| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,¶1
,§,1
* Departments of Nutritional Sciences,
Pathology and Laboratory Medicine,
Medicine, and the
§ Wisconsin Regional Primate Research Center, University of Wisconsin, Madison, Wisconsin 53705, USA;
¶ Pathology and Laboratory Medicine Service and
Geriatric Research, Education and Clinical Center, William S. Middleton VA Hospital, Madison, Wisconsin 53705, USA;
Obesity Research Center, St. Luke’s/Roosevelt Hospital, Columbia University College of Physicians & Surgeons, New York, New York; and
** Department of Physiology and Biophysics, School of Medicine, Case Western Reserve University, Cleveland, Ohio 44106, USA
1Correspondence: T.D.O., Pathology and Laboratory Medicine Service, Room A35, William S. Middleton Memorial Veterans Hospital, 2500 Overlook Terr., Madison, WI 53705, USA. E-mail: toberley{at}facstaff.wisc.edu; R.W., Department of Medicine, William S. Middleton Memorial Veterans Hospital (GRECC 4D), 2500 Overlook Terr., Madison, WI 53705, USA. E-mail: rhweindr{at}facstaff.wisc.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
twofold with
aging. Comparing 17- to 23-year-old normally fed to age-matched monkeys
subjected to CR for 10 years, levels of HNE-modified proteins,
carbonyls, and nitrotyrosine in skeletal muscle from the CR group were
significantly less than control group values. Oxidative damage largely
localized to myofibrils, with lesser labeling in other subcellular
compartments. Accumulation of lipid peroxidation-derived aldehydes,
such as malondialdehyde and 4-hydroxy-2-alkenals, and protein carbonyls
were measured biochemically and confirmed the morphological data. Our
study is the first to quantify morphologically and localize the
age-dependent accrual of oxidative damage in mammalian skeletal muscle
and to demonstrate that oxidative damage in primates is lowered by
CR.—Zainal, T. A., Oberley, T. D., Allison, D. B.,
Szweda, L. I., Weindruch, R. Caloric restriction of rhesus monkeys
lowers oxidative damage in skeletal muscle.
Key Words: aging • free radicals • immunogold • lipid peroxidation • reactive oxygen species • sarcopenia
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. One possibility being
investigated is that CR lowers age-dependent increases in oxidative
stress and damage (1
2
3)
. With regard to human health,
investigation of the ability of CR to retard aging and diseases in a
long-lived primate species may provide data more relevant than that
derived from rodent studies. In response, a long-term CR study in adult
(8- to 14-year-old) male rhesus monkeys was initiated in 1989 at the
Wisconsin Regional Primate Research Center (WRPRC) (4)
.
The maximum life span of rhesus monkeys is thought to be 40 years.
This ongoing study, along with another (5)
, provides
opportunities to investigate the ability of adult-onset CR to influence
longevity, disease patterns, and other expressions of biological aging
in a nonhuman primate species.
The loss of skeletal muscle mass during aging, often referred to as
sarcopenia, is of great public health significance (6)
but
of unknown etiology (7)
. The accumulation of oxidative
damage in skeletal muscle with age as a result of increased oxidative
stress has been hypothesized to contribute to the development of
sarcopenia (8)
. Age-dependent increases in markers of
oxidative damage to DNA, lipids, and proteins determined biochemically
in mammalian skeletal muscle have been widely reported
(9
10
11
12
13
14
15
16
17
18
19)
, thereby supporting a role of oxidative damage in
the etiology of sarcopenia. To date, however, the accumulation of
oxidative damage in mammalian skeletal muscle during aging has not been
investigated histologically, so that the localization of the accrual of
oxidative damage is not well defined.
Caloric restriction provides a tool to investigate the importance of
oxidative damage in the development of sarcopenia. In rats, CR imposed
at late middle age (17 months) opposed the age-dependent loss of muscle
fibers and the development of mitochondrial abnormalities in rats
studied at 30–32 months of age (20)
. In mice, CR started
early in life largely prevents the age-dependent accumulation of
oxidative damage in skeletal muscle, as measured biochemically
(13
, 15
, 17)
. Therefore, investigation of the mechanism by
which CR delays these physiological and biochemical changes in skeletal
muscle is important because it may resolve the etiology of sarcopenia.
In an attempt to reveal the subcellular localization of the inhibition
by CR and to see whether it occurs in primates, we used immunogold
light and electron microscopic (EM) techniques, using
well-characterized antibodies raised against 4-hydroxy-2-nonenal (HNE)
-modified proteins (21
22
23)
, dinitrophenol (DNP)
(24
25)
, and nitrotyrosine (NT) (26
27
28)
to
compare vastus lateralis biopsies obtained from normally fed vs.
calorie-restricted rhesus monkeys. Biochemical assays were used to
investigate correlations among biochemical and morphological
techniques, as shown previously in a model of iron-induced oxidative
stress (23)
.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
were examined. These monkeys
(ranging in age from 17 to 23 years) had been on CR for 10 years when
biopsies of vastus lateralis were obtained for this study. No animal
had any clinical or experimental history expected to differentially
affect skeletal muscle. This protocol was carried out with the approval
of the Institutional Animal Care and Use Committee of the University of
Wisconsin.
For the aging study, animals were fed standard monkey laboratory chow
(No. 5038, Purina, St. Louis, Mo.). For the long-term CR study, both
control and calorie-restricted animals were fed a semipurified diet
(No. 85387, Teklad, Madison, Wis.) as described previously (4
, 29)
. The CR animals had 30% lower food intakes than the
controls. This restriction was achieved by randomly assigning animals
to a treatment group after a 3- to 5-month period of baseline
assessment, during which food intake of the experimental diet was
determined for individual animals. Food intake of the animals assigned
to CR was reduced from their baseline period averages by 10% per month
for 3 months and then maintained at this 30% restriction level. The
controls continued to have free access to food. In 1994, the CR monkeys
were switched to a modified Teklad diet (No. 93131), which is enriched
by 30% in vitamins and minerals. These animals were housed
individually for accurate measurement of daily food intake, but had
extensive auditory and visual contact with other monkeys housed in the
same room. For all monkeys, temperature was maintained at 21°C
with average relative humidity of 50–65%. Room lighting was
automatically controlled to provide alternating 12 h periods of
light and darkness.
Animals consuming the standard laboratory chow were often not allowed true ad libitum access to food. These animals were fed in the morning, and by afternoon many of the animals had consumed all of their food. In contrast, the normally fed animals on the Teklad diet always had food remaining in their cages by late afternoon. Recent measurements of total body percent fat tissue mass (%FTM) using dual-energy X-ray absorptiometry confirmed that animals fed standard laboratory chow ad libitum had %FTM comparable to CR animals rather than normally fed (data not shown). Therefore, the animals of diverse ages consuming standard lab chow were in fact mildly restricted calorically in comparison to the normally fed animals given a semipurified diet.
Tissue preparation
A small portion (200 mg) of the vastus lateralis was removed by
the veterinary staff of the WRPRC under aseptic conditions and general
anesthesia. A portion of each skeletal muscle biopsy was diced and
either snap-frozen in liquid nitrogen and stored for subsequent
biochemical analyses or processed for ultrastructural analysis and
immunogold EM. The remainder of the muscle biopsy was processed for
histological analysis and immunogold light microscopy (LM).
Antibody specificity
The cross-reactivity of polyclonal antibody raised to
HNE-modified keyhole limpet hemocyanin was tested toward compounds
structurally similar to HNE such as malondialdehyde (MDA),
Na-acetylcysteine derivatives of acrolein,
trans-2-pentenal, and trans-2-nonenal.
Competitive Western blot experiments with these potential competitors
were performed. Results indicate that the anti-HNE antibody is highly
specific to HNE-derived modifications to protein, exhibiting no binding
to Michael adducts such as Na-acetylcysteine
derivatives of acrolein, trans-2-pentenal, and
trans-2-nonenal, or to MDA (22)
. Antibody
binding is thus dependent on the presence of the 4-hydroxyl group and
is sensitive to the chain length of the modifying HNE
(22)
. Furthermore, as judged by competitive enzyme-linked
immunosorbent assay experiments, antibody binding is not affected by
the amino acid portion of the adducts (21)
. Therefore, the
epitope recognized by the antibody is the hemiacetal form of the
HNE-derived portion of protein-HNE adducts (21)
. The
specificity of both the DNP (24
25)
and NT
(26
27
28)
antibodies has been well characterized by other
groups. Preincubation of NT with anti-NT blocked labeling of specific
bands in Western blot analyses of kidney tissue homogenates (T. A.
Zainal and T. D. Oberley, unpublished observations).
Biochemical assays
Colorimetric assay for lipid peroxidation
Concentrations of MDA and 4-hydroxy-2-alkenals (4-HDA) were
determined using a colorimetric assay kit (LPO-586, OXIS International,
Portland, Oreg.) according to the manufacturer’s instructions with one
exception: sufficient 2,[6]-di-tert-butyl-p-cresol (Sigma, St. Louis,
Mo.) was added immediately prior to homogenization to achieve a 5 mM
final concentration in order to inhibit further oxidation. The protein
concentration of each sample was measured using a BCA protein assay
reagent (Pierce Chemical, Rockford, Ill.).
Protein carbonyl assay
Protein carbonyl concentrations were measured using the
2,4-dinitrophenylhydrazine (DNPH) procedure of Levine et al.
(30)
, as modified by Sohal et al. (31)
. The
protein concentration of each sample was measured using a BCA protein
assay reagent (Pierce).
Histology
Routine histology was performed as described previously
(23)
.
Ultrastructural analysis
Routine ultrastructural analysis was performed as described
previously (23)
.
Immunogold LM
Immunogold LM was performed as described previously
(23)
with the following exceptions. For carbonyl analysis,
sections were incubated for 1 h in 0.01% DNPH, followed by
extensive rinsing with Tris-buffered saline (TBS: 0.05 M Tris, 0.9%
NaCl, pH 7.6). Sections were incubated with anti-HNE-modified proteins
(1:60), anti-DNP (1:40; LO-DNP-2, Zymed, San Francisco, Calif.), or
anti-NT (1:100; 06–284, Upstate Biotechnology, Lake Placid, N.Y.)
primary antibody overnight at 4°C. The slides were incubated
with diluted (1:100) gold-conjugated goat anti-rabbit
immunoglobulin G (IgG) (LM.GAR5, Goldmark Biologicals,
Phillipsburg, N.J.) or goat anti-rat IgG (LM.GAT5, Goldmark
Biologicals) for 1 h at room temperature.
Immunogold LM quantitation
For quantification using image analysis, black and white prints
for each specimen were prepared using a 2x objective lens. Black and
white images were then obtained as PICT files with a Microtek ScanMaker
V300 flatbed scanner. Image analysis was performed using NIH Image
freeware (version 1.61, available for download from
http://rsb.info.nih.gov/nih-image). The PICT files were opened in
gray scale mode by NIH Image. To make a manual area measurement, a
region of interest (0.3 mm2) was outlined using
the rectangular selection tool. The measure command that would compute
the area, mean gray value, and the minimum and maximum gray value was
then selected. Ten regions were randomly selected for each PICT file.
Mean gray value for each PICT file was obtained by taking the mean
value of the 10 measurements.
Immunogold EM
Immunogold EM was performed as described previously
(23)
with the following exceptions. For carbonyl analysis,
sections were incubated for 1 h in 0.01% DNPH, followed by
extensive rinsing with TBS (0.05 M Tris, 0.9% NaCl, pH 7.6). Sections
were incubated with anti-HNE-modified proteins (1:60), anti-DNP (1:400;
LO-DNP-2, Zymed), or anti-NT (1:1000; 06–284, Upstate Biotechnology)
primary antibody overnight at 4°C. The grids were incubated with
diluted (1:75) gold-conjugated goat anti-rabbit IgG (EM.GAR15, Goldmark
Biologicals) or goat anti-rat IgG (EM.GAT15, Goldmark Biologicals) for
90 min at room temperature.
Immunogold EM quantitation
For quantification using counting of gold beads, a magnification
of 25,000x was used. To take a measurement, a region of interest (0.1
µm2) was outlined using the picture field. Gold
beads within the region were then counted. The total value for each
region was calculated by adding myofibrillar, sarcoplasmic, and
mitochondrial counts. However, there was no mitochondrial labeling with
antibody to HNE-modified proteins. For all antibodies, labeling in
nuclei was rare in comparison to other subcellular compartments. Using
low magnification so that beads were not discernible, 10 regions were
randomly selected for each sample. Mean bead count for each sample was
obtained by taking the mean value of the 10 measurements.
Data analysis
For biochemical assays, statistical analyses were performed
using the unpaired t test, which was modified for unequal
variances when necessary. Data are reported as mean ± standard
error. For morphological studies, preliminary and primary data analyses
were conducted as described below.
Preliminary data analysis
For measurements of myofibrillar, sarcoplasmic, and
mitochondrial labeling at the EM level, 10 measurements for each
subcellular category were made for each animal, allowing us to evaluate
whether or not the total 30 measurements were the result of a single
factor or, in contrast, more than one factor. Nuclear labeling was not
measured, because random sampling of cells with cytoplasm as large as
skeletal muscle cells only infrequently contained nuclei in the grid
sections. We subjected the 30 measurements to principal components
analysis. Based on examination of the scree plot
(32)
and the rotated factor pattern matrices, it seemed
that there were three relatively independent components explaining
variation among the 30 measurements. These components were clearly the
myofibrillar, sarcoplasmic, and mitochondrial measurements. Therefore,
in addition to pooling all 30 measurements together, separate analyses
were conducted for the myofibrillar, sarcoplasmic, and mitochondrial
measurements. However, all 10 measurements for each subcellular
category were averaged to increase the measurement reliability.
Primary analysis
For EM, four primary analyses were conducted: one for the
myofibrillar measurement, one for the sarcoplasmic measurement, one for
the mitochondrial measurement, and one for the integrated response.
Last, total values were also analyzed. For each analysis, we used
ordinary least squares (OLS) regression to assess simultaneously the
independent effects of age and diet on the dependent variable, gold
beads/0.1 µm2. Independent variables included
in the model were age and diet. In each case, interactions between age
and diet condition were tested. However, in no case were such
interactions even close to being statistically significant; therefore,
they were not retained in the model. Assumptions of the parametric
statistical analysis (i.e., normality of residuals and
homoscedasticity) were tested. At most, minor departures were detected.
To evaluate the extent to which this might be producing false
positives, we reconducted the analyses using ranked data per the
recommendation of Puri and Sen (33)
. In no case did the
results change in terms of statistical significance or general
magnitude of effect, which gives substantial confidence in the
significance levels we report. Therefore, we report only the parametric
statistical analyses here. Pairwise differences were tested following
significant omnibus tests of the group effect using the Fisher LSD
approach, which appropriately holds the family-wise alpha level to the
nominal alpha level when there are three, and only three, groups as in
the present study (34)
.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Levels of MDA + 4-HDA in muscle from both adult (15 years, 54.6±1.6
nmol/100 mg protein) and old (26 years, 64.1±0.9 nmol/100 mg protein)
monkeys were significantly greater than in young (4 years, 35.9±1.1
nmol/100 mg protein) animals. As shown in Fig. 2
, protein carbonyl content, determined spectrophotometrically, increased
progressively and significantly with age. A maximum age-dependent
increase of twofold was observed. Protein carbonyl content in muscle
from both adult (4.71±0.01 nmol/mg protein) and old (5.58±0.05
nmol/mg protein) animals was significantly greater than in young
(2.61±0.04 nmol/mg protein) animals. Furthermore, both adult and old
groups differed significantly (P<0.05) from each other.
|
|
Histology
Regardless of age or diet, no intramuscular fibrosis was noted in
hematoxylin- and eosin-stained sections (data not shown). However,
starting at age 16, minimal to moderate deposition of mature adipose
tissue commenced, replacing normal muscle fibers (data not shown).
Calorically restricted animals had a slight tendency toward less fat
deposition (data not shown).
Ultrastructural analysis
EM examination of rhesus monkey vastus lateralis skeletal muscle
showed normal cellular architecture, regardless of age or diet (data
not shown).
Immunogold LM
Labeling in rhesus monkey skeletal muscle was negative when normal
rabbit serum was used in place of primary antibody raised in rabbit
(data not shown). Using a polyclonal antibody raised against
HNE-modified proteins, skeletal muscle from a 4-year-old animal gave a
weak antibody reaction (Fig. 3A
). However, staining in tissue from a representative
26-year-old animal showed much greater reactivity (Fig. 3B
).
These data are representative of the age-dependent increase in labeling
observed when monkeys fed standard lab chow and ranging in age from 2
to 34 years were compared. In all cases, labeling specifically
localized to the sarcolemma, the sarcoplasm, and nuclei. The staining
in the sarcoplasm was granular in nature. When comparing skeletal
muscle from animals in the long-term CR study, the age-dependent
accumulation of HNE-modified proteins in rhesus monkey skeletal muscle
was shown to be attenuated by CR. To illustrate, labeling in a
17-year-old calorically restricted animal (Fig. 3C
) was less
than that observed in a normally fed animal of the same age (Fig. 3D
). Using the OLS regression model with age and diet as
predictors and mean gray value as the dependent variable (Fig. 3E
), both explanatory variables had statistically
significant regression coefficients (P<0.001), with no
interaction of age and group. Furthermore, pairwise comparisons
revealed that the normally fed purified diet group was statistically
different (P<0.001) from each of the other two groups.
|
Incubation of tissue with DNPH results in derivatization of carbonyl
groups leading to DNP adduction to carbonyl-containing
biomacromolecules. The specificity of this method to detect carbonyl
groups was demonstrated by chemical and immunochemical controls. In the
former case, reduction of carbonyl groups with
NaBH4 blocked DNPH binding (data not shown).
Immunochemical validation was demonstrated by omission of DNPH
treatment or substitution of normal rat serum for anti-DNP primary
antibody raised in rat. In both cases, labeling was negative (data not
shown). Using a monoclonal antibody raised against DNP, muscle from a
4-year-old animal (Fig. 4A
) showed considerably less reactivity than tissue from a
26-year-old animal (Fig. 4B
). These data exemplify the
age-dependent increase in labeling observed when monkeys fed standard
lab chow and ranging in age from 2 to 34 years were compared. In all
cases, labeling specifically localized to the sarcolemma, the
sarcoplasm, and nuclei. The staining in the sarcoplasm was granular in
nature. When comparing muscle from control vs. CR animals, carbonyl
levels were not significantly different. For example, labeling in a
17-year-old CR animal (Fig. 4C
) was less than that observed
in a normally fed animal of the same age (Fig. 4D
). Using
OLS regression with age and diet as predictors and mean gray value as
the dependent variable (Fig. 4E
), only age had a
statistically significant regression coefficient (P<0.001).
No interaction of group and age was observed.
|
Using a polyclonal antibody raised against NT, skeletal muscle from
both a 4-year-old (Fig. 5A
) and a 26-year-old (Fig. 5B
) animal gave similar
moderate antibody reactions. Previous results from our laboratory
(T. A. Zainal and T. D. Oberley, unpublished observations)
showed that preincubation of nitrotyrosine with the anti-NT polyclonal
antibody blocked binding in Western blot studies of kidney homogenates.
Labeling localized to the sarcolemma, the sarcoplasm, and nuclei. The
staining in the sarcoplasm was granular in nature. When comparing
skeletal muscle from animals in the long-term CR study, nitrotyrosine
reactivity was lowered by CR. For instance, labeling in a 17-year-old
CR animal (Fig. 5C
) was less than that observed in a
normally fed animal of the same age (Fig. 5D
). Labeling was
negative when normal rabbit serum was used in place of anti-NT primary
antibody (data not shown). Using OLS regression (Fig. 5E
),
both age (P=0.015) and diet (P<0.001) had
statistically significant regression coefficients using mean gray value
as the dependent variable. OLS regression analysis showed that NT
levels decreased significantly with age. Pairwise comparisons revealed
that the CR group was statistically significantly different
(P<0.001) from each of the normally fed groups.
|
Immunogold EM
Labeling in normal rabbit serum-stained skeletal muscle was
negative (data not shown). Weak labeling using antibody raised against
HNE-modified proteins was present in skeletal muscle from a 4-year-old
rhesus monkey (Fig. 6A
). As displayed in Fig. 6B
, increased antibody
labeling was detected in a 26-year-old animal. These data are
representative of the age-dependent increase in labeling observed when
rhesus monkeys fed standard lab chow and ranging in age from 2 to 34
years were compared. In all cases, staining was localized to
myofibrils, nuclei, and the sarcoplasmic reticulum. Labeling was
primarily localized to myofibrils. No mitochondrial labeling was
observed. When comparing biopsies from animals in the long-term CR
study, the age-dependent accumulation of HNE-modified proteins in
rhesus monkey skeletal muscle was shown to be attenuated by CR. To
illustrate, labeling in a representative 17-year-old CR animal (Fig. 6C
) was considerably less than that observed in a normally
fed animal of the same age (Fig. 6D
). Using an OLS
regression model with age and diet as predictors and myofibrillar,
sarcoplasmic, or total gold bead count as the dependent variable (Fig. 6E
), both explanatory variables had statistically
significant regression coefficients (P<0.001), with no
interaction of age and group. Furthermore, pairwise comparisons
revealed statistically significant differences (P<0.001)
between all three diet groups regardless of count classification.
|
For immunogold EM analysis of carbonyl groups, labeling was negative
when the controls described in the immunogold LM results section were
used (data not shown). Using anti-DNP antibody, skeletal muscle from a
4-year-old animal (Fig. 7A
) displayed lower reactivity than tissue from a
representative 26-year-old animal (Fig. 7B
). These data
exemplify the age-dependent increase in labeling observed when animals
fed standard lab chow and ranging in age from 2 to 34 years were
compared. In all cases, staining was primarily localized to myofibrils,
but was also found in mitochondria, nuclei, and the sarcoplasmic
reticulum. The age-dependent accrual of carbonyl groups in rhesus
monkey skeletal muscle was shown to be attenuated by CR (e.g.,
17-year-old CR animal [Fig. 7C
] vs. control of the same
age [Fig. 7D
]). Using OLS regression (Fig. 7E
),
both age and diet had statistically significant regression coefficients
(P<0.001) when using myofibrillar or total gold bead count
as the dependent variable. For sarcoplasmic counts, both age
(P<0.001) and diet (P<0.05) also had
statistically significant regression coefficients. However,
coefficients for age and diet were not statistically significant for
mitochondrial measurements. Pairwise comparisons revealed that the
normally fed semipurified diet group was statistically significantly
different (P<0.001) from each of the other two groups for
the myofibrillar and total gold bead count classifications. For
sarcoplasmic counts, both normally fed groups were statistically
significantly different (<0.01) from each other. No differences were
observed for mitochondrial measurements.
|
Skeletal muscle from both a 4-year-old (Fig. 8A
) and a 26-year-old (Fig. 8B
) animal gave similar
moderate antibody reactions when anti-NT was used. Generally, no
age-dependent change in labeling was observed when animals fed lab chow
and ranging in age from 2 to 34 years were compared. In all cases,
staining was localized to mitochondria, myofibrils, nuclei, and the
sarcoplasmic reticulum. Labeling was primarily localized to myofibrils.
When comparing skeletal muscle from animals in the CR study,
nitrotyrosine reactivity was lowered by CR. For instance, labeling in a
17-year-old CR animal (Fig. 8C
) was considerably less than
that observed in an age-matched control (Fig. 8D
). Using OLS
regression (Fig. 8E
), diet but not age had a statistically
significant regression coefficient (P<0.001) using
myofibrillar, sarcoplasmic, or total gold bead counts as the dependent
variable. Mitochondrial measurements showed no statistical
significance. It should be noted that the influence of age approached
statistical significance for the myofibrillar (P=0.073) and
total (P=0.079) measurements. Pairwise comparisons revealed
statistically significant differences (P<0.001) between all
three diet groups for myofibrillar and total counts. For sarcoplasmic
counts, the lab chow group was statistically significantly different
(P<0.05) from each of the other groups. Again,
mitochondrial measurements showed no statistical significance.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In general, mitochondria are presumed to be the primary cellular sites of reactive oxygen species production during aging; however, our study demonstrates that oxidative damage does not localize primarily to that subcellular compartment in rhesus monkey skeletal muscle. Enhanced antioxidant protection relative to other subcellular compartments and the diffusion of free radicals out of mitochondria to cause damage elsewhere may explain these results. Note that mitochondria did label with anti-DNP and anti-NT antibodies, but not with anti-HNE modified proteins. The explanation for this is not clear, though we have identified strong mitochondrial labeling of mitochondria with antibody to HNE-modified proteins in other systems, including kidney and cardiac muscle. One possible explanation is that HNE performs a specific physiological function in skeletal muscle mitochondria, with enzymes designed to cleave it specifically from that location. Future studies will be necessary to test this hypothesis.
In the present study, biochemically measured lipid peroxidation-derived
aldehydes, such as MDA and 4-HDA, were significantly elevated in muscle
from rhesus monkeys aged 15 and 26 years vs. 4-year-old animals. These
data on lipid peroxidation agree with previous studies (9
10
11
12
, 17
18
19)
demonstrating that tissue content of thiobarbituric acid
reactive substances (TBARS), a marker of endogenous lipid peroxidation,
increases with age in mammalian skeletal muscle. Comparing the
concentration of TBARS in the mitochondrial fraction isolated from
upper hind limb skeletal muscle of mice ranging in age from 7 to 29
months, Lass et al. (17)
showed a sigmoid increase with
age. A rapid increase occurred between 12 and 14 months of age with no
additional elevation thereafter. However, Lass et al. (17)
also reported that the concentration of TBARS in mitochondria from mice
subjected to CR was unaltered during an equivalent period of the life
span. When vastus medialis or lateralis from 66 humans aged 25 to 93
years was compared, there was a significant age-dependent increase in
levels of MDA (18)
. Also, examination of vastus lateralis,
rectus abdominis, and gluteus maximus muscles from 117 humans aged 17
to 91 years revealed a significant increase in MDA and 4-HNE levels
when comparing 66- to 75- and 76- to 85-year-old subjects with those
aged 17–25 years (19)
.
Our biochemical data on protein carbonyl content concur with previous
reports (17
18)
of age-dependent increases in mammalian
skeletal muscle. Using mitochondria isolated from upper hind limb
skeletal muscle, Lass et al. (17)
observed an
age-dependent increase in mitochondrial carbonyl content when comparing
mice from 7 to 29 months of age. In contrast, during a comparable range
of age, Lass et al. (17)
also reported that mitochondria
from mice on CR exhibited no discernible increase in carbonyl content.
When vastus medialis or lateralis from 66 humans aged 25 to 93 years
was compared, there was a significant age-dependent increase in protein
carbonyl content (18)
. In this study, Mecocci et al.
(18)
found that skeletal muscle protein carbonyl levels
from humans aged >60 years were statistically significantly greater
than subjects <50 years in age. Although not statistically
significant, Viner et al. (16)
also reported that carbonyl
groups per mole of Ca-ATPase in sarcoplasmic reticulum (SR) vesicles
isolated from hind limb muscles of male Fisher 344 rats aged 5 or 28
months increased with aging.
As a result of its high reactivity with biological molecules (particularly protein), HNE appears to be an important aldehyde generated by the peroxidation of cellular membrane lipids. In this study, an approximate twofold maximal increase in MDA and 4-HDA was measured biochemically in vastus lateralis from animals aged 4, 15, and 26 years. The biochemical assay we used measures free 4-HDA, of which HNE is a major species, whereas the antibody used for histological analysis was raised specifically against HNE-modified proteins. A comparison between biochemical and morphological measurements therefore is limited since free HNE and HNE adducts have distinct chemical properties. Immunogold LM analysis of skeletal muscle from monkeys ranging in age from 2 to 34 years old revealed a twofold maximal increase in levels of HNE-modified proteins as well. However, a fourfold maximal increase was observed using immunogold EM analysis. At the light level, HNE-modified protein labeling in skeletal muscle from calorically restricted animals was 20% less than in tissue from normally fed animals, whereas immunogold EM revealed a 50% inhibition. Increased sensitivity due to enhanced antigen preservation with immunogold EM and inherent discrepancies in software analysis with immunogold LM may explain these differences.
Incubation of tissue with DNPH results in derivatization of free carbonyls on macromolecules, leading to DNP adduction to the carbonyl-containing molecule. Therefore, unlike the antibody used to detect HNE-modified proteins, a direct comparison can be made between biochemical and morphological analysis of carbonyl groups. In our investigation, an approximate twofold maximal increase in protein carbonyl content was measured biochemically in muscle from animals aged 4, 15, and 26 years. In agreement, a similar increase was observed using both immunogold light and EM analysis and an antibody raised against DNP. Using immunogold LM, carbonyl labeling in muscle from animals on long-term CR was only 5% less than in tissue from normally fed animals, whereas immunogold EM revealed a 15% inhibition. In regard to the NT antibody, immunogold LM demonstrated an age-dependent decrease in NT levels whereas immunogold EM analysis showed no change with age. As previously mentioned, increased sensitivity due to enhanced antigen preservation with immunogold EM and inherent discrepancies in software analysis with immunogold LM may explain these differences. Nitrotyrosine levels were lowered by 20% by CR at both the light and EM level.
It should be noted that animals consuming the standard laboratory chow were not allowed true ad libitum access to food. These animals were fed in the morning and, by the afternoon, many had consumed all of their food. In contrast, the normally fed animals on the semipurified diet always had food remaining in their cages by late afternoon. Measurements of total body %FTM using dual-energy X-ray absorptiometry confirmed that animals fed standard laboratory chow had a total body %FTM comparable to CR and significantly different from normally fed animals (data not shown). Therefore, the animals consuming standard lab chow were, in fact, calorically restricted in comparison to the normally fed animals given semipurified diet, thus providing a likely explanation for the difference in accrual of oxidative damage with age.
The present histological data demonstrate that the subcellular
localization of oxidative damage in rhesus vastus lateralis is
generally myofibrillar in nature, regardless of age or diet. For all
antibodies, staining was largely localized to myofibrils, nuclei, and
the sarcoplasmic reticulum. However, staining was localized primarily
to myofibrils. The bulk of skeletal muscle intracellular protein is
comprised of myofibrillar proteins (35)
, which are
necessary for the generation and transmission of contractile force. The
sheer abundance of myofibrillar proteins may explain in part the
subcellular localization of oxidative damage to myofibrils observed in
this study. On the other hand, turnover of individual proteins in
skeletal muscle occurs at widely differing rates. For example,
myofibrillar protein turnover is different from that found in the
mitochondrial, nuclear, and sarcoplasmic reticulum compartments. The
myofibrillar proteins actin and myosin are continuously degraded and
replaced at rates that vary according to muscle fiber type and the age
of the animal (36)
. However, the myofibrillar proteins
actin and myosin are long-lived, with a turnover of 1–2% per day,
hence a half-life of 30–60 days (37)
. In contrast, a
recent study (38)
using dual isotope analysis of rat
skeletal muscle has shown that that calcium regulatory proteins of the
SR, such as Ca-ATPase, the ryanodine receptor (RyR), and calsequestrin,
and the abundant SR protein—the 53 kDa glycoprotein—display
half-lives of less than 14 days. These investigators report that
turnover rates among individual SR proteins differed, thereby
suggesting individual protein susceptibility to accrual of oxidative
damage. Therefore, in contrast to other intracellular proteins in
skeletal muscle, myofibrillar proteins likely pose a target for
accumulation of oxidative damage due to their sheer abundance and
slower turnover.
An age-dependent decrease in mammalian skeletal muscle protein turnover
appears to occur (39)
. The decline in skeletal muscle
protein turnover in rats from weaning to old age has been carefully
described (40
, 41)
. Goldspink et al. (42)
found that between weaning and senescence, rat diaphragm and extensor
digitorum longus skeletal muscles exhibit progressive decreases in
their fractional rates of growth, protein synthesis, and protein
turnover. Also, using measurement of urinary excretion of
3-methylhistidine (3-MH), an index of myofibrillar protein breakdown,
and rate of L-[14C]-tyrosine incorporation into
gastrocnemius skeletal muscle, Fruhbeck et al.
(43)
demonstrated an age-dependent reduction in the rate
of skeletal muscle protein turnover in male Wistar rats. In that study,
degradation rates were slightly faster than synthesis reduction. No
significant change with age in the urinary 3-MH-creatinine ratio was
observed in an investigation comparing human subjects 18–25 years of
age with those 67–91 years (44)
. However, Yarasheski et
al. (45)
reported a significant age-dependent decrease in
postabsorptive muscle protein synthesis when subjects 23–25 and 60–73
years old were compared. Myofibrillar protein synthesis was also shown
to decrease with age in skeletal muscle from humans aged 21–31 vs.
62–81 years (46)
. Balagopal et al. (47)
reported a progressive decline in myosin heavy-chain, but not
sarcoplasmic, protein synthesis rate with advancing age in humans.
CaATPase and the RyR exhibit significantly slower turnover than other
SR proteins in aged rat skeletal muscle (38)
. These
authors postulate that these two sarcoplasmic proteins are prone to
greater accumulation of modifications with age. This is of note since
the myofibrillar proteins actin and myosin have longer half-lives than
CaATPase and the RyR. An age-dependent decrease in skeletal muscle
protein turnover could account, in part, for the age-dependent accrual
of HNE-modified proteins and carbonyls reported here.
Nitric oxide (NO) performs many regulatory functions in skeletal muscle
(48
, 49)
. Along with other reactive nitrogen species
generated as a result of the interaction between oxygen and NO,
peroxynitrite catalyzes the nitration of protein tyrosine residues to
form NT adducts. Our immunogold light and EM NT results are consistent
with a report (50)
describing no accumulation of
nitrotyrosine adducts with advancing age. Using isotope dilution gas
chromatography-mass spectroscopy, Leeuwenburgh et al. (50)
observed no change in 3-nitrotyrosine levels when comparing vastus
lateralis from rats aged 9 and 24 months. These authors speculated that
their analysis of whole tissue homogenates yielded an average value for
all proteins, thereby failing to detect a possible selective increase
in the oxidation of specific proteins. This is a significant point
since differential accumulation of NT has been shown to occur in
isoforms of sarcoplasmic reticulum Ca-ATPase in rate skeletal muscle
during aging (14)
. Specifically, Viner et al.
(51)
demonstrated that only the SERCA2a isoform contained
significant amounts with approximately one and four NT residues per
young and old Ca-ATPase, respectively. Nitrotyrosine adduct formation
is dependent on equivalent production of both superoxide radical and
NO. Since NO has numerous regulatory functions, we suggest that our NT
findings may simply indicate that production of NO does not increase
with age along with superoxide radical.
Accrual of protein oxidative damage is a function of protein synthesis,
turnover, and repair. Our histological data illustrate that regardless
of antibody, CR lowers oxidative damage in rhesus monkey vastus
lateralis. We speculate that, to some extent, CR may do so by altering
regulation of protein turnover rate. Comparing gastrocnemius muscle
from 30-month-old CR and normally fed C57BL/6 mice, Lee et al.
(52)
used high-density oligonucleotide arrays,
representing 6347 genes, to demonstrate that 16% of transcripts highly
induced by CR encode proteins involved in protein synthesis and
turnover. Measurements of heart (53)
and whole body
(54)
protein turnover rates in calorically restricted rats
have been shown to be age-for-age higher than for normally fed animals,
despite slower growth. In contrast, muscle sizes and total, but not
fractional, synthetic rates were shown to be consistently decreased by
chronic CR (42
, 55)
. Furthermore, urinary excretion of
3-MH has recently been observed to be unaffected by CR (56
, 57)
. However, the effect of CR on turnover rates of individual
skeletal muscle proteins during aging has yet to be investigated, thus
entertaining the possibility that increased turnover of a specific
protein may be induced by CR.
In summary, this is the first report to localize and quantify the age-dependent accumulation of oxidative damage in mammalian skeletal muscle and to characterize its attenuation by CR. Also new is the finding that CR lowers oxidative damage in aging primates. The data suggest that if an age-dependent increase in oxidative stress/damage is pivotal for the development of sarcopenia in mammalian skeletal muscle, the target of oxidative stress is more crucial than the subcellular site of free radical production. Similar to previous studies that biochemically measured markers of oxidative damage in whole tissue homogenates, our study yields an average value for all proteins; therefore, our techniques would fail to detect a selective increase in the oxidation of specific proteins. It appears timely that future studies should not only identify individual oxidized skeletal muscle proteins, but should also attempt to demonstrate precise localization of oxidative damage.
|
ACKNOWLEDGMENTS |
|---|
Received for publication October 5, 1999.
Revision received February 14, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. M. Huffman, D. R. Moellering, W. E. Grizzle, C. R. Stockard, M. S. Johnson, and T. R. Nagy Effect of exercise and calorie restriction on biomarkers of aging in mice Am J Physiol Regulatory Integrative Comp Physiol, May 1, 2008; 294(5): R1618 - R1627. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. R. Heydari, A. Unnikrishnan, L. V. Lucente, and A. Richardson Caloric restriction and genomic stability Nucleic Acids Res., December 3, 2007; 35(22): 7485 - 7496. [Abstract] [Full Text] [PDF]
|
|
