Vasodilatory Effects of Combined Traditional Chinese Medicinal Herbs in Optimized Ratio

Abstract

Recently, a new syndromic disease combination theory of traditional Chinese medicine (TCM) for hypertensive treatment has been introduced. In the wake of this new concept, a new science-based TCM formula that counteracts various syndromes is needed. The objective of this study was to develop such a formula. Five of the most clinically prescribed TCM herbs that work on different syndromes, namely Gastrodia elata, Uncaria rhynchophylla, Pueraria thomsonii, Panax notoginseng, and Alisma orientale, were selected for this study. The fingerprints of these five herbs were analyzed by tri-step Fourier transform infrared spectroscopy. Three different solvents, 95% ethanol, 50% ethanol, and distilled water, were used for the maceration of the herbs and their vasodilatory effects were studied using in vitro precontracted aortic ring model. Among these, the 50% ethanolic extracts of G. elata (GE50) and A. orientale (AO50), and 95% ethanolic extracts of U. rhynchophylla (UR95), P. thomsonii (PT95), and P. notoginseng (PN95) were found to be the most effective for eliciting vasodilation. Thus, these five extracts were used for orthogonal stimulus–response compatibility group studies by using L₂₅ (5⁵) formula. The best combination ratio for GE50, UR95, PT95, PN95, and AO50, which was assigned as Formula 1 (F1), was found at EC₀, EC₂₅, EC₂₀, EC₂₀, and EC₁₀, respectively. The vasodilatory effect of the extracts prepared from different extraction methods using F1 ratio was also studied. From the results, the EC₅₀ and R_max of total 50% ethanolic extract of five herbs using F1 ratio (F1-2) were 0.028 ± 0.005 mg/mL and 101.71% ± 3.64%, with better values than F1 (0.104 ± 0.014 mg/mL and 97.80% ± 3.12%, respectively). In conclusion, the optimum ratio and appropriate extraction method (F1-2) for the new TCM formula were revealed.

Introduction

Hypertension is well known as the “silent killer” and presents a worldwide public health challenge due to its high frequency of causing concomitant diseases, especially in the kidney and cardiovascular systems.¹ Even though synthetic antihypertensive drugs have higher efficacy on specific mechanism of action, current studies show that combination therapies with different mechanisms of action mediated may be more effective than single drugs since multiple signaling pathways are involved in hypertension.^2,3 Nowadays, pharmacological studies on antihypertensive drugs are focused on the holistic and multitargeting therapeutic agents due to the variety of concomitant diseases caused by hypertension.^4,5 Traditional Chinese medicine (TCM) fulfills these criteria and has attracted the attention of scientists worldwide for studying hypertension treatment.³ In addition, TCM preparations are highly effective for regulating physiological pathways and have high safety profiles in clinical settings.^6,7

According to the Chinese Medicine Guideline for Hypertension Management, TCM principles defined the syndromes of hypertension into three main types, which mainly target the liver, kidney, and spleen.⁸ The first disharmony pattern is the fire syndrome, which has been categorized into four types such as the liver, stomach, heart, and intestinal fire. The second pattern is the later stage of hypertension that would cause phlegm-fluid retention syndrome. The third syndrome of hypertension is the deficiency syndrome, such as kidney and spleen deficiencies.^9,10 Therefore, the commonly used TCM prescription formulae were established according to their ability to counteract the different syndromes of hypertension.¹¹

In this study, five TCM herbs that were most frequently used in clinical prescriptions to counteract different obligate syndromes of hypertension were chosen to achieve antihypertensive effects with higher efficacies. Gastrodia elata and Uncaria rhynchophylla belong to the category of herbs that calm the liver and extinguish wind based on TCM principles. G. elata is known to extinguish wind, arrest spasms, relieve vertigo with pain, and soothe the channels and collaterals.^12

–17 U. rhynchophylla is known to clear heat, calm the liver, decrease blood pressure, and extinguish vertigo and spasms.^18
–20 Pueraria thomsonii is categorized as a cool-acrid herb that releases the exterior. It is best in releasing flesh, arresting fever, promoting fluid production to quench thirst, and relieving neck rigidity^2,21
–23; Panax notoginseng is categorized as a herb that stops bleeding and specializes in staunching bleeding, invigorating blood, and relieving pain. It is famous for its use in trauma treatment^{22,24

–28}; and Alisma orientale is a well-known diuretic agent belonging to the category of herbs that drain water and dampness, promote urination, lower blood lipid, and relieve vertigo.^29,30 According to Loh et al., studies of its vasodilatory effects have mostly utilized isolated rat aortic ring experiments, as it is well known as the “golden tool” for antihypertensive drug development due to its high reliability.^31,32 Thus, the isolated rat aortic ring model was chosen for this study. It aimed to discover the best combination ratio of five TCM herbs for efficacy in treating hypertension.

Materials and Methods

Plants and chemicals

The G. elata, U. rhynchophylla, P. thomsonii, P. notoginseng, and A. orientale herbs were purchased from the local medical hall. The TCM herbs were authenticated by Dr. M.F.Y. (specialist in TCM herbal authentication, Universiti Sains Malaysia). Acetylcholine (Ach) and phenylephrine (PE) were bought from Acros Organics (Belgium). Both chemicals were dissolved in distilled water before use. The herbal extracts were dissolved in distilled water at a concentration of 128 mg/mL to form as stock and stored in freezer at −20°C (Pensonic, PFZ-230).

Tri-step Fourier transform infrared identification and characterization of herbs

The five herbs were ground into powder and passed through a 200-mesh sieve to obtain fine powder before the tri-step Fourier transform infrared (FTIR) spectroscopic analysis. A tablet of potassium bromide (KBr) was used as the blank. About 1 mg of herb powder was mixed with 100 mg of KBr and compressed into a small thin tablet by applying a pressure of ∼10 psi. The spectra of five herbs were recorded at an optical path speed of 0.2 cm/s with 16 co-added scans in a resolution of 4 cm⁻¹ between wavelengths of 4000–400 cm⁻¹ using the Spectrum 400 FTIR spectrometer (v 6.3.5) equipped with a DTGS detector (Perkin-Elmer, USA). The disturbances caused by carbon dioxide and water were directly removed online while scanning. The results were considered valid when transmission of 60% or more was achieved, or else the test has to be repeated by adding either more sample or KBr.³³

The second-derivative infrared (SD-IR) was performed by using the Savitzky–Golay polynomial fitting (13-point smoothing). The two-dimensional infrared (2D-IR) spectra were obtained by placing the sample tablet into the sample holder with a programmable heated jacket controller (Model GS20730; Specac). The dynamic spectra were recorded at temperatures ranging from 50°C to 120°C at intervals of 10°C. The 2D-IR correlation spectra were obtained by treating the series of dynamic spectra with 2D-IR correlation analysis software developed by Tsinghua University (Beijing, China).^33
–35

Preparation of herb extracts

The G. elata, U. rhynchophylla, P. thomsonii, P. notoginseng, and A. orientale herbs were cut into smaller pieces and dried in the oven at 50°C for 3 days until a constant weight was achieved. The herbs were ground into fine powder by using mill. Approximately 200 g of herb powder was prepared in three different containers, and each herb was soaked with 95% ethanol, 50% ethanol, or distilled water, individually. The five powdered herbs macerated with 95% ethanol were labeled as GE95, UR95, PT95, PN95, and AO95. Those macerated with 50% ethanol were labeled as GE50, UR50, PT50, PN50, and AO50, whereas those macerated with distilled water were labeled as GEW, URW, PTW, PNW, and AOW. The maceration process was executed at a temperature of 50°C for 48 h and repeated three times. They were then filtered. The filtrate obtained was concentrated by using rotary evaporator at 50°C under reduced pressure. The concentrated filtrates were known as herbal extracts. All the extracts were freeze-dried and kept in desiccators at 4°C for future use. The yields of GE95, GE50, GEW, UR95, UR50, URW, PT95, PT50, PTW, PN95, PN50, PNW, AO95, AO50, and AOW were 10.78%, 20.10%, 50.93%, 5.09%, 8.93%, 5.98%, 7.16%, 18.31%, 17.81%, 15.11%, 23.64%, 10.69%, 13.22%, 11.77%, and 9.15%, respectively.

Animals

The experiments were performed on male Sprague Dawley (SD) rats aging between 8 and 10 weeks and weighing 180 g to 250 g. The rats were housed at room temperature with a 12-hour light/12-hour dark cycle with free access to food and water. The investigation strictly followed the Guidelines in the Care and Use of Laboratory Animals by Universiti Sains Malaysia and the protocol was approved in advance, USM/Animal Ethics Approval/2016/(103) (777).

Preparation of rat aortic rings and in vitro assays

Before the experiment, the Krebs–Henseleit (K-H) (118.0 mM NaCl, 4.7 mM KCl, 25.0 mM NaHCO₃, 2.5 mM CaCl₂, 1.2 mM KH₂PO₄, 1.2 mM MgSO₄, and 11.0 mM glucose, pH 7.4) solution was prepared in a Petri dish and aerated with carbogen (95% O₂ and 5% of CO₂). A male SD rat was euthanized with an overdose of CO₂ gas inhalation. An incision was made through its sternum to locate the aorta. The aorta was excised gently and carefully and immediately immersed in the Petri dish containing K-H solution. The adhering connective tissues, fat, and blood on the isolated aorta were carefully discarded, and the aorta was cut into 3–5-mm-long sections and mounted carefully (to prevent damage of endothelium) in a tissue bath chamber containing 10 mL of K-H solution by using an L-shaped brace and a small needle hook. The tissue bath chamber was continuously aerated with carbogen gas and the temperature was maintained at 37°C. The needle hook was connected to a force-electricity transducer (GRASS Force-Displacement Transducer FT03 C Isometric Measurements). The suspended aortic ring was then left to equilibrate for 45 min at a resting tension of 1.0 g. The K-H solution was replaced at 15-min intervals, and the tension was readjusted to 1.0 g if necessary. The integrity of the aortic rings was determined by exposing them to contractile agent, PE (1 μM) to achieve at least 0.6 g of contraction, and relaxing agent, Ach (1 μM) to achieve at least 60% of relaxation. The aortic ring was rinsed with K-H solution three times and the resting tension was readjusted back to 1.0 g. At this stage, 1 μM of PE was applied to induce precontraction. Subsequently, cumulative concentrations of the herbal extracts (100 μL) were applied into the tissue bath chamber at 20-min intervals.³⁶ The contractile response was measured by the force-electricity transducer, amplified by Quad bridge amp (AD instrument Australia), and then converted into digital signals by PowerLab 26T (AD instrument Australia). Cumulative concentrations (0.0025–1.28 mg/mL in tissue bath) of extracts were applied after PE precontraction. The concentration–response curves of each herbal extract were constructed as shown to compare the vasodilatory effects of extracts.^15,36,37 The extract of each herb, which exerted the highest vasodilatory effect, was selected for further studies. The EC₁₀, EC₁₅, EC₂₀, and EC₂₅ of selected extracts were calculated from their respective concentration–response curves and used for orthogonal stimulus–response compatibility group studies.

Orthogonal stimulus–response compatibility group studies

The orthogonal stimulus–response compatibility group screening was arranged by using L₂₅ (5⁵) formula and forming 25 sets of experiments at which their ratios would not be repeated in other sets of experiments. Their expected relaxation (ER) values, corresponding to each set of the experiments, should not exceed 100% so that the resulting actual relaxation (AR) is valid. The efficacy of each set of the experiments was calculated as (AR/ER) × 100%. The formula that exhibited the highest efficacy would be assumed to be the most optimum ratio (F1). The comparison studies of the calculated optimum ratio (calculated from orthogonal stimulus–response compatibility table), experimental optimum ratio, and formulae that exhibited a vasodilatory effect close to the experimental optimum ratio were carried out to confirm the potential of F1 in terms of vasodilatory effect.³⁸

Efficacy of extracts using different extraction methods following the F1 ratio

After the formulation of F1, the vasodilatory effects of the extracts prepared by using different extraction solvents and extraction methods of the five herbs following the ratio of F1 were investigated. The powders of five raw herbs were prepared in exactly the same ratio as F1 and total extraction at 50°C was conducted using distilled water, 50% ethanol, and 95% ethanol, separately. The extracts were labeled as F1-1, F1-2, and F1-3, respectively. Second, U. rhynchophylla, P. thomsonii, and P. notoginseng were prepared following the ratio as in F1, extracted at 50°C using 95% ethanol, subsequently mixed with the 50% ethanolic extract of A. orientale (following the ratio of F1), and labeled as F1-4. Finally, a total extraction of the five herbs following the ratio of F1 was performed by using distilled water at 100°C and labeled as F1-5. The vasodilatory effects were determined by using the in vitro isolated rat aortic ring model. The concentration–response curves were compared among F1 and other F1 derivative groups.

Statistical analysis

All data are expressed as mean ± SEM. Statistical analysis was performed using one-way analysis of variance using SPSS version 22 software. All tests were two tailed and the significance was set at P < .05.

Results

Tri-step FTIR macro-fingerprint analysis

Tri-step FTIR spectroscopic analysis was performed to identify the fingerprints of G. elata, U. rhynchophylla, P. thomsonii, P. notoginseng, and A. orientale. Their 1D-IR, SD-IR, and 2D-IR spectra were compared with the atlas of Sun et al. to verify the authenticity of these five herbs. The main components and the holistic variations of the chemical constituents were also reported. As shown in Figure 1a, the conventional FTIR spectra of all the five herbs are identical to the spectra in the atlas of TCM published by Sun et al., which proved the authenticity of these herbs. The peak characteristics of each herb in their respective spectrum have been categorized accordingly and listed in Table 1. To further distinguish the differences between the peak characteristics, SD-IR spectroscopy was performed and the spectra within the range of 800–1550 cm⁻¹ are shown in Figure 1b. It can be seen that among the herbs, the SD-IR spectrum of U. rhynchophylla was the most unique in terms of peak intensity rather than peak position, which indicated the possibility of a higher content of vasoactive constituents. In addition, the relationship between the chemical constituents in each herb could be determined by applying external perturbation. For instance, different temperatures were applied to the 2D-IR spectroscopic analysis of these herbs and the resulting dynamic spectra of G. elata, U. rhynchophylla, P. thomsonii, P. notoginseng, and A. orientale are shown in Figure 2a–e, respectively. The dynamic spectra obtained were completely different from one another, which facilitated the identification among different herbs and their chemical structures.

FIG. 1.

Representative 1D-IR (a) and SD-IR (b) spectra in the ranges of 400–4000 and 800–1550 cm⁻¹ for five raw herb powders Gastrodia elata (i), Uncaria rhynchophylla (ii), Pueraria thomsonii (iii), Panax notoginseng (iv), and Alisma orientale (v). ID-IR, one-dimensional infrared; SD-IR, second-derivative infrared.

FIG. 2.

The autopeak curves (i), 2D-IR (ii), and 3D-IR (iii) in the range of 875–1150 cm⁻¹ and autopeak curves (iv), 2D-IR (v), and 3D-IR (vi) in the range of 1160–1520 cm⁻¹ for G. elata (a); the autopeak curves (i), 2D-IR (ii), and 3D-IR (iii) in the range of 850–1100 cm⁻¹ and the autopeak curves (iv), 2D-IR (v), and 3D-IR (vi) in the range of 1200–1500 cm⁻¹ for U. rhynchophylla (b). The autopeak curves (i), 2D-IR (ii), and 3D-IR (iii) in the range of 870–1160 cm⁻¹ and autopeak curves (iv), 2D-IR (v), and 3D-IR (vi) in the range of 1160–1550 cm⁻¹ for P. thomsonii (c); the autopeak curves (i), 2D-IR (ii), and 3D-IR (iii) in the range of 850–1180 cm⁻¹ and autopeak curves (iv), 2D-IR (v), and 3D-IR (vi) in the range of 1180–1520 cm⁻¹ for P. notoginseng (d). The autopeak curves (i), 2D-IR (ii), and 3D-IR (iii) in the range of 860–1170 cm⁻¹ and autopeak curves (iv), 2D-IR (v), and 3D-IR (vi) in the range of 1170–1520 cm⁻¹ for A. orientale (e). 2D-IR, two-dimensional infrared.

Table 1.

The Fourier Transform Infrared Spectra Peak Characterization of Raw Powder of Gastrodia elata, Uncaria rhynchophylla, Pueraria thomsonii, Panax notoginseng, and Alisma orientale

Peak (cm ⁻¹ )
G. elata	U. rhynchophylla	P. thomsonii	P. notoginseng	A. orientale	Primary structure	Possible compounds
3393	3399	3358	3380	3360	O–H, ν	Various
2929	2918	2931	2928	2930	C–H, ν_as	—
	2850				C–H, ν_s	—
1727	1736	1726			C = O, ν	Ester
1649		1647	1649	1650	N–H, δ	Amide I
	1623				C–O, ν_as	Calcium oxalate
				1545	N–H, δ	Amide II
1515	1515				Ring	Aromatic
	1453	1457	1452	1457	Ring	Aromatic
1423		1423	1421	1420	(O) C–H, δ	Saccharides
1370	1376	1369	1376		(O) C–H, δ	Ester
				1343	(O) C–H, δ	Saccharides
	1319				C–O, ν_s	Calcium oxalate
1238	1248	1243	1242	1242	C–O, ν	Ester, glycoside
1156	1158	1157	1156	1155	C–O, ν	Saccharides
1080	1110	1081	1079	1079	C–O, ν	Saccharides
	1056			1045	C–O, ν	Saccharides
1020		1015	1020	1018	C–O, ν	Saccharides

ν, stretching; ν_s, symmetrical stretching; ν_as, asymmetrical stretching; δ, bending.

Vasodilatory effects of different herbal solvent extracts on PE precontractile tone

Three types of solvents were used for the herbal extractions: 95% ethanol, 50% ethanol, and distilled water. According to Figure 3a, the GE95 and GEW showed similar trends of concentration-dependent vasodilatory effects on the isolated aortic rings. However, the GE50 clearly showed the highest vasodilatory effects on the aortic rings compared to other two extracts from the same herb, with lowest EC₅₀ and highest R_max values as shown in Table 2. According to Figure 3b–d, the UR95, PT95, and PN95 have exhibited the highest vasodilatory effects among the extracts of each herb, and their EC₅₀ and R_max values are shown in Table 2 as well. The vasodilatory effects of AO95 and AO50 as shown in Figure 3e have similar R_max values. However, the EC₅₀ value of AO50 was much lower than that of AO95. Thus, AO50 was considered to have the most potent vasodilatory effects of the three extracts. Generally, GE50, UR95, PT95, PN95, and AO50 were selected for further orthogonal stimulus–response compatibility group screening for vasodilatory effects.

FIG. 3.

Vasodilatory effect of (a) 95% ethanolic (GE95), 50% ethanolic (GE50), and distilled water (GEW) extracts of G. elata with cumulative concentration of 0.0025–2.5575 mg/mL; (b) 95% ethanolic (UR95), 50% ethanolic (UR50), and distilled water (URW) extracts of U. rhynchophylla with cumulative concentration of 0.0025–0.1575 mg/mL; (c) 95% ethanolic (PT95), 50% ethanolic (PT50), and distilled water (PTW) extracts of P. thomsonii with cumulative concentration of 0.0025–2.5575 mg/mL; (d) 95% ethanolic (PN95), 50% ethanolic (PN50), and distilled water (PNW) extracts of P. notoginseng with cumulative concentration of 0.0025–0.6375 mg/mL; (e) 95% ethanolic (AO95), 50% ethanolic (AO50), and distilled water (AOW) extracts of A. orientale with cumulative concentration of 0.0025–2.5575 mg/mL, on 1 μM PE precontracted isolated endothelium-intact SD rat aortic rings (n = 8). SD, Sprague Dawley.

Table 2.

Comparison of EC₅₀ and R_max Values Between Different Solvent Extracts of G. elata, U. rhynchophylla, P. thomsonii, P. notoginseng, and A. orientale

Herbs	Concentrations (mg/mL)	Types of solvent	EC₅₀ values (mg/mL)	R_max values (%)
G. elata	0.0025–1.28	95% ethanol	0.131 ± 0.027	101.07 ± 1.92
		50% ethanol	0.024 ± 0.004	106.13 ± 0.91
		dH₂O	0.105 ± 0.018	98.90 ± 2.78
U. rhynchophylla	0.0025–0.08	95% ethanol	0.025 ± 0.003	102.55 ± 6.28
		50% ethanol	0.032 ± 0.006	75.33 ± 5.10
		dH₂O	0.513 ± 0.155	52.36 ± 5.22
P. thomsonii	0.0025–1.28	95% ethanol	0.023 ± 0.003	97.69 ± 3.46
		50% ethanol	4.072 ± 1.354	61.02 ± 4.58
		dH₂O	2.587 ± 0.577	73.40 ± 3.72
P. notoginseng	0.0025–0.32	95% ethanol	0.034 ± 0.007	94.15 ± 2.79
		50% ethanol	0.031 ± 0.004	88.35 ± 0.69
		dH₂O	0.781 ± 0.214	48.27 ± 6.36
A. orientale	0.0025–1.28	95% ethanol	0.099 ± 0.021	97.32 ± 3.20
		50% ethanol	0.085 ± 0.013	97.80 ± 3.68
		dH₂O	33.762 ± 7.539	56.41 ± 3.42

EC₅₀, half of maximal effective concentration; R_max, maximal relaxation. Data are represented as mean ± SEM and plotted in Figure 5 (n = 8 for each different solvent extract).

Orthogonal stimulus–response compatibility group studies

The effective concentrations EC₀, EC₁₀, EC₁₅, EC₂₀, and EC₂₅ of the UR95, GE50, PT95, PN95, and AO50 extracts were obtained from their respective concentration–response curves and shown in Table 3. The compatibility groups were formed using the L₂₅ (5⁵) formula arrangement as shown in Table 4 and the vasodilatory effects were determined on isolated endothelium-intact rat aortic rings. Then, the average relaxation of GE50, UR95, PT95, PN95, and AO50 that corresponds to their EC was further calculated and stated in Table 5. The results showed that the highest average relaxation of GE50, UR95, PT95, PN95, and AO50 was achieved at EC₀, EC₂₅, EC₂₀, EC₂₀, and EC₁₀ (named as F1), respectively. The potency of F1 was further assured by comparing it with the 18th compatibility group (named as F4) with 49% efficacy. Figure 4 shows that the concentration–response curve of vasodilatory effects of F1 was much higher than F4 at a final concentration of 1.2775 mg/mL with EC₅₀ and R_max values shown in Table 7. The potency of F1 was further confirmed by comparing it with F2 and F3. The efficacy of their vasodilatory effects is shown in Table 6. With F1 exhibiting the highest efficacy at 51%, it once again proved its highest potency in exerting vasodilatory effect on the isolated endothelium-intact rat aortic rings, thus it was used for further analysis.

FIG. 4.

Vasodilatory effects of F1 and F4 in cumulative concentration of 0.0025–1.2275 mg/mL on 1 μM PE precontracted isolated endothelium-intact SD rat aortic rings (n = 8). F1, extracts GE50, UR95, PT95, PN95, and AO50 combined in EC: 0, 25, 20, 20, and 10, respectively. F4, extracts GE50, UR95, PT95, PN95, and AO50 combined in EC: 20, 15, 0, 20, and 10, respectively. The maximum percentage of relaxation (R_max) was compared among each other to ensure the efficacy of F1.

Table 3.

The Effective Concentration Selected from Concentration–Response Curves of Each Herbal Extract with the Highest Vasodilatory Effect for Orthogonal Stimulus–Response Compatibility Group Screening

	Concentration (mg/mL)
Effective concentration (%)	A	B	C	D	E
Without extract	0	0	0	0	0
10	0.254	0.722	0.238	1.028	0.253
15	0.348	0.886	0.381	1.542	0.395
20	0.477	1.086	0.610	2.313	0.618
25	0.653	1.331	0.976	3.470	0.965

Concentration stated was obtained from the concentration–response curve of each extract that corresponded to respective effective concentration. Concentrations stated are not the final concentration in tissue bath.

A, 50% ethanolic extract of G. elata; B, 95% ethanolic extract of U. rhynchophylla; C, 95% ethanolic extract of P. thomsonii; D, 95% ethanolic extract of P. notoginseng; E, 50% ethanolic extract of A. orientale.

Table 4.

Orthogonal Stimulus–Response Compatibility Group Studies

	Compatibility groups
Experiment	A	B	C	D	E	ER (%)	AR (%)	Efficacy (%)
1	1	1	1	1	1	0	0	0
2	1	2	2	2	2	40	11.23 ± 1.20	28
3	1	3	3	3	3	60	15.57 ± 2.50	26
4	1	4	4	4	4	80	31.95 ± 2.50	40
5	1	5	5	5	5	100	32.78 ± 1.95	33
6	2	1	2	3	4	55	7.47 ± 1.73	14
7	2	2	3	4	5	80	14.43 ± 2.16	18
8	2	3	4	5	1	70	13.07 ± 1.70	19
9	2	4	5	1	2	65	11.66 ± 2.80	18
10	2	5	1	2	3	60	22.6 ± 1.20	38
11	3	1	3	5	2	65	24.49 ± 2.27	38
12	3	2	4	1	3	60	12.43 ± 1.48	21
13	3	3	5	2	4	85	7.09 ± 0.62	8
14	3	4	1	3	5	75	6.5 ± 0.53	9
15	3	5	2	4	1	70	10.37 ± 1.93	15
16	4	1	4	2	5	75	10.21 ± 1.33	14
17	4	2	5	3	1	70	6.8 ± 2.07	10
18	4	3	1	4	2	65	32.07 ± 1.71	49
19	4	4	2	5	3	90	14.57 ± 1.15	16
20	4	5	3	1	4	80	12.05 ± 1.97	15
21	5	1	5	4	3	85	4.79 ± 1.18	6
22	5	2	1	5	4	80	8.53 ± 1.39	11
23	5	3	2	1	5	75	9.78 ± 0.99	13
24	5	4	3	2	1	70	10.12 ± 1.97	14
25	5	5	4	3	2	95	17.61 ± 2.21	19

Concentration used in orthogonal stimulus–response compatibility group screening: 1, EC₀; 2, EC₁₀; 3, EC₁₅; 4, EC₂₀; 5, EC₂₅; derived from Table 3. Efficacy, (AR/ER) × 100%; ER, expected relaxation; AR, actual relaxation; A, 50% ethanolic extract of G. elata; B, 95% ethanolic extract of U. rhynchophylla; C, 95% ethanolic extract of P. thomsonii; D, 95% ethanolic extract of P. notoginseng; E, 50% ethanolic extract of A. orientale. Data are represented as mean ± SEM (n = 6 for each experiment).

Table 5.

Average Relaxation of Aortic Rings Obtained from Orthogonal Stimulus–Response Compatibility Group Screening

	Average relaxation (%)
Effective concentration (%)	A	B	C	D	E
0	18.31	9.39	13.94	9.18	8.07
10	13.85	10.68	10.68	12.25	19.41
15	12.18	15.52	15.33	10.79	13.99
20	15.14	14.96	17.05	18.72	13.42
25	10.17	19.08	12.62	18.69	14.74

Table 6.

Confirmation Test on Optimum Combination Ratio of Herbal Extracts in Exerting Highest Vasodilatory Effects

	Optimum combination ratio
Formulae	A	B	C	D	E	ER (%)	AR (%)	Efficacy (%)
F1	EC₀	EC₂₅	EC₂₀	EC₂₀	EC₁₀	75	38.15 ± 2.05	51
F2	EC₀	EC₂₅	EC₂₀	EC₂₅	EC₁₀	80	11.79 ± 2.05	15
F3	EC₂₀	EC₂₅	EC₂₀	EC₂₀	EC₁₀	95	11.18 ± 1.01	12

Efficacy, (AR/ER) × 100%; A, 50% ethanolic extract of G. elata; B, 95% ethanolic extract of U. rhynchophylla; C, 95% ethanolic extract of P. thomsonii; D, 95% ethanolic extract of P. notoginseng; E, 50% ethanolic extract of A. orientale. Data are represented as mean ± SEM (n = 8 for each formula).

Table 7.

The EC₅₀ and R_max Values of Different Solvent Extracted Formulae

Formulae	Doses (mg/mL)	EC₅₀ values (mg/mL)	R_max values (%)
F1	0.0025–0.64	0.104 ± 0.014	97.80 ± 3.12
F4	0.0025–0.64	0.225 ± 0.010	81.28 ± 5.25
F1-1	0.0025–1.28	0.250 ± 0.101	95.77 ± 2.75
F1-2	0.0025–0.16	0.028 ± 0.005	101.71 ± 3.64
F1-3	0.0025–1.28	0.254 ± 0.067	93.25 ± 3.98
F1-4	0.0025–1.28	0.171 ± 0.022	99.69 ± 2.12
F1-5	0.0025–1.28	0.377 ± 0.124	92.74 ± 3.38

F1, extracts A, B, C, D, and E combined in EC: 0, 25, 20, 20, and 10, respectively. F4, extracts A, B, C, D, and E combined in EC: 20, 15, 0, 20, and 10, respectively. F1-1, distilled water extract of five herbs using F1 ratio at 50°C; F1-2, 50% ethanolic extract of five herbs using F1 ratio at 50°C; F1-3, 95% ethanolic extract of five herbs using F1 ratio at 50°C; F1-4, total 95% of ethanolic extraction of U. rhynchophylla, P. thomsonii, and P. notoginseng (F1 ratio) combined with D (F1 ratio) at 50°C; F1-5, distilled water extract of five herbs following F1 ratio at 100°C. EC₅₀, half of maximal effective concentration; R_max, maximal relaxation. A, 50% ethanolic extract of G. elata; B, 95% ethanolic extract of U. rhynchophylla; C, 95% ethanolic extract of P. thomsonii; D, 95% ethanolic extract of P. notoginseng; E, 50% ethanolic extract of A. orientale. Data are represented as mean ± SEM and plotted in Figure 4 and 5 (n = 8 for each formula).

Vasodilatory effects of different solvent extracts of F1 formula

F1 was further analyzed by using different preparation and solvent extraction methods as described previously in the Materials and Methods section. The cumulative concentrations of F1, F1-1, F1-2, F1-3, F1-4, and F1-5 (0.0025–0.3175 mg/mL in tissue bath) were applied to the isolated endothelium-intact rat aortic rings and the resulting concentration–response curves were compared among each other, as shown in Figure 5. All formulae elicited concentration-dependent vasodilatory effects on the isolated aortic rings. The EC₅₀ and R_max values of F1-1, F1-3, F1-4, and F1-5 were mildly significant (P < .05) compared to F1 as shown in Table 7. However, F1-2 has dramatically increased the R_max value to 101.71% ± 3.64% with a final concentration of 0.3175 mg/mL and the lowest EC₅₀ value at 0.028 ± 0.005 mg/mL (P < .05), which indicated it is the most effective in terms of vasodilatory effects compared to the other F1 derivatives.

FIG. 5.

Vasodilatory effects of F1, F1-1, F1-2, F1-3, F1-4, and F1-5 with cumulative concentration of 0.0025–0.3175 mg/mL on 1 μM PE precontracted isolated endothelium-intact SD rat aortic rings (n = 8) (^* P < .05, ^** P < .01, ^*** P < .001). F1-1, distilled water extract of five herbs using F1 ratio at 50°C; F1-2, 50% ethanolic extract of five herbs using F1 ratio at 50°C; F1-3, 95% ethanolic extract of five herbs using F1 ratio at 50°C; F1-4, total 95% of ethanolic extraction of U. rhynchophylla, P. thomsonii, and P. notoginseng (using F1 ratio) combined with D (using F1 ratio) at 50°C; F1–5, distilled water extract of five herbs using F1 ratio at 100°C. The maximum percentage of relaxation (R_max) was compared among each other to determine the formula with the highest efficacy in terms of exerting vasodilatory effects.

Discussion

G. elata, U. rhynchophylla, P. thomsonii, P. notoginseng, and A. orientale were selected in this antihypertensive study mainly due to their respective abilities to counteract a variety of hypertensive symptoms such as pacifying the liver, extinguishing wind, improving blood circulation, removing blood stasis, reducing headaches, arresting convulsion, and acting as diuretic agents.^18
–20 Before the commencement of the experiment, tri-step FTIR was used to authenticate these herbs. Tri-step FTIR consists of three fingerprint identification methods, which are the conventional FTIR (1D-IR), SD-IR, and 2D-correlation IR (2D-IR) at which the resolution of the fingerprint features would be gradually amplified and enhanced. The simple and rapid analysis system, as well as the highly reproducible results of the tri-step FTIR spectroscopy, makes it the gold standard for determining the authenticity of TCM materials. By using the tri-step FTIR spectroscopic analysis, the relationship among different functional active components in each herb can be determined without the need to expose the samples to destructive chemical or physical perturbation, which would cause the original samples to behave differently.³⁹ However, some of the characteristic peaks would be too small due to the low concentration of a particular chemical constituent within the sample, or had been unintentionally masked by bigger peaks, thus those peaks were invisible for interpreting in the one-dimensional IR (1D-IR) spectra. Therefore, the SD-IR spectroscopy with its higher resolution was subsequently performed to solve this problem and emphasize the hidden peaks in the SD-IR spectra. In addition to 1D-IR and SD-IR spectroscopic identification, the 2D-IR spectroscopic identification was performed to compensate for the drawbacks of other analytical systems by exposing the samples to external perturbation, and therefore altering the intra- and intermolecular relationship of the chemical constituents within the samples. Subsequently, the relationship between functional groups could be analyzed from the dynamic spectra obtained.

According to the results, the peak intensities shown in the 1D-IR spectra of G. elata, U. rhynchophylla, P. thomsonii, P. notoginseng, and A. orientale were very similar to one another in wavelengths ranging between 2500–3500 and 950–1200 cm⁻¹. The ladder-shaped absorption peaks in the range of 950–1200 cm⁻¹ indicated that all of the five herbs contained a high concentration of starch. To further distinguish the differences among the five herbs, other ranges of absorption peaks were examined such as the peaks at 1623 and 1319 cm⁻¹ that only appeared in the spectrum of U. rhynchophylla, and not in the spectra of other herbs. Both these peaks indicated that U. rhynchophylla consists of certain amounts of calcium oxalate. In addition, there were a pair of amide absorption peaks visible in the 1D-IR spectrum of A. orientale at 1650 and 1545 cm⁻¹, both of which correspond to the characteristic vibration of amide I and amide II and showed that A. orientale contains large numbers of secondary amide proteins. There was only one amide absorption peak at 1649 cm⁻¹ in the spectra of other three herbs (G. elata, P. thomsonii, and P. notoginseng). The absence of absorption peak around 1550 cm⁻¹ proved that G. elata, U. rhynchophylla, P. thomsonii, and P. notoginseng contain mostly tertiary amide protein.⁴⁰ This once again proved that tri-step FTIR spectroscopic analysis not only provided the basis for herbal discrimination but also revealed the characteristics of the chemical constituents contained within the herbs.

The higher resolution SD-IR spectroscopy tends to separate and emphasize the hidden and unintentionally masked overlapping peaks. The SD-IR spectra of the five herbs shown in the figures are mostly composed of the characteristic peaks of starch, such as at the wavelengths at ∼1160, 1107, 1078, 1055, 1018, 987, 922, and 861 cm⁻¹.³⁹ In the spectrum of U. rhynchophylla, the absorption peaks for starch were not obvious, indicating the lower starch content in U. rhynchophylla compared to other herbs. The characteristic peaks of aromatic compounds, which might represent the alkaloids, were observed at around 1515 and 1468 cm⁻¹. These alkaloids are speculated to contribute to the herbal vasodilatory effects. The intensities of both these peaks are in descending order of U. rhynchophylla, G. elata, A. orientale, P. thomsonii, and P. notoginseng. The results implied that the U. rhynchophylla may contain the highest amount of vasoactive compounds among the herbs, thus exerting the highest vasodilatory effect with the lowest EC₅₀ value compared to others.

Furthermore, the 2D-IR spectroscopic analysis of these five herbs was performed, shown in Figure 2. The different chemical constituents of the herbs would result in different thermal stabilities, and hence, the different temperatures applied during the 2D-IR spectroscopic analysis would improve the resolution of IR spectroscopy through the resulting autopeaks, 2D, and 3D-dynamic spectra. From the 2D-IR spectra of U. rhynchophylla, a unique wave can be observed that made it easily distinguishable from other herbs. The other four herbs have high starch content, leading to similar autopeak features such as similar peak numbers and positions especially in autopeaks of P. thomsonii and P. notoginseng. However, there was a big difference observed between the relative intensity of autopeaks. For example, in the range of 870–970 cm⁻¹, the autopeak intensities of P. notoginseng were strongest, whereas G. elata had the weakest autopeak intensities at this wavelength. Moreover, the autopeak at around 1470 cm⁻¹ (C-H banding of aromatic ring) in the 2D-IR spectra of G. elata, P. notoginseng, and A. orientale formed negative corresponding cross peaks with other autopeaks, which indicated that the thermal stability of the aromatic rings was different from one another.³⁹

Ethanol was chosen as the extraction solvent instead of methanol so that the final product could be readily consumed by patients and the further active compounds' fractionation processes could be eliminated for cost–effectiveness and time-saving purposes. In addition, the folk practice of herbal prescription preparations was considered, leading to the use of crude herbs in this study instead of directly using the pure compounds of each herb.⁴¹ The in vitro rat aorta model was used throughout the vasodilatory effect study to follow the current trend of antihypertensive drug research. This method is well known as the “golden tool” in antihypertensive agent development and vascular tone pharmacological researches.^31,32 Isolated aortic rings were used instead of other parts of the blood vessels to minimize the variation in smooth muscle cell orientation and to avoid the damage of endothelium to obtain highly reproducible results.

Three types of solvents were used to extract G. elata, U. rhynchophylla, P. thomsonii, P. notoginseng, and A. orientale, and subsequently, all the extracts obtained were screened for their vasodilatory effects with isolated endothelium-intact rat aortic rings. The results showed that GE50 and AO50 exhibited the highest vasodilatory effects compared to their respective 95% ethanolic and distilled water extracts. The results implied that the vasoactive compounds present in G. elata and A. orientale could be extracted by both distilled water and ethanol. However, the vasoactive compounds extracted by ethanol and distilled water should be presented in equivalent ratios to produce the highest vasodilatory effects. The final concentration applied to the aortic rings for both GE50 and AO50 in vasodilatory effect studies was 2.5575 mg/mL. Moreover, UR95, PT95, and PN95 have exhibited the highest vasodilatory effects on isolated rat aortic rings compared to extracts of these herbs from other solvents. These results indicated that the vasoactive compounds present in U. rhynchophylla, P. thomsonii, and P. notoginseng were extractable by higher percentage of ethanol rather than distilled water, and thus, less vasodilatory effects were obtained from 50% ethanolic and distilled water extracts. Interestingly, the final concentration of UR95 and PN95 required to reach the maximal vasorelaxation percentage was 0.1575 and 0.6375 mg/mL, respectively, lower than that of GE50, PT95, and AO50. Therefore, the potency of the five herbs in exerting vasodilatory effects on isolated rat aortic rings could be arranged in ascending order as such AO50<PT95<GE50<PN95<UR95. Hence, these five herbal extracts were selected for compatibility group screening.

Five different ECs of each herb were selected for the orthogonal stimulus–response compatibility group studies based on the concept where the expected R_max values of each set of compatibility group studies should not exceed 100% after combining the five extracts, as shown in Table 3. The concentrations of each herb were calculated from their representative concentration–response curves as shown in Figure 3. The 25 compatibility groups were arranged using the formula L₂₅ (5⁵), which was same as the previous study regarding the combination of hypotensive components of TCM.⁴² None of the ratios of herbal extracts was repeated in the 25 sets of compatibility group studies by using this formula. Visually, none of the efficacies of the compatibility group studies exceeded 50%. This phenomenon could be explained by the fact that the EC selected for the compatibility group studies was very low in concentration even after combining all of five extracts. Moreover, there might be unknown interactions that occurred after combining all the five extracts in different ratios, hence exhibiting a lower AR than expected. However, the main purpose of performing compatibility group studies was not to directly compare their actual percentage of vasodilation among each other, but to calculate the average relaxation of each EC corresponding to their extracts. According to the results shown, the highest average relaxation of GE50, UR95, PT95, PN95, and AO50 corresponds to EC₀, EC₂₅, EC₂₀, EC₂₀, and EC₁₀ (F1), respectively, with an efficacy of 51%.

There was one of the combination ratios of five extracts, compatibility group 18th (F4), which exhibited a higher potency as F1 with efficacy of 49%. Nonetheless, after comparing the cumulative concentration–response curves among F1 and F4, this apprehension could be rebutted. At the same time, to confirm that GE50 could be ignored, F3 was established by taking the highest average relaxation of GE50 at EC₂₀ to replace EC₀ in the F1 and formulated as GE50, UR95, PT95, PN95, and AO50 with EC₂₀, EC_25, EC₂₀, EC₂₀, and EC₁₀, respectively. Furthermore, the average relaxations of PN95 at EC₂₀ and EC₂₅, as shown in Table 5, were close to each other. Therefore, to further verify the efficacy of F1, F2 was established by replacing the EC₂₀ of PN95 in F1 to EC₂₅, hence the F2 was formulated as GE50, UR95, PT95, PN95, and AO50 with EC₀, EC₂₅, EC₂₀, EC₂₅, and EC₁₀. Both the vasodilatory effects of F2 and F3 were compared to F1. Results showed that the efficacy of F1 formula was the highest, which implied that the presence of GE50 could be eliminated. Hence, once again assuring that the efficacy of F1 in exerting vasodilatory effect was the highest.

At this point, F1 was prepared and again extracted with different types of solvents, including the decoction extraction method, which is the most frequently used method for TCM prescription preparations. Fundamentally, the TCM principle was used to emphasize that all the herbs were formulated as TCM prescription should be extracted in total and not individually, to be functionally active as a whole. Therefore, extra sets of experiments were carried out. The results showed that neither 95% of ethanol (F1-3) nor distilled water (F1-1) extraction at 50°C could act as the best solvents to extract the vasoactive compounds, not even in F1-4 extract. However, the vasodilatory effect of the 50% ethanolic extract of F1 (named as F1-2) was significantly different from that of F1 as well as any of its derivatives. This showed that it was the most effective in terms of vasodilatory effects compared to others. Moreover, the R_max value of F1-2 was obtained at the final concentration of 0.3175 mg/mL, which was far smaller than that of any other single herb extracts used in this formula except UR95. Furthermore, the most frequently used TCM prescription preparation method, the decoction, did not result in the highest R_max value when used to extract these herbs using F1 ratio (F1-5), but produced the highest EC₅₀ value compared to other formulae. Obviously, this preparation method is not suitable to extract these five herbs to obtain the best vasodilatory effects. This phenomenon could be explained by the fact that the vasoactive compounds present in the F1 are thermolabile or lipid soluble, hence cannot be extracted by using distilled water. Thus, the reliability of the TCM theory was proven to hold true at least in the case of vasoactive compound extraction.

In the present study, the 95% ethanolic extracts of U. rhynchophylla, P. thomsonii, and P. notoginseng, and the 50% ethanolic extracts of G. elata and A. orientale had the highest potency for exerting vasodilatory effects compared to other extracts using other solvents. The best combination ratio of these five herbal extracts GE50, UR95, PT95, PN95, and AO50 was at EC₀, EC₂₅, EC₂₀, EC₂₀, and EC₁₀, respectively, which was named as formula 1 (F1). Furthermore, 50% ethanol was found to be the best solvent for extracting the vasoactive compounds of F1 to exhibit the highest efficacy of vasodilatory effects with the extraction formula named as formula 1–2 (F1-2). Therefore, F1-2 was determined to be the best formula for exerting vasodilatory effects in isolated rat aortic rings. Further in vitro and in vivo experiments should be carried out to elucidate the signaling mechanism pathways used by F1-2 in exerting its vasodilatory effects coupled with toxicity study, to confirm its safety and efficacy as a new antihypertensive drug.

Footnotes

Acknowledgment

The author expresses the deepest appreciation to all authors who have made contributions to this article.

Author Disclosure Statement

No competing financial interests exist.

References

Kearney

, Whelton

, Reynolds

, Muntner

, Whelton

, He

: Global burden of hypertension: Analysis of worldwide data. Lancet, 2005; 365:217–223.

, Koon

, Cheung

, et al.: The anti-hypertensive effect of Danshen (Salvia miltiorrhiza) and Gegen (Pueraria lobata) formula in rats and its underlying mechanisms of vasorelaxation. J Ethnopharmacol, 2011; 137:1366–1372.

Xiong

, Yang

, Liu

, Zhang

, Wang

: Chinese herbal formulas for treating hypertension in traditional Chinese medicine: Perspective of modern science. Hypertens Res, 2013; 36:570–579.

Xiong

, Chu

, Li

, He

: Clinical application of the TCM classic formulae for treating chronic bronchitis. J Tradit Chin Med, 2011; 31:69–72.

, Chen

: Integrative medicine: The experience from China. J Altern Complement Med, 2008; 14:3–7.

, Zhang

, Gong

, He

, Deng

: Simultaneous determination of Gastrodin and Ligustrazine hydrochloride in dog plasma by gradient high-performance liquid chromatography. J Pharm Biomed Anal, 2006; 41:1083–1087.

Williamson

, Lorenc

, Booker

, Robinson

: The rise of traditional Chinese medicine and its materia medica: A comparison of the frequency and safety of materials and species used in Europe and China. J Ethnopharmacol, 2013; 149:453–462.

Wang

, Xiong

, Liu

: Traditional chinese medicine syndromes for essential hypertension: A literature analysis of 13,272 patients. Evid Based Complement Alternat Med, 2014; 2014:418206.

Qian

, Zhang

, Zhou

: Relationship of blood pressure variability and TCM constitution classification in essential hypertension patients. J Chin J Integr Tradit West Med, 2003; 23:88–90.

10.

Wang

, Xiong

: Evidence-based chinese medicine for hypertension. Evid Based Complement Alternat Med, 2013; 2013:978398.

11.

Luiz

, Cordovil

, Filho

, Ferreira

: Zangfu zheng (patterns) are associated with clinical manifestations of zang shang (target-organ damage) in arterial hypertension. Chin Med, 2011; 6:23.

12.

Al-Shehabi

, Iratni

, Eid

: Anti-atherosclerotic plants which modulate the phenotype of vascular smooth muscle cells. Phytomedicine, 2015; 23:1068–1081.

13.

Guo

, Wang

, Lin

, et al.: 4-Hydroxybenzyl-substituted amino acid derivatives from Gastrodia elata . Acta Pharmaceutica Sinica B, 2015; 5:350–357.

14.

Hayashi

, Sekine

, Deguchi

, et al.: Phenolic compounds from Gastrodia rhizome and relaxant effects of related compounds on isolated smooth muscle preparation. Phytochemistry, 2002; 59:513–519.

15.

Kim

, Moon

, Oh

, Kim

, Lee

: Ether fraction of methanol extracts of Gastrodia elata, a traditional medicinal herb, protects against kainic acid-induced neuronal damage in the mouse hippocampus. Neurosci Lett, 2001; 314:65–68.

16.

, Wang

, Ouyang

, et al.: New compounds with neuroprotective activities from Gastrodia elata . Phytochem Lett, 2016; 15:94–97.

17.

Ojemann

, Nelson

, Shin

, Rowe

, Buchanan

: Tian ma, an ancient Chinese herb, offers new options for the treatment of epilepsy and other conditions. Epilepsy Behav, 2006; 8:376–383.

18.

Hou

, Lin

, Chen

, Lee

: Monoamine oxidase B (MAO-B) inhibition by active principles from Uncaria rhynchophylla . J Ethnopharmacol, 2005; 100:216–220.

19.

Shim

, Kim

, Ju

, Choi

, Jeong

, Oh

: Effects of the hook of Uncaria rhynchophylla on neurotoxicity in the 6-hydroxydopamine model of Parkinson's disease. J Ethnopharmacol, 2009; 126:361–365.

20.

Zhang

, Zhao

, Xu

, Feng

, Qu

: Medicinal uses, phytochemistry and pharmacology of the genus Uncaria . J Ethnopharmacol, 2015; 173:48–80.

21.

Chen

, Song

, Wang

, et al.: Metabolic differentiations of Pueraria lobata and Pueraria thomsonii using 1 H NMR spectroscopy and multivariate statistical analysis. J Pharm Biomed Anal, 2014; 93:51–58.

22.

Dang

, Miao

, Chen

, et al.: The antithrombotic effect of RSNK in blood-stasis model rats. J Ethnopharmacol, 2015; 173:266–272.

23.

Wong

, Li

, Razmovski-Naumovski

, Chan

: Kudzu root: Traditional uses and potential medicinal benefits in diabetes and cardiovascular diseases. J Ethnopharmacol, 2011; 134:584–607.

24.

Cai

, Li

, Chen

, et al.: Ginsenoside-Rd, a new voltage-independent Ca²⁺ entry blocker, reverses basilar hypertrophic remodeling in stroke-prone renovascular hypertensive rats. Eur J Pharmacol, 2009; 606:142–149.

25.

Guan

, Zhou

, Zhang

, et al.: Ginsenoside-Rd from Panax notoginseng blocks Ca²⁺ influx through receptor- and store-operated Ca²⁺ channels in vascular smooth muscle cells. Eur J Pharmacol, 2006; 548:129–136.

26.

Lee

, Kim

: A review on the medicinal potentials of ginseng and ginsenosides on cardiovascular diseases. J Ginseng Res, 2014; 38:161–166.

27.

, Leung

GPH

, Kwan

, et al.: Suppression of diet-induced hypercholesterolaemia by saponins from Panax notoginseng in rats. J Funct Foods, 2013; 5:1159–1169.

28.

, Liu

, Lu

, Pang

: Effects of Panax notoginseng saponins on proliferation and apoptosis of vascular smooth muscle cells. J Ethnopharmacol, 2011; 137:226–230.

29.

Feng

, Chen

, Tian

, Chen

, Zhao

, Lin

: Diuretic and anti-diuretic activities of the ethanol and aqueous extracts of Alismatis rhizoma. J Ethnopharmacol, 2014; 154:386–390.

30.

Zhao

, Xu

, Che

: Alisolide, alisols O and P from the rhizome of Alisma orientale . Phytochemistry, 2008; 69:527–532.

31.

Loh

, Tan

, Ch'ng

, Ahmad

, Asmawi

, Yam

: Overview of antagonists used for determining the mechanisms of action employed by potential vasodilators with their suggested signaling pathways. Molecules, 2016; 21:495.

32.

Rameshrad

, Babaei

, Azarmi

, Fouladia

: Rat aorta as a pharmacological tool for in vitro and in vivo studies. Life Sci, 2016; 145:190–204.

33.

Choong

, Sun

, Zhou

, Ismail

, Rashid

BAA

, Tao

: Determination of storage stability of the crude extracts of Ganoderma lucidum using FTIR and 2D-IR spectroscopy. Vib Spectrosc, 2011; 57:87–96.

34.

, Sun

, Wang

, et al.: Differentiation of five species of Danggui raw materials by FTIR combined with 2D-COS IR. J Mol Struct, 2014; 1069:229–235.

35.

, Cheng

, Li

, Wang

, Huang

, Sun

: Structural characterization of steam-heat treated Tectona grandis wood analyzed by FT-IR and 2D-IR correlation spectroscopy. Chin Chem Lett, 2015; 26:221–225.

36.

Ameer

, Salman

, Siddiqui

MJA

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

37.

Wang

, Shi

, Wang

, Che

, Yeung

: Mechanisms of the vasorelaxant effect of 1-hydroxy-2, 3, 5-trimethoxy-xanthone, isolated from a Tibetan herb, Halenia elliptica, on rat coronary artery. Life Sci, 2007; 81:1016–1023.

38.

, Zhang

, Bian

, Yu

, Xu

, Li

: Vasodilative action of combinations of hypotensive components of traditional Chinese medicines on rat isolated thoracic aorta. Chin J Exp Traditional Med Formulae, 2012; 18:147–151.

39.

Sun

, Zhou

, Chen

: Infrared Spectroscopy for Complex Mixtures. Chemical Industry, Beijing, 2011, pp. 24–66.

40.

Sun

, Zhou

, Qin

: Atlas of Two-Dimensional Correlation Infrared Spectroscopy for Traditional Chinese Medicine Identification. Chemical Industry Press, Beijing, 2003.

41.

Taniguchi

, Kuroda

, Doi

, et al.: Evaluation of gambir quality based on quantitative analysis of polyphenolic constituents. Yakugaku Zasshi, 2007; 127:1291–1300.

42.

Yang

, Duan

, Du

, Yang

, Li

, Yao

: Geissoschizine methyl ether, a corynanthean-type indole alkaloid from Uncaria rhynchophylla as a potential acetylcholinesterase inhibitor. Nat Prod Res, 2012; 26:22–28.