Serial Review: Oxidatively Modified Proteins in Aging and Disease

Guest Editor: Earl Stadtman

ROLE OF OXIDATIVE STRESS AND PROTEIN OXIDATION IN THE

AGING PROCESS

R

AJINDAR

S. S

OHAL

Department of Molecular Pharmacology and Toxicology, University of Southern California, Los Angeles, CA, USA

(Received 28 December 2001; Revised 25 March 2002; Accepted 27 March 2002)

Abstract—The hypothesis that the rate of oxygen consumption and the ensuing accrual of molecular oxidative damage
constitute a fundamental mechanism governing the rate of aging is supported by several lines of evidence: (i) life spans
of cold blooded animals and mammals with unstable basal metabolic rate (BMR) are extended and oxidative damage
(OxD) is attenuated by an experimental decrease in metabolic rate; (ii) single gene mutations in Drosophila and
Caenorhabditis elegans that extend life span almost invariably result in a generalized slowing of physiological activities,
albeit via different mechanisms, affecting a decrease in OxD; (iii) caloric restriction decreases body temperature and
OxD; and, (iv) results of studies on the effects of transgenic overexpressions of antioxidant enzymes are generally
supportive, but quite ambiguous. It is suggested that oxidative damage to proteins plays a crucial role in aging because
oxidized proteins lose catalytic function and are preferentially hydrolyzed. It is hypothesized that oxidative damage to
specific proteins constitutes one of the mechanisms linking oxidative stress/damage and age-associated losses in
physiological functions.

© 2002 Elsevier Science Inc.

Keywords—Oxidative stress, Protein oxidation, Free radical hypothesis of aging, Senescence, Mechanisms of aging,
Metabolic rate, Free radicals

INTRODUCTION

The postreproductive phase of life of virtually all multi-
cellular species is characterized by the progressive de-
cline in the efficiency of various physiological functions,

whereby the ability to maintain homeostasis is corre-
spondingly attenuated, leading eventually to the death of
the organism. Although many hypotheses have been
advanced, the nature of the causal mechanisms that ini-
tiate the deleterious alterations underlying this phenom-
enon, often referred to as “senescence” or the “aging
process,” remains controversial. It has been advocated
previously that any causal hypothesis should explain: (i)
the mechanistic basis of the age-associated losses in
physiological capacity in individual organisms, (ii) vari-
ations in the rates of progression of senescent alterations
among different individuals and species, and (iii) the
possibility that life span can be extended in some species
by experimental manipulations such as caloric restriction
in rodents or decrease in metabolic rate in poikilotherms,
and single gene mutations in Caenorhabditis elegans and
Drosophila melanogaster [1]. A currently popular hy-
pothesis postulates that a progressive accumulation of
macromolecular oxidative damage is the fundamental
underlying cause of senescence-associated deleterious
alterations. The nature of the evidence supporting the

This article is part of a series of reviews on “Oxidatively Modified

Proteins in Aging and Disease.” The full list of papers may be found on
the homepage of the journal.

Rajindar S. Sohal is the current holder of the Timothy M. Chan

Professorship in the Department of Molecular Pharmacology and Tox-
icology at the University of Southern California. He was previously a
University Distinguished Professor in the Department of Biological
Sciences at Southern Methodist University in Dallas, Texas. He was a
Senior Scholar in the Department of Zoology, University of Cam-
bridge, where he was a member of Wolfson College; a Guest Professor
at University of Dusseldorf, Germany; and a Visiting Professor at
Linkoping University, Sweden, where he also received a Doctor of
Medicine (Honoris causa). He received B.Sc. (Honors) and M.Sc.
(honors) degrees in Zoology from Panjab University, Chandigarh,
India, and Ph.D. degree in Zoology from Tulane University, New
Orleans.

Address correspondence to: Dr. Rajindar S. Sohal, University of

Southern California, Department of Molecular Pharmacology and Tox-
icology, 1985 Zonal Avenue, Los Angeles, CA 90033, USA; Tel: (323)
442-1860; Fax: [phone]; E-Mail: [e-mail].

Free Radical Biology & Medicine, Vol. 33, No. 1, pp. 37– 44, 2002

Copyright © 2002 Elsevier Science Inc.

Printed in the USA. All rights reserved

0891-5849/02/$–see front matter

PII S0891-5849(02)00856-0

37

role of molecular oxidative damage in general, and pro-
tein oxidation in particular, in the aging process is dis-
cussed in this article.

The idea that the rate of energy utilization is a key

determinant of the rate of aging was first proffered by
Pearl [2] on the basis of a series of studies on the effects
of starvation on the survival of D. melanogaster and
growth of cantaloupe seedlings. His main conclusions
were that duration of life is a function of two variables:
(i) the genetically determined constitution of the individ-
ual or “vitality,” and (ii) the average rate of metabolism
during life (pp. 139, 140). The second inference was
expressed by the phrase that “in general the duration of
life varies inversely as the rate of energy expenditure
during life” (p. 145), which has subsequently become
known as the “rate of living” hypothesis. The most
unambiguous support for the rate of living hypothesis
has been provided by studies in poikilotherms, where
lowering the rate of metabolism by virtually any nonle-
thal means, such as a decrease in the ambient tempera-
ture, mutational inactivation of enzymes, decreased level
of physical activity, have all been found to result in
increased life spans [3–5]. Among mammals, life spans
of certain species, such as Turkish hamster, can be ex-
tended experimentally by induced hibernation [6].

Critics of the “rate of living” hypothesis have fre-

quently cited the example of birds, which are both rela-
tively long-lived as well as having a high metabolic rate
as compared to the mammals, to discredit the validity of
this hypothesis. This argument, however, misrepresents
Pearl’s point of view. Variations noted in the rate of
growth of different seedlings and length of survival of
individual flies in response to starvation indeed formed
the basis of his first postulate, namely that different
individuals have different sums of vitality. In fact, Pearl
explicitly stated that the relationship between duration of
life and rate of metabolism applied only to the same
individual. The notion that different species should ex-
pend identical amounts of energy during life was thus
implicitly excluded. Furthermore, this notion also con-
tradicts his explicitly stated postulate that the duration of
life also depends on the genetically determined constitu-
tion of the individual. Thus, the genetic constitution
determines the total amount of energy consumed during
life, or metabolic potential, while the rate at which that
energy is consumed, the metabolic rate, determines the
length of life. Subsequent studies have indicated that the
total amount of energy consumed during life or meta-
bolic potential, varies in different phylogenetic groups
[3,7]. However, within each group, there is a clearly
demonstrable inverse relationship between metabolic
rate and species-specific life span [7]. Although the basis
of such a relationship was unknown in Pearl’s era, the
subsequent demonstration that oxygen consumption by

aerobes entails the generation of reactive oxygen species
(ROS) has provided a clear link between Pearl’s rate of
living hypothesis and the oxidative stress hypothesis of
aging [8,9].

OXIDATIVE STRESS HYPOTHESIS OF AGING

An initial question that is highly pertinent to the

evaluation of oxidative stress hypothesis is whether the
existing evidence supports or refutes the various predic-
tions of the hypothesis.

Is organismic senescence due to accumulation of oxida-
tive damage? It has now been amply documented that
there is an age-associated increase in the steady state
amounts of the products of free radical attacks on mac-
romolecules such as lipids, proteins, and DNA in various
tissues [1,10 –12]. In general, the oxidative damage,
which has often been measured in tissue homogenates,
increases exponentially with age [10,13]. Tissues that are
composed of long-lived, postmitotic cells, such as the
brain, heart, and skeletal muscle, tend to accrue relatively
greater amounts of damage than those composed of
short-lived nonmitotic cells [1,11,14]. Studies on possi-
ble causes of age-associated increase in oxidative dam-
age have indicated that activities of some of the enzy-
matic antioxidative defenses may decrease, while others
increase or remain unaltered at different ages, suggesting
the absence of any generalizable pattern [15–17]. On the
other hand, rates of mitochondrial O

2

⫺

and H

2

O

2

gener-

ation have been found to increase quite consistently with
age [17–21]. Although the existing information on age-
related changes in the efficiency of degradation of oxi-
dized molecules is relatively limited, a decrease in pro-
teolytic activity with age has been noted in some, but not
in other tissues [22–25]. To conclude, the available data
tend to favor the view that the increased production of
ROS is the primary factor responsible for age-related
accrual of molecular oxidative damage. It is possible that
elevation in ROS generation may secondarily damage
the proteosomal system [24,25].

Are interspecies variations in the rate of aging related to
corresponding differences in rate of accrual of oxidative
damage? Interspecies comparisons among nonprimate
mammalian species such as mouse, rat, rabbit, pig, and
horse, among others, have indicated that maximum life
span (MLS) of the species is not correlated with the
activities of antioxidative enzymes, such as superoxide
dismutase, catalase, and glutathione peroxidase [26,27].
On the other hand, rates of mitochondrial O

2

⫺

and H

2

O

2

generation are inversely related to MLS of the species
[28 –30]. Notably, the relationship between MLS of the

38

R. S. S

OHAL

nonprimate mammalian species and basal metabolic rate
(BMR) is superimposable on that between their MLS and
the rates of mitochondrial O

2

⫺

/H

2

O

2

generation [28].

A similar inverse relationship also exists between MLS
of different nonprimate mammalian species and the
steady state amounts of mitochondrial 8-hydroxydeox-
yguanosine, a product of DNA oxidation [31].

Different species of dipteran flies, that vary 2-fold in

average life span, also exhibited inverse relationships
between average life span and both the rates of mito-
chondrial O

2

⫺

and H

2

O

2

generation and the protein car-

bonyl content of tissues [32]. Some phylogenetic groups,
such as the birds and the primates, live relatively longer
than nonprimate mammalian species that have compara-
ble metabolic rates. For instance, the pigeon and the rat
have similar metabolic rates but differ 7- to 8-fold in
MLS, meaning that they have different metabolic poten-
tials. Comparisons between these two species have indi-
cated that mitochondrial rates of O

2

⫺

and H

2

O

2

genera-

tion are considerably lower, whereas activities of some
antioxidative enzymes were higher in the pigeon than in
the rat [33,34]. In summary, MLS of the species, which
have a similar metabolic potential, seem to be inversely
associated with rates of ROS generation and apparently
unrelated to antioxidative defenses, whereas species that
differ in metabolic potential as well as metabolic rate,
MLS is negatively correlated with ROS generation and
positively related to antioxidant defenses.

Are experimental extensions in life span accompanied by
corresponding attenuations of oxidative damage? Avail-
able evidence suggests that, almost invariably, experi-
mental regimens that extend the life spans of poikilo-
therms also result in decreases in the rates of metabolism
and accumulation of oxidative damage [1]. For instance,
elimination of flying activity prolongs the life span of
flies up to 3-fold, while the rate of accumulation of
protein and DNA oxidative damage is correspondingly
diminished [13,55]. Among mammals, especially rodents
such as the rat and the mouse, life span is extended if
caloric intake is decreased from the level consumed by
ad libitum (AL) fed animals. The body temperature of
calorically restricted (CR) animals is daily transiently
lowered up to 4°C in rat and up to 13°C in the mouse,
indicating that these species have unstableness in the rate
of basal metabolism, which is responsive to caloric in-
take [reviewed in 1,36]. This issue has, however, become
controversial because direct comparisons of rates of ox-
ygen consumption per unit body mass were found not to
exhibit any significant differences between AL and DR
groups [37]. In counterpoint, it has been argued that due
to the differences in the body composition and the rela-
tive organ size between the AL and CR animals, body
weight can not be used in this model system for normal-

ization of the rate of oxygen consumption [36]. For
example, fats constitute 18% of the total body weight in
AL and only 7% in CR rats, meaning that lean body mass
cannot be calculated on the basis of the assumption that
it is a fixed proportion of the body weight, as done by
McCarter and Palmer [34]. Notwithstanding this debate,
there is a large body of experimental data indicating that,
among rodents, decrease in the rate of oxygen consump-
tion and body temperature are widely employed physio-
logical mechanisms for adaptation to scarcity of food
[38,39].

The rates of mitochondrial O

2

⫺

and H

2

O

2

generation

as well as steady state concentrations of the products of
free radical attacks on macromolecules, such as lipids,
proteins, and DNA, are lower in tissues of CR than in the
AL-fed animals [1,36]. In general, there are no consistent
or significant differences in activities of antioxidant en-
zymes between the two groups. In summary, the existing
information can be reasonably interpreted to suggest that
in the species that have variable basal metabolic rates,
life span can be experimentally extended by lowering
metabolic activity via a variety of manipulations. Low-
ering of the metabolic rate demonstrably results in a
corresponding attenuation of oxidative damage and an
increase in life span.

Can the rate of aging be altered by overexpression of
antioxidative enzymes or single gene mutations? Results
of studies on overexpression of antioxidative genes have
been quite variable and somewhat ambiguous. Trans-
genic overexpression of Cu, ZnSOD has no life-prolong-
ing effect in mouse [40] and, according to most authors,
in D. melanogaster [41– 43]. Overexpression of Mn-
SOD [44], catalase [45], or a putative thioredoxin reduc-
tase [46] alone also have no effect on life span of
Drosophila, albeit in some cases the experimental ani-
mals exhibit relatively higher resistance to induced oxi-
dative stress. In another study, simultaneous overexpres-
sion of Cu, Zn-SOD and catalase was found to result in
an increase of up to 34% in life span and an attenuation
of macromolecular oxidative damage in D. melanogaster
[47]. In a recent critique of the various transgenic studies
on aging in Drosophila, it was pointed out that the life
span prolongation effects of overexpression of antioxi-
dative genes may be limited to certain genetic back-
grounds only [48]. In general, life spans of transgenic
overexpressors of antioxidative enzymes may be ex-
tended if the life spans of their respective controls are
relatively short. For instance, the longest reported exten-
sion of life span in Cu-Zn SOD overexpressing D. mela-
nogaster (48%) was reported in a strain with an average
life span of 24 d [49], which is about one-third of the
length among healthy, relatively long-lived strains of this
species. In this particular study, the life span extensions

39

Oxidative stress and aging

seemed to be inversely related to Cu, Zn-SOD activity as
well as the life spans of the respective controls. Thus, it
is imperative that life span extensions of transgenic an-
imals should be evaluated in the context of the life spans
of the control strain(s) used vs. those of the robust
long-lived strains of the same species. In most instances,
life span extensions of the transgenic overexpressors
may merely be due to amelioration of a deficiency in an
unhealthy stock rather than retardation of the aging pro-
cess.

Recently, there has been a spate of reports about

extensions of life spans in mutants of C. elegans, D.
melanogaster, and mice, which have been interpreted to
suggest that the rate of aging is controlled by a limited
number of genes [59]. A common feature of virtually all
such life-prolonging mutations seems to be the loss of
specific function, which leads to the decline of the nor-
mal tempo of bioenergetic metabolism with ensuing hy-
pometabolic state [45,48]. Although various novel and
complicated mechanisms have been proposed by the
authors to explain such life span extensions, direct evi-
dence has already been presented for a simpler interpre-
tation that these extensions occur due to a decrease in the
rate of metabolism and consequently of ROS generation
[4,5]. Metabolic defects resulting from a variety of mu-
tations result in phenotypes, which have relatively slower
rates of physiological activities. Hypometabolic state is
known to increase resistance to various stresses and to
enhance antioxidative defenses. These responses are the
effects of hypometabolism rather than the cause of ex-
tended life spans [48].

To summarize, the view that oxidative stress is causal

to the induction of senescence is presently supported
primarily by correlative evidence, such as that amounts
of macromolecular oxidative damage and rates of gen-
eration of mitochondrial ROS increase with age and are
inversely related to the MLS of the species. Extension of
life span following overexpression of antioxidative en-
zymes seems to be inversely related to life span of the
respective controls. Prolongation of life span in single
gene mutants is almost invariably associated with atten-
uations in metabolic rate and/or fecundity. Unfortu-
nately, most of the authors have either ignored the role of
rate of metabolism or have used unsuitable procedures
for its measurement [4]

PROTEIN OXIDATIVE DAMAGE AND AGING

Among the various types of macromolecular oxida-

tive damage that accrues during aging, oxidative modi-
fications of intracellular proteins have been suggested to
play a key role in the causation of senescence-associated
losses in physiological functions because oxidized pro-
teins often lose catalytic function and undergo selective

degradation [10,50,51]. Oxidative damage to a specific
protein, especially at the active site, can induce a pro-
gressive loss of a particular biochemical function. Pio-
neering studies by Stadtman, Levine, and their associates
have documented the relevance of protein oxidative
damage in the aging process and in the etiology of
certain pathological conditions [50,52]. Several types of
ROS-induced protein modifications have been demon-
strated [50,51,53], including the loss of sulfhyryl (

⫺SH)

groups, formation of carbonyls, disulphide crosslinks,
methionine sulfoxide, dityrosine cross-links, nitroty-
rosine, and glyoxidation and lipid peroxidation adducts,
among others.

Loss of protein

⫺SH groups can be induced by a wide

array of ROS and is one of the most immediate responses
to an elevation in the level of oxidative stress [50].
Functional consequences of

⫺SH loss include protein

misfolding, catalytic inactivation, decreased antioxida-
tive capacity, and loss of certain specific functions, such
as binding of heavy metals and sulfur-containing amino
acids by albumin, among others [53]. Age-associated
losses in protein

⫺SH content have been reported in a

variety of tissues and species, including homogenates of
brain, heart, skeletal muscle, and kidney of rodents and
houseflies [20,54,55]. Caloric restriction attenuates

⫺SH

loss, whereas hyperoxia has an opposite effect [54,56].

Another oxidative modification in proteins is the for-

mation of dityrosine crosslinks, which apparently arises
following reaction between two tyrosyl radicals, gener-
ated by peroxidases and other heme proteins. Dityrosine
cross-linking of proteins has been found to increase with
age in mouse skeletal muscle and heart, but not in the
brain or liver; caloric restriction attenuates these in-
creases [57].

Addition of carbonyl-containing adducts to the side

chains of amino acid residues, such as lysine, arginine,
proline, and threonine, is arguably the most well charac-
terized, age-associated, post-translational structural alter-
ation in proteins [10,50,51]. Carbonylation can be caused
by a variety of ROS; however, site-specific metal-cata-
lyzed oxidation, involving the formation of hydroxyl/
ferryl radical via Fenton-type scission of H

2

O

2

, seems to

be the most plausible in vivo mechanism [58]. A series of
studies by Stadtman and his associates have demon-
strated that amounts of protein carbonyl content in-
creases with age in several different mammalian tissues
[24,58].

A relationship between protein carbonylation and the

rate of aging or life expectancy of animals has also been
demonstrated in a number of organs and species. For
instance, the level of protein carbonyls in cultured human
fibroblasts increases exponentially with age of the donors
[58]. Similarly, in the housefly, protein carbonyl content
in whole body homogenates increased with age in an

40

R. S. S

OHAL

exponential fashion [35]. In flies whose life span was
extended around 2- to 3-fold by elimination of flying
activity, the accrual of protein carbonyls with age was
slower than in flies that were permitted to fly and had a
relatively shorter life span [35]. In another experiment
using flying ability as a criterion, aged flies were divided
into two groups, the “crawlers” and the “fliers,” which,
respectively, have relatively short and long life expect-
ancies. At comparable ages, the crawlers were found to
contain a higher concentration of protein carbonyls than
the fliers, suggesting that onset of senescence is associ-
ated with more rapid accrual of protein oxidative damage
[35]. Overall, the findings in insects that protein carbon-
yls increase with age and their steady state concentration
is inversely related to life expectancy have been con-
firmed in mammals [58].

Another approach to examine the relationship be-

tween life expectancy and protein carbonylation in-
volved a comparison between ad libitum (AL) fed and
calorically restricted (CR) mice, which had a

⬃35%

longer life span than the AL mice. The age-related
accrual of protein carbonyl content in homogenates of
various tissues was considerably slower in the CR than
the AL mice [17,54,56]. Similarly, whole body ho-
mogenates of relatively long-lived transgenic D.
melagonaster, overexpressing Cu, Zn-SOD and cata-
lase, accrued protein carbonyls at a slower rate than
the control [42].

Variations in the rate of accrual of oxidatively

modified proteins in vivo have been variously hypoth-
esized to be due to corresponding differences in rates
of oxidant generation, antioxidative defenses, repair
and degradative capacity, or susceptibility to oxidative
modifications [61,62]. For instance, exposure of flies
to hyperoxia or relatively higher physical activity,
which increases oxidant generation, also increase the
rate of accumulation of protein carbonyls [35]. Evi-
dence supporting the modulatory role of antioxidative
defenses in the accumulation of protein carbonyl is
based on studies in transgenic Drosophila overex-
pressing Cu, Zn-SOD and catalase [42]. Decreases in
the ability to degrade carbonylated proteins is corre-
lated in some tissues with age-related increases in the
amount of carbonylated proteins [50,51]. The tissues
of the aged animals also seem to be more susceptible
to sustain protein oxidation in response to experimen-
tally induced oxidative stress than those of the young
animals. For instance, aged, live houseflies, when ex-
posed to x-rays, exhibited a higher net gain of protein
carbonyls than the younger flies [52]. Comparisons
among different species with varying longevities also
indicate that tissues of relatively longer-lived species,
such as the pigeon and the cow, are less susceptible to
radiographically induced protein carbonylation than

the tissues of shorter-lived species like mice and rats
[52]. To conclude, it seems that a combination of
several factors determines the steady state amounts of
protein oxidation products, Nevertheless, rates of ox-
idant generation may be the most dominant factor,
since variations in rates of ROS generation are most
closely associated with differences in oxidative dam-
age and MLS of different species [31,32].

SELECTIVITY OF PROTEIN OXIDATIVE DAMAGE

Although there is strong evidence indicating that the

steady state amounts of the products of free radical attack
on macromolecules, such as protein carbonyls, increase
in tissue homogenates during aging, such evidence does
not elucidate the specific mechanisms that cause losses in
particular cellular functions. It was originally thought
that free radical attacks on proteins and other macromol-
ecules occur randomly because such interactions are
uncatalyzed events. However, it is well documented that
activities of most enzymes do not decline during aging
[63,64]. To understand the basis of this apparent discrep-
ancy and to identify the possible mechanisms underlying
biochemical losses during aging, we hypothesized that
age-associated oxidative damage to proteins and the con-
sequent loss of their function was a selective rather than
a random phenomenon. This hypothesis was initially
tested in mitochondria of the flight muscles of houseflies
and D. melanogaster by using Western blot analysis for
the detection of specific proteins exhibiting carbonyla-
tion [60,61,65].

It was found that only aconitase and adenine nucle-

otide translocase (ANT) exhibited a detectable age-
associated increase in carbonylation and a correspond-
ing loss in functional activity, suggesting that protein
carbonylation during aging is selective [60,61,65].
Both aconitase and ANT were also found to be par-
ticularly sensitive to carbonylation under conditions of
oxidative stress, induced by exposure to hyperoxia.
Studies on rodent plasma also indicate that only a
small fraction of proteins exhibit discernible carbony-
lation at any age [66], suggesting that the specificity of
protein oxidative damage during aging is not limited
to insects. ROS probably act in a random fashion;
however, the sensitivities and proximities of potential
targets differ. The factors that affect selectivity of
oxidative damage to proteins include the presence of a
metal-binding site, molecular conformation, rate of
proteolysis, and relative abundance of amino acid res-
idues susceptible to metal-catalyzed oxidation, among
others [65,66]. Selectivity of protein carbonylation
was further indicated by the findings that molecular
mass of cytochrome c in flies remained unchanged
during aging as well as in response to hyperoxia [67].

41

Oxidative stress and aging

Similarly, malate dehydrogenase, which copurifies
with aconitase, did not exhibit any discernible car-
bonylation in the aged flies [68]. Such findings are
thus inconsistent with the previously prevalent view
that protein carbonylation during aging is a general
phenomenon. This concept also accords with a large
body of data indicating that a vast majority of enzymes
do not lose catalytic activity during aging.

Damage to specific proteins has been hypothesized

to constitute an important mechanism linking oxida-
tive stress and loss of physiological function during
the aging process. Damage to such proteins has been
suggested to be a marker of physiological age or life
expectancy of organisms [69]. Support for these ideas
is provided by studies in which flies were prevented
from flight activity. Prevention of flying was found to
simultaneously result in a 3-fold increase in life span
and a decrease in: (i) rate of oxygen consumption of
flies, (ii) rate of mitochondrial H

2

0

2

generation, (iii)

carbonyl content of aconitase and ANT, and (iv) the
loss in the activity of these proteins [69]. Similarly, an
increase in ambient temperature of D. melanogaster,
which elevates the rate of oxygen consumption and
decreases life span of flies, was found to accelerate
aconitase carbonylation and loss of enzyme activity.
Taken together, findings in the housefly and D. mela-
nogaster suggest that carbonylation of aconitase and
ANT is associated directly with the rate of metabolism
and is inversely related to the life span of the insects.

What is the putative mechanism by which selective

protein oxidative damage, such as that observed in
insect mitochondria, plays a causal role in senescence?
It can be reasoned that damage to key proteins can
potentially result in a myriad of secondary deleterious
alterations. For instance, some of the potential conse-
quenses of loss of aconitase activity can be the slow-
ing down of glycolysis and the tricarboxylic (TCA)
acid cycle, with consequent decrease in flow of elec-
trons to oxygen, leading to depression of oxidative
phosphorylation. Since TCA cycle reduces NAD

⫹ to

NADH at several points, a pro-oxidizing shift may
occur in NAD

⫹/NADH ratio. Another consequence of

the loss of aconitase activity would be the accumula-
tion of citrate, which has been documented in insects
and rats [70,71]. Citrate can bind Fe

2

⫹

, which, in turn,

can cause scission of H

2

O

2

to generate the highly

reactive hydroxyl free radical. Similarly, a decrease in
ANT activity may have broad physiological conse-
quences that are similar to those occurring normally
during senescence, such as a decrease in the maximal
ADP-stimulated state 3 respiration and a consequent
increase in mitochondrial H

2

O

2

generation due to en-

hanced autoxidation of electron transport carrier mol-
ecules. In short, it seems that oxidative damage to

specific protein targets may be an important mecha-
nism linking oxidative stress to age-associated physi-
ological losses.

REFERENCES

[1] Sohal, R. S.; Weindruch, R. Oxidative stress, caloric restriction,

and aging. Science 273:59 – 63; 1996.

[2] Pearl, R. The rate of living. New York: Alfred A. Knopf; 1928.
[3] Sohal, R. S. The rate of living theory: a contemporary interpre-

tation. In: Collatz, K. G.; Sohal, R. S., eds. Insect aging. Heidel-
berg, Germany: Springer-Verlag; 1986:23– 44.

[4] Van Voorhies, W. A.; Ward, S. Genetic and environmental con-

ditions that increase longevity in Caenorhabditis elegans decrease
metabolic rate. Proc. Natl. Acad. Sci. USA 96:11399 –11403;
1999.

[5] Feng, J.; Bussiere, F.; Hekimi, S. Mitochondrial electron transport

is a key determinant of life span in Caenorhabditis elegans.
Develop. Cell 1:633– 644; 2001.

[6] Lyman, C. P.; O’Brian, R. C.; Greene, G. C.; Papafrangos, E. D.

Hibernation and longevity in the Turkish hamster Mesocricetus
brandti. Science 212:668 – 670; 1981.

[7] Cutler, R. G. Antioxidants, aging, and longevity. In: Pryor W. A.,

ed. Free radicals in biology. New York: Academic Press; 1984:
371– 428.

[8] Fridovich, I. The biology of oxygen radicals. Science 201:875–

880; 1978.

[9] Chance, B.; Sies, H.; Boveris, A. Hydroperoxide metabolism in

mammalian organs. Physiol. Rev. 59:527– 605; 1979.

[10] Stadtman, E. R. Protein oxidation and aging. Science 257:1220 –

1224; 1992.

[11] Beckman, K. B.; Ames, B. N. The free radical theory of aging

matures. Physiol. Rev. 78:547–581; 1998.

[12] Grune, T.; Davies, K. J. A. Oxidative processes in aging. In:

Masoro, E. J.; Austad, S. N., eds. Handbook of the biology of
aging. San Diego: Academic Press; 2001:25–58.

[13] Agarwal, S.; Sohal, R. S. DNA oxidative damage and life expect-

ancy in houseflies. Proc. Natl. Acad. Sci. USA 91:12332–12335;
1994.

[14] Sohal, R. S.; Agarwal, S.; Candas, M.; Forster, M.; Lal, H. Effect

of age and caloric restriction on DNA oxidative damage in dif-
ferent tissues of C57BL/6 mice. Mech. Aging Develop. 76:215–
224; 1994.

[15] Sohal, R. S. The free radical hypothesis of aging: an appraisal of

the current status. Aging Clin. Exp. Res. 5:3–17; 1993.

[16] Rikans, L. E.; Moore, D. R.; Snowden, C. D. Sex-dependent

differences in the effects of aging on antioxidant defense mech-
anisms of rat liver. Biochim. Biophys. Acta 1074:195–200; 1991.

[17] Sohal, R. S.; Ku, H.-H.; Agarwal, S.; Forster, M. J.; Lal, H.

Oxidative damage, mitochondrial oxidant generation, and antiox-
idant defenses during aging and in response to food restriction in
the mouse. Mech. Aging Develop. 74:121–133; 1994.

[18] Nohl, H.; Hegner, D. Do mitochondria produce oxygen radicals in

vivo? Eur. J. Biochem. 82:563–567; 1978.

[19] Sohal, R. S.; Arnold, L. A.; Sohal, B. H. Age-related changes in

antioxidant enzymes and pro-oxidant generation in tissues of the
rat with special reference to parameters in two insect species. Free
Radic. Biol. Med. 9:495–500; 1990.

[20] Sohal, R. S.; Agarwal, S.; Sohal, B. H. Oxidative stress and aging

in Mongolian gerbil (Meriones unguiculatus). Mech. Aging De-
velop. 81:15–25; 1995.

[21] Sohal, R. S.; Sohal, B. H. Hydrogen peroxide release by mito-

chondria increases during aging. Mech. Aging Develop. 57:187–
202; 1991.

[22] Starke-Reed, P. E.; Oliver, C. N. Protein oxidation and proteolysis

during aging and oxidative stress. Arch. Biochem. Biophys. 275:
559 –567; 1989.

[23] Agarwal, S.; Sohal, R. S. Aging and proteolysis of oxidized

proteins. Arch. Biochem. Biophys. 309:24 –28; 1994.

[24] Stadtman, E. R. The status of oxidatively modified proteins as a

42

R. S. S

OHAL

marker of aging. In: Esser, K.; Martin, G. M., eds. Molecular
aspects of aging. New York: John Wiley & Sons; 1995:129 –142.

[25] Grune, T.; Shringarpure, R.; Sitte, N.; Davies, K. Age-related

changes in protein oxidation and proteolysis in mammalian cells.
J. Geront. 56:B459 –B467; 2001.

[26] Sohal, R. S.; Sohal, B. H.; Brunk, U. T. Relationship between

antioxidant defenses and longevity in different mammalian spe-
cies. Mech. Aging Develop. 53:217–227; 1990.

[27] Perez-Campo, R.; Lopez-Torres, M.; Rojas, C.; Cadenas, S.;

Barja, G. Longevity and antioxidant enzymes, nonenzymatic an-
tioxidants, and oxidative stress in the vertebrate lung: a compar-
ative study. J. Comp. Phys. 163B:682– 689; 1994.

[28] Ku, H.-H.; Brunk, U. T.; Sohal, R. S. Relationship between

mitochondrial superoxide and hydrogen peroxide production and
longevity of mammalian species. Free Radic. Biol. Med. 15:621–
627; 1994.

[29] Sohal, R. S.; Lass, A.; Yan, L.-J.; Kwong, L. K. Mitochondrial

generation of reactive oxygen species and oxidative damage dur-
ing aging: roles of coenzyme Q and tocopherol. In: Cadenas, E.;
Packer, L., eds. Understanding the process of aging. New York:
Marcel Dekker; 1999:

[30] Barja, G. Mitochondrial free radical production and aging in

mammals and birds. Ann. NY Acad. Sci. 854:224 –238; 1998.

[31] Barja, G.; Herrero, A. Oxidative damage to mitochondrial DNA is

inversely related to maximum life span of mammals. FASEB J.
14:312–318; 2000.

[32] Sohal, R. S.; Sohal, B. H.; Orr, W. C. Mitochondrial superoxide

and hydrogen peroxide generation, protein oxidative damage, and
longevity in different species of flies. Free Radic. Biol. Med.
19:499 –504; 1995.

[33] Ku, H.-H.; Sohal, R. S. Comparison of mitochondrial pro-oxidant

generation and antioxidant defenses between rat and pigeon:
possible basis of variation in longevity and metabolic potential.
Mech. Aging Develop. 72:67–76; 1993.

[34] Barja, G.; Cadenas, S.; Rojas, C.; Perez-Campo, R.; Lopez-

Torres, M. Low mitochondrial free radical production per unit O

2

consumption can explain the simultaneous presence of high lon-
gevity and high aerobic metabolic rate in birds. Free Radic. Res.
21:317–328; 1994.

[35] Sohal, R. S.; Agarwal, S.; Dubey, A.; Orr, W. C. Protein oxidative

damage is associated with life expectancy of houseflies. Proc.
Natl. Acad. Sci. USA 90:7255–7259; 1993.

[36] Weindruch, R.; Sohal, R. S. Caloric intake and aging. N. Engl.

J. Med. 337:986 –994; 1997.

[37] McCarter R. J.; Palmer, J. Energy metabolism and aging: a

lifelong study of Fisher 344 rats. Amer. J. Physiol. 263:E448 –
E452; 1922.

[38] Hand, S. C. Hardewig, I. Downregulation of cellular metabolism

during environmental stress: mechanisms and implications. Ann.
Rev. Physiol. 58:539 –563; 1996.

[39] Wilz, M.; Heldmair, G. Comparison of hibernation, estivation,

and daily torpor in the edible dormouse, Glis glis. J. Comp.
Physiol. 170B:511–521; 2000.

[40] Huang, T. T.; Carlson, E. J.; Gillespie, A. M.; Shi, Y.; Epstein,

C. J. Ubiquitous overexpression of CnZn superoxide dismutase
does not extend life span in mice. J. Geront. 55B:5–9; 2000.

[41] Seto, N. O. L.; Hayashi, S.; Tener, G. M. Overexpression of Cu,

Zn-superoxide dismutase in Drosophila does not affect life span.
Proc. Natl. Acad. Sci. USA 87:4270 – 4274; 1990.

[42] Orr, W. C.; Sohal, R. S. The effects of Cu, Zn-superoxide dis-

mutase gene overexpression on life span and resistance to oxida-
tive stress in transgenic Drosophila melanogaster. Arch. Biochem.
Biophys. 301:34 – 40; 1993.

[43] Reveillaud, I.; Niedzwiecki, A.; Bensch, K. G.; Fleming, J. E.

Expression of bovine superoxide dismutase in Drosophila mela-
nogaster augments resistance to oxidative stress. Mol. Cell. Biol.
11:632– 640; 1991.

[44] Mockett, R. J.; Orr, W. C.; Rahmandar, J. F.; Benes, J. J.; Radyuk,

S. N.; Klichko, V. I.; Sohal, R. S. Overexpression of Mn-contain-

ing superoxide dismutase in trangenic Drosophila melanogaster.
Arch. Biochem. Biophys. 371:260 –269; 1999.

[45] Orr, W. C.; Sohal, R. S. The effects of catalase gene overexpres-

sion on life span and resistance to oxidative stress in transgenic
Drosophila melanogaster. Arch. Biochem. Biophys. 297:35– 41;
1992.

[46] Mockett, R. J.; Sohal, R. S.; Orr, W. C. Overexpression of

glutathione reductase extends survival in transgenic Drosophila
melagonaster. FASEB J. 13:1733–1742; 1999.

[47] Orr, W. C.; Sohal, R. S. Extension of life span by overexpression

of superoxide dismutase and catalase in Drosophila melano-
gaster. Science 263:1128 –1130; 1994.

[48] Sohal, R. S.; Mockett, R. J.; Orr, W. C. Current issues concerning

the role of oxidative stress in aging: a perspective. Results Prob.
Cell Differen. 29:45– 66; 2000.

[49] Sun, J.; Tower, J. FLP recombinase-mediated induction of Cu/

Zn-superoxide dismutase transgene expression can extend the life
span of adult Drosophila melanogaster flies. Mol. Cell. Biol.
19:216 –228; 1999.

[50] Stadtman, E. R.; Berlett, B. S. Reactive oxygen-mediated protein

oxidation in aging and disease. Chem. Res. Toxicol. 10:485– 494;
1997.

[51] Stadtman, E. R.; Levine, R. L. Protein oxidation. Ann. NY Acad.

Sci. 899:191–208; 2000.

[52] Levine, R. L.; Stadtman, E. R. Protein modifications with aging.

In: Schneider, E. L.; Rower, J. W., eds. Handbook of the biology
of aging. Orlando, FL: Academic Press; 1996:184 –197.

[53] Shacter, E. Quantification and significance of protein oxidation in

biological samples. Drug Metab. Rev. 32:302–326; 2000.

[54] Dubey, A.; Forster, M. J.; Sohal, R. S. Effect of age and caloric

intake on protein oxidation in different brain regions and on
behavioral functions of the mouse. Arch. Biochem. Biophys. 333:
189 –197; 1995.

[55] Agarwal, S.; Sohal, R. S. Aging and protein oxidative damage.

Mech. Aging Develop. 75:11–19; 1994.

[56] Forster, M.; Sohal, B. H.; Sohal, R. S. Reversible effects of

long-term caloric restriction on protein oxidative damage. J.
Geront. Biol. Sci. 55:B522–B529; 2000.

[57] Leeuwenburgh, C.; Wagner, P.; Hollosky, J. O.; Sohal, R. S.;

Heinecke, J. W. Caloric restriction attenuates dityrosine cross-
linking of cardiac and skeletal muscle proteins in aging mice.
Arch. Biochem. Biophys. 346:74 – 80; 1997.

[58] Stadtman, E. R. Metal ion-catalyzed oxidation of proteins: bio-

chemical mechanism and biological consequences [published er-
ratum appears in Free Radic. Biol. Med. 1991;10(3-4):249]. Free
Radic. Biol. Med. 9:315–325; 1990.

[59] Hekimi, S., ed. The molecular genetics of aging. Heidelberg,

Germany: Springer-Verlag; 2000.

[60] Yan, L.-J.; Levine, R. L.; Sohal, R. S. Oxidative damage during

aging targets mitochondrial aconitase. Proc. Natl. Acad. Sci. USA
94:11168 –11172; 1997.

[61] Yan, L.-J.; Sohal, R. S. Mitochondrial adenine nucleotide trans-

locase is oxidatively modified during aging. Proc. Natl. Acad. Sci.
USA 95:12896 –12901; 1998.

[62] Agarwal, S.; Sohal, R. S. Relationship between susceptibility to

protein oxidation, aging, and maximum life span potential of
different species. Exp. Geront. 31:365–372; 1996.

[63] Rothstein, M. Biochemical approaches to aging. New York:

Academic Press; 1982.

[64] Kanungo, M. S. Biochemistry of aging. London: Academic Press;

1980.

[65] Das, N.; Levine, R. L.; Orr, W. C.; Sohal, R. S. Selectivity of

protein oxidative damage during aging in Drosophila melano-
gaster. Biochem. J. 360:209 –216; 2001.

[66] Jana, C.; Das, N.; Sohal, R. S. Specificity of age-related carbony-

lation of plasma proteins in the mouse and rat. Arch. Biochem.
Biophys. 397:433– 439; 2002.

[67] Yan, L.-J.; Levine, R. L.; Sohal, R. S. Effects of aging and

hyperoxia on oxidative damage to cytochrome c in the housefly,
Musca domestica. Free Radic. Biol. Med. 29:90 –97; 2000.

[68] Yan, L.-J.; Sohal, R. S. Gel electrophoretic quantitation of protein

43

Oxidative stress and aging

carbonyls derivatized with tritiated borohydride. Anal. Biochem.
265:176 –182; 1998.

[69] Yan, L.-J.; Sohal, R. S. Prevention of flight activity prolongs the

life span of the housefly, Musca domestica, and attenuates the
age-associated oxidative damage to specific mitochondrial pro-
teins. Free Radic. Biol. Med. 29:1143–1150; 2000.

[70] Zahavi, M.; Tahori, A. S. Citric acid accumulation with age in

houseflies and other Diptera. J. Insect Physiol. 11:811– 816; 1965.

[71] Spencer, A. F.; Lowenstein, J. M. Citrate content of liver and

kidney of rat in various metabolic states and in fluoroacetate
poisoning. Biochem. J. 103:342–348; 1967.

ABBREVIATIONS

AL—ad libitum
CR— caloric restriction
MLS—maximum life span
OHdG— hydroxydeoxyguanine
PBN—N-tert-Butyl-

␣-phenylnitrone

ROS—reactive oxygen species
SOD—superoxide dismutase

44

R. S. S

OHAL