Share this article on:

Apolipoprotein E polymorphism and the relationships of physical fitness to plasma lipoprotein-lipid levels in men and women


Medicine & Science in Sports & Exercise: May 1999 - Volume 31 - Issue 5 - p 692-697

Apolipoprotein E polymorphism and the relationships of physical fitness to plasma lipoprotein-lipid levels in men and women. Med. Sci. Sports Exerc., Vol. 31, No. 5, pp. 692-697, 1999.

Purpose: A high level of cardiovascular fitness is generally associated with a plasma lipoprotein-lipid profile predictive of a low cardiovascular disease risk. We have investigated whether apolipoprotein (apo) E polymorphism could alter the relationships of physical fitness to plasma lipoprotein-lipid levels in a sample of healthy untrained subjects (64 premenopausal women and 65 men).

Methods: Subjects were grouped according to gender and apo E phenotype determined by isoelectric focusing electrophoresis.

Results: In both genders, O2max expressed in mL·kg−1·min−1 was negatively correlated with plasma triglyceride levels in apo E2 carriers and apo E3 homozygotes (−0.55 ≤ r ≤ − 0.31; P < 0.05), whereas these associations were not found in apo E4 groups. Plasma low-density lipoprotein (LDL)-C levels were negatively associated with O2max (r = −0.39; P < 0.05) only in women homozygotes for apo E3 whereas O2max was positively correlated with plasma high-density lipoprotein (HDL)2-C levels only in men (r = 0.51; P < 0.001) and women (r = 0.65; P < 0.001) who were apo E3 homozygotes. A control for concomitant association with body fat mass and glucose intolerance performed by partial correlation analyses revealed that, with the exception of the plasma HDL2-C levels in the apo E3 homozygotes, most of the significant associations between O2max (mL·kg−1·min−1) and plasma lipoprotein-lipid levels were mediated by concomitant variation in body fatness and glucose tolerance.

Conclusions: These results suggest that the magnitude of the relationships between O2max and plasma lipoprotein-lipid levels is influenced by the apo E polymorphism. Thus, apo E2 carriers may be particularly responsive to improved fitness, thereby preventing the development of hypertriglyceridemia and type III dyslipoproteinemia.

Lipid Research Center, Physical Activity Sciences Laboratory, and Diabetes Research Unit, Laval University, Québec, CANADA

Submitted for publication June 1997.

Accepted for publication October 1998.

The authors express their gratitude to the subjects for their excellent collaboration and to the staff of the Physical Activity Sciences Laboratory, the Lipid Research Center, and the Diabetes Research Unit for their dedicated work.

This work was supported by the Québec Heart and Stroke Foundation and by the Medical Research Council of Canada. J. St.-Amand is a FCAR and Promutuel-vie fellow.

Sital Moorjani is deceased.

Address correspondence to: Jean-Pierre Després, Ph.D., Director, Lipid Research Center, Laval University Medical Research Center, CHUL, 2705 Blvd Laurier, Ste-Foy, Québec, Canada G1V 4G2. E-mail:

A high level of physical fitness is an independent predictor of a low cardiovascular disease risk (2,19). Physically active individuals are also characterized by lower plasma triglyceride, total cholesterol, and LDL-C levels and by higher plasma HDL-C and HDL2-C levels than untrained individuals (15,18,22,23). Moreover, physical fitness has been reported to be a significant correlate of plasma HDL-C, HDL2-C and triglyceride levels (13,15,23).

However, additional factors such as body composition, insulin sensitivity, gender, and heredity are also known correlates of plasma lipoprotein-lipid profile (3,6,11,14,26). Indeed, obese individuals usually have higher plasma triglyceride and LDL-C levels and lower plasma HDL-C levels than nonobese individuals (3). Individuals who have a very high energy expenditure because of a demanding exercise training program are generally leaner than untrained individuals (15,22). Thus, the lipoprotein-lipid profile of physically active individuals could also be partly attributed to their leanness.

Another characteristic that could be responsible for the "protective" plasma lipoprotein-lipid profile of physically active subjects is their higher level of insulin sensitivity compared with that of untrained individuals (1,18). For example, insulin sensitivity has been negatively correlated with plasma triglyceride and positively correlated with HDL-C levels, whereas glucose intolerance has been associated with elevated plasma triglyceride and LDL-C concentrations, and with reductions in plasma HDL-C and HDL2-C levels (8,12).

Although a favorable plasma lipoprotein profile is generally found among subjects with a high level of physical fitness, considerable heterogeneity remains among well-trained subjects (4). Furthermore, genetic factors could modulate plasma lipoprotein levels and the responsiveness of the lipoprotein profile to physical training (4). For instance, it is well known that plasma lipoprotein-lipid levels are also influenced by apolipoprotein (apo) E polymorphism, since this protein, found on the membrane of triglyceride-rich lipoproteins and HDL, is the ligand for the LDL receptor, LDL-receptor-related protein (LRP), and very-low-density lipoprotein (VLDL) receptor (27,29). The apo E2 and apo E4 isoforms differ from apo E3 by a single amino acid substitution in position 158 and 112, respectively. As in apo E2 homozygotes, apo E2 carriers generally have lower plasma total cholesterol and LDL-C levels than apo E3 homozygotes. Apo E2 carriers are also characterized by an accumulation of chylomicron and VLDL remnants caused by an impaired lipoprotein receptor-binding activity of apo E2 (29). In contrast, apo E4 has an enhanced catabolic rate, and apo E4 carriers generally have higher plasma total cholesterol, LDL-C levels, and risk of cardiovascular disease than apo E3 homozygotes (7,29). Thus, apo E isoform could give an inborn susceptibility to altered plasma lipoprotein-lipid profile which could be amplified by or prevented by exercise.

However, no study has investigated the potential relationships between physical fitness and plasma lipoprotein-lipid profile while taking into consideration the apo E polymorphism. We have therefore examined the relationships between cardiovascular fitness and plasma lipoprotein-lipid levels in groups of men and women defined on the basis of apo E polymorphism, with and without control for the concomitant variation in body composition and glucose tolerance in an attempt to assess the independent contribution of cardiorespiratory fitness to plasma lipoprotein-lipid levels.

Back to Top | Article Outline


Subjects. Sixty-four healthy premenopausal women (25 to 48 yr) and 65 men (30 to 42 yr) were recruited by solicitation in the media. They gave written consent to participate in this study, which had been approved by the Medical Ethics Committee of Laval University. This study has been conducted in conformance with the policy statement of the American College of Sports Medicine. Each participant underwent a complete medical examination carried out by a physician. Individuals with cardiovascular disease, diabetes, endocrine disorders, or on medication were excluded. No subject was heavy drinker. Since physical activity is a known correlate of plasma lipoprotein levels, subjects had to be untrained (not more than one session of 30 min continuous endurance exercise per week assessed by a 3-d record) to be included in the study. Body weight was stable for at least 2 months before the study and smokers were excluded.

Fitness. Maximal aerobic power was measured using a progressive test to exhaustion on a treadmill, as previously described (5). Oxygen consumption (O2) was recorded using an open gas circuit system (Metabolic Measurement Cart, Beckman, Anaheim, CA), and O2max was considered as the highest O2 recorded during the test for 1 min. O2max was expressed in mL O2·min−1·kg−1 of body weight.

Body fatness. The mean of six hydrostatic weighing measurements was used for the estimation of percent body fat from density, using the Siri equation (25). Pulmonary residual volume was determined before immersion in the hydrostatic tank with the closed circuit helium dilution method of Meneely and Kaltreider (16). Fat mass was calculated by multiplying percentage of body fat by body weight. Waist and hip circumferences were measured and the waist:hip ratio (WHR) was calculated as previously described (26).

Oral glucose tolerance test (OGTT). A 75-g oral glucose tolerance test was performed in the morning after an overnight fast. Blood samples were collected through a venous catheter from an antecubital vein at −15, 0, 15, 30, 45, 60, 90, 120, 150, and 180 min for the determination of plasma glucose concentrations. Plasma glucose was enzymatically measured (21), and the glucose total area under the curve during the OGTT was determined with the trapezoid method.

Plasma lipoprotein-lipid analyses. Blood samples were collected before the maximal aerobic power test in the morning after 12 h inactive fasting for determination of lipoprotein-lipid levels. Venous blood from an antecubital vein was drawn into vacutainer tubes containing EDTA while subjects were in a supine position. Blood samples were obtained while women were in their follicular phase, between the fifth and twelfth days of the menstrual cycle. Triglyceride and cholesterol concentrations in whole plasma and in lipoprotein fractions were measured using automated techniques as previously described (6). Ultracentrifugation was used to isolate plasma VLDL (d < 1.006 g·mL−1). HDL was isolated by precipitation of the infranatant (d > 1.06 g·mL−1) with heparin and MnCl2. The cholesterol content of HDL2 subfraction prepared by precipitation was also determined (10).

Apolipoprotein E phenotyping. Apo E phenotypes were determined as previously described (6). In brief, the VLDL fraction was delipidated with acetone-ethanol 1:1 (v/v) followed by diethyl ether at −20°C. The colorless precipitated protein was dried under N2 at room temperature and stored at −20°C. VLDL-protein was solubilized in 10 mM Tris-HCL (pH 8.2), containing 8 M urea and 30 mM dithiothreitol, immediately before electrophoresis. The apo E isoproteins were separated by isoelectric focusing electrophoresis in 7.5% polyacrylamide gels containing 8 M urea and a 2% mixture of ampholytes, pH 4-6 (LKB-Produkter AB, Bromma, Sweden) and 5-7 (Bio-Rad, Richmond, CA) in the proportion of 4:1. Gels were polymerized in cylindrical tubes and run in a water-cooled column disk electrophoresis apparatus (Hoefer Scientific Instruments, San Francisco, CA) at 4°C for 16 h at 150 V.

Statistical analyses. In each gender, the subjects were classified according to their apo E phenotype: 1) the apo E2 group (15 men and 22 women), including individuals carrying either the E2/2 or E3/2 phenotypes; 2) the apo E3 group (38 men and 25 women), composed of individuals homozygous for E3/3 phenotype; and 3) the apo E4 group (12 men and 17 women), composed of individuals carrying either the E3/4 or E4/4 phenotypes. Subjects carrying the E2/4 phenotype (3 men and 2 women) were not included in the analyses. The allele frequencies for ∊2, ∊3, and ∊4 evaluated by gene-counting method were 0.17, 0.68, and 0.15, respectively. The higher frequency of ∊2 and ∊4 in the French Canadian population is responsible for the enrichment of these phenotypes in the sample (9). All variables were normally distributed. The apo E phenotype group effect was tested by one-way analysis of variance (ANOVA). The associations between two variables were quantified by using the Pearson product-moment correlation coefficient. Partial correlation analyses were performed to estimate the contributions of independent variables to the variance of plasma lipoprotein-lipid levels. The Statistical Analysis System (SAS Institute, Cary, NC) was used for all analyses.

Back to Top | Article Outline


In both genders, no difference among groups defined according to apo E phenotypes were found for age, body mass index, fat mass, physical fitness (O2max expressed in mL·kg−1·min−1), and glucose tolerance (plasma glucose area) (Tables 1 and 2). However, the VLDL-C/triglyceride ratio, which is suggestive of remnant accumulation, was higher in male apo E2 carriers. ANOVA indicated no other group difference for plasma lipoprotein-lipid levels in either men or women, although the apo E4 group tended to have higher plasma LDL-C levels than the apo E2 group.





Figure 1 shows the relationships between O2max expressed in mL·kg−1·min−1 and plasma triglyceride levels in subgroups of women defined on the basis of apo E phenotype. Plasma triglyceride levels were associated with O2max only in women carrying the apo E2 isoform or homozygous for apo E3. In women homozygous for apo E3, O2max was significantly correlated with all plasma lipoprotein-cholesterol levels (Table 3). In contrast, O2max was not correlated with plasma lipoprotein-cholesterol levels in women carrying the apo E4 isoform, with the exception of HDL-C levels. Significant relationships were found between O2max and plasma VLDL-C levels as well as HDL-C/cholesterol ratio in female apo E2 carriers. O2max was significantly correlated (−0.42 ≤ r ≤ + 0.55; P < 0.05) with all plasma lipoprotein-lipid levels in the whole sample of 64 women (results not shown). Intercorrelations between the other variables are reported in our previous papers (3,5,6,8,20,26).

Figure 1

Figure 1



In men carrying the apo E2 isoform, O2max showed a negative correlation with plasma triglyceride levels (Fig. 2). In men homozygous for apo E3, O2max was associated with plasma triglyceride, HDL-C, and HDL2-C levels as well as with the HDL-C/cholesterol ratio (Table 4). However, no significant correlation was found among male apo E4 carriers. O2max was significantly correlated (−0.24 ≤ r ≤ + 0.34; P < 0.05) with plasma triglyceride, HDL-C, and HDL2-C levels and with the HDL-C/cholesterol ratio in the whole sample of 65 men (results not shown).

Figure 2

Figure 2



Partial correlation analyses were performed in an attempt to quantify the independent contribution of O2max to the variation of the plasma lipoprotein-lipid profile after adjustment for two important correlates of plasma lipoprotein levels: body fatness and the integrated concentrations of plasma glucose measured after a 75-g oral glucose challenge (glucose area). In both genders no significant residual relationship was found between O2max and plasma lipoprotein-lipid levels after correction for body fat mass and glucose tolerance in apo E2 and E4 groups (Table 5). In both women and men homozygous for the apo E3 isoform, O2max remained correlated with plasma HDL2-C levels after statistical adjustment for body fat mass and glucose tolerance.



Back to Top | Article Outline


Many studies have reported significant relationships between physical fitness and plasma lipoprotein-lipid levels (13,15,23). However, to our knowledge no study has considered the contribution of apo E polymorphism to these relationships. The aim of the present study was therefore to verify whether the magnitude of the expected correlations between physical fitness and plasma lipoprotein-lipid levels could be modulated by apo E polymorphism in men and women.

We report here for the first time that the apo E polymorphism alters the relationships between O2max (expressed in mL·kg−1·min−1) and plasma lipoprotein-lipid levels. Indeed, significant correlations between O2max and plasma lipoprotein-lipid levels were mainly noted among individuals homozygous for the apo E3 isoform. In apo E2 carriers, O2max was significantly related with plasma triglyceride levels, whereas no relationship between O2max and the lipoprotein-lipid profile could be noted in men carrying the apo E4 isoform. In women who were apo E4 carriers, only plasma HDL-C levels showed a significant correlation with O2max. The correlations reported herein between O2max and plasma triglyceride and HDL-C levels are concordant with previous studies that did not control for apo E polymorphism (13,15,23) since apo E3 is the most prevalent form of apo E (29). The smaller sample size in the other groups of subjects carrying less frequent polymorphism makes the correlation analysis less powerful in these groups. However, our results indicate that physical fitness may have an important influence on plasma triglyceride levels in individuals carriers of apo E2 since genetic susceptibility to develop type III dyslipoproteinemia is given by apo E2 (28). Thus, it is postulated that apo E2 carriers may be particularly responsive to improved fitness, thereby preventing the development of hypertriglyceridemia and type III dyslipoproteinemia. On the other hand, apo E4 carriers could be characterized as nonrespondent to the influences of exercise on plasma lipoprotein-lipid profile in men as well as in women. The results of the present cross-sectional analysis are therefore in agreement with our previous observation that genetic factors can modulate the effects of physical training on plasma lipoprotein-lipid levels (4).

On the other hand, the relationships of physical fitness to plasma lipoprotein-lipid levels could be attributed to concomitant variation in other factors affecting plasma lipoprotein-lipid concentrations. As body fatness and glucose tolerance are significant correlates of plasma lipoprotein levels and physical fitness (3-6,8,12,20), we have used partial correlation analyses to quantify the independent relationships between physical fitness and plasma lipoprotein-lipid profile after control for the contributions of these two variables. No association between O2max and plasma lipoprotein-lipid levels was found after adjustment for body fat mass and glucose tolerance, with the exception of plasma HDL2-C levels which remained significantly correlated with O2max in apo E3 homozygotes. These results suggest that body fatness and glucose tolerance largely mediate the associations between O2max and plasma lipoprotein-lipid levels, including HDL-C/cholesterol ratio. The independent association between O2max and plasma HDL2-C levels is concordant with the known relationship of cardiorespiratory fitness to the HDL2 subfraction (24). It is also in agreement with results of Kuusi et al. (13) who have reported that physical fitness is an independent correlate of plasma HDL2-C levels after adjustment for body fat and fasting plasma insulin levels in a sample of 27 young men not defined on the basis of apo E polymorphism. In the study of Nakamura et al. (17), regular joggers had higher plasma HDL-C levels and an elevated HDL-C/cholesterol ratio compared with untrained controls matched for sex, age, relative body weight, total cholesterol, and triglyceride levels. Nikkilä et al. (18) have also reported a significant difference between plasma HDL-C levels of long distance runners and controls matched for body weight. However, in the study of Williams et al. (30) the increase of HDL2b by physical training was no longer significant after adjustment for the concomitant decrease in the BMI, emphasizing the importance of the concomitant variation in body fatness associated with a high level of cardiorespiratory fitness.

In conclusion, a high level of physical fitness was, in the present study, associated with a favorable plasma lipoprotein-lipid profile mainly among individuals homozygous for the apo E3 isoform and with reduced plasma triglyceride levels in individuals who were apo E2 carriers. However, these associations were largely attributable to the concomitant variation of body fatness and glucose tolerance, with the exception of plasma HDL2-C levels which were independently related to O2max among apo E3 homozygotes. These results emphasize the importance of apo E polymorphism as a relevant factor explaining individual differences in the plasma lipoprotein-lipid profile of individuals showing comparable levels of physical fitness. Controlled interventional studies are needed to support the current results.

Back to Top | Article Outline


1. Björntorp, P., M. Fahlen, C. Grimby, et al. Carbohydrate and lipid metabolism in middle-aged, physically well-trained men. Metabolism 21:1037-1044, 1972.
2. Blair, S. N., J. B. Kampert, H. W. Kohl, et al. Influences of cardiorespiratory fitness and other precursors on cardiovascular disease and all-cause mortality in men and women. JAMA 276:205-210, 1996.
3. Després, J.-P., S. Moorjani, A. Tremblay, et al. Relation of high plasma triglyceride levels associated with obesity and regional adipose tissue distribution to plasma lipoprotein-lipid composition in premenopausal women. Clin. Invest. Med. 12:374-380, 1989.
4. Després, J.-P., S. Moorjani, A. Tremblay, et al. Heredity and changes in plasma lipids and lipoproteins after short-term exercise training in men. Arteriosclerosis 8:402-409, 1988.
5. Després J.-P., M. C. Pouliot, S. Moorjani, et al. Loss of abdominal fat and metabolic response to exercise training in obese women. Am. J. Physiol. 261:E159-E167, 1991.
6. Després, J.-P., M.-F. Verdon, S. Moorjani, et al. Apolipoprotein E polymorphism modifies relation of hyperinsulinemia to hypertriglyceridemia. Diabetes 42:1474-1481, 1993.
7. Eichner, J. E., L. H. Kuller, T. J. Orchard, et al. Relation of apolipoprotein E phenotype to myocardial infarction and mortality from coronary artery disease. Am. J. Cardiol. 71:160-165, 1993.
8. Ferland, M., J.-P. Després, A. Nadeau, et al. Contribution of glucose tolerance and plasma insulin levels to the relationships between body fat distribution and plasma lipoprotein levels in women. Int. J. Obes. 15:677-688, 1991.
9. Gaudet, D., S. Moorjani, C. Gagné, et al. Premature coronary artery disease and high prevalence of monogenic traits for lipid disorders among French Canadians in a North Eastern region of Québec province. Atherosclerosis 109:203, 1994.
10. Gidez, L. I., G. J. Miller, M. Burstein, S. Slagle, and H. A. Eder. Separation and quantitation of subclasses of human plasma HDLs by a simple precipitation procedure. J. Lipid. Res. 23:1206-1223, 1982.
11. Godsland, I. F., V. Wynn, D. Crook, and N. E. Miller. Sex, plasma lipoproteins, and atherosclerosis: prevailing assumptions and outstanding questions. Am. Heart J. 114:1467-1503, 1987.
12. Johansen, K. and O. Munck. The relationship between maximal oxygen uptake and glucose tolerance/insulin response ratio in normal young men. Horm. Metab. Res. 11:424-427, 1979.
13. Kuusi, T., E. A. Nikkilä, P. Saarinen, P. Varjo, and L. A. Laitinen. Plasma high density lipoproteins HDL2, HDL3, and postheparin plasma lipases in relation to parameters of physical fitness. Atherosclerosis 41:209-219, 1982.
14. Lusis, A. J. Genetic factors affecting blood lipoproteins: the candidate gene approach. J. Lipid. Res. 29:397-429, 1988.
15. Marti, B., M. Knobloch, W. F. Riesen, and H. Howald. Fifteen-year changes in exercise, aerobic power, abdominal fat, and serum lipids in runners and controls. Med. Sci. Sports Exerc. 23:115-122, 1991.
16. Meneely, G. R. and N. L. Kaltreider. Volume of the lung determined by helium dilution. J. Clin. Invest. 28:129-139, 1949.
17. Nakamura, N., H. Uzawa, H. Maeda, and T. Inomoto. Physical fitness: its contribution to serum HDL. Atherosclerosis 48:173-183, 1983.
18. Nikkilä, E. A., M. R. Taskinen, S. Rehunen, and M. Harkonen. Lipoprotein lipase activity in adipose tissue and skeletal muscle of runners: relation to serum lipoproteins. Metabolism 27:1661-1671, 1978.
19. Peters, R. K., L. D. Cady, D. P. Bischoff, L. Bernstein, and M. C. Pike. Physical fitness and subsequent myocardial infarction in healthy workers. JAMA 249:3052-3056, 1983.
20. Pouliot M.-C., J.-P. Després, A. Nadeau, et al. Visceral obesity in men: associations with glucose tolerance, plasma insulin, and lipoprotein levels. Diabetes 41:826-834, 1992.
21. Richterich, R. and H. Dauwalder. Zur bestimmung der plasma-glukosekonzentration mit der hexokinase-glucose-6-phosphat-dehydrogenase-method. Schweiz. Med. Wochenschr. 101:615-618, 1971.
22. Sady, S. P., E. M. Cullinane, A. Saritelli, D. Bernier, and P. J. Thompson. Elevated high-density lipoprotein cholesterol in endurance athletes is related to enhanced plasma triglyceride clearance. Metabolism 37:568-572, 1988.
23. Schnabel, A. and W. Kindermann. Effect of maximal oxygen uptake and different forms of physical training on serum lipoproteins. Eur. J. Appl. Physiol. 48:263-277, 1982.
24. Schwartz, R. S. The independent effects of dietary weight loss and aerobic training on HDLs and apolipoprotein A-I concentrations in obese men. Metabolism 36:165-171, 1987.
25. Siri, W. E. The gross composition of the body. Adv. Biol. Med. Phys. 4:239-280, 1956.
26. St.-Amand J., J.-P. Després, S. Lemieux, et al. Does lipoprotein or hepatic lipase activity explain the protective lipoprotein profile of premenopausal women? Metabolism 44:491-498, 1995.
27. St. Clair, R. W. and U. Beisiegel. What do all the apolipoprotein E receptors do? Curr. Opin. Lipidol. 8:243-245, 1997.
28. Utermann, G., K. H. Vogelberg, A. Steinmetz, et al. Polymorphism of apolipoprotein E: II. genetics of hyperlipoproteinemia type III. Clin. Genet. 15:37-62, 1979.
29. Weisgraber, K. H. Apolipoprotein E: structure-function relationships. Adv. Prot. Chem. 45:249-302, 1994.
30. Williams, P. T., R. M. Krauss, K. M. Vranizan, J. J. Albers, and P. D. S. Wood. Effects of weight-loss by exercise and by diet on apolipoproteins A-I and A-II and the particle-size distribution of high-density lipoproteins in men. Metabolism 41:441-449, 1992.


© 1999 Lippincott Williams & Wilkins, Inc.