Phosphatidic Acid Sequesters Sec18p from cis-SNARE Complexes to Inhibit Priming

Abstract

Yeast vacuole fusion requires the activation of cis-SNARE complexes through priming carried out by Sec18p/N-ethylmaleimide sensitive factor and Sec17p/α-SNAP. The association of Sec18p with vacuolar cis-SNAREs is regulated in part by phosphatidic acid (PA) phosphatase production of diacylglycerol (DAG). Inhibition of PA phosphatase activity blocks the transfer of membrane-associated Sec18p to SNAREs. Thus, we hypothesized that Sec18p associates with PA-rich membrane microdomains before transferring to cis-SNARE complexes upon PA phosphatase activity. Here, we examined the direct binding of Sec18p to liposomes containing PA or DAG. We found that Sec18p preferentially bound to liposomes containing PA compared with those containing DAG by approximately fivefold. Additionally, using a specific PA-binding domain blocked Sec18p binding to PA-liposomes and displaced endogenous Sec18p from isolated vacuoles. Moreover, the direct addition of excess PA blocked the priming activity of isolated vacuoles in a manner similar to chemically inhibiting PA phosphatase activity. These data suggest that the conversion of PA to DAG facilitates the recruitment of Sec18p to cis-SNAREs. Purified vacuoles from yeast lacking the PA phosphatase Pah1p showed reduced Sec18p association with cis-SNAREs and complementation with plasmid-encoded PAH1 or recombinant Pah1p restored the interaction. Taken together, this demonstrates that regulating PA concentrations by Pah1p activity controls SNARE priming by Sec18p.

Keywords: Pah1p; SNARE; Sec17; diacylglycerol; fusion; lipin; phosphatidic acid; priming.

Figure 1. Sec18p preferentially binds to liposomes containing phosphatidic acid

Recombinant His₆-Sec18p (2 µg) was incubated with liposomes of the indicated compositions for 10 min at 30°C. Liposomes were isolated by centrifugation and washed with 1X PBS before bound proteins were resolved by SDS-PAGE. Bar graphs show average normalized densitometry values measured for 3 separate experiments. Binding was observed in liposomes with PA or DAG (A), with increasing concentrations of PA or DAG (B), with increasing concentrations of PA or PS (C), and with no liposomes (D). * P<0.05; ** P<0.001; *** P<0.0001

Figure 2. The PA-binding domain DEP reduces Sec18p binding to membranes

(A) Liposomes of the indicated compositions were incubated without Dvl2-DEP or with 20 µM Dvl2-DEP for 5 min at 30°C. Recombinant His₆-Sec18p (2 µg) was added to the sample and incubated for 10 min at 30°C. (B) GST-DEP titration and Sec18p binding to liposomes was performed as described above in the presence of the indicated concentrations of GST-DEP. (C) Vacuoles were harvested from wild type DKY6281 and incubated with GST-DEP, GST-C1b domain, or GST-ENTH domain. Bound and unbound fractions were separated by centrifugation and resolved by SDS PAGE. Bar graphs show the percentage of unbound Sec18p in each sample normalized to wild type control. * P<0.05; ** P<0.001; *** P<0.0001.

Figure 3. Addition of exogenous phosphatidic acid or N-ethylmaleimide causes a severe membrane fusion defect

Vacuoles were harvested from wild type BJ3505 and DKY6281 and tested for fusion by luminal mixing and proPho8p maturation. Fusion reactions containing 3 µg of each vacuole type were incubated in the presence of diC8-PA (A), or diC8-DAG (B) at the indicated concentrations. (C) Fusion reactions containing 100 µg/ml diC8-PA were incubated in the presence of 188 nM recombinant Sec18p. Fusion reactions containing 3 µg of each vacuole type were incubated in the presence of GST-DEP (D) or NEM (E) at the indicated concentrations. Fusion results were normalized to untreated wild type vacuoles at standard conditions. (F) Exogenous recombinant Vam7p (200 nM) was used to bypass fusion inhibitors. (G) Gain of resistance kinetics assays were performed in the presence of 140 µg/ml α-Sec17p IgG, 2 mM propranolol, 300 µM diC8-PA, 1 mM NEM, or PS buffer. Data was fit using first-order exponential decay with weights and errors. (H) Calculated half-times from first-order exponential decay fit. ** P<0.001.

Figure 4. Addition of exogenous phosphatidic acid but not diacylglycerol inhibits the priming activity of Sec18p

(A) Vacuoles from BJ3505 yeast were assayed for priming activity as detected by the release of Sec17p from the membrane fraction. Fusion reactions of 3 µg of vacuoles were incubated in the presence of buffer, 1 mM NEM, 300 µM diC8-PA, or 300 µM diC8-DAG. Vacuoles were pelleted by centrifugation at the indicated times and proteins in the supernatant fraction were resolved by SDS-PAGE and imaged by Western blot. Densitometry values were normalized against input sample for each condition. Graphs show the normalized averages (n=3). (B) Vacuoles from BJ3505 and pah1Δ yeast were assayed for priming activity as described in (A). Graph shows the normalized averages (n=3).

Figure 5. Vacuoles from a pah1Δ deletion strain have a decrease in the level of Sec18p bound to cis-SNARE complexes

Vacuoles were harvested from BJ3505, RFY17 (BJ3505 pah1Δ), and RFY19 (BJ3505 pah1Δ, pPAH1) strains and assayed for Sec18p binding to cis-SNARE complexes. (A) Recombinant GST-Vam7p was added to fusion reactions and incubated at 27°C for 30 min in the absence of ATP to allow formation of cis-SNARE complexes containing GST-Vam7p. Next, ATP regenerating system was added and the samples were incubated at 27°C for the indicated times before vacuoles were isolated by centrifugation and solubilized. Glutathione Sepharose was used to pull down GST-Vam7p and attached proteins were resolved by SDS-PAGE and imaged by Western blot. (B) WT versus pah1Δ vacuoles. (C) WT versus pah1Δ + pPAH1 vacuoles. (D) WT versus pah1Δ vacuoles + 200 µg/ml rPah1p. Densitometry values for Sec18p pull down were normalized against the corresponding pull down Vam7p value. Graph shows the normalized average ratios (n=3). The white bars represent the pah1Δ pulldown data from panel B. This is to facilitate visualization of the effect of adding recombinant Pah1. * P<0.05; ** P<0.001.

Figure 6. Vacuoles treated with the Pah1p inhibitor propranolol have a decreased level of Sec18p bound to cis-SNARE complexes

Vacuoles were harvested from BJ3505 yeast expressing CBP-Vam3p and assayed for Sec18p binding to cis-SNARE complexes. Fusion reactions were incubated at 27°C for the indicated times in the presence of propranolol or ATPγS or fusion buffer. At the indicated times, vacuoles were isolated by centrifugation and solubilized. Calmodulin agarose was used to pull down CBP-Vam3p and attached proteins were resolved by SDS-PAGE and imaged by immunoblotting. Fusion reactions were treated with PS buffer, 2 mM propranolol (A) or 1 mM ATPγS (B). (C) Liposome binding and effect of the nucleotide-binding state of Sec18p was tested using liposomes containing 10% PA. His₆-Sec18p (2 µg) was incubated with PA liposomes in binding buffer alone or in the presence of 1 mM Mg²⁺, 1 mM EDTA, or 1 mM ATPγS. Densitometry values for Sec18p pull down were normalized against the corresponding pull down Vam7p value. Graph shows the normalized average ratios (n=3). * P<0.05; ** P<0.001; *** P<0.0001.

Figure 7. Vacuoles treated with diC8-PA have a decrease in the level of Sec18p bound to cis-SNARE complexes

Vacuoles were harvested from BJ3505 and assayed for Sec18p binding to cis-SNARE complexes. Recombinant GST-Vam7p was added to fusion reactions and incubated at 27°C for 30 min in the absence of ATP to allow formation of cis-SNARE complexes containing GST-Vam7p. Next, ATP regenerating system was added and the samples were incubated at 27°C for the indicated times before vacuoles were isolated by centrifugation and solubilized. Glutathione Sepharose was used to pull down GST-Vam7p and protein complexes were resolved by SDS-PAGE and examined by Western blot. Fusion reactions were treated with PS buffer, 300 µM diC8-PA (A) or 300 µM diC8-DAG (B). Densitometry values for Sec18p pull down were normalized against the corresponding pull down Vam7p value. Graph shows the normalized average ratios (n=3). * P<0.05; ** P<0.001.

Figure 8. Working model of Sec18p regulation by PA

Sec18p associates with vacuole membranes through direct interactions with PA. Upon PA hydrolysis by Pah1p, Sec18p is no longer sequestered on the membrane away from cis-SNAREs and is recruited to them to carry out priming activity. Red ovals depict PA’s phosphate headgroup. Blue ovals depict generic lipid head groups.