IL-27 and TCRγδ+ T lymphocytes play critical roles in both innate and adaptive immune responses in health and disease, including infection and tumors. Although the activity of IL-27 is well characterized in different human immune cells, no information is available on the role of IL-27 in human TCRγδ+ T lymphocytes. Here, we provide the first evidence that TCRγδ+ T lymphocytes express both gp130 and WSX-1 chains of IL-27R, and that IL-27 may function in TCRγδ+ T cells by (i) inducing STAT1 and STAT3 phosphorylation, (ii) stimulating cytotoxicity against tumor cells through upregulation of cytotoxic granules production, (iii) reducing the release of Th2-related cytokines, such as IL-5 and IL-13, and inducing IFN-γ production, and (iv) upregulating the expression of CD62L. These results highlighted a novel immunoregulatory property of human IL-27 that may be relevant in the immune response against tumors. Our results may offer new perspectives for the development of future clinical trials using IL-27 and TCRγδ+ cells for cancer immunotherapy.
IL-27 is an heterodimeric cytokine of the IL-12 family [[1, 2]] that binds to a heterodimeric receptor composed of the gp130 and WSX-1 chains []. It is predominantly produced by APCs and plays critical roles in the regulation of human T- and B-cell functions through the activation of STAT molecules [[1, 2, 4, 5]]. IL-27 is considered both a pro- and antiinflammatory cytokine due to its ability to induce (i) CD4+ cell proliferation, (ii) early Th1 differentiation, (iii) generation of Tr-1 and CTL, and (iv) suppression of Th-17 and Th-2 cells []. Although IL-27 was extensively investigated in conventional T cells [[2, 5]], its role on TCRγδ+ T lymphocytes remains unexplored. The latter cells, which are mainly Vγ9Vδ2+ in human peripheral blood and poorly represented in physiological conditions (1–5% of circulating lymphocytes), may be strongly activated and expanded by nonpeptide phosphoantigens expressed by transformed or pathogen-infected cells [[6-9]]. In this context, we recently demonstrated that IL-27 acts as antitumor agent by targeting directly human hematological tumors including multiple myeloma, B-acute lymphoblastic leukemia, and B-cell lymphoma of germinal center origin [[10, 23, 24]]. However, it has been reported that TCRγδ+ T lymphocytes kill a vast repertoire of tumor cell lines and primary samples in vitro including leukemia, lymphoma, melanoma, neuroblastoma, and different types of carcinoma, thus raising great interest in targeting TCRγδ+ T cells for cancer immunotherapy. In addition, TCRγδ+ T lymphocytes interplay with conventional T cells, B cells, NK cells and dendritic cells, neutrophils, and macrophages, thus representing a T-cell population with a critical role in both innate and adaptive immunity [[6, 11-22]].
With this in mind, we investigated the functional role of IL-27 on human TCRγδ+ T lymphocytes, either freshly isolated from peripheral blood of normal subjects or expanded in vitro upon PBMC stimulation with zoledronic acid, and asked whether IL-27 could modulate the functional properties of TCRγδ+ T cells.
Results and discussion
TCRγδ+T cells express functional IL-27R
Resting and activated Vγ9Vδ2+ T cells expressed WSX-1 (mean relative of fluorescence intensity (MRFI) ± SD: resting 1.76 ± 0.005, activated 3.97 ± 0.56, Fig. 1A and B) and gp130 (MRFI ± SD: resting 3.11 ± 0.15, activated 2.63 ± 0.02, Fig. 1A and B) chains, thus indicating that both cell populations may be responsive to IL-27. The complete IL-27R was functional in these cells, as witnessed by the ability of IL-27 to significantly induce STAT1 (MRFI ± SD: medium 1.87 ± 0.02, IL-27 13.99 ± 0.24, p < 0.0001), STAT3 (MRFI ± SD: medium 1.56 ± 0.32, IL-27 2.97 ± 0.11, p = 0.006), but not STAT5 (MRFI ± SD: medium 1.25 ± 0.01, IL-27 1.3 ± 0.02) (Fig. 1C and D) phosphorylation. Thus, TCRγδ+ T cells show a similar behavior to classical T lymphocytes in terms of IL-27R expression and IL-27-driven signaling pathway [[1, 2]].
Finally, the significant differences in WSX-1 (p = 0.03) and gp130 (p = 0.05) expression between resting and activated Vγ9Vδ2+ T cells may be conceivably related to the different experimental conditions used, that is, in vitro expansion by zoledronic acid versus direct isolation of TCRγδ+ T cells from peripheral blood (PB). However, such differences did not significantly impact on STAT-1, STAT-3, or STAT-5 activation (not shown) or other functional responses to IL-27 (i.e. cytotoxicity, see below).
TCRγδ+T-cell mediated cytotoxicity is increased by IL-27 treatment
Next, we investigated whether IL-27 may function on TCRγδ+ T cells, both purified from PBMCs and in vitro expanded, by modulating their cytotoxic activity and consequently their ability to kill tumor target cells. To this end, we used two human cell lines as targets: (i) the HTLA-230 neuroblastoma cells that display a low basal sensitivity to TCRγδ+ T cell-mediated lysis and (ii) the DAUDI Burkitt lymphoma cells that show high sensitivity to TCRγδ+ T-cell mediated lysis.
As shown in Fig. 2A, IL-27 pretreatment rendered activated Vγ9Vδ2+ T cells more effective in HTLA-230 cell lysis at different E:T ratios (E:T ratio, percent specific lysis, medium versus IL-27: 50:1, 38.5 versus 55.5, p < 0.001; 25:1, 33.25 versus 46.5, p < 0.01; 12:1, 27 versus 36.5, p < 0.05; 6:1, 18.25 versus 28.5, p < 0.05; 3:1, 13 versus 22.75, p < 0.05). The addition of anti-TCR Vγ9, but not of anti-NKG2D blocking mAb, inhibited target cell lysis, thus indicating that HTLA-230 cell line recognition was mediated by TCR (Fig. 2A, inset). Furthermore, IL-27 pretreatment rendered both resting and activated Vγ9Vδ2+ T cells more effectively against DAUDI target cells (Fig. 2B, E:T ratio, percent specific lysis, medium versus IL-27: activated: 25:1, 80 versus 96, p < 0.001; 12.5:1, 80 versus 96, p < 0.001; 6:1, 69 versus 92, p < 0001; 3:1, 60 versus 91, p < 0.001; 1.5:1, 55 versus 82, p < 0.001; resting: 25:1, 21.5 versus 33.5, p < 0.01; 12.5:1, 16 versus 28, p < 0.01; 6:1, 11 versus 21.5, p < 0.01; 3:1, 6.5 versus 9.5, ns; 1.5:1, 3 versus 3.5, ns). As shown in Fig. 2C and D, IL-27-mediated increase of TCRγδ+ T cell cytotoxicity was closely related to the stimulation of cytotoxic granules production, as demonstrated by significant increase of Granzyme B (MRFI mean ± SD: activated Vγ9Vδ2+ T cells treated with medium versus IL-27 = 84.61 ± 2.29 versus 124.6 ± 12.87, p = 0.04; resting Vγ9Vδ2+ T cells treated with medium versus IL-27 = 63.01 ± 7.57 versus 94.29 ± 16.28, p = 0.04) and perforin (MRFI mean ± SD: activated Vγ9Vδ2+ T cells treated with medium versus IL-27 = 1.29 ± 0.02 versus 3.08 ± 0.09, p = 0.0003; resting Vγ9Vδ2+ T cells treated with medium versus IL-27 = 10.28 ± 0.69 versus 16.14 ± 0.53, p = 0.003). Finally, IL-27 significantly increased Granzyme A in resting Vγ9Vδ2+ T cells (MRFI mean ± SD: medium versus IL-27-treated cells = 12.76 ± 1.05, versus 16.77 ± 2.01, p = 0.04) but not in activated Vγ9Vδ2+ T cells (MRFI mean ± SD: medium versus IL-27-treated cells = 9,43 ± 1.49 versus 10.45 ± 1.19) (Fig. 2C and D).
IL-27 modulates cytokine release and CD62L expression on TCRγδ+T cells
Finally, the IL-27 role on TCRγδ+ T-cell function was investigated in terms of modulation of (i) cytokine release and (ii) expression of chemokine receptors (CXCR3, CCR5, and CCR6), activating/inhibitory receptors (CD16, TCRγδ, NKG2A), and of the adhesion molecule CD62L. These experiments revealed that IL-27 significantly downregulated Th2-type cytokine secretion in activated Vγ9Vδ2+ T cells, as demonstrated by the inhibition of IL-5 (pg/mL ± SD: medium 177.6 ± 34.22, IL-27 108.5 ± 41.02, p = 0.04) and IL-13 (pg/mL ± SD: medium 1969 ± 313.3, IL-27 1382 ± 166.3, p = 0.04) release. All the other cytokines included in the FlowCytomix Multiplex assay such as IL-2, IL-4, IL-17A, IL-22, IFN-γ, IL-10, IL-12p70, IL-6, IL-9, IL-1β, and TNF-α were not significantly modulated by IL-27 (Fig. 3A). In additional experiments using both resting and activated Vγ9Vδ2+ T cells, IFN-γ and IL-10 production was tested by ELISA. These experiments revealed that, accordingly with results obtained using FlowCytomix assay, IFN-γ secretion was very low and not modulated by IL-27 in activated Vγ9Vδ2+ T cells (Fig. 3B, pg/mL ± SD: medium 28.4 ± 3.5, IL-27 37.3 ± 3.04). By contrast, IL-27 significantly increased IFN-γ production in resting Vγ9Vδ2+ T cells (Fig. 3B pg/mL ± SD: medium 359.8 ± 51.8, IL-27 819.6 ± 96.14, p = 0.01). IL-10 was undetectable in both cell populations and not modulated by IL-27 (not shown).
Furthermore, CD62L, a key adhesion molecule involved in transmigration of TCRγδ+ T cells into inflamed tissues, was significantly upregulated by IL-27 (MRFI ± SD; activated Vγ9Vδ2+ T cells: medium 16.96 ± 2,.09, IL-27 21.03 ± 2.91, p = 0.01; resting Vγ9Vδ2+ T cells: medium 141.8 ± 16.8, IL-27 181.9 ± 12.26, p = 0.04). No modulation of activating/inhibitory receptors or chemokine receptors expression was observed (Fig. 3C).
Taken together, our data provide the first demonstration that human TCRγδ+ T cells, both circulating and in vitro expanded with zoledronic acid, express complete and functional IL-27R and that IL-27 may modulate TCRγδ+ T-cell functions, thus highlighting a novel immunomodulatory role of IL-27, that may be relevant in the immune response against tumors. In this context, we recently reported that IL-27 acts as multifunctional antitumor agent against different human hematological malignancies [[23, 24]] that have been reported to be immunotargeted by peripheral blood TCRγδ+ T cells []. Thus, we may envisage that the presence of exogenous IL-27 in the tumor microenvironment may be crucial for dampening tumor progression, by (i) directly inhibiting tumor cell proliferation and angiogenesis [[23, 24]], (ii) driving Th1 differentiation, generating CTL responses and stimulating NK cells [[2, 26]], (iii) stimulating TCRγδ+ T cell cytotoxicity, and (iv) inducing Th1-type factor release (i.e. IFN-γ [[1, 2, 5]]) and downregulating Th2 cytokines (i.e. IL-5 and IL-13) that may, in turn, sustain unfavorable microenvironmental condition for tumor progression. Finally, such antitumor responses may be amplified by TCRγδ+ T cells persistence in the tumor microenvironment, mediated by IL-27-driven upregulation of surface CD62L. A note of caution that need to be taken into account and evaluated in future studies should be that ability of IL-27 to induce differentiation of immunosuppressive Tr1 lymphocytes, as reported in murine models.
Materials and methods
Expansion of TCR Vγ9Vδ2+T cells and isolation of TCRγδ+T cells
This study was approved by G. Gaslini Institute Ethical Committee. PBMCs were obtained from six healthy donors after informed consent and separated by Ficoll density gradient centrifugation (Sigma Aldrich). Cells were resuspended in RPMI 1640 with 10% pooled human AB sera. Activated Vγ9Vδ2+ T cells were obtained by in vitro PBMC stimulation with 5 μM Zoledronic acid (Enzo Life Sciences, Inc.) in the presence of 50 U/mL of human recombinant (hr) IL-2 (PROLEUKIN, Novartis Farma S.p.A) for 10–15 days. Cultures containing more than 95% TCR Vδ2+ cells were used for further studies. Resting Vγ9Vδ2+ T cells were purified as Vδ2+ cells from PBMCs (n = 4) by immunomagnetic selection, using purified anti-Vδ2 mAb (Pierce) as primary reagent and rat anti-mouse IgG1 beads (Miltenyi Biotec), following manufacturer's protocol.
mAbs and flow cytometry
WSX-1 and gp130 expression was investigated on total PBMCs (gating on TCRγδ+ T cells) and activated Vγ9Vδ2+ T cells by flow cytometry. The following mAbs were used: anti-TCRγδ PE (clone #V65, BD Biosciences), anti-WSX1 PE (clone# 191115, R&D System Inc.), and anti-gp130 FITC (clone # B-R3, AbD Serotec). IL-27 signaling pathway was investigated in resting or activated Vγ9Vδ2+cells cultured 30 min with or without hrIL-27 (R&D Systems, 100 ng/mL) using Alexa 488-conjugated anti phospho-STAT1 (clone #58D6), anti phospho-STAT3 (clone #D3A7), and anti phospho-STAT5 (clone #C71E5, Cell Signaling Technology, Inc.) mAbs, as described []. Surface phenotype of resting or activated Vγ9Vδ2+cells cultured 36 h with or without hrIL-27 (100 ng/mL) was investigated using anti-CXCR3 FITC (clone#49801), anti-CCR5 PE (clone#45531, R&D Systems), anti-CCR6 PE-Cy7 (Beckman Coulter), anti-CD16 FITC (clone#LNK16) and anti-CD62L APC (clone#LT-TD180, Immunotools), and anti-TCRγδ PE (clone#V65) mAbs. Purified anti-NKG2D (clone BAT221) mAb was kindly provided by Dr. Cristina Bottino (Università di Genova, Genova, Italy). PE-conjugated goat anti-mouse IgG1 mAb (Beckman Coulter) was used as secondary reagent. Isotype- and florochrome-matched irrelevant mAbs (Beckman Coulter) were used as controls. Cells were run on Gallios cytometer (Beckman Coulter). 104 events were acquired and analyzed using Kaluza software (Beckman Coulter). Results are expressed as MRFI calculated as MFI of specific mAb/MFI of irrelevant isotype-matched mAb.
Cytokine secretion was investigated on supernatants from activated Vγ9Vδ2+ cells cultured 36 h with or without 100 ng/mL hrIL-27, using the Human Th1/Th2/Th9/Th17/Th22 13plex FlowCytomix Multiplex (eBioscience, Inc.), following manufacturer's protocol. Data were collected using Gallios cytometer and analyzed by Flow cytomix software (eBiosciences). IFN-γ and IL-10 production by activated (n = 4) and purified resting (n = 4) Vγ9Vδ2+ T cells treated or not with IL-27 was assessed using ELISA kits by Immunotools.
51Cr-release cytotoxicity assay was performed as described [], using resting or activated Vγ9Vδ2+cells (cultured 36 h with or without 100 ng/mL hrIL-27) as effector cells and the HTLA-230 human neuroblastoma cell line or DAUDI Burkitt lymphoma cell line, as targets. Results are expressed as percent specific lysis. In some experiments, Vγ9Vδ2+ T cells were preincubated with anti-TCR Vg9 (clone 7A5; Pierce Endogen) or anti-NKG2D blocking mAbs (clone 149810; R&D Systems) before being added to 51Cr-labeled target cells.
Intracellular expression of cytotoxic granules was investigated by intracellular staining and flow cytometry on the same effector cells, using PE-conjugated anti-Granzyme B (clone FGB12, Invitrogen), anti-Granzyme A (clone CB9, BD Biosciences), and anti-perforin (dG9, Ancell) mAbs.
Data were analyzed by GraphPad Prism Software 5.0 (GraphPad Software Inc.) using Mann–Whitney test. A p value of less than 0.05 was considered significant.
This work was supported by grants from Associazione Italiana Ricerca sul Cancro (A.I.R.C.) Milano, Italy (grant number 4014 to I.A.), from Finanziamento Ricerca Corrente, Ministero della Salute, anno 2011 and Progetto Strategico Oncologico 2006 rif070701.
Conflict of Interest
The authors declare no financial or commercial conflict of interest.