Search

A highly purified adenosine 3′,5′-monophosphate-dependent protein kinase from bovine brain has been found to catalyze its own phosphorylation. The incorporated phosphate was shown to be associated with the cyclic AMP-binding subunit (R-protein) of the protein kinase. The catalytic subunit exhibited no detectable incorporation of phosphate into itself, but was required for the phosphorylation of R-protein. The molecular weight of R-protein was determined by polyacrylamide gel electrophoresis to be about 48,000 in the presence of sodium dodecyl sulfate. Cyclic AMP strikingly inhibited the rate of autophosphorylation observed in the presence of ZnCl2, CaCl2, NiCl2, or FeCl2, but had no significant effect in the presence of MgCl2 or CoCl2. The concentration of cyclic AMP required to give half-maximal inhibition of phosphorylation was 3 × 10−7m in the presence of either CaCl2 or ZnCl2. Guanosine 3′,5′-monophosphate was far less effective under the same experimental conditions than cyclic AMP. R-protein appears to be similar to a phosphoprotein recently discovered in synaptic membrane fractions from rat and bovine cerebral cortex.

DPNH-rubredoxin reductase, a flavoprotein which functions as an electron carrier in an enzyme system from Pseudomonas oleovorans capable of hydroxylating fatty acids and hydrocarbons, has been isolated in homogeneous form. The apparent molecular weight is 55,000 as judged by sedimentation and diffusion measurements; a similar result was obtained by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate and mercaptoethanol, thereby suggesting the presence of a single polypeptide chain. FAD was identified as the prosthetic group and was shown to be present in the amount of 1 mole per mole of enzyme. The oxidized protein has absorption maxima at 450, 378, and 272 nm, with an A272:A450 ratio of 6.6 and a molar extinction coefficient at 450 nm of 11.0 x 103. The spectrum is that of a flavoprotein apparently free of other prosthetic groups. Metals could not be detected by atomic absorption spectrophotometry. The amino acid composition shows a high content of hydrophobic residues, particularly leucine, isoleucine, and valine. The purified enzyme catalyzes the rubredoxin-dependent reduction of cytochrome c in the presence of DPNH and also exhibits diaphorase activity toward ferricyanide.

Fatty acid and alkane hydroxylation in Pseudomonas oleovorans requires three protein components: DPNH-rubredoxin reductase, rubredoxin, and the ω-hydroxylase. Homogeneous preparations of the reductase catalyze electron transfer not only to the rubredoxin of P. oleovorans but also to the lower molecular weight rubredoxins of the anaerobic bacteria, Peptostreptococcus elsdenii, Clostridium pasteurianum, and Desulfovibrio gigas. In contrast, the reductase is inactive with nonheme iron proteins such as spinach ferredoxin, putidaredoxin, and adrenodoxin, all of which contain labile sulfide. The reductase transfers electrons directly to ferricyanide and dichlorophenolindophenol, and indirectly to cytochrome c; the latter reaction is completely dependent upon the presence of rubredoxin. DPNH is highly superior to TPNH as an electron donor. Since two electron equivalents are accepted per mole of reductase, as shown by anaerobic titrations with DPNH or dithionite, and since 1 mole of FAD is present, it is concluded that no active oxidation-reduction group is present other than the flavin. The absorbance noted at long wave lengths when the enzyme is quantitatively reduced by DPNH or by EDTA and light in the presence of DPN, but not when it is reduced by dithionite or by EDTA and light in the absence of DPN, suggests the formation of a charge transfer complex between the reduced flavoprotein and DPN. A stable semiquinone was not formed during the titrations, as judged by the spectral changes observed. An interaction between rubredoxin and the reductase was demonstrated both by spectrophotometric and fluorimetric techniques. Measurements of the decrease in fluorescence of the reductase in the presence of rubredoxin indicated that these components form a complex in a 1:1 ratio; the dissociation constant of the complex was determined to be 2.1 x 10-7 m. The energy of activation of the enzyme-catalyzed reduction of rubredoxin by DPNH was found to be 5400 cal per mole.

Synaptic membrane preparations contain a protein with an apparent molecular weight of 49,000, designated Protein II, whose endogenous phosphorylation and dephosphorylation are regulated by cyclic AMP. Protein II and the enzymes which catalyze the cyclic AMP-dependent phosphorylation and dephosphorylation of Protein II have been obtained in a soluble form from these synaptic membrane preparations by treatment with either 0.25% Triton X-100 or 0.1 m NH4Cl. In the Triton X-100 solubilized enzyme system, the regulation of the endogenous phosphorylation of Protein II by cyclic AMP was similar to that observed in untreated synaptic membrane preparations. Thus, the effect of cyclic AMP on the phosphorylation of Protein II was stimulatory in the presence of Mg2+ and inhibitory in the presence of Zn2+. The stimulation by cyclic AMP of the phosphorylation of Protein II observed in the presence of Mg2+ could be converted to an inhibitory effect either by increased ionic strength, by Zn2+, or by a thiol reagent, dithionitrobenzoate. In the NH4Cl-solubilized enzyme system, cyclic AMP-dependent phosphorylation was also observed, but the mode of regulation by cyclic AMP of the phosphorylation of Protein II was partially altered: the effect of cyclic AMP was inhibitory in the presence of either Mg2+ or Zn2+. In both the Triton X-100-solubilized and the NH4Cl-solubilized enzyme systems, cyclic AMP facilitated the endogenous dephosphorylation of Protein II in the absence of any divalent cation, as it did in untreated membrane preparations. The kinetic data obtained for the endogenous phosphorylation and endogenous dephosphorylation of Protein II, in both the Triton extract and the NH4Cl extract, are compatible with the idea that Protein II, its cyclic AMP-dependent protein kinase, and its cyclic AMP-dependent protein phosphatase, may exist in, and be extractable from, synaptic membranes in the form of a complex.

The endogenous phosphorylation of protein in various membranous subcellular fractions from rat cerebrum has been studied in vitro. The endogenous phosphorylation of one minor protein component of synaptic membrane fractions showed a marked dependence upon the presence of added adenosine 3',5'-monophosphate (cyclic AMP). The apparent molecular weight of this phosphoprotein was approximately 100,000. Under the experimental conditions used, the selective stimulation of phosphorylation of a membrane protein by cyclic AMP was observed only in those subcellular fractions enriched in synaptic membranes. The results support other evidence suggesting that certain types of synaptic transmission may involve cyclic AMP-dependent phosphorylation of a specific protein constituent of synaptic membranes.

The phosphorylation of specific membrane proteins by endogenous protein kinase in synaptic membrane fractions from rat cerebrum has been studied using the technique of acrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. The endogenous phosphorylation of two specific membrane proteins was found to be regulated by adenosine 3':5'-monophosphate (cyclic AMP). The minimal molecular weights of these proteins were determined, by gel electrophoresis in the presence of sodium dodecyl sulfate, to be 86,000 (Protein I) and 48,000 (Protein II). The endogenous phosphorylation of Protein I and Protein II was stimulated by cyclic AMP in the presence of magnesium ions. Approximately 80% and 75% of the 32P incorporated into Protein I and Protein II, respectively, was found in phosphoserine. The concentration of cyclic AMP required for maximal stimulation of the phosphorylation of both Protein I and Protein II was approximately 5 x 10-6 m. Guanosine 3':5'-monophosphate had no significant effect, at concentrations up to 10-4 m, on the phosphorylation of either of these proteins. The endogenous phosphorylation occurred rapidly, the phosphorylation of both Protein I and Protein II reaching maximal levels within 5 sec. After reaching a maximal level of phosphorylation, Protein II underwent rapid dephosphorylation. The cyclic AMP-stimulated endogenous phosphorylation of a protein corresponding in position on acrylamide gel electrophoresis to Protein I was also observed in membrane fractions prepared from other neural tissues known to contain synapses, but membrane fractions prepared from several neural and non-neural tissues devoid of synapses, including lingual nerve, lung, liver, spleen, and kidney, all failed to reveal cyclic AMP-dependent phosphorylation of a protein band corresponding in position to Protein I. In contrast, cyclic AMP-dependent phosphorylation of a protein band corresponding in position to Protein II was observed in several neural and non-neural tissues. The results are compatible with the hypothesis that the mediation of certain types of synaptic transmission may involve regulation by cyclic AMP of the level of phosphorylation of specific protein constituents of the synaptic membranes.