A reliable procedure for comparison of antioxidants in rat brain homogenates

Abstract

Lipid peroxidation is a major consequence of oxidative stress and an important cause of neuronal damage in ischaemic injuries and neurodegenerative disorders such as Parkinson’s disease. Recent research has focused on the development of antioxidant drugs which may delay or minimize neurodegeneration. Rapid and reliable assays are therefore necessary in order to evaluate novel antioxidant compounds. A widely adopted method for measurement of lipid peroxidation is the thiobarbituric acid reacting substances (TBARS) assay. Several variations of this method have appeared in the literature, some of which have been tested by us without success. We have therefore established a reliable procedure which takes into account the most important factors previously found to influence the TBARS method. Briefly, various concentrations of drug were added to rat brain homogenates (10% w/v in 20 mM Tris-HCl buffer, pH 7.4) and incubated at 37°C for 10 min before addition of ammonium ferric sulphate (100 or 1000 μM) and a further incubation at 37°C for 30 min. Proteins were then precipitated with 8.1% sodium dodecyl sulphate, the reaction stopped with 20% acetic acid, and the samples were then centrifuged for 15 min. Aliquots of supernatant were added to an equal volume of thiobarbituric acid (0.8%), samples were heated at 95°C for 30 min, and then cooled on ice before reading at 532 nm. The present adaptation represents a simple and highly reproducible assay which does not require difficult extraction procedures with hazardous chemicals and results in a stable chromagen. The method has been evaluated using a number of structurally distinct antioxidants and iron chelators. IC₅₀ values (μM) for percentage inhibition of TBARS formation were as follows: desferroxamine (1.1), U83836E (1.7), butylated hydroxytoluene (13), U74500A (20), LY231617 (22), idebenone (89), and Trolox (110). This order of potency was comparable to that found with a commercially available, but expensive kit designed to specifically measure malondialdehyde (Spearman’s rank correlation coefficient, p < 0.01).

Introduction

Lipid peroxidation is a major consequence of oxidative stress and is generally thought to result from an interaction of reactive oxygen species with polyunsaturated lipids in cell membranes. There are subsequent changes in the structure, function, and permeability of the membrane, which ultimately result in cellular death (Halliwell and Gutteridge, 1985). Lipid peroxidation is an important cause of neuronal damage, for example, in ischaemic injuries, neurotrauma, and neurodegenerative disorders such as Parkinson’s disease (Coyle and Puttfarcken, 1993). Rapid and reliable assays are therefore necessary to evaluate novel antioxidant compounds, capable of inhibiting lipid peroxidation, which might be suitable for therapeutic intervention in these neurological conditions.

A widely adopted and sensitive method for measurement of lipid peroxidation is the thiobarbituric acid reacting substances (TBARS) assay. The assay is based on the reaction of breakdown products of lipid peroxidation such as malondialdehyde (MDA) with thiobarbituric acid (TBA) to produce a pink chromogen when heated at low pH (Esterbauer and Cheeseman, 1990).

The TBARS assay was originally designed to measure the endogenous lipid peroxidation of the test sample, however, baseline lipid peroxidation in brain homogenates from normal animals is minimal and hence is relatively unsuitable for manipulative purposes. While the exact reaction mechanism remains controversial, lipid peroxidation can be stimulated by the addition of iron ions Halliwell 1978, Braughler et al 1986, Auroma et al 1989. Iron has long been thought to catalyse the formation of the highly reactive hydroxyl (^·OH) radical via the Fenton reaction with the ^·OH causing lipid peroxidation by abstracting a hydrogen atom from membrane phospholipids initiating a chain reaction of lipid radical formation Halliwell and Gutteridge 1985, Halliwell and Gutteridge 1988. Iron is known to be involved in oxygen-mediated brain injury responses in vivo Halliwell 1992, Jellinger et al 1992, Olanow 1993. Normally, iron is bound to storage proteins such as transferrin and ferritin (Aisen and Listowsky, 1980), but under conditions of oxidative stress is released by reductants such as superoxide anion and peroxides Koster and Slee 1986, Puppo and Halliwell 1988. Iron released from ferritin can produce ^·OH (Biemond et al., 1986) and catalyse lipid peroxidation Koster and Slee 1986, Kukielka and Cederbaum 1996. Iron released when tissue is injured or through bleeding may be the triggering event in free radical-induced damage and lipid peroxidation in head injury, ischemia, and reperfusion injury Siesjo et al 1989, Willmore and Triggs 1991. Abnormally high concentrations of iron have been found in the substantia nigra of Parkinsonian brains Ben-Shachar et al 1991, Halliwell 1992 and the capacity of brain tissue to bind iron is diminished with age (Barkai et al., 1991).

Many variations of the TBARS method have been reported in the literature since it was first described. Some of these methods have been tested by us without success. Hence, we have adopted aspects of several different methods and, with modifications, established our own procedure. While many criticisms of the TBARS assay have appeared in the literature (e.g., Janero, 1990), when used correctly and with the present modifications, it provides a simple and reliable method for comparison of the potencies of antioxidant compounds and is currently being used to evaluate novel free radical scavengers. In addition, our assay enables compounds which act as iron chelators to be distinguished from those with antioxidant activity. The present method has been evaluated using a range of structurally diverse antioxidants and iron chelators.