Methyltransferase like 3 promotes colorectal cancer proliferation by stabilizing CCNE1 mRNA in an m6A‐dependent manner

Abstract m6A modification is the most prevalent RNA modification in eukaryotes. As the critical N6‐methyladenosine (m6A) methyltransferase, the roles of methyltransferase like 3 (METTL3) in colorectal cancer (CRC) are controversial. Here, we confirmed that METTL3, a critical m6A methyltransferase, could facilitate CRC progression in vitro and in vivo. Further, we found METTL3 promoted CRC cell proliferation by methylating the m6A site in 3′‐untranslated region (UTR) of CCNE1 mRNA to stabilize it. Moreover, we found butyrate, a classical intestinal microbial metabolite, could down‐regulate the expression of METTL3 and related cyclin E1 to inhibit CRC development. METTL3 promotes CRC proliferation by stabilizing CCNE1 mRNA in an m6A‐dependent manner, representing a promising therapeutic strategy for the treatment of CRC.


| INTRODUC TI ON
Colorectal cancer (CRC) is one of the frequent cancers in the world. It remains the second leading cause of cancer-related deaths with high morbidity. 1 For early diagnosis and reasonable treatment of CRC, the 5-year survival rate can reach to 90%. 2 However, for CRC patients with advanced metastases, the 5-year survival even reduces to 8%. 2 Therefore, it is essential to understand how CRC develops and find vital therapeutic targets for the treatment of CRC.
Some studies reported that METTL3 played an oncogenic role in myeloid leukaemia, 29 liver cancer, 28 breast cancer, 30 glioblastoma, 31 bladder cancer 11 and lung cancer. 33 Other studies indicated that METTL3 played a tumour suppressor in renal cell carcinoma 34 and glioblastoma. 35 These studies suggested the effect of METTL3 on tumour development may be tumour-specific. Abnormal cell proliferation is the main characteristic in cancer cells, which damage the genes that directly regulate their cell cycles. Carcinogenesis made their greatest effect by targeting the regulators of G1 phase progression. 36 It was found that METTL3 could participate in tumour growth and progression by regulating the cell cycle of cancer cells.
For example, METTL3 was identified as an essential oncogene to promote cell cycles and growth in acute myeloid leukaemia by targeting transcription factor SP1. 29 METTL3 acted as a tumour suppressor in renal cell carcinoma cells to regulate cell cycle arrest in G1 phase by targeting PI3K/Akt/mTOR signalling pathway. 34 In CRC, a recent research has verified that METTL3 could promote cell self-renewal, stem cell frequency and migration through an m6A-IGF2BP2-dependent mechanism. 37 However, another study showed that METTL3 could suppress colorectal cancer proliferation and migration through p38/ERK pathways. 38 Therefore, the role of METTL3 in CRC might be controversial and need to be further explored. In our present study, we confirmed that METTL3 could promote CRC cells proliferation and revealed a new mechanism by directly regulating cyclin proteins in an m6A-dependent manner.
Human intestinal microbiota can promote or suppress CRC, not only due to the carcinogenic activities of pathogenic bacterium but also caused by the complex effect of wider microbial community, particularly their metabolites, such as deoxycholic acid (DCA), ursodeoxycholic acid and butyrate. 39 Recent studies demonstrated that microbiomes have strong effects on the m6A modification and METTL3 expression in mouse brain and intestine. 40 However, whether human intestinal microbiota and their metabolites can affect the CRC m6A modification was still unknown. As a classical intestinal microbial metabolite, butyrate is an abundant (up to >10 mmol/L) short-chain fatty acid that is transported into the colonic epithelium and localizes within two subcellular compartments. 41,42 It undergoes β-oxidation inside the mitochondria and accounts for ≥70% of the energy used by normal colonocytes, 43 and it also functions as a histone deacetylase (HDAC) inhibitor inside the nucleus to epigenetically regulate gene expression. 44,45 In a study by Scheppach et al, 46 human colonic biopsies were exposed to butyrate ex vivo for 4 hours, which revealed that butyrate increased the proliferation rate at the basal 60% area of the crypt. 47 Butyrate also has the function of resisting inflammation and tumours. [48][49][50][51][52] In the present study, we found that METTL3 was up-regulated in CRC tissues and associated with poor survival. It promoted CRC cell proliferation by targeting cyclin E1 by methylating the m6A site in 3′-untranslated region (UTR) of CCNE1 mRNA. Moreover, we found butyrate could down-regulate the expression of METTL3 and related cyclin E1, inhibiting CRC development. All cells were cultured in DMEM medium supplemented with 10% foetal bovine serum (Bioind), 1% penicillin-streptomycin (Invitrogen) and maintained at 37°C with 5% CO 2 in a humidified atmosphere.

| Transfection
Lentivirus constructing of METTL3 knockdown or overexpression was obtained from Obio Technology Corp. Cells were plated in six wells dishes at 50% confluence and infected with METTL3 overexpression lentivirus (termed as METTL3), a negative control (termed as NC), METTL3 knockdown lentivirus (termed as shMETTL3#1, shMETTL3#2) and a scramble control (termed as shNC) in HT29 and LoVo cells. Pools of stable transfections were generated by selection using puromycin (4 μg/mL) for 2 weeks. CCNE1 plasmids and negative control were obtained from GeneCopoeia (GeneCopoeia). Transfections were performed using the Lipofectamine 3000 kit (Invitrogen) according to the manufacturer's instructions.

| RNA isolation and quantitative real-time PCR (qRT-PCR)
Total RNAs were extracted from tissues and cell lines using Trizol reagent (Invitrogen

| Cell proliferation and colony formation assay
HT29 and LoVo cells were seeded on a 96-well plate with the density of 2000/3000 cells per well. Cell proliferation was assessed using cell counting kit-8 (Obio Technology Corp) after 1, 2, 3, 4 and 5 days.
After 1 hour, the absorbance at 450 nm was read on the microplate reader at 37°C. For colony formation assay, 300 cells were seeded into 6-well plates and maintained in DMEM medium containing 10% FBS for 10-14 days. The colonies were fixed in 4% paraformaldehyde for 30 minutes and stained with 0.1% crystal violet. Then visible colonies were photographed and counted.

| Xenograft formation in vivo
The animal study was approved by the Animal Research Ethics

| Immunohistochemistry analysis
Colorectal cancer tissues and their adjacent nontumorous samples were obtained from patients of the First Affiliated Hospital of Nanjing Medical University. All specimens were wrapped in paraffin and then sectioned. The sections were subjected to heat-induced antigen retrieval, followed by blocking with 5% BSA solution and incubated with the indicated primary antibodies overnight at 4°C.
Subsequently, the sections were incubated with secondary antibody at 37°C for 1 hour and stained with diaminobenzidine and counterstained by haematoxylin. The IHC localization was scored in a semi-quantitative fashion incorporating both the intensity and distribution of specific staining. Survival curves were generated using

Primers name
Sequence (5′-3′) the Kaplan-Meier method and compared using the log-rank test. All patients are involved in the calculation.

| Methylated RNA immunoprecipitation
The protocol of methylated RNA immunoprecipitation (MeRIP) was referred to Meng's study. 54

| Luciferase reporter assay
Stable METTL3 knockdown (shMETTL3) or METTL3 control (shNC) HT29 and LoVo cells were co-transfected with plasmids containing 3′-UTR of wild or mutant fragments from CCNE1 using Lipofectamine 3000 (Invitrogen) according to the manufacturer's protocols. Luciferase activity was measured using the dual-luciferase reporter assay system (Promega) after 48 hours of incubation.
Finally, relative luciferase activity was normalized to the renilla and each assay was repeated in three independent experiments.

| RNA stability assay
To

| Bioinformatic analysis
Clinical data for bioinformatic analysis were downloaded from da-

| Butyrate treatment
To observe the effect of butyrate on CRC cells, endogenous butyrate (2 mmol/L or 4 mmol/L) was added into the CRC cell culture medium.
Total RNAs or proteins were harvested at the times indicated, and then subjected to qRT-PCR analysis or Western blot. Cell proliferation assay and colony formation were also taken after butyrate treatment.

| Statistical analysis
All data were presented as the means ± standard deviation. The comparisons between the groups were analysed by Student's t test.
Statistical analysis was performed using the SPSS 19.0 software (SPSS), and graphical presentations were conducted with GraphPad Prism 5 software. Pearson's correlation coefficient analysis was used to analyse the correlations. P < .05 was considered statistically significant. Survival package was used to analyse survival from GEO database. showed that high METTL3 expression worsened overall prognosis in CRC patients ( Figure S1B).

| METTL3 promoted CRC cells proliferation in vitro and in vivo
First of all, qRT-PCR and Western blot analysis revealed that METTL3 was stably knockdown and overexpressed in HT29 and LoVo cells ( Figure S2). CCK-8 and colony formation assays indicated that METTL3 knockdown significantly inhibited CRC cells proliferation and colony formation, whereas METTL3 overexpression promoted CRC cells proliferation and colony formation (Figure 2A-D). We can find OD values of METTL3 overexpression group were higher than that of NC group, and it was significant only in the Day 5. We thought it may be due to the oncogenic activity of METTL3 in CRC. The function of its overexpression in cells may be not as obvious as that in its knockdown. Cell cycle assays indicated that METTL3 knockdown induced G1-S cell cycle arrest, and METTL3 overexpression decreased percentage of G1 phase ( Figure 2E,F and Figure S3A,B). In subcutaneous implantation experiment, METTL3 knockdown could inhibit CRC tumour size and weight in mice ( Figure 2G and Figure S4).

| METTL3 regulated cyclin E1 expression in CRC cells and correlated with CCNE1 expression in human CRC tissues
To investigate the potential mechanism of CRC cells proliferation regulated by METTL3, we used Western blot and qRT-PCR to investigate the expression of cell cycle-related proteins, cyclin D1, cyclin E1, CDK2, CDK4 and CDK6 by the knockdown or overexpression of METTL3. We found that knockdown of METTL3 could prominently decrease cyclin E1 and CCNE1 expression ( Figure 3A,D and Figure   S5C), whereas METTL3 overexpression increasing cyclin E1 and CCNE1 expression, compared with the control groups ( Figure 3B,D and Figure S5C). It indicated that METTL3 could regulate cyclin E1 and CCNE1 expression in CRC cells. We also found that the expression of cyclin E1 significantly decreased in METTL3 knockdown CRC xenografts tissues in nude mice ( Figure 3C). Moreover, CCNE1 expression was up-regulated in CRC tissues ( Figure 3E) and positively correlated with METTL3 expression in CRC tissues and adjacent noncancerous tissues by qRT-PCR and IHC ( Figure 3F and Figure S7).
Furthermore, TCGA database analysis reached the same results that CCNE1 was positively correlated with METTL3 expression in human CRC tissues ( Figure S6).

| METTL3 exerted tumour-promoting functions in CRC by regulating cyclin E1 expression
To assess the potential function of cyclin E1 in the tumour-promoting functions of METTL3 in CRC, we transfected CCNE1 plasmids into HT29 and LoVo cells ( Figure 3G). As shown in Figure 3H Seq from Lin's study identified that the m6A level in CCNE1 mRNA could be affected by METTL3. 55 Our further MeRIP assays confirmed that METTL3 knockdown caused a significant decrease in the m6A levels of CCNE1 mRNA in LoVo cells ( Figure 4E). It was reported m6A sites existed mostly in the 3′-UTRs of mRNAs, 56 which can affect mRNA stability. 57 We also found a METTL3 catalysing motif site (GGACU) in the 3′-UTR of CCNE1 mRNA. So, we further conducted dual-luciferase assays to elucidate whether this m6A site was required for the CCNE1 mRNA stability. We  Figure   S8A). The similar effect of butyrate was also observed in HT29 cells ( Figure 5B). Moreover, METTL3 and CCNE1 mRNAs were both decreased in HT29 and LoVo cells significantly, when treated by butyrate (4 mmol/L) for 48 hours ( Figure 5C and Figure S8B). In addition, METTL3 overexpression could reverse the inhibition of CRC cell growth and colony formation caused by butyrate treatment ( Figure 5D,E and Figure S8C).

| D ISCUSS I ON
In this study, we gradually confirmed that METTL3, a critical m6A methyltransferase, played an oncogenic role in CRC. 37 45,73,74 Interestingly, cancer cells appear to be more sensitive to the actions of HDACi than nontransformed cells, but the mechanistic basis for this apparent selectivity is poorly understood. 75 Butyrate participates in the m6A modification and METTL3 expression may be conducted by one or more pathway mentioned above. It would be investigated in the further study.
In summary, our work confirmed Li's study that METTL3, a critical m6A methyltransferase, could facilitates CRC progression, and further revealed a new mechanism that METTL3 could promote CRC cells proliferation by directly stabilizing CCNE1 mRNA in m6A-dependent manner, representing a promising therapeutic strategy for the treatment of CRC. Moreover, the effects on m6A modification and expression of METTL3 may be involved in CRC cell proliferation affected by butyrate ( Figure 6).

CO N FLI C T S O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data sets used and analysed in the current study are available from the corresponding author in response to reasonable requests.