Search

During the induction of long-term potentiation (LTP) in hippocampal slices adenosine triphosphate (ATP) is secreted into the synaptic cleft, and a 48 kDa/50 kDa protein duplex becomes phosphorylated by extracellular ATP. All the criteria required as evidence that these two proteins serve as principal substrates of ecto-protein kinase activity on the surface of hippocampal pyramidal neurons have been fulfilled. This phosphorylation activity was detected on the surface of pyramidal neurons assayed after synaptogenesis, but not in immature neurons nor in glial cells. Addition to the extracellular medium of a monoclonal antibody termed mAb 1.9, directed to the catalytic domain of protein kinase C (PKC), inhibited selectively this surface protein phosphorylation activity and blocked the stabilization of LTP induced by high frequency stimulation (HFS) in hippocampal slices. This antibody did not interfere with routine synaptic transmission nor prevent the initial enhancement of synaptic responses observed during the 1-5 min period immediately after the application of HFS (the induction phase of LTP). However, the initial increase in the slope of excitatory postsynaptic potentials, as well as the elevated amplitude of the population spike induced by HFS, both declined gradually and returned to prestimulus values within 30-40 min after HFS was applied in the presence of mAb 1.9. A control antibody that binds to PKC but does not inhibit its activity had no effect on LTP. The selective inhibitory effects observed with mAb 1.9 provide the first direct evidence of a causal role for ecto-PK in the maintenance of stable LTP, an event implicated in the process of learning and the formation of memory in the brain.

The effects of (-)-isoproterenol on adenylate cyclase activity were studied in rat cerebral cortical membranes prepared and assayed in the presence of calcium ions. In assays carried out in the presence of high Mg2+ concentrations (5-10 mM) and of Ca2+ in the micromolar range, addition of 1-100 micro M (-)-isoproterenol caused over 50% inhibition of adenylate cyclase activity. Since these conditions are optimal for supporting endogenous phosphorylative activity in synaptic membranes, we tested whether the observed effects are mediated by changes in the phosphorylation of specific proteins in these membranes. This was done by preincubation of lysed synaptosomes under phosphorylating conditions in the presence and absence of isoproterenol followed by extensive washes and analysis of cyclic AMP formation in resuspended membranes. Addition of (-)-isoproterenol to the preincubation resulted in a 30% decrease of adenylate cyclase activity in the reincubation. Inclusion of [gamma-32P]ATP in the preincubation and examination of the phosphorylation state of specific proteins in membranes entering the reincubation revealed that (-)-isoproterenol inhibited the phosphorylation of a specific protein band with apparent molecular weight of 47,000 (designated band F). These results support the hypothesis that alterations in membrane protein phosphorylation induced by neurotransmitters play a role in the regulation of adenylate cyclase activity.

ATP is secreted in association with neurotransmitters at certain synapses and neuromuscular junctions. Extracellular ATP is known to exert potent effects on the activity of cells in the nervous system, where it can act as a neurotransmitter or as a modulator regulating the activity of other neurohormones. We have suggested that such modulation may involve the activity of extracellular protein phosphorylation systems. It is well known that intracellular protein kinases are important in the regulation of various neuronal functions, but protein kinases which use extracellular ATP to phosphorylate proteins localized at the external surface of the plasma membrane (ecto-protein kinases) have not been demonstrated in neuronal cells. Here we present direct evidence for the existence of an ecto-protein kinase and demonstrate endogenous substrates for its activity at the surface of intact neural cells. The phosphorylation of one of these surface proteins is selectively stimulated during cell depolarization. In addition, neuronal cell adhesion molecules (N-CAMs) appear to be among the substrates of ecto-protein kinase activity. These results suggest a role for surface protein phosphorylation in regulating specific functions of developing and mature neurones.

The effects of a training experience that involves foot shock on the endogenous phosphorylation of membrane-bound proteins from brain were studied. Crude membrane fractions were prepared from the cerebral cortex and neostriatum of animals that had been sacrificed by quick freezing. In vitro incubation of the membranes with gamma-32P-ATP, follwed by SDS-gel electrophoresis of the phosphorylated substrates, revealed that the phosphorylation of two protein components (designated F and H-I) increased in preparations from animals that were subjected to a training experience 24 hr prior to sacrifice. These effects were greater in preparations from the neostriatum than from the cerebral cortex, and were observed in experiments using both rats and mice. Although all trained animals showed a high phosphorylation of bands F and H-I, control animals showed a greater variability in the phosphorylation of these bands. The results indicate that the phosphorylation of specific proteins may play a mediatory role in the processing of experiential information.

Endogenous phosphorylation of 4 membrane-bound protein components of cerebral cortex was studied in vitro. Cyclic AMP was shown to stimulate phosphorylation to a major extent in protein components designated D, E and G, but to a minor extent in protein component F. The time course of phosphorylation of each of these components is affected differently by (a) the presence of cyclic nucleotides or zinc ions and (b) the concentration of ATP and magnesium ions in the reaction mixture. Although a different membrane preparation was used here than by Ueda et al., molecular weight determination and response to divalent cations, magnesium and zinc, suggest that components D and E may be the same as their proteins I and II. The newly described phosphoproteins, F and G, are respectively the least and most stimulated by the addition of cyclic AMP. Under certain conditions phosphate content in protein E can be increased by cyclic AMP and decreased by cyclic GMP.23Author