Anti-PD1 does not improve pyroptosis induced by γδ T cells but promotes tumor regression in a pleural mesothelioma mouse model

Introduction Mesothelioma is an aggressive tumor in the pleural cavity that is difficult to treat. Diagnosis is usually late with minimal treatment options available for the patients and with unfavorable outcomes. However, recent advances in immunotherapy using γδ T cells may have potential against mesothelioma, given its ample tumoricidal and tumor-migratory properties could allow its infiltration to the widespread tumor mass. Thus, we hypothesize that Vδ2 T cells can perform cytotoxic activities against mesothelioma especially when combined with immune checkpoint blocker against PD-1. Methods Human Vδ2 T cells were expanded from peripheral blood mononuclear cells using Tetrakis‐pivaloyloxymethyl 2‐(thiazole‐2‐ylamino) ethylidene‐1,1‐bisphosphonate (PTA) plus IL-2 for 13 days, before used to test for cytotoxicity against mesothelioma cell lines. Mesothelioma-bearing mice was established by Intrapleural administration of mesothelioma cell lines to test for the efficacy of Vδ2 T cells plus anti-PD-1 antibody combination treatment. Pyroptosis was evaluated by cell morphology, western blot analysis, and ELISA experiments. Flow cytometry was used to examine expression of BTN2A1, BTN3A1, PD-L1, PD-L2 on mesothelioma cell lines. Immunofluorescence staining was performed to detect Vδ2 T cells post adoptive transfer and characteristics of pyroptosis in ex vivo mesothelioma tissue sections. Results Indeed, our data demonstrated that Vδ2 T cells killing mesothelioma can be enhanced by anti-PD-1 antibody in vitro, especially for high PD-1 expressing cells, and in vivo in the intrapleural mesothelioma mice model established by us. Adoptive transfer of Vδ2 T cells into these mice leads to tumor regression by 30-40% compared to control. Immunofluorescence of the tumor section confirmed infiltration of Vδ2 T cells into the tumor, especially to cells with BTN2A1 expression (a Vδ2 T cell activating molecule) despite PD-L1 co-localization. Interestingly, these cells co-expressed cleaved gasdermin D, suggesting that pyroptosis was induced by Vδ2 T cells. This was verified by Vδ2 T/mesothelioma co-culture experiments demonstrating membrane ballooning morphology, increased cleaved caspase-3 and gasdermin E, and upregulated IL-1β and IL-18. Discussion Vδ2 T cells plus anti-PD1 exhibited cytotoxicity against mesothelioma in vivo. However, we found no advantage for anti-PD-1 against PD-1 high expressing Vδ2 T cells in promoting pyroptosis. Taken together, our work demonstrated that Vδ2 T cells combined with anti-PD-1 antibody can be developed as a potential combination immunotherapy for mesothelioma.


Introduction
Mesothelioma is an aggressive cancer that occurs in the mesothelial lining of the pleura, pericardium, and peritoneum (1).It is primarily caused by exposure to asbestos in construction materials during the 1950s, leading to the transformation of mesothelium cells into tumor cells with a latency period of over 30 years (1)(2)(3).In 2020, at least 26,000 mesothelioma-related deaths were reported globally (4), and the incidence and mortality rates are expected to rise, especially in undeveloped countries or cities where asbestos is still used.Unfortunately, the survival rate for mesothelioma patients remains low, and the available treatment options, including chemotherapy, surgical resection, and certain immunotherapies, only provide limited improvements in patient lifespan (5).Moreover, traditional therapies often result in unsatisfactory outcomes, with serious complications such as empyema leading to death (6,7).Additionally, the lack of accurate and reliable biomarkers for mesothelioma detection hinders the widespread use of advanced immunotherapies like CAR CD8 + T cells, which require high antigen specificity (8).In light of these challenges, alternative natural immune cells with tumoricidal properties that do not rely on antigen recognition may hold potential against mesothelioma.
Pyroptosis is a form of programmed cell death mediated by inflammatory caspases like Caspase 1, 3, 4, 5, 11, which enables an inflammatory response and the release of active IL-1b and IL-18 (7,9).Pyroptosis is characterized by membrane blebbing, where cleaved gasdermin D and E proteins form membrane pores that induce cell swelling, rupture, and the release of cytokines (10)(11)(12) The cleavage of gasdermin D is mediated by the non-canonical activation of caspase 4, 5 or 11, while gasdermin E is cleaved by canonical pathway activated caspase 1 or 3 (10,(12)(13)(14)(15)(16).Pyroptotic cell death activates anti-tumor immune responses, making it a focus of cancer treatment research.Recent studies have shown upregulated pyroptosis-related genes in mesothelioma compared to other cancers, correlating with the susceptibility of mesothelioma cells to pyroptosis induction (15,17).Therefore, utilizing Vd2 T cells in our model may hold therapeutic potential for inducing pyroptosis and improving clinical outcomes for mesothelioma patients.
The Vd2 subset of gamma-delta T cells (gd T cells) possesses ample cytotoxicity against various cancers, including cholangiocarcinoma, pancreatic cancer, and lung cancer (18)(19)(20).We recently shown in a nasopharyngeal carcinoma mice model that Vd2 T cells can infiltrate into tumor mass, particularly to areas of cells that express BTN2A1/ BTN3A1 (21).These molecules facilitate the presentation of phosphoantigen (pAg) and subsequent activation of Vd2 T cells, with increased pAg found in tumor cells.Importantly, Vd2 T cells can be robustly expanded in vitro using a prodrug called tetrakispivaloxloxymethyl 2-(thiazole-2-ylamino) ethylidene-1,1bisphosphonate (PTA), which exhibits cytotoxicity against mesothelioma through three distinct mechanisms (21)(22)(23).
Despite the advantages of Vd2 T cells, they can become "exhausted" in the tumor microenvironment due to immune checkpoint molecules such as PD-1 and the ligands PD-L1 and PD-L2.Immune checkpoint inhibitors like nivolumab (anti-PD-1 antibody) or durvalumab (anti-PD-L1 antibody) are used clinically to restore the cytotoxic function of immune cells by blocking the PD-1/PD-L1 interaction (9,24).These inhibitors have shown to extend the lifespan of mesothelioma patients by at least 10 months (25)(26)(27)(28)(29)(30).Thus, we hypothesized that combining anti-PD-1 antibody with Vd2 T cells can enhance the efficacy against mesothelioma.
Our data shows that anti-PD-1 antibody (nivolumab) can enhance the anti-tumor ability of Vd2 T cells against mesothelioma in vitro in a pleural mesothelioma mice model, especially those cells with high PD-1 expression.Immunofluorescence staining of tissue sections revealed an increased number of tumor infiltrating Vd2 T cells.Interestingly, liveimaging of Vd2 T cells co-cultured with mesothelioma showed the induction of pyroptosis in the cells, which was confirmed by the detection of the active Caspase 3 and gasdermin E protein expressoion, as well as increased IL-1b and IL-18.However, the pyroptotic effect was not enhanced by anti-PD-1 antibody.

Cell lines
Human mesothelioma cell lines MSTO-211H (hereafter referred to as MSTO) and NCI-H2052 cells (hereafter referred to as H2052) were purchased from ATCC, were cultured with RPMI 1640 medium (ATCC modification) (Cat.no.A1049101, GIBCO) supplemented with 10% fetal bovine serum (FBS) (Cat.no.10270106, GIBCO).Cells were incubated at 37°C in 5% CO 2 .MSTO and H2052 cells are derived from the lung of male patients who suffered from biphasic and stage 4 mesothelioma, respectively.

Construction of luciferase reporter mesothelioma cell lines
pLV-Fluc-mCherry-Puro plasmid encoding luciferase reporter gene and mCherry gene (provided by Yue Jianbo, City University of Hong Kong) and the two packaging plasmids, pMD2.G (Cat.no.12259, Addgene) and psPAX2 (Cat.no.12260, Addgene) were cotransfected into 293T cells to generate lentiviruses.After 48 h posttransfection, the supernatant containing lentiviruses was collected and used to transduce the luciferase gene into MSTO or H2052 cells.Polybrene (10 mg/ml) was added to the cells to improve the lentiviral infection efficiency.After overnight incubation, MSTO or H2052 cells that tested positive for mCherry were selected by replacing culture medium supplemented with puromycin (2 mg/ ml) after 2 days post-infection.Single-cell clones were further obtained by limiting dilution methods, and the luciferase activity was verified by Perkin Elmer EnSight Microplate Reader.The stable luciferase reporter mesothelioma cell lines -MSTO-luc and H2052luc were then established.

Flow cytometry
Cells were collected and washed with FACS buffer (1% FBS in PBS).The cells were then incubated with the appropriate conjugated antibodies, as indicated in Supplementary Table 1 in 100 ml FACS buffer at 4°C for 30 min for surface protein staining.Cells were analyzed using BD FACSCanto ™ II or BD FACSymphony ™ A1 flow cytometer.

Cytotoxicity assay
Mesothelioma cells were pre-labelled with 2 mM Calcein AM (Cat.No. C3100MP, Invitrogen).Luciferase reporter mesothelioma cells were co-cultured with Vd2 T cells with or without pretreatment of nivolumab (Anti-PD-1 antibody).Different ratios between Vd2 T (effector): mesothelioma cells (target) (E:T) were performed in U-round bottom 96 well plate (Cat.No. 168136, Thermo Fischer Scientific) at 37°C in 5% CO 2 incubator for 4 h.1% Triton X-100 was used as the positive control to induce maximum cell death.After centrifuging the plate at 400 xg for 5 minutes, the supernatant was transferred to a black 96-well plate, followed by measuring the released Calcein fluorescence signal at excitation wavelengths 495 nm and emission 515 nm wavelengths.Cell lysis percentage was calculated by (calcein of sample release−spontaneous release) (maximum release−spontaneous release) Â 100 %.Maximum release refers to the positive control.Spontaneous release refers to Calcein-labelled target cells only.

Live cell imaging using Incucyte S3
To monitor total live and dead mesothelioma cells, mesothelioma cells were pre-labelled with 2 mM Calcein AM.Reporter mesothelioma cells were co-cultured with Vd2 T cells with or without nivolumab pre-treatment at different E:T ratios in a 96-well flat bottom culture plate (Cat.No. CS016 -0096, ExCell Bio) and incubated at 37°C in 5% CO 2 incubator.To track cell death, 250 nM Cytotox Red Reagent (Cat.No. 4632, Sartorius) was added to the culture medium, and the cells were monitored over a 4hour period.Images were captured at regular intervals of 20-30 minutes using the IncuCyte S3 Live-Cell Analysis System (Sartorius) and analyzed using the Incucyte software (Sartorius).

Enzyme-linked immunosorbent assay
Different ratios of Vd2 T cells to MSTO-luc and H2052-luc cells were co-cultured.Vd2 T cells were pre-treated (or not) with nivolumab for 6 hours in a 96-well plate at various ratios.Controls include tumor cells only, raptinal or terfenadine treatment.After co-culture, the plate was centrifuge at 400 xg for 5 min before collecting the supernatants.Supernatants were stored at -80°C until they were used for the IL-1b ELISA kit, following the manufacturer's instructions (Cat.No. HSLB00D, R&D Systems).Optical density was measured at 450 nm with the wavelength correction at 540 nm using the microplate reader (BioTek Absorbance Microplate Reader), and the concentration of IL-1b in the samples was calculated.

Immunofluorescence staining of tumor tissue sections
Tumor were fixed with 10% formalin solution (Cat.no.HT501128, Sigma-Aldrich) followed by dehydration of a series of 70%, 80%, 95%, 100% ethanol and xylene (Cat.no.1330-20-7, RCI labscan), as well as paraffin (Cat.no.P3808, Sigma-Aldrich) for embedding.5 mm tumor sections made using the microtome (Shandon Finesse 325 Rotary Microtome, Thermo Fisher Scientific), placed onto adhesive microscope glass slides (Cat.No. 0810501, Marienfeld) and kept at room temperature and in the dark until used.Tumor tissue sections were dewaxed, and antigen retrieval was performed using citrate-based antigen unmasking solution (Cat.No. H-3300, Vector Laboratories).Following blocking with 10% normal goat serum, the tumor sections were stained with unconjugated primary antibodies and conjugated secondary antibodies shown as Supplementary Table 1.Images were acquired by Stellaris confocal microscope (Leica).Counting of nucleated cells was based on Hoechst 33258 staining, with combinations three independent experiments.

Statistical analysis
Statistical analyses were performed using Student's t-test, oneway or two-way analysis of variance (ANOVA), unless otherwise indicated.P< 0.05 is considered statistically significant.

Human Vd2 T cells exerts cytotoxicity against mesothelioma cell lines
Freshly isolated human PBMCs were used to expand Vd2 T cells for 13 days using PTA and IL-2 following the previously described protocol (21), where cell clusters are formed and expanded (Supplementary Figure 1A), with cell numbers that can be increased by 100-1000 fold (Supplementary Figure 1B).The purity of the CD3 + Vd2 + cells reached approximately 80% after expansion and was further enriched to >95% using microbeads (Supplementary Figures 1C-F).These cells were then used for cytotoxicity assays against two human mesothelioma cell lines with transduced luciferase expression (MSTO-luc and H2052-luc) (Supplementary Figure 2).Considering that PD-1 serves as an indicator of "exhausted" function in Vd2 T cells, we first analyzed its expression by flow cytometry (Figure 1A).The median expression of PD-1 was found to be ~5.99%,which was used as the cut-off to distinguish PD-1 lo and PD-1 hi Vd2 T cells.As shown in (Figures 1B, C), PD-1 lo Vd2 T cells exhibited cytotoxic responses up to ~25% against MSTO-luc and H2052-luc.In contrast, PD-1 hi Vd2 T cells resulted in ~12% and ~10% of cytotoxicity against MSTO-luc and H2052-luc, respectively.Thus, we tested the effect of pre-treating the cells with aPD-1 (nivolumab) to improve the cytotoxic functions.Indeed, aPD-1 could boost the cytotoxic effect of PD-1 hi Vd2 T cells against MSTO-luc and H2052-luc by ~2-fold significantly (Figure 1B).However, the antibody only modestly enhanced cytotoxicity without statistical significance (Figure 1C).Therefore, determining the level of PD-1 expression on Vd2 T cells is crucial in justifying the use of anti-PD-1 immune checkpoint inhibitors as immunotherapy for mesothelioma.

Vd2 T cells plus aPD-1 retarded mesothelioma tumor growth in vivo
To test whether the combination of Vd2 T cells and aPD-1 could be effective in vivo, we first established a mouse model of pleural mesothelioma.We injected NOD.Cg-Prkdc scid Il2rg tm1Wjl /SzJ (NSG) immunodeficient mice, aged 4-6 weeks, with 5 x 10 6 MSTOluc or H2052-luc cells intrapleurally (i.pl).In vivo imaging showed xenograft detection in the upper body as early as two days postinjection, with signals increasing over time (Figures 2A-C, Supplementary Figures 3A, 4A).Tumor masses were observed in the pleural cavity, mesothelium, pleural lining, and pericardial lining (Supplementary Figures 3C, 4C).
Having established this model, we aimed to test the effectiveness of Vd2 T cells combined with or without aPD-1 in these MSTO-luc and H2052-luc bearing mice.To validate the in vitro cytotoxicity data, we injected PD-1 lo and PD-1 hi Vd2 T cells into the MSTO-luc bearing mice on day 3 and/or day 6 post-tumor injection (Figure 2A).One dose of intravenous (i.v.) injection of PD-1 lo or PD-1 hi Vd2 T cells reduced tumor growth by ~48% and ~36%, respectively, as measured by tumor luciferase activity (Figures 2B, C).However, two doses of PD-1 lo Vd2 T cells further reduced tumor by another ~10%, whilst PD-1 hi Vd2 T cells improved it by ~12% (Figure 2C).Next, we sought to determine the effect of intraperitoneal (i.p.) aPD-1 treatment in combination with a 2dose Vd2 T cell adoptive immunotherapy in MSTO-luc mice in a separate experiment (Figure 2A).As shown in Figure 2D, two doses of PD-1 hi Vd2 T cells with aPD-1 resulted in ~58% reduction in tumor growth compared to PBS control.Two doses of PD-1 lo Vd2 T cells with aPD-1 resulted in ~67% reduction in tumor growth significantly.Control mice with tumor survived for 17 days, but those that received the PD-1 lo or PD-1 hi Vd2 T cells with aPD-1 treatments survived for up to 21 days (Figure 2E), despite the significant tumor regression.
In the H2052-luc tumor-bearing mice, we tested the effectiveness of aPD-1 treatment in improving PD-1 hi Vd2 T cells adoptive immunotherapy (Figure 2F).Two doses of PD-1 hi Vd2 T cells showed an advantage over one dose in reducing tumor growth, by ~37% and ~19%, respectively.Further, one and two doses of PD-1 hi Vd2 T cells/aPD-1 were able to decrease tumor growth by ~28% and ~49%, respectively.These data suggest that immune checkpoint blockade for Vd2 T cells with high PD-1 expression confers a better prognosis for decreasing mesothelioma tumor mass.However, survival was only prolonged for one day with two doses of combined treatment in 33% of mice compared to control (Figure 2G).For the mice experiments, no significant weight loss was detected (Supplementary Figures 3B, 4B).

Expression of Vd2 T cell activating and inhibitory molecules in mesothelioma
BTN2A1/BTN3A1 are essential molecules for the activation of Vd2 T cells (32), as we have previously shown their detection in solid tumors (21).As the results suggest that Vd2 T cells have certain effectiveness against mesothelioma, we next examined the expression of BTN2A1, BTN3A1, and PD-L1 in mesothelioma tumor tissue sections obtained from the mice experiments described earlier, at the time of death between 17-21 days.
As shown in Supplementary Figure 5A, BTN2A1, BTN3A1, and PD-L1 appears to express differentially on cells and in different regions by immunofluorescence and tile scan from confocal microscopy.Vd2-TCR + cells were observed to have infiltrated into the tumor, with the cells present in 14% of the tumor area for the twodose treatment, compared to 9% for the one-dose treatment (Figure 3A).Moreover, Vd2-TCR + cells seemed to be located in regions where tumor cells expressed BTN2A1 (Figure 3A) and were found to coincide with cells that co-expressed PD-L1 (Supplementary Figure 5A).However, there are also regions in the tumor that only expressed BTN2A1, BTN3A1/PD-L1 or BTN2A1/BTN3A1/PD-L1 (Supplementary Figure 5A).In the tumors of mice that also received aPD-1 treatment along with Vd2 T cells, Vd2 T cells tended to localize to BTN2A1 + cells, similar to those mice that received Vd2 T cells only (Figure 3A).To further illustrate the characteristics of the tumor cells in the microenvironment, we quantified the cells expressing combinations of BTN2A1, PD-L1, and/or PD-L2 in the tile scan of the tumor section from the two-dose treatment (Supplementary 5B).Out of the 24,636 nucleated cells counted, 29.7% did not express these markers, while 26.8% expressed all three markers.Interestingly, cells expressing BTN2A1 or BTN3A1 alone accounted for 27.5% and 2.0% of cells, respectively.5.9% of cells coexpressed BTN2A1/BTN3A1, suggesting that Vd2 T cells with full functional capacity could potentially target 32.7% of cells in the tumor if unhindered by the presence of PD-L1 or blocked by aPD-1.By flow cytometry, we analyzed MSTO-luc and H2052-luc cell lines for the expression of BTN2A1, BTN3A1, PD-L1 and PD-L2 (Figure 3B), BTN2A1 expression accounted for ~6.2% of MSTOluc and ~4.6% of H2052-luc cells (Figure 3C).However, only<1% of cells co-expressed surface BTN2A1 and BTN3A1 (Figure 3D).A higher frequency of cells with PD-L1 expression was found in BTN2A1 -(MSTO-luc: ~20%, H2052-luc: ~20%) compared to BTN2A1 + (MSTO-luc: ~0.91%, H2052-luc: ~0.79%) subpopulations (Figure 3E).PD-L2 expression was found on ~2.1% of BTN2A1 + and ~3.4% of BTN2A1 -cells for MSTO-luc, and ~1.5% of BTN2A1 + and ~8.0% of BTN2A1 -cells for H2052-luc (Figure 3F).The expression of these molecules may explain the difference in the susceptibility of the cell lines towards Vd2 T cell cytotoxicity (Figure 1).Moreover, we considered the expression of PD-1 on the mesothelioma cells (33), but it was found in only ~10% of cells (Figure 3G), and the expanded Vd2 T cells exhibited an average 1.7 ±2.2% PD-L1 expression.Therefore, our analysis reveals that the expression of BTN2A1/BTN3A1 is low on the tumor cells in mesothelioma, where a high percentage of cells express PD-L1/PD-L2, which could hinder the effectiveness of Vd2 T cells for immunotherapy.This suggests the necessity of aPD-1 co-treatment.

Vd2 T cells induce pyroptotic cell death in mesothelioma cells
By performing live-imaging of the co-culture between Vd2 T cells and mesothelioma cells, we observed the membrane blebbing (or "ballooning") phenotype while the tumor cells were undergoing cell death (Supplementary Video 1 and Supplementary Figure 6), which is indicative of pyroptosis.To confirm whether Vd2 T cells are capable of inducing pyroptosis, western blot was used to determine whether the cleavage forms of gasdermin D (GasD), gasdermin E (GasE), caspase 3, 4 could be induced by co-culture of Vd2 T cells with MSTO-luc cells.Tumor cells only served as negative control.Raptinal and terfenadine treatments served as positive controls for caspase 3 and caspase 4 activation, respectively.After 6 h of co-culture, there was increased level of cleaved caspase 3, particularly at the 10:1 E:T ratio (Figure 4A).aPD-1 treatment resulted in higher level of cleaved caspase 3 for the 10:1 ratio, but to a lesser extent with 5:1 and 2.5:1 ratios (Figure 4A).While cleavage of GasE (which can be mediated by caspase 3) seems to have a corresponding effect at 10:1 (Figure 4A).Band intensity values and the ratio of cleaved GasE over full-length GasE, and other proteins, are shown in Supplementary Figure 7.There was a small increase in cleaved GasD compared to control, particularly at the 5:1 ratios for Vd2 T and/or aPD-1 (Figure 4A).However, cleaved caspase 4, and full-length forms of caspase 3, caspase 4, GasD and GasE were similar between the treatments (Figure 4A), suggesting that coculture of Vd2 T and MSTO-luc induced the active form of GasE likely due to caspase 3, and active form of GasD not due to caspase 4. While raptinal induced clear caspase 3 and GasE activation, terfenadine was not successful for caspase 4 and GasD in the mesothelioma cell lines (Figure 4A).In contrast, the effect on the active form of IL-18 was more obvious at 10:1 ratio co-culture, with a coupled decrease of pro-IL-18 (Figure 4B).However, aPD-1 did not increase the level of active IL-18 detected even at the 10:1 ratio.In concordance, there was an increased level of IL-1b released at 10:1 ratio compared to target cells alone, but addition of aPD-1 treatment had no improvement (Figure 4C).For H2052-luc cocultured with Vd2, cleavage of caspase 3, GasE and IL-18 can be found only at the 10:1 ratio but not for GasD, caspase 4 or IL-1b (Supplementary Figures 8A, B).Interestingly, Vd2 T cells did not induce cleaved form of caspase 3 in nasopharyngeal carcinoma (NPC) cell lines (Supplementary Figure 9).Taken together, Vd2 T cells can induce pyroptosis in mesothelioma cells via the canonical pathway but only induce active IL-18 and IL-1b in MSTO-luc but not H2052-luc cells, which may suggest the higher resistance of H2052-luc cells towards Vd2 T cell killing.
To verify this in vivo, we examined H2052-luc tumor tissue sections from control mice or mice that received one or two doses of PD-1 hi Vd2 T cell and/or aPD-1 treatment, for the expression of Vd2-TCR and cleaved GasD from different treatment groups.Indeed, higher frequency of cleaved GasD was found adjacent to Vd2-TCR + cells in those that received two-dose treatment (19%) compared to one-dose treatment (8%) (Figure 5).However, we did not find noticeable difference for one or two doses of Vd2 T cell treatment with aPD-1 to without aPD-1 (Figure 5).These data may suggest that the greater effect of tumor regression could be related to the level of pyroptosis induced by Vd2 T cells.

Discussion
Mesothelioma, which commonly affects the pleural cavity, poses a challenge for treatment due to its widespread nature and late-stage diagnosis.While immune checkpoint inhibitors have shown limited efficacy in treating this cancer (25,26,34), another promising approach is the use of chimeric antigen receptor (CAR)-T cell therapy, which has demonstrated high effectiveness against leukemia and certain solid tumors like neuroblastoma (35,36).However, the lack of well-defined antigens specific to mesothelioma poses a significant obstacle in successfully implementing CAR-T cell therapy for this cancer.Nevertheless, clinical trials combining aPD-1 antibodies with CAR-T cells have shown promise, with median overall survival of mesothelioma patients potentially extended to around 20 months (37).
Alternatively, gamma-delta T cells have garnered attention as a non-MHC-restricted cell therapy for cancer due to their inherent tumor-killing and tumor-migratory properties (38).In the context of mesothelioma, a recent study demonstrated that Vd2 T cells, when protection in CT-26 tumor-bearing mice (43), and TIM-3 has been implicated in inhibiting the cytotoxic function of Vd2 T cells by suppressing the release of granzyme B and perforin (44).Moreover, LAG-3 and PD-1 have been found to synergistically promote tumoral immune escape in melanoma and colon adenocarcinoma (45).Therefore, targeting these immune checkpoint molecules could help ensure the optimal function of Vd2 T cells in vivo.
The induction of pyroptosis in mesothelioma cells by Vd2 T cells is surprising.Pyroptosis is a type of inflammatory cell death that is characterized by the onset of inflammasome leading to the activation of caspase 3. Subsequently, cleaved caspase 3 activates GasE proteins, which migrate to the cell membrane, forming pores that allow osmosis and the formation of membrane blebbing due to cell swelling, ultimately resulting in cell rupture (10,12).Simultaneously, the activated inflammasome caspase 1 cleaves pro-IL-18 and pro-IL-1b, releasing their active forms outside the cell through the GasE pore (10,12,14,16).Alternatively, the non-canonical pathway can also lead to the formation of membrane pores through the activation of caspase 4, 5, or 11, which cleave GasD.In our study, we observed that co-culture of Vd2 T cells with MSTO cells induced pyroptosis, as evidenced by the expression of caspase 3, GasE, GasD, IL-18, and IL-1b, which was confirmed through live-imaging.However, we did not observe stimulation of caspase 4 cleavage by Vd2 T cells, suggesting that the cleaved GasD in MSTO cells may be mediated by caspase 5, caspase 8, or other mechanisms.While activation of pyroptosis has been demonstrated using various small molecules, chemotherapy drugs, or cell death inducers, this is the first instance where cell-cell interaction leading to pyroptosis has been shown for innate immune cells, particularly Vd2 T cells.It would be worthwhile to investigate the cell surface proteins responsible for activating pyroptosis, as this knowledge could provide insights into enhancing the effectiveness of immune cell therapy by incorporating such proteins into the arsenal for combating cancer cells.

1
FIGURE 1 Anti-PD-1 enhances PD-1 hi Vd2 T cells cytotoxicity towards mesothelioma cell lines.(A) PD-1 expression on CD3 + Vd2 + cells analyzed by flow cytometry shown as a violin plot.Red line indicates a cut-off of 5.99% as the median.(B) PD-1 hi Vd2 T cells and (C) PD-1 lo Vd2 T cells were cocultured with luciferase reporter-transduced mesothelioma cell lines (MSTO-luc and H2052-luc) at different effector: target (E:T) ratios in a cytotoxicity assay.Data represents mean ± SEM from ≥ 3 independent experiments.Student's t-test was performed.*P< 0.05, **P< 0.01.

2
FIGURE 2 Vd2 T cells and anti-PD-1 retard mesothelioma tumor growth in vivo.(A) Timeline of both MSTO-luc and H2052-luc mice experiment indicating the day for cell or aPD-1 injection, or measuring tumor size based on luciferase activity in tumor bearing mice.(B) Representative images of luciferase activities of MSTO-luc and H2052-luc bearing mice receiving PD-1 lo Vd2 T with or without aPD-1 at 1-dose or 2-dose regimen.PBS injection served as control.(C, D) Tumor growth measured by luciferase activities over time compared to day 2 post MSTO-luc tumor injection for different treatment groups.Each group contains 3 to 6 mice.(E) Kaplan-Meier plot for MSTO-luc mice receiving 2-dose of PD-1 lo or PD-1 hi Vd2 T with aPD-1 over time.(F) Tumor growth of H2052-luc mice receiving one or two doses of PD-1 hi Vd2 T with or without aPD-1 over time, with survival as Kaplan-Meier plot (G).Each group contains 4-5 mice.Data represents mean ± SEM.Two-way ANOVA statistical test was used for analyzing tumor growth curves.*P< 0.05, **P< 0.01, ***P< 0.001.

4
FIGURE 4 Vd2 T cells induce pyroptosis in mesothelioma cells.(A) Analysis of protein expression level of gasdermin (Gas)E, caspase 3, GasD, caspase 4, and (B) IL-18, following 6 h of co-culture between Vd2 T cells and MSTO-luc.Controls include raptinal and terfenadine, or cells only.Arrows indicate expected band size of the full-length or cleaved proteins.Numbers under the bands represent band intensity normalized to b-actin.Representative immunoblots are shown.(C) ELISA analysis of IL-1b release following 6 h of co-culture between Vd2 T cells and MSTO-luc at 10:1 ratio.plotted as a column graph of mean ± SEM.Data from three independent experiments are shown.One-way ANOVA statistical test was used.*P< 0.05.