We have previously shown that substitution of the D-Tic-Oic dipeptide by a (3S)-[amino]-5-(carbonylmethyl)-2,3-dihydro-1, 5-benzothiazepin-4(5H)-one (D-BT) moiety in the bradykinin B(2) receptor antagonist HOE 140 resulted in a full potent and selective bradykinin B(2) receptor agonist (H-DArg-Arg-Pro-Hyp-Gly-Thi-Ser-D-BT-Arg-OH, JMV1116) exhibiting a high affinity for the human receptor (K(i) 0.7 nM). In the present study, we have investigated the effects of replacement of the D-Tic-Oic moiety by various constrained dipeptide mimetics. The resulting compounds were tested for their binding affinity toward the cloned human B(2) receptor and for their functional interaction with the bradykinin-induced contraction of isolated human umbilical vein. Subsequently, we have designed novel bradykinin B(2) receptor agonists which are likely to be resistant to enzymatic cleavage by endopeptidases and which might represent interesting new pharmacological tools. In an attempt to increase the potency of compound JMV1116, both its N-terminal part and the D-BT moiety were modified. Substitution of the D-arginine residue by a L-lysine residue led to a 10-fold more potent bradykinin B(2) ligand [compound 22 (JMV1465) (K(i) 0.07 nM)], retaining full agonist activity on human umbilical vein. Substitution of the D-BT moiety by a (3S)-[amino]-5-(carbonylmethyl)-2,3-dihydro-8-methyl-1, 5-benzothiazepin-4(5H)-one [D-BT(Me)] moiety led to compound 23 (JMV1609) which exhibited a higher agonist activity (pD(2) = 7.4) than JMV1116 (pD(2) = 6.8).