A 2-Phase Liquid Scintillation Assay for Glycolipid Synthetases

Glycolipid synthetases can be assayed conveniently by incubating the lipid substrate with the radiosugar-labeled nucleotide in a small plastic scintillation vial. At the end of the incubation period, water and perchloric acid are added, then n-butanol, then a toluene-based scintillation cocktail. The radioactive lipid partitions into the scintillation fluid, leaving excess sugar nucleotide in the aqueous phase. Only a small fraction of the total radioactivity in the aqueous layer is detectable. This method is illustrated for ceramide:UDP-glucose glucosyltransferase. The approach should be applicable to other lipid synthetases that can be assayed with a radioactive liydrophilic substrate. In the many widely used radiometric enzyme assay techniques, it is necessary to separate the radioactive product of enzyme action from the radioactive substrate. In the case of lipids, where the labeled substrate may be nonlipoidal, solvent partitioning is commonly done, typically with chloroform/methanol/water. The chloroform layer must be evaporated to dryness (after multiple washings) before the lipid can be counted by liquid scintillation, since chloroform is a quenching agent. Potter (1) used a partition method to separate the labeled acetate formed by acetylcholinesterase, in which the so lvent to luene/ isoamyl a lcohol-was not a quencher and could then be added to a scintillation fluid directly. This approach was improved by Sankaran and Pogell (2), who simply incubated the assay mixture in a scintillation vial, partitioned the labeled product directly into a scintillation fluid, and counted the entire assembly in the normal way. Because the water in the lower phase absorbed most of the/3-radiation coming from the unused tritiated substrate, the observed background activity in the scintillation fluid was not too high. In the case of 14C, as opposed to 3H, there was enough penetration of the water layer to raise the background to an unpleasant level and the lower layer had to be removed. A few additional examples of the scintillation partitioning method have been published since then (3-7) but none have involved lipids, which ought to be particularly suited to the approach. In our initial attempts to use the approach, however, we found excessive differences between duplicate samples and variable drift of observed activities as a function of time. These were alleviated by centrifuging, which presumably brought down small amounts of the lower phase that were adhering to the walls of the vial in contact with scintillation fluid. Additional improvement was obtained by denaturing the proteins with perchloric acid. Another problem was the size of the boiledenzyme blank, which resulted in part from radiation entering the scintillation fluid from the lower phase. This was reduced by lowering the specific activity (sp act) of the lower phase