Singlet oxygen (O-1(2)) can react with cholesterol (Ch) to give three possible ene-addition hydroperoxides: 3-beta-hydroxy-5-alpha-cholest-6-ene-5-hydroperoxide (5-alpha-OOH), 3-beta-hydroxycholest-4-ene-6-alpha-hydroperoxide (6-alpha-OOH), and 3-beta-hydroxycholest-4-ene-6-beta-hydroperoxide (6-beta-OOH). The rates of dye-sensitized photogeneration and also the fates of 5-alpha-OOH and 6-beta-OOH in membrane bilayers have been studied and compared. Irradiation of unilamellar [C-14]Ch/phospholipid vesicles in the presence of aluminum phthalocyanine tetrasulfonate or merocyanine 540 resulted in formation of 5-alpha-OOH and 6-beta-OOH, as determined by high performance liquid chromatography with radiochemical or electrochemical detection. The initial rate of 6-beta-OOH formation was 30-35% that of 5-alpha-OOH in a variety of liposomal systems. However, after a lag, 5-alpha-OOH invariably decayed via allylic rearrangement to 7-alpha-OOH (also known to be a free radical product), whereas 6-beta-OOH accumulated in unabated fashion until Ch depletion became limiting, Photooxidation of Ch in an isolated natural membrane (erythrocyte ghost) or in L1210 leukemia cells gave similar results. When the reaction was carried out in pyridine or methanol, the rate of 6-beta-OOH formation relative to 5-alpha-OOH was reduced by approximately half, with essentially no isomerization of the latter to 7-alpha-OOH. These results suggest that (i) environmental factors in a lipid bilayer somehow make photogeneration of 6-beta-OOH more favorable than in homogeneous solution; and (ii) due to its relative stability, 6-beta-OOH may be a more reliable probe of O-1(2) intermediacy than 5-alpha-OOH.