K+ secretion by the cortical collecting duct (CCD) is stimulated at high flow rates. Patch-clamp analysis has identified a small-conductance secretory K+ (SK) and a high-conductance Ca(2+)-activated K+ (maxi-K) channel in the apical membrane of the CCD. The SK channel, encoded by ROMK, is believed to mediate baseline K+ secretion. The role of the stretch- and Ca2+-activated maxi-K channel is still uncertain. The purpose of this study was to identify the K+ channel mediating flow-dependent K+ secretion in the CCD. Segments isolated from New Zealand White rabbits were microperfused in the absence and presence of luminal tetraethylammonium (TEA) or charybdotoxin, both inhibitors of maxi-K but not SK channels, or apamin, an inhibitor of small-conductance maxi-K+ channels. Net K+ secretion and Na+ absorption were measured at varying flow rates. In the absence of TEA, net K+ secretion increased from 8.3 +/- 1.0 to 23.4 +/- 4.7 pmol. min(-1). mm(-1) (P < 0.03) as the tubular flow rate was increased from 0.5 to 6 nl. min(-1). mm(-1). Flow stimulation of net K+ secretion was blocked by luminal TEA (8.2 +/- 1.2 vs. 9.9 +/- 2.7 pmol. min(-1). mm(-1) at 0.6 and 6 nl. min(-1). mm(-1) flow rates, respectively) or charybdotoxin (6.8 +/- 1.6 vs. 8.3 +/- 1.6 pmol. min(-1). mm(-1) at 1 and 4 nl. min(-1). mm(-1) flow rates, respectively) but not by apamin. These results suggest that flow-dependent K+ secretion is mediated by a maxi-K channel, whereas baseline K+ secretion occurs through a TEA- and charybdotoxin-insensitive SK (ROMK) channel.