SFEBES2017 Oral Communications Obesity, Diabetes Thyroid (6 abstracts)
1Vascular Research Group, University of Manchester, Manchester, UK; 2Northwest Diabetes Research Network, Salford Royal NHS Foundation Trust, Salford, UK; 3Department of Obesity in the Medicine School of Instituto Politecnico Nacional, Mexico City, Mexico; 4Department of Obesity and Endocrinology, University of Liverpool, Liverpool, UK; 5The School of Medicine and Manchester Academic Health Sciences Centre, University of Manchester, Manchester, UK; 6Centre for Biostatistics, Institute of Population Health, University of Manchester, Manchester, UK.
Introduction: T2DM is one of the main causes of morbidity and mortality in the UK. Epigenetic mechanisms impact gene expression and could predispose individuals to a particular metabolic phenotype. We investigated association of LINE-1 methylation (a surrogate of global DNA methylation) with cardiometabolic parameters in a longitudinal cohort of T2DM patients.
Methods: Global LINE-1 DNA methylation in blood-derived DNA samples from 795 (men 60%, women 40%) T2DM patients was quantified by pyrosequencing using PyroMark Q96 CpG LINE-1 (Qiagen). Mean age 59.2 years (men 58.6 years, women 60.0 years). Longitudinal data was collected from year 2002 to 2016.
Results: Mean methylation quantified at 4 CpG sites was 75.78±3.37%. There was no significant methylation difference between men 75.93±3.33% and women 75.56±3.41%, P=0.130 or by age (P=0.123). Cross-sectional analysis at baseline year 2002: Linear regression analysis at baseline showed LINE-1 methylation could statistically significantly predict diastolic BP (adjusted coefficient −0.35 (95% CI −0.59 −0.11, P=0.004)) and eGFR (−0.55 (95% CI −0.92 −0.18, P=0.004)). A ten percent increase in LINE-1 methylation resulted in reduction of diastolic BP by 3.5 mmHg and a reduction in eGFR by 9.2 ml/min/1.73 m2. There was no association with lipid parameters or with HbA1c. Longitudinal analysis over 14 years: Global LINE-1 methylation was negatively associated with BMI in women (−0.25 (95% CI −0.45 −0.05, P=0.013)), and with less weight gain over time. A ten percent increase in LINE-1 methylation was associated with 2.5 kg/m2 reduced BMI in 2016 compared with the population mean. The relation with BP diminished over time, likely due to treatment to target effects. Logistic regression analysis showed no association with mortality and cardiovascular events.
Conclusion: In a 14 year longitudinal cohort of T2DM individuals, we have demonstrated that higher degree of LINE-1 methylation is predictive of less weight gain over time/lower future BMI in women, and relates to lower baseline diastolic BP. Methylation status may thus influence weight trajectory in this group. Potential effects of pharmacological intervention on the relation between methylation and BP/eGFR require further investigation.