The effect of Hypericum perforatum on isolated rat aorta

Abstract

Context: Different Hypericum species such as Hypericum perforatum (HP) L. and Hypericum triquetrifolium Turra are well known and widely used traditional medicine in Turkey.

Objectives: We investigated the effect of standardized HP extract on endothelium and vascular function.

Materials and methods: After suspending the aortas with endothelium in organ baths containing Krebs solution, contractile and relaxant responses were assessed in the absence and presence of HP (0.05 mg/ml).

Results: Although there were significant reductions in the contractile responses to phenylephrine (1113.73 ± 164.11; 477.40 ± 39.94; p < 0.05) and potassium chloride (745.58 ± 66.73; 112.58 ± 26.58; p < 0.05), no differences in the relaxant responses to acetylcholine (94.61 ± 2.65; 87.79 ± 9.40) and sodium nitroprusside (108.82 ± 5.06; 106.43 ± 7.45) were observed.

Discussion and conclusion: These data suggest that even the high dose of HP intervention does not bring any harmful effect on endothelium and smooth muscle function; meanwhile it might be beneficial on some of diseases accompanied with increased vascular contraction.

Keywords::

• Endothelium
• smooth muscle
• vessel
• vasodilatation
• vasoconstriction

Introduction

The vascular endothelium has a crucial role in the modulation of the vascular functions such as contraction and relaxation of the vascular smooth muscle (Demirci et al., 2005; Moncada and Higgs, 2006; Demirci et al., 2008; Villar et al., 2008). This delicate balance is achieved by synthesizing and secreting vasoactive agents, mainly nitric oxide (NO), prostaglandins, endothelium-derived hyperpolarizing factor (EDHF) (Moncada & Higgs, 2006; Villar et al., 2008) and endothelins (Whittle & Moncada, 1990). Endothelial dysfunction contributes some pathologic conditions, including vasoconstriction, hypertension, atherosclerosis, platelet aggregation and inflammation in association with impaired production or activity of NO, reactive oxygen species (ROS) and cytokines (Herman & Moncada, 2005). Therefore, the protection of the endothelium has a great impact on the function of the whole cardiovascular system.

The popularity of traditional herbal therapy with Hypericum species is increasing as a new self-medication approach for many diseases such as depression (Apaydin et al., 2001; Mennini & Gobbi, 2004; Pilkington et al., 2006) wound healing (Cirak et al., 2010; Samadi et al., 2010), anti-inflammatory (Mukerjee et al., 2008; Hammer et al., 2010) and antiviral (Mukerjee et al., 2008). Additionally, our previous study shows the protective effect of Hypericum perforatum (HP) on trinitrobenzene sulfonic acid-induced inflammatory bowel disease (Dost et al., 2009). The extract contains vast many compounds, mainly hypericin, pseudohypericin, hyperforin, xanthone derivatives, flavanoid derivatives, rutin and biapigenin (Hammer et al., 2007; Dost et al., 2009). Each Hypericum compound has also drawn the attention of scientific researches for more than 50 years. It has been demonstrated in the cell culture studies that the antidepressant effect of HP itself or its ingredients is not only mediated via serotonin receptor (Muller & Rossol, 1994), but also via the inhibition of serotonin and noradrenalin uptake (Neary & Bu, 1999; Neary et al., 2001). A significant proportion of their anti-inflammatory activity has been widely attributed to alteration in the oxidant status (Hunt et al., 2001; Lu et al., 2004; Zou et al., 2004; De Paola et al., 2005; Genovese et al., 2006a,b; Dost et al., 2009), however the suppression of the formation of 5-lipoxygenase and cyclooxygenase-1 (COX-1) products (Albert et al., 2002), inhibition of prostaglandin E₂ production (Hammer et al., 2007), inhibition of inducible NO synthase (iNOS) expression (Tedeschi et al., 2003; Saad et al., 2008; Kraus et al., 2010) and tumor necrosis factor-α inhibition (Saad et al., 2008) have also been proposed to play a role.

In regard to cardiovascular system studies with HP, recent studies were only focused on the effect of Hypericum species on the contractile functions; diminished vascular contractility has been reported with histamine and prostaglandin F_2α (Melzer et al., 1991), as well as phenylephrine (Phe) and potassium chloride (KCl) (Apaydin et al., 2001; Khan et al., 2009). Khan et al. (2009) have reported that the hypotensive effect of the plant is dose dependent in the normotensive rats. Contrarily, acute exposure of over-the-counter HP has been reported at the web database, Micromedex 1.0 (http://www.thomsonhc.com) that hypertension is theoretically possible following an overdose.

In spite of the numerous studies carried out to understand the mechanism of the HP, studies establishing the effect of HP on the relaxant functions of smooth muscle or endothelial layer are limited. This study was performed to evaluate the vascular system function related to HP on the rat aorta.

Materials and methods

Experimental animals

Male Wistar rats (380–400 g) were obtained from Experimental Animal Center of Adnan Menderes University and all experiments were performed according to the principles and guidelines of the Animal Ethical Committee of Adnan Menderes University after approval by the Committee.

Chemicals and equipments

Phe hydrochloride, acetylcholine hydrochloride (Ach), sodium nitroprusside (SNP) and KCl were purchased from Sigma Chemicals (Interlab, Izmir, Turkey). After dissolving standardized dry extract of HP (Indena S.p.A; Italy) in 100 µl dimethylsulfoxide (DMSO), 900 µl distilled water has been added to prepare the stock solution. Then 100 µl stock solution has been applied into 20 ml tissue chamber. In this way, the concentration of DMSO was diluted to 0.05%.

Measurement of isometric force was recorded by a force transducer (MAY FDT 05, Commat Ltd. Ankara, Turkey) and a data acquisition system (MP 150, Biopac Systems, Inc., Commat Ltd., Ankara, Turkey).

Experimental design

Under the Ketamine and Xylasine (50 and 5 mg/kg, respectively) anesthesia, aortic rings of ~3 mm in length were mounted in 20 ml organ baths containing Krebs–Henseleit solution of the following composition (in mM): NaCl 118.1, KCl 4.56, CaCl₂ 1.22, MgSO₄ 1.22, KH₂P04 1.1, NaHCo₃ 25, d-glucose 10.1. The bath solution was maintained at 37 ± 1°C and constantly oxygenated with carbogen. After 60–90 min equilibration period under 2 g resting tension (Demirci et al., 2008), cumulative concentration response curves were obtained by adding increasing concentration of Phe (0.001–30 µM) or a single concentration of KCl (40 mM) in to the bath. Relaxant responses were determined by using cumulative concentrations of Ach (0.001–30 µM) or SNP (0.001–3 µM). Following 30 min incubation with 0.05 mg/ml HP, the responses to four pharmacologic agents mentioned above were reassessed. On these experiments, the response of each treated ring was compared with its own untreated control.

In the experiments using high KCl solution, the equimolar amount of Na⁺ was replaced by K⁺ to maintain constant ion strength.

Statistics

Data were presented as mean ± SEM. Variance analysis was used to determine the significance among groups, E_max and pD₂ values were assessed by using paired t-test. p Values <0.05 were considered significant.

Results

Contractile responses

Phe-induced (0.001–30 µM) contractile responses are illustrated in Figure 1 (n = 6). pD₂ and E_max values elicited by Phe and the tension developed to 40 mM KCl are shown in Table 1. The contractile responses to Phe and KCl were significantly decreased with HP intervention when compared with the untreated control (p < 0.05). Additionally pD₂ value of Phe was significantly diminished (p < 0.05).

Table 1.  Contractions (mg) to phenylephrine and KCl and relaxation (%) to acetylcholine and sodium nitroprusside in aortic rings with endothelium from control and HP treated animals.

		Phe	KCl	Ach	SNP
Control	Emax	1131.73 ± 164.11	745.58 ± 66.73	94.61 ± 2.65	108.82 ± 5.06
	pD2	7.19 ± 0.14		7.61 ± 0.26	8.35 ± 0.10
Hypericum	Emax	477.40 ± 39.94*	112.58 ± 26.58^†	87.79 ± 9.40	106.43 ± 7.45
	pD2	6.77 ± 0.07*		7.42 ± 0.21	8.32 ± 0.10

Values are mean ± SEM (n = 6).

*p < 0.05, ^†p < 0.001 control vs. Hypericum.

Phe, phenylephrine; Ach, acetylcholine; SNP, sodium nitroprusside.

Figure 1.  Effect of HP (0.05 mg/ml) on phenylephrine contraction in aortic rings with endothelium. *p < 0.05 control vs. Hypericum.

Relaxant responses

After precontracted with Phe, Ach (0.001–30 µM) (n = 6) and SNP (0.001–3 µM) (n = 6) induced relaxations in all groups (Figures 2 and 3, respectively). The pD₂ values and the maximum relaxations expressed as the percentage of Phe precontraction are given in Table 1. There were no significant differences between the E_max and pD₂ values to Ach and SNP in HP treated group.

Figure 2.  Effect of HP (0.05 mg/ml) on acetylcholine relaxation in aortic rings with endothelium.

Figure 3.  Effect of HP (0.05 mg/ml) on sodium nitroprusside relaxation in aortic rings with endothelium.

Discussion

Even though a number of studies have been conducted to understand the mechanism of HP treatment on various disease conditions (Albert et al., 2002; Tedeschi et al., 2003; De Paola et al., 2005; Genovese et al., 2006a,b; Hammer et al., 2007; Saad et al., 2008; Dost et al., 2009; Cirak et al., 2010; Kraus et al., 2010; Samadi et al., 2010) only a few of them have reported the effect of HP on contractility of vascular system (Melzer et al., 1991; Apaydin et al., 2001; Gilani et al., 2005; Khan et al., 2009, 2010), however, there are no scientific reports with HP detailing its modulatory effects on the vascular tone. Hence, this study has investigated the vascular effects of HP at both smooth muscle and endothelial level.

α-Adrenoceptor-1 agonist activation by Phe and high potassium induced membrane depolarization are widely used to determine the calcium status of the vascular smooth muscle in experimental studies (Karaki et al., 1997). α-Receptors coupled with G protein family and this coupling stimulates phospholipase C which hydrolysis phosphatidylinositol into inositol 1,4,5-triphosphate (IP3) and diacylglycerol. These secondary messengers alter the calcium movements and distribution in the cytoplasm, and calcium regulates contractile elements of the cell (Karaki et al., 1997). Our present finding that HP decreases Phe contraction is in agreement with previous results (Apaydin et al., 2001; Khan et al., 2009), suggesting that changes in the calcium homeostasis of the cell is responsible for the vascular responsiveness to HP. Additionally, Melzer et al. (1991) studied procanidin fractions of HP on isolated coronary arteries and reported that they prevent histamine and prostaglandin F2-α induced arterial contractions possibly through the inhibition of cellular phosphodiesterase (Gilani et al., 2005) that leads to the accumulation of muscle relaxant cyclic AMP, hence calcium sensitivity of the contractile elements is decreased (Karaki et al., 1997; Khan et al., 2010). High potassium induces contractions through the opening of voltage-dependent calcium channels (Karaki et al., 1997) and this phenomenon is assessed in this study as another good indicator of cellular calcium balance. We have shown that the response of KCl is significantly diminished, indicating that calcium influx also inhibited by HP. Calcium antagonistic features of the plant has also been shown on rat and human vas deferens (Capasso et al., 2005), rabbit jejunum and guina pig trachea (Gilani et al., 2005; Khan et al., 2009, 2010). Our results further supported existing reports suggesting that HP treatment attenuates the smooth muscle hyper-reactivity to various agents in vitro by altering calcium homeostasis.

It is well known that NO sythase (NOS) has three isoforms, neuronal (nNOS) and endothelial forms (eNOS) continuously involve in physiological functions, while inducible form (iNOS) exist on the inflammatory site as a response to several mediators (Herman & Moncada, 2005). Ach exerts its effect on endothelium and stimulates the release of NO which is produced by constitutive eNOS, and further results in vasorelaxation of the vascular smooth muscle in normal condition (Herman & Moncada, 2005; Moncada & Higgs, 2006). In our study, HP had no demonstrable effect on Ach induced relaxation in endothelium intact rings after short term incubation. This finding may be explained by the absence of inhibition of constitutive eNOS on rat thoracic aorta. Additionally, NO attenuates α-1 adrenoceptor mediated vasoconstriction (Klabunde et al., 2007). If the endothelium had been pharmacologically denuded, in other words, if NO production or bioavailability has been disrupted by HP, it might have seen an increased Phe contraction after HP intervention. This point also supports that the endothelium was kept intact throughout the study. Along with the controlling of the vascular tone, endothelium actively contributes the inflammatory process by synthesizing and releasing endothelial derived factors and shows close interaction with leukocytes and platelets. Another origin of NO is iNOS which is responsible for the detrimental effect of NO in the inflammatory process. Previous researchers have described the inhibition of iNOS by HP (Tedeschi et al., 2003; Saad et al., 2008; Kraus et al., 2010). Endothelial cell xantine oxidase forms ROS (Villar et al., 2008). A set of experiments has been done on xantine oxidase system of vascular tissue by using commercially available form of HP and this study has shown a dose–dependant ROS suppression (Hunt et al., 2001). Meanwhile, prostaglandin E₂ inhibitory effect of HP proposed the basic mechanism of anti-inflammatory action (Hammer et al., 2007). Albert et al. (2002) have reported that hyperforin is a supressor of another constitutive enzyme, COX-1, inhibiting the synthesis of COX products from platelets; this effect is dose dependent and HP can be 18-fold more potent than aspirin. Same group has also shown that HP inhibits 5-LO product as well. The studies clearly demonstrate that HP modulates the pathways of many vasoactive substances that are associated with endothelial functions without changing the eNOS.

NO-donor SNP directly activates soluble guanylyl cyclase and produces vascular smooth muscle relaxation via formation of cyclic guanosine monophosphate (cGMP) (Guimarães & Moura, 2001). This experiment clearly proved that HP treatment did not alter cGMP concentration in vascular smooth muscle.

Searching the literature reports on the plant, it has been noticed that the doses of Hypericum species used highly differs among studies range from 0.5 to 10.0 mg/mL (Gilani et al., 2005; Fronza et al., 2009; Khan et al., 2009). Relatively high dose of HP incubation has been chosen in the current study that such high dose exposure can be seen while the plant consumes daily without any consideration of its side effects. Although some evidence has been reported that >0.5 µg/mL concentrations of Hypericum oil is cytotoxic on Swiss 3T3 albino mouse fibroblasts (Fronza et al., 2009) and hyperforin, an extract of HP, shows collapses of mitochondrial membrane potential on cortical neurons (Tu et al., 2010), in our study we have not seen any detrimental effects of 0.05 mg/ml HP on endothelium.

Conclusion

This study has demonstrated that HP intervention induces hypo-reactivity to Phe and KCl in the rat aorta, in other words, the plant shows Ca antagonistic properties on vascular smooth muscle. It has also been shown for the first time that HP does not impair the constitutive NOS and NO release or cGMP pathway in the rat aorta.

Although tricyclic antidepressants might be responsible for serious cardiac effects, the uses of HP for the treatment of mild to moderate depression is a common practice of modern physicians and in folkloric medicine in many countries. The present observations lend pharmacological support to the safety of HP as an over-the-counter agent.

Acknowledgement

We wish to thank Indena S.p A. (Italy) for gifting us with standardized extract of Hypericum perforatum.

Declaration of interest

The research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.