METTL3-mediated m⁶A modification of lncRNA MALAT1 promotes colorectal cancer progression by activating the NF-κB signaling pathway

Abstract

Background

Colorectal cancer (CRC), a prevalent malignancy of the digestive system, is often characterized by poor patient prognosis due to high rates of recurrence and metastasis. N6-methyladenosine (m⁶A) modification represents a crucial epigenetic mechanism regulating RNA (including lncRNAs) function and is increasingly implicated in tumor progression. However, the underlying regulatory mechanism of m⁶A modification in CRC pathogenesis remain incompletely understood.

Methods

We first integrated bioinformatic analysis data, which revealed a significant positive correlation and co-enrichment of methyltransferase-Like 3 (METTL3) and the lncRNA metastasis-associated lung adenocarcinoma transcript 1 (MALAT1) in a high-risk prognostic signature. Functional validation included in vitro assessments of cellular proliferation, migration, invasion, apoptosis, and m⁶A methylation levels following METTL3 or MALAT1 suppression, alongside in vivo xenograft tumor models to evaluate the effects of a METTL3 inhibitor on tumor growth dynamics and m⁶A modification levels. Finally, mechanistic profiling was conducted to dissect NF-κB pathway activity.

Results

METTL3 and MALAT1 were markedly upregulated in CRC tissues and cell lines. Molecular docking identified trihydroxy methylcyclohexane aminoglycoside (TMA) as a high-affinity METTL3-binding compound. Inhibition of METTL3 or MALAT1 significantly impaired cellular proliferation, migration, and invasion, while promoting apoptosis. RNA immunoprecipitation quantitative PCR (RIP-qPCR) and RNA stability assays established that METTL3 maintains MALAT1 stability via m⁶A modification. In vivo, suppression of METTL3 potently inhibited xenograft tumor growth without observable toxicity. Western blot and immunohistochemistry (IHC) analyses further verified that this METTL3-MALAT1 axis activates the NF-κB pathway by modulating IκBα degradation and p65 nuclear translocation.

Conclusions

Collectively, our work elucidates the METTL3-m⁶A-MALAT1-NF-κB regulatory axis in CRC, advancing the understanding of epigenetic mechanisms in cancer progression and providing a theoretical foundation for developing METTL3-targeted therapeutic strategies.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-026-08114-0.

Background

Colorectal cancer (CRC) is among the malignancies causing the heaviest global cancer burden, and its complex pathogenesis remains a major research focus [1]. Recent advances in epigenetics have increasingly highlighted the significance of RNA modifications, especially N6-methyladenosine (m⁶A) modification, in tumor development [2]. As the most abundant RNA modification in eukaryotic organisms, m⁶A dynamics involve multiple functional components including methyltransferases such as the METTL3/METTL14 complex, demethylases including FTO and ALKBH5, and reader proteins like the YTHDF family [3, 4]. While m⁶A modification critically influences cancer progression through regulating diverse aspects of RNA metabolism encompassing stability, splicing, and translation [5], the regulatory networks orchestrated by the core methyltransferase METTL3 in CRC remain incompletely understood. Specifically, its downstream effector molecules and associated pathways present significant unresolved questions.

Long non-coding RNAs (lncRNAs) constitute a class of non-coding transcripts exceeding 200 nucleotides in length that play critical roles in epigenetic regulatory networks [6]. These RNA molecules participate extensively in fundamental biological processes including cell cycle regulation and differentiation [7], and are closely linked to the pathogenesis of major diseases [8]. Notably, they exert significant regulatory functions during tumor initiation, progression, and metastasis [9]. Among numerous lncRNAs, MALAT1 has garnered substantial attention due to its aberrant overexpression in multiple malignancies [10]. Research demonstrates that MALAT1 significantly promotes tumor progression by regulating cancer cell proliferation, migration, and apoptosis [11]. Recent studies suggest that m⁶A methylation may be involved in MALAT1 expression regulation [12, 13]. These findings imply that METTL3, as the core catalytic enzyme for m⁶A modification, might participate in tumor progression through regulating MALAT1 m⁶A modification. However, whether METTL3 governs MALAT1 expression via m⁶A-dependent mechanisms in CRC and how such regulation influences malignant phenotypes require further investigation.

Furthermore, inflammatory responses play pivotal roles in CRC pathogenesis. Chronic inflammation not only fosters tumor microenvironment formation but also accelerates cancer cell proliferation and metastasis by activating multiple signaling pathways [14, 15]. At the cellular level, persistent inflammatory stimulation induces genomic instability and epigenetic alterations, thereby driving malignant transformation of normal intestinal epithelial cells [16, 17]. The NF-κB signaling pathway serves as a central hub connecting inflammation and tumorigenesis, with its aberrant activation strongly associated with CRC invasion and metastasis [18, 19]. While studies indicate that lncRNAs can promote tumor development through NF-κB pathway activation [20], whether MALAT1 modulates NF-κB signaling via m⁶A modification-dependent mechanisms remains unreported. Therefore, the molecular mechanism by which METTL3-mediated m⁶A modification regulates MALAT1 to influence NF-κB pathway activity in CRC warrants elucidation.

Against this background, we hypothesized that an METTL3-m⁶A-LncRNA regulatory network plays a critical role in CRC progression. To test this hypothesis and identify the most pivotal lncRNA candidate, we employed an integrated approach combining bioinformatic analyses, clinical specimens, cellular models, and animal experiments to investigate the METTL3-m⁶A-MALAT1-NF-κB regulatory axis in CRC. We demonstrate for the first time that METTL3 stabilizes MALAT1 through m⁶A modification, thereby activating the NF-κB pathway to promote CRC proliferation and metastasis. Additionally, we identified a small-molecule inhibitor targeting METTL3, suggesting a novel therapeutic strategy for CRC.

Methods

Bioinformatic analysis

Transcriptome data from 468 CRC patients and 44 healthy subjects were obtained from TCGA, normalized to TPM. We identified m⁶A-related lncRNAs (Pearson |R| >0.4, FDR < 0.05) and constructed an m⁶A-lncRNA co-expression network. Key prognostic lncRNAs were selected using LASSO-Cox regression to establish a risk score model. Patients were stratified into high-risk and low-risk groups based on the median risk score. Group differences were assessed via Principal Component Analysis (PCA). Prognostic power was evaluated by Kaplan-Meier survival analysis and time-dependent ROC curves. Immune cell infiltration was estimated with TIMER2.0 and analyzed via GSVA. Key molecule expression was validated across 33 TCGA cancer types using Wilcoxon tests. All analyses were performed in R (v4.2.0) with significance set at P < 0.05.

Clinical samples

This study included 10 paired tumor and adjacent non-tumorous tissue specimens collected from CRC patients treated at Fujian Medical University Affiliated Longyan First Hospital between 2018 and 2020. All tissue samples were cryopreserved at -150 °C until further processing. The study protocol received approval from the Institutional Review Board of Fujian Medical University Affiliated Longyan First Hospital (Approval No. LYREC2024-k027-01), with written informed consent obtained from all participants.

Cell culture and treatment

The human normal colonic mucosal epithelial cell line NCM460 and various colorectal cancer cell lines (HT-29, Caco-2, SW480, LS174T, DLD-1) were obtained from Cellverse Co., Ltd. (Shanghai, China). Cells were cultured in their respective media supplemented with 10% fetal bovine serum (FBS; Gibco, USA) and 1% penicillin-streptomycin at 37 °C under 5% CO₂.

Cell proliferation assay

Cell proliferation was assessed using the Cell Counting Kit-8 (CCK-8; Dojindo, Japan) according to manufacturer protocols. LS174T cells were seeded in 96-well plates at a density of 5 × 10⁴ cells/mL (100 µL per well). Following incubation periods of 6, 24, 48, and 72 h at 37 °C with 5% CO₂, 10 µL of CCK-8 solution was added to each well. Plates were further incubated for 2 h before measuring absorbance at 450 nm using a microplate reader (Thermo Scientific Multiskan GO).

Transwell

LS174T cells in logarithmic growth phase were suspended in serum-free medium at a density of 4 × 10⁵ cells/mL. A 600 µL aliquot of complete medium containing 20% FBS was added to the lower chamber of a 24-well plate as the chemoattractant. Then, 100 µL of cell suspension was added to the upper chamber of a Transwell insert (8 μm pore). After 24 h of incubation at 37 °C under 5% CO₂, cells that migrated to the underside of the membrane were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet. Non-migrated cells on the upper surface were gently removed with a cotton swab. The membranes were imaged using an inverted microscope at 100× magnification, and migrated cells were quantified from five random fields per membrane. All experiments were performed with three technical replicates.

5’-ethynyl-2’-deoxyuridine (EdU) assay

Cell proliferation was assessed using an EdU Cell Proliferation Kit (Beyotime, C0078S). LS174T cells were seeded in 24-well plates, labeled with 10 µM EdU for 2 h, then fixed and permeabilized. After Click reaction and DAPI counterstaining, fluorescent images were captured using an inverted microscope (Leica DMi8, Germany).

Flow cytometric analysis

Apoptosis was assessed using Annexin V-FITC/PI double staining while cell cycle distribution was analyzed via PI single staining. Treated LS174T cells were collected, digested with EDTA-free trypsin, washed twice with ice-cold PBS, and resuspended. For apoptosis detection, 1 × 10⁶ cells were resuspended in 100 µL 1× Binding Buffer. Cells were stained with 5 µL Annexin V-FITC (Beyotime, C1062S) and 10 µL propidium iodide (PI), incubated at room temperature for 15 min protected from light, and immediately analyzed. For cell cycle analysis, another 1 × 10⁶ cells were fixed in 70% ice-cold ethanol at 4 °C overnight. After PBS washing, cells were stained with 500 µL PI solution at 37 °C for 30 min in the dark. All samples were analyzed using a BD FACSCanto II flow cytometer.

Quantitative real-time PCR (qPCR)

Total RNA was isolated from clinical tissues and cultured cells using TRIzol Reagent (Invitrogen, USA). Reverse transcription was performed with the SweScript RT Kit (Takara, Japan). Quantitative PCR assays utilized the ArtiCanATM SYBR qPCR Mix (Takara, Japan) on a QuantReady K9600 Real-Time PCR System (Tianlong, Gentier 96). Gene expression levels were normalized to GAPDH and calculated via the 2^−ΔΔCt method.

Primer Sequences: MALAT1-F: 5’-CTGACCCAGGTGCTACACAG-3’, MALAT1-R: 5’-GCTTGCTCCTCAGTCCTAGC-3’; METTL3-F: 5’-GTGATCGTAGCTGAGGTTCGT-3’, METTL3-R: 5’-GGGTTGCACATTGTGTGGTC-3’; GAPDH-F: 5’-AGGCCGGATGTGTTCGC-3’, GAPDH-R: 5’-CAAATCCGTTGACTCCGACC-3’.

Western blot assay

Protein extracts were prepared using RIPA lysis buffer supplemented with protease inhibitor cocktail (Beyotime, China). Proteins were separated by SDS-PAGE electrophoresis and transferred to PVDF membranes (Merck Millipore). Membranes were blocked with 5% skim milk for 1 h at room temperature, followed by overnight incubation at 4 °C with primary antibodies against specific targets. After washing, membranes were incubated with HRP-conjugated Goat Anti-Rabbit IgG secondary antibody (Servicebio, GB23303, 1:10000) for 1 h at room temperature. Protein bands were visualized using the SH-Advance523 ultrasensitive imaging system and quantified by densitometry with ImageJ software.

Primary antibodies: METTL3 (Abcam, ab195352, 1:1000), p-p65 (Proteintech, 82335-1-RR, 1:1000), p65 (Proteintech, 80979-1-RR, 1:2000), p-IκBα (Abcam, ab92700, 1:1000), IκBα (Abcam, ab32518, 1:2000), GAPDH (ABclonal, A19056, 1:5000).

Enzyme-linked immunosorbent assay (ELISA)

Serum from experimental and control mice was coagulated for 30 min at room temperature and centrifuged at 3000 × g for 15 min. Cell culture supernatants from colorectal cancer cells underwent centrifugation at 12,000 × g for 10 min to remove particulates. Concentrations of IL-6, TNF-α, and IL-1β were quantified using commercial ELISA kits (Elabscience) following manufacturer protocols. Absorbance was immediately measured at 450 nm using a SpectraMax M5 microplate reader with reference wavelength correction at 570 nm.

m⁶A RNA methylation level detection

Total m⁶A modification levels in colorectal cancer cells were quantified using the EpiQuik™ m⁶A RNA Methylation Quantification Kit (Epigentek). Total RNA was extracted from treated cells using TRIzol reagent, purified via chloroform phase separation and isopropanol precipitation, and treated with DNase I to remove genomic DNA. RNA purity and concentration were confirmed by spectrophotometry (A260/A280 ≈ 2.0). For the assay, 200 ng of RNA was added to antibody-precoated wells, incubated, and sequentially probed with capture and HRP-conjugated detection antibodies. After TMB color development, absorbance was measured at 450 nm, and the relative m⁶A content was calculated based on the standard curve.

RNA immunoprecipitation (RIP)

For m⁶A-specific RNA immunoprecipitation (MeRIP), total RNA was extracted from 1 × 10⁷ LS174T cells using TRIzol Reagent, fragmented to 100–500 nt, and immunoprecipitated overnight at 4 °C using 5 µg anti-m⁶A antibody (Abcam, ab208577) or equivalent IgG control (CST, 5415 S) bound to Protein A/G magnetic beads. Following washes, co-precipitated RNA was extracted, reverse transcribed, and analyzed by qPCR to quantify MALAT1 m⁶A enrichment using the 2^−ΔΔCt method relative to Input. For RNA-binding protein immunoprecipitation (RIP), cell lysates were incubated for 4 h at 4 °C with 4 µg anti-METTL3 antibody (Abcam, ab195352) or isotype IgG control (Santa Cruz, sc-2025) bound to Protein A/G beads. After washing, bound RNA was extracted, DNase-treated, reverse transcribed, and subjected to qPCR to assess METTL3-bound MALAT1 enrichment.

Molecular docking

This study employed computer-aided drug design to screen small-molecule inhibitors targeting METTL3. The crystal structure of METTL3 (PDB ID: 7L2T) was retrieved from the RCSB PDB database (https://www.rcsb.org/). Hydrogen addition and protonation state optimization were performed using MOE software (Chemical Computing Group), identifying the S-adenosylmethionine (SAM) binding site as the active pocket. Drug-like compound libraries were obtained from ZINC15 (https://zinc15.docking.org/) and ChEMBL (https://www.ebi.ac.uk/chembl/) databases, with a specific focus on compounds exhibiting known or potential anti-CRC activity. The libraries were filtered by Lipinski’s Rule of Five for docking analysis. Molecular docking was executed with AutoDock Vina, with grid centers covering the SAM binding site. Compounds exhibiting binding free energy (ΔG) ≤ -8.0 kcal/mol were selected for interaction analysis.

Actinomycin D assay for RNA stability

To assess the effect of METTL3 on MALAT1 RNA stability, an actinomycin D assay was performed by treating cells with 5 µg/mL actinomycin D (Sigma, A1410) to inhibit transcription. Cells were collected at 0, 3, and 6 h post-treatment, followed by total RNA extraction using TRIzol reagent and DNase I digestion. cDNA was synthesized from 1 µg RNA and subjected to qPCR analysis using SYBR Green Master Mix with GAPDH as reference. The half-life (t_1/2) of MALAT1 RNA was calculated by applying linear regression analysis to the natural logarithm (ln) of the remaining MALAT1% over time.

Mouse tumor formation experiment

4 weeks old female BALB/c-nu nude mice were obtained from Jiangsu GemPharmatech Co., Ltd (Jiangsu, China). Mice were randomly divided into two groups: the Model group (untreated) and Model+METTL3 inhibitor group. Logarithmic-phase cells were resuspended at 5 × 10⁶ cells/mL in PBS and subcutaneously injected into the right flank (100 µL suspension per mouse). Tumor volumes were measured every 3 d using digital calipers once xenografts reached approximately 50 mm³, calculated as (length × width²)/2. At study completion, mice were euthanized by CO₂ asphyxiation followed by cervical dislocation. Tumors were dissected intact for downstream analyses.

Hematoxylin and eosin (H&E) staining

Tumor tissues from mice were fixed in 4% paraformaldehyde, dehydrated, and paraffin-embedded, then sectioned at 4 μm. After deparaffinization and rehydration, sections were stained with hematoxylin and eosin using a commercial kit (ebiogo, China), followed by dehydration, clearing, and mounting with neutral resin. Histological analysis was performed using an optical microscope (OLYMPUS BX53).

Immunohistochemistry (IHC)

Paraffin-embedded tumor tissue Sect.  (4 μm thickness) mounted on glass slides were deparaffinized in xylene and rehydrated through graded ethanol solutions. Antigen retrieval was performed using citrate buffer (pH 6.0) at 95 °C for 20 min. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide for 15 min at room temperature. After serum blocking, sections were incubated overnight at 4 °C with primary antibodies: anti-Ki67 (1:300, Abcam, ab15580), anti-METTL3 (1:200, Proteintech, 15073-1-AP), and anti-p65 (1:100, Proteintech, 10745-1-AP). HRP-conjugated secondary antibody (1:5000, Abcam, ab6721) incubation proceeded for 1 h at room temperature. DAB substrate was applied for 1–3 min for chromogenic detection, followed by hematoxylin counterstaining. Slides were mounted with resin mounting medium and imaged using an optical microscope (OLYMPUS BX53, Japan) at 400× magnification.

TUNEL staining

Paraffin-embedded tumor Section  (4 μm) were deparaffinized in xylene and rehydrated through graded ethanol. Sections were sequentially stained with Weigert’s iron hematoxylin (Sigma, USA), ponceau-acid fuchsin (Sigma, USA), and aniline blue (Sigma, USA), with differentiation in 1% acid alcohol and 1% phosphomolybdic acid. After dehydration, clearing, and mounting with neutral resin (Sakura, Japan), slides were observed and recorded using a Zeiss Axio Imager microscope (Germany) with Axiocam 506 camera (Zeiss, Germany).

Statistical analysis

All experiments were performed in triplicate unless otherwise specified. Data are presented as mean ± standard deviation (SD). Statistical significance was determined using Student’s t-test or one-way ANOVA with Tukey’s post hoc test, as appropriate, in GraphPad Prism 9 software (GraphPad Software, USA). P-values < 0.05 were considered statistically significant.

Results

Identification of a prognostic m⁶A-related lncRNA signature in CRC

We identified m⁶A-associated lncRNAs relevant to CRC progression using RNA sequencing data from 408 CRC patients and 19 healthy subjects obtained from TCGA. Differential expression analysis of lncRNAs data, applying thresholds of |log2FC| > 1 and FDR < 0.05, revealed 1213 lncRNAs significantly dysregulated between NC and CRC groups (Fig. 1A-B). Correlating these 1213 lncRNAs with 23 established m⁶A regulators [21] using a stringent threshold (|R| > 0.6, P < 0.001) yielded 176 candidate m⁶A-related lncRNAs (Fig. 1C).

Fig. 1.

Identification of prognostic m⁶A-related lncRNA signature in CRC. (A) Volcano plot illustrating differentially expressed lncRNAs between NC and CRC. (B) Heatmap depicting hierarchical clustering of dysregulated lncRNAs in NC versus CRC cohorts. (C) Venn diagram. (D-E) LASSO-cox feature selection. (F) Signature lncRNAs expression across risk groups. (G) Kaplan-Meier survival curves by risk stratification. (H) ROC curves

Subsequent dimensionality reduction via LASSO-Cox regression analysis identified a core set of 10 prognosis-associated m⁶A-related lncRNAs (LINC01354, LINC01836, FRMD6-AS1, MALAT1, FAM198A2, ZNF667-AS1, ZDFAS1, LINC00324, ACTA2-AS1, LINC01559) (Fig. 1D-E), which formed the basis of a prognostic risk model. The expression patterns of these 10 lncRNAs across the defined risk groups are visualized in Fig. 1F. Kaplan-Meier analysis demonstrated significantly poorer overall survival (OS) for patients in the high-risk group compared to the low-risk group (P < 0.001, Fig. 1G). Time-dependent ROC analysis showed predictive capability for the risk score model, yielding AUC values approximately 0.6 at 1, 3, and 5 years (Fig. 1H). Principal component analysis (PCA) based on the 10-lncRNA signature confirmed a clear separation between the high-risk and low-risk groups (Supplementary Fig. 1A), and the risk model exhibited superior diagnostic utility compared to standard clinical features (Supplementary Fig. 1B).

Notably, advanced T stage, lymph node metastasis, and higher AJCC stage were significantly enriched within the high-risk group (Supplementary Fig. 1C). Correlation analysis integrating the 10 prognostic lncRNAs, the model risk score, and the 23 m⁶A regulatory genes revealed a predominant positive association, with correlation coefficients mainly distributed around 0.5 (Supplementary Fig. 1D). A correlation heatmap further substantiated this trend and highlighted significant positive correlations: specifically, METTL3 showed strong positive correlations with MALAT1 (P < 0.001). Based on the strength of their association and their established roles in cancer biology [13, 22], and considering METTL3’s central function as the core m⁶A methyltransferase [23], METTL3 and MALAT1 were selected for further functional investigation into their cooperative role in CRC progression.

Immune microenvironment features and pan-cancer expression patterns of core regulators

To comprehensively delineate immune characteristics distinguishing the high-risk and low-risk groups, we analyzed immune cell infiltration estimates for colorectal cancer (COAD) patients sourced from the TIMER2.0 database, integrating data derived from seven distinct algorithms (TIMER, CIBERSORT, CIBERSORT-ABS, QUANTISEQ, MCPCOUNTER, XCELL, and EPIC). Significant differences in immune cell composition were observed between the two groups. The high-risk group exhibited pronounced enrichment of immunosuppressive cell populations, including regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs). Conversely, the low-risk group demonstrated significant enrichment of anti-tumor immune effector cells, including T cells (CD8+), T cells (CD4+), T follicular helper cells, and natural killer (NK) cells (Fig. 2A-B). Subsequent evaluation of immunotherapeutic response potential showed consistent and statistically significant differences in Immunophenoscore (IPS) across risk groups for both anti-CTLA-4 and anti-PD-1 immune checkpoint inhibitors, with the high-risk group exhibiting substantially lower IPS values - aligning with its immunosuppressive cellular profile (Fig. 2C). Pan-cancer analysis of the core regulators METTL3 and MALAT1 revealed consistent and statistically significant overexpression in tumor tissues compared to normal tissues across multiple malignancies, including colorectal cancer, gastric cancer, and hepatocellular carcinoma (Fig. 2D-E).

Fig. 2.

Immune microenvironment features and pan-cancer expression patterns of core regulators. (A) Immune infiltration heatmap. (B) Differential analysis of immune infiltration between high- and low-risk groups. (C) Therapeutic efficacy analysis of PD-1 and CTLA-4 monoclonal antibodies in high-risk and low-risk patient groups. (D) Pan-cancer expression analysis of METTL3. (E) Pan-cancer expression analysis of MALAT1

MALAT1 and METTL3 are highly expressed in colorectal cancer

To elucidate the roles of MALAT1 and METTL3 in CRC development, we first analyzed their expression levels in CRC tissues versus adjacent normal tissues using qPCR. Results demonstrated significantly higher expression of both MALAT1 and METTL3 in CRC tissues compared to normal controls (Fig. 3A-B). IHC analysis further validated elevated METTL3 protein expression in CRC tissues (Fig. 3C). To determine whether this phenomenon extends to cellular models, we examined MALAT1 expression across six CRC cell lines (HT-29, SW480, Caco-2, DLD-1, LS174T, RKO) and normal mucosal epithelial cells (NCM460). All CRC cell lines exhibited significantly increased MALAT1 expression relative to NCM460 cells, with LS174T showing the highest expression level (Fig. 3D). These findings collectively indicate that MALAT1 and METTL3 are consistently overexpressed in CRC tissues and cell lines.

Fig. 3.

MALAT1 exhibits abnormal high expression in colorectal cancer tissues and cells. (A-B) qPCR detection of MALAT1 (A) and METTL3 (B) mRNA expression levels in colorectal cancer tissues. (C) IHC analysis of METTL3 protein expression level in colorectal cancer tissues. Scale bar: 100 μm. (D) qPCR detection of MALAT1 mRNA expression quantity in normal mucosal epithelial cells (NCM460) and colorectal cancer cell lines (HT-29, SW480, Caco-2, DLD-1, LS174T, RKO). *P < 0.05, **P < 0.01, ***P < 0.001

Effect of MALAT1 on biological behaviors of colorectal cancer cells

To further evaluate the biological function of MALAT1 in CRC, LS174T cells were treated with a MALAT1-specific inhibitor. qPCR confirmed significantly reduced MALAT1 expression in inhibitor-treated cells versus controls (Fig. 4A). Functional assays demonstrated that MALAT1 inhibition substantially attenuated cellular proliferation, as evidenced by decreased CCK-8 absorbance (Fig. 4B) and reduced EdU-positive cell rates after 72 h treatment (Fig. 4C). Transwell analysis revealed impaired migration and invasion capacities in the inhibitor group (Fig. 4D). Flow cytometry simultaneously detected a significant increase in apoptotic rate (Fig. 4E) and cell cycle redistribution characterized by G1-phase accumulation with concomitant S-phase reduction (Fig. 4F), indicating that MALAT1 inhibition arrests cell cycle progression. These results collectively demonstrate that MALAT1 suppression inhibits proliferation, migration, and invasion while promoting apoptosis in CRC cells.

Fig. 4.

Effect of MALAT1 inhibition on biological behaviors of colorectal cancer cells. (A) qPCR detection of MALAT1 mRNA expression post-MALAT1 inhibitor treatment. (B) CCK-8 detection of cell proliferation after MALAT1 inhibitor treatment at 24, 48, and 72 h. (C) EdU assessment of cell proliferation capacity. (D) Transwell detection of cell migration and invasion capabilities post-MALAT1 inhibitor treatment. Scale bar: 100 μm. (E) Flow cytometry detection of apoptosis after MALAT1 inhibitor treatment. (F) Flow cytometry detection of cell cycle distribution. *P < 0.05, **P < 0.01, ***P < 0.001

METTL3 maintains MALAT1 stability via m⁶A-dependent modification

To elucidate the regulatory mechanism underlying MALAT1 overexpression in colorectal cancer, we observed significantly elevated global m⁶A modification levels in CRC cell lines (HT-29, SW480, Caco-2, DLD-1, LS174T, RKO) compared to normal intestinal epithelial cells NCM460, with LS174T exhibiting the most pronounced increase (Fig. 5A). MeRIP-qPCR further confirmed enhanced m⁶A modification specifically on MALAT1 in all cancer cell lines, again showing maximal modification in LS174T cells (Fig. 5B). Concordantly, qPCR and Western blot analyses demonstrated consistently upregulated expression of both MALAT1 and METTL3 in these cell lines relative to NCM460, with expression patterns mirroring MALAT1 m⁶A modification levels (Fig. 5C-E), suggesting METTL3 regulates MALAT1 via m⁶A modification. Leveraging this regulatory relationship, we employed molecular docking to screen potential METTL3-targeting inhibitors. Trihydroxy methylcyclohexene aminoglycoside (TMA, binding energy: -12.550 kcal/mol) and pitavastatin (binding energy: -9.234 kcal/mol) demonstrated strong binding affinity to the METTL3 active site (Fig. 5F, detailed parameters in Table 1). Subsequent CCK-8 cytotoxicity assays validated TMA as the most potent inhibitor of CRC cell growth (Fig. 5G). Therefore, TMA was selected for further investigation.

Fig. 5.

METTL3 regulates MALAT1 level through m⁶A modification. (A) Total m⁶A level in cells. (B) MeRIP-qPCR detection of MALAT1 methylation level in colorectal cancer cells. (C-D) qPCR detection of MALAT1 (C) and METTL3 (D) mRNA expression quantity in normal mucosal epithelial cells (NCM460) and colorectal cancer cell lines (HT-29, SW480, Caco-2, DLD-1, LS174T, RKO). (E) Representative Western blot images and quantitative analysis of METTL3 protein in normal mucosal epithelial cells (NCM460) and colorectal cancer cell lines (HT-29, SW480, Caco-2, DLD-1, LS174T, RKO). (F) Molecular docking results of small-molecule compounds with METTL3. (G) CCK-8 detection of inhibition rates in LS174T cells treated with different small-molecule compounds. *P < 0.05, ***P < 0.001

Table 1.

Docking scores of METTL3 with small molecule compounds

Compound CID	Compound Name	Binding Affinity (kcal/mol)
44,891,366	Trihydroxy methylcyclohexene aminoglycoside	-12.550
5,282,452	Pitavastatin	-9.234
24,848,419	pitavastatin(1-)	-9.140
4612	oxamniquine	-7.398
11,450,633	Alogliptin	-7.382

Effect of METTL3 on biological behaviors of colorectal cancer cells through regulating MALAT1

To investigate METTL3-mediated regulation of MALAT1, LS174T cells were treated with the METTL3 inhibitor Compound TMA. qPCR analysis revealed significant downregulation of METTL3 (Fig. 6A) and MALAT1 mRNA levels (Fig. 6B). Western blot confirmed decreased METTL3 protein expression (Fig. 6C). Consistent with METTL3’s role as the core methyltransferase, total m⁶A modification levels were reduced (Fig. 6D), and MeRIP-qPCR demonstrated significantly diminished MALAT1-specific m⁶A modification (Fig. 6E). RIP-qPCR verified direct binding between METTL3 and MALAT1 (Fig. 6F), while RNA stability assays showed that TMA treatment substantially shortened the half-life of MALAT1 following actinomycin D exposure (Fig. 6G), indicating METTL3 maintains MALAT1 stability via m⁶A modification. Functionally, TMA treatment significantly impaired cellular proliferation (Fig. 6H-I) and reduced migration/invasion capacities (Fig. 6J). Concurrently, flow cytometry detected increased apoptosis (Fig. 7A) and G1-phase cell accumulation with S-phase reduction (Fig. 7B), confirming G1-phase cell cycle arrest. Downstream pro-inflammatory cytokine analysis showed significantly decreased expression of IL-6 (Fig. 7C), IL-1β (Fig. 7D), and TNF-α (Fig. 7E). These results collectively demonstrate that METTL3 regulates MALAT1 expression and stability through m⁶A modification, thereby modulating malignant behaviors in colorectal cancer cells.

Fig. 6.

Effects of METTL3 on proliferation and migration/invasion of colorectal cancer cells through regulating MALAT1. (A-B) qPCR detection of METTL3 (A) and MALAT1 (B) mRNA expression levels in cells treated with METTL3 inhibitor (Compound TMA). (C) Representative Western blot images and quantitative analysis of METTL3 protein in cells. (D) Total m⁶A level. (E) MeRIP-qPCR detection of MALAT1 m⁶A methylation level. (F) RIP-qPCR detection of endogenous MALAT1 expression in cells. (G) Assessment of METTL3 effect on MALAT1 stability via Actinomycin D treatment. (H) CCK-8 detection of cell proliferation after METTL3 inhibitor treatment at 24, 48, and 72 h. (I) EdU assessment of cell proliferation. Scale bar: 50 μm. (J) Transwell detection of cell migration and invasion capabilities post-METTL3 inhibitor treatment. Scale bar: 100 μm. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 7.

Effects of METTL3 on apoptosis and inflammation of colorectal cancer cells through regulating MALAT1. (A) Flow cytometry detection of apoptosis after METTL3 inhibitor treatment. (B) Flow cytometry detection of cell cycle distribution. *P < 0.05, **P < 0.01, ***P < 0.001

In vivo study on METTL3 regulating colorectal cancer growth via MALAT1

To evaluate the therapeutic efficacy of METTL3 inhibition in vivo, a xenograft model was established by implanting CRC cells in nude mice. Longitudinal monitoring revealed significantly attenuated tumor volume progression in the inhibitor-treated group (Fig. 8A-B). At the experimental endpoint (day 20), tumors from this group exhibited substantially reduced weight compared to controls (Fig. 8C), while no significant body weight differences were observed between groups (Fig. 8D), indicating favorable in vivo safety. Molecular analysis of excised tumors demonstrated downregulated MALAT1 expression (Fig. 8E), decreased global m⁶A levels (Fig. 8F), and reduced MALAT1-specific m⁶A modification (Fig. 8G) in the inhibitor group. Histopathological examination by H&E staining showed improved tumor morphology with reduced volume and diminished inflammatory infiltration (Fig. 8H). IHC revealed significantly fewer Ki67-positive proliferating cells (Fig. 8I), while TUNEL staining confirmed increased apoptotic indices (Fig. 8J), indicating that METTL3 inhibition suppresses tumor growth through proliferation blockade and apoptosis induction. These in vivo findings align with in vitro results, collectively demonstrating that METTL3 targeting inhibits CRC xenograft growth through MALAT1 regulation.

Fig. 8.

Effect of METTL3 on colorectal cancer growth through regulating MALAT1. (A) Representative tumor tissue images from each group of mice. (B) Tumor volume in each group of mice. (C) Body weight in each group of mice. (D) Tumor weight in each group of mice. (E) qPCR detection of MALAT1 mRNA level in tumor tissues post-METTL3 treatment. (F) Total m⁶A level in tumor tissues. (G) MeRIP-qPCR detection of MALAT1 methylation level. (H) H&E staining of tumor tissues. Scale bar: 50 μm. (I) IHC detection of Ki67 expression level in tumor tissues. Scale bar: 50 μm. (J) TUNEL staining. Scale bar: 50 μm. **P < 0.01, ***P < 0.001

METTL3 inhibition suppresses colorectal cancer progression by modulating the MALAT1/NF-κB signaling axis

To elucidate the downstream regulatory mechanism of the METTL3-MALAT1 axis, we investigated changes in NF-κB signaling. In LS174T cells, METTL3 inhibitor treatment significantly downregulated p65 mRNA while upregulating IκBα mRNA (Fig. 9A-B). Western blot analysis showed reduced phosphorylation levels of key NF-κB pathway components p-IκBα and p-p65 (Fig. 9C), whereas total IκBα and p65 protein expression remained unchanged. Concurrently, secretion of inflammatory cytokines IL-6, IL-1β, and TNF-α was significantly diminished (Fig. 9F). This regulatory pattern was recapitulated in mouse xenograft models. IHC revealed decreased protein expression of both METTL3 and nuclear p65 in inhibitor-treated tumors (Fig. 9D). Western blot further confirmed reduced phosphorylation of p-IκBα and p-p65 (Fig. 9E), while ELISA demonstrated parallel decreases in IL-6, IL-1β, and TNF-α secretion (Fig. 9G). These results collectively demonstrate that METTL3 modulates NF-κB pathway activation through MALAT1 regulation, establishing the METTL3-MALAT1-NF-κB axis as a critical driver of colorectal cancer progression.

Fig. 9.

METTL3 inhibition activates NF-κB signaling pathway through regulating MALAT1. (A-B) qPCR detection of p65 (A) and IκBα (B) mRNA expression levels in cells. (C) Representative Western blot images and quantitative analysis of p-p65, p65, p-IκBα and IκBα proteins in cells. (D) IHC analysis of METTL3 and p65 protein expression levels in tumor tissues. Scale bar: 50 μm. (E) Representative Western blot images and quantitative analysis of p-p65, p65, p-IκBα and IκBα proteins in tumor tissues. (F) ELISA detection of IL-6, IL-1β and TNF-α secretion levels in cell supernatants. (G) ELISA detection of IL-6, IL-1β and TNF-α secretion levels in mouse serum. *P < 0.05, **P < 0.01, ***P < 0.001

Discussion

CRC pathogenesis is closely linked to dysregulated epigenetic control [24]. Our study provides a systematic elucidation of the molecular mechanism by which METTL3 promotes CRC progression through m⁶A-mediated regulation of MALAT1, leading to NF-κB pathway activation. This discovery establishes a novel epigenetic regulatory axis in CRC.

m⁶A modification represents is the most prevalent internal chemical modification in eukaryotic mRNA and lncRNAs. This modification is dynamically regulated by a group of proteins termed “writers” (methyltransferases), “erasers” (demethylases), and “readers” (recognition proteins) [25]. The core m⁶A methyltransferase complex primarily consists of METTL3 and METTL14, which form a stable heterodimer in a 1:1 stoichiometry and co-localize within nuclear speckles [26]. Within this complex, METTL3 serves as the catalytic subunit, while METTL14 provides structural support and enhances RNA-binding affinity [27]. Accumulating evidence indicates that aberrant m⁶A modification is critically involved in the pathogenesis of multiple human diseases, including cardiovascular disorders, neurological conditions, and various cancers [28–30]. For instance, METTL3-mediated m⁶A modification of lncRNA E230001N04Rik has been shown to promote senescence of lung epithelial cells [31], whereas in papillary thyroid cancer, it stabilizes lncRNA LINC01125, consequently suppressing malignant progression, it stabilizes lncRNA LINC01125, consequently suppressing malignant progression [32]. Despite these advances, the specific regulatory mechanisms and functional consequences of METTL3-mediated m⁶A modification on lncRNAs in CRC remain largely unexplored.

Our study fills this knowledge gap by demonstrating, through comprehensive TCGA-based prognostic modeling and experimental validation, that METTL3 specifically recognizes and catalyzes m⁶A modification on MALAT1, thereby enhancing its RNA stability and promoting CRC progression. This finding not only confirms the pivotal role of METTL3 in lncRNAs regulation but also provides mechanistic insights into the METTL3-m⁶A-MALAT1 axis in colorectal carcinogenesis.

The NF-κB pathway represents a classically well-established signaling cascade implicated in inflammation and cancer [33, 34]. A growing body of evidence indicates that the expression and regulation of m⁶A methylation-related genes are closely associated with various immune signaling pathways, particularly the NF-κB pathway [35, 36]. For instance, METTL3 has been reported to regulate LPS-induced inflammatory responses in macrophages via the NF-κB pathway [37], and the m⁶A reader YTHDC2 was shown to influence lung cell malignancy through the CYLD/NF-κB axis [38]. Furthermore, m⁶A methylation has been found to regulate the ERK/NF-κB/AKT signaling pathway in endometrial cancer [39], underscoring the functional link between m⁶A modification and NF-κB activity.

In this study, bioinformatic analysis of TCGA data not only revealed a significant co-expression and prognostic value of METTL3 and MALAT1 in CRC but also identified a distinct immunosuppressive microenvironment in the high-risk patient group characterized by METTL3/MALAT1 enrichment. Building on this clinical association, we delineated the functional METTL3–m⁶A–MALAT1 axis and its link to NF-κB signaling. Through integrated experiments, we demonstrate that METTL3-mediated m⁶A modification stabilizes MALAT1, which in turn activates the NF-κB pathway by promoting IκBα degradation and p65 nuclear translocation. This axis exhibits a distinct immunosuppressive phenotype in CRC, providing novel mechanistic insights into how epitranscriptomic regulation modulates both oncogenic signaling and the tumor immune microenvironment.

Despite these significant findings, several limitations warrant further investigation. First, while we confirmed METTL3 regulates MALAT1 expression via m⁶A modification, the specific modification sites and their functional impacts require deeper characterization. Second, the in vivo observation period was relatively short, precluding assessment of long-term therapeutic efficacy and survival outcomes. Additionally, although the screened small-molecule inhibitor demonstrates promising inhibitory effects, its target selectivity and pharmacokinetic properties need optimization. Finally, tumor microenvironment factors, particularly immune cell interactions with the METTL3-MALAT1-NF-κB axis, were not fully explored and merit dedicated investigation.

Conclusions

In summary, this study systematically demonstrates the pivotal role of the METTL3-m⁶A-MALAT1-NF-κB axis in colorectal cancer pathogenesis. Our findings establish that METTL3 stabilizes MALAT1 through m⁶A modification, thereby activating the NF-κB signaling pathway to drive tumor progression. This discovery not only deepens our understanding of epigenetic regulation in CRC but also provides critical foundations for developing novel diagnostic biomarkers and therapeutic targets. Future research will further explore the clinical translation potential of this regulatory axis to advance precision medicine approaches for colorectal cancer.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ALKBH5

AlkB Homolog 5

AUC

Area Under the Curve

CCK-8

Cell Counting Kit-8

COAD

Colon Adenocarcinoma

Cox

Cox regression

CRC

Colorectal cancer

DAPI

4’,6-diamidino-2-phenylindole

DMEM

Dulbecco’s Modified Eagle Medium

EdU

5-ethynyl-2’-deoxyuridine

ELISA

Enzyme-Linked Immunosorbent Assay

FBS

Fetal Bovine Serum

FITC

Fluorescein Isothiocyanate

FTO

Fat mass and obesity-associated protein

GSVA

Gene Set Variation Analysis

H&E

Hematoxylin and Eosin

HRP

Horseradish Peroxidase

IHC

Immunohistochemistry

LASSO

Least Absolute Shrinkage and Selection Operator

LncRNA

Long non-coding RNA

MALAT1

metastasis-associated lung adenocarcinoma transcript 1

MDSCs

Myeloid-Derived Suppressor Cells

MEM

Minimum Essential Medium

MeRIP

Methylated RNA Immunoprecipitation

METTL3

methyltransferase-like 3

m⁶A

N6-methyladenosine

MOE

Molecular Operating Environment

NF-κB

Nuclear Factor Kappa B

NK

Natural Killer cells

OS

Overall Survival

PCA

Principal Component Analysis

PDB

Protein Data Bank

PI

Propidium Iodide

qPCR

quantitative PCR

RIP

RNA Immunoprecipitation

RIPA

Radioimmunoprecipitation Assay buffer

ROC

Receiver Operating Characteristic

RPMI

Roswell Park Memorial Institute medium

RT

Reverse Transcription

SAM

S-adenosylmethionine

SDS-PAGE

Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis

TCGA

The Cancer Genome Atlas

TMA

trihydroxy methylcyclohexane aminoglycoside

Tregs

Regulatory T cells

TUNEL

Terminal deoxynucleotidyl transferase dUTP nick end labeling

YTHDF

YTH Domain Family

Author contributions

Conceived and designed the study: Dongbo Xu, Shuangming Lin. Performed the literature search and data extraction: Shuangming Lin, Hao Zeng, Xueyi Xue, Baodong Liao. Analyzed the data: Shuangming Lin, Hao Zeng, Yuhui Wang. Drafted the manuscript: Dongbo Xu, Shuangming Lin, Hao Zeng, Hao Lin. All authors contributed to critical revision of the manuscript, read, and approved the final submitted version.

Funding

This work was financially supported by Lanyan City Joint Funding (2022LYF17094), Longyan Clinical Medical Research Center of Digestive Diseases (2025LYF1002).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Clinical study ethics approval and consent to participate

This study was approved by the Ethics Committee of Fujian Medical University Affiliated Longyan First Hospital (Approval No.: LYREC2024-k027-01). All procedures were conducted in accordance with the Declaration of Helsinki and national ethical guidelines for medical research. Written informed consent was obtained from all participants or their legal guardians prior to enrollment, ensuring voluntary participation and confidentiality of personal health information. The study design prioritized patient safety and privacy throughout the research process.

Animal study ethical statement

All animal experiments were performed in compliance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by Gannan Innovation and Translational Medicine Research Institute (Approval No.: MI-20250051). All efforts were made to minimize animal suffering and to reduce the number of animals used in the experiments.

Consent for publication

All authors have reviewed and approved the final version of this manuscript for publication. Written informed consent was obtained from all participants (or their legal guardians) for the publication of anonymized data derived from their clinical samples in this study.

Competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.