The obligatory involvement of a C 21 intermediate in the biosynthesis of cardenolides from cholesterol

Abstract

Previous studies have indicated that the biosynthesis of the three main cardenolides—digitoxigenin, gitoxigenin and digoxigenin—from preformed C 27 precursors could proceed by different pathways and that, in contrast to the previous assumptions of other workers, apparently neither pregnenolone nor progesterone are the sole or even the major biosynthetic intermediates in this sequence. Formally, two major biosynthetic routes may be envisaged from C 27 precursors—one involving “modified” C 21 intermediates arising from the cleavage of the cholesterol side-chain between C-20 and -22, and the other involving C 23 intermediates formed by cleavage between C-23 and C-24. Experiments designed to test the operation of these two routes have been carried out, and are described. The administration of doubly labeled 7- 3H-(1,7,15,22,26)- 14C-cholesterol to a Digitalis lanata plant led to the formation of radioactive digitoxigenin, gitoxigenin and digoxigenin. The 3H/ 14C ratio of these products, relative to the precursor, indicated that only three positions of these cardenolides were labeled with carbon-14. It was shown by chemical degradation that no 14C was present in the lactone ring. Hence, it may be concluded that the carbons 22 and 23 of cardenolides do not originate from the exogenous cholesterol. Similarly, when digitoxigenin was biosynthesized from 4R-4- 3H-2- 14C-mevalonic acid, the 3H/ 14C ratio confirmed that the carbons 22 and 23 of the cardenolides do not originate from MVA. The 3H/ 14C ratio of tigogenin isolated from the same experiment indicated that only two positions, rather than three, were labeled with tritium. The absence of tritium at C-20 was established. The loss of this tritium atom could have occurred either during the biosynthesis or in the isolation processes.