- Letter to the Editor
- Open Access
Molecular characterization and clinical relevance of m6A regulators across 33 cancer types
Molecular Cancer volume 18, Article number: 137 (2019)
The methylation of N6 adenosine (m6A) plays a critical role in diverse biological processes. However, knowledge regarding the reconstitution of m6A across cancer types is still lacking. Here, we systematically analyzed the molecular alterations and clinical relevance of m6A regulators across > 10,000 subjects representing 33 cancer types. We found that there are widespread genetic alterations to m6A regulators, and that their expression levels are significantly correlated with the activity of cancer hallmark-related pathways. Moreover, m6A regulators were found to be potentially useful for prognostic stratification, and we identified IGF2BP3 as a potential oncogene across multiple cancer types. Our results provide a valuable resource that will guide both mechanistic and therapeutic analyses of the role of m6A regulators in cancer.
Methylation of N6 adenosine (m6A) is the most common type of RNA modification, and it plays crucial roles in the development and progression cancer [1, 2]. RNA methylation, similar to DNA or protein modification, is regulated by different types of regulators, including methyltransferases (‘writers’), RNA binding proteins (‘readers’), and demethylases (‘erasers’). Discovery of these different m6A regulators has dramatically increased our understanding of the role of RNA methylation in the regulation of gene expression [3, 4]. In addition, m6A perturbations mediated by these regulators have been shown to dysregulate cell death and cell proliferation, contributing to multiple different human diseases [5, 6]. A comprehensive understanding of the genetic alterations and expression perturbations underlying cancer cell heterogeneity is necessary to elucidate RNA methylation-based therapeutic targets.
In this study, we aimed to systematically characterize the molecular alterations and clinical relevance of m6A RNA regulators across 33 cancer types . We found that there exist widespread genetic alterations (including mutations and copy number variations) in m6A regulators across cancer types. We also aimed to assess whether perturbations in the expression of m6A regulators was correlated with the activity of cancer pathways. Moreover, we explored the clinical prognostic value of m6A regulators, and found that m6A regulators are potentially useful markers for prognostic stratification. Our analysis highlights the importance of m6A regulators in cancer development, and lays a foundation for the development of therapeutic strategies based on RNA methylation.
Results and discussion
Widespread genetic alterations of m6A regulators across cancer types
The numbers of m6A regulators have been identified, and they can be broadly classified as readers, writers, and erasers (Fig. 1a). We reviewed the literature and curated a catalog of 20 genes that function mainly as regulators of RNA methylation (Fig. 1b), including 11 readers, seven writers, and two erasers. We first determined the prevalence of m6A regulator alterations across 33 cancer types (Additional file 1: Table S1) by integrating data on somatic mutations and copy number variations (CNVs). The overall average mutation frequency of m6A regulators was low, ranging from 0.02–8.07% (Fig. 1c and Additional file 1: Table S2). Cancer types with a higher global mutation burden (such as UCEC and SKCM) also exhibited a higher mutation frequency in m6A regulators. We found that YTHDC1, IGF2BP1, YTHDC2 and FTO showed higher mutation frequencies (Fig. 1c). Moreover, we found that several cancer types exhibited relatively few mutations in m6A regulators compared to other cancers, such as PCPG, THCA and UVM. We next collected the mutation data for 967 cell lines across 23 cancers from the Cancer Cell Line Encyclopedia (CCLE) and 652 cell lines across 22 cancers from the Genomics of Drug Sensitivity in Cancer (GDSC) database. We found that the writers had relatively high mutation frequencies across cancer types (Additional file 2: Figure S1).
We next investigated the CNV alteration frequency for all m6A regulators, and found that CNV alterations are prevalent. IGF2BP1/2/3, YTHDF1/3, HNRNPA2B1, and VIRMA showed widespread CNV amplification across cancer types (Fig. 1d and Additional file 1: Table S3). In contrast, YTHDC1/2, METTL14, RBM15B, and ALKBH5 had prevalent CNV deletions (Fig. 1d and Additional file 1: Table S4). There were also prevalent CNV alterations in m6A regulators across cell lines (Additional file 2: Figure S2). An intriguing question is whether these genetic alterations affect the expression of m6A regulators. We therefore explored the expression perturbations of m6A regulators across 17 cancer types with at least five normal controls. We found that CNV alterations are most likely one of the prominent mechanisms leading to perturbations in the expression of m6A regulators (Fig. 1e). The m6A regulators with CNV amplification showed significantly higher expression in cancer cells when compared to normal cells (e.g. IGF2BP1 and IGF2BP3), while the regulators with CNV deletion showed significantly lower expression (e.g. METTL14 and ALKBH5). In particular, we found that IGF2BP3 showed significantly higher expression in 15 cancer types (Fig. 1f). Moreover, we analyzed the expression of m6A regulators across another ~ 7400 samples, representing 11 cancer types, and found that IGF2BP3 also showed higher expression in cancer cells (Additional file 2: Figure S3). These results reveal a highly heterogeneous genetic and expression alteration landscape of m6A regulators across cancer types, suggesting that m6A regulator dysregulation is of importance in different cancer contexts.
Oncogenic pathways regulated by m6A regulators
To further understand the molecular mechanisms by which m6A regulators are involved in cancer, we examined the correlation between the expression of individual m6A regulators and the activity of 50 cancer hallmark-related pathways (Additional file 2: Methods). We found that the expression of m6A regulators is correlated with the activation or inhibition of multiple oncogenic pathways (Fig. 2a and Additional file 1: Table S5). The expression of YTHDF2, RBMX, and RBM15 was correlated with a higher number of activated pathways, such as the PI3K-AKT-MTOR, G2M checkpoint, and P53 pathways. In particular, we found that the m6A reader YTHDF3 was correlated with the activation of several pathways (Fig. 2b), including protein secretion (in 18/33 cancers), androgen response (in 8/33 cancers) and the TGFb signaling pathway (in 6/33 cancers). In contrast, we found that the expression of HNRPA2B1 was correlated with the activity of several cancer pathways (Fig. 2b), including the E2F, G2M checkpoint, and KRAS signaling pathways. These correlations were further validated using HNRPA2B1 knockdown experiments (Additional file 2: Figure S4). In addition, different readers, writers, or erasers were associated with distinct cancer pathway alterations, suggesting different functional effects of m6A regulators within the same functional class.
Moreover, genes do not function in isolation, and evidence has shown that collaboration among writers, erasers, and readers exists in the context of cancer [8, 9]. We thus investigated the co-occurrence of genetic alterations and expression correlation among m6A regulators. We found not only that genes within the same functional class showed significant co-occurrences of genetic alterations and highly correlated expression patterns (Additional file 2: Figure S5), but that a high correlation also existed among writers, erasers, and readers (Fig. 2c). For instance, the reader YTHDC1 was significantly correlated with writers, such as METTL3 and METTL14. We also found that there were higher correlations among genes in the same protein complex, such as RBM15 and WTAP of the spliceosome complex (Fig. 2c, R = 0.51 and p-value < 2.2E-16). Moreover, we found that these writers, erasers, and readers interacted with each other frequently in protein-protein interaction networks (Fig. 2d). There was an especially high number of interactions among the writers. Taken together, these results suggest that cross-talk among the writers, readers, and erasers of RNA methylation, play critical roles in the development and progression of different types of cancers.
Clinical relevance of m6A regulators across cancer types
The prevalent genetic and expression alterations in m6A regulators in various types of cancers, may provide important insight into translational medicine developments. First, we found that m6A regulators were more likely to be essential genes across cell lines, suggesting that they play critical roles in cell growth (Additional file 2: Figure S6). Next, we focused on m6A regulators that showed significant association with patient survival across the 33 cancer types. We found that all of the m6A regulators were associated with the overall survival of patients in at least one cancer type (Fig. 3a). Several m6A regulator genes showed oncogenic features, such as IGF2BP1 and IGF2BP3, and higher expression of these genes was associated with worse survival across cancer types. In particular, high expression of IGF2BP3 was correlated with worse survival in 13 cancer types (Fig. 3b), including KIRC (log-rank p = 2.11E-7), KIRP (log-rank p = 5.68E-8), and LGG (log-rank p = 1.16E-9). Moreover, we collected another 13 datasets across seven tissues from Gene Expression Omnibus (GEO), and found that high expression of IGF2BP3 was associated with poor patient survival (Fig. 3b and Additional file 2: Figure S7). These observations indicate that IGF2BP3 might function as an oncogene across cancer types. In contrast, we found that several m6A regulators also showed features of tumor suppressors, such as METTL14. Higher expression of METTL14 was significantly associated with better survival in seven cancer types (Fig. 3a and Additional file 2: Figure S8).
Moreover, we found more m6A regulators that were correlated with patient survival in two types of kidney cancer. We thus explored whether the expression of m6A regulators could contribute to the stratification of kidney cancer. Based on the global expression pattern of m6A regulators, we identified two subgroups of kidney cancer patients (Fig. 3c). The first subgroup consisted of 572 patients that showed higher expression of m6A regulators (RM-high), and the second of 223 patients with low expression (RM-low). Compared to the RM-low subgroup, patients in the RM-high subgroup had significantly better survival rates (Fig. 3d, log-rank p = 0.005). To further understand the clinical implications of m6A regulators, we examined the correlation between m6A regulators and 150 clinically actionable genes , and observed that m6A regulators frequently interacted with these genes (Additional file 2: Figure S9). Moreover, we manually searched the literature and found that the majority of these regulators have been found to play critical roles in cell growth, proliferation, and metastasis (Additional file 1: Table S6). However, the function of several regulators still require further validation in low throughput experiments. Together, these results suggest a diverse potential of m6A regulators in the prognostic stratification of specific types of cancer and in the development of novel treatment strategies.
We have demonstrated the prevalent genetic and expression alterations of RNA methylation regulators across cancer types. These m6A regulators are tightly correlated with the activation and inhibition of cancer pathways, and are also correlated with prognostically relevant tumor subtypes. In conclusion, this systematic analysis of the landscape of molecular alterations and clinical relevance of m6A regulators lays a critical foundation for understanding the dysregulation of RNA methylation. It will also provide insights into the development of related therapeutic targets.
Availability of data and materials
The gene expression profiles and clinical data can be found at the GDC portal (https://portal.gdc.cancer.gov/). The mutation data can be downloaded from Synapse (https://www.synapse.org/#!Synapse:syn4977808). The copy number data were downloaded from Broad GDAC Firehose (https://gdac.broadinstitute.org/). Software and resources used for the analyses are described in each method section. All results generated in this study can be found in supplementary tables.
Copy number variation
Kidney Renal Clear Cell Carcinoma
Kidney renal papillary cell carcinoma
Brain low grade glioma
Methylation of N6 adenosine
Pheochromocytoma and Paraganglioma
Skin Cutaneous Melanoma
Uterine Corpus Endometrial Carcinoma
Yang Y, Hsu PJ, Chen YS, Yang YG. Dynamic transcriptomic m (6) a decoration: writers, erasers, readers and functions in RNA metabolism. Cell Res. 2018;28:616–24.
Wang S, Chai P, Jia R, Jia R. Novel insights on m (6) a RNA methylation in tumorigenesis: a double-edged sword. Mol Cancer. 2018;17:101.
Pinello N, Sun S, Wong JJ. Aberrant expression of enzymes regulating m (6) a mRNA methylation: implication in cancer. Cancer Biol Med. 2018;15:323–34.
Roundtree IA, Evans ME, Pan T, He C. Dynamic RNA modifications in gene expression regulation. Cell. 2017;169:1187–200.
Fu Y, Dominissini D, Rechavi G, He C. Gene expression regulation mediated through reversible m (6) a RNA methylation. Nat Rev Genet. 2014;15:293–306.
Delaunay S, Frye M. RNA modifications regulating cell fate in cancer. Nat Cell Biol. 2019;21:552–9.
Hoadley KA, Yau C, Hinoue T, Wolf DM, Lazar AJ, Drill E, Shen R, Taylor AM, Cherniack AD, Thorsson V, et al. Cell-of-origin patterns dominate the molecular classification of 10,000 tumors from 33 types of Cancer. Cell. 2018;173:291–304 e296.
Panneerdoss S, Eedunuri VK, Yadav P, Timilsina S, Rajamanickam S, Viswanadhapalli S, Abdelfattah N, Onyeagucha BC, Cui X, Lai Z, et al. Cross-talk among writers, readers, and erasers of m (6) a regulates cancer growth and progression. Sci Adv. 2018;4:eaar8263.
Barabasi AL, Gulbahce N, Loscalzo J. Network medicine: a network-based approach to human disease. Nat Rev Genet. 2011;12:56–68.
Li J, Han L, Roebuck P, Diao L, Liu L, Yuan Y, Weinstein JN, Liang H. TANRIC: an interactive open platform to explore the function of lncRNAs in Cancer. Cancer Res. 2015;75:3728–37.
The results here are in whole based upon data generated by the TCGA Research Network.
This work was supported by the National Natural Science Foundation of China (grants 31571331, 31871338, 61603116, 31970646), Natural Science Foundation for Distinguished Young Scholars of Heilongjiang Province (JQ2019C004) and University Nursing Program for Young Scholars with Creative Talents in Heilongjiang Province. The funders played no roles in the study design, decision to publish, or preparation of the manuscript.
Ethics approval and consent to participate
Patient data we used were acquired by publicly available datasets that were collected with patients’ informed consent.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.