Abstract

Structure determination for membrane proteins remains not only technically challenging but is further complicated by potentially non-native conformations resulting from removal of stabilizing lipids. This is especially true for mechanosensitive channels, which are designed to undergo gating transitions under membrane tension that negates part of the bilayer's lateral pressure. At tensions near its activation threshold (5-7 mN/m), the MscS channel slowly transitions from the resting to the inactivated state, which is tension insensitive. This interconversion between the two non-conductive states is accompanied by an 8 nm2 lateral expansion, revealing that the resting state must be more compact. We attempt to relate these two functional states with existing MscS structures experimentally and computationally. Molecular dynamics simulations used to verify experimental conductance and in-plane area expansion revealed that the many MscS structures are non-conductive or semi-conductive. These non-conductive structures share a splayed conformation of lipid-facing TM1-TM2 helical pairs, deep crevices between TM2 and TM3, and a de-coupling of the TM3a gate from the TM1-TM2 tension-transmitting domains. Yet, functional studies demonstrate the necessity of close hydrophobic TM2-TM3 contact for the resting to open transition; hydrophilic mutations lead to immediate inactivation instead of opening. This prompted the modeling of a “compact” resting state characterized by “upright” rather than splayed TM1-TM2 pairs and tighter TM2-TM3 contact that would reconnect the gate to TM1-TM2 helices. Simulated transitions from structures with splayed TM1-TM2 pairs to this compact state revealed two conserved stabilizing salt bridges: R59-D67 and D62-R131. Charge-inverting and neutralizing mutations at these residues had drastic effects on opening and closing rates, inactivation, and especially recovery from inactivation. Disulfide pairs introduced at these salt bridge positions support their proximity as predicted by the model. A cryo-EM characterization of the disulfide cross-linked channel is in progress.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call