Fig 1.
BoNT/A-cleaved SNAP-25 is heavily phosphorylated at the Ser187 residue.
(A) SH-SY5Y cells, PC-12 cells, and HEK293 cells were treated with 10 nM BoNT/A holotoxin for 96 hours. Immunoblots, shown in lower and higher exposures, include two technical replicates for each condition and are representative of 3 independent biological replicates. The bar graph demonstrates the % of total pSNAP-25 Ser187 phosphorylation levels of PC-12 cells, normalized to total SNAP-25 levels. (B) mESC-derived motor neurons were treated with 500 pM BoNT/A for 4 hours, and Toosendanin and Bafilomycin were utilized as BoNT inhibitor controls. Representative blots of 3 independent experiments are shown. (C) HEK293 cells were transfected with WT SNAP-25 plasmid for 4 hours, followed by 10 nM BoNT/A holotoxin treatment for 72 hours. Western blot images are representative of 4 independent biological replicates. (D) hESC-derived motor neurons were treated with 500 pM BoNT/A holotoxin for 4 hours, and Toosendanin was utilized as a BoNT/A inhibitor. The immunoblots are representative of 5 independent biological replicates. GAPDH and β-Actin served as loading controls. In all conditions, the extent of % pSNAP-25 Ser187 phosphorylation levels was measured, and normalized to total SNAP-25 levels. *: P ≤ 0.05; **: P ≤ 0.01; ***: P ≤ 0.001.
Fig 2.
Ser187 phosphorylated SNAP-25 is efficiently cleaved in cells.
(A) mESC-derived motor neurons were treated with 1 µM PMA for 30 min and then intoxicated with 500 pM BoNT/A holotoxin. Cells were collected at the indicated time points, respectively to the intoxication start time point, processed, and subjected to western blotting. Representative blots include two technical replicates for each condition and are representative of 3 independent biological replicates. The graph represents quantified % SNAP-25 cleavage in the presence and absence of PMA treatment at the indicated time points (B) HEK293 cells were transfected with WT SNAP-25 and/or YFP-tagged BoNT/A LC plasmids for 18 hours (n = 3). 1 µM PMA was supplied to PMA+ samples 1 hour following the transfection initiated. (C) WT SNAP-25, SNAP-25 S187A, and SNAP-25 S187D plasmids were individually or co-transfected with BoNT/A LC in HEK293 cells for 18 hours (n = 3). (D) mESC-derived motor neurons were treated with 500 pM BoNT/A holotoxin for 4 hours (n = 3). Toosendanin and Bafilomycin, well-known BoNT inhibitors, and PKC activator PMA were used as controls and added to the culture 30 min before intoxication started. (E) mESC-derived motor neurons were treated with increasing concentrations of BoNT/A holotoxin (0.5 nM – 20 nM) for 4 hours, and Toosendanin was supplied to the relevant sample 30 min before the intoxication (n = 3). At the end of each experiment, cell lysates were prepared and subjected to immunoblotting. ns: non-significant; *: P ≤ 0.05; **: P ≤ 0.01; ***: P ≤ 0.001.
Fig 3.
SNAP-25 (1-197) is a better substrate for Ser187 phosphorylation.
HEK293 cells were transfected with either WT SNAP-25 or SNAP-25 (1-197) for 18h, and treated without (A) (n = 4), and with (B) (n = 4) PMA, to determine SNAP-25 Ser187 phosphorylation levels via western blotting. PMA was supplied to the cells 1 hour following the transfection initiated. Data was normalized to respective loading controls, WT SNAP-25 or SNAP-25 (1-197). (C) To measure the stability of Ser187 phosphorylation as compared to that of Thr138, PC12 cells were treated with 1 µM PMA, and samples were collected at the indicated time points (0.5h - 18h) following the PMA treatment, and were analyzed with immunoblotting (n = 3). (D) HEK293 cells were transfected with either WT SNAP-25 or SNAP-25 (1-197) for 18h and treated with 1 µM PMA. The samples were collected at indicated time points (0.5h – 12h) following the PMA treatment and subjected to western blotting (n = 3). ns: non-significant; *: P ≤ 0.05; **: P ≤ 0.01; ***: P ≤ 0.001.
Fig 4.
SNAP-25 (1-197) localizes to the membrane and is heavily phosphorylated at Ser187.
(A) Subcellular distribution of SNAP25 and SNAP-25 (1-197) was analyzed by cell fractionation. HEK293 cells were co-transfected with WT SNAP-25 and SNAP-25 (1-197) and treated with and without 1 µM PMA for 1.5 hours. Cells were homogenized, followed by membrane extraction. The total lysate, as well as the soluble membrane and the cytosolic fractions, were analyzed by western blotting with the indicated antibodies. Data representative of 3 independent experiments. The bar graph demonstrates % pSNAP-25 Ser187 levels on the membrane upon PMA treatment, normalized to respective loading controls, WT SNAP-25 or SNAP-25 (1-197). (B) PC12 cells, which endogenously express SNAP-25, were transfected with BoNT/A LC in conditions to achieve a low-level SNAP-25 cleavage and treated with and without 1 µM PMA for 1.5 hours. The total lysate, the soluble membrane, cytosol, and non-soluble fractions were subjected to western blotting with the indicated antibodies. Representative blots of 3 independent experiments are shown. % pSNAP-25 Ser187 levels on the membrane upon PMA treatment were calculated upon normalization to respective loading controls, WT SNAP-25 or SNAP-25 (1-197), which is presented in the bar graph. (C) BoNT/A - cleaved SNAP-25 is detected in HEK293 cells longer than BoNT/A LC. HEK293 cells were transiently transfected with plasmids encoding YFP-tagged BoNT/A LC (n = 4), WT SNAP-25 (n = 3), or SNAP-25 (1-197) (n = 3), and cell lysates were collected at the indicated time points and analyzed with western blotting with the specified antibodies. Western blot band quantification results were normalized to loading controls (β-Actin) that were reprobed in each blot, and the results are presented as mean ± SD. (D) SNAP-25 (1-197) persists longer than BoNT/A LC in PC12 cells. PC12 cells were transiently transfected with YFP-tagged BoNT/A LC, samples were collected at the indicated time points, and subjected to immunoblotting. BoNT/A LC (detected by GFP) levels were determined upon normalization to the loading control. % SNAP-25 (1-197) compared to total SNAP-25 was calculated by measuring the intensities of SNAP-25 WT and SNAP-25 (1-197) bands in each time point. The statistical significance shown in the bar graphs represents the comparison of protein levels at each time point to the Day 1 level. ns: non-significant; *: P ≤ 0.05; **: P ≤ 0.01; ***: P ≤ 0.001.
Fig 5.
SNAP-25 (1-197) binds to syntaxin-1A, and the interaction is enhanced by Ser187 phosphorylation.
(A) PC12 cells were transfected with an HA-tagged SNAP-25 (1-197) plasmid and then treated with 1 µM PMA (a total of 1.5 hours). Immunoprecipitation (IP) was conducted using rabbit polyclonal anti-HA antibodies or normal rabbit IgGs. Input and IP samples were subjected to immunoblotting to detect the levels of the indicated markers. Blots are representative of 4 independent experiments. Syntaxin-1A band intensities were normalized to HA levels that were reprobed on the same IP blots. Syntaxin-1A intensity in the PMA-treated HA immunoprecipitates was compared to non-PMA-treated conditions. (B) PC12 cells were transfected with the BoNT/A LC plasmid, and 1 µM PMA treatment was conducted in the indicated samples (a total of 1.5 hours). Immunoprecipitation was carried out using mouse monoclonal anti-SNAP-25 or normal mouse IgGs. Western blotting was conducted to detect the levels of the indicated markers in both input and IP samples. Representative blots of 4 independent experiments are shown. Syntaxin-1A intensity normalized to SNAP-25 levels in each IP blot was quantified. ns: non-significant; *: P ≤ 0.05; **: P ≤ 0.01; ***: P ≤ 0.001. (C and D) PMA treatment further decreases neurotransmitter release when a low level of SNAP-25 cleavage is present. PC12 cells were transfected with 1 µg YFP-tagged BoNT/A LC and treated with and without 1 µM PMA. Dopamine release was stimulated by high K+ solution treatment, and supernatant was analyzed via a triple quadruple LC-MS/MS system (Waters TQ-S Micro triple quadruple LC-MS/MS). (C) Cell lysates were collected following the high K+ solution treatment and subjected to immunoblotting. The percentage of cleaved SNAP-25 was calculated in each condition and presented in the bar graph. A separate, identical gel was run to assess PMA-mediated SNAP-25 phosphorylation. β-Actin was used as a loading control. (D) Mass spectrometry analysis of dopamine release from PC12 cells. The bar graph demonstrates the percentage of dopamine release relative to vehicle (DMSO) control. Data are representative of 3 independent experiments, and results are presented as mean ± SD. *: P ≤ 0.05; **: P ≤ 0.01; ***: P ≤ 0.001.
Fig 6.
Molecular modeling suggests phosphorylation of Ser187 does not impact SNAP-25-SNARE complex binding, but changes its local dynamics and electrostatic signature.
The four panels summarize the results of molecular modeling on the SNARE complexes. (A) Focus on the local changes induced in the interaction pattern around Ser187 of SNAP-25 and the distribution of relevant inter-side chains distances along the simulations. (B) Binding energy is calculated for each fragment against the other three in the complex (1 vs 3) and for each pair of fragments individually (1 vs 1). The latter is given in a matrix representation with an averaged value and corresponding standard deviation for each considered SNARE complex. The residues included in the calculation of binding energy are highlighted in the upper part of the panel. (C) Focus on the local changes induced in the electrostatic potential around Ser187 of SNAP-25. (D) Free energy profiles for the unwinding process of a small peptide, representative of the sequence around Ser187 in SNAP-25.
Fig 7.
Proposed model of the mechanism underlying the dominant negative effect of BoNT/A cleaved SNAP-25 on neurotransmitter release.
(1) BoNT/A LC cleaves SNAP-25; (2) Cleaved fragment, SNAP-25 (1-197), stays on the membrane; (3) SNAP-25 (1-197) gets efficiently phosphorylated at Ser187 residue; (4) Ser187 phosphorylation provides a competitive advantage for SNAP-25 (1-197) for syntaxin-1A binding over WT SNAP-25, which can lead to a dominant-negative effect.