Polyethylene glycol‐based ultrasonic‐assisted enzymatic extraction, characterization, and antioxidant activity in vitro and in vivo of polysaccharides from Lonicerae japonica leaves

Abstract In this paper, polyethylene glycol (PEG)‐based ultrasonic‐assisted enzymatic extraction (UAEE) was employed in polysaccharides extraction from Lonicerae japonica leaves (LJLP). The optimal extraction conditions (extraction time of 33 min, PEG concentration of 30%, and ultrasonic power of 191 W) were obtained by Box–Behnken design (BBD). Under this condition, the LJLP yield was 14.76%. Furthermore, LJLP was identified as a typical heteropolysaccharide considering the main constitutive monosaccharides include galactose (32.3%), glucose (20.9%), and ribose (15.2%). Moreover, LJLP exhibited high total reducing power and considerable scavenging activities on superoxide radicals, in a concentration‐dependent manner in vitro. In addition, antioxidant analysis in vivo revealed that for the LJLP‐treated mice, the superoxide dismutase (SOD), glutathione peroxidase (GSH‐Px), catalase (CAT) activities, and total antioxidant capacity (TAOC) were significantly increased, while the level of malondialdehyde (MDA) was decreased in both serum and liver.

method, ultrasound-assisted extraction (UAE) has been investigated in the extraction of LJLP, and enzymes have been introduced to disrupt the plant cell walls, which help to enhance the extraction yields.
After preparation through the extracting conditions, the crude LJLP was purified by ethanol precipitation. Then, the preliminary characterization of these polysaccharides was studied by high-performance liquid chromatography (HPLC) and Fourier transform infrared spectroscopy (FT-IR). Moreover, antioxidant activities of LJLP were also evaluated in vitro and in vivo.

| Materials and chemicals
Lonicera Japonica leaves were provided by Hubei Chutianshu Pharmaceutical Co., Ltd. The dried leaves were filtered by a 40 mesh screen after ground and set aside. The power was subjected to degrease by petroleum ether (60-90°C), 80% ethanol help to remove some potential impurities such as flavonoids, lipids, pigments, monosaccharides, oligosaccharides, and so on. The chemical reagents used in experiments were all analytically pure and purchased from Sinopharm Chemical Reagent Co., Ltd. Distilled water was used throughout the study.

| Extraction of polysaccharides
In this study, ultrasound-assisted extraction was performed in an ultrasonic-microwave combined reaction system (Nanjing Atpio Instruments Manufacturer). 5.0 g prepared samples were immersed in 100 ml PEG aqueous solution, and enzyme solution (cellulose:pectase:trypsin was 2:2:1) was added to improve the efficiency of wall-breaking. Through ultrasonic extraction, centrifugation (7104 g, 10 min), the supernatant was sufficiently mixed with 4 times volumes of anhydrous ethanol and stored at 4°C overnight. The precipitate was harvested by centrifugation and dried to gain the raw LJLP. The content of LJLP was determined by the phenol-sulfuric acid method with D-glucose as the standard substance (Dubois, Gilles, Hamilton, Rebers, & Smith, 1956). The formula of extraction yield (%) of LJLP was as follows: LJLP yield (%) = LJLP weight (g)/sample weight (g).

| Optimization of extraction conditions by BBD
According to the results of the single-factor experimental data, a Box-Behnken design of three levels and three independent variables (PEG concentration, X 1 ; ultrasonic power, X 2 ; and extraction time, X 3 ) was applied to investigate the influence on the extraction yield (Y, %) in this optimization study. Furthermore, each variable was set at three levels (high, medium, and low), coding + 1, 0, and −1, respectively. The BBD design was composed of 15 experiments, which include 12 factorial points and 3 center points in a random permutation (Table 1). A statistical software, named Design-Expert, was used for data analysis by mathematical modeling and regression analysis.
All experimental data were analyzed by multiple regressions to fit the following empirical second-order polynomial model: where Y is the response variable; β 0 is the constant; and β i , β ii , and β ij are the linear coefficient, quadratic coefficient, and the cross-product coefficient, respectively. X i and X j are the different independent variables. (1)

| Analysis of monosaccharide composition
The monosaccharide compositions of crud LJLP were measured by HPLC with procedure of previous report (Liu et al., 2017), which had a few modifications. 10 mg of LJLP was treated with 3 mol/L trifluoroacetic acid (TFA) at 100°C for 6 hr. After removal of TFA, other impurities were stripped from the residue by washing with methanol for 3 times and then redissolved in 2 ml water. After adding 0.2 ml 0.5 mol/L 1-phenyl-3-methyl-5-pyrazolone (PMP) methanol solution, 0.2 ml 0.3 mol/L NaOH solution was mixed with the hydrolysate and incubated for 45 min at 65°C. The next step is the acid neutralization treatment of the mixture, followed by adding 1 ml trichloromethane and extracting for 3 times. The aqueous phase was collected to be determined by HPLC.
The determination process was carried out with an Agilent 1,260 HPLC system. The chromatographic conditions were as follows: Phenomenex Gemini C 18 column (250 mm × 4.6 mm, 5 μm) with the temperature of 35°C; flow rate 0.8 ml/min; and detector wavelength 245 nm.
As for the eluting solvents, it was a mixture of acetonitrile (82:18, v/v) and 0.05 M phosphate buffer (pH 6.8). The injection volume was 10 μl.

| FT-IR spectrometric analysis
The IR spectrum of crude LJLP was recorded with a Fourier Transform Infrared Spectroscopy (FT-IR) (Bruker Equinox55, Germany) with the frequency range of 4000-400 cm −1 . The sample was analyzed as KBr pellets.

| Superoxide radical assay
The superoxide anion radical scavenging activity of LJLP was meas- respectively, and incubated for 5 min at room temperature, followed by determination at 560 nm. Ascorbic acid was used as standard.

| Total reducing activity
The total reducing power of LJLP was assessed according to a reported procedure (Sun, Yang, Lu, Wang, & Zhao, 2013). 2.5 ml 0.2 M phosphate buffer saline (pH 6.6) and 2.5 ml 1% (w/v) K 3 Fe(CN) 6 solution were added to 1 ml sample solution and mixed completely. After incubation for 20 min at 50°C, the mixture was mixed with 2.5 ml of 10% (w/v) trichloroacetic acid (TCA) solution. 2 ml of the supernatant was collected by centrifugation, followed by adding 2 ml distilled water and 0.5 ml 0.1% (w/v) FeCl 3 . The sample was determined at 700 nm, using ascorbic acid as a standard.

| Effects of PEG molecule weigh on LJLP yield
Obtaining the polysaccharides usually includes two steps: (1) Polysaccharides were extracted from raw materials with solvents; (2) the precipitation of polysaccharides from the extraction solution.
PEG forms the hydrogen bond with polysaccharides, which could help to achieve higher solubility of polysaccharides. So PEG shows a positive effect on diffusion coefficients of extraction solvent (Zhou, Liu, Ma, & Zhang, 2014). In our study, PEG-200, PEG-400, PEG-600, PEG-800, and PEG-1000 aqueous solutions (30%, v/v) were tested as extracting solvents of LJLP with UAEE, respectively. As shown in Figure 1a, a PEG-600 aqueous solution supported the optimal yield of LJLP (14.25 ± 0.16%, n = 3). It may be explained by the fact that both the viscosity and polarity of PEG are closely related to its molecule weight (Zhou et al., 2011), and PEG-600 can help to gain the maximal extraction yield of LJLP by providing a better solution condition, compared with other PEG aqueous solutions. Therefore, PEG-600 was selected for the further studies.

| Effects of PEG concentration weigh on LJLP yield
To find suitable polarity of extraction solvent for the extraction of LJLP, different PEG concentrations were investigated in this study.
As shown from Figure 1b, the extraction yields started to increase with increasing PEG concentration, and similar change rules of extracting polysaccharides from Ginkgo biloba leaves had been reported  and reached a maximum at the 30%, then followed by an obvious decrease with further increases in PEG concentration. It was because that a relatively lower PEG concentration might support a more suitable viscosity of PEG aqueous solution, which may increase the efficiency of extraction and energy transferring in solution. Thus, the PEG-600 concentration of 30% was the most suitable choice.

| Effects of Ultrasonic power on LJLP yield
Appropriate ultrasonic power is a benefit to improve the efficiency of disrupting cell wall, which has a crucial influence on the extraction rate of polysaccharide. The ultrasonic power in the UAEE was studied in the range from 120 to 280 W, while keeping the other extraction conditions at fixed levels. As can be seen in the Figure 1c, there was a significant increase in the extraction rate of LJLP with increasing ultrasound power from 120 to 200 W, which peaked at 200 W and then decreased. It may be the reason that proper ultrasound power could accelerate solvent penetration into the plant cells and benefit dissolution of the components to be extracted. But over high level of ultrasonic power may destroy the polysaccharide (Li et al., 2013;Zhang, Guo, Wang, & He, 2015). Therefore, the optimal ultrasonic power should be 200 W.

| Effects of enzyme concentration on LJLP yield
To enhance the efficiency of cell disrupting, which benefits polysaccharide releasing from the cell, complex enzymes (cellulose, pectinase, and papain) were used in this study. As shown in Figure 1d,

| Effects of extraction temperature on LJLP yield
The

| Effects of extraction time on LJLP yield
To determine the shortest time for sufficient recovery for LJLP in a single extraction, the extraction time was assessed in this test.

| Optimization of the LJLP extraction
Based on the results of single-factor experiment, three experimental factors (PEG concentration, ultrasonic power, and extraction time) were selected as the independent variables of BBD, while other factors (extraction temperature, enzyme concentration, and PEG molecule weight) were fixed at 60 • C, 2.0% (w/v), and PEG-600, respectively. From BBD and experimental results of extraction yield of LJLP (Table 1), a second-order polynomial regression model was as follows: From the analysis of variance (ANOVA) results (Table 2) It is well known that the model coefficient has a statistical significance considering that the value of "Prob > F" is less than 0.05 (Liu et al., 2013). Based on the level of p-value, the linear variables X 3 , and the quadratic variables X 1 2 , X 2 2 , and X 3 2 were statistically very significant at p < .0001, whereas the linear variables X 1 and X 2 , and two-variable interaction X 1 X 2 , X 1 X 3 , and X 2 X 3 had significant influence (p < .05) on the extraction yield of LJLP. Therefore, three independent variables (PEG concentration, ultrasonic power, and extraction time) were all significantly associated with the extraction yield of LJLP.
To uncover how experimental factors and their interactions affect the extraction rate, Design-Expert was used to plot response surfaces (Zhang, Guo, et al., 2015). The results of extraction yield of LJLP influenced by extraction time (X 1 ), PEG concentration (X 2 ), and ultrasonic power (X 3 ) are shown in Figure 2a

| Infrared spectra analysis
As shown in Figure

| Superoxide radicals scavenging activity
Superoxide radical has harmful effects on DNA and membrane lipid of the cell which may lead to cell damage (Macdonald, Galley, & Webster, 2003). The comparisons were made between LJLP and ascorbic acid based on their superoxide radical scavenging activity.
As shown in Figure 5a, LJLP demonstrated that scavenging activity on superoxide radicals was in direct proportion to its concentration.
LJLP at a concentration of 6 mg/ml supported the maximum scavenging ability (81.56%), while ascorbic acid could maximally take the scavenging activity (98.52%) at 2 mg/ml. In relatively low concentration, ascorbic acid exhibited much higher scavenging activity F I G U R E 2 Response surface plots for the effect of independent variables on the extraction yield of LJLP than LJLP. However, as concentration went up to 1.0 mg/ml, LJLP exhibited much stronger scavenging ability. The above results revealed that LJLP had a significant effect on scavenging superoxide radicals.

| Reducing power assay
In this experiment, the Fe 3+ /ferricyanide complex was reduced into the ferrous form in the presence of reducers (i.e., antioxidants). Thus, The HPLC chromatograms of 8 standard monosaccharides (a) and component monosaccharides released from LJLP (b). Peaks 1. mannose, 2. ribose, 3. rhamnose, 4. glucose, 5. xylose, 6. galactose, 7. arabinose, and 8. fucose F I G U R E 4 FT-IR spectroscopy of LJLP between 4,000 and 400 cm −1 the Fe 2+ concentration was determined by measuring the formation of Perl's Prussian blue at 700 nm (Jin et al., 2012). Figure 5b describes the relationship of the reducing power of LJLP versus their concentration.
The higher the absorbance value, the higher the total reducing power of LJLP. It can be seen from Figure 5b that the total reducing power of LJLP gradually increased with the increasing concentration and had a good correlation with polysaccharide concentration. The reducing power of LJLP was much lower than ascorbic acid. It may be the reason that the ketones of LJLP react with the specific peroxide precursors, thus preventing the formation of peroxide (Elmastaş et al., 2006).

| Antioxidant activities in vivo of LJLP
In this research, the antioxidant activity of LJLP in vivo was evaluated in a mice aging model, which was established by d-galactose. Tables 3 and 4, the activities of CAT, SOD, GSH-PX, and TAOC of the aging model group were significantly lower than those of the normal control group (NCG) in serum and liver, while the contents of MDA in serum and liver were significantly higher than those in the NCG. The above results suggested that establishment of the aging mice model was successful in this study.

As shown in
In recent years, more and more evidence showed that the development of aging was largely due to the cumulative damage, which was caused by intracellular reactive oxygen species (ROS) (Jing et al., 2016).

CO N FLI C T S O F I NTE R E S T
The authors declare no conflict of interest.

E TH I C A L A PPROVA L
All the animal experimental protocols were performed strictly in compliance with the Ethics Committee Guide of China Huanggang Normal University.