1992 SYNTEX AWARD LECTURE - MODEL BIOMEMBRANES - QUANTITATIVE STUDIES OF PEROXIDATION, ANTIOXIDANT ACTION, PARTITIONING, AND OXIDATIVE STRESS

This review outlines a decade of research that employs quantitative kinetic methods of autooxidation to phospholipids aggregated into membranes. The classical rate law for autoxidation: d[O2]/dt = k(p)/(2k(t))1/2 [R-H]R(i)1/2 was found to apply to heterogeneous systems of bilayers and micelles, where k(p) and 2k(t) are the rate constants for chain propagation and termination, respectively, and R(i), the rate of chain initiation, is controlled by thermal initiators. The oxidizability of a typical lipid chain, linoleate 18:2, at 30-degrees-C was similar (0.02-0.04 M-1/2 s-1/2) in different media (solution, micelles, bilayers) and for different procedures using water-soluble or lipid-soluble initiators and inhibitors. A reduction in the absolute rate constant for termination, 2(k)t by a hundredfold in bilayers of dilinoleoylphosphatidyl choline (DLPC) compared to that in tert-butyl alcohol solution, and a drop in the k(p) by a factor of five are interpreted in terms of diffusion of polarized peroxyl radicals from the hydrophobic bilayer phase to the aqueous surface, where peroxyls are strongly solvated by water. Such phenomena may also account for significant changes as observed by P-31 NMR spectra in bilayer lamellar structure accompanying extensive peroxidation. Analysis of the hydroperoxides formed by peroxidation of mixed bilayers of DLPC + DPPC (16:0) initiated by a water-soluble initiator, azobis(2-amidinopropane.HCl) (ABAP), showed a linear trend between the ratio of cis,trans to trans,trans geometrical isomeric hydroperoxides and [DPLC] consistent with a peroxidation mechanism proposed in homogeneous solution. The antioxidant activities, k(inh), of three classes of antioxidants: (a) polyalkyl-6-hydroxychromans (e.g., vitamin E), (b) polyalkyl-4-methoxyphenols, and (c) trialkylphenols, were measured in DLPC membranes. The results show an overall leveling and depression of k(inh) values in DLPC membranes in the series (a) (by several orders of magnitude), (b), and (c) compared to known values in solution in chlorobenzene. In aqueous bilayers it is proposed that k(inh) values are attenuated by hydrogen bonding by water at both the para ether oxygen and phenolic groups. Restricted diffusion (e.g., of alpha-tocopherol) may also reduce antioxidant activities in membranes. A synergistic effect between ascorbic acid and alpha-tocopherol was discovered under conditions of inhibited peroxidation of linoleate in SDS micelles. The natural peptide glutathione, GSH, however, acts as a co-antioxidant with vitamin E by trapping peroxyls in the aqueous phase. Solid cholesterol was found to partition directly into PC lipsomes by shaking, above or below the phase transition temperature, and membrane-bound cholesterol, unlike the solid, undergoes facile peroxidation. A water-soluble form of alpha-tocopherol complexed with bovine serum albumin (alpha-toc:BSA) is an effective antioxidant for autoxidations of linoleate in SDS micelles. In contrast, alpha-toc:BSA required a long equilibration time (e.g., 12 h) with liposomes (DLPC) before the alpha-tocopherol was transferred to the liposomes to provide effective antioxidant action. Carbon radicals from thermolysis under nitrogen of the lipidic azo initiator, 6,6'-azobis(6-cyanododecanoic acid), 6.6', in unilamellar liposomes were spin-trapped by phenyl tert-butyl nitrone (PBN) to give membrane-bound trapped radicals as observed by broad anisotropic ESR spectra. However, under oxygen, 6,6' in liposomes formed derived peroxyl radicals, which were trapped in the aqueous phase by water-soluble 4-trimethylammonium-PBN. The Total (peroxyl) Radical Trapping Antioxidant Parameter (TRAP) of a biological fluid, the number of peroxyl radicals trapped per volume of fluid under ''oxidative stress'' under known conditions of chain initiation (R(i)), was determined for blood plasmas of humans. A control group (n = 6) gave an average TRAP of 775 muM peroxyls trapped compared to athletes (n = 5) where the average TRAP of 1226 muM was 45% higher. Vigorous exercise was found to increase the uric acid contribution to TRAP.