Morroniside Regulates Endothelial Cell Function via the EphrinB Signaling Pathway after Oxygen-Glucose Deprivation In Vitro

Proangiogenic treatment is a potential treatment for acute myocardial infarction (AMI). Morroniside was previously discovered to increase post-AMI angiogenesis in rats as well as the proliferation of rat coronary artery endothelial cells (RCAECs). However, the effects of morroniside on other endothelial cell (EC) functions and underlying mechanisms are unknown. To further clarify the vascular biological activity of morroniside, this work focused on investigating how morroniside influenced endothelial cell functions, such as cell viability, tube formation capacity, migration, and adhesion, and to explore the signaling pathway. Oxygen-glucose deprivation causes ischemic damage in RCAECs (OGD). In vitro investigations were carried out to explore the involvement of morroniside in EC function and pathways mediated by ephrinB. The results revealed that the number of BrdU+ cells and cell viability in the high-dose group were considerably greater than in the OGD group (P < 0.05). The ability of tube formation evaluated by total tube length, tube-like structural junction, and tube area was significantly higher in the morroniside group than in the OGD group (P < 0.001). Morroniside considerably improved migration and adhesion abilities compared to OGD group (P < 0.05, P < 0.01, P < 0.001). The protein expression levels of the ephrinB reverse signaling pathway were substantially greater in the morroniside group than in the OGD group (P < 0.05, P < 0.01). In conclusion, the current study demonstrated that morroniside modulates endothelial cell function via ephrinB reverse signaling pathways and provided a novel insight and therapeutic strategy into vascular biology.


Introduction
Cardiovascular disease (CVD) has been identifed as the major factor resulting in mortality worldwide [1]. Ischemic cardiomyopathy, as a frequently-occurring myopathy of the heart, is primarily caused by inadequate supplies of nutrients and oxygen, resulting in an infarcted myocardium. Despite notable breakthroughs in revascularization attempts, such as percutaneous catheter intervention and surgical revascularization. Terapeutic angiogenesis has gained extensive academic attention [2]. Improving neovascularization is a therapeutic strategy for rescuing tissues from critical ischemia. Angiogenesis, the process through which endothelial cells (EC) sprout in the existing vasculature, followed by migration, proliferation, and tube creation, is critical in microvascular development and revascularization following myocardial ischemia (MI) [3]. Understanding angiogenesis is critical for developing novel therapeutics for MI injuries.
Tere are several routes involved in the pathological process of angiogenesis. Increasing evidence has shown that ephrinB reverse signaling participates in the regulation of the angiogenic process. Upon activation, the intracellular domains of ephrinB ligands can be phosphorylated via Src family kinases (SFKs), allowing them to bind to adaptor proteins such as Nck2 and activate downstream signaling molecules [4]. Many of the signaling pathways initiated by ephrinB signaling in endothelial cells have been identifed and linked to various endothelial cell functions [5].
Morroniside is one of the most abundant iridoid glycosides extracted from Cornus ofcinalis. Morroniside has previously been shown in vivo to promote endothelial progenitor cell proliferation, increase vessel density, and improve cardiac function after acute myocardial infarction (AMI). Besides, morroniside was also corroborated to stimulate the rat coronary artery endothelial cells (RCAECs) proliferation in vitro [6]. However, little is known about the morroniside's impact on other endothelial cell activities and the underlying mechanisms. In this work, we conducted a further study on cell viability, capillary tube formation, endothelial cell migration, adhesion, and related signaling pathways. Tese fndings provide an experimental foundation for better understanding the cardiac angiogenesisregulatory mechanisms of morroniside.

Oxygen Glucose Deprivation (OGD).
For the in vitro ischemia simulation, cells were subjected to hypoxic exposure (95% N 2 and 5% CO 2 ) and subsequently an 8 h incubation in glucose-free DMEM (Gibco, Termo Fisher Scientifc). Control cells, on the other hand, were incubated in normoxic (5% CO 2 and 95% air) context for 8 h. A group of the cells was accomplished under fve groups. Control cells were subjected to normal culture, while OGD cells were subjected to 8 h OGD treatment. For the three OGD + morroniside groups, the cells were pretreated with 1, 10, or 100 μM morroniside for 24 h and then subjected to OGD for 8 h.

Fluorescence Assay.
In immunofuorescence staining, the RCAECs were seeded into 48-well plates at a density of 20,000 cells/well. For 5-bromo-2-deoxyuridine (BrdU) assay, this work immobilized cells following an 8 h incubation using BrdU (10 μM, Sigma-Aldrich, St. Louis, MO, USA). Ten, the present work adopted immunocytochemical approach for quantifying the BrdU-positive cells. As a frst step, RCAECs were immobilized for 20 min in paraformaldehyde (4%), and then treated for 17 min using 2 N HCl at 37°C. After thrice PBS washing and a 1 h blockage at an ambient temperature using the donkey serum (5%, v/v, Jackson ImmunoResearch, Philadelphia, USA), an overnight incubation proceeded using the primary anti-BrdU mouse antibody (1 : 200; Roche, Indianapolis, USA) within 5% donkey serum involving PBS under 4°C. Cells were later incubated for a 2 h period using ALexa Fluor 594-labelled secondary antibody (1 : 400, Life Technologies, Carlsbad, USA) at an ambient temperature. Afterwards, the cellular nuclei were labelled using mounting medium with 4,6-diami-dino-2-phenylindole (DAPI) (Abcam, Cambridge, USA). To label F-actin, cells were fxed with 4% formaldehyde, followed by permeabilization using 0.1% Triton X-100. Fluorescein 555-labeled phalloidin (Abcam) was added to stain F-actin. Following the cellular nucleus labelling using mounting medium with DAPI (Abcam), a Ni-U fuorescence microscope (Nikon) was utilized for the cellular visualization.
2.6. Tube Formation Assay. 150 μl of Matrigel basement membrane matrix (growth factor enriched) (Becton, Dickinson and Company, Franklin Lakes, NJ, USA) was coated into each well of the 48-well plates in a sterile environment, and no air bubble was introduced, followed by 60-min incubation under ambient temperature for allowing the transformation of Matrigel into a gel. Subsequently, Matrigel was added with RCAECs (2 × 10 4 /well). At last, the plates were added with 300 μl serum-free medium for further incubation under 37°C and 5% CO 2 conditions. Each experiment was carried out twice to take the average for separate experiments. At 8 h postincubation, four felds were randomly selected from each well to take the images, then Image-Pro Plus software was utilized to determine tube length, branch points, and tube area.

Scratch Assay.
Scratch assay was conducted to assess RCAECs migration. Briefy, this work grew RCAECs (1 × 10 5 / well) within the 6-well plates to approximately 90% density, and each well was scratched with a sterile 200 μl pipette tip to make a straight wound approximately 1 mm wide in the middle of the cells after synchronizing the cells for 24 h. For determining cell migration, images were captured using an Olympus X71 fuorescence microscope at the indicated time. Te area of scratch was determined with ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Cell
Migration. Te 8 μm 24-well Transwell chambers (Corning, Lowell, MA, USA) were employed for evaluating cell migration. In brief, 5 × 10 4 cells were resuspended in serum-free medium before being put on top of each chamber insert. 10% FBS-containing DMEM was added into the bottom chamber to be the chemoattractant. Tereafter, one cotton swab was utilized to remove cells there were still on the upper membrane. After 24 h, methanol was used to fx the cells migrating via pores, followed by crystal violet staining.
2.9. Cell Adhesion. 3 μg/ml fbronectin was added onto the 96-well plates for overnight incubation under 4°C, followed by 2 h blocking using 1% BSA under 37°C. Later, cells (5 × 10 4 /well) were added into the 96-well plate involving serum-free bufer of the reagent under test (100 μl/well), and then subjected to a 30 min incubation under 37°C. Following mild plate rinsing using PBS at 100 μl per well, 100% ice-cold methanol (100 μl/well) was utilized to immobilize the adherent cells at 4°C for 5 min. After one wash with distilled water, the cells were stained for 15 min with crystal violet (50 μl/well; 0.5% (w/v) in 20% (v/v) methanol) at room temperature. Subsequently, the wells were washed several times with water and left to dry overnight on a bench. Crystal violet was extracted for 10 min at room temperature by adding 50 μl/well of 10% (v/v) acetic acid, and the absorbance was read at 562 nm. An equally-treated plate was utilized for determining background, albeit without cells. Cell adhesion in the absence of a reagent was considered 100% [8]. For visualization, DAPI (Sigma-Aldrich) was added for cell counterstaining, followed by observation with an Olympus X71 fuorescence microscope.

Statistical
Analysis. Data are expressed as mean-± standard deviation (SD) and were statistically analyzed using SPSS 20.0. A one-way ANOVA followed by Tukey's test was used for comparisons between multiple groups. P < 0.05 was considered statistically signifcant.

Morroniside Regulates Proliferation, Viability, and Tube
Formation of RCAECs. We previously demonstrated that morroniside increased the number of Ki67 + RCAECs in vitro [6]. Ki67 immunohistochemistry, on the other hand, can only yield information on the proliferative state of the cells because Ki67 is expressed in all phases of the cell cycle except for G0 [9]. To further estimate the rate of the cell proliferation, we labelled S-phase with BrdU. Te 100 μM morroniside group exhibited pronouncedly elevated BrdU + cell quantity in contrast to the OGD group (Figures 1(a) and  1(b)). Te viability of RCAECs following morroniside therapy was then evaluated. As indicated in Figures 1(c) and 1(d), morroniside dramatically promoted cell viability compared to the OGD group. Tube formation experiments are a key method for evaluating the ability to undergo angiogenesis in vitro. To further determine the proangiogenic efects of morroniside, this work measured total tube length, tube-like structural junction count, and tube area to evaluate the tube formation. As illustration in Figures 1(e)-1(h), compared to the OGD group, the tube formation signifcantly increased following morroniside at concentrations of 10 μM and 100 μM treatment, respectively. Tese results provide direct evidence that morroniside promoted proliferation, cell viability, and tube formation in RCAECs.

Morroniside Regulates Migration of RCAECs.
Te migration of RCAECs treated with morroniside in vitro was determined using a wound healing model. Morroniside signifcantly increased RCAEC migration (Figures 2(a) and 2(c)). We also performed a Transwell migration assay to study the migratory ability of RCAECs (Figures 2(b) and 2(d)). Tese results were consistent with the migration assay conducted using the wound healing model. Compared with the OGD group, Transwell migration was prominently enhanced in the 10 μM and 100 μM morroniside groups. Tese results indicated that morroniside promoted the migration of RCAECs.

Morroniside Regulates Adhesion of RCAECs.
Angiogenesis is a multistep process that includes cell-matrix adhesion, cell migration, and tube formation. As tube formation and cell migration depend on cell-extracellular matrix adhesiveness, we evaluated the efects of morroniside on endothelial cell adhesion to fbronectin, a major substrate for endothelial cells. Results show that treatment with morroniside increased the fbronectin-adherent RCAECs in a dosedependent manner (Figures 3(a) and 3(b)). Cell adhesion is the interaction of cells with the actin cytoskeleton that connects them to the extracellular matrix. To investigate whether morroniside induced F-actin alignment, fuorescently-labeled phalloidin was utilized to visualize F-actin cytoskeleton. As shown in Figure 3(c), the control group had the majority of the F-actin bundles. Oxygen-glucose deprivation model resulted Evidence-Based Complementary and Alternative Medicine in the disruption of the F-actin network, resulting in actin clumps within the cytoplasm. Cells treated with morroniside showed fewer F-actin clumps and more F-actin bundles than the OGD group. As a result, morroniside enhanced RCAECs adhesion and stabilized F-actin flaments.

Morroniside Regulates ephrinB Reverse Signaling
Pathway-Related Protein Levels. Compelling evidence indicates that the ephrinB reverse signaling pathway drives endothelial cellular processes, such as proliferation, adhesion, and migration after initiation [5,10]. In this work, the p-ephrinB level decreased remarkably following OGD  . Cellular viability analysis of RCAECs using a CCK-8 assay after the treatment with morroniside 24 h and 48 h (n � 6) (c, d). Typical tube formation images obtained at 8 h after cell seeding onto Matrigel (e). Te image J was utilized to quantify branch points, total tube length, and tube area (n � 5) (f-h). Results were represented by means ± SD. ### P < 0.001 in comparison with the control. * P < 0.05, * * P < 0.01, and * * * P < 0.001 relative to the OGD group. compared to the control group. Te p-ephrinB level was elevated pronouncedly by treating with morroniside ( Figure 4(a)). Te phosphorylation of ephrinB ligands enables the binding of the adaptor protein Nck2 and activates the downstream proteins of focal adhesion kinase (FAK), VE-cadherin, and integrinα5 [4]. Next, the Nck2, p-FAK, VE-cadherin, and integrin α5 protein expressions were assessed. Resembling the expression outcome for p-ephrinB, the control group had decreased expressions of Nck2, p-FAK, VE-cadherin, and integrinα5 compared with the sham group. Morroniside administration signifcantly enhanced the levels of these proteins (Figures 4(b)-4(e)). As implied by the foregoing fndings, the ephrinB reverse signaling probably serves as a mediator for morroniside's role in the endothelial cell function.

Discussion
Re-establishment of the blood supply to the myocardium after AMI is partly dependent on angiogenesis. Neovascularization is a critical element in the ischemic myocardium repair [11,12]. In our prior work, morroniside has been corroborated to promote endothelial progenitor cell proliferation, increase vessel density, and improve cardiac function following AMI in rats [6]. Te focus of this work were morroniside's efects on the endothelial cell function in vitro and its potential mechanisms. Our fndings suggest that morroniside appears to facilitate the functions of proliferation, cell viability, tube formation, migration, and adhesion of RCAECs. Te ephrinB reverse signaling pathway may be involved in the processes underlying these efects.
ECs are responsible for forming capillary tubes, which is their distinguishing properties and also the essential condition to establish the blood fow-routing continuous vascular lumen. Tube formation can be the rapid and quantitative approach for determining in vitro angiogenesis ability [13]. Our results demonstrated that a capillary-like tubular structure grew more thickly after treatment with morroniside. Furthermore, the morroniside treatment also increased the amount of BrdU + cells and cell viability following OGD. Tese results support our previous fndings that morroniside enhances in-vivo angiogenesis [6]. Based on the general concept, ECs migration is an essential Evidence-Based Complementary and Alternative Medicine component of angiogenesis [14]. We explored how morroniside impacted the migration and discovered that the migration ability of RACECs was diminished after OGD injury and that the treatment with morroniside facilitates cell migration. Aside from the tube formation and migration, another crucial function of ECs is adhesion property. Te dynamic integrin-mediated adhesion of ECs to the surrounding cell-extracellular matrix is critical for angiogenesis in the pathological condition. A wealth of evidence indicates that angiogenic migration persistently depends on the endothelial cell adherence to the cell-extracellular matrix [15]. Ruan and colleagues demonstrated that the enhanced focal adhesion contributes to the angiogenesis after OGD injury [16]. Following that, we evaluate the efect of morroniside on RCAECs adhesion to fbronectin and found that RACECs treated with morroniside were more adhesive than those in the OGD group. Due to the concurrent research, fbronectin participates in the cell attachment to the substrate localized adjacent to actin flament bundles and their termini [17]. Communication between the actin flaments and focal adhesions is crucial for cell adhesion and migration. Our fndings revealed that F-actin was malaligned following OGD, while treatment with morroniside stabilized F-actin flaments. Our in vitro study, for the frst time, clearly demonstrated morroniside's role in the RCAEC viability, tube formation, adhesion, and migration. Reverse signaling of ephrinB has an important impact on regulating EC function. Inducing ephrinB reverse signaling by EphB receptor promotes angiogenesis [10]. In the current study, a marked p-ephrinB elevation was obvious after the treatment with morroniside. Park et al. revealed that ephrinB enhanced integrin-induced cells adhesion; moreover, integrins represented the transmembrane receptors activating signaling while causing cell invasion [18]. Te quantifcation of adherent cell number by 100% of the control group (n � 6) (b). Representative images of F-actin structures stained by fuorescently labeled phalloidin as well as nuclei stained by DAPI (c). Te arrow stands for F-actin bundles, whereas arrowhead stands for F-actin clumps. Results were represented by means ± SD. ### P < 0.001 relative to the control; * P < 0.05, * * P < 0.01, and * * * P < 0.001 relative to the OGD group.
Furthermore, activating integrins accelerate focal adhesion to connect with the actin cytoskeleton, which is required for driving cell migration [19]. Te ephrinB-Nck2 complex can activate focal adhesion kinase (FAK) and has an important impact on the integrin signal [20]. Depending on our fndings shown in Figure 4, the p-FAK, Nck2, and integrin α5 levels were prominently raised following the morroniside therapy. It is well known that VE-cadherin represents the critical adhesive molecule in endothelial adhesive junctions, and there are several processes that infuence the VE-cadherin-targeting endothelial junction stability [21]. Te morroniside group exhibited evidently elevated VEcadherin levels compared with the OGD group, which supported our research on the adhesion ability and F-actin flaments. As implied by these fndings, the ephrinB reverse signaling is a potential mediator for the endothelial cell function-regulatory mechanism of morroniside. Combining our prior study of the regulation of morroniside on the VEGF/VEGF receptor 2 signaling pathway to promote cell proliferation at the cellular level by culturing RCAECs [6], we hypothesize that ephrinB signaling pathway may be a novel important target of morroniside. Hence, further research is needed to fully understand the underlying mechanisms of morroniside's action and to clarify its biologically active components more in depth.

Conclusions
Collectively, we demonstrated that morroniside promoted endothelial cell functions of proliferation, tube formation, migration, and adhesion. Morroniside-induced increases in ephrinB reverse signaling components are thought to have mediated the function regulation mechanism. Based on our previous research, this study further clarifes the vascular biological activity of morroniside through regulating the endothelial cell function and revealed the prospective targets.

Data Availability
Te data utilized for supporting our results are included in this work.

Conflicts of Interest
Te authors declare that they have no conficts of interest.

Authors' Contributions
TTL contributed to data collection analysis and manuscript drafting. SYZ contributed to data collection and analysis. FLS was responsible for revising the manuscript. XT, ZXZ, WRZ, and YFW contributed to data collection.  Figure 4: Morroniside promotes the expression of ephrinB reverse pathway-related proteins. Typical WB images and p-ephrinB expression quantifed relative to GAPDH (n � 4) (a). Typical WB images and Nck-2 expression quantifed relative to GAPDH (n � 4) (b). Typical WB images and p-FAK expression quantifed relative to GAPDH (n � 4) (c). Typical WB images and VE-cadherin expression quantifed relative to GAPDH (n � 4) (d). Typical WB images and integrinα5 expression quantifed relative to GAPDH (n � 4) (e). Results were represented by means ± SD. # P < 0.05 and ## P < 0.01 in comparison with the control group. * P < 0.05 and * * P < 0.01 in comparison with the OGD group.
Evidence-Based Complementary and Alternative Medicine JGX and WW contributed to study design and manuscript revision.