957
J. Sep. Sci. 2014, 37, 957–965
Zhisheng Xie1,2 Yongjiang Sun3 Shingchung Lam1,2 Mingqian Zhao1,2 Zhikun Liang1,2 Xiaoxue Yu1,2 Depo Yang1,2 Xinjun Xu1,2 1 School
of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou, China 2 Laboratory for Quality Control of Traditional Chinese Medicine, Guangdong Technology Research Centre for Advanced Chinese Medicine, Guangzhou, China 3 Quality Administration and Risk Control Center, Biozen Pharmaceutical, Xuchang, China
Received December 13, 2013 Revised February 1, 2014 Accepted February 1, 2014
Research Article
Extraction and isolation of flavonoid glycosides from Flos Sophorae Immaturus using ultrasonic-assisted extraction followed by high-speed countercurrent chromatography A method of ultrasonic-assisted extraction followed by high-speed countercurrent chromatography was established for the extraction and isolation of three flavonoid glycosides, i.e. rutin, narcissin, and nicotiflorin from Flos Sophorae Immaturus. The effects of ultrasonicassisted extraction factors for the main flavonoid compound (rutin) from Flos Sophorae Immaturus were optimized using Box–Behnken design combined with response surface methodology. The optimum conditions were determined as ultrasonic power 83% (600 W), solvent-to-material ratio 56:1, methanol concentration 82% v/v, and extraction time 60 min. Three bioactive flavonol glucosides, rutin, narcissin, and nicotiflorin were isolated from Flos Sophorae Immaturus using high-speed countercurrent chromatography. The separation was performed with a two-phase solvent system containing ethyl acetate/ n-butanol/methanol/water (4:0.9:0.2:5, v/v). Amounts of 87 mg of rutin, 10.8 mg of narcissin, and 1.8 mg of nicotiflorin were isolated from 302 mg of crude extract of Flos Sophorae Immaturus in a one-step separation within 160 min with purities of 99.3, 98.0, and 95.1%, respectively, as determined by HPLC with diode array detection. Their structures were characterized by UV, MS, and NMR spectroscopy. It was demonstrated that the established method was simple, fast, and convenient, which was feasible to extract and isolate active flavonoid glycosides from Flos Sophorae Immaturus. Keywords: Flavonoids / Flos Sophorae Immaturus / Isolation / Response surface methodology / Ultrasonic-assisted extraction DOI 10.1002/jssc.201301340
Additional supporting information may be found in the online version of this article at the publisher’s web-site
1 Introduction Flos Sophorae Immaturus, the dried flower buds of Sophora japonica L. (Leguminosae), is a well-known traditional Chinese medicine, which is produced in many provinces of China, such as Hebei, Shandong, Henan, and Jiangsu, etc. [1]. It has been commonly used to cure the diseases including hemafecia, hemorrhoids blood, blood flux, uterine bleeding, hematemesis, and so forth [2]. Studies have revealed that the main bioactive components in Flos Sophorae Immaturus are Correspondence: Dr. Xinjun Xu, Guangzhou Higher Education Mega Centre, No. 132, East Waihuan Road, Guangzhou 510006, China E-mail: [email protected] Fax: +86-20-39943000
Abbreviations: BBD, Box–Behnken design; DAD, diode array detection; HRE, heat-reflux extraction; HSCCC, highspeed countercurrent chromatography; RSM, response surface methodology; UAE, ultrasonic-assisted extraction C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
flavonoid compounds, including rutin, narcissin, and nicotiflorin (Fig. 1), which have been shown to possess a variety of biological activities, such as antiallergic, anti-inflammatory, antioxidative, free-radical scavenging, and antimutagenic activities [3–9]. In view of these pharmacological effects, it is important to use an efficient method for the extraction, isolation, and purification of the flavonoids from Flos Sophorae Immaturus. Conventional extraction methods, such as heating, boiling, or refluxing extraction can be used to extract flavonoids compounds; however, the disadvantages are the loss of flavonoids compounds due to oxidation, ionization, and hydrolysis during extraction as well as the long extraction time. Ultrasonic-assisted extraction (UAE), an inexpensive, simple, and efficient extraction technique, is very useful for the extraction of thermolabile and unstable compounds and has been widely used in recent years for the extraction of various compounds from natural products [10–12]. The mechanical Colour Online: See the article online to view Fig. 2 in colour. www.jss-journal.com
958
J. Sep. Sci. 2014, 37, 957–965
Z. Xie et al.
as the dried flower buds of S. japonica L. by Associate Professor Xu (Xinjun Xu, Sun Yat-sen University). Authentic standard of rutin was purchased from the National Institutes for Food and Drug Control (Batch number: 10080–200707). Methanol and acetonitrile used in HPLC analysis were of chromatographic grade and purchased from SK Chemicals (Seoul, Korea). Ultrapure water was obtained from a Milli-Q RG purification unit (ELGA Lab Water, UK). Other solvents were of analytical grade and purchased from Tianjin Damao Chemical Reagent Factory (Tianjin, China). DMSO-d6 , which was made by Cambridge Isotope Laboratories, was used as the solvent for NMR spectroscopy. Figure 1. Chemical structures of three flavonoid glycosides in Flos Sophorae Immaturus.
2.2 Apparatus action, cavitation, and thermal efficacies of ultrasound lead to the disruption of cell walls of plants and dissolution of components from cells; thus, UAE can improve the whole extraction efficacy [13]. Conventional silica column chromatography, preparative HPLC, and preparative TLC are usually utilized for the separation and purification of effective compounds from medicinal plants. However, these techniques are tedious, time consuming, waste solvent, and often cause irreversible adsorption of the sample onto the solid phase [14]. High-speed countercurrent chromatography (HSCCC) is a unique support-free liquid–liquid partition chromatography, which shows high recovery and efficiency as well as elimination of irreversible adsorption of the sample on the solid stationary phase in the conventional chromatographic column and low risk of sample denaturation [15–18]. It is obtaining increasing popularity as a means for the separation of pharmaceutical substances, especially for the preparative separation of bioactive components from natural products [19–24]. Solvent extraction has been traditionally used to extract flavonoids from Flos Sophorae Immaturus [25–27]. However, the extraction process is usually time consuming. Recently, several new extraction methods, such as ionic liquid extraction [28, 29] and far-IR-assisted solvent extraction [30], have been applied to extract flavonoids from Flos Sophorae Immaturus. These methods required special solvents or apparatus, which were not easily attainable. In the study, an effective and simple method, which was based on UAE for extraction and one-step HSCCC for the separation and purification of three flavonoids with high purities from Flos Sophorae Immaturus has been developed. The experimental conditions of UAE were optimized by response surface methodology (RSM) and the HSCCC procedure was also optimized.
2 Materials and methods 2.1 Reagents and materials Crude Flos Sophorae Immaturus was purchased from Baozhilin Pharmacy (Guangzhou, China) and authenticated C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
The UAE experiments were performed on an SB25–12DTD ultrasound bath (Scientz Biotechnical, Ningbo, China). Ultrasonic (40 KHz) energy was provided by a transducer that could be adjusted from 0 to 600 W. The HSCCC instrument was QuikPrepTM Chassis Mk5 HSCCC (the unit utilized had four identical coils of 3.2 mm od and 2.16 mm id. Each coil had a volume of approximately 120 mL. Quattro CCC was manufactured by AECS-QuikPrep, Bristol, UK) with a series II HPLC pump (SSI, USA), a sample injection valve with a 10 mL sample loop, and an SPD-10AVP detector (SHIMADZU, Japan). The work station was N2000 (Zhejiang University, China). HPLC analysis was performed with a Lab alliance HPLC system (1500 pump, AS1000 autosampler, UV6000 detector, SSI, USA). An electronic balance (KERN ABT 220–5DM, 0.1 mg, Germany), a DF-101S heating magnetic stirrer (Gongyi Yuhua) and an RE-300 rotational vacuum concentrator (Shanghai Yarong biochemistry instrument factory, China) were utilized for sample preparation.
2.3 Optimization of UAE extraction by RSM and HPLC analysis Box–Behnken design (BBD), an experimental design for RSM, is a class of rotatable or nearly rotatable secondorder design based on three-level incomplete factorial design [31, 32]. In this study, BBD was adopted for the optimization of the UAE process and rutin was selected as the model compound. On the basis of our preliminary single factor experimental results, methanol was selected as the solvent as it produced the highest extraction yield of rutin among methanol, ethanol, and water. The ranges of independent variables including ultrasonic power (A), solvent-to-material ratio (B), methanol concentration (C), and extraction time (D) were also investigated and the four factors, three levels BBD are summarized in Table 1. The main effects, interaction effects, and quadratic effects of the extraction condition were statistically evaluated by a 29-run experiment in a random order including five replicates at the central point shown in Table 2. The data from BBD were analyzed by multiple www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 957–965 Table 1. Independent variables and their levels for BBD
Variables
959
Table 3. Validation of the HPLC method for determination of rutin from Flos Sophorae Immaturus
Levels
A: ultrasonic power (%, 600 W) B: solvent-to-material ratio (mL/g) C: methanol concentration (%) D: extraction time (min)
−1
0
1
80 40 70 40
85 50 80 50
90 60 90 60
Table 2. Response surface of the BBD (uncoded) and results for extraction yields of rutin from Flos Sophorae Immaturus
Standard
Run
A
B
C
D
Y (%)
3 9 15 17 12 23 2 13 7 4 19 24 29 11 5 8 28 22 1 25 14 18 27 20 21 10 6 16 26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
80 80 85 80 90 85 90 85 85 90 80 85 85 80 85 85 85 85 80 85 85 90 85 90 85 90 85 85 85
60 50 40 50 50 40 40 40 50 60 50 60 50 50 50 50 50 60 40 50 60 50 50 50 40 50 50 60 50
80 80 90 70 80 80 80 70 70 80 90 80 80 80 70 90 80 80 80 80 70 70 80 90 80 80 90 90 80
50 40 50 50 60 60 50 50 60 50 50 60 50 60 40 60 50 40 50 50 50 50 50 50 40 40 40 50 50
20.23 19.51 18.56 18.81 19.05 19.83 17.71 16.98 18.58 17.70 17.79 20.38 19.56 19.96 18.32 19.92 20.14 19.62 17.94 20.29 17.70 13.86 19.81 18.79 18.15 18.90 19.86 20.12 19.67
Items
Value
Regression equation (y, peak area; X, concentration (g/mL)) Linear range (g/mL) Correlation coefficient, r Precision (RSD%) Repeatability (RSD%) Stability (RSD%) Recovery (n = 9)
y = (1E+08)X − 197 292 5.5∼220 0.9998 1.00 2.61 0.88 98.06% (RSD% = 2.35%)
regression coefficients were considered to be statistically significant when p values 99% purity was used, with sheath gas 40 arbitrary units; auxiliary gas 20 arbitrary units. The full scan covered the mass range from m/z 150 to 1000. Control and data acquisition were carried out with the Xcalibur 2.0 data system.
2.7 HSCCC separation procedure The multilayer coil column was first entirely filled with the upper phase (stationary phase), then the lower phase (mobile phase) was pumped into the column at the flow rate of 1.5 mL/min in the head-to-tail elution mode; the column rotated at 860 rpm. After the mobile phase front emerged and the hydrodynamic equilibrium was established in the column, the sample solution was injected; effluent was continuously monitored with a UV detector at 254 nm and collected into test tubes with a fraction collector set at 3 min for each tube. The fractions were analyzed by HPLC with diode array detection (DAD) and all fractions of the same purified compound were pooled, evaporated under reduced pressure at 40⬚C to give dried powder, and then dissolved in methanol for subsequent HPLC analysis.
2.8 HPLC analysis of the crude extract and isolated fractions The crude extract of Flos Sophorae Immaturus and each peak fraction from HSCCC were analyzed by HPLC. A Dikma Diamonsil C18 column (250 × 4.6 mm, 5 m) preceded by a Diamonsil C18 guard column (10 × 4.6 mm, 5 m; set at
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3 Results and discussion 3.1 Optimization of UAE conditions Several important parameters of UAE, such as ultrasonic power, solvent-to-material ratio, methanol concentration, and extraction time were optimized. Since rutin was the main effective compound in Flos Sophorae Immaturus, the extraction conditions of UAE were first optimized with 0.5 g of original sample and rutin was selected as the model compound. RSM is a collection of mathematical and statistical techniques based on the fit of a polynomial equation to the experimental data, which must describe the behavior of a dataset with the objective of making statistical previsions [32]. It can evaluate multiple parameters and their interactions and effectively optimize complex extraction procedures in a statistical way, thus reducing the number of experimental trials required [33]. Xu et al. [34] successfully applied RSM to optimize extraction of flavonoids from Fructus Sophorae. In our study, modeling of factors and response was performed by RSM using BBD to predict the highest extraction yield of rutin. The design matrix and the corresponding results
www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 957–965
961
of RSM experiments are shown in Table 2. Multiple linear regression analysis based on those data was performed to obtain the following predicted model: Yield = 19.895 − 0.686 × A + 0.549 × B + 0.900 × C + 0.280 × D − 0.576 × A × B + 1.487 × A × C − 0.076 × A × D − 0.208 × B × C − 0.232 × B × D − 0.051 × C × D − 1.095 ∗ A2 − 0.510 × B 2 − 1.213 × C 2 + 0.384 × D2
(3)
The analysis of variance indicated that the quadratic regression model was extremely statistically significant (p < 0.01) (Table 4), with a satisfactory coefficient of determination (R2 = 0.9444), which showed an agreement between the experimental results and the theoretical values predicted by the polynomial model. Moreover, a low coefficient of variation (CV 2.41%) value revealed a high degree of precision and great reliability of the experimental values. Therefore, the model was prominent and adequate to explain the actual relationship between the response and the significant variables in the range of experimental variables. The significance of each coefficient of Eq. 3 had also been examined to evaluate the effect of each variable and interactions between the variables. The smaller the p value was, the more significant the corresponding coefficient was. As shown in Table 4, items that had small p values (p < 0.05) including A, B, C, AB, AC, A2 , B2 , and C2 had a significant impact on extraction yield of rutin. Especially A, B, C, AC, A2 , and C2 , whose p values were altogether
J. Sep. Sci. 2014, 37, 957–965
Zhisheng Xie1,2 Yongjiang Sun3 Shingchung Lam1,2 Mingqian Zhao1,2 Zhikun Liang1,2 Xiaoxue Yu1,2 Depo Yang1,2 Xinjun Xu1,2 1 School
of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou, China 2 Laboratory for Quality Control of Traditional Chinese Medicine, Guangdong Technology Research Centre for Advanced Chinese Medicine, Guangzhou, China 3 Quality Administration and Risk Control Center, Biozen Pharmaceutical, Xuchang, China
Received December 13, 2013 Revised February 1, 2014 Accepted February 1, 2014
Research Article
Extraction and isolation of flavonoid glycosides from Flos Sophorae Immaturus using ultrasonic-assisted extraction followed by high-speed countercurrent chromatography A method of ultrasonic-assisted extraction followed by high-speed countercurrent chromatography was established for the extraction and isolation of three flavonoid glycosides, i.e. rutin, narcissin, and nicotiflorin from Flos Sophorae Immaturus. The effects of ultrasonicassisted extraction factors for the main flavonoid compound (rutin) from Flos Sophorae Immaturus were optimized using Box–Behnken design combined with response surface methodology. The optimum conditions were determined as ultrasonic power 83% (600 W), solvent-to-material ratio 56:1, methanol concentration 82% v/v, and extraction time 60 min. Three bioactive flavonol glucosides, rutin, narcissin, and nicotiflorin were isolated from Flos Sophorae Immaturus using high-speed countercurrent chromatography. The separation was performed with a two-phase solvent system containing ethyl acetate/ n-butanol/methanol/water (4:0.9:0.2:5, v/v). Amounts of 87 mg of rutin, 10.8 mg of narcissin, and 1.8 mg of nicotiflorin were isolated from 302 mg of crude extract of Flos Sophorae Immaturus in a one-step separation within 160 min with purities of 99.3, 98.0, and 95.1%, respectively, as determined by HPLC with diode array detection. Their structures were characterized by UV, MS, and NMR spectroscopy. It was demonstrated that the established method was simple, fast, and convenient, which was feasible to extract and isolate active flavonoid glycosides from Flos Sophorae Immaturus. Keywords: Flavonoids / Flos Sophorae Immaturus / Isolation / Response surface methodology / Ultrasonic-assisted extraction DOI 10.1002/jssc.201301340
Additional supporting information may be found in the online version of this article at the publisher’s web-site
1 Introduction Flos Sophorae Immaturus, the dried flower buds of Sophora japonica L. (Leguminosae), is a well-known traditional Chinese medicine, which is produced in many provinces of China, such as Hebei, Shandong, Henan, and Jiangsu, etc. [1]. It has been commonly used to cure the diseases including hemafecia, hemorrhoids blood, blood flux, uterine bleeding, hematemesis, and so forth [2]. Studies have revealed that the main bioactive components in Flos Sophorae Immaturus are Correspondence: Dr. Xinjun Xu, Guangzhou Higher Education Mega Centre, No. 132, East Waihuan Road, Guangzhou 510006, China E-mail: [email protected] Fax: +86-20-39943000
Abbreviations: BBD, Box–Behnken design; DAD, diode array detection; HRE, heat-reflux extraction; HSCCC, highspeed countercurrent chromatography; RSM, response surface methodology; UAE, ultrasonic-assisted extraction C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
flavonoid compounds, including rutin, narcissin, and nicotiflorin (Fig. 1), which have been shown to possess a variety of biological activities, such as antiallergic, anti-inflammatory, antioxidative, free-radical scavenging, and antimutagenic activities [3–9]. In view of these pharmacological effects, it is important to use an efficient method for the extraction, isolation, and purification of the flavonoids from Flos Sophorae Immaturus. Conventional extraction methods, such as heating, boiling, or refluxing extraction can be used to extract flavonoids compounds; however, the disadvantages are the loss of flavonoids compounds due to oxidation, ionization, and hydrolysis during extraction as well as the long extraction time. Ultrasonic-assisted extraction (UAE), an inexpensive, simple, and efficient extraction technique, is very useful for the extraction of thermolabile and unstable compounds and has been widely used in recent years for the extraction of various compounds from natural products [10–12]. The mechanical Colour Online: See the article online to view Fig. 2 in colour. www.jss-journal.com
958
J. Sep. Sci. 2014, 37, 957–965
Z. Xie et al.
as the dried flower buds of S. japonica L. by Associate Professor Xu (Xinjun Xu, Sun Yat-sen University). Authentic standard of rutin was purchased from the National Institutes for Food and Drug Control (Batch number: 10080–200707). Methanol and acetonitrile used in HPLC analysis were of chromatographic grade and purchased from SK Chemicals (Seoul, Korea). Ultrapure water was obtained from a Milli-Q RG purification unit (ELGA Lab Water, UK). Other solvents were of analytical grade and purchased from Tianjin Damao Chemical Reagent Factory (Tianjin, China). DMSO-d6 , which was made by Cambridge Isotope Laboratories, was used as the solvent for NMR spectroscopy. Figure 1. Chemical structures of three flavonoid glycosides in Flos Sophorae Immaturus.
2.2 Apparatus action, cavitation, and thermal efficacies of ultrasound lead to the disruption of cell walls of plants and dissolution of components from cells; thus, UAE can improve the whole extraction efficacy [13]. Conventional silica column chromatography, preparative HPLC, and preparative TLC are usually utilized for the separation and purification of effective compounds from medicinal plants. However, these techniques are tedious, time consuming, waste solvent, and often cause irreversible adsorption of the sample onto the solid phase [14]. High-speed countercurrent chromatography (HSCCC) is a unique support-free liquid–liquid partition chromatography, which shows high recovery and efficiency as well as elimination of irreversible adsorption of the sample on the solid stationary phase in the conventional chromatographic column and low risk of sample denaturation [15–18]. It is obtaining increasing popularity as a means for the separation of pharmaceutical substances, especially for the preparative separation of bioactive components from natural products [19–24]. Solvent extraction has been traditionally used to extract flavonoids from Flos Sophorae Immaturus [25–27]. However, the extraction process is usually time consuming. Recently, several new extraction methods, such as ionic liquid extraction [28, 29] and far-IR-assisted solvent extraction [30], have been applied to extract flavonoids from Flos Sophorae Immaturus. These methods required special solvents or apparatus, which were not easily attainable. In the study, an effective and simple method, which was based on UAE for extraction and one-step HSCCC for the separation and purification of three flavonoids with high purities from Flos Sophorae Immaturus has been developed. The experimental conditions of UAE were optimized by response surface methodology (RSM) and the HSCCC procedure was also optimized.
2 Materials and methods 2.1 Reagents and materials Crude Flos Sophorae Immaturus was purchased from Baozhilin Pharmacy (Guangzhou, China) and authenticated C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
The UAE experiments were performed on an SB25–12DTD ultrasound bath (Scientz Biotechnical, Ningbo, China). Ultrasonic (40 KHz) energy was provided by a transducer that could be adjusted from 0 to 600 W. The HSCCC instrument was QuikPrepTM Chassis Mk5 HSCCC (the unit utilized had four identical coils of 3.2 mm od and 2.16 mm id. Each coil had a volume of approximately 120 mL. Quattro CCC was manufactured by AECS-QuikPrep, Bristol, UK) with a series II HPLC pump (SSI, USA), a sample injection valve with a 10 mL sample loop, and an SPD-10AVP detector (SHIMADZU, Japan). The work station was N2000 (Zhejiang University, China). HPLC analysis was performed with a Lab alliance HPLC system (1500 pump, AS1000 autosampler, UV6000 detector, SSI, USA). An electronic balance (KERN ABT 220–5DM, 0.1 mg, Germany), a DF-101S heating magnetic stirrer (Gongyi Yuhua) and an RE-300 rotational vacuum concentrator (Shanghai Yarong biochemistry instrument factory, China) were utilized for sample preparation.
2.3 Optimization of UAE extraction by RSM and HPLC analysis Box–Behnken design (BBD), an experimental design for RSM, is a class of rotatable or nearly rotatable secondorder design based on three-level incomplete factorial design [31, 32]. In this study, BBD was adopted for the optimization of the UAE process and rutin was selected as the model compound. On the basis of our preliminary single factor experimental results, methanol was selected as the solvent as it produced the highest extraction yield of rutin among methanol, ethanol, and water. The ranges of independent variables including ultrasonic power (A), solvent-to-material ratio (B), methanol concentration (C), and extraction time (D) were also investigated and the four factors, three levels BBD are summarized in Table 1. The main effects, interaction effects, and quadratic effects of the extraction condition were statistically evaluated by a 29-run experiment in a random order including five replicates at the central point shown in Table 2. The data from BBD were analyzed by multiple www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 957–965 Table 1. Independent variables and their levels for BBD
Variables
959
Table 3. Validation of the HPLC method for determination of rutin from Flos Sophorae Immaturus
Levels
A: ultrasonic power (%, 600 W) B: solvent-to-material ratio (mL/g) C: methanol concentration (%) D: extraction time (min)
−1
0
1
80 40 70 40
85 50 80 50
90 60 90 60
Table 2. Response surface of the BBD (uncoded) and results for extraction yields of rutin from Flos Sophorae Immaturus
Standard
Run
A
B
C
D
Y (%)
3 9 15 17 12 23 2 13 7 4 19 24 29 11 5 8 28 22 1 25 14 18 27 20 21 10 6 16 26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
80 80 85 80 90 85 90 85 85 90 80 85 85 80 85 85 85 85 80 85 85 90 85 90 85 90 85 85 85
60 50 40 50 50 40 40 40 50 60 50 60 50 50 50 50 50 60 40 50 60 50 50 50 40 50 50 60 50
80 80 90 70 80 80 80 70 70 80 90 80 80 80 70 90 80 80 80 80 70 70 80 90 80 80 90 90 80
50 40 50 50 60 60 50 50 60 50 50 60 50 60 40 60 50 40 50 50 50 50 50 50 40 40 40 50 50
20.23 19.51 18.56 18.81 19.05 19.83 17.71 16.98 18.58 17.70 17.79 20.38 19.56 19.96 18.32 19.92 20.14 19.62 17.94 20.29 17.70 13.86 19.81 18.79 18.15 18.90 19.86 20.12 19.67
Items
Value
Regression equation (y, peak area; X, concentration (g/mL)) Linear range (g/mL) Correlation coefficient, r Precision (RSD%) Repeatability (RSD%) Stability (RSD%) Recovery (n = 9)
y = (1E+08)X − 197 292 5.5∼220 0.9998 1.00 2.61 0.88 98.06% (RSD% = 2.35%)
regression coefficients were considered to be statistically significant when p values 99% purity was used, with sheath gas 40 arbitrary units; auxiliary gas 20 arbitrary units. The full scan covered the mass range from m/z 150 to 1000. Control and data acquisition were carried out with the Xcalibur 2.0 data system.
2.7 HSCCC separation procedure The multilayer coil column was first entirely filled with the upper phase (stationary phase), then the lower phase (mobile phase) was pumped into the column at the flow rate of 1.5 mL/min in the head-to-tail elution mode; the column rotated at 860 rpm. After the mobile phase front emerged and the hydrodynamic equilibrium was established in the column, the sample solution was injected; effluent was continuously monitored with a UV detector at 254 nm and collected into test tubes with a fraction collector set at 3 min for each tube. The fractions were analyzed by HPLC with diode array detection (DAD) and all fractions of the same purified compound were pooled, evaporated under reduced pressure at 40⬚C to give dried powder, and then dissolved in methanol for subsequent HPLC analysis.
2.8 HPLC analysis of the crude extract and isolated fractions The crude extract of Flos Sophorae Immaturus and each peak fraction from HSCCC were analyzed by HPLC. A Dikma Diamonsil C18 column (250 × 4.6 mm, 5 m) preceded by a Diamonsil C18 guard column (10 × 4.6 mm, 5 m; set at
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3 Results and discussion 3.1 Optimization of UAE conditions Several important parameters of UAE, such as ultrasonic power, solvent-to-material ratio, methanol concentration, and extraction time were optimized. Since rutin was the main effective compound in Flos Sophorae Immaturus, the extraction conditions of UAE were first optimized with 0.5 g of original sample and rutin was selected as the model compound. RSM is a collection of mathematical and statistical techniques based on the fit of a polynomial equation to the experimental data, which must describe the behavior of a dataset with the objective of making statistical previsions [32]. It can evaluate multiple parameters and their interactions and effectively optimize complex extraction procedures in a statistical way, thus reducing the number of experimental trials required [33]. Xu et al. [34] successfully applied RSM to optimize extraction of flavonoids from Fructus Sophorae. In our study, modeling of factors and response was performed by RSM using BBD to predict the highest extraction yield of rutin. The design matrix and the corresponding results
www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 957–965
961
of RSM experiments are shown in Table 2. Multiple linear regression analysis based on those data was performed to obtain the following predicted model: Yield = 19.895 − 0.686 × A + 0.549 × B + 0.900 × C + 0.280 × D − 0.576 × A × B + 1.487 × A × C − 0.076 × A × D − 0.208 × B × C − 0.232 × B × D − 0.051 × C × D − 1.095 ∗ A2 − 0.510 × B 2 − 1.213 × C 2 + 0.384 × D2
(3)
The analysis of variance indicated that the quadratic regression model was extremely statistically significant (p < 0.01) (Table 4), with a satisfactory coefficient of determination (R2 = 0.9444), which showed an agreement between the experimental results and the theoretical values predicted by the polynomial model. Moreover, a low coefficient of variation (CV 2.41%) value revealed a high degree of precision and great reliability of the experimental values. Therefore, the model was prominent and adequate to explain the actual relationship between the response and the significant variables in the range of experimental variables. The significance of each coefficient of Eq. 3 had also been examined to evaluate the effect of each variable and interactions between the variables. The smaller the p value was, the more significant the corresponding coefficient was. As shown in Table 4, items that had small p values (p < 0.05) including A, B, C, AB, AC, A2 , B2 , and C2 had a significant impact on extraction yield of rutin. Especially A, B, C, AC, A2 , and C2 , whose p values were altogether
