We have previously demonstrated that significant luminal acidification in the K-replete rabbit inner stripe of the outer medullary collecting duct (OMCD(i)) occurs via a renal H-K-adenosinetriphosphatase (H-K-ATPase) sensitive to several of the gastric H-K-ATPase inhibitors. To investigate further the mechanism of K-dependent luminal acidification in K-replete OMCD(i), we examined the effects of luminal K removal, luminal addition of Ba in the presence and absence of luminal 5.0 nM bafilomycin A(1) (BAF), and basolateral addition of Ba on net bicarbonate flux (J(tCO2), pmol . mm(-1). min(-1)) and transepithelial voltage (V-T, mV). Removal of K from the perfusate inhibited J(tCO2) by 74% (13.4 +/- 4.0 for control, 3.5 +/- 1.4 pmol . mm(-1). min(-1) for experimental, P < 0.05) and was statistically equivalent to the degree of inhibition previously observed under identical experimental conditions by either 10 mu M Sch-28080 or 10 mu M A-80915A. Approximately 50% inhibition of J(tCO2) was observed following luminal application of 2.0 mM Ba2+, and the degree of inhibition was statistically equivalent regardless of whether BAF was present (12.2 +/- 2.7 for control, 6.0 +/- 1.4 pmol . mm(-1). min(-1) for 2.0 mM Ba2+, P < 0.05; 9.6 +/- 1.2 for control with 5 nM BAF, 5.7 +/- 1.3 pmol . mm(-1). min(-1) for 2.0 mM Ba2+ with 5 nM BAF, P < 0.05). Additionally, increasing the luminal concentration of Ba2+ from 2.0 to 4.0 mM resulted in no further inhibition of J(tCO2) (8.5 +/- 1.7 for control, 3.9 +/- 1.3 pmol . mm(-1). min(-1) for 4.0 mM Ba2+, P = 0.05). Neither 2.0 nor 4.0 mM basolateral Ba2+, however, resulted in any significant inhibition of J(tCO2) [8.9 +/- 2.1 for control, 5.5 +/- 2.2 pmol . mm(-1). min(-1) for 2.0 mM Ba2+, P = not significant (NS); 8.0 +/- 2.4 for control, 5.5 +/- 1.7 pmol . mm(-1). min(-1) for 4.0 mM Ba2+, P = NS]. A consistent and immediate but small VT response to 2 mM luminal Ba2+ (3.0 +/- 0.9 for control, 1.2 +/- 0.8 mV for experimental, P < 0.05) was observed, but no significant effect on VT was observed with basolaterally applied Ba. Under the conditions of these studies, we conclude that luminal reduction of K concentration profoundly inhibits J(tCO2), further supporting the role of H-K-ATPase in luminal acidification in the K-replete OMCD(i). Moreover, a component of J(tCO2) attributable to an H-K-ATPase is sensitive to luminal Ba but not sensitive to basolateral Ba to a statistically significant extent. This observation suggests that an apically oriented K-exit mechanism, possibly a K channel, is primarily responsible for H-K-ATPase-dependent K exit in the K-replete OMCD(i).