Substrate localization creates specificity in calcium/calmodulin-dependent protein kinase II signaling at synapses

Calcium/calmodulin-dependent protein kinase II (CaMKII), a major component of the postsynaptic density (PSD) of excitatory synapses, plays a key role in the regulation of synaptic function in the mammalian brain. Although many postsynaptic substrates for CaMKII have been characterized in vitro, relatively little is known about their phosphorylation in vivo. By tagging synaptic proteins with a peptide substrate specific for CaMKII and expressing them in cultured neurons, we have visualized substrate phosphorylation by CaMKII at intact synapses. All substrates tested were strongly phosphorylated by CaMKII in HEK293 cells. However, activity-dependent phosphorylation of substrates at synapses was highly selective in that the glutamate receptor subunits NR2B and GluR1 were poorly phosphorylated whereas PSD-95 and Stargazin, proteins implicated in the scaffolding and trafficking of AMPA receptors, were robustly phosphorylated. Phosphatase activity limited phosphorylation of Stargazin but not NR2B and GluR1. These results suggest that the unique molecular architecture of the PSD results in highly selective substrate discrimination by CaMKII.     

Nitric oxide-mediated modulation of calcium/calmodulin-dependent protein kinase II

The mechanisms of NO inhibition of CaMK [Ca(2+)/CaM (calmodulin)-dependent protein kinase] II activity were studied. In rat pituitary tumour GH3 cells, TRH [thyrotrophin (TSH)-releasing hormone]-stimulated phosphorylation of nNOS [neuronal NOS (NO synthase)] at Ser(847) was sensitive to an inhibitor of CaMKs, KN-93, and was enhanced by inhibition of nNOS with 7NI (7-nitroindazole). Enzyme activity of CaMKII following in situ treatment with 7NI was also increased. The in vitro activity of CaMKII was inhibited by co-incubation either with nNOS and L-arginine or with NO donors SNAP (S-nitroso-N-acetyl-DL-penicillamine) and DEA-NONOate [diethylamine-NONOate (diazeniumdiolate)]. Once inhibited by these treatments, CaMKII was observed to undergo full reactivation on the addition of a reducing reagent, DTT (dithiothreitol). In transfected cells expressing CaMKII and nNOS, treatment with the calcium ionophore A23187 further revealed nNOS phosphorylation at Ser(847), which was enhanced by 7NI and CaMKII S-nitrosylation. Mutated CaMKII (C6A), in which Cys(6) was substituted with an alanine residue, was refractory to 7NI-induced enhancement of nNOS phosphorylation or to CaMKII S-nitrosylation. Furthermore, we could identify Cys(6) as a direct target for S-nitrosylation of CaMKII using MS. In addition, treatment with glutamate caused an increase in CaMKII S-nitrosylation in rat hippocampal slices. This glutamate-induced S-nitrosylation was blocked by 7NI. These results suggest that inactivation of CaMKII mediated by S-nitrosylation at Cys(6) may contribute to NO-induced neurotoxicity in the brain.     

Conformational changes underlying calcium/calmodulin-dependent protein kinase II activation

Calcium/calmodulin-dependent protein kinase II (CaMKII) interprets information conveyed by the amplitude and frequency of calcium transients by a controlled transition from an autoinhibited basal intermediate to an autonomously active phosphorylated intermediate (De Koninck and Schulman, 1998). We used spin labelling and electron paramagnetic resonance spectroscopy to elucidate the structural and dynamic bases of autoinhibition and activation of the kinase domain of CaMKII. In contrast to existing models, we find that autoinhibition involves a conformeric equilibrium of the regulatory domain, modulating substrate and nucleotide access. Binding of calmodulin to the regulatory domain induces conformational changes that release the catalytic cleft, activating the kinase and exposing an otherwise inaccessible phosphorylation site, threonine 286. Autophosphorylation at Thr286 further disrupts the interactions between the catalytic and regulatory domains, enhancing the interaction with calmodulin, but maintains the regulatory domain in a dynamic unstructured conformation following dissociation of calmodulin, sustaining activation. These findings support a mechanistic model of the CaMKII holoenzyme grounded in a dynamic understanding of autoregulation that is consistent with a wealth of biochemical and functional data.     

Identification of membrane-bound calcium, calmodulin-dependent protein kinase II in canine heart

Phospholamban, the putative regulatory proteolipid of the Ca2+/Mg2+ ATPase in cardiac sarcoplasmic reticulum, was selectively phosphorylated by a Ca2+/calmodulin (CaM)-dependent protein kinase associated with a cardiac membrane preparation. This kinase also catalyzed the phosphorylation of two exogenous proteins known to be phosphorylated by the multifunctional Ca2+/CaM-dependent protein kinase II (Ca2+/CaM-kinase II), i.e., smooth muscle myosin light chains and glycogen synthase a. The latter protein was phosphorylated at sites previously shown to be phosphorylated by the purified multifunctional Ca2+/CaM-kinase II from liver and brain. The membrane-bound kinase did not phosphorylate phosphorylase b or cardiac myosin light chains, although these proteins were phosphorylated by appropriate, specific calmodulin-dependent protein kinases added exogenously. In addition to phospholamban, several other membrane-associated proteins were phosphorylated in a calmodulin-dependent manner. The principal one exhibited a Mr of approximately 56,000, a value similar to that of the major protein (57,000) in a partially purified preparation of Ca2+/CaM-kinase II from the soluble fraction of canine heart that was autophosphorylated in a calmodulin-dependent manner. These data indicate that the membrane-bound, calmodulin-dependent protein kinase that phosphorylates phospholamban in cardiac membranes is not a specific calmodulin-dependent kinase, but resembles the multifunctional Ca2+/CaM-kinase II. Our data indicate that this kinase may be present in both the particulate and soluble fractions of canine heart.     

Molecular characterization of calmodulin trapping by calcium/calmodulin-dependent protein kinase II

Autophosphorylation of alpha-Ca(2+)/calmodulin-dependent protein kinase II (CaM kinase II) at Thr(286) results in calmodulin (CaM) trapping, a >10,000-fold decrease in the dissociation rate of CaM from the enzyme. Here we present the first site-directed mutagenesis study on the dissociation of the high affinity complex between CaM and full-length CaM kinase II. We measured dissociation kinetics of CaM and CaM kinase II proteins by using a fluorescently modified CaM that is sensitive to binding to target proteins. In low [Ca(2+)], the phosphorylated mutant kinase F293A and the CaM mutant E120A/M124A exhibited deficient trapping compared with wild-type. In high [Ca(2+)], the CaM mutations E120A, M124A, and E120A/M124A and the CaM kinase II mutations F293A, F293E, N294A, N294P, and R297E increased dissociation rate constants by factors ranging from 2.3 to 116. We have also identified residues in CaM and CaM kinase II that interact in the trapped state by mutant cycle-based analysis, which suggests that interactions between Phe(293) in the kinase and Glu(120) and Met(124) in CaM specifically stabilize the trapped CaM-CaM kinase II complex. Our studies further show that Phe(293) and Asn(294) in CaM kinase II play dual roles, because they likely destabilize the low affinity state of CaM complexed to unphosphorylated kinase but stabilize the trapped state of CaM bound to phosphorylated kinase.     

Protein kinase C and calcium/calmodulin-dependent protein kinase II phosphorylate three-repeat and four-repeat tau isoforms at different rates

All six isoforms of the microtubule-associated protein tau are present in hyperphosphorylated states in the brains of patients with Alzheimer's disease (AD). It is presently unclear how such hyperphosphorylation of tau is controlled. In a previous study (Singh et al. Arch Biochem Biophys 328: 43-50, 1996) we have shown that three-repeat taus containing two N-terminal inserts were phosphorylated to higher levels and at different sites compared to those either lacking or containing only one such insert. We have extended these observations in this study by comparing the phosphorylation of tau isoforms containing three-repeats (t3, t3L) and four-repeats (t4, t4L). In the absence of N-terminal inserts in tau structure (t3, t4) both CaM kinase II and C-kinase phosphorylated four-repeat tau (t4) to a higher extent than three-repeat tau (t3). When two N-terminal inserts are present in tau structure (t3L, t4L), then three-repeat tau (t3L) is phosphorylated to a higher extent than four-repeat tau (t4L) by these kinases. CK-1 and GSK-3 phosphorylated each of the above pairs of three-repeat and four-repeat taus to the same extents. However, after an initial prephosphorylation of the taus by CaM kinase II, GSK-3 differentially phosphorylated three-repeat and four-repeat taus. Under these conditions thr 231, ser 235, ser 396, and ser 404 were phosphorylated to greater extents in four-repeat tau (t4) compared to three-repeat tau (t3) in the absence of N-terminal inserts. In the presence of such inserts these sites were phosphorylated to greater extents in three-repeat (t3L) compared to four-repeat (t4L) tau. Our results indicate that the extents to which tau isoforms are phosphorylated in normal and AD brain depends on (a) the number of repeats (3 or 4), (b) the number of N-terminal inserts (0, 1, or 2), and (c) the initial phosphorylation state of tau.     

Phosphorylation of chicken cardiac C-protein by calcium/calmodulin-dependent protein kinase II.

Chicken cardiac C-protein was readily phosphorylated by purified calcium/calmodulin-dependent protein kinase II (CaM-kinase II). Maximum incorporation was about 4 mol of 32P/mol of C-protein subunit. Peptide mapping indicated that some of the sites phosphorylated by CaM-kinase II were located on the same phosphopeptides obtained when C-protein was phosphorylated by the cAMP-dependent protein kinase (peptides T1, T2, and T3). There was a fourth peptide (T3a) which was unique to CaM-kinase II phosphorylation. 32P-Amino acid analysis showed that essentially all of the 32P of peptides T1, T2, and T3a was in phosphoserine. cAMP-dependent protein kinase incorporated 32P only into threonine of peptide T3. Threonine was the preferred site of phosphorylation by CaM-kinase II, but there was significant phosphorylation of a serine in peptide T3. Partially purified C-protein preparations contained an associated calcium/calmodulin-dependent protein kinase. Peptide maps obtained from C-protein phosphorylated by the endogenous kinase were similar to those obtained from C-protein phosphorylated by CaM-kinase II. However, the ratio of phosphothreonine to phosphoserine in peptide T3 was lower. This was due to a contaminating phosphatase in the partially purified C-protein which preferentially dephosphorylated the phosphothreonine of peptide T3. It is suggested that the calcium/calmodulin-dependent protein kinase associated with C-protein is similar or identical to CaM-kinase II and that CaM-kinase II may play a role in the phosphorylation of C-protein in the heart.     

Regulation of endothelial cell barrier function by calcium/calmodulin-dependent protein kinase II

Thrombin-induced endothelial cell barrier dysfunction is tightly linked to Ca(2+)-dependent cytoskeletal protein reorganization. In this study, we found that thrombin increased Ca(2+)/calmodulin-dependent protein kinase II (CaM kinase II) activities in a Ca(2+)- and time-dependent manner in bovine pulmonary endothelium with maximal activity at 5 min. Pretreatment with KN-93, a specific CaM kinase II inhibitor, attenuated both thrombin-induced increases in monolayer permeability to albumin and decreases in transendothelial electrical resistance (TER). We next explored potential thrombin-induced CaM kinase II cytoskeletal targets and found that thrombin causes translocation and significant phosphorylation of nonmuscle filamin (ABP-280), which was attenuated by KN-93, whereas thrombin-induced myosin light chain phosphorylation was unaffected. Furthermore, a cell-permeable N-myristoylated synthetic filamin peptide (containing the COOH-terminal CaM kinase II phosphorylation site) attenuated both thrombin-induced filamin phosphorylation and decreases in TER. Together, these studies indicate that CaM kinase II activation and filamin phosphorylation may participate in thrombin-induced cytoskeletal reorganization and endothelial barrier dysfunction.     

Phosphorylation by calcium calmodulin-dependent protein kinase II and protein kinase C modulates the activity of nitric oxide synthase.

Nitric oxide synthase purified from rat brain, which is Ca2+ and calmodulin dependent, was phosphorylated by calcium calmodulin-dependent protein kinase II as well as protein kinase C. Phosphorylation by calcium calmodulin-dependent protein kinase II resulted in a marked decrease in enzyme activity (33% of control) without changing the co-factor requirements, whereas a moderate increase in enzyme activity (140% of control) was observed after phosphorylation by protein kinase C. These findings indicate that brain nitric oxide synthase activity may be regulated not only by Ca2+/calmodulin and several co-factors, but also by phosphorylation.     

Modulation of the phosphorylation and activity of calcium/calmodulin-dependent protein kinase II by zinc

Calcium/calmodulin-dependent protein kinase II (CaMPK-II) is a key regulatory enzyme in living cells. Modulation of its activity, therefore, could have a major impact on many cellular processes. We found that Zn(2+) has multiple functional effects on CaMPK-II. Zn(2+) generated a Ca(2+)/CaM-independent activity that correlated with the autophosphorylation of Thr(286), inhibited Ca(2+)/CaM binding that correlated with the autophosphorylation of Thr(306), and inhibited CaMPK-II activity at high concentrations that correlated with the autophosphorylation of Ser(279). The relative level of autophosphorylation of these three sites was dependent on the concentration of zinc used. The autophosphorylation of at least these three sites, together with Zn(2+) binding, generated an increased mobility form of CaMPK-II on sodium dodecyl sulfate gels. Overall, autophosphorylation induced by Zn(2+) converts CaMPK-II into a different form than the binding of Ca(2+)/CaM. In certain nerve terminals, where Zn(2+) has been shown to play a neuromodulatory role and is present in high concentrations, Zn(2+) may turn CaMPK-II into a form that would be unable to respond to calcium signals.     

Kinetics of the inhibition of calcium/calmodulin-dependent protein kinase II by pea protein-derived peptides

Calcium/calmodulin-dependent protein kinase II (CaMKII) catalyzes the phosphorylation of various cellular proteins and excessive activities have been implicated in the pathogenesis of various chronic diseases. We hypothesized that positively charged peptides can be produced through enzymatic hydrolysis of pea proteins; such peptides could then bind to negatively charged calmodulin (CaM) at a physiological pH level and inhibit CaMKII activity. Pea protein isolate was hydrolyzed with an alkaline protease (alcalase) and filtered through a 1000-mol wt cutoff membrane. The permeate, which contained low-molecular weight peptides, was used to isolate cationic peptides on an SP-Sepharose column by ion exchange chromatography. Separation of the permeate on the SP-Sepharose column yielded two fractions with net positive charges that were subsequently used for enzyme inhibition studies. Fraction I eluted earlier from the column and contained lower contents of lysine and arginine than Fraction II, which eluted later. Results show that both peptide fractions inhibited CaMKII activity mostly in a competitive manner, although kinetic data suggested that inhibition by Fraction II may be of the mixed type. Kinetic analysis (K(m) and K(i)) showed that affinity of peptides in Fraction II for CaM was more than that in Fraction I, which was directly correlated with the higher inhibitory properties of Fraction II against CaMKII. The results suggest that it may be possible to use pea protein-derived cationic peptides to modulate CaMKII activities.     

Purification and characterization of calmodulin-dependent protein kinase II from rat spleen: a new type of calmodulin-dependent protein kinase II

A calmodulin-dependent protein kinase has been purified from rat spleen. The enzyme showed a remarkably similar substrate specificity and kinetic parameters to those of rat brain calmodulin-dependent protein kinase II, and exhibited cross-reactivity to a monoclonal antibody against rat brain calmodulin-dependent protein kinase II, indicating that the enzyme might be a calmodulin-dependent protein kinase II isozyme. The sedimentation coefficient was 13.9S, the Stokes radius was 67 A, and the molecular weight was calculated to be 380,000. The purified enzyme gave five polypeptides bands, corresponding to molecular weights of 51,000, 50,000, 21,000, 20,000, and 18,000, on sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Incubation of the purified enzyme with Ca2+, calmodulin, and ATP under phosphorylating conditions induced the phosphorylation of all five polypeptides. When the logarithm of the velocity of the phosphorylation was plotted against the logarithm of the enzyme concentration (van't Hoff plot), slopes of 0.89, 0.94, and 1.1 were obtained for the phosphorylation of the 50/51-kDa doublet, 20/21-kDa doublet, and 18-kDa polypeptide, respectively. These results indicate that the phosphorylation of the five polypeptides is an intramolecular process, and further indicate that all five polypeptides are subunits of this enzyme. Of the five polypeptides, only the 50- and 51-kDa polypeptides bound to [125I]calmodulin, the other polypeptides not binding to it. A number of isozymic forms of calmodulin-dependent protein kinase II so far demonstrated in various tissues are known to be composed of subunits with molecular weights of 50,000 to 60,000 which can bind to calmodulin. Thus a new type of calmodulin-dependent protein kinase II was demonstrated in the present study.     

Liprinalpha1 degradation by calcium/calmodulin-dependent protein kinase II regulates LAR receptor tyrosine phosphatase distribution and dendrite development

Neural activity regulates dendrite and synapse development, but the underlying molecular mechanisms are unclear. Ca(2+)/calmodulin-dependent protein kinase II (CaMKII) is an important sensor of synaptic activity, and the scaffold protein liprinalpha1 is involved in pre- and postsynaptic maturation. Here we show that synaptic activity can suppress liprinalpha1 protein level by two pathways: CaMKII-mediated degradation and the ubiquitin-proteasome system. In hippocampal neurons, liprinalpha1 mutants that are immune to CaMKII degradation impair dendrite arborization, reduce spine and synapse number, and inhibit dendritic targeting of receptor tyrosine phosphatase LAR, which is important for dendrite development. Thus, regulated degradation of liprinalpha1 is important for proper LAR receptor distribution, and could provide a mechanism for localized control of dendrite and synapse morphogenesis by activity and CaMKII.     

Autoactivation of calmodulin-dependent protein kinase II by autophosphorylation

Autophosphorylation of calmodulin (CaM)-dependent protein kinase II (CaM-kinase II) under limiting conditions (2 microM ATP) decreased progressively with increasing concentrations of a substrate, Pro-Leu-Ala-Arg-Thr-Leu-Ser-Val-Ala-Gly-Leu-Pro-Gly-Lys-Lys (syntide-2), suggesting a competition between the substrate and the autophosphorylation site(s) of the enzyme. The rate and extent of the generation of Ca2+/CaM-independent activity of the enzyme by autophosphorylation were also decreased by the presence of syntide-2. The syntide-2 phosphorylation in the presence of Ca2+/CaM under the limiting conditions reached a steady state, after a lag, when the Ca2+/CaM-independent activity reached a plateau. A linear relationship was observed between the activities in the presence and absence of Ca2+/CaM of the enzyme which had undergone various degrees of autophosphorylation, and the extrapolation of activity in the absence of Ca2+/CaM to zero gave 15-20% of the maximum activity. The steady-state rate of syntide-2 phosphorylation in the presence of Ca2+/CaM by the enzyme that had not undergone prior autophosphorylation was decreased by high concentrations of syntide-2 which suppressed autophosphorylation as well as the generation of Ca2+/CaM-independent activity. These results suggest that although the nonautophosphorylated enzyme possesses a basal low level of Ca2+/CaM-dependent activity, autophosphorylation is required for full activation.     

The eag potassium channel binds and locally activates calcium/calmodulin-dependent protein kinase II

Ca(2+)/calmodulin-dependent protein kinase II (CaMKII) has been implicated in the regulation of neuronal excitability in many systems. Recent studies suggest that local regulation of membrane potential can have important computational consequences for neuronal function. In Drosophila, CaMKII regulates the eag potassium channel, but if and how this regulation was spatially restricted was unknown. Using coimmunoprecipitation from head extracts and in vitro binding assays, we show that CaMKII and Eag form a stable complex and that association with Eag activates CaMKII independently of CaM and autophosphorylation. Ca(2+)/CaM is necessary to initiate binding of CaMKII to Eag but not to sustain association because binding persists when CaM is removed. The Eag CaMKII-binding domain has homology to the CaMKII autoregulatory region, and the constitutively active CaMKII mutant, T287D, binds Eag Ca(2+)-independently in vitro and in vivo. These results favor a model in which the CaMKII-binding domain of Eag displaces the CaMKII autoinhibitory region. Displacement results in autophosphorylation-independent activation of CaMKII which persists even when Ca(2+) levels have gone down. Activity-dependent binding to this potassium channel substrate allows CaMKII to remain locally active even when Ca(2+) levels have dropped, providing a novel mechanism by which CaMKII can regulate excitability locally over long time scales.     

ATP-conjugated peptide inhibitors for calmodulin-dependent protein kinase II

Substrate analog peptides of CaMKII with varying degrees of the inhibitory potency were linked to ATPgammaS either by considering a phosphoryl transfer mechanism or simply by using a relatively long flexible linker. The latter bisubstrate inhibitors showed relatively little effects while the former ones improved inhibitory potency to different levels depending on the binding affinities of the peptide moieties. One of the mechanism-based bisubstrate inhibitors was then utilized to demonstrate an ATP-competitive but peptide substrate-uncompetitive inhibition, supporting an ordered binding mechanism for CaMKII.