HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sicard, H. Articles by Romagné, F. Search for Related Content PubMed PubMed Citation Articles by Sicard, H. Articles by Romagné, F.

In Vivo Immunomanipulation of V9V2 T Cells with a Synthetic Phosphoantigen in a Preclinical Nonhuman Primate Model

Hélène Sicard^1,*, Sophie Ingoure^*, Béatrice Luciani^*, Claire Serraz^*, Jean-Jacques Fournié, Marc Bonneville, Jérôme Tiollier^* and François Romagné^*

^* Innate Pharma, Marseilles, France; Institut National de la Santé et de la Recherche Médicale Unité 463, Centre Hospitalier Universitaire Nantes, Quai Moncousu, Nantes, France; and Institut National de la Santé et de la Recherche Médicale Unité 563, CHU Purpan, Toulouse, France

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
V9V2⁺ cells represent the major population of T cells in primate blood and react in an MHC-unrestricted fashion to a set of low m.w. nonpeptide phosphoantigens. Two types of structurally related agonists have been discovered so far: the natural phosphoantigens (hydroxydimethyl allyl-pyrophosphate or isopentenyl-pyrophosphate (IPP)) acting directly on V9V2⁺ TCR and aminobisphosphonates, which block the mevalonate pathway in target cells, leading to accumulation of natural phosphoantigens that in turn activate V9V2⁺ cells. We demonstrate in the cynomolgus monkey that V9V2 can be manipulated in vivo with bromohydrin pyrophosphate (BrHPP)/Phosphostim, a potent synthetic agonist for which the mechanism of action is similar to natural phosphoantigens. Although of very short half-life, injection of BrHPP leads to strong activation of V9V2, inducing production of a high level of Th1 cytokines. Combination of BrHPP with low-dose rhIL-2 induces specific amplification of effector-memory peripheral V9V2 in blood in a dose-dependant manner. This transient response returns to baseline within 10–15 days. Successive infusions of BrHPP and rhIL-2 induce less vigorous expansions, suggesting a progressive exhaustion of the response. As no toxicity is detected with or without IL-2, this scheme represents a promising immunotherapeutic strategy for induction of systemic Th1 cytokines and massive expansion of T cell subset with antitumor and anti-infectious properties.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
The vast majority of peripheral blood T cells in humans express TCR composed of a restricted set of variable regions called V9 and V2, and are generally devoid of CD4 and CD8 coreceptors. V9V2 T cells, which classically represent 0.5–5.0% of whole peripheral blood T cells, participate with dendritic cells (DC),² NK cells, and NK-T cells in the innate immunity response to pathogens or cancer.

V9V2 T cells are known to be preferentially expanded in vivo in the context of various infectious diseases such as tuberculosis (1), legionellosis (2), tularemia (3), brucellosis (4), malaria (5), mononucleosis (6), or infections by Listeria (7) or Salmonella (8). The stimulating bacterial Ags have been shown to be small nonpeptidic compounds classically referred to as "phosphoantigens" (3, 9, 10, 11, 12), owing to the presence of phosphate groups in most instances. Natural phosphoantigens are produced through an isoprenoid biosynthetic pathway called the "Rohmer" or "nonmevalonate" pathway, which is specific to pro- and eukaryotic microorganisms (13, 14, 15, 16). In vitro, V9/V2⁺ lymphocytes react in the nanomolar range to these metabolites (hydroxydimethyl allyl-pyrophosphate) (17, 18). They also react, although in the micromolar range, to metabolites of the mevalonate pathway of normal eukaryotic cells (isopentenyl-pyrophosphate, dimethyl allyl-pyrophosphate) (14). Discrimination between these variously bioactive compounds probably accounts for the specific reactivity of V9V2 T cells toward bacteria and parasites (19).

Beside their anti-infectious activity, it was shown in short-term cytotoxicity assays that V9V2 T cells are able to kill a wide variety of tumor cell lines from very diverse origins: lymphoma and leukemia from B cell, T cell, or myeloid lineages (20, 21, 22, 23, 24), breast carcinoma (25), glioblastoma (26), renal cell carcinoma (27, 28, 29, 30), nasopharyngeal carcinoma (22), lung adenocarcinoma (31). In many cases, this direct in vitro anti-tumoral activity was demonstrated in autologous contexts and with fresh tumor cells or short-term cancer cell lines (26, 27, 28, 29, 30, 32).

Finally, V9V2 T cells are also potent producers of cytokines such as TNF- and IFN-, which can either directly inhibit tumor cell growth or potentiate the anti-tumor activity of other lymphoid effectors. It has also been proposed that cytokine production by T cells influences adaptive immunity by inducing DC maturation in the early phase of the immune response (33, 34).

Because activated V9V2 T cells are potent cytolytic effectors and Th1 cytokine producers, they may play important roles in defense against tumors and represent attractive targets for manipulation in oncology. Such approaches can be designed through the use of low m.w. agonists. Synthetic compounds, such as bromohydrin pyrophosphate (BrHPP), the active pharmaceutical ingredient in Phosphostim used in the present study, with the same mechanism of action and structurally related to the natural phosphoantigens described above, have recently been designed to this end (10, 35). When tested in vitro against whole human blood lymphocytes, BrHPP (10 nM with exogenous IL-2) enabled the selective outgrowth of the V9V2⁺ T cells, which made up to 90% of viable lymphocytes after a 3-wk culture (35). As for natural phosphoantigen, EC₅₀ is obtained on isolated T cell clones or purified V9/V2 T cell populations. V9V2 T cells have already shown in vitro their potential to improve cell response toward various tumor cell types (23, 30). We now intend to use them as immunomodulators of T cells in anti-cancer therapy.

Nonhuman primates (NHP) were demonstrated to represent good animal models for the study of V9V2 cells in vivo, as they carry T cell subsets highly homologous to human phosphoantigen-reactive cells (36, 37). In this study, we present the first evidence that V9V2 cells can be manipulated in vivo with a synthetic phosphoantigen, BrHPP. BrHPP is one of the most potent phosphoantigens available to date (35). This model of primate cell activation by the first synthetic phosphoantigen available at the GMP/clinical grade gives new insights into the physiology of this cell subset upon in vivo stimulation and allows preclinical pharmacodynamics evaluation of this candidate-medicine.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Animals

Three groups of animals were studied.

Group 1 consisted of five purpose-bred healthy male cynomolgus monkeys (Macaca fascicularis). At the beginning of the study, body weights ranged from 3.7 to 4.6 kg.

Group 2 consisted of 10 purpose-bred healthy cynomolgus monkeys (5 males and 5 females). At the beginning of the study, body weights ranged from 1.8 to 3.5 kg and ages from 24 to 36 mo.

Group 3 consisted of 24 purpose-bred healthy cynomolgus monkeys (20 males and 4 females). At the beginning of the study, body weights ranged from 1.8 to 3.5 kg and male ages from 23 to 30 mo. Female ages were unknown (capture animals).

All animals were supplied by C.R.P. Le Vallon, Ferney S.E. Husbandry conditions conformed to the European requirements, comprising monitored temperature, humidity, air change, and lighting cycle. Group 1 and 3 animals were housed in Biomatech Namsa and group 2 animals were housed in MDS Pharma Services. All experiments were subjected to and approved by the local ethical committee before processing.

Phosphoantigens

The synthesis and characterization of trisodium (R,S)-3-(bromomethyl)-3-butanol-1-yl-diphosphate BrHPP has been described previously (35). The batch used for the experiments described in this study was manufactured and characterized under GMP conditions by PCAS-SELOC. Sterilization and clinical unit preparation was conducted under good laboratory practices by Axcell Biotechnologies. Titration of BrHPP in sterile aqueous solution was achieved by high performance anion-exchange chromatography with conductimetric detection (Dionex DX600 system; Ref.35).

Drug administration and blood sampling

Group 1 and 3 animals were anesthetized with i.m. injection of 6 mg/kg of ZoletilND 100 (Tiletamine-Zolazepam; Virbac) before any injection or blood sampling. Injections and blood sampling in group 2 animals were performed on manually restrained nonanesthetized animals.

BrHPP was diluted in saline to the appropriate final concentration (depending on the dose to inject and on the last recorded body weight) so as to inject always 50 ml. BrHPP was administered as a 30-min infusion (or 60 min, for pharmacokinetics experiments) using a microflex infusion set introduced into cephalic or external saphenous vein.

Recombinant human (rh)IL-2 (Proleukin; Chiron) was resuspended in 1 ml of sterile water and diluted in 10 ml of saline for a final concentration of 1.8 million IU/ml. IL-2 (0.3 million U) was administered twice daily by s.c. injection.

Twice weekly, blood samples (1–4 ml) were withdrawn from femoral vessels into EDTA-containing tubes. Tubes were shipped overnight at room temperature (RT) before flow cytometry analyses.

In vitro culture of NHP PBMC

Whole heparin-treated blood from the following NHP species were obtained from the respective providers: rhesus macaque (Macaca mulatta), African green monkey (Chlorocebus aethiops), marmoset (Callithrix jacchus), and saimiri (Saimiri sciureus) provided by the "Centre de Primatologie" (Université Louis Pasteur); baboon (Papio hamadryas), provided by the "Station de Primatologie du Centre National de la Recherche Scientifique"; and cynomolgus monkey (M. fascicularis), provided by Biomatech Namsa. PBMC were separated on Ficoll cushion (Paque-Plus; Amersham) and cultured in RPMI 1640 (Invitrogen Life Technologies) with 10% FCS (Fetal Clone 2; HyClone) in the presence of 3 µM BrHPP and 100 IU/ml rhIL-2 (Proleukin; Chiron) starting at 1.5 x 10⁶ cells/ml. At day 8, cell ratio to viable cells in the culture was determined by flow cytometry, with anti-CD3-PE, anti-V9-FITC (see above), and anti-CD69-PC5 (clone FN50; Immunotech-Beckman Coulter).

Flow cytometry

Primate peripheral T lymphocytes were analyzed twice weekly by flow cytometry on whole blood, after dual staining with anti-CD3-PE Ab and anti-V9-FITC, anti-V9-PC5, or anti-V2-FITC Abs (CD3-PE: SP34 clone, BD Pharmingen; anti-V2-FITC: 15D clone, Endogen). The anti-V9 mAb 7B6 rose against human V9⁺ T cells and cross-reacts with cynomolgus cells (M. A. Peyrat and M. Bonneville, unpublished observations). It was purified by affinity chromatography on protein A and coupled to FITC or to PE-Cy5.

Cynomolgus cell staining conditions were as follows: 50 µl of monkey blood was incubated 15 min at RT with 5 µl of anti-CD3-PE and 6 µl of anti-V2-FITC or 10 µl of anti-Vg9-FITC mAb. Abs were washed with 3 ml of 1x PBS and centrifuged for 4 min at 1300 rpm at RT, and the supernatant was discarded. Red cells were lysed with the OptiLyse C reagent (Immunotech-Beckman Coulter) according to the manufacturer’s instructions. At the final step, stained white blood cells were recovered by centrifugation and resuspended in 300 µl PBS plus 0.2% paraformaldehyde. Immediately before analysis, 50 µl of calibrated Flow Count Fluorospheres (Immunotech-Beckman Coulter) were added to the cells for determination of absolute numbers of cellular populations of interest. Thus, amplification of V9V2 T cells at day x is defined as the following ratio: (calculated number of cells per cubic millimeter of blood at day x of treatment)/(calculated number of cells per cubic millimeter of blood at day 0). For follow up of subpopulations in group 3 animals, whole blood was also stained with anti-V9-PC5, anti-CD45RA-FITC, and anti-CD27-PE (5H9 and M-T271 clones, respectively; from BD Pharmingen, 10 µl of each) and further processed as described above.

We compared 7B6 mAb to the commercial TCR2732 mAb (Endogen). On whole blood primate cells, 7B6 gave systematically a better signal-to-noise ratio compared with TCR2732, but the correlation between the numbers of cells stained by the two Abs is excellent (data not shown). Together with results showing that most monkey V9 cells are V2 positive (37), both Abs can be used to track V9V2 population.

Flow cytometry was performed on an Epics XL-MCL apparatus (Beckman Coulter) with the Expo32 software after gating on live cells.

BrHPP blood concentration determination

Immediately after blood sampling, pharmacokinetics plasma samples were precipitated with Acetonitrile (Carlo Erba) 75% in water and stored at –20°C before direct analysis by HPLC-tandem mass spectrometry (MS₂). A HP1100 Series system (Agilent Biotechnologies) coupled with API2000 mass spectrometer (PE Sciex; Applied Biosystems) was used. The HPLC system was equipped with a HPLC Daisogel C18B column (4.6 x 250 mm internal dimensions, 5 µm; Daiso; A.I.T. France). Ion pair separation was conducted by using isocratic elution of 5 mM ammonium acetate at a flow rate of 1 ml/min. Under these conditions, BrHPP standard eluted from the column at a retention time of 4 min. MS/MS analysis was performed with an electrospray ion source in negative ionization mode. Quantification was achieved in the precursor ion mode by monitoring the [M-H]^– precursor ion at m/z 341 and the product ion at m/z 261.

Cytokine detection

Serum cytokines (TNF- and IFN-) were detected and quantified with the BioSource Cytoscreen ELISA monkey TNF- and Cytoscreen ELISA monkey IFN-, respectively (purchased from CliniSciences), according to the manufacturer’s instructions.

For intracellular cytokine detection, whole monkey blood was incubated 5 h at 37°C in a CO₂ incubator with 30 µM BrHPP or 1 µg/ml ionomycine plus 40 ng/ml PMA and 10 µg/ml brefeldin A (Sigma-Aldrich). Cells were stained for extracellular markers (V9-PC5 and CD3-FITC) before fixation and permeabilization (FACS permeabilization solution 2; BD Pharmingen). Anti-TNF--PE Ab (20 µl at 6.25 µg/ml, Mab11 clone; BD Pharmingen) was then incubated 45 min before washing and detection of TNF-⁺ cells by flow cytometry.

Hematology and serum clinical chemistry

Classical blood parameters including RBC, platelets total and differential white blood cells counts, hemoglobin, mean corpuscular hemoglobin, and mean corpuscular hemoglobin concentration were monitored just before and after each administration and then twice weekly.

On each animal in group 2, 48 or 72 h after each injection, the following blood biochemistry parameters were measured: sodium, potassium, chloride, calcium, inorganic phosphorus, glucose, urea, total cholesterol, total bilirubin, total protein, albumin, globulin, creatinin, alkaline phosphatase, aspartate aminotransferase, and alkaline aminotransferase.

Vital parameters follow up

The animals were observed daily and during and after each injection for any change in vital and clinical parameters (general behavior, skin, hair, respiratory system, CNS). Animals were weighed every 3 days (groups 1 and 3) or weekly (group 2). Body temperature (on vigil animals of group 2) was measured before and at the end of each BrHPP/rhIL-2 (or rhIL-2 alone) infusion and once daily during the 5 days after administration. Heart rate and blood pressure were recorded for all animals in group 2, before and at the end of each administration. All animals were observed at least twice daily for signs of morbidity/mortality. No mortality that could be related to the treatment with the test compound ever occurred during the course of these experiments.

	Results

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
PBMC from six different NHP species proliferate in vitro in response to BrHPP and IL-2

A relevant animal model was searched for further studies of cell activation in vivo by BrHPP. We first tested the in vitro proliferation of PBMC from six different NHP species in response to this compound. After 8 days of culture in the presence of 3 µM BrHPP and rhIL-2, PBMC from all NHP species tested had proliferated to reach 60% and more cells (Fig. 1). Only marmoset cell proliferative response appeared somewhat weaker than for other species, as they only reached 11%. We evaluated the activation status of NHP cells in vitro through staining with an anti-human CD69 Ab and found them highly CD69 positive at day 8 of culture (data not shown). Only marmoset and saimiri cells appeared negative for this marker, presumably for the lack of Ab cross-reactivity to these genetically more distant species (data not shown). These data indicated that NHP in general represent adequate models for the study of cell response to their specific phosphoantigens.

View larger version (45K): [in this window] [in a new window]   FIGURE 1. In vitro amplification of V9V2 T cells from six primate species PBMC in culture with 3 µM BrHPP and IL-2. PBMC from six primate species, differently genetically related to humans, were cultured for 8 days in the presence of BrHPP and renewed IL-2. V9+ T cell population was monitored by flow cytometry using specific anti-human Ab, cross-reacting with the primate V9 receptor chain. Primate T cells specifically expand in every culture to become the major T cell population in most cases.

For the dosage of BrHPP in primate blood after i.v. administration, we developed a specific HPLC-MS₂ method. We performed preliminary pharmacokinetics experiments in cynomolgus monkeys to evaluate BrHPP concentration and plasma half-life in vivo, and thus animal exposure to the compound. Fig. 2 shows typical blood pharmacokinetics curves obtained after i.v. administration of 6 and 48 mg/kg of BrHPP to cynomolgus monkeys as 1-h infusion. Blood samples were taken during and after infusion. BrHPP is detected in vivo at concentrations below 1 µM after administration of 6 mg/kg (corresponding to an administered amount around 60 µmol), and that exposure to BrHPP lasts only the time of infusion. Structurally unchanged BrHPP is always undetectable at time points 30 min after the end of infusion, even after administration of higher doses (up to 545 mg/kg, data not shown). Moreover, steady-state concentration of BrHPP is reached very rapidly after the start of infusion (3–4.5 µM after infusion of 48 mg/kg, corresponding to the administration of 490 µmol BrHPP). These results confirmed that BrHPP half-life in vivo is actually very short, probably not exceeding a few minutes.

View larger version (18K): [in this window] [in a new window]   FIGURE 2. BrHPP plasma pharmacokinetics in cynomolgus monkeys. Two cynomolgus monkeys were administered 6 or 48 mg/kg body weight of BrHPP through 60-min i.v. infusion. Blood samples were collected before, during, and after BrHPP infusion. After adequate sample processing, BrHPP was dosed with a validated HPLC-MS2 method for the determination of its pharmacokinetic parameters (half-life, maximal concentration) and for preliminary evaluation of animal exposure to unchanged drug.

Requirement for IL-2 in cell proliferative response to BrHPP

Although cell proliferation in response to BrHPP is dependent on the presence of IL-2 in vitro, we first evaluated in vivo response of cynomolgus cells to BrHPP alone. In a first series of experiments, four monkeys were treated by five daily injections of 0.2 mg/kg BrHPP, and four animals were treated with saline. Injection of this dose of BrHPP did not result in any toxicity in the treated NHP as assessed by vital signs or body temperature. The percentages of V9V2 remained comparable to the background values (data not shown). As this first tested BrHPP dose might have been too low to induce a biological response, we decided to inject to nine other animals (from group 3), comprising five males and four females, with 48 mg of BrHPP per kilogram of body weight. Flow cytometry on whole blood was performed every 2–4 days, until 28 days after injection, without any evidence for modification in peripheral cell proportion in PBMC (Fig. 3A).

View larger version (17K): [in this window] [in a new window]   FIGURE 3. A, V9V2 T cell in vivo amplification upon Phosphostim treatment requires IL-2. Nine cynomolgus monkeys (five males and four females) were administered 48 mg/kg BrHPP as 30-min i.v. infusion. Blood lymphocyte populations, and among them V9+ cells, were monitored by flow cytometry using specific Abs, every 2–4 days, until day 28. No changes were noted in the target cell ratio among CD3+. B, BrHPP plus IL-2 cotreatment induces transient and reproducible increase in blood cells. Two cynomolgus monkeys received 12 mg/kg BrHPP i.v. as 30-min infusions and concomitant daily treatment with 0.3 million IU IL-2 s.c. twice a day for 5 days. Blood cells were monitored by flow cytometry every 2–4 days. As a control, another animal received the same regimen of IL-2 alone, and after a 15-day flow cytometry survey without apparent effect, it was administered BrHPP plus IL-2 same as the first two animals. Kinetics and level of blood cell increase were the same in the three treated animals.

BrHPP-induced amplification of cells to in the cynomolgus monkey is specific and depends on low-dose rhIL-2

As BrHPP alone, even at high dose, failed at inducing response in the cynomolgus monkey, we tested whether s.c. simultaneous administration of rhIL-2 would induce a significant peripheral amplification of cells.

Two animals (from group 1) received 12 mg/kg of BrHPP i.v. plus 0.3 million IU IL-2 s.c. twice daily for 5 days. As a control for IL-2 effect, a third animal was first treated with saline and 5-day IL-2 s.c. Fourteen days later, it received a single shot of 12 mg/kg BrHPP and a second course of 5-day IL-2 to verify its ability to respond to the test compound. Peripheral cells were monitored by flow cytometry twice a week. Hematological follow up of the animals revealed an increase of lymphocyte counts by 2- to 4-fold in BrHPP-treated and control animals, presumably due to rhIL-2 administration. Specific expansion both in percentage among CD3-positive cells and absolute numbers of V9V2 cells was observed in all BrHPP-treated animals (Fig. 3B). This increase was transient, as the expansion started on day 3, peaked at day 7, and V9V2 cell percentage and counts returned around baseline values by day 10. The control animal treated with IL-2 alone showed a small increase in absolute numbers of V9V2 cells but no increase in their percentage, suggesting that peripheral V9V2 cells, like other blood T cells, were slightly expanded upon IL-2 administration. Subsequent administration of BrHPP plus IL-2 in this control animal resulted in V9V2 cell expansion similar to that of the two first treated animals (Fig. 3B), thus validating this animal as a control in the experiment.

Dose-range effect of BrHPP in cynomolgus monkeys

To evaluate the in vivo dose-response to BrHPP, 10 new animals (group 2, subgroups of 2 animals, 1 male and 1 female) were administered increasing doses of BrHPP (0, 0.12, 2.4, 12, 48 mg/kg) and cotreated with 0.6 million IU IL-2 s.c. once daily for 7 days.

Again, animals treated with IL-2 alone experienced a slight increase of V9V2 cells, of the same order of magnitude as compared with whole lymphocyte population. The percentage and absolute numbers of V9V2 cells in animals of the 0.2 mg/kg BrHPP dose-group were undistinguishable from those of the control animals. A dose-range effect was observed both in percent and absolute counts of V9V2 cells from 2.4 to 48 mg/kg BrHPP (Fig. 4, upper panel) at day 7. Again percent and absolute counts rapidly came back to pretreatment levels. As a plateau was not reached during this treatment, the same animals were treated a second time 22 days after the first injection, with higher doses of BrHPP (12, 48, 72, and 97 mg/kg). To minimize potential effect of the first treatment on the second, the highest BrHPP doses (72 and 97 mg/kg) were given to animals injected with the lowest doses in the first treatment (i.e., animals treated with 0.12 and 2.4 mg/kg; Fig. 4A, lower panel). The time line of proliferation after this second injection was about the same as described before. Addition of an earlier time-point postinjection showed that the maximum V9V2 cell numbers were reached around day 5, whereas at day 7, their frequency already began to decline in all animals (Fig. 4A, lower panel). At the highest BrHPP concentrations, the levels of circulating V9V2 cells were around 80% of circulating CD3-positive cells (Fig. 4B), with absolute numbers reaching a mean of 10,000 V9V2 cells per cubic millimeter. The numbers and percentages of V9V2 cells declined after the peak between day 5 and 7, although more slowly than for the two highest doses. Pooled data from injection 1 and injection 2 gave a clear dose-range effect both in terms of percentage and absolute numbers of V9V2 cells at the peak of response (Fig. 4C). This dose-range effect was observed with a threshold for response above 0.12 mg/kg.

View larger version (28K): [in this window] [in a new window]   FIGURE 4. A, Dose effect of Phosphostim treatment on cell increase in vivo. Ten animals were administered increasing doses of BrHPP, with one female and one male per dose group, as 30-min i.v. infusions, with concomitant IL-2 treatment. BrHPP doses ranged from 0.12 to 48 mg/kg at the first administration, and then from 12 to 97 at the second administration. Two animals also received twice IL-2 only as negative controls. Blood samplings were performed every 3–4 days for flow cytometry analyses of the percentage of cells among CD3+ cells. B, Typical flow cytometry analysis of cynomolgus blood before and 7 days after administration of high dose (97 mg/kg) of BrHPP. Whole blood is dual stained with anti-hCD3 and anti-hV9 Abs to show the remarkable expansion of V9+ cells. C, Dose-dependent amplification of circulating absolute counts in Phosphostim-treated macaques. Flow cytometry follow up of lymphocyte populations in these BrHPP-treated animals included numeration of absolute cell counts. Fold increase in cells is defined as the ratio between the highest number of circulating cells per cubic millimeter of blood after treatment (day 7) on the number of circulating cells before treatment. The amplification of primate cells is clearly function of the administered dose of BrHPP, both in terms of percentage of V9+ cells among CD3+ and of cell absolute counts. Strikingly, the highest administered dose provides a >200-fold increase in blood cells.

Prolongation of IL-2 administration does not sustain peripheral V9V2 cell expansion induced by BrHPP.

The increase of peripheral V9V2 observed in the above experiment was repetitively transient. Peripheral cell percentages started to increase at day 3, peaked at day 5, but already began to decline at day 7 to preinjection levels by day 10.

We questioned whether increasing duration of daily IL-2 administration would improve the time-length of peripheral amplification. Fig. 5 compiles all data obtained with different IL-2 regimen after injection of 12 mg/kg of BrHPP, i.e., the lowest dose yielding consistent and easily detectable expansion of V9V2 cells. Even 12 days of IL-2 treatment did not influence the time course of V9V2 cell increase and decline. So long, prolongation of IL-2 treatment did not seem to allow maintenance of elevated numbers of peripheral V9V2 cells in vivo. We also tested shorter IL-2 treatment schemes, for example, by administrating IL-2 only on the first day of BrHPP injection, but it resulted in less efficient amplification of V9V2 cells in vivo (data not shown).

View larger version (22K): [in this window] [in a new window]   FIGURE 5. Pharmacological impact of increasing the duration of IL-2 cotreatment. As amplification in the blood of Phosphostim/IL-2-treated animals is only transient, we lengthened the time-length of daily IL-2 treatment. All the animals presented in this study received a fixed dose of 12 mg/kg of BrHPP, with IL-2 treatment ranging from 5 to 12 days. Blood cells were monitored by flow cytometry, according to the usual kinetics, for 17 days. The time-length of IL-2 daily cotreatment did not seem to improve the duration of cell increase, systematically going back to basis around day 11.

Amplitude of cynomolgus cell response to BrHPP and low-dose IL-2 decreases with repeated treatments

We observed in the previous dose-range effect study that the amplitude in the response to a second administration of BrHPP was lower than that measured at the first administration.

We designed a long-term study with 12 animals (group 3) to assess whether cell proliferative response decreases with consecutive treatments, and whether this decrease is dependent on the dose and/or on the time-length between the injections. Three groups of four males received (with concurrent low-dose IL-2 s.c. treatment) four successive administrations of 48 mg/kg of BrHPP 8 wk apart, or 12 mg/kg BrHPP 8 wk apart, or 12 mg/kg BrHPP 4 wk apart. During each cycle of treatment, blood cells were monitored at days 0, 4, 5, 7, 9, 11, and 14. Fig. 6 shows, over a 200-day period, the mean amplification waves of cynomolgus peripheral cells in the three groups of animals. Amplitude of in vivo response to BrHPP decreased in time, whatever dose or time-length between injections was tested. This exhaustion of response was dramatically clear cut at the third and fourth injections (Fig. 6). In every animal, the absolute count in circulating cells went back to basis after each treatment, with no apparent decrease in the circulating pool of cells.

View larger version (21K): [in this window] [in a new window]   FIGURE 6. In vivo pharmacological effects of successive coadministrations of Phosphostim and IL-2 and long-term exhaustion of the response. Three groups of four cynomolgus monkeys were administered four times Phosphostim plus IL-2, with various doses (24 or 48 mg/kg, i.e., low or high) and at various time intervals (4 or 8 wk, i.e., short or long). Blood samples were taken every 1–4 days during the 2 wk after each treatment, for flow cytometry follow-up of successive amplifications of cells. The exhaustion of the response with successive injections was compared between high and low doses, and between short and long time-lengths between treatments. As cell responses in all three groups were dramatically impaired after the fourth treatment, some animals (group 12 mg/kg, 4 wk apart) were left untreated for 12 more weeks. They received then a fifth Phosphostim plus IL-2 administration, with two different doses (12 and 60 mg/kg, i.e., low and very high) and were subjected to the usual flow cytometry follow up.

Systemic TNF- and IFN- production by V9V2 T cells upon in vivo challenge with BrHPP

V9V2 cells are known producers of TNF- and IFN- upon in vitro activation. To evaluate whether these cells could be a source of these cytokines in vivo, sera of several animals were collected shortly after BrHPP injection. Samples of sera were collected from group 2 animals during the first dose-range assay (0–48 mg/kg with two animals per dose) just before injection and then 1 and 4 h postinjection, and assayed by ELISA specific for primate TNF- and IFN-. IFN- remained at background serum levels in all treated animals (data not shown). TNF- was detected in the sera of animals 1 h after injection of 48 mg/kg of BrHPP (Fig. 7A, left panel). The serum level of TNF- rapidly decreased as it was no longer detectable 4 h after BrHPP injection, and remained undetectable during the V9V2 cell expansion phase in the blood. This suggests that TNF- is rapidly released by V9V2 cells presumably from an intracellular preformed pool of protein. We also treated two other animals (from group 3) with 48 mg/kg of BrHPP without IL-2 cotreatment to address whether BrHPP alone could induce early systemic cytokine production: both animals showed significant increase in blood TNF- levels, starting 90 min after treatment and maintained up to 3 h after (Fig. 7A, right panel). Thus, although BrHPP alone does not sustain proliferation in vivo, it induces cytokine production by its target cells.

View larger version (32K): [in this window] [in a new window]   FIGURE 7. Production of systemic Th1 cytokines in vivo after Phosphostim administration. A, Systemic Th1 cytokine release early after administration of BrHPP plus IL-2 or BrHPP alone. TNF- was detected by ELISA in the serum of the monkeys first treated with IL-2 and increasing doses of BrHPP (0–48 mg/kg, with two animals per dose), 1 and 4 h after Phosphostim infusion (left panel). The same cytokine was also dosed in the serum of two other animals, 1–3 h after they received 48 mg/kg BrHPP alone (right panel). In each case, the administration of 48 mg/kg BrHPP, with or without concomitant IL-2 treatment, results in early systemic release of TNF-. B, Induction of Th1 cytokines release in vivo from peripheral Phosphostim-amplified primate cells. At their fourth Phosphostim/IL-2 cotreatment, females 2032 and 2034 received 48 mg/kg of BrHPP, which induced 3- and 7.9-fold increases in cell absolute counts, respectively. At the time of cell peripheral peak (day 7 after this fourth injection), both animals were reinjected with 48 mg/kg of BrHPP (without IL-2) and serum samples were collected predose and 1 and 2 h after treatment. IFN- and TNF- were detected in these samples by ELISA. High amounts of each cytokine were found in both animals, correlated with their individual level of circulating cells. C, Ex vivo induction by Phosphostim of intracellular TNF- in primate cells. Fresh cynomolgus blood was incubated with various stimuli to induce intracellular production of cytokines. A mixture of PMA and ionomycin was used as a positive control of unspecific activation, and primate blood was also incubated with BrHPP alone. Intracellular TNF- was detected after permeabilization with a specific Ab, and cells were surface-stained with anti-V9 Ab to identify the population of interest. Flow cytometry percentages of V9+TNF-+ double-positive cells are shown on the upper sides of the quadrants, whereas V9–TNF-+ cell percentages, representing unspecific stimulation, are shown on the lower quadrants.

To demonstrate that the systemic cytokines dosed in the previous experiments are produced specifically by cells, intracellular TNF- production was evaluated at the single-cell level on whole blood stimulated ex vivo by BrHPP. Cynomolgus fresh blood was incubated with 30 µM BrHPP or 40 ng/ml PMA plus 1 µg/ml ionomycin as a positive control, in the presence of brefeldin A. PBMC were then surface-stained with anti-V9 and anti-CD3 Abs for detection of the population of interest, and intracellular TNF- was revealed with anti-TNF- Ab. High induction of intracellular TNF- in the positive control confirms the functionality of cynomolgus cells (V9⁺ as well as V9^–) in this assay (Fig. 7C), whereas BrHPP only induces TNF- production in V9⁺ cells, confirming the specificity of this Ag toward cells ex vivo.

Evolution of subpopulations upon BrHPP treatment in vivo

Recent studies have described different subpopulations, exhibiting distinct functional properties and defined, among other criteria, by their respective expression of the surface markers CD27 and CD45RA (38, 39, 40). On new animals treated with 12 mg/kg of BrHPP plus low-dose IL-2 s.c., we added a CD27/CD45RA double staining to flow cytometry analyses of peripheral cell amplification kinetics (days 0, 4, 5, 7, 11, and 14). Fig. 8 shows a representative result of this follow up. Before treatment, cells had, in majority, a central memory CD27⁺CD45RA^– phenotype ( T_CM,) with a low proportion of naive cells (double-positive CD27⁺CD45RA⁺), depending on the individual. At days 4–5 of treatment, this distribution progressively shifted to one-third effector memory cells (T_EM, CD27^–CD45RA^–) and two-thirds T_CM. As the cell rate went back to baseline (days 7–14), T_EM cells disappeared and the proportion of naive cells increased; final phenotype distribution was identical with what was observed before treatment (T_CM in majority and a few naive cells), with no apparent phenotypic signature of the previous activation (Fig. 8). In two other animals treated with phosphoantigen four times every 5–6 wk, we observed the same transient phenotypical changes after each treatment, but as total number of blood cells went back to basis, they exhibited no sign of increase in effector subset as compared with before treatment (data not shown). In none of the treated animals could we detect the cytotoxic effector memory-RA⁺ subset (T_EMRA, CD27^–CD45RA⁺) nor an increase in T_EM cells (data not shown). Together with this absence of change of into effector cells, we found that systemic IFN- production, although intense after the first administration, dramatically dropped after the second, the third and the fourth ones (data not shown).

View larger version (24K): [in this window] [in a new window]   FIGURE 8. In vivo changes in effector/memory phenotypes on Phosphostim-activated cells. The effector/memory surface markers CD27 and CD45RA were detected by flow cytometry on gated V9+ cells, after coadministration of Phosphostim plus IL-2 to cynomolgus monkeys. This figure is representative of the phenotypic changes obtained on of every treated animal, whatever the dose or the rank of administration. subpopulations are defined as described for all other T lymphocytes, namely: naive (CD45RA+CD27+), central memory (C45RA–CD27+), effector memory (CD45RA–CD27–) and RA+ effector memory (CD45RA+CD27–). The ratios of the respective subpopulations are shown in the corresponding sides of the "CD45RA/CD27" panels, at each time-point before and after treatment (days 0–14), accompanied by the representation of cell percentages on the same days ("SS lin/9" panels).

No alteration of clinical or blood biochemical parameters (see Materials and Methods) were seen in any of the animals treated either with BrHPP alone or in association with IL-2 along all the experiments described above. From a hematological point of view, in all animals treated with IL-2, a transient increase (two to five times) of lymphocytes was observed. In animals treated with the highest dose of BrHPP, slight lymphocytosis was observed, which was concomitant with the peak of V9V2 cells in the periphery. It should be noted that body temperature never increased in any animal treated, despite the presence of detectable levels of TNF- and IFN- inflammatory cytokines in the sera of animals treated with the highest doses.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
This study provides the first extensive description of the V9V2 T cell subset response in vivo to a synthetic phosphoantigen, BrHPP/Phosphostim. When tested in the presence of rhIL-2, this molecule was previously shown to trigger in vitro proliferation of human polyclonal V9V2 T cells with EC₅₀ of 10 nM (35), whereas the molecule alone is sufficient to induce cytokine release and activation markers. In this study, we demonstrate that NHP provides a relevant model for the study of agonists; after i.v. administration to M. fascicularis, BrHPP, although its turn-over in vivo is very rapid, induces reproducible and dose-related proliferation and systemic cytokine secretion by cells in the presence of low-dose rhIL-2.

As suspected from its low m.w., its highly cationic nature and its intrinsic sensitivity to blood phosphatases, BrHPP is very labile in vivo. To test that, we developed a specific HPLC-MS₂ method for the dosage of BrHPP in macaque blood, which has since been successfully transferred to the dosage of BrHPP in human blood for pharmacokinetics survey in clinical trials. Despite its short plasma half-life, BrHPP is able to induce reproducible activation of V9/V2 T cells after an i.v. 30-min infusion. Interestingly, the level of V9/V2 pharmacological response in vivo is identical when BrHPP is administered through bolus i.v. injection (S. Ingoure, unpublished data), which means that the time of contact between the phosphoantigens and their target cells is extremely short.

From the in vivo data presented in this study, it appears that although a single injection of 48 mg/kg of BrHPP alone to an adult M. fascicularis is insufficient to amplify its blood T cells, its coinjection with IL-2 leads to a dramatic expansion of this cell subset. The IL-2 dosage used for in vivo experiments, 0.3 million U s.c. twice daily per animal, was derived from a previously published adoptive immunotherapeutic protocol in humans (1 million IU/m²/day (41)) taking into account the relative body surface of humans and cynomolgus. This dose is considered as low in humans, addressing mainly cells harboring the high affinity receptor for IL-2. Thus, low IL-2 doses are sufficient to support the expansion of V9V2 cells in vivo, suggesting that T cells after BrHPP treatment express the high affinity IL-2 receptor. This is in line with the experiment showing that BrHPP alone induces CD25 on human cells in vitro. We could not detect CD25 on cynomolgus cells, neither in vitro nor in vivo, strongly suggesting that the reagent used is not adequate for the follow up of cynomolgus cells.

Increasing BrHPP dose while injecting a constant amount of IL-2 unambiguously pinpointed the linear relationship between phosphoantigen concentration and increase of absolute numbers of circulating T cells. All the animals that received above 96 mg/kg of BrHPP showed massive expansions (above 200–fold increase in cell numbers) of their T cell population. The injection of the molecule at these dosages (above 48 mg/kg), alone or in combination with IL-2, also led to detectable serum level of IFN- and TNF- within the first few hours after injection. These reactions were transient, however, with IFN- and TNF- being rapidly cleared, and V9V2 T cell ratio returned to the baseline after 2–3 wk in all animals.

Repeating BrHPP injections and rhIL-2 cotreatments was able to support new amplifications of the subset. However, the data demonstrated that recall T cell responses are of lower magnitude compared with initial exposure to the Ag. Accordingly, after up to four injections of BrHPP every 4 or 8 wk, amplification of T cell subset decreased dramatically. This apparent anergy or possible deletion of the target cell population is a long-lasting feature, as a 12-wk recovery period was not sufficient to restore the initial level of response. This result contrasts with a former study involving Macaccus injected with viable mycobacteria, suggesting that mycobacterial challenge induced stronger recall responses of systemic V9V2 T cells, which was associated with more effective clearance of mycobacteria (37). Although this work clearly illustrated the ability of T cells to contribute to the setting of adaptive immunity to mycobacterial infection, our system did not address a further challenge with pathogens. The discrepancy in recall effects obtained in these two studies could be most likely attributed to differences in the Ags administered, namely whole live mycobacteria vs a pure agonist.

It was recently described that, like other T cells (42, 43), V9/V2 T cell subtype is composed of several subpopulations, of distinct migration and functional properties (40, 44). Although the phenotypical description of these naive, memory, and/or effector subpopulations is not restrained to—and more complex than—CD27 and CD45RA expression (38), these two markers were used for the first time in vivo in the context of a specific nonpathogenic phosphoantigen-induced activation of V9/V2 T cells. As expected, in the periphery, V9/V2 cells shift from a mixed naive/central memory phenotype to central/effector memory as they peak. However, when their circulating number goes back to baseline, V9/V2 phenotype returns also to the initial apparently unmodified naive/central memory distribution. These transient phenotypic changes were observed in every individual tested, whichever dose was administered, and were the same after any successive injection, from the first treatment up to the fifth. This is in clear contrast with what was observed by Dieli et al. (39), who showed that, in cancer patients treated several times with zoledronic acid, the proportion of effector V9V2 T cells increases, showing a reduced proliferative capacity and an increase in cytokine production. In our model, although its proliferative capacity reduces with injections, the "visible" pool of peripheral cells remains phenotypically unchanged and shows no increased ability to produce Th1 cytokines. The major difference between their scheme of administration and ours is that we systematically use IL-2 and thus induce transient increases in cell pool. This may result in different ways of activation/differentiation of cells than when they do not amplify. Furthermore, we believe that analysis of blood cells only provides a limited view on phosphoantigen effects in vivo and we are currently trying to study cells in other organs as important migration mechanisms are certainly induced.

On a more clinical standpoint, as V9V2 cells are present in the vast majority of humans, targeting these cells by pharmacological means could lead to a simple and potent immunotherapy both as a source of Th1 cytokines and effector cells in cancer and infectious disease. V9/V2⁺ cells have been shown to respond to synthetic drugs named therapeutic aminobisphosphonates (pamidronate, zoledronate,...) used for the treatment of osteoporosis (45, 46, 47). Although the structures of aminobisphosphonates are related to those of natural phosphoantigens, the mechanism of activation of V9/V2⁺ cells by aminobisphosphonates has turned out to be different from that of natural phosphoantigens. The elucidation of aminobisphosphonates mechanism of action and the definition of their molecular target provided a clue for understanding their activity on cells. Aminobisphosphonates target intracellular enzymes of the mevalonate pathway, and particularly the farnesyl transferase. Metabolic blockade in downstream steps of IPP by aminobiphosphonate presumably leads to accumulation of some upstream mevalonate metabolites such as IPP and determine the phosphoantigen content of cancer cell lines such as Daudi (48). As bisphosphonates were available as clinical grade molecules, they have been used to manipulate V9V2 directly in humans. Pamidronate in combination with low doses of IL-2 results in in vivo increase of the V9V2 population correlated to improvement of clinical outcome (49). However, pamidronate, and more generally bisphosphonates, may not be ideal drugs to induce in vivo activation of V9V2 cells. First, their activity in vitro is in the micromolar range, compared with nanomolar range for the best known phosphoantigen agonists. Second, unlike phosphoantigens, aminobisphosphonates activate V9V2 T cells in an indirect fashion as discussed above, through the blockade of metabolic enzyme of mevalonate that may have unwanted side effects if only manipulation is sought. Finally, bisphosphonates rapidly become localized in the bone with very long terminal half-life that may be detrimental to stimulation purposes (50, 51). Although an therapeutic action of bisphosphonates has been suggested in humans, their effect on V9V2 cell frequency is highly variable from one patient to another, several treated patients showing no detectable expansion or activation of their V9V2 cells (49).

A key issue for the relevance of immunotherapeutic approaches using such protocol is the trafficking of these cells to infected or tumor sites. In our experiments, proliferation of T cells was only explored in the periphery. However, because it has been shown that V9V2, upon phosphoantigen challenge, can express CCR5 chemokine receptor and are capable of diapedesis (52, 53), activated V9V2 cells should be able to migrate to inflammatory sites. In the absence of such inflammatory site in the primates studied, the cells may simply die and return to basal level. As mentioned above, other experiments studying the trafficking of these cells in vivo are under way in our laboratory.

The lead phosphoantigenic compound BrHPP, active substance of the drug product Phosphostim, is currently tested in a phase I clinical trial in oncology patients. Preclinical evaluation of tolerance to Phosphostim, thoroughly studied in regulatory acute toxicity assays in primates, was excellent. Its scheme of administration to volunteer patients was established according to the data generated in the preclinical primate model presented in this study. We chose to administer Phosphostim three consecutive times at 4-wk intervals before the first clinical evaluation. The first cycle was performed without IL-2, for the evaluation of tolerance to Phosphostim alone. Pharmacological response was carefully monitored all along these first three cycles. Then, based upon tolerance, pharmacological and clinical data altogether, it could be decided to administer the patient again for further cycles. During all these phases, the follow up of Phosphostim modulatory properties, in terms of blood cell increase and systemic cytokine production, was imitated from those successfully performed in treated primates.

	Acknowledgments

 
We thank Monique Moyne at Biomatech Namsa, Marielle Christ at MDS Pharma Services, and Roger Groult at the Centre International de Toxicologie for their fruitful collaboration to this work.

	Disclosures

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
H. Sicard, S. Ingoure, B. Luciani, C. Serraz, J. Tiollier, and F. Romagné are current employees of Innate Pharma.

	Footnotes

 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ Address correspondence and reprint requests to Dr. Hélène Sicard, Innate Pharma, 119-121 ancien chemin de Cassis, Marseilles, France. E-mail address: sicard{at}innate-pharma.fr

² Abbreviations used in this paper: DC, dendritic cell; IPP, isopentenyl-pyrophosphate; BrHPP, bromohydrin pyrophosphate; rh, recombinant human; RT, room temperature; NHP, nonhuman primate; MS₂, tandem mass spectrometry.

Received for publication May 20, 2005. Accepted for publication August 2, 2005.

	References

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 

1. Balbi, B., M. T. Valle, S. Oddera, D. Giunti, F. Manca, G. A. Rossi, L. Allegra. 1993. T-lymphocytes with ⁺ V2⁺ antigen receptors are present in increased proportions in a fraction of patients with tuberculosis or with sarcoidosis. Am. Rev. Respir. Dis. 148:1685.-1690. [Medline]
2. Kroca, M., A. Johansson, A. Sjostedt, A. Tarnvik. 2001. V9V2 T cells in human legionellosis. Clin. Diagn. Lab. Immunol. 8:949.-954. [Abstract/Free Full Text]
3. Poquet, Y., M. Kroca, F. Halary, S. Stenmark, M. A. Peyrat, M. Bonneville, J. J. Fournie, A. Sjostedt. 1998. Expansion of V9 V2 T cells is triggered by Francisella tularensis-derived phosphoantigens in tularemia but not after tularemia vaccination. Infect. Immun. 66:2107.-2114. [Abstract/Free Full Text]
4. Bertotto, A., R. Gerli, F. Spinozzi, C. Muscat, F. Scalise, G. Castellucci, M. Sposito, F. Candio, R. Vaccaro. 1993. Lymphocytes bearing the T cell receptor in acute Brucella melitensis infection. Eur. J. Immunol. 23:1177.-1180. [Medline]
5. Roussilhon, C., M. Agrapart, P. Guglielmi, A. Bensussan, P. Brasseur, J. J. Ballet. 1994. Human TcR ⁺ lymphocyte response on primary exposure to Plasmodium falciparum. Clin. Exp. Immunol. 95:91.-97. [Medline]
6. Hassan, J., C. Feighery, B. Bresnihan, A. Whelan. 1991. Elevated T cell receptor ⁺ T cells in patients with infectious mononucleosis. Br. J. Haematol. 77:255.-256. [Medline]
7. Jouen-Beades, F., E. Paris, C. Dieulois, J. F. Lemeland, V. Barre-Dezelus, S. Marret, G. Humbert, J. Leroy, F. Tron. 1997. In vivo and in vitro activation and expansion of T cells during Listeria monocytogenes infection in humans. Infect. Immun. 65:4267.-4272. [Abstract]
8. Hara, T., Y. Mizuno, K. Takaki, H. Takada, H. Akeda, T. Aoki, M. Nagata, K. Ueda, G. Matsuzaki, Y. Yoshikai, et al 1992. Predominant activation and expansion of V9-bearing T cells in vivo as well as in vitro in Salmonella infection. J. Clin. Invest. 90:204.-210.
9. Behr, C., R. Poupot, M. A. Peyrat, Y. Poquet, P. Constant, P. Dubois, M. Bonneville, J. J. Fournie. 1996. Plasmodium falciparum stimuli for human T cells are related to phosphorylated antigens of mycobacteria. Infect. Immun. 64:2892.-2896. [Abstract]
10. Belmant, C., E. Espinosa, F. Halary, Y. Tang, M. A. Peyrat, H. Sicard, A. Kozikowski, R. Buelow, R. Poupot, M. Bonneville, J. J. Fournie. 2000. A chemical basis for selective recognition of nonpeptide antigens by human T cells. FASEB J. 14:1669.-1670. [Abstract/Free Full Text]
11. Constant, P., F. Davodeau, M. A. Peyrat, Y. Poquet, G. Puzo, M. Bonneville, J. J. Fournie. 1994. Stimulation of human T cells by nonpeptidic mycobacterial ligands. Science 264:267.-270. [Abstract/Free Full Text]
12. Tanaka, Y., C. T. Morita, E. Nieves, M. B. Brenner, B. R. Bloom. 1995. Natural and synthetic non-peptide antigens recognized by human T cells. Nature 375:155.-158. [Medline]
13. Rohmer, M., M. Knani, P. Simonin, B. Sutter, H. Sahm. 1993. Isoprenoid biosynthesis in bacteria: a novel pathway for the early steps leading to isopentenyl diphosphate. Biochem. J. 295:(Pt. 2):517.-524.
14. Jomaa, H., J. Feurle, K. Luhs, V. Kunzmann, H. P. Tony, M. Herderich, M. Wilhelm. 1999. V9V2 T cell activation induced by bacterial low molecular mass compounds depends on the 1-deoxy-D-xylulose 5-phosphate pathway of isoprenoid biosynthesis. FEMS Immunol. Med. Microbiol. 25:371.-378. [Medline]