Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Sep 1.
Published in final edited form as: Structure. 2015 Jul 30;23(9):1655–1664. doi: 10.1016/j.str.2015.06.020

Role of the fourth transmembrane α-helix in the allosteric modulation of pentameric ligand-gated ion channels

Casey L Carswell 1,4, Camille M Hénault 1,4, Sruthi Murlidaran 2, JP Daniel Therien 1, Peter F Juranka 1, Julian A Surujballi 1, Grace Brannigan 2,3, John E Baenziger 1,*
PMCID: PMC4824752  NIHMSID: NIHMS773868  PMID: 26235032

Summary

The gating of pentameric ligand gated ion channels is sensitive to a variety of allosteric modulators that act on structures that are peripheral to those involved in the allosteric pathway leading from the agonist site to the channel gate. One such structure, the lipid-exposed transmembrane α-helix, M4, is the target of lipids, neurosteroids, and disease-causing mutations. Here we show that M4 interactions with the adjacent transmembrane α-helices, M1 and M3, modulate pLGIC function. Enhanced M4 interactions promote, while ineffective interactions reduce channel function. The interface chemistry governs the intrinsic strength of M4-M1/M3 inter-helical interactions, both influencing channel gating and imparting distinct susceptibilities to the potentiating effects of a lipid-facing M4 congenital myasthenic syndrome mutation. Through aromatic substitutions, functional studies, and molecular dynamics simulations, we elucidate a mechanism by which M4 modulates channel function.

Pentameric ligand-gated ion channels (pLGICs), such as the nicotinic acetylcholine receptor (nAChR), respond to neurotransmitter binding by transiently opening either cation- or anion-selective channels across the post-synaptic membrane. The sites for agonist binding are located at the interfaces between subunits in the extracellular domain (ECD), which extends away from the membrane surface into the synaptic cleft (Fig. 1)(Unwin, 2005). Agonist binding likely induces rigid body motions, which are translated into transient movements of the pore-lining M2 α-helices of the transmembrane domain (TMD) by a series of loops at the ECD/TMD interface (Althoff et al., 2014; Sauguet et al., 2014; Unwin and Fujiyoshi, 2012). Considerable attention has focused on elucidating the gating movements of these interfacial loops, which form the primary allosteric path leading from the agonist site to the channel gate (Grutter et al., 2005; Jha et al., 2007; Lee and Sine, 2005; Lummis et al., 2005). In contrast, structures not directly involved in the primary allosteric path have received less attention, even though a number of allosteric modulators influence gating via these auxiliary sites (Fig. 1a).

Figure 1. Structures of pLGICs with bound modulators.

Figure 1

Open in a new tab

A) Structures of the nAChR (2BG9), GluCl (3RFI), and GLIC (3P50). In each case, a single subunit is shown as a dark blue cartoon with the M4 α-helix highlighted in red. In the nAChR (left), residues in the agonist site are highlighted as red spheres, while those forming the transmembrane gate are highlighted as yellow spheres. In GluCl (center), the agonist glutamine, the positive modulator ivermectin, the open channel blocker picrotoxin (aligned using 3RI5), and a bound detergent molecule are highlighted as red, orange, yellow, and marine spheres, respectively. In GLIC (right), the inhibitor propofol and bound lipids (aligned from 3EAM) are shown as yellow and marine spheres, respectively. B) A single TMD subunit of ELIC (left, 2VL0) and GLIC (right, 4HFI) with residues at the M4-M1/M3 interface shown as spheres. Aromatic, polar hydrogen bonding, negative, positive, and aliphatic residues are highlighted in yellow, green, red, blue, and tan, respectively. The marine sphere corresponds to a water molecule.

The fourth transmembrane α-helix, M4, is located on the periphery of the TMD and is the target of both lipids and neurosteroids (Baenziger et al., 2015; Barrantes, 2003, 2015; Bocquet et al., 2009; Henault et al., 2014; Hosie et al., 2006; Paradiso et al., 2001). Lipid-facing mutations in M4 of the muscle-type nAChR influence channel gating, with at least one leading to a congenital myasthenic syndrome (CMS)(Bouzat et al., 1998; Lasalde et al., 1996; Lee et al., 1994; Li et al., 1992; Shen et al., 2006; Tamamizu et al., 2000). M4 extends beyond the bilayer to interact directly with the β6–β7 loop (often referred to as the Cys-loop), a key structure at the ECD/TMD interface that participates in channel gating. One model proposes that interactions between M4 and the adjacent α-helices, M1 and M3, are dynamic in that effective M4-M1/M3 interactions lead to M4/Cys-loop interactions that promote channel function, while ineffective M4-M1/M3 interactions eliminate M4/Cys-loop interactions to reduce channel function (Fig. 1b) (daCosta and Baenziger, 2009; daCosta et al., 2013). In this context, it is intriguing to note that of the four transmembrane α-helices, M1-M4, M4 exhibits the greatest sequence variability amongst the various Torpedo and human nAChR subunits. This variability should lead to subunit-specific interactions at the interface between M4 and M1/M3, resulting in variable interaction energies. If strong M4-M1/M3 interactions promote coupling between the agonist site and channel gate, then variable M4-M1/M3 interaction energies should lead to variable coupling efficiencies. nAChR subunits with weak M4-M1/M3 interactions should also be more sensitive to allosteric modulators that act on M4.

The two structurally well-characterized prokaryotic pLGICs, GLIC and ELIC (Fig. 1)(Bocquet et al., 2009; Hilf and Dutzler, 2008, 2009; Pan et al., 2012; Sauguet et al., 2013), are excellent models for probing the role of M4 in pLGIC function as both share a similar tertiary/quaternary fold, yet have distinct M4 conformations. In GLIC, M4 interacts tightly with M1/M3 along its entire length. In ELIC, the C-terminal half of M4 tilts away from M1/M3 with the final five residues unresolved in the crystal structure. Aromatic interactions are key determinants that energetically drive the binding of M4 to M1/M3 during folding of the homologous glycine receptor (Haeger et al., 2010). GLIC exhibits an extensive network of interacting aromatic residues at the M4-M1/M3 interface, including a cluster of three aromatic residues that may be essential for linking the C-terminus of M4 to both M1/M3 and the β6–β7 loop (see Figs. 1B and 3). Intriguingly, this C-terminal M4 aromatic cluster is absent in ELIC. Through aromatic substitutions, functional studies, and molecular dynamics simulations, we examine here the effects of aromatic residues at the M4-M1/M3 interface on the conformation of M4, and how the resulting changes in conformation influence channel function. We also examine whether TMD modulators influence pLGIC function by modulating M4-M1/M3 interactions.

Figure 3. Enhanced M4 binding to M1/M3 potentiates pLGIC function.

Figure 3

Open in a new tab

A) The TMD of a single subunit of an ELIC homology model (based on GLIC, 3EHZ). Aromatic residues conserved in GLIC and ELIC are shown in yellow, while aromatics removed from GLIC or inserted into ELIC are shown in orange, superimposed on WT ELIC residues (red). The M4 sequence alignments highlight key aromatic residues (boxed), identical (*), conserved (:) and semi-conserved (.) residues. B) Two-electrode data for GLIC (upper) and ELIC (lower), with representative mutants. Ligand concentration jumps (protons or cysteamine for GLIC and ELIC, respectively) are indicated with the horizontal bar. C) Dose response curves obtained for single aromatic-to-Ala substitutions in GLIC (upper panel) and from either single and multiple aliphatic-to-aromatic substitutions in ELIC (lower panels). Error bars represent SE. D) Changes in EC50 relative to the WT for GLIC (pH 5.03) and ELIC (0.92mM cys). NC, no current. Multiple aromatic substitutions were not generated for GLIC (X). Error bars represent SD. See table S2 for EC50 values. See also Tables S2 and Fig S3.

RESULTS

Aromatic residues promote M4-M1/M3 interactions

GLIC exhibits nine aromatic residues at the M4-M1/M3 interface (Fig. 1B) (labeled as aromatics (1) M4 F315, (2) M4 F314, (3) M3 Y254, (4) M4 F303, (5) M3 F265, (6) M3 Y266, (7) M1 W213, (8) M4 F299, and (9) M1 F216; see Fig. 3). Although the aromatics at positions 4, 7, 8, and 9 are conserved in ELIC, the entire M4 C-terminal aromatic cluster (aromatics 1–3) and the aromatic side chains at positions 5 and 6 on M3 are absent (see Figs. 1B and 3). We postulated that the distinct profiles of aromatic interactions at the M4-M1/M3 interface in GLIC and ELIC lead to the different conformations of M4 observed in the crystal structures. The different M4 conformations, however, could also result from differential crystal packing and/or detergent-solubilisation effects prior to crystallization (daCosta and Baenziger, 2013).

To probe whether aromatic substitutions at the M4-M1/M3 interfaces influence the conformation of M4 in a folded pLGIC structure located within a membrane environment, we turned to molecular dynamics simulations. Simulations were run for both wild type GLIC (WT-GLIC) and a mutant where the five non-conserved aromatic residues were mutated to Ala (5Ala-GLIC: aromatic-to-Ala substitutions at positions 1, 2, 3, 5, and 6). In both cases, simulations were performed using intact pentamers, as well as with a single-subunit-TMD. The latter simulations revealed intriguing lipid binding poses, which are discussed below. Finally, simulations run for both wild type ELIC and an ELIC mutant with aliphatic-to-aromatic substitutions at the same positions in the M4-M1/M3 interface were not informative because affected residues in M4 C-terminal are not defined in the ELIC crystal structure, thus precluding a defined starting conformation.

Consistent with our hypothesis, the simulations show that aromatic residues at the M4-M1/M3 interface influence the interactions of M4 with M1/M3. Specifically, the close contacts between residues along the entire length of M4 and those on M1/M3 in the GLIC crystal structure are maintained throughout the simulations with WT-GLIC. In contrast, the loss of the C-terminal aromatic cluster leads to a consistent tilting of the C-terminal half of M4 away from M1/M3, with closest Cα -Cα carbon atom contacts on M4-M1 and M4-M3 increasing by roughly 2 Å (Fig. 2A and Fig. S1 – the latter compares directly distances between Y/A254 on M3 and both F/A314 and F/A317 on M4). The differences in M4-M1/M3 interactions are statistically significant based on standard errors calculated across the five subunits. The magnitudes of the separations are larger than typical root-mean-squared-deviations among the transmembrane Cα atoms of different pLGICs (Bocquet et al., 2009; Hibbs and Gouaux, 2011; Hilf and Dutzler, 2009; Miller and Aricescu, 2014) or between different conformations of GLIC (Sauguet et al., 2013). The Cα carbon atom separations observed in the M4 C-terminal region contrast those observed between M4 and M1/M3 near the cytoplasmic side of the bilayer, where the Cα carbon atom separations in 5Ala-GLIC and WT-GLIC converge (Cα atoms separation differences are less than 1 Å) (Fig. 2A – 2C). The latter suggests that the aromatic interactions remaining in the intracellular leaflet of 5Ala-GLIC are sufficient to maintain effective M4-M1/M3 interactions in this region.

Figure 2. Aromatic residues promote M4-M1/M3 interactions.

Figure 2

Open in a new tab

A) Transmembrane domain of WT-GLIC and 5Ala-GLIC following 300 ns of simulation. Helices are shown in surface representation (M1, green; M2, blue; M3, dark gray; M4, orange), with substituted residues shown in stick representation (Y/A254, orange; F/A314, purple; F/A315, green). The M2–M3 loop (yellow) and a portion of the ECD (purple) closest to the interface with the TMD is also shown in surface representation. POPC lipid density, averaged over the final 200 ns of the simulation, is represented by a translucent blue isosurface; the submerged appearance of the 5Ala-GLIC M4 α-helices relative to those of WT-GLIC reflects significant lipid penetration of the 5Ala-GLIC subunits. B) Representative subunit from WT-GLIC (left) and 5Ala-GLIC (center), showing POPC lipids bound to the M1-M4 interface in stick representation with blue acyl chains and red PC headgroups. A rotated view (right) shows a second POPC molecule bound to 5Ala-GLIC, with cyan acyl chains and pink PC headgroups, straddling the 5Ala-GLIC M3-M4 interface. Lines represent the membrane-water interface. C) Average distances between Cα atoms on residues at similar register on opposing helices. Substituted residues are highlighted in red. Error bars represent the standard error across the five subunits. D) Free energy landscape (potential of mean force) for configurations of F314 and F315 relative to Y254 in four of five WT-GLIC subunits, as a function of angle between planar groups and centroid-centroid distance (Equations 1 and 2). In a fifth subunit, the aromatic cluster disassociates early in the simulation, as discussed further in Supplemental Information Fig. 2. The blue symbol in each panel indicates the value of the corresponding angle and distance determined from the crystal structure (PDB code: 4HFI). See also Table S1 and Figs. S1 and S2.

The tilt of the C-terminal half of M4 observed in the simulations of the 5Ala-GLIC mutant suggest that aromatic residues are essential for promoting effective M4-M1/M3 interactions. Note that the tilt of M4 away from M1/M3 observed in simulations of 5Ala-GLIC are similar, but of lesser magnitude to the tilt of M4 observed in the crystal structure of ELIC. In the ELIC crystal structure, the terminal 5 residues are unresolved, suggesting weak, if any, interactions between the M4 C-terminus and M1/M3. The ELIC crystal structure supports the conclusion that aromatic residues at the M4-M1/M3 interface promote M4 interactions with M1/M3. Detergent solubilization and the removal of lipids may perturb the intrinsically weak interactions between the M4 C-terminus and M1/M3 leading to a greater disruption of M4 conformation than observed in the simulations.

The aromatic-to-Ala substitutions have a substantial effect on the energetics of M4-M1/M3 interactions. In WT-GLIC, M4 C-terminal aromatics 1, 2, and 3 are involved in pairwise interactions within the aromatic cluster, with Ala substitutions of these residues leading to energetic penalties of >1 kcal/mol (Table S1). For four of the five subunits, the two-dimensional free energy landscape (Fig. 2C) calculated from the relative orientations of M3:Y254 (3) and M4:F315 (1) aromatic groups has a strong angular dependence consistent with π-π stacking interactions, with the minima near 90° indicating a T-shaped conformation similar to that found in the crystal structure. In the fifth subunit, F315 dissociates from the cluster and rotates to face the lipids, as also observed in the single-subunit-TMD simulations (Fig. S1).

M4 C-terminal aromatics 1, 2, and 3 are also involved in pair-wise interactions with other residues that strengthen M4-M3 interactions. An additional strong energy penalty is associated with the loss of a hydrogen bond between the tyrosine hydroxyl of M3:Y254 (aromatic 3) and the carbonyl oxygen of M4:N307. These two residues hydrogen bond via a bridging water in four out of five chains of the highest resolution crystal structure for GLIC (Sauguet et al., 2013). In the simulations, bridging waters are observed transiently, interspersed with direct hydrogen bonding between the two residues. Also, the mutations weakened several pairwise interactions involving aromatic substituents and non-aromatic polar or hydrophobic residues (Table S1).

One intriguing finding of the simulations is that the tilt of the M4 C-terminus away from M1/M3 and/or the reduced side chain volume in 5Ala-GLIC leads to a change in lipid binding. In WT-GLIC, POPC molecules adopt poses at the edge of the M4-M1 and M4-M3 interfaces, as in the GLIC crystal structures. In 5Ala-GLIC, the POPC molecules penetrate deeper into the M1/M3/M4 α-helical bundle. Entire acyl chains become embedded at the M1-M4 interface (Figure 2C and 2D). A second POPC assumes a pose in a cavity formed by the C-terminal end of M4, the M2–M3 loop, and the β6–β7 loop (Figure 2D), where one acyl chain fills the increased free volume vacated by M4 F314 and F315 upon mutation of each residue to Ala, while the other acyl chain remains in contact with the bulk membrane (Figure 2D). Such interactions are seen across subunits, with one lipid at least partially buried in each of these sites. Buried lipids could potentially mediate some of the effects of the aromatic substitutions, by direct interactions with the M2–M3 and β6–β7 loop and/or indirectly by stabilizing M4 in a conformation with reduced interactions with the ECD. This possibility underscores the potential significance of even slight conformational changes in M4, particularly if they increase the internal free volume at the M4-M1/M3 interface above the volume required to accommodate a buried lipid. Note also that in the single-subunit-TMD simulations, the absence steric conflicts with the ECD allows even deeper penetration of the lipid. It appears that even subtle changes in conformation can dramatically alter lipid binding (Fig. S3).

Weakened M4-M1/M3 interactions inhibit channel function

To test experimentally whether the M4 conformation influences channel function, non-conserved aromatic residues in GLIC were individually mutated to Ala to weaken M4-M1/M3 interactions, and the effects of the individual substitutions on channel function assessed using the two electrode voltage clamp apparatus. WT-GLIC gates open in response to protons, with a pH value required to elicit half maximal channel gating of pH50 = 5.03 ± 0.08 (n = 38). Each individual aromatic-to-Ala mutation led to a right-shift in the dose response to protons, with the pH50 values decreasing by ~0.4 to ~0.9 pH units - the Y266A mutation at position 6 (Fig. 3) gave rise to the largest shift down to a pH50 = 4.12 ± 0.07 (n = 8). The shifts in pH50 correspond to two- to eight-fold increases in the concentrations of protons required for activation (Fig. 3 and Table S2). Simultaneous aromatic-to-Ala substitutions of interacting aromatic pairs were also generated, but none of the double mutants gave observable proton-activated currents. The absence of current could reflect impaired channel function and/or folding and then trafficking to the cell surface (Haeger et al., 2010).

Note that the pH50 values derived from macroscopic currents depend on both the affinity of the agonist for its binding site and the equilibrium constant governing transitions from closed to open states. In addition, desensitization kinetics can influence the measurement of pH50 values. Most of the mutations have little effect on the macroscopic desensitization rates (Fig. S3). Given that the proton binding sites for activation are mainly distant from the TMD (Duret et al., 2011), the majority of the changes in pH50 likely reflect changes to the equilibrium constant governing channel gating – the decreased pH50 values thus likely reflect impaired coupling between agonist binding and channel gating. This interpretation, however, is not equivocal, as a His235 located on the adjacent M2 α-helix influences proton activation of GLIC (Rienzo et al., 2014; Wang et al., 2012). In particular, the Y266A could directly influence the pH50 for gating via this intramembrane protonation site.

Enhanced M4-M1/M3 interactions potentiate channel function

In contrast to the mutations in GLIC, individual aliphatic-to-aromatic substitutions introduced at the M4-M1/M3 interface to enhance M4-M1/M3 interactions in ELIC each left-shifted the dose response to cysteamine, whether or not interacting aromatic partners on the adjacent transmembrane α-helices were present (Fig. 3 and Table S2). Wild type ELIC required a concentration of cysteamine to elicit half maximal channel gating of EC50 = 0.94 ± 0.16 mM cysteamine (n = 11). Of the individual aliphatic-to-aromatic mutations, G218F led to the largest reduction in EC50 = 0.44 ± 0.09 mM cysteamine (n = 13). The changes in EC50 values correspond to between 70% and 50% reductions in the concentrations of cysteamine required to elicit half maximal channel gating. Multiple aromatic additions were also introduced, and these led to even further left-shifts in the EC50 values. Engineering either the entire M4 C-terminal aromatic cluster (EC50 = 0.18 ± 0.02 mM cysteamine (n = 9) or the entire aromatic network of GLIC into ELIC (EC50 = 0.15 ± 0.04 mM cysteamine (n = 7)) shifted the EC50 down to a value approaching 10% of the EC50 value of wild type ELIC. In fact, the largest reductions in EC50 were observed with just two interacting aromatic partners engineered into the M4 C-terminal region, one on M3 and the other on M4. Both the I319F/V260Y and G318F/V260Y double mutants gave EC50 values of 0.13 mM cysteamine.

The left-shifts in the dose response observed with aromatic “additions” in ELIC contrast the right-shifts observed with aromatic “deletions” in GLIC, with left-shifts reflecting a gain, as opposed to a loss of channel function. In contrast to GLIC where the proton-sensitive intramembrane His235 complicates the interpretation of pH50 values, the agonist binding site in ELIC is greater than 30Å distant from even the closest mutations at the M4-M1/M3 interface, indicating that the mutations do not directly influence the chemistry of the agonist site and thus agonist affinity. Furthermore, mutations in M4 of the nAChR have been shown to have no effect on agonist affinity (Bouzat et al., 2000; Mitra et al., 2004; Shen et al., 2006). The changes in EC50 detected here therefore reflect enhanced channel function, i.e. enhanced coupling between agonist binding and channel gating. The gain of function mutations show that residues along the M4-M1/M3 interface in wild type ELIC are not optimized to promote M4-M1/M3 interactions that support channel function, and that improving the effectiveness of these interactions promotes coupling between the agonist site and channel gate.

We considered the possibility that aromatic additions to the M4-M1/M3 interface promote more effective interactions with bound lipids to enhance function. Interactions between the F315 aromatic residue in GLIC and lipids are observed in both the pentamer and single-subunit-TMD simulations. The lipid-facing F317 and F312 residues were also mutated to alanine, leading to gain-of-function and loss-of-function phenotypes, respectively (F317A pH50 = 5.41±0.05, n=8, F312A pH50 = 4.50±0.01, n=6). These results show that it is impossible to predict how interactions between aromatic residues and lipids will influence channel function.

Finally, an important feature of our results is the consistency of the entire data set. Every aromatic-to-Ala substitution at the M4-M1/M3 interface in GLIC led to reduced channel function while every aliphatic-to-aromatic substitution at the same interface in ELIC led to enhanced channel function. The latter is particularly compelling given that although optimal aromatic interactions enhance inter-α-helical interactions, the insertion of aromatic side chains at the M4-M1/M3 interface could lead to structural and/or chemical conflicts and thus a loss of channel function. The data highlight the ease with which effective interactions between M4 and M1/M3 in ELIC can be formed to enhance channel function. The consistency of the data suggests that the changes in function are not due to localized changes in structure, which would be expected to have random effects. The molecular dynamics simulations confirm that aromatic residues at the M4-M1/M3 interface enhance M4-M1/M3 interactions. The gains-of-function observed with aliphatic-to-aromatic residue substitutions in ELIC thus result, at least in part, from enhanced M4 interactions with M1/M3.

A lipid-facing CMS mutation potentiates function by enhancing M4-M1/M3 interactions

If M4-M1/M3 interactions in ELIC are intrinsically weaker than in GLIC leading to relatively poor coupling between the agonist site and channel gate, then ELIC may exhibit a greater capacity for potentiation by allosteric modulators that enhance M4-M1/M3 interactions. To test this hypothesis, we focused on a CMS mutation that occurs on the lipid-facing surface of αM4 in the human muscle-type nAChR (C418W). C418W potentiates nAChR channel function roughly 25-fold (EC50 = 9.11 ± 1.45 μM acetylcholine (n = 31) for wild type, EC50 = 0.34 ± 0.08 μM acetylcholine (n = 11) for C418W) by directly altering M4-lipid interactions, although the mutation must ultimately influence M4’s interactions with the remainder of the TMD (Fig. 4 and Table S3).

Figure 4. A CMS Trp mutation potentiates channel function by enhancing M4 binding.

Figure 4

Open in a new tab

A) Side and top views of the TMD of a single subunit of human α1ChR (homology model based on 2BG9). Aromatics at the M4-M1/M3 interface are shown in yellow. The lipid-facing residue αC418 is shown as a solid orange sphere, with the potentiating αC418W mutation superimposed as an orange sphere/stick transparent combo. B) Proposed mechanism of function for the potentiating effect of αC418W via M4 binding to M1/M3. In the absence of M4 C-terminal aromatic contacts, interaction of the bulky αC418W with the lipid bilayer causes M4 to bind more tightly to M1/M3, potentiating activity. In the presence of M4 C-terminal aromatics, M4 is tightly bound, and αC418W has no effect. C–E) Two-electrode data (left) and dose response curves (right) demonstrating the effect of the αC418 (or equivalent) mutation on C) human muscle-type nAChR, D) ELIC (L308W), and E) GLIC (L304W). Error represented as SE. F) Effect of the L308W mutation on aromatic-substituted ELIC mutants, shown as change in EC50 relative to WT ELIC. Error represented as SD, mutant numbers correspond to those in Fig. 2A. See table S3 for complete EC50 values. See also Table S3.

A leucine residue is found in both GLIC and ELIC at the equivalent position. This Leu residue (L304) residue in GLIC was changed to both Cys and Trp, but neither substitution had any effect on channel function (pH50 = 5.03 ± 0.02 (n = 6) and pH50 = 5.06 ± 0.03 (n = 6), respectively, possibly because the extensive aromatic network at the M4-M1/M3 interface already promotes effective M4-M1/M3 interactions (Fig. 4). In contrast, the same L308C and L308W mutations in ELIC both led to gain of function phenotypes, with the magnitude of the L308W gain of function (EC50 = 0.29 ± 0.05 mM cysteamine (n = 10)) approaching 5-fold relative to the wild type ELIC. Significantly, the introduction of interacting aromatic residues to enhance intrinsic M4-M1/M3 interactions reduced the potentiating effects of this CMS mutation. In fact, L308W had no further effect on the gating of ELIC mutants containing either the three M4 C-terminal cluster aromatics or all five aromatic substitutions. L308W thus potentiates ELIC function in a manner similar to that observed with the introduction of aromatic residues at the M4-M1/M3 interface. These results show that 1) altered lipid-protein interactions promote channel function by enhancing M4-M1/M3 interactions, and 2) the intrinsic strength of M4-M1/M3 interactions influences the functional sensitivity of a pLGIC to altered protein-lipid interactions, including mutations at the lipid-protein interface. The CMS mutation was also superimposed onto various aromatic-to-Ala substituted GLIC mutants, but none of these gave proton-activated currents (data not shown).

Propofol inhibits channel function via an M4 independent mechanism

M4-M1/M3 interactions might play a role in the allosteric effects of other TMD modulators, such as the inhibitory drug propofol. Propofol binds to GLIC near the extracellular surface of the TMD in a cavity delineated by the four transmembrane α-helices and capped by the β6–β7 loop - with the most extensive interactions occurring between M1 and M3 (Nury et al., 2011). We tested the possibility that propofol inhibits effective M4-M1/M3 interactions, by investigating the effects of propofol on several of the aromatic-substituted mutants of both GLIC and ELIC. None of the aromatic-to-Ala substitutions at the M4-M1/M3 interface of GLIC or the aliphatic-to-aromatic substitutions at the same interface in ELIC, however, had a major effect on propofol inhibition (Fig. S4). In contrast to our hypothesis, propofol does not inhibit gating by modulating M4-M1/M3 interactions - consistent with both mutational and simulation studies, which suggest that propofol inhibition results from binding closer to the M2 pore lining α-helix (Nury et al., 2011), or even from within the channel pore (LeBard et al., 2012).

Discussion

Although there are likely other sites of action (Althoff et al., 2014; Brannigan et al., 2008; Jones and McNamee, 1988), a role for M4 in lipid-sensing is highlighted by the identification of M4-bound lipids in the crystal structure of GLIC (Bocquet et al., 2009), as well as by mutagenesis data showing that changes in nAChR M4-lipid interactions influence channel function (Bouzat et al., 1998; Lasalde et al., 1996; Lee et al., 1994; Li et al., 1992; Shen et al., 2006; Tamamizu et al., 2000). M4 is also the site of action for neurosteroids (Hosie et al., 2006; Paradiso et al., 2001). In addition, a lipid-facing mutation on M4 in the muscle-type nAChR potentiates channel activity leading to a congenital myasthenic syndrome (Shen et al., 2006). M4, however, is distant from the channel-lining M2 α-helix, as well as key structures that form the primary allosteric path between the agonist site and the channel gate (i.e. the β1–β2 and β6–β7 loops, the M2–M3 linker), thus raising the question of how changes in M4 structure alter channel function.

Our data show that enhanced M4-M1/M3 interactions potentiate, while reduced interactions inhibit pLGIC function. This conclusion is based on four observations: First, molecular dynamics simulations show that aromatic residues at the M4-M1/M3 interface promote strong M4-M1/M3 interactions, with the elimination these aromatic residues leading to increased Cα-Cα carbon atom separations between M4 and M1/M3. Second, aromatic substitutions that promote M4-M1/M3 interactions enhance, while aromatic substitutions that weaken M4-M1/M3 interactions reduce channel function. Third, aromatic substitutions that modulate M4-M1/M3 interactions influence the potentiating effects of a lipid-facing M4 CMS mutation. No potentiation was observed when the CMS mutation was introduced into GLIC, which exhibits intrinsically effective M4-M1/M3 interactions, while strong potentiation was observed with ELIC, which lacks M4-M1/M3 stabilizing aromatic interactions. Significantly, engineering aromatic interactions into the M4-M1/M3 interface in ELIC abrogates the potentiating response. Finally, the strength of M4-M1/M3 interactions has no effect on the inhibitory effects of the drug propofol, which acts at a TMD site that does not directly involve M4 (LeBard et al., 2012; Nury et al., 2011).

Our proposed model of M4 action is supported by biophysical studies, which have shown that the orientation of nAChR M4, and thus presumably the interactions between M4 and M1/M3, is sensitive to its surrounding lipid environment (Antollini et al., 2005; Xu et al., 2005). The nAChR M4 moves halfway along the reaction coordinate between agonist binding and the open state (Mitra et al., 2004). Motion of M4 has also been detected during desensitization of GLIC (Velisetty et al., 2014), consistent with the desensitization effects observed here with some of the M4-M1/M3 interface aromatic residue substitutions (Supplementary Fig. 4 and 5).

There appears to be a particularly important role for the M4 C-terminus in pLGIC function, in agreement with the proposed role of the M4 C-terminus in lipid sensing by the muscle-type nAChR from Torpedo. Increasing levels of cholesterol and anionic lipids stabilize increasing proportions of agonist-responsive nAChRs (Baenziger et al., 2000; daCosta et al., 2009; daCosta et al., 2002; Hamouda et al., 2006). In the absence of these activating lipids, the nAChR adopts an uncoupled conformation that exhibits resting-state-like agonist binding, but does not usually undergo agonist-induced conformational transitions (Baenziger et al., 2008; daCosta and Baenziger, 2009; daCosta et al., 2013). The M4 C-terminus in both the nAChR and GLIC interacts directly with the β6–β7 loop, an important link between the agonist site and the transmembrane gate (Jha et al., 2007; Lee et al., 2009). The M4 C-terminus also interacts with M3 adjacent to the M2–M3 linker, a structure that controls the orientation of the M2 gating α-helix. Tighter M4 interactions with M1/M3 may facilitate interactions between the M4 C-terminus and the β6–β7-loop, to form a β6–β7-loop conformation that participates optimally in channel gating (daCosta and Baenziger, 2009). Interestingly, the M4 C-terminal does not interact directly with the β6–β7 loop in the ELIC crystal structure, which is significant because crystallized ELIC does not undergo channel gating (Gonzalez-Gutierrez et al., 2012). Weak M4 C-terminal interactions with M1/M3, as a consequence of detergent solubilisation, may lead ELIC to adopt an uncoupled conformation (daCosta and Baenziger, 2013).

Finally, a key finding of our study is the demonstration that variable chemistry at the interface between M4 and M1/M3 in different pLGICs leads to variable M4-M1/M3 interactions, different “efficiencies” of coupling binding to gating, and different susceptibilities to potentiation by allosteric modulators, in this case a CMS mutation that acts on M4. GLIC has an extensive aromatic network at this interface that leads to effective M4-M1/M3 interactions along the entire length of M4, rendering the TMD less malleable and less sensitive to M4 targeting modulators. GLIC is insensitive to the potentiating effects of the lipid-facing M4 CMS mutation. GLIC also maintains efficient gating in lipid environments that stabilize an uncoupled nAChR (Labriola et al., 2013). ELIC, with no aromatic interactions in the C-terminal half of M4 exhibits weak M4-M1/M3 interactions in this region. ELIC is more sensitive than GLIC to M4 targeting modulators, such as the CMS mutation and lipids, although aromatic substitutions at the M4-M1/M3 abrogates sensitivity to both (Carswell et al., 2015). The nAChR with relatively few inter-α-helix aromatic interactions, likely exhibits relatively weak M4-M1/M3 interactions along the entire length of M4, and is even more sensitive than ELIC to both the CMS mutation and lipids. Note that although the Torpedo nAChR structure does not exhibit tight interactions between M4 and M1/M3, M4 is not tilted away from M1/M3 as it is in the ELIC structure. The nAChR structure, however, was solved by cryo-electron microscopy using native nAChR membranes (Unwin, 2005; Unwin and Fujiyoshi, 2012), while the ELIC structure was solved by X-ray diffraction using crystals formed from detergent-solubilized ELIC (Hilf and Dutzler, 2008; Pan et al., 2012). In the native Torpedo membranes, there are “activating” lipids (cholesterol, anionic lipids, etc.) that stabilize a functional conformation, where M4 may associate effectively with M1/M3.

The chemistry at the M4-M1/M3 interface varies across human nAChR subunits, suggesting that human nAChRs exhibit variable M4-M1/M3 interactions, and thus possibly different sensitivities to allosteric modulators that act on M4. Knowledge of the subunit-specific roles of M4 in nAChR function may prove to be important for understanding the mechanisms by which cholinergic activity is modulated by changes in lipid composition that occur during the course of neurodegenerative disease.

Experimental Methods

RNA constructs for oocyte expression

GLIC-pMT3 was kindly provided by Dr. Pierre-Jean Corringer (Bocquet et al., 2009). The GLIC coding sequence was transferred to pSP64 without the C-terminal hemagluttinin tag. ELIC-pTLN was kindly provided by Dr. Raimund Dutzler (Zimmermann et al., 2012). A C-terminal Ala, a cloning artifact not present in the Genbank sequence (accession number POC7B7), was removed. Both the GLIC and ELIC plasmids have the α7 nAChR signal sequence followed by the GLIC or ELIC coding sequence. ELIC-pTLN and GLIC-pSP64 were linearized by MluI and EcoRI, respectively and used to produce capped cRNA by in vitro transcription using the mMESSAGE mMACHINE® SP6 kit (Ambion). All mutants were created using QuikChange™ Site Directed Mutagenesis kits (Agilent) and verified by sequencing.

Electrophysiology

Stage V-VI oocytes were isolated as previously described(Laitko et al., 2006). Oocytes were injected with the indicated amount of mRNA and allowed to incubate one to four days at 16°C in ND96+ buffer (5mM HEPES, 96mM NaCl, 2mM KCl, 1mM MgCl2, 1mM CaCl2, and 2mM Pyruvate). Injected oocytes were placed in a RC-1Z oocyte chamber (Harvard Apparatus; Hamden, CT) containing the appropriate buffer (see below). Whole cell currents were recorded using a two-electrode voltage clamp apparatus (OC-725C oocyte clamp; Holliston, MA). The whole cell currents were recorded while flowing the appropriate buffer through the oocyte chamber at a rate of 5 – 10 ml/min.

For GLIC, whole cell currents were recorded from injected oocytes (3 ng – 13 ng cRNA) immersed in MES buffer (140mM NaCl, 2.8mM KCl, 2mM MgCl2, and 10mM MES). Currents through the plasma membrane in response to pH jumps (pH7.3 down to the indicated pH values) were measured with the transmembrane voltage clamped at voltages between −10mV and −60mV depending on the level of expression of each mutant GLIC. In the majority of cases, the holding potential was −20 mV. For ELIC, whole cell currents were recorded from injected oocytes (0.2 ng – 10 ng cRNA) immersed in HEPES buffer (150mM NaCl, 0.5mM BaCl2 and 10mM HEPES, pH7.0). In most cases, currents through the plasma membrane in response to cysteamine concentration jumps (from 0mM up to the indicated values) were measured with the transmembrane voltage clamped at −40mV.

Propofol (2,6 diisopropylphenol) was obtained from Aldrich (D126608). A stock solution was made by diluting liquid propofol to 1M in dimethyl sulphoxide. This solution was stored in glass in the dark, and diluted in MES or HEPES buffer immediately before use. Each oocyte was exposed to at most two different propofol concentrations. To avoid cumulative inhibitory effects, GLIC and ELIC IC50 values were obtained through repeated measurement of relative inhibition caused by one concentration of propofol on single oocyte, multiple times. The average inhibitory values at each concentration were used in calculation of the IC50 using Prism’s log(inhibitor) vs response (three parameter) analysis.

Dose responses for each mutant were acquired from at least two different batches of oocytes. Each individual dose-response experiment was fit with a variable slope sigmoidal dose-response and the individual EC50 and Hill coefficients from each experiment averaged to give the values ± standard deviation. For the presented dose-response curves, the individual dose responses for each experiment were normalized, and then each data point averaged. Curve fits of the averaged data are presented, with the error bars referring to the standard error.

Molecular Dynamics Simulations

Two systems containing GLIC from PDB 4HFI (Sauget etal., 2013) were prepared by protonating residues according to their standard states at pH 4.6, followed by either no mutations (WT-GLIC) or five simultaneous mutations (5Ala-GLIC) corresponding to the sites 1, 2, 3, 5 and 6 investigated in the experiments. Resolved lipids in the PDB structure were not included, although after about 50 ns of simulation, lipids bound to WT-GLIC in poses similar to those in 4HFI. For each system, the intact pentamer was placed in a 110Å × 110Å POPC membrane aligned parallel to the xy plane using CHARMM-GUI Membrane Builder (Jo et al., 2009). The system was solvated with a total height in z of 155 Å and neutralized, for a total of about 175,000 atoms per system. Atomistic molecular dynamics simulations were run with NAMD v2.9 (Phillips et al., 2005). The CHARMM36 forcefield was used for protein (Best et al., 2012; MacKerell et al., 1998) and phospholipid (Klauda et al., 2010) parameters, with parameters for TIP3P waters (Jorgensen et al., 1983) and ions (Beglov and Roux, 1994) corresponding to those traditionally used with CHARMM-based forcefields. All simulations used periodic boundary conditions and particle mesh Ewald (PME) electrostatics. For more details, see Supplementary Information.

Supplementary Material

Acknowledgments

This research was funded by research grant 111243 to JEB from the Canadian Institutes of Health Research (CIHR) and with support from the CIHR Training Program in Neurodegenerative Lipidomics (TGF-96121), and by research grants MCB1330728 and P01GM55876-14A1 from the National Science Foundation and National Institutes of Health Research, respectively, to GB. This project was supported with computational resources from the National Science Foundation XSEDE program through allocation NSF-MCB110149 as well as a local cluster funded by NSF-DBI1126052.

Footnotes

Author contributions: All authors contributed to the design of the research. CLC, CH, JPDT, PFJ, and JAS created and/or characterized the mutants. CH also performed the propofol experiments. Molecular dynamics simulations were performed by SM and GB. JEB, CH, GB, and CLC wrote the paper. CH, JPDT, GB, and CLC created the figures.

Conflict of interest: The authors declare no conflict of interest.

References

  1. Althoff T, Hibbs RE, Banerjee S, Gouaux E. X-ray structures of GluCl in apo states reveal a gating mechanism of Cys-loop receptors. Nature. 2014;512:333–337. doi: 10.1038/nature13669. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Antollini SS, Xu Y, Jiang H, Barrantes FJ. Fluorescence and molecular dynamics studies of the acetylcholine receptor γM4 transmembrane peptide in reconstituted systems. Mol Membr Biol. 2005;22:471–483. doi: 10.1080/09687860500367915. [DOI] [PubMed] [Google Scholar]
  3. Baenziger JE, Henault CM, Therien JP, Sun J. Nicotinic acetylcholine receptor-lipid interactions: Mechanistic insight and biological function. Biochim Biophys Acta. 2015 doi: 10.1016/j.bbamem.2015.03.010. [DOI] [PubMed] [Google Scholar]
  4. Baenziger JE, Morris ML, Darsaut TE, Ryan SE. Effect of membrane lipid composition on the conformational equilibria of the nicotinic acetylcholine receptor. J Biol Chem. 2000;275:777–784. doi: 10.1074/jbc.275.2.777. [DOI] [PubMed] [Google Scholar]
  5. Baenziger JE, Ryan SE, Goodreid MM, Vuong NQ, Sturgeon RM, daCosta CJ. Lipid composition alters drug action at the nicotinic acetylcholine receptor. Mol Pharmacol. 2008;73:880–890. doi: 10.1124/mol.107.039008. [DOI] [PubMed] [Google Scholar]
  6. Barrantes FJ. Modulation of nicotinic acetylcholine receptor function through the outer and middle rings of transmembrane domains. Curr Opin Drug Discov Devel. 2003;6:620–632. [PubMed] [Google Scholar]
  7. Barrantes FJ. Phylogenetic conservation of protein-lipid motifs in pentameric ligand-gated ion channels. Biochim Biophys Acta. 2015 doi: 10.1016/j.bbamem.2015.03.028. [DOI] [PubMed] [Google Scholar]
  8. Beglov D, Roux B. Finite representation of an infinite bulk system: solvent boundary potential for computer simulations. J Chem Phys. 1994;100:9050–9063. [Google Scholar]
  9. Best RB, Zhu X, Shim J, Lopes PE, Mittal J, Feig M, Mackerell AD., Jr Optimization of the additive CHARMM all-atom protein force field targeting improved sampling of the backbone phi, psi and side-chain chi(1) and chi(2) dihedral angles. J Chem Theory Comput. 2012;8:3257–3273. doi: 10.1021/ct300400x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bocquet N, Nury H, Baaden M, Le Poupon C, Changeux JP, Delarue M, Corringer PJ. X-ray structure of a pentameric ligand-gated ion channel in an apparently open conformation. Nature. 2009;457:111–114. doi: 10.1038/nature07462. [DOI] [PubMed] [Google Scholar]
  11. Bouzat C, Barrantes F, Sine S. Nicotinic receptor fourth transmembrane domain: hydrogen bonding by conserved threonine contributes to channel gating kinetics. J Gen Physiol. 2000;115:663–672. doi: 10.1085/jgp.115.5.663. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Bouzat C, Roccamo AM, Garbus I, Barrantes FJ. Mutations at lipid-exposed residues of the acetylcholine receptor affect its gating kinetics. Mol Pharmacol. 1998;54:146–153. doi: 10.1124/mol.54.1.146. [DOI] [PubMed] [Google Scholar]
  13. Brannigan G, Henin J, Law R, Eckenhoff R, Klein ML. Embedded cholesterol in the nicotinic acetylcholine receptor. Proc Natl Acad Sci U S A. 2008;105:14418–14423. doi: 10.1073/pnas.0803029105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Carswell CL, Sun J, Baenziger JE. Intramembrane Aromatic Interactions Influence the Lipid Sensitivities of Pentameric Ligand-gated Ion Channels. J Biol Chem. 2015;290:2496–2507. doi: 10.1074/jbc.M114.624395. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. daCosta CJ, Baenziger JE. A lipid-dependent uncoupled conformation of the acetylcholine receptor. J Biol Chem. 2009;284:17819–17825. doi: 10.1074/jbc.M900030200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. daCosta CJ, Baenziger JE. Gating of pentameric ligand-gated ion channels: structural insights and ambiguities. Structure. 2013;21:1271–1283. doi: 10.1016/j.str.2013.06.019. [DOI] [PubMed] [Google Scholar]
  17. daCosta CJ, Dey L, Therien JP, Baenziger JE. A distinct mechanism for activating uncoupled nicotinic acetylcholine receptors. Nat Chem Biol. 2013;9:701–707. doi: 10.1038/nchembio.1338. [DOI] [PubMed] [Google Scholar]
  18. daCosta CJ, Medaglia SA, Lavigne N, Wang S, Carswell CL, Baenziger JE. Anionic lipids allosterically modulate multiple nicotinic acetylcholine receptor conformational equilibria. J Biol Chem. 2009;284:33841–33849. doi: 10.1074/jbc.M109.048280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. daCosta CJ, Ogrel AA, McCardy EA, Blanton MP, Baenziger JE. Lipid-protein interactions at the nicotinic acetylcholine receptor. A functional coupling between nicotinic receptors and phosphatidic acid-containing lipid bilayers. J Biol Chem. 2002;277:201–208. doi: 10.1074/jbc.M108341200. [DOI] [PubMed] [Google Scholar]
  20. Duret G, Van Renterghem C, Weng Y, Prevost M, Moraga-Cid G, Huon C, Sonner JM, Corringer PJ. Functional prokaryotic-eukaryotic chimera from the pentameric ligand-gated ion channel family. Proc Natl Acad Sci U S A. 2011;108:12143–12148. doi: 10.1073/pnas.1104494108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gonzalez-Gutierrez G, Lukk T, Agarwal V, Papke D, Nair SK, Grosman C. Mutations that stabilize the open state of the Erwinia chrisanthemi ligand-gated ion channel fail to change the conformation of the pore domain in crystals. Proc Natl Acad Sci U S A. 2012;109:6331–6336. doi: 10.1073/pnas.1119268109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Grutter T, de Carvalho LP, Dufresne V, Taly A, Edelstein SJ, Changeux JP. Molecular tuning of fast gating in pentameric ligand-gated ion channels. Proc Natl Acad Sci U S A. 2005;102:18207–18212. doi: 10.1073/pnas.0509024102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Haeger S, Kuzmin D, Detro-Dassen S, Lang N, Kilb M, Tsetlin V, Betz H, Laube B, Schmalzing G. An intramembrane aromatic network determines pentameric assembly of Cys-loop receptors. Nat Struct Mol Biol. 2010;17:90–98. doi: 10.1038/nsmb.1721. [DOI] [PubMed] [Google Scholar]
  24. Hamouda AK, Sanghvi M, Sauls D, Machu TK, Blanton MP. Assessing the lipid requirements of the Torpedo californica nicotinic acetylcholine receptor. Biochemistry. 2006;45:4327–4337. doi: 10.1021/bi052281z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Henault CM, Sun J, Therien JP, daCosta CJ, Carswell CL, Labriola JM, Juranka PJ, Baenziger JE. The role of the M4 lipid-sensor in the folding, trafficking, and allosteric modulation of nicotinic acetylcholine receptors. Neuropharmacology. 2014 doi: 10.1016/j.neuropharm.2014.11.011. [DOI] [PubMed] [Google Scholar]
  26. Hibbs RE, Gouaux E. Principles of activation and permeation in an anion-selective Cys-loop receptor. Nature. 2011;474:54–60. doi: 10.1038/nature10139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Hilf RJ, Dutzler R. X-ray structure of a prokaryotic pentameric ligand-gated ion channel. Nature. 2008;452:375–379. doi: 10.1038/nature06717. [DOI] [PubMed] [Google Scholar]
  28. Hilf RJ, Dutzler R. Structure of a potentially open state of a proton-activated pentameric ligand-gated ion channel. Nature. 2009;457:115–118. doi: 10.1038/nature07461. [DOI] [PubMed] [Google Scholar]
  29. Hosie AM, Wilkins ME, da Silva HM, Smart TG. Endogenous neurosteroids regulate GABAA receptors through two discrete transmembrane sites. Nature. 2006;444:486–489. doi: 10.1038/nature05324. [DOI] [PubMed] [Google Scholar]
  30. Jha A, Cadugan DJ, Purohit P, Auerbach A. Acetylcholine receptor gating at extracellular transmembrane domain interface: the cys-loop and M2–M3 linker. J Gen Physiol. 2007;130:547–558. doi: 10.1085/jgp.200709856. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Jo S, Lim JB, Klauda JB, Im W. CHARMM-GUI Membrane Builder for mixed bilayers and its application to yeast membranes. Biophys J. 2009;97:50–58. doi: 10.1016/j.bpj.2009.04.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Jones OT, McNamee MG. Annular and nonannular binding sites for cholesterol associated with the nicotinic acetylcholine receptor. Biochemistry. 1988;27:2364–2374. doi: 10.1021/bi00407a018. [DOI] [PubMed] [Google Scholar]
  33. Jorgensen WL, Chandrasekhar J, Madura JD, Impey RW, Klein ML. Comparison of simple potential functions for simulating liquid water. J Chem Phys. 1983;79:926–935. [Google Scholar]
  34. Klauda JB, Venable RM, Freites JA, O’Connor JW, Tobias DJ, Mondragon-Ramirez C, Vorobyov I, MacKerell AD, Jr, Pastor RW. Update of the CHARMM all-atom additive force field for lipids: validation on six lipid types. J Phys Chem B. 2010;114:7830–7843. doi: 10.1021/jp101759q. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Labriola JM, Pandhare A, Jansen M, Blanton MP, Corringer PJ, Baenziger JE. Structural sensitivity of a prokaryotic pentameric ligand-gated ion channel to its membrane environment. J Biol Chem. 2013;288:11294–11303. doi: 10.1074/jbc.M113.458133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Laitko U, Juranka PF, Morris CE. Membrane stretch slows the concerted step prior to opening in a Kv channel. J Gen Physiol. 2006;127:687–701. doi: 10.1085/jgp.200509394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Lasalde JA, Tamamizu S, Butler DH, Vibat CR, Hung B, McNamee MG. Tryptophan substitutions at the lipid-exposed transmembrane segment M4 of Torpedo californica acetylcholine receptor govern channel gating. Biochemistry. 1996;35:14139–14148. doi: 10.1021/bi961583l. [DOI] [PubMed] [Google Scholar]
  38. LeBard DN, Henin J, Eckenhoff RG, Klein ML, Brannigan G. General anesthetics predicted to block the GLIC pore with micromolar affinity. PLoS Comput Biol. 2012;8:e1002532. doi: 10.1371/journal.pcbi.1002532. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Lee WY, Free CR, Sine SM. Binding to gating transduction in nicotinic receptors: Cys-loop energetically couples to pre-M1 and M2–M3 regions. J Neurosci. 2009;29:3189–3199. doi: 10.1523/JNEUROSCI.6185-08.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Lee WY, Sine SM. Principal pathway coupling agonist binding to channel gating in nicotinic receptors. Nature. 2005;438:243–247. doi: 10.1038/nature04156. [DOI] [PubMed] [Google Scholar]
  41. Lee YH, Li L, Lasalde J, Rojas L, McNamee M, Ortiz-Miranda SI, Pappone P. Mutations in the M4 domain of Torpedo californica acetylcholine receptor dramatically alter ion channel function. Biophys J. 1994;66:646–653. doi: 10.1016/s0006-3495(94)80838-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Li L, Lee YH, Pappone P, Palma A, McNamee MG. Site-specific mutations of nicotinic acetylcholine receptor at the lipid-protein interface dramatically alter ion channel gating. Biophys J. 1992;62:61–63. doi: 10.1016/S0006-3495(92)81779-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Lummis SC, Beene DL, Lee LW, Lester HA, Broadhurst RW, Dougherty DA. Cis-trans isomerization at a proline opens the pore of a neurotransmitter-gated ion channel. Nature. 2005;438:248–252. doi: 10.1038/nature04130. [DOI] [PubMed] [Google Scholar]
  44. MacKerell AD, Bashford D, Bellott M, Dunbrack RL, Evanseck JD, Field MJ, Fischer S, Gao J, Guo H, Ha S, et al. All-atom empirical potential for molecular modeling and dynamics studies of proteins. J Phys Chem B. 1998;102:3586–3616. doi: 10.1021/jp973084f. [DOI] [PubMed] [Google Scholar]
  45. Miller PS, Aricescu AR. Crystal structure of a human GABA receptor. Nature. 2014 doi: 10.1038/nature13293. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Mitra A, Bailey TD, Auerbach AL. Structural dynamics of the M4 transmembrane segment during acetylcholine receptor gating. Structure. 2004;12:1909–1918. doi: 10.1016/j.str.2004.08.004. [DOI] [PubMed] [Google Scholar]
  47. Nury H, Van Renterghem C, Weng Y, Tran A, Baaden M, Dufresne V, Changeux JP, Sonner JM, Delarue M, Corringer PJ. X-ray structures of general anaesthetics bound to a pentameric ligand-gated ion channel. Nature. 2011;469:428–431. doi: 10.1038/nature09647. [DOI] [PubMed] [Google Scholar]
  48. Pan J, Chen Q, Willenbring D, Yoshida K, Tillman T, Kashlan OB, Cohen A, Kong XP, Xu Y, Tang P. Structure of the pentameric ligand-gated ion channel ELIC cocrystallized with its competitive antagonist acetylcholine. Nat Commun. 2012;3:714. doi: 10.1038/ncomms1703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Paradiso K, Zhang J, Steinbach JH. The C terminus of the human nicotinic alpha4beta2 receptor forms a binding site required for potentiation by an estrogenic steroid. J Neurosci. 2001;21:6561–6568. doi: 10.1523/JNEUROSCI.21-17-06561.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Phillips JC, Braun R, Wang W, Gumbart J, Tajkhorshid E, Villa E, Chipot C, Skeel RD, Kale L, Schulten K. Scalable molecular dynamics with NAMD. J Comput Chem. 2005;26:1781–1802. doi: 10.1002/jcc.20289. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Rienzo M, Lummis SC, Dougherty DA. Structural requirements in the transmembrane domain of GLIC revealed by incorporation of noncanonical histidine analogs. Chemistry & biology. 2014;21:1700–1706. doi: 10.1016/j.chembiol.2014.10.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Sauguet L, Poitevin F, Murail S, Van Renterghem C, Moraga-Cid G, Malherbe L, Thompson AW, Koehl P, Corringer PJ, Baaden M, et al. Structural basis for ion permeation mechanism in pentameric ligand-gated ion channels. EMBO J. 2013;32:728–741. doi: 10.1038/emboj.2013.17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Sauguet L, Shahsavar A, Poitevin F, Huon C, Menny A, Nemecz A, Haouz A, Changeux JP, Corringer PJ, Delarue M. Crystal structures of a pentameric ligand-gated ion channel provide a mechanism for activation. Proc Natl Acad Sci U S A. 2014;111:966–971. doi: 10.1073/pnas.1314997111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Shen XM, Deymeer F, Sine SM, Engel AG. Slow-channel mutation in acetylcholine receptor alphaM4 domain and its efficient knockdown. Ann Neurol. 2006;60:128–136. doi: 10.1002/ana.20861. [DOI] [PubMed] [Google Scholar]
  55. Tamamizu S, Guzman GR, Santiago J, Rojas LV, McNamee MG, Lasalde-Dominicci JA. Functional effects of periodic tryptophan substitutions in the alpha M4 transmembrane domain of the Torpedo californica nicotinic acetylcholine receptor. Biochemistry. 2000;39:4666–4673. doi: 10.1021/bi992835w. [DOI] [PubMed] [Google Scholar]
  56. Unwin N. Refined structure of the nicotinic acetylcholine receptor at 4Å resolution. J Mol Biol. 2005;346:967–989. doi: 10.1016/j.jmb.2004.12.031. [DOI] [PubMed] [Google Scholar]
  57. Unwin N, Fujiyoshi Y. Gating movement of acetylcholine receptor caught by plunge-freezing. J Mol Biol. 2012;422:617–634. doi: 10.1016/j.jmb.2012.07.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Velisetty P, Chalamalasetti SV, Chakrapani S. Structural basis for allosteric coupling at the membrane-protein interface in Gloeobacter violaceus ligand-gated ion channel (GLIC) J Biol Chem. 2014;289:3013–3025. doi: 10.1074/jbc.M113.523050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Wang HL, Cheng X, Sine SM. Intramembrane proton binding site linked to activation of bacterial pentameric ion channel. J Biol Chem. 2012;287:6482–6489. doi: 10.1074/jbc.M111.305839. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Xu Y, Barrantes FJ, Luo X, Chen K, Shen J, Jiang H. Conformational dynamics of the nicotinic acetylcholine receptor channel: a 35-ns molecular dynamics simulation study. J Am Chem Soc. 2005;127:1291–1299. doi: 10.1021/ja044577i. [DOI] [PubMed] [Google Scholar]
  61. Zimmermann I, Marabelli A, Bertozzi C, Sivilotti LG, Dutzler R. Inhibition of the prokaryotic pentameric ligand-gated ion channel ELIC by divalent cations. PLoS Biol. 2012;10:e1001429. doi: 10.1371/journal.pbio.1001429. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

NIHMS773868-supplement-supplement_1.pdf (1.4MB, pdf)

RESOURCES