Identification of Biomarkers for Type 2 Diabetes Yuvaraj Mahendran Identification of Biomarkers for Type 2 Diabetes

Type 2 diabetes (T2D) is a complex disorder characterized by insulin resistance and pancreatic β-cell dysfunction. The incidence and prevalence of T2D have doubled in recent decades, this phenomenon being attributable to obesity, sedentary lifestyle and unhealthy diet. Both genetic and environmental factors are major determinants of this disease. Genome-wide association studies have identified several risk loci for T2D and hyperglycemia, but the biological role of most of these variants remains unknown. The early diagnosis of diabetes is important in order to avoid long-term microand macrovascular complications in individuals at high risk of T2D. Therefore, the identification of biomarkers that accurately predict incident T2D is of great interest. The main aim of this study was to identify non-genetic and genetic biomarkers that would predict hyperglycemia and incident T2D in a prospective follow-up of the populationbased METSIM (METabolic Syndrome In Men) cohort. We also investigated the significance of insulin sensitivity and insulin secretion as mediators in the associations of metabolites with the deterioration of hyperglycemia and incident T2D. We found that high fasting levels of glycerol, free fatty acids, monounsaturated fatty acids (FAs), and saturated FAs, and omega-7 and -9 FAs associated with increased risk of the development of hyperglycemia and T2D, whereas high levels of omega-6 FAs were associated with reduced risk of hyperglycemia and T2D. Insulin resistance explained these associations at least in part. With respect to erythrocyte membrane FAs, palmitoleic acid, dihomo-gamma-linolenic acid, and the ratios of 16:1n-7/16:0 and 20:3n6/18:2n-6 associated with the worsening of hyperglycemia, whereas the linoleic acid level and the ratio of 18:1n-7/16:1n-7 were associated with decreases in the hyperglycemia. Palmitoleic acid and the ratio of 16:1n-7/16:0 nominally predicted incident T2D, whereas linoleic acid had an opposite association. These associations were largely independent of insulin sensitivity, insulin secretion and glucose levels. Finally, high levels of acetoacetate and β-hydroxybutyrate predicted the worsening of hyperglycemia, and acetoacetate predicted incident T2D. Impaired insulin secretion, but not insulin resistance, explained these associations. One common variant rs780094 of the glucokinase regulatory protein gene was significantly associated with βhydroxybutyrate levels. In conclusion, this study identified several novel biomarkers predicting the worsening of hyperglycemia and incident T2D which could be used in clinical practice. National Library of Medicine Classification: QU 84, QU 90, WK 810, WK 820, WK 880 Medical Subject Headings: Diabetes Mellitus, Type 2; Biological Markers; Genetics; Metabolomics; Hyperglycemia; Insulin; Insulin Resistance; Glycerol; Fatty Acids; Ketone Bodies; Cohort Studies