1359
Exercise Training Decreases Plasma Cholesteryl
Ester Transfer Protein
Richard L. Seip, Phillipe Moulin, Thomas Cocke, Alan Tall, Wendy M. Kohrt, Keith Mankowitz,
Clay F. Semenkovich, Richard Ostlund, Gustav Schonfeld
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
To assess the effect of exercise on the plasma concentration of cholesterol ester transfer protein (CETP)
and its possible influence in mediating the exercise-associated redistribution of cholesterol among plasma
lipoproteins, we measured plasma CETP in 57 healthy normolipidemic men and women before and after
9 to 12 months of exercise training. The training protocol resulted in significant changes in Vo^.,,
(mean±SD, +5-3±3.5 mL • kg"1 • min"1), body weight (-2.5±3.5 kg), plasma triglycerides (-25.7±36J
mg/dL), high-density lipoprotein cholesterol (HDL-C) (+2.6±62 mg/dL), and ratios of total cholesterol
to HDL-C (-0.30±0.52) and low-density lipoprotein cholesterol (LDL-C) to HDL-C (-0.18±0.45) (all
P^.05) but no change in lipoprotein(a). CETP concentration (in milligrams per liter) fell significantly in
response to training in both men (n=28, 2.47±0.66 to 2.12±0.43; % A=14.2%; P<.005) and women
(n=29, 2.72±1.01 to 2-J6±0.76; % A=13.2%; P<M7). The CETP change was observed both in subjects
who lost weight (n=28, A mean weight=-5.0 kg; A CETP=-0.42±0.79; % A=15.4%; P<.009) and in
those who were weight stable (n=29, A mean weight=-0.12 kg; A CETP =-0.29±0.78; % A=10.4%;
P<.055). Pretraining plasma CETP concentration predicted training-associated changes in HDL-C
(r= -.27, P<M) and ratio of LDL-C to HDL-C (r=+.40, P<.002). In a smaller study of 15 men, exercise
training was associated with a decrease in levels of CETP, an increase in plasma postheparin lipoprotein
lipase (LPL) activity, and a decrease in hepatic trigfyceride lipase (HTGL) activity. Overall, the data
suggest that basal plasma CETP concentrations, in addition to LPL and HTGL activities, may contribute
to determining the extent to which exercise redistributes cholesterol among plasma lipoproteins.
(Arterioscler Thromb. 1993;13:1359-1367.)
KEY WORDS • body composition • percent body fat • maximal oxygen uptake • exercise training
• lipoprotein (a) • insulin • glucose tolerance • apolipoproteins
L
ong-term exercise training affects numerous metabolic and body compositional parameters.
Lipid metabolic changes include a reduction of
plasma triglycerides1 and the redistribution of plasma
total cholesterol from the very-low-density lipoprotein
(VLDL) fraction to the high-density lipoprotein (HDL)
fraction.2 The processes responsible for the redistribution of plasma cholesterol to give higher HDL cholesterol (HDL-C) levels are not fully understood. Several
enzymes, including lipoprotein lipase (LPL), hepatic
triglyceride lipase (HTGL), lecithin: cholesterol acyltransferase (LCAT), and the cholesterol ester transfer
protein (CETP), modify plasma lipoprotein particles
and may play a physiological role in HDL metabolism.3
In relation to exercise, an increase in plasma LPL
activity and a decrease in HTGL activity4-3 probably
contribute to the rise in HDL-C. On the other hand,
exercise has no effect on LCAT mass6 and only a small
and temporary effect on LCAT activity.7 The effect of
exercise on plasma CETP mass or activity is not known.
Received December 21, 1992; revision accepted June 9, 1993.
From the Departments of Internal Medicine, Washington University School of Medicine, St Louis, Mo (R.L.S., W.M.K., K.M.,
C.F.S., R.O., G.S.), and The College of Physicians and Surgeons,
Columbia University, New York, NY (P.M., T.C., A.T.).
Reprint requests to Richard L. Seip, PhD, Washington University School of Medicine, Campus Box 8046, 660 South Euclid Ave,
St Louis, MO 63110-1093.
CETP catalyzes the net flux of esterified cholesterol
from cholesterol ester-rich particles (ie, HDL) to
larger, triglyceride-rich acceptor plasma particles (ie,
VLDL and chylomicrons) and to low-density lipoprotein (LDL).3-8 In humans, CETP deficiency is associated
with extremely high HDL-C levels,910 low LDL-C levels,10 and a low ratio of plasma total cholesterol to
HDL-C.10 Tissue sources of CETP include the liver,
intestine, fat, and muscle.11-12 Dietary cholesterol intake
increases plasma CETP mass and activity,12 but there
are few other data to provide insight into factors
associated with fluctuations in CETP mass in humans.
Because exercise increases HDL-C, we hypothesized
that exercise may also decrease CETP mass. We therefore compared plasma CETP concentrations in adult
men and women before with those after as much as 1
year of exercise training. In addition, before and after
exercise training, we related plasma CETP concentration to parameters such as body composition, plasma
lipid and apolipoprotein levels, glucose tolerance and
insulin responses to an oral glucose load, and maximal
oxygen uptake (VOi^), many of which are altered by
exercise training.
Methods
Subjects
In the main study, 57 previously sedentary men
(n=28) and women (n=29) 60 to 72 years old were
1360
Arteriosclerosis and Thrombosis
Vol 13, No 9 September 1993
TABLE 1. Baseline Characteristics of Men and Women in
Exercise Study
No. of subjects
Age(y)
Body weight (kg)
Sum of skinfolds (mm)
Body fat (%)
Fat-free mass (kg)
Fat mass (kg)
Ratio of fat-free mass to fat mass
V o ^ (mL • kg"1 • mirr1)
V o ^ (IVmin)
Women
Men
29
63.8±2.8
67.3 ±12.8
166±45
37.8±4.7
(n=18)
45.0±8.0
(n=18)
27.6±6.8
(n=18)
1.69±0.38
(n=18)
22.4±3.2
1.53 ±0.29
28
64.1±3.2
81.9±12.1
128±46
28.6±6.5
(n=25)
58.4±5.6
(n=25)
24.3 ±9.0
(n=25)
2.65±0.75
(n=25)
28.2±4.4
2.27±0.32
Values are mean±SD. VOina indicates maximal oxygen uptake.
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
studied. All had given informed consent to participate
in studies of exercise training effects approved by the
Human Studies Committee of the Washington University School of Medicine. The subjects were healthy
nonsmokers who normally were active but had not
engaged in exercise training (defined as 30 minutes of
aerobic activity at least 2 days per week). Health status
was evaluated from the following screening procedures:
medical history, physical examination, SMA-12, chest
radiograph, resting electrocardiogram (ECG), and a
Bruce treadmill exercise test with ECG and blood
pressure monitoring. Subjects were excluded from the
study if they had medical problems that could interfere
with interpretation of the results and/or performance of
vigorous exercise. Those with evidence of impaired
glucose tolerance (2-hour oral glucose-tolerance test
plasma glucose level between 140 and 200 mg/dL)13
were excluded. Four of the women received estrogen
replacement therapy throughout the study period. Subject characteristics at baseline are shown in Table 1. In
a follow-up experiment to evaluate exercise-related
changes in plasma CETP, postheparin plasma lipase
activities, and plasma lipoproteins, 15 healthy men were
studied before and after short-term training. Five young
normolipidemic men (age range, 26 to 40 years; percent
body fat [%BF] = 15.9±5.4%; Vo 2 m i I = 56.4±4.7
mL • kg"1 • min"1) and 10 healthy older men (age range,
56 to 72 years; %BF=27.9±4.0%; Vo2mM=27.6±2.1
mL • kg"1 • min"1) participated in this study. Of the older
men, 4 had fasting plasma triglyceride levels of the 90th
percentile or higher, and 2 had HDL-C levels in the fifth
percentile or lower. The others were normolipidemic.
Exercise Training
In the main study of 57 subjects, the subjects were
studied after a 2-month flexibility exercise program that
preceded the endurance exercise and again after 9 to 12
months of endurance exercise training. The endurance
exercise training consisted of three to five supervised
45-minute to 1-hour sessions of exercise (walking, jogging, and/or cycle ergometry) per week for 9 to 12
months. Exercise intensity gradually increased from
65% of maximal heart rate initially to 80% to 85% of
maximal heart rate at the end of the training period.14
All subjects averaged at least three supervised sessions
per week; the mean±SD number of sessions per week
attended was 4.0 ±0.6.
Exercise Energy Expenditure
Exercise energy expenditure rates were estimated
from walking velocities, running velocities, and cycle
power outputs recorded for 4 weeks near the end of
exercise training, using formulas published by the
American College of Sports Medicine.14 One liter of
oxygen consumption was assumed to require 5 kcal of
energy expenditure.
Dietary Monitoring
Subjects completed 7-day food records on four occasions, twice before beginning the endurance training
program and then once at the midpoint and once at the
completion of the training program. Foods were
weighed by the subjects and recorded in household
measures, and the registered dietician who instructed
the subjects on how to record food intake also conducted an interview to assess the accuracy of the
portions that were recorded. The food intake records
were analyzed using the DATADIET NUTRIENT ANALYSIS
program (IPC Datadiet, Camarillo, Calif).
Blood-Sampling Procedure
Blood was sampled before and after the training
period from a subcutaneous arm vein with the subject
seated or in a supine position after an overnight fast and
16 hours after the most recent exercise bout. Plasma was
separated from cells via centrifugation (1000gx20 minutes) at 4°C. Plasma samples were frozen at -70°C for
later analysis.
Plasma Lipid, Lipoprotein, and
Apolipoprotein Measurements
Lipoproteins were measured in the Washington University Lipid Research Clinic Core Laboratory, which
participates in the lipid standardization program of the
Centers for Disease Control and Prevention.15 Total
HDL-C, HDL2-C, and HDL,-C were determined after
selective precipitation of whole plasma using the heparin-manganese procedure.1617 CETP mass was measured with solid-phase competition radioimmunoassay18
using recombinant CETP derived from transfected mammalian cells to coat the wells and a calibrated plasma
pool as the standard. For any given subject, pretraining
and posttraining samples were analyzed in the same
assay. Three assays were used to assay all the samples.
The mean intra-assay coefficient of variation was 6.3%.
Apolipoproteins AI, B, and E were measured by enzymelinked immunosorbent assay (ELISA) as previously described.19-20 Lipoprotein(a) [Lp(a)] was determined by
ELISA21 using the Macra Lp(a) immunoassay kit (Terumo Medical Corporation, Elkton, Md).
Postheparin Plasma Lipase Activity Measurements
(Follow-up Study Only)
Plasma samples for lipase activity analysis were collected 15 minutes after administration of 60 U heparin/kg body wt IV. Pretraining and posttraining samples
were stored at -70°C before being assayed for lipase
activity in triplicate in the same assay, as described
Seip et al
CETP and Exercise
1361
TABLE 2. Plasma LJpid and LJpoproteln Changes With Training
Triglyce rides
VLDL triglycerides
LDL+HDL triglycerides
Total cholesterol
VLDL-C
LDL-C
HDL-C
HDL2-C (n=39)
HDL3-C (n=39)
Pretraining
Posttraining
A
P
127.4+.52.9
85.9±49.4
44.1+8.9
(n=46)
201.8+27.1
17.1 + 12.7
129.0+24.7
52.8+13.9
17.4+9.8
32.9±6.7
101.7±38J
61.2±34.9
44.2±7.7
-25.7±36.3
-24.8±34.9
-0.87±5.22
(n=46)
-8.1±22.6
-2.7±21.3
-5.9±22.0
+2.6±6.2
+0.3±5.3
+1.5±5.5
.0001*
.0001'
.265
193.7±29.8
14.5±20.3
124.1 ±28.6
55.3±15.5
17.7±11.0
34.4±62
.009*
.349
.049*
.003*
.766
.092
VLDL indicates very-low-density lipoprotein; LDL, low-density lipoprotein; HDL, high-density lipoprotein; C,
cholesterol. A, Change between pretraining and posttraining.
Values in mg/dL, mean±SD. n=57 unless otherwise indicated.
•Significant difference from pretraining to posttraining.
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
previously.22 Plasma aliquots (10 to 40 /iL) were assayed
in 200 fiL (final volume) of either high- (0.75 mol/L) or
low- (0.10 mol/L) salt (NaCl) concentration assay mixture. Both high-salt and low-salt assay mixtures contained 30 fiL tris(hydroxymethyl)aminomethane (Tris)HC1 buffer (0.223 mol/L, pH 8.5), 10 jiL heatinactivated rat serum, and 120 /JLL triglyceride substrate
emulsion. The substrate emulsion was made fresh in
6-mL aliquots containing 2.5 /xCi [uC]triolein (Amersham, Arlington Heights, 111), 50 mg nonradioactive
triolein (Sigma Chemical, St Louis, Mo), 6 mg lecithin
(Sigma Chemical), and 550 mg albumin, sonicated in
Tris-HCl (0.223 mol/L, pH 8.5) and 0.13 mol/L NaCl.23
Hepatic lipase activity was taken as the salt-insensitive
activity. Plasma LPL activity was taken as the saltsensitive activity (total activity minus salt-insensitive
activity).
cient, r=.99) using Lange calipers.28 Sites measured
included pectoral, subscapular, midaxillary, umbilical,
suprailiac, and thigh. All measurements were performed in duplicate. If values differed by more than 1.0
mm, additional measurements were taken. Outlying
values (more than 4.0 mm) for a site were eliminated,
and the remaining values were averaged.
was measured during a graded treadmill walking
and jogging protocol designed to increment exercise intensity by 3 to 4 mL O2 • kg • min"1 and elicit fatigue within
6 to 12 minutes.24 Cardiorespiratory data were collected at
30-second intervals using a computerized system that
included a Parkinson-Cowan CD-4 dry gas meter, O2
(Applied Electrochemistry S3-A) and CO2 (Beckman
LB-2) gas analyzers, and a 5-L mixing chamber. ^ 0 2 ^
was taken as the highest V02 recorded over two consecutive 30-second periods. To ensure that V c ^ had been
attained, at least two of the following criteria had to be
satisfied: plateau in Vo^, heart rate within 10 beats per
minute of age-predicted maximum heart rate, and respiratory exchange ratio of more than 1.10.
Statistical Analysis
Student's paired t test was used to test for differences
before and after training. For nonnormaUy distributed
variables, the Wilcoxon nonparametric sign test was
used. Pearson's correlations and partial correlations
were computed using SAS statistical software (SAS Institute, Cary, NC).
Percent Body Fat
Percent body fat was estimated from underwater
weight taken at partial exhalation.25 Lung residual volume was measured outside the hydrostatic weighing
tank using the oxygen dilution procedure.26 Body density was converted to percent body fat using the equation of Brozek et al.27 The method became available in
time to test the last 17 women and 24 men entering the
study.
Skinfold Measurements
Skinfold thicknesses were assessed by one of two
experienced technicians (interrater reliability coeffi-
Oral Glucose-Tolerance Test
After an overnight fast, subjects ingested 75 g glucose.
Plasma samples obtained at 0, 30, 60, 90, 120, and 180
minutes were enzymatically analyzed for glucose (model
23A glucose analyzer; Yellow Springs Instruments, Yellow Springs, Ohio), and insulin was analyzed by radioimmunoassay.29 Areas under the plasma glucose and
insulin concentration curves were determined using the
trapezoidal rule.
Results
During the last 4 weeks of the training period, training
bouts required 1196 ±343 kcal/wk for women and
1814±669 kcal/wk for men. Expressed relative to body
weight, energy expenditure was 18.8±6.6 kcal • kg"1 • wk"1
for women and 24.2±8.9 kcal • kg"1 • wk"1 for men. Exercise intensity, expressed as percent of maximal heart rate,
was similar for both men and women (81 ±4% in women
and 82±6% in men). Basal V02mo values are given in
Table 1. As reported previously,24 training increased
V c w by 4.2 mL O 2 • kg"1 • min"1 (18.8%) (/><.0001) in
women and 6 3 mL O2 • kg"1 • min"1 (223%) (P<.0001) in
men. The respective changes expressed as liters per
minute were 0.22 (13.7%) and 0.41 (18.1%) (P<.0001).
Mean body weight decreased by 2.4±3.9 kg (f<.002) in
women and 2.6±33 kg (P<.003) in men. Body composition analyses of a subset of 17 women and 24 men
indicated that fat loss accounted for 71% of the weight loss
in women and 92% of the weight loss in men.
1362
Arteriosclerosis and Thrombosis
Vol 13, No 9 September 1993
n=28 mm
fc 3
o
p<0.005
p<0.047
FlG 1. Plots of plasma cholesterol ester transfer protein (CETP) mass concentration before and after exercise in women (A) and men (B). *Significant difference
from pretraining to posttraining.
1
i
PRE
PRE
POST
Exercise Training
Plasma Lipoprotein
Lipid
POST
Exercise Training
Changes
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
For the group as a whole, exercise significantly decreased plasma triglycerides by 20.2%, VLDL trigrycerides by 28.8%, and plasma total cholesterol by 4.0%
(Table 2). HDL-C increased significantly by 4.9%, and
LDL-C decreased by 4.6%. In subjects with complete
pretraining and posttraining measurements of HDL
subtractions, the increase in HDL3-C (1.5 mg/dL) accounted for most of the increase in total HDL-C (1.8
mg/dL). Patterns of response were identical for men
and women; for example, mean triglycerides decreased
by 30.5 mg/dL in women and 22.5 mg/dL in men. The
respective changes in women and men were —8.4 and
-8.6 for total cholesterol, -6.2 and -6.1 for LDL-C,
and +3.3 and +2.2 for HDL-C (all in mg/dL).
Plasma CETP, Lp(a), and Apolipoprotein Changes
Training significantly affected plasma CETP concentration (Fig 1 and Table 3) as well as plasma apolipoprotein B concentration (Table 3). CETP concentration
decreased by 13.5% (/><.001), and apolipoprotein B
concentration decreased by 17.3% (F<.0001). However,
Lp(a) and apolipoproteins E and AI did not change
significantly.
Effects of Sex and Weight Loss on
Plasma CETP Change
Mean levels of CETP decreased significantly in both
men (from 2.47±0.66 to 2.12±0.43 mg/dL; 14.2%;
/>=.005) and women (from 2.72±1.01 to 2.36±0.72
mg/dL; 13.2%; P=.O12) (Fig 1). Because weight changes
within the subjects ranged from +2.7 to -17.9 kg
(mean±SD change, -2.5 ±3.5 kg), CETP data were
further analyzed according to weight loss. Fifteen
women and 14 men were weight stable (ie, lost less than
2.1 kg and had mean±SD change of -0.12±1.30 kg). In
20 of these 29 subjects, mean fat mass decreased by 0.5
kg and lean mass increased by 0.2 kg according to body
composition analyses. The mean CETP level decreased
in these 29 subjects from 2.80±0.87 to 2.54±0.73 mg/dL
(9.3%) (Fig 2). Fourteen women and 14 men lost more
than 2.1 kg (mean±SD change, -4.98±3.43 kg). Body
composition analyses performed in 21 of these 28 subjects showed that fat mass decreased by 3.9 kg. Mean
CETP levels also decreased in this group, from
2.72+1.15 to 2.16±0.41 mg/dL (20.6%) (Fig 2). Thus,
neither weight loss nor sex affected the exercise-related
CETP response. Lp(a) was unaffected by exercise regardless of sex.
Plasma Lipoprotein Lipid Distribution
and Composition
Exercise training changed the distribution of plasma
total cholesterol among the major lipoproteins as well as
the compositions of some lipoproteins (Table 4). The
ratios of total plasma cholesterol to HDL-C and of LDL
to HDL decreased by -8.0% and -7.4%, respectively,
but the ratio of VLDL-C to HDL-C remained unchanged. In a subset of 39 subjects in whom the HDL
subfractions were quantified, training decreased the
ratio of VLDL-C to HDL2-C by 36% and the ratio of
VLDL-C to HDL3-C by 31%, but the ratio of HDL 2 -C
to HDL3-C was unaffected. Relative to apolipoprotein B
content, VLDL and LDL contained less cholesterol
after training (Table 4, bottom). In addition, the ratio of
HDL-C to apolipoprotein AI was increased after
training.
Triglyceride distribution also was affected by training
(Table 4). The depletion of VLDL triglycerides was
greater than the depletion of non-VLDL triglycerides,
resulting in a decrease of the ratio of VLDL triglycerides to non-VLDL triglycerides by 22.9% and of the
ratio of VLDL triglycerides to plasma triglycerides by
9.6%. Although the depletion of plasma triglycerides
TABLE 3. Apolipoprotein Changes With Training
A
P
2.71 ±2.38
65.8±24.9
-0.72±3.06
-13.7±21.6
129.7±25.8
9.5±11.6
-3.5±16.1
-0.6±3.9
-0.35±0.78
.081
.0001'
.102
Pretraining
Posttraining
Apo B
3.43±2.46
79.4 ±33.0
Apo AI
126.2±27.2
ApoE
Lp(a)
CETP
10.1±12.9
2.60±0.86
2.24±0.63
.221
.001*
Apo indicates apolipoprotcin; Lp(a), lipoprotein(a); and CETP, cholesteryl ester transfer protein.
Values in mg/dL except CETP, which is jig/mL; mean±SD. n=57. See Table 2 for explanation of A
and *.
Seip et al
1.5 r
1.0 0.5
o.o
-0.5
p<0.055
-1.0
<u
en
o
.c
°9
o
o
p<0.009
-1.5
-2.0
o
-3.5
WEIGHT
STABLE
WEIGHT
LOSERS
Group
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
FIG 2. Bar graph of lack of effect of weight loss on decrease
in plasma cholesterol ester transfer protein (CETP) mass
concentration associated with exercise. ^Significant difference
from pretraining to posttraining.
was larger than that of plasma cholesterol after training,
the ratios of triglyceride to cholesterol in both the
d< 1.006 (VLDL) and the d> 1.006 (LDL+HDL) fractions were not affected.
Correlations Between Plasma CETP Concentration
and Other Parameters
The directions (— or +) of the correlation coefficients
(r) were the same before (Fig 3A) and after (Fig 3B)
training for the significant correlations. Percent body fat
was positively correlated with CETP after training
(r=386, P<.009), and lean body mass was negatively
correlated before training. However, there were no significant correlations between either basal CETP or A CETP
and A % fat mass or A lean body mass (A indicates change
between before and after training). This was true even in
the subgroup that lost more than 2.1 kg. The sum of
CETP and Exercise
skinfold measurements was not correlated with CETP.
Vo2mM was inversely related to plasma CETP after training
(r= - .291, P< .03). Plasma total cholesterol was correlated
to plasma CETP after training (r=.262, P<.05). None of
the measured apolipoproteins or triglyceride parameters
were significantly related to plasma CETP concentrations
(-.04>r>+.13, data not shown). Of the plasma cholesterol ratios, none were significantly related to CETP at
either time point (-.20<r<.20). Insulin response to an
oral glucose load was not significantly correlated to plasma
CETP either before (r=.2O9, P<.13) or after (r=.255,
P<.01) training.
Prediction of Plasma Cholesterol Redistribution From
Plasma CETP Concentration
Of particular interest to us were the exercise-induced
changes in the distribution of plasma cholesterol with
training (eg, A [LDL-C/HDL-C]) and their possible
relationship to CETP concentration (Fig 4 and Table 5).
Basal (ie, pretraining) CETP levels, which ranged from
1.20 to 5.18 mg/L, were negatively correlated with the
exercise-associated change in HDL-C ( r = - . 2 6 7 ,
P<.04) (Fig 4A) and positively correlated with the
change in the ratio of LDL-C to HDL-C (r=.396,
/><.002) (Fig 4B). Removing the effects of basal and
training-associated changes in cholesterol and triglycerides through partial correlations strengthened each of
these relationships with basal CETP (Table 5). Thus,
removing the effects of A triglycerides and A total
cholesterol changed the correlation between basal
CETP and A HDL-C from -0.267 to -0.335. The r
value for A LDL-C/HDL-C changed from +0.396 to
+0.418, and the rvalue for A total cholesterol/HDL-C
increased from +0.241 to +0.390.
Correlations between pretraining vs posttraining
changes for A CETP, A HDL-C, A total cholesterol/
HDL-C, and A triglycerides ranged from -.196 to
+ .340. For example, A triglycerides vs A HDL-C was
-.048 (P=NS). The only significant correlation was
r=.34O (P<.010), for A triglycerides vs A total choles-
TABLE 4. Changes in Plasma Cholesterol and Triglyceride Distribution and Lipoprotein Composition
With Training
Cholesterol
Total cholesterol/HDL
LDL/HDL
VLD1VHDL
VLDUHDL: (n=39)
VLDlVHDLj (n=39)
HDWHDLj (n=39)
Triglycerides
VLDL/plasma
VLDL/non-VLDL (n=46)
Compositions
VLDL: triglycerides/C
Non-VLDL:trigh/cerides/C (n=46)
(VLDL-C+LDL-C)/apo B
HDL-C/apo AI
1363
Pretraining
Posttraining
4.03 + 1.00
2.58+0.77
0.38±0.39
1.89±2.69
0.61 ±0.50
0.52±0.24
3.73±1.03
2.40±0.88
0J0±0.41
1.22±1.45
0.42±0.38
0.50±0.28
0.299+0.520
0.181 ±0.453
0.077 ±0.426
0.675±1.781
0.188±0.333
0.018±0.241
.0001
.004'
.180
.024*
.001 •
.631
0.64±0.12
2.00±1.34
0.57±0.11
1.38±0.76
0.068±0.094
0.618±0.967
.0001
.0001
5.72*1.57
0.24±0.06
2.10±0.75
0.42 ±0.09
5.71±1.68
0.25 ±0.06
1.98±0.77
0.44±0.11
0.006 ±1.93
0.006±0.045
0.122±0.385
0.020±0.049
.981
.385
.020*
.0003
HDL indicates high-density lipoprotein ; LDL, low-density lipoprotein; VLDL, very-low-density lipoprotein; C,
cholesterol; and apo, apolipoprotein.
Values are mean±SD of mass ratios. n=:57 unless otherwise indicated. See Table 2 for explanation of A and *.
1364
Arteriosclerosis and Thrombosis
Vol 13, No 9 September 1993
B. Post Training
A. Pre Training
Variable:
BODY WT
LBM
XBF
SUM SK
Og.ml -kg «mln
TC
VLDL-C
LDL-C
HDL-C
TC/HOL-C
LDL-C/HDL-C
(VLDL-C+LDL-C)/HDL-C
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
VLDL-C/HDL-C
VLDL-C/HDLj-C
HDLJ-C/HDLJ-C
INS AREA
-0.4 -0.2
0.0
0.2
0.4
-0.4
Pearson r value
-0.2
0.0
0.2
0.4
Pearson r value
FIG 3. Graphs of Pearson's correlations between plasma cholesterol ester transfer protein mass concentration and various
laboratory measurements before (A) and after (B) training (n=57 except where noted in parentheses). *P<.05. WT indicates
weight; LBM, lean body mass; BF, body fat; SK, skinfold measurements; TC, total cholesterol; VLDL, very-low-density
lipoprotein; LDL, low-density lipoprotein; HDL, high-density lipoprotein; C, cholesterol; and INS, insulin.
terol/HDL-C. The respective r values for A CETP vs A
HDL-C, A total cholesterol/HDL-C, and A LDL-C/
HDL-C were +.229 (/>=.090), +.184 (P=.174), and
-.008 (/»=.952). For percent A CETP vs percent A
HDL-C, r=.104 (P<.446).
Results of Follow-up Study
Given the correlation between basal CETP and A HDL
in the original 57 subjects, we wanted to compare basal
CETP with changes in plasma lipase activities as factors
contributing to the exercise-induced increase in HDL-C.
Therefore, an additional study was conducted in which the
responses of five runners (running at least 20 miles per
week) who stopped training for 2 weeks were combined
with the responses of 10 healthy sedentary men who
exercised for 10 to 13 consecutive days for more than 1
hour per day at 60% to 70% of maximal capacity. The
results were consistent with those found in the original 57
subjects. For example, exercise training decreased plasma
CETP from 2.59 to 2.13 mg/L (17.8%) (P<.008). Also,
plasma triglycerides decreased by 46 mg/dL (/)<.O7) (percent change, 15.5%; /><.0007), HDL-C increased by 4.0
mg/dL (P<.018), HDL 2 -C increased by 1.6 mg/dL
(P<.008), LPL activity increased by 2.67 /unol free fatty
FTG 4. Scatterplots showing the
relationship of plasma cholesterol
ester transfer protein (CETP)
mass concentration before training to exercise-associated changes
in plasma high-density lipoprotein
cholesterol (HDL-C) concentration (A) and plasma ratio of lowdensity lipoprotein cholesterol
(LDL-C)IHDL-C (B).
2
3
4Plaima CETP [mg/L]
2
3
4
Plaima CETP [mg/L]
Seip et al
CETP and Exercise
1365
TABLE 5. Pearson Partial Correlations Showing Relationships Between Baseline (Pretralning) Plasma CETP Mass Concentration and
Exercise-Associated Changes in HDL-C, Total Cholesterol /HDL-C, and LDL-C/HDL-C With Effects of Plasma Triglycerides and
Cholesterol Removed
Effect removed
Basal plasma triglycerides
Basal plasma total cholesterol
A Triglycerides
A Total cholesterol
A Plasma triglycerides, A plasma total cholesterol
Baseline CETP
vs A HDL-C
-.268*
-.246
-.295*
-.288*
-.335*
Baseline CETP vs A plasma total
cholesterol/HDL-C
.241
.267*
.356t
.273*
.390f
Baseline CETP vs
A LDL-C/HDL-C
.398t
.431t
.370t
.466t
.418t
CETP indicates cholesteryl ester transfer protein; HDL, high-density lipoprotein; LDL, low-density lipoprotein; and C, cholesterol. See
Table 2 for explanation of A and '.
•P<.05; f/'<.01.
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
acids • h ' • mL" 1 (P<.01), and hepatic lipase activity decreased by 1.38 ^.mol free fatty acids • h" 1 • mL" 1
(/)<.OO2). Pearson product moment correlations are
shown in Table 6. Correlation coefficients between basal
CETP with A HDL-C, A HDL 2 -C, and A HDL 2 -C/
HDL3-C were negative, as before, but not significant.
There was no correlation between A LDL-C/HDL-C or A
total cholesterol/HDL-C and basal CETP. Lipase activity
changes, expressed as percent changes, were strongly
related to the changes in fasting triglycerides but were not
correlated with H D L - C changes. Removing the effects of
LPL or H T G L activity did not significantly change the
correlation between baseline CETP and A HDL-C. Similarly, the relationships between changes in LPL or H T G L
activity with changes in H D L - C were not changed by
removing the effects of either or both lipase activity
changes.
Discussion
The main finding of the present study was that 9 to 12
months of aerobic exercise training, performed under
supervision four or five times weekly at an intensity
level of 8 1 ± 5 % maximum heart rate by healthy nondiabetic persons over the age of 60, significantly lowers
the plasma concentration of CETP in both men and
women. In concordance with the findings of many other
longitudinal exercise lipid studies in younger populations, exercise increased H D L - C by 2.6 mg/dL and
decreased triglycerides by 20.2%, with the directions
and magnitudes of changes similar for men and women.
The present findings therefore provide evidence that
both men and women in this age group (60 to 70 years
old) can improve their fasting lipid profiles through
regular exercise training.
Although plasma volume was not measured in the
present study, it is possible that part of the decrease in
CETP is attributable to plasma volume expansion. Exercise training expands plasma volume, with the magnitude
of the expansion depending on a number of factors,
including rate and amount of energy expenditure. The
plasma volume response of subjects 60 to 70 years old is
not known, but in younger men (mean age, 32 to 37 years),
comparable long-term training at a higher absolute intensity30 or requiring a twofold greater weekly caloric expenditure 31 expanded plasma volume by 6% to 8% (and
increased V o ^ by 18% to 26%). If older subjects (ages,
60 to 70 years) respond similarly, the 15.9% increase
(L/min) in Vo2m« observed in the present study makes it
reasonable to assume a plasma volume expansion of 5 % in
the present subjects. If such a change in plasma volume
had occurred, it would increase the total plasma CETP
concentration following training by 5 % ([2.24 fig/
mL]x[1.05 L]=2.37), leaving a net 8.8% decrease in
plasma CETP due to exercise.
Plasma CETP concentrations decreased in a subset of
subjects whose weight remained stable (mean change,
—0.1 kg) as well as in subjects who lost weight (mean
change, - 5 . 0 kg). The absence of a significant correlation between A CETP and A lean body mass or A
percent body fat, even in the subgroup that lost more
than 2.1 kg, suggests that depletion of adipocyte triglyceride per se may not affect C E T P production. Thus, the
lowering of CETP concentration through exercise occurred independently of weight change or sex. Previous
reports have shown that plasma CETP concentration
increases in response to a high-fat, high-cholesterol
diet, 32 indicating a link to nutritional status. In the
present study, reported dietary fat intake was typical of
TABLE 6. Results of Follow-up Study of 15 Men Showing Correlations Between Changes in Plasma HDL-C
Parameters and Factors That May Influence Response of HDL-C to Exercise
Variable
A Plasma triglycerides
A HDL-C
AHDLrC(n = 12)
A HDL3-C (n=12)
A HDL2-C/HDLj-C (n-12)
A Plasma total cholesterol/LDL-C
A LDL-C/HDL-C
A LPL activity (%)
-.77 (/><.001)
-.08
-.10
-.03
-.09
.04
-.04
A HTGL activity (%)
0.24
0.35 (P<.20)
-0.01
0.19
-0.06
0.03
0.18
Baseline CETP
-0.24
-0.12
-0.38 (/><.24)
0.34 (/><.28)
-0.47 (P<.12)
-0.05
-0.11
LPL indicates lipoprotein lipase; HTGL, hepatic triglyceride lipase; CETP, cholesteryl ester transfer protein; HDL,
high-density lipoprotein; LDL, low-density lipoprotein; and C, cholesterol. See Table 2 for explanation of A and *.
1366
Arteriosclerosis and Thrombosis
Vol 13, No 9 September 1993
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
most Americans (not shown) and remained constant
during training.28 That CETP decreased during exercise
regardless of sex, weight change, and reported dietary
status argues for a specific effect of exercise on plasma
CETP levels.
Percent body fat was weakly positively and lean body
mass was weakly negatively correlated with CETP before and after exercise training (r range, .3 to .4). In
mammals, fat tissue contains mRNA for CETP,12 and
synthesis of CETP by fat tissue has been postulated.
The present data advance the idea that fat mass may be
an important determinant of human plasma CETP. We
subsequently have examined the relationship between
percent body fat and plasma CETP levels in 28 men who
varied in body fat (range, 6% to 34% of body weight)
and found a significant correlation (r=.44, P<.04) (unpublished data), further supporting this hypothesis. By
contrast, the sum of skinfold measurements, which is an
index of subcutaneous fat deposition, was not related to
CETP either before or after exercise training. That
overall body fatness but not subcutaneous fat is related
to plasma CETP concentration may mean that intraabdominalry deposited fat is the more important determinant of plasma CETP concentration.
Exercise significantly reduced the insulin response to
a glucose challenge. Such a response is consistent with a
reduction in peripheral insulin resistance. It is tempting
to link the decrease in CETP to an exercise-related
decrease in peripheral insulin resistance, perhaps at the
adipose tissue level. However, this speculation must be
tempered by the weakness of the correlations among
CETP, adipose tissue, and insulin resistance. Clearly,
more definitive studies are required.
The exercise-induced increase in HDL-C has been
attributed to changes in plasma lipase activities. In
highly fit runners, whose plasma HDL-C levels generally are high, a significant portion of the variance in
HDL-C levels is explained by variance in postheparin
plasma LPL activity.5-33-34 In sedentary adults, negative
correlations between hepatic lipase activity and HDL-C
levels,3337 particularly HDL2-C levels,35-37-38 often are
stronger than LPL activity vs HDL-C relationships.
Exercise intervention increases postheparin plasma
LPL activity,4-31-33-36-39 may decrease HTGL activity,4-31-33-36-39 and increases plasma HDL-C,4-22-31-33-36-39
yet significant correlations between A HDL-C and
exercise-associated changes in lipase activities have not
been demonstrated.4-33-36-39 The highest correlations
(r=.33, .34; />=NS) between changes in HDL-C and
LPL activity have been reported by Peltonen et al.36 The
results of our follow-up study agree with results of these
previous studies. We detected significant increases in
plasma trigrycerides, HDL-C, and LPL activity and a
decrease in HTGL activity, but lipase changes were not
correlated with HDL-C changes.
We hypothesized that plasma CETP concentration
may be an additional factor related to the exerciseassociated increase in plasma HDL-C. In our sample of
57 healthy men and women, we found that basal CETP
concentrations were negatively and significantly correlated with the exercise-associated changes in HDL-C
(r=-.27, P<M) and the ratio of LDL-C to HDL-C
(r=-.4O, P<.002). These negative correlations are consistent with the physiological role of CETP in the
redistribution of cholesterol esters from HDL to larger
trigryceride-rich particles. Removing the effects of baseline cholesterol and of the change in trigrycerides
through statistical means further strengthened the relationships of HDL-C, LDL-C to HDL-C, and total
cholesterol to HDL-C with basal CETP mass. In the
additional study of 15 men, the plasma CETP decrease
with training seen in the main study was confirmed, and
the correlations between A HDL-C and A LDL-C to
HDL-C with baseline CETP were in the same direction
as those found in 57 subjects. Given the difficulty of
detecting significant correlations between A HDL-C
and lipase activity changes, it is not surprising to see
only trends between either A HDL-C or A HDL 2 -C and
baseline CETP in our follow-up study.
The predictive value of basal CETP concentration on
the HDL-C and ratio of LDL-C to HDL-C responses
may reflect the exponential relationship between
plasma CETP concentration and HDL-C, as defined by
human genetic CETP deficiency.10 This relationship
predicts that equal reductions in CETP will have a much
larger effect on HDL-C when baseline CETP levels are
lower. Based on these observations, we speculate that a
high basal CETP level predisposes to a small exerciseinduced increase in HDL-C and a small decrease in
LDL-C to HDL-C. Alternatively, with a low basal
plasma CETP level, a wider range of plasma cholesterol
distribution responses is possible, including a large
increase in HDL-C, which may depend on the extent to
which plasma LPL activity increases. Obviously, these
speculations need to be tested.
In summary, long-term exercise decreased plasma
CETP levels in both men and women, independent of
weight change. Positive relationships between plasma
CETP and LDL-C and between plasma CETP and total
cholesterol, observed both before and after exercise
training, support the link between CETP and cholesterol metabolism. We also reported a positive correlation between body fat and plasma CETP concentration,
which may be evidence that adipose tissue in humans is
an important source of CETP. With respect to the
increase in HDL-C associated with exercise, low basal
plasma CETP concentration, but not change in CETP,
was found to be a significant predictor of the HDL-C
increase associated with exercise and a stronger predictor of the decrease in the ratio of LDL-C to HDL-C.
Because the mechanism of exercise-induced HDL-C
increase remains incompletely understood, the role of
CETP in exercise-induced changes of lipoprotein cholesterol distribution warrants further study.
Acknowledgments
Supported by National Institutes of Health (NIH) Program
Project Grant AG-05562; NIH grants HL-21006, HL-29229,
and HL-47436; Diabetes Research and Training Center Grant
AM-20579; and a Qinical Research grant from the American
Diabetes Association. R.L.S. was the recipient of NIH Service
Award 1-F32-HL-084O9-01. The authors thank Dr Thomas
Cole and the Lipid Research Center for plasma lipid analyses,
Dr John O. Holloszy for his suggestions regarding the manuscript, and the staff of the Applied Physiology Laboratory at
Washington University School of Medicine for conducting
exercise testing, exercise training, and body compositional
analyses on the subjects. Computing facilities were provided by
the General Clinical Research Center.
Seip et al CETP and Exercise
References
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
1. Holloszy JO, Skinner JS, Toro G, Cureton TK. Effects of a six
month program of endurance exercise on the serum lipids of
middle-aged men. Am J Cardiol 1964;14:753-760.
2. Haskell WL. The influence of exercise training on plasma lipids
and lipoproteins in health and disease. Ada Med Scand SuppL
1986;711:25-37.
3. Tall AR. Plasma high density lipoproteins. / Clin Invest. 1990;86:
379-384.
4. Kantor MA, Cullinane EM, Sady SP, Herbert PN, Thompson PD.
Exercise acutely increases high density lipoprotein cholesterol and
lipoprotein lipase activity in trained and untrained men.
Metabolism. 1984;36:188-192.
5. Sady S, Cullinane EM, Herbert PN, Kantor MA, Thompson PD.
Training, diet, and physical characteristics of distance runners with
tow or high concentrations of high density lipoprotein cholesterol.
Atherosclerosis. 1984;53:273-281.
6. Thomas TR, Adeniran PW, lltis PW, Aquiar CA, Albers JJ. Effects
of interval and continuous running on HDL-cholesterol, apoproteins A-l and B, and LCAT. Can J Appl Sport Sd. 1985;10:
52-59.
7. Frey I, Baumstark MW, Berg A, Keul J. Influence of acute
maximal exercise on lecithin: cholesterol acyltransferase activity in
healthy adults of differing aerobic performance. EurJAppI PhysioL
1991;62:31-35.
8. Barter PJ, Hopkins GJ, Ha YC. The role of lipid transfer proteins
in plasma lipoprotein metabolism. Am Heart J. 1987;113:538-542.
9. Brown ML, Inazu A, Hesler CB, Agellon LB, Mann C, Whitlock
ME, Marcel YL, Milne RW, Koizumi J, Mabuchi H, Takeda R,
Tall AR. Molecular basis of lipid transfer protein deficiency in a
family with increased high-density lipoproteins. Nature. 1989^342:
448-451.
10. Inazu A, Brown ML, Hesler CB, Agellon LB, Koizumi J, Takata K,
Maruhama Y, Mabuchi H, Tall AR. Increased high-density lipoprotein levels caused by a common cholesterol ester transfer
protein gene mutation. NEnglJMed. 1990;323:1234-1238.
11. Drayna D, Jarnagin AS, McLean J, Henzel W, Kohr W, Fielding C,
Lawn R. Cloning and sequencing of human cholesteryl ester
transfer protein cDNA. Nature. 1987;327:632-634.
12. Jiang XC, Moulin P, Quinet E, Goldberg IJ, Yacoub LK, Agellon
LB, Compton D, Schnitzer-Polokoff R, Tall AR. Mammalian
adipose tissue and muscle are major sources of lipid transfer
protein mRNA. J Biol Chan. 1991;266:4631-4639.
13. American Diabetes Association. Physician's Guide to Non-InsulinDependent (Type II) Diabetes: Diagnosis and Treatment. 2nd ed.
Alexandria, Va: American Diabetes Association, 1988.
14. American College of Sports Medicine. Guidelines for Graded
Exercise Testing and Exercise Prescription. 3rd ed. Philadelphia, Pa:
Lea & Febiger, 1986.
15. Lipid Research Clinics Program. Manual of Laboratory Operations.
Washington, DC: US Government Printing Office, DHEW publication No. NIH 75-628, 1974.
16. Warnick GR, Albers JJ. A comprehensive evaluation of the
heparin-manganese precipitation procedure for estimating high
density lipoprotein cholesterol. J Lipid Res. 1978;19:65-76.
17. Gidez LI, Miller GJ, Burstein M, Slagle S, Eder HA. Separation
and quantitation of subclasses of human plasma high density lipoproteins by a simple precipitation procedure. J Lipid Res. 1982^24:
1206-1224.
18. Marcel YL, McPherson R, Hogue M, Czarnecka H, Zawadzki Z,
Weech PK, Whitlock ME, Tall AR, Milne RW. Distribution and
concentration of cholesterol ester transfer protein in plasma of
normolipidemic subjects. / Clin Invest 1990;85:10-17.
19. Ordovas JM, Peterson JP, Santaniello P, Cohn JS, Wilson PWF,
Schaefer EJ. Enzyme-linked immunosorbent assay for human
plasma apolipoprotein B. J Lipid Res. 1987^28:1216-1224.
1367
20. Schaefer EJ, Ordovas JM. Metabolism of apolipoproteins A-I,
A-II, and A-IV. Methods EnzymoL 1986;129:420-443.
21. Labeur C, Michiels G, Bury J, Usher DC, Rosseneu M. Lipoprotein(a) quantified by an enzyme-linked immunosorbent assay with
monoclonal antibodies. Clin Chem. 1989^35:1380-1384.
22. Mankowitz K, Seip R, Semenkovich CF, Schonfeld G. Short term
interruption of training affects both fasting and post-prandial
lipemia. Atherosclerosis. 1992;95:181-189.
23. Iverius P-H, Ostlund-Linquist A-M. Preparation, characterization,
and measurement of lipoprotein lipase. Methods EnzymoL 1986;
129:691-704.
24. Kohrt WM, Malley MT, Coggan AR, Spina RJ, Ogawa T, Ehsani
AA, Bourey RE, Martin WH III, Holloszy JO. Effects of gender,
age, and fitness level on response of VO^^ to training in 60-71 yr
olds. J Appl PhysioL 1991;71:2OO4-2011.
25. Kohrt WM, Malley MT, Dalsky GP, Holloszy JO. Body composition of healthy sedentary and trained, young and older men and
women. Med Sci Sports Exerc. 1992;24:832-837.
26. Wilmore JH. A simplified technique for determination of residual
lung volumes. J Appl PhysioL 1969;27:96-100.
27. Brozek J, Grande F, Anderson JT, Keys A. Densitometric analysis
of body composition: revision of some quantitative assumptions.
Ann N YAcad Sci. 1963;110:113-140.
28. Kohrt WM, Obert KA, Holloszy JO. Exercise training improves fat
distribution patterns in 60- to 70-year-old men and women.
J GerontoL 1992;47:99-105.
29. Morgan CR, Lazarow A. Immunoassay of insulin: two antibody
system. Plasma insulin levels in normal, subdiabetic, and diabetic
rats. Diabetes. 1963;12:115-126.
30. Oscai LB, Williams BT, Hertig BA. Effect of exercise on blood
volume. J Appl PhysioL 1968;24:622-624.
31. Thompson PD, Cullinane EM, Sady SP, Ftynn MM, Bernier DN,
Kantor MA, Saritelli AL, Herbert PN. Modest changes in highdensity lipoprotein concentration and metabolism with exercise
training. Circulation. 1988;78:25-34.
32. Quinet EM, Agellion LB, Kroon PA, Marcel YL, Lee Y-C,
Whitlock ME, Tall AR. Atherogenic diet increases cholesteryl
ester transfer protein messenger RNA levels in rabbit liver. J Clin
Invest. 1990;85:357-366.
33. Sady S, Cullinane EM, Saritelli A, Bernier D, Thompson PD.
Elevated high-density lipoprotein cholesterol in endurance
athletes is related to enhanced plasma triglyceride clearance.
Metabolism. 1988^7:568-572.
34. Sady S, Thompson PD, Cullinane EM, Kantor MA, Domagala E,
Herbert PN. Prolonged exercise augments plasma triglyceride
clearance. JAMA. 1986;256:2552-2555.
35. Patsch JP, Prasad S, Gotto AM, Patsch W. High density lipoproteinj: relationship of the plasma levels of this lipoprotein
species to its composition, to the magnitude of postprandial
lipemia and to the activities of lipoprotein lipase and hepatic
lipase. J Clin Invest. 1987;80:341-347.
36. Peltonen P, Marniemi J, Hietanen E, Vuori I, Ehnholm C.
Changes in serum serum lipids, lipoproteins, and heparin
releasable lipoh/tic enzymes during moderate physical training in
man: a longitudinal study. Metabolism. 1981;30:518-526.
37. Despres J-P, Ferland M, Moorjani S, Nadeau A, Tremblay A,
Lupien PJ, Theriault G, Bouchard C. Role of hepatic-trigh/ceride
lipase activity in the association between intra-abdominal fat and
plasma HDL cholesterol in obese women. Arteriosclerosis. 1989;9:
485-492.
38. Kuusi T, Ehnholm C, Viikari J, Harkonen R, Vartiainen E, Puska
P, Taskinen M-R. Postheparin plasma lipoprotein and hepatic
lipase are determinants of hypo- and hyperalphalipoproteinemia.
J Lipid Res. 1989;30:1117-1126.
39. Thompson PD, Cullinane EM, Eshelman R, Sady S, Herbert PN.
The effects of caloric restriction or exercise cessation on the serum
lipid and lipoprotein concentrations of endurance athletes.
Metabolism. 1984;33:943-950.
Downloaded from http://atvb.ahajournals.org/ by guest on July 31, 2017
Exercise training decreases plasma cholesteryl ester transfer protein.
R L Seip, P Moulin, T Cocke, A Tall, W M Kohrt, K Mankowitz, C F Semenkovich, R Ostlund
and G Schonfeld
Arterioscler Thromb Vasc Biol. 1993;13:1359-1367
doi: 10.1161/01.ATV.13.9.1359
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272 Greenville
Avenue, Dallas, TX 75231
Copyright © 1993 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/13/9/1359
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/