Omics Approaches in Diabetes: Exploring new biomarkers

Over the years, the prevalence of diabetes has increased globally, reaching epidemic proportions. Diabetes mellitus is a serious and lifelong condition commonly characterized by abnormally elevated blood glucose levels due to a failure in insulin production or a decrease in insulin sensitivity and function, and it is classified as one of the leading causes of high mortality and morbidity rates. Diabetes mellitus is a heterogeneous disease characterized by hyperglycemia, resulting from the body’s inability or impairment in the production and secretion of insulin, but insulin action is often compromised. Several types of diabetes have been identified, the most common is type 1 diabetes (T1D), which results from β-cell destruction due to autoimmunity, and type 2 diabetes (T2D), characterized by impaired insulin secretion and action, but preserved β-cell activity, although defective. Today it’s known that a proportion of patients with adult-onset diabetes share genetic, immunologic, and metabolic features of both T1D and T2D. This is a slowly evolving form of autoimmune diabetes, described as latent autoimmune diabetes of adults (LADA), often mistakenly diagnosed as T2D because it does not require insulin treatment, but it has immunogenetic markers associated with T1D. The aim of my thesis was to identify which metabolites and lipids are characteristic of the different types of diabetes and can be used as biomarkers for the kind of diabetes and the response to treatment. The first part of the thesis focused on alterations and differences in autoimmune diabetes (LADA and T1D) compared to rheumatoid arthritis (RA), as a control autoimmune disease, and compared to T2D. The second part focused on the effect of dapagliflozin, an inhibitor of the sodium-glucose cotransporter-2 (SGLT2i) widely used for the treatment of hyperglycemia, but also for heart failure, on the lipidomic and metabolomic profile of plasma and urine, in order to identify further mechanisms of action that explain how the effects of SGLT2i go beyond the improvement of hyperglycemia and are independent of changes in diuresis. The results showed that autoimmune diabetes is heterogeneous with different metabotypes in LADA and T1D. In particular, autoimmune diabetes is characterized by a low Kynurenine/Tryptophan ratio (Kyn/Tr) (lowest in T1D, followed by LADA and then RA). Kyn/Tr was directly associated with fasting C-peptide levels in LADA identifying Kyn/Tr as a novel marker of beta-cell in autoimmune diabetes, also confirmed by results in human pancreatic islets treated with proinflammatory cytokines. A clear metabolic clustering was observed between LADA and T1D in terms of lipids and metabolites, with an overlap for the profile of T1D and T2D. Compared to LADA and T2D, T1D had higher values of ceramides and some lysophosphatidylcholines. Among the metabolites, aromatic amino acids (AAA) were higher in T1D, in particular a higher concentration of tryptophan, underlining its role in diabetes. The 4 weeks treatment of the SGLT2i dapagliflozin (DAPA) changed only few plasma metabolites, i.e., increased in isoleucine, methionine, citrate, β-hydroxybutyrate and reduced lactate while there was a significant increase in plasma concentrations of unsaturated free fatty acids (FFA), and an increase in some sphingomyelins (SMs) and lysophosphatidylcholines (LPC) containing these fatty acids. A significant change was observed in medium- and short-chain acylcarnitines positively correlated with changes in plasma FFAs and b-hydroxybutyrate. In addition, DAPA significantly increased (independently of increased diuresis) the 24-h urinary excretion of many AA, lactate, TCA cycle metabolites, β-hydroxybutyrate and electrolytes, except for a decrease in malate excretion showing that DAPA treatment has effects that go beyond reduction of hyperglycemia and fatty acid oxidation, including a possible role in the excretion of AA, which should be monitored since it could potentially lead to excess AA loss. In conclusion, the use of metabolomic and lipidomic analyses are useful tools to identify biomarkers and alterations of metabolic diseases such as diabetes and also to identify the mechanisms of action of drugs approved for the treatment of hyperglycemia. These results help to better understand the mechanisms that regulate glucose and lipid metabolism potentially responsible for the development and progression of diabetes and to identify metabolic improvements achieved by pharmacological and non-pharmacological interventions that can be used for personalized interventions in a precision medicine approach.