Original contributionCell-mediated reduction of protein and peptide hydroperoxides to reactive free radicals
Introduction
It is well established that exposure of proteins to free radicals in the presence of O2 can result in physical and chemical changes, loss of structural or enzymatic activity, and biological perturbations 1, 2, 3, 4, 5. These alterations include oxidation of side chain groups, backbone cleavage or fragmentation, cross-linking, unfolding, changes in hydrophobicity and conformation, altered susceptibility to proteolytic enzymes, and the formation of new reactive groups such as 3,4-dihydroxyphenylalanine (DOPA), hydroperoxides, and carbonyls (both protein-bound and released). Hydroperoxides are generated in high yields at both side-chain and backbone (α-carbon) sites 6, 7, 8, 9. Thus, reaction of HO• with Val and Leu side chains, in the presence of O2, yields 3-, and 4-hydroperoxyvaline and 4- and 5-hydroperoxyleucines, respectively 10, 11, whereas reaction with Ala peptides yields primarily backbone peroxides [9]. While other products are also formed, including alcohols and carbonyls, the high yields of hydroperoxides formed presumably reflect the diffusion-controlled addition of O2 to the initial carbon-centered radical, rapid reaction of the resulting peroxyl radical with other hydrogen-atom donors (particularly on proteins where dimerization/disproportionation reactions are disfavored), and the stability of the resulting hydroperoxide.
Hydroperoxides decompose readily in the presence of light, heat, reducing agents, and metal ions, but are reasonably stable in the absence of these agents 6, 7, 12, 13. One-electron reduction of hydroperoxides (e.g., by Cu+ or Fe2+), present at either α-carbon or side-chain sites, gives alkoxyl radicals (reaction 1), that can undergo subsequent rearrangement (1,2-hydrogen shift, reaction 2), fragmentation (β-scission, reaction 3), and hydrogen-atom abstraction reactions yielding further carbon-centered radicals, as well as products such as alcohols (reaction 4) and carbonyls (reaction 3) 8, 9. The carbon-centred radicals generated by such fragmentation/rearrangement reactions can react further with O2 to give peroxyl species (reaction 5) that can then give rise to further hydroperoxides via hydrogen-atom abstraction reactions (reaction 6). These reactions are believed to constitute the chain reaction of protein oxidation. Alcohols can also be generated, in addition to the one-electron processes shown in reactions 1 followed by 4, by direct two-electron reduction (e.g., by NaBH4 or 2-mercaptoethanol; reaction 7). The alcohols formed on reduction of Val, Leu, and Lys hydroperoxides have been shown to be stable to further degradation 10, 11, 14, and have been employed as markers of oxidative stress in vitro and in vivo 5, 15.
Amino acid, peptide, and protein hydroperoxides can be reduced by cellular components including enzymes and low-molecular-weight species. Reduction of Val hydroperoxides has been demonstrated to occur at accelerated rates in the presence of GSH peroxidase plus GSH [13] with the corresponding alcohols as major products. In contrast, BSA-hydroperoxides are reduced slowly by this system with ∼80% of the initial peroxides remaining after 1 h [13]. Reduction by low-molecular-weight molecules, including GSH and ascorbate, but not NADH or NADPH, can occur [6], with the fastest rates of reduction observed with the smallest peroxides; this presumably reflects steric effects.
The fate of peroxides in biological fluids and cellular systems has also been examined 13, 16. Human plasma can degrade both Val and Leu peptide hydroperoxides to the alcohols, though not in stoichiometric amounts [13]. Mouse macrophage-like J774 cells can also reduce hydroperoxides, while cell-free conditions yielded little degradation [13]. Erythrocytes have also been shown to react with a variety of hydroperoxides [16]. Although the last of these studies did not include direct measurements of peroxide degradation, isolated erythrocyte membranes (ghosts) showed a loss of membrane protein thiols, and aggregation of membrane proteins was observed due to interpeptide disulphide bond formation. Inhibition of Na+/K+- and Ca2+-ATPases, whose activity depend on key thiol groups, was also observed [16]. There was no effect on the nonthiol-dependent enzyme acetylcholinesterase, suggesting that thiols may be important targets for protein peroxides. Comparison of the effects of various amino acid peroxides and tert-BuOOH showed the latter had a more marked effect on the membrane-bound enzymes at identical concentrations, suggesting that the permeability of the peroxide into the cell is a major factor. The polar and charged amino acids peroxides would be expected to be less efficiently taken up by cells than small, neutral peroxides such as tert-BuOOH [16]. Peptide and protein peroxides would be expected to be taken up by cells even less readily, though there is little data available on this point.
Three possible types of process may give rise to the observed peroxide reduction with cells. First, intracellular reduction after penetration of the peroxide into the cell; second, reduction at the cell surface; and third, reduction by reducing equivalents released by the cell. The data available to date, and particularly the observation of membrane protein alterations, and the absence of reduction in cell-free controls, suggest that the last of these three alternatives is the least likely. This is in accord with recent studies on the oxidation of low-density lipoprotein induced by J774 cells [17]. In these studies little removal of lipid peroxides was observed with cell-conditioned media (i.e., media which had been in contact with cells for extended periods and the cells subsequently removed). The oxidation observed in the complete, cell-containing system has been postulated to arise primarily via the reduction of trace transition metal ions to a more reactive form by a cell-surface reduction system, such as a trans-plasma membrane electron transport (TPMET) chain. This system, whose activity can be stimulated by ascorbate loading of the cells, has been shown to stimulate the reduction of some Fe3+ complexes to the Fe2+ form, Cu2+ to Cu+ [18], as well as a range of other redox-sensitive agents. This membrane electron transport chain, and analogous NADH-dependent systems, have been shown to be present in a wide range of mammalian cell types, including endothelial, monocyte, macrophage, and erythrocyte cells 17, 19, 20, 21, 22, 23. Furthermore, erythrocytes have recently been shown to reduce extracellular ascorbate free radicals using intracellular ascorbate as the electron donor [24].
In light of the above data we were interested in exploring the hypothesis that such electron transport systems might mediate the extracellular reduction of amino acid, peptide, and protein hydroperoxides to reactive radicals either directly, or via the redox cycling of extracellular trace transition metal ions. We have therefore investigated the reduction of various peroxides, formed on aliphatic substrates, by viable THP-1 cells (a human monocyte-like cell line) both with and without ascorbate preloading. Peroxide loss has been quantified under various conditions and radical formation examined by EPR spin trapping. In addition, oxidative stress in these cells, as a result of peroxide exposure, has been examined by monitoring the signal of the ascorbate free radical.
Section snippets
Materials
Protected amino acids and peptides were obtained from Bachem (Bubendorf, Switzerland) or Sigma-Aldrich (Castle Hill, NSW, Australia), and used as received. Amino acids and peptide hydroperoxides were generated from the materials by 60Co radiolysis in the presence of O2 25, 26. Irradiated solutions were treated with catalase (5 mg ml−1) immediately after the cessation of radiolysis to remove radiation-generated H2O2 prior to the determination of hydroperoxide yields and further use.
Hydroperoxide yields in γ-irradiated substrate solutions
The peroxide concentrations in γ-irradiated or oxidized amino acid, peptide and protein solutions were determined by the FOX assay and the results are given in Table 1. The corresponding hydroperoxide yields for the parent, nonirradiated, samples were negligible (< 1.0 μM).
Effect of amino acid, peptide, and protein hydroperoxides on THP-1 cell viability and proliferation
The viability data for THP-1 cells treated with various amino acid and peptide solutions over a 4 h period, as assessed by LDH activity, are shown in Fig. 1a. For the hydroperoxide-containing substrates, small but
Discussion
The addition of amino acid and peptide hydroperoxides at the concentrations employed to THP-1 cells resulted in small, but significant, changes in cell viability and cell proliferation compared to cells treated with the nonoxidized (parent) materials (Fig. 1). Under these conditions THP-1 cells stimulated the reduction of some, but not all, of the added hydroperoxides, as determined by the FOX assay (Fig. 2). The most marked stimulation in hydroperoxide loss was detected with those materials
Abbreviations
Asc—ascorbate
BSA—bovine serum albumin
DFO—deferoxamine (desferrioxamine) mesylate salt
DHA—dehydroascorbate
DMPO—5,5-dimethyl-1-pyrroline N-oxide
EDTA—the tetra sodium salt of ethylenediamine tetraacetic acid
EPR—electron paramagnetic resonance
HBSS—Hank’s balanced salt solution
[Ni(en)3]2+—Tris(ethylenediamine)nickel(II) complex
PBN—N-tert-butyl-α-phenylnitrone
PBS—phosphate-buffered saline
Protein-C•—protein carbon-centered radical
Protein-O•—protein alkoxyl radical
Protein-OO•—protein peroxyl radical
Acknowledgements
The authors are grateful to the Australian Research Council, the Association for International Cancer Research, and the Wellcome Trust for financial support. We also thank Prof. Roger Dean for helpful discussions.
References (37)
- et al.
Proteins damaged by oxygen radicals are rapidly degraded in extracts of red blood cells
J Biol Chem
(1987) Protein and peptide alkoxyl radicals can give rise to C-terminal decarboxylation and backbone cleavage
Arch Biochem Biophys
(1996)- et al.
Structural identification of valine hydroperoxides and hydroxides on radical-damaged amino acid, peptide, and protein molecules
Free Radic Biol Med
(1995) - et al.
Stable markers of oxidant damage to proteins and their application in the study of human disease
Free Radic Biol Med
(1999) - et al.
Effect of amino acid peroxides on the erythrocyte
Free Radic Biol Med
(1996) - et al.
Direct copper reduction by macrophages. Its role in low density lipoprotein oxidation
J Biol Chem
(1997) - et al.
Ascorbate-dependent electron transfer across the human erythrocyte membrane
Biochim. Biophys. Acta
(1999) - et al.
Erythrocytes reduce extracellular ascorbate free radicals using intracellular ascorbate as an electron donor
J Biol Chem
(2000) - et al.
Beta-scission of side-chain alkoxyl radicals on peptides and proteins results in the loss of side chains as aldehydes and ketones
Free Radic Biol Med
(2002) - et al.
Hydroperoxide assay with the ferric-xylenol orange complex
Anal Biochem
(1999)