Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Cancer Immunol Immunother. Author manuscript; available in PMC 2010 October 4.
Published in final edited form as:
PMCID: PMC2948954

Individual Vγ2-Jγ1.2+ T cells respond to both isopentenylpyrophosphate and Daudi cell stimulation: Generating tumor effectors with low molecular weight phosphoantigens


Human Vγ2Vδ2 T cells exhibit T cell receptor-dependent, MHC-unrestricted recognition of antigen and play important roles in tumor and pathogen immunity. To characterize antigen recognition by the Vγ2Vδ2 TCR, we used the combined approach of spectratyping and CDR3 sequence analysis that measures changes in the TCR repertoire before and after stimulation with a phosphoantigen (isopentenyl pyrophosphate) or an irradiated tumor cell line (Daudi B lymphoma). Here we describe common Vγ2 chains that are substantially involved in the response to both phosphoantigens and tumor cells. The recognition properties of common Vγ2 chains explains the observation that Vγ2Vδ2 T cells expanded by phosphoantigen stimulation specifically recognize and kill some but not all tumor cell lines. Our studies further justify efforts to stimulate tumor immunity by administering low molecular weight phosphoantigens and boosting the frequency and tumor effector functions of circulating Vγ2Vδ2 T cells.

Keywords: Human, Tumor Immunity, T cells, Cytotoxic, Repertoire Development, T Cell Receptors, Gammadelta


A subset of CD3+ T cells marked by expression of the γδ T cell receptor (TCR) is important for tumor surveillance [20] and for control of tolerance, inflammation and pathogen immunity [1]. Human γδ T cells comprise 1-10% of circulating lymphocytes [5,7,13,34] and the major subset expresses the Vγ2Vδ2 TCR but not the lineage determinants CD4 or CD8 [2]. The Vγ2Vδ2 T cell population is largely effector memory [23] and is potently cytotoxic for the Daudi B cell lymphoma [30] and other tumor targets [10,21,26,39]. Because Vγ2Vδ2 T cells represent an abundant effector memory population that is cytotoxic for tumor cells, there is a growing interest in understanding these cells and exploiting their capabilities in the development of new immunotherapies against malignancies [27,41].

The lack of information about tumor antigens recognized by the Vγ2Vδ2 TCR impedes our understanding of γδ T cell regulation. We know that Vγ2Vδ2 T cells respond to stimulation with low molecular weight, non-peptidic antigens including intermediates of sterol biosynthesis (such as isopentenyl pyrophosphate) [9,16,18,35], aminobisphosphonate drugs [11,27] and alkylamines [3]. Vγ2Vδ2 T cells proliferate after stimulation with these agents and respond similarly to some tumor cell lines including the Daudi B cell lymphoma [4,22,25,30]. Tumor recognition is independent of classical MHC molecules, there is no defined requirement for antigen processing [13], and various tumor ligands have been implicated including HSP60 [28] and an F1 ATPase-related structure [37]. Cell lines expanded after stimulation with phosphoantigens or tumors are enriched for the Vγ2-Jγ1.2 rearrangement [18] and are potently cytotoxic for some tumors [10,38]. Stimulation also induces expression of Type 1 cytokines (IFN-γ, TNF-α, RANTES, MIP-1α, MIP-1β and Lymphotactin among others) [8,14,29,40] and can lead to surface expression of CD107a, a marker for cytotoxic effector cells [14]. Experimental data argues against T cell receptor binding to phosphoantigens, aminobisphosphonates or alkylamines and we do not understand why some (but not all) tumor cell lines are stimulatory or why Vγ2Vδ2 T cell lines are potently cytotoxic for tumor cells. Part of the problem reflects unusual features of the γδ TCR repertoire.

Beginning early in life, a strong bias emerges for the Vγ2Vδ2 TCR in blood γδ T cell populations [7]. Within the Vγ2 repertoire there is selection for chains expressing the Jγ1.2 segment [17], one of 5 functional Jγ segments. Repertoire bias is a product of chronic positive selection for Vγ2-Jγ1.2 that is driven by self or ubiquitous antigens. The same bias is common to most human beings irrespective of haplotype [12,17] and is nearly identical in human and nonhuman primates [6,18,36]. In contrast, the highly diverse αβ TCR repertoire is shaped by MHC-dependent positive and negative selection mechanisms in the thymus and does not show an obvious bias in the absence of infection or immunization [33].

Studies on Vγ2Vδ2 T cells are hampered by having few of the tools that are used commonly to study αβ TCR recognition. A lack of known presenting molecules coupled with the inability to define tumor antigens has prevented the use of tetramers, epitope peptides, or similar approaches to study the γδ TCR. Accordingly, we looked directly at the Vγ2 repertoire and its changes after treatment in vitro with model tumor cell (Daudi B cell lymphoma) or phosphoantigen (isopentenyl pyrophosphate) stimulators to understand the mechanism for antigen recognition. We define similarities and differences in the response to tumor cells or phosphoantigens by looking for Vγ2 chain sequences that are expanded preferentially after exposure to either stimulus. What emerges is a map of the complex responses for an entire T cell population exposed to two different antigens. The balance of common and specific recognition describes the capacity of this unique T cell subset to be triggered by phosphoantigens and respond with tumor cytotoxicity

Materials and Methods

Collection and preparation of lymphocytes

Heparinized blood was collected from six healthy volunteers with approval from the Institutional Review Board at the University of Maryland Baltimore and informed consent of the donors. Peripheral blood mononuclear cells were isolated by centrifugation over Ficoll-Paque density gradients as described by the manufacturer (Pharmacia, Uppsala, Sweden). Approximately 5×105 cells were cultured in supplemented RPMI-1640 medium (R10) containing 10% fetal bovine serum (FBS), 2mM L-glutamine (Gibco, Grand Island, NY), 1 U/mL penicillin/streptomycin (GIBCO, Grand Island, NY) and 100 U/mL recombinant human interleukin-2 (rhIL-2) (Tecin, Biological Resources Branch, NIH, Bethesda, Maryland). Isopentenyl pyrophosphate (IPP) (Sigma, St. Louis, MO) was added to a final concentration of 15μM or irradiated (120 Gy) Daudi B cells were added at a ratio of 2:1 Daudi: PBMC. Cultures were incubated for two weeks at 37°C with 5% CO2 and were replenished every 3-4 days by adding IL2-supplemented medium without IPP or Daudi.

Tumor Cell Lines

Daudi B cells (CCL-213, ATCC) and MOLT -4 T cells (CRL-1582, ATCC) were cultured in R10 additionally supplemented with 4.5 g/L glucose, 1.5 g/L NaHCO3, 10mM HEPES and 1mM sodium pyruvate. Several tumor cell lines were selected because of their derivation from HIV-associated malignancies, including: BC-1, BC-2, BL-5, BL-8, IBL-4 (provided kindly by William Harrington, University of Miami) and 2F7. These lines contain no HIV proviral DNA and express no detectable HIV antigens. BC-1 (CRL-2230, ATCC) and BC-2 (CRL-2231, ATCC) were cultured in RPMI-1640 supplemented with 2mM L-glutamine, 1 U/mL penicillin/streptomycin and 20% non-heat inactivated FBS. B cell tumor lines BL-5, BL-8 and IBL-4 were kindly provided by Dr. William Harrington (University of Miami, Miami, FL) and cultured in IMDM supplemented with 2mM L-glutamine, 1 U/mL penicillin/streptomycin and 10% non-heat inactivated FBS. T cell line 2F7 (CRL-10237, ATCC) is cultured in R10 supplemented with 0.05mM β-mercaptoethanol (GIBCO, Grand Island, NY).

Calcein-release Cytotoxicity Assay

To quantify the cytotoxic capacity of Vγ2Vδ2 T cells we used a non-radioactive, fluorometric cytotoxicity assay involving the dye calcein-acetoxymethyl (calcein-AM)[32]. Tumor targets were labeled for 15 minutes with 2μM calcein-AM (Molecular Probes, Eugene, Oregon) at 37°C in 5% CO2 and then washed once with phosphate-buffered saline (PBS). Cells were combined at various effector:target (E:T) ratios in 96 well V-bottom microtiter plates (Corning Inc., Corning, New York) and incubated at 37°C in 5% CO2 for four hours; assays were performed in triplicate. Following incubation, supernatants were transferred to a 96-well flat-bottomed microtiter plates and calcein content is measured using a Wallac Victor2 1420 multi-channel counter (λ485/535nm). Percent specific lysis is calculated as: [(test release – spontaneous release)/(maximum release – spontaneous release)] × 100.

Flow Cytometry

Two weeks after Daudi+IL2, IPP+IL2 or control (IL2 alone) stimulation, cultures were harvested and viable counts were performed by the trypan blue exclusion assay. 3×105 cells were washed once in RPMI-1640 and stained at 4°C with fluoroscein isothiocyanate (FITC)-conjugated CD3 (Clone UCHT1; BD Biosciences, San Diego, CA) and phycoerythrin (PE)-conjugated anti-Vδ2 (clone B6; BD Biosciences, San Diego, CA) or the appropriate isotype controls. After 20 minutes, cells were washed once with RPMI-1640 and resuspended in PBS containing 2% paraformaldehyde. At least 104 lymphocytes (gated on the basis of forward and side scatter profiles) were acquired for each sample on a FACSCalibur (BD Biosciences, San Diego, CA). Flow cytometry data were analyzed with FloJo software (Tree Star, San Carlos, CA). Stimulation index (SI) represents the fold increase for Vδ2 lymphocytes after IPP or Daudi stimulation compared to IL2 alone. SI is calculated as the ratio of the absolute number of Vδ2 lymphocytes on Day 14 to the absolute number of Vδ2 lymphocytes expanded with IL2 alone.

RNA Extraction and reverse transcription-polymerase chain reaction (RT-PCR)

Total RNA was extracted from at least 106 cells using the RNeasy Mini Kit as described by the manufacturer (Qiagen, Valencia, CA). One microgram of total RNA was then converted into cDNA using the Reverse Transcription System (Promega, Madison, WI) in a reaction containing: 500 ng of oligonucleotide A (T15V), 1 mM deoxynucleotriphosphates, 5 mM MgCl2, 10mM Tris-HCL pH 8.8, 50 mM KCL, 0.1% Triton X-100, 18 units of avian myeloblastosis virus reverse transcriptase and 10 units of RNasin ribonuclease inhibitor. Each reaction was incubated at 42°C for 2 hours, then cDNA were diluted to 100 μL by adding to the reaction 80 μL deionized water. PCR reactions were performed using 5 μL cDNA as template and 500 nM each of formard and reverse primers, 0.2 mM dNTPs, 2 mM MgCl2, 10 mM Tris-HCl pH 8.8, 50 mM KCl, 0.1% Triton X-100 and 1 unit of AmpliTaq Gold (Applied Biosystems, Foster City, CA). The following primers were used: oligo-Vγ2 (5’-ATCAACGCTGGCAGTCC-3’), oligo-Cγ1 (5’-GTTGCTCTTCTTTTCTTGCC-3’), 5’ β-actin (5’-GTGGGGCGCCCCAGGCACCA-3’) and 3’ β-actin (5’-CTCCTTAATGTCACGCACGATTTC-3’). PCR was run with the following profile: denaturation for 1 minute at 94°C; 5 minutes at 68°C; 45 cycles (45 seconds at 94°C, 1 minute at 60°C, 1 minute at 72°C); extension for 10 minutes at 72°C. PCR products were separated on a 2% agarose/Tris-acetate-ethylenediaminetetraacetic acid buffer (TAE) gels containing 0.5 μg/mL ethidium bromide.

Spectratype Analysis

Primer extension reactions were performed as described previously [18]. Each reaction contained 1 μL PCR product, 3 mM MgCl2, 0.2 mM dNTP, 0.2 units Taq DNA polymerase (Promega, Madison, WI), 10 mM Tris-HCL pH 8.8, 50 mM KCl, and 0.1 μM –carboxyfluorescein (6-FAM)-labeled primer (Cγ6-FAM for Vγ2 chains: 5’-AATAGTGGGCTTGGGGGAAAC-3’; Cδ16-FAM for Vγ2 chains: 5’-ACGGATGGTTTGGTATGAGG-3’). Four microliters of run-off products were diluted in deionized formamide and 1 μL of N,N,N’,N’-trimethyl-6-rhodamine-labelled molecular size standard was added to each sample. After a denaturation step (5 minutes at 95°C followed by immediate quenching on ice), products were loaded on an Applied Biosystem microcapillary genetic analyzer (Perkin-Elmer, Foster City, CA) and run for either 27 (Vγ2 chains) or 24 (Vγ2 chains) minutes on a performance optimized polymer (POP-4). Molecular size and relative frequency of extension products were determined using Genescan software (Perkin-Elmer, Foster City, CA). To standardize the data irrespective of the run-off primer position, CDR3 length variation was expressed in terms of the total Vγ2 or Vγ2 coding region lengths. Run-off product lengths were corrected by adding the length of the known Vγ2 or Vδ2 mRNA coding regions outside the run-off product. According to this calculation, the major peak for Vγ2 chains is 996 nucleotides and for Vδ2 chains is 1029 nucleotides with corresponding run-off product lengths of 447 and 193 nucleotides, respectively.

Cloning and Sequencing

PCR products were purified by gel extraction using a Qiaquick Gel Extraction kit (Qiagen, Valencia, CA) according to the manufacturer's instructions. Purified products were denatured (1 minute at 94°C), incubated for 30 minutes at 72°C with 2 mM MgCl2, 0.2 mM dATP and 2.5 units Amplitaq Gold (AB, Foster City, CA), then ligated into the pCR2.1 TOPO TA cloning vector (Invitrogen, Carlsbad, CA). Ligated vector was transfected into TOP10F’ competent cells using the TA Cloning kit (Invitrogen) and bacterial colonies were grown overnight on agar plates containing 50 μg/mL ampicillin, 500 μM isopropylthiogalactoside (IPTG) and 80 μg/mL bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-gal). Colonies containing recombinant plasmids were cultured overnight and plasmid DNA were purified using the Wizard Plus Minipreps DNA purification kit (Qiagen, Valencia, CA). Sequencing reactions were done with the Bigdye v3.1 fluorescent sequencing kit (AB, Foster City, CA) using both M13F and M13R oligonucleotide primers for each sample. Sequences were analyzed on an automated sequencer ABI3700 and aligned using Sequencher (GeneCodes Corp, Ann Arbor, MI) and MacClade software.

Vγ2, Jγ1.1, Jγ1.2, Jγ1.3/2.3, Jγ2.1, and Cγ1 were identified by comparison to GenBank accession numbers X07205, X08084, M12950, M12960/M12961, M16016, M14996, respectively and TCRγ locus reference sequences NG_001336 and AF159056.


The Adult Human Vγ2Vδ2 Repertoire is Skewed Toward Longer Vg2 Chain Lengths

Over several years we collected data on the Vγ2 repertoire in healthy, HIV-negative adults. At present, we have data for 44 healthy adult volunteers residing in North America or Africa; the distribution of Vγ2 and Vδ2 chain lengths were determined by spectratype analysis. Similar to previous results, we found that in healthy adult humans the Vγ2 repertoire is skewed toward longer Vγ2 chain lengths (Figure 1), without an overall bias in the distribution of Vδ2 chains (data not shown). Selection for longer Vγ2 chains also coincides with the specific accumulation of Vγ2-Jγ1.2 chains and is estimated by calculating the frequency of Vγ2 chains greater than 987 nucleotides from spectratype data. Taken together, approximately 68% of all Vγ2 chains co-express the Jγ1.2 segment in adults. Thus, selection for the Jγ1.2 segment is a common feature of the adult Vγ2 repertoire and occurs independently of ethnic or geographic background. Part of these data were reported previously [17,18].

Figure 1
The Biased Vγ2 Chain Repertoire in Healthy Adults

Isopentenyl Pyrophosphate and Daudi Tumor Cells Expand Vγ2+ lymphocytes

We selected six representative donors and stimulated PBMC in vitro with interleukin-2 (IL2) alone, 15 μM IPP and IL2, or irradiated (120 Gy) Daudi tumor cells plus IL2. Healthy donors varied in age, sex and ethnicity (Table 1). After two weeks in culture (and periodic supplementation with medium containing IL2), there were strong proliferative to IPP or Daudi stimulation (Table 1). Day 14 cultures were comprised of greater than 60% lymphocytes as determined by forward and side scatter flow cytometric profiles, with the remainder consisting of dead or dying cells. Staining for CD3+CD4+ and CD3+CD8+ lymphocytes showed no significant change in the absolute number of these subsets (data not shown) although we observed elevated NK lymphocytes (CD3-CD8+) only after Daudi stimulation. Additionally, we used irradiated cells from several tumor cell lines (BC-1, BC-2, 2F7, MOLT -4, BL-5, BL-8 and IBL-4) to stimulate Vγ2Vδ2 lymphocytes from PBMC. Using the conditions described here, only Daudi and IBL4 elicited proliferative responses.

Table 1
Expansion Profile of Vδ2 T cells from Normal, Healthy Donors

Both IPP or Daudi Expand Potent Vγ2Vδ2 Tumor Effectors

We expanded Vγ2Vδ2 T cells from PBMC using IPP or Daudi stimulation and tested each line for the ability to lyse three tumor cell targets: Daudi B lymphoma cells, MOLT-4 T cells and IBL4 B lymphoma cells. Effector cells stimulated with either IPP or irradiated Daudi were cytotoxic for all three tumor cell lines tested though we noted differences in potency (Figure 2). MOLT-4 was unusual in that it did not stimulate Vγ2Vδ2 proliferation, but was lysed by Vγ2Vδ2 effector cells. The majority of lymphoma cell lines tested by us and others [38] neither stimulate Vγ2Vδ2 proliferation nor are killed by expanded effector cells. It is important to note that some expanded Vγ2Vδ2 lines contained other lymphocytes including NK cells that might contribute to tumor cytolysis. However, we have observed high levels of tumor killing with lines that were highly enriched (>95%) for Vγ2Vδ2 T cells and effectively lysed tumor cells with effector to target ratios below 1. Further, the potency of effectors (in terms of specific lysis at a given E:T ratio) was the same after IPP or Daudi stimulation despite slightly higher NK cell levels in the Daudi-expanded cultures.

Figure 2
IPP- or Daudi-generated Vγ2Vδ2 Effectors Potently Lyse Tumor Cell Line Targets

IPP or Daudi Selectively Skew Vγ2 Chains Toward Longer Chain Lengths

To understand how IPP or Daudi stimulation remodeled the Vγ2Vδ2 repertoire, we performed spectratype analyses to assess the size distribution for Vγ2 and Vδ2 chains after alkylphosphate or tumor stimulation. In all donors, both IPP and Daudi stimulation and subsequent cell proliferation skewed the Vγ2 repertoire toward longer chain lengths but did not alter the length distribution for Vδ2 chains. This was evident when cumulative frequencies for both Vγ2 and Vδ2 chains from all six donors were averaged and plotted (Figure 3). Both IPP and Daudi skewed the Vγ2 repertoire toward chain lengths greater than 987 nucleotides as evidenced by rightward shifts in the cumulative frequency plots. Comparisons of Vγ2 chain lengths before and after stimulation for all six donors did not reach statistical significance. However, the starting Vγ2 repertoires of donors A and E were almost exclusively Vγ2 chains longer than 987 nucleotides and it was difficult to detect a shift in these highly biased samples (data not shown). When these donors were excluded from the statistical analysis, we observed a significant shift for the remaining four donors after Daudi stimulation (p<0.05). Vγ2 chain lengths were normally distributed in all samples and cumulative frequency plots for Vδ2 were similar before and after stimulation with IL2, IPP or Daudi (Figure 3), whether or not donors A and E were excluded. In terms of chain length, the effects of antigen stimulation were most apparent in the Vγ2 repertoire and we continued our study with emphasis on that chain.

Figure 3
CDR3 Length Distribution in the Population of Vγ2 or Vδ2 Chains Before and after Stimulation with IL2 alone, 15 μM IPP and IL2 or irradiated Daudi and IL2

IPP or Daudi Selectively Expand Vγ2-Jγ1.2+ Cells

We sequenced Vγ2 chains before or after IPP or Daudi stimulation for 4 donors, all of which had significant differences between the starting and expanded spectratypes. The occurrence of discrete Vγ2 CDR3 nucleotide sequences (nucleotypes), predicted amino acid sequences (clonotypes), Jγ segment usage and predicted chain lengths were determined for each donor before and after stimulation. Confirming and extending our previous results [18], we observed a positive selection for Vγ2-Jγ1.2 chains following stimulation with either IPP or Daudi (Figure 4). This result was most dramatic in donors whose starting repertoire had a lower abundance of Vγ2-Jγ1.2 chains. In all cases, sequence analyses confirmed the results obtained by spectratyping (data not shown).

Figure 4
IPP or Daudi Selectively Expand Vγ2-Jγ1.2 chains

Unlike αβ T cell repertoire studies where starting frequencies of particular clonotypes are below the threshold of detection by spectratyping or DNA sequencing, the Vγ2 model provides an unique opportunity to observe repertoire dynamics. We calculated the relative change for each clonotype during a proliferative expansion. The ratio of clonotype frequency before and after stimulation was multiplied by the stimulation index for that culture (Table 1). This produced a relative stimulation index for each clonotype that measured the strength of response for cells expressing individual Vγ2 chains (Figure 5A). Plotting the relative stimulation index for each clonotype after IPP or Daudi treatment (Figure 5A) demonstrated that the population contained clonotypes that varied greatly in their response to each stimulus. As expected, most responding clonotypes and all of those responding to both IPP and Daudi, used the Vγ2-Jγ1.2 rearrangement (data not shown). Additionally, Vγ2 clonotypes that responded to both stimuli (common recognition) were often public chains because they were present in multiple, unrelated donors (Figure 5B).

Figure 5
Examples of Common or Specific Recognition of IPP and Daudi by Vγ2-Jγ1.2 Chains

The Vγ2 chains that responded to both IPP and Daudi comprised a significant fraction of the starting Vγ2 repertoire for each of these four donors. There was a trend among all four donors that IPP or Daudi promoted the expansion of common recognition Vγ2 clonotypes, many of which were also public chains.

Sequences of Vγ2-Jγ1.2 Chains that Respond to IPP, Daudi or Both

We divided all Vγ2-Jγ1.2 clonotypes into 3 groups based on the calculated stimulation indices to IPP or Daudi. Groups comprised clonotypes involved in common (Figure 6A), Daudi-specific (Figure 6B) or IPP-specific (Figure 6C) recognition. Relative stimulation indices varied from 1.6 (Figure 6B, clonotype 3-02) to 1865 (Figure 6A, clonotype 4-01). Although the absolute values varied, the ratios of Daudi SI to IPP SI were generally consistent for individual clonotypes from different donors. For example, the ratio of stimulation indices to IPP versus Daudi were similar in donors A and F for the clonotype 4-01, LWETQELG (Figure 6A). A similar consistency was observed for clonotype 4-03 in donors A and F, even though this sequence behaved differently for donors B and C. Finally, the common clonotype 4-06, LWEVRELG, reacted more to Daudi than to IPP for donors C and F. Common recognition clonotypes were encoded by 3 nucleotypes on average, while the majority of IPP- (6/7) or Daudi-(4/6) specific clonotypes were encoded by a single nucleotype. The overall trend was for increased nucleotype redundancy among the common recognition Vγ2-Jγ1.2 clonotypes.

Figure 6
Multiple Vγ2 Nucleotypes Converge on Clonotypes that Recognize both IPP and Daudi

The three designated subsets of Vγ2 clonotypes were aligned separately to search for consensus sequences in their CDR3 regions. We could not define consensus Vγ2 CDR3 that predicted common or specific recognition (Figure 6).

The Vγ2 Repertoire is Relatively Stable for Years

In order to justify the development of therapies that target Vγ2Vδ2 T cells, it is important to know that the pool of potential tumor effector cells is stable over extended intervals. For these studies, we relied on DNA sequencing data to compare the Vγ2 repertoire among individual donors at more than one time point. Initially, we compared the Vγ2 repertoire in eight donors for samples collected two years apart (Fig. 7A). On average, we had approximately 40 sequences for each time point. We identified Vγ2 nucleotypes that were present in both samples and then calculated the proportion of the Vγ2 repertoire occupied by these nucleotypes at each time point. Two donors (C and E) had conserved sequences occupying between 25 and 50% of the total repertoire and represented the lowest stability. Other donors including A and D were most stable with more than 75% of the Vγ2 repertoire occupied by conserved sequences.

Figure 7
Vγ2 Repertoire is Stable For Many Years

We performed a similar analysis for three donors where Vγ2 sequence data existed for samples collected 7 years earlier (Figure 7B). Data for the early timepoint were reported previously [18]; data for the recent timepoint were generated during this study. On average, we had approximately 45 sequences for each time point. Remarkably, these three donors also showed a high degree of repertoire stability overt his extended interval, with greater than 50% of all detected Vγ2 sequences being conserved. Overall, the Vγ2 repertoire is highly stable in healthy adults. It will be important for future studies to show whether changes in this repertoire might presage an increased risk for malignancy.


Circulating Vγ2Vδ2 T cells exhibit robust proliferative responses to IPP or Daudi stimulation and cells expanded in either condition are comparably cytotoxic for the same Daudi cells. Analysis of the Vγ2 chain repertoire showed a substantial, though incomplete overlap between clones expanded by either stimulus and this accounted for the reported ability to expand γδ T cells from PBMC with phosphoantigens and obtain a line equipped for cytolytic recognition of Daudi [38]. Individual Vγ2 clonotypes that responded in both stimulatory conditions were designated “common recognition” chains; they constituted a substantial portion of the starting repertoire in multiple donors and were expanded preferentially in stimulated cultures.

The ability to distinguish common (responding to both stimuli) and specific (responding to one stimulus) recognition clonotypes derives from unique features of the Vγ2 repertoire. In adults, the major population of Vγ2 chains in circulating T cells is the product of chronic selection for cells expressing the Vγ2-Jγ1.2 rearrangement, likely in response to self or ubiquitous antigens [24]. Cells expressing other Vγ2-Jγ segment rearrangements dominate the immature thymic or cord blood γδ repertoire, but comprise only a minor portion of the mature Vγ2 repertoire (unpublished observations and [7]). Selective pressures favoring the Vγ2-Jγ1.2 rearrangement reduce diversity to such an extent that identical clonotypes are often present among unrelated donors. The Vγ2 repertoire is relatively stable over extended intervals but we also note that rare adult donors have few Vγ2-Jγ1.2 cells and consequently poor functional responses, despite being otherwise healthy.

The abundance of particular Vγ2 clonotypes in the starting γδ TCR repertoire provides an unique opportunity to determine the magnitude of change during an expansion. Such measurements are not usually possible for αβ T cells, because the starting frequency for any given clone is in the range of 10-5 to 10-6, and it is impractical to find these rare events by a bulk cDNA sequencing effort. We were able to measure the extent of proliferation for individual Vγ2 T cell clones and determine whether they respond better to one or both of the model stimuli. Many Vγ2 clonotypes were expanded after IPP or Daudi stimulation. In most cases, the ratio of proliferation after IPP versus Daudi was similar among donors, even though the absolute stimulation indices varied. We interpret this to mean that the response was TCR-dependent but cell expansion required other factors in the culture, likely including monocytic and dendritic cell subsets that are known to promote Vγ2Vδ2 T cell proliferation [15,31]. It is important to note that few studies have ever compared the performance of identical, naturally occurring T cell receptor clonotypes among different individuals. The degree of consistent responses is a remarkable finding and suggests a fundamental role for Vγ2Vδ2 T cells.

The dominant selection for common recognition Vγ2 clonotypes explains why IPP or other phosphoantigen stimulators produce potent antitumor effectors. Quite apart from suggesting that the Vγ2 repertoire is an example of degenerate or non-specific recognition, these data show a high degree of specificity as evidenced by the reproducible behavior of public clonotypes in different donor PBMC.

This work highlights important differences between TCR repertoire for γδ and αβ T cell populations. Unlike the starting αβ T cell repertoire, we found numerous examples of public Vγ2 clonotypes in the starting repertoire nearly all of which expressed the Jγ1.2 segment; selection produced an astonishingly similar repertoire among unrelated individuals. Similar to what has been observed for public αβ T cell repertoire, public Vγ2 clonotypes have minimal CDR3 diversity and are predominantly composed of germline-encoded residues [19]. Historically, the disproportionate engagement of MHC molecules with CDR1/2 of the αβ TCR was proposed to explain the phenomenon of public responses with minimal CDR3 diversity in these cells [19]. Our observation that both IPP and Daudi expand public Vγ2 clonotypes irrespective of haplotype argues against the MHC-imprinting hypothesis for γδ T cells and may reflect the inherent capacity of germline-encoded TCR to recognize conserved, non-MHC restricted antigens.

The mature Vγ2Vδ2 T cell repertoire encodes an effector subset that is primed for rapid proliferative and cytolytic responses to phosphoantigens, tumors and infected cells. This finding supports the efforts to stimulate antitumor responses in cancer patients by treatment with phosphoantigens [27,41]. The Vγ2 repertoire evolves in response to stimulation by ubiquitous or self molecules, possibly including normal metabolites like IPP. The common selection in most persons and the absence of MHC restriction creates a repertoire dominated by public clonotypes that recognize diverse stimuli. Vγ2Vδ2 T cells provide a model for understanding chronic T cell responses that defy peripheral tolerance, and are poised for responses to pathogens and tumor cells.


We thank William Harrington (University of Miami) for kindly providing the BL-5, BL-8 and IBL4 tumor cell lines and we are grateful to Dr. Martin Flajnik, Dr. Maria Salvato and Dr. Adrian Hayday for helpful discussions. Supported by PHS grants AI51212 and CA113261 (C.D.P.).

List of Works Cited

1. Boismenu R, Havran WL. An innate view of gamma delta T cells. Curr Opin Immunol. 1997;9:57–63. [PubMed]
2. Brenner MB, McLean J, Scheft H, Riberdy J, Ang SL, Seidman JG, Devlin P, Krangel MS. Two forms of the T-cell receptor gamma protein found on peripheral blood cytotoxic T lymphocytes. Nature. 1987;325:689–694. [PubMed]
3. Bukowski JF, Morita CT, Brenner M. Human gamma delta T cells recognize alkylamines derived from microbes, edible plants, and tea: implications for innate immunity. Immunity. 1999;11:57–65. [PubMed]
4. Burns J, Lobo S, Bartholomew B. Requirement for CD4+ T cells in the gammadelta T cell proliferative response to Daudi Burkitt's lymphoma. Cell Immunol. 1996;174:19–24. [PubMed]
5. Caccamo N, Dieli F, Wesch D, Jomaa H, Eberl M. Sex-specific phenotypical and functional differences in peripheral human Vgamma9/Vdelta2 T cells. J Leukoc Biol. 2006;79:663–666. [PubMed]
6. Cairo C, Propp N, Hebbeler AM, Colizzi V, Pauza CD. The Vgamma2/Vdelta2 T-cell repertoire in Macaca fascicularis: functional responses to phosphoantigen stimulation by the Vgamma2/Jgamma1.2 subset. Immunology. 2005;115:197–205. [PubMed]
7. Casorati G, De Libero G, Lanzavecchia A, Migone N. Molecular analysis of human gamma/delta+ clones from thymus and peripheral blood. J Exp Med. 1989;170:1521–1535. [PMC free article] [PubMed]
8. Cipriani B, Borsellino G, Poccia F, Placido R, Tramonti D, Bach S, Battistini L, Brosnan CF. Activation of C-C beta-chemokines in human peripheral blood gammadelta T cells by isopentenyl pyrophosphate and regulation by cytokines. Blood. 2000;95:39–47. [PubMed]
9. Constant P, Davodeau F, Peyrat MA, Poquet Y, Puzo G, Bonneville M, Fournie JJ. Stimulation of human gamma delta T cells by nonpeptidic mycobacterial ligands. Science. 1994;264:267–270. [PubMed]
10. Corvaisier M, Moreau-Aubry A, Diez E, Bennouna J, Mosnier JF, Scotet E, Bonneville M, Jotereau F. V gamma 9V delta 2 T cell response to colon carcinoma cells. J Immunol. 2005;175:5481–5488. [PubMed]
11. Das H, Wang L, Kamath A, Bukowski JF. Vgamma2Vdelta2 T-cell receptor-mediated recognition of aminobisphosphonates. Blood. 2001;98:1616–1618. [PubMed]
12. Davodeau F, Peyrat MA, Hallet MM, Gaschet J, Houde I, Vivien R, Vie H, Bonneville M. Close correlation between Daudi and mycobacterial antigen recognition by human gamma delta T cells and expression of V9JPC1 gamma/V2DJC delta-encoded T cell receptors. J Immunol. 1993;151:1214–1223. [PubMed]
13. De Libero G, Casorati G, Giachino C, Carbonara C, Migone N, Matzinger P, Lanzavecchia A. Selection by two powerful antigens may account for the presence of the major population of human peripheral g/d T cells. J Exp Med. 1991;173:1311–1322. [PMC free article] [PubMed]
14. Deetz C, Hebbeler A, Propp N, Cairo C, Tikhonov I, Pauza C. Interferon-gamma Secretion by Human Vg2Vd2 T Cells After Stimulation with Antibody Against the T Cell Receptor plus the Toll-like Receptor 2 Agonist, Pam3Cys. Infection and Immunity. 2006 in press. [PMC free article] [PubMed]
15. Devilder MC, Maillet S, Bouyge-Moreau I, Donnadieu E, Bonneville M, Scotet E. Potentiation of antigen-stimulated V gamma 9V delta 2 T cell cytokine production by immature dendritic cells (DC) and reciprocal effect on DC maturation. J Immunol. 2006;176:1386–1393. [PubMed]
16. Eberl M, Hintz M, Reichenberg A, Kollas AK, Wiesner J, Jomaa H. Microbial isoprenoid biosynthesis and human gammadelta T cell activation. FEBS Lett. 2003;544:4–10. [PubMed]
17. Enders PJ, Yin C, Martini F, Evans PS, Propp N, Poccia F, Pauza CD. HIV-Mediated gammadelta T Cell Depletion Is Specific for Vgamma2(+) Cells Expressing the Jgamma1.2 Segment. AIDS Res Hum Retroviruses. 2003;19:21–29. [PubMed]
18. Evans PS, Enders PJ, Yin C, Ruckwardt T, Malkovsky M, Pauza CD. In vitro stimulation with a non-peptidic alkylphosphate antigen expands cells expressing the Vγ2-Jγ1.2/Vδ2 T cell receptor. Immunology. 2001;104:19–27. [PubMed]
19. Fazilleau N, Cabaniols JP, Lemaitre F, Motta I, Kourilsky P, Kanellopoulos JM. Valpha and Vbeta public repertoires are highly conserved in terminal deoxynucleotidyl transferase-deficient mice. J Immunol. 2005;174:345–355. [PubMed]
20. Ferrarini M, Ferrero E, Dagna L, Poggi A, Zocchi MR. Human gammadelta T cells: a nonredundant system in the immune-surveillance against cancer. Trends Immunol. 2002;23:14–18. [PubMed]
21. Fisch P, Malkovsky M, Braakman E, Sturm E, Bolhuis RL, Prieve A, Sosman JA, Lam VA, Sondel PM. Gamma/delta T cell clones and natural killer cell clones mediate distinct patterns of non-major histocompatibility complex-restricted cytolysis. Journal of Experimental Medicine. 1990;171:1567–1579. [PMC free article] [PubMed]
22. Fisch P, Malkovsky M, Kovats S, Sturm E, Braakman E, Klein BS, Voss SD, Morrissey LW, DeMars R, Welch WJ, Bolhuis RLH, Sondel PM. Recognition by human V gamma 9/V delta 2 T cells of a GroEL homolog on Daudi Burkitt's lymphoma cells. Science. 1990;250:1269–1273. [PubMed]
23. Gioia C, Agrati C, Casetti R, Cairo C, Borsellino G, Battistini L, Mancino G, Goletti D, Colizzi V, Pucillo LP, Poccia F. Lack of CD27-CD45RA-V gamma 9V delta 2+ T cell effectors in immunocompromised hosts and during active pulmonary tuberculosis. J Immunol. 2002;168:1484–1489. [PubMed]
24. Halary F, Fournie JJ, Bonneville M. Activation and control of self-reactive gammadelta T cells. Microbes Infect. 1999;1:247–253. [PubMed]
25. Kaur I, de Jong J, Schell K, Hank J, Fisch P, Sondel PM. Human peripheral gamma delta T cells are stimulated by Daudi Burkitt's lymphoma and not by any other Burkitt's lymphoma tested. Cell Immunol. 1994;156:54–61. [PubMed]
26. Kobayashi H, Tanaka Y, Yagi J, Toma H, Uchiyama T. Gamma/delta T cells provide innate immunity against renal cell carcinoma. Cancer Immunol Immunother. 2001;50:115–124. [PubMed]
27. Kunzman V, Feurle J, Weissinger F, Tony H, Wilhelm M. Stimulation of gd T cells by aminobisphosphonates and induction of antiplasma cell activity in multiple myeloma. Blood. 2000;96:384–392. [PubMed]
28. Laad AD, Thomas ML, Fakih AR, Chiplunkar SV. Human gamma delta T cells recognize heat shock protein-60 on oral tumor cells. Int J Cancer. 1999;80:709–714. [PubMed]
29. Lang F, Peyrat MA, Constant P, Davodeau F, David-Ameline J, Poquet Y, Vie H, Fournie JJ, Bonneville M. Early activation of human V gamma 9V delta 2 T cell broad cytotoxicity and TNF production by nonpeptidic mycobacterial ligands. J Immunol. 1995;154:5986–5994. [PubMed]
30. Malkovska V, Cigel F, Storer BE. Human T cells in hu-PBL-SCID mice proliferate in response to Daudi lymphoma and confer anti-tumour immunity. Clin Exp Immunol. 1994;96:158–165. [PubMed]
31. Miyagawa F, Tanaka Y, Yamashita S, Minato N. Essential requirement of antigen presentation by monocyte lineage cells for the activation of primary human gamma delta T cells by aminobisphosphonate antigen. J Immunol. 2001;166:5508–5514. [PubMed]
32. Neri S, Mariani E, Meneghetti A, Cattini L, Facchini A. Calcein-acetyoxymethyl cytotoxicity assay: standardization of a method allowing additional analyses on recovered effector cells and supernatants. Clin Diagn Lab Immunol. 2001;8:1131–1135. [PMC free article] [PubMed]
33. Nikolich-Zugich J, Slifka MK, Messaoudi I. The many important facets of T-cell repertoire diversity. Nat Rev Immunol. 2004;4:123–132. [PubMed]
34. Panchamoorthy G, McLean J, Modlin RL, MOrita CT, Ishikawa S, Brenner MB, Band H. A predominance of the T cell receptor Vgamma2/Vdelta2 subset in human mycobacterial-responsive T cells suggests germline gene encoded recognition. J Immunol. 1991;147:3360–3369. [PubMed]
35. Poccia F, Malkovsky M, Pollak A, Colizzi V, Sireci G, Salerno A, Dieli F. In vivo gammadelta T cell priming to mycobacterial antigens by primary Mycobacterium tuberculosis infection and exposure to nonpeptidic ligands. Molecular Medicine. 1999;5:471–476. [PMC free article] [PubMed]
36. Rakasz EG, McDougall A, Zayas M, Helgeland JL, Ruckwardt T, Hatfield G, Dykhuizen M, Mitchen JL, Evans PS, Pauza CD. gd T cell receptor repertoire in blood and colonic mucosa of Rhesus macaques. J Med Primatol. 2000;29:387–396. [PubMed]
37. Scotet E, Martinez LO, Grant E, Barbaras R, Jeno P, Guiraud M, Monsarrat B, Saulquin X, Maillet S, Esteve JP, Lopez F, Perret B, Collet X, Bonneville M, Champagne E. Tumor recognition following Vgamma9Vdelta2 T cell receptor interactions with a surface F1-ATPase-related structure and apolipoprotein A-I. Immunity. 2005;22:71–80. [PubMed]
38. Sicard H, Al Saati T, Delsol G, Fournie JJ. Synthetic phosphoantigens enhance human Vgamma9Vdelta2 T lymphocytes killing of non-Hodgkin's B lymphoma. Mol Med. 2001;7:711–722. [PMC free article] [PubMed]
39. Thomas ML, Samant UC, Deshpande RK, Chiplunkar SV. gammadelta T cells lyse autologous and allogenic oesophageal tumours: involvement of heat-shock proteins in the tumour cell lysis. Cancer Immunol Immunother. 2000;48:653–659. [PubMed]
40. Tikhonov I, Deetz CO, Paca R, Berg S, Lukyanenko V, Lim JK, Pauza CD. Human V{gamma}2V{delta}2 T cells contain cytoplasmic RANTES. Int Immunol. 2006 [PMC free article] [PubMed]
41. Wilhelm M, Kunzmann V, Eckstein S, Reimer P, Weissinger F, Ruediger T, Tony HP. Gammadelta T cells for immune therapy of patients with lymphoid malignancies. Blood. 2003;102:200–206. [PubMed]