Gilling · Neuropharmacology 2009 · in vitro electrophysiological and radioligand binding study · n=?

Potency, voltage-dependency, agonist concentration-dependency, blocking kinetics and partial untrapping of the uncompetitive N-methyl-D-aspartate (NMDA) channel blocker memantine at human NMDA (GluN1/GluN2A) receptors.

Cited 74 times in the scientific literature.

Level 5 - mechanism / opinion, no new human data

In vitro bench electrophysiology and binding assay using recombinant human receptors and post-mortem tissue.

PubMed 19371579 · doi:10.1016/j.neuropharm.2009.01.012 · record verified 2026-08-28

What was done

The authors compared the binding potency and biophysical properties of the uncompetitive NMDA channel blockers memantine, ketamine, and (+)MK-801. They evaluated binding to NMDA receptors in post-mortem human cortical tissue, antagonism of intracellular Ca2+ responses in HEK-293 cells expressing human GluN1/GluN2A receptors, and patch-clamp electrophysiology in these transfected cells to measure blocking kinetics, voltage dependency, and channel trapping.

What was found

Memantine exhibited moderate affinity at human GluN1/GluN2A receptors with an IC50 at -70 mV of 0.79 ± 0.02 µM (Hill coefficient 0.92 ± 0.02) and strong voltage dependency (delta = 0.90 ± 0.09). At 10 µM, memantine showed rapid double-exponential blocking kinetics (onset tau_fast = 273 ± 25 ms [69% weight], onset tau_slow = 2756 ± 296 ms; offset tau_fast = 415 ± 82 ms [38% weight], offset tau_slow = 5107 ± 1204 ms) and partial untrapping of approximately 20%. Ketamine showed similar potency (IC50 = 0.71 ± 0.03 µM, Hill = 0.84 ± 0.02) but lower voltage dependency (delta = 0.79 ± 0.04), slower single-exponential kinetics (kon = 0.15 ± 0.05 × 10^6 M^-1 s^-1; koff = 0.22 ± 0.05 s^-1 vs memantine normalized kon = 0.32 ± 0.11 × 10^6 M^-1 s^-1; koff = 0.53 ± 0.10 s^-1), and was fully trapped.

Why it matters

The study verifies that memantine's distinctive biophysical properties—fast kinetics, strong voltage-dependency, and partial untrapping—apply to human GluN1/GluN2A receptors just as in rodent models, supporting its proposed mechanism for clinical tolerability.

Limits

This was an in vitro study performed on recombinant GluN1/GluN2A receptors in HEK-293 cells and post-mortem cortical membrane preparations, which does not capture in vivo neural network dynamics, other human NMDA receptor subunit combinations (e.g., GluN2B, GluN2C, GluN2D), or clinical endpoints. The abstract does not report the sample size of human tissue donors or number of recorded cells.

Cited by