Comprehensive analysis of m(6)A circRNAs identified in colorectal cancer by MeRIP sequencing

PURPOSE: To characterize the entire profile of m(6)A modifications and differential expression patterns for circRNAs in colorectal cancer (CRC). METHODS: First, High-throughput MeRIP-sequencing and RNA-sequencing was used to determine the difference in m(6)A methylome and expression of circRNA between CRC tissues and tumor-adjacent normal control (NC) tissues. Then, GO and KEGG analysis detected pathways involved in differentially methylated and differentially expressed circRNAs (DEGs). The correlations between m(6)A status and expression level were calculated using a Pearson correlation analysis. Next, the networks of circRNA-miRNA-mRNA were visualized using the Target Scan and miRanda software. Finally, We describe the relationship of distance between the m(6)A peak and internal ribosome entry site (IRES) and protein coding potential of circRNAs. RESULTS: A total of 4340 m(6)A peaks of circRNAs in CRC tissue and 3216 m(6)A peaks of circRNAs in NC tissues were detected. A total of 2561 m(6)A circRNAs in CRC tissues and 2129 m(6)A circRNAs in NC tissues were detected. Pathway analysis detected that differentially methylated and expressed circRNAs were closely related to cancer. The conjoint analysis of MeRIP-seq and RNA-seq data discovered 30 circRNAs with differentially m(6)A methylated and synchronously differential expression. RT-qPCR showned circRNAs (has_circ_0032821, has_circ_0019079, has_circ_0093688) were upregulated and circRNAs (hsa_circ_0026782, hsa_circ_0108457) were downregulated in CRC. In the ceRNA network, the 10 hyper-up circRNAs were shown to be associated with 19 miRNAs and regulate 16 mRNAs, 14 hypo-down circRNAs were associated with 30 miRNAs and regulated 27 mRNAs. There was no significant correlation between the level of m(6)A and the expression of circRNAs. The distance between the m(6)A peak and IRES was not significantly related to the protein coding potential of circRNAs. CONCLUSION: Our study found that there were significant differences in the m(6)A methylation patterns of circRNAs between CRC and NC tissues. M(6)A methylation may affect circRNA-miRNA-mRNA co-expression in CRC and further affect the regulation of cancer-related target genes.