Allosteric modulation of ligand-gated ion channels provides a powerful mechanism to fine-tune their activity without competing with endogenous ligands. In the case of NMDA receptors (NMDARs), which are critical for excitatory neurotransmission and synaptic plasticity, allosteric modulators represent potential therapeutic tools, particularly in conditions involving NMDAR hypofunction. Here, we characterize EAR-20, a 17-amino-acid peptide derived from the marine cone snail toxin Conantokin-G, as a novel positive allosteric modulator (PAM) of NMDARs. Using molecular docking, whole-cell and single-channel patch-clamp electrophysiology, and recordings in cultured hippocampal neurons, we show that EAR-20 enhances receptor function by increasing channel open probability and reducing desensitization, and can even activate NMDARs in the absence of exogenous glutamate and glycine, albeit to a lower extent. EAR-20 decreased desensitization, potentiating GluN1-GluN2A and GluN1-GluN2B receptors more than twofold, modestly enhanced (∼25%) GluN1-GluN2A-GluN2B tri-heteromers, and increased NMDAR-mediated currents in primary hippocampal neurons. Molecular docking identified a binding site at the GluN1-GluN2B interface, with Ser773 in GluN1 being critical for the modulatory effect. Importantly, EAR-20 partially rescued hypofunctional NMDARs carrying patient-derived loss-of-function mutations. Together, these findings identify EAR-20 as a novel subunit-dependent positive allosteric modulator with the potential to inspire the development of small molecules targeting the same binding site, offering proof of concept for therapeutic strategies to treat neurological and neurodevelopmental disorders.
The amino acid glutamate is the main excitatory neurotransmitter in the mammalian central nervous system (CNS) acting on two types of receptors: ionotropic receptors (cation-permeable ligand-gated ion channels; iGluRs) and metabotropic receptors (G protein-coupled receptors; mGluRs). While metabotropic receptors mediate long-term modulatory effects, iGluRs are responsible for the fast glutamatergic neurotransmission in the CNS, which is essential for normal brain development and function. iGluRs can be classified into three subfamilies based on gene sequence similarities: AMPA receptors (composed of GluA1–4 subunits), kainate receptors (composed of GluK1–5 subunits), and NMDA receptors (composed of GluN1, GluN2A–D, and GluN3A–B subunits), each with distinct biophysical properties and functional roles in the CNS.
NMDARs are heterotetramers composed of four subunits that can be assembled into either di-heteromeric or tri-heteromeric forms with two obligatory glycine-binding GluN1 subunits and two other identical or different glutamate-binding GluN2 subunits, namely GluN2A, GluN2B, GluN2C, or GluN2D. Each NMDAR subunit is composed of four distinct domains: the amino-terminal domain (ATD), the ligand-binding domain (LBD), the transmembrane domain (TMD), and the carboxy-terminal domain (CTD). Subunit composition provides unique characteristics to each type of NMDAR, resulting in different functional properties including the open time of the channel, channel conductance, permeability to Ca 2+, sensitivity to blockage by Mg 2+, affinity for agonists glycine or D-serine and L-glutamate, and sensitivity and sensibility to allosteric modulators.
NMDA receptors are unique among glutamate-gated ion channels because they require the binding of two agonists–glutamate and glycine (or D-serine) – in addition to membrane depolarization for activation. At resting membrane potentials, extracellular Mg 2+ ions enter and occupy a site in the channel pore, producing a strong voltage-dependent block. When synaptic depolarization occurs (typically through AMPAR activation), the driving force for Mg 2+ entry is reduced and the ion is expelled from the pore, thereby relieving the block and allowing current to flow. Besides, NMDARs also differ from other ionotropic glutamate receptors in their higher affinity for glutamate, a property that contributes to their relatively slow activation kinetics. On the other side, the glycine-binding site, located on the GluN1 subunit, has an even higher affinity and very low concentrations of glycine are enough to saturate it. In fact, ambient extracellular glycine is often sufficient for activating them, although at many forebrain synapses D-serine is the predominant co-agonist. Taken together, these properties establish NMDA receptors as coincidence detectors that respond only when presynaptic glutamate release coincides with postsynaptic depolarization.
NMDARs not only contribute to fast excitatory synaptic transmission but also trigger signaling events upon Ca 2+ entry through their channel pore. These calcium-mediated processes are essential for the activation of physiological functions such as neuronal development and synaptic plasticity, which are fundamental to learning and memory processes. Their crucial role is highlighted in numerous studies showing that dysfunction of NMDARs–due to hypo- or hyperactivation–is associated with neurodevelopmental disorders and neurological diseases. One example of such importance is found in a group of neurodevelopmental disorders known as GRIN-related disorders, resulting from mutations of GRIN genes that encode for the GluN2 subunits of the NMDAR. These genetic alterations lead to intellectual disability along with other neurological and neurodevelopmental impairments. The pathological outcome of primary and secondary NMDAR dysfunction stimulated pharmacological research over the years, testing the use of agonists and antagonists, such as memantine, for treating certain NMDAR related disorders. Although memantine’s side effects are generally mild, agonists (e.g. glutamate analogs) and antagonists (e.g. ketamine or PCP) of NMDARs can cause more serious side effects, such as dissociative symptoms, cognitive impairment, psychosis, seizures or even cardiovascular problems. That turned special attention on NMDAR allosteric modulators, both positive (PAMs) and negative (NAMs), as a safer and more effective alternative as a potential therapeutic strategy for these neurological disorders. Interestingly, PAMs ability to enhance NMDAR activity might be suitable for personalized medicine in patients harboring GRIN loss-of-function, potentially restoring or mitigating these deleterious effects.
The EAR-20 peptide was originally designed as a scrambled control peptide derived from a set of modified peptides based on Conantokin-G (Con-G), a well-known NMDAR antagonist, which specifically targets the GluN2B ligand binding domain. These modified peptides were developed as part of a strategy to finely regulate NMDAR activity through structural variants inspired by Con-G. EAR-20 is not a direct scrambled version of Con-G itself, but rather a scrambled sequence of one of the Con-G–based modified peptides (see methods), and initially, the EAR-20 peptide was intended as a control for electrophysiological tests. Surprisingly, it was observed that EAR-20 activated currents in hippocampal neurons in the absence of added glycine and glutamate. This unexpected finding prompted us to further investigate the effects of EAR-20 on NMDA receptors using electrophysiological techniques. In recombinant receptors expressed in HEK-293T cell line, we have found that EAR-20 peptide exhibits a small but significant agonistic activity in GluN1-GluN2B di-heteromers, as well as a considerable PAM activity of GluN1-GluN2A, GluN1-GluN2B and GluN1-GluN2D receptors. This allosteric interaction enhances the effects of the co-agonists glutamate and glycine by increasing the channel open probability. EAR-20 peptide also potentiates NMDARs in mouse hippocampal neurons, suggesting that the peptide may act as a functional PAM in native receptors. Moreover, EAR-20 exerts a rescuing effect on hypofunctional NMDARs, as tested in loss-of-function variants where receptor activity is compromised, further supporting its role as a functional modulator with potential therapeutic relevance.
EAR-20 peptide (1 KLGMRSELQIDNDQDAD 17) was synthesized manually by solid-phase peptide synthesis (SPPS), using for this purpose the alpha nitrogen of the amino acids protected with the base labile Fmoc group as described in García Díaz (2024). The synthesis method is also detailed in Reyes-Montaño et al. (2023).
The expression plasmids for rat GluN1 and GFP-GluN2A and GFP-GluN2B were provided by S. Vicini (Georgetown University Medical Center, Washington, United States). Dr. J.W. Johnson (University of Pittsburgh, Pittsburgh, United States of America) provided HA-GluN2C and HA-GluN2D plasmids. The plasmids used to analyze triheteromeric NMDARs and N-terminal-deleted GluN2 (GluN2 ΔNTD) were kindly provided by Dr. Pierre Paoletti [École Normale Supérieure Paris, France, EU]. Nucleotide changes for producing GRIN variants were achieved by oligonucleotide-directed mutagenesis, using the QuikChange II XL protocol with turbo Pfu DNA polymerase (Stratagene La Jolla, CA, United States) to replicate the parental DNA strand with the desired mismatch incorporated into the primer. Methylated parental DNA was digested with DpnI for 1 h at 37°C, and the nicked mutant DNA was transformed into XL1-Blue Super Competent Cells (Stratagene La Jolla, CA, United States). The Bacteria were spun down, and plasmid DNA isolated using the Qiagen Spin Miniprep kit (Hilden, Germany). Sequences were verified by Sanger sequencing (STAB vida, Caparica, Portugal).
HEK-293T cell lines were obtained from the American Type Culture Collection and maintained in Dulbecco’s modified Eagle’s mediu
