METTL3 promotes pancreatic cancer immune escape by m6A modification of lncRNA AFAP1-AS1 to enhance EGR2 stability

Abstract

Background and aims

Pancreatic cancer (PC) exhibits significant immune evasion. The roles of N6-methyladenosine (m6A) RNA modification and long non-coding RNAs (lncRNAs) in PC immunity are emerging. We aimed to elucidate the function and mechanism of METTL3-mediated m6A modification of lncRNA AFAP1-AS1 in PC progression and immune escape, focusing on its interaction with early growth response 2 (EGR2).

Methods

Expression of METTL3, AFAP1-AS1, and EGR2, along with AFAP1-AS1 m6A levels, were measured in human PC tissues/cell lines via quantitative reverse transcription PCR, Western blot, and methylated RNA immunoprecipitation. Functional assays (proliferation, invasion, CD8⁺ T cell co-culture) were performed using knockdown/overexpression strategies in PANC-1 cells. Mechanistic insights were gained through RNA immunoprecipitation, stability assays, chromatin immunoprecipitation (ChIP), and fluorescence in situ hybridization. An in vivo murine subcutaneous tumor model validated key findings, with comprehensive immune microenvironment profiling including regulatory T cells (Tregs) and tumor-associated macrophages (TAMs).

Results

METTL3, AFAP1-AS1, and EGR2 were significantly upregulated in PC, where AFAP1-AS1 exhibited increased m6A modification and correlated with poor prognosis. METTL3 directly mediated AFAP1-AS1 m6A modification, enhancing its stability through the m6A readers YTHDF2 and IGF2BP1. Functionally, AFAP1-AS1 promoted PC cell malignancy and immune evasion, characterized by resistance to T cell killing, suppressed T cell function, and increased programmed death-ligand 1 (PD-L1). Mechanistically, AFAP1-AS1 stabilized EGR2 mRNA through direct binding, elevating EGR2 protein expression. ChIP assays confirmed that EGR2 directly binds to the promoters of PD-L1 and Snail. EGR2 knockdown phenocopied AFAP1-AS1 knockdown effects, and EGR2 overexpression rescued the AFAP1-AS1 knockdown phenotype. In vivo, METTL3 depletion inhibited tumor growth and enhanced anti-tumor immunity, characterized by decreased Tregs and M2 macrophages infiltration, effects partially reversed by EGR2 restoration.

Conclusion

METTL3 enhances AFAP1-AS1 stability via m6A modification recognized by YTHDF2 and IGF2BP1, which subsequently increases EGR2 expression by improving its mRNA stability. EGR2 directly activates transcription of immune checkpoint and epithelial-mesenchymal transition (EMT)-related genes. This METTL3/AFAP1-AS1/EGR2 signaling axis drives PC malignancy and immune resistance, offering potential therapeutic targets for PC treatment.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40001-026-03963-3.

Introduction

Pancreatic cancer (PC), as one of the most aggressive types of malignant tumors in the digestive system, poses a serious public health challenge globally with its extremely high mortality rate and difficulty in treatment [1]. Despite the advances in modern medical technology, the five-year survival rate of PC is still less than 10%, making it one of the solid tumors with the worst prognosis [2]. This worrying clinical situation mainly stems from the biological characteristics of PC, including the difficulty of early diagnosis, highly invasive nature and significant treatment resistance [3]. In recent years, an increasing number of studies have shown that immune escape mechanisms play a central role in PC treatment failure [4]. The microenvironment of PC forms a typical “cold” tumor ecosystem characterized by a dense stromal barrier, an abnormal vascular network, and an infiltration of immunosuppressive cells, which collectively limit the penetration and function of immune cells, thus providing a refuge for tumor cells to escape from immune surveillance [5]. The complex interplay between tumor cells and the immune microenvironment in PC involves multiple layers of regulation, including epigenetic modifications that reshape both tumor intrinsic properties and immune cell functions [6]. Therefore, unraveling and intervening in the molecular mechanisms of immune escape in PC has become a key breakthrough for improving therapeutic outcomes.

At the molecular level, epigenetic regulation plays a crucial role in tumorigenesis and development [7]. As an important branch of epitranscriptome research, RNA modification has received extensive attention in recent years. Among them, N6-methyladenine (m6A) modification, as the most abundant and functionally diverse mRNA modification, has been shown to be involved in the regulation of key processes, such as RNA splicing, translocation, translation and degradation [8]. The dynamic balance of m6A modification is precisely regulated by three types of proteins: “writers” (e.g., METTL3, METTL14) responsible for adding methyl groups, ‘erasers’ (e.g., FTO, ALKBH5) responsible for removing methyl groups, and “readers” (e.g., YTHDF1-3) responsible for recognizing the modification sites [9]. As the catalytic core of the m6A methyltransferase complex, METTL3 is abnormally expressed in a variety of tumors, and its regulation of tumor biological behavior has become a hot research topic [10, 11]. Notably, METTL3 not only directly affects the malignant phenotype of tumor cells, but also participates in the process of tumor immune escape by remodeling the tumor microenvironment and regulating T cell function [12, 13]. Recent studies have demonstrated that m6A modification plays a pivotal role in regulating the tumor immune microenvironment, influencing both innate and adaptive immune responses through modulation of immune cell infiltration and function [14]. However, the specific mechanism of METTL3’s role in PC immune escape remains unclear, especially the downstream effector molecular network it regulates needs to be further elucidated.

Long-stranded non-coding RNAs (lncRNAs) are a class of RNA molecules that are more than 200 nucleotides in length and do not encode proteins, which regulate gene expression at the transcriptional and post-transcriptional levels [15]. AFAP1-AS1 is a tumor-associated lncRNA that is significantly overexpressed in a variety of malignant tumors, including PC [16]. AFAP1-AS1 promotes the proliferation, migration and invasive ability of tumor cells by regulating the EMT process, cell cycle, and multiple signaling pathways [17]. LncRNAs can be similarly regulated by m6A modifications, which can significantly affect the stability, subcellular localization, and interaction ability of lncRNAs, thereby altering their biological functions [18]. Importantly, accumulating evidence suggests that m6A-modified lncRNAs serve as critical mediators in tumor progression and immune modulation, representing a novel layer of epitranscriptomic regulation in cancer [19]. This finding provides a new perspective for understanding the mechanism of lncRNAs in tumors. However, whether m6A modification regulates the expression and function of AFAP1-AS1, and especially the potential role of this regulation in PC immune escape remains an urgent scientific question to be explored.

The role of transcription factors is gaining attention in the study of molecular mechanisms of tumor immune escape [20, 21]. Early growth response factor 2 (EGR2) is a zinc-finger transcription factor involved in the regulation of gene expression networks in multiple cell types [22]. In the immune system, EGR2 has a dual function: on the one hand, it is involved in T-cell activation, and on the other hand, it promotes T-cell depletion and immune tolerance in response to sustained antigenic stimulation [23]. There is increasing evidence that EGR2 upregulates the expression of key immune checkpoint molecules, such as PD-1 and PD-L1, in the tumor microenvironment and inhibits the anti-tumor activity of T cells, thus creating an immunosuppressive environment for tumor cells [24]. In PC, the aberrant expression of immune checkpoint molecules is closely related to its typical “cold” tumor characteristics and immunotherapy resistance [25, 26]. Although the importance of EGR2 in immune regulation has been recognized, the upstream molecular mechanisms regulating EGR2 expression, especially the regulatory network in the context of immune escape from PC, remain unclear.

This study aims to systematically elucidate the role and molecular mechanism of METTL3-mediated m6A modification of AFAP1-AS1 in PC progression and immune escape, with a special focus on the complete regulatory network of METTL3/AFAP1-AS1/EGR2 signaling axis. This study is not only expected to answer this basic scientific question, but also to provide new molecular targets for precision immunotherapy of PC, which will ultimately improve the prognosis and survival quality of PC patients.

Materials and methods

Clinical samples

Fifty-six pairs of pathologically-confirmed PC tissues and paired adjacent normal tissues (> 2 cm from the cancerous tissues) were collected at Affiliated Jinling Hospital, Medical School of Nanjing University between 2022 and 2024. None of the patients had received radiotherapy, chemotherapy, or other relevant antitumor therapy prior to surgery. The study was approved by the Ethics Committee of Affiliated Jinling Hospital, Medical School of Nanjing University and all patients signed a written informed consent. Tissue samples were rapidly frozen in liquid nitrogen immediately and subsequently transferred to an ultra-low temperature refrigerator at − 80 °C for subsequent RNA and protein extraction analysis.

Cell culture

Human PC cell lines PANC-1, Capan-1, BxPC-3 and normal human pancreatic ductal epithelial cell line HPNE were purchased from National Collection of Authenticated Cell Cultures (Shanghai, China). Murine PC cell line Panc02 was obtained from ATCC (Manassas, VA, USA). All cells were cultured in DMEM (Gibco) containing 10% fetal bovine serum (Gibco) and 1% penicillin–streptomycin solution (Gibco). The culture conditions were 37 °C, 5% CO₂ in a saturated humidity incubator. All cell lines were characterized by STR before use and were regularly tested for mycoplasma contamination.

Quantitative reverse transcription PCR (RT-qPCR)

Total RNA was extracted from cell or tissue samples using TRIzol™ Reagent (Invitrogen, CA, USA). RNA concentration and purity were determined using a NanoDrop™ One microspectrophotometer (Thermo Fisher Scientific, MA, USA). For mRNA and lncRNA assays, 1 μg of total RNA was reverse transcribed to cDNA using the PrimeScript™ RT Reagent Kit with gDNA Eraser (Takara, Japan). RT-qPCR reactions were performed on an ABI 7500 Fast Real-Time PCR System (Applied Biosystems) using TB Green^® Premix Ex Taq™ II (Tli RNaseH Plus) (Takara). GAPDH was taken as an internal reference gene for normalization. The relative expression levels of the genes were calculated using the 2^−ΔΔCt method. All primers were synthesized by Sangon (Shanghai, China).

Western blot

Cells or tissues were lysed on ice using RIPA lysis buffer (Beyotime, Shanghai, China) containing protease and phosphatase inhibitora. Protein concentration was determined using the BCA Protein Concentration Assay Kit (Beyotime). An equal amount of protein (30 μg) was taken for SDS-PAGE, followed by transferring the protein to a PVDF membrane (MilliporeSigma, MA, USA). The membrane was blocked for 1 h at room temperature in TBST buffer containing 5% skim milk powder and incubated overnight at 4 °C with the corresponding primary antibodies: anti-METTL3 (Cat# ab195352, Abcam), anti-EGR2 (Cat# ab108399, Abcam), anti-GAPDH (Cat# 60,004–1-Ig, Proteintech), PD-1 (Cat# ab52587, Abcam), PD-L1 (Cat# 13,684, Cell Signaling Technology), anti-E-cadherin (Cat# 3195, Cell Signaling Technology), anti-N-cadherin (Cat# 13,116, Cell Signaling Technology), anti-YTHDF2 (Cat# ab220163, Abcam), anti-IGF2BP1 (Cat# ab184305, Abcam), anti-Ubiquitin (FK2) (Cat# 04–263, MilliporeSigma), anti-phospho-Serine/Threonine (Cat# ab17464, Abcam). After TBST washing, the membrane was incubated with horseradish peroxidase (HRP)-labeled secondary antibody (IgG, Cat# SA00001-1, Proteintech) for 1 h at room temperature. Protein bands were detected on the ChemiDoc™ Imaging System (Bio-Rad, CA, USA) using enhanced chemiluminescence detection reagents (Thermo Fisher Scientific). Bands were analyzed in gray scale values using ImageJ software, normalized using GAPDH as an internal reference.

Lentiviral infection

To construct PC cell lines that stably knockdown or overexpress the target genes, lentivirus-mediated gene transfer techniques were applied. Lentiviral vectors (carrying shRNAs targeting specific sequences) for stable knockdown of METTL3, AFAP1-AS1, YTHDF2, IGF2BP1 and EGR2, as well as a negative control shRNA vector (sh-NC) were purchased from Genechem (Shanghai, China). Lentiviral expression vectors for stable overexpression of METTL3 (CDS), AFAP1-AS1 (full-length sequence), and EGR2 (CDS) and their corresponding empty vector controls (Vector) were also purchased from Genechem. All vectors contained the Puromycin resistance gene. Transfection was performed using Lipofectamine™ 3000 (Invitrogen). Cell culture supernatants containing lentiviral particles were collected 48 h after transfection and filtered through a 0.45 μm filter to remove cellular debris. Viral fluids were concentrated by Virus Concentration Reagent (Lenti-X™ Concentrator, Takara), and viral titers were determined by gradient dilution of infected HEK-293T cells and combined with puromycin screening counts. PANC-1 cells were inoculated in 6-well plates. When the confluence reached approximately 50–70%, the medium was replaced with fresh complete medium containing lentiviral particles and polybrene (Sigma-Aldrich, final concentration at 8 μg/mL). After 24 h of infection, the medium was replaced a fresh complete medium. After 38 h of infection, puromycin-containing medium was added for screening. Fresh puromycin-containing medium was replaced every 2–3 days for 2 weeks until the uninfected cells were completely dead. The surviving cells were the cell population that stably expressed shRNA or overexpressed the target gene.

The down-regulation efficiency of the target genes (METTL3, AFAP1-AS1, YTHDF2, IGF2BP1, EGR2) in the stable knockdown cell lines and the up-regulation efficiency in the stable overexpression cell lines were detected by RT-qPCR. For genes encoding proteins (METTL3, YTHDF2, IGF2BP1, EGR2), corresponding changes in protein levels were detected by Western Blot to confirm the knockdown or overexpression effects. The validated stable cell lines were used for subsequent functional experiments and animal model studies.

Nucleoplasmic separation assay

Nuclei and cytoplasmic fractions of PANC-1 cells were separated using the PARIS™ Kit Protein and RNA Isolation System (Thermo Fisher Scientific). Cell precipitates were resuspended with ice-cold Cell Separation Buffer, and nuclei were precipitated by low-speed centrifugation after cell lysis. The supernatant was aspirated. The nuclei were lysed with Cell Breaking Buffer. RNA was extracted from cytoplasmic and nuclear fractions, respectively. The relative distribution of AFAP1-AS1 in the nucleus and cytoplasm was detected by RT-qPCR. GAPDH and U6 were used as cytoplasmic and nuclear controls, respectively.

Fluorescence in situ hybridization (FISH) and immunofluorescence (IF) co-localization

To detect the intracellular co-localization of AFAP1-AS1 and METTL3, FISH combined with IF was used. PANC-1 cells were inoculated in culture plates, fixed with 4% paraformaldehyde (Beyotime) at room temperature, and washed with PBS. The cells were permeabilized with 0.5% Triton X-100 (Sigma-Aldrich) at room temperature, closed with pre-hybridization buffer for 30 min at 37 °C, and added with AFAP1-AS1 specific probe labeled with Cy3 (RiboBio, Guangzhou, China) for hybridization overnight at 37 °C protected from light. On the following day, unbound probes were removed by washing with SSC buffer. Subsequently, IF staining was performed. The membrane was blocked with PBST blocking solution containing 5% BSA for 1 h at room temperature, and anti-METTL3 primary antibody was added and incubated overnight at 4 °C. After washing with PBS, the membrane was incubated with fluorescent secondary antibody (Invitrogen) for 1 h at room temperature, protected from light, sealed with a sealer containing DAPI (Invitrogen), and observed under a laser confocal microscope (Zeiss LSM 880, Carl Zeiss AG, Germany).

RIP analysis

RIP experiments were performed using Magna RIP™ RNA-Binding Protein Immunoprecipitation Kit (17–700, MilliporeSigma). Cells (1 × 10⁷) were lysed with RIP lysis buffer. A portion of the lysate was taken as Input control. The remaining lysate was incubated with magnetic beads pre-coated with anti-METTL3 or negative control IgG (2729, Cell Signaling Technology) at 4 °C overnight. RNA was released by treating the magnetic beads with proteinase K digestion buffer and purified using TRIzol. Fold enrichment of AFAP1-AS1 (relative to Input and IgG controls) was detected by RT-qPCR.

MeRIP analysis

The m6A modification level of AFAP1-AS1 was detected using the Magna MeRIP™ m6A Kit (Cat# 17–10,499, MilliporeSigma). Total RNA was extracted and broken using RNA fragmentation buffer. A portion of the fragmented RNA was taken as Input control. The remaining fragmented RNA was incubated with magnetic beads pre-coated with anti-m6A (Cat# 202,003, Synaptic Systems, Goettingen, Germany) or negative control IgG at 4 °C overnight. The m6A-modified RNA fragments were eluted with elution buffer. After purification of RNA, m6A enrichment of AFAP1-AS1 was detected by RT-qPCR.

m6A dot blot analysis

Total cellular RNA was extracted, and an equal amount of RNA was diluted in DEPC solution, denatured at 70 °C for 5 min and then quickly placed on ice. RNA was spotted onto Hybond-N⁺ nylon membrane (Amersham/GE Healthcare, IL, USA) and crosslinked at 254 nm using a UV crosslinker (Stratalinker UV Crosslinker, Stratagene, CA, USA). The membrane was blocked with TBST containing 5% skim milk powder for 1 h at room temperature, followed by overnight incubation with anti-m6A (Synaptic Systems, Goettingen, Germany) at 4 °C. After TBST washing, the membrane was incubated with HRP-labeled secondary antibody for 1 h at room temperature and detected using ECL reagent.

Luciferase reporter gene assay

The AFAP1-AS1 fragment (WT) containing the predicted m6A site was cloned into the luciferase reporter vector (psiCHECK™−2 Vector, Promega, WI, USA) of the firefly luciferase or renilla luciferase genes in the 3′-UTR. The m6A site was mutated by targeted mutagenesis and a mutant (MUT) reporter plasmid was constructed. WT or MUT reporter plasmids were transfected into PC cells stably knocking down or overexpressing METTL3. After 48 h, cells were lysed and assayed using the Dual-Luciferase^® Reporter Assay System (Cat# E1910, Promega). The effect of METTL3-mediated m6A modification on the stability of the AFAP1-AS1 fragment was assessed by calculating the relative luciferase activity.

mRNA stability analysis

To directly assess the effect of AFAP1-AS1 on EGR2 mRNA stability, transcriptional repression assays were performed. PC cells pre-stabilized by lentiviral infection to knockdown or overexpress AFAP1-AS1 were cultured to logarithmic growth phase. The transcriptional inhibitor Actinomycin D (Sigma-Aldrich, 5 μg/mL) was added to block mRNA synthesis, and cells were collected at different time points (0, 2, 4, 6 h). EGR2 mRNA at each time point was detected by RT-qPCR.

RNA pull-down assay

Biotin-labeled AFAP1-AS1 probe and its corresponding antisense probe (Antisense) were purchased from RiboBio. PC cell lysates were incubated with equimolar amounts of biotin-labeled probes. The probes and their bound RNA complexes were captured using Streptavidin magnetic beads (Invitrogen™ Dynabeads™ MyOne™ Streptavidin C1). The bound RNA was eluted, and the enrichment of EGR2 was detected by RT-qPCR.

Protein stability assay

To directly assess the effect of AFAP1-AS1 on the stability of EGR2 proteins, the CHX Chase Assay was performed. PC cells were cultured to logarithmic growth phase, treated by adding the protein synthesis inhibitor cycloheximide (Sigma-Aldrich, 50 μg/mL), and collected at different time points (0, 2, 4, 6 h). EGR2 protein at each time point was detected by Western Blot.

Chromatin immunoprecipitation (ChIP) assay

ChIP assays were performed using the SimpleChIP^® Enzymatic Chromatin IP Kit (Cat# 9003, Cell Signaling Technology) according to the manufacturer’s instructions. Briefly, PANC-1 cells (1 × 10⁷) were cross-linked with 1% formaldehyde for 10 min at room temperature, followed by quenching with glycine. Cells were lysed and chromatin was digested with Micrococcal Nuclease to obtain DNA fragments of 150–900 bp. The chromatin was immunoprecipitated overnight at 4 °C with anti-EGR2 antibody (Cat# ab108399, Abcam) or normal rabbit IgG (negative control). Protein G magnetic beads were used to capture the antibody-chromatin complexes. After washing and elution, cross-links were reversed and DNA was purified. The enrichment of EGR2 at the promoter regions of target genes (PD-L1, Snail, Slug) was analyzed by qPCR. Primers targeting the EGR2 binding sites in the promoter regions were designed based on bioinformatics prediction. The results were normalized to input DNA and expressed as fold enrichment relative to IgG control.

mmunoprecipitation (IP) for post-translational modifications

To detect ubiquitination and phosphorylation of EGR2 protein, IP experiments were performed. PANC-1 cells with stable knockdown or overexpression of AFAP1-AS1 were lysed in IP lysis buffer (Thermo Fisher Scientific) containing protease inhibitors, phosphatase inhibitors, and 10 mM N-ethylmaleimide (NEM; Sigma-Aldrich) to preserve ubiquitin modifications. Cell lysates (1 mg total protein) were pre-cleared with Protein A/G magnetic beads (Thermo Fisher Scientific) for 1 h at 4 °C. The pre-cleared lysates were incubated with anti-EGR2 antibody (Cat# ab108399, Abcam) or normal rabbit IgG overnight at 4 °C, followed by incubation with Protein A/G magnetic beads for 2 h at 4 °C. The beads were washed five times with IP lysis buffer and eluted with 2 × SDS sample buffer. The immunoprecipitated proteins were analyzed by Western blot using anti-Ubiquitin (FK2) antibody (Cat# 04–263, MilliporeSigma) to detect ubiquitination and anti-phospho-Serine/Threonine antibody (Cat# ab17464, Abcam) to detect phosphorylation.

Isolation, activation and co-culture of CD8⁺ T cells

Peripheral blood mononuclear cells (PBMCs) were isolated from peripheral blood of healthy volunteers using density gradient centrifugation. CD8⁺ T cells were purified by magnetic bead sorting using CD8⁺ T Cell Isolation Kit, human (Miltenyi Biotec, Bergisch Gladbach, Germany). Sorted cells were detected by flow cytometry (BD FACSCanto™ II, BD Biosciences, CA, USA) using anti-CD3-FITC (Cat# 300,406, BioLegend, CA, USA) and anti-CD8-APC (Cat# 301,014, BioLegend) to detect the purity of the sorted cells. Purified CD8⁺ T cells were cultured in RPMI-1640 medium containing 10% FBS and 1% penicillin–streptomycin. CD8⁺ T cells were activated using anti-CD3 (Cat# 16–0037-85,/Thermo Fisher Scientific, 1 μg/mL) and anti-CD28 (Cat# 16–0289-85, eBioscience, 1 μg/mL) as well as rhIL-2 (Cat# 200-02, PeproTech, Cranbury, USA, 20 ng/mL).

The infected PANC-1 cells were used as target cells. The activated CD8⁺ T cells were mixed with the target cells in a 10:1 ratio and co-cultured in 6-well plates for 48 h. At the end of co-culture, the T cells were gently washed away, the remaining adherent cancer cells were stained with 0.1% crystal violet (Beyotime), and the number of surviving cancer cells was photographed and quantitatively analyzed.

T cell proliferation

Activated CD8⁺ T cells were labeled with 5 μM CFSE (Cayman Chemical, Ann Arbor, USA), co-cultured with cancer cells, and stained with CD8 surface markers. T cell proliferation was assessed by flow cytometry to detect the dilution of CFSE fluorescence in CD8⁺ T cells.

ELISA

Co-culture supernatants were collected and assayed for concentrations of key cytokines (IFN-γ, TNF-α) and cytotoxic molecules (Perforin, Granzyme B) using ELISA kits (Invitrogen).

Animal experiments

All animal experiments were performed in strict compliance with the protocol approved by the Animal Care and Use Committee of Affiliated Jinling Hospital, Medical School of Nanjing University. Six-eight weeks old male C57BL/6J mice (immunocompetent mice) were selected (Hunan SJA Laboratory Animal Co., Ltd). Since no conserved genes were found for AFAP1-AS1 in mice, only the effects of METTL3 and EGR2 on tumor immune escape in mice were evaluated. A murine-derived PC cell line (Panc02, ATCC) expressing the corresponding shRNA or overexpression vector was used. Cells (1 × 10⁶ cells) were injected subcutaneously into the right axilla of mice. Groups were randomized (n = 6/group). Mouse body weight and health status were monitored. At the endpoint of the experiment (4 weeks), mice were euthanized, tumor tissues were collected, weighed, and photographed. PD-L1 in tumor tissues was detected by western blot.

Flow cytometry analysis of tumor-infiltrating immune cells

At the experimental endpoint, tumor tissues were harvested and processed into single-cell suspensions. Briefly, tumors were minced and digested with 1 mg/mL collagenase IV (Sigma-Aldrich) and 100 μg/mL DNase I (Roche, Basel, Switzerland) in RPMI-1640 medium at 37 °C for 45 min with gentle agitation. The digested tissues were filtered through 70 μm cell strainers and red blood cells were lysed using ACK lysis buffer (Gibco). The resulting single-cell suspensions were washed twice with PBS containing 2% FBS. For surface staining, cells were incubated with Fc receptor blocking antibody (anti-CD16/CD32, Cat# 553,142, BD Biosciences) for 10 min at 4 °C, followed by staining with fluorochrome-conjugated antibodies for 30 min at 4 °C in the dark. For intracellular Foxp3 staining, cells were fixed and permeabilized using the Foxp3/Transcription Factor Staining Buffer Set (Cat# 00–5523-00, eBioscience) according to the manufacturer’s protocol. The following antibodies were used: anti-CD45-BV510 (Cat# 103,138, BioLegend), anti-CD3-FITC (Cat# 100,204, BioLegend), anti-CD4-PE-Cy7 (Cat# 100,422, BioLegend), anti-CD8a-APC (Cat# 100,712, BioLegend), anti-CD25-PE (Cat# 102,008, BioLegend), anti-Foxp3-APC (Cat# 126,406, BioLegend), anti-CD11b-FITC (Cat# 101,206, BioLegend), anti-F4/80-PE (Cat# 123,110, BioLegend), anti-CD206-APC (Cat# 141,708, BioLegend), and anti-iNOS-PE-Cy7 (Cat# 25–5920-82, eBioscience). Data were acquired on a BD LSRFortessa flow cytometer (BD Biosciences) and analyzed using FlowJo software (Tree Star, Ashland, OR, USA). Tregs were identified as CD45⁺CD3⁺CD4⁺CD25⁺Foxp3⁺ cells, M1 macrophages as CD45⁺CD11b⁺F4/80⁺iNOS⁺ cells, and M2 macrophages as CD45⁺CD11b⁺F4/80⁺CD206⁺ cells.

Statistical analysis

All experiments were performed with at least three biological replicates. Data are expressed as mean ± standard deviation and analyzed using GraphPad Prism software (San Diego, CA, USA). Comparisons between two groups were made using Student’s t-test (two-tailed). Comparisons among multiple groups were performed using one-way ANOVA with Tukey’s HSD test as needed. Survival analyses were performed using the Kaplan–Meier method. P < 0.05 was considered statistically significant.

Results

AFAP1-AS1 shows high expression and significant m6A modification in PC, correlating with poor prognosis

To explore the role of AFAP1-AS1 in PC, we first analyzed its clinical relevance. By querying a public database (http://www.noncode.org), we determined that AFAP1-AS1 is located on chromosome chr4: 7,754,076–7,778,928 [+], with a length of 6808 bp (Fig. 1A). RT-qPCR assay results showed that compared with adjacent normal tissues, AFAP1-AS1 was significantly up-regulated in PC tissues (Fig. 1B). Further detection by MeRIP assay revealed that the m6A modification level of AFAP1-AS1 was likewise significantly higher in PC tissues than in adjacent normal tissues (Fig. 1C). AFAP1-AS1 expression was also significantly elevated in three PC cell lines compared with the normal human pancreatic ductal epithelial cell line HPNE (Fig. 1D). Analysis using the GEPIA database further confirmed that AFAP1-AS1 was abnormally highly expressed in pancreatic adenocarcinoma (PAAD) tissues (Fig. 1E), and its high expression level was significantly associated with worse survival prognosis of patients (Fig. 1F). AFAP1-AS1 showed an increasing trend with the progression of PAAD tumor stage (stage I-IV) (Fig. 1G). Taken together, these results suggest that AFAP1-AS1 expression is upregulated and m6A modification is enhanced in PC, and is closely associated with disease progression and poor patient prognosis.

Fig. 1.

AFAP1-AS1 shows high expression and significant m6A modification in PC, correlating with poor prognosis. A Gene information of AFAP1-AS1 in the bioinformatics website http://www.noncode.org. B RT-qPCR detected the expression of AFAP1-AS1 in PC tissues and adjacent normal tissues (n = 56 pairs). C MeRIP detected m6A modification of AFAP1-AS1 in PC tissues and adjacent normal tissues (n = 56 pairs). D RT-qPCR detected AFAP1-AS1 expression in 3 PC cell lines and normal human pancreatic ductal epithelial cell line HPNE (n = 3). E GEPIA database analysis of AFAP1-AS1 expression in PAAD. F GEPIA database analysis of the correlation between AFAP1-AS1 expression and survival prognosis in PAAD. G GEPIA database analysis of the correlation between AFAP1-AS1 expression and tumor stage (I-IV). Data are expressed as mean ± SD. *P < 0.05

AFAP1-AS1 promotes malignant biological behavior and immune escape in PC cells

To elucidate the function of AFAP1-AS1 in PC progression, we constructed cell models with stable knockdown and overexpression of AFAP1-AS1 in PANC-1 cells (Fig. 2A). Functional experiments showed that AFAP1-AS1 overexpression significantly promoted the clone-forming ability (Fig. 2B) and proliferative activity (Fig. 2C) of PANC-1 cells, whereas AFAP1-AS1 knockdown produced the opposite effect. Transwell assay results further showed that AFAP1-AS1 overexpression enhanced cell invasion and migration abilities, and AFAP1-AS1 knockdown inhibited these abilities (Fig. 2D). Western blot assay revealed that AFAP1-AS1 overexpression downregulated the epithelial marker E-cadherin, while upregulating the mesenchymal marker N-cadherin, suggesting that AFAP1-AS1 may induce EMT (Fig. 2E). Next, we explored the effect of AFAP1-AS1 on the tumor immune microenvironment. CD3⁺CD8⁺ T cells were successfully isolated and characterized (Fig. 2F). In the co-culture system of cancer cells and T cells, the number of surviving cancer cells with AFAP1-AS1 overexpression was increased, suggesting that their ability to resist T cell killing was enhanced, while knockdown of AFAP1-AS1 made cancer cells more sensitive to T cell killing (Fig. 2G). Flow cytometry detection of CFSE dilution showed that the proliferation of CD8⁺ T cells was inhibited when co-cultured with AFAP1-AS1-overexpressing cancer cells, whereas co-culturing with cancer cells knocked down AFAP1-AS1 promoted T cell proliferation (Fig. 2H). ELISA of the co-culture supernatants revealed that AFAP1-AS1 overexpression led to a significant decrease in the levels of all cytotoxic factors secreted by T cells, IFN-γ, TNF-α, Perforin, and Granzyme B. AFAP1-AS1 knockdown led to an increase in the levels of these factors (Fig. 2I). Importantly, Western blot assay showed that AFAP1-AS1 overexpression upregulated PD-L1 in T cells (Fig. 2J). Together, these data reveal that AFAP1-AS1 not only promotes the proliferation, invasion and migration of PC cells, but also promotes immune escape.

Fig. 2.

AFAP1-AS1 promotes PC cell proliferation, invasion, migration and immune escape. PANC-1 cell lines with stable knockdown or overexpression of AFAP1-AS1 were constructed. A RT-qPCR verified the stable knockdown and overexpression of AFAP1-AS1 in PANC-1 cells (n = 3). B Colony formation assay detected the number of cloned cells (n = 3). C EdU assay detected cell proliferation (n = 3). D Transwell assay detected cell invasion and migration ability (n = 3). E Western blot detected protein expression of cellular E-cadherin and N-cadherin (n = 3). F Flow cytometry sorted CD3⁺CD8⁺-positive cells. G Crystal violet staining assessed the number of cancer cells surviving in the co-culture of T cells and cancer cells (n = 3). H Flow cytometry detected the dilution of CFSE fluorescence in CD8⁺ T cells and assessed the proliferation of T cells (n = 3). I ELISA detected the levels of IFN-γ, TNF-α, Perforin, and Granzyme B in the co-culture system (n = 3). J Western blot detected protein expression of PD-L1 in T cells (n = 3). Data are expressed as mean ± SD. *P < 0.05

METTL3 mediates m6A modification of AFAP1-AS1

Considering that the m6A modification level of AFAP1-AS1 was elevated in PC tissues, we further explored the regulatory mechanism, focusing on the m6A methyltransferase METTL3. Western blot and RT-qPCR analyses showed that METTL3 expression was higher in cancer tissues (Fig. 3A, B). Similarly, METTL3 protein expression was higher in PC cell lines than in normal HPNE cells (Fig. 3C). To determine the potential interactions between METTL3 and AFAP1-AS1, we performed subcellular localization analysis. Both nucleoplasmic separation experiments and FISH experiments confirmed that METTL3 protein and AFAP1-AS1 co-localized in the cytoplasm in PANC-1 cells (Fig. 3D, E). Functionally, knockdown of METTL3 resulted in decreased m6A levels of total RNA in PANC-1 cells as detected by m6A antibody spot blot (Fig. 3F). In addition, MeRIP-qPCR further confirmed that knockdown of METTL3 significantly suppressed the m6A modification level of AFAP1-AS1 (Fig. 3G). Luciferase reporter assays further showed that knockdown or overexpression of METTL3-mediated changes in m6A modification affected the stability of AFAP1-AS1 (Fig. 3H). To confirm the direct interaction, RIP experiments confirmed that the METTL3 protein was able to directly bind AFAP1-AS1 (Fig. 3I). Thus, these results suggest that METTL3 expression is up-regulated in PC and directly interacts with AFAP1-AS1 in the cytoplasm, and that METTL3 regulates the m6A modification level of AFAP1-AS1 and its stability.

Fig. 3.

METTL3 mediates AFAP1-AS1 m6A modification. A Western blot detected METTL3 protein expression in PC tissues and adjacent normal tissues (n = 56 pairs). B RT-qPCR detected METTL3 expression in PC tissues and adjacent normal tissues (n = 56 pairs). C Western blot detected METTL3 protein expression in three PC cell lines and the normal human pancreatic ductal epithelial cell line HPNE (n = 3). D Nucleoplasmic separation assay confirmed the co-localization of METTL3 and AFAP1-AS1 in PANC-1 cells (n = 3). E FISH assay confirmed the co-localization of METTL3 and AFAP1-AS1 in PANC-1 cells (n = 3). F m6A antibody dot blot detected m6A level of total RNA in PANC-1 cells with knockdown of METTL3 (n = 3). G MeRIP detected METTL3-mediated AFAP1-AS1 m6A modification (n = 3). H Luciferase reporter assay analyzed the effect of knockdown or overexpression of METTL3-mediated m6A modification on AFAP1-AS1 stability (n = 3). I RIP assay detected the direct interaction between METTL3 and AFAP1-AS1 (n = 3). Data are expressed as mean ± SD. *P < 0.05

YTHDF2 and IGF2BP1 mediate the stabilization of m6A-modified AFAP1-AS1

To identify the specific m6A readers mediating the stabilization of AFAP1-AS1, we performed systematic RIP experiments with antibodies against major m6A reader proteins. The results showed that among the YTHDF family members (YTHDF1, YTHDF2, YTHDF3) and IGF2BP family members (IGF2BP1, IGF2BP2, IGF2BP3), YTHDF2 and IGF2BP1 exhibited significant enrichment of AFAP1-AS1 compared to IgG control in PANC-1 cells (Supplementary Figure S1A). Other readers showed minimal or no significant binding to AFAP1-AS1. Importantly, knockdown of METTL3 significantly reduced the binding of both YTHDF2 and IGF2BP1 to AFAP1-AS1 (Supplementary Figure S1B), indicating that this interaction is dependent on m6A modification.

To further validate the functional roles of YTHDF2 and IGF2BP1 in AFAP1-AS1 stability, we constructed PANC-1 cells with stable knockdown of YTHDF2 and IGF2BP1 (Supplementary Figure S1C, D). Actinomycin D treatment assays revealed that knockdown of either YTHDF2 or IGF2BP1 alone partially decreased AFAP1-AS1 stability (Supplementary Figure S1E, F). Correspondingly, the expression levels of AFAP1-AS1 and its downstream target EGR2 were decreased upon knockdown of YTHDF2 or IGF2BP1 (Supplementary Figure S1G, H). These results demonstrate that YTHDF2 and IGF2BP1 cooperatively recognize m6A-modified AFAP1-AS1 and enhance its stability, thereby mediating the regulatory function of METTL3 on AFAP1-AS1.

AFAP1-AS1 mediates the regulatory role of METTL3 on malignant phenotype and immune escape in PC cells

Based on the regulatory role of METTL3 on the m6A modification of AFAP1-AS1, we designed functional rescue experiments to verify whether AFAP1-AS1 is a downstream effector molecule for METTL3. We concurrently overexpressed AFAP1-AS1 in PANC-1 cells knocking down METTL3 (Fig. 4A). Functional analysis showed that knockdown of METTL3 alone significantly inhibited the clone formation (Fig. 4B), proliferation (Fig. 4C), invasion and migration ability (Fig. 4D) of the cells and reversed the EMT phenotype (elevating E-cadherin and decreasing N-cadherin) (Fig. 4E). However, overexpression of AFAP1-AS1 along with METTL3 knockdown was able to significantly reverse the above inhibitory effects induced by METTL3 knockdown (Fig. 4B–E). In terms of tumor immunity, knockdown of METTL3 alone enhanced the sensitivity of cancer cells to T cell killing (Fig. 4F), promoted T cell proliferation in the co-culture system (Fig. 4G), increased the secretion of cytotoxicity factors (IFN-γ, TNF-α, Perforin, Granzyme B) (Fig. 4H), and decreased PD-L1 in T cells (Fig. 4I). Similarly, overexpression of AFAP1-AS1 significantly reversed these anti-tumor immune-enhancing effects induced by METTL3 knockdown (Fig. 4F–I). The results of these experiments strongly demonstrate that AFAP1-AS1 mediates the regulatory role of METTL3 on malignant phenotype and immune escape in PC cells.

Fig. 4.

METTL3 promotes PC cell proliferation, invasive migration and immune escape through mediating AFAP1-AS1 m6A modification. Overexpression of AFAP1-AS1 was performed while knocking down METTL3 to construct a co-infected PANC-1 cell line. A Western blot or RT-qPCR to detect METTL3 and AFAP1-AS1 expression (n = 3). B Colony formation assay detected the number of cloned cells (n = 3). C EdU assay detected cell proliferation (n = 3). D Transwell assay detected cell invasion and migration ability (n = 3). E Western blot detected protein expression of cellular E-cadherin and N-cadherin (n = 3). F Crystal violet staining assessed the number of cancer cells surviving in the co-culture of T cells and cancer cells (n = 3). G Flow cytometry detected the dilution of CFSE fluorescence in CD8⁺ T cells and assessed the proliferation of T cells (n = 3). H ELISA detected the levels of IFN-γ, TNF-α, Perforin, and Granzyme B in the co-culture system (n = 3). I Western blot detected protein expression of PD-L1 in T cells (n = 3). Data are expressed as mean ± SD. *P < 0.05

AFAP1-AS1 acts by enhancing EGR2 mRNA stability

To explore the downstream molecular mechanisms by which AFAP1-AS1 exerts its function, we used bioinformatics tools to predict its potential RNA-binding targets, and found that EGR2 is a potential intercalating molecule (Fig. 5A). Western blot and RT-qPCR results showed that both protein and mRNA levels of EGR2 were significantly elevated in PC tissues (Fig. 5B, C). EGR2 protein expression was also higher in PC cell lines than in normal HPNE cells (Fig. 5D). To verify the regulatory role of AFAP1-AS1 on EGR2, we performed RNA stability experiments. Actinomycin D treatment showed that knockdown of AFAP1-AS1 significantly decreased the stability of EGR2 mRNA, whereas overexpression of AFAP1-AS1 enhanced its stability (Fig. 5E). RNA pull-down experiments further confirmed the existence of direct binding between AFAP1-AS1 RNA and EGR2 mRNA (Fig. 5F). However, CHX Chase Assay showed that changes in AFAP1-AS1 expression did not affect the stability of EGR2 protein (Fig. 5G). Regarding clinical data, GEPIA database analysis confirmed that EGR2 was abnormally highly expressed in PAAD (Fig. 5H) and its expression level increased with tumor stage (Fig. 5J), but database analysis showed that EGR2 expression did not significantly correlate with the overall survival prognosis of PAAD patients (Fig. 5I). Taken together, these data suggest that AFAP1-AS1 upregulates EGR2 expression in PC by directly binding to EGR2 mRNA and enhancing its stability.

Fig. 5.

AFAP1-AS1 enhances EGR2 mRNA stability. A Bioinformatics website ENCORI (http://starbase.sysu.edu.cn/) detected the potential binding sites of AFAP1-AS1 and EGR2. B Western blot detected protein expression of EGR2 in PC tissues and adjacent normal tissues (n = 56 pairs). C RT-qPCR detected EGR2 expression in three PC cell lines and normal human pancreatic cell line HPNE (n = 3). D Western blot detected EGR2 expression in three PC cell lines and normal human pancreatic cell line HPNE (n = 3). E Actinomycin D assay assessed the effect of knockdown or overexpression of AFAP1-AS1 on EGR2 mRNA stability (n = 3). F RNA pull-down assay detected the direct binding relationship between AFAP1-AS1 and EGR2 (n = 3). G CHX chase assay assessed the effect of AFAP1-AS1 on EGR2 protein stability (n = 3). H GEPIA database analysis of EGR2 expression in PAAD. I GEPIA database analysis of the correlation between EGR2 and survival prognosis in PAAD patients. J GEPIA database analysis of the correlation between EGR2 expression and tumor stage (I-IV). Data are expressed as mean ± SD. *P < 0.05

AFAP1-AS1 does not affect EGR2 post-translational modifications

Given that AFAP1-AS1 did not affect EGR2 protein stability, we further investigated whether AFAP1-AS1 might regulate EGR2 function through modulating its post-translational modifications, specifically ubiquitination and phosphorylation. Immunoprecipitation experiments were performed using anti-EGR2 antibody in PANC-1 cells with stable knockdown or overexpression of AFAP1-AS1. Western blot analysis of the immunoprecipitated EGR2 protein with anti-ubiquitin (FK2) antibody revealed that knockdown or overexpression of AFAP1-AS1 significantly altered the ubiquitination levels of EGR2 (Supplementary Figure S2A, B). Similarly, detection with anti-phospho-Serine/Threonine antibody showed that AFAP1-AS1 manipulation affect the phosphorylation status of EGR2 protein (Supplementary Figure S2C, D). These results confirm that AFAP1-AS1 regulates EGR2 primarily at the post-transcriptional level by enhancing mRNA stability, rather than through modulation of protein post-translational modifications.

Knockdown of EGR2 inhibits malignant progression and immune escape in PC cells

Given that AFAP1-AS1 can regulate EGR2 expression and that EGR2 is highly expressed in PC, we next investigated the function of EGR2 itself. We constructed cell lines that stably knocked down EGR2 in PANC-1 cells (Fig. 6A). Functional experiments showed that knockdown of EGR2 significantly inhibited the clone formation ability (Fig. 6B) and cell proliferation (Fig. 6C) of PANC-1 cells. Meanwhile, Transwell assays demonstrated that knockdown of EGR2 also significantly impaired cell invasion and migration (Fig. 6D). EGR2 knockdown resulted in up-regulation of E-cadherin expression and down-regulation of N-cadherin expression, suggesting that it may be involved in regulating the EMT process (Fig. 6E). In terms of tumor immunity, cancer cells with knockdown of EGR2 survived in reduced numbers when co-cultured with T cells, indicating that they were more sensitive to T cell-mediated killing (Fig. 6F). Accordingly, EGR2 knockdown promoted the proliferation of CD8⁺ T cells in the co-culture system (Fig. 6G) and significantly increased IFN-γ, TNF-α, Perforin and Granzyme B in the supernatant (Fig. 6H). In addition, knockdown of EGR2 also resulted in downregulation of PD-L1 protein expression in T cells (Fig. 6I). These results clearly indicate that EGR2 plays a critical role in promoting proliferation, invasive migration, and immune escape in PC cells.

Fig. 6.

Knockdown of EGR2 inhibits PC cell proliferation, invasion, migration and immune escape. PANC-1 cell line with stable knockdown of EGR2 was constructed. A Western blot detected the knockdown efficiency of EGR2 (n = 3). B Colony formation assay detected the number of cloned cells (n = 3). C EdU assay detected cell proliferation (n = 3). D Transwell assay detected cell invasion and migration ability (n = 3). E Western blot detected protein expression of cellular E-cadherin and N-cadherin (n = 3). F Crystal violet staining assessed the number of cancer cells surviving in the co-culture of T cells and cancer cells (n = 3). G Flow cytometry detected the dilution of CFSE fluorescence in CD8⁺ T cells and assessed the proliferation of T cells (n = 3). H ELISA detected the levels of IFN-γ, TNF-α, Perforin, and Granzyme B in the co-culture system (n = 3). I Western blot detected protein expression of PD-L1 in T cells (n = 3). Data are expressed as mean ± SD. *P < 0.05

EGR2 directly binds to the promoters of PD-L1 and Snail

To elucidate the direct transcriptional regulatory mechanism by which EGR2 promotes immune escape and EMT, we performed ChIP-qPCR assays to examine EGR2 binding to the promoter regions of its potential target genes. Bioinformatics analysis predicted multiple EGR2 binding motifs in the promoter regions of PD-L1 (CD274) and the EMT-related transcription factors Snail and Slug. ChIP-qPCR results demonstrated that anti-EGR2 antibody significantly enriched the PD-L1 promoter region containing the predicted EGR2 binding sites, with approximately 8-fold enrichment compared to IgG control (Supplementary Figure S3A). Similarly, substantial enrichment was observed at the Snail promoter region (approximately 6-fold enrichment) (Supplementary Figure S3B). However, only modest enrichment was detected at the Slug promoter region (approximately 2-fold enrichment) (Supplementary Figure S3C). Furthermore, knockdown of EGR2 significantly reduced the mRNA levels of PD-L1 and Snail, while the reduction in Slug mRNA was less pronounced (Supplementary Figure S3D). These results provide direct evidence that EGR2 functions as a transcription factor that directly activates the transcription of PD-L1 and Snail, thereby promoting immune escape and EMT in PC cells.

AFAP1-AS1 mediates its pro-oncogenic function by regulating EGR2 expression

To determine whether EGR2 is a downstream mediator for AFAP1-AS1 to exert its function, we performed another set of functional rescue experiments. We concurrently overexpressed EGR2 in PANC-1 cells that stably knocked down AFAP1-AS1 (Fig. 7A). The results showed that inhibition of cell clone formation (Fig. 7B), proliferation (Fig. 7C), invasion and migration ability (Fig. 7D), and EMT phenotype (Fig. 7E) caused by knockdown of AFAP1-AS1 alone were significantly restored by overexpression of EGR2. In terms of immune function, increased sensitivity of cancer cells to T-cell killing (Fig. 7F), enhanced T-cell proliferation (Fig. 7G), increased secretion of cytotoxic factors (Fig. 7H), and decreased expression of PD-L1 (Fig. 7I) induced by knockdown of AFAP1-AS1 alone were also significantly reversed by overexpression of EGR2. These results strongly demonstrated that EGR2 is a key downstream effector molecule in the AFAP1-AS1 signaling pathway, and that AFAP1-AS1 promotes the proliferation, invasive migration, and immune escape of PC cells mainly by upregulating EGR2.

Fig. 7.

AFAP1-AS1 mediates PC cell proliferation, invasion, migration and immune escape by promoting EGR2 expression. Overexpression of EGR2 was performed while knocking down AFAP1-AS1 to construct a co-infected PANC-1 cell line. A Western blot or RT-qPCR detected the expression of EGR2 and AFAP1-AS1 (n = 3). B Colony formation assay detected the number of cloned cells (n = 3). C EdU assay detected cell proliferation (n = 3). D Transwell assay detected cell invasion and migration ability (n = 3). E Western blot detected protein expression of cellular E-cadherin and N-cadherin (n = 3). F Crystal violet staining assessed the number of cancer cells surviving in the co-culture of T cells and cancer cells (n = 3). G Flow cytometry detected the dilution of CFSE fluorescence in CD8⁺ T cells and assessed the proliferation of T cells (n = 3). H ELISA detected the levels of IFN-γ, TNF-α, Perforin, and Granzyme B in the co-culture system (n = 3). I Western blot detected protein expression of PD-L1 in T cells (n = 3). Data are expressed as mean ± SD. *P < 0.05

Targeting METTL3 or EGR2 inhibits PC growth in vivo and enhances anti-tumor immunity

Finally, to verify the in vitro findings, we inoculated the murine-derived PC cell line Panc02, which is stably knocked down METTL3 or EGR2, into immunocompetent mice. The results showed that the tumor volume was significantly reduced in the sh-METTL3 group (Fig. 8A) and the weight was significantly reduced (Fig. 8B), while overexpression of EGR2 was able to partially reverse the tumor growth inhibitory effect caused by METTL3 knockdown (Fig. 8A–C). Western blot analysis confirmed that both METTL3 and EGR2 protein expression were decreased in tumor tissues in the sh-METTL3 group, along with a significant reduction in the PD-L1 expression level; whereas in the sh-METTL3 + oe-EGR2 group, both EGR2 and PD-L1 protein expression were restored to the control level (Fig. 8D). These in vivo experimental results fully confirmed the molecular mechanism by which METTL3 promotes PC immune escape via EGR2, suggesting that targeting the METTL3-EGR2 axis may be a potential therapeutic strategy to overcome PC immune tolerance.

Fig. 8.

METTL3 enhances tumor immune escape by promoting EGR2 expression. The murine-derived PC cell line Panc02, which stably knocked down METTL3 and EGR2, was inoculated into immunocompetent mice. A Tumor volume (n = 6/group). B Tumor weight (n = 6/group). C Representative image of the tumors. D Western blot detected protein expression of METTL3, EGR2, and PD-L1 (n = 3). Data are expressed as mean ± SD. *P < 0.05

METTL3 knockdown remodels the tumor immune microenvironment

To comprehensively characterize the impact of the METTL3/EGR2 axis on the tumor immune microenvironment, we performed flow cytometry analysis of tumor-infiltrating immune cell populations in the subcutaneous tumor models. Analysis of regulatory T cells (Tregs, defined as CD4⁺CD25⁺Foxp3⁺ cells) revealed that METTL3 knockdown significantly decreased the proportion of Tregs in tumor tissues compared to the control group, while EGR2 overexpression partially restored Treg infiltration (Supplementary Figure S4A, B).

We further analyzed tumor-associated macrophages (TAMs). The results showed that METTL3 knockdown led to a significant decrease in the proportion of M2-polarized macrophages (CD11b⁺F4/80⁺CD206⁺), which are known to promote immunosuppression and tumor progression (Supplementary Figure S4C, D). Conversely, the proportion of M1-polarized macrophages (CD11b⁺F4/80⁺iNOS⁺), which exhibit anti-tumor activity, was significantly increased in the sh-METTL3 group (Supplementary Figure S4E, F). EGR2 overexpression in the context of METTL3 knockdown partially reversed these changes in macrophage polarization, increasing M2 macrophages and decreasing M1 macrophages (Supplementary Figure S4C-F).

Additionally, we confirmed that CD8⁺ T cell infiltration was significantly enhanced in tumors with METTL3 knockdown, and this increase was partially attenuated by EGR2 overexpression (Supplementary Figure S4G, H). These comprehensive immune profiling results demonstrate that the METTL3/EGR2 axis not only affects CD8⁺ T cell function but also broadly remodels the tumor immune microenvironment by regulating the recruitment and polarization of immunosuppressive cell populations, including Tregs and M2 macrophages.

Discussion

The present study reveals that METTL3 enhances the stability and function of AFAP1-AS1 by mediating the m6A modification of AFAP1-AS1 in the cytoplasm, and then promotes PC cell proliferation, invasion/migration, and immune escape by binding EGR2 mRNA. This “METTL3-AFAP1-AS1-EGR2” axis not only constitutes a molecular basis for the malignant phenotype of PC, but also provides new ideas for immunotherapy targeting m6A modification and lncRNA regulation.

The expression and m6A modification levels of AFAP1-AS1 in PC tissues and cell lines were significantly higher than those of normal controls and showed a trend of increasing with tumor stage progression, and this expression pattern was closely associated with poor patient prognosis. Meanwhile, METTL3 also showed high expression in PC, suggesting that m6A modification may be a key mechanism for the regulation of AFAP1-AS1 expression. Nucleoplasmic separation experiments and FISH experiments confirmed that METTL3 co-localized with AFAP1-AS1 in the cytoplasm, challenging the conventional view that m6A modification occurs mainly in the nucleus [27]. This co-localization may reflect that METTL3 has under-recognized cytoplasmic functions, such as participating in the assembly of RNA degradosome or protecting modified RNA from degradation.

In terms of the specific mechanism by which m6A modification mediates the stability of AFAP1-AS1, our study provides direct evidence that YTHDF2 and IGF2BP1 are the key m6A readers responsible for recognizing and stabilizing m6A-modified AFAP1-AS1. The m6A site on AFAP1-AS1 recruits these specific recognition proteins, which form RNA–protein complexes to resist degradation by cytoplasmic exonucleases [28, 29]. The finding that both YTHDF2 and IGF2BP1 are involved in stabilizing AFAP1-AS1 is particularly intriguing, as YTHDF2 has traditionally been associated with promoting mRNA decay [30]. However, recent studies have revealed context-dependent roles of YTHDF2 in different cellular environments and RNA targets, where it can also contribute to RNA stability under certain conditions [31]. The cooperative action of YTHDF2 and IGF2BP1 in stabilizing AFAP1-AS1 suggests a complex regulatory network where multiple m6A readers work synergistically to determine RNA fate. This is consistent with emerging evidence that m6A readers often function in a coordinated manner rather than independently [32]. m6A modifications may also expose or hide key protein-binding sites by regulating lncRNA secondary structure or affect the subcellular localization of AFAP1-AS1 and its interaction with nucleoplasmic transport proteins [33–35]. Luciferase reporter assays further confirmed that METTL3-mediated m6A modification affects the stability of the AFAP1-AS1 fragment, reinforcing this regulatory relationship. Functional studies revealed that AFAP1-AS1 promoted the proliferation and invasive migration of PC cells while significantly enhancing tumor immune escape. In T-cell co-culture assays, AFAP1-AS1 overexpression significantly inhibited the proliferative activity and secretion of cytotoxic factors (IFN-γ, TNF-α, Perforin, and Granzyme B) in T-cells, while up-regulating PD-L1 expression. This immunosuppressive effect implies that AFAP1-AS1 may remodel the tumor immune microenvironment through multiple mechanisms: directly affecting the expression profiles of immunoregulatory molecules on the surface of cancer cells; being secreted into the tumor microenvironment and then uptaken by immune cells, which directly modulates the immune cell function; or indirectly enhancing the immune escape ability of tumors by inducing EMT. Functional rescue experiments further confirmed that AFAP1-AS1 is a downstream effector molecule of METTL3 in promoting malignant progression and immune escape in PC, establishing a direct link between m6A modification and tumor immune escape. These findings are consistent with recent reports highlighting the critical role of m6A modification in regulating tumor-immune interactions [36].

In exploring the downstream effector molecules of AFAP1-AS1, this study confirmed that the transcription factor EGR2 is a key target of AFAP1-AS1. EGR2was highly expressed in PC tissues and cells, and its expression pattern was highly consistent with that of AFAP1-AS1. The RNA stability assay and RNA pull-down assay confirmed that AFAP1-AS1 enhances its stability by directly binding to the EGR2 mRNA. Notably, although AFAP1-AS1 affected EGR2 mRNA stability, it had no significant effect on its protein stability, suggesting that AFAP1-AS1 regulates EGR2 expression mainly at the post-transcriptional level rather than the post-translational level. The specific mechanism by which AFAP1-AS1 mediates EGR2 mRNA stabilization may involve multiple mechanisms. As a lncRNA, AFAP1-AS1 may be involved in the protection of target mRNAs by forming RNA-RNA duplexes or by recruiting RNA-binding proteins. HuR is known to bind to the AU-enriched 3′UTR region and inhibit mRNA degradation [37–39]. AFAP1-AS1 may act as a “molecular scaffold” that masks miRNA binding sites (e.g., miR-29a/b) on the 3′UTR of EGR2 on the one hand, and on the other hand, recruits stabilizing factors, such as HuR or IGF2BP1, to synergistically inhibit the CCR4-NOT complex-mediated deadenylation process. This hypothesis is based on the fact that both HuR and IGF2BP1 have been reported to positively correlate with mRNA stability [40]. The function of EGR2 as a transcription factor in PC replicates the proliferation-promoting, invasive-migratory, and immune-escape phenotypes of AFAP1-AS1, and the functional rescue experiments further confirmed that EGR2 is a key effector molecule downstream of AFAP1-AS1. Importantly, our ChIP-qPCR experiments provided direct evidence that EGR2 binds to the promoter regions of PD-L1 and Snail, establishing EGR2 as a direct transcriptional activator of these key immune checkpoint and EMT-related genes. EGR2 may promote the malignant phenotype of PC through multiple mechanisms. The direct binding of EGR2 to the PD-L1 promoter explains the molecular mechanism by which this axis promotes immune evasion, while its binding to the Snail promoter accounts for the EMT-promoting effects observed in our functional studies. EGR2 activates downstream effectors of the MAPK/ERK pathway, such as c-Myc or Cyclin D1, to further enhance cell cycle progression and proliferation. EGR2 establishes a more complex immunosuppressive network by regulating the secretion of tumor microenvironmental factors (e.g., TGF-β, IL-10). These hypotheses are based on the fact that EGR family members have been reported to regulate immune checkpoint molecules and EMT-related genes. Recent studies have further highlighted the importance of transcription factor networks in shaping the tumor immune landscape, supporting the central role of EGR2 in PC immune escape [41].

Mouse models further confirmed the significant inhibitory effect of targeting the METTL3-EGR2 axis on PC growth and immune escape. Knockdown of METTL3 significantly reduced tumor volume and decreased EGR2 and PD-L1 expression. Comprehensive immune profiling revealed that METTL3 knockdown not only enhanced CD8⁺ T cell infiltration but also significantly reduced the proportion of immunosuppressive cell populations, including Tregs and M2 macrophages, while increasing M1 macrophage infiltration. This broad remodeling of the tumor immune microenvironment suggests that the METTL3/EGR2 axis functions as a master regulator of immunosuppression in PC. In contrast, EGR2 overexpression partially reversed the antitumor effect of METTL3 knockdown, including the restoration of immunosuppressive cell infiltration, a fact that not only validates the functional relevance of this molecular axis in vivo, but also implies that METTL3 may promote PC progression through other pathways in addition to the AFAP1-AS1-EGR2 axis.

METTL3 overexpression may be associated with the molecular environment specific to PC. It is well known that KRAS mutations in PC can enhance METTL3 [42] transcription via the MAPK pathway or upregulate m6A enzyme expression via HIF-1α in a hypoxic microenvironment [43]. Thus, METTL3 overexpression may not be an isolated event, but rather a result of a concerted effort with tumor-driven mutations and microenvironmental responses. This upstream regulation provides new perspectives for understanding aberrant m6A modifications in PC. From the perspective of epitranscriptome regulation, the METTL3-AFAP1-AS1-EGR2 axis connects RNA modification, noncoding RNA regulation, and transcription factor networks, and represents a novel paradigm of epitranscriptome regulation. This multilayered regulatory network exemplifies the intricate interplay between epigenetic and epitranscriptomic mechanisms in cancer progression and immune evasion [44]. In this model, m6A modifications not only affect the fate of coding RNAs, but also regulate the function of non-coding RNAs, thereby expanding the scope of their regulatory network. Particularly in the context of tumor immunity, the discovery of this molecular axis provides a new perspective for understanding how tumor cells remodel the immune microenvironment through intrinsic epigenetic mechanisms, explaining that acquired resistance to immune checkpoint inhibitors may be related to tumor cells responding to immune selective pressures by upregulating METTL3 expression. The comprehensive remodeling of the tumor immune microenvironment by the METTL3/EGR2 axis, affecting not only cytotoxic T cells but also Tregs and macrophage polarization, underscores the therapeutic potential of targeting this pathway to overcome the immunosuppressive features of PC [45].

Based on the findings of this study, several targeted intervention strategies can be envisioned: development of METTL3-specific small molecule inhibitors; design of antisense oligonucleotides or siRNAs targeting AFAP1-AS1; development of small molecules interfering with the binding of AFAP1-AS1 to EGR2 mRNA; and development of protein degradation chimeric technologies targeting EGR2. Of more practical value are combination strategies, whereby combining METTL3 inhibitors or AFAP1-AS1-targeted drugs with PD-1/PD-L1 inhibitors may overcome PC’s inherent resistance to immune checkpoint inhibitors. The identification of YTHDF2 and IGF2BP1 as key mediators of AFAP1-AS1 stability also opens new avenues for therapeutic intervention by targeting these m6A readers [46]. Despite the important findings provided by the present study, several limitations remain. Since there is no homologue of AFAP1-AS1 in mice, in vivo studies were unable to directly assess the lncRNA component of this signaling axis. Although our xenograft models using human cancer cells with AFAP1-AS1 manipulation and the clinical sample analyses provide compelling evidence for the role of AFAP1-AS1 in vivo, future studies utilizing patient-derived xenograft (PDX) models or humanized mouse models would be valuable to further validate these findings in a more physiologically relevant context. The study relied mainly on a limited number of cell line models, and validation of its generalizability in more patient-derived primary cell and organoid models is yet to be carried out. The specific type of m6A reader and the AFAP1-AS1 stabilizing complexes have not been clarified, affecting the in-depth interpretation of the fine mechanisms. Studies of the tumor immune microenvironment have focused on CD8⁺ T cell function, and have not yet fully explored their effects on other immune cells.

In summary, this study elucidated the regulatory role of the METTL3-AFAP1-AS1-EGR2 axis in PC, demonstrating that METTL3-mediated m6A modification of AFAP1-AS1 is recognized by YTHDF2 and IGF2BP1 to enhance its stability, which in turn stabilizes EGR2 mRNA. EGR2 then directly activates transcription of PD-L1 and Snail to promote immune escape and EMT. This not only deepens the understanding of m6A modification-lncRNA-mRNA regulatory network in the malignant progression of PC and immune escape, but also provides a new strategy for remodeling the tumor immune microenvironment by interfering with the stability of the m6A enzymes or lncRNAs that has important theoretical significance and clinical translational potential.

Author contributions

DaoJun Zhu: Conceptualization, Methodology, Investigation, Soft-ware, Validation, Visualization, Writing—original draft. RongXi Shen: Methodology, Investigation, Validation, Visualization Min Li: Formal analysis, Data curation, Supervision, Writing—review & editing, Project administration.

Funding

Not applicable.

Data availability

Data is available from the corresponding author on request.

Declarations

Ethics approval and consent to participate

All procedures performed in this study involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. All subjects were approved by Affiliated Jinling Hospital, Medical School of Nanjing University (No.2022JL3881). And the animal experiments were complied with the ARRIVE guidelines and performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The experiments were approved by the Institutional Animal Care and Use Committee of Affiliated Jinling Hospital, Medical School of Nanjing University (No.2022JL7194). Written informed consent was obtained from each subject.

Competing interests

The authors declare no competing interests.