Interactions between cholesterol and lipids in bilayer membranes. Role of lipid headgroup and hydrocarbon chain-backbone linkage

We have employed four lipids in the present study, of which two are cationic and two bear phosphatidylcholine (PC) headgroups. Unlike dipalmitoylphosphatidylcholine, the other lipids employed herein do not have any ester linkage between the hydrocarbon chains and the respective lipid backbones. Small unilamellar vesicles formed from each of the PC and cationic lipids with or without varying amounts of cholesterol have been examined using the steady-state fluorescence anisotropy method as a function of temperature. The anisotropy data clearly indicate that the order in the lipid bilayer packing is strongly affected upon inclusion of cholesterol. This effect is similar irrespective of the electrostatic character of the lipid employed. The influence of cholesterol inclusion on multi-lamellar lipid dispersions has also been examined by 1H-nuclear magnetic resonance spectroscopy above the phase transition temperatures. With all the lipids, the line widths of (CH2)n protons of hydrocarbon chains in the NMR spectra respond to the addition of cholesterol to membranes. The influence on the bilayer widths of various lipids upon inclusion of cholesterol was determined from X-ray diffraction studies of the cast films of the lipid-cholesterol coaggregates in water. The effect of cholesterol on the efflux rates of entrapped carboxyfluorescein (CF) from the phospholipid vesicles was determined. Upon incremental incorporation of cholesterol into the phospholipid vesicles, the CF leakage rates were progressively reduced. Independent experiments measuring trans-membrane OH- ion permeation rates from cholesterol-doped cationic lipid vesicles using entrapped dye riboflavin also demonstrated that the addition of cholesterol into the cationic lipid vesicles reduced the leakage rates irrespective of lipid molecular structure. It was found that the cholesterol induced changes on the membrane properties such as lipid order, linewidth broadening, efflux rates, bilayer widths, etc., did not depend on the ability of the lipids to participate in the hydrogen bonding interactions with the 3β-OH of cholesterol. These findings emphasize the importance of hydrophobic interaction between lipid and cholesterol and demonstrate that it is not necessary to explain the observed cholesterol induced effects on the basis of the presence of hydrogen bonding between the 3β-OH of cholesterol and the lipid chain-backbone linkage region or headgroup region