Distinct roles of MK2 and MK5 in cAMP/PKA- and stress/p38MAPK-induced heat shock protein 27 phosphorylation

Background

Classical mammalian mitogen-activated protein kinase (MAPK) pathways consist of a cascade of three successive phosphorylation events resulting in the phosphorylation of a variety of substrates, including another class of protein kinases referred to as MAPK-activating protein kinases (MAPKAPKs). The MAPKAPKs MK2, MK3 and MK5 are closely related, but MK2 and MK3 are the major downstream targets of the p38^MAPK pathway, while MK5 can be activated by the atypical MAPK ERK3 and ERK4, protein kinase A (PKA), and maybe p38^MAPK. MK2, MK3, and MK5 can phosphorylate the common substrate small heat shock protein 27 (HSP27), a modification that regulates the role of HSP27 in actin polymerization. Both stress and cAMP elevating stimuli can cause F-actin remodeling, but whereas the in vivo role of p38^MAPK-MK2 in stress-triggered HSP27 phosphorylation and actin reorganization is well established, it is not known whether MK2 is involved in cAMP/PKA-induced F-actin rearrangements. On the other hand, MK5 can phosphorylate HSP27 and cause cytoskeletal changes in a cAMP/PKA-dependent manner, but its role as HSP27 kinase in stress-induced F-actin remodeling is disputed. Therefore, we wanted to investigate the implication of MK2 and MK5 in stress- and PKA-induced HSP27 phosphorylation.

Results

Using HEK293 cells, we show that MK2, MK3, and MK5 are expressed in these cells, but MK3 protein levels are very moderate. Stress- and cAMP-elevating stimuli, as well as ectopic expression of active MKK6 plus p38^MAPK or the catalytic subunit of PKA trigger HSP27 phosphorylation, and specific inhibitors of p38^MAPK and PKA prevent this phosphorylation. Depletion of MK2, but not MK3 and MK5 diminished stress-induced HSP27 phosphorylation, while only knockdown of MK5 reduced PKA-induced phosphoHSP27 levels. Stimulation of the p38^MAPK, but not the PKA pathway, caused activation of MK2.

Conclusion

Our results suggest that in HEK293 cells MK2 is the HSP27 kinase engaged in stress-induced, but not cAMP-induced phosphorylation of HSP27, while MK5 seems to be the sole MK to mediate HSP27 phosphorylation in response to stimulation of the PKA pathway. Thus, despite the same substrate specificity towards HSP27, MK2 and MK5 are implicated in different signaling pathways causing actin reorganization.     

PKA-induced F-actin rearrangement requires phosphorylation of Hsp27 by the MAPKAP kinase MK5

Mitogen-activated protein kinase (MAPK) pathways can play a role in F-actin dynamics. In particular, the p38 MAPK/MAPK-activated protein kinase 2 (MK2)/heat shock protein 27 (Hsp27) pathway is involved in F-actin alternations. Previously, we showed that MK5 is implicated in F-actin rearrangement induced by the cAMP/cAMP-dependent protein kinase pathway in PC12 cells, while others found Hsp27 to be a good in vitro MK5 substrate. Here we demonstrate that MK5 can specifically interact with Hsp27 in vivo and can induce phosphorylation at serine residues 78 and 82 in cells. siRNA-mediated depletion of Hsp27 protein levels, as well as overexpression of the non-phosphorylatable Hsp27-3A mutant prevented forskolin-induced F-actin reorganization. While ectopic expression of a constitutive active MK5 mutant was sufficient to induce F-actin rearrangement in PC12 cells, co-expression of Hsp27-3A could ablate this process. Our results imply that MK5 is involved in Hsp27-controlled F-actin dynamics in response to activation of the cAMP-dependent protein kinase pathway. These findings render the MK5/Hsp27 connection into a putative therapeutic target for conditions with aberrant Hsp27 phosphorylation such as metastasis, cardiovascular diseases, muscle atrophy, autoimmune skin disease and neuropathology.     

Mechanism of inhibition of Raf-1 by protein kinase A.

The cytoplasmic Raf-1 kinase is essential for mitogenic signalling by growth factors, which couple to tyrosine kinases, and by tumor-promoting phorbol esters such as 12-O-tetradecanoylphorbol-13-acetate, which activate protein kinase C (PKC). Signalling by the Raf-1 kinase can be blocked by activation of the cyclic AMP (cAMP)-dependent protein kinase A (PKA). The molecular mechanism of this inhibition is not precisely known but has been suggested to involve attenuation of Raf-1 binding to Ras. Using purified proteins, we show that in addition to weakening the interaction of Raf-1 with Ras, PKA can inhibit Raf-1 function directly via phosphorylation of the Raf-1 kinase domain. Phosphorylation by PKA interferes with the activation of Raf-1 by either PKC alpha or the tyrosine kinase Lck and even can downregulate the kinase activity of Raf-1 previously activated by PKC alpha or amino-terminal truncation. This type of inhibition can be dissociated from the ability of Raf-1 to associate with Ras, since (i) the isolated Raf-1 kinase domain, which lacks the Ras binding domain, is still susceptible to inhibition by PKA, (ii) phosphorylation of Raf-1 by PKC alpha alleviates the PKA-induced reduction of Ras binding but does not prevent the downregulation of Raf-1 kinase activity by PKA and (iii) cAMP agonists antagonize transformation by v-Raf, which is Ras independent.     

Arachidonic acid promotes phosphorylation of 5-lipoxygenase at Ser-271 by MAPK-activated protein kinase 2 (MK2)

We demonstrated previously that 5-lipoxygenase (5-LO), a key enzyme in leukotriene biosynthesis, can be phosphorylated by p38 MAPK-regulated MAPKAP kinases (MKs). Here we show that mutation of Ser-271 to Ala in 5-LO abolished MK2 catalyzed phosphorylation and clearly reduced phosphorylation by kinases prepared from stimulated polymorphonuclear leukocytes and Mono Mac 6 cells. Compared with heat shock protein 27 (Hsp-27), 5-LO was a weak substrate for MK2. However, the addition of unsaturated fatty acids (i.e. arachidonate 1-50 microm) up-regulated phosphorylation of 5-LO, but not of Hsp-27, by active MK2 in vitro, resulting in a similar phosphorylation as for Hsp-27. 5-LO was phosphorylated also by other serine/threonine kinases recognizing the motif Arg-Xaa-Xaa-Ser (protein kinase A, Ca(2+)/calmodulin-dependent kinase II), but these activities were not increased by fatty acids. HeLa cells expressing wild type 5-LO or S271A-5-LO, showed prominent 5-LO activity when incubated with Ca(2+)-ionophore plus arachidonate. However, when stimulated with only exogenous arachidonic acid, activity for the S271A mutant was significantly lower as compared with wild type 5-LO. It appears that phosphorylation at Ser-271 is more important for 5-LO activity induced by a stimulus that does not prominently increase intracellular Ca(2+) and that arachidonic acid stimulates leukotriene biosynthesis also by promoting this MK2-catalyzed phosphorylation.     

Transcription factor AP-2 activity is modulated by protein kinase A-mediated phosphorylation

We recently reported that APOE promoter activity is stimulated by cAMP, this effect being mediated by factor AP-2 [Garcia et al. (1996) J. Neurosci. 16, 7550-7556]. Here, we study whether cAMP-induced phosphorylation modulates the activity of AP-2. Recombinant AP-2 was phosphorylated in vitro by protein kinase A (PKA) at Ser239. Mutation of Ser239 to Ala abolished in vitro phosphorylation of AP-2 by PKA, but not the DNA binding activity of AP-2. Cotransfection studies showed that PKA stimulated the effect of AP-2 on the APOE promoter, but not that of the S239A mutant. Therefore, cAMP may modulate AP-2 activity by PKA-induced phosphorylation of this factor.     

Regulation of Ca(2+)/calmodulin-dependent protein kinase kinase alpha by cAMP-dependent protein kinase: I. Biochemical analysis

Ca(2+)/calmodulin-dependent protein kinases (CaM-kinases) I and IV are activated upon phosphorylation of their Thr(177) and Thr(196), respectively, by the upstream Ca(2+)/calmodulin-dependent protein kinases CaM-kinase kinase alpha and beta, and deactivated upon dephosphorylation by protein phosphatases such as CaM-kinase phosphatase. Recent studies demonstrated that the activity of CaM-kinase kinase alpha is decreased upon phosphorylation by cAMP-dependent protein kinase (PKA), and the relationship between the inhibition and phosphorylation of CaM-kinase kinase alpha by PKA has been studied. In the present study, we demonstrate that the activity of CaM-kinase kinase alpha toward PKIV peptide, which contains the sequence surrounding Thr(196) of CaM-kinase IV, is increased by incubation with PKA in the presence of Ca(2+)/calmodulin but decreased in its absence, while the activity toward CaM-kinase IV is decreased by incubation with PKA in both the presence and absence of Ca(2+)/calmodulin. Six phosphorylation sites on CaM-kinase kinase alpha, Ser(24) for autophosphorylation, and Ser(52), Ser(74), Thr(108), Ser(458), and Ser(475) for phosphorylation by PKA, were identified by amino acid sequence analysis of the phosphopeptides purified from the tryptic digest of the phosphorylated enzymes. The presence of Ca(2+)/calmodulin suppresses phosphorylation on Ser(52), Ser(74), Thr(108), and Ser(458) by PKA, but accelerates phosphorylation on Ser(475). The changes in the activity of the enzyme upon phosphorylation appear to occur as a result of conformational changes induced by phosphorylation on several sites.     

Ethanol-induced translocation of cAMP-dependent protein kinase to the nucleus. Mechanism and functional consequences.

Ethanol induces translocation of the catalytic subunit (Calpha) of cAMP-dependent protein kinase (PKA) from the Golgi area to the nucleus in NG108-15 cells. Ethanol also induces translocation of the RIIbeta regulatory subunit of PKA to the nucleus; RI and Cbeta are not translocated. Nuclear PKA activity in ethanol-treated cells is no longer regulated by cAMP. Gel filtration and immunoprecipitation analysis confirm that ethanol blocks the reassociation of Calpha with RII but does not induce dissociation of these subunits. Ethanol also reduces inhibition of Calpha by the PKA inhibitor PKI. Pre-incubation of Calpha with ethanol decreases phosphorylation of Leu-Arg-Arg-Ala-Ser-Leu-Gly (Kemptide) and casein but has no effect on the phosphorylation of highly charged molecules such as histone H1 or protamine. cAMP-response element-binding protein (CREB) phosphorylation by Calpha is also increased in ethanol-treated cells. This increase in CREB phosphorylation is inhibited by the PKA antagonist (R(p))-cAMPS and by an adenosine receptor antagonist. These results suggest that ethanol affects a cascade of events allowing for sustained nuclear localization of Calpha and prolonged CREB phosphorylation. These events may account for ethanol-induced changes in cAMP-dependent gene expression.     

Enhanced dephosphorylation of cAMP-dependent protein kinase by oxidation and thiol modification

The catalytic subunit of cAMP-dependent protein kinase (PKA) is phosphorylated at threonine 197 and serine 338. Phosphorylation of threonine 197, located in the activation loop, is required for coordinating the active site conformation and optimal enzymatic activity. However, this phosphorylation has not been widely appreciated as a regulatory site because of the apparent constitutive nature of the phosphorylation and the general resistance of the kinase to phosphatase treatment. We demonstrate here that the observed resistance of the catalytic subunit to dephosphorylation is due, in part, to the presence of the highly nucleophilic cysteine 199 located proximal to the phosphate on threonine 197. Experiments performed in vitro demonstrated that mutation (cysteine 199 to alanine), oxidation, such as by glutathionylation or internal disulfide bond formation, or alkylation of the C-subunit enhanced its ability to be dephosphorylated. Furthermore, rephosphorylation of reduced C-subunit by PDK1 created a cycle whereby the inactive kinase could be reactivated. To demonstrate that thiol modification of PKA can lead to enhanced dephosphorylation in vivo, PC12 cells were treated with N-ethylmaleimide (NEM). Such treatment resulted in complete PKA inactivation and dephosphorylation of threonine 197. This effect of NEM was contingent upon prior treatment of the cells with PKA activators, demonstrating the resistance of the holoenzyme to thiol alkylation-mediated dephosphorylation. Our results also demonstrated that NEM treatment of PC12 cells enhanced the dephosphorylation of the protein kinase Calpha activation loop, suggesting a common mechanism of regulation among members of the AGC family of kinases.     

Mechanism of protein kinase B activation by cyclic AMP-dependent protein kinase.

Activation of protein kinase B (PKB) by growth factors and hormones has been demonstrated to proceed via phosphatidylinositol 3-kinase (PI3-kinase). In this report, we show that PKB can also be activated by PKA (cyclic AMP [cAMP]-dependent protein kinase) through a PI3-kinase-independent pathway. Although this activation required phosphorylation of PKB, PKB is not likely to be a physiological substrate of PKA since a mutation in the sole PKA consensus phosphorylation site of PKB did not abolish PKA-induced activation of PKB. In addition, mechanistically, this activation was different from that of growth factors since it did not require phosphorylation of the S473 residue, which is essential for full PKB activation induced by insulin. These data were supported by the fact that mutation of residue S473 of PKB to alanine did not prevent it from being activated by forskolin. Moreover, phosphopeptide maps of overexpressed PKB from COS cells showed differences between insulin- and forskolin-stimulated cells that pointed to distinct activation mechanisms of PKB depending on whether insulin or cAMP was used. We looked at events downstream of PKB and found that PKA activation of PKB led to the phosphorylation and inhibition of glycogen synthase kinase-3 (GSK-3) activity, a known in vivo substrate of PKB. Overexpression of a dominant negative PKB led to the loss of inhibition of GSK-3 in both insulin- and forskolin-treated cells, demonstrating that PKB was responsible for this inhibition in both cases. Finally, we show by confocal microscopy that forskolin, similar to insulin, was able to induce translocation of PKB to the plasma membrane. This process was inhibited by high concentrations of wortmannin (300 nM), suggesting that forskolin-induced PKB movement may require phospholipids, which are probably not generated by class I or class III PI3-kinase. However, high concentrations of wortmannin did not abolish PKB activation, which demonstrates that translocation per se is not important for PKA-induced PKB activation.     

Characterization of the cAMP-dependent protein kinase catalytic subunit Cgamma expressed and purified from sf9 cells

The Cgamma and Calpha subunits of the cAMP-dependent protein kinase (PKA) contain 350 amino acids that are highly homologous (83% amino acid sequence), with 91% homology within the catalytic domain (a.a. 40-300). Unlike Cgamma, the Calpha subunit has been readily purified and characterized as a recombinant protein in vitro, in intact cells, and in vivo. This report describes for the first time the expression, purification, and characterization of Cgamma. The expression of active Cgamma was eukaryote-specific, from mammalian and insect cells, but not bacteria. Active recombinant Cgamma was optimally expressed and purified to homogeneity from Sf9 cells with a 273-fold increase in specific activity and a 21% recovery after sequential CM-Sepharose and Sephacryl S-300 chromatography. The specific activity of pure Cgamma was 0.31 and 0.81 U/mg with kemptide and histone as substrates, respectively. Physical characterization showed Cgamma had a lower apparent molecular weight and Stokes radii than Calpha, suggesting differences in tertiary structures. Steady-state kinetics demonstrated that like Calpha and Cbeta, Cgamma phosphorylates substrates requiring basic amino acids at P-3 and P-2. However, Cgamma generally exhibited a lower Km and Vmax than Calpha for peptide substrates tested. Cgamma also exhibited a distinct pseudosubstrate specificity showing inhibition by homogeneous preparations of RIalpha and RIIalpha-subunits, but not by pure recombinant protein kinase inhibitors PKIalpha and PKIbeta, PKA-specific inhibitors. These studies suggest that Cgamma and Calpha exhibit differences in structure and function in vitro, supporting the hypothesis that functionally different C-subunit isozymes could diversify and/or fine-tune cAMP signal transduction downstream of PKA activation.     

Cyclic AMP-dependent protein kinase phosphorylates merlin at serine 518 independently of p21-activated kinase and promotes merlin-ezrin heterodimerization

Mutations in the NF2 tumor suppressor gene encoding merlin induce the development of tumors of the nervous system. Merlin is highly homologous to the ERM (ezrin-radixin-moesin) family of membrane/cytoskeleton linker proteins. However, the mechanism for the tumor suppressing activity of merlin is not well understood. Previously, we characterized a novel role for merlin as a protein kinase A (PKA)-anchoring protein, which links merlin to the cAMP/PKA signaling pathway. In this study we show that merlin is also a target for PKA-induced phosphorylation. In vitro [gamma-(33)P]ATP labeling revealed that both the merlin N and C termini are phosphorylated by PKA. Furthermore, both in vitro and in vivo phosphorylation studies of the wild-type and mutated C termini demonstrated that PKA can phosphorylate merlin at serine 518, a site that is phosphorylated also by p21-activated kinases (PAKs). Merlin was phosphorylated by PKA in cells in which PAK activity was suppressed, indicating that the two kinases function independently. Both in vitro and in vivo interaction studies indicated that phosphorylation of serine 518 promotes heterodimerization between merlin and ezrin, an event suggested to convert merlin from a growth-suppressive to a growth-permissive state. This study provides further evidence on the connection between merlin and cAMP/PKA signaling and suggests a role for merlin in the cAMP/PKA transduction pathway.     

Elimination of protein kinase MK5/PRAK activity by targeted homologous recombination

MK5 (mitogen-activated protein kinase [MAPK]-activated protein kinase 5), also designated PRAK (p38-regulated and -activated kinase), was deleted from mice by homologous recombination. Although no MK5 full-length protein and kinase activity was detected in the MK5 knockout mice, the animals were viable and fertile and did not display abnormalities in tissue morphology or behavior. In addition, these mice did not show increased resistance to endotoxic shock or decreased lipopolysaccharide-induced cytokine production. Hence, MK5 deletion resulted in a phenotype very different from the complex inflammation-impaired phenotype of mice deficient in MK2, although MK2 and MK5 exhibit evolutional, structural, and apparent extensive functional similarities. To explain this discrepancy, we used wild-type cells and embryonic fibroblasts from both MK2 and MK5 knockout mice as controls to reexamine the mechanism of activation, the interaction with endogenous p38 MAPK, and the substrate specificity of both enzymes. In contrast to MK2, which shows interaction with and chaperoning properties for p38 MAPK and which is activated by extracellular stresses such as arsenite or sorbitol treatment, endogenous MK5 did not show these properties. Furthermore, endogenous MK5 is not able to phosphorylate Hsp27 in vitro and in vivo. We conclude that the differences between the phenotypes of MK5- and MK2-deficient mice result from clearly different functional properties of both enzymes.     

Induction of immediate early genes by Ca2+ influx requires cAMP-dependent protein kinase in PC12 cells

Agents that activate cAMP-dependent protein kinase (PKA) as well as agents that increase intracellular calcium induce the expression of certain immediate early genes (IEGs). Recently, it has been demonstrated that the same cis-acting element in the 5' region of the c-fos gene has the ability to mediate both cAMP- and calcium-induced c-fos expression in PC12 cells (Sheng, M., McFadden, G., and Greenberg, M. (1990) Neuron 4, 571-582). Here we demonstrate that both cAMP- and calcium-mediated induction of c-fos and egr1 are dependent on PKA activity. Addition of either depolarizing concentrations of KCl or the calcium ionophore, ionomycin, to PC12 cells increased the expression of both c-fos and egr1, but these inductions were dramatically reduced in three PKA-deficient cell lines, 123.7, AB.11, and A126-1B2. Furthermore, pretreatment of PC12 cells with 20 microM H89, a specific inhibitor of PKA, inhibited forskolin, dibutyryl cAMP, and KCl-induced c-fos and egr1 induction, while having no effect on NGF induction. Likewise, in the PKA-deficient cells, NGF or an activator of protein kinase C induced c-fos and egr1 normally. To determine if PKA deficiency modifies the ability of Ca2+ to activate calcium-dependent kinases, autophosphorylation of multifunctional Ca2+/calmodulin-dependent protein kinase (CaM kinase) in response to Ca2+ influx was determined. In parental PC12 cells, PC12 cells pretreated with H89, and PKA-deficient cell lines, CaM kinase was activated equivalently in response to KCl depolarization. These results suggest that PKA is not required for Ca(2+)-induced increase in CaM kinase activity and that the induction of IEGs in response to Ca2+ influx is PKA-dependent. Thus, the requirement for PKA resides at a point distal to the activation of calmodulin-dependent processes.     

Physiological roles of mitogen-activated-protein-kinase-activated p38-regulated/activated protein kinase

Mitogen-activated protein kinases (MAPKs) are a family of proteins that constitute signaling pathways involved in processes that control gene expression, cell division, cell survival, apoptosis, metabolism, differentiation and motility. The MAPK pathways can be divided into conventional and atypical MAPK pathways. The first group converts a signal into a cellular response through a relay of three consecutive phosphorylation events exerted by MAPK kinase kinases, MAPK kinase, and MAPK. Atypical MAPK pathways are not organized into this three-tiered cascade. MAPK that belongs to both conventional and atypical MAPK pathways can phosphorylate both non-protein kinase substrates and other protein kinases. The latter are referred to as MAPK-activated protein kinases. This review focuses on one such MAPK-activated protein kinase, MAPK-activated protein kinase 5 (MK5) or p38-regulated/activated protein kinase (PRAK). This protein is highly conserved throughout the animal kingdom and seems to be the target of both conventional and atypical MAPK pathways. Recent findings on the regulation of the activity and subcellular localization, bona fide interaction partners and physiological roles of MK5/PRAK are discussed.     

Dissecting the role of 5'-AMP for allosteric stimulation, activation, and deactivation of AMP-activated protein kinase

AMP-activated protein kinase (AMPK) is a heterotrimeric protein kinase that is crucial for cellular energy homeostasis of eukaryotic cells and organisms. Here we report on the activation of AMPK alpha1beta1gamma1 and alpha2beta2gamma1 by their upstream kinases (Ca(2+)/calmodulin-dependent protein kinase kinase-beta and LKB1-MO25alpha-STRADalpha), the deactivation by protein phosphatase 2Calpha, and on the extent of stimulation of AMPK by its allosteric activator AMP, using purified recombinant enzyme preparations. An accurate high pressure liquid chromatography-based method for AMPK activity measurements was established, which allowed for direct quantitation of the unphosphorylated and phosphorylated artificial peptide substrate, as well as the adenine nucleotides. Our results show a 1000-fold activation of AMPK by the combined effects of upstream kinase and saturating concentrations of AMP. The two AMPK isoforms exhibit similar specific activities (6 mumol/min/mg) and do not differ significantly by their responsiveness to AMP. Due to the inherent instability of ATP and ADP, it proved impossible to assay AMPK activity in the absolute absence of AMP. However, the half-maximal stimulatory effect of AMP is reached below 2 microm. AMP does not appear to augment phosphorylation by upstream kinases in the purified in vitro system, but deactivation by dephosphorylation of AMPK alpha-subunits at Thr-172 by protein phosphatase 2Calpha is attenuated by AMP. Furthermore, it is shown that neither purified NAD(+) nor NADH alters the activity of AMPK in a concentration range of 0-300 microm, respectively. Finally, evidence is provided that ZMP, a compound formed in 5-aminoimidazole-4-carboxamide-1-beta-D-ribofuranoside-treated cells to activate AMPK in vivo, allosterically activates purified AMPK in vitro, but compared with AMP, maximal activity is not reached. These data shed new light on physiologically important aspects of AMPK regulation.     

Phosphorylation and activation of brain aromatic L-amino acid decarboxylase by cyclic AMP-dependent protein kinase

Aromatic L-amino acid decarboxylase (AAAD), an enzyme required for the synthesis of catecholamines, indoleamines, and trace amines, is rapidly activated by cyclic AMP-dependent pathways in striatum and midbrain in vivo, suggesting enzyme phosphorylation. We now report that the catalytic subunit of cyclic AMP-dependent protein kinase (PKA) directly phosphorylated AAAD immunoprecipitated from homogenates prepared from the mouse striatum and midbrain in vitro. Under the same phosphorylation conditions, the catalytic subunit of PKA also phosphorylated a recombinant AAAD protein expressed in Escherichia coli transfected with an AAAD cDNA isolated from the bovine adrenal gland. The PKA-induced AAAD phosphorylation of immunoprecipitates from striatum and midbrain was time and concentration dependent and blocked by a specific PKA peptide inhibitor. Incubation of the catalytic subunit of PKA with striatal homogenates increased enzyme activity by approximately 20% in a time- and concentration-dependent manner. Moreover, incubation of the catalytic subunit of PKA with recombinant AAAD increased activity by approximately 70%. A direct phosphorylation of AAAD protein by PKA might underlie the cyclic AMP-induced rapid and transient activation of AAAD in vivo.