Mark J. Arcario

Open-channel structures of multiple pentameric ligand-gated ion channels (pLGICs) have been determined, including the prokaryotic model pLGIC, Erwinia ligand-gated ion channel (ELIC). For many of these structures, it remains uncertain whether they represent a physiologic open-channel state because the conditions used for structure determination do not match those of functional measurements in cell membranes. Here, MD simulation is used to examine the ion conduction properties of the ELIC open-channel structure, which was determined using a non-desensitizing mutant called ELIC5. Results from simulations show that the pore remains stably open on the microsecond timescale, but computational electrophysiology measurements demonstrate a large outward rectification and an inward conductance that is significantly lower than experiment. This discrepancy is attributed to a constricted extracellular domain (ECD), which restricts the passage of ions between the ECD vestibule and extracellular solution. Unbiased MD simulation of the ELIC5 structure demonstrates spontaneous widening of an intersubunit space in the ECD to expose a lateral fenestration, which becomes the dominant ion conduction pathway. Computational electrophysiology of the ELIC5 MD-refined structures with a widened lateral fenestration shows better agreement with experimental single-channel recordings. Mutations of residues along the lateral ion conduction pathway show reduced single-channel conductance, supporting the importance of the lateral fenestration for ion conduction in a cation-selective pLGIC.

Open-channel structures of multiple pentameric ligand-gated ion channels (pLGIC) have been determined, including the prokaryotic model pLGIC, ELIC ( Erwinia ligand-gated ion channel). For many of these structures, it remains uncertain whether they represent the physiologic open-channel state because the conditions used for structure determination do not match those of functional measurements in cell membranes. Here, molecular dynamics (MD) simulation is used to examine the ion conduction properties of the ELIC open-channel structure, which was determined using a non-desensitizing mutant called ELIC5. Results from simulations show that the pore remains stably open on the microsecond timescale, but computational electrophysiology measurements demonstrate a large outward rectification, and an inward conductance that is significantly lower than experiment. This discrepancy is attributed to a constricted extracellular domain (ECD), which restricts the passage of ions between the ECD vestibule and extracellular solution. Unbiased MD simulation of the ELIC5 structure demonstrates spontaneous widening of an intersubunit space in the ECD to expose a lateral fenestration which becomes the dominant ion conduction pathway. Computational electrophysiology of the ELIC5 MD-refined structures with a widened lateral fenestration shows better agreement with experimental single-channel recordings. Mutations of residues along the lateral ion conduction pathway show reduced single-channel conductance, supporting the importance of the lateral fenestration for ion conduction in a cation-selective pLGIC.

Nanodiscs, soluble membrane mimetics composed of an amphipathic membrane scaffold protein encircling a lipid bilayer, are widely used in biophysical and structural studies of membrane proteins. Because many membrane proteins are responsive to their membrane environment, through specific protein-lipid interactions and bulk membrane shape and structure, it is important to understand the properties of lipid bilayers contained within nanodiscs in order to interpret studies using this technology. Nanodiscs are known to alter lipid properties, such as membrane thickness and melting temperature, and interactions with the nanodisc rim have been hypothesized to produce local perturbations in lipid structure and dynamics. Larger nanodiscs should compensate for this effect with a larger unperturbed area. To test this hypothesis, we examined the lipid bilayer properties of several lipids (DMPC, DPPC, POPC, DSPC) and soy polar extract in circularized nanodiscs of 11 nm to 50 nm diameter using the environmentally-sensitive fluorophore, Laurdan. In nanodiscs containing a single lipid type, as nanodisc size increased, lipid packing, melting temperature, and cooperativity better approximated the properties of that lipid in large unilamellar vesicles (LUVs). In spNW50 (50 nm nanodisc), the lipid packing and melting temperature were identical to LUVs. However, nanodiscs containing soy polar lipids did not follow this trend suggesting that complex lipid mixtures may produce preferential incorporation of lipids into the nanodisc or nonhomogeneous distribution of lipids within the nanodisc.

Lipid nanodiscs have become a standard tool for studying membrane proteins, including using single particle cryo-electron microscopy (cryo-EM). We find that reconstituting the pentameric ligand-gated ion channel (pLGIC), Erwinia ligand-gated ion channel (ELIC), in different nanodiscs produces distinct structures by cryo-EM. The effect of the nanodisc on ELIC structure extends to the extracellular domain and agonist binding site. Additionally, molecular dynamic simulations indicate that nanodiscs of different size impact ELIC structure and that the nanodisc scaffold directly interacts with ELIC. These findings suggest that the nanodisc plays a crucial role in determining the structure of pLGICs, and that reconstitution of ion channels in larger nanodiscs may better approximate a lipid membrane environment.

Polyunsaturated fatty acids (PUFAs) inhibit pentameric ligand-gated ion channels (pLGICs) but the mechanism of inhibition is not well understood. The PUFA, docosahexaenoic acid (DHA), inhibits agonist responses of the pLGIC, ELIC, more effectively than palmitic acid, similar to the effects observed in the GABA_A receptor and nicotinic acetylcholine receptor. Using photo-affinity labeling and coarse-grained molecular dynamics simulations, we identified two fatty acid binding sites in the outer transmembrane domain (TMD) of ELIC. Fatty acid binding to the photolabeled sites is selective for DHA over palmitic acid, and specific for an agonist-bound state. Hexadecyl-methanethiosulfonate modification of one of the two fatty acid binding sites in the outer TMD recapitulates the inhibitory effect of PUFAs in ELIC. The results demonstrate that DHA selectively binds to multiple sites in the outer TMD of ELIC, but that state-dependent binding to a single intrasubunit site mediates DHA inhibition of ELIC.

Polyunsaturated fatty acids (PUFAs) inhibit pentameric ligand-gated ion channels (pLGICs) but the mechanism of inhibition is not well understood. The PUFA, docosahexaenoic acid (DHA), inhibits agonist responses of the pLGIC, ELIC, more effectively than palmitic acid, similar to the effects observed in the GABA _A receptor and nicotinic acetylcholine receptor. Using photo-affinity labeling and coarse-grained molecular dynamics simulations, we identified two fatty acid binding sites in the outer transmembrane domain (TMD) of ELIC. Fatty acid binding to the photolabeled sites is selective for DHA over palmitic acid, and specific for an agonist-bound state. Hexadecyl-methanethiosulfonate modification of one of the two fatty acid binding sites in the outer TMD recapitulates the inhibitory effect of PUFAs in ELIC. The results demonstrate that DHA selectively binds to multiple sites in the outer TMD of ELIC, but that state-dependent binding to a single intrasubunit site mediates DHA inhibition of ELIC.

While small molecules have been used to induce anesthesia in a clinical setting for well over a century, a detailed understanding of the molecular mechanism remains elusive. In this study, we utilize ab initio calculations to develop a novel set of CHARMM-compatible parameters for the ubiquitous modern anesthetics desflurane, isoflurane, sevoflurane, and propofol for use in molecular dynamics (MD) simulations. The parameters generated were rigorously tested against known experimental physicochemical properties including dipole moment, density, enthalpy of vaporization, and free energy of solvation. In all cases, the anesthetic parameters were able to reproduce experimental measurements, signifying the robustness and accuracy of the atomistic models developed. The models were then used to study the interaction of anesthetics with the membrane. Calculation of the potential of mean force for inserting the molecules into a POPC bilayer revealed a distinct energetic minimum of 4–5 kcal/mol relative to aqueous solution at the level of the glycerol backbone in the membrane. The location of this minimum within the membrane suggests that anesthetics partition to the membrane prior to binding their ion channel targets, giving context to the Meyer–Overton correlation. Moreover, MD simulations of these drugs in the membrane give rise to computed membrane structural parameters, including atomic distribution, deuterium order parameters, dipole potential, and lateral stress profile, that indicate partitioning of anesthetics into the membrane at the concentration range studied here, which does not appear to perturb the structural integrity of the lipid bilayer. These results signify that an indirect, membrane-mediated mechanism of channel modulation is unlikely.