Ethylenediaminetetraacetic acid (EDTA) was purchased from Generay Biotech Co. Ltd. (Shanghai, China). 2′,7′-Dichlorofluorescein diacetate (DCFH-DA) was obtained from Molecular Probes (Thermo Fisher Scientific, Inc., Waltham, MA, USA). The protease inhibitor mixture was purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Dihydrorhodamine 123 (DHR123) and diethylene triamine penta-acetic acid (DTPA) were purchased from Sigma-Aldrich (Merck Millipore, Darmstadt, Germany). Isoflurane (1%) was purchased from Piramal Critical Care, Inc., (Bethlehem, PA, USA). The Bio-Rad protein assay kit and pure nitrocellulose membrane were supplied by Bio-Rad Laboratories (Seoul, Korea). Phenylmethylsulfonyl fluoride (PMSF) and sodium dodecyl sulfate (SDS) were acquired from (Amresco, LLC, Solon, OH, USA). PBS-Tween (P2006) was obtained from Biosesang, Inc., (Gyeonggi-do, Korea). Rabbit polyclonal antibodies against Nox-4 (cat. no. sc-30141), p47^phox (cat. no. sc-14015) and NF-κBp65 (cat. no. sc-372), and mouse monoclonal antibodies against COX-2 (cat. no. sc-19999), iNOS (cat. no. sc-372), histone (cat. no. sc-8030) and β-actin (cat. no. sc-4778), and goat polyclonal antibodies against transforming growth factor (TGF)-β (cat. no. sc-146) and tumor necrosis factor (TNF)-α (cat. no. sc-1351) were purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). Rabbit polyclonal anti-AP-1 (cat. no. 2315S) was obtained from Cell Signaling Technology, Inc. (Danvers, MA, USA). Goat anti-rabbit (cat. no. sc-2774) and goat anti-mouse (cat. no. sc-2005) immunoglobulin G (IgG) horseradish peroxidase (HRP)-conjugated secondary antibodies were acquired from Santa Cruz Biotechnology, Inc. ECL Western Blotting Detection Reagents were supplied by GE Healthcare (Piscataway, NJ, USA).

The aerial parts of O. javanica were collected from agricultural farms in Pyeongyang-ri, Cheongdo-gun, Gyoungsanbuk-do, South Korea in April 2011, and identified by Professor Tae Hoon Kim (Daegu University, Daegu, South Korea). Freshly milled O. javanica plant material (1.4 kg) was extracted with 70% ethanol (EtOH; 10 liters, thrice) at room temperature. The solvent was then evaporated under vacuum, and the combined crude EtOH extract (81.5 g) was suspended in 20% methanol (MeOH; 3 liters), and partitioned sequentially against n-hexane (3 liters, thrice), ethyl acetate (EtOAc; 3 liters, thrice), and n-butanol (BuOH; 3 liters, thrice) to yield dried n-hexane-(5.5 g), EtOAc-(1.8 g), n-BuOH (10.6 g) and H₂O-soluble (53.6 g) residues. A portion (1.7 g) of the EtOAc extract was chromatographed using a column containing Toyopearl HW-40 [coarse grade; 3.0 cm internal diameter (i.d.)x48 cm] using water containing increasing amounts of MeOH in a stepwise manner. The 100% H₂O eluate obtained was subjected to column chromatography using YMC GEL ODS AQ 120-50S (1.6 cm i.d.x37 cm) using aqueous MeOH, to yield pure compound 1 (75.9 mg; retention time, 10.1 min). HPLC analysis was carried out using a YMC-Pack ODS A-302 column (4.6 mm i.d.x150 mm; YMC Co., Kyoto, Japan) with a linear gradient of 10% (v/v) acetonitrile (MeCN) in 0.1% formic acid/H₂O (detection, UV 280 nm; flow rate, 1.0 ml/min; 40°C), which was increased to 90% MeCN over 30 min and then to 100% MeCN over 5 min. Compound 1 was identified and characterized as persicarin by ¹H and ¹³C NMR and by comparing peaks with literature values (18). The ¹H NMR results were as follows: (600 MHz, DMSO-d₆) δ 8.06 (1H, d, J= 2.2 Hz, H-2′), 7.61 (1H, d, J=8.8, 2.2 Hz, H-6′), 6.92 (1H, d, J=8.8 Hz, H-5′), 6.43 (1H, d, J=2.2 Hz, H-8), 6.22 (1H, d, J=2.2 Hz, H-6), 3.83 (3H, s, MeO-3′). The ¹³C NMR results were as follows: (150 MHz, DMSO-d₆) δ 177.1 (C-4), 166.5 (C-7), 161.1 (C-5), 156.2 (C-2), 155.5 (C-9), 149.6 (C-3′), 147.1 (C-4′), 131.8 (C-5′), 121.2 (C-6′), 121.1 (C-1′), 115.2 (C-2′), 103.1 (C-10), 99.7 (C-6), 93.6 (C-8), 55.6 (MeO).

Animal experiments were performed according to the Guidelines for Animal Experimentation and approved by Daegu Haany University (approval no. DHU2015-011). A total of 24, 5-week-old male ICR mice (23–28 g) were purchased from Orient Bio Inc. (Gyeonggi, Korea). Mice were maintained under a 12-h light/dark cycle, and housed at a controlled temperature (22±2°C) and humidity (55±5%) with free access to food and water. After several days of adaptation, the mice were randomly separated into normal control (n=6) and diabetic groups. Mice in the diabetic group were injected intraperitoneally with STZ (Sigma-Aldrich; 120 mg/kg body weight) in 10 mM citrate buffer (pH 4.5). After 7 days of STZ injection, the glucose levels of blood taken from the tail vein were measured, and then the STZ-induced diabetic mice were divided into three groups (each n=6). Treatment with persicarin was initiated after confirming the induction of hyperglycemia in the diabetic mice by weight (33.8±0.5 g) and serum glucose level (288.3±6.5 mg/dl). Mice in the diabetic control group (Veh group) were given water orally, while those in the other two diabetic groups were orally treated with persicarin extracts daily for 10 days at a low or high dose (2.5 and 5 mg/kg body weight, respectively). The diabetic groups were compared with the normal (non-diabetic) control group. Body weight, food intake and water intake were determined every day during the experimental period. After administration for 10 days, mice were anesthetized with 1% isoflurane and blood samples were collected from the abdominal aorta of anesthetized mice. Serum was separated immediately by centrifugation. Subsequently, each mouse was perfused with ice-cold physiological saline, and then the liver was harvested, snap-frozen in liquid nitrogen and stored at −80°C until analyses were performed.

The serum glucose level was measured using a commercial kit (Glucose Test, cat. no. AM201; Asan Pharm. Co., Ltd., Hwaseong, South Korea). Hepatic functional parameters [alanine aminotransferase (ALT) and aspartate aminotransferase (AST)] were measured using a Transaminase CII-Test kit (cat. no. 431-30901; Wako Pure Chemical Industries, Ltd.).

The hepatic glucose level was determined using the method of Momose et al (19), with minor modifications. Hepatic tissue was homogenized with ice-cold 0.9% NaCl buffer, and then the homogenate was deproteinized with 0.15 M Ba(OH)₂ and 5% ZnSO₄. The supernatant was obtained by centrifugation at 1,670 × g for 15 min, and then the glucose level was determined using the aforementioned glucose test kit.

ROS generation was measured using the method of Ali et al (20). Hepatic tissue was homogenized on ice with 1 mM EDTA-50 mM sodium phosphate buffer (pH 7.4), and then 25 mM DCFH-DA was added to the homogenates. After incubation for 30 min, the changes in fluorescence values were determined at an excitation wavelength of 486 nm and emission wavelength of 530 nm. The TBA-reactive substance content was determined using the method of Mihara and Uchiyama (21).

ONOO⁻ was measured by the method of Kooy et al (22). Each sample was mixed with rhodamine buffer (pH 7.4), 5 mM DTPA and 5 mM DHR123. After incubation for 5 min at 37°C, the fluorescence intensity of the oxidized DHR123 was measured with a microplate fluorescence reader at excitation and emission wavelengths of 485 and 530 nm, respectively.

Nuclear protein extraction was performed using the method reported by Komatsu (23). Briefly, liver tissue was homogenized with ice-cold lysis buffer containing 5 mM Tris-HCl (pH 7.5), 2 mM MgCl₂, 15 mM CaCl₂ and 1.5 M sucrose, followed by the addition of a 0.1 M dithiothreitol (DTT) and a protease inhibitor mixture. After centrifugation (10,500 × g for 20 min at 4°C), the pellet was suspended with an extraction buffer containing 20 mM 2-[4-(2-hydroxyethyl)-1-piperazyl] ethanesulfonic acid (pH 7.9), 1.5 mM MgCl₂, 0.42 M NaCl, 0.2 mM EDTA and 25% (v/v) glycerol, followed by the addition of a 0.1 M DTT and protease inhibitor mixture. The mixture was placed on ice for 30 min. The nuclear fraction was prepared by centrifugation at 20,500 × g for 5 min at 4°C.

The post-nuclear fraction was extracted from the kidneys of each mouse. Briefly, liver tissue was homogenized with ice-cold lysis buffer (pH 7.4) containing 137 mM NaCl, 20 mM Tris-HCl, 1% Tween 20, 10% glycerol, 1 mM PMSF, and a protease inhibitor mixture. The homogenate was then centrifuged at 2,000 × g for 10 min at 4°C, and the protein concentration in each fraction was determined using a Bio-Rad protein assay kit.

To determine the expression of NF-κBp65, AP-1 and histone, 10 µg protein from each nuclear fraction was separated by 8% SDS-PAGE. For Nox-4/p47^phox, COX-2, iNOS, TGF-β and β-actin , 10 µg protein of each post-nuclear fraction was separated by 8–15% SDS-PAGE. The separated proteins were transferred to a nitrocellulose membrane, blocked with a 5% (w/v) skimmed milk solution for 1 h, and incubated separately with the primary antibodies (NF-κBp65, AP-1, histone, Nox-4/p47^phox, COX-2, iNOS, TGF-β and β-actin) overnight at 4°C at a dilution of 1:1,000. Following washing with PBS-Tween, the blots were incubated with the anti-rabbit or anti-mouse IgG HRP-conjugated secondary antibody for 1 h at room temperature at a dilution of 1:3,000. Each antigen-antibody complex was visualized using ECL Western Blotting Detection Reagents and detected using SENSI-Q2000 (Lugen Sci. Co., Ltd., Gyeonggi, South Korea). The band densities were determined using ATTO Densitograph software (CS Analyzer 2.0; ATTO Corporation, Tokyo, Japan), and quantified as a ratio to β-actin. The protein levels of the groups are expressed relative to those of the normal mice (represented as 1).

The excised liver samples were immediately fixed in 10% neutral-buffered formalin and, after embedding in paraffin, they were cut into 5-µm sections. After staining with hematoxylin and eosin (H&E), the sections were examined with a light microscope.

Data are expressed as the mean ± standard error of the mean. Statistical comparisons were performed using one-way analysis of variance followed by a Dunnett's test using SPSS 11.5.1 for Windows, 2002 (SPSS Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

The body weight gains, liver weight, food intake and water intake during the experimental period are shown in Table I. The diabetic control mice showed a significant reduction in body weight in comparison with non-diabetic mice. However, the body weights of mice treated with persicarin were notably higher than those diabetic control mice. In addition, the administration of persicarin did not affect food and water intake, but the liver weight was significantly increased by ~1.19- and 1.2-fold in the low and high dose groups, respectively.

Table II shows that the serum glucose level was significantly increased in diabetic control mice; the increase was ~4.9-fold in comparison with that in non-diabetic mice. Persicarin treatment led to a notable reduction of glucose level in a dose-dependent manner. Hepatic functional parameters, namely ALT and AST levels in the serum, of vehicle- and persicarin-treated diabetic mice were investigated. The activities of ALT and AST in the diabetic mice were significantly higher than those of normal mice, while their activities in the persicarin treatment groups were markedly reduced in a dose-dependent manner.

As Table II demonstrates, persicarin exhibited an effect on hepatic glucose. Persicarin administration markedly reduced hepatic glucose levels at a dose of 5 mg/kg.

As shown in Fig. 1, the biomarkers of oxidative stress, namely ROS, ONOO⁻ and TBARS in diabetic control mice were notably elevated compared with those in normal mice. However, the elevated levels were diminished by oral treatment with persicarin.

The results of western blot analysis of hepatic Nox-4 and p47^phox protein in the hepatic tissues of the four groups are shown in Fig. 2. Persicarin administration showed a tendency to decrease the Nox-4 level (although the reduction was not statistically significant), whereas the p47^phox protein expression level was significantly decreased.

As is shown in Figs. 3 and 4, the levels of expression of various inflammation-related proteins, specifically, NF-κBp65, AP-1, iNOS, COX-2 and TGF-β were significantly increased in the livers of diabetic control mice compared with those in non-diabetic mice. NF-κBp65 and AP-1 protein levels were reduced significantly to become comparable with those of normal mice following treatment with persicarin. In addition, TGF-β and COX-2 protein expression levels were notably decreased by the administration of persicarin at a dose of 5 mg/kg. Furthermore, iNOS protein expression was significantly decreased in a dose-dependent manner by treatment with persicarin.

Histological evaluation of the hepatocellular damage was conducted (Fig. 5). The level of hepatocellular damage was higher in the vehicle-treated diabetic mice compared with normal mice. However, the administration of persicarin was observed to attenuate the hepatocellular damage in STZ-treated diabetic mice.