Circulating levels and characterization of microparticles in patients with different degrees of glucose tolerance

Background Microparticles (MPs) are vesicular structures shed from endothelial or circulating blood cells, after activation or apoptosis, and can be considered markers of vascular damage. We aimed to determine the levels of circulating MPs, their content of miRNA-126-3p and 5p, and their relationship with early endothelial activation/damage, in patients with different degree of glucose tolerance. Methods CD62E+, CD62P+, CD142+, CD45+ circulating MPs, their apoptotic (AnnexinV+) fractions, and miRNA-126 expression were determined in 39 prediabetic (PreDM), 68 type 2 diabetic (T2DM), and 53 control (NGT) subjects, along with main anthropometric and biochemical measurements. MPs were analysed by flow cytometry. miRNA-126 was measured by quantitative real-time PCR. Plasma antioxidant capacity was determined by electronic spin resonance; ICAM-1, and VCAM-1 by ELISA. Results Activated endothelial cell-derived MPs (CD62E+) were significantly increased in PreDM and T2DM in comparison to NGT (p < 0.0001). AnnexinV+/CD62E+ MPs and Annexin V+ MPs were significantly increased in T2DM compared to PreDM and NGT (p < 0.001); other MPs were not significantly different among groups. Plasma antioxidant capacity was significantly decreased in PreDM and T2DM compared to NGT (p = 0.001); VCAM-1 significantly increased in PreDM and T2DM in comparison to NGT (p = 0.001). miR-126-3p expression, but not miR-126-5p, in MPs, decreased significantly and progressively from NGT, to PreDM, and T2DM (p < 0.001). In PreDM and T2DM, CD62E+ MPs level was significantly and negatively associated with plasma glucose (p = 0.004). Conclusion We show for the first time that circulating CD62E+ MPs level and miR-126-3p content in MPs are abnormal in subjects with pre-diabetes; the content of miR-126-3p correlates with markers of endothelial inflammation, such as VCAM-1, plasma antioxidant capacity, and microparticles, well-accepted markers of endothelial dysfunction.

eliciting cellular responses: several studies have demonstrated the ability of MPs to promote vascular inflammation, and to interfere with coagulation pathway [6,7]. Diamant and colleagues reported that MPs number and origin were comparable between healthy subjects and well-controlled type 2 diabetic patients; however, the latter group had significantly higher tissue factorpositive MPs [8]. In the Framingham Heart Study, circulating MPs levels were associated with the presence of cardio-metabolic risk factors, particularly dyslipidemia. Intriguingly, in this study, MPs level was not significantly associated with overt diabetes, but the association between triglycerides and circulating MPs was more pronounced in diabetic patients compared to non-diabetic subjects [9].
Recently, it has been demonstrated that, in vitro, elevated glucose induces a marked release of MPs with a distinct molecular composition by endothelial cells with respect to normal glucose treatment [10]. However, the effect, in vivo, of different condition in carbohydrate metabolism on the release of MPs and on the endothelial activation or dysfunction has not been completely investigated.
MPs are one of the major transport vehicles of micro-RNA (miRNAs), and contribute to the transfer of micro-RNAs from the parent cell to various target cells by direct cell-to-cell interaction, modulating the expression of specific genes; miRNAs contained in MPs may in turn influence vascular homeostasis [11][12][13].
Endothelium-derived MPs promote vascular endothelial repair by delivering miR-126-3p into recipient cells; however, MPs obtained in hyperglycaemic conditions show reduced regenerative capacity both in vivo, and in vitro [14]. A reduced expression of miRNA-126 has been shown in susceptible individuals, and in patients with type 2 diabetes mellitus [15,16]; this suggests that the vascular protection of miRNA-126 is lost in the early phases of glucose intolerance.
The precursor miRNA pre-miR-126 gives rise to two mature strands, miR-126-3p and miR-126-5p, which are the most abundant miRNA in endothelial cells [17]. Deletion of gene encoding pre-miR-126 induces the loss of vascular integrity, and impairment of ischemic neovascularization [18]. The silencing of miR-126-3p induces endothelial inflammatory activation, in vitro [19]; furthermore, transfer of miR-126-3p by MPs released from apoptotic endothelial cells limits the atherosclerotic process, indicating an essential role of miRNA-126 in the endothelial stress response [20,21].
Glucose metabolism abnormalities are strongly associated with the development of endothelial dysfunction and atherosclerosis [22,23]. Alterations in carbohydrate metabolism form a continuum, which progressively worsens the cardiovascular health. Early hyperglycemia through a "metabolic memory" triggers endothelial activation as the initial vascular abnormality [24]. In this context, MPs release induced by early hyperglycemia may be considered novel biomarkers of endothelial activation/ dysfunction.
Several studies reported that miRNAs represent a very promising tool for diagnosis and prognostic purposes in type 2 diabetes. However, while circulating miRNAs are modulated by a wide range of factors, including sex, age, and drug treatments, miRNAs within MPs can represent a cleaner source of information, mainly due to their nature of communication between tissues, either contained in exosome [25], or extracellular vesicles [24]. Taken together, these findings suggest that circulating MP-packaged miRNAs, in addition to their role as biomarkers, could act as functional mediators in vascular and metabolic diseases.
Therefore, the aim of this study was to determine the level of circulating MPs, related to endothelial activation/ dysfunction and to assess the expression of miR-126-3p and miR-126-5p in MPs, across the spectrum of glucose intolerance, in non-diabetic, pre-diabetic and type 2 diabetic patients. We also investigated the relationships among MPs, their miRNA-126 content, metabolic parameters and early endothelial activation markers.
Anthropometric and blood pressure measurements were performed in all the subjects and a blood sample in fasting condition was obtained for the determination of glucose, lipid profile, ICAM-1, VCAM-1, plasma antioxidant capacity and MPs.
This study was carried out in accordance with the International Ethical Guidelines and the principles of the Declaration of Helsinki, and was approved by the local Institutional Review Board of the University of Padua Medical Centre. Patients filled out a complete lifestyle questionnaire regarding medical history, current therapy, smoking habits, and physical activity. All subjects signed the informed consent.

Metabolic parameters, adhesion molecule determinations and antioxidant capacity of plasma
Plasma glucose, insulin, C-peptide, and lipids were determined by enzymatic methods. Plasma levels of soluble adhesion molecules, Intercellular adhesion molecule-1 (ICAM-1), and Vascular cell adhesion molecule (VCAM-1) were measured as markers of early endothelial dysfunction by using a high-sensitivity ELISA assay, according to the manufacturer's instructions (BioVision, CA, USA). Results were expressed in ng/ml. The intra and inter-assay coefficient of variation were below 10%. All samples were coded for a blinded analysis, and each plasma sample was determined in duplicate.
The antioxidant capacity (AOC) of plasma was evaluated with Electron Spin Resonance (ESR) spectroscopy, by measuring the ESR intensity decay of the nitroxide probe 4-hydroxy-2,2,6,6-tetramethyl-piperidine-1-oxyl TEMPOL (Aldrich, USA), added to the plasma (0.25 mM). The samples were mixed with TEMPOL in quartz capillaries and rapidly inserted into the cavity of a Bruker ER200 ESR spectrometer at room temperature. ESR spectra of the nitroxide probe were recorded at 1 min interval for the first 15 min with the following experimental parameters: magnetic field sweep 60 G, microwave power 1 mW, modulation amplitude 1 G, scan time 40 s as previously described [28]. The initial ESR intensity decay rate is proportional to the antioxidant level. To convert the decay rate into an ascorbate equivalent concentration, we prepared a calibration curve with ascorbate-TEMPOL samples at known ascorbate concentration, to give antioxidant levels [29].

Microparticles assessment and characterization
Platelet-poor plasma (PPP) was prepared within 3 h of blood collection by double centrifugation (3000g for 15 min). 1 ml of PPP was centrifuged at 18,000g for 40 min at 4 °C to obtain microparticles (MPs). MPs were resuspended in 200 μl of phosphate-buffered saline (PBS, Sigma, USA) and stored at −80 °C until use. Samples, analyzed only after a single freeze-thaw cycle, were thawed by incubation for 5 min in a water bath at 37 °C immediately before assay.

miRNA-126 expression in microparticles
Microparticles from plasma of all the patients were isolated, as above described. Supernatants were removed, and the pellet composed of MPs was used for the determination of the miR-126-5p and miR-126-3p expression. RNA extraction from MPs was determined using RNeasy MinElute Cleanup kit (Qiagen, Hilden, Germany). cDNA was synthesized using specific miRNA primers (Applied Biosystems, USA) for miR-126-3p and miR-126-5p in TaqMan microRNA Reverse Transcription kit (Applied Biosystems,USA). Expression of miR-126-3p and miR-126-5p were assessed by quantitative real-time RT-PCR (qPCR), using a CFX96 PCR detection system (Biorad CFX 96). The RNU48 snRNA was used as an internal control and to normalize miR-126-5p and miR-126-3p expression (Applied Biosystems, USA) [32].

Statistical analysis
Continuous variables are expressed as mean ± standard error and categorical variables as percentage. Nonnormal variables at the Kolmogorov-Smirnov test were log-transformed before analysis. Comparisons between 2 or more groups were performed with the Student's t test or ANOVA, respectively for continuous data. The Bonferroni post hoc test was applied. To determine the association between MPs type and studied variables, univariate analyses were first run and then multiple linear regression analysis was run entering MPs as dependent variable, and explanatory covariates chosen among those showing significant (post hoc p < 0.05) associations in univariate group analysis. Statistical significance was accepted at p < 0.05, and SPSS ver. 22 was used.

Clinical characteristics and adhesion molecules
Main demographic and clinical characteristics of the study subjects are reported in Table 1. Subjects were divided into three categories according to their degree of glucose tolerance. T2DM and PreDM subjects showed significantly increased BMI, systolic blood pressure values, heart rate, and fasting plasma glucose levels in comparison to controls. PreDM subjects showed significantly lower plasma glucose levels and higher blood pressure values compared to T2DM. PreDM was associated with a worse insulin-resistant state, compared to NGT, as demonstrated by Si (8.2 ± 1.6 vs 16.5 ± 2.2 10 −4 dl/kg/ min µU/ml; p = 0.017) and HOMA index (4.0 ± 0.5 vs 1.2 ± 0.2; p < 0.001) measurements.

Microparticles levels
We determined the levels of circulating MPs obtained from activated endothelial cells (CD62E + ), tissue factor-bearing MPs (CD142 + ), leukocyte-derived MPs (CD45 + ), and activated platelet-derived (CD62P + ) MPs according to degree of glucose tolerance ( Table 2). The concentration of circulating CD62E + MPs was significantly increased in PreDM and in T2DM compared to NGT subjects. T2DM patients showed significantly higher values of AnnexinV + /CD62E + and AnnexinV + MPs compared to NGT and PreDM. No other significant differences were observed for all the other kinds of MPs investigated. Figure 1 shows a representative scatter plot from cytometer of circulating CD62E + MPs and the apoptotic fraction from NGT, PreDM and T2DM. PreDM and T2DM exhibited a larger population of MPs in quadrant I1 (CD62E + MPs) than NGT, as well as T2DM showed more positive MPs in quadrant I2 and I4 (respectively AnnexinV + /CD62E + and AnnexinV + MPs) compared to NGT and PreDM.

Plasma antioxidant level and early adhesion markers
The antioxidant level in plasma from all the subjects was assessed by the measurement of the ascorbate-equivalent concentration with ESR spectroscopy. As shown in Fig. 2a, a large decrease of the antioxidant level was found in plasma from PreDM, and T2DM in comparison to NGT (1.4 ± 0.1 and 1.5 ± 0.07 vs 2.3 ± 0.1; p = 0.001). Plasma antioxidant level was negatively correlated with CD62E + MPs (r = −0.31; p = 0.001) and with plasma glucose (r = −0.26; p = 0.004) (Fig. 2b, c).
Then, in all the subjects, we also measured the amount of plasma VCAM-1 and ICAM-1, as early endothelial activation markers. VCAM-1 levels showed a significant increase in PreDM and T2DM in comparison to NGT, while ICAM-1 level was significantly increased in T2DM in comparison to NGT and PreDM subjects (Fig. 2d, e).

miRNA-126 expression in microparticles
Since MPs are the major carriers of miRNAs in circulation, we analysed the expression of miRNA-126, i.e. of its two strands miR-126-3p and miR-126-5p in MPs isolated from the plasma of PreDM, T2DM and control subjects. The level of miR-126-3p was significantly higher in control compared to PreDM (p < 0.001 after Bonferroni correction) and T2DM (p < 0.001) subjects, while it was comparable between PreDM and T2DM patients (Fig. 3a). On the other hand, we did not observe any significant difference for miR-126-5p concentrations among all the three groups (Fig. 3b).
On the other hand, we did not find any significant correlation with miR-126-5p.

Discussion
The present study was designed to characterize and to define the role of several circulating levels of MPs involved in endothelial activation/dysfunction in subjects with different degrees of abnormal glucose metabolism. To this purpose, we determined: (a) CD62E + (E-selectin) MPs that are released by the activated endothelium [33,34]; (b) CD62P + (P-selectin) MPs that are mainly shed by platelets in response to cell activation, and are involved in inflammation and thrombosis [7,35,36]; (c) CD45 + MPs that derive from lymphocytes, and have been associated with the type of subclinical atherosclerotic lesions [37]; and (d) CD142 + (tissue factor) MPs which expression is induced on the surface of damaged or activated endothelial cells [38].

MPs pattern in different degrees of glucose tolerance
We demonstrated that only circulating CD62E + MPs levels were significantly increased in T2DM and in PreDM compared to NGT subjects, while the levels of AnnexinV + /CD62E + MPs and AnnexinV + MPs were higher in T2DM, but not in PreDM, suggesting only in these patients an increased apoptosis. This indicates not only that CD62E + MPs can be considered an early marker of endothelial dysfunction in hyperglycaemic patients, but also that an early activated endothelial state is already present in the prediabetic condition.
The association between CD62E + MPs and reduced antioxidant capacity and increased production of adhesion molecules, both established signs of endothelial dysfunction, strengthen their (pre)atherosclerotic role. The observed strong positive correlation between CD62E + MPs and plasma glucose confirms the relationship between hyperglycemia and endothelium dysfunction. Furthermore, multivariable analyses indicate that hyperglycemia is the primary metabolic defect associated with the release of CD62E + MPs as the correlation persisted after adjusting for BMI and blood pressure. Finally, we show that the content of miR-126-3p was markedly reduced in MPs from PreDM and T2DM in comparison to NGT, and negatively correlated with plasma glucose. These findings support the hypothesis of an important role of MPs as carriers of miRNAs, in the regulation of vascular health, which is altered by hyperglycemia, and in particular in diabetic state. To our knowledge, this is the first study reporting a detailed analysis of metabolic parameters in relation to circulating MPs levels, their miRNA content, and to markers of endothelial dysfunction, in a cohort of non-diabetic, pre-diabetic and diabetic patients.
The abnormal release of CD62E + MPs in PreDM could represent a marker of an altered endothelial phenotype that may progressively result to an increase of the apoptotic fraction of MPs levels, as observed in type 2 diabetic patients. Therefore, while pre-diabetes and type 2 diabetes can be seen as a metabolic continuum, the persistence of hyperglycemia (T2DM) appears to be associated with a different pattern of circulating MPs, characterized by the presence of an increased apoptosis, which may reflect a change from an endothelial activation to dysfunction. Our findings back those of Tramontano et al. who showed a marked increase of apoptotic fraction of circulating endothelial MPs from T2DM in comparison to NGT [39].

MPs and markers of endothelial inflammation
We found that circulating CD62E + MPs are inversely associated with the plasma antioxidant capacity, while there is a direct correlation with the level of VCAM-1; this suggests that the reduction of plasma antioxidant capacity may promote the release of VCAM-1 through circulating CD62E + MPs signalling. In our study, VCAM-1 levels showed a significant increase in PreDM, and T2DM, although the significance of VCAM-1 levels in prediction of T2DM should be improved by the addition of other inflammatory markers, such as E-selectin [40]. Overall these findings support the hypothesis that the excess of CD62E + MPs release from endothelial cells is an early abnormality of vascular disease as it can be observed in prediabetic subjects who show already higher CD62E + MPs levels than controls.

miR-126-3p content in MPs from patients with different degrees of glucose tolerance
Several studies have described MPs as "cargos" of information in blood vessel wall under pathological situations such as hypertension, myocardial infarction, and metabolic syndrome [4]. In this context, we observe that the expression of miR-126-3p in MPs reduces progressively from PreDM to T2DM, while miR-126-5p content is similar to control subjects. miR-126-3p levels exert protective vascular effects promoting vascular endothelial growth factor (VEGF) signalling [41], and through other protective mechanisms [15]. Our results therefore provide new evidence for the existence of a link between MPs release and miRNA-126 content in hyperglycaemic conditions. Moreover, we found for the first time an inverse correlation between miR-126-3p content in MPs and fasting glucose level, indicating a strict glycemia-sensitivity of this biomarker. In accordance with our clinical data, Jansen et al. demonstrated that MP-bound miR-126 was significantly reduced in diabetic patients, and endothelial cell-derived MPs were shown to be the major source of miR-126-containing MPs. They also verified that miR-126 content in endothelial cell-derived MPs was significantly reduced in high glucose condition, in vitro, strengthening our findings [42]. Although we can not provide any direct evidence of MPs in vascular repair capacity, the negative associations between miR-126-3p and CD62E + MPs, and between miR-126-3p and plasma VCAM-1 levels suggest a possible role of endothelial MPs as modulators of endothelial activation/dysfunction, through their miR-126-3p content. Moreover, we have also found a positive correlation between miR-126-3p and the plasma antioxidant capacity, confirming the presence of an association between a pro-oxidant and inflammatory imbalance and abnormal circulating CD62E + MPs.
Mechanistically, we still do not know whether an increase of CD62E + MPs from endothelial cells causes endothelial dysfunction per se or whether it represents simply a marker of endothelial activation/dysfunction in patients with altered glucose metabolism. However, several studies suggest a causal effect of hyperglycemia in altering the physical composition and molecular assessment of endothelial MPs, increasing their capacity for interaction with the vasculature, for the induction of coagulation, and oxidative stress, thus impairing vascular reactivity [10,43,44].

Limitations
This study has limitations: its cross-sectional nature does not allow to draw definite conclusions on MPs evolution at the single-patient level. The sample size is relatively small, but analyses of the levels of MPs are quite robust and reach statistical significance. The loss of miR-126 expression in plasma of diabetic patients has been extensively reproduced, therefore, we did not perform the same analysis in total plasma, to confirm this datum in our samples. On the other hand, a comparative study could strengthen our hypothesis, showing that this trend is mainly attributable to microparticles. Moreover, the sorting of different pattern of subpopulations of MPs may be useful in future research to understand the different epigenetic information present in MPs, particularly miR-NAs. Notwithstanding these limitations, our data may have pathophysiological implications: they demonstrate that MPs expression of endothelial activation may represent a reliable tool for early detection of vascular dysfunction in prevention screening.

Conclusion
We show for the first time that circulating CD62E + MPs level and miR-126-3p content in MPs are abnormal in subjects with different degrees of glucose tolerance; miR-126-3p correlates with markers of endothelial inflammation, such as VCAM-1, plasma antioxidant capacity, and microparticles, well-accepted markers of endothelial dysfunction.
Authors' contributions AG, AA, SVK, and GC participated in study conception and design. AG, CMR, and LF performed the acquisition of data. AG, SVK and GC participated in analysis and interpretation of data. AG, SVK and GC drafted the manuscript and CMR, EC, PS and AA helped in critical review of the manuscript. All authors read and approved the final manuscript. 1 Department of Medicine-DIMED, University of Padova, Padua, Italy. 2 Department of Chemical Sciences, University of Padova, Padua, Italy.