Diabetes-induced endothelial dysfunction is a critical initiating factor in the development of cardiovascular complications. Treatment with relaxin improves tumour necrosis factor α-induced endothelial dysfunction by enhancing endothelial nitric oxide synthase (eNOS) activity and restoring superoxide dismutase 1 protein in rat aortic rings ex vivo. It is, therefore, possible that relaxin treatment could alleviate endothelial dysfunction in diabetes. This study aimed to test the hypothesis that serelaxin (recombinant human relaxin-2) prevents high glucose-induced vascular dysfunction in the mouse aorta. Abdominal aortae were isolated from C57BL/6 male mice and incubated in M199 media for 3 days with either normal glucose (5.5 mM) or high glucose (30 mM), and co-incubated with placebo (20 mM sodium acetate) or 10 nM serelaxin at 37 °C in 5% CO2. Vascular function was analysed using wire-myography. High glucose significantly reduced the sensitivity to the endothelium-dependent agonist, acetylcholine (ACh) (pEC50; normal glucose = 7.66 ± 0.10 vs high glucose = 7.29 ± 0.10, n = 11–12, P < 0.05) and the contraction induced by NOS inhibitor, L-NAME (200 μM) (normal glucose = 59.9 ± 8.3% vs high glucose = 38.7 ± 4.3%, n = 6, P < 0.05), but had no effect on the endothelium-independent agonist, sodium nitroprusside (SNP)-mediated relaxation. Treatment with serelaxin restored endothelial function (pEC50; 7.83 ± 0.11, n = 11) but not NO availability. The presence of the cyclooxygenase (COX) inhibitor, indomethacin (1 μM) (pEC50; control = 7.29 ± 0.10 vs indo = 7.74 ± 0.18, n = 6–12, P < 0.05) and a superoxide dismutase mimetic, tempol (10 μM) (pEC50; control = 7.29 ± 0.10 vs tempol = 7.82 ± 0.05, n = 6–12, P < 0.01) significantly improved sensitivity to ACh in high glucose treated aortae, but had no effect in serelaxin treated aortae. This suggests that high glucose incubation alters the superoxide and COX-sensitive pathway, which was normalized by co-incubation with serelaxin. Neither high glucose incubation nor serelaxin treatment had an effect on cyclooxygenase 1 and 2 (Ptgs1, Ptgs2), prostacyclin synthase (PTGIS) and receptor (Ptgir) as well as thromboxane A2 receptor (Tbxa2r) mRNA expression. Importantly, production of prostacyclin was significantly (P < 0.05) attenuated in high glucose treated aortae, which was prevented by serelaxin treatment. Our data show that serelaxin treatment for 3 days restores high glucose-induced endothelial dysfunction by ameliorating vasodilator prostacyclin production and possibly through the reduction of superoxide in the mouse aorta.