This cross-sectional study included 144 relatively healthy, non-diabetic, non-smoking, postmenopausal women who visited the primary healthcare clinic of the Chung-Ang University Hospital between October 2009 and May 2010. None of the participants had menstruated for over a year, and none were being treated with any hormonal therapy. Medical records were used to collect data regarding previous and current medical conditions and medications taken. Individuals with a history of cardiovascular disease, including myocardial infarction, transient ischemic attack, or cerebral infarction, were excluded from the study.

Regular exercise was defined as at least 150 minutes per week of moderate-intensity activity, or an equivalent combination of moderate and vigorous-intensity activity,¹³⁾ over a duration of at least 6 months. The no exercise group comprised the study participants who did not meet these criteria. Alcohol consumption was defined as the consumption of at least 72 g of alcohol per week.¹⁴⁾ This study was approved by the institutional review board of the Chung-Ang University Hospital (IRB number C2009051 [238]). All participants provided written informed consent.

Anthropometric factors, including body weight and height, were obtained by a single, well-trained examiner and were measured to the nearest tenth of the appropriate unit. Weight (kg) and height (cm) were measured using DS 101 (Dongsahn Jenix, Seoul, Korea), and subjects were dressed in light clothing without shoes. Body mass index (BMI) was calculated as weight divided by the square of the height (kg/m²). Blood pressure was measured after resting for 10 minutes in a seated position. Waist circumference (cm) was measured at the midpoint between the superior margin of the iliac crest and the inferior margin of the last rib in a horizontal plane. Waist-to-hip ratio (WHR) was estimated by the InBody 720 body composition analyzer (Biospace, Seoul, Korea).

All blood samples were obtained after 12 hours of overnight fasting. Serum levels of uric acid, glucose, total cholesterol, high-density lipoprotein cholesterol (HDL), low-density lipoprotein cholesterol (LDL), triglycerides, and high-sensitivity C-reactive protein (hs-CRP) were measured using the ADVIA 1650 chemistry analyzer (Siemens, Munich, Germany). Levels of fasting insulin were measured using an electrochemiluminescence immunoassay system according to the supplier's protocol (Roche, Indianapolis, IN, USA). Insulin resistance was estimated using the homeostasis model assessment of insulin resistance (HOMA-IR) index, defined as follows:

Plasma concentrations of adiponectin and homocysteine were measured using either AdipoGen (Seoul, Korea) or Immulite 2000 (Siemens) immunoassay kits, respectively.

mtDNA was extracted from the peripheral leukocytes in 1 mL of whole blood using a QIAamp DNA kit (Qiagen, Valencia, CA, USA). The mtDNA copy number was determined by real-time polymerase chain reaction, using the LightCycler FastStart DNA Master SYBR Green I kit according to the manufacturer's instructions (Roche, Pleasanton, CA, USA). The quantity of mtDNA was normalized to the expression of nuclear β-globin. Forward and reverse primers for the β-globin gene were 5′-GAAGAGCCAAGGACAGGTAC-3′ and 5′-CAACTTCATCCACGTTCACC-3′, respectively, and forward and reverse primers for the mitochondrial ND1 gene were 5′-AACATACCCATGGCCAACCT-3′ and 5′-AGCGAAGGGTTGTAGTAGCCC-3′, respectively. After denaturation at 95℃ for 300 seconds, samples were subjected to 40 cycles of incubation at 95℃ for 0.1 seconds, 58℃ for 6 seconds, and 72℃ for 18 seconds. The threshold cycle number (Ct) was defined as the number of polymerase chain reaction cycles needed to produce 20 ng of DNA product. The mtDNA copy number was calculated using the following equation: relative copy number=2^ΔCt(ΔCt=Ct_β-globin−Ct_ND1).

Data are presented as mean±standard deviation or number (n, %). The significance of any differences in baseline characteristics between the two groups were assessed using either a t-test or a chi-square or Fisher's exact test, for continuous variables or categorical variables, respectively. The leukocyte mtDNA copy numbers, the concentrations of adiponectin, hs-CRP, insulin, and homocysteine, and the HOMA-IR values were logarithmically transformed to approximate a normal distribution before statistical analyses were performed. An independent association between leukocyte mtDNA copy number and exercise was confirmed using a stepwise multiple linear regression model with a significance threshold of 0.15. Potential confounding variables were defined as those that were significantly associated (P<0.1) with leukocyte mtDNA copy number using Pearson's correlation analysis, either in this study (triglyceride, WHR, waist circumference, and homocysteine level), or in previously reported studies (age,¹⁵⁾ BMI,¹⁶⁾ systolic and diastolic blood pressure, HOMA-IR, and levels of triglyceride and HDL cholesterol^17,18)). Multicollinearity between independent variables was evaluated using Pearson's correlation coefficients. If baseline variables were correlated (r>0.6 and P<0.05), only one variable was included in the analysis; for instance, as WHR and waist circumference showed multicollinearity, waist circumference was excluded from the analysis. The leukocyte mtDNA copy numbers of the exercise and no exercise groups, presented as the least squares mean±standard error, were compared using the analysis of covariance model while adjusting for age and WHR. All statistical analyses were performed using the SAS ver. 9.1 statistics package (SAS Institute Inc., Cary, NC, USA).

The mean age of the participants was 57.4±6.1 years, and ranged from 50 to 72 years. The clinical characteristics of the study participants in the no exercise and regular exercise groups are shown in Table 1. Waist circumference, WHR, and triglyceride level were all significantly higher in the no exercise group than in the regular exercise group (P<0.01). Conversely, no significant differences in age, BMI, blood pressure, alcohol consumption, antihypertensive medication use, or concentrations of fasting glucose, total cholesterol, HDL cholesterol, LDL cholesterol, uric acid, log-transformed hs-CRP, homocysteine, adiponectin, or fasting insulin, HOMA-IR were observed between the groups.

Peripheral leukocyte mtDNA copy number was significantly greater in the regular exercise group (1.33±0.02) than in the no exercise group (1.05±0.02, P<0.01) (Figure 1).

Table 2 shows the independent association between exercise and mtDNA copy number in peripheral leukocytes. Stepwise multiple regression analysis showed that regular exercise was independently associated with mtDNA copy number (β=0.25, P<0.01) after adjusting for age, BMI, WHR, systolic and diastolic blood pressure, HDL cholesterol level, triglyceride level, HOMA-IR, and homocysteine level.