Conservation of Nonpeptide Antigen Recognition by Rhesus Monkey Vγ2Vδ2 T Cells¹

Gamma delta (γδ) T cells are a second subset of T cells with unique functional and recognition properties (1). Murine γδ T cells play important roles in immunity to infections and tumors (2), tissue homeostasis (3), immunoregulation (4), and in controlling autoimmune αβ T cell responses (5). In so doing, γδ T cells likely function as a bridge between the innate and adaptive immune systems. Although few Ags have been defined for murine γδ T cells, human γδ T cells can recognize nonpeptide Ags. γδ T cells expressing Vδ1 TCRs can recognize self and foreign lipids presented by CD1 (6) and the nonclassical MHC class Ib protein, MHC class I-related chain A/MHC class I-related chain B (7). Most human γδ T cells express Vγ2Vδ2 TCRs that recognize nonpeptide prenyl pyrophosphates (8, 9, 10), bisphosphonates (11, 12, 13), and alkylamines (14). The recognition of these nonpeptide Ags requires the pairing of Vγ2 (also termed Vγ9) with Vδ2 (15). Although cell-cell contact is required for recognition (16), recognition does not require classical MHC class I or class II molecules (16), prior antigenic exposure (8, 17, 18), or Ag processing (16).

Large expansions of Vγ2Vδ2⁺ T cells occur during some bacterial and parasitic infections (reviewed in Ref. 1). In some patients, these γδ T cell expansions can be extremely large such that almost all (in erhlichiosis) or half (in salmonellosis and tularemia) of peripheral blood T cells are γδ T cells. The γδ T cell expansion can occur rapidly after infection (as early as 7 days; Ref. 19) and can persist for up to a year (20). Vγ2Vδ2⁺ T cells recognize cells infected with bacteria (21, 22, 23) and can reduce bacterial viability through the release of granulysin (6, 24). Vγ2Vδ2⁺ T cells also produce large amounts of the inflammatory TNF-α and IFN-γ cytokines when stimulated by nonpeptide Ags (25). Moreover, Vγ2Vδ2⁺ T cells activated by nonpeptide Ags protect SCID mice from in vivo bacterial infections by reducing bacterial numbers (26). These results suggest that human γδ T cells have important roles in immunity to these pathogens similar to and perhaps greater than the roles played by γδ T cells in mice.

Determining the importance of γδ T cells in human immunity has been hampered by the lack of an experimental animal model system to study this T cell subset. Few foreign Ags have been discovered for γδ T cells from other species. Other nonprimate animals, including mice and rats, do not have V genes homologous to Vγ2 and Vδ2 genes and do not respond to nonpeptide Ags that stimulate human Vγ2Vδ2⁺ T cells. Recently, we demonstrated that Vγ2Vδ2⁺ T cells in rhesus monkeys can expand in response to infection with Mycobacterium bovis bacillus Calmette-Guérin (BCG)⁴ (27). Monkey Vγ2Vδ2⁺ T cells mounted adaptive (memory) responses because reinfection of the monkeys with BCG resulted in earlier and larger γδ T cell expansions. Importantly, this capacity to rapidly expand coincided with a clearance of BCG bacteremia and immunity to fatal tuberculosis in BCG-vaccinated rhesus monkeys.

To extend our study of rhesus monkey γδ T cells, we have analyzed rhesus monkey γδ T cell responses to the described nonpeptide Ags and superantigens. Like human γδ T cells, rhesus monkey γδ T cells expanded when exposed to prenyl pyrophosphates and other phosphoantigens and to alkylamine Ags. These expansions were limited to γδ T cells expressing Vγ2Vδ2 TCRs. Rhesus monkey Vγ2Vδ2⁺ T cells also responded to the superantigen, staphylococcal enterotoxin A (SEA), and to an unknown Ag expressed by the B cell lymphoma, RPMI 8226. The amino acid sequence of the rhesus monkey Vγ2Vδ2 TCR has strong homology to human Vγ2Vδ2 TCR. A molecular model of the rhesus monkey Vγ2Vδ2 TCR shows a basic region in the complementarity-determining region (CDR)3 binding groove that is similar to that seen in the human Vγ2Vδ2 TCR. These data demonstrate that recognition of nonpeptide prenyl pyrophosphate and alkylamines is conserved in primates providing an animal model for human γδ T cell responses.

Prenyl pyrophosphates and alkylamines were purchased from Sigma-Aldrich (St. Louis, MO). Phosphoantigens including monoethyl phosphate (MEP) and nucleotide-conjugated compounds were synthesized as described (10, 28). The SEA superantigen was produced as a recombinant protein as described (29).

Mononuclear cells were prepared from heparinized rhesus monkey blood by centrifugation over density gradients of Ficoll-Hypaque (Pharmacia Fine Chemicals, Piscataway, NJ). PBMC were incubated with media, isopentenyl pyrophosphate (IPP) (50 μM), or partially purified (E)-4-hydroxy-3-methyl-but-2-enyl pyrophosphate (HMBPP) (1/1000 dilution of a concentrated, <1 kDa Mycobacterium fortuitum supernatant). IL-2 was added to 1 nM on day 3. The cells were harvested on days 7–10 and stained with the indicated mAbs for flow cytometric analysis as detailed below. For the alkylamines, 1 million PBMC in 1 ml of medium were cultured for 10 days with 400 μM sec- or iso-butylamine, followed by double-staining with anti-Vδ2/Jδ2 (15D) and anti-Vγ2 (7A5) mAbs and analysis by flow cytometry. Four monkeys were tested.

Rhesus monkey γδ T cell clones were derived from the blood of a normal rhesus monkey. PBMC were stimulated with a mycobacterial supernatant from M. fortuitum (8). After 2 wk, γδ T cells comprised 96% of T cells. γδ T cells were cloned by limiting dilution at 5 cells/well. T cell clones were maintained by periodic restimulation with PHA-P (30).

For proliferation assays, T cells were plated in triplicate in round-bottom 96-well plates at 5–10 × 10⁴ T cells per well with 1 × 10⁵ irradiated (7000 rad) human PBMC feeder cells, the EBV-transformed B cell line, DG.EBV, PBMC, or glutaraldehyde-fixed Va-2 tumor cells with various Ags. Prenyl pyrophosphates in ammonium hydroxide and methanol were dried by N₂ gas and dissolved in media by sonicating in an ultrasonic water bath for 5 min. The cultures were pulsed with 1 μCi [³H]thymidine (2 Ci/mmol) on day 1 and harvested 16–18 h later.

mAbs reactive with human γδ TCR were tested to identify those that cross-react with rhesus monkey γδ T cells using either rhesus monkey PBMC or rhesus monkey Vγ2Vδ2 T cell lines. The cross reactivity for mAbs for γδ TCR are detailed in Table I. Based on these results, the following anti-TCR mAbs were used: anti-TCRδ1 (pan anti-TCRγδ), δTCS1 or TS8 (anti-Vδ1/Jδ1/2), 15D (anti-Vδ2/Jδ2/3), 7A5 (anti-Vγ2 (Vγ2 is also called Vγ9)), 23D12 (anti-Vγ1.2, 1.3, and 1.4), and 4A11 (anti-Vγ1.4). mAb to surface molecules were FN18 (anti-monkey CD3), OKT4 (anti-CD4), OKT8 (anti-CD8α), and OKT11 (anti-CD2). Isotype-matched control mAbs were used as controls.

Mononuclear cells were prepared from heparinized rhesus monkey blood by centrifugation over density gradients of Ficoll-Hypaque (Pharmacia Fine Chemicals). Isolated mononuclear cells, T cell lines, and T cell clones were analyzed by one- or two-color immunofluorescence after staining with the appropriate mAb as described (31). Cells were incubated with mouse mAbs on ice for 30 min, washed, and stained with FITC-conjugated F(ab′)2 goat anti-mouse IgG and IgM antisera (Tago Scientific, Burlingame, CA) for an additional 30 min on ice. After washing, the cells were resuspended in propidium iodide and analyzed by flow cytometry. To prevent background staining of rhesus monkey B cells by anti-mouse Ig antisera, 2% monkey serum was included in the staining buffer. Flow cytometry was performed with a FACScan flow cytometer using CellQuest software (BD Biosciences, Palo Alto, CA).

RNA was isolated from rhesus monkey T cell lines (Micro RNA isolation kit; Stratagene, La Jolla, CA) followed by cDNA synthesis using SuperScript II reverse transcriptase and random hexamers (SuperScript first-strand synthesis system for RT-PCR; Life Technologies, Gaithersburg, MD). PCR was done with Platinum Taq High Fidelity DNA polymerase (Life Technologies). PCR primers used to derive full-length Vγ2Cγ and Vδ2Cδ chains were as described previously (6) except that for the Vδ2Cδ chain the following primer was used to introduce a KpnI restriction site into the 5′ region of the Vδ2Cδ2 chain for cloning: 5′-gggggtaccCAGGCAGAAGGTGGTTGAGAG-3′ Vδ2 5′ untranslated region. The Vγ2Cγ and Vδ2Cδ PCR products were cloned into pREP7 and pREP9 vectors through KpnI-XhoI and KpnI-BamHI sites (Invitrogen, Carlsbad, CA), respectively. Sequencing was done using an automated sequencer using the pREP forward and reverse primers along with the following reverse primers: Cγ, 3′UT ATGGCCTCCTTGTGCCACCG; Cγ internal, TGTGTCGTTAGTCTTCATGG; Cδ, 3′UT GGAGTGTAGCTTCCTCATGC; and Cδ internal, GACAATAGCAGGATCAAACT. Nucleotide sequences have been deposited into GenBank under the accession numbers of AY190025 (RM.2.32γ), AY190026 (RM.2.32δ), AY190027 (RM.2.14γ), and AY190028 (RM.2.14δ).

Chimeric rhesus monkey Vγ2/human Vδ2 TCR transfectants were derived by electroporation of the Jurkat mutant, J.RT3-T3.5 with the rhesus monkey RM2.14 TCR-γ chain cDNA (cloned in the pREP-7 vector) and the human DG.SF68 TCR-δ chain cDNA (cloned in the pREP-9 vector) as described previously (32). Chimeric human Vγ2/rhesus monkey Vδ2 TCR transfectants were derived using a similar strategy with human DG.SF13 TCR-γ chain cDNA and rhesus monkey RM2.32 TCR-δ chain cDNA except that human Cδ was substituted for rhesus monkey Cδ (see below). Human Vγ2 and Vδ2 TCR cDNAs were inserted into pREP-7 and pREP-9, respectively (15). In initial experiments, neither rhesus monkey Vγ2Vδ2 TCRs nor chimeric human Vγ2/rhesus monkey Vδ2 TCRs were expressed when transfected into J.RT3-T3.5, presumably due to the inability of rhesus monkey Cδ to associate with the human CD3 complex. To allow expression of a chimeric receptor, human Cδ was substituted for rhesus monkey Cδ. A rhesus monkey Vδ2 fragment from the RM2.32 T cell clone and a human Cδ fragment were produced by PCR using Vδ2 forward with rhesus monkey Vδ2 5′ UT reverse primers and human Cδ forward with Cδ 3′ UT reverse primers, respectively. PCR primers were as above except for: rhesus monkey Vδ2, 5′ UT reverse, CAGTCACACGGGTCCCTTTTCCAAAGATG; human Cδ forward, CATCTTTGGAAAAGGAACCCGTGTGACTG.

A complete cDNA from the fragments was then produced by annealing PCR under the following conditions: a 5-min denaturation step at 94°C followed by five cycles of annealing (94°C for 1 min, 50°C for 4 min, 72°C for 1.5 min), 40 cycles of amplification (94°C for 30 s, 55°C for 1 min, 72°C for 1 min), and a final extension at 72°C for 10 min. The rhesus monkey Vδ2/human Cδ cDNA was then cloned into pREP-9.

Stimulation of TCR transfectants for IL-2 release was performed as described (16). Briefly, 1 × 10⁵ transfectants were cultured in triplicate with the indicated Ag in the presence of 1 × 10⁵ glutaraldehyde-fixed Va-2 cells and 10 ng/ml PMA. After 24 h, supernatants were harvested, frozen, thawed, and used at a 1/8 dilution to stimulate the proliferation of the IL-2-dependent cell line, HT-2.

A model of the rhesus monkey Vγ2Vδ2TCR was built using homology modeling as described for the Vγ2 chain (29) using the coordinates for the human Vγ2Vδ2TCR (kindly provided by Dr. D. Garboczi, National Institutes of Health, Bethesda, MD). The sequences of monkey and human Vγ2Vδ2 TCRs were aligned using a GeneMine package (Molecular Applications Group, Palo Alto, CA). Because they show >85% sequence identity, the CARA module of LOOK homology modeling package (33) was used to generate the model of the monkey Vγ2Vδ2 TCR using the coordinates of the crystal structure of human Vγ2Vδ2 TCR as the initial template. There are two insertions in the TCR CDR3. To model this loop, the SEGMOD module of GeneMine (34) was used. The figures were generated by Molscript (35) and Raster3D (36) (see Fig. 7, A–C) and GRASP (see Fig. 7 D).

γδ T cells in humans at birth are present at low levels with a predominance of Vδ1⁺ cells. Between the ages of 3 and 10 years, Vγ2Vδ2⁺ T cells expand in response to environmental factors resulting in the predominance of Vγ2Vδ2⁺ T cells that is found in most adults. To determine whether rhesus monkeys maintained in closed, specific pathogen-free colonies also exhibit a predominance of Vγ2Vδ2⁺ T cells, PBMC from rhesus monkeys were analyzed for V gene segment expression by flow cytometry. Unlike most human adults, adolescent and adult rhesus monkeys exhibited a predominance of Vδ1⁺ T cells (Table II and Fig. 1). This pattern of V gene expression is seen in human newborns and in human infants <3–10 years of age (31, 37). Because the change in the proportion of Vδ1⁺ to Vδ2⁺ T cells in humans is due to an environmental influence, this suggests that rhesus monkeys maintained in closed colonies may not be exposed to the same pathogens or natural flora as humans.

To determine whether γδ T cells from rhesus monkeys respond to nonpeptide Ags, PBMC from rhesus monkeys were stimulated with the major bacterial phosphoantigen, HMBPP, from M. fortuitum and the alkylamines, sec-butylamine and iso-butylamine. Exposure of monkey PBMC to HMBPP resulted in an expansion of Vγ2Vδ2⁺ T cells to between 20 and 97% of CD3⁺ T cells (Fig. 2,A). Similarly, exposure of PBMC from other rhesus monkeys to sec-butylamine and iso-butylamine resulted in Vγ2Vδ2⁺ T cell expansions of between 10 and 43% and 32 and 46% of CD3⁺ T cells, respectively (Fig. 2 B). Note that one monkey did not respond to sec-butylamine and a second different monkey did not respond to iso-butylamine. A similar lack of response to nonpeptide Ags has been observed in ∼10–20% of human adults (17). Thus, rhesus monkey Vγ2Vδ2⁺ cells, like human Vγ2Vδ2⁺ T cells, proliferate in response to nonpeptide phosphoantigens and alkylamine Ags.

Human Vγ2Vδ2⁺ T cells respond to nonpeptide phosphoantigens and the superantigen, SEA. To determine the response of rhesus monkey Vγ2Vδ2⁺ T cells at the clonal level, rhesus monkey Vγ2Vδ2⁺ T cell clones were derived and tested for recognition of the MEP prenyl pyrophosphate analog and the SEA superantigen. A rhesus monkey Vγ2Vδ2⁺ T cell line was derived by stimulating PBMC with HMBPP followed by limited dilution cloning. γδ T cells expressed CD8αα homodimers (CD8α staining was 88% whereas CD8β was 0.9%) and CD2 (99%) (data not shown). Ninety-four percent (51 of 54 clones) of the monkey Vγ2Vδ2⁺ T cell clones derived from this line responded to the prenyl pyrophosphate analog, MEP. Like human γδ T cells (29), the rhesus monkey Vγ2Vδ2⁺ T cell clones also responded to the superantigen, SEA (representative examples are shown in Fig. 3).

To determine whether the conservation of nonpeptide Ag reactivity reflects conservation of the Vγ2Vδ2 TCR sequence, we cloned and sequenced full-length Vδ2 and Vγ2 cDNA from several rhesus monkey clones. The Vδ2 gene segment was highly conserved showing 88% similarity to the human Vδ2 gene segment (Fig. 4,A). Few amino acid changes were noted in CDR1 and CDR2. The Jδ1 and Jδ2 segments were highly conserved with no amino acid differences noted. Some Vγ2Vδ2⁺ T cell clones did not react with the 15D mAb that is specific for Vδ2 in humans (data not shown and Ref. 38). Sequencing the Vδ2 chain from these clones revealed that clones with weak or absent reactivity with the 15D mAb expressed the Vδ2 gene segment in conjunction with the Jδ1 junctional segment (Fig. 4,B). Vγ2Vδ2⁺ T cell clones strongly reactive with the 15D mAb expressed the Vδ2 gene segment in conjunction with the Jδ2 junctional segment. This suggests that the 15D mAb has a restricted reactivity with rhesus monkey γδ T cells preferentially reacting with Vδ2Jδ2 and possibly Jδ3 chains. Other anti-human Vδ2 mAbs lack reactivity to the rhesus monkey Vδ2 gene segment (eight of nine Vδ2-specific mAbs; Table I) suggesting that they may react with the same epitope(s) that are lost in rhesus monkeys.

Comparison of the amino acid sequence of the rhesus monkey, Aotus monkey, chimpanzee, and the human Vγ2 gene segment revealed 91% similarity between the rhesus monkey and human sequence (Fig. 4,C). The CDR1 and CDR2 of Vγ2 had three and two amino acid differences, respectively, that were mostly conservative amino acid changes (Fig. 4,E). The CDR3 region showed at least four amino acid differences in the Jγ1.2 segment that were all conservative changes, including an arginine for lysine difference (Fig. 4,D). We and others have proposed that basic residues in the Jγ1.2 segment may constitute contact residues for pyrophosphate binding (28, 39). The rhesus monkey Vγ2 TCR also showed conservation in Vγ2 CDR3 length with all three Vγ2 CDR3s within two amino acids in length (Fig. 4 D). This is similar to human Vγ2 TCR where 98% of adult Vγ2Vδ2 TCR have Vγ2 CDR3 lengths of ±1 of a modal value (40).

The Cγ and Cδ gene segments were also highly conserved. The Cδ segment showed 87% similarity with the human Cδ segment (Fig. 4,F) and the Cγ segment showed 93% similarity with the human Cγ segment (Fig. 4 G). Thus, rhesus monkey and human Vγ2Vδ2 TCRs are highly conserved in amino acid sequence with only minor differences noted in CDRs.

Because human and monkey Vγ2Vδ2 TCR differ in their CDRs (Fig. 4), we sought to determine whether these changes affect the fine Ag specificity of rhesus monkey Vγ2Vδ2⁺ T cells. A variety of different prenyl pyrophosphate Ags and analogs were used to stimulate human and rhesus monkey Vγ2Vδ2⁺ T cell clones. Despite the CDR differences between human and rhesus monkey Vγ2Vδ2 TCR, the relative potency of different analogs was identical such that the hierarchy of phosphoantigen bioactivity did not change (Fig. 5). For example, similar strong reactivity to IPP and ethyl pyrophosphate (EPP) and low reactivity to phenylethyl- and iso-amyl pyrophosphate was noted for both monkey and human Vγ2Vδ2⁺ T cells (Fig. 5). The rhesus monkey Vγ2Vδ2⁺ T cell clones shown were more sensitive to Ag stimulation as compared with human Vγ2Vδ2⁺ T cell clones but this was not a consistent finding with all monkey clones (such variation is also seen with human clones). Thus, despite differences in CDR sequence between human and monkey Vγ2Vδ2⁺ T cells, fine phosphoantigen specificity was unchanged suggesting that these differences do not affect nonpeptide Ag recognition.

To further study the conservation in recognition of nonpeptide Ag between rhesus monkey and human Vγ2Vδ2 TCR, human/monkey chimeric γδ TCR transfectants were derived by cotransfection and tested for their response to nonpeptide Ags by IL-2 release. TCR transfectants expressing either monkey Vγ2 paired with human Vδ2 or human Vγ2 paired with monkey Vδ2 were found to respond to HMBPP, EPP, and the bisphosphonate, risedronate, in a similar manner (Fig. 6,A). Detailed testing of the TCR transfectant expressing human Vγ2 paired with monkey Vδ2 revealed responsiveness to all of the major Ags defined for human Vγ2Vδ2⁺ T cells including prenyl pyrophosphate Ags and analogs, alkylamines, bisphosphonates, the bacterial phosphoantigen, HMBPP, and the B cell lymphoma, RPMI 8226 (Fig. 6,B). Moreover, the rank order of the response to the different prenyl pyrophosphate Ags and analogs was similar (compare Fig. 6,B, top panel, with Fig. 5). Thus, the rhesus monkey and human Vγ2 and Vδ2 gene segments can be interchanged without loss or major alteration in reactivity to any of the known nonpeptide Ags recognized by the Vγ2Vδ2 TCR.

Because there was close sequence similarity between the rhesus monkey and the human Vγ2Vδ2 TCR (Fig. 4), we made a homology model of the rhesus monkey Vγ2Vδ2 TCR based on the crystal structure of the human Vγ2Vδ2 TCR (39). Consistent with their close sequence similarities, there were few differences between the two TCRs when the carbon backbone was superimposed (Fig. 7, A and B). Despite the differences in the sequence of the CDRs, the position of the CDR loops showed little variation. The amino acid differences (colored red) in the CDR1 and CDR2 of the monkey and human Vγ2 chain are located in areas that are not the highest points on the CDR loops (Fig. 7 C). Moreover, when the surface potential of the two receptors was compared, both have similar basic (positively charged) regions in the CDR3 groove encoded by a lysine (K109) from the Jγ1.2 region of the CDR3 of Vγ2 and an arginine (R51) from the CDR2 loop of Vδ2. These two basic residues are conserved between rhesus monkeys, humans, and Aotus monkeys and are postulated to form a potential binding site for the pyrophosphate residues of phosphoantigens (28, 39). Consistent with this hypothesis, this region is required for Ag recognition because alterations in the CDR3 of the Vγ2 chain or direct mutation of the K109 lysine results in the loss of Ag recognition (41, 42). Another basic lysine residue, K108, that is required for phosphoantigen recognition (our unpublished observation and Ref. 42), exhibits a conservative change because the corresponding residue is an arginine in the rhesus monkey TCR. Thus, close similarities are noted in the structure of the rhesus monkey and human Vγ2Vδ2 TCRs with conservation of the CDR loop topology and in a basic region in the CDR3 groove.

In this study, we demonstrate that recognition of nonpeptide Ags, superantigens, and B cell lymphomas is conserved in rhesus monkey γδ T cells. Like in humans, phosphoantigens, bisphosphonates, and alkylamines stimulate the rhesus monkey Vγ2Vδ2⁺ T cell subset. Moreover, even the fine specificity for phosphoantigens exhibited by rhesus monkey Vγ2Vδ2⁺ T cells is identical to that of human Vγ2Vδ2⁺ T cells. Comparison of the primary amino acid sequence of rhesus monkey Vγ2 and Vδ2 gene segments revealed close similarities with amino acid conservation in all of the CDRs. This sequence similarity was sufficient to allow the expression of chimeric human/monkey Vγ2Vδ2 TCRs that recognized phosphoantigens identically to human Vγ2Vδ2 TCRs. Modeling of the rhesus monkey Vγ2Vδ2 TCR, revealed a similar topology to the CDR loops with identical amino acids in the highest points of the loops. These results suggest that the conservation of γδ T cell recognition extends to lower primates allowing their use as an animal model for human γδ T cells responses.

Rhesus monkey Vγ2Vδ2⁺ T cells showed complete conservation of nonpeptide Ag, lymphoma, and superantigen reactivity. Thus, responses were noted to prenyl pyrophosphates and analogs including IPP, EPP, and HMBPP. Moreover, there were minimal differences in the relative potency of the different analogs such that the hierarchy of reactivity did not significantly differ between rhesus monkey and human Vγ2Vδ2⁺ T cell clones. Similarly, reactivity to the other nonpeptide Ags, the alkylamines and the bisphosphonates (Figs. 2,B and 6, A and B) was conserved as was reactivity to the superantigen, SEA (Fig. 3). The conservation of Ag and superantigen reactivity probably reflects the conservation of the amino acid sequence of the Vγ2 CDR1, CDR2, and HV4 regions (Fig. 4,E) where there are few changes in the predicted critical contact residues (29). Reactivity to the uncharacterized Ag on the B cell lymphomas, Daudi (43), and RPMI 8226 (Fig. 6 B), was also conserved. Thus, reactivity to all of the known Ags for Vγ2Vδ2⁺ T cells was conserved in rhesus monkeys.

The Vγ2 and Vδ2 variable regions are highly conserved between the rhesus monkey and humans (91 and 88%, respectively). Vβ genes show a similar level of amino acid similarity because 17 different Vβ genes averaged 88% similarity to their human counterparts (44). The amino acid differences noted in the CDR1 and CDR2 of Vγ2 and Vδ2 variable regions are mainly conservative changes that are localized to less exposed regions of the TCR on our structural model (Fig. 7,C). Thus, the basic arginine residue (residue 43) from the CDR2 of the Vδ2 gene and the lysine residue from the Jγ1.2 region that makes up the basic region located in the groove of the Vγ2Vδ2 TCR are conserved in the three primates (Fig. 4,E). Rhesus (Fig. 4 B and Ref. 45) and Aotus monkeys (46) also express a hydrophobic residue at codon 109 in the CDR3 of the Vδ2 gene that is found in most human Vγ2Vδ2⁺ T cells (47).

Despite the importance of the CDR3 of the Vγ2 gene in determining reactivity to nonpeptide Ags (41, 42), the rhesus monkey Jγ1.2 gene differs at a number of amino acid residue from human and Aotus Jγ1.2 genes (Fig. 4 and Ref. 45). The first basic residue in the rhesus monkey Jγ1.2 gene segment is an arginine rather than the lysine found in humans and Aotus monkeys. This residue is located near the basic region composed of the second lysine residue of the Jγ1.2 gene with a basic residue from the CDR2 of Vδ2 (28, 39). Both of the basic residues in the human Jγ1.2 gene segment are critical for nonpeptide Ag recognition (42) and thus have been conserved in rhesus monkeys. Three other residues exhibit conservative replacements whereas a fourth residue, the aspartic acid found in the third position, is not present in rhesus monkeys. Although differing in amino acid sequence, rhesus monkey Vγ2Vδ2⁺ T cells are similar to human Vγ2Vδ2⁺ T cells (40, 47) in that they conserve the length of the CDR3 (Fig. 4 D and Refs. 45, 48). Thus, the length of the CDR3 may be more important than certain CDR3 amino acid residues.

The similarities in rhesus monkey and human Vγ2Vδ2 TCRs also extend to their fine specificity to phosphoantigens. Different analogs of IPP showed similar or identical hierarchies of potencies when different rhesus monkey Vγ2Vδ2 clones were compared with human Vγ2Vδ2 clones. Therefore, γδ T cell recognition of phosphoantigens differs from αβ T cell recognition of nonpeptide haptens and drugs presented by MHC class I and class II molecules. These αβ T cells commonly show differences in fine specificity for Ag between clones (49, 50) as well as specificity to the peptide conjugated to the hapten (51). In contrast, the differences in the CDR1, CDR2, and CDR3 seen between rhesus monkey and human Vγ2Vδ2 TCRs do not affect the fine Ag specificity. We speculate that if an Ag-presenting element for prenyl pyrophosphate Ags exists, it is relatively nonpolymorphic and may bind to the Vγ2Vδ2 TCR such that the Ag contacts a relatively small portion of a germline encoded region of the TCR. In this way, amino acid differences in the CDR3 and in other CDRs would not affect fine Ag specificity.

The conservation of nonpeptide Ag recognition in primates suggests that this type of recognition plays an important role in primate immune systems. Like rhesus monkeys, reactivity to phosphoantigens is conserved in Aotus monkeys (46) and recognition of the B cell lymphoma, Daudi, is conserved in chimpanzees (52). Thus, both Old World (Catarrhini) and New World (Platyrrhini) monkeys that separated ∼60 million years ago, as well as higher primates (chimpanzees and humans), are able to recognize nonpeptide Ags.

We recently demonstrated that, unlike innate immune cells, Vγ2Vδ2⁺ T cells can mount adaptive immune responses in response to infection with M. bovis BCG (27). We now show that the majority of γδ T cells from adolescent and adult rhesus monkeys from the New England Primate Center (Southborough, MA) express TCRs using Vδ1 (Fig. 1 and Table II). A similar predominance of Vδ1 T cells is seen in children before an environmentally driven expansion of Vγ2Vδ2⁺ T cells that occurs in most people between ages 3 and 10 (31). This expansion is presumably due to a common bacterial infection(s) that occurs in humans but that does not occur in the closed monkey colony. When the rhesus monkeys were infected with BCG, Vγ2Vδ2⁺ T cells expanded peaking at days 20–30 post infection. Secondary infection with BCG lead to a rapid expansion (as early as 5 days) that peaked earlier and with higher numbers of cells (up to 25–35% of all peripheral blood T cells). Accompanying this expansion of γδ T cells in the blood were higher levels of Vγ2Vδ2⁺ T cells in the pulmonary alveoli and intestinal mucosa and higher levels of expansion in vitro to IPP and mycobacterial prenyl pyrophosphate Ags. Importantly, the expansion of Vγ2Vδ2⁺ T cells correlated with the clearance of BCG organisms from the blood. Moreover, when monkeys were vaccinated with BCG, they could survive aerosol infection with Mycobacterium tuberculosis that rapidly killed unvaccinated monkeys. Protection was associated with earlier increases in alveolar Vγ2Vδ2⁺ T cells as compared with naive animals.

The rapid expansion of Vγ2Vδ2⁺ T cells upon infection of rhesus monkeys is identical to that seen in many human bacterial and protozoal parasitic infections (reviewed in Ref. 1). This broad reactivity of Vγ2Vδ2⁺ T cells is likely due to their recognition of a common metabolic intermediate, HMBPP. HMBPP is an intermediate in the deoxyxylulose synthetic pathway for IPP (our unpublished data and Ref. 53) that is highly active. This pathway is found only in Eubacteria, Apicomplexan parasites, and the chloroplasts of plants. Thus, the presence of this Ag in the blood or tissue is a strong indication of an infection. Because almost all adult Vγ2Vδ2⁺ T cells respond to nonpeptide Ags, a remarkably high precursor frequency of 1 T cell in 20 to 1 in 50 can respond to the prenyl pyrophosphates and alkylamines produced by bacteria or parasites. This type of recognition shares great similarities with innate immune recognition (54, 55) and suggest that γδ T cells are using their Ag receptors like pattern recognition receptors (1).

The advantage of this innate T cell recognition is the ability to focus T cell responses earlier in immune responses and more broadly than would be normally possible. Although the function of Vγ2Vδ2⁺ T cells is difficult to assess, Vγ2Vδ2⁺ T cells can kill infected cells and bacteria (22, 23) and secrete large amounts of TNF-α and IFN-γ, two important type 1 cytokines (25, 56). Human Vγ2Vδ2⁺ T cells stimulated with nonpeptide Ags protect “humanized” SCID mice from several species of bacteria including Staphylococcus aureus, Morganella morganii, and Escherichia coli (26). Vγ2Vδ2⁺ T cells can recognize a potential nonpeptide Ag in B cell lymphomas (57, 58) suggesting a role for these cells in tumor immunity. Vγ2Vδ2⁺ T cells also express NK receptors (59, 60), including NKG-2D (61), that may allow Vγ2Vδ2⁺ T cells to play a similar role to NK cells in antitumor and antiviral immunity including immunity to HIV/SIV (43, 62, 63). Furthermore, although few Ags for murine γδ T cells have been discovered, murine γδ T cells are required for optimal protection from infection with several bacterial species including Klebsiella pneumoniae (64), M. tuberculosis (65), Listeria monocytogenes (66), and Nocardia asteroides (67). Murine γδ T cells also decrease the severity of several autoimmune diseases (5) and inflammatory responses to infections (68, 69). Therefore, these studies suggest a role for γδ T cells in immunity to infection, tumors, and in autoimmunity.

The development of an animal model for Vγ2Vδ2⁺ T cells would greatly help the development of nonpeptide Ag vaccines that target these cells for their broad antibacterial and antiparasitic effects. Mice, rats, rabbits, guinea pigs, and ruminants do not appear to recognize known nonpeptide Ags because there is no reactivity seen in direct testing and because none of these animal species have V genes homologous to primate Vγ2 or Vδ2 genes. Our results show that rhesus monkeys can serve as animal models to study all facets of nonpeptide Ag recognition and Vγ2Vδ2⁺ T cell function in vivo. Indeed, i.v. infusion of rhesus monkeys with the weak phosphoantigen diphosphoglycerate is reported to increase in vitro expansions and production of IFN-γ and TNF-α by Vγ2Vδ2⁺ T cells upon IPP stimulation although this effect only persisted for 60 days (70). A more potent nonpeptide Ag vaccine that stimulates an adaptive Vγ2Vδ2⁺ T cell response could play an important role in a multicomponent vaccine for tuberculosis, leprosy, or anthrax infections or for protection from drug-resistant bacteria.

We thank Dr. Kia-Joo Puan and Chenggang Jin for critical reading of this manuscript. We thank Pamela Bruellman and Zhimei Fang for technical assistance. We thank Drs. Marc Bonneville, Genaro de Libero, Klaus Pfeffer, Simon Carding, Dieter Kabelitz, Frederick Treibel, and Lorenzo Moretta for the providing mAbs used in this report. We thank Dr. David Garboczi for providing the crystal coordinates for human Vγ2Vδ2 TCR, G115.