A1 Refereed original research article in a scientific journal
BMI is positively associated with accelerated epigenetic aging in twin pairs discordant for body mass index
Authors: Lundgren Sara, Kuitunen Sara, Pietiläinen Kirsi H., Hurme Mikko, Kähönen Mika, Männistö Satu, Perola Markus, Lehtimäki Terho, Raitakari Olli, Kaprio Jaakko, Ollikainen Miina
Publisher: WILEY
Publication year: 2022
Journal: Journal of Internal Medicine
Journal name in source: JOURNAL OF INTERNAL MEDICINE
Journal acronym: J INTERN MED
Volume: 292
Issue: 2
First page : 627
Last page: 640
Number of pages: 14
ISSN: 0954-6820
eISSN: 1365-2796
DOI: https://doi.org/10.1111/joim.13528
Web address : https://onlinelibrary.wiley.com/doi/10.1111/joim.13528
Self-archived copy’s web address: https://research.utu.fi/converis/portal/detail/Publication/176166174
Background
Obesity is a heritable complex phenotype that can increase the risk of age-related outcomes. Biological age can be estimated from DNA methylation (DNAm) using various "epigenetic clocks." Previous work suggests individuals with elevated weight also display accelerated aging, but results vary by epigenetic clock and population. Here, we utilize the new epigenetic clock GrimAge, which closely correlates with mortality.
Objectives
We aimed to assess the cross-sectional association of body mass index (BMI) with age acceleration in twins to limit confounding by genetics and shared environment.
Methods and results
Participants were from the Finnish Twin Cohort (FTC; n = 1424), including monozygotic (MZ) and dizygotic (DZ) twin pairs, and DNAm was measured using the Illumina 450K array. Multivariate linear mixed effects models including MZ and DZ twins showed an accelerated epigenetic age of 1.02 months (p-value = 6.1 x 10(-12)) per one-unit BMI increase. Additionally, heavier twins in a BMI-discordant MZ twin pair (Delta BMI >3 kg/m(2)) had an epigenetic age 5.2 months older than their lighter cotwin (p-value = 0.0074). We also found a positive association between log (homeostatic model assessment of insulin resistance) and age acceleration, confirmed by a meta-analysis of the FTC and two other Finnish cohorts (overall effect = 0.45 years, p-value = 4.1 x 10(-25)) from adjusted models.
Conclusion
We identified significant associations of BMI and insulin resistance with age acceleration based on GrimAge, which were not due to genetic effects on BMI and aging. Overall, these results support a role of BMI in aging, potentially in part due to the effects of insulin resistance.
Downloadable publication This is an electronic reprint of the original article. |