Full text of DOI:10.1093/gerona/56.10.B449

Journal of Gerontology: BIOLOGICAL SCIENCES
Copyright 2001 by The Gerontological Society of America
2001, Vol. 56A, No. 10, B449–B455
Exercise Training Eliminates Age-Related Differences
in Skeletal Muscle Insulin Receptor and IRS-1
Abundance in Rats
Edward B. Arias,1 Luc E. Gosselin,2 and Gregory D. Cartee1
1Biodynamics Laboratory and Department of Kinesiology, University of Wisconsin–Madison.
2Department of Physical Therapy, Exercise, and Nutritional Sciences, State University of New York at Buffalo.
Insulin resistance is common in old age, and exercise training can improve insulin sensitivity. The
purpose of this study was to determine the influence of age (6 vs 26 months) and exercise training (10
weeks of treadmill running) on insulin signaling protein abundance in skeletal muscle from male
Fisher 344 rats. Muscle levels of insulin receptor (IR), insulin receptor substrate-1 (IRS-1), phos-
phatidylinositol 3-kinase (PI3K), and Akt1, a serine–threonine kinase, were determined. IRS-1 was
reduced with aging, IR and PI3K abundance was greater in old rats, and Akt1 was unchanged.
IRS-1 was increased by training in old but not young rats, and IR was increased by training in young
but not old rats. PI3K tended to increase and Akt1 did not change with training, regardless of age.
Aging does not uniformly affect insulin signaling protein abundance, and exercise differentially alters
IR and IRS-1 in young and old rats, thereby eliminating age-related differences in these proteins.
DVANCING age is often associated with insulin re-
sistance (1,2). This outcome may not, however, be in-
dressed age-related changes in the expression of insulin sig-
naling proteins (18,19). Prior to this study, to our knowledge,
postreceptor signaling protein abundance in skeletal muscle
had not been reported for rats older than 18–20 months of age.
Many studies have indicated that exercise training can en-
hance insulin sensitivity for glucose disposal by skeletal
muscle (20,21). Exercise-induced enhancement of glucose
disposal is preserved in 25-month-old rats (22). Recently, in-
vestigators have evaluated the effect of exercise training on
the abundance of insulin signaling proteins, including the in-
sulin receptor, IRS-1, PI3K, and Akt, in the muscle of young
and adult rats (6,23). Previous research with young rats (that
were approximately 4–27 weeks old) has indicated that exer-
cise training can result in increased insulin receptor (6) and
PI3K (23) abundance in skeletal muscle. Differing results
(no change, increased, or decreased levels) have been re-
ported for skeletal muscle IRS-1 abundance after training
(6,23). Chibalin and colleagues (6) found no change in Akt
levels in muscle after exercise training. However, apparently
no studies have been published addressing the effects of ex-
ercise training on the abundance of insulin signaling proteins
in rats older than 5 months of age. Accordingly, the goal of
this study was to determine the influence of exercise training
on the abundance of several key insulin signaling proteins
(insulin receptor, IRS-1, PI3K, and Akt1) in skeletal muscle
from adult (6 months) and old (26 months) rats.
A
evitable, and it is clearly influenced by aspects of life style,
including physical activity (3). Skeletal muscle, an impor-
tant target tissue for insulin, is highly responsive to exer-
cise-induced adaptations, for example the increased abun-
dance and activity of enzymes in the Krebs cycle (4,5). In
addition, at least in muscles from young animals, exercise
training can influence skeletal muscle proteins that are in-
volved in insulin signaling and action (6).
In recent years, much has been revealed about the signal-
ing pathway that links insulin to its multiple actions. The
first step in the insulin signaling pathway is insulin’s binding
to its receptor, followed by receptor autophosphorylation on
specific tyrosines, which in turn activates the receptor’s tyro-
sine kinase to phosphorylate downstream insulin receptor
substrate (IRS) proteins (7). IRS-1 is the predominant IRS
protein expressed by skeletal muscle (8–11). Phosphoryla-
tion of multiple tyrosine residues on IRS proteins is likened
to a “molecular switch” that turns on insulin action: tyrosine-
phosphorylated motifs within IRS proteins serve as docking
sites for binding to downstream signaling proteins that pos-
sess SH2 domains (8). One of these downstream proteins,
phosphatidylinositol 3-kinase (PI3K), is a heterodimer con-
sisting of a regulatory subunit and a catalytic subunit (12).
The PI3K regulatory subunit engages specific, tyrosine-
phosphorylated motifs of IRS, resulting in enhanced PI3K
enzyme activity. A serine–threonine kinase known as Akt
(also known as protein kinase B, or PKB) is an important
signaling protein that is downstream of PI3K (13,14).
METHODS
Materials
Several studies have addressed the influence of age on insu-
lin binding in skeletal muscle (15–17), but only a few have ad-
Pure nitrocellulose membranes (Osmonics, Inc.) were
purchased from Fisher Scientific (Itasca, IL). Anti-IRS-1
B449

B450
ARIAS ET AL.
(06-248) and anti-PI3K p85 (06-497) were purchased from
Upstate Biotechnology, Inc. (Lake Placid, NY), and anti-
ꢀ-insulin receptor (sc-711) and anti-Akt1 (sc-1618) were pur-
chased from Santa Cruz Biotechnology, Inc. (Santa Cruz,
CA). Enhanced chemiluminescence (ECL) kits were pur-
chased from Amersham Pharmacia Biotechnology (Piscat-
away, NJ). Total protein concentrations were performed by
using the bicinchoninic acid method (BCA; Pierce, Rock-
ford, IL), and all other chemicals were purchased from
Sigma Chemical Co. (St. Louis, MO).
standard was similarly prepared from four male Brown Nor-
way rats (retired breeders of indeterminate age; sedentary)
and used for PI3K and Akt1 measurements. Supernatants
were stored at ꢂ80ꢃC until used. Protein concentrations of
the supernatants were determined by the BCA method (26).
Immunoblotting
Aliquots of supernatants (100 ꢄg protein) were solubi-
lized in sodium dodecyl sulfate (SDS) sample buffer, boiled
for 3 minutes, and subjected to 7% (insulin receptor and
IRS-1) or 10% (PI3K and Akt1) SDS–polyacrylamide gel
electrophoresis. A five-point standard curve using pooled
standard (50–150 or 75–125 ꢄg of total protein) was also
loaded on gel and used to normalize signal intensities be-
tween blots. Resolved proteins were transferred to nitrocel-
lulose paper in electrotransfer buffer overnight at a constant
current of 150 mA/transfer apparatus in electrotransfer
buffer (20 mM Tris, pH 8.0, 150 mM glycine, 0.025% SDS,
10% methanol). Nitrocellulose blots were incubated in
blocking solution, which consisted of 5% nonfat milk pro-
tein (Blotto, Santa Cruz Biotech., Inc, Santa Cruz, CA) and
4% bovine serum albumin (BSA) in phosphate-buffered sa-
line plus 0.05% Tween-20 (PBST), pH 7.5, for 2 hours at
room temperature or overnight at 4ꢃC. Blots were then
washed (1 ꢆ 15 min, 2 ꢆ 5 min) in PBST and subsequently
incubated in a 1% Blotto/0.8% BSA PBST solution with
anti-ꢀ-IR (1:1000), anti-IRS-1 (1:1000), anti-PI3K (1:1000),
or anti-Akt1 (1:1000) overnight at 4ꢃC. Blots were then
washed again and incubated with an appropriate second-
ary antibody (horseradish peroxidase-conjugated anti-rabbit
IgG or anti-goat IgG, 1:5000) for 2 hours at room tempera-
ture. Blots were again washed of excess antibody, subjected
to ECL, and immunoreactive protein was quantified by den-
sitometry. Protein abundance is reported as relative concen-
tration based on linear regression from the standard curves
that were run with each gel.
Treatment of Rats
Young (age 3 months) and old (age 23 months) male Fi-
scher 344 rats were obtained from the National Institute on
Aging (Bethesda, MD) and housed two per cage. The ani-
mals were maintained on a 12:12 (hour) light–dark cycle
and provided with Purina rat chow and water ad libitum.
Upon arrival at the animal facility at the University of Buf-
falo, the animals were acclimated to their surroundings for 5
days (24).
Young and old rats were randomly assigned to either
untrained controls (young untrained or YUT, n ꢁ 9; old
untrained or OUT, n ꢁ 8) or exercise-trained animals
(young trained or YT, n ꢁ 8; old trained or OT, n ꢁ 7). The
training protocol was similar to that previously described
(24). Animals were familiarized to treadmill walking (8–10
m/min) over a 1-week period. Following this period, the rats
were weight matched and randomly assigned to untrained
controls or exercise-trained groups. The exercise-trained
groups were subjected to a 10-week progressive training
regimen, ultimately running at speeds corresponding to
ꢀ70% of maximal oxygen consumption for 45 min/day, 5
days/week (25). During the final 4 weeks of training, YT
and OT rats were running 27 and 15 m/min, respectively, on
a 15% incline. The animals were anesthetized, 24–48 hours
following the final exercise bout, with a mixture of xylazine
(10 mg/kg) and ketamine (60 mg/kg) given intraperitoneally
(0.1 ml/100 g). Upon loss of pedal reflexes, plantaris mus-
cles were quickly excised, frozen in liquid nitrogen, and
stored at ꢂ80ꢃC until processing.
Statistical Analysis
All data are expressed as mean ꢇ standard error. Differ-
ences among treatment groups were determined by a two-
way (Age ꢆ Training) analysis of variance using Sigma-
STAT (SPSS, Inc., Chicago, IL). A value of p ꢈ .05 was
considered statistically significant. Pairwise multiple com-
parison procedures (Tukey’s post hoc test) were used to
identify the source of significant variance among groups.
Muscle Preparation
Frozen plantaris muscle was weighed, transferred to pre-
chilled sterile polypropylene tubes, and homogenized (Tis-
sumizer; Tekmar Co., Cincinnati, OH) in ice-cold lysis
buffer (1 ml/100 ꢄg of muscle tissue). The buffer contained 50
mM N-[2-Hydroxyethyl]piperazine-Nꢅ-[2-ethanesulfonic] acid
(HEPES), pH 7.4, 1% Triton X-100 (vol/vol), 10 mM ethyl-
enediamine tetra-acetic acid, 10 mM Na pyrophosphate,
100 mM NaF, 10 mM Na3VO4, 5 ꢄg/ml leupeptin, 0.5 ꢄg/
ml pepstatin, 10 ꢄg/ml aprotinin, and 1 mM phenylmethyl-
sulfonyl fluoride (PMSF). Homogenates were transferred
to microfuge tubes, rotated for 1 hour at 4ꢃC, and then cen-
trifuged at 12,000 ꢆ g for 1 hour to remove insoluble mate-
rial. Rat plantaris muscles to be used for standard curves
(pooled standard) were also prepared in the same fashion,
and supernatants were pooled and stored in frozen aliquots.
For insulin receptor and IRS-1, muscles from eight male
Fischer 344–Brown Norway rats (ꢀ8 months old; seden-
tary) were used for the pooled standard. A second pooled
RESULTS
An important strength of this study was that protein abun-
dance was quantified based on a standard curve included
with each Western blot. The amount of pooled standard
loaded on each gel was linearly related to the relative densi-
tometric units for each of the proteins evaluated: insulin re-
ceptor (R2 ꢁ .930–.998), IRS-1 (R2 ꢁ .972–.999), PI3K
(R2 ꢁ .876–.995), and Akt1 (R2 ꢁ .904–.996). Therefore,
the magnitude of relative changes can be quantitatively in-
terpreted (e.g., a 25% increase in densitometric units can be
interpreted as a 25% increase in immunoreactive protein).
There were significant main effects of age (p ꢁ .03) and
training (p ꢁ .04) on insulin receptor abundance (Figure 1).
A post hoc analysis indicated that the 23% greater value for

AGING, EXERCISE, AND INSULIN SIGNALING
B451
Figure 1. Insulin receptor (ꢀ subunit) protein in skeletal muscle:
100 ꢄg of plantaris muscle homogenate from old untrained (OUT; n ꢁ
8), old trained (OT; n ꢁ 7), young untrained (YUT; n ꢁ 9), or young
trained (YT; n ꢁ 8) rats were immunoblotted with anti-insulin recep-
tor (ꢀ-subunit). Densitometric values were quantitated relative to the
standard curve (pooled plantaris muscle extracts, 75–125 ꢄg) and ex-
pressed as the mean ꢇ SE. Age main effect, p ꢁ .03; training main ef-
fect, p ꢁ .04. *p ꢁ .02, YUT vs YT; †p ꢁ .02, YUT vs OUT, post hoc
analysis.
Figure 2. Insulin receptor substrate-1 protein in skeletal muscle
from old untrained (OUT; n ꢁ 8); old trained (OT; n ꢁ 7); young un-
trained (YUT; n ꢁ 9), or young trained (YT; n ꢁ 8) rats. Densitomet-
ric values were quantitated relative to the standard curve (pooled
plantaris muscle extracts, 75–125 ꢄg) and expressed as the mean ꢇ
SE. Age main effect, p ꢉ .001; training main effect, p ꢁ .035. *p ꢁ
.01, OUT vs OT; †p ꢉ .001, OUT vs YUT, post hoc analysis.
(6,28,29). Insulin signaling is markedly different in these
young, rapidly growing rats as compared with adult or older
animals (18,19). Therefore, the primary purpose of this
study was to determine the influence of age and exercise
training on the abundance of several key insulin signaling
proteins (IR, IRS-1, PI3K, and Akt1) in adult (6 months)
and old (26 months) rats. The most important findings were
that, among these proteins, only IRS-1 was lower in 26-
month-old versus 6-month-old rats, and this age-related dif-
ference was virtually eliminated in the endurance-trained
groups. The difference in insulin receptor in untrained rats
(old slightly greater than young) was also essentially elimi-
nated by endurance training.
OUT rats compared with YUT rats was significant (p ꢁ
.02), as was the 22% greater value for YT rats compared
with YUT rats (p ꢁ .02). Unlike the young rats, in the old
group, a post hoc analysis did not reveal a statistically sig-
nificant effect of training, and no significant difference was
detected between the YT and OT groups.
There were significant main effects of age (p ꢉ .001) and
training (p ꢁ .035) on IRS-1 abundance (Figure 2). There
was a nonsignificant trend for an Age ꢆ Training interac-
tion (p ꢁ .08). A post hoc analysis indicated that the 30%
decreased value for YUT versus OUT rats was significant
(p ꢉ .001). The 31% greater IRS-1 level for OT compared
with OUT rats was also statistically significant based on the
post hoc analysis (p ꢁ .01). In contrast with results for old
animals, the post hoc analysis did not reveal a significant
training effect in the young rats, and no age-related signifi-
cant difference was detected between the YT and OT
groups.
There was a significant main effect of age (p ꢁ .01), but
not of training (p ꢁ .07), on PI3K abundance (Figure 3). A
post hoc analysis indicated an aging effect when rats were
matched for training status: OUT values were 21% greater
than YUT values (p ꢁ .05), and OT values were 25%
greater than YT values (p ꢁ .04).
There were no significant effects of age (p ꢁ .28) or
training (p ꢁ .42) on protein abundance of Akt1 in skeletal
muscle (Figure 4).
DISCUSSION
Both aging (18,19,27) and exercise (6,16,22,23,28–31)
can influence insulin signaling in skeletal muscle. However,
little is known about the influence of exercise training on
insulin signaling in older animals (16,22). Most published
studies regarding exercise and insulin signaling have fo-
cused on very young rats, typically 1–3 months old.
Figure 3. The p85-phosphatidylinositol 3-kinase (PI3-kinase) pro-
tein in skeletal muscle from old untrained (OUT; n ꢁ 8), old trained
(OT; n ꢁ 7), young untrained (YUT; n ꢁ 9), or young trained (YT;
n ꢁ 8) rats. Densitometric values were quantitated relative to the
standard curve (pooled plantaris muscle extract, 50–150 ꢄg) and ex-
pressed as the mean ꢇ SE. Age main effect, p ꢁ .01. *p ꢁ .05, YUT
vs OUT; †p ꢁ .04, YT vs OT, post hoc analysis.

B452
ARIAS ET AL.
values, and standard errors) were not provided. In mouse
muscle, p85-PI3K protein levels tended to be higher in old
(25 months) compared with young (5 and 12 months)
groups (27). Collectively, these data demonstrate that PI3K
abundance does not decline with aging. Rather, there is evi-
dence for higher values in older animals. Other isoforms of
the regulatory subunit of PI3K have been identified (36),
but neither these isoforms nor the catalytic subunit of PI3K
(37) have been studied in relation to aging.
Of the three isoforms of Akt known to be expressed by
skeletal muscle (Akt1, Akt2, and Akt3, also known as
PKBꢊ, PKBꢀ, and PKBꢋ, respectively), insulin activation
is greatest for Akt1 (38). Accordingly, we assessed Akt1
and found that its concentration did not differ with age. Ap-
parently, this is the first study to assess Akt1 concentration
and aging. The influence of age on muscle levels of Akt2
and Akt3 has not been described.
Figure 4. The serine–threonine kinase (Akt1) protein in skeletal
muscle from old untrained (OUT; n ꢁ 8), old trained (OT; n ꢁ 7),
young untrained (YUT; n ꢁ 9), or young trained (YT; n ꢁ 8) rats.
Densitometric values were quantitated relative to the standard curve
(pooled plantaris muscle extract, 50–150 ꢄg) and expressed as the
mean ꢇ SE.
Influence of Exercise Training on the Abundance of
Insulin Signaling Proteins in Skeletal Muscle
We found a small training-induced increase in insulin re-
ceptor abundance. Chibalin and colleagues (6) found an in-
crease in receptor abundance after 1 or 5 days of a swim-
ming protocol (2 ꢆ 3 hour exercise sessions per day) by
female Wistar rats of approximately 6 weeks of age (esti-
mated based on reported body weight). However, Nagasaki
and colleagues (23) did not detect an exercise training effect
on insulin receptor abundance in muscle from 4- and 27-
week-old female, Wistar rats that had access to a voluntary
running wheel beginning at 2 weeks of age. Several studies
have demonstrated an increase in insulin receptor binding
with exercise training (treadmill or voluntary wheel run-
ning) in young (ꢀ3 months) rats (28,29,39). Other studies
reported unchanged insulin binding with exercise training
(treadmill running) in rats of various ages: ꢀ4–5 months
(40), 12 months, and 24 months (16). Interestingly, Willis
and colleagues (41) found a decline in abundance of insulin
receptor ꢀ subunit in muscle from 24-month-old mice after
they performed voluntary wheel running. However, there is
no evidence for reduced insulin receptor quantity in muscle
from rats with exercise training, and many studies have
found evidence that training elicits increased receptor levels.
Exercise training essentially eliminated the age-related
decrease in IRS-1 abundance, with increased levels found in
the OT rats. There was no apparent training effect on IRS-1
in young (6-month-old) rats. Nagasaki and colleagues (23)
also found the effect of exercise (voluntary wheel running
activity) on muscle IRS-1 depended on age: they found in-
creased IRS-1 concentration in muscle from 27-week-old,
but not 4-week-old rats. Chibalin reported that 5 days of
swim training led to a significant decrease in muscle IRS-1
content in approximately 6-week-old rats (6). Age and per-
haps the exercise mode and/or duration are likely determi-
nants of the training response for muscle IRS-1.
Influence of Age on the Abundance of Insulin Signaling
Proteins in Skeletal Muscle
We found a small but significant increase in the abun-
dance of insulin receptor in plantaris muscle from old com-
pared with young rats. Carvalho and colleagues (18) re-
ported no significant effect of age (2, 5, 12, and 18 months)
on insulin receptor of rat skeletal muscle, based on the use
of a Western blot analysis. Previous studies in adult rats
have demonstrated no age-related change (32,33), or mod-
est reductions in insulin receptor number as assessed by in-
sulin binding measures (15–17). Martineau and colleagues
(27) observed no significant age-related changes in insulin
receptor protein levels of skeletal muscle from mice at 5,
12, and 25 months of age. Although results have varied
somewhat, it is clear that there is not a large change in insu-
lin receptor content of rat skeletal muscle between adult-
hood and old age.
IRS-1 protein abundance was 36% lower in skeletal mus-
cle from old compared with young rats. Previous research
has indicated a substantial reduction in IRS-1 in the muscle
of 5-month-old versus 1- to 2-month-old animals (18,23),
with no further decreases in IRS-1 abundance in animals
that were 5, 12, and 18 months of age (18). Our data, taken
together with data from Carvalho (18), suggest that IRS-1
content in rat skeletal muscle declines with advancing age
in a biphasic manner: the first reduction occurs during de-
velopment, and after a long period of stability, the content
declines again during old age. The functional consequences
of this later decline are uncertain. Several IRS isoforms
have been identified (11,34,35), including two in skeletal
muscle (IRS-1 and IRS-2). IRS-1 is recognized as the pre-
dominant isoform involved with insulin-mediated glucose
uptake in skeletal muscle (8,19).
Muscle p85-PI3K content tended to be increased with ex-
ercise training, regardless of age. Nagasakai and colleagues
(23) reported a significant training-induced increase in p85-
PI3K abundance in skeletal muscle from 4- and 27-week-
old rats. Taken together, these results suggest that exercise
Abundance of the p85 regulatory subunit of PI3K was
39% higher in old versus young rats. Carvalho (18) stated
that p85-PI3K was not changed with aging (2, 5, 12, and 20
months); however, numerical data (i.e., sample sizes, mean

AGING, EXERCISE, AND INSULIN SIGNALING
B453
can enhance p85-PI3K content in muscle across a wide
range of ages. The influence of exercise on the isoforms of
the regulatory and catalytic subunits of PI3K has apparently
not been reported.
Akt1 abundance was unaffected by exercise training, re-
gardless of age. Studying much younger rats and using an
antibody that reacts with all three isoforms of Akt, Chibalin
and colleagues (6) also found no exercise-induced change in
protein concentration in muscle.
cle GLUT4 glucose transporter content in old (24–28
months of age) rats. In contrast, many studies have indicated
that exercise training led to increased GLUT4 content in
muscle from young to middle-aged (ꢀ1.5–16 months of age)
rats (6,47,48,50–52). Apparently several key proteins for in-
sulin action, including GLUT4, insulin receptor, and IRS-1,
respond to training differently in young and old rats. Despite
the lack of increased GLUT4 content in old rats, Han and
colleagues (22) found that exercise training led to enhanced
insulin sensitivity, as assessed with a hyperinsulinemic-
euglycemic clamp. Our results suggest that increased abun-
dance of key insulin signaling proteins, including IRS-1,
may contribute to the enhanced insulin action in exercise-
trained old rats. It should be noted that, unlike most studies
with rats, a training-induced increase in GLUT4 abundance
of muscle has been reported for older (ꢀ60–65 years) hu-
mans (53–55) and old (24 months) mice (41).
Implications of Age- and Exercise-Induced Differences
in Insulin Signaling Proteins
A reduction in insulin-stimulated glucose disposal (13–
40%) in rats at 18–25 months versus 3–4 months of age has
been repeatedly described (2,32,42–44). Would the decre-
ment in IRS-1 we observed be expected to diminish insulin
signaling and/or action? There is evidence that a substantial
reduction in IRS-1 alone can diminish insulin signaling
steps yet not elicit insulin resistance. Transgenic mice that
were heterozygous for the null allele in the IRS-1 gene had
an ꢀ60% decline in IRS-1 protein expression compared
with wild-type littermates, accompanied by a comparable
decline in tyrosine phosphorylation of IRS-1 and p85-PI3K-
associated IRS-1 with insulin stimulation (45). However,
animals from the two genotypes had similar glucose re-
sponses to an insulin tolerance test.
Although the reduction in IRS-1 we observed was likely
insufficient to greatly diminish insulin action, this change
together with other deficits in the insulin signaling pathway
might be important. When mice that were double heterozy-
gous for null alleles in both the IRS-1 and insulin receptor
genes were studied, the transgenic mice became hyperin-
sulinemic compared with wild-type controls, and the magni-
tude of hyperinsulinemia increased dramatically with in-
creasing age (2 vs 4–6 months of age) (10). Approximately
40% of the older double heterozygotes became diabetic,
demonstrating an interaction between age and downregula-
tion of multiple insulin signaling proteins. Kido and col-
leagues (46) further assessed the interactions among insulin
signaling genes by producing transgenic mice with com-
bined heterozygous null mutations in the insulin receptor,
IRS-1, and/or IRS-2 genes. Metabolic differences emerged
as mice became older (between 2 and 6 months of age; data
not reported for older animals). Carvalho and colleagues
(18) found, in 18- to 20-month-old compared with 2-month-
old rats, that a decline in IRS-1 was accompanied by a rela-
tively much greater decline in tyrosine phosphorylation of
IRS-1 and IRS-1-bound PI3K in insulin-stimulated muscle.
The disproportionate change in signaling, relative to the de-
cline in IRS-1, suggests that IRS-1 was not the only relevant
change. Our results indicate that the other putative age-
related changes are not reductions in abundance of insulin
receptor, p85-PI3K, or Akt1.
It is essential to recognize that the functional activities of
signaling proteins (e.g., IRS-1-associated PI3K activity or
Akt1 activity) may be altered in the absence of changes in
their abundance. The influence of age and exercise on these
aspects of insulin signaling should be assessed in future ex-
periments.
Conclusions
Aging did not elicit a uniform decline in insulin signaling
protein abundance. Among the proteins studied, a reduction
in old rats occurred only for IRS-1, and this decrease was
accompanied by an increase in p85-PI3K (the binding part-
ner of IRS-1) and insulin receptor concentration. The de-
cline in IRS-1 may contribute to the insulin resistance that
occurs in rats between approximately 6 and 26 months of
age, but changes in proteins other than the insulin receptor,
p85-PI3K, and Akt1 are probably also necessary for im-
paired insulin action. The exercise effect on insulin signal-
ing protein abundance was different in young and old rats.
IRS-1 levels were increased with training in the old rats, but
unaffected in young rats. Conversely, insulin receptor abun-
dance was increased in YT but not OT rats. Regardless of
age, abundance of p85-PI3K tended to be higher after train-
ing, and Akt1 was unresponsive to exercise. Previous stud-
ies have demonstrated that both young and old rats can
achieve enhanced insulin sensitivity with exercise training
(22,56–58), and the current findings suggest that the under-
lying mechanisms may not be identical.
Acknowledgments
This research was supported by The National Institutes of Health (NIH)
under Grant NIH-AG10026 (G. Cartee) and the American Federation for
Aging Research (L. Gosselin). E. B. Arias was supported by institutional
training grant NIH-AG00213.
The authors thank Josh Rodman and Chris Adams for their technical help.
Address correspondence to G. D. Cartee, Biodynamics Laboratory, De-
partment of Kinesiology, 2000 Observatory Drive, Madison, WI 53706.
E-mail: cartee@education.wisc.edu
In old rats, the most striking exercise effect was an in-
crease in IRS-1 content. In addition, we found that p85-PI3K
content tended to be increased by exercise training. Han and
colleagues (22) found that exercise-trained old (25 months of
age) rats had increased IRS-1-bound p85-PI3K in insulin-
stimulated muscle. Most (22,47,48), but not all, studies (49)
have found that exercise training does not increase the mus-
References
1. Harris MI, Flegal KM, Cowie CC, et al. Prevalence of diabetes, im-
paired fasting glucose, and impaired glucose tolerance in US adults—
The Third National Health and Nutrition Examination Survey, 1988–
1994. Diabetes Care. 1998;21:518–524.

B454
ARIAS ET AL.
2. Goodman MN, Dluz SM, McElaney MA, Belur E, Ruderman NB.
Glucose uptake and insulin sensitivity in rat muscle—changes during
3–96 weeks of age. Am J Physiol. 1983;244:E93–E100.
3. Ivy JL. Role of exercise training in the prevention and treatment of in-
sulin resistance and non-insulin-dependent diabetes mellitus. Sports
Med. 1997;24:323–338.
4. Cartee GD. Influence of age on skeletal muscle glucose transport and
glycogen metabolism. Med Sci Sports Exerc. 1994;26:577–585.
5. Booth FW, Thomason DB. Molecular and cellular adaptation of mus-
cle in response to exercise—perspectives of various models. Physiol
Rev. 1991;71:541–585.
receptors, and PI3-kinase protein contents in the C57BL/6 mouse.
Mech Ageing Dev. 1999;106:217–232.
28. Bonen A, Clune PA, Tan MH. Chronic exercise increases insulin bind-
ing in muscles but not liver. Am J Physiol. 1986;251:E196–E203.
29. Santos RF, Mondon CE, Reaven GM, Azhar S. Effects of exercise
training on the relationship between insulin binding and insulin-stimu-
lated tyrosine kinase activity in rat skeletal muscle. Metabolism. 1989;
38:376–386.
30. Zhou Q, Dohm GL. Treadmill running increases phosphaditylinositol
3-kinase activity in rat skeletal muscle. Biochem Biophys Res Com.
1997; 236:647–650.
6. Chibalin AV, Yu M, Ryder JW, et al. Exercise-induced changes in ex-
pression and activity of proteins involved in insulin signal transduction
in skeletal muscle: differential effects on insulin-receptor substrates 1
and 2. Proc Natl Acad Sci USA. 2000;97:38–43.
7. White MF, Kahn CR. The insulin signaling system. J Biol Chem.
1994;269:1–4.
31. Koval JA, Maezono K, Patti ME, Pendergrass M, Defronzo RA, Man-
darino LJ. Effects of exercise and insulin on insulin signaling proteins
in human skeletal muscle. Med Sci Sports Exerc. 1999;31:998–1004.
32. Nishimura H, Kuzuya H, Okamoto M, et al. Change of insulin action
with aging in conscious rats determined by euglycemic clamp. Am J
Physiol. 1988;254:E92–E98.
8. Myers MG, Sun XJ, White MF. The IRS-1 signaling system. Trends
Biochem Sci. 1994;19:289–293.
9. Sciacchitano S, Taylor SI. Cloning, tissue expression, and chromo-
somal localization of the mouse IRS-3 gene. Endocrinology. 1997;
138:4931–4940.
10. Tamemoto H, Kadowaki T, Tobe K, et al. Insulin resistance and
growth retardation in mice lacking insulin receptor substrate-1. Na-
ture. 1994;372:182–186.
11. Sun XJ, Wang LM, Zhang YT, et al. Role of IRS-2 in insulin and cyto-
kine signaling. Nature. 1995;377:173–177.
33. Wang ZQ, Bell-Farrow AD, Sonntag W, Cefalu WT. Effect of age and
caloric restriction on insulin receptor binding and glucose transporter
levels in aging rats. Exp Geront. 1997;32:671–684.
34. Sun XJ, Rothenberg P, Kahn CR, et al. Structure of the insulin-recep-
tor substrate IRS-1 defines a unique signal transduction protein.
Nature. 1991;352:73–77.
35. Lavan BE, Fantin VR, Chang ET, Lane WS, Keller SR, Lienhard GE.
A novel 160-kDa phosphotyrosine protein in insulin-treated embry-
onic kidney cells is a new member of the insulin receptor substrate
family. J Biol Chem. 1997;272:21,403–21,407.
12. Shepherd PR, Withers DJ, Siddle K. Phosphoinositide 3-kinase: the
key switch mechanism in insulin signalling. Biochem J. 1998;333:
471–490.
13. Cohen P, Alessi DR, Cross DAE. The Tenth Datta Lecture—PDK1,
one of the missing links in insulin signal transduction? FEBS Lett.
1997;410:3–10.
14. Summers SA, Birnbaum MJ. A role for the serine/threonine kinase,
Akt, in insulin-stimulated glucose uptake. Biochem Soc Trans. 1997;
25:981–988.
15. Barnard RJ, Youngren JF. Regulation of glucose transport in skeletal
muscle. FASEB J. 1992;6:3238–3244.
16. Eiffert KC, McDonald RB, Stern JS. High sucrose diet and exercise—
effects on insulin-receptor function of 12-month and 24-month-old
Spague-Dawley rats. J Nutr. 1991;121:1081–1089.
17. Kono S, Kuzuya H, Okamoto M, et al. Changes in insulin receptor ki-
nase with aging in rat skeletal muscle and liver. Am J Physiol. 1990;
259:E27–E35.
18. Carvalho CRO, Brenelli SL, Silva AC, Nunes ALB, Velloso LA, Saad
MJA. Effect of aging on insulin receptor, insulin receptor substrate-1,
and phosphatidylinositol 3-kinase in liver and muscle of rats. Endocri-
nology. 1996;137:151–159.
19. Carvalho CRO, Maeda L, Brenelli SL, Saad MJA. Tissue-specific reg-
ulation of IRS-2/PI 3-kinase association in aged rats. Biol Chem. 2000;
381:75–78.
20. Goodyear LJ, Kahn BB. Exercise, glucose transport, and insulin sensi-
tivity. Annu Rev Med. 1998;49:235–261.
21. Wallberg-Henriksson H, Rincon J, Zierath JR. Exercise in the man-
agement of non-insulin-dependent diabetes mellitus. Sports Med.
1998;25:25–35.
22. Han DH, Hansen PA, Chen MM, Holloszy JO. DHEA treatment re-
duces fat accumulation and protects against insulin resistance in male
rats. J Gerontol Biol Sci. 1998;53A:B19–B24.
23. Nagasaki M, Nakai N, Oshida Y, et al. Exercise training prevents mat-
uration-induced decreases in insulin receptor substrate-1 and
phosphatidylinositol 3-kinase in rat skeletal muscle. Metabolism.
2000;49:954–958.
24. Gosselin LE, Betlach M, Vailas AC, Thomas DP. Training-induced al-
terations in young and senescent rat diaphragm muscle. J Appl
Physiol. 1992;72:1506–1511.
25. Mazzeo RS, Brooks GA, Horvath SM. Effects of age on metabolic re-
sponses to endurance training in rats. J Appl Physiol. 1984;60:1369–
1374.
26. Smith PK, Krohn RI, Hermanson GT, et al. Measurement of protein
using bicinchoninic acid. Anal Biochem. 1985;150:76–85.
27. Martineau LC, Chadan SG, Parkhouse WS. Age-associated alterations
in cardiac and skeletal muscle glucose transporters, insulin and IGF-1
36. Inukai K, Funaki M, Ogihara T, et al. p85 alpha gene generates three
isoforms of regulatory subunit for phosphatidylinositol 3-kinase (PI
3-kinase), p50 alpha, p55 alpha, and p85 alpha, with different PI 3-kinase
activity elevating responses to insulin. J Biol Chem. 1997;272:7873–
7882.
37. Funaki M, Katagiri H, Inukai K, Kikuchi M, Asano T. Structure and func-
tion of phosphatidylinositol-3,4 kinase. Cell Signal. 2000;12:135–142.
38. Turinsky J, Damrau-Abney A. Akt kinases and 2-deoxyglucose uptake
in rat skeletal muscles in vivo: study with insulin and exercise. Am J
Physiol. 1999;45:R277–R282.
39. Dohm GL, Sinha MK, Caro JF. Insulin-receptor binding and protein-
kinase activity in muscles of trained rats. Am J Physiol. 1987;252:
E170–E175.
40. Grimditch GK, Barnard RJ, Kaplan SA, Sternlicht E. Effect of training
on insulin binding to rat skeletal muscle sarcolemmal vesicles. Am J
Physiol. 1986;250:E570–E575.
41. Willis PE, Chadan SG, Baracos V, Parkhouse WS. Restoration of in-
sulin-like growth factor I action in skeletal muscle of old mice. Am J
Physiol. 1998;275:E525–E530.
42. Cartee GD, Briggs-Tung CA, Kietzke EW. Persistent effects of exer-
cise on skeletal muscle glucose transport across the lifespan of rats. J
Appl Physiol. 1993;75:972–978.
43. Barzilai N, Banerjee S, Hawkins M, Chang CJ, Chen W, Rossetti L.
The effect of age-dependent increase in fat mass on peripheral insulin
action is saturable. J Gerontol Biol Sci. 1998;53A:B141–B146.
44. Escriva F, Agote M, Rubio E, et al. In vivo insulin-dependent glucose
uptake of specific tissues is decreased during aging of mature Wistar
rats. Endocrinology. 1997;138:49–54.
45. Bruning JC, Winnay J, Bonner-Weir S, Taylor SI, Accili D, Kahn CR.
Development of a novel polygenic model of NIDDM in mice het-
erozygous for IR and IRS-1 null alleles. Cell. 1997;88:561–572.
46. Kido Y, Burks DJ, Withers D, et al. Tissue-specific insulin resistance
in mice with mutations in the insulin receptor, IRS-1, and IRS-2. J
Clin Invest. 2000;105:199–205.
47. Gulve EA, Henricksen EJ, Rodnick KJ, Youn JH, Holloszy JO. Glu-
cose transporters and glucose transport in skeletal muscles of 1-mo-old
and 25-mo-old rats. Am J Physiol. 1993;264:E319–E327.
48. Kern M, Dolan PL, Mazzeo RS, Wells JA, Dohm GL. Effect of aging
and exercise on GLUT-4 glucose transporters in muscle. Am J Physiol.
1992;263:E362–E367.
49. Youngren JF, Barnard RJ. Effects of acute and chronic exercise on
skeletal muscle glucose transport in aged rats. J Appl Physiol. 1995;
78:1750–1756.
50. Kawanaka K, Tabata I, Katsuta S, Higuchi M. Changes in insulin-
stimulated glucose transport and GLUT-4 protein in rat skeletal mus-
cle after training. J Appl Physiol. 1997;83:2043–2047.

AGING, EXERCISE, AND INSULIN SIGNALING
B455
51. Slentz A, Gulve EA, Rodnick KJ, Henriksen EJ, Youn JH, Holloszy
JO. Glucose transporters and maximal transport are increased in en-
durance-trained rat soleus. J Appl Physiol. 1992;73:486–492.
52. Etgen GJ, Brozinick JT, Kang HY, Ivy JL. Effects of exercise training
on skeletal muscle glucose uptake and transport. Am J Physiol. 1993;
264:C727–C733.
53. Hughes VA, Fiatarone MA, Fielding RA, et al. Exercise increases
muscle GLUT-4 levels and insulin action in subjects with impaired
glucose tolerance. Am J Physiol. 1993;264:E855–E862.
56. Ivy JL, Young JC, Craig BW, Kohrt WM, Holloszy JO. Aging, exer-
cise, and food restriction—effects on skeletal muscle glucose uptake.
Mech Ageing Dev. 1991;61:123–133.
57. James DE, Burleigh KM, Kraegen EW, Chisholm DJ. Effect of acute
exercise and prolonged training on insulin response to intravenous glu-
cose in vivo in rat. J Appl Physiol. 1983;55:1660–1664.
58. James DE, Kraegen EW, Chisholm DJ. Effects of exercise training on
in vivo insulin action in individual tissues of the rat. J Clin Invest.
1985;76:657–666.
54. Dela F, Ploug T, Handberg A, et al. Physical training increases muscle
GLUT4 protein and messenger RNA in patients with NIDDM. Diabe-
tes. 1994;43:862–865.
55. Cox JH, Cortright RN, Dohm GL, Houmard JA. Effect of aging on re-
sponse to exercise training in humans: skeletal muscle GLUT-4 and
insulin sensitivity. J Appl Physiol. 1999;86:2019–2025.
Received February 22, 2001
Accepted May 17, 2001
Decision Editor: John Faulkner, PhD