Constraining Gating Actions of Specific Subunits During NMDA Receptor Activation

Abstract

Listen

Translate article

NMDA receptor activation gating initiates at its extracellular ligand-binding domain (LBD) and culminates in opening of the associated ion channel. Kinetically, this gating transition occurs through sequential intermediaries with at minimum three pre-open and two open states. The overall gating transition occurs in a concerted manner only after each subunit has bound its respective ligand. Functional NMDA receptors, however, are obligate heterotetramers, typically composed of two glycine-binding GluN1 and two glutamate-binding GluN2 subunits. To study subunit-specific contributions to gating kinetics, we constrained relative movements of the M3/M3-S2 central channel gating axis through intra-subunit disulfide cross-linking of partner positions in the M3-S2 and S2-M4 LBD-ion channel linkers, GluN1(C,C) and GluN2A(C,C). Analysis of single-channel recordings revealed the cross-linked receptors to be inefficient in activation gating, with a significantly lower Po [mean SEM; 0.02 ± 0.005 for GluN1(C,C)-GluN2A vs. 0.61 ± 0.05 for GluN1-GluN2A]. This 30-fold reduction in Po is manifested by both a 6-fold decrease in mean open duration (1.3 ± 0.2 vs. 8.3 ± 0.7 ms) and an 18-fold increase in mean closed duration (94 ± 20 vs. 5.1 ± 0.7 ms). The cross-linked receptors undergo only flickery openings to a short-lived open state (τO1, 0.25 ± 0.05 vs. 0.12 ± 0.01 ms) and virtually do not visit the main long-lived (∼ 8.7 ms) open state. Instead, cross-linked receptor activity is locked into two long-lived pre-open states (τC3, 38 ± 9 vs. 4.2 ± 0.4; τC4, 208 ± 45 vs. 52 ± 3 ms). Thus, GluN1 subunit-specific constraint of its M3/M3-S2 gating axis most strongly affects the kinetic end states of activation gating, with minimal, if any, perturbation of the intermediate states. These results support a tightly coupled LBD to ion channel gating machinery in the NMDA receptor.