Ellagic acid (EA), a tannin was isolated from Eucalyptus citriodora leaves and its anti-inflammatory activity

A tannin, EA (1), and other nine non-tannins compounds, gallic acid (2), quercetin (3), myricetin (4), 3-O-methylellagic acid-4′-O-α-L-rhamnopyranoside (5), quercetin-3-O-β-D-galactopyranoside (6), kaempferol-3-O-β-D-glucoside (7), quercetin-3-O-β-D-glucuronide (8), quercetin-3-O-rutinoside (9), 3,3′,4-tri-O-methylellagic acid-4′-O-β-D-glucopyranosyl (10) were isolated from a valuable medicinal plant, Eucalyptus citriodora. Structural identification of these compounds was conducted using 1H NMR and 13C NMR spectroscopy and comparing their spectral data with those previously reported in literatures. The anti-inflammatory effects of EA were evaluated in ethanol-induced acute gastric ulcer mice models in our study. The result demonstrated that the intragastric administration of EA significantly prevented the gastric ulceration caused by ethanol treatments. Especially, the gastric tissue in the middle-dose EA (100 mg/kg) showed few ulcerations with only slight focal congestion which indicated that it has a significant protective effect on gastric ulcer by increasing the IL-10 and PGE2 levels, and reducing the IL-6, TNF-α, GAS and COX-2 levels. In addition, the middle-dose EA has no adverse effect on liver and kidney. These findings imply that EA exerts gastroprotective effects by means of its anti-inflammatory effects and may be a potential drug for anti-ulcer treatment.


Introduction
The gastric ulcer is a common disorder with multiple causes including the infection from Helicobacter pylori, indiscriminate use of nonsteroidal anti-inflammatory drugs, irregular diet, excessive alcohol consumption and pepsin and gastric acid secretion and is mainly occurs in the stomach and duodenum [1]. Nowadays, it is commonly believed that gastric ulcer therapy is generally focused on inhibiting gastric acid secretion involved in suppressing the secretion of interleukin-6 (IL-6), tumor necrosis factoralpha (TNF-α), and stimulate mucosal defense mechanism [2]. However, many of existed medicines have side effects and limitations, including the antibiotics used to eradicate Helicobacter pylori and the proton pump inhibitors [3].
In recent years, a lot of research has been accomplished to develop natural products with anti-ulcer activity, which may provide a rich source of new anti-ulcer drugs [4]. A few plant extracts and plant-derived compounds have been identified and confirmed to have powerful anti-ulcer activity with safety [5]. Among a broad reach of natural molecules, dietary polyphenols have a variety of biological mechanisms and play a pivotal role in the treatment of gastric and duodenal ulcers.
Eucalyptus citriodora is a medicinal plant that belongs to the Myrtle family and is found in Australia and some oriental countries, especially China. Eucalyptus citriodora leaves extract has a variety of biological activities such as anti-bacterial, anti-fungal, anti-hyperglycemia, anti-inflammatory, and antioxidation for it contains a rich resource of health-promoting bio-active compounds such as flavonoids and polyphenol [6]. Therefore, Eucalyptus citriodora is used in many herbal preparations to treat a range of diseases, including colds, fever, flu and inflammations [7]. Dharawal people are Indigenous Australians and they traditionally used Eucalyptus citriodora leaves to treat inflammatory, wounds and fungal infections. Among the polyphenol compounds extracted from Eucalyptus citriodora, Ellagic acid (EA), a phenolic lactone compound, exhibits anti-inflammatory, antioxidant, hepatoprotective, and anti-mutagenic effects. EA could decrease gastric ulcer and bleeding induced by ethanol in which attributing to the enhancement of NO production [8]. Furthermore, EA have abilities to reduce stress gastric injury and inhibit gastric acid secretion significantly [9]. In the present study, we evaluated the antiulcerogenic effects of different doses of EA extracted from Eucalyptus citriodora against ethanolinduced gastric ulcer and to investigate its effects on inflammatory cytokines, including IL-6, IL-10, TNF-α, GAS, COX-2, and PGE 2 . In parallel, we evaluated the effect of EA on the hepatic and nephridial by the activities of alanine aminotransferase (ALT), aspartate aminotransferase (AST), and creatinine (Cr) in serum.

Identification of isolated compounds
Ten compounds were isolated and identified in the 95% EtOH extracts of the Eucalyptus citriodora leaves (Fig. 1). These compounds were identified by spectroscopic analyses, including MS and NMR spectroscopy and comparing their spectral data with those previously reported in literature.
Among the ten compounds, EA is a plant-derived polyphenol with anti-inflammatory, antioxidant, gastroprotective, hepatoprotective and anti-mutagenic effects. Its gastric protective mechanisms in ethanol-induced acute gastric ulcers have not been specifically elucidated and described. Therefore, the anti-ulcer activities of EA were evaluated in ethanolinduced acute gastric ulcer models in KM mice.

Effect of EA on inflammatory factors in gastric ulcer tissues of mice
After administration with anhydrous ethanol, the IL-10 level in gastric tissues were significantly decreased while the IL-6 and TNF-α levels were on the contrary (P < 0.05). In addition, the IL-10 level in the gastric tissues of EA were markedly increased in comparison to that of the model group (P < 0.05 or P < 0.01), while the IL-6 and TNF-α levels were significantly reduced (P < 0.05). Obviously, the reduction of TNF-α in the middle-dose of EA was more significant than that in the low-dose and high-dose EA groups (Fig. 2).
Effects of EA on GAS, COX-2 and PGE 2 in gastric ulcer tissues of mice The serum GAS contents of model group were significantly higher than those of the normal control group, however, the COX-2 and PGE 2 contents were on the contrary (P < 0.05). Additionally, the administration of middle-dose and high-dose EA showed a noticeable reduction in the GAS levels of serum (P < 0.05), while the PGE 2 levels were significantly elevated compared with that of model group (P < 0.01) (Fig. 3).

Effects of EA on ALT, AST and Cr in gastric ulcer tissues of mice
The ALT, AST, and Cr activities were significantly elevated in the model group in comparison to the normal control group (P < 0.05). The ALT and AST levels in the high-dose EA group were significantly lower than that of the normal control group, model group and omeprazole group (P < 0.01). Moreover, the ALT, AST, and Cr levels in the middle-dose EA group were close to the positive control omeprazole group (P < 0.05) (Fig. 4).

Gross appearance of the gastric mucosa
The gastric mucosal membrane surface of the normal control group (Fig. 5A) was smooth and intact with no hyperemia. In model group (Fig. 5B), a number of punctual ulcers distributed in mucosa were observed. There were some hyperemia points and a small ulcer area can be found in the omeprazole group (Fig. 5C). In the EA administration group (Fig. 5D, E, F), the degree of punctual bleeding, damage of gastric mucosa, and the area of gastric ulcers in mice were attenuated, suggesting that the EA has a protective effect on ethanol-induced acute gastric ulcer in the mice.

Effects of EA on gastric lesions histology
In order to evaluate the status of gastric mucosa in ulcer mice, H&E staining was performed on stomach sections of different groups. The gastric tissue of the normal control group (Fig. 6A, a) is dense and well-structured, the mucosal layer and muscular layer are clearly visible, and there are no apparent pathological changes such as tissue hemorrhage or inflammatory cell infiltration. The model group mice (Fig. 6B, b) administrated absolute ethanol without medical therapy produced several necrotic cell fragments and a large number of infiltrated inflammatory cells. Besides, the smooth muscle structure in the model group was disordered, the superficial gastric epithelium was disrupted and exfoliated and there were vacuoles and cell damage appear in the local gastric glands. In the omeprazole group (Fig. 6C,  c), the glands were arranged consistently and there is a small amount of inflammatory cell infiltration. The mice that administrated pretreatment with EA displayed significant protection of the gastric mucosa (Fig. 6D, d, E, e, F, f), most areas of the stomach tissue of the mice are clearly structured, the cells are arranged neatly, and the degree of gland damage is less, which is significantly enhanced versus with the model group. Among them, the gastric tissues of mice in the low-dose EA group have no difference from the normal group and the mucosal and muscle layer are clearly visible and textured, and only a few inflammatory cells were infiltrated. Figure 7 shows the results of H & E staining of liver tissue in ulcer mice. The liver tissue structure of the mice in normal group (Fig. 7A, a) was complete with clear layout, round and full hepatocytes, and obvious boundary between the liver sinusoid and liver plate. In the model group (Fig. 7B, b), the liver tissue structure of mice was still clear, however, the liver cells were loose, the liver sinusoids were narrowed, and there were significant inflammatory cell infiltration and inflammatory response. The liver tissue of the mice in the omeprazole group (Fig. 7C, c) were arranged neatly, and a small number of vacuoles could be seen, but no obvious inflammatory cell infiltration was observed. Compared with the model group, the liver tissue of mice in the EA group was intact, with no obvious liver damage, and the degree of inflammatory cell infiltration was improved in a dose-dependent manner. The low dose of EA had a significant protective effect on the liver. Effects of EA on nephridial histology As shown in Fig. 8, no significant histological changes were observed in the kidney of mice in the normal group, the omeprazole group and the groups administrated with different doses of EA. The kidneys of the model group had a larger glomerulus and wider lumen of the renal tubules, and the epithelial cells of renal tubulars were markedly damaged. Therefore, EA administered to mice was safe and no drug-related toxicity was detected even at the highest doses investigated.

Discussion
It has been reported that the natural secondary metabolites found in ethno-medicinal plants have antioxidants, especially polyphenols, which have a variety of functions to counteract the free radicals and regulate oxidative stress in cells [20]. And there is growing evidence that the use of medicinal plants is an alternative for the treatment of several gastrointestinal disorders, including gastric ulcers. The bio-active natural products with gastroprotective properties and a few side effects could substitute existing synthetic drugs. EA is a common polyphenol metabolite present in many medicinal plants, fruits and vegetables. EA has been reported to have antioxidant, anti-inflammatory, anti-proliferative, anti-diabetic and anti-viral properties. Therefore, it is a very promising agent for the treatment of ulcerative colitis, cancer and diabetes [21]. Notably, EA its antioxidant property is related to most of its pharmacological activities, including reducing the lipid metabolism and altering inflammatory mediators (IL-6, IL-1β, and TNF-α) [22].
In this study, we have isolated and identified ten compounds from Eucalyptus citriodora leaves and investigated the protective effects of EA extracted from Eucalyptus citriodora leaves on ethanol-induced acute gastric ulcer in mice. The mice administrated anhydrous ethanol (0.1 mL/10 g) had produced acute gastric ulcers and inflammatory responses, such as bleeding, edema, and erosion in the gastric tissues. However, the mice pretreated with various concentrations of EA significantly caused a reduction in the gastric injury, and the pathological changes of gastric ulcer were improved considerably, especially in the middle-dose group.
Inflammation is an important pathogenesis of gastric ulcers. The inflammatory cells will be affected to overreact and produce inflammatory factors (TNF-α, IL-6, and IL-10) when the gastrointestinal tract is stimulated by alcohol [23]. The pro-inflammatory cytokine TNF-α is associated with inflammation and injury of various tissues, and it will be secreted at the onset of gastric ulcer [24]. It is reported that increasing TNF-α will lead to severe damage to gastrointestinal mucosal by stimulating the neutrophil accumulation [25]. IL-6 is a multifunctional cytokine and is a direct immune response to inflammation caused by infection and tissue damage [26]. During gastric ulcer, elevated IL-6 level will stimulate lymphocytes and neutrophils which result in damages [27]. IL-10 is a counter-regulatory cytokine in inflammatory responses, and it can slow down the production of pro-inflammatory cytokines to damp down inflammation [28]. In this study, the TNF-α and IL-6 levels in the gastric tissues of the absolute ethanol group significantly increased, while the IL-10 level were considerably reduced. The increased levels of TNF-α and IL-6, and the decreased level of the IL-10 indicated injury and inflammation in the gastrointestinal tissues. Reversely, the TNF-α and IL-6 levels were significantly reduced, and the level of the IL-10 were expressively increased in the gastric tissues of the ellagic acid groups. These results are concordant with those from a study carried out by Beserra et al. [29] who reported that EA reduced the pro-inflammatory cytokine IL-6 and increased the anti-inflammatory cytokine IL-10 in the ethanol-initiated acute gastric ulcer. Furthermore, the mice pretreated with EA showed a significant reduction in the degree of hemorrhagic damage and infiltration of inflammatory cells, which manifested EA has a good anti-inflammatory effect and protective effect on ethanol-induced acute gastric ulcer in the mice, especially in the middle-dose (100 mg/kg) EA group. The anti-inflammation effect of EA was more significant at the middle dose; however, the high-dose EA (200 mg/kg) showed a more positive effect on reducing the IL-6 level and increasing the IL-10 level. The exact mechanism of this effect has not figure out yet. Therefore, we proposed various hypotheses to explain this phenomenon. Firstly, EA exhibited protective effects against free radical damage in gastric ulcers, which may account for its antioxidant activity [30]. EA has the antioxidant activity of scavenging free radicals and inhibiting lipid peroxidation, which depends on EA concentration. Secondly, the high-dose EA may stimulate physiological response mechanism which can lead to the changes in many physiological pathways, thus may have adverse effects on the anti-ulcer activity [31]. Thirdly, EA is known as a good chelator, and the chelating ability is positive correlated with its concentration. So that, high-dose EA can more chelate with calcium ion in the extracellular or cytoplasm to change net concentration of free calcium. Meanwhile, the calcium gradient is associated with many signaling cascades, it means that high-dose EA can more significantly stimulate these signaling cascades to affect the potency on gastric ulcer [32]. Gastrin (GAS) is a polypeptide hormone with an important role in stimulating gastric acid secretion, intensifying cellular metabolism and increase the proliferative activity of cells in the gastrointestinal tract [33], but excessive gastric acid secretion will lead to changes in gastric mucosal permeability and accelerate the formation of gastric ulcers [34]. In this research, the level of GAS significantly increased in the anhydrous ethanol group while GAS level decreased significantly in the middle-dose EA group, and it is shown that middledose EA has an ability on protecting gastric mucosa by decreasing GAS level. This may be a negative feedback mechanism and we speculate that the observed decrease in plasma circulating gastrin may help to explain the protective activity of EA in the stomach injured by ethanol.
COX-2 is an important inducing enzyme triggered by a number of cytokines and inflammatory mediators present in various inflammatory cells. In the inflammatory states, induced COX-2 is primarily responsible for the synthesis of prostaglandin E 2 (PGE 2 ) [35]. PGE 2 is a bio-active lipid and plays a critical role in guiding and governing various aspects of the inflammatory response. In the initial phase of the inflammatory response, PGE 2 participates in the pro-inflammatory process, but in the latter phases, PGE 2 exerts control over a number of mechanisms that lead to the resolution of inflammation [36]. In parallel, PGE 2 is established clearly the important contributions in the protection of gastric mucosa [37]. In this study, the COX-2 contents of the model group were significantly higher than those of the normal control group while the PGE 2 contents were significantly lower than those of the normal group. These data are consistent with a previous study conducted by Tu et al. [38] where absolute ethanol, administration stimulate the expressions of COX-2 and inhibit the secretion of PGE 2 . However, pretreated with EA in our study, the COX-2 level significantly decreased while the PGE 2 level significantly increased, and both were dose-dependent. Consistent with these results, H&E staining showed that absolute alcohol induced gastric mucosal damage and inflammatory cell infiltration, while the cells in stomach tissue are arranged neatly after pretreat with EA. These results confirmed that EA has a beneficial effect against ethanol-induced gastric injury and inflammation.
Drug toxicity is the most common cause of acute liver and kidney failure; thus, it is necessary to identify and monitor the toxicity signals during clinical practice. ALT and AST are found mainly in the liver. When the liver is damaged, ALT and AST are released into the blood, which increases serum ALT and AST concentration [39]. Therefore, the levels of ALT and AST in serum are predictive of the likelihood of drug-induced liver injury [40]. Creatinine is a cyclic anhydride of creatine and serum creatinine is the most reliable indicator of nephropathy [41][42][43]. In this study, the ALT, AST, and Cr activities in the low-dose and middle-dose of EA group were close to the normal control group and positive control omeprazole group. The results showed that the low-dose and middle-dose of EA has no effects on the hepatic and nephridial.

Conclusions
Eucalyptus citriodora leaves are the rich source of polyphenol and flavonoid compounds and exhibit strong antioxidant and anti-inflammatory properties, which have potential to develop the drugs of antioxidant, gastroprotective and anti-inflammatory. We have extracted and identified ten compounds from Eucalyptus citriodora leaves and investigated the effects on anti-ulcer and antiinflammatory against ethanol-induced acute gastric ulcer of ellagic acid that is a polyphenol compound of extraction. The present findings evidence that middle-dose (100 mg/kg) EA have significant effect on anti-gastric ulcer and antiinflammatory through increase the IL-10 and PGE 2 levels, and reduce the IL-6, TNF-α, GAS, and COX-2 levels. Furthermore, the ALT, AST, and Cr contents in the middledose EA have few differences from the normal group, which indicate that it has no adverse effect on liver and kidney. These results uncover that ellagic acid may be a good candidate for development as an anti-ulcer drug.

Plant material
Eucalyptus citriodora leaves was collected at the Lingnan Normal University, Guangdong, China. It was authenticated by one of the authors (Zhongliu Zhou) and a voucher specimen (EC 2020/9) was deposited at the Natural Medicinal Chemistry Lab of the university.

Animals
The animal experiments were conducted according to the rules of animal experiment and the protocol also followed the rules of the local Animal Ethics Committee (Ethic Permit TCMF1-2014025). Forty-eight male mice (20-25 g) were purchased from the Guangdong Medical Laboratory Animal Center (Guangdong, China). The mice were placed in suitable cages with inert husk materials as bedding and were maintained under controlled environment conditions of light and dark cycle (light 12 h, dark 12 h, temperature 23 ± 2°C and relative humidity 55 ± 10%) for a period of 1 week before starting the experiments. The animals were provided free access to water and standard chow pellets.

Experimental design
The mice were randomly assigned to six groups, containing normal group, model group, omeprazole group and high/ meddle/low dose of Ellagic acid. All groups successively received corresponding drugs once a day (0.1 mL/10 g) for 14 days by a gastric gavage. (Normal group and model group administrated normal saline).

Induction of acute gastric ulcer and sample collection
On the last day of giving drugs, 60 min after drug administration, the mice were not in normal group received absolute ethanol (0.2 mL/10 g) by a gastric gavage, while normal group received normal saline [44]. After 1 h, removed the eyeballs of the mice and the blood was collected under 4°C and centrifuged at 4000 rpm for 10 min to obtain serum preserving at −20°C for biochemical estimation. The mice were sacrificed by cervical dislocation, the stomach was removed, opened along the greater curvature and rinsed gently in normal saline. The stomach, kidney and liver tissues were collected and rinsed with normal saline and then fixed into 4% paraformaldehyde solution, processed for histological, immunohistochemical, and biochemical assessments.

Histopathological evaluation
After the stomach, kidney and liver tissues were fixed in 4% paraformaldehyde solution for 24 h, these tissues were dehydrated using a series of alcohol, cleared in xylene and then treated with paraffin imbedding. Tissues were cut into 5 µm thickness, stained with Hematoxylin-eosin and were performed to observe the histopathological changes of these samples under a light microscope.

Liver tissue levels in mice
The livers from the mice were processed into piece and added pre-cooled saline (5 mL/10 g) and then homogenate in the ice-bath and centrifuged under 4°C at 4000 rpm for 10 min to obtain serum. The levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in the liver tissue were determined according to the kit instructions (Shanghai MLBIO Biotechnology Co. Ltd.).

Serum levels in mice
The level of IL-6, IL-10, TNF-α, GAS, COX-2, PGE 2 and Cr in the serum were determined according to the kit instructions (Shanghai MLBIO Biotechnology Co. Ltd.).

Statistical analysis
The results from each group were calculated as mean ± standard error of mean (S.E.M.). They were analyzed by one-way analysis of variance, and the statistical evaluation between two groups was determined by LSD on SPSS 20.0. Probability (P) values < 0.05 (P < 0.05) were considered to be statistically significant.