Interactions among GYKI-52466, cyclothiazide, and aniracetam at recombinant AMPA and kainate receptors.

We examined the actions of cyclothiazide, aniracetam, and 1-(4-aminophenyl)-4-methyl-7,8-methylenedioxy-5H-2,3-benzodiazepine (GYKI-52466) on recombinant alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA) and kainate receptors. Receptors expressed in Xenopus oocytes or human embryonic kidney 293 cells were characterized using voltage and patch-clamp electrophysiology. Aniracetam and cyclothiazide potentiated AMPA receptor currents by slowing or blocking desensitization. Cyclothiazide was more potent at receptors consisting of flip subunits compared with receptors consisting of flop subunits, whereas aniracetam appeared to be more efficacious at flop receptors. The potency of GYKI-52466 did not differ in heteromeric flip or flop containing AMPA receptors, but GYKI-52466 was less potent at homomeric GluRAi and GluRDi receptors. At heteromeric AMPA receptors, 50 microM cyclothiazide increased the IC50 value for GYKI-52466 significantly. The increase was largest in GluRBi/Di receptors where the IC50 value shifted from 21.9 microM (95% confidence interval, 12.0-39.8 microM) to 126 microM (95% confidence interval, 72.4-214 microM) in the presence of cyclothiazide. In contrast, 100 microM GYKI-52466 did not alter the EC50 of cyclothiazide at GluRBi/Di receptors nor did it markedly change the maximal potentiation induced by cyclothiazide. At GluRBi/Di receptors transiently expressed in human embryonic kidney 293 cells, 30 microM GYKI-52466 inhibited the steady state and the peak current evoked by 300 microns L-glutamate to the same extent (34.5 +/- 12% and 27.3 +/- 13.0%, respectively; five experiments), and GYKI-52466 did not alter the apparent rate of desensitization (tau = 15.7 +/- 4.7 and 17.5 +/- 8.3 msec in the absence and presence of GYKI-52466, respectively; five experiments). GYKI-52466 inhibited L-glutamate currents in the presence and absence of 10 microM cyclothiazide, but GYKI-52466 never restored the desensitization that was blocked by cyclothiazide. Furthermore, GYKI-52466 inhibited L-glutamate currents mediated by homomeric Glu6 receptors, which are not potentiated by cyclothiazide. Our data suggest that the effect of cyclothiazide on the affinity of GYKI-52466 for its binding site is allosteric and that the positive modulatory effect of cyclothiazide and the negative modulatory effect of GYKI-52466 result from binding to separate sites on recombinant subunits.