Vanadate Induces Calcium Signaling, Ca²⁺ Release-Activated Ca²⁺ Channel Activation, and Gene Expression in T Lymphocytes and RBL-2H3 Mast Cells Via Thiol Oxidation¹

The role of Ca²⁺ signaling in the activation of T lymphocytes is well established (1, 2). Binding of the TCR to Ag associated with the MHC initiates a cascade of intracellular signaling that culminates in the production of cytokines and T cell proliferation. Among the proximal signaling steps that take place at the plasma membrane, tyrosine phosphorylation of substrates, including phospholipase C, generates two key second messengers, diacylglycerol, leading to the activation of protein kinase C, and 1,4,5-inositol trisphosphate (IP₃),⁵ leading to the initiation of the Ca²⁺ signal. The Ca²⁺ signal is sustained by the activation of a specific type of store-operated Ca²⁺ channel in the plasma membrane called the Ca²⁺ release-activated Ca²⁺ (CRAC) channel (1). Ca²⁺ signaling must be oscillatory or sustained at the level of individual cells to activate gene transcription via the NF-AT pathway, leading to cytokine production (3, 4, 5).

The level of cellular tyrosine phosphorylation and subsequent activation events are controlled by a balance between phosphorylation and dephosphorylation reactions. Thus, compounds that alter the balance of kinase and phosphatase activity may profoundly influence cellular activation in the immune response. Sodium orthovanadate (Na₃VO₄) is a potent inhibitor of protein tyrosine phosphatase (PTPs) and other phosphoryl transfer enzymes (6). In the +5 oxidation state, vanadate adopts a trigonal bipyramidal configuration that mimics the phosphate ion (7). In this configuration, Na₃VO₄ competes with the γ-phosphate of ATP for binding to the active site of the Na⁺,K⁺-ATPase, thus inhibiting the enzyme (8, 9). Na₃VO₄ inhibits PTPs by a similar mechanism (reviewed in Refs. 10, 11, 12). Formation of a phosphate-thioester bond with a conserved cysteine at the active site of PTPs is critical to the dephosphorylation reaction. Na₃VO₄ reversibly displaces phosphate from the active site thiol (13). In addition, thiol oxidation may permanently inactivate PTPs during prolonged treatment of cells with Na₃VO₄ (14). Cysteine could be oxidized directly by Na₃VO₄ or indirectly following the intracellular formation of pervanadate. Pervandate is a complex mixture of peroxovanadium compounds formed in vitro by mixing H₂O₂ and Na₃VO₄ (reviewed in Ref. 10). Pervanadate produces an irreversible inhibition of PTPs that can be prevented by preincubation with thiol-reducing agents, suggesting that the effect of pervanadate is mediated by the oxidation of cysteine (15). Although vanadate and pervanadate have been widely used as general inhibitors of PTPs and clearly enhance tyrosine phosphorylation in immune cells (16, 17, 18, 19, 20), the effects of vanadium compounds on other cellular targets as a result of their thiol reactivity have not been fully explored.

The modulation of intracellular signaling by vanadium compounds takes on a greater significance because they are being considered as potential therapeutic agents for use in both insulin-dependent and noninsulin-dependent forms of diabetes (21, 22, 23, 24, 25). Their insulin-mimetic actions in vivo and in vitro correlate with their ability to inhibit PTPs and induce phosphorylation of the β subunit of the insulin receptor (26, 27, 28, 29, 30). Recent studies have shown that concentrations of vanadium compounds that increase glucose transport and normalize blood glucose levels (21, 31, 32) also produce tyrosine phosphorylation and Ca²⁺ mobilization in Jurkat T cells (16, 17, 33) and RBL cells (18). Thus, potential actions by these compounds on the immune system must also be considered.

Previous attempts to characterize the action of vanadate on cells of the immune system have yielded conflicting results. In Jurkat T cells, pervanadate was reported to induce Ca²⁺ influx independent of store release (17). However, in RBL cells pervanadate appeared to release Ca²⁺ from intracellular stores (18). In many cell types, Ca²⁺ influx is coupled to the depletion of intracellular Ca²⁺ stores by a process termed capacitative Ca²⁺ entry (34, 35). In lymphocytes and RBL cells store release leads to CRAC channel activation (36, 37, 38, 39, 40, 41, 42). To clarify the mechanisms of action of vanadate, we have investigated its effects on [Ca²⁺]_i signaling in Jurkat T lymphocytes and RBL cells using single-cell imaging and electrophysiological techniques. Our experiments were designed to show whether the normal Ca²⁺ signaling mechanisms used by cells of the immune system are usurped by vanadate stimulation, whether Ca²⁺ channel activation is direct or secondary to release from Ca²⁺ stores, and whether nonphysiological mechanisms are evoked.

The human leukemia T cell line, Jurkat E6-1, and the rat basophilic leukemia cell line, RBL-2H3, were obtained from American Type Culture Collection (Manassas, VA). Jurkat-NZ cells containing a β-galactosidase reporter gene construct (lacZ) under the control of the NF-AT promoter were derived from the Jurkat subline NZDipA.1.5.22 and have been previously described and characterized (4). Jurkat E6-1 and Jurkat-NZ cells were grown in RPMI 1640 medium containing 10% heat-inactivated FBS, 10 mM HEPES, and 2 mM glutamine. Cells were cultured in 25-ml flasks (Costar, Cambridge, MA) at 37°C in 5% CO₂ in a humidified incubator. RBL cells were maintained in Eagle’s MEM supplemented with 20% FCS and 2 mM glutamine. RBL cells were plated onto glass coverslips 1–2 days before use.

Thapsigargin (TG) and dipotassium oxodiperoxo (pyridine-2-carboxylato) vanadate (bpV(pic)) were obtained from Alexis Biochemicals (San Diego, CA). 5′-Adenlylimidodiphosphate tetralithium salt was obtained from Roche Molecular Biochemicals (Basel, Switzerland), and adenosine 5′-o-(3-thiotriphosphate) tetralithium salt (ATPγS) was obtained from Calbiochem (San Diego, CA). All other salts and reagents were obtained from Sigma-Aldrich (St. Louis, MO).

Jurkat T cells were loaded with 3 μM fura-2-AM (Molecular Probes, Eugene, OR) for 30–40 min at room temperature (20–25°C). The cells were then washed three times with RPMI/10% FCS. RBL cells were loaded in medium containing 1 μM fura-2-AM and 2.5 mM probenecid for 30 min at 37°C in the incubator. These cells were then washed three times with Eagle’s MEM containing 2.5 mM probenecid and returned to the incubator for an additional 30 min to complete the hydrolysis of the fura-2 ester. After fura-2 loading, all cells were stored at room temperature in the dark until use. Before imaging, media were exchanged with mammalian Ringer containing: 160 mM NaCl, 4.5 mM KCl, 2 mM CaCl₂, 1 mM MgCl₂, 10 mM glucose, and 10 mM HEPES (pH 7.4; osmolality, 290–310 mOsm/kg). Illumination was provided by a xenon arc lamp (Carl Zeiss, New York, NY) and transmitted through a filter wheel unit (Lambda 10, Axon Instruments, Foster City, CA) containing 350- and 385-nm excitation filters. The filtered light was reflected by a 400-nm dichroic mirror through a ×63 oil immersion objective to illuminate cells. Emitted light above 480 nm was received by an intensified CCD camera (C2400, Hamamatsu Photonics, Bridgewater, NJ), and the video information was relayed to an image processing system (Videoprobe, ETM Systems, Petaluma, CA). Full field-of-view 8-bit images, averaged over 16 frames, were collected at 350- and 385-nm wavelengths. Digitally stored 350/385 ratios were constructed from background-corrected 350- and 385-nm images. Single-cell measurements of [Ca²⁺]_i were calculated from the 350/385 ratios using the equation of Grynkiewicz et al. (43) and a K_d of 250 nM for fura-2. The minimum 350/385 ratio was measured in single cells after incubation for 10 min in Ca²⁺-free Ringer containing 2 mM EGTA. Maximum ratio values were obtained after perfusion with Ringer containing 10 mM Ca²⁺, 1 μM TG, and 10 μM ionomycin.

The sodium salt of low molecular mass heparin (average m.w., 3 kDa; Sigma, St. Louis, MO) was made up to a concentration of 25 mg/ml in a 100 mM KCl solution containing 0.1% tetramethylrhodamine dextran (Molecular Probes). The heparin solution was loaded into prepulled glass capillaries (Femtotip, Eppendorf, Hamburg, Germany). Cell microinjection was performed using a pressure injector (model 5246, Eppendorf) and micromanipulator (model 5171, Eppendorf) as previously described (44). Microinjections were performed 2 h before the calcium imaging experiments.

Membrane currents were measured in Jurkat T cells using the whole-cell configuration of the patch-clamp technique (45, 46). An EPC-9 amplifier (HEKA, Lambrecht, Germany) interfaced to a Macintosh Quadra 700 computer (Apple Computer, Cupertino, CA) was used for pulse application and data recording. Membrane voltages were corrected for liquid junction potentials, and current recordings were corrected for leak and capacitative currents. Patch pipettes were pulled from Accu-fill 90 Micropets (Becton Dickinson, Parsippany, NJ) using a P87 micropipette puller (Sutter Instruments, Novato, CA). Pipettes were coated with Sylgard (Dow Corning, Midland, MI) and heat polished to a final resistances of 2–5 MΩ. Patch-clamp experiments were performed at room temperature (20–25°C). Unless otherwise indicated the membrane currents were filtered at 1.5 kHz. Data analysis was performed using the program Pulse (HEKA, Lambrecht, Germany). During the whole-cell recordings, the membrane potential was held at 0 mV, and the Ca²⁺ current through CRAC channels was measured during 200-ms voltage ramps from −100 to +40 mV applied at 1-s intervals. The pipette solution contained 128 mM Cs⁺ aspartate, 12 mM bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetate (BAPTA), 0.9 mM CaCl₂, 3.16 mM MgCl₂, and 10 mM HEPES; was titrated to pH 7.2 with CsOH; and had an estimated free Ca²⁺ concentration of 10 nM. The external solution had the following composition: 150 mM NaMeSO₃, 20 mM CaCl₂, 10 mM glucose, and 10 mM HEPES and was titrated to pH 7.4 with NaOH.

The expression of lacZ was measured using a fluorescence assay as previously described (47). In brief, Jurkat-NZ cells were seeded at 1 × 10⁵ cells/well in 96-well plates containing culture medium alone or with addition of 1 μM TG or varying concentrations of bpV(pic), each condition in the presence or the absence of 50 nM PMA. The cells were activated in a humidified incubator at 37°C with 5% CO₂ for 10 h, then lysed and incubated with 3 mM 4-methylumbelliferyl β-d-galactopyranoside (Molecular Probes). The fluorescence signal measured in a multiwell plate reader (CytoFluor Series 4000, Perspective Biosystems, Framingham, MA; 400 nm excitation, 505 nm emission) is proportional to lacZ activity driven by the NF-AT pathway.

Numerical values for single-cell [Ca²⁺]_i traces were analyzed with Origin (Microcal, Northhampton, MA). Statistical analysis was performed on data sets using Excel v5.0 (Microsoft, Redmond, WA). Data are reported as the mean ± SD. Multisample hypotheses were tested using a single-factor ANOVA and performing a Dunnett’s test to determine the significance of differences from control values. Data were considered statistically different at p < 0.05.

Previous studies (17, 18) demonstrated that application of peroxovanadium compounds increased [Ca²⁺]_i in both RBL cells and Jurkat T cells. However, different mechanisms have been postulated for the Ca²⁺ increase in each cell type. These differences may have been due to cell-specific effects of pervanadate or to different mixtures of peroxyvanadium compounds being used each study. To test the hypothesis that peroxyvanadium compounds act by a similar mechanism on T cells and mast cells we tested the effects of the cell-permeant peroxyvanadium compound, bpV(pic) on Jurkat T cells and RBL mast cells. When exposed to bpV(pic) (200 μM) in the presence of extracellular Ca²⁺, most Jurkat T cells (80 ± 10%) responded after a variable delay with an abrupt increase in [Ca²⁺]_i. In the majority of responding cells the initial rise in [Ca²⁺]_i was followed by slow oscillations with a 73 ± 4-s period and a peak [Ca²⁺]_i near 700 nM (Fig. 1,A). The average Ca²⁺ response from all observed cells (Fig. 1,A, bottom graph) rose smoothly from the resting level of 66 ± 41 nM to a plateau of 488 ± 210 nM. This population response reflects the asynchronous Ca²⁺ oscillations in single cells and the fact that the average includes nonresponding cells. Our data confirm earlier observations that peroxovanadium compounds increase [Ca²⁺]_i in Jurkat T cells (16, 17) and reveal at the single-cell level that bpV(pic) can produce Ca²⁺ oscillations analogous to those produced during Ag presentation (48). Extracellular Ca²⁺ is required to maintain the bpV(pic)-induced oscillations, because exchanging the bath solution with nominally Ca²⁺-free Ringer’s solution rapidly and reversibly terminated the oscillations (Fig. 1 A, application bar).

Treatment of RBL cells with pervanadate results in Ca²⁺ responses that vary with the peroxovanadium compounds formed in the mixture (18). Using bpV(pic), we characterized the effects of a single form of peroxovanadate on RBL cells. The average response of RBL cells to 200 μM bpV(pic) closely resembled that of the Jurkat T cells (Fig. 1,B). The [Ca²⁺]_i began to increase 200 s after the application of bpV(pic) and gradually approached a plateau of 500 ± 170 nM after an additional 1200 s. The sustained Ca²⁺ plateau required Ca²⁺ influx, because [Ca²⁺]_i returned to baseline during perfusion with nominally Ca²⁺-free Ringer’s solution (Fig. 1 B, application bar). Examination of the responses of individual cells revealed that RBL cells were less prone to oscillate than the Jurkat T cells and appeared to have a biphasic response to bpV(pic) consisting of an initial transient [Ca²⁺]_i rise followed by a secondary [Ca²⁺]_i increase to a plateau.

To separate the contributions of Ca²⁺ release and Ca²⁺ influx, we applied bpV(pic) in nominally Ca²⁺-free Ringer’s solution. For both Jurkat T and RBL cells, single-cell records show that application of bpV(pic) elicited an abrupt [Ca²⁺]_i transient following a variable delay averaging 257 ± 177 s in Jurkat T cells and 579 ± 260 s in RBL cells (Fig. 2). In the averaged response (Fig. 2, bottom graphs), the Ca²⁺ transient appeared blunted due to the variation in response times between cells. In some experiments, the variation in latency resulted in the masking of initial [Ca²⁺]_i transients in the averaged response. However, analysis of the single-cell responses clearly showed that bpV(pic) releases Ca²⁺ from intracellular stores. Ca²⁺ readdition to Ca²⁺-depleted cells caused [Ca²⁺]_i to increase rapidly to a plateau level (750 ± 40 nM in Jurkat T cells and 530 ± 140 nM in RBL cells). Thus, bpV(pic) appears to release Ca²⁺ from intracellular stores and induce capacitative Ca²⁺ influx. To determine the extent of Ca²⁺ store depletion by bpV(pic), we probed for residual Ca²⁺ in the intracellular stores with TG, a specific inhibitor of the microsomal Ca²⁺-ATPase that depletes Ca²⁺ stores in the endoplasmic reticulum. When TG (1 μM) is applied alone in nominally Ca²⁺-free Ringer’s solution, it normally produces a rapid and complete emptying of the endoplasmic Ca²⁺ stores, resulting in a Ca²⁺ transient (49). When applied after perfusion with bpV(pic) in nominally Ca²⁺-free Ringer, TG could not elicit a Ca²⁺ transient (data not shown), indicating that bpV(pic) had completely depleted the TG-sensitive Ca²⁺ store.

To assess the long term effects of vanadate treatment, we monitored transcriptional activation of the Ca²⁺-sensitive transcription factor, NF-AT. Jurkat-NZ cells contain a reporter construct in which the β-galactosidase gene (lacZ) is driven by a triple, tandem repeat of the NF-AT binding element found in the T cell IL-2 promoter. When Jurkat-NZ cells are activated, the translocation of NF-AT to the nucleus causes a dramatic increase in β-galactosidase production that can be assessed by fluorogenic substrates (4). Stimulation of Jurkat-NZ cells with PMA and TG caused a large increase in NF-AT activity, whereas PMA or TG alone had little effect (Fig. 3). Parallel experiments with bpV(pic) yielded a smaller, but still significant, increase in NF-AT activity over that in untreated cells regardless of the presence of PMA (Fig. 3). The ability of vanadate to induce NF-AT activity in the absence of PMA contrasts with the PMA requirement for gene expression triggered by TG or ionomycin (4, 50). These results suggest that vanadate may also act to induce PKC activity.

If the effect of bpV(pic) on [Ca²⁺]_i requires IP₃-evoked Ca²⁺ release, blockade of the IP₃ receptor (IP₃R) with the competitive antagonist heparin (51) should prevent the action of bpV(pic). On the other hand, if vanadate directly activates Ca²⁺ influx, as suggested previously (17), heparin should be without effect. To test the efficacy of heparin microinjection, we made use of the calcium response in RBL cells produced by P2 purinergic receptor stimulation (52). RBL cells were stimulated with the P2 receptor agonist ATP (10 μM) in the presence and the absence of microinjected low m.w. heparin. Fig. 4,A shows that purinergic stimulation produces an abrupt rise in [Ca²⁺]_i and several oscillations before [Ca²⁺]_i returns to baseline levels. As expected, heparin-injected RBL cells (Fig. 4,B) did not respond to purinergic stimulation. Stimulation by bpV(pic) was inhibited by heparin injection (Fig. 5). Both the initial Ca²⁺ transient and the delayed Ca²⁺ plateau were completely blocked. In contrast, heparin preinjection did not inhibit Ca²⁺ responses evoked by TG or ionomycin (data not shown), demonstrating that heparin did not interfere directly with CRAC channel function. Microinjection with control dextrans had no effect on the Ca²⁺ response to bpV(pic). We conclude that both transient and sustained bpV(pic)-mediated Ca²⁺ signals require IP₃-dependent release of Ca²⁺.

In Jurkat T cells, depletion of intracellular Ca²⁺ stores results in the opening of CRAC channels and Ca²⁺ influx. During passive store depletion by intracellular dialysis with heavily buffered low Ca²⁺ solutions, an initial lag period was followed by the development of inward Ca²⁺ current through CRAC channels (Fig. 6). These currents exhibit inward rectification, voltage-independent gating, and a reversal potential beyond +40 mV. Inclusion of bpV(pic) in the pipette solution accelerated the activation of the CRAC currents without altering their rectification or reversal potential (Fig. 6). Using the amplitude of the current at −80 mV, we compared the effects of bpV(pic) and vanadate with that of IP₃ on the activation of CRAC currents (Fig. 7). When the channels were activated by passive store depletion alone, the current reached its maximum (−1.18 ± 0.6 pA/pF) within 260 s (t_1/2 = 122 ± 48 s; Fig. 7,A and Table I). Addition of IP₃ (10 μM) to the intracellular solution accelerated activation of the CRAC current, reducing the t_1/2 of activation to 14 ± 7 s, but did not alter the final amplitude of the current (Fig. 7,B and Table I). Dialysis of the cell with solutions containing bpV(pic) or Na₃VO₄ (10 μM) also accelerated the activation of CRAC channels (Fig. 7, C and D) reducing the t_1/2 to 14 ± 5 and 29 ± 18 s, respectively. Thus, the enhanced rate of activation of CRAC currents by Na₃VO₄ and bpV(pic) mimicked the well-studied effects of IP₃ on CRAC currents.

The actions of Na₃VO₄ and bpV(pic) on CRAC channels could result from increased tyrosine phosphorylation secondary to phosphatase inhibition. We used nucleotide replacement to test whether the effects of NaVO₄ could be duplicated by changes in the level of cellular phosphorylation. ATPγS is an ATP analogue that is readily used by kinases (53, 54, 55, 56). When the thiophosphoryl group from ATPγS is transferred to a protein, the resulting bond is resistant to the action of phosphatases. Therefore, nucleotide substitution of ATPγS for ATP should mimic the effects of inhibiting intracellular phosphatases. Intracellular dialysis with 5 mM ATPγS did not alter the amplitude or rate of activation of CRAC currents (Table I). Furthermore, intracellular dialysis with adenlylimidodiphosphate tetralithium salt (5 mM), a nonhydrolyzable analogue of ATP that should reduce protein phosphorylation, did not reduce the rate of CRAC channel activation. In addition, the serine phosphatase inhibitors okadaic acid and calyculin did not increase the rate of CRAC channel activation (Table I). Thus, treatments that shift the phosphorylation status of cellular proteins did not duplicate the effects of vanadate on CRAC channels.

To determine whether the effects of vanadate on CRAC currents are mediated by interaction with cysteine residues, we tested whether the action of Na₃VO₄ could be prevented by the thiol-reducing agent DTT. Fig. 8, A–C, shows that DTT (5 mM; in pipette and bath) prevented the activation of CRAC currents by Na₃VO₄ (10 μM). Comparison of rates of activation shows that DTT can prevent activation of CRAC currents by Na₃VO₄, but not IP₃ (Table I). To test whether thiol oxidation alone could enhance the activation of CRAC channels, we included chloramine T, an agent that selectively oxidizes methionine and cysteine residues (57), in the intracellular solution. The rate of activation of CRAC channels was enhanced by chloramine T with respect to activation during passive store depletion (Fig. 8, C and D, and Table I). These data demonstrate that intracellular dialysis with vanadate or chloramine T produces an increased rate of activation of CRAC currents. Shifting the intracellular redox state to a more reducing environment eliminates the activity of both compounds, suggesting that the effects of vanadate and chloramine T are mediated by thiol oxidation. We further evaluated the hypothesis that chloramine T and vanadate act via the same mechanism. Fig. 9,A shows that chloramine T, like pbV(pic) (Fig. 5), produces a gradual rise in [Ca²⁺]_i from baseline to micromolar levels that was inhibited by preinjection with heparin. Thus, Ca²⁺ signaling induced by either vanadate or chloramine T requires IP₃-dependent intracellular Ca²⁺ release.

The data presented above demonstrate that treatment with the peroxovanadium compound, bpV(pic), induces Ca²⁺ release from intracellular stores and Ca²⁺ entry in Jurkat T lymphocytes and RBL cells. Single-cell [Ca²⁺]_i imaging revealed oscillations as well as sustained [Ca²⁺]_i signals when extracellular Ca²⁺ was present. In the absence of extracellular Ca²⁺, Ca²⁺ release transients evoked by vanadate led to complete depletion of the TG-sensitive store. Heparin, which blocks the association of IP₃ with its receptor, blocked bpV(pic)-induced Ca²⁺ signaling. Remarkably, intracellular dialysis with bpV(pic) or Na₃VO₄ activated CRAC channels as rapidly as dialysis with IP₃. CRAC channel activation by vanadium compounds was prevented by incubation with the thiol-reducing agent DTT and was mimicked by the thiol oxidant chloramine T, suggesting that a redox reaction contributes to their mechanism of action. Ca²⁺ signaling induced by bpV(pic) resulted in downstream effects on gene expression by activating the transcription factor, NF-AT, without a requirement for costimulation by addition of PMA.

Where does vanadate act to promote Ca²⁺ signaling? In principle, vanadate could produce Ca²⁺ influx directly by activating Ca²⁺ channels in the plasma membrane or indirectly by releasing Ca²⁺ from intracellular stores. The CRAC activation by vanadate was extremely rapid (Table I), suggesting that vanadate bypasses the membrane-delimited, rate-limiting steps in Ca²⁺ influx activation following TCR engagement. However, rapid kinetics alone do not indicate that CRAC channels are directly affected by vanadate. Our data clearly demonstrate that vanadate elicits the release of Ca²⁺ from intracellular stores in both RBL and Jurkat T cells (Fig. 2). Furthermore, heparin inhibited vanadate-induced Ca²⁺ entry in RBL cells (Fig. 5), indicating that Ca²⁺ entry is secondary to intracellular Ca²⁺ store depletion and requires the binding of IP₃ to its receptor in the endoplasmic reticulum. Finally, currents activated by vanadate were indistinguishable in every characteristic from those activated by dialysis with IP₃ or by passive Ca²⁺ store depletion (Fig. 6). Taken together, our results eliminate the possibility that vanadate produces Ca²⁺ influx by direct activation of CRAC channels or any other Ca²⁺ influx mechanism. These data are consistent with earlier studies showing pervanadate-induced Ca²⁺ entry secondary to store release in single RBL cells (18), but conflict with results in Jurkat T cells suggesting that pervanadate directly activates membrane Ca²⁺ channels (17). The latter conclusion was drawn when a rise in [Ca²⁺]_i was not observed in the absence of extracellular Ca²⁺. However, that study was performed on cell suspensions using a spectrophotometer in which the average Ca²⁺ response from many cells was sampled simultaneously. Our results using video imaging of individual cells show that Ca²⁺ transients occur asynchronously following treatment with bpV(pic). Small transients in the absence of extracellular Ca²⁺ would be missed in population studies (see, for example, Fig. 2 B). Therefore, we conclude that vanadate induces Ca²⁺ influx secondarily to promoting IP₃ production and Ca²⁺ store release, consistent with previous results demonstrating increased PLCγ activity and higher cytosolic IP₃ levels in intact cells (16, 18).

How does vanadate act chemically inside the cell to promote Ca²⁺ signaling? Vanadate and bpV(pic) are multifunctional reagents that can interact with a number of cellular components in two primary modes of action: 1) mimicking the transition state complex formed by the phosphate ion during phosphoryl transfer reactions; and 2) directly oxidizing cysteine thiols (6). Although Na₃VO₄ does not shift the total cellular redox state in lymphocytes, which would appear to emphasize the importance of the first mechanism (19), measurements of the total cellular redox state principally assess the status of nonprotein thiols; thus, the selective oxidation of active site cysteines may not be reflected in those experiments. In fact, a number of studies demonstrate that thiol oxidation plays an important role in the mechanism by which vanadium compounds inhibit PTPs. First, inactivation of PTP1B by vanadium compounds is prevented, but not reversed, by the reducing agent DTT (15). Second, pervanadate directly oxidizes the active site cysteine to cysteic acid in PTP1B (15). Third, Na₃VO₄ and pervanadate oxidize cysteine sulfhydryls (58). Other agents that cause thiol oxidation, such as thimerosal, have been reported to enhance Ca²⁺ release from intracellular stores through direct interaction with the IP₃R (59, 60, 61). Thimerosal was shown to increase both the single-channel conductance and the mean open time of the reconstituted Ca²⁺ release channel (62). In addition, thimerosal-induced Ca²⁺ influx has been linked to the release of Ca²⁺ from intracellular stores and the activation of CRAC channels in the plasma membrane of RBL-2H3 cells (63). Our data show that chloramine T mimics and DTT reverses the activation of CRAC channels by Na₃VO₄ and bpV(pic) (Figs. 8 and 9). Thus, our results favor oxidation as vanadate’s mode of action. The primary target must be more proximal than the IP₃ receptor, because heparin blocks bpV(pic)-induced Ca²⁺ signaling. Furthermore, vanadate-induced gene expression did not require the presence of phorbol ester (Fig. 3), suggesting a direct action of vanadate on PLCγ to generate adequate diacylglycerol for PKC stimulation. Additional effects that might potentiate activity of the IP₃R activity are not excluded.

Of the myriad of cellular actions demonstrated with vanadium compounds, their insulin-mimetic effects have garnered the most clinical interest. Micromolar concentrations of vanadate and peroxovanadium compounds stimulate hexose uptake, glucose oxidation, and lipogenesis in vivo and in vitro (26, 27, 28, 31, 64). Clinical trials demonstrating that sodium metavanadate and vanadyl sulfate improve insulin sensitivity and fasting blood glucose levels have led to suggestions for use of these agents in adjunctive therapy in diabetes (22, 23, 24, 25). The clinically applied forms of vanadate have been shown to interconvert within the cell to the +5 oxidation state and to pervandate depending upon the intracellular redox state (reviewed in Refs. 10, 11 , and 65). Thus, the effects of vanadate on Ca²⁺ signaling may contribute to the insulin-mimetic properties of these compounds. Unfortunately, beyond their beneficial metabolic effects, long term mitogenic and potentially tumorigenic effects of vanadium compounds must be considered (reviewed in Ref. 60). When applied to lymphocytes at clinically relevant concentrations, vanadium compounds mimic receptor-mediated activation. Our data show that bpV(pic) activates NF-AT-dependent gene expression in Jurkat T cells. Previous reports have demonstrated that peroxyvanadate compounds activate c-Jun and c-Fos and induce nuclear translocation of NF-κB in lymphocytes (17, 19). Recently, peroxovanadium compounds were shown to stimulate HIV-1 production in latently infected cell lines in conjunction with T cell activation (20). Thus, studies of the mechanisms by which vanadium compounds interact with lymphocytes may have relevance to its proposed clinical application and yield new approaches to other diseases of the immune system.

We appreciate the expert technical assistance of Dr. Luette Forrest during this project. The use of the Optical Biology Shared Resource of the Chao Family Comprehensive Cancer Center is gratefully acknowledged.