Hawthorn berry ( crataegus songarica ) causes endothelium-dependent relaxation of the porcine coronary artery: Role of Estrogen receptors

Abstract

BACKGROUND:

Fruits of Crataegus songarica are commonly used for the treatment of vascular insufficiency and heart problems.

OBJECTIVE:

Our aim was to determine the effect of C. songarica on vascular tone and to determine the mechanisms underlying the vasorelaxant properties.

METHODS:

Extracts of C. songarica were tested for vasodilator activity of porcine coronary artery after pre-contraction with the thromboxane mimetic U46619 in the presence or absence of inhibitors of intracellular signaling cascades. Reactive oxygen species were assessed by dihydroethidine staining and the level of eNOS and AKT phosphorylation was measured by immunohistochemical staining.

RESULTS:

Extracts of C. songarica berries produced endothelium dependent vasorelaxation, with most significant effect induced by aqueous fraction (AS-CS). This vasorelaxant effect of AS-CS was reduced by inhibition of nitric oxide pathways and inhibition of potassium channels. Inhibition of phosphatidylinositol 3- kinase and Src tyrosine kinase, as well as scavenging of reactive oxygen species, produced an attenuation of the relaxation response. Estrogen receptor antagonists

(tamoxifen and ICI 182,782) reduced the AS-CS mediated vasorelaxation. AS-CS also stimulated the endothelial formation of ROS and phosphorylation of Akt and eNOS.

CONCLUSION:

The data indicated that C. songarica produces an endothelium-dependent vasorelaxation, which is partly dependent upon estrogen receptors, and sensitive to inhibition of ROS/Src/PI3K/NO pathways.

Keywords

Crataegus endothelium-dependent relaxation reactive oxygen species PI3-kinase/Akt coronary artery

1 Introduction

Crataegus songarica, locally known as Goni, is native to Pakistan and widely distributed in mountains of Chitral, Swat, Astor, Gilgit, Kashmir and Muree. Fruits of C. songarica are used commonly for different cardiovascular disorders like hypertension and heart failure [1 –3], hypertrophy and vascular insufficiency [4] and gastrointestinal conditions [5]. A previous research study from our laboratory has revealed that C. songarica produces diuretic and blood pressure lowering effects through activation of NO/cGMP pathway and renin-angiotensin in rats [6]. Berries of the Crataegus (hawthorn) species are known to have one of the highest content of phenolic constituents including procyanidins, anthocyanins C-glycosyl flavones and phenolic acids. Crataegus encompasses approximately 300 species, many of which have been reported to possess cardioprotective, hypolipidemic, vasorelaxant, antioxidant and anti-arrhythmic properties [7 –11]. As C. songarica is thought to have a beneficial role in cardiovascular diseases and as it contains an enriched polyphenolic profile, this study was designed to determine the effect of C. songarica on vascular tone.

Endothelial cells (ECs) play an important role in controlling blood flow, blood pressure, thrombosis, platelet aggregation and in generation and migration of new blood vessels [12]. Various antihypertensive drugs reduce blood pressure by directly acting on blood vessels and causing relaxation. The endothelium regulates the vascular tonicity and maintains a fine balance between constriction and relaxation of vessels by secreting both vasoconstrictors (ET-1 and Angiotensin II) and vasodilators (NO and prostacyclin) [13, 14]. The prerequisite for a physiologically proficient endothelium is to release NO which causes vasodilatation and inhibits the release of endothelin-1 which causes vasoconstriction along with inflammatory, oxidative and proliferative functions [15]. The balance between production of vasoconstrictors and vasodilators is disrupted during endothelial dysfunction, which makes the vessels more susceptible to atherosclerosis and thrombosis. Endothelial dysfunction is a major participant in a variety of diseases like diabetes, hypertension, atherosclerosis and dyslipidemia [14, 16]. The severity of endothelial dysfunction is thought to be direct marker of different kinds of cardiovascular events in hypertensive patients [17]. Therefore, the maintenance of endothelial function is a therapeutic target for treatment of cardiovascular disease [18].

In order to determine whether C. songarica has vasodilator properties, experiments were performed to determine the effect of extracts of C. songarica berries on the level of tone in porcine isolated coronary arteries and to characterize the relaxing factors involved. The data presented indicate that aqueous extracts of C. songarica produce an endothelium-dependent relaxation through activation of estrogen receptors and subsequent activation of a pathway involving PI3 kinase and NOS/NO activation.

2 Material and methods

2.1 Chemicals and drugs

Most chemicals used for this research work i.e n-hexane, methanol, dichloromethane, n-butanol, ethyl acetate, L-NAME, indomethacin, atropine, bradykinin, U46619, tetraethylammonium (TEA), 4-aminopyridine, glibenclamide, barium chloride, iberiotoxin, apamin, TRAM-34, PEG-catalase, diethylthiocarbamate (DETCA), N-acetyl cysteine and EGTA were purchased from Sigma-Aldrich, UK. The chemicals (NaCl, KCl, KH2PO4, MgSO4.7H2O, CaCl2, NaHCO3, and glucose) were also obtained Fisher Scientific, Loughborough, UK and were utilized to formulate Krebs Hensleit buffer solution. ODQ, tamoxifin, PP2, ly249002, L760735, ICI 182782, win64338 Hcl, were obtained from Tocris Bioscience (Bristol, UK). All the chemicals of standard analytical grade were utilized in this research.

2.2 Plant material used

The fruits of Crataegus songarica were obtained from Upper Boni (Chitral), Khyber Pakhton Khawa (Pakistan) during the months of september-october 2016, which were further identified by Dr. Amin-Ullah Shah (plant taxonomist, UOS). A voucher specimen (# W-6030) was submitted to the Herbarium, Department of Botany, University of Sargodha for further research reference.

2.3 Extraction and fractionation

Fruits (4 kg) were shade dried and their peal was isolated, which was pulverized to a coarse constituency for extraction. Refined powder was soaked in 3L water and methanol (30:70, v/v), followed by filtration with muslin cloth and filter papers. This process was repeated three times and the filtrate obtained was concentrated at 40 °C using a rotary evaporator and stored in a cool place (4–6 °C). The % yield of crude extract of C. songarica (CS-Cr) was 27%. For activity guided fractionation, 100 g crude extract of C. songarica (CS-Cr) was mixed with distilled water and solvent–solvent extraction was performed with different polarity based absolute solvents; hexane, dichloromethane, ethyl acetate and n-butanol. Obtained fractions were concentrated at 40 °C using a rotary evaporator. This process resulted in 13.8 g butanol fraction (BS-CS) and 86.17 g aqueous fraction (AS-CS), whereas, ethyl acetate, hexane and dichloromethane fractions were absent. All the samples were placed at 4°C in a refrigerator for further pharmacological and phytochemical studies [19].

2.4 Screening of vasorelaxant activity of extract/fractions of C. songarica

2.4.1 Preparation of rings of distal coronary arteries

Briefly, porcine hearts were acquired from local abattoir and transported to laboratory in boxes filled with ice-cold Krebs-Henseleit solution (NaCl 118, KCl 4.8, CaCl₂·H₂O 1.3, NaHCO₃ 25.0, KH₂PO4 1.2, MgSO₄·7H₂O, glucose 11.1 (in mM) previously gassed with a mixture of 95% O₂ and 5% CO₂) [pH 7.4]. Porcine coronary arteries (PCAs) were dissected out, placed in Krebs-Henseleit solution and left overnight at 4°C. Next morning, distal portion of coronary arteries were cleaned to remove adjacent fatty tissues and cut into 2 mm long rings with mean vessel dimension of 0.81±0.02 mm for further studies.

2.4.2 Wire myograph study

The PCA rings were attached under a basal tension of 24.5 mM in a multichannel wire myograph (DMT 610 M, Denmark) attached to a PowerLab data acquisition system (AD instruments, UK). Channels were filled with 5 ml Krebs solution, gassed with 5% CO₂ and 95% O₂ and maintained at 37°C. After 60 min of stabilization, PCA rings were contracted twice with 60 mM KCl, washed and then contracted with U46619 (1 nM –50μM) to get 80 % of the contraction produced with 60 mM KCl. The rings were tested for endothelium integrity with a single concentration of bradykinin (0.3μM). Tissues were then washed and left for a further 30 min stabilization period before rings were contracted again with U46619. Once the contraction had stabilised, extract/fractions (0.0001–10 mg/ml) of C. songarica were screened for their vasorelaxant activity in both endothelium intact and denuded arteries. Fractions were added cumulatively to create a concentration response curve. The endothelium of the coronary artery rings was removed by gentle rubbing of the inside the lumen of rings using forceps. Successful removal of the endothelium was confirmed by a lack of relaxation with bradykinin (0.3μM). The percentage relaxation produced by the extract/fractions was calculated as a percentage change in level of tone from the U46619-induced contraction [20, 21].

2.5 Determination of the mechanism underlying the endothelium dependent vasorelaxation of the aqueous soluble fraction of C. songarica (AS-CS)

Crude extract/fractions of C. songarica produced endothelium dependent vasorelaxation in coronary rings with maximum vasorelaxation effect induced by aqueous soluble fraction (AS-CS). Therefore, the following experiments were performed to understand the mechanism underlying the endothelium dependent relaxation response to aqueous soluble fraction.

2.5.1 Effect of L-NAME, ODQ and indomethacin on the relaxation response to AS-CS

The rings were incubated, with L-NAME (300μM, [20]) - NO synthase inhibitor, indomethacin (10μM, [20]) - the cyclooxygenase inhibitor or both for 45–60minutes before the contraction with U46619. In some experiments, rings were incubated with (ODQ) (10μM) to estimate the involvement of NO/cGMP pathway in the relaxation to AS-CS. Then, a concentration response curve was made with cumulative additions of AS-CS (0.0001 –5 mg/ml) in the absence and presence of inhibitors and percentage relaxation was measured.

2.5.2 Effect of KCl and K+ channel blockers on the relaxation response to AS-CS

A concentration response curve to AS-CS was constructed in coronary arteries pre-contracted with 20 mM KCl and compared with relaxation response in U46619 pre contracted rings. In other experiments, the rings were incubated for 45 minutes with tetraethylammonium (TEA; 10 mM, [20]) - non-selective K+ channel inhibitor, glibenclamide (1μM, [20]) - the K_ATP channel inhibitor or barium chloride (30μM, [20]) - an inwardly rectifying K+ channel (Kir) inhibitor, before the contraction with U46619.

Some rings were incubated with iberiotoxin (BK_Ca inhibitor, 100 nM), TRAM-34 (IK_Ca inhibitor, 10μM) and Apamin (SK_Ca inhibitor, 500 nM) respectively [20]. In addition, rings were also incubated with combination of apamin and Tram-34 or iberotoxin in the presence and absence of both indomethacin and L-NAME. Then, a concentration response curve was made with cumulative additions of AS-CS (0.0001 –5 mg/ml) in the absence and presence of inhibitors and percentage relaxation was measured.

2.5.3 Determination of the involvement of muscarinic, bradykinin, takykinin and estrogen receptors in the relaxation to AS-CS

Rings were incubated for 45 minutes with muscarinic receptor inhibitor, atropine (100 nM), the bradykinin receptor B2 inhibitor, Win 64338 (1μM; [22]), tachykinin NK1 receptor inhibitor, L760735 (100 nM, [23]), and non-selective estrogen receptor inhibitor, tamoxifen (10μM or 50μM) or a selective ER-α inhibitor, ICI 182, 782 (1μM). In some experiments, rings were incubated with tamoxifen and combination of PP2 and LY249002 to determine any co-relation between estrogen receptor and Src/PI3-kinase/Akt Pathways. Then, a concentration response curve was made with cumulative additions of AS-CS (0.0001 –5 mg/ml) in the absence and presence of inhibitors and percentage relaxation was measured. In other experiments, tissues were pre-contracted with either U46619 or 30 mM KCl before concentration response curves to 17β estradiol (1 nM to 30μM) were carried out.

2.5.4 Role of calcium in the relaxation response to AS-CS

Tissues were incubated in Krebs solution containing EGTA (2 mM) for 15 minutes prior to U46619 contraction to remove any extracellular calcium, whereas, in control rings tissues were incubated with normal krebs solution. Then concentration response curve was constructed with cumulative additions of AS-CS in both segments and percentage relaxation was compared.

Additionally, a set of experiments were run in which the rings were incubated with 5 mg/ml AC-CS in calcium free Krebs solution and exposed to 60 mM KCl. Then cumulative addition of calcium (1μM to 3 mM) was made to construct concentration response curve.

2.5.5 Effect of reactive oxygen species (ROS) in the relaxation response to AS-CS

Rings pre-incubated with SOD mimetic - MnTMPyP (100 uM), SOD inhibitor - diethylthiocarbamate (DETCA, 10 mM), PEG-catalase (300 Uml–1), which inhibits the decomposition of H₂O₂ and an antioxidant N-acetylcysteine (10 mM) [24], for 45 minutes before U46619 contraction. Then, a concentration response curve was made with cumulative additions of AS-CS (0.0001 –5 mg/ml) in the absence and presence of inhibitors and percentage relaxation was measured.

2.5.6 Involvement of Src/PI3-kinase/Akt pathway in relaxation to AS-CS

Rings were incubated with PP2 (10μM)-Src-kinase inhibitor and LY249002 (50μM)-PI3-kinase inhibitor (25), whereas, the control tissues were incubated with vehicle - DMSO (0.1% v/v). In some experiments, rings were pre incubated with combination of both PP2 and LY204002 with and without L-NAME (300μM) for 45–60 before U46619 contraction. After that, a concentration response curve was made with cumulative additions of AS-CS (0.0001 –5 mg/ml) in the absence and presence of inhibitors and percentage relaxation was measured.

2.5.7 Immunohistochemical determination of phosphorylated eNOS and AKT in response to AS-CS

In order to measure the phosphorylation of eNOS and AKT in response to AS-CS in coronary arteries, coronary artery rings were treated with 5 mg/ml AC-CS at 37°C for 30 minutes, then embedded and frozen in OCT. The frozen coronary arteries (at 14μm) were cryosectioned and air-dried for 10 minutes and placed at –80°C for further use. Then coronary rings were affixed with 4% paraformaldehyde and washed followed by 2 hr incubation with 10% milk in phosphate-buffered saline (PBS) at 25°C to block nonspecific binding. Then the rings were exposed to p-eNOS (1/100) and p-AKT (1/100) antibody overnight at 4°C. Next day, the rings were washed followed by 2 h incubation at room temperature in dark with secondary antibody (1/400). After 2 h rings were washed with PBS and mounted with fluorescence mounting medium and cover-slipped. Primary antibodies were omitted in negative control rings. Then fluorescence signals were measured with confocal laser-scanning microscope (20M lens) and quantified with Image J software [26].

2.5.8 In situ recognition of reactive oxygen species in response to AS-CS

Porcine coronary artery rings were exposed to 5 mg/ml AC-CS at 37°C for 30 minutes. Then the rings were embedded in OCT and frozen in a liquid nitrogen. Then 25μm thick sections of frozen rings were made and affixed on slides coated with polylysine. The rings were treated with fluorescent dye dihydroethidium (DHE) for 30 minutes at 37°C followed by treatment with AS-CS (5 mg/mL) for 10 minutes with and without PEG-catalase (500 U/mL), MnTMPyP (100 mM) and NAC (10 mM). DHE fluorescence was measured by confocal microscope with 20 magnification lens and quantified with Image J software (version 1.49p) [27].

2.6 Data analysis

Relaxation data were expressed as a percentage change from the level of pre-contraction and presented as means±standard error of the mean of n experiments, where n = the number of different animals used. Statistical analysis was carried out by using 2-way ANOVA followed by Bonferroni’s post-hoc test. P < 0.05 was considered as significant. EC50values were calculated by non-linear regression using GraphPad Prism software (GraphPad, San Diego, Ca, USA).

3 Results

3.1 Endothelium- dependent effect of crude extract of C. songarica in isolated porcine coronary arteries (PCAs)

The crude extract of C. songarica (CS-Cr) (0.0001 to 5 mg/ml) produced a concentration dependent vasorelaxation in U46619 contracted porcine coronary artery (PCA) rings with more effect in endothelium intact rings. Removal of the endothelium significantly reduced the vasorelaxant activity (Fig. 1). CS-Cr at 3 mg/ml produced 94±4.9% relaxation in endothelium intact rings as compared to only 74.3±5.9% relaxation in endothelium denuded rings at 5 mg/ml concentration (Fig. 1).

Fig. 1

Vasorelaxant effect of, (A) crude extract (CS-Cr), (B) butanol soluble fraction (BS-CS), C) aqueous soluble fraction (AS-CS) of C. songarica in U46619-precontracted endothelium denuded (E-) and intact (E+) porcine coronary rings. Data are expressed as a percentage change from the U46619-induced contraction and are means±SEM of six separate experiments. ^***= P < 0.001 v control, 2-way ANOVA, followed by a Bonferroni post-hoc test. D & E example traces showing responses to cumulative additions of aqueous soluble fraction (AS-CS) in endothelium intact (D) and endothelium denuded (E) segments of porcine coronary artery. Concentrations added are in mg/ml.

3.2 Comparative vasorelaxant effect of various fractions of C. songarica in isolated porcine coronary artery rings

Both the butanol (BS-CS) and the aqueous fraction (AS-CS) obtained from the crude extract were analyzed for their endothelium dependent and independent effects in PCA rings. The aqueous fraction of C. songarica (AS-CS) was found to be most effective in producing vasorelaxant effect (Fig. 1). AS-CS produced a concentration and endothelium dependent vasorelaxation. AS-CS at the concentration of 5 mg/ml produced 92.5±6.2% vasorelaxation in endothelium intact rings, whereas, it produced only 44.3±3.2% relaxation in endothelium denuded rings at the maximum concentration of 10 mg/ml. The butanol fraction (BS-CS) at the highest concentration, 10 mg/ml, produced 73.5±4.2% relaxation in endothelium intact rings, whereas, in endothelium denuded rings, at same concentration, it produced only 59.2±3.9% relaxation (Fig. 1).

The above results indicate that the aqueous soluble fraction of C. songarica (AS-CS) is responsible for the potent endothelium dependent effects. Therefore, the aqueous fraction was further investigated to determine the underlying endothelium dependent vasorelaxant mechanism.

3.3 AS-CS produces vasorelaxation via activation of the NO/cGMP pathway

Nitric oxide and prostacyclin have an essential role in endothelium dependent vasorelaxation. Pre-treatment with the nitric oxide synthase inhibitor L-NAME (300μM), significantly prevented the vasorelaxant response of AS-CS (Fig. 2A). However, pre-incubation with indomethacin (10μM), had no effect on AS-CS induced relaxation (Fig. 2B). Pre incubation with a combination of both L-NAME and indomethacin reduced AS-CS induced relaxation in a similar way to pre-incubation with L-NAME alone (Fig. 2C), indicating that nitric oxide mediates the endothelium-dependent relaxation response. Furthermore, pre-incubation with ODQ- a specific guanylyl cyclase inhibitor, significantly reduced the relaxation produced by AS-CS demonstrating the involvement of NO-GC pathway in the AS-CS induced vasorelaxation (Fig. 2D).

Fig. 2

Vasorelaxant effect of aqueous soluble fraction of C. songarica (AS-CS) in U46619-pre-contracted coronary rings in the absence and presence of, A) L-NAME (300μM), B) Indomethacin (10μM), C) combination of L-NAME and indomethacin and D) ODQ (10μM). Figure E shows the relaxation responses to AS-CS after pre-contraction with KCl in comparison with pre-contraction with U46619. Data are expressed as a percentage change from the level of pre-contraction and are means±SEM of six separate experiments. ^*= p < 0.05, and ^***= p < 0.001 v control, 2-way ANOVA, followed by a Bonferroni post-hoc test.

3.4 AS-CS induces vasorelaxation by activation of K⁺ channels

To determine the role of K+ channels in the vasorelaxant response of AS-CS, some experiments were performed in KCl pre-contracted PCA rings, in comparison with U46619 pre-constricted PCAs. KCl pre-constriction significantly reduced the AS-CS induced relaxation (Fig. 2E). There was only 25.4±6.7% relaxation in KCl pre-contracted rings at 5 mg/ml AS-CS compared to 92.4±2.6% relaxation in U46619-precontracted rings. Pre-treatment with the non-selective potassium channel blocker (TEA, 10 mM) also significantly decreased the AS-CS induced relaxation. On the other hand, K_ATP blocker glibenclamide (3μM), K_ir blocker barium chloride (1 mM), or the selective blockers of small conductance (SK_Ca) Apamin (500 nM), intermediate conductance (IK_Ca) TRAM-34 (10μM) or large-conductance (BKCa), iberiotoxin (100 nM), Ca^2 +-activated potassium channels did not alter the relaxation induced by AS-CS (Fig. 3).

Fig. 3

Vasorelaxant effect of aqueous soluble fraction of C. songarica (AS-CS) in U46619-pre-contracted coronary rings in the absence and presence of, A) tetraethylammonium (TEA, 10 mM), B) glibenclamide (1 uM), C) barium chloride (30μM)) D) Iberiotoxin (100 nm), E) Iberiotoxin in the presence of L-NAME and indomethacin and F) in the presence of L-NAME (300μM), indomethacin (10μM), apamin (500 nM) and/or TRAM-34 (10μM). Data are expressed as a percentage change from the U46619-induced contraction and are means±SEM of six separate experiments. ^*= p < 0.05 and ^***= p < 0.001 v control, 2-way ANOVA, followed by a Bonferroni post-hoc test.

3.5 AS-CS induces endothelium-dependent relaxations through Src and PI3-kinase/Akt pathways

Inhibition of Src tyrosine kinase with PP2 significantly reduced relaxations to AS-CS (Fig. 4A). The LY249002 (PI3-kinase inhibitor) also inhibited AS-CS induced relaxations significantly (Fig. 4B). Moreover, the combination of PP2 and LY294002 had the same effect as inhibition with L-NAME alone (Fig. 4C). Furthermore, the combination of PP2 with LY249002 had the same effect as L-NAME alone (Fig. 4C). Similarly, the combination of PP2, LY294002 and L-NAME also had the same effect as L-NAME alone (Fig. 4C).

Fig. 4

Vasorelaxant effect of aqueous soluble fraction of C. songarica (AS-CS) in U46619-pre-contracted coronary rings in the absence and presence of, A) LY249002, (50μM), B) PP2 (10μM), C) LY294002 + PP2 in the presence and absence of 300μM L-NAME. Data are expressed as a percentage change from the U46619-induced contraction and are means±SEM of six separate experiments. ^*= p < 0.05 and ^***= p < 0.001 v control, 2-way ANOVA, followed by a Bonferroni post-hoc test

3.6 AS-CS induces endothelium-dependent relaxations through estrogen receptors

The estrogen receptor antagonist tamoxifen (50μM) significantly reduced the vasorelaxant response to AS-CS (Fig. 5A). Similarly the selective ER-α antagonist, ICI 182, 782 also reduced the vasorelaxant response to AS-CS, indicating a role for estrogen receptors in the AS-CS induced vasorelaxant effect (Fig. 5C). PP2 and LY249002 had no inhibitory effect on the AS-CS relaxation in the presence of tamoxifen, indicating that they are part of the same pathway (Fig. 5B). Atropine, WIN64338 and L760735 did not significantly affect the vasorelaxant response to AS-CS, indicating the muscarinic receptors (Fig. 5D), bradykinin receptors (Fig. 5E) and tachykinin receptors, respectively, are not involved (Fig. 5F).

Fig. 5

Vasorelaxant effect of aqueous soluble fraction of C. songarica (AS-CS) in U46619-pre-contracted coronary rings in the absence and presence of, A) tamoxifen (10 or 50μM), B) tamoxifen (50μM) in the presence of 50μM LY294002 & 10μM PP2, C) ICI 182,782 (1μM), D) Atropine (100 nM), E) WIN 64338 (1μM) and F) L760735 (100 nM). Data are expressed as a percentage change from the U46619-induced contraction and are means±SEM of six separate experiments. ^*= p < 0.05 and ^***= p < 0.001 v control, 2-way ANOVA, followed by a Bonferroni post-hoc test.

3.7 AS-CS induces redox-sensitive endothelium-dependent vasorelaxation

Relaxations to AS-CS were significantly reduced when PCA rings were pre-incubated with an inhibitor of superoxide dismutase, DETCA (10 mM), as well as a superoxide dismutase mimetic MnTMPyP (100 uM) (Fig. 6). Similarly, in the presence of, PEG-catalase (H₂O₂ inhibitor, 300 U/ml), AS-CS induced relaxation were significantly reduced compared to relaxation produced in control rings (Fig. 6). In addition, AS-CS induced relaxations were also inhibited significantly by pre-incubation of PCA rings with the antioxidant N-acetylcysteine (10 mM) (Fig. 6).

Fig. 6

A to D show Vasorelaxant effect of aqueous soluble fraction of C. songarica (AS-CS) in U46619-pre-contracted coronary rings in the absence and presence of, A) sodium diethyldithiocarbamate (DETC; 10 mM), B) PEG-catalase (300 Uml^-1), C) N-acetylcysteine (10 mM) D) MnTMPyP (100 uM). Data are expressed as a percentage change from the U46619-induced contraction and are means±SEM of six separate experiments. ^**= p < 0.01 and ^***= p < 0.001 v control, 2-way ANOVA, followed by a Bonferroni post-hoc test. Figure E to F show effects of AS-CS on in situ formation of ROS on coronary rings. Dihydroethidium (DHE) loaded coronary rings were treated with AS-CS (5 mg/kg) in the absence and presence of, PEG-catalase, MnTMPyP and N-acetylcystine (NAC). The fluorescence was monitored with confocal microscope. ^***= p < 0.001 vs. control, a = p < 0.001 vs. AS-CS (n = 4). Pictures are showing dihydroethidium (DHE) fluorescence in untreated (control) and AS-CS treated sections.

3.8 AS-CS causes the endothelial formation of ROS in coronary arteries

In situ formation of ROS in coronary arteries was measured with the redox-sensitive fluorescent probe dihydroethidium, using confocal microscope. Treatment of coronary artery sections with AS-CS (5 mg/ml) significantly enhanced the fluorescent signal which was primarily related to the endothelium exhibiting formation of reactive oxygen species. This response was reduced in the presence of PEG-catalase, MnTMPyP and NAC (Fig. 6E & F).

3.9 Role of extracellular calcium in AS-CS induced relaxation

A high concentration of AS-CS (5 mg/ml) reduced the contractile response to CaCl₂ (1μM–3 mM) in PCAs that had been pre-stimulated with high K+ (60 mM) in the absence of extracellular Ca² ⁺ . The maximum contraction produced by 3 mM CaCl₂ was 1.43±0.12 g, which was reduced to 0.99±0.05 g in the presence of AS-CS (Fig. 7A). A significant reduction in AS-CS induced relaxation was observed in the absence of extracellular calcium. In calcium free Kreb’s-Henseleit buffer, 5 mg/ml AS-CS produced only 48.4±2.4% relaxation as compared to 83.7±5.7% relaxation in the presence of calcium (Fig. 7B). Pre-incubation with L-NAME had no further effect on the relaxation response in the absence of calcium (Fig. 7B).

Fig. 7

A) Inhibitory effect of aqueous soluble fraction of Crataegus songarica (AS-CS, 5 mg/ml) on contractions induced by cumulative addition of CaCl₂ (1μM to 3 mM) expressed a change in grams weight. B) Relaxation response to AS-CS in calcium free Krebs-Henseleit buffer in absence and presence of L-NAME. C) Relaxation response to 17β estradiol after pre-contraction with U46619 or 30 mM KCl. Data are expressed as a percentage change from the U46619-induced contraction and are means±SEM of 6–7 separate experiments. ^**= p < 0.01 and ^***= p < 0.001, 2-way ANOVA, followed by a Bonferroni post-hoc test.

As the estrogen receptor antagonists caused partial inhibition of the relaxation response to AS-CS, the effects of estradiol on vascular tone were determined. 17β estradiol produced a concentration-dependent relaxation of the PCA after pre-contraction with U46619. This was reduced slightly after pre-contraction with KCl (Fig. 7C).

3.10 AS-CS causes phosphorylation of eNOS and AKT in coronary arteries

Pronounced fluorescence signals for the phosphorylation of both Akt and eNOS were observed in PCAs incubated with AS-CS (5 mg/ml). In contrast, only a very faint fluorescence signal was observed in control rings (Fig. 8).

Fig. 8

Panel A) Aqueous soluble fraction of C. songarica (AS-CS, 5 mg/kg) induced phosphorylation of eNOS in coronary artery sections monitored with confocal microscope. Upper panel (graph) shows cumulative data and lower panels (pictures) show representative immunofluorescence staining. ^***= p < 0.001 vs. control (n = 3). eNOS = endothelial nitric oxide synthase. Panel B) Aqueous soluble fraction of C. songarica (AS-CS, 5 mg/kg) induced phosphorylation of AKT in coronary artery sections monitored with confocal microscope. Upper panel (graph) show cumulative data and lower panels (pictures) show representative immunofluorescence staining. ^***= p < 0.001 vs. control (n = 3).

4 Discussion

Fruits of Crataegus songarica have been used for the treatment of vascular disorders, including hypertension. In this study we performed an initial screen using butanol and aqueous extracts of C. songarica compared to crude extract. All the extract/fractions produced endothelium dependent vasorelaxation in coronary artery rings with maximum vasorelaxation effect induced by the aqueous soluble fraction of C. songarica (AS-CS). The fact that the aqueous extract produced a similar response to the crude extract, and a greater response than the butanol extract, suggests that the aqueous extract contains more of the active compounds. Therefore further experiments were designed to determine mechanism underlying the endothelium dependent vasorelaxation of AS-CS.

Endothelial cells modulate vascular homeostasis [28, 29] and respond to endocrine and physical stimuli by releasing endothelial mediators which includes endothelium derived hyperpolarizing factor (EDH), nitric oxide (NO) and prostacyclin that produce vasodilation of vessels [30, 31]. Among these factors NO, synthesized by endothelial nitric oxide synthase (eNOS) has attracted much attention because of its multiple vasoprotective properties [32]. In this study, L-NAME (eNOS inhibitor), ODQ (guanylyl cyclase inhibitor) and indomethacin (cyclooxygenase inhibitor) were used to evaluate the endothelium dependent relaxation effect of AS-CS. The results revealed that the relaxation response produced by AS-CS was significantly reduced by pretreatment with L-NAME and ODQ, whereas indomethacin had no effect on the AS-CS vascular effect, indicating that prostacyclin is not involved. Moreover, pretreatment with both indomethacin and L-NAME had no further effect beyond that seen with L-NAME alone, suggesting the role of NO/sGC/cGMP pathway in the AS-CS induced relaxation response. The remaining relaxation in the presence of L-NAME was similar to that seen after removal of the endothelium, indicating that NO is likely to be the sole mediator of the endothelium-dependent relaxation response [33]. It is well recognized that in endothelium, acetylcholine through M3 receptors, bradykinin through B2 receptors, substance P through tachykinin NK1 receptors and estrogen through ER-α receptors causes activation of endothelial nitric oxide (NO) synthase which in turn lead to enhanced bioavailability of NO [34 –37]. In our study, to determine whether components of the AS-CS extract could be acting through endothelial receptors, the effects of different receptor antagonists on the relaxation response were determined. PCA rings were pre-incubated with antagonists of these four receptors. Neither atropine, win64338, nor L760735 inhibited the relaxation to AS-CS, indicating the neither muscarinic, bradykinin B2, nor tachykinin NK1 receptors are involved in the relaxation response. Whereas, presence of both tamoxifen and ICI 182, 782 significantly decreased the AS-CS induced relaxation response in PCAs, indicating the involvement of estrogen receptors. A number of studies have demonstrated that estrogens and estrogen receptors can produce relaxation of blood vessels. In rat arteries, estradiol produces a relaxation through a potassium channel dependent mechanism [38]. In this present study, 17β estradiol produced a concentration-dependent relaxation of the PCA, which was reduced slightly after pre-contraction with KCl, similar to the effects seen with AS-CS.

It has also been recognized that various stimuli including estrogens, corticosteroids, shear stress and vascular endothelial growth factor activate a Src kinase/PI3-kinase pathway leading to activation of Akt protein kinase and phosphorylation of endothelial NO synthase resulting in enhanced synthesis of nitric oxide [39 –42]. The Src family of kinases are redox-sensitive kinases that act as upstream activators of the PI3-kinase/Akt pathway leading to activation of eNOS. PI3-kinase converts the plasma membrane lipid phosphatidylinositol-4,5-biphosphate to phosphatidylinositol- 3,4,5-triphosphate with subsequent activation of Akt [32, 43]. In our study, experiments were performed to determine the role of both Src kinase and the PI3-kinase in the relaxation response to AS-CS. Data obtained revealed that LY249002 (PI3-kinase inhibitor) and PP2 (Src tyrosine kinase inhibitor), both attenuated the AS-CS-induced vasorelaxation. A combination of the inhibitors had no further effect, indicating that they are part of the same pathway. Furthermore, addition of L-NAME alongside LY294002 and PP2 also had no further effect, lending support to the idea that the AS-CS induced relaxation response is mediated through stimulation of a PI3 kinase/Akt-eNOS-cGMP signaling pathway. Moreover, PP2 and LY294002 had no inhibitory effect on the AS-CS relaxation response in the presence of tamoxifen, demonstrating that the Src/PI-3K pathway is likely to be downstream of the estrogen receptor. These data were further confirmed by experiments showing an increase in phosphorylation of Akt in the presence of AS-CS, indicating an increase in levels of activated enzyme. Similarly, levels of phosphorylated eNOS were also increased in the presence of AS-CS, indicating increased levels of activated nitric oxide synthase. It is well established that estrogen and estrogen receptors produce relaxation of blood vessels (38, 44). Furthermore, a study in rat aorta has demonstrated that the relaxation to estradiol is attenuated by inhibition of PI3 kinase, similar to the response to AS-CS [45].

Synthesis of reactive oxygen species (ROS), particularly superoxide anions, activates the PI3-kinase/Akt signaling pathway in endothelial cells and enhances the formation of NO [46]. Previously, red wine polyphenols have been reported to trigger redox-sensitive activation of the PI3-kinase/Akt signaling pathway that leads to eNOS phosphorylation at Ser 1177 and subsequent formation of nitric oxide [27, 47]. Moreover, it has also been reported that hydrogen peroxide (H₂O₂) also acts as an upstream stimulator of the Src kinase/PI3-kinase/Akt pathway to cause phosphorylation of eNOS in endothelial cells [27 , 49]. In the current study, PEG-catalase (H₂O₂ inhibitor), diethyldithiocarbamate (DETCA, superoxide dismutase inhibitor) and MnTMPyP (SOD mimetic), prevented the AS-CS induced vasorelaxation, supporting a role for reactive oxygen species. Furthermore, to confirm these data, formation of ROS in response to AS-CS was detected with confocal microscope in porcine coronary rings. These data clearly indicated that AS-CS increased the formation of ROS in the endothelium of control rings, while the presence of various antioxidants reduced the formation of ROS in response to AS-CS. These data also rule out a role for antioxidants with the AS-CS extracts in the relaxation response.

It is well known that membrane potential also regulates vascular tone and this potential is regulated by movement of various ions (Na⁺, K+, Ca^2 +) through membrane by different ion channels i.e. potassium and calcium channels [29, 32]. Membrane potassium ion channels are potential target for anti-hypertensive drugs because of their contribution in the maintenance of vascular tone. cGMP activates Ca²⁺ activated potassium channels via protein kinase G phosphorylation that will lead to opening of K+ channels, therefore activation of K+ channels can be downstream of NO production [50 –52]. Opening of K+ channels in vascular smooth muscle cells results in an increase in K+ permeability, leading to K+ efflux. The exit of K+ from the vascular smooth muscle cells induces membrane hyperpolarization and consequently the closing of voltage-operated Ca^2 + channels, leading to vasorelaxation [53 –55]. Two sets of experiments suggest the contribution of K+ channels in the vasorelaxation effect of AS-CS. First, the relaxation to AS-CS was almost completely abolished in the PCAs pre-constricted with a high concentration of KCl [55, 56]. Second, in the presence of TEA (10 mM), a nonselective potassium channel blocker, the vasorelaxant response induced by increasing concentrations of AS-CS was significantly attenuated, validating our observation that AS-CS significantly affect K+ channels to induce relaxation response. On the other hand, the relaxation response was not affected by any selective K+ channel blockers (glibenclamide, barium chloride, iberiotoxin, TRAM-34 and apamin), suggesting that multiple types of K+ channels may be involved in relaxation response to AS-CS and inhibition of all K+ channels might be required to see a significant inhibition [32, 57].

Relaxation responses to AS-CS were also attenuated in the absence of extracellular calcium. Under these conditions, L-NAME had no further affect, indicating that the NO-dependent relaxation pathway also involves inhibition of calcium influx, possibly through a cGMP/PKG inhibition of calcium channels either directly, or through activation of K+ channels, as discussed above. These data were supported by experiments showing that AS-CS inhibits calcium-induced contraction of the PCA.

Plants containing polyphenols display a number of pharmacological properties; anti-hypertensive, vasorelaxant and antioxidant potential [58]. Polyphenols from black tea have been shown to activate the estrogen ERα receptor downstream of p38 MAP kinase, leading to stimulation of an Akt/ eNOS pathway independently of ligand activation of the receptor [59]. The authors suggest that the ERα has an adaptor function, similar to receptor transactivation seen with some tyrosine kinase receptors. Furthermore, activation of the ERα receptor in this way is thought to be different from that seen with estradiol. Therefore, although estradiol produced a concentration-dependent relaxation of the PCA in this present study, activation by plant polyphenols could potentially lead to stimulation of a different intracellular signaling pathway. It is not clear from the study by Anter et al which polyphenols lead to activation of the ERα receptor [59]. Previous analysis of the constituents of C. songarica identified a number of polyphenols that have been reported to have estrogenic activity [6]. These include metabolites of caffeic acid, which might activate ERα receptors [60]. Similarly metabolites of quercetin were also identified, which has been reported to activate ERα receptors in bovine endothelial cells [61].

5 Conclusion

The data presented in this study demonstate that the aqueous fraction of C. songarica berries produces an endothelium-dependent vasorelaxation of porcine coronary artery rings. The data indicate that the relaxation is sensitive to antagonism of estrogen receptors, as well as inhibition of Src, PI3K and nitric oxide pathways, and an increase in ROS. Part of the response involves inhibition of calcium-induced contractions. A caveat to this study is that we only tested the effects of C. songarica on relaxation responses in porcine coronary arteries. Studies in other blood vessels are required to confirm that the responses seen are common to all blood vessels.

Footnotes

Acknowledgment

The authors are thankful to Higher Education Commission of Pakistan, for providing IRSIP scholarship to Waqas Younis for visiting University of Nottingham for his doctoral research work. The authors are also thankful to Ministry of Foreign Affairs and International Development (MAEDI) France and the Ministry of Higher Education and Research (MESR) of France for funding this project through PERIDOT Research Program.

Authors’ contributions

Waqas Younis perform experiments and collect data of various activities. Richard Roberts supervised this study and provide the necessary facilities for experiments. Alamgeer and V.B. Schini-kerth provide help to design this study and in the correction of the manuscript. Muhamamd Akmal farooq helped in Immunohistochemical studies. All authors read and approved the final manuscript.

Conflicts of interest

The authors declare that they have no conflict of interests.

References

Lone

. Tree diversity and economic importance of forest trees of Kashmir (Jammu and Kashmir), India. Int J Fundam Appl Sci. 2013;2(4):56–63.

Lone

, Bhardwaj

, Shah

, et al. Ethnobotanical survey of some threatened medicinal plants of Kashmir Himalaya, India. J Med. Plant Res. 2014;8(47):1362–13673.

Ullah

, Rashid

, Parveen

. Medicinal plants used in the isolated region of Bumburet, Kalash Valley, District Chitral, Pakistan. Pak J Weed Sci Res. 2014;20(3).

Shah

, Khan

. Common medicinal folk recipes of siran valley, Mansehra, Pakistan. Ethnobot Leaflets. 2006;2006(1):5.

Kumar

, Paul

, Anand

. An ethnobotanical study of medicinal plants used by the locals in Kishtwar, Jammu and Kashmir, India. Ethnobot Leaflets. 2009;2009(10):5.

Younis

, Alamgeer , Schini-Kerth

, et al. Role of the NO/cGMP pathway and renin-angiotensin system in the hypotensive and diuretic effects of aqueous soluble fraction from Crataegus songarica K. Koch. J Ethnopharmacol. 2020;249:112400.

Koçyõldõz

ZÇ

, Birman

, Olgaç

, et al. Crataegus tanacetifolia leaf extract prevents L-NAME-induced hypertension in rats: a morphological study. Phytother Res. 2006;20(1):66–70.

Bernatonien≐

, Masteikova

, Majien≐

, et al. Free radical-scavenging activities of Crataegus monogyna extracts. Medicina. 2008;44(9):706.

Kuo

, Yeh

, Shieh

, et al. Effect of ShanZha, a Chinese herbal product, on obesity and dyslipidemia in hamsters receiving high-fat diet. J Ethnopharmacol. 2009;124(3):544–50.

10.

Wang

, Xiong

, Feng

. Effect of crataegus usage in cardiovascular disease prevention: an evidence-based approach. Evid. Based Complementary Altern. Med. 2013;2013.

11.

Hernández-Pérez

, Bah

, Ibarra-Alvarado

, et al. Aortic relaxant activity of Crataegus gracilior Phipps and identification of some of its chemical constituents. Molecules. 2014;19(12):20962–74.

12.

Forte

, Conti

, Damato

, et al. Targeting nitric oxide with natural derived compounds as a therapeutic strategy in vascular diseases. Oxid Med Cell Longev. 2016.

13.

Savoia

, Sada

, Zezza

, et al. Vascular inflammation and endothelial dysfunction in experimental hypertension. Int J Hypertens. 2011;2011.

14.

Incalza

, D’Oria

, Natalicchio

, et al. Oxidative stress and reactive oxygen species in endothelial dysfunction associated with cardiovascular and metabolic diseases. Vasc Pharmacol. 2018;100:1–9.

15.

Halcox

, Schenke

, Zalos

, et al. Prognostic value of coronary vascular endothelial dysfunction. Circulation. 2002;106(6):653–8.

16.

Cai

, Harrison

. Endothelial dysfunction in cardiovascular diseases: the role of oxidant stress. Circ Res. 2000;87(10):840–4.

17.

Griendling

, Sorescu

, Ushio-Fukai

. NAD (P) H oxidase: role in cardiovascular biology and disease. Cir res. 2000;86(5):494–501.

18.

Daiber

, Steven

, Weber

, et al. Targeting vascular (endothelial) dysfunction. Br. J. Pharmacol. 2017;174(12):1591–619.

19.

Alamgeer , Auger

, Chabert

, et al. Mechanisms underlying vasorelaxation induced in the porcine coronary arteries by Thymus linearis, Benth. J Ethnopharmacol. 2018;225:211–9.

20.

Wong

, Roberts

, Randall

. Sex differences in endothelial function in porcine coronary arteries: a role for H2O2 and gap junctions. Br J Pharmacol. 2014;171:2751–66.

21.

Wong

, Roberts

, Randall

. Sex differences in the role of transient receptor potential (TRP) channels in endothelium-dependent vasorelaxation in porcine isolated coronary arteries. Eur J Pharmacol. 2015;750:108–17.

22.

Dodson

, Rhoden

. Bradykinin increases Na+-K+pump activity in cultured guinea-pig tracheal smooth muscle cells. Br J Pharmacol. 2001;133(8):1339–45.

23.

Harrison

, Owens

, Williams

, et al. An orally active, water-soluble neurokinin-1 receptor antagonist suitable for both intravenous and oral clinical administration. J Med Chem. 2001;44(24):4296–9.

24.

Wong

, Garle

, Alexander

, et al. A role for the sodium pump in H2O2-induced vasorelaxation in porcine isolated coronary arteries. Pharmacol Res. 2014;90(1):25–35.

25.

Roberts

. α2 adrenoceptor-mediated vasoconstriction in porcine palmar lateral vein: role of phosphatidylinositol 3-kinase and EGF receptor transactivation Br J Pharmacol. 2003;138(1):107–16

26.

Rashid

, Idris-Khodja

, Auger

, et al. Polyphenol-rich blackcurrant juice prevents endothelial dysfunction in the mesenteric artery of cirrhotic rats with portal hypertension: role of oxidative stress and the angiotensin system. J Med Food. 2018;21(4):390–9.

27.

Anselm

, Chataigneau

, Ndiaye

, et al. Grape juice causes endothelium-dependent relaxation via a redox-sensitive Src-and Akt-dependent activation of eNOS. Cardiovasc. Res. 2007;73(2):404–13.

28.

França-Silva

, Luciano

, Ribeiro

, et al. The 2-nitrate-1, 3-dibuthoxypropan, a new nitric oxide donor, induces vasorelaxation in mesenteric arteries of the rat. Eur J Pharmacol. 2012;690(1-3):170–5.

29.

Tan

, Ch’ng

, Loh

, et al. Vasorelaxation effect of Glycyrrhizae uralensis through the endothelium-dependent pathway. J Ethnopharmacol. 2017;199:149–60.

30.

Ameer

, Salman

, Siddiqui

, et al. Pharmacological mechanisms underlying the vascular activities of Loranthus ferrugineus Roxb. in rat thoracic aorta. J Ethnopharmacol. 2010;127(1):19–25.

31.

Jin

, Wen

, Li

, et al. The mechanism of vasorelaxation induced by ethanol extract of Sophora flavescens in rat aorta. J Ethnopharmacol. 2011;137(1):547–52.

32.

Jin

, Wen

, Kim

, et al. Vascular relaxation by ethanol extract of Xanthoceras sorbifolia via Akt-and SOCE-eNOS-cGMP pathways. J Ethnopharmacol. 2010;132(1):240–5.

33.

De AF Da

, Crestani

, de Souza

, et al. Endothelium-dependent and independent vasorelaxation induced by an n-butanolic fraction of bark of Scutia buxifolia Reiss (Rhamanaceae). J Ethnopharmacol. 2012;141(3):997–1004.

34.

Xia

, Krukoff

. Estrogen induces nitric oxide production via activation of constitutive nitric oxide synthases in human neuroblastoma cells. Endocrinology. 2004;145(10):4550–7.

35.

Karabucak

, Walsch

, Jou

, et al. The role of endothelial nitric oxide in the Substance P induced vasodilation in bovine dental pulp. J Endod. 2005;31(10):733–6.

36.

Junior

, Gasparotto

, Lourenço

, et al. Antihypertensive effects of isoquercitrin and extracts from Tropaeolum majus L.: evidence for the inhibition of angiotensin converting enzyme. J Ethnopharmacol. 2011;134(2):363–72.

37.

Lara

, Bourgeois

, El-Dahr

, et al. Bradykinin/B2 receptor activation regulates renin in M-1 cells via protein kinase C and nitric oxide. Physiol Rep. 2017;5(7):e13211.

38.

Tep-areenan

, Kendall

, Randall

. Mechanisms of vasorelaxation to 17β-oestradiol in rat arteries. Eur. J. Pharmacol. 2003;22;476(1-2):139–49.

39.

Dimmeler

, Fleming

, Fisslthaler

, et al. Activation of nitric oxide synthase in endothelial cells by Akt-dependent phosphorylation. Nature. 1999;399(6736):601–5.

40.

Fulton

, Gratton

, McCabe

, et al. Regulation of endothelium-derived nitric oxide production by the protein kinase Akt. Nature. 1999;399(6736):597–601.

41.

Simoncini

, Hafezi-Moghadam

, Brazil

, et al. Interaction of oestrogen receptor with the regulatory subunit of phosphatidylinositol-3-OH kinase. Nature. 2000;407(6803):538–41.

42.

Ndiaye

, Chataigneau

, et al. Red wine polyphenols induce EDHF-mediated relaxations in porcine coronary arteries through the redox-sensitive activation of the PI3-kinase/Akt pathway. Br J Pharmacol. 2004;142(7):1131–6.

43.

Cantley

. The phosphoinositide 3-kinase pathway. Science. 2002;296(5573):1655–7.

44.

Teoh

, Quan

, Leung

, Man

. Vascular effects of estrone and diethylstilbestrol in porcine coronary arteries. Menopause. 2009;1;16(1):104–9.

45.

Bucci

, Roviezzo

, Cicala

, et al. 17-β-oestradiol-induced vasorelaxation in vitro is mediated by eNOS through hsp90 and akt/pkb dependent mechanism. Br J Pharmacol. 2002;135(7):1695–700.

46.

Alhosin

, Anselm

, Rashid

, et al. Redox-sensitive up-regulation of eNOS by purple grape juice in endothelial cells: role of PI3-kinase/Akt, p38 MAPK, JNK, FoxO1 and FoxO3a. PLoS One. 2013;8(3):e57883.

47.

Ndiaye

, Chataigneau

, Lobysheva

, et al. Red wine polyphenols-induced, endothelium-dependent NO-mediated relaxation is due to the redox-sensitive PI3-kinase/Akt-dependent phosphorylation of endothelial NO-synthase in the isolated porcine coronary artery. FASEB J. 2005;19(3):1–20.

48.

Drummond

, Cai

, Davis

, et al. Transcriptional and posttranscriptional regulation of endothelial nitric oxide synthase expression by hydrogen peroxide. Circ Res. 2000;86(3):347–54.

49.

Kim

, Formoso

, Li

, et al. Epigallocatechin gallate, a green tea polyphenol, mediates NO-dependent vasodilation using signaling pathways in vascular endothelium requiring reactive oxygen species and Fyn. J Biol Chem. 2007;282(18):13736–45.

50.

Kubo

, Nakaya

, Matsuoka

, et al. Atrial natriuretic factor and isosorbide dinitrate modulate the gating of ATP-sensitive K+channels in cultured vascular smooth muscle cells. Circ Res. 1994;74(3):471–6.

51.

Nelson

, Quayle

. Physiological roles and properties of potassium channels in arterial smooth muscle. Am J Physiol. 1995;268(4):C799–C822.

52.

Baranowska-Kuczko

, Kozłowska

, Kozłowski

, et al. Mechanisms of endothelium-dependent relaxation evoked by anandamide in isolated human pulmonary arteries. N-S Arch Pharmacol. 2014;387(5):477–86.

53.

, Han

, Jung

, et al. Physiological roles of K+channels in vascular smooth muscle cells. J Smooth Muscle Res. 2008;44(2):65–81.

54.

Dick

, Tune

. Role of potassium channels in coronary vasodilation. Exp Biol Med. 2010;235(1):10–22.

55.

Reis

, Oliveira Filho

, Rodrigues

, et al. Involvement of potassium channels in vasorelaxant effect induced by Valeriana prionophylla Standl. in rat mesenteric artery. Evid. Based Complementary Altern. Med. 2013;2013.

56.

Kozłowska

, Baranowska

, Schlicker

, et al. Virodhamine relaxes the human pulmonary artery through the endothelial cannabinoid receptor and indirectly through a COX product. Br J Pharmacol. 2008;155(7):1034–42.

57.

Kim

, Kang

, Kim

, et al. Procyanidins in crataegus extract evoke endothelium-dependent vasorelaxation in rat aorta. Life Sci. 2000;67(2):121–31.

58.

Aparecida

, Vergutz Menetrier

, Luiz Botelho Lourenco

, Gasparotto Junior

. Cellular and molecular mechanisms of diuretic plants: an overview. Curr Pharm Des. 2017;23(8):1247–52.

59.

Anter

, Chen

, Shapira

, et al. p38 mitogen-activated protein kinase activates eNOS in endothelial cells by an estrogen receptor α-dependent pathway in response to black tea polyphenols. Circ Res. 2005;96:1072–8.

60.

Kiyama

. Estrogenic activity of coffee constituents. Nutrients. 2019;11:1401.

61.

Liu

, He

, Wang

, et al. Selected phytoestrogens distinguish roles of Era transactivation and ligand binding of anti-inflammatory activity. Endocrinology. 2018;159:3351–64.