Telomere length (TL) is a specific DNA-protein complex of repetitive hexanucleotides at the end of chromosomes. Sufficient TL is necessary for complete replication of DNA. It plays a key role in promoting the integrity and stability of chromosomes by protecting them from nuclease degradation, end-to-end fusion, and cellular senescence.¹

In normal cell processes, TL gradually decreases with each subsequent mitotic cell division due to incomplete replication of the lagging strand during partial DNA synthesis. When the TL becomes shorter, either cell growth is inhibited or cell death is encouraged by a cell-replication aging signal.² Human peripheral leukocyte telomeres gradually shorten with age. Because age and TL are inversely related, TL has been suggested as an indicator of biological aging.²

In South Korea, the prevalence of obesity has increased over the last decade, from 29.7% in 2009 to 36.3% in 2019. The obesity rate of women is 27.3%, that of men has increased to 46.2%. The obesity rates of men in their 20s, 30s, 40s, 50s, and 60s were 28.4%, 52.2%, 50.8%, 46.1%, and 42.4%, respectively, indicating that obesity was most prevalent in men in their 30s.³

Obesity is a major risk factor for various chronic diseases related to aging, including cardiovascular disease (CVD), diabetes, and specific cancers, making it a major cause of preventable deaths worldwide.⁴ Multiple previous studies have reported that obesity increases systemic inflammation and oxidative stress⁵ and promotes age-related diseases by causing various dysfunctions in metabolism and immunity.⁶ Because inflammation and oxidative stress are major causes of aging and aging-related diseases and play essential roles in telomere attrition,⁷ obesity, which is a condition of chronic inflammation and oxidative stress, is related to TL. However, the correlation between obesity and telomeres remains unclear; although body mass index (BMI) and TL are inversely correlated,⁸ as are current weight and TL,^9,10 obesity and TL are unrelated.^11-13

This inconsistency with previous studies can be attributed to the fact that the relationship between obesity and TL is also influenced by the presence or absence of disease. That is, obesity is a risk factor for metabolic syndrome or CVD, and people with CVD have shorter TLs compared with healthy people.^14,15 TL has also been reported to be inversely correlated with CVD risk factors such as age, BMI, waist circumference (WC), hypertension, triglyceride (TG) levels, and fasting blood glucose (FBG) levels.¹³

Regular exercise reportedly decreases the risk of obesity, hypertension, type 2 diabetes mellitus, and CVD. Similarly, prolonged exercise and high cardiorespiratory fitness (CRF; as measured by the maximum amount of oxygen a body can use during a specified period of intense exercise, or maximum rate of oxygen utilization [VO₂max]) are correlated with good health and high survival. Physically healthy individuals had longer TLs compared with those of the same age group,¹⁶ and physically fit elders had longer TLs compared with those of the same age group.¹³ However, self-reporting physical activity questionnaires can be subject to recall bias. Obese people in particular often overestimate the amount of exercise they engage in.¹⁷

In contrast to the problems posed by self-reported questionnaires of physical activity, CRF is an objective and reproducible metric that can measured in a laboratory. Although CRF is influenced by age, sex, health, and genes, it reportedly reflects actual patterns of physical activities and is negatively correlated with CVD and mortality.¹⁸ A similar correlation was also observed among obese but fit people (those with high CRF). The Aerobic Center Longitudinal Study, which involved 21,925 men aged 30 to 83 years, found that the mortality rate of obese but fit men was lower than that of lean but unfit men.¹⁹ However, inconsistent results regarding the correlation between CRF and TL have been reported. For example, several studies have reported that exercise has a protective effect on TL,^16,20-22 while others found no such correlation.^23,24 This is because, although TL decreases over time, it is stable from childhood to young adulthood and begins to shorten only in later adulthood, and large differences among individuals have been reporter.²⁵ The correlation between CRF and TL varies with age: CRF and TL were unrelated in a group of 15-year-old²⁶ and a young group (18 to 32 years of age).^22,27 Differences in TL by CRF have been reported only in a middle-aged group (55 to 72 years).²²

To determine the correlation between TL and obesity, fitness level and CVD risk factors should be analyzed together. However, it is necessary to distinguish among age groups because the correlation between BMI and TL length differs by sex,^9,28 and the correlation between CRF and TL length also differs by age.

This study aimed to examine differences in TL length by obesity, CVD risk factors, and fitness levels in an adolescent group (10 to 20 years), a young-adult group (21 to 34 years), and a middle-adulthood group (>35 years) of male South Koreans among whom obesity rates were rising.

Subjects

The subjects of this study were males between the ages of 10 and 59 years (n=249). Each was assigned to one of three groups according to age (adolescents, young adults, and middle-aged), with no history of musculoskeletal injuries within the past year, functional limitations, or history of cancer. Recruitment was carried out by advertising in local newspapers and social media. In this study, only males were included because women are affected by menopause.

Measurement of obesity indices

The height, weight, BMI, muscle mass, and % body fat of the participants were measured using multi-frequency bioelectrical impedance analysis (InBody 770; Biospace Co. Ltd.). WC (cm) was measured between the ribs and iliac crest after the end of normal exhalation.

Analysis of CVD risk factors

For blood tests, 10 mL of blood was drawn from a forearm vein at 9:00 AM in the morning after the subjects had fasted for 12 hours. The blood samples were centrifuged at 3,000 rpm for 10 minutes using a centrifuge and analyzed using plasma.

CVD risk factors were selected and added by referring to the items proposed by the World Health Organization,²⁹ including age, sex, BMI, smoking, systolic blood pressure (SBP), the presence or absence of diabetes, total cholesterol (TC), and by the National Cholesterol Education Program Adult Treatment Panel III,³⁰ including age, family history of premature coronary heart disease, smoking, hypertension, low high-density lipoprotein cholesterol (HDL-C), high TC, low-density lipoprotein cholesterol (LDL-C), and diabetes mellitus. The disease diagnosis criteria according to risk factors were as follows. An SBP of 140 mmHg or higher or a diastolic blood pressure (DBP) of 90 mmHg or higher was classified as hypertension, and an FBG level of 126 mg/dL or higher or was diagnosed as diabetes. A TC level of 240 mg/dL or a TG level of 200 mg/dL or a LDL-C level of 160 mg/dL or higher or a HDL-C level of <40 mg/dL was diagnosed as dyslipidemia. Patients who have been taking medicines for hypertension, diabetes, or dyslipidemia were was defined as having a comorbidity. Overweight status and obesity were classified by BMI: those with a BMI of 23 to 24.9 kg/m² were classified as overweight, and those with a BMI of 25 kg/m² or higher were classified as obese. In addition, subjects with a WC of 90 cm or higher³ were classified as obese to calculate the number of CVD risk factors. Blood sugar was analyzed using the glucose oxidase method (747 automatic analyzer; Hitachi), and TC, TG, LDL-C, and HDL-C levels were analyzed with an automatic analyzer (COBAS integra 800; Roche) using enzymatic colorimetry. The blood pressure at rest was measured on the left upper arm using TENSOVAL (Hartmann), and the average of two measurements was used.

Analysis of telomere length

Leukocyte telomere length (LTL) was analyzed using the principle of repeated replication of the DNA template using DNA polymerase. For quantification, the original DNA length was first measured by obtaining the relative DNA repeat copy number for the single-copy gene number after treating the DNA with fluorescence. To quantify LTL, the relative telomere-repeat copy number for the single-copy gene number (T/S ratio) was determined using a method described by Cawthon³¹ that incorporated real-time polymerase chain reaction (PCR) amplification. To quantify real-time PCR, two DNA samples were diluted at five ratios, and a standard curve was determined. The same DNA sample was analyzed three times in the same well position. After the melt and standard curves were drawn, the length of the reference DNA was analyzed by obtaining the correlation coefficient of the mean terminal restriction fragment lengths of DNA according to the diluted concentration.

Measurement of cardiorespiratory fitness

The maximum oxygen intake test for CRF was conducted using the Bruce protocol (1973) and a gas analyzer (Vmax ST 1.0; Quinn Company). Each subject’s resting condition was verified by measuring the resting heart rate and respiratory rate before taking the test. The maximum CRF was set to the case where the VO2 or heart rate of subjects remained at a steady state without increasing even when the exercise load was increased or no more exercise could be performed due to fatigue.³² The subjects’ CRF levels were classified by the age criteria set by the American College of Sports Medicine³² (high, medium, and unfit).

Data processing

SPSS WIN version 23.0 (IBM Co.) was used for data analysis, and the mean and standard deviation were calculated for each measurement item. One-way analysis of variance was used to examine differences in body composition, CVD risk factors, CRF, and LTL by group. The correlations of obesity, CVD risk factors, and fitness level with LTL were analyzed using the Pearson correlation. In addition, regression analysis was performed to analyze the influence of obesity, CRF, and CVD risk factors on LTL, and all data were adjusted according to the variables, age, and the presence or absence of disease. The statistical significance of the differences was set to α=0.05.

Ethical approval and consent to participate

Research involving human subjects, human material, or human data must be performed in accordance with the Declaration of Helsinki. The subjects were sufficiently briefed on the purpose and proceedings of the study and voluntarily submitted their written consent before participating in this study. This study was approved by Institutional Review Board of Dankook University (DKU 2021-11-006).

Characteristics of study participants

The characteristics of the study participants are summarized in Table 1. Significant differences were found among the LTL, obesity indices, CVD risk factors, and VO₂max according to age. The LTLs of the adolescents were longer than and those of the young adults and the middle-adulthood group (P<0.001), and the LTLs of the young adults were longer than the middle-adulthood group (P<0.01). The obesity indices (BMI, % fat, and WC) increased with age (P<0.001), as did all the CVD risk factors (DBP, TC, TG, HDL-C, LDL-C, and FBG [P<0.001]). However, VO₂max levels decreased with age (P<0.001) and the muscle mass of the adolescents was lower than that of young adults (P<0.05).

Telomere length according to related factors for obesity, CVD risk factors, and VO₂max

Values for LTL according to obesity, CVD risk factors, and CRF are summarized in Table 2. Members of the non-obese group had a longer LTL, with a T/S ratio of 1.3±0.2, compared with the obesity group’s T/S ratio of 1.1±0.2 (P<0.001). With hypertension, the non-hypertension group had a longer LTL, with a T/S ratio of 1.2±0.2, compared with the hypertension group’s T/S ratio of 1.1±0.2 (P<0.001). The non-diabetes group had a longer LTL, with a T/S ratio of 1.2±0.2 compared with the diabetes group’s T/S ratio of 1.1±0.2 (P<0.01). Similarly, the non-dyslipidemia group had a longer LTL, with a T/S ratio of 1.3±0.2, compared with the dyslipidemia group’s T/S ratio of 1.1±0.2 (P<0.001). In terms of VO₂max, the unfit, medium-fit, and high-fit groups had T/S ratios of 1.1±0.2, 1.2±0.2, and 1.3±0.2, respectively (P<0.001)—the higher the VO₂max, the higher the LTL.

Correlation analysis between telomere length and related factors

The results of the analysis of the correlations of LTL with obesity, CVD risk factors, and CRF are summarized in Table 3. LTL showed statistically significant negative correlations with obesity indices, including BMI, % fat, and WC (P<0.001), and with CVD risk factors, including DBP (P<0.001), TC (P<0.001), TG (P<0.001), LDL-C (P<0.001), and FBG (P<0.05). In contrast, LTL showed statistically significant positive correlations with HDL-C and VO₂max (P<0.001). BMI also showed statistically significant positive correlations with CVD risk factors, including SBP, DBP, TC, TG, LDL-C, and FBG (P<0.001), and VO₂max (P<0.001), but a statistically significant negative correlation with HDL-C (P<0.001). VO₂max showed statistically significant negative correlations with age (P<0.001) and obesity indicators, including BMI, % fat, and WC (P<0.001), and with CVD risk factors, including SBP (P<0.01), DBP (P<0.001), TC (P<0.001), TG (P<0.001), LDL-C (P<0.001), and FBG (P<0.001), but a statistically significant positive correlation with HDL-C (P<0.001).

The results of the analysis of the regression between LTL and related variables are summarized in Table 4. With the exception of muscle mass, the factors of age, obesity index, CVD risk factors, and VO₂max were all correlated with LTL (P<0.01). With the exception of SBP, LTL decreased in each increment of age, obesity indices, CVD risk factors, and CRF. After the values were adjusted according to all variables, LTL statistically significantly decreased by a T/S ratio of −0.006 when the age increased by 1 year. As for the obesity indicators, when CVD risk factors including TC and LDL-C levels, increased by 1 unit, LTL significantly decreased from a T/S ratio of −0.001 to −0.0011. However, when muscle mass increased by 1 kg, LTL significantly increased, with a T/S ration of 0.003. Furthermore, as VO₂max increased by 1 mL/kg/min, LTL significantly increased, with a T/S ratio of 0.004.

The results in terms of age and CVD risk factors that influence LTL adjusted according to obesity and CRF are summarized in Table 5. For the obesity group, LTL became shorter with a T/S ratio of 0.006 for each age increment of 1 year (P<0.001) and by 0.001 when the TC and LDL-C level increased by 1 mg/dL (P<0.01), whereas LTL became longer with T/S ratios of 0.001 and 0.006 when muscle mass increased by 1 kg and VO₂max increased by 1 mL/kg/min (P<0.01), respectively. For the non-obesity group, LTL decreased by a T/S ratio of 0.004 each time the age increased by a step of 1 year (P<0.01). In the unfit group, LTL decreased by a T/S ratio of 0.007 when the age increased by 1 year (P<0.001), and by 0.001 each time the TC and LDL-C level increased by 1 mg/dL (P<0.01). In contrast, LTL increased by T/S ratios of 0.004 and 0.009 when the muscle mass increased by 1 kg (P<0.05) and the VO₂max level increased by 1 mL/kg/min (P<0.01), respectively. For the medium-fit group, LTL became shorter with a T/S ratio of 0.004 each time the WC increased by 1 cm (P<0.05), whereas LTL became longer by T/S ratios of 0.006 and 0.019 when the muscle mass increased by 1 kg (P<0.05) and the VO₂max level increased by 1 mL/kg/min (P<0.001), respectively For the high-fit group, LTL became shorter with a T/S ratio of 0.002 when the SBP increased by 1 mmHg (P<0.05), whereas LTL became longer with T/S ratios of 0.005 and 0.004 when the muscle mass increased by 1 kg (P<0.05) and the VO₂max level increased by 1 mL/kg/min, respectively.

The results of the analysis of the regression of LTL by age according to the presence of absence of disease are summarized in Table 6. For the subjects 20 years of age or younger, disease status did not affect LTL. For subjects 21 to 34 years old, LTL decreased by a T/S ratio of 0.152 for those with dyslipidemia. For subjects 35 years or older, LTL lengthened by a T/S ratio of 0.148 as VO₂max increased by 1 mL/kg/min.

Both LTL and CRF decreased with age, whereas both the levels and numbers of body composition factors and CVD risk factors increased with age. Furthermore, LTL showed differences depending on obesity, dyslipidemia, and CRF levels. When all the variables influencing LTL were adjusted, LTL became shorter as age and TC and LDL-C levels increased, and it became longer as the muscle mass and VO₂max increased.

Excessive fat reportedly decreases LTL by inducing inflammation and oxidative stress. Endocrine organs in adipose tissues accelerate inflammation by directly releasing pro-inflammatory mediators and increasing the level of adipocytokines. However, studies of the correlation between obesity and LTL have produced conflicting results. Valdes et al.¹⁰ reported that obese women had shorter LTLs compared with thin women. Lee et al.⁸ reported that body fat and abdominal fat were inversely correlated with LTL in non-Hispanic whites aged 8 to 80 years. However, Diaz et al.¹² claimed that abdominal fat was not correlated with LTL in men and women aged 40 to 64 years, and that diabetes, CVD, and cancers were not correlated with LTL. Other inconsistent results have been reported, such as the existence of inverse correlations between obesity and body size classified by BMI and LTL^8-10 and some studies have found no correlations.^11-13

However, a recent meta-analysis of 87 studies reported that BMI and LTL were inversely correlated and that classification by age and ethnicity showed a strong inverse correlation between BMI and LTL in a group of people aged 18 to 61 years.³³ This study also found statistically significant negative correlations between LTL and age and obesity indices. An obesity group and a normal-weight group showed a difference in T/S ratios of approximately 0.2. Gielen et al.³³ used cross-sectional data to estimate that LTL in adults decreases by approximately 25 base pairs per year or by a T/S ratio of approximately 0.01 on average each year,^10,34 and this was equal to an increase in biological age of approximately 1 year. Based on this estimation, the subjects of this study showed an increase in biological age of approximately 2 years according to obesity. However, when adjusted for all variables, this difference had no significant effect on BMI. These findings can be attributed to the increase in BMI with age.

A previous study reported that BMI and LTL showed a largely inverse correlation in younger populations, and the correlation decreased in older populations.³⁵ This is because BMI is an accurate indicator of adiposity in those aged 60 years or younger. However, in the group of subjects aged 65 or older, it is difficult to use obesity alone to reflect BMI because muscle mass, bone density, and height also decrease. The most interesting result of this study is that BMI, % fat, and WC all increased with age. However, in a regression analysis adjusted for all variables, BMI, % fat, and WC exhibited no statistically significant effect on LTL. This suggests that, after height growth was completed in subjects through their 10s and 20s, weight increased in later adulthood, which increased their BMI, % fat, and WC. However, no influence was seen on LTL unless there were changes in the other variables. When the effects of obesity on LTL were analyzed after adjusting for age and CVD risk factors, it was found that the LTL of members of the obesity group decreased more than that of those in the non-obesity group. Moreover, as muscle mass and VO₂max increased, LTL in the obesity group increased. Maintaining an appropriate weight and increasing muscle mass and CRF is therefore a possible method of preventing the loss of LTL with age.

Manson et al.³⁶ reported that obesity, as a cause of type 2 diabetes mellitus in individuals in their 50s, can induce myocardial infarction and heart failure, and it is a cause of weakness and muscle wasting due to weight loss in the 60s. In particular, the occurrence of chronic diseases increases with age, and this is related to weight loss in older populations.³⁷ The correlation between obesity indices and LTL has been suggested as a factor that should be considered along with the effect of disease according to age.³⁸ In this study, the results of a regression analysis adjusted for all variables also showed that, for the obesity group, the LTL became shorter when the TC and LDL-C levels increased, whereas for the non-obesity group, the LTL became shorter when the WC increased. LTL is reportedly correlated with atherosclerosis or CVD.^14,39 Congestive heart disease patients have an LTL that is equal to that of a healthy population aged 11 years, which indicates biological aging of the blood vessel walls.³⁹ Therefore, although obesity indices increase and LTL decreases with age, if LDL-C, a main cause of coronary artery disease, is managed in the obese group and abdominal obesity is prevented in the non-obesity group, LTL loss may be avoidable.

In this study, the CRF decreased with age, showing significant differences in the obesity group, and also had a significant negative correlation with BMI. Furthermore, LTL increased with CRF, particularly among those 35 years of age or older. An increase in VO₂max prevented the loss of LTL and resulted in elongation of LTL as subjects aged. This result is consistent with the findings of a study of the correlation between CRF and LTL in 994 patients with coronary artery disease: those with a higher CRF level had a longer LTL, and those with a high CRF had an LTL approximately twice the length of those with a low CRF.⁴⁰ Furthermore, the unfit group showed that LTL shortened as LDL-C levels increased, whereas the CVD risk factors did not influence LTL in the medium-fit and high-fit groups. This suggests that an intermediate or high CRF can lower the effect of CVD risk factors and is correlated with LTL maintenance and elongation.

We found that, although obesity influences LTL by increasing the level of CVD risk factors and decreasing CRF, maintaining a high CRF can decrease the effects of obesity and CVD risk factors according to age and influence the maintenance and elongation of LTL. The primary limitation of this study is that it included only 249 men in their 10s to 50s. This is relatively small number of subjects, and those 65 years or older were not included. However, we discovered a correlation between obesity and LTL in men whose obesity rate was rising by directly measuring the CRF of various age groups, including youth, and analyzing the effects of CVD risk factors related to obesity. These results suggest a need for CRF management, together with preventive efforts to address increasing obesity with age.

Yun-A Shin is an editorial board member of the journal, but she was not involved in peer reviewer selection, evaluation, or decision process for this article. No other potential conflicts of interest relevant to this article were reported.

We thank all the participants in the Aging & Fitness Cohort study. This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF-2020S1A5A2A01040040).

Study concept and design: YAS; analysis and interpretation of data: JHK; drafting of the manuscript: YAS; critical revision of the manuscript: YAS; statistical analysis: JHK; obtained funding: YAS; administrative, technical, or material support: YAS and JHK; and study supervision: YAS.