Modal affinities of endplate acetylcholine receptors caused by loop C mutations

Abstract

The time course of the endplate current is determined by the rate and equilibrium constants for acetylcholine receptor (AChR) activation. We measured these constants in single-channel currents from AChRs with mutations at the neurotransmitter-binding sites, in loop C. The main findings are: (a) Almost all perturbations of loop C generate heterogeneity in the channel open probability ("modes"). (b) Modes are generated by different affinities for ACh that can be either higher or lower than in the wild-type receptors. (c) The modes are stable, in so far as each receptor maintains its affinity for at least several minutes. (d) Different agonists show different degrees of modal activity. With the loop C mutation αP197A, there are four modes with ACh but only two with partial agonists. (e) The affinity variations arise exclusively from the αδ-binding site. (f) Substituting four γ-subunit residues into the δ subunit (three in loop E and one in the β5-β5' linker) reduces modal activity. (g) At each neurotransmitter-binding site, affinity is determined by a core of five aromatic residues. Modes are eliminated by an alanine mutation at δW57 but not at the other aromatics. (h) Modes are eliminated by a phenylalanine substitution at all core aromatics except αY93. The results suggest that, at the αδ agonist site, loop C and the complementary subunit surface can each adopt alternative conformations and interact with each other to influence the position of δW57 with respect to the aromatic core and, hence, affinity.

Figure 1.

The ligand-binding site of an acetylcholine binding protein. The agonist-binding sites are at subunit interfaces; the principal side (α subunit in AChRs) is white, and the complementary side (δ, ε, or γ subunit) is tan. The structure is Lymnaea stagnalis (Protein Data Bank accession no. 3WIP; Olsen et al., 2014), and residue numbers are mouse endplate AChRs. Green, aromatic core; tan spheres, αC atoms of γ-subunit substitutions (see Fig. 6); red sphere, structural water; dashed lines, H bonds.

Figure 2.

Loop C proline mutations generate P_O modes. WT AChR clusters are homogenous. (A; left) Single-channel currents (30 µM ACh; −100 mV; open is down). Clusters of openings are binding-gating events (C-O) and long silent intervals between clusters are desensitization (D). (Right) Histograms of cluster P_O (fitted by a single Gaussian) and intra-cluster interval durations (fitted by a single exponential). (Inset) An example cluster. In both adult and fetal AChRs, there is only a single P_O population (0.70 ± 0.06, 187, and 0.93 ± 0.03, 156; mean ± SD; n clusters). (B) Loop C mutations αP197A and αP194A in adult AChRs induce modes. P_O histograms for αP197A have four populations (0.01 ± 0.01, 83; 0.20 ± 0.04, 6; 0.67 ± 0.06, 8; and 0.96 ± 0.01, 110), and multiple exponentials are required to describe the interval duration distributions.

Figure 3.

Gating of loop C proline mutants in the absence of agonists is homogeneous. (A; left) Single-channel clusters from unliganded AChRs (adult type, −100 mV; background mutation were added to increase constitutive activity; see Materials and methods). (Right) Intra-cluster interval duration histograms and an example cluster. WT, αP194A, and αP197A show both brief and long (arrow) unliganded openings. The briefer, main component of each distribution represents C-O gating; the longer open component(s) is characteristic of unliganded activity but is distinct from C-O and of unknown origin (Purohit and Auerbach, 2009). (B) Adding the agonist site mutation αY93F eliminates long unliganded openings but does not affect the gating open interval time constant (Purohit and Auerbach, 2010).

Figure 4.

Partial agonists generate fewer modes than does ACh. (A; left) Example clusters for αP197A adult AChRs ([agonist] = 1 mM; −100 mV; open is down). (Right) Cluster P_O histogram for TMA with peaks at 0.04 ± 0.01 and 0.87 ± 0.10. Arrow, P_O for TMA in WT AChRs (0.30 ± 0.06). The full agonist ACh has four P_O populations (Fig. 2 B, bottom). (B) αP197A adult AChRs activated by a saturating [Cho] (100 mM; 70 mV; open is up). (Top) Two modes are apparent. (Bottom left) Interval duration histograms and example clusters of each mode. (C) Phi (slope) analysis of αP197A choline modes. The fold-change in the di-liganded gating equilibrium constant is caused mainly by a similar fold-change in the channel-opening rate constant.

Figure 5.

Modes are stable. (A; left) Low time-resolution view of αP197A (+αY93A) clusters from a patch having a single AChR. (Right) Corresponding histograms (adult AChR, 30 µM ACh; −100 mV; open is down). For ∼30 min, P_O was homogeneous. (Right, bottom) The slowest shut component (arrow; 162 ± 70 s), which represents recovery from deep desensitization, is similar to the WT value (270 s; Elenes and Auerbach, 2002), indicating that the patch probably had only one AChR. (B) A multichannel patch of the same construct shows P_O modes.

Figure 6.

Modes are generated only at the αδ agonist site. (A) With only a functional αδ site there, are four P_O^ACh and two P_O^TMA populations, as in the WT. (B) Modes are absent in AChRs with only a functional αε or αγ site. (C) Substituting side chains from the γ subunit into the δ subunit (three in loop E and one in the β5–β5′ linker; see Fig. 1) reduces modal activity. Open is down for ACh (−100 mV; 100 µM for αδ and αε, and 30 µM for αγ) and up for TMA (70 mV; 5 mM).

Figure 7.

Example currents for α-subunit loop C perturbations. (A) Modes with different substitutions at αP197. For each mutation, examples of the various modes are shown directly below the mutant designation. Only two modes are apparent with larger side chains (see Fig. 2 for Ala mutation). (B) Alanine scan of loop C. Modes are apparent with an alanine substitution at every loop C position except αY190 and αY198 (boxed). (C) Deletions. Modes are present with deletions of some loop C residues. (D) Modes are present with αT196P added to the αP197A background. All experiments: adult AChRs, 30 µM ACh; −100 mV; open is down.

Figure 8.

Alanine scan of the aromatic core. (A) δW57A. (Top) Example currents from a single-site, αδ-only AChR (3 mM ACh; 70 mV; open is up). (Bottom) Cluster P_O and interval duration histograms. Only one P_O population is apparent (K_d^ACh given in Table 1). (B) Example currents from other alanine mutants. Multiple modes are apparent (two-site, adult AChRs; 30 µM ACh; −100 mV; open is down). αP197A was present for all.

Figure 9.

Phenylalanine scan of the aromatic core. (A) All F substitutions eliminate αP197A modes except for αY93F (two-site, adult AChRs; 30 µM ACh; −100 mV; open is down). Calibration, 1 s/7 pA (low resolution traces) and 250 ms, 5 pA (high resolution clusters). (B) F substitutions at one-site, αδ-only AChRs (3 mM ACh; 70 mV; open is up). Clusters from αY190F, αY198F, and δW57F are homogeneous.