METTL3's role in cervical cancer development through m6A modification

N6‐methylated adenosine (m6A) is a crucial RNA modification in eukaryotes, particularly in cancer. However, its role in cervical cancer (CC) is unclear. We aimed to elucidate the part of m6A in CC by analyzing methyltransferase‐like 3 (METTL3) expression, identifying downstream targets, and exploring the underlying mechanism. We assessed METTL3 expression in CC using western blotting, quantitative polymerase chain reaction (qPCR), and immunohistochemistry. In vitro and in vivo experiments examined METTL3's role in CC. We employed RNA sequencing, methylated RNA immunoprecipitation sequencing, qPCR, and RNA immunoprecipitation qPCR to explore METTL3's mechanism in CC. METTL3 expression was upregulated in CC, promoting cell proliferation and metastasis. METTL3 knockdown inhibited human cervical cancer by inactivating AKT/mTOR signaling pathway. METTL3‐mediated m6A modification was observed in CC cells, targeting phosphodiesterase 3A (PDE3A). METTL3 catalyzed m6A modification on PDE3A mRNA through YTH domain family protein 3 (YTHDF3). Our study indicated the mechanism of m6A modification in CC and suggested the METTL3/YTHDF3/PDE3A axis as a potential clinical target for CC treatment.


| INTRODUCTION
Cervical cancer (CC) is one of the most common malignancies in women worldwide, with approximately 500 000 new cases diagnosed yearly. 1,2CC exhibits no evident clinical manifestation in the early stage, and is often diagnosed late, missing the optimal treatment window. 3The occurrence and progress of CC involve multiple "omics" interactions. 4Although considerable effort has been invested in treating CC, the effectiveness of current therapies remains limited.Therefore, exploring the mechanism underlying CC onset and progression is important for improving treatment strategies.
The methylation of the sixth nitrogen (N) atom of the adenine nucleotide (m 6 A methylation) is catalyzed by the methyltransferase complex.This modification is frequent in the RRACH (R = guanine [G] or A, H = A, cytosine [C], or uracil [U]) motif in RNA molecules.m 6 A methylation functions in various physiological processes, including cell differentiation, circadian rhythm, and development. 5][8] m 6 A modifications significantly affect CC cell metastasis; however, the underlying mechanism remains unknown.Exploring the potential link between m 6 A modification and CC will expand our understanding of RNA modification-related processes and present innovative avenues for diagnosing CC.
Phosphodiesterase 3A (PDE3A) belongs to the cyclic nucleotide phosphodiesterase 3 (PDE3) family and is primarily expressed in the heart, pancreas, endometrium, fallopian tube, oocyte, kidney, and other organs. 9,10DE3A is crucial in cancer development.PDE3A is connected with cell proliferation and metastasis in gastrointestinal stromal tumors. 11PDE3A arrests cells in the G0/G1 phase and inhibits cell cycle progression in malignant salivary gland cells. 12Additionally, PDE3A promotes breast cancer invasion and metastasis by inducing the inflammatory nuclear factor kappa-light-chainenhancer of activated B cells (NF-κB) signaling pathway by inhibiting cyclic adenosine monophosphate/protein kinase A. 13 Numerous PDE inhibitors are currently in clinical use and trials for colorectal, ovarian, and endometrial cancers. 14However, no study has explored PDE3A expression in CC.
Therefore, our study aimed to reveal the role of N6methylated adenosine in CC by analyzing the clinical significance of METTL3 expression.We aimed to explore the downstream target of METTL3 and elucidate the intricate mechanism underlying their interaction.Collectively, this study revealed novel findings on the METTL3/YTHDF3/PDE3A axis, providing promising insights into the development of new strategies for targeted CC therapy.

| Cell culture
The SiHa and C33A cell lines were procured from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China).Then, they were cultured in highglucose Dulbecco's Modified Eagle Medium (DMEM, Gibco, Carlsbad, CA, USA) with 10% fetal bovine serum (FBS, Gibco).The HeLa and CaSki cell lines were gifted by the Qilu Hospital Laboratory, Shandong University.HeLa and CaSki cells were cultured in DMEM with 10% FBS.All cells are cultured in an incubator at 37°C and 5% CO 2 .

| Cell transfection
METTL3 overexpression and knockdown lentiviruses were acquired from GeneChem Co. (Shanghai, China).Stably infected cell lines were selected using puromycin (2 mg/mL; GeneChem Co.).The multiplicity of infection was 20, and transfection efficiency was verified using reverse transcription-quantitative polymerase chain reaction (RT-qPCR) and western blotting.

| RNA extraction and RT-qPCR
Total RNA was extracted using TRIzol reagent (Accurate Biology, Hunan, China), and 1 μg of RNA was reverse transcribed into cDNA using a SPARKscript II All-in-one RT SuperMix for qPCR (Sparkjade Biotechnology Co., Ltd.Shandong, China), following the manufacturer's protocols.The relative expression of METTL3 mRNA was analyzed following the 2 −ΔΔCt method. 15βactin was used as an internal control.

| Western blotting
After the cells were washed with 1x PBS, 100 μL RIPA lysis buffer (Beyotime, Beijing, China) was added for lysis.Subsequently, the cells were subjected to a bicinchoninic acid protein assay (Thermo Fisher Scientific, Waltham, MA, USA).Transfer proteins to polyvinylidene fluoride membranes and incubate overnight at 4°C with primary antibodies (1:1000).Then, incubated with horseradish peroxidase-conjugated secondary antibody (1:5000) for 1 h, and the bands were visualized using an electrochemiluminescence detection reagent (Enzyme, Nanjing, China).

| Cell growth and proliferation analysis
The viability of the cells was determined using a cell counting kit-8 (CCK8) assay.Cells stably expressing METTL3 (SiHa/CaSki) were seeded in 96-well plates at 3000 cells/well density and cultured at 37°C incubator, with three replicate wells in each group.After 0, 24, 48, and 72 h, 10% CCK8 solution (Dojindo, Tokyo, Japan) was added.The optical density at 450 nm was measured by a microplate reader.
For colony formation assays, CC cells were seeded in 6-well plates at a density of 1000 cells/well.Then cultured at a 37°C incubator for two weeks with regular medium changes.Staining was performed when the colony number reached 50, and colony formation efficiency was calculated.

| Transwell migration/invasion assay
Cells were scraped with a sterile pipette tip for the woundhealing experiment upon reaching >95% confluency.The isolated cells were washed with PBS and placed in a 37°C incubator for 48 h.Finally, photographs were captured at 0 and 48 h, and we calculated the percentage of wound closure.
For the migration assay, 4 × 10 4 cells (SiHa/CaSki) were seeded in the upper chamber of a transwell insert (Corning Inc., Corning, NY, USA), and 650 μL of DMEM containing 20% FBS was added to the lower chamber.Cells were stained with 0.1% crystal violet (Sigma-Aldrich, St. Louis, MO, USA) after incubation at 37°C for 24 h.For the invasion experiment, Matrigel (BD Biosciences, Franklin Lakes, NJ, USA) was diluted with DMEM and added into the upper chamber.A cell suspension was then added after the matrigel solidified.The rest of the experimental steps were consistent with those of the migration assay.Migrating or invading cells were then photographed under 10× magnification using a microscope (EVOS M7000; Invitrogen, Carlsbad, CA, USA).

| Methylated RNA immunoprecipitation sequencing (MeRIP-seq)
Total RNA was immunoprecipitated using a GenSeq® m 6 A-IP kit (GenSeq Inc., Shanghai, China), following the manufacturer's protocols.Briefly, RNA was fragmented into ~100nt fragments.Meanwhile, protein A/G magnetic beads were cultured with m 6 A antibodies at 4°C for 1 h.RNA-seq libraries were constructed using a GenSeq® Low Input Whole RNA Library Prep kit (GenSeq Inc.), following the manufacturer's instructions.Library sequencing was executed using an Illumina NovaSeq platform with 150 bp paired-end reads.

| RNA sequencing (RNA-seq)
Ribosomal RNA (rRNA) was isolated from samples using a GenSeq® rRNA Removal kit (GenSeq, Inc.), following the manufacturer's instructions.The GenSeq® Low Input RNA Library Prep kit (GenSeq Inc.) was then used to establish a sequencing library following the manufacturer's protocols.Quality control and quantification of the constructed sequencing library were performed by a BioAnalyzer 2100 system (Agilent Technologies, Santa Clara, CA, USA).Library sequencing was executed using an Illumina NovaSeq platform with 150 bp paired-end reads.

| MeRIP-qPCR
MeRIP-qPCR was performed using a MeRIP kit (BersinBio™, Guangzhou, China), according to the manufacturer's protocols.Briefly, total RNA was fragmented into ~100nt fragments.Subsequently, 4 μg m 6 A antibody was used for co-immunoprecipitation, and A/G magnetic beads were used to elute the m 6 A-modified fragments.Finally, RNA enrichment was assessed using real-time qPCR.

| RIP-qPCR
RIP was conducted using a GenSeq® RIP kit (GenSeq, Inc.), following the manufacturer's protocols.In brief, SiHa cells were lysed and incubated with antibody-conjugated magnetic beads overnight at 4°C.After washing and purification, the RNA expression of the target gene was quantified using RT-qPCR.

| mRNA stability
Act-D (5 mg/mL) was added to each group of cells.After 0, 5, and 10 h, the cells were washed with pre-cooled PBS.RNA extraction and RT-qPCR were performed to detect PDE3A mRNA expression.

| Animal assays
Female BALB/c nude mice (4-6 weeks old, Vital River Laboratory Animal Technology Co., Ltd.Beijing, China) were randomly divided into two groups (n = 6 per group) and cultured in a specific pathogen-free environment.A suspension of CaSki cells stably expressing Lv-mock or Lv-METTL3 (~3 × 10 6 cells) was injected into the right axilla of the mice.The mice were subjected to regular weight and tumor size measurements every three days to monitor growth.Twenty-four days after injection (when the tumor size was ~1 cm), the mice were anesthetized with tribromoethanol (Avertin), and the tumors were harvested.The tumor volume was calculated as follows: V (mm 3 ) = 0.5 × L × W 2 .The tumor tissues were fixed for subsequent experiments.

| Statistical analysis
All statistical analyses were performed using GraphPad Prism software (GraphPad Software Inc., La Jolla, CA, USA).The Kaplan-Meier method and log-rank test were used to calculate the survival curves of patients with CC.Western blot band intensities were quantified via ImageJ software (version: 2.3.0).Data are expressed as the mean ± standard deviation of at least three independent experiments.Student's t-test was used to determine statistical differences between two groups.Statistical significance was set at p < .05.

| METTL3 is highly expressed in CC tissues and promotes CC progression
We performed a bioinformatics analysis of METTL3 expression using CC samples and the Gene Expression Profiling Interactive Analysis (GEPIA) database (http:// gepia.cance r-pku.cn/ index.html).Tumor samples had higher METTL3 expression than normal samples (Figure 1A), indicating that METTL3 is involved in CC progression.Immunohistochemistry (IHC) analysis showed that CC tissues had remarkably increased METTL3 expression than adjacent non-tumor tissues (Figure 1B).Western blotting indicated that CC tissues had higher METTL3 expression than adjacent non-tumor tissues (Figure 1C).Furthermore, METTL3 mRNA and protein were highly expressed in all CC cell lines investigated (Figure 1D, E).These results reveal that METTL3 plays a role in the progression of CC.
We selected SiHa, C33A, and CaSki cell lines for subsequent experiments.We constructed an overexpression/ silencing lentivirus targeting METTL3 and evaluated its efficiency in overexpressing/silencing METTL3 at the mRNA and protein levels (Figure 1F-I).RT-qPCR indicated that METTL3 mRNA expression was notably reduced in SiHa and C33A cells (p < .05)(Figure 1F).Hence, shRNA-3 was used in subsequent experiments.

| METTL3 promotes CC cell proliferation and metastasis in vitro
We then performed cell function experiments to ascertain the contribution of METTL3 in CC.EDU experiments indicated that METTL3 overexpression increased notably the proliferation of CC cells, whereas its knockdown decreased (Figure 2A).CCK8 experiments showed that METTL3 overexpression promoted cell viability, whereas its inhibition had the opposite effect (Figure 2B).The results of the colony formation experiments had the same trend with those of the EDU experiments (Figure 2C).Additionally, we demonstrated the effect of METTL3 on CC cell metastasis by scratch and transwell experiments, showing consistent results.METTL3 overexpression promoted cell metastasis and wound-healing abilities, whereas METTL3 knockdown showed the opposite effect (Figure 3A,B).Cell cycle experiments showed that METTL3 knockdown induced cell cycle arrest in the G0/G1 phase (Figure 4A).Therefore, METTL3 may promote the development of CC in vitro.

| METTL3 overexpression promotes CC tumor growth in nude mice
Given the in vitro oncogenic effect of METTL3 on CC cells, we conducted an in vivo assay to explore the involvement of METTL3 in CC tumor growth.METTL3 overexpression remarkably promoted tumor growth (volume and weight) compared to the blank control group (Figure 4B-D).Western blotting revealed that tumor tissues had notably higher METTL3 expression than the controls (Figure 4E).These findings reveal that METTL3 overexpression promotes CC tumor growth in vivo.

| METTL3 knockdown inhibits human cervical cancer by inactivating the AKT/mTOR signaling pathway
Emerging evidence shows that the AKT/mTOR pathway is a key signaling pathway in the regulation of various biological processes and increases cell growth, 16 survival, 17 and apoptosis. 18Here, we explored whether METTL3 knockdown affected the AKT/mTOR signaling pathway using western blotting experiments.Compared with the NC group, METTL3 knockdown decreased the levels of p-AKT and p-mTOR and the expression of downstream effector p70S6K (Figure 5A), suggesting that the AKT/ mTOR signaling pathway was inactivated.Furthermore, to investigate whether the phenotypic cell changes induced by METTL3 knockdown involved the AKT/mTOR pathway, a rescue experiment was performed using an AKT pathway activator, IGF-1.The results of the CCK8 and transwell assays demonstrated that IGF-1 could reverse the inhibitory effects of METTL3 knockdown on the proliferation and invasion of human cervical cancer cells (Figure 5B,C).Overall, the data indicate that the pro-oncogenic function of METTL3 in human cervical cancer cells was mediated by the AKT/mTOR pathway.

| MeRIP-seq combined with RNA-seq identified PDE3A as a target of METTL3
To investigate potential target genes of METTL3 in CC, we conducted RNA-seq to analyze changes in the transcriptome of SiHa cells.Following METTL3 knockdown, 7556 7526 genes were downregulated and upregulated, respectively (Figure 6A,B).To explore whether this resulted from METTL3-mediated m 6 A modification, we conducted MeRIP-seq and observed a reduction in 20 446 peaks (Figure 6C).Gene Ontology analysis revealed that the differentially expressed m 6 A-modified transcripts were enriched in genes associated with trigeminal nerve development, axon guidance, and the regulation of synaptic structure modification, revealing the possible involvement of m 6 A in neuronal meta-regulation (Figure 6D).Pathway analysis revealed that the transforming growth factor-beta signaling pathway, prostate cancer, FoxO signaling pathway, and transcriptional dysregulation in cancer are linked to METTL3-mediated m 6 A modification (Figure 6E), corroborating the role of METTL3 in CC tumorigenesis.MeRIP-seq analysis suggested that the m 6 A peaks were distributed in the start, stop, and coding sequence regions (Figure 6F); the m 6 A peak was mainly located near the stop codon (Figure 6G).
The results of RNA-seq combined with MeRIP-seq are illustrated in Figure 7A.We identified 87 transcriptionally downregulated and hypomethylated (p < .05)and 631 transcriptionally upregulated and hypermethylated m 6 A genes (p < .05).Given the possible role of METTL3 in CC, mRNA transcripts exhibiting reduced methylation following METTL3 knockdown may be viable targets.Among the downregulated transcripts, we selected four genes (NTR4A3, COL5A1, PDE3A, and NFATC2) closely related to cancer development for subsequent experimental verification (Figure 7B).Furthermore, the Integrative Genomics Viewer tool (IGV) revealed that the methylation peaks of these genes were downregulated compared to the controls (Figure 7C).
To validate the MeRIP-seq and RNA-seq results, we conducted MeRIP-qPCR and RT-qPCR analyses.The methylation and mRNA levels of PDE3A, NR4A3, and COL5A1 were notably downregulated (Figure 7D,E).However, NFATC2 methylation levels were inversely correlated with its mRNA levels, suggesting that METTL3 may be involved in other processes, such as the degradation of NFATC2.Next, we confirmed that METTL3 directly interacts with these genes using RIP-qPCR (Figure 7F).Only PDE3A a notably high enrichment factor (≥2000), warranting further investigation.MeRIP-seq revealed that the consensus motif of METTL3 was recognized during m 6 A modification (GGACC) of the 3′ untranslated region (UTR) of PDE3A mRNA (Figure 7G).Therefore, PDE3A may be a potential target gene for METTL3-mediated m 6 A modification.

| METTL3 exerts oncogenic effects by targeting PDE3A
To investigate the mechanism underlying the METTL3mediated regulation of PDE3A expression, we silenced or overexpressed METTL3 in SiHa and CaSki cells.PDE3A expression showed a consistent trend at both the protein and mRNA levels (Figure 8A,B).We subsequently evaluated PDE3A expression in CC tissues and corresponding controls through IHC.CC tumors had remarkably higher PDE3A expression than the controls (Figure 8C).Next, we constructed a PDE3A overexpression plasmid and investigated its overexpression efficiency using RT-qPCR and western blotting (Figure 8D,E).PDE3A mRNA and protein were notably overexpressed.We then conducted colony formation and transwell assays to demonstrate the effect of PDE3A on the functional phenotype of CC cells.The colony formation and metastatic abilities of CC cells were increased notably when PDE3A was overexpressed (Figure 8F,G).
Subsequently, we conducted a rescue assay to demonstrate the role of PDE3A and METTL3 in CC development.Both colony formation and transwell assays indicated that PDE3A overexpression restored the METTL3 knockdown-induced attenuation of the colony-forming and metastatic abilities of SiHa cells (Figure 8H,I).These results suggest that PDE3A mediates the oncogenic processes induced by METTL3 in CC cells and METTL3 exerts its oncogenic effects by targeting PDE3A.

| METTL3 regulates PDE3A mRNA stability through YTHDF3
Although we have demonstrated that METTL3 targets PDE3A, the underlying mechanism by which METTL3 promotes PDE3A expression remains clear.METTL3 recruits YTHDF3 to enhance the stability of its target transcripts. 19Therefore, we speculated that METTL3 also enhances PDE3A stability through YTHDF3.First, small interfering RNAs targeting YTHDF3 were constructed, and their mRNA and protein efficiencies were determined (Figure 9A,B).The interference efficiency of the YTHDF3-3 construct was the highest, so it was selected for subsequent experiments.We then demonstrated that YTHDF3 silencing reduced PDE3A mRNA expression using RT-qPCR and found that YTHDF3 may be involved in the epigenetic regulation of PDE3A (Figure 9C).RIP-PCR revealed that YTHDF3 and PDE3A mRNA are directly bound (Figure 9D).Western blotting revealed that YTHDF3 silencing reduced PDE3A protein expression (Figure 9E).The RNA decay rate assays suggested that, compared to that in the corresponding control, the half-life of HK2 mRNA was remarkably shortened following both METTL3 and YTHDF3 knockdown (Figure 9F,G).Finally, we silenced YTHDF3 based on the overexpression of METTL3 and detected changes in PDE3A protein content using a western blot assay.Our findings indicate that the upregulation of PDE3A expression caused by METTL3 overexpression was significantly reversed by the reduction of YTHDF3 (Figure 4).This suggests that YTHDF3 plays a role in regulating PDE3A expression through m 6 A modification mediated by METTL3 (Figure 9H).Finally, we knocked down YTHDF3 after overexpressing METTL3 and detected changes in PDE3A level.The overexpression of METTL3 upregulated PDE3A expression, and this effect was significantly reversed by downregulating YTHDF3 expression (Figure 9I).The findings suggest that YTHDF3 is involved in the regulation of PDE3A expression through m6A modification mediated by METTL3.These results demonstrate that YTHDF3 recognizes methylated PDE3A mRNA and METTL3/YTHDF3 enhances its stability. 4| DISCUSSION m 6 A modifications are the most prevalent chemical alterations in RNA and play multiple roles in gene transcription. 20The involvement of m 6 A in various diseases 21,22 has been gradually unveiled, revealing its association with the development of malignant tumors. 23,24Moreover, m 6 A critically mRNA splicing, 25 stability, 26 translation efficiency, 27 and nuclear localization. 28he methyltransferase METTL3, a core component of the METTL3-METTL14-WTAP complex, is a crucial factor in tumor progression and the pathogenesis of various diseases. 1Several studies have recently demonstrated the role of m 6 A in CC.METTL3 could promote the occurrence and development of CC through IGF2BP3. 29Additionally, METTL3 promotes CC progression by inhibiting endoplasmic reticulum stress through the m 6 A modification of TXNDC5 mRNA. 30However, the specific mechanism of m 6 A action in CC remains unknown, given the diverse range of its functions.
In this study, we found notable upregulation of METTL3 in CC tissues and cells.Functional experiments indicated that METTL3 promotes the proliferation and metastasis of CC cells, suggesting its involvement in CC development.In addition, we observed that METTL3 knockdown inhibits the progression of human cervical cancer through inactivation of the AKT/mTOR signaling pathway, including decreased levels of p-AKT, p-mTOR, and p70S6K.Through RNA-seq and MeRIP-seq analyses, we identified a prominent m 6 A site in the 3′UTR of PDE3A mRNA.Silencing METTL3 significantly downregulated the methylation peak of this site, indicating that METTL3 mediates m 6 A modification of the adenosine in the GGACC sequence of the PDE3A 3′UTR, potentially leading to PDE3A mRNA degradation.Mechanistic experiments demonstrated that METTL3 regulates m 6 A methylation at the A residue within the GGACC motif of the PDE3A 3′UTR.By recognizing the m 6 A site on PDE3A mRNA, METTL3 enhances its stability and promotes PDE3A protein expression.Rescue assays further confirmed that PDE3A overexpression significantly restores the METTL3-induced inhibition of the CC phenotype, indicating that METTL3 promotes CC pathogenesis through PDE3A.
The m 6 A-mediated regulation of gene expression relies on reader proteins that recognize and bind to RNA molecules with m 6 A modifications.METTL3, the primary m 6 A writer, interacts with reader proteins, influencing transcript metabolism.For instance, in esophageal squamous cell carcinoma, 19 METTL3 increases the methylation level of EGR1 mRNA and enhances its stability in a YTHDF3-dependent manner.METTL3 is a typical m 6 A methyltransferase implicated in tumorigenesis, with high expression associated with poor prognosis in various tumors.In non-small cell lung cancer, METTL3 promotes malignancy by targeting and inhibiting SFRP2, activating the βcatenin signaling pathway. 31In endometrial cancer, METTL3 facilitates immune surveillance by inhibiting YTHDF2-mediated degradation of NLRC5 mRNA, thus attenuating immune escape cancer 32 In our study, we revealed that the interaction between METTL3 and PDE3A is mediated by the m 6 A reader protein YTHDF3.YTHDF3 recognizes the m 6 A site in PDE3A mRNA and facilitates METTL3 binding, thereby enhancing its stability.YTHDF3 is responsible for regulating RNA translation 33 and degradation rates. 34In our experiments, YTHDF3 effectively reduced PDE3A mRNA levels.Immunohistochemical staining of CC surgical specimens confirmed significantly higher levels of METTL3 and PDE3A in CC specimens compared to control specimens, consistent with the cell assay results.Together, these findings highlight the significance of METTL3 as a writer protein in the m 6 A-mediated regulation of PDE3A expression.This discovery not only expands our knowledge and addresses gaps in m 6 A-related research but also enhances our understanding of the molecular mechanisms underlying CC metastasis.
Our study had limitations, indicating that the METTL3/YTHDF3/PDE3A pathway needs to be investigated further.Although we observed that METTL3 knockdown inhibited human cervical cancer by activating the AKT/mTOR signaling pathway, we did not establish that PDE3A functions through the AKT/mTOR pathway.This aspect will be explored in subsequent experiments.Herein, we have identified METTL3 as an oncogene involved in CC development through the YTHDF3/PDE3A pathway, which presents a potential therapeutic target for human CC treatment.

| CONCLUSIONS
In summary, this study is the first to report that METTL3 promotes CC progression by targeting PDE3A via YTHDF3.Furthermore, we demonstrated the upregulation of METTL3 in CC samples, confirming its clinical relevance and suggesting its potential prognostic value.These findings imply the clinical implications of METTL3 as a prognostic marker for CC.Additionally, understanding the role of the METTL3/YTHDF3/PDE3A axis may offer new perspectives on CC-targeted therapies and prospects for advancing anticancer drugs.

F I G U R E 1
METTL3 expression in cervical cancer.(A) Expression level of the methyltransferase METTL3 based on the GEPIA database.(B) Immunohistochemistry staining images showing METTL3 expression in cervical cancer (CC) tissues and paired adjacent non-tumor tissues.(C) Western blot analysis of METTL3 expression in CC specimens and adjacent non-tumor tissues.(D, E) Protein and mRNA expression levels of METTL3 in CC cell lines (SiHa, CaSki, C33A, and HeLa) were assessed using western blotting and reverse transcriptionquantitative polymerase chain reaction (RT-qPCR), respectively.(F-I) Silencing/overexpression efficiency of transfected METTL3 lentivirus assessed using western blotting and RT-qPCR.***p < .001,**p < .01,*p < .05.F I G U R E 2 METTL3 promotes the proliferation of cervical cancer (CC) cells.(A, B) Proliferative ability of cells in which METTL3 was knocked down or overexpressed was assessed using EDU and the cell counting kit-8 assays.(C) Number of clones of CC cells was assessed using a colony formation assay.***p < .001.F I G U R E 3 METTL3 promotes the migration and invasion abilities of cervical cancer (CC) cells.(A) Wound-healing ability of CC cells in which METTL3 was knocked down or overexpressed was assessed using cell scratch experiments.(B) Migration and invasion abilities of CC cells in which METTL3 was knocked down or overexpressed were assessed using transwell experiments.***p < .001,**p < .01.F G U R E 4 METTL3 promotes tumor growth in vivo.(A) Cellular distribution at the G0/G1, S, and G2/M phases was assessed using flow cytometry.(B) Representative images of subcutaneous SiHa tumor growth in xenografted BALB/c nude mice.Each group of mice was ectopically implanted with 5 × 10 6 cancer cells into the left axilla (n = 6).Cells were transfected with a mock or overexpression lentiviral vector (MOCK/METTL3).(C) Tumor weight.Tumors were harvested 21 days after cell injection, and their weights were measured.Data are expressed as the mean ± standard error of the mean (SEM).(D) Tumor size in the mouse model was monitored every three days.Data are expressed as the mean ± SEM of the tumor volume.(E) METTL3 protein expression in the tumors of nude mice in the two groups, assessed using western blotting.βactin was used as a loading control.***p < .001,**p < .01,*p < .05.F I G U R E 5 METTL3 knockdown-induced cell phenotype changes are mediated by the AKT/mTOR pathway.(A) AKT/mTOR/p70S6K protein expression after knockdown/overexpression of METTL3, assessed using western blotting.(B) Proliferative ability of cells in which METTL3 was knocked down.IGF-1 incubation was assessed using the Cell Counting Kit-8 assays.(C) Invasion ability of cells in which METTL3 was knocked down.IGF-1 incubation was assessed using the transwell assay.***p < .001,**p < .01,*p < .05.

F I G U R E 6
Transcriptome profile regulated by METTL3 in cervical cancer (CC) cells.(A, B) Differentially expressed genes in CC cells after METTL3 knockdown compared with sh-NC controls, identified using RNA-seq.(C) m 6 A peak in CC cells after METTL3 knockdown.(D) Gene ontology analysis showing the target pathways for CC.(E) Pathway analysis in CC. (F) m 6 A peak distribution plot for SiHa cells.(G) Metagene map showing the m 6 A map in SiHa cells, including the 3′ untranslated region (UTR), coding sequence region (CDS), and the 5′UTR.

F
I G U R E 7 PDE3A is a direct target of METTL3-mediated m 6 A modification.(A) Distribution of genes with significant changes in both m 6 A and gene expression levels in SiHa cells after METTL3 silencing.(B) Candidate target genes for METTL3.(C) m 6 A peak distribution of PDE3A, NR4A3, NFATC2, and COL5A mRNA, revealed using the Integrative Genomics Viewer tool (IGV).(D) m 6 A modification of the four genes is mediated by METTL3, determined using methylated RNA immunoprecipitation sequencing.(E) Transcript modification of four genes is mediated by METTL3, detected using reverse transcription-quantitative polymerase chain reaction.(F) Enrichment of METTL3 binding to the PDE3A, NR4A3, NFATC2, and COL5A1 m 6 A modification sites, revealed using RIP-qPCR.(G) Schematics of the m 6 A motif of METTL3 and the m 6 A site in the 3′UTR of PDE3A mRNA (near the stop codon).****p < .0001,***p < .001,**p < .01,*p < .05.

F I G U R E 8
METTL3 exerts oncogenic effects by targeting PDE3A.(A) PDE3A protein expression after knockdown/overexpression of METTL3, assessed using western blotting.(B) PDE3A mRNA expression after the knockdown/overexpression of METTL3, assessed using RT-qPCR.(C) Immunohistochemistry staining of PDE3A expression in cervical cancer (CC) tissues and paired adjacent non-tumor tissues.(D, E) PDE3A mRNA and protein overexpression efficiency, verified using RT-qPCR and western blotting, respectively.(F) Effect of PDE3A overexpression on the migration and invasion ability of CC cells, assessed using transwell experiments.(G) Effect of PDE3A overexpression on the colony formation ability of CC cells, assessed using the colony formation assay.(H, I) PDE3A overexpression promotes CC progression and enhances the migration, invasion, and colony-forming abilities of CC cells after METTL3 knockdown.***p < .001,**p < .01.

F
I G U R E 9 YTHDF3 mediates the stability of PDE3A in an m 6 A-dependent manner.(A, B) YTHDF3 knockdown efficiency at both the mRNA and protein levels, verified using reverse transcription-quantitative polymerase chain reaction (RT-qPCR) and western blotting, respectively.(C) Effect of YTHDF3 silencing on PDE3A mRNA expression, assessed using RT-qPCR.(D) Direct binding of YTHDF3 and PDE3A mRNA demonstrated using RIP-PCR.(E) Effect of YTHDF3 disruption on PDE3A protein levels, elucidated using western blotting.(F) PDE3A mRNA half-life (t 1/2 ) after the silencing of METTL3 and YTHDF3.RNA stability assays were performed after incubation with actinomycin D (5 mg/mL).(G) Effect of overexpression of METTL3 and knockdown of YTDHF3 on PDE3A protein levels, elucidated using western blotting.***p< .001,**p < .01.