Can RNA chemical modifications inside tumor cells control which immune cells get recruited to the tumor microenvironment?

RNA chemical modifications within tumor cells serve as master regulators of the tumor immune microenvironment (TME) by controlling the recruitment and exclusion of specific immune cell populations (Direct, High; PMID: 41280938). Modifications such as $N^6$-methyladenosine ($m^6A$) and 5-methylcytosine ($m^5C$) dictate the expression of chemokines, cytokines, and metabolic pathways that determine whether a tumor adopts an "immune-inflamed," "immune-excluded," or "immune-desert" phenotype.

$m^6A$ Modification and Chemokine-Mediated Recruitment

The $m^6A$ modification landscape inside tumor cells directly influences the secretion of chemokines responsible for attracting cytotoxic lymphocytes and suppressive myeloid cells.

• Effector T Cell Recruitment: Depletion of the $m^6A$ writers METTL3 or METTL14 in colorectal cancer (CRC) and melanoma cells significantly increases the recruitment of $CD8^+$ T cells by enhancing the stability of Stat1 and Irf1 mRNAs, leading to increased production of the T-cell-trafficking chemokines CXCL9 and CXCL10 (Direct, High; PMID: 32964498).
• Transcriptional Silencing of Chemokines: In non-small cell lung cancer (NSCLC), tumor-intrinsic circular RNA circFNDC3B disrupts the TFII-I/STAT1 complex, which inhibits the transcriptional activation of CXCL10 and CXCL11, thereby creating an immune-excluded microenvironment by preventing $CD8^+$ T cell infiltration (Direct, High; PMID: 40341878).
• ALKBH5 and Macrophage Infiltration: The $m^6A$ eraser ALKBH5 upregulates the secretion of CCL2 and CXCL10 in NSCLC, which recruits $PD-L1^+$ tumor-associated macrophages (TAMs) and skews them toward a pro-tumor M2 phenotype (Direct, High; PMID: 38872221).

Epitranscriptomic Control of Immunosuppressive Cell Accumulation

RNA modifications often promote an immunosuppressive TME by facilitating the recruitment and maintenance of myeloid-derived suppressor cells (MDSCs) and regulatory T cells (Tregs).

• MDSC Recruitment Axis: The $m^6A$ reader YTHDF1 promotes the translation of p65/RELA, which activates the NF-$\kappa$B pathway to induce the expression of CXCL1. This secretion recruits MDSCs via the CXCL1-CXCR2 axis in CRC, which subsequently antagonizes effector T cell function (Direct, High; PMID: 36717220).
• RBM15 and MDSC Accumulation: RBM15 facilitates the stability of the lncRNA CBR3-AS1 in an IGF2BP3-dependent manner, which sponges miR-409-3p to induce CXCL1 expression, resulting in MDSC recruitment and radioresistance in lung cancer models (Direct, High; PMID: 39962467).
• Metabolic Signaling: ALKBH5-mediated $m^6A$ demethylation in tumors regulates the lactate efflux program (via MCT4/SLC16A3). High lactate levels in the TME promote the recruitment of MDSCs and Tregs, reducing the efficacy of anti-PD-1 therapy (Direct, High; PMID: 41280938).

Recruitment Through Antigen Visibility and Metabolic Reprogramming

Modifications inside tumor cells also control immune cell movement by altering the metabolic landscape and the visibility of tumor antigens.

• MHC-I Stabilization: Tumor-intrinsic YTHDF1 enhances the translation of lysosomal protease genes; its deficiency limits the degradation of tumor antigens and MHC-I molecules. This increase in surface MHC-I expression converts "cold" immune-desert tumors into "hot" tumors with robust $CD4^+$ and $CD8^+$ T cell infiltration (Direct, High; PMID: 36650153).
• $m^5C$ and Myeloid Recruitment: Deficiency of the $m^5C$ eraser TET2 leads to increased IL-6 production in tumor tissues, which stimulates the recruitment of granulocytic MDSCs and reduces $CD8^+$ T cell numbers in the TME (Direct, High; PMID: 37488178).

Summary of Clinical Significance

Epitranscriptomic signatures, such as the "m6Ascore" or "m5Cscore," have been developed to quantify these modification patterns in individual patients (Direct, High; PMID: 37488178). Low m6A/m5C scores are generally associated with "immune-inflamed" phenotypes characterized by high lymphocyte infiltration and better responses to immune checkpoint inhibitors, whereas high scores often represent "immune-desert" or "immune-excluded" niches with poor clinical outcomes (Direct, High; PMID: 37488178).

How do $m^6A$ reader proteins specifically distinguish between transcripts that recruit effector cells versus those that recruit suppressor cells?

What experimental readouts best validate the causal link between tumor-intrinsic RNA modifications and in vivo lymphocyte trafficking?

What evidence exists for small-molecule inhibitors of RNA writers or erasers as adjuvant agents to improve immune checkpoint blockade response?

Unverified Citations

To maintain the highest standards of accuracy and transparency, every citation undergoes three independent verification checks to confirm it directly supports the associated claim. The references below did not satisfy all verification stages. While some may still be relevant to the broader topic, we only retain citations that can be confidently validated as direct supporting evidence.

• PMID:36658237 — High lactate levels in the TME promote the recruitment of MDSCs and Tregs, reducing the efficacy of anti-PD-1 therapy
  Failed: conclusion — The paper does not mention lactate levels or recruitment of MDSCs/Tregs via lactate; it focuses on STAT1 stability in macrophages.
• PMID:32164750 — Epitranscriptomic signatures, such as the "m6Ascore" or "m5Cscore," have been developed to quantify these modification p...
  Failed: entities — The paper only develops an m6Ascore and does not mention an m5Cscore.

Hypothesis 1

The METTL3-METTL14 writer complex functions as an integrated epitranscriptomic hub in colorectal cancer to maintain immune exclusion by simultaneously triggering YTHDF2-mediated decay of T-cell-recruiting transcripts and YTHDF1-mediated translation of MDSC-recruiting transcripts.

Mechanistic rationale

• The loss of the m6A writers METTL3 or METTL14 stabilizes Stat1 and Irf1 mRNAs in CRC, which normally undergo m6A-dependent degradation, thereby increasing the secretion of the CD8+ T-cell-recruiting chemokines CXCL9 and CXCL10. (Derived, Low; PMID: 32964498)
• Simultaneously, the m6A reader YTHDF1 recognizes m6A-modified RELA (p65) transcripts to enhance their translation, which activates the NF-κB pathway to induce CXCL1-mediated recruitment of immunosuppressive G-MDSCs in the colorectal TME. (Direct, High; PMID: 36717220)
• The coordination between these two axes—reducing effector cell recruitment via Stat1/Irf1 instability and increasing suppressor cell recruitment via p65/RELA translation—collectively establishes an immune-excluded or immune-desert phenotype that characterizes CRC resistance to PD-1 blockade. (Derived, Medium; PMID: 32964498, PMID: 36717220, PMID: 33500720)

Predictions

• Genetic knockout of METTL3 in microsatellite stable (MSS) CRC cells will lead to a simultaneous decrease in CXCL1 protein levels and an increase in CXCL9 and CXCL10 mRNA half-life. (Derived, Medium; PMID: 38187373)

Study design

Utilize CRISPR/Cas9 to generate METTL3-knockout and METTL3-catalytically-dead (D395A) variants in MSS CRC cell lines (e.g., CT26). Perform Ribo-seq and RNA-seq to measure translational efficiency of p65/RELA and stability of Stat1/Irf1. Evaluate the immune landscape using syngeneic mouse allografts, assessing the infiltration of CD8+ T cells and G-MDSCs via multi-color flow cytometry and spatial transcriptomics. Treat tumor-bearing mice with anti-PD-1 therapy to correlate recruitment shifts with therapeutic responsiveness. (Derived, Medium; PMID: 32964498, PMID: 36717220, PMID: 36650153)

Confounders & controls

• To control for m6A-independent functions of METTL3, include a catalytically inactive METTL3 mutant rescue to determine if catalytic writer activity is essential for both chemokine-altering pathways. (Indirect, Low; PMID: 41164187)
• Neutralize CD8+ T cells and MDSCs separately using depleting antibodies (anti-CD8 and anti-Ly6G) to quantify their relative contributions to tumor regression in METTL3-deficient models. (Derived, Medium; PMID: 41280938)

Risks/limitations

• The dual effects of m6A reader competition (YTHDF1 vs. YTHDF2) may vary across cell types or under specific stressors like hypoxia, potentially masking the phenotypic switch in certain CRC subtypes. (Indirect, Low; PMID: 41280938)

Falsification criteria

• The hypothesis is falsified if silencing METTL3 increases both CXCL1-mediated MDSC recruitment and CXCL9/10-mediated CD8+ T-cell recruitment, indicating it does not act as a selective recruitment switch. (Derived, Medium; PMID: 41280938)

Unverified Citations

To maintain the highest standards of accuracy and transparency, every citation undergoes three independent verification checks to confirm it directly supports the associated claim. The references below did not satisfy all verification stages. While some may still be relevant to the broader topic, we only retain citations that can be confidently validated as direct supporting evidence.

• PMID: 32964498 — Flow cytometric analysis of syngeneic CRC tumors lacking METTL3 will show a significantly reduced frequency of G-MDSCs (...
  Failed: entities,conclusion — The paper reports increased total CD8+ T cells but does not mention G-MDSCs, Ly6G, or effector memory T cell subsets.
  Possible alternatives (unverified): PMID:36658237 (77% topic match); PMID:39962467 (67% topic match)
• PMID: 36717220 — Flow cytometric analysis of syngeneic CRC tumors lacking METTL3 will show a significantly reduced frequency of G-MDSCs (...
  Failed: entities,conclusion — This paper provides data for G-MDSCs following YTHDF1 knockout, but not for METTL3 knockout.
  Possible alternatives (unverified): PMID:36658237 (77% topic match); PMID:39962467 (67% topic match)

Hypothesis 2

Tumor-intrinsic METTL3 establishes an immune-excluded tumor microenvironment in colorectal cancer by acting as a dual-action recruitment toggle that initiates YTHDF2-dependent degradation of the Stat1/Irf1 effector-chemoattractant axis and YTHDF1-dependent translational activation of the RELA/p65 suppressor-chemoattractant axis.

Mechanistic rationale

• The METTL3-METTL14 complex is the primary methyltransferase responsible for m6A deposition on transcripts in colorectal cancer cells. (Derived, Low; PMID: 32964498, PMID: 33500720)
• In CRC, m6A modification of Stat1 and Irf1 mRNAs leads to their destabilization and degradation via the m6A reader YTHDF2, which results in significantly reduced secretion of the effector-recruiting chemokines CXCL9 and CXCL10. (Derived, Medium; PMID: 32964498, PMID: 38187373)
• Simultaneously, RELA (p65) mRNA contains m6A sites that are recognized by the reader YTHDF1 to enhance its translational efficiency. (Direct, High; PMID: 36717220)
• Increased p65 protein levels activate the NF-kB pathway to induce the expression and secretion of CXCL1, which promotes the recruitment of immunosuppressive G-MDSCs to the colorectal tumor microenvironment. (Direct, High; PMID: 36717220)
• The coordination between these two axes—reducing CD8+ T-cell attractants and increasing MDSC attractants—collectively establishes the immune-excluded phenotype observed in high m6A-score colorectal tumors. (Derived, Medium; PMID: 33500720, PMID: 32964498, PMID: 36717220)

Predictions

• Genetic knockout of METTL3 in hot (microsatellite instability-high) CRC cells will switch their secretome from a CXCL1-high/CXCL10-low state to a CXCL1-low/CXCL10-high state. (Derived, Medium; PMID: 38187373)
• Inhibiting USP5 will decrease RELA (p65) protein expression and subsequent CXCL1 secretion by destabilizing YTHDF1, thereby reducing MDSC recruitment in vivo. (Indirect, Low; PMID: 39900921, PMID: 36717220)

Study design

Utilize CRISPR/Cas9 to generate METTL3-knockout and METTL3-catalytically-dead (D395A) variants in MSS CRC cell lines (e.g., CT26). Perform Ribo-seq to measure the translational efficiency of RELA (p65) and RNA-seq to determine the stability and half-life of Stat1/Irf1. Evaluate the immune landscape using syngeneic mouse allografts, assessing the infiltration of CD8+ T cells and G-MDSCs via multi-color flow cytometry and spatial transcriptomics. Correlate shifts in recruitment axes with therapeutic responsiveness to anti-PD-1 treatment. (Derived, Medium; PMID: 32964498, PMID: 36717220, PMID: 36650153, PMID: 37488178)

Confounders & controls

• Use a catalytically inactive METTL3 mutant rescue to distinguish m6A-dependent regulation from m6A-independent scaffolding functions. (Indirect, Low; PMID: 41164187)
• Perform CD8+ T cell and MDSC depletion using neutralizing antibodies (anti-CD8 and anti-Ly6G) to quantify the relative contribution of each axis to the tumor-suppressive effect of METTL3 inhibition. (Derived, Medium; PMID: 32964498, PMID: 36717220, PMID: 36658237)

Risks/limitations

• Competition between m6A readers (e.g., YTHDF1 vs. IGF2BPs) for the same transcript binding sites may vary under stress conditions like hypoxia, potentially masking the phenotypic outcome. (Indirect, Low; PMID: 41280938, PMID: 41049442)

Falsification criteria

• The hypothesis is falsified if silencing METTL3 simultaneously increases both CXCL1-mediated MDSC recruitment and CXCL10-mediated CD8+ recruitment, indicating it does not act as a selective recruitment switch. (Indirect, Low; PMID: 41280938)