trans-SNARE complex assembly and yeast vacuole membrane fusion

Abstract

Regulated membrane fusion is essential for cell compartmentation. Intracellular fusion requires Rab-family GTPases, Rab-effector complexes, Sec1/Munc18 proteins, key regulatory lipids, and SNARE proteins (1). Most SNARE proteins are membrane-bound by a C-terminal apolar region or by a prenyl tail. SNARE proteins have membrane-proximal heptad repeat sequences, termed the SNARE motif. These proteins associate in alpha-helical, coiled-coil bundles as heteromeric SNARE complexes (2). Three glutamine residues and one arginine residue at the center of the four associated SNARE motifs, termed the zero layer, have a conserved role in SNARE function and categorize each SNARE as either Q- or R-SNARE (3). SNARE complexes are in cis when their apolar anchors are all in the same membrane bilayer or in trans when these anchors are in closely apposed membranes, poised for fusion (4). SNAREs and SNARE complexes associate with other factors, including Sec1/Munc18 proteins (5, 6), Ca^{-sensors such as synaptotagmin (}7–9), and others (10–12), for fusion. trans-SNARE pairs may promote membrane fusion by inducing local physical stress on the bilayer (13), destabilizing bilayer structure through their slanted transmembrane domains (14), or enriching membrane destabilizing lipids such as diacylglycerol at the fusion site (15). Despite their importance, there have been few reports of direct physical assay of trans-SNARE pairs (4, 16, 17).

We study membrane fusion with vacuoles from Saccharomyces cerevisiae. Vacuole fusion requires the Rab GTPase Ypt7p, its hexameric effector HOPS (homotypic fusion and vacuole protein sorting) complex, three Q-SNAREs (Vam3p, Vti1p, and Vam7p), one R-SNARE (Nyv1p), and key “regulatory” lipids (ergosterol, diacylglycerol, and phosphoinositides). At the start of in vitro vacuole fusion reactions, the chaperones Sec18p (yeast NSF) and Sec17p (α-SNAP) disassemble cis-SNARE complexes, freeing the SNAREs for association in trans. Vacuoles tether, supported by Ypt7p (18) and HOPS (19), and are drawn together until each pair of tethered vacuoles has disk-like regions of “boundary” membrane that are tightly apposed (20). Each of the key fusion factors (Ypt7p, HOPS, the SNAREs, and the regulatory lipids) become enriched at a ring-shaped microdomain surrounding the boundary membrane, termed the vertex ring (15, 20). SNARE pairing follows some time later and leads to complete fusion.

Yeast vacuoles isolated from vam3Δ or nyv1Δ strains cannot undergo homotypic fusion (21). However, vacuoles from vam3Δ strains fuse slowly with vacuoles from nyv1Δ strains, suggesting that these SNAREs pair in trans (21). Assays of the physical association of Vam3p and Nyv1p from vacuoles from nyv1Δ and vam3Δ strains, respectively, offered a direct assay of trans-SNARE pairing (4). However, these studies did not adequately distinguish SNARE pairs that were truly in trans from those that may have been trans but had been rendered cis by fusion, or from those that were formed de novo in cis after fusion. Engineering epitope tags on different SNAREs may permit the distinction between cis and trans complexes with vacuoles that are otherwise wild type in their fusion activity (17). We now report an assay of trans-SNARE complex formation with vacuoles that undergo normal rates and extents of fusion. Each vacuolar constituent that is needed for vertex ring enrichment is needed for trans-SNARE pairing, but either the phosphoinositide ligand myristoylated alanine-rich C kinase substrate effector domain (MED) or an excess of the SNARE chaperone Sec17p permit trans-SNARE complex formation while blocking the progression to fusion.

Acknowledgments

Abbreviations

Footnotes