Vasorelaxant and Hypotensive Effects of an Ethanolic Extract of Eulophia macrobulbon and Its Main Compound 1-(4′-Hydroxybenzyl)-4,8-Dimethoxyphenanthrene-2,7-Diol

Background: Ethnopharmacological studies demonstrated the potential for Eulophia species to treat inflammation, cancer, and cardio-metabolic diseases. The aim of the study was to investigate the vasorelaxant effect of ethanolic Eulophia macrobulbon (EM) extract and its main phenanthrene on rat isolated mesenteric artery and to investigate the hypotensive effect of EM. Methods: The vasorelaxant effects of EM extract or phenanthrene and the underlying mechanisms were evaluated on second-order mesenteric arteries from Sprague Dawley rats. In addition, the acute hypotensive effect was evaluated in anesthetized rats infused with cumulative concentrations of the EM extract. Results: Both EM extract (10-4–1 mg/ml) and phenanthrene (10-7–10-4 M) relaxed endothelium-intact arteries, an effect that was partly reduced by endothelium removal (p < 0.001). A significant decrease in the relaxant effect of the extract and the phenanthrene was observed with L-NAME and apamin/charybdotoxin in endothelium-intact vessels, and with iberiotoxin in denuded vessels. SNP (sodium nitroprusside)-induced relaxation was significantly enhanced by EM extract and phenanthrene. By contrast, ODQ (1H-[1,2,4]oxadiazolo[4,3-a]quinoxaline-1-one), 4-aminopyridine and glibenclamide (endothelium-denuded vessels) and indomethacin (endothelium-intact vessels) had no effect. In calcium-free solution, both the EM extract and phenanthrene inhibited extracellular Ca2+-induced contraction in high KCl and phenylephrine (PE) pre-contracted rings. They also inhibited the intracellular Ca2+ release sensitive to PE. The acute infusion of EM extract (20 and 70 mg/kg) induced an immediate and transient dose-dependent hypotensive effect. Conclusion: The ethanolic extract of EM tubers and its main active compound, 1-(4′-hydroxybenzyl)-4,8-dimethoxyphenanthrene-2,7-diol (phenanthrene) induced vasorelaxant effects on rat resistance vessels, through pleiotropic effects including endothelium-dependent effects (NOS activation, enhanced EDH production) and endothelium-independent effects (opening of KCa channels, inhibition of Ca2+ channels, inhibition of intracellular Ca2+ release and PDE inhibition).


INTRODUCTION
Orchids belong to the plant family Orchidaceae, one of the most diverse group among the angiosperm with near 25,000 species (Behera et al., 2013). Aside from their ornamental value, orchids are also acknowledged for their use in traditional medicines (Bulpitt, 2005;Behera et al., 2013;Pant, 2013;Patil and Mahajan, 2013;Narkhede et al., 2016). Chinese medicine was probably the first to describe orchids for their medicinal use. Other pharmacopeias from India and many countries from South-Asia such as Taiwan, Singapore, Vietnam, Sri Lanka, Thailand, Myanmar, use orchids in traditional medicine since the ancient time (Suresh Kumar et al., 2000;Hernández-Romero et al., 2005;Luo et al., 2007). Likewise, the use of orchids in America also has a long history. Phytochemically, orchids have been reported to contain alkaloids, triterpenoids, flavonoids and stilbenoids (Kong et al., 2003;Kovács et al., 2008;Singh et al., 2012). Surprisingly, limited information on medicinal values of orchids regarding their therapeutic properties in different parts of the world is available, whereas such information would be helpful for sustainable management of resources.
As PDE enzymes are widely expressed in smooth muscle cells including vascular smooth muscle cells (VSMC) (Komas et al., 1991), the aim of the present study was to investigate the vasorelaxant effect of an ethanolic extract of EM and its main constituent, 1-(4 -hydroxybenzyl)-4,8-dimethoxyphenanthrene-2,7-diol, and to unravel the mechanisms involved on isolated rat mesenteric artery. To determine whether the in vitro vasorelaxant properties translate into an in vivo effect, the acute hypotensive effect of EM extract was determined in anesthetized rats.

Preparation of the Plant Extract
Fresh tubers of EM were collected from Prachinburi province, Thailand. The voucher specimen (No. 002716) was identified by Associate Professor Dr. Anupan Kongbungkerd, Department of Biology, Faculty of Science, Naresuan University, and kept at Faculty of Science, Naresuan University, Phitsanulok, Thailand. As previously described (Temkitthawon et al., 2017), the tubers of EM were cut and dried at 55 • C. The dried material (2 kg) was ground into a fine powder, macerated with 95% ethanol (14 L) for 3 days/time (two times). Then, it was filtered and evaporated under vacuum until dryness to give the crude ethanolic extract with the yield of 15.78% (w/w). This extract contained 0.52% (w/w) of a phenanthrene, 1-(4hydroxybenzyl)-4,8-dimethoxyphenanthrene-2,7-diol, which was used as a bioactive marker and a major compound of the extract (Temkitthawon et al., 2017). This compound was isolated from EM extract as described by Temkitthawon et al. (2017). Briefly, this compound was isolated and purified by using column chromatography, preparative thin layer chromatography, size exclusion chromatography, high performance liquid chromatography (HPLC) and recrystallization techniques. The EM extract and the phenanthrene were stored at −20 • C until used.

Animals
Male Sprague Dawley rats (8-12 weeks-old) were purchased from Janvier (Le Genest Saint Isle, France) for the in vitro study of vascular reactivity and from National Laboratory Animal Center (Mahidol University, Salaya, Thailand) for the in vivo study. Animals were kept under 12-12 h light: dark cycle, at 22 ± 1 • C and allowed free access to standard food and water. The investigation complied with the ARRIVE animal research: reporting in vivo experiments. All protocols were approved by the local committees for ethics in animal experimentation No. 2015/001-CD/5PR of University of Franche-Comté (Besançon, France) and Naresuan University Animal Care and Use Committee (NUACUC, Naresuan University, Phitsanulok, Thailand, ethic approval number: NU-AE591025).

Mesenteric Arteries Preparation
Rats were anesthetized with sodium pentobarbital (Ceva Santé Animale, France) (60 mg/kg, i.p.). Second order branches of mesenteric arteries were placed in Krebs solution with the following composition (mM): NaCl 118, KCl 4.7, KH 2 PO 4 1.2, MgSO 4 1.2, CaCl 2 2.5, NaHCO 3 25, glucose 12, maintained at pH of 7.4, 37 • C, and continuously aerated with 95% O 2 , 5% CO 2 . Then 2-mm segments of artery were mounted in organ chambers and threaded on two 40-µm diameter stainless steel wires. To measure isometric force, the artery segments were connected into a Multi Myograph System (Model 610 M v.2.2, DMT A/S, Denmark). The data were recorded using Chart TM Ver.7 (ADInstruments, France). After a 15-min equilibration period, segments were stretched to their optimal lumen diameter for active tension development. Optimal lumen diameter was determined based on the internal circumference/wall tension ratio of the segments by setting the internal circumference to 90% of what the vessels would have if they were exposed to a passive tension equivalent to that produced by a transmural pressure of 100 mmHg. Optimal lumen diameter was determined using specific software for normalization of resistance arteries (DMT Normalization Module; ADInstruments). After an initial equilibration period of 30 min, viability of vessels was tested from their vasoconstriction to high extracellular KCl (100 mM). The presence of functional endothelium was confirmed by more than 60% relaxation to the endothelium-dependent agonist acetylcholine (ACh 10 −5 M) after pre-constriction with phenylephrine (PE 10 −5 M). In some rings, endothelium was mechanically removed by gently rubbing inside the vessel with a small wire. The completeness of endothelial denudation was confirmed by the absence of relaxation to ACh (10 −5 M). Arteries were again allowed to equilibrate 30 min before the start of the experiments.

Vasorelaxant Effect of EM Extract and Its Main Compound
To investigate the relaxant effect of the EM extract and 1-(4 -hydroxybenzyl)-4,8-dimethoxyphenanthrene-2,7-diol (phenanthrene), mesenteric rings were pre-contracted with submaximal concentration of PE (10 −5 M). When the contraction reached a plateau, the EM extract (10 −4 -1 mg/ml) or phenanthrene (10 −7 -10 −4 M) were added cumulatively to endothelium-intact rings as well as to endothelium-denuded rings to obtain a concentration-response curve. The relaxation effect was calculated as the percentage of the contraction in response to PE. The effect of the solvent, dimethyl sulfoxide (DMSO 0.23% for the EM extract, 0.18% for the phenanthrene), was evaluated in the same conditions.

Role of Endothelium-Dependent Pathways
To evaluate the role of endothelial nitric oxide synthase (eNOS), cyclo-oxygenase (COX) and endothelium-derived hyperpolarizing factor (EDH) pathways in the vasorelaxant activity of the EM extract or its main compound, endotheliumintact mesenteric rings were incubated with the NOS inhibitor, N G -nitro-L-arginine methyl ester (L-NAME, 10 −4 M), the COX inhibitor, indomethacin (10 −5 M) or with a combination of the small calcium-activated potassium channel blocker, apamin (10 −7 M) and the intermediate and large conductance calcium-activated potassium channel blocker, charybdotoxin (10 −7 M) for 30 min before pre-contraction with PE (10 −5 M), respectively. Then, cumulative concentrations of the EM extract or phenanthrene were added.

Role of K + Channels From Vascular Smooth Muscle
To investigate the contribution of K + Channels to the relaxant effect of the EM extract and phenanthrene, endotheliumdenuded mesenteric rings were pre-incubated with the voltagegated potassium channel (K V ) blocker, 4-aminopyridine (4-AP, 10 −3 M), the ATP-sensitive potassium channel (K ATP ) blocker, glibenclamide (10 −5 M), or the large conductance Ca 2+activated K + channels (K Ca ) blocker, iberiotoxin (10 −7 M), Frontiers in Pharmacology | www.frontiersin.org for 30 min before pre-contraction with PE (10 −5 M). Then, cumulative concentrations of the EM extract were added.

Role of sGC/cGMP Pathway
To assess whether the relaxant effect of extract and phenanthrene might be dependent on PDE inhibition, we determined if they induced changes in the soluble guanylyl cyclase (sGC)/cyclic guanosine monophosphate (cGMP) pathway in vascular smooth muscle cell (VSMC). Endothelium-denuded (E−) mesenteric rings were incubated with the vehicle (0.06% for the EM extract, 0.1% for the phenanthrene), EM extract (125 µg/ml) or phenanthrene (35 µM) for 10 min. Then rings were contracted with 10 −5 M PE (for the EM extract) or 100 mM KCl [for the phenanthrene, as rings pre-incubated with phenanthrene did not contract to 10 −5 M PE or low KCl (30 mM)], and then exposed to cumulative concentrations (10 −11 -10 −4 M) of sodium nitroprusside (SNP), a NO donor. To determine whether the EM extract or phenanthrene directly activates sGC or rather acts downstream of sGC, i.e., on PDE, endotheliumdenuded rings were incubated with the selective sGC inhibitor, 1H-[1,2,4]oxadiazolo[4,3-a]quinoxaline-1-one (ODQ, 10 −5 M) for 30 min, before contraction with 10 −5 M PE. Then, cumulative concentrations of the EM extract or phenanthrene were added.

Role of Calcium Channels
To assess the role of the extracellular calcium (Ca 2+ ) influx in the extract-or phenanthrene-induced relaxations, endotheliumdenuded mesenteric rings were incubated with a Ca 2+ -free Krebs solution containing 1 mM, methylene glycol-bis (2aminoethylether)-N,N,N' ,N'-tetraacetic acid (EGTA) for 40 min. Rings were firstly contracted with 10 −5 M PE to deplete intracellular Ca 2+ store from sarcoplasmic reticulum (SR) and washed four times every 10 min with the Ca 2+ -free Krebs solution. Then, rings were incubated with the vehicle or the EM extract at the EC 50 (125 µg/ml) or phenanthrene at the EC 50 (35 µM) for 10 min before 10 −5 M PE or 8x10 −2 M KCl was applied. Then, 10 −2 M CaCl 2 was added to evoke a contractile response (Senejoux et al., 2013).
To determine the effect of EM extract and phenanthrene on intracellular Ca 2+ release, the endothelium-denuded mesenteric rings were incubated with the L-type voltage dependent Ca 2+ channel inhibitor verapamil (10 −7 M) for 30 min. Then, 10 −5 M PE was added to the bath, and vehicle, EM extract (125 µg/ml) or phenanthrene (35 µM) were added. Under these conditions, the observed PE-induced contractions are caused by intracellular Ca 2+ release via the opening of inositol 1,4,5 trisphosphate (IP 3 ) receptors from SR (Tom et al., 2010).

Acute Effect of EM Extract on Blood Pressure and Heart Rate
To investigate whether the direct in vitro vascular effect of EM extract on resistance vessels translates into an in vivo effect, we studied the acute hypotensive effect of EM extract in anesthetized rats. The effect of phenanthrene was not studied because of the low quantity and poor dissolution of the isolated compound. Normotensive male Sprague Dawley rats were anesthetized with . After a 15-min stabilization period, systolic arterial blood pressure (SBP), diastolic blood pressure (DBP), and heart rate (HR) were recorded before and during intravenous infusion (at 1 ml/min) of 1 ml/kg of saline, vehicle (a saline solution containing 5% DMSO), EM extract (1, 5, 20, 70 mg/kg), or the NO donor SNP (25 µg/kg). The dose of SNP was chosen as a dose inducing a maximum lowering effect on blood pressure (Pagani et al., 1978;Kamkaew et al., 2013). The EM extract was dissolved in DMSO, diluted in saline and filtered through 0.2 µM syringe filter. Each subsequent infusion was administered when baseline values were fully recovered and waited for 15 min before the next infusion.

Statistical Analysis
All data were expressed as means ± standard error of mean (SEM) of n animals (for in vivo studies) or n vessel segments (for in vitro studies). The EM extract or phenanthrene induced vasorelaxation (%) was calculated as the percentage of the contraction to PE at 10 −5 M. The EC 50 values (defined as the concentration of extract that induced 50% of the maximal relaxation) and E max values (values of maximal relaxation) were determined by fitting the original concentration-response curves using GraphPad Prism software (version 5.0). Concentrationresponse relationships were analyzed using two-way ANOVA followed by the Bonferroni's test. Unpaired Student's t-test was used for two group comparisons. The multiple comparisons were analyzed using one-way ANOVA followed by Tukey's test. p < 0.05 was considered statistically significant.

Vasorelaxant Effect of the EM Extract and Its Main Compound
As shown in Figure 1 and Table 1, both EM extract ( Figure 1A) and phenanthrene ( Figure 1B) induced a dose-dependent vasorelaxation in endothelium-intact (E+, EC 50 = 0.05 ± 0.01 mg/ml and 8.1 ± 2.0 µM, respectively) and endothelium-denuded rings (E−, EC 50 = 0.12 ± 0.01 mg/ml and 35.1 ± 4.8 µM, respectively). The removal of the endothelium significantly blunted the relaxant effect of EM extract and phenanthrene (p < 0.01, Figure 1), as confirmed by the 2.3-and 4.3-fold increase in the EC 50 values, respectively ( Table 1). The solvent had no vasorelaxant effect (Figure 1). In summary, EM extract and phenanthrene induced both endothelium-dependent and-independent vasorelaxant effects.

Endothelium-Dependent Pathways Involved in EM Extract-and Phenanthrene-Induced Relaxation
As shown in Figure 2 and Table 1, the inhibition of NOS by L-NAME and EDH production by the combination of apamin and charybdotoxin significantly reduced the relaxation induced by EM extract (Figures 2A,C), and by the phenanthrene (Figures 2D,F), leading to an increase in the EC 50 values of EM extract and phenanthrene without changes of the E max values ( Table 1). By contrast, indomethacin did not change the relaxant effect of EM extract and phenanthrene (Figures 2B,E).

Involvement of K + Channels in EM Extract-and Phenanthrene-Induced Relaxation
As presented in Figure 3 and Table 1, iberiotoxin (a blocker of K Ca channels) significantly reduced the EM extractand phenanthrene-induced relaxation in endotheliumdenuded rings (Figures 3C,F), while other K + channel blockers, e.g., 4-AP (a blocker of K V channels) or glibenclamide (a blocker of K ATP channels) had no effect (Figures 3A,B,D,E).

Involvement of sGC/cGMP Pathway in EM Extract-and Phenanthrene-Induced Relaxation
To further understand the possible mechanisms underlying the endothelium-independent effect of EM extract and phenanthrene, the role of sGC/cGMP pathway was assessed in endothelium-denuded rings. As shown in Figures 4A,B

Role of Intracellular Ca 2+ Release
To determine if the EM extract-or phenanthrene-induced relaxation was related to the inhibition of intracellular Ca 2+ release, endothelium-denuded rings were incubated with 10 −7 M verapamil followed by the application of PE and EM extract (125 µg/ml) or phenanthrene (35 µM) or vehicle. As shown in Figure 6, the EM extract ( Figure 6A) and phenanthrene ( Figure 6B) significantly decreased the contraction to PE (p < 0.001).
Acute Effect of EM Extract on Blood Pressure Levels Figure 7A and Table 2 shows the effects of intravenous injection of the vehicle, EM extract and SNP on blood pressure. The values for the control (infusion of saline) were SBP = 109 ± 3 mmHg, DBP = 67 ± 2 mmHg, MAP = 81 ± 2 mmHg, HR = 296 ± 10 beats per minute (BPM). After infusion of the vehicle (DMSO 5%), the values were SBP = 109 ± 3 mmHg, DBP = 67 ± 2, MAP = 81 ± 2 mmHg, HR = 295 ± 13 BPM (no significant vs. control values). As compared to vehicle, infusion of EM extract at the doses of 20 and 70 mg/kg resulted in a significant dose-dependent reduction in SBP, DBP and MAP (Figures 7B-D and Table 2). Conversely, HR was unchanged whatever the dose of extract ( Figure 7E). The blood pressure lowering effect of EM extract was significantly lower than that of SNP (25 µg/kg).

DISCUSSION
The new findings of this study are that the ethanolic extract of Eulophia macrobulbon (EM) tubers and its main active compound, 1-(4 -hydroxybenzyl)-4,8-dimethoxyphenanthrene-2,7-diol (phenanthrene) induced a vasorelaxant effect of resistance vessels and acutely decreased blood pressure. In the cardiovascular system, the tone of vascular smooth muscle of small arteries and arterioles determines the vascular resistance, vascular function and blood pressure (Jackson, 2000;Sandoo et al., 2010). The endothelium is an important regulator of vascular tone, through the release of vasodilatory mediators such as NO, EDH, and prostacyclin (PGI 2 ) (Sandoo et al., 2010). In addition, ion channels in the plasma membrane of VSMC including different types of K + channels and Ca 2+ channels play a seminal role by determining cytosolic Ca 2+ concentration and the sensitivity of contractile machinery to Ca 2+ (Jackson, 2000;Webb, 2003). The results of the present study demonstrated that both the extract and the pure phenanthrene induced a relaxant effect that combined endothelium-dependent and endotheliumindependent mechanisms. As regards endothelium-dependent pathways, the data showed that increased production of EDH and to a lesser extent activation of NOS is involved in the effect of FIGURE 7 | Acute effect of EM extract on blood pressure in normotensive anesthetized rats. (A) blood pressure traces showing the effect of single intravenous injections of vehicle (5% DMSO in saline), EM extract (1, 5, 20, 70 mg/kg BW and SNP 25 µg/kg BW, over 20 s). The bars show the percentage reduction in systolic (B), diastolic (C), mean arterial blood pressure (D) and the percent change in heart rate (E), calculated from the values measured at the peak effect of DMSO, EM extract, and SNP at the doses indicated, as compared to the baseline value before each infusion. Data are expressed as means ± SEM of n animals. * p < 0.05, * * p < 0.01, * * * p < 0.001 vs. vehicle, † p < 0.001 vs. EM extract 20 mg/kg, # p < 0.001 vs. EM extract 70 mg/kg, (n = 10-16).
EM extract, whereas both production of EDH and activation of NOS contribute to the effect of phenanthrene. By contrast, COX activation is unchanged by both extract and pure compound. This latter finding contrasts with that of Chinsamy et al. (2014) showing that a dichloromethane extract of tubers from Eulophia hereroensis exhibited a potent in vitro COX inhibitory effect. By contrast, a phenanthrene isolated from Eulophia ochreata (9,10-dihydro-2,5-dimethoxyphenanthrene-1,7-diol) prevented LPS-induced COX-2 expression in isolated monocytes (Datla et al., 2010). Contrary to large vessels in which NO is the major endothelium-derived factor (Wu et al., 1993), EDH is thought to play a major role in the endothelium-dependent vasorelaxation in resistance vessels (Félétou and Vanhoutte, 1999;Chauhan et al., 2003). Of interest, albeit beneficial vascular effects such as a relaxant effect on aorta (Rendón-Vallejo et al., 2012) or an inhibitory effect on expression of adhesion molecules in isolated VSMC have been reported with various phenanthrene derivatives (Huang et al., 2008;Choi et al., 2012;Lo et al., 2017), to the best of our knowledge, our study is the first to show that a benzylated phenanthrene derivative improves vascular function. Our data revealed that the relaxant effect of EM extract and its main compound are mainly due to endothelium-independent mechanisms. The recent demonstration that EM extract as well as its main phenanthrene are potent PDE 5 inhibitors (Temkitthawon et al., 2017) led us to investigate their effect on the sGC/cGMP pathway. After NO release from endothelial cells, NO activates sGC in VSMC, resulting in the generation of cGMP. The increase in intracellular cGMP concentration activates cGMPdependent protein kinase (PKG), which causes vasorelaxation via the modulation of Ca 2+ channels as well as by decreasing the Ca 2+ sensitivity of the vascular smooth muscle contractile proteins (Morgado et al., 2012). Then intracellular cGMP is The systolic (SBP), diastolic (DBP), mean arterial blood pressure (MAP), and heart rate (HR) were measured during the peak effect of single intravenous injections of vehicle, EM extract, and SNP at the doses indicated. Bpm: beats per minute. All data are expressed as means ± SEM of n animals. * p < 0.05, * * p < 0.01, * * * p < 0.001 vs. vehicle. rapidly inactivated to GMP by the activity of PDEs. Therefore, cGMP concentration in smooth muscle cells is mainly dependent on the balance between its production by sGC and its breakdown by PDE (Rybalkin et al., 2003;Francis et al., 2010;Morgado et al., 2012;Maurice et al., 2014). Our results showed that incubation of denuded rings with the EM extract or its main compound significantly enhanced the relaxant effect of the NO donor SNP, a direct activator of sGC, suggesting that either sGC activity or cGMP production were enhanced by Eulophia. The fact that ODQ, a selective sGC inhibitor, had no effect on the relaxation induced by EM extract or its main compound, discarded the first hypothesis and suggested that they act downstream by increasing the cGMP availability likely via PDE inhibition. Indeed, rat resistance arteries express four major types of PDE including PDE3 and PDE1 inhibition which hydrolyzes cGMP, PDE4 which hydrolyzes cAMP and PDE3 which mainly hydrolyzes cAMP but also cGMP (Komas et al., 1991;Sampson et al., 2001;Lugnier, 2006). Our previous in vitro experiments showed that EM extract/phenanthrene have a greater inhibitory effect on PDE5 as compared to PDE1 (Temkitthawon et al., 2017), suggesting that its effect on cGMP pathway is mainly due to PDE5 inhibition. However, its capacity to inhibit cAMP-dependent PDE is unknown so it cannot be excluded that its relaxant effect partly relates on PDE3 or PDE4 inhibition and the subsequent increase in cAMP-PKA signaling.
Even though further studies are required to confirm the role of PDE5 inhibition in the relaxant effect of EM extract and phenanthrene, the contribution of this mechanism seems of moderate importance as compared to other endotheliumindependent mechanisms implying VSMC K + and Ca 2+ channels. Among the K + channels inhibitors tested, only iberiotoxin significantly hampered extract-and phenanthreneinduced relaxation, indicating that they were both able to stimulate large-conductance calcium-activated K + channels which are extensively expressed in mesenteric arteries (Beleznai et al., 2011). As these channels are activated by intracellular Ca 2+ (Beleznai et al., 2011), we determined whether EM extract and phenanthrene might modulate Ca 2+ homeostasis in VSMC. To assess whether EM extract and phenanthrene modified the extracellular Ca 2+ influx, experiments were conducted on rings contracted with PE or KCl in Ca 2+ -free Krebs solution in which Ca 2+ was added subsequently. The fact that the extract and phenanthrene dramatically reduced Ca 2+ -induced contraction after both PE and KCl exposure indicates that the blockade of both ROCCs and VGCCs is involved in the vasodilating effects. Moreover, our results showing that EM extract and phenanthrene also decreased the contraction induced by PE in the presence of verapamil, an inhibitor of VGCCs and in a lesser extent of ROCCs (Striggow and Bohnensack, 1993;Shin et al., 2005), demonstrate that they also inhibit the intracellular Ca 2+ release from the IP 3 -sensitive stores of SR in mesenteric rings.
The present study was conducted on small branches of mesenteric arteries, i.e., on only one of the vascular beds contributing to the global peripheral vascular resistances. Whether the mechanisms would be the same in other arterioles requires further studies. However, the above data obtained in isolated vessels suggested that in vivo, EM extract would induce a hypotensive effect. Our results showed that this was the case as the cumulative infusion of the EM extract dose-dependently decreased blood pressure without changing heart rate, indicating that EM extract had a strong effect on various resistance vessels. Whether this hypotensive effect combined with the enhanced production of endothelium-derived relaxing factors observed in vitro would translate into positive effects on vascular diseases warrants further investigation.

CONCLUSION
The ethanolic extract of Eulophia macrobulbon (EM) tubers and its main active compound, 1-(4 -hydroxybenzyl)-4,8dimethoxyphenanthrene-2,7-diol (phenanthrene) induced vasorelaxant effects on resistance vessels, through pleiotropic effects including mainly endothelium-independent effects (inhibition of Ca 2+ fluxes, opening of K Ca channels, increase in cGMP signaling) and to a lesser extent endotheliumdependent effects (NOS activation, enhanced EDH production). The mechanisms of the extract are similar to those of the phenanthrene suggesting that this compound is mainly responsible for the vascular effects of EM extract.

AUTHOR CONTRIBUTIONS
SW: experimental design, in vivo and in vitro experiments, data analysis and interpretation, and manuscript drafting. KC: experimental design, data analysis and interpretation, and manuscript drafting. HM: experimental design and manuscript drafting. KI: preparation of the extract and isolated pure compound, and experimental design. PrT: preparation of the extract and isolated pure compound, and experimental design. PeT: experimental design, in vitro experiments, and data analysis and interpretation. CD: experimental design, data analysis and interpretation, and manuscript drafting.