The DAP-kinase family of proteins: study of a novel group of calcium-regulated death-promoting kinases

DAP-kinase (DAPk) is a Ca(2+)/calmodulin (CaM)-regulated Ser/Thr kinase that functions as a positive mediator of programmed cell death. It associates with actin microfilament and has a unique multidomain structure. One of the substrates of DAPk was identified as myosin light chain (MLC), the phosphorylation of which mediates membrane blebbing. Four additional kinases have been identified based on the high homology of their catalytic domain to that of DAPk. Yet, they differ in the structure of their extracatalytic domains and in their intracellular localization. One member of this family, DRP-1, also shares with DAPk both the property of activation by Ca(2+)/CaM and a specific phosphorylation-based regulatory mechanism. The latter involves an inhibitory type of autophosphorylation on a conserved serine at position 308, in the CaM regulatory domains of these two kinases. This phosphorylation, which occurs in growing cells, restrains the death-promoting effects of these kinases, and is specifically removed upon exposure of cells to various apoptotic stimuli. The dephosphorylation at this site increases the binding and sensitivity of each of these two kinases to their common activator-CaM. In DAPk, the dephosphorylation of serine 308 also increases the Ca(2+)/CaM-independent substrate phosphorylation. In DPR-1, it also promotes the formation of homodimers necessary for its full activity. These results are consistent with a molecular model in which phosphorylation on serine 308 stabilizes a locked conformation of the CaM regulatory domain within the catalytic cleft and simultaneously also interferes with CaM binding. In DRP-1, it introduces an additional locking device by preventing homodimerization. We propose that this unique mechanism of autoinhibition, evolved to keep these death-promoting kinases silent in healthy cells and ensures their activation only in response to apoptotic signals.     

The regulation of death-associated protein (DAP) kinase in apoptosis

DAP-kinase is a calcium/calmodulin (Ca2+/CaM) serine/threonine kinase which positively mediates programmed cell death in a variety of cell systems. The kinase is localized to the actin microfilament and has a unique, multidomain structure consisting of ankyrin repeats and a death domain. One of the substrates of DAP-kinase was identified as myosin light chain (MLC), the phosphorylation of which mediates membrane blebbing. Another arm in its mode of action leads to the formation of autophagic vesicles. Recent work addressed its mode of regulation and identified a mechanism which restrains its apoptotic function in growing cells and enables its activation during cell death. It involves an inhibitory type of autophosphorylation on serine 308 within the CaM regulatory domain. This negative phosphorylation takes place in growing cells and is strongly reduced upon their exposure to the apoptotic stimulus of C6-ceramide. The substitution of serine 308 to alanine, which mimics the ceramide-induced dephosphorylation at this site, increases Ca2+/CaM-independent substrate phosphorylation, as well as binding and overall sensitivity of the kinase to CaM. At the cellular level, it strongly enhances the death-promoting activity of the kinase. These results are consistent with a molecular model in which phosphorylation on serine 308 stabilizes a locked conformation of the CaM regulatory domain within the catalytic cleft and, simultaneously, also interferes with CaM binding. We propose that this unique mechanism of auto-inhibition evolved to impose a locking device which keeps DAP-kinase silent in healthy cells and ensures its activation only in response to apoptotic signals.     

The pro-apoptotic function of death-associated protein kinase is controlled by a unique inhibitory autophosphorylation-based mechanism

Death-associated protein kinase is a calcium/calmodulin serine/threonine kinase, which positively mediates programmed cell death in a variety of systems. Here we addressed its mode of regulation and identified a mechanism that restrains its apoptotic function in growing cells and enables its activation during cell death. It involves autophosphorylation of Ser(308) within the calmodulin (CaM)-regulatory domain, which occurs at basal state, in the absence of Ca(2+)/CaM, and is inversely correlated with substrate phosphorylation. This type of phosphorylation takes place in growing cells and is strongly reduced upon their exposure to the apoptotic stimulus of C(6)-ceramide. The substitution of Ser(308) to alanine, which mimics the ceramide-induced dephosphorylation at this site, increases Ca(2+)/CaM-independent substrate phosphorylation as well as binding and overall sensitivity of the kinase to CaM. At the cellular level, it strongly enhances the death-promoting activity of the kinase. Conversely, mutation to aspartic acid reduces the binding of the protein to CaM and abrogates almost completely the death-promoting function of the protein. These results are consistent with a molecular model in which phosphorylation on Ser(308) stabilizes a locked conformation of the CaM-regulatory domain within the catalytic cleft and simultaneously also interferes with CaM binding. We propose that this unique mechanism of auto-inhibition evolved to impose a locking device, which keeps death-associated protein kinase silent in healthy cells and ensures its activation only in response to apoptotic signals.     

Death-associated protein kinase-related protein 1, a novel serine/threonine kinase involved in apoptosis

In this study we describe the identification and structure-function analysis of a novel death-associated protein (DAP) kinase-related protein, DRP-1. DRP-1 is a 42-kDa Ca(2+)/calmodulin (CaM)-regulated serine threonine kinase which shows high degree of homology to DAP kinase. The region of homology spans the catalytic domain and the CaM-regulatory region, whereas the remaining C-terminal part of the protein differs completely from DAP kinase and displays no homology to any known protein. The catalytic domain is also homologous to the recently identified ZIP kinase and to a lesser extent to the catalytic domains of DRAK1 and -2. Thus, DAP kinase DRP-1, ZIP kinase, and DRAK1/2 together form a novel subfamily of serine/threonine kinases. DRP-1 is localized to the cytoplasm, as shown by immunostaining and cellular fractionation assays. It binds to CaM, undergoes autophosphorylation, and phosphorylates an exogenous substrate, the myosin light chain, in a Ca(2+)/CaM-dependent manner. The truncated protein, deleted of the CaM-regulatory domain, was converted into a constitutively active kinase. Ectopically expressed DRP-1 induced apoptosis in various types of cells. Cell killing by DRP-1 was dependent on two features: the status of the catalytic activity, and the presence of the C-terminal 40 amino acids shown to be required for self-dimerization of the kinase. Interestingly, further deletion of the CaM-regulatory region could override the indispensable role of the C-terminal tail in apoptosis and generated a "superkiller" mutant. A dominant negative fragment of DAP kinase encompassing the death domain was found to block apoptosis induced by DRP-1. Conversely, a catalytically inactive mutant of DRP-1, which functioned in a dominant negative manner, was significantly less effective in blocking cell death induced by DAP kinase. Possible functional connections between DAP kinase and DRP-1 are discussed.     

Regulation of Drosophila Ca2+/Calmodulin-Dependent Protein Kinase II by Autophosphorylation Analyzed by Site-Directed Mutagenesis

Abstract: In this study we demonstrate that Drosophila calcium/calmodulin-dependent protein kinase II (CaMKII) is capable of complex regulation by autophosphorylation of the three threonines within its regulatory domain. Specifically, we show that autophosphorylation of threonine-287 in Drosophila CaMKII is equivalent to phosphorylation of threonine-286 in rat α CaMKII both in its ability to confer calcium independence on the enzyme and in the mechanistic details of how it becomes phosphorylated. Autophosphorylation of this residue occurs only within the holoenzyme structure and requires calmodulin (CaM) to be bound to the substrate subunit. Phosphorylation of threonine-306 and threonine-307 in the CaM binding domain of the Drosophila kinase occurs only in the absence of CaM, and this phosphorylation is capable of inhibiting further CaM binding. Additionally, our findings suggest that phosphorylation of threonine-306 and threonine-307 does not mimic bound CaM to alleviate the requirement for CaM binding to the substrate subunit for intermolecular threonine-287 phosphorylation. These results demonstrate that the mechanism of regulatory autophosphorylation of this kinase predates the split between invertebrates and vertebrates.     

The DAPK family: a structure–function analysis

DAP-kinase (DAPK) is the founding member of a family of highly related, death associated Ser/Thr kinases that belongs to the calmodulin (CaM)-regulated kinase superfamily. The family includes DRP-1 and ZIP-kinase (ZIPK), both of which share significant homology within the common N-terminal kinase domain, but differ in their extra-catalytic domains. Both DAPK and DRP-1 possess a conserved CaM autoregulatory domain, and are regulated by calcium-activated CaM and by an inhibitory auto-phosphorylation within the domain. ZIPK’s activity is independent of CaM but can be activated by DAPK. The three kinases share some common functions and substrates, such as induction of autophagy and phosphorylation of myosin regulatory light chain leading to membrane blebbing. Furthermore, all can function as tumor suppressors. However, they also each possess unique functions and intracellular localizations, which may arise from the divergence in structure in their respective C-termini. In this review we will introduce the DAPK family, and present a structure/function analysis for each individual member, and for the family as a whole. Emphasis will be placed on the various domains, and how they mediate interactions with additional proteins and/or regulation of kinase function.     

Intrasteric regulation of myosin light chain kinase: the pseudosubstrate prototope binds to the active site.

We previously proposed a molecular mechanism for the activation of smooth muscle myosin light chain kinase (smMLCK) by calmodulin (CaM). According to this model, smMLCK is autoinhibited in the absence of Ca2+/CaM due to the interaction of a pseudosubstrate prototope, contained within the CaM binding/regulatory region, with the active site of the enzyme. Binding of Ca2+/CaM releases the autoinhibition and allows access of the protein substrate to the active site of the enzyme, resulting in phosphorylation of the myosin light chains. We now provide direct experimental evidence that the pseudosubstrate prototope can associate with the active site. We constructed a smMLCK mutant in which the five-amino acid phosphorylation site of the myosin light chain substrate was inserted into the pseudosubstrate sequence of the CaM binding domain without disrupting the ability of the enzyme to bind Ca2+/CaM. We demonstrate that this mutant undergoes intramolecular autophosphorylation at the appropriate inserted serine residue in the absence of CaM and that this autophosphorylation activates the enzyme. Binding of Ca2+/CaM to the mutant enzyme stimulated myosin light chain substrate phosphorylation but strongly inhibited autophosphorylation, presumably by removing the pseudosubstrate from the active site. These results confirm that the pseudosubstrate sequence has access to the catalytic site and that the activation of the enzyme is accompanied by its removal from this position due to Ca2+/CaM binding as predicted by the model.     

Mechanistic link between PKR dimerization, autophosphorylation, and eIF2alpha substrate recognition

The antiviral protein kinase PKR inhibits protein synthesis by phosphorylating the translation initiation factor eIF2alpha on Ser51. Binding of double-stranded RNA to the regulatory domains of PKR promotes dimerization, autophosphorylation, and the functional activation of the kinase. Herein, we identify mutations that activate PKR in the absence of its regulatory domains and map the mutations to a recently identified dimerization surface on the kinase catalytic domain. Mutations of other residues on this surface block PKR autophosphorylation and eIF2alpha phosphorylation, while mutating Thr446, an autophosphorylation site within the catalytic-domain activation segment, impairs eIF2alpha phosphorylation and viral pseudosubstrate binding. Mutational analysis of catalytic-domain residues preferentially conserved in the eIF2alpha kinase family identifies helix alphaG as critical for the specific recognition of eIF2alpha. We propose an ordered mechanism of PKR activation in which catalytic-domain dimerization triggers Thr446 autophosphorylation and specific eIF2alpha substrate recognition.     

Novel aspects of calmodulin target recognition and activation.

Several crystal and NMR structures of calmodulin (CaM) in complex with fragments derived from CaM-regulated proteins have been reported recently and reveal novel ways for CaM to interact with its targets. This review will discuss and compare features of the interaction between CaM and its target domains derived from the plasma membrane Ca2+-pump, the Ca2+-activated K+-channel, the Ca2+/CaM-dependent kinase kinase and the anthrax exotoxin. Unexpected aspects of CaM/target interaction observed in these complexes include: (a) binding of the Ca2+-pump domain to only the C-terminal part of CaM (b) dimer formation with fragments of the K+-channel (c) insertion of CaM between two domains of the anthrax exotoxin (d) binding of Ca2+ ions to only one EF-hand pair and (e) binding of CaM in an extended conformation to some of its targets. The mode of interaction between CaM and these targets differs from binding conformations previously observed between CaM and peptides derived from myosin light chain kinase (MLCK) and CaM-dependent kinase IIalpha (CaMKIIalpha). In the latter complexes, CaM engulfs the CaM-binding domain peptide with its two Ca2+-binding lobes and forms a compact, ellipsoid-like complex. In the early 1990s, a model for the activation of CaM-regulated proteins was developed based on this observation and postulated activation through the displacement of an autoinhibitory or regulatory domain from the target protein upon binding of CaM. The novel structures of CaM-target complexes discussed here demonstrate that this mechanism of activation may be less general than previously believed and seems to be not valid for the anthrax exotoxin, the CaM-regulated K+-channel and possibly also not for the Ca2+-pump.     

Substrate-selective and calcium-independent activation of CaMKII by α-actinin

Protein-protein interactions are thought to modulate the efficiency and specificity of Ca(2+)/calmodulin (CaM)-dependent protein kinase II (CaMKII) signaling in specific subcellular compartments. Here we show that the F-actin-binding protein α-actinin targets CaMKIIα to F-actin in cells by binding to the CaMKII regulatory domain, mimicking CaM. The interaction with α-actinin is blocked by CaMKII autophosphorylation at Thr-306, but not by autophosphorylation at Thr-305, whereas autophosphorylation at either site blocks Ca(2+)/CaM binding. The binding of α-actinin to CaMKII is Ca(2+)-independent and activates the phosphorylation of a subset of substrates in vitro. In intact cells, α-actinin selectively stabilizes CaMKII association with GluN2B-containing glutamate receptors and enhances phosphorylation of Ser-1303 in GluN2B, but inhibits CaMKII phosphorylation of Ser-831 in glutamate receptor GluA1 subunits by competing for activation by Ca(2+)/CaM. These data show that Ca(2+)-independent binding of α-actinin to CaMKII differentially modulates the phosphorylation of physiological targets that play key roles in long-term synaptic plasticity.     

Domain Swapping Used To Investigate the Mechanism of Protein Kinase B Regulation by 3-Phosphoinositide-Dependent Protein Kinase 1 and Ser473 Kinase

Protein kinase B (PKB or Akt), a downstream effector of phosphoinositide 3-kinase (PI 3-kinase), has been implicated in insulin signaling and cell survival. PKB is regulated by phosphorylation on Thr308 by 3-phosphoinositide-dependent protein kinase 1 (PDK1) and on Ser473 by an unidentified kinase. We have used chimeric molecules of PKB to define different steps in the activation mechanism. A chimera which allows inducible membrane translocation by lipid second messengers that activate in vivo protein kinase C and not PKB was created. Following membrane attachment, the PKB fusion protein was rapidly activated and phosphorylated at the two key regulatory sites, Ser473 and Thr308, in the absence of further cell stimulation. This finding indicated that both PDK1 and the Ser473 kinase may be localized at the membrane of unstimulated cells, which was confirmed for PDK1 by immunofluorescence studies. Significantly, PI 3-kinase inhibitors prevent the phosphorylation of both regulatory sites of the membrane-targeted PKB chimera. Furthermore, we show that PKB activated at the membrane was rapidly dephosphorylated following inhibition of PI 3-kinase, with Ser473 being a better substrate for protein phosphatase. Overall, the results demonstrate that PKB is stringently regulated by signaling pathways that control both phosphorylation/activation and dephosphorylation/inactivation of this pivotal protein kinase.     

Chemical quenched flow kinetic studies indicate an intraholoenzyme autophosphorylation mechanism for Ca2+/calmodulin-dependent protein kinase II

Autophosphorylation of alpha-Ca(2+)/calmodulin-dependent protein kinase II (CaM kinase II) at Thr-286 generates Ca(2+)-independent activity that outlasts the initial Ca(2+) stimulus. Previous studies suggested that this autophosphorylation occurs between subunits within each CaM kinase II holoenzyme. However, electron microscopy studies have questioned this mechanism because a large distance separates a kinase domain from its neighboring subunit. Moreover, the recently discovered ability of CaM kinase II holoenzymes to self-associate has raised questions about data interpretation in previous investigations of autophosphorylation. In this work, we characterize the mechanism of CaM kinase II autophosphorylation. To eliminate ambiguity arising from kinase aggregation, we used dynamic light scattering to establish the monodispersity of all enzyme solutions. We then found using chemical quenched flow kinetics that the autophosphorylation rate was independent of the CaM kinase II concentration, results corroborating intraholoenzyme activation. Experiments with a monomeric CaM kinase II showed that phosphorylation of this construct is intermolecular, supporting intersubunit phosphorylation within a holoenzyme. The autophosphorylation rate at 30 degrees C was approximately 12 s(-1), more than 10-fold faster than past estimates. The ability of CaM kinase II to autophosphorylate through an intraholoenzyme, intersubunit mechanism is likely central to its functions of decoding Ca(2+) spike frequency and providing a sustained response to Ca(2+) signals.     

Inactivation and dephosphorylation of protein kinase Balpha (PKBalpha) promoted by hyperosmotic stress.

To study the role of protein kinase B (PKB) in response to cellular stress, we examined PKBalpha activity following different stress treatments. Hyperosmotic but not chemical stress resulted in inactivation of PKBalpha and prevented activation by pervanadate and mitogens. Hyperosmotic shock did not affect the MAP kinase pathway, suggesting that this inhibitory effect was specific for PKB. Our data further indicate that downregulation occurs via dephosphorylation of Thr308 and Ser473, the major regulatory phosphorylation sites of PKBalpha. Indeed, calyculin A, which inhibits protein phosphatases 1 and 2A, effectively blocked hyperosmotic stress-mediated inactivation (dephosphorylation) of PKBalpha. High osmolarity did not affect phosphatidylinositol 3-kinase activity but led to a marked increase in PI(3,4,5)P3 and a decrease in PI(3,4)P2 formation after pervanadate stimulation, suggesting that hyperosmotic stress has an inhibitory effect on a phosphatidylinositol 5-phosphatase which converts PI(3,4,5)P3 into PI(3,4)P2. Immunofluorescence studies revealed that membrane translocation, a prerequisite for PKB activation, was not affected by hyperosmotic stress. Our results indicate that hyperosmotic stress can act at two levels: (i) inhibition of phosphorylation of Thr308 and Ser473 by upstream kinases and (ii) by promoting rapid dephosphorylation of these regulatory sites.     

Agonist-induced formation of FGFR1 homodimers and signaling differ among members of the FGF family

Fibroblast growth factor receptor 1 (FGFR1) is known to be activated by homodimerization in the presence of both the FGF agonist ligand and heparan sulfate glycosaminoglycan. FGFR1 homodimers in turn trigger a variety of downstream signaling cascades via autophosphorylation of tyrosine residues in the cytoplasmic domain of FGFR1. By means of Bioluminescence Energy Resonance Transfer (BRET) as a sign of FGFR1 homodimerization, we evaluated in HEK293T cells the effects of all known FGF agonist ligands on homodimer formation. A significant correlation between BRET² signaling and ERK1/2 phosphorylation was observed, leading to a further characterization of the binding and signaling properties of the FGF subfamilies. FGF agonist ligand-FGFR1 binding interactions appear as the main mechanism for the control of FGFR1 homodimerization and MAPK signaling which demonstrated a high correlation. The bioinformatic analysis demonstrates the interface of the two pro-triplets SSS (Ser–Ser–Ser) and YGS (Tyr–Gly–Ser) located in the extracellular and intracellular domain of the FGFR1. These pro-triplets are postulated participate in the FGFR1 homodimerization interface interaction. The findings also reveal that FGF agonist ligands within the same subfamily of the FGF gene family produced similar increases in FGFR1 homodimer formation and MAPK signaling. Thus, the evolutionary relationship within this gene family appears to have a distinct functional relevance.     

Calmodulin is involved in the Ca2+-dependent activation of ceramide kinase as a calcium sensor

We recently demonstrated that the activation of ceramide kinase (CERK) and the formation of its product, ceramide 1-phosphate (C1P), are necessary for the degranulation pathway in mast cells and that the kinase activity of this enzyme is completely dependent on the intracellular concentration of Ca(2+) (Mitsutake, S., Kim, T.-J., Inagaki, Y., Kato, M., Yamashita, T., and Igarashi, Y. (2004) J. Biol. Chem. 279, 17570-17577). Despite the demonstrated importance of Ca(2+) as a regulator of CERK activity, there are no apparent binding domains in the enzyme and the regulatory mechanism has not been well understood. In the present study, we found that calmodulin (CaM) is involved in the Ca(2+)-dependent activation of CERK. The CaM antagonist W-7 decreased both CERK activity and intracellular C1P formation. Additionally, exogenously added CaM enhanced CERK activity even at low concentrations of Ca(2+). The CERK protein was co-immunoprecipitated with an anti-CaM antibody, indicating formation of intracellular CaM.CERK complexes. An in vitro CaM binding assay also demonstrated Ca(2+)-dependent binding of CaM to CERK. These results strongly suggest that CaM acts as a Ca(2+) sensor for CERK. Furthermore, a CaM binding assay using various mutants of CERK revealed that the binding site of CERK is located within amino acids 422-435. This region appears to include a type 1-8-14B CaM binding motif and is predicted to form an amphipathic helical wheel, which is utilized in CaM recognition. The expression of a deletion mutant of CERK that contained the CaM binding domain but lost CERK activity inhibited the Ca(2+)-dependent C1P formation. These results suggest that this domain could saturate the CaM and hence block Ca(2+)-dependent activation of CERK. Finally, we reveal that in mast cell degranulation CERK acts downstream of CaM, similar to CaM-dependent protein kinase II, which had been assumed to be the main target of CaM in mast cells.     

Inhibitory autophosphorylation of multifunctional Ca2+/calmodulin-dependent protein kinase analyzed by site-directed mutagenesis.

Initial autophosphorylation of multifunctional Ca2+/calmodulin-dependent protein kinase (CaM kinase) occurs at Thr286 (the "autonomy" site) and converts the kinase from a Ca(2+)-dependent to a partially Ca(2+)-independent or autonomous enzyme. After removal of Ca2+/calmodulin, the autonomous kinase undergoes a "burst" of inhibitory autophosphorylation at sites distinct from the autonomy site which may be masked in the presence of bound calmodulin. This burst of Ca(2+)-independent autophosphorylation blocks the ability of calmodulin to activate the kinase. We have used site-directed mutagenesis to replace putative inhibitory autophosphorylation sites within the calmodulin binding domain of recombinant alpha-CaM kinase with nonphosphorylatable alanines and examined the effects on autophosphorylation, kinase activity, and calmodulin binding. Although prominent Ca(2+)-independent autophosphorylation occurs within the calmodulin binding domain at Thr305, Thr306, and Ser314 in wild-type alpha-CaM kinase, the inhibitory effect on kinase activity and calmodulin binding is retained in mutants lacking any one of these three sites. However, when both Thr305 and Thr306 are converted to alanines the kinase does not display inhibition of either activity or calmodulin binding. Autophosphorylation at either Thr305 or Thr306 is therefore sufficient to block both binding and activation of the kinase by Ca2+/calmodulin. Thr306 is also slowly autophosphorylated in a basal reaction in the continuous absence of Ca2+/calmodulin. Autophosphorylation of Thr306 by the kinase in either its basal or autonomous state suggests that in the absence of bound calmodulin, the region of the autoregulatory domain surrounding Thr306, rather than the region near the autonomy site, lies nearest the peptide substrate binding site of the kinase.     

The kinase activation loop is the key to mixed lineage kinase-3 activation via both autophosphorylation and hematopoietic progenitor kinase 1 phosphorylation

We have demonstrated previously that Cdc42 induced MLK-3 homodimerization leads to both autophosphorylation and activation of MLK-3 and postulated that autophosphorylation is an intermediate step of MLK-3 activation following its dimerization. In this report we sought to refine further the mechanism of MLK-3 activation and study the role of the putative kinase activation loop in MLK-3 activation. First we mutated the three potential phosphorylation sites in MLK-3 putative activation loop to alanine in an effort to abrogate MLK-3 autophosphorylation. Mutant T277A displayed almost no autophosphorylation activity and was nearly nonfunctional; mutant S281A, that displayed a low level of autophosphorylation, only slightly activated its downstream targets, whereas the T278A mutant, that exhibited autophosphorylation comparable to that of the wild type, was almost fully functional. Thus, these residues within the activation loop are critical for MLK-3 autophosphorylation and activation. In addition, when the Thr277 and Ser281 residues were mutated to negatively charged glutamic acid to mimic phosphorylated serine/threonine residues, the resulting mutants were fully functional, implying that these two residues may serve as the autophosphorylation sites. Interestingly, HPK1 also phosphorylated MLK-3 activation loop in vitro, and Ser281 was found to be the major phosphorylation site, indicating that HPK1 also activates MLK-3 via phosphorylation of the kinase activation loop.     

Substitution of the autophosphorylation site Thr516 with a negatively charged residue confers constitutive activity to mouse 3-phosphoinositide-dependent protein kinase-1 in cells

3-Phosphoinositide-dependent protein kinase-1 (PDK-1)is a serine/threonine kinase that has been found to phosphorylate and activate several members of the AGC protein kinase family including protein kinase B (Akt), p70 S6 kinase, and protein kinase Czeta. However, the mechanism(s) by which PDK-1 is regulated remains unclear. Here we show that mouse PDK-1 (mPDK-1) undergoes autophosphorylation in vitro on both serine and threonine residues. In addition, we have identified Ser(399) and Thr(516) as the major mPDK-1 autophosphorylation sites in vitro. Furthermore, we have found that these two residues, as well as Ser(244) in the activation loop, are phosphorylated in cells and demonstrated that Ser(244) is a major in vivo phosphorylation site. Abolishment of phosphorylation at Ser(244), but not at Ser(399) or Thr(516), led to a significant decrease of mPDK-1 autophosphorylation and kinase activity in vitro, indicating that autophosphorylation at Ser(399) or Thr(516) is not essential for mPDK-1 autokinase activity. However, overexpression of mPDK-1(T516E), but not of mPDK-1(S244E) or mPDK-1(S399D), in Chinese hamster ovary and HEK293 cells was sufficient to induce Akt phosphorylation at Thr(308) to a level similar to that of insulin stimulation. Furthermore, this increase in phosphorylation was independent of the Pleckstrin homology domain of Akt. Taken together, our results suggest that mPDK-1 undergoes autophosphorylation at multiple sites and that this phosphorylation may be essential for PDK-1 to interact with and phosphorylate its downstream substrates in vivo.     

Structure of the complex of calmodulin with the target sequence of calmodulin-dependent protein kinase I: studies of the kinase activation mechanism

Calcium-saturated calmodulin (CaM) directly activates CaM-dependent protein kinase I (CaMKI) by binding to a region in the C-terminal regulatory sequence of the enzyme to relieve autoinhibition. The structure of CaM in a high-affinity complex with a 25-residue peptide of CaMKI (residues 294-318) has been determined by X-ray crystallography at 1.7 A resolution. Upon complex formation, the CaMKI peptide adopts an alpha-helical conformation, while changes in the CaM domain linker enable both its N- and C-domains to wrap around the peptide helix. Target peptide residues Trp-303 (interacting with the CaM C-domain) and Met-316 (with the CaM N-domain) define the mode of binding as 1-14. In addition, two basic patches on the peptide form complementary charge interactions with CaM. The CaM-peptide affinity is approximately 1 pM, compared with 30 nM for the CaM-kinase complex, indicating that activation of autoinhibited CaMKI by CaM requires a costly energetic disruption of the interactions between the CaM-binding sequence and the rest of the enzyme. We present biochemical and structural evidence indicating the involvement of both CaM domains in the activation process: while the C-domain exhibits tight binding toward the regulatory sequence, the N-domain is necessary for activation. Our crystal structure also enables us to identify the full CaM-binding sequence. Residues Lys-296 and Phe-298 from the target peptide interact directly with CaM, demonstrating overlap between the autoinhibitory and CaM-binding sequences. Thus, the kinase activation mechanism involves the binding of CaM to residues associated with the inhibitory pseudosubstrate sequence.