John A.P. Rostas

A rapid method for isolation of synaptosomes on Percoll gradients

A new rapid method for fractionation of crude synaptosomes (postmitochondrial pellet, P2) on a discontinuous 4-step Percoll gradient is described. The homogeneity and integrity of the 5 major subcellular fractions were determined by analysis of the distribution of protein, lactate dehydrogenase, cytochrome oxidase, pyruvate dehydrogenase, synapsin I (a synaptic vesicle marker) and the myelin basic proteins. The biochemical results were substantiated by quantitative electron microscopy. Fractions 3, 4 and 5 were enriched in synaptosomes and contained 19.7, 40.6 and 19.5% of the intact, identifiable synaptosomes in P2, respectively. Fraction 1 was enriched in membranous material, fraction 2 in myelin and fraction 5 in extrasynaptosomal mitochondria. The synaptosomes in fractions 3, 4 and 5 differed in their size, and their content of mitochondria, synapsin I and neurotransmitters. These results suggest that partial separation of different pools of synaptosomes has been achieved. The synaptosomes in fractions 3, 4 and 5 are viable, as they take up calcium, phosphate and noradrenaline; they are metabolically normal as judged by their ability to perform protein phosphorylation and they respond normally to depolarization by increasing calcium uptake, protein phosphorylation and neurotransmitter release. The synaptosomes in fraction 4 are relatively homogeneous and appear to be free of contamination from lysed synaptosomes and synaptic plasma membranes. This constitutes a major advantage of the Percoll method over traditional procedures which involve centrifugation to equilibrium. We have therefore confirmed (J. Neurochem., 43 (1984) 1114-1123) the advantages of Percoll use over traditional procedures, while further reducing the time taken, and extended our analysis to show that the present procedure provides a fractionation of synaptosomes into different pools of viable synaptosomes.

A rapid Percoll gradient procedure for isolation of synaptosomes directly from an S1 fraction: homogeneity and morphology of subcellular fractions

A method for preparation of synaptosomes from rat cerebral cortex, on a discontinuous Percoll gradient, was previously developed for use with a P2 pellet (Brain Research, 372 (1986) 115-129). Here the Percoll method has been adapted for use with an S1-supernatant which eliminates a potentially damaging resuspension step and saves over 30 min, representing a third of the total preparation time. The homogeneity of the synaptosomes in each of the 5 subcellular fractions obtained with the S1-Percoll method was determined biochemically by analysis of the distribution of total protein, myelin basic protein, synapsin I and pyruvate dehydrogenase across the gradient. Electron microscopy was also used to determine the homogeneity of the synaptosomes, as well as to determine their morphological characteristics. Fraction 4 was the most enriched in synaptosomes and contained the lowest level of contamination by myelin, extrasynaptosomal mitochondria and plasma membranes. The yield of synaptosomes in fraction 4 with the S1-Percoll method was 1.4-fold greater than with the P2-Percoll method. While all other fractions contained some synaptosomes the major additional content in fractions 1-3 and 5 was, respectively, unidentified small membranes, myelin, synaptic plasma membranes and extrasynaptosomal mitochondria. Fraction 1 was enriched for very small synaptosomes (0.34 micron mean diameter) only 8% of which contained mitochondria, while fractions 2-4 progressively included larger synaptosomes containing more mitochondria. Fraction 5 synaptosomes were approximately the same size as those in fraction 4 (0.63 micron mean diameter), but 83% contained mitochondria, significantly more than in fraction 4. The synaptosomes in fraction 5 were found to be relatively resistant to hypotonic lysis, explaining a previously observed lack of phosphorylation of synapsin I in this fraction. The differences in homogeneity and morphological characteristics of the synaptosomes in fractions 1-5 suggest that the basis for their fractionation on Percoll gradients is different from that achieved with the more traditional procedures for isolating synaptosomes and that unique synaptosomal fractions are obtained with the S1-Percoll procedure.

Subcellular Localisation of 14‐3‐3 Isoforms in Rat Brain Using Specific Antibodies

Abstract: The 14‐3‐3 protein family, which is present at particularly high concentrations in mammalian brain, is known to be involved in various cellular functions, including protein kinase C regulation and exocytosis. Despite the fact that most of the 14‐3‐3 proteins are cytosolic, a small but significant proportion of 14‐3‐3 in brain is tightly and selectively associated with some membranes. Using a panel of isoform‐specific antisera we find that the ε, η, γ, β, and ζ isoforms are all present in purified synaptic membranes but absent from mitochondrial and myelin membranes. In addition, the η, ε, and γ isoforms but not the β and ζ isoforms are associated with isolated synaptic junctions. When different populations of synaptosomes were fractionated by a nonequilibrium Percoll gradient procedure, the ε and γ isoforms were present and the β and ζ isoforms were absent from the membranes of synaptosomes sedimenting in the more dense parts of the gradient. The finding that these proteins are associated with different populations of synaptic membranes suggests that they are selectively expressed in different classes of neurones and raises the possibility that some or all of them may influence neurotransmission by regulating exocytosis and/or phosphorylation.

Autonomous activity of CaMKII is only transiently increased following the induction of long-term potentiation in the rat hippocampus

A major role has been postulated for a maintained increase in the autonomous activity of CaMKII in the expression of long‐term potentiation (LTP). However, attempts to inhibit the expression of LTP with CaMKII inhibitors have yielded inconsistent results. Here we compare the changes in CaMKII autonomous activity and phosphorylation at Thr286 of αCaMKII in rat hippocampal slices using chemical or tetanic stimulation to produce either LTP or short‐term potentiation (STP). Tetanus‐induced LTP in area CA1 requires CaMKII activation and Thr286 phosphorylation of αCaMKII, but we did not observe an increase in autonomous activity. Next we induced LTP by 10 min exposure to 25 mm tetraethyl‐ammonium (TEA) or 5 min exposure to 41 mm potassium (K) after pretreatment with calyculin A. Exposure to K alone produced STP. These protocols allowed us to monitor temporal changes in autonomous activity during and after exposure to the potentiating chemical stimulus. In chemically induced LTP, autonomous activity was maximally increased within 30 s whereas this increase was significantly delayed in STP. However, in both LTP and STP the two‐fold increase in autonomous activity measured immediately after stimulation was short‐lived, returning to baseline within 2–5 min after re‐exposure to normal ACSF. In LTP, but not in STP, the phosphorylation of αCaMKII at Thr286 persisted for at least 60 min after stimulation. These results confirm that LTP is associated with a maintained increase in autophosphorylation at Thr286 but indicate that a persistent increase in the autonomous activity οf CaMKII is not required for the expression of LTP.

Septin 3 (G‐septin) is a developmentally regulated phosphoprotein enriched in presynaptic nerve terminals

The septins are GTPase enzymes with multiple roles in cytokinesis, cell polarity or exocytosis. The proteins from the mammalian septin genes are called Sept1–10. Most are expressed in multiple tissues, but the mRNA for Sept5 (CDCrel‐1) and Sept3 (G‐septin) appear to be primarily expressed in brain. Sept3 is phosphorylated by cGMP‐dependent protein kinase I (PKG‐I) and the cGMP/PKG pathway is involved in presynaptic plasticity. Therefore to determine whether Sept3 specifically associates with neurones and nerve terminals we investigated its distribution in rat brain and neuronal cultures. Sept3 protein was detected only in brain by immunoblot, but not in 12 other tissues examined. Levels were high in all adult brain regions, and reduced in those enriched in white matter. Expression was developmentally regulated, being absent in the early embryo, low in late embryonic rat brain and increasing after birth. Like dynamin I, Sept3 was specifically enriched in synaptosomes compared with whole brain, and was only found in a peripheral membrane extract and not in the soluble or membrane extracts. Sept3 was particularly abundant in mossy fibre nerve terminals in the hippocampus. In primary cultured hippocampal neurones Sept3 immunoreactivity was punctate in neurites and predominantly localized to presynaptic terminals, strongly colocalizing with synaptophysin and dynamin I. The specific nerve terminal localization was confirmed by immunogold electron microscopy. Together this shows that Sept3 is a neurone‐specific protein highly enriched in nerve terminals which supports a secretory role in synaptic vesicle recycling.

Protein and glycoprotein composition of synaptic junctions prepared from discrete synaptic regions and different species.

Synaptic junction (SJ) were isolated by subcellular fractionation from different areas of the steer brain and from the brains of different species (steer, rat, chicken and human) for the purpose of comparing their protein and glycoprotein composition. The synaptic junction fractions from different brain regions and species were of comparable morphological purity. Analysis of the polypeptide composition of isolated synaptic junction fractions via SDS polyacrylamide gel electrophoresis showed that the major polypeptides were represented in all junctional fractions independent of their source. Tubulin, actin, the major 52,000 Mr postsynaptic density protein and a group of proteins with a molecular weight of 200-250,000 Mr were all similarly represented. Most other components were also similar but quantitative differences were found for a few polypeptides. Interspecies differences were more prevalent than those between different brain areas of the same species. The protein compositions of different brain areas were similar even when an area consisting of primarily one neuronal type was compared to areas containing a mixture of neuronal types. However, two-dimensional gel electrophoresis revealed distinct (but usually minor) polypeptides in enriched quantities in one or more brain areas. Tryptic peptide maps were carried out on the major postsynaptic density protein of different species. These maps showed a high degree of conservation in this proteins primary structure among all species studied. The glycoproteins of isolated synaptic junctions which bind the plant lectin concanavalin A (Con A) were also examined. To identify individual Con A binding components, SJ fractions were solubilized and constituent glycoproteins were separated by SDS gel electrophoresis. Gels were then incubated in 125I-Con A. The glycoproteins which bound Con A in gels were few in number and were not the major Coomassie blue staining bands. The great majority of the Con A binding glycoproteins were similar between species and among brain areas of the same species.

Modulation of the phosphorylation and activity of calcium/calmodulin- dependent protein kinase II by zinc

Calcium/calmodulin‐dependent protein kinase II (CaMPK‐II) is a key regulatory enzyme in living cells. Modulation of its activity, therefore, could have a major impact on many cellular processes. We found that Zn2+ has multiple functional effects on CaMPK‐II. Zn2+ generated a Ca2+/CaM‐independent activity that correlated with the autophosphorylation of Thr286, inhibited Ca2+/CaM binding that correlated with the autophosphorylation of Thr306, and inhibited CaMPK‐II activity at high concentrations that correlated with the autophosphorylation of Ser279. The relative level of autophosphorylation of these three sites was dependent on the concentration of zinc used. The autophosphorylation of at least these three sites, together with Zn2+ binding, generated an increased mobility form of CaMPK‐II on sodium dodecyl sulfate gels. Overall, autophosphorylation induced by Zn2+ converts CaMPK‐II into a different form than the binding of Ca2+/CaM. In certain nerve terminals, where Zn2+ has been shown to play a neuromodulatory role and is present in high concentrations, Zn2+ may turn CaMPK‐II into a form that would be unable to respond to calcium signals.

Controlling the cell cycle: the role of calcium/calmodulin-stimulated protein kinases I and II

Many studies have implicated Ca2+ and calmodulin (CaM) as regulators of the cell cycle. Ca2+/CaM-stimulated proteins, including the family of multifunctional Ca2+/CaM-stimulated protein kinases (CaMK), have also been identified as mediators of cell cycle progression. CaMKII is the best-characterized member of this family, and is regulated by multi-site phosphorylation and targeting. Using pharmacological inhibitors that were believed to be specific for CaMKII, CaMKII has been implicated in every phase of the cell cycle. However, these ‘specific’ inhibitors also produce effects on other CaMKs. These additional effects are usually ignored, and the effects of the inhibitors are normally attributed to CaMKII without further investigation. Using new specific molecular techniques, it has become clear that CaMKI is an important regulator of G1, whereas CaMKII is essential for regulating G2/M and the metaphase-anaphase transition. If the mechanisms controlling these events can be fully elucidated, new targets for controlling proliferative diseases may be identified.

Multiple Forms and Distribution of Calcium/Calmodulin-Stimulated Protein Kinase II in Brain

In recent years, the enzyme Ca2+/calmodulin‐stimulated protein kinase II1 (CaM‐PK II) as attracted a great deal of interest. CaM‐PK II is the most abundant calmodulin‐stimulated protein kinase in brain, where it is particularly enriched in neurons (Ouimet et al., 1984; Erondu and Kennedy, 1985; Lin et al., 1987; Scholz et al., 1988). Neuronal CaM‐PK II has been suggested to be involved in several phenomena associated with synaptic plasticity (Lisman and Goldring, 1988; Kelly, 1992), including long‐term potentiation (Malinow et al., 1988; Malenka et al.,1989), neurotransmission (Nichols et al., 1990; Siekevitz, 1991), and learning (for review, see Rostas, 1991). This enzyme has also been postulated to be selectively vulnerable in several pathological condition, including epilepsy/kindling (Bronstein et al.,1990; Wu et al., 1990), cerebral ischemia (Taft et al., 1988), and organophosphorus toxicity (Abou‐Donia and Lapadula, 1990).

The role of serine/threonine protein phosphatases in exocytosis.

Modulation of exocytosis is integral to the regulation of cellular signalling, and a variety of disorders (such as epilepsy, hypertension, diabetes and asthma) are closely associated with pathological modulation of exocytosis. Emerging evidence points to protein phosphatases as key regulators of exocytosis in many cells and, therefore, as potential targets for the design of novel therapies to treat these diseases. Diverse yet exquisite regulatory mechanisms have evolved to direct the specificity of these enzymes in controlling particular cell processes, and functionally driven studies have demonstrated differential regulation of exocytosis by individual protein phosphatases. This Review discusses the evidence for the regulation of exocytosis by protein phosphatases in three major secretory systems, (1) mast cells, in which the regulation of exocytosis of inflammatory mediators plays a major role in the respiratory response to antigens, (2) insulin-secreting cells in which regulation of exocytosis is essential for metabolic control, and (3) neurons, in which regulation of exocytosis is perhaps the most complex and is essential for effective neurotransmission.