Properties of Flavonoids Isolated from the Bark of Eysenhardtia polystachya and Their Effect on Oxidative Stress in Streptozotocin-Induced Diabetes Mellitus in Mice

Six new flavonoids 2′,4′-dihydroxychalcone-6′-O-β-d-glucopyranoside (1), α,3,2′,4′-tetrahydroxy-4-methoxy-dihydrochalcone-3′-C-β-glucopyranosy-6′-O-β-d-glucopyranoside (2), 7-hydroxy-5,8′-dimethoxy-6′α-l-rhamnopyranosyl-8-(3-phenyl-trans-acryloyl)-1-benzopyran-2-one (3), 6′7-dihydroxy-5,8-dimethoxy-8(3-phenyl-trans-acryloyl)-1-benzopyran-2-one (4), 9-hydroxy-3,8-dimethoxy-4-prenylpterocarpan (5), and α,4,4′-trihydroxydihydrochalcone-2′-O-β-d-glucopyranoside (6) were isolated from bark of Eysenhardtia polystachya. Antidiabetic activity of compounds 1–5 in terms of their cellular antioxidant and free radical scavenging and also in streptozotocin- (STZ-) induced diabetic mice was evaluated on liver transaminases, lipid peroxidation, total bilirubin, total protein, superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase (CSH-Px), and glutathione reductase (GSH). Results indicated that 1–5 scavenged 2,2-diphenyl-1-picrylhydrazyl (DPPH), hydroxyl (∙OH), nitric oxide radicals (NO∙), superoxide anion radical (O2 ∙−), radical cation (ABTS∙+), and hydrogen peroxide (H2O2) radical, and protection against H2O2 induced BSA damage was also observed. Furthermore, 1–5 showed ability to decrease the oxidative stress in H9c2 cell. Diabetic mice present high levels of lipid peroxide, total protein, SGPT, SGOT, ALP, and TB. However, treatment of STZ-induced diabetes in mice with 1–5 reduced levels of these enzymes leading to protector effect of liver. In addition, with treatment with 1–5, increases in radical scavenging enzymes of CSH-Px, SOD, GSH, and CAT have also been observed in diabetic mice. The antioxidant properties of compounds 1–5 are a promising strategy for ameliorating therapeutic effects by avoiding disorders in the normal redox reactions in healthy cells which consequently could alleviate complications of diabetes.


Introduction
Oxidative stress is a consequence of the increased reactive oxygen species (ROS) production and/or a decrease in their elimination. ROS such as hydrogen peroxide (H 2 O 2 ), superoxide anion (O 2 •− ), hydroxyl radical (OH), nitrogen oxide (NO), and lipid peroxides are formed in aerobic metabolism as normal products but also are produced under pathophysiological conditions in elevated rates. ROS can be responsible for the attack to biological macromolecules such as nucleic acids, proteins, membrane lipids, and carbohydrates, causing damage in the cell, which has been implicated in cardiovascular diseases, cancer, neurodegenerative disorders, and diabetes [1]. Natural antioxidants present in the majority of plants help in preventing mutagenesis, aging, and carcinogenesis and reduce oxidative damage due to their radical scavenging effect [2].
The tree Eysenhardtia polystachya (Ortega) Sarg, belonging to the Leguminosae family, is known as "palo azul" and has widely been used as antirheumatic, for the treatment of 2 Oxidative Medicine and Cellular Longevity nephrolithiasis and bladder disorders developed in diabetes [3]. Phytochemical studies indicate that E. polystachya contains polyphenols [4]. In another study, isoflavones displayed moderate cytotoxic activity against KB cell lines [5]. In previous studies, methanol-water extract from the bark of E. polystachya was evaluated in in vitro assays, showing antioxidant potential, hypoglycemic, and AGEs inhibition capacity [6]. In this study, we investigated the antioxidant properties and protective activity of 5 flavonoids isolates from the bark of E. polystachya in in vitro and in vivo assays.

General Experimental
Procedures. IR spectra were determined on a Perkin-Elmer 1720 FTIR. l H-l H, l3 C, DEPT, ID, DQF-COSY, TOCSY, and HMBC experiments were recorded on a Bruker DRX-300 NMR spectrometer, operating at 599. 19 MHz for 1 H and 150.86 MHz for l3 C, using the UXNMR software package; chemical shifts are expressed in (ppm) using TMS as an internal standard [7]. HREIMS were recorded on a JEOL HX 110 mass spectrometer (JEOL, Tokyo, Japan). TLC and column chromatography were carried out using precoated TLC silica gel 60 F254 aluminium sheets from Sigma-Aldrich (St. Louis, USA) and silica gel 60 (230-400 mesh, Merck Co., New Jersey, USA); solvents used as eluents were purchased from Fermont (California, USA). DCFH-DA were purchased from Degussa (Cincinnati, USA). All other reagents were obtained from Sigma-Aldrich (St. Louis, USA).

Antioxidant Activity In Vitro
2.4.1. Scavenging Effects on DPPH Radicals. 1 mL solution of 2,2-diphenyl-1-picrylhydrazyl (DPPH, 0.2 mmol/L) in methanol was added to 4 mL of sample. The mixture was shaken vigorously and was allowed to stand at room temperature for 30 min. Absorbance of the resulting solution was measured at 517 nm using a spectrophotometer (UV-1800, Shimadzu, Kyoto, Japan) [8].

Chelating Activity on Metal
Ions. 40 mol/L FeCl 2 was reacted with flavonoids for 5 min. Then ferrozine (200 mol/ L) was added and the mixture was left to stand for another 10 min. The absorbance at 562 nm was determined spectrophotometrically (UV-1800, Shimadzu, Kyoto, Japan) [8].

Trolox Equivalent Antioxidant Capacity (TEAC) Assay.
2 mmol/L H 2 O 2 alone in 30 mmol/L acetate buffer with pH 3.6 was incubated for 30 min at room temperature producing 10 mmol/L of ABTS radical cation (ABTS •+ ). 80 L of this solution was added to 0.4 mol/L acetate buffer (pH 5.8, 800 L) and to 20 L of the sample solutions; after exactly 5 min the absorbance was read at 734 nm. A dose-response curve was plotted for Trolox and antioxidant ability was expressed as TEAC value [9].

Scavenging Effects on Nitrite Oxide (NO).
In the test tubes to 0.1 mL of sodium nitroprusside solution (25 mmol/L) was added 150 g/mL of the sample; then the tubes were incubated for 150 min at room temperature. Griess reagent (0.3 mL of 1 g/L naphthylethyl-enediamine dihydrochloride and 0.3 mL of 10 g/L sulfanilamide in 5% H 3 PO 4 ) was added to the incubation solution. The absorbance was immediately read at 570 nm and referred to the absorbance of standard solutions of sodium nitrite salt treated in the same way with Griess reagent [10].

Effects on the Oxidation of BSA Induced by H
0.2 mL of samples was incubated with 1 mL BSA (40 mg/mL), 0.4 mL H 2 O 2 (20 mmol/L), and 0.4 mL of 20 mmol/L phosphate buffer (pH 7.4), for 2 h at 37 ∘ C. 5,5-Dithio-bis(2-nitrobenzoic acid) (DTNB, 1 mL, 2 mmol/L) was then added and the mixture was left to stand for another 30 min. The absorbance was measured at 410 nm. The free thiol concentration of samples was calculated based on the standard curve prepared by using various concentrations of l-cysteine.

Superoxide Radical Scavenging Activity (O 2
•− ). The reaction mixture is carried out in final volume of 3 mL containing 6 mM EDTA, 20 M riboflavin, 58 mM phosphate buffer at pH 7.6, and 50 M of nitroblue tetrazolium chloride (NBT).
Oxidative Medicine and Cellular Longevity 3 The reaction mixture is exposed for 15 min to 40 V under fluorescence lamp to initiate the reaction. The absorbance was determined at 560 nm [11].

Cellular Oxidative Stress Inhibition.
The intracellular accumulation of ROS in the H9c2 rat cardiomyoblast cell line was determined with carboxy-2 ,7 -dichloro-dihydrofluorescein diacetate (DCFHDA) method [14]. Cells H9c2 rat heart-derived embryonic myocytes are obtained from American Type Culture Collection, Manassas, VA, USA (CRL-1446), which was cultured with DMEM/F12 supplemented with 100 U/mL penicillin G, 10% (v/v) foetal bovine serum, 2 mM/L-glutamine, and 100 mg/mL streptomycin. Cells were incubated using 5% CO 2 and 95% air at 37 ∘ C, after the cells were plated in 35-mm culture dishes at 5.0 × 10 −4 cells/cm 2 . Then, the medium is changed with a new one. Cells seeded on 96-well plates were incubated with DCFHDA probe for 40 min. At the end of this period, medium was removed and cells were exposed to the flavonoids under investigation at a concentration of 10, 50, and 100 g/mL. After incubating exposed cells at 37 ∘ C for 24 h, fluorescence was measured at 488 nm (excitation) and 535 nm (emission) wavelengths on a microplate reader (Molecular Devices Spectra MAX Gemini X).

Evaluation of Oxidative Stress Markers In Vivo
2.13.1. Animals Care Conditions. The study was conducted in male CD1 mice, weighing about 25-30 g. Before and during the experiment, animals were fed a standard laboratory diet (Mouse Chow 5015, Purina) with free access to water. Mice were procured from the bioterium of ENCB and were housed in microloan boxes in a controlled environment (temperature 25 ± 2 ∘ C). Animals were acclimatized for a period of three days in their new environment before the initiation of experiment. Litter in cages was renewed three times a week to ensure hygiene and maximum comfort for animals. The experiments reported in this study were following the guidelines stated in Principles of Laboratory Animal Care (NIH publication 85-23, revised 1985) and the Mexican Official Normativity (NOM-062-Z00-1999). All animals procedures were performed in accordance with the recommendations for the care and use of laboratory animals (756/lab/ENCB).
2.14. Induction of Mild Diabetes (MD). Mild diabetes type 2 was induced in overnight fasted mice by a single intraperitoneal injection of 45 mg/kg streptozotocin (STZ) dissolved in 0.1 mol/L cold citrate buffer (pH 4.5), 15 min after the intraperitoneal administration of 120 mg/kg nicotinamide. The STZ treated animals were allowed to drink 5% glucose solution overnight to overcome drug induced hypoglycemia. After 7 days of development of diabetes, mice with moderate diabetes having persistent hyperglycaemia with more than 200 mg/dL were used for further experimentation [15].

Experimental Design.
Eight groups ( = 10) of diabetic mice were used to determine the chronic effect of 1-5. Each group was submitted to a specific treatment, as follows. Normal control and mild diabetic rats groups were fed with normal diet and drinking water ad libitum and were given saline by gastric gavage Mice with mild diabetes daily treated with the compounds isolated by an oral gavage at doses of 20 mg/kg of weight. The flavonoids were dissolved in distilled water and administered orally as a daily dose for four weeks. Mice were fasted overnight and sacrificed by cervical dislocation. Blood samples were collected in EDTA coated tubes. Livers, kidney, and pancreas were removed, washed with cold saline, and stored at −80 ∘ C.

Antioxidant Parameters Levels in Serum
, Liver, Pancreas, and Kidney. Activities of superoxide dismutase (SOD), serum catalase (CAT), glutathione peroxidase (CSH-Px), and glutathione reductase (GSH) were measured by assay kits purchased from Cayman Chemical (Michigan, USA), and the procedures were according to the kits instructions. The protein concentration was determined by the Bradford method [16] as described in the Bio-Rad protein assay kit. Lipid peroxidation (LPO) is used as an indicator of the oxidative stress in tissues and was estimated by the method of Fraga et al. [17], expressed as M/g of liver and kidney tissue. Malondialdehyde (MDA) as thiobarbituric acid reactive substance was measured at 532 nm spectrophotometrically [18]. Serum glutamate oxaloacetate transaminase (SGOT), glutamate pyruvate transaminase (SGPT), serum alkaline phosphatase (SALP), total bilirubin (TB), and total protein were determined by using a commercial Diagnostic Kit Biocompare, BioVision, Biocompare, and Thermo Scientific, respectively.
Compound 2 was also isolated as dark yellow amorphous powder. The aliphatic hydroxyl group of compound 3 has position similarly to 2. The singlet at H 3.85 was assigned to an aromatic methoxyl group which showed long range correlation with C-4 ( C 145.05) in the HMBC spectrum indicated to be linked at position C-4, whereas the doublets at H 4.85 and H 5.01 were attributed to two -glucopyranose moieties on the basis of the coupling constant (7.5 and 7.3 Hz). In the HMBC spectrum of 2, the long range correlations of anomeric proton H-1 ( H 4.85) with C-2 ( c 164.13), C-3 ( c 107.88), C-4 ( c 165.34), C-2 ( c 74.73), C-3 ( c 78.85), and C-5 ( c 77.32) were observed. These signals suggested that glucose unit should be directly linked at position C-3 . Also the anomeric proton H-1 ( H 5.01) correlation with C-6 ( c 164.04), C-2 ( C 78.65), and C-5 ( c 78.01) indicated that the other glucose unit should be linked to C-6 through a glycosidic bond (Table 1). 1 H NMR spectrum showed the typical pattern of a 2 ,3 ,4 ,6 -tetrasubstituted and 3,4-disubstituted chalcone with a singlet for one proton at H 6.68 could be assigned to the C-5 indicating that this position is free. Analysis of the HMBC and COSY spectra of 3 allowed for the complete assignment of all protons and carbons ( Figure 2). Consequently, the structure of compound 3 was elucidated as ,3,2 ,4 -tetrahydroxy-4-methoxy-dihydrochalcone-3 -C--glucopyranosy-6 -O--d-glucopyranoside ( Figure 1).    Table 1.
Compound 5 was isolated as yellow amorphous powder. Its molecular formula was assigned as C 22 H 24 O 5 on the basis of HRESIMS and 13 C NMR data. In IR spectrum were observed absorption bands for hydroxyl groups at 3389 cm −1 and aromatic groups at 1619, 1547, 1483, and 1429 cm −1 . Its molecular formula was established by HRESIMS and 13 C NMR data as C 22 H 24 O 5 . 13 C NMR and DEPT spectrum indicated that it contains 22 carbons, including four methyls, two methylenes, seven methines, and nine quaternary carbons. A pterocarpan structure was indicated to 1 H MR spectrum due to the splitting pattern of the protons at  cis (6.6 Hz) and trans (13.4) [21]. Key correlations in 5 are shown in Figure 2. Therefore, the structure of compound 6 is 9-hydroxy-3,8-dimethoxy-4-prenylpterocarpan ( Figure 1). ,4,4 -Trihydroxydihydrochalcone-2 -O--d-glucopyranoside (6) H-) suggesting the presence of -CO-CH (OH)-CH2-moiety. Compound 6 showed the presence of a glucopyranosyl moiety which was linked to C-3 according to the HSQC correlation of anomeric proton at H 5.02 (br, d, = 7.3 Hz, H-1 ) with C-3 ( 164.03). The anomeric configuration was assigned as for the glucopyranosyl group from their coupling constants. The downfield shift of C-4 at 144.59 was indicative that it has a hydroxy group. Aromatic protons H-3 ( H 6.68), H-5 ( H 6.44), and H-6 ( H 7.71) suggested a 2 ,4 -dioxygenation in a trisubstituted ring, which is corroborated by the NOESY and the HMBC spectrum ( Figure 2). The AA XX spin system ( H 6.71, 7.22) of A ring indicates 1,4-disubstitution, whereas the chemical shift of C-4 ( 164.03) reveals hydroxyl group at this position. Therefore, the structure of compound 6 was identified as ,4,4 -trihydroxydihydrochalcone-2 -O-d-glucopyranoside (Figure 1).

Antioxidant Activity.
Radical scavenging capacities of compounds 1-5 isolated from E. polystachya are given in Table 3. However, compound 6 displayed lower scavenging value and the data are not presented here. Chalcones 4, 3, 5, and 1 showed significant scavenging activity in DPPH radical with IC 50 values of 10.21 to 19.39 g/mL. These results suggested that these compounds showed the highest hydrogen-donating capacity towards the DPPH radical.
Hydroxyl radicals are found in living systems that react with almost all biomolecules as they are high oxidant. The orders of reactivity of the flavonoids were similar to that of DPPH. Compound 4 (33.17 g/mL) was found to exhibit the highest hydroxyl radical scavenging effect followed by 5 (36.52 g/mL) and 3 (37.45 g/mL). Scavenging of hydrogen peroxide by the isolated compound was found to be better compared to the ascorbic acid (IC 50 443.76 g/mL). Among the tested flavonoids, the best scavenging of hydrogen peroxide activity was presented by 3 (IC 50 80.73 g/mL), 4 (IC 50 87.68 g/mL), and 1 (IC 50 91.13 g/mL). The presence in the -OH group in ring A and -OCH3 in ring B showed good antioxidant activity on scavenging of hydrogen peroxide activity compared to electron-withdrawing substituents at paraposition.
Superoxide radical is a free radical harmful to cellular components, precursor for many reactive oxygen species. Superoxide radical scavenging capacity followed an activity order of chalcones 4, 3, and 5.
As evident from Table 3, the isolated compound exerted nitric oxide scavenging activities with values of IC 50 of 25.47 to 75.04 g/mL. Compound 4 showed the highest scavenging effects on nitric oxide, followed by 3, 5, and 1. Reactive nitrogen species (N 2 O 4 , NO 3 , N 3 O 4 , NO, and NO 2 ) are formed during the reaction with superoxides or oxygen which are responsible for altering the functional and structural cellular components.
The ability to chelate and deactivate transition metals is an important mechanism of antioxidant activity, which catalyzes Fenton-type reactions and hydroperoxide decomposition. Therefore, it was considered of importance to study chelating ability of the isolated compound with the metal. The chelating activity of flavonoids to suppress the formation of ferrozine-Fe 2+ complex decreased in the order of 4 > 5 > 3 > 1 > 2.
Chalcones 4 and 3 and pterocarpan 5 also showed good significant scavenging activities of Trolox (16.23 to 20.34 nM) indicating their good antioxidant effect.
The inhibition of the isolated compound on the oxidation of BSA induced by H 2 O 2 was also determined. Exposure of H 2 O 2 to BSA produced a decrease of free thiol groups level. Results are shown in Table 3; the isolated compound significantly inhibited the oxidation of BSA induced by H 2 O 2 in a dose-dependent manner. Among the five compounds tested, 4 showed the strongest antioxidant activity at a concentration of 300 mg/mL preventing the 97% of the depletion of protein thiol groups. These results indicated that hydrophilic antioxidant moiety in the compounds has the ability to scavenge H 2 O 2 reducing thiol groups oxidation in BSA.
The antioxidant activity in all tested assays indicates that the compound with chalcone skeleton showed the best inhibitory activity when the dihydrochalcone as in compounds 4, 3, and 1 indicated that chalcones having electron donor group in the phenyl ring A in orthoposition to carbonyl moiety showed better antioxidant activity. Moreover, the activity decreasing obviously with glycosidation such as only 1 showed a moderate inhibitory activity, and the other glycoside 2 showed a low inhibitory effect. However, pterocarpan showed a good antioxidant activity in various oxidation systems. Antioxidants compounds have been shown to delay, inhibit, and prevent the oxidation, possibly through the mechanisms interacting with biological systems as scavenging free radicals, absorbing oxygen radicals, and chelating of the metal ions [29].

Flavonoids Prevent Palmitate-Induced ROS Production and Oxidative Stress in H9c2
Cells. The ability of 1-5 to decrease the oxidative stress in cells was determined. It is well known that concentrations of palmitate at 500 m increase oxidative stress and stimulate ROS production in H9c2 cell. H 2 O 2 model has been used to determine the oxidative stress because it can generate superoxide anion radical. Pretreatment with flavonoids for 1 hr at 20 M significantly decreased ROS production preventing oxidative stress in H9c2 cell and palmitate-induced ROS production using DCFH-HA (for H 2 O 2 − ) and DHE (for O 2 − ) test. The oxidative stress reduction potential of 1-5 is given in Table 4. The capacity of 10, 50, and 100 g/mL flavonoids to reduce the oxidative stress was compared with that of 25 g/mL of ascorbic acid. The values of the mean fluorescent intensity indicated that the isolated compound significantly reduces the increase of ROS in palmitate-induced cells. The antioxidant effectiveness in vitro of the isolated compound can be possibly due to their ability to act as free radical scavengers, quenchers of singlet

Effects of Flavonoids on Oxidative Stress Markers and Aminotransferases
Activities. There were no changes of blood glucose level after 30 days of administration of 1-5. There was a significant increase of lipid peroxide in the liver, kidney, and pancreas in diabetes mice. However, the administration of 1-5 improved these levels in the treated groups with respect to the diabetic control group as shown in Table 5. The production of ROS including hydroxyl, superoxide anions and hydrogen peroxide, nitric oxide, DPPH, ABTS radicals, protein oxidation, and lipid peroxidation can be enhanced by treatment with isolated 1-5, and increases in radical scavenging enzymes of CSH-Px, SOD, GSH, and CAT have also been observed in liver, pancreas, and kidney in the group of diabetic animals treated. Elevation of serum biomarker enzymes such as total protein, SGPT, SGOT, ALP, and TB was observed in diabetic mice indicating deterioration in liver function (Table 6) which may be because STZ produced liver damage causing leakage of these enzymes into the blood [31]. However, treatment of STZ-induced diabetes in mice with 1-5 reduced levels of these enzymes leading to hepatoprotective effect.

Effects of Flavonoids on TBARS Levels.
In diabetic mice, elevated levels of oxidative stress are due to protein glycation, autoxidation of glucose, low activities of antioxidant  All values are expressed as mean ± SEM, = 10. a < 0.05 when compared to normal control group. b < 0.01 when compared to diabetic control group.
enzymes, and lipid peroxidation [32]. Consistent with this finding, decreased activities of antioxidant enzymes, such as GSH, SOD, GST, and GPx, and a marked increase in the concentration of TBARS indicate increase lipid peroxidation leading to decrease of the antioxidant defense mechanisms to avoid overproduction of ROS which leads to tissue injury [33]. Supplementation to diabetic mice with the isolated compound resulted in a significant ( < 0.05) diminution in lipid peroxidation levels, in liver and kidney compared with diabetic control, and the observed reduction of TBARS towards normal levels (Table 7). These results suggest a marked inhibition of the isolated compound on ROS generation in STZinduced diabetic mice. TBARS levels are an index of oxidative stress and endogenous lipid peroxidation which intensifies with increasing free radicals production. Therefore, in diabetic patients measurement of TBA-reactive substance levels is used to determine the level of oxidative stress. In addition, increased lipid peroxidation in diabetic mice might be due to stimulation of hepatic triglyceride synthesis [34].