In isolated rat hearts perfused at constant coronary flow and heart rate the amount of shortening in their major axis (L), the developed tension (T) and their time derivatives (L and T) were measured after different interventions. Relative changes of maximal velocities of contraction (+L, +T) and relaxation (-L, -T) were assessed by the ratio between both velocities (+L/-L, +T/-T). After the interventions, cAMP intracellular levels and cAMP dependent protein kinase activity were measured. The negative inotropic effect of verapamil (10(-7) M) and nifedipine (5 X 10(-7) M) was accompanied by a relatively greater decrease in maximal velocity of relaxation. Consequently the ratio +L/-L increased. Verapamil increased +L/-L from 1.27 +/- 0.07 to 1.57 +/- 0.08 (P less than 0.001). Nifedipine increased +L/-L from 1.25 +/- 0.05 to 1.77 +/- 0.12 (P less than 0.001). Whereas nifedipine increased intracellular cAMP levels from 0.403 +/- 0.036 pmol/mg wet weight to 0.534 +/- 0.047 (P less than 0.05), verapamil did not alter them. Neither verapamil nor nifedipine affected cAMP protein kinase activity. Increasing Ca2+ in the perfusate did not change the ratio +L/-L. A decrease in extracellular Ca2+, on the other hand, produced a greater decrease in -L than in +L, so that the ratio +L/-L increased from 1.27 +/- 0.05 to 1.49 +/- 0.10 (P less than 0.05). No changes were detected in cAMP levels or its protein kinase activity. Similar results were obtained when +T/-T was analyzed. To offset the negative inotropic effect caused by calcium antagonists, either increased extracellular Ca2+ or isoproterenol can be used.(ABSTRACT TRUNCATED AT 250 WORDS)