Programmed cell death-1 (PD-1) is an inhibitory receptor and plays an important role in the regulation of αβ T cells. Little is known, however, about the role of PD-1 in γδ T cells. In this study, we investigated the expression and function of PD-1 in human γδ T cells. Expression of PD-1 was rapidly induced in primary γδ T cells following antigenic stimulation, and the PD-1+ γδ T cells produced IL-2. When PD-1+ γδ T cells were stimulated with Daudi cells with and without programmed cell death ligand-1 (PD-L1) expression, the levels of IFN-γ production and cytotoxicity in response to PD-L1+ Daudi cells were diminished compared to the levels seen in response to PD-L1− Daudi cells. The attenuated effector functions were reversed by anti-PD-L1 mAb. When PD-1+ γδ T cells were challenged by PD-L1+ tumors pretreated with zoledronate (Zol), which induced γδ TCR-mediated signaling, the resulting reduction in cytokine production was only slight to moderate compared to the reduction seen when PD-1+ γδ T cells were challenged by PD-L1− tumors. In addition, cytotoxic activity of PD-1+ γδ T cells against Zol-treated PD-L1+ tumors was comparable to that against Zol-treated PD-L1− tumors. These results suggest that TCR triggering may partially overcome the inhibitory effect of PD-1 in γδ T cells.
Human Vγ2Jγ1.2Vδ2 (also termed Vγ9JPVδ2)-bearing γδ T cells recognize the so-called phosphoantigens, a group consisting of isopentenyl pyrophosphate (IPP) and related metabolites derived from microbial pathogens in a γδ TCR-dependent manner 1–4. One of the most potent naturally occurring phosphoantigens is (E)-4-hydroxy-3-methylbut-2-enyl pyrophosphate (HMB-PP), which is derived from the 2-C-methy-D-erythritol-4-phosphate/1-deoxy-D-xylulose-5-phosphate pathway, an isoprenoid-biosynthetic pathway unique to certain microbes and plants 5, 6; we and others have previously reported that the subset of γδ T cells stimulated by HMB-PP is also potently stimulated by human tumor cells pretreated with nitrogen-containing bisphosphonates (N-BPs) 7, 8. A growing body of evidence shows that N-BPs inhibit farnesyl pyrophosphate synthase downstream of IPP in the mammalian mevalonate pathway 9, 10. It has been suggested that the resulting intracellular accumulation of IPP in the tumor cells allows γδ T cells to recognize the tumor cells 11, 12, although the exact mechanisms whereby this may occur remain to be identified. An increase in intracellular IPP may also occur spontaneously in certain tumor cells 13, 14. Based on these results, it has been proposed that γδ T cells may be involved in surveillance for cellular metabolic stress 15, 16. Activated γδ T cells produce various cytokines including IFN-γ and TNF-α and also exhibit potent cytotoxic activity 17, 18, and thus may serve as potential effector cells against tumors 19.
The membrane protein known as programmed cell death-1 (PD-1) is a member of the immunoglobulin superfamily, which is induced in αβ T cells following antigenic stimulation 20. Upon engagement with its specific ligands PD-L1 and PD-L2 via a characteristic molecular interaction 21, 22, PD-1 delivers an inhibitory signal for TCR-mediated activation 23, 24. PD-1 ligands are expressed on DC as well as various tissue cells, and as such the PD-1/PD-Ls system plays a crucial role in the maintenance of self-tolerance and prevention of potential autoimmunity 25–27. Moreover, we and others have previously reported that the specific cytotoxic activity of αβ T cells against tumor cells expressing PD-L1 was attenuated in several murine model systems 28–30. Accumulating evidence indicates that significant proportions of human tumor cells of various types express PD-Ls 31–33; more recently, we have reported that the expression of PD-L1 in ovarian tumors is a predictive factor for poor prognosis among human ovarian cancer patients 34. These results strongly suggest that the PD-1/PD-Ls system is one of the crucial factors that determine T-cell-mediated immunity against tumors. The expression and function of PD-1 in human γδ T cells have not been explored, however. In the present study, we examined the expression of PD-1 in human γδ T cells and its effects on these cells' effector functions against tumor cells.
Expression of PD-1 and PD-L1 in human γδ T cells stimulated with phosphoantigen
γδ T cells in freshly isolated PBMC from healthy donors expressed minimal or undetectable levels of PD-1 and PD-L1. Upon stimulation in vitro with an optimal concentration of HMB-PP (10 μM), however, significant proportions of Vδ2+ γδ T cells (ranging from about 40 to 95%) showed prominent expression of PD-1 as early as day 3 (Fig. 1A). The PD-1 expression gradually declined thereafter as the numbers of γδ T cells robustly increased; nevertheless, a substantial proportion remained PD-1+ even on day 10, though their expression level was reduced (Fig. 1A). Concomitantly, nearly 100% of γδ T cells from healthy donors exhibited high levels of PD-L1 expression on day 3 of culture (Fig. 1B). PD-L1 expression declined rather rapidly thereafter, and only a negligible proportion remained PD-L1+ on day 10 (Fig. 1B).
To examine whether the expression of PD-1 and PD-L1 is determined by γδ T cells per se or by other third party cells, we stimulated PBMC with the nonpeptide antigen, sorted PD-1+ and PD-1−Vδ2+ cells on day 3, maintained the sorted γδ T cells, and analyzed the expression of PD-1 and PD-L1 on days 6 and 10. As shown in Fig. 2A, the level of PD-1 and PD-L1 expression on the sorted PD-1+Vδ2+ cells declined as the culture progressed and the PD-1−Vδ2+ cells remained negative for PD-1 on day 10. We next examined the expression of PD-1 and PD-L1 during secondary stimulation with HMB-PP. After primary stimulation of PBMC with HMB-PP for 10 days, the cells, of which more than 96% were Vδ2+, were restimulated with various concentrations of HMB-PP. The PD-1 expression was augmented (MFI ca. 260) within 24 h in a dose-dependent manner after restimulation with 10 μM HMB-PP. In stark contrast, expression of PD-L1 was detected at only marginal levels (MFI ca. 25), even among cells restimulated with the highest dose (100 μM) of HMB-PP (Fig. 2B, left panel). Similar results were obtained regardless of whether the γδ T cells came from healthy volunteers or breast cancer patients (Fig. 2B, right panel). To further characterize the properties of γδ T cells with respect to PD-1 expression, we cultured PD-1+Vδ2+ cells which had been sorted on day 3, purified PD-1−Vδ2+ cells on day 10, and analyzed the PD-1 re-induction by HMB-PP. As illustrated in Fig. 2C, γδ T cells which had lost the expression of PD-1 remained negative for PD-1 even after restimulation with an optimal concentration of HMB-PP.
Comparison of cell surface markers and cytokine profiles between PD-1+ and PD-1− γδ T cells
We next examined whether PD-1+ γδ T cells had any unique features that were not shared by their PD-1− counterparts. Ten days after stimulation with HMB-PP, PD-1+ and PD-1− γδ T-cell populations exhibited essentially identical cell surface phenotypes: both were CD27+CD44+CD45RO+CD25−CD62L−, which corresponds to the phenotype of memory γδ T cells (Fig. 3A). Upon restimulation with HMB-PP, the primed γδ T cells produced cytokines, including IFN-γ, TNF-α, and IL-13, in a dose-dependent manner (Fig. 3B). When PD-1+ and PD-1− γδ T-cell populations were stimulated separately, both populations produced substantially comparable amounts of these cytokines (Fig. 3C and Supporting Information Fig. 1). The primed γδ T cells also secreted IL-2 in response to HMB-PP in a dose-dependent manner (Fig. 3D). It is of note that IL-2 was produced predominantly by PD-1+ γδ T cells (Fig. 3E). To determine whether PD-1/PD-L1 engagement was involved in the IL-2 production by γδ T cells, we analyzed the secretion of IL-2 in response to EJ-1 pretreated with zoledronate (Zol), a representative N-BP, in the presence or absence of anti-PD-L1 mAb. As shown in Fig. 3F, the production of IL-2 by PD-1+γδ T cells was not influenced by the presence of mAb. We next examined whether PD-1− γδ T cells that had lost PD-1 expression retained the capacity for IL-2 secretion. PBMC were stimulated with HMB-PP and PD-1+Vδ2+ cells were purified on day 3 and maintained as described in Fig. 2A. On day 10, PD-1+ and PD-1−Vδ2+ cells were separated and restimulated with HMB-PP, and the culture supernatants were examined for IL-2 content. As shown in Fig. 3G, PD-1−Vδ2+ cells failed to produce a significant amount of IL-2.
Ligand engagement of PD-1 on γδ T cells attenuates their effector functions against Daudi cells
To examine the effect of PD-1 engagement on the effector functions of γδ T cells, we established a Daudi cell line stably expressing PD-L1 (Daudi/PD-L1) (Fig. 4A). PD-1+ and PD-1− γδ T cells were isolated from PBMC that had been stimulated with HMB-PP and challenged by Daudi or Daudi/PD-L1. Although PD-1− γδ T cells produced approximately the same amount of IFN-γ in response to Daudi and Daudi/PD-L1, PD-1+ γδ T cells produced significantly less IFN-γ in response to Daudi/PD-L1 than in response to Daudi (Fig. 4B). This effect was reversed, however, by the inclusion of anti-PD-L1 mAb to the culture (Fig. 4B). Similarly, the cytotoxic activity of PD-1+ γδ T cells against Daudi/PD-L1 was significantly lower than that against Daudi; again, this inhibitory effect was reversed by the addition of anti-PD-L1 mAb to the assay culture (Fig. 4C). Essentially the same result was obtained in the standard CD107a degranulation assay (Fig. 4D). These results strongly suggest that the PD-1/PD-L1 system is fully functional in γδ T cells.
Sensitization of tumor cells with Zol overcomes the inhibitory effect of PD-1
We and others have previously reported that pretreating various tumor cells with N-BPs promoted TCR-mediated activation of γδ T cells 7, 8. Here, we examined the effect of PD-1 engagement on γδ T-cell activation in response to Zol-treated tumor cells. As we had anticipated, the cytotoxic activity of PD-1+ γδ T cells against Zol-treated Daudi was markedly enhanced, compared to that against untreated Daudi cells (Figs. 4C and 5A, left panel). Interestingly, PD-1+ γδ T cells exhibited a significant level of cytotoxicity against Zol-treated Daudi/PD-L1, only slightly less potent than that against Zol-treated Daudi; the inhibitory effect was completely reversed by the addition of anti-PD-L1 mAb (Fig. 5A, right panel). To confirm these findings, we conducted similar experiments using T-47-D, a mammary carcinoma cell line that per se lacks the capacity of inducing cytokine production from γδ T cells. T-47-D cells initially expressed no detectable PD-L1, but we established a line stably expressing a high level of PD-L1 (T-47-D/PD-L1) (Fig. 5B). Although primed γδ T cells produced undetectable amounts of IFN-γ in response to either T-47-D or T-47-D/PD-L1 (Fig. 5C), PD-1+ and PD-1− γδ T cells exhibited strong IFN-γ production in response to both T-47-D and T-47-D/PD-L1 pretreated with Zol (Fig. 5D). Although PD-1+ γδ T cells produced slightly less IFN-γ in response to Zol-treated T-47-D/PD-L1 than T-47-D cells, this effect was reversed by the addition of anti-PD-L1 mAb (Fig. 5D). This inhibitory effect was less apparent than that in the case of untreated Daudi/PD-L1 (Fig. 5D, left panel, see also Fig. 4B, left panel). Moreover, PD-1+ γδ T cells exhibited potent cytotoxic activity at essentially identical levels in response to Zol-treated T-47-D/PD-L1 and T-47-D, regardless of the presence or absence of anti-PD-L1 mAb (Fig. 5E). These results suggest that strong γδ-TCR signaling may overcome the inhibitory effect of PD-1 on the anti-tumor activity of γδ T cells against PD-L1+ tumor targets.
Endogenous expression of PD-L1 on tumor cells does not affect anti-tumor activity of γδ T cells
Finally, we investigated whether the expression of endogenous PD-L1 modulates γδ T-cell cytotoxicity against Zol-treated tumor cells. We selected solid tumor cell lines with various expression levels of endogenous PD-L1: PC-3 (prostatic adenocarcinoma), which has strong PD-L1 expression; EJ-1 (bladder carcinoma) and MRK-nu-1 (mammary carcinoma), which have weak PD-L1 expression; and MKN45 (gastric adenocarcinoma), which has no detectable PD-L1 expression (Fig. 6A). When these cells were pretreated with Zol, Jurkat cells functionally expressing γδ TCR recognized the tumor cells in a γδ TCR-dependent manner, confirming the γδ TCR-mediated recognition of Zol-treated tumor cells (Fig. 6B). We then examined the cytotoxic activity of γδ T cells in response to Zol-treated tumor cells. γδ T cells showed strong cytotoxicity against Zol-treated PC-3 cells, moderate activity against Zol-treated EJ-1, and weaker yet significant activity against MRK-nu-1 cells regardless of the presence or absence of anti-PD-L1 mAb (Fig. 6C). Against Zol-treated MKN45, in contrast, γδ T cells showed no detectable cytotoxicity despite the absence of PD-L1 expression (Fig. 6C). This observation was confirmed by the standard CD107a degranulation assay (Fig. 6D). Since γδ T cells express NK cell stimulatory receptors, we analyzed the tumor cells for the expression of their ligands. As shown in Supporting Information Figs. 2–6, MICA/B, CD112, and ULBP1-4 expression levels were no different after treatment of tumor cells with Zol than before, suggesting that the augmented cytotoxicity of γδ T cells was attributable to γδ TCR-dependent recognition of Zol-treated tumor cells. These results further confirm that γδ TCR-mediated signaling may overcome the PD-1/PD-Ls signaling in certain circumstances.
Although the function of PD-1 has been extensively studied in mouse and human αβ T cells 35–39, little is known about the roles of PD-1/PD-L1 signaling in human γδ T cells. Since the anti-tumor activity of γδ T cells is dependent on cell-to-cell contact, modulation by costimulatory and inhibitory signals may play a role in γδ T-cell responses to tumors. In this study, we revealed that human peripheral blood γδ T cells expressed PD-1 upon stimulation with nonpeptide antigens and that this expression reached a maximum within 3 days, gradually declining thereafter. Since the time course of PD-1 and PD-L1 expression was not influenced by third party cells as evidenced by our cell sorting results, it seems to be an intrinsic property of γδ T cells. It is intriguing that γδ T cells which have once expressed PD-1 and have subsequently lost PD-1 expression are likely to be defective in their susceptibility to PD-1 re-induction.
Although there was no phenotypic difference between PD-1+ and PD-1− γδ T cells, it was found that PD-1+ γδ T cells produced a significantly higher level of IL-2 in response to an optimal concentration of HMB-PP than PD-1− γδ T cells did. This clearly shows that PD-1+ γδ T cells are functionally more active than PD-1− γδ T cells. Since γδ T cells are among the innate immune cells, and since they require IL-2 for their clonal expansion before the CD4+ αβ T cells, a class of adaptive immune cells, produce it, the highly activated PD-1+ γδ T cells may be responsible for the production of IL-2 through an autocrine process at the very early stage of the first line of defense. These highly activated γδ T cells exhibit a high level of natural killing activity, however, and could result in unfavorable immune reactions against normal tissues. It is thus reasonable to hypothesize that functional PD-1 is expressed on the highly active γδ T cells to avoid unnecessary immune reactions against normal tissues expressing PD-L1. Based on the present results, γδ T cells per se express PD-L1 immediately after primary stimulation with phosphoantigens. This PD-L1 expression seems to be a means for γδ T cells to escape the cytotoxic activity mediated by other highly activated PD-1+ γδ T cells.
In addition, most of the PD-1+ γδ T cells induced by primary stimulation lose PD-1 expression as the culture progresses and the resulting PD-1− γδ T cells no longer re-express PD-1 or produce IL-2 under the conditions used in this study. Although PD-1 signaling seems not to be involved directly in the IL-2 secretion pathway, as evidenced by the result of adding anti-PD-L1 mAb to the culture system, PD-1 status is a good marker of fully functional γδ T cells. Since γδ T cells are considered to be on the immune system's first line of defense, it is reasonable for γδ T cells to respond vigorously to primary stimulation, then, after a certain period of time, lose the capacity for IL-2 production, which is required for clonal expansion in an autocrine manner in the early stages of immune response.
It is of note that γδ T cells have γδ TCR-dependent anti-tumor activity in addition to natural killer-like activity. When tumor cells are treated with Zol, the sensitized tumor cells can be recognized by γδ T cells in a γδ TCR-dependent manner. As Jurkat gene transfer experiments have shown, when a strong signal is transduced via γδ TCR, the cytotoxic activity of γδ T cells is augmented and the γδ TCR-mediated signaling seems to partially overcome the inhibitory effects of PD-1/PD-L1 signaling. In fact, endogenous expression of PD-L1 on tumor cells fails to modulate γδ TCR-mediated cytotoxicity of γδ T cells. Since a high level of PD-L1/PD-L2 expression was observed on malignant tumors in patients 34, strong signaling via PD-1/PD-L1 engagement could negatively modulate the γδ TCR-mediated signaling. Even if this is the case, the abrogation of PD-1/PD-Ls engagement by specific mAbs or recombinant proteins 29 may lead to the promotion of anti-tumor activity because PD-1 on highly activated γδ T cells is functional.
This finding is encouraging for the development of cancer immunotherapy using γδ T cells and Zol 40–42. Several clinical trials harnessing human γδ T cells have been conducted based on these cells' potential anti-tumor activity, and beneficial effects of γδ T cells and Zol seemed to have played a role in the prevention of tumor progression 43–46. Recently, it has been reported in a large-scale clinical trial that the addition of Zol to regular endocrine therapy in the treatment of premenopausal patients with estrogen-responsive early breast cancer significantly improved disease-free survival 47. Although the mechanisms underlying these effects remain to be identified, it is likely that the effects are attributable to γδ T cells, at least in part. We and others have also conducted cancer immunotherapy using ex vivo expanded γδ T cells and Zol and have observed beneficial effects 48. The present results strongly suggest that adoptive transfer of human γδ T cells in combination with N-BPs may serve as an effective means of immunotherapy for patients with malignant tumors.
Materials and methods
Phosphoantigens and tumor cell lines
Phosphoantigens used in this study are listed in Supporting Information Fig. 7. Tumor cell lines used in this study are described in Supporting Information Fig. 8.
Stimulation of γδ T cells
Peripheral blood samples were obtained from healthy donors and cancer patients after institutional review board approval and with written informed consent. PBMC were prepared and stimulated with phosphoantigens as described in Supporting Information Fig. 8.
Flow cytometric analysis was performed on a FACSCalibur system (Becton Dickinson, Franklin Lakes, NJ, USA) 49. The following antibodies were used: allophycocyanin (APC)-conjugated anti-CD3, anti-CD25, anti-CD44, anti-CD45RO, anti-CD62L mAbs (BD Pharmingen, San Diego, CA, USA), APC-eFluorTM 780-conjugated anti-CD27 mAb (eBioscience, San Diego CA, USA), fluorescein isothiocyanate (FITC)-conjugated anti-TCR-δ2 mAb (Immunotech, Prague, Czech Republic), anti-PD-1 and anti-PD-L1 mAbs (Medical & Biological Laboratories, Nagoya, Aichi, Japan), Alexa 488-conjugated anti-PD-L1 (Alexa 488-conjugation kit, Invitrogen Life Technology, Carlsbad, CA, USA), R-phycoerythrin (RPE)-conjugated anti-mouse Ig (Dako Denmark A/S, Glostrup, Denmark), PE-conjugated anti-CD107a mAb, isotype control Abs (BioLegend, San Diego, CA, USA), and propidium iodide (PI, Sigma-Aldrich, St. Louis, MO, USA). The gating strategy is depicted in Supporting Information Fig. 9–18.
Cell separation using cell sorter
After stimulation of PBMC with HMB-PP for 3 days, the cells were stained with anti-PD-1 mAb, RPE-conjugated anti-mouse Ig Ab, and FITC-conjugated anti-Vδ2 mAb. PD-1+Vδ2+ cells and PD-1−Vδ2+ cells were purified using a FACSAria cell sorter (Becton Dickinson). The sorted cells were maintained in Yssel's medium containing 100 U/mL IL-2 at 2×106 cells/mL and examined for expression of PD-1, PD-L1, and Vδ2 on days 6 and 10.
Cell separation using MACS beads
PBMC cultured with HMB-PP (10 μM) for 10 days, at which point 96% or more were confirmed to be Vδ2+, or FACSAria-sorted PD-1+Vδ2+ cells cultured until day 10 (1×108 cells each) were incubated in 500 μL of 20 μg/mL anti-PD-1 mAb for 30 min at 4°C and then in 1 mL of 1:100-diluted RPE-conjugated anti-mouse Ig Ab for 30 min at 4°C. After being washed with PBS containing 2% FBS, the cells were incubated with 1 mL of PBS containing 2% FBS and 100 μL of anti-PE MicroBeads (Miltenyi Biotec, Bergisch Gladbach, Germany) for 15 min at 4°C; PD-1+ and PD-1− γδ T cells were then purified using an AutoMACS Pro separator (Miltenyi Biotec) in a depl025 mode according to the manufacturer's instructions.
Analysis of PD-1 re-induction
For analysis of PD-1 re-induction in PD-1−Vδ2+ cells which had lost PD-1 expression, PD-1+Vδ2+ cells sorted on day 3 were cultured until day 10 and Vδ2+ cells having lost PD-1 expression were purified using magnetic beads. The resulting 3×106 PD-1−Vδ2+ cells were stimulated with various concentrations of HMB-PP in 1.5 mL of Yssel's medium containing 10% human AB serum for 24 h, stained with anti-PD-1 mAb, RPE-conjugated anti-mouse Ig Ab, and FITC-conjugated anti-Vδ2 mAb, and examined for the expression of PD-1.
Phenotypic analysis of PD-1+ and PD-1− γδ T cells
PBMC stimulated with HMB-PP for 10 days were 3-color stained with anti-PD-1 mAb, RPE-conjugated anti-mouse Ig Ab, FITC-conjugated anti-Vδ2 mAb, and one of the following mAbs conjugated with APC: CD25, CD27, CD44, CD45RO, or CD62L. Cell populations in a Vδ2+ gate were analyzed and depicted as dot plot diagrams.
Cytokine production in response to HMB-PP
Either bulk γδ T cells or magnetic beads-purified PD-1+ or PD-1− γδ T cells were stimulated with various concentrations of HMB-PP for 24 h in triplicate at 3×106 cells/1.5 mL in Yssel's medium in a 24-well culture plate or 1×105 cells/200 μL in a round bottom 96-well plate. Cytokines (TNF-α, IFN-γ, IL-13, and IL-2) in the culture supernatants were measured using their corresponding ELISA kits (PeproTech, Rocky Hill, NJ, USA) according to the manufacturer's instructions.
Effect of mAb on IL-2 production
After stimulation of PBMC with HMB-PP for 10 days, PD-1+ and PD-1− γδ T cells were separated using magnetic beads. The resulting PD-1+ and PD-1− γδ T cells, 3×106 of each, were stimulated with EJ-1 cells pretreated with 100 μM Zol for 24 h in the presence or absence of 12.5 μg/mL of anti-PD-L1 mAb. The culture supernatants were examined for IL-2 content.
Effect of PD-1 status on IL-2 production
After stimulation of PBMC with HMB-PP for 3 days, PD-1+Vδ2+ cells were FACSAria-sorted and cultured until day 10. PD-1+Vδ2+ and PD-1−Vδ2+ cells were then separated using magnetic beads. The resulting cells, 3×106 of each, were stimulated with 10 μM HMB-PP for 24 h. The culture supernatants were examined for IL-2 content.
Effect of anti-PD-L1 mAb on IFN-γ production
After stimulation of PBMC with HMB-PP for 10 days, PD-1+ and PD-1− γδ T cells were purified using magnetic beads. The 1×105 cells thus separated were stimulated in triplicate with 1×105 Daudi or Daudi/PD-L1 cells in the presence or absence of 12.5 μg/mL of anti-PD-L1 mAb in 200 μL of complete RPMI1640 medium. After 24 h, IFN-γ in the culture supernatants was measured through ELISA according to the manufacturer's instructions.
Either PD-1+ or PD-1− γδ T cells (1×104 cells/well) were incubated in triplicate with tumor cells (1×104 cells/well) in 100 μL of complete RPMI1640 medium in the presence or absence of anti-PD-L1 mAb (12.5 μg/mL) in a 96-well round bottom plate for various periods of time (4, 8, and 12 h). The specific tumor cell lysis was measured using the CytoTox 96 Non-Radioactive Cytotoxicity Assay Kit (Promega, Madison, WI, USA) according to the manufacturer's instructions.
CD107a degranulation assay
PD-1+ γδ T cells (1×105 cells/well) were incubated with tumor cells at an effector-to-target ratio of 1:0.1, 1:0.3, 1:1, or 1:3 for Daudi or Daudi/PD-L1 or 1:0.2, 1:0.6, 1:2, or 1:6 for PC-3, EJ-1, MRK-nu-1, and MKN45 in 100 μL of complete RPMI1640 medium in a 96-well culture plate in the presence of 10 μL of PE-conjugated anti-CD107a mAb at 37×C with 5% CO2. After 4 h of incubation, the cells were stained with 50 μL of 20-fold diluted FITC-conjugated anti-Vδ2 mAb and analyzed using a FACSCalibur flow cytometer.
γδ TCR gene transfer and IL-2 assay
Jurkat cells expressing Vγ2Vδ2-TCR were prepared and examined for IL-2 production as described previously 50. In brief, the transfectant cells at 2×105 cells/100 μL and tumor cells at 2×105 cells/100 μL were incubated in triplicate. After 16 h, the supernatants were examined for IL-2 using a CTLL-2 and Cell Titer Glo reagent (Promega), and the luminescence was measured by an ARVO luminometer (PerkinElmer, Foster City, CA, USA).
Statistical analysis was performed using Student's t-test (p<0.05 is considered significant).
The authors are grateful to Ms. Chiyomi Inoue for excellent technical assistance and to Dr. Shigekazu Nagata, Kyoto University, Japan, for providing pEF-BOS vectors. This work was supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, Culture, Sports, and Technology of Japan (MEXT) (to Y. T. and N. M.), and by “Special Coordination Funds for Promoting Science and Technology” from MEXT and Astellas Pharma Inc. through the “Formation of Center for Innovation by Fusion of Advanced Technologies” program (to Y. T. and N. M.).
Conflict of interest: The authors declare no financial or commercial conflict of interest.