Peripheral blood monocytes are responsible for γδ T cell activation induced by zoledronic acid through accumulation of IPP/DMAPP

Nitrogen-containing bisphosphonates indirectly activate Vγ9Vδ2 T cells through inhibition of farnesyl pyrophosphate synthase and intracellular accumulation of isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP), but the cells responsible for Vγ9Vδ2 T cell activation through IPP/DMAPP accumulation are unknown. Treatment of human peripheral blood mononuclear cells (PBMCs) with a pharmacologically relevant concentration of zoledronic acid induced accumulation of IPP/DMAPP selectively in monocytes, which correlated with efficient drug uptake by these cells. Furthermore, zoledronic acid-pulsed monocytes triggered activation of γδ T cells in a cell contact-dependent manner. These observations identify monocytes as the cell type directly affected by bisphosphonates responsible for Vγ9Vδ2 T cell activation.

Nitrogen-containing bisphosphonate drugs (N-BPs) are effective anti-resorptive agents commonly used for the treatment of postmenopausal osteoporosis and tumour-induced osteolysis. Intravenous administration of N-BPs is commonly associated with a flu-like acute-phase reaction involving the activation of Vc9Vd2 T cells (Kunzmann et al, 2000). As Vc9Vd2 T cells have potent anti-tumoural properties, activation and expansion of Vc9Vd2 T cells by N-BPs offers a promising strategy for cancer immunotherapy (Bonneville & Scotet, 2006). N-BPs, like dietary alkylamines, activate Vc9Vd2 T cells through inhibition of farnesyl diphosphate (FPP) synthase (Gober et al, 2003;Thompson & Rogers, 2004;Hewitt et al, 2005;Thompson et al, 2006a), most likely via an accumulation of the substrates of FPP synthase, isopentenyl diphosphate (IPP) and its stereoisomer dimethylallyl diphosphate (DMAPP), which are agonists of the Vc9Vd2 T cell receptor (TCR) (Tanaka et al, 1995). However, the exact cell type in peripheral blood that accumulates IPP/ DMAPP after exposure to N-BP remains unknown. Monocytes have previously been shown to be required for activation of Vc9Vd2 T cells by the N-BP pamidronate (Miyagawa et al, 2001), although the role of monocytes was thought to be direct presentation of N-BP to the Vc9Vd2 TCR (Miyagawa et al, 2001). It has since become clear that the activation of Vc9Vd2 T cells by N-BPs involves intracellular N-BP uptake and IPP/DMAPP accumulation (Gober et al, 2003;Thompson & Rogers, 2004). We hypothesized that monocytes, due to their high endocytic activity, are the peripheral blood cells that efficiently internalize N-BP and accumulate IPP/DMAPP, and therefore are directly responsible for triggering Vc9Vd2 T cell activation.

Reagents
Zoledronic acid (ZOL) and CGP-58318 (an analogue of ZOL; Legay et al, 2002) were provided by Novartis Pharma AG (Basel, Switzerland). All other reagents were from Sigma Chemical Company (Poole, Dorset, UK), unless otherwise stated. Mevastatin (MEV) was converted from the lactone form by dissolving 5 mg in 100 ll of ethanol and 100 ll 1 N NaOH, adding 1 ml phosphate buffered saline (PBS), and adjusting the pH to 8 using 1 N HCl. CGP-58318 was conjugated through the primary amine group to Alexa Fluor 488 (AF488-BP) or Alexa Fluor 680 (AF680-BP) as previously described for alendronate (Thompson et al, 2006b).

Cell culture and treatment
This study was approved by the North of Scotland Research Ethics Committee. Informed consent was obtained for the collection of peripheral blood from healthy volunteers in accordance with the Declaration of Helsinki. Peripheral blood mononuclear cells (PBMCs) were isolated using Lymphoprep (Axis-Shield, Oslo, Norway) density gradient centrifugation and cultured in a-minimal essential medium supplemented with 100 U/ml penicillin, 100 lg/ml streptomycin, 2 mmol/l glutamine, 10% fetal bovine serum, and 10 U/ml recombinant human interleukin 2, except for detection of drug uptake where PBMCs were isolated using PharmLyse (BD Biosciences, Oxford, UK). Murine J774AE2 macrophage-like cells were cultured as previously described (Thompson et al, 2006b), and plated at 5 · 10 5 cells/well in 6-well plates and allowed to adhere overnight. The cells were then treated as indicated in the figure legends for 24 h before harvesting.

Detection and quantification of IPP/DMAPP and ApppI
Human PBMCs (2 · 10 6 cells/ml in 75 cm 2 flasks) or J774AE2 macrophages (5 · 10 5 cells/well in 6-well plates) were treated as indicated in the figure legends, and T cells or monocytes were purified from human PBMCs by negative selection using MACS beads (Miltenyi Biotech GmbH, Bergisch Gladbach, Germany) according to manufacturer's instructions. IPP/ DMAPP and ApppI were quantified by high performance liquid chromatography negative ion electrospray ionization mass spectrometry (HPLC-ESI-MS) in acetonitrile extracts, as previously described (Mönkkönen et al, 2006), and results were expressed per milligram protein, determined by Bradford assay.

Detection of drug uptake
Human PBMCs (1 · 10 6 cells/well in 24-well plates) were treated as indicated in the legend of Fig 1, washed twice in icecold PBS, and stained with anti-CD14 (Miltenyi Biotech) or anti-CD3 antibodies (Beckman Coulter Immunotech, Palo Alto, CA, USA). Formaldehyde-fixed cells were analysed on an LSRII (BD Biosciences) flow cytometer using facsdiva software (BD Biosciences), or cytospun onto slides and counterstained with SYTOX Orange (Invitrogen, Paisley, UK) prior to analysis on a Zeiss LSM510 META system (Carl Zeiss Ltd, Welwyn Garden City, UK) using a 63· objective lens, and aim software (Carl Zeiss Ltd) for image acquisition and analysis.

Co-culture of ZOL-treated monocytes with cd T cells
Monocytes were isolated from human PBMCs using anti-CD14 magnetic bead separation (Miltenyi Biotec GmbH) according to the manufacturer's instructions. Purity of the monocyte fraction was ‡93%, as assessed by CD14-labelling and flow cytometric analysis. Isolated monocytes were plated out at 5 · 10 5 cells/well in 24-well plates, pulse-treated with 1 lmol/l ZOL or vehicle for 2 h, then washed with PBS. Vehicle-or ZOL-treated monocytes (1 · 10 5 cells/well) were then co-cultured with the monocytedepleted PBMCs (5 · 10 5 cells/well) or with cd T cells (1 · 10 4 cells/well) isolated from the monocyte-depleted fraction using a negative cd T cell isolation kit (Miltenyi Biotec GmbH), in a final volume of 500 ll medium in 24-well plates. To assess the role of cell-cell contact, transwell inserts (pore-size 0AE4 lm) were used to separate monocyte-depleted PBMCs or cd T cells from monocytes.

Detection of interferon-c release
After 72 h, conditioned medium was collected and the concentration of interferon (IFN)-c was determined using an IFN-c Quantikine enzyme-linked immunosorbent assay kit (R&D Systems, Minneapolis, MN, USA) according to the manufacturer's instructions.

Detection of cd T cell proliferation
After 7 d, the percentage of Vd2-TCR + T cells in the PBMC co-cultures was determined by immunostaining with anti-CD3 and anti-Vd2-TCR antibodies and flow cytometric analysis, as previously described (Thompson et al, 2006a).

ZOL and sec-butylamine induce IPP/DMAPP accumulation in J774AE2 macrophages and this is prevented by mevastatin
Recent studies have suggested that accumulation of IPP/ DMAPP triggers Vc9Vd2 T cell activation by N-BPs as well as alkylamines (Gober et al, 2003;Thompson & Rogers, 2004;Hewitt et al, 2005;Thompson et al, 2006a). However, the IPP/ DMAPP accumulation in response to pharmacological inhibition of FPP synthase is not well characterized. We used quantitative HPLC-ESI-MS to assess intracellular IPP/DMAPP levels in response to ZOL treatment in J774AE2 macrophages. Low concentrations of ZOL ( ‡0AE5 lmol/l) were sufficient to cause detectable accumulation of IPP/DMAPP (Fig S1A). In addition, the alkylamine sec-butylamine (SBA) also induced IPP/DMAPP accumulation at concentrations ‡5 mmol/l ( Fig S1A). The relatively high concentration of SBA required to induce IPP/ DMAPP accumulation correlated with the far lower potency of alkylamines for inhibiting FPP synthase than ZOL (Thompson et al, 2006a). Simultaneous treatment with MEV almost com- pletely prevented the ZOL-or SBA-induced IPP/DMAPP accumulation (Fig S1A). These findings provide further evidence for the notion that statins, through inhibiting HMG-CoA reductase upstream of FPP synthase, prevent N-BP-and alkylamine-induced Vc9Vd2 T cell activation by preventing the accumulation of IPP/DMAPP resulting from FPP synthase inhibition (Gober et al, 2003;Thompson & Rogers, 2004;Hewitt et al, 2005;Thompson et al, 2006a), and further confirm the mechanism of Vc9Vd2 T cell activation and proliferation by FPP synthase inhibitors. The accumulation of IPP/DMAPP in J774AE2 cells in response to both ZOL and SBA was associated with a concurrent accumulation of the recently identified novel IPP metabolite ApppI (Mönkkönen et al, 2006), and this was also blocked by MEV (Fig S1B). Whether the metabolism of IPP into ApppI affects the activation of Vc9Vd2 T cells by N-BPs or alkylamines is presently unknown.

Monocytes accumulate IPP/DMAPP in response to ZOL in human PBMC cultures
To investigate which cell type in peripheral blood is responsible for the IPP/DMAPP accumulation and resulting Vc9Vd2 T cell activation by N-BPs, we assessed intracellular IPP/DMAPP levels in human PBMC cultures in response to ZOL treatment. IPP/ DMAPP could not be detected in untreated cells, but when human PBMCs were pulsed for 2 h with 10 lmol/l ZOL, IPP/ DMAPP was clearly detectable in the non-T cell fraction, whereas T cells showed only minor levels of IPP/DMAPP ( Fig  S2). The MS/MS spectrum from the peak in the non-T cell fraction was identical to the MS/MS spectrum of synthetic IPP (not shown). When human PBMCs were pulsed with 1 lmol/l ZOL for 2 h (to mimic the estimated circulating concentration and exposure of peripheral blood cells following a standard dose of ZOL in vivo; Chen et al, 2002), ZOL-induced IPP/DMAPP accumulation was clearly detectable in monocytes purified by magnetic bead separation, but hardly detectable in the nonmonocyte fraction (Fig 1A). Isolation of monocytes based on cell adherence yielded similar results (not shown).

Peripheral blood monocytes efficiently internalize N-BP
We have shown recently that fluid-phase endocytosis is the major route by which N-BPs are internalized in J774AE2 macrophages and osteoclasts (Thompson et al, 2006b). Since peripheral blood monocytes are highly endocytic, we investigated whether the selective accumulation of IPP/DMAPP in this cell type correlated with relatively high levels of drug uptake. Indeed, in cultures of PBMCs, only CD14 + monocytes internalized large amounts of fluorescently labelled N-BP, while little or no uptake by other cell types, including CD3 + lymphocytes, could be detected at the concentrations and treatment duration used (Fig 1B-D). Similarly, only monocytes internalized large amounts of dextran (Fig 1E), a marker of fluid-phase endocytosis, and confocal microscopy demonstrated intracellular co-localization of fluorescently labelled N-BP with dextran ( Fig 1F). This suggests that N-BP uptake by peripheral blood monocytes, similar to that by J774AE2 macrophages and osteoclasts (Thompson et al, 2006b), predominantly occurs via fluid-phase endocytosis. Together, these findings indicate that, following an intravenous infusion of ZOL, only highly endocytic monocytes in peripheral blood are likely to internalize sufficient ZOL to cause a significant accumulation of IPP/DMAPP.

ZOL-pretreated monocytes activate Vc9Vd2 T cells in human PBMC cultures, and this is dependent on cell-cell contact
Having shown that monocytes accumulate detectable levels of IPP/DMAPP following treatment with a clinically relevant pulse of ZOL (1 lmol/l for 2 h), we investigated whether purified peripheral blood monocytes pretreated with the same concentration and duration of ZOL activated Vc9Vd2 T cells in co-cultures with untreated (monocyte-depleted) PBMCs. ZOL-treated monocytes induced a 2AE5 ± 1AE2-fold (n = 3) increase in IFN-c release when co-cultured with monocytedepleted PBMCs, as compared to co-culture of monocytedepleted PBMCs with vehicle-treated monocytes (Fig 2A), although this did not reach statistical significance. The increase in IFN-c correlated with a 2AE1 ± 0AE5-fold (n = 3) increase (P < 0AE05) in Vd2 + T cell number after a 7 d culture (Fig 2B), and an increase in cell clustering (not shown) in co-cultures with ZOL-treated monocytes as compared to co-cultures with vehicle treated-monocytes. Both IFN-c release and Vd2 + T cell proliferation were prevented when the monocyte-depleted PBMCs were separated from monocytes using transwell inserts, indicating that cell-cell contact is required for the activation of Vc9Vd2 T cells by ZOL-treated monocytes.

ZOL-pretreated monocytes activate purified cd T cells
Finally, we investigated whether ZOL-pretreated monocytes activated Vc9Vd2 T cells in enriched cd T cell cultures. After 72 h, IFN-c release was significantly increased in co-cultures of cd T cells with ZOL-treated monocytes, compared to co-cultures with vehicle-treated monocytes (P < 0AE05; Fig 2C).
As observed in the co-cultures of monocytes with monocytedepleted PBMCs, the increase in IFN-c release was abrogated when cd T cells were separated from monocytes using transwell inserts (Fig 2C). This strongly suggests that ZOLtreated monocytes can directly activate cd T cells via a cell contact-dependent mechanism.

Conclusions
This study demonstrated that ZOL induced IPP/DMAPP accumulation selectively in monocytes following treatment with a clinically-relevant dose and duration of ZOL, most probably a result of efficient intracellular drug uptake by this highly endocytic cell type. Furthermore, peripheral blood ATP analog (ApppI) inhibits the mitochondrial adenine nucleotide translocase (ANT) and is responsible for the apoptosis induced by nitrogen-containing bisphosphonates. British Journal of Pharmacology, 147, 437-445. Tanaka, Y., Morita, C.T., Tanaka

Supporting information
Additional Supporting Information may be found in the online version of this article: Fig S1. Mevastatin prevents both ZOL-and SBA-induced IPP/DMAPP and ApppI accumulation in J774AE2 macrophages. Fig S2. ZOL-induced IPP/DMAPP accumulation occurs predominantly in non-T cells in human PBMC cultures.
Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.
Short Report ª 2008 The Authors