The recombinant peptide, DBAYL, a promising therapeutic peptide for type 2 diabetes, is a new, potent, and highly selective agonist for VPAC2 generated through site-directed mutagenesis based on sequence alignments of pituitary adenylate cyclase-activating peptide (PACAP), vasoactive intestinal peptide (VIP), and related analogs. The recombinant DBAYL was used to evaluate its effect and mechanism in blood glucose metabolism and utilization. As much as 28.9 mg recombinant DBAYL peptide with purity over 98% can be obtained from 1 l of Luria-Bertani medium culture by the method established in this study and the prepared DBAYL with four mutations (N10Q, V18L, N29Q, and M added to the N-terminal) were much more stable than BAY55-9837. The half-life of recombinant DBAYL was about 25 folds compared with that of BAY55-9837 in vitro. The bioactivity assay of DBAYL showed that it displaced [(125)I]PACAP38 and [(125)I]VIP from VPAC2 with a half-maximal inhibitory concentration of 48.4 ± 6.9 and 47.1 ± 4.9 nM, respectively, which were significantly lower than that of BAY55-9837, one established VPAC2 agonists. DBAYL enhances the cAMP accumulation in CHO cells expressing human VPAC2 with a half-maximal stimulatory concentration (EC(50)) of 0.68 nM, whereas the receptor potency of DBAYL at human VPAC1 (EC(50) of 737 nM) was only 1/1083 of that at human VPAC2, and DBAYL had no activity toward human PAC1 receptor. Western blot analysis of the key proteins of insulin receptor signaling pathway: insulin receptor substrate 1 (IRS-1) and glucose transporter 4 (GLUT4) indicated that the DBAYL could significantly induce the insulin-stimulated IRS-1 and GLUT4 expression more efficiently than BAY55-9837 and VIP in adipocytes. Compared with BAY55-9837 and PACAP38, the recombinant peptide DBAYL can more efficiently promote insulin release and decrease plasma glucose level in Institute of Cancer Research (ICR) mice. These results suggested that DBAYL could efficiently improve glucose uptake and glucose-dependent insulin secretion by VPAC2-mediated effect.