Vasorelaxant Activity of Salvia hispanica L.: Involvement of the Nitric Oxide Pathway in Its Pharmacological Mechanism

Salvia hispanica L., commonly known as chía, and its seeds have been used since ancient times to prepare different beverages. Due to its nutritional content, it is considered a dietary ingredient and has been reported with many health benefits. Chia seed components are helpful in cardiovascular disease (CVD) by reducing blood pressure, platelet aggregation, cholesterol, and oxidation. Still, its vasodilator effects on the vascular system were not reported yet. The hexanic (HESh), dichloromethanic (DESh), and methanolic (MESh) extracts obtained from chía seeds were evaluated on an aortic ring ex-vivo experimental model. The vasorelaxant efficacy and mechanism of action were determined. Also, phytochemical data was obtained through 13C NMR-based dereplication. The MESh extract showed the highest efficacy (Emax = 87%), and its effect was partially endothelium-dependent. The mechanism of action was determined experimentally, and the vasorelaxant curves were modified in the presence of L-NAME, ODQ, and potassium channel blockers. MESh caused a relaxing effect on KCl 80 mM-induced contraction and was less potent than nifedipine. The CaCl2-induced contraction was significantly decreased compared with the control curve. Phytochemical analysis of MESh suggests the presence of mannitol, previously reported as a vasodilator on aortic rings. Our findings suggest NO-cGMP pathway participation as a vasodilator mechanism of action of S. hispanica seeds; this effect can be attributed, in part, to the mannitol presence. S. hispanica could be used in future research focused on antihypertensive therapies.


Introduction
Salvia L. genus is the great abundant taxonomic group of the Lamiaceae family with ca. 1000 species [1]. The Salvia spp., have a wide worldwide distribution [2] and has been used in ancient traditional medicine, as food, and even in the production of cosmetics in different parts of the world [3][4][5], e.g., Salvia hispanica L., commonly known as chía, is an

Mechanism of Action
The MESh showed the major vasorelaxant effect, so its mechanism of action was assessed. The MESh effect was partially endothelium dependent. This suggests that endothelium-derived relaxant factors (EDRF) such as nitric oxide (NO), prostacyclin (PGI2), or Endothelium-Derived Hyperpolarizing Factor (EDHF) and direct mechanisms on vascular smooth muscle cells as antagonism of the adrenergic receptor or calcium channel blocking are involved in the MESh-induced relaxation [24]. Based on earlier data, the participation of EDRF as NO and PGI2 was first investigated.
The MESh extract showed the major vasorelaxant effect, so its mechanism of action was assessed. A MESh relaxation curve was shifted to the right in the presence of L-NAME (Emax = 18.48%; EC50 = >1000 µg/mL) and ODQ (Emax = 28.32%; EC50 = >1000 µg/mL) and efficacy was significantly lowered (Figure 2a), while relaxation curve in the presence of indomethacin or atropine were not altered respect to the control (MESh: Emax = 87.69%; CE50 = 124 µg/mL) (Figure 2b). The curve in the presence of potassium channel blockers such as 2-aminopyridine, glibenclamide, and tetraethylammonium were significantly

Mechanism of Action
The MESh showed the major vasorelaxant effect, so its mechanism of action was assessed. The MESh effect was partially endothelium dependent. This suggests that endothelium-derived relaxant factors (EDRF) such as nitric oxide (NO), prostacyclin (PGI 2 ), or Endothelium-Derived Hyperpolarizing Factor (EDHF) and direct mechanisms on vascular smooth muscle cells as antagonism of the adrenergic receptor or calcium channel blocking are involved in the MESh-induced relaxation [24]. Based on earlier data, the participation of EDRF as NO and PGI 2 was first investigated.
The MESh extract showed the major vasorelaxant effect, so its mechanism of action was assessed. A MESh relaxation curve was shifted to the right in the presence of L-NAME (E max = 18.48%; EC 50 = >1000 µg/mL) and ODQ (E max = 28.32%; EC 50 = >1000 µg/mL) and efficacy was significantly lowered (Figure 2a), while relaxation curve in the presence of indomethacin or atropine were not altered respect to the control (MESh: E max = 87.69%; CE 50 = 124 µg/mL) (Figure 2b). The curve in the presence of potassium channel blockers such as 2-aminopyridine, glibenclamide, and tetraethylammonium were significantly modified (Figure 3a). The same shows the relaxant curve of MESh in aortic rings contracted with high potassium solution (KCl; 80 mM) (Figure 3b). The MESh extract was efficient (E max = 70%) compared with a calcium channel blocker (nifedipine, E max = 99%). While Figure 3c shows the CaCl 2 -induced contraction curve. The MESh significantly caused inhibition of calcium contraction as well as nifedipine. The efficacy of norepinephrine in the presence of MESh was not modified.
Molecules 2023, 28, x FOR PEER REVIEW 4 of 13 modified ( Figure 3a). The same shows the relaxant curve of MESh in aortic rings contracted with high potassium solution (KCl; 80 mM) ( Figure 3b). The MESh extract was efficient (Emax = 70%) compared with a calcium channel blocker (nifedipine, Emax = 99%). While Figure 3c shows the CaCl2-induced contraction curve. The MESh significantly caused inhibition of calcium contraction as well as nifedipine. The efficacy of norepinephrine in the presence of MESh was not modified.
(a) (b)  Molecules 2023, 28, x FOR PEER REVIEW 4 of 13 modified ( Figure 3a). The same shows the relaxant curve of MESh in aortic rings contracted with high potassium solution (KCl; 80 mM) (Figure 3b). The MESh extract was efficient (Emax = 70%) compared with a calcium channel blocker (nifedipine, Emax = 99%). While Figure 3c shows the CaCl2-induced contraction curve. The MESh significantly caused inhibition of calcium contraction as well as nifedipine. The efficacy of norepinephrine in the presence of MESh was not modified.

Phytochemical Analysis
The 13 C NMR-based dereplication was performed by comparing the chemical shift (δC) experimental values from the crude extracts (Figures S1-S3) with those predicted chemical shifts contained in a dedicated database Salvia/DB. Based on the score obtained from comparing the experimental vs. predicted chemical shift, the dereplication analysis performed with HESh and DESh extract suggested the presence of alcohols, saturated and unsaturated fatty acids in the first 20 results (Tables 2 and 3), of which linoleic acid (18/18 matching carbons, score 1) was confirmed in HESh by comparing their experimental chemical shift vs. reported chemical shift [25] (Table S1). Similarly, the results of the MESh extract show the presence of sugars, glycosides, and polyalcohols (Table 4); of these, mannitol (6/6 matching carbons, score 1) was confirmed [26] (Table S1). The presence of fatty acids in DESh was not confirmed. Table 2. First 20 NPs predicted from the results of the 13 C NMR-based dereplication analysis, of the HESh extract.

Phytochemical Analysis
The 13 C NMR-based dereplication was performed by comparing the chemical shift (δ C ) experimental values from the crude extracts (Figures S1-S3) with those predicted chemical shifts contained in a dedicated database Salvia/DB. Based on the score obtained from comparing the experimental vs. predicted chemical shift, the dereplication analysis performed with HESh and DESh extract suggested the presence of alcohols, saturated and unsaturated fatty acids in the first 20 results (Tables 2 and 3), of which linoleic acid (18/18 matching carbons, score 1) was confirmed in HESh by comparing their experimental chemical shift vs. reported chemical shift [25] (Table S1). Similarly, the results of the MESh extract show the presence of sugars, glycosides, and polyalcohols (Table 4); of these, mannitol (6/6 matching carbons, score 1) was confirmed [26] (Table S1). The presence of fatty acids in DESh was not confirmed. Table 2. First 20 NPs predicted from the results of the 13 C NMR-based dereplication analysis, of the HESh extract.

Rank
Name Score (δ C Match)

Discussion
Many research groups are investigated the nutraceutical properties and pharmacology efficacy of S. hispanica as an antioxidant [16], anti-diabetic [14,17], a weight loss adjuvant [18], and hypertensive prevent [19,20]. Thus, this work was focused to obtained evidence of the vasorelaxant direct effect and phytochemistry composition of S. hispanica seed. Three extracts of S. hispanica seeds were prepared: HESh, DESh, and MESh, and their preclinical pharmacology efficacy were tested. The MESh extract showed more efficiency, and vasorelaxant efficacy was partially endothelium-dependent, so its mechanism of action was attributed to a dual mechanism correspondent to endothelium factors' presence and targets in smooth muscle cells [24,27].
The role of the vascular endothelium is affected by the presence of membrane-bound receptors like cholinergic, mainly M 3 receptors, which when activated by an agonist induces nitric oxide production and release [28]. The MESh curve in the presence of atropine (an M 3 antagonist) was not significantly modified. Thus, behavior discarded M 3 receptor activation as the first trigger of the extract vasorelaxant effect. The major endotheliumderived releasing factor (EDRF) is nitric oxide (NO), which diffuses to smooth muscle cells where the guanylyl cyclase enzyme is activated, thus catalyzing GMP to cGMP conversion. To identify the participation of nitric oxide pathway activation, endothelium-intact aortic rings were pre-incubated with L-NAME (an NO synthetase inhibitor) or ODQ (a guanylyl cyclase inhibitor), respectively, and after that, the MESh relaxant curve was obtained. Both curves were lowered in the presence of these inhibitors, thus suggesting that the NO-cGMP pathway could be involved in MESh-induced relaxation [29]. On the other hand, it is well known that PGI 2 endothelium releases can evoke a vasorelaxant effect [30]. Thus, with the purpose to identified prostaglandins-participation in the MESh effect, aortic rings were pre-incubated with indomethacin (an unspecific prostaglandins inhibitor); however, the MESh efficacy and potency were not modified, thus discarding PGI 2 -participation in MESh vasorelaxant effect [30].
As described, EDRF can increase the second's messengers as cGMP or cAMP and thus cause PKG or PKA activation, which participates in the opening of KATP, potassium channel calcium-activation (K Ca ), and repolarization. In consequence, the relaxation process is evoked [31]. In this context, the relaxant effect of MESh, in the presence of tetraethylammonium (K Ca channels blocker), glibenclamide (an ATP-sensitive potassium channel blocker), and or 2-aminopyridine (an inhibitor of KV), was significantly decreased by a half percent (Figure 3a), and the curve was shifted to the right respect to control (indicating a loss in potency), suggesting potassium channels opening because of NO-production MESh induced. The opening of potassium channels will cause cell repolarization, and consequently, calcium channel blocking occurs [32].
To assess if Ca 2+ channel blockade was involved in the vasorelaxant effect of the MESh, the assay was carried out in Ca 2+ -free of RKH solution, and a curve for contraction with CaCl 2 was obtained (control curve). The contractile effect induced by CaCl 2 was compared in the absence and presence of the MESh (EC 50 = 124.7 µg/mL). The CaCl 2induced contraction was significantly decreased by MESh-like nifedipine used as the positive control ( Figure 3c). Moreover, the MESh (3.03 to 1000 µg/mL) produced a moderate vasorelaxant effect on the contraction induced with high potassium Krebs solution (E max = 65%) (Figure 3b). These behaviors suggest a half blocker of L-type Ca 2+ channels in the membrane or due to the repolarization process [33]. Changes in the intracellular Ca 2+ concentration and membrane depolarization stimulate large-conductance Ca 2+ -activated K + (BKCa 2+ ) channels, which are thought to play an essential role in maintaining the membrane potential of vascular smooth muscle cells [34].
Meanwhile, phytochemical analyses were performed through the 13 C-NMR dereplication method as a spectroscopic strategy that has proven to help identify NPs in complex mixtures, even without purification [35]. This technique offers advantages over the dereplication analyses using mass spectrometry (MS), liquid chromatography (LC), gas chromatography (GC), or combined instrumentation such as LC-UV, LC-MS, or GC-MS [36]. Here, we have used 13 C-NMR chemical shift values obtained from crude extracts for the dereplication process to obtain a phytochemical approximation of S. hipanica seeds obtained in Jalisco, México. Based on the comparison of experimental vs. reported chemical shifts (Table S1), the results showed the presence of linoleic acid (Table 2), previously reported in aerial parts of Salvia triloba L. f., [37]; and mannitol (Table 4), reported in leaves of S. officinalis [38]. This is the first report of mannitol on S. hispanica seeds. Also, the endothelium-dependent vasorelaxant effect of mannitol on mousse mesenteric arterioles was reported and attributed to hyperosmotic action. Therefore, this compound could be responsible for MESh vasorelaxant effect [38]. However, further studies are needed to determine the presence of mannitol and its relationship with the reported biological activity.

Chemicals and Solution Preparation
(+/-)-norepinephrine bitartrate hydrate (NE), carbamylcholine chloride (carbachol), indomethacin, L-NG-Nitroarginine methyl ester (L-NAME), 1H- [1,2,4] All other reagents and solvents were analytical grade. Stock solutions of all the chemicals were made in distilled water, except for the extracts, which were dissolved in DMSO (10%). Fresh dilutions were made at the day of the experiment.

Experimental Subjects
Adult male Wistar rats (250-300 g bodyweight) were obtained from the Universidad Anáhuac-Mayab animal house in Mérida, Yucatan, México. Animals were housed in polycarbonate cages and maintained under standard laboratory conditions (12-h light/dark cycle, 25 ± 2 • C and humidity 45-65%) and were fed with a standard rodent diet and water ad libitum. All animal procedures were conducted by Secretaría de Agricultura, Ganadería, Desarrollo Rural, Pesca y Alimentación of México (SAGARPA, 1999) [39] and approved by the Institutional Animal Care and Use Committee (UJAT-0001-2017). All experiments were carried out using six animals per group. All study animals were sacrificed by cervical dislocation after deep anesthesia with phenobarbital (65 mg/kg, i.p.).

General Experimental Procedures
The Adult male Wistar rats were sacrificed accordingly to the method described by Sánchez-Recillas et al. (2020) [40]. Cervical dislocation and thoracic dissection were carried out to extract the thoracic aorta. It was cleaned from adjacent and connective tissue and then cut into strips 3 mm long (we use a Vernier measurement instrument). In addition, for some aortic rings, the endothelium layer was gently removed by cotton rub manual procedures. The aortic rings were assembled in chambers at 37 • C containing Krebs-Henseleit solution (KHS) at pH 7.4 using stainless steel hooks under an optimal tension of 3 g. After that, aortic rings were submitted to stabilize period for 20 and were constantly bubbled with an O 2 :CO 2 (95:5%) mixture. Changes in tension were recorded by force transducers Grass-FT03 (Astromed, West Warwick, RI, USA) connected to an analyzer MP-150 (BIOPAC 4.1 Instruments, Santa Barbara, CA, USA).
A sensitization process was carried out after the stabilization period. The tissues were stimulated with noradrenaline (NE, 0.1 µM) for 15 min after washed with fresh KHS, and allowed to stabilize for 15 min, this procedure was repeat three times. The absence of endothelium layer was confirmed by the lack of the relaxant response (>50%) induced by carbachol (CCH; 1 µM) in the last contraction with NE before washing with fresh KHS to assess viability.

Ex Vivo Vasorelaxant Evaluation
After sensitization period, the tissues were allowed to stabilize for 20 min and then, contracted with NE (0.1 µM). Extracts (3.03 to 1000 µg/mL), vehicle (100% final concentration) or the positive controls CCH for endothelium-intact aortic rings (E+; 0.303 to 100 µg/mL) and nifedipine for endothelium-denuded aortic rings (E-; 3.89 × 10 −5 to 3.46 µg/mL) were added to the chamber in quarter-logarithm dilutions and cumulative concentration-response curves (CRC). The relaxant effect of the samples was determined by their ability to reduce the maximal vascular contraction (E max ) and potency (EC 50 ) effects induced by NE comparing the tissue tension before and after their addition.

Mechanism of Action Approach
In order to establish the mechanism of action of methanol extract of S. hispanica (MESh), the following experiments were conducted.
(a) To establish a possible antagonism of adrenergic receptors or disruption of the NE pathway, the following procedures were performed on E-aortic rings. A cumulative NE-induced contraction (4.15 × 10 −11 to 3.6 × 10 −5 M) of CRC was made as the positive control (control CRC). In another experiment, aortic rings were pre-incubated with MESh (EC 50 = 125 µg/mL) for 15 min. Then the CRC to NE-induced contraction was performed to compare the contraction induced by NE in the absence and presence of MESh.
(b) To know the role of endothelium-derived relaxing factors as nitric oxide (NO) or prostacyclin (PGI 2 ), the E+ aortic rings were pre-incubated with NG-nitro-L-arginine methyl ester (L-NAME, nitric oxide synthase inhibitor (100 µM) or indomethacin and cyclooxygenases inhibitor (10 µM), respectively for 15 min, before the contraction with NE (0.1 µM). The relaxation CRC of MESh (3.03 to 1000 µg/mL) was built as described in the vasorelaxant set experiments. The maximal relaxing effect of the MESh was compared in the absence and presence of L-NAME or indomethacin, respectively.
(d) To know the role of K + channels on extract-induced vasorelaxant effect, the E+ aortic rings were pre-incubated with tetraethylammonium (TEA, non-selective K Ca channels blocker (10 µM), 2-Aminopyridine (2AP; 100 µM) an inhibitor of voltage-gated potassium channels (KV) or glibenclamide (10 µM) an ATP-sensitive potassium channel blocker (K ATP ) for 15 min, previous to the contraction with NE (1 µM). The relaxation CRC of MESh (3.03 to 1000 µg/mL) was built as described before. The maximal relaxing effect of the MESh was compared in the absence and presence of K ATP .
(e) To determine whether inhibition of extracellular Ca 2+ influx was involved in the extract-induced vasorelaxation, the experiments were carried out in Ca 2+ -free KHS. After sensitization, endothelium-intact aortic rings were washed with Ca 2+ -free KHS containing KCl (80 mM) and stabilized for 15 min. Then, a CRC for CaCl 2 -induced contraction was obtained without the MESh (control group). Once the maximal contraction was reached, tissue was washed with Ca 2+ -free, KCl (80 mM), and KHS, and allowed to stabilize for 20 min. Finally, after 15 min incubation with the (EC 50 = 125 µg/mL), another CRC for CaCl 2 -induced contraction was obtained. The contractile effect induced by CaCl 2 was compared in the absence and presence of the MESh.

Natural Products Databases for 13 C NMR Dereplication
A database (DB) of natural products (NPs) for 13 C NMR-based dereplication process was prepared with some modifications to the method described by Bruguière et al. (2021) [41]. Briefly, a search was carried out for previously NPs reported in Salvia spp., using LO-TUS: Natural Products Online, available at https://lotus.naturalproducts.net/ (accessed on 20 March 2023). The structures resulting from the LOTUS research were exported in .sdf format, containing 2392 NPs. The 13 C NMR chemical shifts (δ C ) were predicted for each NPs' methyl, methylene, methine, and quaternary carbons using the algorithm described by Nuzillard (2021) [42] and the ACD NMR predictor (ACD/Labs) to obtain Salvia/DB in the required format for MixONat software (v.1.0.1., SONAS, France).

13 C NMR-Based Dereplication
Dereplication analyses were carried out using MixONat software available at https://sourceforge.net/projects/mixonat/ (accessed on 24 April 2023) [43]. Previously, 50 mg of the crude plant extracts were dissolved in chloroform-d4 or methanol-d4 (600 µL). Carbon spectra ( 13 C-NMR, 150 MHz) were obtained on a Varian-Agilent AR Premium Compact spectrometer (Santa Clara, CA, USA). The spectra were acquired with 6000 scans, and the spectral width was 230 ppm. Phase and baseline correction of spectra was performed automatically using MestReNova software (v. 12.0.0, Mestrelab Research, Santiago de Compostela, Spain). A minimum intensity threshold was then used to collect positive 13 C-NMR signals. Each spectrum's experimental δ C list and intensities data were exported to a .csv file using Excel Microsoft Office (Microsoft, Redmond, WA, USA) as an input file in MixONat. The software ranked the putative NPs contained in the mixture with a range score between 1 and 0, according to the number of matching experimental δ C obtained from extracts vs. the predicted δ C-SDF values of NPs in the Salvia/DB. The predicted NPs were then confirmed by comparing the experimental vs. with that δ C reported in the literature analyzed in the same deuterated solvent, with a displacement tolerance range of ±0.5 ppm.

Statistical Analysis
The results are expressed as the standard error of the mean (n = 5) ± SEM. Concentrationresponse curves (CRC) were plotted, and the experimental data from the CRC were adjusted by the nonlinear Hill equation with a curve-fitting program (ORIGIN 8.0 MICROCAL). The statistical significance of differences between means was assessed by a one-way analysis of variance (ANOVA) followed by Tukey's post hoc test; p-values lower less than 0.05 (* p < 0.05) were statistically significant [44,45].

Conclusions
The vasorelaxant effect of S. hispanica seems to be characterized, and our results suggest NO-cGMP pathway participation as a vasodilator mechanism of action. Also, the preliminary phytochemical report was presented. This work is the first report on ex vivo pharmacology analysis of S. hipanica species, which could be used in future research focused on antihypertensive therapies.