Protein-tyrosine kinases activate while protein-tyrosine phosphatases inhibit L-type calcium channel activity in pituitary GH3 cells.

The aim of this study was to evaluate the effect of protein-tyrosine kinase (PTK) and protein tyrosine phosphatase (PTP) inhibitors on Ca2+ channels in GH3 cells. The activity of Ca2+ channels was monitored either by single-cell microfluorometry or by the whole-cell configuration of the patch-clamp technique. Genistein (20-200 micron) and herbimycin A (1-15 micron) inhibited [Ca2+]i rise induced either by 55 mM K+ or 10 micron Bay K 8644. In addition, genistein and lavendustin A inhibited whole-cell Ba2+ currents. By contrast, daidzein, a genistein analogue devoid of PTK inhibitory properties, did not modify Ca2+ channel activity. The inhibitory action of genistein on the [Ca2+]i increase was completely counteracted by the PTP inhibitor vanadate (100 micron). Furthermore, vanadate alone potentiated -Ca2+-i response to both 55 mM K+ and 10 micron Bay K 8644. The possibility that genistein could decrease the [Ca2+]i elevation by enhancing Ca2+ removal from the cytosol seems unlikely since genistein also reduced the increase in fura-2 fluorescence ratio induced by Ba2+, a cation that enters into the cells through Ca2+ channels but cannot be pumped out by Ca2+ extrusion mechanisms. Finally, in unstimulated GH3 cells, genistein caused a decline of [Ca2+]i and the disappearance of [Ca2+]i oscillations, whereas vanadate induced an increase of [Ca2+]i and the appearance of [Ca2+]i oscillations in otherwise non-oscillating cells. The present results suggest that in GH3 cells PTK activation causes an increase of L-type Ca2+ channel function, whereas PTPs exert an inhibitory role.

It has been largely demonstrated that the activity of L-type Ca 2ϩ channels can be regulated by different types of kinases, such as protein kinase A (PKA) 1 (1,2) and protein kinase C (PKC) (3,4). These two kinases phosphorylate serine (Ser) and threonine (Thr) residues on the ␣and ␤-subunits of these channel proteins (5,6). Recently, a great deal of interest in the literature has been devoted to another class of kinases, the protein-tyrosine kinases (PTKs) (7)(8)(9). These enzymes, which exist both in transmembrane receptor-linked (7) or non-transmembrane forms (8,9), phosphorylate tyrosine (Tyr) residues on several cellular proteins. Since it has been recently reported that in non-excitable cells such as T-lymphocytes the overexpression of PTK activity, obtained transfecting these cells with the PTK-encoding oncogene v-src, induces a remarkable increase of basal and stimulated [Ca 2ϩ ] i levels (10), it appeared of interest to explore the possibility that PTKs could modulate the activity of L-type Ca 2ϩ channels. For this purpose, the effect of the specific PTK inhibitors genistein (11,12), herbimycin A (13), and lavendustin A (14) on the function of L-type Ca 2ϩ channels was evaluated in pituitary GH 3 cells (15) by singlecell microfluorometry and patch-clamp electrophysiology.
On the other hand, since PTK activity is functionally counteracted by protein-tyrosine phosphatases (PTPs) (16,17), the possible effect of the PTP inhibitor vanadate (18) on L-type Ca 2ϩ channels was also investigated.

EXPERIMENTAL PROCEDURES
Cell Culture-GH 3 cells were obtained from Flow Laboratories (Irvine, Scotland) and grown on plastic dishes in Ham's F-10 medium (Life Technologies, Inc., San Giuliano Milanese, Italy) with 15% horse serum (Flow, Irvine, Scotland), 2.5% fetal calf serum (HyClone, Logan, UT), 100 IU of penicillin/ml, and 100 g of streptomycin/ml. Cells were cultured in a humidified 5% CO 2 atmosphere. Culture medium was changed every 2 days. For microfluorometric studies, cells were seeded on glass coverslips (Fisher) coated with poly-L-lysine (30 g/ml) (Sigma). All the experiments were performed 2-4 days after seeding. The cells were at a culture passage between 34 and 60.
Intracellular Calcium Measurements-Intracellular calcium levels were measured using a microfluorometric technique, as reported previously (19). Briefly, the cells, grown on glass coverslips, were loaded with 5 M fura-2/AM for 1 h at room temperature in Krebs-Ringer saline solution (5.5 mM KCl, 160 mM NaCl, 1.2 mM MgCl 2 , 1.5 mM CaCl 2 , 10 mM glucose, 0.2% bovine serum albumin, and 10 mM Hepes/NaOH, pH 7.4). At the end of fura-2/AM loading, the coverslip was mounted in a perfusion chamber (Medical System Co., Greenvale, NY) on an inverted Nikon Diaphot fluorescence microscope. Throughout the experiment, the cells were superfused continuously with Krebs-Ringer saline solution using a peristaltic pump (Gilson, France) and a microtube, positioned with a macromanipulator on the cells under observation (Narishige, Japan). The perfusion medium was removed continuously from the perfusion chamber by suction using a microaspirator (Medical System Co.) connected with a vacuum pump (Hoofer, San Francisco). All drugs tested were introduced into the superfusion line using an injection loop and a two-way valve (Thomson, Springfield, VA). A 100watt xenon lamp (Osram, Germany) with a computer-operated filter wheel bearing two different interference filters (340 and 380 nm) illuminated the microscopic field with uv light alternatively at the wavelength of 340 and 380 nm, with an interval of 500 ms between lighting at 340 and 380 nm. The interval between each couple of lighting and the next was chosen according to the experimental protocol. Emitted light was passed through a 400 nm dichroic mirror, filtered at 510 nm, and collected by a CCD camera (Photonic Science, Robertsbridge, East Sus-* This work was supported by Consiglio Nazionale delle Ricerche Grants 93.02003-CT14, 93.04222-CT04, and 94.02525-CT04 and Ministero dell' Università e della Ricerca Scientifica e Tecnologica (by 40 and 60% grants) (to L. A. and G. d. R.). 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.
Electrophysiological Recordings-Ba 2ϩ currents were recorded using the whole-cell configuration of the patch-clamp technique. All experiments were performed at room temperature (20 -22°C) using a List EPC7 patch-clamp amplifier (Darmstadt, Germany). The patch pipette was filled with standard internal solution (110 mM CsCl, 10 mM tetramethylammonium chloride, 2 mM MgCl 2 , 10 mM EGTA, 8 mM glucose, 10 mM Hepes, pH 7.3). Two mM ATP and 0.25 mM cAMP were added to the intracellular solution to prevent channel rundown. The cells were perfused continuously with a 10 mM Ba 2ϩ external solution (125 mM NaCl, 10 mM BaCl 2 , 1 mM MgCl 2 , 10 mM Hepes, 0.3 M tetrodotoxin, pH 7.3) both with and without the drugs to be tested. The final pipette tip resistance was 3-5 megaohms when filled with the internal solution. The currents were filtered at 5 kHz. The sampling interval was 80 s. From a holding potential of Ϫ90 mV the cells were depolarized to various potentials at a frequency of 0.2 Hz to minimize Ca 2ϩ channel rundown. The compensation of capacitative transients and leakage currents was performed both on-line by the clamp amplifier settings and off-line by subtracting Cd 2ϩ -insensitive currents (200 M Cd 2ϩ ).
Materials-All chemicals were of analytical grade and were purchased from Sigma. Genistein was obtained from BIOMOL Research Labs Inc. (Plymouth, PA). Lavendustin A, herbimycin A, Bay K 8644, daidzein, and fura-2/AM were purchased from Calbiochem. Nifedipine was a kind gift of Bayer AG (Germany).
Statistical Analysis-Data were analyzed by means of Student's t test for paired data or by analysis of variance followed by Scheffè test. Data are expressed as mean values Ϯ S.E.

Inhibition of Ca 2ϩ Channel Activity by Genistein, Herbimycin A, and Lavendustin A-In GH 3 cells genistein, which in-
hibits PTKs by competing with ATP for the binding on these enzymes (11), caused a dose-dependent (20 -200 M) inhibition of [Ca 2ϩ ] i elevation elicited by a superfusion medium containing 55 mM K ϩ (Fig. 1, A-C). The apparent IC 50 of genistein effect (30 M) on [Ca 2ϩ ] i is similar to that on PTK activity (11). Herbimycin A, another PTK inhibitor that acts on these enzymes by a completely different mechanism, namely by direct binding on its reactive SH groups (13), exerted a similar concentration-dependent inhibition on the [Ca 2ϩ ] i increase induced by 55 mM K ϩ (Fig. 1D).
On the other hand, when GH 3 cells were superfused with two 10 M consecutive pulses of the dihydropyridine activator of L-type Ca 2ϩ channels Bay K 8644 (20,21), two equivalent elevations of [Ca 2ϩ ] i occurred ( Fig. 2A). However, if genistein (200 M) was superfused 5 min before the second pulse with the L-type Ca 2ϩ channel activator, a 40% reduction of the [Ca 2ϩ ] i increase was observed (Fig. 2B).
To identify more directly the target of PTK inhibition, Ba 2ϩ currents through Ca 2ϩ channels were recorded in GH 3 cells by means of the whole-cell configuration of the patch-clamp technique. From the holding potential of Ϫ90 mV, test potentials above Ϫ60 mV elicited large inward Ba 2ϩ currents, which peaked around Ϫ35 mV (Fig. 3E). At all the test potentials the currents displayed less than 10% inactivation during the 100-ms pulse duration (Fig. 3A). These properties suggest the presence of a large population of L-type Ca 2ϩ channels. This was further confirmed by the ability of the selective L-type Ca 2ϩ channel blocker nifedipine to inhibit approximately 80% of the whole-cell Ba 2ϩ currents (Fig. 4C). Perfusing GH 3 cells with the PTK inhibitor genistein (100 M) caused a 50% reduction of the currents at all potentials tested (Fig. 3B). Complete suppression of the currents was achieved with 200 M Cd 2ϩ (Fig. 3C). Upon extensive washout (5 min) Ba 2ϩ currents recovered (Fig. 3D). The extent of genistein-induced inhibition of Ba 2ϩ currents was comparable to that observed in microfluorometric studies (Fig. 4D). Lavendustin A (25 M), another PTK inhibitor which could not be studied microfluorometrically because of its intrinsic fluorescence, also inhibited Ba 2ϩ currents (Fig. 4, A and D). By contrast, daidzein, the inactive analogue of genistein (12), did not exert any influence on Ba 2ϩ currents (Fig. 4, C and D). It should be underlined that although nifedipine inhibition of Ba 2ϩ currents occurred with a very short latency (10 s), the effect of genistein required a longer period of time (30 s) (Fig. 4E).
Effect of Genistein on the Fura-2 Fluorescence Ratio Increase Induced by Extracellular Ba 2ϩ -It is well known that Ba 2ϩ ions enter into the cells through Ca 2ϩ channels, bind fura-2, increase its fluorescence ratio, and cannot be extruded through Ca 2ϩ efflux pathways (22,23). For this reason, an increase in the fura-2 fluorescence ratio after extracellular Ba 2ϩ exposure is a specific index of cation influx through Ca 2ϩ channels. When GH 3 cells were superfused in a Ca 2ϩ -free medium, two consecutive exposures to 1 mM extracellular Ba 2ϩ caused comparable increases in the fura-2 fluorescence ratio. When genistein (200 M) was superfused 5 min before and during the second Ba 2ϩ exposure, a significant decrease in the fura-2 fluorescence ratio occurred. In fact, the S 2 /S 1 ratio was 0.99 Ϯ 0.001 in control cells and 0.73 Ϯ 0.001 in genistein-treated cells (p Ͻ 0.01).
The  The holding potential was Ϫ90 mV, and 100-ms depolarization steps from Ϫ80 to ϩ25 in 15-mV increments were delivered. The data are shown without any leak subtraction procedure. Panel E shows the current to voltage (I/V) relationship for genistein-induced inhibition of voltage-dependent Ba 2ϩ currents. Current values were taken at the end of the depolarizing steps. Each point is the mean of three different cells recorded in the same experimental conditions. The data have been normalized to the peak value of the control I/V (Ϫ35 mV) for each cell, to facilitate comparison.

FIG. 4. Comparison among the effects of genistein, lavendustin A, daidzein, and nifedipine on Ba 2؉ currents in GH 3 cells. Panel
A shows single-current traces obtained from a cell depolarized to Ϫ40 mV from a holding potential of Ϫ90 mV. As indicated, the same cell was subsequently recorded in control solution, after a 2-min exposure to 25 M lavendustin A, and after a 1-min exposure to 200 M Cd 2ϩ . Panel B shows single-current traces obtained from a cell depolarized to Ϫ30 mV from a holding potential of Ϫ90 mV. As indicated, the same cell was subsequently recorded in control solution, after a 3-min exposure to 100 M daidzein, after a 3-min exposure to 100 M genistein, and after a 1-min exposure to 200 M Cd 2ϩ . Panel C shows single-current traces obtained from a cell depolarized to Ϫ40 mV from a holding potential of Ϫ90 mV. As indicated, the same cell was subsequently recorded in control solution, after a 2-min exposure to 5 M nifedipine, and after a 1-min exposure to 200 M Cd 2ϩ . Each trace is shown without any leak subtraction procedure. In panel D is reported the percent of inhibition of the Ba 2ϩ currents at

Elicited by L-type Ca 2ϩ Channel-activating Stimuli and Reverses Genistein Inhibition of 55 mM K ϩ -induced [Ca 2ϩ ] i
Increase-When the PTP inhibitor vanadate (100 M) was superfused for 15 min before the 55 mM K ϩ pulse, a 30% increase of the [Ca 2ϩ ] i response was observed (Fig. 5A). A similar potentiation of the [Ca 2ϩ ] i response was also observed when the cells were exposed to 10 M Bay K 8644 (Fig. 5B).
In addition, the superfusion of GH 3 cells with 100 M vanadate for 15 min completely abolished the inhibition of the [Ca 2ϩ ] i response to 55 mM K ϩ which follows the exposure of these cells to 200 M genistein for 2 min (Fig. 6, A and B). 3 Cells-In unstimulated conditions, 20% (12/53) of GH 3 cells displayed oscillations of [Ca 2ϩ ] i , defined as an increase of [Ca 2ϩ ] i above the mean of the basal values ϩ 2 S.D. occurring with a frequency higher than one peak every 3 min. The remaining cells (41/53, i.e. 80%) that did not display these characteristics were defined as nonoscillating. In non-oscillating cells, the superfusion with 200 M genistein caused a 30% decline of basal [Ca 2ϩ ] i (Fig. 7A) (Fig. 7,  C and D). DISCUSSION The results of the present study, obtained by means of singlecell microfluorometry and whole-cell patch-clamp techniques, demonstrate that the activity of Ca 2ϩ channels in GH 3 cells can be influenced by the interplay between PTK and PTP activity: PTK activation seems to cause an increase, whereas PTP activation appears to exert an inhibitory role on this ion channel.

Effect of the PTK Inhibitor Genistein and of the PTP Inhibitor Vanadate on Basal [Ca 2ϩ ] i in Resting GH
The hypothesis that the L-type Ca 2ϩ channel is the target of PTK and PTP modulation derives from the results showing that the increase of [Ca 2ϩ ] i elicited by the specific L-type Ca 2ϩ channel activator Bay K 8644 and high K ϩ concentrations was reduced by the PTK inhibitor genistein and enhanced by the PTP blocker vanadate. A further support to this idea is the ability of genistein and lavendustin A to inhibit Ba 2ϩ currents through Ca 2ϩ channels that displayed biophysical and pharmacological features of the L-type. On the other hand, the possibility that the action of PTK inhibitors is exerted on the T-type Ca 2ϩ channels, which have been described in GH 3 cells, seems unlikely since this Ca 2ϩ channel type does not play a significant role in the [Ca 2ϩ ] i elevation elicited by strong activating stimuli (55 mM K ϩ or Bay K 8644) (24,25). In addition, the biophysical features of Ba 2ϩ currents recorded in GH 3 cells in the present study do not show the presence of a significant population of this Ca 2ϩ channel type. Furthermore, the remarkable inhibition of Ba 2ϩ currents by the L-type blocker nifedipine suggests that the largest population of Ca 2ϩ channels is represented by the L-type.
The possibility that the genistein-induced reduction of the [Ca 2ϩ ] i increase elicited by high K ϩ concentrations could be due to an increase of Ca 2ϩ removal from the cytoplasm to the extracellular space or into the intracellular Ca 2ϩ stores seems not to be compatible with the results of the present study. In fact, genistein also reduced the entrance of Ba 2ϩ ions, a cation that is known to be unable to substitute for Ca 2ϩ in the extrusion mechanisms. In support of this interpretation, the entity of the genistein-induced inhibition of the [Ca 2ϩ ] i rise induced by 55 mM K ϩ and 10 M Bay K 8644 was comparable to the inhibition observed in electrophysiological experiments.
Since it has been reported that genistein, besides inhibiting PTKs, can also block other protein kinases such as PKA and PKC (11,12), which are known to modulate L-type Ca 2ϩ channels (1)(2)(3)(4), the possibility exists that its effects on the activity of L-type Ca 2ϩ channels could occur via PKA or PKC inhibition. However, this hypothesis seems unlikely since herbimycin A and lavendustin A, two other specific PTK inhibitors devoid of PKA or PKC inhibitory action (14,26) and structurally unrelated to genistein, effectively inhibited Ca 2ϩ channel activity in GH 3 cells. This evidence strongly suggests that PKA or PKC inhibition is not involved in the genistein action on Ca 2ϩ channels. In addition, the IC 50 for genistein inhibition of Ca 2ϩ channels (30 M) was very similar to that for PTK inhibition and much lower than that for PKA and PKC blockade (11). The specificity of genistein action on Ca 2ϩ channels via PTKs was confirmed further by the inability of the genistein analogue daidzein, which lacks PTK inhibitory properties, to modify Ca 2ϩ channel activity in electrophysiological recordings.
The existence of a PTK regulation of L-type Ca 2ϩ channels in GH 3 cells is also supported by the fact that PTPs, which physiologically counteract the activity of PTKs (16,17), exert an opposite modulation on L-type Ca 2ϩ channel activity. In fact, the inhibition of PTPs by orthovanadate, a well known inhibitor of these enzymes (18), was able to enhance the [Ca 2ϩ ] i increase induced by high K ϩ concentrations and to counteract the inhibitory effect of genistein on this response.
The modulation exerted by PTKs and PTPs seems to occur not only when L-type Ca 2ϩ channels are activated by high depolarizing stimuli, but also in resting conditions. In fact, the inhibition of PTKs by genistein caused a decline of [Ca 2ϩ ] i and a disappearance of [Ca 2ϩ ] i oscillations in oscillating GH 3 cells, whereas the blockade of PTPs by vanadate induced an increase of [Ca 2ϩ ] i or the appearance of [Ca 2ϩ ] i oscillations. These findings were not unexpected since in unstimulated conditions, L-type Ca 2ϩ channels of GH 3 cells are spontaneously active, as shown by the fact that spontaneous action potentials have been detected (27) and that these potentials are coupled to oscillations of [Ca 2ϩ ] i , which can be abolished by the specific L-type Ca 2ϩ channel blocker nifedipine (15).
The results of the present study showing that PTKs exert a stimulatory modulation on L-type Ca 2ϩ channels are in line with the recent report that genistein induces a concentrationdependent inhibition of Ca 2ϩ channel currents in vascular smooth muscle cells (28). In addition, evidence has been provided that the inhibition of PTKs can also reduce Ca 2ϩ influx through plasma membrane "refilling" channels (29 -31) and that different types of receptor-operated channels, like the nicotinic, N-methyl-D-aspartic acid, and ␥-aminobutyric acid receptor channels, can be modulated by PTKs (32)(33)(34).
The results of the present study could be of interest to explain the Ca 2ϩ dependence of certain biological responses elicited by some growth factors (35). In fact, the stimulation of many growth factor receptors, such as those for the epidermal growth factor, recognize as a signaling pathway the activation of a receptor-linked PTK (7). Since the results of the present study indicated that PTK activation leads to Ca 2ϩ entrance through L-type Ca 2ϩ channels into the cells, the Ca 2ϩ -dependent epidermal growth factor-induced differentiation of GH 3 cells toward the lactotroph phenotype (36) could be the consequence of the activation of L-type Ca 2ϩ channels, especially if one considers that in a different pituitary cell line, epidermal growth factor induces an increase of [Ca 2ϩ ] i which is independent of phospholipase C␥1-dependent inositol 1,4,5-trisphosphate generation (37). In conclusion, all of these results suggest that L-type Ca 2ϩ channels are modulated by the PTK/PTP system in GH 3 cells. The molecular mechanism of this modulation remains to be clarified. However, a possible working hypothesis to explain the effect of PTK inhibitors on Ca 2ϩ channel function could be that phosphorylation by PTKs exerts a permissive role on the activation of Ca 2ϩ channels elicited by both the dihydropyridine agonist Bay K 8644 and depolarizing stimuli. Such a model has already been proposed by Armstrong et al. (2) to explain the effect of PKA on Ca 2ϩ channel activation.