Phosphorylation of the Inositol 1,4,5-Trisphosphate Receptor by Cyclic Nucleotide-dependent Kinases in Vitroand in Rat Cerebellar Slices in Situ *

We have examined cyclic nucleotide-regulated phosphorylation of the neuronal type I inositol 1,4,5-trisphosphate (IP3) receptor immunopurified from rat cerebellar membranes in vitro and in rat cerebellar slices in situ. The isolated IP3 receptor protein was phosphorylated by both cAMP- and cGMP-dependent protein kinases on two distinct sites as determined by thermolytic phosphopeptide mapping, phosphopeptide 1, representing Ser-1589, and phosphopeptide 2, representing Ser-1756 in the rat protein (Ferris, C. D., Cameron, A. M., Bredt, D. S., Huganir, R. L., and Snyder, S. H. (1991) Biochem. Biophys. Res. Commun. 175, 192–198). Phosphopeptide maps show that cAMP-dependent protein kinase (PKA) labeled both sites with the same time course and same stoichiometry, whereas cGMP-dependent protein kinase (PKG) phosphorylated Ser-1756 with a higher velocity and a higher stoichiometry than Ser-1589. Synthetic decapeptides corresponding to the two phosphorylation sites (peptide 1, AARRDSVLAA (Ser-1589), and peptide 2, SGRRESLTSF (Ser-1756)) were used to determine kinetic constants for the phosphorylation by PKG and PKA, and the catalytic efficiencies were in agreement with the results obtained by in vitro phosphorylation of the intact protein. In cerebellar slices prelabeled with [32P]orthophosphate, activation of endogenous kinases by incubation in the presence of cAMP/cGMP analogues and specific inhibitors of PKG and PKA induced in both cases a 3-fold increase in phosphorylation of the IP3 receptor. Thermolytic phosphopeptide mapping of in situ labeled IP3 receptor by PKA showed labeling on the same sites (Ser-1589 and Ser-1756) as in vitro. In contrast to the findings in vitro, PKG preferentially phosphorylated Ser-1589 in situ. Because both PKG and the IP3receptor are specifically enriched in cerebellar Purkinje cells, PKG may be an important IP3 receptor regulator in vivo.

Activation of intracellular signal transduction cascades frequently involves increased phosphoinositide hydrolysis following stimulation of phospholipase C. Inositol 1,4,5-trisphosphate (IP 3 ), 1 a second messenger produced by phosphoinositide hydrolysis, mediates Ca 2ϩ release from intracellular stores by binding to IP 3 -sensitive Ca 2ϩ channels, thereby increasing their "open" probability (1). IP 3 receptors derive from at least three different genes, constituting types I, II, and III, which are approximately 70% identical at the amino acid level but differ in distribution and regulation (reviewed in Refs. 1-3). An assembly of four 260-kDa subunits forms the receptor. Each subunit consists of a cytoplasmic, amino-terminal IP 3 binding domain, a coupling domain, and a Ca 2ϩ channel pore of six transmembrane segments (2)(3)(4). Type I is further diversified by alternative RNA splicing, resulting in two main forms, of which the longest (SIIϩ, containing the 40 amino acid residues 1693-1732) is specifically expressed in neurons (2). One or more IP 3 receptor forms have been found in virtually all cell types examined (reviewed in Ref. 2), but particularly high amounts of type I IP 3 receptor are seen in smooth muscle cells and in cerebellar Purkinje neurons. Calcium release mediated by IP 3 receptors appears to be an essential step for the induction of long term depression (LTD) in Purkinje cells (5).
A number of different mechanisms modulates IP 3 receptor function, including binding of ATP, fatty acids, and calcium (reviewed in Ref. 2); a number of neurodegenerative processes (6,7); and phosphorylation of the IP 3 receptor by specific protein kinases. cAMP-dependent protein kinase (PKA) phosphorylates the type I IP 3 receptor both in vitro and in vivo (8 -11) and has also been reported to phosphorylate type II and III in intact cells (12). Ca 2ϩ /calmodulin-dependent protein kinase II, protein kinase C and the tyrosine kinase Fyn have also been reported to phosphorylate the type I IP 3 receptor (13)(14)(15)(16)(17). In addition, the receptor may undergo autophosphorylation (18). Early work indicating that the neuronal IP 3 receptor (SIIϩ) can be phosphorylated by cGMP-dependent protein kinase (PKG) (8) was later confirmed by in vitro experiments (19 -21). Likewise, the nonneuronal type I IP 3 receptor (SIIϪ) found in smooth muscle cells, also termed the G 0 protein (22,23), is a substrate for phosphorylation by both PKA (10) and PKG (19,20,(23)(24)(25). Recent reports of IP 3 receptor phosphorylation by PKA and PKG in hepatocytes (26 -28), kidney cells (11), and platelets (29,30) support these observations. Phosphorylation of the IP 3 receptor by PKA and PKG represents a possible mechanism for cross-talk whereby cyclic nucleotides can modulate IP 3 -mediated regulation of Ca 2ϩ levels (20,31). Because cAMP and cGMP levels in most cells are regulated by various extracellular signals, identification of phosphorylation sites labeled by these kinases is of interest. Amino acid sequencing indicated that PKA phosphorylates the rat neuronal receptor on Ser-1756 and, less efficiently, on Ser-1589 (9). Similarly, PKG appears to phosphorylate mainly Ser-1756 in vitro (19). In contrast, the smooth muscle IP 3 receptor is preferentially phosphorylated by PKA on Ser-1589 in vitro (10), whereas Ser-1756 is more prominently phosphorylated by PKA in kidney cells in vivo (11). Hence, the preferred substrate sites for PKA-mediated phosphorylation of the IP 3 receptor differ according to receptor isoforms and tissues, and the factors determining these responses remain unclear. Moreover, IP 3 receptor phosphorylation by PKG in neuronal cells has not been characterized in detail.
In the present study, we have examined in vitro phosphorylation catalyzed by the cyclic nucleotide-dependent protein kinases both of the immunoisolated cerebellar IP 3 receptor and of synthetic peptides corresponding to the phosphorylation sites. We have further characterized the in situ phosphorylation of the IP 3 receptor in cerebellar slices stimulated with cyclic nucleotide analogues. The results confirm earlier reports stating that the cerebellar IP 3 receptor can be efficiently phosphorylated by PKA and PKG on two distinct sites in vitro. In addition, we show that both PKA and PKG can phosphorylate the cerebellar IP 3 receptor in situ on the same two sites, albeit with distinct time courses and kinetics.
Some of these data have been reported in abstract form (21,32).
Preparation of Peptides-A synthetic peptide comprising the carboxyl-terminal 18 amino acid residues of the mouse IP 3 receptor (33), used to raise antibodies to the IP 3 receptor (6,7,34), and two 10-amino acid peptides (peptide 1, with the sequence AARRDSVLAA, corresponding to residues 1584 -1593, and peptide 2, SGRRESLTSF, corresponding to residues 1751-1760) were synthesized at the Biotechnology Center, University of Oslo. A 30-amino acid synthetic peptide, termed D32-((Ser-34)8 -38), encompassing amino acids 8 -38 from DARPP-32 (dopamine-and cAMP-regulated phosphoprotein, 32 kDa) in which the serine residue at position 34 is efficiently phosphorylated by both PKA and PKG (35,36) Immunopurification of the IP 3 Receptor for in Vitro Phosphorylation Assays-The IP 3 receptor was isolated from rat cerebellum homogenized in 10 volumes of ice-cold buffer containing 0.32 M sucrose, 20 mM EPPS (pH 8.5), 1 mM EDTA, and the protease inhibitors phenylmethylsulfonyl fluoride (0.1 mM) and leupeptin (10 g/ml). Centrifugation and extraction with 1% (w/v) deoxycholate was performed as described previously (34). The resulting supernatant (containing approximately 5 mg/ml protein) is referred to as the deoxycholate extract. Protein A-Sepharose beads were washed and incubated with anti-IP 3 receptor antiserum and subsequently with deoxycholate extract (diluted 1:3 in TBS-Tween) as described (34), and the final immunocomplexes were isolated by centrifugation and washing of the beads.
Phosphorylation of the IP 3 Receptor-Immunopurified IP 3 receptor bound to protein A-Sepharose was phosphorylated in the presence of exogenous protein kinases. Phosphorylation with the catalytic subunit of PKA was performed at 30°C in 20 mM Tris-HCl (pH 7.4), 10 mM MgCl 2 , 1 mM EGTA, 10 M 8-Br-cAMP, the catalytic subunit of PKA (0.5 M), and phosphatase inhibitors (including 6 mM p-nitrophenylphosphate, 12 mM ␤-glycerophosphate, and 0.02 mM sodium vanadate). Phosphorylation by PKG was performed in the same buffer but with 10 M 8-Br-cGMP instead of 8-Br-cAMP, with the purified holoenzyme of PKG (0.16 M) instead of PKA, and with addition of 10 M peptide inhibitor of PKA (Walsh inhibitor). Phosphorylation was initiated by addition of [␥-32 P]ATP (final concentration, 8.5 M, containing 12 Ci/ mmol) and terminated after various incubation times (1-20 min) by addition of 40 mM EDTA and washing twice in TBS-Tween. Following boiling of the protein A-Sepharose beads in SDS-containing stop solution (37), proteins were separated by SDS-PAGE, the protein content of Coomassie-stained gel bands was analyzed densitometrically, and phosphoproteins were visualized by autoradiography. For quantitation, bands of interest were located with the autoradiograms as guides and excised from the gels. Radioactivity was measured by Cerenkov counting.
For analysis of sequential addition of the two cyclic nucleotide-dependent protein kinases, initial phosphorylation of the immunoisolated IP 3 receptor was performed as described above, but employing nonradioactive ATP (8.5 M) for 20 min. Following washing of the beads (twice) in TBS-Tween, the second incubation was initiated by addition of exogenous kinase and [␥-32 P]ATP (8.5 M, 12 Ci/mmol). The reaction was terminated after 5 min by EDTA, and the samples were washed and analyzed by SDS-PAGE as described above.
Thermolytic Digestion and Phosphopeptide Mapping of 32 P-Labeled Protein-Following SDS-PAGE, the 32 P-labeled gel bands containing the IP 3 receptor were detected by autoradiography, excised from the gels, washed, and subjected to proteolysis with thermolysin (1 mg/ml) for 24 h at 37°C. The thermolytic phosphopeptides were separated by electrophoresis and ascending chromatography as described (38), and the resulting phosphopeptide maps were visualized by autoradiography.
Preparation and Incubation of Rat Cerebellar Slices-Wistar rats (100 -200 g) were sacrificed with halothane, and the cerebellum was quickly removed and cooled in artificial cerebrospinal fluid (ACSF) at 0 -4°C of the following composition: NaCl, 124 mM; KCl, 2 mM; KH 2 PO 4 , 1.25 mM; MgSO 4 , 2 mM; NaHCO 3 , 26 mM; glucose, 10 mM; bubbled with 95% O 2 /5% CO 2 (pH 7.4). The hemispheres were glued to a mounting block, and sagittal slices (400 m) were cut with a vibroslice in cold oxygenated ACSF. The slices were placed in an interface chamber exposed to humidified gas and maintained in ACSF at a temperature of 25-27°C for at least 2 h. After equilibration, the slices were carefully transferred to wells (12-well cell culture cluster, Costar) containing 500 l of ACSF, where incubations with different reagents were performed in an O 2 -enriched atmosphere at room temperature. The slices were continuously kept in calcium-free solutions to diminish disturbances induced by calcium influx, e.g. proteolysis by calpains (34), protein kinase C or Ca 2ϩ /calmodulin-dependent protein kinase II phosphorylation (12)(13)(14)(15)30), and protein phosphatase 2B dephosphorylation (40). Incubations were terminated by removal of the reaction buffer and addition of 400 l of ice-cold buffer (containing 0.25 M sucrose, 20 mM Tris-HCl (pH 7.4), 2 mM EDTA, 0.1 mM phenylmethylsulfonyl fluoride, 10 g/ml leupeptin, and the phosphatase inhibitors ␤-glycerophosphate (60 mM), sodium vanadate (0.1 mM) and sodium fluoride (0.1 mM)) and homogenized by sonication. Following centrifugation at 1,000 ϫ g for 15 min at 4°C, deoxycholate was added to the supernatant (1% w/v, final concentration), and extraction on ice for 30 -45 min was followed by centrifugation for 30 min at 27,000 ϫ g. The resulting supernatant was incubated with protein A-Sepharose beads preincubated with antiserum to immunoisolate the IP 3 receptor as described above.
In Situ Phosphorylation of the IP 3 Receptor in Rat Cerebellar Slices-Cerebellar slices were incubated with 625 Ci of [ 32 P]orthophosphate and cyclic nucleotide analogues (final concentration, 3.2 mM 8-pCPT-cAMP or 1.6 mM 8-pCPT-cGMP) after equilibration in phosphate-free ACSF. The specific PKA inhibitor KT 5720 (final concentration, 2.5 M) or the PKG inhibitor KT 5823 (final concentration, 1.6 M) was present during the [ 32 P]orthophosphate incubation of control slices and was added to the other slices 20 min prior to addition of cyclic nucleotide analogues (20). Reactions were terminated by homogenization of the slices as described above, and immunoisolation of the IP 3 receptor was followed by SDS-PAGE. Incorporation of 32 P into the IP 3 receptor was visualized by autoradiography and quantified by densitometry and Cerenkov counting. The phosphoprotein was further characterized by thermolytic peptide mapping as described above.
For the back-phosphorylation experiments, immunopurified samples from cerebellar slices that had been incubated with protein kinase inhibitors and cyclic nucleotide analogues in phosphate-containing ACSF as described above were in vitro phosphorylated at room temperature for 20 min using exogenous PKA and [␥-32 P]ATP. The reactions were terminated by EDTA and washing in TBS-Tween, followed by SDS-PAGE, as described above.
Miscellaneous Methods-SDS-PAGE was performed using the buffers of Laemmli (37). Protein content was analyzed by the bicinchoninic acid method (41). When phosphorylation stoichiometry was examined, the amount of IP 3 receptor in the sample was quantified by densitometry of the gels following staining with Coomassie Blue, using standard amounts of bovine serum albumin in the same gels to construct standard curves. Images of autoradiograms were prepared using a Hewlett-Packard ScanJet IIC/ADF scanner, and Desk Scan II (version 1.0, Hewlett-Packard) together with Adobe Photoshop 2.5 or Corel 4.0 software. Statistical analysis and Hanes-Woolf-plots were obtained using Graph-Pad Prism 2.01.

Phosphorylation of the Cerebellar IP 3 Receptor in Vitro-
Previous work has shown that the ϳ260 kDa (apparent from SDS-PAGE) protein band phosphorylated by PKA in cerebellum mostly represents the IP 3 receptor (8,34,42). When the IP 3 receptor, extracted from rat cerebellar membranes and immunoprecipitated with rabbit antibodies raised against the carboxyl-terminal end of the brain type I IP 3 receptor, was used as substrate for PKA and PKG in vitro (Fig. 1A), we found similar phosphorylation reactions. PKA rapidly phosphorylated the immunoisolated IP 3 receptor during 5-min incubations (Fig. 1B); longer incubation produced only marginal additional effects (not shown). PKG phosphorylated the immunoisolated IP 3 receptor more slowly and to a lower maximum level (Fig. 1B).
To estimate the extent of phosphorylation, 0.60 g of IP 3 receptor was phosphorylated for 5 min. Employing the predicted molecular mass of 313 kDa for the immunopurified receptor subunit (33), our results indicated that PKA phosphorylated the immunoisolated IP 3 receptor to a stoichiometry of 1.02 mol of phosphate per mol of protein, whereas PKG phosphorylated the IP 3 receptor to a stoichiometry of 0.48 mol of phosphate per mol of protein under these conditions (see also Fig. 4, a and b).
Site-specific Phosphorylation of the Immunopurified IP 3 Receptor-The phosphorylated domains of the IP 3 receptor were characterized by phosphopeptide mapping. When the phosphorylated protein was subjected to extensive thermolytic digestion and peptide separation in two dimensions on silica plates, two major (phosphopeptides 1 and 2) and two minor phos-phopeptides were seen following incubation with PKA ( Fig.  2A). In contrast, incubation with PKG resulted in one major and one minor phosphopeptide only; the former comigrated with phosphopeptide 2, and the latter comigrated with phosphopeptide 1 (Fig. 2B). Quantitation of the 32 P-labeled phosphopeptides showed that PKA induced a rapid phosphorylation of both the major phosphorylation sites and that these two phosphorylation sites were phosphorylated to the same extent during the different incubation times (not shown). PKG induced a rapid phosphorylation of site 2 but a more protracted and less complete phosphorylation of site 1 as a function of time (Fig. 3). This is in accordance with the time course for the phosphorylation of the isolated protein, shown in Fig. 1B.
Additive Phosphorylation of the Immunopurified IP 3 Receptor-The relation between the serine residues phosphorylated by PKA and PKG was further examined by phosphorylating the isolated IP 3 receptor with sequential addition of the two kinases. Incubation with nonradioactive ATP in the presence of PKA for 20 min prevented the protein from subsequent phosphorylation with [␥-32 P]ATP in the presence of PKG (Fig. 4). In contrast, incubation of the IP 3 receptor with nonradioactive ATP and PKG for 20 min allowed the protein to become further phosphorylated by subsequent addition of [␥-32 P]ATP in the presence of PKA, with the final 32 P labeling representing approximately half of that obtained by PKA alone (Fig. 4). Thermolytic phosphopeptide mapping of this sample showed that incubation with nonradioactive ATP and PKG led to a considerable decrease in the subsequent PKA-catalyzed 32 P labeling of phosphorylation site 2, but not site 1 (Fig. 2C).
Taken together, the phosphopeptide mapping of the IP 3 receptor indicates that PKA catalyzed in vitro phosphorylation of at least two residues within 5 min, whereas PKG catalyzed phosphorylation of one residue within the first minute, with a subsequent and slower phosphorylation of the second residue (Fig. 3). Moreover, the residues phosphorylated by PKA and PKG are identical.
Phosphorylation of Synthetic IP 3 Receptor Peptides-PKA has been shown to phosphorylate Ser-1756 and, less efficiently, Ser-1589, in the neuronal IP 3 receptor in vitro (9). To confirm that the same two serines are phosphorylated by PKG and to study the kinetics and phosphorylation efficiency of these in vitro phosphorylation reactions, we used two synthetic decapeptides as substrates for PKA and PKG: peptide 1, encompassing Ser-1589 (AARRDSVLAA), and peptide 2, encompassing Ser-1756 (SGRRESLTSF) (10,19). The results were compared with those obtained with a 30-amino acid peptide derived from DARPP-32, D32-((Ser-34)8 -38), which is known to represent a good substrate for both cyclic nucleotide-depend- ent kinases (36). The reactions followed Michaelis-Menten kinetics at low substrate concentrations, but exhibited substrate inhibition at higher substrate concentrations (not shown). Hanes-Woolf analysis indicated that peptide 1, containing Ser-1589, was phosphorylated with the highest V max by both PKG and PKA (Table I). In contrast, peptide 2, containing Ser-1756, was phosphorylated by both PKA and PKG with relatively low V max , but with a lower K m , indicating that both kinases have higher affinity for peptide 2 in vitro. PKA, which phosphorylated sites 1 and 2 of the immunopurified protein to a similar extent ( Fig. 2A), phosphorylated the synthetic peptides with almost equal catalytic efficiencies (K cat /K m ; Table I). PKG, which phosphorylated site 2 of the immunopurified protein more extensively than site 1 (Figs. 2B and 3), phosphorylated the synthetic peptide 2 with a higher catalytic efficiency than the synthetic peptide 1 ( Table I). The kinetic data were thus in agreement with the in vitro phosphorylation of the intact IP 3 receptor. The DARPP-32 peptide showed kinetic properties similar to those of peptide 2 for both PKA and PKG (Table I).
Phosphorylation of IP 3 Receptor in Situ in Rat Cerebellar Slices-It was of interest to ascertain that these kinases are able to phosphorylate the IP 3 receptor in neurons in situ. Because the IP 3 receptor is specifically abundant in Purkinje neurons of the rat cerebellum (2), we utilized cerebellar slice preparations to study these neurons. The slices were incubated for 2-2.5 h in ACSF before stimulation to ensure equilibration of physiological processes (43). Preparation and incubation of slices were in accordance with previously published techniques for electrophysiological slice experiments (44).
Incorporation of [ 32 P]Orthophosphate into Stimulated Slices from Rat Cerebellum-Rat cerebellar slices were preincubated in [ 32 P]orthophosphate, followed by addition of 8-pCPT-cGMP or 8-pCPT-cAMP together with the specific PKA and PKG inhibitors, KT 5720 and KT 5823, respectively. Immunopurification of the IP 3 receptor and quantitation of the radioactive  gel bands revealed a considerable increase in phosphate incorporation (Fig. 5a) in stimulated slices versus control slices. Phosphopeptide mapping showed the extent to which both the phosphopeptides were phosphorylated (Fig. 5b). In slices treated with PKA inhibitor alone (Fig. 5b, A), the IP 3 receptor was weakly phosphorylated. Stimulation of the slices with 8-pCPT-cAMP resulted in preferential incorporation of [ 32 P]orthophosphate in phosphopeptide 2 (Ser-1756; Fig. 5b, B). However, as opposed to the in vitro observations, stimulation of slices with 8-pCPT-cGMP led to preferential incorporation of 32 P in phosphopeptide 1 (Ser-1589; Fig. 5b, C).
Back-phosphorylation of IP 3 Receptor from Stimulated Cerebellar Slices-Slices of rat cerebellum (equilibrated for 2.5 h in phosphate-containing ACSF) were stimulated for 10 min with 8-pCPT-cGMP or 8-pCPT-cAMP in the presence of the specific inhibitors of PKA and PKG, respectively. Following homogenization and immunopurification, the IP 3 receptor was backphosphorylated in vitro with PKA and [␥-32 P]ATP. The decrease in radioactivity seen when back-phosphorylated receptor is compared with controls ( Fig. 6) represents the increase in in situ phosphorylation induced by the pharmacological agents applied. Stimulation with 8-pCPT-cGMP reduced back-phosphorylation to 75.5 Ϯ 15.5% of control (p ϭ 0.029, Mann-Whitney U test), 2 and stimulation with 8-pCPT-cAMP reduced the back-phosphorylation to 75.3 Ϯ 8.1% of control (p ϭ 0.012, Student's t test).

DISCUSSION
Phosphorylation of the IP 3 receptor has been observed in different cell types, involving different signaling systems and protein kinases. In the present work, we show that both PKA and PKG mediate in vitro phosphorylation of at least two sites in the neuronal IP 3 receptor immunopurified from rat cerebellum. When the intact neuronal IP 3 receptor is phosphorylated in vitro, low concentrations of PKA label Ser-1756, whereas high levels of PKA label both Ser-1756 and Ser-1589 (9). In vitro phosphorylation by PKG appears to mimic the effect of low PKA, because a strong preference for Ser-1756 is observed (20). In contrast, the smooth muscle IP 3 receptor is phosphorylated by PKA in vitro only on Ser-1589 (10). The smooth muscle G 0 protein, later reported to represent an IP 3 receptor (23), is preferentially phosphorylated by endogenous PKG in vitro on a phosphopeptide comigrating on phosphopeptide maps with the IP 3 receptor phosphopeptide containing Ser-1589 (25). A proposed cause for the differential phosphorylation between neuronal and smooth muscle IP 3 receptors is the 40amino acid stretch comprising the alternatively spliced residues 1693-1732, which are not found in the nonneuronal IP 3 receptor subtype from smooth muscle (SIIϪ).
Under the experimental conditions used in this work, we have found that the two phosphorylation sites showed distinct kinetic features when their in vitro phosphorylation was studied by the use of synthetic peptide substrates. Phosphorylation of synthetic decapeptides containing the putative phosphorylation sites for PKA and PKG showed that both enzymes display higher values for V max and K cat for the Ser-1589-containing peptide 1 than for the Ser-1756-containing peptide 2 (Table I).

FIG. 5. Incorporation of [ 32 P]orthophosphate into IP 3 receptor in rat cerebellar slices.
Rat cerebellar slices preincubated with [ 32 P]orthophosphate for 90 min and the specific protein kinase inhibitors KT 5720 (PKA inhibitor) and KT 5823 (PKG inhibitor) for 20 min were stimulated to induce phosphorylation of the IP 3 receptor. Following stimulation, the slices were homogenized, the IP 3 receptor was immunoisolated, and SDS-PAGE and phosphopeptide mapping were performed as described under "Experimental Procedures." The experiment was performed three times with similar results. a, bar graph representing incorporated radioactivity per amount of protein in the IP 3 receptor gel band. Each bar represents the mean Ϯ S.D. of two slices, relative to the control slices. A, control slices, incubated only with KT 5720 throughout the 32 P prelabeling; B, slices incubated with KT 5823 followed by 8-pCPT-cAMP for 10 min; C, slices incubated with KT 5720 followed by 8-pCPT-cGMP for 10 min. b, phosphopeptide mapping of the gel bands illlustrated in a. Phosphopeptides (1 and 2) and application points (q) are indicated.
Despite the lower maximal rate of phosphorylation, PKG had approximately 4-fold higher affinity for peptide 2 than peptide 1, whereas PKA exhibited 1.5-fold higher affinity for peptide 2 compared with peptide 1. Hence, the catalytic efficiency (K cat / K m ) of PKG was higher for peptide 2, indicating a preference for Ser-1756 by this kinase (Table I), which is in agreement with earlier results (19). However, the differences in catalytic efficiency were small, and both IP 3 receptor-derived peptides appeared to be phosphorylated by the two kinases with a higher catalytic efficiency than D32-((Ser-34)8 -38) ( Table I). The latter peptide represents the phosphorylation domain of DARPP-32, a protein that is a substrate for both PKA (35) and PKG (45) in situ. Thus, from these data we cannot exclude any of these IP 3 receptor peptides as possible targets for cyclic nucleotideactivated protein kinases in vivo.
Our in situ data appear to contradict the report from Komalavilas and Lincoln (19), in which Ser-1589 was proposed not to represent a physiological substrate site for PKG. Our experiments employing [ 32 P]orthophosphate-incorporated cerebellar slices, which were stimulated with a highly specific cGMP analogue, showed that the IP 3 receptor was phosphorylated mostly on Ser-1589 in situ (Fig. 5), whereas a cAMP analogue mainly produced the Ser-1756 phosphopeptide. This difference in the phosphorylation pattern, which apparently is not caused by distinct IP 3 receptor subtypes, may be due to the properties of the protein kinases and their localization in the Purkinje cells (46). These results emphasize the care needed when extrapolation from in vitro experiments is used to identify physiologically important processes.
Further evidence for in situ phosphorylation of the IP 3 receptor in Purkinje cells came from back-phosphorylation experiments. Following both cAMP and cGMP analogue stimulation, such experiments decreased the subsequent in vitro phosphorylation induced by PKA by 25%. Because the IP 3 receptor was extracted and isolated in the presence of phosphatase inhibitors, these results indicate that the various treatments had decreased the amount of dephospho-IP 3 receptor. Hence, although phosphorylation/dephosphorylation reactions are rapid and dynamic processes, and the stimulation with cyclic nucle-otide analogues took place without phosphatase inhibitors present, these results demonstrate notable changes in IP 3 receptor phosphorylation caused by in situ stimulation.
Recent work has provided increasing knowledge about the functional consequences of IP 3 receptor phosphorylation. Cyclic nucleotides mediate regulation of intracellular Ca 2ϩ levels, e.g. by phosphorylation of Ca 2ϩ channels, Ca 2ϩ transporters, and regulatory proteins (47,48). Phosphorylation of the IP 3 receptor by PKA regulates the sensitivity of the IP 3 receptor to IP 3 (49), although the effect of phosphorylation remains uncertain: in microsomal membranes from rat brain and human platelets, cAMP-dependent phosphorylation has been reported to inhibit Ca 2ϩ release (29,30,40,50,51), whereas phosphorylation of purified rat brain IP 3 receptor increases Ca 2ϩ flux in reconstituted lipid vesicles (52). Increased Ca 2ϩ flux is seen when IP 3 receptor is phosphorylated by PKA in permeabilized SH-SY5Y human neuroblastoma cells (12), and Ca 2ϩ release and oscillations are stimulated by cAMP and IP 3 in rat hepatocytes (28). Furthermore, Ca 2ϩ /calmodulin-dependent protein kinase II and protein kinase C phosphorylation of the IP 3 receptor are reported to increase Ca 2ϩ mobilization in rat brain, liver nuclei, and fibroblasts (15,40,53), whereas endogenous kinases increase Ca 2ϩ release through IP 3 receptors in platelets (30), and the tyrosine kinase Fyn increases the open probability of IP 3 receptors in T-lymphocytes (17).
Multiple mechanisms may also mediate the effects of cGMP on intracellular Ca 2ϩ (54), including cGMP-mediated regulation of smooth muscle plasma membrane Ca 2ϩ -ATPase (23) and PKG-induced decreases in agonist-induced IP 3 generation (55). The consequences of PKG-phosphorylation of the IP 3 receptor are not well defined, although in rat and guinea pig hepatocytes, cGMP potentiates Ca 2ϩ release and induces oscillations (26,27). PKG also appears to inhibit IP 3 receptor activity in platelets (29), similar to the effects of PKA. However, functional effects of IP 3 receptor phosphorylation in brain remain uncertain, because both decreased and increased Ca 2ϩ release have been observed (12,50,52).
In the cerebellum, the Purkinje cells contain large amounts of both IP 3 receptor and PKG (8,46). The effect of conjunction of parallel fiber activation and climbing fiber-induced depolarization in Purkinje cells is a long lasting synaptic depression (LTD). Recent studies have proposed that climbing fiber activity may induce calcium release from intracellular stores via IP 3 receptors, which in turn may sensitize IP 3 receptors to the IP 3 released as a consequence of parallel fiber activation (56,57). Induction of LTD is prevented by a specific, blocking antibody to the IP 3 receptor, and LTD fails to develop in mice with a disrupted IP 3 receptor type I gene (5). One underlying molecular mechanism whereby climbing fiber-induced nitric oxide release (58), the resulting generation of cGMP, and the activation of PKG may regulate LTD in these synapses is the phosphorylation of the IP 3 receptor.
FIG. 6. Back-phosphorylation of IP 3 receptor from cerebellar slices stimulated in situ. Rat cerebellar slices were preincubated in nonradioactive, phosphate-containing ACSF for 2.5 h, followed by incubation with the PKG inhibitor KT 5823 (control) (a), KT 5823 for 20 min and 8-pCPT-cAMP for 10 min (b), the PKA inhibitor KT 5720 (control) (c), KT 5720 for 20 min and 8-pCPT-cGMP for 10 min (d). Homogenization of the slices and immunopurification of the IP 3 receptor was followed by in vitro phosphorylation with exogenous PKA and [␥-32 P]ATP for 20 min. Termination of the reaction, SDS-PAGE, and quantitation of the IP 3 receptor was performed as described. The bars represent mean Ϯ S.D. of four slices; the experiment was performed three times with similar results. *, significantly different from control, p Ͻ 0.05 (Student's t test); ૽, p Ͻ 0.05 (Mann-Whitney U test).