Minimal Shortening of Leukocyte Telomere Length

Journals of Gerontology: Biological Sciences cite as: J Gerontol A Biol Sci Med Sci, 2018, Vol. 73, No. 11, 1448–1452 doi:10.1093/gerona/gly071 Advance Access publication 21 April 2018

Brief Report

Philip M. C. Davy, PhD,1,† D. Craig Willcox, PhD,2,3,4,† Michio Shimabukuro, MD,5,6,7 Timothy A. Donlon, PhD,8 Trevor Torigoe, BSc,1 Makoto Suzuki, MD, PhD,4 Moritake Higa, MD,7 Hiroaki Masuzaki, MD,9 Masataka Sata, MD,10 Randi Chen, MSc,2 Rachel L. Murkofsky, MD, MPH,11 Brian J. Morris, DSc PhD,2,12 Eunjung Lim, PhD,13 Richard C. Allsopp, PhD,1,14,† and Bradley J. Willcox, MD2,4,14,† 1 Institute for Biogenesis Research, University of Hawaii, Honolulu. 2Department of Geriatric Medicine, John A. Burns School of Medicine, University of Hawaii, Honolulu. 3Department of Human Welfare, Okinawa International University, Ginowan, Japan. 4Okinawa Research Center for Longevity Science, Japan. 5Department of Diabetes, Endocrinology and Metabolism, School of Medicine, Fukushima Medical University, Japan. 6Department of Cardio-Diabetes Medicine, Institute of Biomedical Sciences, Tokushima University Graduate School, Japan. 7 Diabetes and Life-Style Related Disease Center, Tomishiro Central Hospital, Okinawa, Japan. 8Ohana Genetics, Honolulu, Hawaii. 9Division of Endocrinology, Diabetes and Metabolism, Hematology, Rheumatology (Second Department of Internal Medicine), Graduate School of Medicine, University of the Ryukyus, Okinawa, Japan. 10Department of Cardiovascular Medicine, Institute of Biomedical Sciences, Tokushima University Graduate School, Japan. 11Spinal Cord Injury and Disorders Program, VA Pacific Islands Health Care System, Honolulu, Hawaii. 12School of Medical Sciences and Bosch Institute, University of Sydney, New South Wales, Australia. 13Office of Biostatistics and Quantitative Sciences, John A. Burns School of Medicine, University of Hawaii, Honolulu. 14Pacific Health Research and Education Institute, Honolulu, Hawaii.

These authors contributed equally to this work.

†

Address correspondence to: Richard C. Allsopp, PhD, Institute for Biogenesis Research, John A Burns School of Medicine, University of Hawaii, Honolulu, HI 96813, USA. E-mail: [email protected] Received: June 16, 2017; Editorial Decision Date: March 15, 2018 Decision Editor: Rafael de Cabo, PhD

Abstract FOXO3 is one of the most prominent genes demonstrating a consistently reproducible genetic association with human longevity. The mechanisms by which these individual gene variants confer greater organismal lifespan are not well understood. We assessed the effect of longevity-associated FOXO3 alleles on age-related leukocyte telomere dynamics in a cross-sectional study comprised of samples from 121 healthy Okinawan-Japanese donors aged 21–95 years. We found that telomere length for carriers of the longevity associated allele of FOXO3 single nucleotide polymorphism rs2802292 displayed no significant correlation with age, an effect that was most pronounced in older (>50 years of age) participants. This is the first validated longevity gene variant identified to date showing an association with negligible loss of telomere length with age in humans in a cross-sectional study. Reduced telomere attrition may be a key mechanism for the longevity-promoting effect of the FOXO3 genotype studied. Keywords: Human genetics, Lifespan, Telomere dynamics

Age is the strongest risk factor for death and for most chronic diseases associated with human mortality (1). Despite many theories concerning the mechanisms responsible for human aging, an

understanding of causality has been hampered by ethical, fiscal, and complexity constraints associated with performing necessary longterm controlled clinical studies. Research into the genetics of human

© The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: [email protected].

1448

Downloaded from https://academic.oup.com/biomedgerontology/article-abstract/73/11/1448/4980820 by guest on 24 November 2018

Minimal Shortening of Leukocyte Telomere Length Across Age Groups in a Cross-Sectional Study for Carriers of a Longevity-Associated FOXO3 Allele

Journals of Gerontology: BIOLOGICAL SCIENCES, 2018, Vol. 73, No. 11

Materials and Methods Okinawan-Japanese annual health exam participants (n = 121) ranging in age from 21 to 95 years were recruited, with prior informed consent, through Tomishiro Central Hospital (Tomigusuku, Okinawa, Japan) and its associated clinics throughout the Okinawa prefecture as well as Central Hospital (Naha, Okinawa, Japan) and its associated facilities. In Japan, essentially everyone undergoes an annual health screening exam (kenkko shindan or kenshin) that is mandated by the Ministry of Health, Labour, and Welfare. Therefore

there is no potential bias towards healthier individuals. All subjects were recruited between January 2013 and March 2016. Inclusion criteria were healthy individuals over the age of 20 years. Exclusion criteria were (a) age < 20 years, (b) recent heart attack, stroke, or cancer treatment, (c) severe dementia (meaning they were unable to understand the informed consent form), (d) any known genetic disease or disability, and (e) any patient whose attending physician deemed them inappropriate for the study. Blood (5 mL) was collected as part of a routine health screening examination. All samples were de-identified. The study was conducted following Veterans Affairs VA Institutional Review Board approval as well as approval from the Ethics Committee from Tomishiro Central Hospital, Tokushima University and Okinawa International University. Peripheral blood was collected in EDTA vacuum tubes and stored at 4°C for 2–10 days before processing. Mononuclear cells were separated utilizing Ficoll-Paque PREMIUM following the manufacturer’s guidelines. Cells (5 × 105) were aliquoted from the mononuclear fraction, washed in phosphate buffered saline, and lysed with CHAPS buffer. Extracts were assessed for telomerase activity using the telomere repeat amplification protocol (TRAP). Genomic DNA was isolated from mononuclear cells (~2.5–3.0 × 106) for genotyping and telomere length analysis. Fifty samples were either too degraded or had too little DNA or poor-quality DNA for accurate analysis of telomere length and were therefore excluded. Most of the samples that were excluded were in the initial phase of sample collection when it was unclear how long the genomic DNA would retain high quality from collection to shipping time. At that time, we were collecting samples from men, women, young, and old of various genotypes so there was little likelihood of bias. A total 121 high quality DNA samples were obtained that had sufficient DNA for Southern blot analysis of telomere length. Procedures to analyze all blood samples were performed at John A Burns School of Medicine, University of Hawaii. For more details, see Supplementary Methods.

Statistical Analysis Least squares linear regression (Generalized Linear Model) was used to determine the age effects on telomere length (group mean telomere length changes) for each FOXO3 genotype group (SAS version 9.2; SAS Institute, Inc., Cary, North Carolina), after assessing models with or without adjusting for either sex, smoking status, diabetes, BMI, or waist circumference. Differences in LTL between FOXO3 genotype by age bracket (ages 25–50 years) and telomerase activity as a function of genotype was assessed using Student’s t-test.

Results We assessed the effect of FOXO3 genotypes on mean telomere length and telomerase activity levels, as a function of age, in leukocytes from a cross-sectional population sample of healthy Okinawan-Japanese men and women (n = 121) aged 21–95 years. There was a significant interaction between age and FOXO3 genotype (p = .0005), consistent with prior studies (2). Table 1 summarizes standard clinical measurements as a function of FOXO3 genotype. The clinical measurements were similar between the groups. As expected, we found that mean telomere length (LTL) decreased with increasing donor age at a rate of loss ~22 base pairs per year (bp/year) for the total population (Figure 1; Table 2). Carriers of the longevity-associated FOXO3 G allele of SNP rs2802292 were significantly protected against aged-related telomere length loss relative to carriers of the


longevity have identified hundreds of genes associated with longevity, but polymorphisms in only two genes (APOE and FOXO3) have demonstrated strong and consistent replications across multiple, diverse human populations (2). We were the first to demonstrate an association of alleles of FOXO3 single nucleotide polymorphisms (SNPs) with longevity in a population of American men of Japanese ancestry (3). This finding was subsequently replicated in over 10 independent studies (2). The most robust FOXO3 SNP associated with human longevity is rs2802292 (2–4). In model organisms, many evolutionarily conserved genes associated with longevity encode protein involved in stress-tolerance (5) and energy homeostasis (1) pathways. A common connection between these pathways is FOXO3 homeostatic regulators (2). Indeed, FoxO3 appears to be a necessary factor for the most robust non-genetic intervention that enhances longevity— energy restriction—since genetic deletion of the FOXO3 homolog in mice abolishes the effect (6). Telomere attrition is an important mechanism that affects cellular lifespan (7–10). The reparative ribonucleoprotein, telomerase, protects the ends of linear chromosomes from loss of genetic material and helps maintain genome stability. The attrition of telomeres in mitotically active human somatic cells has now been shown to be ultimately responsible for their limited replicative lifespan (11). The reduction of length of one or more telomeres to a size that abrogates function leads to cell senescence by signaling a DNA damage response. In telomerase mutant mice, telomeres shorten with each successive generation, eventually leading to sterility and numerous aging-like phenotypes, including anemia, gray hair, and reduced mobility (12). Furthermore, re-activation of telomerase in telomerase deficient mice can extend telomere length, “rescue” proliferative defects of certain tissues, reduce indications of organismal aging, and restore lifespan (7). In humans, telomeres shorten in proliferative tissues, such as peripheral blood and bone marrow, during aging (8). Clinical studies of diverse populations have demonstrated a strong inverse correlation between telomere length and age in proliferating tissues (10). Furthermore, lifestyle stresses (13) and age-associated diseases (14), including cardiovascular disease (15) and stroke (16), significantly correlate with accelerated telomere shortening. Telomere length analysis in leukocytes has also been shown to be predictive of the development of dementia (16) and possibly certain types of cancer, for example prostate cancer (17). In this study, our primary objective was to evaluate the relationship between leukocyte telomere length (LTL) and longevity-associated alleles of FOXO3. To achieve this, we recruited a cross-sectional cohort of adults from a young age (20 years) to very old individuals (85–95 years of age) from a population with a high percentage of long-lived individuals—the Okinawan Japanese. The OkinawanJapanese are a relatively homogeneous population and have a high prevalence of oldest-old, including centenarians (18) so was a highly suitable population for the current study.

1449


1450

Table 1. Comparison of Age and Clinical Variables as a Function of FOXO3 Genotype GG

GT

TT

n Age (years)

11 59 ± 17.4 (42–87) 44 11 26.5 ± 4.26 89.2 ± 11.3 78

51 56.7 ± 17.4 (30–87) 57 18 26.4 ± 6.79 88.8 ± 13.8 40

59 60.2 ± 19.6 (25–94) 36 12 25.9 ± 2.7 90.9 ± 10.1 45

Gender (% male) Current smoking (%) BMI (kg/m2) Waist circumference (cm) Diabetes (% prevalence)

p*

.083 .091 .69 .73 .88 .26

Note: *p is for test of homogeneity among genotypes.

common TT genotype (p = .0005) (Figure 1A; Table 2). Strikingly, there was no significant correlation between telomere length and age throughout adulthood from age 21 to 95 (p > .1) for carriers of the longevity-associated G allele. In an exploratory analysis, no difference in the correlation between telomere length and age was observed as a function of allele carrier status amongst APOE genotypes containing alleles ε2, ε3, and ε4 (Supplementary Figure 1), although this analysis was limited in power due to low numbers of ε2 carriers and therefore requires further study. To control for possible confounder effects, we assessed multiple models with or without BMI, waist circumference, diabetes, gender, and smoking status. These factors did not influence the interaction of telomere length and age as a function of FOXO3 allele of SNP rs2802292, except non-smoking, which marginally reduced the significance of the rate of change in telomere length for G-allele carriers as compared to TT allele carriers (p = .01). To assess whether the longevity-associated G allele affects telomere length at different stages of life, we partitioned the analysis into young (21–50; n = 44) and old (51–95; n = 77) cohorts. Younger participants appeared to display similar telomere length dynamics irrespective of FOXO3 genotype (Figure 1B), exhibiting a significant negative correlation with age (p < .05), whereas older participants showed a striking difference in that G-allele carriers showed no association between telomere length and age (p > .05) as compared to TT-allele carriers, in whom there was a significant negative correlation with age (p = .008; Figure 1C). We also compared the average telomere length for subjects in age bins of 20–40, 40–60, 60–84, and 85–95 years. The results indicated differences in telomere length between G allele carriers and TT genotype, with a tendency of TT carriers to have longer telomeres early in life and shorter telomeres at very old ages (Figure 1D). This observation will require future investigation with additional longitudinal analysis to confirm a protective effect of the longevity associated G allele on telomere length. To assess whether the effect of FOXO3 genotype on telomere length across age may be attributable to genotype-related differences in telomerase activity, we measured leukocyte telomerase activity in the same samples used for genotyping. Normalized telomerase activity (mean ± SE) did not differ between FOXO3 genotypes (G carriers vs TT genotype: p > .1 [n = 87]; Supplementary Figure 2).

Discussion The present study is the first to report an association between telomere length and longevity-associated FOXO3 genotype as a function of age in a cross-sectional Okinawan-Japanese cohort. We found that the average telomere length for young (21–50-year-old) participants with the TT FOXO3 genotype was significantly higher (p ≤

.05) than that observed for G-allele carriers (FOXO3 genotype GG or GT). Conversely, telomere length was higher for elderly G-allele carriers (85–95 years of age). This was consistent with the lack of detection of any appreciable association between age and telomere length across the cohort for G-allele carriers, at least from age 50 onwards (Figure 1A and C). A convergence of telomere length in the elderly was reported previously, in that mean telomere length was diminished amongst the elderly, as compared to younger subjects (19). Extremely old persons, who exhibit an enrichment in longevity-associated alleles, tend to have longer telomeres (20). Consistent with protection against mortality, the frequency of the G-allele (rs2802292) has been shown to increase approximately 2-fold from age 70–100 years (21). We propose that the reduced rate of shortening of telomere length in the elderly may, in part, be explained by the increase in population frequency of protective FOXO3 G-allele carriers in older persons (21). The current association between FOXO3 genotype and telomere length in humans supports a prior finding in C. elegans, that in worms overexpressing telomere-binding HRP-1 protein, the lifespan-extending effect of elongated telomeres was dependent on the FOXO3 C. elegans homolog daf-16 (22). Others reported reduced telomere attrition with age in individuals having a genetic variant of SIRT1, a putative longevity gene whose encoded protein is in the pathway upstream of FoxO3, and that interacts with FoxO3 under conditions of stress (23). More recently, in a study of human fibroblast senescence that assessed the SIRT1-FoxO3-hTERT pathway, FoxO3 was a downstream activator between SIRT1 and hTERT that activated the transcription of c-MYC, and induced increased transcription of hTERT in human fibroblasts, delaying their senescence (24). Therefore, this pathway may provide a putative biological basis for the current findings of a reduced rate of loss of telomere length from young to old adults who are G-allele carriers. In support, a recent study of Japanese centenarian offspring, who tend to be healthier than average, possibly due to enrichment in protective alleles of longevity genes, also demonstrated negligible telomere attrition with age, in contrast to offspring of average-lived parents, who demonstrated the usual telomere attrition with age (20). The study also demonstrated longer telomeres in centenarians and super centenarians. The specific gene(s) responsible were not assessed in that study, but, as mentioned above, FOXO3 G-allele frequency is known to more than double from age 70 years to age 100 years and becomes the major allele in very old Japanese (2). In fact, by age 90 years, most Japanese, whites and blacks are G-allele carriers (21). In conclusion, the present study demonstrates, for the first time, lack of a correlation between telomere length and age, that is observed for the general population, for carriers of a validated


Variable


1451

Table 2. Effect of Age on Telomere Length, by FOXO3 Genotype FOXO3 Genotype*

Rate (bp/year)

95% CI

p†

GG GT G-carrier TT

9 2 4 −33

−53 to 71 −19 to 23 −16 to 23 −51 to −15

.79 .83 .70 .0008

human longevity-associated allele (over a range of 21–95 years of age). This suggests one potential mechanism for the protective effect of FOXO3 genotype on human lifespan and health span could possibly be through slowing the rate of telomere attrition in replicative tissues during aging. Our data further suggest that telomerase enzyme levels do not play a major role in this phenomenon. Thus, other factor(s) may be responsible for the ability of FOXO3 genotype to maintain telomere length well into old age. For example, carriers of the FOXO3 G allele might have substantially reduced rates of hematopoietic cell turnover during aging, possibly due to reduced levels of chronic inflammation (21). Future studies are warranted to further elucidate the cellular and molecular basis for these findings, and to better understand the interplay between FOXO3 and its encoded protein, with age-related telomere attrition.

Supplementary Material Supplementary data is available at The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences online.

Funding This study was supported by grants from the US National Institute on Aging (R21OAG04298: R.C.A., D.C.W., B.J.W.) and the Japan Ministry of Education, Culture, Sports, Science and Technology contract 16K01823 (M.S.). E.L. was partially supported by the National Institutes of Health (NIH) under the grants U54MD007584 and G12MD007601.

Acknowledgements We acknowledge the cooperation of staff from Diabetes and Life-Style Related Disease Center (Mss. Saya Kawakami and Saeko Nakamura) and Health Care Center (Dr. Masaki Takara) at Tomishiro Central Hospital, Chatan Hospital (Dr. Susumu Kinjo), Okinawa Central Hospital (Dr. Yoshiichi Onaka) and Central Care Village Utopia Okinawa (Mr. Masaaki Gima and Mr. Takanobu Okuma), Okinawa International University (Mr. Takeda Shinzato and Dean Shotoku Yasura), Pacific Health Research and Education Institute, and Okinawa Research Center for Longevity Science, for assistance with sample collection and/or research coordination and administration. We thank Ms. Sayaka Mitsuhashi for editorial assistance and all study participants and their families for their participation. Figure 1. (A) Regression analysis of LTL versus participant age for FOXO3 (rs2802292) genotype (n = 121; TT = 58, G-carriers = 63). (B) Regression analysis for relatively young age group (age 20–50 years) (n = 44; TT = 20, G-carriers = 24). (C) Regression analysis for older age group (age 51–95) (n = 77; TT = 38, G-carriers = 39). (Carriers of the longevity-associated FOXO3 G allele of SNP rs2802292 were significantly protected against a decrease in mean telomere length with age relative to common TT genotype; p = .0005 for comparison of regression curves). (D) Comparison of average telomere length for TT versus G-allele genotypes in age group bins from age 20 to 95. See Table 2 for analysis of interaction of age on telomere length as a function of FOXO3 genotype.

Conflict of Interest R.C.A., D.C.W., and B.J.W. serve on the Journal editorial board.

References 1. Kaeberlein M, Rabinovitch PS, Martin GM. Healthy aging: the ultimate preventative medicine. Science. 2015;350:1191–1193. doi:10.1126/science.aad3267


Note: CI = confidence interval. *Single nucleotide polymorphism rs2802292; Genetic model used was rare dominant, based on our findings in men of Japanese ancestry. †p – probability of linear regression coefficient compared to zero.

1452


14. Cawthon RM, Smith KR, O’Brien E, Sivatchenko A, Kerber RA. Association between telomere length in blood and mortality in people aged 60 years or older. Lancet. 2003;361:393–395. doi:10.1016/S0140-6736(03)12384-7. 15. Epel ES, Merkin SS, Cawthon R, et al. The rate of leukocyte telomere shortening predicts mortality from cardiovascular disease in elderly men. Aging (Albany NY). 2008;1:81–88. doi:10.18632/aging.100007. 16. Martin-Ruiz C, Dickinson HO, Keys B, Rowan E, Kenny RA, Von Zglinicki T. Telomere length predicts poststroke mortality, dementia, and cognitive decline. Ann Neurol. 2006;60:174–180. doi: 10.1002/ana.20869. 17. Hurwitz LM, Heaphy CM, Joshu CE, et al. Telomere length as a risk factor for hereditary prostate cancer. Prostate. 2014;74:359–364. doi:10.1002/ pros.22755. 18. Bendjilali N, Hsueh W-C, He Q, et al. Who are the Okinawans? ancestry, genome diversity, and implications for the genetic study of human longevity from a geographically isolated population. J Gerontol A Biol Sci Med Sci. 2014;69:1474–1484. doi:10.1093/gerona/glt203 19. Halaschek-Wiener J, Vulto I, Fornika D, et al. Reduced telomere length variation in healthy oldest old. Mech Ageing Dev. 2008;129:638–641. doi:10.1016/j.mad.2008.07.004 20. Arai Y, Martin-Ruiz CM, Takayama M, et al. Inflammation, but not telomere length, predicts successful ageing at extreme old age: a longitudinal study of semi-supercentenarians. EBioMedicine. 2015;2:1549–1558. doi:10.1016/j.ebiom.2015.07.029 21. Willcox BJ, Tranah GJ, Chen R, et al. The FoxO3 gene and cause-specific mortality. Aging Cell. 2016;15:617–624. doi:10.1111/acel.12452 22. Joeng KS, Song EJ, Lee KJ, Lee J. Long lifespan in worms with long telomeric DNA. Nat Genet. 2004;36:607–611. doi:10.1038/ng1356 23. Kim S, Bi X, Czarny-Ratajczak M, et al. Telomere maintenance genes SIRT1 and XRCC6 impact age-related decline in telomere length but only SIRT1 is associated with human longevity. Biogerontology. 2012;13:119– 131. doi:10.1007/s10522-011-9360-5 24. Yamashita S, Ogawa K, Ikei T, Fujiki T, Katakura Y. FOXO3a potentiates hTERT gene expression by activating c-MYC and extends the replicative life-span of human fibroblast. PLoS One. 2014;9:e101864. doi:10.1371/ journal.pone.0101864


2. Morris BJ, Willcox DC, Donlon TA, Willcox BJ. FOXO3: a major gene for human longevity–a mini-review. Gerontology. 2015;61:515–525. doi:10.1159/000375235 3. Willcox BJ, Donlon TA, He Q, et al. FOXO3A genotype is strongly associated with human longevity. Proc Natl Acad Sci USA. 2008;105:13987– 13992. doi:10.1073/pnas.0801030105 4. Broer L, Buchman AS, Deelen J, et al. GWAS of longevity in CHARGE consortium confirms APOE and FOXO3 candidacy. J Gerontol A Biol Sci Med Sci. 2015;70:110–118. doi:10.1093/gerona/glu166 5. Dato S, Soerensen M, Lagani V, et al. Contribution of genetic polymorphisms on functional status at very old age: a gene-based analysis of 38 genes (311 SNPs) in the oxidative stress pathway. Exp Gerontol. 2014;52:23–29. doi:10.1016/j.exger.2014.01.014 6. Shimokawa I, Komatsu T, Hayashi N, et al. The life-extending effect of dietary restriction requires Foxo3 in mice. Aging Cell. 2015;14:707–709. doi:10.1111/acel.12340 7. Jaskelioff M, Muller FL, Paik J-H, et al. Telomerase reactivation reverses tissue degeneration in aged telomerase deficient mice. Nature. 2011;469:102–106. doi:10.1038/nature09603 8. von Zglinicki T, Martin-Ruiz CM. Telomeres as biomarkers for ageing and age-related diseases. Curr Mol Med. 2005;5:197–203. 9. Li JSZ, Fuste JM, Simavorian T, et al. TZAP: a telomere-associated protein involved in telomere length control. Science. 2017;355:638–641. doi:10.1126/science.aah6752 10. Blackburn EH, Epel ES, Lin J. Human telomere biology: a contributory and interactive factor in aging, disease risks, and protection. Science. 2015;350:1193–1198. doi:10.1126/science.aab3389. 11. Bodnar AG, Ouellette M, Frolkis M, et al. Extension of life-span by introduction of telomerase into normal human cells. Science. 1998;279:349–352. 12. Blasco MA, Lee HW, Hande MP, et al. Telomere shortening and tumor formation by mouse cells lacking telomerase RNA. Cell. 1997;91:25–34. 13. Epel ES. Psychological and metabolic stress: a recipe for accelerated cellular aging? Hormones (Athens). 2009;8:7–22.