Lipid Peroxidation

Lipids, whether in bulk solution or as components of biological membranes, can undergo chemical modifications and degradation by a series of free-radical-mediated chain reactions known collectively as lipid peroxidation. In food products – typically those rich in vegetable oils – peroxidation is associated with the onset of rancidity. In biological systems, lipid peroxidation is recognised as playing an important, underlying role in many pathological processes. Peroxidation of the lipids in low density lipoprotein (LDL), for example, is involved in the generation of a modified form of the particle (oxidised LDL or 'oxLDL') that is believed to play a key role in the development of atherosclerosis. Lipid peroxidation in biological membranes can lead to cell death and, even at low doses, can modify cell signalling through its effects on membrane fluidity.

A diverse range of non-radical products, many of which are potential toxins (e.g.carbonyl compounds and alkylhydroperoxides), is generated from lipids during their peroxidation. Whereas the lipid peroxidation seen under pathological conditions occurs through what are essentially random, uncontrolled chain reactions, a controlled enzymatic form of peroxidation is the means by which a whole range of prostaglandins and related chemical messengers are generated from fatty acids.

Antioxidants play an important role in the prevention (or, at least, suppression) of lipid peroxidation – for further details, see 'Antioxidant Chemistry'. In this section, the focus will be on the mechanisms through which lipid peroxidation is initiated, propagated (amplified) and terminated – the three 'classical' stages of free-radical chain reactions (seen also, for example, in combustion chemistry). The section concludes with a description of the expertise and services offered by Westcott Research and Consulting in this research area.

1 Mechanisms of lipid peroxidation

1.1 Initiation
Although there is no reason why the glycerol-based head group of a triglyceride or modified diglyceride cannot also be damaged by free radicals, lipid peroxidation refers specifically to the reactions which occur in the fatty-acid components of these molecules (or indeed in free fatty acids). Intact fatty acids do not contain unpaired electrons (they are not free radicals), therefore to initiate lipid peroxidation – i.e. to generate a fatty-acid centred radical – one of the following must take place:

  1. A species containing an unpaired electron – i.e. a free radical – must add to the fatty acid (an odd number added to an even number gives an odd number). The hydroxyl radical, for example, is an electrophile and will add across a double bond in an unsaturated fatty acid, resulting in the formation of a carbon-centred radical:

  2. An oxidant must remove a single electron from the fatty acid (an even number minus one gives an odd number). Although powerful one-electron oxidants, such peroxidase compound I-type species, the sulphate radical anion (SO4•–) and the Ce4+ ion, are capable of abstracting the electron, in most cases this is removed along with an accompanying proton: the overall effect is the removal of a hydrogen atom (e– + H+ = H atom) and the processes is described as a 'hydrogen atom abstraction' (or just 'hydrogen abstraction'). The hydroxyl radical, for example, can abstract a H-atom from a methylene group (–CH2–) in a saturated fatty acid:

    –CH2–CH2–CH2–CH2–     +     •OH     →    –CH2–•CH–CH2–CH2–     +     H2O

    Even relatively weak oxidants, such as HO2• (the protonated superoxide radical, for which the pKa value is 4.8), can abstract hydrogen from polyunsaturated fatty acids. This is because the resultant carbon-centred radicals are stabilised through delocalisation of the unpaired electron across the π-system:

    Oxygen readily adds to carbon-centred radicals (R•), giving peroxyl radicals (ROO•). Peroxyl radicals can themselves abstract hydrogen from fatty acids (RH), thereby initiating molecular damage at new sites, whether in a bulk lipid (e.g. in the interior of an LDL particle) or within a lipid bilayer. This transfer and amplification of damage is referred to as the propagation stage of lipid peroxidation:

    R• + O2 → ROO•

    ROO• + RH → ROOH + R•

    1.2 Propagation reactions
    The carbon-centred radicals (R•) generated during the propagation phase can themselves react with oxygen to form peroxyl radicals and attack yet more fatty acids. The lipid hydroperoxides (ROOH) also generated in the above reaction can react with redox-active metal ions (typically those of Fe and Cu), to generate peroxyl and alkoxyl (RO•) radicals. Trace amounts of iron ions, for example, can act catalytically in the following propagation reactions:

    Fe3+ + ROOH → Fe2+ + ROO• + H+

    Fe2+ + ROOH → Fe3+ + RO• + OH–

    In laboratory experiments, it is common practice to add metal ions to stimulate lipid peroxidation (e.g. in liposomes or LDL particles). Although there are mechanisms through which the direct addition of metal ions can bring about the initiation of lipid peroxidation (see below), in many cases the stimulation of peroxidation results from the reactions of the metal ions with preformed lipid hydroperoxides, which for various reasons are invariably present in lipid extracts. In such situations, the metal is not strictly initiating lipid peroxidation, but rather acting at the propagation phase.

    1.3 Chain termination reactions
    Lipid peroxidation can result in extensive damage to bulk lipids and biological membranes. The reactions come to a halt when either: (i) a radical is intercepted by a radical-scavenging antioxidant, as exemplified by the chemical repair of a peroxyl radical by vitamin E (see 'Antioxidant Chemistry'); or (ii) two radicals combine to form non-radical products e.g.:

    R• + R• → R-R

    2. Expertise and services offered by Westcott Research and Consulting
    We offer expertise and guidance in all areas involving lipid peroxidation: from the elucidation of basic reaction mechanisms through to the evaluation of the role of lipid peroxidation in biological systems (e.g. in specific disease scenarios, such as atherosclerosis) and in problems relating to food technology. Dr Burkitt has some 20 years' experience of research at the cutting-edge of lipid peroxidation. Some of the contributions he has made to the understanding of the process are described below:

    2.1 Mechanisms of iron-induced lipid peroxidation
    In the 1980s a number of researchers challenged the view that the hydroxyl radical is responsible for the initiation of iron-induced lipid peroxidation in micelles, liposomes, microsomes and related systems. Much of the evidence forwarded by these researchers was based on the 'unexpected' effects of chelating agents, radical-scavenging antioxidants, catalase, hydrogen peroxide and [Fe(II)]:[Fe(III)] ratios on the process. These observations led to the proposal that various species, including elusive Fe2+-Fe3+ and Fe2+-O2-Fe3+ complexes, ferryl intermediates and even 'cryto-hydroxyl radicals' are responsible for initiating lipid peroxidation by low-molecular-weight iron complexes (typically those of citrate, ADP and EDTA).

    In his early studies under the supervision of Prof. B. C. Gilbert at the University of York, Dr Burkitt showed how the observations which led to these proposals could be explained using conventional Fenton chemistry, in which the hydroxyl radical is generated through the reaction of Fe(II) with hydrogen peroxide, the latter being either added directly, generated enzymatically or formed via the direction reduction of oxygen by the reduced metal ion (see 'Metal ions in Health and Disease'). The effects of chelating agents, antioxidants and the like were rationalised in terms of a two-phase system involving distribution of the metal chelate and other reactants between the aqueous and lipid phases. Catalase, for example, which is confined to the aqueous phase, can enhance lipid peroxidation by preventing the oxidation of Fe(II) before it enters the lipid phase, where radical generation must take place if lipid peroxidation is to be induced. In subsequent studies, Dr Burkitt focused his efforts on the elucidation of the mechanisms of copper-induced lipid peroxidation – due largely to the interest there has been in the role of copper in the promotion of low density lipoprotein (LDL) oxidation during atherosclerosis.

    2.2 Initiation of lipid peroxidation by Cu2+ ions
    Cu(II) ions (e.g. from CuCl2) are often used to induce lipid peroxidation in liposomes and isolated LDL particles. Such preparations typically contain traces of lipid hydroperoxides (LOOH), which it is widely assumed react with the metal in redox-cycling reactions analogous to those described above for iron:

    Cu2+ + LOOH → Cu+ + ROO• + H+

    Cu+ + ROOH → Cu2+ + RO• + OH–

    Whilst the latter of these two reactions is thermodynamically favourable and proceeds readily for many Cu-complexes (for which the reduction potential of the Cu2+/Cu+ couple is some 0.15 V, i.e. well below that for LOOH/RO•,OH– at around 2.0 V at pH 7), the former reaction is thermodynamically unfavourable. Confusion and problems have arisen because it has been common practice to use the complexing agents bathocuproine and neocuproine to monitor this reaction.

    When copper ions bind to either of these agents, the reduction potential for the Cu2+/Cu+ couple is raised to about 0.6 V, which is far higher than the value for copper complexes of biological relevance. (In other words, Cu2+ becomes a much more powerful oxidising agent in the presence of either bathocuproine or neocuproine.) Researchers have done this because bathocuproine and neocuproine form stable, red-coloured complexes with Cu+ ions, which are easily quantified in a spectrophotometer. Although this provides a convenient means of monitoring the reduction of Cu2+ by LOOH (the former reaction), what is being followed is a reaction artefact: the inclusion of bathocuproine or neocuproine in the reaction system allows a reaction to happen which would otherwise not take place! Considerations along such lines led Dr Burkitt to propose (and then demonstrate) an alternative reaction involving the formation of Cu3+:

    Cu2+ + ROOH → Cu3+ + RO• + OH–

    This reaction is feasible for copper ions bound to peptides because the +3 oxidation state of the metal is stabilised by deprotonation of the amino acid ligands, resulting in a lowering of the Cu3+/Cu2+ potential to around 1.0 V. This favours (thermodynamically) the oxidation of Cu2+ to Cu3+ by LOOH according to the above reaction (where LOOH/RO•,OH– is around 2.0 V at pH 7). In a series of EPR spin-trapping experiments, Dr Burkitt and his assistant were able to prove the operation of this reaction (see Jones and Burkitt, 2003).

    Although the above reactions have been referred to as mechanisms for the initiation of lipid peroxidation, this is not strictly correct because the decomposition of alkylhydroperoxides occurs during the propagation phase. It is perhaps more accurate to say that the added metal ions are being used to re-start peroxidation that has stalled, due perhaps to the absence of metal ions (depending how the LOOH had been generated in the first place). We now turn our attention to a system in which copper ions are acting at the initiation stage of lipid peroxidation.

    2.3 Stimulation of copper-dependent lipid peroxidation by vitamin E
    The addition of Cu(II) ions to lipids devoid of hydroperoxides (LOOH and H2O2) would not normally result in lipid peroxidation. However if α-tocopherol (the major form of vitamin E) is present, Cu(II) will induce peroxidation. This is believed to involve the one-electron oxidation of α-tocopherol to the α-tocopheroxyl radical by the metal, as shown below where α-Toc-OH is the vitamin and α-Toc-O• its radical (see 'Antioxidant Chemistry' for the structures of these species):

    Cu2+ + α-Toc-OH → Cu+ + α-Toc-O• + H+

    Many researchers have assumed that because Cu(I) ions cannot form reactive species in the absence of a peroxide (whether H2O2 or LOOH), the species responsible for the initiation of peroxidation must be (however counterintuitive) the α-tocopheroxyl radical. As described under 'Antioxidant Chemistry', the α-Toc-O• radical is poorly reactive and therefore its formation usually results in the cessation of lipid peroxidation. Although the radical is able to abstract hydrogen atoms from polyunsaturated fatty acids (where the resultant carbon-centred lipid radical is somewhat stabilised by conjugation), the reaction is very slow (k ~ 0.1 dm3 mol–1 s–1).

    Those who have applied the so-called 'Tocopherol-Mediated Peroxidation' model of lipid peroxidation to such systems have ignored far too hastily the fate of the Cu(I) ions generated in the above reaction: Cu(I) ions are extremely unstable (unless, misguidedly, bathocuproine or neocuproine have been added to 'monitor' the reaction!) and will react rapidly with oxygen to give a series of reactive species, including the hydroxyl radical:

    Cu+ + O2 → Cu2+ + O2•–

    H+ + O2•– → HO2•

    Cu+ + HO2• + H+ → Cu2+ + H2O2

    Cu+ + H2O2 → Cu2+ + •OH + OH–

    The demonstration of the reduction of oxygen through to H2O2 by Cu(I) using an oxygen electrode is a trivial matter indeed. The subsequent reduction of the peroxide to •OH has been demonstrated by EPR spin-trapping (see Burkitt and Milne, 1996). Given that •OH is the most powerful one-electron oxidant that can be generated in a living organism, and reacts with all biological molecules at diffusion-controlled rates (k ~ 1010 dm3 mol–1 s–1), there can be simply no justification to attributing such a role to the α-Toc-O• radical, which attacks lipids with a rate constant some hundred trillion-times lower (and only polyunsaturated fatty acids at that).

    2.4 Spin-trapping radicals in lipid bilayers
    Dr Burkitt designed a test system in which to assess the performance of a series of novel EPR spin-trapping agents synthesised by Dr Richard Hartley and a graduate student at the University of Glasgow. These spin traps consist of a long, hydrophobic chain to which is attached a nitrone group and (usually) a polar head-group.

    Following their incorporation into lipid bilayers, these traps were shown form nitroxide radical-adducts with radicals trapped at depths within the membrane determined by the position of the nitrone function on the alkyl chain. Due to their partial immobilisation within membranes, the EPR spectra of these adducts display anisoptropic features typical of the nitroxide spin-probes used in membrane spin-labelling studies. Various spin traps were assessed and their spectra interpreted in terms of the extent to which their radical adducts were able to undergo 'bobbing' and 'wagging' motions in lipid bilayers (see Hay et al. 2005).

    In addition to their utility in the detection of radicals in membranes, these spin-traps - through their interception of radicals - are antioxidants with potential pharmacological applications: they can be targeted to intercept radicals at specific locations in biological membranes, such as those of mitochondria and other important sites.

    Selected publications showing the scope of Dr Burkitt's experience in this area: