The influence of cell membrane and SNAP25 linker loop on the dynamics and unzipping of SNARE complex

Abstract

The soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) complex is composed of three neuronal proteins VAMP2, Syntaxin and SNAP25, which plays a core role during the process of membrane fusion. The zipping assembly of the SNARE complex releases energies and drives the vesicle and cell membrane into close proximity. In this study, we use all-atom molecular dynamics simulations to probe the dynamics of SNARE and its unzipping process in the context of membrane at the atomistic details. Our results indicated that the NTD of SNARE core domain is relatively more stable than CTD, which is in agreement with previous experiments. More importantly, possible interactions between the linker loop (LL) region of SNAP25 and VAMP2 are observed, suggests that the LL region may facilitate VAMP2 binding and SNARE initiation. The forced unzipping of SNARE in the presence of membrane and LL of SNAP25 reveals the possible pathway for energy generation of SNARE zipping, provides information to understand how force may regulate the cooperativity between the membrane and the SNARE complex.

Fig 1. SNARE complex and the simulation setup.

(A)Schematic of ternary SNARE four helix bundle with Syntaxin(yellow), VAMP2(red), SNAP25(green). The notation of SNARE layers is indicated. The sequence of linker loop of SNAP25 is shown, 4 palmitoylated cysteines anchored to the membrane were shown in red. (B)Assembly of SNARE complex with the membrane, with the palmitoylated cysteines on SNAP25 highlighted. (C)The initial simulaiton system. The SNARE core domain is placed ~ 40° reletive to the membrane.

Fig 2. The free molecular dynamics simulations.

(A)The RMSD of three different simulations, resutls for SNARE core domain and LL region are shown. (B)The change of the angle between SNARE core domain and the membrane. (C)Change of the angle between the transmembrane domain of Syntaxin and the membrane.

Fig 3. LL region of SNAP25 forms interactions with VAMP2.

(A)snapshot of the free MD simulation. The interactions on the N-terminal residues of NTD region are highlighted, with VAMP2 residues 25 to 41 shown in gray and SNAP25 residues 125–130 shown in purple. A specific salt bridge between VAMP2 E41 and SNAP25 R124 is also indicated. (B)The time-course of the distance VAMP2 E41 and SNAP25 R124 for Free2, a distance below 3.2Å can be viewed as salt bridge fromed. (C)Number of contacts between of SNAP25 with VAMP2 residues 24 to 41.

Fig 4. The dynamics of each SNARE layer.

(A)Averaged root mean squared fluctuation of each layer in the simulations. (B) Time-course of the numbers of hydrogen bond between the layer residues with their surroundings for each layers. (C)Average number of Hydrogen bond between the layer residues with their surroundings for the last 20 ns of the two simulation (80–100 ns).

Fig 5. Steered molecular dynamics simulations.

(A)Snapshots for a representative unzipping simulation (FU2). (B)First panel: the force curve. Second panel: the number of contacts between VAMP2 residues 25 to 41 with SNAP 25. Third panel: the distance between VAMP2 E41 and SNAP25 R124 salt bridge. Fourth panel: the RMSD of the rest part of SNARE core domain when VAMP2 is pulled away. The pink area indicated the final detachment between VAMP2 and the rest of SNARE core domain, four dashed line represents the time of snapshots in (Fig 5A).

Fig 6. Analysis of SMD simulations.

(A)Force-extension curves of SMD simulations FU1, FU2 and FU4. The yellow, shaded area represents zero layer position. (B)The change of the SNARE orientation with the membrane during the simulations.