BTN3A1 regulate the function of Vγ9Vδ2 T cells through TCR signaling pathway in ovarian cancer Authors: Jian Liu, Yuqin Wang, Min Wu, Hui Wang Published: 2026 ABSTRACT BackgroundAdoptive cell transfer (ACT) therapy shows great promise in ovarian cancer. In addition to the classic αβ T cells, γδ T cells have become a very important anti-tumor weapon in tumor immunotherapy, especially in hematological tumors, due to their high in vitro expansion efficiency and high tumor killing activity. However, in solid tumors, especially cold tumors such as ovarian cancer, the anti-tumor effect of γδ T cells and the key factors affecting and regulating the anti-tumor ability of γδ T cells are still unclear.MethodThe functional characteristics of Vγ9Vδ2 T cells in ovarian cancer, along with the molecules and pathways mediating their interaction with tumor cells, were preliminarily explored. This was done using databases such as The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO), in addition to RNA sequencing (RNA-seq). Furthermore, Vγ9Vδ2 T cells were co-cultured with ovarian cancer tumor cell lines to validate the molecules and pathways affecting Vγ9Vδ2 T cell function through flow cytometry.ResultsVγ9Vδ2 T cells in ovarian cancer have strong anti-tumor properties. High expression of BTN3A1 in tumor cells reduces the activation of expanded Vγ9Vδ2 T cells, increases expression of exhaustion related molecules, and raises apoptosis levels in these cells. Blocking the T cell receptor (TCR) can eliminate the effect of BTN3A1 on exhaustion of Vγ9Vδ2 T cells.ConclusionBTN3A1 activates Vγ9Vδ2 T cells but also induces their exhaustion through the TCR signaling pathway. BTN3A1 may serve as a regulator of activation-induced exhaustion in γδ T cells. Targeting downstream related molecules of BTN3A1 is of great significance for enhancing adoptive Vγ9Vδ2 T cells transfer therapy.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s13048-026-02180-0. Background Since the mid-1970s, the mortality rate of ovarian cancer (OC) has decreased by over 30%. However, less than half of women survive for more than 5 years after diagnosis, partly because invasive high-grade serous carcinoma is the predominant subtype [1]. The American Cancer Society estimates that the case fatality rate of invasive OC in 2025 will reach 60.9%, accounting for 4% of all cancer deaths in women. Moreover, the 5-year relative survival rate of advanced metastatic OC is only 31% [2]. Currently, OC treatment primarily involves surgery, radiotherapy, chemotherapy, and targeted therapy. Exploring more effective therapies is crucial to improving prognosis of OC [3]. The presence of tumor-infiltrating lymphocytes (TILs), particularly T cells, is associated with improved progression-free survival and overall survival in patients with OC [4]. Adoptive T-cell therapy has been demonstrated to be a promising therapeutic approach for OC [5]. ACT using genetically modified T cells has been evaluated in OC, including TCR engineered T cells and chimeric antigen receptor (CAR) modified T cells. The former can recognize peptides presented by human leukocyte antigen (HLA) molecules on the cell surface, while the latter can recognize antigens on the surface of cancer cells [3]. However, most CAR-T research on OC is still in the preclinical stage, and their therapeutic function remains limited, mainly due to the lack of qualified target antigens in CAR-T therapy for solid tumors [6]. In addition to conventional αβ T cells, γδ T cells represent another subset with remarkable antitumor activity. Notably, their cytotoxic effects are independent of neoantigen presentation and traditional major histocompatibility complex (MHC)-dependent antigen presentation, thereby partially compensating for the above limitations of αβ T cells. Beyond their direct tumor-killing capacity, γδ T cells have been shown to coordinate both innate and adaptive immune responses, such as promoting the activation and function of NK cells, inducing dendritic cell (DC) maturation, and activating CD4⁺ and CD8⁺ T cells through antigen presentation [7, 8]. Moreover, their combination of typical characteristics of T cells and natural killer (NK) cells, as well as their ease of expansion in vitro, makes them very attractive for adoptive transfer therapy in tumor [9]. Adoptive transfer of γδ T cells has shown promising results in the treatment of hematological malignancies. However, its application in solid tumors is still limited due to lack of functional persistence, an exhaustion phenotype, and impaired during transport to the tumor microenvironment (TME) [10]. Notably, the infiltration of γδ T cells in tumors correlates with the duration of disease-free interval in patients with advanced serous OC [11]. Specifically, Vδ2 T cells, a key subset of γδ T cells, have been reported to not only inhibit the growth of ovarian tumor cells but also completely eradicate pre-existing tumors with an initial size of less than 5 mm [12]. In addition, Vδ2 T cells are highly effective in killing human ovarian cancer stem cells, reducing their proliferation rate to approximately 40% of the original level, and enhancing the sensitivity of tumor cells to chemotherapeutic agents [5]. Although γδ T cells are abundant in ovarian cancer tissues and can be recruited by tumor-derived factors, their secretion of IFN-γ is reduced and their cytotoxic activity is partially impaired [13]. Moreover, γδ T cells in malignant ascites and tumor tissues of OC have been found to specifically express inhibitory receptors such as TIGIT, PD-1, and TIM-3 [14]. These findings suggest that γδ T cells in OC are likely to be in an exhausted state with suppressed antitumor functions; however, the underlying mechanisms remain largely unclear. Therefore, exploring the exhaustion mechanism of γδ T cells in OC is of great significance for improving ACT therapy. This study preliminarily explored the mechanisms underlying Vδ2 T-cell exhaustion following contact with tumor cells. The activation molecule BTN3A1 was found to influence the effector functions of Vδ2 T cells. Interestingly, blocking the T-cell receptor (TCR) was shown to abolish the effect of BTN3A1 on the exhaustion of Vγ9Vδ2 T cells. These findings provide important clues for further improving the efficacy of ACT. Materials and methods Cell lines and cell culture Ovarian cancer cell line OVCAR8, ES-2, A2780, SKOV3 were cultivated in RPMI DMEM medium supplemented with 10% fetal bovine serum (FBS) (Gibco, Life Technologies). The BTN3A1 plasmid package with lentivirus (pRP[CRISPR]-EGFP/Puro-hCas9-U6 > hBTN3A1[gRNA#839]) was purchased from Yunzhou Biotechnology (Guangzhou, China) and transduced into the OVCAR8 cell line for 24 h. The cells were screened with 2 µg/ml puromycin (Selleck) and the surviving tumor cells were amplified for subsequent experiments (Supplement Fig. 1). For the amplification of Vδ2 T cells, PBMCs were cultured in X-ViVo medium supplemented with 10% FBS. A working concentration of 5µM zoledronic acid salt (Sigma) and 1000 IU/ml recombinant human IL-2 (GeneScript) were added to the culture medium to stimulate Vδ2 T cell expansion. On the 12th day of cultivation, a certain number of cultured cells were harvested and co-cultured with tumor cells to detect the exhaustion, activation, killing, and anti-tumor potential of Vδ2 T cells. Flow cytometry A pipette was used to aspirate a certain amount of tumor tissue or co-cultured single-cell suspension into a flow cytometer, and the total number of cells was adjusted to be consistent in each sample. 3 ml of flow cytometry buffer was added, followed by centrifugation at 1500 rpm at 4 °C for 5 min. The supernatant was discarded, and the cells was incubated with the active dyes FVS (BD) and Fc block (B313423, Biolegend) for 15 min each. The cells were then mixed with fluorescein-labeled antibodies and incubated at 4 °C for 30 min. For intracellular staining, cells were fixed and permeabilized with the Cytofix/Cytoperm Fixation/Permeabilization Kit (554714, BD) after surface protein staining. For nuclear staining, cells were fixed and nuclear-permeabilized using Transcription Factor Buffer Set (562574, BD) after surface protein staining, followed by labeling of intracellular proteins with fluorescein labeled antibodies. For detecting the apoptosis level of Vδ2 T cells after co-culturing, Annexin V-PE/7-AAD apoptosis detection kit (559763, BD) staining was applied to detect the apoptosis level of the Vδ2 T cells after surface staining. The specific staining antibody included CD3 (SK7, BD), Vδ2 (REA771, Miltenyl), γδ TCR (B1, Biolegend), PD-1 (NAT105, Biolegend), TIM3 (F38-212, Biolegend), LAG3 (11C3C65, Biolegend), NR4A2 (F-5, SCBT), NR4A3 (H-7, SCBT), IFN-γ (4 S.B3, BD), TNF-α (MAb11, BD), Perforin (dG9, Biolegend), Granzyme B (GB11, BD), PD-L1 (MIH1,BD), HLA-DR (G46-6, BD), CD86 (IT2.2, Biolegend). Western blot RIPA Lysis Buffer (G2002, Servicebio, China) was used to separate proteins from cell lysate, and the proteins were separated by electrophoresis. They were then transferred onto a PVDF membrane and sealed in 5% skim milk for 1 h. The membrane was incubated with specific antibody overnight at 4 °C. The antibodies used for WB detection: BTN3A1 pAb (25221-1-AP, Proteintech), ZAP70 mAb (#2705S, CST), p-ZAP70 mAb (#2717S, CST), CD3ζ mAb (#88083S, CST), p-CD3ζ mAb (#67748S, CST), NFATc1 (D43B1, CST), HRP Goat anti-Rabbit IgG (AS014, ABclonal), and HRP Goat anti-Mouse IgG (AS003, ABclonal). The membrane was then incubated with secondary antibodies for 1 h at 25 °C. Protein expression levels were observed using Western Bright ECL HRP substrate (K-12045-D50, Advansta) and a chemiluminescence imaging system (Bio-Rad). Colony formation A total of 250/500 cells were seeded in a six well plate and cultured in DMEM containing 10% FBS for 2 weeks. The formed clones were fixed with 4% paraformaldehyde and stained with crystal violet before being photographed. Bioinformatics analysis RNA-seq data processed by the STAR pipeline for the OV (ovarian serous cystadenocarcinoma) project were downloaded and organized from the TCGA database (https://portal.gdc.cancer.gov), GEO database (https://www.ncbi.nlm.nih.gov/geo/) and ICGC database (https://dcc.icgc.org). TPM format data were extracted and further converted to log2 format. Correlation analysis of the main variables (TRDV2, TRGV9, BTN3A1) in the data was performed using the Spearman statistical method with R (4.2.1). Based on the ssGSEA algorithm provided in the R package (GSVA [1.46.0]) [15], the immune infiltration status corresponding to the dataset was calculated using markers of 24 immune cells provided by Bindea et al. [16]. The matrix and immune scores of the corresponding dataset were calculated through the R package estimate [1.0.13] [17]. The analysis results were visualized using the R package ggplot2 [3.4.4]. In order to analyze the effect of BTN3A1 on OC, TCGA data were divided into high and low expression groups according to the expression of BTN3A1. The DESeq2 package was used to perform differential analysis on the raw counts-matrix [18]. Genes with logfc > 1.0 were selected for enrichment analysis using the cluster Profiler [4.4.4] package [19], and the Zscore value corresponding to each enrichment entry was calculated using the GOplot [1.0.2] package [20]. The analysis results were visualized using the R package ggplot2 [3.4.4]. GSEA analysis and immunomodulator analysis of DEGs related to BTN3A1 were performed using the BEST database [21]. Correlation analysis between the abundance of γδ T cells in OC patients in the TCGA database and their cumulative survival rate were performed using the TIMER 3.0 analysis [22]. RNA sequencing and analysis Vδ2 T cells were purified after co-culturing with tumor cells. After extracting total RNA, an mRNA library was established using the Illumina technology platform. RNA-seq and quality control were performed on the BGI-SEQ platform to obtain the corresponding sequencing data. Combined fold change (fc) analysis of Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) was performed using the clusterProfiler [4.4.4] package and GOplot [1.0.2] package. ID conversion of molecules in the input data was performed using the MSigDB database, and Gene Set Enrichment Analysis (GSEA) was conducted using the clusterProfiler package [23]. The analysis results were visualized using the R package (ggplot2 [3.4.4]). Statistical analysis Flow staining data were analyzed and plotted using Flow Jo 10.4 software, and data was analyzed and plotted by Graphpad software. Data in the figure were presented as mean ± standard deviation or standard error. Two-tailed Student’s t-test was used for statistical analysis. P < 0.05 was considered statistically significant, and statistical differences were indicated by asterisks: ****P < 0.0001, ***P < 0.001, **P < 0.01, *P < 0.05). Results Vδ2 T cells exhibit anti-tumor properties in ovarian cancer Gentle et al. found in their study based on 39 types of tumors and approximately 18,000 samples that γδ T cells are the subpopulation most strongly correlated with benign prognosis in patients [24]. Moreover, γδ T cells abundantly infiltrate OC tumor tissue, and the proportion of reactive Vδ2 T cells is positively correlated with the overall survival rate and residual tumor burden of patients [25]. Based on these reports, we first investigated the association between Vγ9Vδ2 T cell infiltration and activation of various immune cells in OC. In the analysis based on the ESTIMATE algorithm, which assesses ImmunoScore and StromalScore [26], we found that the representative markers of Vγ9Vδ2 T cells (TRDV2 and TRGV9) were positively correlated with the infiltration levels of immune cells in the TME (Fig. 1A, B). Specifically, both TRDV2 and TRGV9 in ovarian cancer tissues were strongly associated with the infiltration of multiple anti-tumor immune cells, including total T cells, central memory T cells, effector memory T cells, cytotoxic T cells, helper T cells, and NK cells (Fig. 1C, D). These results suggest that Vγ9Vδ2 T cells are likely to play an important regulatory role in the infiltration, activation, and anti-tumor immune response of immune cells in OC. Although previous studies have reported that in vitro-expanded Vγ9Vδ2 T cells can effectively kill ovarian cancer cells (such as HTB75, OVCAR3, SKOV3, A2780, and TOV112D) [12], the effects of ovarian cancer cells on Vγ9Vδ2 T cells remain unclear. The activation level and functional state of Vγ9Vδ2 T cells, which are regulated by various signals upon entering tumor tissues, have been identified as key factors influencing their immune regulatory functions. Based on this, we aimed to further investigate potential changes in the functional status of Vγ9Vδ2 T cells in the context of OC. To achieve this, we expanded Vγ9Vδ2 T cells from normal human peripheral blood in vitro, co-cultured them with the ovarian cancer cell line OVCAR8, and employed RNA sequencing to explore the changes in these T cells. We found that, compared to the control group, Vγ9Vδ2 T cells co-cultured with ovarian cancer cells showed higher expression of anti-tumor T cell activation-related genes, including TNFRSF9, IFNG, LTA, and IL2RA (Fig. 1E), which encode 4-1BB, IFN-γ, TNF-β, and IL-2 receptor proteins, respectively. GO and KEGG enrichment analysis revealed that genes highly expressed in Vγ9Vδ2 T cells after co-culture with ovarian cancer cells were mainly enriched in multiple immune response-related pathways, such as regulation of T cell activation, cytokine-mediated signaling pathway, signaling receptor activator activity (Fig. 1F), cytokine-cytokine receptor interaction, TNF signaling pathway, and calcium signaling pathway (Fig. 1G). However, it is noteworthy that we also observed significant upregulation of genes associated with immune cell exhaustion and functional impairment, such as PD-1, TIM3, and TIGIT in Vγ9Vδ2 T cells after co-culture with ovarian cancer cells (Fig. 1H). Therefore, these results suggest that while ovarian cancer cells activate Vγ9Vδ2 T cells, they may also induce changes in their functional state, including immune exhaustion. Fig. 1Vδ2T cells in ovarian cancer tissue have anti-tumor function. A, B Use Estimate algorithm [1.0.13] and spearman statistics to analyze the correlation between TRGV9 (C) and TRDV2 (D) expression levels and ImmunoScore, ESTIMATEScore and StromalScoreindices in ovarian tumors in the TCGA dataset. C, D Use ssGSEA algorithm and Spearman statistics to analyze the correlation between the of TRGV9 (A) and TRDV2 (B) expression levels and the infiltration of various immune cell subsets in ovarian cancer tissues in the TCGA dataset. E Volcano plot of differentially expressed genes between Vδ2 T cells co-cultured with OVCAR8 cells for 24 hours (E: T = 1:1) and amplified Vδ2 T cells. F, G GO enrichment analysis (F) and KEGG enrichment analysis (G) of differentially expressed genes between Vδ2 T cells co-cultured with OVCAR8 cells for 24 hours (E: T=1:1) and amplified Vδ2 T cells. BP: Biological Process; CC: Cellular Component; MF: Molecular Function. H mRNA expression levels of PDCD1, HAVCR2, TIGHT in Vδ2 T cells and after co-culture with OVCAR8 tumor cell lines (E: T = 1: 1) for 24 hours Vδ2T cells in ovarian cancer tissue have anti-tumor function. A, B Use Estimate algorithm [1.0.13] and spearman statistics to analyze the correlation between TRGV9 (C) and TRDV2 (D) expression levels and ImmunoScore, ESTIMATEScore and StromalScoreindices in ovarian tumors in the TCGA dataset. C, D Use ssGSEA algorithm and Spearman statistics to analyze the correlation between the of TRGV9 (A) and TRDV2 (B) expression levels and the infiltration of various immune cell subsets in ovarian cancer tissues in the TCGA dataset. E Volcano plot of differentially expressed genes between Vδ2 T cells co-cultured with OVCAR8 cells for 24 hours (E: T = 1:1) and amplified Vδ2 T cells. F, G GO enrichment analysis (F) and KEGG enrichment analysis (G) of differentially expressed genes between Vδ2 T cells co-cultured with OVCAR8 cells for 24 hours (E: T=1:1) and amplified Vδ2 T cells. BP: Biological Process; CC: Cellular Component; MF: Molecular Function. H mRNA expression levels of PDCD1, HAVCR2, TIGHT in Vδ2 T cells and after co-culture with OVCAR8 tumor cell lines (E: T = 1: 1) for 24 hours BTN3A1 is a candidate for regulating Vδ2 T cell immune function To further evaluate the functional changes of Vδ2 T cells in the context of adoptive transfer, we expanded Vδ2 T cells in vitro and the cell purity reached around 95% on Day 12 (Fig. 2A, B). Subsequently performed co-culture assays with tumor cells to assess their functional activity. Flow cytometry analysis revealed that Vδ2 T cells can be activated by tumor cells, secreting activating factors (Fig. 2C), which is consistent with previous research. But interestingly, the expression level of exhausted molecules was also upregulated (Fig. 2D). This prompts us to further contemplate the possible mechanisms involved. Fig. 2Functional changes of Vδ2 T cells expanded in vitro after contact with tumor cells. A, B The proportion of Vδ2+T cells in all living cells (A) and the MFI of Vδ2 expression (B) during the expansion of Vγ9Vδ2 T cells (Day0, Day8, Day11, Day14) from healthy PBMCs in vitro. C, D Vδ2 T cells cultured alone or co cultured with OVCAR8 cells (E: T = 1: 1) for 24 h. The percentage of IFN-γ+ TNF-α+ cells (C) and PD-1+ TIM3+ cells (D) in Vδ2 T cells was detected by flow cytometry Functional changes of Vδ2 T cells expanded in vitro after contact with tumor cells. A, B The proportion of Vδ2+T cells in all living cells (A) and the MFI of Vδ2 expression (B) during the expansion of Vγ9Vδ2 T cells (Day0, Day8, Day11, Day14) from healthy PBMCs in vitro. C, D Vδ2 T cells cultured alone or co cultured with OVCAR8 cells (E: T = 1: 1) for 24 h. The percentage of IFN-γ+ TNF-α+ cells (C) and PD-1+ TIM3+ cells (D) in Vδ2 T cells was detected by flow cytometry Numerous studies have shown that Vδ2 T cells bind to BTN3A1 through the γδ TCR and induce conformation changs in the intracellular B30.2 domain of BTN3A1 [27]. Together, they sense the specific accumulation of phosphoantigens (pAgs) in tumor cells, which leading to rapid activation and proliferation of Vδ2 T cells, secretion of inflammatory cytokines, and targeted killing of tumor cells [28, 29]. Therefore, BTN3A1 derived from tumor cells may regulate Vδ2 T cells in OC. We divided ovarian cancer tissues into high and low expression groups of BTN3A1 and analyzed the DEGs between the two groups. GSEA-Hallmark analysis shows that BTN3A1 regulates pathways such as interferon gamma response, inflammatory response (Fig. 3A). GO analysis is mainly enriched in modules such as immune response, immune system process and MHC protein complex (Fig. 3B). Further GO combined with FC analysis (log2fc > 1) revealed that DEGs were enriched in immune modules such as lymphocyte mediated immunity, regulation of T cell activation et al. (Fig. 3C). All of the above indicate the important role of BTN3A1 in immune regulation of OC. Fig. 3BTN3A1 regulates anti-tumor related immune pathways in ovarian cancer. A, B GSEA-Hallmark analysis (A) and GO enrichment analysis (B) of the differentially expressed genes between high and low BTN3A1 expression groups in TCGA dataset. C GO combined with FC (logfc>1) enrichment analysis of differentially expressed genes between high and low BTN3A1 expression groups in the TCGA dataset. D Using the Estimate algorithm [1.0.13] and Spearman statistics, analyze the correlation between BTN3A1 expression levels and ImmunoScore, ESTIMATESCore, and StromalScore indices in ovarian tumors in the TCGA dataset. E Pearson correlation analysis was applied to identify molecules in the TCGA dataset that are highly correlated with BTN3A1 expression. Genes with R>0.4 were screened for GO and KEGG enrichment analysis. F Correlation heatmap between immune module related genes and BTN3A1 in ovarian cancer tissue sequencing data from ICGC, TCGA, GEO databases. G The protein expression of BTN3A1 in ovarian cancer cell lines OVCAR8, ES-2, A2780 and SKOV3 and the constructed BTN3A1 knockout cell lines were detected by WB. H Place 250 and 500 OVCAR8 cells in a six well plate, respectively, and count the number of clones formed after 2 weeks (>50 cells/clone) BTN3A1 regulates anti-tumor related immune pathways in ovarian cancer. A, B GSEA-Hallmark analysis (A) and GO enrichment analysis (B) of the differentially expressed genes between high and low BTN3A1 expression groups in TCGA dataset. C GO combined with FC (logfc>1) enrichment analysis of differentially expressed genes between high and low BTN3A1 expression groups in the TCGA dataset. D Using the Estimate algorithm [1.0.13] and Spearman statistics, analyze the correlation between BTN3A1 expression levels and ImmunoScore, ESTIMATESCore, and StromalScore indices in ovarian tumors in the TCGA dataset. E Pearson correlation analysis was applied to identify molecules in the TCGA dataset that are highly correlated with BTN3A1 expression. Genes with R>0.4 were screened for GO and KEGG enrichment analysis. F Correlation heatmap between immune module related genes and BTN3A1 in ovarian cancer tissue sequencing data from ICGC, TCGA, GEO databases. G The protein expression of BTN3A1 in ovarian cancer cell lines OVCAR8, ES-2, A2780 and SKOV3 and the constructed BTN3A1 knockout cell lines were detected by WB. H Place 250 and 500 OVCAR8 cells in a six well plate, respectively, and count the number of clones formed after 2 weeks (>50 cells/clone) ImmuneScore analysis based on TCGA ovarian cancer transcriptomic data revealed a positive correlation between BTN3A1 expression and computationally inferred immune cell infiltration in OC (Fig. 3D). We conducted enrichment analysis on genes strongly correlated with BTN3A1 (R > 0.4) in OC, which are concentrated in immune activation related pathway such as immune response-regulating signaling pathway, antigen processing and presentation, positive regulation of cytokine production (Fig. 3E). Furthermore, we analyzed the correlation between different immune module related genes and BTN3A1 in ovarian cancer tissue sequencing data from different databases (ICGC, TCGA, GEO) (Fig. 3F). BTN3A1 is positively correlated with various immune activating molecules (TNFSF13, CXCL12, TNFRSF4, KLRK1, CD48, IL2RA), but also various immune suppressive molecules (LAG3, HAVCR2, IDO1, TIGIT, PDCD1), demonstrating its complex role in immune regulation and antigen presentation. Furthermore, we conducted a joint survival analysis of BTN3A1 and γδ T cells. As the follow-up time increased, we found that patients with low BTN3A1 expression and high infiltration of γδ T cells had the highest survival rate (Supplementary Fig. 2). BTN3A1 regulates multiple anti-tumor related pathways in Vδ2 T cells BTN3A1 is expressed in various ovarian cancer cell lines (Fig. 3G). It is also highly expressed in ovarian cancer tissues and negatively correlated with patient prognosis [30]. To further demonstrate the effect of BTN3A1 on Vδ2 T cells, we constructed a BTN3A1 knockout (BTN3A1-KO) cell line (Fig. 3G, Supplement Fig. 3) and confirmed that BTN3A1-KO does not affect tumor cell proliferation (Fig. 3H). Expanded Vδ2 T cells were co-cultured with tumor cells (BTN3A1-WT, BTN3A1-KO) in vitro and were sorted for RNA-seq. After co-culturing for 24 h, 538 DEGs were identified, while 335 DEGs were obtained after 48 h (Fig. 4A). After co-culturing for 24 h, DEGs were mainly enriched in cellular components and molecular functions related to cell adhesion, such as “cell-substrate junction”, “integrin binding”, “focal adhesion” (Fig. 4B). After 48 h, DEGs were more enriched in immune activation pathways, including “positive regulation of cytokine production” and “cytokine receptor binding” (Fig. 4C). GSEA-Hallmark analysis showed that both time points involved the “Apical Junction”, suggesting that BTN3A1 requires intercellular binding to exert its effects on Vδ2 T cells (Fig. 4D, E). In the Hallmark pathways enriched with DEGs over 24 h, “MYC targets V2” was downregulated, and “Notch signaling” was upregulated (Fig. 4D). Myc regulates metabolic reprogramming of activated T lymphocytes [31], while the Notch pathway controls the initiation of CD8 + T cells and the activation of established effector T cells. Activation induced Notch signaling is necessary for IFN-γ release [32]. After 48 h, DEGs were enriched in the “IL6/JAK/STAT3 Signaling”, “Interferon Gamma Response” and “TNFa Signaling Via Nfkb” pathways. These pathways demonstrate the activation of Vδ2 T cells and further promote the killing function of Vδ2 T cells against tumor cells (Fig. 4E). Fig. 4BTN3A1 regulates the function-related pathways of Vδ2 T cells. Vδ2 T cells and OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines were co-cultured (E: T = 1: 1) for 24 hours and 48 hours respectively, and then the separated Vδ2 T cells were subjected to RNA-seq. A Wayne diagrams of up-regulated and down-regulated differential genes. B-E Combined KEGG and GO enrichment analysis of the differentially expressed genes co-cultured at 24h (B) and 48h (C), and GSEA-Hallmark analysis of the differentially expressed genes co-cultured at 24h (D) and 48h (E). F, G GO (F) and KEGG (G) enrichment analysis combined with FC (logfc>2) of all differentially expressed genes co-cultured for 24h and 48h. H GO and KEGG co-enrichment analysis of common differentially expressed genes obtained from co-culture for 24 and 48 hours BTN3A1 regulates the function-related pathways of Vδ2 T cells. Vδ2 T cells and OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines were co-cultured (E: T = 1: 1) for 24 hours and 48 hours respectively, and then the separated Vδ2 T cells were subjected to RNA-seq. A Wayne diagrams of up-regulated and down-regulated differential genes. B-E Combined KEGG and GO enrichment analysis of the differentially expressed genes co-cultured at 24h (B) and 48h (C), and GSEA-Hallmark analysis of the differentially expressed genes co-cultured at 24h (D) and 48h (E). F, G GO (F) and KEGG (G) enrichment analysis combined with FC (logfc>2) of all differentially expressed genes co-cultured for 24h and 48h. H GO and KEGG co-enrichment analysis of common differentially expressed genes obtained from co-culture for 24 and 48 hours In order to better understand the overall changes in the contact between Vδ2 T cells and tumors, we conducted enrichment analysis by combining DEGs at 24 h and 48 h. GO and KEGG analysis confirmed that BTN3A1 mainly activates Vδ2 T cells by regulating “PI3K-AKT signaling pathway” and “Calcium signaling pathway” through mutual contact with the membrane surface of Vδ2 T cells (Fig. 4F, G). The 146 DEGs enriched in the “Focal adhesion”, “CXCR chemokine receptor binding”, and “cytokine activity” modules at both 24 h and 48 h (Fig. 4H), further indicating that BTN3A1 regulates Vδ2 T cells through direct contact, mainly by activating Vδ2 T cells to secrete cytokines and chemokines to kill tumor cells. BTN3A1 weakened the function of activated Vδ2 T cells Through the TCGA database, we found that BTN3A1 was significantly positively correlated with the cytokines IFNG, TNF, as well as the cytotoxic molecules GZMB and PRF1 in ovarian cancer tissue (Fig. 5A). Interestingly, when we co-cultured expanded Vδ2 T cells with tumor cells expressing different levels of BTN3A1, we found that the mRNA levels of IFN-γ and TNF in Vδ2 T cells were higher in the BTN3A1-KO group (Fig. 5B). Meanwhile, TNFSF9 (an activated immune checkpoint molecule), CD69 (an early T cell activation marker), and LAMP1 (which encodes the CD107A protein) were all higher in the BTN3A1-KO group, but BTN3A1 did not affect the expression of KLRK1 (which encodes the NKG2D protein) (Fig. 5C). Due to the correlation analysis indicating a significant association between BTN3A1 and antigen-presenting related molecules in ovarian cancer tissues (Fig. 3F), we further examined the expression of co-stimulatory molecules and found no changes in the molecules CD28 and CD86, with only slight upregulation of HLA-DR (Fig. 5D). Fig. 5BTN3A1 attenuated the functional level of activated Vδ2 T cells. A Correlation between BTN3A1 and IFNG, TNF, PRF1 and GZNB mRNA expression levels in patients with ovarian cancer in TCGA database. B mRNA expression levels of IFNG and TNF after co-culture of Vδ2 T cells with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1: 1) for 24 hours and 48 hours. C, D mRNA expression levels of activated molecules (TNFSF9, CD69, LAMP1, KLRK1) and co-stimulatory molecules (CD80, CD28) and antigen presenting molecules (HLA-DR) after 24h of co-culture of Vδ2 T cells with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1: 1). E-G Vδ2 T cells and OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines were co-cultured (E: T = 1: 1) for 24 h, 48 h and 72h respectively. The percentage of IFN-γ+, TNF-α+ and Perforin+ cells (E) and HLA-DR+, CD80+ and CD28+ cells (F, G) in Vδ2 T cells was detected by flow cytometry BTN3A1 attenuated the functional level of activated Vδ2 T cells. A Correlation between BTN3A1 and IFNG, TNF, PRF1 and GZNB mRNA expression levels in patients with ovarian cancer in TCGA database. B mRNA expression levels of IFNG and TNF after co-culture of Vδ2 T cells with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1: 1) for 24 hours and 48 hours. C, D mRNA expression levels of activated molecules (TNFSF9, CD69, LAMP1, KLRK1) and co-stimulatory molecules (CD80, CD28) and antigen presenting molecules (HLA-DR) after 24h of co-culture of Vδ2 T cells with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1: 1). E-G Vδ2 T cells and OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines were co-cultured (E: T = 1: 1) for 24 h, 48 h and 72h respectively. The percentage of IFN-γ+, TNF-α+ and Perforin+ cells (E) and HLA-DR+, CD80+ and CD28+ cells (F, G) in Vδ2 T cells was detected by flow cytometry To verify the above results, we used flow cytometry to detect protein-level expression of IFN-γ, TNF, and PRF1. These proteins were also expressed at higher levels in the BTN3A1-KO group (Fig. 5E). Therefore, when BTN3A1 acted on activated Vδ2 T cells, it promoted functional downregulation of these cells. As the co-culture time increased from 24 to 72 h, the expression levels of IFN-γ and TNF gradually decreased, indicating that ovarian cancer cells induce Vδ2 T cell exhaustion over time (Fig. 5E). Meanwhile, Vδ2 T cells express high levels of antigen-presenting molecules, however, BTN3A1 does not affect CD28 expression. The effect on HLA-DR expression only appears at 72 h, but its level does not change significantly over time (Fig. 5F, G). Continuous stimulation of BTN3A1 promotes exhaustion and death of activated Vδ2 T cells Due to the reduced expression of Vδ2 T cell activation related molecules by BTN3A1, we further explored the effect of BTN3A1 on exhaustion related molecules in Vδ2 T cells. There is also a strong positive correlation between BTN3A1 and immune checkpoint molecules such as PDCD1, HAVCR2, LAG3, CTLA4, TIGIT, CD274 in tumor tissues (Fig. 6A). After co-culturing in vitro, we found that the exhaustion related transcription factors (TFs) TOX and EOMES were expressed at higher levels in Vδ2 T cells from the BTN3A1-WT group, while TCF7 expression showed no significant difference (Fig. 6B). Meanwhile, the expression of immunosuppressive molecules was also higher in Vδ2 T cells from the BTN3A1-WT group (Fig. 6C, D). At the protein level, the expression levels of PD-1, TIM3, and LAG3 were also higher in the BTN3A1-WT group (Fig. 6E). This is consistent with the previous statement that BTN3A1 is positively correlated with both activating molecules and immunosuppressive molecules (Fig. 3F). The expression of immunosuppressive molecules in Vδ2 T cells gradually decreases with the prolongation of co-culture time, which may be related to increased Vδ2 T cell death during the co-culturing. Fig. 6BTN3A1 promoted exhaustion and death of activated Vδ2T cells. A Correlation between BTN3A1 and PDCD1, HAVCR2, LAG3, CTLA4, TIGHT and CD274 mRNA expression levels in patients with ovarian cancer in TCGA database. B mRNA expression levels of exhaustion-associated transcription factors TOX, EOMES, and activating transcription factor TCF7 after 24 h of co-culture of Vδ2 T cells with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1:1). C, D mRNA expression levels of immunosuppression-related molecules PCDC1, HAVCR2, LAG3, ENTPD1, TIGHT, and CTLA4 after 24 h of co-culture of Vδ2 T cells with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1:1). E, F Vδ2 T cells were co-cultured with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1:1) for 24 h, 48 h, 72 h, and 96 h, respectively. The percentage of PD-1+, Tim3+, Lag3+ cells (E) and Annexin V-7-AAD- cells (F) in Vδ2 T cells was detected by flow cytometry after receiving cell culture medium BTN3A1 promoted exhaustion and death of activated Vδ2T cells. A Correlation between BTN3A1 and PDCD1, HAVCR2, LAG3, CTLA4, TIGHT and CD274 mRNA expression levels in patients with ovarian cancer in TCGA database. B mRNA expression levels of exhaustion-associated transcription factors TOX, EOMES, and activating transcription factor TCF7 after 24 h of co-culture of Vδ2 T cells with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1:1). C, D mRNA expression levels of immunosuppression-related molecules PCDC1, HAVCR2, LAG3, ENTPD1, TIGHT, and CTLA4 after 24 h of co-culture of Vδ2 T cells with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1:1). E, F Vδ2 T cells were co-cultured with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1:1) for 24 h, 48 h, 72 h, and 96 h, respectively. The percentage of PD-1+, Tim3+, Lag3+ cells (E) and Annexin V-7-AAD- cells (F) in Vδ2 T cells was detected by flow cytometry after receiving cell culture medium When co-cultured with ovarian cancer cells, Vδ2 T cells show enrichment in genes related to the regulation of the apoptotic signaling pathway (Fig. 1F). Additionally, BTN3A1 regulates the apoptosis pathway in OC (Fig. 3A). Therefore, we measured the apoptosis levels of Vδ2 T cells after co-culture and found that the BTN3A1-WT group, which highly expressed exhaustion markers, exhibited increased Vδ2 T cell apoptosis (Fig. 6F). BTN3A1 affects the function of Vδ2 T cells through the TCR signaling pathway The enrichment analysis in the previous section indicates that BTN3A1 mainly exerts regulatory effects through intercellular binding (Fig. 4F). Previous studies also have shown that continuous antigen stimulation of TCR triggers T cell exhaustion. Since BTN3A1 can activate Vδ2 T cells via Vγ9Vδ2 TCR [28], it may also promote the expression of exhaustion markers through Vγ9Vδ2 TCR. We found that the higher the level of BTN3A1 expression in tumor cells, the higher the activation level of the TCR signaling pathway (Supplementary Fig. 4). Therefore, BTN3A1 activates the TCR signaling pathway of Vδ2 T cells. Then, we added Vγ9 antibody to the co-culture system of ovarian cancer cells and Vδ2 T cells to block the binding of BTN3A1 to Vδ2 T cells, resulting in the disappearance of differential PD-1 and TIM3 expression (Fig. 7A). Fig. 7Membrane-expressed BTN3A1 promoted Vδ2 T cell exhaustion and death-related molecule expression via TCR signaling pathway. A Cell culture supernatants were collected 24 h after Vδ2 T cells were co-cultured with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines with or without Vγ9 neutralizing antibody, and the percentage of PD-1+, TIM3+ cells in Vδ2 T cells was detected by flow cytometry. B OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cells were plated on a six-well plate and irradiated (50 GY) after cell adhesion. After 6 h, OVCAR8 tumor cells were co-cultured with Vδ2 T cells (E: T = 1:1). After 48 h, the cell culture medium was collected to detect the percentage of PD-1+ and Tim3+ cells in Vδ2 T cells by flow cytometry. C Vδ2 T cells were co-cultured with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1:1) for 24 h, 48 h, 72 h, and 96 h, respectively. Cell culture medium was collected and the percentage of NR4A1+, NR4A2+, NR4A3+ cells in Vδ2 T cells was detected by flow cytometry. D, E Vδ2 T cells were co-cultured with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1:1) for 24 h, 48 h, 72 h, and 96 h, respectively. The percentage of PD-L1+ and FASL+ cells in Vδ2 T cells was detected by flow cytometry after receiving the cell culture medium Membrane-expressed BTN3A1 promoted Vδ2 T cell exhaustion and death-related molecule expression via TCR signaling pathway. A Cell culture supernatants were collected 24 h after Vδ2 T cells were co-cultured with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines with or without Vγ9 neutralizing antibody, and the percentage of PD-1+, TIM3+ cells in Vδ2 T cells was detected by flow cytometry. B OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cells were plated on a six-well plate and irradiated (50 GY) after cell adhesion. After 6 h, OVCAR8 tumor cells were co-cultured with Vδ2 T cells (E: T = 1:1). After 48 h, the cell culture medium was collected to detect the percentage of PD-1+ and Tim3+ cells in Vδ2 T cells by flow cytometry. C Vδ2 T cells were co-cultured with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1:1) for 24 h, 48 h, 72 h, and 96 h, respectively. Cell culture medium was collected and the percentage of NR4A1+, NR4A2+, NR4A3+ cells in Vδ2 T cells was detected by flow cytometry. D, E Vδ2 T cells were co-cultured with OVCAR8 (BTN3A1-WT, BTN3A1-KO) tumor cell lines (E: T = 1:1) for 24 h, 48 h, 72 h, and 96 h, respectively. The percentage of PD-L1+ and FASL+ cells in Vδ2 T cells was detected by flow cytometry after receiving the cell culture medium To exclude changes in tumor cell membrane proteins caused by activated Vδ2 T cells, we first irradiated tumor cells to inhibit their activity. Then, we co-cultured these irradiated tumor cells with Vδ2 T cells. We found that the exhaustion effect of BTN3A1 on Vδ2 T cells still existed (Fig. 7B). Due to the fact that TF NR4As promote T cell exhaustion and apoptosis [33, 34], we also detected TF NR4As in Vδ2 T cells and found that after 48 h, BTN3A1 significantly promoted the expression of TF NR4As (Fig. 7C). Therefore, BTN3A1 on the membrane surface of tumor cell promotes the expression of immunosuppressive molecules in Vδ2 T cells through TCR signaling. BTN3A1 both activates Vδ2 T cells and promotes upregulation of PD-1, it is highly likely that TCR activation promotes the FAS/FASL [35] and PD-1/PD-L1 [36] pathways, leading to Vδ2 T cell exhaustion and death. After co-culturing, we detected the expression of PD-L1 and FASL in Vδ2 T cells. BTN3A1 also promotes the expression of PD-L1 and FASL (Fig. 7D, E), which further accelerates the exhaustion and death of Vδ2 T cells. Discussion At present, γδ T cells exhibit strong cytotoxicity against various cancer cells in an MHC-unrestricted manner. Consequently, many studies have performed in vitro and in vivo expansion and adoptive transfer for tumor immunotherapy [9]. The adoptive transfer of amplified Vδ2 T cells completely eliminated smaller ovarian cancer tumors in NSG mice [12]. CDR3δ-grafted Vδ2 T cells also showed clear inhibitory effects on the growth of ovarian cancer tumors, significantly prolonging the survival of tumor-bearing mice [37, 38]. When liposomal alendronic acid is injected into the peritoneal cavity as a sensitizing agent, it utilizes the “enhanced permeability and retention effect” to make late-stage tumors vulnerable to attack by adoptively transfered γδ T cells, promoting the regression of formed ovarian tumors by more than an order of magnitude [39]. Antibodies targeting tumor cells in combination with γδ T cells can also exert effective anti-tumor effects. For example, the TCR γ9δ2 (OT3)-Fc fusion protein possesses both TCR γδ antigen recognition properties and Fc-mediated antibody effector functions, significantly inhibiting tumor growth and improving survival rates in human ovarian cancer xenograft models [40]. Bispecific monoclonal antibodies were constructed by combining monoclonal antibodies (GI or A13) that stimulate TCR γ/δ with Mov19 monoclonal antibodies specific to ovarian cancer cells. These bispecific antibodies effectively target TCR γ/δ lymphocytes and ovarian cancer cells, facilitating targeted tumor cell lysis [41]. Therefore, Vδ2 T cells have great therapeutic potential in OC, and exploring their functional changes upon entering the TME is of great significance for enhancing their therapeutic ability. This study preliminarily explored the effect of tumor cells on Vδ2 T cells. We found that prolonged BTN3A1-TCR binding leads to exhaustion of Vδ2 T cells. The downstream signal transduction cascade downstream of TCR junction is extremely complex, involving the interconnection of multiple signaling pathways. For example, TCR participates in activating the PKCθ and MAPK/ERK pathways, both of which can promote the activity of TF NF-κB, which in turn can trigger the calcium signaling pathway, affect the activation of TF NFAT, and affect the expression of related TFs [42]. This complexity is also reflected in our research results, that is, the BTN3A1-TCR interaction can affect downstream TFs NR4A2/3. Meanwhile, the existence of cross regulation in these downstream pathways remains an unresolved issue. Therefore, further research is needed to confirm the detailed regulatory pathways. At the same time, our research shows that tumor cells expressing BTN3A1 can indeed induce the upregulation of exhaustion molecules in Vδ2 T cells expanded in vitro through TCR receptors within 24 h, but Vδ2 T cells still retain strong killing ability. Therefore, further exploration is needed to determine the extent to which the BTN3A1 expressed by tumor cells affects the functional exhaustion of Vδ2 T cells, whether it is persistent, and whether it is reversible. BTN3A1 is highly expressed in OC and significantly negatively correlated with patient survival rate. BTN3A1 inhibits tumor reactive αβ-TCR activation by preventing N-glycosylated CD45 from separating from immune synapses [43]. In the PD-L1 humanized system, targeting BTN3A1 can overcome the highly immunosuppressive microenvironment of OC and is more effective than neutralizing the PD-L1/PD-1 checkpoint in delaying malignant progression [43]. IFN-γ can induce the expression of BTN3A1 in a STAT1 dependent manner. The infiltration levels of CD4, CD8 T cells, and Vδ2 T cells, as well as the expression levels of IFN-γ, are positively correlated with BTN3A1 expression, and this correlation is consistent with PD-L1 [44]. Therefore, the expression pattern and regulatory mechanism of BTN3A1 make it highly similar to PD-L1, suggesting that BTN3A1 may be a molecule with partial features of immune checkpoint pathways in tumors. However, because the TME is a complex network, Vδ2 T cells extensively interact with various other immune cells. For example, γδ T cells in ovarian cancer tissue promote the immunosuppressive function of naive CD4 T cells [13], and those producing IL-17 A mobilize macrophages to promote tumor cell growth [45]. Therefore, further exploration is needed to investigate the mutual influence between other immune cells and Vδ2 T cells on functional changes. At the same time, BTN3A1 also regulates the immune activity of CD4 T, CD8 T, monocytes, and NK cells, playing an important role in regulating the immune system [46]. Therefore, different expression levels and expression sites of BTN3A1 may have different effects on Vδ2 T cells. Given the lack of Vδ2 T cell counterparts in mice and significant interspecific differences in BTN3A1, it is difficult to construct a suitable in vivo tumor model to further explore the relationship between them. Therefore, more physiologically relevant systems—such as humanized mouse models or organoid-based platforms—would be required to fully capture these interactions. This is also the direction of our further efforts, in order to achieve better clinical translation. Conclusion This study preliminarily explored Vδ2 T cells as an ideal candidate for therapy. These cells become active upon contact with ovarian cancer cells but also initiate exhaustion and apoptosis programs. BTN3A1, an important molecule activating Vδ2 T cells, also contributes to their exhaustion by regulating the TCR signaling pathway. Therefore, BTN3A1 may serve as a regulator of activation-induced exhaustion in γδ T cells [44]. Supplementary Information Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2.