Identifying key circulating biomarkers that mediate the association between dynamic physical activity patterns and obesity risk

Background
The effectiveness of physical activity (PA) in reducing obesity risk is well established; however, the impact of PA timing throughout the day on obesity remains unclear. This study investigates the relationship between dynamic PA patterns and risk of obesity, aiming to identify potential circulating biomarkers that mediate this relationship.

Methods
We analyzed 24 h PA data from 92,072 UK Biobank participants (aged 42 to 79 years), collected via accelerometers between 2013 and 2015. Functional principal component analysis (fPCA) extracted circadian PA patterns. Multivariable logistic regression models related fPCA scores to obesity risk. Mediation analyses identified significant biomarkers mediating the PA-obesity relationship.

Results
Three distinct PA profiles explained 99.7% of the total variation. Significant associations were observed between rest-activity profiles and sociodemographic characteristics. A higher fPC1 score was inversely associated with obesity risk (beta = 0.77; 95% CI 0.76–0.78, p < 0.001). Early rise PA (fPC2) showed an inverse association with obesity (beta = 0.97; 95% CI 0.96–0.98, p < 0.001). A high fPC3 score also demonstrated inverse associations with obesity (beta = 0.93; 95% CI 0.92–0.94, p < 0.001). Biomarkers such as HDL cholesterol, triglycerides, LDL cholesterol, glucose, HbA1c, SHBG, GGT, CRP, IL-6, GlycA and SOD were identified as mediators. The mediation proportions for fPC1, fPC2, and fPC3 were 62.82%, 64.59%, and 54.85%, respectively.ConclusionOur results suggest that both overall PA and its timing are crucial for reducing obesity risk. Blood lipids, glucose, oxidative stress markers (GGT, SOD), inflammation biomarkers (CRP, IL-6, GlycA) and hormone SHBG are key circulating biomarkers mediating the relationship between physical activity and obesity.