ORIGINAL ARTICLE Year : 2014  Volume : 10  Issue : 39  Page : 292303 Optimization of simultaneous ultrasonicassisted extraction of watersoluble and fatsoluble characteristic constituents from Forsythiae Fructus Using response surface methodology and highperformance liquid chromatography YongGang Xia, BingYou Yang, Jun Liang, Di Wang, Qi Yang, HaiXue Kuang School of Pharmacy, Key Laboratory of Chinese Materia Medica (Heilongjiang University of Chinese Medicine), Ministry of Education, Harbin, 150040, P.R., China Correspondence Address: Background: The compounds (+)pinoresinolβglucoside (1) forsythiaside, (2) phillyrin (3) and phillygenin (4) were elucidated to be the characteristic constituents for quality control of Forsythiae Fructus extract by chromatographic fingerprint in 2010 edition of Chinese Pharmacopoeia due to their numerous important pharmacological actions. It is of great interest to extract these medicinally active constituents from Forsythiae Fructus simultaneously. Materials and Methods: In this study, a new ultrasoundassisted extraction (UAE) method was developed for the simultaneous extraction of biological components 14 in Forsythiae Fructus. The quantitative effects of extraction time, ratio of liquid to solid, extraction temperature, and methanol concentration on yield of these four important biological constituents from Forsythiae Fructus were investigated using response surface methodology with BoxBehnken design. The compounds 14 extracted by UAE were quantitative analysis by highperformance liquid chromatographyphotodiode array detect (HPLCPAD), and overall desirability (OD), the geometric mean of the contents of four major biological components, was used as a marker to evaluate the extraction efficiency. Results: By solving the regression equation and analyzing 3D plots, the optimum condition was at extraction temperature 70°C, time 60 min, ratio of liquid to solid 20, and methanol concentration 76.6%. Under these conditions, extraction yields of compounds 14 were 2.92 mg/g, 52.10 mg/g, 0.90 mg/g and 0.57 mg/g, respectively, which were in good agreement with the predicted OD values. In order to achieve a similar yield as UAE, soxhlet extraction required at least 6 h and maceration extraction required much longer time of 24 h. Established UAE method has been successfully applied to sample preparation for the quality control of Forsythiae Fructus. Additionally, a quadrupole timeofflight mass spectrometry was applied to the structural confirmation of analytes from the complex matrices acquired by UAE. Conclusion: The results indicated that UAE is an effective alternative method for extracting bioactive constituents, which may facilitate a deeper understanding of the extract of active constituents in Forsythiae Fructus from the raw material to its extract for providing the theoretical references.
Introduction Forsythiae Fructus, the fruit of Forsythia suspensa (Oleaceae), [1] is one of the most famous Chinese herbal medicines listed in the Pharmacopoeia of the People's Republic of China. [2] Forsythiae Fructus has been utilized as an antipyretic, detoxicant, antioxidant and antiinflammatory agent for the treatment of various infectious diseases. [3],[4] Furthermore, Forsythiae Fructus extract have been used for a long time as traditional Asian medicines to treat gonorrhea, erysipedas, inflammation, and pharyngitis. [5],[6] More than 40 Chinese medicinal preparations containing Forsythiae Fructus or its extract are listed in 2010 edition of Chinese pharmacopoeia, such as ''Shuanghuanglian Injection,'' ''Kangbingdu oral solution,'' ''Yin Qiao tablet,'' etc., [2] Many studies have elucidated that caffeoyl glycosides, and lignans are the main bioactive components responsible for the various biological activities of Forsythiae Fructus. [7],[8] For example, forsythoside is responsible for the antibacterial and the antioxidant activities of the herb. [3],[9],[10] Phillyrin could enhance immunological function and alleviate delayed hypersensitivity in mice. [6],[11] Phillygenin and (+)pinoresinolβglucoside showed their protective effects against peroxynitriteinduced oxidative stress in LLCPK1 cells. [12] Therefore, (+) pinoresinolβglucoside, (1) forsythiaside, (2) phillyrin (3) and phillygenin (4) possess important pharmacological activities and may become promising phytopharmaceuticals. Extraction and determination of the compounds 14 in Forsythiae Fructus are also essential for the quality evaluation of the traditional Chinese medicine. [8],[13],[14],[15] By our previous studies, the compounds 14 were elucidated to be the characteristic constituents for quality control of Forsythiae Fructus extract by chromatographic fingerprint in 2010 edition of Chinese pharmacopoeia due to their numerous important pharmacological actions. [8] Besides, a number of studies found that the quality of Forsythiae Fructus often varied considerably according to it's the harvest season, storage, geographic origin or other growing conditions. [8],[13],[14],[15] Though determining the contents of 14 have been widely described no study has been reported on the extraction of these four ''characteristic constituents'' simultaneously. Conventional soxhlet extraction (SE) and maceration extraction (ME) are very timeconsuming and require relatively high extraction temperature for SE, which have been used for several decades. [16],[17] Thus, there is an increasing demand for a novel extraction technique with the shortened extraction time, reduced extraction temperature, and increased extraction efficiency. Ultrasonic technique is being used widely in analytical chemistry, facilitating different steps in the analytical process, particularly in sample preparation. [18],[19],[20] Ultrasonicassisted extraction (UAE) is an expeditious, inexpensive, and efficient alternative to traditional extraction techniques. [21],[22] UAE may enhance the extraction efficiency due to disruption of cell walls, particlesize reduction, and enhancing mass transfer of the cell contents as a result of cavitation bubble collapse. [20],[21],[22] UAE of phillyrin from the seeds of F. suspensa has been performed by classical univariate approach (onevariableatatime), [23] but this method cannot determine the interactions between parameters and find the most suitable UAE condition, so some "experimental design" was adopted to detect the influencing factors while the number of trials can be kept to a minimum. So far, there is no report for the simultaneous extraction of compounds 14 using ultrasonic technique coupled with BoxBehnken statistical design (BBD). BBD provides efficient solutions compared with a threelevel fullfactorial design, reducing the number of required experiments by confounding higherorder interactions, which becomes more significant as the number of factors increases. [24],[25],[26] It provides information about the relative significance of main effects, as well as information about interaction effects that cannot be predicted by univariate techniques. The model results are easily interpreted and visualized in response surface plots. [27],[28] The goals of this study are to optimize the UAE condition for extracting these four characteristic compounds 14 from Forsythiae Fructus coupled with BBD. The overall desirability (OD), the geometric mean of the contents of four major biological components was used as a marker to evaluate the extraction efficiency. [29],[30] BBD was applied to fit and to exploit a mathematical model representing the relationship between the responses and the variables (i.e, extraction time, ratio of liquid to solid, temperature, and methanol concentration). This study may facilitate a deeper understanding of the extract of these active constituents from the raw material to its extract and provide theoretical references for industrial production of Forsythiae Fructus extract. Materials and Methods Materials and reagents The unripe fruits of F. suspensa were collected in March 2010 from Henan Province, China and identified by Prof. Wang ZhenYue of Heilongjiang University of Chinese Medicine, The voucher specimen (2010013) was deposited at Herbarium of Heilongjiang University of Chinese Medicine, Harbin, China. The crude drug was pulverized into powder form by a disintegrator (Weinengda Instrument Company, Lanxi, China), and then sieved with stainless steel sieves to classify the particle size. The powdered sample was kept in a dry and dark place until use. All organic solvents used for extraction were of analytical grade and purchased from Tianjin Chemical Factory, Tianjin, China. The highperformance liquid chromatography (HPLC) grade methanol was purchased from Dikama Technology Corporation (Richmond Hill, USA). Deionized water was purified by MilliQ system (Millipore, Bedford, MA, USA). Standards, namely (+)pinoresinolβglucoside, forsythiaside, phillyrin, and phillygenin were isolated by the author from the fruits of F. suspensa. Structures of the standards are shown in [Figure 1].{Figure 1} UAE For the UAE experiments, an ultrasonic bath was used as an ultrasound source. The bath (KQ500DB, Kunshan Ultrasound Co. Ltd., China) was a rectangular container (540 mm × 320 mm × 350 mm). The bath power rating was 500 W on the scale of 410. The extraction temperature was controlled and maintained at the desired value by circulating external water from a thermostatic water bath into the cleaning bath. The sample beakers were immersed into the ultrasonic cleaning bath for irradiation under different extraction conditions including solvents: Ethanol, methanol, acetonitrile, and water; percentage of methanol in water of 20100%; solvent to solid ratio of 2080 mL/g; temperature of 3070°C; extraction time of 1560 min; particle size of 4060 mesh. Finally, extracts were filtered off through 0.22 μm membrane filter and the filtrate was collected for HPLC analyses. All samples were prepared and analyzed in triplicate. ME ME method was performed with 1.0 g (50 mesh) of dried samples and 20 mL of 76.6% methanol extracted 3 times, each for 24 h, and then mixed them at room temperature. The extracts were combined and concentrated by a rotary vacuum evaporator. All solutions were filtered through 0.22 μm membrane filter before direct injection into the HPLC system. All samples were prepared and analyzed in triplicate. SE SE was performed in a soxhlet apparatus. The powders of 1.0 g (50 mesh) and 20 mL of 76.6% methanol were placed into the soxhlet apparatus. The exhaustive extraction was performed for 6 h at 70°C. All solutions were filtered with 0.22 μm membrane filter before the HPLC analysis. All samples were prepared and analyzed in triplicate. HPLC analysis The analyses were performed using waters e2695 liquid chromatography system, equipped with a quaternary solvent delivery system, a waters 600 controller, two waters 600 pumps, a 2695 auto sampler, a waters 2998 photodiode array detector, and a waters 2695 column oven. The separation was carried out on waters symmetry C 18 column (4.6 mm × 150 mm, 5 μm). The gradient elution was employed using solvent A (MeOH) and solvent B (water) at 30°C; the gradient program was used as follows: initial 010 min, linear change from A to B (10:90, v/v) to AB (25:75, v/v); 1040 min, linear change to AB (35:65, v/v); 4060 min, linear change to AB (60:40, v/v). The flow rate was set at 1.0 mL/min, and the injection volume was 10 μL. Due to the different UV characteristic of these components, the detection wavelength was set at 332 and 270 nm for quantitative analysis of the caffeoyl glycoside and lignans, respectively. Quantification was carried out by the integration of the peak using external standard method by means of a six point calibration curve. The regression equations, correlation coefficient and linear range are listed in our previous published papers. [8] The extraction yields of (+)pinoresinolβglucoside, forsythiaside, phillyrin, and phillygenin by UAE, ME and SE methods were calculated using the following equation: extraction yields (%) = weight of compounds extracted (g)/weight of dried sample (g) ×100%. Qualitative analysis based on ultra performance liquid chromatographyquadrupole timeofflight The UPLCMS analysis was performed on a Waters ACQUITY UPLC system (Waters Corporation, Milford, USA) coupled with a Waters Xevo QTOF equipped with electrospray ionization. For the reversedphase UPLC analysis, the ACQUITY UPLC TM HSS C18 column (100 mm × 2.1 mm i.d., 1.7 μm, Waters Corp, Milford, USA) was used. The column temperature was maintained at 40°C; the flow rate of the mobile phase was 0.40 mL/min; the injection volume was fixed at 2.0 μL. Mobile phase A consisted of 0.1% formic acid in methanol while mobile phase B consisted of 0.1% formic acid in water. The column was eluted with a linear gradient of 218% B over initial to 6.0 min, 1824% B over 6.013.0 min, 2429% B over 13.018.0 min, 2942% B over 18.023.0 min, 4250% B over 2324.5 min, 5098% B over 24.526 min, and returned to 2% B for 1.0 min and then held for 1.0 min at an eluent flow rate of 0.40 mL/min. For the UPLCXevo QTOF analysis, the mass spectrometric fullscan data were acquired in the negative ion mode from 100 to 1000 Da with a 0.1 s scan time. Other conditions were as follows: capillary voltage of 2.4 kV, desolvation temperature of 400°C, sample cone voltage of 25 V, extraction cone voltage of 4.0 V, collision energy of 30 eV, source temperature of 120°C, cone gas flow of 50 L/h and desolvation gas flow of 400 L/h for negative ion mode. Data were centroided, and mass was corrected during acquisition using an external reference (LockSpray TM ) consisting of a 0.2 ng/ml solution of leucineenkephalin infused at a flow rate of 20 μL/min via a lockspray interface, generating a reference ion for negative ion mode ([M − H]− =554.2615) to ensure the accuracy during the MS analysis. Experimental design After determining the preliminary range of the extraction variables through singlefactor test, a 29run BBD consisting of four variables at three levels was established to optimize the four marker constituents extraction conditions from F. suspensa, including extraction temperature, extraction time, solvent concentration, and ratio of liquid to solid, which significantly influenced the extraction yield. As shown in [Table 1], the four factors chosen for this study were designated as X1 , X2 , X3 , and X4 and prescribed into three levels, coded + 1, 0, −1 for high, intermediate and low value, respectively. The four variables were coded according to the following equation:{Table 1} [INLINE:1] Where X i is a coded value of the variable; x i is the actual value of the variable; x 0 is the actual value of the independent variable at the center point, and is the step change of the variable. A 2 nd order polynomial model corresponding to the BBD was fitted to correlate the relationship between the independent variables and the response (extraction yield, polysaccharide yield and uronic acid yield) to predict the optimized conditions. The computergenerated quadratic model is given as: [INLINE:2] Where Y is the predicted response; X i and X j are the coded independent variables; β0 is the intercept coefficient; βi is the linear coefficient; βii is the squared coefficient, and βij is the interaction coefficient. Analysis of the experimental design data and calculation of predicted responses were carried out using Design Expert software (version 8.0, StatEase, Inc., Minneapolis, USA). Statistical analyses DesignExpert 8.0, trial version was used for the analysis of variance (ANOVA) analysis of the experimental data obtained. The quality of the fit of the polynomial model equation was expressed by the coefficient of determination R2, and the values of adjustedR2 of models were evaluated to check the model adequacies. The significance of each term in the equation is to estimate the goodness of fit in each case. The ANOVA tables were generated, and effect and the regression coefficients of individual linear, quadratic and interaction terms were determined. The P values of less than 0.05 were considered to be statistically significant. The regression coefficients were then used to make the statistical calculation to generate contour and dimensional maps from the regression models. Results and Discussion Singlefactor experiments Optimizing UAE conditions should consider the interaction of different extraction factors and the linear relationship between response and variables. In order to reveal the complicated interaction and relationship, a statistical analysis method, BBD was selected to optimize UAE parameters. The OD, the geometric mean of the contents of 4 target compounds was used as a marker to evaluate the extraction efficiency. [29],[30] Before BBD optimizing UAE parameters, a preliminary experiment has been performed. In the preliminary experiment, extraction factors including solvent type, particle size, ultrasonic power, temperature, time, methanol concentration and ratio of liquid to solid were studied. The four different solvents (ethanol, methanol, acetonitrile and water) were tested under the same conditions: sample of 1.0 g, temperature of 70°C, solvent to solid ratio of 20 mL/g, ultrasound frequency 60 kHz, extraction time of 30 min and particle size of 50 meshes. The results showed that methanol gave the highest extraction yields, followed by ethanol, acetonitrile and water. The different extraction efficiencies of these solvents could be attributed to their different polarities and viscosities. The effect of ultrasonic frequency on UAE was explored with solvent to solid ratio at 20 ml/g and other conditions fixed as mentioned previously (sample: 1.0 g, solvent: methanol, temperature: 70°C, extraction time: 30 min and particle size: 50 mesh). Results indicated an obvious increase of extraction yields when the ultrasound frequency was increased from 40 kHz to 60 kHz. When the ultrasound frequency was increased from 60 kHz to 100 kHz, no significant differences between the extraction yields were detected (P > 0.05). Therefore, an ultrasound frequency of 60 kHz should be optimum to extract these four characteristic constituents. Particles with the size 40, 50, 60 meshes were collected for the study. The result showed that there were no significant differences from 40 to 60 meshes for the extraction yields. Therefore, 50 meshes were selected for following experiments. Thus, the other four factors, extraction time, ratio of liquid to solid, extraction temperature, and methanol concentration were selected as BBD factors and the ultrasound frequency and particle size were set at 60 kHz and 50 mesh, respectively. Statistical analysis and the model fitting The effects of four process variables (i.e. extraction time [X1 ], ratio of liquid to solid [X2 ], extraction temperature [X3 ] and methanol concentration [X4 ]) were studied during the experimentation. An optimum process should be investigated in order to obtain high extraction yield. The results of 29 runs using BBD design are presented in [Table 2] that includes the design, experimental values and the OD values. BBD with four factors and three levels, including five replicates at the center point, was used to fit a 2 nd order response surface in order to optimize the extraction conditions. The five center point runs were carried out to measure the process stability and inherent variability.{Table 2} By applying the multiple regression analysis on the experimental data, the results of the BBD were fitted with a 2 nd order polynomial equation. Thus, a mathematical regression model for a total content fitted in the coded factors was given as following: [INLINE:3] where Y was the response, that was the OD of four phenolic compounds contents and X1 , X2 , X3 and X4 were the coded values of the test variables extraction time, ratio of liquid to solid, extraction temperature and methanol concentration, respectively. The significance of each coefficient was determined using P value, which is used as a tool to check the significance of each coefficient and the interaction strength between each independent variable. [31] If P value is the smaller it is the bigger the significance of the corresponding coefficient. The significance of the F value depends on the number of degrees of freedom in the model and is shown in the P value column (95% confidence level). In general, the effects lower than 0.05 are significant. The ANOVA [Table 3] showed that this regression model was highly significant (P < 0.0001) with F value of 13.65. The F value of 1.05 for lack of fit implies that it is not significant comparing to the pure error. The fitness of the model was further confirmed by a satisfactory value of determination coefficient, which was calculated to be 0.9221, indicating that 92.21% of the variability in the response could be predicted by the model [Table 3]. The value of the adjusted determination coefficient (adjusted R2 = 0.8545) also confirmed that the model was highly significant. The "Pred RSquared" of 0.6731 is in reasonable agreement with the "Adj RSquared" of 0.8545 [Table 4]. "Adeq Precision" measures the signal to noise ratio. A ratio greater than 4 is desirable. The ratio of 12.314 indicates an adequate signal. This model can be used to navigate the design space. As shown in [Table 3], the variable with the largest effect was the X4 (P < 0.0001), followed by the other linear terms of X2 and X3 (P < 0.01). The other term coefficients were not significant (P > 0.05).{Table 3}{Table 4} Analysis of response surfaces The 3D plot in [Figure 2]a showed the effects of extraction time (X1 ) and ratio of liquid to solid (X2 ) on extraction yield (Y) at fixed extraction temperature (50°C) and methanol concentration (60%). There was a rapid rise in extraction yield with an increase in extraction time (X1 ) and ratio of liquid to solid (X2 ). [Figure 2]b show the 3D response surface plot at varying extraction time (X1 ) and extraction temperature (X3 ) at fixed the ratio of liquid to solid (50:1) and methanol concentration (60%). There was a slight upsurge in extraction yield (Y1 ) with an increase in extraction time (X1 ) and extraction temperature (X3 ). Like [Figure 2]ac showed the effects of extraction time (X1 ) and methanol concentration (X4 ) on extraction yield (Y) at fixed extraction temperature (50°C) and the ratio of liquid to solid (50:1). The extraction yield increased rapidly within the methanol concentration from 20% to 76.6%, but when beyond 76.6%, the extraction yield reached the plateau region where the yield was maximized and did not increase any more, and the yield increased slightly with the increase of the extraction temperature. In [Figure 2]d, when the 3D response surface plot were developed for the extraction yield with varying ratio of liquid to solid (X2 ) and extraction temperature (X3 ) at fixed the extraction time (37.5 min) and methanol concentration (60%). It indicated that a rapid rise in extraction yield with an increase in the ratio of liquid to solid (X2 ) and extraction temperature (X3 ). In [Figure 2]e, when the 3D response surface plot was developed for the extraction yield with varying ratio of liquid to solid (X2 ) and methanol concentration (X4 ) at fixed extraction temperature (50°C) and extraction time (37.5 min). It indicated methanol concentration (X4 ) exhibited an important effect on the extraction yield, which did not continue to increase significantly until the methanol concentration was over 76.6%. And further increase in methanol concentration, results in slow decrease in the extraction yield. With respect to the ratio of liquid to solid (X2 ), the influence of this parameter was not as significant as that of methanol concentration. As shown in this figure, extraction yield showed a slight rise with an increase in the ratio of liquid to solid (X2 ). The 3D response surface plot based on independent variables extraction temperature (X3 ) and methanol concentration (X4 ) are shown in [Figure 2]f while the other two independent variables, extraction time and ratio of liquid to solid were kept at 37.5 min and 50, respectively. It can be seen that the extraction yield increased with the increase of methanol concentration (X4 ) from 20% to 76.6%, then dropped slightly from 76.6% to 100% and the yield increased very gently with the increase of the ratio of liquid to solid from 20 to 80.{Figure 2} Optimization of extracting parameters and validation of the model By solving the regression equation (3) and analyzing 3D plots, it can be concluded that the optimal extraction conditions from Forsythiae Fructus are extraction temperature 70°C, time 60 min, ratio of liquid to solid 20, and methanol concentration 76.6%. Among the four extraction parameters that have been studied, methanol concentration (X4 ) was the most significant factor that affects the yield followed by the ratio of liquid to solid (X2 ), extraction temperature (X3 ) and extraction time (X1 ) according to the regression coefficients significance of the quadratic polynomial model and gradient of slope in the 3D response surface plot. A possible explanation is that the four characteristic compounds 14 possess different polarity and solubility; hence control of methanol concentration is very important to extract them simultaneously. However, with the increase in the ratio of liquid to solid, the extraction yield of compounds 14 is down gradually, which is in agreement with previous results. [23] This phenomenon could be explained that some substances, which have dissolved in the extracting solvent, could increase the solubility of compounds 14. If the volume of extracting solvent was increased markedly, the concentrations of those substances dissolved in the extracting solvent would also decrease obviously, and the ability to increase the solubility of compounds 14 would also decrease distinctly, which results in a decreased extracting yield of compounds 14. The predicted extraction yield (OD) that was given by the Design Expert software under the above conditions was 0.672. The optimum extraction conditions were applied to 3 independent replicates to verify the prediction from the model. The mean experimental extraction yield of (+)pinoresinolβglucoside, (1) forsythiaside, (2) phillyrin, (3) and phillygenin, (4) were 2.92 mg/g, 52.10 mg/g, 0.90 mg/g and 0.57 mg/g, respectively, namely OD of 0.670, which were in good agreement with the predicted value of the model equation, confirming that the response model was adequate for the optimization. Besides, extraction temperature (30, 50, 70, and 90°C), extraction time (20, 40, 60, and 80 min), ratio of liquid to solid (10, 20, 30 and 40), and methanol concentration (25%, 50%, 75% and 100%) were validated using univariate method. When one of the parameters, including temperature, time, ratio of liquid to solid and methanol concentration was optimized, the others were set at the predication optimization value (temperature 70°C, time 60 min, ratio of liquid to solid 20, and methanol concentration 76.6%). The results showed the optimization values of temperature, time and the ratio of liquid to solid were the same as the results of BBD [Figure 3]ac, but methanol concentration of univariate method was 75% [Figure 3]d because 75% ethanol concentration was a real experiment result rather than predicted value.{Figure 3} Comparison of UAE with ME and SE The selection of an extraction method mainly depends on the advantages and disadvantages of the processes, such as extraction yield, production cost complexity, environmental friendliness, and safety. In general, ME and SE are the most frequently used traditional extraction methods. The results of extraction yields of UAE, ME and SE listed in [Table 5] showed that ME obtained the highest yields of forsythiaside and phillygenin, and UAE got the highest yield of (+)pinoresinolβglucoside and phillyrin. However, for total yield of these four characteristic constituents, ME and UAE had no significant differences (P > 0.05). Therefore, the UAE was more suitable for simultaneous extraction of (+)pinoresinolβglucoside, forsythiaside, phillyrin, and phillygenin than ME and SE method, due to the fact that UAE can save a lot of time, reduce solvent consumption and bring the higher yield of total contents.{Table 5} Application of UAE method to Forsythia samples coupled with HPLC for quality control The developed UAE sample preparation method was then successfully applied to simultaneously determine the four characteristic components in 12 batches of samples obtained from different sources under the optimal UAE condition and quantified by HPLC, which was a validated and feasible method for evaluating and controlling quality of Forsythiae Fructus. The sample HPLC chromatography can be seen in [Figure 4]. A large variation of the contents of the four characteristic compounds was found among these samples from different origins and the harvest time. The percentage of forsythiaside ranging from 1.09% to 17.82% in Forsythiae Fructus was the highest, followed by (+)pinoresinolβglucoside from 0.08% to 1.09%, phillyrin from 0.02% to 0.17% and phillygenin from 0.013% to 0.06%. The variation is possibly attributed to several factors, such as plant origins, harvesting time, storage conditions, etc., [8],[15] The variation in contents of these characteristic biological components may cause changes in clinical efficacy. To ensure the quality of Forsythiae Fructus extract, this suggests that each procedure involved should be standardized.{Figure 4} Qualitative analysis of four characteristic constituents extracted by UAE based on quadrupole timeofflight mass spectrometry UAE method has been successfully applied to sample preparation for the quality control of Forsythiae Fructus. Being a high resolution mass spectrum, QTOFMS could perform accurate mass measurement, which gives elemental composition of parent and fragment ions. Furthermore, the insource collision induced dissociation technique was applied in our experiment to acquire sufficient structure information from QTOFMS. The four targeted chromatographic peaks (14) were unambiguously identified as (+)pinoresinolβglucoside, forsythiaside, phillyrin and phillygenin by comparison of their retention time, UV spectra and fragmentation behaviors with those of the reference. Peak 1 generated [MH]  ions at m/z 519.1878 (calc. for C 26 H 31 O 11 519.1866) indicating a molecular formula of C 26 H 32 O 11 . The fragment ion at m/z 357.1313 [MGlcH]  was attributed to its aglycone of (+)pinoresinol [Figure 5]a. Peak 2, forsythiaside, generated [MH]− ions at m/z 623.1974 (calc. for C 29 H 35 O 15 623.1976) indicating a molecular formula of C 29 H 36 O 15 . A series of the diagnostic ions from the precursor ion at m/z 461.1633 (calc. for C 20 H 29 O 12 461.4659), 477.1396 (calc. for C 23 H 25 O 11 477.1397), 487.1459 (calc. for C 21 H 27 O 13 487.1452), 443.1531 (calc. for C 20 H 27 O 11 443.1553) and 315.1061 (calc. for C 14 H 19 O 8 315.1080) were attributed to [MCaffeoylH]− , [MRhaH]− , [MphenylethanolH]− , [MCaffeoylH 2 OH]− and [MCaffeoylRhaH]− , respectively [Figure 5]b. Peak 3, phillyrin, gave [MH]− ions at m/z 533.2042 (calc. for C 27 H 33 O 11 533.2023) indicating a molecular formula of C 27 H 33 O 11 . Two diagnostic ions from the precursor ion at m/z 371.1461 (calc. for C 21 H 23 O 6 371.1495) and 356.1200 (calc. for C 20 H 20 O 6 356.1260) were observed, which can be identified as [MGlcH]− and [MGlcCH 3 H]− , respectively [Figure 5]c. Peak 4 showed [MH]− and (2 [MH]H)− ions at 371.1461 (calc. for C 21 H 23 O 6 371.1495) and 741.2953 (calc. for C 42 H 45 O 12 741.2911) indicating a molecular formula of C 21 H 24 O 6 . diagnostic ions [MCH3 H]− from the precursor ion at m/z 356.1218 (calc. for C20 H20 O6 356.1260) was observed, which can be identified as phillygenin [ [Figure 5]d.{Figure 5} Conclusions The UAE method reported here can offer an effective alternative for simultaneous extraction of (+)pinoresinolβglucoside, forsythiaside, phillyrin, and phillygenin. It can be applied to sample preparation for the quality control of the Forsythiae Fructus. Additionally, it might serve as a promising industrial extraction protocol of the four biological characteristic compounds. Compared with ME and SE, UAE has been proved to be easy, efficient, and inexpensive method with low toxicity and high reproducibility. It is important to note that the methanol concentration was the most significant factor that affects the yield according to the regression coefficients significance of the quadratic polynomial model and gradient of slope in the 3D response surface plot. The optimal UAE conditions found were: extraction temperature 70°C, time 60 min, ratio of liquid to solid 20, and methanol concentration 76.6%. The extraction yields of (+)pinoresinolβglucoside, forsythiaside, phillyrin, and phillygenin were 2.92 mg/g, 52.10 mg/g, 0.90 mg/g and 0.57 mg/g, respectively. The results indicated that UAE may facilitate a deeper understanding of the extract of active constituents in Forsythiae Fructus from the raw material to its extract for providing theoretical references. Our work was financially supported by the Program for the New Century Excellent Talents in Heilongjiang Provincial University, Heilongjiang province postdoctoral special funding, Innovative Talents Funding of Heilongjiang University of Chinese Medicine, and State Key Creative New Drug Project of China, Standard research platform construction of Traditional Chinese Medicine extract 2009ZX09308004 and 2012ZX09304005004. References


