Gastroprotective Effects of β-sitosterol and β-sitosterol-3-O-β- D-glucopyranoside from Bridelia ferruginea Stem Bark

Corresponding Author: Adaobi Chioma Ezike Department of Pharmacology and Toxicology, Faculty of Pharmaceutical Sciences, University of Nigeria, Nsukka, 410001, Enugu State, Nigeria Email: adaobi.ezike@unn.edu.ng Abstract: The methanol extract of Bridelia ferruginea Benth. (Euphorbiaceae) stem bark (BFME) was partitioned in chloroformmethanol-water (2:2:1) mixture to obtain the Chloroform (CF) and Aqueous Methanol (AMF) fractions. The BFME, CF and AMF were screened for antiulcer activity using indomethacin-induced ulcer as activity guide. The CF provided the highest gastroprotection and was subsequently fractionated in a silica gel (60-200 mesh) column eluted with different mixtures of n-hexane and ethyl acetate (100:0; 95:5; 90:10; 80:20) to obtain six fractions (I-VI). Fractions III and VI offered the highest protection against indomethacin-induced ulcer and were further purified in a sephadex LH-20 column eluted with methanol to yield two compounds, BF1 and BF2. Using Nuclear Magnetic Resonance (HNMR, C-NMR) and electron impact mass spectroscopies, BF1 and BF2 were confirmed to be β-sitosterol and β-sitosterol-3-O-βD glucopyranoside respectively. The BFME, fractions, β-sitosterol and βsitosterol-3-O-β-D-glucopyranoside elicited dose-related and significant (P<0.05) protection against various ulcers in rats. β-sitosterol, 100 and 300 mg/kg, produced 79.70, 82.18, 42.31, 44.87, 65.97, 70.83, 80.22 and 87.91% gastroprotection; while β-sitosterol-3-O-β-D-glucopyranoside, 100 and 300 mg/kg, caused 69.80 and 74.26, 33.33, 35.26, 84.03, 95.83, 83.52 and 85.71% gastroprotection against indomethacin-, ethanol-, cold restraint stressand pylorus ligationinduced ulcers, respectively. Results demonstrated gastroprotective effects of B. ferruginea stem bark, attributable to β-sitosterol and β-sitosterol-3-O-β-D-glucopyranoside.


Introduction
Peptic ulcer is usually due to an imbalance between the injurious (acid, pepsin and Helicobacter pylori) and defensive (mucus, bicarbonate, prostaglandins, nitric oxide, some peptides) factors of the gastric mucosa (Wallace and Sharkey, 2011;Love and Thoma, 2014;Turner, 2015). To re-establish the balance, pharmacotherapeutic agents are used to inhibit gastric acid secretion, eliminate H. pylori infection, or enhance the mucosal defensive mechanisms.
Numerous plants are used in ethnomedicine to treat peptic ulcer disease. Various scientific studies have demonstrated the antiulcer efficacy of some of these medicinal plants (Falcão et al., 2008;Ezike et al., 2009;Vimala and Shoba, 2014). One of such plants with antiulcer activity is Bridelia ferruginea Benth (Euphorbiaceae).
Earlier studies in our Laboratory demonstrated the antiulcer (Ezike et al., 2011) and wound healing (Ezike et al., 2015) potentials of B. ferruginea stem bark. To further elucidate the antiulcer properties of the plant, this study sought to isolate and characterize the antiulcer compound(s) from extracts of B. ferruginea stem bark.

Chemicals, Reagents and Drugs
Solvents including those used for extraction, fractionation, separation and isolation of the pure compounds were purchased from Sigma-Aldrich (Darmstadt, Germany). Indomethacin was obtained from Sigma Aldrich (St. Louis, MO, USA), while cimetidine was procured from Jiangxi Xierkangtai Pharmaceutical Co. Ltd. (North Zone, High-New Technology Industrial Zone, Pingxiang, Jiangxi, China).
Animals: Adult Sprague Dawley rats (120-150 g) of either gender (the female rats were 12-13 weeks old, while the male rats were 11-12 weeks old) bred in the Laboratory Animal Facility of the Department of Pharmacology and Toxicology, University of Nigeria, Nsukka were used for the study. The animals were maintained ad libitum on standard pellets and water. All animal experiments were in compliance with National Institute of Health Guide for Care and Use of Laboratory Animals (Pub No. 85-23, revised 1985) and with prior permission from the National Health Research Ethics Committee (NHREC) of the University of Nigeria, with protocol clearance number NHREC/05/01/2012C.

Preparation of Extract
Fresh stem bark of B. ferruginea trees growing in Ezimo, Enugu State, Nigeria were collected in October. The plant was identified and authenticated by a taxonomist at the International Center for Ethnomedicine and Drug Development (InterCEDD), Nsukka, Nigeria. A voucher specimen (specimen number: BFSB2011) of the plant was kept in the Herbarium of the Department of Pharmacognosy, Faculty of Pharmaceutical Sciences, University of Nigeria, Nsukka. The bark was separated from the wood stem, cleared of the dead outer parts and thorns, cut into smaller pieces and dried under shade for seven days. The dried plant material was pulverized to powder using a milling machine.
The powdered plant material (8 kg) was extracted by cold maceration in methanol for 48 h at room temperature (28±1°C) and the mixture filtered. Subsequently, the plant material was repeatedly washed with methanol until the filtrate became clear. Removal of the solvent in vacuo using a rotary evaporator (40°C) afforded 2.01 kg of the methanol extract (BFME; 25.125% w/w).

Fractionation of the Extract
The BFME (2 kg) was partitioned, in eight separate batches, in a chloroform-methanol-water (2:2:1) mixture. The resulting chloroform layer was concentrated in vacuo using a rotary evaporator (40°C) to obtain 38.1 g of the Chloroform Fraction (CF; 1.905% w/w). While the aqueous-methanol portion was freezedried to afford 126.5 g of the Aqueous Methanol Fraction (AMF; 6.325% w/w).

Biological Activity Guided Studies and Isolation of Bioactive Constituents
The BFME and fractions were subjected to biological activity-guided studies using indomethacin-induced ulcer as activity guide. The CF offered higher protection than the AMF. Subsequently, CF (38 g) was further fractionated in a silica gel (60-200 mesh) column eluted with gradient mixtures of n-hexane and ethyl acetate (100: 0; 95: 5; 90:10; 80: 20). About 37 fractions of 100 ml each were collected. The fractions were pooled based on the similarity of constituents visualized on silica gel coated TLC plates developed with n-hexane-ethyl acetate (7:3) mixture to afford 6 broad fractions: fraction I (1.08 g; 2.84% w/w), fraction II (1.2 g; 3.16% w/w), fraction III (8.7 g; 22.89% w/w), fraction IV (1.5 g; 3.95% w/w), fraction V (1.65 g; 4.34% w/w) and fraction VI (11.1 g; 29.21% w/w). The fractions were also screened for antiulcer activity. Fractions III and VI gave comparably higher protection against indomethacin-induced ulcer. Further purification of fractions III (8.0 g) and VI (10.5 g), separately, by column chromatography using sephadex LH-20 with methanol as eluent yielded two white waxy amorphous powders; compound I (BF1; 2.60 g; 32.5% w/w) and compound II (BF2; 2.85 g; 27.14% w/w) respectively. The antiulcer activities of BF1 and BF2 were confirmed using the activity guide.

Characterization of Isolated Compounds
The molecular weights of the isolated compounds BF1 and BF2 were determined by Electron Impact Mass Spectroscopy (EI-MS). Electron impact mass spectra were measured on a Finnigan MAT 8430 mass spectrometer which uses energy for ionization (70 eV) achieved by accelerating the electrons through a potential drop of 70 V.
The molecular structures of the compounds were elucidated using 1 H-NMR and 13 C-NMR, as applicable. The spectra were recorded at 300 K on ARX 600 MHz NMR spectrometers. All the one-dimensional and twodimensional spectra were obtained using the standard Bruker software with Tetramethylsilane (TMS) as internal standard reference signal. The observed chemical shifts (δ) were recorded in parts per million (ppm) and the coupling constants (J) were recorded in Hertz (Hz).
Proton signals of BF1 were captured and assigned using 1 H-1 H Correlation Spectroscopy ( 1 H-1 H COSY). Also, proton and carbon signals of BF2 were captured and assigned using 1 H-1 H COSY, Distortionless Enhancement by Polarization (DEPT), Heteronuclear Single Quantum Coherence Spectroscopy/Heteronuclear Single Quantum Coherence Correlation (HSQC) and Heteronuclear Multiple Bond Correlation Spectroscopy (HMBC) as applicable.

Indomethacin-Induced Ulcer Test
This was carried out according to the method described by (Santin et al., 2010) with some modifications. Rats starved of food for 24 h prior to the experiment, but with free access to water, were randomly placed into 17 groups (n = 5). Groups 1-15 received oral administrations of one of BFME, CF, AMF, fractions I -VI, BF1 and BF2 at 100 and/or 300 mg/kg respectively. Groups 16 and 17 were the control groups and received cimetidine (100 mg/kg) and 1% Tween-80 (5 mL/kg), respectively. One hour after treatment, indomethacin (100 mg/kg) was given to the animals orally. Four hours later, the animals were sacrificed by cervical dislocation and their stomachs removed and opened along the greater curvature. The stomachs were rinsed carefully under running water, pinned on a cork board and examined with a hand lens (10). Using a modification of the method described by (Main and Whittle, 1975), ulcers formed on the glandular portion of the stomach were observed and each graded on a 0-3 scale based on the length; 1 = <1 mm; 2 = 1-3 mm; and 3 = >3 mm. The Ulcer Index (UI) was calculated as (1× number of ulcers grade 1) + (2× number of ulcers grade 2) + (3× number of ulcers grade 3). The overall score was divided by a factor of 10 and the mean score for each group calculated. Ulcer protection (%) of the treated groups were calculated using the relation:     % 100 1 / Ulcer protection y z  Where: y = Ulcer index of treated group z = Ulcer index of control group (Ezike et al., 2014)

Absolute Ethanol-Induced Ulcer Test
This was carried out according to the method described by (Santin et al., 2010) with some modifications. Rats starved of food for 24 h prior to the experiment, but with free access to water, were randomly divided into 17 groups (n = 5). Groups (1-15) received oral administrations of one of BFME, CF, AMF, fractions I-VI, BF1 and BF2 at 100 and/or 300 mg/kg respectively. Groups 16 and 17 were the control groups and received cimetidine (100 mg/kg) and 1% Tween-80 (5 mL/kg), respectively. One hour after treatment, each animal received 1 mL of absolute ethanol orally. One hour after ethanol administration, the animals were sacrificed by cervical dislocation and their stomachs removed and opened along the greater curvature. The stomachs were rinsed carefully under running water, pinned on a cork board and examined with a hand lens (10). Ulcers formed on the glandular portion of each stomach were observed and scored using a 0-7 scale, as described by (Ezike et al., 2014). Where 0 = No ulcer, 1 = One slight ulcer, 2 = More than one grade 1 ulcer, 3 = One ulcer of length ≤0.5 cm, 4 = One ulcer of length >0.5 cm, 5 = More than one grade 3 ulcer, 6 = More than one grade 4 ulcer, 7 = Complete hemorrhagic lesion of the mucosa. The overall score was divided by a factor of 10. Mean ulcer score for each group was calculated and expressed as the Ulcer Index (UI). Ulcer protection (%) of the treated groups were calculated as described for indomethacin-induced ulcer.

Cold Restraint Stress-Induced Gastric Ulcer Test
This was carried out according to the method described by (Viana et al., 2013) with some modifications. Rats starved of food for 24 h prior to the experiment, but with free access to water, were randomly placed into twelve groups (n = 5). Groups 1-10 received oral administrations of one of BFME, fraction III, fraction VI, BF1 and BF2 at 100 and/or 300 mg/kg respectively. Groups 11 and 12 were the control groups and received cimetidine (100 mg/kg) and 1% Tween-80 (5 mL/kg), respectively. One hour after treatment, each rat was restrained in a closed cylindrical cage maintained at 4-6°C. After 4 h, the animals were sacrificed by cervical dislocation and their stomachs removed and opened along the greater curvature. The stomachs were rinsed carefully under running water, pinned on a cork board and examined with a hand lens (10). Ulcers were graded, also UI and ulcer protection (%) were determined as described for indomethacin-induced ulcer.

Pylorus Ligation-Induced Ulcer Test
The pylorus ligation-induced ulcer test was performed using the method described by (Shay, 1945) with some modifications. Rats starved of food for 24 h prior to the experiment, but with free access to water, were randomly placed into six groups (n = 5). The animals were anesthetized i.p. with phenobarbital sodium (35 mg/kg), the abdomen incised and the pylorus ligated. Immediately after pylorus ligation, groups 1-4 received one of BF1 and BF2 at 100 or 300 mg/kg administered intraduodenally respectively, while control animals (groups 5 and 6) received either 1% Tween-80 (5 mL/kg) or cimetidine (100 mg/kg). The stomachs were replaced carefully and the animals were allowed to recover. Four hours later, the animals were sacrificed by cervical dislocation, the abdomens opened and another ligature placed around the esophagus close to the diaphragm. The stomachs were removed, opened along the greater curvature, rinsed carefully under running water and pinned on a cork board for examination with a hand lens (10). Ulcers were graded, also UI and ulcer protection (%) were determined as described for indomethacin-induced ulcer.

Statistical Analysis
Data were analyzed using One-Way ANOVA in GraphPad Prism 7.0. Results were expressed as Mean ± SEM. Differences between means were determined using Bonferroni's post hoc test for multiple comparison and regarded significant at P<0.05.

Phytochemistry Physical Characteristics, Spectral Data and Structure of BF1
β-sitosterol (BF1) was isolated as a white, waxy amorphous powder, soluble in chloroform, but insoluble in water. The molar mass was deduced as 414 g/mol based on the EI-MS m/z peaks at 414, 415 and 416.
The major fragment at m/z 396 occurred by a loss of 18 amu (loss of H2O molecule). This suggested the presence of OH group in the compound. The BF1 was identified to be a steroid or triterpene with hydroxyl group at position 3. These observations suggested that BF1 was most likely a β-sitosterol.

Discussion
Assessment of the antiulcer activity of B. ferruginea stem bark showed that the methanol extract, its fractions and isolated compounds (β-sitosterol and β-sitosterol-3-O-β-D-glucopyranoside), significantly protected the gastric mucosa against indomethacin-, absolute ethanol-, cold restraint stress-and pylorus ligation-induced ulcers.
The extract, fractions and isolated compounds significantly protected the rat stomach against indomethacin-induced lesions. Indomethacin, a Non-Steroidal Anti-Inflammatory Drug (NSAID), causes gastroduodenal ulceration by suppressing the production of prostaglandins (Wallace, 2001), increasing the secretion of gastric acid (Wallace, 2001) and generation of free radicals (Lichtenberger, 1995). Endogenous prostaglandins help to maintain mucosal integrity under injurious and adverse conditions by exerting negative feedback inhibition of acid secretion, mediating functional vasodilation in the gastric mucosa (Main and Whittle, 1973;1975), maintaining mucosal blood flow and gastric microcirculation (Vane, 1971) and production of bicarbonate and mucus (Robert, 1982). Therefore, suppression of prostaglandins synthesis by indomethacin and other NSAIDs impedes the vasodilator, antisecretory and other protective mechanisms of prostaglandins, causing increased susceptibility to gastric mucosal lesions. Thus, indomethacin produces erosion and bleeding in the gastric mucosa by reduction in mucosal blood flow and increase in gastric acid production (Main and Whittle, 1975). Also, mucosal damage elicited by NSAIDs is characterized by changes in its permeability to ions, Ulcer index 2.5 2 1.5 1 0.5 0 Treatment 100 mg/kg 300 mg/kg water and protein, which further erode the mucosal integrity. Hence the ability of the bioactive constituents of B. ferruginea stem bark to protect against indomethacin-induced ulcer suggests antisecretory and cytoprotective activity. Ethanol produces severe hemorrhagic erosion on the glandular part of the stomach (Ezike et al., 2009) by reduction of bicarbonate and mucus, increase in leukotrienes and free radicals which cause lipid peroxidation and cell damage (Peskar et al., 1986;Glavin and Szabo, 1992;Ezike et al., 2009). It also reduces endogenous glutathione and prostaglandin levels and increases the release of histamine and influx of calcium ions (Glavin and Szabo, 1992). The observed effects of the bioactive constituents of B. ferruginea stem bark on ethanol-induced ulcer indicate gastroprotection through augmentation of the defensive factors.
The pathology of stress ulcer involves increased gastric acid secretion due to histamine release (Kitagawa et al., 1979), oxidative damages mainly due to hydroxyl radicals (Das and Banerjee, 1993;Liu et al., 1996;Das et al., 1997;Bandyopadhyay et al., 2002;Brzozowski et al., 2008), reduction of blood flow and mucus production in the gastric mucosa (Kitagawa et al., 1979), pancreatic juice reflux (Guth, 1972), inhibition of gastric mucosal prostaglandin synthesis (Brzozowski et al., 2008;Nur Azlina et al., 2013), inflammatory responses (Bregonzio et al., 2003) and increased vagal activity (Brodie and Hanson, 1960;Grijalva and Novin, 1990). Stress ulcers in humans and rats share similar pathophysiology (Konturek et al., 2003). Stress and distress decrease blood flow in the upper gastrointestinal tract (Kauffman Jr, 1997) and may render the human stomach and duodenum more susceptible to damage (Levenstein, 2002). Furthermore, human studies have shown that stress, anxiety and depression increase acid secretion (Feldman et al., 1992), impair ulcer healing and promote ulcer relapse (Levenstein, 2002). Stress also causes an increase in gastrointestinal motility resulting in folds in the stomach (Peters and Richardson, 1983) which increase susceptibility to damage by acid (Brodie and Hanson, 1960). Stress also decreases the quality and amount of mucus adhering to the gastric mucosa. Due to the significant role of mucus in protection and enhanced healing of the stomach membranes, stressinduced ulcer models are usually deployed to evaluate gastroprotective agents. The bioactive constituents of B. ferruginea stem bark protected the rat gastric mucosa against stress ulcers suggesting cytoprotective activity, potential to reduce acid secretion and shield against oxidative damage and enhancement of mucosal barrier and healing mechanisms.
Gastric acid accumulates in the stomach when the pylorus is ligated (Brodie, 1966), causing formation of gastric ulcers. Ulceration produced by pylorus ligation is due to auto-digestion and subsequent collapse of gastric mucosal barrier, secondary to high concentration of acid (Tovey, 2015). The bioactive constituents of B. ferruginea stem bark elicited remarkable and significant gastroprotective effects in rats subjected to pylorus ligation suggesting decrease in total acidity and volume of gastric secretion, hence antisecretory activity.
Bioactivity-guided fractionation technique was employed to aid the isolation of the antiulcer constituents of B. ferruginea stem bark. Successive separation and pharmacological screening using indomethacin-induced ulcer as bioactivity guide revealed that two steroids isolated from the plant, β-sitosterol and β-sitosterol-3-Oβ-Dglucopyranoside possess gastroprotective activity. This is the first report of the isolation of these compounds from the stem bark extract of B. ferruginea. They are phytosterols which are steroid compounds composed of plant sterols and stanols related to cholesterol and differ in carbon chains and or presence or absence of double bond. β-sitosterol was more active against indomethacin-and ethanol-induced ulcers, while β-sitosterol-3-O-β-Dglucopyranoside elicited more protection against cold restraint stress-induced ulcer. However, there was hardly any difference in the degree of protection against pylorus ligation-induced ulcer elicited by both compounds. Earlier studies reported the gastroprotective effects of these and related phytosterols (Arrieta et al., 2003;Xiao et al., 1992;Navarrete et al., 2002). Phytosterols have been demonstrated to reduce the permeability of phosphatidyl-choline bilayers to water and enhance stability of phospholipid monolayers (Hąc-Wydro et al., 2007;Tovey, 2015), reduce the leakage of proton and sodium ions from cell membranes (Haines, 2001) and inhibit the release of histamine from peritoneal mast cells (Shoji et al., 1994); these may partly account for the observed cytoprotective effects of β-sitosterol and β-sitosterol-3-O-β-glucopyranoside isolated from B. ferruginea stem bark.

Conclusion
The results from this study reveal β-sitosterol and βsitosterol-3-O-β-D-glucopyranoside as antiulcer constituents of B. ferruginea stem bark and account for antiulcer effects of the plant. These phytosterols may serve as lead compounds to develop novel antiulcer agent(s) with improved efficacy and safety profile.

Author's Contributions
Nnamani EM: Participated in the study design, conducted the experiments and data analysis.
Akah PA and Okoli CO: Participated in the study design.
Ezike AC: Participated in the study design, experiments and data analysis.
Tchineme M: Participated in phytochemistry and some other aspects of the experiments.
All the authors participated in the preparation of the manuscript.

Ethics
This article is original and contains unpublished material. The corresponding author confirms that all of the other authors have read and approved the manuscript and no ethical issues involved.