Interrelationships and determinants of aging biomarkers in cord blood

Abstract

Background: Increasing evidence supports the concept of prenatal programming as an early factor in the aging process. DNA methylation age (DNAm age), global genome-wide DNA methylation (global methylation), telomere length (TL), and mitochondrial DNA content (mtDNA content) have independently been shown to be markers of aging, but their interrelationship and determinants at birth remain uncertain.

Methods: We assessed the inter-correlation between the aging biomarkers DNAm age, global methylation, TL and mtDNA content using Pearson's correlation in 190 cord blood samples of the ENVIRONAGE birth cohort. TL and mtDNA content was measured via qPCR, while the DNA methylome was determined using the human 450K methylation Illumina microarray. Subsequently, DNAm age was calculated according to Horvath's epigenetic clock, and mean global, promoter, gene-body, and intergenic DNA methylation were determined. Path analysis, a form of structural equation modeling, was performed to disentangle the complex causal relationships among the aging biomarkers and their potential determinants.

Results: DNAm age was inversely correlated with global methylation (r = -0.64, p < 0.001) and mtDNA content (r = - 0.16, p = 0.027). Cord blood TL was correlated with mtDNA content (r = 0.26, p < 0.001) but not with global methylation or DNAm age. Path analysis showed the strongest effect for global methylation on DNAm age with a decrease of 0.64 standard deviations (SD) in DNAm age for each SD (0.01%) increase in global methylation (p < 0.001). Among the applied covariates, newborn sex and season of delivery were the strongest determinants of aging biomarkers.

Conclusions: We provide insight into molecular aging signatures at the start of life, including their interrelations and determinants, showing that cord blood DNAm age is inversely associated with global methylation and mtDNA content but not with newborn telomere length. Our findings demonstrate that cord blood TL and DNAm age relate to different pathways/mechanisms of biological aging and can be influenced by environmental factors already at the start of life. These findings are relevant for understanding fetal programming and for the early prevention of noncommunicable diseases.

Keywords: Aging biomarkers; Cord blood; DNAm age; telomere length; mitochondrial DNA content; ENVIRONAGE cohort; Global methylation.

Fig. 1

Overview of the identified studies investigating the interrelationship between the aging biomarkers telomere length (TL)), DNA methylation age or DNA methylation age acceleration (DNAm age/DNAm AA), mitochondrial DNA content (mtDNA content) or global DNA methylation (global m.) in the age groups of adolescents (mean age ≥ 10), adults (mean age > 19) and older people (mean age > 65). The arrows in the ovals show which correlations/associations have been investigated in the different age groups

Fig. 2

Flow chart visualizing the sample selection. Initially, 199 mother-newborn pairs participating in the ENVIRONAGE birth cohort between July 2014 and June 2015, with epigenome-wide DNA methylation data, were selected. The final number of participating mother-newborn pairs included in the analysis was 190

Fig. 3

Pearson correlations between DNAm Age = epigenetic age calculated according to Horvath [9], TL = relative telomere length, mtDNAcontent = relative mitochondrial DNA content, globalmean = mean global DNA methylation, promotermean = mean methylation of the promoter gene-region, bodymean = mean methylation of the gene-body and intergenicmean = mean methylation of the intergenic region. In the top right corner, the correlation coefficients, and in the bottom left corner scatterplots of the correlation with regression line and 0.95% confidence interval are shown. The diagonal density plots display the distribution of observations. 5mC = 5-methylcytosine; mtDNA = mitochondrial DNA; nDNA = nuclear DNA; T/S = telomere/ single copy gene ratio * p < 0.05; ** p < 0.001

Fig. 4

Graphical display of the path analysis model showing only the p < 0.05 significant standardized estimates for the multiple regression analyses with the four markers of biological age as endogenous variables (bottom). The analysis's exogenous variables comprised the other respective aging biomarkers, sex, gestational age, newborn ethnicity, birthweight, maternal smoking, maternal education, maternal early-pregnancy BMI, and parity. For TL and mtDNA content, additionally, white blood cell count, season of delivery and parental age, and for global methylation cell-type distribution according to Bakulski [51], were included as covariates. The coefficients in the figure were standardized, representing a 1 SD change in each exposure pathway. Red color stands for negative and green color for positive associations. The arrow's width indicates the degree of correlation, with wider arrows indicating higher correlation. Significant associations between cord blood cell composition and white blood cell count are not shown for the sake of clarity. DNAmAge = epigenetic age calculated according to Horvath [9]; Gestat.Age = gestational age in days; Globalm = global DNA methylation; TL = relative telomere length; mtDNA = mitochondrial DNA content; Warm Season = warm season (April 1 – September 30). *p < 0.05; **p < 0.001

Fig. 5

Associations between the cord blood aging biomarkers investigated in this study and health outcomes at birth and later in life were found in previous studies. A green check mark indicates that the study reported an association, and a grey cross indicates the lack of an association. A slash indicates that no study investigating the relationship with the health outcome was identified. Y = years of age