DNA methylation provides an essential epigenetic mechanism that influences a wide range of biological functions including transcription activity, chromatin organisation, imprinting, cellular differentiation, and chromosome stability. Previous studies have established that the patterns of DNA methylation at specific genetic loci display tissue [12
] and gender differences [13
] and can be modified by environmental influences such as diet [44
], famine [49
], or exposure to sun, tobacco, alcohol, arsenic, and asbestos [11
]. In animal studies, maternal behaviour involving high levels of maternal licking and grooming of pups is associated with changes in CpG methylation in the pups [50
]. It has also been reported that DNA methylation profiles can change with age [9
In this study, we have investigated whether tai chi practice is associated with changes in DNA methylation and, if so, whether the changes are beneficial (as defined by the slowing of age-related changes). To address these questions, we have determined and compared the methylation profiles of 60 CpGs from six different chromosomes between female tai chi practitioners and the matched controls. We have identified six CpG marks originating from chromosomes 5, 17, and X that show a significant difference (P
= 0.001–0.031) between the two cohorts. Four of these CpGs show losses in DNA methylation while the remaining two demonstrate gains in DNA methylation with age in the controls. The results indicate that the two adjacently located CpGs (G6PD_6 and G6PD_7, 20
bp apart; see ) within the promoter region of the G6PD gene both give decreasing normal DNA methylation trends with age. In contrast, within the Rad50 gene promoter region, while a decreasing normal methylation trend is seen for the Rad50_2, an increasing trend is seen for the Rad50_10 situated 147
bp downstream (). Previous studies have described variability in DNA methylation of distinct CpG marks that are situated in very close proximity [51
Notably, for all six CpGs, we have observed a slowing of the age-related trends for DNA methylation losses or gains in the tai chi cohort compared to the control cohort. The maximum amount of slowing ranged from approximately 5% to 70% for the different marks, with the Rad50_10 mark from the promoter region of the DNA double-strand break repair protein gene Rad50 [53
] showing the greatest effect of slowing. There also appear to be two noticeable timing patterns with respect to when a CpG shows a response to the effects of tai chi. For Rad50_2 and Xp13_1, the effects of tai chi are detectable from (and presumably before) 45 years of age and thereafter across the 45–88 age range. For the remaining four marks, significant methylation differences between the two cohorts do not become apparent until after 50–55 years of age. One possible explanation for this observation is that tai chi has no effect on the methylation profiles of these four CpGs until after this age. An alternative and tantalising explanation is that a significant deterioration of the normal DNA methylation profiles for these four CpGs does not occur until after the postmenopausal age of 50–55 years and that tai chi is effective in slowing this deterioration when it occurs.
Other studies have described nonuniform changes of DNA methylation in different CpGs over time [9
]. The changes can result in, or increase the risk of, disease such as Beckwith-Wiedemann, Rett, and ICF (Immunodeficiency, Centromere instability and Facial anomalies) syndromes and cancers [54
]. Age-related decline of DNA methylation reflects the gradual deterioration of important epigenetic regulatory functions and genetic control, providing a mechanism for late onset of common diseases including cancer [9
]. Our observed slower cross-sectional decline in the DNA methylation of all six CpGs in the tai chi cohort compared to those of the controls suggests that tai chi practice has a beneficial effect in protecting against the decay of epigenetic functions with age.
How tai chi brings about such a protection is unclear. Numerous studies have pointed to a positive effect of tai chi on a significant number of physical, physiological, and psychosocial functions including balance control, flexibility, cardiovascular fitness, falls in the elderly, musculoskeletal conditions such as osteoarthritis and rheumatoid arthritis, hypertension, bone mineral density, neuromuscular function, the cardiovascular, respiratory, endocrine, and immune systems, psychological responses including depression and perceived health status, and quality of life including activity tolerance, pain management, kinaesthetic sense, sleep quality, and stress reduction [1
]. Studies have also shown the benefit of tai chi on bone mineral density and neuromuscular function in postmenopausal women [25
]. Where tai chi was compared to other forms of exercise, such as stretching and/or resistance-training exercises used in the fibromyalgia and Parkinson's Disease studies [5
], significantly better health outcomes were reported for tai chi; however, a systematic comparison of tai chi and other forms of exercise has not been done. Our study was not designed to answer the question of whether general exercise is related to epigenetic changes. We collected data on exercise to adjust for possible confounding. As indicates, a similar proportion of tai chi and control participants (86% and 85%, resp.; P
= 0.99) regularly performed non-tai chi exercise. There was a range of non-tai chi exercise being done by participants and we do not have enough details about types and intensities of exercise to further analyse whether general exercise is related to epigenetic changes. Our hypothesis related specifically to mind-body exercise, that is, tai chi.
As a complex intervention tool, tai chi is thought to impart its beneficial effects through the integrated improvement of the well-being of the individual at the physical, psychosocial, physiological, emotional, behavioural, and spiritual levels. As the results of this preliminary study suggest, such a holistic intervention approach may, at a more fundamental level, impact directly on the epigenetic and molecular wellbeing of the cells.
This study has a number of limitations. As we performed simultaneous statistical testing on multiple CpGs, the chance of false positives arising is expected to increase. However, the consistently positive effects of tai chi—seen in small to moderate DNA methylation change over time—for all six marks showing a significant difference between the two cohorts lend support that the changes observed reflect a true, positive biological benefit of the tai chi practice. The nonrandomised design of this study also makes it more difficult to control for the effects of confounders. These effects have been minimised by the well-matched general characteristics of the two cohorts and by our statistical model including potential confounders that may be related to DNA methylation. However, further randomised studies would be useful. This study investigated loci-specific CpGs changes. Further work will benefit from looking at a genome-wide DNA methylation profile of tai chi. In our study, we chose “beginners” versus “3 or more years” to try to ensure that we have two sufficiently distinct groups—one with significant tai chi experience and one with virtually no experience. More precise determination of tai chi experience examining the effects of intensity and duration of tai chi practice on epigenetic outcomes would be of interest. In addition, it would be informative to investigate other parameters such as tissue types, younger individuals, the males, the effects of short-term practice, and the styles of tai chi and methods of teaching used.