of aerobic training and gender on HDL-C and LDL-C subfractions in Yucatan
WILLIAM B. KIST, TOM R. THOMAS,
KRISTEN E. HORNER, and
M. HAROLD LAUGHLIN
Departments of Food Science
and Human Nutrition and Veterinary Biomedical Sciences, University of Missouri,
Columbia MO 65211
WILLIAM B. KIST, TOM R. THOMAS,
KRISTEN E. HORNER, andM. HAROLD LAUGHLIN. Effects
of aerobic training and gender on HDL-C and LDL-C subfractions in Yucatan
minature swine. JEPonline
2 No 2 1999. The purpose of this study
was to determine if either aerobic training or gender influence HDL-C and
LDL-C subfractions in miniature swine. Thirty-five Yucatan miniature
swine were randomly assigned to either a sedentary group (n = 19, 8 males,
11 females) or a training group (n = 16, 7 males, 9 females). Swine
progressed in training to one hour a day, 5/days week, for 16 weeks.
Swine were fed Purina mini pig chow throughout the study. At the
conclusion of training, deltoid citrate synthase (CS), heart weight to
body weight (HW/BW), plasma triglyceride (TG), total cholesterol (TC),
HDL-C, HDL2-C, HDL3-C,
LDL-C, LDL1-C, LDL2-C,
and LDL3-C were assessed.
ANOVA (training x gender) with post hoc Tukey tests indicated that trained
swine had significantly elevated CS and HW/BW values (p < 0.05).
Training had no significant effect on either HDL-C or LDL-C subfractions.
In contrast, there were significant gender differences with females exhibiting
greater post-exercise values on TC (69.9 > 49.0 mg . dl-1), LDL-C (27.4
> 17.5), LDL1-C (2.3
> 0.7), LDL2-C (11.6
> 6.3), LDL3-C (13.5
> 10.4), and HDL-C (36.2 > 31.4). Results suggest that 16 weeks of
aerobic training was not effective in altering the lipoprotein subfractions
of miniature swine. Female miniature swine demonstrated higher lipoprotein
values than males, regardless of training status.
KEY WORDS: PIGS, ENDURANCE,
EXERCISE, LIPOPROTEIN PROFILE, CHOLESTEROL
Although the role of plasma triglycerides
(TG) cardiovascular disease is not clearly defined, the relationship between
plasma total cholesterol (TC) and cardiovascular disease is more completely
characterized (1,2). Specifically,
atherosclerosis is thought to be related to decreased high density lipoprotein
cholesterol (HDL-C), decreased HDL2-C,
and elevated low density lipoprotein cholesterol (LDL-C) subfractions (3,4).
is considered to be important in reverse cholesterol transport, with higher
levels thought to be protective against the development of atherosclerosis
(5). Conversely, LDL-C, the major carrier
of cholesterol, is positively correlated with atherosclerosis (6).
When blood concentrations of the small dense LDL subfraction, LDL3-C,
is elevated it has been demonstrated that there is a three fold greater
risk for atherosclerosis (3, 6).
Humans with low HDL2-C and elevated LDL3-C
are thought to be at risk for atherosclerosis. Humans who exercise
train aerobically generally demonstrate an increase in HDL2-C
and a decrease in LDL-C and experience associated increased longevity (7,8,9).
Swine generally demonstrate human-like
lipoprotein profiles, aerobic capacities and cardiac responses to training,
and are therefore considered a good animal model for the study of atherosclerosis
Despite the high degree of similarity in cardiovascular function between
humans and swine, research of how exercise training influences lipoprotein
parameters in swine have yielded inconsistent findings. Neither Pedersoli
(TC, TG) (13) nor Van Oort et al. (TC, HDL-C, LDL-C,
TG) (14) were able to demonstrate significant lipoprotein
differences among swine with training. Likewise, Link et al. (15)
exercise trained swine and did not see significant training effects on
TC or TG. In contrast, Forsythe et al. (16) did
observe significant percentage changes in TC and HDL-C with exercise training,
while Stucchi et al. (17) demonstrated absolute increases
in HDL2-C. Differences in exercise
training intensity, duration, and/or frequency could account for discrepancies
among these studies.
In contrast, gender effects upon some swine
lipoprotein parameters were found to be consistent in two studies (13,15).
Both Link et al. (15) and Pedersoli (13)
noted that female miniature swine demonstrated higher lipoprotein values
(TG, TC) than males. Neither study investigated the effects of gender
on HDL-C and LDL-C subfractions. It is noteworthy that Forsythe et
al. (16) used castrated male swine while Stucchi et
al. (17) used female swine in their exercise training
studies. It is plausible that swine with low levels of testosterone
demonstrate a better response to training. This effect has not been
investigated. Understanding gender effects may aid in interpreting
training effects in the swine model of atherosclerosis. The combined
effects of gender and training on HDL-C and LDL-C subfractions in swine
have not been thoroughly investigated and need to be characterized (10).
Therefore the purpose of this study was to determine if either aerobic
training or gender influence HDL-C and LDL-C subfractions in miniature
Animals and training protocol.
Yucatan miniature swine (n = 35) (Charles
Rivers Laboratories) were used to test the responses of HDL-C and LDL-C
subfractions to training. The swine were approximately one year of
age (12-14 months) when the study began and were randomly divided into
sedentary (S) (n = 19, 8 males, 11 females) and training (T) (n = 16, 7
males, 9 females) groups. Thus, a total of 15 male and 20 female
pigs were utilized in this investigation. This study was approved
and conducted in conformance with the University of Missouri’s animal care
committee’s guidelines on animal research.
The aerobic endurance training period lasted
16 weeks with swine training five consecutive days a week on a treadmill.
The training protocol began at approximately 30 minutes (week 1) progressing
to 60 minutes at 5 m/hr (week 16). A 10% grade was used throughout
the training program. At the conclusion of the training program swine
exceeded 20 miles/week, total mileage. S swine were not trained.
The training program has been previously detailed (18).
All swine received a typical pig diet (Purina, mini pig chow, PMI Feeds
Inc., St Louis, MO 63144) throughout the study. The diet consisted
of 16% protein, 2.5% fat, 14% fiber, 8% ash, 3% minerals and the balance
nitrogen free extract (3.89 kilocalories/g). Food intake was adjusted
to prevent a weight loss or gain. Although individual food consumption
was not recorded, the trained animals probably received a greater quantity
of pig chow to offset increased energy expenditure. Water was given
Markers of training and termination
At the conclusion of the training period,
a final exercise stress test was performed (18). “Stage
3” heart rate (HR) and “Stage 4” endurance duration were recorded.
Animals were subsequently sacrificed via removal of the heart after administration
of intravenous ketamine (1ml/2.85 kg), rompun (1ml/44 kg), and thiopental
(1ml/2.5 kg). Citrate synthase (CS) from the deltoid muscle (um/min/g
tissue) and wet heart weight (HW) to live body weight (BW) ratio (HW/BW)
were obtained to document physiological and anatomical evidence of training
effects (19, 20).
HDL-C and LDL-C subfractions.
At the conclusion of training, jugular
blood samples for lipoprotein assay were obtained via venipuncture and
collected in vacutainers containing EDTA. Blood samples were obtained
following a 12 hour fast. Samples from females were drawn without
regard for estrous cycle variations. Plasma for lipoprotein measurement
was separated by centrifugation (Beckman TJ-6R centrifuge, Palo Alto, CA)
at 4oC for 15 minutes at 3750 rev/m.
All plasma was stored at -70oC until analyzed.
TG was assayed spectrophotometrically (Beckman
model DU-20, Fullerton, CA 92634) using a Sigma Diagnostic kit (Triglyceride
GPO-Trinder #339, St. Louis, MO 63145). HDL fractions were assayed
for cholesterol content using a Sigma Diagnostic kit (Cholesterol, # 352).
HDL-C was determined following a heparin manganese precipitation process
to remove very low density lipoprotein cholesterol (VLDL-C) and LDL-C from
the plasma (21). This was followed by precipitation
of HDL2-C using dextran sulfate to determine
HDL3-C (21). HDL2-C
was deduced as HDL-C minus HDL3-C (HDL2-C
= HDL-C - HDL3-C). LDL subfractionation
was performed by separating plasma LDL into three levels using density
gradient ultracentrifugation (160,000g for 21 hours) with the LDL subfractions
individually analyzed for cholesterol content (12).
Statistical analysis was performed (Sigma
Stat, Jandel Scientific, San Rafael, CA) using a two way ANOVA (training
x gender). Post hoc Tukey tests were performed when significant
F values (p < 0.05) were noted. Values are reported as means
± standard deviations (SD).
Anatomical and physiological effects
CS, HW/BW, stage 3 HR, and stage 4 endurance
time means are reported in Table 1. CS means were significantly elevated
in the T animals. Likewise, HW/BW ratios were significantly elevated
in the T animals. Similarly, significant training effects were demonstrated
by lowered stage 3 HR and prolonged stage 4 endurance times.
1. Markers of training
Stage 3 HR
15.57 ± 3.37
4.51 ± 0.58
271.8 ± 15
20.77 ± 3.45
20.03 ± 3.59*
5.57 ± 0.86*
251.0 ± 33*
30.15 ± 3.66*
CS = citrate synthase (deltoid),
HW/BW = heart weight/body weight ratio (g/kg),
Stage 3 HR = Submaximal
heart rate in stage 3 of stress protocol,, Endurance time = exercise time,stage
4 of stress protocol. Values are means ± SD. *
Indicates significant difference, trained vs. sedentary (p
HDL and LDL subfractions, effects
of training and gender.
TG and LDL-C subfraction means and standard
deviations are reported in Table 2. There was no significant training
effect for any parameter. There was, however, a significant interaction
on TG for both males and females. The interaction was inconsistent,
as S males had lower TG values than T males, while S females had higher
TG than T females. TG and LDL-C pooled means (training status and
gender) are reported in
Table 3. There was
a significant main effect of gender on TG, TC, LDL-C, LDL1-C,
LDL2-C, and LDL3-C
with females consistently demonstrating greater values.
2. LDL-C subfractions
17.3 ± 8.5a
28.0 ± 8.4bc
33.0 ± 16.5b
26.4 ± 8.7c
47.0 ± 13.9
51.7 ± 8.2
67.0 ± 9.9
73.5 ± 9.2
16.0 ± 3.3
19.2 ± 3.5
26.4 ± 6.6
28.5 ± 7.7
0.7 ± 0.4
0.7 ± 0.4
2.3 ± 2.7
2.3 ± 1.9
6.1 ± 1.8
6.6 ± 1.3
10.4 ± 4.3
12.8 ± 4.8
11.7 ± 2.5
13.7 ± 3.0
13.3 ± 3.4
TG = triglycerides, TC =
total cholesterol, LDL-C = low density lipoprotein cholesterol.
Values are means ±
SD. Values with different superscripts are statistically different
a vs. trained males,
b vs. trained females, c vs. sedentary females. (p < 0.05).
3. LDL-C subfractions pooled by group and gender
26.4 ± 15.6
27.0 ± 8.3
21.8 ± 9.8
30.0 ± 13.7
58.6 ± 15.2
64.8 ± 14.0
49.0 ± 11.6
69.9 ± 9.9*
21.5 ± 7.4
24.4 ± 7.7
17.5 ± 3.7
27.4 ± 7.0*
1.6 ± 2.0
1.6 ± 1.7
0.7 ± 0.4
2.3 ± 2.2*
8.4 ± 3.9
10.1 ± 4.8
6.3 ± 1.6
11.6 ± 4.6*
11.6 ± 3.4
12.6 ± 3.0
10.4 ± 2.4
13.5 ± 3.1*
TG = triglycerides, TC = total cholesterol, LDL-C = low
density lipoprotein cholesterol.
Values are means ±
SD. * Indicates significant difference by gender (p < 0.05).
HDL-C subfraction means and standard deviations
are illustrated in
Figure 1. There was no significant
training effect on any subfraction. HDL-C pooled subfraction means
are illustrated in Figure 2. There was a significant gender effect
on HDL-C with females exhibiting greater values (females = 36.2 ±
5.8 mg/dL, males = 31.4 ± 8.1).
1. HDL-C subfractions by group and gender
HDL-C = high density cholesterol.
Values are means (mg/dL) ± SD.
There were no significant
differences by training (p<0.05).
2. HDL-C subfractions pooled by group and gender
HDL-C = high density cholesterol.
Values are means (mg/dL) ± SD.
* significant differences
by gender (p<0.05).
Data of statistical power (1-B)
for specific statistical mean comparisons are reported in Table 4.
Power greater than or equal to 0.80 was considered adequate. Power
analysis for gender demonstrated that Stage 4 endurance, TC, LDL-C, LDL2-C,
and LDL3-C were adequate. Power
analysis for training effects demonstrated that HW/BW, CS, Stage 3 HR,
and Stage 4 endurance were adequate. For the interaction of gender
and training, power was inadequate for all parameters.
4. Statistical power data for key variables
STAGE 3 HR
HW/BW = heart weight to body weight ratio, CS = citrate
synthase, STAGE 3 HR = heart rate during third stage of exercise test,
STAGE 4 endurance = duration of last stage of exercise test, TG = triglycerides,
TC = total cholesterol, LDL-C = low density lipoprotein cholesterol,HDL-C
= high density cholesterol. Power values calculated for two way ANOVA
(gender x training) with
p < 0.05.
Training effects on anatomical and
Trained swine in this study exhibited
a significantly higher HW/BW, lower stage 3 HR, longer stage 4 endurance,
and higher CS (Table 1). The HW/BW is a relative
anatomical indicator of heart size that was used to reveal cardiac muscle
hypertrophy in response to training (14, 16,18).
Hypertrophy was conclusively demonstrated as T swine had significantly
heavier hearts relative to body size. Stage 3 HR values demonstrated
that T animals had significantly lower HR at this submaximal exercise intensity
compared to the S group. Consistent with this, the T group was able
to continue significantly longer during Stage 4 of the exercise test.
Both a lower submaximal HR and greater exercise duration would be considered
positive markers of a training effect (14, 19).
Consistent with these parameters was the significant physiological increase
in muscle CS in T swine (Table 1). Increased
CS reflects increased oxidative capacity of a trained muscle (19,20).
Thus, in the T swine the efficacy of the aerobic training was both anatomically
and physiologically demonstrated.
Training effects on HDL-C and LDL-C
Despite the obvious aerobic training effect,
trained swine did not demonstrate a significant training effect on lipoprotein
subfractions (Tables 2, 3,
and Figures 1, 2). These findings were consistent with those of
Pedersoli (13), Van Oort et al. (14),
and Link et al. (15). In contrast, Forsythe et
al. (16) demonstrated a 16% increase in HDL-C, a 12%
decrease in LDL-C, and a lower TC in T animals. However, these differences
were based upon percentages and not absolute values. For example,
there was no difference in the absolute amount of plasma HDL-C found, but
the percentage of cholesterol carried in HDL was significantly greater
in the T than in the S swine. Thus, the results of the present study
are comparable to other swine studies in that major lipoprotein classes
are resistant to change by training (13,14,15).
The reason for the lack of lipoprotein
response in the present study was not ascertained. Intensity of the
exercise program seems an unlikely reason as evidenced by training induced
changes in physiological and anatomical markers (Table
1). Stucchi et al. (17) trained Yucatan swine
at 75% maximum HR using a similar protocol and had comparable CS findings
(Stucchi et al. S = 17.5, T = 22.5, present study S = 15.6 and T = 20.0
It therefore seems likely that the swine of the present study were trained
near 75% maximal HR.
In humans, training near 75% maximum HR
for 30 minutes would be considered adequate for lipoprotein changes to
occur. Superko (22) reviewed evidence from several
studies and deduced that 15 miles per week of aerobic activity is sufficient
to modify human lipoprotein profiles. The T swine in this study exceeded
20 miles/week. Thus, if swine and human lipoprotein profiles
behave similarly to an equivalent stimulus, the training intensity should
have been adequate in this study. The findings of the present study
and other studies may suggest that in swine lipoprotein changes may lag
behind other training indicators.
The length of the training period is a
more plausible explanation for the lack of demonstrable lipoprotein subfraction
changes in the present study. Neither Pedersoli (13)
nor Van Oort et al. (14) demonstrated significant training
effects, despite training periods of approximately eight months.
In humans, it may take a considerable period of time for HDL and LDL to
change significantly (8, 23).
Therefore, if Yucatan swine HDL and LDL subfractions behave similar to
human counterparts, a longer training period may be necessary to demonstrate
a significant training effect in HDL and LDL subfractions. This duration
hypothesis is supported by the results of Stucchi et al. (17).
Only after two years of endurance training was a significant decrease of
TG in LDL1, LDL2
and a four-fold increase in HDL2-C demonstrated.
Thus, sixteen weeks of training in the present study may be inadequate
to induce a significant change despite adequate intensity.
The frequency of training does not appear
to contribute to the negative training findings. Forsythe et al.
(16) trained swine 7 days/week, Van Oort et al. (14),
4 days/week, Link et al. (15), 5 days/week, and Pedersoli (13),
5 days/week, and all showed an absolute lack of lipoprotein changes with
training. Consistent with this, the present study used 5 days/week
of training and no significant lipoprotein changes were noted.
The diet used in the present study could
contribute to the lack of demonstrable lipoprotein changes. In the
present study TC, HDL-C, LDL-C, and TG plasma values were normolipidemic
(Tables 2, 3, and Figures
1, 2) and were comparable, although generally lower, to Pedersoli (13)
(pre-atherogenic diet trial) and Van Oort et al. (14)
(control diet, S and T) values. It is reasonable to assume that the
normal values of the present study were directly related to the amount
of fat (2.5%) ingested. Thus, without extreme training protocols,
subtle lipoprotein effects may not have been demonstrated due to the initial
low normal values. It is plausible to suggest that if the swine of
the present study had atherogenic lipoprotein values, a lipoprotein training
effect may have been demonstrated. A training study using an atherogenic
diet (high fat/cholesterol) is in progress.
Finally, the statistical power of the present
study may have been inadequate to detect subtle training differences (Table
4) in lipoprotein parameters. The crux of the problem for
some parameters appears to be large inter-animal variability. For
example, in the present study, the range of HDL2-C
values of the S animals was 0.1 to 6.9 mg/dl. The range in the T
animals was 0.1 to 9.5. With such large variances (Figures 1 and
2) within the groups, detection of differences between the groups is difficult.
Related to this, in the negative lipoprotein
training studies of Pedersoli (13), Van Oort et al.
(14), Forsythe et al. (16), and Link
et al. (15), all employed a between subjects design.
Only Stucchi et al. (17) demonstrated significant training
differences (HDL2-C) using a between subjects
design. It is noteworthy that Stucchi et al. (17)
used only nine (n = 9) animals (5 control and 4 training). Considering
the large HDL2-C inter-animal variance
documented in the present study, and the almost nonexistent levels of HDL2-C
in some Yucatan swine, it is plausible that the results from Stucchi et
al. (17) may have been statistically insignificant with
a larger population of animals.
Gender effects on HDL and LDL subfractions.
There was a significant effect of gender
on TC, LDL-C, LDL1-C, LDL2-C,
LDL3-C (Table 3)
and HDL-C (Figure 2). The gender effects demonstrated that females
had greater values than males in both S and T animals. This gender
effect was consistent with Pedersoli (13) (TC, TG) and
Link et al. (15) (TC, TG). Thus, it appears that
female swine generally have elevated lipoprotein profiles compared to their
male counterparts. This contrasts with human premenopausal females
who typically have healthier lipoprotein profiles than males. In
human females, HDL-C subfractions are higher and LDL-C subfractions are
lower, as demonstrated by Ziogas et al. (24). Thus,
miniature swine simply may not parallel human HDL-C and LDL-C gender subfraction
The only parameter which demonstrated a
significant interaction between gender and training was TG (Table
2). This finding, however, was inconsistent. S female swine
had greater TG values than T females, as would be expected based upon human
studies (24). Interestingly, Merservey et al. (25)
noted that the food intake of T female Yucatan swine was slightly less
than S females. A lower food intake could impact TG values.
In contrast, S males had lower TG levels than T males. Pedersoli
(13) also found that S male miniature swine had lower
TG values than T males. Pedersoli attributed this finding to
the fact that the S males ate less food than T males and therefore had
lower TG values. A lower TG value in S swine conflicts with the typical
human TG response where S individuals have higher TG values than T individuals
(24). Similar TG responses in the present study
could be attributed to food intake, however, food intake was not recorded
and this suspicion cannot be confirmed. Thus, overall, HDL-C and LDL-C
subfractions responses to training does not appear to be gender specific.
In conclusion, these results indicate
that a 16 week aerobic training program was not effective in altering HDL-C
and LDL-C subfractions in either male or female miniature swine fed a normal
diet. The results also suggest that human and swine lipoprotein profiles
may respond differently to training. However, the limited duration
of the training program in this study may have contributed to the lack
of change in blood lipoproteins. Additionally, our results indicate
that female miniature swine generally have higher HDL-C and LDL-C subfractions
than males, regardless of training status.
K, Pinson J, Weart C. The triglyceride connection in atherosclerosis.
Ann Pharmacother 1992:26:1109-17.
N. Diet, lipids, and coronary heart disease. Endocrinol
and Metab Clin of N. America. 1990:19:321-44.
M, Breslow J, Hennekens J, Buring J, Willett W, Krauss R. Low
density lipoprotein patterns and risk of myocardial infarction. JAMA
H. Lipoprotein metabolism and it=s relationship to atherosclerosis.
Med Clin N Amer 1994:78:1-20.
Y, Glangeaud M, Fainaru M, Stein O. The removal of cholesterol from
aortic smooth muscle cells in culture and Landschutz ascites cells by human
high density apolipoproteins. Biochim Biophys Acta 1975:380:106-18.
C, Fortmann S, and Krauss R. Association of small low density lipoprotein
particles with the incidence of coronary artery disease in men and women.
T, Robertson R, Goss F, Metz K, Laporte R. Effect of repeated exercise
bouts on high density lipoprotein cholesterol and its subfractions HDL2-C
and HDL3-C. Int
J Sports Med 1993:14:196-201.
L, Elliott D. The effect of exercise on lipid metabolism in men and women.
Sports Med 1987:4:307-21.
R. Contributions of epidemiology to exercise science and cardiovascular
health. Med Sci Sports Exerc 1988:20:426-38.
G. The swine as a model for studying exercised induced changes in
lipid metabolism. Med Sci Sports Exerc 1997:29:1437-43.
C, Lakas C, Zambraskie E. Oxygen consumption in treadmill exercised
Yucatan miniature swine. Med Sci Sports Exerc 1982:14:467-70.
D, Demacker N. Comparative studies on the low density lipoprotein
subfractions from pig and man. Comp Biochem Physiol 1988:90:297-300.
W. Physical exercise, atherogenic diet, and serum lipids in swine.
Curr Ther Res 1978:23:464-73.
Oort G, Gross D, Spiekerman A. Effects of eight weeks of physical
conditioning on atherosclerotic plaque in swine. Am J Vet Res
R, Pedersoli W, Safanie A. Effect of exercise on development of atherosclerosis
in swine. Ather
W, Miller E, Curry B, Bennink M. Aerobic exercise effects on lipoprotein
and tissue lipids in young pigs. Ather 1981:38:327-37.
A, Terpstra A, Foxall T, Nicolosi R, Smith S. The effect of exercise on
plasma lipids and LDL subclass metabolism in miniature swine. Med
Sci Sports Exerc 1991:23:552-61.
M, Overholster K, Bhatte M. Exercise training increases coronary
transport reserve in miniature swine. J Appl Physiol 1989:67:1140-9.
M, Hale C, Novela L, Gute D, N. Hamilton N, Ianuzzo C. Biochemical
characterization of exercise-trained porcine myocardium. J
Appl Physiol 1991:71:229-35.
P. Citrate synthase methods. Meth Enzymol 1969:13:3-11.
G, Nguyen T, Albers A. Comparison of improved precipitation methods
for quantification of high density lipoprotein cholesterol.
Clin Chem 1985:31:217-22.
H. Exercise training, serum lipids, and lipoprotein particles: Is
there a change threshold? Med Sci Sports Exerc 1991:23:677-85.
G, Janssen E, Hermans J, Wersh J. Effects of long term exercise
of moderate intensity on anthropometric values and serum lipids. Eur
J Clin Chem Clin Biochem 1995:33:121-6.
G, Thomas T, Harris W. Exercise training, postprandial hypertriglyceridemia,
and LDL subfraction distribution. Med Sci Sports Exerc
C, Carey G. Dietary fat saturation and endurance exercise alter lipolytic
sensitivity of adipocytes isolated from yucatan miniature swine.
J. Nutr 1994:124:2235-42.
for this study was provided by Food for the 21st Century (summer internship)
and by NIH Grant HL-52490 (M. H. Laughlin).
The authors have no professional
or financial interest in any products/equipment cited herein.
The results of the present study do not constitute endorsement of any product
by the authors.
For reprints: Dr.
Tom R. Thomas, 104 Rothwell Gym, University of Missouri, Columbia MO
Society of Exercise Physiologists
All Rights Reserved