Accepted: 6 March 2017

DOI: 10.1111/sms.12881

ORIGINAL ARTICLE

Effects of 6-­month aerobic interval training on skeletal muscle

metabolism in middle-­aged metabolic syndrome patients

A. Guadalupe-Grau1,2   |  V. E. Fernández-Elías3,4  |  J. F. Ortega3  |  F. Dela1  | 

J. W. Helge1  |  R. Mora-Rodriguez3

1
Xlab, Department of Biomedical
Sciences, Faculty of Health
Aerobic interval training (AIT) improves the health of metabolic syndrome patients
Sciences, Center for Healthy (MetS) more than moderate intensity continuous training. However, AIT has not been
Aging, University of Copenhagen, shown to reverse all metabolic syndrome risk factors, possibly due to the limited dura-
Copenhagen, Denmark
2
tion of the training programs. Thus, we assessed the effects of 6 months of AIT on
ImFINE Research Group, Department of
Health and Human Performance, Technical cardio-­metabolic health and muscle metabolism in middle-­aged MetS. Eleven MetS
University of Madrid, Madrid, Spain (54.5±0.7 years old) underwent 6 months of 3 days a week supervised AIT program
3
Exercise Physiology Laboratory at on a cycle ergometer. Cardio-­metabolic health was assessed, and muscle biopsies
Toledo, University of Castilla-La Mancha,
were collected from the vastus lateralis prior and at the end of the program. Body fat
Toledo, Spain
4
Department of Sport Science, European
mass (−3.8%), waist circumference (−1.8%), systolic (−10.1%), and diastolic (−9.3%)
University of Madrid, Madrid, Spain blood pressure were reduced, whereas maximal fat oxidation rate and VO2peak were
significantly increased (38.9% and 8.0%, respectively; all P<.05). The remaining
Correspondence
Ricardo Mora-Rodríguez, Exercise components of cardio-­metabolic health measured (body weight, blood cholesterol,
Physiology Laboratory at Toledo, triglycerides, and glucose) were not changed after the intervention, and likewise, insu-
University of Castilla-La Mancha, Toledo,
lin sensitivity (CSi) remained unchanged. Total AMPK (23.4%), GLUT4 (20.5%),
Spain.
Email: [email protected] endothelial lipase (33.3%) protein expression, and citrate synthase activity (26.0%)
Funding information
increased with training (P<.05). Six months of AIT in MetS raises capacity for fat oxi-
Spanish Ministry of Economy and dation during exercise and increases VO2peak in combination with skeletal muscle im-
Competiveness, Grant/Award Number: provements in mitochondrial enzyme activity. Muscle proteins involved in glucose,
DEP2014-52930-R
fat metabolism, and energy cell balance improved, although this was not reflected by
parallel improvements in insulin sensitivity or blood lipid profile.

KEYWORDS
aerobic interval training, metabolic syndrome, muscle metabolism

1  |   IN T RO D U C T ION counterparts.2 Low aerobic capacity is also frequently re-

ported in MetS patients, making exercise training a core com-
The combination of increased life expectancy and seden- ponent in the treatment of the syndrome. Exercise training
tary behaviors is alarmingly augmenting the incidence of has proven its efficacy to increase metabolic flexibility and
age-­related metabolic pathologies, such as metabolic syn- cardiovascular fitness, especially those exercise interventions
drome (MetS).1 MetS is a disorder characterized by the co-­ involving high-­intensity interval training.3-5 Part of those
occurrence of obesity, hyperlipidemia, and insulin resistance. exercise-­induced improvements are thought to occur at the
In consequence, patients with MetS have an increased risk skeletal muscle level. In addition to its locomotive function,
for type 2 diabetes, cardiovascular disease, hypertension, skeletal muscle is a major determinant of whole-­body aerobic
and even all-­cause mortality compared with age-­matched capacity,6 and plays a crucial role in the regulation of fatty

Scand J Med Sci Sports. 2018;28:585–595. wileyonlinelibrary.com/journal/sms © 2017 John Wiley & Sons A/S.     585 |
Published by John Wiley & Sons Ltd
 |
586       GUADALUPE-­GRAU et al.

acid (FA) uptake, mitochondrial FA oxidation,7 basal met- phosphorylation, glycogen content, hexokinase II (HKII), EL,
abolic rate and is the primary tissue responsible for whole-­ LPL, and HSL). We also aimed to test whether the intervention
body glucose disposal.8 Studies performed by us and others is effective to induce mitochondrial biogenesis (citrate synthase
suggest that skeletal muscle energy metabolism may be dys- activity, CS), and to increase the ability to oxidize fat at the
regulated in metabolic disease patients due to decreased whole-­body level (maximal fat oxidation capacity, MFO) and
mitochondrial function and/or density,9,10 or impaired FA the mitochondrial level (β-­hydroxyacyl-­CoA-­dehydrogenase
oxidation that results in the accumulation of intramuscular activity, HAD) in a group of middle-­aged MetS with a previ-
lipids.11,12 Interestingly, 4 months of aerobic interval training ous experience in AIT (4 months). Ultimately, results of this
(AIT), that combines periods of light to moderate intensity study may help to design training based lifestyle intervention
(~70% HRmax) with sets of vigorous intensity (~90% HRmax), programs aimed at improving health among MetS patients.
is a time efficient way to increase aerobic capacity, reverse
risk factors, and induce mitochondrial biogenesis in middle-­
aged MetS individuals.9,13 Other AIT studies report mixed
2  |  M ATERIAL AND M ETHOD S
results regarding muscle insulin sensitivity, reflected by dif-
2.1  |  Participants and study design
ferent effects on the total protein levels of the glucose trans-
porter GLUT4, glycogen, hexokinase II (HKII), and insulin Eleven metabolic syndrome patients (eight men and three
receptor mediated signaling.14-16 However, the main skeletal post-­
menopausal women) with a previous experience in
muscle signaling cascades activated by exercise have not AIT9 volunteered to participate in the study. This study was
been sufficiently investigated in response to aerobic interval conducted between December 2012 and June 2013, with a
training in MetS patients. During muscle contraction, the rate minimum detraining period of 2 months from the previous
of ATP turnover increases and AMPKα phosphorylation on AIT program. Their mean±SD age, height, body weight,
its threonine residue 172 (pThr172-­AMPKα) can be elicited BMI, body fat mass, and VO2peak were 54.5±8.7 years old,
by both moderate17 and high-­intensity exercise.18-20 When 1.65±0.18 m, 90.1±14.5 kg, 32.8±1.9 kg·m−2, 31.8±3.4 kg,
the AMPK signaling cascade is activated, a higher peroxi- and 25.6±7.2 mL·kg−1·min−1. Written consent was given by
some proliferator-­activated receptor gamma coactivator 1-­α the participants after they were fully informed about the ex-
(PGC-­1α) mRNA expression is induced leading to a stimula- ercise program and the experimental procedures, and the pos-
tion of mitochondrial biogenesis and fiber type switching to- sible benefits and risks associated with the experiment. All
ward type I fibers.21,22 Another well-­characterized upstream the participants reported no cardiovascular or renal disease,
modulator of PGC-­1α is the calcium/calmodulin-­dependent peripheral vascular disease, and any disease associated with
protein kinase II (CaMKII) signaling pathway, which is ac- exercise intolerance. The study was approved by the local
tivated by the increased intracellular calcium flux with ex- Hospital’s Ethics Committee.
ercise.20,23 This signaling cascade is also thought to mediate Owing to the special characteristics of our participants
the expression of genes controlling FA oxidation, oxidative (middle-­aged MetS willing to have a muscle biopsy taken),
enzymes, and GLUT4 among others.20 it was not possible to gather a sufficient number of patients
On the other hand, muscle lipases hydrolyze blood tri- to conduct a randomized controlled trial. In consequence, a
glycerides, triglyceride-­rich lipoproteins, and chylomicrons quasi-­experimental reversal design was used, in which each
supplying the myocyte with fatty acids to oxidize or store. subject acted as their own control.27
Upon production by the underlying parenchymal cells, lipo-
protein lipase (LPL) is transported and attached to the capil-
2.2  |  Preliminary testing
lary endothelium by the protein GPIHBP1 forming endothelial
lipase (EL). Overexpression of EL decreases high-­ density All volunteers underwent medical screening to exclude indi-
lipoprotein (HDL) cholesterol levels, whereas blocking its viduals with symptoms or signs of cardiorespiratory disease.
action increases concentrations of HDL cholesterol.24 EL Prior to the start of the experiment, participants underwent
has been shown to be higher in skeletal muscle of endurance-­ a cycling graded exercise test until volitional exhaustion
trained middle-­aged men compared to age-­matched sedentary in an electrically braked cycle ergometer (Ergoselect 200;
men.25,26 However, the potential effects of long-­term exercise Ergoline, Bitz, Germany). Integrated standard 12-­lead ECG
training on LPL, EL, and other lipases important for intramus- (Quark T12, Cosmed, Rome, Italy) and blood pressure were
cular lipid droplets mobilization like the hormone-­sensitive monitored in every stage to ensure that all subjects had a nor-
lipase (HSL) are not fully elucidated in MetS patients. mal cardiovascular response to exercise. During the maximal
In this study, we tested if 6 months of AIT improves test, O2 consumption was measured by indirect calorim-
metabolic risk factors, insulin sensitivity and upregulates etry (Quark B2; Cosmed), and peak oxygen consumption
key proteins in skeletal muscle energy metabolism (Thr172-­ (VO2peak) and peak heart rate (HRpeak) were assessed as ex-
AMPK/Ser221-­ ACCβ phosphorylation, Thr286-­CaMKII plained elsewhere.9
GUADALUPE-­GRAU et al.   
|
   587

using immune-­ turbidimetry tests (iCV; 0.7%±2.1%).

2.3  |  Exercise training
HDL-­c using accelerator selective detergent method (iCV;
Participants underwent supervised aerobic interval train- 1.7%±2.9%). Blood triglycerides (TG) with glycerol-­ 3-­
ing (AIT) with a frequency of three times per week during phosphate oxidize method (iCV; 0.8%±1.7%). Total serum
6 months. Training consisted of pedaling for 10 minutes as cholesterol (TC) and high-­density lipoprotein cholesterol
warm up at 70% HRmax followed by 4×4-­minutes intervals (HDL-­c) by an enzymatic method with a single aqueous rea-
at 90% HRmax interspersed with 3-­minutes active recovery gent (iCV; 1.1%±1.4%). Plasma leptin and adiponectin con-
at 70% HRmax and a 5-­minutes cool-­down period for a total centration were measured using a specific high-­sensitivity
of 43 minutes. Exercise intensity was increased as train- human ELISA kit (R&D systems, Abingdon, UK).
ing adaptations developed to maintain the target heart rate
(Accurex coded, Polar, Kempele, Finland). Participants were
required to attend at least 85% of all the exercise sessions and
2.7  |  Insulin sensitivity
instructed to maintain their regular dietary patterns during the A 50-­minutes-­long IVGTT was used to measure insulin sen-
duration of the study. sitivity as proposed by Tura et al.29 (ie, CSI) index following
the recommendations of the ICARUS group.30 IVGTT was
performed using a glucose load of 0.5 g·kg−1 body mass with
2.4  |  Clinical investigation
a maximal dose of 35 g of glucose for participants surpassing
Before and after the 6 months training program, all the par- 70 kg of body weight. Upon drawing a baseline blood sam-
ticipants were tested for body composition, anthropometry ple, we used a 30% glucose solution (Glucosada 30%, Grifols,
(weight and waist circumference), resting blood pressure, Barcelona, Spain) manually infused at an even rate over 3 min-
blood metabolites (fasting glucose, glycated hemoglobin, utes using two 60-­mL syringes (BD Plastipak, Casarrubios
cholesterol, triglycerides, leptin, and adiponectin), muscle del Monte, Spain). Next, 5-­mL blood samples were obtained
glycogen and protein, exercise MFO, and peak oxygen con- every 10 minutes (ie, 10, 20, 30, 40, and 50 minutes), and the
sumption (VO2peak) using a graded exercise test. Blood was catheter was flushed with 3 mL 0.9% saline after every sample
drawn in the morning after a 10-­hours overnight fast. Post-­ to ensure patency. Insulin concentration was measured in du-
training tests were scheduled at least 72 hours after the last plicate using chemiluminescent microparticle immunoassay
exercise training session to avoid measuring the acute effects (iCV; 2.0%±2.8%) in an automated immunoassay analyzer
of the last exercise bout rather than the chronic effects of (Architect ci4100; Abbott Laboratories, Irving, Texas, USA).
the exercise training program. Percent body fat, trunk body The homeostasis model assessment (HOMA) was also calcu-
fat, and fat-­free mass were determined by dual energy X-­ray lated following standard procedures.31
absorptiometry (DXA Hologic Serie Discovery Wi QDR).
Supine resting blood pressure was recorded using a handheld
aneroid sphygmomanometer (Gamma GST; Heine, Munich,
2.8  |  Muscle biopsies analysis
Germany) as the average of four measurements. Muscle biopsies were obtained from m. vastus lateralis after
30 minutes of rest in the supine position using the Bergström
muscle biopsy needle technique modified to include suc-
2.5  |  Cardio-­metabolic health
tion. The muscle specimens were immediately inspected to
Peak aerobic capacity (VO2peak) was assessed on an elec- remove any visible blood, fat, or connective tissue. After
tronically braked cycle ergometer (Ergoselect 200; Ergoline) this, the muscle tissue was frozen in liquid nitrogen within
during a graded exercise testing using indirect calorimetry, <30 seconds of sampling and stored at −80°C until further
(Quark b2; Cosmed) with 12 lead ECG monitoring (Quark analysis. Glycogen concentration was determined from the
T12; Cosmed). The highest heart rate value obtained during measurement of glucose after acid hydrolysis, as previ-
the test was considered HRpeak. Maximal fat oxidation was ously reported.32 The maximal activity of the enzymes β-­
assessed in a fasted state using a graded exercise test with hydroxyacyl-­CoA-­dehydrogenase and citrate synthase was
3-­minutes stages until respiratory exchange ratio exceeded determined fluorometrically, using the methodology de-
1.0. The last minute of each stage was averaged to calculate scribed previously.33
non protein respiratory quotient and fat oxidation rate.28

2.9  |  Total protein extraction,

2.6  |  Blood analysis electrophoresis, and Western blot analysis
Plasma glucose was analyzed using the glucose oxidase-­ Muscle protein extracts were prepared as described previ-
peroxidase method with intra-­to interassay coefficient of var- ously,34 and total protein content was quantified using the
iation (iCV) of 0.9%±1.2%. Glycated hemoglobin (HbA1c) bicinchoninic acid assay.35 Briefly, proteins were solubilized
 |
588       GUADALUPE-­GRAU et al.

in sample buffer containing 0.0625 mol·L−1 Tris-­HCl, pH

2.10  |  Statistical analysis
6.8, 2.3% (wt/vol) sodium dodecyl sulfate (SDS), 10%
(vol/vol) glycerol, 5% (vol/vol) beta-­ mercaptoethanol, Data are presented as mean±SD. Normally distributed data
and 0.001% (wt/vol) bromophenol blue and separated were analyzed using Student’s two-­tailed paired t test (pre-­
on 10% criterion polyacrylamide precast gels (Bio-­ Rad, to post-­training comparison) and differences established by
Copenhagen, Denmark). After SDS electrophoresis, the gels Tukey’s post hoc analysis. Statistical significance level was
were electrophoretically transferred to polyvinylidene fluo- set at P<.05. All statistical analyses were performed using
ride (PVDF) membranes (0.2 μm pores; Bio-­Rad) using the SPSS software for windows (v.18; IBM , Chicago, Illinois,).
Trans-­Blot Turbo Transfer System (25 V in 7-­minutes pro- To determine the magnitude and meaningfulness of find-
tocol; Bio-­Rad) with Trans-­Blot Turbo Midi Transfer Packs ings, effect size statistics were calculated using Cohen’s
(Bio-­Rad). The membranes were blocked for 1.5 hours at d (G*Power Version 3.1.2 , Heinrich Heine, Düsseldorf,
room temperature with either 5% skim milk diluted in Tris-­ Germany), being the cutoffs small (from 0.2 to 0.5), medium
buffered saline (10 mmol·L−1 Tris Base, 150 mmol·L−1 (from 0.5 to 0.8), or large (over 0.8).
NaCl, pH 7.4) with 0.05% Tween-­20 (TBS-­T), or 5% bo-
vine serum albumin (BSA) TBS-­T. The membranes were
then incubated with the primary antibody overnight at 4°C.
3  |  RESULTS
To determine Thr172-­AMPKα, Ser221-­ACCβ, and Thr286-­
3.1  |  Metabolic syndrome factors
CaMKII phosphorylation levels, antibodies directed against
the phosphorylated amino acid (MWs 62, 280 and 50 KDa, Participants responses to exercise training did not differ be-
respectively, Cell Signalling, Danvers, MA, USA) and tween sex (46% males and 54% females), and thus, data
total form of these kinases (Cell Signalling for AMPK and were analyzed as a group without sex distinctions. Changes
ACCβ and Santa Cruz Biotechnology, Santa Cruz, CA for in metabolic syndrome factors after 6 months of AIT are de-
CaMKII) were diluted in 5% TBS-­T; BSA-­blocking buffer. tailed in Table 1 and Table S1. Body fat mass (−3.8%±0.4%;
GLUT4, HKII, EL, HSL, and LPL total protein expression P<.05; Cohen’s d=0.45), waist circumference (−1.8%±0.2%;
were assessed in membranes incubated with their specific P<.05; Cohen’s d=0.30), and blood pressure (−10.1±0.6
antibodies (MWs 50, 102, 70, 88, and 56 KDa, respectively, and 9.3%±0.4%, respectively, for SBP and DBP; P<.05;
Fischer Scientific, Roskilde, Denmark for GLUT4, Cell Cohen’s d=0.91 and 1.20, respectively) were the variables
Signalling, for HKII, Sigma-­Aldrich, Saint Louis, MA for that significantly improved after the training program, whereas
EL and Santa Cruz Biotechnology for HSL and LPL) di- body weight tended to be lower after the intervention (from
luted in 5% BSA-­TBS-­T blocking buffer. To control for dif- 90.1±14.5 to 88.6±14.1 kg, P=.09; Cohen’s d=0.11). In con-
ferences in loading and transfer efficiency, the membranes trast, fasting blood glucose, total cholesterol, HDL, and triglyc-
were incubated with a monoclonal mouse anti-­alpha-­tubulin erides did not change significantly after the 6 months of AIT.
(MW: 50 KDa; Biosigma, St. Louis, MO, USA) antibody
diluted in 5% skim milk TBS-­T blocking buffer. No sig-
3.2  |  Insulin sensitivity, blood parameters,
nificant changes were observed in alpha-­ tubulin protein
and exercise analysis
levels during the experiments. Antibody-­specific labeling
was revealed by incubation with a horseradish peroxidase Calculated insulin sensitivity index (CSI) from the 50-­minutes
(HRP)-­conjugated goat anti-­rabbit antibody and the HRP-­ IVGTT was not significantly altered with 6-­ month AIT.
conjugated donkey anti-­ mouse antibodies from Jackson Accordingly, HOMA values also remained stable with train-
ImmunoResearch (West Grove, PA, USA) both diluted in ing (Table 2). HbA1c, leptin, and adiponectin were not
5% skim milk TBS-­T blocking buffer and visualized with changed after 6 months of aerobic interval training (Table 2).
the ECL Western blotting detection system (Amersham Regarding exercise variables, VO2peak increased with the
Biosciences, Buckinghamshire, UK) using a CCD system intervention by 8.0% (P<.05; Cohen’s d=0.22; Table 2).
(LAS-­3000 Luminescent Image Analyser; Fujifilm, Tokyo, Maximal pedaling workload (Wmax) and MFO were increased
Japan) and quantified by the “Multi Gauge” analysis soft- (9.9%±0.8% and 38.9%±4.6%, respectively; P<.05; Cohen’s
ware (ver. 3.0; Fujifilm). All proteins were measured in du- d=0.88 and 0.27; Table 2 and Table S2).
plicate, and the variation coefficient was below 15%.
Pre-­and post-­intervention samples for each individual were
3.3  |  Mitochondrial biogenesis and fatty
loaded in each gel (26 wells) to reduce intergel variance be-
acid oxidation capacity
tween groups. In addition, two human muscle samples obtained
from a healthy young man were loaded as internal controls on Mitochondrial enzyme citrate synthase activity was in-
all gels to control for intergel variability. Overall, the variation creased 26.0% (P<.05; Cohen’s d=1.75) after the training
coefficient for the controls loaded in all gels was 11%. program, whereas mitochondrial β-­oxidation estimated from
GUADALUPE-­GRAU et al.   
   589
|
T A B L E   1   Metabolic syndrome factors
Pre-­training After 6 mo of AIT P value Cohen’s d
before and after 6 mo of aerobic interval
training (AIT) Body weight (kg) 90.1±14.5 88.6±14.1 .09 0.11
Body fat mass (kg) 31.8±3.4 30.6±2.6* .03 0.45
Waist circumference 107.2±6.8 105.3±6.7* .02 0.30
(cm)
Glucose (mmol·L−1) 6.32±0.81 6.32±0.92 .94 0.00
Triglycerides 1.15±0.40 1.39±0.80 .24 0.40
(mmol·L−1)
Total Cholesterol 4.61±0.90 4.55±0.90 .70 0.07
(mmol·L−1)
HDL (mmol·L−1) 1.00±0.12 1.15±0.10 .36 1.42
SBP (mm Hg) 134.0±17.8 120.5±13.1* .00 0.91
DBP (mm Hg) 81.5±7.9 74.0±4.9* .00 1.20
HDL, high-­ density lipoprotein; SBP, systolic blood pressure; DBP, diastolic blood pressure. Values are
mean±SD.
*Significantly different from pre-­training (P<.05)

T A B L E   2   Blood and exercise

Pre-­training After 6 mo of AIT P value Cohen’s d
variables before and after 6 mo of aerobic
interval training (AIT) HbA1c (%) 5.9±0.5 5.9±0.4 .21 0.00
HOMA 3.6±0.8 4.8±2.1 .21 0.79
CSI (×10−4 min−1 4.0±0.7 4.1±0.7 .90 0.00
[μU·mL]−1)
Leptin (pg·mL−1) 23 910±19 421 23 889±15 991 1.00 0.00
Adiponectin (pg·mL−1) 10 789±4420 9949±4424 .40 0.20
VO2peak(L·min−1) 2.3±0.9 2.5±1.0* .04 0.22
VO2peak(mL·kg−1·min−1) 25.6±7.2 27.9±8.0 .10 0.32
Maximal workload (W) 194±78 217±97* .01 0.27
−1
MFO (g·min ) 0.21±0.06 0.29±0.12* .02 0.88
HbA1c, glycated hemoglobin; MFO, maximal fat oxidation. Values are mean±SD.
*Significantly different from pre-­training (P<.05)

enzymatic β-­hydroxy-­acyl-­CoA-­dehydrogenase activity was significant when comparing pre to post-­training (Figure 2B).
not changed significantly (Figure 1A and B, respectively). Neither ACCβ total protein expression nor Ser221-­ACCβ
phosphorylation changed with the exercise intervention
(Figure 2C). Moreover, 24 weeks of stationary cycle aerobic
3.4  |  Muscle glucose metabolism
interval training did not alter the basal expression of total and
Participants muscle glycogen content remained unaltered phosphorylated Thr286-­CamKII levels (Figure 2D).
after 6 months of AIT (223±58 vs 263±75 mmol·mg−1 dry
muscle). Similarly, total protein expression of HKII was sim-
3.6  |  Muscle lipases
ilar after the intervention (Figure 2E). However, the total pro-
tein amount of the glucose transporter GLUT4 was increased EL expression showed a significant increase (33.3%±0.4%;
by 20.5% (P<.05; Cohen’s d=0.52; Figure 2F). Figure 3A; P<.05; Cohen’s d=0.85) after training compared
to baseline values, with no differences in the HSL and LPL
protein levels (Figure 3B-­C).
3.5  |  Basal skeletal muscle Thr172-­AMPKa/
Ser221-­ACCβ and Thr286-­CaMKII levels
Total AMPKα protein expression was a 23.4% higher after the 4  |  DISCUSSION
intervention compared to pre-­training levels (P<.05; Cohen’s
d=0.57; Figure  2A). The Thr172AMPKα/Total AMPKα The current study reveals that long-­term (24 weeks) aerobic
(fractional phosphorylation of AMPKα) was not statistically interval training induces a number of health benefits related
 |
590       GUADALUPE-­GRAU et al.

F I G U R E   1   Enzyme activity of (A)

3-­hydroxy acetyl-­coA-­dehydrogenase
(HAD) and (B) citrate synthase (CS) activity
before (PRE) and after (POST) 6 mo of AIT.
Solid lines represent individual values for
each participant. *Significantly higher from
pre-­training (P<.05)

F I G U R E   2   Muscle protein expression

before (PRE) and after (POST) 6 mo of
AIT. (A) Total AMPKα expression; (B)
Thr172-­AMPKα phosphorylation level; (C)
Ser221-­ACCβ phosphorylation level; (D)
CamKII phosphorylation level; (E) HKII
expression and (F) GLUT4 expression. Solid
lines represent individual values for each
participant. *Significantly higher from pre-­
training (P<.05)

to metabolic risk factors in middle-­aged MetS patients fed

4.1  |  Changes in mitochondrial biogenesis
“ad libitum.” Namely, subjects reduce their body fat, waist
markers in muscle (CS and upstream
circumference, and blood pressure (Table 1) while slightly
modulators AMPK and CaMKII) and whole-­
increasing their cardiovascular (ie, VO2peak) and more con-
body VO2peak
sistently metabolic (MFO) fitness during exercise (Table 2).
However, 24 weeks of aerobic interval training (AIT) was not In healthy subjects, it has been established that the main
enough to improve their whole-­body insulin sensitivity or their limiting factor determining maximal aerobic capacity is car-
blood lipid profile. In addition, using muscle biopsy in the vas- diac output rather than skeletal muscle oxygen utilization.36
tus lateralis, we analyzed the effects of AIT on key skeletal However, metabolically challenged persons (ie, obese, type 2
muscle proteins regulating muscle metabolism to gain insight diabetics) present impaired skeletal muscle oxidative capac-
on how exercise training may attenuate metabolic abnormali- ity, often shown by a decrease in mitochondrial density but not
ties associated with the metabolic syndrome. ­function.9,37 This misbalance could be counteracted by exercise
GUADALUPE-­GRAU et al.   
|
   591

F I G U R E   3   Muscle lipase expression before (PRE) and after (POST) 6 mo of AIT. (A) endothelial lipase expression (EL); (B) hormone-­
sensitive lipase expression (HSL) and (C) lipoprotein lipase expression (LPL). Solid lines represent individual values for each participant.
*Significantly higher from pre-­training (P<.05)

training, because MetS patients experience the habitual cardio- There is, to our knowledge, no data about the effects of long-­
vascular adaptations that improve oxygen kinetics in healthy term intermittent perturbations of the cellular energy state (eg,
subjects.38 It is somehow surprising the significant but small 24-­week interval cycling exercise training) in the skeletal mus-
effect that the training intervention exerted on the aerobic ca- cle cellular signals and enzymatic adaptation of MetS patients.
pacity of our subjects (ie, 8.0% increase in VO2peak), in com- We observed that this training increases total AMPKα protein
parison with other studies performed on MetS patients were level (Figure 2A). Nielsen and colleagues demonstrated an
they observed around a 35% increase in VO2peak.16 Among 85% higher level of total AMPKα protein content in endurance
the possible causes for this low responsiveness is the fact that trained vs sedentary healthy subjects.45 Moreover, 8 weeks of
the subjects recruited for this study participated in a previous strength training induced a significant increase in total AMPKα
4 months AIT study performed by our research group,9 where in MetS patients.46 We observed a 20% higher total AMPKα
they showed a mean increase in VO2peak of 21%. Another expla- protein levels after the training intervention, suggesting that an
nation could be that the selected training intervention may not increase in total AMPK protein content is an early event occur-
be the best therapeutic exercise training regime for longer than ring in the adaptation of MetS skeletal muscle toward a more
4 months studies, and additional volume and/or intensities are oxidative muscle phenotype. It remains unknown what training
required to increase aerobic capacity. On the other hand, there program intensity, volume, and duration would exert the best
is a positive relationship between the relative change in VO2max adaptations in MetS individuals, although the metabolic fluc-
in response to physical training and CS activity (as a surro- tuations induced by intermittent exercise appear to be a potent
gate of mitochondrial biogenesis).39 Our participants increased stimulus for PGC-­1α upstream kinases such as AMPK.
their VO2peak by 8.0%, accompanied by a 26.0% increase in Regarding calcium-­ mediated muscle signaling, existing
CS activity (Pearson correlation r=.63; P<.05). Therefore, it literature states that acute high-­intensity exercise is able to in-
seems that AIT-­induced improvements in VO2peak are coupled duce Thr286-­CaMKII activation,20,47 and short term (3 weeks)
to mitochondrial biogenesis in MetS patients. PGC-­1α mRNA of one-­legged endurance exercise training provokes a onefold
expression is also frequently used as a marker of mitochon- increase in maximal CaMKII activity and CaMKII kinase iso-
drial biogenesis, due its action as coactivator of transcription form expression in young healthy men.48 Conversely, the rest-
factors, and nuclear and mitochondrial genes required for the ing skeletal muscle levels of either pThr286-­CaMKII or total
organelle’s synthesis.40 PGC-­1α expression and activity are CaMKII remained unchanged in the present investigation. The
regulated by upstream signaling pathways of protein kinases, as lack of available data about long-­term adaptations on CaMKII
AMPK and CaMKII.23,41 We have observed that basal pThr172-­ signaling pathway in MetS individuals and the different exer-
AMPKα fractional phosphorylation did not change with the cise protocols applied make comparisons difficult, and further
training intervention, maybe because muscle samples were studies on the effects of AIT on MetS regarding calcium-­
taken in resting conditions where there is lack of an increased mediated muscle signaling are therefore warranted.
energy demand (ie, reduced ATP:AMP ratio) that would stimu-
late AMPK activation. In support for our results, other authors
4.2  |  Changes in muscle fat metabolism
have shown similar basal pThr172-­AMPKα phosphorylation
markers (β-­hydroxyacyl-­CoA-­dehydrogenase,
levels in muscle biopsies obtained from resting vastus lateralis
EL, ACC) and maximal fat oxidation induced
muscle of obese and lean subjects. 42,43 Furthermore, Kjøbsted
by exercise
et al.44 have recently reported an intact regulation of the AMPK
signaling network in response to acute exercise in skeletal mus- Our data support the classical metabolic adaptations to aero-
cle of male patients with type 2 diabetes. bic exercise training that increase the reliance on fat as energy
 |
592       GUADALUPE-­GRAU et al.

substrate during exercise,49 confirmed by a significant 36% investigation. Consequently, more mechanistically driven
increase in maximal fat oxidation capacity in our patients. studies (ie, combining in vivo protein activity and function
At the muscle cell level, activation of AMPK phosphorylates and ex vivo mitochondrial FAO respiration) are needed to
and inhibits acetyl-­coenzyme A carboxylase (ACC), lead- elucidate the physiological role of EL in human skeletal
ing to reduced malonyl-­coenzyme A and increased FA flux muscle. Moreover, although we inspected our muscle sam-
into the mitochondria via carnitine palmitoyl transferase-­1.50 ples under the microscope to remove any visible blood, fat,
ACC phosphorylation is therefore considered a marker of in- or connective tissue, it is not possible to ensure that the EL
creased FA oxidation in the muscle. In concordance with the expression detected in the blots is all from myocyte tissue.
abovementioned lack of changes in the degree of pThr172-­
AMPKα levels, pSer221-­ ACCβ remained unchanged after
4.3  |  Changes in muscle carbohydrate
24 weeks of AIT. Moreover, the independent index of mito-
metabolism muscle markers (HK, GLUT4,
chondrial fatty acid β-­oxidation, this is, hydroxy-­acyl-­CoA-­
glycogen) and insulin sensitivity (CSi, HOMA-­
dehydrogenase enzymatic activity confirmed the lack of
IR, blood glucose)
adaptation at the resting state. Our results imply that in meta-
bolic syndrome patients, the increase in the maximal capacity Most of the studies analyzing insulin sensitivity responses to
to oxidize fat during exercise is not explained by an increase AIT report small or no changes on steady state glucose infu-
in resting mitochondrial FA oxidation capacity or basal acti- sion rate during an euglycemic clamp,15 oral glucose toler-
vation of key proteins governing skeletal muscle metabolic ance tests, and insulin sensitivity indexes.9,13,38 Accordingly,
regulation (AMPK-­ACC). It can be alternatively explained the data included in this report (lack of changes in Csi,
by an increased oxygen carrying and delivery capacity of HbA1c, HOMA-­IR, and fasting blood glucose concentration)
the blood,51 or increased capacity of FA uptake or release suggest that 24 weeks of increasing intensity AIT alone is
from the lipid droplets for oxidation. We acknowledge that largely futile at decreasing insulin resistance at the whole-­
phosphorylation of AMPK-­ACC in response to acute exer- body level and that additional interventions (ie, diet or medi-
cise after a training intervention could result in substantial cine) are necessary to increase insulin sensitivity in metabolic
changes in the degree of pThr172-­AMPKα and pSer221-­ACCβ syndrome patients. This lack of adaptation was also reflected
levels, and such adaptations could also influence the results at the muscle level, where glycogen content and HKII, a key
presented in the study. We made a large effort to obtain mus- protein in the regulation of glucose metabolism remained
cle biopsies before and after the training intervention, but a unchanged. Interestingly, we observed a 20.5% increase in
limitation is the lack of acute post-­exercise muscle biopsies. the intramuscular levels of the glucose transporter GLUT4,
Eight weeks endurance training in obese subjects in- suggesting that discrete adaptations at the cellular level occur
creases the activity of lipases involved in the intramuscular before a visible change in whole-­body insulin sensitivity can
lipolysis regulation, as HSL and ATGL, but not total HSL be observed.
protein content,52 and 2 weeks physical inactivity (induced
by leg immobilization) and age negatively affect intramus-
4.4  |  Metabolic syndrome factors are not
cular lipolytic capacity in men, leading to intramuscular tri-
further reduced with 24 weeks of AIT
glycerides accumulation.25 In agreement, we observed no
differences in total HSL and LPL protein content with train- We have previously reported that in sedentary MetS pa-
ing, although of novelty, we observed higher protein amount tients 16 weeks of an AIT program could reverse metabolic
of EL with 24 weeks AIT in our MetS patients. This lipase syndrome factors in roughly one-­third of this population
encoded by the LIPG gene and synthesized by endothelial despite the absence of a diet intervention. In the present in-
cells was initially reported not to be expressed in skeletal vestigation, a 27.3% of the participants reversed metabolic
muscle53; however, a recent study suggests a possible role for syndrome factors <3. The most important changes were
EL in endothelial tissue of human skeletal muscle. Vigelsø observed in blood pressure (12% decrease), blood HDL-­c
et al. showed a 100% higher EL expression in endurance concentrations (12% increase), and to a lower extent in
trained compared to healthy middle-­aged men and suggested waist circumference (3.8%). However, plasma glucose and
that EL might be an alternative pathway for FA uptake in triglycerides concentrations were not significantly reduced.9
skeletal muscle.25,26 This EL increase can also be explained The question raised from this investigation was whether an
by a higher microvascular density with endurance training, AIT program with a longer duration could be more adequate
which has also been observed after sprint interval training in to revert the remaining MetS components to normal clini-
previously sedentary males.54 Although we found a higher cal values. Surprisingly, we have observed that 24 weeks of
expression of muscle EL with aerobic interval training, this AIT significantly reduced blood pressure and waist circum-
was not coupled to an increased basal mitochondrial HAD ference to a similar extent (10.1 and 1.8%, respectively) than
activity and we did not measure angiogenesis in the present shorter duration AIT protocols,9,16,55,56 and tended to reduce
GUADALUPE-­GRAU et al.   
|
   593

body weight with no significant effects on HDL-­c, fasting 2. Ford ES. Risks for all-­cause mortality, cardiovascular disease,
plasma glucose, and triglyceride concentrations, similarly and diabetes associated with the metabolic syndrome: a summary
to shorter studies.9,56 In addition, whole-­body fat mass was of the evidence. Diabetes Care. 2005;28:1769‐1778.
3. Galgani JE, Moro C, Ravussin E. Metabolic flexibil-
significantly reduced by 3.8%, although systemic concen-
ity and insulin resistance. Am J Physiol Endocrinol Metab.
trations of leptin and adiponectin remained unchanged. 2008;295:E1009‐E1017.
This lack of correlation between changes in fat mass stores 4. Storlien L, Oakes ND, Kelley DE. Metabolic flexibility. Proc
and leptin concentration has also been observed before in Nutr Soc. 2004;63:363‐368.
healthy men and women in response to acute and chronic 5. Weston KS, Wisloff U, Coombes JS. High-­ intensity interval
exercise,57,58 and indicates that maybe a greater fat mass loss training in patients with lifestyle-­induced cardiometabolic dis-
is needed to induce either a reduction in leptin or an increase ease: a systematic review and meta-­analysis. Br J Sports Med.
2014;48:1227‐1234.
in adiponectin levels.
6. Ivy JL, Costill DL, Maxwell BD. Skeletal muscle determinants
This study lacked a control group. However, it is important
of maximum aerobic power in man. Eur J Appl Physiol. 1980;44:
to say that our study population is very specific and difficult 1‐8.
to find (middle-­aged MetS willing to donate muscle sam- 7. Zhang L, Keung W, Samokhvalov V, Wang W, Lopaschuk GD.
ples), and that there is enough evidence in the literature about Role of fatty acid uptake and fatty acid beta-­oxidation in medi-
the beneficial effects of exercise training on MetS9,13,15,16 to ating insulin resistance in heart and skeletal muscle. Biochem
suggest that a randomized trial that condemned the patients Biophys Acta. 2010;1801:1‐22.
willing to participate in the intervention to a passive control 8. Herman MA, Kahn BB. Glucose transport and sensing in the
maintenance of glucose homeostasis and metabolic harmony. J
group was not ethically justified. Our results suggest that re-
Clin Invest. 2006;116:1767‐1775.
ducing metabolic syndrome factors is not a matter of longer 9. Mora-Rodriguez R, Ortega JF, Hamouti N, et  al. Time-­course
training programs, but maybe training programs with higher effects of aerobic interval training and detraining in pa-
training volume or intensity in each bout and/or concomitant tients with metabolic syndrome. Nutr Metab Cardiovasc Dis.
diet-­induced weight loss.13 In consequence, there is definitely 2014;24:792‐798.
a need for longer (≥24 weeks) AIT-­based studies, combining 10. Romanello V, Sandri M. Mitochondrial quality control and mus-
“ad libitum” and caloric restriction feeding conditions to con- cle mass maintenance. Front Physiol. 2015;6:422.
firm our results. 11. Schenk S, Horowitz JF. Coimmunoprecipitation of FAT/CD36
and CPT I in skeletal muscle increases proportionally with fat oxi-
dation after endurance exercise training. Am J Physiol Endocrinol
Metab. 2006;291:E254‐E260.
5  |   P ER S P E C T IV E S 12. Schenk S, Horowitz JF. Acute exercise increases triglyceride syn-
thesis in skeletal muscle and prevents fatty acid-­induced insulin
Overall, our findings show that in middle-­aged MetS men resistance. J Clin Invest. 2007;117:1690‐1698.
and women familiarized with exercise-training and fed “ad 13. Mora-Rodriguez R, Ortega JF, Guio de Prada V, et al. Effects
libitum,” 6 months of AIT slightly raises their capacity to oxi- of simultaneous or sequential weight loss diet and aerobic in-
terval training on metabolic syndrome. Int J Sports Med.
dize fat and consume oxygen during exercise in combination
2016;37:274‐281.
with skeletal muscle improvements in mitochondrial enzyme
14. Roberts CK, Hevener AL, Barnard RJ. Metabolic syndrome and
activity. Moreover, muscle proteins involved in glucose me- insulin resistance: underlying causes and modification by exercise
tabolism and energy cell balance also improved although this training. Compr Physiol. 2013;3:1‐58.
was not reflected in enhanced whole body insulin sensitiv- 15. Stuart CA, South MA, Lee ML, et al. Insulin responsiveness in
ity or blood lipid profile. Therefore, there is certainly a need metabolic syndrome after eight weeks of cycle training. Med Sci
for additional AIT studies combined with diet and/or medi- Sports Exerc. 2013;45:2021‐2029.
cation to better design exercise training interventions able to 16. Tjonna AE, Lee SJ, Rognmo O, et  al. Aerobic interval training
versus continuous moderate exercise as a treatment for the meta-
improve glucose and fat utilization at the muscle and whole-­
bolic syndrome: a pilot study. Circulation. 2008;118:346‐354.
body level. On the other hand, mechanistically driven studies 17. Roepstorff C, Thiele M, Hillig T, et al. Higher skeletal muscle al-
(ie, combining in vivo and ex vivo experiments, or in animal pha2AMPK activation and lower energy charge and fat oxidation
knockout models) are needed to elucidate the physiological in men than in women during submaximal exercise. J Physiol.
role of EL in human skeletal muscle. 2006;574:125‐138.
18. Guerra B, Guadalupe-Grau A, Fuentes T, et  al. SIRT1, AMP-­
activated protein kinase phosphorylation and downstream kinases
R E F E R E NC E S
in response to a single bout of sprint exercise: influence of glu-
1. Tzanetakou IP, Katsilambros NL, Benetos A, Mikhailidis cose ingestion. Eur J Appl Physiol. 2010;109:731‐743.
DP, Perrea DN. “Is obesity linked to aging?” Adipose tis- 19. Morales-Alamo D, Ponce-Gonzalez JG, Guadalupe-Grau A,
sue and the role of telomeres. Ageing Res Rev. 2012;11: et  al. Increased oxidative stress and anaerobic energy release,
220‐229. but blunted Thr172-­AMPKalpha phosphorylation, in response to
 |
594       GUADALUPE-­GRAU et al.

sprint exercise in severe acute hypoxia in humans. J Appl Physiol. 38. Pedersen BK, Saltin B. Exercise as medicine—evidence for
2012;113:917‐928. prescribing exercise as therapy in 26 different chronic diseases.
20. Morales-Alamo D, Ponce-Gonzalez JG, Guadalupe-Grau A, et al. Scand J Med Sci Sports. 2015;25(Suppl 3):1‐72.
Critical role for free radicals on sprint exercise-­induced CaMKII 39. Vigelso A, Andersen NB, Dela F. The relationship between skele-
and AMPKalpha phosphorylation in human skeletal muscle. J tal muscle mitochondrial citrate synthase activity and whole body
Appl Physiol. 2013;114:566‐577. oxygen uptake adaptations in response to exercise training. Int J
21. Richter EA, Ruderman NB. AMPK and the biochemistry of ex- Physiol Pathophysiol Pharmacol. 2014;6:84‐101.
ercise: implications for human health and disease. Biochem J. 40. Yan Z, Lira VA, Greene NP. Exercise training-­induced regulation
2009;418:261‐275. of mitochondrial quality. Exerc Sport Sci Rev. 2012;40:159‐164.
22. Steinberg GR, Jorgensen SB. The AMP-­activated protein kinase: 41. Lee WJ, Kim M, Park HS, et al. AMPK activation increases fatty
role in regulation of skeletal muscle metabolism and insulin sen- acid oxidation in skeletal muscle by activating PPARalpha and
sitivity. Mini Rev Med Chem. 2007;7:519‐526. PGC-­1. Biochem Biophys Res Commun. 2006;340:291‐295.
23. Abbott MJ, Edelman AM, Turcotte LP. CaMKK is an upstream 42. Bandyopadhyay GK, Yu JG, Ofrecio J, Olefsky JM. Increased
signal of AMP-­activated protein kinase in regulation of substrate malonyl-­CoA levels in muscle from obese and type 2 diabetic
metabolism in contracting skeletal muscle. Am J Physiol Regul subjects lead to decreased fatty acid oxidation and increased li-
Integr Comp Physiol. 2009;297:R1724‐R1732. pogenesis; thiazolidinedione treatment reverses these defects.
24. Das UN. Long-­ chain polyunsaturated fatty acids, endothelial Diabetes. 2006;55:2277‐2285.
lipase and atherosclerosis. Prostaglandins Leukot Essent Fatty 43. Fuentes T, Ara I, Guadalupe-Grau A, et  al. Leptin receptor
Acids. 2005;72:173‐179. 170  kDa (OB-­ R170) protein expression is reduced in obese
25. Vigelso A, Gram M, Wiuff C, et  al. Effects of immobilization human skeletal muscle: a potential mechanism of leptin resis-
and aerobic training on proteins related to intramuscular substrate tance. Exp Physiol. 2010;95:160‐171.
storage and metabolism in young and older men. Eur J Appl 44. Kjobsted R, Pedersen AJ, Hingst JR, et  al. Intact regulation of
Physiol. 2016;116:481‐494. the AMPK signaling network in response to exercise and insulin
26. Vigelso A, Prats C, Ploug T, Dela F, Helge JW. Higher muscle in skeletal muscle of male patients with type 2 diabetes: illumi-
content of perilipin 5 and endothelial lipase protein in trained than nation of AMPK activation in recovery from exercise. Diabetes.
untrained middle-­aged men. Physiol Res. 2016;65:293‐302. 2016;65:1219‐1230.
27. Masse LC, Dassa C, Gauvin L, Giles-Corti B, Motl R. Emerging 45. Nielsen JN, Mustard KJ, Graham DA, et  al. 5′-­AMP-­activated
measurement and statistical methods in physical activity research. protein kinase activity and subunit expression in exercise-­trained
Am J Prev Med. 2002;23:44‐55. human skeletal muscle. J Appl Physiol. 2003;94:631‐641.
28. Frayn KN. Calculation of substrate oxidation rates in vivo from 46. Layne AS, Nasrallah S, South MA, et al. Impaired muscle AMPK
gaseous exchange. J Appl Physiol Respir Environ Exerc Physiol. activation in the metabolic syndrome may attenuate improved
1983;55:628‐634. insulin action after exercise training. J Clin Endocrinol Metab.
29. Tura A, Sbrignadello S, Succurro E, Groop L, Sesti G, Pacini 2011;96:1815‐1826.
G. An empirical index of insulin sensitivity from short IVGTT: 47. Combes A, Dekerle J, Webborn N, Watt P, Bougault V, Daussin
validation against the minimal model and glucose clamp indices FN. Exercise-­induced metabolic fluctuations influence AMPK,
in patients with different clinical characteristics. Diabetologia. p38-­MAPK and CaMKII phosphorylation in human skeletal mus-
2010;53:144‐152. cle. Physiol Rep. 2015;3:e12462.
30. Bingley PJ, Colman P, Eisenbarth GS, et  al. Standardization of 48. Rose AJ, Frosig C, Kiens B, Wojtaszewski JF, Richter EA. Effect
IVGTT to predict IDDM. Diabetes Care. 1992;15:1313‐1316. of endurance exercise training on Ca2+ calmodulin-­dependent
31. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, protein kinase II expression and signalling in skeletal muscle of
Turner RC. Homeostasis model assessment: insulin resistance humans. J Physiol. 2007;583:785‐795.
and beta-­cell function from fasting plasma glucose and insulin 49. Holloszy JO, Coyle EF. Adaptations of skeletal muscle to endur-
concentrations in man. Diabetologia. 1985;28:412‐419. ance exercise and their metabolic consequences. J Appl Physiol.
32. Passonneau JV, Lauderdale VR. A comparison of three meth- 1984;56:831‐838.
ods of glycogen measurement in tissues. Anal Biochem. 50. Ruderman NB, Saha AK, Vavvas D, Witters LA. Malonyl-­
1974;60:405‐412. CoA, fuel sensing, and insulin resistance. Am J Physiol.
33. Andersen JL, Schjerling P, Andersen LL, Dela F. Resistance 1999;276:E1‐E18.
training and insulin action in humans: effects of de-­training. J 51. Boushel R, Saltin B. Ex vivo measures of muscle mitochondrial
Physiol. 2003;551:1049‐1058. capacity reveal quantitative limits of oxygen delivery by the cir-
34. Guerra B, Santana A, Fuentes T, et al. Leptin receptors in human culation during exercise. Int J Biochem Cell Biol. 2013;45:68‐75.
skeletal muscle. J Appl Physiol. 2007;102:1786‐1792. 52. Louche K, Badin PM, Montastier E, et  al. Endurance exercise
35. Smith PK, Krohn RI, Hermanson GT, et  al. Measurement of training up-­regulates lipolytic proteins and reduces triglyceride
protein using bicinchoninic acid. Anal Biochem. 1985;150: content in skeletal muscle of obese subjects. J Clin Endocrinol
76‐85. Metab. 2013;98:4863‐4871.
36. Andersen P, Saltin B. Maximal perfusion of skeletal muscle in 53. Jaye M, Lynch KJ, Krawiec J, et  al. A novel endothelial-­
man. J Physiol. 1985;366:233‐249. derived lipase that modulates HDL metabolism. Nat Genet.
37. Larsen S, Stride N, Hey-Mogensen M, et al. Increased mitochon- 1999;21:424‐428.
drial substrate sensitivity in skeletal muscle of patients with type 54. Cocks M, Shaw CS, Shepherd SO, et  al. Sprint interval and
2 diabetes. Diabetologia. 2011;54:1427‐1436. endurance training are equally effective in increasing muscle
GUADALUPE-­GRAU et al.   
|
   595

microvascular density and eNOS content in sedentary males. J SUPPORTING INFORMATION

Physiol. 2013;591:641‐656.
55. Sari-Sarraf V, Aliasgarzadeh A, Naderali MM, Esmaeili H, Additional Supporting Information may be found online in the
Naderali EK. A combined continuous and interval aerobic train- supporting information tab for this article.  
ing improves metabolic syndrome risk factors in men. Int J Gen
Med. 2015;8:203‐210.
56. Stensvold D, Tjonna AE, Skaug EA, et al. Strength training ver- How to cite this article: Guadalupe-Grau A,
sus aerobic interval training to modify risk factors of metabolic Fernández-Elías VE, Ortega JF, Dela F, Helge JW,
syndrome. J Appl Physiol. 2010;108:804‐810. Mora Rodríguez R. Effects of 6-­month aerobic
57. Guadalupe-Grau A, Ara I, Dorado C, et  al. Osteocalcin as interval training on skeletal muscle metabolism in
a negative regulator of serum leptin concentration in hu-
middle-­aged metabolic syndrome patients. Scand J
mans: insight from triathlon competitions. Eur J Appl Physiol.
2010;110:635‐643.
Med Sci Sports. 2018;28:585–595. https://doi.org/
58. Guadalupe-Grau A, Perez-Gomez J, Olmedillas H, et  al. 10.1111/sms.12881
Strength training combined with plyometric jumps in adults:
sex differences in fat-­bone axis adaptations. J Appl Physiol.
2009;106:1100‐1111.