Epigenetic inactivation of the RASSF1A 3p21.3 tumor suppressor gene in both clear cell and papillary renal cell carcinoma

Renal cell carcinoma (RCC), the most common adult kidney neoplasm, is histopathologically heterogeneous, with most sporadic RCCs (similar to 80%) classified as clear cell (CC) tumors. Chromosome 3p allele loss is the most frequent genetic alteration in RCC but is associated specifically with sporadic and hereditary forms of clear cell RCC (CC-RCC) and is not a feature of non-CC-RCC, such as papillary (chromophilic) RCC. The VHL tumor suppressor gene (TSG) maps to chromosome 3p25, and somatic inactivation of the VHL gene occurs in up to 70% of CC-RCC tumors and cell lines. However, VHL inactivation is not sufficient for CC-RCC tumorigenesis, and inactivation of 3p12-p21 TSG(s) appears to be necessary in CC-RCC irrespective of VHL gene inactivation status. Recently, we demonstrated that the candidate 3p21 TSG, RASSF1A, is hypermethylated in most small cell lung cancers. We have now investigated the role of RASSF1A inactivation in primary RCC tumors. RASSF1A promoter methylation was detected in 23% (32 of 138) of primary CC-RCC tumors. In CC-RCC cell lines, RASSF1A methylation was associated with silencing of RASSF1A expression and restoration of expression after treatment with 5'-azacytidine. The frequency of RASSF1A methylation was similar in CC-RCC with and without VHL gene inactivation (24% versus 21%), and there was no association between epigenetic silencing of the RASSF1A and VHL TSGs, because 0 of 6 tumors with VHL hypermethylation had RASSF1A methylation, and VHL was not methylated in 26 CC-RCCs with RASSF1A methylation. Although 3p allele loss has been reported rarely in papillary RCC, we identified RASSF1A methylation in 44% (12 of 27) of papillary RCCs analyzed. Thus: (a) inactivation of RASSF1A is a frequent event in both CC-RCC and papillary RCC tumors; (b) there is no relationship between epigenetic silencing of RASSF1A and VHL inactivation status in CC-RCC. Fifty-four CC-RCCs analyzed for RASSF1A methylation were informative for 3p21 allele loss, and 20% (7 of 35) with 3p21 allele loss demonstrated RASSF1A methylation. All informative CC-RCCs with 3p21 allele loss and no RASSF1A methylation also demonstrated allele losses at other regions of 3p so that tumorigenesis in these cases may result from: (a) haploinsufficiency of RASSF1A; (b) inactivation of other 3p21 TSGs; or (c) inactivation of 3p TSGs from outside of 3p21. RASSF1A is the first TSG to be inactivated frequently in both papillary and CC-RCCs. The finding of frequent epigenetic inactivation of RASSF1A in papillary RCCs despite previous studies reporting infrequent 3p21 allele loss in this tumor type illustrates how the systematic identification of all major human cancer genes will require detailed analysis of the cancer genome and epigenome.