Commentary

SNARE proteins mediate fusion

Jason B. Bock* and Richard H. Scheller

Howard Hughes Medical Institute, Department of Molecular and Cellular Physiology, Stanford University School of Medicine, Stanford, CA 94305-5428

ipid bilayers compartmentalize eukary- tions were hypothesized to bring vesicles orientation of SNAREs within the com- Lotic cells into distinct organelles. This into close proximity with the fusion signal plex was not known. However, several compartmentalization allows for specializa- Ca2ϩ and to possibly serve as a scaffold for structural studies, culminating in the tion of diverse cellular processes, from DNA the soluble membrane trafficking factors crystal structure of the minimal interact- polymerization to zymogen proteolysis. ␣-SNAP and NSF (see below) (4) (Fig. 1A). ing domains of the 7S complex, demon- While the specific complement of proteins Concurrent with these studies, a cell- strated that the SNAREs align in a par- present in each organelle defines its func- free assay system for Golgi membrane allel fashion (10–12). This finding im- tion, there is a dynamic flux between these fusion was developed. With this system plies that in a trans-SNARE state the organelles. The selective transport of pro- two proteins previously discovered as es- vesicle and target membranes will be teins between organelles is the central pro- sential for secretion in yeast, ␣-SNAP brought into close apposition (Fig. 1C). cess in the organization of membrane com- (Sec17p) and NSF (Sec18p), were charac- This realization sparked the hypothesis partments. This process is largely mediated terized from mammalian systems. It sub- that the energy released from formation by the budding of transport vesicles from a sequently was found that these proteins of the stable complex, in concert with the donor compartment followed by the vec- interacted with the SNAREs syntaxin, physical bringing together of lipid bilay- toral trafficking to and fusion with an ac- SNAP-25, and VAMP (vesicle-associated ers, might mediate the membrane fusion ceptor compartment. Over the last several ) (5). These three itself (Fig. 1D). years considerable effort has been exerted SNARE proteins were found to form a The ultimate test of this model is to to uncover the molecular mechanisms un- stable complex termed 7S for its size mi- faithfully reconstitute the process from pu- derlying this process. A set of protein fam- gration on glycerol gradients (6) (Fig. 1B). rified components. In an attempt to accom- ilies known as SNAREs [soluble N-ethyl- The central role of these proteins in a late plish this goal, a liposome fusion system maleimide-sensitive factor (NSF) attach- stage of vesicle trafficking was solidified dependent on SNARE proteins was devel- ment protein receptors] is at the crux of by several distinct genetic, biochemical, oped (13). Syntaxin and SNAP-25 were vesicle docking and͞or fusion. Members of and cell-free assay systems. When it was reconstituted in one pool of liposomes, these families reside on transport vesicles found that these SNAREs were proto- whereas VAMP was reconstituted in lipo- COMMENTARY (v-SNARES) and on target membranes (t- types of protein families whose members somes containing fluorophore-bound lipids, SNAREs) and bind to each other. While were distributed throughout the secretory which at high concentrations allow fluores- previously implicated in conferring specific- pathway (7), the hypothesis arose that cent resonance energy transfer to occur. ity to vesicle trafficking, more recent studies these proteins mediated the specificity of Fusion can be measured as the concentra- have demonstrated that these proteins also vesicle trafficking by defining membranes tion of fluorescent lipids changes as bilayers may mediate the fusion process itself. In two that were compatible for docking and mix. This system demonstrated that papers in this issue of PNAS, Rothman and fusion. With the understanding that the SNAREs could mediate lipid mixing, and colleagues (1, 2) demonstrate that lipo- cytosolic proteins ␣-SNAP and NSF could the authors concluded that SNAREs indeed somes reconstituted with only v- and t- hydrolyze ATP and dissociate SNARE comprise a minimal fusion machinery. SNAREs can fuse and mix luminal contents complexes, the specific docking hypothe- However, the slow kinetics and lack of an with reasonably physiological kinetics. sis was further extended such that disso- assay for luminal content mixing left linger- These results further implicate SNAREs as ciation of the complex led to bilayer fu- ing doubts about the physiological relevance the core fusion machinery for intracellular sion. of this claim. vesicle trafficking. This hypothesis then was modified by a To further address whether SNARE In the 1960s and 1970s with the advent of series of experiments using an in vitro yeast complex formation leads directly to fusion, electron microscopy, the hypothesis of a vacuole fusion assay. One conclusion from a novel twist on a well-established cell per- secretory pathway was developed, in which these experiments was that ␣-SNAP and meabilized assay system was developed (14). proteins to be secreted initially are synthe- NSF act significantly before the fusion The power of the new approach came from sized in the endoplasmic reticulum, moved event, likely in a SNARE-priming role to the use of botulinum neurotoxins, which to and then through the Golgi apparatus, dissociate cis-SNARE complexes (8). Fol- specifically cleave members of the core 7S packaged into secretory granules, and fi- low-up experiments also confirmed that a v- complex. By using these toxins in conjunc- nally secreted (3). Further it was proposed and t-SNARE are required on opposite tion with rescuing peptides, those authors that these various trafficking steps were membranes before fusion (9). Thus while found, in contrast to the conclusions drawn mediated by transport vesicles. A molecular redefining the role of ␣-SNAP and NSF in from the vacuole fusion studies, that stable understanding of the role of SNAREs in vesicle trafficking as primarily SNARE trans-SNARE complexes form only after this process began in the early 1990s with primers, these experiments underscored in Ca2ϩ is present. By making mutations in studies on the mammalian presynaptic yet another system the central role of trans- residues present in the core of the SNARE nerve terminal. Immunoprecipitations with SNARE complexes. complex, they also demonstrated that the an against the synaptic vesicle pro- It is the nature of these trans-SNARE tein synaptotagmin yielded the protein syn- pairs that has been the focus of investi- taxin (a t-SNARE), which also was found to gation over the last several years. At the See companion articles on pages 12565 and 12571. ϩ interact with Ca2 channels. These interac- time of the initial docking hypothesis, the *To whom reprint requests should be addressed.

PNAS ͉ October 26, 1999 ͉ vol. 96 ͉ no. 22 ͉ 12227–12229 Downloaded by guest on September 28, 2021 Fig. 1. Historical hypothesized role of SNAREs in membrane trafficking. (A) The synaptic vesicle protein synaptotagmin was found to interact with the plasma membrane protein syntaxin, which also interacts with Caϩϩ channels. (B) Trans-SNARE complexes dock vesicles and provide a scaffold for the assembly of the fusion apparatus. (C) Parallel alignment of SNARES forces membranes into close apposition. (D) Current hypothesis that SNARE complex formation is regulated by proteins such as n-sec1. Membrane fusion occurs in concert with trans-SNARE complex formation. ␣-SNAP and NSF then break apart cis-SNARE complexes to reset the system for another round of fusion.

stability of the SNARE complex correlates min. If the liposomes are preincubated at The conclusion from these papers is that with the extent of membrane fusion. In 4°C for 16 hr before incubation at 37°C, the SNAREs in an isolated system can fuse addition, both SNARE complex formation initial kinetics are significantly faster, with membranes, which contrasts with recent and Ca2ϩ triggering are closely coupled in the first round of fusion complete after 7 data using the vacuolar fusion assay. Exper- time to . Thus, these results min. However, after 15 min this reaction iments from this system suggest that while support the model that membrane fusion follows the same, slower kinetics as the SNARE complex formation is an essential occurs in concert with SNARE complex reaction without preincubation. The au- intermediate in vesicle docking and fusion, formation, and that stable, irreversible thors propose the preincubation allows trans-SNARE pairing can be uncoupled SNARE complex only exists in cis as the trans-SNARE complexes to form, which from fusion itself (16). Thus these authors product of the fusion reaction itself (15). permits a rapid first round of fusion. If the believe that trans-SNARE complex forma- Two papers in this issue of PNAS further minimal SNARE interacting domain of syn- tion sends a signal to release calcium from taxin, the H3 domain, is used in place of the address the mechanism of membrane fusion the yeast vacuole, which in turn signals an full-length protein with no preincubation, by extending the previously developed lipo- as-yet-uncharacterized fusion machinery the kinetics are linear throughout the 2-hr some fusion assay. Nickel et al. (1) use a experiment, with the first-half round of fu- (16, 17). Obviously in the reconstituted li- novel complementary oligonucleotide assay sion occurring in 10 min. The N-terminal posome assay there are no proteins that system to clearly demonstrate that in the domain, which was cleaved away from syn- could be downstream of SNAREs to medi- SNARE-mediated liposome fusion system taxin, has been previously shown to retard ate fusion. So either the SNARE-mediated the contents of v- and t-SNARE vacuoles the binding of SNAP-25 to syntaxin, which liposome fusion is not translatable to what merge, thus the SNAREs mediate complete is the rate-limiting step for SNARE com- occurs in vivo or the vacuole studies are not bilayer fusion, rather than just hemifusion. plex formation. However, in this system interpreted correctly. Although other pro- In the second paper (2), when full-length syntaxin and SNAP-25 are purified and teins such as NSF and annexins have been SNAREs are incorporated into liposomes reconstituted into liposomes as a complex. found to be fusogenic in a liposome system, and then incubated at 37°C, fusion occurs Thus the N terminus of syntaxin must have it seems unlikely that SNAREs, which are with linear kinetics for at least 2 hr with the an additional regulatory role in complex present at the correct time and place for first-half round of fusion complete in 40 formation or fusion itself. fusion in both cell-free and cell-permeabi-

12228 ͉ www.pnas.org Bock and Scheller Downloaded by guest on September 28, 2021 against syntaxin 1 were able to precipitate SNAP-25 and VAMP, whereas syntaxin 5, present on the intermediate compartment and Golgi, was found to in- teract with endoplasmic reticulum and Golgi v-SNAREs. To examine these inter- actions in a more isolated system, in vitro binding studies using recombinant proteins were initiated (18, 19). In those experiments recombinant members of the syntaxin, VAMP, and SNAP-25 families were as- sayed for complex formation and complex stability. Surprisingly, almost any combina- tion of SNAREs was able to form a complex that was SDS-resistant and had similar ther- mal stability. These results allow for three possibilities. One is that SNAREs are not involved in determining specificity of vesicle trafficking. The second is that thermal sta- bility may not be an accurate barometer of intracellular events. Perhaps there is a sig- nificant kinetic difference in the formation of these SNARE complexes, which would be overlooked by a thermodynamic mea- surement. Third, the protein–protein inter- actions of the SNAREs themselves may not be specific in vitro, yet in concert with other layers of protein interactions, the overall fidelity of vesicle trafficking can be estab- lished. Thus the information for the speci- ficity of membrane trafficking may not re- side in the ability of the SNAREs to pair with each other, while in vivo specific pairing may indeed occur. It may well be that the translocation of vesicles from donor to ac- Fig. 2. Localization of SNARE proteins within the secretory pathway of a typical eukaryotic cell. Members COMMENTARY of the syntaxin family are colored red. SNAP-25 family members are green, and VAMP family members are ceptor sites via specific cytoskeletal and blue. N, nucleus; ER, endoplasmic reticulum; SER, smooth ER; IC, intermediate compartment; TGN, motor protein interactions, as well as rab- trans-Golgi network; V, vesicles; DCV, dense core vesicles; EE, early endosome; LE, late endosome; L, effector interactions, leads to specific ; CV, clathrin-coated vesicles. SNARE complex formation, and thus fidel- ity in vesicle trafficking. In summary the field of intracellular ves- lized assays, are clearly necessary for a late a relatively large number of SNAREs to icle trafficking has become fascinatingly in- step in vesicle trafficking, and are sufficient distinguish between the various vesicle traf- tricate from its beginnings with morpholog- for fusion in vitro, do not carry out that ficking steps. Second, SNAREs should have ical descriptions from electron micrographs. function in vivo. distinct subcellular localizations. It appears With the recent convergence of cell-free, Thus while converging lines of evidence these two criteria have been met, with many cell-permeabilized, and reconstituted in implicate SNAREs as the core fusion ma- SNARES already identified and localized to vitro assay systems, it appears we are ap- chinery, other recent studies address the specific subcellular locations (Fig. 2). proaching an understanding of the central role of SNAREs in underpinning the spec- The final and most important criterion is players in vesicle trafficking. The next chal- ificity of vesicle trafficking. One would ex- that SNAREs form selective complexes. lenge lies in decoding the cross-talk between pect three criteria to be fulfilled if SNAREs Initial immunoprecipitations of SNAREs these central players and the bounty of other perform this function. First, there should be appeared to isolate specific sets of proteins. proteins associated with vesicle trafficking.

1. Nickel, W., Weber, T., McNew, J. A., Parlati, F., 409–418. Westermann, B., Gmachl, M., Parlati, F., So¨llner, So¨llner, T. H. & Rothman, J. E. (1999) Proc. Natl. 7. Hardwick, K. G. & Pelham, H. R. (1992) J. Cell. T. H. & Rothman, J. E. (1998) Cell 92, 759–772. Acad. Sci. USA 96, 12571–12576. Biol. 119, 513–521. 14. Chen, Y. A., Scales, S. J., Patel, S. M., Doung, 2. Parlati, F., Weber, T., McNew, J. A., Westermann, 8. Mayer, A., Wickner, W. & Haas, A. (1996) Cell 85, Y. C. & Scheller, R. H. (1999) Cell 97, 165–174. B., So¨llner, T. H. & Rothman, J. E. (1999) Proc. 83–94. 15. Xu, T., Binz, T., Niemann, H. & Neher, E. (1998) Natl. Acad. Sci. USA 96, 12565–12570. 9. Nichols, B. J., Ungermann, C., Pelham, H. R., Nat. Neurosci. 1, 192–200. 3. Palade, G. (1975) Science 189, 347–358. Wickner, W. T. & Haas, A. (1997) Nature (Lon- 16. Ungermann, C., Sato, K. & Wickner, W. (1998) Nature (London) 396, 543–548. 4. Bennett, M. K., Calakos, N. & Scheller, R. H. don) 387, 199–202. 17. Peters, C. & Mayer, A. (1998) Nature (London) Science 257, 19, (1992) 255–259. 10. Lin, R. C. & Scheller, R. H. (1997) Neuron 396, 575–580. 5. So¨llner, T., Whiteheart, S. W., Brunner, M., Er- 1087–1094. 18. Yang, B., Gonzalez, L., Jr., Prekeris, R., Steeg- djument-Bromage, H., Geromanos, S., Tempst, P. 11. Hanson, P. I., Roth, R., Morisaki, H., Jahn, R. & maier, M., Advani, R. J. & Scheller, R. H. (1999) & Rothman, J. E. (1993) Nature (London) 362, Heuser, J. E. (1997) Cell 90, 523–535. J. Biol. Chem. 274, 5649–5653. 318–324. 12. Sutton, R. B., Fasshauer, D., Jahn, R. & Brunger, 19. Fasshauer, D., Antonin, W., Margittai, M., Pabst, 6. So¨llner, T., Bennett, M. K., Whiteheart, S. W., A. T. (1998) Nature (London) 395, 347–353. S. & Jahn, R. (1999) J. Biol. Chem. 274, 15440– Scheller, R. H. & Rothman, J. E. (1993) Cell 75, 13. Weber, T., Zemelman, B. V., McNew, J. A., 15446.

Bock and Scheller PNAS ͉ October 26, 1999 ͉ vol. 96 ͉ no. 22 ͉ 12229 Downloaded by guest on September 28, 2021