The existence of functionally distinct intracellular Ca(2+) stores has been proposed in some types of smooth muscle. In this study, we sought to examine Ca(2+) stores in the gallbladder by measuring intracellular Ca(2+) concentration ([Ca(2+)](i)) in fura 2-loaded isolated myocytes, membrane potential in intact smooth muscle, and isometric contractions in whole mount preparations. Exposure of isolated myocytes to 10 nM CCK caused a transient elevation in [Ca(2+)](i) that persisted in Ca(2+)-free medium and was inhibited by 2-aminoethoxydiphenylborane (2-APB). Application of caffeine induced a rapid spike-like elevation in [Ca(2+)](i) that was insensitive to 2-APB but was abolished by pretreatment with 10 muM ryanodine. These data support the idea that both inositol trisphosphate (IP(3)) receptors (IP(3)R) and ryanodine receptors (RyR) are present in this tissue. When caffeine was applied in Ca(2+)-free solution, the [Ca(2+)](i) transients decreased as the interval between Ca(2+) removal and caffeine application was increased, indicating a possible leakage of Ca(2+) in these stores. The refilling of caffeine-sensitive stores involved sarcoendoplasmic reticulum Ca(2+)-ATPase activation, similar to IP(3)-sensitive stores. The moderate Ca(2+) elevation caused by CCK was associated with a gallbladder contraction, but caffeine or ryanodine failed to induce gallbladder contraction. Nevertheless, caffeine caused a concentration-dependent relaxation in gallbladder strips either under resting tone conditions or precontracted with 1 muM CCK. Taken together, these results suggest that, in gallbladder smooth muscle, multiple pharmacologically distinct Ca(2+) pools do not exist, but IP(3)R and RyR must be spatially separated because Ca(2+) release via these pathways leads to opposite responses.