TY - JOUR
AU - Zhong,, Fangli
AB - Abstract A potential method called microwave-assisted aqueous two-phase extraction (MA-ATPE) was developed for concurrent extraction and purification of gentiopicroside from Gentiana scabra Bunge. Formation characteristics of aqueous two-phase system (ATPS) composed of ethanol and 25 kinds of salts were investigated; K2HPO4 (w/w, 21.71%) and ethanol (w/w, 40.72%) were determined to be the optimal compositions of ATPS. Response surface methodology based on Box–Behnken design was used to investigate the extraction conditions, the optimal parameters were summarized as follows: 80°C of extraction temperature, 31 s of extraction time, 11:1 (mL/g) of liquid-to-solid ratio, 100 meshes of particle size and 806 W of microwave power. Under these conditions, the extraction yield of gentiopicroside was 65.32 ± 0.24 mg/g with a recovery of 96.51%. Compared with other four methods, the purity of gentiopicroside in the crude extracts reached 17.16 ± 0.25%, which was significantly higher than that of smashing tissue extraction, microwave assisted-extraction, ultrasonic-assisted extraction and heat reflux extraction, respectively. In addition, the phase-forming salt can be recyclable. Therefore, MA-ATPE was an excellent and alternative technique to the conventional extraction approaches of gentiopicroside. Introduction The genus Gentiana of the family gentianaceae comprises over 400 species that are widely distributed in alpine habitats of temperate regions of Asia, Europe and America. Some species also thrive in Northwest Africa, Eastern Australia and New Zealand (1). Gentiana scabra Bunge (G. scabra) listed in the Chinese Pharmacopoeia as “Longdan” (2) is one of the most famous traditional Chinese herbal medicines; it had been reported to be effective in treating liver dysfunction, anorexia, inflammation and indigestions (3). Phytochemical studies have shown G. scabra contained various chemical constituents such as secoiridoids, gentiopicroside, triterpenoids, flavonoids, essential oil, alkaloids and polysaccharides, etc. (4–5). Among these compounds, gentiopicroside (Fig. 1) have been proved to be the primary chemical constituents for its noticeable pharmacological actions in Gentiana plants (6–8), and thus the content of gentiopicroside had been used as a main quality indicator for evaluating different Gentiana species. Moreover, the huge demand of domestic and international market for gentiopicroside led to the further investigation aimed at the improvement of extraction and purification techniques (9–10). The extraction of gentiopicroside were usually conducted by classical methods such as heat reflux extraction (HRE) and solvent soaking extraction (11), which suffered from some limitations and shortcomings such as lengthy extraction time, high solvent consumption, low extraction yield, toxic solvent residuals and degradation of the unstable product, etc. (12–13). In the last few years, various novel methods including ultrasonic-assisted extraction (UAE) (14) and smashing tissue extraction (STE) (15) had been developed to extract gentiopicroside, the extraction time and energy consumption was reduced, and extraction yield was also increased, but the purity of gentiopicroside in the crude extracts was lower. In addition, microwave assisted-extraction (MAE), a field-intensified technique, also have been extensively applied for the extraction of bioactivity components from plants matrix due to its unique advantages of quick heating, low quantity solvent used, decreased energy consumption and pollution (13, 16). Aqueous two-phase extraction (ATPE) was first introduced in the mid-1950s by Albertson in order to achieve the preliminary separation of biomolecules (17). Nowadays, it has been widely used in the separation and purification of natural compounds (12–13, 18). The enrichment performance of ATPE mainly depended on the biphasic extraction capacity and selectivity of aqueous two-phase system (ATPS), in which target constituents and impurities could be extracted into the top and bottom phase, respectively, and thus the selection of phase-forming components played an important role in ATPS (19–20). According to the previous reports, ATPS can form spontaneously by mixing two different hydrophilic polymers (e.g., polyethylene glycol (PEG) and dextran (21–23)), a hydrophilic polymer and a salt (e.g., PEG and NaSO4 or phosphate (21, 24)) or a non-ionic surfactant micellar system (e.g., TritonX-114 ((25)) with water. However, ATPS composed of polymers or surfactants can often not form transparent solutions and are too viscous for the subsequent process. Our recent experimental investigations (26) have proved that ATPS based on short chain alcohols and salt was a promising liquid–liquid extraction technique. In comparison with pure organic solvents, it possessed noticeable advantages such as low viscosity, low cost, easy demixing, easy recovery of phase-forming components, relatively environment-friendly. Similar results can also be found in the latest literatures on enrichment of flavonoids (27) and chlorogenic acid (28). Partition behavior of target components and its co-exist impurities is different in ATPS due to physicochemical properties of molecules, and thus ATPS have been employed to extract and enrichment various target compounds from natural plants. Nowadays, ethanol/salt ATPS have gained increasing attention in extracting biological components from medicinal plants (18). During the extraction process of MAE, microwave energy can lead to the disruption of weak bonds existed in target components. Hence, the penetration of solvent into matrix is enhanced and release of bioactive constituents is accelerated (29). Nevertheless, extraction capability of MAE is mainly depended on the dielectric susceptibility of extraction agent and solid material, and thus co-exist impurities can also be extracted during the extraction process, resulting in complicated procedures as for sample pretreatment before qualitative or quantitative analyze (13). As microwave energy can be absorbed more strong when a great deal of water was contained in the multiphase extraction agent (ATPS), a mass of thermal effect could be produced, leading to significant increase of temperature, enhancement of penetration of extraction agent into plant tissue and easy release of target components into the surrounding extraction solvent. Therefore, with the integration of MAE and ATPS, microwave-assisted aqueous two-phase extraction (MA-ATPE) may be a promising, powerful and innovative technique; it combined field-intensified effect with ATPS in one step aimed at concurrent extraction and enrichment of target compound. However, the related study for extracting and purifying gentiopicroside using MA-ATPE was very scarce until now. In the present study, an effective, cheap and novel method have been developed for the concurrent extraction and purification of gentiopicroside from G. scabra by integrating MAE and ATPE in one-step procedure. After the compositions of ATPS (25 kinds of salts and ethanol) and several influential parameters including solvent/solid ratio, microwave irradiation power, extraction temperature, extraction time and particle size were investigated, MA-ATPE extraction process was optimized by response surface methodology (RSM) base on Box–Behnken design (BBD). In addition, another four methods including STE, MAE, UAE and HRE were compared with MA-ATPE to evaluate the method proposed in this study. Furthermore, the morphologies of both G. scabra residuals after extraction of gentiopicroside and the raw material matrix were observed by scanning electron microscope (SEM) in order to well understand the extraction mechanism. To the best of our knowledge, this is the first report that microwave-assisted ethanol/K2HPO4 ATPS is employed to obtain gentiopicroside from G. scabra. The present work would be helpful for the further exploration of other natural medicinal plant. Materials and methods Materials and reagents Air-dried stems of G. scabra (specimen number 140913) were collected from Huadian City located in Northeast of China in 2014, which were authenticated by Prof. Xin Sui from College of biological and food engineering, Jilin Institute of Chemical Technology, China. The dried samples were smashed by a disintegrator (HX-200A, Yongkang Hardware and Medical Instrument Plant, China) and then sieved with 40, 60, 80, 100 and 120 meshes. The standard of Gentiopicroside (MUST-14052205) was 98% purity at least and purchased from Chengdu Must Bio-Tech Co., Ltd. (Chengdu, China). Acetonitrile and methanol of HPLC (High-performance liquid Chromatography) grade were obtained from Yongda Chemical Reagent Co., Ltd. (Tianjin, China). Ammonium sulfate, sodium dihydrogen phosphate, dipotassium hydrogen phosphate, trisodium citrate, sodium carbonate and other salts were all analytical grade and purchased from Aladdin (Shanghai, China). Ethanol and other solvents were analytical grade and purchased from Damao Reagents Co. (Tianjin, China). The preparation of ATPSs Absolute ethanol, salt and distilled water were used to form ATPS in the present work. Especially, the commonly used 25 kinds of salts were investigated. The detailed preparation process of ATPS can be described as follows: a certain amount of salt was firstly dissolved in distilled water, and then a predetermined volume of absolute ethanol was added to the salt solution. Finally, the mixture was mixed thoroughly by a full stirring at 20 ± 0.1°C on a heating magnetic stirrer (79-1, Nanjing, China) and held until two phase were formed. The concrete properties of each ATPS was observed and noted down, such as the time required stabilizing the phases and the formation features. Phase diagram Once the general selection of the ATPS had been achieved, a phase diagram was needed. The phase diagram was prepared by turbidity titration method at 20 ± 0.1°C according to the turbidimetric titration method previously described by Nemati-Knade (30) as well as the study of our research team (26). Firstly, a certain amount of absolute ethanol was put into a 25 mL conical flask. Then, a certain quantity of salt solution of known mass fraction, which can form ATPS with ethanol easily, was added into the flask dropwise and well mixed up on a heating magnetic stirring apparatus (79-1, Nanjing, China) based on experimental results of section “2.2”. It can be observed that the solution was clear at first, but after a certain amount of salt solution was added, one further drop made the solution turbid and separated into two phases spontaneously. Both the mass fractions of ethanol and salt were recorded exactly. Finally, a few drops of distilled water was added to make the mixture clear again, and the above procedures were repeated many times in order to acquire enough data to construct the phase diagrams. The extraction procedures of gentiopicroside with ATPS To a 15 mL centrifuge tube marked with a scale line (the minimum value was 0.1 mL), 2.0 mL of distilled water, 0.4 mL of crude gentiopicroside solution with a known concentration, a certain volume of anhydrous ethanol and a certain mass of five kinds of salts, including ammonium sulfate, dipotassium hydrogen phosphate, sodium dihydrogen phosphate, sodium carbonate and trisodium citric were added, respectively. Then the whole system was stirred thoroughly using a heating magnetic stirrer (79-1, Nanjing, China) at 20 ± 0.1°C until the salt was dissolved completely, the spontaneous phase separation can be observed during a few seconds (31) because of the incompatibility of ethanol and salt solution as well as the low viscosity of the system. In order to gain the complete phase-forming, a centrifuge procedure at speed of 3,000 rpm for 10 min was performed at 20 ± 0.1°C. The top phase was mainly composed of ethanol and gentiopicroside, and the bottom phase was the salt-rich solution containing polysaccharides, color materials and other impurities. After the volumes of top and bottom phase were determined by direct visual observation to tube marked, both of the two phases were collected and the concentration of gentiopicroside was analyzed by High-performance liquid Chromatography (HPLC) with triplicate. The partition coefficient (K) of the gentiopicroside was determined using equation (1): $$\begin{equation} K={C}_t/{C}_b\end{equation}$$ (1) where Ct and Cb were equilibrium concentration of gentiopicroside in the top and bottom phase, respectively. The phase ratio (R) was determined using equation (2): $$\begin{equation} R={V}_t/{V}_b \end{equation}$$ (2) where Vt and Vb were the volumes of equilibrium alcohol-rich phase and salt-rich phase, respectively. The recovery of gentiopicroside (i.e., in the top phase) was determine using equation (3): $$\begin{equation} Recovery\ \left(\%\right)=\frac{C_t{V}_t}{C_t{V}_t+{C}_b{V}_b}\times 100\end{equation}$$ (3) where Cb and Vb were the measured concentration (μg/mL) of gentiopicroside and volume (mL) of the bottom phase, respectively, and Ct and Vt were the corresponding value of the top phase. Extraction methods MA-ATPE procedures The dried sample (2.0 g), passed through different mesh sieve (40, 60, 80, 100 and 120), was accurately weighed and put into a 150 mL extraction vessel contained ATPS composed of ethanol and K2HPO4, then the extraction vessel was placed into a microwave synthesis/extractor (XH-100, Beijing, China), followed by the microwave irradiation under a given condition according to the experimental design. The detailed extraction procedures were described as follows: G. scabra powder sieved with certain meshes were added to a certain volume of ethanol/K2HPO4 ATPS (liquid/solid ratios in the range of 5:1 to 25:1) and then irradiated under certain power (microwave irradiation power ranging from 600 to 1,000 W) at the specified temperature (extraction temperature ranging from 50 to 90°C) for some seconds (extraction time ranging from 15 to 90 s). After the extraction was completed, the mixture was filtered using filter paper to remove the stem residues, and the filtrate was placed into a 50.0 mL colorimeter tube. The filtrate was left undisturbed to allow phase separation for 1 h at 20 ± 0.1°C on a heating magnetic stirrer (79-1, Nanjing, China). Subsequently, the top and bottom phases were firstly collected using pipette and then filtered through 0.22 μm nylon membrane. Finally, the concentration of gentiopicroside was analyzed by HPLC, the extraction yield of gentiopicroside in G. scabra was determined using the equation (4). $$\begin{equation} Extraction\ yield\ \left( mg/g\right)=\frac{C_t{V}_t+{C}_b{V}_b}{M_s}\end{equation}$$ (4) where Ct and Vt were the measured concentration (μg/mL) and volume (mL) of the top phase, respectively, Cb and Vb were the measured concentration (μg/mL) and volume (mL) of the bottom phase, respectively, and MS was the mass of stem powders used (g). At the same time, in order to confirm the purify reliability of ATPS method established in this work, after the concentration of gentiopicroside distributed in the top phase was determined, the top phase was added to 5-fold dehydrated ethanol to make salt-out. When the supernatant and precipitation were allowed to stand for 1 h, K2HPO4 was removed by centrifugation at 3,000 rpm for 10 min. The resulting supernatant was freeze dried at −90°C in a lyophilizer (SJIA-10 N, Zhejiang, China). The purity of the gentiopicroside was calculated using the equation (5) $$\begin{equation} Purity\ \left(\%\right)=\frac{C_t{V}_t}{M_g}\times 100\end{equation}$$ (5) where Ct and Vt were the same values mentioned in equation (4). Mg was the mass of freeze dried gentiopicroside extracts (g). HRE procedures According to the previous reports (32), the dried sample (2.0), passed through 100-mesh sieve, was accurately weighed and added into a round-bottom flask contained 24 mL of 70% ethanol. After being connected with cooling water, the flask was put into a water-bath with temperature maintained at 60°C, the extraction was conducted for three times with extraction time of 2 h, 1 h and 1 h, respectively. When the extraction was finished, the resulting mixture was centrifuged and insoluble residue was removed, then the filtrate was collected and put into a 250 mL volumetric flask, which subsequently was filled with the extraction solvent. Finally, a small quantity of the sample solution (about 1–2 mL) was filtered using 0.22 μm nylon micropore membrane and analyzed by HPLC, the rest solution was concentrated to remove ethanol aqueous at 50°C in a rotary evaporator (RE-52A, Shanghai, China) under reduced pressure, followed by the thorough dryness using a lyophilizer (SJIA-10 N, Zhejiang, China). The extraction yield and the purity of gentiopicroside were evaluated, respectively. All experiments were performed in triplicate. Microwave-assisted extraction procedures Microwave-assisted extraction (MAE) was carried out in the microwave synthesis/extractor (XH-100, Beijing, China). The dried sample (2.0 g), passed through 100-mesh sieve, was accurately weighed and mixed with a certain amount of ethanol aqueous in a round-bottom flask. Then the flask was placed into the microwave extractor and irradiated under the specified microwave power at given temperature for some time. The final extractions parameters were the same as MAE-ATPE method optimized in this work except for the extraction solvent. After the extraction was finished, the resulting mixture was conducted as described in section “2.5.2”. The extraction yield and the purity of gentiopicroside were evaluated, respectively. All experiments were performed in triplicate. UAE procedures For UAE, the extraction procedures were conducted according to the optimized method determined in previous literatures reported by our research team (33). In the UAE experiments, an ultrasonic cleaner (SY-800, Shanghai, China) was used as an ultrasonic generator, to which the working frequency was fixed at 45 kHz and the power was 300 W on the scale of 0–100%. To be brief, the dried sample (2.0 g), passed through 100-mesh sieve, was accurately weighed and put into a 150 mL flask, after adding 60 mL of 70% ethanol aqueous, the flask was partially immersed into the ultrasonic cleaner and extracted by ultrasonic at the power of 250 W at 35°C for 45 min. After the extraction was completed, all the following procedures were the same as described in section “2.5.2”. The extraction yield and the purity of gentiopicroside were evaluated, respectively. All experiments were conducted in triplicate. STE procedures STE was conducted on a smashing tissue extractor (JHBE-50S, Zhengzhou, China), and the specified extraction conditions was set based on the previous study of our research group (34). The dried sample (2.0), passed through 100-mesh sieve, was accurately weighed and added into a 200 mL extraction cell. After adding 60 mL of 70% ethanol aqueous, the liquid mixture was extracted under the voltage of 150 V at room temperature (25°C) for 35 s. When the extraction was finished, the following procedures were the same as described in section “2.5.2”. The extraction yield and the purity of gentiopicroside were evaluated, respectively. All experiments were conducted in triplicate. Experimental design and statistical analysis On the basis of investigating single factors experiments, in which the preliminary range of five extraction variables were determined, RSM was adopted to determine the optimum extraction of MA-ATPE based on our previous study on optimization of extraction technology (31, 35). The BBD with four independent variables, including microwave irradiation time, microwave power, extraction temperature and ratio of solid to liquid (g/mL) at three levels were conducted. All data were obtained from triplicates by HPLC. The detailed range and levels of the independent variables were presented in Table I. Taking the evaluation of pure error sum of squares into consideration (36), five replicates at the center of the complete experimental design made up of 29 experimental points were performed. Furthermore, in order to achieve the minimization of the effects caused by some unexpected variables, the experiment was deliberately conducted in a random order (37). Data from the BBD were fitted to a quadratic polynomial model, by using the Design-Expert software of 8.0.6, the following quadratic equation (6) can be right for explaining the model. Figure 1 Open in new tabDownload slide Chemical structures of gentiopicroside. Figure 1 Open in new tabDownload slide Chemical structures of gentiopicroside. Table I The Box–Behnken Design Matrix of Four Variables for Extraction of Gentiopicroside Run X1a X2b X3c X4d Extraction time (s) Liquid-solid ratio (mL/g) Microwave power (W) Temperature (°C) Extraction yield (mg/g) 1 1 0 0 −1 60 10 800 70 49.23 2 1 −1 0 0 60 5 800 80 41.67 3 0 0 0 0 45 10 800 80 68.24 4 −1 0 0 1 30 10 800 90 40.62 5 0 1 0 1 45 15 800 90 51.85 6 1 0 1 0 60 10 900 80 52.74 7 0 −1 1 0 45 5 900 80 43.06 8 0 0 0 0 45 10 800 80 64.32 9 0 0 0 0 45 10 800 80 68.97 10 0 0 1 1 45 10 900 90 51.11 11 1 1 0 0 60 15 800 80 49.67 12 0 0 −1 −1 45 10 700 70 48.61 13 0 0 0 0 45 10 800 80 65.22 14 −1 −1 0 0 30 5 800 80 46.83 15 1 0 −1 0 60 10 700 80 46.07 16 0 1 1 0 45 15 900 80 50.65 17 0 0 −1 1 45 10 700 90 51.30 18 −1 0 1 0 30 10 900 80 51.71 19 0 0 1 −1 45 10 900 70 49.05 20 0 1 0 −1 45 15 800 70 52.86 21 0 −1 0 1 45 5 800 90 51.07 22 −1 0 0 0 30 15 800 80 49.59 23 1 0 0 1 60 10 800 90 52.61 24 0 −1 0 −1 45 5 800 70 52.25 25 0 0 0 0 45 10 800 80 66.64 26 −1 0 0 −1 30 10 800 70 46.61 27 −1 0 −1 0 30 10 700 80 49.40 28 0 0 −1 0 45 15 700 80 49.70 29 0 −1 −1 0 45 5 700 80 41.94 Run X1a X2b X3c X4d Extraction time (s) Liquid-solid ratio (mL/g) Microwave power (W) Temperature (°C) Extraction yield (mg/g) 1 1 0 0 −1 60 10 800 70 49.23 2 1 −1 0 0 60 5 800 80 41.67 3 0 0 0 0 45 10 800 80 68.24 4 −1 0 0 1 30 10 800 90 40.62 5 0 1 0 1 45 15 800 90 51.85 6 1 0 1 0 60 10 900 80 52.74 7 0 −1 1 0 45 5 900 80 43.06 8 0 0 0 0 45 10 800 80 64.32 9 0 0 0 0 45 10 800 80 68.97 10 0 0 1 1 45 10 900 90 51.11 11 1 1 0 0 60 15 800 80 49.67 12 0 0 −1 −1 45 10 700 70 48.61 13 0 0 0 0 45 10 800 80 65.22 14 −1 −1 0 0 30 5 800 80 46.83 15 1 0 −1 0 60 10 700 80 46.07 16 0 1 1 0 45 15 900 80 50.65 17 0 0 −1 1 45 10 700 90 51.30 18 −1 0 1 0 30 10 900 80 51.71 19 0 0 1 −1 45 10 900 70 49.05 20 0 1 0 −1 45 15 800 70 52.86 21 0 −1 0 1 45 5 800 90 51.07 22 −1 0 0 0 30 15 800 80 49.59 23 1 0 0 1 60 10 800 90 52.61 24 0 −1 0 −1 45 5 800 70 52.25 25 0 0 0 0 45 10 800 80 66.64 26 −1 0 0 −1 30 10 800 70 46.61 27 −1 0 −1 0 30 10 700 80 49.40 28 0 0 −1 0 45 15 700 80 49.70 29 0 −1 −1 0 45 5 700 80 41.94 a aX1 was extraction time (s), the values for each level (−1, 0, 1) was 15, 30 and 45, respectively. b bX2 was liquid/solid ratio (mL/g), the values for each level (−1, 0, 1) was 5:1, 10:1 and 15:1, respectively. c cX3 was microwave irradiation power (W), the values for each level (−1, 0, 1) was 700, 800 and 900, respectively. d dX4 was extraction temperature (°C), the values for each level (−1, 0, 1) was 70, 80 and 90, respectively. Open in new tab Table I The Box–Behnken Design Matrix of Four Variables for Extraction of Gentiopicroside Run X1a X2b X3c X4d Extraction time (s) Liquid-solid ratio (mL/g) Microwave power (W) Temperature (°C) Extraction yield (mg/g) 1 1 0 0 −1 60 10 800 70 49.23 2 1 −1 0 0 60 5 800 80 41.67 3 0 0 0 0 45 10 800 80 68.24 4 −1 0 0 1 30 10 800 90 40.62 5 0 1 0 1 45 15 800 90 51.85 6 1 0 1 0 60 10 900 80 52.74 7 0 −1 1 0 45 5 900 80 43.06 8 0 0 0 0 45 10 800 80 64.32 9 0 0 0 0 45 10 800 80 68.97 10 0 0 1 1 45 10 900 90 51.11 11 1 1 0 0 60 15 800 80 49.67 12 0 0 −1 −1 45 10 700 70 48.61 13 0 0 0 0 45 10 800 80 65.22 14 −1 −1 0 0 30 5 800 80 46.83 15 1 0 −1 0 60 10 700 80 46.07 16 0 1 1 0 45 15 900 80 50.65 17 0 0 −1 1 45 10 700 90 51.30 18 −1 0 1 0 30 10 900 80 51.71 19 0 0 1 −1 45 10 900 70 49.05 20 0 1 0 −1 45 15 800 70 52.86 21 0 −1 0 1 45 5 800 90 51.07 22 −1 0 0 0 30 15 800 80 49.59 23 1 0 0 1 60 10 800 90 52.61 24 0 −1 0 −1 45 5 800 70 52.25 25 0 0 0 0 45 10 800 80 66.64 26 −1 0 0 −1 30 10 800 70 46.61 27 −1 0 −1 0 30 10 700 80 49.40 28 0 0 −1 0 45 15 700 80 49.70 29 0 −1 −1 0 45 5 700 80 41.94 Run X1a X2b X3c X4d Extraction time (s) Liquid-solid ratio (mL/g) Microwave power (W) Temperature (°C) Extraction yield (mg/g) 1 1 0 0 −1 60 10 800 70 49.23 2 1 −1 0 0 60 5 800 80 41.67 3 0 0 0 0 45 10 800 80 68.24 4 −1 0 0 1 30 10 800 90 40.62 5 0 1 0 1 45 15 800 90 51.85 6 1 0 1 0 60 10 900 80 52.74 7 0 −1 1 0 45 5 900 80 43.06 8 0 0 0 0 45 10 800 80 64.32 9 0 0 0 0 45 10 800 80 68.97 10 0 0 1 1 45 10 900 90 51.11 11 1 1 0 0 60 15 800 80 49.67 12 0 0 −1 −1 45 10 700 70 48.61 13 0 0 0 0 45 10 800 80 65.22 14 −1 −1 0 0 30 5 800 80 46.83 15 1 0 −1 0 60 10 700 80 46.07 16 0 1 1 0 45 15 900 80 50.65 17 0 0 −1 1 45 10 700 90 51.30 18 −1 0 1 0 30 10 900 80 51.71 19 0 0 1 −1 45 10 900 70 49.05 20 0 1 0 −1 45 15 800 70 52.86 21 0 −1 0 1 45 5 800 90 51.07 22 −1 0 0 0 30 15 800 80 49.59 23 1 0 0 1 60 10 800 90 52.61 24 0 −1 0 −1 45 5 800 70 52.25 25 0 0 0 0 45 10 800 80 66.64 26 −1 0 0 −1 30 10 800 70 46.61 27 −1 0 −1 0 30 10 700 80 49.40 28 0 0 −1 0 45 15 700 80 49.70 29 0 −1 −1 0 45 5 700 80 41.94 a aX1 was extraction time (s), the values for each level (−1, 0, 1) was 15, 30 and 45, respectively. b bX2 was liquid/solid ratio (mL/g), the values for each level (−1, 0, 1) was 5:1, 10:1 and 15:1, respectively. c cX3 was microwave irradiation power (W), the values for each level (−1, 0, 1) was 700, 800 and 900, respectively. d dX4 was extraction temperature (°C), the values for each level (−1, 0, 1) was 70, 80 and 90, respectively. Open in new tab $$\begin{equation} Y=\beta k 0+\sum \limits_{i=1}^4\beta k iXi+\sum \limits_{i=1}^4\beta{kiiXi}^2+\sum \limits_{i<j=2}^4\beta k ijXiXj,\end{equation}$$ (6) where Y was the extraction yield of gentiopicroside (the dependent variables), βk0 was the constant coefficient, βi, βii, and βij were the coefficients that represented the linear, quadratic and cross-product effects of the factors including X1, X2, X3 and X4, respectively, where X1 was microwave irradiation time, X2 was ratio of solid to liquid (g/mL), X3 was microwave power and X4 was extraction temperature. HPLC analysis According to our recent study (33), the concentration of gentiopicroside in the extracts was analyzed by HPLC using an ELITE apparatus equipped with an UV230+. Chromatographic separation was performed on a Hypersil ODS column (250 mm × 4.6 mm, 5 μm) and maintained at 25°C under the desired detection wavelength of 270 nm. The mobile phase was composed of acetonitrile (A) and 0.4% (w/v) water-formic acid. Gentiopicroside was eluted at a flow rate of 1.0 mL/min in the following gradient mode: 0~10 min, 5~28% A, 10~16 min, 28~30% A, 16~16.1 min: 30~100% A and 16.1~30 min, 100% A. After a 15 min equilibration period, the next sample was injected, and the injected volume was 20 μL. The chromatographic peak of the gentiopicroside was confirmed by comparing the retention times and UV spectrum with that of the reference compounds. Seven experimental points in the range of 9.12~456.0 μg/mL were employed for establishing a calibration curve. The strong regression lines for gentiopicroside was y = 117.62 + 15.451x (r2 = 0.9999), where y was the peak area of analyte, and x was the concentration of analyte (μg/mL). All solution and sample were filtered through 0.22 μm nylon membranes prior to HPLC analysis. Scanning electron microscopy analysis In order to further reveal the mechanism of extraction, the effects of different extraction methods on the microstructure of samples were investigated. The raw materials of G. scabra and the residue after extraction of gentiopicroside were observed by the use of scanning electron microscopy (SEM) analysis. They were fixed on the silicon wafer and sputtered with gold to a thickness of about 100 nm. The shape and the surface characters tested under high vacuum conditions at an accelerating voltage of 15.0 KV were observed and recorded on the SEM (Quanta-200, FEI Ltd., The Netherlands). Result Preliminary screening of the ATPS In the screening experiments of phase-forming components, ethanol was directly chosen as phase-forming organic alcohol because of its low toxicity and moderate boiling point, which was more suitable for large-scale industrial production compared with other organic solvent. Then, 25 kinds of salts including sodium molybdate (Na2MoO4), sodium tetraborate (Na2B4O7), sodium oxalate (Na2C2O4), ammonium molybdate ((NH4)2Mo4O13), sodium acetate (CH3COONa), potassium hydrogen phthalate (KOOCC6H4COOK), sodium sulfate (NaSO4), potassium carbonate (K2CO3), sodium benzoate (C7H5NaO2), trisodium citrate (C6H5Na3O7), ammonium oxalate (3HNOOC-COONH3), ferrous sulfate (FeSO4), ammonium ferrous sulfate (NH4(SO4)2FeSO4), sodium dihydrogen phosphate (NaH2PO4), disodium hydrogen phosphate (Na2HPO4), trisodium phosphate (Na3PO4), ammonium sulfate [(NH4)2SO4)], dipotassium hydrogen phosphate (K2HPO4), sodium carbonate (Na2CO3), sodium chloride (NaCl), potassium dihydrogen phosphate (KH2PO4), potassium bromate (KBrO3), triammonium citrate (C6H17N3O7), sodium tartrate (Na2C4H4O6) and potassium sodium tartrate (KNaC4H4O6) were investigated based on their phase demixing and formation features. As can be seen from Table II, ethanol/K2HPO4, ethanol/(NH4)2SO4, ethanol/NaH2PO4, ethanol/C6H5Na3O7, ethanol/KNaC4H4O6 and ethanol/Na2CO3 could formed the desired two phases with a transparent system and fast demixing (less than 3 min); ethanol/Na2MoO4, ethanol/C6H17N3O7 and ethanol/Na2C4H4O6 can also form ATPS, but the phase separation required a relatively extended time, meanwhile, one of the two phases was also turbid, which restricted the widely application seriously, and the rest 15 kinds of salts-ethanol cannot form ATPS at all. Table II The Tested Formation and Features of Various ATPSs Phase compositiona Phase ratio Phase demixing Formation features Ethanol NaSO4 No No Precipitation in bottom Na2B4O7 No No Flocculated precipitate in the system (NH4)2SO4 0.62 90 s Stable and transparent NaH2PO4 5.80 2–3 min Stable and transparent Na2HPO4 No No Flocculated precipitate in the system KH2PO4 No No Transparent in the system K2HPO4 2.84 20 s Stable and transparent CH3COONa No No Transparent in the system Na3PO4 No No Mash in the system NaCl No No Transparent in the system FeSO4 No No Transparent in the system KBrO3 No No Salted out in the bottom Na2CO3 11.1 50 s Stable and transparent K2CO3 0.88 1–2 min Slightly turbid in the top phase NH4(SO4)2FeSO4 No No Transparent in the system Na2C2O4 (sodium oxalate) No No Turbid in the system 3HNOOC-COONH3 (ammonium oxalate) No No Transparent in the system KOOCC6H4COOK (potassium biphthalate) No No Transparent in the system C6H17N3O7 (triammonium citrate) 1.27 3–4 min Slightly turbid in the top phase C6H5Na3O7 (trisodium citrate) 1 2–3 min Transparent in the system C7H5NaO2 (sodium benzoate) No No Transparent in the system Na2C4H4O6 (sodium tartrate) 3.5 4–5 min Instable, feculent precipitation in bottom (KNaC4H4O6) Potassium sodium tartrate 1.8 2 min Stable and transparent (NH4)2Mo4O13 (ammonium molybdate) No No Flocculent precipitate in bottom phase Na2MoO4 (sodium molybdate) 0.74 10–11 min Stable and transparent Phase compositiona Phase ratio Phase demixing Formation features Ethanol NaSO4 No No Precipitation in bottom Na2B4O7 No No Flocculated precipitate in the system (NH4)2SO4 0.62 90 s Stable and transparent NaH2PO4 5.80 2–3 min Stable and transparent Na2HPO4 No No Flocculated precipitate in the system KH2PO4 No No Transparent in the system K2HPO4 2.84 20 s Stable and transparent CH3COONa No No Transparent in the system Na3PO4 No No Mash in the system NaCl No No Transparent in the system FeSO4 No No Transparent in the system KBrO3 No No Salted out in the bottom Na2CO3 11.1 50 s Stable and transparent K2CO3 0.88 1–2 min Slightly turbid in the top phase NH4(SO4)2FeSO4 No No Transparent in the system Na2C2O4 (sodium oxalate) No No Turbid in the system 3HNOOC-COONH3 (ammonium oxalate) No No Transparent in the system KOOCC6H4COOK (potassium biphthalate) No No Transparent in the system C6H17N3O7 (triammonium citrate) 1.27 3–4 min Slightly turbid in the top phase C6H5Na3O7 (trisodium citrate) 1 2–3 min Transparent in the system C7H5NaO2 (sodium benzoate) No No Transparent in the system Na2C4H4O6 (sodium tartrate) 3.5 4–5 min Instable, feculent precipitation in bottom (KNaC4H4O6) Potassium sodium tartrate 1.8 2 min Stable and transparent (NH4)2Mo4O13 (ammonium molybdate) No No Flocculent precipitate in bottom phase Na2MoO4 (sodium molybdate) 0.74 10–11 min Stable and transparent Open in new tab Table II The Tested Formation and Features of Various ATPSs Phase compositiona Phase ratio Phase demixing Formation features Ethanol NaSO4 No No Precipitation in bottom Na2B4O7 No No Flocculated precipitate in the system (NH4)2SO4 0.62 90 s Stable and transparent NaH2PO4 5.80 2–3 min Stable and transparent Na2HPO4 No No Flocculated precipitate in the system KH2PO4 No No Transparent in the system K2HPO4 2.84 20 s Stable and transparent CH3COONa No No Transparent in the system Na3PO4 No No Mash in the system NaCl No No Transparent in the system FeSO4 No No Transparent in the system KBrO3 No No Salted out in the bottom Na2CO3 11.1 50 s Stable and transparent K2CO3 0.88 1–2 min Slightly turbid in the top phase NH4(SO4)2FeSO4 No No Transparent in the system Na2C2O4 (sodium oxalate) No No Turbid in the system 3HNOOC-COONH3 (ammonium oxalate) No No Transparent in the system KOOCC6H4COOK (potassium biphthalate) No No Transparent in the system C6H17N3O7 (triammonium citrate) 1.27 3–4 min Slightly turbid in the top phase C6H5Na3O7 (trisodium citrate) 1 2–3 min Transparent in the system C7H5NaO2 (sodium benzoate) No No Transparent in the system Na2C4H4O6 (sodium tartrate) 3.5 4–5 min Instable, feculent precipitation in bottom (KNaC4H4O6) Potassium sodium tartrate 1.8 2 min Stable and transparent (NH4)2Mo4O13 (ammonium molybdate) No No Flocculent precipitate in bottom phase Na2MoO4 (sodium molybdate) 0.74 10–11 min Stable and transparent Phase compositiona Phase ratio Phase demixing Formation features Ethanol NaSO4 No No Precipitation in bottom Na2B4O7 No No Flocculated precipitate in the system (NH4)2SO4 0.62 90 s Stable and transparent NaH2PO4 5.80 2–3 min Stable and transparent Na2HPO4 No No Flocculated precipitate in the system KH2PO4 No No Transparent in the system K2HPO4 2.84 20 s Stable and transparent CH3COONa No No Transparent in the system Na3PO4 No No Mash in the system NaCl No No Transparent in the system FeSO4 No No Transparent in the system KBrO3 No No Salted out in the bottom Na2CO3 11.1 50 s Stable and transparent K2CO3 0.88 1–2 min Slightly turbid in the top phase NH4(SO4)2FeSO4 No No Transparent in the system Na2C2O4 (sodium oxalate) No No Turbid in the system 3HNOOC-COONH3 (ammonium oxalate) No No Transparent in the system KOOCC6H4COOK (potassium biphthalate) No No Transparent in the system C6H17N3O7 (triammonium citrate) 1.27 3–4 min Slightly turbid in the top phase C6H5Na3O7 (trisodium citrate) 1 2–3 min Transparent in the system C7H5NaO2 (sodium benzoate) No No Transparent in the system Na2C4H4O6 (sodium tartrate) 3.5 4–5 min Instable, feculent precipitation in bottom (KNaC4H4O6) Potassium sodium tartrate 1.8 2 min Stable and transparent (NH4)2Mo4O13 (ammonium molybdate) No No Flocculent precipitate in bottom phase Na2MoO4 (sodium molybdate) 0.74 10–11 min Stable and transparent Open in new tab In consideration of the difficult level of phase-forming as well as the convenience of manipulation, five kinds of commonly used salts including K2HPO4, (NH4)2SO4, NaH2PO4, Na2CO3 and C6H5Na3O7 were preliminarily employed for the formation of ATPS. The phase diagrams composed of the five kinds of salt and ethanol were investigated in the following experiments. Phase diagrams of the selected ATPSs In order to determine the phase-forming components, phase diagrams of five kinds of ATPSs including ethanol/Na2CO3, ethanol/K2HPO4, ethanol/(NH4)2SO4, ethanol/C6H5Na3O7 and ethanol/NaH2PO4 were investigated, respectively. As shown in Figure 2, two zones delineated by a binodal curve can be observed. Above the curves, there were two-phase areas that mainly contained ethanol-rich aqueous in the top phase and salt-rich aqueous in the bottom phase, respectively. When the mass fractions of salts were maintained at a high level, the curve would result in the extreme points of forming ATPS with the continuous increase of ethanol concentration. Correspondingly, the excessive phase-forming salt would also separate out at a high level of ethanol concentration. Therefore, the mass fraction of ethanol and salts mentioned above should be firstly chosen in the region above the binodal curve in the following experiment. The results illustrated in Figure 2 indicated the upper limit of mass fractions of the salt was ~40% and that of ethanol was ~35% in the ATPS composed of ethanol/K2HPO4. This conclusion was consistent with the recent report in literatures (13, 29). Figure 2 Open in new tabDownload slide The phase diagrams of five kinds of ethanol/salt system at 20 ± 0.1°C. Figure 2 Open in new tabDownload slide The phase diagrams of five kinds of ethanol/salt system at 20 ± 0.1°C. Selection of the optimal ethanol/salt ATPSs On the basis of section “3.2”, in order to achieve better extraction of gentiopicroside, five kinds of combinations of ATPSs were further optimized to determine the salt type and the mass fractions of phase-forming components, the detailed experimental results were shown in Table III. It can be seen that the system formed by ethanol/K2HPO4 had the highest recovery (95.26%) indicating the enrichment ability of ATPS made up of ethanol and K2HPO4 was the strongest for gentiopicroside. The best performance of ethanol/K2HPO4 system can be further confirmed by the chromatograms (Figure 3), it was clear that the gentiopicroside was mainly distributed in the top phase rather than bottom phase. Therefore, ethanol/K2HPO4 was adopted as the optimized ATPS and further investigated in the present study. Table III Extraction of Gentiopicroside Using Different Ethanol/Salts (Each System Contained 2.0 mL of Distilled Water, 3.0 mL of Ethanol, 2.0 mL of Crude Gentiopicroside Sample Solution and a Certain Amount of Salt) Ethanol/salt system Salt added (g) Mass fractions of ethanol/salt (%) Phase ratio (R) partition coefficient (K) Recovery (R, %) Ethanol/K2HPO4a 0.8 50.55/9.80 5.21 0.46 70.77 1.2 47.77/13.89 3.32 1.43 82.66 1.6 45.28/15.32 3.11 3.29 91.14 2.0 43.04/20.86 2.85 7.03 95.26 2.4 41.01/23.85 2.43 2.21 84.35 Ethanol/NaH2PO4b 1.4 30.50/18.02 4.43 0.20 46.98 1.6 29.74/20.75 3.39 1.54 83.95 1.8 29.01/22.03 2.90 1.67 82.86 2.0 28.32/23.89 2.64 1.65 81.27 2.2 27.82/25.61 1.67 2.26 79.05 Ethanol/(NH4)2SO4 1.0 32.16/13.57 2.80 0.87 81.56 1.2 31.31/15.85 1.71 1.66 73.96 1.4 30.50/18.02 1.35 5.68 88.49 1.6 29.74/20.75 1.05 3.84 80.17 1.8 29.01/22.03 1.22 1.97 70.65 Ethanol/Na2CO3 0.7 33.52/10.01 3.31 0.18 36.92 0.8 33.05/11.61 2.89 0.17 32.44 0.9 32.60/12.38 2.13 0.23 33.57 1.0 32.16/13.57 1.85 0.29 34.65 1.1 31.73/14.73 1.80 0.36 39.35 Ethanol/C6H5Na3O7 (trisodium citrate) 1.2 31.31/15.85 3.31 1.06 67.91 1.4 30.50/18.02 2.89 1.25 68.91 1.6 29.74/20.75 2.13 1.71 72.77 1.8 29.01/22.03 1.85 1.73 74.32 2.0 28.32/23.89 1.80 2.20 68.79 Ethanol/salt system Salt added (g) Mass fractions of ethanol/salt (%) Phase ratio (R) partition coefficient (K) Recovery (R, %) Ethanol/K2HPO4a 0.8 50.55/9.80 5.21 0.46 70.77 1.2 47.77/13.89 3.32 1.43 82.66 1.6 45.28/15.32 3.11 3.29 91.14 2.0 43.04/20.86 2.85 7.03 95.26 2.4 41.01/23.85 2.43 2.21 84.35 Ethanol/NaH2PO4b 1.4 30.50/18.02 4.43 0.20 46.98 1.6 29.74/20.75 3.39 1.54 83.95 1.8 29.01/22.03 2.90 1.67 82.86 2.0 28.32/23.89 2.64 1.65 81.27 2.2 27.82/25.61 1.67 2.26 79.05 Ethanol/(NH4)2SO4 1.0 32.16/13.57 2.80 0.87 81.56 1.2 31.31/15.85 1.71 1.66 73.96 1.4 30.50/18.02 1.35 5.68 88.49 1.6 29.74/20.75 1.05 3.84 80.17 1.8 29.01/22.03 1.22 1.97 70.65 Ethanol/Na2CO3 0.7 33.52/10.01 3.31 0.18 36.92 0.8 33.05/11.61 2.89 0.17 32.44 0.9 32.60/12.38 2.13 0.23 33.57 1.0 32.16/13.57 1.85 0.29 34.65 1.1 31.73/14.73 1.80 0.36 39.35 Ethanol/C6H5Na3O7 (trisodium citrate) 1.2 31.31/15.85 3.31 1.06 67.91 1.4 30.50/18.02 2.89 1.25 68.91 1.6 29.74/20.75 2.13 1.71 72.77 1.8 29.01/22.03 1.85 1.73 74.32 2.0 28.32/23.89 1.80 2.20 68.79 a aK2HPO4 was added to the system in the form of K2HPO4·3H2O. b bNaH2PO4 was added to the system in the form of NaH2PO4·2H2O. Open in new tab Table III Extraction of Gentiopicroside Using Different Ethanol/Salts (Each System Contained 2.0 mL of Distilled Water, 3.0 mL of Ethanol, 2.0 mL of Crude Gentiopicroside Sample Solution and a Certain Amount of Salt) Ethanol/salt system Salt added (g) Mass fractions of ethanol/salt (%) Phase ratio (R) partition coefficient (K) Recovery (R, %) Ethanol/K2HPO4a 0.8 50.55/9.80 5.21 0.46 70.77 1.2 47.77/13.89 3.32 1.43 82.66 1.6 45.28/15.32 3.11 3.29 91.14 2.0 43.04/20.86 2.85 7.03 95.26 2.4 41.01/23.85 2.43 2.21 84.35 Ethanol/NaH2PO4b 1.4 30.50/18.02 4.43 0.20 46.98 1.6 29.74/20.75 3.39 1.54 83.95 1.8 29.01/22.03 2.90 1.67 82.86 2.0 28.32/23.89 2.64 1.65 81.27 2.2 27.82/25.61 1.67 2.26 79.05 Ethanol/(NH4)2SO4 1.0 32.16/13.57 2.80 0.87 81.56 1.2 31.31/15.85 1.71 1.66 73.96 1.4 30.50/18.02 1.35 5.68 88.49 1.6 29.74/20.75 1.05 3.84 80.17 1.8 29.01/22.03 1.22 1.97 70.65 Ethanol/Na2CO3 0.7 33.52/10.01 3.31 0.18 36.92 0.8 33.05/11.61 2.89 0.17 32.44 0.9 32.60/12.38 2.13 0.23 33.57 1.0 32.16/13.57 1.85 0.29 34.65 1.1 31.73/14.73 1.80 0.36 39.35 Ethanol/C6H5Na3O7 (trisodium citrate) 1.2 31.31/15.85 3.31 1.06 67.91 1.4 30.50/18.02 2.89 1.25 68.91 1.6 29.74/20.75 2.13 1.71 72.77 1.8 29.01/22.03 1.85 1.73 74.32 2.0 28.32/23.89 1.80 2.20 68.79 Ethanol/salt system Salt added (g) Mass fractions of ethanol/salt (%) Phase ratio (R) partition coefficient (K) Recovery (R, %) Ethanol/K2HPO4a 0.8 50.55/9.80 5.21 0.46 70.77 1.2 47.77/13.89 3.32 1.43 82.66 1.6 45.28/15.32 3.11 3.29 91.14 2.0 43.04/20.86 2.85 7.03 95.26 2.4 41.01/23.85 2.43 2.21 84.35 Ethanol/NaH2PO4b 1.4 30.50/18.02 4.43 0.20 46.98 1.6 29.74/20.75 3.39 1.54 83.95 1.8 29.01/22.03 2.90 1.67 82.86 2.0 28.32/23.89 2.64 1.65 81.27 2.2 27.82/25.61 1.67 2.26 79.05 Ethanol/(NH4)2SO4 1.0 32.16/13.57 2.80 0.87 81.56 1.2 31.31/15.85 1.71 1.66 73.96 1.4 30.50/18.02 1.35 5.68 88.49 1.6 29.74/20.75 1.05 3.84 80.17 1.8 29.01/22.03 1.22 1.97 70.65 Ethanol/Na2CO3 0.7 33.52/10.01 3.31 0.18 36.92 0.8 33.05/11.61 2.89 0.17 32.44 0.9 32.60/12.38 2.13 0.23 33.57 1.0 32.16/13.57 1.85 0.29 34.65 1.1 31.73/14.73 1.80 0.36 39.35 Ethanol/C6H5Na3O7 (trisodium citrate) 1.2 31.31/15.85 3.31 1.06 67.91 1.4 30.50/18.02 2.89 1.25 68.91 1.6 29.74/20.75 2.13 1.71 72.77 1.8 29.01/22.03 1.85 1.73 74.32 2.0 28.32/23.89 1.80 2.20 68.79 a aK2HPO4 was added to the system in the form of K2HPO4·3H2O. b bNaH2PO4 was added to the system in the form of NaH2PO4·2H2O. Open in new tab Figure 3 Open in new tabDownload slide Chromatograms for the determination of gentiopicroside for standard (A) and sample solution (B). Figure 3 Open in new tabDownload slide Chromatograms for the determination of gentiopicroside for standard (A) and sample solution (B). Partition behaviors of gentiopicroside in ATPS were important for the enrichment. The crude gentiopicroside extraction solution was added into the ATPS composed of ethanol/K2HPO4 at different mass fractions of salt and ethanol, respectively. The effects of K2HPO4 mass fraction on partition coefficient and recovery were firstly studied in this work. As indicated in Table III, the recovery of gentiopicroside increased with the increase of mass fractions of K2HPO4 (from 9.80% to 20.86%), while there was a little decrease when the concentration of K2HPO4 was increased to 23.85%. Actually, the increasing of salt concentration can increase the salting-out ability, which is beneficial for extracting gentiopicroside. However, higher salt concentration will reach saturation and more water will be pulled into salt-rich phase, which can probably make the alcohol-rich phase become not ideal solvent for dissolving gentiopicroside. Therefore, 2.0 g of K2HPO4·3H2O was preliminary used in the further experiments. The effects of ethanol mass fractions in ATPS on recovery were also investigated base on the experimental results of K2HPO4 mass fractions. As can be seen from Table IV, the recovery of gentiopicroside was increased from 89.27% to 97.11% when absolute ethanol was changed from 2.2 mL to 2.6 mL, but the recovery exhibited a tendency to reduce slightly following with continuous increase of ethanol mass fraction, and thus 2.6 mL of absolute ethanol and 2.0 g K2HPO4·3H2O was employed for the following ATPS tests. Gentiopicroside can enter into the top phase (rich in ethanol) easily and achieved higher recovery as expected. Hence, the optimal amount of absolute ethanol was determined as 2.6 mL, i.e., ATPS composed of 40.72% (w/w) ethanol and 21.71% (w/w) K2HPO4 was judged to be the best and used in all the following experiments. Table IV Extraction of Gentiopicroside Using Different Ethanol/K2HPO4 (Each System Contained 2.0 mL of Distilled Water, 2.0 g of K2HPO4·3H2O, 2.0 mL of Crude Gentiopicroside Extraction Solution and a Certain Volume of Ethanol) Ethanol/salt system Ethanol added (mL) Mass fractions of ethanol/salt (%) Phase ratio (R) Partition coefficient (K) Recovery (R, %) Ethanol/K2HPO4 2.2 38.21/22.63 2.44 3.41 89.27 2.6 40.72/21.71 2.86 12.43 97.11 3.0 43.04/19.40 3.02 7.73 95.91 3.4 47.33/20.07 3.17 4.35 93.15 Ethanol/salt system Ethanol added (mL) Mass fractions of ethanol/salt (%) Phase ratio (R) Partition coefficient (K) Recovery (R, %) Ethanol/K2HPO4 2.2 38.21/22.63 2.44 3.41 89.27 2.6 40.72/21.71 2.86 12.43 97.11 3.0 43.04/19.40 3.02 7.73 95.91 3.4 47.33/20.07 3.17 4.35 93.15 Open in new tab Table IV Extraction of Gentiopicroside Using Different Ethanol/K2HPO4 (Each System Contained 2.0 mL of Distilled Water, 2.0 g of K2HPO4·3H2O, 2.0 mL of Crude Gentiopicroside Extraction Solution and a Certain Volume of Ethanol) Ethanol/salt system Ethanol added (mL) Mass fractions of ethanol/salt (%) Phase ratio (R) Partition coefficient (K) Recovery (R, %) Ethanol/K2HPO4 2.2 38.21/22.63 2.44 3.41 89.27 2.6 40.72/21.71 2.86 12.43 97.11 3.0 43.04/19.40 3.02 7.73 95.91 3.4 47.33/20.07 3.17 4.35 93.15 Ethanol/salt system Ethanol added (mL) Mass fractions of ethanol/salt (%) Phase ratio (R) Partition coefficient (K) Recovery (R, %) Ethanol/K2HPO4 2.2 38.21/22.63 2.44 3.41 89.27 2.6 40.72/21.71 2.86 12.43 97.11 3.0 43.04/19.40 3.02 7.73 95.91 3.4 47.33/20.07 3.17 4.35 93.15 Open in new tab Optimization of the extraction process by MA-ATPE In order to improve the extraction process, single factor experiments were conducted after optimum compositions of ethanol/K2HPO4 ATPS was determined. In the present work, five highly influential factors, including solid/liquid ratio (g/mL), microwave irradiation power, extraction time, particle size of sample and extraction temperature were tested, respectively. Different solid-liquid ratios were investigated as the following experiments, 2.0 g of ground dried G. scabra stems (60 mesh) and a certain amount of ATPS were put into a flask, the other parameters were fixed at microwave power 700 W, extraction temperature 60°C and extraction time 60 s. As Figure 4A illustrated, the extraction yield of gentiopicroside increased drastically when liquid/solid (mL/g) ratio increased from 5:1 to 10:1, but it decreased with the continuous increase of liquid/solid ratios. The reason may be ATPS interacts with the electromagnetic field, which directly influences the microwave energy transfer process. A large amount of heat would dissipate in ATPS when the liquid/solid ratio was too high in the extraction process, and only a little amount of heat can be absorbed by raw material. However, if the liquid/solid ratio was low, the extraction solvent would be too little to dissolve the target compound. Therefore, solid-liquid ratios between 1:5 and 1:15 were further screened in the optimized experiments. Figure 4 Open in new tabDownload slide Effects of the different single factors on the extraction yield and recovery of gentiopicroside, in which the ATPS systems were composed of 40.72% (w/w) ethanol and 21.71% (w/w) K2HPO4. (A) Liquid/solid ratio (mL/g), (B) extraction time (s), (C) microwave irradiation power (W), (D) extraction temperature (°C), (E) particle size of sample. Figure 4 Open in new tabDownload slide Effects of the different single factors on the extraction yield and recovery of gentiopicroside, in which the ATPS systems were composed of 40.72% (w/w) ethanol and 21.71% (w/w) K2HPO4. (A) Liquid/solid ratio (mL/g), (B) extraction time (s), (C) microwave irradiation power (W), (D) extraction temperature (°C), (E) particle size of sample. The effects of extraction time on extraction yield of gentiopicroside were studied at the range of 15~90 s when other conditions were fixed as follows: ATPS composed of 21.71% (w/w) K2HPO4 and 40.72% (w/w) ethanol, sample particle size of 60 mesh, microwave irradiation power of 700 W, liquid-solid ratio of 10:1 (mL/g) and extraction temperature of 60°C. As can be seen in Figure 4B, the extraction yield increased drastically when the extraction time was changed from 15 s to 45 s and reached the maximum (45.62 mg/g) with higher recovery of 96.42% at 45 s. Hence, 30–60 s was selected for further optimization. In order to evaluate the effects of microwave irradiation power on extraction yield of gentiopicroside, microwave power at the level of 600 W, 700 W, 800 W, 900 W and 1,000 W were investigated, respectively. The other parameters were set as follows: ATPS composed of 21.71% (w/w) K2HPO4 and 40.72% (w/w) ethanol, sample particle size of 60 mesh, extraction temperature of 60°C, liquid-solid ratio of 10:1 (mL/g) and extraction time of 45 s. As indicated in Figure 4C, it is obvious that the extraction yield almost keep stable and no significant improve can be found within the designed power range mentioned above, only a relatively higher value of 49.12 mg/g with recovery of 96.41% achieve at 800 W. Microwave power could affect the rate of irradiation energy, higher microwave power result in more electromagnetic energy to be transferred to extraction system. However, there was a temperature receptor in the microwave extractor that was used to control the irradiation of energy, and no electromagnetic energy was absorbed by the plant material when actual temperature was above the designed temperature. Similar results have been reported in previous literatures on microwave-assisted extraction of flavonoids from pigeon pea roots (27). Finally, microwave power of 700 W~900 W was further optimized in the following experiments. Generally speaking, high temperature can enhance extraction since it can increase the solubility of target components. Meanwhile, the cell wall can be easily destroyed at a higher temperature, which led to relatively free release of compounds from plant tissue. Extraction at different temperatures (40, 50. 60, 70, 80 and 90°C) was investigated while the other parameters were fixed at 21.71% K2HPO4 (w/w), 40.72% (w/w) ethanol, sample particle size of 60 mesh, microwave irradiation power of 800 W, liquid-solid of 10:1 (mL/g) and extraction time of 45 s. As shown in Figure 4D, the temperature has a strong effect on extraction yield of gentiopicroside. The extraction yield increase drastically with the increase of temperature in the range of 50–80°C. Although higher temperature is good for ideal extraction, it cannot be increased infinitely. As shown in Figure 4D, the extraction yield decline slightly when temperature is increased to 90°C. These results are consistent with the previous reports in microwave-assisted extraction of isoflavonoids from Dalbergia odorifera T. Chen leaves (29). Therefore, 70~90°C is chosen for further optimization. Particle size of sample is another important factor for extraction of compounds from natural products. Generally speaking, the extraction yield can be improved when sample particle is small due to the increase of surface area, high permeability and diffusivity of solvent into material. In the present study, sample passed through 40, 60, 80, 100 and 120 meshes were investigated when the other parameters were the same as mentioned above (ATPS made up of 21.71% (w/w) K2HPO4 and 40.72% (w/w) ethanol, microwave irradiation power of 800 W, liquid-solid of 10: 1 (mL/g), extraction temperature of 80°C and extraction time of 45 s). As shown in Figure 4E, the extraction yield increased continuously within the whole tested range, and the maximum extraction yields achieved when sample was sieved with 120 meshes screen. However, it just increased slightly when the particle size was changed from 100 to 120 meshes. In consideration of the increased difficult level of subsequent filtration caused by small particles, particle sieved with 100 meshes was small enough and therefore used in the optimization tests. These results were the same as that of our previous study on extraction of lignans from Schisandra chinensis Baill (34). Optimization of MA-ATPE conditions with RSM Statistical analysis and model fitting RSM employed in this study involved four individual variables with three levels, in which five replicates at the center point were conducted to investigate the inherent variability and process stability, i.e., the pure error sum of squares can be obtained finally. The 29 different experimental conditions and the response (extraction yield of gentiopicroside) were shown in Table I. Using multiple regression analysis on the experimental data, the response variable Y (extraction yield) can be characterized by the following second-order polynomial equation (7): $$\begin{eqnarray} Y&=&66.86+0.60{X}_1+2.29{X}_2+0.94{X}_3-4.17E-003{X}_4\nonumber\\ && +1.31{X}_1{X}_2+1.09{X}_1{X}_3+2.34{X}_1{X}_4\nonumber\\ && -0.04{X}_2{X}_3-0.04{X}_2{X}_4-0.16{X}_3{X}_4-10.00{X_1}^2\nonumber\\ &&-9.45{X_2}^2-8.93{X_3}^2-7.45{X_4}^2\end{eqnarray}$$ (7) where Y was the extraction yield of gentiopicroside, and X1, X2, X3 and X4 were extraction time (s), liquid-solid ratio (mL/g), microwave irradiation power (W) and extraction temperature (°C), respectively. The statistical significance of the regression equation was evaluated using the F-test and P-value. The results of analysis of variance (ANOVA) for the response surface quadratic model were shown in Table V. As can be seen from Table V, the high F-value (7.82) and low P-value (0.0002) of model indicate regression equation is ideal and the model is also highly statistically significant. Furthermore, the credibility analysis of the regression equation is also conducted; the value of determination coefficient (R2) in this model is 88.67%, indicating the correlation between actual and predicted values is satisfactory. At the same time, the value of coefficient of variation (C.V) is relatively low (7.06%), which imply a better reliability of experiments values. Therefore, the accuracy and the general availability of the polynomial model are adequate in the present paper. Table V Test of Significance for Regression Coefficienta Source Sum of squares df Mean square F-value P-value Modelb 1469.34 14 104.95 7.82 0.0002 X1 4.36 1 4.36 0.32 0.5778 X2 63.02 1 63.02 4.70 0.0479 X3 10.64 1 10.64 0.79 0.3882 X4 2.083-E004 1 2.083-E004 1.553-E005 0.9969 X1 X2 6.86 1 6.86 0.51 0.4862 X1 X3 4.75 1 4.75 0.35 0.5612 X1 X4 21.95 1 21.95 1.64 0.2217 X2 X3 7.225-E003 1 7.225-E003 5.385-E004 0.9818 X2 X4 7.225-E003 1 7.225-E003 5.385-E004 0.9818 X3 X4 0.099 1 0.099 7.396-E003 0.9327 X12 649.11 1 649.11 48.38 <0.0001 X22 579.85 1 579.85 43.22 <0.0001 X32 517.24 1 517.24 38.55 <0.0001 X42 360.12 1 360.12 26.84 0.0001 Residual 187.83 14 13.42 Lack of fit 172.45 10 17.24 4.48 0.0806 Pure Error 15.38 4 3.85 Cor total 1657.17 28 Source Sum of squares df Mean square F-value P-value Modelb 1469.34 14 104.95 7.82 0.0002 X1 4.36 1 4.36 0.32 0.5778 X2 63.02 1 63.02 4.70 0.0479 X3 10.64 1 10.64 0.79 0.3882 X4 2.083-E004 1 2.083-E004 1.553-E005 0.9969 X1 X2 6.86 1 6.86 0.51 0.4862 X1 X3 4.75 1 4.75 0.35 0.5612 X1 X4 21.95 1 21.95 1.64 0.2217 X2 X3 7.225-E003 1 7.225-E003 5.385-E004 0.9818 X2 X4 7.225-E003 1 7.225-E003 5.385-E004 0.9818 X3 X4 0.099 1 0.099 7.396-E003 0.9327 X12 649.11 1 649.11 48.38 <0.0001 X22 579.85 1 579.85 43.22 <0.0001 X32 517.24 1 517.24 38.55 <0.0001 X42 360.12 1 360.12 26.84 0.0001 Residual 187.83 14 13.42 Lack of fit 172.45 10 17.24 4.48 0.0806 Pure Error 15.38 4 3.85 Cor total 1657.17 28 a aThe results was obtained with the Design Expert 8.0.6 software. b bX1 was extraction time (s); X2 was liquid/solid ratio (mL/g); X3 was microwave irradiation power (W); X4 was extraction temperature (°C). Open in new tab Table V Test of Significance for Regression Coefficienta Source Sum of squares df Mean square F-value P-value Modelb 1469.34 14 104.95 7.82 0.0002 X1 4.36 1 4.36 0.32 0.5778 X2 63.02 1 63.02 4.70 0.0479 X3 10.64 1 10.64 0.79 0.3882 X4 2.083-E004 1 2.083-E004 1.553-E005 0.9969 X1 X2 6.86 1 6.86 0.51 0.4862 X1 X3 4.75 1 4.75 0.35 0.5612 X1 X4 21.95 1 21.95 1.64 0.2217 X2 X3 7.225-E003 1 7.225-E003 5.385-E004 0.9818 X2 X4 7.225-E003 1 7.225-E003 5.385-E004 0.9818 X3 X4 0.099 1 0.099 7.396-E003 0.9327 X12 649.11 1 649.11 48.38 <0.0001 X22 579.85 1 579.85 43.22 <0.0001 X32 517.24 1 517.24 38.55 <0.0001 X42 360.12 1 360.12 26.84 0.0001 Residual 187.83 14 13.42 Lack of fit 172.45 10 17.24 4.48 0.0806 Pure Error 15.38 4 3.85 Cor total 1657.17 28 Source Sum of squares df Mean square F-value P-value Modelb 1469.34 14 104.95 7.82 0.0002 X1 4.36 1 4.36 0.32 0.5778 X2 63.02 1 63.02 4.70 0.0479 X3 10.64 1 10.64 0.79 0.3882 X4 2.083-E004 1 2.083-E004 1.553-E005 0.9969 X1 X2 6.86 1 6.86 0.51 0.4862 X1 X3 4.75 1 4.75 0.35 0.5612 X1 X4 21.95 1 21.95 1.64 0.2217 X2 X3 7.225-E003 1 7.225-E003 5.385-E004 0.9818 X2 X4 7.225-E003 1 7.225-E003 5.385-E004 0.9818 X3 X4 0.099 1 0.099 7.396-E003 0.9327 X12 649.11 1 649.11 48.38 <0.0001 X22 579.85 1 579.85 43.22 <0.0001 X32 517.24 1 517.24 38.55 <0.0001 X42 360.12 1 360.12 26.84 0.0001 Residual 187.83 14 13.42 Lack of fit 172.45 10 17.24 4.48 0.0806 Pure Error 15.38 4 3.85 Cor total 1657.17 28 a aThe results was obtained with the Design Expert 8.0.6 software. b bX1 was extraction time (s); X2 was liquid/solid ratio (mL/g); X3 was microwave irradiation power (W); X4 was extraction temperature (°C). Open in new tab Significance of the model is also examined by lack-of-fit test that could measure the failure of the model by means of presenting the data in experiment domain at points that are not included in the regression, as shown in Table V, the F-value of 4.48 and P-value of 0.0806 suggest the lack-of-fit is not significant. The P-value is usually employed as a tool for evaluating the significance of each coefficient, which in turn may imply the pattern of the interaction effects between the variables. The smaller of the P value is, more significant of the corresponding coefficient is. As can be seen from Table V, the lineal coefficient (X2) and all the quadratic term coefficient (X12, X22, X32 and X42) are significant because all the P values are <0.05, indicating the ratio of liquid to solid and quadratic effects between every two factors played an important part in the process of MA-ATPE. Hence, ratio of liquid to solid should be taken seriously when MA-ATPE was employed for extracting and enrich natural bioactive compounds. On the other hand, all the P values obtained from interaction terms are above 0.05, signified all the interaction effects between every two investigated factors, including microwave irradiation power, liquid/solid ratio, extraction time and temperature are non-significant in the present work. To further understand the relationships among the four experimental variables, the response as well as the interactions of experimental variables, the three-dimensional response surface plots were constructed as the function of two tested factors, in which the third one was fixed at its center level, Figure 5A–F indicate the response surface plots obtained from the extraction yield of gentiopicroside. The optimized conditions could be predicted as follows based on the Design-Expert software 8.0: microwave irradiation power of 806 W, extraction time of 31 s, extraction temperature of 80°C and liquid/solid ratio of 11:1 (mL/g). Under these optimal conditions, the extraction yield of gentiopicroside could reach as high as 66.86 mg/g. Figure 5 Open in new tabDownload slide The response surface for the effect of independent variables on the extraction yield of gentiopicroside. (A) Extraction time (s) vs. liquid/solid ratio (mL/g), microwave power (W) and temperature (°C) were held at its optimum, (B) extraction time (s) vs. microwave power (W), liquid/solid ratio (mL/g) and temperature (°C) were held at its optimum, (C) extraction time (s) vs. temperature (°C), liquid/solid ratio (mL/g) and microwave power (W) were held at its optimum, (D) microwave power (W) vs. liquid/solid ratio (mL/g), extraction time (s) and temperature (°C) were held at its optimum, (E) liquid/solid ratio (mL/g) vs. temperature (°C), extraction time (s) and microwave power (W) were held at its optimum, (F) temperature (°C) vs. microwave power (W), liquid/solid ratio (mL/g) and extraction time (s) were held at its optimum. Figure 5 Open in new tabDownload slide The response surface for the effect of independent variables on the extraction yield of gentiopicroside. (A) Extraction time (s) vs. liquid/solid ratio (mL/g), microwave power (W) and temperature (°C) were held at its optimum, (B) extraction time (s) vs. microwave power (W), liquid/solid ratio (mL/g) and temperature (°C) were held at its optimum, (C) extraction time (s) vs. temperature (°C), liquid/solid ratio (mL/g) and microwave power (W) were held at its optimum, (D) microwave power (W) vs. liquid/solid ratio (mL/g), extraction time (s) and temperature (°C) were held at its optimum, (E) liquid/solid ratio (mL/g) vs. temperature (°C), extraction time (s) and microwave power (W) were held at its optimum, (F) temperature (°C) vs. microwave power (W), liquid/solid ratio (mL/g) and extraction time (s) were held at its optimum. Verification of the predictive model There is a great need to evaluate the accuracy and reliability of the model equation. The confirmatory experiments were performed in triplicate under the following procedure: 2.0 g of dried G. scabra powder sieved 100-mesh was accurately weighed and mixed with 22.20 mL of ATPS composed of K2HPO4 (21.71%, w/w) and ethanol (40.72%, w/w) in a extraction vessel, then the vessel was placed into a microwave extraction instrument, the extraction was conducted when other parameters were set as follows: microwave irradiation power of 806 W, extraction time of 31 s and extraction temperature of 80°C. The actual extraction yield of gentiopicroside have reached 65.32 ± 0.24 mg/g (n = 3) with the recovery of 96.51% (gentiopicroside dispersed in the top phase). Figure 6 Open in new tabDownload slide SEM images of untreated G. scabra sample (A), sample after MA-ATPE (B), sample after microwave-assisted extraction (C), sample after STE (D), sample after UAE (E) and sample after HRE (F). Figure 6 Open in new tabDownload slide SEM images of untreated G. scabra sample (A), sample after MA-ATPE (B), sample after microwave-assisted extraction (C), sample after STE (D), sample after UAE (E) and sample after HRE (F). Analysis of microscopic changes The extraction yields of gentiopicroside extracted by various methods are very different, these phenomena may be related to the specific extraction characteristics, and thus SEM was employed to observe the microscopic changes of G. scabra residual and crude sample. Results of microscopic changes of untreated G. scabra sample and five kinds of residuals were indicated in Figure 6. As can be seen from Figure 6A, the parenchyma is complete nubbly, no destroy on cell walls can be observed for the G. scabra raw material. Contrasted with the untreated material, all the residual show an illustration of parenchyma disorganization, only the damage level of cell walls are different. As depicted in Figure 6B–F, an increasing level of cell damage increased in the order of HRE < UAE < STE < MAE. A crumbly and very thin texture can be found in Figure 6B and C, which may be related with the sudden temperature rise and internal pressure increase as a result of microwave irradiation (38). After STE treatment, the cell is also damaged and its tissue become thinner (Figure 6D) due to high-velocity mechanical shearing power and ultra-dynamic molecular percolation effects (34), and therefore the morphology change of residual is similar to that of MAE, only the destroy level is relatively slight. The morphology of sample treated by UAE can be seen from Figure 6E, from which we can know the cell wall is partly destroyed and tissue disorganization occurred because of cavitation effects. As for HRE, no significant difference can be observed from Figure 6F compared with other extraction methods, only a slight change of samples surfaces is found clearly, these phenomena may be the result of long heating time (39). Compared MAE with MAE-ATPE, both cell walls of the two samples are greatly ruptured (Figure 6D–E), but more significant impact on the morphology changes can be observed from the residual obtained from MAE-ATPE, the reasons may be microwave energy adsorption is greatly enhanced in the presence the inorganic salt. The result mentioned above implies MAE-ATPE is superior to MAE, STE, UAE and HRE. MAE-ATPE enhances the absorption of microwave irradiation and makes the parenchyma of plant sample ultrathin and disorganized. Hence, the chemical substances within the cell can be released easily into the surrounding solvents in comparison with other extraction techniques. Comparison of different extraction methods In the present work, the novel MA-ATPE approach for concurrent extraction and purification of gentiopicroside from stems of G. scabra is compared with other four techniques including HRE, STE, MAE and UAE. The results are shown in Table VI, from which we can know extraction yield of gentiopicroside decreased in the order of MA-ATPE > MAE > STE > UAE > HRE. Meanwhile, the highest extraction yield can be found to be 65.32 ± 0.24 mg/g, while the lowest extraction yield was only 49.93 ± 0.37 mg/g. Table VI Comparisons of the Extraction Yield and Purity of Gentiopicroside by Different Methods Methoda MA-ATPEb MAEc STEc UAEc HREc Extraction temperature (°C) 80 80 25 65 100 Extraction time 31 s 31 s 35 s 45 min 2 h1 h1 h Ultrasonic power (W) 250 Microwave power (W) 806 806 Extraction voltage (V) 150 Extraction times 1 1 1 1 3 Solvent/material ratio (mL/g) 11:1 11:1 30:1 30:1 36:1 Extraction yield (mg/g) 65.32 ± 0.24 64.16 ± 0.17 53.70 ± 0.11 52.12 ± 0.08 49.93 ± 0.37 Purity (%) 17.16 ± 0.25 12.33 ± 0.34 10.94 ± 0.11 11.20 ± 0.17 8.87 ± 0.21 Methoda MA-ATPEb MAEc STEc UAEc HREc Extraction temperature (°C) 80 80 25 65 100 Extraction time 31 s 31 s 35 s 45 min 2 h1 h1 h Ultrasonic power (W) 250 Microwave power (W) 806 806 Extraction voltage (V) 150 Extraction times 1 1 1 1 3 Solvent/material ratio (mL/g) 11:1 11:1 30:1 30:1 36:1 Extraction yield (mg/g) 65.32 ± 0.24 64.16 ± 0.17 53.70 ± 0.11 52.12 ± 0.08 49.93 ± 0.37 Purity (%) 17.16 ± 0.25 12.33 ± 0.34 10.94 ± 0.11 11.20 ± 0.17 8.87 ± 0.21 a aExtraction yields of gentiopicroside (mg/g) = mean ± SD (n = 3) and RSD were all lower than 3.5%. b bThe extraction solvent was ATPS composed of K2HPO4 (21.71%, w/w) and ethanol (40.72%, w/w). c cThe extraction solvent was 70% ethanol (v/v). Open in new tab Table VI Comparisons of the Extraction Yield and Purity of Gentiopicroside by Different Methods Methoda MA-ATPEb MAEc STEc UAEc HREc Extraction temperature (°C) 80 80 25 65 100 Extraction time 31 s 31 s 35 s 45 min 2 h1 h1 h Ultrasonic power (W) 250 Microwave power (W) 806 806 Extraction voltage (V) 150 Extraction times 1 1 1 1 3 Solvent/material ratio (mL/g) 11:1 11:1 30:1 30:1 36:1 Extraction yield (mg/g) 65.32 ± 0.24 64.16 ± 0.17 53.70 ± 0.11 52.12 ± 0.08 49.93 ± 0.37 Purity (%) 17.16 ± 0.25 12.33 ± 0.34 10.94 ± 0.11 11.20 ± 0.17 8.87 ± 0.21 Methoda MA-ATPEb MAEc STEc UAEc HREc Extraction temperature (°C) 80 80 25 65 100 Extraction time 31 s 31 s 35 s 45 min 2 h1 h1 h Ultrasonic power (W) 250 Microwave power (W) 806 806 Extraction voltage (V) 150 Extraction times 1 1 1 1 3 Solvent/material ratio (mL/g) 11:1 11:1 30:1 30:1 36:1 Extraction yield (mg/g) 65.32 ± 0.24 64.16 ± 0.17 53.70 ± 0.11 52.12 ± 0.08 49.93 ± 0.37 Purity (%) 17.16 ± 0.25 12.33 ± 0.34 10.94 ± 0.11 11.20 ± 0.17 8.87 ± 0.21 a aExtraction yields of gentiopicroside (mg/g) = mean ± SD (n = 3) and RSD were all lower than 3.5%. b bThe extraction solvent was ATPS composed of K2HPO4 (21.71%, w/w) and ethanol (40.72%, w/w). c cThe extraction solvent was 70% ethanol (v/v). Open in new tab Regeneration of salt and purification of gentiopicroside In order to reduce the cost, protect environment and purify gentiopicroside, inorganic salt (K2HPO4) should be removed from the product or recycled for further industrial application, a traditional method named dilution crystallization is applied in this work. On the basis of previous study (26), the resulting mixtures obtain from MA-ATPE is subjected to filtration with filter paper, the filtrate was collected and standing for 1 h at 20 ± 0.1°C. After two phases were separated from each other, the ethanol-rich phase was added to 5-fold of absolute ethanol and kept in 4°C overnight，K2HPO4 can be precipitated and further eliminated by filtration. By means of performing this simple salt-removing operation, the gentiopicroside with a relatively high purity (5.93%) in the crude extracts can be obtained. Discussion In the preliminary screening of the ATPS, K2HPO4 and (NH4)2SO4 have been acknowledged as the two most effective and promising salts that have been extensively used in the compositions of ATPS in the last few years (15–16, 32); therefore, the conclusions drew from the present study were consistent with that of previous literatures mentioned above. While among the numerous salts investigated in this work, most of them cannot form ATPS successfully, indicating the salt used for forming ATPS was specific and very limited. The reasons for these phenomena may be related with the charge of the salt ion as well as the size, but the exact mechanism for phase-forming salt is complex, a more advisable and reasonable explanation need to be further explored in the future. However, a certain relationship between the phase separations and performance of salt can be concluded as follows: (i) the salt with low water solubility in water could be difficult to form ATPS with ethanol, in which the salted-out and flocculent turbidity can be observed obviously in the system, such as Na2HPO4, KBrO3, Na2B4O7, (NH4)2Mo4O13 and Na2C2O4; (ii) the salt contained multivariate acid radical ion can form ATPS more easily, such as (NH4)2Mo4O13, C6H5Na3O7, KNaC4H4O6. Therefore, the salt that contained multivariate acid radical ion and possess higher water solubility simultaneously should be given priority in the screening of ATPS compositions. When a tested response is influenced by various factors and the possible interactions among independent variables, RSM is an effective statistical technique for optimizing the complicated processes originally described by Box and Wilson (40). Compared with orthogonal tests and other optimizing techniques, RSM allows more efficient, easier arrangement and interpretation of experiments and possesses lots of advantages including reducing materials, saving time and decreasing sample expenses (34). More importantly, the response surface mapping appeared in RSM design facilitated an improved understanding of the factor–response relationship and interactions associated with them (41). Hence, RSM are receiving growing interest and has been extensively applied for the technological optimization (41–42). In this study, the actual extraction yield of gentiopicroside (65.32 ± 0.24 mg/g) is very close to the predicted value, indicating the predicted model is accordant with the actual value. Therefore, the model can be used for predicting extraction yield of gentiopicroside in G. scabra successfully. In order to evaluate the probabilities of the proposed extraction approach for major target compounds, the extraction efficiency of gentiopicroside by MA-ATPE was compared with those of STE, UAE, HRE, MA-ATPE and MAE. Results showed extraction yield was improved obviously when MAE was adopted. However, the purity differed greatly between MA-ATPE and MAE, the contents of gentiopicroside in the crude extracts obtained from MA-ATPE reached 17.16 ± 0.25%, which was significantly superior to that of MAE, indicating the purity of gentiopicroside can be improved when ATPS was selected as extraction solvent. This result is similar to that of reported literature for extraction of polysaccharides from G. scabra (43). Therefore, MA-ATPE not only can increase the extraction yield of gentiopicroside in a shortened extraction time, but also gain a higher purity. The ideal extraction yield with higher purity could be simultaneously obtained because this novel technique combined the advantages of MAE for extraction and ATPS for enrichment. Conclusions In the present study, a relatively green and potential MA-ATPE method has been developed for concurrent enhancement of extraction yield and purity of gentiopicroside from G. scabra. ATPS composed of K2HPO4 (w/w, 21.71%) and ethanol (w/w, 40.72%) were proved to possess the optimal extraction ability for gentiopicroside based on investigation of 25 kinds of salts. RSM is successfully applied for the extraction technology optimization of gentiopicroside. The optimized conditions are as follows: 100 meshes of raw material sample, 806 W of microwave irradiation power, 31 s of extraction time, 80°C of extraction temperature and 11:1 (g/mL) of liquid/solid ratio. Under these conditions, the extraction yield of gentiopicroside reach 65.32 ± 0.24 mg/g with recovery of 96.51%. Compared with STE, MAE, UAE and HRE, not only the extraction yield of gentiopicroside is improved significantly, but also purity is enhanced as expected. The purity of gentiopicroside obtained from MA-ATPE reach 17.16 ± 0.25%, which are significantly superior to that of STE, MAE, UAE and HRE. Therefore, the combinations of aqueous-two-phase and microwave-assisted extraction are a promising technique for extracting gentiopicroside from G. scabra. As an excellent and alternative technique, MA-ATPE composed of K2HPO4 and ethanol can also be potential for extraction of natural compounds from other medicinal plants. Acknowledgments This work was financially supported by Heilongjiang University Ph.D. Innovation Research Project of Heilongjiang Province of China (YJSCX2019-02HLJU), Natural Science Foundation of Jilin Province of China (20170101211JC and 20170204001YY), Youth Foundation of Jilin Province of China (20160520130JH), Science and Technology Project of Jilin Provincial Department of Education (JJKH20170214KJ) and Science Foundation of Jilin Institute of Chemical Technology (201635 and 2018019). References [1] Georgieva , E. , Handjieva , N. , Popov , S. , Evstatieva , L. ; Comparative analysis of the volatiles from flowers and leaves of three Gentiana species ; Biochemical Systematics and Ecology , ( 2005 ); 33 : 938 – 947 . Google Scholar Crossref Search ADS WorldCat [2] The State Pharmacopoeia Committee of China ; The pharmacopoeia of the People’s Republic of China . Chemical Industry Press ( part 1 ), Beijing , ( 2015 ), p. 96 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC [3] Kim , J.A. , Son , N.S. , Son , J.K. , Jahng , Y. , Chang , H.W. , Jang , T.S. ; Two new secoiridoid glycosides from the rhizomes of Gentiana scabra Bunge ; Archives of Pharmacal Research , ( 2009 ); 32 : 863 – 867 . Google Scholar Crossref Search ADS PubMed WorldCat [4] Cai , W.R. , Xu , H.L. , Xie , L.L. , Sun , J. , Sun , T.T. , Wu , X.Y. , et al. ; Purification, characterization and in vitro anticoagulant activity of polysaccharides from Gentiana scabra Bunge roots ; Carbohydrate Polymers , ( 2016 ); 140 : 308 – 313 . Google Scholar Crossref Search ADS PubMed WorldCat [5] Wang , C.Y. , Wang , Y. , Zhang , J. , Wang , Z.Y. ; Optimization for the extraction of polysaccharides from Gentiana scabra Bunge and their antioxidant in vitro and anti-tumor activity in vivo ; Journal of the Taiwan Institute of Chemical Engineers , ( 2014 ); 45 : 1126 – 1132 . Google Scholar Crossref Search ADS WorldCat [5] Chang-Liao , W.L. , Chien , C.F. , Lin , L.C. , Tsai , T.H. ; Isolation of gentiopicroside from Gentianae Radix and its pharmacokinetics on liver ischemia/reperfusion rats ; Journal of Ethnopharmacology , ( 2012 ); 141 : 668 – 673 . Google Scholar Crossref Search ADS PubMed WorldCat [6] Lian , L.H. , Wu , Y.L. , Wan , Y. , Li , X. , Xie , W.X. , Nan , J.X. ; Anti-apoptotic activity of gentiopicroside in d-galactosamine/lipopolysaccharide-induced murine fulminant hepatic failure ; Chemico-Biological Interactions , ( 2010 ); 188 : 127 – 133 . Google Scholar Crossref Search ADS PubMed WorldCat [7] Wang , C.H. , Cheng , X.M. , He , Y.Q. , White , K.N. , Bligh , S.W. , White , B.F. ; Pharmacokinetic behavior of gentiopicroside from decoction of radix gentianae, gentiana macrophylla after oral administration in rats: a pharmacokinetic comparison with gentiopicroside after oral and intravenous administration alone ; Archives of Pharmacal Research , ( 2007 ); 30 : 1149 – 1154 . Google Scholar Crossref Search ADS PubMed WorldCat [8] Duan , B.Z. , Hu , J.Y. , Huang , L.F. , Yang , X.Y. , Chen , F.Y. ; Chemical fingerprint analysis of Gentianae Radix et Rhizoma by high-performance liquid chromatography ; Acta Pharmaceutica Sinica B , ( 2012 ); 2 : 46 – 52 . Google Scholar Crossref Search ADS WorldCat [9] Wang , Z.Y. , Wang , C.Y. , Su , T.T. , Zhang , J. ; Antioxidant and immunological activities of polysaccharides from Gentiana scabra Bunge roots ; Carbohydrate Polymers , ( 2014 ); 112 : 114 – 118 . Google Scholar Crossref Search ADS PubMed WorldCat [10] Zheng , P. , Zhang , K.J. , Wang , Z.Z. ; Genetic diversity and gentiopicroside content of four Gentiana species in China revealed by ISSR and HPLC methods ; Biochemical Systematics and Ecology , ( 2011 ); 39 : 704 – 710 . Google Scholar Crossref Search ADS WorldCat [11] Guo , X.C. , Zhu , H.Y. , Zhou , H.B. , Bing , L. , Wang , S. , Zang , J.J. , et al. ; Optimization on extraction process of gentiopicroside by orthogonal design ; Journal of Jilin Medical College , ( 2012 ); 33 : 77 – 79 . OpenURL Placeholder Text WorldCat [12] Cao , C.L. , Huang , Q. , Zhang , B. , Li , C. , Fu , X. ; Physicochemical characterization and in vitro hypoglycemic activities of polysaccharides from Sargassum pallidum by microwave-assisted aqueous two-phase extraction ; International Journal of Biological Macromolecules , ( 2018 ); 109 : 357 – 368 . Google Scholar Crossref Search ADS PubMed WorldCat [13] Zhang , W. , Zhu , D. , Fan , H.J. , Liu , X.Q. , Wan , Q. , Wu , X.H. , et al. ; Simultaneous extraction and purification of alkaloids from Sophora flavescens Ait. by microwave-assisted aqueous two-phase extraction with ethanol/ammonia sulfate system ; Separation and Purification Technology , ( 2015 ); 141 : 113 – 123 . Google Scholar Crossref Search ADS WorldCat [14] Li , C. , Han , Y.P. ; Optimization of extraction of conditions for gentiopicrin from gentian by response surface methodology ; Journal of Southwest University for Nationalities , ( 2013 ); 39 : 204 – 208 . OpenURL Placeholder Text WorldCat [15] Zhu , H.Y. , Zhang , L. , Feng , B. ; Comparison of ultrasonic method and smashing tissue extraction process of Gentiopicroside from Gentianae Radix et Rhizoma ; Chinese Journal of Modern Applied Pharmacy , ( 2011 ); 28 : 395 – 398 . OpenURL Placeholder Text WorldCat [16] Chan , C.H. , Yusoff , R. , Ngoh , G.C. , Kung , F.W. ; Microwave-assisted extractions of active ingredients from plants ; Journal of Chromatography A , ( 2011 ); 1218 : 6213 – 6225 . Google Scholar Crossref Search ADS PubMed WorldCat [17] Albertsson , P.A. ; Partition of cell particles and macromolecules , third ed. Wiley , New York , ( 1986 ). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC [18] Zhang , X.F. , Teng , G.X. , Zhang , J. ; Ethanol/salt aqueous two-phase system based ultrasonically assisted extraction of polysaccharides from Lilium davidiivar. unicolor Salisb: physicochemical characterization and antiglycation properties ; Journal of Molecular Liquids , ( 2018 ); 256 : 497 – 506 . Google Scholar Crossref Search ADS WorldCat [19] Xie , X.J. , Zhu , D. , Zhang , W. , Huai , W.B. , Wang , K. , Huang , X.W. , et al. ; Microwave-assisted aqueous two-phase extraction coupled with high performance liquid chromatography for simultaneous extraction and determination of four flavonoids in Crotalaria sessiliflora L ; Industrial Crops and Products , ( 2017 ); 95 : 632 – 642 . Google Scholar Crossref Search ADS WorldCat [20] Zhang , W. , Liu , X.Q. , Fan , H.J. , Zhu , D. , Wu , X.H. , Huang , X.W. , et al. ; Separation and purification of alkaloids from Sophora flavescens Ait. by focused microwave-assisted aqueous two-phase extraction coupled with reversed micellar extraction ; Industrial Crops and Products , ( 2016 ); 86 : 231 – 238 . Google Scholar Crossref Search ADS WorldCat [21] Ooi , C.W. , Hi , S.L. , Kamal , S.M.M. , Ariff , A. , Ling , T.C. ; Extractive fermentation using aqueous two-phase systems for integrated production and purification of extracellular lipase derived from Burkholderia pseudomallei ; Process Biochemistry , ( 2011 ); 46 : 68 – 73 . Google Scholar Crossref Search ADS WorldCat [22] Antov , M. , Omorjan , R. ; Pectinase partitioning in polyethylene glycol 1000/Na2SO4 aqueous two-phase system: statistical modeling of the experimental results ; Bioprocess and Biosystems Engineering , ( 2009 ); 32 : 235 – 240 . Google Scholar Crossref Search ADS PubMed WorldCat [23] Pimentel , M.C.B. , Araujo , A.L. , Figueiredo , Z.M.B. , Silva , R.A. , Cavalcanti , M.T.H. , Moreira , K.A. , et al. ; Aqueous two-phase system for citrinin from fermentation broth ; Separation and Purification Technology , ( 2013 ); 110 : 158 – 163 . Google Scholar Crossref Search ADS WorldCat [24] Rahimpou , F. , Hatti-Kaul , R. , Mamo , G. ; Response surface methodology and artificial neural network modeling of an aqueous two-phase system for purification of a recombinant alkaline active xylanase ; Process Biochemistry , ( 2016 ); 51 : 452 – 462 . Google Scholar Crossref Search ADS WorldCat [25] Xing , J.M. , Li , F.F. ; Chiral separation of mandelic acid by temperature-induced aqueous two-phase system ; Journal of Chemical Technology and Biotechnology , ( 2012 ); 87 : 346 – 350 . Google Scholar Crossref Search ADS WorldCat [26] Cheng , Z.Y. , Cheng , L.Q. , Song , H.Y. , Yu , L.Y. , Zhong , F.L. , Shen , Q.H. , et al. ; Aqueous two-phase system for preliminary purification of lignans from fruits of Schisandra chinensis Baill ; Separation and Purification Technology , ( 2016 ); 166 : 16 – 25 . Google Scholar Crossref Search ADS WorldCat [27] Zhang , D.Y. , Zu , Y.G. , Fu , Y.J. , Wang , W. , Zhang , L. , Luo , M. , et al. ; Aqueous two-phase extraction and enrichment of two main flavonoids from pigeon pea roots and the antioxidant activity ; Separation and Purification Technology , ( 2013 ); 102 : 26 – 33 . Google Scholar Crossref Search ADS WorldCat [28] Tan , Z.J. , Wang , C.Y. , Yi , Y.J. , Wang , H.Y. , Li , M. , Zhou , W.L. , et al. ; Extraction and purification of chlorogenic acid from ramie (Boehmeria nivea L. Gaud) leaf using an ethanol/salt aqueous two-phase system ; Separation and Purification Technology , ( 2014 ); 132 : 396 – 400 . Google Scholar Crossref Search ADS WorldCat [29] Ma , F.Y. , Gu , C.B. , Li , C.Y. , Luo , M. , Wang , W. , Zu , Y.G. , et al. ; Microwave-assisted aqueous two-phase extraction of isoflavonoids from Dalbergia odorifera T. Chen leaves ; Separation and Purification Technology , ( 2013 ); 115 : 136 – 144 . Google Scholar Crossref Search ADS WorldCat [30] Nemati-Knade , E. , Shekaari , H. , Jafari , S.A. ; Thermodynamic study of aqueous two phase systems for some aliphatic alcohols plus sodium thiosulfate plus water ; Fluid Phase Equilibria , ( 2012 ); 321 : 64 – 72 . Google Scholar Crossref Search ADS WorldCat [31] Cheng , Z.Y. , Song , H.Y. , Yang , Y.J. , Liu , Y. , Liu , Z.G. , Hu , H.B. , et al. ; Optimization of microwave-assisted enzymatic extraction of polysaccharides from the fruit of Schisandra chinensis Baill ; International Journal of Biological Macromolecules , ( 2015 ); 76 : 161 – 168 . Google Scholar Crossref Search ADS PubMed WorldCat [32] Xiang , Z.Y. , Li , X.F. , Luo , K.P. , Luo , J. , Yang , L. , Lin , H. , et al. ; Study on the extraction technology of total gentiin in Longdan by orthogonal design method ; Pharmacy and Clinics of Chinese Materia Medica , ( 2015 ); 6 : 23 – 25 . OpenURL Placeholder Text WorldCat [33] Cheng , Z.Y. , Song , H.Y. , Yang , Y.J. , Jiang , G.Z. , Yu , S.H. , Xue , J.L. ; HPLC analytical and ultrasound-assisted extraction technology of gentiopicroside ; Journal of Henan University of Technology , ( 2015 ); 36 : 64 – 68 . OpenURL Placeholder Text WorldCat [34] Cheng , Z.Y. , Song , H.Y. , Yang , Y.J. , Zhou , H.L. , Liu , Y. , Liu , Z.G. ; Smashing tissue extraction of five lignans from the fruit of Schisandrin chinensis ; Journal of Chromatographic Science , ( 2016 ); 54 ( 2 ): 246 – 256 . Google Scholar Crossref Search ADS PubMed WorldCat [35] Cheng , Z.Y. , Yang , Y.J. , Liu , Y. , Liu , Z.G. , Zhou , H.L. , Hu , H.B. ; Two-steps extraction of essential oil, polysaccharides and biphenyl cyclooctene lignans from Schisandra chinensis Baill fruits ; Journal of Pharmaceutical and Biomedical Analysis , ( 2014 ); 96 : 162 – 169 . Google Scholar Crossref Search ADS PubMed WorldCat [36] Ji , Y.B. , Dong , F. , Dong , B. , Miao , J. , Jin , L.N. , Liu , J.F. , et al. ; Optimizing the extraction of anti-tumor polysaccharides from the fruit of Capparis spionosa L. by response surface methodology ; Molecules , ( 2012 ); 17 : 7323 – 7335 . Google Scholar Crossref Search ADS PubMed WorldCat [37] Chen , R.Z. , Li , S.Z. , Liu , C.M. , Yang , S.M. , Li , X.L. ; Ultrasound complex enzymes assisted extraction and biochemical activities of polysaccharides from Epimedium leaves ; Process Biochemistry , ( 2012 ); 47 : 2040 – 2050 . Google Scholar Crossref Search ADS WorldCat [38] Li , F.F. , Li , Q. , Wu , S.G. , Tan , Z.J. ; Salting-out extraction of allicin from garlic (Allium sativum L.) based on ethanol/ammonium sulfate in laboratory and pilot scale ; Food Chemistry , ( 2017 ); 217 : 91 – 97 . Google Scholar Crossref Search ADS PubMed WorldCat [39] Yoshida , T. , Tsubaki , S. , Teramoto , Y. , Azuma , J.I. ; Optimization of microwave-assisted extraction of carbohydrates from industrial waste of corn starch production using response surface methodology ; Bioresource Technology , ( 2010 ); 101 : 7820 – 7826 . Google Scholar Crossref Search ADS PubMed WorldCat [40] Box , G.E.P. , Wilson , K.B. ; On the experimental attainment of optimum conditions ; Journal of the Royal Statistical Society. Series A (General) , ( 1951 ); 13 : 1 – 45 . OpenURL Placeholder Text WorldCat [41] Sandhu , P.S. , Beg , S. , Katare , O.P. , Singh , B. ; QbD-driven development and validation of a HPLC method for estimation of tamoxifen citrate with improved performance ; Journal of Chromatographic Science , ( 2016 ); 54 ( 8 ): 1373 – 1384 . Google Scholar Crossref Search ADS PubMed WorldCat [42] Hasnain , M.S. , Ansari , S.A. , Rao , S. , Tabish , M. , Singh , M. , Abdullah , M.S. , et al. ; QbD-driven development and validation of liquid chromatography tandem mass spectrometric method for the quantitation of sildenafil in human plasma ; Journal of Chromatographic Science , ( 2017 ); 55 ( 6 ): 587 – 594 . Google Scholar Crossref Search ADS PubMed WorldCat [43] Cheng , Z.Y. , Song , H.Y. , Cao , X.L. , Shen , Q.H. , Han , D.D. , Zhong , F.L. , et al. ; Simultaneous extraction and purification of polysaccharides from Gentiana scabra Bunge by microwave-assisted ethanol-salt aqueous two-phase system ; Industrial Crops and Products , ( 2017 ); 102 : 75 – 87 . Google Scholar Crossref Search ADS WorldCat © The Author(s) 2019. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Concurrent Extraction and Purification of Gentiopicroside from Gentiana scabra Bunge Using Microwave-Assisted Ethanol-Salt Aqueous Two-Phase Systems
JO - Journal of Chromatographic Science
DO - 10.1093/chromsci/bmz101
DA - 2020-01-01
UR - https://www.deepdyve.com/lp/oxford-university-press/concurrent-extraction-and-purification-of-gentiopicroside-from-Bg0P5nraTU
DP - DeepDyve
ER -