Protein phosphorylation in isolated plasma membranes and postsynaptic junctional structures from brain synapses

Abstract

The intrinsic protein kinase and phosphoprotein phosphatase activities of rat brain synaptic plasma membranes were investigated by following the incorporation of 32P from [γ- 32P]adenosine 5′-triphosphate into endogenous membrane proteins. The relationship of both these enzymes and of the phosphorylated proteins to brain synaptic junctions was determined by comparing the distribution of both the enzyme activities and the 32P-labelled polypeptides in subcellular fractions containing either synaptic membranes, isolated postsynaptic densities or the postsynaptic junctional lattice, a structure composed of interlacing protein fibres which underlies the postsynaptic density. After labelling synaptic membranes and then isolating the two junctional subfractions the specific activity of proteinbound 32P was two-fold higher in the postsynaptic density than in synaptic membranes and more than three-fold higher in the postsynaptic lattice. The enzyme activities were elevated two-fold in the postsynaptic density compared to synaptic membranes but in the postsynaptic lattice they were less than half the synaptic membrane level. The low levels of both enzymes in the postsynaptic lattice could be the result of inactivation by exposure to the sodium deoxycholate used to prepare the lattice fraction rather than indicating that the enzyme molecules are absent. Each subcellular fraction was phosphorylated and individual labelled polypeptides were identified by electrophoresis and autoradiography. All the 32P-labelled bands in synaptic membranes were also present in the postsynaptic density and postsynaptic lattice with the exception of material of molecular weight 47,000 which, although shown to be present by protein staining, was not phosphorylated in either junctional fraction. A 32P-labelled band of molecular weight 26,000 which was found in both postsynaptic junctional fractions was below the limit of detection in synaptic membranes. The two lowest molecular weight polypeptides, of 21,000 and 19,000, were labelled with specific activities respectively nine-fold and six-fold higher in isolated postsynaptic densities than in the synaptic membranes from which they were derived. Most of the postsynaptic density polypeptides exhibited a higher incorporation of 32P when labelling was performed with the addition of 50μm cyclic adenosine monophosphate. However, the three smallest (molecular weights: 19,000, 21,000 and 26,000) showed an inverse response, the incorporation of 32p into them being markedly depressed when cyclic adenosine monophosphate was present. Thus most of the [ 32P]phosphate incorporated into synaptic membranes via the intrinsic kinase is associated with the postsynaptic junctional lattice, which is probably a structural framework for the postsynaptic organelle ( MATUS & WALTERS, 1975 ). This suggests that rather than being exclusively involved in synaptic transmission the phosphorylation of postsynaptic junctional proteins may be part of a more general metabolic response to afferent stimulation.