Significance Of Serum MicroRNAs During The Pathogenesis Of Type2Diabetes:a Clinical Study

Research backgroundPre-diabetes refers to an abnormal glycaemia status, characterized by impaired glucose tolerance (IGT) and/or impaired fasting glucose (IFG). Pre-diabetes is highly prevalent and pre-diabetic patients have much higher chance to develop to overt diabetes; on the other hand, interventions in pre-diabetic period (lifestyle or drugs) can effectively delay or even convert the development to overt T2D (type2diabetes), so pre-diabetic stage is very significant in T2D prevention and treatment.Pre-diabetes is the pathogenetic procedures from diabetes-susceptible individuals to overt T2D patients. Accumulation of the abnormal expression of T2D related genes finally triggers some key pathophysiological changes, which make the pathogenetic procedures unconvertible. How to evaluate if the pathogenesis of pre-diatetic patients is still reversible, will not only be helpful for a deeper understanding of the molecular biological basis of pre-diabetes, but also has important clinical significance. In order to determine if the accumulation of related abnormal genes expressionhas reached the "critical point",FPG or2-hour PG or other external phenotypic index are not reliable. Ideally the judgement should be made based on the expression of related genes.MicroRNAs (miRNAs) are a class of endogenous, small (21-23nucleotides in length), non-coding but functional RNAs, serving as the posttranscriptional regulators of gene expression through complementary base pairing to messenger RNAs. Since being discovered, miRNAs have been proven to play significant roles in a wide spectrum of diseases including T2D. Altered miRNA expression profiles are observed in a range of different diseases as well.MiRNAs are present in human peripheral blood at consistent and reproducible levels. For T2D research, clinical tissue samples (esp. pancreas) can hardly be obtained, so using peripheral miRNAs as biomarkers in T2D patients would be of special significance.This study explores during the pathogenesis of T2D, expression levels and patterns of7T2D related circulating miRNAs in peripheral blood. Mir-9negatively regulates the release of insulin; expression of miR-124a2decreases mRNA level of insulin; Over-expression of miR-375suppresses glucose-induced insulin secretion, and conversely, inhibition of endogenous miR-375function enhances insulin secretion; Elevated expression of mir-29exists in three important target tissues of insulin action: muscle, fat, and liver of diabetic rats, over-expression of miR-29a/b/c in3T3-L1adipocytes largely represses insulin-stimulated glucose uptake, and causes insulin resistance; Over-expression of miR-30d increases insulin gene expression, while inhibition of miR-30d abolishes glucose-stimulated insulin gene transcription; Elevated miR-34a impairs nutrient-induced secretion, while miR-146contributes to increased apoptosis, detrimental effects of palmitate on (3-cells could be related to mir-34a and miR-146.ObjectivesTo explore during the pathogenesis of T2D, from susceptible individuals to pre-diabetes and finally to overt T2D, expression and clinical significance of7circulating T2D-related miRNAs (miR-9, miR-29a, miR-30d, miR-34a, miR-124a, miR-146a, miR-375), and study the expressing features of T2D-relatd miRNAs in pre-diabetic stages, and the feasibility of predicting the development of pre-diabetes by the expression patterns of T2D-related miRNAs.Materials and methodsResearch subjects:Fifty-six subjects were recruited to this study:18cases of newly diagnosed T2D (n-T2D) patients,19cases of pre-diabetes individuals (impaired glucose tolerance [IGT] and/or impaired fasting glucose [IFG]) and19cases of T2D-susceptible individuals with normal glucose tolerance (s-NGT).Diagnosis:A2-hour OGTT (Oral Glucose Tolerance Test,75g of glucose) was performed in this study whilst samples for plasma glucose (PG) were drawn at0and120minutes. Glucose tolerance was evaluated based on OGTT as follows:NGT (FPG <5.6mmol/1and2-hour PG<7.8mmol/1), isolated IFG (FPG5.6-6.9mmol/1and2-hour PG<7.8mmol/1), isolated IGT (FPG<5.6mmol/1and2-hour PG between7.8and11.0mmol/1), IFG+IGT (FPG5.6-6.9mmol/1and2-hour PG7.8-11.0mmol/1), and newly diagnosed type2diabetic subjects (FPG≥7.0mmol/1and/or2-hour PG≥11.1mmol/1). Measurement:Total serum RNA was extracted by mirVana Column, serum miRNAs were detected via real-time RT-PCR by using TaqMan(?) MicroRNA Assays kits from ABI.Statistics:All statistical tests were two-sided. The statistical significance was defined as P<0.05. Statistic analysis was performed by SPSS software package for Windows version16.0. One-way ANOVA or Chi-square was employed to compare age, BMI, blood glucose, serum lipid and sex distribution among three subject groups. For qRT-PCR data, the difference of threshold cycle (Ct) between the target miRNAs and RNU6B (△Ct) was employed to show the relatively expression levels of7target miRNAs. The statistical P value was generated by the one-way ANOVA and the following LSD post hoc multiple comparisons. One-way ANOVA was also employed to compare the RNA concentrations among the three groups of subjects. To carry out an agglomerative hierarchical cluster analysis of the subjects based on miRNAs expression, we used a between-groups linkage algorithm and a squared Euclidean distance measure. For canonical discriminant analysis expression levels of all seven miRNAs were included to generate Canonical Discriminant Functions predicting the classification of three groups based on miRNAs expression.ResultsClinical data:No significant difference in sex distribution, age, BMI was found among three subjects groups. For OGTT results, both FPG and2-hour PG showed significantly higher stepwise trends from s-NGT to pre-diabetes and then n-T2D. Triglyceride (TG) also showed significantly higher stepwise trends from s-NGT to n-T2D.Seven miRNAs were detectable in serum samples from three groups:Total RNA containing miRNAs, which was qualified for the following qRT-PCR analysis, was obtained from all subjects. No significant differences in RNA concentrations was discovered among three groups (s-NGT:7.311±0.6852; pre-diabetes:9.429±1.880; n-T2D:7.407±1.223; ng/μl; mean±SEM; P=0.4626, one-way ANOVA). In the following real-time RT-PCR step, all seven miRNAs including RNU6B showed reliable threshold cycle (Ct) values in most of56samples. In about3%of the assays, the fluorescent signals of miRNAs fail to reach the set threshold after40cycles. In these assays, Ct values were calculated as40.Expression levels of seven serum miRNAs were significantly elevated in n-T2D patients:All7miRNAs studied showed significant up-regulation in sera from n-T2D group compared with the s-NGT group.4miRNAs showed P value below0.01(miR-29a, P=0.0030; miR-34a, P=0.0000; miR-146a, P=0.0080; miR-375,P=0.0040; one-way ANOVA and LSD post hoc multiple comparisons). The biggest difference was seen in miR-34a expression.Expression patterns of seven serum miRNAs in pre-diabetes were similar to NGT than T2D:The expression patterns of7miRNAs in the pre-diabetes group were similar to the s-NGT group. Except for miR-375, all other6miRNAs showed elevated expression to some extent, but their expression levels were so close to those in the s-NGT groups in that no significant difference, even of borderline statistical significance, was observed. When the expression of these7miRNAs in the pre-diabetes group were compared with those in the n-T2D group,5miRNAs showed significantly lower expression:miR-9(P=0.0210), miR-29a (P=0.0030), miR-34a (P=0.0010), miR-146a (P=0.0100) and miR-375(P=0.0020), with the difference between means of these5miRNAs ranged from-2.377±1.552to-3.747±1.285(mean±SEM).n-T2D features specific expression signatures of seven serum miRNAs:Canonical discriminant analysis showed that among all seven miRNAs, miR-34a showed the highest standardized Canonical Discriminant Function Coefficients, which was in accordance with the previous results from student t-test, indicating its most significant difference. In general,67.3%of original grouped cases were correctly classified, among which70.6%of n-T2D subjects had a correct judgement, while the number of pre-diabetes and s-NGT subjects was63.2%and68.4%, respectively. The scatter plot shows that in Function1(P=0.0151, Wilks'Lambda) the centroids of s-NGT (-0.4130) and pre-diabetes (-0.4980) are very close, while the centroid of n-T2D (1.018) is separated from them. In the Hierarchical Clustering analysis of the study subjects from all three groups, two sections were generated based on the expression signatures of seven serum miRNAs, and12out of17n-T2D subjects were clustered together within the left section, while the distribution of s-NGT and pre-diabetes subjects mixed with each other, and no significant cluster of either group can be found. These results suggest that N-T2D featured relatively specific expression signatures of seven serum miRNAs compared with s-NGT and pre-diabetes.ConclusionsIn pre-diabetes and s-NGT groups, expression levels of peripheral T2D-related miRNAs were similar, no significant difference were found; compared with pre-diabetes and s-NGT groups, when the pathogenesis reached over T2D stage (n-T2D group), expression levels of serum T2D-related miRNAs increase significantly, showing significant difference comparing to the previous two groups. When put7miRNAs into a panel and analyze their expression patterns, we found similar expression patterns in pre-diabetes and s-NGT groups, the Canonical Discriminant Function cannot make reliable judgement between them, and these two groups mixed together in the Hierarchical Clustering analysis. While for n-T2D group, the Canonical Discriminant Function can make reliable judgement, and most of them clustered in the same group in the Hierarchical Clustering analysis. These results proved that in the pre-diabetic stage during the pathogenesis of T2D, although the homeostasis of blood glucose has changed, as the key regulator of gene expression, expression levels and patterns of miRNAs have not changed significantly, but remained similar to the individuals with normal glucose tolerance. Till finally when the pathogenesis reaches to the over T2D, significant changes happen in the levels and patterns of miRNAs. This can partly explain why interventions in pre-diabetic period (lifestyle or drugs) can effectively delay or even convert the development to overt T2D.