IGF2BP3 Promotes Head and Neck Squamous Cell Carcinoma Progression via the M6A-LDHA-Lactate-H4K8la Axis and E2F2 Activation

Abstract

Introduction

Head and neck squamous cell carcinoma (HNSCC) is characterized by metabolic reprogramming and poor prognosis. While lactate accumulates in HNSCC, how upstream RNA regulation coordinates lactate-associated epigenetic alterations during tumor progression remains unclear.

Methods

We integrated TCGA-HNSCC analyses, paired clinical specimens, and in vitro functional assays with mechanistic readouts including RIP-qPCR and LDHA 3′UTR luciferase reporters. Lactate was quantified in culture supernatants, H4K8la was assessed by immunoblotting and tissue IF, and H4K8la CUT&Tag was performed as an exploratory chromatin profiling assay.

Results

IGF2BP3 was upregulated in HNSCC and associated with adverse survival in public datasets. IGF2BP3 silencing inhibited proliferation, migration, and invasion. Mechanistically, IGF2BP3 bound LDHA mRNA and promoted LDHA expression via an m⁶A-site-dependent LDHA 3′UTR mechanism, increasing lactate and H4K8la; exogenous lactate partially restored H4K8la under pH-matched conditions. Exploratory H4K8la CUT&Tag suggested increased H4K8la signal at the E2F2 locus with enrichment of cell-cycle programs upon lactate treatment.

Conclusion

These findings support an IGF2BP3-LDHA-lactate-H4K8la axis linking post-transcriptional regulation to metabolic and chromatin remodeling in HNSCC.

Introduction

Head and neck squamous cell carcinoma (HNSCC) is the major histological subtype of head and neck cancers and represents a substantial global health burden.^1,2 Owing to the lack of reliable early-stage biomarkers, many patients are diagnosed at locally advanced or metastatic stages, where high recurrence rates and treatment resistance limit further gains in overall survival.^1-3 Accordingly, delineating actionable molecular pathways that drive HNSCC progression is critical for risk stratification and the development of effective targeted therapies.^1,2

Metabolic reprogramming is a hallmark of cancer. ⁴ In HNSCC, aerobic glycolysis is frequently enhanced, leading to elevated lactate production. ⁵ Beyond shaping an acidic and immunosuppressive tumor microenvironment, lactate can act as a signaling metabolite that influences gene expression programs.^6,7 Lactate dehydrogenase A (LDHA), a key enzyme catalyzing the conversion of pyruvate to lactate, is often aberrantly activated and correlates with tumor invasiveness and adverse clinical outcomes.^5,8 However, the upstream post-transcriptional mechanisms that sustain LDHA overexpression and lactate accumulation in HNSCC remain insufficiently defined. ⁹

The emerging non-metabolic functions of lactate further suggest a mechanistic bridge between metabolic remodeling and epigenetic regulation. Since the discovery of histone lysine lactylation, lactate-driven chromatin modification has been proposed as a framework linking cellular metabolic states to chromatin architecture and transcriptional outputs.^10,11 Accumulating evidence indicates that histone lactylation contributes to tumor-associated transcriptional regulation.^7,11,12 However, in HNSCC, key mechanistic links remain unclear, particularly how upstream RNA modifications translate into site-specific lactylation changes such as H4K8la and how these epigenetic events connect to proliferation-related transcriptional programs, including E2F transcription factor 2 (E2F2)-associated cell-cycle regulation, at promoter regions.^11,12

Insulin-like growth factor 2 mRNA-binding protein 3 (IGF2BP3) is an oncofetal RNA-binding protein that functions as a canonical m⁶A ‘reader’ to enhance the stability and translation of m⁶A-modified transcripts. ¹³ Aberrant IGF2BP3 upregulation has been linked to aggressive phenotypes and adverse outcomes across multiple malignancies, including squamous cell carcinomas,^13-15 yet the downstream metabolic-epigenetic consequences of IGF2BP3 activity in HNSCC remain insufficiently defined.

In this study, we tested the hypothesis that IGF2BP3, as an m⁶A reader, promotes lactate accumulation by enhancing LDHA expression through an m⁶A-dependent, 3′UTR-mediated post-transcriptional mechanism, thereby inducing H4K8la and engaging E2F2-associated cell-cycle programs during HNSCC progression.^{9-11,13,16,17}

Materials and Methods

Study Design

This study integrated paired clinical specimens (n = 30 pairs), analysis of the TCGA-HNSC cohort, and in vitro functional and mechanistic experiments to characterize the expression, clinical relevance, and prognostic value of the m⁶A reader IGF2BP3 in HNSCC. We further investigated whether IGF2BP3 regulates glycolytic flux and lactate-associated H4K8la to promote malignant phenotypes. In vivo, the impact of exogenous lactate on tumor growth was evaluated using TU177 xenografts (n = 5 per group). CUT&Tag profiling was performed as an exploratory assay to map H4K8la occupancy changes.

Human Specimens and Ethical Considerations

Between 2024 and 2025, 30 paired, pathologically confirmed HNSCC tumor tissues and matched adjacent normal tissues (NAT) were collected from the First Affiliated Hospital of Anhui Medical University. NAT was defined as grossly normal-appearing tissue obtained ≥2 cm from the tumor margin and was histologically confirmed to be tumor-free by two independent pathologists. None of the patients received radiotherapy, chemotherapy, immunotherapy, or targeted therapy prior to surgery. Specimens were fixed in formalin, paraffin-embedded, and sectioned for immunohistochemistry (IHC) and immunofluorescence (IF). This study was approved by the Medical Ethics Committee, The First Affiliated Hospital of Anhui Medical University (Hefei, Anhui, China; Approval No. PJ2023-12-37; approved on 2023-10-18)and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants.

Public Dataset and Survival Analysis

We downloaded STAR-counts data and corresponding clinical information for HNSCC from the TCGA database (https://portal.gdc.cancer.gov). We then extracted data in TPM format and performed normalization using the log2(TPM+1) transformation. After retaining samples that included both RNAseq data and clinical information, we ultimately selected 504 samples for further analysis. Survival differences between the two groups were compared using the log-rank test in Kaplan–Meier (KM) survival analysis. For Kaplan–Meier curves, the P value was derived from the log-rank test, and the hazard ratio (HR) with its 95% confidence interval (CI) was estimated using univariate Cox regression. All analytical methods above and R packages were performed using R software version v4.0.3. p < 0.05 was considered as statistically significant.

Cell Lines and Culture

The HNSCC cell lines TU177 and LIU-LSC-1 were used for in vitro experiments, and HEK293T cells were used for dual-luciferase reporter assays. TU177 was purchased from Huatuo Bioscience, whereas LIU-LSC-1 is a patient-derived laryngeal squamous cell carcinoma cell line established and characterized in our previous study. ¹⁸ Cell identity was verified by short tandem repeat (STR) profiling, and all cell lines were routinely tested for mycoplasma contamination. TU177 and LIU-LSC-1 cells were cultured in RPMI-1640 supplemented with 10% fetal bovine serum (FBS). HEK293T cells were maintained in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% FBS. All cells were incubated at 37°C in a humidified atmosphere containing 5% CO₂.

Small Interfering RNA (siRNA) Transfection

siRNA targeting IGF2BP3 (sense: 5′-CCU​UGA​AAG​UAG​CCU​AUA​UTT-3′; antisense: 5′-AUA​UAG​GCU​ACU​UUC​AAG​GTT-3′) and negative control siRNA (siNC; sense: 5′-UUC​UCC​GAA​CGU​GUC​ACG​UTT-3′; antisense: 5′-ACG​UGA​CAC​GUU​CGG​AGA​ATT-3′) were synthesized by GenePharma (Shanghai, China). Cells at 50–60% confluence were transfected using Lipofectamine RNAiMAX (Thermo Fisher Scientific) at a final siRNA concentration of 50 nM. Knockdown efficiency was validated by qRT-PCR and Western blotting at 24 h and 48 h post-transfection. Functional experiments were performed within 72 h post-transfection. Independent biological replicates (≥3) were performed unless otherwise stated.

Quantitative Real-Time PCR

Total RNA was extracted using TRIzol reagent (Invitrogen), reverse-transcribed into cDNA, and analyzed by SYBR Green-based qRT-PCR. Relative gene expression was calculated using the 2^−ΔΔCt method, with GAPDH as the internal control.

Western Blotting and Histone Extraction

Cells were lysed in RIPA buffer supplemented with protease and phosphatase inhibitors. Protein concentrations were determined using the bicinchoninic acid (BCA) assay. Equal amounts of protein were separated by SDS-PAGE and transferred onto PVDF membranes. After blocking, the membranes were incubated overnight at 4 °C with primary antibodies against IGF2BP3 (Proteintech, Cat# 14642-1-AP, 1:2000), LDHA (Cell Signaling Technology, Cat# 3582, 1:1000), GAPDH (Proteintech, Cat# 60004-1-Ig, 1:5000), Histone H4 (Cell Signaling Technology, Cat# 2935, 1:1000), and H4K8la (PTM BIO, Cat# PTM-1415RM, 1:2000). The membranes were then incubated with HRP-conjugated goat anti-rabbit IgG (H+L) and HRP-conjugated goat anti-mouse IgG (H+L) secondary antibodies (Yeasen, Cat# 33101ES60 and 33201ES60, respectively) at 1:5000, followed by visualization using an enhanced chemiluminescence (ECL) system. IGF2BP3 and LDHA signals were normalized to GAPDH, and H4K8la was normalized to total Histone H4. For histone extraction, cells were incubated with 0.2 N HCl overnight at 4 °C, followed by centrifugation, and the supernatant was collected for H4K8la analysis.

Immunohistochemistry and H-Score Assessment

Paraffin sections (5 μm) were deparaffinized in xylene and rehydrated through graded ethanol. Antigen retrieval was performed by heating slides in 10 mM citrate buffer (pH 6.0) using a microwave oven at 92-98 °C for 15 min, followed by cooling to room temperature. Endogenous peroxidase activity was quenched with 3% H₂O₂ (in methanol) for 10 min. Sections were blocked with normal goat serum for 20 min at room temperature, and then incubated with primary antibodies against IGF2BP3 (14642-1-AP, Proteintech; 1:400) at 4 °C overnight. After washing, sections were incubated with a secondary antibody (biotinylated goat anti-rabbit IgG, SP-9001) and visualized using DAB (30-60 s), followed by hematoxylin counterstaining. IGF2BP3 staining was quantified using the H-score, calculated as (1 × % weak) + (2 × % moderate) + (3 × % strong), yielding a range of 0-300. Two pathologists blinded to clinical data independently scored all slides, and the mean score was used for statistical analyses. High and low expression groups were defined by the cohort median H-score.

Immunofluorescence and Quantification

Paraffin sections (4 μm) were deparaffinized in xylene and rehydrated through graded ethanol. Antigen retrieval was performed in 10 mM sodium citrate buffer (pH 6.0) by microwave heating for 10 min (at or just below boiling), followed by cooling to room temperature. Sections were permeabilized with 0.3% Triton X-100 in PBS for 10 min and blocked with 5% normal goat serum + 1% BSA in PBS for 1 h at room temperature. Sections were then incubated with primary antibodies against H4K8la (PTM BIO, Cat# PTM-1415RM; 1:200) at 4 °C overnight, followed by incubation with an Alexa Fluor-conjugated secondary antibody (goat anti-rabbit IgG, Alexa Fluor 594; Invitrogen, Cat# A-11012; 1:500) for 1 h at room temperature in the dark. Nuclei were counterstained with DAPI (1 μg/mL, 5 min) and slides were mounted using an antifade mounting medium. Images were captured using a fluorescence microscope (20× objective; 200× final magnification) under identical exposure settings across groups (scale bar, 50 μm). For each sample, five random fields were quantified in ImageJ after background subtraction; mean fluorescence intensity was normalized to the number of DAPI-positive nuclei.

Cell Functional Assays

Cell proliferation was assessed using Cell Counting Kit-8 (CCK-8), 5-ethynyl-2′-deoxyuridine (EdU) incorporation, and colony formation assays, whereas migration and invasion were assessed using Transwell assays with or without Matrigel. For CCK-8 assays, cells were seeded at 1 × 10⁴ cells/well in 96-well plates, and absorbance at 450 nm was measured at 24, 48, and 72 h. For EdU assays, cells were incubated with EdU at 50 μM for 2 h, and EdU-positive nuclei were quantified from five random fields per condition. For Transwell assays, 1 × 10⁵ cells/insert were seeded; for invasion assays, inserts were coated with Matrigel (1:8 dilution in Dulbecco’s modified Eagle medium [DMEM]), and migrated or invaded cells were counted in five random fields per insert.

Lactate Treatment and Measurement

Cells were treated with 20 mM sodium L-lactate for 24 h. Controls received equimolar NaCl. To minimize pH-related artifacts, lactate and control media were pH-matched to 7.4. Extracellular lactate was quantified using a Lactate Assay Kit (Abcam) according to the manufacturer’s protocol and normalized to total protein content (BCA assay). For lactate quantification, a standard curve was generated using 0, 2, 4, 6, 8 and 10 nmol/well of lactate standard (in 50 μL per well) and measured in duplicate. Each sample was assayed in technical triplicates, and experiments were repeated with at least three independent biological replicates.

RNA Sequencing and Bioinformatics Analysis

RNA sequencing was performed on three biological replicates from tumor and adjacent normal tissues using the Illumina platform. Raw reads were quality-controlled, aligned to the GRCh38/hg38 reference genome, and analyzed for differential expression using DESeq2 (FDR < 0.05). Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses, as well as gene set enrichment analysis (GSEA), were conducted using the clusterProfiler package.

Proteomics Analysis and Visualization

Quantitative proteomics data from tumor (T) and normal (N) groups (n = 3 biological replicates each) were normalized and analyzed for differential expression. Fold changes were expressed as log2(T/N) ratios, with statistical significance determined by two-tailed tests. Results were visualized as volcano plots (x-axis: log2 fold change; y-axis: −log10 P-value). Differentially expressed proteins were defined as those with |log2FC| ≥ 1 (corresponding to fold change ≥ 2 or ≤ 0.5) and P < 0.05. Upregulated, downregulated, and non-significant proteins were color-coded. Volcano plots were generated using GraphPad Prism, with key proteins annotated.

RNA Immunoprecipitation (RIP)-qPCR

RIP assays were performed using the Magna RIP Kit (MilliporeSigma) following the manufacturer’s protocol. Briefly, ∼1 × 10⁷ cells were lysed in RIP lysis buffer supplemented with RNase inhibitor. Lysates were incubated with magnetic beads conjugated to anti-IGF2BP3 antibody (Proteintech, Cat# 14642-1-AP; 5 μg per reaction) or normal rabbit IgG (5 μg) as a negative control at 4 °C overnight with rotation. After stringent washing, immunoprecipitated complexes were treated with proteinase K to release RNA. Coprecipitated RNA was purified, reverse-transcribed, and analyzed by qRT-PCR to quantify enrichment of target transcripts (e.g., LDHA mRNA). RIP enrichment was calculated as percent input and further expressed as fold enrichment over IgG. Each RIP-qPCR was performed with technical duplicates and repeated in ≥3 independent biological replicates.

Dual-Luciferase Reporter Assay

Wild-type (WT) and m⁶A-site mutant (Mut) LDHA 3′UTR reporter constructs were used for dual-luciferase assays in HEK293T cells. Cells were seeded in 24-well plates and transfected at 70-80% confluence. To achieve efficient knockdown, cells were transfected with siRNA (final 50 nM) and, 24 h later, transfected with the WT and Mut reporter plasmids (500 ng/well) together with a Renilla control plasmid (50 ng/well) using a standard transfection reagent. An empty reporter vector (without the LDHA 3′UTR insert) was included as a baseline control. At 48 h after reporter transfection, firefly and Renilla luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega). Reporter activity was expressed as the ratio of firefly to Renilla luciferase activity and normalized to the control group. Each condition was measured in technical triplicates and repeated in ≥3 independent experiments.

Cleavage Under Targets and Tagmentation (CUT&Tag) Analysis

CUT&Tag assays for H4K8la were performed in control and lactate-treated cells using a commercial CUT&Tag kit following the manufacturer’s protocol. Briefly, 5 × 10⁴-1 × 10⁵ cells per condition were immobilized on concanavalin A beads, incubated with anti-H4K8la primary antibody (PTM BIO, Cat# PTM-1415RM; 1:50-1:100), followed by secondary antibody and pA-Tn5 tagmentation. Libraries were PCR-amplified, purified, and sequenced. Raw reads were adapter-trimmed and aligned to the hg38 genome; duplicates and low-quality reads were removed. Peaks were called using a uniform pipeline, and promoter regions were defined as TSS ± 3 kb. Peaks were annotated to nearby genes for exploratory functional enrichment and motif analyses. Because CUT&Tag was performed once per condition, chromatin profiling results were treated as exploratory and interpreted primarily at the track-level and gene-set level.

Xenograft Model

Four-week-old male BALB/c nude mice were purchased from GemPharmatech Co., Ltd. (Nanjing, China) and used for subcutaneous xenograft experiments. Mice were maintained under specific pathogen-free (SPF) conditions at 20–26°C and 40%–70% relative humidity, on a 12-h light/dark cycle, with ad libitum access to sterilized food and water. Mice were acclimatized for one week before tumor inoculation. TU177 cells (1 × 10⁶ cells in 100 μL PBS) were injected subcutaneously into the flank of each mouse. ¹⁹ Mice were randomized into groups (n = 5 per group) when tumors became palpable. Treatment administration, tumor measurement, and data analysis were performed in a blinded manner. Predefined exclusion criteria included failed tumor engraftment, accidental injection failure, or severe non-study-related illness; however, no animals, experimental units, or data points met these criteria or were excluded from the analysis. The group size was prespecified to minimize animal use while maintaining the ability to detect biologically meaningful differences. Tumor volume was calculated as V = (length × width²)/2, and tumor size was monitored every 3 days. The lactate group received intraperitoneal sodium L-lactate (350 mg/kg, every other day) dissolved in PBS (pH 7.4), whereas controls received equal volumes of freshly prepared pH-matched PBS. This regimen was selected as a conservative lower-intensity dosing schedule adapted from previously reported in vivo lactate administration paradigms in tumor-bearing mice.^20,21 To minimize animal suffering, mice were closely monitored throughout the study and predefined humane endpoints were applied (e.g., tumor volume ≥1500 mm³, ulceration, or >15% body-weight loss). At study termination or when predefined humane endpoints were reached, mice were deeply anesthetized with isoflurane delivered via an induction chamber at 5% in oxygen at 1 L/min until loss of pedal withdrawal reflex and respiratory arrest, and euthanasia was then completed by cervical dislocation. Death was confirmed before tissue collection. Tumors were then excised and weighed. The reporting of this study conforms to ARRIVE 2.0 guidelines ²² and the Guide for the Care and Use of Laboratory Animals. ²³ All procedures were approved by the Institutional Animal Care and Use Committee (IACUC), Anhui Medical University (Hefei, Anhui, China; Approval No. LLSC20241411; approved on 2024-03-06).

Statistical Analysis

All statistical analyses were performed using IBM SPSS (version 26.0) and GraphPad Prism. Data are presented as mean ± standard deviation (SD) unless otherwise stated. Normality was assessed using the Shapiro-Wilk test. For two-group comparisons, an unpaired two-tailed Student’s t-test was used for normally distributed data; otherwise, the Mann-Whitney U test was applied. For comparisons among three or more groups, one-way analysis of variance (ANOVA) or the Kruskal-Wallis test was used as appropriate, followed by the corresponding post hoc multiple-comparison test. For experiments involving two factors (e.g., treatment × time), two-way ANOVA was applied. For xenograft tumor growth curves with repeated measurements over time, two-way repeated-measures ANOVA was used; if missing values occurred, a mixed-effects model was applied. Endpoint tumor weight was analyzed using an unpaired two-tailed Student’s t-test. Survival differences were evaluated using the log-rank test. All P values were two-sided, and P < 0.05 was considered statistically significant. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Results

IGF2BP3 is Upregulated in HNSCC and Associated With Poor Prognosis

In an exploratory DIA-based proteomics analysis of paired tumor and adjacent normal tissues (n = 3 pairs), IGF2BP3 was among the proteins upregulated in tumors (Figure 1A). Consistently, analysis of the TCGA-HNSC cohort showed that patients with high IGF2BP3 expression (median split) had poorer overall survival than those with low expression (log-rank P = 0.032; Figure 1B). Immunohistochemistry in an independent paired cohort (n = 30) further confirmed higher IGF2BP3 H-scores in tumor tissues compared with adjacent normal tissues (Wilcoxon P < 0.0001; Figure 1C). Together, these findings identify IGF2BP3 as an upregulated and clinically relevant factor associated with poor prognosis in HNSCC (Figure 1).

Figure 1.

IGF2BP3 is upregulated in HNSCC and predicts poor prognosis. (A)Volcano plot showing differential expression between HNSCC tumors and adjacent normal tissues (T/N); IGF2BP3 is highlighted. (B) Kaplan-Meier overall survival analysis stratified by IGF2BP3 expression (high vs low); log-rank P value is indicated. (C) Representative IHC staining of IGF2BP3 in paired adjacent normal and tumor tissues, with paired H-score quantification. Representative IHC images were acquired at 200× magnification (scale bar, 50 μm). H-scores were evaluated from tumor epithelial regions by two blinded pathologists, and paired comparisons were performed across n = 30 matched samples.

IGF2BP3 Promotes Malignant Phenotypes in HNSCC Cells in Vitro

In TU177 and LIU-LSC-1 cells, siRNA-mediated IGF2BP3 knockdown markedly reduced IGF2BP3 mRNA and protein levels (Figure 2A and B). Functionally, IGF2BP3 depletion decreased cell viability in CCK-8 assays, reduced EdU incorporation, and suppressed Transwell migration and Matrigel invasion (Figure 2C-E), supporting a pro-proliferative and pro-migratory role of IGF2BP3 in HNSCC cells. Collectively, these data indicate that IGF2BP3 promotes proliferative, migratory, and invasive phenotypes in HNSCC cells in vitro (Figure 2).

Figure 2.

IGF2BP3 promotes proliferation, migration, and invasion of HNSCC cells in vitro. (A) qRT-PCR analysis confirming IGF2BP3 knockdown in TU177 and LIU-LSC-1 cells transfected with siIGF2BP3 or negative control siRNA (siNC). (B) Western blotting validation of IGF2BP3 knockdown, with heatmap summarizing relative protein levels across replicates. (C) CCK-8 assays showing reduced cell viability and proliferation following IGF2BP3 silencing. (D) Transwell migration and invasion assays in TU177 and LIU-LSC-1 cells after IGF2BP3 knockdown, with quantification. (E) EdU incorporation assays showing decreased DNA synthesis after IGF2BP3 silencing (EdU, green; DAPI, blue), with quantification of EdU-positive cells. Unless otherwise stated, quantitative data are presented as mean ± SD from 3 independent biological experiments.

IGF2BP3 Binds LDHA mRNA and Enhances LDHA Expression in an m⁶A-Dependent

Transcriptome analysis indicated enrichment of glycolysis/gluconeogenesis and other metabolic pathways among tumor-associated gene expression changes (Figure 3A). RIP-qPCR demonstrated enrichment of LDHA mRNA in IGF2BP3 immunoprecipitates compared with IgG controls (Figure 3B). In dual-luciferase assays, IGF2BP3 knockdown reduced luciferase activity driven by the WT LDHA 3′UTR but not by the empty vector or the m⁶A-site mutant 3′UTR reporter (Figure 3C), supporting an m⁶A-dependent 3′UTR mechanism. Consistently, IGF2BP3 knockdown decreased LDHA mRNA and protein levels in both TU177 and LIU-LSC-1 cells (Figure 3D,E). Taken together, these results support LDHA as a downstream target of IGF2BP3 and indicate that IGF2BP3 enhances LDHA expression through an m⁶A-dependent 3′UTR mechanism (Figure 3).

Figure 3.

IGF2BP3 binds LDHA mRNA and promotes LDHA expression via an m⁶A-dependent 3′UTR mechanism. (A) KEGG pathway enrichment analysis of differentially expressed genes identified from RNA-seq comparing HNSCC tumors versus paired NAT, highlighting glycolysis/gluconeogenesis. (B) RIP-qPCR showing enrichment of LDHA mRNA in anti-IGF2BP3 immunoprecipitates compared with IgG control, with GAPDH mRNA as a negative control. (C) Dual-luciferase reporter assays using an empty vector control, the LDHA-WT 3′UTR reporter, or the m⁶A-site-mutated LDHA-MUT 3′UTR reporter in siNC versus siIGF2BP3 conditions. (D) qRT-PCR analysis of LDHA mRNA levels following IGF2BP3 knockdown in TU177 and LIU-LSC-1 cells. (E) Western blotting of LDHA in TU177 and LIU-LSC-1 cells after IGF2BP3 knockdown (siNC vs siIGF2BP3), with heatmap summarizing relative LDHA protein levels across independent experiments. RIP-qPCR and dual-luciferase assays were repeated in 3 independent biological experiments; luciferase values were measured in technical triplicates within each experiment.

IGF2BP3 Enhances Lactate Production and Promotes H4K8la

IGF2BP3 knockdown reduced extracellular lactate levels in culture supernatants and decreased H4K8la (Figure 4A and B). Supplementation with sodium L-lactate (20 mM, 24 h) partially restored H4K8la levels in IGF2BP3-silenced cells under pH-matched conditions (Figure 4C). In paired clinical samples (n = 10 pairs), immunofluorescence revealed stronger nuclear H4K8la signals in tumor tissues than in adjacent normal tissues (Figure 4D). Together, these findings suggest that IGF2BP3 contributes to lactate accumulation and lactate-associated H4K8la in HNSCC (Figure 4).

Figure 4.

IGF2BP3 promotes lactate-driven H4K8la in HNSCC. (A) Extracellular lactate levels in TU177 and LIU-LSC-1 cells after siIGF2BP3 versus siNC. (B) Western blotting of H4K8la in TU177 and LIU-LSC-1 cells after IGF2BP3 silencing, with total H4 as loading control and heatmap summarizing relative H4K8la levels across replicates. (C) Sodium L-lactate stimulation experiments showing lactate-induced H4K8la and its modulation by IGF2BP3 knockdown under the indicated conditions (siNC + Vehicle, siIGF2BP3 + Vehicle, siIGF2BP3 + Sodium L-lactate, siNC + Sodium L-lactate), with heatmap summarizing relative H4K8la levels. (D) Representative IF staining of H4K8la in paired adjacent normal and tumor tissues (H4K8la, red; DAPI, blue), with paired quantification of H4K8la IF intensity (n = 10 paired samples). Cell-based quantitative analyses were performed in at least 3 independent biological experiments; paired tissue IF quantification was based on n = 10 matched samples.

Lactate-Associated H4K8la Chromatin Changes Suggest Activation of Cell-Cycle Programs and Promote Tumor Growth in Vivo

Exploratory CUT&Tag profiling (one sample per condition) showed globally increased H4K8la signal in lactate-treated cells (Figure 5A). Increased signal was observed around the E2F2 transcription start site and additional promoter regions (Figure 5B). Gene annotation of lactate-associated peaks followed by exploratory enrichment and motif analyses suggested overrepresentation of cell-cycle and DNA replication programs (Figure 5C). In vivo, lactate administration accelerated TU177 xenograft growth and increased endpoint tumor weights compared with controls (n = 5 per group; Figure 5D). Overall, exploratory chromatin profiling and xenograft experiments suggest that lactate-associated H4K8la remodeling is linked to cell-cycle-related programs and enhanced tumor growth in vivo (Figure 5).

Figure 5.

Lactate remodels the H4K8la chromatin landscape toward cell-cycle programs and promotes HNSCC tumor growth in vivo. (A) Representative metagene profiles and heatmaps showing global distribution of H4K8la chromatin signals across gene regions (TSS to TES) under NC versus lactate conditions. The metagene plot and heatmap shown in panel A were generated from the same CUT&Tag analysis workflow and assembled only during final figure layout; no splicing was performed within any individual source image. (B) Genome browser tracks showing H4K8la enrichment at the E2F2 locus in Lactate (LAC) versus NC conditions, with TSS indicated. (C) Exploratory KEGG pathway enrichment analysis of lactate-associated H4K8la programs, highlighting cell-cycle and related signaling pathways. (D) In vivo tumor growth assays comparing Vehicle and lactate groups, including tumor growth curves, representative excised tumors, and tumor weights (n = 5 mice per group).

Working Model

Collectively, the data support a model in which IGF2BP3 enhances LDHA expression via an m⁶A-dependent 3′UTR mechanism, increases lactate production, and promotes H4K8la. Lactate-associated H4K8la chromatin changes are accompanied by transcriptional programs related to cell-cycle regulation and tumor growth in vivo (Figure 6). Thus, the proposed working model integrates the IGF2BP3-LDHA-lactate-H4K8la axis with E2F2-related cell-cycle programs during HNSCC progression (Figure 6).

Figure 6.

Proposed working model: the IGF2BP3-m⁶A-LDHA axis may elevate lactate and is associated with H4K8la enrichment near the E2F2 locus and cell-cycle programs in HNSCC. Schematic illustrating that elevated IGF2BP3 in HNSCC promotes LDHA expression via an m⁶A-dependent 3′UTR-mediated post-transcriptional mechanism, thereby increasing LDHA expression, glycolytic flux, and glycolysis-derived lactate production. Based on CUT&Tag profiling of H4K8la, lactate treatment is associated with an increased H4K8la signal at the E2F2 promoter and transcription start site, consistent with activation of cell-cycle transcriptional programs. Collectively, this IGF2BP3-LDHA-lactate-H4K8la-E2F2 cascade supports malignant phenotypes, including proliferation, migration and invasion, and tumor growth in vivo, and is concordant with poor clinical prognosis.

Discussion

This study elucidates how metabolic reprogramming may couple post-transcriptional regulation with chromatin modification in HNSCC. Specifically, it outlines a mechanistic framework spanning the m⁶A reader IGF2BP3, the glycolytic enzyme LDHA, and lactate-associated histone lactylation, together with downstream engagement of cell-cycle transcriptional programs.^9,10,13

IGF2BP3 binds LDHA mRNA and enhances LDHA expression in an m⁶A-dependent manner, which is associated with increased glycolytic flux and lactate output, elevated H4K8la, and increased H4K8la signal at E2F2-associated promoter regions, consistent with malignant phenotypes in HNSCC. This framework integrates RNA modification, metabolic flux, histone modification, and transcriptional output into a coherent model and provides a rationale for how lactate-derived metabolic signals could contribute to nuclear transcriptional alterations.^9,10

Clinical and functional evidence collectively supports an oncogenic association of IGF2BP3 with HNSCC progression. Previous studies have established that the IGF2BP family comprises oncofetal RNA-binding proteins associated with invasive phenotypes, recurrence, and adverse prognosis across multiple malignancies. ¹³ Elevated IGF2BP3 expression in oral squamous cell carcinoma similarly correlates with poor prognosis and protumorigenic features.^14,15 Although IGF2BP3-mediated upregulation of LDHA via m⁶A-dependent pathways has been reported in other cancers, direct evidence in HNSCC has been limited. ¹⁷ The present in vitro experiments show that IGF2BP3 knockdown suppresses HNSCC cell proliferation, migration, and invasion, supporting its pro-tumorigenic role. Moreover, exogenous lactate supplementation in xenograft models accelerates tumor growth (Figure 5D), supporting the tumor-promoting potential of lactate in this setting. Together, these findings position IGF2BP3 as a plausible upstream regulator of the proposed metabolic-epigenetic cascade, while further in vivo genetic studies will be required to establish causality.

This study identifies the IGF2BP3-m⁶A-LDHA axis as a key link between post-transcriptional regulation and metabolic flux. As m⁶A reader proteins, IGF2BP1/2/3 recognize m⁶A modifications and enhance target mRNA stability and translation, thereby increasing protein expression. ¹³ Here, this conserved mechanism is applied to LDHA, a key glycolytic enzyme, supporting a cascade of m⁶A-dependent mRNA regulation, increased glycolytic activity, and lactate accumulation. ⁹ In line with prior observations linking m⁶A-mediated LDHA upregulation to metabolic reprogramming and therapy resistance, the current work further connects this axis to lactate-associated chromatin modifications, providing a framework that links Warburg-like metabolic remodeling with nuclear epigenetic regulation. ²⁴

At the nuclear level, lactate-driven histone lactylation serves as a direct conduit linking metabolic status to gene expression. Histone lactylation, a lactate-derived post-translational modification on lysine residues, has been implicated in coupling metabolism to transcriptional control. ¹⁰ In this study, H4K8la is employed as a primary epigenetic readout, with consistent alterations observed in response to lactate modulation and perturbations of the IGF2BP3-LDHA-lactate axis, indicating that lactate accumulation promotes remodeling of specific chromatin marks. Concurrently, the enzymatic machinery underlying lactylation “writers” is being elucidated, with histone acetyltransferase binding to ORC1 (HBO1) identified as a histone lysine lactyltransferase. ²⁵ These insights provide actionable avenues for genetic or pharmacological interference with lactylation processes to assess the functional necessity of H4K8la in this pathway.

Moreover, this study forges a specific link between lactylation-dependent epigenetic alterations and cell cycle transcriptional programs. The E2F family acts as central regulators of cell cycle progression and DNA replication, influencing molecular subtypes, immune infiltration, and prognosis in HNSCC. ²⁶ CUT&Tag profiling suggested H4K8la enrichment at E2F2-associated promoter and regulatory regions, with directional changes aligned with proliferative phenotypes, thus suggesting a potential link between chromatin remodeling, transcriptional reprogramming, and proliferative phenotypes. To bolster causal inference at this node, future experiments could interrogate E2F2 loss-of-function contexts to determine whether lactate- or LDHA-mediated rescue persists, complemented by locus-specific validations (e.g., qPCR at peak summits versus control regions or promoter-luciferase reporters) to substantiate the link between site-specific enrichment and transcriptional activation.

Within the broader literature, the principal contributions of this study are the construction of a cross-hierarchical regulatory loop and enhanced testability at the downstream output: it extends the canonical m⁶A reader function of IGF2BP3-mediated mRNA stabilization to LDHA, further linking it to lactate-mediated histone lactylation, thereby establishing a seamless mechanism spanning RNA modification to chromatin regulation^9,10,13; concurrently, it delineates the functional consequences of lactylation on cell cycle transcriptional programs (exemplified by the E2F2 module), rendering the connection between epigenetic alterations and tumorigenic phenotypes more amenable to experimental validation. ²⁶ Furthermore, emerging evidence suggests that IGF2BP3 itself may undergo lactylation, contributing to stress tolerance and metabolic adaptation. ²⁷ These observations complement the lactate-to-epigenetic axis described herein and imply that lactate signaling operates at multiple levels to facilitate tumor adaptation, although integrated mechanisms in HNSCC await further empirical elucidation. ¹⁰

Regarding translational implications, this study highlights multiple nodes that could be considered for combinatorial intervention. HNSCC recurrence and therapy resistance are intimately linked to tumor microenvironment dynamics, wherein lactate accumulation and associated acidification modulate the immune landscape and treatment responses.^28,29 Accordingly, targeting IGF2BP3, LDHA, and lactylation-associated enzymes represents a plausible strategy for mechanistic interrogation and potential patient stratification. However, the current evidence is primarily preclinical, and clinical efficacy will require validation in model systems and patient cohorts that more closely represent clinical settings.^28,29

This study has several limitations. First, the modest cohort size and population heterogeneity may compromise the robustness of prognostic associations, warranting validation in larger, independent, or multicenter cohorts with multivariable adjustment for confounders. Second, reliance on cell lines and subcutaneous xenografts limits recapitulation of the native head and neck microenvironment and immune interactions; future studies could employ orthotopic transplantation, patient-derived organoids, or immunocompetent models to evaluate axis functionality across therapeutic contexts. Third, the H4K8la-E2F2 association currently rests primarily on locus enrichment and phenotypic correlation, and the CUT&Tag profile was generated once per condition; downstream loss-of-function studies, together with modulation of lactylation writer and eraser enzymes, would strengthen causal linkages within the pathway. Fourth, only two HNSCC cell lines were included for in vitro validation in the current study, which may not fully capture the biological heterogeneity of HNSCC; therefore, further confirmation in additional cell models will be important to strengthen the generalizability of the findings. Fifth, only one validated siRNA sequence was available and used in the current study. Nevertheless, the knockdown effect was consistently supported at the mRNA, protein, and functional levels in two HNSCC cell lines. Additional independent siRNAs or orthogonal genetic approaches would further strengthen on-target confidence and should be incorporated in future studies.^25,26

Conclusion

This study supports an IGF2BP3-m⁶A-LDHA-lactate-H4K8la axis in HNSCC and suggests that lactate-associated chromatin remodeling may contribute to E2F2-related cell-cycle programs and tumor progression. These findings provide a mechanistic framework linking post-transcriptional regulation, metabolic remodeling, and chromatin changes in HNSCC. However, further validation in larger clinical cohorts and more representative in vivo models is required to strengthen causal inference and assess translational relevance.