It was established in animal models that vasoactive intestinal peptide (VIP) significantly increases blood flow in the salivary glands, potentiates vasodilatatory response to acetylcholine, and stimulates salivation. We hypothesized that endothelium can modulate vascular response of the human submandibular artery to VIP, therefore the aim of the study was to evaluate the contributions of endothelial diffusible vasodilatatory factors on VIP-induced relaxation. This was achieved using following substances: the nitric oxide synthase inhibitor N(G)-nitro-L-arginine (L-NOARG), the potassium channel blocker 4-aminopyridine (4-AP), and the cyclo-oxygenase inhibitor indomethacin. Seven patients (ages 30-45 yrs; four females and three males), undergoing surgery for salivary duct calculus excision, were enrolled in this study. During surgery, submandibular arteries were ligated and dissected into segments (2 segments from each patient). Segments were immediately immersed in ice-cold Krebs-Ringer bicarbonate solution and transported to in vitro experimental set-up within 20 minutes. All experimental procedures were conducted after obtaining written informed patient consent and approval of the University of Nis Faculty of Medicine Ethical Committee and were performed in accordance with the Declaration of Helsinki. VIP (3 x 10-9 – 3 x 10-7 mol/L) induced a concentration-dependent relaxation in endothelium-containing arterial segments (pEC50 = 7.62 ± 0.01; maximal relaxation = 80.4 ± 4.2%, n = 7), but not in endothelium-denuded segments of the human submandibular artery. This VIP-induced relaxant response was significantly reduced after pretreatment with L-NOARG (10-5 mol/L) (pEC50 =7.50 ± 0.10 vs. 7.30 ± 0.09; maximal relaxation = 79.3 ± 6.1% vs. 39.3 ± 3.8%, n = 5, p < 0.05) and almost completely inhibited after pretreatment with 4-AP (10-5 mol/L) (pEC50 = 7.67 ± 0.03 vs. 7.58 ± 0.13; maximal relaxation = 84.5 ± 1.5% vs. 18.0 ± 2.7%, n = 5, p < 0.01). On the other hand, pretreatment of the arterial segments with indomethacin (10-5 mol/L) failed to induce any effect (pEC50 = 7.56 ± 0.14 vs. 7.66 ± 0.05; maximal relaxation = 78.0 ± 3.1% vs. 85.5 ± 2.6%, n = 4, p > 0.05). Since vascular effects of VIP are usually mediated through cAMP signaling pathway, they were tested in the presence of an adenylate cyclase inhibitor, forskolin (10-5 mol/L). Our results shown that relaxant effect of VIP was not significantly changed. We concluded that VIP has direct action on the human submandibular artery inducing endothelium-dependent relaxation presumably by releasing the endothelium-derived hyperpolarizing factor and NO.