Force distribution during the push-to-open gating process of NOMPC. (a) The simulation systems of NOMPC. The NOMPC molecule was divided into two subsystems, denoted by the orange and purple rectangular boxes, for molecular dynamics (MD) simulations. The interface between the Linker helices (LH) and the TRP domains was shown in a circular inset (bottom view). (b) The membrane-parallel (purple line) and membrane-normal (yellow line) components of the net forces on the TRP domain exerted by the LH domain. The error bars denote the standard deviations from two compressive trajectories (CI-1/2) and two free trajectories (FI-1/2). (c) The net forces on the TRP domain exerted by the LH domain. The grey quiver on each residue showed the net force exerted by the LH domain. The black showed the resultant force on the residues of the TRP domain. Two major points of force application, E1571 and R1581, were marked. ⊗ represented the inward force component. The data were averaged for the four subunits by rotationally symmetrized around the protein center.

A membrane-parallel torsion force can open NOMPC. (a-b) The side view (a) and bottom view (b) of the net-FDA of the E1571 and R1581 on the TRP domain exerted from the LH domain, which was decomposed into the membrane-parallel and membrane-normal components. ⊗ represented membrane-normal forces pointing into the membrane. (c-d) The transmembrane pore size evolution in the simulations with membrane-parallel forces (c) or membrane-normal forces (d), 3 repeats for each condition (trajectories: MPI-1/2/3, MNI-1/2/3, see Methods).

Torsional mechanical properties of the AR bundle. (a) The sketch of the steered molecular dynamics simulation. The dotted arrows revealed the direction of twisting. (b) The relationships between the torque on AR1-8 (left Y-axis), the length of the AR domain (right Y-axis), and the rotational angle of AR1-8. This was calculated from the rotating trajectories RII (0-4).

Force transmission along the AR domain. (a) The quiver plot of net-FDA (XY planar) of each (i +1)th AR unit exerted from the ith AR unit showed the force transmission from the intracellular terminal to the transmembrane side of the AR domain viewing from AR1 toward AR29. ⊙represented the outward Z-component force. (b) The distribution of the pairwise residue-residue net-FDA within AR9 to AR29 (weak interaction under 5 pN neglected). The inset showed local information over 70 pN. (c) The representative interaction heatmaps (AR13-AR15) showed the magnitude of non-bonded pairwise residue-residue net-FDA (left) and hydrogen bonds occupancy (right). (d) The Pearson correlation coefficient (PCC) between net-FDA and hbonds in AR units. Inset showed the mean value and standard deviation of Intra-AR (diagonal) and Inter-AR (sub-diagonal) PCCs.

A schematic diagram of the twist-to-open model. The side view (left) and bottom view (right) of the twist-to-open model. Both the compression and CW twisting of the AR domain can generate a membrane-parallel twisting force on the TRP domain, which is the key component for gating.

MD simulation systems.

Simulation trajectories of system I.

Simulation trajectories of system II.

Kinematic analysis of the push-to-open trajectory CI-1. (a) The simulation systems and domain structures. (b) The pore dilation process in the push-to-open trajectory. (c) The motion tendency of the LH domain. The purple arrows represent each α-C, and the orange arrows represent the mass center of the LH domain. The data were averaged for the four subunits by rotationally symmetrized around the protein center. (d) The rotation of the TRP domain, projected in XY plane (light purple line, positive values a clockwise rotation viewing from the intracellular side) and Z direction (dark purple line, the positive value meant a tilt-up rotation from intracellular side to extracellular side), respectively.

The pairwise FDA results of the push-to-open trajectory CI-1 between the AR29 domain, the LH domain, and the TRP domain. The size and color of each marker denoted the magnitude of the corresponding pairwise FDA results labeled in the two sides. The circle marker and square marker showed the FDA results between AR29-LH and TRP-LH, respectively.

The FDA results of the push-to-open trajectories between the LH domain and the TRP domain on the XY plane (TRP). The superimposed quiver plot of the FDA results on each residue of the TRP domain exerted from the LH domain, calculated from trajectories FI-1/2, CI-1/2.

Water molecules around pore region. (a) A snapshot of trajectory MPI-1 at 800 ns. The pore was dilated so water molecules (red) could spontaneously pass through. (b) The number of water molecules in the gate region around I1554, which was the narrowest part. We counted the water molecules whose oxygen atom was within 4 Å in the Z-direction from the α-C of I1554.

The torque (a) and rotational angle (b) versus time of trajectory RII, respectively. The black line illustrated the trajectory RII-0 (the second step of the rotation of system II), while the other colored lines starting from a circle depicted the third step of the rotation of system II at every 2.5-degree rotation for a duration of 100 ns in the trajectory RII-1/2/3/4. The solid line corresponded to the data of the AR1-8, whereas the dotted line corresponded to the LH domain (Fig. 3).

Location of AR13-15 in the AR spring.