In addition to blocking cyclooxygenases, members of the fenamate group of nonsteroidal anti-inflammatory drugs have been proposed to affect brain GABA A receptors. Using quantitative autoradiography with GABA A receptor-associated ionophore ligand [35 S]t-butylbicyclophosphorothionate (TBPS) on rat brain sections, one of the fenamates, niflumate, at micromolar concentration was found to potentiate GABA actions in most brain areas, whereas being in the cerebellar granule cell layer an efficient antagonist similar to furosemide. With recombinant GABA A receptors expressed in Xenopus laevis oocytes, we found that niflumate potentiated 3 M GABA responses up to 160% and shifted the GABA concentration-response curve to the left in ␣1␤2␥2 receptors, the predominant GABA A receptor subtype in the brain. This effect needed the ␥2 subunit, because on ␣1␤2 receptors, niflumate exhibited solely an antagonistic effect at high concentrations. The potentiation was not abolished by the specific benzodiazepine site antagonist flumazenil. Niflumate acted as a potent antagonist of ␣6␤2 receptors (with or without ␥2 subunit) and of ␣X␤2␥2 receptors containing a chimeric ␣1 to ␣6 subunit, which suggests that niflumate antagonism is dependent on the same transmembrane domain 1-and 2-including fragment of the ␣6 subunit as furosemide antagonism. This antagonism was noncompetitive because the maximal GABA response, but not the potency, was reduced by niflumate. These data show receptor subtypedependent positive and negative modulatory actions of niflumate on GABA A receptors at clinically relevant concentrations, and they suggest the existence of a novel positive modulatory site on ␣1␤2␥2 receptors that is dependent on the ␥2 subunit but not associated with the benzodiazepine binding site.The structure and function of ligand-gated ion channels and their integral receptors is not well understood; therefore, it is important to establish new tools to probe various receptor subunit interactions, which can lead to novel ideas for the development of receptor subtype-selective drugs. GABA is the major inhibitory neurotransmitter in the mammalian brain. Its fast actions are mediated through ligand-gated anion channels, GABA type A (GABA A ) receptors, which are distributed throughout the brain. Receptor activation at cell membrane alters the conformation of the receptor normally leading to opening of the ionophore and anion flux, hyperpolarization of the cell, and inhibition of neuronal excitability. The GABA A receptor subtypes are formed by temporal and spatial regulation of subunit [␣1-6, ␤1-3, ␥1-3, ␦, ⑀, , 1-3, and ; Barnard et al., 1998;Bonnert et al., 1999;Sinkkonen et al., 2000) expression in brain regions and/or by cellular regulation of assembly to pentameric receptor complexes. GABA A ergic drugs, such as benzodiazepines, barbiturates, and volatile anesthetics, enhance the actions of GABA (Sieghart, 1995) and are used to treat, e.g., anxiety, insomnia, and epilepsy, and in general anesthesia. Many drug effects on GABA A rece...

The nonsteroidal anti-inflammatory drug (NSAID) niflumic acid, a fenamate in structure, has many molecular targets, one of them being specific subtypes of the main inhibitory ligand-gated anion channel, the GABA A receptor. Here, we report on the effects of other fenamates and other classes of NSAIDs on brain picrotoxinin-sensitive GABA A receptors, using an autoradiographic assay with [ 35 S]TBPS as a ligand on mouse brain sections. We found that the other fenamates studied (flufenamic acid, meclofenamic acid, mefenamic acid and tolfenamic acid) affected the autoradiographic signal at low micromolar concentrations in a facilitatory-like allosteric fashion, i.e., without having affinity to the [ 35 S]TBPS binding site. Unlike niflumic acid that shows clear preference for inhibiting cerebellar granule cell layer GABA A receptors, the other fenamates showed little brain regional selectivity, indicating that their actions are not receptor-subtype selective. Of the non-fenamate NSAIDs studied at 100 µM concentration, diclofenac induced the greatest inhibition of the binding, which is not surprising as it has close structural similarity with the potent fenamate meclofenamic acid. Using two-electrode voltage-clamp assays on Xenopus oocytes, the effect of niflumic acid was found to be dependent on the β subunit variant and the presence of γ2 subunit in rat recombinant α1β and α1βγ2 GABA A receptors, with the β1 allowing the niflumic acid inhibition and β3 the stimulation of the receptor-mediated currents. In summary, the fenamate NSAIDs constitute an interesting class of compounds that could be used for development of potent GABA A receptor allosteric agonists with other targets to moderate inflammation, pain and associated anxiety/depression.