Abstract LB-378: The outlier approach discriminates drivers from passengers among genes methylated in cancers

• Published in: Cancer research (Chicago, Ill.) Ill.), 2012-04, Vol.72 (8_Supplement), p.LB-378-LB-378
• Main Authors: Kikuyama, Mizuho, Takeshima, Hideyuki, Kinoshita, Takayuki, Okochi-Takada, Eriko, Wakabayashi, Mika, Akashi-Tanaka, Sadako, Ogawa, Toshihisa, Seto, Yasuyuki, Ushijima, Toshikazu
• Format: Article
• Language: English
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.AM2012-LB-378

A large number of genes are methylated in cancers, and most of them are methylated as passengers of carcinogenesis, and this fact hampers identification of drivers, tumor-suppressor genes (TSGs), silenced by aberrant methylation of promoter CpG islands (CGIs). To overcome this issue, we focused on the rule that the vast majority of genes methylated in cancers lack, in normal cells, RNA polymerase II (Pol II) and have trimethylation of histone H3 lysine 27 (H3K27me3) in their promoter CGIs. However, approximately 5% of the genes methylated in cancers are against this general rule, constituting a group of “outliers”. It is expected that TSGs belong to this group of outliers since they are expressed or ready to be expressed in normal cells. Here, we aimed to demonstrate that TSGs belong to the group of outliers, and that searching for outliers enables us to identify TSGs. First, we demonstrated that some of known TSGs in breast and colon cancers had Pol II and lacked H3K27me3 in normal cells, being outliers of the rule. We then made a genome-wide search for outlier genes in breast cancers. Based on Pol II binding and H3K27me3 statuses in normal cells and DNA methylation statuses in five breast cancer cell lines, 14 outlier genes were identified from 280 methylated genes. Among these genes, four genes were confirmed to be methylated in primary breast cancer samples, and two (HOXA5 and FBN2) of these were known TSGs. Among the remaining two genes, DZIP1 was shown to suppress growth of breast cancer cells, suggesting it to be a novel TSG. These results showed that some of known TSGs are the outlier genes and that TSGs can be efficiently identified by searching for outliers against the rule, the outlier approach. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr LB-378. doi:1538-7445.AM2012-LB-378