Proc. Nati. Acad. Sci. USA Vol. 88, pp. 2232-2235, March 1991 Biochemistry trisphosphate receptor: Phosphorylation by protein C and -dependent protein in reconstituted vesicles (cAMP/inositol turnover/diacylglycerol/serine/threonine) CHRISTOPHER D. FERRIS*, RICHARD L. HUGANIR*t, DAVID S. BREDTt, ANDREW M. CAMERON*, AND SOLOMON H. SNYDER*t§¶ Departments of *Neuroscience, tPharmacology and Molecular Sciences, and §Psychiatry and Behavioral Sciences, and tThe Howard Hughes Medical Institute, Johns Hopkins University School of Medicine, Baltimore, MD 21205 Contributed by Solomon H. Snyder, December 19, 1990

ABSTRACT We have previously demonstrated that the concentrations were 10 mM for MgCl2 and 50 ,uM for ATP. inositol 1,4,5-trisphosphate (IP3) receptor is phosphorylated by The final concentrations of phosphorylating enzymes were 2 cyclic AMP-dependent protein kinase (PKA). In the present pug/ml, 3 ug/ml, and 1 pug/ml for PKC, CaM kinase II, and study, phosphorylation of IP3 receptors has been examined catalytic subunit ofPKA, respectively. For both CaM kinase with purified receptor protein reconstituted in liposomes to II and PKC phosphorylation, 1 mM CaCl2 was added, while remove detergent that can inhibit protein kinases. The IP3 a final calmodulin concentration of 0.05 mg/ml was present receptor is stoichiometrically phosphorylated by protein kinase for CaM kinase II phosphorylation. Phosphorylated amino C (PKC) and Ca2' calmodulin-dependent protein kinase H acids and phosphopeptide maps were determined as de- (CaM kinase H) as well as by PKA. Phosphorylation by the scribed (8). three enzymes is additive and involves different peptide se- quences. Phosphorylation by PKC, which is stimulated by RESULTS Ca2+ and diacylglycerol, and by CaM kinase H, which requires Ca2+, provides means whereby Ca2+ and diacylglycerol, In a previous study ofthe phosphorylation ofthe IP3 receptor formed during inositol phospholipid turnover, may regulate by cAMP-dependent protein kinase, we reported that the IP3 receptor physiology. purified IP3 receptor protein was not phosphorylated by PKC or CaM kinase 11 (5). In addition, before identification of the The inositol phospholipid acts IP3 receptor protein, Greengard and coworkers (9, 10) had through the formation of inositol 1,4,5-trisphosphate (IP3), described a phosphorylated protein in Purkinje cells of the which triggers the release of calcium from portions of the , designated PCPP-260, which was phosphoryl- (1, 2). The purified IP3 receptor ated by PKA and which we have shown to be identical to the protein (3), reconstituted into lipid vesicles, mediates IP3 IP3 receptor (S. Supattapone and S.H.S., unpublished data). alterations of Ca2" flux, indicating that the protein contains PCPP-260 did not appear to be phosphorylated by CaM both the IP3 recognition site and the Ca2' ion channel (4). The kinase II (9, 10). In these earlier studies, the proteins were IP3 receptor is stoichiometrically phosphorylated by cAMP- purified in the presence of detergents, which were not dependent protein kinase (PKA), resulting in diminished removed from the purified preparation for phosphorylation potency of IP3 in releasing Ca2` from brain membranes (5). experiments. Many detergents inhibit protein kinases and This provides a mechanism for "cross-talk" between the may also denature substrates. Recently, we have reconsti- cAMP and inositol phospholipid systems. Utilizing reconsti- tuted the IP3 receptor protein into lipid vesicles (4) in which tuted lipid vesicles, we now demonstrate specific, selective, the detergent is completely removed, and the receptor is and stoichiometric phosphorylation of the IP3 receptor by restored to a lipid environment. Accordingly, we have reex- protein kinase C (PKC) and Ca2' calmodulin-dependent amined the phosphorylation of IP3 receptors by various protein kinase II (CaM kinase II). protein kinase in these vesicles (Fig. 1, Table 1). Purified and reconstituted IP3 receptor protein is phospho- MATERIALS AND METHODS rylated by PKC, CaM kinase II, and PKA. SDS/PAGE analysis reveals a single phosphorylated protein band of 260 Materials. [32P]ATP was purchased from Du Pont/NEN. kDa with all three phosphorylating enzymes (data not shown). CaM kinase II was the generous gift of Mary Kennedy and All three enzymes phosphorylate the IP3 receptor stoichio- Bruce Patton (California Institute of Technology, Pasadena, metrically (Table 1). At 300C, the time course of phosphory- CA). Trypsin and thermolysin were obtained from Boehr- lation appears somewhat different for the three enzymes. CaM inger Mannheim. All other chemicals were purchased from kinase II produces about half-maximal phosphorylation at 10 Sigma. IP3 receptor was purified (3, 4) and reconstituted into min, slightly more rapid than PKA, while PKC requires 30 min lipid vesicles (4) as described. Catalytic subunit of protein for half-maximal phosphorylation (Fig. 1). kinase A was purified from bovine heart as described (6). We wondered whether the three phosphorylating enzymes PKC was purified from rat brain as described (7). act at the same or different sites. To examine this question, Phosphorylation ofPurified and Reconstituted IP3 Receptor. we measured the stoichiometry of phosphorylation in the Reconstituted proteoliposomes (50 ,ul unless otherwise indi- presence of mixtures of the enzymes. Phosphorylation ap- cated) were incubated at 30°C for 60 min or as indicated in a pears to be additive for the three enzymes (Table 1); in the final incubation vol of 100 ,ul. In all experiments, fnal Abbreviations: IP3, inositol 1,4,5-trisphosphate; PKA, cyclic AMP- The publication costs of this article were defrayed in part by page charge dependent protein kinase; PKC, protein kinase C; CaM kinase II, payment. This article must therefore be hereby marked "advertisement" calcium calmodulin-dependent protein kinase II. in accordance with 18 U.S.C. §1734 solely to indicate this fact. ITo whom reprint requests should be addressed. 2232 Downloaded by guest on September 28, 2021 Biochemistry: Ferris et al. Proc. Natl. Acad. Sci. USA 88 (1991) 2233

* - Ser -- - Ser - o Cr ~~~~PKA - Thr - - Thr - - Tyr - Tyr - E~0.6- PKC _5 0.48 E 0.2- * -- 0r -- - Or - as PKA PKC CaM 10 20 30 40 50 60 kinase 11 Time(min) FIG. 2. Determination of phosphorylated amino acids for PKA, FIG. 1. Time course of phosphorylation of purified and recon- PKC, and CaM kinase II phosphorylation of purified and reconsti- stituted IP3 receptor by PKA (catalytic subunit), PKC, and CaM tuted IP3 receptor. Phosphorylation ofIP3 receptor was performed as kinase II. Phosphorylation reactions were performed as described. described. After SDS/PAGE, the IP3 receptor was cut out of the gel At the indicated times, phosphorylation was stopped by the addition and digested by thermolysin as described in the legend to Fig. 3 and of 50 1fl of 3x sample buffer for SDS/PAGE. All samples were ref. 7, and the peptides were hydrolyzed to amino acids by incubation electrophoresed on 7.5% polyacrylamide gels. The gels were stained in 6 M HCI at 105°C as described (7). This experiment was repeated with Coomassie brilliant blue, the IP3 receptor band was cut out, and with the same result. Or, origin. radioactivity was determined by Cerenkov counting ofthe gel bands. These experiments have been repeated three times with essentially toward the cathode slightly ahead ofthe marker basic fuchsin the same results. (Fig. 3A). PKC produces one major phosphopeptide that migrates partially to the anode (Fig. 3B). A minor product of presence of all three enzymes, the extent of phosphorylation is -3 times greater than with any of the enzymes alone. The A specificity of the three enzymes is ensured by experiments showing that PKC activity is eliminated in the absence of I Ca2", while CaM kinase II activity is dependent on both Ca2l >. and calmodulin. Confirming our earlier observations, PKA 0cr phosphorylation of the IP3 receptor is blocked by the Walsh 0 a peptide inhibitor of this enzyme (5). E II can on either 0 PKC, PKA, and CaM kinase phosphorylate f- serine or threonine. To distinguish these alternatives, we 0 analyzed by electrophoresis phosphorylated amino acids 0 after acid of phosphopeptides (Fig. 2). With all three enzymes, a single major is phosphorylated, which corresponds to phosphoserine. In the case of CaM kinase II phosphorylation, some phosphothreonine is pres- Electrophoresis ent, although it represents <5% of the total incorporated phosphate. B To identify phosphorylation sites of these enzymes, we performed two-dimensional phosphopeptide analysis (Figs. 3 and 4). The IP3 receptor, phosphorylated by PKA, PKC, or CI CaM kinase II, was digested with thermolysin or trypsin and 0 0c the resultant phosphopeptides were analyzed by electropho- 0 resis and thin-layer chromatography in two dimensions. With 0 E PKA, we observe a single basic phosphopeptide that migrates 0 Table 1. Additive phosphorylation of IP3 receptor by PKA, PKC, and CaM kinase II 32p/IP3 receptor, Phosphorylating condition mol/mol _ Electrophoresis PKA 1.0 + Walsh inhibitor <0.1 FIG. 3. Two-dimensional phosphopeptide map of IP3 receptor PKC 0.9 phosphorylated by PKA (A) and PKC (B). IP3 receptor was phos- - Ca2+ <0.1 phorylated as described. After SDS/PAGE, the IP3 receptor band CaM kinase II 1.25 was cut out ofthe gel and digested by incubating the gel band at 30°C - Ca2+ <0.1 for 20 hr in a final vol of 1 ml in the presence of thermolysin (300 - Calmodulin <0.1 ,ug/ml) and 50 mM NH4HCO3. After lyophilization, the peptides were separated by electrophoresis and chromatography as described PKA + PKC + CaM kinase II 2.7 (7). Phenol red and basic fuschin were used as negative and positive Specific and stoichiometric phosphorylation of purified and re- markers, respectively. Electrophoresis was carried out at 500 V until constituted IP3 receptor by PKA (catalytic subunit), PKC, and CaM the markers had migrated 6 cm from the origin. Chromatography was kinase II. Phosphorylation was performed as described. The incu- done as described (7) and was stopped when the markers had bation was for 60 min at 30"C. This experiment was replicated with migrated 6 cm. These experiments have been repeated three times

13. Yamamoto, H., Maeda, N., Niinobe, M., Miyamoto, E. & tapone, S. & Snyder, S. H. (1989) Nature (London) 339, Mikoshiba, K. (1989) J. Neurochem. 53, 917-923. 468-470. 14. Worley, P. F., Baraban, J. M., Colvin, J. S. & Snyder, S. H. 19. Satoh, T., Ross, C. A., Villa, A., Supattapone, S., Pozzan, T., (1987) Nature (London) 325, 159-161. Snyder, S. H. & Meldolesi, J. (1990) J. CellBiol. 111, 615-624. 15. Ouimet, C. C., McGuiness, T. L. & Greengard, P. (1984) Proc. 20. C. D., Huganir, R. L. & Snyder, S. H. (1990) Proc. Natl. Acad. Sci. USA 81, 5604-5608. Ferris, 16. Walaas, S. I., Lai, Y., Gorelick, F. S., DeCamilli, P., Moretti, Natl. Acad. Sci. USA 87, 2147-2151. M. & Greengard, P. (1988) Brain Res. 464, 233-242. 21. Worley, P. F., Baraban, J. M., Supattapone, S. & Snyder, 17. Fukunaga, K., Goto, S. & Miyamoto, E. (1988) J. Neurochem. S. H. (1987) J. Biol. Chem. 262, 12132-12136. 51, 1070-1078. 22. Danoff, S. K., Supattapone, S. & Snyder, S. H. (1988) Bio- 18. Ross, C. A., Meldolesi, J., Milner, T. A., Satoh, T., Supat- chem. J. 254, 701-705. Downloaded by guest on September 28, 2021