The HERITAGE Family Study

HERITAGE --Genetics, Response to Exercise, Risk Factors


It is widely recognized that individuals can respond quite differently to a given intervention, such as drugs, diet, or exercise. For instance, there are considerable individual differences in improvement in maximal oxygen uptake (VO2max; measure of aerobic endurance capacity) with aerobic training.

Studies conducted with young or older adults have typically reported gains in VO2max ranging from almost 0% to 50%, even though all the subjects completed exactly the same training program under close supervision. Scientists had previously assumed that these variations result from differing degrees of compliance with the training program, i.e., good compliers have the highest percentage of improvement and poor compliers show little or no improvement. However, it is now clear that even when there is full compliance with the program, substantial variations occur in the percentage improvements in VO2max values of different people. The same principle is also thought to apply to other physical activity-related phenotypes, including differences in response of the various risk factors for cardiovascular disease and type 2 diabetes.

Moreover, previous studies conducted with identical twins have suggested that heredity plays a major role in determining to what degree the body adapts to an intervention such as an exercise training program. All these data were available by the late 1980s primarily as a result of the research of C. Bouchard and his colleagues at Laval University in Quebec City when the planning for the HERITAGE Family Study began.

The HERITAGE Family Study

The HERITAGE Family Study began to be funded in 1992. The justification for the study was that: a) regular endurance exercise has favorable effects on the risk profile for CVD and type 2 diabetes; b) there are considerable individual differences in the response to regular exercise; and c) genes are thought to play an important role in determining the general benefits accrued from participating in regular physical activity. The main goal of the multicenter project is to study the role of the genotype in the cardiovascular and metabolic responses to aerobic exercise training and the changes brought about by regular exercise for several cardiovascular disease and diabetes risk factors. The research is funded by grants from the National Heart, Lung and Blood Institute of the National Institutes of Health of the US government.

The HERITAGE Family Study consortium is composed of research teams from the following five institutions: Indiana University (James S. Skinner, Ph.D.), University of Minnesota (Arthur S. Leon, M.D.), Texas A&M University (Jack H. Wilmore, Ph.D.), Washington University (D.C. Rao, Ph.D.) and the Pennington Biomedical Research Center (Claude Bouchard, Ph.D.). The consortium was put together by C. Bouchard who was then on the faculty at Laval University in Quebec City. The Project Director during the first phase was Jacques Gagnon, Ph.D. When the coordination of the Consortium moved to the Pennington Biomedical Research Center, Tuomo Rankinen, Ph.D., became Project Director.
Phase 1 of the HERITAGE Family Study (1992 - 1997) was devoted to developing a high-quality database. During this phase, 742 healthy, sedentary subjects (483 Whites and 259 Blacks), 17 to 65 years old, were recruited, tested, exercise-trained in the laboratory under supervision with the same program for 20 weeks, and re-tested. These subjects came from families of White descent with both parents and biological adult offspring and from families or pairs of first-degree relatives of Black ancestry.

Responses to submaximal and maximal exercise of such cardiovascular and metabolic variables as oxygen uptake, blood pressure, heart rate, stroke volume, and cardiac output, as well as blood lactate, glucose, glycerol and free-fatty acids, were measured before and after training. Plasma lipids, lipoproteins and apolipoproteins, and insulin, glucose and C-peptide responses to an intravenous glucose load, plasma sex steroids and glucocorticoids, resting blood pressure, and body fat and regional fat distribution (including abdominal visceral fat) were also assessed both before and after training. Dietary intake, smoking habits, level of habitual physical activity and other lifestyle components were assessed by questionnaires. Blood was collected to establish a permanent cell line for each subject.

The focus of Phase 2 (1997-2001) was on data analysis and a series of molecular genetic studies. Data analysis and publication projects have focused primarily on: a) nongenetic determinants of cardiorespiratory endurance and CVD and type 2 diabetes risk factors and their responses to the endurance exercise program; b) genetic epidemiology issues pertaining to exercise phenotypes and CVD and type 2 diabetes risk factors. On the other hand, the molecular studies involve candidate gene investigations, differential display of skeletal muscle transcripts to identify new candidate genes and a genome wide scan to support the genetic dissection of the response to the endurance exercise program for the HERITAGE phenotypes.

Phase 3 (2001-2004) of the HERITAGE Family Study started September 1, 2001. The main focus of Phase 3 is to expand and further refine the search for genes and mutations affecting cardiorespiratory endurance and cardiovascular disease and type 2 diabetes risk factors as well as their response to regular exercise.


As of September 1, 2004, there have beenmore than 120 publications in refereed journals. Because of the complexity and extent of the data obtained, it is not possible to summarize all or even a large part of the results. Nevertheless, results from one phenotype (VO2max ) should demonstrate the types and importance of findings in the HERITAGE Family Study.

One question asked was whether families had similar levels before training began. Relative to the phenotype of VO2max adjusted for age, sex, body mass and body composition, genetic factors explained about 40% of the variation. There was a large variation in response to training at each of the centers. Although the average increase in VO2max was 19%, about 5% of the subjects had little or no change (<5%) and about 5% had an increase of 40 to >50%. This large variation occurred at all ages and at all levels of initial fitness and was similar for Blacks and Whites and for women and men. There was about 2.5 times more variance between families than within families for the gains in VO2max, and the maximal heritability reached 47%. However, there was no relationship between initial level of VO2max and the change in VO2max after training (r = 0.08). In other words, it appears that one set of genes influenced the initial level and another set of genes influenced the response to training. Hence, part of the genetic component for VO2max expresses itself only in response to an active lifestyle. We could not find non-genetic variables measured before training that would differentiate between responders and non-responders.

Because DNA is available from all subjects, studies have been undertaken to identify genes and mutations that are associated with the variability in responsiveness to training as well as VO2max in the sedentary state. For instance, several candidate genes have been investigated and a few have yielded positive results. One of them was the muscle form of the creatine kinase gene (CKM). A DNA sequence variant in this gene was shown to be associated with the trainability of VO2max. For example, about 30% of the low responders (the lowest decile of delta VO2max) were homozygotes for this variant while there were no homozygotes among the high responders (upper decile). A genomic scan study performed with highly polymorphic markers has yielded a few promising chromosomal regions that may harbor genes for VO2max in the sedentary state or its trainability. These chromosomal regions are being further investigated with denser sets of markers. Similar projects are being pursued for all HERITAGE phenotypes including blood pressure, lipids and lipoproteins, glucose and insulin phenotypes, abdominal visceral fat, stroke volume, cardiac output and others.

Results from the HERITAGE Family Study have important implications for understanding human variation in a variety of physiological variables and risk factors for common chronic diseases. In addition, because of the large number of subjects who completed the same standardized training program, data can be analyzed for the effects of or interactions with sex, race, age and initial levels of fitness.