Variabilità genomica e di espressione di miRNA candidati nella neuropatia diabetica

Diabetic neuropathy (DN) is the most common microangiopathic complication of type 2 diabetes (T2D), leading to the greatest morbidity and mortality. The neuropathies developing in patients with diabetes are heterogeneous in terms of symptoms, pattern of neurological involvement, pathological alterations and underlying mechanisms. However, the most common type of DN are diabetic peripheral neuropathy (DPN) and cardiovascular autonomic neuropathy (CAN). DPN e CAN affect respectively 30% and 20% of diabetic patients, with a prevalence that reaches values of more than 50% in patients with longer duration or higher age. The etiology and many aspects of the pathogenesis of diabetic neuropathy are still unknown; however, based on the current pathogenic model, the interaction between genetic, epigenetic and environmental factors seems to contribute to this diabetes complication. Among the environmental factors, it is known that hyperglycemia is the main risk factor which leads to increased oxidative stress, mitochondrial dysfunction and cell death in the affected organs, including peripheral nerves. Genetic factors also play an important role: several genetic variants associated with DN development are reported in literature. In addition, epigenetic factors have emerged in recent years, including microRNAs (miRNAs), which play an active role in the development of diabetes and its complications. Until now, however, few studies have been conducted to analyze possible miRNA alterations in relation to DN development. The aim of the present study was to investigate a possible involvement of candidates miRNAs in DN. Specifically, the objectives of this doctoral project were the following: 1. to evaluate, in a group of 150 T2D patients undergoing neurological evaluation, the possible association of the rs3746444 polymorphism of MIR499a with the susceptibility to DPN and CAN and any correlations with disease phenotypes and clinical parameters; 2. to identify any differences in the expression of candidates miRNAs between positive and negative patients for the two forms of diabetic neuropathy and evaluate the ability of the expression levels of the associated observed miRNAs, to discriminate between the various neuropathic groups, through the construction of ROC curves; 5 3. to verify if the SNPs localized in the examined miRNA genes can influence the miRNAs expression levels; 4. in light of the results obtained, to investigate the relationship between diabetic neuropathy, oxidative stress and MIR499a, evaluating any differences in the number of copies of mitochondrial DNA (mtDNA) between positive and negative DPN and CAN patients and compared to healthy controls and investigate a possible correlation of the number of copies of mtDNA with the rs3746444 polymorphism of the MIR499a. We observed that the GG genotype of rs3746444 was associated with a higher risk of developing CAN (P=0.002 and OR=16.08, P=0.0005 and OR=35.02, for early and confirmed CAN, respectively) and DPN (P=0.037 and OR=6.56), after correction for BMI, sex, age, HbA1c and disease duration. Moreover, the GG genotype was associated with worse values of MDNS (Michigan Diabetic Neuropatic Score) (P=0.017), VPT (Vibration Perception Threshold) (P=0.01), thermal thresholds (P=0.01), and CAN score (P=0.001). A logistic multivariate analysis confirmed that MIR499A GG genotype, disease duration and HbA1c contributed to early CAN (R2 =0.26), while the same variables and age contributed to DPN (R2 =0.21). With a multiple linear regression, we observed that GG genotype (P=0.001) and disease duration (P=0.035) were the main variables contributing to the CAN score (R2 =0.35). Then, we analysed the expression of six candidate miRNAs in blood of patients evaluated for diabetic neuropathy. In DPN patients we observed a higher expression of miR-128aa (P=0.015), while miR-155 and miR-499a seem to be down-expressed (P=0.04 and P=0.05, respectively). We observed a lower expression of miR-155 (P=0.05) in patients with CAN with respect to CAN-negative patients. Furthermore, we observed that the rs767649 polymorphism variant allele in the miR-155 promoter region was associated with a higher expression of this miRNA (P=0.003) compared to the wild-type allele. Moreover, we performed the ROC analysis for the miRNAs that showed significant differences in expression levels in patients with T2D according to the presence of DPN and CAN, with the aim to evaluate the capacity of these miRNAs to identify the patients with T2D at higher risk to develop neuropathy. The ROC analysis suggested that miR-128aa and miR155 have the ability to discriminate the participants with T2D at higher risk to develop DN (AUC = 0.801) and DPN (AUC = 0.802). 6 Based on the interesting results obtained on the miR-499a, we investigated one of the pathways in which it is involved. Mir-499a targets calcinenurin (CnA) and inhibits its expression, attenuating dynamin-like protein 1 (Drp1) activity, which regulates mitochondrial fission. In view of this, we have decided to investigate the amount of mtDNA in PBMC of T2D subjects, evaluated for CAN and DPN, and of healthy subjects, and then to analyze the possible correlations between the number of mtDNA copies and the rs3746444 SNP genotype in MIR499A gene. We found a significant mtDNA decrease in patients with DPN compared with those without (P=0.02), while no differences were observed between subjects with and without CAN. Furthermore, the homozygous variant genotype for the polymorphism rs3746444 of MIR499A correlated with a decrease in the number of mtDNA copies, particularly in T2D participants (P=0.009). The present study highlighted the interest aroused by microRNAs, their functions and their alterations. On the hand, the identification of specific miRNAs expression differences between diabetic patients with or without neuropathy and the observation that some of these differences are related to a specific form of neuropathy, suggest that these molecules could potentially be evaluated as biomarkers for identify the subjects at higher risk of developing DN among diabetic patients. At the same time, the analysis of the pathways allowed to speculate on the possible role that the three miRNAs described above could play in the pathogenesis of diabetic neuropathy and provided useful information to generate a new hypothesis about a possible mechanism to be investigated. Indeed, starting from the results about miR-499a, we decided to investigate the mechanism in which one of its target genes was involved, providing further evidence of the central role occupied by oxidative stress in DN. The identification of microRNAs associated with a specific pathological state can therefore provide indications on the physiological mechanisms altered or involved in the etiology of the disease. Understanding the role played by these RNA molecules within the pathology examined is also the first necessary step to propose them also as a possible therapeutic target