Heterologous expression and nonsense suppression provide insights into agonist behavior at α6β2 nicotinic acetylcholine receptors

Highlights

• α6L9′Sβ2L9′S_LFM/AAQA nAChR yields high expression levels in Xenopus oocytes.

• Expression levels are high enough to permit nonsense suppression studies.

• Structure function studies show ACh makes a cation-π interaction with TrpB.

• Nicotine, though more potent than ACh, makes no cation-π interaction with TrpB.

Abstract

The α6-containing subtypes of the nicotinic acetylcholine receptor (nAChR) are localized to presynaptic terminals of the dopaminergic pathways of the central nervous system. Selective ligands for these nAChRs are potentially useful in both Parkinson's disease and addiction. For these and other goals, it is important to distinguish the binding behavior of agonists at the α6-β2 binding site versus other subtypes. To study this problem, we apply nonsense suppression-based non-canonical amino acid mutagenesis. We report a combination of four mutations in α6β2 that yield high-level heterologous expression in Xenopus oocytes. By varying mRNA injection ratios, two populations were observed with unique characteristics, likely due to differing stoichiometries. Responses to nine known nAChR agonists were analyzed at the receptor, and their corresponding EC₅₀ values and efficacies are reported. The system is compatible with nonsense suppression, allowing structure–function studies between Trp149 – a conserved residue on loop B found to make a cation-π interaction at several nAChR subtypes – and several agonists. These studies reveal that acetylcholine forms a strong cation-π interaction with the conserved tryptophan, while nicotine and TC299423 do not, suggesting a unique pharmacology for the α6β2 nAChR.

Introduction

The nicotinic acetylcholine receptors (nAChR) are a type of pentameric ligand gated ion channel activated by the neurotransmitter acetylcholine, as well as nicotine and a wide array of related small molecules (Smart and Paoletti, 2012). In addition to its role at the neuromuscular junction, the neuronal nAChRs are widely distributed throughout the CNS. To date, twelve different subunits of the neuronal nAChR have been identified, α2–α10, and β2–β4 (Le Novère et al., 2002). These subunits assemble in various patterns to form different subtypes with distinct localizations, pharmacological characteristics, and functions in the nervous system (Gotti et al., 2006, Zoli et al., 2014). The most commonly and widely expressed neuronal nAChRs in the brain are the α4β2 and α7 subtypes, and these have been studied in depth (Holladay et al., 1997).

The α6 subunit, which is primarily localized to the ventral tegmental area and substantia nigra pars compacta, is thought to form α6β2 pentamers as well as complex subtypes with three or more different subunits, such as α6α4β2, α6β2β3, and α6α4β2β3 (Gotti et al., 2006, Grady et al., 2010, Gerzanich et al., 1997). In some regions, such as the locus coerulus, α6β4 nAChRs also form (Azam et al., 2010). Subtypes of nAChRs containing the α6 subunit have been of recent interest, as they are found in dopaminergic presynaptic terminals and thus influence the release of dopamine in both the nigrostriatal and mesocorticolimbic pathways (Holladay et al., 1997, Quik and Wonnacott, 2011, Quik and McIntosh, 2006, Yang et al., 2009). As such, finding agonists that are selective at these subtypes, specifically the α6-β2 binding site, could be important in studies of both Parkinson's disease and addiction.

A number of structural features are well established for nAChRs. Each subunit has an N-terminal extracellular ligand-binding domain followed by four transmembrane helices, M1–M4 (Miyazawa et al., 2003). Of note are the M2 helix, which lines the channel pore, (Jha et al., 2009) and the intracellular M3–M4 loop, which is thought to be involved in the trafficking of the receptor from the endoplasmic reticulum (ER) to the membrane surface (Kracun et al., 2008). At the interface of two adjacent subunits in the extracellular domain is the ligand binding site, comprised of six loops. Loops A–C are contributed by the primary (α) subunit and D–F by the complementary (β) subunit (Corringer et al., 2000). These loops contribute five conserved residues – TyrA (α6:Y93), TrpB (α6:W149), TyrC1 (α6:Y190), TyrC2 (α6:Y197), and TrpD (β2:W57) – that form an aromatic box responsible for binding the cationic moiety of agonists and antagonists. Previous studies have shown that TrpB in the α4-β2 interface and TyrC2 in the α7–α7 interface make a cation-π interaction with acetylcholine (Van Arnam and Dougherty, 2014). These results contributed to a pharmacophore model of the α4β2 and α7 subtypes and advanced our understanding of the differences in pharmacology among nAChR subtypes.

Various derivatives of α-conotoxins, disulfide-rich peptide antagonists of nAChRs, provide selective antagonism among α6-containing nAChRs (Azam et al., 2010, Hone et al., 2013, Hone et al., 2012). These selective antagonists have provided rich information about the roles of α6-containing subtypes in physiology and behavior. It is thought that additional information can be gained, and perhaps useful drugs found, among selective agonists. However developing agonists selective for α6-containing subtypes requires a deeper understanding of the ligand site, specifically how the α6-β2 binding site differs from those previously studied.

High precision studies of agonist binding in the α4β2 and α7 receptors have utilized nonsense suppression-based non-canonical amino acid mutagenesis in a Xenopus laevis oocyte expression system (Dougherty and Van Arnam, 2014). Nonsense suppression is, however, relatively inefficient, with agonist-induced currents roughly an order of magnitude lower than produced by conventional mutagenesis, making previously reported α6-expression systems such as chimeric subunits, and concatenated subunits unsuitable for this technique (Yang et al., 2009, Kuryatov et al., 2000, Letchworth and Whiteaker, 2011, Wang et al., 2014, Papke et al., 2008, Capelli et al., 2011, Kuryatov and Lindstrom, 2011, Ley et al., 2014). Here, we report a combination of four mutations that result in the controlled and consistent expression of α6β2 at the high levels that permit nonsense suppression and thus incorporation of non-canonical amino acids. Results from such experiments allow preliminary development of a binding model for agonists at α6β2-containing nAChRs.