IL-12 Expands and Differentiates Human Vγ2Vδ2 T Effector Cells Producing Antimicrobial Cytokines and Inhibiting Intracellular Mycobacterial Growth

Abstract

While IL-12 plays a key role in differentiation of protective CD4⁺ Th1 response, little is known about mechanisms whereby IL-12 differentiates other T-cell populations. Published studies suggest that predominant Vγ2Vδ2 T cells in humans/nonhuman primates (NHP) are a fast-acting T-cell subset, with capacities to rapidly expand and produce Th1 and cytotoxic cytokines in response to phosphoantigen (E)-4-hydroxy-3-methyl-but-2-enyl pyrophosphate (HMBPP) produced by Mycobacterium tuberculosis (Mtb) or others. However, whether IL-12 signaling pathway mediates fast-acting and Th1 or anti-microbial features of Vγ2Vδ2 T cells remains poorly defined. Here, we show that IL-12, but not other IL-12 family members IL-27/IL-35, apparently expanded HMBPP-activated Vγ2Vδ2 T cells. Although IL-12 and IL-2 similarly expanded HMBPP-activated Vγ2Vδ2 T-cell clones, the IL-12-induced expansion did not require endogenous IL-2 or IL-2 co-signaling during HMBPP + IL-12 co-treatment. IL-12-induced expansion of Vγ2Vδ2 T cells required the PI3K/AKT and STAT4 activation pathways and endogenous TNF-α signaling but did not involve p38/MAPK or IFN-γ signals. IL-12-expanded Vγ2Vδ2 T cells exhibited central/effector memory phenotypes and differentiated into polyfunctional effector cell subtypes which expressed TBX21/T-bet, antimicrobial cytokines IFN-γ, TNF-α, GM-CSF, and cytotoxic granule molecules. Furthermore, the IL-12-expanded Vγ2Vδ2 T cells inhibited the growth of intracellular mycobacteria in IFN-γ- or TNF-α-dependent fashion. Our findings support the concept that IL-12 drives early development of fast-acting Vγ2Vδ2 T effector cells in antimicrobial immune responses.

Keywords: IL-12; Vγ2Vδ2 T cells; anti-tuberculosis; differentiation; proliferation.

Figure 1

IL-12, but not IL-27 or IL-35, significantly drives the proliferation and expansion of HMBPP-activated Vγ2Vδ2 T cells. (A) Representative flow cytometric plots shown cellular population augmentation (upper panels) and cellular division (lower panels) of Vγ2Vδ2 T cells after 7-day ex vivo co-culture with media, HMBPP (10 ng/ml), IL-12 (25 ng/ml), or the combination of HMBPP + IL-12. Cells gated on Live cells, lymphocytes, single not doublet/triplet, and then CD3⁺ T cells. Vγ2Vδ2 T cells were expressed as percentage in total CD3⁺ T cells. Cellular division or proliferation was determined by the percentage of diluted/lower CFSE fluorescence intensity of Vγ2Vδ2 T cells. (B) Kinetics of absolute number of Vγ2Vδ2 T cells during expansion treated by HMBPP + IL-12 or HMBPP + IL-2. Data shown as mean ± SEM of three independent experiments pooled from 12 healthy controls. (C) Both IL-27 and IL-35 failed to induce the expansion of Vγ2Vδ2 T cells in PBMC followed by HMBPP stimulation. PBMCs from healthy controls were treated by HMBPP (10 ng/ml) with or without either IL-27 or IL-35 (25 ng/ml) for 7 days. Percentages of Vγ2Vδ2 T cells in PBMC were determined by flow cytometry. The bar plot shown the percentages of Vγ2Vδ2 T cells in CD3⁺ T cells were similar in HMBPP + IL-27 or IL-35 vs. HMBPP only cultures. Data shown as mean ± SEM of five independent experiments pooled from 25 healthy controls. NS, not significant (p > 0.05) for statistics. ^****p < 0.0001, when HMBPP + IL-12 group is compared to media group (ANOVA, Dunnett's test). The specific bioactivity of recombinant IL-12 cytokine was validated in our previous publication (24) in the blockade assay using the anti-IL-12 neutralizing mAb, and this anti-IL-12 neutralizing mAb can significantly reduce HMBPP + IL-12 expansion (not shown).

Figure 2

IL-12 and IL-2 similarly expand predominant clones of HMBPP-activated Vγ2Vδ2 T cells, but the IL-12-induced expansion does not require endogenous IL-2. (A–D) A comparison of the percentage of CDR3 length (A,B) and analysis of VDJ sequences (C,D) in Vγ2- and Vδ2-bearing TCR expressed by cells expanded by HMBPP + IL-12 and HMBPP + IL-2, respectively. RT-PCR was used to specifically amplify the CDR3 regions of Vγ2- and Vδ2-bearing TCR cDNA. (E) Endogenous IL-2 is not involved in IL-12 induced expansion of HMBPP-activated Vγ2Vδ2 T cells. The bar graph shows that anti-IL-2 neutralizing mAbs from two sources (αIL-2-BD and αIL-2-RD) fail to block the ability of IL-12 to expand HMBPP-activated Vγ2Vδ2 T cells (left panel), but these anti-IL-2 mAbs not isotype controls (ISO-BD/ISO-RD) significantly reduced the IL-2 expansion of HMBPP-activated Vγ2Vδ2 T cells in the HMBPP + IL-2 culture. PBMC were co-cultured with HMBPP (10 ng/ml) plus IL-2 (5 ng/ml) or IL-12 (25 ng/ml) in the presence or absence of 5 ug/ml anti-IL-2/IL-12 antibody or isotype controls for 7 days. Data shown as mean ± SEM of four independent experiments pooled from 20 healthy controls, ^****p < 0.0001, t-test.

Figure 3

The IL-12-induced expansion of HMBPP-activated Vγ2Vδ2 T cells requires endogenous TNF-α, but not IFN-γ. Pooled graph data of the percentage of Vγ2Vδ2 T cells in CD3⁺ T cells expanded after the 7-day in the PBMC culture treated with HMBPP + IL-12 in the presence or absence of 10 ug/ml cytokine-neutralizing antibodies of anti-IFN-γ (A) or anti-TNF-α mAb (C). Data are mean ± SEM of three independent experiments pooled from 12 healthy controls ^****p < 0.0001, paired t-test. (B) Exogenous TNF-α did not enhance the ability of IL-12 to expand HMBPP-activated Vγ2Vδ2 T cells. TNF-α (25 or 50 ng/ml) was added to HMBPP + IL-12 cultures, and cells were cultured for 7 days prior to measuring expansion of Vγ2Vδ2 T cells. Data shown as mean ± SEM of three independent experiments pooled from 12 healthy controls.

Figure 4

Both PI3K/AKT and STAT4 pathways, but not p38/MAPK, are involved in the IL-12-induced expansion of HMBPP-activated Vγ2Vδ2 T cells. (A–C) are graph data showing that PI3K/AKT and STAT4, but not P38/MAPK pathway, were required for HMBPP + IL-12 expansion of Vγ2Vδ2 T cells was reduced by the chemical blockers for PI3K/AKT and STAT4 pathways, but not those for P38/MAPK. PBMCs were co-cultured for 7 days with HMBPP + IL-2 or HMBPP + IL-12 in the presence of escalating doses (1, 5, 10 μM) of LY2944002 (inhibitor for PI3K/AKT), SB203580 (inhibitor for p38/MAPK), or doses (10, 20, 30, 40, 50 μM) of LSF (inhibitor for STAT4) or DMSO. Data are mean ± SEM of three independent experiments pooled from 12 healthy controls. ^**p < 0.01, vs. media group (ANOVA, Dunnett's test). (D) Exogenous TGF-β significantly reduce the ability of IL-12, but not IL-2, to expand HMBPP-activated Vγ2Vδ2 T cells. TGF-β (10 or 100 ng/ml) was added to the PBMC cultures treated with HMBPP + IL-12 or HMBPP + IL-2 (n = 12), and cells were cultured for 7 days prior to the flow-based analysis of the expansion of Vγ2Vδ2 T cells. Each dot represents one healthy control. ^**p < 0.01 vs. control group (paired t-test).

Figure 5

Vγ2Vδ2 T cells expanded by HMBPP + IL-12 exhibit central/effector memory phenotypes and maintain tissue trafficking markers. (A) Shown in the left are representative flow-cytometry quadrats displaying memory surrogate markers of Vγ2Vδ2 T cells based on CD27 and CD45RA expression in HMBPP + IL-12 cultures at day 0 and 7. Data at day 0 are similar to those in the medium only control culture. Shown in the right are representative flow cytometry histograms uncovering the phenotype of the αβ T cells that are cultured and gated in cytometry similarly to Vγ2Vδ2 T cells. (B) Bar graph shows that Vγ2Vδ2 T cells expanded by HMBPP + IL-12 express effector and central memory phenotypes. (C) Flow histogram and graph data show that expanded Vγ2⁺Vδ2⁺ T cells maintain expression of tissue-trafficking/residence markers CCR5 and LAF-1. Flowcytometry data are gated on live individual lymphocytes, CD3⁺, then Vγ2Vδ2⁺, and then representative markers. Data shown are mean ± SEM of three independent experiments pooled from 15 healthy controls.

Figure 6

Vγ2Vδ2 T cells expanded by HMBPP + IL-12 differentiate into polyfunctional effector cells expressing antimicrobial IFN-γ, TNF-α, GM-CSF, and tri-CTL cytokines. (A) Vγ2Vδ2 T cells expanded by HMBPP + IL-12 expressed higher level of TBX21 than those expanded by HMBPP/IL-2. PBMC were cultured with HMBPP + IL-12 (white bars) or HMBPP + IL-2 (black bars) for 7 days, and expanded Vγ2Vδ2 T cells were purified. Then, enriched Vγ2Vδ2 T cells were used for RNA isolation and determination of the expression levels of TBX21 and FOXP3. Cells from HMBPP + IL-2 treatment served as control setting. Shown are mean ± SEM of three independent experiments pooled from 15 healthy controls, ^****p < 0.0001, t-test. (B) Pooled flow-cytometry data (12 healthy controls) show the frequencies of T-bet⁺ cells (left) and mean fluorescence index (MFI) expression of Foxp3 (right) in gated Vγ2Vδ2 T cells expanded by HMBPP + IL-12 or HMBPP + IL-2. Each dot represents one healthy control. ^***p < 0.001, paired t-test. (C) Percentage numbers of Vγ2Vδ2 T cells expanded by HMBPP + IL-12 or HMBPP + IL-2 could produce CD107a, IFN-γ, TNF-α, and GM-CSF in 15 combinations in response to HMBPP stimulation. Using Boolean analysis, the bar graph shows percentages of individual multi-functional effector subsets for Vγ2Vδ2 T cells expanded by HMBPP + IL-12 (white bars) and HMBPP + IL-2 (black bars). Gating was on individual live lymphocytes, CD3⁺ T, then Vγ2⁺Vδ2⁺ T cells and then those cytokine markers. Shown are mean ± SEM of three independent experiments pooled from 15 healthy controls, ^****p < 0.0001, ^***p < 0.001, t-test. (D) Bar graph shows fold changes in expression levels of GZMA, GZMB, GNLY, and PRF in HMBPP + IL-12-expanded Vγ2Vδ2 T cells after HMBPP or media stimulation. Data are from four independent experiments pooled from 15 healthy controls. ^****p < 0.0001, ^***p < 0.001, ^**p < 0.01, t-test. Representative flow cytometric histograms are in Supplementary Figure 5A.

Figure 7

Vγ2Vδ2 T cells expanded by HMBPP + IL-12 inhibit intracellular mycobacteria growth in IFN-γ- and TNF-α-dependent fashions. (A) HMBPP + IL-12-expanded Vγ2Vδ2 T cells inhibit intracellular BCG growth in THP-1 and hMDM cells, respectively. Vγ2Vδ2 T cells enriched from HMBPP + IL-12 or HMBB + IL-2 co-cultures by positive selection (see Supplementary Figure 1D for the purity) were co-cultured with BCG-infected THP-1 (left panel) or hMDM (right panel) cells at an E: T ratio = 10:1 for 72 h. The reduction of CFU counts induced by Vγ2Vδ2 T cells was significantly more striking than that by B cells or media only. (B) Blocking assays using neutralizing mAb show that Vγ2Vδ2 T cells enriched from HMBPP + IL-12 cultures require effector molecules, IFN-γ and TNF-α, to inhibit intracellular mycobacteria. Vγ2Vδ2 T cells enriched from HMBPP + IL-12 cultures were incubated with BCG-infected THP-1 and hMDM cells at an E: T ratio of 10: 1 for 3 days in the presence/absence of 10 ug/ml neutralizing antibodies against IFN-γ, TNF-α, or matched Isotype. Data are mean ± SEM of three independent experiments pooled from 15 healthy controls, ^****p < 0.0001, ^***p < 0.001 vs. control (ANOVA, Dunnett's test).