Upregulated miR-206 Aggravates Deep Vein Thrombosis by Regulating GJA1-Mediated Autophagy of Endothelial Progenitor Cells

Background Deep vein thrombosis (DVT) is the third most prevalent vascular disease worldwide. MicroRNAs (miRNAs) play regulatory roles in functions of endothelial progenitor cells (EPCs), which is becoming a promising therapeutic choice for thrombus resolution. Nevertheless, the role of miR-206 in EPCs is unclear. Methods EPCs were isolated from the peripheral blood of patients with DVT. In DVT mouse models, DVT was induced by stenosis of the inferior vena cava (IVC). The levels of miR-206 and gap junction protein alpha 1 (GJA1) in EPCs and vascular tissues of DVT mice were detected by reverse transcription-quantitative polymerase chain reaction (RT-qPCR). The proliferation, migration, apoptosis, and angiogenesis were tested by cell counting kit-8 (CCK-8) assay, Transwell assay, flow cytometry analysis, and in vitro tube formation assay. The levels of autophagy-related proteins as well as the level of GJA1 in EPCs and vascular tissues were evaluated by western blotting. DVT formation in vivo was observed through hematoxylin-eosin (HE) staining. The expression of thrombus resolution markers, CD34 molecule (CD34) and matrix metallopeptidase 2 (MMP2), in the thrombi was measured by immunofluorescence staining. Results miR-206 overexpression inhibited proliferation, migration, and angiogenesis and promoted apoptosis of EPCs, while miR-206 knockdown exerted an opposite effect on EPC phenotypes. Downregulation of GJA1, the target of miR-206, abolished the influence of miR-206 on EPC phenotypes. Furthermore, silencing of miR-206 suppressed the autophagy of EPCs via upregulating GJA1. miR-206 knockdown repressed thrombus formation, enhanced the homing ability of EPCs to the thrombosis site, and facilitated thrombus resolution in DVT mouse models. Additionally, miR-206 was upregulated while GJA1 was downregulated in vascular tissues of DVT mice. miR-206 knockdown elevated GJA1 expression in vascular tissues of DVT mice. The expression of miR-206 was negatively correlated with that of GJA1 in DVT mice. Conclusion miR-206 knockdown upregulates GJA1 to inhibit autophagy of EPCs and then promote EPC proliferation, migration, and angiogenesis, thereby enhancing EPC homing to thrombi and facilitating thrombus resolution.


Introduction
Deep vein thrombosis (DVT), with an annual incidence of about 10 million people, is the third most prevalent vascular disease worldwide [1]. DVT refers to the formation of blood clots in the deep venous lumen, which leads to blood flow disorders [2]. Clinical symptoms include limb swelling, superficial vein expansion, chronic pain, and distal venous hypertension [3]. At present, effective treatment options for DVT include surgical thrombectomy, pharmacologic thrombolysis, and anticoagulation [4]. Low-molecular-weight heparin (LMWH) is a drug with anticoagulant and antithrombotic effects, which has been widely used in DVT treatment [5]. However, these therapeutic methods also have great side effects, such as wound complications and risk of major hemorrhage [6]. Thus, it is essential to explore the molecular mechanism of DVT to improve its treatment.
Endothelial progenitor cells (EPCs) are vascular endothelial precursor cells, which have been demonstrated to show effective therapeutic applications in cardiovascularand vascular-related diseases [7]. EPCs have the capacity to proliferate, migrate to peripheral blood, and further differentiate into endothelial cells to participate in angiogenesis [8]. It has been reported that EPCs can be recruited to the thrombus to accelerate thrombus resolution, making EPCs emerge as a promising direction for DVT treatment [9]. However, the clinical applications of EPCs are faced with many challenges [10]. Thus, to explore the mechanism regulating EPC functions is crucial.
MicroRNAs (miRNAs) are small noncoding RNAs of 21-24 nucleotide long, which participate in posttranscriptional modulation of gene expression through suppressing RNA translation or facilitating mRNA degradation [11]. Numerous miRNAs including miR-143-3p, miR-195, and miR-130a-3p play pivotal roles in modulating the functions of EPCs such as cell proliferation, migration, angiogenesis, and autophagy and further influencing DVT recanalization and resolution [12][13][14]. miR-206 was reported to be implicated in the pathogenesis of many cardiovascular disorders, such as atherosclerosis, pulmonary arterial hypertension, and coronary artery disease (CAD) [15][16][17][18]. In a previous study, miR-206 was found to be markedly upregulated in peripheral blood EPCs from patients with CAD, and miR-206 downregulated phosphatidylinositol-4-phosphate 3kinase catalytic subunit type 2 alpha (PIK3C2α) expression to repress the migratory and vasculogenic abilities of EPCs and therefore result in the pathophysiology of CAD [19]. However, whether miR-206 adjusts the functions of EPCs from patients with DVT and thrombus resolution is unclear. Furthermore, it has been widely reported that miR-206 exerts its biological effects in various pathological processes in human diseases via targeting gap junction protein alpha 1 (GJA1), usually known as connexin 43 (Cx43) [20][21][22]. Gap junctions (GJs) mediate the exchange of metabolites, ions, and secondary messenger molecules between adjacent cells, being a significant contributor to direct cell-to-cell communication [23]. GJs are composed of connexins, and Cx43 has been identified as a major GJ protein in multiple cells, which is crucial for regulating stem cell differentiation [24]. Blocking Cx43 expression has been reported to suppress GJ-mediated signal transfer between neighboring cells, thereby impairing EPC proliferation, migration, and angiogenesis [25]. Therefore, our study investigated whether miR-206 affects EPC proliferation, migration, and angiogenesis in the pathogenesis of DVT via targeting GJA1.
Here, we intended to explore the role and associated molecular mechanism of miR-206 in regulating EPC functions and thrombus resolution. We hypothesized that miR-206 might target GJA1 to mediate the autophagy, proliferation, migration, and angiogenesis of EPCs, thereby regulating thrombus resolution. Our study might provide a new therapeutic target for DVT.

Materials and Methods
2.1. Cell Isolation and Culture. Ten patients aging 18-60 years old, who were diagnosed with first idiopathic DVT of the lower limbs at Nanjing First Hospital, Nanjing Medical University (Jiangsu, China), were included in this study. None of the patients had received anticoagulant drugs or undergone surgery prior to blood collection. DVT were identified by color Doppler ultrasound and lower extremity angiography. Patients with a history of diabetes mellitus, hypertension, and other chronic diseases were excluded. Peripheral blood was collected (80 ml each subject) from ten patients for isolating EPCs. The written informed consents were obtained from all participants. The study was approved by the Ethics Committee of Nanjing First Hospital, Nanjing Medical University (Jiangsu, China).
Human EPCs were extracted as previously described [26]. In brief, mononuclear cells were extracted from the peripheral blood of DVT patients by density-gradient centrifugation. Then, the cells were plated on fibronectin (Corning, NY, USA)-coated T-75 cell culture flask (Thermo Fisher Scientific, Waltham, MA, USA), followed by incubation with endothelial growth medium-2 (Lonza, Walkersville, Maryland, USA) containing 10% fetal bovine serum (FBS; Gibco, Carlsbad, CA, USA). After 4 days of incubation, nonadherent cells were removed by washing with phosphate-buffered saline (PBS) and the medium was changed. Early EPCs developed a spindle-shaped morphology after 7 days of culture. Colonies of endothelial-like cells were grown to 80% confluence, harvested with 0.25% trypsin, and passaged at a ratio of 1 : 2. The third or fourth passages of EPCs were used for the subsequent experiments.

DVT Mouse Model Establishment.
Eight-week-old male C57BL/6J mice were purchased from Beijing Vital River Laboratory Animal Technology Co. Ltd. Animal experiment was approved by the Institutional Animal Care and Use Committee of Nanjing First Hospital, Nanjing Medical University (Jiangsu, China).
The inferior vena cava (IVC) stenosis-induced DVT mouse model was established as described previously [27]. Briefly, mice were anesthetized with isoflurane-oxygen mixture. An incision was made along the midline of the abdomen, and intestines were exteriorized and soaked in warm saline. The IVC was gently separated from the aorta, and all IVC side branches were ligated. Subsequently, a 7.0 polypropylene suture was placed over the IVC and ligated over a spacer (30-gauge needle), and the spacer was then removed. This procedure led to about 90% closure of the vessel lumen without endothelial denudation. Finally, the peritoneum was closed with silk suture and skin with staples. After 24 h, thrombus formation in the IVC was checked.    3 Cardiovascular Therapeutics 2.11. Tube Formation Assay. The angiogenic capacity of EPCs was detected through in vitro tube formation assay. Approximately 40 μl of Matrigel (R&D Systems, MN, USA) was added into each well of 96-well plates, followed by 30 min of culture for solidification. Then, 100 μl of EPCs was inoculated into the coated plates. After culturing at 37°C for 8 h, capillary-like structures could be observed under a microscope.
2.14. Hematoxylin and Eosin (HE) Staining. After the mice were sacrificed, sections of the specimens 2 mm below the IVC ligation were removed and fixed with 4% paraformaldehyde for histological analysis. The fixed specimens were dehydrated using gradient alcohol, cleared up with xylene, and embedded in paraffin. The embedded tissue blocks were sliced at 4 μm intervals. These sections were later stained with hematoxylin for 15 min, washed with water, counterstained with eosin for 5 min, dehydrated with alcohol, hyalinized, and mounted with neutral resins. Finally, an inverted microscope (Olympus, Tokyo, Japan) was used to capture all histological images.
2.15. Statistical Analysis. Experimental data from at least three independent experiments were analyzed using SPSS 20.0 (SPSS, Chicago, USA) and are presented as the mean ± standard deviation. Student's t-test or one-way analysis of variance (ANOVA) followed by the Tukey post hoc test was employed to compare differences between groups or among groups. Gene correlation was analyzed by Pearson correlation analysis. In all analyses, p < 0:05 was considered significant.

miR-206
Affects EPC Proliferation, Apoptosis, Migration, and Angiogenesis. To investigate the influence of miR-206 on the phenotypes of EPCs, we first transfected EPCs with miR-206 inhibitor or mimics. PCR analysis indicated that miR-206 expression in EPCs was reduced after miR-206 downregulation but was elevated after miR-206 overexpression (Figure 1(a)). Then, a series of loss-of-function or gain-offunction experiments were conducted. As shown by the CCK-8 assay, miR-206 downregulation facilitated while miR-206 overexpression inhibited the EPC proliferation (Figure 1(b)). The results of flow cytometry analysis demonstrated that the percentage of apoptotic cells in the miR-206 inhibitor group was decreased, while the apoptotic cell population in the miR-206 mimic group was increased (Figure 1(c)). Transwell migration assay was conducted to evaluate EPC migration, which demonstrated that the number of EPCs passing through the membrane into the lower chamber was increased after miR-206 knockdown, while this number was significantly reduced after overexpressing miR-206 (Figure 1(d)). Additionally, to investigate whether miR-206 affects the angiogenesis of EPCs, a tube formation assay was carried out. Compared with the miR-NC group, EPCs transfected with miR-206 inhibitor displayed increased capillary-like structures and tube number, while EPCs transfected with miR-206 mimics showed decreased capillary-like structures and tube number (Figure 1(e)). This suggested that miR-206 exerted a negative effect on EPC angiogenesis. In conclusion, miR-206 knockdown promotes the proliferation, migration, and angiogenesis as well as inhibits the apoptosis of EPCs, while miR-206 overexpression exerts the opposite effects.
3.2. miR-206 Directly Targets GJA1. Since the influence of miR-206 on EPC phenotypes was confirmed, we subsequently investigated the underlying mechanism by which miR-206 regulates the behavior of EPCs. Based on the miRDB database (http://mirdb.org/), the target genes of miR-206 were predicted and the top ten genes were selected for PCR and western blotting analysis (Figure 2(a)). We    7 Cardiovascular Therapeutics position 478-485 of GJA1 3 ′ UTR, and the binding site is highly conserved among multiple species (Figure 2(e)). To further validate whether miR-206 binds to the GJA1 3 ′ UTR, a luciferase reporter assay was conducted. Wild-type or mutant 3 ′ UTR of GJA1 containing the miR-206 binding site were cloned into pmiR-RB-REPORT luciferase reporter vector for recombinant vectors pmiR-RB-REPORT-GJA1-3 ′ UTR-Wt/Mut (Figure 2(f  )). miR-206 knockdown remarkably enhanced the luciferase activity of pmiR-RB-REPORT-GJA1-3 ′ UTR-Wt, while there was no significant change in the luciferase activity of pmiR-RB-REPORT-GJA1-3 ′ UTR-Mut (Figure 2(g)). Overall, GJA1 is a target of miR-206.   GJA1 silencing attenuates the influence of miR-206 downregulation on EPC proliferation, apoptosis, migration, and angiogenesis.
To further explore whether miR-206 regulates the biological behaviors of EPCs by targeting GJA1, GJA1 was knocked down using sh-GJA1. GJA1 protein level was reduced in EPCs after GJA1 silencing (Figure 3(a)). Subsequently, EPCs were cotransfected with sh-GJA1 and miR-206 inhibitor. We discovered that compared with the miR-NC group, the miR-206 inhibitor group displayed enhanced cell proliferation, migration, and angiogenesis and decreased cell apoptosis (Figures 3(b)-3(e)). However, compared to the miR-206 inhibitor group, the miR-206 inhibitor+sh-GJA1 group showed decreased cell proliferation, migration, and angiogenesis and enhanced cell apoptosis (Figures 3(b)-3(e)). Therefore, cotransfection of miR-206 inhibitor and sh-GJA1 in EPCs effectively antagonized the effects of miR-206 inhibitor alone on EPC behaviors, which suggests that miR-206 regulates EPC behaviors by targeting GJA1.

miR-206
Modulates the Autophagy of EPCs via Targeting GJA1. Autophagy is an important metabolic way, and existing studies have demonstrated that autophagy can regulate numerous cell biological activities, including cell growth, apoptosis, migration, and angiogenesis [29]. Since we have confirmed the influence of miR-206 on EPC biological behaviors, then we investigated whether miR-206 regulates autophagy of EPCs by detecting levels of autophagy-related proteins. Western blotting analysis revealed that miR-206 knockdown significantly decreased the protein levels of ATG5 and ATG7 and the ratio of LC3B II/I as well as increased the protein level of p62, while miR-206 overexpression showed the opposite results (Figure 4(a)). These findings indicated that miR-206 knockdown repressed while miR-206 overexpression promoted the autophagy of EPCs. However, we also discovered that GJA1 silencing reversed the influence of miR-206 downregulation on the level of these proteins, which suggested that GJA1 silencing decreased the inhibition of miR-206 downregulation on the EPC autophagy. In summary, miR-206 modulates the autophagy of EPCs via targeting GJA1.

miR-206 Knockdown Represses Thrombus Formation in
Mice with DVT. Since the influence of miR-206 on EPC biological behaviors in vitro has been confirmed, then we next established the DVT mouse model to detect whether miR-206 affects the thrombosis in IVC in vivo. Through HE staining, we discovered that there was no thrombosis in IVC of mice in the Sham group. Compared with the Sham group, mice in the DVT group displayed obvious thrombosis in IVC 24 h after IVC ligation ( Figure 5(a)). However, after injection with antagomir-206, DVT mice showed alleviated blood clots in IVC (Figure 5(a)). In addition, both thrombus length and weight were significantly increased in the DVT group versus the Sham group but were markedly decreased in the DVT+antagomir-206 group versus the DVT+antagomir-NC group (Figures 5(b) and 5(c)). This suggested that miR-206 knockdown reduced the length and weight of the

miR-206 Knockdown Enhances EPC Homing Ability and
Facilitates Thrombus Recanalization and Resolution in Mice with DVT. It was reported that EPCs can be recruited to the thrombosis site and contribute to neovascularization and thrombus resolution [30]. Then, we explored whether miR-206 knockdown affects EPC homing ability to the thrombosis site in IVC of DVT mouse models and participates in thrombus recanalization and resolution. EPCs transfected with NC inhibitor or miR-206 inhibitor were injected into DVT mouse models. After 7 days, the thrombi with vessel walls were harvested. The thrombus size in the GFP-miR-206 inhibitor-EPC group was smaller than that in the GFP-NC inhibitor-EPCs group (Figure 6(a)). The thrombus weight was also significantly reduced after injection with GFP-miR-206 inhibitor-EPCs ( Figure 6(b)). Furthermore, we discovered that the number of EPCs in the thrombosis site in the GFP-miR-206 inhibitor-EPC group was more than that in the GFP-NC inhibitor-EPCs group (Figures 6(c) and 6(d)). Additionally, immunofluorescence staining was performed to evaluate the expression of CD34, the marker of angiogenesis in the thrombus, and MMP2, a crucial modulator of early thrombus resolution, in the thrombi. The results showed that CD34 and MMP2 displayed higher expression in the thrombi of the GFP-miR-206 inhibitor-EPC group versus the GFP-NC inhibitor-EPC group (Figure 6(e)).

miR-206 Negatively Regulates GJA1 Expression In Vivo.
Finally, we investigated the expression of miR-206 and GJA1 in vascular tissues of DVT mice. PCR analysis revealed that miR-206 was upregulated while GJA1 was downregulated in vascular tissues of mice in the DVT group. The results of PCR in the DVT+NC group showed no significant difference from those in the DVT group (Figures 7(a) and 7(b)). However, miR-206 expression was reduced, while GJA1 expression was elevated after downregulating miR-206 (Figures 7(a) and 7(b)). Furthermore, the protein level of GJA1 in vascular tissues was also reduced in the DVT group versus the Sham group but was increased in the DVT+antagomir-206 group (Figure 7(d)). Additionally, Pearson correlation analysis demonstrated that miR-206 expression was negatively correlated to GJA1 expression in vascular tissues of DVT mice (Figure 7(d)). In summary, miR-206 negatively regulates GJA1 expression in vivo.

Discussion
DVT is one common venous thromboembolic disorder and will result in deadly pulmonary embolisms if undetected and untreated timely [1]. The existing treatments might also bring unsatisfactory side effects despite their effective therapeutic outcomes [1]. Increasing studies have demonstrated the significant role of EPCs in thrombus resolution and recanalization [4]. miRNAs exert significant effects on the angiogenic capacity of EPCs, which subsequently mediate thrombus recanalization [31]. In this study, we discovered that miR-206 downregulation promotes EPC autophagy, proliferation, migration, and angiogenesis and thereby enhances thrombus resolution by targeting GJA1.

Cardiovascular Therapeutics
Accumulating studies have shown that miRNAs regulate diverse biological processes, including cell differentiation, apoptosis, survival, and proliferation [32]. miR-206 was previously reported to regulate cellular processes of EPCs, whose angiogenic ability is closely related to the pathology of cardiovascular diseases [33]. For example, miR-206 represses the viability and migration and enhances the apoptosis of EPCs from patients with CAD through inhibiting vascular endothelial growth factor (VEGF) expression, which suggests that miR-206 can be a novel biomarker for CAD [16]. Furthermore, miR-206 inhibits EPC migration, tube formation, and angiogenesis of patients with CAD [19]. In this study, we similarly discovered that miR-206 repressed the growth, migration, and angiogenesis of EPCs from patients with DVT. GJA1, also known as Cx43, is a significant gap junction protein that acts as a communication channel linking cardiomyocytes [34]. Cx43 is closely correlated with the enhanced migration of several cell types such as glioma cells, or neural crest cells, HeLa cells, and EPCs [35]. By far, Cx43 is the dominant gap junction protein in the EPCs [25]. It was reported that blocking GJA1 expression impairs EPC proliferation, migration, and angiogenesis [35]. In a previous study, VEGF induces EPC differentiation via Cx43-induced GJs and then contributes to vascular repair [36]. In hind limb ischemia mice, GJA1 downregulation reduces VEGF expression and attenuates the proliferation and migration activity as well as the angiogenic potential of EPCs [25]. In this study, we discovered that GJA1 silencing reversed the influence of miR-206 silencing on EPC biological behaviors. GJA1 knockdown promoted the apoptosis while inhibiting the growth, migration, and angiogenesis of EPCs. Therefore, miR-206 contributes to EPC dysfunction by downregulating GJA1.
GJA1 was reported to maintain cell junction structure and cell homeostasis via modulating autophagy [37]. Autophagy is a catabolic process for the degradation and recycling of cytosolic components by lysosomes [38]. Autophagy not only participates in cellular development and homeostasis but also plays a dual role in various physiological processes [39]. Previous studies have demonstrated that autophagy could either negatively or positively regulate cell proliferation, migration, and angiogenesis [40]. For example, miR-143-3p targets autophagy-related 2B (ATG2B) to suppress autophagy and thereby promotes EPC viability, migration, and tube formation [14]. miR-195 13 Cardiovascular Therapeutics inhibits autophagy and thereby attenuates EPC proliferation, migration, and angiogenesis via targeting GABA type A receptor-associated protein-like 1 (GABARAPL1) [12]. In this study, miR-206 knockdown markedly reduced the expression of ATG5 and ATG7 and the ratio of LC3B II/I as well as increased the expression of the ratio of p62, while miR-206 overexpression brought the opposite results. Overall, miR-206 knockdown inhibited while miR-206 overexpression promoted the autophagy of EPCs. However, GJA1 downregulation reversed the effects of miR-206 knockdown on autophagy. Therefore, our study indicated that miR-206 targeted GJA1 to affect autophagy, thereby modulating EPC proliferation, apoptosis, migration, and angiogenesis.
Since the enhanced proliferation, migration, and angiogenesis of EPCs contributes to thrombus resolution, which alleviates the development of DVT, then DVT mouse models were established for investigating the influence of miR-206 on DVT formation and resolution in vivo, respectively. We discovered that miR-206 downregulation not only inhibited thrombus formation but also enhanced EPC homing to the thrombosis site and promoted thrombus resolution. This result is in consistent with that in the previous studies [26,41,42]. Additionally, upregulated miR-206 and downregulated GJA1 along with a negative correlation between their expression were discovered in vascular tissues of DVT mice, which also demonstrated that miR-206 inhibited GJA1 expression in vivo.

Conclusion
In summary, this study elucidated that miR-206 knockdown targets GJA1 to inhibit EPC autophagy and thereby promote EPC proliferation, migration, and angiogenesis as well as suppress EPC apoptosis, which eventually enhances thrombus resolution and alleviates DVT development. This research might provide a promising target for DVT treatment.

Data Availability
The datasets used during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest
The authors declare that there exists no conflict of interest.