We investigated the effect of human alpha-atrial natriuretic peptide (alpha-hANP) on osmotically stimulated arginine vasopressin (AVP) secretion and thirst appreciation. Seven normal male volunteers were studied on two occasions: synthetic alpha-hANP-(99-126) (2 pmol.kg-1.min-1) or control was infused intravenously for 30 min before and for the first 60 min of a 120-min hypertonic saline (855 mmol/l) infusion (0.06 ml.kg-1.min-1). Plasma ANP did not alter significantly during infusion of control and hypertonic saline (C + HS) but rose to steady-state concentrations of 17.4 +/- 3.2 pmol/l during infusion of ANP and hypertonic saline (ANP + HS). Plasma osmolality increased on both study days [ANP + HS: 284.4 +/- 0.6 to 299.7 +/- 1.1 mosmol/kgH2O (P < 0.01), C + HS: 283.6 +/- 1.2 to 299.1 +/- 1.6 mosmol/kgH2O (P < 0.01)], as did plasma sodium [ANP + HS: 139.0 +/- 0.6 to 148.0 +/- 0.4 mmol/l (P < 0.01), C+HS: 137.6 +/- 0.75 to 145.8 +/- 0.7 mmol/l (P < 0.01)] and blood volume (ANP+HS: 7.7 +/- 0.6%, C+HS: 9.4 +/- 1.0%). The increase in plasma osmolality was accompanied by an increase in plasma AVP [ANP+HS: 1.4 +/- 0.3 to 8.3 +/- 1.2 pmol/l (P < 0.01), C+HS: 1.6 +/- 0.4 to 7.8 +/- 1.5 pmol/l (P < 0.01)]. The difference in thirst from the initial value at -30 min increased significantly over the time course of both studies (P < 0.01), but the effect of ANP compared with control infusion was to significantly decrease thirst at 30 and 60 min [30 min, ANP+HS: 0.3 +/- 0.3 cm, C+HS: 1.6 +/- 0.4 cm (P < 0.05); 60 min, ANP+HS: 1.3 +/- 0.7 cm, C+HS: 3.4 +/- 0.8 cm (P < 0.01)]. We conclude that physiological increases in plasma ANP concentrations do not inhibit osmotically stimulated AVP secretion but will blunt osmotically stimulated thirst appreciation in humans.