Regulatory volume decrease (RVD) following hypo-osmotic stimulation was studied using videometric methods in isolated proximal renal tubules from trout (Salmo trutta). The relative tubule diameter increased by 132.0+/-4.8 % (maximum swelling within 1 min at 15 and 25 degrees C and within 4 min at 10 degrees C) following a change from iso-osmotic (290 mosmol kg-1) to hypo-osmotic (160 mosmol kg-1) Ringer's solution. The tubule diameter subsequently decreased to approximately one-quarter of the maximal value. Ouabain (1 mmol l-1) reduced cell swelling and inhibited the RVD response by 28.0+/-10.5 %. Furthermore, increasing the bath K+ concentration by 30 mmol l-1 inhibited RVD by 76.5+/-3.6 %. The K+ channel blocker quinine, but not Ba2+ (1 and 2 mmol l-1), significantly decreased the RVD response (by 25.0+/-5.4 and 72.3+/-5.1 % at 0.1 and 0.5 mmol l-1, respectively). Similarly, increasing the Cl- concentration in the bath from 47 to 102 mmol l-1 induced a significant reduction (45. 2+/-7.9 %) in RVD. The RVD response was also markedly reduced (by 54. 7+/-5.3 %) by the Cl- channel blocker indacrinone (MK-196; 0.5 mmol l-1), but only marginally by 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB; 1, 5, 8 and 10 micromol l-1). Addition of the K+/Cl- symport inhibitor furosemide (0.1 mmol l-1) resulted in a 39.8+/-3.9 % inhibition of RVD. This inhibition could be completely overcome by simultaneous administration of 1 micromol l-1 tributyltin (anion exchanger) and furosemide. Chelation of either extracellular (1 mmol l-1 EGTA) or both extra- and intracellular Ca2+ (1 mmol l-1 EGTA, 10 micromol l-1 A23187) had no effect on this RVD process. Furthermore, as measured using the fluorescent dye Fura-2/AM, there was no increase in the intracellular free Ca2+ concentration upon hypo-osmotic stimulation. Administration of the 5-lipoxygenase antagonist ETH 615-139 (20 micromol l-1), however, induced a 60 % inhibition of RVD. Simultaneous addition of ETH-615 and either the K+ ionophore gramicidin (0.5 mmol l-1) or the anion exchanger tributyltin (1 micromol l-1) could not reverse the ETH 615-139 inhibition. Finally, administration of the cycloxygenase inhibitor indomethacin had only a small, but significant, effect on RVD. We conclude that RVD following hypo-osmotic swelling is in these cells a temperature- and ouabain-sensitive process that appears to be the result of K+ efflux through quinine-sensitive, Ba2+-insensitive K+ channels and Cl- efflux through an MK-196- and furosemide-sensitive Cl- conductance that is relatively unaffected by NPPB. This KCl efflux seems to be regulated by eicosanoids produced by the 5-lipoxygenase. Arachidonic acid metabolites from the cycloxygenase pathway are not involved in this process. Similarly, neither extra- nor intracellular Ca2+ appears to be important for the signalling of RVD.