| BackgroundGlobal cancer statistics for 2020 show that gastric cancer remains the fifth most common cancer worldwide and the fourth most common cancer causing death after lung,colorectal and liver cancers.Helicobacter pylori(H.pylori)is a primary risk factor for gastric cancer,and the rate of H.pylori infection in adults in areas with high incidence of gastric cancer in China is over 60%.Since the early symptoms of gastric cancer are not obvious,most patients are diagnosed at an advanced stage,and drug therapy becomes the main means at this time.Traditional chemotherapeutic drugs have limited efficacy and high side effects,resulting in a low five-year survival rate of gastric cancer patients.Therefore,it is important to reveal the mechanism of gastric cancer development,develop targeted drugs for gastric cancer treatment,and establish a robust prognostic model for gastric cancer to guide the clinical use of drugs to prolong the survival of gastric cancer patients.Epigenetic regulation plays an important role in the process of gastric cancer development and progression.The study of epigenetic modifications to explore the mechanism of gastric cancer development can provide theoretical support for early intervention and late treatment of the disease.Because epigenetic modifications are reversible,epigenetic molecules have the potential to become potential targets for gastric cancer therapy.Histone demethylation modification as an important epigenetic modification has been a hot spot in oncology research.Through data analysis,we found that histone demethylation KDM6B was highly expressed in gastric cancer samples,and high expression of KDM6B in gastric cancer patients was associated with poor patient prognosis.Studies have reported that aberrant expression of KDM6B can promote or inhibit the progression of different tumors,however,the function and regulatory mechanism of KDM6B in gastric cancer progression have not been elucidated.The prognosis prediction of gastric cancer is important for the development of gastric cancer treatment plan and evaluation of treatment effect.Currently,AJCC(American Joint Committee on Cancer)TNM staging is still the most basic prognostic prediction tool for gastric cancer,and high staging suggests poor prognosis.However,due to the heterogeneity of gastric cancer and patients,patients with the same TNM stage may have different prognosis.Therefore,it is necessary to personalize the prediction according to patients' own characteristics and considering other important factors.In recent years,bioinformatics has developed at a high speed,and biological big data,such as gene chip data and RNA-seq data,have increased dramatically.Using bioinformatics analysis methods and analysis software to analyze biological big data has great advantages in formulating personalized treatment plans and establishing accurate prognostic prediction models,and has promising applications.Objective1.To analyze the role and mechanism of histone demethylase KDM6B in gastric carcinogenesis and development.2.To construct a multifactorial Cox risk model that can predict gastric cancer survival based on published data of gastric cancer using bioinformatics methods.Methods and Results1.KDM6B promotes gastric cancer cell proliferation and metastasis by demethylation of H3K27me3 near the CXCR4 promoter region through its enzymatic activity,which in turn upregulates CXCR4 expression.Helicobacter pylori can induce the expression of KDM6B.(1)By performing WGCNA(weighted gene co-expression network analysis)on the gastric cancer dataset GSE15460,three gene modules were found to be positively correlated with tumor staging.(2)We intersected a total of 1258 genes with 451 epigenetic genes to obtain a total of 8 genes in the three modules positively associated with tumor staging in result 1 above.Differential expression analysis of these eight genes in the GEPIA database revealed that ARID1A and KDM6B were highly expressed in gastric cancer tissues compared with normal tissues.By comparing Kaplan-Meier plotter survival curves,KDM6B was found to have a higher prognostic value for patients with gastric cancer.(3)Analysis by GEO datasets GSE13911 and GSE19826 revealed that KDM6B expression was significantly upregulated in gastric cancer samples.ROC curves(Receiver Operating Characteristic curve)were plotted in the datasets GSE13911 and GSE19826,and it was found that KDM6B expression could specifically distinguish gastric cancer tissues from paraneoplastic tissues.Real-time quantitative PCR revealed that the expression level of KDM6B mRNA was significantly higher in gastric cancer tissues than in paraneoplastic tissues.western blot assay and immunohistochemical assay both revealed that the expression level of KDM6B protein was significantly higher in gastric cancer samples than in paraneoplastic tissues.(4)In vitro clone formation and Edu assays revealed that knockdown with KDM6B or addition of GSK-J4,an inhibitor of KDM6B enzyme activity,could inhibit the proliferation of gastric cancer cells,and overexpression of KDM6B could promote the proliferation of gastric cancer cells,but overexpression of KDM6B with mutated enzyme activity had no such effect.Subcutaneous tumorigenesis experiments in nude mice revealed that knockdown of KDM6B inhibited the proliferation of gastric cancer cells in vivo.(5)Tanswell and scratch assay demonstrated that knockdown with KDM6B or addition of GSK-J4,an inhibitor of KDM6B enzymatic activity,could inhibit the metastasis of gastric cancer cells,and overexpression of KDM6B could promote the metastasis of gastric cancer cells,but overexpression of KDM6B with mutated enzymatic activity did not have such an effect.Western blot experiments revealed that interference with KDM6B or addition of GSK-J4,an inhibitor of KDM6B enzyme activity,down-regulated the expression of N-cadherin protein,a mesenchymal marker,and up-regulated the expression of E-cadherin and Claudin-1 protein,both epithelial markers,in gastric cancer cells.and Claudin-1 protein expression,and overexpression of KDM6B increased the expression of the mesenchymal marker N-cadheri protein and decreased the expression of the epithelial markers E-cadherin and Claudin-1 protein in gastric cancer cells,but KDM6B with mutations in enzymatic activity did not have this effect.(6)To investigate the specific mechanism by which KDM6B functions in gastric cancer,after gastric cancer cells interfered with KDM6B,we performed RNA-Seq,which resulted in the identification of 769 differentially expressed genes.Pathway enrichment analysis revealed that a variety of tumors and signaling pathways associated with tumor development were enriched.By PPI analysis,we obtained the differential gene protein interaction network.Using MCC to score the protein interaction network,we identified CXCR4 as a possible downstream core gene regulated by KDM6B.Correlation analysis of KDM6B and CXCR4 in the GEPI database revealed that their mRNA expression levels were positively correlated.RT-QPCR and Western blot assays showed that knocking down or adding GSK-J4,a KDM6B inhibitor,both downregulated CXCR4 expression in gastric cancer cells,and overexpression of KDM6B could promote CXCR4 expression,but KDM6B with mutated enzyme activity did not have this effect.(7)Western blot results revealed that knockdown of KDM6B or addition of its enzymatic activity inhibitor GSK-J4 upregulated the protein level of H3K27me3 in gastric cancer cells.The presence of an enriched region of H3K27me3 near CXCR4 promoter region was found by WASHU website.Chip-qPCR results showed that knockdown of KDM6B significantly increased the enrichment level of H3K27me3 near CXCR4 promoter region in gastric cancer cells compared with the control group.(8)Transwell reversion assay revealed that interfering with CXCR4 could reverse the promotion of proliferation ability of gastric cancer cells due to overexpression of KDM6B.(9)RT-qPCR and Western blot assays showed that H.pylori infection promoted the expression of KDM6B mRNA and protein in gastric cancer cells with a timedependent trend.2.Based on public data,a 7-gene cluster risk model was constructed using a bioinformatics approach and validated in multiple datasets,further combined with age staging to construct a column line graph and perform performance evaluation.(1)WGCNA(weighted gene co-expression network analysis)analysis using publicly available data of gastric cancer revealed that there were dark blue gene modules positively associated with tumor stage.By PPI analysis and MCOD score,we obtained two gene sub-networks.Nine core genes were obtained by taking the intersection of the genes with MM>0.8 and GS>0.2 in the dark blue module and subnetwork 1 genes.Analysis using the GSE27411 and TCGA datasets revealed that the nine core genes were highly expressed in H.pylori-positive gastritis and gastric cancer samples compared with H.pylori-negative samples.(2)Based on the expression of nine core genes,we divided the gastric cancer samples into two groups.Comparison of survival curves revealed that patients in group 2 had unfavorable prognosis compared to group 1.By comparing the clinical characteristics of the two groups of patients,we found that group 1 samples were mainly metabolic and value-added,and group 2 samples were infiltrative.By comparing the tumor microenvironment,more immune and stromal cell infiltrates were found in group 2 samples and higher tumor purity in group 1.GSVA analysis revealed that multiple signaling pathways related to immune response,signaling,H.pylori infection,epithelial-mesenchymal transition(EMT),hypoxia and tumor progression pathways were enriched in group 2 samples.(3)The differential gene expression analysis was performed on the two groups of samples.7 genes were obtained by univariate Cox regression analysis,LASSO regression analysis and multivariate Cox regression analysis of differentially expressed genes:APOD,APOE,CCDC80,CTHRC1,FERMT2,GXYLT2 and SMPX.based on the expression and survival data of the 7 genes,we constructed a 7-gene multivariate Cox prognostic risk model was constructed and risk scores were obtained.Survival and ROC curves showed that patients in the high-risk group had poor prognosis in the training set and multiple validation sets,and the risk score could predict patient survival more accurately.Univariate Cox regression analysis in both the training set and multiple validation found that the risk score had significant prognostic significance.The results of multivariate Cox regression analysis showed that the risk score could be used as an independent prognostic factor.(4)By integrating age,staging and risk score,we constructed a column line graph.Calibration plots,ROC curve plots were evaluated and found that the column line graph showed good predictive performance at 1,3 and 5 years.By comparing with three previously reported prognostic models for gastric cancer,our column nomogram was found to exhibit superior predictive ability.The expression levels of seven genes in 16 pairs of gastric and paracancerous clinical tissues were examined by real-time quantitative PCR,which showed that APOE,CTHRC1 and GXYLT2 were highly expressed in gastric cancer tissues,APOD was lowly expressed in gastric cancer tissues,and there was no significant difference in CCDC80,FERMT2 and SMPX expression.Survival curves were plotted using the Kaplan-Meier plotter database and found that all 7 genes had significant prognostic value.(5)Analysis of patient risk scores and clinical characteristics revealed that patients with high tumor stage also had higher risk scores,patients with mixed and diffuse gastric cancer had higher risk scores,patients with lymphovascular infiltration had high risk scores,patients with mesenchymal phenotype(MP)had higher risk scores than those with epithelial phenotype(EP),and patients with distant metastasis(M1)had higher risk scores than those without metastasis(M0).GSEA analysis revealed that the low-risk group was mainly enriched to pathways associated with DNA replication,base excision and homologous recombination,and the high-risk group was mainly enriched to pathways associated with angiogenesis,epithelial mesenchymalization and tumor metastasis.Correlation analysis revealed that the risk scores were positively correlated with the expression of mesenchymal molecules CDH2,SNAIL1,SNAIL2 and VIM,and negatively correlated with the expression of epithelial molecule CDH1.(6)Comparing the tumor microenvironment between the high and low groups revealed that the immune score and stromal score of the samples in the high-risk group were higher than those in the high-risk group,with risk score negatively correlated with tumor purity The results of chi-square analysis showed that the proportion of patients with high immunogenicity was high in the high-risk group,while the proportion of patients with low immunogenicity was high in the low-risk group.Comparing the differences in 29 immune label scores between samples from the high and low risk groups,it was found that most immune labels,such as immune response-related labels,CD8+T cells,NK cells,immune checkpoints,TILs and IFN(interferon)responses were scored higher in samples from the high risk group.By comparing the expression levels of immune checkpoint molecules,it was found that the expression levels of immune checkpoint molecules were higher in the high-risk group samples than in the low-risk group.By analyzing the differential genes in the high-and low-group samples,we used the CMap database to obtain nine drugs with potential to treat patients in the high-risk group.Conclusion(1)In this study,we found that KDM6B expression is upregulated in gastric cancer and patients with high expression of gastric cancer have unfavorable prognosis.kdm6 can transcriptionally regulate the expression of downstream target gene CXCR4 through H3K27me3 demethylase activity,thus promoting the development and progression of gastric cancer.Our research results can provide experimental support for the development of new target drugs for the treatment of gastric cancer.(2)In this section,we constructed a 7-gene risk model associated with tumor progression using public data on gastric cancer.The risk model combined with clinical factors to construct a line graph can be used clinically to predict the prognosis of gastric cancer patients in order to guide clinical drug use. |
PDF Full Text Request