Differential Effects of a Calcineurin Inhibitor on Glutamate-induced Phosphorylation of Ca2+/Calmodulin-dependent Protein Kinases in Cultured Rat Hippocampal Neurons*

Calcium/calmodulin-dependent protein kinases (CaM kinases) are major multifunctional enzymes that play important roles in calcium-mediated signal transduction. To characterize their regulatory mechanisms in neurons, we compared glutamate-induced phosphorylation of CaM kinase IV and CaM kinase II in cultured rat hippocampal neurons. We observed that dephosphorylation of these kinases followed different time courses, suggesting different regulatory mechanisms for each kinase. Okadaic acid, an inhibitor of protein phosphatase (PP) 1 and PP2A, increased the phosphorylation of both kinases. In contrast, cyclosporin A, an inhibitor of calcineurin, showed different effects: the phosphorylation and activity of CaM kinase IV were significantly increased with this inhibitor, but those of CaM kinase II were not significantly increased. Cyclosporin A treatment of neurons increased phosphorylation of Thr196 of CaM kinase IV, the activated form with CaM kinase kinase, which was recognized with an anti-phospho-Thr196 antibody. Moreover, recombinant CaM kinase IV was dephosphorylated and inactivated with calcineurin as well as with PP1, PP2A, and PP2C in vitro. These results suggest that CaM kinase IV, but not CaM kinase II, is directly regulated with calcineurin.

terized. It is abundantly expressed in the brain and plays an important functional role in hippocampal long-term potentiations (1)(2)(3)(4). After activation by Ca 2ϩ /CaM (5), CaM kinase II shows Ca 2ϩ -independent activity, which prolongs its activity even after intracellular Ca 2ϩ concentrations return to basal levels. CaM kinase IV is also abundant in the brain, including the cerebral cortex, the hippocampus, and particularly the nuclei of cerebellar granule cells (6 -10). Because of its expression in the latter, this enzyme is also called CaM kinase Gr (granule). In addition, CaM kinase IV is found in the thymus and testis (7), suggesting functional importance not only in the brain but in the immune system and reproductive cells. The genomic structure of CaM kinase IV is quite unique: both CaM kinase IV and calspermin, a male germ cell-specific CaM-binding protein, are derived from a single CaM kinase IV gene (7,11). CaM kinase IV has a broad substrate specificity and is thought to be multifunctional (8), as is CaM kinase II. Although less is known about the function of CaM kinase IV than that of CaM kinase II, recent work shows important biochemical and molecular differences between both kinases. CaM kinase IV functions as a monomer rather than an oligomer (8). CaM kinase IV requires phosphorylation at Thr 196 with another CaM kinase, CaM kinase kinase, for full activation (12)(13)(14)(15), rather than autophosphorylation, as typically observed in CaM kinase II. CaM kinase IV is localized in the nuclei of neurons (9), possibly associating with chromatin by the polyglutamate sequence in its C-terminal region, whereas the limited isoforms of CaM kinase II (␣ B and ␦ B isoforms) exist in the nuclei. The nuclear localization of CaM kinase IV suggests a role in the regulation of gene expression. CaM kinase IV can phosphorylate Ser 133 , the essential site for activation of CREB (16), which is essential for the activation of CRE-containing promoters. In an assay using cultured hippocampal neurons, it was reported that CREB phosphorylation induced by synaptic stimulations was mediated with CaM kinase IV (17). Moreover, Ca 2ϩ -induced expression of brain-derived neurotrophic factor that has a CRE motif in its promoter region (18) was reported to be mediated by CaM kinase IV (19), suggesting that Ca 2ϩ -and CRE-mediated gene expression is controlled with CaM kinase IV in vivo. Given the functional differences in these kinases, it is likely that CaM kinase IV and CaM kinase II are regulated by different mechanisms.
To characterize the differences between CaM kinase IV and CaM kinase II in the central nervous system and focus on the differences in the regulatory mechanisms of these kinases, we investigated glutamate-induced phosphorylation and the effects of protein phosphatase (PP) inhibitors, using primary cultured rat hippocampal pyramidal neurons. We also examined the susceptibility of recombinant CaM kinase IV to protein phosphatases. * This work was supported in part by Grants-in-Aid for Scientific Research and for Scientific Research on Priority from the Ministry of Education, Science, Sports and Culture of Japan and by a research grant from the Human Frontier Science Program (to E. M. and K. F.). 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 U.S.C. Section 1734 solely to indicate this fact.

EXPERIMENTAL PROCEDURES
Materials-The following chemicals and materials were obtained from the indicated sources: [␥- 32 Barnes et al. (20) in our laboratory; bovine purified calcineurin, UBI; and recombinant rat protein phosphatase 2C, a generous gift from Dr. S. Tamura (Tohoku University, Institute of Development, Aging and Cancer, Sendai, Japan) (21).
Preparation of Polyclonal Antibodies-The following peptides derived from rat CaM kinase IV were synthesized commercially by Fujiya and served as the immunogen: (a) the C terminus 10 amino acids (PQQDAILPEY), (b) Thr 196 -phosphorylated peptides that correspond to amino acids 184 -197 (LSKIVEHQVLMKT(p)V), and (c) a peptide corresponding to unphosphorylated amino acids 184 -197. These peptides (except for the latter) were conjugated to keyhole limpet hemocyanin, mixed with Freund's complete adjuvant, and injected into Japanese white rabbits. The methods used to prepare the polyclonal antibody (IgG fraction) were described previously (22). After repeated injection of the antigen, the anti-serum to each peptide was prepared and purified as the pellet of an ammonium sulfate cut. The anti-Thr 196 -phosphorylated CaM kinase IV antibody (anti-pT196) was purified using an affinity peptide column using the SulfoLink kit (Pierce) according to the manufacturer's protocol. Preparation of the anti-CaM kinase II polyclonal antibody was reported previously (22).
Cell Culture-Rat hippocampal pyramidal neurons were cultured as reported previously (23). Hippocampi were dissected from Wistar rats at postnatal day 1 and mechanically dispersed. Neurons were filtered through nylon mesh, plated onto collagen-coated dishes, and cultured in Eagle's Minimum Essential Medium containing 10% fetal calf serum. The medium was changed every 3 days.
Cell Stimulation with Glutamate-Rat hippocampal neurons cultured for 8 days were stimulated using the following protocol. Culture medium was aspirated, and the neurons were preincubated with KRH buffer (23) for 30 min and then stimulated with 10 M glutamate and 10 M glycine for indicated periods. In the experiments using the phosphatase inhibitors, 5 M cyclosporin A (CsA) was added throughout the preincubation and stimulation or 1 M okadaic acid was added for 15 min of the preincubation and throughout the stimulation. Cells were frozen in liquid nitrogen immediately after stimulation, harvested with homogenizing buffer (50 mM Tris-HCl, pH 7.5, 0.5 M NaCl, 0.5% Triton X-100, 10 mM EDTA, 4 mM EGTA, 1 mM Na 3 VO 4 , 30 mM Na 4 P 2 O 7 , 50 mM NaF, 0.1% SDS, 0.1 mM leupeptin, 0.075 mM pepstatin A, 0.05 mg/ml trypsin inhibitor, 1 mM phenylmethanesulfonyl fluoride, and 1 mM dithiothreitol), and homogenized with brief bursts of ultrasonication using a Branson Sonifier 250. The homogenate was centrifuged at 10,000 ϫ g for 10 min to exclude insoluble materials.
Immunoblot Analysis-The cell lysate, brain lysate, or fusion proteins were mixed with Laemmli's SDS-sample buffer (24) and boiled for 5 min. Samples were subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to a nitrocellulose membrane at 70 V for 3.5 h using a Trans Blot Cell (Bio-Rad) in a cold room (4°C). After blocking for 1 h at room temperature with TTBS solution containing 2.5% bovine serum albumin, the membranes were incubated overnight at 4°C with the first antibody diluted with TTBS and bovine serum albumin (1:1000 for the anti-CaM kinase IV C-terminal antibody and 1:5 for the anti-pT196 antibody). Bound antibodies were visualized with 125 I-protein A (0.1 mCi/ml) and analyzed by a Bio-Imaging Analyzer BA100 or FLA-2000 (Fuji film).
Labeling of Cells and Quantitation of Phosphorylation-Rat hippocampal neurons cultured for 8 days were washed once with phosphate-free and serum-free Dulbecco's modified Eagle's media and labeled in 1.0 ml of this medium containing carrier-free [ 32 P] orthophosphate (0.25 mCi/ml) for 5 h. The cells were then preincubated, stimulated, and homogenized as described above. The homogenate was precleared with 50 l of protein A-Sepharose CL-4B (50% v/v). Immunoprecipitation of CaM kinase IV and CaM kinase II was performed essentially as reported previously (23). Ten g of anti-CaM kinase IV C-terminal antibody and 50 l of protein A-Sepharose were mixed with the precleared homogenate at 4°C for 4 h. The immunocomplex was separated by centrifugation at 10,000 ϫ g for 2 min, washed four times with the homogenizing buffer, and then prepared as a sample for SDS-PAGE. The resulting supernatant was mixed with 5 g of the anti-CaM kinase II antibody and protein A-Sepharose for another 4 h and then prepared as a sample for SDS-PAGE. After SDS-PAGE, the gels were dried and analyzed by a Bio-Imaging Analyzer (FLA-2000; Fuji Film).
Assay for Immunoprecipitated CaM Kinase IV-Neurons were stimulated and homogenized, and the immunoprecipitation was performed with the anti-CaM kinase IV antibody as described above, except for the 32 P labeling procedures. After incubation for 4 h with the anti-CaM kinase IV antibody, the immunocomplex was washed four times with buffer containing of 50 mM HEPES, pH 7.5, 1 mM EDTA, and 1 mM dithiothreitol. Immunoprecipitated CaM kinase IV was assayed using peptide-␥ (8) (synthesized commercially by Fujiya) in 100 l of a total volume of the reaction mixture containing 50 mM HEPES (pH 7.5), 10 mM Mg(OAc) 2 , 0.5 mM CaCl 2 , 0.3 M CaM, 40 M peptide-␥, and 0.1 mM [␥-32 P]ATP (3000 -5000 cpm/pmol). After incubation at 30°C for 10 min with gentle tapping, reactions were terminated by centrifugation of the immunocomplex. Aliquots of the supernatants were spotted onto Whatman P81 P-cellulose paper. The papers were washed in 75 mM phosphoric acid buffer, and the radioactivity on the paper was measured by scintillation spectrometry.
All primers contained restriction sites (XhoI for CaM kinase IV ␣ primers and EcoRI for CaM kinase kinase ␣ primers). Polymerase chain reaction products were subcloned into plasmid pGEX5X-1 (Pharmacia) using a DNA ligation kit (Takara), sequenced to confirm that no mutation was introduced, and then transfected into Escherichia coli BL21 DE3. Expression and purification of the fusion proteins were performed using GST Purification Modules (Pharmacia) according to the manufacturer's protocols.
Phosphorylation of Fusion Proteins-Approximately 250 -500 g of recombinant CaM kinase IV were incubated at 30°C for 60 min with approximately 10 g of recombinant CaM kinase kinase in buffer containing 50 mM HEPES (pH 7.5), 10 mM Mg(OAc) 2 , 0.5 mM CaCl 2 , 0.3 M CaM, 0.3 mg of bovine serum albumin, and 0.1 mM [␥-32 P]ATP (3000 -5000 cpm/pmol) in a total volume of 500 l. The reactions were terminated by the addition of EDTA to a final concentration of 20 mM. Phosphorylated CaM kinase IV was purified as the pellet of an 80% ammonium sulfate cut. These samples were dialyzed against 5 liters of buffer containing 50 mM HEPES (pH 7.5), 1 mM EDTA, 1 mM dithiothreitol, and 10% (v/v) glycerol.
Assay for Protein Phosphatase-The protein phosphatases we used had the specific activities listed below, using microtubule-associated protein 2 phosphorylated with CaM kinase II as a substrate: (a) protein phosphatase 1 catalytic subunit, 30 nmol/min/mg; (b) purified rat brain protein phosphatase 2A holoenzyme, 1.5 nmol/min/mg; (c) bovine calcineurin, 2.4 nmol/min/mg; and (d) recombinant protein phosphatase 2C␣ (21), 9.5 nmol/min/mg. We adjusted sample amounts of each protein phosphatase to the activities such that a 50% dephosphorylation of phosphorylated microtubule-associated protein 2 occurred in 15 min. Phosphorylated CaM kinase IV (1 g) was incubated at 30°C for the indicated periods with protein phosphatases under the following buffer conditions (total, 50 l): (a) protein phosphatases 1 and 2A, 50 mM HEPES (pH 7.5) and 1 mM MnCl 2 ; (b) calcineurin, 50 mM HEPES (pH 7.5), 1 mM MnCl 2 , 1 mM CaCl 2 , and 0.3 M CaM; and (c) protein phosphatase 2C, 50 mM HEPES (pH 7.5) and 10 mM Mg(OAc) 2 . Reactions were terminated by (a) the addition of 5 l of SDS-sample buffer in the case of SDS-PAGE, (b) the addition of 20 l of 50% trichloroacetic acid and 150 g of bovine serum albumin, followed by centrifugation at 10,000 ϫ g for 10 min and measurement of the radioactivity of the released phosphate by scintillation spectrometry, or (c) the addition of termination buffer (50 mM Tris-HCl, pH 7.5, 0.15 M NaCl, 0.1% Triton X-100, 10 mM EDTA, and 1 mM dithiothreitol), followed by purification of GST-CaM kinase IV with glutathione-Sepharose 4B. To determine changes in CaM kinase IV activity by phosphorylation and dephosphorylation, GST-CaM kinase IV was previously phosphorylated with 0.1 mM cold ATP and then assayed for activity using [␥-32 P]ATP and peptide-␥ as a substrate.

Characterization of Anti-CaM Kinase IV Polyclonal
Antibodies-We prepared a rabbit anti-rat CaM kinase IV polyclonal antibody. This antibody recognized a 63-kDa protein (CaM kinase IV␣) found in a rat hippocampal lysate and a 90-kDa GST-fusion protein of rat CaM kinase IV (Fig. 1A). Using the antibody, we performed immunoprecipitation of the cell lysates obtained from 32 P-labeled hippocampal neurons (Fig. 1B). In our previous study (23), we demonstrated that the activation of CaM kinase II by glutamate stimulation was mediated by a NMDA-type glutamate receptor. We examined whether this observation was also applicable to the activation of CaM kinase IV. The incorporation of phosphate into the 63-kDa protein was significantly increased by stimulation with glutamate for 3 min (182.6 Ϯ 10%; n ϭ 4) compared with a control and was reduced by the application of 20 M MK801 (a noncompetitive inhibitor of NMDA-type glutamate receptor; 92.7 Ϯ 2% of control) (Fig.  1B), indicating that phosphorylation of CaM kinase IV depended on NMDA-type glutamate receptors. The application of MK801 without glutamate did not show significant changes in the phosphorylation levels of CaM kinase II and CaM kinase IV compared with controls (89% and 87%, respectively). Other inhibitors of glutamate receptors (20 M CNQX (a specific inhibitor of AMPA-type glutamate receptors; 212.7 Ϯ 11%) and 1 mM AP3 (an inhibitor of metabotropic glutamate receptors; 191.5 Ϯ 7%) did not show inhibitory effects on CaM kinase IV phosphorylation.
We also raised an antibody against CaM kinase IV phosphorylated at Thr 196 with CaM kinase kinase designated anti-pT196. This antibody recognized the 63-kDa protein in rat hippocampal lysate (Fig. 1C, lane 5) when it was phosphorylated with Ca 2ϩ /CaM and recognized 90-kDa GST-CaM kinase IV (Fig. 1C, lane 8) when it was phosphorylated with GST-CaM kinase kinase, indicating that the antibody was specific for CaM kinase IV phosphorylated at Thr 196 (Fig. 1C).
Glutamate-induced Phosphorylation of CaM Kinase IV and CaM Kinase II-We next examined the time courses of glutamate-induced phosphorylation of CaM kinases IV and II in cultured neurons, using a labeling assay technique. As shown in Fig. 2, A and B, inset, glutamate-induced phosphorylation of CaM kinase IV was transient without CsA ( Fig. 2A and inset). The maximal peak was obtained at 3 min and declined after-  5-7). B, immunoprecipitated CaM kinase IV with the anti-CaM kinase IV antibody from an extract of 32 P-labeled neurons. Radiolabeled neurons were stimulated with control (none), 10 M glutamate, or glutamate plus 20 M MK801 in KRH buffer for 3 min. The medium was then aspirated, and cells were frozen in liquid N 2 . Cells were homogenized, and the Triton X-100-soluble fraction was immunoprecipitated with the anti-CaM kinase IV antibody. The immunocomplex was subjected to SDS-PAGE as duplicate samples, followed by autoradiography. The application of MK801 without glutamate did not show significant differences compared with control (data not shown). C, immunoblot was performed using the anti-CaM kinase IV antibody (lanes 1 and 2), the anti-pT196 antibody (lanes 3-8), or the anti-pT196 antibody preabsorbed with the antigen (lanes 9 and 10) ward. In contrast, the glutamate-induced phosphorylation of CaM kinase II was relatively sustained without CsA after phosphorylation ( Fig. 2B and inset). The difference in the phosphorylation level of CaM kinase IV between 3 and 10 min was significant (p Ͻ 0.01, Student's t test), whereas that of CaM kinase II was not significant. These results suggest the existence of different dephosphorylation mechanisms for CaM kinase IV and CaM kinase II.
The Effects of CsA on Phosphorylation-To test the above-mentioned idea, we examined the effects of protein phosphatase inhibitors using these points of the time in the following experiments. After a 30-min pretreatment of cells with 5 M CsA, an inhibitor of calcineurin, glutamate-induced phosphorylation of CaM kinase IV increased significantly ( Fig. 2A), whereas CaM kinase II phosphorylation was not significantly affected (Fig. 2B). The addition of CsA alone had no effect on the phosphorylation of either CaM kinase IV ( Fig. 2A) or CaM kinase II (Fig. 2B). These results suggest that calcineurin is  (n ϭ 4). Okadaic acid was added to the medium 15 min before stimulation. Immunoprecipitation, SDS-PAGE, and quantitation of phosphorylation were performed as described above and under "Experimental Procedures." Lanes 1-4 indicate the autoradiographs of duplicate samples stimulated with control (none), glutamate, okadaic acid plus glutamate, and okadaic acid alone, respectively, for 3-and 10-min stimulation periods. involved in CaM kinase IV dephosphorylation but not in CaM kinase II dephosphorylation.
The Effects of Okadaic Acid on Phosphorylation-Unlike the results obtained using CsA, a 15-min treatment of neurons with 1 M okadaic acid inhibits both protein phosphatases 1 and 2A and significantly increases the phosphorylation of both CaM kinase IV and CaM kinase II (Fig. 2, C and D), showing that both enzymes are regulated with protein phosphatase 1 and/or 2A. Protein phosphatase 2A has been reported to regulate both CaM kinase II (25) and CaM kinase IV (13,26) and to be associated with CaM kinase IV (27). The increased level of CaM kinase IV phosphorylation after treatment with okadaic acid was about 2-fold higher than that seen after treatment with CsA alone. It seems that CaM kinase IV is more strictly regulated with okadaic acid-sensitive protein phosphatases than with CsA-sensitive protein phosphatases.
Effects of CsA on the Activity of CaM Kinase IV-Although the regulation of CaM kinase IV with an okadaic acid-sensitive protein phosphatase (e.g. PP2A) has been reported previously (26), the effect of CsA on CaM kinase IV activity has not been reported previously. We examined whether increased phosphorylation of CaM kinase IV by CsA treatment resulted in an increase in CaM kinase IV activity. As shown in Fig. 3, a 3-min application of glutamate to neurons significantly increased Ca 2ϩ /CaM-dependent CaM kinase IV activity. In comparison, cells stimulated for 10 min showed decreased activity. Moreover, CaM kinase IV activities after stimulation with glutamate plus CsA for both 3 and 10 min were significantly increased compared with those of cells stimulated for the same periods with glutamate or CsA alone. Because the phosphorylation of Thr 196 of CaM kinase IV with CaM kinase kinase significantly increased both Ca 2ϩ /CaM-dependent and -independent activities in vitro (15,28), the results indicate that phosphorylation of Thr 196 of CaM kinase IV and subsequent autophosphorylation in the N terminus may occur. Thus, calcineurin is involved in the dephosphorylation of at least Thr 196 .
Examination of the Phosphorylation of Thr 196 of CaM Kinase IV-Phosphorylation of CaM kinase IV by treatment with glutamate alone and glutamate plus CsA increased the kinase activity, as described above (Fig. 3). Because this finding indi-cates that Thr 196 of CaM kinase IV is phosphorylated, we asked whether the phosphorylation level of Thr 196 of CaM kinase IV increases after treatment with glutamate and/or CsA using the anti-pT196 antibody. CsA alone had no effect on the phosphorylation of Thr 196 of CaM kinase IV. Treating neurons with glutamate significantly increased phosphorylation, and an additional increase was observed when neurons were treated with glutamate plus CsA (Fig. 4). The total protein levels of CaM kinase IV were not altered by stimulation with glutamate and/or CsA. These results show that Thr 196 of CaM kinase IV is phosphorylated with CaM kinase kinase in response to glutamate and/or CsA.
In Vitro Dephosphorylation of Recombinant CaM Kinase IV with Protein Phosphatases-To further examine whether CaM kinase IV is phosphorylated with CaM kinase kinase and dephosphorylated with protein phosphatases, recombinant GST fusions of CaM kinase kinase and CaM kinase IV were used (Fig. 5). GST-CaM kinase IV phosphorylated with CaM kinase kinase with an incorporation of 43 pmol phosphate/g CaM kinase IV was dephosphorylated with protein phosphatase 1, protein phosphatase 2A, calcineurin, and protein phosphatase 2C in a time-dependent manner (Figs. 5, A and B). To quantitate the relative potency of each protein phosphatase activity, we adjusted the protein levels of protein phosphatases to that which can dephosphorylate an equivalent amount of phosphate from microtubule-associated protein 2. Among the protein phosphatases tested, protein phosphatase 1 was the most active in dephosphorylating CaM kinase IV phosphorylated with CaM kinase kinase, followed by calcineurin and protein phosphatases 2A and 2C (Fig. 5B).
The dephosphorylation of CaM kinase IV with protein phosphatases was correlated with decreases in the activities of CaM kinase IV, which were determined using peptide-␥ as a substrate (Fig. 5C). Because CaM kinase IV is activated by the phosphorylation of Thr 196 with CaM kinase kinase, these results show that protein phosphatase 1, protein phosphatase 2A, calcineurin, and protein phosphatase 2C can each dephosphorylate this site and regulate kinase activity. DISCUSSION In previous reports, we have investigated the activation of CaM kinase II by autophosphorylation in response to neurotransmitters, growth factors, and hormones in various cells (rat embryo fibroblasts (29), hippocampal neurons (23), cerebellar granule cells (5), NG 108-15 cells (30), hippocampal slices (1), rat cortical astrocytes (31), MIN 6 cells (32), and bovine adrenal chromaffin cells (33)). This study describes the activation of CaM kinase IV in hippocampal neurons by stimulation with an excitatory neurotransmitter, glutamate. In this connection, Bito et al. (17) reported that CREB phosphorylation caused by synaptic stimulation or by high potassium-induced depolarization was mediated by CaM kinase IV activation in hippocampal neurons. Furthermore, Park and Soderling (26) reported activation of CaM kinase IV in Jurkat T lymphocytes by stimulation of the antigen receptor CD3. Thus, CaM kinase IV as well as CaM kinase II in living cells is activated in response to neurotransmitters. This is the first report to show directly that CaM kinase IV is activated in neurons by glutamate stimulation, an observation lending support to the idea that CaM kinase IV is involved in Ca 2ϩ -mediated gene expression.
Although CsA alone had no effect on the phosphorylation of CaM kinase IV in cultured hippocampal neurons, the combina-tion of glutamate and CsA greatly increased its phosphorylation compared with the increases observed with glutamate alone ( Fig. 2A). In contrast, no difference in phosphorylation of CaM kinase II was observed when stimulation with glutamate alone was compared with stimulation with glutamate plus CsA. This finding suggested that calcineurin could dephosphorylate CaM kinase IV but not CaM kinase II. The CaM kinase II results were consistent with in vitro observations made in a previous report (34). Recombinant GST-CaM kinase IV phosphorylated with GST-CaM kinase kinase could be dephosphorylated with calcineurin purified from rat brain. In addition, FK506, a calcineurin inhibitor, was reported to delay the time course of dephosphorylation of CREB phosphorylated with CaM kinase IV (17), suggesting the regulation of CaM kinase IV or CaM kinase kinase with calcineurin. On the other hand, okadaic acid combined with glutamate stimulated the phosphorylation of both CaM kinases IV and II (Figs. 1, C and D). This suggests that okadaic acid-sensitive protein phosphatases are involved in the dephosphorylation of CaM kinases IV and II. CaM kinase II could be dephosphorylated with protein phosphatases 1 (35), 2A (36), and 2C (21). As predicted, GST-CaM kinase IV could also be dephosphorylated with purified protein phosphatases 1, 2A, and 2C in addition to calcineurin. In contrast to this study, Park and Soderling (26) reported that the catalytic subunit of protein phosphatase 1 could not dephosphorylate CaM kinase IV. The difference between our results and their data has not yet been clarified. We suppose that it FIG. 5. In vitro dephosphorylation of GST-CaM kinase IV and correlation of activity and dephosphorylation. A, 1 g of GST-CaM kinase IV phosphorylated with [ 32 P]ATP by GST-CaM kinase kinase was dephosphorylated with protein phosphatases as indicated for 0, 3, 10, and 30 min. CaMKIV, CaM kinase IV. After incubation, the reaction mixtures were subjected to SDS-PAGE as duplicate samples, followed by autoradiography. Minor bands with molecular masses lower than that of GST-CaM kinase IV may be degradation products. B, protein phosphatase activities were assayed for 32 P-labeled phosphorylated GST-CaM kinase IV with protein phosphatase 1 (q), protein phosphatase 2A (OE), calcineurin (f), and protein phosphatase 2C (Ⅺ) during a time course. E, a sample incubated without protein phosphatases. Values represent the means of triplicate determinations. Similar experiments were performed at least three times, and a representative experiment is shown. C, effects of dephosphorylation on phosphorylated GST-CaM kinase IV activity. One g of GST-CaM kinase IV phosphorylated with GST-CaM kinase kinase using cold ATP was dephosphorylated with control (none), PP1, PP2A, calcineurin, and PP2C for 30 min. After incubation, GST-CaM kinase IV was immunoprecipitated with the anti-CaM kinase IV antibody, and the immunocomplexes were assayed for CaM kinase IV activity as described under "Experimental Procedures." n ϭ 4, *, p Ͻ 0.01 (Student's t test). may be due to the difference in the isoforms of protein phosphatase 1; they used an ␣-isoform, whereas we used a human ␥-isoform that was purchased commercially. Because both of the catalytic subunits of protein phosphatase 1 are recombinant, the preparations have no contamination by other protein phosphatases.
CaM kinase IV has multiple autophosphorylation sites in the N-terminal region (37) and in a calmodulin-binding domain in the C-terminal region (38) in addition to the Thr 196 phosphorylated with CaM kinase kinase. Although the relationship between kinase activity and N-terminal autophosphorylation is not yet clearly understood, C-terminal autophosphorylation was reported to cause inactivation of CaM kinase IV (38). Because the physiological significance of this phosphorylation is still unclear, it is important to determine the protein phosphatases responsible for the dephosphorylation of the autophosphorylated C-terminal because dephosphorylation of this site may reactivate CaM kinase IV.
Because it was shown in this study that CaM kinase IV was dephosphorylated with calcineurin ( Fig. 5), the interaction between CaM kinase IV and calcineurin is predicted to occur in neurons. CaM kinase IV, which is primarily localized to the nuclei of neurons (17), has also been seen in the axons of neurons (9). Although calcineurin is thought to be localized to the cytoplasm, it has been reported that calcineurin may translocate to the nucleus with NF-AT after ionomycin treatment (39), which shows Ca 2ϩ -mediated translocation of calcineurin to the nucleus. It is unclear how the phosphorylation state of nuclear CaM kinase IV compares with that of the putative cytoplasmic form and whether translocation of calcineurin with glutamate stimulation occurs in neurons. How these enzymes interact should be elucidated in a further study.
This study focused on the dephosphorylation of CaM kinase IV in cultured hippocampal neurons. Because CaM kinase IV is activated with CaM kinase kinase in a Ca 2ϩ /CaM-dependent manner, CaM kinase kinase is also activated in response to glutamate. Because it was reported that CaM kinase kinase showed strong autophosphorylation in the presence of Ca 2ϩ / CaM (40), we do not exclude the possibility that activation of CaM kinase IV may be controlled by the regulation of CaM kinase kinase with protein phosphatases.