Minireview

Protein Oxidation in Aging, Disease, and Oxidative Stress*

Barbara S. Berlett and Earl R. Stadtman From the Laboratory of Biochemistry, NHLBI, National Institutes of Health, Bethesda, Maryland 20892 The demonstration that oxidatively modified forms of proteins accumulate during aging, oxidative stress, and in some pathological conditions has focused attention on physiological and nonphysiological mechanisms for the generation of reactive oxygen species (ROS)1 and on the modification of biological molecules by various kinds of ROS. Basic principles that govern the oxidation of proteins by ROS were established in the pioneering studies of Swallow (1), Garrison (2, 3), and Scheussler and Schilling (4) who characterized reaction products formed when proteins were ex. , or posed to ionizing radiation under conditions where only OH, O2 a mixture of both was made available. Results of these studies demonstrated that the modification of proteins is initiated mainly by reactions with OH; however, the course of the oxidation process . or its protonated is determined by the availability of O2 and O2 ). Collectively, these ROS can lead to oxidation of amino form (HO2 acid residue side chains, formation of protein-protein cross-linkages, and oxidation of the protein backbone resulting in protein fragmentation. In the meantime, it has been shown that other forms of ROS may yield similar products and that transition metal . in some of the reactions.2 ions can substitute for OH and O2 Oxidation of the Protein Backbone As is illustrated in Fig. 1, oxidative attack of the polypeptide backbone is initiated by the OH-dependent abstraction of the -hydrogen atom of an amino acid residue to form a carbon-centered radical (Fig. 1, Reaction c). The OH needed for this reaction may be obtained by radiolysis of water or by metal-catalyzed cleavage of H2O2 (Reactions a and b). The carbon-centered radical thus formed reacts rapidly with O2 to form an alkylperoxyl radical intermediate (Reaction d), which can give rise to the alkylperoxide (Reaction f), followed by formation of an alkoxyl radical (Reaction h), which may be converted to a hydroxyl protein derivative (Reaction j). Significantly, many of the steps in this pathway that are mediated by can be catalyzed also by Fe2 (Reactions e, g, interactions with HO2 and i)2 or by Cu (not shown). The alkyl, alkylperoxyl, and alkoxyl radical intermediates in this pathway may undergo side reactions with other amino acid residues in the same or a different protein molecule to generate a new carbon-centered radical (Reaction 1) capable of undergoing reactions similar to those illustrated in Fig. 1.
P

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 272, No. 33, Issue of August 15, pp. 2031320316, 1997 Printed in U.S.A.

R1C R2C 3 R1CCR2

P P

REACTION 2

Protein Fragmentation The generation of alkoxyl radicals (Fig. 1, Reactions h and g) sets the stage for cleavage of the peptide bond by either the diamide or -amidation pathways. Upon cleavage by the diamide pathway (Fig. 2, Pathway a), the peptide fragment derived from the N-terminal portion of the protein possesses a diamide structure at the C-terminal end, whereas the peptide derived from the C-terminal portion of the protein possesses an isocyanate structure at the N-terminal end. In contrast, upon cleavage by the -amidation pathway (Fig. 2, Pathway b), the peptide fragment obtained from the N-terminal portion of the protein possesses an amide group at the C-terminal end, whereas the N-terminal amino acid residue of the fragment derived from the C-terminal portion of the protein exists as an N--ketoacyl derivative. Upon acid hydrolysis, the peptide fragments obtained by the diamide pathway will yield CO2, NH3, and a free carboxylic acid, whereas hydrolysis of the fragment obtained by the -amidation pathway yields NH3 and a free -ketocarboxylic acid. Peptide bond cleavage can occur also as a result of ROS attack of glutamyl, aspartyl, and prolyl side chains. As described by Garrison (2), OH-dependent abstraction of a hydrogen atom from the -carbon atom of a glutamyl residue, followed by reactions analogous to Reactions d, f, and h in Fig. 1, will lead eventually to peptide bond cleavage by a mechanism in which oxalic acid is formed and the N-terminal amino acid of the peptide derived from the C-terminal portion of the protein will exist as an N-pyruvyl derivative (Reaction 3). Based on the observation that the number of peptides formed during radiolysis of proteins is approximately equal to the number of prolyl residues, Schuessler and Schilling (4) proposed that oxidation of prolyl residues would lead to peptide bond cleavage. This was verified by studies of Uchida et al. (5) showing that oxidation of proline residues leads to the formation of 2-pyrrolidone and concomitant peptide bond cleavage (Reaction 4). Because acid hydrolysis of 2-pyrrolidone yields 4-aminobutyric acid, the presence of 4-aminobutyric acid in protein hydrolysates is presumptive evidence for peptide bond cleavage by the proline oxidation pathway. Oxidation of Amino Acid Side Chains All amino acid residues of proteins are susceptible to oxidation by OH. However, the products formed in the oxidation of some residues have not been fully characterized. Table I lists some of the products formed during the oxidation of the residues that are most susceptible to oxidation. Oxidation of Sulfur-containing Amino Acid ResiduesCysteine and methionine residues are particularly sensitive to oxidation by almost all forms of ROS. Under even mild conditions cysteine residues are converted to disulfides and methionine residues are converted to methionine sulfoxide (MeSOX) residues. Most biological systems contain disulfide reductases and MeSOX reductases that can convert the oxidized forms of cysteine and methionine residues back to their unmodified forms. These are the only oxidative modifications of proteins that can be repaired. Based on the observation that preferential oxidation of several exposed methionine residues in some proteins has little effect on their biological function, it was proposed that the cyclic oxidation-reduction of methionine residues serves as a built-in ROS scavenger system to protect such proteins from more extensive irreversible oxidative modifications (6). This proposition is supported by results of recent studies showing that a knock-out strain of yeast lacking MeSOX reductase is more sensitive to H2O2 toxicity than the wild-type strain and that, when grown in the presence of H2O2, the protein and free amino acid pool of the mutant strain contain higher levels

R1C or R1OO or R1O R2 C H 3 R2C R1H or R1OOH or R1OH

P P P

REACTION 1

Moreover, in the absence of oxygen, when Reaction d in Fig. 1 is prevented, the carbon-centered radical may react with another carbon-centered radical to form a protein-protein cross-linked derivative (Reaction 2).
* This minireview will be reprinted in the 1997 Minireview Compendium, which will be available in December, 1997. This is the fourth article of five in the Oxidative Modification of Macromolecules Minireview Series. To whom correspondence should be addressed: Laboratory of Biochemistry, Bldg. 3, Rm. 222, NHLBI, National Institutes of Health, Bethesda, MD 20892. Tel.: 301-496-4096; Fax: 301-496-0599. 1 The abbreviations used are: ROS, reactive oxygen species; MeSOX, methionine sulfoxide reductase; PN, peroxynitrite; GS, glutamine synthetase. 2 For review, see Donald C. Borg and Karen M. Schaich (1988) in Oxygen Radicals and Tissue Injury (Halliwell, B., ed) pp. 20 26, Proceedings of an Upjohn Symposium, Federation of American Societies for Experimental Biology, Bethesda, MD. This paper is available on line at http://www.jbc.org

20313

20314

Minireview: Protein Oxidation

interconversion between phosphorylated and unphosphorylated forms (7) or between nucleotidylated and unmodified forms (8). Accordingly, nitration would compromise one of the most important mechanisms of cellular regulation of key enzyme activities and of signal transduction networks (7). This possibility is underscored by the demonstration that nitration of tyrosine residues in model substrates prevents the phosphorylation of these residues by protein tyrosine kinases (9, 10) and by the demonstration that nitration of tyrosine residues in Escherichia coli glutamine synthetase (GS) converts the enzyme to a form with regulatory properties similar to those obtained by in vivo enzyme-catalyzed adenylylation of a single tyrosine residue in each subunit of the enzyme (11). The enzyme-catalyzed cyclic adenylylation and deadenylylation of GS is the basis of an exquisite mechanism for the feedback regulation of GS activity by diverse end products of glutamine metabolism (8). In contrast, the nitration of tyrosine residues is an irreversible process and therefore locks the enzyme into a relatively inactive configuration. The ability of PN to nitrate tyrosine residues and oxidize methionine residues of proteins is dependent upon the availability of CO2. In the absence of CO2, PN is in equilibrium with an activated form (PN*) of unknown structure that reacts rapidly with methionine residues to form MeSOX (12), but in the presence of CO2, PN is almost instantly converted to a derivative (possibly ONOOCO 2 or O2NOCO 2 ) that can nitrate aromatic compounds (1316). Accordingly, the nitration of tyrosine and the oxidation of methionine residues of proteins are mutually exclusive processes that are differentially regulated by the availability of CO2, as illustrated in Scheme 1. Curiously, in the case of GS, there is little or no nitration of tyrosine residues in the complete absence of CO2, and no oxidation of methionine residues occurs in the presence of physiological concentrations of CO2 (i.e. 5% CO2, pH 7.4). Even so, the PN-dependent oxidation of methionine residues in the absence of CO2 and the nitration of tyrosine residues in the presence of CO2 both convert GS to a form with regulatory properties similar to those obtained by enzyme-catalyzed adenylylation of a single tyrosine in each subunit of the enzyme (17). Generation of Protein Carbonyl Derivatives As already noted, oxidative cleavage of proteins by either the -amidation pathway (Fig. 2) or by oxidation of glutamyl side chains (Reaction 3) leads to formation of a peptide in which the N-terminal amino acid is blocked by an -ketoacyl derivative. However, as shown in Table I, direct oxidation of lysine, arginine, proline, and threonine residues may also yield carbonyl derivatives. In addition, carbonyl groups may be introduced into proteins by reactions with aldehydes (4-hydroxy-2-nonenal, malondialdehyde) produced during lipid peroxidation (Fig. 3A) (18 20) or with reactive carbonyl derivatives (ketoamines, ketoaldehydes, deoxyosones) generated as a consequence of the reaction of reducing sugars or their oxidation products with lysine residues of proteins (2123) (glycation and glycoxidation reactions) (Fig. 3B). The presence of carbonyl groups in proteins has therefore been used as a marker of ROS-mediated protein oxidation, and several sensitive methods for the detection and quantitation of protein carbonyl groups have been developed (24). As judged by the presence of carbonyl groups, it has been established that protein oxidation is associated with aging, oxidative stress, and a number of diseases. Oxidative Stress-induced Protein OxidationElevated levels of oxidized protein are present in animals and cell cultures following their exposure to various conditions of oxidative stress. Thus, exposure of animals or cell cultures to either hyperoxia, forced exercise, ischemia-reperfusion, rapid correction of hyponatremia, paraquat toxicity, magnesium deficiency, ozone, neutrophil activation, cigarette smoking, x-radiation, chronic alcohol treatment, or mixed function oxidation systems leads to an increase in the level of oxidized protein.4 Protein Oxidation and AgingAging is associated with the accumulation of inactive or less active, more heat-labile forms of numerous enzymes (25, 26). The possibility that these age-related changes are due, at least in part, to oxidative modification is
4 For review, see E. R. Stadtman and B. S. Berlett (1997) Chem. Res. Toxicol. 10, 485 494.

FIG. 1. Oxygen free radical-mediated oxidation of proteins.

FIG. 2. Peptide bond cleavage by the (a) diamide and (b) -amidation pathways.

REACTION 3

REACTION 4

of MeSOX than are present in the wild type strain.3 Aromatic Amino Acid ResiduesAromatic amino acid residues are among the preferred targets for ROS attack. As shown in Table I, tryptophan residues are readily oxidized to formylkynurenine and kynurenine and to various hydroxy derivatives; phenylalanine and tyrosine residues yield a number of hydroxy derivatives; histidine residues are converted to 2-oxohistidine, asparagine, and aspartic acid residues. Reactions with PeroxynitriteWith the discovery that nitric oxide is a normal product of arginine metabolism and that it reacts . to form peroxynitrite (Reaction 5), the biological rapidly with O2 effects of peroxynitrite (PN) have been extensively studied.
. NO 3 ONOO O2 REACTION 5

Methionine and cysteine residues of proteins are particularly vulnerable to oxidation by PN, and tyrosine and tryptophan residues are selective targets for PN-dependent nitration. The nitration of tyrosine residues may be of singular importance since nitration precludes the ability of tyrosine residues to undergo cyclic
3 J. Moskovitz, B. S. Berlett, M. J. Poston, R. L. Levine, and E. R. Stadtman, unpublished results.

Minireview: Protein Oxidation

TABLE I Amino acids most susceptible to oxidation
Amino acids Cysteine Methionine Tryptophan Phenylalanine Tyrosine Histidine Arginine Lysine Proline Threonine Glutamyl Oxidation products

20315

Disulfides, cysteic acid Methionine sulfoxide, methionine sulfone 2-, 4-, 5-, 6-, and 7-Hydroxytryptophan, nitrotryptophan, kynurenine, 3-hydroxykynurinine, formylkynurinine 2,3-Dihydroxyphenylalanine, 2-, 3-, and 4-hydroxyphenylalanine 3,4-Dihydroxyphenylalanine, tyrosine-tyrosine cross-linkages, Tyr-O-Tyr, cross-linked nitrotyrosine 2-Oxohistidine, asparagine, aspartic acid Glutamic semialdehyde -Aminoadipic semialdehyde 2-Pyrrolidone, 4- and 5-hydroxyproline pyroglutamic acid, glutamic semialdehyde 2-Amino-3-ketobutyric acid Oxalic acid, pyruvic acid

Protein Oxidation and DiseaseAccumulation of oxidized protein (protein carbonyls) is associated with a number of diseases, including amyotrophic lateral sclerosis, Alzheimers disease, respiratory distress syndrome, muscular dystrophy, cataractogenesis, rheumatoid arthritis, progeria, and Werners syndrome.4 Although the level of carbonyl has not been directly determined, there is reason to believe that oxidative modification of proteins is implicated also in atherosclerosis, diabetes, Parkinsons disease, essential hypertension, cystic fibrosis, and ulcerative colitis.4
SCHEME 1

FIG. 3. Formation of protein carbonyls by glycation, glycoxidation, and by reactions with peroxidation products of polyunsaturated fatty acids (PUFA). A, reactions of sugars with protein lysyl amino groups (P-NH2). B, Michael addition of 4-hydroxy-2-nonenal to protein lysine (PNH2), histidine (P-His), or cysteine (PSH) residues. C, reaction of protein amino groups (PNH2) with the lipid peroxidation product, malondialdehyde.

indicated by the facts. (a) In vitro exposure of enzymes to ROS elicits changes in catalytic activity, heat stability, and proteolytic susceptibility similar to those that occur during aging (2730). (b) Brief exposure of animals to oxidative stress leads to alterations in enzymes similar to that associated with aging (31, 32). (c) Old animals are more susceptible than young animals to protein damage during oxidative stress, e.g. x-radiation, H2O2 (33, 34). (d) There is an age-related increase in the carbonyl content of protein in human brain (35), gerbil brain (36), eye lens (37), rat hepatocytes (32), whole body protein of flies (38), and human red blood cells (28). (e) The carbonyl content of protein in cultured human fibroblasts increases exponentially as a function of the age of the fibroblast donor (28). (f) There is an inverse relationship between regimens that lead to an increase in life span and regimens that lead to an increase in protein carbonyl content and vice versa (39). For example, diet (caloric) restriction of rats (39) and mice (40) leads to an increase in life span and to a decrease in the level of protein carbonyls. When compared at the same chronological age, strains of short lived houseflies contain higher levels of oxidized proteins than their longer lived cohorts (41).

Accumulation of Oxidized Protein The intracellular level of oxidized protein reflects the balance between the rate of protein oxidation and the rate of oxidized protein degradation. This balance is a complex function of numerous factors that lead to the generation of ROS, on the one hand, and of multiple factors that determine the concentrations and/or activities of the proteases that degrade oxidatively damaged protein, on the other. As illustrated in Fig. 4, many different physiological and environmental processes lead to the formation of ROS. Collectively, these processes can promote the generation of a battery of ROS, . , R, ROO, RO, NO, including a number of free radicals (OH, O2 RS, ROS, RSOO, and RSSR . ), various non-radical oxygen derivatives (H2O2, ROOH, 1O2, O3, HOCl, ONOO, ONOCO 2, O2NOCO 2 , N2O2, NO2 , and highly reactive lipid- or carbohydratederived carbonyl compounds, viz. 4-hydroxy-2-nonenal, malondialdehyde ketoamines, ketoaldehydes, and deoxyosones. Any one of these ROS is capable of promoting the modification of proteins. However, as shown in Fig. 4, their abilities to do so are dependent upon the concentrations of a myriad of enzymic and non-enzymic factors (antioxidants) that can either inhibit the formation of ROS or facilitate their conversion to inactive derivatives. For example, . formed by several pro-oxidant systems shown in Fig. 4 is the O2 readily converted to H2O2 by the action of superoxide dismutase. This H2O2 together with H2O2 produced by various oxidases and metal-catalyzed oxidation systems is readily degraded by catalase, glutathione peroxidase, thiol-specific antioxidant enzymes, and other peroxidases. However, if in the course of metabolism the concentrations of these antioxidant activities become insufficient to decompose all of the H2O2 formed, the H2O2 may undergo metal ion-catalyzed cleavage by the Fenton reaction to generate the even more toxic OH. This reaction is dependent upon the availability of iron and copper, which is determined by the concentrations of metal-binding proteins (ferritin, transferrin, lactoferrin, and ceruloplasmin), and of multiple factors (iron-responsive elements, etc.) that control the intracellular concentrations of these proteins, as well as factors that influence the binding and/or the release of metal ions from these binding proteins. The level of ROS is also a function of the concentrations of vitamins (A, C, and E) and of metabolites (uric acid, bilirubin, etc.) that are capable of either scavenging free radicals directly or of facilitating the regeneration of metabolites that do so. Finally, metal ion chelators can either suppress or enhance the rates of ROS generation by forming complexes with iron or copper that inhibit their ability to catalyze ROS formation or alter their redox potentials and therefore their ability to undergo cyclic interconversion between oxidized and reduced states. Furthermore, other divalent cations (Mg2, Mn2, and Zn2) may compete with Fe(II) or Cu(I) for binding to metal binding sites on proteins and thereby prevent site-specific generation of OH, which is likely the most important mechanism of protein damage (42). In addition, Mn(II) is able to inhibit the reduction of Fe(III) to Fe(II) (43) and thus

20316

Minireview: Protein Oxidation

specific processes summarized in Fig. 4 are exaggerated, leading to unique manifestations that are characteristic of the disorder.
REFERENCES
1. Swallow, A. J. (1960) in Radiation Chemistry of Organic Compounds (Swallow, A. J., ed) pp. 211224, Pergamon Press, New York 2. Garrison, W. M. (1987) Chem. Rev. 87, 381398 3. Garrison, W. M., Jayko, M. E., and Bennett, W. (1962) Radiat. Res. 16, 487502 4. Schuessler, H., and Schilling, K. (1984) Int. J. Radiat. Biol. 45, 267281 5. Uchida, K., Kato, Y., and Kawakishi, S. (1990) Biochem. Biophys. Res. Commun. 169, 265271 6. Levine, R. L., Mosoni, L., Berlett, B. S., and Stadtman, E. R. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 15036 15040 7. Hunter, T. (1995) Cell 80, 225236 8. Chock, P. B., and Stadtman, E. R. (1996) in Principles of Medical Biology (Bittar, E. E., and Bittar, N., eds) Vol. 4, pp 201220, Jai Press, Greenwich, CT 9. Kong, S.-K., Yim, M. B., Stadtman, E. R., and Chock, P. B. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 33773382 10. Gow, A. J., Duran, D., Malcolm, S., and Ischiropoulos, H. (1996) FEBS Lett. 385, 63 66 11. Berlett, B. S., Friguet, B., Yim, M. B., Chock, P. B., and Stadtman, E. R. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 1776 1780 12. Pryor, W. A., Jin, X., and Squadrito, G. L. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 1117311177 13. Lymar, S. V., and Hurst, J. K. (1995) J. Am. Chem. Soc. 117, 8867 8868 14. Lymar, S. V., Jiang, Q., and Hurst, J. K. (1996) Biochemistry 35, 78557861 15. Uppu, R. M., Squadrito, G. L., and Pryor, W. A. (1996) Arch. Biochem. Biophys. 327, 335343 16. Denicola, A., Freeman, B. A., Trujillo, M., and Radi, R. (1996) Arch. Biochem. Biophys. 333, 49 58 17. Berlett, B. S., and Stadtman, E. R. (1996) FASEB J. 10, A1100 (Abstr. 585) 18. Schuenstein, E., and Esterbauer, H. (1979) CIBA Found. Symp. 67, 225244 19. Esterbauer, H., Schaur, R. J., and Zolner, H. (1991) Free Radical Biol. Med. 11, 81128 20. Uchida, K., and Stadtman, E. R. (1993) J. Biol. Chem. 268, 6388 6393 21. Kristal, B. S., and Yu, B. P. (1992) J. Gerontol. 47, B104 B107 22. Baynes, J. W. (1996) Diabetes 40, 405 411 23. Monnier, V., Gerhardinger, C., Marion, M. S., and Taneda, S. (1995) in Oxidative Stress and Aging (Cutler, R. G., Packer, L., Bertram, J., and Mori, A., eds) pp. 141149, Birkhauser Verlag, Basel, Switzerland 24. Levine, R. L., Williams, J. A., Stadtman, E. R., and Schacter, E. (1994) Methods Enzymol. 233, 346 357 25. Dreyfus, J. C., Kahn, A., and Schapira, F. (1978) Curr. Top. Cell. Regul. 14, 243297 26. Rothstein, M. (1977) Mech. Ageing Dev. 6, 241257 27. Takahashi, R., and Goto, S. (1990) Arch. Biochem. Biophys. 277, 228 233 28. Oliver, C. N., Ahn, B.-W., Moerman, E. J., Goldstein, S., and Stadtman, E. R. (1987) J. Biol. Chem. 262, 5488 5491 29. Zhou, J. Q., and Gafni, A. (1991) J. Gerontol. 46, B217B221 30. Oliver, C. N., Fucci, L., Levine, R. L., Wittenberger, M. E., and Stadtman, E. R. (1982) in Cytochrome P-450, Biochemistry, Biophysics, and Environmental Implications (Heitanen, E., Laitinen, M., and Hanninen, O., eds) pp. 531539, Elsevier Biomedical Press, Amsterdam 31. Starke, P. E., Oliver, C. N., and Stadtman, E. R. (1987) FASEB J. 1, 36 39 32. Starke-Reed, P. E., and Oliver, C. N. (1989) Arch. Biochem. Biophys. 275, 559 567 33. Agarwal, S., and Sohal, R S. (1993) Biochem. Biophys. Res. Commun. 194, 12031206 34. Sohal, R. S., Agarwal, S., and Sohal, B. H. (1995) Mech. Ageing Dev. 81, 1525 35. Smith, C. D., Carney, J. M., Starke-Reed, P. E., Oliver, C. N., Stadtman, E. R., and Floyd, R. A. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 10540 10543 36. Carney, J. M., Starke-Reed, P. E., Oliver, C. N., Landum, R. W., Cheng, M. S., Wu, J. F., and Floyd, R. A. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 36333636 37. Garland, D., Russell, P., and Zigler, J. S. (1988) in Oxygen Radicals in Biology and Medicine (Simic, M. G., Taylor, K. S., Ward, J. F., and von Sontag, V., eds) pp. 347353, Plenum Publishing Corp., New York 38. Sohal, R. S., Agarwal, S., Dubey, A., and Orr, W. C. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 72557259 39. Sohal, R. S., Ku, H.-H., Agarwal, S., Forster, M. J., and Lal, H. (1994) Mech. Ageing Dev. 79, 121133 40. Youngman, L. D., Park, J.-Y. K., and Ames, B. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 91129116 41. Sohal, R. S., Ku, H.-H., and Agarwal, S. (1993) Biochem. Biophys. Res. Commun. 196, 711 42. Stadtman, E. R. (1990) Free Radical Biol. Med. 9, 315325 43. Nakamura, K., Oliver, C., and Stadtman, E. R. (1985) Arch. Biochem. Biophys. 240, 319 329 44. Friguet, B., Szweda, L., and Stadtman, E. R. (1994) Arch. Biochem. Biophys. 311, 168 173 45. Grune, T., Reinheckel, T., Joshi, M., and Davies, K. J. A. (1995) J. Biol. Chem. 270, 2344 2351 46. Grant, A. J., Jessup, W., and Dean, R. J. (1993) Free Rad. Res. Comms. 18, 259 267 47. Sua rez, G., Etlinger, J. D., Maturana, J., and Weitman, D. (1995) Arch. Biochem. Biophys. 321, 209 213 48. Rivett, A. J. (1986) Curr. Top. Cell. Regul. 28, 291337 49. Yu, C. E., Wijsman, E. M., Nakura, J., Miki, T., Piussan, C., Matthews, S., Fu, Y. H., Mulligan, J., Martin, G. M., and Schellenberg, G. D. (1997) Am. J. Hum. Genet. 60, 330 341

FIG. 4. Accumulation of oxidized protein is dependent upon the balance between pro-oxidant, antioxidant, and proteolytic activities. MSR, methionine sulfoxide reductase; GPx, glutathione peroxidase; CAT, catalase; RSH-Px, thiol-specific peroxidase; NOS, nitric oxide synthetase; SOD, superoxide dismutase; GST, glutathione transferase.

prevent its ability to promote formation of OH by the Fenton reaction as well as the generation of other forms of ROS as illustrated in Fig. 1. As noted above, the accumulation of oxidized protein reflects not only the rate of protein oxidation but also the rate of oxidized protein degradation, which (as shown in Fig. 4) is also dependent upon many variables, including the concentrations of proteases that preferentially degrade oxidized proteins and numerous factors (metal ions, inhibitors, activators, and regulatory proteins) that affect their proteolytic activities. For example, oxidized forms of some proteins (e.g. cross-linked proteins (44 46)) and proteins modified by glycation (47) or lipid peroxidation products (44) are not only resistant to proteolysis but, in fact, can inhibit the ability of proteases to degrade the oxidized forms of other proteins (44, 48). The foregoing discussion and some of the points illustrated in Fig. 4 are by no means comprehensive. They are intended to call attention to the extraordinary complexity of ROS biochemistry. Numerous other factors not discussed are certainly important in determining the steady-state level of oxidative damage under varying physiological and environmental conditions. It is our belief that during aging there is a progressive accumulation of errors at the level of DNA that affect any one or more of the factors that govern the dynamics of protein oxidation and oxidized protein degradation. This leads to a shift in the balance between these processes in favor of oxidized protein accumulation and attendant loss of biological function. Two observations are consistent with this hypothesis. (a) The level of oxidized protein in cultured fibroblasts is a function of the age of the fibroblast donor and is independent of the cell passage number. (b) Chronic injection of the free radical scavenger tert-butyl--phenylnitrone leads to a reversal of some agerelated changes in the gerbil brain, but when the tert-butyl-phenylnitrone treatment is discontinued the age-related changes reappear (36). Both observations indicate that the level of oxidative damage is determined by the genetic make-up of the cell, which changes with age. According to this proposition, the aged phenotype could be expressed (a) by a single point mutation that could impair a biological activity that occupies a central role in biological functions, such an alteration of helicase as occurs in Werners syndrome (49), or (b) by the accumulation over time of numerous errors leading to deficiencies in the synthesis and/or activities of a multiplicity of the factors that govern the balance between protein oxidation and degradation. From this perspective, aging could be looked upon as a degenerative process (disease?) that might include aberrations that contribute to the development of other pathologies, such as Alzheimers disease, amyotrophic lateral sclerosis, diabetes, etc., in which the accumulation of oxidatively modified protein has been demonstrated. Perhaps in these diseases one or more of the