A screening method for ternary solvent system in countercurrent chromatography
Through the establishment of high-performance liquid chromatography detection and functional relationship, the time-consuming and labor-consuming problem of ternary solvent selection of countercurrent chromatography is solved, and the selection and optimization of high-efficiency solvent system for HAWat and EBuWat systems is realized. It is suitable for the separation of natural complex samples and provides The automation and intelligent basic data of the next generation of CCC instruments.
Patent Information
- Application Number
- CN202411132907.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-19
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-08-19
AI Technical Summary
The ternary solvent selection method for countercurrent chromatography in the prior art requires multiple shake flask experiments, which is time-consuming and labor-intensive. The existing mathematical prediction model is mainly applicable to the HEMWat system. The research on HMWat and EBuWat systems is limited, making it difficult to effectively predict complex samples. solvent system.
Using a screening method of countercurrent chromatography ternary solvent system, the partition coefficient and compound polarity of template compounds were detected by high-performance liquid chromatography, and the functional relationship between the ratio of alcohol solvents and the partition coefficient and compound polarity was established, and the suitable alcohol was calculated. Solvent-like ratios are used to select and optimize ternary solvent systems in HAWat and EBuWat systems.
It realizes efficient, simple and efficient selection and optimization of countercurrent chromatography solvent system for natural complex samples, saves time, improves screening efficiency, and provides basic data for the automation and intelligence of next-generation CCC instruments.
Smart Images

Figure CN118641676B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of separation and extraction of natural products, and particularly relates to a method for screening a countercurrent chromatography solvent. Background Art
[0002] The disclosure of this background information is intended to enhance understanding of the general background of the invention and should not necessarily be taken as an acknowledgment or any form of suggestion that this information constitutes the prior art already known to a person skilled in the art.
[0003] Countercurrent chromatography (CCC) is a unique liquid-liquid partition chromatography technique that can be used to separate and purify various active ingredients from natural products because it avoids irreversible adsorption losses of samples, provides high recovery and high reproducibility. A suitable two-phase solvent system is an essential requirement for the successful implementation of CCC separations. However, the selection of a suitable two-phase liquid system is time-consuming, which has become a major obstacle to the widespread adoption of CCC.
[0004] In the 1980s, with the design and application of expert systems for chromatography system optimization, liquid chromatography or gas chromatography became more intelligent and automated. To date, shake flask experiments based on static and dynamic equilibrium are still the main means of developing countercurrent chromatography methods, which limits the automation process of countercurrent chromatography. In recent years, the rapid development of thermodynamic prediction models, including the conductor screening model of real solvents (COSMO-RS), universal functional group activity coefficients (UNIFAC), non-random two-liquid (NRTL) and quantitative structure-activity relationship (QSAR). These thermodynamic models can estimate the composition of the upper and lower phases of a two-phase solvent system and describe the interaction between solvent and solute molecules to predict the solvent system. This transfers the iterative process of solvent system selection from the laboratory to the computer, saving a lot of experimental work. However, due to the need to provide structural information of the analyte, its application in natural products is limited in the case of unknown molecules. The development of mathematical prediction functions meets the need for rapid prediction of CCC of natural products with unknown structures. Han et al. used mathematical equations such as linear, exponential or logarithmic to describe the relationship between solvent ratio and partition coefficient in n-hexane / ethyl acetate / methanol / water (HEMWat) and n-hexane / methanol / water (HMWat) systems to predict and optimize two-phase solvent systems. Wu et al. designed a 9×9 graph for HEMWat, demonstrated that there was an obvious linear correlation between the partition coefficient and the number of systems, and proposed a “four-point screening” strategy. Tong et al. used retention models such as the linear solvent strength model (LSS) and the quadratic solvent strength model to describe the mathematical relationship between the partition coefficient and solvent strength in HEMWat and EBuWat systems to predict the optimal two-phase solvent system. Koichi Inoue et al. used experimental design (DOE) as a predictive model for the HEMWat system and successfully applied it to the purification of milbemycin. These mathematical prediction methods save a lot of manpower for CCC separation. However, due to the widespread use and clear regularity of the HEMWat system, most mathematical prediction models are developed for it. In contrast, there are limited studies on other common solvent systems such as the HMWat system and the EBuWat system. The above models accurately predict the distribution of single compounds in CCC solvent systems, but it becomes challenging to predict natural complex samples with multiple components. In previous studies, a linear prediction function based on compound polarity (B%) and solvent system composition was proposed for the HEMWat system, providing a simple selection strategy for this solvent system without the need for an iterative modeling process. However, the developed method was only used for the HEMWat system, and its applicability to other solvent systems has not been verified. Therefore, it is still worthwhile to explore a simple, general, and effective method to predict CCC solvent systems other than the HEMWat system.
[0005] HEMWat, HMWat, and EBuWat systems have similar solvent compositions, but changes in lipophilic solvents such as n-hexane and ethyl acetate lead to significant differences in the applicability and sample distribution behavior of the three systems. In terms of applicability, it is generally believed that the HEMWat system is suitable for separating samples of medium polarity. However, the n-hexane / alcohol solvent (methanol, ethanol, isopropanol) / water (HAWat) system is a common solvent system for separating non-polar compounds; the ethyl acetate / n-butanol / water (EBuWat) system is the preferred method for separating more polar glycosides. In terms of sample distribution rules, the EBuWat system and the HAWat system are quite different from the modeled HEMWat system. A linear correlation between log K and system number is observed in the 9×9 plot for the HEMWat system. However, the relationship between the K value of the compound in the HMWat system and the composition of the solvent system can be characterized by an exponential function, while the relationship between the logarithmic K value of the compound in the EBuWat system and the solvent composition satisfies a quadratic function relationship. Therefore, although the concept and method for rapid selection of HEMWat solvent system have been developed, establishing mathematical prediction models for HMWat and EBuWat systems remains an unexplored but meaningful attempt, which can provide basic data for the automation and intelligence of the next generation of CCC instruments. Summary of the invention
[0006] In view of the problem that the ternary solvent selection method in the prior art needs to be combined with multiple shake bottle experiments, which is time-consuming and labor-intensive, the present invention provides a screening method for a countercurrent chromatography ternary solvent system that does not require multiple shake bottle experiments.
[0007] To achieve the above purpose, the present invention adopts the following technical solution.
[0008] A method for screening a ternary solvent system containing an alcohol solvent in countercurrent chromatography comprises the following steps:
[0009] (1) extracting the template compound with ternary solvents having different alcohol solvent ratios to obtain an upper phase solution and a lower phase solution;
[0010] (2) The content of the template compound in the upper phase solution or the lower phase solution is detected by high performance liquid chromatography (HPLC) to obtain the distribution coefficient K:
[0011] K=
[0012] Among them, A U is the peak area of the sample in the upper phase solution, A L is the peak area of the sample in the lower phase solution;
[0013] and, compound polarity B:
[0014] B=
[0015] Among them, t R is the retention time, t D is the dead time, κ is the gradient slope, and B0 is the initial methanol percentage in the mobile phase;
[0016] (3) Establishing the functional relationship between the template compound distribution coefficient K and the alcohol solvent ratio X, and calculating the alcohol solvent ratio X1 in the ternary solvent when K is 1;
[0017] (4) Establish the functional relationship between the alcohol solvent ratio X1 and the compound polarity B in the ternary solvent system;
[0018] (5) Separating and detecting the main components of the analyte by HPLC, obtaining the average value of the compound polarity B of each main component, substituting the average value of the compound polarity B into the functional relationship of step (4) to calculate the alcohol solvent ratio X in the ternary solvent system;
[0019] The ternary solvent is selected from n-hexane / alcohol / water (HAWat) or ethyl acetate / n-butanol / water (EBuWat); the alcohol is any one of methanol, ethanol and isopropanol.
[0020] In step (1), the template compound of HAWat is selected from: Honokiol (CAS: 35354-74-6), Osthole (CAS: 484-12-8), Tangeretin (CAS: 481-53-8), Dehydrocostus Lactone (CAS: 477-43-0), Isoimpreatorin (CAS: 482-45-1), Magnolol (CAS: 528-43-8), Curcumol (CAS: 4847-97-0), Schizandrin B (CAS: 61281-37-6), Tanshinone II A (CAS: 484-97-0), IIA, CAS: 568-72-9), Bakuchiol (CAS: 10309-37-2), Vitamin A (CAS: 68-26-8);
[0021] The template compound of EBuWat is selected from:
[0022] Rutin (CAS: 153-18-4), Mangiferin (CAS: 4773-96-0), Salidroside (CAS: 10338-51-9), Daidzin (CAS: 552-66-9), Picroside II (CAS: 39012-20-9), Puerarin (CAS: 3681-99-0), Paeoniflorin (CAS: 23180-57-6), Geniposide (CAS: 24512-63-8), Naringin (CAS: 10236-47-2), Ginsenoside Rg1 (CAS: 22427-39-0).
[0023] In step (2), the chromatographic column of the high performance liquid chromatography is a reverse phase C18 column, and the mobile phase is methanol-water.
[0024] In order to ensure accurate results, in step (3), the correlation coefficient R of the functional relationship is not less than 0.95; further, the correlation coefficient R is not less than 0.99.
[0025] For the HAWat system, since three alcohols can be selected, multiple ratios will be generated. In order to select a more accurate ratio, a correction step (6) is also included after step (5):
[0026] Prepare the ternary solvent according to the alcohol solvent ratio X calculated in step (5), and obtain the measured distribution coefficient K of the analyte according to steps (1) and (2), and then calculate the selectivity factor α:
[0027] α=
[0028] Among them, K n+1 and K n are the distribution coefficients K of the components of adjacent peaks in HPLC, and K n+1 ≥K n ;
[0029] And, the correction value M (mean absolute error):
[0030] M= ;
[0031] And filter and adjust according to the following methods:
[0032] a) Screening the measured distribution coefficient K value between 0.25 and 2.5 and satisfying the maximum value of the selection factor α;
[0033] b) When the K value is less than 0.25 and the selectivity factor α is maximized, the alcohol solvent ratio for the n-hexane / alcohol / water system is (XM), and the alcohol solvent ratio for the ethyl acetate / n-butanol / water system is (X+M);
[0034] c) When the K value is greater than 2.5 and the selectivity factor α is maximized, the alcohol solvent ratio for the n-hexane / alcohol / water system is (X+M), and the alcohol solvent ratio for the ethyl acetate / n-butanol / water system is (XM).
[0035] The present invention has the following advantages:
[0036] The present invention provides an effective, simple and efficient method for selecting a countercurrent chromatography solvent system for natural complex samples. By establishing a mathematical model, a new high-throughput selection and optimization strategy for a ternary solvent system is proposed. The average concentration of methanol at the inlet of the C18 column when the target compound reaches a peak value represents the compound polarity (B), and then the value of B is input into the mathematical model to calculate the solvent system conditions required for the target compound. Through a high-performance liquid chromatography analysis and simple adjustment, the solvent system suitable for sample separation is determined. From these suitable solvent systems, the optimal separation conditions of the solvent system can be easily determined. This method can effectively select and optimize the countercurrent chromatography solvent system of natural compound samples, save time, and improve screening efficiency. The present invention proposes a mathematical prediction model of the HAWat and EBuWat systems, which provides basic data for the construction of the automation and intelligent process of the next generation of CCC instruments. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 It is the structural formula of the template compound used to establish the ternary solvent screening method of the HAWat system;
[0038] Figure 2 is the fitting curve of the functional relationship between logK of the template compound and the alcohol solvent ratio X in the HAWat system, wherein ac are the linear fitting curves of n-hexane / methanol / water, n-hexane / ethanol / water and n-hexane / isopropanol / water, and df are the quadratic function fitting curves of n-hexane / methanol / water, n-hexane / ethanol / water and n-hexane / isopropanol / water;
[0039] Figure 3 This is the countercurrent chromatogram of the volatile oil of Curcuma zedoariae;
[0040] Figure 4 HPLC spectra of volatile oil of Curcuma zedoariae (a) and the purified target compound (b);
[0041] Figure 5 is the structural formula of the target compound obtained by separation of volatile oil of Curcuma zedoariae;
[0042] Figure 6 It is the structural formula of the template compound used to establish the ternary solvent screening method of the EBuWat system;
[0043] Figure 7 is the fitting curve of the template compound logK and the ratio of n-butanol X in the EBuWat system;
[0044] Figure 8 is the linear mathematical model of the EBuWat solvent system (X=1.8);
[0045] Fig. 9 This is the countercurrent chromatogram of the crude extract of Pueraria root;
[0046] Fig.10 HPLC spectra of the crude extract of Pueraria root (a) and the purified target compound (b);
[0047] Fig.11 It is the structural formula of the target compound obtained by separation of Pueraria lobata crude extract. DETAILED DESCRIPTION
[0048] The present invention will be further described below in conjunction with embodiments and drawings, but the present invention is not limited by the following embodiments.
[0049] Example 1 Establishment of the mathematical model of the ternary solvent of the HAWat system
[0050] 1. Extraction
[0051] Honokiol (CAS: 35354-74-6), Osthole (CAS: 484-12-8), Tangeretin (CAS: 481-53-8), Dehydrocostus Lactone (CAS: 477-43-0), Isoimpreatorin (CAS: 482-45-1), Magnolol (CAS: 528-43-8), Curcumol (CAS: 4847-97-0), Schizandrin B (CAS: 61281-37-6), Tanshinone IIA (CAS: 568-72-9), Bakuchiol (CAS: 10309-37-2), Vitamin A (CAS: 10309-37-2), and 1,2-dihydro-1,2-dihydro-1,2-dihydro-1,2-dihydro-1,2-dihydro-1,2-dihydro-1,2-dihydro-1,2-dihydro-1,2-dihydro-1,2-dihydro-1,2-dihydro-1,2-dihydro-1 A, CAS: 68-26-8) as a template compound, its molecular structure is as follows Figure 1 As shown;
[0052] N-hexane (H), methanol (M) / ethanol (E) / isopropanol (I) and water (Wat) were added into a separation funnel in a ratio of 10:A:(10-A)(A=1, 2, 3, 4, 5, 6, 7, 8, 9), mixed vigorously, and equilibrated at room temperature to prepare 9 ternary solvents in different ratios, which were abbreviated as HMWat, HEWat, and HIWat according to the solvent abbreviations.
[0053] The template compound was extracted with ternary solvents (shake flask experiment) to obtain an upper phase solution and a lower phase solution.
[0054] 2. Determination of the partition coefficient K
[0055] Five columns were used in the Agilent 1260 HPLC system: Agilent ZORBAX SB-C 18 (5 μm, 4.6×250 mm) chromatographic column, Aglient TC-C 18 (5 μm, 4.6×250 mm) chromatographic column, Agilent Extend -C 18 (5 μm, 4.6 × 250 mm) columns, Thermo Hypersil GOLD (5 μm, 4.6 × 250 mm) columns, and ThermoSyncronis C 18 The chromatographic column (5 μm, 4.6×250 mm) was used with methanol-water as the mobile phase. The content of the template compound in different upper and lower phase solutions was determined by gradient elution. The distribution coefficient K was calculated by the integrated peak area:
[0056] K=
[0057] in,
[0058] A U is the peak area of the sample in the upper phase solution,
[0059] A L is the peak area of the sample in the lower phase solution;
[0060] The polarity of the compound B is calculated based on the retention time, the gradient slope of the mobile phase, and the initial methanol percentage in the mobile phase:
[0061] B=
[0062] in,
[0063] t R is the retention time;
[0064] t D is the dead time, which is 0.8 min;
[0065] κ is the gradient slope, which is 2;
[0066] B0 is the initial methanol percentage in the mobile phase, which is 10%;
[0067] Gradient elution conditions: 0-45 min, methanol 10%→100%; 45-55 min, methanol 100%;
[0068] Flow rate: 1.0 mL / min;
[0069] The column temperature was 30°C;
[0070] The injection volume was 10 μL.
[0071] Table 1 Retention time (t R ) and C 18 Methanol concentration at column inlet (B)
[0072]
[0073] The retention time and methanol concentration at the inlet of each template compound determined by HPLC are shown in Table 1. The results show that for Agilent ZORBAX SB-C 18 (5 μm, 4.6×250 mm) column HAWat system is suitable for separating compounds in the range of B>83%. 18 (5 μm, 4.6×250 mm) chromatographic column HAWat system is suitable for separating compounds in the range of B>78%. 18 (5 μm, 4.6×250 mm) chromatographic column HAWat system is suitable for separating compounds with B>82% range, for Thermo Hypersil GOLD (5 μm, 4.6×250 mm) chromatographic column HAWat system is suitable for separating compounds with B>81% range, for Thermo Syncronis C 18 (5 μm, 4.6×250 mm) column HAWat system is suitable for separating compounds in the range of B>89%.
[0074] Table 2 Distribution coefficient K of template compounds in HAWat system
[0075]
[0076] Note: - represents K value is too large or too small; each value is repeated three times.
[0077] 3. Functional relationship fitting of distribution coefficient K and alcohol solvent ratio X
[0078] The K values in Table 2 were log-converted, and the quantitative relationship between logK and X was fitted according to a linear equation and a quadratic equation, respectively.
[0079] Table 3 The linear equation of log K of template compound and alcohol solvent ratio X in HAWat system
[0080]
[0081] Table 4 Quadratic equation of logK of template compound and alcohol solvent ratio X in HAWat system
[0082]
[0083] As shown in Table 3-4, the two models have an ideal correlation coefficient R between the logK of 11 template compounds and the proportion of alcohol solvents in the water phase. 2 Specifically, when a linear function is used to fit the logK of the template compound to the apparent content of the alcohol solvent in the aqueous phase, the R 2 It can be above 0.9474. With the change of alcohol solvent type, the fitting effect of the one-dimensional linear function on the n-hexane / methanol / water system is better than that on the n-hexane / ethanol / water and n-hexane / isopropanol / water systems. This indicates that the change of alcohol solvent may cause the change of hydrophilic and hydrophobic interactions in the solvent system, thereby affecting the distribution behavior of the template compound. When fitting with a quadratic function, it can be found that the R of the quadratic function of all template compounds 2 All of them are above 0.9927, and some can even reach 0.9999. This shows that in the HAWat system, both models have good fitting effects on describing the relationship between the logK of the template compound and the proportion of alcohol solvent in the aqueous phase. The fitting curves of the 11 template compounds for linear functions and quadratic functions are shown in Figure 2. Figure 2 shown.
[0084] 4. Functional relationship fitting between alcohol solvent ratio X and compound polarity B
[0085] According to the fitting equations in Tables 3 and 4, when K=1, the alcohol solvent proportion X is calculated, and the quantitative relationship between X and B is fitted according to a linear equation and a quadratic equation, respectively.
[0086] Table 5 Linear fitting mathematical model equations of different chromatographic columns in the HAWat system
[0087]
[0088] Table 6 Nonlinear fitting mathematical model equations of different chromatographic columns in the HAWat system
[0089]
[0090] The results are shown in Tables 5 and 6. By analyzing these one-variable linear function models and quadratic function models, it can be found that the square of the correlation coefficient (R 2 ) are almost higher than the model established by a linear function, but the difference between the two is small. Therefore, this study believes that both a linear function and a quadratic function can predict the system in the mathematical modeling of the n-hexane / alcohol solvent / water solvent system.
[0091] 5. Screening and calibration
[0092] According to experience, a K value of 0.25-2.5 has a better effect on compound extraction. For samples with large deviations outside this range, the predicted results need to be adjusted and corrected.
[0093] The predicted value of the alcohol ratio was calculated by Table 5, and the actual value was calculated by Table 3 to compare the two sets of data; the predicted value was calculated by Table 6, and the actual value was calculated by Table 4. The difference and average value of the predicted value and actual value of the template compound with large error when K=1 were calculated, and statistical analysis was performed. The results showed that in the one-dimensional linear function prediction model, the difference between the predicted value and the actual value of 15 prediction models was relatively stable, and the average absolute error of 15 models was 1.9. At the same time, in the quadratic function prediction model, the difference between the predicted value and the actual value of 15 prediction models was also relatively stable, and the average absolute error of 15 models was 1.8. This shows that although the two types of prediction models established have poor prediction effects for a few compounds, the prediction results can be corrected according to the distribution of the samples in the two phases to meet the separation requirements of the countercurrent chromatography solvent system. That is, for K values less than 0.25, the alcohol solvent ratio is (X-1.9 or 1.8); for K values greater than 2.5, the alcohol solvent ratio is (X+1.9 or 1.8).
[0094] Example 2 Screening and countercurrent chromatography separation of volatile oil from Curcuma zedoariae using HAWat ternary solvent system
[0095] An appropriate amount of Curcuma zedoaria slices (100 g) were crushed, soaked in 10 times the amount of water for half an hour, and then heated under reflux for extraction. After continuous slight boiling for 10 hours, the heating was stopped and the volatile oil of Curcuma zedoaria was collected.
[0096] 1. Screening of HAWat ternary solvent system
[0097] The volatile oil of Zedoariae was isolated by Agilent ZORBAX SB-C 18The column (5 μm, 4.6×250 mm) was used for gradient elution with methanol-water as the mobile phase: 0-45 min, methanol from 10% to 100%; 45-55 min, methanol was maintained at 100%. The flow rate was 1.0 mL / min, the column temperature was 30°C, and the detection wavelength was 214 nm.
[0098] The results showed that the volatile oil of Curcuma zedoaria had five main chromatographic peaks at 37.60 min, 38.05 min, 38.97 min, 39.73 min and 41.74 min. The average value of the compound polarity B of target compounds 1-5 was calculated to be 86.84% with these five components as target compounds 1-5.
[0099] According to the one-dimensional linear prediction model and the quadratic function prediction model in Tables 5 and 6, the average value of B was substituted into the prediction model to obtain 6 solvent systems of different types and proportions. The prediction results of the one-dimensional linear model are HMWat=10:6.2:3.8, HEWat=10:5.5:4.5 and HIWat=10:6.5:3.5; while the prediction results of the quadratic function model are HMWat=10:6.7:3.3, HEWat=10:5.8:4.2 and HIWat=10:6.7:3.3.
[0100] In order to verify these solvent systems, the K value of each target compound was measured by a shake flask experiment. The specific K values of the five target compounds are shown in Tables 7 and 8.
[0101] Table 7 Verification and correction of the predicted solvent system for target compounds in volatile oil of Curcuma zedoariae (linear function prediction model)
[0102]
[0103] Table 8 Verification and correction of the predicted solvent system for target compounds in volatile oil of Curcuma zedoaria (quadratic function prediction model)
[0104]
[0105] Since the K values of the target compounds in the predicted solvent system are generally too large, the predicted solvent system needs to be corrected. According to the above predicted values and the actual discreteness, the solvent system after the linear model correction is HMWat=10:8.1:1.9, HEWat=10:7.4:2.6 and HIWat=10:8.4:1.6; the solvent system after the quadratic function model correction is HMWat=10:8.5:1.5, HEWat=10:7.6:2.4 and HIWat=10:8.5:1.5.
[0106] Although the actual K values of the target compounds in the HIWat system are all near K=1, the α value between the two target compounds is too low, which will cause the target peaks to be unable to be separated on the HSCCC instrument. In the HMWat=10:8.1:1.9 and HEWat=10:7.4:2.6 solvent systems, the K value of target compound 5 is larger, which will have a longer separation time and reduce the resolution of the chromatographic peaks. At the HEWat=10:7.6:2.4 ratio, the separation degree of target compound 1 and target compound 2 is poor. According to the K value and α value of the target compound, HMWat=10:8.5:1.5 was finally selected for the solvent system conditions for the subsequent separation of the five main chromatographic peaks in the volatile oil of Curcuma zedoaria.
[0107] 2. Countercurrent chromatography separation of volatile oil from Curcuma zedoaria
[0108] 150 mg of volatile oil of Curcuma zedoaria was dissolved in 2 mL of upper phase and 2 mL of lower phase, the upper phase was used as stationary phase and the lower phase as mobile phase for HSCCC separation.
[0109] Completely fill the multilayer column with the upper phase at a flow rate of 30 mL / min. After the stationary phase fills the chromatographic spiral tube, pump the mobile phase at a flow rate of 3 mL / min and collect it in a measuring cylinder to calculate the retention rate of the stationary phase. Inject the sample solution into the separation column, turn on the speed controller, and rotate the separation column clockwise at a speed of 800 rpm. Turn on the UV detector and continuously monitor the column effluent using the UV detection method. Collect the chromatographic peak components based on the UV analysis results and the chromatogram. Stationary phase retention rate ( S r ) is calculated as:
[0110] S r = ×100%
[0111] Where Vs is the stationary phase volume and Vc is the total column capacity.
[0112] Under the conditions of flow rate 3 mL / min and rotation speed 800 rpm, the stationary phase retention rate was 76.7%, and all target compounds were separated within 400 min. The HSCCC chromatogram is shown in Figure 3 As a result, 109.23 mg of compound 1, 13.95 mg of compound 2, 16.25 mg of compound 3, 10.66 mg of compound 4 and 8.33 mg of compound 5 were separated from 300 mg of volatile oil of Curcuma zedoariae.
[0113] After one HSCCC separation, the purities of compounds 1-5 were 96.2%, 86.8%, 90.5%, 87.7% and 97.2%, respectively. Compounds 2-4 were subjected to preparative liquid phase purification. After purification, compounds 1-3 were light yellow oily liquids, and compounds 4 and 5 were white crystals. The HPLC spectrum of the volatile oil of Curcuma zedoaria is shown in FIG. Figure 4 a, wherein the labels 1-5 correspond to target compounds 1-5 respectively; the HPLC spectra of the purified target compounds 1-5 are shown in Figure 4 As shown in b, the structure is 1 H NMR, 13 C NMR was used to identify and compare with previously published values. The structures of compounds 1-5 are shown in Figure 5 As shown, they are Curzerenone (Compound 1), Isofuranodienone (Compound 2), Furanodienone (Compound 3), Isocurcumenol (Compound 4) and Germacrone (Compound 5).
[0114] Example 3 Establishment of the ternary solvent screening method of the EBuWat system
[0115] 1. Extraction
[0116] Rutin (CAS: 153-18-4), mangiferin (CAS: 4773-96-0), salidroside (CAS: 10338-51-9), daidzin (CAS: 552-66-9), picroside II (CAS: 39012-20-9), puerarin (CAS: 3681-99-0), paeoniflorin (CAS: 23180-57-6), geniposide (CAS: 24512-63-8), naringin (CAS: 10236-47-2), ginsenoside Rg1 (CAS: 22427-39-0) were used as template compounds, and their molecular structures are shown in the following formulas: Figure 6 As shown;
[0117] Ethyl acetate (E), n-butanol (Bu) and water (Wat) were added into a separation funnel in the ratio of (10-Bu):Bu:10, Bu=1, 2, 3, 4, 5, 6, 7, 8, 9, mixed vigorously, and equilibrated at room temperature to prepare 9 ternary solvents in different ratios, which were abbreviated as EBuWat according to the solvent abbreviation.
[0118] The template compound is extracted with a ternary solvent to obtain an upper phase solution and a lower phase solution.
[0119] 2. Determination of the partition coefficient K
[0120] Five columns were used in the Agilent 1260 HPLC system: Agilent ZORBAX SB-C 18 (5 μm, 4.6×250 mm) chromatographic column, Aglient TC-C 18 (5 μm, 4.6×250 mm) chromatographic column, Agilent Extend -C 18 (5 μm, 4.6 × 250 mm) columns, Thermo Hypersil GOLD (5 μm, 4.6 × 250 mm) columns, and ThermoSyncronis C 18 The chromatographic column (5 μm, 4.6×250 mm) was used with methanol-water as the mobile phase. The content of the template compound in different upper and lower phase solutions was determined by gradient elution. The distribution coefficient K was calculated by the integrated peak area:
[0121] K=
[0122] in,
[0123] A U is the peak area of the sample in the upper phase solution,
[0124] A L is the peak area of the sample in the lower phase solution;
[0125] The polarity of the compound B is calculated based on the retention time, the gradient slope of the mobile phase, and the initial methanol percentage in the mobile phase:
[0126] B=
[0127] in,
[0128] t R is the retention time;
[0129] t D is the dead time, which is 0.8 min;
[0130] κ is the gradient slope, which is 2;
[0131] B0 is the initial methanol percentage in the mobile phase, which is 10%;
[0132] Gradient elution conditions: water (A) - methanol (B), 0 - 45 min, 10% → 100% B; 45 - 55 min, 100% B;
[0133] Flow rate: 1.0 mL / min;
[0134] Column temperature was 30 °C;
[0135] Sample injection volume was 10 μL.
[0136] Table 9 Retention times (t R ) and C 18 Methanol concentration (B) at the column inlet
[0137]
[0138] Ten selected template compounds B were analyzed. Since salidroside and geniposide showed special distribution behaviors, the applicable range of the solvent system for other flavonoid glycosides was described. For the Agilent ZORBAX SB - C 18 (5 μm, 4.6×250 mm) column, the EBuWat system could separate flavonoid glycosides in the range of 40% < B < 55%. For the Aglient TC - C18 (5 μm, 4.6×150 mm), Agilent Extend - C 18 (5 μm, 4.6×250 mm) and Thermo Hypersil GOLD (5 μm, 4.6×250 mm) columns, the EBuWat system could be used to separate flavonoid glycosides in the range of 36% < B < 50%. While for the Thermo Syncronis C 18 (5 μm, 4.6×250 mm) column, the EBuWat system could separate flavonoid glycosides in the range of 43% < B < 57%.
[0139] Table 10 Partition coefficients K of the template compounds in the EBuWat system
[0140]
[0141] The K values of the 10 template compounds in the EBuWat system are shown in Table 10: When the apparent content of n - butanol in the organic phase is less than 60%, as the proportion of n - butanol increases, the K values of the template compounds show an upward trend. However, when the apparent content of n - butanol in the organic phase is higher than 60%, the K values of the template compounds tend to be constant.
[0142] 3. Fitting of the functional relationship between the partition coefficient K and the proportion X of alcohol solvents
[0143] The K values in Table 10 were log-transformed, and the quantitative relationship between logK and X was fitted.
[0144] Table 11 Quadratic equation of logK of template compound and alcohol solvent ratio X in EBuWat system
[0145]
[0146] As shown in Table 11, Figure 7 As shown in the figure, there is an obvious quadratic correlation between the logK of the 10 template compounds and the ratio of n-butanol in the organic phase X. The correlation coefficients of all template compounds R 2 All of them are above 0.97, which shows that the quadratic function can well describe the distribution behavior of compounds in the EBuWat system.
[0147] 4. Functional relationship fitting between n-butanol ratio X and compound polarity B
[0148] According to the fitting equation in Table 11, the proportion of n-butanol X when K=1 is calculated, and the quantitative relationship between X and B is fitted. The results are shown in Figure 8 As shown in the figure, the ratio of n-butanol is a constant X = 1.8. Except for two special terpene glycosides, salidroside and gardenia glycoside, the remaining compounds can find a solvent system ratio suitable for compound separation when the n-butanol concentration is between 0% and 40%.
[0149] 5. Screening and calibration
[0150] The percentage of n-butanol in the remaining template compounds except salidroside and gardenia glycoside when K=1 was statistically analyzed, and the percentage of n-butanol in the template compound when K=1 fluctuated within the range of 1.8±0.9. That is, for K values less than 0.25, the alcohol solvent ratio is (X+0.9); for K values greater than 2.5, the alcohol solvent ratio is (X-0.9).
[0151] Example 4 Screening of Pueraria extracts suitable for EBuWat ternary solvent system and countercurrent chromatography separation
[0152] The Pueraria root slices (100 g) were crushed appropriately and extracted with 1000 mL of 95% ethanol by ultrasonic for 3 times at room temperature for 1 h each time. After filtration, the extract was distilled under reduced pressure to obtain 10.7 g of Pueraria root crude extract.
[0153] 1. Screening of HAWat ternary solvent system
[0154] The crude extract of Pueraria lobata was purified by Agilent ZORBAX SB-C 18The column (5 μm, 4.6×250 mm) was gradient eluted with methanol-water as the mobile phase: 0-45 min, methanol from 10% to 100%; 45-55 min, methanol was maintained at 100%. The flow rate was 1.0 mL / min, the column temperature was 30°C, and the detection wavelength was 280 nm.
[0155] The results showed that the crude extract of Pueraria lobata had five main chromatographic peaks at 13.36min, 15.82min, 16.46min, 16.83min, and 18.07min. The EBuWat solvent system was prepared according to 8.2:1.8:10, and the K values of each target compound were measured by shaking bottle experiment. The specific K values of the five target compounds are shown in Table 12.
[0156] Table 12 Verification and correction of predicted solvent systems for target compounds in Pueraria lobata extract
[0157]
[0158] The K value of the target compound in the kudzu root crude extract was determined according to the EBuWat system mathematical model. Since the K value of the target compound in the initial solvent system was too small, the initial solvent system was corrected (n-butanol ratio + 0.9). The K value of the target compound in the corrected solvent system met the separation requirements of countercurrent chromatography, so the countercurrent chromatography separation experiment was carried out with EBuWat = 7.3:2.7:10 as the ternary solvent.
[0159] 2. Countercurrent Chromatographic Separation of Pueraria lobata Extract
[0160] 150 mg of the crude extract was dissolved in 2 mL of the upper phase and 2 mL of the lower phase, the lower phase was the mobile phase and the upper phase was the stationary phase for HSCCC separation.
[0161] The upper phase is completely filled into the multilayer column at a flow rate of 30 mL / min. After the stationary phase fills the chromatographic spiral tube, the mobile phase is pumped in at a flow rate of 3 mL / min and collected in a measuring cylinder. The retention rate of the stationary phase is calculated. The sample solution is injected into the separation column, and the speed controller is turned on to rotate the separation column at a speed of 800 rpm in a clockwise direction. The UV detector is turned on and the column effluent is continuously monitored by UV detection. The chromatographic peak components are collected according to the UV analysis results and the chromatogram. The HSCCC chromatogram is shown in Figure 1. Fig. 9 shown.
[0162] After one HSCCC separation, the purities of compounds 6-10 were 98.0%, 97.8%, 93.2%, 97.7% and 83.0%, respectively. Compound 10 was purified by preparative liquid phase. The HPLC spectrum of the crude extract of Pueraria root is shown in Fig.10a, wherein the labels 6-10 correspond to the target compounds 6-10 respectively; the HPLC spectrum of the purified target compound 6-10 is as shown in Fig.10 As shown in b, the structure is 1 H NMR, 13 C NMR was used to identify and compare with previously published values. The structures of compounds 6-10 are shown in Fig.11 As shown, they are 3'-Hydroxypueraria glycoside (Compound 6), Puerarin (Compound 7), 3'-Methoxypuerarin (Compound 8), Mirificin (Compound 9) and Daidzin (Compound 10).
[0163] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims
1. A method for screening a ternary solvent system containing an alcohol solvent in countercurrent chromatography, characterized in that: The following steps are involved: (1) extracting the template compound with ternary solvents having different alcohol solvent ratios to obtain an upper phase solution and a lower phase solution; (2) The content of the template compound in the upper phase solution or the lower phase solution is detected by high performance liquid chromatography, and the distribution coefficient K is obtained: K= Among them, A U is the peak area of the sample in the upper phase solution, A L is the peak area of the sample in the lower phase solution; and, compound polarity B: B= Among them, t R is the retention time, t D is the dead time, κ is the gradient slope, and B0 is the initial methanol percentage in the mobile phase; (3) Establishing the functional relationship between the template compound distribution coefficient K and the alcohol solvent ratio X, and calculating the alcohol solvent ratio X1 in the ternary solvent when K is 1; (4) Establish the functional relationship between the alcohol solvent ratio X1 and the compound polarity B in the ternary solvent system; (5) Separating and detecting the main components of the analyte by high performance liquid chromatography, obtaining the average value of the compound polarity B of each main component, substituting the average value of the compound polarity B into the functional relationship of step (4) to calculate the alcohol solvent ratio X in the ternary solvent system; The ternary solvent is selected from n-hexane / alcohol / water or ethyl acetate / n-butanol / water; the alcohol is any one of methanol, ethanol and isopropanol; Step (5) also includes a correction step (6): Prepare the ternary solvent according to the alcohol solvent ratio X calculated in step (5), and obtain the measured distribution coefficient K of the analyte according to steps (1) and (2), and then calculate the selectivity factor α: α= Among them, K n+1 and K n are the distribution coefficients K of the components of adjacent peaks in HPLC, and K n+1 ≥K n ; And, correction value M: M= ; And filter and adjust according to the following methods: a) Screening the measured distribution coefficient K value between 0.25 and 2.5 and satisfying the maximum value of the selection factor α; b) When the K value is less than 0.25 and the selectivity factor α is maximized, the alcohol solvent ratio for the n-hexane / alcohol / water system is (XM), and the alcohol solvent ratio for the ethyl acetate / n-butanol / water system is (X+M); c) When the K value is greater than 2.5 and the selectivity factor α is maximized, the alcohol solvent ratio for the n-hexane / alcohol / water system is (X+M), and the alcohol solvent ratio for the ethyl acetate / n-butanol / water system is (XM).
2. The screening method according to claim 1, characterized in that In step (1), the template compound of n-hexane / alcohol / water is selected from the group consisting of: honokiol, osthole, hesperidin, dehydrocostus lactone, isoimperatorin, magnolol, curcumol, schisandrin B, tanshinone II A, bakuchiol, and vitamin A; The template compound of ethyl acetate / n-butanol / water is selected from: rutin, mangiferin, salidroside, daidzin, picroside II, puerarin, paeoniflorin, gardenia jasminoides, naringin, and ginsenoside Rg1.
3. The screening method according to claim 1, characterized in that In step (2), the chromatographic column of the high performance liquid chromatography is a reverse phase C18 column, and the mobile phase is methanol-water.
4. The screening method according to claim 1, characterized in that In step (3), the correlation coefficient R of the functional relationship is not less than 0.
95.
5. The screening method according to claim 1, characterized in that In step (3), the correlation coefficient R of the functional relationship is not less than 0.99.
