Desferrioxamine and membrane oxidation: radical scavenger or iron chelator?

meter affecting the efficiency of the whole process is the chemical structure of the drug molecule. The saturated or unsaturated character o f PL acyl chains seems t o play a secondary role. T h e reason for this low incidence could he thc high content of cholesterol in the DRV preparations. It is well known that this molecule modulates the fluidity of the bilayers, having a dual behaviour, depending on the initial ordered state of the PL ('gel' o r 'fluid'). In the present case, as the starting phospholipids have extreme packing characteristics it is logical to assume that cholesterol fluidizes the hydrogenated PL bilayers and introduces order in the natural phosphatidylcholine bilayers, thus rendering both lipid mixtures in a similar physicochemical state. T h e influence of chemical nature of the drug in the encapsulation efficiency can be explained by differences in hydrophobicity of both molecules. Taking into account the ph' values o f gentamicin and morphine, at pH 7.4. morphine is partly nonprotonatcd and is therefore more hydrophobic than gentamicin. This apparently iinonialous behaviour has been reported for two states of the same molcculc pilocarpine free base/pilocarpine hydrochloride [ 4 I and codeine free base/codeine hydrochloride 12 I. One would expect ;I direct relationship between hydrophobicity and encapsulation efficiency, but on the contrary it seems that extrenic values o f hydrophobicity o r hydrophilicity give better yields. T h e same trends wcre observed in the stability studies. T h e amount o f free morphine released from liposomes rcprcsented. alter 5 weeks of storage at 4"C, Y S % ) and YO'%) o f the total drug initially encapsulated in liposomes formed by saturated and unsaturated lecithins. respectively. These values were 30(!h and 1 O'%). respectively, when the encapsulated drug WIS gentamicin. All these results show the need to optimize stitdies in order to find the best working conditions for each drug.