Reply from M. Mori, K. Higuchi and H. Nose

Abstract

We thank G. Atkinson, C. E. Taylor and H. Jones for their interest and Letter to the Editor regarding our recent symposium review (Mori et al. 2009). The authors of the letter emphasize the necessity to control for artifacts, especially for regression to the mean, when examining any individual differences in the response to interventions including exercise, a point that is well taken and appreciated by us. While it is well recognized that exercise decreases the risks, or alleviates the symptoms, of lifestyle-related diseases, elucidation of mechanisms for the inter-individual differences in response to exercise interventions remains an important unsolved issue. Many studies have been done to examine whether the difference in the genetic composition of individuals (gene polymorphism) was involved in the mechanisms. In few cases, however, have the researchers been aware of the importance of the issue raised by the authors. In our review, high-intensity interval walking training had a profound effect on physiological regulation in older individuals, especially in the least fit (Joyner & Nose, 2009; Nose et al. 2009). Therefore, we agree with the authors’ view that the results of genetic association analyses in the population could include some artifacts by regression to the mean and that even though initial values were not statistically different between genotypes, we could not exclude the possibility that the explained variance due to polymorphism appeared small because the effects of the artifact were large. However, in another study using the same experimental cohort, we investigated any influence of vasopressin V1a receptor genotype for the rs1042615 polymorphism on the exercise-induced responses of diastolic blood pressure (DBP) and blood LDL cholesterol level (Masuki et al. 2010). In the study, the difference in the magnitude of response between genotypes remained statistically significant even after analysis of covariance with initial values included as a covariate. Based on this observation, we concluded that the V1a receptor polymorphism might alter the sensitivity of DBP and LDL cholesterol responses to exercise training. Thus, while we agree that many factors could compromise the interpretation, we believe that it should be possible to elucidate the genetic basis of inter-individual variability in the effects of exercise on the alleviation of lifestyle-related diseases by the study appropriately controlling the factors. We presume that the authors might share this goal with us. In this regard, the authors’ letter is appreciated for their scientific advice to all the researchers in exercise science.