Clinical information, somatic mutation and gene expression data of patients with pancreatic cancer were obtained from two previously published cancer genome studies (11,12). The somatic mutation and clinical information from 483 patients with pancreatic cancer was obtained from data published by Bailey et al (11) and Waddell et al (12). The survival time of 159 patients of the total 483 patients was available, with survival status indicated as ‘alive’, ‘deceased’ or ‘loss’ at the final follow-up. Gene expression data of 96 patients were obtained from the Gene Expression Omnibus (13) (dataset, GSE36924) (11), while 96 patients have survival information. The samples with follow-up information were used for survival analysis.

Survival analysis of genes mutated in ≥10 (out of 159) patients was performed. Patients were divided into two groups according to the mutation status of each gene. The difference in overall survival between the two groups was estimated using a Kaplan-Meier curve and tested for significance using a log-rank test. A gene was considered to be associated with patient clinical outcomes when a P<0.05 value was obtained. Following this, patients were additionally divided into two groups, based on mutation and gene expression levels, according to the median expression values of each gene. Survival analysis was performed using gene-expression data obtained from the Gene Expression Omnibus (dataset, GSE36924) (11). It was identified that somatic mutation and gene expression of TRPM2 were significantly associated with the patient's overall survival (Tables I and II). Thus, further experimental assays were performed to investigate the role of TRPM2 in pancreatic cancer.

It was hypothesized that genes involved in a similar biological process may exhibit similar expression patterns. Gene expression correlation analysis was performed to reveal genes functionally associated with TRPM2 in patients with pancreatic cancer. Pearson's correlation coefficient between TRPM2 and each of the genes of interest was calculated. Following this, the top genes with a correlation coefficient of >0.50 were retained (Table III).

PANC-1 cells (American Type Culture Collection, Manassas, VA, USA) were seeded in 6-well plates until 80% confluency was reached following 24 h of culture at 37°C. Hilymax-Trpm2 siRNA complex (20 pmol/l), Hilymax-Trpm2 OverExp vector (4 µg/well), empty vector and scramble siRNA (Generay Biotech Co., Ltd., Shanghai, China) were added to the cell culture, using the transfection agent Hilymax (Dojindo Molecular Technologies, Inc., Kumamoto, Japan) well as per the manufacturer's instructions, and incubated at 37°C in a 5% CO2 incubator for 4 h. The medium was subsequently replaced with fresh medium (RPMI 1640; 10% fetal bovine serum, Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) and incubated for a further 48 h. The untransfected group was subjected to normal culture without transfection reagent, the empty vector represented the control for TRPM2 overexpression, and scramble siRNA represented the control for TRPM2 silencing. Cells were prepared for the subsequent reverse transcription-quantitative polymerase chain reaction (RT-qPCR) and Transwell assays. The Trpm2 siRNA and Trpm2 OverExp plasmid were designed and synthesized by Generay Biotech Co., Ltd. (Shanghai, China; Table IV).

Total RNA from cells was extracted using Cell Culture and Tissue Total RNA Extraction and Preparation Mini kit (Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. The quantity and quality of RNA were confirmed using a NanoDrop 1000. The primers were designed using Primer Premier 5.0 software (Premier Biosoft International, Palo Alto, CA, USA) and synthesized by Generay Biotech Co., Ltd. (Table IV). RT-qPCR was performed using the KAPA SYBR Green Supermix PCR kit with the AriaMx Real-Time PCR System (both Agilent Technologies, Inc., Santa Clara, CA, USA). RT was performed at 50°C for 30 min. qPCR conditions: Denaturing at 95°C, 10 min, then 40 cycles of denaturing at 95°C for 15 sec, annealing at 60°C for 1 min, and elongation at 95°C for 15 sec and 60°C for 15 sec. The relative expression levels among the different genes were determined using the 2^−ΔΔCq method (14).

PANC-1 cell proliferation was investigated using the Cell Counting Kit-8 (CCK-8; Dojindo Molecular Technologies, Inc.). PANC-1 cells were seeded onto a 96-well microplate at a density of 3×104 cells/well, and the cells were transfected with either Hilymax-Trpm2 OverExp vector (0.2 µg/well), Hilymax-Trpm2 siRNA (20 pmol/l), empty vector (0.2 µg/well) and scramble siRNA (20 pmol/l). The cells were subsequently cultured for 0, 24 and 48 h. Following this, 5 µl CCK-8 solution was added to each well and incubated at 37°C for a further 2 h. Optical density was determined at 450 nm using a microplate spectrophotometer (BioTek Instruments, Inc., Winooski, VT, USA).

PANC-1 cells were seeded in 24-well plates at a density of 1×105 cells per well and incubated overnight at 37°C. Following this, the cells were transfected with Hilymax-Trpm2 OverExp vector (1 µg/well), Hilymax-Trpm2 siRNA (20 pmol/l), empty vector and scramble siRNA. The cells were subsequently cultured for 24 h. The cells at the bottom of wells were then scratched using a 10 µl tip and the floating cells were gently washed twice using RPMI 1640 medium. Images of each well were captured using a microscope (magnification, ×100) at 0 and 24 h time intervals post-injury (Olympus X51; Olympus Corporation, Tokyo, Japan).

PANC-1 cells were transfected with TRPM2 Vector, TRPM2 siRNA, empty vector and scramble siRNA. A total of 24 h post-transfection, the cells were seeded into Matrigel-plated upper wells with RPMI 1640 medium without serum (5×104 cells/well), and the lower wells contained 500 µl complete medium (RPMI 1640 and 10% fetal bovine serum), following a routine procedure. Following incubation for 48 h at 37°C, cells that did not migrate through the pores were gently removed with a cotton swab. Cells on the lower side of the insert filter were fixed by 5% glutaraldehyde for 10 min and stained with 1% crystal violet in 2% ethanol at room temperature for 20 min. Numbers of cells on the underside of the filter from five randomly selected microscopic views (magnification, ×100) were counted.

The difference of overall survival between the two groups was estimated using Kaplan-Meier curve and tested for significance using log-rank test by SPSS 17.0 software (SPSS, Inc., Chicago, IL, USA). GraphPad Prism 5 (GraphPad Software, USA) was used to perform statistical analysis for the results of RT-qPCR, CCK-8, wound healing and Transwell assays. Significance between groups was evaluated by one way analysis of variance followed by a Bonferroni post hoc test. P<0.05 was considered statistically significant.

Following the data analysis of genes using the Kaplan-Meier curve and the log-rank test, TRPM2 was selected as a potential candidate marker of prognosis. The survival analysis revealed that in 159 patients, the mutation status of TRPM2 was significantly correlated with patient survival (P=1.0416×10-2; Table I). The somatic mutations of TRPM2 in this cohort are detailed in Table II. Furthermore, the enhanced expression level affected survival among 91 patients (P=4.2253×10-2; Fig. 1). Gene expression correlation analysis revealed that TRPM2 was strongly correlated with the expression of probable phospholipid-transporting ATPase IM (ATP8B4), γ-parvin (PARVG), tudor domain containing 9 (TDRD9), Toll-like receptor (TLR7) and Scm-like with four MBT domains protein 2 (SFMBT2) (Table III).

RT-qPCR analysis demonstrated that compared with the empty vector group, TRPM2 was successfully overexpressed in the Hilymax-TRPM2 OverExp group (P<0.05; Fig. 2). In addition, the expression level of TRPM2 was successfully knocked down in the Hilymax-TRPM2 siRNA group compared with the scramble siRNA group (P<0.05; Fig. 2).

The results of the CCK-8 assay demonstrated that TRPM2 successfully enhanced PANC-1 cell proliferation in the Hilymax-Trpm2 OverExp group compared with the empty vector group (P<0.05; Fig. 3) at the 48-h time interval, whereas the Hilymax-Trpm2 siRNA group exhibited an inhibitory effect compared with the scramble siRNA group (P<0.05; Fig. 3). There were no significant differences between the vector group and the empty vector group at the 24 h time interval, however there was a significant difference between the TRPM2 siRNA group compared with the scramble siRNA group (Fig. 3).

The results of the scratch wound-healing assay revealed that the migratory ability of PANC-1 cells was significantly enhanced following overexpression of TRPM2 (P<0.01; Fig. 4), however, this was significantly suppressed in the RNA interfering group (P<0.001; Fig. 4).

Invasive ability was determined via Transwell assays using Matrigel, and the results demonstrated a significantly enhanced invasive ability in the OverExp group compared with the empty vector group (P<0.01; Fig. 5). The siRNA group exhibited fewer invaded cells in the lower chamber of the microwell (P<0.01; Fig. 5).