Nitric-Oxide-Mediated Vasodilation of Bioactive Compounds Isolated from Hypericum revolutum in Rat Aorta

Simple Summary Hypericum revolutum (HR) is reported to produce vasodilating activity in phenylephrine-precontracted aortae, where the chloroform fraction is the most potent. Chemical investigation of this fraction yielded two new compounds, revolutin (1) and hyperevolutin C (2), along with three known metabolites, β-sitosterol (3), euxanthone (4), and 2,3,4-tirmethoxy xanthone (5). Isolated compounds 1, 2, 3, and 5 produce vasodilation activities that are dependent on endothelial nitric oxide release. Abstract Vasodilators are an important class in the management of hypertension and related cardiovascular disorders. In this regard, the chloroform fraction of Hypericum revolutum (HR) has been reported to produce vasodilating activity in phenylephrine-precontracted aortae. The current work aims to identify the active metabolites in the chloroform fraction of HR and illustrate the possible mechanism of action. The vasodilation activities were investigated using the isolated artery technique. NO vascular release was assessed by utilizing the NO-sensitive fluorescent probe DAF-FM. Free radical scavenging capacity was assessed utilizing DPPH. Chemical investigation of this fraction yielded two new compounds, revolutin (1) and hyperevolutin C (2), along with three known metabolites, β-sitosterol (3), euxanthone (4), and 2,3,4-tirmethoxy xanthone (5). Compounds 1, 2, 3, and 5 showed significant vasodilation activities that were blocked by either endothelial denudation or L-NAME (nitric oxide synthase inhibitor), pointing towards a role of endothelial nitric oxide in their activities. In confirmation of this role, compounds 1–3 showed a significant release of NO from isolated vessels, as indicated by DAF-FM. On the other hand, only compound 5 showed free radical scavenging activities, as indicated by DPPH. In conclusion, isolated compounds 1, 2, 3, and 5 produce vasodilation activities that are dependent on endothelial nitric oxide release.


Introduction
Elevated blood pressure is a serious disorder that underlies other cardiovascular diseases and is a direct complication of metabolic disorders such as diabetes and metabolic syndrome. Hypertension can be due to increased heart stimulation or, most likely, increased peripheral resistance and endothelial dysfunction [1]. Endothelial dysfunction has a crucial role in the progression of hypertension by affecting vascular relaxation and constriction. The endothelium-dependent vasodilatation regulatory system controls vascular function

Plant Material Extraction
Dried powdered aerial parts (1 Kg) were extracted at room temperature with 5 L methanol (four times), utilizing Ultraturrax, until exhausted. The under-vacuum concentration of the total extract produced a brown residue (40 g). The residue was mixed with 200 ml of water, partitioned with 500 ml chloroform (4 times), and vaporized to furnish Fraction I (10 g).

Animals
Seven-week-old male Wistar rats (180-200 g) were used (King Fahd Medical Research Center, KAU, KSA). The animals were housed with access to standard rodent pellets and purified water in clear polypropylene cages (4 rats each). Constant housing conditions were applied, including alternating 12 h light and dark, 22 ± 3 • C temperature, sufficient ventilation, and 50-60% relative humidity. The research ethics committee of King Abdulaziz University approved the study (approval number 126-1439). The study was carried out according to the Saudi Arabia Research Bioethics Guidelines and Regulations, which are in accordance with the Animals in Research: Reporting In Vivo Experiments (ARRIVE) guidelines for research involving animals [15]. The animals were executed by decapitation using a rodent guillotine administered by qualified personnel in the animal housing. The method is acceptable to induce a rapid loss of consciousness, according to the AVMA Guidelines for the Euthanasia of Animals: 2020 Edition (section M3.7) [16], and the descending thoracic aorta was precisely removed and washed from connective tissues and fats.

Evaluating the Chloroform Fraction and Isolated Metabolites' Direct Relaxant Effect
Vasodilating capacities were assessed using the isolated artery method, as formerly reported [17,18]. Briefly, the aorta was removed, cleansed of any connective tissue and fats, and sliced into rings (3 mm). Each ring was hung in Krebs Henseleit buffer channels (4.8 mM KCl, 118 mM NaCl, 1.2 mM MgSO 4 , 1.2 mM KH 2 PO 4 , 2.5 mM CaCl 2 , 11.1 mM glucose, and 25 mM NaHCO 3 ) at 37 • C, with continuous aeration with gas (5% CO 2 and 95% O 2 ). Every 30 min, the channel buffer solution was exchanged. Quantification of the aortic tension was accomplished using an isometric force transducer, and the results were presented through a PowerLab data interface module linked to a PC running Chart software v8 (ADI Instruments).
The aortic rings were set aside for 30 min for equilibration at a 1500 mg ± 50 resting tension. Initial aorta contraction and relaxation were then carried out by the addition of PE, followed by ACh (both at 10 µM). After the tension was reverted to the rest state, accumulative concentrations of 1-10 µg/mL and 1-10 µM for the chloroform fraction or the pure metabolites, respectively, were added to the organ bath precontracted (PE, 10 µM)isolated aortae. Tension reduction was estimated as a measurement of vasodilating actions. In other sets of experiments for investigating the role of the endothelium in the vasodilating effect, it was mechanically made bare. Additionally, L-NAME (100 µM) was added in the organ bath 15 min before adding the isolated metabolites or chloroform fraction to explore the role of nitric oxide in the vasodilating influence on different sets of experiments.

Examining the Effect of Chloroform Fraction and Isolated Metabolites on Nitric Oxide (NO) Production
The separated aorta intracellular induction of NO production by the tested metabolites and the chloroform fraction was examined employing the DAF-FM fluorescence probe, as similarly outlined in our former work [19]. Similar to the earlier technique, the thoracic aorta was removed, the fats were washed off, and it was sliced into approximately 6 mm pieces. Every piece was put in a 96-well black plate separate well, maintained in dim light, having made a 2.5 µM DAF-FM/KHB mixture (37 • C) immediately before starting the procedure; 100 µL were accurately drawn and transported to the wells' neighboring column; after that, ACh (10 µM) was included in one of the aortic segments, and the chloroform fraction (10 µg/mL) or separated metabolites (10 µM) was put in the other wells after 3 min. Again, after 3 min, 100 µL were taken from the wells with the aortae and transmitted to the nearby well columns. For the blank, a row without an aorta was retained for each ACh concentration, which was handled identically. The withdrawn volumes' fluorescence intensity (and not the column including the aortae) was then assessed at λem = 515 nm and λex = 485 nm using a SpectraMax ® M3 Monochromator plate reader.

Studying the ROS Scavenging Potential of the Isolated Metabolites and the Chloroform Fraction
ROS scavenging potential was assessed, as formerly stated in previous work from our laboratories [20]. In a 96-well clear plate, the pure metabolites (1-10 µM) or Fraction I (1-10 µg/mL) in MeOH was added to a DPPH (240 µM) solution in MeOH/tris (1:1 v/v). For the control (C), MeOH was utilized instead of the fraction or metabolites. DPPH was directly prepared before adding to the plate. The absorbance was estimated every minute at 520 nm for 10 min using a SpectraMax ® M3-Monochromator plate reader.

Statistical Analysis
Experimental values are depicted as the mean ± standard error of the mean (SEM). For statistical analysis, one-or two-way ANOVA (analysis of variance) was applied, as designated in the figure legends, succeeded by Dunnett's posthoc test utilizing GraphPad Instat software version 5. If p < 0.05, the differences were recognized as significant.

Compound 2
Compound 2 was a yellow amorphous solid, having two pseudo-molecular ion peaks at m/z 537.2821 (C 30  Its HSQC and 13 C (Figures S8 and S9) spectra exhibited 3 methylenes, 10 methyls, 6 methines (from which there were one oxymethine and two olefinics), and 11 quaternaries comprising 4 carbonyls. 1 Figure S10) of H-25 to C-27, C-24, and C-6 secured the connection of C-6 and C-26, C-24, and C-25 and affirmed the linking of the 1-oxo-4-methyl-pent-3-eneyl moiety at C-6 ( Figure 3b). Further, the linkage of γ,γ-dimethyl allyl at C-25 was assured by the relations of H-26 to H-29 and H-28. Furthermore, the cross-peaks between H-5/C-7 and C-24 assured the linkage between C-5, C-6, and C-7. The cross-peaks of H-16/C-18, C-15 and C-17, H-14 and H-13/C-8 and C-6, and H-16/C-8, besides relation H-6/C-8, assured the connection between C-8/C-7 and C-7/C-6 and secured the location of the 1-oxo-2-methyl propyl moiety at C-8. The correlations of H-4/C-4a, C-9a, C-10, C-5, and C-2, and H-12 and H-11/C-2 in HMBC proved the 3,4-dihydropyran moiety and the connection between C-10 and C-4a. This evidence revealed the fusion of the pyran ring at C-4a and C-9a. The linkage of the γ,γ-dimethyl ally group to C-9a via a C-19 oxymethine was assured by the  (Figure 3c). Therefore, the structure of 2 was elucidated as represented and named hyperevolutin C. The nomenclature of 2 was given based on its structural similarity to hyperevolutins A and B, which were previously separated from H. perforatum [11]. downfield shift of C-3/HC-3 established the existence of a chloride at C-3. Further, the nuclear Overhauser effect spectroscopy (NOESY) spectrum of 2 ( Figure S12 (Figure 3c). Therefore, the structure of 2 was elucidated as represented and named hyperevolutin C. The nomenclature of 2 was given based on its structural similarity to hyperevolutins A and B, which were previously separated from H. perforatum [11].

Chloroform Fraction Vasodilating Activity
The chloroform fraction exhibited reductions in tension and consequent concentration-dependent vasodilation of PE (10 μM)-precontracted separated aortae that approached statistical significance compared with time control (at conc. 3 and 10 µ g/mL, both at p < 0.05). Removal of the endothelial layer (aorta denudation) prevented the chloroform fraction vasodilating effect, as obvious from the considerable prohibition (at conc. 1, 3, and 10 µ g/mL, all at p < 0.05). Additionally, L-NAME (1 mM) blocked the vasodilating effect of the chloroform fraction, as manifest from the remarkable inhibition (at conc. 1, 3, and 10 µ g/mL of the choroform fraction, all at p < 0.05) (Figure 4).

Vasodilating Activity of the Isolated Compounds
The compound 3 addition resulted in concentration-dependent vasodilation of PE (10 µM)-precontracted separated aortae ( Figure 5C). At a concentration of 10 µM, the vasodilating effect of compound 3 attained a statistically significant level (p < 0.05) compared with time control. This effect was entirely blocked by L-NAME or endothelial denudation, as apparent from the notable prohibition (both at conc. 3 and 10 µM, all at p < 0.05; Figure 6A). Nevertheless, compound 4 did not possess any notable vasodilation of PE (10 µM)-precontracted separated aortae. Likewise, L-NAME or endothelial denudation had no significant influences ( Figure 5D). Figure 5E demonstrated that the compound 5 addition gave rise to a decline in tension and, consequently, concentration-dependent vasodilation of PE-precontracted separated aortae. This effect at 10 µM attained a statistically significant (p < 0.05) level. Moreover, this effect was inhibited by endothelial denudation (at conc. 10 µM) or L-NAME (at both conc.3 and 10 µM, all at p < 0.05; Figure 5E).

Effect on Vascular NO-Production
The addition of 10 µM Ach at 37 • C to the aortic rings caused a marked NO production (p < 0.05) in comparison to control that was quantified and detected using 2.5 µM DAF-FM reagent. The chloroform fraction addition (conc. 10 µg/mL) brought about a similar increase in NO production, as evident by a notable increase in the fluorescence of DAF-FM compared with the control values (p < 0.05). The isolated metabolites 1-3 provoked NO production that attained a statistically noteworthy level (all at p < 0.05) at a 10 µM concentration ( Figure 6). nificant influences ( Figure 5D). Figure 5E demonstrated that the compound 5 addition gave rise to a decline in tension and, consequently, concentration-dependent vasodilation of PE-precontracted separated aortae. This effect at 10 µ M attained a statistically significant (p < 0.05) level. Moreover, this effect was inhibited by endothelial denudation (at conc. 10 µ M) or L-NAME (at both conc.3 and 10 µ M, all at p < 0.05; Figure 5E).

Effect on Vascular NO-Production
The addition of 10 µ M Ach at 37 °C to the aortic rings caused a marked NO production (p < 0.05) in comparison to control that was quantified and detected using 2.5 µ M DAF-FM reagent. The chloroform fraction addition (conc. 10 µ g/mL) brought about a similar increase in NO production, as evident by a notable increase in the fluorescence of DAF-FM compared with the control values (p < 0.05). The isolated metabolites 1-3 provoked NO production that attained a statistically noteworthy level (all at p < 0.05) at a 10 µ M concentration ( Figure 6).

Free Radical Scavenging (FRS) Capacities
In the 10 min reaction results between 240 µ M DPPH, the chloroform fraction, and compounds 1-5 (conc. 1, 3, and 10 µ M), only 5 exhibited remarkable FRS activity. The other compounds, 1-4, did not have any significant FRS activities (data not shown). Figure  7 revealed that 5 (conc. 10 µ M) possessed DPPH free radical scavenging activity that was translated into an antioxidant potential, which is clear from the noticeable variations from control, beginning from the 1st minute until the 10th minute (p < 0.05).

Free Radical Scavenging (FRS) Capacities
In the 10 min reaction results between 240 µM DPPH, the chloroform fraction, and compounds 1-5 (conc. 1, 3, and 10 µM), only 5 exhibited remarkable FRS activity. The other compounds, 1-4, did not have any significant FRS activities (data not shown). Figure 7 revealed that 5 (conc. 10 µM) possessed DPPH free radical scavenging activity that was translated into an antioxidant potential, which is clear from the noticeable variations from control, beginning from the 1st minute until the 10th minute (p < 0.05). 0.05) by one way-ANOVA and Newmans-Keuls posthoc test.

Free Radical Scavenging (FRS) Capacities
In the 10 min reaction results between 240 µ M DPPH, the chloroform fraction, and compounds 1-5 (conc. 1, 3, and 10 µ M), only 5 exhibited remarkable FRS activity. The other compounds, 1-4, did not have any significant FRS activities (data not shown). Figure  7 revealed that 5 (conc. 10 µ M) possessed DPPH free radical scavenging activity that was translated into an antioxidant potential, which is clear from the noticeable variations from control, beginning from the 1st minute until the 10th minute (p < 0.05).

Discussion
Natural products can exert vasodilation through either the endothelium (NO generation) or by acting on smooth muscle (Ca + and K + channels) [33]. The current work represents the first evaluation of the bioactive compounds from H. revolutum that are responsible for vasodilation activities; we also investigate the mechanism of action. A previous report from our laboratory [14] proved that the total methanol extract of H. revolutum gave rise to concentration-dependent vasodilation of phenylephrine-precontracted isolated aortae. The bio-guided fractions indicated that the chloroform fraction is accountable for the noticed total extract vasodilation potential. In this regard, similar vasodilating activities have been reported for other plant extracts; the methanol extract of Garcina mangostana as well as Mentha longifolia were reported to produce a direct vasorelaxant effect in phenylephrine-induced vasoconstriction and in an experimental model of angina, respectively [34,35]. Moreover, different phytoconstituents are known for their vasodilation activity. Phenolic compounds are the most important class of vasodilators. For example, flavone can exert vasodilation by acting on the Ca + channel and NO generation [33].
Compounds 1, 2, and 3 show significant vasodilation activities that were blocked by endothelial denudation (Figure 8). This points to the key role of the endothelium in mediating their vasodilating activities. In addition, L-NAME (NO synthase inhibitor) completely blocked the mentioned compounds' activities, suggesting endothelial nitric oxide stimulation as a major mechanism of activity. It is well established that the endothelium has a crucial role in controlling arterial tone via the release of the key vasorelaxant molecule NO [36]. In order to confirm that endothelial NO stimulation is the main vasodilating mechanism of compounds 1, 2, and 3, the endothelial release of NO upon compound addition was measured by the NO-sensitive fluorescent probe, DAF-FM. The current study confirms the release of NO from the vascular endothelium upon the addition of compounds 1, 2, and 3.
has a crucial role in controlling arterial tone via the release of the key vasorelaxant molecule NO [36]. In order to confirm that endothelial NO stimulation is the main vasodilating mechanism of compounds 1, 2, and 3, the endothelial release of NO upon compound addition was measured by the NO-sensitive fluorescent probe, DAF-FM. The current study confirms the release of NO from the vascular endothelium upon the addition of compounds 1, 2, and 3. Revolutin (1) belongs to the phloroglucinol group of compounds, which are known for their ability to improve NO generation. Previous reports have reported its ability to increase NO levels, leading to a decrease in blood pressure in vivo [37]. Meanwhile, hyperevolutin C (2) is a novel terpenoid structure closely related to garcinielliptone G, which was previously isolated from Garcinia subelliptica [38] but not previously tested for its biological effects. β-Sitosterol (3) has previously exhibited hepatoprotective and cardiopro- Revolutin (1) belongs to the phloroglucinol group of compounds, which are known for their ability to improve NO generation. Previous reports have reported its ability to increase NO levels, leading to a decrease in blood pressure in vivo [37]. Meanwhile, hyperevolutin C (2) is a novel terpenoid structure closely related to garcinielliptone G, which was previously isolated from Garcinia subelliptica [38] but not previously tested for its biological effects. β-Sitosterol (3) has previously exhibited hepatoprotective and cardioprotective effects in CdCl 2 -induced hypertensive rats [39]. Euxanthone (4) was previously investigated and showed a pronounced vasodilator effect through the release of endothelial factors such as NO and COX-derived factors. Additionally, it provoked the prohibition of a Ca +2 -sensitive mechanism initiated by protein kinase C instead of repression of a contraction-dependent release of the intracellular Ca + stores or prohibiting voltage-operated Ca + channels [40].
Free radical scavenging (FRS) activity is another important way to preserve the released NO from quenching by superoxides and subsequent conversion into nitrites or nitrates. Only compound 5 among the tested metabolites 1-5 has substantial FRS capacities. We take into consideration that compound 5 showed moderate vasodilation that was endothelial-dependent and inhibited by L-NAME but was not associated with NO generation. These data suggest preserving NO bioavailability rather than stimulating NO generation as the major mechanism of action of compound 5. In the meantime, we cannot exclude the possibility that compound 5 may stimulate endothelium-derived hyperpolarizing factors or other vasodilators generated in the endothelium, such as prostacyclin. While compound 5 has not been previously investigated for its vasodilator effect, the xanthone group of compounds was reported to show vasorelaxant and antihypertensive activities [41]. While previous reports have shown the ability of flavonoids and benzophenone nuclei to enhance vasodilatation through NO production [35,42], our study is the first to introduce a similar effect for the phloroglucinol nucleus. Additionally, our study is the first to report the vasodilation activity of the xanthone nucleus through the inhibition of NO degradation. This finding is significant in terms of increased drug potency when NO production is intended. Interestingly, El-bassossy et al. [43] have reported a similar NO-protective mechanism by heme oxygenase-1. It is noteworthy to mention that the chloroform fraction of H. revolutum showed the highest potency relative to each isolated compound, which can be attributed to the synergistic effect of the bioactive compounds. The proposed pharmacological mechanism is illustrated in Figure 9.
to report the vasodilation activity of the xanthone nucleus through the inhibition of NO degradation. This finding is significant in terms of increased drug potency when NO production is intended. Interestingly, El-bassossy et al. [43] have reported a similar NO-protective mechanism by heme oxygenase-1. It is noteworthy to mention that the chloroform fraction of H. revolutum showed the highest potency relative to each isolated compound, which can be attributed to the synergistic effect of the bioactive compounds. The proposed pharmacological mechanism is illustrated in Figure 9.  The main limitation of this study is exploring vasodilation only in the thoracic aortic model; other arteries such as cerebral and abdominal arteries were not investigated. Additionally, the study concentrates only on NO from the endothelium as the main mechanism; other mechanisms such as Ca and K channels need to be investigated. Additionally, vasodilator activity was proven only in vitro; therefore, in vivo studies on animals are required, in detail, to assess the toxicity of these compounds as well as their metabolites and their effects on blood vessels. Finally, planning to assess the activity of the isolated compounds in humans should be a final step after the detailed study of these compounds.

Conclusions
The phytochemical investigation of the chloroform fraction of H. revolutum yielded two new compounds that were identified as revolutin (1) and hyperevolutin C (2), together with three known metabolites (3)(4)(5). Compounds 1-3 and 5 showed significant vasodilation in isolated aortae. The observed vasodilation of compounds 1-3 seems to be mediated via NO generation, as blocked by endothelial removal and L-NAME, and approves DAF-FM NO release. Compound 5 vasodilation is thought to be mediated by its free radical scavenging activities that protect the released NO from quenching by superoxides. Due to the multifactorial nature of cardiovascular diseases such as hypertension, knowing the mechanisms of the vasodilation action of these compounds is a crucial element for developing and planning different therapeutic strategies. Concretely, the observed vasodilation ability of these metabolites may reveal their potential therapeutic use against high-blood-pressure-related cardiovascular diseases.

Patents
This work resulted in US Patent number 10,780,139, 2020.