The fetal origins of metabolic health: exploring the association between newborn biological age and metabolism hormones in childhood

Abstract

BACKGROUND: Telomere length (TL), mitochondrial DNA copy number (mtDNAcn), and DNA methylation age (DNAmAge) are common aging biomarkers. However, research on the associations between these three markers at birth and subsequent metabolic status was limited. This study aimed to evaluate the association between TL, mtDNAcn, and DNAmAge in newborns and the variation in metabolic hormones of children at 3 years old. METHODS: This research involved 895 mother-child pairs from a birth cohort in China, with TL and mtDNAcn measured using quantitative real-time PCR, DNA methylation (DNAm) assessed using Infinium MethylationEPIC Beadchip, and DNAm age (DNAmAge) determined using Horvath's epigenetic clock. Insulin and leptin levels were measured via electrochemiluminescence assay. Multivariable adjusted linear regression and restricted cubic spline (RCS) analysis were utilized to examine the association between aging markers and metabolic hormones. RESULTS: The linear regression analysis indicated the percentage change of metabolism hormones for per doubling of aging biomarkers alterations and found significant associations between DNAmAge and insulin levels (adjusted percent change (95% CI), - 13.22 (- 23.21 to - 1.94)), TL and leptin levels (adjusted percent change (95% CI), 15.32 (1.32 to 31.24)), and mtDNAcn and leptin levels (adjusted percent change (95% CI), - 14.13 (- 21.59 to - 5.95)). The RCS analysis revealed significant non-linear associations between TL (Ln transformed) and insulin (Ln transformed) (P = 0.024 for nonlinearity), as well as DNAmAge (Ln transformed) and leptin (Ln transformed) (P = 0.043 for nonlinearity). Specifically, for TL and insulin, a positive association was observed when TL (Ln transformed) was less than - 0.05, which transitioned to an inverse association when TL (Ln transformed) was greater than - 0.05. Regarding DNAmAge and leptin, there was a sharp decline when DNAmAge (Ln transformed) was less than - 1.35, followed by a plateau between - 1.35 and - 0.67 and then a further decline when DNAmAge (Ln transformed) was greater than - 0.67. CONCLUSIONS: In this prospective birth cohort study, variation in metabolic hormones of children at 3 years old was associated with TL, mtDNAcn, and DNAmAge at birth. These findings suggested that TL, mtDNAcn, and DNAmAge might play a role in the biological programming of metabolic health from birth.