Novel recombinant protein PK5-Gal-3C enhances the anti-tumor activity of T cells via binding with TRPV2 Authors: Guopeng Zhang, Wei Zhang, Xiaohuan Wei, Zhenyu Wang, Feitong Wang, Shishuo Sun, Haiheng Xu, Xiangye Liu, Qing Zhang, Xiaoge Gao Published: 2026 ABSTRACT AbstractInvestigating strategies to enhance T cell effector function can improve the adoptive immune responses to tumors and complement existing tumor immunotherapies. Here, we present a novel artificially designed recombinant protein, PK5-Gal-3C, which is composed of the fifth kringle domain of plasminogen (PK5) and the C-terminal carbohydrate-recognition domain of galectin-3 (Gal-3C). This protein exhibited potent anti-tumor activity by significantly enhancing T cell effector function. Specifically, PK5-Gal-3C directly activated T cells by binding to glycosylated TRPV2 via Gal-3C, a thermosensitive calcium-permeable cation channel, thereby promoting the influx of calcium ions to enhance T cells cytotoxicity via the activation of c-Jun. Correspondingly, inhibition of TRPV2 or c-Jun impaired the cytotoxicity of T cell mediated by PK5-Gal-3C. Additionally, PK5-Gal-3C demonstrated significant anti-tumor activity by enhancing T cell tumor infiltration and cytotoxicity in a mouse model, as well as improving the anti-tumor efficacy of CAR-T cells in solid tumors. In summary, PK5-Gal-3C is a safe and potent anti-tumor agent with promising potential for T cell-mediated cancer immunotherapy.Graphical abstract Supplementary InformationThe online version contains supplementary material available at 10.1007/s00262-026-04401-3.Keywords: Novel recombinant protein PK5-Gal-3C; T cell; Transient receptor potential cation channel, subfamily V member 2; Immunotherapy; Solid tumor Introduction Immunotherapy has rapidly developed as relative novel cancer therapeutic approaches in cancer treatment, aiming to reactivate immune cells or transfer engineered immune cells to recognize and eliminate tumor cells [1]. Until now, immunotherapy has achieved considerable success in the clinical therapy, particularly through the immune checkpoint blockade (ICB) and adoptive cell therapy (ACT) [2]. Unfortunately, a great majority of patients failed to response to ICB, especially for those “cold” tumor with paucity of T cell infiltration [3]. Although ACT has achieved remarkable clinical success in the clinical therapy of hematologic malignancies, a considerable challenge of ACT remains that there is not outstanding breakthrough in the therapy of solid tumors [4]. Under physiological conditions, T cells, as the principal effector cells of the immune system, are positively correlated with patient survival when enriched in tumor tissues [5]. Simultaneously, T cells serve as the primary target of ICB [6, 7] and constitute the main effector cells for ACT [4, 8]. Therefore, it is very important and valuable to investigate the strategies for enhancing the T cell effector function. PK5, the fifth kringle domain of plasminogen, exhibits anti-angiogenic activity by inhibiting endothelial cells growth and migration in vivo and in vitro [9, 10]. Galectin-3, a member of the mammalian galectin family, specifically binds to the β-galactosides and is composed of the N-terminal regulatory peptide and the C-terminal carbohydrate-recognition domain (Gal-3C) [11]. More evidences suggested that recombinant Gal-3C acts as a potent inhibitor of full-length Gal-3 via competitively binding with its endogenous ligands [12]. Previously, we engineered a recombinant protein, PK5-Gal-3C, by the fusion of PK5 to the N-terminus of Gal-3C, aiming to develop a novel fusion protein with enhanced anti-tumor efficacy. PK5-Gal-3C exhibited increased anti-angiogenic activity, as demonstrated in our previous investigations [13]. The transient receptor potential cation channel, subfamily V member 2 ( TRPV2) is a member of the transient receptor potential (TRP) channel family and is capable of detecting noxious temperatures exceeding 52 °C, as well as various stimuli such as osmotic pressure, chemical agents, and mechanical forces [14, 15]. TRPV2 is composed of a large N-terminal cytoplasmic domain, a C-terminal cytoplasmic domain, and six transmembrane segments that include a pore-forming loop exhibiting high permeability to calcium ions [16]. Increasing evidences demonstrates that TRPV2 plays a pivotal role in modulating both innate and adoptive immune responses, encompassing processes such as phagocytosis of macrophage [17], mast cells degranulation [18], and B cells’ activation and function [19]. Notably, in 2006, Sauer and Jegla first proposed in a patent application that TRPV2 may serve as a potential target for the identification of compounds capable of regulating T cell activation and function [20]. Consequently, further investigation into TRPV2 elucidates their regulatory effects on T cell functions. In the present study, we reported that PK5-Gal-3C could specifically bind with the glycosylated TRPV2 in T cells via the Gal-3C domain, facilitating the influx of calcium ions, which in turn directly activated T cells to enhance their effector function. PK5-Gal-3C also exhibited significant anti-tumor activity in a murine model of colon cancer by augmenting both the infiltration and cytotoxicity of T cells within the tumor microenvironments (TMEs). Additionally, it improved the anti-tumor efficacy of human CAR-T cells in solid tumors using an immunosuppressed mouse model. These results indicated that PK5-Gal-3C represents a safe and potent anti-tumor agent and may serve as a valuable adjunct to these existing T cell-mediated tumor immunotherapies aimed at improving therapeutic efficacy and response rates. Materials and methods Cell culture The human colorectal cancer cell lines (HCT116 and HCT-8) and human renal cancer lines (OSRC-2 and 786-O) used in this study were maintained in the RPMI-1640 medium (Bio-Channel, Nanjing, China). Conversely, the human liver cancer cell lines (Huh7 and PLC/PRF/5) and the mouse colon cancer cell line MC38 were cultured in high-glucose Dulbecco’s modified eagle medium (DMEM, KeyGen Biotech, Nanjing, China). Human peripheral blood mononuclear cells (PBMCs) were collected from healthy donors following established protocols [21]. For T cell preparation, PBMCs were activated using CD3/CD28 Dynabeads™ (Gibco) for five days, followed by further expansion for an additional 3–5 days in the X-VIVO medium supplemented with recombinant human interleukin-2 (IL-2; Novoprotein, Suzhou, China). Subsequently, T cells were cryopreserved in the liquid nitrogen for subsequent in vitro and in vivo application. All culture media were supplemented with 10% fetal bovine serum (FBS; Bio-Channel, Nanjing, China) and 1% penicillin/streptomycin. All cell cultures were maintained in a humidified incubator at 37 °C with 5% CO2. Antibodies and reagents The following antibodies for flow cytometry were purchased from BioLegend (Shanghai, China) and included the following: PE-conjugated anti-human CD3, APC-conjugated anti-human CD4, APC-Cy7-conjugated anti-human CD8, FITC-conjugated anti-human CD25, PE-Cy7-conjugated anti-human CD25, and PE-Cy7-conjugated anti-human CD69, and PerCP-conjugated anti-mouse CD45, PE-Cy5 conjugated anti-mouse CD3, FITC-conjugated anti-mouse CD4, and BV786-conjugated anti-mouse CD8. The antibodies targeting phosphorylated c-Jun and total c-Jun were obtained from Cell Signaling Technology (Shanghai, China). The antibodies of TRPV2 and galectin-3 were obtained from Proteintech (Wuhan, China). Horseradish peroxidase (HRP)-labeled secondary antibodies were all purchased from Beyotime Biotechnology (Shanghai, China). Gold Band Plus 3-color Regular Range Protein Marker (8–180 kDa) was acquired from Yeasen Biotechnology (Shanghai, China). Human ELISA kits for TNF-α (RD-TNFα-Hu), for granzyme B (GzmB, CT211A), and for IFN-γ (RD-IFNγ-Hu) were all obtained from Mabtech (Shanghai, China). Additionally, BAPTA/AM, a membrane permeable calcium chelator, and SP600125, a JNK inhibitor, were purchased from MCE (Shanghai, China). SET2, a TRPV2 antagonist, was obtained from TargetMol (Shanghai, China). Animals The female C57NL/6 J mice aged six to eight weeks old, as well as NCG (NOD/ShiLtJGpt-Prkdcem26Cd52Il2rgem26Cd22/Gpt) mice, were purchased from GemPharmatech Co., Ltd (Nanjing, China). Additionally, female NPG (NOD.Cg-Prkdcscid Il2rgtm1/Vst) mice were purchased from Vitalstar Biotechnology Co., Ltd (Beijing, China). All mice were housed within an individually ventilated cage (IVC) system at the Animal Center of Xuzhou Medical University, maintained under controlled environmental conditions with temperatures ranging from 21 to 22 °C and relative humidity levels between 50 and 60%. The mice were provided with a standardized diet and water following autoclaving. These parameters including body weight, food intake, and water consumption were recorded at least once daily. To reduce the risk of infection, each mouse will be provided with a separate syringe, and local disinfection will be conducted prior to injection. Mice will be euthanized if they exhibiting a sudden loss of body weight for three consecutive days, or if they presenting with severe abdominal fluid accumulation or tumor ulceration. The animal experiments were conducted from March 2023 to June 2025, in accordance with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines 2.0, and were approved by the Experimental Animal Ethics Committee of Xuzhou Medical University (Permit number: 202303T010). Preparation of recombinant protein PK5-Gal-3C Recombinant proteins PK5-Gal-3C and RFP were expressed in BL21(DE3) following induction with IPTG, and subsequently purified via affinity chromatography according to the previously reported methods [13]. Preparation of CAR-T cells targeting carbonic anhydrase IX (CAIX) The human PBMCs were activated by CD3/CD28 Dynabeads (11132D, Gibco), and then infected with lentiviral constructs encoding CAIX-targeting CAR with costimulatory domain 41BB according to our previously reported protocols [21]. The lentivirus-infected T cells were continuously expanded for additional five days prior to their application in subsequent in vitro and in vivo assays. Generation of wide-type and mutant TRPV2 plasmids The cDNA of human TRPV2 (pOTB7-TRPV2, P16770) was obtained from MiaoLingBio, China, and subsequently cloned into the pCDNA3.1 eukaryotic expression vector, with a Myc tag added at the N-terminus. To construct the mutant TRPV2, partially reverse complementary primers were designed for the mutated site. The cDNA of mutant TRPV2 was amplified using a super-fidelity DNA polymerase and then incubated with the linearized pCDNA3.1 vector in a specific ratio according to the ClonExpress® MultiS one-step cloning kit (Vazyme, Nanjing, China) protocol. The recombinant product was transformed into competent cells for further amplification, purification, and sequencing. Subcutaneous wide-type mice model The MC38 cells (1 × 106/mouse) were collected and slowly injected into the right back flask of anesthetized C57NL/6 J mice using tribromoethyl alcohol. After seven days, all tumor-bearing mice were tagged and randomly subdivided into three groups of six mice each and treated by intraperitoneal injection with PBS, RFP, and PK5-Gal-3C every two days at the dose of 25 mg/kg mouse body weight, respectively. All mice were housed in a controlled environment and treated at the same time each day, albeit in a randomized order. The mice body weight and tumor size were measured every three days, and the tumor volume was calculated based on the formula: length × width2/2. After three weeks, all tumor-bearing mice were anesthetized by intraperitoneal injection of tribromoethyl alcohol and subsequently euthanized by cervical dislocation. The peripheral blood and tumor tissue were removed and collected, and the tumor was cut into small pieces. The single-cell suspension was processed using the Miltenyi gentleMACS dissociator, which was equipped with a heater set to 37 °C. The tumor-infiltrating T cells were stained by antibodies and analyzed or sorted using flow cytometry for further in vitro cytotoxicity assessment. Deletion of CD8 + T cell in vivo model As described above, MC38 cells (1 × 106/mouse) were injected into the right back flask of C57NL/6 J mice. All mice were randomly grouped of six mice each and treated with InVivoMAb anti-mouse CD8α antibody (clone: 53–6.7, BioXcell) (CD8αAb) by intraperitoneal injection at the dose of 200 μg/mouse twice a week since the fourth day according to the manufacture’s protocol. The population of CD8 + T cells in peripheral blood was analyzed by flow cytometry 24 h after intraperitoneal injection of CD8α antibody. Then, PK5-Gal-3C was administrated every two days at the dose of 25 mg/kg mouse body weight after one week. The body weight and tumor size were measured every three days. After four weeks, all tumor-bearing mice were anesthetized by intraperitoneal injection of tribromoethyl alcohol and subsequently euthanized by cervical dislocation. The tumor tissue was collected and weighted. Subcutaneous NPG/NCG mice model According to our previously reported protocols, OSRC-2 cells (2 × 106/mouse) were collected and injected into the right back flank region of NPG or NCG mice with six to eight weeks old. To further explore the direct effects of PK5-Gal-3C on T cell effector functions, activated T cells were incubated with fresh medium every day containing PK5-Gal-3C or RFP (2 μg/ml) for five consecutive days. Then, pretreated T cells (1 × 107/mouse) in 100 µl of PBS buffer were injected into tumor-bearing NPG mice via the tail vein every two weeks, respectively. To further measure the synergistic anti-tumor effect, tumor-bearing mice were randomly grouped of six mice each and treated with CAR-T cells targeting CAIX (1 × 107/mouse) in 100 µl of PBS buffer by tail intravenous injection in the presence or absence of PK5-Gal-3C treatment (25 mg/kg mouse body weight, every two days). The body weight and tumor size were measured every three days. Flow cytometry assay Flow cytometry was utilized to detect the expression of cell surface markers. According to our previously reported protocols [21], cells were collected and incubated with specific antibodies for flow cytometry at 4 °C for 30 min, followed by staining with live/dead dyes (APC). The cells were subsequently analyzed using flow cytometry. Additionally, CountBright™ absolute counting beads (Invitrogen) were employed to assess the number of T cells. CFSE Cell proliferation assay Firstly, activated T cells were stained with CFSE solution and measured by flow cytometry as the control. The stained T cells were cultured for another 48, 72, and 96 h and measured with flow cytometry. Calcium flux assay Firstly, the cells were stained using the Fluo-4 calcium assay kit (Beyotime, Shanghai) and then washed three times. Subsequently, the stained cells were stimulated with PK5-Gal-3C under the specified conditions and analyzed using a flow cytometer for 2–3 min. Real-time cellular analysis (RTCA) The RTCA was performed using the xCELLigence RTCA S16 instrument (Agilent Technologies) to analyze the cytotoxicity of T cells in vitro. Firstly, 50 μl of cell medium was added to an E-plate 16 for measuring the background impedance. The target cells were seeded in the E-plate for adhesion, after which T cells (effector cells) and recombinant PK5-Gal-3C protein were added at the specified effector-to-target (E:T) ratio and concentration. The E-plate instrument operates within a standard cell culture incubator, allowing for real-time monitoring of the dynamic impedance of target cells. The normalized cellular index values at the endpoint were then determined and analyzed using the RTCA data analysis software 1.0. Enzyme-linked immunosorbent assay (ELISA) The ELISA was performed to analyze the concentration of cytokines in the cultured supernatant collected from the incubation of tumor cells and T cells, including IFN-γ, granzyme B (GzmB), and TNF-α following the manufacturers’ protocols, respectively. The absorbance at 450 nm was measured using Cytation 3 (BioTeck) reader. The concentration of cytokines was calculated according to the standard curves. Immunofluorescence (IF) assay Firstly, T cells were incubated with PK5-Gal-3C proteins with Flag tag for 2 h. Then, T cells were collected and washed three times with pre-chilled phosphate-buffered saline (PBS). T cells were fixed in 4% paraformaldehyde and incubated with primary antibodies targeting Flag and TRPV2, respectively. The second antibodies labeled with fluorescence were added for further 1-h incubation. Additionally, DAPI was added to stain the nucleus of T cells. Finally, T cells were mounted and viewed under a confocal microscope (ZEISS LSM880). Western Blotting assay Firstly, the cells were treated under the indicated conditions and subsequently collected for lysis in the RIPA buffer with diluted protease inhibitor cocktails. The supernatant was collected and quantified with BCA protein assay kit according to the manufacturer’s directions. Proteins were separated on SDS–polyacrylamide gel electrophoresis, transferred onto membranes (GE HealthCare), and probed with the specific primary antibodies followed by HRP-labeled secondary antibodies. Positive signals were detected using Tanon imaging system (Tanon Science & Technology Co., Ltd, Shanghai, China). For immunoprecipitation (IP) assay, T cells were incubated with or without PK5-Gal-3C bearing a Flag tag for 2 h, followed by three washes with chilled PBS. The lysed protein supernatant was incubated with the anti-Flag magnetic beads overnight. The beads were washed and collected for subsequent analysis using shotgun liquid chromatography-tandem mass spectrometry (LC–MS/MS), which was performed using a Q Exactive Plus mass spectrometer coupled with an EASY-nLC (Thermo Fisher Scientific, MA, USA) in the GenecCem Co., Ltd (Shanghai, China). Statistical Analysis All in vitro data are expressed as means ± standard deviation (SD), while in vivo data are presented as means ± standard error of the mean (SEM). Comparisons between two groups, assuming equal variances, were conducted using both paired and unpaired Student’s t tests. Comparisons among three or more groups were analyzed using one-way or two-way ANOVA, following normality and log-normality tests. All data analyses were performed with GraphPad Prism 8.0.1 (GraphPad Software, Inc., CA, USA), and a p value of less than 0.05 was considered statistically significant. Results The recombinant fusion protein PK5-Gal-3C inhibits tumor growth through T cell-dependent anti-tumor immunity The subcutaneous (s.c.) MC38 tumor model was established according to the methods. All mice were randomly assigned to groups and subsequently treated with PK5-Gal-3C, the control recombinant protein RFP, or PBS under standardized conditions, respectively (Fig. 1A). Interestingly, treatment with PK5-Gal-3C effectively inhibited tumor growth, with an average tumor weight of 0.27 g, compared to 0.43 g and 0.58 g observed in the RFP and PBS groups, respectively (Fig. 1 B-D). Besides, PK5-Gal-3C or RFP treament did not affect the body weight of tumor-bearing mice compared to the PBS group (Fig. S1A). These findings are consistent with previously reported effects in a liver cancer model [13]. We next determined the subset of tumor-infiltrated CD8 + T cells that responded to PK5-Gal-3C treatment. Flow cytometry analysis revealed a significant increase in both the frequency (26.2%) and absolute number (4.4 × 104 per gram of tumor tissue) of CD8 + T cells within the tumors from PK5-Gal-3C-treated group relative to those treated with RFP (13.2%, 1.6 × 104 per gram of tumor tissue) and PBS (16.7%, 1.3 × 104 per gram of tumor tissue) (Figs. 1E-G and S1B). Recombinant fusion protein PK5-Gal-3C enhanced the anti-tumor activity of tumor-infiltrating T cells in vivo. (A) Schematic diagram of the mouse model. C57BL/6 J mice were administered MC38 cells intraperitoneally, followed by treatment with the recombinant fusion protein PK5-Gal-3C, the control protein RFP, or the solvent PBS, every two days. (B) Tumor growth curves were recorded every three days. (C and D) Tumors were excised and weighed at the study’s conclusion (n = 6). (E) Representative flow cytometry plots demonstrating the tumor-infiltrating CD8 + T cells, with the gating strategy outlined in Fig. S1B. (F and G) Frequency and counts of CD8 + T cells within the tumor tissues. (H) Schematic representation of the mouse model. C57BL/6 J mice with subcutaneous xenograft tumors were treated with a specific anti-CD8α antibody to deplete CD8 + T cells, either in the presence or absence of PK5-Gal-3C. (I) Tumor growth curves were recorded every three days. The tumor mass was excised (J) and weighed (K) at the conclusion of the experiment (n = 7). After the first intraperitoneal injection of the anti-CD8α antibody, one mouse from the combination group died, possibly related to the experimental procedure. (L) Schematic diagram of tumor-infiltrating T cells was sorting and analyzed their cytotoxicity with RTCA in vitro. (M) RTCA was performed to assess the cytotoxicity of sorted T cells isolated from tumor tissues. MC38 cells were seeded in the E-plate as target cells, followed by the addition of sorted T cells under specific conditions. (N) Statistical analysis of the normalized cell index was conducted using the RTCA software (RTCA data analysis software 1.0). *p < 0.05, **p < 0.01, ***p < 0.001 These findings suggested that the anti-tumor efficacy of the recombinant fusion protein PK5-Gal-3C may be contingent upon T cell involvement. To confirm this hypothesis, CD8 + T cells, which serve as primary effector cells, were depleted using CD8αAb in the presence or absence of PK5-Gal-3C in vivo (Fig. 1H). Treatment with CD8αAb effectively reduced the population of peripheral CD8 + T cells to 10.5% compared to 33.2% observed in PK5-Gal-3C-treated tumor-bearing mice (Fig. S1C-E), and did not affect the body weight of tumor-bearing mice (Fig. S1F). Correspondingly, the depletion of CD8 + T cells significantly reduced the tumor growth inhibition induced by PK5-Gal-3C, as evidenced by an increase in average tumor weight to 1.52 g relative to that in the PK5-Gal-3C group (Fig. 1I-K). Furthermore, the cytotoxicity of sorted tumor-infiltrating T cells was reassessed in vitro using RTCA (Fig. 1L). The results indicated that T cells derived from PK5-Gal-3C-treated cohort exhibited significantly enhanced cytotoxicity against MC38 cells, with the inhibitory ratio of 93% and 90% compared to those from the RFP and PBS groups, respectively (Fig. 1M and N). Collectively, these data underscored the pivotal role of T cell-mediated anti-tumor responses in the tumor growth inhibition suppression elicited by PK5-Gal-3C. PK5-Gal-3C could directly enhance the activation, proliferation, and effector functions of T cells in vitro As previously demonstrated, T cell-mediated anti-tumor activity plays a crucial role in tumor growth inhibition induced by PK5-Gal-3C. To further elucidate the direct impact of PK5-Gal-3C on T cell functionality, we examined its effects on T cell activation, proliferation, and effector functions in vitro, respectively. For T cell activation, the expression of CD69 and CD25, well-recognized markers of T cell activation, was assessed. Firstly, PBMCs, representing the naïve T cells, were incubated with PK5-Gal-3C for 24 h and then analyzed with flow cytometry to quantify the expression of CD69 and CD25. As shown in Fig. 2A and B, PK5-Gal-3C treatment could increase the proportion of CD69 + (18.1%) and CD25 + (8.13%) T cells within PBMC population (p < 0.001). Moreover, PK5-Gal-3C treatment also could enhance the frequency of CD69 + (18.3%) and CD25 + (40.5%) T cells among T cells pre-activated with anti-CD3/CD28 antibodies (Fig. 2C and D). PK5-Gal-3C could promote the activation and proliferation of T cells in vitro. PK5-Gal-3C could promote the activation of T cells indicated by the increase in the frequency of CD69 + and CD25 + T cells and the intracellular calcium ions. Both PBMCs (A and B) and activated T cells (C and D) were incubated with PK5-Gal-3C, respectively. The frequencies of CD69 + and CD25 + T cell subsets were subsequently analyzed using flow cytometry. The gating strategy is illustrated in Fig. S2. (E) PK5-Gal-3C increased the intracellular calcium levels of T cells, but not PK5 or Gal-3C alone. T cells were stained with Fluo-4 and then incubated with PK5-Gal-3C, PK5, and Gal-3C, respectively. Flow cytometry was performed to assess T cell fluorescence. (F) PK5-Gal-3C-induced influx of calcium ions from extracellular medium. T cells were stained with Fluo-4 and then incubated with in HBSS buffer, with or without calcium ions. Flow cytometry was performed to assess T cell fluorescence. (G) PK5-Gal-3C promoted T cell proliferation in vitro. Activated T cells were stained with CFSE and incubated with PK5-Gal-3C for further 48 h, 72 h, and 96 h. Flow cytometry was performed to assess the mean fluorescence intensity (MFI) under the indicated time points. (H) T cell numbers were counted with the countess II automated cell counter under the indicated time points. * p < 0.05, ***p < 0.001 It is well documented that elevated cytosolic calcium ion concentrations are critical for T cell activation, proliferation, differentiation, and cytotoxic functions, serving as a hallmark of activated T cells [22, 23]. As shown in Fig. 2E, PK5-Gal-3C induced a significant increase in intracellular calcium ion levels in T cells, an effect not observed with PK5 or Gal-3C alone. Furthermore, depletion of extracellular calcium ions from the HBSS buffer substantially abrogated PK5-Gal-3C- induced calcium influx (Fig. 2F), indicating that PK5-Gal-3C promoted calcium inflow from the extracellular medium. For the T cell proliferation assay, activated T cells were labeled with CFSE and subsequently cultured for further 48 h, 72 h, and 96 h, respectively. All T cells were further analyzed with flow cytometry and counted under the indicated time points. Incubation with PK5-Gal-3C resulted in 37.7% reduction in mean fluorescence intensity (MFI) of CFSE and 49.2% increase in the total T cell numbers relative to the control group (Fig. 2F and G). Taken together, these above results demonstrated that PK5-Gal-3C protein could directly promote the activation and proliferation of T cells. To further evaluate the cytotoxic function and cytokine secretion of T cells induced by PK5-Gal-3C, we performed RTCA and ELISA assays, respectively. As shown in Fig. 3A-D, neither T cells nor PK5-Gal-3C alone significantly affected tumor cells growth under the specified conditions. However, the presence of PK5-Gal-3C markedly facilitated T cell-mediated cytotoxicity against HCT116 and HCT-8 tumor cells. Furthermore, ELISA results demonstrated that PK5-Gal-3C significantly enhanced the release of IFN-γ, GzmB, and TNF-α from T cells (p < 0.001). Comparable enhancements in cytotoxic activity and cytokine release were also observed in liver and renal cancer cell lines (Fig. S3). Collectively, these data demonstrated that the fusion protein PK5-Gal-3C directly potentiated the cytotoxicity of T cells. PK5-Gal-3C enhanced the cytotoxicity of T cells in vitro. RTCA and ELISA were performed to evaluate the killing efficacy of T cells and the secretion of cytokines IFN-γ, GzmB, and TNF-α by T cells in the presence or absence of PK5-Gal-3C. PK5-Gal-3C promoted the death of target cells and significantly increased the release of IFN-γ, GzmB, and TNF-α, including HCT-8 (A and B) and HCT116 (C and D). (E) Schematic diagram of the mouse model. The subcutaneous xenograft model NPG mice were constructed using the OSRC-2 cells, followed by treatment with T cells every two weeks, which were pre-incubated with PK5-Gal-3C for consecutive 5 days. (F) Tumor growth curves were recorded every three days. (G and H) Tumors were excised and weighed at the end of experiments (n = 6). ***p < 0.001 To further investigate the direct impact of PK5-Gal-3C on T cell effector functions, activated T cells were incubated with PK5-Gal-3C for a duration of five consecutive days. Subsequently, these T cells (1 × 107 per mouse) were administered via tail vein injection into tumor-bearing NPG mice (Fig. 3E). As shown in Fig. 3F–H, T cells pretreated with PK5-Gal-3C significantly suppressed tumor growth, yielding an average tumor weight of 0.23 g, in contrast to 0.56 g observed in control group. Collectively, these findings indicated that PK5-Gal-3C directly augments the anti-tumor efficacy of T cells in vivo. PK5-Gal-3C could bind to N-glycosylated TRPV2 of T cells via the Gal-3C domain The findings presented above demonstrate that PK5-Gal-3C can directly promote the activation, proliferation, and cytotoxic functions of T cells. To further elucidate the mechanisms by which PK5-Gal-3C modulates T cell activity, IP/MS assay was performed to identify potential molecular targets. The mass spectrometry analysis identified that TRPV2 serves as a putative receptor for PK5-Gal-3C (Fig. 4A and B). TRPV2, as a cation-permeable channel, was further validated as an interacting partner through IP results (Fig. 4C). Moreover, Flag-tagged PK5-Gal-3C was incubated with T cells for two h and followed by immunostaining with the antibodies specific to Flag and TRPV2, respectively. IF assay revealed that PK5-Gal-3C treatment enhanced the membrane localization of TRPV2 and demonstrated co-localization of PK5-Gal-3C with TRPV2 at the cell membrane (Fig. 4D). Collectively, all these results confirmed that PK5-Gal-3C could bind to TRPV2, implicating this interaction in the regulation of T cell function. PK5-Gal-3C bound to the cation-permeable channel TRPV2 of T cells via the Gal-3C domain. (A) Schematic representation of IP and LC–MS/MS. T cells were incubated with PK5-Gal-3C, followed by lysis within RIPA buffer for subsequent IP and LC–MS/MS analysis. (B) The top five potential receptors of PK5-Gal-3C identified from the LC–MS/MS results. (C) PK5-Gal-3C was found to bind to TRPV2, as demonstrated by IP and Western Blotting results. (D) T cells were incubated with PK5-Gal-3C with Flag tag and then stained with specific antibodies targeting Flag and TRPV2, respectively. (E) IP assay was performed to analyze the interaction of PK5, Gal-3C, and PK5-Gal-3C with TRPV2 in T cells. (F and G) the NetNGlyc 1.0 tool was utilized to analyze the potential N-glycosylation sites of TRPV2, revealing that the Asn residue at position 570 (Asn-570) exhibited the highest probability of glycosylation. (H) Mutation of Asn-570 blocked the binding of PK5-Gal-3C to TRPV2. The plasmids encoding the wild-type and mutant forms of Asn-570 TRPV2 were transfected into 293 T cells. Subsequently, the cells were incubated with PK5-Gal-3C for further IP and Western blotting assays. I Comparison of molecular weight of wide-type and mutant TRPV2 in the absence or presence of PNGase F We conducted a detailed investigation into the interaction between PK5-Gal-3C and TRPV2. As shown in Fig. 4E, the Gal-3C domain alone demonstrated binding affinity to TRPV2, whereas the PK5 domain did not exhibit such interaction. This observation suggested the potential involvement of TRPV2 glycosylation in mediating its association with the Gal-3C domain. Furthermore, the NetNGlyc 1.0 analysis (https://services.healthtech.dtu.dk/services/NetNGlyc-1.0/) identified a putative N-glycosylation site at asparagine residue, 570 (Asn-570) (Fig. 4F). Notably, Asn-570 is located in close proximity to the cation pore of TRPV2 (Fig. 4G), implying that the glycosylation of Asn-570 may modulate the functional properties of the ion channel. To evaluate the role of N-glycosylation in the binding between PK5-Gal-3C and TRPV2, a site-directed mutation substituting Asn-570 with glutamine (N570Q) was generated. Both wild-type (WT) and mutant (MUT) TRPV2, tagged with c-Myc, were over-expressed in 293 T cells and subsequently incubated with PK5-Gal-3C. Western Blotting analysis revealed that WT TRPV2 presented two distinct bands, whereas the MUT TRPV2 exhibited a single band of lower molecular weight, consistent with the absence of glycosylation. And IP results further demonstrated that PK5-Gal-3C could bind with WT but not MUT (Fig. 4H). Additionally, treatment with PNGase F, an amidase enzyme that cleaves N-linked oligosaccharide chains, could reduce the molecular weight of WT TRPV2, but had no effect on the mutant form (Fig. 4I). Collectively, these findings indicated that N-glycosylation of TRPV2 at Asn-570 is critical for its interaction with the Gal-3C domain of PK5-Gal-3C. Inhibition of TRPV2 restricted the enhancement of effector function in T cells induced by PK5-Gal-3C As reported previously, TRPV2 functions as a thermosensor and is also activated by ligands in both adaptive and innate immune cells [15]. To investigate whether TRPV2 plays a role in modulating T cell function induced by PK5-Gal-3C, we employed SET2, a selective antagonist of TRPV2, to inhibit its activity. As shown in Fig. 5A and B, SET2 could attenuate the cytotoxicity induced by PK5-Gal-3C, reducing it by 62.8% as measured by RTCA. Furthermore, SET2 markedly decreased the secretion of key cytokines, including IFN-γ, GzmB, and TNF-α by T cells (p < 0.001) (Fig. 5C). These findings collectively indicated that TRPV2 was involved in the regulation of T cell effector functions mediated by PK5-Gal-3C. TRPV2 was involved in the regulation of T cell effector function via the activation of c-Jun mediated by PK5-Gal-3C. (A and B) SET2, a specific antagonist of TRPV2, inhibited the cytotoxic function of RTCA T cells activated by PK5-Gal-3C. (C) SET2 reduced the secretion of GzmB, IFN-γ, and TNF-α from T cells stimulated by PK5-Gal-3C. (D) Western blotting was used to measure the phosphorylation of c-Jun and its upstream kinase JNK1/2 induced by PK5-Gal-3C. (E and F) SP600125, a specific inhibitor of JNK1/2, weakened the killing function of T cells induced by PK5-Gal-3C, as demonstrated by the RTCA results. (G) SP600125 inhibited the release of cytokines IFN-γ, GzmB, and TNF-α from T cells induced by PK5-Gal-3C. H Western blotting was used to measure the phosphorylation of c-Jun in the presence of SET2 induced by PK5-Gal-3C. (I) Western blotting was used to measure the phosphorylation of c-Jun in the presence of BAPTA/AM, a permeable calcium chelator, induced by PK5-Gal-3C. The grayscale values of Western blotting bands were analyzed using ImageJ software and normalized against the control group. **p < 0.01, ***p < 0.001 Previous studies have demonstrated that the activation of c-Jun can rejuvenate exhausted T cells, thereby augmenting their anti-tumor efficacy [23, 24]. Notably, our results revealed that PK5-Gal-3C could induce the phosphorylation of c-Jun. Correspondingly, PK5-Gal-3C was observed to promote the phosphorylation of JNK1/2, the upstream kinases responsible for c-Jun activation (Fig. 5D). As illustrated in Fig. 5E and F, SP600125, an ATP-competitive JNK inhibitor, effectively abrogated the cytotoxic activity of T cells against HCT116 cells induced by PK5-Gal-3C, reducing it by 77.8%, as evidenced by RTCA. Furthermore, ELISA assays demonstrated that SP600125 significantly suppressed the PK5-Gal-3C-mediated secretion of IFN-γ, GzmB, and TNF-α (p < 0.001) (Fig. 5G). To further elucidate the involvement of TRPV2 in the activation of c-Jun mediated by PK5-Gal-3C, SET2 was utilized to disrupt the phosphorylation of c-Jun in the presence of PK5-Gal-3C. As shown in Fig. 5H, SET2 partially inhibited the PK5-Gal-3C induced phosphorylation of c-Jun. Moreover, BAPTA/AM, a cell-permeable calcium chelator, also suppressed the phosphorylation of c-Jun induced by PK5-Gal-3C (Fig. 5I). Collectively, these results indicated that TRPV2 facilitates c-Jun activation via the calcium signaling, thereby augmenting the cytotoxic response of T cells elicited by PK5-Gal-3C. PK5-Gal-3C potentiated the anti-tumor efficacy of CAR-T cells in solid tumor According to these results presented above, PK5-Gal-3C exhibits significant capacity to enhance the T cell cytotoxicity both in vivo and in vitro. Consequently, we evaluated the synergistic anti-tumor effects of PK5-Gal-3C in combination with CAR-T cells targeting CAIX within a renal cancer mouse model (Fig. 6A). The positive ratio and cytotoxicity of CAR-T cells analyzed with RTCA were shown in Fig. S4. Notably, administration of PK5-Gal-3C alone did not produce a discernible anti-tumor effect in the NPG mouse model (Fig. 6B-D), which might be attributable to the absence of functional T cells in this system. More importantly, combined treatment of PK5-Gal-3C and CAR-T cells resulted in a substantial inhibition of tumor growth (73.8%) relative to CAR-T cell treatment alone (Fig. 6B-D). Furthermore, these therapeutic interventions did not significantly impact the body weight of tumor-bearing mice (Fig. 6E). Collectively, these findings indicated that PK5-Gal-3C has the potential to potentiate the anti-tumor efficacy of CAR-T cells in solid tumors, highlighting its prospective utility in tumor immunotherapy. Nevertheless, further investigations employing additional solid tumor models are warranted to comprehensively assess the synergistic effects of PK5-Gal-3C and CAR-T cells in future studies. PK5-Gal-3C enhanced the anti-tumor activity of CAR-T cells targeting CAIX in the NPG model of renal caner. (A) Schematic diagram of the NPG mouse model. The subcutaneous xenograft model NCG mice were constructed using the OSRC-2 cells. (B) Tumor growth curves were recorded every three days. The tumor-bearing mice were treated with PBS and CAR-T cells in the presence or absence of PK5-Gal-3C protein, respectively (n = 6). (C and D) Tumors were excised and weighed at the end of experiment. (E) The mouse weight was recorded every three days. ns: no significant. ***p < 0.001 Discussion T cells play a vital role in the host immunity and T cells-mediated anti-tumor therapies. In the present study, we demonstrated that the fusion protein PK5-Gal-3C significantly facilitated the tumor infiltration of T cells and augmented their effector functions, resulting in pronounced anti-tumor efficacy. Furthermore, our findings indicated that PK5-Gal-3C could directly enhance the anti-tumor activity of T cells through its interaction with TRPV2, which promoted the influx of calcium ions. Besides, calcium ions are involved in the phosphorylation of c-Jun, thereby amplifying the cytotoxic response of T cells induced by PK5-Gal-3C. Notably, the combination of PK5-Gal-3C and CAR-T cell therapy markedly suppressed tumor growth. Furthermore, PK5-Gal-3C was able to enhance the secretion of cytokines (IFN-γ, GzmB, and TNF-α) when tumor cells were present (Fig. 3), indicating that PK5-Gal-3C activates T cells in a manner dependent on tumor antigens. Besides CD8 + T cells, Th1-type CD4 + T cells may also contribute to the release of these cytokines, which requires additional confirmation. Moreover, treatment with PK5-Gal-3C did not cause significant changes in the body weight of tumor-bearing mice, aligning with previous findings [13]. These findings suggest that PK5-Gal-3C treatment is unlikely to cause serious side effects. However, further studies are needed to evaluate the safety and pharmacokinetics of PK5-Gal-3C therapy. Typically, T cells are initially activated when their T cell receptors (TCRs) recognize antigens, a process that is then amplified by costimulatory signals and cytokines. This activation leads to T cell proliferation, differentiation, cytokine production, and effector function [25, 26]. Therefore, T cell activation represents the first step in anti-tumor immunity. Here, we found that PK5-Gal-3C could directly activate T cells accompanying the increase in intracellular calcium ions. It is well known that calcium ions can activate calcineurin, which can promote the nuclear translocation of transcription factors, and subsequently upregulate the expression of genes related to T cell activation, differentiation, and cytokines production. Conversely, blocking of calcium ions can inhibit T cell differentiation and effector functions [22, 23, 27]. The increase in intracellular calcium ions induced by PK5-Gal-3C was due to the calcium inflow from extracellular space, suggesting that the plasma membrane calcium-permeable channels were involved. For T cells, numerous types of plasma membrane permeable channels that control calcium ion influx have been identified, including store-operated calcium channels, transient receptor potential channels, purinergic ionotropic receptors, and voltage activated calcium channels, all of which are activated and regulated by complicated concentration and electrochemical gradient [22]. Herein, we demonstrate that the fusion protein PK5-Gal-3C could specifically recognize and bind to TRPV2, a member of the transient receptor potential channel family. Human TRPV2 exhibits the highest expression levels compared to TRPV1, TRPV3, and TRPV4 in T cells [28]. Additionally, TRPV2 functions as a polymodal receptor that converts various environmental stimuli—such as noxious heat, mechanical forces, chemical signaling molecules, and membrane stretch—into an ion flux across the cell membrane [29, 30]. Achieving a temperature as high as 52 °C is generally difficult under both physiological and pathological conditions. However, TMEs often contain an abundance of abnormal cytokines and stressors that might serve as stimulator of TRPV2. It was reported that the inhibition of TRPV2 disrupts TCR receptor signaling and calcium ion pathways in Jurkat cells, an immortalized T lymphocyte cell line originating from T cell leukemia [31]. However, there remains a significant challenge regarding the limited research on the role of TRPV2 in primary T cells. N-glycosylation is one of the important post-translational modifications that plays a crucial role in regulating the protein localization and function [32, 33]. Although there is currently no reported information regarding the glycosylation of human TRPV2, studies on rat TRPV2 and other members of the TRPV family suggest that glycosylation similarly affects the trafficking and function of TRPV2 [34, 35]. Our results showed that PK5-Gal-3C could specifically bind with glycosylated TRPV2. And the Asn-570 was the potential and only N-glycosylated sites. However, the precise N-glycosylation pattern is unknown. Cholesterol, as one of the main components of cell membranes, is reported to serve as endogenous inhibitory modulators of TRPV2 via binding to the vanilloid binding pocket [36]. Weak carbonic acids, such as acetic acid, lactic acid, and CO2, can activate and sensitize TRPV2, whereas weak hydrochloric acid does not have this effect [14]. Additionally, endogenous reactive oxygen species [37] and phosphorylation [38] could also modulate TRPV2. In the present study, the Gal-3C domain was identified as responsible for binding to glycosylated TRPV2, suggesting that full-length Gal-3 may also act as a novel endogenous modulator of glycosylated TRPV2. However, the detailed regulatory functions require further investigation. Although current research indicates that there is no significant binding between PK5 and T cells, PK5 is essential for enhancing the activation ability of the fusion protein PK5-Gal-3C on T cells. Furthermore, previous studies have reported that PK5 is crucial for the anti-angiogenic activity of fusion protein PK5-Gal-3C [13], which may optimize the TMEs and further promote the activation of T cells induced by PK5-Gal-3C. Increasing evidences have reported that over-expression of c-Jun could improve or restore CAR-T cell efficacy [24, 39]. Our findings indicate that PK5-Gal-3C activated c-Jun in a calcium ions-dependent manner to promote the cytotoxicity of T cells. However, the precise mechanisms by which calcium ions activate c-Jun mediating by PK5-Gal-3C require further investigation. Conclusion In this study, we demonstrated that the recombinant fusion protein PK5-Gal-3C exhibited significant anti-tumor activity that is dependent on T cells. PK5-Gal-3C can directly stimulate the activation and proliferation of T cells and enhance their cytotoxicity. These findings are not only beneficial for drug development but also provide novel strategies for synergistic therapy involving the PK5-Gal-3C protein and T cell-targeted immunotherapy. Supplementary Information Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 1380 KB) Acknowledgements The authors thank Qianqian Dai, Cancer Institute, Xuzhou Medical University, for helping with flow cytometry operation and data acquisition. The graphical abstract was created with BioGDP.com [40]. Authors’ contributions Xiaoge Gao, Qing zhang, Xiangye Liu, conceived and designed the experiments. Xiaoge Gao, Guopeng Zhang, Wei Zhang, Xiaohuan Wei, Zhenyu Wang, and Feitong Wang performed the experiments. Xiaoge Gao, Qing zhang, Shishuo Sun, Feitong Wang, and Haiheng Xu analyzed the data. All authors contributed reagents, materials, and analysis tools. Xiaoge Gao and Xiangye Liu wrote the paper. All authors read and approved the final version of the manuscript. Funding This work was supported by the National Natural Science Foundation of China (Grant Nos. 81972242 and 82473333), the Natural Science Foundation of Jiangsu Province (Grant No. BK20150216), the Natural Science Key Project of Jiangsu Provincial Education Department (Grant No.20KJA320006), the Jiangsu Province Social Development Projects (Grant No. BE2023689), the XZHMU-QL Joint Research Fund (Grant No. QL-YB032), the Science and Technology Project of Xuzhou Municipal Health Commission (Grant No. XWKYHT20240123), and the Construction Project of High Level Hospital of Jiangsu Province (Grant No. LCZX202501). Data availability The datasets analyzed in the current study are available on request to the corresponding author. Declarations Conflict of interest The authors declare no competing interests. Ethics approval and consent to participate The animal studies were conducted from March 2023 to June 2025, in accordance with the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines 2.0, and were approved by the Experimental Animal Ethics Committee of Xuzhou Medical University (Permit number: 202303T010). Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Guopeng Zhang and Wei Zhang have contributed equally to this work. Change history 8/19/2026 A Correction to this paper has been published: 10.1007/s00262-026-04499-5 Contributor Information Xiangye Liu, Email: liuxy83@xzhmu.edu.cn. Qing Zhang, Email: qingzhang@xzhmu.edu.cn. Xiaoge Gao, Email: gaoxiaoge@xzhmu.edu.cn. References Associated Data Supplementary Materials Supplementary file1 (DOCX 1380 KB) Data Availability Statement The datasets analyzed in the current study are available on request to the corresponding author.