Physiology & Behavior 87 (2006) 360 – 367
Metabolic effects of voluntary wheel running in young and old
Syrian golden hamsters
Agnes E. Coutinho, Sergiu Fediuc, Jonathan E. Campbell, Michael C. Riddell *
School of Kinesiology and Health Science, Faculty of Pure and Applied Science, York University, 4700 Keele Street, Toronto, ON, Canada, M3J 1P3
Received 27 September 2004; received in revised form 25 October 2005; accepted 26 October 2005
Abstract
To explore the metabolic effects of high volume wheel running in the Syrian golden hamster, 6-week old (YOUNG) and 6-month old (OLD)
male animals were randomly divided into sedentary (i.e., YOUNG-S or OLD-S) or running wheel (i.e., YOUNG-RW or OLD-RW) groups (n = 8/
group). RW groups had 24-h access to activity wheels while S were housed in standard rodent cages. At the start of wheel exposure, the number of
revolutions were similar in both groups, but by day 15 were nearly two-fold higher in the YOUNG vs. OLD. OLD ate more than YOUNG and
wheel running increased food intake by ¨50%. YOUNG-RW maintained the same total body mass as YOUNG-S, while OLD-RW had a transient
weight loss of ¨10 g. Perirenal fat mass was smaller in YOUNG- and OLD-RW groups compared with S groups (45% and 66%, respectively.
Plantaris muscle cytochome c oxidase activity was also ¨2-fold higher in YOUNG-RW than in YOUNG-S hamsters but was similar between
OLD-RW and OLD-S groups. Plasma leptin levels were ¨60% lower in YOUNG-RW compared with YOUNG-S and correlated significantly with
visceral fat pad mass (r 2 = 0.58, p = 0.001). Corticosterone levels were lower in YOUNG-RW (13.0 T 0.36 ng/ml) than in YOUNG-S (16.4 T 0.83
ng/ml) hamsters and higher in OLD-RW (22.62 T 0.47 ng/ml) than in OLD-S (15.54 T 0.13 ng/ml) hamsters. These observations reveal that the
hamster is a suitable model for accelerating the effects of exercise on body composition and metabolic alterations associated with training and that
the training adaptations are more pronounced in younger compared with older hamsters, possibly as a result of the higher voluntary wheel activity
in the former group.
D 2005 Elsevier Inc. All rights reserved.
Keywords: Body composition; Visceral fat; Exercise; Leptin; Glucocorticoids; Aging; Rodents
1. Introduction
The metabolic syndrome, classified by excess visceral
adipose tissue, insulin resistance, hypertension, elevated blood
triglyceride levels and decreased high density lipoprotein
levels, has been linked to coronary heart disease and the
development of type 2 diabetes mellitus [25]. Both large scale
epidemiological studies [5,19] and smaller randomized control
trials [16,26] indicate that regular exercise prevents and/or
delays the development of the metabolic syndrome and type 2
diabetes, thus improving health and quality of life. However,
the mechanisms by which exercise achieves these beneficial
effects are poorly understood. Experimental designs using
rodent models have been used to report that endurance exercise
* Corresponding author. Tel.: +1 416 736 2100x40493.
E-mail address: [email protected] (M.C. Riddell).
0031-9384/$ - see front matter D 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.physbeh.2005.10.006
training improves insulin sensitivity, possibly by altering
skeletal muscle insulin sensitivity [reviewed in [15]]. Many
of these studies may be flawed, however, since rodents (usually
rats) are often forced to exercise with treadmill running or
swimming and the exercise is usually performed during the
‘‘lights on’’ cycle when animals would normally sleep. The
effects of forced exercise and the disruption of diurnal sleep
patterns are known to hyperactivate the hypothalamic pituitary
adrenal axis (HPA-axis) [3], which may cause insulin resistance
and disrupt the beneficial effects of regular exercise on the
metabolic syndrome.
The Syrian golden hamster (Mesocricetus auratus) is
commonly used as a model of dietary induced insulin resistance
[12,22,27]. Three weeks of high fructose feeding induces
significant hyperlipidemia, hyperglyceridemia, hyperinsulinemia and the development of whole-body insulin resistance,
similar to what is observed in obese humans with the metabolic
syndrome [27]. The usefulness of the hamster model, rather
A.E. Coutinho et al. / Physiology & Behavior 87 (2006) 360 – 367
than the rat model, for investigating the metabolic syndrome, is
strengthened by the observations that fat metabolism of the
hamster closely resembles that of a human [18], thus making it
a suitable model for investigation of the hormonal and
nutritional regulation of glucose metabolism.
Currently, the effects of voluntary exercise on glucose
metabolism and the development of the metabolic syndrome in
the Syrian golden hamster are unknown. The purpose of this
study is to explore the metabolic effects of voluntary training
on young and old Syrian golden hamsters. Our hypothesis is
that hamsters will demonstrate a high volume of volitional
exercise training, which will dramatically influence body
composition and other measurable parameters that are associated with the development of insulin resistance and type 2
diabetes. We also predict that the metabolic effects of voluntary
exercise will be more pronounced in the older animals when
the attenuation in visceral fat accumulation will be more
notable. These findings hope to establish the Syrian golden
hamster as a suitable animal model for the study of high
volume exercise and the metabolic syndrome.
361
2.3. Experimental procedures
2.3.1. Blood sampling
Non-fasted blood samples were taken via saphenous vein
puncture on the last day of every week at approximately 1330
h, during the circadian trough of their hypothalamo-pituitary
adrenal axis activity, for determination of whole blood glucose
and plasma corticosterone (CORT) concentrations. To do this, a
one-centimeter area overlying the right saphenous vein was
shaven, wiped with alcohol and lightly covered with petroleum
jelly to allow blood droplet accumulation. A 23G size needle
was then used to puncture the vein. The first drop of blood
collected (¨15 Al) was used for whole blood glucose
determination using a standard blood glucose meter (Glucometer Elite XL, Bayer Inc.). Approximately 175 Al of whole
blood was collected into Potassium-EDTA coated microvettes
(SARSTEDT Inc., Montreal, QC, Canada) and centrifuged at 4
-C and 2000 RPM for 5 min prior to plasma collection. The
entire animal preparation and blood collection took approximately 3 min per animal. Plasma samples were frozen and
stored at 20 -C until further analysis (Section 2.3.3.).
2. Materials and methods
2.1. Subjects
Sixteen, 24-day old (i.e., YOUNG) and sixteen 6-month old
(i.e., OLD), male Syrian golden hamsters (Mesocricetus
auratus, Charles River, Montreal, QC, Canada) with an initial
weight of 46.6 T 0.83 g (mean T SEM) and 153.6 T 4.0 g,
respectively, were individually housed in clear cages and kept
in a temperature (23 –25 -C) and humidity (40 – 60%) controlled
room. The animals were allowed to acclimatize to daily human
handling and to a 12 h – light/dark cycle (on 0700 to 1900) for 4
days prior to the start of a 31-day experimental protocol (see
Section 2.2.). They were given access to standard rodent chow
(Purina 5001, 4.3 kcals/g metabolizable energy) and water ad
libitum. Food consumption and body weight were monitored
daily. For food intake, the amount of food remaining in the food
hopper and scattered within the cage, if any, was subtracted from
the amount supplied during the previous 24 h. The experimental
protocol was approved by the York University Animal Care
Committee prior to the start of the experiment.
2.2. Research design
YOUNG and OLD hamsters were randomly assigned to one
of two groups (n = 8/group): sedentary (i.e., YOUNG-S; OLDS) or running wheel (YOUNG-RW; OLD-RW) after the 4 days
of habituation (see above). Hamsters in the RW groups were
individually housed in cages (height: 36.4 cm, width: 26.8 cm,
depth 50 cm) with unrestricted, 24-h access to activity wheels
(outside diameter: 34.5, width: 9 cm) for the 31 day protocol.
Hamsters in the S groups were housed in similar cages for the
same time period but without activity wheels. Wheel revolutions were recorded daily and running distance was calculated
as the wheel circumference times the revolutions. Cages were
cleaned and fresh bedding was replaced once per week.
2.3.2. Euthanization
At the end of five weeks, non-fasted animals were
euthanized by decapitation under terminal CO2 anesthesia
between 0930 and 1300 h. After decapitation, visceral adipose
tissue (i.e., right perirenal fat pad), right plantaris muscle and
the right adrenal gland were removed, weighed and then snap
frozen on dry ice.
2.3.3. Metabolic hormone concentrations
Plasma insulin, leptin, glucagon-like peptide 1(GLP-1),
glucagon and amylin hormone concentrations were determined
from the week four saphenous vein blood samples using the
Luminex i 100 instrument and the LINCOplexi Well Plate
Assay rat/mouse endocrine panel (LINCO Research, INC, St.
Charles, MO, United States).
Plasma CORT concentration was measured in weekly
saphenous vein blood samples using a commercially available
RIA kit (ICN Biomedicals Inc., Costa Mesa, CA, United
States).
2.3.4. Skeletal muscle cytochrome c oxidase activity
Cytochrome c oxidase (COX) activity in the plantaris
muscle was determined from sacrificed animals, as previously
described [11]. Enzyme activity was determined as the
maximal rate of oxidation of fully reduced cytochrome c,
measured by the change in absorbance at 550 nm in a
Microplate Reader (ELx800 Universal, Bio-tek instruments).
2.4. Data analysis
Statistical analysis was completed using Statistica 6.0
statistical software with P 0.05 as the criterion for statistical
significance. All data is expressed as mean T standard error of
mean (SEM). The effect of age and wheel running activity on
anthropometric values, COX activity levels and LINCO plasma
A.E. Coutinho et al. / Physiology & Behavior 87 (2006) 360 – 367
hormone concentration was assessed using a two way (age and
treatment) mixed analysis of variance (ANOVA). A three-way
mixed ANOVA (age by treatment by time) was used to assess
food consumption, weekly blood glucose and weekly plasma
CORT levels. A two-way ANOVA was used for analysis of
wheel running distance.
3. Results
15.0
Food intake (g/day)
362
3.1. Subject characteristics
12.5
10.0
7.5
5.0
Young S
Young RW
Wheel revolutions and running distances are shown in Fig. 1.
A significant main effect of Age [ F(1,14) = 47.76, p < 0.001] and
a significant Age Day interaction [ F(30, 420) = 8.28,
p < 0.001] revealed that wheel revolutions were higher in the
YOUNG vs. OLD group and that the number of revolutions
increased with the duration of exposure in the YOUNG-RW
group but remained relatively stable in the OLD-RW group. At
the start of wheel exposure, the number of revolutions were
similar in both groups (averaging 11337 T 1055, pooled data) but
by day 15, were significantly higher in the YOUNG vs. OLD
hamsters ( p < 0.05). In the YOUNG group, the wheel revolutions continued to increase until reaching a maximum of
20364 T 675 (22.4 T 0.74 km/day) by day 22.
Daily food intake is shown in Fig. 2. At the start of treatment,
food intake was similar between RW and S hamsters within each
age group. Significant main effects for Age [ F(1, 28) = 70.87,
p < 0.001] and for Treatment [ F(1, 28) = 109.73, p < 0.001]
established that OLD ate more than YOUNG hamsters and that
wheel running increased food intake. A significant Day TreatTreatment interaction [ F(33, 924) = 7.09, p < 0.001] revealed
that RW hamsters initially increased their food intake with time
until about week 2, while the S hamsters demonstrated a
decrease in food intake which stabilized by approximately the
third week of treatment. After day five, food intake was higher in
YOUNG-RW compared with YOUNG-S hamsters and after day
6, it was higher in the OLD-RW compared with the OLD-S
Old S
Old RW
2.5
-5
0
5
10
15
20
25
30
35
Day
Fig. 2. Average daily food consumption (g/day). Food intake was measured
daily during three days of habituation (starting at day minus 3) to standard
rodent cages and after exposure to either a novel standard cage or to an activity
wheel cage (day 0). Values are mean T SEM. S=sedentary; RW=running wheel.
A significant Day Treatment interaction [ F(33, 924) = 7.09, p < 0.001]
revealed that RW hamsters increased their food intake with time, while the S
hamsters maintained their food intake.
hamsters. There was no significant Age Treatment Day
interaction [ F(33, 924) = 1.35, p = 0.09].
Body mass is shown in Fig. 3. A three way ANOVA
revealed main effects for Age [ F(1, 28) = 422.23, p < 0.001],
indicating that OLD hamsters were heavier than YOUNG
hamsters and an interaction between Age Day [ F(34,
952) = 225.82, p < 0.001] established that YOUNG hamsters
gained mass while OLD hamsters weight remained more
stable. An Age Treatment Day [ F (34, 952) = 5.10,
p < 0.001] interaction revealed that YOUNG-RW maintained
the same mass as YOUNG-S hamsters during treatment but
that OLD-RW had a small, but significant transient weight loss.
Post hoc analysis of the interaction revealed that the body mass
of the OLD-RW group between days 8– 29 inclusive was
25
Young RW
Old RW
20000
15000
15
10
10000
5
5000
0
Revolutions
Distance (km)
20
0
-5
0
5
10
15
Day
20
25
30
35
Fig. 1. Average daily running distance for the YOUNG-RW (n = 8) and OLDRW (n = 8) groups starting immediately after activity wheel cage exposure (day
0). Values are mean T SEM. A significant main effect of Age [ F(1, 14) = 47.76,
p < 0.001] and a significant Age Day interaction [ F(30, 420) = 8.28,
p < 0.001] were found. Post hoc analysis indicated that the number of
revolutions were higher in YOUNG-RW compared with OLD on and after
day 15.
Body mass (g)
200
150
100
50
Old S
Old RW
Young S
Young RW
0
-5
0
5
10
15
20
25
30
35
Day
Fig. 3. Body mass (g). Values are mean T SEM. Body mass was measured daily
during 3 days of habituation (starting at day minus 3) to standard rodent cages
and after exposure to either a novel standard cage or to an activity wheel cage
(day 0). Values are mean T SEM. S=sedentary; RW=running wheel. An
Age Treatment Day [ F(34, 952) = 5.10, P < 0.001] interaction was found.
Post hoc analysis revealed that the body mass of the OLD-RW group between
days 8 – 29 inclusive was significantly less than day minus 3.
A.E. Coutinho et al. / Physiology & Behavior 87 (2006) 360 – 367
363
Table 1
Organ weights (measured on the right side of the animal), expressed in absolute (Abs) and relative (Rel) to total body weight
Visceral fat pad
Adrenal
Plantaris
Abs (g)
Rel (%BW)
Abs (g)
Rel (%BW)
Abs (g)
Rel (%BW)
Young-S
Young-RW
Old-S
Old-RW
Significance
0.72 T 0.06
0.78 T 0.05
0.010 T 0.0005 y
0.011 T 0.0003
0.021 T 0.002 g
0.023 T 0.003
0.24 T 0.02 a
0.26 T 0.02 a
0.011 T 0.001 y
0.012 T 0.001
0.031 T 0.05
0.034 T 0.005 q
0.71 T 0.11
0.47 T 0.07 h
0.017 T 0.0010
0.011 T 0.0003
0.040 T 0.002
0.026 T 0.001
0.39 T 0.03 a
0.26 T 0.02 a
0.018 T 0.0010
0.012 T 0.0004
0.039 T 0.002
0.026 T 0.002
a, p < 0.01 vs. S
h, p < 0.01 vs. Young-S
y, p < 0.001 vs. Old
ns
g, p < 0.05 vs. all groups
q, p = 0.06 vs. Young-S
BW=body weight; S=sedentary group; RW=running wheel group; ns=Not significantly different.
significantly less than the OLD-S group at day minus 3, by
approximately 10 g.
Absolute and relative right visceral fat pad mass, adrenal
gland mass and plantaris muscle mass at the end of the treatment
period are shown in Table 1. One-way ANOVAs revealed that
absolute [ F(3, 28) = 15.02, p < 0.001] and relative [ F(3,
28) = 35.23, p < 0.001] visceral fat pad mass was considerably
smaller than in the RW groups compared with S groups. Post hoc
analysis revealed that relative visceral fat pad mass was also
lower in the OLD-S compared to the YOUNG-S animals
( P < 0.01). Absolute adrenal gland mass was higher in OLD
compared to YOUNG hamsters [main effect of group; F(3,
28) = 28.13, p < 0.001], but no difference existed between S and
RW groups or between relative adrenal gland mass in any of the
four groups. A one-way ANOVA revealed that absolute plantaris
mass was less in the YOUNG-S group compared with the other
three groups [ F(3, 28) = 8.30, p < 0.001]. A one-way ANOVA
revealed a trend ( P = 0.06) toward a higher relative plantaris
mass in YOUNG-RW compared with YOUNG-S.
3.2. Plasma glucose and hormone concentrations
There was no difference in blood glucose concentrations
between the four groups over the entire duration of the study,
with values ranging within the euglycemic range in all groups
(4.6 T 0.3 –5.1 T 0.1 mM). Table 2 shows metabolic hormone
concentrations in all groups at the end of treatment. A one-way
ANOVA revealed that plasma leptin levels were higher in
YOUNG-S than in the other three groups [ F(3, 24) = 8.71,
p < 0.001]. As shown in Fig. 4A, plasma leptin concentration
correlated significantly with visceral fat pad mass (r 2 = 0.17,
p = 0.03), although this was mainly due to the high correlation
within the YOUNG animals only (r 2 = 0.58, p < 0.001).
Average food intake during the last week of treatment was
significantly correlated with plasma leptin levels measured at
that time (r 2 = 0.27, p = 0.01, Fig. 4B). Plasma glucagon, GLP-1
and amylin levels were not detectable in sufficient number of
animals to allow for a statistical comparison between the two
groups (data not shown). A one-way ANOVA revealed [ F(3,
28) = 46.57, p < 0.001] that plasma corticosterone levels were
higher in OLD-RW compared with the other three groups.
3.3. Cytochrome c oxidase activity
At the end of treatment, plantaris muscle COX activity was
approximately 2-fold higher ( p < 0.01) in the YOUNG-RW
(8.3 T 0.31 Amol/min/g) than in the YOUNG-S (4.1 T 0.22
Amol/min/g) animals, but not significantly different between
the OLD-RW (4.91 T 0.25 Amol/min/g) and OLD-S (5.37 T 0.25
Amol/min/g) groups ( p > 0.05).
4. Discussion
The worldwide incidence of insulin resistance and the
metabolic syndrome is currently increasing at epidemic rates
[29]. Although several studies show improvement of symptoms
with regular exercise (i.e., training), including increased insulin
sensitivity and reductions in plasma lipid levels, the mechanisms responsible for this phenomenon remain unclear. Both
human and rat models have been utilized to investigate the
effects of various exercise training protocols on the development of insulin resistance and type 2 diabetes mellitus
[reviewed in [15]]. Although human studies may be important
to show that regular exercise attenuates the development of the
metabolic syndrome and type 2 diabetes over several decades
[5,19], animal models are also useful to help examine the
physiological mechanisms behind the protective effects of
exercise. Many of the previous studies used forced exercise
models (e.g., treadmill running or swimming) with the exercise
performed during the ‘‘lights on cycle’’ when rodents are
normally sleeping. To our best knowledge, this study is the first
attempt to investigate the impact of voluntary exercise training
Table 2
Plasma fed hormone concentrations
Insulin pM
Leptin pM
Cort ng/ml
Young-S
Young-RW
Old-S
Old-RW
Significance
291.56 T 50
92.24 T 47.1 a
16.37 T 0.83
283.99 T 51
35.5 T 7.3
12.96 T 0.36 y
210.63 T 53
34.01 T 7.4
15.54 T 0.13
176.85 T 37
20.96 T 7.16
22.62 T 0.47 a
ns
a, p < 0.001 vs. all groups
y, p < 0.05 vs. Young-S and Old-S
a, p < 0.01 vs. all groups
Cort=corticosterone; S=sedentary group; RW=running wheel group. Blood samples were taken during week four at 0100 h.
364
A.E. Coutinho et al. / Physiology & Behavior 87 (2006) 360 – 367
Leptin Concentration (pM)
A
Young S
Young RW
200
175
Old S
r2=0.17
p=0.03
150
Old RW
125
100
75
50
25
0
0.00
0.25
0.50
0.75
1.00
1.25
Visceral Fat Mass (g)
B
Food Intake (g)
15
10
5
r2=0.27
p=0.010
0
0
50
100
150
200
Leptin Concentration (pM)
Fig. 4. A) Plasma leptin concentrations vs. visceral fat pad mass in all animals
at the end of treatment. B) Average food intake during the last week of
treatment vs. plasma leptin levels on the final day of treatment in all animals.
on body composition and a variety of metabolic markers
related to the development of insulin resistance and the
metabolic syndrome in the Syrian golden hamster, an animal
model that exhibits a high volume of voluntary wheel running.
In this study, we found that young Syrian golden
hamsters are a suitable animal model for accelerating the
metabolic effects of regular voluntary-type endurance exercise. Specifically, we found that young hamsters who were
allowed continuous access to running wheels for 31 days,
had considerably less visceral adipose tissue, a tendency for
a higher skeletal muscle mass (Table 1) and higher skeletal
muscle cytochrome C oxidase activity, a cellular marker of
muscle oxidative capacity. The number of wheel revolutions
performed by the YOUNG-RW group doubled throughout
the treatment protocol from 11900 T 1057 revolutions/day
(13.09 T 1.163 km/day) to a peak of 20364 T 675 revolutions
(22.4 T 0.74 km/day) by day 22, which likely provided a
continuous stimulus for these metabolic changes (Fig. 1). In
this model, these morphologic changes occur despite similar
increases in total body mass and adrenal mass gain in the S
and RW groups (Fig. 3). We also found that, compared with
YOUNG-S animals, YOUNG-RW animals had lower circulating basal (i.e., unstimulated) glucocorticoid concentration
and markedly lower leptin levels by the end of treatment
(Table 2). These morphological and metabolic adaptations
associated with high volume wheel running in this young
hamster model may be protective against the development of
insulin resistance and type 2 diabetes, which can be initiated
by high fructose or high fat feeding in this animal model
[22].
Although the impact of volitional wheel running on several
metabolic parameters in YOUNG hamsters was evident, we
also wished to determine if there may be more pronounced
metabolic difference with wheel running in older animals. Our
specific aim was to determine if visceral fat mass and insulin
resistance could be reversed with voluntary wheel running in
animals that start the exercise later in life. This was
hypothesised since regular exercise is already known to
attenuate visceral fat mass gain and maintain insulin sensitivity
in rodents exposed to volitional wheel running throughout their
lifespan [13]. To test this, we exposed 6-month old (i.e., middle
aged) hamsters to the identical experimental protocol that was
used for the younger hamsters. We found that volitional wheel
running later in life causes significant reductions in visceral fat
mass but no change in skeletal muscle mass, or in skeletal
muscle cytochrome C oxidase activity. Moreover, to our
surprise, we did not see any improvements in fed plasma
insulin or glucose concentrations nor changes in plasma leptin
levels with wheel running in these older hamsters. Paradoxically, we observed increases in basal (i.e., nadir) plasma
corticosterone levels (Table 2) in the older hamsters despite the
observation that the number of wheel revolutions remained
relatively stable throughout of the treatment protocol (Fig. 1). It
appears, therefore, that the metabolic adaptations to volitional
wheel running in older animals differ somewhat from younger
hamsters, either because the amount of wheel running is less
(see Fig. 1) or because there are age-related differences in
adaptation to regular exercise.
The high volume of voluntary wheel running (¨20 km/day)
in our YOUNG hamsters is consistent with reports from other
investigations using young hamsters [10,28]. Previous reports
have shown that hamsters [10,28] exhibit substantially higher
volumes of volitional wheel running, in comparison to the 5– 6
km/day range reported for young rats [14]. Interestingly, the
circadian rhythms of wild and laboratory hamsters have been
examined by others [9,28] and reveal that the robust activity
levels observed in the laboratory are comparable to that of the
wild type hamsters of a similar age. These observations also
demonstrate that YOUNG hamsters are relatively homogenous
in their activity and food intake patterns, which contrasts to
what has been observed for out-bred strains of rats [21]. We
also found that older hamsters, exposed to an identical
environment, initially ran a similar distance as the younger
group but did not increase their wheel running volume over
time (Fig. 1). We were not able to find any published studies on
the wheel running behavior of rats exposed to wheel running
later in life, although several studies have shown that there is a
decrease in activity patterns as young animals age while they
are exposed to running wheels [17]. Preliminary investigations
suggest that previously sedentary 23-month old rats exposed to
wheel running accumulate less than 100 m per night of running
A.E. Coutinho et al. / Physiology & Behavior 87 (2006) 360 – 367
(personal communications with D.A. Hood, York University,
Toronto). It may be, therefore, that older hamsters are a useful
model for the study of various metabolic and other adaptations
to voluntary exercise in previously sedentary aging mammals.
In addition to exhibiting considerably higher levels of wheel
running compared to rats, hamsters appear to be considerably
more homogeneous in their running wheel activity and appear
to maintain similar total body mass to control despite high
volumes of exercise. For example, the range of distance ran by
our YOUNG hamsters during the final week of treatment was
18 –24 km/per day compared to 1 –12 km reported for rats for
approximately the same training duration [14]. Interestingly,
this high amount of wheel running does not suppress food
intake or lower body mass in hamsters, as has been consistently
reported in rats [21]. In older hamsters, we observed only a
small (5 –8 g) transient reduction in body mass with wheel
running (Fig. 1) while food intake dramatically and immediately increased to compensate for the increased energy
expenditure. These behavioural differences in running volume,
food intake and maintenance of total body mass in hamsters
compared with rats may be particularly relevant when
examining the metabolic responses to endurance exercise
training.
Our study provides some information about body composition and the energy cost of high volume wheel running in
this animal model. Since the S and RW groups had similar
body masses throughout the treatment period (Fig. 3), the
energy cost of running, expressed in grams of food, may be
estimated from the differences in food consumption between
the two groups [1]. For example, during the last week of
running, the YOUNG-RW hamsters, averaging 20.3 km/day,
consumed 3.22 g more food than the S animals, and weighed
¨90 g. As such, the estimated cost of running per gram of
body mass is ¨0.0018 g/km. Similarly, OLD-RW hamsters ran
9.6 km/day, consumed 3.5 g more food than the S group, and
weighed 153 g, making the energy cost of running, expressed
per gram of body mass, equal to 0.0024 g/km, similar to the
0.0020 g/km reported for mice [20] and 0.0018 g/km reported
for rats [1].
As mentioned above, we observed profound differences in
visceral adipose tissue mass between the RW and S animals at
the end of treatment (Table 1). To date, most rodent studies fail
to report any indexes of body composition in exercise training
studies and tend to report that total body mass may be lowered
with voluntary wheel running at least in rats [1]. The
attenuation in visceral adipose tissue gain in the YOUNGRW group was particularly dramatic, likely because of the
relatively high energy expenditure in this group. Together, a
higher muscle mass and lower visceral mass in the YOUNGRW compared with the YOUNG-S animals suggests a major
change in body composition, despite no difference in total body
mass. This change in body composition (seen here as decreased
visceral fat pad mass and increased lean mass) due to increased
energy expenditure in the RW animals would be expected to
enhance insulin sensitivity in these animals, perhaps by
lowering portal release of free fatty acids and/or by altering
the secretion of fat-derived metabolically active factors such as
365
resistin, leptin, ACRP30, tumor necrosis factor-a and fatderived glucocorticoids. Indeed, visceral adipose tissue and
skeletal muscle mass are the main determinants of whole body
insulin sensitivity and the surgical removal of visceral fat has
been shown to prevent insulin resistance associated with aging
and type 2 diabetes in Zucker Diabetic Fatty rats [8]. Although
we also observed a smaller reduction in visceral fat pad mass in
the OLD-RW group, we failed to see any differences in skeletal
muscle mass, possibly because the training volume was less in
the older vs. younger animals.
Surprisingly, we observed no difference between RW and S
groups (YOUNG or OLD) in fed plasma insulin or fed glucose
concentrations (Table 2). The failure to observe differences in
glucose or insulin levels between RW and S groups, despite
major differences in body composition, may be related to the
higher food intake in the RW groups and/or the time of day that
these measurements were taken. Indeed, a potential limitation
of our study is that blood samples were taken in the early
afternoon and the animals had not been fasted during the
evening before. Typically, insulin sensitivity is measured in the
fasted state or following a standardized glucose challenge. It is
particularly relevant to note that, compared with S animals, the
RW animals consumed 50% more food daily and yet had
similar insulin levels (Fig. 2), suggesting that the later group
had higher insulin sensitivity. Further studies are needed,
therefore, to determine if high volume wheel running influences glucose tolerance in older hamsters fed an insulin
resistant, pair-fed diet.
In line with the attenuated visceral adipose tissue mass in
the YOUNG-RW hamsters, we observed that plasma leptin
levels in these animals were two-thirds lower than that
observed for the S animals by the end of the protocol (Table
2). Other studies have also reported lower circulating leptin
levels in trained compared to untrained animals [21];
however, it is still unclear whether or not this adaptation is
a direct result of training or simply a reflection of the reduced
adipose tissue content. We also observed that plasma leptin
levels correlated modestly with visceral fat pad mass (Fig.
4a), although the strength in this relationship was determined
primarily by the strong correlation in the YOUNG groups
(r 2 = 0.58). Leptin, an adipokine secreted from adipose tissue
in proportion to the amount of fat stored, causes satiety and
helps regulate glucose and fat metabolism in mammals [7].
High levels of circulating leptin concentration and leptin
resistance have been associated with insulin resistance in
humans and in rodent models of obesity [7]. In our study, we
also observed a modest correlation between food intake
during the last week of treatment and circulating leptin levels
(Fig. 4B). It is clear, however, that factors other than leptin
levels influenced food intake in these animals. Furthermore,
since leptin is secreted primarily from subcutaneous stores,
which were not measured in our study, it is of no surprise that
this relationship is modest at best. Interestingly, exercise per
se, or via some intermediary signal likely influences food
consumption, since the RW groups had significantly higher
food intake than S groups for any given level of circulating
leptin (Fig. 4). The associations between leptin levels and
366
A.E. Coutinho et al. / Physiology & Behavior 87 (2006) 360 – 367
visceral fat mass and between food intake and leptin
concentration in our study should be viewed with caution,
since the number of animals in each group is small (n = 8 in
each) and because there appear to be some baseline
differences in leptin levels between animals.
Another novel finding of this study is that after approximately four weeks of regular exercise, the YOUNG-RW
hamsters had similar basal plasma CORT concentration in
comparison to their sedentary counterparts, in spite of a high
volume of wheel running, while the OLD-RW animals had a
higher CORT level compared with their sedentary counterparts
(Table 2). The low basal CORT levels in the YOUNG-RW
group is particularly interesting since four weeks of wheel
running normally causes hyperactivity of the hypothalamopituitary axis at the onset of the dark light cycle but a normal
HPA activity in the morning in mice [4]. We have recently
shown that in rats, there is an initial upregulation, followed by a
gradual attenuation, in diurnal HPA axis activity during 4 weeks
of voluntary wheel running [6] or forced swimming [24].
Together these observations reveal that there is a reduction of
basal pituitary– adrenal cortical axis with training and/or a
reduction in glucocorticoid secretion from non-adrenal sites,
such as visceral fat [2]. Although a larger adrenal gland mass
has been reported with voluntary wheel running [4] and with
forced exercise [23] in other rodent models, we did not find that
the adrenal weight was altered significantly by wheel running in
OLD or YOUNG hamsters. In older hamsters we observed an
increase in plasma corticosterone concentration suggesting
continued hyperactivity of the HPA axis following 31 days of
training (Table 2). It may be, therefore, that older animals fail to
adapt to the daily stress of running, although further investigations are needed to confirm these findings.
In summary, our findings indicate that the hamster is a
reasonable model to study the accelerated effects of high
volume voluntary training on various metabolic parameters
related to glucose homeostasis. In this model, regular exercise
results in considerably lower visceral fat mass gain and an
increase in lean mass and mitochondrial oxidative capacity. In
addition, high volume wheel running appears to lower basal
glucocorticoid levels and increase leptin levels, possibly as a
result of the reduction in visceral adiposity. These morphological and metabolic adaptations associated with high volume
wheel running may be protective against the development of
insulin resistance and type 2 diabetes, which can be initiated
in this animal model with various nutritional interventions.
We also conclude that exercise training starting later in life
also influences the aging hamster in a positive way, although
to a lesser extent, possibly because of a lower volume of
voluntary physical activity and an increase in pituitary –
adrenal activity.
Acknowledgments
A. E. Coutinho is a recipient of the Natural Science and
Engineering Research Council of Canada (NSERC) Postgraduate Scholarship. This project was generously supported by
NSERC operating grant (261306) and by infrastructure grants
from the Canadian Foundation for Innovation and the Ontario
Innovation Trust.
References
[1] Afonso VM, Eikelboom R. Relationship between wheel running, feeding,
drinking, and body weight in male rats. Physiol Behav 2003;80:19 – 26.
[2] Cavagnini F, Croci M, Putignano P, Petroni ML, Invitti C. Glucocorticoids
and neuroendocrine function. Int J Obes Relat Metab Disord 2000;
24(Suppl 2):S77 – 9.
[3] Chennaoui M, Gomez MD, Lesage J, Drogou C, Guezennec CY. Effects
of moderate and intensive training on the hypothalamo – pituitary – adrenal
axis in rats. Acta Physiol Scand 2002;175:113 – 21.
[4] Droste SK, Gesing A, Ulbricht S, Muller MB, Linthorst AC, Reul JM.
Effects of long-term voluntary exercise on the mouse hypothalamic –
pituitary – adrenocortical axis. Endocrinology 2003;144:3012 – 23.
[5] Eriksson KF, Lindgarde F. Prevention of type 2 (non-insulin-dependent)
diabetes mellitus by diet and physical exercise. The 6-year Malmo
feasibility study. Diabetologia 1991;34:891 – 8.
[6] Fediuc S, Campbell J, Riddell MC. Effect of voluntary training on
glucocorticoid responsiveness to restraint stress in Sprague – Dawley rats.
Med Sci Sports Exerc 2005;37:S240.
[7] Friedman JM, Halaas JL. Leptin and the regulation of body weight in
mammals. Nature 1998;395:763 – 70.
[8] Gabriely I, Ma XH, Yang XM, Atzmon G, Rajala MW, Berg AH, et al.
Removal of visceral fat prevents insulin resistance and glucose
intolerance of aging: an adipokine-mediated process? Diabetes 2002;51:
2951 – 8.
[9] Gattermann R, Fritzsche P, Weinandy R, Neumann K. Comparative
studies of body mass, body measurements and organ weights of wildderived and laboratory golden hamsters (Mesocricetus auratus). Lab
Anim 2002;36:445 – 54.
[10] Gibbs FP, Petterborg LJ. Exercise reduces gonadal atrophy caused by short
photoperiod or blinding of hamsters. Physiol Behav 1986;37:159 – 62.
[11] Gordon JW, Rungi AA, Inagaki H, Hood DA. Effects of contractile
activity on mitochondrial transcription factor A expression in skeletal
muscle. J Appl Physiol 2001;90:389 – 96.
[12] Goulinet S, Chapman MJ. Plasma lipoproteins in the golden Syrian
hamster (Mesocricetus auratus): heterogeneity of apoB- and apoA-Icontaining particles. J Lipid Res 1993;34:943 – 59.
[13] Gulve EA, Rodnick KJ, Henriksen EJ, Holloszy JO. Effects of wheel
running on glucose transporter (GLUT4) concentration in skeletal muscle
of young adult and old rats. Mech Ageing Dev 1993;67:187 – 200.
[14] Halseth AE, Fogt DL, Fregosi RF, Henriksen EJ. Metabolic responses of
rat respiratory muscles to voluntary exercise training. J Appl Physiol
1995;79:902 – 7.
[15] Henriksen EJ. Invited review: effects of acute exercise and exercise
training on insulin resistance. J Appl Physiol 2002;93:788 – 96.
[16] Holm G, Bjorntorp P. Metabolic effects of physical training. Acta Paediatr
Scand Suppl 1980;283:9 – 14.
[17] Ingram DK. Age-related decline in physical activity: generalization to
nonhumans. Med Sci Sports Exerc 2000;32:1623 – 9.
[18] Kasim-Karakas SE, Vriend H, Almario R, Chow LC, Goodman MN.
Effects of dietary carbohydrates on glucose and lipid metabolism in
golden Syrian hamsters. J Lab Clin Med 1996;128:208 – 13.
[19] Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM,
Walker EA, et al. Reduction in the incidence of type 2 diabetes with
lifestyle intervention or metformin. N Engl J Med 2002;346:393 – 403.
[20] Koteja P, Swallow JG, Carter PA, Garland Jr T. Energy cost of wheel
running in house mice: implications for coadaptation of locomotion and
energy budgets. Physiol Biochem Zool 1999;72:238 – 49.
[21] Kraemer RR, Chu H, Castracane VD. Leptin and exercise. Exp Biol Med
(Maywood, NJ) 2002;227:701 – 8.
[22] Leung N, Naples M, Uffelman K, Szeto L, Adeli K, Lewis GF.
Rosiglitazone improves intestinal lipoprotein overproduction in the fatfed Syrian Golden hamster, an animal model of nutritionally-induced
insulin resistance. Atherosclerosis 2004;174:235 – 41.
A.E. Coutinho et al. / Physiology & Behavior 87 (2006) 360 – 367
[23] Moraska A, Deak T, Spencer RL, Roth D, Fleshner M. Treadmill
running produces both positive and negative physiological adaptations
in Sprague – Dawley rats. Am J Physiol Regul Integr Comp Physiol
2000;279:R1321 – 9.
[24] Park E, Chan O, Li Q, Kiraly M, Matthews SG, Vranic M, et al. Changes
in basal hypothalamo – pituitary – adrenal activity during exercise training
are centrally mediated. Am J Physiol Regul Integr Comp Physiol 2005;
289:R1360 – 71.
[25] Reaven GM. Role of insulin resistance in human-disease. Diabetes 1988;
37:1595 – 607.
[26] Ruderman NB, Ganda OP, Johansen K. The effect of physical training on
glucose tolerance and plasma lipids in maturity-onset diabetes. Diabetes
1979;28(Suppl 1):89 – 92.
367
[27] Taghibiglou C, Carpentier A, Van Iderstine SC, Chen B, Rudy D, Aiton A,
et al. Mechanisms of hepatic very low density lipoprotein overproduction
in insulin resistance. Evidence for enhanced lipoprotein assembly, reduced
intracellular ApoB degradation, and increased microsomal triglyceride
transfer protein in a fructose-fed hamster model. J Biol Chem 2000;
275:8416 – 25.
[28] Weinert D, Fritzsche P, Gattermann R. Activity rhythms of wild and
laboratory golden hamsters (Mesocricetus auratus) under entrained and
free-running conditions. Chronobiol Int 2001;18:921 – 32.
[29] Zimmet PZ, McCarty DJ, de Court. The global epidemiology of noninsulin-dependent diabetes mellitus and the metabolic syndrome. J
Diabetes Complications 1997;11:60 – 8.