Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Immunol. Author manuscript; available in PMC 2010 May 6.
Published in final edited form as:
PMCID: PMC2865221

Prolonged (E)-4-Hydroxy-3-Methyl-But-2-Enyl Pyrophosphate-Driven Antimicrobial and Cytotoxic Responses of Pulmonary and Systemic Vγ2Vδ2 T Cells in Macaques1


Although phosphoantigen-specific Vγ2Vδ2 T cells appear to play a role in antimicrobial and anticancer immunity, mucosal immune responses and effector functions of these γδ T cells during infection or phospholigand treatment remain poorly characterized. In this study, we demonstrate that the microbial phosphoantigen (E)-4-hydroxy-3-methyl-but-2-enyl pyrophosphate (HMBPP) plus IL-2 treatment of macaques induced a prolonged major expansion of circulating Vγ2Vδ2 T cells that expressed CD8 and produced cytotoxic perforin during their peak expansion. Interestingly, HMBPP-activated Vγ2Vδ2 T cells underwent an extraordinary pulmonary accumulation, which lasted for 3–4 mo, although circulating Vγ2Vδ2 T cells had returned to baseline levels weeks prior. The Vγ2Vδ2 T cells that accumulated in the lung following HMBPP/IL-2 cotreatment displayed an effector memory phenotype, as follows: CCR5+CCR7CD45RACD27+ and were able to re-recognize phosphoantigen and produce copious amounts of IFN-γ up to 15 wk after treatment. Furthermore, the capacity of massively expanded Vγ2Vδ2 T cells to produce cytokines in vivo coincided with an increase in numbers of CD4+ and CD8+ αβ T cells after HMBPP/IL-2 cotreatment as well as substantial perforin expression by CD3+Vγ2 T cells. Thus, the prolonged HMBPP-driven antimicrobial and cytotoxic responses of pulmonary and systemic Vγ2Vδ2 T cells may confer immunotherapeutics against infectious diseases and cancers.

Accumulating evidence suggests that human Vγ2Vδ2 (also called V γ9Vδ2) T cells play a role in mediating immunity against microbial pathogens (18) and tumors (9, 10). Vγ2Vδ2 T cells are a major circulating γδ T cell subset that normally constitutes 2–5% of peripheral blood T cells, and are unique in their ability to massively expand during various bacterial and protozoal infections (11) and notably increase in patients with certain cancers (10, 12). Vγ2Vδ2 T cell expansion appears to be specifically mediated by certain low m.w. foreign- and self-nonpeptidic phosphorylated metabolites of isoprenoid biosynthesis (e.g., (E)-4-hydroxy-3-methyl-but-2-enyl pyrophosphate (HMBPP),3 isopentenyl pyrophosphate (IPP), and its isomer dimethylallyl pyrophosphate) (1315) commonly referred to as phosphoantigens. HMBPP is produced in the newly discovered 2-C-methyl-d-erythritol-4-phosphate pathway of isoprenoid biosynthesis of most eubacteria, apicomplexan protozoa, plant chloroplasts, and algae, but not in vertebrates and thus not in the human host (16). Although IPP is also produced in humans, its bioactivity is ~104 lower than that of HMBPP, which is the most potent Vγ2Vδ2 T cell activator known, with an EC50 of 0.1 nM (13, 17, 18). Thus, much higher levels of endogenous IPP (e.g., those produced during cellular stress or transformation) (10, 19, 20) are most likely needed to trigger IPP-specific Vγ2Vδ2 T cell responses. Although the mechanism of phosphoantigen recognition by Vγ2Vδ2T cells is not well defined, it appears to be mediated by the Vγ2Vδ2 TCR (21) and requires cell-cell contact (22).

We and others have recently demonstrated that nonhuman primates can serve as a useful animal model to study the immune biology of human Vγ2Vδ2 T cells in response to phosphoantigen-producing pathogens or synthetic phospholigands (1, 23, 24). The remarkable expansion of Vγ2Vδ2 T cells after infection or treatment with synthetic Vγ2Vδ2 TCR ligands has indeed raised an exciting possibility to explore the potential of activated Vγ2Vδ2 T cells as immunotherapeutics against cancers and infectious diseases. Because the majority of pathogenic infections occur as a result of airborne, oral, or sexual transmission, it is crucial to characterize phosphoantigen-specific Vγ2Vδ2 T cell immune responses at these mucosal sites. However, many immunological questions regarding mucosal Vγ2Vδ2 T cell responses after infection or phospholigand treatment have not been addressed. Particularly, the ability of massively expanded Vγ2Vδ2 T cells to undergo mucosal migration and accumulate at these sites upon phospholigand treatment or infection remains largely unknown. In addition, the ability of activated Vγ2Vδ2 T cells to recognize naturally occurring phosphoantigen and exert effector functions in mucosae or the circulation has not been well characterized. Because the microbial phosphoantigen HMBPP is the most potent Vγ2Vδ2 T cell-activating ligand known, we have used HMBPP plus IL-2 treatment regimens to address a series of open questions as to whether HMBPP-activated Vγ2Vδ2 T cells are able to undergo prolonged expansion as seen in various infections, migrate to or accumulate in pulmonary and other mucosae in the context of phenotypic changes in homing or memory markers, re-recognize phosphoantigen and exert effector functions, and impact αβ T cell responses.

Materials and Methods


Four- to 8-year-old, 3- to 4-kg, cynomolgus macaques (Macaca fascicularis) were used in this study. All animals were maintained and used in accordance with guidelines of the institutional animal care and use committee. Animals were anesthetized with 10 mg/kg ketamine HCl (Fort Dodge Animal Health) i.m. for all blood sampling and treatments. EDTA-anticoagulated blood was collected at various time points before and after treatment. Day 0 blood was drawn immediately before treatment.

HMBPP and IL-2 administration

HMBPP was synthesized using a previously described procedure (25), which was scaled up to a final yield of 10 g. The purity determined by ion chromatography was >98%. Immediately before injection, HMBPP was reconstituted with saline and sterile filtered. Human rIL-2 (Proleukin; Chiron) was reconstituted with sterile ddH2O immediately before injection. Based on the in vitro bioactivity of HMBPP and with the objective of having a high in vivo exposure to HMBPP with a single injection, two groups of cynomolgus macaques received a 1-ml i.m. injection of either 10 mg/kg (n = 4) or 50 mg/kg (n = 4) HMBPP. These animals also received 0.5-ml s.c. injections of 1 million IU of IL-2 once daily for 5 consecutive days beginning on the day of HMBPP treatment. One macaque receiving 10 mg/kg HMBPP was only given 0.6 million IU of IL-2/day. As a control group, HMBPP alone was given at the following concentrations: 50 mg/kg (n = 4), 10 mg/kg (n = 2), and 5 mg/kg HMBPP (n = 1). Another control group was given 1 million IU of IL-2/day for 5 consecutive days (n = 2). No acute or long-term adverse effects were exhibited by any animal, including fever or chills, after HMBPP or IL-2 treatment. The general behavior and health of the animals remained normal throughout the study and after its conclusion.

Bronchoalveolar lavage (BAL), gingival, and rectal mucosae sampling

The procedures were modified from our previously described protocols (1). Before BAL and rectal biopsy sampling, animals were subjected to overnight or 24-h fasting, respectively, and were tranquilized i.m. with 1–2 mg/kg xylazine (Ben Venue Laboratories) and 10 mg/kg ketamine HCl. For BAL, animals also received 0.05 mg/kg atropine (Phoenix Scientific) i.m. as an anticholinergic and were restrained in an upright position. A pediatric feeding tube was inserted down the larynx, into the trachea through direct visualization with a laryngoscope to the level of the carina. Saline (10 ml) was instilled into the trachea and immediately withdrawn and repeated a maximum of three times until a total of 12–15 ml of BAL fluid was retrieved. For rectal biopsies, animals were restrained in ventral recumbency with the pelvic area supported and elevated ~4–5 above remainder of body. With the aid of a speculum standard, 3 × 5-mm biopsy forceps were used to collect two to three tissue pieces at each time point. Gingival mucosa samples were performed on anesthetized animals by atraumatically brushing their gum line with small sterile soft-bristled toothbrushes and rinsing the area with ~50 ml of saline and collecting the oral rinse.

Isolation of lymphocytes from blood, BAL fluid, gingival, and rectal mucosae

PBL were isolated from freshly collected EDTA blood by Ficoll-PaquePlus (Amersham) density gradient centrifugation before analysis. Lymphocyte isolation from freshly collected rectal mucosal biopsies was done according to previously described procedures (26) with minor changes. Briefly, biopsies were collected in RPMI 1640 containing 5% FBS (Invitrogen Life Technologies), washed, incubated for 30 min (37°C, 300 rpm) in 5% FBS-HBSS plus 5 mM EDTA, and, upon washing, incubated in 5% FBS-RPMI 1640 plus 90 U/ml collagenase (Sigma-Aldrich) for 1 h (37°C, 300 rpm). Samples were repeatedly aspirated with a 16-gauge needle to disrupt tissue and filtered through 70-µm cell strainers before layering on Ficoll-Paque-Plus and centrifuging at 3000 rpm for 20 min, after which the mononuclear cell layers were collected and washed with 10% FBS-RPMI 1640 before analysis. Freshly collected BAL and gingival mucosa samples were washed with 5% FBS-PBS and filtered through 40-µm cell strainers (BD Biosciences) before analysis.

Immunofluorescent staining and flow cytometric analysis

For cell surface staining, 100 µl of EDTA blood was treated with RBC lysing buffer (Sigma-Aldrich) and washed twice with 5% FBS-PBS before staining. PBMC, BAL, rectal, and gingival mucosa cells were stained with up to five Abs (conjugated to FITC, PE, allophycocyanin, Pacific Blue, and PE-Cy5 or allophycocyanin-Cy7) for at least 15 min. After staining, cells were fixed with 2% formaldehyde-PBS (Protocol Formalin) before analysis on a CyAn ADP flow cytometer (DakoCytomation). Lymphocytes were gated based on forward and side scatters, and pulse-width and at least 40,000 gated events were generally analyzed using Summit Data Acquisition and Analysis Software (DakoCytomation). Absolute cell numbers were calculated based on flow cytometry data and complete blood counts performed on a hematology system (Advia 120; Siemens).

The following mouse mAbs were used: Vγ9 (7A5), Vδ2 (15D), Vδ1 (TS8.2), and Pan γδ (5A6.E9) (Pierce); Vδ3 (P11.5B) (Beckman Coulter); CD3 (SP34, SP34-2), CD4 (L200), CD8 (RPA-T8), CD27 (M-T271), CD28 (CD28.2), CD45RA (5H9), CD49d (9F10), CCR5 (3A9), and IFN-γ (4S.B3) (BD Pharmingen); CD4 (OKT4) and CD27 (O323) (eBioscience); CD8 (DK25) (DakoCytomation); CCR7 (150503) (R&D Systems); perforin-biotin (Pf-344) (Mabtech); and rat mAb to integrin β7 (FIB504) (BD Pharmingen). The following secondary Abs were used for indirect staining: Pacific Blue-conjugated streptavidin (Invitrogen Life Technologies) and PE-conjugated goat F(ab′)2 anti-mouse IgG (Fcγ) (Beckman Coulter). Staining panels were as follows: CD3/CD4/CD8/Vγ2/Vδ2; CD3/Vγ2/ CD27/CD45RA; CD3/panγδ/Vδ1 or Vδ3; CD3/Vγ2/CCR5/CCR7; CD3/Vγ2/α47.

Intracellular cytokine staining

For intracellular cytokine staining, 105-106 BAL cells or 106 PBL plus costimulatory mAbs CD28 (1 µg/ml) and CD49d (1 µg/ml) were incubated with HMBPP (40 ng/ml) or medium alone in 200 µl final volume for 1 h at 37°C, 5% CO2, followed by an additional 5-h incubation in the presence of brefeldin A (GolgiPlug; BD Biosciences). After staining cell surface CD3 and Vγ2 for at least 15 min, cells were permeabilized for 45 min (Cytofix/Cytoperm; BD Biosciences) and stained another 45 min for IFN-γ and perforin before resuspending in 2% formaldehyde-PBS.

Statistical analysis

Statistical analysis was done using ANOVA and Student’s t test, as previously described (1).


HMBPP/IL-2 cotreatment induced prolonged expansion of Vγ2Vδ2 T cells and a transient increase in CD4+ and CD8+ αβ T cells in the circulation

Because nonmicrobial phospholigands tested to date appear to stimulate a shorter-term expansion of Vγ2Vδ2 T cells than those observed during human and macaque infections with phosphoantigen-producing microbes (23, 24, 27), we sought to determine whether the microbial phosphoantigen HMBPP can induce a prolonged expansion of Vγ2Vδ2T cells in monkeys. HMBPP (n = 7) or IL-2 (n = 2) treatment alone did not induce detectable Vγ2Vδ2 T cell expansion (Fig. 1a). However, after a single 10 mg/kg i.m. HMBPP injection plus low-dose IL-2 treatment, circulating Vγ2Vδ2 T cells specifically expanded to 41–62% of total CD3+ T cells (41- to 57-fold above baseline) and 1250–9900 cells/µl (102- to 442-fold above baseline) at day 4 after the treatment and remained 4- to 7-fold above baseline in percentage of total T cells at day 14 before returning to baseline at day 56 (n = 3) (Fig. 1, b–d). Upon increasing the HMBPP dose to 50 mg/kg plus IL-2, the duration of the Vγ2Vδ2 T cell expansion was significantly increased in half of the animals tested (animals 7235 and 7318) as compared with those treated with 10 mg of HMBPP/kg (Fig. 1e). Vγ2Vδ2 T cell levels remained 4- and 2.3-fold above baseline at day 56 in relative percentage of CD3+ T cells in animals 7235 and 7318, respectively, and did not return to baseline until day 105 (Fig. 1e). Such a prolonged expansion of Vγ2Vδ2 T cells appeared to result from greater increases in their absolute numbers during peak expansion, as follows: 12,200 and 41,700 cells/µl (127- and 914-fold above baseline) in the two animals, respectively (Fig. 1f). Vγ2Vδ2 T cells in the other two animals that received the 50 mg/kg HMBPP dose (7311 and 7317) expanded at day 4 and returned to baseline by day 21. Interestingly, majority of expanded Vγ2Vδ2 T cells in HMBPP/IL-2-cotreated animals expressed CD8 on their cell surface (Fig. 2, a–c). Furthermore, in these cotreated animals, absolute numbers of circulating CD4+ and CD8+ αβ T cells increased 3.5- and 3.7-fold on average, respectively, at day 7 after the HMBPP/IL-2 cotreatment and then returned near baseline levels on day 11, whereas HMBPP or IL-2 alone resulted in no or only subtle increases in CD4+ and CD8+ αβ T cell numbers (Fig. 2, d and e). Also, there was a marked lymphopenia of all examined T cells at day 1 after HMBPP treatment before the massive increases in Vγ2Vδ2 and αβ CD4+ and CD8+ T cells in cotreated animals (Fig. 1 and Fig. 2). These results indicate that the microbial phosphoantigen HMBPP is unique compared with other previously described phospholigands in its ability to stimulate a prolonged, major expansion of circulating Vγ2Vδ2 T cells, and also a transient increase in circulating CD4+ and CD8+ αβ T cells when coadministered with IL-2.

HMBPP/IL-2 cotreatment induced a prolonged and massive expansion of circulating Vγ2Vδ2 T cells in vivo. Relative percentages of CD3+ T cells that express Vγ2Vδ2 after treatment with HMBPP or IL-2 alone are shown over time ...
HMBPP/IL-2 cotreatment induced an expansion of circulating CD8- and CD4-expressing T cells in vivo. Cotreated animals were treated with 50 mg/kg (open symbols) or 10 mg/kg (filled symbols) HMBPP plus IL-2. As controls, four animals received 50 mg/kg HMBPP ...

HMBPP/IL-2 cotreatment induced a prolonged accumulation of Vγ2Vδ2 T cells in pulmonary mucosa and their short-term increase in gingival and rectal mucosae

Given the possibility that HMBPP/Vγ2Vδ2 T cell-based immunotherapeutics against infections might rely on the ability of these cells to migrate to the mucosal interface after their systemic expansion, we sought to examine whether massively expanded Vγ2Vδ2 T cells can accumulate in pulmonary and other mucosae after HMBPP/IL-2 cotreatment. HMBPP-activated Vγ2Vδ2 T cells appeared to readily undergo pulmonary migration during their massive expansion after HMBPP/IL-2 cotreatment, accounting for up to 80% of total CD3+ T cells and up to 4 million in absolute number in the BAL fluid at the peak accumulation time, days 4–11 (Figs. 3, a–c). Importantly, Vγ2Vδ2 T cell levels remained up to 46% of total CD3+ T cells (~ 10.4-fold above baseline) in the BAL fluid of all examined animals at day 56, whereas at this time blood Vγ2Vδ2 T cells were <5% of CD3+ T cells (Fig. 3b and Fig. 1, c and e). Sixteen weeks after the single HMBPP/IL-2 cotreatment, Vγ2Vδ2 T cells were as high as 13% of total BAL fluid CD3+ T cells in the five monkeys, whereas blood Vγ2Vδ2 T cells had returned to baseline several weeks prior. Also, the majority of pulmonary accumulated Vγ2Vδ2 T cells expressed CD8 (data not shown). On average, HMBPP/IL-2 cotreatment also led to 6- and 2-fold increases in pulmonary CD4+ and CD8+Vγ2 αβ T cell numbers, respectively, at day 4 (Fig. 3, d and e). Interestingly, although HMBPP treatment alone did not result in detectable increases in peripheral αβ T cell numbers (Fig. 2, d and e), we did detect 2- and 4-fold increases in pulmonary CD4+ and CD8+Vγ2 T cell numbers, respectively, at day 4, whereas the Vγ2Vδ2 T cell levels remained the same posttreatment (Fig. 3, c–e). Thus, although HMBPP alone did not result in marked expansion of peripheral or pulmonary Vγ2Vδ2 T cells, it may have been sufficient to activate these cells and affect pulmonary αβ T cell levels.

HMBPP/IL-2 cotreatment induced a prolonged accumulation of Vγ2Vδ2 T cells and a transient increase in αβ T cells in pulmonary mucosa. Freshly collected cells from BAL fluid of animals treated with 50 mg/kg (open symbols) ...

We then examined whether the massively expanded Vγ2Vδ2 T cells could similarly accumulate in rectal and gingival mucosae after the HMBPP/IL-2 cotreatment. Baseline Vγ2Vδ2 T cell levels in the rectal and gingival mucosae were on average 0.4 and 3.4% of the total CD3+ T cells at these sites, respectively (Fig. 4, b and d). On day 4, Vγ2Vδ2 T cells accumulated to make up 12–70% of total CD3+ T cells in the gingival mucosa of all tested animals except animal number 7317, whose Vγ2Vδ2 T cell expansion level in the blood and lung was also the lowest among the animals examined (Fig. 4, a and b). In the rectal mucosa, an increase in Vγ2Vδ2 T cells was prominently apparent in animals 7318 and 7319 on day 4 (Fig. 4, c and d). Vγ2Vδ2 T cells returned to baseline levels at day 21 or 28 at both of these sites. Also, HMBPP alone did not result in detectable changes in gingival or rectal Vγ2Vδ2 T cell levels (Fig. 4, b and d). Altogether, these results demonstrate that HMBPP/ IL-2 cotreatment induced a prolonged accumulation of Vγ2Vδ2 T cells in pulmonary mucosa and also their short-term increase in gingival and rectal mucosae.

HMBPP/IL-2 cotreatment induced a short-term increase in Vγ2Vδ2 T cells in gingival and rectal mucosae. Freshly collected cells from gingival mucosa (a and b) and rectal mucosal biopsies (c and d) from animals cotreated with 50 mg/kg (open ...

Effector memory Vγ2Vδ2 T cells that accumulated in the lung following HMBPP/IL-2 cotreatment were CCR5+CCR7CD45RACD27+

The high levels of Vγ2Vδ2 T cells that accumulated at the pulmonary mucosa upon HMBPP/IL-2 activation most likely underwent transendothelial migration from the circulation by specific interactions with the endothelium (27). Given that CCR7 and integrins (28) play an important role in driving lymphoid or mucosal migration of αβ T cells, and that CCR5 is implicated in endothelial migration of human γδ T cells in vitro (29), we focused on evaluation of CCR7, integrins α47, and CCR5 expression by pulmonary and circulating Vγ2Vδ2 T cells. Interestingly, whereas integrin α4 was expressed by virtually all pulmonary and circulating Vγ2Vδ2 T cells before and after treatment, integrin β7 expression increased from ~46 to ~68% on circulating Vγ2Vδ2 T cells and decreased from ~63 to ~23% on pulmonary Vγ2Vδ2 T cells (Fig. 5, a and b). Before treatment, circulating Vγ2Vδ2 T cells expressed over 3-fold higher levels of CCR5 on their cell surface as compared with those Vγ2Vδ2 T cells in the lung, whereas pulmonary Vγ2Vδ2 T cells expressed over 4-fold higher levels posttreatment (Fig. 5, a and b). In contrast, whereas <10% of pulmonary Vγ2Vδ2 T cells expressed CCR7 before and after treatment, blood Vγ2Vδ2 T cells expressed significantly higher levels of CCR7 posttreatment (Fig. 5, a and b). These results suggest that pulmonary Vγ2Vδ2 T cells that accumulated after HMBPP/IL-2 cotreatment possess a proinflammatory phenotype. Conversely, blood Vγ2Vδ2 T cells may maintain the ability to migrate to lymph nodes or other lymphoid tissues due to the predominant expression of CCR7 during HMBPP/IL-2-mediated massive expansion, which potentially explains the Vγ2Vδ2 lymphopenia observed on day 1 post-HMBPP treatment (Fig. 1, c and e).

Migration and memory marker expression by pulmonary and circulating Vγ2Vδ2 T cells. BAL fluid and whole blood from the following cotreated animals: 7311, 7315, 7317, 7318, and 7319 were examined. CD3- and Vγ2-gated BAL cells or ...

We then compared the potential memory status of pulmonary and circulating Vγ2Vδ2 T cells based on their expression of the surrogate memory markers CD27 and CD45RA at various time points post-HMBPP treatment. On average, ~82–95% of the Vγ2Vδ2 T cells in the lung displayed a memory phenotype (CD27+CD45RA) before and after treatment (Fig. 5c). In addition, on average, pulmonary CD27+CD45RA and CD27CD45RA Vγ2Vδ2 T cells increased from 7.9 × 105 to 2.1 × 106 (2.7-fold) and from 1.4 × 104 to 5.9 × 105 (40-fold) in absolute number on day 4, respectively (Fig. 5d). Vγ2Vδ2 T cells in the blood also predominantly exhibited the CD27+CD45RA memory phenotype at most time points except at peak expansion, whereby the effector memory phenotype (CD27CD45RA) significantly increased (Fig. 5e). However, majority of these circulating γδ T cells, not those in the pulmonary compartment, maintained the expression of the lymph node homing receptor CCR7 (Fig. 5b). Thus, although Vγ2Vδ2 T cells in both pulmonary and blood compartments predominantly displayed the CD27+CD45RA memory phenotype, an expression of CCR5 or CCR7 potentially distinguished effector memory from central memory phenotypes of these cells. In this regard, Vγ2Vδ2 T cells that migrated to the lungs and remained there for months after HMBPP/IL-2 cotreatment appeared to be CCR5+CCR7CD27+CD45RA effector memory cells, as majority of them were able to exert effector function (see Fig. 8 below).

Massive antimicrobial responses of long-lived pulmonary Vγ2Vδ2 T cells. Freshly collected BAL cells from cotreated animals at 12 and 15 wk post-in vivo HMBPP treatment were incubated for 6 h with HMBPP plus anti-CD28 and anti-CD49d mAbs ...

HMBPP-activated Vγ2Vδ2 T cells in the lung and circulation were able to re-recognize phosphoantigen and exert effector functions of IFN-γ and perforin production

To determine whether the Vγ2Vδ2 T cells that profoundly increased in the blood and pulmonary mucosa after HMBPP/IL-2 cotreatment could potentially re-recognize phosphoantigen produced by pathogens and mount antimicrobial immune responses, Vγ2Vδ2 T cells in blood or BAL fluid were assessed for their ability to produce cytokines in vivo as well as to re-recognize phosphoantigen and exert those effector functions ex vivo. Up to 35% of circulating Vγ2 T cells that expanded on day 4 after HMBPP/IL-2 cotreatment were able to produce the cytotoxic molecule perforin even without ex vivo HMBPP restimulation (Fig. 6, a and b), whereas 0.5–3.8% of expanded Vγ2 T cells produced the antimicrobial cytokine IFN-γ without further HMBPP exposure (Fig. 7, a and b). The in vivo production of these effector cytokines by Vγ2Vδ2 T cells at their peak expansion coincided with the apparent increases in numbers of CD4+ and CD8+ αβ T cells at day 7 (Fig. 2, d and e). Interestingly, the entire circulating Vγ2 T cell subpopulation that expanded in vivo at day 4 after HMBPP/ IL-2 cotreatment produced perforin following ex vivo HMBPP restimulation (Fig. 6, c–e), whereas 13–61% of blood Vγ2 T cells (4.9–10.6% of total CD3+) produced IFN-γ upon HMBPP restimulation at this time point (Fig. 7, c–e). Perforin expression by 4–15 and 2–6% of circulating CD3+Vγ2 T cells was also detected upon HMBPP restimulation on days 4 and 28, respectively (Fig. 6c). However, perforin-producing Vγ2 T cells were not detected at time points after day 4, suggesting the cytotoxic function of HMBPP-specific Vγ2Vδ2 T cells may be a transient feature. The percentage of circulating Vγ2 T cells that produced IFN-γ upon HMBPP restimulation varied among the animals tested, but were detectable at significant levels at all time points (Fig. 7e), suggesting the antimicrobial function of these cells is a lasting feature.

Cytotoxic responses of circulating Vγ2Vβ2 T cells. Freshly collected PBL from cotreated animals were stimulated for 6 h with HMBPP (c–e) or medium alone (a and b) plus anti-CD28 and anti-CD49d mAbs before staining. Perforin expression ...
Antimicrobial responses of circulating Vγ2Vδ2 T cells. Freshly collected PBL from cotreated animals were stimulated for 6 h with HMBPP (c–e) or medium alone (a and b) plus anti-CD28 and anti-CD49d mAbs before staining. IFN-γ ...

Surprisingly, even at 12 and 15 wkpost-HMBPP treatment, majority of the Vγ2 T cells that accumulated at high levels in the pulmonary mucosa were able to re-recognize the phosphoantigen HMBPP and exert effector function of cytokine production. Although up to 20% of BAL CD3+ T cells were Vγ2Vδ2 T cells at weeks 12 and 15, on average 84.9 and 82.8% of these BAL Vγ2+ T cells produced massive amounts of IFN-γ upon HMBPP restimulation at these time points, respectively (Fig. 8, a and b). This was in contrast to the low percentage of IFN-γ-producing Vγ2Vδ2 T cells in the blood even at their peak expansion after HMBPP/ IL-2 cotreatment. Also, ~1.5 and 3.1% of pulmonary Vγ2+ T cells were able to produce perforin upon HMBPP restimulation at weeks 12 and 15, respectively (Fig. 8c). Furthermore, cytokine production by CD3+Vγ2 T cells was detected in the lung (Fig. 8). These results indicate that the long-lived HMBPP-specific Vγ2 T cells in the blood and lung are not anergic, but rather, are capable of responding to further antigenic stimulation to become IFN-γ-and/or perforin-producing effectors and may regulate αβ T cell responses.


This is the first study using the microbial phosphoantigen HMBPP for potential Vγ2Vδ2 T cell-based immunotherapeutics in macaques. In comparison with recent studies using a nonmicrobial phospholigand (23, 24), several new observations have been made in our extensive study using HMBPP, as follows: 1) HMBPP/IL-2 cotreatment can induce a longer massive expansion of Vγ2Vδ2 T cells, perhaps due to the fact that HMBPP is the most potent Vγ2Vδ2-TCR ligand tested to date (13). 2) HMBPP-activated Vγ2Vδ2 T cells can undergo an extraordinary pulmonary accumulation or migration, which lasts for months even after the expansion of these cells becomes undetectable in the blood. 3) These pulmonary Vγ2Vδ2 T cells displayed an effector memory pheno-type, as follows: CCR5+CCR7CD45RACD27+.4) Although the peak peripheral expansion of Vγ2Vδ2 T cells confers upon these cells the ability to produce cytotoxic perforin in response to HMBPP, most VγVδ2 T cells that migrated to the lung after HMBPP/IL-2 cotreatment are able to re-recognize microbial phosphoantigen and mount effector function via IFN-γ production. 5) The capacity of massively expanded Vγ2Vδ2 T cells to autonomously produce cytokines without ex vivo restimulation coincides with an increase in numbers of CD4+ and CD8+ αβ T cells in the blood and lung after HMBPP/IL-2 cotreatment.

The prolonged accumulation of Vγ2Vδ2 T cells in the lung airspaces implies that the pulmonary mucosa is the favorable migration site for phosphoantigen-specific Vγ2Vδ2 T cells. In general, the magnitude and duration of Vγ2Vδ2 T cell expansion in the lung are much greater than in the blood and gingival or rectal mucosae after the HMBPP/IL-2 cotreatment. This is consistent with our earlier studies demonstrating that a major expansion of Vγ2Vδ2 T cells after pulmonary Mycobacterium tuberculosis infection occurs in the pulmonary compartment, but not in the blood (1). Importantly, the current study indicates that the pulmonary mucosa can selectively recruit a large number of Vγ2Vδ2 T cells that predominantly display an effector memory phenotype and produce large quantities of IFN-γ after HMBPP restimulation. The preferential expression of CCR5 by pulmonary Vγ2Vδ2 T cells suggests that CCR5 may contribute in recruiting HMBPP-activated Vγ2Vδ2 T cells to the lung. This scenario is indeed supported by the in vitro migration study describing a role of CCR5 and its ligands (MIP-1α, MIP-1β, and RANTES) in the transendothelial migration of human Vγ2 T cells (29). Although the CCR5-driven migration may account for the prolonged accumulation of Vγ2Vδ2 T cells in the lung after the single HMBPP/IL-2 cotreatment, we cannot exclude the possibility that proliferation of these γδ T cells in pulmonary/bronchial mucosal lymphoid follicles contributes to their long-lasting accumulation. It is worth mentioning that the timing of the drop in peripheral Vγ2Vδ2 and αβ T cell numbers after expansion does not directly correspond to the increases of these cells in the lung. Thus, the decline of these cells in the circulation after their expansion may be due to activation-induced cell death.

Another extraordinary observation found in the present study is that over 80% of Vγ2Vδ2 T cells that migrated to the lung can re-recognize phosphoantigen and produce copious amounts of the antimicrobial cytokine IFN-γ even at 12–15 wk after the single HMBPP/IL-2 cotreatment. This observation suggests that the HMBPP/IL-2 regimen can confer Vγ2Vδ2 T cell-based immunotherapeutics against a variety of pulmonary infections induced by phosphoantigen-producing pathogens. Presumably, these HMBPP-activated Vγ2Vδ2 T cells can readily sense or recognize phosphoantigen-producing pathogens and mount IFN-γ-mediated antimicrobial immune responses in the mucosa and lymphoid-tissue interface. Because HMBPP-activated Vγ2Vδ2 T cells can express granulysin (L. Shao and Z. W. Chen, unpublished study), they may directly mediate bactericidal effects. Furthermore, HMBPP-activated Vγ2Vδ2 T cells may readily attack target cells infected with phosphoantigen-producing pathogens, because some of these γδ T cells can produce the cytotoxic molecule perforin.

The long-lived pulmonary Vγ2Vδ2 T cell response induced by a single HMBPP/IL-2 cotreatment may be sufficient to provide some degree of effect against various pathogens. However, multiple treatments may be needed to obtain optimal effects. Although an exhaustive response of Vγ2Vδ2 T cells after repeated phospholigand/human IL-2 treatments has been previously reported (23), it is yet to be determined whether this is a true exhaustive phenomenon or simply attributed to the development of tolerance to human IL-2 by the monkeys instead of Vγ2Vδ2 T cell exhaustion after repeated phospholigand/human IL-2 cotreatments.

We also found that the majority of HMBPP-activated Vγ2Vδ2 T cells express CD8. Although under conditions in which these cells are not undergoing expansion, they constitute a minor proportion of total CD8+ T cells, our study demonstrates that these cells expand to make up >50% of the total CD8+ T cell population when Vγ2Vδ2 T cell phosphoantigen is present along with IL-2. Thus, during situations in which Vγ2Vδ2 T cells may be induced (e.g., during most bacterial or protozoal infections), it is important to distinguish CD8+ T cells based on their TCR to accurately determine potentially distinct responses of these two T cell populations. Furthermore, HMBPP-activated Vγ2Vδ2 T cells appear to impact αβ T cell responses in vivo. In animals whose Vγ2Vδ2 T cells underwent massive expansion and produced detectable levels of cytokines in vivo, circulating αβ T cells increased almost 4-fold. The increase in αβ T cells may have resulted from a non-TCR-mediated expansion of total αβ T cells (e.g., via IL-2R-, CD27/28-, α47-, or other adhesion molecule-mediated signaling). The increase that we observed was transient most likely due to the absence of cognate αβ TCR Ags, yet circulating and pulmonary CD3+Vγ2 T cells produced considerable levels of perforin and IFN-γ. Thus, massively expanded Vγ2Vδ2 T cells after the HMBPP/IL-2 cotreatment may confer an adjuvant effect on the development of adaptive αβ T cell responses in the setting of an infection or immunization with a vector-based vaccine (1, 30, 31).


We thank B. Paige, J. Graves, and Dr. K. Hagen for technical assistance with flow cytometry, and the University of Illinois Biological Resources Laboratory staff for animal care and technical assistance with complete blood counts.


1This work was supported by National Institutes of Health R01 Grants HL64560 and RR13601 (both to Z.W.C.) and Deutsche Forschungsgemeinschaft Grant JO565/1-1 (to H. J.).

3Abbreviations used in this paper: HMBPP, (E)-4-hydroxy-3-methyl-but-2-enyl pyrophosphate; BAL, bronchoalveolar lavage; IPP, isopentenyl pyrophosphate.


The authors have no financial conflict of interest.


1. Shen Y, Zhou D, Qiu L, Lai X, Simon M, Shen L, Kou Z, Wang Q, Jiang L, Estep J, et al. Adaptive immune response of Vγ2Vδ2+ T cells during mycobacterial infections. Science. 2002;295:2255–2258. [PMC free article] [PubMed]
2. Wang L, Kamath A, Das H, Li L, Bukowski JF. Antibacterial effect of human Vγ2Vδ2 T cells in vivo. J. Clin. Invest. 2001;108:1349–1357. [PMC free article] [PubMed]
3. Poccia F, Battistini L, Cipriani B, Mancino G, Martini F, Gougeon ML, Colizzi V. Phosphoantigen-reactive Vγ9Vδ2 T lymphocytes suppress in vitro human immunodeficiency virus type 1 replication by cell-released antiviral factors including CC chemokines. J. Infect. Dis. 1999;180:858–861. [PubMed]
4. Dieli F, Troye-Blomberg M, Ivanyi J, Fournie JJ, Bonneville M, Peyrat MA, Sireci G, Salerno A. Vγ9/Vδ2 T lymphocytes reduce the viability of intracellular Mycobacterium tuberculosis. Eur. J. Immunol. 2000;30:1512–1519. [PubMed]
5. Ottones F, Dornand J, Naroeni A, Liautard JP, Favero J. Vγ9Vδ2 T cells impair intracellular multiplication of Brucella suis in autologous monocytes through soluble factor release and contact-dependent cytotoxic effect. J. Immunol. 2000;165:7133–7139. [PubMed]
6. Poccia F, Agrati C, Martini F, Capobianchi MR, Wallace M, Malkovsky M. Antiviral reactivities of γδ T cells. Microbes Infect. 2005;7:518–528. [PubMed]
7. Farouk S, Mincheva-Nilsson EL, Krensky AM, Dieli F, Troye-Blomberg M. γδ T cells inhibit in vitro growth of the asexual blood stages of Plasmodium falciparum by a granule exocytosis-dependent cytotoxic pathway that requires granulysin. Eur. J. Immunol. 2004;34:2248–2256. [PubMed]
8. Troye-Blomberg M, Worku S, Tangteerawatana P, Jamshaid R, Soderstrom K, Elghazali G, Moretta L, Hammarstrom M, Mincheva-Nilsson L. Human γδ T cells that inhibit the in vitro growth of the asexual blood stages of the Plasmodium falciparum parasite express cytolytic and proinflammatory molecules. Scand. J. Immunol. 1999;50:642–650. [PubMed]
9. Liu Z, Guo BL, Gehrs BC, Nan L, Lopez RD. Ex vivo expanded human Vγ9Vδ2+ γδ-T cells mediate innate antitumor activity against human prostate cancer cells in vitro. J. Urol. 2005;173:1552–1556. [PubMed]
10. Corvaisier M, Moreau-Aubry A, Diez E, Bennouna J, Mosnier JF, Scotet E, Bonneville M, Jotereau F. V γ9Vδ2 T cell response to colon carcinoma cells. J. Immunol. 2005;175:5481–5488. [PubMed]
11. Chen ZW, Letvin NL. Vγ2Vδ2+ T cells and anti-microbial immune responses. Microbes Infect. 2003;5:491–498. [PMC free article] [PubMed]
12. McClanahan J, Fukushima PI, Stetler-Stevenson M. Increased peripheral blood γδ T-cells in patients with lymphoid neoplasia: a diagnostic dilemma in flow cytometry. Cytometry. 1999;38:280–285. [PubMed]
13. Hintz M, Reichenberg A, Altincicek B, Bahr U, Gschwind RM, Kollas AK, Beck E, Wiesner J, Eberl M, Jomaa H. Identification of (E)-4-hydroxy-3-methyl-but-2-enyl pyrophosphate as a major activator for human γδ T cells in Escherichia coli. FEBS Lett. 2001;509:317–322. [PubMed]
14. Tanaka Y, Morita CT, Tanaka Y, Nieves E, Brenner MB, Bloom BR. Natural and synthetic nonpeptide antigens recognized by human γδ T cells. Nature. 1995;375:155–158. [PubMed]
15. Altincicek B, Moll J, Campos N, Foerster G, Beck E, Hoeffler JF, Grosdemange-Billiard C, Rodriguez-Concepcion M, Rohmer M, Boronat A, et al. Cutting edge: human γδ T cells are activated by intermediates of the 2-C-methyl-D-erythritol 4-phosphate pathway of isoprenoid biosynthesis. J. Immunol. 2001;166:3655–3658. [PubMed]
16. Eberl M, Hintz M, Reichenberg A, Kollas AK, Wiesner J, Jomaa H. Microbial isoprenoid biosynthesis and human γδ T cell activation. FEBS Lett. 2003;544:4–10. [PubMed]
17. Eberl M, Hintz M, Jamba Z, Beck E, Jomaa H, Christiansen G. Mycoplasma penetrans is capable of activating Vγ9/Vδ2 T cells while other human pathogenic mycoplasmas fail to do so. Infect. Immun. 2004;72:4881–4883. [PMC free article] [PubMed]
18. Zhang Y, Song Y, Yin F, Broderick E, Siegel K, Goddard A, Nieves E, Pasa-Tolic L, Tanaka Y, Wang H, et al. Structural studies of Vγ2Vδ2 T cell phosphoantigens. Chem. Biol. 2006;13:985–992. [PubMed]
19. Gober HJ, Kistowska M, Angman L, Jeno P, Mori L, De Libero G. Human T cell receptor γδ cells recognize endogenous mevalonate metabolites in tumor cells. J. Exp. Med. 2003;197:163–168. [PMC free article] [PubMed]
20. Bonneville M, Fournie JJ. Sensing cell stress and transformation through Vγ9Vδ2 T cell-mediated recognition of the isoprenoid pathway metabolites. Microbes Infect. 2005;7:503–509. [PubMed]
21. Bukowski JF, Morita CT, Tanaka Y, Bloom BR, Brenner MB, Band H. Vγ2Vδ2 TCR-dependent recognition of non-peptide antigens and Daudi cells analyzed by TCR gene transfer. J. Immunol. 1995;154:998–1006. [PubMed]
22. Morita CT, Beckman EM, Bukowski JF, Tanaka Y, Band H, Bloom BR, Golan DE, Brenner MB. Direct presentation of nonpeptide prenyl pyrophosphate antigens to human γδ T cells. Immunity. 1995;3:495–507. [PubMed]
23. Sicard H, Ingoure S, Luciani B, Serraz C, Fournie JJ, Bonneville M, Tiollier J, Romagne F. In vivo immunomanipulation of Vγ9Vδ2 T cells with a synthetic phosphoantigen in a preclinical nonhuman primate model. J. Immunol. 2005;175:5471–5480. [PubMed]
24. Casetti R, Perretta G, Taglioni A, Mattei M, Colizzi V, Dieli F, D’Offizi G, Malkovsky M, Poccia F. Drug-induced expansion and differentiation of Vγ9Vδ2> T cells in vivo: the role of exogenous IL-2. J. Immunol. 2005;175:1593–1598. [PubMed]
25. Hecht S, Amslinger S, Jauch J, Kis K, Trentinaglia V, Adam P, Eisenreich W, Bacher A, Rohdich FF. Studies on the non-meval-onate isoprenoid biosynthetic pathway: simple methods for preparation of isotope-labeled (E)-1-hydroxy-2-methylbut-2-enyl 4-diphosphate. Tetrahedron Lett. 2002;43:8929–8933.
26. Veazey RS, DeMaria M, Chalifoux LV, Shvetz DE, Pauley DR, Knight HL, Rosenzweig M, Johnson RP, Desrosiers RC, Lackner AA. Gastrointestinal tract as a major site of CD4+ T cell depletion and viral replication in SIV infection. Science. 1998;280:427–431. [PubMed]
27. Chen ZW, Letvin NL. Adaptive immune response of Vγ2Vδ2 T cells: a new paradigm. Trends Immunol. 2003;24:213–219. [PMC free article] [PubMed]
28. Lefrancois L, Puddington L. Intestinal and pulmonary mucosal T cells: local heroes fight to maintain the status quo. Annu. Rev. Immunol. 2006;24:681–704. [PubMed]
29. Roth SJ, Diacovo TG, Brenner MB, Rosat JP, Buccola J, Morita CT, Springer TA. Transendothelial chemotaxis of human α/β and γ/δ T lymphocytes to chemokines. Eur. J. Immunol. 1998;28:104–113. [PubMed]
30. Cendron D, Ingoure S, Martino A, Casetti R, Horand F, Romagne F, Sicard H, Fournie JJ, Poccia F. A tuberculosis vaccine based on phosphoantigens and fusion proteins induces distinct γδ and α/β T cell responses in primates. Eur. J. Immunol. 2007;37:549–565. [PubMed]
31. Martino A, Casetti R, Poccia F. Enhancement of BCG-induced Th1 immune response through Vγ9Vδ2 T cell activation with non-peptidic drugs. Vaccine. 2007;25:1023–1029. [PubMed]