A method for ultrasonically extracting effective components of Astragalus and Safflower herb pair using natural deep eutectic solvent

Abstract

The present invention provides a method for ultrasonically extracting the effective components of Astragalus and Safflower drug pair using a natural deep eutectic solvent. This patent uses the contents of 9 effective components in the Astragalus and Safflower drug pair as evaluation indicators, assigns weights using the entropy weight method, calculates the comprehensive score, and uses response surface design and genetic neural network optimization to obtain a method for ultrasonically extracting the effective components of the Astragalus and Safflower drug pair using a natural deep eutectic solvent. Using the established UPLC‑MS / MS method, the pharmacokinetic study of hydroxysafflor yellow A, syringin, calycosin, formononetin and calycosin in the deep eutectic solvent extract in rats showed that compared with the water extract, the deep eutectic extraction significantly improved the bioavailability of the five components in rats (P<0.05).

Description

Technical Field

[0001] The invention provides a method for ultrasonically extracting effective components of astragalus and safflower herb pairs by using a natural low eutectic solvent. Background Art

[0002] Stroke is the leading cause of death and disability among Chinese residents, seriously affecting the health and quality of life of patients. Astragalus-safflower is a classic Chinese medicine pair in clinical practice. It is used in combination in Astragalus and Safflower Decoction, Buyang Huanwu Decoction and Naoxintong Capsules, and has a definite effect on ischemic cerebrovascular disease. Astragalus is the most frequently used Qi-tonifying medicine in the treatment of stroke. It can replenish Qi and blood, relieve stagnation and relieve numbness, and improve the symptoms of Qi and blood deficiency in patients; Safflower is a classic blood-activating and stasis-removing medicine that can regulate menstruation, activate blood circulation, stop bleeding and relieve pain. The two medicines have many similarities in many mechanisms and pathways of protecting cerebral ischemic damage, such as inhibiting cell apoptosis and inflammatory response, anti-oxidation, scavenging free radicals, and regulating signal pathways to play a brain protective role.

[0003] The active ingredients contained in traditional Chinese medicine or natural medicine are the material basis for treating diseases. How to effectively extract the active ingredients has always been the focus of research. Although the commonly used solvents for extracting traditional Chinese medicine have the advantages of simple operation and low cost, most solvents also have toxicity problems and are likely to have an impact on the environment and drugs. As environmental protection awareness continues to gain popularity, people's requirements for health and green are getting higher and higher. The concept of green extraction has also been proposed. Deep eutectic solvents (DESs) are a new type of green solvents composed of two types of compounds: hydrogen bond acceptors (HBAs) and hydrogen bond donors (HBDs). The two may promote the formation of mixed systems through hydrogen bonding and have good solvent properties. They are non-toxic, low volatile, and have high thermal stability. Natural deep eutectic solvents (NADESs) are completely composed of natural products. Among them, HBA is usually amines, and HBD is usually organic acids or carbohydrates. NADESs have high solubility for natural products, especially compounds with poor water solubility and macromolecular substances, and have great potential in traditional Chinese medicine extraction. Summary of the invention

[0004] The invention provides a method for ultrasonically extracting effective components of astragalus and safflower herb pairs by using a natural low eutectic solvent.

[0005] A method for ultrasonically extracting effective components of astragalus and safflower drug pair using a natural low eutectic solvent. The method comprises: accurately weighing astragalus and safflower powders in proportion, adding a natural low eutectic solvent, and ultrasonically extracting; after centrifuging the extract, taking the supernatant, diluting it with ultrapure water, and passing it through a porous filter membrane; and quantifying 9 effective components in each extract by HPLC; the natural low eutectic solvent is a Bet-Lac solvent.

[0006] Preferably, the extraction time is 85-90 min, the temperature is 65-75° C., the Bet-Lac concentration is 70-80%, and the liquid-to-solid ratio is 48-52 mg / mL.

[0007] Preferably, the extraction time is 90 min, the temperature is 65° C., the Bet-Lac concentration is 80%, and the liquid-to-solid ratio is 50 mg / mL.

[0008] Preferably, astragalus and safflower coarse powder are accurately weighed in a ratio of 3:1.

[0009] Preferably, the natural low eutectic solvent is prepared by mixing Bet-Lac and water in a molar ratio of 1:1, heating in an oil bath and maintaining at 80° C. until a clear liquid is formed, and maintaining magnetic stirring during heating to ensure that the melted liquid is evenly mixed.

[0010] This application method includes the following steps:

[0011] (1) Preparation of test solution

[0012] Astragalus and safflower powders were accurately weighed according to the commonly used clinical dosage ratio, natural low eutectic solvent was added, and ultrasonic extraction was performed; after the extract was centrifuged, the supernatant was diluted with ultrapure water and passed through a porous filter membrane to obtain the test solution;

[0013] (2) Preparation of reference solution

[0014] Accurately weigh the reference substances of calycosin, calycosin glycosides, formononetin, hydroxysafflower yellow A, dehydrated safflower yellow B, meditartin, syringin, kaempferol-3-O-rutinoside and safflower red pigment, and add methanol to prepare reference substance solutions respectively; accurately pipette each reference substance solution to prepare a mixed reference substance solution;

[0015] (3) HPLC-DAD content determination.

[0016] in,

[0017] Step (2): Accurately weigh appropriate amounts of calycosin, calycosin glycosides, formononetin, hydroxysafflor yellow A, dehydrated safflor yellow B, medipteroside, syringin, kaempferol-3-O-rutinoside and safflower red pigment reference substance, and add methanol to prepare reference substance solutions of 2.330, 2.000, 2.556, 4.450, 3.880, 2.330, 1.005, 2.330 and 4.000 mg / mL, respectively. Accurately pipette each reference solution to prepare a mixed reference solution containing 210 μg of calycosin, 180 μg of calycosin glycosides, 230 μg of formononetin, 400 μg of hydroxysafflor yellow A, 350 μg of dehydrated safflor yellow B, 210 μg of meditartin, 90 μg of syringin, 210 μg of kaempferol-3-O-rutinoside, and 400 μg of safflor red pigment per 1 mL of solution.

[0018] Step (3) HPLC-DAD content determination chromatographic conditions are:

[0019] Column: Alltech Alltima C 18 Chromatographic column (250 mm × 4.6 mm, 5 μm); Guard column: Alltech Alltima C 18 Guard column (12.5 mm × 4.6 mm); flow rate: 1.0 mL / min; column temperature: 30 °C; mobile phase: acetonitrile (A), 0.01% trifluoroacetic acid aqueous solution (B); gradient elution program: 0-7 min, 10%-14% A; 7-10 min, 14%-18% A; 10-20 min, 18%-20% A; 20-27 min, 20%-20% A; 27-30 min, 20%-23 %A; 30-36min, 23%-26%A; 36-44min, 26%-34%A; 44-53min, 34%-50%A; Detection wavelength: 403nm to detect hydroxysafflor yellow A and dehydrated safflor yellow B, 260nm to detect calycosin, calycosin glycosides, formononetin, kaempferol-3-O-rutinoside and syringin, 210nm to detect medipteroside, 520nm to detect saffron red pigment.

[0020] In addition, the method of the present application also provides a parameter optimization method: using a genetic neural network model to optimize parameters and predict optimal extraction conditions and comprehensive scores.

[0021] This patent uses the contents of 9 active ingredients (hydroxysafflor yellow A, dehydrated safflor yellow B, syringin, calycosin, kaempferol-3-O-rutinoside, formononetin, calycosin, meditartin and safflower red pigment) in the Astragalus saffron herb pair as evaluation indicators, assigns weights using the entropy weight method, calculates the comprehensive score, and uses response surface design and genetic neural network optimization to obtain a method for ultrasonic extraction of active ingredients of the Astragalus saffron herb pair using natural deep eutectic solvents. The established UPLC-MS / MS method was used to study the pharmacokinetics of hydroxysafflor yellow A, syringin, calycosin, calycosin and calycosin in the deep eutectic solvent extract in rats. The results showed that compared with the water extract, the low eutectic extraction significantly improved the bioavailability of the five ingredients in rats (P<0.05). BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is the HPLC chromatogram of the mixed reference solution and the test solution.

[0023] Among them, 1. Hydroxysafflor yellow A; 2. Dehydrated safflor yellow B; 3. Syringin; 4. Calycosin isoflavone glycoside; 5. Kaempferol-3-O-rutinoside; 6. Formononetin; 7. Calycosin isoflavone; 8. Medidatin; 9. Safflower red pigment

[0024] A, B, C, and D are the chromatograms of the mixed reference solution at detection wavelengths of 403, 260, 210, and 520 nm, respectively; E, F, G, and H are the chromatograms of the test solution at detection wavelengths of 403, 260, 210, and 520 nm, respectively.

[0025] Figure 2 The influence of each factor on the total extraction rate.

[0026] Figure 3 It is a neural network structure.

[0027] Figure 4 Neural network prediction model results.

[0028] Figure 5 UPLC-MS / MS chromatograms of blank plasma plus mixed standards and internal standards;

[0029] Among them, 1. syringin; 2. calycosin isoflavone glycoside; 3. formononetin; 4. calycosin isoflavone; 5. hydroxysafflor yellow A; 6. digoxin; 7. puerarin. DETAILED DESCRIPTION

[0030] 1. Instruments and reagents

[0031] 1.1 Instrument

[0032] Swinging high-speed universal pulverizer (DFY-500, Wenling Linda Machinery Co., Ltd.); 1 / 10000 analytical balance (AL104, Mettler Toledo); ultrasonic cleaner (KQ5200B, Kunshan Ultrasonic Cleaning Instrument Co., Ltd.); centrifuge (3K15, Yangzhou Xima Centrifuge Co., Ltd.); ultrapure water instrument (Millipore); high-performance liquid chromatography (Agilent1260infinity, Agilent Technologies, USA) equipped with G1311C quaternary pump, G1316A column oven and G1315D detector. The liquid-mass spectrometer system consists of DGU-20A5R(C)HPLC (SHIMADZU CORPORATION, Japan) and API4500Q-TRAP triple quadrupole mass spectrometer (AB SCIEX, USA).

[0033] 1.2 Drug testing

[0034] Reference substances of calycosin (batch number: DST200609-012), calycosin glycosides (batch number: DST200619-013), formononetin (batch number: DST201129-044), meditartin (batch number: DST200206-159), hydroxysafflor yellow A (batch number: DSTDQ001702) were purchased from Chengdu Desite; syringin (batch number: 19111241, Shanghai Yuanye); kaempferol-3-O-rutinoside (batch number: 21070121, Shanghai Tongtian); safflower red pigment and dehydrated safflor yellow B were homemade in the laboratory. The purity of the above reference substances was ≥98%. Trifluoroacetic acid (analytical grade); ultrapure water; acetonitrile (chromatographic grade); methanol (chromatographic grade).

[0035] Astragalus (batch number: 190702) and safflower (batch number: 200313) were purchased from the famous Chinese medicine clinic of Zhejiang Chinese Medical University. They were identified by Professor Huang Shengwu of the School of Pharmacy of Zhejiang Chinese Medical University. Astragalus is the dried root of Astragalus membranaceus (Fisch.) Bge., a plant of the leguminous family, and safflower is the dried flower of Carthamus tinctorius L., a plant of the Asteraceae family. The pieces were crushed and passed through a 20-mesh sieve to obtain the powders of Astragalus and safflower.

[0036] 2 Methods and Results

[0037] 2.1 Preparation of test solution

[0038] Weigh astragalus and safflower powders accurately according to the clinically commonly used dosage ratio of 3:1, place in a conical flask, add a certain concentration of NADESs according to a certain liquid-to-solid ratio, seal, shake well, and ultrasonically extract for a certain time at a certain temperature and power. Centrifuge the extract at 4000rpm for 15min, take the supernatant and dilute it 3 times with ultrapure water, filter it through a 0.22μm microporous membrane, and obtain the test solution.

[0039] 2.2 Preparation of reference solution

[0040] Accurately weigh appropriate amounts of calycosin, calycosin glycosides, formononetin, hydroxysafflor yellow A, dehydrated safflor yellow B, medipteroside, syringin, kaempferol-3-O-rutinoside and safflower red pigment reference substances, and add methanol to prepare reference substance solutions of 2.330, 2.000, 2.556, 4.450, 3.880, 2.330, 1.005, 2.330 and 4.000 mg / mL, respectively. Accurately pipette each reference solution to prepare a mixed reference solution containing 210 μg of calycosin, 180 μg of calycosin glycosides, 230 μg of formononetin, 400 μg of hydroxysafflor yellow A, 350 μg of dehydrated safflor yellow B, 210 μg of meditartin, 90 μg of syringin, 210 μg of kaempferol-3-O-rutinoside, and 400 μg of safflor red pigment per 1 mL of solution.

[0041] 2.3 HPLC-DAD content determination chromatographic conditions

[0042] Column: Alltech Alltima C 18 Chromatographic column (250 mm × 4.6 mm, 5 μm); Guard column: Alltech Alltima C 18 Guard column (12.5 mm × 4.6 mm); flow rate: 1.0 mL / min; column temperature: 30 °C; mobile phase: acetonitrile (A), 0.01% trifluoroacetic acid aqueous solution (B); gradient elution program: 0-7 min, 10%-14% A; 7-10 min, 14%-18% A; 10-20 min, 18%-20% A; 20-27 min, 20%-20% A; 27-30 min, 20%-23 %A; 30-36min, 23%-26%A; 36-44min, 26%-34%A; 44-53min, 34%-50%A; Detection wavelength: 403nm to detect hydroxysafflor yellow A and dehydrated safflor yellow B, 260nm to detect calycosin, calycosin glycosides, formononetin, kaempferol-3-O-rutinoside and syringin, 210nm to detect medipteroside, 520nm to detect saffron red pigment.

[0043] 2.4 HPLC-DAD content determination methodology

[0044] 2.4.1 Specificity test

[0045] Accurately pipette the mixed reference solution and test solution respectively, inject 20 μL according to the chromatographic conditions under "2.3", and measure to obtain the corresponding chromatogram ( Figure 1 ). Under the chromatographic conditions, the nine active ingredients, hydroxysafflor yellow A, dehydrated safflor yellow B, syringin, calycosin, kaempferol-3-O-rutinoside, formononetin, calycosin, medipterostilbene and safflor red pigment, were well separated, and other components in the test solution did not interfere with the determination of the nine components.

[0046] 2.4.2 Linear range investigation

[0047] The mixed reference solution under "2.2" was diluted step by step to obtain a mixed reference solution with gradient concentrations, and the determination was performed according to the chromatographic conditions under "2.3". A linear regression was performed with the peak area of ​​the chromatographic peak as the ordinate (Y) and the concentration of each component in the reference solution as the abscissa (X) to obtain the regression equation. The results are shown in Table 1.

[0048] Table 1 Results of linear relationship investigation of 9 components

[0049]

[0050]

[0051] 2.4.3 Precision test

[0052] Accurately draw the mixed reference solution under "2.2", repeat the injection 6 times according to the chromatographic conditions, the injection volume is 20μL, and record the corresponding peak area. The RSDs of the peak areas of hydroxysafflor yellow A, dehydrated safflor yellow B, safflor red pigment, syringin, calycosin isoflavone glycosides, kaempferol-3-O-rutinoside, formononetin, calycosin isoflavones, and medi-pterostilbene were calculated to be 1.61%, 0.97%, 1.84%, 0.94%, 1.28%, 1.28%, 1.99%, 1.54%, and 1.76%, respectively. The results show that the instrument has good precision.

[0053] 2.4.4 Repeatability test

[0054] Take 6 samples from the same batch, prepare the test solution according to the operation of "2.1", measure according to the chromatographic conditions, and record the corresponding peak areas. The RSDs of the peak areas of hydroxysafflor yellow A, dehydrated safflor yellow B, safflor red pigment, syringin, verbascoside, kaempferol-3-O-rutinoside, formononetin, verbascoside and meditartin were 0.97%, 1.56%, 1.47%, 1.90%, 2.03%, 0.84%, 0.96%, 1.01% and 0.96%, respectively.

[0055] 2.4.5 Stability test

[0056] The test solutions from the same batch were accurately aspirated and stored at 4°C in the dark. The measurements were carried out at 0, 4, 8, 12 and 24 h, and the corresponding peak areas were recorded. The RSDs of the peak areas of hydroxysafflor yellow A, dehydrated safflor yellow B, safflor red pigment, syringin, calycosin, kaempferol-3-O-rutinoside, formononetin, calycosin and medipterostilbene were 2.64%, 1.44%, 2.50%, 2.64%, 1.27%, 1.02%, 2.53%, 2.54% and 2.15%, respectively. The results showed that the test solution was stable at 4°C in the dark within 24 h.

[0057] 2.4.6 Sample recovery test

[0058] Take 6 portions of 75 mg of Astragalus powder and 25 mg of Carthamus tinctorius with measured content, weigh them accurately, add reference solution with equal content to each portion, and prepare the test sample under the extraction conditions of ultrasonic 75 min, temperature of 65°C, DES concentration of 70%, and solid-liquid ratio of 50 mg / ml. Samples were injected respectively to calculate the sample recovery rate and RSD. The average recoveries of hydroxysafflor yellow A, dehydrated safflor yellow B, safflor red, syringin, verbascoside, kaempferol-3-O-rutinoside, formononetin, verbascoside and pteranoside were 95.87% (RSD=1.12%), 97.19% (RSD=1.56%), 101.31% (RSD=2.69%), 99.50% (RSD=2.08%), 101.27% (RSD=1.98%), 102.32% (RSD=3.07%), 100.73% (RSD=2.01%), 98.93% (RSD=3.11%) and 100.46% (RSD=2.15%).

[0059] 2.5 Preparation and screening of natural deep eutectic solvents (NADESs)

[0060] 2.5.1 Preparation of NADESs

[0061] Nine types of NADESs, including Choline Chloride-Lactic acid (ChCl-Lac), Choline Chloride-Urea (ChCl-Ur), Choline Chloride-Glycerinum (ChCl-Gly), Proline-urea (L-Proline-Urea, Pro-Ur), Proline-lactic acid (L-Proline-Lacticacid, Pro-Lac), Proline-glycerol (L-Proline-Glycerinum, Pro-Gly), Betaine-Glycerinum (Bet-Gly), Betaine-Lactic acid (Bet-Lac) and Betaine-urea (Bet-Urea, Bet-Ur), were selected for preparation. HBA, HBD and water were mixed according to the molar ratios in Table 2, heated in an oil bath and maintained at 80°C until a clear liquid was formed. During the heating period, magnetic stirring was maintained to ensure that the molten liquid was evenly mixed.

[0062] Table 2 Different types of NADESs

[0063]

[0064] #The molar ratio of water is in brackets

[0065] 2.5.2 Screening of NADESs

[0066] In order to investigate the extraction rate of 9 target components in the Astragalus and Safflower drug pair by different types of NADESs, the best NADES type was selected, and Astragalus and Safflower powder were accurately weighed in a ratio of 3:1, placed in a conical flask, and the above 9 NADESs were respectively transferred, and 70% NADESs were added at a solid-liquid ratio of 70 mg / mL, sealed, shaken, and ultrasonically extracted at 70°C for 45 minutes. The extract was centrifuged at 4500rpm for 15 minutes, and the supernatant was taken and diluted with an equal volume of water, and passed through a 0.22μm microporous filter membrane. Finally, HPLC was used to quantify the 9 active ingredients in each extract, calculate the extraction rate, and screen out the best extraction solvent. The extraction results are shown in Table 3.

[0067] Among the 9 target ingredients, hydroxysafflor yellow A, anhydrosafflor yellow B, safflower red pigment, syringin, calycosin, kaempferol-3-O-rutinoside and formononetin reached the highest extraction rates in Bet-Lac solvent, which were 39.438, 17.403, 7.377, 0.423, 0.182, 1.130 and 0.113 mg / g, respectively; medipteroside and calycosin reached the maximum extraction rates in Pro-Gly and Chcl-Lac solvents, which were 0.185 and 0.453 mg / g, respectively. In addition, the extraction rates of hydroxysafflor yellow A, dehydrated safflor yellow B, kaempferol-3-O-rutinoside, formononetin and calycosin (39.438, 17.403, 1.130, 0.113 and 0.368 mg / g, respectively) by Bet-Lac solvent were higher than those by hot water extraction (based on literature: Chen Guangwei, Jin Weifeng, Xu Shouchao, Zhang Ling, Wang Yu, He Yu*. Spherical symmetry design combined with genetic neuroscience The water extraction process of Astragalus membranaceus-Safflower drug pair was optimized through the Internet. Chinese Herbal Medicine, 2021, 52(8): 2257-2266. The extraction conditions were as follows: liquid-to-solid ratio of 22.7:1, extraction time of 30 min, extraction temperature of 87 °C, and extraction times of 2 times. The extraction rates of each component were 38.952, 4.797, 0.685, 0.036, and 0.064, respectively. Fat-soluble components that could not be extracted by water extraction, such as safflower red pigment, can also be extracted.

[0068] In the ultrasonic extraction method, the extraction rates of the nine components extracted by Bet-Lac were higher than those extracted by using only water, 70% ethanol or 95% ethanol as solvents. Therefore, Bet-Lac solvent was selected as the optimal extraction solvent for the following experiments.

[0069] Table 3 Extraction rate of active ingredients of Astragalus safflower herb by different types of NADESs (mg / g)

[0070]

[0071] 2.6 Single factor test (the influence of each factor on the total extraction rate is shown in Figure 2 )

[0072] 2.6.1 Effect of NADES concentration on total yield

[0073] Weigh the astragalus and safflower powder in a ratio of 3:1, place in a conical flask, add 50%, 60%, 70%, 80%, 90% Bet-Lac at a liquid-to-solid ratio of 70 mg / mL, seal, shake, and extract at 70℃ ultrasonic for 45 minutes. Centrifuge the extract at 4500rpm for 15 minutes, take the supernatant and dilute it with an equal volume of ultrapure water, filter it through a 0.22μm microporous membrane, detect the 9 active ingredients, and calculate the total extraction rate. The results are as follows Figure 2A. At 70% concentration of Bet-Lac, the extraction rate reached the maximum, which was 72.99 mg / g, so 60, 70, and 80% were selected as the three levels of NADES concentration factor.

[0074] 2.6.2 Effect of ultrasonic time on total extraction rate

[0075] Weigh the astragalus and safflower powder in a ratio of 3:1, place in a conical flask, add 70% Bet-Lac at a liquid-to-solid ratio of 70 mg / ml, seal, shake, and extract by ultrasonic at 70°C for 45, 60, 75, 90, and 105 min. Centrifuge the extract at 4500 rpm for 15 min, take the supernatant and dilute with an equal volume of ultrapure water, filter through a 0.22 μm microporous membrane, and detect the 9 active ingredients by HPLC and calculate the total extraction rate. The results are shown in the figure. Figure 2 D. When the extraction time reached 90 min, the extraction rate reached the maximum, which was 81.83 mg / g, so 75, 90, and 105 min were selected as the three levels of the ultrasonic time factor.

[0076] 2.6.3 Effect of ultrasonic temperature on total extraction rate

[0077] Weigh the astragalus and safflower powder accurately in a ratio of 3:1, place in a conical flask, add 70% Bet-Lac at a liquid-to-solid ratio of 70 mg / mL, seal, shake well, and perform ultrasonic extraction for 45 min at 60°C, 65°C, 70°C, 75°C, and 80°C. Centrifuge the extract at 4500 rpm for 15 min, take the supernatant and dilute with an equal volume of ultrapure water, filter through a 0.22 μm microporous membrane, and detect 9 active ingredients and calculate the total extraction rate. Figure 2 C. When the extraction temperature reached 70°C, the extraction rate reached the maximum, which was 73.48 mg / g. 65, 70, and 75°C were selected as the three levels of the temperature factor.

[0078] 2.6.4 Effect of material-liquid ratio on total extraction rate

[0079] Weigh the astragalus and safflower powder in a ratio of 3:1, place in a conical flask, add 70% Bet-Lac at a solid-liquid ratio of 30, 40, 50, 60 and 70 mg / mL, plug, shake, and extract at 70℃ for 45 minutes. Centrifuge the extract at 4500rpm for 15 minutes, take the supernatant and dilute with an equal volume of ultrapure water, filter through a 0.22μm microporous membrane, and detect the 9 active ingredients by HPLC and calculate the total extraction rate. The results are shown in Figure 2 B. When the solid-liquid ratio reached 50 mg / mL, the extraction rate reached the maximum, which was 89.16 mg / g, so 40, 50, and 60 mg / mL were selected as the three levels of the solid-liquid ratio.

[0080] 2.7 Experimental Design and Results

[0081] Based on the above single-factor test results, the corresponding single factor was selected as the independent variable, and the comprehensive score of the 9 components by the entropy weight method was used as the response value. The Design-Expert software was used to carry out a combined design experiment according to the Box-Behnken design. Each group was measured three times in parallel and the average value was taken.

[0082] Table 4 Factor levels of process optimization

[0083]

[0084] Table 5 Response surface test scheme and test results

[0085]

[0086] 2.8 Overall Rating

[0087] Since there are many components measured in this experiment, and the content varies greatly, it is not convenient for subsequent optimization, so the weighting method is chosen for comprehensive evaluation. The subjective weighting method has a certain degree of arbitrariness and subjectivity, and is not suitable for weighting objective things, so the objective weighting method is chosen. Criteria Importance Through Intercrieria Correlation (CRITIC Weighting Method). In addition to considering the amount of information contained in the indicators, the correlation between the indicators is also considered. The two indicators of contrast intensity and conflict are used to characterize these two types of information, and the objective properties of the data are fully utilized for scientific evaluation. Among them, the contrast intensity is represented by the mean square error. The larger the mean square error, the greater the weight; the conflict is represented by the correlation coefficient between the indicators. The larger the correlation coefficient, the stronger the correlation, and the correspondingly lower the conflict, the smaller the weight. The specific calculation steps are as follows:

[0088] (1) Standardize the original indicators

[0089]

[0090] (2) Calculate the standard deviation s of each index component j And the correlation coefficient matrix R = (r ij ) m×n

[0091]

[0092]

[0093] (3) Calculate the conflict between indicators c j

[0094]

[0095] (4) Calculate the information amount I of each indicator j

[0096] I j =s j ·c j

[0097] (5) Weight w of each indicator j

[0098]

[0099] (6) Comprehensive score Y extracted from each group i The formula is as follows

[0100]

[0101] In this experiment, the weights of hydroxysafflor yellow A, dehydrated safflor yellow B, safflor red pigment, syringin, calycosin, kaempferol-3-O-rutinoside, formononetin, calycosin, and pterostilbene were 0.1038, 0.0980, 0.0884, 0.0693, 0.1433, 0.0833, 0.1393, 0.1496, and 0.1251, respectively. Therefore, the calculation formula of the comprehensive score is as follows:

[0102] Y=0.1038c 1 +0.0980c 2 +0.0884c 3 +0.0693c 4 +0.1433c 5 +0.0833c 6 +0.1393c 7 +0.1496c 8 +0.1251c 9

[0103] 2.9 Condition Optimization and Verification

[0104] 2.9.1 Response Surface Model

[0105] Design-Expert 8.0.6.1 software was used to analyze and process the experimental data obtained, and the model with the highest quadratic term was selected for fitting. At the same time, variance analysis was performed on the obtained equation. The fitting formula is as follows:

[0106]

[0107] in, is the predicted value of the total extraction rate.

[0108] The results of variance analysis (Table 6) showed that the equation obtained by response surface optimization was not significant (p>0.05), indicating that the response surface model could not well explain the relationship between extraction conditions and results, so the genetic neural network model was subsequently tried.

[0109] Table 6 Analysis of variance of response surface model

[0110]

[0111] 2.9.2 Genetic Neural Network Model

[0112] Construct a 3-layer genetic neural network model, and use 22 groups of experimental data as training sets of the neural network model, and 5 groups as test sets. Used to train and verify the neural network. The minimum mean absolute error is used as the objective function, the four conditions of time, temperature, NaDES concentration, and material-liquid ratio are used as input neurons, and the comprehensive score is used as the output neuron. The number of hidden layer neurons cannot be determined, so we can only keep trying to adjust the number of hidden layer neurons, compare the fit gap between the training set and the validation set, and determine whether the model has "overfitting" or "underfitting". Finally, the number of hidden layer neurons was selected as 3, and the neural network structure was 4*3*1, see Figure 3 Using the Matlab toolbox, the population size was set to 20, the maximum genetic generation was set to 5, and other parameters were kept as default for training. The training set and the prediction set R 2 The neural network model parameters are shown in Table 7. The neural network prediction model results are shown in Table 7. Figure 4 .

[0113] The obtained model parameters were used to optimize and predict the optimal extraction conditions and comprehensive scores, and the extraction conditions were as follows: extraction time 90.312 min, temperature 67.154 °C, Bet-Lac concentration 80.262%, liquid-to-solid ratio 50.378 mg / mL, and the comprehensive score was 0.5930.

[0114] Table 7 Connection weights and thresholds in the neural network model

[0115]

[0116] Note: X is the input layer neuron; H is the hidden layer neuron.

[0117] 2.10 Model Validation

[0118] Considering the practical feasibility, the obtained optimized conditions are approximated as extraction time 90min, temperature 65℃, Bet-Lac concentration 80%, and liquid-to-solid ratio 50mg / mL. According to the above-obtained optimal extraction conditions, the test solution was prepared according to the operation of item "2.1", and the chromatographic conditions under item "2.3" were measured and the contents of 9 target components were calculated, and the comprehensive value was calculated. The results are shown in Table 8. Under the optimal process conditions obtained by the genetic neural network, the average comprehensive score was 0.5813, which was only 1.97% different from the predicted score of 0.5930 given by the model.

[0119] Table 8 Experimental results of genetic neural network model optimization condition verification

[0120]

[0121] 2.11 Experimental animals and preparation of middle artery embolism model

[0122] Twenty-four SPF SD rats, half male and half female, were housed in the Experimental Animal Center of Zhejiang Chinese Medical University (certificate number: SYXK 2018-0012): temperature 24 ± 2 °C, humidity 45 ± 5%, and 12 h light / dark cycle.

[0123] After one week of adaptive feeding, the model group rats were subjected to middle cerebral artery occlusion (MCAO) model according to the Longa suture method. After 1 hour of ischemia caused by suture insertion, the suture was removed to achieve reperfusion. The sham operation group was operated in the same manner except that the suture was not inserted.

[0124] 2.12 Rat grouping, drug administration and blood sampling

[0125] Twelve female and 12 male rats were randomly divided into 4 groups (sham-operated rats + water extract group, sham-operated rats + Bet-Lac extract group, model rats + water extract group and model rats + Bet-Lac extract group, 3 rats in each group). All rats were fasted overnight before surgery or administration and had free access to water. Eight groups of SD rats were orally administered with the water extract of Astragalus and Safflower herbal pair or the Bet-Lac low eutectic solvent extract at a dose of 10.8 g / kg. After oral administration, blood samples of about 0.3 mL each time were collected from the orbital venous plexus at 5, 15, 30, 45, 60, 120, 240, 360, 480, and 720 min and transferred to heparinized centrifuge tubes. During the collection period, the blood volume was supplemented by intraperitoneal injection of physiological saline twice. Each blood sample was immediately centrifuged at 4000 rpm for 10 min, and the supernatant was stored at -80°C.

[0126] 2.13 Sample preparation

[0127] Accurately pipette 100 μL of plasma sample into a 1.5 mL centrifuge tube, accurately add 10 μL of puerarin and digoxin internal standard (IS) solution, vortex for 1 min, and mix thoroughly; add 0.5 mL of methanol, vortex for 1 min, centrifuge at 12000 r / min for 10 min, and transfer 560 μL of supernatant to a new centrifuge tube. Blow dry with nitrogen, re-dissolve with 100 μL of methanol, vortex for 1 min, centrifuge at 12000 rpm for 10 min, and take the supernatant for determination.

[0128] 2.13 UPLC-MS / MS conditions

[0129] The MS detection sensitivity and in vivo content of 9 compounds quantified during the extraction stage were investigated, and 5 components, namely syringin, calycosin, formononetin, calycosin and hydroxysafflor yellow A, were selected for pharmacokinetic analysis.

[0130] The determination of the five components in rat plasma was performed using a liquid-mass spectrometer, using positive and negative ion modes, positive ion detection mode, ionization voltage 5500V, negative ion detection mode, ionization voltage -4500V. The ion source temperature was 500°C, the nebulizer was 50psi, the auxiliary heating gas was 50psi, the air curtain gas was 50psi, and the collision gas was high-purity nitrogen. The ion scan range was set to m / z 100-1000. Chromatographic separation was performed using CAPCELL PAK ADME HR C 18 Column (4.6 mm × 150 mm, 3 μm). The mobile phase was 0.1% formic acid (A)-acetonitrile (B). The gradient elution program was as follows: 0-2 min, 10-31% B, 2-6 min, 31-45% B, 6-18 min, 45-49% B, 18-20 min, 49-95% B, 20-22 min, 95-10% B. Column temperature 40 ° C, flow rate 0.5 mL / min, injection volume 10 μL.

[0131] Table 9 Fragmentation conditions of each detection component in positive and negative ion mode

[0132]

[0133] DP: Declustering potential; EP: Entry potential; CXP: Collision cell exit potential

[0134] 2.12 UPLC-MS / MS Methodology Review

[0135] 2.12.1 Specificity inspection

[0136] Compare the LC-MS spectra of blank plasma samples, rat plasma samples after administration, and blank plasma samples with mixed standards and internal standard samples added ( Figure 5). The blank plasma samples showed no interference with the five target analytes detected.

[0137] 2.12.2 Linear Relationship

[0138] Accurately pipette 100 μL of blank rat plasma, and accurately add 50uL of mixed reference solution with gradient concentrations, 10μL of puerarin and digoxin internal standard solution respectively, to obtain reference solution with a series of concentrations. Analyze according to the above liquid quality conditions, use the peak area ratio X of reference substance to internal standard as the horizontal coordinate, and the reference concentration Y as the vertical coordinate, and calculate the regression equation and correlation coefficient (r).

[0139] Table 10 Standard curves and linear ranges of each target component

[0140]

[0141] 2.12.3 Precision and accuracy assessment

[0142] High, medium and low concentration plasma quality control samples were prepared to determine intra-day precision, inter-day precision and accuracy.

[0143] Table 11 Precision and accuracy inspection results

[0144]

[0145] 2.12.4. Stability study

[0146] Six plasma quality control samples of high, medium and low concentrations were collected to examine the content of each component under the following three storage conditions.

[0147] Table 12 Stability Study

[0148]

[0149]

[0150] 2.12.5 Investigation of recovery rate and matrix effect

[0151] The results of the recovery rate and matrix effect are shown in Table 13.

[0152] Table 13 Recovery and matrix effect

[0153]

[0154] 2.13 Pharmacokinetics of Bet Lac Extract and Water Extract in Rats

[0155] Due to the low volatility of NaDES, in order to facilitate oral administration to rats, the solid-liquid ratio was adjusted to 300 mg / mL, while other extraction parameters remained unchanged. The data of each time point and the corresponding concentration were processed using DAS 2.0 software, and the statistical moment model was selected to obtain the relevant pharmacokinetic parameters, as shown in Table 14. max , C max , AUC 0-∞ , MRT 0-t ) as the independent variable, each control factor (extraction solvent, gender and surgery) as the dependent variable, and body weight as the covariate of multivariate ANOVA. The results of multivariate ANOVA are shown in Table 15.

[0156] The results in Table 15 show that body weight and gender do not affect metabolism overall (P>0.05), and surgery and solvent have significant effects on the pharmacokinetic parameters of bioactive components in rats, so the paired comparison of solvent and surgery was further studied, and the results are shown in Table 16. The extraction solvent has a significant effect on the metabolism of each component in rats. Compared with water as the extraction solvent, Bet-Lac increased the peak time (T max ) was advanced by an average of 1.089 h (CI = [0.531, 1.647], P < 0.01), resulting in a retention time (MRT 0-t ) was shortened by an average of 0.719 h (CI = [0.415, 1.023], P < 0.01), and the maximum blood concentration (C max ) increased by 11834.492 ng / mL (CI = [10954.162, 12714.822], P < 0.01), and the area under the curve (AUC 0-∞ ) increased by 61775.528 ng / mL*h (CI=[52077.516,71473.540], P<0.01).

[0157] The effects of surgery and solvent on the pharmacokinetic parameters of each tested component were further analyzed, and the results are shown in Table 17. For example, for syringin, Bet-Lac increased C max The average increase was 50.530 ng / mL (CI = [23.121, 77.940], P < 0.01), AUC 0-∞ The average increase was 246.261 ng / mL*h (CI = [119.108, 373.415], P < 0.01), MRT 0-t The decrease was 0.802 h (CI = [0.287, 1.318], P < 0.01). In addition, Bet-Lac also had similar effects on other components, that is, the low eutectic solvent significantly improved the oral bioavailability of the main components of the Astragalus and Safflower combination.

[0158] Table 14 Pharmacokinetic parameters of the components of the Astragalus and Safflower drug pair after administration with different solvents

[0159]

[0160] Table 15 Multivariate ANOVA results

[0161]

[0162]

[0163] Table 16 Pairwise comparisons of statistically significant control variables: solvent and surgery

[0164]

[0165] *P<0.05, **P<0.01

[0166] Table 17 Comparison of solvent and surgical pairings for each target analyte

[0167]

[0168]

[0169] *P<0.05.

Claims

1. A method for ultrasonically extracting effective components of Astragalus and Safflower drug pair using a natural deep eutectic solvent, the method comprising: accurately weighing Astragalus and Safflower powder in proportion, adding a natural deep eutectic solvent, and ultrasonically extracting; after centrifuging the extract, taking the supernatant, diluting it with ultrapure water, and passing it through a porous filter membrane; quantifying 9 effective components in each extract by HPLC; the natural deep eutectic solvent is a betaine-lactic acid solvent; in, Extraction time 85-90min, temperature 65-75°C, betaine-lactic acid concentration 70-80%, solid-liquid ratio 48-52mg / mL; HPLC-DAD content determination chromatographic conditions are: Chromatographic column: Alltech Alltima C 18 Chromatographic column, 250 mm × 4.6 mm, 5 μm; guard column: Alltech Alltima C 18 Guard column, 12.5 mm × 4.6 mm; flow rate: 1.0 mL / min; column temperature: 30°C; mobile phase A: acetonitrile, mobile phase B: 0.01% trifluoroacetic acid aqueous solution; gradient elution program: 0-7 min, 10%-14% A; 7-10 min, 14%-18% A; 10-20min, 18%-20%A; 20-27min, 20%-20%A; 27-30min, 20%-23%A; 30-36min, 23%-26%A; 36-44min, 26%-34%A; 44-53min, 34%-50%A; Detection wavelength: 403nm to detect hydroxysafflor yellow A and dehydrated safflor yellow B, 260nm to detect calycosin, calycosin glycosides, formononetin, kaempferol-3-O-rutinoside and syringin, 210nm to detect medipteroside, 520nm to detect safflor red pigment; The preparation method of the natural deep eutectic solvent is as follows: betaine, lactic acid and water are mixed in a molar ratio of 1:1:1, respectively, and heated in an oil bath and maintained at 80° C. until a clear liquid is formed, and magnetic stirring is maintained during the heating to ensure that the melted liquid is evenly mixed; The nine active ingredients are calycosin, calycosin glycosides, formononetin, hydroxysafflor yellow A, dehydrated safflor yellow B, meditartin, syringin, kaempferol-3-O-rutinoside and safflor red pigment.

2. The method according to claim 1, Features: The extraction time was 90 min, the temperature was 65° C., the betaine-lactic acid concentration was 80%, and the solid-liquid ratio was 50 mg / mL.

3. The method according to claim 1, Features: Accurately weigh astragalus and safflower coarse powder in a ratio of 3:

1.

4. The method according to claim 1, 2 or 3, Features: The steps include: (1) Preparation of test solution Astragalus and safflower powders were accurately weighed according to the commonly used clinical dosage ratio, natural low eutectic solvent was added, and ultrasonic extraction was performed; after the extract was centrifuged, the supernatant was diluted with ultrapure water and passed through a porous filter membrane to obtain the test solution; (2) Preparation of reference solution Accurately weigh the reference substances of calycosin, calycosin glycosides, formononetin, hydroxysafflower yellow A, dehydrated safflower yellow B, meditartin, syringin, kaempferol-3-O-rutinoside and safflower red pigment, and add methanol to prepare reference substance solutions respectively; accurately pipette each reference substance solution to prepare a mixed reference substance solution; (3) HPLC-DAD content determination.

5. The method according to claim 4, Features: Step (2): Weigh appropriate amounts of formononetin, formononetin glucoside, ononin, hydroxysafflor yellow A, anhydro safflor yellow B, medicarpin, syringin, kaempferol-3-O-rutinoside and safflower red pigment reference substances accurately, and prepare reference substance solutions with methanol at concentrations of 2.330, 2.000, 2.556, 4.450, 3.880, 2.330, 1.005, 2.330, 4.000 mg / mL respectively; accurately pipette each reference substance solution and prepare a mixed reference substance solution containing 210 μg of formononetin, 180 μg of formononetin glucoside, 230 μg of ononin, 400 μg of hydroxysafflor yellow A, 350 μg of anhydro safflor yellow B, 210 μg of medicarpin, 90 μg of syringin, 210 μg of kaempferol-3-O-rutinoside and 400 μg of safflower red pigment in every 1 mL of the solution.

6. According to the method described in claim 1, it is characterized in that: it further includes the step of optimizing and predicting the optimal extraction conditions and comprehensive score by using a genetic neural network model.

Images