Response Surface Methodology for Optimizing the Ultrasound-Assisted Extraction of Polysaccharides from Acanthopanax giraldii

Abstract

The aim of this study was to investigate and optimize the most important factors affecting the extraction of Acanthopanax giraldii HARMS polysaccharides (AHPs) by ultrasound-assisted extraction (UAE) technology in a systemic manner. The ranges of four factors, including extraction temperature, liquid/solid ratio, extraction time, and ultrasonic power, were first determined by a single-factor experiment, followed by optimization of the UAE conditions using the Box–Behnken design (BBD) for maximum AHPs production. In our study, the models developed from the experimental design predicted the experimental data well and had a high determination coefficient (R²=0.9387). The optimized conditions for AHPs extraction were as follows: extraction temperature, 58°C; liquid/solid ratio, 25 : 1; extraction time, 73 min; and ultrasonic power, 85 W. Under these optimized conditions, the polysaccharide yield was 1.532±0.037% (n=3), being very close to the predicted value of 1.546% by the model. In addition, to investigate whether there was a difference of AHPs content between UAE and traditional hot water extraction (THWE), Fourier-transform (FT) IR spectral analyses was performed. The results showed that the functional groups of the polysaccharides extracted by either UAE or THWE were fundamentally identical. Furthermore, AHPs extracted by UAE could promote macrophage activation, such as enhanced phagocytosis and increased cytokine (interleukin-6 (IL-6), and tumor necrosis factor-α (TNF-α)) secretion in RAW264.7 cells. In conclusion, optimization of the UAE conditions by response surface methodology (RSM) was a promising method to improve the extraction yield of AHPs. AHPs extracted by the optimized UAE method can maintain their polysaccharide structure and biological activity.

Acanthopanax giraldii HARMS belongs to the plant of Araliaceae, which is mainly distributed in northwest regions of China.^1,2) In traditional Chinese medicine, the stem bark of A. giraldii HARMS is used to strengthen bones and muscles as well as to ease back and leg pain by relieving rheumatism. A variety of such natural bioactive compounds, including polysaccharides, volatile oils, and saponins, have already been separated from A. giraldii HARMS, among which A. giraldii HARMS polysaccharides (AHPs) are the main active ingredients. AHPs exhibit many physiological properties, including antitumor, anti-inflammatory, antiviral, liver protection, immune system boosting, and other effects.³⁾ However, research reports on AHPs extraction methods are limited.

Recently, increasing numbers of polysaccharides from plant sources are being used in medicine and health foods. However, the traditional extraction methods of plant polysaccharides, such as hot water extraction, are unsatisfactory due to the following shortcomings: long extraction time, high temperature, low selectivity, and low extraction percentage. The development of new advanced techniques has significantly improved the production of natural bioactive compounds, including plant polysaccharides. For example, ultrasound-assisted extraction (UAE) is a more effective and rapid extraction technique, compared with the traditional methods, and has been applied to extract natural bioactive compounds from different sources.⁴⁾ Compared to conventional and other modern extraction techniques, UAE can significantly increase the extraction yield of natural bioactive compounds due to utilization of high-intensity shock waves and cavitations that can constrain the chemical constituents into the solvent at a reduced temperature.^5–7) However, the UAE method also has its disadvantages, such as a high power of ultrasonication, which can cause damage to the polysaccharide structure, leading to difficult separation. Therefore, it is essential to optimize the UAE method to produce higher yields.

Among a variety of tools for yield optimization, response surface methodology (RSM), a novel method comprised of mathematical and statistical techniques, is currently widely used for optimizing the extraction processes in the pharmaceutical and food industries. Since extraction processes are often affected by many factors and their interactions,⁸⁾ it is better to understand the effects of those factors alone or their interactions on the extraction yield. Considering its high efficiency and ease of operation, the Box–Behnken design (BBD) is the best type of RSM that is commonly used to optimize technical parameters.^9,10) Therefore, in this study, we selected the BBD method to optimize the UAE conditions for AHPs, including the liquid/solid ratio, extraction temperature, extraction time, ultrasonic power, and interactions between these conditions.

Experimental

Plant Material

Bark of A. giraldii HARMS was purchased from Beijing Tong Ren Tang Group Co., Ltd. (China). The raw material of A. giraldii HARMS used for this study has been identified by Associate Professor Jinbao Tang, College of Pharmacy, Weifang Medical University (China). Prior to the extraction process, the gross sample was crushed in a grinder to reduce the particle size and passed through a 100-mesh sieve to obtain a fine powder. The powder was dried by heating at 70°C until a constant weight was reached.

Ultrasound-Assisted Extraction (UAE)

Four grams of dried sample powder was soaked in distilled water for 5 h. Samples were then extracted in an ultrasonic water bath (Kunshan Ultrasonic instruments Co., Ltd., China, KQ-100VDE) under a set of designed extraction temperatures (40–80°C), liquid/solid ratios (15 : 1 to 30 : 1 mL/g), extraction times (15 to 105 min), and ultrasonic power (40–100 W). After the UAE procedure, the extraction solution was abstracted from the insoluble residue by centrifugation at 5000 rpm for 20 min. To remove protein, the supernatants were treated three times by the Sevag method.^11,12) The resulting solutions were intensively dialyzed for 2 d against distilled water (cut-off molecular weight (MW)=7000 Da) and then mixed with four volumes of ethanol overnight at 4°C to isolate the polysaccharides. The precipitates were washed with ethanol.

Determination of Extraction Yield

The total polysaccharide content was determined by the phenol-sulfuric acid method, in which D-glucose was taken as the standard.¹¹⁾ The total polysaccharide extraction yield was calculated as the percentage of pretreated sample dry plant powder weight, according to the following equation:   

(1)

Experimental Design with Statistical Analysis

After determining the preliminary scope of the extraction variables by using a single-test design in our study, the BBD with four variables and three levels⁸⁾ was utilized to optimize the elements and maximize the extraction rate. Four independent parameters were selected for optimization of AHPs extraction, including extraction temperature (°C, x₁), liquid/solid ratio (mL/g, x₂), extraction time (min, x₃), and ultrasonic power (W, x₄). For statistical analysis, as exhibited in Table 1, the parameters were coded according to Eq. 2:   

(2)

where X_i stands for the variable’s coded value; x_i represents an independent variable’s actual value; x₀ is an independent variable’s actual value at the center point; and Δx_i is an independent variable’s rate change value. A total of 29 experiments were then designed (Table 2). Three replicate runs at the design center were performed to allow pure error estimation.¹³⁾ Experimental runs were randomized to minimize unexpected variable effects in the observed responses.

Table 1. Code and Level of Independent Variables Chosen for the Box–Behnken Design

Variable	Symbol		Level
	Coded	Uncoded	−1	0	1
Extraction temperature (°C)	X1	x1	40	60	80
Liquid/solid ratio (mL·g−1)	X2	x2	20	25	30
Extraction time (min)	X3	x3	60	75	90
Ultrasonic power (W)	X4	x4	60	80	100

Table 2. Box–Behnken Experimental Design and Results for the Yield of AHPs

Run	Coded variable levels				Yield of AHPs (%)
	X1	X2	X3	X4
	Extraction temperature (°C)	Liquid/solid ratio (mL·g−1)	Extraction time (min)	Ultrasonic power (W)
1	60	30	75	60	1.4217
2	70	30	75	80	1.3786
3	70	25	60	80	1.4079
4	60	25	90	100	1.531
5	60	20	75	60	1.3172
6	60	20	75	100	1.5041
7	60	25	90	60	1.4377
8	60	20	90	80	1.3258
9	50	25	75	100	1.505
10	60	25	75	80	1.4259
11	60	25	75	80	1.3513
12	60	25	75	80	1.444
13	60	30	75	100	1.3797
14	60	30	90	80	1.2054
15	50	25	90	80	1.3238
16	70	25	90	80	1.3175
17	60	25	60	60	1.3815
18	50	30	75	80	1.4629
19	50	25	75	60	1.5092
20	70	25	75	60	1.3164
21	70	25	75	100	1.3586
22	50	20	75	80	1.534
23	60	20	60	80	1.3948
24	60	25	60	100	1.3555
25	60	25	75	80	1.5212
26	60	25	75	80	1.5107
27	50	25	60	80	1.4151
28	60	30	60	80	1.5051
29	70	20	75	80	1.3979

Experimental data were well-matched to the predicted values by a quadratic polynomial model, and the regression coefficient was calculated. The nonlinear computer-generated quadratic model that was utilized in the response surface was as follows:   

(3)

where Y represents the dependent variable (actual polysaccharide yield) and β₀, β_i, β_ii, and β_ij are regression coefficients for the intercept, linear, quadratic, and interaction terms, respectively. X_i and X_j are coded values of the independent parameters, while k is equivalent to the number of tested factors (k=4).

The experimental data analysis was carried out using Design-Expert (version 8.0.5b) software. The model mentioned above was established by statistical analysis, especially regression analysis (ANOVA). The suitability of the polynomial model equation was evaluated by the coefficient of determination R². The lack of fit and the coefficient were examined using the F-test and p-value.

Comparisons between UAE and Traditional Hot Water Extraction (THWE)

Based on the results of the preliminary experiments, the pretreated dried sample powder was soaked in distilled water for 5 h and was reflux-extracted twice in a water bath at an extraction time of 75 min, liquid/solid ratio of 25 : 1, and extraction temperature of 58°C. According to the procedure described above in the “UAE” section, the AHPs were obtained and the extraction yield was determined.

UV and Fourier-Transform (FT) IR Spectral Analyses

The characteristic absorption of AHPs was identified by UV and FT-IR spectroscopy. Ultraviolet spectra were recorded with a UV-visible spectrophotometer (Beijing Purkinje General Instrument Co., Ltd., China) in the range of 200–400 nm. The FT-IR spectra were determined using an FT-IR spectrometer (Nicolet 5700, Thermo) in the range of 400–4000 cm⁻¹.

Immunomodulatory Activities of AHPs on Macrophages

RAW264.7 cells (a murine macrophage line) were purchased from the American Type Culture Collection (Bethesda, MD, U.S.A.). Cells were grown in Dulbecco’s modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS, Gibco, Grand Island, NY, U.S.A.) in a cell incubator containing 5% CO₂ at 37°C. AHPs were dissolved in DMEM at final concentration of 10.0 mg/mL and the solution was sterilized by filtration through 0.22 µm membrane filter and kept at 4°C.

The phagocytosis of RAW264.7 cells was determined with a Neutral Red Cell Proliferation and Cytotoxicology Assay Kit (Beyotime, China.¹³⁾ RAW264.7 cells were plated in a 96-well plate (5×10⁴ cells/well) in culture medium containing 10% FBS and were treated with AHPs (0 to 200 µg/mL) or lipopolysaccharide (LPS, 1 µg/mL) for 24 h. After the supernatants were discarded, 150 µL of fresh medium and 20 µL of neutral red solution were added to the cells, and the cells were incubated for an additional 2 h. Next, after washing the cells with phosphate-buffered saline solution three times, 200 µL of cell lysis solution was added to the cells. The results were recorded with a spectrophotometer at a test wavelength of 540 nm.

To determine the cytokine concentration in the cell supernatant, RAW264.7 cells were plated in a 96-well plate (5×10⁴ cells/well) in culture medium containing 10% FBS and were treated with AHPs (0 to 200 µg/mL) or LPS (1 µg/mL) for 48 h. The supernatant was collected, and the concentrations of tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6) were determined using an enzyme-linked immunosorbent assay kit (R&D Systems, eBioscience), according to the manufacturer’s instructions.

Statistical Analysis

ANOVA was used to evaluate the differences between the control and test groups. The results were described as means±standard error. Values of p>0.05, p<0.05, and p<0.01 were statistically regarded as not significant, significant, and extremely significant, respectively.

Results and Discussion

Effect of Extraction Temperature on the Polysaccharide Yield

To improve the extraction efficiency of polysaccharides using the UAE method, optimization of the extraction temperature should be included because the polysaccharide solubility increases with an increasing extraction temperature.¹⁴⁾ The effect of different temperatures on the extraction yield of polysaccharides is shown in Fig. 1a. The other extraction variables were fixed as follows: ultrasonic power of 80 W, extraction time of 75 min, and liquid/solid ratio of 20 : 1 (mL/g). When the extraction temperature was in the range of 40–60°C, the yield increased with an increasing extraction temperature, reaching a peak at 60°C. Thereafter, the polysaccharide yield slowly decreased as the temperature increased continually. Therefore, 60°C was considered as the optimal extraction temperature.

Fig. 1. Effect of Extraction Temperature (a), Liquid/Solid Ratio (b), Extraction Time (c), and Ultrasonic Power (d) on the Polysaccharide Yield

Effect of the Liquid/Solid Ratio on the Polysaccharide Yield

The liquid/solid ratio is another factor that affects the extraction yield. In this study, the effect of different liquid/solid ratios on the extraction yield of polysaccharides was investigated. The results showed that at an extraction temperature of 60°C and a fixed ultrasonic power of 80 W, the polysaccharide yield continued to increase as the proportion of water to crude material ranged from 15 : 1 to 25 : 1 (mL/g). However, the yield curve reached a plateau as the ratio increased continually (Fig. 1b). A reasonable explanation for this phenomenon is that there might be less solvent at a higher liquid/solid ratio. From an economical point of view, in this study, 25 : 1 (mL/g) was used as the best liquid/solid ratio.

Effect of Extraction Time on the Polysaccharide Yield

The extraction time is the third important factor for optimization of UAE conditions. Our study found that at a temperature of 60°C, a liquid/solid ratio of 25 : 1 (mL/g), and an ultrasonic power of 80 W, the AHPs extraction yield increased with an increasing time ranging from 15 min to 75 min. After 75 min, the yield slowly decreased (Fig. 1c). The extraction of polysaccharides from raw materials needs a certain amount of time for the liquid to penetrate into the dry powder material, dissolve it, and then spread out from the material.¹⁵⁾ However, a longer extraction time will decrease the polysaccharide yield, possibly due to the heating effect, which destroys the polysaccharide structure and leads to degradation of the extracts. So, in this study, the optimized extraction time was determined to be 75 min.

Effect of Ultrasonic Power on the Polysaccharide Yield

To a certain extent, the ultrasonic power plays a crucial role in UAE. As shown in Fig. 1d, once the other conditions were fixed (60°C, liquid/solid ratio of 25 : 1 (mL/g), and 75 min), the AHPs extraction yield increased as the ultrasonic power increased from 40 to 80 W, and it reached a peak at 80 W, beyond which the yield decreased. Ultrasonic treatment disrupts the cell walls of plants; therefore, the AHPs can dissolve faster into water. Nevertheless, degradation of the polysaccharide structure may also be caused by excessive ultrasonic power.¹⁶⁾ Thus, 80 W was selected as the optimal extraction power for the model fitting.

Statistical Analysis and Model Fitting

There were a total of 29 experimental points for optimizing the four individual parameters in the BBD. The experimental conditions and the AHPs yield based on the BBD method are shown in Table 2. To show whether the test variables and response variable are related by multiple regression analysis, the following second-order polynomial equation in terms of coded factors was used:   

(4)

where Y is the polysaccharide yield, and X₁, X₂, X₃, and X₄ are the coded variables for the extraction temperature, liquid/solid ratio, extraction time, and ultrasonic power, respectively.

A fitted quadratic polynomial model of AHPs extraction was identified by the statistical method ANOVA. As shown in Table 3, the goodness-of-fit of the regression model was evaluated by the determination coefficient (R²), while the goodness of the predictive ability of the model was determined by the predicted R value. Higher values of R² and the predicted R² indicate a better fit and the predictive ability of the model, respectively. For the quadratic regression model, the R² value was 0.9387, indicating that the model could explain 93.87% of the variation and that only 6.13% of the total variations could not be explained. As shown in Table 3, the predicted R² value was 0.9774, meaning that only 2.26% of the total variations were not illustrated by the model. Also shown, the coefficient of variation (CV) value (4.45%) was relatively low, indicating a high degree of precision and a good reliability of the experimental values. The lack-of-fit value of 0.0931 was not statistically significant relative to the pure error, implying that the model equation was able to forecast the AHPs extraction rate in the range of the experimental variables. The significance of each coefficient could be determined by the p values, which in turn may indicate the pattern of interaction between the variables. A smaller p value indicated that the corresponding coefficient was more obvious. As shown in Table 3, the F value of the model was 2.83 and its p value was 0.00307. This p value means that the regression model is very significant and that the chance that a “Model F Value” could occur due to noise is only 0.307%. According to this p value, the model mentioned above is acceptable. If the value of “Prob > F” is less than 0.01, the model terms are considered to be statistically significant.

Table 3. ANOVA for the Response Surface Quadratic Model

Source	Sum of squares	Degrees of freedom	Mean square	F Value	p Value
Model	0.16	14	0.011	2.83	0.00307
X1	7.86×10−3	1	7.86×10−3	1.99	0.1804
X2	1.26×10−5	1	1.26×10−5	3.19×10−3	0.9558
X3	0.014	1	0.014	3.47	0.0336
X4	0.085	1	0.085	3.17	0.0425
X1X2	7.92×10−4	1	7.92×10−4	0.2	0.6612
X1X3	2.03×10−7	1	2.03×10−7	5.12×10−5	0.9944
X1X4	3.11×10−3	1	3.11×10−3	0.79	0.3902
X2X3	0.013	1	0.013	3.37	0.0879
X2X4	0.013	1	0.013	3.31	0.0902
X3X4	0.261	1	0.261	2.66	0.008
X12	0.036	1	0.036	9.2	0.0089
X22	0.03	1	0.03	7.5	0.016
X32	0.056	1	0.056	14.26	0.002
X42	0.024	1	0.024	6.09	0.0271
Residual	0.055	14	3.95 ×10−3
Lack of Fit	0.05	10	5.04 ×10−3	4.1	0.0931
Pure Error	4.92×10−3	4	1.23 ×10−3
Cor Total	0.21	28

R²=0.9387; adjusted R²=0.9774; predicted R²=0.9077; CV=4.45%

The regression coefficient values of Eq. (4) are shown in Table 3. The linear coefficients (X₃ and X₄), the quadratic term coefficients (X₁², X₂², and X₃²), and the cross-product coefficient (X₃X₄) of the model were significant, with very low p values (p<0.05). Meanwhile, the other term coefficients did not significantly affect the AHPs extraction yield (p>0.05). By observing the quadratic and linear coefficients, we concluded that the order of factors that will affect the response value of the AHPs extraction yield is as follows: extraction time>ultrasonic power>extraction temperature>liquid/solid ratio. Furthermore, the regression model established in Eq. (4) could also be made in three dimensions. By this way, the relationships between the independent and dependent variables can be observed by the contour plots.

Analysis of the Response Surface

The contour lines and response surfaces were accessed by Design-Expert software. The graphical results of the AHPs yield are shown in Figs. 2 and 3, from which the effects of the variables including extraction temperature, liquid/solid ratio, ultrasonic power, and extraction time were clearly demonstrated. RSM plays a vital role in the optimization of independent variables. On the other hand, with this method, dependent variables could achieve the maximum response. The AHPs yield was revealed in pace with two uninterrupted variables; at the same time, another variable was fixed at the zero level in the contour plot and the three-dimensional response surface plot. In Figs. 2 and 3, the maximum predictive value that was displayed by the three-dimensional response surface was limited to the smallest ellipse in the contour line. When there was an ideal interaction between the two independent variables, ellipse profiles might be acquired.^17,18) Furthermore, the independent variables and maximal predictive value obtained from the figures paralleled the optimal values of the dependent variables (responses) calculated by the equations.^19,20)

Fig. 2. The Graphical Results (Three-Dimensional Response Surface) of the Polysaccharide Yield Affected by the Extraction Temperature, Liquid/Solid Ratio, Extraction Time, and Ultrasonic Power

Fig. 3. The Graphical Results (Contour Plots) of the Polysaccharide Yield Affected by the Extraction Temperature, Liquid/Solid Ratio, Extraction Time, and Ultrasonic Power

Moreover, the reciprocal interaction between the extraction temperature and the liquid/solid ratio on the extraction yield is shown in Figs. 2a and 3a, in which the extraction time and ultrasonic power were immobilized at a level of zero (Table 1). When the extraction temperature increased from 50 to 57°C, the AHPs extraction yield increased evidently, but it decreased with an increasing temperature when the extraction temperature reached beyond 57°C. The AHPs yield increased evidently with an increase of the liquid/solid ratio from 20 to 25.12; but beyond 25.12, the AHPs yield decreased slowly as the liquid/solid ratio increased.

As shown in Figs. 2b and 3b, the extraction temperature and the extraction time were varied, respectively, with the liquid/solid ratio and ultrasonic power set at a level of zero (Table 1). These figures demonstrated the reciprocal interaction of the extraction temperature and the extraction time on the AHPs extraction yield.

Figures 2c and 3c show the impact of the extraction temperature and ultrasonic power on the AHPs extraction yield. The AHPs extraction yield increased as the extraction temperature increased from 40 to 57°C and the ultrasonic power increased from 40 to 85 W. However, the impact of the extraction temperature on the AHPs yield was less, indicating that the interaction of the extraction temperature with ultrasonic power did not significantly affect the AHPs yield.

As shown in Figs. 2d and 3d, the AHPs yield increased at first and then decreased with an increase of the liquid/solid ratio and the extraction time, when the extraction temperature and ultrasonic power were fixed. Figures 2e and 3e show the impact of the liquid/solid ratio and ultrasonic power on the AHPs yield. The plots reveal that at the point where both the liquid/solid ratio and ultrasonic power have the mean values, the AHPs yield dramatically increased. Figures 2f and 3f exhibit the impacts of extraction time and ultrasonic power on the AHPs yield. The response plots demonstrate that the yield reached a small peak at the point of the average value of either the ultrasonic power or extraction time.

Optimization and Verification of the Predictive Model

The optimal AHPs extraction conditions were determined by analysis using Design-Expert 8.0.5 software. The AHPs extraction conditions included an extraction temperature of 57.81°C, an extraction time of 72.61 min, an ultrasonic power of 85.48 W, and a liquid/solid ratio of 24.84 : 1. The estimated value of 1.55% for Y in Eq. 4 was obtained under these conditions. To confirm the model accuracy, verification experiments were performed under the optimized analytical conditions. The adjusted extraction conditions were as follows: extraction temperature of 58°C, liquid/solid ratio of 25 : 1, extraction time of 75 min, and ultrasonic power of 80 W. The AHPs extraction yield in the verification experiments was 1.542±0.037% (n=3), which was much closer to the predicted one (1.55%). The accuracy of the model was evaluated by triplicate experiments under the optimized extraction conditions. Taken together, this regression model was defined to be reliable and accurate for predicting the AHPs extraction yield.

Comparisons between UAE and THWE

Compared to THWE, the application of UAE provides an improved extraction yield of AHPs (Table 4). The extraction time of UAE was significantly shortened, almost half as much as that of THWE. This is because ultrasound radiation can accelerate the extraction process and may improve the extraction of bioactive compounds. It was confirmed that ultrasound extraction is an appropriate and effective extraction technique for polysaccharides from A. giraldii HARMS.

Table 4. Comparison of Extraction Conditions and Yield of Polysaccharides from A. giraldii HARMS between Ultrasound-Assisted Extraction (UAE) and Traditional Hot Water Extraction (THWE) (n=3)

Method	Extraction temperature (°C)	Liquid/solid ratio (mL·g−1)	Extraction time (min)	Yield (%)
UAE	58	25 : 1	75	1.542±0.037
THWE	58	25 : 1	75×2	1.014±0.045

UV and FT-IR Spectral Analysis

To examine the effect of UAE or THWE on the functional groups of AHPs, UV and FT-IR spectral analyses of AHPs that were extracted by either UAE or THWE were recorded. The UV spectra of AHPs extracted by both UAE and THWE were similar, with an absorption peak at 190 nm only and without absorption peaks at 260 and 280 nm, indicating that the AHPs did not contain nucleic acids or proteins. Furthermore, as a whole, there was no significant difference between the FT-IR spectra of AHPs extracted by either UAE or THWE (Fig. 4). In the FT-IR spectra of AHPs extracted by the two different methods, a large and strong peak was detected at approximately 3389 and 3382 cm⁻¹, which are characteristic signals of the hydroxyl group due to inter- and intramolecular interactions of the polysaccharide chains.²¹⁾ The weak peaks at approximately 2918 cm⁻¹ corresponded to the characteristic absorption of the C–H antisymmetrical stretching vibration.²¹⁾ The band at around 1614 cm⁻¹ was due to the carbon–oxygen double bond (C=O) asymmetric stretching vibration absorption peak.²²⁾ In addition, the peak at 1384 cm⁻¹ was an indication of the presence of a carboxyl group with symmetrical C=O stretching vibration. The intense characteristic band at approximately 1045 cm⁻¹ in the FT-IR spectra indicated the stretching vibration of C-O-C of the glycosidic structure.²¹⁾ According to the FT-IR spectral analysis, the functional groups of polysaccharides extracted by either UAE or THWE are fundamentally identical. However, UAE is more efficient than THWE for the extraction of AHPs.

Fig. 4. FT-IR and UV Spectra of Polysaccharides from A. giraldii HARMS ((a) AHPs by UAE; (b) AHPs by THWE)

Immunomodulatory Activities of AHPs on Macrophages

Neutral Red Phagocytosis Assay

One of the most distinct features of macrophage activation is an increase in phagocytic activity.²³⁾ As shown in Fig. 5, neutral red phagocytosis of RAW264.7 cells incubated with AHPs was significantly enhanced within the experimental concentration range (from 50 to 200 µg/mL, p<0.05), compared to the control group. These results suggest that the administration of AHPs may stimulate and enhance the immune response against foreign materials such as pathogens and tumors. At the same time, there was no significant difference between AHPs extracted by UAE and by THWE.

Fig. 5. The Influence of AHPs on the Phagocytosis of RAW264.7 Cells

The results are expressed as means±standard error for two independent experiments performed in triplicate. *Significant difference from the control group was designated as p<0.05.

Cytokine Expression Analysis

Cytokines, a series of potent molecules in the immune system, can cause changes in cell proliferation, differentiation, apoptosis, inflammation, and migration.²⁴⁾ Previous studies have reported that macrophage can secrete cytokines such as- IL-6, and TNF-α, and the secretion ability can be increased by polysaccharides.^25,26) The effects of AHPs on the production of TNF-α and IL-6 were determined in the culture supernatants of macrophages cultivated for 48 h. As shown in Fig. 6A, treatment with AHPs at concentrations from 50 µg/mL to 200 µg/mL significantly improved the production of TNF-α (p<0.01), compared to the control group, in a concentration-dependent manner. The IL-6 production following the treatment of macrophages is shown in Fig. 6B. At a concentration of 200 µg/mL, AHPs treatment caused a 133% increase in IL-6 production, compared to the untreated cells. And there was no significant difference between AHPs extracted by UAE and by THWE.

Fig. 6. TNF-α and IL-6 Production by RAW264.7 Cells Stimulated by AHPs

The results are expressed as means±standard error for two independent experiments performed in triplicate. Significant difference from the control group was designated as * p<0.05 and ** p<0.01.

Progress in the Extraction Methods of Polysaccharides

In recent years, several extraction methods have been used for the extraction of polysaccharides from plants, such as THWE, UAE, enzymatic hydrolysis extraction, and microwave-assisted extraction. Polysaccharides are a class of macromolecular compounds containing many hydroxyl groups; therefore, they are soluble in water because they easily form hydrogen bonds with water molecules. On the other hand, polysaccharides will precipitate in a higher concentration of alcohol or ether due to a lower solubility. Although the combination of water extraction and alcohol precipitation is a simple method that is convenient, low-cost, nonpolluting, and suitable for industrial production, this method is rarely used currently due to its several disadvantages as mentioned in the Introduction. The other extraction methods have their own disadvantages such as strict pH control (acid or alkaline extraction), high cost (enzymatic method), as well as energy consumption and structural damage to the polysaccharide (microwave-assisted extraction). It is worth mentioning that UAE promotes the dissolution of the active ingredients within the cell via tissue deformation and rupture, thus releasing the cell contents. The UAE method has been widely used for the extraction of plant polysaccharides due to its high efficiency. However, it still has the disadvantage of requiring a high power of ultrasonication that can destroy the polysaccharide structure, resulting in a difficult separation.²⁷⁾ To avoid this issue, it is necessary to optimize the UAE conditions by RSM, an effective statistical technique that can reduce the number of experimental trials required for the evaluation of multiple factors and their interactions. In our study, FT-IR spectral analysis confirmed that optimization of the UAE conditions by RSM could avoid damage to the polysaccharide structure in the process of AHPs extraction. The use of RSM has been recently reported for optimization of the processing parameters or extraction process variables and the interactions of these variables.^28–32) As the best type of RSM, the BBD is the most efficient and easiest to arrange and interpret the results, compared with other methods.^33,34) In the present study, we used the BBD method to optimize the conditions of UAE for AHPs extraction. All of the analytical methods, especially the BBD used in this study, will provide a reference for improvement of the extraction yield of polysaccharides from plants in industrial settings.

Conclusion

To improve the AHPs extraction yield and maintain the integrity of the polysaccharide structure, we investigated the UAE conditions on the AHPs extraction yield using the RSM method for optimization of the experimental parameters, including extraction temperature, liquid/solid ratio, extraction time, and ultrasonic power. Based on single-factor experiments, using the BBD, a quadratic polynomial parameter mathematical model was established for the AHPs extraction process. Using ANOVA, this model was found to be reasonable and reliable for predicting responses. In this model, the optimized conditions (extraction temperature of 58°C, extraction time of 73 min, ultrasonic power of 85 W, and liquid/solid ratio of 25 : 1) were selected according to the regression equation. Under these optimized conditions, we reached the maximum experimental value (1.54%) of the AHPs extraction yield, which was very close to the predicted one (1.55%). FT-IR and UV spectral analyses confirmed that optimization of the UAE conditions by RSM could avoid damage to the polysaccharide structure in the process of AHPs extraction. The enhanced phagocytosis and increased cytokine secretion in macrophages treated with AHPs support that AHPs extracted by the optimized UAE technology still maintain their biological activity. Further evaluation of the immunomodulatory functions of AHPs is in progress in our laboratory.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (No. 81303198, No. 81471048) and the Natural Science Foundation of Shandong Province (No. ZR2011HQ044, No. ZR2015HM028).

Conflict of Interest

The authors declare no conflict of interest.

References

• 1) Drug Administration Department of the Ministry of health CBMI. “Modern practical Materia Medical” (Part 1), People’s Medical Publishing House, Beijing, 1997, p. 496.
• 2) Zhou R. H., “Resources science of traditional Chinese medicine. Chinese medical science and Technology Press,” Beijing, 1993, p. 345.
• 3) Liu W. G., Xu H. W., Wu S. Q., Jouranl of Shandong University of TCM, 28, 71–73 (2004).
• 4) Ma J. W., Qiao Z. Y., Xiang X., Carbohydr. Polym., 84, 1061–1068 (2011).
• 5) Mardones C., Von Baer D., Hidalgo A., Contreras A., Sepúlveda C., J. Sep. Sci., 31, 1124–1129 (2008).
• 6) Orozco-Solano M., Ruiz-Jiménez J., Castro L. D., J. Chromatogr. A, 1217, 1227–1235 (2010).
• 7) Tadeo J. L., Sanchez-Brunete C., Albero B., García-Valcárcel A. I., J. Chromatogr. A, 1217, 2415–2440 (2010).
• 8) Myers R. H., Montgomery D. C., “Response Surface Methodology: Process and Product in Optimization Using Designed Experiments.” J. Wiley, Hoboken, NJ, 1995.
• 9) Ferreira S. L., Bruns R. E., Ferreira H. S., Matos G. D., David J. M., Brandão G. C., da Silva E. G., Portugal L. A., dos Reis P. S., Souza A. S., dos Santos W. N., Anal. Chim. Acta, 597, 179–186 (2007).
• 10) Liu H., Du X., Yuan Q., Zhu L., Phytochem. Anal., 20, 475–483 (2009).
• 11) Sevag M., Lackman D. B., Smolens J., J. Biol. Chem., 124, 425–436 (1938).
• 12) Staub A. M., Methods Carbohydr. Chem., 5, 5–6 (1965).
• 13) Chu B. S., Quek S. Y., Baharin B. S., Food Chem., 80, 295–302 (2003).
• 14) Li W., Wang Z., Sun Y. S., Chen L., Han L. K., Zheng Y. N., Phytochem. Anal., 22, 313–321 (2011).
• 15) Wang Y. J., Zhong C., Mao J. W., Fan M., Wu X., Carbohydr. Polym., 77, 713–717 (2009).
• 16) Ying Z., Han X., Li J., Food Chem., 127, 1273–1279 (2011).
• 17) Muralidhar R. V., Chirumamila R. R., Marchant R., Nigam P., Biochem. Eng. J., 9, 17–23 (2001).
• 18) Sun Y. X., Liu J. C., Kennedy J. F., Carbohydr. Polym., 80, 949–953 (2010).
• 19) Ballard T. S., Mallikarjunan P., Zhou K., O’Keefe S., Food Chem., 120, 1185–1192 (2010).
• 20) Yang B., Liu X., Gao Y. X., Innov. Food Sci. Emerg. Technol., 10, 610–615 (2009).
• 21) Xu Y., Cai F., Yu Z., Zhang L., Li X., Yang Y., Liu G., Food Chem., 194, 650–658 (2016).
• 22) Černá M., Barros A. S., Nunes A., Rocha S. M., Delgadillo I., Čopíková, J., Coimbra M.A., Carbohydr. Polym., 51, 383–389 (2003).
• 23) Chen Y., Xie M. Y., Nie S. P., Li C., Wang Y. X., Food Chem., 107, 231–241 (2008).
• 24) Moradali M. F., Mostafavi H., Ghods S., Hedjaroude G. A., Int. Immunopharmacol., 7, 701–724 (2007).
• 25) Bai Y., Zhang P., Chen G., Cao J., Huang T., Chen K., Int. Immunopharmacol., 12, 611–617 (2012).
• 26) Zhang C. X., Dai Z. R., Food Chem. Toxicol., 49, 2961–2967 (2011).
• 27) Zhou C. S., Yu X. J., Zhang Y. Z., He R., Ma H., Carbohydr. Polym., 87, 2046–2051 (2012).
• 28) Chen Y. Y., Gu X. Y., Huang S. Q., Li J., Wang X., Tang J., Int. J. Biol. Macromol., 46, 429–435 (2010).
• 29) Huang S. Q., Ning Z. X., Int. J. Biol. Macromol., 47, 336–341 (2010).
• 30) Tian Y. T., Zeng H. L., Xu Z. B., Zheng B., Lin Y., Gan C., Lo Y. M., Carbohydr. Polym., 88, 522–529 (2012).
• 31) Zhong K., Wang Q., Carbohydr. Polym., 80, 19–25 (2010).
• 32) Zhang B. Z., Yan P. S., Chen H., He J., Carbohydr. Polym., 87, 2569–2575 (2012).
• 33) Fan T., Hu J., Fu L., Zhang L., Carbohydr. Polym., 115, 701–706 (2015).
• 34) Kurd F., Samavati V., Int. J. Biol. Macromol., 74, 498–506 (2015).