The objectives of this study were (i) to determine whether the reduced absorption of captopril from the colon of humans also occurs in rats and (ii), after confirmation of the relevance of a new rat model, to evaluate the intestinal absorption of captopril and several of its analogs. A model was developed and validated in which specific sites within the GI tract of rats were surgically implanted with a cannula such that animals could be dosed while conscious and unrestrained. The absorption of captopril after administration into the lower GI tract of rats was significantly reduced relative to the upper GI tract, which was consistent with results reported previously in humans. In rats, the absorption of the S-benzoyl thioester prodrug of captopril (SQ-25868) from the lower GI tract was substantially greater than that of captopril. However, the absorption of the S-benzoyl thioester prodrug of 4-phenyl thio-captopril (SQ-26991) from the lower GI tract was only marginally better than that of captopril. In additional studies in dogs, a 12h controlled-release formulation of SQ-25868 provided sustained blood levels of captopril while maintaining acceptable bioavailability (> 80%). Two approaches were tried, without success, to stabilize captopril in vivo: (i) complexation with zinc (SQ-26284) and (ii) use of ascorbic-acid-buffered (pH 3.5) vehicle. The zinc complex might have failed because it has very low solubility, whereas the pH-3.5-buffered vehicle was quickly neutralized within the colonic lumen in rats, and did not stabilize captopril against oxidation. Rapid neutralization might explain why the colonic bioavailability of captopril was not substantially increased when this pH-3.5-buffered vehicle was tried in humans.