Plasma renin activity (PRA), urinary excretions of PGE2, 6-keto-PGF1 alpha (6KPGF), TXB2 and renal function were determined in healthy women both in normal potassium balance (N, n = 14) and in experimental potassium depletion (KD). KD was induced by natriuretic treatment--associated to replacement of net NaCl and water losses--in the presence of either normal (congruent to 50 mmol/d) or low (less than or equal to 10 mmol/d) dietary potassium intake. By using different depletive patterns, three groups with estimated cumulative potassium deficit (mean +/- SEM) of 124 +/- 38 (KD0, n = 8), 160 +/- 43 (KD1, n = 8) and 198 +/- 22 mmol (KD2, n = 6), respectively, were obtained. Renal function by the clearance (cl.) method and urinary prostanoid concentrations by the RIA method were estimated during hypotonic polyuria (oral water load) and subsequent moderate antidiuresis induced by a low-dose infusion of lysine-8-vasopressin. 1. In KD0 group the potassium depletive treatment was inefficacious in significantly reducing either the plasma potassium concentration (PK) or the urinary potassium excretion (UKV). The reductions of PK and UKV as well as the enhancement of PRA became significant in KD1 and KD2 groups. 2. The urinary prostanoid excretions were not significantly changed in the KD0 and KD1 groups while in the KD2 group they were reduced, mainly concerning the urinary 6KPGF excretion. 3. Furthermore in the KD2 group, with larger potassium depletion, some of the typical hypokalemic renal dysfunctions appeared. The data suggest that a pathophysiologically critical degree of potassium depletion is associated with an inhibited renal prostanoid synthesis as well as an increased renin secretion.