Identification of markers that distinguish adipose tissue and glucose and insulin metabolism using a multi-modal machine learning approach.

The study of metabolomics has improved our knowledge of the biology behind type 2 diabetes and its related metabolic physiology. We aimed to investigate markers of adipose tissue morphology, as well as insulin and glucose metabolism in 53 non-obese male individuals. The participants underwent extensive clinical, biochemical and magnetic resonance imaging phenotyping, and we also investigated non-targeted serum metabolites. We used a multi-modal machine learning approach to evaluate which serum metabolomic compounds predicted markers of glucose and insulin metabolism, adipose tissue morphology and distribution. Fasting glucose was associated with metabolites of intracellular insulin action and beta-cell dysfunction, namely cysteine-s-sulphate and n-acetylgarginine, whereas fasting insulin was predicted by myristoleoylcarnitine, propionylcarnitine and other metabolites of beta-oxidation of fatty acids. OGTT-glucose levels at 30 min were predicted by 7-Hoca, a microbiota derived metabolite, as well as eugenol, a fatty acid. Both insulin clamp and HOMA-IR were predicted by metabolites involved in beta-oxidation of fatty acids and biodegradation of triacylglycerol, namely tartrate and 3-phosphoglycerate, as well as pyruvate, xanthine and liver fat. OGTT glucose area under curve (AUC) and OGTT insulin AUC, was associated with bile acid metabolites, subcutaneous adipocyte cell size, liver fat and fatty chain acids and derivates, such as isovalerylcarnitine. Finally, subcutaneous adipocyte size was associated with long chain fatty acids, markers of sphingolipid metabolism, increasing liver fat and dopamine-sulfate 1. Ectopic liver fat was predicted by methylmalonate, adipocyte cell size, glutathione derived metabolites and fatty chain acids. Ectopic heart fat was predicted visceral fat, gamma-glutamyl tyrosine and 2-acetamidophenol sulfate. Adipocyte cell size, age, alpha-tocopherol and blood pressure were associated with visceral fat. We identified several biomarkers associated with adipose tissue pathophysiology and insulin and glucose metabolism using a multi-modal machine learning approach. Our approach demonstrated the relative importance of serum metabolites and they outperformed traditional clinical and biochemical variables for most endpoints.

Adipose tissue morphology, imaging and metabolomics predicting cardiometabolic risk and family history of type 2 diabetes in non-obese men.

We evaluated the importance of body composition, amount of subcutaneous and visceral fat, liver and heart ectopic fat, adipose tissue distribution and cell size as predictors of cardio-metabolic risk in 53 non-obese male individuals. Known family history of type 2 diabetes was identified in 25 individuals. The participants also underwent extensive phenotyping together with measuring different biomarkers and non-targeted serum metabolomics. We used ensemble learning and other machine learning approaches to identify predictors with considerable relative importance and their intricate interactions. Visceral fat and age were strong individual predictors of ectopic fat accumulation in liver and heart along with markers of lipid oxidation and reduced glucose tolerance. Subcutaneous adipose cell size was the strongest individual predictor of whole-body insulin sensitivity and also a marker of visceral and ectopic fat accumulation. The metabolite 3-MOB along with related branched-chain amino acids demonstrated strong predictability for family history of type 2 diabetes.

Metabolic characteristics of individuals at a high risk of type 2 diabetes - a comparative cross-sectional study.

BACKGROUND: Type 2 diabetes (T2D) is associated with substantial morbidity and mortality. Individuals with a family history of T2D are at an increased risk of developing the disease. The aim of this study was to assess metabolic differences between first-degree relatives (FDR) of T2D patients and persons with no known family history of T2D (non-FDR). METHODS: In 200 FDR and 73 non-FDR, we compared anthropometrics, glucose tolerance status, different measurements of insulin secretion, insulin resistance, as well as blood lipids and other blood analyses. RESULTS: In the FDR group, 30 individuals had impaired glucose tolerance or T2D. Among the non-FDR, two individuals had impaired glucose tolerance. In unadjusted data, the FDR were older, had stronger heredity for coronary heart disease, lower body mass index and weight, higher OGTT plasma glucose concentrations, and impaired insulin secretion (all p < 0.05). Using propensity score, we matched the groups, resulting in significantly stronger heredity of coronary heart disease, higher OGTT plasma glucose at 60 and 90 min, larger glucose area under curve during the OGTT and higher serum creatinine among the FDR. Using least squares means, OGTT glucose at 60 and 120 min, as well as the area under curve, and OGTT insulin levels at 60 min were significantly higher. Body mass index was negatively correlated with insulin sensitivity (MI) and positively correlated with HOMA-ß, a measurement of insulin secretion. CONCLUSIONS: We show that FDR are more likely to have impaired glucose tolerance and display higher OGTT plasma glucose and insulin, indicating an unfavorable metabolic profile. We conclude that OGTT is a simple and yet informative metabolic assessment in the FDR group. In both groups, we saw a negative correlation between body mass index and MI, confirming the role of body mass index in insulin resistance.

Metabolic predictors of impaired glucose tolerance and type 2 diabetes in a predisposed population--A prospective cohort study.

BACKGROUND: We characterized in detail (oral and intravenous glucose tolerance tests (OGTT and IVGTT), euglycemic hyperinsulinemic clamp, adipose tissue biopsy), healthy first-degree relatives (FDR) of individuals with type 2 diabetes (T2D), to examine predictive factors for future development of impaired glucose tolerance (IGT) or T2D. METHODS: Non-diabetic FDR (n = 138, mean age 40.5 ± 6.5 years, 57 % women) underwent an extended OGTT every 3 years to assess any deterioration in glucose tolerance status. Differences between groups were assessed by logistic fit for continuous variables and by contingency analysis for categorical variables. Multiple logistic regression analysis was applied to adjust for confounding variables. RESULTS: At follow-up (mean 5.6 ± 2.4 years) 19 subjects had IGT and 4 had T2D. At baseline these 23 subjects had more family members with T2D, higher fasting plasma glucose, higher OGTT plasma glucose at 120 min, higher HbA1c, lower M-value and higher total cholesterol compared to subjects with normal glucose tolerance (NGT). There were significantly larger changes in weight, BMI, fasting plasma glucose, OGTT plasma glucose at 120 min and HbA1c in individuals developing IGT or T2D during the follow-up period than the subjects remaining NGT. Crude predictors of deteriorating glucose tolerance were age, family history of diabetes and of hypertension, OGTT plasma glucose levels at 60 min, 90 min, and 120 min, as well as serum bilirubin, ALP and creatinine (p-values <0.05). A multiple nominal logistic regression model revealed that male sex, low M-value and high physical exercise (p-values <0.05) predicted development of IGT/T2DM. CONCLUSION: In sum, genetically predisposed individuals for T2D with deteriorating glucose tolerance exhibit insulin resistance as well as beta-cell and signs of adipose tissue dysfunction, emphasizing the multifactorial pathophysiology in the development of IGT and T2D.