Heredity and Changes in Plasma Lipids and
Lipoproteins after Short-term Exercise Training in Men
Jean-Pierre Despres, Sital Moorjani, Angelo Tremblay, Eric T. Poehlman,
Paul J. Lupien, Andre Nadeau, and Claude Bouchard
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
The alms of this controlled experiment were to Investigate the effects of short-term
aerobic exercise training on plasma llpld and llpoproteln concentrations and the role
of heredity In determining the Individual variation observed in the llpoprotein-llpid
response. Six pairs of male monozygotic (MZ) twins were subjected to an exercise
training program that Induced a 22 000 kcal energy deficit after 22 consecutive days
of training. This program significantly reduced body weight, percent body fat, and
subcutaneous fat and significantly increased maximal oxygen consumption (V0 2 max)
(p<0.005). The plasma Insulin response to an oral glucose challenge was markedly
reduced after training (p<0.001). Plasma triglycerlde concentration decreased and
the high density llpoproteln cholesterol (HDL-CHOL)/CHOL ratio Increased with
training (p<0.05). Subjects also displayed substantial Individual variation In their
response to exercise training, but the changes In plasma CHOL, apollpoproteln
(apo) B low density llpoproteln cholesterol (LDL-CHOL), HDL-CHOL, and the HDLCHOL/CHOL ratio tended to be similar within MZ twin pairs (0.67§/i§0.92;
0.05>p<0.0001) thus Indicating a significant effect of heredity on the sensitivity of
plasma llplds and lipoproteins to exercise training. In addition, Individual changes In
plasma llplds and lipoproteins were correlated with neither changes In percent body
fat nor changes In VOzmax, whereas changes In plasma CHOL, apo B, and LDL-CHOL
were significantly correlated with changes In the proportion of trunk fat and with the
Insullnogenlc Index (0.63§/^0.74; 0.05>p<0.01). Control for Initial level of body fat
distribution or its alteration during exercise training eliminated the wrthln-palr
resemblance In CHOL, apo B, and LDL-CHOL responses. These findings Indicate that
some of the Individual variation In the response of plasma lipoproteins to exercise
training Is Influenced by heredity and Is partly mediated by alterations In subcutaneous adipose tissue distribution and associated changes in Insulin metabolism.
(Arteriosclerosis 8:402-409, July/August 1988)
N
umerous studies have reported associations between
plasma llpld and lipoprotein concentrations, plasma
lipoprotein-lipid composition, plasma apoprotein levels,
and coronary heart disease. 1 - 4 Since the altered metabolism of plasma lipoproteins is an important factor in the
development of cardiovascular disease (CVD), the study
of genetic and environmental factors affecting the concentration of plasma lipoproteins is, therefore, of importance.
In that respect, various effects on lipoprotein metabolism
of nutrition 5 and lifestyle factors such as cigarette
smoking, 67 alcohol consumption, 78 and exercise9 have
been documented. Accordingly, the role of heredity in
determining plasma lipoprotein variation has been studied, and significant genetic effects have been reported,
ranging from monogenic defects (i.e., familial
hypercholesterolemia)10 to polygenlc variations.11-15 An
issue of equal importance, which has received lesser
attention, however, is the genotype-environment
interaction;11 that is, the role of heredity in determining the
sensitivity of plasma lipoproteins to respond to a given
treatment. We recently studied one aspect of this interaction by overfeeding monozygotic (MZ) twins for a period of
22 days.16 Substantial individual variation in the concentration of plasma lipids and lipoproteins was observed in
this experiment, but subjects of the same twin pair displayed comparable changes in serum triglycerides (TG),
high density lipoprotein cholesterol (HDL-CHOL), and
HDL-CHOL/total CHOL ratio.18 The results suggest that
heredity plays an important role in determining individual
variations in plasma lipoproteins, as well as in determining
the sensitivity to develop an atherogenic profile in response
to a caloric excess.
From the Physical Activity Sciences Laboratory, the Upld
Research Center, and the Department of Medicine, Laval University, Quebec, Canada.
This research was supported by NSERC of Canada, FCACQuebec, and Fitness Canada. Dr. Despres Is the recipient of a
"Fonds de la recherche en sante du Quebec" scholarship.
Address for reprints: Jean-Pierre Despres, Ph.D., Physical
Activity Sciences Laboratory, PEPS, Laval University, Ste-Foy,
Quebec, Canada G1K 7P4.
Received July 29,1987; revision accepted February 26,1988.
We do not know, however, whether an individual's
response to measures that are aimed at reducing the risk
of CVD, such as an improvement of the lipoprotein profile,
will also be associated with genetic characteristics. Exercise training is recognized as having significant effects on
plasma lipoproteins,8 a factor that could help explain the
increased CVD risk often seen in sedentary subjects.9
However, the question of whether the response of plasma
lipids and lipoproteins to exercise training is associated
402
EXERCISE TRAINING AND PLASMA LIPOPROTEINS IN TWINS
with heredity has, to our knowledge, not been studied. To
investigate this question, six pairs of MZ twins were
subjected to a controlled experiment during which an
aerobic exercise training program of 22 days was used to
generate a negative energy balance of 22 000 kcal. The
plasma lipoprotein-lipid profile, along with other potentially
associated metabolic indices, was measured.
Methods
Subjects
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
Six pairs of young adult male MZ twins gave their
written consent to participate in this study, which received
the approval of the Laval University Medical Ethics Committee. The participants were subjected to a complete
physical examination and were required to be healthy but
sedentary. No subject had a history of recent illness,
diabetes, hyperiipidemla, or other endocrine disorders.
The zygosity was established from a questionnaire and on
the basis of genetic concordance for 11 red blood cell
antigens and enzymes (15 loci) and for the A, B, and C
loci of the HLA system. Twins that were discordant for one
of these loci were excluded from the study.
Experimental Protocol
A few weeks before the beginning of the experiment,
the subjects were Instructed to complete a dietary record
during three days that included a weekend day as previously described.17 The reliability of the assessment of
mean energy intake by this method has already been
reported.17 During the experiment, the subjects were
housed in an experimental station located in a park about
80 km from the laboratory. The men were instructed to
refrain from exercise and their energy intake was carefully
monitored by nutritionists during a 4-day baseline period.
The daily caloric Intake during the experimental protocol
was determined from the mean energy intake obtained
from the 3-day dietary journal and the 4-day observation
period and corresponded to 14.4 ±0.8 MJ (3419 ±200
kcal) per day. This value represents the mean daily
energy intake during the entire exercise training program.
Once the assessment of basal energy needs was completed, the men were trained on bicycle ergometers for 22
consecutive days at 58% of their maximal oxygen uptake
(V02max). The exercise duration was 116 min/day and
was calculated to induce a daily energy deficit of 4.2 MJ
(1000 kcal). Two exercise sessions were performed each
day to produce this energy deficit. During each exercise
session, heart rate was recorded at every 3 minutes. All
meals during the training program were prepared and
measured by nutritionists and no additional food was
available other than that distributed by nutritionists. The
mean contribution of carbohydrate, fat, and protein intake
corresponded to 45%, 38%, and 17% of the daily energy
intake, respectively.
The effect of heredity in determining the sensitivity of
various metabolic variables to exercise training and
genotype-training interaction effects have been reported
for carbohydrates18 and adipose tissue metabolisms,19 as
well as for resting metabolic rate and dietary-induced
thermogenesis.20 Thus, an extensive description of the
DesprSs et al.
403
experimental protocol has already been reported.20 The
present study focuses on plasma lipid and lipoprotein
changes and on the effect of heredity in determining the
magnitude of their responses to exercise training.
Laboratory Measurements
The following measurements were made at the beginning and at the end of the experiment. VO2max was
assessed by a progressive test to exhaustion on a bicycle
ergometer. The initial work load was 50 watts and was
increased by 25 watts until exhaustion. VO2 was recorded
using an open gas circuit system and V0;>max was
considered as the highest VO2 recorded during the test for
1 minute.21
Subcutaneous sklnfolds were measured on the left
side of the body using the Harpenden skinfold caliper.
Biceps, triceps, subscapular, calf, abdominal, and suprailiac skinfold thicknesses were measured by using the
procedures of the International Biological Program. 22
The sum of these six skinfolds was used as an indicator
of subcutaneous fat, 23 and the ratio of trunk skinfolds
(subscapular, abdomen, supralliac) divided by extremity
skinfolds (biceps, triceps, calf) was used as an indicator
of subcutaneous adipose tissue distribution.24-25 Body
density was determined by underwater weighing 28 and
the percent body fat was calculated with the equation of
Slri.27 The residual lung volume was assessed by the
method of Wllmore et al. 28
Blood samples were obtained in the morning after a
12-hour fast for the measurement of plasma lipids and
lipoproteins. Plasma TG and CHOL,28 as well as HDLCHOL,30 were measured, whereas low density lipoprotein
cholesterol (LDL-CHOL) was estimated by using the
equation of Friedewald et al. 31 Plasma apolipoprotein B
(apo B) was measured by the electroimmunoassay of
Laurell32 as previously described. 1933 An oral glucose
tolerance test (OGTT) was also performed after an overnight fast, and plasma glucose and insulin concentrations
were measured as previously described.18 Glucose and
insulin areas were calculated by the trapezoid method.
The effects of the present exercise training program on
plasma glucose and insulin levels during the OGTT have
already been reported.18 In the present study, these
variables are used only as independent variables.
Statistical Analysis
The effect of exercise training was analyzed by using a
two-way analysis of variance for repeated measures on
one factor as previously described.20 The intraclass correlation coefficient for relative changes, which measures
the intrapair resemblance in the response to exercise
training, was calculated from the between- and the withintwin pairs variances as outlined by Haggard.34 Pearson
product-moment coefficients were also computed to estimate the associations between the variables. As subjects
of the same twin pair were not independent from each
other, these correlations should be considered as trends,
rather than as accurate measurements of the relations
between the variables.
ARTERIOSCLEROSIS
404
VOL 8, No 4, JULY/AUGUST 1988
Table 1. Effects of 22 Days of Exercise Training on Body Fatness, Maximal Aerobic Power, and
Carbohydrate Metabolism In Six Pairs of Male MZ Twins
Variable
Before
Weight (kg)
% body fat (%)
Sum of six skJnfolds (mm)
T/E skinfolds ratio
VOynax (l/mln)
V0;>max (ml/kg/min)
Glucose area
Insulin area
Insulin/glucose areas
77.3 ±7.0
18.3 ±3.2
67.7 ±12.7
1.58 ±0.29
3.53±0.13
48.4 ±2.8
16410±664
13569±1397
0.84 ±0.09
After
74.9 ± 7.1 f
15.9 ±3.5*
60.5+12.6*
1.61 ±0.24™
3.78±0.11*
53.1 ± 3 . 1 *
17073±493 n>
7556±1336t
0.45 ±0.09*
% change (range)
- 3 % ( - 6 % to - 1 % )
-18%(-49%tO -1%)
- 1 3 % ( - 2 1 % to - 2 % )
+ 3% ( - 4 % to +16%)
+ 7%(0%to +14%)
+ 1 0 % ( - 1 % t O +17%)
+ 5% ( - 1 7 % to +27%)
- 4 6 % ( - 6 2 % t o +2%)
- 4 9 % ( - 6 3 % to - 1 4 % )
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
The values are means ±SEM.
The effect of training was assessed by a two-way analysis of variance for repeated measures on one factor
(exercise training) wtth twins nested in pairs. T/E skJnfolds ratio=trunk skinfolds (subscapular, abdomen, and
suprailiac)/extremity skinfolds (triceps, biceps, and calf) ratio; Sum of six skinfolds = sum of trunk and extremity
skinfolds; VOjmax = maximal oxygen consumption; Glucose area = sum of plasma glucose concentrations (mg/dl/180
min) during the oral glucose tolerance test; Insulin area = sum of plasma insulin concentrations (/iU/ml/180 min)
during the oral glucose tolerance test; Insulin/glucose areas = ratio of the insulin area over the glucose area.
The exercise training produced a daily caloric deficit of 1000 kcal.
*p<0.005, tP<0.001, +-P<0.0001, ns = not significant.
Table 2. Effects of 22 Days of Exercise Training on Plasma LIplds and Upoprotelns In Six Pairs
of Male MZ Twins
Before
Variable
Plasma CHOL
Plasma TG
Plasma apo B
Plasma LDL-CHOL
Plasma HDL-CHOL
Plasma HDL-CHOL/CHOL
After
% change (range)
154.8 ±7.2
104.5±19.9
63.0 ±2.8
136.9±8.3™
72.0 ±15.5+
55.1 ±3.3™
103.0±4.6
85.2 ±6.5™
37.3 ±3.0™
0.28 ±0.02*
- 1 1 % ( - 2 7 % t o +21%)
-32%(-60%to -2%)
- 1 2 % ( - 2 7 % to +18%)
- 1 6 % ( - 4 5 % to +22%)
+ 1 6 % ( - 3 5 % t O +50%)
+ 32% ( - 2 2 % to +80%)
32.9 ±2.6
0.22 ±0.02
Values are given as mg/dl and are the means ± SEM.
The effect of training was assessed by a two-way analysis of variance for repeated measures on one factor
(exercise training) with twins nested in pairs. CHOL = cholesterol, TG = trtglyceride, apo = apoprotein, LDL = tow
density llpoproteln, HDL = high density llpoprotein.
The exercise training produced a daily caloric deficit of 1000 kcal.
*p<0.05, tP<0.01, ns = not significant.
Results
The 22-day exercise training program significantly
increased V0 2 max expressed in absolute values (l/min) or
as per kilogram of body weight (Table 1). Body fatness
was also significantly reduced as indicated by the significant decreases in body weight, percent of body fat, and in
the sum of skinfolds. The proportion of subcutaneous fat
in the trunk did not change significantly, as reflected by
the lack of change in the sum of trunk/sum of extremity
skinfolds ratio. There was no effect of exercise training on
the plasma glucose area during the glucose tolerance
test, but a substantial reduction (p<0.001) occurred in the
plasma insulin area after training. The reduction in the
ratio of the insulin over glucose areas (insulinogenic
index) was also highly significant (p<0.0001).
Of interest is the substantial individual variation in the
response of these variables to training. Indeed, changes
in percent of body fat ranged from a decrease of near 50%
to a decrease of only 1%, and the increase in VOjmax
ranged from 0% to 14%. Thus, subjects displayed various
levels of sensitivity when exposed to the same exercise
training prescription. A variation in response for the
changes in Insulin area after oral glucose ingestion (from
- 6 2 % to +2%) and in the change in the insulin/glucose
areas (from - 6 3 % to - 1 4 % ) was also noted.
Table 2 presents the effects of the training program on
the concentration of plasma lipids and lipoproteins. Plasma
TG decreased significantly in response to the exercise
program, whereas the trends for decreases in plasma
CHOL and LDL-CHOL (decreases of 1 1 % and 16%
respectively) and for an increase in plasma HDL-CHOL
(increase of 16%) were not significant. However, these
trends contributed to a significant increase in the HDLCHOL/CHOL ratio (increase of 32%, p<0.05). Plasma
apo B concentration did not change significantly.
Although only changes in plasma TG and in the HDLCHOL/CHOL ratio were significant, the men again displayed individual variations in their responses to the
training program. Thus, changes in plasma CHOL ranged
from a decrease of 27% to an increase of 21 %, and the
response of plasma TG ranged from a decrease of 60% to
a decrease of only 2%. A comparable variation in changes
EXERCISE TRAINING AND PLASMA LIPOPROTEINS IN TWINS
Despr6s et al.
405
Table 3. F Ratios for wrthIn-Twin Pair Resemblance for Effect of Exercise Training, and
wlthln-Palr Similarity In Response to Exercise Training
Variable
Resemblance
before
exercise
training
Exercise
training
effect
Genotypeexercise
training
interaction*
Intraclass
coefficient
for relative
changes
3.0™
23.2*
4.4™
23.811
0.61"
16.0§
0.921|
-0.28m
18.3§
10.9*
Plasma CHOL
Plasma TG
0.90§
Plasma apo B
Plasma LDL-CHOL
Plasma HDL-CHOL
0.87§
0.48™
3.7™
1.7™
HDL-CHOL/CHOL
0.65f
6.3t
0.67t
0.89§
0.88§
0.90§
5.1f
0.67f
0.83*
T h i s F ratio Is for wtthin-MZ pair similarity in response to overfeeding.
CHOL = cholesterol, TG = triglyceride, apo = apoprotein, LDL = low density llpoprotein, HDL = high density llpoprotein.
tpS0.05, * p £ 0 . 0 1 , §pS0.005, ||p50.0001, ns = not significant.
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
Table 4. Correlation Coefficients between Relative Changes In VCVnax, Percent Body Fat,
Subcutaneous Fat Distribution, and Relative Changes In Carbohydrate Metabolism, as well as
Relative Changes In Plasma LJplds and LIpoprotelns after Exercise Training
Relative changes in
Relative changes in
VCynax
% Body fat
Plasma CHOL
Plasma TG
Plasma apo B
Plasma LDL-CHOL
Plasma HDL CHOL
HDL-CHOL/CHOL
-0.24
0.15
-0.27
Plasma glucose area
Plasma Insulin area
Insulin/glucose areas
0.00
-0.09
-0.12
-0.05
-0.15
-0.18
-0.04
0.17
0.28
0.33
0.14
-0.24
-0.17
-0.20
-0.08
Trunk fat
Extremity fat
0.70f
-0.04
0.79*
-0.19
0.39
0.20
0.12
0.33
0.35
-0.15
-0.29
-0.37
-0.16
0.53
0.06
0.78*
0.18
-0.39
T/E ratio
0.74f
-0.19
0.72t
0.68*
0.40
-0.22
0.15
0.73f
0.66*
CHOL = cholesterol, TG = trigfyceride, apo = apoprotein, LDL = low density lipoprotein, HDL = high density lipoprotein, Glucose area = sum of plasma glucose concentrations (mg/dl/180 mln) during the oral glucose tolerance test,
Insulin area = sum of plasma insulin concentrations (/xU/ml/180 mln) during the oral glucose tolerance test,
Insulin/glucose areas => ratio of the insulin area over the glucose area in plasma during the oral glucose tolerance test,
VCVnax = maximal oxygen consumption, Trunk fat = sum of subscapular, abdominal, and suprailiac skinfolds;
Extremity fat = sum of biceps, triceps, and calf skinfolds; T/E ratio = trunk/extremity skinfolds ratio.
*p<0.05, fp<0.01, *p<0.005, otherwise coefficients are not significant.
in plasma apo B ( - 2 7 % to +18%), HDL-CHOL ( - 3 5 %
to +50%), and in the HDL-CHOL7CHOL ratio ( - 2 2 % to
+ 80%) was observed.
Table 3 suggests that this variation in response to
exercise training was associated with the genotypic characteristics of the subjects. Before training, there was a
significant within-palr resemblance for all plasma variables (0.65§/j§0.90, 0.05>p<0.005) except plasma HDLCHOL. There was also a significant intrapair resemblance
in the relative changes in plasma CHOL, apo B, LDLCHOL, HDL-CHOL, and the HDL-CHOL/CHOL ratio
(0.67§/jS0.92, 0.05>p<0.0001), Indicating a high level
of similarity within MZ twin pairs in the magnitude of
changes in plasma lipoprotein profile in response to
training.
Table 4 shows that changes in plasma lipids and lipoproteins were not correlated with changes in the percent of
body fat nor with changes in vXVnax. However, when the
relative loss of trunk fat was considered, significant correlations were observed between the individual changes in
the relative proportion of trunk fat and changes in plasma
CHOL, LDL-CHOL, and apo B. We previously reported that
this training protocol significantly reduced the insulin
response to an oral glucose challenge.18 In the present
study, the indicators of carbohydrate metabolism were
analyzed as independent variables in an attempt to further
understand the mechanisms by which alterations in body
fat distribution are associated with changes In plasma
lipoproteins. Table 4 indicates that changes in the plasma
glucose and insulin areas during the glucose tolerance test,
as well as the changes in the ratio of insulin/glucose areas,
were not correlated with the magnitude of changes in
percent body fat nor with the changes in VOzmax. Changes
in the trunk/extremity skinfolds ratio were, however, significantly correlated with alterations in both the insulin area
(r=0.73, p<0.01) and the ratio of insulin/glucose areas
(r= 0.66, p<0.05). These results indicate that subjects who
lost more trunk fat compared to other individuals were
those who also showed the greatest improvements In their
carbohydrate metabolism.
Figure 1 shows that changes in the ratio of insulin/glucose areas after training were correlated with changes in
406
ARTERIOSCLEROSIS
2
VOL 8, No 4, JULY/AUGUST 1988
to
-10r-
2 -io
a.
r=0.63'
Uj .
O
o
u
3 -30 h
o
•j -40
o
3 -30
o
-
| -40
to
\ -50
-50
z
at
S -60
z
Vi -60
I "7
r = 0.06n
M
-30
-20
X .70
_J_
_J_
_1_
-10
0
+10
+20
-40
-60
+30
-20
% CHANGES IN PLASMA TO
% CHANGES IN PLASMA CHOL
A
B
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
Figure 1. Correlations between relative changes in the plasma insulin/glucose areas during an oral
glucose tolerance test and (A) relative changes in plasma cholesterol (CMOL) or (B) relative changes
in plasma triglycertde (TG) after 22 days of exercise training, which induced a daily caloric deficit of
1000 kcal. *p<0.05, ns = not significant.
plasma CHOL (Figure 1 A) but not with changes in plasma
TG (Figure 1B). Figure 2 Indicates that changes in the
insulin/glucose areas were correlated with changes in
plasma apo B (Figure 2A) and in plasma LDL-CHOL
(Figure 2B), suggesting a synchronized response to exercise training for body fat distribution, carbohydrate metabolism, and some plasma llpoproteins.
Because changes in plasma lipoproteins were correlated with changes in body fat distribution (Table 4) and
with the initial level of body fat distribution ( r = - 0 . 8 6 ,
- 0 . 8 2 , and - 0 . 8 8 between initial level of trunk/extremity
skinfolds ratio and changes in CHOL, apo B, and LDLCHOL, respectively, p<0.01), the within-pair similarity in
lipoprotein response was studied after control for relative
change in body fat distribution or for the initial level of
adipose tissue distribution (Table 5). After these controls,
the within-pair similarity in CHOL, apo B, and LDL-CHOL
responses was no longer observed, suggesting that the
genetic effect on the response of these plasma variables
was mediated by a similar initial level or by response in
body fat distribution.
The within-pair resemblance in HDL-CHOL/CHOL
response to training remained, however, statistically significant, suggesting that this genotype-training interaction
effect is not mediated by the within-pair resemblance in
body fat distribution.
Discussion
The results of this study indicate that biological inheritance is an important determinant of the heterogeneous
response of plasma lipkjs and lipoproteins to an exercise
training program that generates a chronic negative energy
balance for 22 days. Indeed, whereas only plasma TG
concentration and the HDL-CHOL/CHOL ratio were significantly modified by this short-term, exercise training program, the subjects showed substantial Individual variations
in plasma lipid and lipoprotein responses to the program.
Thus, with the exception of serum TG, changes in CHOL,
LDL-CHOL, HDL-CHOL, HDL-CHOL/CHOL, and apo B
tended to be similar within twin pairs (0.05>p<0.0001).
The responses of plasma CHOL, apo B, and LDL-CHOL
-10
0.64*
-20
m
O
r = 0.74f
20
o
-30
3 -JO
o
-40
=! - 40
at
-50
-60
-70
-30
-20
-10
0
+10
+20
o
z
<
X
o
*
-SO
% CHANGES IN PLASMA APO B
-40
-M
-20
-10
+10 '20
% CHANGES IN PLASMA LDL-CHOL
B
Figure 2. Correlations between relative changes in the plasma insulin/glucose areas during an oral
glucose tolerance test and (A) relative changes in plasma apollpoprotein (apo) B or (B) relative
changes In plasma low density Upoprotein cholesterol (LDL-CHOL) levels after 22 days of exercise
training, which Induced a daily caloric deficit of 1000 kcal. *p<0.05; tP<0.01.
EXERCISE TRAINING AND PLASMA LIPOPROTEINS IN TWINS
Table 5. Intraclass Correlation Coefficients and F
Ratios for wtthln-Palr Similarity in Response to Exercise Training after Control for Relative Changes or
Initial Level of Trunk/Extremity Sklnfolds Ratio
Controlled for
relative change in
T/E ratio
Controlled for initial
level of T/E ratio
Intraclass
coefficient
F ratio
Intraclass
coefficient
CHOL
0.13
1.29
TG
ApoB
-0.40
0.41
0.32
0.43
2.37
1.94
Variable
LDL-CHOL
HDL-CHOL
HDL-CHOL/CHOL
0.28
0.87*
1.78
14.42*
0.19
-0.27
-0.07
-0.23
0.72f
0.59*
F ratio
1.47
0.58
0.86
0.63
6.02t
3.88*
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
CHOL = cholesterol, TG = triglyceride, Apo = apoprotein,
LDL = low density lipoprotein, HDL = high density lipoprotein.
*pS0.07;
were, however, correlated with the initial level and with the
change in body fat distribution. After statistical control for
body fat distribution (initial level or change), the within-pair
similarity in CHOL, apo B, and LDL-CHOL responses was
no longer observed. These results suggest that the
genotype-training interaction effect noted for the response
of CHOL, apo B, and LDL-CHOL is mediated by the effect
of heredity on body fat distribution. The response of the
HDL-CHOL/CHOL ratio to exercise training, which was
also influenced by heredity, was not due to the twins'
similarity in initial levels nor to changes in body fat distribution. However, with only six different genotypes studied,
these statistical controls for resemblance in body fat distribution should be interpreted with caution.
Numerous exercise training studies have reported
changes in plasma lipids and lipoproteins, 93536 and
these changes in the lipoprotein-lipid profile are considered to be favorable in reducing the potential risk of
CVD. 9 In the present study, along with a decrease in
serum triglyceride levels, we also noted a significant
increase in the proportion of plasma CHOL carried in the
HDL fraction (HDL-CHOL7CHOL). Although the atherogenicity of high serum TG concentrations is still a
subject of controversy,37 its reduction after training
indicates a decreased secretion of TG-rich particles by
the liver38 combined with a higher capacity of the
endothelium to catabolize TG in circulating lipoproteins.39
The increase in the HDL-CHOLVCHOL ratio is undoubtedly a beneficial effect of the exercise program, because
this ratio is considered a foremost predictor of cardiovascular disease events.2
Many studies have reported reduction in plasma
CHOL and an increase in HDL-CHOL after exercise
training, although not all studies are concordant. 93536 In
the present experiment, we also observed similar trends
for these two variables, but they did not attain statistical
significance. The magnitude of decreases in the concentrations of plasma CHOL (11%), apo B (12%), and
LDL-CHOL (16%) and the concomitant increase in HDLCHOL concentration (16%) is comparable to the exercise training effects generally reported. 933 - 38 The non-
Despr6s et al.
407
significance of these changes could be attributed partly
to the small number of subjects. In addition, although
severe (2 hours of aerobic exercise per day), the exercise program was of short duration. In addition, the initial
condition of the men in this study (young, nonobese,
with normal lipoprotein profiles) was undoubtedly a
factor in minimizing the effect of exercise training on
plasma lipoproteins. The initial levels of lipoproteins and
lipids have already been reported to be an important
variable in the determination of the magnitude of the
lipoprotein responses to an exercise training program. 36
It is logical to speculate that such a training program in
subjects with abnormal lipoprotein profiles would have
produced more substantial changes. Despite the controlled environment (restricted caloric intake and exercise training), an individual variation in the response to
training was clearly evident. However, further work is
required to identify the mechanisms responsible for the
presence of low and high responders to intervention
measures aimed at reduction of the CVD risk.
it is noteworthy that individual changes in plasma
lipoproteins and lipids were dissociated from changes in
percent body fat and V0 2 max. A reduction in the proportion of subcutaneous trunk fat, however, appeared to be
associated with the response of postglucose plasma
insulin and of plasma lipoproteins to exercise training.
Indeed, the results indicate that subjects who responded
with the greatest reduction in their proportion of trunk fat
as reflected by the trunk/extremity skinfolds ratio were the
ones who also showed the greatest reductions in plasma
CHOL, LDL-CHOL, and apo B levels and in whom there
were the largest reductions in postglucose plasma insulin
and insulin/glucose areas. These results suggest a dynamic
association between fat distribution and plasma lipoproteins, as well as between changes in body fat topography
and changes in the carbohydrate metabolism. Associations between body fat distribution and indicators of
carbohydrate metabolism have been reported, 40 - 47 and
recent data have also demonstrated that body fat topography is associated with plasma lipoprotein and lipid
concentrations. 2 3 4 1 * 3 - 4 7 Therefore, the association
between body fat distribution and metabolic complications 2340 - 47 could help to explain the relationship between
adipose tissue localization and CVD events in recent
prospective studies. 484950
To our knowledge, the relationship between body fat
localization and CVD risk factors has not been studied in
a controlled study where weight loss occurs. Thus, we
wondered whether the subjects showing the greatest loss
of trunk fat were also those showing the most substantial
changes in insulin metabolism and in lipoprotein-lipid
profile. Our results show that this was, indeed, the case,
as indicated by significant correlations between changes
in the trunk/extremity skinfolds ratio and changes in the
insulinogenic index (postglucose plasma insulin/glucose
areas). Furthermore, decreases in the insulinogenic index
were associated with reductions in plasma CHOL, LDLCHOL, and apo B concentrations. Ail these changes were
also associated with alterations in fat distribution but were
dissociated from changes in total body fat. The importance of body fat distribution in the response of CHOL,
408
ARTERIOSCLEROSIS
V O L 8, No 4, JULY/AUGUST 1988
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
LDL-CHOL, and apo B to exercise training was such that,
after control for initial level or change in body fat distribution, no within-pair similarity was observed in the response
of these plasma variables. Because the insulinogenic
index was closely associated with changes in CHOL,
LDL-CHOL, apo B, and in body fat distribution, these
results suggest that the genetically determined response
of CHOL, LDL-CHOL and apo B was due to the within-pair
resemblance in body fat distribution and was mediated by
changes in plasma Insulin levels. The mechanism relating
body fat distribution to insulin metabolism has been recently
reviewed. 4 7 5 1 5 2 The mechanism by which body fat localization is correlated with plasma lipoprotein lipids remains
to be established. However, the present data suggest that
some of the effects of fat distribution on plasma lipoproteins are mediated by changes in the metabolism of
insulin. The relationship between TG-rich lipoproteins and
plasma Insulin levels is well documented, 1 8 4 5 4 7 6 2 but the
effects on CHOL-rich lipoproteins are less well understood. The significant correlation observed between
changes in the insulin/glucose areas and LDL-CHOL
suggests an augmented uptake of LDL particles by the
apolipoprotein B,E receptor10 because it has been shown
that insulin can modulate the receptor uptake and catabolism of LDL.53 Therefore, It is possible that in situations of
improved insulin sensitivity such as in exercise trained
subjects, the LDL receptor activity Is enhanced. The
similar correlation between insulin/glucose areas and
plasma apo B levels also supports the direct removal of
LDL particles, but does not rule out a similar removal
mechanism for the particles of intermediary density that
are produced during increased TG hydrolysis. These
changes could also be attributed to exercise trainingassociated variations in plasma insulin levels. Further
work is required to substantiate this hypothesis.
Acknowledgments
We would like to express our gratitude to Fernand Landry,
Marcel Boulay, Pierre Lagasse, and Germain Theriault for their
collaboration. We also thank dieticians France Bollard, MarieFrance Breton, and Denise Mitchell for the meal planning and
compilation of the data on energy intake. The expert technical
assistance of Guy Foumfer, Henri Bessette, Johanne Bergeron,
Lucle Turcotte, Pierre Laliberte, Marie Martin, and Rachel
Duchesne is also acknowledged. We express our gratitude to
Claude Leblanc for the statistical analyses.
References
1. Miller GJ, Miller NE. Plasma hlgh-denslty-lipoprotein concentration and development of ischaemic heart-disease.
Lancet 1975;1:16-19
2. Castelll WP. Epidemiology of coronary heart disease: The
Framingham Study. Am J Med 1984;76:4-12
3. Upld Research Clinics Program. The Upid Research
Clinics Coronary Primary Prevention Trial Results. II: The
relationship of reduction in incidence of coronary heart
disease to cholesterol lowering. JAMA 1984;251:365-374
4. Thompson Q. Apoproteins: determinants of lipoprotein metabolism and indices of coronary risk. Br Heart J 1984;
51:585-588
5. Grundy SM, Bllhelmer D, Blackburn H, et al. Rationale of
the diet-heart statement of the American Heart Association.
Report of the AHA Nutrition Committee. Arteriosclerosis
1982:2:177-191
6. Kannel WB. Update on the role of cigarette smoking in
coronary heart disease. Am Heart J 1981 ;101:319-328
7. Haffner SM, Applebaum-Bowden D, Wahl PW, et al.
Eptdemlological correlates of high density lipoprotein subfractions, apolipoproteins A-l, A-ll, and D, and lecithin cholesterol acyttransferase. Effects of smoking, alcohol, and
adiposity. Arteriosclerosis 1985:5:169-177
8. Castelll WP, Doyle JT, Gordon T, et al. Alcohol and blood
lipids. The Cooperative Lipoprotein Phenotyping Study. Lancet 19772:153-155
9. Haskell Wl_ Exercise-Induced changes in plasma lipids and
llpoproteins. Prev Med 1984;1323-36
10. Brown MS, Goldstein J L A receptor-mediated pathway for
cholesterol homeostatis. Science 1986:23234-47
11. Motulsky AG. The genetic hyperllpidemias. N Engl J Med
1976:294:823-827
12. Namboodlrl KK, Green PP, Kaplan EB, et al. Family
aggregation of high density lipoprotein cholesterol. Collaborative Upid Research Clinics Program Family Study. Arteriosclerosis 1983:3:616-628
13. Isellus L, Carlson LA, Morton NE, Efendlc S, Undsten J,
Luft R. Genetic and environmental determinants for lipoprotein concentrations In blood. Acta Med Scand 1985;
217:161-170
14. Amos Ct, Wilson AF, Rosenbaum PA, et al. An approach
to the murtivariate analysis of hlgh-denstty-lipoproteln cholesterol in a large kindred: The Bogalusa Heart Study. Genet
Epidemiol 1986:3:255-267
15. Austin MA, King MC, Bawd RD, Hulley SB, Friedman
GD. Risk factors for coronary heart disease in adult female
twins. Genetic herrtability and shared environmental influences. Am J Epidemiol 1987;125:308-318
16. Despres JP, Poehlman ET, Tremblay A, et al. Genotypeinfluenced changes in serum HDL cholesterol after shortterm overfeeding in man: Association with plasma insulin
and triglyceride levels. Metabolism 1987;36:363-368
17. Tremblay A, Sevtgny J, Leblanc C, Bouchard C. The
reproduclbilrty of a three-day dietary record. Nutr Res 1983;
3:819-830
18. Tremblay A, Poehlman E, Nadeau A, Perusse L, Bouchard C. Is the response of plasma glucose and insulin to
short-term exercise-training genetically determined? Horm
Metab Res 1987:19:65-67
19. Poehlman ET, Tremblay A, Marcotte M, Perusse L, TherlauH G, Bouchard C. Heredity and changes in body composition and adipose tissue metabolism after short-term
exercise-training. Eur J Appl Physiol 1987;56:398-402
20. Poehlman ET, Tremblay A, Nadeau A, DussauK J, Thertautt G, Bourchard C. Heredity and changes in hormones
and metabolic rates wtth short-term training. Am J Physiol
(Endocrinol Metab) 1986;250:E711-717
21. Prud'homme D, Bouchard C, Leblanc C, Landry F,
Fontaine E. Sensitivity of maximal aerobic power to training is genotype-dependent. Med Scl Sports Exerc 1984;
16:489-493
22. Welner US, Lourie JA. Human biology, a guide to field
methods. Oxford: Blakwell Scientific, 1969:1-76
23. Despres JP, AJIard C, Tremblay A, Talbot J, Bouchard C.
Evidence for a regional component of body fatness in the
association with serum lipids in men and women. Metabolism 1985;34:967-973
24. Bouchard C. Inheritance of fat distribution and adipose
tissue metabolism. In: Vague J, Bjomtorp P, Guy-Grand B,
eds. Metabolic complications of human obesities. Amsterdam: Elsevier Science Publishers, 1985:87-96
25. Despres JP, Bouchard C, Tremblay A, Savard R, Marcotte M. Effects of aerobic training on fat distribution in male
subjects. Med S d Sports Exerc 1985;17:113-118
26. Behnke AR, Wilmore JH. Evaluation and regulation of body
build and composition. Englewood Cliffs, NJ: Prenctlce-Hall,
1974:20-37
27. Slrl WE. The gross composition of the body. Adv Biol Med
Phys 1956:4:239-280
28. Wilmore JH, Vodak PA, Parr RB, Glrandola RN, Billing
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
Downloaded from http://atvb.ahajournals.org/ by guest on June 17, 2017
Heredity and changes in plasma lipids and lipoproteins after short-term exercise training in
men.
J P Després, S Moorjani, A Tremblay, E T Poehlman, P J Lupien, A Nadeau and C Bouchard
Arterioscler Thromb Vasc Biol. 1988;8:402-409
doi: 10.1161/01.ATV.8.4.402
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272 Greenville
Avenue, Dallas, TX 75231
Copyright © 1988 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/8/4/402
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for which permission
is being requested is located, click Request Permissions in the middle column of the Web page under Services.
Further information about this process is available in the Permissions and Rights Question and Answerdocument.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/