METTL3 promotes an immunosuppressive microenvironment in bladder cancer via m6A-dependent CXCL5/CCL5 regulation

Abstract

Background: Bladder cancer (BLCA) is a challenging malignancy with a poor prognosis, particularly in muscle-invasive cases. Despite recent advancements in immunotherapy, response rates remain suboptimal. This study investigates the role of METTL3, an m6A RNA methylation "writer," in regulating the immune microenvironment of BLCA.

Methods: Through bioinformatics analysis, we identified METTL3 as being associated with the formation of an immunosuppressive microenvironment in BLCA and poor response to immunotherapy. Subsequently, we silenced METTL3 expression in BLCA cells using short hairpin RNA (shRNA) or inhibited its function with STM2457. The effectiveness of these interventions in remodeling the BLCA tumor microenvironment (TME) was confirmed through animal experiments and flow cytometry. Mechanistically, RNA sequencing and methylated RNA immunoprecipitation (MeRIP) sequencing revealed the molecular pathways by which METTL3 regulates the TME. This was further validated using in vitro cell co-culture, immunoprecipitation, ELISA, and RNA degradation assays. The synergistic effect of METTL3 with anti-Programmed Cell Death Protein 1 (PD-1) treatment in BLCA was confirmed in both orthotopic and ectopic BLCA animal models.

Results: METTL3 was found to increase CXCL5 levels and suppress CCL5 expression in an m6A-dependent manner, leading to increased recruitment of myeloid-derived suppressor cells (MDSCs) and reduced infiltration of CD8+T cells. Silencing METTL3 or inhibiting its function restored immune cell balance and significantly enhanced the efficacy of anti-PD-1 therapy. Clinically, METTL3 overexpression correlated with poor complete response rate to immune checkpoint inhibitors (ICIs) therapy, associated with an immunosuppressive microenvironment characterized by elevated MDSC levels and reduced CD8+T cell infiltration.

Conclusions: These findings highlight METTL3 as a key regulator of the immune microenvironment in BLCA and a promising therapeutic target to improve immunotherapy outcomes. Targeting METTL3 could potentially enhance the efficacy of ICIs in patients with BLCA.

Keywords: Bladder Cancer; Cytokine; Immunotherapy; Myeloid-derived suppressor cell - MDSC; T cell.

Figure 1. METTL3 is highly expressed in tumors and is associated with an immunosuppressive microenvironment. (A) Flowchart for screening key N6-methyladenosine (m6A) modification genes related to immunotherapy response in bladder cancer (BLCA). (B) Pearson correlation analysis bar chart of the 10 target genes with the percentage of complete response (CR) patients to immunotherapy in the IMvigor210 cohort, and a scatter plot of METTL3 expression level versus CR patient percentage. (C) Proportion of immunotherapy responses among different Lund subtypes in the IMvigor210 cohort. (D) Violin plot of METTL3 expression levels in bladder tissues of patients with different Lund subtypes. (E–F) Expression and statistical analysis of METTL3 in normal and tumor cells from single-cell sequencing of clinical bladder cancer samples. Histogram of METTL3 expression levels in cancer tissues versus adjacent normal tissues in (G) non-paired samples and (H) paired samples from the The Cancer Genome Atlas (TCGA) bladder cancer cohort. (I) Representative immunohistochemistry staining of METTL3 in clinical BLCA samples. (J–K) Scatter plots of METTL3 expression levels with CD8+T cell, cytotoxic cell, and myeloid-derived suppressor cell (MDSC) infiltration levels based on ssGSEA algorithm and TIMER V.2.0 database. (L) Statistical plot of METTL3 expression levels and immune scores in BLCA from the CAMOIP database. *p<0.05; **p<0.01; ***p<0.001.

Figure 2. Silencing METTL3 inhibits bladder cancer progression by reshaping the tumor microenvironment. (A) Western blot analysis of METTL3 expression levels in MB49 cells. (B) Reverse Transcription Quantitative Polymerase Chain Reaction (RT-qPCR) analysis of METTL3 expression levels in MB49 cells. (C) Schematic of the animal experiment: control, METTL3-silenced, or METTL3-overexpressing MB49 cells were subcutaneously injected into mice, and on day 12, flow cytometry was performed to analyze the tumor immune microenvironment (n=5), along with the assessment of tumor growth in mice (n=5). (D) Images of bladder cancer tumors in mice. (E) Growth curves of bladder cancer tumors in mice. (F) Tumor weights of bladder cancer tumors in mice. (G–L) Analysis results of the immune microenvironment in bladder cancer tumors from mice. (M) Schematic of the in situ bladder cancer model: control, METTL3-silenced, or METTL3-overexpressing MB49 cells were injected into the bladder wall of mice, and in vivo imaging system (IVIS) was used to monitor tumor growth in mice (n=5). (N) IVIS images of in situ bladder cancer tumors. (O) Images of in situ bladder cancer tumors in mice. (P) Tumor volumes of in situ bladder cancer tumors in mice. (Q) Tumor weights of in situ bladder cancer tumors in mice. ns, no significance. *p<0.05; **p<0.01; ***p<0.001.

Figure 3. METTL3 regulates the expression and secretion of CXCL5 and CCL5 in bladder cancer. (A) Schematic of the transcriptome sequencing workflow following METTL3 knockdown in MB49 cells. (B) Heatmap of differentially expressed genes after METTL3 knockdown (criteria for differential genes: p<0.05, fold change >1.5 or <0.67. (C) Volcano plot of differentially expressed genes after METTL3 knockdown. (D) Network diagram of GO enrichment analysis of differentially expressed genes. (E) Chemokines with differential expression following METTL3 silencing. (F–G) Reverse Transcription Quantitative Polymerase Chain Reaction (RT-qPCR) analysis of CXCL5 and CCL5 mRNA expression levels following METTL3 overexpression or knockdown in MB49 cells. (H) RT-qPCR analysis of CXCL5 and CCL5 mRNA expression levels in MB49 cells treated with DMSO or STM2457 for 72 hours. (I) ELISA of CXCL5 and CCL5 secretion levels in the culture supernatant of MB49 cell lines; ELISA of CXCL5 and CCL5 levels in (J) mouse tumor tissues and (K) in peripheral blood serum. RT-qPCR analysis of CXCL5 and CCL5 mRNA expression levels in (L) 5637 cells and (M) T24 cells treated with DMSO or STM2457 for 72 hours. *p<0.05; **p<0.01; ***p<0.001.

Figure 4. METTL3 regulates CXCL5 expression through N6-methyladenosine (m6A) modification, thereby promoting myeloid-derived suppressor cell (MDSC) chemotaxis. (A) Schematic diagram of the workflow used for Methylated RNA Immunoprecipitation (MeRIP) sequencing and data analysis in METTL3-knockdown MB49 stable cell lines. (B) Bar chart showing the number of m6A modification sites identified in MeRIP sequencing results. (C) Venn diagram of downstream target gene screening for METTL3, intersecting chemokines significantly altered in RNA sequencing with those showing significant downregulation in m6A modification levels in m6A sequencing. (D) m6A peak map of CXCL5 mRNA modification sites. (E) Bar chart of MeRIP-qPCR results showing the m6A modification level of CXCL5 mRNA in MB49 cells after METTL3 knockdown. (F) RNA Immunoprecipitation (RIP) assay detecting the interaction between METTL3 and CXCL5 mRNA. (G) RNA degradation assay showing CXCL5 mRNA stability after silencing METTL3. (H) RNA degradation assay showing CXCL5 mRNA stability after treatment with METTL3 inhibitor STM2457 (2 µg/mL, 72 hours) in MB49 cells. (I) Reverse Transcription Quantitative Polymerase Chain Reaction (RT-qPCR) analysis of IGF2BP1 and CXCL5 mRNA expression levels in MB49 cells after silencing IGF2BP1. (J) RT-qPCR analysis of IGF2BP2 and CXCL5 mRNA expression levels in MB49 cells after silencing IGF2BP2. (K) RT-qPCR analysis of METTL3, IGF2BP1, and CXCL5 mRNA expression levels in MB49 cells after overexpression of METTL3 and/or silencing of IGF2BP1. (L) Schematic of the animal experiment. (M) Images of bladder cancer tumors in mice. (N) Growth curves of bladder cancer tumors in mice. (O) Tumor weights of bladder cancer tumors in mice. ns, no significance. *p<0.05; **p<0.01; ***p<0.001.

Figure 5. METTL3 regulates bladder cancer progression by chemotactic CD8+T cell infiltration through the IGF2BP1-AHR-CCL5 axis. (A) Venn diagram illustrating the screening process for key transcription factors regulated by METTL3-mediated m6A modification and involved in CCL5 transcription. (B) Reverse Transcription Quantitative Polymerase Chain Reaction (RT-qPCR) analysis of AHR and CCL5 mRNA expression levels after AHR knockdown in MB49 cells. (C) Assessment of CCL5 mRNA expression levels after overexpression of METTL3 and/or knockdown of AHR in MB49 cells. (D) Schematic representation of AHR binding sites within the CCL5 promoter region as predicted by JASPAR. (E) CHIP-qPCR analysis of AHR enrichment at the CCL5 promoter region. (F) mRNA and (G) protein expression levels of AHR after METTL3 knockdown in MB49 cells. (H) Peak plot of m6A modification sites in AHR in MB49 cells. (I) MeRIP-qPCR analysis showing changes in AHR m6A modification levels following METTL3 knockdown in MB49 cells. (J) RIP-qPCR analysis of METTL3 enrichment in AHR mRNA in MB49 cells. (K) MeRIP-qPCR showing changes in AHR m6A modification levels after treatment with the METTL3 inhibitor STM2457 in MB49 cells. (L) RT-qPCR analysis of AHR mRNA levels after STM2457 treatment to inhibit METTL3 in MB49 cells. (M) RNA decay assay showing AHR mRNA stability after silencing METTL3. (N) RNA decay assay showing AHR mRNA stability after treatment with METTL3 inhibitor STM2457 (2 µg/mL, 72 hours) in MB49 cells. (O) RT-qPCR analysis of IGF2BP1 and AHR mRNA expression levels in MB49 cells after silencing IGF2BP1. (P) RT-qPCR analysis of IGF2BP2 and AHR mRNA expression levels in MB49 cells after silencing IGF2BP2. (Q) RT-qPCR analysis of METTL3, IGF2BP1, and AHR mRNA expression levels in MB49 cells after overexpression of METTL3 and/or silencing of IGF2BP1. (R) Images of tumors formed by MB49 stable cell lines (control, AHR overexpression, METTL3 knockdown, METTL3 knockdown with AHR overexpression) subcutaneously implanted into the backs of C57BL/6J mice. (S) Growth curves of mouse bladder cancer tumors. (T) Volume of mouse bladder cancer tumors. (U) Schematic of the animal experiment. (V) Images of bladder cancer tumors in mice. (W) Growth curves of bladder cancer tumors in mice. (X) Tumor weights of bladder cancer tumors in mice; ns, no significance. *p<0.05; **p<0.01; ***p<0.001.

Figure 6. Targeting METTL3 enhances the efficacy of anti-Programmed Cell Death Protein 1 (PD-1) immunotherapy in bladder cancer. (A) Control and METTL3-knockdown MB49 stable cell lines were subcutaneously injected into mice. Anti-PD-1 antibody (200 µg/mouse, every 3 days) was administered intraperitoneally starting on day 6. Tumors were harvested on day 12 for flow cytometric analysis of the immune microenvironment (n=5). (B–D) Images, growth curves, and tumor weights of subcutaneous bladder cancer tumors in mice. (E–F) Flow cytometric analysis of MDSCs and CD8+T cell infiltration levels in the tumor tissues of mouse bladder cancer. (G) Wild-type MB49 cells were subcutaneously injected into mice, and on day 6, the mice were randomly divided into groups. Treatment included anti-PD-1 antibody (200 µg/mouse, every 3 days, intraperitoneally), IgG antibody (200 µg/mouse, every 3 days, intraperitoneally), the METTL3 inhibitor STM2457 (250 µg/tumor, once daily, intratumorally), and a combination of STM2457 and anti-PD-1 antibody. (H, J) Images, growth curves, and tumor weights of bladder cancer tumors in mice. (K) Control or METTL3 knockdown MB49 stable cell lines were orthotopically injected into the mouse bladder wall to establish an orthotopic bladder cancer model. Anti-PD-1 antibody (200 µg/mouse, every 3 days, intraperitoneally) or IgG antibody (200 µg/mouse, every 3 days, intraperitoneally) was administered starting on day 6 (n=5). (L) In vivo imaging system (IVIS) Living imaging of tumor growth in the orthotopic bladder cancer model. (M) Images of orthotopic bladder cancer tumors in mice. (N) Statistical analysis of fluorescence signal values from IVIS Living imaging on day 16. (O) Tumor volume in the orthotopic bladder cancer model. (P) Tumor weight in the orthotopic bladder cancer model. (Q) Schematic diagram of the study content. ns, no significance. *p<0.05; **p<0.01; ***p<0.001.