Additivities of compounds that increase the numbers of high affinity [3H]muscimol binding sites by different amounts define more than 9 GABA(A) receptor complexes in rat forebrain: implications for schizophrenia and clozapine research.

BACKGROUND

The numbers of [3H]MUS binding sites were reported to be elevated in layers II and III, but not V or VI, in cingulate cortex of schizophrenic brains post mortem. These increases in [3H]MUS binding sites are probably due to compensatory up-regulation of GABA(A) receptors on pyramidal cells as a consequence of a selective loss of GABAergic interneurons in layer II of cingulate cortex. The number of [3H]flunitrazepam binding sites was reported to be reduced in schizophrenic cingulate cortex, and this may directly reflect the loss of GABAergic interneurons. Chronic administration of clozapine to rats was reported to significantly reduce the numbers of [3H]MUS binding sites in temporal cortex and hippocampus which may be due to selective blockade of GABA(A) receptors on GABAergic interneurons that make synaptic contact with pyramidal cells. Basket cells are GABAergic interneurons that make synaptic contact with pyramidal cells as well as other interneurons. Basket cells can also generate both theta and gamma oscillations. Clozapine increases the power of theta and gamma EEG. Schizophrenic patients show reduced EEG power at 40 Hz (gamma frequency) but not at lower frequencies during auditory stimulation. The GABA(A) receptor blocker bicuculline at 10 nM, but not 10 microM, was reported to increase the amplitude of slow oscillations (< or =1 Hz) in rat hippocampal slices. It therefore seems possible that clozapine, by selectively blocking another GABA(A) receptor, could increase the amplitude of gamma oscillations.

FINDINGS

Twenty-six compounds that inhibit [35S]TBPS binding in ways that are reversible by 10 nM R-5135 were found to increase [3H]MUS binding to membranes prepared from rat whole forebrain. In almost all cases the increases in binding were due to increases in the number of binding sites with little effect on affinity (Kd) for [3H]MUS. Concentration-response curves for the compounds revealed maximum increases in [3H]MUS (Esat) binding ranging from 140% (for meclizine) to 313% of control for honokiol. Additivity experiments showed that propofol (44% above control) and diflunisal (50% above control) were almost entirely additive, but there was also a small, but significant overlap, suggesting the existence of three groups of [3H]MUS binding sites defined by propofol and diflunisal. Meclizine was entirely additive with both propofol and diflunisal, indicating the existence of a fourth [3H]MUS binding site. Alphaxalone is also completely additive with meclizine, and has an Esat value significantly larger than that for propofol + diflunisal suggesting a fifth [3H]MUS binding site. The Esat for mefenamate is significantly greater than the Esat for alphaxalone, and mefenamate is also completely additive with meclizine, suggesting the existence of a sixth [3H]MUS binding site. The Esat for magnolol is significantly greater then the Esat, for mefenamate, and the Esat for honokiol is greater than that for magnolol, suggesting, but not proving, the existence of a seventh and an eighth group of [3H]MUS binding sites. The binding of [3H]MUS alone, without enhancers may represent a ninth group of binding sites which is probably heterogeneous as indicated by the very low pseudo Hill coefficients for bicuculline and strychnine in displacing [3H]MUS without enhancer. Altogether, our results suggest the existence of more than 9 different [3H]MUS binding sites. Clozapine was a very weak overall displacer of [3H]MUS (IC50 = 280 microM). However, 5 microM clozapine reduced [3H]MUS binding 6% (P < 0.0001, n = 10) and significantly reduced [3H]MUS binding enhanced by propofol (approximately 14%) or clotrimazole (approximately 17%) but not 17 other compounds tested.

TENTATIVE CONCLUSIONS

In the absence of enhancers [3H]MUS may bind preferentially to GABA(A) receptors on pyramidal cells and less to interneurons in cerebral cortex. Conversely, [3H]flunitrazepam may bind preferentially to GABA(A) receptors (allosterically) on interneurons and less to pyramidal cells. Clozapine appears to selectively block a small fraction (10-20%) of [3H]MUS binding sites with an IC50 value in the low micromolar range. This fraction may be preferentially located on certain GABAergic interneurons (basket cells?) that make synaptic contact with pyramidal cells. The blockade of these GABA(A) receptors by clozapine would be expected to increase the firing rate of the interneurons and the release of GABA onto pyramidal cells. Such blockade would also increase the generation of gamma oscillations by the basket cells. Some of these interneurons appear to be destroyed selectively, probably during the second trimester of gestation by a non-paralytic polio virus, in individuals who wil