Table S1. Cell samples used for CAGE profiling. Table S2. Public datasets used in this study. Table S3. Differentially expressed promoters in FANTOM5 NSCLC cell lines. Table S4. Up-regulated/hypomethylated promoters in NSCLC cell lines. Table S5. Epi-markers in NSCLC. Table S6. Extended information for Table 2. Table S7. Univariate and multivariate Cox regression models of TCGA LUAD and LUSC datasets. Table S8. Gene expression profiling of SAECs treated with 5-aza-dC and TSA. Table S9. Up-regulated/hypomethylated genes in NSCLC cell lines that show up-regulation by 5-aza-dC and TSA in SAECs. Table S10. Promoters and methylation array probes that overlap nine families of repetitive elements. Table S11. Promoters and methylation array probes that overlap REP522 repetitive elements. Table S12. Transcripts commonly down-regulated by two different MYEOV siRNAs in A549 cells. Table S13. The relationship between expression of 22 epi-markers and EGFR/KRAS mutation in TCGA LUAD dataset.
ARTICLE ABSTRACT
Lung cancer is the leading cause of cancer-related deaths worldwide. The majority of cancer driver mutations have been identified; however, relevant epigenetic regulation involved in tumorigenesis has only been fragmentarily analyzed. Epigenetically regulated genes have a great theranostic potential, especially in tumors with no apparent driver mutations. Here, epigenetically regulated genes were identified in lung cancer by an integrative analysis of promoter-level expression profiles from Cap Analysis of Gene Expression (CAGE) of 16 non–small cell lung cancer (NSCLC) cell lines and 16 normal lung primary cell specimens with DNA methylation data of 69 NSCLC cell lines and 6 normal lung epithelial cells. A core set of 49 coding genes and 10 long noncoding RNAs (lncRNA), which are upregulated in NSCLC cell lines due to promoter hypomethylation, was uncovered. Twenty-two epigenetically regulated genes were validated (upregulated genes with hypomethylated promoters) in the adenocarcinoma and squamous cell cancer subtypes of lung cancer using The Cancer Genome Atlas data. Furthermore, it was demonstrated that multiple copies of the REP522 DNA repeat family are prominently upregulated due to hypomethylation in NSCLC cell lines, which leads to cancer-specific expression of lncRNAs, such as RP1-90G24.10, AL022344.4, and PCAT7. Finally, Myeloma Overexpressed (MYEOV) was identified as the most promising candidate. Functional studies demonstrated that MYEOV promotes cell proliferation, survival, and invasion. Moreover, high MYEOV expression levels were associated with poor prognosis.Implications: This report identifies a robust list of 22 candidate driver genes that are epigenetically regulated in lung cancer; such genes may complement the known mutational drivers.Visual Overview: http://mcr.aacrjournals.org/content/molcanres/15/10/1354/F1.large.jpg. Mol Cancer Res; 15(10); 1354–65. ©2017 AACR.
