The Role And Mechanism Of PENK Expression In Gastrointestinal Stromal Tumor

OBJECTIVES: The relationship was investigated between the role of PENK expression in gastrointestinal stromal tumor and the malignant behavior of the tumor through the analysis of widely used biological database and the analysis of clinical GIST tissues validation combined with GIST cell lines function test in order to clarify the association between PENK gene expression and clinical risk stratification in GIST,and also the study was aimed to elucidate the mechanism of PENK in gastrointestinal stromal tumors and the function in GIST-T1 cell secondary to imatinib mesylate.METHODS: 1)PENK expression was analyzed between the GIST tumor tissue samples and the matched normal tissue samples through the GEO datasets,the Oncomine online datasets(https://www.oncomine.org)and the c Bio Portal online datasets(http://www.cbioportal.org).A set of different expression genes were screened from the datasets by Fc(Fold change)≥2 and CNA≥5% combined with the results of q RT-PCR of different risk stratification of GIST fresh tissue specimens(10 cases of high risk,5 cases of median risk,5 cases of low risk and 14 normal gastric tissues(10cases paired with high risk GISTs,2 cases paired with low risk GISTs,2 cases paired with median risk GISTs)from the department of gastrointestinal surgery of Ren Ji Hospital.The gene PENK was selected because of its decreasing expression level combined with increasing NIH GIST stratification risk.At the same time,the results of unsupervised hierarchical clustering analysis of risk-related differential genes was studied in GIST from Ren Ji Hospital,School of Medicine Shanghai Jiaotong University,and we found that PENK was highly expressed in low to median risk GISTs,but low expression in high risk GISTs.And so,the gene PENK was selected as our research object.2)The tissue microarray of GIST patients(2002.01.01 ～ 2011.12.31)from Renji Hospital,School of Medicine,Shanghai Jiaotong University by immunohistochemistry was massively screened to turn out whether the expression of PENK was lower at high risk GISTs compared with low to median risk GISTs,at the same time the correlation between the expression of PENK and the clinicopathological grade was investigated,and also the prognosis of the patients was compared with the clinicopathological data and the follow-up results.3)GIST-882,GIST-T1 cell lines function test: The GIST-882 cell line,GIST-T1 cell line were managed to verify the expression of PENK,and then PENK stable interference and overexpression in GISTT1 cell lines were constructed: The proliferation ability of GIST-T1 cells was confirmed by CCK-8 experiment and cell clone formation experiment.GIST-T1 migration and invasion ability were verified by scratch experiment and Transwell experiment.The cell cycle and apoptosis were tested to verify the influences by PENK.The effect of PENK on the tumorigenicity of GIST-T1 cells in nude mice was confirmed by constructing subcutaneous tumor-bearing GIST-T1 cells.4)The mechanism of PENK in GIST process were studied and analyzed through analysis of upstream regulation of transcription factor prediction and the downstream signaling pathway activation.5)At the same time,a novel GIST-T1 imatinib mesylate resistant cell line was constructed to investigate the function of PENK in the GIST-T1 IM resistant cell line.RESULTS: 1)The screened genes were differentially expressed in the database of paired GISTs in GEO,TCGA,Oncomine by the Fold change value was greater than 2,and also the gene screening was determined by the copy number amplification.The selected genes which were all high expressed appeared identically in the two datasets(TOX,CHEK1,FOXG1,CEP170,BDNF,KIF1 A,PEG10,SATB2,TCF19,MCPH1,SIX1,RFTN2,GNG2,PENK).2)q RT-PCR were performed on the 14 differentially expressed genes.The results turned out that PENK showed low expression in normal tissues but showed high expression in GIST tissues.There was significant difference between the two groups(P=0.0188).Further comparison in the GIST groups showed PENK was highly expressed in the low and median risk GIST groups,but low in high risk GIST groups(statistically different,P=0.00014).There was no significant difference in the low and median risk groups(P = 0.837).The expression of PENK was significantly lower in the high risk GIST group than that in the normal tissues(statistically different,P = 0.015).3)Immunohistochemical staining results of 268 GIST tissue microarrays showed the negative expression rate of PENK in high GIST tissues was 41%,which was significantly higher than that in low-risk GIST tissues(22.8%),and there was significant difference between the two groups(P=0.001).To further correlate PENK expression with NIH risk stratification we found that PENK protein expression level was negatively associated with tumor size(P=0.001),mitotic phase(P=0.0002),tumor rupture and hemorrhage(P=0.026),NIH risk classification(P=0.001).With the increase of GIST tumor volume,the increase risk of tumor rupture,the increase of tumor mitotic phase and the increase risk of NIH,the positive rate of PENK protein staining in GIST tumor tissue decreased,and the negative rate increased significantly,suggesting that PENK in GIST progress may play as a tumor suppressor gene.4)Based on the analysis of survival rate and disease recurrence rate of 268 GIST patients,we found that the survival rate of patients with high PENK expression group was significantly higher than that in low PENK expression group,the difference was statistically significant(P=0.036).The contrary result occurred in GIST patients with disease recurrence rate.The recurrence rate of GIST in patients with high expression of PENK was significantly lower than that in low PENK expression group,the difference was statistically significant(P=0.001).The same conclusion was drawn in 268 GIST patients with different levels of PENK expression combined with the patient’s followup data and the survival curve by Logistic regression analysis.High PENK expression significantly improved the OS(overall survival)rate of patients with GIST,so did the RFS(recurrence free survival)rate,the difference was statistically significant,P values were 0.034 and 0.033 respectively,which meant that high PENK expression showed better prognosis but not the lower one.At the same time,we also found that among the268 GIST patients the OS of the low and median risk patients was significantly better than high risk patients according to NIH risk classification(very low risk/low risk,median risk,and high risk),so did the RFS,P=0.001,which showed high expression level of PENK positively correlated with low to median NIH stratification and positively correlated with better prognosis,otherwise the result is on the contrary.In addition to specify the effects of imatinib on OS and RFS in 268 GIST patients,we performed Logistic regression analysis and found that taking IM and not taking IM did not affect OS and RFS in 268 patients,P value was greater than 0.05.5)In the further validation of cell function experiment,we found that PENK overexpression GIST-T1 cells and stable PENK interference GIST-T1 cells showed a diametrically opposite function: In cell proliferation experiments,PENK stable interference GIST-T1 cells grew faster than PENK overexpressing ones(P <0.05).In the identification of cell migration,invasive ability of Transwell experiments,PENK overexpression of GISTT1 cell lines displayed weaker in the scratches of cell migration,so did the invasion in Transwell experiments.In the cell cycle test,PENK overexpression or stable interference GIST-T1 cell line had little effect on GIST-T1 cell cycle,there was no significant difference.In the apoptosis experiments,the ability of PENK to induce apoptosis in overexpression GIST-T1 cell lines was significantly higher than that in stable interference GIST-T1 cell lines.In vitro experiments,stable PENK interference GIST-T1 cells induced tumorigenic ability was significantly higher than that of control groups.Therefore,we validated PENK as a tumor suppressor gene in inhibition of tumor proliferation in vivo and in vitro.6)In the study of PENK how to trigger its anticancer mechanism research,we found that the promotion of PENK was modulated by AP-1 gene.PENK could not inhibit the cell cycle,but the enzymatic hydrolysis of PENK,OGF(opiod growth factor)could arrest the cell cycle by activating the tumor suppressor gene P16 to inhibit the binding of Cdk4 and Cyclin D1,the cell cycle were arrested in G0/G1 phase,and thus showed the inhibition of GIST proliferation,meanwhile PENK also could activate apoptosis-related pathways to increase GIST apoptosis.In this study,we confirmed that PENK could not inhibit the growth of imatinib-resistant GIST-T1 cell line but the enzymatic hydrolysis of PENK,OGF could inhibit the growth of imatinib-resistant GIST-T1 cell line and thus laid the foundation for further study on the mechanism of imatinib-resistant GIST-T1 cells.CONCLUSIONS: 1)PENK was highly expressed in low and median risk GIST tumor tissues and was low in high risk GIST tumor tissues,and the expression level of PENK in high risk GISTs was lower than that of normal tissues.The expression level of PENK was positively correlated with the OS and RFS rate.2)PENK showed its inhibition of proliferation as a tumor suppressor gene in GIST cells in vivo and in vitro.3)It was OGF but not PENK that could inhibit the cell cycle in G0/G1 phase by activating tumor suppressor gene P16.PENK could activate apoptosis-related pathways to increase GIST apoptosis.4)PENK could not inhibit the growth of imatinib-resistant GIST-T1 cells but OGF could inhibit the proliferation in vitro.