SNARE-Mediated Lipid Mixing Depends on the Physical Stateof the VesiclesXiaocheng Chen,* Demet Aracx,* Tzu-Ming Wang,yzChristopher J. Gilpin,§Joshua Zimmerberg,{and Josep Rizo**Departments of Biochemistry and Pharmacology,yDepartment of Physiology,zCenter for Basic Neuroscience, and§Department of CellBiology, University of Texas Southwestern Medical Center, Dallas, Texas 75390; and{Laboratory of Cellular and Molecular Biophysics,National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland 20892ABSTRACT Reconstitution experiments have suggested that N-ethylmaleimide sensitive factor attachment protein receptor(SNARE) proteins constitute a minimal membrane fusion machinery but have yielded contradictory results, and it is unclearwhether the mechanism of membrane merger is related to the stalk mechanism that underlies physiological membrane fusion.Here we show that reconstitution of solubilized neuronal SNAREs into preformed 100 nm liposomes (direct method) yieldsproteoliposomes with more homogeneous sizes and protein densities than the standard reconstitution method involvingdetergent cosolubilization of proteins and lipids. Standard reconstitutions yield slow but efficient lipid mixing at high proteindensities and variable amounts of lipid mixing at moderate protein densities. However, the larger, more homogenousproteoliposomes prepared by the direct method yield almost no lipid mixing at moderate protein densities. These resultssuggest that the lipid mixing observed for standard reconstitutions is dominated by the physical state of the membrane, perhapsdue to populations of small vesicles (or micelles) with high protein densities and curvature stress created upon reconstitution.Accordingly, changing membrane spontaneous curvature by adding lysophospholipids inhibits the lipid mixing observed forstandard reconstitutions. Our data indicate that the lipid mixing caused by high SNARE densities and/or curvature stress occursby a stalk mechanism resembling the mechanism of fusion between biological membranes, but the neuronal SNAREs arelargely unable to induce lipid mixing at physiological protein densities and limited curvature stress.INTRODUCTIONFusion of two membranes into a single membrane is a criticalevent for a wide variety of biological processes, includingintracellular transport, fertilization, and viral entry into a hostcell. Multiple types of functional and genetic evidence haveshown that N-ethylmaleimide sensitive factor attachmentprotein receptors (SNAREs) play a key role in all steps of thesecretory and endocytic pathways in eukaryotic cells (reviewedin (1–5)). Studies of the synaptic vesicle SNARE synaptobrevin/vesicle associated membrane protein and the presynapticplasma membrane SNAREs syntaxin and SNAP-25 showedthat they form a tight complex known as the SNARE complexthrough coiled-coil sequences called SNARE motifs (6–10).The discovery that this complex involves a parallel interac-tion between the SNARE motifs of synaptobrevin andsyntaxin, which are adjacent to their transmembrane (TM)regions, showed that assembly of the SNARE complex mustbring the synaptic vesicle and plasma membranes togetherand led to the hypothesis that SNARE complex formationmay provide the energy for membrane fusion (11,12). Theneuronal SNARE complex was later shown to consist of abundle of four parallel a-helices (two from SNAP-25 andone each from syntaxin and synaptobrevin) (13,14), andcharacterization of SNAREs from diverse membrane com-partments (e.g., Antonin et al. (15)) has indicated that allSNARE complexes adopt similar four-helix bundle struc-tures. In addition, reconstitution experiments revealed lipidmixing between liposomes containing synaptobrevin andliposomes containing syntaxin/SNAP-25 (16), and similarresults were obtained with yeast SNAREs involved indifferent membrane traffic processes (17).These and other observations have led to a widespreadmodel whereby the SNAREs constitute a minimal machineryfor intracellular membrane fusion, but the validity of thismodel is still under intense debate (18–20). A key problemwith this model is that it does not account for the strictrequirement of Sec1/Munc18-1 homologs for all types of in-tracellular membrane fusion (reviewed in Rizo and Sudhof(3)). Moreover, additional proteins appear to act downstreamof SNAREs in fusion of egg cortical vesicles (21). The de-bate over the precise function of the SNAREs also arises inpart because of the intrinsic in vitro nature of the reconsti-tution experiments, which hinders conclusive demonstrationof protein function in the absence of strict correlations within vivo data. In addition, the initial reconstitution experi-ments with neuronal SNAREs (16) were performed withliposomes containing an exceedingly high amount of synapto-brevin, and the observed rate of lipid mixing (minutes-hours)was very slow compared to the timescale of neurotransmitterrelease (,0.5 ms). Whereas liposome fusion mediated byneuronal SNAREs was later observed at lower protein/lipidratios (22), other experiments using different reconstitutionschemes revealed no lipid mixing, which was attributed toSubmitted July 24, 2005, and accepted for publication November 28, 2005.Address reprint requests to Josep Rizo, Tel.: 214-645-6360; Fax: 214-645-6353; E-mail: [email protected].Ó 2006 by the Biophysical Society0006-3495/06/03/2062/13 $2.00doi: 10.1529/biophysj.105.0714152062 Biophysical Journal Volume 90 March 2006 2062–2074membrane sequestration of part of the synaptobrevin SNAREmotif (23,24). Note also that when SNARE-mediated lipidmixing is observed, the number of rounds of fusion cal-culated (see Parlati et al. (25)) is significantly lower thanexpected from the stoichiometry of the reaction mixture. It isalso unclear why SNARE-induced fusion of individualvesicles with planar bilayers occurs at much faster timescales(,1 s) than the overall rate of liposome-liposome fusion andwhy only a small fraction of the vesicles fused in these ex-periments (26). Particularly worrisome was also the obser-vation that significant lipid mixing was still observed whenlong linkers that can span 100 A˚were introduced betweenthe SNARE motifs and TM regions of syntaxin andsynaptobrevin (27), since the linkers would have been ex-pected to uncouple SNARE complex assembly from fusionand it seems highly unlikely that these mutants would befunctional in vivo (28).Despite these caveats and unanswered