Introduction
The world’s ageing population continues to increase with the number of persons aged 60 years and over expected to exceed the number of children in the world by 2045 [
1]. Increased longevity is supporting marked growth in the proportion of adults aged over 85 years [
2]. As chronic disease burden increases with age, it is important that health and function are maintained to complement increased longevity. Diet plays a key role in maintaining health in adulthood and may impact on markers of cellular ageing such as telomere length.
Telomeres are repetitive DNA sequences at the ends of eukaryotic chromosomes that undergo attrition each time a somatic cell divides [
3]. They prevent loss of genomic DNA at ends of linear chromosomes and protect their physical integrity with cell division and replication, thereby protecting against cell death. Telomere length decreases with age and is considered a biomarker of accelerated cellular ageing. Shorter telomere length is associated with decreased life expectancy and increased risk of chronic disease [
4] and was recently described as one of the nine “hallmarks of ageing” [
5]. As telomere attrition rate varies between individuals and is increased by inflammation and oxidative stress [
4], it is a proposed modifiable lifestyle risk factor for health and longevity in older age. Although there is some genetic regulation of variability of telomere length, it may also be partially explained by lifestyle behaviours, including dietary intake [
6]. The intake of various individual nutrients including folate, vitamins C, E, D, A, and β-carotene, magnesium and omega-3 polyunsaturated fatty acids is associated with longer telomere length [
6,
7]. However, there has been little investigation into the association between whole diet, or dietary patterns and telomere length.
Whilst previous research into nutrition and health in older age has focussed on the role of individual nutrients or foods, there is increasing interest in dietary pattern analysis as a chronic disease determinant [
8]. Dietary patterns can be defined by two approaches: multivariate statistical techniques such as factor or cluster analysis (data-driven approaches), and dietary scoring methods informed by a priori guidelines and recommendations, or diet quality indices. Diet quality indices can assess adherence to dietary guidelines [
9], or a particular type of diet such as the Mediterranean diet [
10]. Diet quality assessed by adherence to dietary guidelines has been associated with cardiometabolic risk factors [
11], and physical and mental health [
12,
13], whilst diet quality assessed by adherence to a Mediterranean diet and higher dietary variety has been associated with reduced risk of mortality [
14] in older people.
Despite the wide variety of diet quality indices available, few studies have investigated dietary patterns and telomere length in older people [
15,
16], and only one has included multiple indices of diet quality [
17]. Considering the limited research to date, assessing multiple indices of diet quality may give greater insights into which components or aspects of diet quality have greatest influence on telomere length. The aim of this study was to examine associations between three measures of diet quality and relative telomere length in men and women aged 57 years and over.
Discussion
In this study of 679 community-dwelling older men and women from the WELL study, greater diet quality assessed via the DGI-2013, RFS, and MDS was not cross-sectionally associated with longer relative telomere length. Previous studies have reported cross-sectional associations between higher intakes of vegetables and lower intakes of saturated fat and sugar-sweetened soda and longer telomere length [
40,
41]. Reported associations between diet quality and telomere length have been limited, with only one study investigating associations between multiple dietary indices and telomere length. In 4676 women, aged 42–70 years from the Nurses’ Health Study, higher scores on the Mediterranean diet and Alternate Healthy Eating Index were associated with longer telomere length. However, no associations between prudent and Western dietary patterns and telomere length were observed [
17]. A cross-sectional study of 217 men and women aged 71–87 years reported that a greater adherence to a Mediterranean diet was associated with longer leucocyte telomere length and higher telomerase activity, a relationship moderated by inflammation [
15]. Following a 5-year Mediterranean diet intervention, an association between TL and changes in adiposity indices was observed in 521 men and women aged 55–80 years [
42]. Furthermore, in the same cohort, telomere shortening appeared to be attenuated with greater adherence to the Mediterranean diet in people carrying the Pro12Ala polymorphism in the peroxisome proliferator-activated receptor gamma 2 (
PPARy2) gene [
43].
Studies investigating dietary patterns and telomere length longitudinally have been limited to date. A randomised crossover intervention of 20 older adults found reduced telomere shortening following a Mediterranean diet for 4 weeks compared to a saturated fatty acid diet and low-fat high-carbohydrate diet [
44]. More recently, a randomised intervention trial of the Mediterranean diet in 520 participants aged 55–80 years reported greater adherence to a Mediterranean diet was associated longer telomeres in women only at baseline [
45]. However, no beneficial effects for the Mediterranean diet on telomere length were observed at 5-year follow-up, and a higher risk of telomere shortening was observed in the Mediterranean diet and nuts group.
Associations between inflammatory dietary patterns derived by principal component analysis and leucocyte telomere length in 840 men and women aged 45–84 years have been investigated [
16]. No relationship between telomere length and a “fats and processed meat” pattern or a “whole grains and fruit” pattern was observed, despite these patterns being associated with C-reactive protein, interleukin-6, and homocysteine [
16]. However, an association between intake of processed meat and shorter telomeres was reported. A recent Mediterranean diet intervention study of 520 participants at high cardiovascular risk reported a pro-inflammatory diet characterised by a dietary inflammatory index was associated with shorter telomeres cross-sectionally and after a 5-year follow-up [
46].
Previous studies have also investigated associations between healthy lifestyle and telomere length. In 612 patients with advanced prostate cancer and 1049 matched controls, the sum score of a healthier lifestyle and diet factors (low or no cigarette use, higher fruit and vegetable intake, lower BMI, more physical activity), correlated with longer telomere length; no correlations between telomere length were observed when these factors were analysed individually [
47]. In patients with low-risk prostate cancer, a comprehensive lifestyle intervention (changes to diet, physical activity, stress management, and social support) resulted in increased telomere length and telomerase activity, when comparing baseline and 5-year follow-up values. Adherence to the lifestyle changes was also significantly associated with relative telomere length when adjusted for age and length of follow-up [
48]. Similarly, in a cohort of 5862 women (aged mean 58.7 ± 0.09 years), increased telomere length was associated with adherence to a healthy lifestyle; the later defined using five components including smoking, physical activity, adiposity, alcohol use, and diet [
49].
Previous studies have reported inverse correlations between age and telomere length [
15‐
17,
50,
51]. However, no association was observed between age and telomere length in the current study. The age range of our sample of 57–68 years was considerably narrower than that of previous studies, with reported age ranges of 42–70 [
17], 71–87 [
15], 45–84 [
16], 1–96 [
50], and 18–90 [
51] years. In the present study, the
T/
S values ranged between 0.43 and 1.40 which are similar to values observed previously within the same age range [
50]. Our lack of observation of an association between age and telomere length is most likely due to the small age range. This is not without precedence and there appears to be limited correlation between age and telomere length when analysed within narrow age ranges of 40–44 and 60–64 years [
52].
Diet quality has been shown to be associated with mortality [
14] and reduced risk of chronic diseases such as cardiovascular disease and diabetes [
53], all outcomes that have been associated with shortening of telomeres [
4]. Previously, it has been proposed that consumption of a healthier diet would protect against age-related telomere loss through reduced oxidative stress and inflammation [
6]. The DGI was associated with lower intakes of energy, total fat and saturated fat, and higher intakes of fibre, β-carotene, vitamin C, folate, calcium, and iron previously [
9]. The RFS was associated with intake of similar nutrients and antioxidants and also reduced markers of chronic disease and inflammation, including serum homocysteine, C-reactive protein, plasma glucose, total serum cholesterol, and blood pressure in previous research [
31]. The Mediterranean diet is also proposed to reduce disease risk through its anti-inflammatory and anti-oxidative properties [
54]. In summary, although previous research demonstrates the potential for all three indices used in the current study to assess long-term adherence to diets associated with reduced inflammation and oxidative stress, no significant associations with telomere length were observed.
These findings are generalisable to community-dwelling men and women around the “peri-retirement” stage. However, it should be noted that WELL study participants reported higher scores on the RAND 36-item survey [
13] compared to other Australian population-based samples [
55,
56], and may therefore represent a group with better health status than the Australian population. The samples are also not reflective of the oldest old or specific clinical populations, and therefore, these findings may not be applicable across the older age range in a broader sense. However, there are challenges to assessing telomere length in older age, as the upper age limit for reliable assessment of telomere length is 75 years [
57]. Telomere length instability and “survivor bias”, where people who live longer are suspected to possess longer telomeres and some resilience to chronic disease, may contribute to a lack of association between telomere length and mortality observed above this age [
57,
58]. However, this is unlikely to have played a role in the lack of associations observed in the current study.
The strengths of our study include the well-characterised sample, detailed assessment of dietary intake and potential confounders and the use of three validated diet quality indices. However, the study has some limitations that should be considered. Only a subsample of the original cohort provided a blood sample, which may have created a selection bias. Dietary intake was assessed by a non-quantified FFQ, which did not allow for adjustment of energy intake in statistical analysis. However, key determinants of energy intake were taken into account by adjusting for sex, BMI, and physical activity. Although a range of potential covariates were tested for inclusion in the regression models, it is possible some residual confounding remained. Finally, the relatively modest sample size may have limited our ability to detect the small associations reported by previous studies, although this is unlikely since some of those studies were conducted in a sample of similar size to the current study [
15,
17].
A potential limitation of studies investigating the Mediterranean diet in non-Mediterranean samples is the lack of foods and eating behaviours truly reflective of a traditional Mediterranean diet. However, studies in Australian [
59,
60] and UK samples [
61] have reported the presence of a Mediterranean-style pattern. Our choice of Mediterranean Diet Index included sample-specific median intakes of vegetables, legumes, fruits and nuts, cereals, and fish as cut points for scoring. Although this enables adaptation of the score for a non-Mediterranean population, it likely reflects a dietary pattern with substantially lower absolute intakes of these foods compared to typical of Mediterranean populations as supported by a recent Australian study [
62]. An olive oil or monounsaturated fat component was not included in the MDS in this study due to limitations in the FFQ. However, some controversy exists regarding the inclusion of this item in Mediterranean diet assessment in non-Mediterranean populations, as olive oil consumption is typically low and monounsaturated-to-saturated fat ratio instead reflects animal fat intake, which is low in the typical Mediterranean diet [
63].
Although telomere length is thought to vary depending on exposure to environmental factors, telomere length is also determined by genetics [
64] and shows inter-individual variation from birth [
65]. Thus, attempts to determine the association between diet and telomere attrition should ideally measure within-person change in telomere length over long periods of time. The current study was cross-sectional and was therefore unable to assess within-person change in telomere length. This may have limited the ability to detect significant associations in the current study. It should be noted that all previous studies investigating dietary patterns and telomere length have also been cross-sectional. Therefore, inconsistencies in the evidence to date including the current study may be related to the fact that telomere length shows inter-individual variation and telomere length at any age relates to a combination of telomere length at birth and subsequent rate of change [
66]. Considering the variation in findings of this and previous studies, it is critical that future studies are longitudinal in study design and include assessment of telomere length over time to determine a true measure of biological ageing.
Conclusion
In a sample of adults residing in Victoria, Australia, men and women aged 57–68 years with better-quality diets did not have longer telomeres. Further investigation of telomere length over time in longitudinal studies will determine whether dietary patterns are associated with cellular ageing in an older population.