Keywords
sex differences - cardiac - exercise - AMPK
Introduction
Sex differences underlie the physiology, pathophysiology, as well as social and
economic responses to several human diseases, including cardiovascular disease [1]. Ischemic heart disease, the leading cause
of death in the US [2], is one of such
diseases characterized by significant differences between biological sexes. Although
many factors likely contribute to these significant sex differences including access
to healthcare, and social and economic circumstances [3], a wealth of data also suggests that
biological and physiological factors result in different pathogenesis and treatment
outcomes in males compared to females [4].
Regular endurance exercise is the most cardioprotective intervention yet identified
[5] and elicits significant differences
between sexes. Given the importance of exercise in cardiovascular disease, it is
imperative that we understand the sex-specific mechanisms by which exercise protects
the heart. 5′-AMP-activated protein kinase (AMPK) is a key regulator of
energy balance in the heart and has recently emerged as a target for
cardioprotection [6]
[7]. AMPK activation by a single bout of
exercise and exercise training both reduce infarct size [8]
[9], highlighting the critical role of cardiac
AMPK in exercise-mediated protection. Though reports of sex differences in AMPK
exist in skeletal muscle in response to endurance exercise [10], sex-specific regulation of AMPK in the
heart has not been previously examined in response to exhaustive exercise and
training. Thus, the purpose of the current investigation was to determine the impact
of sex and endurance training on cardiac AMPK activation in response to exhaustive
exercise.
Materials and Methods
We confirm that this study meets the ethical standards of the journal [11].
Animal care
All methods described in this experiment were approved by the Institutional
Animal Care Users Committee protocol # A-3216–10. Healthy male and
female C57BL6 mice 13–16 weeks old and weighing 20–25 g
were randomly assigned to an experimental group, after which they were caged
with their litter mates. Cages were housed in a temperature-controlled room
(21°C) with 12:12-hour light/dark cycle. Animals were given ad
libitum access to fresh water and Purina® brand Laboratory
Rodent Chow.
Research design
Mice were randomly assigned to one of three groups: sedentary-resting controls
(SED-Rest; males n=5; females n=5),
sedentary-exhausted (SED-Exhausted; males n=9; females
n=8), and trained-exhausted (TR-Exhausted; males
n=10; females n=10). SED-Rest mice received no
exercise training and were sacrificed in a resting state after acclimation to
light handling. SED-Exhausted mice were acclimated to a Stanhope rodent
treadmill at slow speeds (<10 m/min,
10 min/day) for one week. TR-Exhausted mice were subjected to 12
weeks of endurance training as described below. TR-Exhausted and SED-Exhausted
animals were sacrificed immediately after a maximal exhaustive bout of exercise
that occurred 24 h after their last training bout.
Endurance exercise training protocol
TR-exhausted mice began acclimating to treadmill exercise for the first five days
of the protocol (days 0–4) and started training on day 6. Animals in the
TR-Exhausted group were exercised six days per week for a total of
12+weeks. Training sessions took place between 18:00 and 21:00, because
this was the beginning of the animal’s dark cycle. The incline remained
at 25°, and speed and duration were gradually increased until animals
could maintain 22 m/min for 60 min/day (days
36–41). This speed and duration were sustained for the remainder of the
training period ([Fig. 1a])
Fig. 1 Training and exhaustive exercise bout protocols. a
Summary of 12-week training protocol. b Summary of exhaustion
protocol for trained (TR) and sedentary (SED) exhausted mice.
Exhaustive exercise test
An exhaustive exercise test was performed using a modified graded protocol as
previously published [12]
[13]. Briefly, the test began at
15 m/min for the SED group and 20 m/min for TR
group. Treadmill speed was increased by 4 m/min every
3 min. Once at 40 m/min, the speed increased by
2 m/min every 2 min. Termination occurred when the
animal could no longer keep pace with the treadmill and ceased to run. Sacrifice
took place immediately post-exercise. TR-Exhausted animals completed the
exhaustive protocol 24 h after their last training regimen.
SED-Exhausted mice were tested 24 h after their last treadmill
acclimatization session. SED-Rest controls were sacrificed in a rested state.
The exhaustive exercise test is summarized in [Fig. 1b].
Tissue extraction, homogenization, immunoblotting
Tissue samples were collected from the apex of the left ventricle (LV), as well
as from the remaining portion of the LV and right ventricle (“whole
heart”). All animals were sacrificed via cervical dislocation. Hearts
were rapidly excised, dissected, and freeze-clamped at liquid nitrogen
temperatures. Tissue samples were weighed, mechanically homogenized in an
ice-cold lysis buffer (20 mM HEPES pH 7.4, 50 mM
β-glycerol phosphate, 2 mM EGTA, 1 mM DTT, 10 mM
NaF, 1 mM sodium orthovanadate, 1% triton-100, 10%
glycerol, protease inhibitor cocktail), centrifuged at 10 000 g
for 10 min at 4°C. A micro BCA protein assay kit was used to
determine protein concentration.
AMPK expression was measured using standard immunoblotting techniques. Briefly,
samples were loaded on a 10% polyacrylamide gel and separated by sodium
dodecyl sulfate-polyacrylamide gel electrophoresis. Proteins were transferred to
polyvinylidene difluoride membranes and blocked in tris-buffered saline and
5% nonfat milk for 1 h at room temperature. The membranes were
incubated overnight at 4°C with primary antibodies (1:1000 dilution)
specific to either total AMPK or phosphorylation of AMPKα at
Thr172. Membranes were incubated in secondary anti-rabbit
antibodies (1:1000) for 1 h at room temperature, exposed to
chemiluminescence reagents and imaged using autoradiography. Analysis of protein
bands was quantified using ImageJ.
AMPK activity
Apex cell lysates were used for the AMPK activity assay. AMPK activity was
determined using a 32P radioimmunoassay with SAMS
(His-Met-Arg-Ser-A&-Met-Ser-Gly-Leu- His-Leu-Val-Lys- Arg- Arg) as a
substrate as previously described [14]
[15]. Briefly, LV apex AMPK activity was
calculated by nanomoles of ATP incorporated into the SAMS peptide/gram
of muscle per minute.
Statistical analysis
Body weight was compared by a repeated measures 2-way ANOVA (sex ×
training). A 2-way ANOVA (sex × training) with a Tukey post hoc test was
used to compare all other outcome variables. In each case, the independent
effects of sex and training are reported with post-hoc comparisons. A multiple
linear regression model was used to investigate the impact of sex and training
on the correlation between heart weight, AMPK activity, and max running speed
using R statistical software. All other data were analyzed using JMP Version
14© statistical software (JMP, Cary, NC, USA).
P<0.05 was considered indicative of statistical significance and all
data were expressed as means±SEM.
Results
Baseline animal characteristics
Forty-eight C57BL/6 mice (males n=24; females n=24) at
15–19 weeks were used in this study. A total of 20 mice were assigned to
the TR-Exhausted group (males n=10; females n=10), 17 to the
SED-Exhausted group (males n=9; females n=8), and 11 (males
n=5; females n=6) to the SED-Rest group to serve as
non-exercised, non-exhausted controls. All animals were considered healthy as
none displayed signs or symptoms of illness or disease. As expected, male mice
had significantly higher body weights than female mice. SED and TR had similar
body weights at the beginning of the training protocol.
Indicators of training adaptation
Animals (male and female) in the TR-Exhausted groups maintained body weight
throughout the study, gaining 0.31±0.53 g, whereas the
SED-Exhausted groups gained 3.28±0.57 g, and the SED-Rest
controls gained 3.78±0.71 g, a significant effect of training.
Interestingly, male mice in the TR-Exhausted group lost weight over the 12-week
training period, whereas female mice did not ([Fig. 2a]).
Fig. 2 Training adaptations and cardiac morphometrics for male
and female trained (TR) and sedentary (SED) rest and exhausted mice.
a Body weight changes in male and female mice. b Heart
weight in male and female mice. c Heart/body weight. d Max
running speed. e Heart weight normalized to training (max running
speed). Data were analyzed by a 2-way ANOVA with sex and training as
independent variables. n=20 TR-Exhausted (10 male, 10 female);
n=18 SED-Exhausted (9 male, 9 female); and n=10 SED-Rest
(5 male, 5 female). *p<0.05 effect of training,
^p<0.05 significant effect of sex. Data are presented as
means±SEM.
TR-Exhausted animals had significantly higher (25%) maximal running
velocity compared to SED-Exhausted animals (44±0.43 vs.
33±0.68 m/min; [Fig.
2b]). TR-Exhausted animals had a significantly larger heart relative to
body weight (Ht/BW; 5.1±0.13 mg/g) compared to
SED-Exhausted (4.5±0.16 mg/g) and SED-Resting animals
(4.1±0.11 mg/g; [Fig.
2c]). Ht/BW was not influenced by sex. We also normalized
heart weight to max running velocity ([Fig.
2d]). Although we found a significant effect of training, there was no
impact of sex and no interaction between sex and training.
AMPK activity by sex and training
Cardiac AMPK activity was significantly greater in male mice compared to females
and in TR compared to SED. Post-hoc analyses indicated that both sexes activated
cardiac AMPK activity after a bout of exhaustive exercise ([Fig. 3a]). Sex and training differences in
cardiac AMPK activity were likewise found in phospho/total AMPK
expression, with a significant effect of sex and training status and higher
activation of AMPK in TR-Exhausted male than female mice ([Fig. 3b]).
Fig. 3 Left ventricular AMPK activation in trained-exhausted
(TR-Exhausted), sedentary-exhausted (SED-Exhausted), and
sedentary-resting (SED-Rest) groups for both male and female mice.
a LV AMPK activity. n=20 TR-Exhausted (10 male, 10
female); SED-Exhausted (8 male, 8 female); SED-Rest (5 male, 5 female).
*p<0.05 significant effect of training, ^p<0.05
significant effect of sex, #significant difference within sex. b
LV AMPK protein expression with representative image. n=20
TR-Exhausted (6 male, 6 female), SED-Exhausted (4 male, 4 female),
SED-Rest (4 male, 4 female). *p<0.05 significant effect
of training, ^p<0.05 significant effect of sex, #significant
difference within sex. Data were analyzed by 2-way ANOVA (sex ×
training) with post hoc analyses. Data are presented as
means±SEM.
Correlations between indicators of training adaption and cardiac AMPK
activity
We performed multiple linear regressions to determine whether there were
significant associations between indicators of training adaption and cardiac
AMPK activity and how these associations differed by sex and training status.
There was a positive correlation between Ht/BW and max running speed ([Fig. 4a]). AMPK activity correlated with
max running speed in male mice but not in female mice ([Fig. 4b]). There was no significant linear
association between AMPK activity and Ht/BW in either sex ([Fig. 4c]). Together, these associations
suggest that males demonstrate higher cardiac AMPK activity relative to aerobic
adaptation.
Fig. 4 Linear association between cardiac AMPK activity and
indicators of aerobic training in male and female mice. a Heart
weight normalized to body weight and max running speed. b AMPK
activity and max running speed. c AMPK activity and heart weight
normalized to body weight. Data were analyzed by multiple linear
regressions. n=20 TR-Exhausted (10 male, 10 female);
n=18 SED-Exhausted (9 male, 9 female); and n=10 SED-Rest
(5 male, 5 female).
Discussion
The purpose of this study was to investigate the effect of endurance exercise
training on cardiac AMPK activity in male and female C56BL/6 mice after an
exhaustive bout of exercise. The primary goal was to assess sex differences in
cardiac AMPK activity resulting from an acute exercise bout as previously found in
human skeletal muscle [10]. We found striking
fundamental differences in how male and female mice respond to cardiometabolic
stress. Regardless of training status, male mice have significantly higher levels of
cardiac AMPK activity in response to an acute bout of exhaustive exercise compared
to female mice. We also show that trained mice have a greater capacity to activate
cardiac AMPK in response to increased metabolic demand, potentially because they
were able to achieve higher workloads, thus necessitating greater AMPK activation,
although this activation also differed by sex. Consistent with the hypothesis that
greater demand necessitates greater AMPK activation, AMPK activity is suggested to
be activated in a graded fashion by exercise, with higher workloads eliciting
greater activity [16]. It is not known if
higher capacity to activate AMPK in trained mice is a requirement for higher maximal
running velocity, or vice versa. However, our data add novelty to the existing body
of work, demonstrating that despite similar exercise intensity (max running speed)
and cardiac remodeling (hypertrophy), male mice still demonstrate higher AMPK
activity than females.
Our primary finding of sex-specific activation of cardiac AMPK in response to
exhaustive exercise is interesting in light of clinical findings demonstrating
sex-specific outcomes in males and females with heart disease. Our data align with
one previous report in which three weeks of voluntary wheel running increased
cardiac pAMPK expression in male but not female mice [17]. Basal AMPK expression differences between
sexes have been reported to remain into older age, with 22-month old mice
demonstrating differential expression in the kidney and brain but not in the liver,
also suggesting tissue-specific regulation of AMPK by sex [18]. Clinically, how activation of AMPK may be
manipulated to inform disease outcomes in a sex-specific manner remains
understudied. However, our data raise the question whether AMPK activators may
protect males from ischemic disease more effectively than females. Continued sex and
tissue-specific studies of the impact of both exercise on males and females are
warranted to tailor pharmacological or lifestyle approaches for human medicine.
An exciting area of research regarding the effect of exercise on cardioprotection
involves the ability of exercise training to stimulate expression of myocardial
ATP-sensitive K+(KATP) channels. Sex differences have been
observed in the rat heart concerning these KATP channels, with sedentary
females showing significantly higher expression of both subunits. This has been
correlated to significantly lower infarct size in female rats (25%) compared
to males (37%) [19]. Additional
downstream AMPK targets might demonstrate sex-specific outcomes Similar to
downstream targets of AMPK, the mechanisms upstream responsible for sex-specific
regulation of AMPK remain somewhat unclear. Estrogen (E2) is reported to activate
AMPK through the estrogen receptor α (ERα), via phosphorylation of
Thr172 of the α-catalytic subunit [20]. However, the interplay between exercise, AMPK, and estrogen signaling
has not been fully described.
We also found differences between males and females with regard to endurance
training. Interestingly, whereas male mice lost body weight over the 12-week
training period, female mice did not, consistent with epidemiological data in women
undergoing exercise-induced weight loss programs [21]. Previous reports of sex differences in endurance training report
greater cardiac hypertrophy in female mice than males [17]. However, these investigations utilized
free wheel-running whereas our study used forced treadmill training. Indicators of
endurance training adaption suggest that our trained mice underwent sufficient
training, because we report higher maximal running velocities as well as higher
heart weight/body weight for both sexes. Further, previous work from our
group reports elevated citrate synthase activity in the quadriceps femoris of mice
who underwent a similar training protocol [12]. Thus perhaps differences (physiological, psychological) exist between
forced and voluntary exercise that explain the differences between males and females
in our data. Future work should elucidate the impact of sex on these different
exercise modalities.
Limitations and Conclusions
Limitations and Conclusions
The current work did not differentiate cardiac AMPK subunits. AMPK is composed of two
catalytic subunits, α1 and α2, with AMPKα2 being the
predominant catalytic subunit in the myocardium. Previous reports suggest that
AMPKα2 is also the predominant isoform activated following endurance
exercise training in the heart [22]. Although
we were unable to assess isoform specific activation in the current study,
unpublished data from our lab suggests that male TR-Exhausted mice have pronounced
AMPKα2 activation. Whether this remains true in female mice is unknown, and
isoform-specific investigations in both sexes are warranted. Further studies should
also elucidate the impact of age (i. e., do sex differences change across
the life course, does genetic background impact sex differences of AMPK activation),
and the therapeutic outcomes of AMPK activators and inhibitors in models of both
sexes. In conclusion, significant sex differences underlie the pathogenesis of
cardiovascular disease and the response to regular endurance exercise. Cardiac
activation of AMPK is different between male and female mice. Understanding how the
mechanisms up- and downstream of AMPK contribute to sex differences is vital if we
are to better treat heart disease in men and women.
