J Appl Physiol 118: 1502–1509, 2015.
First published August 22, 2014; doi:10.1152/japplphysiol.00588.2014.
Participation in a 1,000-mile race increases the oxidation of carbohydrate in
Alaskan sled dogs
Benjamin F. Miller,1* Joshua C. Drake,1 Frederick F. Peelor III,1 Laurie M. Biela,1 Raymond Geor,2
Kenneth Hinchcliff,3 Michael Davis,4* and Karyn L. Hamilton1*
1
Department of Health and Exercise Science, Colorado State University, Fort Collins, Colorado; 2Large Animal Clinical
Sciences, College of Veterinary Medicine, Michigan State University, East Lansing, Michigan; 3Faculty of Veterinary Science,
The University of Melbourne, Parkville, Victoria, Australia; and 4Center for Veterinary Health Sciences, Oklahoma State
University, Stillwater, Oklahoma
Submitted 1 July 2014; accepted in final form 19 August 2014
canine; stable isotope; exercise; substrate; carbohydrate
CANIS LUPUS FAMILIARIS, THE domesticated dog, is capable of
extreme endurance performance. The subspecies Alaskan
Husky has been specifically bred for endurance performance. It
is uncertain what the true maximal O2 consumption (V̇O2max)
of a well-trained Alaskan Husky is, but a study in previously
untrained yearling sled dogs recorded an average V̇O2max of
198.7 ml·kg⫺1·min⫺1 after 4 wk of moderate-intensity training
(1). At the annual Iditarod and Yukon Quest 1,000 sled dog
races, the winning teams can cover nearly 1,600 km in fewer
than 9 days. When the energetic demands of this extreme
activity are combined with environmental stresses, total energy
expenditure (TEE) approximates 50,000 kJ/day (11). This
value far exceeds, on both a relative (per kg body weight) and
absolute (despite dogs being one-third the weight of humans)
basis, the highest values recorded in humans during the Tour
de France (28). It is possible that, during these sustained
* B. F. Miller, M. Davis, and K. L. Hamilton are co-principal investigators.
Address for reprint requests and other correspondence: B. Miller, Dept. of
Health and Exercise Science, Colorado State Univ., 220 Moby B Complex,
Fort Collins, CO 80523-1582 (e-mail: [email protected]).
1502
periods of exercise, the dogs approach values considered the
maximal metabolizable energy intake (13). Understanding how
the canine athlete sustains extremely high energetic flux over
prolonged periods of time could provide insight into how to
sustain similar efforts in humans (i.e., endurance athletes and
military professionals).
Data on metabolic flux in canine athletes are sparse. Classic
studies of metabolic flux often used canine models, resulting,
for example, in the vast amount of data available on hepatic
metabolism and glucoregulation (4, 24). The classic comparative physiology studies of Weibel, Taylor, and others used
moderately trained dogs (specifically, Labrador Retrievers) as
representative of a highly aerobic species to compare to a
species (Capra hircus, pygmy goats) with relatively low aerobic capacity (22, 26, 27). These studies determined that
maximal lipid oxidation in the dogs, as measured by respiratory exchange ratio (RER), occurred at 40% V̇O2max, which
was not different than that in goats (22). Furthermore, at the
same relative exercise intensity (%V̇O2max), RER was equivalent between dogs and goats, although absolute rates of fatty
acid oxidation were greater in dogs because of the higher
overall flux rates (26). Although these studies used 14C-labeled
free fatty acids to measure oxidation rates (Rox) during exercise (26), no labeled bicarbonate prime was provided, nor were
bicarbonate kinetics characterized. The lack of knowledge of
bicarbonate kinetics is a concern because dogs are capable of
splenic contraction (15), which could increase CO2 retention. It
is well established that factors that change the functional
bicarbonate pool will change labeled carbon recovery, thus
potentially compromising carbon-measured Rox (12, 25).
From these collective data, it has been assumed that longdistance racing sled dogs sustain exercise by high free fatty
acid Rox (10).
In a previous study, our laboratory demonstrated that Alaskan sled dogs had significant glycogen depletion, as would be
expected after a 160-km run (16). Remarkably though, each of
four more successive 160-km runs over the subsequent 4 days
failed to significantly deplete glycogen stores. In addition,
muscle glycogen was maintained while the dogs were consuming a diet low in carbohydrate (CHO; 15% calories of total
energy) (16). From these data, we concluded that canine
athletes have a unique ability to rapidly adapt to an extreme
exercise stress, enabling them to sustain exercise without
reliance on stored CHO.
Given the interest of both occupational and athletic populations in sustaining prolonged submaximal exertion over prolonged periods in humans, we sought to characterize fuel use
8750-7587/15 Copyright © 2015 the American Physiological Society
http://www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 17, 2017
Miller BF, Drake JC, Peelor FF 3rd, Biela LM, Geor R,
Hinchcliff K, Davis M, Hamilton KL. Participation in a 1,000-mile
race increases the oxidation of carbohydrate in Alaskan sled dogs. J
Appl Physiol 118: 1502–1509, 2015. First published August 22, 2014;
doi:10.1152/japplphysiol.00588.2014.—The Alaskan Husky has been
specifically bred for endurance performance and is capable of extreme
endurance performance. We examined sled dogs in the trained state at
the beginning of the race season and after a 1,600-km race (Iditarod).
Our hypothesis was that lipids would be the predominant substrate
during submaximal exercise in long-distance racing sled dogs, and a
1,600-km race would increase the reliance on lipids during an exercise
bout at the same absolute exercise intensity. The experiments were
completed over three testing periods, which were completed in January of two different years before participation in a 1,600-km race, or
in March shortly after completion of a 1,600-km race. After determination of H13CO⫺
3 recovery, the dogs were tested with primed continuous infusions of [1,1,2,3,3-2H]glycerol, [3-13C]lactate, or [6,62
H2]glucose. During exercise, respiratory exchange ratio was significantly higher in raced (0.92 ⫾ 0.01) compared with nonraced (0.87 ⫾
0.01) dogs. During exercise, glucose rate of appearance was potentially sustained by a large glycerol rate of disappearance with an
increase in lactate rates of oxidation after a 1,600-km race. Therefore,
contrary to our hypothesis, the sled dogs were dependent on carbohydrate energy sources, a reliance that increased further after participation in a 1,600-km race.
Energy Metabolism in Sled Dogs
Table 1. Animal characteristics and experimental period
n
Age, yr
Weight, kg
Sex
All dogs
Nonraced
22
8
5⫾1
4⫾1
23.7 ⫾ 3.5
26.4 ⫾ 2.6
13 M, 9 F
6 M, 2 F
Raced
14
5⫾1
22.2 ⫾ 3.1
7 M, 7 F
Dates of Testing
January 2013 and
January 2014
March 2013
Values are means ⫾ SE; n, no. of dogs. M, male; F, female. There were no
significant differences between nonraced and raced dogs.
METHODS
Overall study design. All procedures were approved by the Oklahoma State University Institutional Animal Care and Use Committee
before the start of the experiments. The studies were performed at two
competitive racing kennels in Alaska. To complete multiple stable
isotope infusions, the experiments were completed over 2 yr. In
January 2013 and January 2014, dogs were tested in the trained state
(referred to here as nonraced). Some of the dogs had completed
shorter racing distances (160- to 420-km races), but the January
experiments were before completion of a 1,600-km race. In March
2013, experiments were performed after completion of the 1,600-km
Iditarod sled dog race with all dogs completing the Iditarod, or the
Iditarod and Yukon Quest 1,000 in the previous 2 mo (referred to here
as raced). All raced dogs were tested 5-11 days after the completion
of the Iditarod. Some dogs were used in more than one study period.
Table 1 shows the complete details of age, weight, and sex of the
dogs.
Experimental procedures. Experimentation was performed in custom-designed environmental chambers cooled to ⫺7 to ⫺12°C. Dogs
were either tested after an overnight fast, or allowed a light morning
meal (0700) if tested after 1400. On arrival to the laboratory, dogs
were weighed, and background blood and breath samples were obtained. Following background samples, a catheter was placed in the
cephalic vein (distal forearm) for stable isotope infusions. The dogs
then rested quietly in the environmental chamber for 90 min, followed
by 90 min of treadmill exercise at 10.5 km/h (6.5 mph) and 4% grade.
The same absolute intensity was used in experiments on nonraced and
raced dogs.
Stable isotope trials. The experiments were completed over three
testing periods (Table 2). In January 2013, four nonraced dogs
underwent a primed-continuous infusion of [1,1,2,3,3-2H]glycerol and
[3-13C]lactate with a priming dose of H13CO⫺
3 . In March 2013, eight
raced dogs underwent a primed-continuous infusion of [1,1,2,3,32
H]glycerol, [3-13C]lactate, and [6,6-2H2]glucose with a priming dose
of H13CO⫺
3 . Additionally, two nonraced dogs and four raced dogs
received a primed-continuous infusion of H13CO⫺
3 . Finally, in January 2014, four nonraced dogs underwent a primed-continuous infusion
of [6,6-2H2]glucose and H13CO⫺
3 . The infusion rates for all isotopes
were based on previous studies in humans (5, 17, 18) and were
confirmed for the ability to produce steady-state enrichments in pilot
trials on the dogs. After background samples of blood and breath,
additional blood was sampled by jugular venipuncture at 60, 75, and
90 min of rest, and 30, 45, 60, 75, and 90 min of exercise. Additional
breath samples were collected at 60 and 90 min of rest, and 30, 60, and
90 min of exercise. Blood for lactate analysis was collected in 8%
1503
Miller BF
perchloric acid. All other blood was collected in vacutainers containing EDTA. Blood samples were immediately chilled on ice and
centrifuged at 3,000 g for 18 min, and the supernatant was collected
and frozen until further analysis.
Given the extreme environmental conditions, the following modifications were needed when performing the isotope trials. The isotope
infusion was kept warm in an insulated box, and the lines were
insulated and rapidly thawed with a hairdryer if any occlusion was
noted. Blood sampling was via jugular venipuncture because it was
not possible to keep a jugular catheter open and in place in the cold or
keep catheters fixed in the neck scruff. Enriched breath samples were
collected into respiratory vacutainers using a custom-designed anesthesia mask. During exercise, blood sampling and collection of breath
for isotope enrichment necessitated an ⬃2-min break-in running.
Respiratory gases were collected with a custom-made helmet facemask and to an open-flow calorimetry system (Sable System, Las
Vegas, NV) that subsampled from a 1,200 l/min primary flow from
the helmet worn by the dog. Water vapor pressure, CO2, and O2
partial pressures were measured in the subsampled air, and O2
consumption (V̇O2) and CO2 production (V̇CO2) were calculated using
the equations of Lighton (14). During rest, 2 min of data were
recorded after a steady state of V̇CO2 was achieved. During exercise,
V̇O2 and V̇CO2 data were collected for ⬃3 min while the dogs were
running.
Isotope analysis. Breath samples were analyzed by IRMS for
13
CO2 enrichment by Metabolic Solutions (Nashua, NH). Glucose,
lactate, and glycerol were prepared as the pentaacetate, heptafluorobutyric anhydride, and trimethylsilyl derivatives, respectively, using
previously published procedures (5, 17, 18). Glucose, lactate, and
glycerol enrichments and concentrations were measured with an
Agilent 7890A gas chromatograph coupled to an Agilent 5975C mass
spectrometer using standard curves for both enrichment and concentration. Concentrations of glucose, lactate, and glycerol were determined against standard curves using [13C6]glucose, [13C3]lactate, or
[3H2]glycerol as the internal standards.
Glucose enrichments were measured using methane positive chemical ionization with a DB-5MS, 30-m column. Injector temperature
was set at 250°C. The oven program started at 110°C, held for 1 min,
and increased 35°C/min until 290°C was reached. Analysis was
performed with a 65:1 ml/min split injection ratio with helium as the
carrier gas. Transfer line temperature was set to 290°C, whereas
source and quadrupole temperatures were both 150°C. Ions monitored
for selective ion monitoring (SIM) were 331.0, 332.0, 333.0 and 337.0
for M⫹0, M⫹1, M⫹2 isotopomers, and the [13C6]glucose internal
standard, respectively. Lactate enrichments were measured using
electron ionization with a DB-17, 30-m column. Injector temperature
Table 2. Isotope infusion rates
Isotope
Infusion Rate
2
[1,1,2,3,3- H]glycerol
Prime, mg
Resting, mg/min
Exercise, mg/min
[3-13C]lactate
Prime, mg
Resting, mg/min
Exercise, mg/min
[6,6-2H2]glucose
Prime, mg
Resting, mg/min
Exercise, mg/min
H13CO⫺
3 (prime only), mg
H13CO⫺
3 (primed continuous)
Prime, mg
Resting, mg/min
Exercise, mg/min
J Appl Physiol • doi:10.1152/japplphysiol.00588.2014 • www.jappl.org
Dates of Testing
January 2013 and March 2013
72
0.4
1.6
January 2013 and March 2013
57.5
2.5
10
March 2013 and January 2014
200
2.0
8.0
100
100
1.7
10.3
All dates
March 2013 and January 2014
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 17, 2017
patterns during exercise in long-distance racing sled dogs, a
highly trained aerobic mammal. To do so, we examined sled
dogs in the trained state at the beginning of the race season and
after a 1,600-km race (Iditarod). Our working hypothesis was
that lipids would be the predominant substrate during submaximal exercise in long-distance racing sled dogs, and a 1,600-km
race would increase the reliance on lipids during an exercise
bout at the same absolute exercise intensity.
•
1504
Energy Metabolism in Sled Dogs
Ra ⫽ F ⫺ V
冉
C1 ⫹ C2
2
冉
冉
IE2 ⫺ IE1
t2 ⫺ t1
IE2 ⫹ IE1
Rd ⫽ Ra ⫺ V
MCR ⫽
冊冉
冉
2
C2 ⫺ C1
t2 ⫺ t1
Rd
冊
冊
冊
(1)
(2)
冊
C2 ⫹ C1
2
(3)
⫺1
where Ra and Rd are measured in mg·kg ·min for lactate and
glucose and ␮mol·kg⫺1·min⫺1 for glycerol; MCR is measured in
ml·kg⫺1·min⫺1; F represents the isotope infusion rate (mg·kg⫺1·min⫺1); and
V is the volume distribution of glucose (100 ml/kg), lactate (320
ml/kg), and glycerol (160 ml/kg) (23). C1 and C2 are concentrations at
sampling times t1 and t2; IE1 and IE2 are the excess isotopic enrichments. Measured isotopic enrichments were corrected for background
of blood samples taken before isotope infusion.
13
Recovery of H13CO⫺
CO2 was calculated as:
3 as
关共IE
CO2
兲
兴
⫻ V̇CO2 ⫻ 100
(4)
F
where IECO2 is the isotopic enrichment of expired 13CO2; V̇CO2 is the
volume of CO2 expired per minute; and F is the H13CO⫺
3 infusion
rate.
Lactate Rox was calculated from expired CO2 and IECO2:
Relative lactate oxidation共%兲 ⫽
关共IE
CO2
兲
兴
⫻ V̇CO2 ⫻ 100
Fk
Lactate Rox ⫽ Rd ⫻ relative lactate oxidation
⫺1
⫺1
(5)
(6)
13
where Rox is measured in mg·kg ·min ; F is the [3- C]lactate
infusion rate; and k is the correction factor for the retention of CO2 in
body pools, as determined by relative H13CO⫺
3 recovery (Eq. 4).
RER was calculated from V̇CO2 and V̇O2. The absolute rate of CHO
and lipid oxidation was calculated assuming negligible contribution of
protein oxidation (6):
CHO Rox ⫽
关共4.55 ⫻ V̇CO 兲 ⫺ 共3.21 ⫻ VO2兲兴 ⫻ 1, 000
2
m
Lipid Rox ⫽
关共1.67 ⫻ V̇O2兲 ⫺ 共1.67 ⫻ V̇CO 兲兴 ⫻ 1, 000
2
(8)
m
where CHO and lipid Rox are measured in mg·kg⫺1·min⫺1, V̇O2 and
V̇CO2 are in l/min, and m is dog mass (kg).
Percent oxidation (%) and TEE were calculated assuming negligible contribution of protein oxidation by the equations of Geor (7):
CHO % ⫽
TEE ⫽
冋冉
RER ⫺ 0.71
0.29
⫻ 100
Lipid % ⫽ 100 ⫺ CHO%
CHO%
100
冊
册 冋冉 冊
⫻ V̇O2 ⫻ 21.1 ⫹
Fat%
100
(9)
(10)
册
⫻ V̇O2 ⫻ 19.7
(11)
where TEE is in kJ/min, and V̇O2 is in l/min.
Statistics. Our primary goal was to measure substrate flux during
exercise in Alaskan sled dogs. Primary outcomes were calculated as
the average of the last 15 min of the resting period and the last 30 min
of exercise and are presented as means ⫾ SE. Because the dogs did
not always breathe normally in the mask during the resting period,
which led to variability, we have not presented data that involved
breath sampling at rest. For outcomes that had both resting and
exercise values, we used a two-way analysis of variance to determine
the effect of exercise (rest vs. exercise) and the effect of training
(nonraced vs. raced). If a significant interaction was detected, a least
significant difference post hoc analysis determined where those difference lie. When values were only available for exercise, we used a
Student’s paired t-test. Values were determined to be significant at
P ⱕ 0.05.
RESULTS
⫺1
Relative recovery共%兲 ⫽
Miller BF
(7)
Bicarbonate recovery in the Alaskan Husky was 97 ⫾ 6% in
the nonraced dogs and 109 ⫾ 7% in the raced dogs (data not
shown). Because group means were not significantly different,
data were combined to determine an overall mean of 102 ⫾
5%. Subsequent Rox were, therefore, left uncorrected and
assumed 100% recovery of 13C for the last 30 min of exercise.
The dogs completed the exercise task at a V̇O2 of 54.9 ⫾ 2.1
and 59.5 ⫾ 3.1 ml·kg⫺1·min⫺1 (not significantly different) for
nonraced and raced, respectively (Fig. 1A). During exercise,
RER was significantly higher in raced (0.92 ⫾ 0.01) compared
with nonraced dogs (0.87 ⫾ 0.01) (Fig. 1B). The percent CHO
oxidation was higher in raced dogs (71 ⫾ 5%) compared with
nonraced dogs (57 ⫾ 5%), and this was matched by a lower
percent lipid oxidation in raced dogs (29 ⫾ 5%) vs. nonraced
dogs (43 ⫾ 5%) (Fig. 1C). Finally, TEE was not different
between nonraced vs. raced dogs (Fig. 1D), but absolute rates
of CHO oxidation were higher in the raced dogs compared with
the nonraced dogs. There was a corresponding trend (P ⫽ 0.07)
for a lower absolute rate of lipid oxidation in the raced vs.
nonraced dogs.
Glucose concentrations were significantly greater in raced
dogs compared with nonraced dogs, and there was a significant
interaction (Fig. 2A). Post hoc analysis of the interaction
determined that nonraced exercise glucose concentration was
less than its resting value, and that raced exercise glucose
concentration was greater than nonraced exercise glucose concentration. Exercise increased glucose Ra, Rd, and MCR (Fig.
2, B–D). In addition, there was a significant decrease in MCR
for nonraced vs. raced dogs (Fig. 2D).
J Appl Physiol • doi:10.1152/japplphysiol.00588.2014 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 17, 2017
was set at 250°C. The oven program started at 60°C and then
increased at 30°C/min until 110°C was reached. Analysis was performed with a total flow of 23.5 ml/min and constant pressure of 18.2
psi in splitless mode with helium as the carrier gas. The source and
quadrupole temperatures were 230 and 150°C, respectively. The ions
monitored using SIM were 327.0, 328.0, and 330.0 for M⫹0, M⫹1,
and [13C3]lactate internal standard, respectively. Finally, glycerol
enrichments and concentration were monitored using positive chemical ionization with the injector temperature set at 250°C, and initial
oven temperature was set at 110°C and held for 3 min. Oven
temperature was then gradually increased by 45°C/min until it reached
a final temperature of 250°C. Helium was used as the carrier gas for
all analyses with splitless injection. The transfer line temperature was
set at 280°C, the source temperature was set at 280°C, and the
quadrupole temperature was set at 150°C. The ions monitored using
SIM were 159, 164, and 162 for glycerol, [2H5]glycerol, and
[2H3]glycerol internal standard, respectively.
Calculations. Glucose, lactate, and glycerol turnover [rate of appearance (Ra) and rate of disappearance (Rd)] and metabolic clearance
rates (MCR) were calculated from the equations of Steele modified for
use with stable isotopes (29):
•
Energy Metabolism in Sled Dogs
•
A
•
B
VO2 and VCO2
RER
*
0.95
*
0.90
RER
60
40
0.85
0.80
20
0.75
0
•
C
0.70
•
VO2
VCO2
D
% Oxidation
*
80
*
40
60
30
40
20
p = 0.07
20
10
0
CHO
Lipid
0
CHO
Non-Raced
Fig. 1. Respiratory measurements in nonraced
and raced Alaskan Huskies during exercise. A:
O2 consumption (V̇O2; ml·kg⫺1·min⫺1) and
CO2 production (V̇CO2; ml·kg⫺1·min⫺1). B:
respiratory exchange ratio (RER). C: percent
(%) oxidation. D: absolute rate of oxidation
(Rox; mg·kg⫺1·min⫺1) and total energy expenditure (TEE; kJ/min). Data are representative of
the last 30 min of exercise. Shaded bars are
nonraced; open bars are raced dogs. Values are
means ⫾ SE; n ⫽ 10 –11 dogs. *Significantly
different (P ⬍ 0.05) from nonraced. CHO, carbohydrates.
Lipid
Total
Raced
Lactate concentrations did not change from rest to exercise
in nonraced or raced dogs and were maintained below 2 mM
during exercise (Fig. 3A). Lactate Ra and Rd displayed a trend
(P ⫽ 0.09) for increased flux with exercise (Fig. 3, B and C),
but no difference existed between nonraced and raced. There
were no differences in lactate MCR (Fig. 3D). However,
lactate Rox during exercise was significantly greater in the
raced dogs vs. the nonraced dogs (Fig. 3E).
Glycerol concentration, Ra, and Rd (Fig. 4, A–C, respectively) were significantly greater during exercise compared
with rest. In addition, there was a significant effect of nonraced
vs. raced in that raced dogs had lower glycerol concentration
and turnover compared with nonraced dogs.
DISCUSSION
The mechanisms behind the ability to sustain prolonged
submaximal work are of great interest. We, therefore, sought to
characterize fuel use in animals capable of extremely high
aerobic endurance, Alaskan sled dogs. We hypothesized that
lipids would be the predominant substrate during submaximal
exercise in long-distance racing sled dogs, and that a 1,600-km
race would increase the reliance on lipids during an exercise
bout at the same absolute exercise intensity. Contrary to our
hypothesis, the sled dogs were dependent on CHO energy
sources, a reliance that increased further after participation in a
1,600-km race. In addition, the CHO use appeared to be
sustained by gluconeogenesis from glycerol and increased
lactate oxidation after a 1,600-km race.
Substrate partitioning. Because it is well established that a
variety of factors can change bicarbonate retention (12, 25),
thus compromising 13C-measured Rox, we were concerned
about the lack of bicarbonate priming dose and characterization
of bicarbonate recovery in the previous studies examining
substrate oxidation in dogs (27). Therefore, we used a primed
continuous infusion of H13CO⫺
3 to determine the recovery of
13
C. During exercise, there was a full recovery (100%) of 13C
from the bicarbonate, with no retention of label in a potentially
expanding bicarbonate pool. We established that a H13CO⫺
3
prime by itself is sufficient, with no further correction needed
for the calculation of 13C-measured Rox.
To our surprise, the Alaskan Husky was reliant on CHO
energy sources during low- to moderate-intensity exercise. It
was assumed that long-distance racing sled dogs sustain exercise by high Rox of free fatty acids (10). In addition, we
assumed, given the highly aerobic nature of the canine and
especially the Alaskan Husky, that there would be adaptations
that allow for sustained rates of fatty acid oxidation. In previous studies (22, 26, 27), substrate partitioning in Labrador
Retrievers was compared with pygmy goats as an example of
an evolutionarily evolved aerobic species (Canis lupus) vs. a
species with low aerobic capacity (Capra hircus). The hypothesis was that the highly aerobic canines would have greater
rates of fat oxidation than the goats. Although the researchers
found that the canines had a higher absolute rate of fat
oxidation than the goats at the same relative exercise intensity,
this was only due to the higher overall flux rate in the dogs.
Furthermore, at exercise intensities from 60 to 85% V̇O2max, the
percentage of energy contribution from CHO exceeds that of
lipid in the dogs (22). In the present study, we confirm this
CHO dependence during exercise in the Alaskan Husky.
There are several interesting aspects to the CHO dependence of the Alaskan Husky. Previous studies in yearling
Alaskan Huskies have reported an average V̇O2max of 198.7
ml·kg⫺1·min⫺1 after 4 wk of moderate-intensity training (1),
although in our observations this value could be much higher
in highly trained racing dogs. In the present study, the exercise
was completed at 54.9 ⫾ 2.1 and 59.5 ⫾ 3.1 ml·kg⫺1·min⫺1
for nonraced and raced dogs, respectively. Therefore, the
exercise task was completed at ⬃30% of V̇O2max. Furthermore,
our exercise task was completed at 10.5 km/h, while, during
J Appl Physiol • doi:10.1152/japplphysiol.00588.2014 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 17, 2017
20
40
*
)
TEE (KJ/min
60
0
Rate of Oxidation
80
(mg/kg/min)
(ml/kg/min)
1505
Miller BF
1.00
80
(%)
•
1506
Energy Metabolism in Sled Dogs
Glucose Concentration
8
B
15
Ra (mg/kg/min)
b
6
a
4
2
Glucose Ra
0
10
5
0
Rest
Exercise
Rest
* Significant effect of exercise
# Significant effect of training
$ Significant interaction
C
D
Glucose Rd
15
Glucose MCR
15
MCR (ml/kg/min)
Rd (mg/kg/min)
Exercise
10
5
0
10
5
0
Rest
Exercise
* Significant effect of exercise
Rest
Exercise
* Significant effect of exercise
# Significant effect of training
Non-Raced
the Iditarod, dogs regularly sustain speeds of 13–16 km/h for
up to 6 – 8 h while pulling a sled. It is clear that the experimental exercise task was a low percentage of the maximal
work capacity of the sled dogs. It is well established that lipid
oxidation sustains exercise at low-intensity exercise, and this
switches to a dependence on CHO oxidation at higher intensities (2). Completion of the exercise task at such a low
percentage of maximal capacity of the race dogs makes the
dominance of CHO oxidation a novel finding.
During racing conditions, the dog is fed a diet that is ⬃33%
protein, 50% fat, and 16% CHO (Eagle Ultrapack, NPAL
Analysis). If we assume 10,000 kcal/day intake during racing,
this equates to 3,300 kcal protein, 5,000 kcal fat, and 1,600
kcal CHO, or 804 g protein, 568 g fat, and 390 g CHO. By
energy percent, the 16% CHO is below what is considered a
high-CHO diet, which can reach values of 64% in highly
trained cyclists during competition (20). However, the value of
390 g for a 24-kg dog equates to a CHO intake of 16.3 g/kg
body wt, which exceeds the recommended CHO intake of
10 –12 g/kg for an extreme exercise program in humans (3).
Therefore, although on a relative scale the CHO intake is low,
on an absolute scale it far exceeds human endurance athlete
intake and could possibly sustain the prolonged exercise during
racing competition.
CHO oxidation. In the present study, not only were the dogs
CHO dependent, but, compared with nonraced dogs, the raced
dogs increased CHO oxidation during an exercise bout at the
Raced
same absolute intensity. It is known that CHO stores are
limited, and that glycogen depletion is correlated with fatigue
(8). How then do the racing dogs sustain CHO oxidation, and
how do the dogs accomplish this with relatively little glycogen
depletion previously observed (16)?
To appreciate the potential mechanisms that sustain CHO
oxidation, it is worth comparing substrate flux values between
the dogs and humans (Table 3). During exercise, glucose Ra
and lactate Rd both exceed flux rates in humans exercising at
roughly equivalent relative exercise intensities. What is more
remarkable is that the glycerol Ra and Rd greatly exceeded, by
⬃5–10 times, the rates observed in humans. Glycerol Ra is the
best measure of lipolysis. In addition, glycerol Rd largely
represents glycerol disposal in the liver, because glycerol
kinase is needed to reactivate glycerol for triglyceride synthesis. Also, at the liver, glycerol can be used as a gluconeogenic
substrate. The highly gluconeogenic capacity of dogs, especially from glycerol precursor, has been noted previously (24).
In these studies, it was determined that, when provided an
exogenous glycerol infusion, dogs increased the rate of gluconeogenesis to match glycerol Rd. In addition, the increase in
gluconeogenesis continued unabated so that the rates of conversion of glycerol to glucose far exceed those found in
humans. In the present study, we calculated that a maximum of
62% of the glucose Ra could be accounted for by glycerol Rd.
The data also indicate that the raced dogs have decreased
contribution of glycerol to gluconeogenesis. Although the rate
J Appl Physiol • doi:10.1152/japplphysiol.00588.2014 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 17, 2017
Fig. 2. Glucose concentration and glucose
kinetics during rest and exercise in nonraced
and raced Alaskan Huskies. A: glucose concentration (mM). B: glucose rate of appearance (Ra; mg·kg⫺1·min⫺1). C: glucose rate of
disappearance (Rd; mg·kg⫺1·min⫺1). D: glucose metabolic clearance rate (MCR;
ml·kg⫺1·min⫺1). Data are representative of
the last 15 min of rest and the last 30 min of
exercise. Shaded bars are nonraced; open bars
are raced dogs. Values are means ⫾ SE; n ⫽
3– 8 dogs. Significant effect of *exercise,
#training, and $interaction (P ⬍ 0.05). a Significantly different from nonraced rest (P ⬍
0.05). b Significantly different from nonraced
exercise (P ⬍ 0.05).
Concentration (mM)
A
Miller BF
•
Energy Metabolism in Sled Dogs
B
Lactate Concentration
1507
Miller BF
Lactate Ra
20
4
Ra (mg/kg/min)
Concentration (mM)
A
•
3
2
1
15
10
5
0
0
Rest
Rest
Exercise
Exercise
p = 0.09 effect of exercise
C
D
Lactate Rd
MCR (ml/kg/min)
15
10
5
80
60
40
20
0
0
Rest
Rest
Exercise
Exercise
Fig. 3. Lactate concentration and kinetics during
rest and exercise and lactate oxidation during exercise in nonraced and raced Alaskan Huskies.
A: lactate concentration (mM). B: lactate Ra
(mg·kg⫺1·min⫺1). C: lactate Rd (mg·kg⫺1·min⫺1).
D: lactate MCR (ml·kg⫺1·min⫺1). E: lactate Rox
(mg·kg⫺1·min⫺1). Data are representative of the
last 15 min of rest and the last 30 min of exercise.
Shaded bars are nonraced; open bars are raced
dogs. Values are means ⫾ SE; n ⫽ 4 – 8 dogs.
*Significantly different (P ⬍ 0.05) from nonraced.
p = 0.09 effect of exercise
E
Lacatate Rox
*
Rox (mg/kg/min)
15
10
5
0
Non-Raced
Raced
Non-Raced
Raced
decreased in raced dogs, both raced and nonraced dogs have
much higher rates than humans at comparative exercise intensity (Table 3). However, the change in glycerol Rd from
nonraced to raced suggests alternative mechanisms to sustain
CHO oxidation. Although we propose that gluconeogenesis is
the main fate of glycerol Rd, it is possible that dogs have a high
rate of triglyceride cycling, where hepatic glycerol Rd would
be used for triglyceride synthesis (9). It is also possible that this
futile cycling could be a source of heat production to maintain
body temperature in the cold (19, 21). Further studies are
needed to confirm the fate of glycerol Rd.
Before study, it was thought that lactate oxidation could
represent a significant source of CHO during exercise. During
exercise, there was no change in lactate concentration compared with resting conditions. In addition, lactate Ra and Rd
only showed a trend (P ⫽ 0.09) to increase. However, lactate
Rox doubled in the raced dogs compared with the nonraced
dogs. This increase in oxidation suggests that part of the
decrease in gluconeogenesis from glycerol could be offset by
an increase in lactate oxidation, thus decreasing glucose need.
The tradeoff between lactate and glycerol is interesting, since
lactate is another gluconeogenic substrate that has been shown
to be a preferred gluconeogenic substrate and muscle fuel (17,
18). It is worth noting that, although the changes in lactate
kinetics were relatively unimpressive, the rates of lactate turnover still exceed those found in humans at similar relative
exercise intensities. Finally, the rate of lactate oxidation and
glucose Rd cannot fully account for the amount of CHO
oxidized, as measured by RER. Therefore, there is the possibility that these dogs did not spare glycogen, or alternative
fuels, such as ketones, may change RER values.
Limitations. There are several limitations to the present
study. First, our study was cross sectional. This limitation was
a practical consideration since teams are picked in the days just
before the race. Second, our measurements only took place
over 90 min of exercise in relatively rested dogs, and we were
J Appl Physiol • doi:10.1152/japplphysiol.00588.2014 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 17, 2017
Rd (mg/kg/min)
Lactate MCR
100
20
1508
Energy Metabolism in Sled Dogs
likely at an exercise intensity that differed from racing conditions. During long-distance races, dogs regularly complete 4to 8-h runs at roughly 10 –18 km/h. Therefore, it is unclear how
the prolonged nature of that exercise and the intensity at which
Glycerol Concentration
A
Glycerol (µM)
0.3
0.2
0.1
0.0
Table 3. Comparison of substrate kinetics between Alaskan
Huskies and previously published data in exercising humans
Nonraced
Raced
Human
Glucose Ra,
mg·kg⫺1·min⫺1
Lactate Rd,
mg·kg⫺1·min⫺1
Glycerol Rd,
␮mol·kg⫺1·min⫺1
10.4 ⫾ 1.7
8.1 ⫾ 0.5
7.7 ⫾ 0.4*
10.2 ⫾ 2.5
13.4 ⫾ 1.7
8.7 ⫾ 0.6*
54.0 ⫾ 12.2
27.7 ⫾ 4.7
5.7 ⫾ 0.5†
Exercise
* Significant effect of exercise
# Significant effect of training
Glycerol Ra
B
80
60
40
20
0
Rest
Exercise
* Significant effect of exercise
# Significant effect of training
Glycerol Rd
C
80
60
40
20
0
Rest
Exercise
* Significant effect of exercise
# Significant effect of training
Non-Raced
Raced
Fig. 4. Glycerol concentration and kinetics during rest and exercise in nonraced
and raced Alaskan Huskies. A: glycerol concentration (␮M). B: glycerol Ra
(␮mol·kg⫺1·min⫺1). C: glycerol Rd (␮mol·kg⫺1·min⫺1). Data are representative of the last 15 min of rest and the last 30 min of exercise. Shaded bars are
nonraced; open bars are raced dogs. Values are means ⫾ SE; n ⫽ 4 – 8 dogs.
*Significant effect of exercise (P ⬍ 0.05). #Significant effect of training (P ⬍
0.05).
the exercise is performed changes substrate preference. For
example, changes in hormonal profiles could shift substrate
partitioning, as well as progressive glycogen depletion. Third,
we were limited in the number of experimental trials we could
complete, thus limiting the number of tracer protocols used.
We determined tracer protocols a priori to capture the most
amount of information on substrate flux and carried out the
experiments over a period of 2 yr. Because of the relatively
small number of dogs tested, the constraints placed on the
timing of testing, and the extreme environmental conditions,
our data, although unique and informative, are not yet complete. For example, 13C glucose and fatty acid should be used
to confirm RER measurements. These studies strongly indicate
that sled dogs, which are equivalent to the athletic elite of
aerobic mammals, are dependent on CHO metabolism during
exercise. Finally, total substrate oxidation is not fully accounted for by the tracers used in the present study. As
mentioned, additional carbon-labeled tracers are needed. In
addition, it is possible that alternative substrates, such as
ketones, have marked Rox, or that there is contribution from
glycogen stores in the exercising muscle or other nonexercising
muscle. Future studies will need to explore these other potential contributors.
Conclusion. We studied substrate partitioning in the Alaskan
Husky because of their remarkable ability to sustain prolonged
periods of submaximal exercise, which could provide insight
into how to sustain similar efforts in humans (i.e., endurance
athletes and military professionals). Contrary to our hypothesis, the Alaskan Husky was CHO dependent during a 90-min
bout of exercise, and this dependence increased after a
1,600-km race. It also appears that high rates of glycerol Rd
contribute to a large proportion of hepatic glucose production,
although further studies need to confirm this fate. Future
studies should confirm substrate Rox and determine the potential contribution of muscle glycogen and alternative energy
sources.
ACKNOWLEDGMENTS
We acknowledge the mushers, handlers, and helpers from SP Kennel,
Happy Trails Kennel, and Apex Kennels for the dedication to these studies. In
addition, we acknowledge the contributions of Austin Viall and Shannon
Massie.
GRANTS
This work was supported by US Army Research Office Division of Life
Sciences awards W911NF0910549 and W911NF-13-1-0091.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the author(s).
J Appl Physiol • doi:10.1152/japplphysiol.00588.2014 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 17, 2017
Rest
Ra (µmol/kg/min)
Miller BF
Values are means ⫾ SE. Ra, rate of appearance; Rd, rate of disappearance.
*Exercise at 55% maximal O2 consumption from Ref. 17. †Exercise at 45%
maximal O2 consumption from Ref. 5.
0.4
Rd (µmol/kg/min)
•
Energy Metabolism in Sled Dogs
AUTHOR CONTRIBUTIONS
Author contributions: B.F.M., K.W.H., M.D., and K.L.H. conception and
design of research; B.F.M., J.C.D., L.M.B., M.D., and K.L.H. performed
experiments; B.F.M., J.C.D., F.F.P., L.M.B., and K.L.H. analyzed data;
B.F.M., M.D., and K.L.H. interpreted results of experiments; B.F.M. and
K.L.H. prepared figures; B.F.M. and K.L.H. drafted manuscript; B.F.M., M.D.,
and K.L.H. edited and revised manuscript; B.F.M., J.C.D., F.F.P., L.M.B.,
K.W.H., M.D., and K.L.H. approved final version of manuscript.
REFERENCES
Miller BF
1509
14. Lighton JRB. Measuring Metabolic Rates: A Manual for Scientists.
Oxford, UK: Oxford University Press, 2008.
15. Longhurst JC, Musch TI, Ordway GA. O2 consumption during exercise
in dogs–roles of splenic contraction and alpha-adrenergic vasoconstriction. Am J Physiol Heart Circ Physiol 251: H502–H509, 1986.
16. McKenzie E, Holbrook T, Williamson K, Royer C, Valberg S, Hinchcliff K, Jose-Cunilleras E, Nelson S, Willard M, Davis M. Recovery of
muscle glycogen concentrations in sled dogs during prolonged exercise.
Med Sci Sports Exerc 37: 1307–1312, 2005.
17. Miller BF, Fattor JA, Jacobs KA, Horning MA, Navazio F, Lindinger
MI, Brooks GA. Lactate and glucose interactions during rest and exercise
in men: effect of exogenous lactate infusion. Journal of Physiology 544:
963–975, 2002.
18. Miller BF, Fattor JA, Jacobs KA, Horning MA, Suh SH, Navazio F,
Brooks GA. Metabolic and cardiorespiratory responses to “the lactate
clamp.” Am J Physiol Endocrinol Metab 283: E889 –E898, 2002.
19. Newsholme EA, Crabtree B. Substrate cycles in metabolic regulation and
in heat generation. Biochem Soc Symp 41: 61–109, 1976.
20. Rehrer NJ, Hellemans IJ, Rolleston AK, Rush E, Miller BF. Energy
intake and expenditure during a 6-day cycling stage race. Scandinavian
Journal of Medicine & Science in Sports 20: 609 –618, 2010.
21. Reidy SP, Weber JM. Accelerated substrate cycling: a new energywasting role for leptin in vivo. Am J Physiol Endocrinol Metab 282:
E312–E317, 2002.
22. Roberts TJ, Weber JM, Hoppeler H, Weibel ER, Taylor CR. Design
of the oxygen and substrate pathways. II. Defining the upper limits of
carbohydrate and fat oxidation. J Exp Biol 199: 1651–1658, 1996.
23. Sakurai Y, Zhang XJ, Wolfe RR. TNF directly stimulates glucose
uptake and leucine oxidation and inhibits FFA flux in conscious dogs. Am
J Physiol Endocrinol Metab 270: E864 –E872, 1996.
24. Shaw WA, Issekutz TB, Issekutz B. Gluconeogenesis from glycerol at
rest and during exercise in normal, diabetic, and methylprednisolonetreated dogs. Metab Clin Exp 25: 329 –339, 1976.
25. Trimmer JK, Casazza GA, Horning MA, Brooks GA. Recovery of
13
CO2 during rest and exercise after [1-13C]acetate, [2-13C]acetate, and
NaH13CO3 infusions. Am J Physiol Endocrinol Metab 281: E683–E692,
2001.
26. Weber JM, Brichon G, Zwingelstein G, McClelland G, Saucedo C,
Weibel ER, Taylor CR. Design of the oxygen and substrate pathways.
IV. Partitioning energy provision from fatty acids. J Exp Biol 199:
1667–1674, 1996.
27. Weber JM, Roberts TJ, Vock R, Weibel ER, Taylor CR. Design of the
oxygen and substrate pathways. III. Partitioning energy provision from
carbohydrates. J Exp Biol 199: 1659 –1666, 1996.
28. Westerterp KR, Saris WH, van Es M, Hoor ten F. Use of the doubly
labeled water technique in humans during heavy sustained exercise. J Appl
Physiol 61: 2162–2167, 1986.
29. Wolfe RR, Chinkes DL. Isotope Tracers in Metabolic Research: Principles and Practice of Kinetic Analysis (2nd Ed.). New York: Wiley, 2005.
J Appl Physiol • doi:10.1152/japplphysiol.00588.2014 • www.jappl.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 17, 2017
1. Banse HE, Banse HE, Sides RH, Sides RH, Ruby BC, Ruby BC, Bayly
WM. Effects of endurance training on V̇O2max and submaximal blood
lactate concentrations of untrained sled dogs. Equine Comp Exer Physiol
4: 89 –94, 2007.
2. Brooks GA, Mercier J. Balance of carbohydrate and lipid utilization
during exercise: the “crossover” concept. J Appl Physiol 76: 2253–2261,
1994.
3. Burke LM, Kiens B, Ivy JL. Carbohydrates and fat for training and
recovery. J Sports Sci 22: 15–30, 2004.
4. Camacho RC, Lacy DB, James FD, Coker RH, Wasserman DH.
Hepatic glucose autoregulation: responses to small, non-insulin-induced
changes in arterial glucose. Am J Physiol Endocrinol Metab 287: E269 –
E274, 2004.
5. Casazza GA, Jacobs KA, Suh SH, Miller BF, Horning MA, Brooks
GA. Menstrual cycle phase and oral contraceptive effects on triglyceride
mobilization during exercise. J Appl Physiol 97: 302–309, 2004.
6. Emhoff CAW, Messonnier LA, Horning MA, Fattor JA, Carlson TJ,
Brooks GA. Direct and indirect lactate oxidation in trained and untrained
men. J Appl Physiol 115: 829 –838, 2013.
7. Geor RJ. New methods for assessing substrate utilization in horses during
exercise. In: Advances in Equine Nutrition II. Nottingham, UK: Nottingham University Press, 2001.
8. Green HJ. How important is endogenous muscle glycogen to fatigue in
prolonged exercise? Can J Physiol Pharmacol 69: 290 –297, 1991.
9. Gross RC, Eigenbrodt EH, Farquhar JW. Endogenous triglyceride
turnover in liver and plasma of the dog. Journal of Lipid Research 8:
114 –125, 1967.
10. Hill RC. The nutritional requirements of exercising dogs. J Nutr 128:
2686S–2690S, 1998.
11. Hinchcliff KW, Reinhart GA, Burr JR, Schreier CJ, Swenson RA.
Metabolizable energy intake and sustained energy expenditure of Alaskan
sled dogs during heavy exertion in the cold. Am J Vet Res 58: 1457–1462,
1997.
12. Irving CS, Wong WW, Shulman RJ, Smith EO, Klein PD. [13C]bicarbonate kinetics in humans: intra- vs. interindividual variations. Am J
Physiol Regul Integr Comp Physiol 245: R190 –R202, 1983.
13. Kirkwood Minireview JK. A limit to metabolisable energy intake in
mammals and birds. Comp Biochem Physiol A Comp Physiol 75: 1–3,
1983.
•