Characterization of phenolic compounds from Eugenia supra-axillaris leaf extract using HPLC-PDA-MS/MS and its antioxidant, anti-inflammatory, antipyretic and pain killing activities in vivo

Reactive oxygen species (ROS) are involved in the pathophysiology of several health disorders, among others inflammation. Polyphenols may modulate ROS related disorders. In this work, thirty-two phenolic compounds were tentatively identified in a leaf extract from Eugenia supra-axillaris Spring. ex Mart. using HPLC-MS/MS, five of which were also individually isolated and identified. The extract displayed a substantial in vitro antioxidant potential and was capable of decreasing ROS production and hsp-16.2 expression under oxidative stress conditions in vivo in the Caenorhabditis elegans model. Also, the extract showed higher inhibitory selectivity towards COX-2 than COX-1 in vitro with higher selectivity towards COX-2 than that of diclofenac. The extract also exhibited anti-inflammatory properties: It attenuated the edema thickness in a dose dependent fashion in carrageenan-induced hind-paw odema in rats. In addition, the extract reduced the carrageenan-induced leukocyte migration into the peritoneal cavity at the highest dose. Furthermore, the extract showed antipyretic and analgesic activities in a mouse model. Eugenia supra-axillaris appears to be a promising candidate in treating inflammation, pain and related oxidative stress diseases.

extraction and fractionation. Fresh leaves (1.5 kg) from E. supra-axillaris were homogenized in a MeOH-H 2 O (3:1) mixture (3 × 4 L). The obtained extract was filtered and dried under reduced pressure to give a yield of 120 g dried extract. The obtained extract was re-dissolved in water (500 mL) and defatted with n-hexane (3 × 500 mL). The obtained defatted extract was collected, dried under reduced pressure and frozen at -70 °C and then lyophilized yielding 90 g of a fine dried powder. The defatted extract was also investigated for its phenolic content by two-dimensional paper chromatography (TDPC) using n-butanol: acetic acid: water, 4:1:5 upper layer (BAW) as the first developing solvent and 6% acetic acid as the second solvent.

HpLc-MS/MS. A Thermofinnigan (Thermo Electron Corporation, USA) coupled with an LCQ-Duo ion
trap mass spectrometer with an ESI source (ThermoQuest) system was utilized. A C18 reversed-phase column (Zorbax Eclipse XDB-C18, rapid resolution, 4.6 × 150 mm, 3.5 µm, Agilent, USA) was used to separate the analytes. A gradient of water and acetonitrile (ACN) (0.1% formic acid each) was applied from 5% to 30% ACN over 60 min with flow rate of 1 mL/min with a 1:1 split before the ESI source. The samples were injected automatically using autosampler surveyor ThermoQuest. The instrument was controlled by Xcalibur software (Xcalibur TM 2.0.7, Thermo Scientific). The MS operated in the negative mode as before 8 . The ions were detected in a full scan mode and mass range of 50-2000. Isolation and identification of the phenolic metabolites. The defatted aqueous methanol extract of the leaves (20 gm) was dissolved in water and applied to a polyamide 6 s column and eluted with H 2 O followed by H 2 O-MeOH mixtures of decreasing polarities. The collected fractions were investigated by TDPC using BAW as the first solvent and 6% acetic acid as the second solvent. Compound 1 ( Table 1) was isolated from fraction I by crystallisation. Compounds 7 and 9 (Table 1) were isolated from fraction II by preparative paper chromatography using BAW as an eluent. Similarly, compounds 16 and 23 (Table 1) were separated from fraction III and IV, respectively. The isolates were identified according to their UV spectral data, 1 H-and 13 C-NMR assignments.
In vitro antioxidant evaluation. Determination of total phenolic contents, DPPH radical scavenging activity and FRAP ferric reducing antioxidant power assays were done as described 9 . The total phenolic content was estimated using Folin-Ciocalteu method. In brief, 20 µL solution of the plant extract was mixed with 100 µL Folin-Ciocalteu reagent. At ambient temperature, the mixture was incubated for 5 min and then 80 µL of 7.5% Na 2 CO 3 solution was added. The mixture was then left for 30 min in the dark and the absorbance was monitored using a microplate reader (Biochrom Asys UVM 340) at 735 nm. The results were presented as gallic acid equivalent (GAE) in mg/g of extract. Total antioxidant capacity (TAC) assay was determined using a commercially available TAC ELISA kit (MBS726896, my biosource, Inc., San Diego, CA, USA) according to the manufacturer's instructions with ascorbic acid as the reference standard to estimate TAC as described 10 . In vivo antioxidant activities (Caenorhabditis elegans experiments). The worms were maintained under standard conditions as described before 11 . In brief, age synchronized cultures were obtained by sodium hypochlorite treatment of gravid adults; the eggs were kept in M9 buffer for hatching. The obtained larvae were then transferred to S-media containing E. coli OP50 (D.O 600 = 1.0) 12 . We here used Wild type (N2), TJ375 [Phsp-16.2: GFP(gpls1)] and TJ356 strains to perform the experiments as described before 11 . The worms were provided by the Caenorhabditis Genetic Centre (CGC), University of Minnesota, U.S.A. In brief, Age synchronized worms (TJ356, L1 stage, grown in S-medium) were used to quantify the subcellular DAF-16: GFP localization. The worms were treated with three different doses (50, 100 and 200 µg/mL) at 20 °C for 24 h in S-medium. Fluorescence microscopy was used to collect the images. The worms were classified to cytosolic, intermediate, and nuclear based on the localization of the fusion DAF-16::GFP.
pression potential was analysed according to Abdelall et al. 13 using a lipoxygenase inhibitor screening assay kit (Cayman Chemical, AnnArbor, MI, USA). Cyclooxygenases (COX-1 and COX-2) were measured by an enzyme immunoassay (EIA) kit (Cayman Chemical, AnnArbor, MI, USA) according to the manufacturer's instruction. COX-2 selectivity index (SI values) which is defined as IC 50 (COX-1)/IC 50 (COX-2) was estimated and compared to the reference compounds celecoxib, indomethacin and diclofenac as previously described 10 .
In vivo anti-inflammatory experiments. Adult male Wistar rats (140-160 g) and Swiss albino mice (20-25 g) obtained from the Faculty of Veterinary Medicine, Zagazig University, Egypt were used in this study. They were housed under constant experimental conditions of humidity (55%), temperature (23 °C), and 12 h light/dark cycle. Animals were left for one week for acclimatization before performing the experiments and were supplied with water and a commercially available regular chow diet ad libitum. The study protocol was approved by the Ethical Committee of the Faculty of Pharmacy, Zagazig University for Animal Use (P 9-12-2017) and performed according to the guidelines of the US National Institutes of Health on animal care and use.
Carrageenan-induced hind-paw odema. The vehicle (10 mL/kg), E. supra-axillaris aqueous methanol extract (200 mg/kg and 400 mg/kg, p.o.) or diclofenac (10 mg/kg) were orally gavaged to rats (n = 6/group) 1 h before the induction of inflammation. Afterwards, hind paw odema was induced in the right leg of the rats by injecting freshly prepared carrageenan solution (1% in 0.9% NaCl, 0.1 mL), into the sub plantar tissue. The thickness (mm) of the right leg hind paw was measured in the dorsal-plantar axis by a calliper ruler before and after the carrageenan injection at hourly intervals for 6 h and then at 24 h. The cumulative anti-inflammatory effect during the whole experiment period (0-24 h) was evaluated by detecting the area under the curve of changes in paw thickness-time curve (AUC 0-24 ). www.nature.com/scientificreports www.nature.com/scientificreports/ Migration of leukocyte to peritoneal cavity in mice. This test was carried out according to the method of Silva-Comar et al. 14 . Swiss albino mice (n = 5-8/group), was assigned into four groups and orally given the vehicle (1 mL/100 g, p.o.) or extract (200 mg/kg and 400 mg/kg, p.o.). Thirty min later, the animals were injected intraperitoneally with 0.1 mL carrageenan solution (500 μg/mice) or 0.1 mL sterile saline. Diclofenac (10 mg/kg, p.o.) was used as an anti-inflammatory reference compound. After 3 h, the animals were euthanized, and the peritoneal cavity was washed with 3 mL of phosphate-buffer saline (PBS) containing 1 mM ethylenediamine tetra-acetic acid (EDTA). The number of leukocytes in the peritoneal cavity wash was determined using a haemocytometer and expressed as number of cells/mL.

Analgesic activity experiments.
Peripheral anti-nociceptive activity experiment. Acetic acid writhing test in mice was used to evaluate the peripheral analgesic activity of the extract according to the method of Nakamura et al. 15 . Briefly, mice were assigned into 4 groups (5-7 mice). Group (1) was pre-treated with the vehicle (1% Tween 80, 10 mL/kg) and served as negative control. Groups (2) and (3) pre-treated with the extract (200 and 400 mg/kg, p.o. respectively) and group (4) pre-treated with diclofenac (10 mg/kg, a reference drug) 1 h prior i.p. injection of 0.7% acetic acid (1 mL/100 g). The total number of writhes was recorded for 25 min.
Central anti-nociceptive activity (Hot plate test). The possible central analgesic activity of the extract was tested using hot plate method 16,17 . Four groups of mice (6 each) were used in the present experiment and received either the extract (100 or 200 mg/kg, p.o.) or the vehicle (10 mL/kg, p.o.). The opioid analgesic nalbuphine was given to another group of mice as reference central analgesic. Each mouse was put on a hot plate heated at 55 ± 1 °C, 60 min after giving the drugs or vehicle. Mice were observed for their response to heat of the hot plate (lapping of the fore and hind paws, hind paw rising or jumping). The required time for the animals to show the first response to the heat was recorded, at baseline and at 1, 2, 3, and 4 h following the administration of vehicle or drugs.
Acetic acid-induced mouse vascular permeability. The acetic acid-induced mouse vascular permeability was carried out as previously described 18 . Mice were assigned into four groups (6 mice/each group). They were pre-treated with the extract in two dose levels (200 mg/kg, and 400 mg/kg, p.o.), diclofenac (20 mg/kg) or vehicle. After 1 h of pretreatment, 0.2 mL Evans blue (0.25% solution in normal saline) was injected in the tail vein. Thirty min later, acetic acid (0.6% in normal saline, 1 mL/100 g) was injected in the peritoneal cavity of mice. Another group of mice was injected with normal saline only and served as normal control. After another 30 min, the mice were euthanized by cervical dislocation. The peritoneal cavity of each mouse was washed with 3 mL saline and the washing was then centrifuged at 3000 g for 10 min. Evans blue dye content of the supernatant which is proportional to the vascular permeability was detected at 610 nm using a plate reader (Biotech, Vt, USA).

Induction of pyrexia in mice.
In this experiment 4 groups of mice were used (n = 6). Pyrexia was induced in mice as described previously with modifications 19,20 . The initial body temperature of the rectum was recorded for each mouse using a lubricated digital thermometer. Next, Brewer's yeast suspension was prepared in normal saline (30%) and then mice were injected with yeast suspension (1 mL/100 g) subcutaneously behind the neck. 18 h later rectal temperature was recorded again (T 0 ) and mice showed a higher record by at least 0.5 °C were included in the study. Pyretic mice were then given the extract (200 mg/kg, and 400 mg/kg, p.o.), paracetamol (150 mg/kg) or vehicle. The rectal temperature was reordered again at 30 min, 1, 2, 3 and 24 h post-treatment.
Statistical analysis. The data of the present study are expressed as mean ± SEM. Results of the experiments were statistically analysed using GraphPad Prism software, version 5.00 (GraphPad Software, Inc. La Jolla, CA, USA). The statistical difference among different groups was evaluated using Analysis of Variance (ANOVA) or repeated-measures analysis of variance (RM-ANOVA) followed by Tukey's post hoc test and Student's t-test. The considered level of significance was p < 0.05.

Phenolic profile of E. supra-axillaris extract by LC-MS/MS. Profiling of the polyphenolic metabolites
in E. supra-axillaris leaf extract was performed by HPLC-ESI-MS/MS. In total, thirty-two secondary metabolites were characterized and tentatively identified based on their retention times, molecular weight, and fragmentation pattern as well as comparison with reported data (Fig. 1 Fig. 2). Compound 14 exhibited a [M-H] − at m/z 399 and a main daughter ion at m/z 275; it was tentatively characterized as monolactonoside didecarboxylated valoneoic acid (Fig. 3).
Antioxidant activities. The extract showed substantial antioxidant activities in vitro in DPPH and FRAP assays when compared to positive controls ascorbic acid and quercetin, respectively ( Table 2). The total phenolic content amounted 335 mg GAE/g extract according to the Folin-Ciocalteu method. Additionally, the total antioxidant capacity (TAC) assay of the extract was 1.5-fold higher than that of the solid antioxidant compound, ascorbic acid (Table 2).  www.nature.com/scientificreports www.nature.com/scientificreports/ To examine the protective effects of the extract in vivo, we monitored the survival rate of the wild type C. elegans worms N2 under a lethal dose of 80 µM of the pro-oxidant juglone. Worms which were pretreated with the extract had enhanced survival rates when compared to juglone group, which was treated with juglone alone. Epigallocatechin gallate (EGCG) was used as positive control (Fig. 4a). We also investigated the influence of the extract on the intracellular accumulation of ROS. E. supra-axillaris extract was able to decrease the levels of ROS in vivo in a dose dependent manner (Fig. 4b).
To further investigate the antioxidant activity in vivo of the extract, we used a sensitive oxidative stress sensor hsp-16.2. Under oxidative or heat stress, hsp-16.2 acts as chaperon to recognize and degrade unfolded proteins. After exposing the nematodes to 20 μM juglone for 24 h, hsp-16.2 is highly expressed. The extract was able to reduce hsp-16.2 expression in a concentration-dependant manner (Fig. 4c). Thus, the extract can alleviate oxidative stress in worms and its compounds show bioavailability.  www.nature.com/scientificreports www.nature.com/scientificreports/ DAF-16 (a member of the FOXO transcription factor group) is crucial in several signaling pathways that regulate the stress response, age-related diseases as well as other important biological processes. To get an insight in the molecular mode of action of the tested extract, the transgenic C. elegans strain TJ356 was used to investigate the influence of the extract on the subcellular localization of DAF-16, which is in the cytosol when inactive. The extract caused a translocation of FOXO transcription factor DAF-16 from the cytoplasm to the nucleus indicating that the in vivo antioxidant capacity of the extract may involve the DAF-16/FOXO regulated signalling pathway (Fig. 4d).

Anti-inflammatory effects. In vitro anti-inflammatory activities. Cyclooxygenase (COX) is a key enzyme
in the biosynthesis of prostaglandin from arachidonic acid. COX has two main isoforms: COX-1 and COX-2. Inflammation mediated pathologies are related to COX-2 over-expression. The E. supra-axillaris extract suppressed both COX-1 and COX-2 in vitro with greater selectivity towards COX-2 than COX-1. Interestingly, the extract has a much more COX-2 selectivity than that of diclofenac (Table 3). Moreover, the extract has nearly double the potency of zileuton, the reference LOX inhibitor to inhibit lipoxygenase enzyme in vitro (Table 3).
Inflammation: Effects of the extract on carrageenan-induced paw edema in rats. Animals, injected with 0.1 mL carrageenan (1% in 0.9%, sub-planter), showed increased paw thickness when measured each hour for 5 h and at 24 h after injection. The highest increase was observed 4 h post injection to reach 3.9 ± 0. 28 mm over base line readings (paw thickness measured before carrageenan injection). Rats pretreated 1 h earlier with the extract (200 and 400 mg/kg, p.o.) showed a significant decrease in edema thickness in a dose-dependent fashion with more potent activities than the standard anti-inflammatory drug, diclofenac (Fig. 5).
Effects of E. supra-axillaris extract on mouse carrageenan-induced leukocyte migration. Pretreatment with the extract attenuated carrageenan (500 μg/cavity, i.p., 0.1 mL) induced leukocyte recruitment into the peritoneal cavity in mice. Noteworthy, the response to the extract was more potent than that obtained in animals treated with diclofenac, the reference drug, 1 h prior to carrageenan challenge (Fig. 6). www.nature.com/scientificreports www.nature.com/scientificreports/  Table 3. The in vitro effects of E. supra-axillaris extract on COX-1, COX-2 and 5-LOX. SI is COX selectivity index which is defined as IC 50 (COX-1)/IC 50 (COX-2).  www.nature.com/scientificreports www.nature.com/scientificreports/ Effects of Eugenia supra-auxillaris on acetic acid-induced vascular permeability in mice. As shown in Fig. 6, acetic acid injection in mice (0.6%, ip) increased vascular permeability represented as significantly (P < 0.001) higher Evans blue absorbance in the peritoneal cavity exudate compared to saline injected mice (0.48 ± 0.08 vs 0.07 ± 0.003). This effect was attenuated in mice pretreated with E. supra-auxillaris extract (200 and 400 mg/kg, p.o) 1h r prior acetic acid injection by 73 and 80%, respectively. Additionally, the reference standard, diclofenac sodium achieved 63% lower reading compared to control mice (Fig. 6).

Pain-killing properties: Effects of the extract on acetic acid-induced writhing response in mice.
Oral pretreatment with the extract revealed a weak potency at the lower dose (200 mg/kg, p.o) but completely abolished acetic acid (0.7% acetic acid, 1 mL/100 g) induced writhes in mice when used at a higher dose (400 mg/kg) to achieve zero writhes in the overall 30 min observation period. Additionally, mice pretreated with diclofenac (10 mg/kg, p.o.), the reference standard, showed 67% reduction of the control writhes (Fig. 7).
Analgesic properties: Effects of the extract on hot plate test in mice. Mice pretreated with the extract (200 mg/kg, p.o.) and the reference standard nalbuphine (10 mg/kg, i.p.), showed a longer response latency when measured at 1, 2, 3 and 4 h after administration to reach its peak at the 2 h time point (2.8 and 3.8 fold of control., respectively). While the effect was significant (p < 0.001) at all-time points for nalbuphine, the effect of extract was only significant 2, 3 and 4 h post treatment (Fig. 8).
Antipyretic activities: Effect of the extract on Brewer's yeast induced pyrexia in mice. Brewer's yeast injection in mice raised rectal body temperature of injected mice when measured 18 h after injection. Mice pretreated with E. supra-axillaris extract (200 and 400 mg/kg, p.o.) showed significantly lower rectal temperatures starting from 1 h post treatment (p < 0.001). This antipyretic effect was similar in the two studied doses and comparable to the effect obtained in mice treated with paracetamol (150 mg/kg), the antipyretic standard drug (Table 4).

Discussion
In the current study, a leaf extract from E. supra-axillaris exhibited a plethora of pharmacological activities. It scavenged the free radicals in vitro in DPPH assay, reduced FeSO 4 in FRAP assay and exhibited total antioxidant capacity (TAC) which was 1.5-fold higher than that of the gold antioxidant standard, ascorbic acid.
It also significantly increased the survival rate against the deleterious effects of juglone, reduced the intracellular ROS and juglone induced Phsp-16.2:GFP levels in a dose dependant manner and induced nuclear translocation of DAF-16::GFP in C. elegans. Similar effects have been reported for other plant extracts rich in polyphenols 11 .
Carrageenan induced edema is a common experimental model used to assess the anti-inflammatory effects of natural products. The extract exerted potent inflammatory effect in a dose-dependent fashion with more potent activities than the standard anti-inflammatory drug, diclofenac which may be attributed to the observed COX-2 inhibition. Similar findings had been reported for Eugenia brasiliensis 23 .
We further investigated the anti-inflammatory effect of the extract on carrageenan-induced peritonitis model. In this model, carrageenan, causes the leukocytes recruitment to the peritoneal cavity. Leukocyte recruitment to the inflammation site plays an important role in the development of an inflammatory process. It is a highly regulated process and represents a potential therapeutic target 28 . This study showed that the extract was more potent than diclofenac in reducing the number of inflammatory cells at the inflammation site. Similar findings had been reported for Eugenia jambolana 24 .
The released inflammatory mediators during different phases of inflammation stimulate both peripheral and central nociceptive pathways resulting in pain sensation. Tissue injury leads to activation of both COX and LOX and subsequent formation of prostanoids, cytotoxins as well as leukotrienes that will act in both the development of the inflammatory process and the hypernociceptive signal 29 . Surprisingly, the higher dose of the extract completely abolished acetic acid induced writhes in mice to achieve zero writhes in the overall 30 min observation period while mice pretreated with diclofenac, the reference standard, showed 67% reduction of the control writhes. These results may be attributed to the inhibition of COX and the inhibition of the generation of pain mediators mainly, prostaglandins. Similar findings had been reported for Eugenia jambolana 24 .
The extract not only exerted a peripheral anti-nociceptive effect but also exerted a central anti-nociceptive effect. The hot-plate test is used in the present study to evaluate the central anti-nociceptive effects. The extract showed longer latency in the response of mice to hot plate, which was significant 2, 3 and 4 h post treatment. However, it is less potent than the reference standard nalbuphine.
Brewer's yeast induced pathogenic fever by increasing the synthesis of prostaglandin 30 , in the hypothalamus and is utilized in this study to investigate antipyretic effect of the extract. The extract in both studied dose levels showed significantly lower rectal temperatures starting from 1 h post treatment and their effects is like that of paracetamol. Therefore, it can be postulated that the extract may interfere with the release of prostaglandin (PGE2) and pyrogenic cytokines 31 . This study showed that the extract inhibits both COX-1 and COX-2 enzymes which lead to inhibition of synthesis of prostaglandins (PGE2). The antipyretic activity of the extract may be attributed to the phenolic components of the extract. Noteworthy, the current results are in a good agreement with those reported from other Eugenia species, namely E. jambolana, E. brasiliensis, E. uniflora and other Myrtaceae species Syzygium aqueum, S. jambos and S. samarangense 10,23-27,32 . conclusions Profiling of the phenolic secondary metabolites of E. supra-axillaris leaves was performed utilizing HPLC-MS/MS and led to the dereplication of thirty-two compounds, from which seventeen were reported for the first time in this species. Moreover, five phenolics were individually isolated and identified. The extract exhibited substantial antioxidant, anti-inflammatory, antipyretic and analgesic activities in vivo. In conclusion, the witnessed in vitro and in vivo activities suggest that E. supra-axillaris could exert protective properties against oxidative and free radical damage associated with diverse pathological disorders. Further studies are needed, to explore the corresponding mechanisms.  Table 4. Effect of E. supra-axillaris extract on Brewer's yeast induced pyrexia in mice. Each value represents the mean ± S.E.M (n = 5), *p < 0.001 vs. control valuesusing Analysis of Variance (ANOVA) followed by Tukey post hoc test.