Stimulation of NSF ATPase activity by alpha-SNAP is required for SNARE complex disassembly and exocytosis

Abstract

N-ethylmaleimide-sensitive fusion protein (NSF) and alpha-SNAP play key roles in vesicular traffic through the secretory pathway. In this study, NH2- and COOH-terminal truncation mutants of alpha-SNAP were assayed for ability to bind NSF and stimulate its ATPase activity. Deletion of up to 160 NH2-terminal amino acids had little effect on the ability of alpha-SNAP to stimulate the ATPase activity of NSF. However, deletion of as few as 10 COOH-terminal amino acids resulted in a marked decrease. Both NH2-terminal (1-160) and COOH-terminal (160-295) fragments of alpha-SNAP were able to bind to NSF, suggesting that alpha-SNAP contains distinct NH2- and COOH-terminal binding sites for NSF. Sequence alignment of known SNAPs revealed only leucine 294 to be conserved in the final 10 amino acids of alpha-SNAP. Mutation of leucine 294 to alanine (alpha-SNAP(L294A)) resulted in a decrease in the ability to stimulate NSF ATPase activity but had no effect on the ability of this mutant to bind NSF. alpha-SNAP (1-285) and alpha-SNAP (L294A) were unable to stimulate Ca2+-dependent exocytosis in permeabilized chromaffin cells. In addition, alpha-SNAP (1-285), and alpha-SNAP (L294A) were able to inhibit the stimulation of exocytosis by exogenous alpha-SNAP. alpha-SNAP, alpha-SNAP (1-285), and alpha-SNAP (L294A) were all able to become incorporated into a 20S complex and recruit NSF. In the presence of MgATP, alpha-SNAP (1-285) and alpha-SNAP (L294A) were unable to fully disassemble the 20S complex and did not allow vesicle-associated membrane protein dissociation to any greater level than seen in control incubations. These findings imply that alpha-SNAP stimulation of NSF ATPase activity may be required for 20S complex disassembly and for the alpha-SNAP stimulation of exocytosis.

Figure 1

A schematic diagram of the α-SNAP truncation mutants used in this study and their ability to stimulate the ATPase activity of NSF compared to wild-type α-SNAP. (Top left) Schematic diagram of the α-SNAP truncation mutants used in this study. (Top right) Extent of stimulation of NSF ATPase activity as a percentage of stimulation by wild-type α-SNAP. 2 μg of full-length α-SNAP or α-SNAP truncation mutant were preimmobilized to the surface of plastic microfuge tubes for 20 min on ice. α-SNAP was removed from the tubes, and 1 μg of NSF was added for 1 h at 37°C. NSF ATPase activity was measured using a spectrophotometric assay, and the results were expressed as a percentage of the wild-type α-SNAP stimulation of NSF. Corrections were made for preexisting phosphates and contaminating ATPases by subtracting values from duplicate samples run with NEM-inactivated NSF. Data was pooled from three to seven separate assays for each SNAP mutant. (Bottom) Alignment of database sequences of bovine α-SNAP, β-SNAP, γ-SNAP, C. elegans SNAP, squid SNAP, Drosophila SNAP, and yeast sec17p, showing leucine 294 to be the only conserved residue in the last 10 amino acids of α-SNAP.

Figure 1

A schematic diagram of the α-SNAP truncation mutants used in this study and their ability to stimulate the ATPase activity of NSF compared to wild-type α-SNAP. (Top left) Schematic diagram of the α-SNAP truncation mutants used in this study. (Top right) Extent of stimulation of NSF ATPase activity as a percentage of stimulation by wild-type α-SNAP. 2 μg of full-length α-SNAP or α-SNAP truncation mutant were preimmobilized to the surface of plastic microfuge tubes for 20 min on ice. α-SNAP was removed from the tubes, and 1 μg of NSF was added for 1 h at 37°C. NSF ATPase activity was measured using a spectrophotometric assay, and the results were expressed as a percentage of the wild-type α-SNAP stimulation of NSF. Corrections were made for preexisting phosphates and contaminating ATPases by subtracting values from duplicate samples run with NEM-inactivated NSF. Data was pooled from three to seven separate assays for each SNAP mutant. (Bottom) Alignment of database sequences of bovine α-SNAP, β-SNAP, γ-SNAP, C. elegans SNAP, squid SNAP, Drosophila SNAP, and yeast sec17p, showing leucine 294 to be the only conserved residue in the last 10 amino acids of α-SNAP.

Figure 2

Effect on NSF ATPase activity of α-SNAP, α-SNAP (1–160), α-SNAP (Δ160–200), α-SNAP (1–285), and α-SNAP (L294A) over a range of concentrations. (A) Standard NSF ATPase assays were performed with several concentrations of full-length α-SNAP, α-SNAP (1–160), or α-SNAP (Δ160–200). (B) NSF ATPase assays with α-SNAP, α-SNAP (1–285), or α-SNAP (L294A). The data have been normalized to the maximal stimulation due to full-length α-SNAP. This was calculated by substracting NEM-insensitive ATPase activity and then calculating the relative differences between NSF ATPase activity in the presence or absence of α-SNAPs.

Figure 2

Effect on NSF ATPase activity of α-SNAP, α-SNAP (1–160), α-SNAP (Δ160–200), α-SNAP (1–285), and α-SNAP (L294A) over a range of concentrations. (A) Standard NSF ATPase assays were performed with several concentrations of full-length α-SNAP, α-SNAP (1–160), or α-SNAP (Δ160–200). (B) NSF ATPase assays with α-SNAP, α-SNAP (1–285), or α-SNAP (L294A). The data have been normalized to the maximal stimulation due to full-length α-SNAP. This was calculated by substracting NEM-insensitive ATPase activity and then calculating the relative differences between NSF ATPase activity in the presence or absence of α-SNAPs.

Figure 3

Reversal of the α-SNAP (L294A) mutation restores the ability to stimulate NSF ATPase activity. Standard assays of NSF ATPase activity were performed with 5 μg/tube of α-SNAP, α-SNAP (L294A), or α-SNAP (A294L), shown as α-SNAP^R. The data shows levels of ATP hydrolysis with untreated and with NEM-treated NSF. The intrinsic ATPase activity of NSF was not reduced by α-SNAP (L294A).

Figure 4

Binding of NSF to immobilized α-SNAP mutants. 2 μg of full-length SNAP or α-SNAP mutant was preimmobilized to the surface of polypropylene tubes for 20 min on ice. After washing with 1 ml SWB, the tubes were incubated with 2 μg of NSF in NBB for 10 min. The tubes were washed with 1 ml NBB, and bound proteins solublized with 50 μl of SDS buffer. The samples were then analyzed by SDS-PAGE and detected by silver-staining. A–C show results from different experiments. Note that no NSF binding was detected in control tubes without added SNAPs (first lane of each part).

Figure 4

Binding of NSF to immobilized α-SNAP mutants. 2 μg of full-length SNAP or α-SNAP mutant was preimmobilized to the surface of polypropylene tubes for 20 min on ice. After washing with 1 ml SWB, the tubes were incubated with 2 μg of NSF in NBB for 10 min. The tubes were washed with 1 ml NBB, and bound proteins solublized with 50 μl of SDS buffer. The samples were then analyzed by SDS-PAGE and detected by silver-staining. A–C show results from different experiments. Note that no NSF binding was detected in control tubes without added SNAPs (first lane of each part).

Figure 4

Binding of NSF to immobilized α-SNAP mutants. 2 μg of full-length SNAP or α-SNAP mutant was preimmobilized to the surface of polypropylene tubes for 20 min on ice. After washing with 1 ml SWB, the tubes were incubated with 2 μg of NSF in NBB for 10 min. The tubes were washed with 1 ml NBB, and bound proteins solublized with 50 μl of SDS buffer. The samples were then analyzed by SDS-PAGE and detected by silver-staining. A–C show results from different experiments. Note that no NSF binding was detected in control tubes without added SNAPs (first lane of each part).

Figure 5

α-SNAP (1–285) and α-SNAP (L294A) are unable to stimulate catecholamine release from digitonin-permeabilized adrenal chromaffin cells but inhibit the α-SNAP stimulation of exocytosis. (A) Adrenal chromaffin cells were permeabilized for 45 min with permeabilization buffer and stimulated with 10 μM Ca²⁺ with or without 25 μg/ml of full-length α-SNAP, α-SNAP (1–285), α-SNAP (L294A), or the reverse mutation α-SNAP (A294L) for 30 min, and released catecholamine was assayed. The data were pooled from seven separate experiments, and the means were expressed as a percentage of the 10 μM Ca²⁺ control stimulation ± SEM. (B) Adrenal chromaffin cells were permeabilized for 20 min with permeabilization buffer, incubated with KGEP buffer with or without 25 μg/ml of recombinant α-SNAP (1–160), α-SNAP (1–285), or α-SNAP (L294A) for 25 min, and then subsequently stimulated with 10 μM Ca²⁺ with or without 25 μg/ml full-length α-SNAP for 30 min, and released catecholamine was assayed. The data was pooled from four separate experiments, and the means were expressed as a percentage of the 10 μM Ca²⁺ control stimulation ± SEM.

Figure 5

α-SNAP (1–285) and α-SNAP (L294A) are unable to stimulate catecholamine release from digitonin-permeabilized adrenal chromaffin cells but inhibit the α-SNAP stimulation of exocytosis. (A) Adrenal chromaffin cells were permeabilized for 45 min with permeabilization buffer and stimulated with 10 μM Ca²⁺ with or without 25 μg/ml of full-length α-SNAP, α-SNAP (1–285), α-SNAP (L294A), or the reverse mutation α-SNAP (A294L) for 30 min, and released catecholamine was assayed. The data were pooled from seven separate experiments, and the means were expressed as a percentage of the 10 μM Ca²⁺ control stimulation ± SEM. (B) Adrenal chromaffin cells were permeabilized for 20 min with permeabilization buffer, incubated with KGEP buffer with or without 25 μg/ml of recombinant α-SNAP (1–160), α-SNAP (1–285), or α-SNAP (L294A) for 25 min, and then subsequently stimulated with 10 μM Ca²⁺ with or without 25 μg/ml full-length α-SNAP for 30 min, and released catecholamine was assayed. The data was pooled from four separate experiments, and the means were expressed as a percentage of the 10 μM Ca²⁺ control stimulation ± SEM.

Figure 6

α-SNAP (1–285) and α-SNAP (L294A) associate with but are unable to support dissociation of the 20S complex. (A) A detergent extract of rat brain membrane proteins was incubated with 15 μg of NSF, 30 μg α-SNAPs for 30 min with 0.5 mM MgATP or MgATPγS as indicated at 4°C. Proteins were immunoprecipitated with an antisyntaxin antibody conjugated to protein G–Sepharose, and bound proteins were solubilized with SDS sample buffer and separated on a 12.5% polyacrylamide gel. Proteins were detected using specific antisera to NSF, α-SNAP, syntaxin, and VAMP. (B) Extracts were incubated without (control) or with added α-SNAP and with 15 μg NSF in the presence of 0.5 mM MgATP or MgATPγS as indicated. Endogenous α-SNAP in control incubations was sufficient to recruit exogenous NSF. (C) Extracts were incubated with NSF with no added α-SNAP (control) or with added α-SNAPs as indicated. The presence of VAMP in syntaxin immunoprecipitates was determined by immunoblotting, and the amount of VAMP dissociated in the presence of MgATP was calculated as a percentage of the amount of bound VAMP in MgATPγS incubations.The data shown are the mean values from two experiments.

Figure 6

α-SNAP (1–285) and α-SNAP (L294A) associate with but are unable to support dissociation of the 20S complex. (A) A detergent extract of rat brain membrane proteins was incubated with 15 μg of NSF, 30 μg α-SNAPs for 30 min with 0.5 mM MgATP or MgATPγS as indicated at 4°C. Proteins were immunoprecipitated with an antisyntaxin antibody conjugated to protein G–Sepharose, and bound proteins were solubilized with SDS sample buffer and separated on a 12.5% polyacrylamide gel. Proteins were detected using specific antisera to NSF, α-SNAP, syntaxin, and VAMP. (B) Extracts were incubated without (control) or with added α-SNAP and with 15 μg NSF in the presence of 0.5 mM MgATP or MgATPγS as indicated. Endogenous α-SNAP in control incubations was sufficient to recruit exogenous NSF. (C) Extracts were incubated with NSF with no added α-SNAP (control) or with added α-SNAPs as indicated. The presence of VAMP in syntaxin immunoprecipitates was determined by immunoblotting, and the amount of VAMP dissociated in the presence of MgATP was calculated as a percentage of the amount of bound VAMP in MgATPγS incubations.The data shown are the mean values from two experiments.

Figure 6

α-SNAP (1–285) and α-SNAP (L294A) associate with but are unable to support dissociation of the 20S complex. (A) A detergent extract of rat brain membrane proteins was incubated with 15 μg of NSF, 30 μg α-SNAPs for 30 min with 0.5 mM MgATP or MgATPγS as indicated at 4°C. Proteins were immunoprecipitated with an antisyntaxin antibody conjugated to protein G–Sepharose, and bound proteins were solubilized with SDS sample buffer and separated on a 12.5% polyacrylamide gel. Proteins were detected using specific antisera to NSF, α-SNAP, syntaxin, and VAMP. (B) Extracts were incubated without (control) or with added α-SNAP and with 15 μg NSF in the presence of 0.5 mM MgATP or MgATPγS as indicated. Endogenous α-SNAP in control incubations was sufficient to recruit exogenous NSF. (C) Extracts were incubated with NSF with no added α-SNAP (control) or with added α-SNAPs as indicated. The presence of VAMP in syntaxin immunoprecipitates was determined by immunoblotting, and the amount of VAMP dissociated in the presence of MgATP was calculated as a percentage of the amount of bound VAMP in MgATPγS incubations.The data shown are the mean values from two experiments.