Elevated plasma low density lipoprotein (LDL) or LDL cholesterol(LDL-C) concentration has been considered an independent risk factor for the development of atherosclerosis (9). However, recent research has confirmed the existence of LDL subfractions with major differences in lipid/protein composition, particle diameter, molecular weight, and density (17). At least three major LDL subfractions have been isolated by preparative ultracentrifugation (LDL1, LDL2, and LDL3) (26). LDL3 is the smallest and most dense LDL subfraction (12) containing higher percentages of cholesteryl ester and apoprotein B (apoB) and lower percentages of triglycerides (TG). Subjects with predominantly small LDL particles (i.e., LDL3) have been shown to have a three-fold increased risk for myocardial infarction independent of age, sex, and relative weight(2). The association is equally powerful in men and women(7).
It is well documented that exercise training reduces TC and LDL-C levels. However, few studies have examined the effect of chronic exercise training on LDL subfraction distribution. Results from previous studies suggest that the exercise training effect is mainly a result of the reduction in the plasma concentrations of the smaller, denser LDL3 particles(4,29). The limited literature on the association between exercise training and LDL subclasses has been focused on trained versus sedentary subjects, and no data are available on the influence of recreational type of exercise on the LDL subfraction profile.
A common finding in studies examining LDL subfraction distribution and its determinants is the role of plasma TG as the best predictor of LDL subfraction distribution (19). The rate of TG clearance after a meal varies even among normolipidemic individuals, and postprandial hypetriglyceridemia (PP-HTG) may persist up to 8 h or more depending primarily on the amount of fat in the meal as well as on the individual's ability to break down and clear TG (20). As a result, individuals who are believed to be normotriglyceridemic based on fasting TG analysis may actually be in hypertriglyceridemic state for many hours during the day following meals. Several groups of investigators have suggested that atherosclerosis may be related to postprandial metabolic changes in lipids and lipoproteins (8,22) and may relate to dense LDL formation (16). Thus, even normolipidemic individuals may spend the majority of time in the postprandial state and expose arteries to postprandial lipoproteins which may be related to atherogenesis(13,22).
Few studies have examined the role of exercise training with regard to the lipemic response after a high fat meal. Merrill et al.(20) investigated the TG response of young trained and untrained men after a high fat meal and concluded that endurance training was associated with diminished PP-HTG. Cohen et al. (6) also found trained men to exhibit lower PP-HTG than sedentary men. To our knowledge, no data are available on whether regular, recreational-type of exercise influences PP-HTG. In addition, no studies examined postprandial TG clearance rate and the LDL subfraction distribution in the same group of subjects. Therefore, the major purpose of this study was to examine whether differences exist in postprandial TG levels and LDL subfraction distribution among groups of different fitness levels. It was hypothesized that sedentary individuals will exhibit prolonged elevation of TG levels in response to a fatty meal and also exhibit a less favorable LDL subfraction profile than chronic exercisers (recreational and trained).
Fifty-four apparently healthy, nonsmoking, asymptomatic men and women (30 men and 24 women) subjects were recruited to fit into three groups: 1) sedentary (S), 2) recreational exercisers (R), and 3) endurance trained (T). The fitness requirements for the three groups were determined based on an activity questionnaire. They were: 1) Sedentary group: No regular exercise or exercise not more than 1 time per week for at least 2 yr prior to participation, 2) Recreational group: Regular, noncompetitive type of aerobic exercise for 3-5 times per week for at least 2 yr prior to participation, and 3) Endurance trained: Competition-based intense aerobic exercise training 5 or more times per week for at least 2 yr prior to participation.
Additional criteria for participation in the study were applied in an attempt to select a relatively homogeneous pool of subjects with the exception of exercise involvement. Specifically, each subject was screened using a health history questionnaire and eliminated from the study if he/she had more than one major cardiovascular disease risk factor or any disease symptoms as defined by the American College of Sports Medicine (1). All female participants were premenopausal, nonoral contraceptive users and were not on any estrogen replacement therapy. An initial blood analysis was performed for the purpose of selecting participants who were normolipemic (TG< 130 mg·dL-1, cholesterol < 230 mg·dL-1). Body composition (skinfolds) was assessed for all individuals in an attempt to avoid subjects who had extreme body fat (acceptable ranges: between 10 and 28% for men and between 15 and 33% for women) and extreme lifestyles except for exercise. Three skinfold sites were used for estimating percent body fat in men (abdominal, chest/pectoral, and thigh) and women (triceps, suprailiac, and thigh), and percent body fat was calculated according to the Jackson and Pollock equations (14,15).
Prior to participation in the study, subjects were asked to record a 4-d dietary intake to determine habitual dietary composition. To participate in the study each subject had to have: 1) an average percent fat in the diet 20-40% and 2) monounsaturated, saturated, and unsaturated fat intake each between 4 and 15% of total calories. The dietary intakes were analyzed using the Food Processor IV software program. No subjects were on lipid lowering drugs. One subject was dropped from the study because of caffeine consumption during the 8-h TG sampling period. Subject characteristics are shown inTable 1.
Each subject gave informed consent as approved by the Institutional Review Board of the University of Missouri-Columbia. Subjects were habituated to the use of the treadmill and performed an exhaustive treadmill test to measure maximal oxygen consumption for use as an estimation of cardiovascular fitness level. Following a 3-5 min warm-up at a self-selected pace at level treadmill grade, the initial speed was set at 4, 5, or 6 mph for the S, R, and T groups, respectively, and it was held constant for the first 2 min of testing. Then, speed was increased by 0.5 mph in 1-min increments for 4 min. Thereafter, TM grade was raised 2% every minute, and speed remained constant until exhaustion; then speed was decreased during a 3- to 5-min cooldown period. The highest ˙VO2 value obtained during the test was considered as the˙VO2max.
After the completion of the maximal testing, instructions were given to the subjects concerning the 24-h dietary and exercise restrictions prior to the next visit. Subjects were asked to follow a typical 24-h diet followed by a 12-h overnight fast the day before the third visit. Subjects were asked to eat a snack exactly 12 h prior to the morning blood collection and then to abstain from all food intake to ensure an identical fasting period for all subjects. During the 12-h overnight fast, subjects abstained from drinking caffeine or alcohol. Vigorous physical activity was not permitted for 36 h prior to the experimental session. All female subjects were scheduled for the experimental session visit during the last half of their monthly menstrual cycle in an attempt to avoid any major fluctuations in estrogen levels which may have influenced fasting serum TG or LDL concentrations.
During the experimental session, a 30-mL resting blood sample was drawn from each subject from an arm vein to Vacutainer tubes containing EDTA. The test meal then was given in a milkshake form and consumed within 10 min. The test milkshake consisted of 270 mL of whipping cream (90% calories from fat) and 65 g of banana split ice cream with walnuts (53% calories from fat, predominantly long chain fatty acids-LCFA) for a total of 980 kcal (83% of calories from fat, 100 g fat total). Additional blood samples (10 mL each) were drawn at 2, 4, 6, and 8 h after the meal to assess the rate of postprandial clearance of TG. The subjects consumed only plain water during this period. The magnitude of triglyceridemic response was quantified as follows: the area under the TG curve as defined by 2 lines, one connecting the individual TG values and one originating at the 0-h levels parallel to the abscissa (trapezoidal rule) (11).Equation
where Ln = the triglyceride concentration at N hours. Plasma was separated by centrifugation at 4°C for 15 min at 3750 rpm in a Beckman TJ-6R centrifuge (Palo Alto, CA) and aliquoted into a series of storage vials which were closed with a screw cap to prevent evaporation and stored at -70°C until analyses were performed within 2 months. The premeal plasma was used for total HDL-C, TG, and LDL subfraction separation while baseline (premeal) TG and postmeal plasma samples (2, 4, 6, and 8 h post) were used for determining the rate of postprandial TG clearance.
Whole plasma TG and C were measured by spectrophotometric assay using enzyme protocols (Cholesterol-Procedure #352, and triglyceride GPO-Trinder-Procedure #339, Sigma Diagnostics, St. Louis, MO) using known standards and lipid control. HDL-C was analyzed by spectrophotometric assay(procedure #352-4, Sigma Diagnostics). All samples were analyzed in duplicate. Any duplicates with concentration variation >5% were repeated.
LDL subfractionation was performed by separating fasting plasma LDL into three subfractions of increasing density by density gradient ultracentrifugation as described by Swinkels et al. (26). Specifically, 2.5 mL of plasma were adjusted to 1.10 g·mL-1 with solid NaBr and placed in the bottom of the polyallomer centrifuge tube. This was underlayered with 0.5 mL volume of 1.30 g·mL-1 saline to give the sample a flat (not rounded) bottom. The plasma was then overlayered successively with NaBr solutions: 1.5 mL of 1.063 g·mL-1; 3.0 mL of 1.030 g·mL-1; 3.0 mL of 1.019 g·mL-1; and 1.0 of 1.006 g·mL-1 (top of tube). The tube thus constituted was centrifuged in a SW41 rotor (Beckman, L7-65 centrifuge) at 36,000 rpm (160,000 g) for 20.6 h (including 15 min of acceleration; deceleration time not included) at 20°C. The separate lipoprotein fractions were recovered by puncturing the bottom of the tube and pumping the contents out with a peristaltic pump in separate tubes. Five fractions were collected: 1st fraction, HDL (3.9 mL); 2nd fraction, LDL3 (1.3 mL); 3rd fraction, LDL2 (1.3 mL); 4th fraction LDL1 (1.6 mL); 5th fraction, IDL(1900 mL); and 6th fraction, VLDL plus wash (2.6 mL). The LDL subfraction density ranges were as follows: LDL1, 1.021-1.030; LDL2, 1.030-1.042; and LDL3, 1.042-1.063 g·mL-1. All volumes and densities were previously established following extensive pilot work. The LDL subfractions were analyzed for cholesterol using cholesterol reagent(Boehringer-Mannheim Corp., Indianapolis, IN) at 500 nm wavelength and apoB100 on a Cobas Mira Analyzer (Roche Analytical Instruments, Inc., Nutley, NJ) to quantify the LDL subfractions.
Apoprotein B 100 levels were determined by an immunoturbidometric method using the SPQ antibody reagent kit (Incstar Corp., Stillwater, MN) with known standards following the manufacturer's instructions. Specifically, microvolumes of plasma and antibody diluent were pipetted into individual cuvettes. Following an initial incubation and measurement of blank, undiluted antiserum was added to the cuvettes. The sample (antigen) and undiluted plasma were mixed in the reaction cuvettes resulting in the formation of insoluble antigen-antibody complexes. This led to increased production of turbidity in the mixture; after an incubation period of approximately 5 min, the absorbance of the solution was measured at the analytical wavelength (340 nm). Because there is only one apoB molecule per LDL particle, the amount of apoB reflects the LDL particle number for each subfraction.
The coefficients of variation were the following: Intraassay: LDL1-C= 4.8%, LDL2-C = 3.8%, LDL3-C = 2.8%, TC = 5.3%, HDL-C = 2.7%; Inter-assay: TG = 1.9%. The percent recovery for the cholesterol in the LDL subfractions was 97% ± 1.5%.
Triglyceride concentrations before and at 2, 4, 6, and 8 h after the meal were analyzed using repeated measures ANOVA on one factor (time). When a significant time × group interaction was observed, repeated measures ANOVA was calculated for each significant 2-h segment (0-2, 2-4, 4-6, 6-8 h) and significances (P < 0.05) were followed by Tukeypost-hoc pairwise comparisons to examine which pairs of groups were different within that segment. Subject characteristics, low density lipoprotein subfractions, and HTG scores were analyzed using MANOVA, and significant MANOVAs (Wilks's λ <0.05) were followed by ANOVApost-hoc tests. Significant F ratios (P<0.05) were followed by Tukey tests to determine which pairs of groups were different.
An attempt was made to match the three groups as closely as possible for age, body weight, dietary habits, and fasting lipid (TC, TG) concentration while ensuring at the same time distinct exercise training levels(Table 1). As expected, maximal oxygen consumption and body composition values were different among the groups.
Exercise training appeared to be an influencing factor on the postprandial TG response. (Fig. 1). The repeated measures ANOVA showed that the S group had significantly higher TG concentration at 2, 4, 6, and 8 h after the meal compared with the T group. A significant difference also was found at 4 h when the S group was compared with the R group. There were no statistical differences in the TG curves for the R and T groups.
The overall group MANOVA was significant among groups, Wilks's λ = 0.32, F(28,66) = 1.77, P < 0.03. The mean HTG score (area under the curve) was significantly lower for the R and T groups compared to the S group (P < 0.05) (Fig. 2). However, the means for the R and S groups were not statistically different. Men and women were not significantly different on HTG score.
HDL-C values were not significantly different among the groups(Table 2). Endurance trained subjects had significantly lower LDL3-C and LDL3-apoB100 levels compared with sedentary individuals (Tables 2 and 3). No differences in LDL subfractions were found between the S and R groups or the R and T groups(P> 0.05). Moreover, the particle number and cholesterol content of LDL1, LDL2, and total LDL were similar among the three groups. Men had significantly higher mean total LDL-C and apoB100 than women, primarily owing to the differences in the LDL3 subfraction values(Tables 2 and 3).
Postprandial hypertriglyceridemia. It appears that regular exercise attenuated the TG response to a high fat meal as indicated by the HTG area under the curve score. Sedentary subjects showed a poorer fat tolerance compared with the other two groups. Several other studies reported similar findings (6,20) comparing trained and sedentary subjects. Furthermore, the effect of exercise on PP-HTG was not entirely dose dependent. In the present study, participation in a regular, recreational type of aerobic exercise program was associated with a lower PP-HTG score than the sedentary group, but higher volume, competition-based aerobic training was not associated with further decreases in the HTG score. Most research studies in this area indicated that the key enzyme regulating the TG clearance is lipoprotein lipase (LPL), and the activity of this enzyme has been shown to be elevated as a result of training (18).
The significant differences between the S and R and the S and T curves indicates that exercisers may be able to clear incoming TG faster than sedentary individuals. This finding is supported by Cohen et al.(6) who showed lower chylomicron-TG half-life for trained subjects versus controls suggesting that LPL-induced TG clearance in trained subjects begins as soon as chylomicrons enter the circulation.
Since fat absorption rate was not measured in the present study, the question remains whether the observed differences in the HTG curves among groups reflect differences in intestinal absorption or systemic clearance of TG. Cohen (5) found that tests which rely on absorption(oral fat tolerance and intraduodenal perfusion) where highly correlated with intravenous fat tolerance test (r = 0.79). Cohen et al.(6) also reported that exercise training alters the TG clearance rate following a high fat meal (140 g fat) and the effect is independent of the absorptive process. Therefore, it is likely that the observed PP-HTG responses among groups in the present study reflect clearance rather than absorptive differences.
Lipoprotein subfraction distribution. Endurance trained subjects had lower number and C content of the small LDL3 particles compared with the other two groups even though baseline TC and LDL-C concentrations were similar among groups. This is particularly important considering the growing evidence for the potentially atherogenic role of LDL3 (7). Although the potential mechanisms were not under investigation in this study, it is likely that a decrease in hepatic lipase(HL) and CETP induced by exercise training may prevent the formation of small and cholesterol-rich LDL particles. Seip et al. (23) demonstrated that a regular, supervised aerobic exercise program for 9-12 months decreased the activity of CETP and HL while increasing the LPL activity. The absence of a significant difference between the S and R groups indicates that for significant changes in the small LDL3 to occur, a higher fitness level may be required. This is supported by Williams et al.(29) who reported similar findings after examining the LDL subfractions in sedentary and trained individuals. More evidence comes from Stucchi et al. (24) who showed that two years of aerobic exercise training in pigs produced significant composition changes in LDL particles, with the smaller LDL being more responsive to training.
Dietary fat and total energy intake. In this study, sedentary subjects and recreational exercisers had significantly higher SFA intake compared with the highly trained group (Table 1). O'Hanesian et al. (21) reported that increased total fat and saturated intake was associated with increases in LDL-C concentrations. Likewise Wahrburg et al. (27) found that increased fat intake caused rapid elevations in apoprotein B levels. Thus, it is possible that the differences in LDL subfractions between the S and T groups was partially a result of the differences in dietary fat ingestion.
Gender comparisons. Men and women exhibited similar HTG scores, and resting TG values also were not different between men (67.1 ± 26.8 mg·dL-1) and women (61.5 ± 22.5). PP-HTG may not be associated with cardiovascular disease in women (10), and the role of resting TG values in predicting CV disease risk in women and men is controversial (3,10). In the present study, total LDL-C and apoB100 were lower in women. Other investigators have not observed this difference (25), and LDL-C may not be a strong predictor of CV disease in women (3). However, the presence of the more dense LDL particles has been consistently linked to coronary heart disease in men and women (7). In the present study, most of the difference in LDL was caused by the lower concentrations in the LDL3 subfraction in the women. This gender difference in LDL subfractions has been observed previously(25). Watson et al. (28) observed an increased percent distribution in LDL1 between men and women and a tendency for LDL3 distribution to be lower in women. Thus, it appears that the cholesterol and apoB 100 values in the LDL3 subfraction may be important markers for the CV risk difference between men and women. Exercise training appears to modify these levels in both sexes.
In summary, the comparison among fitness groups in the present study suggests that even in normolipidemic subjects, the TG response to fat ingestion and lipoprotein subfraction differences are present. There is growing evidence for the atherogenic role of PP-HTG, and our data support the potentially protective role of exercise training. A reduction in PP-HTG apparently can be achieved with regular, recreational type of aerobic exercise 3-5 times per week. However, in normolipidemic subjects, competition-based vigorous aerobic training appears to be necessary for reduction in LDL3. Since both prolonged PP-HTG and elevated LDL3 subfraction occur in sedentary individuals, there may be a biological link between these parameters.
These results suggest that recreational and competitive aerobic exercise training attenuate the TG response to a high fat meal but that a higher volume, competitive training regimen may be necessary to alter the LDL subfraction composition.
1. American College of Sports Medicine. Guidelines for Exercise Testing and Prescription
. Baltimore: Williams and Wilkins, 1995.
2. Austin, M. A., J. L. Breslow, J. E. Hennekens, J. E. Buring, W. C. Willett, and R. M. Krauss. Low density lipoprotein patterns and risk of myocardial infarction. JAMA
3. Bass, K. M., C. J. Newschaffer, M. J. Klag, T. L. Bush. Plasma lipoprotein levels as predictors of cardiovascular death in women.Arch. Intern. Med.
4. Berg, A., I. Frey, M. W. Baumstark, M. Halle, and J. Keul. Physical activity and lipoprotein lipid disorders. Sports Med.
5. Cohen, J. C. Chylomicron triglyceride clearance: comparison of three assessment methods. Am. J. Clin. Nutr.
6. Cohen, J. C., T. D. Noakes, and A. J. S. Benade. Postprandial lipemia and chylomicron clearance in athletes and in sedentary men. Am. J. Clin. Nutr.
7. Gardner, D. G., S. P. Fortmann, and R. M. Krauss. Association of small low-density lipoprotein particles with the incidence of coronary artery disease in men and women. JAMA
8. Geurian, K., J. B. Pinson, and C. W. Weart. The triglyceride connection in atherosclerosis. Ann. Pharmacother
. 26:1109-1117, 1992.
9. Ginsberg, H. N. Lipoprotein metabolism and its relationship to atherosclerosis. Med. Clin. North Am.
10. Ginsberg, H. N., J. Jones, W. S. Blaner, et al. Association of postprandial triglyceride and retinyl palmitate responses with newly diagnosed exercise-induced myocardial ischemia in middleaged men and women. Arterioscler. Thromb. Vasc. Biol.
11. Grant, K. I., M. P. Marais, and M. A. Dhansay. Sucrose in a lipid-rich meal amplifies the postprandial excursion of serum and lipoprotein triglyceride and cholesterol concentrations by decreasing triglyceride clearance. Am. J. Clin. Nutr.
12. Griffin, B. A., M. J. Caslake, B. Yip, G. W. Tait, C. J. Packard, and J. Shepherd. Rapid isolation of low density lipoprotein (LDL) subfractions from plasma by density gradient ultracentrifugation.Arteriosclerosis
13. Grundy, S. M. and G. L. Vega. Two different views of the relationship of hypertriglyceridemia to coronary heart disease. Arch. Intern. Med.
14. Jackson, A. S. and M. L. Pollock. Generalized equations for predicting body density of men. Br. J. Nutr.
15. Jackson, A. S. and M. L. Pollock. Generalized equations for predicting body density for women. Med. Sci. Sports Exerc.
16. Karpe, F. P., P. Tornvall, T. Olivecrona, G. Steiner, L. A. Carslon, and A. Hamsten. Composition of human low density lipoprotein: effects of postprandial triglyceride-rich lipoproteins, lipoprotein lipase, hepatic lipase and cholesteryl ester transfer protein.Atherosclerosis
17. Krauss, R. M. and D. J. Burke. Identification of multiple subclasses of plasma low density lipoproteins in normal humans.J. Lipid Res.
18. Ladu, M. J., H. Kapsas, and W. K. Palmer. Regulation of lipoprotein lipase in muscle and adipose tissue during exercise. J. Appl. Physiol
19. McNamara, J. R., J. L. Jenner, Z. Li, P. W. F. Wilson, and E. J. Schaefer. Change in LDL particle size associated with change in plasma triglyceride concentration. Arterioscl. Thromb.
20. Merill, J. R., R. G. Holly, R. L. Anderson, N. Rifai, M. E. King, and R. Demeersman. Hyperlipemic response of young trained and untrained men after a high fat meal. Arteriosclerosis
21. O'Hanesian, B. Rosner, L. M. Bishop, and F. M. Sacks. Effects of inherent responsiveness to diet and dat-to-day diet variation on plasma lipoprotein concentrations. Am. J. Clin. Nutr.
22. Patsch, J. R., G. Miesenbock, and T. Hopferwieser. Relation of triglyceride metabolism and coronary artery disease.Arterioscl. Thromb.
23. Seip, R. L., P. Moulin, and T. Cocke. Exercise training decreases plasma cholesteryl ester transfer protein. Arterioscler. Thromb.
24. Stucchi, A. F., A. H. M. Terpstra, T. L. Foxall, R. J. Nicolosi, and S. C. Smith. The effect of exercise on plasma lipids and LDL subclass metabolism in miniature swine. Med. Sci. Sports. Exerc.
25. Swinkels, D. W., P. N. M. Demacker, J. C. M. Hendriks, and A. Van't Laar. Low density lipoprotein subfractions and relationship to other risk factors for coronary artery disease in healthy individuals.Arteriosclerosis
26. Swinkels, D. W., H. L. M. Hak-Lemmers, and P. N. M. Demacker. Single spin density gradient ultracentrifugation method for the detection and isolation of light and heavy low density lipoprotein subfractions. J. Lipid Res.
27. Wahrburg, U., H. Martin, M. Sandkamp, H. Schulte, and G. Assmann. Comparative effects of a recommended lipid-lowering diet vs a diet rich in monounsaturated fatty acids on serum lipid profiles in healthy young adults. Am. J. Clin. Nutr.
28. Watson, T. D. G., M. J. Caslake, D. J. Freeman, et al. Determinants of LDL subfraction and concentrations in young normolipidemic subjects. Arterioscler. Thromb.
29. Williams, P. T., R. M. Krauss, P. D. Wood, F. T. Lindgren, C. Giotas, and K. M. Vranizan. Lipoprotein subfractions of runners and sedentary men. Metab.