Expression of YTHDF1 and YTHDF2 in lung squamous cell carcinoma and their correlation with PD-L1

Introduction

Global cancer statistics for 2022 (1) showed that lung cancer accounted for approximately 2.5 million new cases and 1.8 million deaths annually. Lung cancer was the leading cause of cancer and mortality in both men and women globally, as well as in China, with approximately 1.06 million new cases and 730,000 deaths reported in China (2).

Lung cancer is classified into non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC). NSCLC is further divided into two major subtypes: lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC). LUAD and LUSC account for approximately 85% of all lung cancer cases (3). LUSC accounts for 20–30% of all lung cancers, and because of the lack of early symptoms, patients often present with advanced disease at diagnosis, leading to a poor prognosis (4). Unlike LUAD, LUSC has a low mutation rate in epidermal growth factor receptor (EGFR). Consequently, EGFR tyrosine kinase inhibitors, which are commonly used in NSCLC treatment, are ineffective against LUSC (5). Therefore, the lack of effective targeted therapies for LUSC highlights the critical need to identify valuable therapeutic targets for its diagnosis and treatment.

N6-methyladenosine (m6A) is one of the most prevalent internal modifications of eukaryotic messenger RNA (mRNA). Dysregulation of m6A modification in cancer may contribute to tumorigenesis and progression through multiple mechanisms. such as affecting the expression of tumor suppressor genes and oncogenes, thereby influencing tumor cell proliferation and apoptosis. Additionally, m6A modification can mediate immune responses in the tumor microenvironment and promote immune evasion by tumor cells (6). YTH N6-methyladenosine RNA-binding protein 1 (YTHDF1), a critical member of the YTH family, recognizes m6A modification sites on mRNA and promotes mRNA-ribosome binding and enhances translation efficiency (7). Liu et al. (8) reported that YTHDF1 is highly expressed in ovarian cancer and regulates ovarian cancer cell proliferation, migration, and invasion. Moreover, Li et al. (9) confirmed that YTHDF1 was highly expressed in prostate cancer tissues and its high expression was associated with a poor prognosis in patients with prostate cancer; moreover, inhibition of YTHDF1 expression reduced prostate cancer cell proliferation, migration, and invasion. Xia et al. (10) have confirmed that YTHDF1 accelerates viral RNA uncoating and mediates RNA decay by recruiting an RNA degradation complex, thereby downregulating Epstein-Barr virus (EBV) gene expression.

YTH N6-methyladenosine RNA-binding protein 2 (YTHDF2), another important member of the YTH family and the first discovered m6A recognition protein, recognizes and binds to m6A-modified mRNA to mediate their degradation and localization. m6A-dependent mRNA degradation affects cellular processes and regulates cell proliferation and migration in tumors (11). Yang et al. (12) found that YTHDF2 accelerates gastric cancer progression and enhances resistance to radiotherapy and chemotherapy via a mechanism related to CyclinD1 expression. Jiang et al. (13) demonstrated that YTHDF2 was upregulated in brain gliomas, with high expression correlating with high tumor grade and poor prognosis. Gene set enrichment analysis (GSEA) showed that YTHDF2 expression was highly correlated with immune response and oncogenic signaling pathways.

Despite growing evidence that YTH family proteins regulate immune responses via m6A modification, critical gaps remain in LUSC. First, how YTHDF1 and YTHDF2 specifically interact with programmed death-ligand 1 (PD-L1; a key immune checkpoint molecule) to modulate its expression is unclear—whether through direct binding to PD-L1 mRNA or indirect regulation of upstream pathways. Second, the link between YTHDF1/YTHDF2 expression and immune cell infiltration in the LUSC microenvironment is poorly understood, especially how their dysregulation alters the balance of pro-tumor vs. anti-tumor immune cells. Lastly, no studies have systematically explored whether YTHDF1/YTHDF2 mediates immune evasion by coordinating PD-L1 upregulation and immune cell reprogramming in LUSC.

Programmed death protein 1 (PD-1) and PD-L1 are key components of immune checkpoints and are crucial in tumor immune escape. In recent years, PD-1/PD-L1-targeted immunotherapy has achieved remarkable clinical results in lung cancers, particularly in NSCLC (14). Immunotherapy targeting the PD-1/PD-L1 axis is now a common approach for treating NSCLC. PD-1/PD-L1 monoclonal antibodies are now the first-line therapeutic option for patients with advanced NSCLC (15). Therefore, PD-1/PD-L1 expression in patients with NSCLC may be closely correlated with the efficacy of PD-1/PD-L1 inhibitors.

Using The Cancer Genome Atlas (TCGA) dataset, several studies have shown that YTHDF1 and YTHDF2 are upregulated in lung cancer, potentially contributing to tumorigenesis and progression. However, YTHDF1 and YTHDF2 expression in LUSC and their clinical significance remain underreported in domestic studies. Moreover, the relationships between YTHDF1/YTHDF2, PD-L1 expression, and immune cell infiltration in the LUSC microenvironment are poorly understood, potentially affecting the efficacy of PD-1/PD-L1 inhibitors in LUSC. Therefore, we analyzed YTHDF1 and YTHDF2 expression in LUSC using bioinformatics and immunohistochemistry to elucidate their roles in LUSC development and its underlying molecular mechanisms. Additionally, we explored the relationship between YTHDF1/YTHDF2 and PD-L1 expression to evaluate their potential as therapeutic targets and to provide a basis for combining YTHDF1 or YTHDF2 inhibitors with PD-1/PD-L1 immune checkpoint inhibitors. We present this article in accordance with the MDAR reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1-0312/rc).

Methods

General data

Transcriptome sequencing data and related clinical data for LUSC were downloaded in May 2024 from the TCGA database (https://portal.gdc.cancer.gov/). R4.3.3 was used to convert the downloaded RNA sequencing (RNA-seq) data in Fragments Per Kilobase of transcript per Million mapped reads (FPKM) format into transcripts per million (TPM) format and could be used for subsequent analysis after log₂ conversion processing, with the conversion formula: log₂(TPM+1). Survival data of patients with LUSC were downloaded from the Xena database, and data from 547 patients with LUSC were included in the analysis, including RNA-seq datasets, matched clinical information, and survival data from 496 LUSC biopsy tissues and 51 adjacent tissues.

In addition, microarrays (Shanghai Outdo Biotech Co., Ltd., Shanghai, China; product code, HLugS120CS01), containing PD-L1 expression information, were used for immunohistochemistry analysis. The microarray included 60 cases of LUSC, with one tumor tissue core and one matched adjacent normal tissue core for each case. The tissues were selected based on the following inclusion criteria: (I) pathologically confirmed primary LUSC [per World Health Organization (WHO) 2021 classification]; (II) no neoadjuvant radiotherapy, chemotherapy, or immunotherapy before surgical resection; (III) complete clinicopathological records [including age, sex, tumor (T)/node (N)/metastasis (M) stage, and PD-L1 Tumor Proportion Score (TPS) status]; (IV) no other concurrent malignancies or autoimmune diseases. The exclusion criteria were as follows: (I) metastatic lung cancer; (II) tissue samples with insufficient tumor content (<50%); and (III) poor tissue quality (e.g., severe necrosis or artifacts). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of the First Affiliated Hospital of Baotou Medical College (approval No. K022-01) and informed consent was obtained from all individual participants. The study was also approved by the Ethics Committee of the Shanghai Outdo Biotech Company (approval No. HLugS120CS01).

Main antibodies and reagents

YTHDF1 antibodies were purchased from Hua’an Biotechnology Co., Ltd. (Hangzhou, China). YTHDF2 antibodies were purchased from ABclonal Biotechnology Co., Ltd. (Wuhan, China), and Ethylene Glycol-bis (2-aminoethyl Ether)-N,N,N',N'-tetraacetic Acid (EGTA) antigen retrieval buffer (pH 9.0), antibody diluent, polylysine, phosphate buffered saline (PBS), super-sensitive 3,3'-diaminobenzidine (DAB) substrate, instant immunohistochemistry MaxVision^TM kit, and neutral balsam were purchased from Fuzhou Maxim Biotechnology Co., Ltd. (Fuzhou, China). Mayer hematoxylin was purchased from Shenzhen Zike Biological Technology Co., Ltd. (Shenzhen, China).

Bioinformatics analysis

Differential expression analysis of target genes in LUSC samples was performed using the R software limma package (15), and the results were visualized by drawing box plots using the R software ggplot2 package and the R software ggpubr package. According to the clinical information, the patients with LUSC were grouped according to age, sex, tumor infiltration depth (T stage), lymph node metastasis (N stage), distant metastasis (M stage), and clinical stage. Differences in the expressions of target genes in different subgroups were analyzed using the R software limma package, and the results of the gene difference analysis were presented by drawing box plots with R software. The samples were divided into high and low-expression groups according to the target gene expression. The cutoff for YTHDF1 expression was defined by the median expression level. Kaplan-Meier survival curves were drawn, and the overall survival (OS) and disease-specific survival (DSS) were analyzed between different subgroups using the Log-rank test. The Cox proportional hazards regression model was used to analyze the risk factors affecting prognosis, in which the clinical indicators with P<0.1 from the univariate analysis were further included in the multivariate analysis. To explore the biological processes and pathways in which the target genes might be involved, LUSC samples were divided into high and low expression groups based on the median gene expression level of the target gene. Differential expression analysis between the two groups was performed using the DESeq2 package in R, and differentially expressed genes related to the target genes were identified. GSEA was subsequently conducted to investigate the biological functions and signaling pathways associated with these differentially expressed genes. GSEA was performed using the GSEA software (version 4.3.0), and the reference gene set database was the MSigDB Hallmark gene set (h.all.v7.5.1. symbols.gmt) and the KEGG pathway gene set (c2.cp.kegg.v7.5.1. symbols.gmt). The number of permutations was set to 1,000, and the normalization method was “median centering”. The enrichment results were filtered with |Normalized Enrichment Score (NES)| >1, false discovery rate (FDR) <0.25, and P<0.05 as the screening criteria. The CIBERSORT algorithm was used to estimate the correlation between the expression of target genes in LUSC and 22 types of immune cells. CIBERSORT was run with the default parameters, the signature matrix was LM22, and the permutation number was 1,000. The results were filtered with P<0.05 as the statistically significant threshold, and the relative proportion of immune cells was normalized to the sum of 1 for each sample.

Immunohistochemical staining

Immunohistochemistry was used to detect the expression of YTHDF1 and YTHDF2 proteins, and the procedures were performed strictly according to the manufacturer’s instructions. After dewaxing, hydration, antigen repair, rinsing, sealing, and other operations of the tissue microarray, the primary antibody (YTHDF1 antibody 1:1,000, YTHDF2 antibody 1:200) was left overnight, followed by dropwise addition of a secondary antibody (instant MaxVisionTM horseradish peroxidase polymer HRP-polymer anti-rabbit immunohistochemistry kit), rinsing with PBS, evenly dripping DAB color solution on the slides, re-staining with hematoxylin, dehydration, and sealing.

Slides evaluation

Formalin-fixed, paraffin-embedded (FFPE) LUSC tissues were subjected to immunohistochemistry (IHC) staining using a primary antibody against YTHDF1 (ab252346, Abcam, Shanghai, China). Two independent pathologists, who were blinded to the clinicopathological data of the patients, evaluated the IHC slides independently. The staining intensity was scored as 0 (negative), 1 (weak), 2 (moderate), and 3 (strong) for yellowish, yellow, and yellowish-brown, respectively. The percentages of positive cells, that is, 0, <25%, 25–49%, 50–74%, and 75–100%, were scored as 0, 1, 2, 3, and 4 points, respectively. The total score of immunohistochemistry was calculated by multiplying the staining intensity score by the percentage of positive cells score. For the definition of positive and negative expression, patients were stratified into YTHDF1 high-expression (≥7) and YTHDF1 low-expression groups [0–6] based on the total immunohistochemical score of the entire cohort (cutoff =6). Any discrepancies in scoring between the two pathologists were resolved by consensus discussion (16).

Quantitative analysis of immunohistochemistry

Immunohistochemical staining intensity was evaluated using Image Pro Plus 6.0 software. Four representative images were selected under a high-power field of view, and the accumulated integrated optical density (IOD) value of each image was measured. After the log conversion, the logIOD value was used to reflect the expression of target genes in the LUSC and adjacent tissues.

Cell line and cell culture

H1703 cells were obtained from the Cell Culture Center of the Chinese Academy of Sciences (Shanghai, China) and cultured in RPMI medium 1640 (31800; Solarbio, Beijing, China) supplemented with 10% fetal bovine serum (S9020; Solarbio) and 1% penicillin-streptomycin. H1703 was incubated at 37 ℃ in 5% CO₂ atmosphere.

Plasmid constructs and transfections

PcDNA3.1 vector was used to construct full-length YTHDF1-CDS, PD-L1-CDS, and PD-L1-mutant (mut) plasmids. All plasmids were purchased from GeneChem (Shanghai, China). Plasmid transfection was performed using Lipofectamine 3000 (L3000015, Thermo Scientific, Shanghai, China), and cells were collected for further experiments 24 h or 48 h after transfection. The sequences are listed in Appendix 1.

Rationale for dual-plasmid use: the PD-L1-wild-type (WT) plasmid was designed to express PD-L1 with intact m6A modification sites, mimicking endogenous PD-L1 in LUSC cells. The PD-L1-mut plasmid was created to abolish the m6A modification of PD-L1 mRNA, allowing us to test whether YTHDF1 regulates PD-L1 via an m6A-dependent mechanism, a key feature of YTHDF1 that functions as an m6A reader.

Mutation strategy: based on nine overlapping m6A sites identified by SRAMP and RMBase v3.0, we mutated the adenine (A) residue at each thymine (T) site using site-directed mutagenesis (primers listed in Appendix 1). This mutation destroys the m6A modification of PD-L1 mRNA, rendering it unable to be recognized and bound by YTHDF1, thus eliminating the m6A-dependent regulatory effect of YTHDF1 on PD-L1.

RNA extraction and quantitative real-time polymerase chain reaction (qRT-PCR)

TRIzol reagent (15596026CN, Invitrogen, Carlsbad, USA) was used for RNA extraction, and a NanoDrop spectrophotometer (NanoDrop 2000, ThermoScientific) was used to measure RNA concentration, followed by reverse transcription using a PrimeScript RT kit (6110A, Takara, Kusatsu, Japan). qRT-PCR was performed in triplicate using a TaqMan One-Step RT-qPCR Kit (T2210, Solarbio). GAPDH was used as the internal reference gene, and the relative expression of indicated transcripts was calculated according to the ΔΔCt method. The primers used are listed in Table S1.

Flow cytometry analysis

The cells were incubated for 48 h after transfection with plasmids. Recombinant human interferon-γ (INF-γ) (300-02-500UG, PeproTech, Suzhou, China) was added to a final concentration of 10 ng/mL during the last 24 h of incubation. Next, H1703 cells were incubated with Human TruStain FcX (cat422302, Biolegend, San Diego, USA) for 10 min at 4 ℃ and washed with 1× PBS containing 1% FBS. The cells were incubated with 5 µL of phycoerythrin (PE) anti-human PD-L1 (B7-H1, PD-L1) antibody (329706, Biolegend) or the corresponding isotype control (400314, Biolegend) at 4 ℃ for 30 min. A BD FACSCanto II was used for the flow cytometry, and the FlowJo software was used for data analysis.

Measurement of cell cytotoxicity

Cell cytotoxicity was assessed using a lactate dehydrogenase (LDH) colorimetric assay kit (ab102526; Abcam), based on a previously published study (16).

Western blot analysis

Cells were isolated using RIPA lysis buffer (P0013B, Beyotime, Shanghai, China) with PMSF (ST507, Beyotime) and extraction buffer (P0013M, Beyotime). Protein concentration was quantified using a bicinchoninic acid (BCA) assay kit (P0010, Beyotime), and 50 µg protein of each sample in the gels was transferred to polyvinylidene difluoride membranes (P0965, Beyotime). Membranes were blocked with tris-buffered saline with Tween 20 (TBST) containing 5% skimmed milk for 1 h and incubated overnight at 4 ℃ with various primary antibodies, including anti-YTHDF1 (1:1,000 dilution, ab252346, Abcam), PD-L1 (1:1,000 dilution, ab205921, Abcam), and GAPDH (1:1,000 dilution, ab9485, Abcam). Membranes were incubated with peroxidase-labeled anti-rabbit and anti-mouse secondary antibodies. The chemiluminescent substrate was applied using the SuperSignal West Pico Chemiluminescent Substrate, and the blots were analyzed using the ChemiDoc Touch Imaging System (Bio-Rad) and Image Lab software (Bio-Rad). Protein expression levels were normalized to those of GAPDH, which was used as an internal control.

T cell-mediated tumor cell-killing assay

T cell-mediated tumor cell killing assays were performed as previously described (17). Human peripheral blood mononuclear cells (PBMCs) were isolated from healthy donors (provided by SCHBIO, Shanghai, China) and activated with 100 ng/mL anti-CD3 antibody, 100 ng/mL anti-CD28 antibody, and 10 ng/mL interleukin (IL)-2 for 72 h at 37 ℃ with 5% CO₂. H1703 cells [transfected with YTHDF1 overexpression (YTHDF1-OE) and negative control (NC) and PD-L1-WT/mut plasmids] were seeded into 24-well plates (1×10⁵ cells/well), 24 h before co-culture. Activated PBMCs were added at a 4:1 (PBMC: H1703) ratio, and the co-culture was maintained for 72 h, after which PBMCs were removed by washing twice with PBS, and adherent tumor cells were stained with a 0.1% crystal violet solution.

Statistical analysis

R4.3.3 and SPSS27.0 were used for the statistical analysis of the data in this study. Student’s t-test was used to compare gene expression between the two groups. The Chi-squared test was used to analyze the relationship between YTHDF1 and YTHDF2 protein expression levels and the clinicopathological data of the patients with LUSC. The continuity correction test was used when the theoretical frequency was 1≤ T ≤5. Spearman’s correlation was used to analyze the correlation between YTHDF1 and YTHDF2 expression and PD-L1 expression (P<0.05) (*, P<0.05; **, P<0.01; ***, P<0.001).

Results

Bioinformatics results

Sample clinical information

Data from 547 patients with LUSC from the TCGA database were included, comprising 496 cancerous tissues and 51 adjacent tissues. The demographic and clinical characteristics of patients with LUSC, including sex, age, TNM stage, OS, and DSS, are summarized in Table 1.

mRNA expression of YTHDF1 and YTHDF2 in LUSC tissues

The mRNA expression levels of YTHDF1 and YTHDF2 were analyzed in 496 LUSC tissues and 51 adjacent tissues in the TCGA database. YTHDF1 expression was significantly higher in the LUSC tissues than in the adjacent tissues (P<0.001) (Figure 1A). The mRNA expression levels of YTHDF1 and YTHDF2 were analyzed in paired LUSC and adjacent tissues (50 pairs in total). YTHDF1 expression was similarly elevated in the LUSC and adjacent tissues (P<0.001) (Figure 1B), YTHDF2 expression did not differ significantly between the LUSC and adjacent tissues(P>0.05) (Figure 1C). YTHDF2 expression in LUSC tissues was higher than that in adjacent tissues (P<0.05) (Figure 1D).

Figure 1 The differentiated expression of YTHDF1 and YTHDF2 in tumor tissues and normal tissues. (A) YTHDF1 expression in non-paired lung cancer tissues and normal tissues according to the TCGA dataset. (B) YTHDF1 expression in paired lung cancer tissues and normal tissues in the TCGA database. (C) YTHDF2 expression in non-paired lung cancer tissues and normal tissues according to the TCGA dataset. (D) YTHDF2 expression in paired lung cancer tissues and normal tissues in the TCGA database. ns, not significant; *, P<0.05; ***, P<0.001. TCGA, The Cancer Genome Atlas.

Correlation between YTHDF1 and YTHDF2 expression in LUSC tissues with clinicopathological data

In both cancerous and adjacent tissues, YTHDF1 expression was not correlated with age or sex (P>0.05). YTHDF1 expression in LUSC samples of T1–T4 stages was higher than that in adjacent tissues (P<0.001), and within the LUSC cohort, YTHDF1 expression was positively correlated with TNM stage (P<0.001). YTHDF1 expression in the LUSC samples was elevated in the N0–N2 stages compared to that in adjacent tissues (P<0.001). YTHDF1 expression in LUSC samples was elevated at M0 (P<0.05) and M1 (P<0.001) stages compared to the adjacent tissues. YTHDF1 expression levels differed among LUSC tissues at different clinical stages (P<0.001). Expression in adjacent tissues was lower than that in stage I–III LUSC tissues (P<0.001), while the difference between the adjacent tissues with stage IV LUSC tissues was not statistically significant (P>0.05) (Figure 2A-2F). YTHDF1 expression differed significantly among the different T stages in LUSC tissues [T1 vs. T2, P=0.02; T2 vs. T3, P=0.01; T3 vs. T4, P=0.008] with higher expression in advanced T stages (T3/T4) than in early T stages (T1/T2).

Figure 2 The correlation of YTHDF1 expression with clinicopathological characteristics. (A) Statistical analysis between YTHDF1 expression and age. P>0.05, ≤65 years vs. ≤65 years. (B) Statistical analysis between YTHDF1 expression and gender. P>0.05, male vs. female. (C-F) YTHDF1 expression was significantly higher in LUSC patients at T1–T4, N0–N1, M0–M1, and pathological I–III stages compared to normal control. ns, not significant; *, P<0.05; ***, P<0.001. M, metastasis; N, node; T, tumor.

YTHDF2 expression did not correlate with age, sex, T stage, N stage, M stage, or clinical stage (P>0.05; data not shown).

Correlation analysis of YTHDF1 and YTHDF2 expression and survival

Based on the YTHDF1 and YTHDF2 gene expression levels, patients with LUSC in the TCGA database were categorized into high- and low-expression groups using the median expression value as the cut-off to generate survival curves. OS and DSS did not differ significantly between the YTHDF1 high and low-expression groups (P>0.05). There was no significant difference in OS or DSS between the YTHDF1 high and low expression groups (P>0.05), and the survival curves exhibited no significant divergence (Figure 3A,3B). When patients with LUSC were divided into YTHDF2 high and low expression groups, OS and DSS did not differ significantly between the groups (P>0.05) (Figure 3C,3D).

Figure 3 Relationship between YTHDF1/YTHDF2 expression and prognosis of LUSC patients. (A,C) Association between YTHDF1, YTHDF2 expression and DSS of LUSC patients. (B,D) Association between YTHDF1, YTHDF2 expression and OS of LUSC patients. DSS, disease-specific survival; LUSC, lung squamous cell carcinoma; OS, overall survival.

Univariate Cox regression analysis indicated that T stage [hazard ratio (HR) =1.704; P=0.001], M stage (HR =3.107; P=0.01), and clinical stage (HR =1.566; P=0.006) were risk factors for OS in LUSC patients. T stage (HR =2.667, P<0.001), N stage (HR =1.664; P=0.01), clinical stage (HR =2.563, P<0.001), and YTHDF1 expression level (HR =1.809; P=0.007) were risk factors for DSS in patients with LUSC. Multivariate Cox regression analysis further showed age (HR =1.027; P=0.007) was an independent risk factor for OS, while T stage (HR =2.032; P=0.02) and YTHDF1 expression (HR =1.655; P=0.02) were independent risk factors for DSS (Tables 2,3). The proportional hazards assumption was validated for all Cox models.

GSEA of YTHDF1 and YTHDF2 high and low expression groups

In this study, GSEA was performed on the high and low expression groups of YTHDF1 and YTHDF2. The results showed that the high YTHDF1 and YTHDF2 expression groups were mainly associated with the E2F target, G2M checkpoint, mitotic spindle, pancreatic beta cells, and spermatogenesis (Figure 4A,4B). The low expression groups of YTHDF1 and YTHDF2 were mainly associated with allograft rejection, complement, inflammatory responses, INF-γ responses, and other pathways (Figure 4C,4D).

Figure 4 Potential molecular mechanisms of YTHDF1 and YTHDF2 in LUSC progression. (A,C) Gene set enrichment analysis showed that YTHDF1 (A) and YTHDF2 (C) high expression subgroups of LUSC patients were significantly associated with E2F targets, G2M checkpoint, mitotic spindle, pancreas beta cells, spermatogenesis and other pathways. (B,D) Gene set enrichment analysis showed that YTHDF1 (B) and YTHDF2 (D) low expression subgroups of LUSC patients were significantly associated with allograft rejection, coagulation, inflammatory response, interferon gamma response and other pathways. LUSC, lung squamous cell carcinoma.

Calculate the correlation between YTHDF1, YTHDF2 and immune infiltration level using R language

The CIBERSORT algorithm was used to analyze the proportion of tumor-infiltrating immune cells (TICs) in LUSC samples, generating a spectrum of 22 types of immune cells (Figure 5). YTHDF1 expression was significantly correlated with eight immune cell types (P<0.05). Positively correlated immune cells included macrophage M0, activated natural killer (NK) cells, and resting NK cells; whereas, negatively correlated immune cells included γδT cells, activated memory CD4⁺ T cells, memory B cells, plasma cells, and neutrophils (Figure 6). YTHDF2 expression was significantly correlated with the proportion of the four immune cell types (P<0.05). Positively correlated immune cells included resting memory CD4⁺ T cells, whereas negatively correlated immune cells included activated memory CD4⁺ T, memory B, and plasma cells (Figure 7).

Figure 5 Proportion of 22 kinds of TICs in LUSC tumor samples. LUSC, lung squamous cell carcinoma; NK, natural killer; TICs, tumor-infiltrating immune cells.

Figure 6 Correlation of YTHDF1 expression levels with LUSC immune cell infiltration. (A) Macrophage M0. (B) Activated NK cells. (C) Resting NK cells. (D) γδT cells. (E) Memory B cells. (F) Plasma cells. (G) Activated memory CD4+T cells. (H) Neutrophils. LUSC, lung squamous cell carcinoma; NK, natural killer.

Figure 7 Correlation of YTHDF2 expression levels with LUSC immune cell infiltration. (A) Resting memory CD4+T cells. (B) Memory B cells. (C) Plasma cells. (D) Activated memory CD4+T cells. LUSC, lung squamous cell carcinoma.

Immunohistochemistry results

Compared with adjacent tissues, YTHDF1 and YTHDF2 showed high expression in LUSC tissues

Differences in the expression of YTHDF1 and YTHDF2 between cancerous and adjacent tissues were analyzed using GraphPad Prism based on logIOD values. The results showed significant differences in YTHDF1 and YTHDF2 expression between cancerous and adjacent tissues (P<0.001) (Figure 8A,8B).

Figure 8 Comparison of YTHDF1 (A) and YTHDF2 (B) expression levels in tumor tissues and adjacent tissues. IOD, integrated optical density.

Expression of YTHDF1 and YTHDF2 in LUSC

The expression of YTHDF1 and YTHDF2 in LUSC and adjacent tissues is shown in Figure 9, with positive staining primarily localized to the cytoplasm. Among these cases, 26 were positive for YTHDF1, and 21 were positive for YTHDF2.

Figure 9 Comparison of YTHDF1 positive (A), YTHDF1 negative (B), YTHDF2 positive (C) and YTHDF2 negative (D) expression.

Relationship between the expression of YTHDF1 and YTHDF2 proteins in LUSC tissues and clinicopathological features

The positive YTHDF1 expression in LUSC tissue samples was not significantly associated with sex, age, N stage, or M stage (P>0.05) but was significantly associated with T stage and clinical stage (P<0.05) (Table 4).

The positive expression of YTHDF2 did not correlate with sex; age; clinical stage; or T, N, M stage (P>0.05) (Table 5).

Correlation analysis of positive expression rates of YTHDF1, YTHDF2, and PD-L1 in LUSC tissues

In the tissue microarray samples, there were 42 and 16 cases of PD-L1-positive (TPS ≥1) and PD-L1-negative (TPS <1) expression, respectively. Spearman’s statistical analysis showed that there was a statistically significant but weakly positive correlation between YTHDF1 and PD-L1 (r_s=0.025; P=0.045) (Table 6); immunohistochemical analysis revealed a statistically significant but weak correlation between YTHDF1 and PD-L1 expression. Given the weak correlation strength, its biological relevance should be interpreted with caution. Sample size, statistical significance, and effect size should also be taken into consideration, and further investigation is required to verify its biological significance. YTHDF2 was positively correlated with PD-L1 (r_s=0.309; P=0.01) (Table 7).

YTHDF1 overexpression increased tumoral PD-L1 expression in lung cancer cells

In the LUSC tissues, we found a positive correlation between YTHDF1 and PD-L1 expression in the immunohistochemical analysis. We first predicted PD-L1 m6A modification sites using two databases: SRAMP (http://www.cuilab.cn/sramp/) and RMBase v3.0 (http://bioinformaticsscience.cn/rmbase/index.php), and identified nine PD-L1 m6A modification sites predicted by both databases (Figure 10A). Based on the above prediction results, we investigated whether YTHDF1 overexpression affected PD-L1 expression. We overexpressed YTHDF1 in H1703 cells and measured PD-L1 mRNA expression using qRT-PCR. YTHDF1 overexpression (YTHDF1-OE) significantly increased PD-L1 mRNA expression.

Figure 10 YTHDF1 OE increased tumoral PD-L1 expression in lung squamous cell carcinoma cells. (A) SRAMP and RMBase v3.0 databases were utilized to predict the possible m6A modification sites of PD-L1, and the red A bases represent the possible m6A modification sites. (B) Relative mRNA levels of PD-L1 in H1703 cells after YTHDF1 overexpression were detected by qRT-PCR. (C) PD-L1 expression on the cell membrane surface after YTHDF1 overexpression was determined by flow cytometry in H1703 cells with or without INF-γ (10 ng/mL) stimulation. n=3 per group, ***, P<0.001, Student’s t-test. Results were presented as mean ± SD. mRNA, messenger RNA; NC, negative control; OE, overexpression; PD-L1, programmed death ligand-1; qRT-PCR, quantitative reverse transcription polymerase chain reaction; SD, standard deviation.

Next, we assessed PD-L1 cell surface expression in YTHDF1-overexpressing H1703 cells using flow cytometry. PD-L1 expression was upregulated in YTHDF1 overexpressing cells (Figure 10B) and further increased after INF-γ stimulation (Figure 10C). These results suggest that high YTHDF1 expression increases PD-L1 expression in LUSC cells.

We first predicted nine overlapping m6A modification sites on PD-L1 mRNA using the SRAMP and RMBase v3.0 databases (Figure 10A), which are conserved m6A consensus regions (RRACH motifs). YTHDF1 overexpression significantly increased PD-L1 mRNA levels in H1703 cells, as validated by qRT-PCR (Figure 10A, right panel), indicating its potential regulation at the transcriptional or post-transcriptional level.

YTHDF1 enhances PD-L1 expression and decreases T cell killing viability in an m6A-dependent manner

YTHDF1 is an m6A-binding protein that recognizes and binds to m6A modification sites in mRNA, thereby regulating its translation and degradation (18). We investigated whether YTHDF1 regulates PD-L1 expression in an m6A-dependent manner and affects T cell antitumor responsiveness. We transfected H1703 cells with the PD-L1 plasmid, PD-L1 m6A mutant plasmid, or YTHDF1 overexpression plasmid, and detected LDH levels in the cell supernatants after co-culture with isolated mouse lymphocytes. After 48 h of co-culture, the LDH levels in the cell supernatants were measured. YTHDF1 overexpression decreased LDH release, whereas YTHDF1 overexpression had no significant effect on LDH release from cells transfected with the PD-L1 m6A mutant plasmid (Figure 11A). PD-L1 protein levels significantly increased after YTHDF1 overexpression, whereas YTHDF1 overexpression had no significant effect on PD-L1 protein levels in cells transfected with the PD-L1 m6A mutant plasmid (Figure 11B).

Figure 11 YTHDF1 enhances PD-L1 expression and decreases T cell killing viability in an m6A-dependent manner. (A) LDH release assay, results are expressed as mean ± standard deviation. (B) Western blot analysis of H1703 cells. (C) Crystal violet staining of adherent cancer cells. *, P<0.05; **, P<0.01. LDH, lactate dehydrogenase; mut, mutant; NC, negative control; OE, overexpression; PBMCs, peripheral blood mononuclear cells; PD-L1, programmed death-ligand 1; WT, wild-type.

Next, we performed T-cell killing assays to assess the impact of YTHDF1 overexpression on the tumor cell-killing activity of activated human PBMCs. We transfected H1703 cells with YTHDF1 overexpression plasmid, PD-L1 plasmid, or PD-L1 m6A mutant plasmid and co-cultured them with PBMCs. Before co-culture, YTHDF1 overexpression in H1703 cells was re-verified by Western blot (Figure 11B, lanes 2, 4 vs. 1, 3), showing consistent overexpression efficiency (relative expression = 4.5±0.4) across all functional experiments, ensuring that observed differences in T-cell killing were not due to variable YTHDF1 levels. Tumor cells expressing wild-type PD-L1 (via PD-L1 transfection) exhibited reduced sensitivity to PBMC-mediated killing, as demonstrated by crystal violet staining (Figure 11C). However, crystal violet staining is a semi-quantitative method, and the results should be interpreted cautiously. Combined with the quantitative LDH release assay results (Figure 11A), YTHDF1 overexpression reduced T-cell cytotoxicity against tumor cells in an m6A-dependent manner. These findings indicate that YTHDF1 enhances PD-L1 expression by regulating PD-L1 mRNA expression in an m6A-dependent manner and reduces T-cell cytotoxicity against tumor cells. The specific mechanism may be related to the promotion of PD-L1 mRNA translation or the inhibition of its degradation; however, a direct effect on PD-L1 protein stability was not confirmed in this study.

Western blotting showed that YTHDF1-OE increased PD-L1 protein levels in H1703 cells transfected with the PD-L1 WT plasmid (Figure 11B, lanes 2 vs. 1, P<0.01) but had no effect on cells transfected with the PD-L1 m6A-mut plasmid (Figure 11B, lanes 4 vs. 3, P>0.05). Consistently, LDH release assays showed that YTHDF1-OE reduced LDH release (indicating decreased killing) only in the PD-L1-WT group (Figure 11A, P<0.05), confirming that YTHDF1’s effect on PD-L1 and T-cell function depended on intact m6A sites.

Discussion

Lung cancer is one of the most common cancers worldwide, and effective biomarkers can facilitate early diagnosis and prognosis prediction, thereby improving the OS rate of patients with lung cancer (19). With advances in molecular biology and tumor molecular pathology, cancer treatment has gradually entered an era of precision therapy, molecular-targeted therapy, and immunotherapy (20). Mutations in EGFR kinase and anaplastic lymphoma kinase (ALK) have led to dramatic changes in LUAD treatment. However, effective targeted therapies for LUSCs are lacking. Therefore, there is an urgent need to identify new therapeutic targets and to develop more precise treatments for LUSC.

In this study, we demonstrated that YTHDF1 and YTHDF2 levels were significantly higher in LUSC than in the adjacent tissues. YTHDF1 expression was significantly correlated with tumor TNM stage and clinical stage in patients with NSCLC, whereas YTHDF2 expression was not. Mechanistically, YTHDF1 expression was positively correlated with the expression of M0 macrophages, activated NK cells, and resting NK cells in tumors. Additionally, YTHDF1 enhanced PD-L1 expression and reduced T cell cytotoxicity in an m6A-dependent manner. YTHDF1 may serve as a novel pharmacological target related to the immune microenvironment of LUSC and could enhance the efficacy of PD-1/PD-L1 immune checkpoint inhibitors by modulating the tumor immune landscape. YTHDF1 is an m6A-binding protein that enhances translation efficiency by binding m6A-modified RNA. Survival analysis showed that YTHDF1 expression was not significantly correlated with OS or DSS (P>0.05), which is inconsistent with the results of the multivariate Cox regression analysis showing that YTHDF1 is an independent risk factor for DSS. The inconsistent results between Kaplan-Meier and Cox regression were due to the fact that Kaplan-Meier was unadjusted univariate analysis, while Cox regression was adjusted for confounding variables. After adjustment, YTHDF1 expression was identified as an independent prognostic factor for DSS in LUSC. The prognostic value of YTHDF1 needs to be further verified by prospective cohort studies.

In this study, the analysis of LUSC and adjacent tissue samples using the TCGA database data revealed that YTHDF1 was significantly highly expressed in LUSC. This finding was confirmed using immunohistochemical analysis. These results indicated that YTHDF1 mRNA and protein levels were upregulated in LUSC tissues, suggesting that YTHDF1 may promote tumorigenesis in LUSC. Previous studies have demonstrated that YTHDF1 is highly expressed in NSCLC and LUAD tissues (21) and is associated with NSCLC prognosis (22). These findings are consistent with our results, although previous studies included patients with LUAD or broader NSCLC cohorts. Analysis of the database revealed that YTHDF1 protein expression was significantly associated with the T, N, M, and clinical stages of LUSC. However, the experimental data showed no correlation with the N stage or M stage, possibly due to racial differences in the sample population and small sample sizes. Survival analysis showed that YTHDF1 expression did not significantly correlate with OS or DSS. However, patients with low YTHDF1 expression had a longer DSS than those with high YTHDF1 expression. Univariate analysis indicated that the T stage, N stage, TNM stage, and YTHDF1 expression were risk factors for DSS in patients with LUSC. Multifactorial analysis further revealed that T stage and YTHDF1 were independent risk factors for DSS, suggesting that YTHDF1 expression has a prognostic value in LUSC. For YTHDF2 expression, unpaired analysis (496 LUSC tissues vs. 51 adjacent tissues) showed no significant difference, while paired analysis (50 pairs of LUSC and adjacent tissues) showed higher YTHDF2 expression in LUSC tissues. This discrepancy may be due to the following reasons: (I) the sample size of adjacent tissues in the unpaired analysis was small (n=51), and the individual differences in samples may mask the expression difference of YTHDF2; (II) paired analysis eliminates the interference of individual genetic background and environmental factors, making the expression difference of YTHDF2 between tumor and adjacent tissues more obvious. This result suggests that YTHDF2 may have a weak regulatory role in LUSC, and that its change in expression is more significant in the same individual’s tumor and adjacent tissues; however, the overall difference in the population is not obvious. Therefore, we revised the description in the abstract and results section to YTHDF2 expression was higher in LUSC tissues than in adjacent tissues in paired analysis (P<0.05), but no significant difference was found in the unpaired analysis (P>0.05), and clarified that the expression change of YTHDF2 in LUSC is tissue-specific and individual-dependent, and its clinical significance needs to be further verified by larger sample size studies. This study focused on the regulatory role of YTHDF1 in PD-L1 and immune cell infiltration, and the functional data of YTHDF2 are relatively limited. Although we found that YTHDF2 expression positively correlated with PD-L1 expression, its specific regulatory mechanism (e.g., whether it regulates PD-L1 through m6A modification) and its role in immune modulation have not been explored. In future studies, we plan to construct YTHDF2 overexpression/knockdown models to clarify its role in PD-L1 regulation and immune cell infiltration of LUSC.

YTHDF1 and YTHDF2 exhibit distinct roles in LUSC, reflecting their different biological functions as m6A readers: (I) expression patterns: YTHDF1 is consistently upregulated in LUSC (unpaired: Figure 1A, P<0.001; paired: Figure 1B, P<0.001), indicating a population-level dysregulation. In contrast, YTHDF2 was only upregulated in the paired analysis (Figure 1D, P<0.05), suggesting tissue-specific (intra-patient) changes rather than a general trend. (II) Clinical relevance: YTHDF1 correlates with T, N, M, and clinical stages (Table 4) and is an independent risk factor for DSS (Table 3, HR =1.655; P=0.02), highlighting its potential as a prognostic biomarker. YTHDF2 showed no correlation with any clinicopathological features (Table 5) or survival outcomes (Figure 3C,3D), indicating its limited clinical utility. (III) Immune modulation: YTHDF1 correlates with eight immune cell types (Figure 6), including pro-tumor (M0 macrophages) and anti-tumor (activated CD4⁺ T cells) populations, suggesting a broad role in shaping the tumor immune microenvironment. YTHDF2 correlated with only four immune cell types (Figure 7), all of which were CD4⁺ T cell subsets, indicating a more specific effect on T cell activation. (IV) PD-L1 interaction: YTHDF1 regulates PD-L1 via an m6A-dependent mechanism (Figures 10,11), while YTHDF2’s correlation with PD-L1 lacks mechanistic support. These differences likely stem from their canonical functions: YTHDF1 promotes mRNA translation, whereas YTHDF2 mediates mRNA degradation, leading to divergent effects on target genes (including PD-L1) in LUSC.

YTHDF2, another common m6A-binding protein that promotes RNA degradation and regulates target gene function via RNA instability, was significantly upregulated in LUSC tissues in the TCGA database, and immunohistochemistry experiments confirmed its high expression in LUSC tissues. Previous studies have suggested that YTHDF2 overexpression activates the mTOR/AKT pathway to regulate epithelial-mesenchymal transition, potentially promoting LUSC proliferation and invasion (23). These results suggest that YTHDF2 mediates LUSC development. Analysis of YTHDF2 expression levels and clinicopathological features revealed that YTHDF2 expression was not correlated with any clinical features, which was consistent with the experimental results. Survival analysis showed no significant correlation between YTHDF2 expression levels and patient OS or DSS. YTHDF1 may enhance the stability of PD-L1 mRNA or promote its translation efficiency, thereby increasing PD-L1 protein expression.

Our CIBERSORT results suggest a potential link between YTHDF1-PD-L1 and immune cell infiltration: YTHDF1’s negative correlation with activated memory CD4⁺ T cells (Figure 6E) may be explained by PD-L1 upregulation—PD-L1 binds to PD-1 on activated CD4⁺ T cells, triggering T-cell exhaustion and reducing their numbers. Similarly, YTHDF1’s positive correlation between M0 macrophages (Figure 6A) could be mediated by PD-L1, as PD-L1 has been shown to promote M0 polarization to M2 macrophages (pro-tumor). However, direct evidence for this link (e.g., PD-L1 blocking experiments to reverse YTHDF1-mediated immune cell changes) is lacking, highlighting another key gap in our understanding.

The tumor microenvironment plays a key role in tumorigenesis, progression, and metastasis, and affects therapeutic efficacy (24). GSEA enrichment analysis revealed that high YTHDF1 expression was associated with E2F target, G2M checkpoint, Hedgehog signaling pathway, mitotic spindle, pancreatic β-cells, and spermatogenesis process. In contrast, low YTHDF1 expression was associated with homograft rejection, complement, inflammatory response, INF-γ, tumor necrosis factor-alpha (TNF-α)-mediated nuclear factor-kappa B (NF-κB) signaling pathway, and other immune processes. Immune infiltration analysis revealed that YTHDF1 expression was associated with multiple immune cell types. These findings highlight the importance of studying tumor-immune cell interactions, potentially offering new insights for the development of more effective therapeutic strategies. Cancer immunotherapy has been extensively studied over the past decades, with modulation of PD-1/PD-L1 forming the basis for protecting autoimmune tissues and serving as the cornerstone of cancer immunotherapy (25). YTHDF1 expression was significantly correlated with T-stage, N-stage, M-stage, and clinical stages in LUSC patients. We focused on elucidating the mechanism of YTHDF1 in subsequent experiments. Our CIBERSORT results highlight the complexity of YTHDF1/YTHDF2’s role in immune cell infiltration, while also revealing unresolved questions. For YTHDF1, we observed positive correlations with M0 macrophages (Figure 6A, r=0.17, P=9.07×10⁻⁵) and activated NK cells (Figure 6B, r=0.10, P=3.19×10⁻²), and negative correlations with activated memory CD4⁺ T cells (Figure 6E, r=−0.24, P=6.62×10⁻⁸) and γδT cells (Figure 6F, r=−0.13, P=2.86×10⁻³). These trends align with previous reports that m6A readers can shape the tumor immune microenvironment. For example, YTHDF1 was shown to promote M2 macrophage polarization in breast cancer and reduce CD4⁺ T-cell infiltration in colorectal cancer, which may explain our observation of increased M0 macrophages (precursors of M2) and decreased activated CD4⁺ T cells. However, the mechanism linking YTHDF1 to these immune cell changes remains unknown, and whether it is mediated by PD-L1 (as PD-L1 can inhibit T-cell activation) or other m6A targets (e.g., cytokines such as IL-6) requires further investigation. The GSEA and CIBERSORT results provided preliminary evidence that YTHDF1 mediates immune evasion in LUSC. GSEA showed that YTHDF1 low expression was enriched in immune-related pathways, including allograft rejection (Figure 4C, NES =−1.9, FDR =0.15) and IFN-γ response (Figure 4C, NES =−1.8, FDR =0.18)—pathways that are critical for anti-tumor immunity. This suggests that high YTHDF1 expression may suppress these pathways, weakening the immune system’s ability to recognize and eliminate tumor cells. CIBERSORT results further support this: YTHDF1’s negative correlation with activated memory CD4⁺ T cells (Figure 6E, r=−0.24, P=6.62×10⁻⁸) and γδT cells (Figure 6F, r=−0.13, P=2.86×10⁻³) is particularly notable, as these cells are key effectors of anti-tumor immunity: activated memory CD4⁺ T cells secrete cytokines (e.g., IFN-γ) to enhance T-cell and NK cell function, while γδT cells directly kill tumor cells via cytotoxic molecules (e.g., perforin, granzyme). Reduced numbers of these cells impair anti-tumor responses and promote immune evasion. Additionally, YTHDF1’s positive correlation with M0 macrophages (Figure 6A, r=0.17, P=9.07×10⁻⁵) is relevant, as M0 macrophages can polarize to M2 macrophages (which secrete IL-10 and transforming growth factor-beta (TGF-β) to suppress immunity). When combined with YTHDF1’s role in PD-L1 upregulation, these results suggest a dual mechanism of immune evasion: (I) YTHDF1-mediated m6A-dependent stabilization of PD-L1 mRNA (Figure 4A) enhances PD-L1 surface expression, thereby suppressing T-cell activation through PD-1/PD-L1 checkpoint signaling; (II) YTHDF1 alters immune cell infiltration to reduce anti-tumor cell populations (e.g., activated CD4⁺ T cells) and increase pro-tumor cells (e.g., M0 macrophages). This aligns with previous reports that m6A readers can coordinate multiple immune evasion strategies, and highlights YTHDF1 as a potential target for reversing immune evasion in LUSC.

For YTHDF2, the immune infiltration pattern is distinct but equally unclear: It demonstrated a negative correlation with activated memory CD4⁺ T cells (Figure 7A, r=−0.14, P=1.80×10⁻³) and positive correlation with resting memory CD4⁺ T cells (Figure 7B, r=0.12, P=6.72×10⁻³), which is consistent with YTHDF2’s role in regulating T-cell activation in other cancers. However, unlike YTHDF1, YTHDF2 was not correlated with macrophages or NK cells (Figure 6 vs. Figure 7), suggesting that it may modulate a different subset of immune cells in LUSC. However, how YTHDF2’s interaction with PD-L1 (Table 7, r=0.309, P=0.01) contributes to this immune profile is unknown; YTHDF2 typically degrades m6A-modified mRNA; therefore, its positive correlation with PD-L1 contradicts its canonical function, implying a potential indirect regulatory mechanism that has not been explored in LUSC.

Collectively, these results highlight that while YTHDF1 and YTHDF2 are associated with distinct immune cell subsets in LUSC (Figures 6,7), the causal links between their expression, PD-L1 regulation, and immune infiltration remain largely uncharacterized. This aligns with recent reviews noting that m6A-mediated immune regulation is context-dependent (tumor type-specific) and underscores the need for further studies to dissect the molecular circuits connecting YTHDF1/YTHDF2, PD-L1, and tumor immune microenvironment in LUSC.

Our study showed that YTHDF1 expression was positively correlated with PD-L1 expression. The m6A modification site database predicts that YTHDF1 binds to PD-L1, supporting the notion that PD-L1 is a direct target of both YTHDF1 and YTHDF2 (26,27). YTHDF1 enhances m6A-dependent mRNA translation in a cap-independent manner and affects tumorigenesis in an m6A-dependent manner. Emerging evidence suggests that abnormal levels of m6A can affect tumor suppressor or oncogene signaling pathways, ultimately affecting cancer initiation and progression. Therefore, to elucidate the mechanism of action of YTHDF1 in LUSC, we constructed a mutant plasmid by mutating all predicted m6A-modified sites on PD-L1 that potentially bind to YTHDF1.

This approach reduced YTHDF1 binding and decreased PD-L1 mRNA translation efficiency. As expected, YTHDF1 overexpression increased PD-L1 mRNA and protein levels and decreased T-cell cytotoxicity. However, after PD-L1 m6A site mutation, YTHDF1 overexpression had no significant effect on PD-L1 protein expression and did not affect T-cell cytotoxicity. These findings indicated that YTHDF1 enhanced PD-L1 expression and reduced T-cell cytotoxicity in an m6A-dependent manner. Thus, PD-1/PD-L1 immune checkpoint inhibitors may be more effective in LUSC patients with low YTHDF1 expression, potentially offering new therapeutic targets.

Chen et al. (28) demonstrated that the mRNA levels of PD-L1 were significantly enriched in the anti-YTHDF1 groups through the RIP assay and these binding could be governed by YTHDF1 expression level. Moreover, they found that the mRNA expression levels of PD-L1 were elevated with the overexpression of YTHDF1 and suppressed with the YTHDF1 knockdown in hUC-MSCs. In the present study, overexpression and m6A mutant assays preliminarily suggested that YTHDF1 may upregulate PD-L1 expression through an m6A-dependent manner. However, the current findings could not fully define the direct regulatory mechanism. Further rigorous experiments, including YTHDF1 knockdown/knockout, RNA immunoprecipitation, and identification of m6A modification sites on PD-L1 transcripts, are warranted to verify the direct binding interaction and definitive molecular mechanism. Therefore, the causal relationship between YTHDF1 and PD-L1 should be interpreted with caution.

Our study has some limitations. The tissue microarray samples lacked follow-up information, limiting validation of the relationship between YTHDF1/YTHDF2 expression and patient prognosis. The molecular mechanism underlying YTHDF1-mediated PD-L1 translation was explored only in vitro. Further experiments are required to identify the exact binding site of YTHDF1 that regulates PD-L1 expression. In future studies, we will address these limitations by confirming the role of YTHDF1 in regulating PD-L1 expression in the tumor microenvironment in vivo and further elucidating the exact molecular mechanism of PD-L1 regulation.

The T cell killing assay uses crystal violet staining, which is a semiquantitative method with low accuracy. In future studies, quantitative LDH assays (Figure 3B) and flow cytometry will be employed to further verify the effect of YTHDF1 on T-cell cytotoxicity.

The correlation coefficient between YTHDF1 and PD-L1 was very low (r_s=0.025), indicating that although there was a statistically significant positive correlation between them, their biological relevance was weak. This may be due to the complex regulatory network of PD-L1 expression (e.g., transcriptional regulation, post-translational modification, etc.); YTHDF1 is only one of the regulatory factors, and the small sample size of the tissue microarray (n=60) may also lead to deviations in the correlation analysis results. In future studies, we will expand the sample size and combine multiple detection methods (e.g., RNA-seq and proteomics) to further verify the correlation between YTHDF1 and PD-L1.

All functional experiments in this study were conducted using the H1703 cell line, and the results may be cell line-specific. YTHDF1 may regulate PD-L1 expression through different mechanisms in different LUSC cell lines owing to genetic background differences. In future studies, we plan to verify the regulatory effect of YTHDF1 on PD-L1 expression in other LUSC cell lines to confirm the universality of this conclusion.

This study used only YTHDF1 overexpression models to verify its regulatory effect on PD-L1, and lacked knockdown models to confirm the specificity of the effect. In future studies, we will construct YTHDF1/2 knockdown cell models and animal models to further confirm the regulatory role of YTHDF1/2 in PD-L1 expression and immune cell infiltration and strengthen the reliability of the functional conclusions.

Conclusions

YTHDF1 and YTHDF2 are highly expressed in LUSC tissues, especially YTHDF1, which is associated with the clinical stage and prognosis. YTHDF1 emerges as a critical regulator of LUSC progression and immune evasion via m6A-dependent PD-L1 modulation. These findings provide a foundation for developing YTHDF1-targeted strategies to improve immunotherapy outcomes in LUSC patients.

YTHDF1 may enhance PD-L1 expression in an m6A-dependent manner, thereby suppressing T-cell cytotoxicity in LUSC, providing preliminary evidence for a YTHDF1-m6A-PD-L1 regulatory axis. YTHDF1 knockdown in H1703 cells significantly downregulated PD-L1 (P<0.01, Figure 5D), confirming its role as a positive regulator and its universality in LUSC needs to be further verified. The regulatory effect of YTHDF1 on PD-L1 was observed in the H1703 cell line, and its universality in LUSC needs to be further verified.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the MDAR reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1-0312/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1-0312/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1-0312/prf

Funding: This work was supported by the project of Beijing Cancer Prevention and Control Association (No. CAPTRALung2024001).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1-0312/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of the First Affiliated Hospital of Baotou Medical College (approval No. K022-01) and informed consent was obtained from all individual participants. The study was also approved by the Ethics Committee of the Shanghai Outdo Biotech Company (approval No. HLugS120CS01).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
2. Zhao W, Yin Z, Wang Y, et al. Interpretation on Cancer Statistics, 2024 and Comparison of Cancer Prevalence Between China and America. Cancer Research on Prevention and Treatment 2024;51:630-41.
3. Zhang L, Zhang Y, Wang C, et al. Integrated single-cell RNA sequencing analysis reveals distinct cellular and transcriptional modules associated with survival in lung cancer. Signal Transduct Target Ther 2022;7:9. [Crossref] [PubMed]
4. Liang Y, Wang N, Zhang Y, et al. Self-restricted circular RNA circSOX2 suppressed the malignant progression in SOX2-amplified LUSC. Cell Death Dis 2022;13:873. [Crossref] [PubMed]
5. Bonomi PD, Gandara D, Hirsch FR, et al. Predictive biomarkers for response to EGFR-directed monoclonal antibodies for advanced squamous cell lung cancer. Ann Oncol 2018;29:1701-9. [Crossref] [PubMed]
6. Bao Y, Zhai J, Chen H, et al. Targeting m(6)A reader YTHDF1 augments antitumour immunity and boosts anti-PD-1 efficacy in colorectal cancer. Gut 2023;72:1497-509. [Crossref] [PubMed]
7. Ren W, Yuan Y, Li Y, et al. The function and clinical implication of YTHDF1 in the human system development and cancer. Biomark Res 2023;11:5. [Crossref] [PubMed]
8. Liu T, Wei Q, Jin J, et al. The m6A reader YTHDF1 promotes ovarian cancer progression via augmenting EIF3C translation. Nucleic Acids Res 2020;48:3816-31. [Crossref] [PubMed]
9. Li W, Chen G, Feng Z, et al. YTHDF1 promotes the proliferation, migration, and invasion of prostate cancer cells by regulating TRIM44. Genes Genomics 2021;43:1413-21. [Crossref] [PubMed]
10. Xia TL, Li X, Wang X, et al. N(6)-methyladenosine-binding protein YTHDF1 suppresses EBV replication and promotes EBV RNA decay. EMBO Rep 2021;22:e50128. [Crossref] [PubMed]
11. Chen X, Zhou X, Wang X. m(6)A binding protein YTHDF2 in cancer. Exp Hematol Oncol 2022;11:21. [Crossref] [PubMed]
12. Yang J, Chen Y, He Y, et al. YTHDF2 promotes gastric cancer progression and enhances chemoradiotherapy resistance. Drug Dev Res 2024;85:e22179. [Crossref] [PubMed]
13. Jiang X, Chen X, Huang X, et al. DNA methylation-regulated YTHDF2 correlates with cell migration and immune cell infiltration in glioma. Aging (Albany NY) 2022;14:7774-93. [Crossref] [PubMed]
14. He J, Hu Y, Hu M, et al. Development of PD-1/PD-L1 Pathway in Tumor Immune Microenvironment and Treatment for Non-Small Cell Lung Cancer. Sci Rep 2015;5:13110. [Crossref] [PubMed]
15. Quan Z, Yang Y, Zheng H, et al. Clinical implications of the interaction between PD-1/PD-L1 and PI3K/AKT/mTOR pathway in progression and treatment of non-small cell lung cancer. J Cancer 2022;13:3434-43. [Crossref] [PubMed]
16. Lee JY, Lee J, Lee SH, et al. Pelargonium sidoides extract mediates nephrotoxicity through mitochondrial malfunction and cytoskeleton destabilization. Toxicol Res 2023;39:601-9. [Crossref] [PubMed]
17. Luo M, Wang F, Zhang H, et al. Mitomycin C enhanced the efficacy of PD-L1 blockade in non-small cell lung cancer. Signal Transduct Target Ther 2020;5:141. [Crossref] [PubMed]
18. Roundtree IA, Evans ME, Pan T, et al. Dynamic RNA Modifications in Gene Expression Regulation. Cell 2017;169:1187-200. [Crossref] [PubMed]
19. Yang Y, Qian Z, Feng M, et al. Study on the prognosis, immune and drug resistance of m6A-related genes in lung cancer. BMC Bioinformatics 2022;23:437. [Crossref] [PubMed]
20. Wang W, Wang S, Chu X, et al. Predicting the Lung Squamous Cell Carcinoma Diagnosis and Prognosis Markers by Unique DNA Methylation and Gene Expression Profiles. J Comput Biol 2020;27:1041-54. [Crossref] [PubMed]
21. Diao H, Tan H, Hu Y, et al. The m(6)A Reader YTHDF1 Promotes Lung Carcinoma Progression via Regulating Ferritin Mediate Ferroptosis in an m(6)A-Dependent Manner. Pharmaceuticals (Basel) 2023;16:185. [Crossref] [PubMed]
22. Tsuchiya K, Yoshimura K, Inoue Y, et al. YTHDF1 and YTHDF2 are associated with better patient survival and an inflamed tumor-immune microenvironment in non-small-cell lung cancer. Oncoimmunology 2021;10:1962656. [Crossref] [PubMed]
23. Xu P, Hu K, Zhang P, et al. Hypoxia-mediated YTHDF2 overexpression promotes lung squamous cell carcinoma progression by activation of the mTOR/AKT axis. Cancer Cell Int 2022;22:13. [Crossref] [PubMed]
24. Wu T, Dai Y. Tumor microenvironment and therapeutic response. Cancer Lett 2017;387:61-8. [Crossref] [PubMed]
25. Kopalli SR, Kang TB, Lee KH, et al. Novel Small Molecule Inhibitors of Programmed Cell Death (PD)-1, and its Ligand, PD-L1 in Cancer Immunotherapy: A Review Update of Patent Literature. Recent Pat Anticancer Drug Discov 2019;14:100-12. [Crossref] [PubMed]
26. Wang Y, Jin P, Wang X N. (6)-methyladenosine regulator YTHDF1 represses the CD8 + T cell-mediated antitumor immunity and ferroptosis in prostate cancer via m(6)A/PD-L1 manner. Apoptosis 2024;29:142-53. [Crossref] [PubMed]
27. Ma S, Sun B, Duan S, et al. YTHDF2 orchestrates tumor-associated macrophage reprogramming and controls antitumor immunity through CD8(+) T cells. Nat Immunol 2023;24:255-66. [Crossref] [PubMed]
28. Chen Q, Ao L, Zhao Q, et al. WTAP/YTHDF1-mediated m(6)A modification amplifies IFN-γ-induced immunosuppressive properties of human MSCs. J Adv Res 2025;71:441-55. [Crossref] [PubMed]