Phosphatase Inhibition Reveals a Calcium Entry Pathway Dependent on Protein Kinase A in Thyroid FRTL-5 Cells

Calcium entry through store-operated calcium channels is an important entry mechanism. In the present report we have described a novel calcium entry pathway that is independent of depletion of intracellular calcium stores. Treatment of the cells with the phosphatase inhibitor calyculin A (caly A), which blocked thapsigargin-evoked store-operated calcium entry (SOCE), induced a potent concentration-dependent calcium entry. In a calcium-free buffer, acute addition of caly A evoked a very modest increase in cytosolic free calcium ([Ca2+]i). This increase was not from the agonist-mobilizable calcium stores, as the thapsigargin-evoked increase in [Ca2+]i was unaltered in caly A-treated cells. The caly A-evoked calcium entry was not blocked by Gd3+ or 2-APB, whereas SOCE was. Caly A enhanced the entry of barium, indicating that the increase in intracellular calcium was not the result of a decreased extrusion of calcium from the cytosol. Jasplakinolide and cytochalasin D had only marginal effects on calcium entry. The protein kinase A (PKA) inhibitor H-89 and an inhibitory peptide for PKA abolished the caly A-evoked entry of both calcium and barium. The SOCE was, however, enhanced in cells treated with H-89. In cells grown in the absence of thyrotropin (TSH), the caly A-evoked entry of calcium was smaller compared with cells grown in TSH-containing buffer. Stimulation of cells grown without TSH with forskolin or TSH restored the calyculin A-evoked calcium entry to that seen in cells grown in TSH-containing buffer. SOCE was decreased in these cells. Our results thus suggest that TSH, through the production of cAMP and activation of PKA, regulates a calcium entry pathway in thyroid cells. The pathway is distinctly different from the SOCE. As TSH is the main regulator of thyroid cells, we suggest that the novel calcium entry pathway participates in the regulation of basal calcium levels in thyroid cells.

Calcium is the key regulator of a multitude of cellular processes, including proliferation, gene transcription, and apoptosis (1). To enable such a broad spectrum of different actions, the cells have evolved a plethora of different mechanisms regulating cellular calcium levels. Calcium may be mobilized from intracellular compartments by the activation of inositol 1,4,5trisphosphate (IP 3 ) 1 receptors or ryanodine receptors. Furthermore, calcium may enter the cells through a multitude of calcium channels, both voltage-dependent and -independent channels. Ca 2ϩ ATPases in both the plasma membrane and in intracellular membranes participate in regulation by transporting calcium out of the cell or into intracellular compartments.
In non-excitable cells, calcium entry is usually evoked as a result of agonist-mediated activation of the phospholipase C-IP 3 pathway, mobilization of sequestered calcium from intracellular stores, and, as a consequence of this depletion, storeoperated calcium entry (SOCE). Calcium entry can also be the result of agonist-evoked activation of receptor channels (i.e. the P2X receptors for ATP) or the activation of non-selective cation channels (i.e. second-messenger channels and stretch-activated channels). Of these different calcium entry pathways, the activation of store-operated calcium channels has attracted substantial interest (2,3). In addition, several investigations show that agonist-evoked calcium entry (e.g. as seen in calcium oscillations) is mediated through a distinct, arachidonic acidregulated calcium channel (4,5).
The SOCE, also called the capacitative calcium entry, is the direct result of depletion of intracellular calcium stores. The emptying of the stores evoked a profound entry of calcium into the cytosol from the extracellular space, through mechanisms not yet fully understood. Three theories are presently at hand. First, the emptying of intracellular stores produced a factor that diffuses to the plasma membrane and opens calcium channels. Second, emptying of the calcium stores causes the fusion of vesicles containing calcium channels with the plasma membrane. In the last model, the depletion of calcium stores causes a conformational coupling of the IP 3 receptor with calcium channels (2,3). Very convincing evidence has been obtained in support of the conformational coupling theory, e.g. the disruption of the actin cytoskeleton potently abrogates store-operated calcium entry (6,7). Thus, the disruption and rearrangement of the cytoskeleton would physically hinder a coupling between the IP 3 receptor and calcium channels located on the plasma membrane. However, evidence obtained in several investigations has provided support for all three mechanisms. The situation is not made easier to fathom considering the wealth of different types of calcium channels (in particular the TRP family of channels) (8) that have been suggested to mediate, at least in part, store-operated calcium entry.
In thyroid FRTL-5 cells, changes in intracellular calcium control several processes, including the regulation of iodide efflux (9 -11) and the regulation of proliferation and synthesis of DNA (12,13). Furthermore, changes in intracellular calcium may modify the TSH-evoked effects in thyroid cells (9,14). In addition to agonist-evoked changes in [Ca 2ϩ ] i (15), the SOCE pathway is probably an important regulator of intracellular calcium signaling (16). In preliminary experiments aimed at further understanding the mechanisms regulating calcium entry, we observed that the phosphatase inhibitor calyculin A, a potent disrupter of the actin cytoskeleton, abolished SOCE in our cells. However, in those experiments we observed that the resting calcium levels were higher compared with that seen in control cells. Further investigations revealed that calyculin A evoked a substantial calcium entry that was distinct from the SOCE. Furthermore, the calyculin A-evoked calcium entry appeared to be crucially dependent on cAMP and PKA. As the cAMP-PKA pathway is a major regulator of thyroid cell function, we suggest that the calcium entry mechanism we have described in the present investigation is of importance in regulating calcium levels in thyroid cells. Cell Culture-Rat thyroid FRTL-5 cells, originally obtained from the Interthyr Foundation (Bethesda, MD), were grown in Coon's modified Ham's F-12 medium supplemented with 5% calf serum and six hormones (6H) (17) (insulin, 10 g/ml; transferrin, 5 g/ml; hydrocortisone, 10 nM; tripeptide Gly-L-His-L-Lys, 10 ng/ml; TSH, 0.3 milliunits/ml; somatostatin, 10 ng/ml) in a water-saturated atmosphere of 5% CO 2 and 95% air at 37°C. Prior to an experiment, cells from one donor culture dish were harvested with a 0.2% trypsin solution and plated onto plastic 100-mm culture dishes. The cells were the grown for 7-8 days, incorporating 2-3 changes of culture medium with the final medium change 24 h prior to an experiment. In some experiments the cells were grown in 5H medium (i.e. medium lacking TSH) for 5 days prior to an experiment.

Materials
Measurement of [Ca 2ϩ ] i -The medium was aspirated, and the cells were harvested with HEPES-buffered saline solution (HBSS, in millimolar concentrations: NaCl, 118; KCl, 4.6; glucose, 10; CaCl 2 , 1.0; HEPES, 20, pH 7.4) lacking Ca 2ϩ but containing 0.02% EDTA and 0.1% trypsin. After washing the cells three times by pelleting, the cells were incubated with 1 M Fura-2/AM for 30 min at 37°C. Following the loading period, the cells were washed twice with HBSS buffer, incubated for at least 15 min at room temperature, and washed once more. Fluorescence was measured with a Hitachi F2000 fluorimeter using excitation wavelengths of 340 and 380 nm and detecting emission at 510 nm. The signal was calibrated by the addition of 1 mM CaCl 2 and Triton X-100 to obtain maximum fluorescence. The chelation of extracellular Ca 2ϩ with 5 mM EGTA and the addition of Tris base to elevate the pH above 8.3 were both used to obtain minimum fluorescence.
[Ca 2ϩ ] i was calculated as described by Grynkiewicz et al. (18) , using a computer program designed for the fluorimeter with a K d value of 224 nM for Fura 2.
For the experiments, calyculin A and the used inhibitors were applied as a dimethyl sulfoxide solution. The concentration of the solute never exceeded 0.2%, and in the control experiments the solvent alone was used.
Calcium Imaging Experiments-Cells from one 100-mm cell culture dish were harvested using 0.02% EDTA-0.1% trypsin solution and plated onto polylysine-coated round coverslips. After 2-4 days of culture, the cells were incubated with Fura-2/AM (final concentration 4 M) for 30 min at room temperature. Following the loading period, the cells were incubated for at least 20 min to ensure complete cleavage of the AM group from Fura-2. After washing the cells once, the coverslip was mounted on an Axiovert 35 inverted microscope with a Fluor ϫ40 objective and perfused with HBSS. The excitation filters were set at 340 and 380 nm, and the excitation light was obtained using an XBO 75W/2 xenon lamp. Emission was measured at 510 nm. The shutter was controlled by a Lambda 10 -2 control device (Sutter Instruments, Novato, CA). Fluorescence images were collected with a SensiCam 12BIT CCD camera (PCO/CD Imaging, Kelheim, Germany). The experiments were performed at room temperature. Images were taken every 2.6 s to avoid bleaching. The system was calibrated using Ca 2ϩ -free HBSS containing Fura-2 (potassium salt Confocal Microscopy of Actin Filaments-The cells were grown as described above. Cells from a 100-mm culture dish were harvested as described for the calcium measurements and plated onto poly(D)lysinecoated coverslips. The cells were cultured for 48 h with a change of culture medium after 24 h. The cells were stimulated with 100 nM calyculin A for 15 and 30 min. Control cells were treated with vehicle only. The coverslips were then washed twice with warm (ϩ37°C) PBS. The fixation of the cells was made by incubating the coverslips for 10 min in 3.7% formaldehyde at room temperature, after which the cells were washed twice with PBS. The cells were permeabilized for 3-5 min in 0.1% Triton X-100 in PBS and washed twice with warm PBS.
Prior to the staining, the cells were incubated with PBS containing 1% bovine serum albumin for 30 min at room temperature. The fluorescein isothiocyanate-conjugated phalloidin (300 units; Molecular Probes) was diluted 1:40 with PBS-bovine serum albumin, and the cells were incubated with the solution for 20 min at room temperature. The cells were then washed twice with warm PBS.
The cells were examined using a Leica TCS SP confocal microscope equipped with an Argon-Krypton laser (Omnichrome, Melles Griot, Carlsbad, CA). The excitation wavelength was 488 nm, and the emission wavelength was 500 -540 nm. Pictures were taken using a PL APO ϫ40/1.25-0.75 oil objective. All figures were acquired with Leica TCS NT software.
Measurement of Cellular cAMP Levels-The cells were harvested as described for the calcium experiments and were preincubated in HBSS for 20 min in the presence of 0.3 mM isobutylmethylxantin (to inhibit phosphodiesterase activity). Then aliquots of cells (0.5 ϫ 10 6 cells/tube) were stimulated with 100 nM calyculin A. After 15 min the reaction was stopped with perchloric acid to a final concentration of 0.5 M. The samples were neutralized by KOH, and the cAMP concentrations were determined using a protein binding method (19).
Electrophysiological Measurements of the E M -The studies were performed using the patch clamp whole-cell technique in voltage and current clamp mode (20). Prior to the experiments, cells were harvested with 0.02% EDTA-trypsin solution and subcultured on coverslips on 24-well plates (BD PharMingen) for 2-5 days. The coverslips were placed in a perfusion chamber. During recordings the cells were continuously perfused with a standard solution containing (in mM): NaCl, 150; KCl, 5.4; MgCl 2 , 1; CaCl 2 , 1.8; HEPES, 5 (pH adjusted to 7.4 with NaOH) at 0.5 ml/min. Calyculin A (100 nM) was added to the superfusate from a stock solution of 100 M.
Patch pipettes were made from GC150TF glass micropipettes (Harvard Apparatus, Kent, UK) and had a resistance of 2-5 M⍀ (when filled with an internal solution containing (in mM): KCl, 150; MgCl 2 , 2; BAPTA, 5; HEPES, 10 (pH was adjusted to 7.2 with KOH). All recordings were made at room temperature. Currents were recorded with an EPC-9 amplifier (HEKA, Lambrecht, Germany). Analyses of the recordings were made using Pulse and Pulse Fit software (HEKA).
Statistics-The results are expressed as the mean Ϯ S.E. Statistical analysis was made using Student's t test for paired observations. When three or more means were tested, analysis of variance was used.

Calcium Entry Mechanisms in FRTL-5 Cells-Previous stud-
ies have shown that FRTL-5 cells respond to depletion of intracellular calcium stores with a substantial entry of calcium (16). In the present study we have shown that the thapsigargin-evoked depletion of calcium stores results in entry of exog-enous calcium and that this entry can be blocked by 30 M 2-APB, 1 M Gd 3ϩ and by pretreatment of the cells with 100 nM of the phosphatase inhibitor calyculin A (Fig. 1, A and B).
We observed that the basal calcium level of cells treated with calyculin A always was higher than in control cells (218 Ϯ 5 nM and 126 Ϯ 5 in calyculin-treated and control cells, respectively, n ϭ 6, p Ͻ0.05, see Fig. 1A). To investigate the reason for this, we stimulated FRTL-5 cells with calyculin A and observed a distinct increase in [Ca 2ϩ ] i ( Fig. 2A). This increase was not blocked by either 30 M 2-APB or 1 M Gd 3ϩ (Fig. 2B). The results thus suggest that, in FRTL-5 cells, calyculin A evokes an increase in [Ca 2ϩ ] i that does not occur through store-operated calcium channels. Calyculin A per se evoked a very modest increase in [Ca 2ϩ ] i in cells in a calcium-free buffer (22 Ϯ 5 nM, n ϭ 4). However, calyculin A did not mobilize calcium from IP 3 -dependent stores, as the thapsigargin-evoked release of sequestered calcium in a calcium-free buffer was 151 Ϯ 27 nM in calyculin A-treated cells and 148 Ϯ 9 nM in control cells, respectively.
To further investigate the effect of calyculin A, we pretreated the cells with calyculin A and suspended the cells in a calciumfree buffer. Then calcium (final concentration 1 mM) was added back to the cells. In these experiments, we observed an increase in [Ca 2ϩ ] i that was dependent on the concentration of calyculin A (Fig. 2, C and D). The increase in [Ca 2ϩ ] i was insensitive to both 2-APB and Gd 3ϩ (Fig. 2E). In contrast to this observation, if FRTL-5 cells were stimulated with 1 M thapsigargin and calcium was then readded, the calcium entry was potently abrogated by 2-APB and Gd 3ϩ (Fig. 3A). However, if cells treated with 100 nM calyculin A were stimulated with 1 M thapsigargin in a calcium-free buffer and calcium was readded, we always obtained a substantial entry of calcium. The initial calcium entry was indistinguishable from that seen in cells stimulated with thapsigargin only (867 Ϯ 91 nM and 817 Ϯ 69 nM in thapsigargin-and calyculin A-treated cells, respectively;  Fig. 3B). The increase in [Ca 2ϩ ] i probably is a combination of the calyculin-evoked calcium entry and a small residual store-operated calcium entry due to the thapsigarginevoked depletion of the endoplasmic reticulum calcium stores. A comparison of the increase in [Ca 2ϩ ] i obtained after addition of calcium to cells treated with calyculin only, or with both calyculin and thapsigargin, is shown in Fig. 3D.
To investigate the effect of calyculin A on the actin cytoskeleton in our cells, we labeled control cells and cells treated with 100 nM calyculin with fluorescein isothiocyanate-conjugated phalloidin. As can be seen in Fig. 4, treatment of the cells resulted in a dramatic reconstruction of the actin filaments in the cells. The effect of calyculin A in our cells was very similar to that observed in other cell types (see Refs. 6 and 7) and thus probably explains why the thapsigargin-evoked SOCE was attenuated in our cells after treatment with calyculin A.
To ensure that calyculin A actually enhanced calcium entry and not increased [Ca 2ϩ ] i by blocking plasma membrane Ca 2ϩ ATPases and calcium extrusion, we measured the entry of Ba 2ϩ in calyculin A-stimulated cells. In these experiments pretreatment with calyculin A clearly enhanced Ba 2ϩ entry (Fig. 5A). For comparison, the thapsigargin-evoked Ba 2ϩ entry is shown (Fig. 5B). The results in Fig. 5 show that the calyculin A-evoked Ba 2ϩ entry (49 Ϯ 3 fluorescense units/30 s) is more pronounced than the thapsigargin-evoked Ba 2ϩ entry (31 Ϯ 4 fluorescense units/30 s; p Ͻ0.05). Furthermore, both calyculin A and thapsigargin enhanced the entry of Sr 2ϩ (results not shown).
Mechanisms of Calyculin A-evoked Calcium Entry-Calyculin A could theoretically enhance calcium entry by hyperpolarizing the membrane potential, thus increasing the electrochemical driving force for calcium. However, in patch clamp experiments in the current clamp mode, we could not show an effect of calyculin A on the membrane potential compared with vehicle-treated cells. The change in E M in control cells was Ϫ5.6 Ϯ 3.6 mV, compared with Ϫ7.7 Ϯ 0.8 mV in cells treated with 100 nM calyculin A as measured over a time span of 6 min.
Calyculin A potently redistributes the actin filaments of the cytoskeleton. To test whether the effects of calyculin A were the result of this redistribution, we treated our cells with either We also tested whether cyclosporin, an inhibitor of the phosphatase calcineurin, could mimic the effect of calyculin A. However, incubating the cells with cyclosporin (1 g/ml for 15 min) did not enhance calcium entry (data not shown).
We have previously shown that protein kinase C can potently modulate calcium entry in FRTL-5 cells (16). Pretreating the cells with either calphostin C (100 nM) or GF109203X (10 M) for 15 min did not attenuate the calyculin A-evoked calcium entry (data not shown). In addition, stimulating calyculin Apretreated cells with 100 nM phorbol 12-myristate 13-acetate did not increase the calcium entry compared with control cells treated with calyculin A only (data not shown). However, when the cells were pretreated with the PKA inhibitor H-89 (10 M for 15 min) or with PKI 14 -22 (10 M for 30 min), both the calyculin A-evoked calcium and barium entry were abolished (Fig. 6, A-C). In sharp contrast to this, we observed that the thapsigargin-evoked calcium entry was enhanced in the presence of 10 M H-89 (Fig. 6D).
To further test the hypothesis that PKA was involved in regulating the calyculin A-evoked calcium entry, we cultured our cells in medium lacking TSH. The calyculin A-evoked calcium entry was significantly reduced in the cells ( increased compared with TSH-depleted control cells (Fig. 7, B  and C). Thus, both forskolin and TSH restored the calyculin A-evoked calcium entry to the same level as seen in cells grown in the presence of 0.3 milliunits of TSH.
To investigate whether a direct activation of PKA without prior treatment with calyculin A could enhance calcium entry in TSH-depleted cells, we stimulated cells with 10 M forskolin and calcium. As can be seen in Fig. 7D, forskolin evoked an entry of calcium (150 Ϯ 7 nM) that was significantly smaller in cells pretreated with 10 M H-89 (109 Ϯ 2 nM, p Ͻ0.05).
We next investigated whether calyculin A increased cellular cAMP levels in FRTL-5 cells. The concentration of cAMP in cells stimulated with 100 nM calyculin A for 15 min was 4.7 Ϯ 0.5 pg/0.5 ϫ 10 6 cells (n ϭ 4), compared with 6.3 Ϯ 1.0 pg/0.5 ϫ 10 6 (n ϭ 4, p Ͼ0.05) in vehicle-treated control cells. Thus, the effect of calyculin A was downstream from the production of cAMP in the cells, suggesting that the effect of calyculin A was the result of an enhanced phosphorylation by PKA.
For comparison, the thapsigargin-evoked increase in [Ca 2ϩ ] i was also measured in TSH-depleted cells. Our results show that the thapsigargin-evoked SOCE was enhanced in cells grown in the absence of TSH compared with cells stimulated with 0.3 mM TSH (Fig. 7, E and F). Thus, an increase in cAMP has an antagonizing effect on SOCE in FRTL-5 cells.
We also wanted to investigate whether calcium entry evoked by G protein-coupled receptors could be enhanced by treatment with calyculin A. When cells were pretreated with 100 nM calyculin A for 15 min, the calcium signal evoked by 100 M ATP was clearly broadened (Fig. 8). The decrease in [Ca 2ϩ ] i was 36 Ϯ 5 nM/10 s at 75% of the downward part of the calcium transient, whereas the value was 65 Ϯ 10 nM/10 s in control cells (p Ͻ0.05). Similar results were obtained in cells pretreated with pertussis toxin (50 ng/ml for 24 h, results not shown). Furthermore, H-89 attenuated the ATP-evoked calcium signal in calyculin A-treated cells (decrease in [Ca 2ϩ ] i was Ͼ Ͼ100 nM/10 s). All experiments were performed in the presence of 1 M Gd 3ϩ to block SOCE. Next we tested whether the ATP-evoked calcium signal in cells not treated with calyculin A was composed of a component dependent on PKA. This seems to be the case, as PKI 14 -22 altered the ATP-evoked calcium spike (decrease in [Ca 2ϩ ] i was 114 Ϯ 19 nM/10 s, p Ͻ0.05; Fig. 8). Thus, the novel calcium entry pathway also seems to be involved in calcium signaling through G protein-coupled receptors.
Effect of Calyculin A on [Ca 2ϩ ] i in Single Cells-As the above experiments were performed with cells in suspension, we wanted to investigate whether calyculin A also enhanced calcium entry in single cells. As can be seen in Fig. 9, pretreatment of single cells with 100 nM calyculin A for 15 min enhanced calcium entry compared with the entry obtained in control cells. The effect of calyculin A was not altered by the presence of 1 M Gd 3ϩ , whereas the thapsigargin-evoked calcium entry was decreased (data not shown) in a manner similar to that observed in cells in suspension. DISCUSSION In the present investigation we have shown that, in addition to store-operated calcium entry (16), thyroid FRTL-5 cells also have a previously unknown calcium entry pathway. This calcium entry pathway seems to depend on the activation of PKA. The novel entry pathway is distinctly different from the SOCE pathway. We have based our conclusion on the following observations. Store-operated calcium entry was almost totally blocked by preincubating the cells with calyculin A, whereas the novel calcium entry pathway was revealed by inhibition of phosphatases with calyculin A. The increase in [Ca 2ϩ ] i evoked by calyculin A was the result of calcium entry, not a decreased extrusion of calcium, as calyculin A potently enhanced the entry of Ba 2ϩ , a poor substrate for the plasma membrane Ca 2ϩ ATPase. The store-operated Ba 2ϩ entry was of smaller magnitude in comparison with the calyculin-evoked Ba 2ϩ entry. SOCE was abrogated by 2-APB and a low concentration of Gd 3ϩ , whereas the calyculin A-evoked entry was not. Finally, the PKA inhibitors H-89 and PKI 14 -22 blocked the calyculin A-evoked calcium entry, whereas H-89 enhanced SOCE. In addition, in cells grown without TSH, the calyculin A-evoked calcium entry was enhanced by the addition of TSH, whereas TSH decreased the SOCE.
The entry of calcium upon depletion of calcium stores has been explained by conformational coupling of the IP 3 receptor with plasma membrane calcium channels. Evidence suggests that these channels belong to the TRP family of channel proteins (possibly the TRPC3, Refs. 22,23). Compelling evidence for conformational coupling has been obtained in experiments where calyculin A-evoked formation of cortical actin layers inhibited calcium entry (6,24,25). However, this effect was considered mainly a function of the reorganization of the actin cytoskeleton, as jasplakinolide also inhibited store-operated calcium entry. Experiments performed with cytochalasin D, another agent forming cortical actin, resulted in similar results. The present calcium entry pathway, on the other hand, is not attenuated after calyculin A-evoked cortical actin formation. Thus, our results strongly point at an effect mediated by an inhibition of a phosphatase, as treatment of the cells with jasplakinolide or cytochalasin D did not evoke a response similar to that obtained by calyculin A. Furthermore, the calcium entry was not the result of store depletion, as calyculin A did not modify agonist-evoked store depletion. Thus, the present calcium entry is probably not the result of a coupling of intracellular stores to calcium channels in the plasma membrane. We suggest that the inhibition of a phosphatase uncouples a block of a calcium entry pathway.
Previous studies have shown that a phosphatase is involved in regulating SOCE. The very first studies on SOCE showed that emptying of intracellular stores evoked a highly calciumselective current (calcium release-activated current, I CRAC ) (26,27). The activation of this current involved a phosphatase (27). In another study it was shown that the effect of a presently unknown calcium influx factor was enhanced by inhibition of phosphatases by okadaic acid (28). Other investigations have shown that calyculin A or methavanadate enhances a SOCElike current in outside-out patches in rabbit portal vein smooth muscle cells, a current similar to that evoked by phorbol esters or diacylglycerol (29,30). This calcium entry is probably mediated by a member of the TRP family of channel proteins as diacylglycerol has been shown to directly activate TRPC3 and TRPC6 channels (31). Furthermore, inhibition of phosphatases enhanced protein kinase C-evoked activation of SOCE in inside-out patches (29). In smooth muscle cells, inhibition of the phosphatase calcineurin prevented the calcium-evoked inhibition of L-type voltage-operated calcium channels (VOCCs) cells (32), and inhibition of phosphatases increased VOCC currents in pancreatic ␤-cells (33). Phosphatases thus seem to have an important role in regulating ion channels and are often excised together with ion channels (34). However, in a study by Lalevée et al. (35), pretreatment of cells with calyculin A was without any effects on either angiotensin II-or high potassium-evoked calcium signals.
The calcium entry mechanism described is probably also distinct from the arachidonic acid-evoked calcium entry pathway. Arachidonic acid-evoked calcium entry has been implicated in regulating calcium entry during calcium oscillations evoked by low concentrations of agonists (36). Interestingly, Gd 3ϩ or 2-APB did not block the arachidonic acid-evoked calcium entry (37). In FRTL-5 cells, arachidonic acid evoked calcium entry (38), but the magnitude was modest compared with that seen in the present investigation. In addition, arachidonic acid depleted intracellular calcium stores (38). Presently no information exists regarding the sensitivity of the arachidonic acid-evoked calcium entry to calyculin A. However, the arachidonic acid-evoked current was inhibited by calcineurin (39), whereas we were unable to evoke any calcium entry by treating the cells with cyclosporin A. Trebak et al. (40) have shown that low concentrations of Gd 3ϩ did not block receptor-activated HTRPC and 2-APB had only a weak effect. Thus, we cannot exclude that treatment of our cells with calyculin A reveals calcium entry through a calcium channel of the TRP family of channels. Furthermore, we cannot exclude the possibility that some (e.g. autocrine) form of receptor-mediated activation of calcium entry may occur after treatment with calyculin A. This is unlikely, as previous studies in FRTL-5 cells have shown that cAMP blocked the coupling of G protein-coupled receptors to at least phospolipase C (21). We did, however, observe that activation of G protein-coupled receptors evoked calcium signals that, in part, were mediated by the novel entry mechanism. These calcium signals could further be enhanced by calyculin A and were blocked by both H-89 and PKI 14 -22 . Our results thus point at a rather complex calcium signaling pattern in response to agonists activating G protein-coupled receptors as ATP also evokes SOCE in FRTL-5 cells (15).
A very interesting observation was that the calcium entry appeared to be dependent on the cAMP-PKA pathway: if PKA were blocked by H-89 or PKI 14 -22 , the calyculin A-evoked calcium entry was blocked. Furthermore, in cells grown in the absence of TSH, the calyculin A-evoked calcium entry was of lower magnitude compared with cells grown in the presence of TSH. Addition of TSH or forskolin to cells grown in TSH-free medium rapidly enhanced calcium entry in cells treated with calyculin, suggesting that phosphorylation of a channel protein (or an accessory protein) by PKA is involved in regulating calcium entry. In sharp contrast to this observation, we showed that the thapsigargin-evoked calcium entry was attenuated by cAMP in cells grown in TSH-free medium. A similar cAMPevoked inhibition of store-operated calcium entry has been shown in platelets (7). The TSH-cAMP-PKA pathway is one of the most important regulators of thyroid cell function. It is tempting to suggest that the calcium entry mechanism revealed in the present investigation participates in the regulation of basal calcium levels in thyroid cells. This regulation probably is under strict control of PKA and a presently unknown phosphatase. Furthermore, the entry pathway is clearly distinct from store-operated calcium entry. Apparently G protein-coupled receptors also may utilize this calcium entry pathway. Further investigations will reveal whether a complex interplay exists between store-operated calcium entry and the novel cAMP-dependent calcium entry.