Pediatric Exercise Science, 2015, 27, 67-76 http://dx.doi.org/10.1123/pes.2014-0008 © 2015 Human Kinetics, Inc.
Aerobic Training Suppresses Exercise-Induced Lipid Peroxidation and Inflammation in Overweight/Obese Adolescent Girls Hala Youssef, Carole Groussard, and Sophie Lemoine-Morel
Joel Pincemail University of Liège
University of Rennes 2
Christophe Jacob, Elie Moussa, and Abdallah Fazah
Josiane Cillard University of Rennes
University of Balamand
Jean-Claude Pineau
Arlette Delamarche
Dynamics of Human Revolution
University of Rennes
This study aimed to determine whether aerobic training could reduce lipid peroxidation and inflammation at rest and after maximal exhaustive exercise in overweight/obese adolescent girls. Thirty-nine adolescent girls (14–19 years old) were classified as nonobese or overweight/obese and then randomly assigned to either the nontrained or trained group (12-week multivariate aerobic training program). Measurements at the beginning of the experiment and at 3 months consisted of body composition, aerobic fitness (VO2peak) and the following blood assays: pre- and postexercise lipid peroxidation (15F2a-isoprostanes [F2-Isop], lipid hydroperoxide [ROOH], oxidized LDL [ox-LDL]) and inflammation (myeloperoxidase [MPO]) markers. In the overweight/ obese group, the training program significantly increased their fat-free mass (FFM) and decreased their percentage of fat mass (%FM) and hip circumference but did not modify their VO2peak. Conversely, in the nontrained overweight/obese group, weight and %FM increased, and VO2peak decreased, during the same period. Training also prevented exercise-induced lipid peroxidation and/or inflammation in overweight/obese girls (F2-Isop, ROOH, ox-LDL, MPO). In addition, in the trained overweight/obese group, exercise-induced changes in ROOH, ox-LDL and F2-Isop were correlated with improvements in anthropometric parameters (waist-to-hip ratio, %FM and FFM). In conclusion aerobic training increased tolerance to exercise-induced oxidative stress in overweight/obese adolescent girls partly as a result of improved body composition. Keywords: aerobic training, exercise, lipid peroxidation, inflammation, obesity, adolescent girls
Youssef, Groussard, Lemoine-Morel, Cillard, and Delamarche are with Laboratory of Movement, Sport, and Health (M2S), University of Rennes 2, Rennes, France. Pincemail is with the Dept. of Cardiovascular Surgery, University of Liege, Liege, Belgium. Jacob, Moussa, and Fazah are with the Laboratory of Physiology and Biomechanics of Motor Performance, University of Balamand, Balamand, Lebanon. Pineau is with the L’UPR 2147, Dynamics of Human Evolution, Paris, France. Address author correspondence to Carole Groussard at carole. [email protected].
The prevalence of adolescent obesity and its comorbidities is increasing worldwide. Adolescent obesity is associated with reduced physical activity, especially in girls (11), and alimentary disorders (22). These behavioral problems are associated with hormonal and metabolic disorders (e.g., insulin resistance and inflammation) that increase weight gain and fat mass in adolescence, especially in young girls (27). Because insulin resistance and inflammation, like obesity, promote oxidative stress (OS), obese adolescent girls are at high risk of OS (18,27). Indeed, several studies have reported higher levels of lipid
67
Downloaded by W VA Univ Library on 09/16/16, Volume 27, Article Number 1
68 Youssef et al.
peroxidation markers (2,4,27) and a lower antioxidant status (14,30) in overweight and obese adolescents compared with their nonobese counterparts. Thus, enhancing physical activity in early adolescence may prevent obesity-induced complications (e.g., cardiovascular risk factors and diabetes) in this population (12). When combined with a restricted calorie intake, aerobic training decreases resting levels of OS and/or inflammation markers (23). However, the role of exercise intervention, per se, in the decrease in OS remains unknown. Only three studies have investigated the isolated effects of aerobic training on resting parameters. Two studies failed to show any modification of OS or inflammation markers in overweight children (6,10), and the other study failed to show any modification of OS and inflammation markers in adolescent girls (16). The beneficial effects of training are not always detectable at rest but can typically be observed in response to exercise (13). Therefore, we will determine whether exercise training will reduce the exercise oxidative stress response in girls compared with no training. In our previous study (33), we demonstrated that maximal exhaustive exercise increases the level of lipid peroxidation (lipid hydroperoxides, ROOH; isoprostanes, F2-Isop) and inflammation (myeloperoxidase, MPO) markers in overweight/ obese adolescent girls but not in nonobese girls. Thus, we hypothesize that 3 months of aerobic training could reduce lipid peroxidation and inflammation after maximal exercise in adolescent girls who are overweight or obese.
Methods Subjects Thirty-nine postmenarcheal adolescent girls (ranging from 14 to 19 years old) were recruited from several high schools via personal contact. Written informed consent was obtained from the parents of each subject before the study, and the survey was approved by the ethical committee on human research of Balamand University, Lebanon. After determining body mass index values (BMI = weight [kg]/(height [m])2), the participants were stratified into two groups, a nonobese group (n = 16) and an overweight/obese group (n = 23; overweight group) based on the criteria proposed by Cole et al. (5). Waist and hip circumferences were also measured to determine the waist-to-hip ratio (WHR), and a validated (by dual energy x-ray absorption) portable device based on an ultrasound technique was used to assess the total body percentage of fat mass (%FM; 21). The nonobese and overweight subjects were randomly distributed into trained (7 nonobese and 14 overweight) and nontrained (9 nonobese and 9 overweight) subgroups. Subjects were randomized using opaque closed envelopes prepared by a researcher not involved in the study The following inclusion criteria were required for participation: 1) no regular physical activity; 2) no metabolic, cardiovascular, or other current chronic health
problems; 3) no regular drug consumption; 4) no regular smoking; 5) no birth control pill use; and 5) no antioxidant supplementation during the training period or within the previous 6 months. The participants were asked to eat and drink normally and not to consume any antioxidant supplements. To evaluate their dietary intake and to ensure that they had not changed their caloric or water consumption, the participants completed two 7-day diet records, the first at the beginning and the last at the end of the 3-month period.
Experimental Protocol The participants came to the laboratory two times: once at the beginning of the protocol (T0) and once 3 months later (T3). Each visit was conducted in the morning after the patients had fasted for 12 hr overnight. After a medical examination (electrocardiogram, blood pressure, and resting heart rate) and anthropometric measurements (height, weight, %FM by dual energy x-ray absorption [QDR-4500WE; Hologic, software version 8.26, wholebody analysis], and hip and waist circumferences), the participants ate a standardized breakfast (with the same relative calorie and macronutrient [percentage] intake in each group). One hour after breakfast, aerobic fitness (peak oxygen consumption: VO2peak) was evaluated using a graded maximal exercise test on an electrically braked cycle ergometer (Monark Ergomedic 839E Electronic Test Cycle, Varberg, Sweden). As recommended by Armstrong and Welsman (1), we used the term VO2peak, which refers to the highest VO2 attained during the graded maximal exercise test, because a plateau in VO2 rarely occurs during such a test, especially in overweight/obese people. During the exercise, the heart rate of each subject was continuously monitored via electrocardiogram (Schiller AT-102 ECG machine, Doral, FL, USA). Oxygen consumption was measured by a breath-by-breath gas analyzer (medical graphics CPX/D, Saint Paul, MN, USA). Aerobic Training Protocol. The physical training
consisted of multivariate aerobic exercises. Based on the results of the VO2peak test, the intensities of the continuous aerobic and interval training exercises were established as follow. For the continuous aerobic program, knowing the evolution of the heart rate during the VO2peak test, we targeted our intensity using the reserve heart rate (70–75% of HRres) based on the formula of Karvonen: %VO2max = %HRres = (HRex—HRrest)/(HRmax—HRrest). Concerning the interval training program, based on the results of the VO2peak test, the trained groups were divided into 4 subgroups to individualize the intensity of the interval training program. During the first field test, the girls had to run (one way and back) on a distance of 20, 25, 30, 35 m (in total 40, 50, 60, 70 m) depending on the subgroups. The back and forth run allows the girls to adapt their speed to their performance in the first lap . The girls had to perform 3 series of ten repetitions. If at the end of the first series the heart rate was less than heart
Downloaded by W VA Univ Library on 09/16/16, Volume 27, Article Number 1
Training and Oxidant Stress in Obese adolescent Girls 69
rate we targeted (always using the formula of Karvonen), the girl was transferred to the upper intensity group. The same field test was done every week to adapt the training intensity. The program was divided into three phases, for a total duration of 12 weeks, with three nonconsecutive training sessions Mondays, Wednesdays, and Fridays (of 2 hr each per week). For phase 1 (weeks 1–3), each week was planned as follows: 1) the first session of the week consisted of educative running exercises and some strengthening exercises without loads and ended with proprioception exercises and stretching exercises; it took an average of 90 min; 2) The second session of the week consisted of continuous running with moderate intensity (70% from HR reserve) with work time equal to the time of recovery work and 3) The third session of the week was devoted to group games (basketball, football, games etc.). For phase 2 (weeks 4–6), the number of sessions per week remained constant but the intensity of the sessions increased and the time decreased slightly. Each week was planned as follows: 1) The content of the first session of the week consisted of short intermittent exercise (15 s running at 70–80% of the HRres followed by 15 s of rest); 2) The second session of the week was continuous running based on dual sport such as badminton, tennis. We always emphasized the dynamism of sessions and limited the recovery; 3) The third session was always composed of team sport games but on larger space and with reduced number of players to increase the working time of each subject. The intensity was increased due to the introduction of small competitions or each team (coaches became players). For phase 3 (weeks 7–12), each week was planned as follows: 1) several exercises based on group games with the same principle of phase 2; 2) several “circuit” training exercises; and 3) 15/15 interval training running (15 s running at 80–100% of the HRres followed by 15 s of rest). The training was conducted by a professional team (members of our research group) with expertise in exercise physiology and training. A heart rate monitor was used to ensure good heart rate equivalent to the exact percentage of HRres that we targeted. Concerning the adherence, we lost 4 overweight trained subjects and 2 nonobese trained subjects for various reasons beyond the research protocol. Blood Sampling. Blood samples were collected at the same time every day in the morning for all individuals (at rest after the standardized breakfast and postexercise after the incremental test), both at the beginning of the experiment (before training—T0) to establish the baseline and again 3 months later to determine the effect of training (T3). To determine the plasma F2-Isop, ROOH, oxidized low-density lipoproteins (ox-LDL), and MPO, blood samples were collected from an antecubital vein into an EDTA Vacutainer before and immediately after exercise. Blood samples were immediately centrifuged at 1500 g for 10 min (at 4 °C) to separate the plasma (ORTO ALRESA mod.Digicen.R, Spain).
For F2-IsoP, 10 μL of an ethanolic solution of butylhydroxytoluene (100 mM) was added to 1 ml of plasma before the sample was frozen at –80 °C. The other aliquots were immediately frozen and stored at –80 °C (Angelantoni Industré SPA Biomedical division Polar 530v, Italy) until analysis. Biochemical Analysis. The inflammation marker MPO was measured using a colorimetric kit with an enzyme-linked immunosorbent assay (Elisa K 6631, Immundiagnostik, Frankfurt am Main, Germany). To evaluate lipid peroxidation markers, ROOH were quantified using the commercial OxyStat kit (Biomedica Gruppe, Divischgasse, Wien, Austria), and ox-LDL levels were determined spectrophotometrically with a competitive enzyme-linked immunosorbent assay (ELISA kit, Immundiagnostik, Germany). Total F2-IsoP was measured by liquid chromatography mass spectrometry (LC/MS), as described previously (33). Briefly, plasma was treated with KOH (15% w/v) for alkali hydrolysis at 37 °C for 1 hr. The mixture was then neutralized using a phosphate buffer, purified with an F2-isoprostane affinity column and analyzed by LC/MS. Dietary Assessment. A 7-day dietary record was provided to each participant at T0 and T3. Before recording their dietary intake, participants were asked to eat normally, not to consume any AO supplementation during the entire experimental period and to return their dietary record 1 week later. The dietary quantification was controlled and validated by an expert interviewing every participant. The intakes in energy, in AO and in macronutrients were analyzed by the same technician, using Nutrilog 1.20b software.
Statistics The results were expressed as the mean ± SEM (s.e.m), and postexercise plasma values were corrected for plasma volume variations. The data were analyzed using Statistica 7.1 software. The normality of the data distribution was determined using the Kolmogorov-Smirnov test. The descriptive variables between groups were compared using unpaired Student’s t-tests for parametric data or Mann-Whitney U tests for nonparametric data. Repeated measurements were compared between groups using two-way ANOVAs (group [nontrained and trained] and exercise [pre and post exercise]), three-way ANOVAs for the effect of training at rest (group, category [nonobese and overweight], and time [T0 and T3]), and four-way ANOVAs for effect of training on exercise response (group, category, exercise, and time). A Wilcoxon test was used for the nonparametric data. If the difference was significant, the post hoc of LSD Fisher was used. The Pearson test (parametric data) or Spearman test (nonparametric data) was used to detect correlations between variables. Correlations were performed to determine associations between training-induced changes in anthropometric/aerobic fitness parameters and training-induced changes in OS markers. p < .05 was considered significant.
Results
Downloaded by W VA Univ Library on 09/16/16, Volume 27, Article Number 1
Subject Characteristics Table 1 reports the characteristics of the subjects, 16 nonobese (9 nontrained and 7 trained) and 23 overweight (9 nontrained and 14 trained) adolescent girls. As expected, at baseline, the overweight girls presented significantly greater body weight, BMI, %FM, fat free mass (FFM), and waist and hip circumference values than the nonobese girls (Table 1). Aerobic fitness (relative VO2peak) and peak power output (power reached at VO2peak expressed in W.kg-1) were significantly lower in the overweight groups than in the nonobese groups (Table 1). It should be noted that at T0, the girls involved in the training program and the nontrained girls in each group (nonobese and overweight) showed similar anthropometric and aerobic fitness values.
The Effects of 3 Months of Training Anthropometric and Aerobic Fitness Assessments. As
reported in Table 1, the trained overweight subgroup showed a significant increase in FFM (3.2%) and significant decreases in %FM (-3.4%; 12 out of 14 have lost fat: range of variation: -12–4.2%) and hip circumference (-4.2%) after the 3-month training program. In the nontrained overweight subjects, significant increases in weight (2.6%) and %FM (5.5%) were observed after 3 months. In addition, their aerobic fitness decreased. The 3-month training program also had beneficial effects in the trained nonobese subgroup, resulting in a significant decrease in %FM (-7.6%; 7 out of 7 have lost fat: range of variation: -12.02% to -4.04%) and increases in FFM (4.5%) and relative peak power output (14.4%). Conversely, the nontrained nonobese group did not show any anthropometric changes after 3 months, and their aerobic fitness decreased. Intragroup comparisons showed that after the training period, the girls in the trained group exhibited a higher peak power output and VO2peak than the girls in the nontrained group. Improvements in aerobic fitness were associated with concomitant beneficial changes in body composition. Indeed, ΔVO2peak was negatively correlated (p < .05) with the 3-month values for weight (r = –0.64), BMI (r = –0.65), % FM (r = –0.72), and waist circumference (r = –0.74).
Lipid Peroxidation and Inflammation at Rest, Before, and After Training. Table 2 shows the values for lipid
peroxidation and inflammation-related markers at rest at T0 and T3. Baseline values of F2-Isop, ROOH and MPO did not differ between the groups, but baseline ox-LDL was higher in the nontrained overweight girls than in their nonobese counterparts (p < .05). The 3-month training program did not affect any of these parameters, except for ox-LDL in the nonobese group. Indeed, training prevented the increase in ox-LDL observed during the 3-month period in the nontrained nonobese subgroup (p < .05).
70
Lipid Peroxidation and Inflammation-Related Parameters in Response to Maximal Exhaustive Exercise. In the
nonobese groups F2-Isop, ROOH, and MPO were not modified by acute exercise, both at T0 and at T3 (data not shown), but ox-LDL was significantly increased at baseline in the trained subgroup (p < .05) (Figure 1). Three months of training considerably reduced the exercise-induced change in ox-LDL (Δexox-LDL) (p < .05) (Figure 1). In the trained overweight subgroup, exercise induced a significant increase in F2-Isop (Figure 2a), ROOH (Figure 2b), and MPO (Figure 2c) (p < .05) at T0. After the 3-month training program, the exercise-induced increases in the above markers were no longer significant. In addition, the postexercise values of F2-Isop and MPO after training were significantly lower than the pretraining values (p < .05). Furthermore, the 3-month training program significantly decreased the exercise-induced Δexox-LDL (Figure 1). In the nontrained overweight subgroup, the exerciseinduced increase in ROOH and MPO at T0 remained significant after 3 months (p < .05) (data not shown). Correlations Between Training-Induced Changes in Anthropometric Parameters and Training-Induced Changes in OS Markers. In both trained subgroups,
the changes in WHR induced by training were correlated with exercise-induced changes in ROOH (r = .52; p < .05), and the training-induced changes in %FM and hip circumference were correlated with the exercise-induced changes in ox-LDL (r = .56 and 0.51, respectively; p < .05). In addition, in the trained overweight subgroup, the variations in %FM induced by training were also positively correlated with the 3-month postexercise F2-Isop values (r = .76; p < .05), and the modifications in FFM that were induced by training were negatively correlated with the 3-month postexercise ROOH values (r = –0.53; p < .05).
Correlations Between Training-Induced Changes in Aerobic Fitness and Training-Induced Changes in OS Markers. In the trained subgroups, the changes in
peak power output induced by training were negatively correlated with the exercise-induced changes in ROOH (r = – 0.46, p < .05) and with postexercise MPO values (r = –0.48, p < .05). Furthermore, this correlation was higher for the trained overweight group than for the other subgroups (r = –0.62, p < .05).
Discussion This is the first study to examine exercise-training effects (not associated with dietary intervention) on exerciseinduced oxidative stress in overweight adolescent girls. These protective effects are associated with improvements in body composition and with a lack of the deterioration of aerobic fitness level that was observed in the nontrained group. These findings are important because many disorders associated with obesity (e.g., cardiovascular disease and diabetes) can be aggravated by higher OS and inflammation in adolescents. If left untreated,
Table 1 Anthropometric Data and Aerobic Fitness Values at T0 and T3 in Nonobese and Overweight Subjects Nonobese (n = 16) Age (years) BMI
Overweight (n = 23)
Nontrained (n = 9)
Trained (n = 7)
Nontrained (n = 9)
Trained (n = 14)
16.6 ± 0.5
17.1 ± 0.3
16.3 ± 0.5
16.1 ± 0.3
(kg/m2)
T0
21.9 ± 0.5
21.0 ± 0.3
28.3 ± 0.9
28.9 ± 1.0
T3
22.2 ± 0.7
21.4 ± 0.2
28.6 ± 0.8
28.9 ± 0.7
Δ
0.3 ± 0.2
0.4 ± 0.2
0.4 ± 0.2
–0.03 ± 0.2
54.2 ± 1.2
53.8 ± 2.2
74.6 ± 2.6$$$$
74.7 ± 2.8$$$
Weight (kg)
Downloaded by W VA Univ Library on 09/16/16, Volume 27, Article Number 1
T0
76.5 ±
2.3**
T3
55.1 ± 1.6
54.7 ± 2.1
75.5 ± 2.7
Δ
0.9 ± 0.6
0.9 ± 0.1
1.9 ± 0.7
0.8 ± 0.5
157.3 ± 0.02
159.8 ± 0.02
162.4 ± 0.01
160.5 ± 0.01
Height (cm) T0 T3
157.6 ± 0.01
160.0 ± 0.02
Δ
0.3 ± 0.001
0.2 ± 0.001
% fat mass
163.3 ±
0.01**
161.6 ± 0.01***
0.9 ± 0.002$
1.1 ± 0.003$
.
T0
27.2 ± 2.2
25.7 ± 0.9
36.1 ± 1.2$$
36.5 ± 1.8$$$
T3
27.9 ± 2.21
23.8 ± 0.98***
38.0 ± 1.89**
35.2 ± 1.84*
0.7 ± 0.7
–1.9 ± 0.3#
1.9 ± 0.5
–1.3 ± 0.4#
T0
39.6 ± 1.3
39.9 ± 1.5
47.4 ± 0.8$$$$
46.9 ± 1.4$$
T3
39.6 ± 1.1
41.6 ± 1.5**
47.2 ± 0.9
48.4 ± 1.3**
–0.25 ± 0.6
1.5 ± 0.4#
Δ Fat-free mass (kg)
Δ
0.5#
0.0 ± 0.5
1.7 ±
T0
95.8 ± 1.1
95.1 ± 1.8
110.8 ± 1.9$$$$
111.3 ± 1.8$$$$
T3
96.2 ± 0.9
95.1 ± 1.8
110.5 ± 1.7
108.5 ± 2.3*
Δ
0.4 ± 0.9
0.0 ± 0.3
–0.3 ± 1.0
–2.8 ± 1.1
1.88 ± 0.05
1.99 ± 0.12
1.65 ± 0.05$
1.60 ± 0.07$
1.54 ± 0.06
1.55 ± 0.05$$$$
Hip circumference (cm)
Power max (W·kg-1) T0
2.28 ±
0.15**#
T3
1.73 ± 0.12
Δ
–0.15 ± 0.12
0.29 ± 0.06#
–0.11 ± 0.07
–0.05 ± 0.04$
32.4 ± 1.7
34.3 ± 2.1
30.0 ± 0.4$
29.1 ± 1.1$
2.1#
0.7*
28.2 ± 1.1$
–2.4 ± 1.0
–0.9 ± 0.7$
V. O2peak (ml·kg-1·min-1) T0 T3
29.5 ±
Δ
1.7**
–2.9 ± 1.9
35.5 ±
1.2 ± 0.9
27.6 ±
Note. Values are in means ± SEM. BMI = body mass index, WHR = waist to hip ratio, W = watt, V. O2peak = peak rate of oxygen consumption. ∆ = the change in the variable from baseline to 3 months. $Significant
difference from nonobese group at T0.
#Significant
difference from nontrained subgroup.
*Significant
difference from T0 to T3.
$p
Aerobic Training Suppresses Exercise-Induced Lipid Peroxidation and Inflammation in Overweight/Obese Adolescent Girls Hala Youssef, Carole Groussard, and Sophie Lemoine-Morel
Joel Pincemail University of Liège
University of Rennes 2
Christophe Jacob, Elie Moussa, and Abdallah Fazah
Josiane Cillard University of Rennes
University of Balamand
Jean-Claude Pineau
Arlette Delamarche
Dynamics of Human Revolution
University of Rennes
This study aimed to determine whether aerobic training could reduce lipid peroxidation and inflammation at rest and after maximal exhaustive exercise in overweight/obese adolescent girls. Thirty-nine adolescent girls (14–19 years old) were classified as nonobese or overweight/obese and then randomly assigned to either the nontrained or trained group (12-week multivariate aerobic training program). Measurements at the beginning of the experiment and at 3 months consisted of body composition, aerobic fitness (VO2peak) and the following blood assays: pre- and postexercise lipid peroxidation (15F2a-isoprostanes [F2-Isop], lipid hydroperoxide [ROOH], oxidized LDL [ox-LDL]) and inflammation (myeloperoxidase [MPO]) markers. In the overweight/ obese group, the training program significantly increased their fat-free mass (FFM) and decreased their percentage of fat mass (%FM) and hip circumference but did not modify their VO2peak. Conversely, in the nontrained overweight/obese group, weight and %FM increased, and VO2peak decreased, during the same period. Training also prevented exercise-induced lipid peroxidation and/or inflammation in overweight/obese girls (F2-Isop, ROOH, ox-LDL, MPO). In addition, in the trained overweight/obese group, exercise-induced changes in ROOH, ox-LDL and F2-Isop were correlated with improvements in anthropometric parameters (waist-to-hip ratio, %FM and FFM). In conclusion aerobic training increased tolerance to exercise-induced oxidative stress in overweight/obese adolescent girls partly as a result of improved body composition. Keywords: aerobic training, exercise, lipid peroxidation, inflammation, obesity, adolescent girls
Youssef, Groussard, Lemoine-Morel, Cillard, and Delamarche are with Laboratory of Movement, Sport, and Health (M2S), University of Rennes 2, Rennes, France. Pincemail is with the Dept. of Cardiovascular Surgery, University of Liege, Liege, Belgium. Jacob, Moussa, and Fazah are with the Laboratory of Physiology and Biomechanics of Motor Performance, University of Balamand, Balamand, Lebanon. Pineau is with the L’UPR 2147, Dynamics of Human Evolution, Paris, France. Address author correspondence to Carole Groussard at carole. [email protected].
The prevalence of adolescent obesity and its comorbidities is increasing worldwide. Adolescent obesity is associated with reduced physical activity, especially in girls (11), and alimentary disorders (22). These behavioral problems are associated with hormonal and metabolic disorders (e.g., insulin resistance and inflammation) that increase weight gain and fat mass in adolescence, especially in young girls (27). Because insulin resistance and inflammation, like obesity, promote oxidative stress (OS), obese adolescent girls are at high risk of OS (18,27). Indeed, several studies have reported higher levels of lipid
67
Downloaded by W VA Univ Library on 09/16/16, Volume 27, Article Number 1
68 Youssef et al.
peroxidation markers (2,4,27) and a lower antioxidant status (14,30) in overweight and obese adolescents compared with their nonobese counterparts. Thus, enhancing physical activity in early adolescence may prevent obesity-induced complications (e.g., cardiovascular risk factors and diabetes) in this population (12). When combined with a restricted calorie intake, aerobic training decreases resting levels of OS and/or inflammation markers (23). However, the role of exercise intervention, per se, in the decrease in OS remains unknown. Only three studies have investigated the isolated effects of aerobic training on resting parameters. Two studies failed to show any modification of OS or inflammation markers in overweight children (6,10), and the other study failed to show any modification of OS and inflammation markers in adolescent girls (16). The beneficial effects of training are not always detectable at rest but can typically be observed in response to exercise (13). Therefore, we will determine whether exercise training will reduce the exercise oxidative stress response in girls compared with no training. In our previous study (33), we demonstrated that maximal exhaustive exercise increases the level of lipid peroxidation (lipid hydroperoxides, ROOH; isoprostanes, F2-Isop) and inflammation (myeloperoxidase, MPO) markers in overweight/ obese adolescent girls but not in nonobese girls. Thus, we hypothesize that 3 months of aerobic training could reduce lipid peroxidation and inflammation after maximal exercise in adolescent girls who are overweight or obese.
Methods Subjects Thirty-nine postmenarcheal adolescent girls (ranging from 14 to 19 years old) were recruited from several high schools via personal contact. Written informed consent was obtained from the parents of each subject before the study, and the survey was approved by the ethical committee on human research of Balamand University, Lebanon. After determining body mass index values (BMI = weight [kg]/(height [m])2), the participants were stratified into two groups, a nonobese group (n = 16) and an overweight/obese group (n = 23; overweight group) based on the criteria proposed by Cole et al. (5). Waist and hip circumferences were also measured to determine the waist-to-hip ratio (WHR), and a validated (by dual energy x-ray absorption) portable device based on an ultrasound technique was used to assess the total body percentage of fat mass (%FM; 21). The nonobese and overweight subjects were randomly distributed into trained (7 nonobese and 14 overweight) and nontrained (9 nonobese and 9 overweight) subgroups. Subjects were randomized using opaque closed envelopes prepared by a researcher not involved in the study The following inclusion criteria were required for participation: 1) no regular physical activity; 2) no metabolic, cardiovascular, or other current chronic health
problems; 3) no regular drug consumption; 4) no regular smoking; 5) no birth control pill use; and 5) no antioxidant supplementation during the training period or within the previous 6 months. The participants were asked to eat and drink normally and not to consume any antioxidant supplements. To evaluate their dietary intake and to ensure that they had not changed their caloric or water consumption, the participants completed two 7-day diet records, the first at the beginning and the last at the end of the 3-month period.
Experimental Protocol The participants came to the laboratory two times: once at the beginning of the protocol (T0) and once 3 months later (T3). Each visit was conducted in the morning after the patients had fasted for 12 hr overnight. After a medical examination (electrocardiogram, blood pressure, and resting heart rate) and anthropometric measurements (height, weight, %FM by dual energy x-ray absorption [QDR-4500WE; Hologic, software version 8.26, wholebody analysis], and hip and waist circumferences), the participants ate a standardized breakfast (with the same relative calorie and macronutrient [percentage] intake in each group). One hour after breakfast, aerobic fitness (peak oxygen consumption: VO2peak) was evaluated using a graded maximal exercise test on an electrically braked cycle ergometer (Monark Ergomedic 839E Electronic Test Cycle, Varberg, Sweden). As recommended by Armstrong and Welsman (1), we used the term VO2peak, which refers to the highest VO2 attained during the graded maximal exercise test, because a plateau in VO2 rarely occurs during such a test, especially in overweight/obese people. During the exercise, the heart rate of each subject was continuously monitored via electrocardiogram (Schiller AT-102 ECG machine, Doral, FL, USA). Oxygen consumption was measured by a breath-by-breath gas analyzer (medical graphics CPX/D, Saint Paul, MN, USA). Aerobic Training Protocol. The physical training
consisted of multivariate aerobic exercises. Based on the results of the VO2peak test, the intensities of the continuous aerobic and interval training exercises were established as follow. For the continuous aerobic program, knowing the evolution of the heart rate during the VO2peak test, we targeted our intensity using the reserve heart rate (70–75% of HRres) based on the formula of Karvonen: %VO2max = %HRres = (HRex—HRrest)/(HRmax—HRrest). Concerning the interval training program, based on the results of the VO2peak test, the trained groups were divided into 4 subgroups to individualize the intensity of the interval training program. During the first field test, the girls had to run (one way and back) on a distance of 20, 25, 30, 35 m (in total 40, 50, 60, 70 m) depending on the subgroups. The back and forth run allows the girls to adapt their speed to their performance in the first lap . The girls had to perform 3 series of ten repetitions. If at the end of the first series the heart rate was less than heart
Downloaded by W VA Univ Library on 09/16/16, Volume 27, Article Number 1
Training and Oxidant Stress in Obese adolescent Girls 69
rate we targeted (always using the formula of Karvonen), the girl was transferred to the upper intensity group. The same field test was done every week to adapt the training intensity. The program was divided into three phases, for a total duration of 12 weeks, with three nonconsecutive training sessions Mondays, Wednesdays, and Fridays (of 2 hr each per week). For phase 1 (weeks 1–3), each week was planned as follows: 1) the first session of the week consisted of educative running exercises and some strengthening exercises without loads and ended with proprioception exercises and stretching exercises; it took an average of 90 min; 2) The second session of the week consisted of continuous running with moderate intensity (70% from HR reserve) with work time equal to the time of recovery work and 3) The third session of the week was devoted to group games (basketball, football, games etc.). For phase 2 (weeks 4–6), the number of sessions per week remained constant but the intensity of the sessions increased and the time decreased slightly. Each week was planned as follows: 1) The content of the first session of the week consisted of short intermittent exercise (15 s running at 70–80% of the HRres followed by 15 s of rest); 2) The second session of the week was continuous running based on dual sport such as badminton, tennis. We always emphasized the dynamism of sessions and limited the recovery; 3) The third session was always composed of team sport games but on larger space and with reduced number of players to increase the working time of each subject. The intensity was increased due to the introduction of small competitions or each team (coaches became players). For phase 3 (weeks 7–12), each week was planned as follows: 1) several exercises based on group games with the same principle of phase 2; 2) several “circuit” training exercises; and 3) 15/15 interval training running (15 s running at 80–100% of the HRres followed by 15 s of rest). The training was conducted by a professional team (members of our research group) with expertise in exercise physiology and training. A heart rate monitor was used to ensure good heart rate equivalent to the exact percentage of HRres that we targeted. Concerning the adherence, we lost 4 overweight trained subjects and 2 nonobese trained subjects for various reasons beyond the research protocol. Blood Sampling. Blood samples were collected at the same time every day in the morning for all individuals (at rest after the standardized breakfast and postexercise after the incremental test), both at the beginning of the experiment (before training—T0) to establish the baseline and again 3 months later to determine the effect of training (T3). To determine the plasma F2-Isop, ROOH, oxidized low-density lipoproteins (ox-LDL), and MPO, blood samples were collected from an antecubital vein into an EDTA Vacutainer before and immediately after exercise. Blood samples were immediately centrifuged at 1500 g for 10 min (at 4 °C) to separate the plasma (ORTO ALRESA mod.Digicen.R, Spain).
For F2-IsoP, 10 μL of an ethanolic solution of butylhydroxytoluene (100 mM) was added to 1 ml of plasma before the sample was frozen at –80 °C. The other aliquots were immediately frozen and stored at –80 °C (Angelantoni Industré SPA Biomedical division Polar 530v, Italy) until analysis. Biochemical Analysis. The inflammation marker MPO was measured using a colorimetric kit with an enzyme-linked immunosorbent assay (Elisa K 6631, Immundiagnostik, Frankfurt am Main, Germany). To evaluate lipid peroxidation markers, ROOH were quantified using the commercial OxyStat kit (Biomedica Gruppe, Divischgasse, Wien, Austria), and ox-LDL levels were determined spectrophotometrically with a competitive enzyme-linked immunosorbent assay (ELISA kit, Immundiagnostik, Germany). Total F2-IsoP was measured by liquid chromatography mass spectrometry (LC/MS), as described previously (33). Briefly, plasma was treated with KOH (15% w/v) for alkali hydrolysis at 37 °C for 1 hr. The mixture was then neutralized using a phosphate buffer, purified with an F2-isoprostane affinity column and analyzed by LC/MS. Dietary Assessment. A 7-day dietary record was provided to each participant at T0 and T3. Before recording their dietary intake, participants were asked to eat normally, not to consume any AO supplementation during the entire experimental period and to return their dietary record 1 week later. The dietary quantification was controlled and validated by an expert interviewing every participant. The intakes in energy, in AO and in macronutrients were analyzed by the same technician, using Nutrilog 1.20b software.
Statistics The results were expressed as the mean ± SEM (s.e.m), and postexercise plasma values were corrected for plasma volume variations. The data were analyzed using Statistica 7.1 software. The normality of the data distribution was determined using the Kolmogorov-Smirnov test. The descriptive variables between groups were compared using unpaired Student’s t-tests for parametric data or Mann-Whitney U tests for nonparametric data. Repeated measurements were compared between groups using two-way ANOVAs (group [nontrained and trained] and exercise [pre and post exercise]), three-way ANOVAs for the effect of training at rest (group, category [nonobese and overweight], and time [T0 and T3]), and four-way ANOVAs for effect of training on exercise response (group, category, exercise, and time). A Wilcoxon test was used for the nonparametric data. If the difference was significant, the post hoc of LSD Fisher was used. The Pearson test (parametric data) or Spearman test (nonparametric data) was used to detect correlations between variables. Correlations were performed to determine associations between training-induced changes in anthropometric/aerobic fitness parameters and training-induced changes in OS markers. p < .05 was considered significant.
Results
Downloaded by W VA Univ Library on 09/16/16, Volume 27, Article Number 1
Subject Characteristics Table 1 reports the characteristics of the subjects, 16 nonobese (9 nontrained and 7 trained) and 23 overweight (9 nontrained and 14 trained) adolescent girls. As expected, at baseline, the overweight girls presented significantly greater body weight, BMI, %FM, fat free mass (FFM), and waist and hip circumference values than the nonobese girls (Table 1). Aerobic fitness (relative VO2peak) and peak power output (power reached at VO2peak expressed in W.kg-1) were significantly lower in the overweight groups than in the nonobese groups (Table 1). It should be noted that at T0, the girls involved in the training program and the nontrained girls in each group (nonobese and overweight) showed similar anthropometric and aerobic fitness values.
The Effects of 3 Months of Training Anthropometric and Aerobic Fitness Assessments. As
reported in Table 1, the trained overweight subgroup showed a significant increase in FFM (3.2%) and significant decreases in %FM (-3.4%; 12 out of 14 have lost fat: range of variation: -12–4.2%) and hip circumference (-4.2%) after the 3-month training program. In the nontrained overweight subjects, significant increases in weight (2.6%) and %FM (5.5%) were observed after 3 months. In addition, their aerobic fitness decreased. The 3-month training program also had beneficial effects in the trained nonobese subgroup, resulting in a significant decrease in %FM (-7.6%; 7 out of 7 have lost fat: range of variation: -12.02% to -4.04%) and increases in FFM (4.5%) and relative peak power output (14.4%). Conversely, the nontrained nonobese group did not show any anthropometric changes after 3 months, and their aerobic fitness decreased. Intragroup comparisons showed that after the training period, the girls in the trained group exhibited a higher peak power output and VO2peak than the girls in the nontrained group. Improvements in aerobic fitness were associated with concomitant beneficial changes in body composition. Indeed, ΔVO2peak was negatively correlated (p < .05) with the 3-month values for weight (r = –0.64), BMI (r = –0.65), % FM (r = –0.72), and waist circumference (r = –0.74).
Lipid Peroxidation and Inflammation at Rest, Before, and After Training. Table 2 shows the values for lipid
peroxidation and inflammation-related markers at rest at T0 and T3. Baseline values of F2-Isop, ROOH and MPO did not differ between the groups, but baseline ox-LDL was higher in the nontrained overweight girls than in their nonobese counterparts (p < .05). The 3-month training program did not affect any of these parameters, except for ox-LDL in the nonobese group. Indeed, training prevented the increase in ox-LDL observed during the 3-month period in the nontrained nonobese subgroup (p < .05).
70
Lipid Peroxidation and Inflammation-Related Parameters in Response to Maximal Exhaustive Exercise. In the
nonobese groups F2-Isop, ROOH, and MPO were not modified by acute exercise, both at T0 and at T3 (data not shown), but ox-LDL was significantly increased at baseline in the trained subgroup (p < .05) (Figure 1). Three months of training considerably reduced the exercise-induced change in ox-LDL (Δexox-LDL) (p < .05) (Figure 1). In the trained overweight subgroup, exercise induced a significant increase in F2-Isop (Figure 2a), ROOH (Figure 2b), and MPO (Figure 2c) (p < .05) at T0. After the 3-month training program, the exercise-induced increases in the above markers were no longer significant. In addition, the postexercise values of F2-Isop and MPO after training were significantly lower than the pretraining values (p < .05). Furthermore, the 3-month training program significantly decreased the exercise-induced Δexox-LDL (Figure 1). In the nontrained overweight subgroup, the exerciseinduced increase in ROOH and MPO at T0 remained significant after 3 months (p < .05) (data not shown). Correlations Between Training-Induced Changes in Anthropometric Parameters and Training-Induced Changes in OS Markers. In both trained subgroups,
the changes in WHR induced by training were correlated with exercise-induced changes in ROOH (r = .52; p < .05), and the training-induced changes in %FM and hip circumference were correlated with the exercise-induced changes in ox-LDL (r = .56 and 0.51, respectively; p < .05). In addition, in the trained overweight subgroup, the variations in %FM induced by training were also positively correlated with the 3-month postexercise F2-Isop values (r = .76; p < .05), and the modifications in FFM that were induced by training were negatively correlated with the 3-month postexercise ROOH values (r = –0.53; p < .05).
Correlations Between Training-Induced Changes in Aerobic Fitness and Training-Induced Changes in OS Markers. In the trained subgroups, the changes in
peak power output induced by training were negatively correlated with the exercise-induced changes in ROOH (r = – 0.46, p < .05) and with postexercise MPO values (r = –0.48, p < .05). Furthermore, this correlation was higher for the trained overweight group than for the other subgroups (r = –0.62, p < .05).
Discussion This is the first study to examine exercise-training effects (not associated with dietary intervention) on exerciseinduced oxidative stress in overweight adolescent girls. These protective effects are associated with improvements in body composition and with a lack of the deterioration of aerobic fitness level that was observed in the nontrained group. These findings are important because many disorders associated with obesity (e.g., cardiovascular disease and diabetes) can be aggravated by higher OS and inflammation in adolescents. If left untreated,
Table 1 Anthropometric Data and Aerobic Fitness Values at T0 and T3 in Nonobese and Overweight Subjects Nonobese (n = 16) Age (years) BMI
Overweight (n = 23)
Nontrained (n = 9)
Trained (n = 7)
Nontrained (n = 9)
Trained (n = 14)
16.6 ± 0.5
17.1 ± 0.3
16.3 ± 0.5
16.1 ± 0.3
(kg/m2)
T0
21.9 ± 0.5
21.0 ± 0.3
28.3 ± 0.9
28.9 ± 1.0
T3
22.2 ± 0.7
21.4 ± 0.2
28.6 ± 0.8
28.9 ± 0.7
Δ
0.3 ± 0.2
0.4 ± 0.2
0.4 ± 0.2
–0.03 ± 0.2
54.2 ± 1.2
53.8 ± 2.2
74.6 ± 2.6$$$$
74.7 ± 2.8$$$
Weight (kg)
Downloaded by W VA Univ Library on 09/16/16, Volume 27, Article Number 1
T0
76.5 ±
2.3**
T3
55.1 ± 1.6
54.7 ± 2.1
75.5 ± 2.7
Δ
0.9 ± 0.6
0.9 ± 0.1
1.9 ± 0.7
0.8 ± 0.5
157.3 ± 0.02
159.8 ± 0.02
162.4 ± 0.01
160.5 ± 0.01
Height (cm) T0 T3
157.6 ± 0.01
160.0 ± 0.02
Δ
0.3 ± 0.001
0.2 ± 0.001
% fat mass
163.3 ±
0.01**
161.6 ± 0.01***
0.9 ± 0.002$
1.1 ± 0.003$
.
T0
27.2 ± 2.2
25.7 ± 0.9
36.1 ± 1.2$$
36.5 ± 1.8$$$
T3
27.9 ± 2.21
23.8 ± 0.98***
38.0 ± 1.89**
35.2 ± 1.84*
0.7 ± 0.7
–1.9 ± 0.3#
1.9 ± 0.5
–1.3 ± 0.4#
T0
39.6 ± 1.3
39.9 ± 1.5
47.4 ± 0.8$$$$
46.9 ± 1.4$$
T3
39.6 ± 1.1
41.6 ± 1.5**
47.2 ± 0.9
48.4 ± 1.3**
–0.25 ± 0.6
1.5 ± 0.4#
Δ Fat-free mass (kg)
Δ
0.5#
0.0 ± 0.5
1.7 ±
T0
95.8 ± 1.1
95.1 ± 1.8
110.8 ± 1.9$$$$
111.3 ± 1.8$$$$
T3
96.2 ± 0.9
95.1 ± 1.8
110.5 ± 1.7
108.5 ± 2.3*
Δ
0.4 ± 0.9
0.0 ± 0.3
–0.3 ± 1.0
–2.8 ± 1.1
1.88 ± 0.05
1.99 ± 0.12
1.65 ± 0.05$
1.60 ± 0.07$
1.54 ± 0.06
1.55 ± 0.05$$$$
Hip circumference (cm)
Power max (W·kg-1) T0
2.28 ±
0.15**#
T3
1.73 ± 0.12
Δ
–0.15 ± 0.12
0.29 ± 0.06#
–0.11 ± 0.07
–0.05 ± 0.04$
32.4 ± 1.7
34.3 ± 2.1
30.0 ± 0.4$
29.1 ± 1.1$
2.1#
0.7*
28.2 ± 1.1$
–2.4 ± 1.0
–0.9 ± 0.7$
V. O2peak (ml·kg-1·min-1) T0 T3
29.5 ±
Δ
1.7**
–2.9 ± 1.9
35.5 ±
1.2 ± 0.9
27.6 ±
Note. Values are in means ± SEM. BMI = body mass index, WHR = waist to hip ratio, W = watt, V. O2peak = peak rate of oxygen consumption. ∆ = the change in the variable from baseline to 3 months. $Significant
difference from nonobese group at T0.
#Significant
difference from nontrained subgroup.
*Significant
difference from T0 to T3.
$p
