Inhibition of serine/threonine phosphatase enhances arachidonic acid-induced [Ca²⁺]_i via protein kinase A

Abstract

Arachidonic acid (AA) regulates intracellular calcium concentration ([Ca²⁺]_i) in a variety of cell types including salivary cells. In the present study, the effects of serine/threonine phosphatases on AA-induced Ca²⁺ signaling in mouse parotid acini were determined. Mice were euthanized with CO₂. Treatment of acini with the serine/threonine phosphatase inhibitor calyculin A blocked both thapsigargin- and carbachol-induced Ca²⁺ entry but resulted in an enhancement of AA-induced Ca²⁺ release and entry. Effects were mimicked by the protein phosphatase-1 (PP1) inhibitor tautomycin but were inhibited by the PP2A inhibitor okadaic acid. The protein kinase A (PKA) inhibitor PKI_(14-22) significantly attenuated AA-induced enhancement of Ca²⁺ release and entry in the presence of calyculin A, whereas it had no effect on calyculin A-induced inhibition of thapsigargin-induced Ca²⁺ responses. The ryanodine receptor (RyR) inhibitor, tetracaine, and StHt-31, a peptide known to competitively inhibit type II PKA regulatory subunit binding to PKA-anchoring protein (AKAP), abolished calyculin A enhancement of AA-induced Ca²⁺ release and entry. StHt-31 also abolished forskolin potentiation of 4-chloro-3-ethylphenol (4-CEP) and AA on Ca²⁺ release but had no effect on 8-(4-methoxyphenylthio)-2′-O-methyladenosine-3′,5′-cAMP potentiation of 4-CEP responses. Results suggest that inhibition of PP1 results in an enhancement of AA-induced [Ca²⁺]_i via PKA, AKAP, and RyRs.

regulation of the intracellular calcium concentration ([Ca²⁺]_i) in nonexcitable cells is mediated by agonists that increase Ca²⁺ entry via depletion of intracellular Ca²⁺ stores, i.e., capacitative Ca²⁺ entry (CCE) (34). In several cell types, including parotid and pancreatic acinar cells (2, 19, 28, 30, 31, 40), Ca²⁺ entry is also regulated via a noncapacitative Ca²⁺ entry (NCCE) pathway. One activator of NCCE is arachidonic acid (AA) (28), an essential polyunsaturated fatty acid that plays an important role in intracellular signaling. Importantly, a distinction of the NCCE pathway is its dependence on low agonist concentrations (40). We recently reported that AA, at a concentration that inhibits the NCCE pathway [the arachidonate-regulated Ca²⁺ (ARC) channel (29)], regulates two Ca²⁺ entry pathways in mouse parotid acini; AA inhibited the CCE pathway and activated a second Ca²⁺ pathway that was not affected by gadolinium (Gd³⁺) but was dependent on depletion of Ca²⁺ from ryanodine-sensitive stores (46).

One way that [Ca²⁺]_i can be regulated in cells is via serine/threonine phosphatases. Calcineurin, a protein phosphatase-2B (PP2B) inhibitor, for example, has been implicated in the modulation of ion channels including N-methyl-d-aspartate receptor channels (13); L-type Ca²⁺ channels (27); NCCE channels, i.e., ARC channels (29); and ryanodine receptor (RyR) channels (4). Less is known about serine/threonine phosphatases PP1 and PP2A, which have been reported to regulate carbachol- and thapsigargin-induced Ca²⁺ entry in parotid acinar cells (38, 43), lacrimal acini (51), and platelets (32). In parotid cells, the mechanism(s) involved was not defined, whereas in lacrimal cells and platelets, protein kinase C (PKC)-dependent phosphorylation of an unidentified target protein(s) was sensitive to PP1/PP2A.

Recent studies demonstrated that, in addition to store-operated Ca²⁺ entry, another pathway exists in thyroid FRTL-5 cells that is activated by calyculin A alone and that is not blocked by Gd³⁺ or 2-aminoethoxydiphenyl borate (14). Furthermore, this novel pathway is dependent on protein kinase A (PKA) (14). The aim of the present study was to determine the role of serine/threonine phosphatases in AA-induced regulation of [Ca²⁺]_i in mouse parotid acini. We report that inhibition of serine/threonine phosphatase by calyculin A enhances AA-induced Ca²⁺ release and Ca²⁺ entry, whereas CCE is inhibited. Data further suggest that enhancement of AA-induced Ca²⁺ responses by calyculin A is dependent on PKA, acting via a PKA-anchoring protein (AKAP), and the release of Ca²⁺ from RyR-sensitive stores.

MATERIALS AND METHODS

Materials.

Materials were obtained as follows: arachidonic acid, thapsigargin, carbachol, forskolin, tetracaine, hyaluronidase, bovine serum albumin (BSA), HEPES, and 4-chloro-3-ethylphenol (4-CEP) from Sigma Chemical (St. Louis, MO); calyculin A, okadaic acid, and tautomycin from Calbiochem (La Jolla, CA); PKA-anchoring protein from Biomol Research Lab (Plymouth Meeting, PA); 8-(4-methoxyphenylthio)-2′-O-methyladenosine-3′,5′-cAMP (8-pMeOPT-2′-O-Me-cAMP) from Axxora (San Diego, CA); collagenase type 2 (CLS2) from Worthington (Freehold, NJ); fura-2 AM from Molecular Probes (Eugene, OR); and StHt-31 from Promega (Madison, WI). Male Swiss Webster mice (27–30 g) were from Taconic (Germantown, NY). Incubation times for AA, thapsigargin, carbachol, and forskolin were based on our previous publications (47); StHt-31 and phosphatase inhibitor incubation times were based on the literature and preliminary studies.

Preparation of parotid acini.

Protocols and all experiments with animals were approved and conducted under the authority of the University of Washington Institutional Animal Care and Use Committee. Small groups of isolated mouse parotid cells (acini) from CO₂-euthanized animals were prepared as described previously by Watson et al. (47).

Measurement of [Ca²⁺]_i in intact cells.

Acini were suspended 1:50 (wt/vol) in Krebs-Henseleit bicarbonate (KHB) buffer containing 0.176 mg/ml ascorbic acid and 0.2% BSA, pH 7.4, and loaded with fura-2 AM at 3.3 μg/ml of cell suspension for 30 min at 37°C with continuous gassing (5% CO₂-95% O₂) and shaking. Fura-2 AM was prepared at 1 mg/ml in DMSO just before use. Loaded cells were washed three times in 0.2% BSA-KHB buffer, resuspended at 1:50 (wt/vol), and maintained at 24°C with gassing and shaking. Because AA binds to BSA (1), thus reducing the effective AA concentration, a concentration of AA (45 μM), in combination with a reduced level of BSA (0.025%), used to maintain cell integrity, was employed. The 0.025% BSA was sufficient to prevent cell leakage as determined by Trypan blue exclusion and illustrated by the thapsigargin studies conducted under identical incubation conditions (46). After a 25-min incubation period, an aliquot of cells was washed twice in the above buffers ± Ca²⁺, diluted 1:10, and placed in ultraviolet grade fluorometric cuvettes (Spectrocel). Calcium was measured with a Filterscan spectrofluorometer system equipped with a magnetic stirrer and constant temperature cuvette holder (Photon Technology International, South Brunswick, NJ). [Ca²⁺]_i was calculated using the equation of Grynkiewicz et al. (15), where K_d = 224 nM. Data are expressed as the ratio of fura-2 fluorescence due to excitation at 340 nm to that due to excitation at 380 nm.

Data analysis.

Calcium release and entry data were collected as the ratio of the peak response, calculated as the average ± SE of n experiments and are expressed as the percent inhibition relative to maximal stimulation above control. Statistical analysis was performed using paired Student's t-test (P < 0.05).

RESULTS

Calyculin A augments AA-induced Ca²⁺ release and Ca²⁺ entry.

To determine whether serine/threonine phosphatases play a role in AA-induced Ca²⁺ release and entry, acini were incubated with specific serine/threonine phosphatase inhibitors, calyculin A, tautomycin, and okadaic acid. When parotid acini were preincubated for 10 min with calyculin A (100 nM), a PP1 and PP2A inhibitor, resting levels of [Ca²⁺]_i were not affected. However, both AA-induced Ca²⁺ release and Ca²⁺ entry were significantly enhanced, i.e., by 30 ± 3.1% and 78.7 ± 7.1%, respectively (Fig. 1, trace b). Pretreatment of acini for 30 min with tautomycin, a PP1 inhibitor, mimicked the effects of calyculin A in enhancing AA-induced Ca²⁺ release by 29.3 ± 1.7% and Ca²⁺ entry by 88.8 ± 11% (Fig. 2A, trace b), whereas preincubation of acini for 45 min with okadaic acid, a PP2A inhibitor, had little effect on Ca²⁺ release, and inhibited Ca²⁺ entry, i.e., by 20 ± 3.9% (Fig. 2B, trace b).

Fig. 1.

Effect of calyculin A (Cal) on arachidonic acid (AA)-induced Ca²⁺ release and Ca²⁺ entry in mouse parotid acini. Acini were incubated in nominally Ca²⁺-free Krebs-Henseleit bicarbonate (KHB) buffer in the absence (trace a) and presence (trace b) of calyculin A (100 nM) for 10 min (not recorded), before the addition, at 90 s, of AA (45 μM). Ca²⁺ (1.28 mM) was reintroduced at 420 s. Control (trace c) represents acini incubated in the presence of calyculin A alone, followed by the reintroduction of 1.28 mM Ca²⁺. Results are representative of 10 experiments (P < 0.05).

Fig. 2.

Effects of tautomycin (Taut) and okadaic acid (OK) on AA-induced Ca²⁺ release and Ca²⁺ entry in mouse parotid acini. Acini were incubated in nominally Ca²⁺-free KHB in the absence (trace a) and presence (trace b) of tautomycin (1 μM) (A) for 30 min (not recorded) and okadaic acid (1 μM) (B) for 30 min (not recorded), before the addition, at 90 s, of AA (45 μM). Ca²⁺ (1.28 mM) was reintroduced at 420 s. Control (trace c) represents acini incubated in the presence of tautomycin (A) and okadaic acid (B) alone, followed by the reintroduction of 1.28 mM Ca²⁺. Results are representative of four experiments (P < 0.05).

Because AA-induced Ca²⁺ entry in mouse parotid acini is distinct from CCE (46), we compared the effects of the serine/threonine phosphatase inhibitor, calyculin A, on AA-induced [Ca²⁺]_i with effects on thapsigargin, a known activator of CCE. As shown in Fig. 3A (trace b), calyculin A (100 nM) attenuated thapsigargin (2 μM)-induced Ca²⁺ release by 29.3 ± 1.7% and Ca²⁺ entry by 84.0 ± 4.5%. A similar inhibiting effect by serine/threonine phosphatase inhibitors on thapsigargin-induced Ca²⁺ responses was previously observed in rat parotid acinar cells (38, 43) and thyroid FRTL-5 cells (14). Calyculin A also inhibited the effects of the G protein-coupled receptor agonist carbachol (10 μM) on Ca²⁺ entry, and it decreased the time required for Ca²⁺ to return to basal values (Fig. 3B, trace b). Whether calyculin A increases Ca²⁺ uptake into intracellular stores or increases removal from the cell has yet to be determined. Importantly, data indicate that inhibition of serine/threonine phosphatase, specifically PP1, results in augmentation of AA-induced [Ca²⁺]_i, suggesting that a phosphorylation event is involved in regulating both Ca²⁺ release and Ca²⁺ entry.

Fig. 3.

Effects of calyculin A on thapsigargin (Thaps) and carbachol (Carb)-induced Ca²⁺ release and Ca²⁺ entry in mouse parotid acini. Acini were incubated in nominally Ca²⁺-free KHB in the absence (trace a) and presence (trace b) of calyculin A (100 nM) for 10 min (not recorded), before the addition, at 90 s, of thapsigargin (2 μM) (A) and carbachol (10 μM) (B). Ca²⁺ (1.28 mM) was reintroduced at 420 s. Control (trace c) represents acini incubated in the presence of calyculin A alone, followed by the reintroduction of 1.28 mM Ca²⁺. Results are representative of four experiments (P < 0.05).

Calyculin A enhancement of AA-induced Ca²⁺ release and Ca²⁺ entry involves PKA.

In the next series of experiments, the mechanism(s) by which calyculin A enhances AA-induced Ca²⁺ release and entry was determined. In thyroid FRTL-5 cells, phosphatase inhibition by calyculin A revealed a Ca²⁺ entry pathway dependent on PKA (14). To determine whether PKA is also involved in calyculin A augmentation of AA-induced Ca²⁺ release and entry in mouse parotid cells, acini were incubated with the PKA inhibitor PKI_(14-22) (1 μM) for 30 min before the addition of AA. In the presence of calyculin A, PKI_(14-22) reversed AA-induced augmentation of Ca²⁺ release and significantly inhibited Ca²⁺ entry by 87.6 ± 2.4% (Fig. 4A, trace c). In the absence of calyculin A, PKI_(14-22) had no effect on either AA-induced Ca²⁺ release or Ca²⁺ entry (Fig. 4B). In contrast to PKA playing a role in calyculin A augmentation of AA-induced Ca²⁺ responses, thapsigargin-induced Ca²⁺ release and entry (Fig. 5, trace b), as well as carbachol-induced Ca²⁺ responses (data not shown), in the presence of calyculin A, were unaffected by PKA.

Fig. 4.

Effects of the PKA inhibitor, PKI_(14-22) on AA-induced Ca²⁺ release and Ca²⁺ entry in the absence and presence of calyculin A in mouse parotid acini. A: acini were incubated in nominally Ca²⁺-free buffer, in the absence (traces a and b) and presence (trace c) of PKI_(14-22) (1 μM) for 30 min (not recorded), and in the absence (trace a) and presence (traces b and c) of calyculin A (100 nM) for an additional 10 min, before the addition, at 60 s, of AA (45 μM). Ca²⁺ (1.28 mM) was reintroduced at 420 s. Control (trace d) represents acini incubated in the presence of calyculin A (100 nM) and PKI_(14-22) (1 μM), followed by the reintroduction of 1.28 mM Ca²⁺. Results are representative of five experiments (P < 0.05). B: acini were incubated in nominally Ca²⁺-free buffer in the absence (trace a) and presence (trace b) of PKI_(14-22) (1 μM) for 30 min (not recorded), before the addition, at 60 s, of AA (45 μM). Ca²⁺ (1.28 mM) was reintroduced at 420 s. Control (trace c) represents acini incubated with PKI_(14-22) (1 μM) followed by the reintroduction of 1.28 mM Ca²⁺. Results are representative of five experiments (P < 0.05).

Fig. 5.

Effects of calyculin A on thapsigargin-induced Ca²⁺ release and Ca²⁺ entry in the absence and presence of the PKA inhibitor PKI_(14-22) in mouse parotid acini. Acini were incubated in nominally Ca²⁺-free buffer, in the absence (traces a and c) and presence (trace b) of PKI_(14-22) (1 μM), for 30 min (not recorded), and in the absence (trace a) and presence (traces b and c) of calyculin A (100 nM) for an additional 10 min (not recorded), before the addition, at 90 s, of thapsigargin (2 μM). Ca²⁺ (1.28 mM) was reintroduced at 420 s. Control (trace d) represents acini incubated in the presence of calyculin A (100 nM) plus PKI_(14-22) (1 μM) followed by the reintroduction of 1.28 mM Ca²⁺. Results are representative of three experiments (P < 0.05).

Enhancement of AA-induced- Ca²⁺ release is dependent on ryanodine-sensitive stores.

Previous studies from our laboratory demonstrated that RyRs are present in mouse parotid acini (9) and that AA induces release of Ca²⁺ from ryanodine-sensitive stores (46). Thus we determined whether AA-induced augmentation of Ca²⁺ entry by calyculin A was dependent on the release of Ca²⁺ from ryanodine-sensitive stores. For these studies, we used the RyR-sensitive store antagonist tetracaine, previously shown to inhibit AA-induced Ca²⁺ increase from parotid acini (46). As shown in Fig. 6 (trace b), calyculin A augmented AA-induced Ca²⁺ release and Ca²⁺ entry as described in Fig. 1. In the presence of tetracaine (500 μM), augmentation of AA-induced Ca²⁺ release and Ca²⁺ entry by calyculin A was reduced by 82.1 ± 3% and 76.4 ± 4%, respectively (Fig. 6, trace c). Tetracaine also blocked AA-induced Ca²⁺ release and Ca²⁺ entry by 82.6 ± 3.8% and 74.7% ± 2.3%, respectively, in the absence of calyculin A (Fig. 6, trace d), as previously reported (46).

Fig. 6.

Effects of tetracaine (Tet) on AA-induced Ca²⁺ release and Ca²⁺ entry in the absence and presence of calyculin A. Acini were incubated in nominally Ca²⁺-free buffer in the absence (traces a and b) and presence (traces c and d) of tetracaine (500 μM) for 5 min (not recorded), and in the absence (traces a and d) and presence (traces b and c) of calyculin A for 10 min (not recorded), before the addition, at 90 s, of AA (45 μM). Ca²⁺ (1.28 mM) was reintroduced at 420 s. Control (trace e) represents acini incubated in the presence of calyculin A (100 nM) plus tetracaine (500 μM) followed by the reintroduction of 1.28 mM Ca²⁺. Results are representative of three experiments (P < 0.05).

Forskolin potentiation of 4-CEP and AA on [Ca²⁺]_i.

Since data suggest that AA releases Ca²⁺ from ryanodine-sensitive stores in parotid acini, we determined whether PKA directly activates RyRs. In initial experiments, we determined whether forskolin, a known activator of PKA, releases Ca²⁺ from RyR-sensitive stores activated by 4-CEP. Acini were treated with forskolin (10 μM) for 10 min, and the effects of 4-CEP on Ca²⁺ release and entry were determined. As shown in Fig. 7A, trace a, 4-CEP (200 μM) produced a slow increase in Ca²⁺ release as previously reported in mouse parotid acini (3, 46). In the presence of forskolin (10 μM), effects of 4-CEP on Ca²⁺ release were potentiated (trace b). Forskolin (10 μM) also potentiated the effects of 4-CEP on Ca²⁺ entry (Fig. 7A, inset trace b). Forskolin alone, however, had no effect on resting [Ca²⁺]. In the second experiment, we determined whether forskolin also potentiates the effects of AA (45 μM) on Ca²⁺ release, because this compound was reported to enhance the binding of [³H]ryanodine to mouse parotid microsomal membranes (9) and to activate RyRs in isolated parotid acini (46). As shown in Fig. 7B (trace b), addition of forskolin (25 μM) to parotid acini also potentiated the effects of AA (45 μM) on Ca²⁺ release. Forskolin (10 μM) also potentiated AA-induced Ca²⁺ release, but to a lesser extent. Results thus suggested that under these conditions, PKA directly releases Ca²⁺ from ryanodine stores in mouse parotid acini.

Fig. 7.

Effects of forskolin (For) on 4-chloro-3-ethylphenol (4-CEP) and AA-induced Ca²⁺ release from mouse parotid acini. A: acini were incubated in a nominally Ca²⁺-free buffer in the absence (trace a) and presence (trace b) of forskolin (10 μM) for 10 min (not recorded), before the addition, at 90 s, of 4-CEP (200 μM). Inset: acini were incubated in a Ca²⁺-containing buffer in the absence (trace a) and presence (trace b) of forskolin (10 μM) for 10 min (not recorded), before the addition of 4-CEP (200 μM). B: acini were incubated in a nominally Ca²⁺-free buffer in the absence (trace a) and presence (trace b) of forskolin (25 μM) for 10 min (not recorded), before the addition, at 90 s, of AA (45 μM). Controls (trace c) represent acini incubated without agonists. Results are representative of three experiments (P < 0.05).

AA-induces Ca²⁺ release via an A-kinase PKA-anchoring protein (AKAP).

RyRs have been reported to exist in a macromolecular signaling complex with PKA, AKAP, and phosphatases (6, 26). Furthermore, PKA is known to be targeted to specific subcellular locations through the association with AKAPs (7, 11, 24, 33, 36). Because results indicated that calyculin A produces augmentation of AA-induced Ca²⁺ entry by releasing Ca²⁺ from ryanodine stores sensitive to PKA, further studies were conducted to evaluate the role of AKAP in PKA regulation of Ca²⁺ release and entry in parotid acini. For these studies, we used the cell- permeable peptide inhibitor StHt-31 (10, 45), which inhibits type II PKA regulatory subunit (RII) binding to AKAP (12). Acini were pretreated with StHt-31 (10 μM) for 30 min before the addition of calyculin A (100 nM) for an additional 10 min before the addition of AA (45 μM). As shown in Fig. 8 (trace c), augmentation of AA-induced Ca²⁺ release and entry by calyculin A (100 nM) was almost completely blocked in the presence of StHt-31. StHt-31, alone, elicited no effect on AA-induced Ca²⁺ responses (data not shown), consistent with the lack of effect of PKA noted under these conditions.

Fig. 8.

Effects of StHt-31 on AA-induced Ca²⁺ release and Ca²⁺ entry in mouse parotid acini. Acini were incubated in a nominally Ca²⁺-free buffer, in the absence (traces a and b) and presence (trace c) of StHt-31 (10 μM) for 30 min (not recorded), and in the absence (trace a) and presence (traces b and c) of calyculin A (100 nM) for an additional 10 min (not recorded), before the addition, at 90 s, of AA (45 μM). Ca²⁺ (1.28 mM) was reintroduced at 420 s. Control (trace d) represents acini incubated in the presence of StHt-31 (10 μM) and calyculin A (100 nM) followed by the reintroduction of 1.28 mM Ca²⁺. Results are representative of three experiments (P < 0.05).

Forskolin potentiates 4-CEP and AA-induced Ca²⁺ release via an A-kinase PKA-anchoring protein (AKAP).

As proof of specificity and association between AKAP and RyRs in parotid cells, we further examined the effects of StHt-31 on forskolin enhancement of 4-CEP, as well as on forskolin enhancement of AA-induced Ca²⁺ release. Acini were treated with StHt-31 (10 μM) for 30 min and forskolin (25 μM) for an additional 10 min before the addition of 4-CEP (200 μM; Fig. 9A) or AA (45 μM; Fig. 9B). StHt-31 reversed forskolin augmentation of both 4-CEP and AA-induced Ca²⁺ release by 88 ± 7% and 94% ± 2.3, respectively (Fig. 9, A and B, trace c). StHt-31, alone, elicited no effect on Ca²⁺ release (Fig. 9, A and B, trace d).

Fig. 9.

Effects of StHt-31 on forskolin augmentation of 4-CEP and AA-induced Ca²⁺ release in mouse parotid acini. Acini were incubated in a nominally Ca²⁺-free buffer, in the absence (traces a and b) and presence (trace c) of StHt-31 (10 μM) for 30 min (not recorded), and in the absence (trace a) and presence (traces b and c) of forskolin (25 μM) for an additional 10 min (not recorded), before the addition, at 60 s, of 4-CEP (200 μM) (A) and AA (45 μM) (B). Controls (trace d) represent acini incubated with StHt-31 alone. Results are representative of four experiments (P < 0.05).

As further proof that forskolin augmentation of 4-CEP-induced Ca²⁺ release is due to PKA, rather than to stimulation of Epac, a cAMP-regulated guanine nucleotide exchange factor reported to affect the release of Ca²⁺ from intracellular stores (17, 20), we examined the effect of StHT-31 on Epac augmentation of 4-CEP-induced Ca²⁺ release. Acini were treated with the Epac activator 8-pMeOPT-2′-O-Me-cAMP (100 μM) for 10 min before the addition of 4-CEP (200 μM). Results presented in Fig. 10 show that 8-pMeOPT-2′-O-Me-cAMP also augmented 4-CEP-stimulated Ca²⁺ release. However, treatment of acini with StHt-31 (10 μM) for 30 min before the addition of 8-pMeOPT-2′-O-Me-cAMP (100 μM) did not abolish this response.

Fig. 10.

Effects of StHt-31 on 8-(4-methoxyphenylthio)-2′-O-methyladenosine-3′,5′-cAMP (8-pMeOPT) augmentation of 4-CEP in mouse parotid acini. Acini were incubated in a nominally Ca²⁺-free buffer, in the absence (traces a and b), and presence (trace c) of StHt-31 (10 μM) for 30 min (not recorded), and in the absence (trace a) and presence (traces b and c) of 8-pMeOPT (100 μM) for an additional 10 min (not recorded), before the addition, at 60 s, of 4-CEP (200 μM). Controls (trace d) represent acini incubated with StHt-31 alone. Results are representative of four experiments (P < 0.05).

DISCUSSION AND CONCLUSIONS

We previously reported that AA regulates Ca²⁺ entry by depletion of Ca²⁺ from ryanodine-sensitive intracellular stores (46). In the present study we used serine/threonine phosphatase inhibitors, i.e., calyculin A, okadaic acid, and tautomycin, to determine the identity and role of the phosphatases in regulating AA-induced Ca²⁺ signaling. The effects of calyculin A appear to be agonist dependent, because calyculin A enhanced AA-induced Ca²⁺ responses but inhibited both thapsigargin and carbachol-induced Ca²⁺ responses. The major finding of the present study is that inhibition of serine/threonine phosphatase PP1 augments AA-induced Ca²⁺ entry in mouse parotid acini by a mechanism involving PKA-mediated Ca²⁺ release from ryanodine-sensitive stores. Inhibition of serine/threonine phosphate by calyculin A, alone, was previously reported to increase Ca²⁺ entry in FRTL-5 cells (14), whereas, in parotid acini, pretreatment of acini with calyculin A for 30 min before the addition of AA did not change basal Ca²⁺ levels. Furthermore, in parotid acini, calyculin A increased AA-induced Ca²⁺ entry by releasing Ca²⁺ from ryanodine-sensitive stores, whereas calyculin A-induced Ca²⁺ entry in FRTL-5 cells was independent of intracellular stores (14).

Importantly, in the present study, PKA had no effect on AA-induced Ca²⁺ release or entry in the absence of calyculin A. Inhibition of endogenous PP1 was required to observe effects of PKA on [Ca²⁺]_i . This suggests that inhibition of serine/threonine phosphatase increases the activity of PKA sufficient to release intracellular Ca²⁺ stores. Common to both FRTL-5 cells (14) and parotid acini, however, is the observation that increases in Ca²⁺ entry by calyculin A are not blocked by Gd³⁺ (data not shown), an inhibitor shown to block store-operated Ca²⁺ channels, but not AA-induced Ca²⁺ entry in mouse parotid acini (46). In both thyroid and FRTL-5 cells and parotid cells, the Ca²⁺ entry channel affected by calyculin A is not known and will require further investigation.

An important aspect of the present study is the dependence of AA-induced Ca²⁺ entry on PKA, observed only in the presence of calyculin A. This suggests that inhibition of serine/threonine phosphatase increases the activity of PKA sufficient to release intracellular Ca²⁺ stores. Protein phosphatases regulate a large number of cellular functions and signal transduction pathways in cooperation with protein kinase; they dephosphorylate a large number of substrates (6). PP1 is found in nearly all cellular compartments and is thought to be targeted to intracellular organelles including the sarcoplasmic and endoplasmic reticulum (8, 18). The free catalytic subunit of PP1 activity has been referred to as promiscuous because of its measured high activity and low specificity (22). High intrinsic activity is necessary to dephosphorylate and thereby downregulate ion channels (42).

A major mechanism for regulating native forms of PPI is phosphorylation of its regulatory subunits by PKA, PKC, myosin light-chain kinase, and casein kinase II (41). PP1 activity is also regulated by endogenous inhibitor proteins (39) as well as exogenous products that inhibit serine/threonine phosphatase 1, e.g., tautomycin (41). Among its many functions, PP1 is an enzyme that promotes the rational use of energy. Reversal of the cell to a basal and/or energy-conserving state plays a key role in recovery from stress, promotes apoptosis when cells are damaged beyond repair, downregulates ion pumps and transporters and ion channels in various tissues, and promotes the exit from mitosis and maintains cells in the G1 or G2 phases of the cell cycle (6). In the present study, ryanodine-associated PP1 is thought to counteract PKA, i.e., to dephosphorylate RyR subunits and the conductance of RyR channels (16, 49). Calcium release channels, i.e., RyR and inositol 1,4,5-trisphosphate (IP₃) receptor (IP₃R), are part of a macromolecular complex containing AKAP-6, PP1, and PP2A (37) and are targets for PP1 in parotid cells. PP1 activity is regulated in this complex to ensure appropriate responses to extracellular signals.

In our study, we found that inhibition of PP1 increased Ca²⁺ release/entry by increasing PKA phosphorylation of the RyR Ca²⁺ release channel. The localization of PKA near a particular substrate appears to be highly important for the regulation of specific physiological events (7). RyRs are known to be regulated by PKA phosphorylation (37). Previous studies provided evidence for PKA phosphorylation of IP₃R following stimulation with physiological levels of CCK in pancreatic acinar cells (23) and carbachol in mouse parotid cells (3). In isolated parotid acinar cells, Bruce et al. (3) reported that forskolin potentiates carbachol-evoked increases in Ca²⁺ from IP₃-sensitive stores, whereas no evidence was obtained to support direct PKA-mediated activation of RyRs. In contrast, calyculin A potentiated AA-induced Ca²⁺ release in the present study, by a mechanism involving PKA regulation of RyR-sensitive stores. That PKA regulates ryanodine-sensitive stores, and not IP₃-sensitive stores, in the presence of AA and calyculin A, is supported by the following data: 1) AA alone increased Ca²⁺ release in mouse parotid acini from ryanodine and not IP₃ stores (46); 2) tetracaine, an inhibitor of RyRs, prevented AA-induced Ca²⁺ release and entry in parotid acini treated without (46) and with calyculin A; 3) AA-enhancement of Ca²⁺ release by calyculin A was inhibited by PKI_(14-22); and 4) forskolin enhanced 4-CEP and AA on Ca²⁺ release, both of which were reversed by StHt-31. A recent study by Luo et al. (25) also indicates that AA, at low and high concentrations, releases Ca²⁺ from RyR stores and increases Ca²⁺ entry in human embryonic kidney 293 cells.

In contrast to our findings, Bruce et al. (3) failed to observe forskolin augmentation of 4-chloro-m-cresol on Ca²⁺ release in mouse parotid cells, at a forskolin concentration used in the present study. Because RyRs are in low abundance, compared with IP₃Rs in nonexcitable cells, including mouse parotid cells (9), Bruce et al. (3) were unable to observe phosphorylation of RyRs by PKA. To circumvent this problem, we determined whether there was an association between PKA and RyRs by demonstrating that forskolin enhancement of 4-CEP, as well as AA, on Ca²⁺ release is blocked by StHt-31. Although mouse parotid acini were used in both studies, there were differences in cell preparations as well as in mouse strains. Over the years, we have found that responses to a given agonist can differ significantly depending on the mouse strain. Differences in responses to forskolin may also be related to the use of cell suspensions in this study, rather than single cells. Cyclic AMP was also reported to release Ca²⁺ from ryanodine stores in rat parotid suspensions (48). Although Bruce et al. (3) reported PKA-mediated phosphorylation of IP₃Rs, but not RyRs, forskolin had no affect on resting [Ca²⁺]_i.

In the present study, we also found that forskolin (10 μM) alone had no effect on [Ca²⁺]_i. However, in the presence of either 4-CEP or AA, forskolin enhanced Ca²⁺ release via an effect of PKA on RyRs. Importantly, although augmentation of 4-CEP was mimicked by the Epac activator 8-pMeOPT-2′-O-Me-cAMP, Ca²⁺ release was not blocked by StHt-31. Therefore, an important question relates to how PKA activates the RyR. Failure to observe PKA phosphorylation of RyR may be related to the fact that RyRs may not be directly phosphorylated. A recent study by Zissimopoulos et al. (50) showed that RyRs, i.e., RyR1, RyR2, and RyR3, bind to snapin, a ubiquitously expressed SNARE-associated protein in nonneuronal cells (5). Further studies will be required to determine whether there is an association between RyRs and snapin in parotid cells, and whether PKA phosphorylation of snapin regulates Ca²⁺ release.

In establishing a role for PKA in calyculin A augmentation of AA-induced Ca²⁺ release and entry, we also found that StHt-31, a peptide known to inhibit RII binding to AKAP (12), significantly blocked enhancement of AA-induced Ca²⁺ release and entry. AKAPs are expressed in salivary cells (21) and are typically part of a complex that includes not only PKA (10, 12, 35) but also serine/threonine phosphatases and IP₃R/RyR (37, 44). Importantly, by anchoring PKA in close proximity to RyR, AKAP regulates RyR function in muscle cells (37). To date, no studies have been reported that address the role of a PKA/AKAP/RyR complex in nonmuscle cells. Our data suggest that a similar complex may be operative in parotid cells and would represent the first report of such a complex existing in nonexcitable cells. Enhanced PKA activity causes a maintained potentiation of local Ca²⁺ release signals.

Data presented show that treatment of acini with calyculin A produced differential effects on AA and thapsigargin/carbachol-induced Ca²⁺ signaling. Whereas calyculin A attenuated thapsigargin- and carbachol-induced Ca²⁺ signaling, AA-induced Ca²⁺ release and entry were augmented, and by a novel mechanism involving the serine/threonine phosphatase, PP1, PKA, and an AKAP. Protein phosphorylation and dephosphorylation are considered to be key steps in the control of secretion. A major function of parotid acinar cells is to secrete water and electrolytes and proteins that are necessary for oral health. Acinar secretion is highly dependent on Ca²⁺. Thus, in our study, inhibition of PP1 by calyculin A and an increase in AA-induced [Ca²⁺]_i would lead to an increase in amylase secretion from mouse parotid acini.

GRANTS

This work was supported by National Institute of Dental and Craniofacial Research Grant DE-05249.