Recently, tissue segment binding method with a hydrophilic radioligand [(3)H]-CGP12177 was developed to detect plasma membrane beta-adrenoceptors in rat heart (Horinouchi et al., 2006). In the present study, propranolol (40 mg kg(-1) day(-1)), atenolol (40 mg kg(-1) day(-1)) and bevantolol (200 mg kg(-1) day(-1)) were administered to rats for 6 weeks, and the changes of plasma membrane beta-adrenoceptors and their mRNA expression in rat ventricle were examined. Chronic administration of propranolol increased the beta(1)-adrenoceptors but decreased the beta(2)-adrenoceptors without changing total amount of plasma membrane beta-adrenoceptors. Atenolol increased both plasma membrane beta(1)- and beta(2)-adrenoceptors, whereas bevantolol had no effect on the beta-adrenoceptor density and their subtype proportions. In contrast, the density of beta-adrenoceptors detected in conventional homogenate binding study was extremely low (approximately 60% of plasma membrane beta-adrenoceptors detected with the tissue segment binding method) and the effects of chronic administration of beta-adrenoceptor antagonists were not necessarily in accord with those at the plasma membrane beta-adrenoceptors. The mRNA levels of beta(1)- and beta(2)-adrenoceptors were not altered by propranolol treatment, while beta(1)-adrenoceptor mRNA significantly decreased after administration of atenolol or bevantolol without changing the level of beta(2)-adrenoceptor mRNA. The present binding study with intact tissue segments clearly shows that the plasma membrane beta(1)- and beta(2)-adrenoceptors of rat heart, in contrast to the homogenate binding sites and the mRNA levels, are differently affected by chronic treatment with three beta-adrenoceptor antagonists; up- and down-regulations of beta(1)- and beta(2)-adrenoceptors, respectively, by propranolol, and up-regulation of both the subtypes by atenolol, but no significant change in both the subtypes by bevantolol.