AIM: To identify genes potentially involved in Helicobacter pylori (H. pylori)-induced gastric carcinogenesis. METHODS: GES-1 cells were co-cultured with H. pylori strains isolated from patients with gastric carcinoma (GC, n = 10) or chronic gastritis (CG, n = 10) for in vitro proliferation and apoptosis assays to identify the most and least virulent strains. These two strains were cagA-genotyped and used for further in vivo carcinogenic virulence assays by infecting Mongolian gerbils for 52 wk, respectively; a broth free of H. pylori was lavaged as control. Genomic profiles of GES-1 cells co-cultured with the most and least virulent strains were determined by microarray analysis. The most differentially expressed genes were further verified using quantitative real-time polymerase chain reaction in GES-1 cells infected with the most and least virulent strains, and by immunohistochemistry in H. pylori positive CG, precancerous diseases, and GC biopsy specimens in an independent experiment. RESULTS: GC-derived H. pylori strains induced a potent proliferative effect in GES-1 cells in co-culture, whereas CG-derived strains did not. The most (from a GC patient) and least (from a CG patient) virulent strains were cagA-positive and negative, respectively. At week 52, CG, atrophy, metaplasia, dysplasia, and GC were observed in 90.0%, 80.0%, 80.0%, 90%, and 60.0%, respectively, of the animals lavaged with the most virulent strain. However, only mild CG was observed in 90% of the animals lavaged with the least virulent strain. On microarray analysis, 800 differentially expressed genes (49 up- and 751 down-regulated), involving those associated with cell cycle regulation, cell apoptosis, cytoskeleton, immune response, and substance and energy metabolisms, were identified in cells co-cultured with the most virulent strain as compared with those co-cultured with the least virulent strain. The six most differentially expressed genes (with a betweenness centrality of 0.1-0.2) were identified among the significant differential gene profile network, including JUN, KRAS, BRCA1, SMAD2, TRAF1, and HDAC6. Quantitative real-time polymerase chain reaction analyses verified that HDAC6 and TRFA1 mRNA expressions were significantly more up-regulated in GES-1 cells co-cultured with the most virulent strain than in those co-cultured with the least virulent strain. Immunohistochemistry of gastric mucosal specimens from H. pylori-positive patients with CG, intestinal metaplasia (IM), dysplasia, and GC showed that moderately positive and strongly positive HDAC6 expression was detected in 21.7% of CG patients, 30.0% of IM patients, 54.5% of dysplasia patients, and 77.8% of GC patients (P < 0.001). The up-regulation of TRAF1 expressions was detected in 34.8%, 53.3%, 72.7%, and 88.9% specimens of CG, IM, dysplasia, and GC, respectively (P < 0.001). CONCLUSION: The overexpression of HDAC6 and TRAF1 in GES-1 cells co-cultured with the GC-derived strain and in H. pylori-positive dysplasia and GC suggests that HDAC6 and TRAF1 may be involved in H. pylori-induced gastric carcinogenesis.
AIM: To identify genes potentially involved in Helicobacter pylori (H. pylori)-induced gastric carcinogenesis. METHODS: GES-1 cells were co-cultured with H. pylori strains isolated from patients with gastric carcinoma (GC, n = 10) or chronic gastritis (CG, n = 10) for in vitro proliferation and apoptosis assays to identify the most and least virulent strains. These two strains were cagA-genotyped and used for further in vivo carcinogenic virulence assays by infecting Mongolian gerbils for 52 wk, respectively; a broth free of H. pylori was lavaged as control. Genomic profiles of GES-1 cells co-cultured with the most and least virulent strains were determined by microarray analysis. The most differentially expressed genes were further verified using quantitative real-time polymerase chain reaction in GES-1 cells infected with the most and least virulent strains, and by immunohistochemistry in H. pylori positive CG, precancerous diseases, and GC biopsy specimens in an independent experiment. RESULTS: GC-derived H. pylori strains induced a potent proliferative effect in GES-1 cells in co-culture, whereas CG-derived strains did not. The most (from a GC patient) and least (from a CG patient) virulent strains were cagA-positive and negative, respectively. At week 52, CG, atrophy, metaplasia, dysplasia, and GC were observed in 90.0%, 80.0%, 80.0%, 90%, and 60.0%, respectively, of the animals lavaged with the most virulent strain. However, only mild CG was observed in 90% of the animals lavaged with the least virulent strain. On microarray analysis, 800 differentially expressed genes (49 up- and 751 down-regulated), involving those associated with cell cycle regulation, cell apoptosis, cytoskeleton, immune response, and substance and energy metabolisms, were identified in cells co-cultured with the most virulent strain as compared with those co-cultured with the least virulent strain. The six most differentially expressed genes (with a betweenness centrality of 0.1-0.2) were identified among the significant differential gene profile network, including JUN, KRAS, BRCA1, SMAD2, TRAF1, and HDAC6. Quantitative real-time polymerase chain reaction analyses verified that HDAC6 and TRFA1 mRNA expressions were significantly more up-regulated in GES-1 cells co-cultured with the most virulent strain than in those co-cultured with the least virulent strain. Immunohistochemistry of gastric mucosal specimens from H. pylori-positive patients with CG, intestinal metaplasia (IM), dysplasia, and GC showed that moderately positive and strongly positive HDAC6 expression was detected in 21.7% of CG patients, 30.0% of IM patients, 54.5% of dysplasiapatients, and 77.8% of GC patients (P < 0.001). The up-regulation of TRAF1 expressions was detected in 34.8%, 53.3%, 72.7%, and 88.9% specimens of CG, IM, dysplasia, and GC, respectively (P < 0.001). CONCLUSION: The overexpression of HDAC6 and TRAF1 in GES-1 cells co-cultured with the GC-derived strain and in H. pylori-positive dysplasia and GC suggests that HDAC6 and TRAF1 may be involved in H. pylori-induced gastric carcinogenesis.