Extraction Optimization, Structural Characterization, and Anti-Hepatoma Activity of Polysaccharides from Scutellaria Barbata D. Don

Wenwen Su Mudanjiang Medical University Leilei Wu Mudanjiang Medical University Qichao Liang Mudanjiang Medical University Xiaoyue Lin Heilongjiang University of Chinese Medicine Xiaoyi Xu Mudanjiang Medicai University Shikai Yu Heilongjiang University of Chinese Medicine Yitong Lin Mudanjiang Medical University Yang Fu Mudanjiang Medical University Jiadong Zhou Mudanjiang Medical University Bo Zhang Mudanjiang Cancer Hospital Li Li Mudanjiang Medical University Dan Li Mudanjiang Medical University Yongkui Yin Mudanjiang Medical University Gaochen Song (  songgaochen32@sina.com ) Mudanjiang Medical University


Background
Liver cancer is the fourth leading cause of cancer deaths and ranks sixth among new cases in the world (1). Because of its early silence and high heterogeneity, most patients with liver cancer are diagnosed in the late stages, and the cure rate is low. According to current trends, the number of new cases and deaths due to liver cancer worldwide are expected to increase from 841,080 and 781,631 in 2018 to 1,361,836 and 1,284,252 in 2040, which constitute increases of 62% and 64%, respectively (2). The methods for treating liver cancer include surgical resection, liver transplantation, arterial chemoembolization, and systemic pharmacological treatment with kinase inhibitors such as sorafenib. However, nearly 70% of patients easily develop recurrent liver cancer or metachronous liver cancer (3), which also causes typical side effects such as hypertension, hypothyroidism, and leucopenia/neutropenia (4). Compared with the burdens of many other major cancers, although the incidence and mortality rates for liver cancer are increasing worldwide, the development of new drugs for liver cancer has historically been lacking. Therefore, the development of low toxicity and high-e ciency antitumor drugs has become the focus of researchers.
Many studies have shown that polysaccharides used in Chinese medicine have attracted interest in the eld of natural macromolecules because of their signi cant biological activities, which include antitumor (5), antioxidant(6) and anticoagulant activities (7), and treatment applications for ischemiareperfusion injury(8), liver protection (9), viral infections(10), control of blood lipids (11), control of blood pressure (12) and immune regulation (13). Scutellaria barbata D. Don is a plant of the family Lamiaceae (14), is listed in the 2015 edition of the Chinese Pharmacopoeia and has the unique effects of clearing away heat and toxic material, diuresis and detumescence. Recent pharmacological studies have provided strong evidence for antitumor, antioxidant, antibacterial, antiviral and immunomodulatory effects (15)(16)(17), and these medicinal values are attributed to the fact that Scutellaria barbata contains a variety of components, including alkaloids, avonoids, steroids and polysaccharides(18). As high molecular weight carbohydrates, polysaccharides are very important in the life activities of organisms. Scutellaria barbata polysaccharide can resist complement (19), inhibit the growth of lung cancer (20,21), and inhibit the proliferation and metastasis of colorectal cancer HT29 cells (22,23) and high glucoseinduced retinal vascular endothelial cells (24). In addition, our previous study found that Scutellaria barbata polysaccharide (SBPs) inhibit tumor growth in hepatoma H22-bearing mice (25). However, most of the studies on SBPs involved the crude polysaccharides, and the isolation, puri cation and antihepatoma activity of SBPs have not been studied. Furthermore, the unique antitumor properties of SBPs limit their potential for clinical application due to the lack of clear structural and mechanistic information.
Based on the above considerations, this study was designed to obtain the crude polysaccharide of Scutellaria barbata by using water extraction and alcohol precipitation, optimize the extraction process with response surface methodology and purify the crude SBP by column chromatography to obtain SBP-1A and SBP-2A. The polysaccharide content, molecular weight, monosaccharide composition and basic structure were preliminarily identi ed. Then, a MTT assay was used to identify the polysaccharide components with anti-hepatoma effects. The antitumor activity of SBP-2A was evaluated by colony formation tests, morphological observations, apoptosis and cell cycle analyses.  (33). The elution solutions were combined, rotated and concentrated to 1/8 of the original volume at 55 ℃; a dialysis bag with molecular weight of 3500 Da was used for 24 h to collect the dialysate, which was prefrozen at -40 ℃ for 24 h, and then freeze-dried into powder (34). Samples were placed on a SephadexG-100 dextran gel column and washed with ultrapure water at 5 ml/tube. The polysaccharide content of each tube was measured by the phenol sulfuric acid method, and elution curves were drawn. If the component showed a single symmetrical elution peak after the SephadexG-100 gel column, it contained a single polysaccharide component. The main peak components were concentrated, frozen and dried to obtain white occulated polysaccharides.

Polysaccharide and protein content
A glucose standard solution, 0.1 mg/ml, was prepared and 0, 0.4, 0.6, 0.8, 1.0, 1.2, 1.4, 1.6, and 1.8 ml of glucose standard solution was added into stoppered test tubes. Distilled water was added to establish a volume of 2 ml, 1 ml of 5% phenol was added and mixed well, and 5 ml of concentrated sulfuric acid was added quickly. The mixture was warmed in a 40 ℃ water bath for 30 min and allowed to stand for 20 min, after cooling to room temperature, the absorbance value at 490 nm was measured with an ultraviolet spectrophotometer; the glucose standard concentration (μg/ml) was the abscissa, and the absorbance value A was the ordinate used to draw the glucose standard curve. One milligram of SBP was weighed and fully dissolved in distilled water. The absorbance value was determined at 490 nm according to the phenol sulfuric acid method (33), and the polysaccharide concentration in the sample was calculated according to the standard curve. Then, according to the number of standards and samples, a BCA working solution was prepared to give a 50:1 ratio of BCA reagent and Cu reagent. A 10 μl sample of BSA standard was diluted to 100 µl (0.5 mg/ml) with PBS. The standard (0, 2, 4, 6, 8, 12, 16, 20 μl) was added to 96 well plates, and PBS was added to increase the volumes to 20 μl. The sample was weighed and diluted with PBS, and 20 µl was added to a 96 well plate with 200 µl/well BCA working solution; the plate was warmed at 37 ℃ for 15-30 min. The absorbance value at 562 nm was measured by enzyme labeling instrument (Multiskan MK3, ThermoFisher, USA), and the protein concentration was calculated according to the standard curve.

Monosaccharide composition
Sixteen monosaccharide standards were prepared as 10 mg/ml standard solutions. The 0.01, 0.1, 0.5, 1, 5, 10 and 20 mg/L gradient concentration standards of monosaccharide standard solutions were labelled as standards 1-7. Ten milligrams of sample was accurately weighed and placed into an ampoule bottle, 10 ml of 3 M TFA was added and the substrate was hydrolyzed at 120 ℃ for 3 h. The acid hydrolysis solution was absorbed, transferred into the tube, blown dry with nitrogen, added to 5 ml of water, vortexed and mixed evenly, diluted to 100 µl, added to 900 µl of deionized water, and centrifuged at 12,000 rpm for 5 min. The supernatant was removed for ion chromatography (IC) analysis(35) with an ICS5000 system (Thermo Fisher, USA).

Molecular weight determination
High-performance liquid chromatography (HPLC) is widely used in determinations of molecular weight because of its high accuracy and e ciency(36). Samples and standards were weighed accurately (molecular weights were 5000, 11600, 2800, 48600, 80900, 148000, 273000, 409800, 667800 and 3693000 Da, respectively), 5 mg/ml solutions were prepared, the solutions were centrifuged at 12000 rpm for 10 min and the supernatant was ltered through 0.22 μm microporous lters, and the sample was transferred into a 1.8 ml injection vial. The chromatographic conditions for HPLC (LC-10A, Shimadzu, Japan) were as follows: chromatographic column: BRT105-104-102 tandem gel column (8×300 mm); mobile phase: 0.05 NaCl solution; ow rate: 0.6 ml/min, column temperature: 40 ℃; injection volume: 20 μl; detector: differential refractive index detector RID-1OA. By taking the logarithm of the molecular weight of the standard (log(Mw)) as the ordinate and the peak time (min) as the abscissa, regression tting of the curve was performed with software to obtain the standard curve for molecular weight distribution. The chromatogram of the sample was obtained with the chromatographic separation conditions described above, the retention time of a single symmetrical peak was recorded, and the molecular weight was calculated.

FT-IR analysis
Samples were analyzed by Fourier transform infrared (FT-IR) spectroscopy(37) in a pressed KBr pellet. Polysaccharide (5 mg) was weighed, put into a mortar with dried KBr, mixed and ground fully, pressed into a thin sheet with a pellet press, and analyzed with a Thermo Scienti c Nicolet iS5 FT-IR spectrophotometer (Thermo Nicolet Co., Madison, WI, USA) with a scan range of 500-4000 cm -1 .  precipitates were obtained by centrifugation, 300 µl of precooled PBS was added to resuspend them, 700 µl of precooled absolute ethanol was slowly added drop by drop with mixing, and then the cells were xed overnight at -4 ℃. The xing solution was washed away with PBS, and 100 µl of RNase A solution was added to the cell precipitate to resuspend it. The mixture was warmed in a water bath at 37 ℃ for 30 min, 400 µl of PI staining solution was added and mixed, and the cells were incubated in the dark at 4 ℃ for 30 min. Finally, the cell cycle of the sample was evaluated by Cytomics FC500 Flow Cytometry CXP (Beckman, USA) and the cycle results were analyzed by ModFit LT 5.0 software.

Statistical analyses
All data were expressed as the mean ± standard deviation (SD). All experiments were repeated three times. One-way ANOVA and multiple comparisons were used, and all statistical analyses were performed using GraphPad Prism 7.0 software and plotted with Origin2021. P < 0.05 was taken to indicate a statistically signi cant difference. ratio. When the liquid-material ratio was more than 1:25, the extraction rate increased rapidly, which may be because the increase in the liquid-to-material ratio allowed the polysaccharides to dissolve more fully(38, 39). Therefore, 1:25 was chosen as the center for the response surface experiment. As shown in Fig. 1(B), the effects of 60 ℃, 70 ℃, 80 ℃, 90 ℃ and 100 ℃ temperatures on the extraction of SBP were investigated. When the extraction temperature was 90 ℃, the polysaccharide extraction rate reached the highest value. When the extraction temperature was varied between 60 ℃ and 90 ℃, the rate of polysaccharide extraction increased. With a further increase in temperature, the extraction rate showed a downward trend. Therefore, an extraction temperature of 90 ℃ was selected as the central point for the response surface experiment. As shown in Fig. 1(C), when the extraction times were 1 h and 2 h, the rate of SBP extraction was higher, and prolonging the extraction time improved the yield of polysaccharide (40); the rate of SBP extraction decreased with extension of the time from 2 h to 4 h, and it is possible that the structures of the polysaccharides changed during the long extraction process (41). Therefore, 2 h was selected as the center for the response surface experiment.

Regression model and analysis of variance
According to the results of the single-factor experiment, a response surface optimization experiment with three factors and three levels was designed, with the rate of polysaccharide extraction as the response value and the liquid-solid ratio (A), extraction temperature (B) and extraction time (C) as the in uencing factors. The experiment was designed with Design Expert 11 software. As shown in Table 1, a binary multiple equation relating the extraction rate of SBP (y) and the three factors was obtained by data analysis: Y=3.91+0.045A+0.125B+0.113C-0.068AB+0.098AC-0.118BC-0.991A 2 -0.271B 2 +0.034C 2 Y is the extraction e ciency of SBP, A is the ratio of material to liquid, B is the extraction time, and C is the extraction temperature.
Signi cance tests were conducted for all models and regression model coe cients, and the results are shown in Table 2. The F value of the model was 143.86, and P < 0.0001, which indicates that the regression model was very signi cant; the model mismatch term p value was 0.3838 (P > 0.05), so it was not signi cant at the level of α=0.05; this indicates that model tting was effective and experimental error was small. The correlation coe cient R 2 was 0.9946, indicating that the simulated value of the model was consistent with the actual predicted value and the prediction of the model was reasonable; the coe cient of variation (CV) was only 1.85%, so the model exhibited good repeatability and high accuracy. From the F values for the three in uencing factors (A, B and C), it can be concluded that the effects of in uencing factors on the extraction rate of SBP decreased in the order extraction time > extraction temperature > solid-liquid ratio.

Response surface analysis and model optimization
The response surface diagram and the contour map derived from the multiple quadratic regression model were used to evaluate the pairwise interactions of experimental factors and impacts on the extraction of SBP and determine the optimal level range for each factor. The steeper the slope of the response surface was, the higher the response sensitivity. The shape of the contour line re ects the strength of the interaction. The contour line for the solid-liquid ratio and the extraction time tended to be oval, indicating that the interactions were signi cant; the interactions of other factors were not signi cant (Fig. 2). The optimum extraction parameters for SBP were a solid-liquid ratio of 1:25.36, extraction time of 120.3 min, and extraction temperature of 100 ℃.

SBP puri cation
Ion chromatography and gel column chromatography are usually used in the separation and puri cation of polysaccharides. The polysaccharide fractions with single peaks obtained by elution with 0.1 M and 0.2 M NaCl solutions were the largest (Fig. 3). Therefore, the eluates with these two elution peaks were collected and named SBP-1 and SBP-2. The two components SBP-1 and SBP-2 were puri ed by a SephadexG-100 gel column. As shown in Fig. 4(A) and (B), the two components exhibited single and symmetrical elution peaks, demonstrating that the two polysaccharides were relatively homogeneous; the samples with the two single elution peaks were collected, freeze-dried into white occulent powders, and named SBP-1A and SBP-2A. Five polysaccharide components were puri ed by ion column chromatography, but only polysaccharide components with high content, high purity and uniformity, namely, SBP-1A and SBP-2A, were retained due to limitations of the experimental conditions.

Determination of polysaccharide content and protein content
Analysis showed that the polysaccharide content of SBP-1A was 93.2% and that of SBP-2A was 95.5%. A standard curve was obtained through determination of the BSA protein standard: y=1.1702x+0.1401, R 2 =0.9902. The protein contents of the polysaccharide samples were determined: the protein content of SBP-1A was 2.87% and that of SBP-2A was 0.87% (Table 3).

Monosaccharide composition
The monosaccharide compositions of SBP-1A and SBP-2A were analyzed by ion chromatography. The peak sequences and retention times of the monosaccharide compositions were compared with those of chromatograms for the monosaccharide standard samples (Fig. 5). For the mixed standard, the peak at 2.0 min was for sodium hydroxide, and that at 40 min was the peak for sodium acetate. Table 3 shows the molar ratios of the monosaccharide samples. SBP-1A was mainly composed of arabinose (30.6%) and galactose (38.4%), and the uronic acid content was 0.7%; SBP-2A was mainly composed of arabinose (36.3%) and galactose (42.7%), and the uronic acid content was 1.2%. The monosaccharide constituents of the two components were fucose, galactosamine hydrochloride, rhamaose, arabinose, glucosamine hydrochloride, galactose, glucose, xylose and mannose, but the molar ratios were different.

FT-IR Infrared spectroscopy
As shown in Fig. 7, the infrared analysis showed that SBP-1A and SBP-2A had characteristic absorption peaks for polysaccharides near 3400 cm -1 , and the strong absorption peak was the result of O-H stretching vibrations in polysaccharides. The peak near 2900 cm -1 was assigned to C-H stretching vibrations. There was a stretching vibration for C=O at 1634 cm -1 , which may be due to a carboxyl or acetyl group, indicating the presence of uronic acid. The absorption peak near 1380 cm -1 was caused by the variable angle vibration of C-H, indicating that the polysaccharide had the β-characteristic absorption peak of dextran. A peak appeared near 1120 cm -1 , indicating that the sugar residue of the polysaccharide was mainly pyranose. rates increased signi cantly to 15.8% and 8.76%, and the proportion of apoptosis induced by 200 μg/ml SBP-2A was more signi cant. As displayed in Fig. 10(B), by analyzing the percentage of DNA content in G1, S and G2 cells of the cell cycle, we found that, compared with the untreated group, increases in the SBP-2A concentration increased the percentage of cells in the G1 phase from 52.28% to 62.03%, and the percentage of cells in the G2 phase was decreased from 34.22% to 26.97% in a dose-dependent manner.

Discussions
The common design methods for optimization of polysaccharide extraction include orthogonal tests and response surface optimization. In the optimization of the SBP extraction process, previous experiments were investigated by orthogonal testing. In this study, the response surface method was used for the rst time to optimize the SBP extraction process. Compared with the orthogonal test, this method excluded random errors from the test and continuously analyzed all levels of the experiment in the optimization process to produce a continuous prediction model. In the single-factor experimental design, the e ciency for extraction of polysaccharide was lower at 100 ℃ than at 90 ℃; this may be due to destruction of the polysaccharide structure when the temperature was too high, resulting in a decrease in the extraction rate (42)(43)(44). To facilitate practical operation, the process was improved as follows: the solid-liquid ratio was 1:25, the extraction time was 2 h, and the extraction temperature was 90 ℃. Under these conditions, three groups of repeated validation experiments were carried out, and the average extraction rate for SBP was 3.851 ± 0.13%, which was close to the predicted value of 4.06%; this indicated that the model had good predictability and that the process optimized by the response surface method was reliable. Two polysaccharides (SBP-1A and SBP-2A) had been isolated from Scutellaria barbata and preliminary characterization of the SBP-1A and SBP-2A was investigated. Infrared spectroscopy is a method used for rapid analysis of polysaccharides, and it can accurately identify speci c absorption peaks (45). The absorption peaks for SBP-1A and SBP-2A found between 4000−500 cm −1 were determined by infrared spectrometry using polysaccharide in a KBr pellet.
Many studies have shown that the low-toxicity natural polysaccharides extracted from Chinese herbal medicines inhibit the proliferation of tumor cells and selectively induce apoptosis(46). Different biological macromolecules can be obtained from different raw materials and extraction methods, and their types and structural characteristics affect the biological activities of polysaccharides (47)(48)(49). Studies have shown that long-term in vitro culturing or drug treatment of cells can inhibit cell colony formation (50,51).
Therefore, we further veri ed the effect of SBP-2A on the inhibition of HepG2 cell proliferation with colony formation assays. In vitro experiments preliminarily con rmed that SBP-2A inhibited the growth of HepG2 cells. We evaluated morphological changes in cells to better determine the induced apoptosis response to drugs (52)(53)(54). The morphological characteristics of HepG2 cells treated with SBP-2A for 48 h were directly observed by inverted microscopy. In this study, Hoechst 33258 staining was used to determine whether HepG2 cells could be induced to undergo apoptosis by SBP-2A. Cell cycle arrest and apoptosis are important processes in programmed cell death (55). To further evaluate the ability of SBP-2A to induce cell apoptosis and cycle arrest, HepG2 cells treated with SBP-2A for 48 h were stained by the annexin V-FITC/PI double staining method and the PI single staining method. The number of apoptotic cells and DNA content were measured quantitatively at each stage of the cell cycle by ow cytometry. In addition, the growth of normal and tumor cells was orderly in different stages of the cell cycle(56), which can induce cell cycle arrest and apoptosis and effectively inhibit the growth of tumor cells. Therefore, these results suggest that SBP-2A induced apoptosis of HepG2 cells and blocked apoptosis in the G1 phase of the cell cycle. The above experimental results showed that SBP-2A isolated and puri ed from Scutellaria barbata may be a candidate drug deserving further evaluation in cancer prevention, which provides information for further studies on the molecular mechanism of its effect on human liver cancer cells. This could provide a basis for determining the effects of SBP against liver cancer and support its potential use as a functional medicinal component. Research on the anti-hepatoma mechanism of SBP-2A and its antitumor activity in H22 tumor-bearing mice is in progress.

Conclusion
In this study, the parameters for SBP extraction were optimized by the response surface method, and two   Response surface and contour map of the three-factor interaction in the SBP extraction rate.    Absorption peaks for SBP-1A and SBP-2A were measured in KBr pellets by an infrared spectrometer with a spectral range of 4000-500 cm-1.

Figure 8
Proliferation inhibition levels for human hepatoma HepG2 cells treated with SBP, SBP-1A and SBP-2A for 48 h were measured by the MTT method.