Kechao Nie1, Zhihua Zheng1, Yi Wen1, Jinglin Pan2, Yufeng Liu3, Xiaotao Jiang1, Yanhua Yan1, Kailin Jiang1, Peng Liu1, Shijie Xu4, Fengbin Liu5, Peiwu Li6. 1. Department of Gastroenterology, The First Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou 510405, Guangdong, China; First Clinical Medical College, Guangzhou University of Chinese Medicine, Guangzhou 510405, Guangdong, China. 2. Department of Gastroenterology, Hainan Provincial Hospital of Traditional Chinese Medicine, Haikou 570100, Hainan, China. 3. Department of Gastroenterology, The First Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou 510405, Guangdong, China. 4. Guangzhou University of Chinese Medicine, Guangzhou 510405, Guangdong, China. Electronic address: xsj@gzucm.edu.cn. 5. Department of Gastroenterology, The First Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou 510405, Guangdong, China. Electronic address: liufb163@vip.163.com. 6. Department of Gastroenterology, The First Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou 510405, Guangdong, China. Electronic address: doctorlipw@gzucm.edu.cn.
Abstract
BACKGROUND: Increasing evidence suggests that the lncRNA-miRNA-mRNA regulatory network is highly correlated with gastric cancer (GC) development. However, a prognosis-associated lncRNA-miRNA-mRNA network remains to be identified in GC. METHODS: Differentially expressed genes (DEGs) were screened by integrating 6 microarray datasets using the RRA method. Hub genes were identified by analysing their degrees in a PPI (protein-protein interaction) network. Upstream miRNAs and lncRNAs of hub genes were predicted by miRTarBase and miRNet, respectively. Key genes, miRNAs and lncRNAs were identified by evaluating their expression and prognosis in GEPIA and Kaplan-Meier plotter, respectively. A key lncRNA-miRNA-mRNA network was constructed in Cytoscape, and the correlations were analysed in the ENCORI database. We also evaluated the mRNA expression of ceRNA axes in the TIMER and Oncomine databases and their correlation with prognosis in GC patients with different clinical features using Kaplan-Meier plotter. In addition, correlations between mRNA and immune infiltrating cells in GC were investigated by the TIMER database. Finally, several experiments were conducted to verify our analyses. RESULTS: Forty-two upregulated and 86 downregulated DEGs were obtained from the "RRA" integrated analysis. Eight of the 20 hub genes were identified as key genes by analysing their expression and prognosis. Seventeen miRNAs were predicted to target key genes, and low expression of 4 miRNAs suggested poor outcome in GC. Furthermore, 155 lncRNAs were predicted to target 4 key miRNAs, and only 5 lncRNAs were highly expressed, suggesting poor outcomes in patients with GC. Then, the H19-miR-29a-3p-COL1A2 axis was constructed by correlation analysis. In addition, COL1A2 was positively correlated with lymphatic metastasis, immune infiltrating cell levels, markers of monocytes, tumour-associated macrophages (TAMs), and M2 macrophages but not M1 macrophages in GC. The experimental results revealed that the H19-miR-29a-3p-COL1A2 axis may promote macrophage polarization from M1 to M2 in GC. CONCLUSIONS: A novel lncRNA-miRNA-mRNA axis was identified and may be involved in regulating immune cell infiltration and macrophage polarization, which may provide new treatment strategies for GC.
BACKGROUND: Increasing evidence suggests that the lncRNA-miRNA-mRNA regulatory network is highly correlated with gastric cancer (GC) development. However, a prognosis-associated lncRNA-miRNA-mRNA network remains to be identified in GC. METHODS: Differentially expressed genes (DEGs) were screened by integrating 6 microarray datasets using the RRA method. Hub genes were identified by analysing their degrees in a PPI (protein-protein interaction) network. Upstream miRNAs and lncRNAs of hub genes were predicted by miRTarBase and miRNet, respectively. Key genes, miRNAs and lncRNAs were identified by evaluating their expression and prognosis in GEPIA and Kaplan-Meier plotter, respectively. A key lncRNA-miRNA-mRNA network was constructed in Cytoscape, and the correlations were analysed in the ENCORI database. We also evaluated the mRNA expression of ceRNA axes in the TIMER and Oncomine databases and their correlation with prognosis in GC patients with different clinical features using Kaplan-Meier plotter. In addition, correlations between mRNA and immune infiltrating cells in GC were investigated by the TIMER database. Finally, several experiments were conducted to verify our analyses. RESULTS: Forty-two upregulated and 86 downregulated DEGs were obtained from the "RRA" integrated analysis. Eight of the 20 hub genes were identified as key genes by analysing their expression and prognosis. Seventeen miRNAs were predicted to target key genes, and low expression of 4 miRNAs suggested poor outcome in GC. Furthermore, 155 lncRNAs were predicted to target 4 key miRNAs, and only 5 lncRNAs were highly expressed, suggesting poor outcomes in patients with GC. Then, the H19-miR-29a-3p-COL1A2 axis was constructed by correlation analysis. In addition, COL1A2 was positively correlated with lymphatic metastasis, immune infiltrating cell levels, markers of monocytes, tumour-associated macrophages (TAMs), and M2 macrophages but not M1 macrophages in GC. The experimental results revealed that the H19-miR-29a-3p-COL1A2 axis may promote macrophage polarization from M1 to M2 in GC. CONCLUSIONS: A novel lncRNA-miRNA-mRNA axis was identified and may be involved in regulating immune cell infiltration and macrophage polarization, which may provide new treatment strategies for GC.