Acetylcholine-receptor channels (AChRs) are allosteric membrane proteins that mediate synaptic transmission by alternatively opening and closing (‘gating’) a cation-selective transmembrane pore. Although ligand binding is not required for the channel to open, the binding of agonists (for example, acetylcholine) increases the closed ⇌ open equilibrium constant because the ion-impermeable → ion-permeable transition of the ion pathway is accompanied by a low → high affinity change at the agonist-binding sites. The fact that the gating conformational change of muscle AChRs can be kinetically modeled as a two-state reaction has paved the way to the experimental characterization of the corresponding transition state, which represents a snapshot of the continuous sequence of molecular events separating the closed and open states. Previous studies of fully (di-) liganded AChRs, combining single-channel kinetic measurements, site-directed mutagenesis, and data analysis in the framework of the linear free-energy relationships of physical organic chemistry, have suggested a transition-state structure that is consistent with channel opening being an asynchronous conformational change that starts at the extracellular agonist-binding sites and propagates towards the intracellular end of the pore. In this paper, I characterize the gating transition state of unliganded AChRs, and report a remarkable difference: unlike that of diliganded gating, the unliganded transition state is not a hybrid of the closed- and open-state structures but, rather, is almost indistinguishable from the open state itself. This displacement of the transition state along the reaction coordinate obscures the mechanism underlying the unliganded closed ⇌ open reaction but brings to light the malleable nature of free-energy landscapes of ion-channel gating.
The muscle acetylcholine receptor channel (AChR)1 is the neurotransmitter-gated ion channel that mediates neuromuscular synaptic transmission in vertebrates (1). Although the structure of this large pentameric transmembrane protein (∼470 residues per subunit) is not known with atomic resolution, a wealth of structural information exists, mainly from mutational studies, affinity labeling, chemical modification of specific residues, electron microscopy, and crystallography (reviewed in ref. 2). As is the case of any other allosteric protein, the dynamic behavior of this receptor-channel can be understood in the framework of thermodynamic cycles, with conformational changes and ligand-binding events as the elementary steps (3-5). Thus, the AChR can adopt a variety of different conformations that can interconvert (closed, open, and desensitized ‘states’), and each conformation has a distinct ligand-binding affinity (low affinity in the closed state and high affinity in the open and desensitized states) and a particular ‘catalytic efficiency’ (ion-impermeable in the closed and desensitized states, and ion-permeable in the open state). To meet the physiological requirement of a small closed ⇌ open (‘gating’) equilibrium constant for the unliganded receptor, and a large gating equilibrium constant for the ACh-diliganded receptor, the affinity of the AChR for ACh must be higher in the open than in the closed conformation (4-6). This follows from the notion that the equilibrium constants governing the different reaction steps (ligand binding and gating) of these cyclic reaction schemes are constrained by the principle of detailed balance.
Hence, irrespective of whether the receptor is diliganded, monoliganded or unliganded, two changes must take place in going from the closed state (low ligand affinity and ion-impermeable) to the open state (high ligand affinity and ion-permeable): a) the pore becomes permeable to ions, and b) the transmitter-binding sites, some 50 Å away from the pore domain (7), increase their affinity for the ligand (with the reverse changes taking place during closing). The apparent lack of stable intermediates between the closed and open conformations, inferred from kinetic modeling of the diliganded-gating reaction (8), suggests that these two changes occur as a result of a one-step, global conformational change. The question, then, arises as to whether this concerted conformational change proceeds synchronously (i.e., every residue of the protein moves ‘in unison’) or asynchronously (i.e., following a sequence of events; ref. 9) and, if the latter were the case, whether multiple, few, or just one sequence of events is actually traversed by the channel to ‘connect’ the end states.
Analysis of the correlation between rate and equilibrium constants of gating in diliganded AChRs has allowed us to address some of these issues by probing the structure of the transition state (8, 10-12), that is, the intermediate species between the end states of a one-step reaction that can be most easily studied. Interpretation of these results in the framework of the classical rate-equilibrium free-energy relationships of physical organic chemistry (13, 14), revealed that AChR diliganded gating is a highly asynchronous reaction, and suggested that the transition-state ensemble is quite homogeneous, as if the crossing of the energy barrier were confined to a narrow pass at the top of the energy landscape. In the opening direction, the conformational rearrangement that leads to the low-to-high affinity change at the extracellular binding sites precedes the conformational rearrangement of the pore that renders the channel ion-permeable. This propagated global conformational change, which we have referred to as a ‘conformational wave’ (11), must reverse during channel closing so that closing starts at the pore and propagates all the way to the binding sites.
It is not at all obvious why the diliganded-gating conformational change starts at the binding sites when the channel opens, nor even why the conformational change propagates at all through the receptor, instead of taking place synchronously throughout the protein. Is there any correlation between the location of the domain that binds agonist and the location of the initiation site for the opening conformational change? Could the latter have started from the intracellular end of the pore, for example, and have propagated to the (extracellular) transmitter-binding sites? What difference does it make to be liganded or unliganded as far as the mechanism of the gating conformational change is concerned? To address these issues, I set out to explore the mechanism of gating in unliganded AChRs by probing the structure of the corresponding transition state using kinetic measurements, site-directed mutagenesis, and the concepts of rate-equilibrium free-energy relationships and Φ-value analysis.
Briefly, a Φ-value can be assigned to any position in the protein by estimating the slope of a ‘Brönsted plot’2 [log (gating rate constant) versus log (gating equilibrium constant)] where each point corresponds to a different amino-acid substitution at that given position. More coarsegrained Φ-values can also be obtained by using different agonists or different transmembrane potentials, for example, as a means of altering the rate and equilibrium constants of gating. Very often, rate-equilibrium plots are linear, and 0 < Φ < 1. A value of Φ = 0 suggests that the position in question (in the case of a mutation series) experiences a closed-state-like environment at the transition state whereas a value of Φ = 1 suggests an open-state-like environment. A fractional Φ-value suggests an environment that is intermediate between those experienced in the closed and open states (16).
Earlier results indicated that the Φ-values obtained by varying the transmembrane potential are different in diliganded and unliganded AChRs. These Φ-values, which are a measure of the closed-state-like versus open-state-like character of the channel’s voltage-sensing elements at the transition state, are 0.070 ± 0.060 in diliganded receptors (17), and 1.025 ± 0.053 in unliganded AChRs (11, 18). The present study reveals that residues at the transmitter-binding sites (Figure 1), the extracellular loop that links the second (M2) and third (M3) transmembrane segments (M2-M3 linker), and the upper and lower half of M2, which during diliganded gating have Φ-values of ∼1 (ref. 11), ∼0.7 (ref. 10), ∼0.35 (refs 8, 11, 12), and ∼0 (ref. 12), respectively, have also Φ-values very close to 1 during unliganded gating. This generalized shift in Φ-values suggests that the diliganded → unliganded perturbation deforms the energy landscape of gating in such a way that the ‘new’ transition state occurs very close to the open state, to such an extent that all tested positions experience an open-state-like environment at the transition state of unliganded gating. Thus, the transition state occurs so ‘late’ (i.e., so close to the open state) that its inferred structure does not provide any clues as to the intermediate stages of this reaction.
Hence, the mechanism of unliganded gating remains obscure. The change in the position of the transition state along a reaction coordinate, as a result of perturbations to the energy landscape, is a very well known phenomenon in organic chemistry (e.g., refs 20-26), and protein folding (e.g., refs 27-34). In this paper, I show that this phenomenon can also take place in the case of allosteric transitions and, therefore, that the structure of the transition state of a global conformational change need not be fixed; rather, it can change depending on the experimental conditions.