C2 domain of protein kinase C alpha: elucidation of the membrane docking surface by site-directed fluorescence and spin labeling.

The C2 domain is a conserved signaling motif that triggers membrane docking in a Ca(2+)-dependent manner, but the membrane docking surfaces of many C2 domains have not yet been identified. Two extreme models can be proposed for the docking of the protein kinase C alpha (PKC alpha) C2 domain to membranes. In the parallel model, the membrane-docking surface includes the Ca(2+) binding loops and an anion binding site on beta-strands 3-4, such that the beta-strands are oriented parallel to the membrane. In the perpendicular model, the docking surface is localized to the Ca(2+) binding loops and the beta-strands are oriented perpendicular to the membrane surface. The present study utilizes site-directed fluorescence and spin-labeling to map out the membrane docking surface of the PKC alpha C2 domain. Single cysteine residues were engineered into 18 locations scattered over all regions of the protein surface, and were used as attachment sites for spectroscopic probes. The environmentally sensitive fluorescein probe identified positions where Ca(2+) activation or membrane docking trigger measurable fluorescence changes. Ca(2+) binding was found to initiate a global conformational change, while membrane docking triggered the largest fluorescein environmental changes at labeling positions on the three Ca(2+) binding loops (CBL), thereby localizing these loops to the membrane docking surface. Complementary EPR power saturation measurements were carried out using a nitroxide spin probe to determine a membrane depth parameter, Phi, for each spin-labeled mutant. Positive membrane depth parameters indicative of membrane insertion were found for three positions, all located on the Ca(2+) binding loops: N189 on CBL 1, and both R249 and R252 on CBL 3. In addition, EPR power saturation revealed that five positions near the anion binding site are partially protected from collisions with an aqueous paramagnetic probe, indicating that the anion binding site lies at or near the surface of the headgroup layer. Together, the fluorescence and EPR results indicate that the Ca(2+) first and third Ca(2+) binding loops insert directly into the lipid headgroup region of the membrane, and that the anion binding site on beta-strands 3-4 lies near the headgroups. The data support a model in which the beta-strands are tilted toward the parallel orientation relative to the membrane surface.

Effects of prenyl pyrophosphates on the binding of PKCgamma with RACK1.

Receptors for activated C kinase (RACKs) are a group of PKC binding proteins that have been shown to mediate isoform-selective functions of PKC and to be crucial in the translocation and subsequent functioning of the PKC isoenzymes on activation. RACK1 cDNA from the shrimp Penaeus japonicus was isolated by homology cloning. The hepatopancreas cDNA from this shrimp was found to encode a 318-residue polypeptide whose predicted amino acid sequence shared 91% homology with human G(beta2)-like proteins. Expression of the cDNA of shrimp RACK1 in vitro yielded a 45-kDa polypeptide with positive reactivity toward the monoclonal antibodies against RACK1 of mammals. The shrimp RACK1 was biotinylated and used to compare the effects of geranylgeranyl pyrophosphate and farnesyl pyrophosphate on its binding with PKCgamma in anti-biotin-IgG precipitates. PKCgammas were isolated from shrimp eyes and mouse brains. Both enzyme preparations were able to inhibit taxol-induced tubulin polymerization. Interestingly, when either geranylgeranyl pyrophosphate or farnesyl pyrophosphate was reduced to the submicrogram level, the recruitment activity of RACK1 with purified PKCgamma was found to increase dramatically. The activation is especially significant for RACK1 and PKCgamma from different species. The observation implies that the deprivation of prenyl pyrophosphate might function as a signal for RACK1 to switch the binding from the conventional isoenzymes of PKC (cPKC) to the novel isoenzymes of PKC (nPKC). A hydrophobic binding pocket for geranylgeranyl pyrophosphate in RACK1 is further revealed via prenylation with protein geranylgeranyl transferase I of shrimp P. japonicus.

Phosphorylation of threonine 638 critically controls the dephosphorylation and inactivation of protein kinase Calpha.

BACKGROUND: It has been widely reported that multisite phosphorylation plays an essential role in the regulation of protein kinases. However, our understanding of how these events modify protein function in vitro and in vivo is poorly understood. Protein kinase C (PKC) affords an interesting example of how phosphorylation control is coupled to effector control. PKC is acutely regulated by the second messenger diacylglycerol; however, it is also known to undergo multisite phosphorylation. Previously, we and others have shown that one site in the 'activation loop' of PKCalpha (a threonine residue at position 497; T497) and PKCbeta (T500) is essential for the catalytic competence of these proteins. More recently, a carboxy-terminal site (T638 in PKCalpha) has been implicated. In this report, we investigate the role of this site and its interaction with the catalytic core site. RESULTS: We have analyzed mutant PKCalpha proteins, in which amino-acid substitutions were made at the T638 site, and shown that phosphorylation at this site affects the conformation of the protein, as judged by thermal stability, and sensitivity to oxidation, trypsin and phosphatase treatment. This supersensitivity to dephosphorylation in vitro was also seen in an agonist-dependent context in vivo. We have also shown that phosphorylation of this site is not essential for catalytic activity of the purified protein. The molecular basis of the control operating through the T638 site was provided by the evidence of a functional interaction with the previously described catalytic core site, T497. This inter-relationship was further established by the demonstration that the E497 mutant protein had a thermal instability and phosphatase supersensitivity similar to that of the A638 and E638 mutants. CONCLUSIONS: The T638 phosphorylation site is not required for the catalytic function of PKCalpha per se, but serves to control the duration of activation by regulating the rate of dephosphorylation and inactivation of the protein. This is achieved through the cooperative interaction between the T638 and T497 sites; if either of these residues is not phosphorylated, the protein is supersensitive to phosphatase action. This model of PKCalpha function is likely to be of general significance to the protein kinase superfamily, where similarly juxtaposed sites exist. We conclude that dephosphorylation of PKCalpha, and, by inference, other protein kinases, is regulated by multisite phosphorylation.

Stimulation of protein phosphatase-1 activity by phorbol esters. Evaluation of the regulatory role of protein kinase C in insulin action.

In this study, we examined the role of insulin, protein kinase C (PKC) and mitogen-activated protein kinase (MAPK) cascade in activation of protein phosphatase-1 (PP-1) by using three complementary approaches. First, differentiated L6 cells were acutely exposed to 12-O-tetradecanoylphorbol-13-acetate (TPA, 400 nM) to activate PKC. In these cells, TPA caused 32% stimulation of PP-1 activity. The PP-1 stimulation by TPA was comparable to stimulation by insulin (t1/2 = 1 min and EC50 = 5 nM) with a maximum effect in 5 min. The effects of insulin and TPA were not additive. Insulin and TPA also stimulated MAPK (> 2-fold increase over basal, with myelin basic protein as a substrate). ML-9, a myosin light chain kinase inhibitor, blocked the effects of insulin and TPA on both MAPK and PP-1 activation. In the second approach, PKC was down-regulated by chronic treatment with TPA. In these cells subsequent effects of insulin on MAPK and PP-1 activation were blocked, without an effect on basal enzyme levels. In the third approach, two selective inhibitors of PKC, calphostin and chelerythrine chloride, were used to inhibit PKC. These inhibitors completely prevented insulin and TPA stimulation of MAPK and PP-1 and blocked insulin-induced translocation of PKC to the plasma membranes. We conclude that PKC plays an important role in insulin stimulation of PP-1 via the activation of MAPK cascade.

A new member of the third class in the protein kinase C family, PKC lambda, expressed dominantly in an undifferentiated mouse embryonal carcinoma cell line and also in many tissues and cells.

Protein kinase C (PKC)-related cDNA clones isolated from cDNA libraries of mouse P19 embryonal carcinoma cells and mouse brain encoded a 67-kDa protein, PKC lambda. PKC lambda shows the highest amino acid sequence identity with PKC zeta (72%), the third class of the PKC family. Northern blot analysis showed that the mRNA for PKC lambda is expressed in a wide variety of cells and tissues, including P19 and NIH 3T3 cells, as well as brain, kidney, testis, and ovary. In undifferentiated P19 cells, the mRNA for PKC lambda is the most abundant among all the PKC family members. The differentiation of P19 cells results in an increase in PKC alpha and epsilon, and a decrease in PKC lambda. Antiserum raised against a peptide of PKC lambda identified a 74-kDa protein in P19 cell extracts as well as in extracts from COS cells transfected with the PKC lambda expression plasmid. Autophosphorylation of the PKC lambda that immunoprecipitated with the specific antiserum was observed, indicating that PKC lambda possesses protein kinase activity. A phorbol ester binding assay using intact COS cells expressing PKC lambda failed to detect binding activity specific to PKC lambda at phorbol dibutylate concentrations up to 300 nM, suggesting that PKC lambda does not possess phorbol ester binding activity. These results, in conjunction with the results obtained in parallel experiments with PKC zeta and other PKC members, suggest a biochemical similarity between PKC lambda and zeta and their clear difference from other PKC members.

[Inhibition of protein kinase C by stilbenoids].

The effect of 15 stilbenoids on protein kinase C (PKC) was studied in order to search for naturally occurring PKC inhibitors. All these compounds were isolated from Chinese medicines. Three oligomeric stilbenes from Caragana sinica, alpha-viniferin, kobophenol A and miyabenol C, were shown to intensely inhibit the activity of partially purified rat brain PKC with IC50 values of 62.5, 52.0, and 27.5 mumol.L-1, respectively. Kinetic analyses revealed that that inhibition was noncompetitive. The other compounds also showed the effect. Monomer stilbenes exhibited PKC inhibitory activity at higher mumol.L-1 concentrations than oligomeric stilbenes. Whenever they are methylated or acetylated perfectly, the inhibition weakens or disappears.

Modulation of rat liver protein kinase C during "in vivo" CC14-induced oxidative stress.

Rat intoxication with a single dose of the hepatotoxin carbon tetrachloride induces a significant modification of liver protein kinase C total activity which depends on the degree of the intrahepatocyte oxidative unbalance provoked by various concentrations of the haloalkane. Low carbon tetrachloride amounts stimulate total protein kinase C activity, while one order of magnitude higher amounts exert strong enzyme inhibition. The latter effect is due to an early inactivation followed with progress of time by a proteolytic degradation of the enzyme. A pathological recruitment of the calcium-dependent protein kinase C regulatory enzymes calpain and calpastatin appears responsible for protein kinase C loss. The prolonged excess of cytosolic calcium which characterizes the single high dose carbon tetrachloride poisoning also leads to inactivation of calpain II and calpastatin in a time-dependent manner.

Mouse protein kinase C-delta, the major isoform expressed in mouse hemopoietic cells: sequence of the cDNA, expression patterns, and characterization of the protein.

A complementary DNA (cDNA) of 2559 bp which encode all 674 amino acids of mouse protein kinase C-delta (PKC-delta) has been isolated from a cDNA library prepared from ABPL-2, a mouse myeloid tumor. The library was screened with a partial PKC-delta cDNA clone that had been created by polymerase chain reaction (PCR) amplification of ABPL-2 RNA using primers that are conserved among all rat PKC isozymes. This approach proved to be a distinct improvement over screening with synthetic oligonucleotides. Similar sets of cDNAs prepared from other hemopoietic cell lines were screened with this PKC-delta cDNA and with probes for the other PKC isoforms. These experiments revealed that the major isoform of PKC expressed in hemopoietic cells is PKC-delta. PKC-delta protein was purified from ABPL-3, a mouse myeloid tumor which expressed principally the delta isoform of PKC. The protein eluted from a hydroxylapatite column in the same position as PKC-beta and -epsilon would elute, if present. The kinase activity of purified PKC-delta showed strict dependence on the presence of phospholipids, but showed no activation by Ca2+.

Expression of a functional protein kinase C-gamma using a baculovirus vector: purification and characterisation of a single protein kinase C iso-enzyme.

The cloning of complementary DNAs for protein kinase C (PKC) has revealed a multi-gene family of closely related protein kinases [Parker et al. (1986) Science 233, 853-859; Coussens et al. (1986) Science 233, 859-866]. In vivo, the distribution of the PKC isoenzymes follows a fairly tissue-specific pattern suggesting that functional differences exist between the members of this kinase family. To initiate a detailed characterisation of the individual isoenzymes, and as an alternative approach to purifying and separating the individual PKC types and their splice variants from mammalian tissues, we have expressed the bovine PKC type gamma in insect cells using a baculovirus expression vector. The bovine protein constitutes one of the major proteins in infected cells and can be purified to near homogeneity by a 2-step procedure. Analysis of the purified protein confirms that it has authentic mammalian PKC characteristics with respect to phospholipid dependence and phorbol ester binding. The bovine PKC gamma purified from infected cells is post-translationally modified and resolves into a doublet of molecular weights 82,000 and 84,000 upon SDS-polyacrylamide gel electrophoresis. These two size classes of polypeptides appear to result from differential phosphorylation as demonstrated by sensitivity to protein phosphatase treatment. The applicability and the potential of this system for the analysis of the various mammalian PKC isoenzymes is discussed.

Purification and characterization of an isoform of protein kinase C from bovine neutrophils.

Protein kinase C (PKC) from bovine neutrophils was purified 1420-fold. Subcellular fractionation analysis of bovine neutrophil homogenate in the presence of EGTA indicated that more than 95% of the PKC activity was present in the soluble fraction. The purification procedure from cytosol involved sequential chromatographic steps on DE-52 cellulose, Mono Q, and phenyl-Sepharose. Whereas bovine brain PKC could be resolved into four isoenzymatic forms by chromatography on a hydroxylapatite column, bovine neutrophil PKC was eluted in a single peak, suggesting that it corresponded to a single isoform. The apparent molecular weight of bovine neutrophil PKC was 82,000, as judged by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. By filtration on Sephadex G-150, a molecular weight of 85,000 was calculated, indicating that bovine neutrophil PKC in solution is monomeric. Its isoelectric point was 5.9 +/- 0.1. Bovine neutrophil PKC was autophosphorylated in the presence of [gamma-32P]ATP, provided that the medium was supplemented with Mg2+, Ca2+, phosphatidylserine, and diacylglycerol; phorbol myristate acetate could substitute for diacylglycerol. Autophosphorylated PKC could be cleaved by trypsin to generate two radiolabeled peptides of Mr 48,000 and 39,000. The labeled amino acids were serine and threonine. During the course of the purification procedure of bovine neutrophil PKC, a protein of Mr 23,000, which was abundant in the cytosolic fraction of the homogenate, was found to exhibit a strong propensity to PKC-dependent phosphorylation in the presence of [gamma-32P]ATP, Mg2+, Ca2+, phosphatidylserine, and diacylglycerol. This protein was recovered together with PKC in one of the two active peaks eluted from the Mono Q column at the second step of PKC purification.(ABSTRACT TRUNCATED AT 250 WORDS)

Purification and properties of protein kinase C from rabbit reticulocyte lysates.

We have previously reported that addition of Ca2+ and phospholipid (PL) inhibits translation in hemin-containing reticulocyte lysates through activation of a eukaryotic protein synthesis initiation factor (eIF-2) kinase. The possibility that this activation was mediated by a Ca2+-PL-dependent protein kinase (protein kinase C, PKC) appeared unlikely by the observation that it was prevented or reversed by NADPH-generating systems. Nevertheless, reticulocyte lysates contain a potent PKC activity and we deemed it desirable to isolate this enzyme to answer unequivocally the question whether it does or does not activate eIF-2 alpha kinase. We have purified reticulocyte PKC to near homogeneity with Mr 95,500 as estimated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The enzyme absolutely depended upon both Ca2+ and phosphatidylserine for activity on histone H1 or the beta-subunit of initiation factor eIF-2 and underwent autophosphorylation in a Ca2+- and PL-dependent manner. Mild treatment with trypsin yielded an Mr 82,000 polypeptide that still required Ca2+ and PL for activity. This Mr agrees with that reported for other PKCs, suggesting that these enzymes may undergo limited degradation during isolation. Further proteolytic treatment converted the reticulocyte enzyme into a Ca2+- and PL-dependent form, as is known for PKCs from other sources. The highly purified PKC had no effect on translation in hemin-supplemented reticulocyte lysates.

The complete primary structure of protein kinase C--the major phorbol ester receptor.

Protein kinase C, the major phorbol ester receptor, was purified from bovine brain and through the use of oligonucleotide probes based on partial amino acid sequence, complementary DNA clones were derived from bovine brain complementary DNA libraries. Thus, the complete amino acid sequence of bovine protein kinase C was determined, revealing a domain structure. At the amino terminal is a cysteine-rich domain with an internal duplication; a putative calcium-binding domain follows, and there is at the carboxyl terminal a domain that shows substantial homology, but not identity, to sequences of other protein kinase.