRPV2 protein expressions in human ESCC tissues were evaluated by the same immunohistochemical staining protocol as our previous study 20 . Paraffin sections of tumor tissues (thickness of 4 μm) were subjected to immunohistochemical staining using the avidin-biotin-peroxidase method 20 . Briefly, paraffin sections were dewaxed with xylene and hydrated with a graded series of alcohol 20 . Endogenous peroxidases were quenched by incubating the sections for 30 min in 0.3% H 2 O 2 20 . The Avidin/Biotin Blocking Kit (Vector Laboratories, Burlingame, CA) was used for the blocking of endogenous biotin, biotin receptors, and avidin-binding sites 20 . Sections were then treated with a protein blocker and incubated at 4 °C overnight with the primary antibody 20 . The avidin-biotin-peroxidase complex (Vectastain ABC Elite kit; Vector Laboratories, Burlingame, CA) was visualized using diaminobenzidine tetrahydrochloride 20 . Sections were counterstained with hematoxylin 20 . These sections were then dehydrated through a graded series of alcohols, cleared in xylene, and mounted 20 . Immunohistochemical samples stained with TRPV2 were graded based on staining intensities, strong or weak expression, and proportions 20 .
Statistical analysis. Statistical analysis was performed using same method of our previous study 20 . Fisher's exact test was used to evaluate differences between proportions, and the Student's t-test was employed to evaluate continuous variables 20 . Survival curves were constructed using the Kaplan-Meier method, and differences in survival were examined using the Log-rank test 20 . A multivariate analysis of the factors influencing survival was performed using Cox's proportional hazard model 20 . Differences were considered to be significant when the relevant p value was <0.05 20 . These analyses were performed using JMP statistical software (version 12, SAS Institute Inc., Cary, NC) 20 .

TRPV2 protein and mRNA expression in ESCC cell lines. Quantitative RT-PCR and Western blotting
were performed to evaluate TRPV2 expression in the human ESCC cell lines, TE5, TE8, TE9, TE15, KYSE70, LYSE150, and KYSE170 (Fig. 1A,B). TRPV2 mRNA was strongly expressed in TE15, and more strongly expressed in KYSE170 than in the other ESCC cell lines. Meanwhile, similar intensity of TRPV2 protein expression was observed in TE5, TE9, TE15, KYSE70 and KYSE170. siRNA knockdown of TRPV2 suppressed the proliferation of ESCC cells. We conducted knockdown experiments using siRNA in the TRPV2-overexpressing ESCC cell lines, TE15 and KYSE170, and evaluated the effects of TRPV2 depletion on cell proliferation. TRPV2 siRNA effectively decreased TRPV2 mRNA and protein expression in TE15 and KYSE170 (Fig. 1C). Cell proliferation was significantly less in TRPV2-depleted cells than in control-siRNA-treated cells (Fig. 1D).

TRPV2 depletion inhibited cell cycle progression and induced apoptosis in ESCC cells. A cell
cycle analysis was performed to investigate the mechanisms underlying suppressed proliferation induced by TRPV2 depletion in ESCC cells (Fig. 1E). The transfection of TRPV2 siRNA partially inhibited cell cycle progression from the G1 to S phase in TE15 and KYSE170. Moreover, the cell population of sub-G1 increased in TPRV2-depleted cells, suggesting that TRPV2 depletion suppressed the proliferation of ESCC cells by inhibiting cell cycle progression or inducing apoptosis. In order to validate the induction of apoptosis by TRPV2 Ingenuity Canonical Pathways p value (−log) Predicted Activation Activated z-score  www.nature.com/scientificreports www.nature.com/scientificreports/ depletion, an apoptosis assay and Western blotting of the apoptosis markers, caspase 3 and cleaved caspase 3 were conducted. The results obtained revealed that the down-regulation of TRPV2 induced early apoptosis (annexin V-positive/PI negative) and late apoptosis (annexin V-positive/PI positive) in TE15 and KYSE170 72 hours after the siRNA knockdown ( Fig. 2A). Western blotting showed that the expression of cleaved caspase 3 increased in TRPV2-depleted ESCC cells (Fig. 2B).

TRPV2 controlled cell migration and invasion in ESCC cells.
The effects of the down-regulation of TRPV2 on cell migration and invasion in ESCC cells were analyzed, and the results obtained revealed that the TRPV2 siRNA knockdown inhibited cell migration and invasion (Fig. 2C). These results indicated that TRPV2 plays an important role in the cancer cell migration and invasion of ESCC.
Gene expression profiles of TRPV2-depleted ESCC cells. The gene expression profile of TRPV2-depleted KYSE170 cells was analyzed using microarrays and a bioinformatic study. Microarray results showed that the expression of 3697 genes displayed a fold change of >2.0 in KYSE170 cells upon the depletion of TRPV2. The number of up-regulated genes was 1372, while that of down-regulated genes was 2325. A list of 20 genes with expression levels that were the most strongly up-or down-regulated in TRPV2-depleted KYSE170 cells was shown in Supplementary Tables 1 and 2. The Ca 2+ -related genes, MICU1 and CACNB2, were strongly The signaling map, which was generated using an Ingenuity Pathway Analysis (IPA) software (Ingenuity Systems, Qiagen, Redwood City, CA), of "WNT/β-Catenin", the top-ranking canonical pathway related to TRPV2 depletion according to an IPA analysis. Red and green indicate genes with expression levels that were higher or lower, respectively, than reference RNA levels. (B) The signaling map, which was generated by IPA software, of "Basal cell carcinoma signaling" (C) Verification of gene expression by real-time quantitative RT-PCR. The expression levels of four selected "WNT/ TGF-βCatenin" and "Basal cell carcinoma signaling"-related genes (WNT10A, TGF-β2, TGF-β2R, and GLI) in TRPV2-depleted TE15 and KYSE170 were compared to those in control siRNA-transfected cells using real-time quantitative RT-PCR. Mean ± SEM. n = 3. *p < 0.05 (significantly different from control siRNA). www.nature.com/scientificreports www.nature.com/scientificreports/ down-regulated in TRPV2-depleted KYSE170 cells. Furthermore, a gene ontology analysis (IPA) showed that the function of "the invasion of cells", "migration of cells", "migration of tumor cell lines", "invasion of tumor cell lines", "cell proliferation of tumor cell lines", "apoptosis", and "cell proliferation of tumor cell lines" were down-regulated (activation z-score, −3.156 to −5.912) ( Table 1). These results were consistent with those of the cancer function, proliferation, cell cycle, apoptosis, and migration/invasion assays.  Table 2. "WNT/β-Catenin signaling", "basal cell Carcinoma signaling", and "regulation of the Epithelial-Mesenchymal Transition Pathway" were the top-ranking canonical pathways related to TRPV2 depletion, while "basal cell carcinoma signaling" and "WNT/β-Catenin signaling" were predicted to be down-regulated pathways (activation z-score: −1.886 and −1.441) (Fig. 3A,B and Supplementary Fig. 1A). Moreover, the pathways associated with stemness, "Human Embryonic Stem Cell Pluripotency", "Role of NANOG in Mammalian Embryonic Stem Cell Pluripotency", and "Wnt/β-catenin Signaling", were the top ranked in the canonical pathway.   Table 4. Prognostic factors of esophageal squamous cell carcinoma according to univariate and multivariate analyses. pT: pathological tumor invasion depth, pN: pathological lymph node metastasis.
In order to confirm the results of the microarray analysis, the expression of several genes included in "WNT/β-Catenin signaling", "basal cell carcinoma signaling", "regulation of the Epithelial-Mesenchymal Transition Pathway", and the relationship between stem cell-related genes and TRPV2 expression were validated in more detail using quantitative RT-PCR. WNT10A, TGFβ2, TGFβR2, GLI, Snai1, Zeb2, CDH2, CD44, and SOX2 mRNA expression levels were lower in TRPV2-depleted TE5 or KYSE170 than in control siRNA-transfected cells (Fig. 3C, Supplementary Figs 1B and 2). These results suggesting that TRPV2 regulated cancer cell function via "WNT/β-Catenin signaling", "basal cell Carcinoma signaling", "regulation of the Epithelial-Mesenchymal Transition Pathway" and the expression of stem cell-related genes.
TRPV2 protein expression in human ESCC samples. In order to evaluate the significance of TRPV2 protein expression in ESCC samples, we performed immunohistochemical staining on 62 tumor samples of Human ESCC using the TRPV2 antibody. Nuclear TRPV2 expression was detected in the middle layer of the non-cancerous stratified squamous epithelium. (Fig. 4A). Although TRPV2 cytoplasmic expression in carcinoma cells varied widely between each ESCC sample, we initially evaluated TRPV2 signal intensities in cancer cells as no, weak, and strong expression (Fig. 4B-D); the proportion of each intensity in tumors was subsequently measured. ESCC samples in which the proportion of strong intensity was 20% or larger were defined as high expression (n = 40), and the others as low expression (n = 22). The relationships between TRPV2 protein expression and patient backgrounds and various clinicopathological parameters were analyzed (Table 3). Tumor sizes were slightly larger in the high group than in the low group. No significant difference was observed in recurrence rate; however, distant recurrence was tended to occur more frequently in TRPV2 high group. Furthermore, the number of cases who underwent radical treatment for recurrence, such as chemoradiotherapy or surgery, was tended to be larger in TRPV2 low group (50.0% vs 38.8%).
The relationship between the expression of TRPV2 and prognosis of ESCC patients was investigated. The 5-year overall survival rate after surgery was significantly worse in the high group than in the low group (59.5% vs 85.2%, p = 0.020) (Fig. 4E). Moreover, the 5-year survival of ESCC patients after post-operative recurrence in the high group was significantly worse than in the low group (46.6% vs 0.0%, p = 0.006) (Fig. 4F). A univariate analysis of survival (5-year overall survival) showed that pathological venous invasion, pathological lymph node metastasis, and the strong expression of TRPV2 were significant. A multivariate analysis with these 3 factors (Cox's proportional hazard model) revealed that the strong expression of TPRV2 was an independent poor prognostic factor (Table 4).

Discussion
A role for TRPV2 in cellular development or morphology was recently reported. Kojima et al. showed that TRPV2 was associated with cell cycle progression via the regulation of its translocation induced by Insulin-Like Growth Factor 1 22 . TRPV2 has been shown to play a role in cellular migration through the regulation of intracellular Ca 2+ concentrations 11 . In the field of oncology, many researchers reported that TRPV2 similarly regulated cell death in cancer cells or cancer migration/invasion 13,15,16,18,23 . They showed that the regulation of Ca 2+ signaling by TRPV2 may affect these cancer functions. Ca 2+ is an essential element for the survival and function of cells. Amplifications in the magnitude and duration of intracellular Ca 2 changes may mean the difference between cellular migration and cell death. In malignant cells, calcium signaling plays important roles in proliferation, apoptosis, tumor stromal interaction, metastasis, and drug resistance 24,25 . In the present study, TRPV2 expression was firstly evaluated, and TRPV2 knockdown experiment was subsequently performed. Although TRPV2 expression in ESCC cell lines was observed, the discrepancy existed between the protein and mRNA expression. Zhang et al. described that the intensity of protein expression was not consistent with mRNA expression in over two-third of molecules which expressed in human colorectal cancer specimens 26 . TRPV2 may be one of the molecules with the inconsistency between gene and protein expression. Knockdown experiments demonstrated that TRPV2 depletion suppressed tumor proliferation, cell cycle progression, and migration/invasion, and also induced apoptosis in ESCC cells (Figs 1 and 2). Moreover, the gene ontology analysis revealed that cancer functions, such as cell invasion, angiogenesis, cell migration, cell proliferation, and apoptosis, were down-regulated in TRPV2-depleted ESCC cells (Table 1). These results were consistent with the previously reported antitumor effects induced by the regulation of Ca 2+ signaling. Therefore, it is plausible that TRPV2 regulates cancer biology via calcium signaling in ESCC.
Furthermore, we performed a pathway analysis to clarify the role of TRPV2 in the cancer signaling of ESCC, and revealed that the depletion of TRPV2 down-regulated "basal cell carcinoma signaling". "Basal cell carcinoma signaling" is a pathway related to proliferation or apoptosis in basal cell carcinoma, in which cross talk between the hedgehog pathway and Wnt/β-Catenin signaling activates several cancer functions 27,28 . The involvement of the hedgehog pathway in ESCC was previously reported in our laboratory 29 . The present results indicated that TRPV2 regulated malignant potentials via cross talk between the hedgehog pathway and Wnt/β-catenin signaling; furthermore, Ca 2+ may act as a second messenger between TRPV2 expression and these pathways. Previous studies revealed that intracellular Ca 2+ plays an important role in the WNT pathway (WNT/calcium pathway) 30,31 . In this pathway, intracellular Ca 2+ act as a second messenger, resulting in the control of cancer-related gene expression. These results and previous findings suggested that TRPV2 controls WNT/β catenin signaling and basal cell carcinoma signaling (cross talk between the hedgehog and WNT pathways) via the regulation of Ca 2+ signals, such as WNT/calcium signaling.
TRPV2 depletion also down-regulated "Wnt/β-catenin signaling" in the pathway analysis, which regulated pluripotency via the translocation of β-catenin into the nucleus. The relationship between this pathway and cancer stem cells has already been reported 32,33 . In the microarray data obtained in the present study, TRPV2 depletion down-regulated the expression of the stem cell markers SOX2 and CD44. Furthermore, the top-ranked pathway contained the stemness-related signals "Human Embryonic Stem Cell Pluripotency" and "Role of NANOG in Mammalian Embryonic Stem Cell Pluripotency". The validation of gene alterations using RT-PCR www.nature.com/scientificreports www.nature.com/scientificreports/ revealed that CD44 and SOX2 were down-regulated in TRPV2-depleted ESCC cells. We previously reported the overexpression of TRPV2 in cancer stem cells derived from ESCC cell lines, and the present results from the pathway analysis were consistent with these findings. Therefore, TRPV2 may also maintain the stemness of ESCC stem cells via these pathways.
TRPV2 depletion down-regulated the gene expression of TGFβ and TGFβR in "Wnt/β-catenin signaling" in contrast to our exceptions. These results may be explained by the TGFβ paradox. TGFβ plays contrasting roles in tumors, acting as a tumor suppressive gene during the first stages of carcinogenesis and as a tumor promoter during the advanced stages of progression 34,35 . Therefore, the paradoxical results of TGFβ and TGFβR were consistent with our hypothesis.
Moreover, the pathway analysis revealed that the TRPV2 knockdown altered the expression of many genes associated with epithelial-mesenchymal transition ( Table 2, rank 2). The validation of gene alterations using RT-PCR revealed that TGFβ, WNT, and SNAIL were down-regulated in TRPV2-depleted ESCC cells. These results suggested that TRPV2 controls the migration/invasion ability of ESCC via epithelial-mesenchymal transition.
Finally, we performed immunohistochemical analysis in order to evaluate clinical significance of TRPV2 expression in ESCC samples. The results revealed that TRPV2 expression was significantly associated with overall survival; however, significant correlations between clinicopathological features and TRPV2 expression were not observed. To clarify the reason of bad prognosis in TRPV2 high group, recurrent patterns and survival after recurrence were analyzed, revealing that distant recurrence were more frequently observed in TRPV2 high group and the prognosis after recurrence in TRPV2 high group was significantly worse than in low group. These findings suggested that TRPV2 affected recurrence pattern or therapeutic effect of therapy for recurrence.
In summary, we herein demonstrated that TRPV2 played a role in the proliferation, cell cycle progression, survival, migration, and invasion of ESCC cells. Microarray results showed that TRPV2 markedly affected the expression of genes related to WNT/β-catenin signaling, basal cell carcinoma signaling (cross talk between the hedgehog and WNT/β-catenin pathways), and regulation of the epithelial-mesenchymal transition pathway. The results of the immunohistochemical analysis of ESCC human samples revealed that the strong expression of TRPV2 was a poor prognostic factor in patients with ESCC. Although further investigations are needed, the present results demonstrate that TRPV2 has potential as a poor prognostic biomarker and novel therapeutic target for ESCC.

Data availability
The data sets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.