Portal de Periódicos da CAPES

Conserved α-Helical Segments on Yeast Homologs of the Synaptobrevin/VAMP Family of v-SNAREs Mediate Exocytic Function

1997; Elsevier BV; Volume: 272; Issue: 26 Linguagem: Inglês

10.1074/jbc.272.26.16591

1083-351X

Jeffrey E. Gerst,

Endoplasmic Reticulum Stress and Disease

We are studying yeast homologs of the synaptobrevin/VAMP family of vesicle-associated membrane proteins, which act as vesicular compartment-solubleN-ethylmaleimide-sensitive factor attachment protein receptors (v-SNAREs) in cells having a capacity for stimulus-coupled secretion, as well as in other cell types. The yeast homologs, Snc1 and Snc2, localize to secretory vesicles and are required for normal bulk secretion in Saccharomyces cerevisiae. Here we have used Snc deletion mutants and chimeric Snc-VAMP proteins to demonstrate that these v-SNAREs can be dissected into regions that are either indispensable or dispensable for exocytic function in vivo.We have found that a region encompassing two predicted amphipathic α-helices (helix 1 and helix 2) (residues 32–85), which are thought to form coiled-coil structures, is essential for conferring exocytosis in yeast. Deletions in either the helix 1 or helix 2 segments result in a complete loss in the ability of the protein to confer secretion competence to snc cells and to interact genetically with components of the proposed fusion complex: the Sec9 and Sso2 t-SNAREs and the Sec17 α-SNAP homolog. In contrast, deletions in either the variable (residues 2–27) or putative intravesicular (residues 115–117) regions have no deleterious effect upon v-SNARE function. This makes it unlikely that sequences in either the amino or carboxyl terminus act in an exocytic capacity. Along with additional studies utilizing chimeric Snc-VAMP proteins, we suggest that although the Snc and synaptobrevin/VAMP proteins have evolved to mediate vastly different exocytic programs, their structural requirements and actions have remained remarkably well-conserved in evolution. We are studying yeast homologs of the synaptobrevin/VAMP family of vesicle-associated membrane proteins, which act as vesicular compartment-solubleN-ethylmaleimide-sensitive factor attachment protein receptors (v-SNAREs) in cells having a capacity for stimulus-coupled secretion, as well as in other cell types. The yeast homologs, Snc1 and Snc2, localize to secretory vesicles and are required for normal bulk secretion in Saccharomyces cerevisiae. Here we have used Snc deletion mutants and chimeric Snc-VAMP proteins to demonstrate that these v-SNAREs can be dissected into regions that are either indispensable or dispensable for exocytic function in vivo. We have found that a region encompassing two predicted amphipathic α-helices (helix 1 and helix 2) (residues 32–85), which are thought to form coiled-coil structures, is essential for conferring exocytosis in yeast. Deletions in either the helix 1 or helix 2 segments result in a complete loss in the ability of the protein to confer secretion competence to snc cells and to interact genetically with components of the proposed fusion complex: the Sec9 and Sso2 t-SNAREs and the Sec17 α-SNAP homolog. In contrast, deletions in either the variable (residues 2–27) or putative intravesicular (residues 115–117) regions have no deleterious effect upon v-SNARE function. This makes it unlikely that sequences in either the amino or carboxyl terminus act in an exocytic capacity. Along with additional studies utilizing chimeric Snc-VAMP proteins, we suggest that although the Snc and synaptobrevin/VAMP proteins have evolved to mediate vastly different exocytic programs, their structural requirements and actions have remained remarkably well-conserved in evolution. Soluble N-ethylmaleimide-sensitive factor attachment protein (SNAP) 1The abbreviations used are: SNAP-25, synaptosomal-associated protein of 25 kilodaltons; HA, influenza virus hemagglutinin antigen; NSF, N-ethylmaleimide-sensitive factor; SNAP, soluble NSF attachment protein; SNARE, soluble NSF attachment protein receptors; v-SNAREs, vesicular compartment; t-SNAREs, target compartment; SNC, suppressor of the null allele of the cyclase-associated protein; SSO, suppressor of Sec1; VAMP, vesicle-associated membrane protein; YPD, yeast extract/Bacto-peptone/dextrose; ts, temperature-sensitive; PCR, polymerase chain reaction. 1The abbreviations used are: SNAP-25, synaptosomal-associated protein of 25 kilodaltons; HA, influenza virus hemagglutinin antigen; NSF, N-ethylmaleimide-sensitive factor; SNAP, soluble NSF attachment protein; SNARE, soluble NSF attachment protein receptors; v-SNAREs, vesicular compartment; t-SNAREs, target compartment; SNC, suppressor of the null allele of the cyclase-associated protein; SSO, suppressor of Sec1; VAMP, vesicle-associated membrane protein; YPD, yeast extract/Bacto-peptone/dextrose; ts, temperature-sensitive; PCR, polymerase chain reaction. receptors (SNAREs) comprise several families of evolutionarily conserved, membrane-localized receptors for components of the vesicle docking and fusion machinery in eukaryotes. These receptors are thought to participate in the fusion of carrier vesicles with their target membranes by facilitating both vesicle docking and bilayer interaction (1Rothman J.E. Warren G. Curr. Biol. 1994; 4: 220-233Abstract Full Text Full Text PDF PubMed Scopus (478) Google Scholar). SNAREs are found on both the vesicular compartment (v-SNAREs), as well as on the acceptor or target compartment (t-SNAREs), and at all levels of the secretory pathway.We have identified two proteins that act as v-SNAREs on carrier vesicles of the late secretory pathway in yeast. These proteins, Snc1 (2Gerst J.E. Rodgers L. Riggs M. Wigler M. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 4338-4342Crossref PubMed Scopus (104) Google Scholar) and Snc2 (3Protopopov V. Govindan B. Novick P. Gerst J.E. Cell. 1993; 74: 855-861Abstract Full Text PDF PubMed Scopus (250) Google Scholar), are homologs of neuronal proteins known as synaptobrevins or VAMPs (4Baumert M. Maycox P.R. Navone F. DeCamilli P. Jahn R. EMBO J. 1989; 8: 379-384Crossref PubMed Scopus (396) Google Scholar, 5Trimble W.S. Cowan D.M. Scheller R.H. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 4538-4542Crossref PubMed Scopus (449) Google Scholar) and a nonneuronal, constitutively expressed form known as cellubrevin (6McMahon H.T. Ushkaryov Y.A. Edelmann L. Link E. Binz T. Niemann H. Südhof T.C. Nature. 1993; 364: 346-349Crossref PubMed Scopus (397) Google Scholar). Synaptobrevin/VAMPs were first identified as components of synaptic vesicles (4Baumert M. Maycox P.R. Navone F. DeCamilli P. Jahn R. EMBO J. 1989; 8: 379-384Crossref PubMed Scopus (396) Google Scholar, 5Trimble W.S. Cowan D.M. Scheller R.H. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 4538-4542Crossref PubMed Scopus (449) Google Scholar) and were also isolated as elements that participate in the binding of SNAP proteinsin vitro (7Söllner T. Whiteheart S.W. Brunner M. Erdjument-Bromage H. Geromanos S. Tempst P. Rothman J.E. Nature. 1993; 362: 318-324Crossref PubMed Scopus (2603) Google Scholar, 8Söllner T. Bennett M.K. Whiteheart S.W. Scheller R.H. Rothman J.E. Cell. 1993; 75: 409-418Abstract Full Text PDF PubMed Scopus (1573) Google Scholar).Genetic studies demonstrate that Snc proteins are required for vesicle docking and fusion (3Protopopov V. Govindan B. Novick P. Gerst J.E. Cell. 1993; 74: 855-861Abstract Full Text PDF PubMed Scopus (250) Google Scholar, 9Couve A. Gerst J.E. J. Biol. Chem. 1994; 269: 23391-23394Abstract Full Text PDF PubMed Google Scholar). Yeast cells lacking Snc protein expression accumulate secretory vesicles and fail to secrete normally. In addition, these cells show conditional-lethal phenotypes that result from the blockage of vesicle fusion. Like their neuronal counterparts, Snc proteins are thought to interact physically with t-SNAREs from the plasma membrane to form a putative SNARE complex (9Couve A. Gerst J.E. J. Biol. Chem. 1994; 269: 23391-23394Abstract Full Text PDF PubMed Google Scholar, 10Brennwald P. Kerns B. Champion K. Keränen S. Bankaitis V. Novick P. Cell. 1994; 79: 245-258Abstract Full Text PDF PubMed Scopus (311) Google Scholar). Specifically, these comprise members of the Sec9/SNAP-25 (10Brennwald P. Kerns B. Champion K. Keränen S. Bankaitis V. Novick P. Cell. 1994; 79: 245-258Abstract Full Text PDF PubMed Scopus (311) Google Scholar, 11Oyler G.A. Higgins G.A. Hart R.A. Battenberg E. Billingsley M. Bloom F.E. Wilson M.C. J. Cell Biol. 1989; 109: 3030-3052Crossref Scopus (687) Google Scholar) and Sso/syntaxin (12Bennett M.K. Calakos N. Scheller R.H. Science. 1992; 257: 255-259Crossref PubMed Scopus (1065) Google Scholar, 13Inoue A. Obata K. Akagawa K. J. Biol. Chem. 1992; 267: 10613-10619Abstract Full Text PDF PubMed Google Scholar, 14Aalto M.K. Ronne H. Keränen S. EMBO J. 1993; 12: 4095-4104Crossref PubMed Scopus (343) Google Scholar) families. Snc proteins co-precipitate with the Sec9 (9Couve A. Gerst J.E. J. Biol. Chem. 1994; 269: 23391-23394Abstract Full Text PDF PubMed Google Scholar, 10Brennwald P. Kerns B. Champion K. Keränen S. Bankaitis V. Novick P. Cell. 1994; 79: 245-258Abstract Full Text PDF PubMed Scopus (311) Google Scholar) and Sso proteins (10Brennwald P. Kerns B. Champion K. Keränen S. Bankaitis V. Novick P. Cell. 1994; 79: 245-258Abstract Full Text PDF PubMed Scopus (311) Google Scholar) from yeast detergent extracts, and genetic studies suggest that Sec9 function in exocytosis may be Snc-dependent (9Couve A. Gerst J.E. J. Biol. Chem. 1994; 269: 23391-23394Abstract Full Text PDF PubMed Google Scholar). Recently, we have determined that Snc proteins are modified by the palmitoylation of unique cysteine residues located proximal to their transmembrane domains (15Couve A. Protopopov V. Gerst J.E. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 5987-5991Crossref PubMed Scopus (48) Google Scholar). Palmitoylation is required to yield the mature form of the protein but is not necessary for exocytic function.Synaptobrevin/VAMP and Snc proteins are conserved type II membrane proteins. They consist of a variable region of ∼30 residues at the amino terminus, followed by a conserved region (∼60 residues) bearing two predicted amphipathic α-helices separated by a nonhelical spacer, and a single membrane-spanning domain at the carboxyl terminus (16Dascher C. Ossig R. Gallwitz D. Schmidt H.D. Mol. Cell. Biol. 1991; 11: 872-885Crossref PubMed Scopus (278) Google Scholar,17Trimble W.S. J. Physiol. ( Paris ). 1993; 87: 107-115Crossref PubMed Scopus (28) Google Scholar). The variable and conserved regions are cytosolic, whereas the 3–4 residues adjacent to the membrane-spanning domain are likely to be intravesicular. The first helix of the conserved region of synaptobrevin/VAMP (known as helix 1 and designated here as “H1”) contains a signal for the targeting to synaptic-like microvesicles in PC12 cells (18Grote E. Hao J.C. Bennett M.K. Kelly R.B. Cell. 1995; 81: 581-589Abstract Full Text PDF PubMed Scopus (144) Google Scholar). The second helix of synaptobrevin/VAMP (known as helix 2 and designated here as “H2”) is thought to participate, along with H1, in the interaction with cognate t-SNAREs from the presynaptic membrane (e.g. SNAP-25 and syntaxins A and B), based uponin vitro studies (18Grote E. Hao J.C. Bennett M.K. Kelly R.B. Cell. 1995; 81: 581-589Abstract Full Text PDF PubMed Scopus (144) Google Scholar, 19Hayashi T. McMahon H. Yamasaki S. Binz T. Hata Y. Binz T. Sudhof T.C. Niemann H. EMBO J. 1994; 13: 5051-5061Crossref PubMed Scopus (661) Google Scholar, 20Chapman E.R. An S. Barton N. Jahn R. J. Biol. Chem. 1994; 269: 27427-27432Abstract Full Text PDF PubMed Google Scholar). These v- and t-SNARE interactions are thought to result from the formation of coiled-coil structures involving the amphipathic α-helices of the individual SNARE partners.To address whether specific regions within the synaptobrevin/VAMPs confer the SNARE-SNARE interactions that lead to exocytosis, we have used the yeast system to perform a structure/function analysis of the Snc v-SNAREs. By using genetic and biochemical assays, we demonstrate that the ability of Snc proteins to confer exocytic function in vivo corresponds to their ability to interact genetically with components of the putative fusion complex (Sec17, Sec9, and Sso2). We also demonstrate that the H1 and H2 segments together mediate the exocytic functions of these v-SNAREs. In addition, there is a distinct requirement for the membrane-spanning domain but no requirement for the intravesicular and variable regions. The latter suggests that this region is unlikely to act in a fusogenic capacity, as proposed for the amino terminus of the hemagglutinin antigen (HA), a viral fusion protein bearing similar structural motifs (reviewed in Ref. 21Stegmann T. Helenius A. Bentz J. Viral Fusion Mechanisms. CRC Press, Inc., Boca Raton, FL1993: 89-112Google Scholar).DISCUSSIONYeast Snc proteins are archetypal v-SNAREs of the late secretory pathway. These proteins share structural homology with members of the synaptobrevin/VAMP/cellubrevin family and engage in similar exocytic functions. Like their brethren, Snc proteins localize to secretory vesicles (3Protopopov V. Govindan B. Novick P. Gerst J.E. Cell. 1993; 74: 855-861Abstract Full Text PDF PubMed Scopus (250) Google Scholar), interact with t-SNAREs from the plasma membrane (9, 10, and this study), and are likely to undergo a dynamic cycle of transport to, and retrieval from, the plasma membrane.Here we have demonstrated that the conserved amphipathic α-helical region (residues 32–85) is essential for conferring the exocytic functions of these v-SNAREs. Deletions or gross substitutions in either of the predicted H1 or H2 segments result in a complete loss of function of the proteins, although they appear to reach secretory vesicles and the plasma membrane. Thus, as was predicted by experiments performed in vitro with mammalian v- and t-SNAREs (18–20, and 33), it would seem likely that both helices are required for SNARE complexing in vivo. Although we cannot say whether the individual helices prefer specific t-SNARE partners, it appears that both helices can interact either with Sec9 or Sso2. This is based upon the result that Snc proteins lacking in either H1 or H2 are unable to suppress ts defects in either Sec9 or Sso2 and inhibit the growth ofsec9 and sso2 cells at permissive temperatures. These proteins, Snc1Δ31-50 and Snc1Δ57-81, appear to exert a dominant negative effect when overexpressed. Thus, it is possible that they compete with the native proteins in undergoing SNARE partnering but are unable to confer secretory functions. Interestingly, the Snc1Δ2-27 mutant was unable to confer suppression of the ts defects in either Sec9 or Sso2, unlike Snc1 or Snc1Δ2-17, suggesting that a portion of the variable domain aids in the rescue of the mutant t-SNAREs.Work by Hayashi et al. (19Hayashi T. McMahon H. Yamasaki S. Binz T. Hata Y. Binz T. Sudhof T.C. Niemann H. EMBO J. 1994; 13: 5051-5061Crossref PubMed Scopus (661) Google Scholar) demonstrated that the domains required for SNARE complex formation in vitro correlate well with the regions predicted to form α-helices. Specifically, synaptobrevin binding to the carboxyl-terminal third of syntaxin or SNAP-25 was shown to involve the entire conserved domain (residues 27–96) (19Hayashi T. McMahon H. Yamasaki S. Binz T. Hata Y. Binz T. Sudhof T.C. Niemann H. EMBO J. 1994; 13: 5051-5061Crossref PubMed Scopus (661) Google Scholar). Earlier studies also showed binary interactions between synaptobrevin/VAMP and syntaxin (33Calakos N. Bennett M.K. Peterson K.E. Scheller R.H. Science. 1994; 263: 1146-1149Crossref PubMed Scopus (364) Google Scholar) and SNAP-25 (20Chapman E.R. An S. Barton N. Jahn R. J. Biol. Chem. 1994; 269: 27427-27432Abstract Full Text PDF PubMed Google Scholar). Deletions of both the variable region as well as the transmembrane domain of synaptobrevin were found to have no deleterious effect upon complex formation and, in contrast, were found to enhance the interaction (19Hayashi T. McMahon H. Yamasaki S. Binz T. Hata Y. Binz T. Sudhof T.C. Niemann H. EMBO J. 1994; 13: 5051-5061Crossref PubMed Scopus (661) Google Scholar). Here, using in vivo studies with mutant Snc proteins, we have shown that the structural requirements that lead to secretion are quite similar to those proposed for SNARE complex formation in vitro. In our study, however, the transmembrane domain of the protein is required for function, probably due to its membrane anchoring capacity. This is likely to be essential for the localization of Snc proteins to secretory vesicles, although it may be unnecessary for SNARE complex formation per se.Importantly, the results obtained in our in vivo study have been collaborated by a recent work in which the regions of VAMP required for conferring Ca2+-mediated exocytosis from permeabilized insulin-secreting HIT-T15 cells were examined (34Regazzi R. Sadoul K. Meda P. Kelly R.B. Halban P.A. Wollheim C.B. EMBO J. 1996; 15: 6951-6959Crossref PubMed Scopus (82) Google Scholar). By using tetanus toxin-resistant mutants of VAMP, the authors demonstrate that deletions in either of the amphipathic α-helices inhibited the ability of these mutants to restore exocytosis in tetanus toxin-treated cells. In addition, deletion of the variable region (residues 2–31) had no effect upon VAMP2 function, as shown here for Snc1. Moreover, the ability of these mutants to bind to either syntaxin Ia or SNAP-25in vitro paralleled their ability to confer exocytic functions. Thus, the results obtained in our mutational analysis of the yeast synaptobrevin/VAMP homologs are identical to those obtained for VAMP using a permeabilized cell system.It has been suggested that the amino terminus of the synaptobrevin/VAMP proteins might act in a fusogenic capacity (35Jahn R. Südhof T.C. Annu. Rev. Neurosci. 1994; 17: 219-246Crossref PubMed Scopus (335) Google Scholar) to confer bilayer fusion in the manner described for the fusion peptides of the influenza hemagglutinin antigen. This study suggests that removal of the amino-terminal variable region of a yeast v-SNARE has no deleterious effect upon its ability to confer exocytosis in snc cells. Thus, the likelihood that this region would act in an essential fusogenic capacity would seem slight. This is supported by the previous study (34Regazzi R. Sadoul K. Meda P. Kelly R.B. Halban P.A. Wollheim C.B. EMBO J. 1996; 15: 6951-6959Crossref PubMed Scopus (82) Google Scholar), yet neither work can rule out the possibility that the H1 segment alone could act in the capacity of a fusion peptide, if such a mechanism were indeed operant.Although no well defined role for the variable region has been described, other studies suggest that it may be involved in neuronal transmission. For example, peptides corresponding to the variable region of VAMP2 (but not VAMP1) inhibit acetylcholine release fromAplysia neurons (36Cornille F. Deloye F. Fournié-Zaluski M.C. Roques B.P. Poulain B. J. Biol. Chem. 1995; 270: 16826-16832Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar) and the interaction between VAMP and synaptophysin in vitro (37Washbourne P. Schiavo G. Montecucco C. Biochem. J. 1995; 305: 721-724Crossref PubMed Scopus (109) Google Scholar). Therefore, the role of the variable region could have additional functions in higher eukaryotes and may well be cell- and isoform-specific. As no synaptophysin/synaptoporin homologs have been identified in yeast thus far, we are unable to determine whether this interaction (38Calakos N. Scheller R.H. J. Biol. Chem. 1994; 269: 24534-24537Abstract Full Text PDF PubMed Google Scholar, 39Edelmann L. Hanson O.I. Chapman E.R. Jahn R. EMBO J. 1995; 14: 224-231Crossref PubMed Scopus (386) Google Scholar) is involved in constitutive exocytosis in lower eukaryotes.Another study has demonstrated the requirements for the sorting of VAMP proteins to synaptic vesicles (18Grote E. Hao J.C. Bennett M.K. Kelly R.B. Cell. 1995; 81: 581-589Abstract Full Text PDF PubMed Scopus (144) Google Scholar). There the authors show that the synaptic vesicle sorting signal for both VAMP and cellubrevin is contained within the H1 segment. Removal of residues within H1 was shown to block targeting to synaptic vesicles, whereas alanine-scanning mutagenesis revealed that certain hydrophobic residues that participate in the heptad repeat of the H1 helix are important for sorting (18Grote E. Hao J.C. Bennett M.K. Kelly R.B. Cell. 1995; 81: 581-589Abstract Full Text PDF PubMed Scopus (144) Google Scholar). These mutations are likely to decrease the overall hydrophobicity of the helix and to lessen its ability to form a coiled-coil structure.In our study, we have found that both VAMP2 and VAMP2snc90-117 are likely to target to secretory vesicles and reach the plasma membrane when expressed in yeast. Yet these proteins are deficient in conferring exocytosis, as well as in localizing in quantity to secretory vesicles, during vesiculation insec6 cells. In contrast, Snc1 and several Snc-VAMP chimeras (e.g. Snc1vamp49-116 and Snc1vamp54-93) associated primarily with the vesicles under the same conditions. Therefore, it would seem likely (although not exclusively) that certain sequences in Snc1 confer protein retrieval from the plasma membrane. In our labeling experiment, we cannot differentiate between proteins contributed from the biosynthetic pool and those recycled from the cell surface. Yet, because those proteins that bore the Snc1 H1 segment underwent a significant re-localization to the vesicle population (upon vesiculation), we are inclined to believe that the H1 of the yeast homologs, like that of synaptobrevin/VAMP, may play a role in protein sorting to vesicles. Correspondingly, those chimeras which possess the VAMP2 H1 segment associated primarily with the plasma membrane and not with vesicles. Perhaps, then, the cellular machinery required for Snc retrieval is unable to recognize efficiently the H1 domain of the mammalian isoforms. More work will be required to fully address these issues.Overall, there is a significant degree of structural as well as functional homology between the Snc and synaptobrevin/VAMP proteins, despite the fact that they confer different exocytic programs in vastly diverged systems. Some differences remain to be resolved, but the beauty of these v-SNAREs lies in their overwhelmingly simple structure and conserved set of interactions and functions. Solving the tertiary and quaternary structures of these proteins will ultimately shed light upon how they act in mediating SNARE complex formation and exocytosis. Soluble N-ethylmaleimide-sensitive factor attachment protein (SNAP) 1The abbreviations used are: SNAP-25, synaptosomal-associated protein of 25 kilodaltons; HA, influenza virus hemagglutinin antigen; NSF, N-ethylmaleimide-sensitive factor; SNAP, soluble NSF attachment protein; SNARE, soluble NSF attachment protein receptors; v-SNAREs, vesicular compartment; t-SNAREs, target compartment; SNC, suppressor of the null allele of the cyclase-associated protein; SSO, suppressor of Sec1; VAMP, vesicle-associated membrane protein; YPD, yeast extract/Bacto-peptone/dextrose; ts, temperature-sensitive; PCR, polymerase chain reaction. 1The abbreviations used are: SNAP-25, synaptosomal-associated protein of 25 kilodaltons; HA, influenza virus hemagglutinin antigen; NSF, N-ethylmaleimide-sensitive factor; SNAP, soluble NSF attachment protein; SNARE, soluble NSF attachment protein receptors; v-SNAREs, vesicular compartment; t-SNAREs, target compartment; SNC, suppressor of the null allele of the cyclase-associated protein; SSO, suppressor of Sec1; VAMP, vesicle-associated membrane protein; YPD, yeast extract/Bacto-peptone/dextrose; ts, temperature-sensitive; PCR, polymerase chain reaction. receptors (SNAREs) comprise several families of evolutionarily conserved, membrane-localized receptors for components of the vesicle docking and fusion machinery in eukaryotes. These receptors are thought to participate in the fusion of carrier vesicles with their target membranes by facilitating both vesicle docking and bilayer interaction (1Rothman J.E. Warren G. Curr. Biol. 1994; 4: 220-233Abstract Full Text Full Text PDF PubMed Scopus (478) Google Scholar). SNAREs are found on both the vesicular compartment (v-SNAREs), as well as on the acceptor or target compartment (t-SNAREs), and at all levels of the secretory pathway. We have identified two proteins that act as v-SNAREs on carrier vesicles of the late secretory pathway in yeast. These proteins, Snc1 (2Gerst J.E. Rodgers L. Riggs M. Wigler M. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 4338-4342Crossref PubMed Scopus (104) Google Scholar) and Snc2 (3Protopopov V. Govindan B. Novick P. Gerst J.E. Cell. 1993; 74: 855-861Abstract Full Text PDF PubMed Scopus (250) Google Scholar), are homologs of neuronal proteins known as synaptobrevins or VAMPs (4Baumert M. Maycox P.R. Navone F. DeCamilli P. Jahn R. EMBO J. 1989; 8: 379-384Crossref PubMed Scopus (396) Google Scholar, 5Trimble W.S. Cowan D.M. Scheller R.H. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 4538-4542Crossref PubMed Scopus (449) Google Scholar) and a nonneuronal, constitutively expressed form known as cellubrevin (6McMahon H.T. Ushkaryov Y.A. Edelmann L. Link E. Binz T. Niemann H. Südhof T.C. Nature. 1993; 364: 346-349Crossref PubMed Scopus (397) Google Scholar). Synaptobrevin/VAMPs were first identified as components of synaptic vesicles (4Baumert M. Maycox P.R. Navone F. DeCamilli P. Jahn R. EMBO J. 1989; 8: 379-384Crossref PubMed Scopus (396) Google Scholar, 5Trimble W.S. Cowan D.M. Scheller R.H. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 4538-4542Crossref PubMed Scopus (449) Google Scholar) and were also isolated as elements that participate in the binding of SNAP proteinsin vitro (7Söllner T. Whiteheart S.W. Brunner M. Erdjument-Bromage H. Geromanos S. Tempst P. Rothman J.E. Nature. 1993; 362: 318-324Crossref PubMed Scopus (2603) Google Scholar, 8Söllner T. Bennett M.K. Whiteheart S.W. Scheller R.H. Rothman J.E. Cell. 1993; 75: 409-418Abstract Full Text PDF PubMed Scopus (1573) Google Scholar). Genetic studies demonstrate that Snc proteins are required for vesicle docking and fusion (3Protopopov V. Govindan B. Novick P. Gerst J.E. Cell. 1993; 74: 855-861Abstract Full Text PDF PubMed Scopus (250) Google Scholar, 9Couve A. Gerst J.E. J. Biol. Chem. 1994; 269: 23391-23394Abstract Full Text PDF PubMed Google Scholar). Yeast cells lacking Snc protein expression accumulate secretory vesicles and fail to secrete normally. In addition, these cells show conditional-lethal phenotypes that result from the blockage of vesicle fusion. Like their neuronal counterparts, Snc proteins are thought to interact physically with t-SNAREs from the plasma membrane to form a putative SNARE complex (9Couve A. Gerst J.E. J. Biol. Chem. 1994; 269: 23391-23394Abstract Full Text PDF PubMed Google Scholar, 10Brennwald P. Kerns B. Champion K. Keränen S. Bankaitis V. Novick P. Cell. 1994; 79: 245-258Abstract Full Text PDF PubMed Scopus (311) Google Scholar). Specifically, these comprise members of the Sec9/SNAP-25 (10Brennwald P. Kerns B. Champion K. Keränen S. Bankaitis V. Novick P. Cell. 1994; 79: 245-258Abstract Full Text PDF PubMed Scopus (311) Google Scholar, 11Oyler G.A. Higgins G.A. Hart R.A. Battenberg E. Billingsley M. Bloom F.E. Wilson M.C. J. Cell Biol. 1989; 109: 3030-3052Crossref Scopus (687) Google Scholar) and Sso/syntaxin (12Bennett M.K. Calakos N. Scheller R.H. Science. 1992; 257: 255-259Crossref PubMed Scopus (1065) Google Scholar, 13Inoue A. Obata K. Akagawa K. J. Biol. Chem. 1992; 267: 10613-10619Abstract Full Text PDF PubMed Google Scholar, 14Aalto M.K. Ronne H. Keränen S. EMBO J. 1993; 12: 4095-4104Crossref PubMed Scopus (343) Google Scholar) families. Snc proteins co-precipitate with the Sec9 (9Couve A. Gerst J.E. J. Biol. Chem. 1994; 269: 23391-23394Abstract Full Text PDF PubMed Google Scholar, 10Brennwald P. Kerns B. Champion K. Keränen S. Bankaitis V. Novick P. Cell. 1994; 79: 245-258Abstract Full Text PDF PubMed Scopus (311) Google Scholar) and Sso proteins (10Brennwald P. Kerns B. Champion K. Keränen S. Bankaitis V. Novick P. Cell. 1994; 79: 245-258Abstract Full Text PDF PubMed Scopus (311) Google Scholar) from yeast detergent extracts, and genetic studies suggest that Sec9 function in exocytosis may be Snc-dependent (9Couve A. Gerst J.E. J. Biol. Chem. 1994; 269: 23391-23394Abstract Full Text PDF PubMed Google Scholar). Recently, we have determined that Snc proteins are modified by the palmitoylation of unique cysteine residues located proximal to their transmembrane domains (15Couve A. Protopopov V. Gerst J.E. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 5987-5991Crossref PubMed Scopus (48) Google Scholar). Palmitoylation is required to yield the mature form of the protein but is not necessary for exocytic function. Synaptobrevin/VAMP and Snc proteins are conserved type II membrane proteins. They consist of a variable region of ∼30 residues at the amino terminus, followed by a conserved region (∼60 residues) bearing two predicted amphipathic α-helices separated by a nonhelical spacer, and a single membrane-spanning domain at the carboxyl terminus (16Dascher C. Ossig R. Gallwitz D. Schmidt H.D. Mol. Cell. Biol. 1991; 11: 872-885Crossref PubMed Scopus (278) Google Scholar,17Trimble W.S. J. Physiol. ( Paris ). 1993; 87: 107-115Crossref PubMed Scopus (28) Google Scholar). The variable and conserved regions are cytosolic, whereas the 3–4 residues adjacent to the membrane-spanning domain are likely to be intravesicular. The first helix of the conserved region of synaptobrevin/VAMP (known as helix 1 and designated here as “H1”) contains a signal for the targeting to synaptic-like microvesicles in PC12 cells (18Grote E. Hao J.C. Bennett M.K. Kelly R.B. Cell. 1995; 81: 581-589Abstract Full Text PDF PubMed Scopus (144) Google Scholar). The second helix of synaptobrevin/VAMP (known as helix 2 and designated here as “H2”) is thought to participate, along with H1, in the interaction with cognate t-SNAREs from the presynaptic membrane (e.g. SNAP-25 and syntaxins A and B), based uponin vitro studies (18Grote E. Hao J.C. Bennett M.K. Kelly R.B. Cell. 1995; 81: 581-589Abstract Full Text PDF PubMed Scopus (144) Google Scholar, 19Hayashi T. McMahon H. Yamasaki S. Binz T. Hata Y. Binz T. Sudhof T.C. Niemann H. EMBO J. 1994; 13: 5051-5061Crossref PubMed Scopus (661) Google Scholar, 20Chapman E.R. An S. Barton N. Jahn R. J. Biol. Chem. 1994; 269: 27427-27432Abstract Full Text PDF PubMed Google Scholar). These v- and t-SNARE interactions are thought to result from the formation of coiled-coil structures involving the amphipathic α-helices of the individual SNARE partners. To address whether specific regions within the synaptobrevin/VAMPs confer the SNARE-SNARE interactions that lead to exocytosis, we have used the yeast system to perform a structure/function analysis of the Snc v-SNAREs. By using genetic and biochemical assays, we demonstrate that the ability of Snc proteins to confer exocytic function in vivo corresponds to their ability to interact genetically with components of the putative fusion complex (Sec17, Sec9, and Sso2). We also demonstrate that the H1 and H2 segments together mediate the exocytic functions of these v-SNAREs. In addition, there is a distinct requirement for the membrane-spanning domain but no requirement for the intravesicular and variable regions. The latter suggests that this region is unlikely to act in a fusogenic capacity, as proposed for the amino terminus of the hemagglutinin antigen (HA), a viral fusion protein bearing similar structural motifs (reviewed in Ref. 21Stegmann T. Helenius A. Bentz J. Viral Fusion Mechanisms. CRC Press, Inc., Boca Raton, FL1993: 89-112Google Scholar). DISCUSSIONYeast Snc proteins are archetypal v-SNAREs of the late secretory pathway. These proteins share structural homology with members of the synaptobrevin/VAMP/cellubrevin family and engage in similar exocytic functions. Like their brethren, Snc proteins localize to secretory vesicles (3Protopopov V. Govindan B. Novick P. Gerst J.E. Cell. 1993; 74: 855-861Abstract Full Text PDF PubMed Scopus (250) Google Scholar), interact with t-SNAREs from the plasma membrane (9, 10, and this study), and are likely to undergo a dynamic cycle of transport to, and retrieval from, the plasma membrane.Here we have demonstrated that the conserved amphipathic α-helical region (residues 32–85) is essential for conferring the exocytic functions of these v-SNAREs. Deletions or gross substitutions in either of the predicted H1 or H2 segments result in a complete loss of function of the proteins, although they appear to reach secretory vesicles and the plasma membrane. Thus, as was predicted by experiments performed in vitro with mammalian v- and t-SNAREs (18–20, and 33), it would seem likely that both helices are required for SNARE complexing in vivo. Although we cannot say whether the individual helices prefer specific t-SNARE partners, it appears that both helices can interact either with Sec9 or Sso2. This is based upon the result that Snc proteins lacking in either H1 or H2 are unable to suppress ts defects in either Sec9 or Sso2 and inhibit the growth ofsec9 and sso2 cells at permissive temperatures. These proteins, Snc1Δ31-50 and Snc1Δ57-81, appear to exert a dominant negative effect when overexpressed. Thus, it is possible that they compete with the native proteins in undergoing SNARE partnering but are unable to confer secretory functions. Interestingly, the Snc1Δ2-27 mutant was unable to confer suppression of the ts defects in either Sec9 or Sso2, unlike Snc1 or Snc1Δ2-17, suggesting that a portion of the variable domain aids in the rescue of the mutant t-SNAREs.Work by Hayashi et al. (19Hayashi T. McMahon H. Yamasaki S. Binz T. Hata Y. Binz T. Sudhof T.C. Niemann H. EMBO J. 1994; 13: 5051-5061Crossref PubMed Scopus (661) Google Scholar) demonstrated that the domains required for SNARE complex formation in vitro correlate well with the regions predicted to form α-helices. Specifically, synaptobrevin binding to the carboxyl-terminal third of syntaxin or SNAP-25 was shown to involve the entire conserved domain (residues 27–96) (19Hayashi T. McMahon H. Yamasaki S. Binz T. Hata Y. Binz T. Sudhof T.C. Niemann H. EMBO J. 1994; 13: 5051-5061Crossref PubMed Scopus (661) Google Scholar). Earlier studies also showed binary interactions between synaptobrevin/VAMP and syntaxin (33Calakos N. Bennett M.K. Peterson K.E. Scheller R.H. Science. 1994; 263: 1146-1149Crossref PubMed Scopus (364) Google Scholar) and SNAP-25 (20Chapman E.R. An S. Barton N. Jahn R. J. Biol. Chem. 1994; 269: 27427-27432Abstract Full Text PDF PubMed Google Scholar). Deletions of both the variable region as well as the transmembrane domain of synaptobrevin were found to have no deleterious effect upon complex formation and, in contrast, were found to enhance the interaction (19Hayashi T. McMahon H. Yamasaki S. Binz T. Hata Y. Binz T. Sudhof T.C. Niemann H. EMBO J. 1994; 13: 5051-5061Crossref PubMed Scopus (661) Google Scholar). Here, using in vivo studies with mutant Snc proteins, we have shown that the structural requirements that lead to secretion are quite similar to those proposed for SNARE complex formation in vitro. In our study, however, the transmembrane domain of the protein is required for function, probably due to its membrane anchoring capacity. This is likely to be essential for the localization of Snc proteins to secretory vesicles, although it may be unnecessary for SNARE complex formation per se.Importantly, the results obtained in our in vivo study have been collaborated by a recent work in which the regions of VAMP required for conferring Ca2+-mediated exocytosis from permeabilized insulin-secreting HIT-T15 cells were examined (34Regazzi R. Sadoul K. Meda P. Kelly R.B. Halban P.A. Wollheim C.B. EMBO J. 1996; 15: 6951-6959Crossref PubMed Scopus (82) Google Scholar). By using tetanus toxin-resistant mutants of VAMP, the authors demonstrate that deletions in either of the amphipathic α-helices inhibited the ability of these mutants to restore exocytosis in tetanus toxin-treated cells. In addition, deletion of the variable region (residues 2–31) had no effect upon VAMP2 function, as shown here for Snc1. Moreover, the ability of these mutants to bind to either syntaxin Ia or SNAP-25in vitro paralleled their ability to confer exocytic functions. Thus, the results obtained in our mutational analysis of the yeast synaptobrevin/VAMP homologs are identical to those obtained for VAMP using a permeabilized cell system.It has been suggested that the amino terminus of the synaptobrevin/VAMP proteins might act in a fusogenic capacity (35Jahn R. Südhof T.C. Annu. Rev. Neurosci. 1994; 17: 219-246Crossref PubMed Scopus (335) Google Scholar) to confer bilayer fusion in the manner described for the fusion peptides of the influenza hemagglutinin antigen. This study suggests that removal of the amino-terminal variable region of a yeast v-SNARE has no deleterious effect upon its ability to confer exocytosis in snc cells. Thus, the likelihood that this region would act in an essential fusogenic capacity would seem slight. This is supported by the previous study (34Regazzi R. Sadoul K. Meda P. Kelly R.B. Halban P.A. Wollheim C.B. EMBO J. 1996; 15: 6951-6959Crossref PubMed Scopus (82) Google Scholar), yet neither work can rule out the possibility that the H1 segment alone could act in the capacity of a fusion peptide, if such a mechanism were indeed operant.Although no well defined role for the variable region has been described, other studies suggest that it may be involved in neuronal transmission. For example, peptides corresponding to the variable region of VAMP2 (but not VAMP1) inhibit acetylcholine release fromAplysia neurons (36Cornille F. Deloye F. Fournié-Zaluski M.C. Roques B.P. Poulain B. J. Biol. Chem. 1995; 270: 16826-16832Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar) and the interaction between VAMP and synaptophysin in vitro (37Washbourne P. Schiavo G. Montecucco C. Biochem. J. 1995; 305: 721-724Crossref PubMed Scopus (109) Google Scholar). Therefore, the role of the variable region could have additional functions in higher eukaryotes and may well be cell- and isoform-specific. As no synaptophysin/synaptoporin homologs have been identified in yeast thus far, we are unable to determine whether this interaction (38Calakos N. Scheller R.H. J. Biol. Chem. 1994; 269: 24534-24537Abstract Full Text PDF PubMed Google Scholar, 39Edelmann L. Hanson O.I. Chapman E.R. Jahn R. EMBO J. 1995; 14: 224-231Crossref PubMed Scopus (386) Google Scholar) is involved in constitutive exocytosis in lower eukaryotes.Another study has demonstrated the requirements for the sorting of VAMP proteins to synaptic vesicles (18Grote E. Hao J.C. Bennett M.K. Kelly R.B. Cell. 1995; 81: 581-589Abstract Full Text PDF PubMed Scopus (144) Google Scholar). There the authors show that the synaptic vesicle sorting signal for both VAMP and cellubrevin is contained within the H1 segment. Removal of residues within H1 was shown to block targeting to synaptic vesicles, whereas alanine-scanning mutagenesis revealed that certain hydrophobic residues that participate in the heptad repeat of the H1 helix are important for sorting (18Grote E. Hao J.C. Bennett M.K. Kelly R.B. Cell. 1995; 81: 581-589Abstract Full Text PDF PubMed Scopus (144) Google Scholar). These mutations are likely to decrease the overall hydrophobicity of the helix and to lessen its ability to form a coiled-coil structure.In our study, we have found that both VAMP2 and VAMP2snc90-117 are likely to target to secretory vesicles and reach the plasma membrane when expressed in yeast. Yet these proteins are deficient in conferring exocytosis, as well as in localizing in quantity to secretory vesicles, during vesiculation insec6 cells. In contrast, Snc1 and several Snc-VAMP chimeras (e.g. Snc1vamp49-116 and Snc1vamp54-93) associated primarily with the vesicles under the same conditions. Therefore, it would seem likely (although not exclusively) that certain sequences in Snc1 confer protein retrieval from the plasma membrane. In our labeling experiment, we cannot differentiate between proteins contributed from the biosynthetic pool and those recycled from the cell surface. Yet, because those proteins that bore the Snc1 H1 segment underwent a significant re-localization to the vesicle population (upon vesiculation), we are inclined to believe that the H1 of the yeast homologs, like that of synaptobrevin/VAMP, may play a role in protein sorting to vesicles. Correspondingly, those chimeras which possess the VAMP2 H1 segment associated primarily with the plasma membrane and not with vesicles. Perhaps, then, the cellular machinery required for Snc retrieval is unable to recognize efficiently the H1 domain of the mammalian isoforms. More work will be required to fully address these issues.Overall, there is a significant degree of structural as well as functional homology between the Snc and synaptobrevin/VAMP proteins, despite the fact that they confer different exocytic programs in vastly diverged systems. Some differences remain to be resolved, but the beauty of these v-SNAREs lies in their overwhelmingly simple structure and conserved set of interactions and functions. Solving the tertiary and quaternary structures of these proteins will ultimately shed light upon how they act in mediating SNARE complex formation and exocytosis. Yeast Snc proteins are archetypal v-SNAREs of the late secretory pathway. These proteins share structural homology with members of the synaptobrevin/VAMP/cellubrevin family and engage in similar exocytic functions. Like their brethren, Snc proteins localize to secretory vesicles (3Protopopov V. Govindan B. Novick P. Gerst J.E. Cell. 1993; 74: 855-861Abstract Full Text PDF PubMed Scopus (250) Google Scholar), interact with t-SNAREs from the plasma membrane (9, 10, and this study), and are likely to undergo a dynamic cycle of transport to, and retrieval from, the plasma membrane. Here we have demonstrated that the conserved amphipathic α-helical region (residues 32–85) is essential for conferring the exocytic functions of these v-SNAREs. Deletions or gross substitutions in either of the predicted H1 or H2 segments result in a complete loss of function of the proteins, although they appear to reach secretory vesicles and the plasma membrane. Thus, as was predicted by experiments performed in vitro with mammalian v- and t-SNAREs (18–20, and 33), it would seem likely that both helices are required for SNARE complexing in vivo. Although we cannot say whether the individual helices prefer specific t-SNARE partners, it appears that both helices can interact either with Sec9 or Sso2. This is based upon the result that Snc proteins lacking in either H1 or H2 are unable to suppress ts defects in either Sec9 or Sso2 and inhibit the growth ofsec9 and sso2 cells at permissive temperatures. These proteins, Snc1Δ31-50 and Snc1Δ57-81, appear to exert a dominant negative effect when overexpressed. Thus, it is possible that they compete with the native proteins in undergoing SNARE partnering but are unable to confer secretory functions. Interestingly, the Snc1Δ2-27 mutant was unable to confer suppression of the ts defects in either Sec9 or Sso2, unlike Snc1 or Snc1Δ2-17, suggesting that a portion of the variable domain aids in the rescue of the mutant t-SNAREs. Work by Hayashi et al. (19Hayashi T. McMahon H. Yamasaki S. Binz T. Hata Y. Binz T. Sudhof T.C. Niemann H. EMBO J. 1994; 13: 5051-5061Crossref PubMed Scopus (661) Google Scholar) demonstrated that the domains required for SNARE complex formation in vitro correlate well with the regions predicted to form α-helices. Specifically, synaptobrevin binding to the carboxyl-terminal third of syntaxin or SNAP-25 was shown to involve the entire conserved domain (residues 27–96) (19Hayashi T. McMahon H. Yamasaki S. Binz T. Hata Y. Binz T. Sudhof T.C. Niemann H. EMBO J. 1994; 13: 5051-5061Crossref PubMed Scopus (661) Google Scholar). Earlier studies also showed binary interactions between synaptobrevin/VAMP and syntaxin (33Calakos N. Bennett M.K. Peterson K.E. Scheller R.H. Science. 1994; 263: 1146-1149Crossref PubMed Scopus (364) Google Scholar) and SNAP-25 (20Chapman E.R. An S. Barton N. Jahn R. J. Biol. Chem. 1994; 269: 27427-27432Abstract Full Text PDF PubMed Google Scholar). Deletions of both the variable region as well as the transmembrane domain of synaptobrevin were found to have no deleterious effect upon complex formation and, in contrast, were found to enhance the interaction (19Hayashi T. McMahon H. Yamasaki S. Binz T. Hata Y. Binz T. Sudhof T.C. Niemann H. EMBO J. 1994; 13: 5051-5061Crossref PubMed Scopus (661) Google Scholar). Here, using in vivo studies with mutant Snc proteins, we have shown that the structural requirements that lead to secretion are quite similar to those proposed for SNARE complex formation in vitro. In our study, however, the transmembrane domain of the protein is required for function, probably due to its membrane anchoring capacity. This is likely to be essential for the localization of Snc proteins to secretory vesicles, although it may be unnecessary for SNARE complex formation per se. Importantly, the results obtained in our in vivo study have been collaborated by a recent work in which the regions of VAMP required for conferring Ca2+-mediated exocytosis from permeabilized insulin-secreting HIT-T15 cells were examined (34Regazzi R. Sadoul K. Meda P. Kelly R.B. Halban P.A. Wollheim C.B. EMBO J. 1996; 15: 6951-6959Crossref PubMed Scopus (82) Google Scholar). By using tetanus toxin-resistant mutants of VAMP, the authors demonstrate that deletions in either of the amphipathic α-helices inhibited the ability of these mutants to restore exocytosis in tetanus toxin-treated cells. In addition, deletion of the variable region (residues 2–31) had no effect upon VAMP2 function, as shown here for Snc1. Moreover, the ability of these mutants to bind to either syntaxin Ia or SNAP-25in vitro paralleled their ability to confer exocytic functions. Thus, the results obtained in our mutational analysis of the yeast synaptobrevin/VAMP homologs are identical to those obtained for VAMP using a permeabilized cell system. It has been suggested that the amino terminus of the synaptobrevin/VAMP proteins might act in a fusogenic capacity (35Jahn R. Südhof T.C. Annu. Rev. Neurosci. 1994; 17: 219-246Crossref PubMed Scopus (335) Google Scholar) to confer bilayer fusion in the manner described for the fusion peptides of the influenza hemagglutinin antigen. This study suggests that removal of the amino-terminal variable region of a yeast v-SNARE has no deleterious effect upon its ability to confer exocytosis in snc cells. Thus, the likelihood that this region would act in an essential fusogenic capacity would seem slight. This is supported by the previous study (34Regazzi R. Sadoul K. Meda P. Kelly R.B. Halban P.A. Wollheim C.B. EMBO J. 1996; 15: 6951-6959Crossref PubMed Scopus (82) Google Scholar), yet neither work can rule out the possibility that the H1 segment alone could act in the capacity of a fusion peptide, if such a mechanism were indeed operant. Although no well defined role for the variable region has been described, other studies suggest that it may be involved in neuronal transmission. For example, peptides corresponding to the variable region of VAMP2 (but not VAMP1) inhibit acetylcholine release fromAplysia neurons (36Cornille F. Deloye F. Fournié-Zaluski M.C. Roques B.P. Poulain B. J. Biol. Chem. 1995; 270: 16826-16832Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar) and the interaction between VAMP and synaptophysin in vitro (37Washbourne P. Schiavo G. Montecucco C. Biochem. J. 1995; 305: 721-724Crossref PubMed Scopus (109) Google Scholar). Therefore, the role of the variable region could have additional functions in higher eukaryotes and may well be cell- and isoform-specific. As no synaptophysin/synaptoporin homologs have been identified in yeast thus far, we are unable to determine whether this interaction (38Calakos N. Scheller R.H. J. Biol. Chem. 1994; 269: 24534-24537Abstract Full Text PDF PubMed Google Scholar, 39Edelmann L. Hanson O.I. Chapman E.R. Jahn R. EMBO J. 1995; 14: 224-231Crossref PubMed Scopus (386) Google Scholar) is involved in constitutive exocytosis in lower eukaryotes. Another study has demonstrated the requirements for the sorting of VAMP proteins to synaptic vesicles (18Grote E. Hao J.C. Bennett M.K. Kelly R.B. Cell. 1995; 81: 581-589Abstract Full Text PDF PubMed Scopus (144) Google Scholar). There the authors show that the synaptic vesicle sorting signal for both VAMP and cellubrevin is contained within the H1 segment. Removal of residues within H1 was shown to block targeting to synaptic vesicles, whereas alanine-scanning mutagenesis revealed that certain hydrophobic residues that participate in the heptad repeat of the H1 helix are important for sorting (18Grote E. Hao J.C. Bennett M.K. Kelly R.B. Cell. 1995; 81: 581-589Abstract Full Text PDF PubMed Scopus (144) Google Scholar). These mutations are likely to decrease the overall hydrophobicity of the helix and to lessen its ability to form a coiled-coil structure. In our study, we have found that both VAMP2 and VAMP2snc90-117 are likely to target to secretory vesicles and reach the plasma membrane when expressed in yeast. Yet these proteins are deficient in conferring exocytosis, as well as in localizing in quantity to secretory vesicles, during vesiculation insec6 cells. In contrast, Snc1 and several Snc-VAMP chimeras (e.g. Snc1vamp49-116 and Snc1vamp54-93) associated primarily with the vesicles under the same conditions. Therefore, it would seem likely (although not exclusively) that certain sequences in Snc1 confer protein retrieval from the plasma membrane. In our labeling experiment, we cannot differentiate between proteins contributed from the biosynthetic pool and those recycled from the cell surface. Yet, because those proteins that bore the Snc1 H1 segment underwent a significant re-localization to the vesicle population (upon vesiculation), we are inclined to believe that the H1 of the yeast homologs, like that of synaptobrevin/VAMP, may play a role in protein sorting to vesicles. Correspondingly, those chimeras which possess the VAMP2 H1 segment associated primarily with the plasma membrane and not with vesicles. Perhaps, then, the cellular machinery required for Snc retrieval is unable to recognize efficiently the H1 domain of the mammalian isoforms. More work will be required to fully address these issues. Overall, there is a significant degree of structural as well as functional homology between the Snc and synaptobrevin/VAMP proteins, despite the fact that they confer different exocytic programs in vastly diverged systems. Some differences remain to be resolved, but the beauty of these v-SNAREs lies in their overwhelmingly simple structure and conserved set of interactions and functions. Solving the tertiary and quaternary structures of these proteins will ultimately shed light upon how they act in mediating SNARE complex formation and exocytosis. I thank those who made contributions to this work including Vladimir Protopopov for electron microscopy and immunogold labeling; Drs. R. Scheller and L. Elferink for the generous gift of plasmids; Drs. P. Novick, H. Ronne, and S. Keränen for yeast strains; and Dr. M. Eisenstein for molecular modeling.