Dependence of Soluble Guanylyl Cyclase Activity on Calcium Signaling in Pituitary Cells*

The role of nitric oxide (NO) in the stimulation of soluble guanylyl cyclase (sGC) is well established, but the mechanism by which the enzyme is inactivated during the prolonged NO stimulation has not been characterized. In this paper we studied the interactions between NO and intracellular Ca2+ in the control of sGC in rat anterior pituitary cells. Experiments were done in cultured cells, which expressed neuronal and endothelial NO synthases, and in cells with elevated NO levels induced by the expression of inducible NO synthase and by the addition of several NO donors. Basal sGC-dependent cGMP production was stimulated by the increase in NO levels in a time-dependent manner. In contrast, depolarization of cells by high K+ and Bay K 8644, an L-type Ca2+ channel agonist, inhibited sGC activity. Depolarization-induced down-regulation of sGC activity was also observed in cells with inhibited cGMP-dependent phosphodiesterases but not in cells bathed in Ca2+-deficient medium. This inhibition was independent from the pattern of Ca2+ signaling (oscillatoryversus nonoscillatory) and NO levels, and was determined by averaged concentration of intracellular Ca2+. These results indicate that inactivation of sGC by intracellular Ca2+serves as a negative feedback to break the stimulatory action of NO on enzyme activity in intact pituitary cells.

proteins that is involved in a broad array of cellular functions in cardiovascular, neuronal, neuroendocrine, and other cell types (2). In this respect, sGC-generated cGMP production allows this enzyme to transmit the NO signal to the downstream elements of the signaling cascade, such as cGMP-dependent protein kinase (3), cyclic nucleotide-gated channels (CNGs) (4), and cGMP-regulated phosphodiesterases (5).
In contrast to the well established role of NO in the activation of sGC, the mechanisms of inactivation for this enzyme have been incompletely characterized. This is especially important for cells in which the highly diffusible NO is intimately involved in the control of Ca 2ϩ signaling by facilitating Ca 2ϩ release and/or Ca 2ϩ influx pathways, as in hepatocytes (6,7), glial cells (8), and neuroendocrine cells (9). Because the elevated intracellular calcium concentration ([Ca 2ϩ ] i ) stimulates the activity of two NO-producing enzymes, neuronal NO synthase (nNOS) and endothelial NOS (eNOS), such up-regulation would lead to overloading the cells with Ca 2ϩ . Thus, it is reasonable to propose that the negative feedback effect of Ca 2ϩ on sGC would provide the necessary mechanism to coordinate the regulation of intracellular cGMP and Ca 2ϩ concentrations.
To test this hypothesis, we chose rat anterior pituitary cells, the majority of which exhibit spontaneous [Ca 2ϩ ] i transients of high amplitude that are sufficient to trigger hormone secretion (10). These cells also express the messages for rod, cone, and olfactory CNGs, which may participate in the generation of spontaneous [Ca 2ϩ ] i transients (11). Although not fully characterized, the NO-signaling pathway is also expressed in these cells (12)(13)(14) and is activated by several G protein-coupled receptors (15)(16)(17). In our study, we initially characterized the NOS subtypes expressed in pituitary cells and their participation in the delivery of NO and sGC-controlled cGMP. We further characterized the role of calcium in the control of sGC activity in intact pituitary cells.

MATERIALS AND METHODS
Cell Cultures and Treatments-Experiments were performed on anterior pituitary cells from normal female Sprague-Dawley rats obtained from Taconic Farms (Germantown, NY). Pituitary cells were dispersed as described previously (18) and cultured in medium 199 containing Earle's salts, sodium bicarbonate, 10% horse serum, and antibiotics. Cell purification was done as described previously (18), and further identification of gonadotrophs and lactotrophs was done by the addition of specific Ca 2ϩ -mobilizing agonists for these cells, gonadotropin-releasing hormone and thyrotropin-releasing hormone (Peninsula, San Carlos, CA), respectively.
To express iNOS, cells (10 6 /well) were treated for 16 h with 30 g/ml lipopolysaccharide ϩ 1000 IU/ml interferon-␥ (LPSϩIFN-␥), both from Sigma. To elevate NO levels, cells were treated with three NO donors: sodium nitroprusside (SNP) from Research Biochemicals (Natick, MA) and N-ethylethanamine:1,1-diethyl-2-hydroxy-2-nitrosohydrazine (DEA) and 3,3Ј-(hydroxynitrosohydrazino)bis-1-propanamine (DPTA), both from Alexis Biochemicals (San Diego, CA). Basal and stimulated NOS activity * 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  cGMP, cAMP, and Nitrite Measurements-Cells (1 million/well) were plated in 24-well plates and incubated overnight at 37°C under 5% CO 2 -air and saturated humidity. Prior to the experiments, the medium was removed and cells were washed with Ca 2ϩ -containing medium 199 and stimulated at 37°C under 5% CO 2 -air and saturated humidity. cAMP and cGMP were measured in the medium and in dialyzed cells as described previously (18), using specific antisera provided by Albert Baukal (NICHD, Bethesda, MD). For measurement of total NO production (NO 2 Ϫ ϩ NO 3 Ϫ ), samples were initially treated with nitrate reductase (Alexis Biochemicals) to convert nitrate to nitrite. Sample aliquots were then mixed with an equal volume of Greiss reagent containing 0.5% sulfanilamide and 0.05% naphthylethylenediamine in 2.5% phosphoric acid (all from Sigma); the mixture was incubated at room temperature for 10 min, and the absorbance was measured at 546 nm (19). In experiments with NO 2 Ϫ measurements, samples were not treated with nitrate reductase. In both measurements, nitrite concentrations were determined relative to a standard curve derived from increasing concentrations of sodium nitrite. Concentrations of cAMP, cGMP, and NO 2 are expressed as combined values in cell content and in medium.
Western Blot Analysis-Postmitochondrial fractions of anterior pituitary tissue and dispersed pituitary cells, cerebellum, and aortic rings were obtained from adult female Sprague-Dawley rats. Concentration of proteins was estimated by the Bradford method using bovine serum albumin as a standard (20). Equal amounts of protein (22 g) from each postmitochondrial fraction were run on one-dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis using a discontinuous buffer system (NOVEX, San Diego, CA). The immunodetections on nNOS and iNOS were done with primary antibodies from Affinity BioReagents, Inc. (Golden, CO); eNOS antibody was from Transduction Laboratories (Lexington, KY); and ␤-actin antibody was from Oncogene Research Products (Boston, MA). The secondary antibody for all assays was an anti-mouse IgG (rabbit) from Life Technologies, Inc. (Gaithersburg, MD) linked to horseradish peroxidase. The reactive bands were always determined with a luminol-based kit, and the reaction was detected by an enhanced chemiluminescence system, using x-ray film. The Characterization of NO-controlled cGMP Production-Basal cGMP production was analyzed in mixed populations of anterior pituitary cells cultured in the absence of phosphodiesterase inhibitors. Under these conditions, cGMP was produced in a time-dependent manner, reaching the steady-state plateau response 60 -90 min after replacing medium in cultured cells (Fig. 1B, open circles). Consistent with the role of the NO signaling pathway in the control of cGMP production, the addition of NO donors, SNP, DEA, and DPTA (21), was accompanied by a significant increase in cGMP accumulation. In cells stimulated with LPSϩIFN-␥ for 16 h, cGMP production was also significantly elevated compared with controls and was comparable with levels observed in cells stimulated with SNP. Nitrite levels, which are commonly used as indicators of NO production (22), were significantly elevated in treated cells. As expected, the profiles of nitrite levels expressed as the sum of both NO 2 Ϫ and NO 3 Ϫ versus NO 2 Ϫ only (without nitrate reductase treatment) were highly comparable. The levels of total nitrite measured 1 h after stimulation were similar in SNP-and LPSϩIFN-␥-stimulated cells and significantly higher in cells treated with DPTA and DEA. Finally, the levels of cGMP progressively increased with the elevation in nitrite levels (Fig. 1C).

The Characterization of NO-derived Enzymes in Pituitary
Both basal and LPSϩIFN-␥-induced cGMP productions were completely inhibited by aminoguanidine, an NOS inhibitor (23), in the micromolar to millimolar concentration range ( Fig.  2A). In parallel to cGMP production, nitrite accumulation in cultured medium was inhibited by aminoguanidine (Fig. 2B). The blockers of sGC, NS 2028 (24), and ODQ (25) also inhibited cGMP productions in a dose-dependent manner. Fig. 3A illustrates the effect of NS 2028 on cGMP accumulation in pituitary cells during the first 60 min of incubation. In the same samples, NS 2028 treatment did not significantly affect cAMP production, indicating that AC is not sensitive to this compound. NS 2028 also inhibited the SNP-induced cGMP production (Fig.  3B) without affecting cAMP production. Likewise, basal and SNP-induced cGMP production, but not cAMP production, was significantly inhibited by ODQ (Fig. 4, A and B).
The Dependence of sGC Activity on [Ca 2ϩ ] i -To study the effects of cytosolic Ca 2ϩ on sGC activity, the [Ca 2ϩ ] i was elevated by depolarizing cells with high K ϩ . Fig. 5A illustrates that depolarization of cells with 50 mM K ϩ led to similar patterns of [Ca 2ϩ ] i response in somatotrophs, lactotrophs, and gonadotrophs (shown on left panels), as well as all other unidentified cells (not shown). A high K ϩ -induced increase in [Ca 2ϩ ] i was accompanied by a significant and dose-dependent inhibition of cGMP production to about 20% of that observed in controls (Fig. 5B, left panel). Both depolarization-induced elevation in [Ca 2ϩ ] i and inhibition of cGMP production were almost abolished in cells bathed in Ca 2ϩ -deficient medium (Fig.  5, A and B, right panels), indicating that depolarization per se is not responsible for the observed effects. Consistent with the role of [Ca 2ϩ ] i in the control of nNOS and eNOS, depletion of extracellular Ca 2ϩ was accompanied with a significant decrease in cGMP production (illustrated by the upper dashed line in Fig. 5B). Manipulation of extracellular Ca 2ϩ concentration did not obviously affect cAMP production (Fig. 5B).
The addition of two inhibitors of cGMP-specific phosphodiesterases, vinpocetine and dipyridamole (26,27), elevated basal cGMP levels (Fig. 6A). In cells with inhibited phosphodiesterases, however, depolarization of cells inhibited cGMP production with a comparable efficiency. High K ϩ -induced inhibition of cGMP production, but not cAMP production, was also observed in cells expressing iNOS (Fig. 6B). Furthermore, high K ϩ inhibited cGMP production in control cells treated with SNP, DEA, and DPTA (Fig. 6C). These results indicate that elevated [Ca 2ϩ ] i inhibits sGC activity and that this inhibition occurs in the presence of elevated NO levels.
High K ϩ -induced elevation in [Ca 2ϩ ] i (50, 75, and 100 mM) occurred in a nonoscillatory manner. In contrast, depolarization of cells with 10 -20 mM K ϩ frequently led to the generation of an oscillatory [Ca 2ϩ ] i response after an initial spike response. Fig. 7A illustrates the effects of 12 mM K ϩ on the pattern of [Ca 2ϩ ] i signaling in identified somatotrophs, lactotrophs, and gonadotrophs. At that particular concentration of K ϩ , a significant inhibition of cGMP production was observed in control cells (Fig. 5B), as well as in cells expressing iNOS (Fig. 6B). In further studies, the effects of low K ϩ were compared with those induced by Bay K 8644, an L-type Ca 2ϩ channel agonist. Addition of Bay K 8644 initiated [Ca 2ϩ ] i oscillations in quiescent somatotrophs, lactotrophs, gonadotrophs (Fig. 7B), and other pituitary cells and increased the amplitude of transients in oscillating cells (not shown). As in K ϩ -stimulated cells, the rise in [Ca 2ϩ ] i induced by Bay K 8644 was accompanied with the inhibition of cGMP production in controls and vinpocetine-treated cells (Fig. 7C, left panels). SNPinduced cGMP production was also inhibited by Bay K 8644 (Fig. 7C, right panel). Finally, Bay K 8644 inhibited cGMP production by LPSϩIFN-␥-treated cells in a dose-dependent manner (Fig. 7D).
To quantify the dependence of sGC activity on [Ca 2ϩ ] i , pituitary cells were stimulated with increasing concentrations of K ϩ (10, 18, 30, 50, 75, and 100 mM) and Bay K 8644 (0.1 and 1 M), and the averaged area [Ca 2ϩ ] i (t) curves during 15 min of stimulation were calculated and compared with cGMP production. Fig. 8, A and B

DISCUSSION
The NO-cGMP signaling pathway exhibits a complex action on Ca 2ϩ signaling in a number of cells. A cross-talk between NO and Ca 2ϩ is critical in cells with intra-and intercellular Ca 2ϩ oscillations, which are dependent on Ca 2ϩ release mechanisms, such as those reported recently in hepatocytes (6). NO-cGMP also facilitates the Ca 2ϩ influx pathway by stimulating CNGs, channels that depolarize cells to the level needed for activation of voltage-gated Ca 2ϩ channels (28). The same pathway may also inhibit Ca 2ϩ influx by activating Ca 2ϩ -controlled K ϩ channels in a cGMP-dependent protein kinase-dependent manner (8,29). Because the rise in [Ca 2ϩ ] i is required to activate nNOS and eNOS, the latter pathway would lead to the coordinated regulation of [Ca 2ϩ ] i and cGMP. That is not the case with cells expressing CNGs, where the positive feedback effects of Ca 2ϩ on NOS would elevate [Ca 2ϩ ] i in an uncontrolled manner.
Two recent reports have indicated the expression of CNGs in pituitary cells (11) and hypothalamic immortalized neurons (30); pituitary somatotrophs express the message for rod CNGs, whereas gonadotropin-releasing hormone-secreting neurons express the olfactory subtype of these channels. In general, activation of these channels by cAMP and cGMP leads to depolarization of cells and the subsequent facilitation of voltage-gated Ca 2ϩ influx (28). These channels may provide a rationale for the dual control of Ca 2ϩ signaling and secretion in neuroendocrine cells by the NO signaling pathway and by AC-coupled receptors. The NO-cGMP pathway is operative in pituitary and hypothalamic cells (12)(13)(14)(15)(16)(17) and provides a rationale for the control of spontaneous [Ca 2ϩ ] i fluctuations, which were observed in both normal and immortalized pituitary cell types (31)(32)(33)(34)(35)(36). Mixed populations of pituitary cells express two enzymes, nNOS and eNOS ( Fig. 1) (17, 37). These cells also express iNOS in response to LPSϩIFN-␥ stimulation (Fig. 1) (38). AC-coupled receptors are also operative in these cells and stimulate Ca 2ϩ influx (39 -42).
It was obvious that in cells where the NO-cGMP pathway is positively coupled to Ca 2ϩ influx, the inactivation of sGC is needed to break the stimulatory action of NO on Ca 2ϩ signaling and Ca 2ϩ -controlled cellular functions. However, it was not clear how sGC activity is controlled. Here we show that sCG activity in anterior pituitary cells is down-regulated by elevation in intracellular Ca 2ϩ and that the inhibition of the enzyme activity is dependent on the [Ca 2ϩ ] i . Our results further indicate that Ca 2ϩ -dependent inhibition of sGC activity occurs in the presence of elevated NO, i.e. independently of the status of NOS, as well as when cGMP-specific phosphodiesterases were inhibited.
Ca 2ϩ -dependent inhibition of sGC was observed in cells exhibiting oscillatory and nonoscillatory signaling.  (11), which control basal hormone secretion (10). The activation of several AC-coupled receptors also leads to the initiation of extracellular Ca 2ϩ -dependent [Ca 2ϩ ] i fluctuations in quiescent cells and to an increase in the frequency of fluctuations in spontaneously active cells (39 -42). Finally, when activated, Ca 2ϩ -mobilizing receptors generate extracellular Ca 2ϩ -independent oscillatory Ca 2ϩ signals in pituitary cells (10).
In accord with our observations concerning intact cells, it has also been shown recently that Ca 2ϩ inhibits sGC in crude cell extract and immunopurified preparations (43). Such a role of Ca 2ϩ signaling is not unique for sGC; intracellular Ca 2ϩ also inhibits two other members of this family of enzymes, particulate guanylyl cyclase (44,45) and adenylyl cyclase, particularly types V and VI (46). The common point in [Ca 2ϩ ] i -dependent inhibition of these enzymes is in the reciprocal regulation of intracellular concentrations of cyclic nucleotides and Ca 2ϩ . The unique characteristic of the action of elevated [Ca 2ϩ ] i on sGC signaling pathway is its dual action; it stimulates cGMP production by activating eNOS and nNOS but inhibits sGC even in the presence of elevated NO. This inhibition is not only required to balance the nNOS-and eNOS-generated NO production but also to protect the cells from overloading with Ca 2ϩ when NO production occurs in a [Ca 2ϩ ] i -independent manner, i.e. by iNOS (22) and phosphorylated eNOS (47).
In summary, here we show that sGC and AC are operative in unstimulated pituitary cells and differently regulated by [Ca 2ϩ ] i . The basal levels of cGMP are 2-4-fold higher than cAMP when estimated in the absence of phosphodiesterase inhibitors. Facilitation of voltage-gated Ca 2ϩ influx and removal of Ca 2ϩ from medium were practically ineffective in modulating cAMP production. This suggests that the participation of [Ca 2ϩ ] i -sensitive AC in unstimulated pituitary cells is minor. In contrast, both treatments affected sGC activity. Consistent with the role of [Ca 2ϩ ] i in facilitating nNOS and eNOS, cGMP production was inhibited but not abolished by culturing cells in Ca 2ϩ -deficient medium. Inhibition was not observed in cells with elevated NO. An increase in [Ca 2ϩ ] i also inhibited cGMP production but independently of NO levels and the phosphodiesterase activity. The low cAMP production and its independence of [Ca 2ϩ ] i and the high level of cGMP and its dependence on [Ca 2ϩ ] i are consistent with the coupling of the NO signaling pathway, but not the AC signaling pathway, to spontaneous Ca 2ϩ signaling in cultured pituitary cells.