Puberty reshapes DNA to worsen or reverse insulin resistance

A longitudinal study shows that puberty may reprogram children’s epigenetic profiles in ways that track worsening or improving insulin resistance, uncovering early molecular signals of future cardiometabolic disease.

Study: Novel epigenetic marks of insulin resistance trajectories in a longitudinal study of childhood obesity. Image credit: Pixel-Shot/Shutterstock.com

The epidemic rise of obesity among children is a threat to public health for generations to come. Obesity is linked to the emergence of insulin resistance, which increases the risk of diabetes and cardiovascular disease. A recent paper in the journal Cardiovascular Diabetology explores the association of epigenetic markers with the presence and trajectory of IR from childhood to adolescence, highlighting puberty as a critical window for early cardiometabolic risk programming.

Puberty marks a turning point in insulin sensitivity

Insulin resistance (IR) is a condition where tissues show decreased responsiveness to insulin and thus fail to metabolize glucose in physiologically appropriate ways. IR contributes to the occurrence of multiple adverse cardiometabolic outcomes, including metabolic syndrome and cardiovascular disease (CVD), in children and adults. IR in childhood and adolescence is most often associated with obesity, which often persists in young adults. Notably, IR tends to persist at puberty in obese children, while in normal-weight healthy children, an initial decline in insulin sensitivity is followed by a return to normal after puberty.

Prior research has shown that persistent IR at puberty is closely linked to young-onset T2DM in youth. Several single-nucleotide polymorphisms (SNPs) are correlated with the risk of T2DM and associated traits like IR, but account for only a small part of the genetic risk. Among other putative factors driving IR, epigenetic changes like DNA methylation form a major component. DNA methylation dynamically regulates gene expression, mediating the effect of environmental changes on disease risk.

An earlier multi-cohort epigenome-wide association study (EWAS) demonstrated that epigenetic-genetic associations explained 32 % of variance in adult body mass index (BMI). This may involve the interaction of specific SNPs with regulatory changes affecting DNA methylation, termed methylation quantitative trait loci (mQTL). Such mQTLs have been shown to alter transcription and insulin secretion in pancreatic tissue.

Multiple metabolic disorders like T2DM and obesity have their own DNA methylation signatures. The current study sought to examine epigenetic alterations at puberty, a critical developmental window with significant changes in DNA methylation.

PUBMEP cohort tracks children through puberty

This research forms part of the PUBMEP study, which assessed the course of cardiometabolic risk factors from pre-puberty to puberty in Spanish children. The investigators used data from 90 children who were followed up from prepuberty to puberty. Other participants were assessed only at prepubertal (99) or pubertal (129) stages, and were added to expand the sample size for cross-sectional analysis.

IR was assessed in each participant using the homeostasis model assessment of IR (HOMA-IR) and the quantitative insulin sensitivity test index (QUICKI) tools. IR status was defined using sex- and pubertal-stage-specific HOMA-IR cut-offs. This was compared against genotype and epigenetic data using linear models. The aim was to identify site-specific methylation status in correlation with IR status and trajectory.

EWAS models were adjusted for age, sex, recruitment centre, estimated blood-cell proportions, and technical batch effects. Further, mQTL analyses were performed to identify potential genetic drivers of IR-associated pubertal epigenetic trajectories and profiles, especially in obese children.

Worsening IR tied to hypermethylation patterns

The researchers stratified the longitudinal sample into five groups based on the trajectories of obesity-related IR. These were combined with the cross-sectional sample to obtain a set of IR-associated CpG (methylated DNA) sites. Comparisons were made within each longitudinal group to capture individual-level methylation changes within each trajectory; between groups to capture contrasts between participants with improving versus persistent or developing IR; and across the cross-sectional sample to compare adolescents with and without IR.

This analysis identified 120 CpG sites where obesity-associated IR at puberty was correlated with methylation changes. The greatest number of sites were identified in the longitudinal between-groups comparison. These comprised a hypermethylation profile that tended to occur with worsening IR (the group that developed IR at puberty), in contrast to a hypomethylation trajectory that was more likely to occur when IR resolved at puberty.

The two groups with new-onset or persistent puberty IR had the highest increases in HOMA-IR, fasting insulin, and in the fat mass index (FMI, fat mass (kg)/height (m)2), compared to the other three groups that did not have puberty IR. Participants with new-onset puberty IR had the greatest increase in leptin, and in the inflammatory markers hs-CRP and t-PAI-1, the lowest levels being observed in the group with improved IR at puberty.

The greatest difference in systolic blood pressure was seen between the groups with puberty IR and the others, underlining the importance of puberty-linked changes in metabolic risk factors.

Interestingly, 14 of these IR-associated methylation sites were also linked to puberty adiposity, high blood sugar, or blood pressure. This suggests that these sites may represent shared epigenetic signatures underlying multiple cardiometabolic abnormalities rather than isolated IR. The current study also mapped other sites to established epigenetic loci associated with adult T2DM and obesity, suggesting the presence of these metabolic perturbations at puberty, albeit subclinically.

Some of the IR-associated methylation sites were located in genes like PEPD, TSC2, EGLN3, EHD2, SLC2A9, and VASN. These have not been reported to be linked to IR in children so far, and represent novel markers in pediatric cohorts. All six were consistently linked to higher IR at puberty. Some of these loci were also correlated with adiposity and blood pressure. In all cases, the direction of change varied with the IR trajectory at puberty.

These genes are expressed in pathways relating to vascular signaling, extracellular matrix remodeling, nutrient sensing, and hypoxia sensing. These may represent biologically plausible changes occurring early in metabolic disturbance. Larger studies with deep epigenetic assessment, possibly coupled with Mendelian randomization and mediation analyses, could help unravel the mechanisms underlying these findings.

Despite the lack of causal inferences, these observations suggest the presence of modifiable epigenetic markers in obesity-related IR. These trajectory-linked methylation patterns were consistent with pubertal cross-sectional differences between IR and non-IR adolescents.

mQTL analysis

For 25 cis SNP–CpG associations corresponding to 20 CpG sites, mQTL analysis identified nearby genetic variants that regulate DNA methylation. These variants may explain part of the variation. Conversely, most CpG sites associated with IR in this study appear to be independent of genetic influence.

Strengths and limitations

The study used longitudinal assessments, corrected for known age, sex, recruitment-centre, blood-cell-composition, and technical confounders, and incorporated multiple cardiometabolic traits with mQTL data into the epigenome data. This enabled the researchers to link epigenetic and clinical trajectories of IR in children.

However, the study was limited by its relatively small sample size and by the use of whole blood to assess DNA methylation, which may not fully capture tissue-specific changes in insulin-sensitive organs such as the liver, muscle, or pancreas. It also lacked independent comparison cohorts for validation, and residual confounding cannot be ruled out, particularly from factors known to influence epigenetic variation, including exposure to pollution, smoking, and maternal body mass index.

Findings open path toward precision prevention

This longitudinal study shows that in children, distinct DNA methylation profiles mirror the presence of IR and its trajectory during the pubertal transition. While a small subset of these IR-associated changes in methylation status may be partly driven by nearby genetic variants, most reflect responses to metabolic and environmental stimuli. This suggests the possibility of targeting epigenetic modification along with metabolic parameters to reduce cardiometabolic risk in the future.

These findings need to be replicated and their functional significance identified. Follow-up research could help understand how obesity drives IR at molecular level, paving the way for cardiometabolic preventive approaches.

Download your PDF copy by clicking here.