Antispasmodic activity from Serjania caracasana fractions and their safety

Abstract

In a previous study, we reported the antispasmodic and gastroprotective effects of the Serjania caracasana (Jacq.) Willd., Sapindaceae, extract. In the present study, we evaluated the LD₅₀, hemolytic and antispasmodic activities of its fractions and characterized its major constituents by isolation and GC–MS. The animals showed non-toxic symptoms with oral doses up to 2000 mg/kg, suggesting a safe oral administration. Furthermore, a low hemolytic activity was detected for the saponin fraction. Antispasmodic activity of the fractions was evaluated through carbachol-induced contractions in rat ileum. The hexane fraction was the most potent (IC₅₀ 68.4 ± 5.9 µg/ml) followed by the dichloromethane fraction (IC₅₀ 161.3.4 ± 40.7 µg/ml). Butanol fraction was the less effective (IC₅₀ 219.8 ± 60.3 µg/ml). The phytochemical study of the S. caracasana fractions afforded the isolation of friedelin, β-amyrin, allantoin and quercitrin. This is the first time that the presence of allantoin and quercitrin in the Serjania genus has been reported. Among the isolated compounds and those characterized by GC–MS, β-amyrin and β-sitosterol were present in the most active fractions, hexane and dichloromethane, and they may be related to its antispasmodic activity. In addition, spathulenol was only found in the hexane fraction and its presence might justify the highest antispasmodic activity observed for this fraction.

Keywords:
Antispasmodic activity; Compounds isolation; GC–MS; Hemolytic assay; Ileum rat; Extract toxicity

Introduction

Serjania caracasana (Jacq.) Willd., Sapindaceae, is used for weaving baskets and rustic ropes, mostly because it is considered ichthyotoxic. This belief comes from its similarity with S. lethalis, which is known to contain hemolytic saponins (Teixeira et al., 1984). In fact, all previous publications about S. caracasana (Aragão and Valle, 1973; Xavier and Mors, 1975; Cordeiro and Valle, 1975; Maia-Braggio et al., 1978) used S. lethalis instead, as it was later confirmed (Teixeira et al., 1984).

Some Serjania species are used in folk medicine such as S. comata (rheumatism), S. lethalis (kidney pain, anti-inflammatory), S. erecta (gastric pain, anti-inflammatory), S. marginata (gastric pain) and S. triquetra (diuretic) (Chávez and Delgado, 1994; Guarin-Neto et al., 2000; Agra et al., 2008; Périco et al., 2015). Based on these popular uses, our group has previously demonstrated that the S. caracasana hydroethanolic extract showed a gastroprotective effect by inhibiting gastric ulcer induction and an in vitro antispasmodic activity (Silva et al., 2012).

In this study, we further evaluated the antispasmodic activity to identify the compounds responsible for this effect. The extract safety was initially evaluated through the determination of its median lethal dose (LD₅₀) and the search for hemolytic saponins in the butanol fraction.

Materials and methods

Chemical materials

Standard sample of saponins of Quillaja saponaria was a kind courtesy of Prof. Dr. Carmen Queiroga. The citrated bovine blood was purchased at the USP Veterinary Hospital (USP-Brazil). Silica gel 60 F₂₅₄ TLC aluminum sheets were purchased from Merck Company. All others reagents and solvents were analytical grade.

Plant material

The aerial parts of Serjania caracasana (Jacq.) Willd., Sapindaceae, were collected at the base of Pico do Jabre (7°15′34.27″ S, 37°23′8.53″ W), Paraiba, Brazil during fructification period (June, 2009) by Dr. Josean F. Tavares (UFPB). The plant material was identified by Prof. Dr. Maria de Fátima Agra (UFPB). A voucher specimen (No. M.F. Agra et al., 6963) was deposited in the Herbarium Prof. Lauro Pires Xavier (JPB), at the same University.

Phytochemical analysis

Spectroscopic properties of the isolated compounds

NMR analysis was performed in an Agilent INOVA-500 (500 MHz) spectrometer. High resolution mass spectra were performed on a MicroTOF LC mass spectrometer from Bruker Daltonics and IR spectra was recorded with a Bomem instrument. The structures of the isolated compounds were determined by ¹H and ¹³C NMR spectroscopy, using one and two-dimensional techniques, together with IR and HRMS data by comparison with data previously published for these compounds.

Gas chromatography–mass spectrometry (GC–MS) analysis

The hexane and CH₂Cl₂ fractions had their chemical composition analyzed by gas chromatography–mass spectrometry (GC/MS), performed using a Shimadzu GCMS-QP2010 Ultra system, with a fused silica capillary column coated with 5% polyphenylsiloxane/95% dimethyl polysiloxane (Rxi^®-5 ms, 10 m × 0.10 mm ID × 0.10 µm film thickness), electron ionization system operating at 70 eV with an interface temperature of 260 °C, scan time of 0.1 scans/s and acquisition mass range of 35.0–500.0 Da. For the GC analysis, the injection temperature was set at 250 °C, the oven temperature program was 100 °C for 1 min, 100–290 °C at 15 °C/min, maintaining 290 °C for 15 min with helium as a carrier gas (0.42 ml/min). The compound identification was performed by comparing the mass spectra with the Wiley library and literature data (Adams, 2007) together with the co-injection of standards. The relative composition was obtained from the electronic integration measurements using flame ionization detection (300 °C) without taking into account relative response factors, in the same conditions described above.

Extract preparation

The powdered air-dried aerial parts of S. caracasana (1916 g) were extracted exhaustively with 96% aqueous ethanol solution at room temperature. The combined extracts were filtered and concentrated under reduced pressure at 40 °C affording a hydroethanolic extract (202.84 g; 10.6% extraction yield) (EtOH).

Extract fractionation and compound isolation

The EtOH extract (200 g) was suspended in MeOH–H₂O (70:30, v/v) mixture, and it was subsequently fractioned with hexane, CH₂Cl₂ and BuOH. Each solvent fraction was then evaporated to dryness under reduced pressure to give hexane (31.53 g; 15.8%), CH₂Cl₂ (21.9 g; 10.96%) and BuOH (44.50 g; 22.2%) fractions. The CH₂Cl₂ and BuOH fractions were separately subjected to CC on silica gel using step gradients of hexane–EtOAc and EtOAc–MeOH. Column fractions that presented only one major compound were resubmitted to the same chromatographic separation to obtain the purified compounds. CH₂Cl₂ fraction yielded four compounds (friedelin (1), β-amyrin (2), stigmasterol and β-sitosterol), while BuOH fraction afforded one compound (β-sitosterol glucoside). A remaining part of the BuOH fraction was suspended with 300 ml of MeOH. The solution was filtered and insoluble part reserved. Diethyl ether was carefully added into the solution until some precipitation started, and then it was placed in the freezer (-22 °C for 24 h). The precipitate formed was removed by filtration, and reserved. This process was repeated twice. The pooled precipitates were solubilized with MeOH and concentrated under reduced pressure at 40 °C, to give BuOH-1 fraction. The remaining organic phase was evaporated under reduced pressure to dryness to give BuOH-2 fraction. The BuOH-2 fraction was subjected to CC on silica gel using step gradients of CHCl₃–MeOH and after fixed solvent system (CHCl₃–acetone–formic acid, 75:16.5:8.5 (v/v)) (Ikeda et al., 1991) to obtain 27 fractions combined according to their TLC profiles into fifteen major fractions (Fr1–Fr15). From fraction Fr7 obtained allantoin (3) as precipitate after solvent drying. Fraction Fr8 was purified by preparative TLC (CHCl₃–MeOH, 80:20 (v/v)) to give quercitrin (4).

Friedelin (1). 5.6 mg, 0.03% yield; amorphous powder; IR ν_max (KBr) cm^-1: 3000, 2848, 1714; ¹H NMR (500 MHz, CDCl₃): 1.16 (3H, s, H-28), 1.03 (3H, s, H-27), 0.99 (3H, s, H-26), 0.98 (3H, s, H-30), 0.93 (3H, s, H-29), 0.86 (3H, sl ap, H-23), 0.85 (3H, s, H-25), 0.70 (3H, s, H-24). ¹³C NMR data were consistent with those previously reported (Akihisa et al., 1992).

β-Amyrin (2). 169.8 mg, 1.05% yield; white powder; IR ν_max (KBr) cm^-1: 3299, 2946, 2852, 1034; ¹H NMR (500 MHz, CDCl₃): 5.16 (1H, t, J = 3.6 Hz, H-12), 3.20 (1H, dd, J = 11.15, 4.5, H-3), 1.11 (3H, s, H-27), 0.97 (3H, s, H-23), 0.94 (3H, s, H-26), 0.91 (3H, s, H-24), 0.80 (3H, s, H-28), 0.76 (3H, s, H-25). ¹³C NMR data were consistent with those previously reported (Dias et al., 2011).

Stigmasterol. 37.8 mg, 0.24% yield; white powder; ¹H NMR (500 MHz; CDCl₃): 5.33 (1H, sl, H-6), 5.13 (1H, dd, H-23), 5.00 (1H, dd, H-22), 3.50 (2H, m, H-3). ¹³C NMR data were consistent with those previously reported (Kojima et al., 1990).

β-Sitosterol. 21.4 mg, 0.13% yield; white powder; ¹H NMR (500 MHz; CDCl₃): 5.33 (1H, sl, H-6), 3.50 (2H, m, H-3). ¹³C NMR data were consistent with those previously reported (Kojima et al., 1990).

β-Sitosterol glucoside. 3.2 mg, 0.04% yield; amorphous powder; ¹H NMR (500 MHz, Py-d₅): 5.32 (1H, sl, H-6), 5.04 (1H, d, H-1′), 4.55 (1H, m ap, H-6′), 4.28 (1H, s ap, H-4′), 3.99 (1H, m, H-3), 2.70/2.45 (1H, m, H-4), 0.98 (3H, d, H-21), 0.94 (3H, sl, H-19), 0.91 (3H, s ap, H-26), 0.87 (3H, d ap, H-29), 0.84 (3H, d, H-27), 0.64 (3H, s, H-18). ¹³C NMR data were consistent with those previously reported (Kojima et al., 1990).

Allantoin (3). 6.5 mg, 0.08% yield; colorless crystal; IR ν_max (KBr) cm^-1: 3439, 3344, 3224, 3063, 1782, 1659; ¹H NMR (500 MHz, DMSO-d₆): 10.51 (1H, br s, H-1), 8.03 (1H, s, H-4), 6.91 (1H, d, J = 13.5 Hz, H-3), 5.78 (2H, s, H-8), 5.24 (1H, d, J = 13.5 Hz, H-6). ¹³C NMR data were consistent with those previously reported (Sripathi et al., 2011). EI-HRMS (negative-ion mode) m/z: 157.03722 [M-H]^- (C₄H₆N₄O₃ requires 157.0315).

Quercitrin (4). 30.9 mg, 0.41% yield; yellow powder; ¹H NMR (500 MHz, CD₃OD): 7.34 (1H, s ap, H-2′), 7.30 (1H, d ap, J = 7.74 Hz, H-6′), 6.91 (1H, d, J = 7.74 Hz, H-5′), 6.36 (1H, s, H-6), 6.19 (1H, s, H-8), 5.34 (1H, s ap, H-1″), 0.94 (1H, d, J = 5.85 Hz, H-6″). ¹³C NMR data were consistent with those previously reported (Shi et al., 2010).

Toxicological assays

Median lethal dose (LD₅₀)

Male Swiss mice (35–40 g) were separated in three groups of five animals for each treatment (n = 5). Animals from each experimental and control groups were hosted in cages maintained at constant room temperature (22 ± 1 °C) and subjected to a 12/12 h light/dark cycle with access to food and water ad libitum. Procedures involving animals and their care were performed in conformity with OECD 420 (2001), adopted in our laboratory, and in compliance with current internationally accepted instructions for the care of laboratory animals and ethical guidelines. Furthermore, clearance for conducting the study was obtained from the Ethics Committee of Animal Use of the Nove de Julho University (approval # AN 0003/11).

In order to determine the oral median lethal dose (LD₅₀) of the EtOH extract, this test followed the method described for Miller and Tainter (1944) with same modifications. The control group received the vehicle (0.1% Tween 20 in distillated water). Two doses (1000 and 2000 mg/kg, in a volume of 1 ml/g) were given orally, by gavage, to two groups of mice for the determination of LD₅₀. The animals were observed during the first 180 min after the treatment and after 24, 48 and 72 h for any toxic signs and symptoms.

Qualitative assay of hemolytic saponins

The qualitative hemolytic activity of the saponins was tested with bovine blood reagent as described by Sharma et al. (2012) with some modifications. Briefly, aliquots of 1 ml of citrated bovine blood were washed three times with 9 ml of saline (0.9%; w/v NaCl) followed by centrifugation at 180 × g for 5 min. The cell suspension was finally prepared by diluting the pellet to 3% (v/v) in saline solution to obtain the blood reagent. Also, to visualize the saponins a comparative spray reagent (Liebermann–Burchard reagent) was prepared (Wagner and Bladt, 1996).

The assay was performed with a 10 µl aliquot of saponins from Quillaja saponaria (SQ) and BuOH fraction (conc. 20 mg/ml, solubilized in MeOH–H₂O, 70:30 (v/v)) applied in TLC plates which were eluted with the solvent system CHCl₃–MeOH–TFA 0.5% (60:40:5 (v/v)) to 6.5 cm from the origin. After the elution, the TLC plates were air dried.

One of the developed TLC plates was immersed for 20 s in a glass dish containing the 3% blood reagent freshly prepared. After this time, the TLC plate was removed and held vertically for 30 s, and subsequently immersed in saline for 30 s. Finally, this TLC plate was held vertically for complete drying and further visualization of the hemolytic spots. The hemolytic spots were compared with the developed TLC sprayed with Liebermann–Burchard reagent (Sharma et al., 2012).

Quantitative assay of hemolytic saponins

The quantitative hemolytic activity of the saponins was evaluated with a bovine blood cell suspension by turbidimetry as described by Xie et al. (2008) with some modifications. Briefly, aliquots of 10 ml of blood were washed three times with saline solution by centrifugation at 180 × g for 2 min. The cell suspension was finally prepared by diluting the pellet to 5% (v/v) in saline solution.

The SQ and BuOH fraction samples were solubilized initially in saline to 25 mg/ml. For the assay, 180 µl of freshly prepared 5% blood cell suspension was mixed with 20 µl of sample or control solution. The sample concentration ranged from 820.0 to 16.4 µg/ml for BuOH fraction, and from 500.0 to 5 µg/ml for SQ. The microplate was incubated for 30 min at 37 °C and centrifuged at 70 × g for 10 min. An aliquot of each supernatant (75 µl) was transferred to a flat-bottom microplate and the free hemoglobin was measured at 540 nm (Silveira et al., 2011). Saline and SQ (100 µg/ml) were considered as minimal and maximal hemolytic controls. The concentration inducing 50% of maximum hemolysis (HC₅₀) was calculated by non-liner regression (GraphPad Prism 5.01). Each experiment included triplicates for each concentration. The results of the quantitative hemolytic activity are presented as mean ± SD and the other results as mean ± SEM.

Pharmacological assay

Antispasmodic activity on ileum isolated rat

Ileum strips were isolated from rat following the methodology described by Walker and Wilson (1979) and suspended in organ baths (5 ml) containing modified Krebs physiological solution, consisting of (mmol/l): NaCl 117.0; KCl 4.7; NaH₂HPO₄·H₂O 1.2; MgSO₄·7H₂O 1.3; CaCl₂·2H₂O 2.5; NaHCO₃ 25.0; glucose 11.0; pH 7.4 (Sun and Benishin, 1994). Furthermore, the clearance for conducting the study was obtained from the Ethics Committee of Animal Use of the Federal University of São Paulo (approval # CEUA 4195060514/14).

The hexane, CH₂Cl₂ and BuOH fractions were dissolved initially in 0.01% Cremophor EL and diluted in MilliQ water to obtain the stock solution of 10 mg/ml which was stored at 0 °C.

The tissues were maintained under 1 g tension, bubbled continuously with O₂ at 37 °C. They were attached to force isometric transducers and connected to a data system AQCD (AVS Projetos, Brazil). Following control contractions with KCl (40 mmol/l) or carbachol (1 µmol/l), and washing with a fresh Krebs solution, tissue strips were exposed to concentration gradient ranging from 500 to 9 µg/ml of hexane, CH₂Cl₂ or BuOH fractions for 15 min (Walker and Wilson, 1979), then stimulated again with the previous referred concentration of carbachol. The antispasmodic activity of the samples was expressed as maximum contraction (E_max value) obtained in the presence of the distinct fraction relative to the maximum contraction in their absence (control). The concentration of fraction that reduces to 50% a maximal response for an agonist (IC₅₀) values were determined from individual concentration–response curves by non-linear regression (Jenkinson et al., 1995).

Statistical analysis

Statistical significance between control and experimental group was evaluated using either Student's t-test or one-way ANOVA following Dunnett's Multiple Comparison Test. Data were considered significant when p < 0.05.

Results and discussion

Median lethal dose (LD₅₀)

In our earlier study, we observed the S. caracasana EtOH extract significantly inhibited KCl pre-contracted ileum, indicating an interesting in vitro antispasmodic activity (Silva et al., 2012). However, due to the possible toxicity to mammals that could be related to the presence of hemolytic saponins, as reported for other Serjania species, we decided to perform some preliminary safety assays. The oral acute toxicity from S. caracasana crude extract was assessed by its LD₅₀. In this assay, two oral doses (1000 and 2000 mg/kg) did not produce any visible signs of toxic symptoms 72 h after receiving the extract. Therefore, no further evaluation was performed and the LD₅₀ was estimated as higher than 2000 mg/kg. This LD₅₀ value places the extract in the Category 5 of the United Nations Globally Harmonized System for toxicity hazards, which leads this extract to be considered as safe after oral acute exposure. Additional animal tests with substances in this category range are strongly discouraged and should only be considered when there is a direct relevance for protecting human health (Bulgheroni et al., 2009). Similar results were found for the chloroform extract from S. erecta and hydroethanolic extract from S. marginata, where mice treated with a single oral dose of 5000 mg/kg had no signs of toxic effects in an acute toxicity study (Arruda et al., 2009; Périco et al., 2015).

Qualitative and quantitative assays of hemolytic saponins

Despite the low acute toxicity of S. caracasana EtOH extract, the presence of hemolytic saponins in Serjania extracts should be investigated. These compounds are considered harmful not only because of their acute toxic effects, as reported to the saponins isolated from S. lethalis, serjanosides A, B and C, that showed toxic acute effects in rats and mice (Teixeira et al., 1984). Also, chronic administration of saponins in animals is known to affect their growth or altering their palatability and, consequently, food consumption or altering the digestion process and absorption (Oleszek, 1996). The TLC chromatogram of the S. caracasana BuOH fraction sprayed with Liebermann–Burchard reagent (Fig. 1B) showed the presence of three main purple zones, characteristic of saponins, at R_f 's 0.3, 0.4 and 0.6, with the latter as the major component. The comparison of this TLC profile with that obtained after spraying with the Blood reagent (Fig. 1A) indicated that the observed saponins were not able to cause hemolysis in the TLC assay in the concentration tested. In contrast, the positive control (SQ) showed some white spots (R_f = 0.1 and 0.2) against a pink background (Fig. 1A), characteristic of hemolysis. However, SQ also presented two other compounds that were non-hemolytic compounds (R_f = 0.4 and 0.6), typically saponins for Liebermann–Burchard reagent (Fig. 1B).

Fig. 1
TLC of BuOH fraction of Serjania caracasana aerial parts (1) and of rich fraction saponins of Quillaja saponaria (SQ) (2). Solvent system: chloroform–methanol–trifluoroacetic acid aqueous solution 0.5% (60:40:5, v/v). Detection: blood reagent 3% (A); Liebermann–Burchard reagent (B).

Similarly, in the quantitative hemolytic activity, the BuOH fraction exhibited a low hemolysis rate (∼2%) until the maximum concentration tested (820 µg/ml) when compared with the positive control (SQ) (100% at 100 µg/ml) (Fig. 2). Once at the maximum concentration tested the hemolysis rates were under 50% of hemolytic activity, the BuOH fraction HC₅₀ value was estimated as >1000 µg/ml. The positive control (SQ) showed high hemolytic activity (HC₅₀ = 24.03 ± 1.03 µg/ml) under the same conditions. Thus, different than would be expected for an ichthyotoxic species, our results indicated that S. caracasana saponins could be considered as non-hemolytic.

Fig. 2
Hemolytic activity of the enriched saponin fraction from Quillaja saponaria (SQ) (A) and of the BuOH fraction of Serjania caracasana aerial parts (B).

Antispasmodic activity on ileum isolated rat

As S. caracasana demonstrated low toxicity and almost no hemolytic effect, we decided to investigate which extract fraction would be responsible for the antispasmodic activity previously reported (Silva et al., 2012).

The antispasmodic activity was analyzed for the S. caracasana hexane, CH₂Cl₂ and BuOH fractions. Fig. 3(A)–(C) shows the overall inhibitory effect of these fractions on rat ileum contractions. The hexane fraction significantly inhibited the maximum ileum contractions at the concentrations of 27, 81, 243 and 500 µg/ml (Fig. 3A), presenting the respective amplitude decrease values of 66.1 ± 7.1, 47.5 ± 3.8, 24.5 ± 2.4 and 23.5 ± 7.3% (Fig. 3A), while for the CH₂Cl₂ fraction a significant inhibition started at the contractions of 81, 243 and 500 µg/ml, with E_max values of 62.7 ± 12.6, 47.7 ± 7.9 and 20.2 ± 3.7% respectively (Fig. 3B). In contrast, the BuOH fraction was only able to significantly reduce the intensity of the contractions at 243 and 500 µg/ml with respective E_max of 44.9 ± 13.1 and 25.8 ± 5.5% (Fig. 3C).

Fig. 3
Effect of hexane (A), CH₂Cl₂ (B) or BuOH (C) fractions of Serjania caracasana aerial parts on pre-contracted ileum by carbachol 1 µM (n = 3). One-way ANOVA followed by Dunnett's Multiple Comparison Test: **p < 0.01; ***p < 0.001 (control × fraction).

These results indicated that the antispasmodic activity is distributed in all S. caracasana fractions. However, the n-hexane fraction was more potent than the crude EtOH extract, that was 46% at the concentration of 81 µg/ml (Silva et al., 2012). The CH₂Cl₂ fraction was less active showing similar results to those obtained with the crude extract, and the BuOH fraction showed even less activity than the extract. Thus, all the fractions were interesting to search for natural antispasmodic compounds.

Phytochemical analysis

In order to determine the possible compounds responsible for the antispasmodic activity, CH₂Cl₂ and BuOH fractions were submitted to column chromatography. From these fractions, we could isolate and characterize seven components, two oleanane triterpenes: friedelin (1) and β-amyrin (2), three steroids: stigmasterol, β-sitosterol and β-sitosterol glucoside, one urate derivative: allantoin (3) and one flavonol: quercitrin (4).

Additionally, the two most active fractions, hexane and CH₂Cl₂, had their chemical composition compared by GC–MS, indicating a similar chemical composition for these fractions (Fig. 4). The mass spectra obtained allowed to characterize six compounds (Table 1) representing 79.8% and 66.6% of the total of compounds detected in the GC–MS. The two fractions showed the same major components, β-amyrin (2) (60.0% and 51.0%) and ethyl palmitate (6.2% and 8.6%), respectively for hexane and CH₂Cl₂ fractions. Additionally, in the hexane fraction was found spathulenol (5) in a representative quantity (4.2%). This sesquiterpene is a common essential oil component in several plant species, it has shown an antispasmodic activity in uterus rings contraction model induced by KCl (Perez-Hernandez et al., 2008). The presence of spathulenol exclusively in the hexane fraction might justify the highest antispasmodic activity observed in comparison to the other fractions.

Fig. 4
GC–MS chromatogram of hexane (A) and CH₂Cl₂ (B) fractions of Serjania caracasana aerial parts.

Table 1
Compounds characterized in the hexane and CH₂Cl₂ fractions of Serjania caracasana aerial parts.

	Hexane fraction			CH2Cl2 fraction
Compound	RT (min)	%		RT (min)	%
Spathulenol (5)	5.8	4.2		-	-
6,10,14-Trimethyl-2-pentadecanone	7.3	1.7		7.3	1.8
Methyl palmitate	7.7	1.7		7.7	2.7
Ethyl palmitate	8.1	6.2		8.1	8.6
β-Sitosterol	14.7	1.2		14.7	2.1
β-Amyrin (2)	15.1	60.0		15.1	51.0
Total		75.0			66.2

In a recent study, Coutinho et al. (2015) reported the seed fatty-acid compositions for sixteen Sapindaceae species. Serjania species, including S. caracasana, presented high levels eicosenoic acid and other palmitic acid esters. In our study, palmitic acid methyl and ethyl esters were found in both hexane and CH₂Cl₂ fractions. Friedelin (1) and β-amyrin (2), together with other triterpenes, are typically isolated within the genus Serjania, such as the presence of 1 and 2 reported in S. salzmanniana (Barbosa-Filho et al., 1988). In this study, we detected for the first time in a Serjania species quercitrin (4), allantoin (3) and spathulenol (5).

According to our results, S. caracasana saponins and quercitrin are not the main metabolites responsible for the antispasmodic effect observed, but they may contribute to the overall effect observed in the crude extract. In similar studies, it has been observed that the fractions enriched with saponins and flavonol derivatives, including quercitrin, showed significant antispasmodic activity in the acetylcholine model (Trute et al., 1997).

In conclusion, our findings demonstrated that although of S. caracasana is considered an ichthyotoxic species, this property, if present, cannot be related to the presence of hemolytic saponins. The toxicity for mammal species was not confirmed by the acute toxicity test, requiring complimentary studies on the subject. In addition, future studies should be carried out with the isolated compounds to verify their spasmolytic activity.

Ethical responsibilities

Protection of human and animal subjects

The authors declare that the procedures followed were in accordance with the regulations of the responsible Clinical Research Ethics Committee and in accordance with those of the World Medical Association and the Helsinki Declaration.

Confidentiality of data

The authors declare that no patient data appears in this article.

Right to privacy and informed consent

The authors declare that no patient data appears in this article.

Acknowledgments

The authors are grateful to CNPq/RENORBIO and CAPES/Brazil for financial support and research fellowships.

References

• Adams, R.P., 2007. Identification of Essential Oil Components by Gas Chromatography/Mass Spectrometry, 4^thed. Allured Publishing Corporation, Illinois, USA.
• Agra, M.F., Silva, K.N., Basílio, I.J.L.D., Freitas, P.F., Barbosa-Filho, J.M., 2008. Survey of medicinal plants used in the region Northeast of Brazil. Rev. Bras. Farmacogn. 18, 472-508.
• Akihisa, T., Yamamoto, K., Tamura, T., Kimura, Y., Iida, T., Nambara, T., Chang, F.C., 1992. Triterpenoid ketones from Lingnania chungii McClure: arborinone, friedelin and glutinone. Chem. Pharm. Bull. 40, 789-791.
• Aragão, J.A., Valle, J.R., 1973. Ictiotixicidade de timbós dos gêneros Serjania, Derris e Tephrosia Ciência e Cultura 25, 643.
• Arruda, A.P.C.C.B.N., Coelho, R.G., Honda, N.K., Ferrazoli, C., Pott, A., Hiruma-Lima, C.A., 2009. Gastroprotective effect of Serjania erecta Radlk (Sapindaceae): involvement of sensory neurons, endogenous nonprotein sulfhydryls, and nitric oxide. J. Med. Food. 12, 1411-1415.
• Barbosa-Filho, J.M., Araujo, V.T., Bhattacharyya, J., 1988. Chemical constituents of Serjania salzmanniana Fitoterapia 59, 430-431.
• Bulgheroni, A., Kinsner-Ovaskainen, A., Hoffmann, S., Hartung, T., Prieto, P., 2009. Estimation of acute oral toxicity using the no observed adverse effect level (NO-AEL) from 28 day repeated dose toxicity studies in rat. Regul. Toxicol. Pharmacol. 53, 16-19.
• Chávez, M.I., Delgado, G., 1994. Isolation and relay systhesis of 11α-hydroperoxy diacetyl hederagenin, a novel triterpenoid derivative from Serjania triquetra (Sapindaceae). Biogenetic implications. Tetrahedron 50, 3869-3878.
• Cordeiro, E.A., Valle, J.R., 1975. Ictiotixicidade comparada da rotenona e do serjanosídeo. Cien. Cultura 27, 561.
• Coutinho, D.J.G., Barbosa, M.O., Silva, R.M., Silva, S.I., Oliveira, A.F.M., 2015. Fatty-acid composition of seeds and chemotaxonomic evaluation of sixteen Sapindaceae species. Chem. Biodivers. 12, 1271-1280.
• Dias, M.O., Hamerski, L., Pinto, A.C., 2011. Separação semipreparativa de α e β-amirina por cromatografia líquida de alta eficiência. Quim. Nova 34, S1-S6.
• Guarin-Neto, G., Santana, S.R., Silva, J.V.B., 2000. Notas etnobotânicas de espécies de Sapindaceae Jussieu. Acta. Bot. Bras. 14, 327-334.
• Ikeda, Y., Sugiura, Y.M., Fukaya, C., Yokoyama, K., Hashimoto, Y., Kawanishi, K., Moriyasu, M., 1991. Periandradulcins A, B and C: phosphodiesterase inhibitors from Periandra dulcis Mart. Chem. Pharm. Bull. 39, 566-571.
• Jenkinson, D.H., Barnard, E.A., Hoyer, D., Humphrey, P.P.A., Leff, P., Shankley, N.P., 1995. Internacional union of pharmacology committee on receptor nomenclature and drug classification. IX. Recommendations on terms and symbols in quantitative pharmacology. Pharm. Rev. 47, 255-266.
• Kojima, H., Sato, N., Hatano, A., Ogura, H., 1990. Sterol glucosides from Prunella vulgaris Phytochemistry 29, 2351-2355.
• Maia-Braggio, M., Lapa, A.J., Valle, J.R., 1978. Mecanismo da ação tóxica da Serjania caracasana (Jacq.) Willd. Ciência e Cultura 30, 455.
• Miller, L.C., Tainter, M.L., 1944. Estimation of the LD50 and its error by means of logarithmic probit graph paper. Proc. Soc. Exp. Biol. Med. 57, 261-264.
• Oleszek, W., 1996. Alfafa Saponins: Structure, Biological Activity, and Chemotaxonomy. In Waller, G.R., Yamasaki, K. (org.) Saponins Used in Food and Agriculture. Plenum Press, New York, pp. 155–170.
• Perez-Hernandez, N., Ponce-Monter, H., Medina, J.A., Joseph-Nathan, P., 2008. Spasmolytic effect of constituents from Lepechinia caulescens on rat uterus. J. Ethnopharmacol. 115, 30-35.
• Périco, L.L., Heredia-Vieira, S.C., Beserra, F.P., dos Santos, R.C., Weiss, M.B., Resende, F.A., Ramos, M.A.S., Bonifácio, B.V., Bauab, T.M., Varanda, E.A., Gobbi, J.I.F., da Rocha, L.R.M., Vilegas, W., Hiruma-Lima, C.A., 2015. Does the gastroprotective action of a medicinal plant ensure healing effects? An integrative study of the biological effects of Serjania marginata Casar. (Sapindaceae) in rats. J. Ethnopharmacol. 172, 312-324.
• Sharma, O.P., Kumar, N., Singh, B., Bhat, T.K., 2012. An improved method for thin layer chromatographic analysis of saponins. Food Chem. 132, 671-674.
• Shi, S.Y., Zhang, Y.P., Zhou, H.H., Huang, K.L., Jiang, X.Y., 2010. Screening and identification of radical scavengers from Neo-Taraxacum siphonanthum by online rapid screening method and nuclear magnetic resonance experiments. J. Immunoassay Immunochem. 31, 233-249.
• Silva, J.L.V., Carvalho, V.S., Silva, F.L., Barbosa-Filho, J.M., Rigoni, V.L.S., Nouailhetas, V.L.A., 2012. Gastrointestinal property of Serjania caracasana (Jacq.) Willd. (Sapindaceae) on rats. Pharmacologyonline S1, 22-26.
• Silveira, F., Rossi, S., Fernández, C., Gosmann, G., Schenkel, E., Ferreira, F., 2011. Alum-type adjuvant effect of non-haemolytic saponins purified from Ilex and Passiflora spp.. Phytother. Res. 25, 1783-1788.
• Sripathi, S.K., Gopal, P., Lalitha, P., 2011. Allantoin from the leaves of Pisonia grandis R. Br. Int. J. Pharm. Life Sci. 2, 815-817.
• Sun, Y.D., Benishin, C.G., 1994. K⁺ channel openers relax longitudinal muscle of guinea-pig ileum. Eur. J. Pharmacol. 271, 453-459.
• Teixeira, J.R.M., Lapa, A.J., Souccar, C., Valle, J.R., 1984. Timbós: ichthyotoxic plants used by Brazilian Indians. J. Ethnopharmacol. 10, 311-318.
• The Organisation of Economic Co-operation Development (OECD), 2001. OECD Guideline for TESTING of chemicals: 420 Acute Oral Toxicology – Fixed Dose Procedure. The Organisation of Economic Co-operation Development (OECD), pp. 1–14.
• Trute, A., Gross, J., Mutschler, E., Nahrstedt, A., 1997. In vitro antispasmodic compounds of the dry extract obtained from Hedera helix Planta Med. 63, 125-129.
• Wagner, H., Bladt, S., 1996. Plant Drug Analysis: A Thin Layer Chromatography Atlas, 2nd edition. Springer-Verlag, Heidelberg, Germany, pp. 305–327.
• Walker, R., Wilson, K.A., 1979. Prostaglandins and the contractile action of bradykinin on the longitudinal muscle of rat isolated ileum. Br. J. Pharmacol. 67, 527-533.
• Xavier, H.S., Mors, W.B., 1975. As saponinas tóxicas de Serjania caracasana Cien. Cultura 27, 179.
• Xie, Y., Ye, Y.-P., Sun, H.-X., Li, D., 2008. Contribution of the glycidic moieties to the haemolytic and adjuvant activity of platycodigenin-type saponins from the root of Platycodon grandiflorum Vaccine 26, 3452-3460.

Publication Dates

• Publication in this collection
  May-Jun 2017

History

• Received
  30 Sept 2016
• Accepted
  20 Dec 2016