Skip main navigation

Plasma MicroRNA Profiling Reveals Loss of Endothelial MiR-126 and Other MicroRNAs in Type 2 Diabetes

Originally publishedhttps://doi.org/10.1161/CIRCRESAHA.110.226357Circulation Research. 2010;107:810–817

Abstract

Rationale:

MicroRNAs (miRNAs) have been implicated in the epigenetic regulation of key metabolic, inflammatory, and antiangiogenic pathways in type 2 diabetes (DM) and may contribute to common disease complications.

Objective:

In this study, we explore plasma miRNA profiles in patients with DM.

Methods and Results:

Total RNA was extracted from plasma samples of the prospective population-based Bruneck study. A total of 13 candidate miRNAs identified by microarray screening and miRNA network inference were quantified by quantitative PCR in all diabetic patients of the Bruneck study and age- and sex-matched controls (1995 evaluation, n=80 each). Quantitative PCR assessment revealed lower plasma levels of miR-20b, miR-21, miR-24, miR-15a, miR-126, miR-191, miR-197, miR-223, miR-320, and miR-486 in prevalent DM, but a modest increase of miR-28-3p. Findings emerged as robust in multivariable analysis and were independent of the standardization procedure applied. For endothelial miR-126, results were confirmed in the entire Bruneck cohort (n=822) in univariate (odds ratio [95% confidence interval], 0.38 [0.26 to 0.55]; P=2.72×10−7) and multivariate analyses (0.57 [0.37 to 0.86]; P=0.0082). Importantly, reduced miR-15a, miR-29b, miR-126, miR-223, and elevated miR-28-3p levels antedated the manifestation of disease. Most differences in miRNA levels were replicated in plasma obtained from hyperglycemic Lepob mice. High glucose concentrations reduced the miR-126 content of endothelial apoptotic bodies. Similarly in patients with DM, the reduction of miR-126 was confined to circulating vesicles in plasma.

Conclusions:

We reveal a plasma miRNA signature for DM that includes loss of endothelial miR-126. These findings might explain the impaired peripheral angiogenic signaling in patients with DM.

MicroRNAs (miRNAs) are a class of small noncoding RNAs that function as translational repressors. They bind through canonical base pairing to a complementary site in the 3′ untranslated region of their target mRNAs and can direct the degradation or translational repression of these transcripts.12 MiRNAs have been shown to play important roles in development, stress responses, angiogenesis, and oncogenesis.34 Accumulating evidence also points to an important role of miRNAs in the cardiovascular system.45 For example, miRNAs modulate endothelial cell function and regulate their inflammatory response and angiogenic potential.67

Recently, Mitchell et al highlighted the presence of miRNAs in plasma.8 These plasma miRNAs are not cell-associated, but packaged in microvesicles that protect them from endogenous RNase activity. Interestingly, plasma miRNAs can display unique expression profiles: specific tumor miRNAs were identified in cancer patients,9 whereas tissue-derived miRNAs constitute a marker for injury.1011 In cardiovascular diseases, circulating miRNAs have been investigated in myocardial injury, coronary artery disease, and heart failure.1215 Type 2 diabetes mellitus (DM) is one of the major risk factors of cardiovascular disease leading to endothelial dysfunction and micro- and macrovascular complications.1617 However, a systematic analysis of plasma miRNAs in DM has not yet been performed.

The present study is the first to reveal a plasma miRNA signature for DM in a large population-based cohort. Our findings may provide new insights into the pathophysiology of DM and the manifestation of its vascular complications.

Methods

An expanded Methods section is available in the Online Data Supplement at .

Study Subjects

The Bruneck study is a prospective population-based survey initially designed to investigate the epidemiology and pathogenesis of atherosclerosis and later extended to study all major human diseases including DM.1820 At the baseline evaluation in 1990, the study population was recruited as a sex- and age-stratified random sample of all inhabitants of Bruneck (Bolzano Province, Italy) 40 to 79 years old. During 1990 and the re-evaluation in 1995 (the first five-year period), a subgroup of 63 individuals died or moved away. In the remaining population follow-up was 96.5% complete (n=822). RNA extraction was performed from plasma specimens collected as part of the 1995 follow-up in 822 individuals. Follow-up in 2000 and 2005 was 100% complete for clinical end points and >90% complete for repeated laboratory examinations. Assessment of the ankle-brachial index and incident peripheral artery disease is described in detail in the Online Data Supplement. The Bruneck population is highly representative of the general community and shows characteristics similar to those of other Western countries. The mean age was 62.9 years, 49.9% were women and the prevalence of DM was 9.7%. Their clinical characteristics are summarized in Online Table I. The protocols of the Bruneck study were approved by the appropriate Ethics Committees, and all study subjects gave their written informed consent before entering the study.

Ascertainment of DM

Presence of DM was established according to World Health Organization criteria, ie, when fasting glucose was ≥7 mmol/L (126 mg/dL) or when the 2-hour oral glucose tolerance test glucose level was ≥11.1 mmol/L (200 mg/dL) or when the subjects had a clinical diagnosis of the disease. Self-reported DM status was obligatorily confirmed by reviewing the medical records of the subject's general practitioners and files of the Bruneck Hospital (for details, see the Online Data Supplement).

MiRNA Expression Profile

RNA extraction was performed using the miRNeasy kit (Qiagen) from plasma specimens collected as part of the 1995 follow-up in 822 individuals. MiRNAs were reverse-transcribed using the Megaplex Primer Pools (Human Pools A v2.1 and B v2.0), and expression was screened using TaqMan miRNA Arrays A and B (all from Applied Biosystems). TaqMan miRNA assays were used to determine the expression of individual miRNAs. Additional details are provided in the Online Data Supplement.

Sampling Strategy and Statistical Analysis

For the initial microarray screening, 2 pools of plasma obtained from 5 individuals with DM each (randomly selected from the 80 patients with DM in the Bruneck study) and 6 pools of plasma obtained from controls identical in age, sex, risk factor profile (low-density lipoprotein, smoking, hypertension), and atherosclerosis status were used. Quantitative (q)PCR was performed for the 13 miRNAs that showed correlation with DM. TaqMan assays were run in duplicates. Two groups of patients were assessed: group 1, all subjects with manifest DM in the Bruneck cohort (1995, n=80); group 2, 19 patients who developed DM between 1995 and 2005 (incident DM). Eighty and 19 matched individuals identical for age and sex with fasting glucose levels of <6.1 mmol/L (110 mg/dL), 2-hour glucose of <7.7 mmol/L (140 mg/dL), and no history of DM served as controls. In the case of several suitable matches, all were numbered in ascending order and one was selected by means of a computer-based random number generator. Finally, miR-126 was measured in the entire population of 822 individuals. For the lack of generally accepted standards all qPCR data were analyzed as unadjusted Ct values and standardized to both miR-454 and RNU6b, a small nuclear RNA, which fulfilled the following criteria: detectable in all samples, low dispersion of expression levels and null association with DM status. Moreover, the miR-454 expression profile showed little association with other miRNAs and positioned outside the coexpression modules of the complex miRNA network in plasma (Figure 1). Data were analyzed using SPSS version 15.0 and STATA version 10 software packages. Continuous variables were presented as means±SD or median (interquartile range) and dichotomous variables as numbers and percentages. Fold changes of individual miRNAs were calculated for each pair of matched (pre)diabetic cases and controls by dividing the standardized expression levels of the miRNAs. The median of fold changes is presented in Figure 2. Differences in miRNA levels between subjects with prevalent or incident DM and corresponding groups of matched controls were analyzed using the nonparametric Wilcoxon test for related samples with computation of exact probability values. To account for the potential confounding effects of lifestyle features and other variables related to DM and to analyze for interactions, we additionally performed logistic regression analyses for matched data that include loge-transformed expression levels of miRNAs (1 per model) and the following variables: social status, family history of DM, body mass index, waist-to-hip ratio, smoking status, alcohol consumption (g/d), physical activity (sports index), and high-sensitivity C-reactive protein. Details on model construction were described by Hosmer and Lemeshow.21 First-order interactions between miRNAs and the above variables, as well as age and sex, were calculated by inclusion of appropriate interaction terms. None of these terms achieved statistical significance. Cox proportional hazard models were used to assess the predictive value of miR-126 for new-onset peripheral artery disease. Participants who experienced an outcome event were censored with respect to subsequent follow-up. The proportional hazard assumption was confirmed for miR-126 by testing the interaction of miR-126 with a function of survival time (Cox models with time-dependent covariates). A total of 37 subjects with symptomatic peripheral artery disease at baseline were excluded, leaving 785 subjects for this analysis. The potential association between miR-126 and low ankle–brachial index was analyzed by means of logistic regression analysis. This analysis was fit to the entire study population. Differences in miR-126 between categories of glucose tolerance were compared with General Linear Models. All probability values presented are 2-sided.

Figure 1.

Figure 1. MiRNA coexpression network and miRNA topology values.A, Weighted and undirected miRNA coexpression network. Nodes correspond to miRNAs and edges (links) indicate similarity in miRNA expression, as measured by the PCC. Strength of similarity is indicated by a red-blue edge color gradient. At PCC values of >0.85, the coexpression network consisted of 120 miRNAs and 1020 coexpression links. B, Relationships of node degrees (number of coexpressing pairs for each miRNA), clustering coefficients (likelihood of miRNA sharing a large number of neighbors), and eigenvector centralities (relative contribution of miRNA to stability of the network) for 120 miRNAs. There were 30 differentially expressed miRNAs (blue), of which 13 (red) occupied locations important to maintenance of the overall network structure.

Figure 2.

Figure 2. Association of plasma miRNAs with manifest DM. Thirteen plasma miRNAs were quantified by qPCR in patients with prevalent DM and matched controls (n=80 each). Bars in the center of each graph represent fold changes of plasma miRNA levels between diabetic patients and controls. Bars on the left provide a comparison (fold changes) between plasma obtained from hyperglycemic Lepob and control mice. Squares and lines on the right represent odds ratios (95% CIs) derived from multivariable logistic regression analyses for matched data (human studies). Probability values were from the nonparametric Wilcoxon test for related samples (human studies, center) and Mann–Whitney U test for unrelated samples (animal studies), and from multivariable logistic regression analyses for matched data (human studies).

MiRNA Coexpression Network Inference and Analysis

Network inference algorithms were applied to evaluate global expression properties of miRNAs in DM. Similarity in miRNA expression profiles was interrogated using either Pearson correlation coefficients (PCCs) or context likelihood of relatedness (CLR) between all possible miRNA pairs.22 Pairs that maintained dependence above a predefined threshold were represented in the form of an undirected weighted network, where nodes correspond to miRNAs and links (edges) correspond to the similarity between them. Although PCC is a way to measure linear relationships between features (miRNAs), CLR relies on a mutual information metric and does not assume linearity,2324 thus possessing some flexibility to detect biological relationships that may otherwise be missed. PCC was used to detect clusters of similarly expressed miRNAs from a high throughput space of expression arrays, whereas CLR was used to identify all nonrandomly associated qPCR-validated miRNA profiles. Although PCC and CLR sometimes yield similar results, CLR was chosen for qPCR-validated miRNAs because it is more sensitive to nonlinear dynamics of miRNA expression than PCC and significantly outperforms other network inference methods (eg, ARACNE) in identifying biologically meaningful relationships.2526 The PCC threshold was set to a point where miRNA coexpression network began to acquire a scale-free architecture, which is a characteristic of most real-world networks, including biological ones.27 To ensure reproducibility, the 30 differentially expressed miRNAs were evaluated by assessing their network properties,28 rather than the magnitude of over- or underexpression. The CLR threshold was chosen such that all 13 miRNAs could be represented in the network while retaining the smallest possible number of links between them.

Topological Analysis

During prescreening, miRNAs were studied by virtue of their topology in the global miRNA coexpression network, as well as individual over- or underexpression. For each miRNA, topological parameters including node degree, clustering coefficient, and eigenvector centrality were systematically calculated. Node degree is defined as the total number of edges that are connected to a given miRNA. Clustering coefficient is the degree to which miRNAs tend to cluster together. Eigenvector centrality is a measure of miRNA importance, such that a particular miRNA receives a greater value if it is strongly correlated with other miRNAs that are themselves central within the network.

Endothelial Cell Culture

Human umbilical vein endothelial cells (HUVECs) were purchased from Cambrex and cultured on gelatin-coated flasks in M199 medium supplemented with 1 ng/mL endothelial cell growth factor (Sigma), 3 μg/mL endothelial growth supplement from bovine neural tissue (Sigma), 10 U/mL heparin, 1.25 μg/mL thymidine, 5% FBS, 100 μg/mL penicillin and streptomycin as described previously.29,30 HUVECs exposed to high glucose (25 mmol/L) were cultured in complete medium for 6 days. As a control, HUVECs were cultured in complete medium (5 mmol/L glucose) supplemented with mannitol, to exclude effects of osmotic stress. On day 5, cells were counted and an equal number of cells was seeded on T75 flasks and incubated for an additional day.

Isolation of Vesicles

Vesicles were isolated as described previously.31 In brief, HUVECs were deprived of serum and growth factors for 24 hours before the conditioned medium was harvested and the cells were lysed in QIAzol reagent. The cellular lysates were stored at −20°C for miRNA expression analysis. The conditioned medium was first precleared by centrifugation for 10 minutes at 800g to remove any floating cells. Then, endothelial particles (apoptotic bodies) were isolated by centrifugation at 10 600 rpm for 20 minutes. Subsequently, an additional centrifugation step was performed at 20 500 rpm for 2 hours to isolate small microparticles (size, <1 μm) shedding from endothelial cells. Identical centrifugation steps were performed to obtain vesicles from plasma. The pelleted vesicles were resuspended in PBS and stored at −80°C. Total RNA was extracted using the miRNeasy kit as described above. RNA was quantified using the NanoDrop spectrophotometer and 20 ng of total RNA were used for reverse transcription.

Results

Comprehensive MiRNA Profiling

For the initial screening, Human TaqMan miRNA arrays (CardA v2.1 and CardB v2.0, Applied Biosystems) covering 754 small noncoding RNAs were applied to 8 pooled samples, 2 consisting of subjects with DM, and 6 of appropriate controls (see methods). Of the 148 miRNAs with Ct values of <36, 130 miRNAs were detected using fluidic Card A and therefore all further analysis focused on this data set, resulting in the identification of 30 differentially expressed plasma miRNAs in patients with DM (Online Table II).

MiRNA Network Analysis

MiRNA pairs with high correlation values (PCC >0.85) were represented in the form of an undirected and weighted network. At this threshold, the network was dominated by a small number of hubs that linked with many loosely connected nodes, a property of many biological networks (Online Figure I). The miRNA network consisted of 120 miRNAs (nodes) and 1020 coexpression links (edges) (Online Table II). The 30 differentially expressed miRNAs were sampled by virtue of their localization in the miRNA coexpression network as marker selection using network topology is more reproducible than assessment of individual over- or underexpression.28 Within the network, the 30 differentially expressed miRNAs were topologically central (Online Figure II; Online Table III). 13 of the 30 differentially expressed miRNAs that displayed extreme spectra of node degrees, clustering coefficients, and eigenvector centrality values were selected (Figure 1B). MiR-454 was the only miRNA that showed no association with the expression of other miRNAs and was positioned outside all network modules (Figure 1A). Thus, it was chosen as an additional normalization control.

Validation by qPCR

The 13 topologically unique miRNAs were further quantified by qPCR. All patients with manifest DM at the time of the 1995 evaluation (n=80) were compared to age- and sex-matched controls (Online Table I). Plasma levels of miR-24, miR-21, miR-20b, miR-15a, miR-126, miR-191, miR-197, miR-223, miR-320, miR-486, miR-150, and miR-29b were lower in diabetic subjects, whereas miR-28-3p tended to be higher (Figure 2). Findings were consistent for expression levels standardized to either miR-454 or RNU6b (Figure 2) and for nonstandardized miRNA levels (data not shown). Nine miRNAs standardized to RNU6b showed significant differences between patients with DM and controls, and 4 remained significant after accounting for the multiple comparisons performed (Bonferroni probability value <0.000133), including endothelial miR-126. Bonferroni correction, however, is overly conservative in this setting because individual miRNAs are not independent of each other but extensively correlated. In multivariate analyses, all miRNAs except miR-29b were significantly associated with manifest DM. Eleven showed an inverse relationship and miR-28-3p a positive one. Findings for miRNAs standardized to miR-454 are summarized in Figure 2 and were similar in diabetic patients with or without drug treatment (mainly sulfonylureas). In another run, miR-126 was quantified in the entire Bruneck cohort (n=822) and standardized against miR-454. In logistic regression analyses, miR-126 emerged again as a significant predictor of manifest DM (odds ratio [95% confidence interval {CI}] for a 1-SD unit decrease of loge-transformed expression level of miR-126, 2.23 [1.69 to 2.96]; P=2.28×10−8), and this association persisted in a multivariable model (odds ratio [95% CI], 1.64 [1.19 to 2.28]; P=0.0027). Moreover, there was a gradual decrease in plasma levels of miR-126 across categories of normal glucose tolerance (n=580), impaired fasting glucose/impaired glucose tolerance (n=162) and manifest DM (n=80) (Figure 3). As expected, levels of miRNAs were inversely correlated with fasting glucose levels in both patients with DM, as well as control subjects (r≈−0.191 to −0.335, P<0.05 each except miR-29b and miR-320) and less consistently with 2-hour glucose levels (r=−0.091 to −0.239) and HbA1c concentrations (r=−0.093 to −0.312). Most of the miRNA changes observed in DM could independently be replicated in plasma samples of 8 to 12 week old hyperglycemic Lepob mice (Figure 2).

Figure 3.

Figure 3. Plasma levels of miR-126 across categories of normal glucose tolerance (NGT), impaired fasting glucose/impaired glucose tolerance (IFG/IGT), and manifest DM.Squares and lines indicate adjusted geometric means and 95% CIs (white squares, values adjusted for age and sex; black squares, values adjusted for age, sex, social status, family history of DM, body mass index, waist-to-hip ratio, smoking status, alcohol consumption (g/d), physical activity (sports index), and high-sensitivity C-reactive protein). This analysis was performed in the entire study population (n=822). Differences in miR-126 between categories of NGT, IFG/IGT and DM were compared with General Linear Models and probability values are for trend. World Health Organization definition of categories: normal glucose tolerance (fasting glucose, <110 mg/dL; 2-hour glucose, <140 mg/dL), impaired fasting glucose/impaired glucose tolerance (110 mg/dL≤fasting glucose<126 mg/dL, 140 mg/dL≤2-hour glucose<200 mg/dL) and manifest DM. American Diabetes Association definition of categories: normal glucose tolerance (fasting glucose, <100 mg/dL; 2-hour glucose, <140 mg/dL), impaired fasting glucose/impaired glucose tolerance (100 mg/dL≤fasting glucose<126 mg/dL, 140 mg/dL≤2-hour glucose<200 mg/dL) and manifest DM.

Incident DM

Importantly, several miRNAs were already altered before manifestation of DM. A total of 19 subjects, who were normoglycemic in 1995, developed DM over the 10-year follow-up period (1995 to 2005 mean interval until diagnosis of DM 79.2 months). Baseline levels of miR-15a, miR-29b, miR-126, and miR-223 were significantly lower in these subjects, whereas miR-28-3p was higher compared to matched controls (Figure 4).

Figure 4.

Figure 4. Association of plasma miRNAs with incident DM. Thirteen plasma miRNAs were quantified by qPCR in patients who developed DM over a 10-year observation period and matched controls. Probability values were derived from the nonparametric Wilcoxon test for related samples. *P<0.05, ***P<0.001.

MiRNAs as Biomarkers in DM

To determine whether miRNAs can correctly distinguish individuals with prevalent or incident DM from healthy controls, we reduced miRNA expression to principal components (PC). Interestingly, 91/99 (92%) controls and 56/80 (70%) DM cases were correctly classified using expression profiles of 5 most significant miRNAs (miR-15a, miR-126, miR-320, miR-223, miR-28-3p) (Figure 5). The 24 DM cases that were classified as normal subjects had significantly lower levels of fasting glucose (mean± SD, 120.0±28.6 mg/dL versus 147.2±55.0 mg/dL, P=0.005) and HbA1c (mean± SD, 6.03±0.75% versus 6.66±1.54%, P=0.016) and represent a selection of patients with well-controlled DM. Importantly, using this model 10/19 (52%) of normoglycemic subjects that developed DM over the 10-year follow-up period were already classified as diabetics. Inclusion of additional miRNAs into the classification did not improve model performance (Figure 5B). Thus, these 5 miRNA can be considered as minimal requirement for classification using a miRNA signature. Further support to the putative value of miRNAs as diagnostic tools in DM was provided by the inference of miRNA relevance network (Online Figure III).

Figure 5.

Figure 5. Classification, PCA, and network properties.A, Classification efficiency of 13 miRNAs across control subjects (n=99), patients with incident DM (n=19), and manifest DM (n=80). Higher score is indicative of a greater degree of differential expression and thus a higher classification potential. B, Classification accuracy. Best classification accuracy was achieved using the top 5 differentially expressed miRNAs. C, PCA, a technique that reduces the data to two or more uncorrelated principal components (PC) that explain most of the variance, was used to determine whether control subjects could be distinguished from cohorts with incident and manifest DM. PCA decomposition of the top 5 miRNAs was sufficient to cluster together 91/99 (92%) controls and 56/80 (70%) patients with manifest DM. Interestingly, 10/19 (52%) patients with incident DM were clustered with cases of manifest disease.

MiRNA-126 in DM

Among the miRNAs most consistently associated with DM was miR-126. This miRNA has previously been shown to be highly enriched in endothelial cells and endothelial apoptotic bodies and to govern the maintenance of vascular integrity, angiogenesis, and wound repair.3435 To determine whether hyperglycemia affects miR-126 release from endothelial cells, miRNA levels of shedding endothelial particles (apoptotic bodies) and microparticles derived under normal (5 mmol/L) and high (25 mmol/L) glucose concentrations were compared. Whereas cellular miRNA concentrations remained unaltered, high glucose significantly reduced the miR-126 content in endothelial apoptotic bodies (Figure 6A). Shedding of other miRNAs except miR-24 was not affected (Online Figure IV). Consistent with these in vitro experiments, the reduction in miR-126 levels in patients with DM was confined to the particulate fraction in plasma (Figure 6B). Finally, evidence from our population cohort suggests that loss of miR-126 in plasma correlates with subclinical and manifest peripheral artery disease. In detail, miR-126 was associated with a low ankle-brachial index (< 0.9, n=77) (unadjusted and age-/sex-adjusted odds ratio [95% CI] for a 1-SD unit decrease of loge-transformed expression level of miR-126, 1.95 [1.48 to 2.57], P<0.001; and 1.39 [1.04 to 1.86], P=0.025) and with new-onset symptomatic peripheral artery disease (1995 to 2005, n=15) (unadjusted and age-/sex-adjusted hazard ratio [95% CI] for a 1-SD unit decrease of loge-transformed expression level of miR-126, 2.63 [1.38 to 5.02], P=0.0032; and 2.15 [1.06 to 4.38], P=0.030). In the latter analysis, 37 subjects with symptomatic peripheral artery disease at baseline were excluded.

Figure 6.

Figure 6. The effect of increased glucose levels on the miR-126 content of vesicles. High glucose concentrations led to a decrease in the miR-126 content of endothelial-derived particles (apoptotic bodies) (A) and circulating vesicles in plasma (B). MiRNA expression was assessed using qPCR. miR-454 was used as a normalization control. The data are from 4 independent experiments and presented as means±SD. *P<0.05.

Discussion

We provide the first evidence for a plasma miRNA signature in patients with DM and a potential prognostic value in this setting. Our findings warrant further investigations into the role of miRNAs in diabetic vascular and myocardial complications.

Plasma MiRNAs in DM

Using differential expression and concepts of network topology, we identified 13 plasma miRNAs in DM including loss of miR-126. Findings were robust in multivariable analyses of patients with DM and age- and sex-matched controls and confirmed in hyperglycemic Lepob mice. Of note, deregulation of several plasma miRNAs antedated the manifestation of DM. Principal component analysis (PCA) of the 13 studied miRNAs indicates that 5 miRNAs (miR-15a, miR-126, miR-320, miR-223, miR-28-3p) with the highest scores are necessary and sufficient for a nonredundant classification. Although the function of the highest scoring miRNA, miR-15a is unknown in DM, miR-15a has been previously implicated in cell cycle control and apoptosis in cancer cells.32 Its expression was shown to inversely correlate with the expression of cyclin D1, although it was also reported to negatively regulate levels of B-cell lymphoma 2 (Bcl-2), a key antiapoptotic protein.33 In Lepob mice, however, miR-15a was not significantly different from wild-type mice, probably attributable to low plasma levels. Clinically relevant questions are whether miRNA levels are capable of assessing the probability of DM manifestation in high-risk individuals like patients with impaired fasting glucose, borderline levels of HbA1c or metabolic syndrome, and whether miRNAs can assist in the prediction of micro- and macrovascular complications in DM. Our data suggest that plasma miRNAs might be a useful predictive tool in DM but await confirmation in larger collectives of patients with DM and prediabetes before more definitive comparisons with other standard risk factors can be made. MiR-126, however, deserves special consideration.

MiR-126 and DM

Plasma levels of miR-126 were determined in the entire Bruneck cohort. This is, to our knowledge, the first time a miRNA has been measured in a large population-based study. Whereas most plasma miRNAs are ubiquitously expressed, miR-126 is highly enriched in endothelial cells and plays a pivotal role in maintaining endothelial homeostasis and vascular integrity.35 It facilitates vascular endothelial growth factor (VEGF) signaling by repressing 2 negative regulators of the VEGF pathway, including the Sprouty-related protein SPRED1 and phosphoinositol-3 kinase regulatory subunit 2 (PIK3R2/p85-β).34 Plasma miRNAs are packaged in membranous vesicles that change in numbers, cellular origin, and composition depending on the disease state.36 Accumulating evidence support the notion that these vesicles are not just byproducts resulting from cell activation or apoptosis. Instead, they constitute a novel type of cell–cell mechanism of communication. For example, miR-126 is the most abundant miRNA in endothelial apoptotic bodies.37 Shedding of miR-126 from endothelial cells has been shown to regulate VEGF responsiveness and to confer vascular protection in a paracrine manner.37 In our study, loss of miR-126 was consistently associated with DM and the miR-126 content in endothelial apoptotic bodies was reduced in a glucose-dependent fashion. Because apoptotic bodies and microparticles can be transferred to other cell types,31,37 low plasma levels might result in reduced delivery of miR-126 to monocytes and contribute to VEGF resistance and endothelial dysfunction. This view is corroborated by previous reports that monocytes of patients with DM show impaired responsiveness to VEGF contributing to defects in collateral vessel development38 and by our finding that loss of miR-126 confers an elevated risk of symptomatic and subclinical peripheral artery disease in the Bruneck population.

Potential Clinical Implications

There are currently no good soluble biomarkers for atherosclerosis and/or endothelial dysfunction. Inflammatory markers such as high-sensitivity C-reactive protein are widely used, but they lack specificity for the vasculature. Moreover, the existing imaging techniques mainly capture the endstage of the disease, eg, the occurrence of plaques, but not its earliest stage such as endothelial dysfunction. If certain plasma miRNAs were uniquely modified by vascular injury, they may be capable of adding to the predictive value of conventional risk factors.

Merits and Limitations

Strengths of our study are its considerable size, representativity for the general community, high methodological standards, control for multiple testing, network analyses, rigorous replication using distinct techniques, various standards (miR-454, RNU6b), and different systems (plasma, cell culture) and species (human, obese mice). As limitations, the particulate fraction in plasma will contain other particles besides endothelial apoptotic bodies and the microarray used for the initial screening did not consider all miRNAs currently known. Accordingly, we cannot claim completeness for the miRNA profile among patients with DM and studies in large collectives of patients with DM and prediabetes are required to assess the potential of the reported miRNA signature and to establish miRNA–drug interactions.

Conclusion

This study provides the first evidence that plasma miRNAs, including endothelial miR-126, are deregulated in patients with DM, which may ultimately lead to novel biomarkers for risk estimation and classification39 and could be exploited for miRNA-based therapeutic interventions of vascular complications associated with this disease.

Non-standard Abbreviations and Acronyms

CI

confidence interval

CLR

context likelihood of relatedness

DM

type 2 diabetes

HUVEC

human umbilical vein endothelial cell

Lepob

Leptin obese

miRNA

microRNA

PCA

principal component analysis

PCC

Pearson correlation coefficient

qPCR

quantitative PCR

VEGF

vascular endothelial growth factor

Sources of Funding

This work was funded by the British Heart Foundation. M.M. is a Senior Research Fellow of the British Heart Foundation.

Disclosures

A.Z., S.K., I.D., J.W., and M.M. filed a patent application, owned by King's College London, that details claims related to described circulating miRNAs as biomarkers for DM.

Footnotes

In June 2010, the average time from submission to first decision for all original research papers submitted to Circulation Research was 14.5 days.

*Both authors contributed equally to this work.

Correspondence to Dr Manuel Mayr,
King's British Heart Foundation Centre, King's College London, 125 Coldharbour Lane, London SE5 9NU, United Kingdom
. E-mail .

References

  • 1. Bartel DP. MicroRNAs: target recognition and regulatory functions. Cell. 2009; 136:215–233.CrossrefMedlineGoogle Scholar
  • 2. Pillai RS, Bhattacharyya SN, Filipowicz W. Repression of protein synthesis by miRNAs: how many mechanisms?Trends Cell Biol. 2007; 17:118–126.CrossrefMedlineGoogle Scholar
  • 3. Kloosterman WP, Plasterk RH. The diverse functions of microRNAs in animal development and disease. Dev Cell. 2006; 11:441–450.CrossrefMedlineGoogle Scholar
  • 4. Latronico MV, Catalucci D, Condorelli G. Emerging role of microRNAs in cardiovascular biology. Circ Res. 2007; 101:1225–1236.LinkGoogle Scholar
  • 5. van Rooij E, Marshall WS, Olson EN. Toward microRNA-based therapeutics for heart disease: the sense in antisense. Circ Res. 2008; 103:919–928.LinkGoogle Scholar
  • 6. Suarez Y, Fernandez-Hernando C, Pober JS, Sessa WC. Dicer dependent microRNAs regulate gene expression and functions in human endothelial cells. Circ Res. 2007; 100:1164–1173.LinkGoogle Scholar
  • 7. Kuehbacher A, Urbich C, Zeiher AM, Dimmeler S. Role of Dicer and Drosha for endothelial microRNA expression and angiogenesis. Circ Res. 2007; 101:59–68.LinkGoogle Scholar
  • 8. Mitchell PS, Parkin RK, Kroh EM, Fritz BR, Wyman SK, Pogosova-Agadjanyan EL, Peterson A, Noteboom J, O'Briant KC, Allen A, Lin DW, Urban N, Drescher CW, Knudsen BS, Stirewalt DL, Gentleman R, Vessella RL, Nelson PS, Martin DB, Tewari M. Circulating microRNAs as stable blood-based markers for cancer detection. Proc Natl Acad Sci U S A. 2008; 105:10513–10518.CrossrefMedlineGoogle Scholar
  • 9. Tanaka M, Oikawa K, Takanashi M, Kudo M, Ohyashiki J, Ohyashiki K, Kuroda M. Down-regulation of miR-92 in human plasma is a novel marker for acute leukemia patients. PLoS One. 2009; 4:e5532.CrossrefMedlineGoogle Scholar
  • 10. Laterza OF, Lim L, Garrett-Engele PW, Vlasakova K, Muniappa N, Tanaka WK, Johnson JM, Sina JF, Fare TL, Sistare FD, Glaab WE. Plasma MicroRNAs as sensitive and specific biomarkers of tissue injury. Clin Chem. 2009; 55:1977–1983.CrossrefMedlineGoogle Scholar
  • 11. Wang K, Zhang S, Marzolf B, Troisch P, Brightman A, Hu Z, Hood LE, Galas DJ. Circulating microRNAs, potential biomarkers for drug-induced liver injury. Proc Natl Acad Sci U S A. 2009; 106:4402–4407.CrossrefMedlineGoogle Scholar
  • 12. Wang GK, Zhu JQ, Zhang JT, Li Q, Li Y, He J, Qin YW, Jing Q. Circulating microRNA: a novel potential biomarker for early diagnosis of acute myocardial infarction in humans. Eur Heart J. 2010; 31:659–666.CrossrefMedlineGoogle Scholar
  • 13. Ji X, Takahashi R, Hiura Y, Hirokawa G, Fukushima Y, Iwai N. Plasma miR-208 as a biomarker of myocardial injury. Clin Chem. 2009; 55:1944–1949.CrossrefMedlineGoogle Scholar
  • 14. Fichtlscherer S, De Rosa S, Fox H, Schwietz T, Fischer A, Liebetrau C, Weber M, Hamm CW, Roxe T, Muller-Ardogan M, Bonauer A, Zeiher AM, Dimmeler S. Circulating microRNAs in patients with coronary artery disease. Circ Res. 2010; 107.LinkGoogle Scholar
  • 15. Tijsen AJ, Creemers EE, Moerland PD, de Windt LJ, van der Wal AC, Kok WE, Pinto YM. MiR423–5p as a circulating biomarker for heart failure. Circ Res. 2010; 106:1035–1039.LinkGoogle Scholar
  • 16. Nathan DM. Long-term complications of diabetes mellitus. N Engl J Med. 1993; 328:1676–1685.CrossrefMedlineGoogle Scholar
  • 17. Frankel DS, Meigs JB, Massaro JM, Wilson PW, O'Donnell CJ, D'Agostino RB, Tofler GH. Von Willebrand factor, type 2 diabetes mellitus, and risk of cardiovascular disease: the Framingham offspring study. Circulation. 2008; 118:2533–2539.LinkGoogle Scholar
  • 18. Kiechl S, Lorenz E, Reindl M, Wiedermann CJ, Oberhollenzer F, Bonora E, Willeit J, Schwartz DA. Toll-like receptor 4 polymorphisms and atherogenesis. N Engl J Med. 2002; 347:185–192.CrossrefMedlineGoogle Scholar
  • 19. Bonora E, Kiechl S, Willeit J, Oberhollenzer F, Egger G, Meigs JB, Bonadonna RC, Muggeo M. Population-based incidence rates and risk factors for type 2 diabetes in white individuals: the Bruneck study. Diabetes. 2004; 53:1782–1789.CrossrefMedlineGoogle Scholar
  • 20. Kiechl S, Schett G, Schwaiger J, Seppi K, Eder P, Egger G, Santer P, Mayr A, Xu Q, Willeit J. Soluble receptor activator of nuclear factor-kappa B ligand and risk for cardiovascular disease. Circulation. 2007; 116:385–391.LinkGoogle Scholar
  • 21. Hosmer DW, Lemeshow S. Applied Logistic Regression. 2nd ed.New York: John Wiley & Sons; 2000.CrossrefGoogle Scholar
  • 22. Ramakers C, Ruijter JM, Deprez RH, Moorman AF. Assumption-free analysis of quantitative real-time polymerase chain reaction (PCR) data. Neurosci Lett. 2003; 339:62–66.CrossrefMedlineGoogle Scholar
  • 23. Roulston M. Significance testing of information theoretic functionals. Physica D. 1997; 10:62–66.CrossrefGoogle Scholar
  • 24. Slonim N, Atwal GS, Tkacik G, Bialek W. Information-based clustering. Proc Natl Acad Sci U S A. 2005; 102:18297–18302.CrossrefMedlineGoogle Scholar
  • 25. Faith JJ, Hayete B, Thaden JT, Mogno I, Wierzbowski J, Cottarel G, Kasif S, Collins JJ, Gardner TS. Large-scale mapping and validation of Escherichia coli transcriptional regulation from a compendium of expression profiles. PLoS Biol. 2007; 5:e8.CrossrefMedlineGoogle Scholar
  • 26. Bansal M, Belcastro V, Ambesi-Impiombato A, di Bernardo D. How to infer gene networks from expression profiles. Mol Syst Biol. 2007; 3:78.CrossrefMedlineGoogle Scholar
  • 27. Barabasi AL. Scale-free networks: a decade and beyond. Science. 2009; 325:412–413.CrossrefMedlineGoogle Scholar
  • 28. Chuang HY, Lee E, Liu YT, Lee D, Ideker T. Network-based classification of breast cancer metastasis. Mol Syst Biol. 2007; 3:140.CrossrefMedlineGoogle Scholar
  • 29. Tunica DG, Yin X, Sidibe A, Stegemann C, Nissum M, Zeng L, Brunet M, Mayr M. Proteomic analysis of the secretome of human umbilical vein endothelial cells using a combination of free-flow electrophoresis and nanoflow LC-MS/MS. Proteomics. 2009; 9:4991–4996.CrossrefMedlineGoogle Scholar
  • 30. Pula G, Mayr U, Evans C, Prokopi M, Vara DS, Yin X, Astroulakis Z, Xiao Q, Hill J, Xu Q, Mayr M. Proteomics identifies thymidine phosphorylase as a key regulator of the angiogenic potential of colony-forming units and endothelial progenitor cell cultures. Circ Res. 2009; 104:32–40.LinkGoogle Scholar
  • 31. Prokopi M, Pula G, Mayr U, Devue C, Gallagher J, Xiao Q, Boulanger CM, Westwood N, Urbich C, Willeit J, Steiner M, Breuss J, Xu Q, Kiechl S, Mayr M. Proteomic analysis reveals presence of platelet microparticles in endothelial progenitor cell cultures. Blood. 2009; 114:723–732.CrossrefMedlineGoogle Scholar
  • 32. Bandi N, Zbinden S, Gugger M, Arnold M, Kocher V, Hasan L, Kappeler A, Brunner T, Vassella E. miR-15a and miR-16 are implicated in cell cycle regulation in a Rb-dependent manner and are frequently deleted or down-regulated in non-small cell lung cancer. Cancer Res. 2009; 69:5553–5559.CrossrefMedlineGoogle Scholar
  • 33. Cimmino A, Calin GA, Fabbri M, Iorio MV, Ferracin M, Shimizu M, Wojcik SE, Aqeilan RI, Zupo S, Dono M, Rassenti L, Alder H, Volinia S, Liu CG, Kipps TJ, Negrini M, Croce CM. miR-15 and miR-16 induce apoptosis by targeting BCL2. Proc Natl Acad Sci U S A. 2005; 102:13944–13949.CrossrefMedlineGoogle Scholar
  • 34. Fish JE, Santoro MM, Morton SU, Yu S, Yeh RF, Wythe JD, Ivey KN, Bruneau BG, Stainier DY, Srivastava D. miR-126 regulates angiogenic signaling and vascular integrity. Dev Cell. 2008; 15:272–284.CrossrefMedlineGoogle Scholar
  • 35. Wang S, Aurora AB, Johnson BA, Qi X, McAnally J, Hill JA, Richardson JA, Bassel-Duby R, Olson EN. The endothelial-specific microRNA miR-126 governs vascular integrity and angiogenesis. Dev Cell. 2008; 15:261–271.CrossrefMedlineGoogle Scholar
  • 36. VanWijk MJ, VanBavel E, Sturk A, Nieuwland R. Microparticles in cardiovascular diseases. Cardiovasc Res. 2003; 59:277–287.CrossrefMedlineGoogle Scholar
  • 37. Zernecke A, Bidzhekov K, Noels H, Shagdarsuren E, Gan L, Denecke B, Hristov M, Köppel T, Jahantigh MN, Lutgens E, Wang S, Olson EN, Schober A, Weber C. Delivery of microRNA-126 by apoptotic bodies induces CXCL12-dependent vascular protection. Sci Signal. 2009; 2:ra81.CrossrefMedlineGoogle Scholar
  • 38. Waltenberger J, Lange J, Kranz A. Vascular endothelial growth factor-A-induced chemotaxis of monocytes is attenuated in patients with diabetes mellitus: a potential predictor for the individual capacity to develop collaterals. Circulation. 2000; 102:185–190.LinkGoogle Scholar
  • 39. Mayr M, Zhang J, Greene AS, Gutterman D, Perloff J, Ping P. Proteomics-based development of biomarkers in cardiovascular disease: mechanistic, clinical, and therapeutic insights. Mol Cell Proteomics. 2006; 5:1853–1864.CrossrefMedlineGoogle Scholar

Novelty and Significance

What Is Known?

  • MicroRNAs are small noncoding regulatory RNAs that alter protein expression.

  • Some microRNAs circulate in plasma vesicles and have been proposed as potential biomarkers for cancer, sepsis, myocardial infarction and heart failure.

What New Information Does This Article Contribute?

  • First evidence that plasma miRNAs are deregulated in patients with type 2 diabetes.

  • The distinct plasma miRNA pattern in diabetes includes loss of miR-126, a miRNA that is highly enriched in endothelial cells and facilitates VEGF signaling.

  • In a population-based study, miR-126 plasma levels negatively correlate with subclinical and manifest peripheral artery disease.

Our study is the first to reveal a plasma miRNA signature in a large population-based cohort of type 2 diabetics. We identified a set of circulating miRNAs that display altered expression in diabetes and characterized dynamic changes in miRNA coexpression networks. This may ultimately lead to novel biomarkers for risk estimation and classification and could be exploited for miRNA-based therapeutic interventions of vascular complications associated with this disease.

eLetters(0)

eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.