We previously reported that stimulation of a-adrenoceptors produced a triphasic time response-curve composed of an initial positive phase followed by a negative phase and then a second positive inotropic phase in the left atria of guinea pigs (1); and the positive phase was preferentially inhibited by prazosin, whereas the negative one was inhibited by yohimbine and phentolamine (2). In view of the reported selectivity of these antagonists on a, and a2-adrenoceptors (3-5), these results may indicate that the positive inotropic effect (PIE) and the negative one induced by stimulation of a-adrenoceptors are mediated by a, and a2-adrenoceptors, respectively.Recently, a newly developed antihyper tensive compound, 2-[4-(n-butyryl)-homo piperazine-1 -yl]-4-amino-6,7-dimethoxy quinazoline (E-643), was reported to be a specific and competitive inhibitor of norepin ephrine at a-adrenoceptors (6). A com parative study of this compound and other a-adrenoceptor antagonists demonstrated its potent and highly antagonistic property on postsynaptic a-adrenoceptors (7). It was also shown that [3H] E-643 binds specifically to a, -adrenoceptors in rat brain and peripheral organs (8).The present study was designed to confirm whether E-643 blocks selectively the a adrenoceptor-mediated PIE in the left atria of guinea pigs as prazosin did.The experimental methods and procedure were essentially the same as described in the previous paper (1). Briefly, the left atria isolated from the hearts of guinea pigs (310 480 g) were suspended in a 50 ml organ bath containing Krebs-Henseleit solution bubbled with 95% 02 and 5% C02 at a temperature of 30±1 °C, and it was driven electrically with rectangular pulses of 1 msec duration and of an intensity twice the threshold at a constant rate. The isometric tension was registered with a transducer (San-ei Sokki 45196) on a penrecorder (Nihon Kohden RJG-3026) through a preamplifier (Nihon Kohden RP-5). The preparations were allowed to equilibrate for 60 min at a fre quency of 1 Hz under the resting tension of 2 g. The cumulative dose-response curves (DRC) for the PIE of phenylephrine (Phe) were determined in the preparations driven at 1 Hz. The time-response curves of Phe were obtained at a stimulation rate of 0.2 Hz since the triphasic pattern in the time response curves of Phe has been shown to be more prominent at this rate of stimulation (1). When sotalol or E-643 was used, they were added 45 min before the administration of Phe. Drugs used were 1 -phenylephrine hydrochloride (Sigma), sotalol hydrochloride (Bristol), and E-643 hydrochloride (Eisai). Statistical significance was estimated by means of the Student's t-test.The shift of the DRC for the PIE of Phe to the right by 5x10-5 M sotalol was not parallel, but it was more pronounced in the upper than lower part of the DRC. On the