Search

hSRBC is a putative tumor suppressor located at 11p15.4, at which frequent genomic loss has been observed in several human malignancies. To explore the candidacy of hSRBC as a suppressor of gastric tumorigenesis, we analyzed the expression and mutation status of hSRBC in gastric tissues and cell lines. hSRBC transcript was expressed in all normal and benign tumor tissues examined, but undetectable or very low in 73% (11/15) cancer cell lines and 41% (46/111) primary tumors. Loss or reduction of hSRBC expression was tumor-specific and correlated with stage and grade of tumors. While allelic loss or somatic mutations of the gene were infrequent, its expression was restored in tumor cells by 5-aza-2'-deoxycytidine treatment and aberrant hypermethylation of 23 CpG sites in the promoter region showed a tight association with altered expression. Transient or stable expression of hSRBC led to a G(1) cell cycle arrest and apoptosis of tumor cells, and strongly suppresses colony forming ability and xenograft tumor growth. In addition, hSRBC elevated apoptotic sensitivity of tumor cells to genotoxic agents, such as 5-FU, etoposide and ultraviolet. Interestingly, hSRBC increased the protein stability of p53 and expression of p53 target genes, such as p21(Waf1), PUMA and NOXA, while hSRBC-mediated cell cycle arrest and apoptosis were abolished by blockade of p53 function. Our findings suggest that hSRBC is a novel tumor suppressor whose epigenetic inactivation contributes to the malignant progression of gastric tumors, in part, through attenuated p53 response to stresses., ((c) 2007 Wiley-Liss, Inc.)

Mutational alterations of PTEN and PIK3CA, which negatively and positively regulate PI3-kinase activity, respectively, have been observed in many types of human cancer. To explore the implication of PTEN and PIK3CA mutations in gastric tumorigenesis, we characterized the expression and mutation status of the genes in 126 gastric tissues and 15 cell lines. Expression of PTEN transcript was abnormally low in 5 of 15 (33%) cell lines and 20 of 55 (36%) primary carcinomas, whereas 0 of 71 noncancerous tissues including 16 benign tumors showed altered expression. Allelotyping analysis using an intragenic polymorphism (IVS4+109) revealed that 14 of 30 (47%) informative cases carried LOH of the gene, which is closely linked to low expression. The LOH rate was significantly higher in advanced tumors [12 of 19 (63%)] compared to early-stage tumors [2 of 11 (18%)] and more frequent in poorly differentiated tumors [9 of 13 (69%)] than well- or moderately differentiated tumors [5 of 17 (29%)]. Interestingly, however, none of the LOH tumors carried mutational disruption of the remaining allele, suggesting haploinsufficiency of PTEN in gastric tumorigenesis. Methylation studies revealed that PTEN pseudogene, but not PTEN, is methylated in cell lines and primary tumors, indicating that PTEN is not a target of epigenetic silencing in gastric cancers and that the pseudogene should be considered more carefully in methylation analysis of the PTEN promoter. Genomic amplification of PIK3CA was found in 9 of 15 (60%) cell lines and 20 of 55 (36.4%) primary tumors but in no noncancerous tissues. Furthermore, PIK3CA amplification was predominantly detected in tumors with no PTEN alterations, suggesting that mutations of PTEN and PIK3CA are mutually exclusive events in gastric tumorigenesis. Amplification of PIK3CA was strongly associated with increased expression of PIK3CA transcript and elevated levels of phospho-AKT. Collectively, our data reveal that 13 of 15 (87%) gastric cell lines and 31 of 55 (56%) primary carcinomas harbored either amplification of PIK3CA or abnormal reduction of PTEN. Mutually exclusive alterations of PTEN and PIK3CA also suggest that mutations of either gene could activate the PI3-kinase/AKT signaling pathway, which is directly linked to the malignant progression of gastric tumor cells., (Copyright 2003 Wiley-Liss, Inc.)

Methylation associated inactivation of RASSF1, a putative tumor suppressor identified at 3p21.3, has been frequently observed in several human malignancies, including lung and breast cancers. To explore the penetrance of RASSF1 in gastric carcinogenesis, we performed expression and mutation analyses of 3 isotypes of RASSF1 (A, B, and C) in 150 gastric specimens, including 15 carcinoma cell lines. RASSF1A and RASSF1B transcripts were not expressed in 60% (9 of 15) and 33% (5 of 15) of gastric carcinoma cell lines, respectively, whereas RASSF1C was detectable in all cell lines. Bisulfite DNA sequencing analysis revealed that the CpG island in the RASSF1A promoter is hypermethylated in all RASSF1A-nonexpressing cell lines. In addition, both RASSF1A and RASSF1B were re-expressed by treatment with the demethylating agent 5-aza-2'-deoxycytidine. Among 90 primary gastric adenocarcinomas examined, 41 (46%) and 19 (21%) expressed no or abnormally low levels of RASSF1A and RASSF1B, respectively, and 12 (13%) tumors showed no expression of both isoforms. Loss or abnormal down-regulation of RASSF1A correlated with tumor stage and grade but not with histological types of tumors. Methylation-specific PCR analysis demonstrated that 95% (39 of 41) of RASSF1A-nonexpressing primary tumors are methylated at the CpG sites in the promoter, whereas none of the adjacent noncancerous or normal tissues are methylated. No somatic mutations were detected in RASSF1 transcripts expressed in unmethylated tumors. However, 10 methylated tumors, including 4 cell lines, showed low genomic levels of RASSF1 and expressed no RASSF1A transcripts, suggesting that RASSF1A inactivation might be caused by both epigenetic and genetic mechanisms in a subset of gastric adenocarcinomas. In conclusion, our data indicate that epigenetic transcriptional silencing of RASSF1, especially RASSF1A isoform, is a frequent event in gastric tumorigenesis and might play an important role in the malignant progression of gastric adenocarcinomas.

Allelic deletion or transcriptional silencing of RASSF1, a putative tumor suppressor at 3p21.3, has been found in a considerable proportion of lung, breast, and ovarian cancers. In this study, we analyzed the expression and mutation status of three RASSF1 isoforms (-A, -B, and -C) in 55 primary bladder carcinomas and 10 bladder and prostate cancer cell lines. The RASSF1A transcript was not found in 80% (4 of 5) and 100% (4 of 4) of bladder and prostate cell lines, respectively. Compared with normal bladder tissues, loss or significant reduction of RASSF1A was identified in 62% (34 of 55) of primary bladder carcinomas and 10 (83%) of 12 matched sets showed tumor-specific alteration of RASSF1A expression. Moreover, loss or abnormal down-regulation of RASSF1A correlated with advanced tumor stage. RASSF1B was undetectable in 60% (3 of 5) of bladder cell lines and in 31% (17 of 55) of primary tumors, but none of these tumors showed altered expression exclusively in RASSF1B. RASSF1C transcript was detected in all cell lines and primary tumors we examined. Expression of RASSF1A and RASSF1B was reactivated in all nonexpressor cell lines by treatment with the demethylating agent 5-aza-2'-deoxycytidine. Bisulfite DNA sequencing analysis revealed that aberrant hypermethylation at the CpG island in the RASSF1A promoter is strongly associated with the loss of RASSF1A expression in cell lines and uncultured primary tumors. Methylation-specific PCR and BstUI digestion analyses also demonstrated that 97% (33 of 34) of RASSF1A-nonexpressing primary tumors are methylated. Although somatic mutations were not identified in RASSF1 transcripts expressed in unmethylated tumors, 24% (9 of 37) of methylated cell lines and primary tumors showed detectable reductions in genomic levels of RASSF1, suggesting that RASSF1A inactivation might be caused by both epigenetic and genetic mechanisms in a subset of bladder tumors. Together, our data suggest that RASSF1A inactivation may play a critical role in the malignant progression of human bladder carcinomas.