Studies on the preferential incorporation of [3H]glycerol over [32P]phosphate into major phospholipids of human platelets

It is well known that platelets readily incorporate radioactive glycerol, but not radioactive phosphate into phosphatidylcholine (PC) and phosphatidylethanolamine (PE) in vitro, thus not in accordance with de novo synthesis according to the Kennedy pathway. In attempts to understand the reason for the discrepancy, gel-filtered platelets were incubated simultaneously with 32P]Pi and 3H]glycerol, and the specific and relative radioactivities of products and intermediates were determined. Both precursors were incorporated into phosphatidylinositol (PI) with a 32P/3H ratio similar to that in glycerol 3-phosphate (in accordance with the Kennedy pathway). However, PC and PE obtained a much lower ratio. The specific 32P radioactivity in phosphorylcholine was similar to that of the gamma-phosphoryl of ATP and 650-times higher than that of PC. The specific 32P radioactivity of phosphorylethanolamine was 20-times less than that of phosphorylcholine. Both mass and 32P labelling of CDP-choline were below the detection limits. It is concluded that the incorporation of 32P]Pi into PC via phosphorylcholine is insignificant while the preferential incorporation of 3H]glycerol could be explained by exchange of diacyl3H]glycerol in the reversible choline phosphotransferase (CDP-choline: 1,2-diacylglycerol cholinephosphotransferase) reaction. The same mechanism would explain the preferential incorporation of 3H over 32P into PE, although dilution of 32P at the phosphorylethanolamine stage would account for part of the feeble 32P incorporation. Although other mechanisms are also possible, our results clearly show that the appearance of 3H]glycerol in PC and PE is not a reliable method of monitoring de novo synthesis of these phospholipids.