THE USE OF N-(7-NITROBENZ-2-OXA-1,3-DIAZOLE-4-YL)-LABELED LIPIDS IN DETERMINING TRANSMEMBRANE LIPID DISTRIBUTION

Transbilayer lipid distribution of small unilamellar vesicles (SUVs) a nd large unilamellar vesicles (LUVs) was measured using P-31-nuclear m agnetic resonance (NMR) spectroscopy, chemical modification with 2,4,6 -trinitrobenzene sulfonic acid (TNBS) and dithionite reduction of N-(7 -nitrobenz-2-oxa-1,3-diazole-4-yl)-labeled lipid (NBD-lipid). The dith ionite assay was the most reproducible of the three assays, with 1.2% error for SUVs and 3.9% error for LUVs. The dithionite assay also agre ed best with theoretical inner:outer leaflet ratios, based on vesicle diameters determined by electron microscopy (Thomas et al. (1989) Bioc hem. Biophys. Acta 978, 85-90). Dithionite assay measurements were wit hin 2.7% of theoretical ratios for SUVs and 2.3% for LUVs, while the N MR assay for SUVs was 14% lower than theoretical ratios and 23% lower for LUVs. The accuracy of NBD-lipids as markers for total transbilayer lipid was investigated. NBD-labeled phosphatidylserine, phosphatidylc holine and phosphatidylglycerol were accurate markers for total transb ilayer lipid distribution, as their distributions were in close agreem ent with theoretical ratios. However, NBD-labeled phosphatidylethanola mine displayed a slight preference for the inner leaflet at low mole f ractions of phosphatidylethanolamine, while native phosphatidylethanol amine showed a preference for the outer leaflet at the same concentrat ion. NBD-labeled phosphatidic acid also showed a slight preference for the inner leaflet. We conclude that although dithionite-based assessm ent of NBD-labeled lipids across membrane bilayers can be a powerful a nalytical tool, caution must be used in the interpretation of results.