Direct phosphorylation of brain tyrosine hydroxylase by cyclic AMP-dependent protein kinase: mechanism of enzyme activation.

Tyrosine hydroxylase [tyrosine monooxygenase, L-tyrosine, tetrahydropteridine:oxygen oxidoreductase (3-hydroxylating), EC 1.14.16.2] was highly purified from rat caudate nuclei. When the pure hydroxylase was phosphorylated by incubation with cyclic AMP-dependent protein kinase and [32P]ATP, 32P and tyrosine hydroxylase activity were detected after polyacrylamide gel electrophoresis in a single protein band. After sodium dodecyl sulfate gel electrophoresis, 32P was detected only in a probably active subunit of tyrosine hydroxylase of molecular weight 62,000. Phosphorylation of the hydroxylase increased its activity by 2-fold, and was associated with an increase in Vm without any change in Km for either substrate or cofactor. We propose that the pool of native tyrosine hydroxylase is composed of a mixture of enzyme molecules in both active and probably inactive forms, that the active form is phosphorylated, and that phosphorylation produces an active form of the enzyme at the expense of an inactive one.     

Evidence for the involvement of a cyclic AMP-independent protein kinase in the activation of soluble tyrosine hydroxylase from rat striatum.

Activation of rat striatal tyrosine hydroxylase [TyrOHase; tyrosine monooxygenase; L-tyrosine, tetrahydropteridine:oxygen oxidoreductase (3-hydroxylating), EC 1.14.16.2] by ATP/Mg2+ and endogenous protein kinase can be produced without the addition of cAMP. This activation is not due to endogenous free catalytic subunit derived from cAMP-dependent protein kinase. In the presence of amounts of protein kinase inhibitor sufficient for complete inhibition of striatal cAMP-dependent protein kinase and the cAMP-mediated activation of TyrOHase, addition of ATP/Mg2+ results in an enhancement of TyrOHase activity. Enzyme activation does not occur when the nonhydrolyzable form of ATP, adenylyl imidodiphosphate, is substituted for ATP. When TyrOHase is assayed in the presence of ATP/Mg2+ and different concentrations of either tyrosine or 6-methyltetrahydropterin co-factor, a 2-fold increase in enzyme Vmax is demonstrable, with no change in the Km for either substrate or cofactor. In contrast, in the presence of cAMP and ATP/Mg2+, both an increase in Vmax and an enhanced affinity for pterin cofactor are demonstrable. In the latter circumstance, the 2-fold increase in Vmax can be attributed entirely to the action of cAMP-independent protein kinase. The addition of either EGTA or CaCl2 does not modify the effect seen in the presence of ATP, suggesting that the effect of ATP/Mg2+ is not mediated by a Ca2+-dependent protein kinase. These data support the existence of a cAMP-independent striatal protein kinase that can catalyze the activation of TyrOHase.     

Calcium/phospholipid-dependent protein kinase (protein kinase C) phosphorylates and activates tyrosine hydroxylase.

Protein kinase C, purified to homogeneity, was found to phosphorylate and activate tyrosine hydroxylase that had been partially purified from pheochromocytoma PC 12 cells. These actions of protein kinase C required the presence of calcium and phospholipid. This phosphorylation of tyrosine hydroxylase reduced the Km for the cofactor 6-methyltetrahydropterine from 0.45 mM to 0.11 mM, increased the Ki for dopamine from 4.2 microM to 47.5 microM, and produced no change in the Km for tyrosine. Little or no change in apparent Vmax was observed. These kinetic changes are similar to those seen upon activation of tyrosine hydroxylase by cAMP-dependent protein kinase. Two-dimensional phosphopeptide maps of tyrosine hydroxylase were identical whether the phosphorylation was catalyzed by protein kinase C or by the catalytic subunit of cAMP-dependent protein kinase. Both protein kinases phosphorylated serine residues. The results suggest that protein kinase C and cAMP-dependent protein kinase phosphorylate the same site(s) on tyrosine hydroxylase and activate tyrosine hydroxylase by the same mechanism.     

Activation of hormone-sensitive lipase and phosphorylase kinase by purified cyclic GMP-dependent protein kinase

Cyclic GMP-dependent protein kinase, purified to homogeneity from bovine lung, was shown to activate hormone-sensitive lipase partially purified from chicken adipose tissue. The degree of activation was the same as that effected by cyclic AMP-dependent protein kinase although higher concentrations of the cyclic GMP-dependent enzyme were required (relative activities expressed in terms of histone H2b phosphorylation units). Activation by cyclic AMP-dependent protein kinase was completely blocked by the heat-stable protein kinase inhibitor protein from skeletal muscle but activation by the cyclic GMP enzyme was not inhibited. Lipase fully activated by cyclic AMP-dependent protein kinase showed no further change in activity when treated with cyclic GMP-dependent protein kinase. Lipase activated by cyclic GMP-dependent protein kinase was reversibly deactivated by purified phosphorylase phosphatase (from bovine heart); full activity was restored by reincubation with cyclic GMP and cyclic GMP-dependent protein kinase. Cholesterol esterase activity in the chicken adipose tissue fraction, previously shown to be activated along with the triglyceride lipase by cyclic AMP-dependent protein kinase, was also activated by cyclic GMP-dependent protein kinase. Crude preparations of hormone-sensitive triglyceride lipase from human or rat adipose tissue and cholesterol esterase from rat adrenal were also activated by cyclic GMP-dependent protein kinase. Purified phosphorylase kinase (rabbit skeletal muscle) was also shown to be activated by cyclic GMP-dependent protein kinase. The present results, together with those of other workers on histone phosphorylation, suggest that the substrate specificities of cyclic GMP-dependent and cyclic AMP-dependent protein kinase may be similar. This is discussed in the light of a model recently proposed with regard to the relationship between the subunit structures of the two kinases. The physiologic significance of the findings remains to be established.     

Regulation of recombinant rat tyrosine hydroxylase by dopamine.

Recombinant rat PC12 tyrosine hydroxylase, also called tyrosine 3-monooxygenase [L-tyrosine, tetrahydropteridine:oxygen oxidoreductase (3-hydroxylating), EC 1.14.16.2], purified from Escherichia coli is in an activated form with a low Km for the tetrahydrobiopterin cofactor and a pH optimum of 6.5. Pretreatment with low levels of the derived product, dopamine, inhibits catalytic activity, increases the Km for the cofactor, and shifts the pH curve towards a more acidic pH optimum. Labeled dopamine binds to tyrosine hydroxylase with high affinity (Kd = 1 microM) but low stoichiometry (r = 0.08 mol/mol of enzyme subunit). The binding of dopamine results in the appearance of a blue-green chromophore with lambda max at approximately 660 nm, which is consistent with the formation of a catecholamine-iron complex. In the absence of dopamine, the recombinant enzyme cannot be further activated by phosphorylation with cAMP-dependent protein kinase, although as much as 1 mol of phosphate is incorporated per mol of subunit. In contrast, the enzyme pretreated with dopamine is activated by phosphorylation in the same fashion and to the same extent as the native hydroxylase. The results suggest that the high-affinity binding of catecholamine products is a pivotal post-translational modification that determines the state of enzyme activation and the response to phosphorylation.     

Phosphorylation of McArdle phosphorylase induces activity.

In McArdle disease, myophosphorylase deficiency, enzyme activity is absent but the presence of an altered enzyme protein can frequently be demonstrated. We have found that phosphorylation of this protein in vitro can result in catalytic activity. We studied muscle of four patients; all lacked myophosphorylase activity, but myophosphorylase protein was demonstrated by immunodiffusion or gel electrophoresis. Incubation of muscle homogenate supernatants with cyclic AMP-dependent protein kinase and ATP resulted in phosphorylase activity. The activated enzyme comigrated with normal human myophosphorylase in gel electrophoresis. Incubation with [gamma-32P]ATP resulted in incorporatin of 32P into the band possessing phosphorylase activity. Activation of phosphorylase by cyclic AMP-dependent protein kinase was inhibited by antibodies to normal human myophosphorylase or by inhibitory protein to cyclic AMP-dependent protein kinase. Incubation of muscle homogenates with phosphorylase b kinase and ATP also resulted in phosphorylase activity. After the action of cyclic AMP-dependent protein kinase, the resulting activity was similar to that of phosphorylase b. However, incubation with phosphorylase kinase resulted in activity similar to that of phosphorylase a. For several reasons, it is not likely that McArdle disease is due to lack of normal phosphorylation, but restoration of activity to the mutant protein by phosphorylation may provide a clue to understanding the mechanism of this genetic defect.     

Activation of purified calcium channels by stoichiometric protein phosphorylation.

Purified dihydropyridine-sensitive calcium channels from rabbit skeletal muscle were reconstituted into phosphatidylcholine vesicles to evaluate the effect of phosphorylation by cyclic AMP-dependent protein kinase (PK-A) on their function. Both the rate and extent of 45Ca2+ uptake into vesicles containing reconstituted calcium channels were increased severalfold after incubation with ATP and PK-A. The degree of stimulation of 45Ca2+ uptake was linearly proportional to the extent of phosphorylation of the alpha 1 and beta subunits of the calcium channel up to a stoichiometry of approximately 1 mol of phosphate incorporated into each subunit. The calcium channels activated by phosphorylation were determined to be incorporated into the reconstituted vesicles in the inside-out orientation and were completely inhibited by low concentrations of dihydropyridines, phenylalkylamines, Cd2+, Ni2+, and Mg2+. The results demonstrate a direct relationship between PK-A-catalyzed phosphorylation of the alpha 1 and beta subunits of the purified calcium channel and activation of the ion conductance activity of the dihydropyridine-sensitive calcium channels.     

Regulation of 6-phosphofructo-2-kinase activity by cyclic AMP-dependent phosphorylation.

Addition of glucagon to isolated rat hepatocytes resulted in inhibition of 6-phosphofructo-2-kinase (ATP:D-fructose-6-phosphate-2-phosphotransferase) activity in extracts of the cells and in a decrease in the intracellular level of fructose 2,6-bisphosphate. The effect on 6-phosphofructo-2-kinase was characterized by a decrease in the affinity of the enzyme for fructose 6-phosphate. To investigate the mechanism of action of glucagon, 6-phosphofructo-2-kinase from rat liver was partially purified by polyethylene glycol precipitation, DEAE-cellulose chromatography, (NH4)2SO4 fractionation, Sephacryl S-200 gel filtration, DEAE-Sephadex chromatography, and Sephadex G-100 gel filtration. Incubation of the purified enzyme with the catalytic subunit of the cyclic AMP-dependent protein kinase from rat liver and [gamma-32P]ATP resulted in 32P incorporation into a protein with a subunit Mr of 49,000 as determined by NaDodSO4 disc gel electrophoresis. Associated with this phosphorylation was an inhibition of 6-phosphofructo-2-kinase activity that was also characterized by a decrease in the affinity of the enzyme for fructose-6-phosphate. Both the phosphorylation and the inhibition of the purified 6-phosphofructo-2-kinase were blocked by addition of the heat-stable protein kinase inhibitor. It is concluded that the glucagon-induced decrease in fructose 2,6-bisphosphate levels observed in isolated hepatocytes is due, at least in part, to cyclic AMP-dependent phosphorylation and inhibition of 6-phosphofructo-2-kinase.     

Catalysis of serine and tyrosine autophosphorylation by the human insulin receptor.

The protein kinase activity of human insulin receptors purified from Sf9 insect cells after infection with a recombinant baculovirus was evaluated. The following experimental observations led to the unexpected conclusion that this receptor protein catalyzes both serine and tyrosine autophosphorylation at significant stoichiometries. (i) Phosphorylation of lectin-purified insulin receptors with [gamma-32P]ATP resulted in rapid receptor tyrosine phosphorylation (7 mol of P per high-affinity binding site) and the delayed onset of insulin-stimulated receptor serine phosphorylation (about 7% of total phosphorylation). The tyrosine kinase inhibitor (hydroxy-2-naphthalenylmethyl)phosphonic acid (HNMPA), which has no effect on protein kinase C or cyclic AMP-dependent protein kinase activities, inhibited both the receptor serine and tyrosine phosphorylation. (ii) Phosphorylation of a synthetic peptide substrate composed of insulin receptor residues 1290-1319 on serines-1305/1306 by partially purified insulin receptors was also inhibited by HNMPA. (iii) Insulin receptors sequentially affinity-purified on immobilized wheat germ agglutinin and immobilized insulin showed no apparent contaminant proteins on silver-stained SDS/polyacrylamide gels yet catalyzed autophosphorylation on receptor serine and tyrosine residues when incubated with [gamma-32P]ATP. These results suggest that the catalytic site of the insulin receptor tyrosine kinase also recognizes receptor serine residues as substrates for the phosphotransfer reaction. Furthermore, insulin-stimulated receptor serine phosphorylation in intact cells may occur in part by an autophosphorylation mechanism subsequent to tyrosine phosphorylation of the insulin receptor.     

Mechanism of action of glucagon on hepatocyte phosphofructokinase activity.

Addition of glucagon to isolated hepatocytes reduced the activity of 6-phosphofructokinase (ATP:D-fructose-6-phosphate 1-phosphotransferase, EC 2.7.1.11) and pyruvate kinase (ATP:pyruvate 2-O-phosphotransferase, EC 2.7.1.40). Phosphorylation contributed to the inhibition of pyruvate kinase, but several lines of evidence indicated that this reaction was not responsible for the inhibition of phosphofructokinase. First, the increase in phosphorylation in intact cells induced by increasing the concentration of glucagon did not correlate well with the decrease in enzyme activity. Second, phosphorylation of phosphofructokinase induced by addition of cyclic AMP and Mg2+-ATP or by addition of Mg2+-ATP and the catalytic subunit of the cyclic AMP-dependent protein kinase to hepatocyte extracts had no effect on enzyme activity. Third, ammonium sulfate precipitation of the enzyme from extracts of cells incubated with glucagon abolished the hormone effect. The effect could be restored, however, by the addition of a phosphofructokinase-free extract from glucagon-treated cells to the ammonium sulfate-treated enzyme from either untreated or glucagon-treated cells. These results suggest that the inhibition of phosphofructokinase by glucagon is due to changes in the level of an allosteric effector(s).     

Inhibition of mammalian protein kinase and phosphodiesterase activities by a cyclic AMP-like compound isolated from higher plants.

A cyclic AMP-like substance has been isolated from higher plant tissues which can be quantitated with the use of a radioimmunoassay similar to that described by A. L. Steiner, D. M. Kipnis, R. Utiger, and C. Parker [(1969) Proc. Natl. Acad. Sci. USA 64, 367-373]. This compound has been extensively purified and is chromatographically distinct from authentic cyclic AMP. This cyclic AMP-like compound inhibited beef heart 3':5'-cyclic-nucleotide phosphodietsterase (3':5'-cyclic-nucleotide 5'-nucleotidohydrolase, EC 3.1.4.17), with half-maximal inhibition occurring at a concentration of 7.6 X 10(-10) M cyclic AMP equivalents. The compound also inhibited cyclic AMP-dependent protein kinase (ATP:protein phosphotransferase; EC 2.7.1.37) from bovine heart, with half-maximal inhibition of mixed histone phosphorylation occurring at 8.0 X 10(-11) M cyclic AMP equivalents. Equipotent inhibition of phosphorylation and associated trace ATPase activity were observed with the purified catalytic subunit of cyclic AMP-dependent protein kinase from calf thymus with a synthetic heptapeptide as substrate. Moreover, steady-state kinetic analysis of this inhibition in the latter system showed it to be nonlinear and noncompetitive versus MgATP.     

Substrate specificity of the cyclic AMP-dependent protein kinase.

The protein substrate specificity of the catalytic subunit of rabbit skeletal muscle cyclic AMP-dependent protein kinase (EC 2.7.1.37; ATP:protein phosphotransferase) has been studied using genetic variants of beta casein. It was found that beta casein-B was phosphorylated at a much greater rate than beta caseins A1, A2, A3, or C. The enhanced phosphorylation of beta casein-B, as compared with the most common variant A2, was attributed to an arginine substitution for a serine at position 122, which caused a nearby residue, serine 124, to become a phosphorylation site for the protein kinase. These results further support the concept that the local primary structure is important in specificity and that arginine may be a specific determinant common to all the local phosphorylation site sequences recognized by the cyclic AMP-dependent protein kinase.     

In vivo and in vitro phosphorylation of rat liver fructose-1,6-bisphosphatase.

Incorporation of 32P from [gamma-32P]ATP into a homogenous preparation of rat hepatic fructose-1,6-bisphosphatase (D-fructose-1,6-bisphosphatase 1-phosphohydrolase, EC 3.1.3.11) was catalyzed by a homogeneous preparation of the catalytic subunit of cyclic AMP-dependent protein kinase from bovine liver. Approximately 4 mol of phosphate were incorporated per mol of the tetrameric enzyme. This phosphorylation was associated with an increase in enzyme activity. In addition, in vivo phosphorylation of the enzyme was observed after injection of radioactive inorganic phosphate into rats and subsequent isolation of the enzyme by conventional purification methods and by immunoprecipitation. All of the labeled phosphate incorporation into the enzyme, both in vitro and in vivo, was precipitated by antibody specific for the enzyme. Furthermore, the 32Pi counts were coincident with the enzyme subunit band when the immunoprecipitates were examined by sodium dodecyl sulfate/disc gel electrophoresis. Acid hydrolysis of the immunoprecipitated enzyme that was phosphorylated in vitro revealed that only seryl residues were labeled. On the basis of the concentration of protein kinase (0.2-1.0 muM) necessary to phosphorylate physiological amounts of fructose-1,6-bisphosphatase (1.0-4.0 muM), it is suggested that cyclic AMP-dependent protein kinase may catalyze the phosphorylation of fructose-1,6-bisphosphatase in vivo.     

Regulation of fructose-6-phosphate 2-kinase by phosphorylation and dephosphorylation: possible mechanism for coordinated control of glycolysis and glycogenolysis.

The kinetic properties and the control mechanism of fructose-6-phosphate 2-kinase (ATP: D-fructose-6-phosphate 2-phosphotransferase) were investigated. The molecular weight of the enzyme is approximately 100,000 as determined by gel filtration. The plot of initial velocity versus ATP concentration is hyperbolic with a Km of 1.2 mM. However, the plot of enzyme activity as a function of fructose-6-phosphate is sigmoidal. The apparent K0.5 for fructose-6-phosphate is 20 microM. Fructose-6-phosphate 2-kinase is inactivated by the catalytic subunit of cyclic AMP-dependent protein kinase, and the inactivation is closely correlated with phosphorylation. The enzyme is also inactivated by phosphorylase kinase in the presence of Ca2+ and calmodulin. The phosphorylated fructose-6-phosphate 2-kinase, which is inactive, is activated by phosphorylase phosphatase and alkaline phosphatase. The possible physiological significance of these observations in the coordinated control of glycogen metabolism and glycolysis is discussed.     

Cyclic AMP-dependent phosphorylation of high molecular mass proteins in pig thyroid cells

Four high molecular mass (H Mr) proteins were found to be phosphorylated in a cyclic AMP-dependent manner in both partially purified pig thyroid membrane fractions and in pig thyroid cells in culture. These phosphoproteins did not correspond to major cellular proteins; they were found in both soluble and particulate subfractions of homogenates from cultured thyroid cells. The molecular mass of the 4 proteins named HMr1 to HMr4 determined by polyacrylamide gel electrophoresis in the in the presence of sodium dodecylsulfate was about 310 000 for HMr1, 250 000 for HMr2, 240 000 for Hmr3 and 220 000 for HMr4. HMr1 comigrated with brain MAP1, whereas HMr3 and HMr4 had the same mobility as alpha-and beta-spectrins, respectively. The 4 high molecular mass phosphoproteins are substrates of cyclic AMP-dependent protein kinase(s) since (a) their phosphorylation was increased by cyclic AMP and not by cyclic GMP or calcium alone or calcium in the presence of calmodulin or phospholipid; (b) the effect of cyclic AMP was prevented by the thermostable inhibitor of cyclic AMP-dependent protein kinases; (c) the purified catalytic subunit of cyclic AMP-dependent protein kinases markedly phosphorylated the 4 HMr proteins. The 32P-labeling of HMr proteins using either endogenous cyclic AMP-dependent protein kinase or the purified catalytic subunit was always lower in cells cultured in the presence of TSH (reassociated in follicle-like structures) than in freshly dispersed cells or cells cultured in basal conditions (cells in monolayer). These results suggest that the 4 high molecular mass thyroid phosphoproteins represent structural components, the phosphorylation of which could vary with the cellular organization.     

Ca2+ -activated K+ conductance in internally perfused snail neurons is enhanced by protein phosphorylation.

Depolarizing voltage steps induce inward and outward currents in voltage-clamped, internally perfused neurons from the snail Helix roseneri. Addition of the catalytic subunit of cyclic AMP-dependent protein kinase (ATP:protein phosphotransferase, EC 2.7.1.37) to the internal perfusing medium results in an increase in the net outward current, with no apparent effect on the inward current. Catalytic subunit inactivated by 5,5'-dithiobis(2-nitrobenzoic acid) is without effect, indicating that the increase in net outward current results from protein phosphorylation rather than an unspecific effect of protein perfusion. Decreasing the external Ca2+ concentration from 10 to 1 mM eliminates the effect of catalytic subunit, suggesting that Ca2+ plays an important role in this response. This suggestion is supported by the fact that the stimulation by catalytic subunit can be mimicked by increasing the Ca2+ concentration in the internal perfusion medium and can be prevented by intracellular perfusion with 10 mM EGTA. The results are consistent with the hypothesis that cyclic AMP-dependent protein phosphorylation regulates the Ca2+-activated K+ conductance in these cells.     

Microinjection of catalytic subunit of cyclic AMP-dependent protein kinase enhances calcium action potentials of bag cell neurons in cell culture

We have found that the calcium action potentials of bag cell neurons from the abdominal ganglion of Aplysia may be enhanced by intracellular microinjection of the catalytic subunit of cyclic AMP-dependent protein kinase (ATP:protein phosphotransferase, EC 2.7.1.37). The catalytic subunit was purified from bovine heart and shown to be effective in stimulating the phosphorylation of bag cell proteins in homogenates at concentrations of 10-50 nM. Intracellular injection into isolated bag cell neurons maintained in primary culture was through pressure applied to microelectrodes filled at the tip with catalytic subunit (5-22 muM). In 11 of 16 injected cells, both the slope of the rising phase and the height of the action potentials evoked by a constant depolarizing current were markedly enhanced relative to the pre-injection control (mean increases, 73% and 35%, respectively). This effect could occur with no change in resting potential or in the latency of the action potential from the onset of the depolarizing pulse. The effect was observed with enzyme dissolved in three different salt solutions (Na phosphate, K phosphate, or KCl). In two experiments, tetrodotoxin (50 muM) added to the extracellular medium had no effect on the enhanced action potentials. Subsequent addition of the calcium antagonist Co(2+), however, diminished or abolished the spikes. In more than half of the experiments, the injection of catalytic subunit was accompanied by an increase in the input resistance of the cells as measured by applying small hyperpolarizing current pulses. In three experiments, subthreshold oscillations in membrane potential resulted from the injections. Control injections (24 cells), carried out either with carrier medium alone or with heat-inactivated enzyme preparations, did not produce spike enhancement, increased input resistance, or oscillations. Our data suggest that the stimulation of intracellular protein phosphorylation by the catalytic subunit of cyclic AMP-dependent protein kinase enhances the excitability of bag cell neurons by modifying calcium and potassium channels or currents.     

Synthetic hexapeptide substrates and inhibitors of 3'':5''-cyclic AMP-dependent protein kinase.

The substrate specificity of the catalytic subunit of rabbit skeletal muscle 3': 5'-cyclic AMP-dependent protein kinase (EC 2.7.1.37; ATP: protein phosphotransferase) has been studied using the synthetic peptide Arg-Gly-Tyr-Ser-Leu-Gly corresponding to the sequence around serine 24, a phosphorylation site in reduced, carboxymethylated, maleylated (RCMM) chicken egg white lysozyme. This peptide served as a substrate for the enzyme and exhibited a 6-fold higher Vmax and a 100-fold higher Km than RCMM-lysozyme. Replacement of the arginine with glycine, histidine, or lysine resulted in a dramatic reduction in the Vmax. These results support the concept that arginine is an important residue in determining the substrate specificity of the protein kinase, predominantly influencing the Vmax of the phosphorylation reaction. Two synthetic peptides in which serine was replaced by an alanine acted as competitive inhibitors of phosphorylation of the synthetic peptide substrate and RCMM-lysozyme.     

Compartmentalization of cyclic AMP-dependent protein kinases in human erythrocytes.

The human erythrocyte contains two types of cyclic AMP-dependent protein kinase. The membrane-associated protein kinase holoenzyme can be released intact by hypotonic treatment at alkaline pH. Chromatography on DEAE-cellulose and molecular weight determinations demonstrate that this enzyme is a type I cyclic AMP-dependent protein kinase, while the predominant cyclic AMP-dependent protein kinase found in the cytoplasm is type II. The catalytic subunits of the two types of kinase found in the erythrocyte are identical, but the regulatory subunits, which distinguish the two types of kinase, determine their localization within the cell. It is proposed that the regulatory subunit of the type I enzyme, in addition to binding the catalytic subunit, interacts specifically with one or more membrane proteins and that this interaction may serve to position the kinase in preferential proximity to protein substrates.     

Intracellular injection of t he catalytic subunit of cyclic AMP-dependent protein kinase simulates facilitation of transmitter release underlying behavioral sensitization in Aplysia.

It has been difficult to establish whether cyclic AMP-mediated protein phosphorylation in nerve cells plays a specific role in synaptic transmission. This difficulty can be overcome in higher invertebrates because their large neurons allow the injection of protein molecules into the cell. We have used intracellular injection to study whether protein phosphorylation is involved in the mechanism of sensitization, a simple form of learning. Sensitization of the gill-withdrawal reflex in Aplysia involves enhancement of transmitter release by presynaptic facilitation at a particular set of synaptic connections between identified sensory neurons and their follower cells. We have found that injection of the catalytic subunit of cyclic AMP-dependent protein kinase (ATP:protein phosphotransferase, EC 2.7.1.37) purified from bovine heart mimics the action of the natural transmitter and of serotonin, the putative transmitter, by simulating the physiological changes that accompany presynaptic facilitation. Intracellular injection of the kinase into a sensory cell (i) broadens the action potential in the presence of tetraethylammonium, indicating an increase in Ca2+ current, (ii) decreases the input conductance of the cell, presumably as a result of a decrease in the K+ current, and (iii) increases the amount of transmitter released by terminals of the sensory cell onto follower neurons.