1. This paper describes the effects of GR117289 (1-[[3-bromo-2-[2-(1H-tetrazol-5-yl)phenyl]-5-benzo-furanyl]methyl ]-2-butyl-4-chloro-1H-imidazole-5-carboxylic acid) at angiotensin receptors and binding sites in rabbit aorta, rat liver and bovine cerebellum preparations in vitro. 2. In rabbit isolated aortic strips, GR117289 (0.3, 1 and 3 nM) caused a concentration-related, insurmountable suppression of the concentration-response curve to angiotensin II (AII). When the contact time was increased, a greater degree of antagonism of AII was observed, suggesting that GR117289 is slow to reach equilibrium. A pKB of 9.8 +/- 0.1 was calculated for GR117289 after 3 h incubation. GR117289 (1 microM) did not affect contractile responses to phenylephrine or 5-hydroxytryptamine (5-HT) in the rabbit aorta. 3. GR117289 (1 nM) alone caused a marked suppression and a slight rightward displacement of the AII concentration-response curve. Co-incubation with the competitive, surmountable AT1 receptor antagonist, losartan (10 nM, 100 nM and 1 microM), resulted in a concentration-related upward and rightward displacement of the concentration-response curve to subsequently administered AII. In separate experiments in which preparations were pre-incubated with GR117289 (1 nM), subsequent addition of losartan (1 microM) for 2, 15 or 45 min caused a further, but similar, rightward displacement of the concentration-response curve to subsequently administered AII with a time-dependent increase in the maximum response.4. Suppression of All-induced contractile responses, caused by superfusion with GRI17289 (0.3, 1 or 3 nM) was not reversed by continuously washing the tissues for 3 h; in fact, the potency of GRI 17289 was slightly enhanced after this period.5. In rat liver membranes, GRI17289 was a potent competitor with [3H]-AII for AT, binding sites(pKi = 8.7 +/- 0.1) but in bovine cerebellum membranes, it was a very weak competitor for AT2 binding sites (pKi<6). Pre-incubation of rat liver membranes with GRI17289 had little effect on its affinity(pKi = 9.1 +/- 0.21), but increasing the concentration of bovine serum albumen in the assay buffer from 0.001% to 0.1% w/v decreased affinity (pKi= 7.5 +/- 0.1).6. In saturation binding experiments in rat liver membranes, GRI 17289 (12 nM) increased the Kd of[3H]-AII from 0.28 +/- 0.06 nM to 0.37 +/- 0.02 nM, and decreased Bm. from 10.0 +/- 0.1 to 5.6 +/-0.3 fmol mg' tissue. In other experiments, GR1 17289 (1 jIM) did not alter the rate of dissociation of[3H]-AII from AT1 binding sites, following addition of excess unlabelled All.7. In rabbit aorta vascular smooth muscle membranes, GR1 17289 competed with ['25I]-Sar'1le8 All for binding to AT, binding sites. In the presence of 0.1% w/v bovine serum albumen, a pIC50 of 7.6 +/- 0.1 was calculated. Under the same conditions, but with rat liver membranes, a pIC50 of 7.8 +/- 0.1 was determined.8. Taken together, these results show that GRI17289 is a potent, specific, selective and insurmountable antagonist at angiotensin AT, receptors. Its profile in the rabbit aorta is consistent with the proposalthat GRI17289 is a slowly reversible (pseudo-irreversible) antagonist at these receptors.