Mechanism of the generation of autonomous activity of Ca2+/calmodulin-dependent protein kinase IV.

Ca2+/calmodulin-dependent protein kinase IV (CaM-KIV) is phosphorylated at Thr196 by Ca2+/calmodulin-dependent protein kinase kinase (CaM-KK), resulting in induction of both autonomous activity and a high level of Ca2+/CaM-dependent activity. We have shown that the kinetics of Thr196 phosphorylation of CaM-KIV by CaM-KK is well correlated with the generation of its autonomous activity, although Thr177 phosphorylation of CaM-KI does not induce its autonomous activity. The activities of CaM-KI chimera mutants fused with C-terminal regions (residues 296-469 and 296-350) of CaM-KIV are completely dependent on Ca2+/CaM, which is also the case for CaM-KI. Unlike wild-type CaM-KI, however, phosphorylation of Thr177 in the chimera mutants by CaM-KK resulted in generation of significant autonomous activities, indicating that the phosphorylation of Thr in the activation loop is sufficient to partially release the autoinhibitory region of CaM-KIV from the catalytic core. Indeed, the CaM-KIV peptide (residues 304-325) containing minimum autoinhibitory sequences (residues 314-321) suppressed the activity of non-phosphorylated CaM-KIV with an IC50 of approximately 50 microm, and this suppression was competitive with respect to the peptide substrate; however, the CaM-KIV peptide was not capable of inhibiting Thr196-phosphorylated CaM-KIV. Taken together, these results indicated that the Thr196 phosphorylation of CaM-KIV by CaM-KK reduced the interaction of the catalytic core with the autoinhibitory region, resulting in generation of the autonomous activity.

ent gene expression by the phosphorylation of transcription factors such as cAMP-responsive element-binding protein (CREB) (7)(8)(9)(10)(11). A recent study of transgenic mice carrying dominant negative CaM-KIV alleles that confer defects in the phosphorylation of CREB indicates that these animals exhibit a disruption of late phase long term potentiation and that they are impaired in the consolidation/retention phase of hippocampus-dependent memory (12). Analysis of mice deficient in CaM-KIV has revealed that the CaM-KIV-mediated pathway plays an important role in the function and development of both the cerebellum and hippocampal CA1 neurons (13,14), and is critical for male and female fertility (15,48).
Previous studies have demonstrated that full activation of CaM-KIV requires not only Ca 2ϩ /CaM binding but also phosphorylation of its activation loop Thr residue (Thr 196 ) by an upstream protein kinase (CaM-KK) (9, 16 -19). Phosphorylation of Thr 196 in CaM-KIV results in the induction of a large amount of Ca 2ϩ /CaM-dependent activity. In addition to the induction of total activity, Ca 2ϩ /CaM-independent activity (autonomous activity) is generated by activation with CaM-KK. This important regulatory mechanism would allow a transient elevation in intracellular Ca 2ϩ levels to produce a prolonged CaM-KIV activation to regulate gene transcription via the phosphorylation of transcription factor(s). Autonomous activity of CaM-KIV has been observed with purified enzyme from rat brain (20), immunoprecipitated enzyme from anti-TCR/CD3 monoclonal antibody-stimulated Jurkat cells (21,22), and ionomycin-treated, transfected COS-7 and HeLa cells (23,24). A recent study suggested that the autonomous activity appears to be required for CaM-KIV to regulate CREB-mediated transcription, indicating that it might be physiologically relevant (25).
We have shown that the induction of the total activity of CaM-KIV by CaM-KK phosphorylation was because of mainly decreasing K m for its substrate (17). However, this does not account for generation of the autonomous activity. At a minimum, Thr 196 phosphorylation by CaM-KK is required for generation of Ca 2ϩ /CaM-independent activity of CaM-KIV, because a mutation of Thr 196 by Ala has been shown to abolish the generation of autonomous activity (19,23). Therefore, it has not been resolved whether the Ca 2ϩ -independent activity is a direct consequence of Thr 196 phosphorylation or whether it is the result of subsequent autophosphorylation.
In contrast, it has been well characterized that CaM-KII, another member of multifunctional CaM-Ks, is converted to the Ca 2ϩ /CaM-independent form by autophosphorylation at Thr 286 in its autoinhibitory region, which suppresses the autoinhibitory function of the enzyme (26 -31). Thr 286 autophosphorylation has been shown to be sensitive to the duration, magnitude, and frequency of the imposed calcium transient (32,33) and also to be important in the regulation of synaptic plasticity and behavior in vivo (34). Although there is an equivalent Thr (Thr 308 ) in the putative autoinhibitory region of CaM-KIV, mutation of Thr 308 to Ala did not impair the generation of autonomous activity, indicating that the Thr 308 is not involved in the generation of autonomous activity (23,35). In addition, multiple autophosphorylation sites have been identified in either the N-terminal region of the catalytic domain or the Cterminal region of CaM-KIV (36 -38), but none of them has been shown to be involved in its autonomous activity (35). Thus the precise mechanism of the autonomous activity of CaM-KIV generated by CaM-KK phosphorylation remains uncertain.
In this report, we have examined the mechanism of generation of the autonomous activity of CaM-KIV by CaM-KK phosphorylation using Escherichia coli-expressed recombinant enzymes, including CaM-KI/CaM-KIV chimera mutants and the autoinhibitory peptide of CaM-KIV, and demonstrated that Thr 196 phosphorylation reduced the affinity of its autoinhibitory region to the catalytic core and is likely sufficient to generate the autonomous activity of the enzyme.

EXPERIMENTAL PROCEDURES
Materials-CaM-KK␣ cDNA (GenBank TM accession number L42810) was obtained from a rat brain cDNA library (18). Recombinant CaM-KK␣ was expressed in E. coli and purified as described previously (39). Recombinant rat CaM was expressed in the E. coli strain BL21(DE3) using pET-CaM (kindly provided by Dr. Nobuhiro Hayashi, Fujita Health University, Toyoake, Japan) and purified by phenyl-Sepharose column chromatography (40). Anti-CaM-KIV antibody and anti-CaM-KI antibody were obtained from Signal Transduction Laboratories and Santa Cruz Biotechnology, Inc., respectively. Anti-phospho-CaM-KIV at Thr 196 and anti-phospho-CaM-KI at Thr 177 monoclonal antibodies were generated against the synthetic phosphopeptides corresponding to residues 189 -203 of rat CaM-KIV (CEHQVLMKT(p)VCGTPGY) and corresponding to residues 169 -184 of rat CaM-KI (CEDPGSVLST-(p)ACGTPGY), respectively. Peptides were conjugated with keyhole limpet hemocyanin via the N terminus cysteine and were injected into BALB/c mice as described previously (41). Biotinylated CaM was purchased from Biomedical Technologies Inc. (Stoughton, MA). A synthetic peptide corresponding to residues 304 -325 of mouse CaM-KIV (VHM-DTAQKKLQEFNARRKLKAA) was obtained from Bio-Synthesis Inc. (Lewisville, TX) with Ͼ95% purity. All other chemicals were from standard commercial sources.
Construction and Expression of CaM-KI and CaM-KIV-GST-fused CaM-KI was constructed by amplification of cDNA fragments using rat CaM-KI cDNA as a template, and a sense primer (5Ј-GCTCTAGAGAT-GCCAGGGGCAGTGGAAGGC-3Ј) and an antisense primer (5Ј-CCGCTC-GAGTCAGTCCATGGCCCTAGAGC-3Ј). GST-fused CaM-KIVs were constructed by amplification of cDNA fragments using mouse CaM-KIV cDNA (17) as a template, and a sense primer (5Ј-GCTCTAGAGATGCT-CAAAGTCACGGTGCCC-3Ј) and the following antisense primers: wild- The PCR products were then ligated into the XbaI/XhoI site of the pGEX-PreS vector (39). GST-fused CaM-KI/IV469 and CaM-KI/IV350 were constructed by replacement of the StuI/XhoI-digested fragment of the wild-type pGEX-PreS-CaM-KI by a PCR fragment amplified using mouse CaM-KIV cDNA as a template, and sense (5Ј-CCTTGCAGCAC-CCATGGGTCACAGGTAAAG-3Ј) and antisense primers (wild-type and CaM-KIV-(1-350), respectively) as described above. The nucleotide sequences of CaM-KIV mutants and CaMKI/KIV chimera mutants, including the wild-type enzymes, were confirmed by an ABI377 automated sequencer. cDNAs carrying GST-fused enzymes were introduced into E. coli JM-109, and the expression of recombinant enzyme was induced by 1 mM isopropyl-␤-D-thiogalactopyranoside.
In Vitro Activation of CaM-KI and CaM-KIV by CaM-KK-Purified recombinant CaM-Ks (0.1 mg/ml) were incubated with CaM-KK␣ (3 g/ml) at 30°C for the indicated periods in a solution containing 50 mM HEPES (pH 7.5), 10 mM Mg(Ac) 2 , 1 mM DTT, 2 mM CaCl 2 , 10 M CaM, and 200 M ATP. The reaction was initiated by addition of ATP and terminated by 10 -20-fold dilution with ice-cold 50 mM HEPES (pH 7.5), 2 mg/ml bovine serum albumin, 10% ethylene glycol, and 2 mM EDTA. Five l of the diluted sample was then subjected to the protein kinase assay or Western blot analysis.
In Vitro Assay for CaM-KI and CaM-KIV Activity-CaM-KI and CaM-KIV were assayed at 30°C for 5-10 min in a solution (25 l) containing 50 mM HEPES (pH 7.5), 10 mM Mg(Ac) 2 , 40 M syntide-2, 1 mM DTT, and 200 M [␥-32 P]ATP (ϳ1000 cpm/pmol) in the presence of either 1 mM EGTA (autonomous activity) or 1 mM CaCl 2 , 4 M CaM. The reaction was initiated by the addition of the enzyme and terminated by spotting aliquots (15 l) onto phosphocellulose paper (Whatman P-81) followed by several washes with 75 mM phosphoric acid (42). Phosphate incorporation into syntide-2 was determined by liquid scintillation counting of the filters.
Others-Western blotting was performed as described previously using horseradish peroxidase-conjugated anti-mouse IgG antibody (Amersham Biosciences) or anti-goat IgG antibody as a secondary antibody and chemiluminescence reagent (PerkinElmer Life Sciences) for detection. CaM overlay was carried out using biotinylated CaM (0.5 g/ml) in the presence of 1 mM CaCl 2 as described previously (17) using chemiluminescence reagent (PerkinElmer Life Sciences) for detection. Protein concentration was estimated by staining with Coomassie Brilliant Blue (Bio-Rad) using bovine serum albumin as a standard (43).

Activation and Phosphorylation of CaM-KIV and CaM-KI by
CaM-KK-It has been shown that CaM-KIV is phosphorylated at Thr 196 by CaM-KK, resulting in induction of a high level of Ca 2ϩ /CaM-dependent activity as well as generation of Ca 2ϩ / CaM-independent activity (17)(18)(19). Because CaM-KIV undergoes autophosphorylation of multiple residues subsequent to activation (36 -38), the relationship between activation of the kinase activity and Thr 196 phosphorylation has not been precisely determined in vitro. Therefore, we have attempted to perform time course experiments on CaM-KIV activation and phosphorylation of Thr 196 using anti-phospho-Thr 196 monoclonal antibody (Fig. 1, A and B). We have expressed GST-fused enzymes and then GST was cleaved by PreScission protease treatment followed by removing GST and GST-fused protease to avoid the effect of GST on the enzyme activities and regulation. Thus the recombinant enzymes used in this study contain five residues (Gly-Pro-Ile-Leu-Glu) fused with the first Met. As shown in Fig. 1A, both Ca 2ϩ /CaM-dependent and -independent activities (ϳ30% of total activity) of CaM-KIV were induced by CaM-KK treatment in the presence of Ca 2ϩ /CaM and Mg-ATP and were saturated for 20 min. This activation kinetics was well correlated with Thr 196 phosphorylation under this condition (t1 ⁄2 ϭ ϳ5 min, Fig. 1B). It is noteworthy that we did not observe a significant time lag between generation of the autonomous activity and phosphorylation of Thr 196 . These characteristic features of the recombinant CaM-KIV used in this study, including generation of autonomous activity by activation and kinetic parameters of the enzyme (K m for ATP ϭ 27 M and K m for syntide-2 ϭ 264 M, Fig. 5C), were similar to the previous results obtained using enzymes expressed in Sf9 cells (16,17,19). We also performed similar experiments using CaM-KI and the same concentration (3 g/ml) of CaM-KK as used for CaM-KIV activation (Fig. 1C). Protein kinase activity assay and Western blot analysis using anti-phospho-Thr 177 monoclonal antibody revealed that, in comparison to CaM-KIV, CaM-KI was more rapidly activated and phosphorylated at Thr 177 by CaM-KK (t1 ⁄2 ϭ Ͻ1 min) than CaM-KIV. However, significant autonomous activity of CaM-KI was not observed during the activation. These results indicated that the autoinhibitory mechanisms might differ between CaM-KIV and CaM-KI, at least for the phosphorylated form of enzymes, which in turn might affect the generation of the autonomous activity of CaM-KIV.
Characterization of CaM-KI/CaM-KIV Chimera Mutants-To examine the role of the autoinhibitory mechanism of CaM-KIV in the generation of autonomous activity, we have constructed, expressed, and purified chimera mutants in which the CaM-KI catalytic domain was fused with various C-terminal portions of CaM-KIV (residues 296 -469 and residues 296 -350; Fig. 2, A and B, upper panel), because it has been shown that CaM-KI does not undergo autophosphorylation in the catalytic domain subsequent to the activation (44) and also does not generate autonomous activity in association with Thr 177 phosphorylation (Fig. 1C). CaM-overlay analysis revealed that the chimera mutants possess the same degree of functional Ca 2ϩ / CaM binding ability as the wild-type CaM-KI (Fig. 2B, lower  panel). Without activation by CaM-KK, CaM-KI/CaM-KIV469 and CaM-KI/CaM-KIV350 were inactive in the absence of Ca 2ϩ /CaM and exhibited kinase activity only in the presence of Ca 2ϩ /CaM, which results are indistinguishable from those for the wild-type CaM-KI (Fig. 2C). This indicates that the autoinhibitory region in the C-terminal of CaM-KIV (within residues 296 -350) functionally suppresses the catalytic activity of CaM-KI, which is conformationally neutralized by Ca 2ϩ /CaM binding. When we treated these CaM-KI chimera mutants with CaM-KK in the presence of Ca 2ϩ /CaM and Mg-ATP for 5 min as described in Fig. 1C, both mutants including wild-type CaM-KI were phosphorylated at Thr 177 (Fig. 2D, inset). However, unlike wild-type CaM-KI, which does not generate significant autonomous activity (Figs. 1C and 2D), CaM-KI/CaM-KIV469 and CaM-KI/CaM-KIV350 showed a high level of autonomous activity after their activation. This result strongly supports the idea that the autoinhibitory mechanism of CaM-KIV was partially disrupted by phosphorylation of the activation loop Thr residue, thereby resulting in generation of autonomous activity, because it has been shown that CaM-KI is phosphorylated at Thr 177 by CaM-KK, but no other autophosphorylation in the catalytic domain has been observed (44).
Mapping of the Autoinhibitory Sequence in CaM-KIV-To clarify the involvement of the autoinhibitory function in generation of the autonomous activity of CaM-KIV, we attempted to precisely map the autoinhibitory sequence in CaM-KIV. Previous studies have demonstrated that the truncation at Leu 313 (16) and a block mutation (Phe 316 -Asn 317 to Asp-Asp) are converted to the Ca 2ϩ /CaM-independent form of CaM-KIV (17), suggesting that the C-terminal region after Leu 313 contains the autoinhibitory sequence of CaM-KIV. Because the precise location of the autoinhibitory sequence has not been determined, we expressed and purified a series of C-terminal truncation mutants to map a minimum autoinhibitory sequence (Fig. 3, A and B, insert upper panel). As shown in Fig.  3B, measurement of the protein kinase activities of these mutants in either the absence or presence of Ca 2ϩ /CaM revealed that truncation after Gly 336 did not alter either the Ca 2ϩ /CaM dependence of the activity or the activity itself. The truncation mutants at Ala 331 , Lys 327 , and Lys 321 are completely inactive in either the absence or presence of Ca 2ϩ /CaM, indicating that these mutants contain a functional autoinhibitory sequence. Semiquantitative analysis of the Ca 2ϩ /CaM binding of these mutants using the CaM overlay method in the presence of Ca 2ϩ (Fig. 3B, insert lower panel) showed that the truncation mutant at Ala 331 fails to bind Ca 2ϩ /CaM as well as other truncation mutants (1-327, 1-321, 1-313, and 1-309), suggesting that the residues between Ser 332 and Gly 336 are the C-terminal end of the Ca 2ϩ /CaM binding sequence. Thus the 1-331, 1-327, and 1-321 mutants were not activated by Ca 2ϩ /CaM. Further trun- cation at Leu 313 generated the constitutively active enzyme to the same degree as did truncation at Ala 309 , consistent with a previous report (16), and which suggested that a minimum autoinhibitory sequence is located between Gln 314 and Lys 321 . This is also consistent with our previous observation that double Asp mutations at Phe 316 and Asn 317 impair the autoinhibitory function of CaM-KIV, resulting in generation of Ca 2ϩ / CaM-independent activity (17).
We also examined whether CaM-KIV-(1-336) generates its autonomous activity by activation, because the mutant lacks the C-terminal portion of CaM-KIV (residues 337-469) but contains the functional regulatory domain, including the autoinhibitory and Ca 2ϩ /CaM binding sequences that completely suppress the catalytic activity of CaM-KIV in the absence of Ca 2ϩ /CaM (Fig. 3). During incubation with CaM-KK in the presence of Mg-ATP and Ca 2ϩ /CaM, Ca 2ϩ /CaM-independent activity was significantly increased in association with the Thr 196 phosphorylation (Fig. 4), which was similar to the results for the wild-type CaM-KIV (Fig. 1, A and B). This result indicated that the C-terminal portion of CaM-KIV after Gly 336 is unlikely to be involved in the generation of its autonomous activity.
Inhibition of CaM-KIV Activity by CaM-KIV Autoinhibitory Peptide-Based on the mapping of the regulatory domain in CaM-KIV (Fig. 3), we were able to synthesize the CaM-KIV peptide (residues 304 -325) containing the minimum autoinhibitory sequence (residues 314 -321) to examine its inhibition of unphosphorylated and Thr 196 -phosphorylated CaM-KIV (Fig. 5A, upper and middle panels). We also confirmed by CaM overlay analysis that the Ca 2ϩ /CaM bindings of two enzymes were indistinguishable (Fig. 5A, lower panel). In the presence of Ca 2ϩ /CaM, the CaM-KIV peptide inhibited the total activity of unphosphorylated CaM-KIV with an IC 50 value of ϳ50 M (Fig. 5B), which is ϳ3-fold less potent than the inhibition by the CaM-KII autoinhibitory peptide (residues 281-302, T286A) (17). Furthermore, inhibition of CaM-KIV by the CaM-KIV peptide (residues 304 -325) was not competitive with ATP but was competitive with the peptide substrate syntide-2 (Fig. 5C). This kinetic profile of enzyme inhibition was similar to that observed for the inhibition of CaM-KIV by the CaM-KII autoinhibitory peptide (residues 281-302, T286A) (17). When we assayed the same concentration (2 g/ml) of phosphorylated CaM-KIV by CaM-KK for 30 min (Fig. 1A), the CaM-KIV peptide at a concentration of up to 250 M was unable to suppress the activity of activated CaM-KIV (Fig. 5B). We performed the same kinase assay as shown in Fig. 5B in the absence of the peptide substrate (syntide-2), but significant phosphate incorporation into the CaM-KIV peptide was not detected (data not shown). We also confirmed by mass spectrometry analysis that the autoinhibitory peptide was not phosphorylated at Thr 308 by activated CaM-KIV (data not shown). This result indicated that the interaction between the catalytic core of CaM-KIV and the autoinhibitory peptide was abolished by Thr 196 phosphorylation, resulting in a loss of the inhibitory potency of the peptide.

DISCUSSION
The enzymatic activities of Ca 2ϩ /CaM-dependent protein kinases are thought to be strictly autoinhibited by intrasteric interaction between their own autoinhibitory segment and the catalytic core when the intracellular Ca 2ϩ concentration is low enough to dissociate from CaM. Once the concentration rises upon stimulation of the cells, the Ca 2ϩ -CaM complex binds to the Ca 2ϩ /CaM binding segment of the kinases, which either overlaps or is adjacent to the autoinhibitory segment. Then Ca 2ϩ /CaM binding conformationally neutralize the autoinhibitory function, resulting in the generation of protein kinase activity. Two members of the CaM-K family, CaM-KII and CaM-KIV, generate Ca 2ϩ /CaM-independent activity in association with autophosphorylation at Thr 286 and trans-phosphorylation at Thr 196 by an upstream kinase (CaM-KK), respectively.
Here we have shown that the phosphorylation of Thr 196 in the activation loop by CaM-KK was likely sufficient to generate the autonomous activity of CaM-KIV because of: 1) the in vitro kinetics of the activation and Thr 196 phosphorylation of CaM-KIV were well correlated; 2) CaM-KI chimera mutants containing the regulatory domain of CaM-KIV generated autonomous activities that were associated with Thr 177 phosphorylation by CaM-KK, whereas the CaM-KI wild-type did not exhibit Ca 2ϩ /CaM-independent activity by Thr 177 phosphorylation and also did not undergo autophosphorylation in the catalytic domain subsequent to the activation (44); 3) as shown in the experiment with a truncation mutant (CaM-KIV-(1-336)), the C-terminal of CaM-KIV (residues 337-469) is not directly involved in generation of autonomous activity; and 4) CaM-KIV autoinhibitory peptide (residues 304 -325) failed to inhibit activated CaM-KIV without phosphorylation by CaM-KIV. Although multiple autophosphorylation sites in the N-terminal region of the catalytic domain of CaM-KIV have been identified (36 -38), a previous study has shown that the deletion of six N-terminal Ser residues (Ser 8 , Ser 10 , Ser 12 , Ser 13 , Ser 15 , and Ser 16 in the human enzyme) did not impair the generation of autonomous activity subsequent to the activation by CaM-KK (35). Collectively, these results suggest that the Thr 196 phosphorylation is most likely sufficient to generate the autonomous activity of CaM-KIV. However, although it is unlikely, we cannot completely exclude the possibility that the autophosphorylation at unidentified residue(s) subsequent to phosphorylation at Thr 196 by CaM-KK may be involved in the generation of autonomous activity. Interestingly, CaM-KI is also phosphorylated on Thr 177 in the activation loop by CaM-KK, resulting in a large increase in Ca 2ϩ /CaM-dependent activity without generation of autonomous activity (45,46). This result, together with our present data, suggests that the autoinhibitory mechanisms of CaM-KI and CaM-KIV are distinct, which is supported by the fact that the CaM-KI chimera mutants fused with the regulatory domain of CaM-KIV, generating Ca 2ϩ /CaM-independent activity by Thr 177 phosphorylation, whereas the CaM-KIV autoinhibitory domain completely suppressed the catalytic activity of unphosphorylated chimera mutants in the absence of Ca 2ϩ / CaM. Therefore, the autoinhibitory segment of CaM-KIV no longer tightly binds to the Thr 177 -phosphorylated CaM-KI catalytic core in the absence of Ca 2ϩ /CaM, but that of CaM-KI does. Consistent with these findings, we directly demonstrated that the inhibitory function of the autoinhibitory peptide of CaM-KIV (residues 304 -325), by preventing the substrate binding, was completely lost toward phosphorylated CaM-KIV on Thr 196 . This observation clearly indicates that the autonomous activity of CaM-KIV is generated by a reduction of the interaction between the catalytic core and the autoinhibitory region. Notably, the Ca 2ϩ /CaM-independent activity of activated CaM-KIV accounts for ϳ30% of the total activity suggesting that the autoinhibitory region is partially but not fully released from the catalytic core. This may indicate that the intramolecular interaction of the catalytic core with the autoinhibitory region is not completely destroyed by Thr 196 phosphorylation, as distinct from our results with the intermolecular inhibition by the autoinhibitory peptide. The crystal structure of the autoinhibited form of CaM-KI revealed that the regulatory segment of CaM-KI (residues 286 -316) interacted with the catalytic core at some distance from Thr 177 (47). This may account for the finding that the autoinhibitory mechanism of CaM-KI appeared not to be affected by Thr 177 phosphorylation. We could therefore speculate that the autoinhibitory domain of CaM-KIV might be located closer to the activation loop in its autoinhibited form of unphosphorylated kinase to prevent the substrate binding. Subsequent to phosphorylation by CaM-KK, the interaction of the autoinhibitory segment with the catalytic core of CaM-KIV could be reduced by either Thr 196 phosphorylation directly or by conformational changes of the catalytic domain mediated by Thr 196 phosphorylation in the absence of Ca 2ϩ /CaM, resulting in the partial release of the autoinhibitory domain from the catalytic core. In contrast to the findings for CaM-KIV, it has been extensively characterized that direct autophosphorylation of the autoinhibitory region at Thr 286 in CaM-KII suppresses the autoinhibitory function that is necessary to generate CaM-KII autonomous activity (26 -31). Therefore, closely related multifunctional CaM-Ks appear to show different modes of regulating the autoinhibitory mechanism in association with the activation. Because the three-dimensional structure of CaM-KIV has not been determined, future structural studies will be needed to clarify the mechanism of the enzymatic regulation of CaM-KIV in greater detail, and to reveal the diverse range of autoinhibitory mechanisms among the CaM-K family.