Rab effector domain peptides stimulate the release of neurotransmitter from cell cultured synapses

Properties and plasticity of paired-pulse depression at a central synapse

Synaptic depression was studied at the axo-axonic connection between the goldfish Mauthner axon and identified cranial relay interneurons using simultaneous presynaptic and postsynaptic recordings and a paired-pulse stimulus paradigm. We used interstimulus intervals (ISIs) ranging from 10 msec to 1 sec and a cycle time of approximately 5 sec. Depression (Delta EPSP/EPSP1) was maximal at the shorter intervals (80%) and decreased exponentially with a tau approximately 400 msec (360 +/- 107 msec, mean +/- SD). We found the amplitudes of the first and second EPSP were not correlated, indicating the magnitude of depression does not depend on the amount of transmitter released by the conditioning stimulus. At short ISIs, the latency of EPSP2 was 23% longer than that of EPSP1 and recovered to control with tau approximately 400 msec, whereas rise time and decay time were not altered significantly. The latency distribution, which is determined by the timing of the first quantum released each trial, was used to derive alpha(t), the rate of evoked exocytosis after an action potential. alpha(t) was biphasic, and both components were consistently delayed during depression. Presynaptic manipulations of putative intracellular regulatory pathways, such as Ca(2+) and GTPgammaS injections, preferentially affected the amplitude of EPSP1 or EPSP2. These results are not consistent with simple depletion of the available pool of synaptic vesicles as the major mechanism underlying depression. They rather suggest that it is attributable to a modification or refractoriness of the release process and that there may be multiple pathways subserving evoked exocytosis.

rab3 mediates cortical granule exocytosis in the sea urchin egg

Egg activation at fertilization in the sea urchin results in the exocytosis of approximately 15,000 cortical granules that are docked at the plasma membrane. Previously, we reported that several integral membrane proteins modeled in the SNARE hypothesis, synaptotagmin, VAMP, and syntaxin, in addition to a small GTPase of the ras superfamily, rab3, were present on cortical granules (Conner, S., Leaf, D., and Wessel, G., Mol. Reprod. Dev. 48, 1-13, 1997). Here we report that rab3 is associated with cortical granules throughout oogenesis, during cortical granule translocation, and while docked at the egg plasma membrane. Following cortical granule exocytosis, however, rab3 reassociates with a different population of vesicles, at least some of which are of endocytic origin. Because of its selective association with cortical granules in eggs and oocytes, we hypothesize that rab3 functions in cortical granule exocytosis. To test this hypothesis, we used a strategy of interfering with rab3 function by peptide competition with its effector domain, a conserved region within specific rab types. We first identified the effector domain sequence in Lytechinus variegatus eggs and find the sequence 94% identical to the effector domain of rab3 in Stronglocentrotus purpuratus. Then, with synthetic peptides to different regions of the rab3 protein, we find that cortical granule exocytosis is inhibited in eggs injected with effector domain peptides, but not with peptides from the hypervariable region or with a scrambled effector peptide. Additionally, effector-peptide-injected eggs injected with IP3 are blocked in their ability to exocytose cortical granules, suggesting that the inhibition is directly on the membrane fusion event and not the result of interference with the signal transduction mechanism leading to calcium release. We interpret these results to mean that rab3 functions in the regulation of cortical granule exocytosis following vesicle docking.

Role of Rab GTPases in membrane traffic

Small GTPases of the Rab subfamily have been known to be key regulators of intracellular membrane traffic since the late 1980s. Today this protein group amounts to more than 40 members in mammalian cells which localize to distinct membrane compartments and exert functions in different trafficking steps on the biosynthetic and endocytic pathways. Recent studies indicate that cycles of GTP binding and hydrolysis by the Rab proteins are linked to the recruitment of specific effector molecules on cellular membranes, which in turn impact on membrane docking/fusion processes. Different Rabs may, nevertheless, have slightly different principles of action. Studies performed in yeast suggest that connections between the Rabs and the SNARE machinery play a central role in membrane docking/fusion. Further elucidation of this linkage is required in order to fully understand the functional mechanisms of Rab GTPases in membrane traffic.

Synthetic peptides containing a BXBXXXB(B) motif activate phospholipase C-beta1

We have recently shown that synthetic peptides of the effector domain of the low-molecular-mass GTP-binding protein Rab3 stimulate inositol 1,4,5-trisphosphate production in various permeabilized cells. To investigate the mechanism of the peptide-induced activation of phospholipase C (PLC) and to identify the PLC isoenzyme(s) targeted by these peptides, isolated pancreatic acinar membranes and cytosol were preincubated with anti-PLC antibodies before examination of PLC activity in response to the Rab3B/D effector-domain peptide (VSTVGIDFKVKTVYRH, peptide P1). Western blot analysis revealed the presence of PLC-beta1, -beta3, -gamma1 and -delta1 in membrane and cytosolic fractions. P1 stimulated PLC activity in both membrane and cytosolic fractions. Anti-(PLC-beta1) antibody inhibited P1-induced PLC activity in both subcellular fractions almost completely. Moreover, P1-induced amylase release in digitonin-permeabilized pancreatic acini was also inhibited. Other immunoneutralizing anti-PLC antibodies had no effect, suggesting that P1 activates PLC-beta1 but not PLC-beta3, -gamma1 or -delta1. P1 also activated recombinant PLC-beta1, indicating direct activation of PLC-beta1 by Rab3 effector-domain peptides. To investigate further the structure-function relationship of the peptides, truncated peptides of P1 were tested for their ability to activate PLC in isolated pancreatic acinar membranes and to stimulate amylase release from digitonin-permeabilized pancreatic acini. Peptides containing a BXBXXXB(B) motif (where B represents a basic residue and X any residue)[KVKTVYRH (EC50 of 1 nM to stimulate amylase release) approximately TVGIDFKVKTVYRH > TVGIDFKVKTVYR] were potent stimulators of amylase release and PLC activity, whereas deletion of the C-terminus (VSTVGIDF), of the two basic C-terminal amino acid residues (VSTVGIDFKVKTVY and KVKTVY), or destruction of the BXB motif (VKTVYR) resulted in inactive peptides. In conclusion, the results of the present study show that short peptides containing a BXBXXXB motif represent promising pharmacological agents to activate the PLC-beta1 isoenzyme.

Facilitation of Ca2+ action potential frequency by a small G protein Rab3A in rat pituitary GH3 cells

GH3 pituitary cells have high tendency to exhibit spontaneous Ca2+ action potentials and their frequency (Ca2+ APF) is increased by treatment with thyrotropin-releasing hormone (TRH). Although spontaneous Ca2+ firing was thought to be significant for the induction of oscillations in cytosolic Ca2+ concentration ([Ca2+]i), little attempt to elucidate the mechanism has been done so far. We demonstrate here that spontaneous Ca2+ APF in GH3 cells was increased 1.5-3 fold, comparable to that for TRH, by injection of guanosine 5'-0-3-thiotriphosphate (GTPgammaS), rab3A effector domain peptide, and phorbol-dibutyrate (PDBu), whereas guanosine 5'-O-(2-thiodiphosphate) (GDPbetaS), H-rab5 peptide, ras peptide, and 4 alpha-phorbol did not. The enhancement of Ca2+ firing by rab3A effector domain peptide was blocked by a protein kinase C (PKC) inhibitor, PKC(19-36). The present study suggests that the spontaneous Ca2+APF may be controlled by small G protein phosphorylated by PKC.

Rat basophilic leukaemia (RBL) cells overexpressing Rab3a have a reversible block in antigen-stimulated exocytosis

The rat basophilic leukaemia (RBL) cell line has been widely used as a convenient model system to study regulated secretion in mast cells. Activation of these cells through the high-affinity receptor for IgE (Fcepsilon-RI) results in degranulation and the extracellular release of mediators. There is good evidence of a role for GTPases in mast cell degranulation, and a number of studies with peptides derived from the Rab3a effector domain have suggested that Rab3a may function in this process. However, in neuroendocrine cells, overexpression of Rab3a can act as a negative regulator of stimulated exocytosis [Holz, Brondyk, Senter, Kuizon and Macara (1994) J. Biol. Chem. 269, 10229-10234; Johanes, Lledo, Roa, Vincent, Henry and Darchen (1994) EMBO J. 13, 2029-2037]. In order to study the function of Rab3a in RBL degranulation, we have generated clones of RBL cells stably expressing Rab3a, and show that in these haematopoietic cells Rab3a can also function as a negative regulator of exocytosis. Overexpression of a mutant form of Rab3a (Asn-135 to Ile), which is predicted to be predominantly GTP-bound, also inhibited degranulation. However, overexpression of a mutant form of Rab3a that was truncated at the C-terminus to remove the sites for geranylgeranylation failed to inhibit degranulation. The effect of Rab3a is specific to secretion, and we observe no effect of Rab3a on receptor-mediated endocytosis. The Rab3a-induced block in degranulation can be bypassed by stimulation of streptolysin-O-permeabilized cells with guanosine 5'-[gamma-thio]triphosphate. We conclude from these studies that Rab3a is implicated in an early stage of granule targeting, whereas fusion of granules with the plasma membrane is regulated by a distinct downstream GTP-binding protein or proteins.

A unifying hypothesis for acetylcholine release

Mediatophore is the only nerve terminal membrane protein known to translocate acetylcholine upon calcium action. It is localized at the active zone. In this review we attempted to describe its role in relation to the vesicular and membrane protein complexes that are formed at the active zone. The model pictures a possible set of sequential steps that lead to exocytosis. The smallest quantal events are attributed to mediatophore opening momentarily, while synaptic vesicles synchronize release by controlling the calcium microdomain. A clear distinction is made between sub-quantal ACh release preserved after Botulinum toxin action, and exocytosis of vesicular contents. A cybernetic model for release and exocytosis related to protein interactions is presented for future works.

Stimulation of inositol 1,4,5-trisphosphate production by peptides corresponding to the effector domain of different Rab3 isoforms and cross-linking of an effector domain peptide target

Rab3 proteins are localized on secretory vesicles and appear to be involved in regulated exocytosis. We have previously shown that a modified peptide corresponding to the effector domain of the small molecular mass GTP-binding protein Rab3A, Rab3AAL, stimulates inositol 1,4,5-trisphosphate [Ins(1,4,5)P3] production and amylase release in digitonin-permeabilized pancreatic acini. Experiments using monoclonal antibodies reveal that the Rab3-like protein present in pancreatic acini is not the Rab3A isoform. However, since the putative effector domains of the four as yet known Rab3 proteins (A, B, C and D) differ only in the C-terminal four amino acid residues, Rab3A effector domain peptide could mimic the action of the pancreas-specific Rab3 isoform. In the present study we report that peptides corresponding to the different Rab3 isoforms stimulate both Ins(1,4,5)P3 production and amylase secretion with an order of potency Rab3B/D > Rab3AAL > Rab3A = Rab3C. For Rab3A, B/D and C effector domain peptides the concentrations causing half-maximal response (EC50) were 3, 0.2 and 3 nM for Ins(1,4,5)P3 accumulation and 0.3, 0.02 and 0.3 nM for amylase release, respectively. A Rab1A effector domain peptide, Rab1AAL, and a scrambled peptide of Rab3AAL were less potent by several orders of magnitude in eliciting these responses compared with native Rab3 effector domain peptides. None of the peptides influenced Ins(1,4,5)P3 production and amylase release in intact acini. Cross-linking of 125I-Rab3B/D peptide to pancreatic acinar membranes showed a band at 70 to 75 kDa with maximum intensity at 75 kDa. Radiolabelling of the substrates could be displaced by unlabelled Rab3B/D peptide, and to a lesser extend by Rab3A peptide, whereas the scrambled peptide of Rab3AAL had no effect. These data suggest that phospholipase C and exocytosis might be regulated by Rab3B-or Rab3D-like proteins in pancreatic acinar cells. A 75 kDa protein that preferentially cross-linked to 125I-Rab3B/D effector domain peptide is a potential candidate as an effector protein of Rab3 effector domain peptides.

Synaptic vesicle depletion in reticulospinal axons is reduced by 5-hydroxytryptamine: direct evidence for presynaptic modulation of glutamatergic transmission

5-hydroxytryptamine (5-HT; serotonin) is known to depress glutamatergic synaptic transmission in the spinal cord of vertebrates. To test directly whether 5-HT inhibits synaptic glutamate release, we examined its effect on the ultrastructure of synaptic vesicle clusters in giant reticulospinal axons in a lower vertebrate (lamprey; Lampetra fluviatilis). The size of these axons makes it possible to selectively expose only a part of the presynaptic element to 5-HT, while another part of the same axon is maintained in control solution. Action potential stimulation at 20 Hz for 20 min caused a marked reduction in the number of synaptic vesicles in active zones maintained in control solution, while in the part exposed to 5-HT (20 microM) the number of synaptic vesicles per active zone was approximately 3-fold higher. In contrast, 5-HT had no effect on the number of vesicles in resting axons. To examine whether 5-HT acts by reducing presynaptic Ca2+ influx, intra-axonal recordings of Ba2+ potentials were performed. No reduction of the axonal Ba2+ potential could be detected after application of 20 or 200 microM 5-HT. The present results show that 5-HT reduces the rate of synaptic exocytosis in reticulospinal axons. The effect appears to be mediated by a mechanism distinct from the presynaptic Ca2+ channels.

The synaptic vesicle and its targets

Synaptic vesicles play the central role in synaptic transmission. They are regarded as key organelles involved in synaptic functions such as uptake, storage and stimulus-dependent release of neurotransmitter. In the last few years our knowledge concerning the molecular components involved in the functioning of synaptic vesicles has grown impressively. Combined biochemical and molecular genetic approaches characterize many constituents of synaptic vesicles in molecular detail and contribute to an elaborate understanding of the organelle responsible for fast neuronal signalling. By studying synaptic vesicles from the electric organ of electric rays and from the mammalian cerebral cortex several proteins have been characterized as functional carriers of vesicle function, including proteins involved in the molecular cascade of exocytosis. The synaptic vesicle specific proteins, their presumptive function and targets of synaptic vesicle proteins will be discussed. This paper focuses on the small synaptic vesicles responsible for fast neuronal transmission. Comparing synaptic vesicles from the peripheral and central nervous systems strengthens the view of a high conservation in the overall composition of synaptic vesicles with a unique set of proteins attributed to this cellular compartment. Synaptic vesicle proteins belong to gene families encoding multiple isoforms present in subpopulations of neurons. The overall architecture of synaptic vesicle proteins is highly conserved during evolution and homologues of these proteins govern the constitutive secretion in yeast. Neurotoxins from different sources helped to identify target proteins of synaptic vesicles and to elucidate the molecular machinery of docking and fusion. Synaptic vesicle proteins and their markers are useful tools for the understanding of the complex life cycle of synaptic vesicles.

The G protein-activating peptide, mastoparan, and the synthetic NH2-terminal ARF peptide, ARFp13, inhibit in vitro Golgi transport by irreversibly damaging membranes

Mastoparan is a cationic amphipathetic peptide that activates trimeric G proteins, and increases binding of the coat protein beta-COP to Golgi membranes. ARFp13 is a cationic amphipathic peptide that is a putative specific inhibitor of ARF function, and inhibits coat protein binding to Golgi membranes. Using a combination of high resolution, three-dimensional electron microscopy and cell-free Golgi transport assays, we show that both of these peptides inhibit in vitro Golgi transport, not by interfering in the normal functioning of GTP-binding proteins, but by damaging membranes. Inhibition of transport is correlated with inhibition of nucleotide sugar uptake and protein glycoslation, a decrease in the fraction of Golgi cisternae exhibiting normal morphology, and a decrease in the density of Golgi-coated buds and vesicles. At peptide concentrations near the IC50 for transport, those cisternae with apparently normal morphology had a higher steady state level of coated buds and vesicles. Kinetic analysis suggests that this increase in density was due to a decrease in the rate of vesicle fission. Pertussis toxin treatment of the membranes appeared to increase the rate of vesicle formation, but did not prevent the membrane damage induced by mastoparan. We conclude that ARFp13 is not a specific inhibitor of ARF function, as originally proposed, and that surface active peptides, such as mastoparan, have the potential for introducing artifacts that complicate the analysis of trimeric G protein involvement in regulation of Golgi vesicle dynamics.

Peptides as probes for G protein signal transduction

Triggered by agonist binding to cell surface receptors, the heterotrimeric G proteins dissociate into alpha and beta gamma subunits, each activating distinct second messenger pathways. Peptides from the primary sequences of receptors, G proteins, and effectors have been used to study the molecular interactions between these proteins. Receptor-derived peptides from the second, third and fourth intracellular loops and certain naturally occurring peptides antagonize G protein interactions and can directly activate G protein. These peptides bind to G protein sites that include the N and C terminal regions of the alpha subunit and a yet to be identified region of the beta subunit. Peptides have also been useful in characterizing G protein-effector interactions. The identification of the contact sites between proteins involved in G protein signal transduction should aid in the development of non-peptide mimetic therapeutics which could specifically modify G protein-mediated cellular responses.

The exocytotic fusion pore and neurotransmitter release

Membrane fusion is ubiquitous in biological systems, occurring in the simplest of unicellular eukaryotes as well as higher eukaryotes. As soon as the first primitive eukaryotic cell utilized a lipid bilayer as an outer membrane, membrane fusion (and fission) became necessary for the traffic of material from the outside to the inside, the inside to the outside, and between different intracellular membrane-bounded compartments. The earliest cells would have made use of the intrinsic ability of lipid bilayers to fuse under certain conditions. Although this fusogenic property of bilayers has been known for some time, it is has become clear only relatively recently that two phospholipid bilayers will fuse spontaneously, owing to a hydrophobic force, when the bilayers are brought close together under conditions of membrane tension or high curvature (Helm and Israelachvili, 1993). The primeval cell would have used proteins to develop the appropriate architecture in which such fusion would occur in a regulated manner. During the course of evolution, ever more sophisticated ways of regulating this basic process would evolve, but the underlying fusion mechanism would remain unchanged. We have proposed that a macromolecular scaffold of proteins is responsible for bringing the plasma membrane close to the secretory granule membranes and creating the architecture that enables the hydrophobic force to cause fusion (Figure 1; Nanavati et al., 1992; Monck and Fernandez, 1992; Oberhauser and Fernandez, 1993). Evidence is now accumulating that there are several highly conserved families of proteins associated with vesicle fusion events, from yeast to mammalian cells, and with intracellular traffic, as well as with regulated exocytosis and synaptic transmission (Bennett and Scheller, 1993; Sollner et al., 1993; Südhof et al., 1993). The molecular structures (or scaffolds) that regulate membrane fusion are likely to contain related proteins and share certain fundamental properties.

Rab proteins in regulated exocytosis

Regulated exocytosis is responsible for neuronal communication, hormone secretion, food digestion, control of glucose uptake and many other basic processes. Despite the structural and functional diversity of the cells undergoing regulated exocytosis, all regulated exocytosis involves specialized vesicles that are stored in the cytoplasm and fuse with the plasma membrane in response to a trigger event. Recent evidence suggests that a subset of small GTP-binding proteins, Rab3 and its relatives, participate in the control of regulated exocytosis.

RT-PCR cloning of Rab3 isoforms expressed in peritoneal mast cells

Rab proteins are ras-like low molecular mass GTP-binding proteins, which are postulated to act as specific regulators of membrane trafficking in exocytosis and endocytosis. We have previously shown that synthetic peptides, corresponding to the effector domain of Rab3 proteins, stimulate a complete exocytotic response in mast cells. We have used a PCR-cloning strategy to investigate the presence of mRNA encoding Rab3 in mast cells. RNA based PCR was then performed on mast cell RNA using degenerate oligonucleotide primers based on two conserved sequences among Rab3 proteins. However, no PCR products were obtained, even for proteins known to be expressed in high copy numbers in mast cells (beta-actin and Fc receptor). We have found that the problem resides in the presence of mast cell secretory granule derived heparin, that copurifies with the RNA; heparin has been shown to inhibit the activity of reverse transcriptase and Taq polymerase in PCR. After treating the RNA (obtained from about 500 mast cells) with heparinase, several PCR products of varying size were obtained using primers specific for Rab3 proteins. These products were cloned and sequenced. We have found clones containing sequences that had a 100% homology at the deduced amino acid level to a portion of Rab3B and Rab3D (amino acids 16 to 83).

GTP-binding proteins: necessary components of the presynaptic terminal for synaptic transmission and its modulation

Using synapses that form between somata of Helisoma neurons in cell culture, we have studied the presynaptic regulation of synaptic transmission. Guanosine 5'-triphosphate (GTP)-binding proteins play critical roles in regulating synaptic transmission. Injection of guanine nucleotide analogues has demonstrated that one or more GTP-binding protein is necessary for transmitter release. Heterotrimeric G proteins continuously regulate the amount of transmitter released at the synapse by modulating potassium and calcium channels, and by controlling the secretory response to calcium. Perturbations of the synapse using guanosine 5'-diphosphate (GDP) beta S, GTP gamma S, and rab effector domain peptides suggest that small GTP-binding proteins also play critical roles in the synapse. We discuss the possibility that rab3, or related proteins, are required for exocytosis, and by cooperating with other proteins maintain vesicles in a docked state in the synapse.