In vitro vasorelaxation mechanisms of bioactive compounds extracted from Hibiscus sabdariffa on rat thoracic aorta

Background In this study, we suggested characterizing the vasodilator effects and the phytochemical characteristics of a plant with food usage also used in traditional treatment of arterial high blood pressure in Senegal. Methods Vascular effects of crude extract of dried and powdered calyces of Hibiscus sabdariffa were evaluated on isolated thoracic aorta of male Wistar rats on organ chambers. The crude extract was also enriched by liquid-liquid extraction. The various cyclohexane, dichloromethane, ethyl acetate, butanol extracts obtained as well as the residual marc were subjected to Sephadex LH-20 column chromatography. The different methanolic eluate fractions were then analyzed by Thin Layer (TLC) and High Performance Liquid Chromatography (HPLC) and their vascular effects also evaluated. Results The H. Sabdariffa crude extract induced mainly endothelium-dependent relaxant effects. The endothelium-dependent relaxations result from NOS activation and those who not dependent to endothelium from activation of smooth muscle potassium channels. The phytochemical analysis revealed the presence of phenolic acids in the ethyl acetate extract and anthocyans in the butanolic extract. The biological efficiency of the various studied extracts, in term of vasorelaxant capacity, showed that: Butanol extract > Crude extract > Residual marc > Ethyl acetate extract. These results suggest that the strong activity of the butanolic extract is essentially due to the presence of anthocyans found in its fractions 43-67. Conclusion These results demonstrate the vasodilator potential of hibiscus sabdariffa and contribute to his valuation as therapeutic alternative.


Background
Cardiovascular pathologies complications (myocardial infarction, stroke...) constitute one of the most important causes of mortality and morbidity in the world [1][2][3]. These complications, often facilitated by arterial high blood pressure, appear among the main causes of death in Africa. Indeed, according to World Health Organisation (W.H.O) experts, high blood pressure and hypercholesterolemia are more frequent in the developing countries than believed. Among the risk factors, except hypercholesterolemia, obesity, smoking addict and diabetes constitute the major contributing factors of these diseases [4]. A future scenario by the W.H.O. reveals a negative trend due to an increase in the rate of morbidity and mortality especially in Emerging Countries [1]. Considering the gravity and the frequency of these conditions, a search for compounds having vascular benefits is intensively pursued [5]. The interest of researchers in the whole world for these compounds encouraged us to study the healing plants of the Senegalese pharmacopoeia. Indeed, an ethnobotanical investigation led by our laboratory had listed several healing plants with antihypertensive potential among which, Hibiscus sabdariffa L. In the Senegalese pharmacopoeia, H. sabdariffa is one of the most-often used plants in the traditional treatment of high arterial blood pressure. Previous studies led by numerous groups of researchers [6][7][8][9][10][11][12][13][14] had already reported scientific proof of the antihypertensive effects traditionally attributed to H. sabdariffa. If these studies allowed demonstrating the therapeutic potential of this plant, so in vitro as in vivo, the underlying mechanisms involved as well as the phytochemical compounds responsible for these effects were not fully documented. So the objective of this study was to contribute to the understanding of such mechanisms and the discovery of bioactive substances responsible for vascular effects of H. sabdariffa. By combining technical preparation (extraction, enrichment, fractionation) and phytochemical characterization (TLC, HPLC) combined with biological characterization methods (organ bath), we strived to identify the phytochemical compounds and estimate their vasorelaxant effects.

Methods
Organic extract preparation H. sabdariffa calyces was obtained from the Tilène market (Dakar). Calyces were dried during a week at room temperature, to avoid the risks of mold formation because of the relative humidity of the plant, and also to facilitate its conservation and its use during the grinding. Dried and powdered calyx (Grinder RM-100, Retsch ® ) of Hibiscus sabdariffa (500 g) was extracted by maceration at room temperature for 2 hours with 60% methanol. The hydroalcoholic extract was then filtered in vacuum conditions (Vacuum pump V-700, Büchi ® ) by means of the phial of Kitassato and evaporated on a rotary evaporator (Rotavapor R-210, Büchi ® ). Methanolic extract evaporation was realized during three successive days until the obtaining of a dry crude extract (136.7 g). Evaporation conditions were as follows: Temperature: +40°C; Cooling: +21°C; Rotation: 4000 tr./min. The methanolic extract, when not evaporated at once, went through those stages of separation with cyclohexane, dichloromethane, ethyl acetate and butanol to end up as an enriched extract after two hours of decantation. It is repeated as often as needed with new solvent until exhaustion (colorless organic phase). The various liquid organic extracts (cyclohexanic, 1.42 g; dichloromethanic, 2.53 g; ethyl acetate, 34.85 g; butanolic, 18.97 g and the residual marc, 79.01 g) were then washed with anhydrous sodium sulphate (Fischer ® ) to fix some residual water, and then filtered.

Organ bath experiments
Experiments were conducted in accordance with the Guide for the Care and Use of Laboratory Animals as promulgated by the Senegalese authorities.
Male Wistar rats weighing 150-200 g were procured from a local Institute (Faculté des Sciences et Techniques, Dakar, Senegal). They were fed on standard rat feed and given free access to water. Thoracic aorta were removed from rats after anaesthesia with pentobarbital (60 mg/kg, i.p.) and cleaned of connective tissue and cut into rings (3-4 mm in length). As indicated, the endothelium was removed by rubbing the intimal surface of rings with a pair of forceps.

Phytochemical analysis -Enriched extracts fractionation
Enriched extracts were fractionated by liquid Chromatography on Lipophilic Sephadex LH20 ® (Sigma-Aldrich) according to the following protocol: 40 g of Sephadex LH20 are conditioned with methanol 20% in a glass column of 2,3 cm diameter provided with a faucet. The flow was adjusted in 32 drip/min. 1,5 g of extract were dissolved in 5 ml of methanol and deposited on the surface of the frost.

-TLC-fingerprint and HPLC analysis
For the TLC analysis, extracts were dissolved in the migration solvent of the ethyl acetate/icy acetic acid/formic acid/water mixture (100:11:11:26). 10 μl of reference solutions and samples (1 mg/mL) were applied to the TLC plate. At the end of the migration, TLC Plates were dried and phytochemical compounds observed under natural light or after revelation by the NEU reagent (= 1% of diphenylboryloxyethylamine in methanol) and observation under UV light in 366 nm. Interpretation of the various chromatograms was made on the basis of those presented in Plant Drug Analysis: 6 am. Wagner, S Bladt (1996) [15]. Fluorescence can be to interpret in the fol-lowing way: blue: phenolic Acids; yellow -Orange: Flavonols; Yellow -Green: Flavones.
With the aim of confirming the chemical composition of the crude extract and determining that of enriched extracts, we proceeded to an HPLC analysis. For that purpose, we used pure reference substances (chlorogenic acid, phenolic acid, delphinidin, cyanidin, etc.) of retention time and length of detection known, for the determination of the phytochemical profile of our various extracts. Extracts were examined in the following conditions: Mobile phase in gradient mode constituted by the mixture anhydrous trifluoroacetic acid 0.1% and acetonitril; debit: 1 ml/min; column C18 (EC 250/4.6 Nucleodur 100-10 C 18 ec); Diode array detector between 191 and 700 nm; injection volume: 10 μL.

Materials
Unless otherwise indicated, drugs were purchased from Sigma Chemical Co or Aldrich (Saint Quentin-Fallavier, France). Norepinephrine (MISR CO) was a generous gift from 'Pharmacie Nationale d'Approvisionnement', Dakar, Senegal). Methanol, butanol, acetic acid, cyclohexane and dichloromethane solutions were purchased from Fischer Scientific.

Statistical Analysis
Values are expressed as mean ± SEM. Statistical evaluation was performed with Student's t test for paired data or ANOVA. Values of p < 0.05 were considered statistically significant.

Regulatory mechanisms of the hibiscus sabdariffa crude extract-induced relaxation
In order to characterize mechanisms involved in the relaxing effects of hibiscus sabdariffa, we conducted vascular reactivity experiments using isolated rat thoracic aortic rings treated or not with inhibitors.
As the endothelium was strongly involved in the observed relaxations, it was necessary to study the role of NO-Synthase (NOS) and Cyclooxgenase (COX), two major enzymes responsible for the release of relaxing factors in vascular beds. Using L-NAME and Indomethacin, two respective inhibitors of these enzymes, our results indicate that only NOS is activated after administration of the crude extract. Indeed, Figure 2A shows that L-NAME significantly reduced the relaxations, whereas Indomethacin does not, suggesting a possible stimulation of NO-sGC-cGMP signaling pathway by the H. sabdariffa crude extract.
Interestingly, as shown in figure 2B, the NO scavenger oxyhemoglobin and the soluble guanylate cyclase inhibitor methylene blue significantly reduce the effect of sGC activation after administration of the H. sabdariffa extract, leading to a decrease of the observed relaxations.

-NOS-NO-sGC pathway activation
The NO pathway was strongly involved in the relaxation induced by the crude extract of H. sabdariffa. An interesting question was how this pathway is activated. Activation of the PI3-kinase/akt pathway leads to phosphorylation of eNOS, as reported by numerous studies [16][17][18][19]. Moreover, cell-derived reactive oxygen species (ROS), when present in biological media at physiological concentrations can activate this pathway [20][21][22]. Our results, as shown in Figure 2C, show that wortmanin, which specifically inhibits Phosphatidyl-inositol-3-kinase (Pi3-K), as well as the SOD mimetic MnTMPyP, were found to reduce significantly the relaxations obtained with the crude extract. It is suggested that activation of the lipid kinase PI3K participate as major regulators in the NOS-initiated cascades of vasorelaxation induced by h. sabdariffa extract.

-Potassium channels activation
Relaxations obtained with the crude extract in vessels without endothelium, even if they are significantly lower compared with those observed in vessels with intact endothelium, led us to think a direct relaxing effect of this extract on vascular muscles. A likely mechanism is an hyperpolarization after direct activation of potassium channels. This has been verified by the non-selective inhibitor of potassium channels, barium chloride (BaCl 2 ). Indeed, after treatment of vessels with this inhibitor, we observed a significant reduction of relaxations both in vessels with endothelium ( Figure 2D), than in   The values indicate the mean ± SEM of the EC 50 and the E max obtained from 6-8 experiments.
those without endothelium (data not shown). Moreover, our results also show that K + -ATP-dependent channels are not responsible for the endothelium-independent relaxation, as Glibenclamide, considered as a selective inhibitor of these channels does not significantly alter relaxations.

Vascular relaxing effects of the various enriched extracts studied
Since the relaxations observed with the crude extract are less than those observed with acetylcholine or sodium nitroprusside, it was necessary to make enriched extracts in order to improve the vasorelaxations. Figure 3A shows that the ethyl acetate extract causes a vasorelaxation significantly less important than the crude extract taken as reference. On the other hand, the results obtained with the butanolic extract show a vasorelaxation significantly more important than those of the crude extract. However, residual marc leads a vasorelaxation not significantly different from those of the crude extract. The biological efficiency of the various studied extracts in terms of vasorelaxant capacity was appreciated on the basis of the EC 50 and of the maximal effect (E max ), As indicated (Table 1 and Figure 3B), the butanolic extract presents a vasorelaxant potential more important than the other extracts. It is important to note that cyclohexanic and dichloromethanic extracts were not characterized because of their very weak return on extraction. Furthermore, these solvent allow to get rid of constituent's generally unwanted fats, chlorophylls and by-products. TLC-fingerprinting of the crude extract of hibiscus sabdariffa

Phytochemical analysis of the various studied extracts -TLC fingerprint and HPLC analysis of the crude extract
To verify the presence of polyphenolic compounds whose vasorelaxant effects have already been the subject of numerous studies, we proceeded to a TLC-fingerprint analysis of the H.    Table 3) and of anthocyans detected in 342 nm in the butanolic extract ( Figure 8, table 4). Finally, the residual marc was not the object of an HPLC analysis because its vasorelaxant capacity is similar to that of the crude extract.

-TLC fingerprint of the butanolic extract
The biggest vasorelaxant capacity of the butanolic extract and its wealth in anthocyans led to us to fractionate this extract with the aim of identifying its compounds. Results (figure 9) show that only fraction 43-67 of the butanolic extract is rich in anthocyans compared with the other fractions which contain all polyphenolic compounds, in particular phenolic and chlorogenic acids or flavonoids.

Discussion
The main results of this study demonstrate and confirm the relaxing effect of hibiscus sabdariffa extracts, especially on the isolated rat aorta. But even more interesting, they have helped characterize the possible mechanisms involved in vasorelaxation while highlighting the link between this effect and responsible phytochemical compounds.
In terms of vasorelaxation effects, analysis of our results shows that hibiscus sabdariffa effects are strongly endothelium-dependent and involve stimulation of NOS enzyme by the Pi3-K/Akt pathway. Indeed, the dominant  role of the endothelium in vessel relaxation by plant polyphenols has already been demonstrated in numerous works [20,[22][23][24][25][26][27][28][29][30][31][32]. Our results also are in agreement with data obtained from red wine polyphenolic compounds (RWPC) which activate enzymes involved in the release of endothelial relaxant factors including eNOS [27,33,34]. Moreover, our results are also in agreement with the idea that RWPC promotes the release of endothelial NO through a redox sensitive PI3/Akt pathway [22]. Finally, these results also contrast with one of our previous study [21] where it was interesting to note that the mechanism by which cognac polyphenolic compounds (CPC) enhances NO production does not involve redox system.
Data obtained with this previous study demonstrate that CPC is able to directly increase NO production without affecting superoxide anions and enhances the bradykinininduced NO production in human endothelial cells.
Our results also show a non-endothelium-dependent relaxation induced by the h. sabdariffa extract, which has not been the case for most other types of polyphenolic extracts. The likely mechanism of this non-endotheliumdependent relaxation is a direct smooth muscle activation. Indeed, H. sabdariffa as shown by our results, can also relax blood vessels without endothelium, and it was also admitted that the endothelium hyperpolarized factor Phytochemical profile of the ethyl acetate extract after HPLC analysis and detection in 270 nm   Our study also showed that on adrenalin-precontracted isolated aortic rings, the crude hydro-alcoholic extract of Hibiscus sabdariffa induced a vasorelaxation. This relaxation is dose-dependent. It reaches a value of 66,57% ± 8,07 for the maximal concentration administrated i.e. 10 -1 mg/ml. These results concur with those of Ajay et al. [9] who showed that for a concentration of 1 mg/ml, he noted a maximal relaxation of 86% ± 4,84. However it is to note that they worked in vitro with a model of spontaneously hypertensive rats while we worked with normal rats.
The observed vasorelaxation is more important with the butanolic extract for which the maximal effect is about 94,3 ± 0,97% at a concentration of 10 -1 mg/ml in comparison with the maximal relaxation observed with the crude extract. However, the effect of the residual extract is less important than that of the butanolic and the crude extracts; and that of the ethyl acetate extract is even less. In light of these results, it appears that our study confirmed that the hydro-methanolic total extract of the dried calyces of H. sabdariffa possess an important vasorelaxant activity; and that the enriched butanolic extract of the dried calyces possess an vasorelaxant activity even more important. Numerous studies reported the presence and the nature of some anthocyans of H. sabdariffa [10,36,41,[45][46][47][48][49][50]. As for the Senegalese variety of H. sabdariffa, we note especially cyanidin by-products such as cyanidin-3-monoglucoside, TLC-fingerprinting of the various fractions of the butanolic extract of hibiscus sabdariffa cyanidin-3,5-diglucoside, cyanidin-3-sambubioside [51]. After analysis, it seems that the strong vasorelaxant activity of the butanolic extract, compared with the other extracts, would be due for many to the presence of these anthocyans in his fraction 43-67. Finally, the weak quantities of extraction of this fraction did not allow us to test it on isolated rat thoracic aorta.
To the best of our knowledge, such results (the link between the vasorelaxant property and the anthocyans present in H. Sabdariffa) have not been reported in the literature. They thus constitute one of the originalities of our work which is to be continued, to characterize and isolate these anthocyans as well as their molecular mechanisms in the induced vasorelaxation.

Conclusion
Hibiscus sabdariffa could be an alternative in the care of vascular diseases in our countries, considering its low cost and its availability. It is also necessary to emphasize the preventive role that Hibiscus sabdariffa could play. This is all the easier to realize as it is about a plant known, and currently used, by local populations. It would be necessary to make the populations aware of these virtues and to encourage its consumption.