Antihypertensive and vasorelaxant effects of Rhizoma corydalis and its active component tetrahydropalmatine via NO/cGMP pathway and calcium channel blockade in isolated rat thoracic aorta

Abstract

Rhizoma corydalis has been used for the treatment of a variety of cardiovascular diseases in China. Tetrahydropalmatine (THP), one of the main active ingredients isolated from Rhizoma corydalis, is reported to have potent analgesic effects and it has been used in Chinese clinical practice for many years. The main roles of the extract of Rhizoma corydalis and THP are their analgesic effect and protective effect towards the cardio-cerebral vascular system. Although there is a lot of research about the cardiovascular protective effects of Rhizoma corydalis and THP, their vasorelaxant effects on thoracic aorta have not been well studied. Therefore, the aim of the present study was designed to observe the acute antihypertensive effect of a 70% ethanol extract of Rhizoma corydalis (RC) and THP as well as investigating the possible mechanisms of vasodilatation induced by RC and THP on isolated rat aorta. The in vivo acute antihypertensive activity was measured on spontaneously hypertensive rats. Measurements (systolic blood pressure and diastolic blood pressure) were recorded before and after RC and THP treatments at 0, 1, 2, 4 and 6 h by a tail cuff method. Isolated rat thoracic aorta were used in vitro, including endothelium-intact and endothelium-denuded aortic rings. Specific inhibitors including L-NAME, ODQ, INDO, TEA, Gli and atropine were used, which were added 20 min before the NA contracted rat thoracic aorta, and then RC and THP were added to induce vasodilatation. The results demonstrated that RC and THP induced vasodilatation in an endothelium dependent and endothelium independent manner. The endothelium dependent pathway was the result of activation of the NO/cGMP signaling pathway. The endothelium independent pathways were involved in the blockade of VDCCs and inhibition of Ca²⁺ mobilization from intracellular stores, as well as the stimulation of the muscarinic receptor. In addition, the K_ATP channel was activated in the vasorelaxant mechanism of THP.

---

Introduction

Arterial hypertension is considered as a serious health problem and represents a major stake of public health in both developed and developing countries. The current and common methods for controlling hypertension rely on the use of long-term therapy associating different drugs which include diuretics, angiotensin converting enzyme inhibitors, angiotensin II receptor blockers, β-blockers, calcium antagonists, α-adrenergic blockers and direct vasodilators.¹ Despite this, pharmaceutical companies have shown that natural product research remains to be a successful tool to produce novel candidates for the treatment of untreated diseases, since natural products represent privileged structures selected by evolutionary mechanisms over millions of years. In addition, according to the World Health Organization (WHO), about 65–80% of the world’s population in developing countries, due to poverty and lack of access to modern medicine, depend essentially on plants for their primary health care.^2–4 During the last 20 years, around 200 vasodilator compounds have been derived from plants. Despite their number and the structural diversity, it is clear that most compounds with vasodilator activity are alkaloids, terpenoids or flavonoids.⁵ Therefore, it is necessary to characterize compounds with vasorelaxant and antihypertensive activity derived from plants.

Vasodilators are useful for the treatment of cerebral vasospasm and hypertension, and also for the improvement of peripheral circulation. Recently, studies on traditional Chinese medicine and its active ingredients in the regulation of nitric oxide (NO) have become more and more popular. The mechanisms of the plants or the active ingredients derived from plants modulate NO action mean that they are promising candidates for vasodilatation, and may have potential to prevent and treat cardiovascular diseases.⁶ It has been reported that there are many other extracts of plants or active components that can exert the vasorelaxant effect involved in the NO-cGMP pathway.^7,8 Furthermore, the contractile response in smooth muscle is caused by Ca²⁺ influx through voltage-dependent Ca²⁺ channels (VDCCs) and/or receptor operated Ca²⁺ channels (ROCCs). The endothelium independent vasodilators, calcium channel antagonists, have been reported to inhibit VDCCs and lead to a decrease in the intracellular Ca²⁺ concentration in smooth muscle, which finally causes vasorelaxation.⁹ Furthermore, the opening or activation of K⁺ channels in vascular smooth muscle cells induces hyperpolarization of the membrane potential, which could lead to the dilatation of arteries.¹⁰

Rhizoma corydalis has been used for the treatment of various cardiovascular diseases in China. It is also a well-known traditional Chinese herb for the treatment of chest pain, epigastria pain, dysmenorrhea, traumatic swelling and pain.^11–13 Tetrahydropalmatine (THP, the structure shown in Fig. 1), one of its main active ingredients, has been demonstrated to have potent analgesic effects and has been used in Chinese clinical practice for many years.^14,15 It has various pharmacology effects including antihypertensive activity, protective effects towards human endothelial cells, anxiolytic properties, protection against methamphetamine-induced spatial learning, memory impairment and protection against myocardial ischaemia-reperfusion injury.^16–20 Dehydrocorydaline, another important ingredient, demonstrates a relaxant effect on thoracic aorta with intact or without endothelium.²¹ Although there is a lot of research about the cardiovascular protective effects of Rhizoma corydalis and THP, their vasodilatory effects have not yet been defined. Therefore, the aim of the present study was to investigate the possibility and to delineate the possible mechanisms of vasodilatation induced by RC and THP on isolated rat aorta.

Fig. 1 Chemical structure of THP, one of the major ingredients existing in RC.

Experimental

Reagents and materials

Rhizoma corydalis was donated by Tianjin Tasly Pharmaceutical Co., Ltd. (Tianjin, China). They were identified by Prof. Gao (Tianjin University, China). The voucher specimens (voucher no. YHS141102) were deposited at the School of Pharmaceutical Science and Technology at Tianjin University, Tianjin, China. Noradrenaline bitartrate injection (NA, 2 mg mL⁻¹) and atropine sulfate injection (0.5 mg mL⁻¹) were provided by the Affiliated Hospital of Logistics College of Chinese People’s Armed Police Forces. THP, tetrahydroberberine (THB, internal standard, IS), baicalein (BAI, IS), verapamil and acetylcholine were obtained from the National Institute for Control of Pharmaceutical and Biological Products (Beijing, China). Glibenclamide (Gli), N^G-nitro-L-arginine methyl ester (L-NAME), indomethacin (INDO), 1H-[1,2,4] oxadiazolo[4,3-a]-quinoxalin-1-one (ODQ), and tetraethylammonium chloride (TEA) were produced by Sigma-Aldrich Co. (St. Louis, MO, USA). Calcium chloride, glucose, magnesium chloride, potassium chloride, sodium bicarbonate, sodium dihydrogen phosphate, sodium chloride, sodium hydroxide and trichloroacetic acid were produced by Tianjin Fengchuan Chemical Reagent Science and Technology Co., Ltd. (Tianjin, China). All chemicals used were of analytical grade and solubilized in distilled H₂O/saline. The vehicles used had no effect in the control experiments.

Rhizoma corydalis (500 g) was powdered and extracted twice with 5 L of 70% ethanol, for 2 h each time. The filtrate was collected and then the solvent was removed under reduced pressure in a rotary evaporator (Buchi B-480) and RC was obtained with a yield of 15.3% (w/w).

Animals

Male Wistar rats, Wistar Kyoto rats (WKY) and spontaneous hypertensive rats (SHR) weighing about 280–330 g used in the experiments were purchased from the Experimental Animal Center, Chinese Academy of Medical Sciences, Peking, SCXK(Jing)-2012-0001. The animal study proposal was approved by the Institutional Animal Care and Use Committee (IACUC) of Institute of Radiation Medicine Chinese Academy of Medical Sciences (Tianjin, China) with the permit number IACUC2014-010. The Animal Ethics Committees of the Faculty of Medicine approved all experimental protocols in accordance with the Principles of Laboratory Animal Care and Use in Research (Ministry of Health, Beijing, China).

Preparation of rat thoracic aortic rings

The male Wistar rats had free access to water but food was withdrawn 24 h before the experiment. After the sacrifice of the animals, thoracic aortic ring segments (4–5 mm) were mounted for tension recording (1 g) in 10 mL organ baths filled with Krebs solution (composition in mM: 115.3 mM NaCl, 4.9 mM KCl, 1.46 mM CaCl₂·2H₂O, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄, 11.1 mM D-glucose and 25 mM NaHCO₃, pH = 7.4), thermoregulated at 37 °C and aerated with carbogen.¹⁶ In some rings, the endothelium was removed by gently rubbing the intimal surface with a cotton bud. The rings were allowed to equilibrate for at least 90 min before the addition of any drug. During this period, the rings were rinsed with fresh solution every 20 min. After equilibrating for 90 min, all aortic rings obtained a maximal response by contracting twice with KCl (60 mM). After restoration of the vessel tension to baseline levels, the rings were exposed to NA (1 × 10⁻⁶ M) to test their contractile responses. The functional endothelium and smooth muscle relaxation capacities of the arteries were checked by acetylcholine (1 × 10⁻⁵ M). Thus, the endothelium was considered intact when 80% or more than 80% relaxation was achieved by acetylcholine in the aorta rings pre-contracted using NA. When the endothelium was fully removed, <1% relaxation in response to acetylcholine could be recorded.¹⁷ The responses of the thoracic aorta were recorded isotonically using a RM6240BD type multichannel physiological signal acquisition system.

Quantitative analysis of THP in RC by HPLC-DAD

5.42 mg THP was accurately weighed and dissolved in 5 mL of methanol in a volumetric flask as a standard solution. 1.0 g RC was accurately weighed and dissolved in 10 mL of methanol in a volumetric flask. The supernatant was filtered through a 0.45 μm syringe filter before HPLC analysis. The THP content was determined using a standard curve. An Agilent 1100 liquid chromatography system (Agilent Technologies, USA) was equipped with a quaternary pump, an online degasser, a column temperature controller and a DAD detector. The analytical column temperature was kept at 35 °C. The samples were separated with a Kromasil C18 column (4.6 mm × 250 mm I.D., 5 μm), using 2% aqueous acetic acid adjusted with triethylamine to pH 5.0 (A) and acetonitrile (B) (4 : 6). The detection wavelength was 280 nm. The injection volume was 20 μL and the flow rate was 1 mL min⁻¹.

Animal treatment and acute antihypertensive activity of RC and THP

Eight WKY rats were orally administrated with sodium carboxymethylcellulose (CMC-Na). 48 SHRs were randomly divided into six groups, with eight in each group, and orally administrated with CMC-Na, valsartan (10 mg kg⁻¹), THP (20, 40 and 80 mg kg⁻¹) and RC (8.5 g kg⁻¹) for six weeks, respectively. Valsartan, RC and THP were suspended in CMC-Na.

In vivo acute antihypertensive activity measurements of RC and THP were performed on SHRs by a tail cuff method, using the pressure transducer and a data acquisition system (Chengdu Taimeng Software Co., Ltd. China). Before the measurements, conscious rats were restrained for 10 to 15 min in a warm chamber in a quiet room and conditioned to numerous cuff inflation–deflation cycles by a trained operator. Measurements (systolic blood pressure: SBP and diastolic blood pressure: DBP) were recorded before and after RC or THP treatment at 0, 1, 2, 4 and 6 h by the tail cuff method, three times each. An average for every trio was calculated for each time. These averages were used for the graphics and tables.

Effects of RC and THP on NA-induced tonic contractions in endothelium-intact and -denuded rat aortic rings. The vasorelaxant effects of RC and THP were investigated on both endothelium-intact and endothelium-denuded aortic rings. After the rings were pre-equilibrated, they were pre-contracted with NA (1 × 10⁻⁶ M) until stability of the tension was established, which was followed by cumulative exposure to RC (0.05, 0.1, 0.2, 0.4, 0.8 and 1.6 mg mL⁻¹) and THP (18.8, 37.6, 75.2, 150.4, 300.8 and 601.6 μg mL⁻¹).²²

Role of endothelium in RC and THP-induced vascular response. To determine whether the endothelium was involved in the relaxant effects of RC and THP, endothelium-intact rings were incubated with L-NAME (1 × 10⁻⁴ M, a NO synthase (NOS) inhibitor), ODQ (1 × 10⁻⁶ M, a selective blocker of sGC) and INDO (1 × 10⁻⁵ M), a cyclooxygenase (COX) inhibitor, for 20 min prior to being pre-contracted with NA. The cumulative concentration-response curves of RC (from 0.05 to 1.6 mg mL⁻¹) and THP (from 18.8 to 601.6 μg mL⁻¹) were then constructed. Non-incubation with the inhibitors was considered as the control group.²³

Role of K⁺ channels in RC and THP-induced vasorelaxation. The role of the K⁺ channels was investigated by the vasorelaxation response upon pre-incubating the endothelium-denuded aortic rings with one of the following specific K⁺ channel blockers: TEA (1 × 10⁻³ M) and Gli (1 × 10⁻⁵ M) for 20 min before NA was added. RC and THP were added cumulatively once the plateau was attained. Non-incubation with the inhibitors was considered as the control group.

Effects of RC and THP on extracellular Ca²⁺-induced contraction activated by KCl. To investigate whether the inhibitory effects of RC (0.1, 0.4, 0.8 and 1.6 mg mL⁻¹) and THP (37.6, 75.2, 150.4 and 300.8 μg mL⁻¹) were involved in inhibiting Ca²⁺ influx through VDCCs, endothelium-denuded aortic rings were exposed to a Ca²⁺-free solution containing EDTA (1 × 10⁻⁴ M) for 30 min in order to remove Ca²⁺ from the tissues. The rings were then rinsed in Ca²⁺-free and high-K⁺ (60 mM) Krebs solution (without EDTA). The cumulative concentration-response curves for CaCl₂ (0.03 to 30 mM) were obtained either in the absence of RC or THP (control group) or after 20 min incubation with RC and THP.

Effects of RC and THP on the sarcoplasmic reticulum calcium release induced by NA. To determine whether the relaxation induced by RC and THP was related to the inhibition of the release of intracellular Ca²⁺, the effects of RC and THP on NA-induced contractions in a Ca²⁺-free medium were investigated. After the usual stabilization time, endothelium-denuded aortic rings were washed with Ca²⁺-free solutions for 20 min. Then, THP was used to pre-incubate the aortic rings for 20 min before NA was added to elicit the release of intracellular Ca²⁺. The maximal tension induced by NA in the control group (without RC or THP) was considered as 100%.²²

Role of muscarinic receptor in RC and THP-induced vascular response. To investigate whether RC and THP exerted vasodilatation by the activation of the muscarinic receptor, endothelium-intact aortic rings were incubated with atropine (a muscarinic receptor antagonist, 1 × 10⁻⁶ M) for 20 min prior to being pre-contracted with NA. RC and THP were added once the plateau was attained. Non-incubation with the muscarinic receptor antagonist was considered as the control group.

Pharmacokinetic study of THP after administration of RC and THP

After overnight fasting (with free access to water), 24 rats were randomly divided into four groups (THP 20, 40, 80 mg kg⁻¹ and RC 8.5 g kg⁻¹) for the pharmacokinetic study of THP. From each group, serial blood samples were collected at 0.083, 0.167, 0.25, 0.5, 1, 1.5, 2.5, 4, 6, 8, 10, 12 and 24 h after a single oral dose of RC or THP. Serial blood samples (0.5 mL each) were collected from the veins of the eye socket into heparinized centrifuge tubes after the study of drug administration. The blood samples were immediately centrifuged at 6000 rpm for 10 min. The plasma was transferred to new tubes, immediately frozen and then stored at −20 °C until analysis.

Rat plasma sample (200 μL) and 10 μL of IS solution (THB 30.5 μg mL⁻¹ is for pure THP or BAI 35.0 μg mL⁻¹ is for RC) were added to a 1.5 mL plastic tube and vortexed for 30 s, followed by the addition of 600 μL of acetonitrile. The mixture was then vortexed for 1 min and centrifuged at 10 000 rpm for 10 min. The supernatant was transferred into another plastic tube and evaporated to dryness under a stream of nitrogen in a water bath at 40 °C. Finally, the residue was dissolved in 100 μL of methanol with vortex-mixing for 60 s. After being centrifuged at 14 000 rpm for 10 min, a 50 μL aliquot of the THP sample or a 20 μL aliquot of the RC sample was injected into the HPLC for analysis. The HPLC condition for THP sample analysis was as follows: the HPLC apparatus was composed of (LC-20AT, Shimadzu, Japan) a diode array detector (SPD-M20A) and a column oven (CTO-20A). The detector wavelength was set at 280 nm, and the analytical column, a Phenomenex C18 column (4.6 mm × 250 mm I.D., 5 μm), was used at 35 °C. The mobile phase was composed of triethylamine aqueous solution (pH = 8.0) and acetonitrile in a ratio of 4 : 6. The flow rate was 1 mL min⁻¹. The HPLC condition for RC sample analysis was as follows: an Agilent 1100 liquid chromatography system (Agilent Technologies, USA) was equipped with a quaternary pump, an online degasser, a column temperature controller and a DAD detector. The analytical column temperature was kept at 35 °C. The samples were separated with a Phenomenex C18 column (4.6 mm × 250 mm I.D., 5 μm), using 2% aqueous acetic acid adjusted with triethylamine to pH 5.0 (A) and acetonitrile (B) under a gradient condition: 0–5 min 30–37% B, 5–10 min 37% B, 10–15 min 37–42% B. The detection wavelength was 280 nm. The flow rate was 1 mL min⁻¹. Good linearity was obtained from 0.52 to 268.70 μg mL⁻¹ for THP. The main pharmacokinetic parameters were calculated using 3p87 software (Chinese Pharmacological Society, Beijing, China).

Statistical analysis

The results are expressed as means ± standard errors (s.e.). One-way analysis of variance (ANOVA), Dunnett’s test or Fisher’s Protected LSD multiple comparison test are used as appropriate. Tests were performed using an SPSS 20.0 system (Chicago, IL); a p value of less than or equal to 0.05 was considered to be statistically significant.

Results

Quantitative analysis of THP in RC by HPLC-DAD

The content of THP was finally quantified using corresponding calibration curves y = 16 095x + 129 969 with good correlation coefficients (r² = 0.9999). The content of THP in RC was 4.7 mg g⁻¹. The representative chromatograms are shown in Fig. 2.

Fig. 2 Representative HPLC chromatograms of (A) THP standard (purity >99%) and (B) RC detected at a wavelength of 280 nm. The content of THP in RC was 4.7 mg g⁻¹.

Acute antihypertensive activity of RC and THP

In WKY, the baseline values of SBP and DBP were 130.36 ± 8.81 mmHg and 109.97 ± 6.86 mmHg, respectively. In the model group, the baseline values of SBP and DBP were 183.44 ± 9.35 mmHg and 151.21 ± 11.63 mmHg, respectively. Gastric gavage of THP 80 mg kg⁻¹ could induce a stable decrease in SBP and DBP for the SHRs (Table 1), which was similar to the valsartan group. In addition, there was a significant decrease of SBP and DBP compared to the model group. Meanwhile, SBP did not show an obvious decline, but DBP decreased significantly after the administration of 8.5 g kg⁻¹ RC to the rats (Table 1).

Table 1 Time-course of systolic blood pressure (SBP) and diastolic blood pressure (DBP) in SHRs before the treatment (0 hour) and after the treatment of RC or THP. SBP and DBP values were measured at 0, 1, 2, 4 and 6 h^a

BP (mmHg)	Time (h)	WKY	Model	VAL	THP (mg kg^−1)			RC (8.5 g kg^−1)
					20	40	80
a Symbols and vertical bars represent means and S.E.M. ANOVA followed by Dunnett’s multiple comparison test. *p < 0.05, **p < 0.01, ***p < 0.001.
SBP	0	130.36 ± 8.81	183.44 ± 9.35	177.27 ± 7.19	181.22 ± 15.56	180.99 ± 10.44	171.97 ± 12.78	179.00 ± 12.38
	1	135.91 ± 12.19	177.84 ± 7.96	154.30 ± 12.30**	172.33 ± 14.22	170.83 ± 15.75	154.88 ± 16.36**	180.64 ± 14.69
	2	133.31 ± 11.35	180.81 ± 8.41	144.86 ± 5.43***	170.25 ± 13.34	164.55 ± 10.98*	148.03 ± 15.54***	175.01 ± 8.61
	4	128.55 ± 19.36	190.08 ± 9.78	148.38 ± 13.62***	165.44 ± 9.98**	157.27 ± 19.73***	139.17 ± 6.86***	171.88 ± 26.26
	6	132.03 ± 12.93	171.60 ± 9.92	145.67 ± 15.44***	167.58 ± 8.97	155.24 ± 10.67*	142.34 ± 7.32***	166.32 ± 5.07
DBP	0	109.97 ± 6.86	151.21 ± 11.63	149.23 ± 5.33	152.42 ± 12.78	150.72 ± 13.29	144.97 ± 17.39	150.33 ± 10.60
	1	112.17 ± 7.19	161.41 ± 7.23	127.79 ± 19.81**	145.33 ± 18.44*	146.85 ± 12.24*	126.02 ± 23.15**	152.67 ± 11.61
	2	101.30 ± 9.78	159.96 ± 6.76	118.67 ± 6.79***	143.59 ± 13.79*	137.55 ± 9.77*	125.10 ± 14.04*	148.76 ± 10.50*
	4	103.88 ± 9.35	173.67 ± 10.10	118.83 ± 12.01***	135.46 ± 17.21**	130.76 ± 8.91***	120.85 ± 5.85***	138.05 ± 17.68**
	6	110.36 ± 12.78	144.28 ± 8.21	120.18 ± 17.58*	134.53 ± 7.13	124.69 ± 10.23**	111.95 ± 8.68***	132.46 ± 4.31

---

In vitro experiments: studies on isolated rat thoracic aorta

Effects of RC and THP on NA-induced tonic contractions in endothelium-intact and -denuded rat aortic rings

In the endothelium-intact aortic ring preparations, the vascular tension induced by NA was markedly reduced by RC and THP in concentration-dependent manners with maximum values of 100.5 ± 7.0% and 99.0 ± 7.0% (n = 5) at the concentrations of 1.6 mg mL⁻¹ and 601.6 μg mL⁻¹ (EC₅₀ = 0.26 mg mL⁻¹ and 110.7 μg mL⁻¹), respectively. While in the endothelium-denuded aortic rings, the vascular tension was reduced by RC and THP also in concentration-dependent manners with maximum values of 81.4 ± 13.0% and 76.1 ± 9.1% (n = 5) at the concentrations of 1.6 mg mL⁻¹ and 601.6 μg mL⁻¹ (EC₅₀ = 0.76 mg mL⁻¹ and 270.3 μg mL⁻¹), respectively, indicating the occurrence of endothelium independent relaxation. The vasorelaxant effects of RC and THP in the endothelium-intact vasculature were more potent than those in the endothelium-denuded vasculature (Fig. 3A–D).

Fig. 3 Influence of different concentrations of (A) RC and (B) THP on NA pre-incubated rat thoracic aorta rings with and without endothelium (endothelium-intact E+, endothelium-denuded E−). Once a plateau was attained, concentration-response curves of RC (0.05, 0.1, 0.2, 0.4, 0.8 and 1.6 mg mL⁻¹) and THP (18.8, 37.6, 75.2, 150.4, 300.8 and 601.6 μg mL⁻¹) induced relaxation were obtained and changes in tension were recorded for 5 min at each concentration. Representative tracings of the relaxation of aortas treated with NA (1 × 10⁻⁶ M) in response to (C) RC and (D) THP. The symbols and vertical bars represent means and S.E.M. ANOVA followed by Dunnett’s multiple comparison test were used. *p < 0.05, **p < 0.01, ***p < 0.001 (E+ vs. E−) (n = 5).

Role of endothelium in RC and THP-induced vascular response

Considering the endothelium dependent effects of RC and THP, we investigated the involvement of the endothelium in RC and THP-induced vasorelaxation in NA-constricted rings. Experiments were performed on aorta after pre-treatment with INDO, L-NAME and ODQ. As illustrated in Fig. 4A–C, RC-induced endothelium dependent relaxation was significantly reduced by pretreatment with ODQ and L-NAME in concentration-dependent manners with maximum values of 64.7 ± 3.0%, 65.0 ± 10.0% (n = 5) at a concentration of 1.6 mg mL⁻¹ (EC₅₀ = 1.229 and 0.884 mg mL⁻¹, p < 0.001). However, pre-incubation with INDO did not show any influence on the vasorelaxation of RC (Fig. 4A and D, p > 0.05).

Fig. 4 Effect of different inhibitors (A) INDO (1 × 10⁻⁵ M, a COX inhibitor), (B) L-NAME (1 × 10⁻⁴ M, a NO synthase inhibitor) and (C) ODQ (1 × 10⁻⁶ M, a selective blocker of sGC) on RC-induced vasodilatation in endothelium-intact rat thoracic aortic rings. The inhibitors were added 20 min before NA-induced contraction and then the added RC accumulated to induce vasodilatation every 5 min. Representative force traces which show the vascular relaxation induced by RC in endothelium-intact aorta rings pre-incubated with (D) INDO, (E) L-NAME and (F) ODQ. The symbols and vertical bars represent means and S.E.M. ANOVA followed by Dunnett’s multiple comparison test were used. *p < 0.05, **p < 0.01, ***p < 0.001 (control vs. INDO/ODQ/L-NAME) (n = 5).

THP-induced endothelium dependent relaxation was significantly reduced by pretreatment with ODQ and L-NAME in concentration-dependent manners with maximum values of 69.1 ± 9.1% and 43.0 ± 9.1% (n = 5) at a concentration of 601.6 μg mL⁻¹ (EC₅₀ = 1.360 and 597.7 μg mL⁻¹, p < 0.001). However, pretreatment with INDO did not show any obvious influence on the vasorelaxation of THP (Fig. 5C, p > 0.05).

Fig. 5 Effect of different inhibitors (A) ODQ (1 × 10⁻⁶ M, a selective blocker of sGC), (B) L-NAME (1 × 10⁻⁴ M, a NO synthase inhibitor) and (C) INDO (1 × 10⁻⁵ M, a COX inhibitor) on THP-induced vasodilatation in endothelium-intact rat thoracic aortic rings. The inhibitors were added 20 min before NA-induced contraction and then the added THP accumulated to induce vasodilatation every 5 min. The symbols and vertical bars represent means and S.E.M. ANOVA followed by Dunnett’s multiple comparison test were used. *p < 0.05, **p < 0.01, ***p < 0.001 (control vs. INDO/ODQ/L-NAME) (n = 5).

Role of K⁺ channels in RC and THP-induced relaxation

To assess whether RC and THP-induced vasodilatation was mediated by the activation of K⁺ channels, aortic rings were incubated with different K⁺ channel inhibitors, including Gli and TEA. As illustrated in Fig. 6A, the pretreatment of endothelium-denuded aortic rings with the K_ATP blocker Gli could significantly attenuate the THP-induced vasorelaxation. However, the K_Ca blocker TEA could not abolish THP-induced vasorelaxation (Fig. 6B). As for RC, neither of the blockers modified the vasorelaxant effect induced by RC (Fig. 6C and D).

Fig. 6 Effect of the pretreatment with different K⁺ channel blockers (A, C) Gli (1 × 10⁻⁵ M, an ATP-sensitive K⁺ channel (K_ATP) blocker) and (B, D) TEA (1 × 10⁻³ M, a non-selective blocker of K_Ca channels) on RC and THP-induced vasodilatation in endothelium-denuded rat thoracic aortic rings. The inhibitors were added 20 min before NA-induced contraction and then the added RC and THP accumulated to induce vasodilatation every 5 min. The symbols and vertical bars represent means and S.E.M. ANOVA followed by Dunnett’s multiple comparison test were used. *p < 0.05, **p < 0.01, ***p < 0.001 (control vs. TEA/Gli) (n = 5).

Role of muscarinic receptor in RC and THP-induced vascular response

As shown in Fig. 7A, the relaxation induced by RC on endothelium-intact rat aortic rings pre-contracted with NA was significantly attenuated in the presence of the muscarinic receptor antagonist atropine (p < 0.05). Meanwhile, the relaxation induced by THP (150.4, 300.8 and 601.6 μg mL⁻¹) on the endothelium-intact rat aortic rings pre-contracted with NA was attenuated in the presence of the muscarinic receptor antagonist atropine (p < 0.05, Fig. 7B).

Fig. 7 Vasorelaxant effects of (A) RC and (B) THP on endothelium-intact aortic rings incubated with atropine (a muscarinic receptor antagonist, 1 × 10⁻⁶ M) prior to their pre-constriction with NA. Endothelium-intact aortic rings were incubated with atropine for 20 min prior to pre-contraction with NA. Once a plateau was attained, RC and THP were added respectively. Non-incubation with the muscarinic receptor antagonist was considered as the control group. The symbols and vertical bars represent means and S.E.M. ANOVA followed by Dunnett’s multiple comparison test were used. Compared with control *p < 0.05, **p < 0.01, ***p < 0.001 (n = 5).

Effects of RC and THP on extracellular Ca²⁺-induced contraction activated by KCl

In order to examine whether the vasorelaxant effects of RC and THP were mediated through the blockade of Ca²⁺ influx, a high dose of K⁺ (60 mM) was used to depolarize the tissue. Verapamil (1 × 10⁻⁶ M), used as a positive control, could induce vasorelaxation (Fig. 8). Like verapamil, a well-known calcium antagonist, pretreatment with RC (0.1, 0.4, 0.8 and 1.6 mg mL⁻¹) or THP (37.6, 150.4, 300.8 and 601.6 μg mL⁻¹) for 20 min produced a rightward parallel displacement of the CaCl₂ curves and reduced the maximum contraction induced by 30 mM CaCl₂ to 75.6 ± 4.7%, 80.8 ± 4.8%, 60.5 ± 6.6% and 41.6 ± 4.4% of RC and 76.8 ± 2.2%, 62.0 ± 3.0%, 55.6 ± 4.1% and 43.6 ± 3.2% of THP, respectively (Fig. 8A–D), while verapamil reduced the contraction to 42.6 ± 1.7%. The CaCl₂ curves of THP at a concentration of 300.8 μg mL⁻¹ almost overlap with the curves of verapamil, while the effect of RC at a concentration of 1.6 mg mL⁻¹ was stronger than that of verapamil, suggesting that Ca²⁺ influx was reduced by RC and THP.

Fig. 8 Dose-effect curves of CaCl₂ on rat thoracic aorta rings in the absence and in the presence of (A) RC and (B) THP (endothelium-denuded E−). The cumulative concentration-response curves for CaCl₂ (0.03 to 30 mM) were obtained in the absence of RC and THP (control group) or after 20 min of incubation with RC (0.1, 0.4, 0.8 and 1.6 mg mL⁻¹) and THP (37.6, 150.4, 300.8 and 601.6 μg mL⁻¹). Finally, the contractile effect induced by CaCl₂ was compared in the absence (control group) and presence of RC and THP. Non-incubation with RC and THP were considered as the control group. Representative tracings of the contraction of the aortas treated with CaCl₂ in response to (C) a control and (D) RC 0.4 mg mL⁻¹. The symbols and vertical bars represent means and S.E.M. ANOVA followed by Dunnett’s multiple comparison test were used. Compared with control *p < 0.05, **p < 0.01, ***p < 0.001 (n = 5).

Effects of RC and THP on the sarcoplasmic reticulum calcium release induced by NA

As shown in Fig. 9, RC and THP dramatically reduced NA-induced contraction in a calcium free buffer. Pre-incubation with RC for 20 min significantly attenuated the NA-induced contraction to 81.1 ± 3.3%, 78.9 ± 6.7%, 75.0 ± 8.5%, 68.6 ± 6.0%, 41.5 ± 3.5% and 15.2 ± 5.3%, respectively (n = 5) (Fig. 9A). Similarly, THP attenuated the NA-induced contraction to 100 ± 0.6%, 90.6 ± 1.1%, 83.4 ± 0.7%, 73.3 ± 1.7%, 56.8 ± 3.4% and 40.5 ± 0.8%, respectively (n = 5) (Fig. 9B), suggesting that RC and THP also exerted their vasorelaxant action by interfering with the calcium released from the sarcoplasmic reticulum.

Fig. 9 Vasorelaxant effects of (A) RC and (B) THP on NA (1 × 10⁻⁶ M) pre-constricted aortic rings (endothelium-denuded) in Ca²⁺-free solution. Endothelium-denuded rings were allowed to stabilize in a Ca²⁺-free Krebs solution for 20 min. Then, RC and THP were used to pre-incubate the aortic rings for 20 min before NA (1 × 10⁻⁶ M) was added to stimulate the release of intracellular Ca²⁺. The maximal tension induced by NA in the control group (without RC and THP) was considered as 100%. The symbols and vertical bars represent means and S.E.M. ANOVA followed by Dunnett’s multiple comparison test were used. Compared with control *p < 0.05, **p < 0.01, ***p < 0.001 (n = 5).

Pharmacokinetic study of THP

Plasma concentrations of THP were determined by a sample and accurate analysis method after oral administration of RC or THP. As the results shown in Fig. 10, under the optimized conditions, all of the analytes were separated with good resolution. The plasma concentration–time profiles of THP in rat plasma are illustrated in Fig. 11 and the pharmacokinetic parameters are summarized in Table 2. As shown in Fig. 11 and Table 2, after a single oral dose of THP to rats, the C_(max) of THP in 20, 40 and 80 mg kg⁻¹ were 7.417 ± 1.007, 14.331 ± 1.401 and 22.520 ± 1.975 μg mL⁻¹, respectively. The plasma concentration of THP increased rapidly in the first half hour after drug administration. However, the C_(max) was 1.742 ± 0.253 μg mL⁻¹ after a single oral dose of RC 8.5 g kg⁻¹ (equal to 40 mg kg⁻¹ THP), which was significantly lower than that of 40 mg kg⁻¹ THP (C_(max) = 14.331 ± 1.401 μg mL⁻¹).

Fig. 10 HPLC chromatograms of (A, D) blank plasma, (B, E) blank plasma spiked with a standard and IS, (C, G) rat plasma after oral administration of THP and RC and (F) RC sample. A, B and C are the results of THP and D, E, F and G are the results of RC.

Fig. 11 Plasma concentrations of THP in rats after oral administration of 20, 40, 80 mg kg⁻¹ THP and 8.5 g kg⁻¹ RC.

Table 2 Pharmacokinetic parameters of THP after oral administration of RC or THP. (n = 6, mean ± SD)

Parameters	RC (8.5 g kg^−1)	THP (mg kg^−1)
		20	40	80
Ke (h^−1)	0.135 ± 0.0314	0.175 ± 0.0275	0.0867 ± 0.0362	0.0687 ± 0.0465
Ka (h^−1)	5.658 ± 1.147	5.985 ± 1.213	6.998 ± 0.798	5.895 ± 1.141
t1/2(Ka) (h)	0.122 ± 0.0256	0.116 ± 0.0474	0.0990 ± 0.0732	0.118 ± 0.0361
t1/2(Ke) (h)	5.119 ± 0.931	3.954 ± 0.236	7.996 ± 1.458	10.085 ± 3.731
T(max) (h)	0.676 ± 0.0222	0.608 ± 0.0201	0.635 ± 0.0234	0.765 ± 0.0341
C(max) (μg mL^−1)	1.742 ± 0.253	7.417 ± 1.007	14.331 ± 1.401	22.520 ± 1.975
AUC (μg mL^−1 h^−1)	14.100 ± 0.327	40.719 ± 8.291	138.116 ± 16.213	267.117 ± 14.352
CL/F(s) (mg h^−1 μg^−1 mL^−1)	0.071 ± 0.0169	0.491 ± 0.0942	0.290 ± 0.0593	0.299 ± 0.0609
V/F(c) (mg μg^−1 mL^−1)	0.524 ± 0.0821	2.802 ± 0.815	3.341 ± 0.533	4.358 ± 0.697

---

Discussion

To the best of our knowledge, this is the first study of the vasorelaxant effects induced by RC and THP. The present study showed that RC and THP, at concentrations of 0.05 to 1.6 mg mL⁻¹ and 18.8 to 601.6 μg mL⁻¹, could relax isolated rat aortic rings treated by NA. The vasorelaxant effects of RC and THP could be related to the endothelium, since vasorelaxation was blocked by L-NAME (a NOS inhibitor) and ODQ (a selective blocker of sGC). RC and THP also induced relaxation in rat aortic rings in an endothelium independent manner. The endothelium dependent pathway was the result of activation of the NO/cGMP signaling pathway. The endothelium independent pathways were involved in the blockade of VDCCs and the inhibition of Ca²⁺ mobilization from intracellular stores, as well as the stimulation of the muscarinic receptor. In addition, the K_ATP channel was activated in the vasorelaxant mechanism of THP. The mechanisms and pathways responsible for RC and THP-induced vasorelaxation are depicted in Fig. 12. Until now, the vasorelaxant effects and the underlying mechanisms of RC and THP on thoracic aorta have not been well studied.

Fig. 12 Schematic mechanisms of the vasorelaxant effects of RC and THP on isolated rat thoracic aorta. VSMC, vascular smooth muscle cell; sGC, soluble guanylyl cyclase; cGMP, cyclic guanosine monophosphate; GTP, guanosine triphosphate; K_Ca, calcium-activated K⁺ channels; K_ATP, ATP-sensitive K⁺ channels; VDCCs, voltage-dependent Ca²⁺ channels; ROCCs, receptor-operated Ca²⁺ channels.

Rhizoma corydalis is a well-known traditional Chinese herb, usually used for the treatment of chest pain, epigastria pain, dysmenorrhea, traumatic swelling and pain. It has also been used for the treatment of various cardiovascular diseases in China. Recent studies reported that Rhizoma corydalis extracts exhibited several biological activities, including anti-allergic, analgesic, anti-amnestic and cardioprotective effects.^24–27 Alkaloids including THP, corydaline, protopine et al., are acknowledged to be the major active components in Rhizoma corydalis.²⁸ THP, one of its main active ingredients, has been reported to have a potent analgesic effect and has been used in Chinese clinical practice for many years.^15,16 It has various pharmacology effects, including antihypertensive activity, a protective effect to human endothelial cells, anxiolytic properties, protection against methamphetamine-induced spatial learning and memory impairment and protection against myocardial ischaemia-reperfusion injury.^16–20 The dosages of THP often used were 10, 20, 40 and 60 mg kg⁻¹ in a lot of pharmacological studies by gavage, which ensures that THP could reach an effective plasma concentration.²⁰ After oral administration of a single dose of 60 mg kg⁻¹ THP, the maximum plasma concentrations were 6.15 ± 2.1 and 7.54 ± 2.9 μg mL⁻¹ for normotensive rats and SHR.²⁹ The present study displayed that the C_(max) was 22.520 ± 1.975 μg mL⁻¹ after a single oral dose of 80 mg kg⁻¹ THP to rats, which was indeed not less than 53 μM (equal to 18.815 μg mL⁻¹), the lowest concentration of THP in the in vitro vasorelaxant experiments. However, the C_(max) was 1.742 ± 0.253 μg mL⁻¹ after a single oral dose of 8.5 g kg⁻¹ RC, which was significantly lower than that of 40 mg kg⁻¹ THP (C_(max) = 14.331 ± 1.401 μg mL⁻¹). Therefore, we proved that the administration of RC markedly decreased the plasma concentration of THP in comparison with a single oral dose of THP itself. According to the result that the antihypertensive effect of RC and the plasma concentration of THP had a positive correlation, the result of the decline of plasma concentration of THP after oral administration of RC comparing with pure THP might result from the influence by ingredients from RC. We considered that compounds existing in RC could decrease the antihypertensive effect of RC, which was related to the decrease of plasma concentration of THP. That was why 40 mg kg⁻¹ THP had a hypotensive effect, but 8.5 g kg⁻¹ RC did not show an antihypertensive effect. The present study also showed that 80 mg kg⁻¹ THP induced a stable decrease in SBP (Table 1) for the SHRs, which was similar to valsartan (an angiotensin II type 1 receptor antagonist). Interestingly, the antihypertensive effect of THP was stronger than that of RC in vivo, but the opposite effect of vasodilation was observed in vitro. 0.05 mg mL⁻¹ RC (containing 2.35 μg mL⁻¹ THP) could induce the vascular dilation, while 2.35 μg mL⁻¹ THP did not show vasodilatation. Therefore, we increased the concentration of THP to 18.8 μg mL⁻¹ (equal to the THP in 0.4 mg mL⁻¹ RC), which could relax the thoracic aorta. The reason why there was such a phenomenon might be related to the complexity of the chemical composition in RC. We hypothesize that one or more elements in RC may play a hypotensive or hypertensive role on blood pressure, which may be the synergistic or antagonistic effect of some chemical constituents and therefore make the antihypertensive effect of THP more potent than that of RC. There is a lot of research about the pharmacological effects of alkaloids in RC, but little is related to the hypotensive effect. Thus, further studies are needed to investigate the chemical compounds that are responsible for these pharmacological observations. In addition, in vivo, hypertension is closely regulated through interactions between the renin angiotensin system, the central nervous system and the peripheral nervous system, which are involved in most of the physiological and pathophysiological processes in hypertension. Besides, disorders of blood pressure are common and cause a major impact on lifestyle and environment. Therefore, we believe that hypertension is the result of integration of multiple factors in vivo.

Endothelial cells can synthesize and release a broad spectrum of vasoactive substances, which play a central role in the regulation of vascular function by controlling vascular tone. Among the main endothelial-derived factors that relax arterial smooth muscle are NO and the PGI₂ mechanism.^30,31 The PGI₂/cAMP pathway is involved in the mechanism used by a low percentage (23%) of vasodilating compounds.³² PGI₂ is an endogenous vasoactive eicosanoid produced by COX from arachidonic acid in endothelial cells. It can cause the relaxation of vascular smooth muscle through stimulation of a G-protein-coupled receptor that, in turn, activates adenylyl cyclase (AC) and thus raises cAMP levels, inducing vasodilation as a result.³³ Participation of PGI₂ is determined by using INDO as an inhibitor of the COX enzyme in our study. As illustrated in the present results, pretreatment with INDO did not show an obvious influence on the vasorelaxation of THP (Fig. 7C), suggesting that the PGI₂ pathway was not involved in the mechanism of action of RC and THP. NO is the principal mediator of endothelium dependent relaxation in vascular smooth muscles. Besides, NO is the main endothelium-derived relaxing factor (EDRF) produced in vascular smooth muscles under activation of eNOS.³⁴ A high percentage (98.4%) is involved the NO/cGMP pathway of vasodilating compounds.³² NO mediates its biological effects by activating sGC and elevates intracellular cGMP synthesis from GTP. The mechanisms of natural products and plant extracts induced vasorelaxation are in close relation to the NO system.⁸ Therefore, the NO/cGMP pathway is involved in this process through experiments using L-NMMA, an inhibitor of eNOS and ODQ, an inhibitor of sGC. The present study showed that the endothelial denudation of aorta rings or the inhibition of NOS greatly attenuated RC and THP-induced vasorelaxation, providing evidence that the activation of endothelial cells and the NO pathway were involved in the endothelium dependent relaxation of the vessels. ODQ could inhibit RC and THP-induced vasorelaxation of thoracic aorta rings. These findings clearly revealed that the vascular relaxation induced by RC and THP was mediated by the endothelium system via the endothelium dependent NOS-NO-cGMP signaling pathway.

As K⁺ channels play determinant roles in regulating vascular tone, we studied their possible involvement in RC and THP-induced vasorelaxation. Previous reports have identified four types of K⁺ channel expressed in vascular smooth muscle cells, including the Ca²⁺-activated K⁺ channel (K_Ca), voltage-dependent K⁺ channel (K_V), ATP-sensitive K⁺ channel (K_ATP) and inward rectifier K⁺ channel (K_ir).³⁵ The outflow of K⁺ through these channels hyperpolarizes the membrane and thereby inhibits the entry of Ca²⁺. This process eventually results in the relaxation of blood vessels.³⁶ A lot of vascularly bioactive agents exert their vasorelaxant effects by opening or closing the K⁺ channels.³⁷ Our research has shown that the effect of RC on the smooth muscle cell membrane does not seem to be due to the opening of K⁺ channels, since the vasorelaxant effects of RC were unaffected significantly by pretreatment with K⁺ channel blockers. However, the K_ATP channel appears to be a molecular target for the relaxant activity of THP. The results in the present study demonstrated that THP-induced relaxation of the aortic ring with the endothelium was significantly attenuated by Gli, an ATP-sensitive K⁺ channels inhibitor, but not by TEA, a non-selective K_Ca channel blocker, suggesting that THP-induced relaxation is closely related to the activation of K_ATP channels. However, some other possible mechanisms unrelated to the K⁺ channels might be involved.

The mechanism of vascular smooth muscle contraction involves the participation of different signal transduction pathways, all of which converge to increase cytoplasmic Ca²⁺ concentrations. Sufficient intracellular activator Ca²⁺ is a prerequisite for vascular contraction.³⁸ The concentration of this cation increases both by extracellular Ca²⁺ entering through VDCCs and ROCCs, and by the release of Ca²⁺ from intracellular stores.³⁹ Therefore, the mechanisms of action associated with vasodilating agents that decrease the intracellular Ca²⁺ concentration involve blocking VDCCs and ROCCs or inhibiting the release of this cation from intracellular stores. Hence, different techniques were used to determine the involvement of VDCCs or the release of intracellular calcium in our present study. The present results displayed that the inhibitory action of RC and THP on VDCCs could be seen as a rightward shift in the dose-response curve for CaCl₂, like verapamil, a well known calcium antagonist. The contribution of Ca²⁺ released from intracellular stores was determined by incubating the tissue in a Krebs solution, free of Ca²⁺, to which NA was added to induce phasic contractions with calcium from the sarcoplasmic reticulum. The results indicated that RC and THP dramatically reduced the NA-induced contraction in a calcium free buffer. Taken together, RC and THP were shown to inhibit CaCl₂-induced vasoconstriction and NA-induced contractions in a calcium-free medium. These results indicated the involvement of VDCCs and inhibition of the outflow of intracellular Ca²⁺ in the vasodilating effect produced by RC and THP.

Conclusion

In conclusion, our data demonstrated endothelium dependent and endothelium independent vasorelaxant effects of RC and THP. The endothelium dependent pathway is the result of activation of the NO/cGMP signaling pathways. The endothelium independent pathway involves the blockade of VDCCs and inhibition of Ca²⁺ mobilization from intracellular stores, as well as the stimulation of the muscarinic receptor. In addition, the vasorelaxant mechanism of THP involves activation of the K_ATP channels.

Conflict of interest

We have no conflict of interest in this research.

Abbreviation

CMC-Na	Carboxymethylcellulose
COX	Cyclooxygenase
DBP	Diastolic blood pressure
Gli	Glibenclamide
INDO	Indomethacin
IP3	Inositol triphosphate
KATP	ATP-sensitive K^+ channels
KCa	Ca^2+-activated K^+ channels
Kir	Inward rectifier K^+ channels
KV	Voltage-dependent K^+ channels
L-NAME	N^G-Nitro-L-arginine methyl ester
NA	Noradrenaline bitartrate injection
NO	Nitric oxide
ODQ	1H-[1,2,4]oxadiazolo[4,3-a]-quinoxalin-1-one
ROCCs	Receptor-operated Ca^2+ channels
SBP	Systolic blood pressure
SHR	Spontaneous hypertensive rats
TEA	Tetraethylammonium chloride
THB	Tetrahydroberberine
THP	Tetrahydropalmatine
VDCCs	Voltage-dependent Ca^2+ channels
WKY	Wistar Kyoto rats

Acknowledgements

The work was supported by the China Postdoctoral Science Foundation, China (Grant No. 2014M562675).

References

1. B. A. Staffileno, J. Cardiovasc. Nurs., 2005, 20, 354–364.
2. A. L. Harvey, Curr. Opin. Chem. Biol., 2007, 11, 480–484.
3. R. E. Smith and M. Ashiya, Nat. Rev. Drug Discovery, 2007, 6, 597–598.
4. J. Vergara-Galicia, R. Ortiz-Andrade, J. Rivera-Leyva, P. Castillo-España, R. Villalobos-Molina, M. Ibarra-Barajas, I. Gallardo-Ortiz and S. Estrada-Soto, Fitoterapia, 2010, 81, 350–357.
5. A. Novakovic, M. Marinko, A. Vranic, G. Jankovic, P. Milojevic, I. Stojanovic, D. Nenezic, N. Ugresic, V. Kanjuh, Q. Yang and G. W. He, Eur. J. Pharmacol., 2015, 762, 306–312.
6. F. S. Monteiro, A. C. Silva, I. R. Martins, A. C. Correia, I. J. Basílio, M. F. Agra, J. Bhattacharyya and B. A. Silva, J. Ethnopharmacol., 2012, 141, 895–900.
7. O. Hernández-Abreu, P. Castillo-España, I. León-Rivera, M. Ibarra-Barajas, R. Villalobos-Molina, J. González-Christen, J. Vergara-Galicia and S. Estrada-Soto, Biochem. Pharmacol., 2009, 78, 54–61.
8. T. T. Wang, G. H. Zhou, J. H. Kho, Y. Y. Sun, J. F. Wen, D. G. Kang, H. S. Lee, K. W. Cho and S. N. Jin, J. Ethnopharmacol., 2013, 148, 655–663.
9. H. Matsuo, R. Okamoto, K. Zaima, Y. Hirasawa, I. S. Ismail, N. H. Lajis and H. Morita, Bioorg. Med. Chem., 2011, 19, 4075–4079.
10. J. Vergara-Galicia, R. Ortiz-Andrade, P. Castillo-España, M. Ibarra-Barajas, I. Gallardo--Ortiz, R. Villalobos-Molina and S. Estrada-Soto, Vasc. Pharmacol., 2008, 49, 26–31.
11. M. Kubo, H. Matsuda, K. Tokuoka, S. Ma and H. Shiomoto, Biol. Pharm. Bull., 1994, 17, 262–265.
12. H. Matsuda, H. Shiomoto, S. Naruto, K. Namba and M. Kubo, Planta Med., 1988, 54, 27–33.
13. J. Yamahara, Nippon Yakurigaku Zasshi, 1976, 72, 909–927.
14. H. Chu, G. Jin, E. Friedman and X. Zhen, Cell Molecular and Neurobiology, 2008, vol. 28, pp. 491–499.
15. X. Z. Zhu, Mem. Inst. Oswaldo Cruz, 1991, 86, 173–175.
16. M. T. Lin, F. Y. Chueh, M. T. Hsieh and C. F. Chen, Clinical Experiment of Pharmacology and Physiology, 1996, vol. 23, pp. 738–745.
17. J. Yu, B. K. Piao, Y. X. Pei, X. Qi and B. J. Hua, J. Life Sci., 2010, 87, 55–63.
18. H. Henkes, M. Franz, O. Kendall, J. Monroe, A. Legaspi, J. LeDoux, C. Haese, D. Williams, S. McCall, A. D. Johnson and T. E. Ceremuga, J. Am. Assoc. Nurse Anesth., 2011, 79, S75–S80.
19. Y. J. Chen, Y. L. Liu, Q. Zhong, Y. F. Yu, H. L. Su, H.A. Toque, Y. H. Dang, F. Chen, M. Xu and T. Chen, Neurosci. Bull., 2012, 28, 222–232.
20. Y. Han, W. Zhang, Y. Tang, W. L. Bai, F. Yang, L. Xie, X. Li, S. Zhou, S. Pan, Q. Chen, A. Ferro and Y. Ji, PLoS One, 2012, 7, e38627.
21. J. Yan, C. L. You, X. Y. Zhang and H. Wang, Chin. J. Pharmacol. Toxicol., 2001, 15, 326–329.
22. Z. Qu, J. Z. Zhang, W. Y. Gao, H. Chen, H. M. Guo, T. T. Wang, H. F. Li and C. X. Liu, J. Ethnopharmacol., 2014, 155, 572–579.
23. L. Tao, H. S. Hu and X. C. Shen, Phytomedicine, 2013, 20, 387–393.
24. S. Y. Saito, M. Tanaka, K. Matsunaga, Y. Li and Y. Ohizumi, Biol. Pharm. Bull., 2004, 27, 1270–1274.
25. Z. Z. Ma, W. Xu, N. H. Jensen, B. L. Roth, L. Y. Liu-Chen and D. Y. Lee, Molecules, 2008, 13, 2303–2312.
26. T. M. Hung, T. M. Ngoc, U. J. Youn, B. S. Min, M. Na, P. T. Thuong and K. Bae, Biol. Pharm. Bull., 2008, 31, 159–162.
27. C. P. Wen, L. M. Wu, H. Y. Ling and L. D. Li, J. Pharm. Pharmacol., 2007, 59, 1159–1165.
28. T. J. Li, S. Wang, X. S. Meng, Y. R. Bao, S. S. Guan, B. Liu, L. Chen, L. Wang and X. R. Ran, PLoS One, 2014, 9, e82499.
29. Z. Y. Hong, G. J. Cai, W. T. Ma, J. Wen, Y. F. Chai and G. R. Fan, Biomed. Chromatogr., 2012, 26, 749–753.
30. T. D. Giles, G. E. Sander, B. D. Nossaman and P. J. Kadowitz, J. Clin. Hypertens., 2012, 14, 198–204.
31. R. M. Palmer, A. G. Ferrige and S. Moncada, Nature, 1987, 327, 524–526.
32. J. Francisco, V. Luna, I. A. César, R. M. Alejandra, R. M. Isela and Z. S. Miguel Ángel, Molecules, 2013, 18, 5814–5857.
33. Y. Tanaka, F. Yamaki, K. Koike and L. Toro, Curr. Med. Chem.: Cardiovasc. Hematol. Agents, 2004, 2, 257–265.
34. A. M. Bruno, L. Ito, S. Claudener, Q. Mayara, F. Alana, A. G. Luiz, S. Kyria, H. Alexandre, M. S. Ana and S. Benildo, Arch. Biochem. Biophys., 2015, 565, 32–39.
35. M. T. Nelson and J. M. Quayle, Am. J. Physiol., 1995, 268, 799–822.
36. E. A. Ko, J. Han, I. D. Jung and W. S. Park, J. Smooth Muscle Res., 2008, 44, 65–81.
37. H. Morita, K. Zaima, I. Koga, A. Saito, H. Tamamoto, H. Okazaki, T. Kaneda, T. Hashimoto and Y. Asakawa, Bioorg. Med. Chem., 2011, 19, 4051–4056.
38. B. R. Wamhoff, D. K. Bowles and G. K. Owens, Circ. Res., 2006, 98, 868–878.
39. S. P. Wang, W. J. Zang, S. S. Kong, X. J. Yu, L. Sun and X. F. Zhao, Eur. J. Pharmacol., 2008, 579, 283–288.

---

Footnote

† These two authors contributed equally to this work.

---