Ulmus parvifolia Modulates Platelet Functions and Inhibits Thrombus Formation by Regulating Integrin αIIbβ3 and cAMP Signaling

Background The prevalence of cardiovascular diseases (CVDs) is increasing at a high rate, and the available treatment options, sometimes, have complications which necessitates the need to develop safer and efficacious approaches. Ethnomedicinal applications reportedly reduce CVD risk. Ulmus parvifolia Jacq. (Ulmaceae) commonly known as Chinese Elm or Lacebark Elm, is native to China, Japan, and Korea. It exhibits anti-inflammatory, antiviral, and anticancer properties, but its anti-platelet properties have not yet been elucidated. Purpose To investigate the pharmacological anti-platelet and anti-thrombotic effects of U. parvifolia bark extract. Study Design and Methods Human and rat washed platelets were prepared; light transmission aggregometry and scanning electron microscopy was performed to assess platelet aggregation and the change in platelet shape, respectively. Intracellular calcium mobilization, ATP release, and thromboxane-B2 production were also measured. Integrin αIIbβ3 activation was analyzed in terms of fibrinogen binding, fibronectin adhesion, and clot retraction. The expression of MAPK, Src, and PI3K/Akt pathway proteins was examined. Cyclic nucleotide signaling pathway was evaluated via cAMP elevation and VASP phosphorylation. Anti-thrombotic activity of the extract was evaluated in vivo using an arteriovenous shunt rat model, whereas its effect on hemostasis in mice was assessed via bleeding time assay. Results U. parvifolia extract significantly inhibited human and rat platelet aggregation in a dose-dependent manner along with inhibition of calcium mobilization, dense granule secretion, and TxB2 production. Integrin αIIbβ3 mediated inside-out and outside-in signaling events, as evidenced by the inhibition of fibrinogen binding, fibronectin adhesion, and clot retraction. The extract significantly reduced phosphorylation of Src, MAPK (ERK, JNK, and p38MAPK), and PI3K/Akt pathway proteins. Cyclic-AMP levels were elevated in U. parvifolia-treated platelets, while PKAαβγ and VASPser157 phosphorylation was enhanced. U. parvifolia reduced thrombus weight in rats and moderately increased bleeding time in mice. Conclusion U. parvifolia modulates platelet responses and inhibit thrombus formation by regulating integrin αIIbβ3 mediated inside-out and outside-in signaling events and cAMP signaling pathway.


INTRODUCTION
Cardiovascular diseases (CVDs) are considered a leading cause of death worldwide. In the United States, one in seven and one in nine deaths occurred owing to coronary heart disease and heart failure, respectively, in 2013 (Mozaffarian et al., 2016). World Health Organization stated that CVD accounted for 30% of all deaths in 2005, and in Europe, it remains the primary cause of 42% mortalities in men and 52% in women (Nichols et al., 2013). Following vascular injury, platelets play a crucial role in maintaining hemostasis and preventing blood loss via thrombus formation; however, pathophysiological hyper-activation of platelets is the major cause underlying thrombotic complications, which contribute toward the development of cardiovascular ailments including atherosclerosis, coronary heart disease, stroke, and heart attack (Andrews and Berndt, 2004).
Pharmacological platelet suppression effectively reduces thrombotic events, and many clinical drugs are available for treating and preventing CVD. However, side effects and complications (e.g., gastric bleeding in case of aspirin and sometimes, thrombocytopenia or aplastic anemia in case of clopidogrel) caused by those drugs may outweigh their benefits (Barrett et al., 2008), while a significant part of population is resistant to most commonly used anti-platelet agents i.e., aspirin and clopidogrel (Wang et al., 2006;Ferguson et al., 2008) which necessitate the development of alternative preventive and therapeutic approaches with no or minimal drug-associated complications. Besides anti-platelet drug-based treatment options for thrombotic disorders and CVD-related complications, there is increasing focus on the use of natural products and their bioactive natural compounds including ethnomedicinal applications for CVD treatment and prevention (Badimon et al., 2010;Kim, 2018); similarly, innumerable natural products, including traditional Mediterranean diet and medicinal plants, have also been proven effective with regard to their cardioprotective and anti-platelet effects in the primary and secondary prevention of CVD (Estruch et al., 2013;Rastogi et al., 2016;Irfan et al., 2018c;Shin et al., 2019).
The genus Ulmus includes several species which produce fine wood, medicinal products, and edible fruit. Several Ulmus species reportedly possess anti-oxidative (Jung et al., 2010) and anti-platelet (Yang et al., 2013) properties. Ulmus parvifolia Jacq., commonly known as the Chinese Elm or Lacebark Elm, is native to Korea, Japan, and China, reportedly possesses anti-inflammatory (Mina et al., 2016), anti-allergic (Kim et al., 2016), anti-viral, and anticancer (Hamed et al., 2015) properties. While investigating newer and safer efficacious ethnomedicinal products, we recognized the medicinal properties of U. parvifolia bark. Here, we aimed to evaluate anti-platelet and anti-thrombotic effects of U. parvifolia bark ethanol extract on human and rat platelets and explored its pharmacological composition.

Extraction and Identification of Active Compounds
Bark of U. parvifolia was obtained from National Institute of Horticultural and Herbal Science (Jeollabuk-do, Republic of Korea). U. parvifolia bark was grounded and extracted with 70% ethanol at 80°C for 3 h, filtered with Whatman ™ filter paper (GE Healthcare, PA, USA), condensed in a rotary evaporator (Rotavapor ® R-100; B.U.CHI Labortechnik, Switzerland), and lyophilized to obtain powdered extract. The powder was stored at −30°C for further use in experiments.

Quantitative Analysis of Identified Compounds
Two main peaks (1 and 2) for catechins were quantified using UV detector at 280 nm wavelength and were calculated using a standard curve obtained from an authentic catechin standards. Standard calibration curves were plotted over a concentration range of 0.0012.5-0.2 mg/ml for five different concentrations of catechin and catechin-7-O-b-D -apiofuranoside (r 2 > 0.998). The amount of compound was finally expressed as mg/g of ethanol extract.

Animals
Sprague-Dawley rats weighing 220-240 g and 7-week-old C57BL/6J male mice weighing 20-22 g were purchased from Orient Co. (Seoul, Republic of Korea); prior to the experiment, they were acclimatized for 1 week in a special animal room maintained at 23 ± 2°C and 50 ± 10% humidity under 12-h light/ dark cycles. Experiments were conducted according to IACUC guidelines, and experimental protocols were approved by the Ethics Committee of College of Veterinary Medicine, Kyungpook National University, Daegu, Korea (Permit number: 2017-0014).

Preparation of Washed Human and Rat Platelets
Human platelet-rich plasma (PRP) collected from healthy volunteers who provided informed consent was obtained from the Korean Red Cross Blood Center (KRBC, Changwon, Korea), and its experimental use was approved by KRBC and the Korea National Institute for Bioethics Policy Public Institutional Review Board (PIRB17-1019-03). Washed human and rat platelets were prepared as previously described (Irfan et al., 2018a).

Platelet Aggregation Assay and Scanning Electron Microscope (SEM) Analysis
To assess platelet aggregation, the standard procedure of lighttransmission aggregometry was performed using a Chrono-log aggregometer (Havertown, PA, USA), as previously described (Irfan et al., 2018b). Briefly, washed platelets were pre-incubated with various concentrations of either U. parvifolia extract or vehicle (DMSO) for 1 min in the presence of 1 mM calcium chloride (CaCl 2 ), followed by stimulation with various agonists (Collagen, ADP, or thrombin) for 5 min with continuous stirring at 37°C. DMSO concentration was maintained at <0.1%.
A field emission SEM was used to assess platelet shape change and aggregation by obtaining ultrastructure images as previously described (Irfan et al., 2018a).

Arteriovenous Shunt Model
The anti-thrombotic activity of U. parvifolia extract was assessed in a rat extracorporeal shunt model as previously described (Irfan et al., 2018c). Rats were orally administered with the vehicle, U. parvifolia extract, or ASA once daily for 3 days. Two hours after the last administration, rats were anesthetized and shunt was placed for 15 min after initiating extracorporeal circulation. Subsequently, blood flow was stopped and the thrombus formed was weighed.

In Vivo Bleeding Assay
Male mice were divided into three treatment groups (n = 5, each). They were intraperitoneally administered with saline, ASA, or U. parvifolia extract once daily for 3 days. One hour after the last administration, mice were anaesthetized, and tail bleeding assay was performed as previously described (Irfan et al., 2018b).

Statistical Analysis
Data were analyzed by one-way analysis of variance (ANOVA), followed by measurement of statistically significant differences using Dunnett's post-hoc test (SAS Institute Inc., Cary, NC, USA). All data are presented as the mean ± standard deviation (SD). A p value of ≤0.05 was considered statistically significant.
Detailed description on; chemicals and reagent sources, preparation of washed human and rat platelets, scanning electron microscope (SEM) analysis, ATP release assay, measurement of [Ca 2+ ] i mobilization, flow cytometry, fibronectin adhesion assay, clot retraction, measurement of cyclic-AMP, AV-shunt model, and in vivo bleeding assay has been included in Supplementary Material.

U. parvifolia Inhibits Agonist-Induced Human and Rat Platelet Aggregation
Initial screening confirmed that U. parvifolia was effective against human and rat platelet aggregation induced by several agonists, including collagen, ADP, AA, and thrombin. Figures 1A, B show that U. parvifolia significantly inhibited agonist-induced aggregation in a dose-dependent manner in rat platelets; similar inhibitory results were obtained in agonist-induced aggregation in human platelets ( Figure 1C). The extract also potently inhibited ADP-induced aggregation measured in rat PRP, in a dose-dependent manner ( Figure 1E). We examined the effect of U. parvifolia extract on platelet shape change and aggregation using SEM. We found a dose-dependent inhibition of collagen-induced platelet shape change and aggregation in U. parvifolia-treated rat platelets compared with that in vehicletreated platelets ( Figure 1D).

U. parvifolia Reduces [Ca 2+ ] i Mobilization, Dense Granule Secretion, and Thromboxane-B2 Production
We assessed the effect of U. parvifolia extract on collagenstimulated platelet intracellular calcium ion ([Ca 2+ ] i ) mobilization, ATP release, and thromboxane production; and found its significant and dose-dependent inhibitory activity against [Ca 2+ ] i mobilization while it also markedly reduced ATP and thromboxane-B2 release (Figure 2).

U. parvifolia Attenuates Integrin a IIb b 3 -Mediated Inside-Out and Outside-In Signaling
We examined whether various concentrations of U. parvifolia extract modulate collagen-stimulated platelet integrin signaling, and found that they significantly attenuated fibrinogen binding to integrin a IIb b 3 in a dose-dependent manner ( Figures 3A, B). We also examined inhibitory activity of U. parvifolia extract on fibronectin adhesion and found that it significantly and dosedependently reduced platelet adhesion on fibronectin-coated surface compared with vehicle control ( Figure 3C). We also found that U. parvifolia extract significantly reduced clot retraction in a dose-dependent manner ( Figures 3D-F).

U. parvifolia Attenuates MAPK, Src, and PI3K/Akt Phosphorylation
We explored the underlying inhibitory mechanism of the extract, and our results revealed that U. parvifolia extract significantly inhibited collagen-mediated GPVI downstream signaling by attenuating the phosphorylation of ERK, JNK, p38 MAPK , Src, and PI3K/Akt pathway molecules (Figure 4).

U. parvifolia Elevates cAMP Levels and Enhances VASP and PKAabg Phosphorylation
We assessed intracellular cAMP levels in collagen-stimulated platelets which were markedly amplified owing to treatment with ascending concentrations of U. parvifolia extract ( Figure 5A). Similarly, phosphorylation of not only VASP but also phospho-VASP ser157 (a preferred site by PKA) was significantly enhanced . Immunoblotting was conducted to analyze the phosphorylation of signaling molecules extracted from the lysates of collagen-stimulated washed platelets that were pre-treated with either U. parvifolia extract or vehicle. Representative immunoblot images and data (mean ± SD) from at least four independent experiments are shown. *p < 0.05, **p < 0.01, and ***p < 0.001 versus agonist-treated group.

A B
FIGURE 5 | U. parvifolia elevates cAMP production and enhances VASP and PKAabg phosphorylation. (A) cAMP immunoassay performed for collagen-stimulated washed platelets pre-treated with various U. parvifolia extract concentrations, vehicle, or forskolin. (B) Immunoblotting analysis conducted for the phosphorylation of total VASP, p-VASP ser157 , and PKAabg extracted from platelets. Representative immunoblot images and graphs plotted for the data (mean ± SD values) obtained from at least four independent experiments are shown. *p < 0.05 and ***p < 0.001 versus agonist-treated group.
in a dose-dependent manner in U. parvifolia-treated platelets. Furthermore, to evaluate whether the observed effect is mediated via cAMP-PKA-VASP ser157 -dependent pathway, we assessed PKAabg phosphorylation; it was also found to be significantly increased in a dose-dependent manner ( Figure 5B). Figure 6A shows that U. parvifolia extract significantly reduced thrombus weight in a dose-dependent manner compared with vehicle control. Similarly, our results of tail bleeding assay in mice also revealed that U. parvifolia extract moderately increased bleeding time compared to vehicle control while time to cessation of bleeding was greatly increased in ASA group ( Figure 6B).

DISCUSSION
Here, we evaluated anti-platelet and anti-thrombotic effects of U. parvifolia bark ethanol extract and determined its potential pharmacological properties involved in the modulation of platelet functions which may be attributed to its phenolic constituents, especially catechins. Collagen, thrombin, and ADP are agonists that induce strong platelet aggregation by triggering downstream signaling events, including granule secretion, via activation of GPVI, proteinaseactivated receptor (PAR), and P2Y12 receptor signaling pathways, respectively. These platelet-signaling events are subdivided into (i) early receptor signaling triggered by agonists or adhesive stimulants; (ii) merging of common pathways and amplification of signaling; (iii) inside-out signaling which causes conformational change of integrin a IIb b 3 structure to allow fibrinogen binding and early phase of platelet adhesion; and (iv) outside-in signaling which augments late phases of adhesion and clot retraction. Moreover, platelets contain dense (d) granules, including Ca 2+ , ATP, and serotonin, as well as alpha (a) granules packed with adhesive ligands including fibrinogen, fibronectin, and P-selectin. Secretion of these granules upon platelet activation further enhances platelet adhesion, shape change, and aggregation ( Figure 8) (Estevez and Du, 2017;Irfan et al., 2020). The unceasing change in shape, which results in full activation of platelets, can be best observed under SEM. Our initial screening results revealed that U. parvifolia extract had strong inhibitory effects against the above mentioned agonists, and it significantly inhibited platelet aggregation and shape change in a dose-dependent manner; these findings indicated that the extract exerts strong antiplatelet activity via inhibition of several platelet-signaling pathways, as detailed in Figure 8. To examine the effects on physiological condition of platelets, we used rat PRP and found that the extract potently inhibited ADP-induced platelet aggregation; indicating that constituents of extract have strong ability to modulate platelet responses and aggregation. COX-1 inhibitor like aspirin and antagonists of the P2Y12 for ADP like clopidogrel and ticagrelor are currently used in coronary interventions and prevention of cardiovascular thrombotic events which are sometimes, impeded by side effects or resistant to some patients (Cattaneo, 2011). Our results show that U. parvifolia extract with several pharmacological constituents could be a potent natural candidate to deal with platelet-related disorders.   To investigate the underlying mechanism, we further examined intracellular [Ca 2+ ] i mobilization, dense granule secretion, and thromboxane production. We found that our extract markedly inhibited [Ca 2+ ] i mobilization, ATP secretion, and thromboxane-B2 production. Similarly, significant inhibition of fibrinogen binding to integrin a IIb b 3 and fibronectin adhesion was observed, indicating that U. parvifolia extract modulates integrin a IIb b 3 -mediated insideout signaling.
When fibrinogen binds to a IIb b 3 , it transduces signal into the cell (termed as outside-in signaling), which further enhances platelet adhesion and spreading, thrombus formation, and clot retraction (Irfan et al., 2018a). Similarly, fibronectin is another adhesive ligand that stabilizes thrombus formation in the vasculature. It binds to integrin a IIb b 3 and augments platelet aggregation by developing cohesive aggregates (Pankov and Yamada, 2002). GTPases regulate platelet adhesion, shape change, and clot retraction through actin cytoskeletal changes. Rho kinases (ROCKs) are downstream regulators which mediate RhoA-stimulated actin cytoskeletal changes via myosin light chain phosphorylation. The role of Rho kinase in facilitating clot retraction has been previously described using Rho kinase or ROCK inhibitor (Y-27632) (Liao et al., 2007). Among SFKs, Src kinase is predominantly expressed in platelets and plays a vital role in integrin a IIb b 3 -mediated signaling and is reportedly also involved in clot retraction (Senis et al., 2014). U. parvifolia extract significantly inhibited clot retraction and reduced Src phosphorylation. This result suggested that U. parvifolia extract modulated integrin a IIb b 3 -mediated outside-in signaling by inhibiting Rho kinase and Src kinase and consequently inhibiting platelet activation. Antagonists of glycoprotein IIb/ IIIa (a IIb b 3 ) such as abciximab and eptifibatide are used to reduce occlusive arterial events in patients with atherosclerosis (Cattaneo, 2011). Our results show that the extract and its constituents could be a great natural alternate to be considered in clinical interventions.
MAPKs (ERK, JNK, and p38 MAPK ) are highly expressed in platelets and become activated when platelets encounter any potential agonists; phosphorylation of these proteins triggers granule secretion which further enhances platelet aggregation (Adam et al., 2008). Src family kinases (SFKs) participate in early signaling of platelet activation and play a central role in mediating platelet functions. Moreover, MAPK and PI3K/Akt are downstream effectors of SFK and play critical roles in platelet activation by influencing calcium mobilization, granule secretion, and platelet aggregation (Senis et al., 2014). Here, our U. parvifolia extract impeded these molecules and significantly reduced the phosphorylation of MAPK and PI3K/ Akt, indicating its possible inhibitory mechanism against platelet functions.
Cyclic-AMP and cyclic nucleotide-dependent protein kinase are suppressed in activated platelets, whereas elevated intracellular cAMP levels inhibit platelet activation and aggregation (Li et al., 2003). Actin dynamics are relegated by vasodilator-stimulated phosphoprotein (VASP) which is a substrate of cyclic nucleotide (cAMP/cGMP)-dependent protein kinases (i.e., PKA/PKG); stimulation of these kinases augment VASP phosphorylation and inhibit platelet activation and aggregation. Elevation of cAMP levels reportedly inhibits platelet activation by activation of VASP via specifically increased phosphorylation of VASP ser157 (Ok et al., 2012). VASP ser157 acts as a substrate for cAMP-dependent PKA; its stimulation inhibits platelet activation via modulation of platelet secretion and adhesion, and its phosphorylation also hinders integrin a IIb b 3 -mediated signaling, thereby inhibiting platelet aggregation (Wentworth et al., 2006). Here, U. parvifolia extract inhibited platelet activation by upregulating cAMP-PKA-VASP ser157 pathway. Figure 8 summarizes the effects of U. parvifolia on platelet intracellular signaling. AV-shunt model is widely used for assessing anti-thrombotic activity (Endale et al., 2012), and well established to assess in vivo anti-thrombotic effects. Umetsu and Sanai (Umetsu and Sanai, 1978) have stated that the shunt comprises activated platelets, fibrin, and trapped erythrocytes and that its formation can be attenuated by anti-platelet agents. A similar thrombus formation occurs in coronary arteries following heart attack or myocardial infarction (Silvain et al., 2011). Our results revealed that as opposed to treatment with vehicle, treatment with U. parvifolia moderately increased bleeding time in mice and a substantially reduced thrombus formation in rats via inhibition of platelet activation.
UPLC is a powerful tool to identify and characterize the chemical profiles of natural products (Liu et al., 2013). Catechins are polyphenolic phytochemicals exhibiting several biological activities in the human body, potentially in the treatment and prevention of cardiovascular ailments through modulation of blood lipid metabolism, cardioprotective effects including vascular endothelium protection and stabilizing blood pressure (Shaterzadeh-Yazdi et al., 2017), while another study has also summarized the beneficial effects of catechins on cardiovascular system (Braicu et al., 2013). Most of the Ulmus species are reported to possess pharmacological properties including antiplatelet effects; due to their polyphenolic contents such as catechin, epicatechin, catechin-7-O-b-D -apiofuranoside, and catechin-7-O-b-D -xylopyranoside (Jung et al., 2010;Yang et al., 2013;Park et al., 2020). Among these catechins, catechin-7-O-b-D -apiofuranoside has been also known for its anti-oxidant, antiinflammatory and anti-fibrotic properties (MinHee et al., 2003;Jung et al., 2010;Kwon et al., 2011;Park et al., 2020). Previous studies have shown that catechins inhibit platelet activity by regulating calcium mobilization (Kang et al., 2001), impeding several signaling kinases including ERK and p38 MAPK (Lill et al., 2003), increasing cAMP levels, and enhancing VASP ser157 phosphorylation in platelets (Ok et al., 2012); they also inhibit ERK, JNK, p38 MAPK , and Akt activation in vascular smooth muscles and endothelial cells. Our UPLC results suggest that the extract predominantly contains catechins, which are most probably responsible for the inhibition of platelet activation and thrombus formation via regulation of MAPK pathway and cAMP signaling. CONCLUSION U. parvifolia substantially inhibits agonist-stimulated platelet aggregation, [Ca 2+ ] i mobilization, dense granule secretion, fibrinogen binding, and fibronectin adhesion as well as reduces clot retraction, which in turn inhibit integrin a IIb b 3 -mediated inside-out and outside-in signaling. It also attenuates the phosphorylation of Src, MAPK's, and PI3K/Akt-signaling m o l e c u l e s ; e l e v a t e s c y c l i c -A M P l e v e l s ; e n h a n c e s phosphorylation of VASP ser157 and PKAabg; and inhibits in vivo thrombus formation. Taken together, we conclude that U. parvifolia modulates platelet functions and inhibits thrombus formation by regulating integrin a IIb b 3 and cyclic nucleotide signaling; thus, it is a potential candidate for preventing and treating platelet-related CVDs.

DATA AVAILABILITY STATEMENT
All datasets generated for this study are included in the article/ Supplementary Material.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Korean Red Cross Blood Center and the Korea National Institute for Bioethics Policy Public Institutional Review Board (PIRB17-1019-03). Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements. The animal study was reviewed and approved by Ethics