Rab3 is a monomeric GTP-binding protein associated with secretory vesicles which has been implicated in the control of regulated exocytosis. We have exploited Rab3 mutant proteins to investigate the function of Rab3 in the process of neurotransmitter release from Aplysia neurons. A GTPase-deficient Rab3 mutant protein was found to inhibit acetylcholine release suggesting that GTP hydrolysis by Rab3 is rate-limiting in the exocytosis process. This effect was abolished by a mutation in the effector domain, and required the association of Rab3 with membranes. In order to determine the step at which Rab3 interferes with the secretory process, tetanus and botulinum type A neurotoxins were applied to Aplysia neurons pre-injected with the GTPase-deficient Rab3 mutant protein. These neurotoxins are Zn2+-dependent proteases that cleave VAMP/synaptobrevin and SNAP-25, two proteins which can form a ternary complex (termed the SNARE complex) with syntaxin and have been implicated in the docking of synaptic vesicles at the plasma membrane. The onset of toxin-induced inhibition of neurotransmitter release was strongly delayed in these cells, indicating that the mutant Rab3 protein led to the accumulation of a toxin-insensitive component of release. Since tetanus and botulinum type A neurotoxins cannot attack their targets, VAMP/synaptobrevin and SNAP-25, when the latter are engaged in the SNARE complex, we propose that Rab3 modulates the activity of the fusion machinery by controlling the formation or the stability of the SNARE complex.

The convergence of genetic and biochemical studies has revealed that the mechanisms underlying constitutive fusion of carrier vesicles with target membranes are conserved from yeast to mammals (Bennett and Scheller, 1993; Ferro-Novick and Jahn, 1994; Rothman, 1994). A similar fusion machinery seems to operate in the tightly regulated process of neurotransmitter release from nerve terminals. However, in neurons additional components prevent fusion from occuring unless it is triggered by an appropriate stimulus (i.e. Ca2+ influx). Among the proteins that have been implicated in the regulation of intracellular membrane transport are the members of the Rab/Ypt/Sec4 family of monomeric GTP-binding proteins. Rab proteins are either found in the cytosol, where they are bound to GDP and complexed to Rab-guanine nucleotide dissociation inhibitor (Rab-GDI), or in association with membranes. Consistent with their effect on the regulation of defined transport steps, Rab proteins are specifically associated with distinct membrane compartments (Simons and Zerial, 1993; Lledo et al., 1994). Interaction of Rab proteins with the donor compartment is followed by release of Rab-GDI into the cytosol and by guanine nucleotide exchange (Pfeffer, 1994).

GTP-bound Rab is then thought to interact with an effector molecule to control vesicle transport, docking or fusion. Following GTP hydrolysis, catalyzed by a GTPase-activating protein (GAP), GDP-bound Rab is solubilized by Rab-GDI and enters a new functional cycle.

Rab3a is a member of the Rab family that is mainly expressed in neurons and neuroendocrine cells where it is localized on secretory vesicles (Darchen et al., 1990, 1995; Fischer von Mollard, 1990). Recently, a role for Rab3a in regulated exocytosis has been demonstrated (Geppert et al., 1994; Holz et al., 1994; Johannes et al., 1994). In bovine chromaffin cells, Rab3a acts as an inhibitory modulator of secretion (Holz et al., 1994; Johannes et al., 1994). However, the mechanism of action of Rab3a and the identity of the step controlled by this GTPase remain to be determined.

Another set of proteins implicated in regulated exocytosis involves vesicle-associated membrane protein (VAMP; also termed synaptobrevin), an integral membrane protein from several classes of secretory vesicles, syntaxin, an integral plasma membrane protein, and SNAP-25 (synaptosomal associated protein of 25 kDa), which is palmitoylated and located at the plasma membrane. It was found that syntaxin, SNAP-25, and VAMP bind tightly to each other (Söllner et al., 1993a,b). The resulting complex is very stable, even in the presence of SDS (Hayashi et al., 1994; Pellegrini et al., 1994, 1995). The complex is also known as the SNAP receptor (abbreviated as SNARE complex) since it binds the N-ethyl-maleimide-sensitive fusion protein (NSF) and the soluble NSF attachment proteins (SNAPs) to form a 20 S particle. NSF and SNAPs were identified by Rothman and colleagues as soluble factors necessary for intracellular membrane fusion (Rothman, 1994). Their homologs in yeast are essential genes involved in vesicular transport along the secretory pathway (Rothman, 1994). NSF is an ATPase, and upon ATP hydrolysis, NSF catalyzes a conformational change of syntaxin, which in turn leads to the disassembly of the 20 S complex (Hanson et al., 1995). This event should precede membrane fusion, because the assembled SNARE complex accumulates in yeast secretory mutants deficient in SEC18 (NSF) activity (Søgaard et al., 1994). The exact function performed by the SNARE complex is not yet clear. However, since it contains membrane proteins from both the carrier vesicle (v-SNAREs) and the target membrane (t-SNAREs), it has been proposed that selectivity in pairing between v- and t-SNAREs may provide specificity for the interaction of a transport vesicle with its target membrane (Rothman, 1994; Rothman and Warren, 1994). According to the SNARE hypothesis, docking and fusion would proceed as a consequence of sequential interactions between v-SNAREs, t-SNAREs, SNAPs and NSF.

Tetanus toxin (TeNT) and seven related botulinum neurotoxins (BoNT/A-G) are highly potent inhibitors of neurotransmitter release. These neurotoxins were recently identified as zinc metallopeptidases specific for the SNAREs. VAMP/synaptobrevin is attacked by TeNT and BoNT/B, D, F, and G; syntaxin is cleaved by BoNT/C; SNAP-25 is a substrate for BoNTs A and E (for review, see Nieman et al., 1994; Montecucco and Schiavo, 1995). The in vitro cleavage of synaptic proteins correlates with the inhibition of neurotransmitter release induced by these toxins. There is also a clear correlation between toxin-induced modification of quantal release, the ability of aminopyridines to counteract toxin effects, and the identity of the toxins’ targets (Poulain et al., 1995). In addition, the inhibitory effect of TeNT or BoNT/B is prevented by peptides spanning the cleavage site or the toxin-binding domain of VAMP and by anti-VAMP antibodies, indicating that the toxin-induced blockade of neurotransmission is mostly due to cleavage of VAMP (Schiavo et al., 1992a; Poulain et al., 1993; Rossetto et al., 1994). Altogether, these findings provide strong evidence that SNARE proteins play an important role in synaptic vesicle exocytosis.

To date, little is known about the functional relationship between Rab proteins and the SNARE complex. Although Rab proteins do not seem to be core components of the SNARE complex, genetic evidence supports a role for the Rab proteins in the assembly of this complex (Lian et al., 1994; Søgaard et al., 1994). With the aim of elucidating mechanistic aspects of Rab3 function and of investigating a possible link between Rab3 proteins and SNAREs, we have characterized the effect of Rab3 mutants on neurotransmitter release. Identified Aplysia neurons making well characterized synapses were used. Indeed, this model system allows an easy access to intracellular compartment and a real-time monitoring of the secretory activity following single or sequential intracellular application of different molecular tools, a procedure that is not readily feasible in mammalian systems (Schiavo et al., 1992a; Poulain et al., 1993; Rossetto et al., 1994). Here, we report that a GTPase-deficient Rab3 mutant inhibits acetylcholine (ACh) release from nerve terminals in Aplysia. Furthermore, this Rab3 mutant delays the onset of action of TeNT and BoNT/A, indicating that VAMP and SNAP-25 are maintained in a toxin-insensitive state that most probably corresponds to the SNARE complex.

Electrical recordings

Experiments were performed at 22°C at identified cholinergic synapses in the buccal ganglion of the marine sea slug Aplysia californica (Marinus Inc., CA, USA). In this ganglion, two presynaptic neurons, B4 and B5, make well-known chloride-dependent inhibitory synapses with the same post-synaptic cells as B3, B6, and others. In order to prevent the bursting activity which characterizes presynaptic neurons in the buccal ganglion, dissected ganglia were superfused continuously (10 ml/hour) with a physiological medium containing an enriched concentration of Ca2+ and Mg2+ ions (460 mM NaCl, 10 mM KCl, 33 mM CaCl2, 50 mM MgCl2, 28 mM MgSO4, 10 mM Tris-HCl, pH 7.8). During the experiments, both presynaptic cells (125-180 μm in diameter) and one postsynaptic neuron, either B3 or B6 (150-250 μm in diameter) were impaled with glass microelectrodes (3 M KCl Ag/AgCl2, 4-10 MegaOhms). The two presynaptic neurons were depolarized alternately to elicit an action potential every 60 seconds. The ensuing postsynaptic response was recorded in the post-synaptic neuron as a current change using a conventional two-electrode voltage-clamp technique. Subsequently, by taking into account its reversal potential, i.e. the null potential for Cl, the post-synaptic response was expressed as an apparent membrane conductance. This value is directly proportional to the amount of ACh released per impulse (for details, see Poulain et al., 1993, 1996).

Neuronal application of Rab proteins or toxins

TeNT and BoNT/A (two chain form), highly purified from Clostridium tetani and Clostridium botulinum (Schiavo et al., 1992b,c), were a kind gift from C. Montecucco. Prior to intracellular administration, Rab or TeNT samples were mixed with a dye (10% v/v; fast green FCF, Sigma Chem. Co.) and air-pressure injected under visual and electrophysiological monitoring by means of a third microelectrode impaled into the presynaptic cell body. After injection, this micropipette was removed. For extracellular application of BoNT/A, the superfusion was stopped and the toxin added to the bath. After 20 minutes, superfusion was resumed and all unbound toxin washed away. Extracellular or intracellular administration of buffer solutions used to prepare Rab proteins or toxins produced no change in transmitter release.

Plasmids

Comparison of human Rab3a and Aplysia Rab3 sequences showed a high degree of homology except in the C-terminal domain. The 24 last amino acids of human Rab3a were replaced by the corresponding residues of Aplysia californica Rab3 using a PCR-based strategy (Johannes et al., 1994). Briefly, the human tagged rab3a cDNA (Zahraoui et al., 1989; Johannes et al., 1994) in pGEM-4Z (pGEM-TRab3a) and the synthetic oligonucleotides (5′-3′) Nhe12 (GGCCGGCAGCTAGCTGACCACCT), Apl1 (GGGTTCTCGG- TGAGCCGCGTCGACTTGGTCGTGTTGTTGACCAGGGTAGG- GTCCGCCGTGTCCAA), Apl2 (CGGCTCACCGAGAACCCC- AACATGAACAGCTCTGGCTGCTCCTGCTGAGAGCCATCCC- ACTCC), and Ext9 (ATTGAATTCGT GTAGTGTTGTCAT), were used to generate a NheI-EcoRI fragment which was inserted into the corresponding restriction sites of pGEM-TRab3a, yielding pGEM-hu/apRab3. A similar approach was followed to delete the 3 last amino acids of hu/apRab3 carrying the isoprenylation motif. The following oligonucleotides were used: Nhe12, ΔC1 (GTGGGATGGCTCTCATCAGGATCAGCCAGAGCTGTTCAT- GTT), ΔC2 (AACATGAACAGCTCTGGCTGATCCTGATGAGA- GCCATCCCAC), and Hind14 (AGTACAAGCTTTGACATCTCC- TAAGG). The mutations T36N, V55E, Q81L, and N135I were introduced into pGEM-hu/apRab3 or pGEM-hu/apRab3aΔC by exchanging the BamHI-NheI fragment of this plasmid with the corresponding fragments of pGEM-TRab3aT36N, pGEM-TRab3aV55E, pGEM-TRab3aQ81L, and pGEM-TRab3aN135I, respectively (Johannes et al., 1994). To construct V55E-containing double mutants the PstI-NheI fragment of pGEM-hu/apRab3V55E was replaced by those of pGEM-hu/apRab3Q81L or pGEM-hu/apRab3N135I. For expression as fusion proteins, the BamHI-SstII fragment of each construct was inserted into the corresponding restrictions sites of pGEX-TRab3a (Johannes et al., 1994).

The cDNA encoding Aplysia Rab3 (apRab3) (GenBank database, accession number U00986) was kindly provided by Y. Hu and R. Baston from Dr Kandel’s laboratory (Columbia University, New York, USA). The coding sequence was cloned via PCR in frame to the COOH terminus of GST using the pGEX-4T1 system (Pharmacia Biotech) and the oligonucleotides ApR3A (5′-GCATAGAATT- CGCTTCCGCAAACGACT-3′) and ApR3B (5′-CTAGGTCTC- GAGTAACGAACGACCCTC-3′). To introduce the Q80L mutation into apRab3, ApR3A and ApR3C (5′-GGTCGGGTACCGCTCC- AGGCCAGC-3′) were used to generate a EcoRI-KpnI PCR fragment that was inserted into the corresponding restriction sites of pGEX-apRab3.

The PCR-amplified regions were verified by dideoxy sequencing (US Biochemicals). Recombinant proteins were expressed in E. coli and purified onto glutathione-Sepharose (Pharmacia Biotech), as described previously (Johannes et al., 1994).

In vitro geranylgeranylation

Geranylgeranyl transferase (GGTase) II activity was determined essentially as described (Cremers et al., 1994). Briefly, each reaction contained in 50 μl: 2 μM of hu/apRab3 or mutants, 0.5 μM [3H]ger-anylgeranyl pyrophosphate (14.8 Ci/mmol, Amersham), 10 μM GDP, 5 mM MgCl2, 1 mM Nonidet-P40, 50 mM Hepes-NaOH pH 7.2, 1 mM DTT, 50 ng each of GGTase II components A and B. The reactions were done in duplicate and incubated at 37°C for the indicated time. At the end of the incubation period, the amount of ethanol/HCl-precipitable radioactivity, corresponding to [3H]geranyl-geranyl transferred to Rab3 proteins was determined by filtration through a glass fiber filter.

Determination of kinetic constants

Binding of [3H]GDP (10 Ci/mmol, Amersham) to purified Rab3 proteins (20 nM) was performed in buffer A (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.05% bovine serum albumin) containing 5 mM EDTA and 100 nM labelled nucleotide. After 15 minutes at 30°C, 1 mM MgCl2 was added and the reaction mixture was further incubated at 30°C for 15 minutes. The labelled complex was diluted 10 times in buffer A containing 10 mM MgCl2 and 50 μM unlabelled GDP. Aliquots (1 ml) of the reaction mix were withdrawn at various intervals, added to 2 ml of ice-cold washing buffer (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 10 mM MgCl2), and filtered through 0.45 μm nitrocellulose filters. The filters were washed twice with the same buffer. Radioactivity remaining on the filters was a measure of bound nucleotide. The natural log of radioactivity bound at time t/radioactivity bound at time zero was plotted versus time and the slopes determined by least square analysis to determine the first-order dissociation rate constants.

GTP-bound Rab3 inhibits ACh release from Aplysia neurons

In neuroendocrine cells, exocytosis of large dense core vesicles is controlled by the monomeric GTPase Rab3 (Lledo et al., 1993; Holz et al., 1994; Johannes et al., 1994). In order to study the function of Rab3 in the neurotransmitter release process, we used identified neurons of the buccal ganglion of Aplysia californica. This model system allows intraneuronal application of recombinant proteins and concomitant electrophysiological quantification of ACh release from injected cells via the measurement of the amplitude of evoked post-synaptic responses (Schiavo et al., 1992a; Poulain et al., 1993; Rossetto et al., 1994). In contrast to mammals where 4 isoforms of rab3 are known, only one rab3 gene has been identified in Aplysia californica (GenBank database accession number U00986). Preliminary experiments indicated that human Rab3a proteins were without any effect on ACh release from Aplysia neurons (see below). Comparison of human and Aplysia Rab3 sequences showed that positions 1 to 193 are highly conserved. However the C-terminal domain is divergent (Fig. 1). This region, known as the hypervariable domain, has been described as a targeting signal responsible for the specific subcellular localization of the Rab proteins (Chavrier et al., 1991; Stenmark et al., 1994). Since the lack of effect of human Rab3a proteins in Aplysia might be due to mistargeting, we exchanged the 24 last amino acids of human Rab3a by the corresponding residues of Aplysia Rab3. The chimeric protein was referred to as hu/apRab3. Previously, we and others found that mutants of Rab3a either defective in guanine nucleotide binding (Rab3aN135I) or in GTPase activity (Rab3aQ81L) inhibited regulated exocytosis of secretory granules in adrenal chromaffin cells (Holz et al., 1994; Johannes et al., 1994). The same mutations were introduced into the sequence of hu/apRab3. The respective recombinant proteins hu/apRab3Q81L and hu/apRab3N135I were expressed in E. coli as glutathione S-transferase (GST) fusion proteins and purified on glutathione-Sepharose. A c-myc tag was also present at the N terminus of hu/apRab3 proteins.

Fig. 1.

Alignment of human Rab3a and Aplysia Rab3 sequences. The overall identity between human Rab3a (huRab3a, accession number J04941) and Aplysia Rab3 (apRab3, Genbank accession number U00986) is 77%, and up to 86% from residues 1 to 192. The C-terminal domain is highly divergent. Identical amino acids are blocked in black and conservative substitutions in grey. The chimeric hu/apRab3 used in this study comprises the main part of huRab3a (residues 1-196), and the 24 last amino acids of apRab3 (from the arrowhead).

Fig. 1.

Alignment of human Rab3a and Aplysia Rab3 sequences. The overall identity between human Rab3a (huRab3a, accession number J04941) and Aplysia Rab3 (apRab3, Genbank accession number U00986) is 77%, and up to 86% from residues 1 to 192. The C-terminal domain is highly divergent. Identical amino acids are blocked in black and conservative substitutions in grey. The chimeric hu/apRab3 used in this study comprises the main part of huRab3a (residues 1-196), and the 24 last amino acids of apRab3 (from the arrowhead).

Microinjection of purified recombinant hu/apRab3, hu/apRab3N135I (Table 1), or hu/apRab3Q81L (Fig. 2 and Table 1) into the cell body of identified presynaptic neurons produced an inhibition of ACh release as detected by a decreased amplitude of the postsynaptic response. Similar results were obtained after separation of GST and Rab3 moieties by thrombin cleavage (Fig. 3A). On a molar basis, at a final intrasomatic concentration of ∼0.4 μM, the mean inhibitory effect of hu/apRab3Q81L (46.4±10%, n=13, mean ± s.d.) was higher than that of hu/apRab3 (16.2±16.6%, n=4). The scattering of the data, especially with hu/apRab3, was greater than generally observed in this system with other agents that affect the secretory activity, suggesting that the variability found in these experiments might reflect differences in the ratio of endogenous/recombinant Rab3 or in the activity of interacting partners.

Table 1.

Recombinant Rab3 proteins in Aplysia neurons

Recombinant Rab3 proteins in Aplysia neurons
Recombinant Rab3 proteins in Aplysia neurons
Fig. 2.

Effect of hu/apRab3 mutants on ACh release from Aplysia neurons. Neurotransmitter release was evoked at identified cholinergic synapses in the buccal ganglion of Aplysia and monitored following microinjection (at time zero) of different Rab3 proteins into the cell body of a presynaptic (either B4 or B5) neuron. (A) hu/apRab3Q81L; (B) hu/apRab3T36N; (C) hu/apRab3(V55E-Q81L); (D) hu/apRab3V55E. Results are given as a percentage of the mean release measured before injection. Typical experiments are shown. In A, the response of another presynaptic neuron making a synapse with the same postsynaptic cell and injected with control buffer is shown (open circles). Note that neurotransmission is very stable throughout the experiment. The mutation V55E in the effector domain (in C and D) not only abolished the inhibitory effect of hu/apRab3 and hu/apRab3Q81L, but also potentiated the postsynaptic response.

Fig. 2.

Effect of hu/apRab3 mutants on ACh release from Aplysia neurons. Neurotransmitter release was evoked at identified cholinergic synapses in the buccal ganglion of Aplysia and monitored following microinjection (at time zero) of different Rab3 proteins into the cell body of a presynaptic (either B4 or B5) neuron. (A) hu/apRab3Q81L; (B) hu/apRab3T36N; (C) hu/apRab3(V55E-Q81L); (D) hu/apRab3V55E. Results are given as a percentage of the mean release measured before injection. Typical experiments are shown. In A, the response of another presynaptic neuron making a synapse with the same postsynaptic cell and injected with control buffer is shown (open circles). Note that neurotransmission is very stable throughout the experiment. The mutation V55E in the effector domain (in C and D) not only abolished the inhibitory effect of hu/apRab3 and hu/apRab3Q81L, but also potentiated the postsynaptic response.

Fig. 3.

Importance of the C-terminal region for the activity of Rab3 proteins. The experiments were done as described in Fig. 2. In A, apRab3Q80L (▫) and hu/apRab3Q81L (▪) were injected after thrombin cleavage of the GST moiety. They have similar effects on ACh release suggesting that the sequence divergence other than the C terminus between these two Rab3 proteins is not critical for Rab3 function. In B, the effect of hu/apRab3Q81L (+) on ACh release is compared to that of hu/apRab3(Q81L-ΔC) (•) or huRab3Q81L (∘). The lack of isoprenylation motif reduces the activity of hu/apRab3(Q81L-ΔC), while the full-length human Rab3Q81L does not change neurotransmission. C, hu/apRab3V55E (∘) potentiates ACh release (as in Fig. 2C), in contrast to hu/apRab3(V55E-ΔC) lacking the isoprenylation motif which has no effect. Calculated respective final intrasomatic concentrations are A: 0.25 μM; B: huRab3Q81L, 0.4 μM, hu/apRab3Q81L, 0.4 μM; hu/apRab3(Q81L-ΔC), 0.8 μM and C: hu/apRab3V55E 0.6 μM; hu/apRab3(V55E-ΔC) 3.2 μM.

Fig. 3.

Importance of the C-terminal region for the activity of Rab3 proteins. The experiments were done as described in Fig. 2. In A, apRab3Q80L (▫) and hu/apRab3Q81L (▪) were injected after thrombin cleavage of the GST moiety. They have similar effects on ACh release suggesting that the sequence divergence other than the C terminus between these two Rab3 proteins is not critical for Rab3 function. In B, the effect of hu/apRab3Q81L (+) on ACh release is compared to that of hu/apRab3(Q81L-ΔC) (•) or huRab3Q81L (∘). The lack of isoprenylation motif reduces the activity of hu/apRab3(Q81L-ΔC), while the full-length human Rab3Q81L does not change neurotransmission. C, hu/apRab3V55E (∘) potentiates ACh release (as in Fig. 2C), in contrast to hu/apRab3(V55E-ΔC) lacking the isoprenylation motif which has no effect. Calculated respective final intrasomatic concentrations are A: 0.25 μM; B: huRab3Q81L, 0.4 μM, hu/apRab3Q81L, 0.4 μM; hu/apRab3(Q81L-ΔC), 0.8 μM and C: hu/apRab3V55E 0.6 μM; hu/apRab3(V55E-ΔC) 3.2 μM.

The Q81L mutation strongly reduces the intrinsic GTPase activity of Rab3a (Brondyk et al., 1993). Thus, the marked inhibitory effect of hu/apRab3Q81L suggests that it is the GTP-bound form of hu/apRab3 that negatively modulates ACh release. In order to address this question further, we created the mutant hu/apRab3T36N. Rab3aT36N has a very low affinity for GTP (Burstein and Macara, 1992). Although its affinity for GDP is also reduced, the protein is thought to prevail in the GDP-bound state. Microinjection of hu/apRab3T36N (∼0.2 μM, final intrasomatic concentration) had no significant effect on the postsynaptic response (Fig. 2B and Table 1). These results suggest that intracellular factors which interact with GDP-bound hu/apRab3 are not limiting under our experimental conditions. Furthermore, they strongly suggest that Rab3 activity may become rate-limiting in the secretory process if the rate of GTP hydrolysis is reduced.

To test the validity of human/Aplysia chimeric Rab3 proteins as probes of Rab3 function in Aplysia neurons, we have analyzed the effect of Aplysia Rab3 carrying a mutation equivalent to Q81L in human Rab3a and referred to as apRab3Q80L The effect of this mutant on ACh release in this series of experiments was very similar to that of hu/apRab3Q81L (43±16%, n=7, mean ± s.d.). Similar results were obtained after cleavage by thrombin (39.5±13%, n=4, mean ± s.d.) (Fig. 3A). Wild-type apRab3 had a slight inhibitory effect on ACh release comparable to that of hu/apRab3 (data not shown). The observation that hu/apRab3Q81L and apRab3Q80L both inhibit ACh release indicates that the sequence divergence other than the C terminus between these two Rab proteins is not critical for Rab3 function.

Suppression of Rab3 activity by a mutation in the effector domain

We have previously shown that the inhibitory effect of over-expressed Rab3a on the secretory activity of PC12 cells is neutralized by a mutation in the loop L2 (Johannes et al., 1994), also called effector loop since it is supposed to mediate interaction with target proteins. Consistent with these results, the same mutation, V55E, also abolished the strong inhibitory effect of the mutants Q81L and N135I in PC12 cells (L. Johannes and F. Darchen, unpublished observations). The double mutant hu/apRab3(V55E-Q81L) was injected into Aplysia neurons (0.25 μM final intrasomatic concentration). This did not inhibit ACh release and even potentiated it by 10.3% (±3.5%, n= 4; Fig. 2C and Table 1), indicating that the effector domain mutation prevents inhibition by hu/apRab3Q81L. ACh release was also enhanced after injection of a similar amount of hu/apRab3V55E (+17.1±12.7%, n=9, mean ± s.d., 3 hours after injection; Figs 2D, 3C and Table 1). In both cases, the effect developed after a delay of 54 minutes (±12 minutes, n=13). This delay was larger than the one observed with hu/apRab3Q81L (11±5 minutes, n=13). Seemingly, effector loop mutants take longer to become competent to enter the functional cycle. In agreement with this interpretation, we found that hu/apRab3V55E is less efficiently geranylgeranylated than wild-type hu/apRab3 protein (Fig. 4). The V55E mutants thus probably reach the membrane of synaptic vesicles more slowly than the wild-type protein. However, the V55E mutation does not change the kinetics of GDP dissociation (Table 1), suggesting that the reduced interaction with geranylgeranyl transferase II or with effector molecules such as Rabphilin-3A (McKiernan et al., 1993) is due to a specific modification of the effector domain rather than to some type of unspecific structural alteration.

Fig. 4.

Post-translational modification of hu/apRab3 and mutants. The mutant hu/apRab3V55E (•) is far less efficiently modified by geranylgeranyl transferase II than wild-type hu/apRab3 (∘). The mutant is nevertheless a substrate for GGTase II activity whereas hu/apRab3ΔC, lacking the isoprenylation motif, is not modified at all (▵). Reactions were done in duplicate. Similar results were obtained in two other experiments.

Fig. 4.

Post-translational modification of hu/apRab3 and mutants. The mutant hu/apRab3V55E (•) is far less efficiently modified by geranylgeranyl transferase II than wild-type hu/apRab3 (∘). The mutant is nevertheless a substrate for GGTase II activity whereas hu/apRab3ΔC, lacking the isoprenylation motif, is not modified at all (▵). Reactions were done in duplicate. Similar results were obtained in two other experiments.

The potentiation of neurotransmitter release induced by the V55E mutants suggests that these mutants can enter the functional cycle and compete with endogenous Rab3. Their inability to interact with a Rab3 effector, such as Rabphilin-3A (McKiernan et al., 1993), would result in a functional neutralization of the inhibitory activity of Rab3 in these cells. These data are thus in agreement with the enhancement of the secretory activity observed previously in chromaffin cells loaded with antisense oligodeoxynucleotides directed to rab3a mRNA (Johannes et al., 1994).

The C-terminal domain and isoprenylation contribute to Rab3 activity

The experiments presented above show that the chimeric Rab3 proteins used in this study are active in Aplysia neurons. In order to determine the importance of the C-terminal region for Rab3 activity, we compared the effect of chimeric hu/apRab3 proteins on ACh release to that of human Rab3a mutants. In contrast to the inhibitory effect of hu/apRab3N135I and hu/apRab3Q81L, neither huRab3aN135I nor huRab3aQ81L significantly changed ACh release (Fig. 3B). These results indicate that the C-terminal hypervariable domain carries functional determinants, probably implicated in the proper targeting of Rab3 to synaptic vesicles, that are critical for Rab3 activity. Another important feature of Rab proteins is the isoprenylation of two carboxyl terminus cysteine residues. This posttranslational processing, catalyzed by geranylgeranyl transferase II, is thought to stabilize membrane attachment of Rab proteins which otherwise lack hydrophobic domains (Musha et al., 1992). We addressed the question of the functional significance of this processing by deleting the last 3 amino acids of Rab3 including the two cysteines that function as acceptors of geranylgeranyl groups. As expected, the resulting purified ΔC proteins were not labelled by [3H]geranylgeranyl pyrophosphate in a reaction containing purified geranylgeranyl transferase II (Fig. 4). ΔC mutants are nevertheless stable, as indicated by their unchanged GDP-dissociation rates (Table 1). Deletion of the isoprenylation motif led to a 50% loss of hu/apRab3Q81L-mediated inhibition of ACh release (Fig. 3B and Table 1), and to a complete loss of hu/apRab3V55E-induced stimulation of neurotransmission (Fig. 3C and Table 1). These results indicate that isoprenylation is important for Rab3 function in neurotransmitter release. The lower activity of ΔC mutants is likely to be a consequence of a weaker association of Rab3 with synaptic vesicles. However, the fact that hu/apRab3(Q81L-ΔC) retains some activity suggests that isoprenylation is not absolutely required for Rab3 function. This has previously been observed for other members of the Rab family (Tisdale et al., 1992; Lombardi et al., 1993; Martinez et al., 1994).

Prior injection of a GTPase-deficient Rab3 mutant delays further inhibition of ACh release by clostridial neurotoxins

In order to gain further insights into the mechanism of action of Rab3 in exocytosis of neurotransmitter, we studied the possible interferences between the inhibitory Rab3 mutant hu/apRab3Q81L and the impairment of neurotransmission by clostridial toxins. These neurotoxins (TeNT and BoNTs A-G) consist of two disulfide-linked polypeptide chains, a light (L) chain and a heavy (H) chain. Binding, internalization, and translocation of the toxins into the cytosol of neuronal cells are mediated by the H chain while intracellular blockade of neurotransmitter release depends on the proteolytic activity of the L chain (Montecucco and Schiavo, 1995). At identified synapses in the buccal ganglion of Aplysia, a toxin-induced inhibitory effect becomes detectable after a short lag which depends on the toxin serotype used (∼5 minutes for BoNT/A; ∼15 minutes for TeNT) (Poulain et al., 1996). The lag is partially due to the time required for diffusion of the toxin from the cell body, where it is injected, to the nerve terminals. Inhibition of neurotransmitter release by TeNT is due to cleavage of VAMP. In order to determine whether Rab3 is acting before or after VAMP in the sequence of events that ultimately lead to neurotransmitter release, we injected sequentially hu/apRab3Q81L and TeNT. As shown above, hu/apRab3Q81L reduced the amplitude of the postsynaptic response. Then, after evoked ACh release had reached a new stable level, TeNT was injected intraneuronally (at 8 nM, final intrasomatic concentration). As shown in Fig. 5A, ACh release remained unchanged for 80±20 minutes (n=4; mean ± s.d.); then it dropped rapidly. This long delay contrasted with that observed with TeNT alone (delay of 14±3 minutes, n=6; mean ± s.d.). The increased latency between toxin injection and subsequent inhibition of ACh release was observed in all the experiments that were performed. A correlation was noted between the magnitude of the inhibitory effect of hu/apRab3Q81L and the duration of the latency period (Fig. 6). Importantly, this action appeared specific for hu/apRab3Q81L because pre-injection of hu/apRab3V55E had no effect on TeNT-induced inhibition of ACh release (Fig. 5C).

Fig. 5.

Prior injection of hu/apRab3Q81L delays the onset of TeNT- and BoNT/A-induced inhibition of ACh release. Representative experiments are shown. (A) A presynaptic cell of the buccal ganglion was sequentially injected with hu/apRab3Q81L (at time zero, 0.4 μM final concentration) and, after the postsynaptic response had reached a plateau, with TeNT (indicated by an arrow). The final intracellular concentration of the toxin was about 8 nM. Results are given as a percentage of control ACh release measured before Rab3Q81L injection. Further inhibition of ACh release was detectable only 80 minutes after TeNT injection. hu/apRab3Q81L (•) delays the onset of TeNT-induced inhibition of ACh release whereas control buffer (○) does not. Results are given as a percentage of the mean value measured before injection of TeNT, to allow direct comparison of the duration of the latency period preceding TeNT-induced inhibition of ACh release. hu/apRab3Q81L or control buffer were injected 100 minutes before TeNT. (C) Similar experiment as in A, showing the lack of effect of hu/apRab3V55E (0.4 μM final concentration) on the kinetics of TeNT-induced blockade of ACh release. (D) hu/apRab3Q81L (▴) or control buffer (▵) were injected into a presynaptic cell followed, 120 minutes later, by BoNT/A (7 nM for 20 minutes) application in the bath at time zero (indicated as a hatched area). Results are given as a percentage of ACh release measured before toxin application as in B.

Fig. 5.

Prior injection of hu/apRab3Q81L delays the onset of TeNT- and BoNT/A-induced inhibition of ACh release. Representative experiments are shown. (A) A presynaptic cell of the buccal ganglion was sequentially injected with hu/apRab3Q81L (at time zero, 0.4 μM final concentration) and, after the postsynaptic response had reached a plateau, with TeNT (indicated by an arrow). The final intracellular concentration of the toxin was about 8 nM. Results are given as a percentage of control ACh release measured before Rab3Q81L injection. Further inhibition of ACh release was detectable only 80 minutes after TeNT injection. hu/apRab3Q81L (•) delays the onset of TeNT-induced inhibition of ACh release whereas control buffer (○) does not. Results are given as a percentage of the mean value measured before injection of TeNT, to allow direct comparison of the duration of the latency period preceding TeNT-induced inhibition of ACh release. hu/apRab3Q81L or control buffer were injected 100 minutes before TeNT. (C) Similar experiment as in A, showing the lack of effect of hu/apRab3V55E (0.4 μM final concentration) on the kinetics of TeNT-induced blockade of ACh release. (D) hu/apRab3Q81L (▴) or control buffer (▵) were injected into a presynaptic cell followed, 120 minutes later, by BoNT/A (7 nM for 20 minutes) application in the bath at time zero (indicated as a hatched area). Results are given as a percentage of ACh release measured before toxin application as in B.

Fig. 6.

Correlation between the inhibitory effect (in % of control) of hu/apRab3Q81L on ACh release and the duration of the latency period preceding TeNT- (▪) and BoNT/A- (▫) induced inhibition of the postsynaptic response. The data were determined in experiments similar to those described in Fig. 5. Linear regression analysis gave y=1.61x+15.01; r=0.82 for TeNT (solid line) and y=1.46x+4.33; r=0.96 for BoNT/A (broken line).

Fig. 6.

Correlation between the inhibitory effect (in % of control) of hu/apRab3Q81L on ACh release and the duration of the latency period preceding TeNT- (▪) and BoNT/A- (▫) induced inhibition of the postsynaptic response. The data were determined in experiments similar to those described in Fig. 5. Linear regression analysis gave y=1.61x+15.01; r=0.82 for TeNT (solid line) and y=1.46x+4.33; r=0.96 for BoNT/A (broken line).

The increased latency period between TeNT injection and inhibition of ACh release suggests that hu/apRab3Q81L prevents free access of TeNT to its target. This might be due either to a direct interaction between Rab3 and VAMP or to a Rab3-induced stabilization of a multimolecular complex involving VAMP, in which VAMP would be protected from proteolytic cleavage by TeNT. Interestingly, VAMP, syntaxin, and SNAP-25 were found to be toxin-insensitive once assembled into the SNARE complex (Hayashi et al., 1994; Pellegrini et al., 1994, 1995). Thus, it is tempting to speculate that hu/apRab3Q81L promotes the accumulation of TeNT-insensitive SNARE complexes. However, VAMP was also found to interact with synaptophysin (Calakos and Scheller, 1994; Edelman et al., 1995; Washbourne et al., 1995) and with VAP-33, a newly discovered synaptic protein of Aplysia (Skehel et al., 1995). In the latter case, it is not clear whether synaptophysin or VAP-33 is able to protect VAMP from TeNT attack or not. In order to discriminate between these different possibilities, we determined whether hu/apRab3Q81L could also protect another component of the SNARE complex. Since BoNT/C is poorly active in Aplysia (B. Poulain, unpublished observations) we focused on BoNT/A that specifically cleaves SNAP-25. The same experimental procedure as for TeNT was followed. Neurons were injected with hu/apRab3Q81L, and after stabilization of the postsynaptic response, BoNT/A (7 nM) was applied to the extracellular medium. As illustrated in Figs 5D and 6, inhibition of ACh release caused by BoNT/A was significantly delayed in these cells (63 minutes ± 21, n=4; mean ± s.d.) as compared to control cells (5 minutes ± 1, n=7). Thus, hu/apRab3Q81L reduced the efficiency of cleavage of the t-SNARE/SNAP-25 as well as that of the v-SNARE/VAMP by clostridial neurotoxins.

Rab3, a rate-limiting factor for exocytosis

In the present study, recombinant Rab3 proteins mutated in several domains were introduced intracellularly with the aim of gaining insight into the role of Rab3 in neurosecretion. We found that the GTPase-deficient mutants apRab3Q80L and hu/apRab3Q81L and, to a lesser extent, apRab3 and hu/apRab3 inhibited ACh release from Aplysia neurons, whereas the preferentially GDP-bound mutant hu/apRab3T36N had no effect. These results together with the observed stimulation of ACh release induced by proteins carrying a mutation in the effector loop (V55E mutants) and the previously reported increase in secretory activity of chromaffin cells injected with antisense oligodeoxynucleotides directed to rab3a mRNA (Johannes et al., 1994) suggest a negative modulatory role for Rab3 in neurotransmitter release. This contention is based on the assumption that the recombinant proteins enter the functional cycle of endogenous Rab3. Several observations support this view: (i) in agreement with earlier studies demonstrating the role of the C-terminal hypervariable domain in the targeting of Rab proteins to the proper intracellular membrane compartment (Chavrier et al., 1991; Stenmark et al., 1994), the presence of the 24 C-terminal residues specific for Aplysia Rab3 was required for the various Rab3 constructs to be active in Aplysia neurons. Moreover, deletion of the isoprenylation motif of hu/apRab3 proteins strongly reduced or abolished their effect. (ii) The mutation V55E in the effector loop, which is supposed to block the interaction of Rab proteins with their effectors suppresses the inhibitory effect of hu/apRab3Q81L. In addition, the V55E mutation not only abolished the activity of recombinant Rab3 proteins, as would be expected if the latter were acting by sequestration of a co-factor, but even enhanced ACh release. (iii) Rab3T36N, which is preferentially bound to GDP, has no inhibitory effect on exocytosis at Aplysia synapses. This observation argues against the titration of a limiting component interacting with GDP-bound Rab3 such as GRF or RabGDI.

The inhibitory effect of the GTPase-deficient mutants indicates that GTP-hydrolysis by Rab3 can be rate-limiting in the exocytosis of neurotransmitter as proposed previously for the process of catecholamine secretion by chromaffin cells (Holz et al., 1994; Johannes et al., 1994). It is tempting to speculate that the regulation of the GTPase activity of Rab3 represents a physiological means to control neurotransmitter release. In the case of YPT1, it has also been suggested that GTP hydrolysis can become limiting for the fusion of ER-derived transport vesicles with Golgi membranes, as demonstrated by means of a non-hydrolysable GTP analogue in an in vitro assay (Lupashin et al., 1996).

A functional link between Rab3 and the SNAREs

The onset of TeNT- and BoNT/A-induced inhibition of ACh release from Aplysia neurons is significantly delayed by prior injection of the GTPase-deficient mutant hu/apRab3Q81L, whereas injection of hu/apRab3V55E carrying a mutation in the effector domain has no effect on the neurotoxins’ action. While the amplitude of the secretory response is decreased by about 50% after injection of hu/apRab3Q81L, the latency period is increased by a factor of 5 or 12 for TeNT and BoNT/A, respectively. The duration of the latency period correlates with the extent of inhibition of neurotransmitter release induced by hu/apRab3Q81L. It is unlikely that the delay preceding the toxin-induced inhibition of ACh release is a consequence of the reduction of neurotransmitter release per se. Indeed, at the same Aplysia synapses, no delay in TeNT action was observed after a ∼50% decrease of neurotransmitter release was induced by the injection of N-terminal fragments of VAMP/synaptobrevin (Cornille et al., 1995). A possibility is that the delay in the toxins’ action reflects a change in the dynamics of intracellular pools of synaptic vesicles. The existence of a pool of toxin-insensitive vesicles corresponding to ready-to-fuse or ‘primed’ vesicles has been previously suggested (McMahon et al., 1992; Bittner et al., 1993; Lawrence et al., 1994). In normal conditions, this pool of vesicles would be small and thus rapidly depleted, leading to fast inhibition of ACh release by either TeNT or BoNT/A. In contrast, the size of this pool would be increased in neurons injected with hu/apRab3Q81L. The recruitment of toxin-insensitive vesicles from this enlarged reserve pool would sustain ACh release for a longer time, thus accounting for the increased latency period. It may be noted that the GTPasedeficient Rab3 mutant did not change the slope of the toxininduced blockade of neurotransmission. This suggests that the number of releasable vesicles becomes rate-limiting only after exhaustion of the hu/apRab3Q81L-induced reserve pool. An increased number of readily releasable vesicles might be evidenced by morphological techniques as an increased number of synaptic vesicles docked at the active zone. However, the complexity of the synaptic arborisation of central neurons in Aplysia hampered any morphometric analysis. Such an analysis will be carried out in another experimental system. Since it is unlikely that hu/apRab3Q81L directly affects the proteolytic activity of TeNT and BoNT/A, the delayed effect of the neurotoxins rather indicates a reduced accessibility of these metallo-proteases to their respective targets, i.e. to VAMP and SNAP-25. Recently, it has been shown that VAMP, SNAP-25, and syntaxin become uncleavable by neurotoxins once they are engaged within the SNARE complex (Hayashi et al., 1994; Pellegrini et al., 1994, 1995). This result provides a simple explanation for the observed effect of hu/apRab3Q81L on TeNT and BoNT/A activity. We suggest that hu/apRab3Q81L induces the accumulation of SNAP-25 and VAMP in a toxin-insensitive state which corresponds to the SNARE complex itself.

Despite many attempts, direct physical interaction between Rab proteins and the SNAREs has not been demonstrated. However, genetic and biochemical evidence in yeast suggest that members of the Rab/Ypt/Sec4 family of GTP-binding proteins are implicated in the formation or stabilization of the SNARE complex (Dascher et al., 1991; Søgaard et al., 1994; Lian et al., 1994; Brennwald et al., 1994). Overexpression of SEC9, a yeast homologue of SNAP-25, was found to suppress an effector domain mutation of SEC4 (Brennwald et al., 1994), and the ER-to-Golgi SNARE complex fails to form in the absence of functional Ypt1 protein (Søgaard et al., 1994; Lian et al., 1994). Thus, in agreement with these data, the accumulation of assembled SNARE complexes that seems to be induced by hu/apRab3Q81L might result from an increased rate of complex assembly. However, the observed inhibition of ACh release by hu/apRab3Q81L cannot be explained by an increase in SNAREs assembly and would point towards another facet of Rab3 function. A possibility that is consistent with the data presented here would be that hu/apRab3Q81L slows the discharge of vesicles from a pool in which the SNARE complex is already formed (as indicated by its insensitivity to either TeNT or BoNT/A). This would account for the observed inhibition of ACh release and would also contribute to the proposed accumulation of a toxin-resistant component of release.

Finally, in order to take into account all the available data, a model might be proposed in which Rab3 would have a dual effect on the secretory process. On one hand, Rab3 would promote the secretory activity, possibly by catalysing the assembly of the SNARE complex. This might account for the defect in synaptic transmission observed, under certain experimental conditions, in mice carrying a rab3a null mutation (Geppert et al., 1994). On the other hand, due to its rate-limiting GTPase activity (Holz et al., 1994; Johannes et al., 1994, this paper), Rab3 would control a step downstream of the SNARE complex formation. Depending upon the respective activities and concentrations of regulatory factors, Rab3 might therefore contribute to the fine tuning of the secretory activity, with integrative properties generally proposed for Ras-like proteins.

We thank M. P. Laran-Chich for technical assistance, Dr C. Montecucco for his kind gift of TeNT and BoNT/A, Dr M. Seabra for providing purified GGTase II, Dr Y. Hu, Dr R. Baston and Dr E. R. Kandel for the cDNA encoding Aplysia Rab3, and Dr E. Meier for revising the manuscript. L.J. is supported by a fellowship from Boehringer Ingelheim Fonds. This work was supported by research contracts from the Association Française de Lutte contre les Myopathies and Fondation pour la Recherche Médicale (to B.P.), and from the Association pour la Recherche sur le Cancer (to J.P.H.).

Bennett
,
M. K.
and
Scheller
,
R. H.
(
1993
).
The molecular machinery for secretion is conserved from yeast to neurons
.
Proc. Nat. Acad. Sci. USA
90
,
2559
2563
.
Bittner
,
M. A.
and
Holz
,
R. W.
(
1993
).
Protein kinase C and clostridial neurotoxins affect discrete and related steps in the secretory pathway
.
Cell. Mol. Neurobiol
.
13
,
649
664
.
Brennwald
,
P.
,
Kearns
,
B.
,
Champion
,
K.
,
Keränen
,
S.
,
Bankaitis
,
V.
and
Novick
,
P.
(
1994
).
Sec9 is a SNAP-25-like component of a yeast SNARE complex that may be the effector of Sec4 function in exocytosis
.
Cell
79
,
245
258
.
Brondyk
,
W. H.
,
McKiernan
,
C. J.
,
Burstein
,
E. S.
and
Macara
,
I. G.
(
1993
).
Mutants of Rab3A analogous to oncogenic Ras mutants
.
J. Biol. Chem
.
268
,
9410
9415
.
Burstein
,
E. S.
and
Macara
,
I. G.
(
1992
).
Interactions of the ras-like protein-p25rab3A with Mg2+ and guanine nucleotides
.
Biochem. J
.
282
,
387
392
.
Calakos
,
N.
and
Scheller
,
R. H.
(
1994
).
Vesicle-associated membrane protein and synaptophysin are associated on the synaptic vesicle
.
J. Biol. Chem
.
269
,
24534
24537
.
Chavrier
,
P.
,
Gorvel
,
J. P.
,
Stelzer
,
E.
,
Simons
,
K.
,
Gruenberg
,
J.
and
Zerial
,
M.
(
1991
).
Hypervariable C-terminal domain of Rab proteins acts as a targeting signal
.
Nature
353
,
769
772
.
Cornille
,
F.
,
Deloye
,
F.
,
Fournie-Zaluski
,
M.-C.
,
Roques
,
B. P.
and
Poulain
,
B.
(
1995
).
Inhibition of neurotransmitter release by synthetic proline-rich peptides shows that the N-terminal domain of vesicle-associated membrane protein/synaptobrevin is critical for neuro-exocytosis
.
J. Biol. Chem
.
270
,
16826
16832
.
Cremers
,
F. P. M.
,
Armstrong
,
S. A.
,
Seabra
,
M. C.
,
Brown
,
M. S.
and
Goldstein
,
J. L.
(
1994
).
REP-2, a Rab escort protein encoded by the choroideremia-like gene
.
J. Biol. Chem
.
269
,
2111
2117
.
Darchen
,
F.
,
Zahraoui
,
A.
,
Hammel
,
F.
,
Monteils
,
M.-P.
,
Tavitian
,
A.
and
Scherman
,
D.
(
1990
).
Association of the GTP-binding protein Rab3A with bovine adrenal chromaffin granules
.
Proc. Nat. Acad. Sci. USA
87
,
5692
5696
.
Darchen
,
F.
,
Senyshyn
,
J.
,
Brondyk
,
W. H.
,
Taatjes
,
D. J.
,
Holz
,
R. W.
,
Henry
,
J.-P.
,
Denizot
,
J.-P.
and
Macara
,
I. G.
(
1995
).
The GTPase Rab3a is associated with large dense core vesicles in bovine chromaffin cells and rat PC12 cells
.
J. Cell Sci
.
108
,
1639
1649
.
Dascher
,
C.
,
Ossig
,
R.
,
Gallwitz
,
D.
and
Schmitt
,
H. D.
(
1991
).
Identification and structure of 4 yeast genes (SLY) that are able to suppress the functional loss of YPT1, a member of the RAS superfamily
.
Mol. Cell. Biol
.
11
,
872
885
.
Edelmann
,
L.
,
Hanson
,
P. I.
,
Chapman
,
E. R.
and
Jahn
,
R.
(
1995
).
Synaptobrevin binding to synaptophysin: a potential mechanism for controlling the exocytotic fusion machine
.
EMBO J
.
14
,
224
231
.
Ferro-Novick
,
S.
and
Jahn
,
R.
(
1994
).
Vesicle fusion from yeast to man
.
Nature
370
,
191
193
.
Fischer von Mollard
,
G.
,
Mignery
,
G. A.
,
Baumert
,
M.
,
Perin
,
M. S.
,
Hanson
,
T. J.
,
Burger
,
P. M.
,
Jahn
,
R.
and
Südhof
,
T. C.
(
1990
).
Rab3 is a small GTP-binding protein exclusively localized to synaptic vesicles
.
Proc. Nat. Acad. Sci. USA
87
,
1988
1992
.
Geppert
,
M.
,
Bolshakov
,
V. Y.
,
Siegelbaum
,
S. A.
,
Takei
,
K.
,
De Camilli
,
P.
,
Hammer
,
R. E.
and
Südhof
,
T. C.
(
1994
).
The role of Rab3A in neurotransmitter release
.
Nature
369
,
493
497
.
Hanson
,
P. I.
,
Otto
,
H.
,
Barton
,
N.
and
Jahn
,
R.
(
1995
).
The N-ethylmaleimide-sensitive fusion protein and α-SNAP induce a conformational change in syntaxin
.
J. Biol. Chem
.
270
,
16955
16961
.
Hayashi
,
T.
,
McMahon
,
H.
,
Yamasaki
,
S.
,
Binz
,
T.
,
Hata
,
Y.
,
Südhof
,
T. C.
and
Niemann
,
H.
(
1994
).
Synaptic vesicle membrane fusion complex: Action of clostridial neurotoxins on assembly
.
EMBO J
.
13
,
5051
5061
.
Holz
,
R. W.
,
Brondyk
,
W. H.
,
Senter
,
R. A.
,
Kuzion
,
L.
and
Macara
,
I. G.
(
1994
).
Evidence for the involvement of rab3a in Ca2+-dependent exocytosis from adrenal chromaffin cells
.
J. Biol. Chem
.
269
,
10229
10234
.
Johannes
,
L.
,
Lledo
,
P.-M.
,
Roa
,
M.
,
Vincent
,
J.-D.
,
Henry
,
J.-P.
and
Darchen
,
F.
(
1994
).
The GTPase Rab3a negatively controls calcium-dependent exocytosis in neuroendocrine cells
.
EMBO J
.
13
,
2029
2037
.
Lawrence
,
G. W.
,
Weller
,
U.
and
Dolly
,
J. O.
(
1994
).
Botulinum A and the light chain of tetanus toxins inhibit distinct stages of Mg.ATP-dependent catecholamine exocytosis from permeabilized chromaffin cells
.
Eur. J. Biochem
.
222
,
325
333
.
Lian
,
J. P.
,
Stone
,
S.
,
Jiang
,
Y.
,
Lyons
,
P.
and
Ferro-Novick
,
S.
(
1994
).
Ypt1p implicated in v-SNARE activation
.
Nature
372
,
698
701
.
Lledo
,
P.-M.
,
Vernier
,
P.
,
Vincent
,
J.-D.
,
Mason
,
W. T.
and
Zorec
,
R.
(
1993
).
Inhibition of Rab3B expression attenuates Ca2+-dependent exocytosis in rat anterior pituitary cells
.
Nature
364
,
540
544
.
Lledo
,
P.-M.
,
Johannes
,
L.
,
Vernier
,
P.
,
Zorec
,
R.
,
Darchen
,
F.
,
Vincent
,
J.-D.
,
Henry
,
J.-P.
and
Mason
,
W. T.
(
1994
).
Rab3 proteins: Key players in the control of exocytosis
.
Trends Neurosci
.
17
,
426
432
.
Lombardi
,
D.
,
Soldati
,
T.
,
Riederer
,
M. A.
,
Goda
,
Y.
,
Zerial
,
M.
and
Pfeffer
,
S. R.
(
1993
).
Rab9 functions in transport between late endosomes and the trans Golgi network
.
EMBO J
.
12
,
677
682
.
Lupashin
,
V. V.
,
Hamamoto
,
S.
and
Scheckman
,
R. W.
(
1996
).
Biochemical requirements for the targeting and fusion of ER-derived transport vesicles with purified yeast Golgi membranes
.
J. Cell Biol
.
132
,
277
289
.
Martinez
,
O.
,
Schmidt
,
A.
,
Salamero
,
J.
,
Hoflack
,
B.
,
Roa
,
M.
and
Goud
,
B.
(
1994
).
The small GTP-binding protein rab6 functions in intra-Golgi transport
.
J. Cell Biol
.
127
,
1575
1588
.
McMahon
,
H. T.
,
Foran
,
P.
,
Dolly
,
J. O.
,
Verhage
,
M.
,
Wiegant
,
V. M.
and
Nicholls
,
D. G.
(
1992
).
Tetanus toxin and botulinum toxins type A and B inhibit glutamate, γ-aminobutyric acid, aspartate, and met-enkephalin release from synaptosomes
.
J. Biol. Chem
.
267
,
21338
21343
.
McKiernan
,
C. J.
,
Brondyk
,
W. H.
and
Macara
,
I. G.
(
1993
).
The Rab3A GTPase interacts with multiple factors through the same effector domain - mutational analysis of cross-linking of Rab3A to a putative target protein
.
J. Biol. Chem
.
268
,
24449
24452
.
Montecucco
,
C.
and
Schiavo
,
G.
(
1995
).
Structure and function of tetanus and botulinum neurotoxins
.
Quart. Rev. Biophys
.
28
,
423
472
.
Musha
,
T.
,
Kawata
,
M.
and
Takai
,
Y.
(
1992
).
The geranylgeranyl moiety but not the methyl moiety of the smg-25A/rab3A protein is essential for the interactions with membrane and its inhibitory GDP/GTP exchange protein
.
J. Biol. Chem
.
267
,
9821
9825
.
Niemann
,
H.
,
Blasi
,
J.
and
Jahn
,
R.
(
1994
).
Clostridial neurotoxins: new tools for dissecting exocytosis
.
Trends Cell Biol
.
4
,
179
185
.
Pellegrini
,
L. L.
,
O’Connor
,
V.
and
Betz
,
H.
(
1994
).
Fusion complex formation protects synaptobrevin against proteolysis by tetanus toxin light chain
.
FEBS Lett
.
353
,
319
323
.
Pellegrini
,
L. L.
,
O’Connor
,
V.
,
Lottspeich
,
F.
and
Betz
,
H.
(
1995
).
Clostridial neurotoxins compromise the stability of a low energy SNARE complex mediating NSF activation of synaptic vesicle fusion
.
EMBO J
.
14
,
4705
4713
Pfeffer
,
S. R.
(
1994
).
Rab GTPases: Master regulators of membrane trafficking
.
Curr. Opin. Cell Biol
.
6
,
522
526
.
Poulain
,
B.
,
Rossetto
,
O.
,
Deloye
,
F.
,
Schiavo
,
G.
,
Tauc
,
L.
and
Montecucco
,
C.
(
1993
).
Antibodies against rat brain vesicle-associated membrane protein (synaptobrevin) prevent inhibition of acetylcholine release by tetanus toxin or botulinum neurotoxin type B
.
J. Neurochem
.
61
,
1175
1178
.
Poulain
,
B.
,
Molgo
,
J.
and
Thesleff
,
S.
(
1995
).
Quantal neurotransmitter release and the clostridial neurotoxins’ targets
.
Curr. Top. Microbiol. Immunol
.
195
,
243
255
.
Poulain
,
B.
,
de Paiva
,
A.
,
Deloye
,
F.
,
Doussau
,
F.
,
Tauc
,
L.
,
Weller
,
U.
and
Dolly
,
J. O.
(
1996
).
Differences in the mutiple step process of inhibition of neurotransmitter release induced by tetanus toxin and botulinum neurotoxins type A and B at Aplysia synapses
.
Neuroscience
70
,
567
576
.
Rossetto
,
O.
,
Schiavo
,
G.
,
Montecucco
,
C.
,
Poulain
,
B.
,
Deloye
,
F.
,
Lozzi
,
L.
and
Shone
,
C.
(
1994
).
SNARE motif and neurotoxins
.
Nature
372
,
415
416
.
Rothman
,
J. E.
(
1994
).
Mechanisms of intracellular protein transport
.
Nature
372
,
55
63
.
Rothman
,
J. E.
and
Warren
,
G.
(
1994
).
Implications of the SNARE hypothesis for intracellular membrane topology and dynamics
.
Curr. Biol
.
4
,
220
233
.
Schiavo
,
G.
,
Benfenati
,
F.
,
Poulain
,
B.
,
Rossetto
,
O.
,
Polverino de Laureto
,
P.
,
DasGupta
,
B. R.
and
Montecucco
,
C.
(
1992a
).
Tetanus and botulinum-B neurotoxins block neurotransmitter release by proteolytic cleavage of synaptobrevin
.
Nature
359
,
832
835
.
Schiavo
,
G.
,
Poulain
,
B.
,
Rossetto
,
O.
,
Benfenati
,
F.
,
Tauc
,
L.
and
Montecucco
,
C.
(
1992b
).
Tetanus toxin is a zinc protein and its inhibition of neurotransmitter release and protease activity depend on zinc
.
EMBO J
.
11
,
3577
3583
.
Schiavo
,
G.
,
Rossetto
,
O.
,
Santucci
,
A.
,
DasGupta
,
B. R.
and
Montecucco
,
C.
(
1992c
).
Botulinum neurotoxins are zinc proteins
.
J. Biol. Chem
.
267
,
23479
23483
.
Simons
,
K.
and
Zerial
,
M.
(
1993
).
Rab proteins and the road maps for intracellular transport
.
Neuron
11
,
789
799
.
Skehel
,
P. A.
,
Martin
,
K. C.
,
Kandel
,
E. R.
and
Bartsch
,
D.
(
1995
).
A VAMP-binding protein from Aplysia required for neurotransmitter release
.
Science
269
,
1580
1583
.
Søgaard
,
M.
,
Tani
,
K.
,
Ye
,
R. R.
,
Geromanos
,
S.
,
Tempst
,
P.
,
Kirchhausen
,
T.
,
Rothman
,
J. E.
and
Söllner
,
T.
(
1994
).
A rab protein is required for the assembly of SNARE complexes in the docking of transport vesicles
.
Cell
78
,
937
948
.
Söllner
,
T.
,
Whitehart
,
S. W.
,
Brunner
,
M.
,
Erdjument-Bromage
,
H.
,
Geromanos
,
S.
,
Tempst
,
P.
and
Rothman
,
J. E.
(
1993a
).
SNAP receptors implicated in vesicle targeting and fusion
.
Nature
362
,
318
324
.
Söllner
,
T.
,
Bennett
,
M. K.
,
Whiteheart
,
S. W.
,
Scheller
,
R. H.
and
Rothman
,
J. E.
(
1993b
).
A protein assembly-disassembly pathway in vitro that may correspond to sequential steps of synaptic vesicle docking, activation, and fusion
.
Cell
75
,
409
418
.
Stenmark
,
H.
,
Valencia
,
A.
,
Martinez
,
O.
,
Ullrich
,
O.
,
Goud
,
B.
and
Zerial
,
B.
(
1994
).
Distinct structural elements of rab5 define its functional specificity
.
EMBO J
.
13
,
575
583
.
Tisdale
,
E. J.
,
Bourne
,
J. R.
,
Khosravi-Far
,
R.
,
Der
,
C. J.
and
Balch
,
W. E.
(
1992
).
GTP-binding mutants of Rab1 and Rab2 are potent inhibitors of vesicular transport from the endoplasmic reticulum to the Golgi complex
.
J. Cell Biol
.
119
,
749
761
.
Washbourne
,
P.
,
Schiavo
,
G.
and
Montecucco
,
C.
(
1995
).
Vesicle-associated membrane protein-2 (synaptobrevin 2) forms a complex with synaptophysin
.
Biochem. J
.
305
,
721
724
.
Zahraoui
,
A.
,
Touchot
,
N.
,
Chardin
,
P.
and
Tavitian
A.
(
1989
).
The human Rab genes encode a family of GTP-binding proteins related to Yeast YPT1 and SEC4 products involved in secretion
.
J. Biol. Chem
.
264
,
12394
12401
.