Introduction

Oral squamous cell carcinoma (OSCC) stands as the predominant malignancy impacting the head and neck (Johnson et al. 2020). Despite the integration of surgery, radiotherapy, chemotherapy, and more recently immune-checkpoint inhibitors (ICIs), the clinical outcome of advanced OSCC remains unsatisfactory, the 5-year survival rate for OSCC patients continues to be lower compared to various solid tumors, with only an approximately 60% survival rate (Mroueh et al. 2020; Zhou et al. 2021; Cramer et al. 2019; Gu et al. 2022).

Although ICIs have transformed cancer therapy in several tumor types, their efficacy in OSCC is limited, largely due to the presence of an immunosuppressive TME (Bejarano et al. 2021; Tiwari et al. 2022). The TME in OSCC exhibits substantial heterogeneity, including non-immune cells, immune cells, and cytokines, chemokines, extracellular matrix, and metabolic byproducts (de Visser and Joyce 2023; Shen and Kang 2018). The bidirectional signaling between neoplastic cells and the surrounding TME profoundly influence immune suppression and tumor progression (Wang et al. 2024; Dong et al. 2024; Gan et al. 2024; Liu et al. 2025). Therefore, targeting these cellular and molecular interactions presents a promising strategy for improving immunotherapy responses in OSCC.

Among the immune components of the TME, T helper 17 (Th17) cells have acted as key players in promoting angiogenesis, inflammation, and immune evasion in various cancers (Fang et al. 2022; Chandra et al. 2024; Neuhaus et al. 2024; Theobald et al. 2021). Nevertheless, the specific function of Th17 cells in the OSCC microenvironment and the upstream regulatory mechanisms driving their polarization remain poorly understood.

Peroxisome proliferator-activated receptor gamma (PPARγ), as part of the nuclear receptor superfamily, acts pleiotropic effects on metabolism, cellular proliferation and differentiation, autoimmune diseases, and cancer (Ishtiaq et al. 2022; Christofides et al. 2021). A previous investigation indicated that PPARγ is prominently expressed in head and neck squamous cell carcinoma (HNSCC) and exhibits a negative correlation with patient prognosis (Wei et al. 2022). Nonetheless, the precise function of PPARγ in shaping the immunosuppressive landscape of OSCC has yet to be fully characterized.

In this study, we identify a novel PPARγ/CEBPA/IL-17C/IL-17A signaling axis that drives Th17 cells polarization and immunosuppression in OSCC. Our findings not only elucidate a previously unrecognized mechanism of immune evasion but also suggest that targeting PPARγ may represent a promising therapeutic strategy to disrupt tumor-promoting inflammation and enhance immunotherapeutic efficacy in OSCC.

Materials and methods

Cell culture

SCC-7 cells were acquired from the State Key Laboratory of Oral Disease, and CD4 + T cells were extracted from mouse spleens. SCC-7 cells were maintained in DMEM (Sigma, USA) with FBS (10%) and penicillin–streptomycin (1%).

Animals

The selected experimental animals were male C3H/HeJ mice aged 5–6 weeks obtained from Spfbiotech (Beijing, CHINA). Each mouse received a subcutaneous injection of 100 μl /1 × 105 cells into the flank using a disposable syringe. Mice were randomly assigned to two experimental groups: the experimental group (incubated with GW9662, 3 mg/kg, Sigma) and the control group (treated with DMSO). Every 3 days, the mice were observed for mental status, changes in body weight, and activity levels. When the animals reached the ethical euthanasia standards, the experiment was terminated, and euthanasia was performed using carbon dioxide until the last mouse model reached the euthanasia standard, marking the end of the experiment.

Patients and follow-up

A total of 104 pathologically confirmed OSCC tissue samples were collected for this study. The study protocol was reviewed and received ethical approval from the West China Hospital of Stomatology Ethics Committee.

CCK-8 assay

A 96-well plate was used to culture SCC-7 cells. For the experimental group, the medium was replaced with complete DMEM containing GW9662 (10 μM, Sigma, USA). At different time points, the original culture medium was aspirated from each well and CCK-8 solution (Beyotime, China) was added. After 30 min, measure the absorbance at 450 nm.

qPCR

Extracting RNA using the Total RNA Extraction Kit (Tiangen Biotech, China). Use the kit (Takara, Japan) to reverse-transcribed RNA into cDNA. SYBR™ Select Master Mix (Thermo Scientific, USA) was added as per the manufacturer's instructions.

Western blot analysis

Proteins were gathered using lysis buffer (Beyotime, China). After SDS-PAGE separation, proteins were transferred to PVDF membranes via electrophoresis. The membranes were then incubated with primary antibodies overnight at 4 °C, followed by incubation with HRP-labeled secondary antibodies (Proteintech, China). Protein bands were subsequently detected using the Amersham Imager 600 system (USA). Primary antibodies: GAPDH (1:5000, Cell Signaling Technology, USA), PPARγ (1:1000, Abcam, UK), CEBPA (1:1000, Proteintech, China), IL-17C (1:1000, Thermo Fisher Scientific, USA), and IL-17A (1:1000, Thermo Fisher Scientific, USA).

Transwell co-culture

SCC-7 cells were added into the lower chamber of a 24-well plate (1 × 104 cells/well). CD4 + T cells were added into the upper chamber of the 24-well plate (2.5 × 105 cells/well). The medium in the upper chamber consisted of 1640 medium supplemented with 10% FBS, glutamine (100x, Gibco), β-mercaptoethanol (5 μM, Gibco, USA), murine IL-6 (30 ng/ml, Gibco, USA), murine TNF-α (10 ng/ml, Gibco, USA), murine IL-1β (10 ng/ml, Gibco, USA), anti-mouse IL-4 (5 μg/ml, Gibco, USA), TGF-β1 (5 ng/ml, Gibco, USA), and anti-mouse IFN-γ (5 μg/ml, Gibco, USA).

In vitro Th17 cells differentiation assay with IL-17C stimulation and blockade

For this studies, recombinant mouse IL-17C (rIL-17C, 600 ng/mL; ACROBiosystems, China) or anti-IL-17C neutralizing antibody (1:500 dilution; Proteintech, China) was added to the culture medium. Cells treated with PBS or IgG as controls. Cells were gathered three days later for flow cytometric analysis.

Flow cytometry

CD4 + T cells were stained with anti-mouse CD3-APC antibody, CD4-FITC antibody, and anti-mouse IL-17A-PE antibody (BioLegend, USA). Cell populations were assessed via flow cytometry.

ELISA

To detect IL-17A in the coculture supernatant, a mouse IL-17/IL-17A ELISA kit (Neobioscience, China) was used. Manufacturer’s guidelines were followed for ELISA implementation.

RNA sequencing and bioinformatics analysis

Enrichment analysis of KEGG pathways was carried out using the clusterProfiler package in R to uncover significantly involved biological processes and signaling cascades. Gene Set Enrichment Analysis (GSEA) was implemented via the fgsea package, utilizing hallmark and immunologic signature gene sets curated from the Molecular Signatures Database (MSigDB). All raw and processed RNA sequencing data have been submitted to the NCBI Sequence Read Archive (SRA) under accession number PRJNA1250126.

Single-cell RNA-seq data acquisition and analysis

This study employed publicly available scRNA-seq datasets retrieved from the Gene Expression Omnibus (GSE172577). The dataset includes a total of 48,007 single cells derived from multiple tumor specimens. Raw gene expression matrices were downloaded and processed using the Seurat R package (v4.3.0). Cells with fewer than 200 detected genes or more than 10% mitochondrial gene expression were filtered out to remove low-quality cells. Gene expression values were log-normalized, and the top 2000 most variable genes were identified for downstream dimensionality reduction. Principal component analysis (PCA) was performed, followed by Uniform Manifold Approximation and Projection (UMAP) for two-dimensional visualization. Cells were clustered using the FindNeighbors and FindClusters functions with a resolution parameter set to 0.5. Cell-type annotation was performed based on canonical marker genes and reference to published HNSCC datasets. Differential expression analysis between PPARG-pos and PPARG-neg epithelial cells was conducted using the FindMarkers function with the Wilcoxon rank-sum test. Genes with adjusted P < 0.05 and |log₂FC|≥ 0.25 were considered differentially expressed. Pathway enrichment analysis of upregulated genes in the PPARG + epithelial population was performed using the clusterProfiler package with KEGG and Hallmark gene sets. Violin plots and dot plots were generated using Seurat and ggplot2 for data visualization.

Immunohistochemistry (IHC)

Paraffin sections were then deparaffinized and rehydrated through a series of graded alcohols. Antigen retrieval was performed, followed by blocking, and then sections were incubated overnight with anti-PPARγ antibody (Abcam, USA). Next day, a universal goat anti-mouse/rabbit secondary antibody (PV9000, China) was applied at 37 °C for 1 h. Visualization was achieved using DAB substrate (ZLI9018, China), followed by hematoxylin counterstaining and a dehydration step. The slides were finally air-dried and coverslipped for analysis.

Multiplex immunohistochemical (mIHC)

mIHC was performed using the TSA® Plus Fluorescence Kit (TSA-RM-247259, PANOVUE) following the manufacturer’s protocol. Serial Sects. (4 μm) from FFPE OSCC tissue microarrays were deparaffinized, rehydrated through graded alcohols, and subjected to antigen retrieval using alkaline retrieval buffer (Tris-EDTA, pH 9.0), followed by incubation with blocking buffer for 30 min. Primary antibodies were sequentially applied and detected using HRP-conjugated secondary antibodies, followed by fluorophore-tyramide signal amplification. The antibodies used included: anti-CEBPA (1:500, Proteintech, China), anti-IL-17C (1:200, HUABIO, China), and anti-IL-17A (1:400, Proteintech, China). Between each staining cycle, slides were subjected to antibody stripping using microwave treatment. After the final round, nuclei were counterstained with DAPI.

CUT&RUN

CUT&RUN was utilized to detect CEBPA interaction with the IL-17C promoter in SCC-7 cells, based on the supplier’s guidelines (Cell Signaling Technology, USA). Briefly, cells were harvested and immobilized on Concanavalin A-coated magnetic beads. Following permeabilization, cells were incubated overnight at 4 °C with anti-CEBPA antibody (Proteintech, China) or normal rabbit IgG. The samples were then incubated with protein A-MNase for 2 h at 4 °C. Targeted chromatin digestion was initiated by adding calcium chloride (2 mM) and stopped after 30 min. DNA fragments were extracted using phenol–chloroform and ethanol precipitation. Then, qPCR was performed with primers specific for the IL-17C promoter region. Relative enrichment was calculated using the percent input method.

Statistical analysis

GraphPad Prism (version 8.0) was used for data processing. Statistical analyses were performed using unpaired Student’s t-test for comparisons between two groups, and one-way ANOVA was applied for evaluating differences among multiple groups. Pearson correlation coefficients were calculated to assess the strength of linear relationships. Overall survival was illustrated by Kaplan-Meier plots and evaluated via the Log-rank test. A P value less than 0.05 was considered statistically significant.

Results

Inhibiting PPARγ can suppress the progression of OSCC both in vitro and in vivo

We selected SCC-7 cells as the in vitro model, with primary epithelial cells from normal mice as controls. Western blot analysis revealed a marked upregulation of PPARγ in SCC-7 cells compared to control cells (Fig. 1A). To validate this in vivo, we established a murine OSCC model using 4-nitroquinoline 1-oxide (4NQO) administered via drinking water—a widely used method for inducing OSCC that closely mimics human disease progression (Fig. 1B). IHC assay results showed significantly higher levels of PPARγ in OSCC tissues (Fig. 1C, D). To determine whether specific inhibition of PPARγ could suppress OSCC progression, we treated SCC-7 cells with GW9662, a selective small-molecule antagonist of PPARγ. GW9662 could competitively bind to PPARγ and blocks its activity, thereby interfering with PPARγ-mediated signaling pathways. CCK-8 assays showed that suppressing PPARγ could markedly suppressed the proliferation of SCC-7 cells (Fig. 1E). To assess whether this effect extends in vivo, we established a subcutaneous tumor model in C3H/HeJ mice (Fig. 1F). On the 16th day after the inoculation of SCC-7 cells, we observed that inhibiting PPARγ activity significantly suppressed the growth of OSCC in mice (Fig. 1G), and no significant differences in body weight or apparent signs of systemic toxicity were observed, suggesting good in vivo tolerability (Fig. 1H, I). Furthermore, mice treated with GW9662 exhibited a better prognosis after tumor formation (Fig. 1J). These findings suggest that GW9662 effectively inhibits OSCC progression in vivo. In conclusion, the results support that inhibiting PPARγ can effectively suppress the onset and progression of OSCC.

Fig. 1
figure 1

Inhibition of PPARγ suppresses the progression of OSCC. A The expression of PPARγ protein between normal mouse tongue cells and SCC-7 cells. B Schematic illustration of 4NQO drinking water method to construct OSCC mouse model. C, D IHC assay showed the expression of PPARγ in mice normal and OSCC tissues (Scale bar = 100 µm.). E CCK-8 assay was used to assess the changes in the proliferation of SCC-7 cells treated with GW9662 (10 μM) for 0, 12, 24, 36, or 48 h. F Experimental design for in vivo tumor xenograft assay. G Tumor growth curve showing significantly reduced tumor volume in GW9662-treated mice compared with control group (N = 6/group). H H&E staining of lung, liver, spleen, kidney, and heart tissues harvested from tumor-bearing mice treated with either DMSO or GW9662 (Scale bars: 100 μm). I Mouse body weight monitoring during treatment of two groups (N = 6/group). J Kaplan–Meier survival analysis of mice bearing SCC-7 tumors treated with DMSO or GW9662 (N = 8/group). The data are presented as the means ± SDs from three or more independent experiments. Asterisks indicate significant differences (*P < 0.05, **P < 0.01, ***P < 0.001)

Full size image

PPARγ regulates the development of OSCC through modulating the IL-17 signaling

To uncover the molecular pathways underlying the anti-tumor effects of GW9662, we reanalyzed previously RNA-seq data from UM1 cells exposed to 10 μM GW9662 (Wei et al. 2022). Differential expression analysis revealed 236 upregulated and 471 downregulated genes in the GW9662-treated group compared to the DMSO control, including IL-17C, which is a key molecule in the IL-17 signaling pathway (Fig. 2A). KEGG pathway enrichment analysis of the downregulated genes demonstrated significant enrichment in immune-related signaling cascades, most notably the IL-17 pathway (Fig. 2B). GSEA further identified negative enrichment of immune-related functions upon PPARγ inhibition, including T cell chemotaxis (Fig. 2C), regulation of Th17-type immune responses (Fig. 2D), and neutrophil chemotaxis (Fig. 2E).

Fig. 2
figure 2

PPARγ regulates IL-17 signaling and Th17-associated immune responses in OSCC. A Volcano plot showing differentially expressed genes in UM1 cells treated with GW9662 (10 μM) versus DMSO control. A total of 236 genes were upregulated and 471 were downregulated upon PPARγ inhibition. Several key genes in the IL-17 signaling pathway, including IL-17C, MMP13, IL-1B, and CCL20, were markedly suppressed following PPARγ inhibition. B KEGG pathway enrichment analysis of downregulated genes after GW9662 treatment. C–E GSEA revealed that PPARγ inhibition negatively regulates immune-related processes, including T cell chemotaxis (C), regulation of Th17-type immune response (D), and neutrophil chemotaxis (E). F UMAP plot showing unsupervised clustering of 48,007 cells from the GSE172577 single-cell RNA-seq dataset, identifying major cell types in OSCC tumors. G Violin plot of PPARG expression across cell types, showing its preferential expression in epithelial cells. H Volcano plot of differentially expressed genes between PPARG-positive and PPARG-negative epithelial cells, highlighting upregulation of genes involved in inflammatory and metabolic processes. I Pathway enrichment analysis of genes upregulated in PPARG + epithelial cells, showing activation of PPAR signaling, IL-17 signaling, TGF-β pathway, and Th17 differentiation

Full size image

To validate these transcriptomic alterations at a single-cell level, we interrogated the publicly available scRNA-seq dataset GSE172577, comprising 48,007 cells. Unsupervised clustering and UMAP visualization delineated diverse cellular populations, including epithelium, keratinocytes, fibroblasts, myeloid cells, NKT cells, B cells, plasma cells, mural and endothelial cells (Fig. 2F). Notably, PPARG (the gene name of PPARγ) expression was predominantly confined to epithelial cells, as shown in the violin plot (Fig. 2G), suggesting cell-type-specific expression patterns and regulatory roles. To further explore the transcriptomic landscape associated with epithelial PPARG expression, we performed differential gene expression analysis between PPARG-positive and PPARG-negative epithelial cells. PPARG-positive epithelium displayed a significant upregulation of genes involved in lipid metabolism and epithelial differentiation, including KRT13, SPRR2, KRT4, and FABP5 (Fig. 2H). Pathway enrichment analysis of the upregulated genes in PPARG-positive cells revealed robust activation of the PPAR signaling pathway, as well as IL-17 signaling, TGF-β signaling, Th17 differentiation (Fig. 2I), suggesting a critical regulatory role of PPARγ in epithelial cells. These observations indicate that PPARγ actively modulates the IL-17 pathway, specifically IL-17C, and is likely to enhance the activation of Th17 cells.

PPARγ promotes Th17 cells differentiation via IL-17C regulation

Th17 cells have been associated with tumor progression by producing cytokines like IL-17A, which enhance angiogenesis, support tumor cell proliferation, and contribute to immunosuppressive TME (Saran et al. 2025; Li et al. 2024). To evaluate whether PPARγ facilitates the Th17 polarization, CD4 + T cells were harvested from murine spleens with the aid of anti-mouse CD4 magnetic beads. The purity of isolated cells was assessed by flow cytometry, confirming that CD4 + T cells constituted approximately 84% of the total cell population (Fig. 3A, B). To elucidate regulatory effect PPARγ in CD4 + T cell differentiation, the CD4 + T cells were co-cultured with SCC-7 cells in Th17-polarizing environment. Flow cytometric analysis revealed that pharmacological inhibition of PPARγ reduced the proportion of Th17 cells, indicating that PPARγ activity is essential for optimal Th17 differentiation (Fig. 3C–E). Consistently, ELISA assays of the co-culture supernatants demonstrated that GW9662-mediated inhibition of PPARγ significantly decreased the secretion of IL-17A, a signature cytokine of Th17 cells (Fig. 3F). The sequencing analysis showed that the expression level of IL-17 pathway-related genes, including IL-17C, was markedly downregulated. We examined the expression of IL-17C in SCC-7 cells following GW9662 treatment. Both qPCR and immunoblotting analyses confirmed that PPARγ inhibition significantly reduced IL-17C expression at both the mRNA and protein levels (Fig. 3G–H). To investigate the functional role of IL-17C in regulating Th17 polarization, CD4 + T cells were maintained under Th17-polarizing conditions and stimulated with recombinant IL-17C (rIL-17C), isotype IgG, or rIL-17C in combination with neutralizing anti-IL-17C antibody. Flow cytometry demonstrated a notable rise in Th17 cells frequency following rIL-17C treatment relative to the control group, and this enhancement was abrogated upon co-treatment with anti–IL-17C antibody, resulting in a marked reduction in Th17 frequency (Fig. 3I). Statistical analysis confirmed that rIL-17C significantly elevated Th17 cells percentages, while blockade of IL-17C reversed this effect (Fig. 3J). These findings imply that PPARγ plays a pivotal function in facilitating Th17 lineage commitment and cytokine production by regulating IL-17C, thereby potentially contributing to tumor-associated immune modulation in the OSCC microenvironment.

Fig. 3
figure 3

Inhibition of PPARγ activity suppresses Th17 cells differentiation in vitro. A Schematic illustration of the experimental workflow. CD4 + T cells were extracted from the mouse spleen. B Flow cytometry revealed that CD4 + T cells accounted for approximately 84% of the total cells. C Transwell co-culture system setup: CD4 + T cells were placed in the upper chamber, and SCC-7 cells were cultured in the lower chamber for 48 hours. D, E Flow cytometry plots showing the proportion of Th17 cells in the DMSO and GW9662-treated groups. F ELISA analysis showing reduced IL-17A levels in the supernatant of the GW9662 group compared to DMSO control. G qPCR analysis showing that Il-17c mRNA expression is significantly reduced following PPARγ inhibition with GW9662 in SCC-7 cells. H Western blot showing decreased protein levels of PPARγ and IL-17C after GW9662 treatment in SCC-7 cells. I Representative flow cytometry plots showing the proportion of Th17 cells under different treatment conditions: Control, rIL-17C, IgG, and rIL-17C combined with anti-IL-17C neutralizing antibody. J Quantification of Th17 cell percentages across the four groups. The data are presented as the means ± SDs from three or more independent experiments. Asterisks indicate significant differences (*P < 0.05, **P < 0.01, ***P < 0.001).

Full size image

PPARγ promotes IL-17C expression through transcriptional regulation by CEBPA

To further explore the transcriptional mechanism by which PPARγ modulates IL-17C expression, we performed an integrative analysis. Using the GeneCards database (https://www.genecards.org/), we identified 11 potential downstream targets of PPARγ, and intersected these with 291 predicted transcription factors of IL-17C (Table S1). Considering their intersection, cytosine-cytosine-adenosine-adenosine-adenosine-thymidine/enhancer binding protein α (CEBPA) was identified as the only overlapping protein (Fig. 4A). CEBPA, a transcription factor characterized by its basic leucine zipper (bZIP) domain, is known to participate in various tumor-related gene expression programs (Chen et al. 2022; Xie et al. 2021). Consistently, SCC-7 cells treated with GW9662 showed marked reductions in both Cebpa mRNA levels (qPCR) and CEBPA protein levels (Fig. 4B, C), supporting the hypothesis that PPARγ promotes IL-17C expression through transcriptional activation of CEBPA. To explore whether CEBPA could directly bind the IL-17C promoter, we used AlphaFold3 to predict CEBPA–IL-17C interactions (Fig. 4D), and JASPAR database analysis revealed a conserved CEBPA-binding motif within the IL-17C promoter region (Fig. 4E), further supporting the transcriptional regulation hypothesis. Whereafter, silencing CEBPA significantly reduces both IL-17C mRNA and protein expression, suggesting that CEBPA regulates IL-17C expression (Fig. 4F, G). To determine whether CEBPA directly binds to the promoter region of IL-17C, we performed CUT&RUN assays. qPCR analysis of CUT&RUN-enriched DNA showed that CEBPA was significantly enriched at the predicted binding site compared to the IgG control (Fig. 4H), indicating a robust and specific interaction. Consistent with this, PCR further validated the presence of the target genomic region in the CEBPA pull-down, whereas no amplification was detected in the IgG group (Fig. 4I). Importantly, to examine the functional relevance of CEBPA in Th17 polarization, we assessed the percentage of Th17 cells following CEBPA knockdown. Flow cytometry revealed a marked decrease in the percentage of Th17 cells in the siCEBPA group in contrast to the siNC controls (Fig. 4J, K). Together, these findings reveal that PPARγ promotes IL-17C expression by transcriptionally activating CEBPA, thereby establishing a PPARγ–CEBPA–IL-17C regulatory axis that may contribute to Th17 cell differentiation and tumor-associated inflammation in OSCC.

Fig. 4
figure 4

PPARγ promotes IL-17C expression via CEBPA in OSCC. A Venn diagram showing the intersection between predicted downstream effectors of PPARγ and transcription factors of IL-17C, identifying CEBPA as a common candidate. B qPCR validation showing that Cebpa and Il-17c expression is significantly reduced in SCC-7 cells following GW9662 treatment. C Western blot confirming downregulation of CEBPA protein levels upon PPARγ inhibition. D AlphaFold3-predicted 3D structure showing the binding of CEBPA to the IL-17C promoter region. (ipTM = 0.18; pTM = 0.21) . E JASPAR motif analysis identifying a conserved CEBPA-binding motif (CAAT box) in the IL-17C promoter. F Quantification of Il-17c mRNA levels in siNC and siCEBPA-treated cells in SCC-7 cells. G Western blot showing reduced CEBPA and IL-17C protein levels in siCEBPA-treated cells compared to siNC in SCC-7 cells. H CUT&RUN-qPCR analysis showing significant enrichment of CEBPA binding at the predicted IL-17C promoter region compared to IgG control. I Agarose gel electrophoresis confirming the presence of IL-17C promoter DNA in the CEBPA pull-down, while the IgG group showed no detectable band. J Representative flow cytometry plots showing the proportion of Th17 cells following CEBPA knockdown (siCEBPA) versus negative control (siNC). K Quantification of Th17 cell percentages. CEBPA silencing significantly reduced the proportion of Th17 cells compared to siNC. The data are presented as the means ± SDs from three or more independent experiments. Asterisks indicate significant differences (*P < 0.05, **P < 0.01, ****P < 0.0001).

Full size image

Targeting PPARγ inhibits Th17 polarization through a CEBPA-dependent IL-17C/IL-17A pathway in vivo

To recapitulate the function of PPARγ in regulating Th17 cells differentiation within the TME of OSCC, we established a subcutaneous tumor model using C3H/HeJ mice. In the experimental group, SCC-7 cells pretreated with GW9662 for 24 h were subcutaneously inoculated. Following tumor establishment, GW9662 was administered via intraperitoneal injection every three days. On day 12, tumor tissues were harvested, and performed flow cytometry analysis (Fig. 5A). Flow cytometric results demonstrated that inhibition of PPARγ activity significantly inhibited the polarization of CD4 + T cells toward the Th17 lineage within OSCC tumors (Fig. 5B, C). In parallel, tumor tissues were collected for gene and protein expression analyses. Consistent with in vitro findings, suppression of PPARγ markedly decreased the expression of CEBPA, IL-17C and IL-17A at both mRNA and protein levels (Fig. 5D, E). To further validate these findings in a more physiologically relevant model, we examined the expression patterns of IL-17C, CEBPA, and IL-17A in 4NQO-induced OSCC tissues, which more closely mimic the native oral tumor microenvironment. Multiplex immunofluorescence (mIHC) staining revealed significantly elevated levels of all three proteins in OSCC lesions compared with normal oral mucosa (Fig. 5F, G). These findings collectively indicate that PPARγ inhibition in vivo suppresses Th17 cells differentiation by downregulating CEBPA and IL-17C, thereby disrupting the immunoregulatory axis within the TME in OSCC.

Fig. 5
figure 5

Inhibition of PPARγ activity suppresses Th17 cells differentiation in vivo. A Schematic diagram of the mouse subcutaneous xenograft tumor model. The experimental group was treated with GW9662 (3 mg/kg). Control group treatment: DMSO (N = 3/group). B, C Flow cytometry analysis of the percentage of CD4+ IL-17A+ cells in mouse tumors. D qPCR revealed that the mRNA levels of Cebpa, Il-17c and Il-17a were decreased after PPARγ activity was inhibited in OSCC tumors. E WB analysis revealed that the protein levels of CEBPA, IL-17C and IL-17A were decreased after PPARγ activity was inhibited in OSCC tumors. F mIHC staining of IL-17C, CEBPA, and IL-17A (red) in normal oral mucosa and 4NQO-induced OSCC tissues. DAPI (blue) marks nuclei. (Scale bar: 20 μm). G Quantification of fluorescence intensity per area revealed significantly elevated expression of IL-17C, CEBPA, and IL-17A in OSCC tissues compared to normal mucosa. The data are presented as the means ± SDs from three or more independent experiments. Asterisks indicate significant differences (**P < 0.01, ****P < 0.0001, ***P < 0.001).

Full size image

PPARγ/CEBPA/IL-17C axis correlates with survival in OSCC

Building on our prior study which found that PPARγ is markedly upregulated in OSCC and negatively associated with overall survival (Wei et al. 2022), we further performed mIHC on 104 paired OSCC tumor specimens to evaluate the spatial distribution and clinical relevance of downstream effectors CEBPA, IL-17C, and IL-17A. The results revealed that IL-17C, CEBPA, and IL-17A were highly expressed in OSCC tissues (Fig. 6A). Quantitative correlation analyses further revealed strong positive associations between PPARγ levels and the expression of IL-17C, CEBPA, and IL-17A (Fig. 6B-D). In addition, IL-17C levels were tightly correlated with both CEBPA and IL-17A expression, reinforcing the existence of a coordinated regulatory cascade (Fig. 6E, F). Kaplan-Meier survival analyses demonstrated elevated levels of IL-17C, CEBPA, and IL-17A was each significantly associated with poor overall survival (Fig. 6G-I). Together, these results provide robust tissue-level validation for the PPARγ/CEBPA/IL-17C/IL-17A axis in OSCC, and highlight its potential prognostic and functional relevance in the tumor immune microenvironment.

Fig. 6
figure 6

PPARγ/CEBPA/IL-17C axis is upregulated in OSCC and correlates with poor patient prognosis. A mIHC staining of 104 serial OSCC sections showing co-expression of IL-17C (cyan), CEBPA (red), and IL-17A (yellow). Representative high and low expression cases are shown. Nuclei were counterstained with DAPI (blue). (Scale bars: 50 μm). B-F Correlation analyses between PPARγ and IL-17C (B), PPARγ and CEBPA (C), PPARγ and IL-17A (D), IL-17C and CEBPA (E), and IL-17C and IL-17A (F) in 104 OSCC clinical specimens. G-I Kaplan-Meier survival curves showing overall survival of OSCC patients stratified by IL-17C (G), CEBPA (H), and IL-17A (I) expression levels. High expression of each marker was significantly associated with poorer survival outcomes.

Full size image

Discussion

OSCC represents the predominant epithelial malignancy affecting the oral and maxillofacial region, comprising approximately 95% of HNSCC, and is prone to recurrence and metastasis. Despite advances in multimodal therapy, OSCC remains prone to recurrence, regional lymph node invasion, and distant metastasis, with unsatisfactory long-term survival rates (Alsaeedi and Aggarwal 2022). PPARγ, a member of the nuclear receptor family that plays multifaceted roles in metabolism, cell proliferation and differentiation, autoimmune diseases, and cancer, is markedly upregulated in OSCC and negatively correlates with patient prognosis (Wei et al. 2022). Although PPARγ has been implicated in tumor biology of OSCC, its contribution to shaping the immune contexture in OSCC has yet to be fully clarified.

Although the role of Th17 cells in tumor progression remains controversial, recent studies have consistently highlighted their pro-tumorigenic effects. Th17 cells represent a subset of CD4 + T helper cells recognized for their secretion of IL-17A. Th17 cells can secrete factors, like IL-17A, IL-22, IL-23 to promote vascular survival, recruit neutrophils, and activate immune-suppressive activities in the TME to facilitate tumor progression (Salazar et al 2020; Knochelmann et al 2018; Mills 2023; Perez et al 2020). However, studies on Th17 cells in the OSCC immune microenvironment have primarily focused on measuring cytokine levels in patient serum and analyzing the distribution of this subset in tissues, with few investigations thoroughly exploring the pro-tumorigenic mechanisms of this cell population at the microenvironmental level (Wang et al. 2020; Shan et al. 2020). Our study reveals that aberrant overexpression of PPARγ in OSCC promotes Th17 cells differentiation, remodels the immune microenvironment, and drives OSCC progression, suggesting that PPARγ may exert its oncogenic function in OSCC by promoting Th17-mediated inflammation. In addition, Recent evidence highlights the potential of PPARγ antagonism as a strategy to reinforce tumor-targeting immunity and improve the efficacy of immunotherapies. For instance, the PPARγ inhibitor GW9662 was found to suppress adipogenesis-driven PD-L1 expression in adipocytes, a mechanism that otherwise compromises anti-PD-1/PD-L1 antibody activity. In murine mammary tumor models, GW9662 significantly enhanced responses to immune checkpoint blockade, accompanied by increased infiltration and activation of intratumoral CD8 + T cells (Wu et al. 2018). Similarly, another study demonstrated that PPARγ inhibition synergized with cytokine-induced killer (CIK) cell therapy, markedly improving tumor cell lysis in glioblastoma and neuroblastoma models. These findings underscore the immunomodulatory capacity of PPARγ inhibition and its compatibility with both checkpoint blockade and adoptive cellular immunotherapy (Vordermark et al. 2024). Consistent with these reports, our study identifies a novel immunosuppressive role for PPARγ via promotion of IL-17C-mediated immune evasion. This mechanistic insight further supports the rationale for targeting PPARγ as a combinatorial approach to overcome tumor-driven immunosuppression and reprogram the tumor immune microenvironment toward a more immunoreactive state.

CEBPA is a bZIP transcription factor participated in an array of physiological functions, including myeloid cell differentiation, immune cell development, metabolism, and inflammatory signaling (Kim et al 2024; Pundhir et al 2018; Nakagawa et al 2024). Research has demonstrated that CEBPA facilitates breast tumorigenesis through transcriptional control of cytokine signaling 2 (SOCS2) (Wang et al. 2025). Mechanistically, we observed CEBPA as a key transcription factor downstream of PPARγ that mediates Th17 cell differentiation. Our multi-omics integrative analysis, together with experimental validation (qPCR, immunoblotting, AlphaFold3 modeling and cut&RUN assay), revealed that PPARγ promotes CEBPA expression, which activates IL-17C transcription. IL-17C subsequently promotes Th17 differentiation. Importantly, we demonstrated both in vitro and in vivo that PPARγ inhibition significantly reduces Th17 cells polarization, IL-17A secretion, and expression of the PPARγ/CEBPA/IL-17C axis in OSCC. This finding introduces a previously unrecognized signaling cascade by which tumor-intrinsic PPARγ activity reprograms the tumor immune microenvironment in favor of Th17-driven immunosuppression and tumor progression. From a translational perspective, targeting PPARγ represents a promising strategy to disrupt Th17-mediated immunosuppression and may complement existing immunotherapies.

Although our results, together with prior studies, support the immunomodulatory potential of PPARγ inhibition, the translational applicability of GW9662 remains to be fully established. Future work is needed to evaluate its safety profile in a dose-dependent manner, its pharmacokinetics, and its therapeutic synergy with immune checkpoint inhibitors (such as PD-L1/PD-1) in clinically relevant OSCC models.

Conclusions

In summary, this study reveals a novel immunoregulatory mechanism in OSCC whereby PPARγ promotes Th17 cells differentiation via PPARγ/CEBPA/IL-17C. Disruption of this signaling axis through PPARγ inhibition offers a compelling strategy to reshape the tumor immune microenvironment and restrain malignant progression. These findings provide key mechanistic insight into the oncogenic role of PPARγ and suggest new avenues for the development of immunomodulatory therapies targeting tumor-immune interactions in OSCC.