Xenovulene A (XR368) is a natural product exhibiting little structural resemblance with classical benzodiazepines yet is able to displace high-affinity ligand binding to the benzodiazepine site of the gamma-aminobutyric acid (GABA)A receptor. We have characterized this compound and an associated congener (XR7009) by use of radioligand binding and electrophysiological methodologies with native neurons and the Xenopus oocyte expression system. Xenovulene A, and the more potent XR7009, inhibited [3H]flunitrazepam binding to rat forebrain with Ki values of 7 and 192 nM, and 1.7 and 42 nM, respectively, each site accounting for approximately 50% of the total specific binding. In cerebellar and spinal cord membranes, these ligands identified only single binding sites. These ligands demonstrated no intrinsic agonist activity at recombinant GABA(A) receptors comprising alpha1beta1gamma2S subunits expressed in Xenopus oocytes, yet at 1 microM both significantly potentiated the GABA-induced response and reduced the GABA EC50 from 10.9 (control) to 5.1 (Xenovulene A) or 2.7 microM (XR7009). The rank potency order for enhancement of the 10 microM GABA response is: XR7009 (EC50, 0.02 microM) > diazepam (0.03) > Xenovulene A (0.05) > flurazepam (0.17). The activity of XR368 and XR7009 was reduced by the benzodiazepine antagonist, flumazenil, and absent in receptors devoid of the gamma2 subunit. These agents exhibited receptor subtype selectivity because alpha3beta1gamma2S receptors were less sensitive to these compounds relative to alpha1 subunit-containing receptors, whereas alpha6beta1gamma2S receptors were completely insensitive. Potentiation of the response to GABA on native GABA(A) receptors in cortical neurons substantiates the profile of the novel structures of Xenovulene A and XR7009 as specific benzodiazepine agonists.