Gastroprotective effect of the aqueous fraction of hydroacetonic leaf extract of Eugenia uniflora L . ( Myrtaceae ) ( pitanga ) against several gastric ulcer models in mice

1 Faculty of Pharmacy, University Center of Anápolis Unievangélica, University Avenue, km 3,5, University City, s/n, 75083-515 Anápolis-GO, Brazil. 2 Department of Pharmacology, Institute of Biological Sciences, Federal University of Goiás, Campus Samambaia, Campus Street, s/n, 74001-97, Goiânia-GO, Brazil. 3 Institute of Chemistry, Federal University of Goiás, Campus Samambaia, Goiânia-GO, Brazil. 4 Institute of Chemistry, Federal University of Goiás, Brazil.


INTRODUCTION
The peptic ulcer disease (PUD) is one of the most common disorders of gastrointestinal tract (TGI) with a prevalence of 4 to 5% in human society (El-Maraghy et al., 2015).The PUD is a gastrointestinal disorder that occurs in the stomach and duodenum and generally is caused by an imbalance between aggressive and protective factors (Santin et al., 2010).It is well known that the major etiological factors involved in the onset of peptic ulcers are Helicobacter pylori infection, prolonged use of non-steroidal anti-inflammatory drugs (NSAIDs), ischemia of the gastric mucosa, age, genetic factors, stress, alcohol, smoking and dietary habits (Caldas et al., 2011).
Eugenia uniflora L. (Myrtaceae) is popularly known as cherry, ibipitanga, pitanga or naganpiri (Consolini and Sarubbio, 2002;Rattmann et al., 2012;Weyerstahl et al., 1988).This species of bushy plant with edible fruit is native to Brazil and widely distributed in South America countries (Lorenzi and Souza, 1999).The fruits are rich in calcium, anthocyanins, flavonoids, carotenoids and vitamin C that conferred high antioxidant property of this species.The pleasant flavor and odor make this specie a desirable content of ice cream, juices, jams, wines and cosmetics (Lima et al., 2002;Lopes, 2008).The E. uniflora leaves are rich in tannins and flavonoids (Auricchio and Bacchi, 2003) and several studies have shown that tannin rich species have been traditionally used for their gastroprotective effects (de Jesus et al., 2012).

Botanical material
The E. uniflora leaves were collected at Anápolis, Goias,Brazil (16°20' 12.8 "S,48°56' 19.3" W).The botanical material was authenticated by Prof. Dr. Heleno Ferreira Diaz of the Botany Department, Federal University of Goiás.A specimen voucher was deposited at the herbarium of the Federal University of Goiás (n°.25477).The concentrations of AFHP and drugs were adjusted to ensure that all treatments respect the volume of 10 mL/kg in purified water for oral treatment or in saline for i.p administrations.

Preparation of aqueous fraction of hydroacetonic leaf extract
Dried and ground leaves of E. uniflora (1.0 kg) were exhaustively extracted with 50% acetone, using an overhead stirrer apparatus at room temperature.The acetone was removed under reduced pressure and the suspended aqueous extract was filtered to eliminate fats and chlorophylls.Following, a liquid-liquid extraction with ethyl acetate (10 × 150 mL) was carried out.The combined organic phase was evaporated to yield an ethyl acetate fraction (15 g).The aqueous layer was freeze-dried to yield a 122 g aqueous fraction, which was stored at -20°C.

Animals
Swiss albino male mice, weighing 25 to 35 g from the Central Animal Laboratory of the Federal University of Goiás were used.The animals were housed at 22±1°C on a 12 h (h) light/dark cycle with free access to food and water.All experiments were approved by the Ethics Committee on Animal Use (Protocol number 038/14).

General pharmacological activity test
The groups of animals (n = 3) were treated orally or intraperitoneally with increasing doses of AFHP (100, 300 or 1000 mg/kg), while the control group received vehicle (distilled water 10 mL/kg, p.o.) or saline (10 mL/kg, i.p.).After the treatments, the animals were observed for 3 min under free ambulation on the flat surface after 5, 10, 20, 30 and 60 min; 4, 8, 24 and 48 h; 4 and 7 days of treatment.The observed behavioral changes that differentiate treated animals of the control group were reported in standard pharmacological screening form ( supplementary Table 1) adapted from Irwin proposal (Irwin, 1968).*Corresponding author.E-mail: jlfarmacia@hotmail.com.
Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License

Indomethacin-induced ulcer
After 16 h of fasting, animals (n = 8/group) were orally treated with vehicle (10 mL/kg), AFHP (100, 300 and 1000 mg/kg) or ranitidine (50 mg/kg).After 60 min of treatment, all animals received indomethacin (30 mg/kg, s.c.) and after 3 h from the administration of the ulcerogenic agent, all treatments were repeated.Animals were euthanized 6 h after the administration of indomethacin to removed the stomachs for the evaluation of lesion index (LI) (Djahanguiri, 1969).The LI and the percentage of gastric ulcer inhibition were calculated according to Rios et al. (2010) (Table 1).

Hypothermic restraint stress ulcer
After 16 h of fasting, the animals (n = 8/group) received vehicle (10 mL/kg, p.o), AFHP (300 mg/kg, p.o) or ranitidine (50 mg/kg, p.o).One hour after treatment, gastric ulceration was induced by immobilizing the animals in a closed cylindrical cage maintained at 4°C.After 2 h, the mice were euthanized to remove the stomach for LI assessment (Senay and Levine, 1967).
The animals were euthanized 1 h after induction of gastric lesions, while the stomachs were removed and opened along the greater curvature.The stomachs were photographed and the area of lesions (%) was measured by AUTOCAD software.

Parameters involved in gastric acid secretion
The pylorus ligature was performed by adaptation of method described by Shay et al. (1945).After 16 h of fasting, animals (n=8/group) were anesthetized and pylorus ligature was carried out.Mice received vehicle, AFHP (300 mg/kg) or ranitidine (50 mg/kg) intraduodenally (i.d.).Four hours later the animals were sacrificed, the stomachs were removed and the gastric luminal contents were centrifuged for 30 min at 2000 g.The supernatant was used to measure the gastric juice volume (mL), total acidity and pH.

Quantification of gastric wall adhered mucus
The modified method of Corne et al. (1974) was used to quantify gastric mucus.After 16 h of fasting, the animals (n=8) received vehicle, AFHP (300 mg/kg) or carbenoxolone (200 mg/kg) by gavage.After 60 min, all groups were orally treated with 60% ethanol solution (10 mL/kg, p.o.).The contents of the stomach was weighed and transferred to a test tube containing 7 mL of 0.1% Alcian blue (0.16 M sucrose in 0.05 M sodium acetate, pH 5.8).After two consecutive rinses with 5 mL of sucrose (0.25 M), 5 mL of MgCl2 (0.5 M) was added in each test tube for the extraction of mucus content with the dye.The glandular segment remained in this solution for 2 h with intermittent agitation.After which 4 mL of the resultant blue solution was agitated vigorously with 4 mL of ethyl ether until the formation of an emulsion and was centrifuged for 10 min at 3600 g.The absorbance of the supernatant was measured at 598 nm using a spectrophotometer.The concentration of Alcian blue was calculated through a calibration curve and the results were expressed in g of Alcian blue/g of glandular tissue.

Quantification of reduced glutathione (GSH)
After ethanol induced gastric ulcer, the other half segment of the glandular stomach area was weighed and transferred to a tube where the homogenate was done with ice-cold 0.02 M ethylenediaminetetraacetic acid 10% (EDTA).According to the method proposed by Sedlak and Lindsay (1968), 400 μL aliquots of homogenate were mixed with 320 μL of distilled water and 80 μL of 50% trichloroacetic acid in Eppendorf tubes and centrifuged at 3000 g for 15 min.Subsequently, the supernatant (400 μL) was mixed with 800 μL Tris HCl (0.4 M, pH 8.9) and 5, 5'-dithiobis-2nitrobenzoic acid (DTNB 0.01 M).The absorbance was read within 3 min at 412 nm.The concentration of GSH was calculated using a standard curve of reduced glutathione (GSH) expressed in g /g of tissue.

HCl/EtOH-induced gastric mucosa ulcer in mice pretreated with L-NAME
This method was performed as described by Matsuda et al. (1999).After a 16 h of fasting, the animals were pretreated with 0.9% saline (10 mL/kg, i.p.) or L-NAME, an inhibitor of NO synthase (20 mg/kg, i.p.).Thirty minutes later, animals received an oral dose of vehicle or AFHP (300 mg/kg).After 60 min, all groups were orally treated with 0.45M HCl/60% ethanol solution (10 mL/kg) to induce gastriculcer.After 1h, animals were euthanized and the stomachs were removed, opened along the greater curvature and gastric damage was determined as described above.

Statistical analysis
Results were expressed on means ± standard error of mean (SEM) absolute or percentage values and were compared using one-way analysis of variance (ANOVA) followed by the Tukey post-hoc test (to analyze more than two groups) or student unpaired "t" test (to analyze two independent groups (Drummond and Tom, 2011a, b).Effects were considered significant at p < 0.05.

Effects of AFHP in the general pharmacological evaluation
AFHP at doses of 100, 300 and 1000 mg/kg administrated intraperitoneally (i.p.) caused a reduction of spontaneous movements and induced writhing which was observed between 5 min to 4 h.The dose of 100 mg/kg caused analgesia and alienation from 5 min until 4 h at a dose of 300 to 1000 mg/kg and in addition to the changes described above catatonia, ataxia, diarrhea were also observed.Death of animals was recorded at the highest dose (i.p) within 24 h.The animals treated with different doses by oral route showed no behavioral changes that differentiate animals treated with vehicle (supplementary Table 1).

Effect of AFHP in gastric ulcer induced by nonsteroidal anti-inflammatory drug (NSAID)
The administration of indomethacin (NSAID) produced Results are expressed as mean ± SEM of the LI for eight mice.Statistical comparison was performed using ANOVA followed by the Tukey.***p ≤ 0.001 compared with the control group.
extensive lesions in the gastric mucosa.Treatment of mice with AFHP (100, 300 or 1000 mg/kg) significantly reduced the LI (Table 2).

Effect of AFHP in gastric ulcer induced by hypothermic restraint stress
In the gastric ulcer induced by stress in the model of hypothermic restraint, treatment with AFHP or ranitidine significantly reduced the LI when compared with the control group (Table 3).

Effect of AFHP on gastric ulcer induced by HCl / ethanol in mice
Administration of HCl/ethanol yielded extensive lesions in the gastric mucosa of the stomach.These lesions were characterized by multiple red or dark brown spots of different sizes along the gastric mucosa (Figure 1).Treatment of mice with AFHP significantly reduced the ulcerated area by 58.9% when compared to the control group (Table 4).

Effect of AFHP on parameters of gastric acid secretion in mice
The treatment with AFHP (300 mg/kg) was unable to decrease the volume of gastric secretion.However, AFHP increased the pH and decreased the total acidity.The treatment with ranitidine (50 mg/kg) caused a decrease in the volume of gastric acid secretion, total acidity and increased pH (Table 5).

Effect of AFHP on gastric adhered mucus
The alcian blue binding capacity of gastric mucus in the control group with lesion (ethanol 60%, 10 mL/kg, p.o.) was significantly reduced compared with the control group without injury.However, the groups of animals with lesions that were pretreated with AFHP or carbenoxolone significantly increased the alcian blue binding capacity of gastric wall mucus (Table 6).

Effect of AFHP on the amount of GSH in the stomach tissue
The GSH content in the control group with lesion (EtOH Results are expressed as mean ± SEM of the LI for eight mice.Statistical comparison was performed using ANOVA followed by the Tukey.*** p ≤ 0.001 compared with the control group.

A B C
Figure 1.Gastroprotective effects of AFHP or carbenoxolone on the HCl/ethanol-induced gastric lesion in mice.Results are expressed as mean ± SEM of the ulcerated area (%) for eight mice.Statistical comparison was performed using ANOVA followed by the Tukey ** p ≤ 0.01 *** p ≤ 0.001 compared with the control group.60%, 10 mL/kg, p.o.) was significantly reduced compared with control group without injury.However, the groups of animals with lesions that were pretreated with AFHP or carbenoxolone significantly increased the GSH content in 48.24 or 11.84%, respectively, when compared with control group with lesion (Table 6).

Effect of AFHP on HCl/ethanol-induced gastric mucosal lesion with L-NAME-pretreated mice
Treatment with HCl/EtOH induced extensive lesions in the gastric mucosa of the stomach.However, treatment with AFHP significantly reduced the ulcerated area in Results are expressed as mean ± SEM for eight mice.Statistical comparison was performed using ANOVA followed by the Tukey test.* p ≤ 0.05, ** p ≤ 0.01, control without lesion vs control with lesion; # p ≤ 0.05, ## p ≤ 0.01, carbenoxolone or AFHP vs control with lesion.64.2%.Pretreatment with L-NAME did not reverse the gastroprotective activity of AFHP (Table 7).

DISCUSSION
Due to the series of side effects associated with the first line of antiulcer drugs, the study of medicinal plants derived compounds for the treatment of various gastrointestinal disorders is becoming important around the world (Zheng et al., 2014).The phytochemical study of Eugenia uniflora allowed the isolation and structure elucidation of several phenolic substances of three types: galoil esters, ellagitannins monomeric and dimeric flavonoids and glycosides.Among the compounds identified in the aqueous fraction of hidroacetonic extract of E. uniflora leaves are: oenothein B, myricitrin, quercitrin, eugeniflorin D2 and camptothin A (Fortes et al., 2015).The present study evaluated the gastroprotective effect of aqueous fraction of hydroacetonic leaf extract of pitanga (AFHP) in different models of experimentally induced gastric ulcers and the possible mechanisms of actions involved in this effect.According to Parmar and Desai (1993) various mechanisms in different experimental models of gastric ulcers make it impossible to think of a single mechanism of gastroprotective activity.Recent studies have shown that the leaf extract of E. dysenterica and E. punicifolia have gastro-protective activity (Basting et al., 2014;Prado et al., 2014).
An important factor involved in the pathogenesis of gastric lesions induced by anti-inflammatory nonsteroidal drugs (NSAIDs) is a deficiency of endogenous prostaglandins (PGs).Prostaglandins, particularly PGE 2 and PGI 2 are described as key mediators in gastric mucosal defense.Its cytoprotective effect has been associated with the stimulation of mucus/bicarbonate secretion, maintenance of mucosal blood flow and inhibition of gastric acid secretion (Martins et al., 2014;Sousa et al., 2013).
The indomethacin (NSAID) inhibits the production of prostaglandins leading to a decrease of the formation and release of mucus/bicarbonate and increases the production of HCl, thereby favoring the appearance of gastrointestinal ulcers (Wallace and Devchand, 2005).Our data suggest that oral treatment with AFHP (100, 300 and 1000 mg/kg), ranitidine (50 mg/kg) protected the gastric mucosal lesions induced by NSAIDs.These results suggest possible involvement of mucus and or PGs.
The stress induced ulcers occur as a result of stressful events such as, burns, sepsis, surgery and trauma (de Almeida et al., 2012).Several studies show that disturbances in gastric secretion, changes in microcirculation and abnormal gastric motility are the possible mechanisms involved in stress-induced ulcers (Amany and Ibrahim, 2013;Batista et al., 2004).Stress reduces endogenous glutathione levels and promotes generation of reactive species such as OH .and inhibit the biosynthesis of mucosal prostaglandins by H 2 O 2 accumulation (Bandyopadhyay et al., 2002).The cold restraint stress model (4°C, 2 h) has been widely used to evaluate anti-ulcer activity (Viana et al., 2013).The results suggest that the gastroprotective activity AFHP in stress -induced gastric injury is probably mediated by its antisecretory and antioxidant activity.GSH is a major mediator involved in maintaining the integrity of the gastric mucosa, due to its antioxidant capacity (Mutoh et al., 1990).It is known that gastric lesions induced by stress and EtOH are associated with significant decrease in mucosal levels of GSH (Szabo and Vattay, 1990).The administration of AFHP significantly increased the GSH levels in gastric mucosa.This result suggests that AFHP has antioxidant activity.
The effect of AFHP was also evaluated in ulcer model induced by HCl/EtOH.It is known that the ingestion of large amounts of ethanol promotes gastric lesions in humans and experimental animals (Siegmund et al., 2003;Teyssen and Singer, 2003).It is well established that the formation of gastric mucosal lesions by necrotizing agents such as HCI/EtOH enhances lesions and reduce the number of gastric defense mechanisms such as disruption of blood flow, degranulation of mast cells, reduction of prostaglandin and mucus/bicarbonate release (Abdel-Salam et al., 2001;Glavin and Szabo, 1992).Oral administration of HCl / EtOH in the control group promoted necrotic lesions and bleeding characteristics.Based on the results obtained from this model, AFHP reduced lesion area.An increase in prostaglandin synthesis or mucus adhered to gastric mucosa may be one of the responsible for the antiulcerogenic mechanisms.These results are in agreement with those found by Viana et al. (2013).Mucus is one of the most important parameters that contribute to the protection of gastric mucosa.Gastric mucus is responsible for the first line of defense of the gastric mucosa and consists of a transparent viscous, elastic, adherent gel that is made up of water and glycoproteins (Martins et al., 2015).The mucus layer is a physical barrier that adheres together with bicarbonate and protects the underlying mucosa from proteolytic digestion (Allen and Flemstrom, 2005).When the mucus barrier is damaged, the gastric mucosa becomes more susceptible to gastric acid induced ulcers (Santin et al., 2010).This study revealed that the amount of adhered gastric mucus was increased by treatment with AFHP and this increase probably contributed to the cytoprotective effect of AFHP.Pyloric ligation model produces biochemical parameters in gastric mucosa such as volume, pH and total acidity.The pyloric ligation interferes with gastric mucosa and changes in the levels of prostaglandins, cytokines and endogenous glutathione (Singh et al., 2008).
In an attempt to determine the gastroprotection mechanisms of AFHP, the parameters of gastric acid secretion were evaluated.We found that intraduodenal administration of AFHP significantly reduced total acidity and increased pH.This result indicates a systemic action in addition to anti-secretory activity of AFHP.The reference drug, ranitidine (50 mg/kg), significantly reduced the volume of gastric juice, total acidity and increased the pH.
It is known that nitric oxide (NO) plays an important role in the defense of the gastric mucosa and is a biological mediator which regulates the secretion of mucus and blood flow (Falcao Hde et al., 2013).In the gastrointestinal tract, NO is also involved in modulating the activity of mast cells together with endogenous prostaglandins (Klein-Junior et al., 2013;Wallace, 2006).Our results showed that the inhibition of NO synthase by L-NAME did not reverse gastroprotection effect of AFHP, thereby suggesting that NO synthesis is not critical to its gastroprotective activity.

Conclusion
Overall findings in the current study showed the effectiveness of AFHP in preventing gastric ulcers against lesions induced in various experimental models.These effects could be associated with the gastric cytoprotective mechanisms including the participation of mucus and GSH levels on the mucosa.Further studies will be focused to elucidate the phytochemical(s) responsible(s) for the antiulcer mechanism of AFHP.

Table 1 .
Score attribution scale for the degree of ulceration.

Table 2 .
Effects of AFHP or ranitidine on indomethacin induced ulcers in mice.

Table 3 .
Effects of AFHP or ranitidine on stress-induced gastric lesions in mice.

Table 4 .
Effects of AFHP or carbenoxolone on HCl/ethanol-induced gastric lesions in mice.

Table 5 .
Effects of AFHP or ranitidine extract, administered intraduodenally, on the biochemical parameters of gastric juice obtained from pylorus ligature in mice.

Table 6 .
Effect of oral treatment of AFHP or carbenoxolone on the gastric adhered mucus and GSH in the model of ethanol (60 %, 10 mL/kg, p.o.) in mice.

Table 7 .
Role of nitric oxide (NO) in the gastroprotective effect of AFHP against HCl/EtOHinduced gastric injury in mice.
Results are expressed as mean ± SEM for eight mice.Statistical comparison was performed using ANOVA followed by the Tukey test.. *** p ≤ 0.001 compared with the control group (treated with saline + vehicle).