According to the global burden of disease study in 2018, colorectal cancer (CRC) is one of the most common cancer types worldwide, and represents the third leading cause of cancer-related deaths (1). Many risk factors have been associated with CRC, including genetic and environmental factors. Hereditary CRC accounts for ~5–10% of all CRC cases, while the tumorigenesis and development of CRC arise through the accumulation of a multistep carcinogenic process that affected by lifestyle and dietary factors (2). In the last decade, increasingly researches have focused on the multiple molecular pathways involved in CRC pathogenesis, especially genetic and epigenetic events (3, 4). It is generally believed that, epigenetic alterations in CRC occur earlier and happened more frequently than genetic alterations (5, 6). Genomic instability, mutational inactivation of tumor suppressor genes, and activation of oncogenes are all involved in its development (6). Nowadays, genome-wide association studies have recently linked CRC to several common genetic variants or single-nucleotide polymorphisms, meanwhile, miRNA profiling of CRCs has identified over 20 up and downregulated miRNAs (7).

Natural RNA molecules contain multiple chemically modified nucleosides (8). Aside of DNA methylation and histone modification, mRNA modification represents another layer of epigenetic regulation of gene expression (9). The extent of RNA modifications is highly sensitive to changes in cellular micro-environment or the switch between different physiological states, and these changes in patterns of RNA modification would in turn affect the regulation for cell function and adaptation (10, 11). A wide range of RNA base modifications, collectively called the epitranscriptome, have been implicated in translation control, RNA splicing defects and many cancer types (12). N6-methyl-adenosine (m6A) is the prevalent modification in eukaryotic mRNAs, whose reversible methylation may have a profound impact on gene expression regulation (13). Abnormal methylation of m6A would cause a range of diseases, including tumors (14, 15), neurological diseases (16, 17), and embryonic retardation (18). Recently, their roles in cancer biology and cancer stem cells have become prominent hotspot in the research field of on tumorigenesis and screening of potential biological targets. However, the cellular and functional dynamics of m6A RNA modifications in the regulation of CRC development remain largely unexplored.

In this study, the epidemiologic information of 522 CRC samples was obtained from the Cancer Genome Atlas (TCGA). We analyzed the association between the expression of m6A regulators and CRC initiation and progression, screened the key regulators whose expression was related with the overall survival (OS) of CRC patients, and constructed a CRC prognosis signature based on these regulators to evaluate the roles of m6A regulators in CRC development and prognosis.

Similar to DNA methylation and histone modification, m6A RNA modifications can be added by writer enzymes and removed by eraser enzymes. The differential expressions of specific RNA m6A methylation regulators are linked to mis-regulated RNAs in tumors, however, the same m6A methylation regulators may have distinct functions in different tumors (19). In this study, we proposed to explore the feasibility of the common m6A regulators in CRC prognosis estimation. Differential expression analysis found that 8 m6A regulators, including METTL14, WTAP, YTHDC1, YTHDC2, ALKBH5, FTO, YTHDF1, and YTHDF2, were significantly differently expressed between adjacent normal and tumor samples. This should preliminarily suggest that those m6A regulators have the potential of influencing CRC initiation and could be used for the subsequent analysis. In addition, among those differential expression regulators, the expressions of WTAP and FTO exhibited significantly sustained elevated and decreased with the progression of CRC, respectively, suggesting enhanced m6A RNA methylation with the increasing tumor grade. METTL3 and METTL14 were both writer m6A regulatory enzymes, and WTAP can improve methylation efficiency in nuclear speckles by translocating the METTL3-METTL14 complex to mRNA targets (20). FTO is the first discovered m6A eraser that can influence the transcription of adjacent genes (21, 22). The up-regulation of WTAP was able to promote the proliferation and survival of tumor cells, while FTO plays an oncogenic role as an m6A demethylase in acute myeloid leukemia (22, 23). Furthermore, the expression trend of WTAP and FTO that same as our result was positively associated with the malignant progression in gliomas (24).

The abnormal methylation of m6A mRNA has shown prognostic value in multiple tumors, such as cervical cancer (25), acute myeloid leukemia (26), pancreatic cancer (27), and hepatocellular carcinoma (28). Kandimalla's group has reported a risk-score derived from a seven gene mRNA expression classifier consisting of METTL3, METTL14, WTAP, YTHDF1, YTHDF2, FTO, and ALKBH5 that associated with poor disease-free survival of CRC patients (29). In this study, YTHDC2 and ALKBH5 were picked out by multivariate Cox proportional hazard model, and applied for the construction of CRC prognosis prediction signature. The result of multivariate Cox regression analysis suggested that the signature based on YTHDC2 and ALKBH5 was able to predict CRC patients prognosis reliably independent of age, gender and stage.

YTHDC2 is a nuclear localized protein that is widely expressed in human cancer cell lines (30, 31). YTHDC2 can promote cancer metastasis via enhancing the translating efficiency of hypoxia-inducible factor (HIF)-1α in colon tumor cells, and may become a diagnostic marker and target gene for treating colon cancer patients (32). Fanale et al. suggested YTHDC2 as a potential candidate for pancreatic cancer susceptibility and a useful marker for early detection (32). However, the regulatory mechanism of YTHDC2 on tumor behavior is still poorly understood. ALKBH5 belongs to the non-heme Fe(ii)- and α-ketoglutarate (KG)-dependent dioxygenase AlkB family of proteins demethylated NANOG mRNA, and stimulated the expressions of HIF-1α and HIF-2α of breast cancer cells when exposure to hypoxia (33, 34). It was reported that, the growth of CRC is associated with the physiological state of hypoxia, and HIF-1α is a key factor for CRC metastasis (35). What's more, aberrant expressions of YTHDC2 and ALKBH5 in the context of CRC initiation and progression have been previously reported, and consistent results as the current study were obtained (32, 36). Therefore, we speculate that, the RNA methylation regulated by YTHDC2 and ALKBH5 might regulate tumor proliferation and metastasis through HIF-mediated hypoxic microenvironment response and then affect the prognosis of patients, which would be confirmed in our future research.

In this study, we comprehensively analyzed associations between mRNA expressions of m6A regulators with the development and prognosis of CRC. Among the m6A regulators, the abnormal expressions of WTAP and FTO were found to be significantly related to the progression of CRC, and YTHDC2 and ALKBH5 were identified as key regulators that could predict the prognosis of patients with CRC independently. This study highlighted the important role of RNA modification in the development of CRC, and provided potential guiding biomarkers for the therapeutic method selection.

Publicly available datasets were analyzed in this study. This data can be found at: The Cancer Genome Atlas (TCGA), Datasets link: https://portal.gdc.cancer.gov/repository.

LJ and SC put forward the ideas of this article, wrote this article, and analyzed the data. LG helped with acquisition of data and analysis and interpretation of data. XZ analyzed the data and helped with revising the manuscript. All authors read and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.