We have investigated the hypothesis that active anion transport drives fluid secretion by the cystic epithelium in autosomal dominant polycystic kidney disease (ADPKD). We prepared monolayers of a primary culture derived from cystic tissue removed from ADPKD patients. The monolayers were grown on permeant supports, and fluid secretion was initiated by forskolin. The results were compared with those obtained with monolayers of Madin-Darby canine kidney (MDCK) cells, known to secrete Cl-. In the absence of the agonist, ADPKD monolayers absorbed fluid (0.20 +/- 0.02 mu l . cm surface area(-2). h(-1)). Forskolin reversed this to secretion (0.60 +/- 0.03 mu l . cm(-2). h(-1)). Control MDCK monolayers did not transport fluid in either direction, but forskolin induced secretion (0.48 +/- 0.03 mu l . cm(-2). h(-1)). The electrical properties of the monolayers were monitored in Ussing chambers. Forskolin increased the transepithelial potential difference (V-te) of ADPKD monolayers (-0.9 +/- 0.1 to -1.1 +/- 0.1 mV) and the short-circuit current (I-sc) (6.6 +/- 0.7 to 9.2 +/- 0.8 mu A/cm(2)). The transepithelial resistance (R(te)) fell (156 +/- 9 to 138 +/- 10 Omega . cm(2)). Similar results were obtained with MDCK monolayers. The polarity of V-te and the direction of the I-sc are compatible with the hypothesis that active secretion of anion drives fluid secretion. Basolateral application of the Na-K-2Cl cotransporter, bumetanide, reduced forskolin-stimulated fluid secretion by ADPKD monolayers (0.56 +/- 0.05 to 0.28 +/- 0.03), depolarized V-te, and inhibited I-sc without affecting R(te). Apical application of the Cl- channel blocker, diphenylamine-2-carboxylate, also inhibited fluid secretion by ADPKD monolayers (0.65 +/- 0.03 to 0.27 +/- 0.02 mu l . cm(-2). h(-1)). It also depolarized V-te and decreased I-sc. The two inhibitors exerted similar effects on MDCK monolayers. These results substantiate the hypothesis that active transepithelial transport of anion may drive fluid secretion into ADPKD cysts.