While cancer as a complex disease is known to arise from a series of aberrant genetic and epigenetic profiles, its growth and disease-specific outcome can often be characterized by status of one oncogene, such as Estrogen Receptor (ER) of breast cancer. Positive ER appears in around 70% of human breast cancer cases and recent studies dissected the Estrogen-mediated gene regulation through specific DNA methylation patterns by comparison between ER+ and ER- tumor samples. However, this simple approach did not take into consideration the influence of other confounding regulation factors due to the heterogeneity of breast cancer and may be limited in uncovering complex underlying mechanism. To address this problem, we first hypothesize that DNA methylation status and its inter-gene similarity are modulated by multiple factors including ER. The disruption of ER will disrupt association of methylation patterns between two genes which are correlated otherwise. The epigenetic signaling cascade maintained by specific ER status can be further interpreted from the “co-methylation” relationship among genes.

Here we proposed a novel algorithm to identify the modulated gene methylation based on the “modulator of regulation” concept. Setting ER as the modulator, an ER modulated gene pair (ER-GP) was defined as pair of genes with differential correlation between ER+ and ER- samples. Fisher transformation on Pearson correlation coefficients was adopted for statistical power adjustment on sample sizes and the differences of absolute correlation coefficients were tested for statistical significance in the Fisher transform domain. By applying the algorithm to the MBDCap-Seq data of 76 breast tumors (47 ER+ and 29 ER-), we identified 5,673 putative ER-GPs of co-methylation (co-methyl. ER-GPs) involving 1,657 unique genes. In order to delineate a comprehensive and systematic view of interaction between identified ER-GPs, the co-methyl. ER-GPs were combined to construct the ER modulated gene methylation network. With an overall connectivity (average number of first-order neighbors of each gene in a network) of 6.85, genes in the methylation network were highly linked to each other and formed a branched signaling cascade. The nucleosome assembly protein 1-like 5 NAP1L5, previously reported as regulated by both estradiol and tamoxifen treatment, was identified as the top hub gene, indicating that it may serve as an important regulator in the ER modulation. Further Gene Ontology analysis revealed that NAP1L5 and its co-methylated genes were involved in nucleotide biosynthetic process, chromosome organization, and cell proliferation.

In the present study, we proposed a novel algorithm to identify the ER modulated methylation regulatory network. The findings provided biological insights into the underlying mechanism of epigenetic signaling governed by ER and may contribute to improvement of breast cancer treatment.

Citation Format: Yu-Chiao Chiu, Tzu-Hung Hsiao, Fei Gu, Yidong Chen, Tim H-M Huang, Eric Y. Chuang. Identification of estrogen receptor modulated gene methylation network in breast cancer. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 2889. doi:10.1158/1538-7445.AM2013-2889

Note: This abstract was not presented at the AACR Annual Meeting 2013 because the presenter was unable to attend.

©2013 American Association for Cancer Research
2013