Uterine vasculature is less responsive than systemic vasculature to angiotensin II (ANG II)-induced vasoconstriction. We hypothesized that pregnancy augments basal and ANG II-stimulated endothelial prostacyclin (PGI2) and/or nitric oxide (NO) production, which locally increase vascular smooth muscle (VSM) adenosine 3',5'-cyclic monophosphate (cAMP) and guanosine 3',5'-cyclic monophosphate (cGMP), respectively. Uterine (UA) and systemic arteries (SA) from pregnant (P) and nonpregnant (NP) sheep were incubated with isobutylmethylxanthine. Basal PGI2, cAMP, and cGMP production was 2.4-, 1.6-, and 5.9-fold greater (P < 0.01) in UA from P vs. NP sheep; endothelium removal lowered (P < 0.05) values 69, 44, and 88%. Basal SA PGI2 and cAMP, but not cGMP, also were elevated by pregnancy. Indomethacin (Indo; 100 microM) decreased PGI2 and cAMP, but not cGMP production; N omega-nitro-L-arginine methyl ester (L-NAME; 10 microM) and methylene blue (MB, 10 microM) only decreased cGMP. Basal UA, but not SA, NO synthase activity (conversion of [3H]arginine to [3H]citrulline), was 1.8-fold higher in pregnancy and decreased (P < 0.01) after endothelium removal and with L-NAME. ANG II (50 nM) increased PGI2 (86%) and cAMP (56%) production only in UA from P sheep (P < 0.05); this was abolished by endothelium removal or Indo. ANG II also increased (P < 0.01) cGMP production by UA from both groups but only by SA from P ewes; this was absent in denuded, L-NAME-, or MB-treated vessels. Stimulation of VSM cGMP production with sodium nitroprusside (50 microM) was inhibited by MB, but not L-NAME or endothelium removal. In pregnancy, endothelial PGI2 and NO production are enhanced and may contribute to attenuated ANG II vasoconstriction via VSM cAMP and cGMP.