Timothy Lynagh

Ionotropic glutamate receptors (iGluRs) mediate rapid signals between neurons by binding extracellular neurotransmitters and conducting ionic current across the cell membrane. Animal iGluRs can be divided into clades of phylogenetically related receptors, only some of which are present in mammals. Despite their name, many receptors are selectively activated by ligands other than glutamate, and despite high-resolution structural data on most mammalian iGluRs, the molecular basis for ligand selectivity in most clades is not clear. We questioned whether closely related iGluRs within phylogenetic clades share similar ligand selectivity profiles by testing several receptors from bilaterians, cnidarians, a placozoan, and a ctenophore, and by probing the molecular basis of ligand selectivity in iGluRs from different clades with site-directed mutagenesis. We identified glutamate-, glycine-, and GABA-gated channels from both the AKDF clade and the Epsilon clade, and a selectively aspartate-gated channel from the Epsilon clade. We show that in some cases, even very closely related channels from within the same clade and from the same animal species have diverged functionally, leading to channels gated by different agonists. Although we find special roles in ligand selectivity for glycine, threonine, and leucine residues in the “αF-GST” and “β10/αI-LE” motifs (mammalian AMPA receptor nomenclature), our results in diverse iGluRs warn that the effects of mutations, and thus the roles of these residues, are not always easy to predict.

Most ionotropic neurotransmitter receptors are hetero-oligomers of three to five homologous subunits assembled into membrane-spanning ion channels. Mammalian AMPA-type glutamate receptors (AMPARs), which emerged in animals with centralized nervous systems and now mediate most of the excitatory synaptic signaling in our brain, readily assemble as homotetramers, yet often occur as heterotetramers of two to three different AMPAR subunits. Here, we looked for unifying functional properties of the AMPAR family using molecular phylogenetics, together with electrophysiological and pharmacological characterization of heterologously expressed AMPARs from the three major lineages within bilaterian animals. Our results suggest that the hallmark of AMPAR evolution is selective activation by glutamate, fast kinetics, and the duplication of AMPAR genes independently in new animal lineages, such that most major bilaterian groups have relatively unique complements of AMPAR genes. Moreover, in diverse bilaterians, these novel AMPAR subunits have (1) come to rely on each other for functional expression, resulting in obligate heterotetrameric AMPARs, and (2) sub-functionalized, with various subunits contributing differently to channel activation or ion permeation. Remarkably, this evolution appears to have independently converged on a complement of calcium-sensitive and calcium-insensitive AMPARs in mammals and worms via biophysically similar but genetically different mechanisms.

Voltage-gated potassium channel KCNQ1 (Kv7.1) plays a critical role in electrical excitability in the heart, gut, and brain. Together with the auxiliary subunit KCNE1, KCNQ1 generates a slow delayed rectifier current (I_Ks) that is essential for cardiac repolarization. Mutations and dysregulation of this channel are found in channelopathies leading to sudden death, long-QT syndrome, atrial fibrillation, epilepsy, deafness, diabetes, and neuropsychiatric disorders. Although KCNQ1 and related potassium channels are promising therapeutic targets, there are few potent, selective, and therapeutically safe inhibitors and activators available for these proteins. A virtual screening of 36,374 compounds was conducted against KCNQ1, followed by in silico analyses that identified eight potential ligand candidates for experimental evaluation using human KCNQ1 coexpressed with KCNE1 in Xenopus laevis oocytes. Electrophysiological recordings showed that the benzodiazepine-based ligand Zinc13732787 was a potent inhibitor of the channel complex, without affecting KCNQ2/KCNQ3. Based on virtual screening and molecular docking, the 1-(3-chlorophenyl)-urea substituent on the benzodiazepine core is important for selective inhibition of KCNQ1/KCNE1, as further supported by structure-activity relationship and stereochemical exploration of Zinc13732787. Furthermore, low concentrations of Zinc13732787 reduced neurite outgrowth in human neuronal stem cells (NSCs), mirroring the phenotype observed in homozygous KCNQ1-knockout cells. Importantly, Zinc13732787 did not affect NSC proliferation, nor did it induce cytotoxicity. In homozygous KCNQ1-knockout NSCs, compound Zinc13732787 had no effect on neurite outgrowth, indicating high target specificity. These findings suggest that this compound is a valuable tool for investigating the physiological and pathological roles of KCNQ1 across various tissues. Additionally, it could be used as a precursor for novel antiarrhythmic agents as well as for epilepsy and neuropsychiatric conditions.

Diarylamidines are a group of widely used small molecule drugs. One common use of diarylamidines is their pharmacological inhibition of ligand-gated cation channels, including tetrameric ionotropic glutamate receptors and trimeric degenerin/epithelial sodium channel channel/acid-sensing ion channels (DEG/ENaC/ASICs). Here, we discover a DEG/ENaC/ASIC channel from the brachiopod (lamp shell) Novocrania anomala , at which diarylamidines act as agonists. The channel is closely related to bile acid-gated, pH-gated, and peptide-gated channels but is not activated by such stimuli. We describe activation of the channel by diminazene, DAPI, and pentamidine, examine several biophysical and pharmacological properties, and briefly explore the molecular determinants of channel activity with site-directed mutagenesis. We term this channel the diarylamidine-activated sodium channel (DiaaNaC).

Rapid signaling between neurons is mediated by ligand-gated ion channels, cell-surface proteins with an extracellular ligand-binding domain and a membrane-spanning ion channel domain. The degenerin/epithelial sodium channel (DEG/ENaC) superfamily is diverse in terms of its gating stimuli, with some DEG/ENaCs gated by neuropeptides, and others gated by pH, mechanical force or enzymatic activity. The mechanism by which ligands bind to and activate DEG/ENaCs is poorly understood. Here we dissected the structural basis for neuropeptide-gated activity of a neuropeptide-gated DEG/ENaC, FMRFamide-gated sodium channel 1 (FaNaC1) from the annelid worm Malacoceros fuliginosus, using cryo-electron microscopy. Structures of FaNaC1 in the ligand-free resting state and in several ligand-bound states reveal the ligand-binding site and capture the ligand-induced conformational changes of channel gating, which we verified with complementary mutagenesis experiments. Our results illuminate channel gating in DEG/ENaCs and offer a structural template for experimental dissection of channel pharmacology and ion conduction.

Ionotropic glutamate receptors (iGluRs) mediate excitatory signals between cells by binding neurotransmitters and conducting cations across the cell membrane. In the mammalian brain, most of these signals are mediated by two types of iGluRs: AMPA and NMDA (i.e. iGluRs sensitive to 2-amino-3-(5-methyl-3-oxo-1,2-oxazol-4-yl)propanoic acid and N-methyl-D-aspartic acid, respectively). Delta-type iGluRs of mammals also form neurotransmitter-binding channels in the cell membrane, but in contrast, their channel is not activated by neurotransmitter binding, raising biophysical questions about iGluR activation and biological questions about the role of delta iGluRs. We therefore investigated the divergence of delta iGluRs from their iGluR cousins using molecular phylogenetics, electrophysiology, and site-directed mutagenesis. We find that delta iGluRs are found in numerous bilaterian animals (e.g., worms, starfish, and vertebrates) and are closely related to AMPA receptors, both genetically and functionally. Surprisingly, we observe that many iGluRs of the delta family are activated by the classical inhibitory neurotransmitter, γ-aminobutyric acid (GABA). Finally, we identify nine amino acid substitutions that likely gave rise to the inactivity of today’s mammalian delta iGluRs, and these mutations abolish activity when engineered into active invertebrate delta iGluRs, partly by inducing receptor desensitization. These results offer biophysical insight into iGluR activity and point to a role for GABA in excitatory signaling in invertebrates.

SUMMARY: Neurotransmitter ligands electrically excite neurons by activating ionotropic glutamate receptor (iGluR) ion channels. Knowledge of the iGluR amino acid residues that dominate ligand-induced activation would enable the prediction of function from sequence. We therefore explored the molecular determinants of activity in rat NMDA-type iGluRs (NMDA receptors), complex heteromeric iGluRs comprising two glycine-binding GluN1 and two glutamate-binding GluN2 subunits, using amino acid sequence analysis, mutagenesis, and electrophysiology. We find that a broadly conserved aspartate residue controls both ligand potency and channel activity, to the extent that certain substitutions at this position bypass the need for ligand-binding in GluN1 subunits, generating NMDA receptors activated solely by glutamate. Furthermore, we identify a homomeric iGluR from the placozoan Trichoplax adhaerens that has utilized native mutations of this crucial residue to evolve into a leak channel that is inhibited by neurotransmitter binding, pointing to a dominant role of this residue throughout the iGluR superfamily.