Vol. 297, Issue 3, 906-914, June 2001

Structural Modifications to an N-Methyl-D-aspartate Receptor Antagonist Result in Large Differences in Trapping Block

G. A. R. Mealing, T. H. Lanthorn, D. L. Small, R. J. Murray, K. C. Mattes, T. M. Comas and P. Morley

Institute for Biological Sciences, National Research Council of Canada, Ottawa, Ontario, Canada (G.A.R.M., D.L.S., T.M.C., P.M.); and AstraZeneca R&D Boston, Worcester, Massachusetts (T.H.L., R.J.M., K.C.M.)

Differences in the degree of trapping of initial block by N-methyl-D-aspartate (NMDA) receptor antagonists may affect their safety and, hence, suitability for clinical trials. In this comparative study, 23 compounds structurally related to the low-affinity, use-dependent NMDA receptor antagonist (S)--phenyl-2-pyridineethanamine dihydrochloride (AR-R15896AR) were examined to determine the degree of trapping block they exhibit. Compounds were tested at concentrations that produced a comparable initial 80% block of NMDA-mediated whole-cell current in rat cortical cultures. A wide range of values of trapping block was found, indicating that trapping is not an all-or-none event. Fifteen of the compounds trapped significantly more than the 54 ± 3% of initial block trapped by AR-R15896AR. The off-rates of these compounds were slower than that of AR-R15896AR. Only 2 of the 23 compounds trapped significantly less than AR-R15896AR. AR-R15808, the piperidine analog of AR-R15896AR, appeared to trap only 8 ± 3% of its initial block, although its fast off-rate confounded accurate quantification of trapping. AR-R26952, which, like AR-R15896AR, contains a pyridine in place of a phenyl group, trapped 40 ± 5% of its initial block and exhibited kinetics comparable with AR-R15896AR. Structure-activity analysis suggested that the presence of two basic nitrogen atoms and decreased hydrophobicity led to decreased trapping. There was no correlation between trapping and lipophilicity as would be expected if closed-channel egress was due to escape through the lipid bilayer. However, there was a positive correlation between off-rate and degree of trapping. Models that can account for partial trapping are presented.