The possible existence of a beta3-adrenergic receptor (beta3-AR) in the human detrusor muscle was investigated by in vitro functional studies and analysis of mRNA expression. Isoprenaline, noradrenaline and adrenaline each produced a concentration-dependent relaxation of the human detrusor. The rank order for their relaxing potencies was isoprenaline (pD2 6.37+/-0.07) > or = noradrenaline (pD2 6.07+/-0.12) > or = adrenaline (pD2 5.88< or =0.11). Neither dobutamine (beta1- and beta2-AR agonist) nor procaterol (beta2-AR agonist) produced any significant relaxation at concentrations up to 10(-5) M. BRL37344A, CL316243 and CGP-12177A (beta3-AR agonists), relaxed the preparations significantly at concentrations higher than 10(-6) M. The pD2 values for BRL37344A, CL316243 and CGP-12177A were 6.42+/-0.25, 5.53+/-0.09 and 5.74+/-0.14, respectively. CGP-20712A (10(-7) - 10(-5) M), a beta1-AR antagonist, did not affect the isoprenaline-induced relaxation. On the other hand, ICI-118,551, a beta2-AR antagonist, produced a rightward parallel shift of the concentration-relaxation curve for isoprenaline only at the highest concentration used (10(-5) > M) and its pKB value was 5.71+/-0.19. Moreover, SR58894A (10(-7) - 10(-5) M), a beta3-AR antagonist, caused a rightward shift of the concentration-relaxation curve for isoprenaline in a concentration-dependent manner. The pA2 value and slope obtained from Schild plots were 6.24+/-0.20 and 0.68+/-0.31. The beta1-, beta2- and beta3-AR mRNAs were all positively expressed in detrusor smooth muscle preparations in a reverse transcription polymerase chain reaction assay. In conclusion, the present results provide the first evidence for the existence of the beta3-AR subtype in the human detrusor. They also suggest that the relaxation induced by adrenergic stimulation of the human detrusor is mediated mainly through beta3-AR activation.