Full text of DOI:10.1093/gerona/56.3.B116

Journal of Gerontology: BIOLOGICAL SCIENCES
Copyright 2001 by The Gerontological Society of America
2001, Vol. 56A, No. 3, B116–B122
Effects of Caloric Restriction on Skeletal Muscle
Mitochondrial Proton Leak in Aging Rats
Shirin B. Lal,1 Jon J. Ramsey,2 Shadi Monemdjou,1 Richard Weindruch,2,3 and Mary-Ellen Harper1
1Department of Biochemistry, Microbiology and Immunology, University of Ottawa, Ontario, Canada.
2Wisconsin Regional Primate Research Center, University of Wisconsin, Madison.
3Department of Medicine, University of Wisconsin, and Veterans Administration Geriatric Research, Education and Clinical
Center, Madison.
Long-term caloric restriction (CR) retards aging processes and increases maximum life span.
We investigated the influence of CR on mitochondrial proton leaks in rat skeletal muscle. Be-
cause CR lowers oxidative damage to mitochondrial membrane lipids and proteins, we hypothe-
sized that leak would be lower in mitochondria from old CR rats than in age-matched controls.
Three groups (n ϭ 12) were studied: 4-month-old “young” control rats (body weight: 404 g Ϯ 7
SEM), 33-month-old CR rats (body weight: 262 g Ϯ 3), and 33-month-old control rats (body
weight: 446 g Ϯ 5). CR rats received 67% of the energy intake of old control rats, with adequate
intakes of all essential nutrients. Maximum leak-dependent O2 consumption (State 4) was 23%
lower in CR rats than in age-matched controls, whereas protonmotive force values were similar,
supporting our hypothesis. The overall kinetics of leak were similar between the two groups of
old rats; in the young, kinetics indicated higher protonmotive force values. The latter indication
is consistent with aging-induced alterations in proton leak kinetics that are independent of di-
etary intervention. There was no influence of age or diet on serum T4 level, whereas T3 was
lower in young than in old control rats. These results support and extend the oxidative stress hy-
pothesis of aging.
N increase in oxidative stress has long been thought to
be one of the causes of aging (1). The basic premise of
from the mitochondria of isolated rat hepatocytes was
greater by old cells than by cells from young rats. Also, it
has been shown that caloric restriction (CR) prevents the
age-associated accumulation of oxidative damage to mouse
skeletal muscle mitochondria (7).
A
the oxidative stress hypothesis of aging is that age-related
loss of function is due to progressive and irreversible accu-
mulation of molecular oxidative damage (2). Further, it is
thought that longer life spans can be achieved by decreasing
the level of oxidative damage incurred over a lifetime (2).
Recent studies in our laboratory have shown that there is
an increased mitochondrial proton leak in hepatocytes of
old versus young mice, as well as altered kinetics of adeno-
sine triphosphate (ATP) turnover reactions and a modified
distribution of metabolic control in oxidative phosphoryla-
tion reactions (3). Oxidative damage to mitochondrial mem-
branes has been implicated in an augmented generation of
reactive oxygen species (ROS), which, in turn, can lead to
further membrane damage. The implications of such a vi-
cious cycle of oxidative stress and damage have been de-
scribed by in vitro studies such as those of Sohal and Sohal
(4), who showed that mitochondrial H2O2 generation in-
creases after exposure to 2,2-azobis(2-amino-propane)dihy-
drochloride.
These recent findings have led to our current investiga-
tion of the effects of CR on age-related changes in proton
leak in rat skeletal muscle. Thyroid hormone status in these
rats was also assessed, as thyroid hormones increase proton
leak and maximum leak-dependent mitochondrial O2 con-
sumption (8–10). Restricting the caloric intake of mammals
is a well-recognized means of extending maximum life
span. Evidence indicates that CR may act, at least in part, by
decreasing oxidative stress and increasing antioxidant de-
fenses and repair mechanisms (2,11). With the use of this
evidence and that of other studies (3,5–7), we hypothesize
that the beneficial effects on aging of CR may be directly
due to lowered rates of mitochondrial proton leak.
Oxidative phosphorylation is the overall process through
which the oxidation of cellular energy substrates is coupled
to ATP synthesis in mitochondria. The activity of the mito-
chondrial respiratory electron transport chain generates the
protonmotive force (⌬p) across the mitochondrial inner
membrane. It is this electrochemical gradient that is used to
drive protons back into the mitochondrial matrix. Protons
are returned into the matrix via two separate pathways: ATP
synthase and mitochondrial proton leak (Figure 1). ATP
synthase catalyses the phosphorylation of adenosine diphos-
Brookes and colleagues (5) have shown that sequential
additions of 200 ␮M of peroxynitrite to rat brain mitochon-
dria significantly stimulated proton leak. Importantly, the
stimulation of proton leak was inhibited by Trolox, a vita-
min E analog, thus supporting the involvement of lipid per-
oxidation in the mechanism of peroxynitrite cytotoxicity.
Hagen and colleagues (6) reported that oxidant generation
B116

MITOCHONDRIAL PROTON LEAK IN AGING RATS
B117
ing relationships among mitochondrial fatty acid composi-
tion, mitochondrial proton leak, and life span.
The findings of this investigation show that maximum
leak-dependent O2 consumption in skeletal muscle mito-
chondria from old rats is significantly lower in those sub-
jected to CR than in control-fed rats. These results support
and extend the oxidative stress hypothesis of aging.
Figure 1. The oxidative phosphorylation system in mitochondria.
The intermediate in the system, protonmotive force (⌬p), is produced
by the substrate oxidation branch of reactions; ⌬p is consumed by the
proton leak and phosphorylating branches of reactions. The proton
leak branch consists of the leak of protons and any cation cycles
across the mitochondrial inner membrane. The phosphorylation
branch consists of the synthesis of ATP (adenosine triphosphate) by
means of ATP synthase and ATP consumption.
METHODS
Treatment of Animals
Male Wistar rats were studied at either 4 or 33 months of
age. The 4-month-old young controls were obtained from
Charles River (St. Constant, PQ) and studied 2–3 weeks af-
ter arrival in Ottawa. These rats were allowed free access to
Purina 5000 Chow (PMI Feeds, St. Louis, MO). The 33-
month-old rats (old control and CR rats) were obtained from
the specific pathogen-free Shared Aging Rodent Facility at
the Veterans Administration Geriatric Research, Education
and Clinical Center (Madison, WI). These rats were initially
purchased at 5 weeks of age from the Lobund Laboratories
at Notre Dame University. Upon arrival in Madison they
were given free access to Purina 5000 Chow.
At approximately 10 months of age, the latter group of
rats were randomly divided into two groups receiving either
a control or energy-restricted diet (Teklad 91349 control
diet and Teklad 91351 CR diet; Harlan Teklad, Madison,
WI). The diet compositions are summarized in Table 1. The
amount of protein, minerals, and vitamins was elevated in
the diet of CR animals in order to ensure that the rats were
not deficient in these nutrients. The energy intakes were
43.6 kcal/day for the controls and 29.2 kcal/day for the CR
rats. For waste to be minimized during feeding, the pow-
dered diets were suspended in a 0.5% agar solution to pro-
duce a semisolid diet. The rats were caged individually at
23ЊC with light from 7 am to 7 pm. All rats were allowed
free access to water. All animals were cared for in accor-
dance with the guidelines of the Canadian Council on Ani-
mal Care, the Institute of Laboratory Animal Resources
(National Research Council, USA), and with “Guiding Prin-
ciples for Research Involving Animals and Human Beings.”
phate (ADP) (transfer of energy) as protons flow down their
gradient. Conversely, the proton leak pathways lead to de-
creased ⌬p without the synthesis of ATP, but with the pro-
duction of heat. Energy is dissipated as the activity of the
mitochondrial respiratory chain increases in response to de-
creased ⌬p. Proton leak pathways are thought to be modu-
lated by thyroid hormones (8–10) and phylogeny (12), and
they are inversely related to body mass in mammals (13).
CR has consistently been shown to increase maximum
life span by 20–40%, with the magnitude of life-span exten-
sion showing an inverse relationship with energy intake
(14). For an increased life span, however, it is crucial that
animals on CR diets receive adequate amounts of protein,
vitamins, and minerals. In this study, we aimed to examine
mitochondrial proton leak in oxidative skeletal muscle tis-
sue, because the greatest attenuation of oxidative damage
caused by CR appears to occur in postmitotic tissues such as
skeletal muscle, brain, and heart (2,15).
Recent reviews describe age-related changes that occur
in mitochondria. Changes include oxidative damage to
mtDNA, oxidation of mitochondrial proteins, decreased
membrane fluidity, and oxidation of membrane lipids
(2,16,17). However, in our previous study of age-related
changes in the oxidative phosphorylation system in hepato-
cytes of mice, we found no significant effects of age on the
overall kinetics of substrate oxidation reactions (3). Sub-
strate oxidation reactions comprised reactions involved in,
and following, the oxidation of glucose, lactate, pyruvate,
and endogenous substrates; it thus includes components of
the electron transport chain—many of which are encoded in
mtDNA.
Our previous findings did nonetheless show that the over-
all kinetics of mitochondrial proton leak were altered such
that leak-dependent oxygen consumption was higher in
hepatocytes from old versus young mice (3). Membrane
lipid damage is likely an important factor contributing to the
increased proton leak with aging, but we cannot exclude
other potential mechanisms. Over the life span of an animal,
there is a natural decline in the amount of linoleic acid
(18:2) and concomitant increases in the amounts of long-
chain polyunsaturated fatty acids (22:4 and 22:5), resulting
in an increased probability of membrane lipid peroxidation
(18,19). Thus, although mechanistic aspects of mitochon-
drial proton leak are as yet unclear, there are several intrigu-
Table 1. Composition of the Diets for the Old Control and Calorie
Restricted Groups
Control Diet
(g/100 g Diet)
Restricted Diet
(g/100 g Diet)
Ingredients
Casein
DL-methionine
Sucrose
Corn starch
Corn oil
Cellulose
Mineral mix, AIN-76
Calcium carbonate
Vitamin mix
Brewer’s yeast
22.0
0.3
27.0
27.0
13.5
5.0
3.5
0.3
1.0
0.4
27.7
0.4
23.5
23.5
13.5
4.9
4.4
0.4
1.3
0.5
Note: The mineral mix is Teklad 170915 and the vitamin mix is Teklad
40060 (Harlan Teklad, Madison, WI).

B118
LAL ET AL.
Isolation of Hind-Limb Skeletal Muscle Mitochondria
Rats were anesthesized with intraperitoneal injections of
0.325 mg of sodium pentobarbital per gram of body weight
prior to decapitation. Skeletal muscle mitochondria were iso-
lated as described by Bhattacharya and colleagues (20). The
isolated mitochondria were resuspended in a medium con-
taining the following: 120 mM of KCl, 20 mM of sucrose, 20
mM of glucose, 10 mM of KH2PO4, 5 mM of N-2-Hydroxy-
ethylpiperazine-NЈ-ethanesulphonic acid (HEPES), 2 mM of
MgCl2, 1 mM of (Ethylenedioxy) diethylenedinitrilotetrace-
tic acid (EGTA), and 0.5% defatted bovine serum albumin
(BSA), with a pH of 7.2 with potassium hydroxide (KOH).
BSA was defatted by using the method of Chen (21). Mito-
chondrial protein concentration was determined by the Biuret
reaction.
rection factor used to correct for potential-independent
binding of TPMPϩ to mitochondria.
All measurements of ⌬p were performed in duplicate and
simultaneously to the above-described measurements of ox-
ygen consumption.
Determination of Thyroid Hormone Status
Serum T3 and T4 assays were performed by using 125I-Radio
Immuno-Assay kits from INCSTAR corporation (Stillwa-
ter, MN).
Materials.—Oligomycin, valinomycin, succinate, mal-
onate, ADP, and ATP were obtained from Sigma Chemicals
(St. Louis, MI). TPMPиBr was obtained from Aldrich (Mil-
waukee, WI). All radioactive compounds were obtained
from Mandel-Dupont NEN (Guelph, ON).
Measurement of Mitochondrial Oxygen Consumption
Mitochondrial oxygen consumption was measured by us-
ing a Clark type of oxygen electrode (Hansatech, Norfolk,
UK) maintained at 37ЊC. Mitochondria (0.5 mg of protein/
ml) were incubated in 1 mL of medium containing 120 mM
of KCl, 20 mM of sucrose, 20 mM of glucose, 10 mM of
KH2PO4, 5mM of HEPES, 2 mM of MgCl2, 1 mM of EGTA,
and 0.5% defatted BSA, with a pH of 7.2 with KOH, plus 5
␮M of rotenone and 0.4 ␮g of nigericin. Rotenone was
added to inhibit oxidation of endogenous reduced nicotina-
mide adenine dinucleotide (NADH)-linked substrates. Ni-
gericin is added to convert the ⌬pH component of the mi-
tochondrial protonmotive force to potential; thus, our
assessments of mitochondrial membrane potential, de-
scribed below in the following paragraphs, reflect total pro-
tonmotive force. Respiration was initiated with 10 mM of
succinate and 100 ␮M of ATP. State 4 oxygen consump-
tion, defined as maximum nonphosphorylating respiration,
was achieved by using saturating amounts of oligomycin
(12 ␮g/mg of mitochondrial protein) to completely inhibit
ATP synthase, and hence the phosphorylation branch of ox-
idative phosphorylation (see Figure 1). The kinetics of the
leak (i.e., response of leak-dependent oxygen consumption
to imposed changes in protonmotive force) were subse-
quently determined by progressively inhibiting the electron
transport chain with malonate (an inhibitor of Complex II)
to give final concentrations between 0.2 and 3.17 mM. All
measurements were performed in duplicate.
Statistical analysis.—Data were analyzed by using an
analysis of variance (ANOVA) followed by Tukey’s post
hoc tests. All data are reported as mean Ϯ standard error of
the mean. A value of p Ͻ .05 was considered statistically
significant.
RESULTS
Body and Tissue Weights
Mean body weight was, as expected, significantly lower
in old CR than old control rats. Body weight was 41% lower
and approximates the 33% CR. Weights of epididymal
white adipose tissue (EWAT) depots were determined as an
index of adiposity. Both groups of older rats had higher
mean total weights of EWAT than did the young controls
(Table 2). However, if EWAT weights are normalized to the
mean body weights, there are no significant differences be-
tween the two groups of old rats.
Because of the importance of brown adipose tissue ther-
mogenesis in rodents with respect to energy balance, the
weights of interscapular brown adipose tissue (IBAT) de-
pots were also measured. The mean IBAT weight was sig-
nificantly lower in old CR rats than in both other groups
(Table 2). For example, the mean IBAT weight of the old
CR rats was only 21% that of age-matched controls. The
mean IBAT weight was significantly greater in old control
Measurement of Mitochondrial Protonmotive
Force (⌬p)
Table 2. Body and Fat Pad Weights for Old Calorie Restricted, Old
Control, and Young Control Groups
⌬p was determined by using a triphenylmethylphospho-
nium (TPMPϩ)-sensitive electrode (22), which was inserted
into the incubation chamber of the oxygen electrode as de-
scribed by Brown and Brand (23). Mitochondrial matrix
volume and TPMPϩ potential-independent binding correc-
tion factor (am) were also determined as described by Brown
and Brand (23). ⌬p can be calculated by using a modified
Nernst equation as follows: ⌬p ϭ 61.5 ϫ log({[TPMPϩ]mat
ϫ am}/[TPMPϩ]ext), where [TPMPϩ]mat refers to the con-
centration of TPMPϩ in the mitochondrial matrix, [TPMPϩ]ext
refers to the concentration of TPMPϩ in the medium sur-
rounding the mitochondria, and am refers to the binding cor-
Body Weight Total EWAT
Total IBAT
(g)
Group
Old
(g)
(g)
Calorie restricted (n ϭ 12)
Control (n ϭ 12)
Young control (n ϭ 12)
262.4 Ϯ 3.1
446.0 Ϯ 4.7
404.3 Ϯ 6.9
0.55 Ϯ 0.04
0.94 Ϯ 0.06
0.35 Ϯ 0.03
1.13 Ϯ 0.17
6.12 Ϯ 0.16
4.94 Ϯ 0.45
Notes: EWAT ϭ epididymal white adipose tissue; IBAT ϭ interscapular
brown adipose tissue. Results are presented as mean Ϯ SE. Within each column,
results from each group of rats were significantly different from every other
group, as determined by the Tukey multiple comparison test. Overall analysis of
variance, p Ͻ .0001; for comparisons between groups, all comparisons yielded
values of p Յ .01.

MITOCHONDRIAL PROTON LEAK IN AGING RATS
B119
than in young control rats, and it probably reflects a greater
proportion of triglyceride in the depots of old rats.
This was not observed in the data of the mitochondria from
old CR or young controls (Figures 2A, and 2C) and is sur-
prising, as all mitochondria were isolated by using the same
protocol.
Mitochondrial Proton Leak
Maximum leak-dependent oxygen consumption (State 4)
was achieved with the addition of a saturating concentration
of oligomycin (12 ␮g/mg of mitochondrial protein). The fur-
thest point on the right of each curve identifies maximum
State 4 oxygen consumption. It was found that the maximum
State 4 oxygen consumption rate was 23% lower (p Ͻ .05)
in mitochondria from old CR rats (Figure 2A) than that from
old control rats (Figure 2B). The actual mean values were
72.5 Ϯ 6.4 (n ϭ 8), 59.0 Ϯ 3.3 (n ϭ 10), and 66.8 Ϯ 5.3
(n ϭ 9) nmol of O2 per milligram of mitochondrial protein
per minute for old control, CR, and young control groups, re-
spectively. These values are presented in Figure 3.
Figure 2 (parts A, B, and C) further depicts the overall ki-
netics of proton leak reactions. Overall leak kinetics were
not markedly different between the two groups of older rats.
However, certain observations in comparison of results
from the old and the young are of interest. Overall leak ki-
netics over a range of oxygen consumption values show that
⌬p is higher in young than in old rats. This is particularly
evident in the comparison between results of young controls
and those from old CR rats. These results indicate that the
oxygen used to balance the leak over a range of ⌬p is lower
(i.e., the leak is lower) in mitochondria from young than
from old control rats, regardless of dietary intervention.
Again, this supports the hypothesis of age-related increases
in mitochondrial proton leaks.
Thyroid Hormone Levels
Thyroid hormone levels (Table 3) were assessed in serum
collected at the time of sacrifice. Average T3 levels were
significantly higher (p Ͻ .05) in old control than in young
control rats. However, average values for CR rats were not
significantly different from either group. Average T4 levels
were not significantly different between groups.
DISCUSSION
We hypothesized that CR lowers the accrual with aging
of oxidative damage to the mitochondrial inner membrane,
which, in turn has been proposed as a cause of age-associ-
ated increases in mitochondrial proton leak (3). Our previ-
ous study (3) showed that mitochondrial proton leak was
significantly higher in hepatocytes of old mice compared
with cells from young mice. The differences were relatively
small; thus a study of leak in a postmitotic tissue seemed ap-
propriate. Skeletal muscle was selected as the site of study
because oxidative stress and mitochondrial dysfunction may
contribute to losses in muscle mass and quality (24).
The age-associated increase in the amount of oxidative
stress can be ascribed to three main factors: an increase in
the rate of generation of reactive oxygen metabolites such
as H2O2, O2иϪ, and OH; a decline in protective antioxida-
и
tive capabilities; and a decline in the efficiency of repair and
removal of damaged molecules. A hypothesis that requires
further study is that increases in mitochondrial proton leak,
It should also be noted that there was a large variation in
data from the mitochondria of old control rats (Figure 2B).
Figure 2. Relationship between protonmotive force (⌬p) and oxygen consumption of hind-limb skeletal muscle mitochondrial proton leaks in
A, old calorie restricted, B, old control, and C, young control rats. Maximum leak-dependent oxygen consumption (State 4) is the furthest point
on the right of the graph (᭡). All measurements were made in the presence of a saturating amount of oligomycin (12 ␮g/mg of mitochondrial
protein). Overall kinetics of the proton leak (᭿) were determined by titrating ⌬p producers (i.e., substrate oxidation reactions) with increasing
amounts of malonate (A: 0.2, 0.33, 0.53, 0.67, 0.86, 1.0, 1.19, 2.0, 2.18, 3.0, and 3.17 mM; B: 0.2, 0.53, 0.67, 0.86, 1.0, 1.19, and 3.17 mM; C: 0.2, 0.33,
0.53, 0.67, 0.86, 1.0, 1.19, 2.0, 2.18, and 3.0 mM). All points represent mean Ϯ SEM of at least five (A), eight (B), and six (C) duplicate assays.

B120
LAL ET AL.
from old CR rats versus old control rats. The decrease in ox-
ygen consumption at State 4 would lead to a decreased gen-
eration of free radical species at the level of the electron
transport chain. It has been shown that mitochondria that are
damaged by oxidative stress generate greater amounts of
H2O2 than nonoxidized mitochondria (4). This leads to a vi-
cious cycle wherein mitochondria that are damaged by oxi-
dative stress become more “leaky” and would then generate
more ROS and thus become further impaired. The fact that
protonmotive force was higher in mitochondria from young
rats than in CR and old control rats is to be noted. This is
presumably an effect of aging that is independent of CR.
Importantly, CR seems to exert its effect through changing
mitochondrial oxygen consumption under State 4 condi-
tions. This suggests that CR may act to increase life span in
part by decreasing oxidative stress (2,25,26), and decreased
proton leak-dependent oxygen consumption through CR
may be one of the mechanisms that extends life span.
Maintenance of normal mitochondrial membrane lipid
composition is an important indication of the health of mito-
chondria. The majority of the age-related changes observed
in mitochondrial membrane lipid composition occur in car-
diolipin (27). Cardiolipin interacts with various mitochon-
drial membrane proteins, including ATP synthase and
ADP–ATP translocator, and it plays an important role in
maintaining their activity (27). Cardiolipin also appears to
play a major role in controlling the permeability of the mito-
chondrial membrane to small molecules and in establishing
proton gradients (27). CR in rodents has been shown to
maintain the amount of 18:2 acyl side chains in the mito-
chondrial membrane and inhibits the oxidation of cardio-
lipin (18). These effects may minimize proton leak in CR
animals and maintain mitochondrial inner membrane integ-
rity. Thus, our observed lower state 4 oxygen consumption
values in CR rats versus old control rats may be an indica-
tion of allayed mitochondrial membrane lipid oxidation in
these rats.
Figure 3. State 4 oxygen consumption of hind-limb skeletal muscle
mitochondria from old control, calorie restricted, and young control
rats. State 4 (maximum leak dependent) oxygen consumption was
measured as that in the presence of a saturating amount of oligomy-
cin (12 ␮g/mg of mitochondrial protein). Different symbols indicate
significant differences (p Ͻ .05) in means by an analysis of variance
with Tukey post hoc tests. Means were calculated by using n ϭ 8 for
old controls, n ϭ 10 for calorie restricted rats, and n ϭ 9 for young
controls; each rat muscle preparation was assessed in duplicate.
may be both a consequence and a cause of oxidative stress,
as an increased leak results in increased mitochondrial oxy-
gen consumption and thus possibly subsequent oxidant pro-
duction (2,3,5). This could then result in further damage to
lipids and proteins in the mitochondrial and extramitochon-
drial environment.
Mitochondria constitute the greatest source of oxidants in
the cell, as they account for 85% of cellular oxygen con-
sumption and oxidants are continually being produced in
the mitochondria as a by-product of substrate oxidation.
Rats may consume up to 1012 molecules of O2 per day and,
because it is estimated that 2% of these oxygen molecules
are not completely reduced to water, this results in the daily
иϪ
generation of 2 ϫ 1010 molecules of H2O2 and O2 (17).
Surprisingly, we did not find increased State 4 oxygen
consumption in old versus young controls. This contradicts
earlier findings from studies conducted in hepatocytes of
mice (3) and rats (6). However, we were surprised in this
study by the fairly large degree of variation in data within
groups, and particularly within the old control group. This
may be related to the fact that mitochondria from old rats
are more fragile than those from young rats and are thus
more easily damaged during isolation (28). Mitochondrial
preparations from old control rats may thus have contained
variable proportions of unhealthy mitochondria, leading to
the observed variable results. Moreover, it is also possible
that there are species differences in oxidant generation (29),
which affect differences in proton leak between rats and
mice. Had we a higher number of observations per group, it
is likely that statistically significant observations would be
obtained. The relative degree to which mitochondria are
damaged during their isolation from old and young species
is a subject that has received little attention, and more de-
tailed studies (e.g., balance studies of marker enzymes) are
needed.
Mass-specific basal metabolic rate is inversely correlated
with life span in mammals (2); thus, smaller mammals have
higher mass-specific metabolic rates and shorter life spans
than larger mammals. There is also, as mentioned earlier,
the intriguing inverse correlation between mitochondrial
proton leak and body mass in mammals (13).
In this investigation of mitochondrial proton leak in the
skeletal muscle of CR, old control, and young rats, we ob-
served that the maximum leak-dependent oxygen consump-
tion was decreased by 23% in skeletal muscle mitochondria
Table 3. Serum Thyroid Hormone Levels of Old Calorie
Restricted, Old Control, and Young Control Groups
Group
(T3) (nmol)
(T4) (nmol)
Old
Calorie restricted (n ϭ 12)
Control (n ϭ 12)
Young control (n ϭ 12)
1.07 Ϯ 0.08
1.13 Ϯ 0.07*
0.83 Ϯ 0.05
52.2 Ϯ 5.0
50.9 Ϯ 2.1
59.4 Ϯ 4.8
Notes: T3 ϭ L-3,3Ј,5-triiodothyronine; T4 ϭ thyroxine. Results are pre-
sented as mean Ϯ SE.
As mentioned earlier, one factor known to affect mito-
chondrial proton leak is thyroid hormone status (8,9,30,31).
*Significant difference (p Ͻ .05) from young control rats.

MITOCHONDRIAL PROTON LEAK IN AGING RATS
B121
number 01-02 from the Madison Veterans Administration Geriatric Re-
search, Education and Clinical Center.
Although thyroxine (T4) levels tend to remain constant with
age in rodents, increases in L-3, 3Ј, 5-triiodothyronine (T3)
levels, such as we saw, have been observed (32). This has
been attributed to defects in the ability of thyroid stimulat-
ing hormone to adequately regulate thyroid hormone levels
(32). Herlihy and colleagues (33) found that CR in rats de-
creases 24-hour mean T3 levels, but that T4 levels are un-
changed. However, we did not observe differences in serum
T3 between old CR and old control rats. It is interesting that
the greatest values of leak-dependent oxygen consumption
and of serum T3 were observed for the old control group,
because thyroid hormone levels have been positively corre-
lated with mitochondrial proton leak (8,9,30,31).
The specific mechanism for the observed increased proton
leak with age has not yet been identified, though, as outlined
above, evidence suggests a mechanism involving membrane
lipid damage. Another obvious possibility is through the
modified activity of the recently identified uncoupling pro-
teins (UCPs). UCPs are mitochondrial inner membrane pro-
teins that may cause a proton leak. Four new UCPs have
been cloned, and these include UCP2 (34,35), UCP3 (36,
37), UCP4 (38), and BMCP1 (39). Although their homolo-
gies to UCP1, and their ability to uncouple oxidative phos-
phorylation in a variety of in vitro expression systems, sug-
gest that they might be uncoupling proteins, the mechanisms
involved and their physiological functions are as yet un-
known. UCP2 is expressed ubiquitously, with highest levels
of expression in brown and white adipose tissues. UCP3 is
expressed almost exclusively in skeletal muscle in humans,
but also in the brown adipose tissue of rodents. UCP4 is ex-
pressed exclusively in the brain, whereas BMCP1 is ex-
pressed predominantly in the brain but is also expressed at
lower levels (10- to 30-fold) in many other tissues. Thus, an
increased UCP-mediated leak during aging is another possi-
ble mechanism for an increased proton leak. It is also impor-
tant that one of the several putative functions for the novel
UCPs is decreased production of ROS (40,41). The mecha-
nism proposed is that UCPs allow increased electron flow
during nonphosphorylating states, which in turn may de-
crease the relative reduction of redox centers in Complex I
and of cytochrome b in Complex III, and thus decrease the
probability of electron transfer to molecular oxygen.
Address correspondence to Dr. M.-E. Harper, Department of Biochem-
istry, Microbiology and Immunology, Faculty of Medicine, University of
Ottawa, 451 Smyth Road, Ottawa, Ontario, K1H 8M5, Canada. E-mail:
mharper@uottawa.ca
References
1. Harman D. Aging: a theory based on free radical and radiation chemis-
try. J Gerontol. 1956;2:298–300.
2. Sohal RS, Weindruch R. Oxidative stress, caloric restriction and ag-
ing. Science. 1996;273:59–63.
3. Harper M-E, Monemdjou S, Ramsey JJ, Weindruch R. Age-related in-
crease in mitochondrial proton leak and decrease in ATP turnover
reactions in mouse hepatocytes. Am J Physiol. 1998;275:E197–E206.
4. Sohal RS, Sohal BH. Hydrogen peroxide release by mitochondria in-
creases during aging. Mech Ageing Dev. 1991;57:187–202.
5. Brookes PS, Lands J, Clark J, Heales S. Peroxynitrite and brain mito-
chondria: evidence for increased proton leak. J Neurochem. 1998;70:
2195–2202.
6. Hagen TM, Yowe DL, Bartholomew JC, et al. Mitochondrial decay in
hepatocytes from old rats: membrane potential declines, heterogeneity
and oxidants increase. Proc Natl Acad Sci USA. 1997;94:3064–3069.
7. Lass A, Sohal BH, Weindruch R, Forster MJ, Sohal RS. Caloric restric-
tion prevents age-associated accrual of oxidative damage to mouse
skeletal muscle mitochondria. Free Rad Biol Med. 1998;25:1089–
1097.
8. Hafner RP, Nobes CD, McGown AD, Brand MD. Altered relationship
between protonmotive force and respiration rate in non-phosphorylat-
ing liver mitochondria isolated from rats of different thyroid hormone
status. Eur J Biochem. 1988;178:511–518.
9. Hafner RP, Leake MJ, Brand MD. Hypothyroidism in rats decreases
mitochondrial inner membrane cation permeability. FEBS Lett. 1989;
248:175–178.
10. Horrum MA, Tobin RA, Ecklund RE. Thyroid hormone effects of the
proton permeability of rat liver mitochondria. Mol Cell Endocrinol.
1990;41:163–169.
11. Yu BP. Aging and oxidative stress: modulation by dietary restriction.
Free Rad Biol Med. 1996;5:651–668.
12. Brand MD, Couture P, Else PL, Withers KW, Hulbert AJ. Evolution
of energy metabolism. Proton permeability of the inner membrane of
liver mitochondria is greater in a mammal than in a reptile. Biochem J.
1991;275:81–86.
13. Porter RK, Brand MD. Body mass dependence of Hϩ leak in mito-
chondria and its relevance to metabolic rate. Nature. 1993;362:628–
630.
14. Weindruch R, Walford RL. The Retardation of Aging and Disease by
Dietary Restriction. Springfield, IL: Charles C Thomas; 1988.
15. Davies KJ. Oxidative stress: the paradox of aerobic life. Biochem Soc
Symp. 1995;61:1–31.
16. Shigenaga M, Hagen T, Ames BN. Oxidative damage and mitochon-
drial decay in aging. Proc Natl Acad Sci USA. 1994;91:10771–10778.
17. Ames BN, Shigenaga M, Hagen T. Oxidants, antioxidants, and the de-
generative diseases of aging. Proc Natl Acad Sci USA. 1993;90:7915–
7922.
18. Laganiere S, Yu BP. Modulation of membrane phospholipid fatty acid
composition by age and food restriction. Gerontology. 1993;39:7–18.
19. Yu BP, Suescun E, Yang S. Effect of age-related lipid peroxidation on
membrane fluidity of phospholipase A2: modulation by dietary restric-
tion. Mech Ageing Dev. 1992;65:17–33.
20. Bhattacharya SK, Thakar JH, Johnson PL, Shanklin DR. Isolation of
skeletal muscle mitochondria from hamsters using an ionic medium
containing ethylenediaminetetraacetic acid and nagarse. Anal Bio-
chem. 1991;192:344–349.
21. Chen RF. Removal of fatty acids from serum albumin by charcoal
treatment. J Biol Chem. 1967;242:173–181.
The major finding from this study is that proton leak-
dependent oxygen consumption in hind-limb skeletal mus-
cle of rats is decreased in very old rats subjected to CR from
10 months of age. These results support previous findings,
specifically those that demonstrate increased proton leak
and oxidative stress with age (3,6). The recent results of
Lass and colleagues (7) show that CR attenuates a 41% age-
associated increase in superoxide production in submito-
chondrial particles isolated from mouse muscle, and these
results are consistent with our findings. Although the mech-
anisms involved still require elucidation, decreased mito-
chondrial proton leak may be an important factor in the re-
tardation of aging by CR.
22. Kamo N, Muratsugu M, Hongoh R, Kobatake Y. Membrane potential
of mitochondria measured with an electrode sensitive to tetraphenyl
phosphonium and relationship between proton electrochemical poten-
tial in steady state. J Memb Biol. 1979;49:105–121.
23. Brown GC, Brand MD. Thermodynamic control of flux through mito-
chondrial cytochrome bc1 complex. Biochem J. 1985;225:399–405.
Acknowledgments
This work was supported by grants from the Natural Science and Engi-
neering Research Council (NSERC) of Canada (MEH) and the National In-
stitutes of Health (P01 AG 11915; T32 AG 00213). This is publication

B122
LAL ET AL.
24. Weindruch R. Interventions based on the possibility that oxidative
stress contributes to sarcopenia. J Gerontol Biol Sci. 1995;50A:B157–
B161.
34. Fleury C, Neverova M, Collins S, et al. Uncoupling protein 2: a novel
gene linked to obesity and hyperinsulinemia. Nature Genet. 1997;15:
269–272.
25. Yu BP, Yang R. Critical evaluation of the free radical theory of aging.
A proposal for the oxidative stress hypothesis. Ann NY Acad Sci. 1996;
786:1–11.
35. Gimeno RE, Dembski M, Weng X, et al. Cloning and characterization
of and uncoupling protein homolog: a potential mediator of human
thermogenesis. Diabetes. 1997;46:900–906.
26. Sohal RS, Ku HH, Agarwal S, Forester MJ, Lal H. Oxidative damage
mitochondrial oxidant generation and antioxidant defenses during ag-
ing in response to food restriction in the mouse. Mech Ageing Dev.
1994;74:121–130.
27. Hoch F. Cardiolipins and biomembrane function. Biochim Biophys
Acta. 1992;1113:71–133.
28. Wilson PD, Franks LM. The effect of age on mitochondrial ultrastruc-
ture and enzymes. Adv Exp Med Biol. 1975;53:171–183.
29. Sohal RS, Sevensson I, Brunk UT. Relationship between antioxidant
defenses and longevity in different mammalian species. Mech Ageing
Dev. 1990;53:209–215.
30. Harper M-E, Brand MD. Hyperthyroidism stimulates mitochondrial
proton leak and ATP turnover in rat hepatocytes but does not change
the overall kinetics of substrate oxidation reactions. Can J Physiol
Pharmacol. 1994;72:899–908.
36. Boss O, Samec S, Paoloni-Giacobino A, et al. Uncoupling protein 3: a
new member of the mitochondrial carrier family with tissue specific
expression. FEBS Lett. 1997;408:39–42.
37. Vidal-Puig A, Solanes G, Grujic D, Flier JS, Lowell BB. UCP3: an un-
coupling protein homologue expressed preferentially and abundantly
in skeletal muscle and brown adipose tissue. Biochem Biophys Res
Comm. 1997;235:79–82.
38. Mao W, Yu XX, Zhong A, et al. UCP4, a novel brain-specific mito-
chondrial protein that reduces membrane potential in mammalian
cells. FEBS Lett. 1999;443:326–330.
39. Sanchis D, Fleury C, Chomiki N, et al. BMCP1, a novel mitochondrial
carrier with high expression in the central nervous system of humans
and rodents, and respiration uncoupling activity in recombinant yeast.
J Biol Chem. 1998;273:34611–34615.
40. Negre-Salvayre A, Hirtz C, Carrera G, et al. A role for uncoupling pro-
tein-2 as a regulator of mitochondrial hydrogen peroxide generation.
FASEB J. 1997;11:809–815.
41. Boss O, Muzzin P, Giacobino J-P. The uncoupling proteins, a review.
Eur J Endocrinol. 1998;139:1–9.
31. Harper M-E, Brand MD. The quantitative contributions of mitochon-
drial proton leak and ATP turnover reactions to the changed
respiration rate of hepatocytes from rats of different thyroid status. J
Biol Chem. 1993;268:14850–14860.
32. Hotta H, Ooka H, Sato A. Changes in basal secretion rates of thyrox-
ine and 3,3Ј,5-triiodothyroxine from the thyroid gland during aging of
the rat. Jpn J Physiol. 1991;41:75–84.
33. Herlihy JT, Stacy C, Bertrand HA. Long-term food restriction de-
presses serum thyroid hormone concentrations in the rat. Mech Ageing
Dev. 1990;53:9–16.
Received February 14, 2000
Accepted October 3, 2000
Decision Editor: John A. Faulkner, PhD