新疆一枝蒿多糖的提取、純化及其結構分析

王旭輝，徐 鑫，王 卉，葉 凱，涂振東，晁群芳

(1.新疆大學生命科學與技術學院，烏魯木齊 830046；2.新疆農業(yè)科學院生物質能源研究所，烏魯木齊 830091；3. 新疆農業(yè)科學院糧食作物研究所，烏魯木齊 830091)

0 引 言

【Research significance】ArtemisiarupestrisL., belonging to theCompositaefamily, is a well-known traditional Uygur medicinal herb of Xinjiang province in China, and is mainly distributed in Kazakhstan, Mongolia and Russia. This plant is commonly used for the treatment of many diseases, such as detoxication, anti-hypersusceptibility, antitumor, antibacterial, antivirus and protecting liver, and it contains many kinds of bioactive chemicals, primarily includingavonoids, terpenoids, alkaloid, rupestonic acid[1]. The recent pharmacology research indicated that it has the functions of antitumor, antimicrobial, antioxidant, and improving cellular immunity.【Previous research progress】Polysaccharides from plant, epiphyte and animals extracts are an interesting source of additives for several industries, in particular food and drug industry[2]. They play an important role in the growth and development of living organisms and have been widely studied in recent years due to their unique biological, chemical and physical properties[3]. Published literatures have indicated that plant polysaccharides in general have strong antioxidant activities and can be explored as novel potential antioxidants[4-6]. Hot water technology is the main and most conventional extraction method for polysaccharides mentioned in previous studies[7, 8]. However, it is usually associated with longer extraction time and higher temperature but lower extraction efciency. Therefore, it is desirable tond an effective and economical method for bioactive compounds extraction. Ultrasonic has been used to increase extraction yield of bioactive substances from natural products, which is mainly attributed to disruption of cell walls, particle-size reduction, and enhanced mass transfer to the cell contents as a result of cavitation bubble collapse[9-12]. It offers high reproducibility at shorter times, simplied manipulation, and lowered energy input, as well as solvent consumption[13]. Optimizing refers to improving the performance of a system, a process, or a product in order to obtain the maximum benet from it. When many factors and interactions affect desired response, response surface methodology (RSM) is an effective tool for optimizing the process, which was originally developed by Box and Wilson in the 50s[14]. The main advantage of RSM is the reduced number of experimental trials needed to evaluate multiple parameters and their interactions. Therefore, it is less laborious and time-consuming than other approaches required to optimize a process[15], and it has successfully been applied in order to optimize the conditions in food and pharmaceutical research[16-20]. 【Entry point】However, so far there is not any information published about the optimization, purication and structural characterization of polysaccharides fromA.rupestrisL.. For that reason, extraction of polysaccharides is an important step for its application or further research and development. 【The key problems intended to be solved】Therefore, the aim of this work was mainly to report on the extraction and purication of polysaccharides fromA.rupestrisL.. Firstly, single-factor experimental designs (ultrasonic power, ratio of water to raw material, ultrasonic time and extracting times) were carried out before RSM experiments. Secondly, three factors (ultrasonic power, ratio of water to raw material and ultrasonic time) were chosen based on single-factor designs for further optimization by employing a three-level, three-variable BBD from RSM. Furthermore, after purication by gelltration chromatography, ultraviolet spectrum (UV) and Fourier transform-infrared spectroscopy (FT-IR) were employed to identify the chemical structure.

1 Materials and methods

1.1 Materials

DriedA.rupestrisL. was obtained from Urumqi, Xinjiang province, China in September, 2010, the whole plants were cut into smaller pieces and further ground into ane powder in a high speed disintegrator (160C, Fei Da Chinese Traditional Medicine Machine Co., Ltd, Zhejiang, China), and passed through a 40 mesh sieve. The materials were stored at room temperature in a desiccator until use(less than two weeks). All other reagents used in this study were of analytical grade.

1.2 Ultrasonic extraction

The extraction of ARP was done according to the reported method with some modications[21-23]. Briey, the extraction process was performed using an ultrasonic cell disrupter (JY92-2D, Ningbo Xinzhi Biological Technology Co., Ltd., Zhejiang, China), with different ultrasonic power, ratios of water to raw material and the ultrasonic time.

DriedA.rupestrisL. powder was defatted in a Soxhlet apparatus with Petroleum ether (boiling point: 60-90℃) and then were extracted for three times with 80% ethanol at 60℃ and 2 h each time to defat and remove some colored materials, oligosaccharides, and some small molecule materials[24, 25]. The pretreated samples were separated from the organic solvent by centrifugation (3,500 g for 15 min). Each dried pretreated sample was extracted by water in a designed ultrasonic power, ratios of water to raw material, the ultrasonic time and number. The water extraction solutions were separated from insoluble residue through the nylon cloth (Pore diameter: 38 nm), was then concentrated to 25 mL with a rotary evaporator (RE-5299, Gongyi Yuhua Instrument Co., Ltd., He’nan, China) at 55℃ under vacuum[6]. The proteins in the extract were removed using the Sevag reagent[26]. After removal of the Sevag reagent, the supernatant was precipitated by the addition of 95% ethanol to anal concentration of 80% (v/v) to obtain polysaccharides, and kept overnight at 4℃. Then the solution was centrifuged at 4,500 rpm/min for 20 min, washed three times with dehydrated ethanol, and then the precipitate was solubilized in deionized water and lyophilized by vacuum freeze drying machine (D-37520, Ostevode, telefon, Germany). The content of the polysaccharides was measured by anthrone-sulfuric method. The percentage polysaccharides extraction yield (%) is calculated as the polysaccharides content of extraction divided by dried sample weight. All experiments were performed at least in duplicate.

1.3 Experimental design and statistical analysis

To explore the effects of independent variables on the response within the range of investigation, proper ranges of ultrasonic power, ratio of water to raw material, ultrasonic time, and extraction number were preliminarily determined. A three-level-three-factor, Box-Behnken factorial design (BBD) employed in this optimization study. Based on the investigations on single-factor experiments, ultrasonic powers (X1), ratio of water to raw material (X2), ultrasonic time (X3) were chosen for independent variables to be optimized for the extraction yield of ARP. Yield of polysaccharides (Y) was taken as the response of the design experiments. For statistical calculation, the variables were coded according to

(1)

Wherexiwas a coded value of the variable;Xiwas the actual value of variable;X0was the actual value of theXion the center point; and △Xwas the step change value. The range of design factors and their levels is presented in Table 1, which was based on the results of preliminary experiments. As seen from Table 2, the whole design consisted of 17 experimental points carried out in random order. Five replicates at the center of the design were used for the estimation of a pure error sum of squares. The response value in each trial was average of duplicates. Table 1, Table 2

Table 1 Response surface methodology design factors and levels

Independent variablesFactor level-101X1: Ultrasonic power (w)600700800X2: Ratio of water to raw material303540X3: Ultrasonic time (min)202530

Table 2 Box-Behnken design and the response value for yield of polysaccharides

RunX1Ultrasonic power(w)X2Ratio of water to raw materialX3Ultrasonic time (min)YPolysaccharides yield (%)1-1-102.6120-1-13.3230003.9240003.9650003.9560-113.3470113.5781103.2890003.93100003.931101-13.5112-1102.721310-13.3114-10-12.73151-102.8516-1012.89171013.25

Based on the experimental data, regression analysis was performed and wastted into an empirical second-order polynomial model:

(2)

WhereYis the response function,A0was constant.Ai,AiiandAijwere the coefcients of the linear, quadratic and interactive terms, respectively. AndXiandXjrepresented the coded independent variables. The coefcients of the second polynomial model and the responses obtained from each set of experimental design were subjected to multiple nonlinear regressions using software Design-Expert 8.0.5b (Trial Version, State-Ease Inc., Minneapolis, MN, USA). Thetness of the polynomial model equation was expressed by the coefcient of determinationR2, and its statistical signicance was checked by F-test at a probability (P) of 0.05. The signicances of the regression coefcients were also tested by F-test. Additional conrmation experiments were subsequently conducted to verify the validity of the experimental design.

1.4 Purication of the polysaccharide

The crude polysaccharide was re-dissolved in 5 ml distilled water,ltered through 4.5 × 10-4mmlters and applied to a DEAE-52 cellulose column (3.0 × 60 cm) equilibrated with distilled water. The polysaccharide was fractionated and eluted with distilled water and different concentrations of stepwise NaCl solution (0,0.1,0.2 ,0.3 and 0.4 M NaCl) at aow rate of 2.0 mL/min. The elutes were concentrated to obtain the main fractions, which were then fractionated by size-exclusion chromatography on a Sephadex G-100 column (2.5 × 40 cm) eluted with 0.2 M NaCl at aow rate of 0.5 mL/min. The relevant fractions were collected, concentrated, dialyzed and lyophilized.

1.5 Ultraviolet analysis

ARP-2 was dissolved and diluted to 2 mg/mL respectively, and the solutions of the polysaccharide fraction were scanned from 200 to 400 nm with a UV-VIS-NIR spectrophotometer (UV3600, Shimadzu Corporation, Japan).

1.6 FT-IR spectrometric analysis

The FT-IR spectra were recorded using the KBr-disk method with a Nicolet Fourier transform infrared (FTIR) spectrometer (EQINOX 55, BRUKER Corporation, Germany) in the range 400-4,000 cm-1.

1.7 Statistical analysis

Comparison of means was performed by one-way analysis of variance (ANOVA) followed by LSD (Least significant difference method). Statistical analyses (P< 0.05) were performed using SPSS 16.0 software. Design Expert 8.0.5b (Trial Version, State-Ease Inc., Minneapolis, MN, USA) was employed for experimental design, analysis of variance (ANOVA), and model building. All analyses were performed in triplicate.

2 Results and discussion

2.1 Effect of different ultrasonic power on extraction yield of ARP

The effect of different ultrasonic power on extraction yield of ARP is shown in Fig. 1(a). Extraction was carried out at different ultrasonic power (200-1,000 W) conditions while other extraction parameters were as follows: ratios of water to raw material of 30, ultrasonic time of 25 min, and number of extraction of 2. The extraction yield of ARP signicantly increased from 2.08% to 3.71% as ultrasonic power increased from 200 to 1,000 W. It can be found that signicant differences were existing among 200 W, 400 W and 600 W, and among 200 W, 400 W, 600 W and 1,000 W (P< 0.05), but there was no signicant difference between 600 W and 800 W, and among 800 W and 1,000 W (P> 0.05). Although the extraction yield of ARP was also high at 1,000 W, increasing ultrasonic power will bring about the increase in cost for the extraction process from the industrialization point of view. Therefore, ultrasonic power range of 600-800 W was considered to be optimal in the present experiment. Fig.1

Fig.1 Effort of ultrasonic power (a), ratio of water to raw material (b), ultrasonic time(c) and number of extraction (d) on the yield of polysaccharides.Values have no significant statistical difference at 0.05 with the same superscript letters

2.2 Effect of different ratio of water to raw material on extraction yield of ARP

The effect of different ratio of water to raw material on extraction yield of ARP is shown in Fig. 1(b). Extraction was carried out at different ratio of water to material (10-50) conditions while other extraction parameters were as follows: ultrasonic power of 700 W, ultrasonic time of 25 min, and number of extraction of 2. The extraction yield of ARP signicantly increased from 1.92% to 3.86%. Results indicated that signicant differences were existing among 10, 20, 30 and 40, and among 10, 20, 30 and 50 (P< 0.05), but there was no signicant difference between 30 and 40 (P>0.05). Therefore, ratio of water to raw material range of 30-40 was considered to be optimal in the present experiment.

2.3 Effect of different ultrasonic time on extraction yield of ARP

Ultrasonic time is a factor that would inuence the extraction efciency and selectivity of theuid. This might be due to the time requirement of the exposure ofA.rupestrisL. to the release medium where the liquid penetrated into the driedA.rupestrisL. powdered, dissolved it and subsequently diffused out from the material. A longer ultrasonic time presents a positive effect on the extraction yield of ARP. The effect of different ultrasonic time on the extraction yield of ARP is shown in Fig. 1(c). Extraction was carried out at different ultrasonic time conditions while other extraction variables were set as follows: ultrasonic power of 700 W, numbers of extraction of 2, and ratio of water to raw material of 30. When ultrasonic time varied from 10 to 30 min, the variance of extraction yield was relatively rapid, and the extraction yield of ARP reached a maximum at 20-30 min, and then no longer changed as the extraction proceeded. Results showed that signicant differences were existing among 10 min, 15 min, 20 min and 25 min, and among 10 min, 15 min, 20 min and 30 min (P< 0.05), but there was no signicant difference between 25 min and 30 min (P> 0.05). This indicated that ultrasonic time of 25 min was sufcient to obtain the polysaccharide production. Thus, extraction of 20-30 min was favorable for producing the polysaccharides.

2.4 Effect of different number of extraction on extraction yield of ARP

The effect of number of extraction on extraction yield of ARP is shown in Fig. 1(d). While other extraction parameters weretted as follows: ultrasonic power 700 W, ultrasonic time 25 min, and ratio of water to raw material 30. From Fig. 1 (d), it can be found that there was an increasing trend in the yield of ARP accompanying the increase of extracting times, but there was not signicant difference (P>0.05) between 2 times and 3 times. Taking the yield and processing cost into consideration, 2 times was sufcient for the extraction of ARP. Thus, 2 times was selected as the extracting times in the next experiments.

According to the single-parameter study, ultrasonic power of 600-800 W, ratio of water to raw material of 30-40 and ultrasonic time of 20-30 min for RSM experiments were adopted.

2.5 Optimization of the procedure

2.5.1 Statistical analysis and the modeltting

The value of responses (the extraction yield of ARP) at different experimental combination for coded variables is given in Table 2. The extraction yield of ARP ranged from 2.61% to 3.96%. By employing multiple regression analysis on the experimental data, the predicted response for the yield of ARP can be obtained by the following second-order polynomial equation:

Y=3.94+0.22×X1+0.12×X2+0.023×X3+0.080×X1×X2-0.055×X1×X3+0.010×X2×X3-0.73×X1×X1-0.34×X2×X2-0.16×X3×X3

(3)

Table 3 Analysis of variance for thetted quadratic polynomial model of extraction of polysaccharides

ItemStd.devMeanC.V.%PressR2R2AdjR2PredAdeq precisionValue0.0283.360.830.0700.998 50.996 60.980 562.574

The Lack of Fit is an indication of the failure for a model representing the experimental data at which points were not included in the regression or variations in the models cannot be accounted for random error[27]. If there is a signicant Lack of Fit which could be indicated by a low probability value, the response predictor is discarded. The Lack of Fit did not result in a signicant P-value for selected variables, meaning that these models were sufciently accurate for predicting the relevant responses. The Lack of Fit F-value of 5.31 implies that the Lack of Fit is signicant. There is only a 7.03% chance that a “Lack of Fit F-value” could occur due to noise, which indicated that the model equation was adequate for predicting the yield of ARP under any combination of values of the variables. Table 4

Table 4 Regression coefcient estimation and their signicance test for the quadratic polynomial model

VariablesSun of squaresDFMean squareF valueP value prob.>FModel3.6190.04521.25<0.000 1X10.3810.38492.41<0.000 1X20.1210.12149.89<0.000 1X34.05×10314.05×1035.270.055 4X1×X12.2512.252 931.44<0.000 1X2×X20.4910.49638.9<0.000 1X3×X30.1110.11142.89<0.000 1X1×X20.02610.02633.310.000 7X1×X30.01210.01215.740.005 4X2×X34×10414×1040.520.494 0Residual5.38×10377.686×104--Lack of fit4.3×10331.433×1035.310.070 3Pure error1.08×10342.7×104--Correlation total3.6116---

The model was found to be adequate for prediction within the range of experimental variables. The regression coefcient values of Eq. (3) were listed in Table 4. The P-values were used as a tool to check the signicance of each coefcient, which in turn may indicate the pattern of the interactions between the variables. The smaller the value of P was, the more signicant the corresponding coefcient was[28]. It can be seen from this table that the linear coefcients (X1,X2), a quadratic term coefcient (X12,X22,X32) and cross product coefcients (X1×X2) were signicant, with very small P-values (P< 0.05). The other term coefcients were not signicant (P> 0.05). Therefore,X1,X2,X12,X22,X32andX1×X2were important factors in the extraction process of the polysaccharides. The full modellled Eq. (3) was made the 3-D response surface plots and contour plots to predict the relationships between the independent and dependent variables.

2.5.2 Optimization of extraction conditions of ARP

The response surface curves were plotted using Design-Expert to study the effects of parameters and their interactions on polysaccharides yield. The 3-D response surface plots were drawn to illustrate the main and interactive effects of the independent variables on the dependent one. They provide a method to visualize the relationship between responses and experimental levels of each variable and the type of interactions between two test variables. These types of plot (circular or elliptical) indicated the mutual interactions between the variables are signicant or not. Circular contour plot indicates that the interactions between the corresponding variables are negligible, while elliptical contour plot indicates that the interactions between the corresponding variables are signicant[29].In the present study, the 3-D response surface plots and the contour plots, as presented in Figs. 2. It is clear that the yield of ARP is sensitive to minor alterations of the test variables (ultrasonic power, ratio of water to raw material and ultrasonic time). Through these 3-D response surface plots and their respective contour plots, it is very easy and convenient to understand the interactions between two variables and to locate their optimum ranges.

The 3-D response surface plot and the contour plot in Figs. 2 (a) and (d), which give the extraction yield of ARP as a function of ultrasonic power and ratio of water to raw material atxed ultrasonic time (0 level), indicated that the extraction yield of ARP increased with increase of ultrasonic power from 600 to 715.78 W, then decreased slightly from 715.78 to 800 W, and increased rapidly with increase of ratio of water to raw material from 30 to 35.98, then dropped from 35.98 to 40. Figs. 2 (b) and (e) shows the 3-D response surface plot and the contour plot at varying ultrasonic power and ultrasonic time atxed ratio of water to raw material (0 level). From twogures, it indicated that the extraction yield of ARP increased with increase of ultrasonic power from 600 to 715.78 W, then decreased from 715.78 to 800 W, and the extraction yield of ARP was found to increase rapidly with increase of ultrasonic time from 20 to 25.25 min, but beyond 25.25 min, the extraction yield of ARP reached the plateau region where the yield was maximized and did not further increase the yield. The extraction yield of ARP affected by different ultrasonic time and ratio of water to raw material was seen in Figs. 2 (c) and (f), when ultrasonic power wasxed at 0 level. It can be seen that the extraction yield of ARP increased with the increase of ratio of water to raw material from 30 to 35.98, then further increased with the increase of ultrasonic time, and reached the maximum value when ultrasonic time at 25.25 min, and beyond this level, the extraction yield of ARP did not further increase.

From Figs. 2, it can be concluded that the optimum extraction conditions of polysaccharides fromA.rupestrisL. are ultrasonic power 715.78 W, ratio of water to material 35.98 and ultrasonic time 25.25 min. The theoretical extraction yield of ARP that was predicted under the above conditions was 3.97%.

Fig.2

The suitability of the model equations for predicting optimum response values was tested under the conditions: ultrasonic power 715.78 W, ratio of water to material 35.98 and ultrasonic time 25.25 min. A predicted value of 3.97% was obtained for yield of polysaccharides under the optimal conditions. In order to facilitate the practical extraction process of ARP, the optimal conditions were modied as follows: ultrasonic power 716 W, ratio of water to material 36 and ultrasonic time 25 min. A predicted value of 3.94% was obtained under the modied conditions. The modied conditions were used to validate the suitability of thetted model equation for accurately predicting the responses values. The results showed that the actual values of polysaccharides yield were 3.92%± 0.065% under the modied conditions (Table 5), which were in agreement with the predict values signicantly (P>0.05). Thus, the model can be used to optimize the process of polysaccharides extraction fromA.rupestrisL.. Table 5

Fig.2 Response surface plots (a-c) and contour plots (d-f) showing the effects of the variables on the yield of polysaccharide

Table 5 Predicted and experimental values of the responses at optimum conditions

Ultrasonic power(W)Ratio of water to raw material(mL/g)Extraction time(min)Extraction yield of ARP(%)(Experimental)Extraction yield of ARP(%)(Predicted)Optimum conditions715.7835.9825.253.963.97Modi ed conditions71636253.923.94

The representative values were obtained from three independent experiments, and the results are presented as mean value ± SD

2.6 Isolation and purication of polysaccharides

GroundA.rupestrisL. was reuxed with ethanol to deactivate the endogenous enzymes and remove some soluble materials, including free sugars, amino acids and some phenols. Then the dried ethanol-extracted residue was extracted with distilled water at 80℃. After dialysis and precipitation with ethanol, a crude water-soluble polysaccharide (ARP) was obtained as a brownish powder. The ARP was separated and puried by DEAE-52 cellulose column. Four fractions of ARP-1, ARP-2, ARP-3 and ARP-4 were eluted. ARP-2 was further puried through Sephadex G-100 column (Fig. 3(a).and (b).). Fig.3

(a) Elution prole of ARP on a DEAE-cellulose ion-exchange column (at 2 mL/min mobile phaseow rate); (b) Elution prole of ARP-2 (from a) on a Sephadex G-100 column (eluted with 0.2 M mobile phase at 0.5 mL/min). Absorbance at 620 nm (A620) represents the total carbohydrate content (relative) determined by the anthrone-sulfuric method. Spectral Analysis of ARP-2:(c)The UV spectra of ARP-2;(d)FT-IR analysis of ARP-2

Fig.3 Fractionation of ARP by column chromatography

2.7 Ultraviolet analysis

ARP-2 had no absorption at 280 and 260 nm in the UV spectrum (Fig.3(c).), indicating the absence of protein and nucleic acid.

2.8 FTIR analysis

The FTIR spectrum of the polysaccharide in the wavelength range of 4,000-400 cm-1was shown in (Fig.3(d).). In the spectrum of polysaccharide, the bands in the region of 3,410 cm-1were due to the hydroxyl stretching vibration of the polysaccharides. The bands in the region of 2,938-1were due to C-H stretching vibration[5,31]. The characteristic absorptions at 917 cm-1in the IR spectra indicated that was β-congurations. Stronger bands occurring between 1,742 cm-1and 1,611 cm-1are derived from the ester carbonyl (-COOR) groups and carboxylate ion stretching band (-COO-)[30].

3 Conclusions

Compared to conventional extraction techniques, ultrasonic technology improves the extraction efciency within a shortened period of time. Ultrasonic technology was performed for the polysaccharides extraction fromA.rupestrisL. in order to increase the yield extraction. Based on the single-factor experiments, Box-Behnken design (BBD) was used to optimizes process parameters and improves polysaccharide extraction. ANOVA showed that the effects of the variables are signicant and quadratic models are obtained for predicting responses. The optimum set of the independent variables was obtained graphically in order to obtain the desired levels of polysaccharides extraction. The optimum extraction conditions were modified as follows: ultrasonic power 716 W, ratio of water to raw material 36 and ultrasonic time 25 min. Under these optimal conditions, the mean polysaccharides extraction yield was 3.92%± 0.065%, which corresponded well with the predicted yield of 3.94%. Then, the crude polysaccharides were puried byltration, DEAE-52 cellulose column and Sephadex G-100 size-exclusion chromatography in that order. A main fraction, ARP-2 was obtained through the extraction and purication steps. Finally, The absorption spectra of the polysaccharides showed that the polysaccharides-related absorption peaks at 917, 1,100, 1,147, 1,332, 1,423, 1,611, 1,742, 2,939 and 3,410 cm-1. This study could become the basis for later research according to the optimization predictions of the extraction. Further study should be carried out to study bioactivity in vitro and in vivo, with the purpose of applying the polysaccharides as a potential natural antioxidant functional ingredient in the food industry.