Glucagon-like peptide-1 (GLP-1) is an incretin hormone that lowers blood glucose after meals in type 2 diabetes mellitus. The therapeutic potential of GLP-1 in diabetes is limited by rapid inactivation by the enzyme dipeptidylpeptidase-4 (DPP-4). Metformin has been reported to inhibit DPP-4. Here we investigated the acute effects of metformin and GLP-1 alone or in combination on plasma DPP-4 activity, active GLP-1 concentrations, and glucose lowering in type 2 diabetes mellitus. Ten subjects with type 2 diabetes mellitus (8 male and 2 female; age, 68.7 ± 2.6 years [mean ± SEM]; body mass index, 29.6 ± 1.7 kg/m2; hemoglobin A1c, 7.0% ± 0.1%) received 1 of 3 combinations after an overnight fast in a randomized crossover design: metformin 1 g orally plus subcutaneous injection saline (Metformin), GLP-1 (1.5 nmol/kg body weight subcutaneously) plus placebo tablet (GLP-1), or metformin 1 g plus GLP-1(Metformin + GLP-1). At 15 minutes, glucose was raised to 15 mmol/L by rapid intravenous infusion of glucose; and responses were assessed over the next 3 hours. This stimulus does not activate the enteroinsular axis and secretion of endogenous GLP-1, enabling the effect of exogenously administered GLP-1 to be examined. Mean area under curve (AUC) 0-180 minutes plasma glucose responses were lowest after Metformin + GLP-1 (mean ± SEM, 1629 ± 90 mmol/[L min]) compared with GLP-1 (1885 ± 86 mmol/[L min], P < .002) and Metformin (2045 ± 115 mmol/[L min], P < .001). Mean AUC serum insulin responses were similar after either Metformin + GLP-1 (5426 ± 498 mU/[L min]) or GLP-1 (5655 ± 854 mU/[L min]) treatment, and both were higher than Metformin (3521 ± 410 mU/[L min]; P < .001 and P < .05, respectively). Mean AUC for plasma DPP-4 activity was lower after Metformin + GLP-1 (1505 ± 2 μmol/[mL min], P < .001) and Metformin (1508 ± 2 μmol/[mL min], P < .002) compared with GLP-1 (1587 ± 3 μmol/[mL min]). Mean AUC measures for plasma active GLP-1 concentrations were higher after Metformin + GLP-1 (820 × 104 ± 51 × 104 pmol/[L min]) compared with GLP-1 (484 × 104 ± 31 × 104 pmol/[L min], P < .001) and Metformin (419 × 104 ± 34 × 104 pmol/[L min], P < .001), respectively. In patients with type 2 diabetes mellitus, metformin inhibits DPP-4 activity and thus increases active GLP-1 concentrations after subcutaneous injection. In combination with GLP-1, metformin significantly lowers plasma glucose concentrations in type 2 diabetes mellitus subjects compared with GLP-1 alone, whereas insulin responses were similar. Metformin enhances serum concentrations of injected active GLP-1(7-36)amide, and the combination results in added glucose-lowering potency.