JBC Advanced Glycation Endproducts

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M500421200 on February 28, 2005

J. Biol. Chem., Vol. 280, Issue 17, 16754-16762, April 29, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/17/16754    most recent
M500421200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Starai, V. J.
Right arrow Articles by Wickner, W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Starai, V. J.
Right arrow Articles by Wickner, W.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Ion Regulation of Homotypic Vacuole Fusion in Saccharomyces cerevisiae*

Vincent J. Starai{ddagger}, Naomi Thorngren, Rutilio A. Fratti§, and William Wickner

From the Department of Biochemistry, Dartmouth Medical School, Hanover, New Hampshire 03755-3844

Received for publication, January 12, 2005 , and in revised form, February 22, 2005.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Biological membrane fusion employs divalent cations as protein cofactors or as signaling ligands. For example, Mg2+ is a cofactor for the N-ethylmaleimide-sensitive factor (NSF) ATPase, and the Ca2+ signal from neuronal membrane depolarization is required for synaptotagmin activation. Divalent cations also regulate liposome fusion, but the role of such ion interactions with lipid bilayers in Rab- and soluble NSF attachment protein receptor (SNARE)-dependent biological membrane fusion is less clear. Yeast vacuole fusion requires Mg2+ for Sec18p ATPase activity, and vacuole docking triggers an efflux of luminal Ca2+. We now report distinct reaction conditions where divalent or monovalent ions interchangeably regulate Rab- and SNARE-dependent vacuole fusion. In reactions with 5 mM Mg2+, other free divalent ions are not needed. Reactions containing low Mg2+ concentrations are strongly inhibited by the rapid Ca2+ chelator BAPTA. However, addition of the soluble SNARE Vam7p relieves BAPTA inhibition as effectively as Ca2+ or Mg2+, suggesting that Ca2+ does not perform a unique signaling function. When the need for Mg2+, ATP, and Sec18p for fusion is bypassed through the addition of Vam7p, vacuole fusion does not require any appreciable free divalent cations and can even be stimulated by their chelators. The similarity of these findings to those with liposomes, and the higher ion specificity of the regulation of proteins, suggests a working model in which ion interactions with bilayer lipids permit Rab- and SNARE-dependent membrane fusion.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Eukaryotic subcellular compartmentation requires selective membrane fusion. This fusion depends on specific lipids and conserved proteins, including SNAREs1 and their chaperones, Rab family GTPases, Rab effectors, and divalent cations (1). For example, membrane fusion at the synapse is triggered by an afferent wave of electrical depolarization, which opens a voltage-gated Ca2+ channel. The Ca2+ that enters binds to intracellular receptors, most notably the C2 domains of synaptotagmin, triggering a conformational change, which alters synaptotagmin interactions with SNAREs and with the apolar domain of the membrane bilayer (2). Although Ca2+ clearly regulates synaptic membrane fusion, in other membrane fusion events the Ca2+ channel, the trigger for Ca2+ flux, the Ca2+ receptor, and the modes of action of the receptors have received less study.

The vacuole (lysosome) of Saccharomyces cerevisiae is the major repository of cellular Ca2+ (3). Calcium is pumped into the vacuole by Pmc1p, an ATP-driven Ca2+ transporter (4), and by Vcx1p, a Ca2+/H+ exchanger (5). Vacuole homotypic fusion occurs in three stages: ATP-dependent priming, docking, and finally bilayer fusion and content mixing (6). Docking is complex and requires Ypt7p (a Rab family GTPase), the HOPS (homotypic fusion and vacuole protein sorting)/Vps Class C complex (a Ypt7p effector that also binds to SNAREs), and the SNARE proteins Vam7p, Vam3p, Vti1p, and Nyv1p. The last stage of docking, the pairing of SNAREs in trans, triggers a dramatic release of luminal Ca2+ from the vacuole (7, 8). Experiments demonstrating the sensitivity of vacuole fusion to the Ca2+ chelator BAPTA, the relief of BAPTA sensitivity by added Ca2+, and the docking-dependent release of Ca2+ from the vacuole lumen (8, 9) have suggested that Ca2+ may signal successful docking and initiate the terminal stage of membrane fusion. Vacuole fusion requires the calcium-binding protein calmodulin (7), strengthening the connection between free Ca2+ and the downstream processes leading to vacuole fusion.

Several observations now lead us to re-examine the role of free Ca2+, and other ions, in vacuole fusion. Although the Ca2+ chelator BAPTA inhibits a late stage of vacuole fusion (10), BAPTA also inhibits the membrane association of two crucial peripheral membrane proteins, Vam7p and the HOPS/Vps class C complex, which do not have Ca2+ binding motifs. BAPTA, a tetravalent ion, may release these proteins from the membrane, and inhibit vacuole fusion, through its contribution to ionic strength. Vacuole fusion is commonly measured by an assay coupled to the activity of the zinc-metalloenzyme Pho8p, which is sensitive to divalent ion chelators (11). We now report that fusion assays with 5 mM Mg2+ show little sensitivity to BAPTA when they are adjusted to a constant ionic strength and are performed in the presence of the heavy metal ion chelator TPEN (to obviate the effect of Zn2+ chelation by BAPTA). At low Mg2+, the inhibition by BAPTA is not simply due to its ionic strength contribution, yet BAPTA inhibition can be bypassed by altering KCl, Mg2+, and Vam7p concentrations. Finally, under conditions that bypass the need for ATP or Sec18p, Ypt7p- and SNARE-dependent vacuole fusion requires moderate (125–250 mM) KCl but does not require appreciable free divalent cations; at certain KCl concentrations, fusion can even be stimulated by divalent ion chelators. We suggest a working model in which Mg2+, Ca2+, and monovalent ions interact with the lipid bilayer to regulate vacuole fusion.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Yeast Strains and Genetic Modifications—Vacuoles were purified from S. cerevisiae strains BJ3505 (MAT{alpha} ura3–52 trp1-{Delta}101 his3-{Delta}200 lys2–801 gal2 (gal3) can1 prb1-{Delta}1.6R pep4::HIS3) (12) and DKY6281 (MAT{alpha} ura3–52 leu2–3,112 trp1-{Delta}901 his3-{Delta}200 lys2–801 suc2-{Delta}9 pho8::TRP1) (13) for fusion assays. Strains CRY1 (MATa ade2–1oc can1–100 his3–11,15 leu2–3,112 trp1–1 ura3–1), JGY149 (MATa ade2–1oc can1–100 cmd1–6 his3–11,15 leu2–3,112 trp1–1 ura3–1) (14), and JGY041 (MATa ade2–1oc can1–100 cmd1–3 his3–11,15 leu2–3,112 trp1–1 ura3–1) (15) (generous gifts from Dr. Trisha Davis, University of Washington, Seattle, WA) bearing alleles of calmodulin that do not appreciably bind Ca2+, were used to generate pep4{Delta} and pho8{Delta} derivatives for studies of vacuole fusion. Briefly, the pep4::HIS3 allele and the pho8::TRP1 allele were PCR-amplified with flanking chromosomal sequence from BJ3505 and DKY6281, respectively. These fragments were transformed into CRY1, JGY041, and JGY149 using the standard LiAc/ss-DNA/PEG transformation method (16), generating VSY3 (=CRY1 pep4::HIS3), VSY4 (=CRY1 pho8::TRP1), VSY5 (=JGY041 pep4::HIS3), VSY6 (=JGY041 pho8::TRP1), VSY7(=JGY149 pep4::HIS3), and VSY8 (=JGY149 pho8::TRP1). Strain constructs were confirmed by PCR analysis, and maintenance of the correct cmd1 allele in each derivative was confirmed by DNA sequencing.

Reagents—Anti-Sec17p, anti-Sec18p (17), anti-Vam3p (18), anti-Vam7p (8), anti-Ypt7p (19), anti-Vps33p (20), and anti-Vps41p (21) were prepared as described and equilibrated in PS buffer (20 mM PIPES-KOH, pH 6.8, 200 mM sorbitol). His6-Sec17p (21), Gdi1p (22), His6-Gyp1–46p (18), GST-Rdi1p (19), C1b domain (23), FYVE domain (24), and myristoylated alanine-rich C kinase substrate effector domain (25) were purified as described. U73122 [GenBank] (Calbiochem), W7 (Calbiochem), 3-nitrocoumarin (a generous gift from Dr. Enzo Martegani, Università di Milano, Italy), and Filipin III (Sigma) were dissolved in Me2SO. TPEN (Molecular Probes) was dissolved in ethanol. BAPTA (Sigma) was freshly dissolved in PS buffer prior to each use.

Vacuole Isolation and in Vitro Fusion Assay Conditions—Vacuoles were isolated as described previously (13). Vacuoles were purified from ade2–1oc strains VSY3, VSY4, VSY5, VSY6, VSY7, and VSY8 after growth in YPD with 0.002% (w/v) adenine hemisulfate (from a 2% stock dissolved in Me2SO).

In vitro fusion reactions contained 3 µg of pep4{Delta} vacuoles (from BJ3505) and 3 µg of pho8{Delta} vacuoles (from DKY6281). The following three fusion reaction conditions were used in this work: (a) high Mg2+ fusion reactions contained 20 mM PIPES-KOH, pH 6.8, 200 mM sorbitol, 125 mM KCl, 6 mM MgCl2, 1 mM ATP (Amersham Biosciences), 1 mg/ml creatine kinase (Roche Applied Science), 29 mM creatine phosphate (Roche Applied Science), 10 µM coenzyme A (Sigma), and 330 nM purified Pbi2p (IB2) (26); (b) low Mg2+ fusion reactions contained 10 mM PIPES-KOH, pH 6.8, 200 mM sorbitol, 125 mM KCl, 0.5 mM MnCl2, 0.5 mM MgCl2, 0.5 mM ATP, 0.5 mg/ml creatine kinase, 14.5 mM creatine phosphate, 10 µM coenzyme A, and 330 nM purified IB2; and (c) bypass fusion, as previously described (21), can occur without ATP when recombinant Vam7p is supplied. Bypass reactions contained 10 mM PIPES-KOH, pH 6.8, 200 mM sorbitol, 125 mM KCl, 10 µM coenzyme A, 2.8 µM recombinant Vam7p, and 330 nM purified IB2.

All reaction components except the vacuoles were mixed on ice. Vacuoles purified from pep4{Delta} and pho8{Delta} strains were premixed in equal amounts on ice and were added last to each reaction (6 µg of total per reaction). Reactions were incubated at 27 °C for 90 min unless otherwise noted.

After fusion, reactions were assayed for alkaline phosphatase activity (27) with 10 mM CaCl2 included in the assay solution. Samples were centrifuged (16,000 x g, 2 min) just prior to spectrophotometric measurement to remove any precipitates. Fusion units are micromolar of para-nitrophenyl phosphate min–1 µg of vacuole–1. Protein concentrations were determined with a Bio-Rad protein assay kit (Bio-Rad), using bovine serum albumin as standard.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
We assay the fusion of yeast vacuoles that are purified from two strains, one that accumulates catalytically inactive proalkaline phosphatase due to the absence of vacuole luminal proteases, and the other that has the normal proteases but is deleted for the PHO8 phosphatase gene (13). Upon vacuole fusion, the proteases gain access to the proPho8p and cleave it to the catalytically active form, Pho8p. This active form can be assayed colorimetrically and is a quantitative measurement of the extent of vacuole fusion. Pho8p has tightly bound Zn2+ and Cu2+, although only the Zn2+ is needed for phosphatase activity (11). The roles of divalent cations in fusion can be explored with chelators such as EDTA, BAPTA, and TPEN, each with their characteristic affinities for divalent cations (Table I). BAPTA inhibits in vitro vacuole fusion (Fig. 1; compare squares to other symbols on the ordinate). Although added Ca2+ did not stimulate fusion, its addition relieved BAPTA inhibition (Fig. 1), suggesting that the BAPTA-mediated inhibition of vacuole fusion may reflect its calcium chelation properties. When BAPTA and calcium were equimolar, however, a modest amount of fusion inhibition remained. To explore the basis of this Ca2+-indpendent inhibition, we sought BAPTA-sensitive aspects of vacuole fusion.


View this table:
[in this window]
[in a new window]
  		TABLE I
Affinity of chelators for divalent cations

All Kd estimations were calculated with WEBMAXC STANDARD (8/26/2004, CMC1002.TCM constants) at T = 27 °C, pH = 6.8, ionic level = 0.125 N, unless otherwise noted. The software is available at www.stanford.edu/~cpatton/maxc.html.

 



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 1.
BAPTA inhibition of vacuole fusion is reversed by Ca2+. High Mg2+ fusion reactions (see "Experimental Procedures") were incubated at 27 °C for 90 min without chelator (squares) or with 1 mM (triangles), 3 mM (circles), or 5 mM (diamonds) BAPTA and the indicated concentrations of CaCl2 (abscissa).

 
BAPTA Releases Peripherally Bound Proteins from the Vacuolar Membrane—BAPTA can affect biological systems in ways other than calcium chelation. For example, BAPTA has been shown to depolymerize microtubules (28) and to bind to proteins and phospholipids (29, 30). To study additional effects of BAPTA on the vacuole, vacuoles were incubated with 2.5 mM BAPTA and separated into pellet (P) and supernatant (S) fractions by centrifugation. As seen in Fig. 2, BAPTA promoted the release of Vam7p, a peripheral membrane SNARE, and HOPS (homotypic fusion and vacuole protein sorting/Vps-Class C), a multisubunit complex that includes Vps18p, Vps33p, and Vps39p, from the vacuole membrane. The retention of these proteins on the vacuole membrane is not thought to directly require Ca2+, because these proteins lack calcium-binding motifs, although Vps11p and Vps18p do contain Zn2+-binding motifs. These data suggest that BAPTA may have multiple effects on in vitro fusion, including mechanisms that are distinct from Ca2+ chelation.



View larger version (54K):
[in this window]
[in a new window]
  		FIG. 2.
BAPTA alters the vacuole association of Vam7p and HOPS. Standard high Mg2+ fusion reactions without inhibitor or with 5 mM BAPTA were incubated for 90 min at 27 °C, then fractionated into membrane pellets and supernatants by centrifugation (13,000 x g, 15 min, 4 °C). Membranes were resuspended in 30 µl of PS buffer with protease inhibitors (1 µM leupeptin, 5 µM pepstatin, and 0.1 µM Pefabloc-SC). Equivalent portions of the pellets and supernatants were mixed with SDS-loading buffer, resolved by SDS-PAGE, and transferred to nitrocellulose. Membranes were probed with antibodies to Vam7p, the HOPS subunits Vps33p, Vps39p and Vps18p, and the GTPases Ypt7p and Rho1p.

 
BAPTA Contribution to Ionic Strength—Because elevated ionic strength is often sufficient to release peripherally bound membrane proteins, BAPTA might inhibit fusion through its contribution to ionic strength. Ionic strength is proportional to the square of the net charge borne by each ionized species in solution, according to the formula, {rho} = 1/2[{Sigma}MiC 2i], where {rho} is the ionic strength, Mi is the molarity of each ion, and Ci is its net charge. Thus, K4BAPTA would contribute 10-times as much to the ionic strength as equimolar KCl. BAPTA might also chelate other metal ions such as Zn2+, which may contribute to fusion or to the catalytic activity of matured Pho8p, and BAPTA might alter the pH of an insufficiently buffered reaction. To test whether part of the BAPTA inhibition derives from its contribution to ionic strength, we made compensatory adjustments to the KCl concentration of the reactions. To control for insufficient buffering capacity, reactions were performed in 50 mM PIPES-KOH instead of the standard 20 mM. To prevent BAPTA from inhibiting through binding Zn2+, the membrane-permeant heavy-metal chelator TPEN (31) was added; all fusion reactions containing 0.1 mM TPEN included 0.1 mM ZnCl2 in the final Pho8p assay solution.

Vacuole fusion is inhibited by high [KCl] (Fig. 3, squares). The inhibitory effect of 3 mM BAPTA (circles) is only seen as the reaction KCl concentration rises over 100 mM and approaches inhibitory levels. During standard in vitro fusion reactions, Ca2+ concentrations rarely rise above several micromolar (8), and thus even 3 mM BAPTA would reduce the levels of free Ca2+ to subnanomolar. However, at the lowest concentration of KCl tested (25 mM), even 9 mM BAPTA had little effect on fusion (Fig. 3, diamonds). Under these reaction conditions, much of the inhibition of vacuole fusion by BAPTA stems from the ionic strength it contributes to the reaction rather than from its Ca2+-binding properties.



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 3.
Low KCl concentrations reduce BAPTA inhibition. High Mg2+ fusion reactions containing 0.1 mM TPEN were incubated at 27 °C either without chelator (filled squares) or with 3 mM (filled circles) or 9 mM (filled diamonds) BAPTA with the indicated KCl concentrations. ZnCl2 (0.1 mM) was added to the Pho8p assay solution.

 
Factors other than ionic strength can also make modest contributions to the BAPTA inhibition of fusion. Even when the ionic strength is kept constant through adjustment of the KCl concentration, fusion inhibition due to BAPTA (Fig. 4, bars 1 versus 2) is somewhat relieved by the inclusion of ZnCl2 in the Pho8p assay buffer (bars 3 and 4). BAPTA solutions may also contribute to adverse pH changes, because increasing the buffer from 20 to 50 mM makes a modest contribution to BAPTA resistance (bars 5 and 6). Finally, in the presence of TPEN, with its extraordinary affinity for zinc, it becomes clear that a small part of the BAPTA inhibition of our assay signal was due to zinc chelation (bars 7 and 8). In sum, salt, pH, and zinc chelation each contribute to the inhibition by BAPTA.



View larger version (25K):
[in this window]
[in a new window]
  		FIG. 4.
Buffering capacity and ZnCl2 influence BAPTA sensitivity. High Mg2+ fusion reactions were performed either with 125 mM KCl and no BAPTA (black bars) or with 65 mM KCl and 6 mM BAPTA (gray bars) with the indicated concentrations of PIPES buffer and TPEN. ZnCl2 was added to 0.1 mM in the Pho8p assay solution for the reactions in lanes 3–8. Numbers above each gray bar are the percent inhibition of fusion caused by 6 mM BAPTA under each condition.

 
Although vacuoles will fuse in the presence of BAPTA upon lowering the KCl concentration, the high vacuole luminal Ca2+ levels and the docking-triggered Ca2+ flux may lead to brief spurts of Ca2+ at the vacuole surface, which could conceivably act even faster than chelation by BAPTA. We therefore tested vacuoles lacking the known Ca2+ ATPase, Pmc1p, and the Ca2+/H+ exchanger, Vcx1p, for their fusion in the presence of BAPTA. These vacuoles accumulate far less luminal Ca2+ (5) yet fuse normally (10). We considered that these vacuoles might be more sensitive to BAPTA under fusion conditions due to the lack of luminal calcium and that simple adjustment of [KCl] would not bypass a sensitivity to BAPTA. However, this was not the case (Fig. 5). When the [KCl] was lowered to compensate for the Ca2+ and BAPTA effects on ionic strength and when TPEN was added to 0.1 mM (Fig. 5), pmc1{Delta} vcx1{Delta} vacuoles fused as well as PMC1 VCX1 vacuoles in the presence of either added Ca2+, BAPTA, or an equimolar mixture of Ca2+·BAPTA. Strikingly, under ionic strength-adjusted conditions, fusion of either vacuole type was only minimally inhibited by concentrations of BAPTA up to 6 mM (Fig. 5B). Fusion in the presence of 6 mM BAPTA was completely sensitive to antibodies to Vam3p, a vacuolar t-SNARE, which is a hallmark of the physiological fusion pathway. Thus fusion of pmc1{Delta} vcx1{Delta} vacuoles proceeds in the presence of 6 mM BAPTA despite the reduction of luminal calcium.



View larger version (15K):
[in this window]
[in a new window]
  		FIG. 5.
Reduced luminal Ca2+ does not confer BAPTA sensitivity. High Mg2+ fusion reactions using vacuoles from BJ3505 and DKY6281 (PMC1 VCX1) or their pmc1 vcx1 deletion derivatives (BJ3505 vcx1::URA3 pmc1::TRP1) and (DKY6281 vcx1::URA3 pmc1::TRP1) (10) were performed with the indicated concentrations of added CaCl2 (A), BAPTA (B), or equimolar BAPTA·CaCl2 (C). The KCl concentration of each reaction was adjusted to compensate for the ionic strength of the divalent ion/chelator addition. All reactions contained 0.1 mM TPEN, and 0.1 mM ZnCl2 was present in the Pho8p assay solution.

 
To ensure that the fusion seen in the presence of BAPTA occurs by the well studied pathway that requires Rab and Rho GTPases, regulatory lipids, SNAREs, and HOPS, we added inhibitory ligands that target these fusion catalysts to reactions with 50 mM PIPES-KOH, pH 6.8, 0.1 mM TPEN, 65 mM KCl, and 6 mM BAPTA. Vacuole fusion in the presence of BAPTA was sensitive to each of the tested inhibitors (Fig. 6) and therefore proceeds via the authentic, physiological pathway. As a second control, we directly determined the effective concentration and the calcium-chelating properties of our BAPTA stock under fusion assay conditions. We exploited the fact that the extinction coefficient of BAPTA at 254 nm changes as it binds calcium (32) to show that our BAPTA stock is ~90% active (Fig. 7A). In addition, the dissociation constant (Kd) of BAPTA for Ca2+ under our standard vacuole fusion conditions is 126 nM (Fig. 7B), in accord with the published Kd of 107 nM (32). Thus BAPTA remains an effective calcium chelator under our fusion reaction conditions.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 6.
Authentic fusion in the presence of BAPTA. High Mg2+ fusion reactions containing 0.1 mM TPEN, 6 mM BAPTA, and 65 mM KCl were incubated either on ice or at 27 °C without inhibitor (top bar) or in the presence of fusion inhibitors, each used at concentrations previously reported (Fig. 2 of (21)). The Pho8p assay solution contained 0.1 mM ZnCl2.

 



View larger version (17K):
[in this window]
[in a new window]
  		FIG. 7.
BAPTA chelates calcium under fusion reaction conditions. BAPTA (150 mM) was freshly dissolved in PS buffer. A, 100 µM BAPTA was mixed with 10 mM MOPS buffer, pH 7.2, 0.1 mM KCl, and CaCl2 (32) (0–200 µM, prepared as a 1.0 M stock from oven-dried CaCl2) and the absorbance of the solution was measured at 254 nm using this buffer without BAPTA or CaCl2 as a blank. The extinction coefficient ({epsilon}) of BAPTA was determined for each [CaCl2], and was plotted against the Ca2+:BAPTA molar ratio. The intersection of the lines, calculated via a linear regression analysis (GraphPad Prism version 4.0, GraphPad Software, Inc.), measures the saturation of BAPTA by Ca2+. B, Hill plot of the change in absorbance of BAPTA at 254 nm in the presence of Ca2+. Reaction mixtures contained 20 mM PIPES-KOH, pH 6.8, 65 mM KCl, 100 µM TPEN, 10 µM coenzyme A, and 100 µM BAPTA. An equimolar CaCl2·EGTA solution and free EGTA were added in the following proportions to give known free ([Ca2+f]), in a manner previously described (32): 0 mM [Ca2+f]f, 10 mM EGTA; 10 nM [Ca2+f], 11.3 µ µM [Ca:EGTA]/9.99 mM EGTA; 32 nM [Ca2+f], 36.3 µm [Ca:EGTA]/9.96 mM EGTA; 100 nM [Ca2+]f, 113 µM [Ca:EGTA]/9.89 mM EGTA; 320 nM [Ca2+]f, 363 µM [Ca:EGTA]/9.64 mM EGTA; 1000 nM [Ca2+f], 1.13 mM [Ca:EGTA]/8.86 EGTA; 1 mM [Ca2+]f, 1 mM CaCl2.

 
BAPTA and High Salt Show Similar Inhibition of Fusion— Although BAPTA inhibits vacuole fusion through its contribution to ionic strength, we have previously reported (10, 19) that BAPTA is a reversible late-acting fusion inhibitor. We therefore tested whether elevated salt can also act as a reversible late-acting fusion inhibitor. Fusion inhibition by 6 mM BAPTA (Fig. 8B, lane 2) was relieved by 6 mM CaCl2 (lane 3). This is consistent with the BAPTA inhibition being largely due to ionic strength; because BAPTA has a charge of –4 and the stable BAPTA·Ca2+ complex has a net charge of –2, the ionic strength of an equimolar mixture of Ca2+ and BAPTA is far lower than the sum of their separate ionic strengths. Fusion in the presence of BAPTA·Ca2+ remained completely sensitive to antibodies (Fig. 8B, compare lane 3 to lanes 4–6) that inhibited the standard reaction (Fig. 8A). However, when the BAPTA inhibition was reversed by the addition of Ca2+ at 35 min, the ensuing fusion (Fig. 8B, lane 8) was fully resistant to the priming inhibitor {alpha}-Sec18p (lane 9) while remaining sensitive to {alpha}-Ypt7p or {alpha}-Vam3p (lanes 10 and 11). In a standard incubation without inhibitors, the reaction becomes resistant to each of these antibodies by 35 min (9, 19). Thus vacuole priming can occur in the presence of BAPTA, but essential Ypt7p Rab and Vam3p SNARE functions are blocked.



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 8.
Comparison of fusion inhibition by BAPTA and KCl. A, standard high Mg2+ fusion reactions were incubated on ice or at 27 °C for 60 min. Antibody inhibitors were present from the start of the reaction. B, fusion inhibition by 6 mM BAPTA was reversed by the addition of 6 mM CaCl2 at the start of the incubation (left) or 35 min after the start of incubation at 27 °C (right). Antibody inhibitors (lanes 4–6, 9–11) were added 5 min prior to the CaCl2 addition in either case. Reactions in lanes 1–6 were incubated at 27 °C for 60 min. The reaction in lane 7 was moved from 27 °C to ice at 30 min and represents the amount of fusion that had occurred when the antibody inhibitors were added for lanes 8–11. Samples in lanes 8–11 were at 27 °C for a total of 95 min to allow 60 min of fusion at 27 °C after CaCl2 addition. C, high Mg2+ fusion reactions with 250 mM KCl were diluted by the addition of 30 µl of fusion reaction buffer (see "Experimental Procedures") lacking KCl (1:2 dilution to a final [KCl] of 125 mM) at either the start of the reaction (left) or 35 min after the start of fusion at 27 °C (right). Antibody inhibitors (lanes 3–5, 9–11) were added 5 min prior to reversal of the salt or BAPTA blocks. The reaction in lane 6 was diluted 1:2 with reaction buffer containing 250 mM KCl and represents the KCl-inhibited fusion signal. The incubation times and temperature for each set are as described for set B.

 
Fusion reactions inhibited by 250 mM KCl behave similarly in this type of experiment (Fig. 8C). Inhibition by 250 mM KCl can be reversed by diluting the reaction 2-fold to a final concentration of 125 mM KCl prior to warming the reaction to 27 °C (Fig. 8C; compare lane 2, diluted to lower the salt, to lane 6, diluted but maintained at 250 mM KCl). Reversal of salt inhibition through dilution at the start left the reaction still sensitive to each antibody (lanes 3–5). When reactions were incubated at 27 °C with 250 mM KCl, mixed with inhibitors at 30 min, then diluted to lower the salt concentration at 35 min, the pattern of fusion resembled calcium-mediated BAPTA reversal. During incubation with 250 mM KCl, the reaction had acquired resistance to {alpha}-Sec18p (compare lanes 8 and 9) but remained largely sensitive to {alpha}-Ypt7p (lane 10) or {alpha}-Vam3p (lane 11). Thus the BAPTA inhibition of fusion closely resembles inhibition caused by elevated ionic strength. We note that these results differ from our earlier studies (10, 19), which had placed the action of BAPTA after the acquisition of resistance to {alpha}-Ypt7p or {alpha}-Vam3p. These earlier studies may not have allowed the inhibitory antibodies sufficient time to act before reversing the BAPTA block.

BAPTA Effects on Fusion under Other Salt Conditions— Earlier studies of BAPTA inhibition of vacuole fusion (7, 10) had employed closely related reaction conditions that differed in the concentrations of buffer and divalent cations. Under these conditions (10 mM PIPES-KOH, pH 6.8, 0.5 mM Mg/ATP, 0.5 mM MnCl2, 125 mM KCl, 200 mM sorbitol, 10 µM coenzyme A, and 330 nM IB2), hereafter referred to as low Mg2+ conditions, vacuole fusion proceeded (Fig. 9A, filled squares) as in high Mg2+ reaction conditions. As reported (7), fusion is blocked by BAPTA, especially at KCl concentrations >100 mM (filled circles). This inhibition is not relieved by the addition of ZnCl2 to the Pho8p assay (Fig. 9A, open symbols). Fusion at high or low Mg2+ remained sensitive to {alpha}-Vam3p (data not shown).



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 9.
Bypass of BAPTA inhibition by CaCl2, rVam 7, or MgCl2. Low Mg2+ reactions (27 °C, 60 min; see "Experimental Procedures") were incubated at the indicated KCl concentrations. Pho8p phosphatase activity was assayed with a 10 µl sample for 15 min at 30 °C. The fusion signal obtained from a reaction left at 0 °C has been subtracted from all curves. A, reactions were incubated either without BAPTA (squares), or with 6 mM BAPTA (circles). After fusion, phosphatase activity was assayed either without ZnCl2 addition (filled symbols) or with 0.14 mM ZnCl2 in the Pho8p assay solution (open symbols). B, fusion reactions containing 0.1 mM TPEN were assayed for Pho8p activity with 0.14 mM ZnCl2 in the Pho8p assay. Fusion reactions contained: open squares, no BAPTA; filled circles, 1 mM BAPTA; filled triangles, 3 mM BAPTA; filled diamonds, 5 mM BAPTA; filled squares, 6 mM BAPTA. C, all reactions contained 0.1 mM TPEN and were assayed for Pho8p activity with 0.14 mM ZnCl2 in the Pho8p assay solution. BAPTA and CaCl2 solutions were premixed before addition to the fusion reaction. Free BAPTA and Ca2+ concentrations were estimated with WEBMAXC STANDARD (63). Fusion reactions contained the following total concentrations of BAPTA and CaCl2: open squares, 0 mM BAPTA/0 mM CaCl2; filled squares, 6 mM BAPTA/0 mM CaCl2; filled triangles, 6 mM BAPTA/100 µM CaCl2; filled inverted triangles, 6 mM BAPTA/1 mM CaCl2; filled diamonds, 6 mM BAPTA/4 mM CaCl2; filled circles, 6 mM BAPTA/6 mM CaCl2. D, all fusion reactions were as in C, except without TPEN. The Pho8p assay solution did not contain ZnCl2. E, all reactions contained 0.1 mM TPEN and were assayed for Pho8p activity with 0.14 mM ZnCl in the Pho8p assay solution. Fusion reactions contained: filled squares, no BAPTA; filled circles, 6 mM BAPTA; open triangles, 6 mM BAPTA and 2.8 µM rVam7p; open diamonds, 6 mM BAPTA and 6 mM MgCl2.

 
To resolve the effects of BAPTA on vacuole fusion from any effects on the availability of Zn2+ for Pho8p activity, BAPTA was added to low Mg2+ condition reactions, which also contained 100 µM TPEN, and Zn2+ was restored during the Pho8p assay. BAPTA caused a dose-dependent inhibition of fusion activity at higher KCl concentrations (Fig. 9B), while having less effect at salt concentrations of 100 mM KCl or less. Only the addition of 6 mM BAPTA to the reaction caused a significant inhibition of fusion at 100 mM KCl (filled squares). These ionic strength effects of BAPTA on vacuole fusion are similar to the results obtained under high Mg2+ conditions (Fig. 3) and suggest a similar inhibitory mechanism between these two reaction conditions.

If the inhibitory effects of BAPTA on fusion resulted from the chelation of free Ca2+, the addition of CaCl2 should completely restore fusion when the free [Ca2+] reaches a level that is required for fusion. Alternatively, fusion might be restored by reducing the amount of total free [BAPTA] available. To distinguish between these possibilities, CaCl2 was titrated into fusion reactions containing 6 mM BAPTA, low Mg2+, and TPEN. 6 mM BAPTA potently inhibits fusion under these conditions, especially at KCl concentrations >100 mM (Fig. 9C, compare open squares to filled squares). When enough calcium was added to reduce the concentration of BAPTAfree to ~2 mM (filled diamonds), fusion at 150 mM KCl was still severely inhibited, although 2 mM BAPTA did little to inhibit the overall reaction under these conditions (see Fig. 9B). Thus fusion is not simply inhibited by [BAPTA]free. Equimolar CaCl2 (6 mM), which yielded a free Ca2+ concentration of ~40 µM, dramatically restored fusion at low KCl concentrations (Fig. 9C). Fusion was not restored at [KCl] of 150 or greater; this may reflect, at least in part, BAPTA inhibition through contribution to ionic strength. These studies show that BAPTA can indeed, as reported (7), inhibit fusion through Ca2+ chelation under certain conditions of limited free magnesium, low ionic strength, and (see below, Fig. 9E) in the absence of free Vam7p.

It was possible that the effect of CaCl2-dependent relief of BAPTA inhibition seen in Fig. 9C was specific to the inclusion of TPEN in the assay, or ZnCl2 in the Pho8p assay. Therefore, we repeated this assay in the absence of TPEN and ZnCl2 (Fig. 9D) and found a similar profile of fusion under these conditions (compare panels D and C). In addition, fusion showed a similar salt profile and total yield in the presence or absence of TPEN (compare Fig. 9A and 9E, filled squares) and yet remained sensitive to BAPTA (filled circles).

Other reaction components were tested for their ability to reverse the BAPTA-mediated inhibition. There was a dramatic reversal of BAPTA inhibition by the addition of 5.5 mM Mg2+ (Fig. 9E, open diamonds), consistent with our fusion studies under high Mg2+ reaction conditions (Figs. 1, 2, 3, 4, 5, 6, 7, 8). BAPTA inhibition is also reversed by added recombinant Vam7p (open triangles), in accord with earlier studies (8). The fusion supported by Vam7p in the presence of BAPTA is sensitive to {alpha}-Vam3p and {alpha}-Ypt7p antibodies (data not shown) and is thus authentic Rab- and SNARE-dependent fusion. Although Vam7p addition can stimulate a Ca2+ efflux of several micromolar (8, 33), this would have little impact on the free Ca2+ concentrations in the presence of 6 mM BAPTA and is thus unlikely to be responsible for the Vam7p-dependent reversal of BAPTA inhibition (Fig. 9E). Lower KCl concentrations are required for efficient Vam7p or Mg2+ bypass of BAPTA than for reactions without BAPTA (filled squares), consistent with the substantial contributions of BAPTA to ionic strength. The drastic increase of fusion activity seen in the presence of 6 mM BAPTA and 6 mM MgCl2 in the absence of KCl is consistent with the high fusion signals seen at high Mg2+ conditions and high [BAPTA] (Fig. 3). This behavior is also strikingly similar to the BAPTA reversal caused by 6 mM CaCl2 (Fig. 9C). Thus, either Mg2+ or Ca2+ can fulfill the need for divalent cation, and BAPTA can at least partially substitute for KCl by providing the needed reaction ionic strength.

Additional studies with the low affinity calcium chelator, 5,5'-dibromo-BAPTA (Molecular Probes, Kd = 1.6 µM) showed that this compound also inhibits vacuole fusion in concentrations similar to BAPTA (data not shown). Unlike BAPTA, however, this inhibition cannot be relieved by the addition of Ca2+, suggesting that its effects on fusion are not all related to Ca2+ chelation.

Calcium Binding to Calmodulin Is Not Required for Vacuole Fusion—It has been suggested that free calcium might activate the essential calcium-binding protein calmodulin to trigger vacuole fusion (7). In light of our current findings, we reevaluated the requirement for Ca2+ binding to calmodulin. We assayed the fusion of vacuoles bearing mutant calmodulins with severely lowered ability to bind calcium. Either pho8{Delta} or pep4{Delta} were introduced into S. cerevisiae harboring either wild-type calmodulin (CMD1) or either of two calmodulin alleles, cmd1–3 or cmd1–6. The proteins encoded by the cmd1–3 and cmd1–6 alleles bind calcium with dissociation constants of at least 300 µM, and possibly higher (15). In comparison, the wild-type CMD1 protein can bind Ca2+ with a Kd of ~3 µM (15, 34). Vacuoles purified from the cmd1–3 and cmd1–6 genetic backgrounds fuse just as well as those purified from the CMD1 background, even when assayed under low Mg2+ conditions (Fig. 10). Although it is formally possible that the particular calmodulin mutants we have employed are in a "locked-on" signaling mode with regard to membrane fusion, this seems unlikely because cmd1–6 strains cannot survive a challenge by {alpha}-factor mating pheromone, and this pheromone operates through a signaling cascade that requires the locked-on, Ca2+-bound conformation of calmodulin (14). The fusion of these vacuoles remained sensitive to known fusion inhibitors and thus represents authentic Ypt7p-, SNARE-, and HOPS complex-dependent homotypic vacuole fusion. The reason for the discrepancy between this study and a previous report that vacuoles from the cmd1–3 background would not fuse (7) is unclear.



View larger version (8K):
[in this window]
[in a new window]
  		FIG. 10.
Vacuole fusion does not require Ca2+-bound calmodulin. Low Mg2+ reactions ("Experimental Procedures") were on ice or at 27 °C for 90 min. Antibody inhibitors were added from the start of the reaction. Top, fusion of vacuoles isolated from VSY3 and VSY4 (CMD1, see "Experimental Procedures"). Middle, fusion of vacuoles isolated from VSY5 and VSY6 (cmd1–3). Bottom, fusion of vacuoles isolated from VSY7 and VSY8 (cmd1–6).

 
Divalent Ions and Chelators Alter Vacuole Fusion at Different Ionic Strengths—These data suggest that divalent ions, although important in fusion, are not required for a specific signaling pathway. To further explore the requirements for ionic strength or divalent ions, we exploited the fact that vacuoles will fuse in the absence of ATP when provided with sufficient Vam7p (21), a condition that we have termed "bypass fusion," because it bypasses the need for Sec18p/ATP-dependent priming. This permits a decoupling of any requirement for Mg2+ ions for fusion from the role of Mg2+ as a cofactor for ATP hydrolysis by Sec18p. During incubation without ATP and with added Vam7p, vacuoles fuse at elevated KCl concentrations (Fig. 11, filled squares). The addition of 6 mM MgCl2 generates a distinct shift in the salt requirement for fusion (Fig. 11, filled circles). The increased fusion afforded by Mg2+ at lower salt is not solely explained by the ionic strength contribution of MgCl2, because fusion at 118 mM KCl (which has the same ionic strength as 100 mM KCl plus 6 mM MgCl2) is not stimulated to the same extent. The addition of 6 mM CaCl2 supported fusion in a manner that was indistinguishable from the addition of MgCl2 (data not shown), suggesting that these ions behave in a similar manner and that they do not promote fusion by a unique Mg2+- or Ca2+-dependent signaling mechanism. When divalent cationic salts were omitted entirely and a mixture of divalent cation chelators was added (0.1 mM TPEN and 1 mM each of EDTA and BAPTA), the salt profile of fusion was significantly altered (Fig. 11, open squares). Remarkably, this chelator mixture actually stimulates fusion at 100 mM KCl. The divalent ions already associated with the vacuole may be inhibitory or the chelators themselves might alter the physical properties of the lipid bilayer to promote fusion. To determine which chelators stimulate under these narrowly defined conditions, we added the chelators singly or in combination. Bypass fusion with 100 mM KCl requires either divalent ions (Fig. 11) or chelators (Fig. 11 and Table II). The addition of low concentrations (0.1–1 mM) of divalent ion chelators in various combinations promotes fusion, with maximal fusion occurring in the presence of all three chelators (TPEN, EDTA, and BAPTA), a condition where any free divalent ions would be at subnanomolar concentrations. Although divalent ions can directly associate with lipid bilayers to promote fusion (35–37), these data suggest that free divalent cations are not required for association with signaling proteins during vacuole membrane fusion.



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 11.
The effects of KCl, divalent ions, and chelators on bypass fusion. Bypass fusion reactions (60 min, 27 °C; "Experimental Procedures") contained neither divalent ions nor chelators (filled squares), 0.1 mM TPEN, 1 mM EDTA, and 1 mM BAPTA (open squares), or 6 mM MgCl2 (filled circles). After 60 min of fusion, reactions were placed on ice, and chelators and magnesium were added to each reaction so that all Pho8p assays contained the same final concentrations of chelators and divalent ions. Each was diluted 1:3 into PS buffer, and 30 µl of each was developed for 15 min at 30 °C with the standard Pho8p assay solution supplemented with 0.14 µM ZnCl2.

 


View this table:
[in this window]
[in a new window]
  		TABLE II
Bypass fusion can be chelator stimulated

After fusion for 60 min at 27 °C, samples are placed on ice and each of the three chelators are added to each sample to the same final concentration. To assay Pho8p activity, samples were diluted 1:3 in PS buffer and a 30-µl aliquot was assayed ("Experimental Procedures"). ZnCl2 (0.14 mM) was added to the assay solution, and the assay time was 15 min. A fusion sample left at 0 °C gave a fusion signal of 0.19 unit, which has been subtracted from the above data. Chelators were present at 0.1 mM (TPEN) and 1 mM (EDTA and BAPTA).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Divalent cations serve as cofactors and signaling ligands for membrane fusion proteins such as NSF and synaptotagmin. Divalent ions can also drive fusion of protein-free liposomes (35–37). It has been unclear how each of these two ion effects relate to physiological membrane fusion. When serving as a protein cofactor, Ca2+ and Mg2+ are usually not interchangeable and cannot be replaced by monovalent salt. Our current studies are consistent with a working model in which divalent ions modulate the capacity of lipids to rearrange during bilayer fusion.

We find that vacuole fusion can occur despite the chelation of Ca2+ by BAPTA if sufficient Mg2+ or Vam7p are present. This suggests that free Ca2+ may not have a unique role in vacuole fusion. Although calmodulin can serve as a Ca2+ sensor for other biological processes, Ca2+ binding by calmodulin is not required for its function during vacuole fusion (Fig. 10). Furthermore, Rab- and SNARE-dependent vacuole fusion can occur in a reaction with sub-nanomolar concentrations of free divalent ions (Fig. 11 and Table II). Fusion can even be stimulated by divalent ion chelators under certain salt and reaction conditions. Finally, vacuoles lacking the major Ca2+ uptake systems, Pmc1p and Vcx1p, fuse normally in the presence of up to 6 mM BAPTA (Fig. 5), thus providing further evidence that a calcium signal is not a unique prerequisite for vacuole fusion, at least in vitro. Although luminal vacuolar Ca2+ is released after trans-pairing of SNAREs (7, 8), and our results do not exclude the possibility that this docking-dependent efflux of vacuolar calcium might trigger fusion under specific intracellular conditions of limited free Mg2+ or Vam7p, they broaden our thinking about possible fusion mechanisms to include the direct fusogenic effects of Ca2+, Mg2+, and monovalent ions on lipid bilayer rearrangements.

In addition to the roles of proteins in catalyzing fusion, the necessary rearrangements of bilayer lipids during fusion are substantial and have received extensive study and consideration. For example, DAG is fusogenic in membrane model systems due to its induction of negative membrane curvature (38) and is required for vacuole fusion as well (33). Ca2+ and Mg2+ can directly stimulate the aggregation and fusion of liposomes composed of phosphatidylserine (PS) or PS-phosphatidic acid (35, 36). Monovalent salts can also promote or, at higher concentrations, inhibit liposome aggregation and fusion (39), as we now report for vacuole fusion. There has been little prior evaluation whether the Rab- and SNARE-dependent fusion of biological membranes may also be governed by the same interactions of ions with membrane lipids as seen in these liposome studies. Four regulatory lipids (ergosterol, phoshphatidylinositol-3-phosphate, phosphatidylinositol-4,5-bisphosphate, and diacylglycerol) function together with the vacuolar Rab, Rab effectors, and SNAREs to assemble a vacuole membrane microdomain for Rab- and SNARE-dependent vacuole fusion (40). Similarly, an enzymatically controlled and spatially distinct interconversion of phosphatidic acid and lysophosphatidic acid has been suggested to control Golgi membrane fission by altering lipid bilayer shape and physical properties (41, 42).

Despite these advances, the relationship between proteins such as SNAREs and specific regulatory lipids at the last stages of membrane fusion remains unclear. Although Mg2+ is required for NSF to couple ATP hydrolysis to SNARE complex disassembly (43, 44), the NSF ATPase is not required for single rounds of in vitro membrane fusion (1, 21). During synaptic vesicle exocytosis, a wave of membrane depolarization triggers Ca2+ influx, then calcium-bound synaptotagmin undergoes a conformational change that alters its interactions with phosphoinositides (45) and SNARE complexes (46–48) as well as mediating synaptotagmin oligomerization (49, 50). These steps link Ca2+ signaling to regulated neuronal membrane fusion, as elegantly reconstituted in a defined model reaction (51), but it remains unclear whether Ca2+ is directly involved in regulating the fusion of other membranes, such as the vacuole, which lack synaptotagmin. One model of bilayer mixing during vacuole fusion posits a radially expanding, proteinaceous "fusion pore" of apposed, oligomerized vacuolar ATPase V0 sectors (52–54). Alternatively, there is evidence for a "hemifusion" transition state in membrane fusion consisting of a predominantly lipidic "neck" (55–57). The driving forces to establish or resolve either of these states may include bilayer strain from trans-pairing of SNAREs (58–61), although ionic conditions alone can drive PC-PA liposomes into the hemifusion state (35). In support of this, in vitro studies of plant vacuole fusion suggest that Mg2+ addition can drive rapid lipid mixing while maintaining separate luminal contents (62). These studies offer a glimpse into the role Mg2+ and Ca2+ can play in directly altering the arrangement of lipid bilayers during membrane fusion. How, then, do these concepts, founded largely on model studies with liposomes, relate to Rab- and SNARE-dependent fusion systems such as the vacuole?

We suggest a working model for the regulation of vacuole fusion by interactions of ions with the bilayer. Although Mg2+ is required for Sec18p ATPase activity, we suggest that Mg2+ and Ca2+ also bind directly to bilayer lipids, thereby regulating the lipid rearrangements, from separate apposed bilayers to hemifusion and on to subsequent pore formation, which are the essence of membrane fusion. This is in accord with recent findings that ligands to regulatory lipids block fusion (40), even though SNARE pairing is unimpeded when fusion is blocked by lipid ligands such as myristoylated alanine-rich C kinase substrate effector domain.2 Direct assays of the lipid rearrangements that occur after SNARE pairing and during bilayer mixing, and measurement of the effects of Ca2+, Mg2+, and KCl on these rearrangements, will be required to test this model. It is not known whether the ionic requirements for vacuole fusion in vivo are regulated by ionic transients (i.e. a rapid Ca2+ or Mg2+ influx/efflux), or whether the resting ionic state of the cell suffices. In either case, ions may have specific, protein-dependent functions, such as Mg2+ for Sec18p ATPase activity, while also regulating the capacity of the bilayer to undergo the fundamental rearrangements required for biological membrane fusion.


   FOOTNOTES
 
* This work was supported in part by a grant from the National Institute of General Medical Sciences. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} Supported by a Damon Runyon Cancer Research Foundation fellowship (DRG-1837). Back

§ Supported by a Helen Hay Whitney Foundation fellowship. Back

To whom correspondence should be addressed: Dept. of Biochemistry, Dartmouth Medical School, 7200 Vail Bldg., Hanover, NH 03755-3844. Tel.: 603-650-1701; Fax: 603-650-1353; E-mail: Bill.Wickner{at}Dartmouth.edu.

1 The abbreviations used are: SNARE, soluble NSF attachment protein receptor; BAPTA, 1,2-bis-(o-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid; TPEN, N,N,N',N'-tetrakis-(2-pryidylmethyl)ethylenediamine; HOPS, homotypic fusion and vacuole protein sorting complex; PS, phosphatidylserine; PC, phosphatidylcholine; NSF, N-ethylmaleimide-sensitive factor. Back

2 K. M. Collins, N. Thorngren, R. A. Fratti, and W. Wickner, submitted for publication. Back


   ACKNOWLEDGMENTS
 
We thank Dr. E. Martegani for reagents, Dr. T. Davis for strains and suggestions, and Chris Patton for helpful discussions regarding the use of the MAXCHELATOR software. We especially thank Drs. Alexey Merz and Andreas Mayer for helpful discussions and strains and Dr. John Silvius for insightful suggestions.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Jahn, R., Lang, T., and Sudhof, T. C. (2003) Cell 112, 519–533[CrossRef][Medline] [Order article via Infotrieve]
  2. Chapman, E. R. (2002) Nat. Rev. Mol. Cell. Biol. 3, 498–508[CrossRef][Medline] [Order article via Infotrieve]
  3. Dunn, T., Gable, K., and Beeler, T. (1994) J. Biol. Chem. 269, 7273–7278[Abstract/Free Full Text]
  4. Cunningham, K. W., and Fink, G. R. (1994) J. Cell Biol. 124, 351–363[Abstract/Free Full Text]
  5. Cunningham, K. W., and Fink, G. R. (1996) Mol. Cell. Biol. 16, 2226–2237[Abstract]
  6. Wickner, W. (2002) EMBO J. 21, 1241–1247[CrossRef][Medline] [Order article via Infotrieve]
  7. Peters, C., and Mayer, A. (1998) Nature 396, 575–580[CrossRef][Medline] [Order article via Infotrieve]
  8. Merz, A. J., and Wickner, W. T. (2004) J. Cell Biol. 164, 195–206[Abstract/Free Full Text]
  9. Eitzen, G., Will, E., Gallwitz, D., Haas, A., and Wickner, W. (2000) EMBO J. 19, 6713–6720[CrossRef][Medline] [Order article via Infotrieve]
  10. Ungermann, C., Wickner, W., and Xu, Z. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 11194–11199[Abstract/Free Full Text]
  11. Plankert, U., Purwin, C., and Holzer, H. (1991) Eur. J. Biochem. 196, 191–196[Medline] [Order article via Infotrieve]
  12. Jones, E. W. (2002) Methods Enzymol. 351, 127–150[Medline] [Order article via Infotrieve]
  13. Haas, A., Conradt, B., and Wickner, W. (1994) J. Cell Biol. 126, 87–97[Abstract/Free Full Text]
  14. Moser, M. J., Geiser, J. R., and Davis, T. N. (1996) Mol. Cell. Biol. 16, 4824–4831[Abstract]
  15. Geiser, J. R., van Tuinen, D., Brockerhoff, S. E., Neff, M. M., and Davis, T. N. (1991) Cell 65, 949–959[CrossRef][Medline] [Order article via Infotrieve]
  16. Gietz, R. D., and Woods, R. A. (2001) BioTechniques 30, 816–820, 822–816, 828, passim[Medline] [Order article via Infotrieve]
  17. Haas, A., and Wickner, W. (1996) EMBO J. 15, 3296–3305[Medline] [Order article via Infotrieve]
  18. Wang, L., Merz, A. J., Collins, K. M., and Wickner, W. (2003) J. Cell Biol. 160, 365–374[Abstract/Free Full Text]
  19. Eitzen, G., Thorngren, N., and Wickner, W. (2001) EMBO J. 20, 5650–5656[CrossRef][Medline] [Order article via Infotrieve]
  20. Seals, D. F., Eitzen, G., Margolis, N., Wickner, W. T., and Price, A. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 9402–9407[Abstract/Free Full Text]
  21. Thorngren, N., Collins, K. M., Fratti, R. A., Wickner, W., and Merz, A. J. (2004) EMBO J. 23, 2765–2776[CrossRef][Medline] [Order article via Infotrieve]
  22. Garrett, M. D., and Novick, P. J. (1995) Methods Enzymol. 257, 232–240[Medline] [Order article via Infotrieve]
  23. Johnson, J. E., Giorgione, J., and Newton, A. C. (2000) Biochemistry 39, 11360–11369[CrossRef][Medline] [Order article via Infotrieve]
  24. Gillooly, D. J., Morrow, I. C., Lindsay, M., Gould, R., Bryant, N. J., Gaullier, J. M., Parton, R. G., and Stenmark, H. (2000) EMBO J. 19, 4577–4588[CrossRef][Medline] [Order article via Infotrieve]
  25. Wang, J., Arbuzova, A., Hangyas-Mihalyne, G., and McLaughlin, S. (2001) J. Biol. Chem. 276, 5012–5019[Abstract/Free Full Text]
  26. Slusarewicz, P., Xu, Z., Seefeld, K., Haas, A., and Wickner, W. T. (1997) Proc. Natl. Acad. Sci. U. S. A. 94, 5582–5587[Abstract/Free Full Text]
  27. Haas, A. (1995) Meth. Cell Sci 17, 283–294[CrossRef]
  28. Saoudi, Y., Rousseau, B., Doussiere, J., Charrasse, S., Gauthier-Rouviere, C., Morin, N., Sautet-Laugier, C., Denarier, E., Scaife, R., Mioskowski, C., and Job, D. (2004) Eur. J. Biochem. 271, 3255–3264[Medline] [Order article via Infotrieve]
  29. Rousset, M., Cens, T., Vanmau, N., and Charnet, P. (2004) FEBS Lett. 576, 41–45[CrossRef][Medline] [Order article via Infotrieve]
  30. Chiancone, E., Thulin, E., Boffi, A., Forsen, S., and Brunori, M. (1986) J. Biol. Chem. 261, 16306–16308[Abstract/Free Full Text]
  31. Arslan, P., Di Virgilio, F., Beltrame, M., Tsien, R. Y., and Pozzan, T. (1985) J. Biol. Chem. 260, 2719–2727[Abstract/Free Full Text]
  32. Tsien, R. Y. (1980) Biochemistry 19, 2396–2404[CrossRef][Medline] [Order article via Infotrieve]
  33. Jun, Y., Fratti, R. A., and Wickner, W. (2004) J. Biol. Chem. 279, 53186–53195[Abstract/Free Full Text]
  34. Luan, Y., Matsuura, I., Yazawa, M., Nakamura, T., and Yagi, K. (1987) J. Biochem. (Tokyo) 102, 1531–1537[Abstract/Free Full Text]
  35. Leventis, R., Gagne, J., Fuller, N., Rand, R. P., and Silvius, J. R. (1986) Biochemistry 25, 6978–6987[CrossRef][Medline] [Order article via Infotrieve]
  36. Duzgunes, N., Nir, S., Wilschut, J., Bentz, J., Newton, C., Portis, A., and Papahadjopoulos, D. (1981) J. Membr. Biol. 59, 115–125[CrossRef][Medline] [Order article via Infotrieve]
  37. Bentz, J., and Duzgunes, N. (1985) Biochemistry 24, 5436–5443