No abstract

The aim of this study was to estimate the incidence of diabetes mellitus in the Basque Country and the risk factors involved in the disease by reassessing an adult population after 7 years of follow-up. In the previous prevalence study, 847 people older than 18 years were randomly selected from all over the Basque Country and were invited to answer a medical questionnaire, followed by a physical examination and an oral glucose tolerance test. In the reassessment, the same variables were collected and the resulting cohort comprised 517 individuals of whom 43 had diabetes at baseline. The cumulative incidence of diabetes was 4.64% in 7 years and the raw incidence rate was 6.56 cases/1000 person-years (95%CI: 4.11–9.93). Among the incident cases, 59% were undiagnosed. The most strongly associated markers by univariate analyses were age > 60 years, dyslipidaemia, prediabetes and insulin resistance. We also found association with hypertension, obesity, family history of diabetes and low education level. Multivariate analysis adjusted for age and sex showed that a set of risk factors assessed together (dyslipidaemia, waist-to-hip-ratio and family history of diabetes) had great predictive value (AUC-ROC = 0.899, 95%CI: 0.846–0.953, p = 0.942), which suggests the need for early intervention before the onset of prediabetes.

Aims/HypothesisFailure in glucose response to insulin is a common pathology associated with obesity. In this study, we analyzed the genome wide DNA methylation profile of visceral adipose tissue (VAT) samples in a population of individuals with obesity and assessed whether differential methylation profiles are associated with the presence of type 2 diabetes (T2D).MethodsMore than 485,000 CpG genome sites from VAT samples from women with obesity undergoing gastric bypass (n = 18), and classified as suffering from type 2 diabetes (T2D) or not (no type 2 diabetes, NT2D), were analyzed using DNA methylation arrays.ResultsWe found significant differential methylation between T2D and NT2D samples in 24 CpGs that map with sixteen genes, one of which, HOOK2, demonstrated a significant correlation between differentially hypermethylated regions on the gene body and the presence of type 2 diabetes. This was validated by pyrosequencing in a population of 91 samples from both males and females with obesity. Furthermore, when these results were analyzed by gender, female T2D samples were found hypermethylated at the cg04657146-region and the cg 11738485-region of HOOK2 gene, whilst, interestingly, male samples were found hypomethylated in this latter region.ConclusionThe differential methylation profile of the HOOK2 gene in individuals with T2D and obesity might be related to the attendant T2D, but further studies are required to identify the potential role of HOOK2 gene in T2D disease. The finding of gender differences in T2D methylation of HOOK2 also warrants further investigation.