Synapsin I bundles F-actin in a phosphorylation-dependent manner

Cycling of actin assembly in synaptosomes and neurotransmitter release

We have investigated the regulation of actin assembly in whole mouse brain synaptosomes and how that regulation modulates neurotransmitter release. During a 30 s depolarization with high K+, filamentous actin (F-actin) levels, monitored by staining with rhodamine phalloidin, increase dramatically (up to 300% in 3 s), decrease, and increase once again. This F-actin cycling is regulated by pathways both dependent and independent of Ca2+ influx and is markedly affected by exposing synaptosomes to Li+, tetrodotoxin, and diacylglycerol. Measurement of [3H]norepinephrine release from synaptosomes containing entrapped agents that modulate actin assembly (DNAase I or phalloidin) indicates that actin depolymerization is necessary for normal release and that repolymerization limits release.

Dynamic properties of the Ca2+/calmodulin-dependent protein kinase in Drosophila: identification of a synapsin I-like protein

Visual adaptation with blue light induces a change in a special light/dark choice behavior in Drosophila. On the molecular level adaptation induces long-term modulation of the in vitro autophosphorylation capacity of a Ca2+/calmodulin-dependent protein kinase. Here I describe a Drosophila phosphoprotein that is a substrate of this protein kinase. The molecular mass and phosphopeptide composition of this protein are similar to those of rat synapsin I. Furthermore, the Drosophila protein shows immunological cross-reactivity with monoclonal antibodies against rat synapsin I. I conclude that this 86-kDa protein in Drosophila is homologous to the vertebrate synapsin I.

Protein phosphorylation and the two stages of pigment organelle dispersion in permeabilized xanthophores: organelle protein phosphorylation alone supports only the first stage

We reported previously that, in cultured goldfish xanthophores, dispersion of aggregated carotenoid droplets (CDs) requires the specific phosphorylation of the CD protein p57 by a cAMP-dependent protein kinase and the presence of cytosol. We report here that, in permeabilized cells, the addition of the catalytic subunit of cAMP-dependent protein kinase and ATP phosphorylates p57 and converts the CDs from an immobile to a mobile state (first stage of CD dispersion). However, the CDs are restricted to the vicinity of the original site of the CD aggregate and do not actually disperse (second stage of CD dispersion) unless cytosol is also added. We propose that this process may be related to aspects of secretory processes.

Localization of actin and tubulin in developing and adult mammalian photoreceptors

In order to define cytoskeletal domains of the mammalian photoreceptor, actin and tubulin were localized in adult retinae of mouse and human. For light-microscopic localization, actin was labeled using fluorescent phalloidin or monoclonal antibodies against actin, and tubulin was labeled using monoclonal antibodies against alpha- and beta-tubulin in an immunocytochemical method. Actin and tubulin were also localized by ultrastructural immunocytochemistry in the mouse. Filamentous actin was present in the retina at the outer limiting membrane and in synaptic terminals, especially of the cones, while globular actin was observed additionally in the inner segments. Müller cell cytoplasm and apical microvilli at the outer limiting membrane were also labeled for filamentous actin. Alpha- and beta-tubulin were evident throughout the photoreceptors, including the inner segments, but not in the synaptic terminals or at the outer limiting membrane. In the early postnatal retina of mouse, actin and tubulin were present at the ventricular surface. This pattern changed as photoreceptors fully elongated and as synaptogenesis occurred in the outer plexiform layer.

Characterization of synapsin I fragments produced by cysteine-specific cleavage: a study of their interactions with F-actin

Synapsin I is a neuron-specific phosphoprotein that is concentrated in the presynaptic nerve terminal in association with the cytoplasmic surface of synaptic vesicles. It has been demonstrated to bundle F-actin in a phosphorylation-dependent manner in vitro, a property consistent with its proposed role in linking synaptic vesicles to the cytoskeleton and its involvement in the regulation of neurotransmitter release. Synapsin I is composed of two distinct domains, a COOH terminal, collagenase-sensitive, hydrophilic, and strongly basic tail region, and an NH2 terminal, collagenase-resistant head region relatively rich in hydrophobic amino acids. To elucidate the structural basis for the interactions between synapsin I and F-actin and how it relates to other characteristics of synapsin I, we have performed a structure-function analysis of fragments of synapsin I produced by cysteine-specific cleavage with 2-nitro-5-thiocyanobenzoic acid. The fragments were identified and aligned with the parent molecule using the deduced primary structure of synapsin I and the known phosphorylation sites as markers. We have purified these fragments and examined their interactions with F-actin. Two distinct fragments, a 29-kD NH2-terminal fragment and a 15-kD middle fragment, were shown to contain F-actin binding sites. A 51/54-kD middle/tail fragment retained the F-actin binding and bundling activity of synapsin I, but the isolated tail fragment did not retain either activity. In contrast to phosphorylation of sites two and three in intact synapsin I, which abolishes F-actin bundling activity, phosphorylation of these sites in the middle/tail fragment failed to abolish this activity. In conclusion, three domains of synapsin I appear to be involved in F-actin binding and bundling.

Interactions of synapsin I with small synaptic vesicles: distinct sites in synapsin I bind to vesicle phospholipids and vesicle proteins

Synapsin I is a major neuron-specific phosphoprotein that is specifically localized to the cytoplasmic surface of small synaptic vesicles. In the present study, the binding of synapsin I to small synaptic vesicles was characterized in detail. The binding of synapsin I was preserved when synaptic vesicles were solubilized and reconstituted in phosphatidylcholine. After separation of the protein and lipid components of synaptic vesicles under nondenaturing conditions, synapsin I bound to both components. The use of hydrophobic labeling procedures allowed the assessment of interactions between phospholipids and synapsin I in intact synaptic vesicles. Hydrophobic photolabeling followed by cysteine-specific cleavage of synapsin I demonstrated that the head domain of synapsin I penetrates into the hydrophobic core of the bilayer. The purified NH2-terminal fragment, derived from the head domain by cysteine-specific cleavage, bound to synaptic vesicles with high affinity confirming the results obtained from hydrophobic photolabeling. Synapsin I binding to synaptic vesicles could be inhibited by the entire molecule or by the combined presence of the NH2-terminal and tail fragments, but not by an excess of either NH2-terminal or tail fragment alone. The purified tail fragment bound with relatively high affinity to synaptic vesicles, though it did not significantly interact with phospholipids. Binding of the tail fragment was competed by holosynapsin I; was greatly decreased by phosphorylation; and was abolished by high ionic strength conditions or protease treatment of synaptic vesicles. The data suggest the existence of two sites of interaction between synapsin I and small synaptic vesicles: binding of the head domain to vesicle phospholipids and of the tail domain to a protein component of the vesicle membrane. The latter interaction is apparently responsible for the salt and phosphorylation dependency of synapsin I binding to small synaptic vesicles.

Electrostatic and hydrophobic interactions of synapsin I and synapsin I fragments with phospholipid bilayers

Synapsin I, a major neuron-specific phosphoprotein, is localized on the cytoplasmic surface of small synaptic vesicles to which it binds with high affinity. It contains a collagenase-resistant head domain and a collagenase-sensitive elongated tail domain. In the present study, the interaction between synapsin I and phospholipid vesicles has been characterized, and the protein domains involved in these interactions have been identified. When lipid vesicles were prepared from cholesterol and phospholipids using a lipid composition similar to that found in native synaptic vesicle membranes (40% phosphatidylcholine, 32% phosphatidylethanolamine, 12% phosphatidylserine, 5% phosphatidylinositol, 10% cholesterol, wt/wt), synapsin I bound with a dissociation constant of 14 nM and a maximal binding capacity of about 160 fmol of synapsin I/microgram of phospholipid. Increasing the ionic strength decreased the affinity without greatly affecting the maximal amount of synapsin I bound. When vesicles containing cholesterol and either phosphatidylcholine or phosphatidylcholine/phosphatidylethanolamine were tested, no significant binding was detected under any conditions examined. On the other hand, phosphatidylcholine vesicles containing either phosphatidylserine or phosphatidylinositol strongly interacted with synapsin I. The amount of synapsin I maximally bound was directly proportional to the percentage of acidic phospholipids present in the lipid bilayer, whereas the Kd value was not affected by varying the phospholipid composition. A study of synapsin I fragments obtained by cysteine-specific cleavage showed that the collagenase-resistant head domain actively bound to phospholipid vesicles; in contrast, the collagenase-sensitive tail domain, though strongly basic, did not significantly interact. Photolabeling of synapsin I was performed with the phosphatidylcholine analogue 1-palmitoyl-2-[11-[4-[3-(trifluoromethyl)diazirinyl]phenyl] [2-3H]undecanoyl]-sn-glycero-3-phosphocholine; this compound generates a highly reactive carbene that selectively interacts with membrane-embedded domains of membrane proteins. Synapsin I was significantly labeled upon photolysis when incubated with lipid vesicles containing acidic phospholipids and trace amounts of the photoactivatable phospholipid. Proteolytic cleavage of photolabeled synapsin I localized the label to the head domain of the molecule.(ABSTRACT TRUNCATED AT 400 WORDS)

Calcium/calmodulin-dependent protein kinase II

Images

cAMP-independent and cAMP-dependent protein phosphorylations by isolated goldfish xanthophore cytoskeletons: evidence for the association of cytoskeleton with a carotenoid droplet protein

Triton-insoluble cytoskeleton of nonpigment cells has bound protein kinase that phosphorylates, with or without added cAMP, tubulins and the intermediate filament proteins p60, p56, p53, and p45a to give multiple charge variants. In the absence of 8-Br-cAMP, Triton-insoluble cytoskeletons from xanthophores also phosphorylate p60, p56, and p45a, but not p53; tubulin phosphorylation may also be reduced. In the presence of 8-Br-cAMP, p53, as well as several other peptides, are phosphorylated. One of these latter peptides was identified as the carotenoid droplet (pigment organelle) protein p57, whose phosphorylation and dephosphorylation precede pigment dispersion and aggregation respectively (Lynch et al.: J. Biol. Chem. 261:4204-4211, 1986). The amount of pp57 produced depends on the state of pigment distribution in the xanthophores used to prepare the cytoskeletons for labeling. With cytoskeletons from xanthophores with aggregated pigment, pp57 is a major labeled phosphoprotein seen in two-dimensional gels. With cytoskeletons prepared from xanthophores with dispersed pigment, the yield of labeled pp57 is greatly reduced (by at least 90%). Together with earlier results, we propose that, in the aggregated state, p57 serves to bind carotenoid droplets to the cytoskeletons, most likely the microtubules. The significance of other cAMP-dependent phosphorylation reactions is unknown but may be related to cAMP-induced cytoskeleton rearrangement in intact xanthophores.

Purification of anterogin, a protein factor necessary for the dispersion of carotenoid droplets in permeabilized xanthophores of goldfish

We reported previously that the dispersion of carotenoid droplets in permeabilized xanthophores requires cAMP, ATP, and a cytosolic factor present in several secretory tissues as well as in xanthophores. We have now purified this factor from beef liver to apparent/near homogeneity. It appears to be a heterodimer with Mr approximately 125,000. The purified factor has little or no ATPase activity, with or without the presence of actin. Nor does it stimulate the ATPase activity of carotenoid droplets. Its exact function in carotenoid droplet dispersion is thus unclear. Since dispersion of carotenoid droplets is an anterograde translocation, we propose the name anterogin for this protein. We also report that yeast cytosol has anterogin activity.

The cytoskeletal architecture of the presynaptic terminal and molecular structure of synapsin 1

We have examined the cytoskeletal architecture and its relationship with synaptic vesicles in synapses by quick-freeze deep-etch electron microscopy (QF.DE). The main cytoskeletal elements in the presynaptic terminals (neuromuscular junction, electric organ, and cerebellar cortex) were actin filaments and microtubules. The actin filaments formed a network and frequently were associated closely with the presynaptic plasma membranes and active zones. Short, linking strands approximately 30 nm long were found between actin and synaptic vesicles, between microtubules and synaptic vesicles. Fine strands (30-60 nm) were also found between synaptic vesicles. Frequently spherical structures existed in the middle of the strands between synaptic vesicles. Another kind of strand (approximately 100 nm long, thinner than the actin filaments) between synaptic vesicles and plasma membranes was also observed. We have examined the molecular structure of synapsin 1 and its relationship with actin filaments, microtubules, and synaptic vesicles in vitro using the low angle rotary shadowing technique and QF.DE. The synapsin 1, approximately 47 nm long, was composed of a head (approximately 14 nm diam) and a tail (approximately 33 nm long), having a tadpole-like appearance. The high resolution provided by QF.DE revealed that a single synapsin 1 cross-linked actin filaments and linked actin filaments with synaptic vesicles, forming approximately 30-nm short strands. The head was on the actin and the tail was attached to the synaptic vesicle or actin filament. Microtubules were also cross-linked by a single synapsin 1, which also connected a microtubule to synaptic vesicles, forming approximately 30 nm strands. The spherical head was on the microtubules and the tail was attached to the synaptic vesicles or to microtubules. Synaptic vesicles incubated with synapsin 1 were linked with each other via fine short fibrils and frequently we identified spherical structures from which two or three fibril radiated and cross-linked synaptic vesicles. We have examined the localization of synapsin 1 using ultracryomicrotomy and colloidal gold-immunocytochemistry of anti-synapsin 1 IgG. Synapsin 1 was exclusively localized in the regions occupied by synaptic vesicles. Statistical analyses indicated that synapsin 1 is located mostly at least approximately 30 nm away from the presynaptic membrane. These data derived via three different approaches suggest that synapsin 1 could be a main element of short linkages between actin filaments and synaptic vesicles, and between microtubules and synaptic vesicles, and between synaptic vesicles in the nerve terminals.(ABSTRACT TRUNCATED AT 400 WORDS)

Neuronal cytomechanics: the actin-based motility of growth cones

The patterns of synaptic connection that underlie brain function depend on the elaborate forms characteristic of neurons. It is therefore a central goal of neuroscience to understand the molecular basis for neuronal shape. Neuronal pathfinding during development is one major determinant of neuronal shape: growing nerve axons and dendrites must navigate, branch, and locate targets in response to extracellular cue molecules within the embryo. The leading tips of growing nerve processes, structures known as growth cones, contain especially high concentrations of the ubiquitous mechanochemical protein actin. Force generation involving this cytoskeletal molecule appears to be essential to the ability of growing nerve fibers to respond structurally to extracellular cues. New results from electronically enhanced light microscopy of living growth cones are helping to show how actin-based forces guide neurite growth and synapse formation.

Differential effect of alpha-latrotoxin on exocytosis from small synaptic vesicles and from large dense-core vesicles containing calcitonin gene-related peptide at the frog neuromuscular junction

The regulatory peptide called calcitonin gene-related peptide (CGRP) was detected by immunofluorescence in frog motor neurons and motor nerve terminals. In motor nerve terminals, CGRP-like immunoreactivity was found to be segregated within large dense-core vesicles. To determine whether exocytosis from acetylcholine-containing small synaptic vesicles and from CGRP-containing large dense-core vesicles can be independently stimulated, nerve-muscle preparations were exposed to alpha-latrotoxin. This toxin induced complete depletion of acetylcholine-containing small synaptic vesicles but did not induce a parallel depletion of CGRP-like immunoreactivity and of large dense-core vesicles. These effects were independent of the presence of extracellular Ca2+ and occurred both at room temperature and at low temperature (1-3 degrees C). These findings suggest that exocytosis from the two vesicle populations is mediated by distinct biochemical mechanisms, which might be differentially regulated by physiological stimuli.

Abolition of actin-bundling by phosphorylation of human erythrocyte protein 4.9

Protein 4.9, first identified as a component of the human erythrocyte membrane skeleton, binds to and bundles actin filaments. Protein 4.9 is a substrate for various kinases, including a cyclic AMP(cAMP)-dependent one, in vivo and in vitro. We show here that phosphorylation of protein 4.9 by the catalytic subunit of cAMP-dependent protein kinase reversibly abolishes its actin-bundling activity, but phosphorylation by protein kinase C has no such effect. A quantitative immunoassay showed that human erythrocytes contain 43,000 trimers of protein 4.9 per cell, which is equivalent to one trimer for each actin oligomer in these red blood cells. As analogues of protein 4.9 have been identified together with analogues of other erythroid skeletal proteins in non-erythroid tissues of numerous vertebrates, phosphorylation and dephosphorylation of protein 4.9 may be the basis for a mechanism that regulates actin bundling in many cells.

Villin sequence and peptide map identify six homologous domains

Site-specific proteases and antisera to the amino terminus of villin have been used to show that villin is organized into seven protease-resistant domains. Six are contained in the amino-terminal Mr 87,000 villin core, a Ca2+-regulated actin-severing fragment, whereas the carboxyl-terminal domain includes the villin "headpiece," a fragment involved in bundling of actin filaments. Ca2+ inhibits proteolytic cleavage between domains in the amino-terminal half of villin. The protein sequence of villin deduced from a single cDNA clone contains a conserved sequence that is repeated six times and is found in each domain of the villin core. The conserved repeats are found in other actin-severing proteins but not in the villin headpiece. Our results suggest that actin-severing proteins are organized around a common Mr 14,000-17,000 domain.

Autophosphorylation of smooth-muscle caldesmon

Caldesmon, a major actin- and calmodulin-binding protein of smooth muscle, has been implicated in regulation of the contractile state of smooth muscle. The isolated protein can be phosphorylated by a co-purifying Ca2+/calmodulin-dependent protein kinase, and phosphorylation blocks inhibition of the actomyosin ATPase by caldesmon [Ngai & Walsh (1987) Biochem. J. 244, 417-425]. We have examined the phosphorylation of caldesmon in more detail. Several lines of evidence indicate that caldesmon itself is a kinase and the reaction is an intermolecular autophosphorylation: (1) caldesmon (141 kDa) and a 93 kDa proteolytic fragment of caldesmon can be separated by ion-exchange chromatography: both retain caldesmon kinase activity, which is Ca2+/calmodulin-dependent; (2) chymotryptic digestion of caldesmon generates a Ca2+/calmodulin-independent form of caldesmon kinase; (3) caldesmon purified to electrophoretic homogeneity retains caldesmon kinase activity, and elution of enzymic activity from a fast-performance-liquid-chromatography ion-exchange column correlates with caldesmon of Mr 141,000; (4) caldesmon is photoaffinity-labelled with 8-azido-[alpha-32P]ATP; labelling is inhibited by ATP, GTP and CTP, indicating a lack of nucleotide specificity; (5) caldesmon binds tightly to Affi-Gel Blue resin, which recognizes proteins having a dinucleotide fold. Autophosphorylation of caldesmon occurs predominantly on serine residues (83.3%), with some threonine (16.7%) and no tyrosine phosphorylation. Autophosphorylation is site-specific: 98% of the phosphate incorporated is recovered in a 26 kDa chymotryptic peptide. Complete tryptic/chymotryptic digestion of this phosphopeptide followed by h.p.l.c. indicates three major phosphorylation sites. Caldesmon exhibits a high degree of substrate specificity: apart from autophosphorylation, brain synapsin I is the only good substrate among many potential substrates examined. These observations indicate that caldesmon may regulate its own function (inhibition of the actomyosin ATPase) by Ca2+/calmodulin-dependent autophosphorylation. Furthermore, caldesmon may regulate other cellular processes, e.g. neurotransmitter release, through the Ca2+/calmodulin-dependent phosphorylation of other proteins such as synapsin I.

Redistribution of synaptic vesicle antigens is correlated with the disappearance of a transient synaptic zone in the developing cerebral cortex

To examine the distribution of synaptic vesicle antigens during development of the cerebral cortex, antibodies against synapsin I and p65 were used on sections of cat cerebral cortex between E40 and adulthood. In the adult, the layers of the cerebral cortex are immunoreactive for each of these antigens, while the white matter is free of staining. In contrast, the fetal and neonatal pattern of immunostaining is reversed: the cortical plate (future cortical layers) is devoid of immunoreactivity, while the marginal (future layer 1) and the intermediate zones (future white matter) are stained. Electron microscopic immunohistochemistry shows that immunolabeling is associated with presynaptic nerve terminals in the adult and during development. These observations suggest that during development the white matter is a transient synaptic neuropil and that a global redistribution of synapses takes place as the mature pattern of connections within the cerebral cortex emerges.

Identification and characterization of the major proteins of mammalian brain synaptic vesicles

Highly purified rat and cow brain synaptic vesicles contain major proteins with molecular weights of approximately 74,000, 60,000, 57,000, 40,000, 38,000, and 34,000 by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The presence of the major proteins on synaptic vesicles was confirmed by immunoprecipitation of intact rat brain synaptic vesicles with a synaptic vesicle-specific monoclonal antibody. The 40,000-Mr protein appeared to be identical to the 38,000-Mr integral membrane glycoprotein, p38 or synaptophysin, previously identified as a major component of mammalian synaptic vesicles. The isoelectric point of the 75,000-Mr proteins from either rat or cow brain synaptic vesicles is 5.0, and the pI of the 57,000-Mr protein is approximately 5.1 in both species. The similarity in size and charge of several major proteins in rat and cow synaptic vesicles suggests a high degree of structure conservation of these proteins in diverse mammalian species and raises the possibility that a set of functions common to most or all mammalian synaptic vesicles is mediated by these proteins.

Regulation of the distribution of carotenoid droplets in goldfish xanthophores and possible implication to secretory processes

In goldfish xanthophores, the formation of pigment aggregate requires: 1) that a pigment organelle (carotenoid droplet) protein p57 be in the unphosphorylated state; 2) that self-association of pigment organelles occur in a microtubule-independent manner; and 3) that pigment organelles via p57 associate with microtubules. In the fully aggregated state, the pigment organelles are completely stationary. Pigment dispersion is initiated by activation of a cAMP-dependent protein kinase, which phosphorylates p57 and allows pigment dispersion via an active process dependent on F-actin and a cytosolic factor. This factor is not an ATPase, and its function is unknown. However, its abundance in different tissues parallels secretory activity of the tissues, suggesting a similarity between secretion and pigment dispersion in xanthophores. The identity of the motor for pigment dispersion is unclear. Experimental results show that pigment organelles isolated from cells with dispersed pigment have associated actin and ATPase activity comparable to myosin ATPase. This ATPase is probably an organelle protein of relative molecular mass approximately 72,000, and unlikely to be an ion pump. Isolated pigment organelles without associated actin have 5x lower ATPase activity. Whether this organelle ATPase is the motor for pigment dispersion is under investigation. The process of pigment aggregation is poorly understood, with conflicting results for and against the involvement of intermediate filaments.

Amino acid sequences surrounding the cAMP-dependent and calcium/calmodulin-dependent phosphorylation sites in rat and bovine synapsin I

The amino acid sequences surrounding three major phosphorylation sites in rat and bovine synapsin I have been determined by employing automated gas-phase sequencing and manual Edman degradation of purified phosphopeptide fragments. Site 1 is a serine residue phosphorylated by cAMP-dependent protein kinase and by calcium/calmodulin-dependent protein kinase I. The sequence around site 1 was derived from tryptic/chymotryptic phosphopeptides and overlapping cyanogen bromide cleavage fragments. This sequence, identical in rat and bovine synapsin I, is Asn-Tyr-Leu-Arg-Arg-Arg-Leu-Ser(P)-Asp-Ser-Asn-Phe-Met. Site 1 is located at the NH2 terminus of the protein, within the collagenase-resistant head region. Sites 2 and 3 are serine residues phosphorylated by calcium/calmodulin-dependent protein kinase II. The sequences surrounding bovine site 2 and site 3 were derived from tryptic phosphopeptides and overlapping fragments generated by cleavage with chymotrypsin, collagenase, and endoproteinase Lys-C. The sequence around bovine site 2 is Thr-Arg-Gln-Thr-Ser(P)-Val-Ser-Gly-Gln-Ala-Pro-Pro-Lys, and the sequence around bovine site 3 is Thr-Arg-Gln-Ala-Ser(P)-Gln-Ala-Gly-Pro-Met-Pro-Arg. Sites 2 and 3 are located within the COOH-terminal, collagenase-sensitive tail region of the molecule, separated by 36 amino acids. The sequences surrounding rat site 2 and site 3 were derived from tryptic phosphopeptides. The sequence around rat site 2 is Gln-Ala-Ser(P)-Ile-Ser-Gly-Pro-Ala-Pro-Pro-Lys, and the sequence around rat site 3 is Gln-Ala-Ser(P)-Gln-Ala-Gly-Pro-Gly-Pro-Arg. Thus, the sequences surrounding the four sites that are phosphorylated by calcium/calmodulin-dependent protein kinase II, namely sites 2 and 3 in rat and bovine synapsin I, exhibit a high degree of homology.