Background m6A modification has been proved to play an important role in many biological processes. METTL3 as the main methyltransferase for methylation process has been found to be upregulated in many cancers, including CRC. Here, we investigate m6A modification and the underlying mechanism of METTL3 in the development of CRC. Methods The expression of METTL3 was detected in large clinical patient samples. To evaluate the function of METTL3 in vitro and in vivo, colony formation, CCK-8, cell migration and invasion assays were performed. To find out the downstream target of METTL3, GEO dataset was re-mined. We analyzed expression and metastasis-related miRNA by Pearson correlation, and miR-1246 was selected. Here, to identify the downstream target of miR-1246, Targetscan and miRWalk were used. RIP and luciferase reporter assay further confirmed SPRED2 as the direct target of miR-1246. Results We found that upregulated METTL3 is responsible for abnormal m6A modification in CRC and correlates positively with tumor metastasis. The gain- and loss-of-function indicates that METTL3 promotes cell migration and invasion in vitro and in vivo. Additionally, we confirmed that METTL3 can methylate pri-miR-1246, which further promotes the maturation of pri-miR-1246. By using bioinformatics tools, anti-oncogene SPRED2 was identified as the downstream target of miR-1246, wherein downregulated SPRED2 further reverses the inhibition of the MAPK pathway. Conclusions The present study demonstrates that the METTL3/miR-1246/SPRED2 axis plays an important role in tumor metastasis and provides a new m6A modification pattern in CRC development. Electronic supplementary material The online version of this article (10.1186/s13046-019-1408-4) contains supplementary material, which is available to authorized users.

No abstract

Background/Aims: Long non-coding RNAs (lncRNAs) acting as competing endogenous RNAs (ceRNAs) play significant roles in the development of tumors, but the functions of specific lncRNAs and lncRNA-related ceRNA networks have not been fully elucidated for colon adenocarcinoma (COAD). In this study, we aimed to clarify the lncRNA-microRNA (miRNA)-mRNA ceRNA network and potential lncRNA biomarkers in COAD. Methods: We extracted data from The Cancer Genome Atlas (TCGA) and identified COAD-specific mRNAs, miRNAs, and lncRNAs. The biological processes in Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) were analyzed for COAD-specific mRNAs. We then constructed a ceRNA network of COAD-specific mRNAs, miRNAs and lncRNAs and analyzed the correlation between expression patterns and clinical features of the lncRNAs involved. After identifying potential mRNA targets of 4 lncRNAs related to overall survival (OS), we conducted stepwise analysis of these targets through GO and KEGG. Using tissue samples from our own patients, we also verified certain analytical results using quantitative real-time PCR (qRT-PCR). Results: Data from 521 samples (480 tumor tissue and 41 adjacent non-tumor tissue samples) were extracted from TCGA. A total of 258 specific lncRNAs, 206 specific miRNAs, and 1467 specific mRNAs were identified (absolute log₂ [fold change] > 2, false discovery rate < 0.01). Analysis of KEGG revealed that specific mRNAs were enriched in cancer-related pathways. The ceRNA network was constructed with 64 lncRNAs, 18 miRNAs, and 42 mRNAs. Among these lncRNAs involved in the network, 3 lncRNAs (LINC00355, HULC, and IGF2-AS) were confirmed to be associated with certain clinical features and 4 lncRNAs (HOTAIR, LINC00355, KCNQ1OT1, and TSSC1-IT1) were found to be negatively linked to OS (log-rank p < 0.05). KEGG showed that the potential mRNA targets of these 4 lncRNAs may be concentrated in the MAPK pathway. Certain results were validated by qRT-PCR. Conclusion: This study providing novel insights into the lncRNA-miRNA-mRNA ceRNA network and reveals potential lncRNA biomarkers in COAD.