Inhibition of the calcium release-activated calcium (CRAC) current in Jurkat T cells by the HIV-1 envelope protein gp160.

The HIV-1 envelope glycoprotein gp120/160 has pleiotropic effects on T cell function. We investigated whether Ca(2+) signaling, a crucial step for T cell activation, was altered by prolonged exposure of Jurkat T cells to gp160. Microfluorometric measurements showed that Jurkat cells incubated with gp160 had smaller (approximately 40%) increases in [Ca(2+)](i) in response to phytohemagglutinin and had a reduced Ca(2+) influx (approximately 25%). gp160 had similar effects on Jurkat cells challenged with thapsigargin. We used the patch clamp technique to record the Ca(2+) current, which is responsible for Ca(2+) influx and has properties of the calcium release-activated Ca(2+) current (I(CRAC)). gp160 reduced I(CRAC) by approximately 40%. The inhibitory effects of gp160 were antagonized by staurosporine (0.1 microm), an inhibitor of protein-tyrosine kinases and protein kinase Cs (PKCs), and by Gö 6976 (5 microm), an inhibitor acting especially on PKC alpha and PKC beta I. 12-O-Tetradecanoyl phorbol 13-acetate (16 nm), a PKC activator, reproduced the effects of gp160 in untreated cells. A Western blotting analysis of PKC isoforms alpha, beta I, delta, and zeta showed that only the cellular distribution of PKC alpha and -beta I were significantly modified by gp160. In addition, gp160 was able to modify the subcellular distribution of PKC alpha and PKC beta I caused by phytohemagglutinin. Therefore the reduction in I(CRAC) caused by prolonged incubation with gp160 is probably mediated by PKC alpha or -beta I.

The functions of CD4 ϩ T cells are impaired early in HIV-1 1 infection (1-2) at a time when very few (less than 1%) cells appear to be productively infected (2)(3). This might be due to the multiple effects of virus proteins such as HIV-1 Tat (4), virus protein R (5-6) and gp120, which are released by infected cells. The major HIV-1 surface glycoprotein, gp120, is present at a high concentration in peripheral lymphoid organs (7) and, thus, may interact with uninfected cells. HIV-1 gp120 and its precursor gp160 have several effects on T lymphocytes. These include membrane depolarization (8), altered protein-tyrosine kinase p56 lck activity (9 -12), and down-modulation of cell surface CD4 molecules (13). The gp120 protein also causes changes in PKC activity and in the intracellular free Ca 2ϩ concentration (14 -16) of T lymphocytes and opens Ca 2ϩ inward currents in astrocytes (17). The induction of gp160 but not that of gp120 has also been reported to increase the intracellular free Ca 2ϩ concentration in CD4 ϩ cells (18). The alterations in Ca 2ϩ signaling due to gp120/160 (14 -16) are particularly noteworthy because of the role of Ca 2ϩ in T cell activation and proliferation (19 -20). Last, gp120 alters the apoptotic/proliferative processes in T cells (12,(21)(22).
Activation of T cell antigen receptors by antigens or mitogens such as the lectin phytohemagglutinin (PHA) leads to a biphasic rise in [Ca 2ϩ ] i resulting from an initial inositol 1,4,5triphosphatedependent release of Ca 2ϩ from intracellular stores followed by an influx of Ca 2ϩ across the plasma membrane (for review, see Ref. 23). Recent studies suggest that the Ca 2ϩ influx across the plasma membrane is carried by a storeoperated Ca 2ϩ current (24) whose biophysical properties are very similar to those of the calcium release-activated calcium current (I CRAC ) of mast cells (Ref. 25, for review, see Ref. 20). Thus the sustained rise in [Ca 2ϩ ] i , which leads to T cell proliferation (15, 26 -27), requires an appropriate I CRAC . Impairment of the Ca 2ϩ influx has been reported to be responsible for severe immunodeficiency (28). We have therefore investigated the prolonged effects of HIV gp160 protein on (i) the intracellular Ca 2ϩ response to PHA and the reticulum Ca 2ϩ -ATPase inhibitor, thapsigargin, (ii) the calcium release-activated calcium current I CRAC , and (iii) the subcellular distribution of several PKC isozymes in Jurkat T cells.
HIV-1 MN/LAI gp160 glycoprotein was provided by Dr. Kieny (Transgene, Strasbourg, France) and Dr. El Habib (Aventis, Marcy l'Etoile, France). Cells were grown in medium containing 25 g/ml gp160 for 5 days. This concentration depolarizes Jurkat cell membranes and decreases the voltage-gated K ϩ current (8). In PKC subcellular distribution studies, PHA 10 g/ml was added to the cell culture media for 30 min.
All reagents and chemicals were from Sigma except Gö 6976 (Calbiochem). Herbimycin A, staurosporine, 12-O-tetradecanoyl phorbol 13acetate (TPA), okadaic acid, and Gö 6976 were added 30 min before recording CRAC currents. * 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.
All anti-PKC antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-rabbit and anti-goat secondary antibodies coupled to peroxidase were from Sigma.
Intracellular free Ca 2ϩ Concentration ([Ca 2ϩ ] i ) Measurements-Cells were first allowed to adhere to a 0.17-mm thick, poly-D-ornithine-coated glass coverslip inserted into a laboratory-made perfusion chamber (ϳ15-l volume). Cells were rinsed twice with PBS, then incubated in phosphate-buffered saline supplemented with 1% bovine serum albumin (Invitrogen) and 4 M fura 2-AM (Molecular Probes, Eugene, OR) for 30 min at room temperature. Then cells were superfused with saline containing 160 mM NaCl, 4.5 mM KCl, 2 mM CaCl 2 , 1 mM MgCl 2 , 5 mM Hepes, pH 7.4/NaOH or the same solution without Ca 2ϩ .
Digital video image ratios (340/380-nm excitation and 510-nm emission) were obtained at room temperature (20 -25°C) with a fluorescence microscope (Diaphot 300; Nikon Corp., Tokyo, Japan) and a dynamic video imaging system (Argus-50; Hamamatsu Photonics K.K., Hamamatsu City, Japan). The 340-and 380-nm image pairs were acquired every 5 or 20 s and stored in computer memory for further processing. The ratios were converted to [Ca 2ϩ ] i via an in vitro calibration procedure using an intracellular-like solution containing 10 mM NaCl, 119 mM KCl, 1.2 mM MgCl 2 , 10 mM Hepes, pH 7.25/KOH, plus 20 M fura-2 (Molecular Probes). Nine free Ca 2ϩ concentrations (1-705 nM) were generated by adding appropriate amounts of EGTA according to stability constants for all reactions between Ca 2ϩ , Mg 2ϩ , H ϩ , and EGTA (29 -30).
Electrophysiology-Cells for patch clamp recordings (31) were allowed to adhere to a poly-D-ornithine-coated plastic culture dish for 10 min. Each dish was then rinsed twice with a Ca 2ϩ -free saline containing 160 mM NaCl, 4.5 mM KCl, 2 mM MgCl 2 , 5 mM Hepes and adjusted to pH 7.4 with NaOH. Pipettes were pulled from borosilicate glass capillaries (GC150, Clark Electromedical Instruments, Pangbourne Reading, England) coated with Sylgard TM (Dow Corning, Midland, MI) and firepolished. The pipette tip resistance was 5-10 megaohms. Membrane currents were recorded with an Axopatch 1-D amplifier (Axon Instruments, Foster City, CA). Voltage clamp protocols were implemented, and data were acquired with pClamp 6.0 software (Axon Instruments). Capacitative currents were cancelled.
I CRAC was measured in the whole-cell configuration using the following pipette solution: 140 mM cesium aspartate, 2 mM MgCl 2 , 10 mM EGTA, 10 mM Hepes, pH 7.2/CsOH (calculated free Ca 2ϩ concentration ϳ 1 nM). At the disruption of the membrane patch, the holding potential (V H ) was set at 0 mV to minimize basal Ca 2ϩ influx. I CRAC was induced by passively depleting intracellular Ca 2ϩ stores (24 -26, 32-33), by incubating cells in Ca 2ϩ -free solution for 10 min, and by dialyzing its interior with a Ca 2ϩ -free solution. A 200-ms ramp-voltage protocol (Ϫ100 to ϩ50 mV) was used to test for voltage dependence. I CRAC was also evoked by applying a 100-mV step during 200 ms. I CRAC time course was monitored over a 10-min time range by measuring I CRAC 10 ms after pulse imposition every 2 s (Figs. 3B and 4A). The time courses were fitted by a Weibull equation with five parameters using Sigma-Plot 5.0 software (SPSS ASC, Erkrath, Germany).
Membrane potential (V m ) values were estimated by a non-invasive method (34). The pipette solution contained 140 mM KCl, 5 mM NaCl, 10 mM EGTA, 5 mM Hepes, pH 7.2/KOH. A Ϫ200 to ϩ100 mV ramp of 500 ms was applied in the cell-attached configuration. Under the condition used, the equilibrium potential for K ϩ across the membrane patch was close to 0 mV, and the potential value corresponding to zero current was taken as an estimate of ϪV m . This technique allowed us to monitor V m in intact cells.
Protein Extraction-After corresponding treatment, Jurkat cells at 5 ϫ 10 6 cells/ml were centrifuged at 900 ϫ g during 10 min. The pellet was resuspended in 0.5 ml of buffer A (20 mM Tris-HCl, pH 7.5, 0.25 M sucrose, 10 mM EGTA, 2 mM EDTA, 1 tablet of Complete TM /10 ml (Mini protease inhibitor from Roche Molecular Biochemicals). The suspension was sonicated for 5 min and centrifuged at 39,000 rpm, 4°C, for 1 h. The supernatant was used as the cytosol fraction. The pellet was resuspended in buffer A plus 1% Triton X-100 and mixed on ice. The suspension was centrifuged at 39,000 rpm, 4°C, for 1 h. This supernatant was used as the Triton-soluble membrane fraction. The second pellet was resuspended in buffer B (150 mM NaCl, 50 mM Tris-HCl, pH 7.5, 1 mM EDTA, 1% Nonidet P-40, 0.1% sodium deoxycholate, 0.1% SDS plus 1 tablet of Complete TM /10 ml), mixed for 10 min at 4°C, and then centrifuged at 39 000 rpm, 4°C, for 1 h. The supernatant was used as Triton-insoluble fraction. Protein concentrations were determined using the Folin reaction and bovine serum albumin as a standard.
Immunoblotting-Equal amounts of cytosolic, membrane, and Triton-insoluble fractions (30 g) were solubilized in an electrophoresis buffer (66 mM Tris-HCl pH 6.8, 10% glycerol, 2.5% SDS, 0.1% bromphe-nol, and 2% ␤-mercaptoethanol). The proteins were subjected to 8% acrylamide gel electrophoresis. Gels were calibrated using prestained molecular mass standards (Bio-Rad). After electrophoresis, proteins were transferred to 0.45-m polyvinylidene difluoride membranes (Immobilon-P, Sigma). Then the membranes were blocked by incubation in phosphate-buffered saline containing 1% milk and 1% bovine serum albumin for 1 h at room temperature. The membranes were washed and incubated in blockade buffer plus PKC isozyme antibody at 1:500 overnight at 4°C. The blots were then washed three times in phosphatebuffered saline containing 1% milk and 1% bovine serum albumin and incubated with an anti-rabbit or anti-goat secondary antibody linked to peroxidase. The blots were washed three times in phosphate-buffered saline. Western blots were revealed with ECL plus (Amersham Biosciences, Inc.). Autoradiograms were analyzed and quantified by Scion Image software (Scion Corporation, Frederick, MA).
Statistical Analysis-Values are the means Ϯ S.E. Student's unpaired t test was used. A p value Ͻ 0.05 was considered significant. n values represent the number of cells investigated, and N is the number of independent microfluorimetric experiments performed. min (n ϭ 75, n ϭ 2). These plateau levels were dramatically dependent on the presence of external CaCl 2 (Fig. 1, A and B), which is consistent with a PHA-induced entry of Ca 2ϩ (26); thus, [Ca 2ϩ ] i decreased immediately when external Ca 2ϩ was removed. The calcium influx rate was measured for the first 10 -20 s after adding extracellular Ca 2ϩ back to the bath (Fig.  1, A and B). During this period, [Ca 2ϩ ] i was still low, and the Ca 2ϩ pump activity was greatly reduced (35). gp160 reduced the Ca 2ϩ influx by 25%; it was 11 Ϯ 1 nM/s in gp160-treated cells (n ϭ 75) and 14.5 Ϯ 1 nM/s in untreated cells (n ϭ 125, p Ͻ 0.05).

PHA-induced [Ca 2ϩ ] i Increase in
Thapsigargin-induced [Ca 2ϩ ] i Increase in Untreated and gp160-treated Jurkat Cells-The reduction of Ca 2ϩ influx caused by gp160 may be due to an effect on the CRAC channels that are responsible for Ca 2ϩ entry or to an alteration in the PHA transduction signal. We bypassed a possible alteration of the T cell antigen receptor complex by gp160 by testing the effects of thapsigargin (TG), which inhibits Ca 2ϩ re-uptake by the endoplasmic reticulum and opens CRAC channels (32).
The changes in Ca 2ϩ in response to TG were first measured in the absence of external Ca 2ϩ . Under this condition, basal  (Fig. 1D). The delays (1.6 Ϯ 0.4 min for control and 1.5 Ϯ 0.4 min for gp160-treated cells) were similar to those in the presence of external Ca 2ϩ (Fig. 1, A and B). This agrees with the fact that the initial rise in [Ca 2ϩ ] i after mitogen stimulation in the presence of external Ca 2ϩ is due to a release of Ca 2ϩ from the endoplasmic reticulum (32). As also shown in Fig. 1, C and D, [Ca 2ϩ ] i spontaneously returned to its basal concentration after 4.1 Ϯ 0.5 min (n ϭ 134, N ϭ 3) in untreated cells and after 4.5 Ϯ 0.7 min (n ϭ 84, N ϭ 3) in gp160-treated cells in the absence of external Ca 2ϩ . This suggests that gp160 did not alter the activity of the plasma membrane Ca 2ϩ -ATPases, as extrusion of Ca 2ϩ from the cytosol is essentially due to plasma membrane Ca 2ϩ -ATPases in the presence of TG.
Adding 2 mM Ca 2ϩ to the bath in the continued presence of 1 M TG induced strikingly different responses in untreated and gp160-treated cells (Fig. 1, C and D). [Ca 2ϩ ] i peaked at 583 Ϯ 66 nM and had a value of 399 Ϯ 15 nM (n ϭ 134, N ϭ 3, Fig. 1C) 7 min after Ca 2ϩ readdition in untreated cells, but the peak was only 405 Ϯ 31 nM, and the value at 7 min was 161 Ϯ 14 nM (n ϭ 84, N ϭ 3, Fig. 1D, p Ͻ 0.01) in gp160-treated Jurkat cells. The calcium influx rate measured after adding extracellular Ca 2ϩ back to the bath in the presence of TG was reduced by 31% in gp160-treated cells (9 Ϯ 1.5 nM/s (n ϭ 84) versus 13 Ϯ 2 nM/s (n ϭ 134), p Ͻ 0.05). This reduction was in the same range as those obtained in PHA experiment. This indicates that gp160 reduces the Ca 2ϩ influx by acting at some point downstream of the binding of PHA to cell surface receptors. CRAC Current Properties in Jurkat Cells-I CRAC was induced by passively depleting intracellular stores (see "Experimental Procedures"). Hyperpolarization of Jurkat cell membranes evoked a weak inward current in the absence of external Ca 2ϩ (Fig. 2A). Adding 10 mM external CaCl 2 resulted in the appearance of an inward, rapidly inactivating current (Fig. 3A) that was further increased by 20 mM external CaCl 2 . The opening of the inward current was independent of voltage ( Fig. 2B). This current was inhibited by the Ca 2ϩ current blockers, 5 mM Ni 2ϩ ions and 1 mM Cd 2ϩ ions (Fig. 2C) (36). These inhibitors act on the same current because their effects are not additive (not shown). These kinetic and pharmacological characteristics identified this current as I CRAC (25). The time course of I CRAC showed slow inactivation. I CRAC , measured by a hyperpolarization pulse, reached a peak of Ϫ39.6 Ϯ 5.1 pA in 28 Ϯ 1 s (n ϭ 5) after adding 10 mM CaCl 2 and then slowly decreased to close to the initial basal level in Ϸ6 min (352 Ϯ 64 s, n ϭ 5, half-inactivation time of 106 Ϯ 12 s, see Fig.  3A). The peak I CRAC obtained in the presence of 20 mM CaCl 2 was Ϸ36% higher than the I CRAC obtained in the presence of 10 mM CaCl 2 (Ϫ53.7 Ϯ 6.4 pA), and it was reached in half the time (12 Ϯ 2 s, n ϭ 5, not shown). The I CRAC slow inactivation was accelerated, as the current returned to its basal amplitude in less than 2 min (82 Ϯ 22 s, n ϭ 5, not shown). A, the peak current evoked by a Ϫ100-mV pulse was plotted against time to monitor the change in the current over 5 min. The inward current increased rapidly upon adding 10 mM CaCl 2 (black bar) and then decreased slowly. Inactivation was more rapid in gp160-treated than in untreated Jurkat T cells. These recordings are representative of five experiments. B, histograms for CRAC peak current (open bar) and half-inactivation time (black bar) in gp160-treated (n ϭ 5) and untreated (n ϭ 5) Jurkat T cells. *, p Ͻ 0.05, significant difference between gp160-treated and untreated cell I CRAC intensity. §, p Ͻ 0.05, significant difference between gp160-treated and untreated cell I CRAC half-inactivation time.
Staurosporine (0.1 M), a PKC and protein-tyrosine kinase inhibitor, increased the peak I CRAC in gp160-treated Jurkat cells (Fig. 4, A and B, n ϭ 5, p Ͻ 0.02); the time needed to reach the current peak was longer (32 Ϯ 3 s) than in controls (14 Ϯ 3 s, n ϭ 5, p Ͻ 0.01), and the return of I CRAC to its basal level was slower (Fig. 4, A and B, p Ͻ 0.01). Thus, staurosporine antagonized all the effects of gp160 on I CRAC . The profound effects of staurosporine and the ineffectiveness of genistein and herbimycin A strongly suggest that a PKC-mediated phosphorylation is involved in the decrease of I CRAC by gp160. Adding TPA (16 nM), a PKC activator, to gp160-treated cells had no effect on the time course of I CRAC (Fig. 4, A and B, peak current of Ϫ23.9 Ϯ 2.1 pA, n ϭ 5, p ϭ 0.32).
Measurements of I CRAC in untreated Jurkat cells also implicate a PKC-mediated phosphorylation in the regulation of this current. Staurosporine (0.1 M) had no effect on I CRAC intensity in untreated cells (p ϭ 0.23, Fig. 4, C and D). In contrast, 16 nM TPA significantly decreased I CRAC in untreated cells (p Ͻ 0.02, Fig. 4, C and D) and accelerated the slow inactivation. Okadaic acid (1 M), which inhibits phosphatases PP1 and PP2A, thus favoring phosphorylation activity, had effects similar to those of TPA and decreased I CRAC (p Ͻ 0.01, Fig. 5D) in untreated cells. Conversely, this inhibitor had no effect on I CRAC (Ϫ19.9 Ϯ 5 pA, n ϭ 5, p ϭ 0.19, Fig. 4B) in gp160-treated cells.
The above results indicate that I CRAC is reduced in gp160treated cells via a staurosporine-sensitive PKC pathway. We investigated the effects of Gö 6976, which is considered as inhibiting preferentially PKC␣ and PKC␤I (37)(38). This agent is interesting because PKC␣ is the predominant isoform in Jurkat cells, and PKC␤I has been implicated in I CRAC downmodulation (39). On gp160-treated cells, Gö 6976 (5 M) increased the time to reach the peak current after the addition of 10 mM CaCl 2 (Fig. 4E) and slowed current inactivation (Fig. 4F). In contrast, Gö 6976 had no effect in untreated cells (Fig. 4F).

Subcellular Distribution of Protein Kinase C Isozymes-Be-
cause HIV-1 envelope protein disturbs the activity of PKC (14 -16), an immunoblotting analysis was performed to investigate the translocation of several PKC isozymes from the cytosol to the membrane and the Triton-insoluble fraction in response to gp160 treatment, PHA stimulation, and PHA stimulation after gp160 treatment. The PKCs were chosen according to their subgroups, which are classical (␣, ␤I), novel (␦), and atypical (). Two independent experiments have been done for the four PKCs under all conditions. The PKC␣ and ␤I appeared at an apparent molecular mass of 80 kDa. PKC␦ antibody detected two bands with apparent molecular masses of 95 and 110 kDa and the antibody against PKC, two bands of 74 and 80 kDa. The results show that gp160 induced large alterations of the cellular distribution of PKC␣ and PKC␤I (Fig. 5, A and  B), whereas PKC␦ and PKC were less affected (Fig. 5A).
In untreated cells, PKC␣ was detected predominantly in the cytosol (Ϸ92% of protein amount), with a weak amount in the membrane (Ϸ4%) and the Triton-insoluble fraction (Ϸ4%, Fig.  5, A and B). In gp160-treated cells, the total amount of PKC␣ was increased by Ϸ25%. The newly synthesized protein appeared in the membrane fraction (Ϸ7-fold increase) and, to a lesser extent, in the Triton-insoluble fraction (Ϸ3-fold increase), since the amount of cytosolic protein was unchanged. PHA treatment (30 min) had no effect on PKC␣ total amount. However, PHA induced a translocation of PKC␣ from cytosol to the membrane (Ϸ6-fold increase) and Triton-insoluble fractions (Ϸ2.5-fold increase). Interestingly, gp160 antagonized the effect of PHA by reducing the amount of PKC␣ in the membrane (Fig. 5, A and B).
PKC␤I was only detected in cytosolic and Triton-insoluble fractions. In untreated cells, Ϸ75% of PKC␤I was detected in the Triton-insoluble fraction, and Ϸ25% was detected in cytosol. gp160 induced a 3-fold increase in total PKC␤I amount; thus, the cytosolic and Triton-insoluble fraction amounts increased by Ϸ400 and Ϸ 150%, respectively (Fig. 5, A and C). In contrast, PHA, which did not modify the total amount of protein, induced a 14-fold increase of protein amount in the cytosol and the almost complete disappearance of the Triton-insoluble PKC␤I. gp160 antagonized the effect of PHA by preventing the disappearance of PKC␤I from the Triton-insoluble fraction (Fig. 5, A and C).
PKC␦ was only detected in cytosolic and Triton-insoluble fractions. In untreated cells, 90% of PKC␦ was cytosolic ( 2 ⁄3 for the 110-kDa band, 1 ⁄3 for the 95-kDa band). gp160 treatment had no significant effect on total PKC␦ amount but induced a slight increase in the Triton-insoluble fraction, keeping the two isoforms in the same proportions. PHA had no effect per se but could partly counteract the effect of gp160; PHA reduced the Triton-insoluble fraction and slightly increased the cytosolic fraction.
The PKC was only detected in the cytosol under the form of Membrane Hyperpolarization Induced by PHA in Untreated and gp160-treated Cells-The resting V m of untreated cells was Ϫ62 Ϯ 2 mV (n ϭ 26), in good agreement with published data on human T lymphocytes (40) and Jurkat cells (8). Representative V m variations in response to PHA are plotted versus time in Fig. 6. Adding 10 g/ml PHA rapidly caused membrane hyperpolarization (Fig. 6) that reached a plateau at Ϫ87 Ϯ 4 mV (⌬V m ϭ Ϫ25 Ϯ 4 mV, n ϭ 5) in 4 Ϯ 1 min (n ϭ 5). Hyperpolarization was maintained in the presence of PHA throughout the recording time (ϳ11-min check, Fig. 6). The resting V m of gp160-treated cells (V m ϭ Ϫ50 Ϯ 3 mV, n ϭ 29; p Ͻ 0.05) was significantly less negative as compared with untreated cells. Although PHA hyperpolarized gp160-treated cells by Ϫ30 Ϯ 5.5 mV (n ϭ 5) within in 3.2 Ϯ 0.8 min (n ϭ 5), the hyperpolarization of gp160-treated cells was transient (Fig.  6). V m rapidly returned to its resting value within 7.3 Ϯ 1.3 min (n ϭ 5) in the continued presence of PHA.

DISCUSSION
Inhibition of I CRAC by gp160 -PHA triggers dramatically different Ca 2ϩ responses in gp160-treated and control cells. Although untreated cells showed a 400 nM increase in [Ca 2ϩ ] i , sufficient for IL-2 synthesis and cell activation (41), [Ca 2ϩ ] i increased by only half that amount in gp160-treated cells. A qualitatively comparable attenuation was reported in human CD4 ϩ T cells challenged with antigen in the presence of gp120 (15). It is well recognized that gp120/160 interacts with cell surface receptors and disturbs the T cell antigen receptor/CD3/ CD4/p56 lck transduction machinery (9,15). But this interaction cannot explain the present effect of gp160, since a similar result was obtained when TG was used in the presence of extracellular Ca 2ϩ instead of PHA. TG, as a specific inhibitor of Ca 2ϩ re-uptake into the endoplasmic reticulum, causes a large Ca 2ϩ response in lymphocytes that is independent of transduction signals.
The attenuated increase in [Ca 2ϩ ] i in the presence of gp160 appears to be due to a decreased Ca 2ϩ entry rather than to an increased Ca 2ϩ pumping out of the cytosol. TG caused the same increase in [Ca 2ϩ ] i in untreated and gp160-treated cells in the absence of bath Ca 2ϩ . In addition, Ca 2ϩ extrusion from the cytosol in the presence of TG is essentially due to the Ca 2ϩ -ATPases of the plasma membrane, since Ca 2ϩ uptake into the endoplasmic reticulum is inhibited. Under extracellular Ca 2ϩfree conditions, [Ca 2ϩ ] i returned to its basal value with the same time course in gp160-treated and untreated cells, sug-gesting that gp160 does not modify the activity of the plasma membrane Ca 2ϩ -ATPases. In contrast, the Ca 2ϩ influx, as measured when extracellular Ca 2ϩ was added back to the bath, was 25% lower in gp160-treated cells than in untreated cells. Direct measurement confirmed that incubating Jurkat cells with gp160 reduces I CRAC , the current responsible for Ca 2ϩ entry. Thus, gp160 was able to accelerate the CRAC current inactivation, resulting in I CRAC intensity decrease and limited entry of Ca 2ϩ ions.
Inhibition of I CRAC Is a PKC-mediated Process-Several signal pathways inhibit I CRAC , depending on the cell type; they are the Fc␥RII receptor in B lymphocytes (42), sphingosine and several ceramides in Jurkat T cells (43), and PKC in mast cells (44) and Jurkat T cells (39). A PKC-dependent process also seems to be involved in the reduction of I CRAC by gp160. Although herbimycin A and genistein (protein-tyrosine kinase inhibitors) had no effect on the gp160-induced reduction of I CRAC , staurosporine, an inhibitor of the classical and novel PKCs and to a lesser extent of protein-tyrosine kinases (45), restored the I CRAC of gp160-treated cells to control values. A similar result was obtained with Gö 6976, a more specific classical PKC inhibitor (37)(38). Thus, gp160 seems to decrease CRAC channel activity by increasing the PKC activity. Indeed a short incubation of untreated Jurkat cells with the phorbolester TPA, a broad spectrum activator of classic and novel PKC isozymes, mimicked the effects of gp160 and reduced the CRAC current. This agrees with a study by Haverstick et al. (39) showing that Jurkat cells incubated with PMA, another phorbol-ester PKC activator, had a reduced increase in [Ca 2ϩ ] i after activation by anti-CD3 antibody or TG.
There are at least 11 PKC isozymes with different requirements for lipid and Ca 2ϩ , and the PKC profile varies from 1 cell type to another (46). Our results mainly implicate the classical PKC␣ and -␤I into the modulation of I CRAC by gp160. First, the electrophysiological experiments showed that Gö 6976 impaired the gp160-induced decrease of I CRAC . Gö 6976 is a preferential inhibitor of the classical subgroup of PKCs (37)(38). In particular, it does not inhibit PKC␦, -⑀, and - (38). This suggests the implication of PKC␣, an abundant PKC component, or PKC␤I, a less abundant isoform in Jurkat cells (39). Second, immunoblotting experiments indicate that gp160 had profound effects on both the protein amounts and the subcellular distribution of PKC␣ and -␤I, whereas PKC␦ and -were only weakly disturbed. gp160 induced a 7-fold increase in the membrane fraction of PKC␣ (and a 3-fold increase in the Triton-insoluble fraction). This suggests that PKC␣ might mediate the effects of gp160. However, gp160 also induced a 3-fold increase in PKC␤I amount, which resulted in the enrichment of the Triton-insoluble (150%) fraction. Very recently, it has been proposed that the Triton-insoluble fraction contains lipid components of the rafts membrane microdomains (47) and, therefore, might represent membrane-bound protein. Accordingly, Nixon and McPhail (48) have recently shown that upon stimulation the levels of PKC␣ and -␤I increased in the Triton-insoluble fraction. In addition, a previous report suggests that PKC␤I is the major regulator of Ca 2ϩ influx in Jurkat cells (39). Thus it cannot be excluded that PKC␤I is also involved in the downregulation of I CRAC by gp160.
gp160 Alters the Subcellular Distribution of PKC Induced by PHA-In our experiments, the stimulation of I CRAC was induced by the depletion of internal calcium stores and, thus, was independent of PHA. Therefore the inhibitory effect of the PKC on I CRAC is not related to an alteration of the PHA transduction pathway. However, immunoblotting experiments suggest that gp160, besides inhibiting I CRAC , also modulates the PHA transduction pathway. PHA is known to induce the activation of Ca 2ϩ -dependent PKCs (49). In our hands, the addition of PHA for 30 min mainly modified the distribution of PKC␣, the main isozyme in Jurkat cells, and PKC␤I. We observed that gp160 pretreatment profoundly altered the effects of PHA on these two isozymes. After exposure to gp160, PHA was no longer able to induce PKC␣ translocation to the membrane, thus preventing the activation of this PKC. Using an antisense strategy, Lopez-Lago et al. (50) show that inhibition of PKC␣ impairs the expression of IL-2 receptor, tumor necrosis factor-␣ production, and the induction of IL-2 gene in stimulated Jurkat cells. gp160 also prevented the major effect of PHA on PKC␤I, which is the disappearance of the protein from the Triton-insoluble fraction. The meaning of this observation is not clear since Koretzky et al. (51) have previously shown that PKC␤ is not necessary to IL-2 secretion in Jurkat cells.
Conclusion-The decreases in CRAC (this study) and Kv1.3 currents (8) by HIV-1 gp160 may be considered within the context of T cell activation and proliferation. Mitogens cause an increase in Ca 2ϩ concentration in both gp160-treated and untreated cells that is sufficient to hyperpolarize the membrane via the opening of Ca 2ϩ -dependent K ϩ channels (52). However, the reduced I CRAC and its accelerated inactivation result in a lower [Ca 2ϩ ] i elevation (at about 250 nM) that cannot support the activity of Ca 2ϩ -dependent K ϩ channels (52). 2 The driving force for Ca 2ϩ is reduced as a result of the lower activity of both Kv1.3-and Ca 2ϩ -dependent K ϩ channels. In addition, we have shown that HIV-1 gp160 alters the PKC activation pattern caused by PHA. Altogether, this may impair the activation of Ca 2ϩ -binding proteins and the synthesis of IL-2 and eventually diminish T cell activation and proliferation.