Abstract

Six tritium-labeled cardiac glycosides (ouabain, dihydro-ouabain, digoxin, digitoxin, convallatoxol and pro-scillaridin, each in a concentration of 10^-7M) were perfused through guinea-pig hearts for 64 min, followed by 8 min of washout. The resultant cardiac uptake of the glycosides varied more than 100-fold, dihydro-ouabain showing the lowest tissue content and proscillaridin the highest. In heart homogenates, marked differences were found in the supernatant/pellet ratio with the various glycosides. Upon fractionation, in all instances the highest concentration of each glycoside was found in the microsomal fraction. When comparing glycosides, however, approximately the same 100-fold range of concentrations was found in the microsomal fraction as was found with the whole tissue uptake. Freshly isolated beef heart sarcoplasmic reticulum fragments (SRF) were also studied for their ability to take up the same six cardiac glycosides. Two buffer systems were employed, a phosphate and a Tris system. The glycoside uptake by SRF was somewhat greater in the phosphate than in the Tris system. The glycoside binding paralleled that demonstrated by the microsomal fraction of guinea pig except for dihydro-ouabain, which showed very low uptake by the whole heart tissue, but uptake equivalent to ouabain by isolated SRF. Marked differences were found in the potassium inhibition of SRF glycoside accumulation, and only slight dependence of glycoside binding on adenosine triphosphate was found. To explain these results, a model has been proposed which includes a cardiac membrane transport system for digitalis and two different SRF binding sites. The poor biologic activity of dihydro-ouabain can be explained by its inability to be transported. Typical glycosides, such as ouabain, digoxin and convallatoxol, bind predominantly to a potassium-sensitive site, while the binding of digitoxin and proscillaridin is mostly potassium-insensitive.