Whole genome survey for epigenetic changes in response to EGCG treatment in lung cancer cells

A157

Background: Polyphenon E (PE), a decaffeinated preparation of green tea shows chemopreventative effects in animal models for several different tissues. Recent studies suggests that (-)-epigalocatechin-3-gallate (EGCG), the major component in Polyphenon E may inhibit DNA methylation and reinstate normal methylation status of tumor suppressor genes in some cancers, as CpG island hypermethylation and the consequential down regulation of gene expression is observed for many genes involved in a diverse range of hallmark functions deregulated in cancer. Drugs that inhibit methylation are used both as a research tool to assess reactivation of genes silenced in cancer by hypermethylation and as a treatment of some hematological malignancies. Although previous studies have demonstrated epigenetic changes at specific genes such as RARβ and CDKN2A in response to EGCG treatment, a global survey on the whole genome scale is needed to identify novel genes that are re-expressed in response to EGCG exposure. Objective: Our goal is to discover genes whose expression is affected by PE treatment to understand the molecular mechanisms by which PE silences or induces transcription. To identify these changes we undertake the first parallel genomic and epigenomic profile comparison of pre and post EGCG treated cancer cells. Materials and Methods: Non small cell lung carcinoma cell line HCC15 was exposed to 20 and 50 µM EGCG over ten passages. Genomic profiles were generated utilizing whole genome tiling path array comparative genomic hybridization (CGH) microarray, consisting of duplicates of 27 000 overlapping clones spanning the entire human genome. Whole genome methylation profiles were determined by Methylation Dependent Immunoprecipitation (MeDIP) microarray CGH. This method uses an anti-5 methyl-cytosine specific antibody to isolate methylated DNA fragments by immunoprecipitation. Expression data will be performed using the Agilent 44K expression array and confirmed by quantitative RTPCR. Results: 20 µM and 50 µM EGCG inhibited cell growth, but provided sufficient quantities of DNA to perform all analysis in duplicate. Methylation analysis by MeDIP microarray CGH was reproducible and revealed distinct alterations corresponding to previously characterized changes in methylation. When aligned with whole genome copy number profiles, DNA from cells treated with this high dosage of EGCG showed regions of differential methylation corresponding to loci of DNA copy number alteration. In addition, hypermethylated regions also occur in the absence of copy number loss for some loci, suggesting methylation may also be the primary mechanism contributing to gene expression changes in response to PE. Therefore changes in methylation and copy number status may synergistically contribute to the anti-carcinogenic properties of EGCG. Conclusions: We demonstrate that parallel genomic and epigenomic analysis on a whole genome scale can be applied to study changes in DNA in response to PE treatment. Further analysis is underway to determine how these events impact gene expression and mediate cell cycle inhibition. Such analysis will be useful in the development of PE as a cancer chemopreventative agent.