European Journal of Applied Physiology

, 104:19

Response of lipid, lipoprotein-cholesterol, and electrophoretic characteristics of lipoproteins following a single bout of aerobic exercise in women

Authors

    • Institute for Women’s HealthTexas Woman’s University
    • Exercise Physiology Laboratory, Department of KinesiologyTexas Woman’s University
  • Kyle D. Biggerstaff
    • Exercise Physiology Laboratory, Department of KinesiologyTexas Woman’s University
  • Caroline Anderson
    • Exercise Physiology Laboratory, Department of KinesiologyTexas Woman’s University
Original Article

DOI: 10.1007/s00421-008-0770-2

Cite this article as:
Wooten, J.S., Biggerstaff, K.D. & Anderson, C. Eur J Appl Physiol (2008) 104: 19. doi:10.1007/s00421-008-0770-2

Abstract

The effects of a single session of moderate intensity (65% VO2max) aerobic exercise expending 500 kcal of energy on serum lipid and lipoprotein-cholesterol concentrations and the electrophoretic characteristics of low-density lipoprotein (LDL) and high-density lipoprotein (HDL) particles were determined in 11 sedentary, eumenorrheic, premenopausal women immediately prior to, and 24 and 48 h following exercise. Repeated measures analysis of variance revealed significant reductions in triglyceride (25.0%), HDL-cholesterol (10.9%), and HDL3-cholesterol (11.9%) concentrations at 48 h post-exercise. Despite these changes in lipid and lipoprotein-cholesterol concentrations, no significant changes were observed in peak LDL or HDL particle sizes or in the distribution of cholesterol within the LDL and HDL subfractions. Accordingly, it appears that a single session of moderate intensity aerobic exercise expending 500 kcal (2,092 kJ) of energy promotes reductions in triglyceride, HDL-C, and HDL3-C concentrations without concomitantly affecting the electrophoretic characteristics of LDL and HDL particles in this sample of women.

Keywords

Low-density lipoproteinHigh-density lipoproteinAcute exercisePeak lipoprotein size

Introduction

Coronary heart disease (CHD) represents the leading cause of death of women in the United States (American Heart Association 2005). One of the many major risk factors for CHD is dyslipidemia. Dyslipidemia is traditionally characterized as perturbations in blood lipid and lipoprotein-cholesterol concentrations (National Cholesterol Education Program 2001). One recently identified example of such a perturbation is the association between small, dense low-density lipoprotein (LDL) and increased CHD risk (Gardner et al. 1996; St-Pierre et al. 2001). It has been observed that the particle size and density of LDL may vary at a given LDL-cholesterol (LDL-C) concentration (Krauss and Burke 1982). In addition, when the majority of the LDL-C is distributed among the small dense LDL particles, an increased risk for CHD has been reported (Gardner et al. 1996; St-Pierre et al. 2001).

Physical activity and exercise training have been reported to reduce triglyceride and increase HDL-cholesterol (HDL-C) concentrations (Durstine et al. 2001). These modifications in blood lipid and lipoprotein-cholesterol concentrations may be related to changes in lipoprotein size and compositional characteristics. Exercise training may increase the peak LDL and high-density lipoprotein (HDL) particle size and promote a shift in the distribution of cholesterol carried by LDL from a smaller, denser particle (≤25.5 nm) to a larger (>25.5 nm), more cholesterol-rich LDL particle (Beard et al. 1996; Kraus et al. 2002). Thus, physical activity may serve as a potential non-pharmacological treatment for these forms of dyslipidemia.

Reductions in triglyceride and increases in HDL-C concentrations 48 h post-exercise have been reported in men who are hypercholesterolemic and normocholesterolemic following moderate intensity exercise expending 350 and 500 kcal (1,464 and 2,092 kJ) (Crouse et al. 1995; Grandjean et al. 2000). Triglyceride and HDL-C concentrations in women who are sedentary performing moderate intensity exercise have not been demonstrated no change post-exercise (Imamura et al. 2000; Pronk et al. 1995). It is important to note that these women performed 350 kcal (1,464 kJ) or less, which may be below the threshold necessary to elicit a post-exercise response.

Lamon-Fava et al. (1989a) reported that physically active women have a larger LDL size when compared to sedentary controls. Investigations reporting the acute response of lipoprotein particle size have been limited to exercise that is exhaustive in nature and in participants that are very active or highly fit. Lamon-Fava et al. (1989b) characterized the acute changes in LDL size following a triathlon using polyacrylamide gradient gel electrophoresis (PAGE). None of the female competitors reported a change in LDL size; however, 7 of the 34 men did demonstrate an increase in particle size post-event. More recently, Yu et al. (1999) quantified the response of LDL and HDL size using proton nuclear magnetic resonance spectroscopy immediately post-triathlon and reported a significant increase (2.7%) in the average size of HDL. In addition, a 16% reduction in the concentration of HDL3 particles with a concomitant increase of 20% in the concentration of HDL2 particles was reported post-race. Despite no change in the average LDL size, there was a 38% reduction of small, dense LDL particle concentration. In contrast, Liu et al. (1999) reported no change in peak LDL particle size following a marathon in trained male and female runners.

The rational for this investigation is that quantifying the response of lipid, lipoprotein-cholesterol and electrophoretic characteristics of LDL and HDL particles may further enhance our understanding of the acute lipid and lipoprotein changes that precede the changes observed following regular aerobic exercise training. In addition, limitations of the previous investigations examining the response of lipoprotein size are that the exercise interventions (Lamon-Fava et al. 1989b; Yu et al. 1999; Liu et al. 1999) do not reflect current exercise prescription guidelines as stated by the American College of Sports Medicine (2006). In addition, blood lipid responses in these studies were limited to immediately post-exercise and were not measured 24 or 48 h post-exercise, as is commonly reported in the lipid and lipoprotein-cholesterol literature. Therefore, the purpose of this investigation was to examine the effects of a single session of moderate intensity (65% VO2max) aerobic exercise expending 500 kcal (2,092 kJ) of energy on serum lipid and lipoprotein-cholesterol concentrations and the electrophoretic characteristics of LDL and HDL particles in premenopausal women who are sedentary.

Materials and methods

Participants

Eleven premenopausal, eumenorrheic, non-smoking, sedentary women were recruited to participate in this investigation. Sedentary was defined as performing less than 20 min of exercise 2 days week−1 for the past 3 months. In addition, participants were not taking prescription medication for birth control, dyslipidemia, diabetes or hypertension for the previous 6 months. Each participant was informed of the risks associated with the study and signed the written informed consent as approved by Texas Woman’s University Institutional Review Board. Following completion of the written informed consent, a detailed medical history questionnaire was completed.

Anthropometrics and body composition

Body weight was measured to the nearest 0.1 kg with a calibrated digital scale (Tanita Corp., Arlington Heights, IL) and height to the nearest 1 mm with a stadiometer (Perspective Enterprises, Kalamazoo, MI). Body mass index was calculated from body weight in kilograms divided by height in meters squared (Garrow and Webster 1985).

Body composition was evaluated by skinfold measurement (Jackson et al. 1980) with a Lange skinfold caliper (VacuMed, Ventura, CA). Body density was calculated by using the summation of the average of the skinfolds from the triceps, abdomen, and thigh (Jackson et al. 1980). Percent body fat was determined from body density by the Brožek equation (Brozek 1966).

Determination of VO2max

Following the anthropometric and body composition measures, VO2max was determined by having each participant complete a graded exercise test on a Quinton Series S65 treadmill (Quinton Inc., Bothell, WA). The graded exercise test consisted of walking at 3.5 mph with stages of increasing intensity (increase of 3.0% grade per 2-min stage). Respiratory gases were analyzed by a ParvoMedics Truemax 2400 metabolic cart (Consentius Technologies, Sandy, UT) following autocalibration using gases of known concentrations.

During all phases of the VO2max test, heart rate (HR) and electrocardiograph recordings were monitored using a Quinton Q4500 ECG (Quinton Inc., Bothell, WA) with electrode lead placement performed according to the Mason-Likar modified 12-lead system for treadmill stress testing (Koppes et al. 1977). Maximal heart rate (HRmax) was defined as the highest heart rate recorded during the graded exercise test. Systolic and diastolic blood pressures were monitored by auscultation at prescribed 2-min intervals from rest throughout the exercise bout, and at 1-min intervals during recovery. The 6–20 Borg scale was used to obtain ratings of perceived exertion at 2-min intervals during exercise. The test was terminated when each participant could no longer maintain pace with the treadmill. Test results were considered satisfactory if the participant’s response met the following criteria of an exercise HR ≥ 85% of age-predicted HRmax (age-predicted HRmax = 220 − age) and respiratory exchange ratio (VCO2/VO2) ≥ 1.1 (American College of Sports Medicine 2006).

Exercise protocol

To provide some control for the influence of ovulation on serum lipoproteins, two consecutive menstrual cycles were tracked. Each participant recorded the dates that menses began and finished for two consecutive months to provide an estimate of when the third menstrual cycle would begin. Additionally, the average duration of participants’ menstrual cycle was 29.1 ± 2.1 days, which is suggestive of eumenorrhea. Exercise sessions were performed during the early to mid-follicular phase, 24 h following the end of menses, of each participant’s third menstrual cycle.

Participants performed treadmill exercise that consisted of walking at 3.5 mph on a motorized treadmill at a grade that elicited an exercise intensity of 65% VO2max. Respiratory gases were measured for the first 15-min of exercise and then for 5 min at 10-min intervals to estimate the duration of exercise required to expend 500 kcal (2,092 kJ) of energy. The rate of caloric expenditure (kcal min−1) was calculated by taking the product of VO2 (l min−1) and the caloric equivalent of the corresponding respiratory exchange ratio. All exercise was performed between 5:00 and 8:00 a.m. after a 10 h fast. During the exercise session, participants were allowed to consume water ad libitum. In addition, participants were given 20 ml of a carbohydrate–electrolyte drink every 15 min to reduce fatigue and prevent hypoglycemia during the exercise session.

Dietary records

Each participant was asked to record all dietary consumption for 2 days prior to the exercise session, the day of exercise, and the day following the exercise session. Due to the acute effect of alcohol on triglyceride and HDL-C concentrations, participants were requested to abstain from alcohol during the week preceding the exercise session (van der Gaag et al. 2000). Food records were analyzed using the Nutritionist Pro software (Axxya Systems, Stafford, TX) to determine the participant’s total caloric consumption and the percentage of total calories derived from fat, saturated fat, protein, and carbohydrate.

Blood collection

Fasting blood samples were collected by a standard venipuncture of an antecubital forearm vein. Blood was collected in serum-separator vacutainer (BD Diagnostics, Franklin Lakes, NJ) tubes (8 ml) immediately prior to, and 24 and 48 h following the exercise session. Using whole blood, hematocrit was measured and recorded using the microhematocrit technique. Participants fasted for at least 10 h prior to each blood collection session. Serum was separated from the blood by low speed centrifugation (1,500g, 15 min, 10°C), transferred into aliquots, and stored at −70°C until analysis.

Analytical methods

Hemoglobin was determined using the cyanmethemoglobin technique (H4390, Sigma, St Louis). Serum was assayed for total cholesterol, HDL-C, HDL-C subfractions, and triglyceride concentrations for descriptive characteristics. Total cholesterol was determined using a standard enzymatic technique (Kit#C507, Teco Diagnostics, Anaheim, CA) (Allain et al. 1974). Measurement of HDL-C concentration was conducted by the precipitation of apolipoprotein B containing lipoproteins followed by enzymatic measurement of the remaining cholesterol (Kit#H511, Teco Diagnostics, Anaheim, CA). A second precipitation technique was employed using dextran sulfate and MgCl2 to determine HDL3-C concentration (Gidez et al. 1982; Warnick et al. 1982). The concentration of HDL2-C was calculated as the difference between HDL-C and HDL3-C concentration. Triglyceride concentration was analyzed using an enzymatic technique (Kit#T532, Teco Diagnostics, Anaheim, CA) (Bucolo and David 1973; McGowan et al. 1983). The Friedewald equation was used to estimate the concentration of LDL-C (Friedewald et al. 1972). Plasma glucose concentration was measured using a glucose analyzer (YSI-2700; Yellow Springs Instruments, Yellow Springs, OH). All lipid and lipoprotein-cholesterol concentrations were corrected for post-exercise plasma volume changes using the Dill and Costill method (Dill and Costill 1974).

A nondenaturing 2–31% gradient polyacrylamide gel electrophoresis (PAGE) procedure was employed to determine peak LDL, HDL2b, HDL2a, HDL3a, and HDL3b particle sizes (Rainwater et al. 1992, 1997, 2004; Rainwater 1998). The peak lipoprotein particle size was defined as the predominant peak for LDL, HDL2b, HDL2a, HDL3a, and HDL3b. Unlike gels with a 2–16% gradient that will resolve several peaks for LDL, a 2–31% gradient will resolve only one predominant peak for LDL.

Gels were stained with 20% Sudan black B to identify the lipoprotein peak particle size and the distribution of cholesterol among the lipoprotein particles as described by Singh et al. (1995). Sudan black B binds neutral lipid in lipoproteins; however, Callais et al. (1987) concluded that the HDL-C quantitation with PAGE was unaffected by the triglyceride concentration. In addition, several studies have demonstrated that Sudan black B accurately reflects cholesterol distributions among size-resolved lipoproteins (Callais et al. 1987; Cheng et al. 1988; Gambert et al. 1988).

The relative distribution of cholesterol among LDL subfraction (LDL1 = 26.5–29.0 nm, LDL2 = 25.6–26.4 nm, and LDL3 = 24.3–25.5 nm) and HDL subfraction (HDL2b = 9.7–12.9 nm, HDL2a = 8.8–9.6 nm, HDL3a = 8.2–8.7 nm, and HDL3b = 7.8–8.1 nm) particles were determined by quantifying the area under each lipoprotein particle subfraction peak with the exception of LDL (Blanche et al. 1981; Rainwater et al. 1997). The areas for each LDL subfraction (i.e., LDL1, LDL2, and LDL3) were determined by dividing the area under the single LDL peak into three sections. The distribution of cholesterol under each lipoprotein subfraction peak was calculated as a percentage of the total area for LDL and HDL classes (e.g., % HDL-C in HDL2b = area of HDL2b/total HDL area).

The coefficient of variation was calculated as the percent difference between samples that were analyzed in duplicate. The coefficient of variation for the peak LDL (0.3%), HDL2b (0.5%), HDL2a (0.4%), HDL3a (0.4%), and HDL3b (0.5%) particle sizes were within acceptable ranges. The coefficient of variation for the distribution of LDL-cholesterol between LDL1, LDL2, and LDL3 was 6.2, 12.5, and 19.2%, respectively. The coefficient of variation for the distribution of HDL-C between HDL2b, HDL2a, HDL3a, and HDL3b was 14.2, 14.5, 14.5, and 12.2%, respectively. The coefficient of variation for lipid and lipoprotein-cholesterol concentrations was 2.7, 4.1 1.9, and 3.4% for HDL-C, HDL3-C, total cholesterol and triglyceride concentrations, respectively.

Statistical analysis

Data are displayed as mean ± standard deviation (SD). A repeated measures ANOVA was used to determine significant changes in lipid and lipoproteins over time. A Bonferonni post hoc procedure was performed to identify differences between the different blood sampling time points. All calculations were performed with SPSS v11.0 statistical software package (SPSS Inc., Chicago, IL). The criterion reference for statistical significance was set at P < 0.05.

Results

Descriptive characteristics of the participants are displayed in Table 1. The mean exercise VO2 was 1.27 l min−1, which corresponded to an exercise intensity of 62.0 ± 3.5% VO2peak. The mean exercise HR was 145.8 ± 12.2 bpm, which corresponded to an exercise intensity of 77.2 ± 5.6% HRmax and 74.4 ± 5.9% age-predicted HRmax. The mean exercise time for the participants was 80.6 ± 5.7 min. Dietary records were collected for 9 of the 11 participants. No day-to-day differences in the measured dietary variables were observed for the participants (Table 2).
Table 1

Descriptive characteristics of participants at baseline (n = 11)

 

Mean ± SD

Range

Age (years)

24.6 ± 7.9

19–45

Height (cm)

161.5 ± 5.9

150.0–170.0

Weight (kg)

65.8 ± 11.3

53.8–85.1

BMI (kg m−2)

25.3 ± 4.7

20.6–34.2

Body fat (%)

28.3 ± 6.3

19.9–39.1

VO2max

 (l min−1)

2.1 ± 0.2

1.8–2.5

 (ml kg−1 min−1)

32.1 ± 6.4

22.8–42.1

Total cholesterol

 (mg dl−1)

167.3 ± 25.3

116.9–195.5

 (mmol l−1)

4.32 ± 0.65

3.02–5.06

Triglyceride

 (mg dl−1)

103.2 ± 61.5

39.2–238.3

 (mmol l−1)

1.18 ± 0.70

0.45–2.72

HDL-C

 (mg dl−1)

51.3 ± 9.4

41.7–74.6

 (mmol l−1)

1.33 ± 0.24

1.08–1.93

HDL2-C

 (mg dl−1)

25.4 ± 8.7

16.5–47.3

 (mmol l−1)

0.66 ± 0.22

0.43–1.22

HDL3-C

 (mg dl−1)

25.8 ± 3.1

18.8–29.4

 (mmol l−1)

0.67 ± 0.08

0.49–0.76

LDL-C

 (mg dl−1)

95.4 ± 24.7

50.4–122.3

 (mmol l−1)

2.47 ± 0.64

1.30–3.16

Glucose

 (mg dl−1)

81.4 ± 11.0

51.5–92.5

 (mmol l−1)

4.52 ± 0.61

2.86–5.13

Table 2

Total energy intake and macronutrients from 4-day dietary record analyzes

 

Day −2

Day −1

Day 0

Day 1

Total

 (kcal d−1)

1,961.7 ± 988.4

1,542.7 ± 787.9

1,811.6 ± 1,226.3

1,513.6 ± 468.7

 (kJ)

8,207.8 ± 4,135.5

6,454.7 ± 3,296.6

7,579.7 ± 5,130.8

6,332.9 ± 1,961.0

Carbohydrates (%)

51.9 ± 6.4

48.7 ± 15.4

53.2 ± 13.0

49.3 ± 12.4

Protein (%)

15.6 ± 4.7

13.9 ± 7.7

17.3 ± 6.7

15.0 ± 4.1

Fat (%)

32.4 ± 6.7

36.3 ± 12.9

29.4 ± 10.4

35.8 ± 10.2

Saturated fat (%)

11.6 ± 4.5

12.1 ± 5.9

10.3 ± 5.0

11.7 ± 4.8

Data are mean ± SD

No significant changes were observed between days

Day −2 two days prior to exercise session

Day −1 day prior to exercise session

Day 0 day of exercise session

Day 1 one day after exercise session

Plasma volumes were observed to be elevated 4.8 and 6.5% at 24 and 48 h post-exercise, respectively. Analysis of variance revealed that these changes in plasma volume were not significant (P = 0.062); however, the elevated plasma volume may have physiological relevance. Therefore, all lipid and lipoprotein-cholesterol concentrations were corrected for shifts in post-exercise plasma volume. A significant change (P = 0.044) in triglyceride concentration was observed over time (Table 3). Post hoc analyses revealed that triglyceride concentration was significantly (P = 0.033) reduced 25.0% at 48 h post-exercise when compared to baseline. In addition, a significant change (P = 0.035) in HDL-C concentration was observed over time. Post hoc analyses revealed that HDL-C concentration was significantly (P = 0.024) reduced by 10.9% at 48 h post-exercise when compared to baseline. A significant change (P = 0.021) in HDL3-C concentration was observed over time. Post hoc analyses revealed that the HDL3-C concentration was significantly (P = 0.015) reduced 11.9% at 48 h post-exercise when compared to baseline. Interestingly, when lipid and lipoprotein-cholesterol concentrations were not corrected for shifts in plasma volume, the significant changes in triglyceride, HDL-C, and HDL3-C concentrations were no longer observed.
Table 3

Lipid and lipoprotein-cholesterol concentrations before and after exercise

 

IPre

+24 h

+48 h

Tg

 (mg dl−1)

103.2 ± 61.5

83.0 ± 41.5

77.4 ± 45.8*,†

 (mmol l−1)

1.18 ± 0.70

0.95 ± 0.47

0.88 ± 0.52*,†

TC

 (mg dl−1)

167.3 ± 25.3

164.5 ± 35.3

154.1 ± 33.8

 (mmol l−1)

4.32 ± 0.65

4.25 ± 0.91

3.98 ± 0.87

LDL-C

 (mg dl−1)

95.4 ± 24.7

98.3 ± 36.7

93.0 ± 27.6

 (mmol l−1)

2.47 ± 0.64

2.54 ± 0.95

2.40 ± 0.71

HDL-C

 (mg dl−1)

51.3 ± 9.4

49.6 ± 8.8

45.7 ± 12.9*,†

 (mmol l−1)

1.33 ± 0.24

1.28 ± 0.23

1.18 ± 0.33*,†

HDL2-C

 (mg dl−1)

25.4 ± 8.7

25.0 ± 8.2

22.9 ± 11.6

 (mmol l−1)

0.66 ± 0.22

0.65 ± 0.21

0.59 ± 0.30

HDL3-C

 (mg dl−1)

25.8 ± 3.1

24.6 ± 4.0

22.7 ± 4.1*,†

 (mmol l−1)

0.67 ± 0.08

0.64 ± 0.10

0.59 ± 0.11*,†

Data are mean ± SD

Tg triglyceride, TC total cholesterol, LDL-C low-density lipoprotein cholesterol, HDL-C high-density lipoprotein cholesterol, HDL2-C high-density lipoprotein-2 cholesterol, HDL3-C high-density lipoprotein-3 cholesterol, IPre immediately prior to exercise session, +24 h 24 h after exercise session, +48 h 48 h after exercise session

Significant effect (P < 0.05)

* Significantly (P < 0.05) different than IPre

Table 4 contains the mean ± SD values of the peak LDL and HDL particle sizes immediately prior to, and 24 and 48 h following exercise. Table 5 contains the relative distribution of cholesterol among LDL and HDL subfractions immediately prior to, and 24 and 48 h following exercise. No significant changes over time were observed for lipoprotein size or the distribution of cholesterol among LDL and HDL particles.
Table 4

Lipoprotein particle size before and after exercise

 

IPre

+24 h

+48 h

LDL (nm)

26.67 ± 0.45

26.69 ± 0.44

26.70 ± 0.46

HDL2b (nm)

10.91 ± 0.64

10.92 ± 0.63

10.89 ± 0.61

HDL2a (nm)

9.05 ± 0.21

9.07 ± 0.26

9.02 ± 0.21

HDL3a (nm)

8.69 ± 0.09

8.66 ± 0.10

8.63 ± 0.09

HDL3b (nm)

7.95 ± 0.06

7.95 ± 0.09

7.94 ± 0.10

Data are mean ± SD

IPre immediately prior to exercise session, +24 h 24 h after exercise session, +48 h 48 h after exercise session

Table 5

Distribution of cholesterol among lipoproteins before and after exercise

 

IPre

+24 h

+48 h

LDL1 (% LDL-C)

67.69 ± 17.42

67.17 ± 14.50

65.96 ± 14.56

LDL2 (% LDL-C)

26.16 ± 16.72

25.51 ± 13.66

27.37 ± 13.59

LDL3 (% LDL-C)

6.15 ± 2.54

7.31 ± 2.62

6.67 ± 3.23

HDL2b (% HDL-C)

45.31 ± 18.67

43.68 ± 14.91

44.25 ± 17.67

HDL2a (% HDL-C)

27.34 ± 9.94

26.82 ± 10.68

23.85 ± 12.63

HDL3a (% HDL-C)

17.87 ± 7.24

19.67 ± 6.54

19.22 ± 6.36

HDL3b (% HDL-C)

9.47 ± 4.53

9.80 ± 3.15

12.68 ± 10.98

Data are mean ± SD

IPre immediately prior to exercise session, +24 h 24 h after exercise session, +48 h 48 h after exercise session

Discussion

This investigation quantified the response of lipid and lipoprotein-cholesterol concentrations, and the electrophoretic characteristics of LDL and HDL particles following a single session of aerobic exercise in premenopausal women who are sedentary. These data suggest that a single aerobic exercise session promoted significant reductions in triglyceride, HDL-C, and HDL3-C concentrations without altering lipoprotein particle size or the distribution of cholesterol among lipoprotein subfractions.

In sedentary men, reductions of 14.7 and 11.9% in triglyceride concentration have been reported following exercise requiring 350 and 500 kcal (1,464 and 2,092 kJ) of energy expenditure, respectively (Crouse et al. 1995; Grandjean et al. 2000). It is important to note that few well-controlled studies have been conducted examining the acute aerobic exercise responses of lipid and lipoproteins in women. In general, these studies report that energy expenditures as great as 350 kcal (1,464 kJ) have promoted no change in triglyceride concentration at 24 and 48 h post-exercise (Imamura et al. 2000; Pronk et al. 1995). In contrast to these studies, the present data demonstrate a significant 25% reduction in triglyceride concentration at 48 h post-exercise. It is likely that this reduction in triglyceride concentration is due to a greater caloric expenditure during exercise than was used in previous investigations. Interestingly, the women in this study showed a greater percent reduction in triglyceride concentration than men (Grandjean et al. 2000) that also expended 500 kcal (2,092 kJ) of energy.

In previous studies examining the effects of 350 and 500 kcal (1,464 and 2,092 kJ) of energy expenditure in sedentary men, increased HDL-C and HDL3-C concentrations were reported at 48 h post-exercise (Crouse et al. 1995; Grandjean et al. 2000). Conversely, exercise requiring 350 kcal (1,464 kJ) of energy expenditure in sedentary, premenopausal women has promoted no change in HDL-C and HDL3-C concentrations 24 and 48 h post-exercise (Imamura et al. 2000; Pronk et al. 1995). In contrast to these previous studies, reductions in HDL-C and HDL3-C concentrations were evident in the present investigation. It is speculated that the reduction of HDL-C concentration was largely due to the concomitant reduction of HDL3-C. Data were examined for possible hyper- (exercise increased HDL-C and HDL3-C concentrations) or hypo- (exercise reduced HDL-C and HDL3-C concentrations) responders to better explain why mean HDL-C and HDL3-C concentrations were reduced post-exercise. All but two participants responded with a reduction in HDL-C and HDL3-C concentration at 48 h post-exercise.

In contrast to previous studies examining the effects of acute exercise on lipoprotein particle size; participants in this investigation were sedentary, accumulating less than 20 min of exercise per week. The exercise performed in this investigation was not exhausting; however, the total time of exercise performed was greater than the upper limit of 60 min for exercise duration recommended for cardiorespiratory fitness by the American College of Sports Medicine (2006). In agreement with previous investigations (Lamon-Fava et al. 1989b; Yu et al. 1999; Liu et al. 1999) examining the effects of a single session of aerobic exercise, no significant change in LDL particle size was observed in the present study 24 and 48 h post-exercise despite the fact that the participants performed more than the recommended dose of aerobic exercise. It is interesting that there were no significant changes in LDL particle size despite a significant 25% reduction in triglyceride concentration. Lamon-Fava et al. (1989b) reported that the male triathletes who demonstrated an increase in LDL particle size were characterized with the smallest LDL particle and highest triglyceride concentration pre-race. In addition, these male triathletes also had the greatest reduction in triglyceride concentration post-race when compared to the other triathletes. In the present study, the pre-exercise triglyceride concentration ranged greatly between participants, which may explain why there was no change in LDL particle size post-exercise. Similar to observations of Lamon-Fava et al. (1989b), the participant in the present study with the highest triglyceride concentration also had the smallest LDL peak particle size and demonstrated the greatest increase in LDL particle size when compared to the other participants. Interestingly, of the five participants that had an increase in LDL particle size (average 0.21 nm increase), there was an average decrease in triglyceride concentration of 38% at 48 h post-exercise, regardless of pre-exercise triglyceride concentration. These observations suggest that an increase in LDL particle size can occur regardless of pre-exercise triglyceride concentration; however, a reduction in triglyceride concentration may be necessary to promote an increase LDL particle size. The change in triglyceride concentration at 48 h post-exercise was unrelated to individual changes in the distribution of cholesterol between LDL subfractions and HDL electrophoretic characteristics.

The changes in HDL-C and HDL3-C concentration were unrelated to changes in HDL particle size. Interestingly, the participants (n = 4) that had the greatest reduction in HDL3-C concentration (average reduction of 22.0%) also had the greatest increase in the distribution of HDL-C in the HDL3a particles (average increase of 42.7%). This observation suggests that the reduction in HDL3-C concentration may be associated in part with a reduced number of lipid-poor HDL3 particles at 48 h post-exercise.

Since enzyme activities were not quantified in the present study, the following proposed mechanisms to explain the observed reductions in triglyceride, HDL-C and HDL3-C concentrations are speculative and warrant further investigation. The mechanism explaining the acute post-exercise reduction in triglyceride concentration likely is due to increased triglyceride hydrolysis by lipoprotein lipase (LPL). Exercise has been reported to elevate LPL activity for up to 48 h in premenopausal women (Gordon et al. 1996). The increased LPL activity post-exercise does not explain the reduction in HDL-C and HDL3-C. It has been suggested that increased hydrolysis of triglyceride from triglyceride-rich-lipoproteins by LPL promotes the transfer of cholesterol and other neutral lipids to HDL leading to an increase in HDL-C concentration. In this investigation, the reduction in HDL-C was due to a decrease in HDL3-C concentration, which may be due to an increased uptake of lipid poor HDL3 particles.

The exercise-induced reduction of HDL-C and HDL3-C concentrations at 48 h post-exercise may be explained by a two step process: (1) the increased enzymatic activities of cholesteryl ester transfer protein (CETP) and hepatic lipase (HL), followed by (2) removal of lipid poor HDL particles by either the kidney or by the LDL related receptor protein (LRP). Increased CETP activity promotes an increase in the exchange of cholesteryl ester from HDL for triglyceride from triglyceride-rich-lipoproteins. As HDL particles become triglyceride rich, the affinity of HL for the triglyceride-rich HDL particle increases, promoting the hydrolysis of triglyceride from HDL to form small HDL3 (HDL3b or HDL3c) particles, lipid-poor HDL particles that contain apo E, and lipid poor HDL particles containing apo A1 with pre-β mobility (pre-β HDL) (Huang et al. 1994; Rye and Barter 2004).

Since it is hypothesized that the reduction in the number of HDL3a particles explains the reduction in HDL-C and HDL3-C concentrations at 48 h post-exercise, the pre-β HDL particles containing apo A1 were unable to be remodeled into mature cholesterol-rich HDL particles. These particles may have been cleared by the kidneys. Alternatively, lipid-poor HDL particles that contain apo E may have been cleared by the LRP (Vassiliou and McPherson 2004). Following the hydrolysis of HDL particles, the remaining HDL3 particles may have been free of apo A1, which would prevent the HDL3 particle from maturing into a cholesterol-rich HDL particle. Since apo A1 is the cofactor for the activation of LCAT, HDL particles without apo AI would be unable to esterify free cholesterol and mature into a cholesterol-rich HDL2 particle.

Following several weeks of aerobic training, HDL-C concentrations are typically increased in both men and women (Durstine et al. 2001; Leon et al. 2000). It remains unknown if an acute reduction in HDL-C and HDL3-C concentrations is a normal progression leading to increased HDL-C concentrations in response to aerobic exercise training in women. This would be an interesting follow-up investigation to better understand how acute changes in HDL metabolism leads to the elevated HDL-C concentrations observed following aerobic exercise training in women. It is important to note that reverse cholesterol transport to the liver is not dictated by plasma HDL or apo A1 concentrations, but rather, by events that occur in the peripheral organs (Jolley et al. 1998). Despite the reductions in HDL-C and HDL3-C concentrations, based on observations of Jolley et al., the magnitude of net sterol movement through the reverse cholesterol pathway may have not changed or may have possibly increased or decreased post-exercise. Therefore, it is difficult to determine if the acute reductions in HDL-C concentrations had an impact on atherogenic risk.

In conclusion, a single session of aerobic exercise requiring 500 kcal (2,092 kJ) of energy expenditure did not affect LDL and HDL electrophoretic characteristics; however, exercise did promote significant changes in triglyceride, HDL-C, and HDL3-C concentrations. The results from this study did suggest that changes in the electrophoretic characteristics might be related to changes in lipid and lipoprotein-cholesterol concentrations. The need is great for well-designed investigations to examine the acute effects of aerobic exercise on lipid and lipoprotein metabolism in women. In addition, the quantification of lipid and lipoprotein related enzymes, as well as using nuclear magnetic resonance spectroscopy to measure changes in the number of LDL and HDL particles post-exercise is warranted. More importantly, there is a need to investigate the benefits of exercise in women who are at increased risk for CHD.

Copyright information

© Springer-Verlag 2008