A quality detection method for traditional Chinese medicine compound based on fingerprint spectrum

Abstract

A quality detection method for traditional Chinese medicine compound based on fingerprint spectrum, where the traditional Chinese medicine compound is Gancaoxiexin Decoction. First, establish the fingerprint spectrum of Gancaoxiexin Decoction, then conduct anti-inflammatory and antibacterial activity tests. Perform PCA analysis on the antibacterial test results to obtain antibacterial comprehensive pharmacodynamic indexes. Establish a spectral-efficacy relationship mathematical model for anti-inflammatory and antibacterial effects through OPLS analysis, calculate the GRA correlation degree, and obtain drug components related to anti-inflammatory and antibacterial effects. By establishing the fingerprint spectrum of the whole formula in Gancaoxiexin Decoction, further establish the UPLC fingerprint spectrum of its water-soluble part and the GC-MS fingerprint spectrum of its fat-soluble part respectively. Conduct anti-inflammatory and antibacterial tests on the whole formula, water-soluble components and fat-soluble components respectively, and combine the spectrum analysis to obtain a mathematical model related to anti-inflammatory and antibacterial effects for each component, quickly and effectively detecting the quality advantages and disadvantages of each batch of Gancaoxiexin Decoction regarding anti-inflammatory and antibacterial effects, and accurately controlling its quality.

Description

Technical Field

[0001] The present invention relates to the technical field of pharmaceutical ingredient detection, and particularly relates to an anti-inflammatory and antibacterial quality detection method for traditional Chinese medicine compound based on fingerprint. Background Art

[0002] Gancao Xiexin Decoction is a name of traditional Chinese medicine prescription, which comes from Treatise on Febrile Diseases. Clinically, Gancao Xiexin Decoction has achieved good therapeutic effects in the treatment of recurrent oral ulcers. Recurrent oral ulcers are considered as non-specific inflammatory manifestations, and bacteria, viruses, etc. are the main causes of inflammation. Therefore, Gancao Xiexin Decoction also has anti-inflammatory and antibacterial effects.

[0003] Ultra-high performance liquid chromatography (UPLC) is a chromatographic analysis method. Compared with traditional HPLC, UPLC can separate more than twice as many chromatographic peaks as HPLC. Under the same conditions, the resolution of UPLC can recognize more chromatographic peaks, with high separation degree, fast detection speed, and significantly improved sensitivity. However, a single UPLC detection method can only reflect the quality of drugs through the detection of drug ingredient types. However, the composition of drug raw materials and extraction methods of Gancao Xiexin Decoction will have a greater impact on its efficacy. Different batches of Gancao Xiexin Decoction cannot guarantee that it can effectively exert excellent anti-inflammatory and antibacterial effects and ensure its quality stability. In order to effectively monitor the drug quality, it is necessary to identify its drug ingredients and determine the effective components with anti-inflammatory and antibacterial effects. Using fingerprint technology to detect drug components has become a more efficient and convenient quality detection method. Summary of the Invention

[0004] The purpose of the present invention is to provide an anti-inflammatory and antibacterial quality detection method for Gancao Xiexin Decoction based on fingerprint, which can effectively and quickly detect the anti-inflammatory and antibacterial effects of any batch of Gancao Xiexin Decoction, and the detection results are highly accurate.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] A quality detection method for traditional Chinese medicine compound based on fingerprint, characterized in that: the traditional Chinese medicine compound is Gancao Xiexin Decoction. First, establish the fingerprint of Gancao Xiexin Decoction, then conduct anti-inflammatory and antibacterial activity tests. Analyze the results of antibacterial tests by principal component analysis (PCA) to obtain antibacterial comprehensive efficacy indicators. Analyze the spectral-effect relationship between the anti-inflammatory efficacy results and antibacterial efficacy results and the fingerprint. Establish a spectral-effect relationship mathematical model for anti-inflammatory and antibacterial through orthogonal projection partial least squares (OPLS) analysis. Calculate the correlation degree by grey relational analysis (GRA) to obtain the drug components related to anti-inflammatory and antibacterial.

[0007] In the present invention, fingerprint spectrum is combined with PCA analysis, OPLS analysis and GRA correlation value joint analysis, and each component detected in the fingerprint spectrum is associated with anti-inflammatory and antibacterial activities, realizing more accurate quantitative and qualitative analysis of complex mixtures. Through complementary advantages, the drug quality can be detected more precisely.

[0008] Further, the method for establishing the UPLC fingerprint spectrum of the whole formula of Gancaoxiexin Decoction is characterized by including the following steps:

[0009] (1) Preparation of test solution

[0010] Weigh the whole formula composition with a single dose of 20 g of Glycyrrhizae Radix Preparata, 15 g of Scutellariae Radix, 15 g of Zingiberis Rhizoma Recens, 12 g of Jujubae Fructus, 13 g of Pinelliae Rhizoma, and 5 g of Coptidis Rhizoma, that is, a total of 80 g for a single dose. Extract the sample using the semi-bionic extraction method. The extraction conditions are as follows: the pH of the first decoction water is 4.0, the pH of the second decoction water is 7.0, the pH of the third decoction water is 8.0, and the total decoction time is 2 h, including 1.0 h for the first decoction, 0.5 h for the second decoction, and 0.5 h for the third decoction. The extraction temperature is 95 °C, the liquid-to-material ratio is 1:10, and reflux extraction is carried out; the above pH water is adjusted with HCl (0.1 mol·L -1 ) and NaOH (0.1 mol·L -1 ); Combine the extraction solutions of the three times, and concentrate them under reduced pressure at 55 °C to 1 g·mL -1 , which is used as the sample mother liquor and stored at 4 °C for later use. At this time, the obtained sample mother liquor is the whole formula sample;

[0011] (2) Preparation of reference solution:

[0012] Take appropriate amounts of 11 references including zingerone, coptisine hydrochloride, liquiritigenin, berberine hydrochloride, wogonoside, baicalein, 6-gingerol, ammonium glycyrrhizinate, baicalin, liquiritin, and wogonin as the mixed reference for the whole formula sample, dissolve them in methanol to a concentration of 1.0 mg·mL -1 , and store them in a refrigerator at 4 °C as the reference solution;

[0013] (3) Optimal UPLC chromatographic conditions

[0014] The chromatographic column is Shim-pack Scepter C18-200 (3 μm, 4.6×250 mm), the detection wavelength is 277 nm, the flow rate is 0.3 mL·min -1 , the injection volume is 3 μL, and the column temperature is 15 °C;

[0015] Gradient elution: A is an aqueous solution of formic acid with a volume fraction of 0.4%, B is acetonitrile. From 0 to 5 min, the volume percentage of mobile phase B increases from 2% to 6%; from 5 to 10 min, the volume percentage of mobile phase B increases from 6% to 8%; from 10 to 15 min, the volume percentage of mobile phase B increases from 8% to 15%; from 15 to 20 min, the volume percentage of mobile phase B increases from 15% to 20%; from 20 to 30 min, the volume percentage of mobile phase B increases from 20% to 23%; from 30 to 40 min, the volume percentage of mobile phase B increases from 23% to 32%; from 40 to 60 min, the volume percentage of mobile phase B increases from 32% to 35%; from 60 to 80 min, the volume percentage of mobile phase B increases from 35% to 50%; from 80 to 85 min, the volume percentage of mobile phase B increases from 50% to 80%; from 85 to 90 min, the volume percentage of mobile phase B increases from 80% to 98%; from 90 to 95 min, the volume percentage of mobile phase B decreases from 98% to 60%; from 95 to 100 min, the volume percentage of mobile phase B decreases from 60% to 35%; from 100 to 105 min, the volume percentage of mobile phase B decreases from 35% to 2%.

[0016] (4) Establishment of fingerprint

[0017] Take 16 batches of the medicinal materials of Gancaoxiexin Decoction, pulverize them, prepare 16 batches of samples of the whole prescription according to the extraction process, inject samples according to the chromatographic conditions, record the UPLC data (CDF format data), import the original data into the software of "Similarity Evaluation System for Chromatographic Fingerprints of Traditional Chinese Medicines" (2012 version), perform common peak matching by the multi-point calibration method to establish the fingerprint; take S11 as the reference fingerprint, the method for generating the control spectrum is the median method, the time window width is 0.5, and the spectrum spacing is 400 to obtain the fingerprints of 16 batches of samples; there are 61 common chromatographic peaks in the whole prescription. From the results of the spectrum, it can be seen that the similarities of the 16 batches of samples with the reference fingerprint (R) are 0.999, 0.999, 0.999, 0.997, 0.997, 0.996, 0.996, 0.997, 0.999, 0.997, 0.996, 0.997, 0.996, 0.996, 0.997, 0.997 in turn. The RSD value is 0.11%, and the similarity is very high, indicating that the differences among these 16 batches of samples are not significant. At the same time, compare the RSD of the elution time of the common peaks. The value is 0.01 - 0.29%, which is less than 1%, indicating that the elution times of the common peaks are consistent; compare the average RSD of the chromatographic peak areas of the common peaks, and the value is 1.22 - 29.80%.

[0018] Furthermore, the whole formula sample of Gancaoxiexin Decoction was used to conduct anti-inflammatory tests on RAW264.7 cells to obtain corresponding anti-inflammatory data. At the same time, antibacterial tests were carried out on Escherichia coli, Salmonella choleraesuis, and Proteus mirabilis to obtain three groups of antibacterial data. The SPSS 25 statistical analysis software was used to perform PCA analysis on the three groups of antibacterial data, and the comprehensive antibacterial data was obtained after standardization. The SIMCA-P 14.1 software was used with the peak areas of the common peaks in the whole formula sample as the independent variable X and the anti-inflammatory efficacy index as the dependent variable Y to perform OPLS analysis, and the regression equation was fitted: Y = a1X1 + a2X2 + … a 61 X 61 , where Y corresponds to the anti-inflammatory or antibacterial efficacy results, X1 to X 61 correspond to the peak areas of the common peaks, and a1 to a 61 correspond to the regression coefficients between the common peaks and the efficacy. A positive regression coefficient indicates a positive correlation with the efficacy, while a negative number indicates a negative correlation with the efficacy. Then, the GRA correlation degree value was calculated through EXCEL. When the correlation degree > 0.6, it can be used as an effective chromatographic peak for anti-inflammatory or antibacterial purposes.

[0019] Furthermore, through the above analysis, with the grey correlation degree > 0.6, the OPLS analysis regression coefficient being positive and the VIP value > 0.7 as the criteria, the chromatographic peaks related to anti-inflammatory in the whole formula part of Gancaoxiexin Decoction were screened out as 5, 6, 10, 13, 15, 20, 22, 24, 25, 37, 39, 44, 46, 47, 48, 51, 53, 54, 56, 58, 59, 60, a total of 22; the chromatographic peaks related to antibacterial were 3, 5, 8, 9, 11, 12, 14, 16, 17, 18, 19, 20, 21, 23, 26, 27, 28, 29, 30, 33, 35, 36, 37, 40, 41, 42, 43, 44, 45, 49, 50, 51, 54, 57, 58, 60, a total of 36; the chromatographic peaks related to both anti-inflammatory and antibacterial were 5, 20, 37, 44, 51, 54, 58, 60, a total of 8.

[0020] Combined with the previous UPLC-MS chromatographic analysis and reference chromatograms, 14 components were identified from the whole formula for the anti-inflammatory active ingredient group: phenylalanine, sucrose, isoliquiritigenin or glabridin or (S)-pinocembrin, liquiritin, magnoflorine, glycyrrhetinic acid, ziziphus jujuba var. spinosa saponin, wogonoside, glycyrin A, berberine hydrochloride, ammonium glycyrrhizinate, baicalein, 6-gingerol, and dihydroxy-dimethoxyflavone. 22 components were identified from the antibacterial active ingredient group: phenylalanine, glycyrrholide, glycyrin D,schaftoside, 5,7-dihydroxy-6-methoxyflavone-7-O-glucuronide, ononin, liquiritigenin-4'-O-β-D-glucopyranosyl-(1→6)-O-β-D-glucopyranoside, glabridin, cyclic adenosine monophosphate, glycyrin B or N, norwogonin-7-O-β-D-glucuronide, baicalin, magnoflorine, 5,7,8-trihydroxyflavone-7-O-glucoside, coptisine, gingerone, ziziphus jujuba var. spinosa saponin, liquiritigenin, ammonium glycyrrhizinate, baicalein, wogonin, dihydroxy-dimethoxyflavone. 6 components were identified as being related to both anti-inflammatory and antibacterial activities, namely phenylalanine, magnoflorine, ziziphus jujuba var. spinosa saponin, ammonium glycyrrhizinate, baicalein, and dihydroxy-dimethoxyflavone. These components belong to glycosides, flavonoids, and alkaloids.

[0021] Further preferably, the above-mentioned whole formula sample is further purified into a water-soluble part sample and a lipid-soluble part sample, and a UPLC fingerprint of the water-soluble part and a GC-MS fingerprint of the lipid-soluble part are established respectively.

[0022] Furthermore, for the further purification of the above-mentioned whole formula sample, 20 mL of the sample mother liquor is taken, 20 mL of ethyl acetate solvent is added, the extraction bottle is shaken for 5 min and then left to stand for 10 min. The upper layer solution is taken as the lipid-soluble part, and the lower layer solution is taken as the water-soluble part. Extraction is carried out 3 times respectively, and the 3 extraction solutions are combined. The lipid-soluble part extraction solution is evaporated to dryness on a rotary evaporator at 45 °C, and the water-soluble part extraction solution is evaporated to dryness on a rotary evaporator at 55 °C. Then they are respectively made up to 20 mL with ethyl acetate and 50% (v / v) methanol solution to obtain the lipid-soluble part sample solution and the water-soluble part sample solution, which are stored at 4 °C for standby.

[0023] Furthermore, for the establishment of the UPLC fingerprint of the water-soluble fraction, the water-soluble sample solution was injected according to the chromatographic conditions of the above-mentioned whole-formula sample, and the UPLC data (CDF format data) was recorded. The original data was imported into the software "Similarity Evaluation System for Chromatographic Fingerprints of Traditional Chinese Medicines" (2012 version), and the common peaks were matched by the multi-point calibration method to establish the fingerprint. There were 46 common characteristic fingerprint chromatographic peaks in the sample. S1-S16 were the water-soluble fraction samples of the 1st to 16th batches respectively. Taking S9 as the reference fingerprint, with a map interval of 400, the water-soluble sample fingerprint was obtained. From the map results, it can be seen that the similarities of the 16 batches of samples with the reference fingerprint (R) were 0.999, 1.000, 0.999, 0.999, 0.999, 0.999, 0.999, 0.998, 0.998, 0.998, 0.998, 0.998, 0.998, 0.998, 0.998, 0.998 in sequence. The RSD value of the similarity was 0.04%, indicating that the differences among the 16 batches of samples were not significant. The RSD value of the elution time of the common characteristic fingerprint chromatographic peaks was in the range of 0.01-0.35%, less than 1%, indicating that the elution times of the common peaks were highly consistent. The RSD value of their chromatographic peak areas was in the range of 1.47-29.69%, indicating that there were certain differences in the peak areas of individual components among samples of different batches.

[0024] Furthermore, the water-soluble fraction samples were used to conduct antibacterial tests against Escherichia coli, Salmonella choleraesuis, and Proteus mirabilis, and at the same time, anti-inflammatory tests were conducted through RAW264.7 cells to obtain the corresponding antibacterial data and anti-inflammatory data. Through PCA analysis, OPLS analysis, and GRA correlation analysis, it was found that the chromatographic peaks related to anti-inflammatory in the water-soluble fraction samples of Gancaoxiexin Decoction were 2, 3, 8, 9, 10, 12, 13, 17, 18, 23, 24, 25, 28, 29, 30, 32, 35, 40, 42, 44, a total of 20; the chromatographic peaks related to antibacterial in the water-soluble samples of Gancaoxiexin Decoction were 1, 9, 10, 11, 15, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 29, 30, 32, 33, 34, 36, 38, 41, 43, 44, a total of 25; the chromatographic peaks related to both anti-inflammatory and antibacterial were 9, 10, 17, 23, 24, 25, 29, 30, 32, 44, a total of 10.

[0025] Furthermore, 13 out of the 20 chromatographic peaks related to anti-inflammatory effects in the above-mentioned water-soluble fraction samples were identified, specifically 5,7-dihydroxy-6-methoxyflavone-7-O-glucuronide, sucrose, liquiritigenin-4'-O-β-D-glucopyranosyl-(1→6)-O-β-D-glucopyranoside, liquiritin, norwogonin-7-O-β-D-glucuronide, baicalin, magnoflorine, 5,7,8-trihydroxyflavone-7-O-glucoside, scutellarin, coptisine, wogonoside, berberine, and ammonium glycyrrhizinate. 14 out of the 25 chromatographic peaks related to antibacterial effects were identified, specifically sucrose, formononetin-7-O-β-D-glucoside, liquiritin, adenosine cyclic phosphate, palmatine hydrochloride, baicalin, 5,7,8-trihydroxyflavone-7-O-glucoside, magnoflorine, scutellarin, coptisine, wogonoside, liquiritigenin, and uralensis saponin P.

[0026] Furthermore, the GC-MS chromatographic and mass spectrometric conditions for establishing the GC-MS fingerprint of the lipophilic sample are as follows:

[0027] The inlet temperature is 260 °C, the interface temperature is 300 °C, in splitless mode, the carrier gas is helium (He), and the flow rate is 0.8 mL / min -1 , the pressure is 70 kPa, the injection volume is 0.5 μL, and the temperature programming is as follows: the initial temperature is 50 °C, held for 20 min, then increased to 150 °C at a rate of 3 °C / min -1 and held for 5 min, then further increased to 250 °C at a rate of 5 °C / min -1 and held for 15 min. The ionization mode is EI (70 eV), in full scan mode (m / z = 12 - 600), the ion source temperature is 230 °C, and the quadrupole temperature is 150 °C; inject the sample according to the GC-MS chromatographic conditions, record the data (in TXT format), import the original data into the "Similarity Evaluation System for Traditional Chinese Medicine Chromatographic Fingerprints" (2012 version) software, perform common peak matching by the multi-point calibration method, and establish the fingerprint; there are 26 common chromatographic peaks in the fingerprint, and S1 - S16 are 16 batches of lipophilic samples; using S10 as the reference fingerprint, the similarities of the 16 batches of samples to the control fingerprint (R) are 0.990, 0.997, 0.980, 0.999, 0.998, 0.998, 0.999, 0.994, 0.999, 0.999, 0.996, 0.975, 0.999, 0.998, 0.986, and 0.998 in sequence. The RSD value of the similarity is 0.73%, indicating that the differences among these 16 batches of samples are not significant. At the same time, the RSD value of the elution time of the common peaks is 0 - 0.04%, showing no difference in elution time; the RSD value of the chromatographic peak area is between 4.02 - 9.98%, so the content differences among different batches of lipophilic fraction samples are not large.

[0028] Furthermore, liposoluble fraction samples were used to conduct anti-inflammatory tests on RAW264.7 cells, and antibacterial tests were conducted using Escherichia coli, Salmonella choleraesuis, and Proteus mirabilis to obtain corresponding anti-inflammatory data and antibacterial data. Through PCA analysis, OPLS analysis, and GRA correlation analysis, 13 chromatographic peaks related to anti-inflammatory effects in the liposoluble fraction samples of Gancaoxiexin Decoction were screened out, namely peaks 2, 7, 9, 10, 13, 14, 16, 18, 19, 20, 21, 23, and 25; 14 chromatographic peaks related to antibacterial effects were peaks 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 18, 20, 23, and 25; and 10 chromatographic peaks related to both anti-inflammatory and antibacterial effects were peaks 7, 9, 10, 13, 14, 16, 18, 20, 23, and 25.

[0029] Furthermore, 10 out of the 13 chromatographic peaks related to anti-inflammatory effects in the above liposoluble fraction samples were identified as hexanal, vanillin lactoside, melezitose, p-hydroxycinnamic acid, ferulic acid, 1-(4-hydroxy-3-methoxyphenyl)-4-octen-3-one, (E)-1-(4-Hydroxy-3-methoxyphenyl)dec-3-en-5-one, 6-gingerol, 6-shogaol, and wogonin. 11 out of the 14 chromatographic peaks related to antibacterial effects were identified as vanillin lactoside, melezitose, vanillic acid, p-hydroxycinnamic acid, ferulic acid, palmitic acid, 1-(4-hydroxy-3-methoxyphenyl)-4-octen-3-one, 5-hydroxy-1-(4-hydroxy-3-methoxyphenyl)octan-3-one, (E)-1-(4-Hydroxy-3-methoxyphenyl)dec-3-en-5-one, 6-shogaol, and wogonin. 8 active ingredient groups related to both anti-inflammatory and antibacterial effects were identified as vanillin lactoside, melezitose, p-hydroxycinnamic acid, ferulic acid, 1-(4-hydroxy-3-methoxyphenyl)-4-octen-3-one, (E)-1-(4-Hydroxy-3-methoxyphenyl)dec-3-en-5-one, 6-shogaol, and wogonin. These ingredients belong to flavonoids, phenolic acids, glycosides, and polysaccharides.

[0030] The present invention has the following technical effects:

[0031] By establishing the fingerprint of the whole formula of Gancaoxiexin Decoction, further establishing the UPLC fingerprint of its water-soluble fraction and the GC-MS fingerprint of its liposoluble fraction, conducting anti-inflammatory and antibacterial tests on the whole formula, water-soluble components, and liposoluble components respectively, and combining with fingerprint analysis to obtain a mathematical model related to anti-inflammatory and antibacterial effects for each component, the present invention can quickly and effectively detect the quality of each batch of Gancaoxiexin Decoction in terms of anti-inflammatory and antibacterial effects, and accurately control its quality. Description of the Drawings

[0032] Figure 1 : Chromatogram of the control sample solution of the whole formula sample.

[0033] Figure 2 : Chromatogram of the reference substance solution used for the water-soluble part sample.

[0034] Figure 3 : Fingerprint of 16 batches of the whole formula of Gancaoxiexin Decoction in the present invention.

[0035] Figure 4 : Fingerprint of 16 batches of the water-soluble sample of Gancaoxiexin Decoction in the present invention.

[0036] Figure 5 : Fingerprint of 16 batches of the lipid-soluble sample of Gancaoxiexin Decoction in the present invention.

[0037] Figure 6 : Regression coefficient of the common peak and the anti-inflammatory efficacy of the whole formula sample.

[0038] Figure 7 : VIP value of the common peak and the anti-inflammatory efficacy of the whole formula sample.

[0039] Figure 8 : Regression coefficient of the common peak and the antibacterial efficacy of the whole formula sample.

[0040] Figure 9 : VIP value of the common peak and the antibacterial efficacy of the whole formula sample.

[0041] Figure 10 : Regression coefficient of the common peak and the anti-inflammatory efficacy of the water-soluble part sample.

[0042] Figure 11 : VIP value of the common peak and the anti-inflammatory efficacy of the water-soluble part sample.

[0043] Figure 12 : Regression coefficient of the common peak and the antibacterial efficacy of the water-soluble part sample.

[0044] Figure 13 : VIP value of the common peak and the antibacterial efficacy of the water-soluble part sample.

[0045] Figure 14 : Regression coefficient of the common peak and the anti-inflammatory efficacy of the lipid-soluble part sample.

[0046] Figure 15 : VIP value of the common peak and the anti-inflammatory efficacy of the lipid-soluble part sample.

[0047] Figure 16 : Regression coefficient of the common peak and the antibacterial efficacy of the lipid-soluble part sample.

[0048] Figure 17:VIP values of the common peaks and the antibacterial efficacy of the liposoluble fraction samples. Specific implementation manners

[0049] The present invention will be specifically described below through examples. It is necessary to point out here that the following examples are only used to further illustrate the present invention and should not be construed as limiting the protection scope of the present invention. Those skilled in the art can make some non-essential improvements and adjustments to the present invention according to the above content of the present invention.

[0050] Example 1:

[0051] The establishment of the UPLC fingerprint of the test solution extracted from the whole formula was carried out according to the following steps:

[0052] (1) Preparation of the test solution

[0053] Weigh according to a single dose of 20 g of Glycyrrhizae Radix et Rhizoma, 15 g of Scutellariae Radix, 15 g of Zingiberis Rhizoma Recens, 12 g of Jujubae Fructus, 13 g of Pinelliae Rhizoma, and 5 g of Coptidis Rhizoma, that is, a total of 80 g for a single dose. Weigh 16 batches of the whole formula, and extract the samples using the semi-bionic extraction method. The extraction conditions are as follows: the pH of the first decoction water is 4.0, the pH of the second decoction water is 7.0, the pH of the third decoction water is 8.0, and the total decoction time is 2 h, including 1.0 h for the first decoction, 0.5 h for the second decoction, and 0.5 h for the third decoction. The extraction temperature is 95 °C, the liquid-to-material ratio is 1:10, and reflux extraction is carried out; the above pH water is adjusted with HCl (0.1 mol·L -1 ) and NaOH (0.1 mol·L -1 ); Combine the extraction solutions of the three times, and concentrate them under reduced pressure at 55 °C to 1 g·mL -1 , which are used as the sample mother liquors respectively, and stored at 4 °C for later use. At this time, the obtained sample mother liquor is the whole formula sample;

[0054] According to the optimized extraction solvent ratio and constant volume solvent ratio, with 1:1 as the extraction solvent ratio to the extraction solution and 1:2 as the constant volume solvent ratio, take 20 mL of the sample mother liquor and 20 mL of ethyl acetate solvent, shake in an extraction bottle for 5 min, let it stand for 10 min, take the upper layer solution as the liposoluble fraction layer, and the lower layer solution as the water-soluble fraction layer. Extract 3 times, combine the extraction solutions of the three times respectively. The liposoluble fraction is evaporated to dryness on a rotary evaporator at 45 °C, and the water-soluble fraction is evaporated to dryness on a rotary evaporator at 55 °C, and then dissolved and fixed to 20 mL with ethyl acetate and 50% (v / v) methanol solution respectively. The mass concentrations of the water-soluble and liposoluble sample solutions are 0.5 g·mL -1 , and stored at 4 °C for later use;

[0055] (2) Preparation of the reference substance solution:

[0056] Weigh appropriate amounts of zingerone, coptisine hydrochloride, liquiritigenin, berberine hydrochloride, wogonoside, wogonin, 6-gingerol, ammonium glycyrrhizinate, baicalin, liquiritin, and wogonin as 11 reference substances for the whole formula sample mixed reference substance. Dissolve them in methanol to a concentration of 1.0 mg·mL -1 , and store them in a refrigerator at 4°C as the reference substance solution for later use. The chromatogram of the whole formula reference substance solution is as shown in Figure 1 . The corresponding peaks 1-11 are liquiritin, baicalin, coptisine hydrochloride, zingerone, wogonoside, berberine hydrochloride, liquiritigenin, liquiritigenin, ammonium glycyrrhizinate, wogonin, 6-gingerol, and wogonin respectively;

[0057] Weigh appropriate amounts of liquiritin, baicalin, wogonoside, coptisine hydrochloride, liquiritigenin, ammonium glycyrrhizinate, and berberine hydrochloride in the same way to prepare the reference substance solution for the water-soluble part. The corresponding chromatogram is as shown in Figure 2 . The corresponding peaks are 1. liquiritin; 2. baicalin; 3. coptisine hydrochloride; 4. wogonoside; 5. berberine hydrochloride; 6. liquiritigenin; 7. ammonium glycyrrhizinate;

[0058] (3) Optimal chromatographic conditions

[0059] UPLC chromatographic conditions for the whole formula sample and the water-soluble part sample: Chromatographic column Shim-pack ScepterC18-200 (3μm, 4.6×250mm), detection wavelength 277nm, flow rate 0.3mL·min -1 , injection volume 3μL, column temperature 15°C;

[0060] Gradient elution: A is an aqueous solution of formic acid with a volume fraction of 0.4%, B is acetonitrile. From 0 to 5 minutes, the volume percentage of mobile phase B increases from 2% to 6%; from 5 to 10 minutes, the volume percentage of mobile phase B increases from 6% to 8%; from 10 to 15 minutes, the volume percentage of mobile phase B increases from 8% to 15%; from 15 to 20 minutes, the volume percentage of mobile phase B increases from 15% to 20%; from 20 to 30 minutes, the volume percentage of mobile phase B increases from 20% to 23%; from 30 to 40 minutes, the volume percentage of mobile phase B increases from 23% to 32%; from 40 to 60 minutes, the volume percentage of mobile phase B increases from 32% to 35%; from 60 to 80 minutes, the volume percentage of mobile phase B increases from 35% to 50%; from 80 to 85 minutes, the volume percentage of mobile phase B increases from 50% to 80%; from 85 to 90 minutes, the volume percentage of mobile phase B increases from 80% to 98%; from 90 to 95 minutes, the volume percentage of mobile phase B decreases from 98% to 60%; from 95 to 100 minutes, the volume percentage of mobile phase B decreases from 60% to 35%; from 100 to 105 minutes, the volume percentage of mobile phase B decreases from 35% to 2%.

[0061] GC-MS chromatographic conditions for the liposoluble part of the sample: injection port temperature 260 °C, interface temperature 300 °C, splitless mode, carrier gas is helium (He), flow rate 0.8 mL·min -1 , pressure 70 kPa, injection volume 0.5 μL, temperature programming: initial temperature 50 °C, hold for 20 min, at 3 °C·min -1 rising to 150 °C, hold for 5 min, then at 5 °C·min -1 rising to 250 °C, hold for 15 min, ionization mode EI (70 eV), full scan mode (m / z = 12 - 600), ion source temperature 230 °C, quadrupole temperature 150 °C

[0062] (4) Establish the UPLC fingerprint

[0063] Take 16 batches of the medicinal materials of Gancaoxiexin Decoction, pulverize them, and prepare 16 batches of the whole formula samples, water-soluble parts, and liposoluble parts according to the extraction process. Among them, the whole formula samples and water-soluble samples are injected according to the above UPLC chromatographic conditions, and the chromatographic data (CDF format data) are recorded. The liposoluble part is injected according to the above GC-MS chromatographic conditions, and the chromatographic data (TXT format data) are recorded. The data are imported into the software of "Similarity Evaluation System for Traditional Chinese Medicine Chromatographic Fingerprints" (2012 version), and the common peaks are matched by the multi-point calibration method. The fingerprints of the whole formula, water-soluble part, and liposoluble part are established successively as Figures 3 - 5 shown. As Figure 3As shown, there are 61 common chromatographic peaks in the whole formula. S1 - S16 are the water-soluble fraction samples of the 1st to 16th batches respectively, and R is the reference chromatogram generated by the software. Taking S11 as the reference chromatogram, the method for generating the reference chromatogram is the median method, the time window width is 0.5, and the chromatogram spacing is 400, obtaining the fingerprint chromatograms of 16 batches of samples. From the chromatogram results, the similarities of the 16 batches of samples to the reference fingerprint chromatogram (R) are 0.999, 0.999, 0.999, 0.997, 0.997, 0.996, 0.996, 0.997, 0.999, 0.997, 0.996, 0.997, 0.996, 0.996, 0.997, 0.997 in sequence. The RSD value is 0.11%, and the similarity is very high, indicating that the differences among these 16 batches of samples are not significant. At the same time, comparing the RSD of the elution times of the common peaks, the value is 0.01 - 0.29%, less than 1%, showing that the elution times of the common peaks are consistent; comparing the RSD of the areas of the common peaks, the value is 1.22 - 29.80%, showing that there are certain differences in the contents among different batches, but the differences are small.

[0064] As Figure 3 shown, in the UPLC fingerprint chromatogram corresponding to the sample solution of the whole formula, there are 61 common peaks. Due to the large number of peaks, some peaks in the chromatogram overlap, and there is a certain interference between them, resulting in slightly lower reproducibility and accuracy.

[0065] As Figure 4 shown, there are 46 characteristic fingerprint chromatographic peaks in the water-soluble fraction samples. S1 - S16 are the water layer samples of the 1st to 16th batches respectively, and R is the reference chromatogram generated by the software. Taking S9 as the reference chromatogram, the time window width is 0.5, and the chromatogram spacing is 400, obtaining the fingerprint chromatograms of the water-soluble samples. From the chromatogram results, the similarities of the 16 batches of samples to the reference fingerprint chromatogram (R) are 0.999, 1.000, 0.999, 0.999, 0.999, 0.999, 0.999, 0.998, 0.998, 0.998, 0.998, 0.998, 0.998, 0.998, 0.998, 0.998 in sequence. The RSD value of the similarity is 0.04%, indicating that the differences among these 16 batches of samples are not significant. The RSD value of the elution times of the characteristic fingerprint peaks is in the range of 0.01 - 0.35%, less than 1%, showing that the elution times of the common peaks are highly consistent. The RSD value of their chromatographic peak areas is in the range of 1.47 - 29.69%, showing that there are certain differences in the peak areas of individual components among different batches of samples, but it does not affect the establishment of the fingerprint chromatogram.

[0066] As Figure 5As shown in the figure, the fingerprint chromatogram of the fat-soluble sample; there are 26 common chromatographic peaks. S1-S16 are 16 batches of fat-soluble samples; taking S10 as the reference chromatogram, the similarities of the 16 batches of samples with the control fingerprint chromatogram (R) are 0.990, 0.997, 0.980, 0.999, 0.998, 0.998, 0.999, 0.994, 0.999, 0.999, 0.996, 0.975, 0.999, 0.998, 0.986, 0.998 in turn. The RSD value of the similarity is 0.73%, indicating that the differences among these 16 batches of samples are not significant. At the same time, the RSD value of the elution time of the common peaks is 0-0.04%, and there is no difference in the elution time; the RSD value of the chromatographic peak area is between 4.02-9.98%. Therefore, the content differences among different batches of samples in the fat-soluble part are not large.

[0067] Example 2

[0068] Perform mass spectrometry analysis on the water-soluble part samples and fat-soluble samples:

[0069] (1) There are 46 common peaks in the water-soluble samples. Among them, the corresponding compound components are analyzed, and the types of these compounds are classified. It is found that they are mainly acids, glycosides, flavonoids, and alkaloids. At the same time, combined with the control chromatogram, the medicinal materials of the 46 components are classified. Combining the preliminary speculation of the attribution of the previous drug components, finally the medicinal materials of 23 common peaks can be determined. Among them, the common peaks of 11, 12, 17, 18, 34, 36, 38, 40 belong to Glycyrrhiza uralensis Fisch., the common peaks of 19, 23, 24, 28, 29, 32, 36, 39 belong to Scutellaria baicalensis Georgi, the common peaks of 22, 30, 35 belong to Coptis chinensis Franch., the common peak of 15 belongs to Pinellia ternata (Thunb.) Breit., and the common peaks of 5, 21, 31 belong to Ziziphus jujuba Mill.

[0070] (2) Perform mass spectrometry analysis on the fat-soluble part samples. 19 compound components may be analyzed from the 26 characteristic fingerprint common peaks, and the types of these 19 compounds are classified. It is found that they are mainly ketones, acids, and phenols. At the same time, the medicinal materials of these 19 components are classified. Combining the preliminary speculation of the attribution of the previous drug components, finally the medicinal materials of 8 common peaks can be determined. Among them, the common peaks of 22 and 23 belong to Scutellaria baicalensis Georgi, the common peak of 11 belongs to Coptis chinensis Franch., and the common peaks of 12, 16, 18, 19, 20 belong to Zingiber officinale Rosc.

[0071] Methodology investigation of the components in the water-soluble part and the fat-soluble part:

[0072] Investigate the feasibility and scientificity of the established method from three aspects: precision, repeatability, and stability.

[0073] (1) Precision investigation: Precisely pipette the sample solution, according to the optimized chromatographic conditions, inject the sample continuously and repeatedly for 6 times, record the chromatographic elution time and peak area, and calculate the RSD of the chromatographic elution time and peak area;

[0074] (2) Repeatability investigation: Six water-soluble samples of Gancaoxiexin Decoction were prepared in parallel. According to the optimized UPLC chromatographic conditions, injection analysis was carried out in sequence, the peak emergence time and peak area were recorded, and the RSD of the chromatographic peak emergence time and peak area was calculated.

[0075] (3) Stability investigation: A precisely measured solution of the water-soluble sample of the same batch was injected at 0, 2, 4, 8, 12, 18, and 24 h respectively. According to the optimized chromatographic conditions, injection analysis was carried out, the peak emergence time and peak area were recorded, and the RSD of the chromatographic peak emergence time and peak area was calculated.

[0076] The results of the precision, repeatability, and stability tests of the water-soluble fraction samples are shown in Table 1.

[0077] Table 1: Precision, repeatability, and stability of the common peaks of the water-soluble fraction samples

[0078]

[0079]

[0080] For the methodological investigation of the established method, as can be seen from Table 1, the RSD values of the precision retention time are in the range of 0.01 - 0.30%, and the RSD values of the peak area are in the range of 0.13 - 8.91%, both less than 10%, meeting the requirements; the RSD values of the repeatability retention time are in the range of 0.01 - 0.32%, and the RSD values of the peak area are in the range of 0.73 - 9.38%, both less than 10%, meeting the requirements; the RSD values of the stability retention time are in the range of 0.01 - 1.32%, and the RSD values of the peak area are in the range of 0.47 - 9.44%, both less than 10%, meeting the requirements.

[0081] According to the above method, the feasibility and scientificity of the method for establishing the GC-MS fingerprint of the lipophilic fraction were investigated from three aspects: precision, repeatability, and stability. The results are shown in Table 2.

[0082] Table 2: Precision, repeatability, and stability of the common peaks of the lipophilic samples

[0083]

[0084]

[0085] The methodological investigation was carried out on the established method. As can be seen from Table 2, the RSD values of the retention time were in the range of 0.01 - 0.30%, and the RSD values of the peak area were in the range of 0.13 - 8.91%, both less than 10%, meeting the requirements; for repeatability, the RSD values of the retention time were in the range of 0.01 - 0.32%, and the RSD values of the peak area were in the range of 0.73 - 9.38%, both less than 10%, meeting the requirements; for stability, the RSD values of the retention time were in the range of 0.01 - 1.32%, and the RSD values of the peak area were in the range of 0.47 - 9.44%, both less than 10%, meeting the requirements.

[0086] Example 3: Anti - inflammatory and antibacterial tests

[0087] The anti - inflammatory test was carried out by constructing an inflammatory model of mouse RAW264.7 macrophages, and the antibacterial test was carried out by inhibiting Escherichia coli, Salmonella choleraesuis and Proteus mirabilis in vitro to evaluate the anti - inflammatory activity and antibacterial activity effect.

[0088] 1. Anti - inflammatory test and results

[0089] Detection of anti - inflammatory activity of the whole - formula sample, water - soluble part sample and lipid - soluble sample

[0090] Sixteen batches of whole - formula samples, water - soluble part samples and lipid - soluble part samples were taken respectively. With a high concentration of 50 mg·L -1 as the drug concentration, a 6 - well plate was taken, and the cell concentration of RAW264.7 cells in each plate was adjusted to 1×10 5 mL -1 , and 10 μL of LPS with a concentration of 15 μg·mL -1 was added to each well to stimulate for 12 h, then 10 μL of the drug was added and cultured in an incubator at 37°C and 5% CO2 for 12 h. The cell supernatant was collected, and the NO content was detected with reference to the NO detection kit.

[0091] Six replicates were set for each drug - concentration experiment, and a model group and a blank group were also set. The test results are shown in Table 3.

[0092] Table 3: NO content of 16 batches of whole - formula samples, water - soluble part samples and lipid - soluble part samples

[0093]

[0094] Note: Compared with the blank group, *P < 0.05, **P < 0.01; compared with the model group, ▲ P < 0.05, ▲▲ P < 0.01.

[0095] As can be seen from Table 3, compared with the blank group, the NO content in the model group increased significantly (P < 0.01), indicating the successful establishment of the inflammatory model. Compared with the model group, the 16 batches of total formula samples, water-soluble fraction samples, and fat-soluble fraction samples all had a significant inhibitory effect on NO (P < 0.01), and there was no significant difference among batches, indicating that the inhibitory effects of the 16 batches of samples were obvious and the effects were not very different. At the same time, the anti-inflammatory effect was water-soluble fraction samples > total formula samples > fat-soluble fraction samples.

[0096] 2. Bacteriostatic test and results

[0097] (1) Escherichia coli inhibition test

[0098] Take 16 batches of total formula samples, water-soluble fraction samples, and fat-soluble fraction samples, and use LB medium to shake culture Escherichia coli at 37°C for 8 h. After culturing, the bacteria are diluted 50 times with the medium, and 200 μL is inoculated on the LB agar plate. After the plate is dried and there is no liquid flow, the Oxford cup is slightly heated over the flame, placed on the plate, and gently pressed to fix the Oxford cup. Add 16 batches of samples (mass concentration 50 mg·L -1 ) to the Oxford cups, add 180 μL to each sample, and incubate at 37°C for 8 h to observe the bacteriostatic effect. The diameter of the inhibition zone D ≥ 20 mm is extremely sensitive, 15 - 19 mm is highly sensitive, 10 - 14 mm is moderately sensitive, and less than 10 mm is lowly sensitive.

[0099] (2) Salmonella choleraesuis inhibition test

[0100] Take 16 batches of total formula samples, water-soluble fraction samples, and fat-soluble fraction samples, and use LB medium to shake culture Salmonella choleraesuis at 37°C for 8 h. After culturing, the bacteria are diluted 50 times with the medium, and 200 μL is inoculated on the LB agar plate. After the plate is dried and there is no liquid flow, the Oxford cup is slightly heated over the flame, placed on the plate, and gently pressed to fix the Oxford cup. Add 16 batches of samples (mass concentration 50 mg·L -1 ) to the Oxford cups, add 180 μL to each sample, and incubate at 37°C for 6 h to observe the bacteriostatic effect. The diameter of the inhibition zone D ≥ 20 mm is extremely sensitive, 15 - 19 mm is highly sensitive, 10 - 14 mm is moderately sensitive, and less than 10 mm is lowly sensitive.

[0101] (3) Proteus mirabilis inhibition test

[0102] Use LB medium to shake culture Proteus mirabilis at 37°C for 8 h. After culturing, the bacteria are diluted 50 times with the medium, and 200 μL is inoculated on the LB agar plate. After the plate is dried and there is no liquid flow, the Oxford cup is slightly heated over the flame, placed on the plate, and gently pressed to fix the Oxford cup. Add 16 batches of samples (mass concentration 50 mg·L -1) Add 180 μL to each sample, incubate at 37 °C for 6 h, and observe the antibacterial effect. An antibacterial zone diameter D ≥ 20 mm indicates extremely sensitive, 15 - 19 mm indicates highly sensitive, 10 - 14 mm indicates moderately sensitive, and less than 10 mm indicates low sensitivity.

[0103] The antibacterial effects of each sample against Escherichia coli, Salmonella choleraesuis, and Proteus mirabilis are shown in Table 4.

[0104] Table 4: Antibacterial effects of 16 batches of whole - formula samples, water - soluble fraction samples, and lipid - soluble fraction samples

[0105]

[0106] Note: An antibacterial zone diameter D ≥ 20 mm indicates extremely sensitive, 15 - 19 mm indicates highly sensitive, 10 - 14 mm indicates moderately sensitive, and less than 10 mm indicates low sensitivity.

[0107] As can be seen from Table 4, the 16 batches of water - soluble samples have inhibitory effects on Escherichia coli, Salmonella choleraesuis, and Proteus mirabilis. The antibacterial zone diameters of the water - soluble samples for Escherichia coli and Proteus mirabilis are greater than 20 nm, reaching extremely sensitive. The antibacterial zone diameters of Salmonella are mostly in the range of 15 - 19 mm, showing highly sensitive. For the lipid - soluble samples, the antibacterial zone diameters of Escherichia coli and Salmonella choleraesuis are mostly less than 10 nm, and the antibacterial zone for Proteus mirabilis is in the range of 15 - 19 mm. The whole - formula samples are between the water - soluble samples and the lipid - soluble samples, that is, the antibacterial effect is water - soluble fraction samples > whole - formula samples > lipid - soluble fraction.

[0108] Example 4

[0109] Establish a mathematical model through data analysis

[0110] 1. Principle of data analysis

[0111] Principal component analysis (PCA) is a method that transforms the original multiple indicators into a few representative comprehensive indicators, simplifying the complex relationships among variables. Its analysis conditions are that there are multiple variables and there is a certain correlation among the variables. At the same time, the number of principal components is determined according to the eigenvalue > 0.7 and the cumulative variance contribution rate reaching more than 85%. Calculate the scores of each principal component, and finally calculate the comprehensive standard value F. The specific formulas are shown in Formulas 1, 2, and 3.

[0112] F i = w i1 *X1 + w i2 *X2 +... + w in *X n (Formula 1)

[0113]

[0114] F = a1F1 + a2F2 +... + a n F n (Equation 3)

[0115] In Equation 1: w in represents the weight of each variable of the i-th principal component, and X1....X n represents the corresponding X value after data dimensionlessization; in Equation 2: θ n represents the coefficient corresponding to each variable in the component matrix, represents the square root of the eigenvalue corresponding to the i-th principal component; in Equation 3: a n represents the variance percentage of the n-th principal component, F represents the comprehensive standard value, and F n represents the standard value of the n-th principal component.

[0116] Orthogonal Projection to Latent Structures (OPLS) is a multivariate statistical analysis method for multiple dependent variables against multiple independent variables. The regression coefficients in OPLS can not only obtain the correlation degree, positive and negative correlation, and regression equation between variables, but also get the comprehensive effect of components on drug efficacy, and realize the verification of different methods. Taking the common peak area as the independent variable X and the drug efficacy index as the dependent variable Y, the spectrum-effect relationship is studied to obtain the regression coefficients of each X corresponding to Y and establish a regression equation. The Variable Importance in Projection (VIP) is an important index reflecting the explanatory ability of independent variables for dependent variables. The larger the VIP value, the stronger the explanatory ability of the independent variable for the dependent variable. Generally, when VIP > 0.7, the independent variable has significant importance in explaining the dependent variable, that is, the pharmacodynamic components have significant importance.

[0117] Grey Relational Analysis (GRA) is a statistical analysis technique mainly used to analyze the closeness of the relationship between the mother factor and the sub-factors in the system, so as to judge the main factors and secondary factors causing the development and change of the system, and it is a quantitative comparative analysis method for the dynamic development trend of the system. The steps of the grey relational analysis method are as follows:

[0118] ① Select the reference sequence (mother sequence) Yj, where j = 1, 2, 3…, m; the comparison sequence (sub-sequence) X i , where i = 1, 2, 3…, n. In the present invention, taking the anti-inflammatory and antibacterial drug efficacy indexes as the mother sequence (Y), then Y j where j = 1, 2; taking the common peak areas of each characteristic fingerprint in the chromatogram as the sub-sequence (X), then for the whole formula sample X i where i = 1, 2,…, 61, for the water-soluble part sample of X i where i = 1, 2,…, 46, and for the fat-soluble part sample of X i where i = 1, 2,…, 26.

[0119] ②Dimensionalize the variables. Common methods include the initial value method, the mean value method, etc. In the present invention, the mean value method is used for data standardization.

[0120] ③Calculate the difference sequence, the maximum difference, and the minimum difference, and the reference sequence Y j and the comparison sequence X i The absolute value of the difference at the k-th index is the difference sequence value △ 0i (k) = |Y j (k) - X i (k)|, k = 1, 2, 3...n; j = 1, 2; i = 1, 2, 3…, n; The maximum difference is: △ max = max|Y j (k) - X i (k)|; The minimum difference △ min = min|Y j (k) - X i (k)|.

[0121] ④Calculate the correlation coefficient ξ, see formula 4.

[0122] ⑤Calculate the correlation degree r i , see formula 5. r i reflects the overall correlation degree between the reference sequence Y j and the comparison sequence X i . It is set that when r i > 0.6, it is the main contributor to the influence spectrum-effect relationship.

[0123] ξ i (k) = [Δ(min) + ρ × Δ(max)] / [Δ 0i (k) + ρ × Δ(max)] (Equation 4)

[0124]

[0125] In Equation 4: k is the peak number = 1, 2, 3…, n; ρ is the resolution coefficient, and the value range is 0 - 1. Usually, 0.5 is selected; △0i(k) is the absolute difference after mean value normalization of the pharmacodynamic index (reference sequence) and the characteristic peak area (comparison sequence); Δ(min) is the minimum difference, and Δ(max) is the maximum difference; In Equation 5: n is the number of data contained in the comparison sequence (i = 1, 2, 3…n; k = 1, 2, 3).

[0126] In the above analysis, the SPSS.25 statistical analysis software is used to perform PCA analysis by means of dimensionality reduction; the SIMCA-P14.1 software is used to perform OPLS analysis; the GRA correlation degree value is calculated through EXCEL.

[0127] 2. Establishment of the mathematical model

[0128] (1) Establishment of mathematical model for all sample data:

[0129] (a) PCA analysis

[0130] SPSS.25 software was used to standardize multiple groups of antibacterial data. The antibacterial effects of the samples on Escherichia coli, Salmonella choleraesuis and Bacillus mirabilis were analyzed by SPSS.25 software. Two principal components were extracted, and the weight coefficient w was calculated. At the same time, the comprehensive standard values of each principal component were deduced. F1=0.70143*X1+0.69459*X2+0.15976*X3, F2=-0.06928*X1-0.15663*X2+0.98495*X3, so the comprehensive index F value=0.45664F1+0.33057F2.

[0131] (b) OPLS analysis: SIMCA-P14.1 software was used to perform OPLS analysis with the common peak area of the fingerprint of the whole sample as the independent variable X and the anti-inflammatory efficacy index as the dependent variable Y. The regression equation obtained by automatic fitting was: Y = a1X1 + a2X2 + ...a 61 X 61 , where Y corresponds to the anti-inflammatory efficacy result, X1~X 61 The peak area corresponding to the common peak, a1~a 61 The regression coefficients corresponding to the common peaks and anti-inflammatory efficacy are as follows: Figure 6 As shown in , a positive regression coefficient indicates that it is positively correlated with the drug efficacy, while a negative value indicates that it is negatively correlated with the drug efficacy. Figure 7 As shown in Figure 1, the VIP value is an important indicator of the ability of the independent variable to explain the dependent variable. The larger the value, the stronger the ability of the independent variable to explain the dependent variable. It is generally believed that when VIP>0.7, the independent variable is significantly important in explaining the dependent variable.

[0132] The anti-inflammatory mathematical model of the whole formula samples was established by regression coefficient:

[0133] Y1=-0.06087X1-0.00864X2-0.01614X3+0.03062X4+0.03137X5+0.01814X6-0.00

[0134] 099X7-0.03144X8-0.01759X9+0.00339X 10 -0.00546X 11 -0.00740X 12 +0.02976X 13 -0.01172X 14 +0.01153X 15 -0.00305X 16-0.01420X 17 -0.00822X 18 -0.01213X 19 +0.00038X 20 -0.00967X 21 +0.01093X 22 -0.00958X 23 +0.06988X 24 +0.03042X 25 -0.01828X 26 -0.02303X 27 -0.02036X 28 -0.01084X 29 -0.02300X 30 -0.00794X 31 -0.00842X 32 -0.00197X 33 -0.01975X 34 -0.00

[0135] 992X 35 -0.02106X 36 +0.03304X 37 -0.00355X 38 +0.05640X 39 -0.03686X 40 -0.01033X 41 -0.01818X 42 -0.00783X 43 +0.00619X 44 -0.01348X 45 +0.00979X 46 +0.03325X 47 +0.00671X 48 -0.04730X 49 -0.02467X 50 +0.00082X 51 +0.02597X 52 +0.02971X 53 +0.07121X 54 -0.02052X 55 +0.00269X 56 -0.04696X 57 +0.00908X 58 +0.02163X 59 +0.01619X 60-0.02325X 61 。

[0136] Antibacterial modeling: The modeling method is the same as above. Through OPLS analysis, the regression coefficients and VIP values of the common peaks in the fingerprint spectrum and the antibacterial efficacy are as Figure 8 and Figure 9 shown. An antibacterial mathematical model of the whole formula part samples is established from the regression coefficients:

[0137] Y2 = 0.16416X1 - 0.23793X2 + 0.07173X3 - 0.09077X4 + 0.12290X5 - 0.02018X6 - 0.029

[0138] 22X7 + 0.11387X8 + 0.04242X9 - 0.09939X 10 + 0.02155X 11 + 0.00488X 12 - 0.13353X 13 + 0.06701X 14 - 0.05998X 15 + 0.00891X 16 + 0.06324X 17 + 0.01286X 18 + 0.02569X 19 + 0.01900X 20 + 0.01930X 21 - 0.04879X 22 + 0.01915X 23 - 0.58633X 24 - 0.07421X 25 + 0.01340X 26 + 0.07190X 27 + 0.05084X 28 + 0.03172X 29 + 0.02086X 30 - 0.02799X 31 - 0.00807X 32 + 0.06805X 33 + 0.01759X 34 + 0.01637X 35 + 0.06923X 36 + 0.01391X 37 - 0.00563X 38 - 0.25253X 39 + 0.17094X 40 + 0.01824X 41 + 0.05043X42 +0.00038X 43 +0.00626X 44 +0.03375X 45 -0.04151X 46 -0.12191X 47 -0.08453X 48 +0.18263X 49 +0.02569X 50 +0.00396X 51 -0.06341X 52 -0.03223X 53 +0.25851X 54 -0.16344X 55 -0.24608X 56 +0.01414X 57 +0.08618X 58 -0.02558X 59 +0.11159X 60 +0.10086X 61 。

[0139] (c) GRA analysis: Calculated by Formulas 4 and 5, the correlation degree values between the total peak area of fingerprints in the whole formula sample and anti-inflammatory and antibacterial effects are obtained. When the correlation degree > 0.6, it can be used as an effective chromatographic peak for anti-inflammatory and antibacterial effects, and the correlation between the chromatographic peak and anti-inflammatory drug efficacy is obtained: r 61 >r1>r4>r 42 >r 46 >r 54 >r 52 >r 24 >r 14 >r 45 >r 41 >r3>r 27 >r 48 >r 44 >r 28 >r8>r2>r 34 >r 17 >r 21 >r 35 >r 33 >r 36 >r9>r 19 >r 39 >r 29 >r 30 >r 26 >r 18 >r 12 >r 38 >r 47 >r 50 >r11 > r 31 > r 53 > r 37 > r 49 > r 10 > r 43 > r 59 > r 22 > r 16 > r5 > r 40 > r 23 > r 32 > r 51 > r7 > r 15 > r 25 > r 20 > r6 > r 55 > r 13 > r 60 > r 58 > r 57 > r 56 ; Correlation between chromatographic peaks and antibacterial efficacy: r 61 > r1 > r 46 > r4 > r 52 > r 42 > r 27 > r 34 > r 14 > r 28 > r3 > r 44 > r 45 > r 41 > r8 > r 17 > r 47 > r 21 > r 29 > r 50 > r 38 > r 33 > r 18 > r 49 > r 35 > r9 > r 53 > r 19 > r 30 > r 43 > r5 > r 31 > r 39 > r 54 > r 59 > r 10 > r 36 > r 26 > r 11 > r 12 > r2 > r 48 > r 22 > r 23 > r 40 > r 37 > r 51 > r 16> r 32 > r 15 > r7 > r 24 > r 20 > r 13 > r6 > r 25 > r 60 > r 55 > r 58 > r 57 > r 56 (r1 - r61 correspond to S1 - S61 in the mass spectrometry diagram respectively).

[0140] (d) Based on the data of OPLS and PCA analysis of the whole - formula sample and GRA analysis, combined with the study of the spectrum - efficacy relationship, taking the grey relational degree > 0.6, the regression coefficient of OPLS analysis being a positive number (positive correlation) and the VIP value > 0.7 as the criteria, it is obtained that the chromatographic peaks related to anti - inflammation in the whole - formula part sample of Gancaoxiexin Decoction are 5, 6, 10, 13, 15, 20, 22, 24, 25, 37, 39, 44, 46, 47, 48, 51, 53, 56, 58, 59, 60, a total of 22; the chromatographic peaks related to antibacterial are 3, 5, 8, 9, 11, 12, 14, 16, 17, 18, 19, 20, 21, 23, 26, 27, 28, 29, 30, 33, 35, 36, 37, 40, 41, 42, 43, 44, 45, 49, 50, 51, 54, 57, 58, 60, a total of 36; the chromatographic peaks related to both anti - inflammation and antibacterial are 5, 20, 37, 44, 51, 54, 58, 60, a total of 8.

[0141] Combined with the previous UPLC-MS chromatographic analysis and reference chromatograms, 14 components were identified from the whole formula's anti-inflammatory active ingredient group: phenylalanine (5), sucrose (15), isoliquiritigenin or glabridin or (S)-pinocembrin (22), liquiritin (24), magnoflorine (37), glycyrrhetinic acid (39), ziziphus jujuba var. spinosa benzyl glycoside (44), wogonoside (46), glycyrrhenin A (47), berberine hydrochloride (48), ammonium glycyrrhizinate (51), baicalein (54), 6-gingerol (56), and dihydroxy-dimethoxyflavone (58). 22 components were identified from the antibacterial active ingredient group: phenylalanine (5), glycyrrholide (8), glycyrrhenin D (9),schaftoside (11), 5,7-dihydroxy-6-methoxyflavone-7-O-glucuronide (12), formononetin-7-O-β-D-glucoside (16), liquiritigenin-4'-O-β-D-glucose-(1→6)-O-β-D-glucoside (18), glabridin (24), cyclic adenosine monophosphate (30), glycyrrhenin B or N (33), norwogonin-7-O-β-D-glucuronide (35), baicalin (36), magnoflorine (37), 5,7,8-trihydroxyflavone-7-O-glucoside (41), coptisine (42), gingerone (43), ziziphus jujuba var. spinosa benzyl glycoside (44), liquiritigenin (49), ammonium glycyrrhizinate (51), baicalein (54), wogonin (57), dihydroxy-dimethoxyflavone (58). 6 components were identified as being related to both anti-inflammatory and antibacterial activities, namely phenylalanine, magnoflorine, ziziphus jujuba var. spinosa benzyl glycoside, ammonium glycyrrhizinate, baicalein, and dihydroxy-dimethoxyflavone, and these components belong to glycosides, flavonoids, and alkaloids.

[0142] (2) Establishment of the mathematical model for the water-soluble part sample data:

[0143] Using the method described above, the antibacterial data of 3 bacteria were standardized by SPSS 25 software, and 3 principal components were extracted, resulting in the final comprehensive index F value = 0.4450F1 + 0.3028F2 + 0.2523F3.

[0144] The anti-inflammatory data were analyzed by OPLS. The regression coefficients and VIP values of the common peaks in the fingerprint chromatogram of the water-soluble part sample and the anti-inflammatory drug effect are as Figure 10 and Figure 11 shown. The anti-inflammatory mathematical model of the water-soluble part sample was established based on the regression coefficients:

[0145] Y3 = -0.04948X1 + 0.11156X2 + 0.03650X3 - 0.02220X4 - 0.20263X5 - 0.04098X6 - 0.010

[0146] 47X7 + 0.02171X8 + 0.00568X9 + 0.02808X10 -0.00712X 11 +0.00462X 12 +0.01531X 13 -0.00430X 14 -0.00014X 15 -0.02298X 16 +0.00815X 17 +0.04172X 18 -0.02876X 19 -0.03912X 20 -0.03708X 21 -0.00866X 22 +0.01419X 23 +0.10221X 24 +0.01647X 25 -0.05146X 26 -0.09424X 27 +0.08429X 28 +0.01675X 29 +0.00824X 30 -0.03241X 31 +0.02193X 32 -0.03117X 33 -0.02253X 34 +0.05572X 35 -0.03793X 36 -0.19747X 37 -0.13831X 38 -0.12111X 39 +0.09264X 40 -0.06455X 41 +0.15164X 42 +0.25815X 43 +0.09535X 44 -0.05168X 45 +0.04917X 46 。

[0147] The comprehensive data indexes of bacteriostasis obtained above were also analyzed by OPLS. The regression coefficients and VIP values of the common peaks in the fingerprint of the water-soluble fraction samples and the bacteriostatic efficacy are as Figure 12 and Figure 13 shown. A bacteriostatic mathematical model for the water-soluble fraction samples was established from the regression coefficients:

[0148] Y4 = 0.03208X1 - 0.57574X2 - 0.14924X3 - 0.30080X4 - 0.07895X5 - 0.01477X6 - 0.0659

[0149] 7X7 - 0.07526X8 + 0.04278X9 + 0.00187X 10 + 0.01920X 11 - 0.02174X 12 - 0.04240X 13 - 0.01153X 14 + 0.01936X 15 + 0.02612X 16 + 0.11386X 17 - 0.14276X 18 + 0.21253X 19 + 0.01227X 20 + 0.19099X 21 + 0.12653X 22 + 0.15968X 23 + 0.01098X 24 + 0.18724X 25 + 0.27711X 26 - 0.21121X 27 - 0.05584X 28 + 0.04803X 29 + 0.01434X 30 - 0.02906X 31 + 0.01967X 32 + 0.23439X 33 + 0.00493X 34 - 0.26261X 35 + 0.53564X 36 - 0.34400X 37 + 0.55724X 38 - 0.08397X 39 - 0.07313X 40 + 0.15903X 41 - 0.23409X 42 + 0.17930X 43 + 0.20873X 44 - 0.09328X 45 + 0.24996X 46 。

[0150] The correlation between the common peaks of the water-soluble fraction samples and the anti-inflammatory efficacy is r 45 >r27 > r 42 > r4 > r 46 > r1 > r 30 > r 24 > r 17 > r6 > r 33 > r 12 > r3 > r 15 > r 14 > r 26 > r9 > r 16 > r8 > r 32 > r 29 > r7 > r 25 > r 11 > r 31 > r 28 > r 13 > r5 > r 20 > r 21 > r 40 > r 10 > r 18 > r 22 > r 43 > r 37 > r 23 > r 19 > r 39 > r 36 > r 41 > r 38 > r 44 > r2 > r 34 > r 35 , and the correlation with the antibacterial efficacy is r 45 > r 27 > r 46 > r4 > r 42 > r1 > r 33 > r 26 > r 30 > r3 > r 12 > r 17 > r 14 > r 15 > r 25 > r6 > r9 > r 16 > r7 > r 24 > r8 > r 13 > r 29 > r 28 > r 31 > r 32 > r5 > r 11 > r 20 > r 21 > r 18 > r 40 > r 22 > r 10 > r 23> r 19 > r 39 > r 36 > r 41 > r 43 > r 37 > r 35 > r 34 > r2 > r 38 > r 44 。

[0151] Based on the data of GRA analysis, OPLS and PCA analysis, combined with the literature reports on the study of spectrum-effect relationship, taking the grey correlation degree > 0.6, the regression coefficient of OPLS analysis being positive (positive correlation) and the VIP value > 0.7 as the criteria, the key component groups with anti-inflammatory and antibacterial effects in the water-soluble part samples of Gancaoxiexin Decoction were screened. The chromatographic peaks related to anti-inflammatory in the water-soluble part samples of Gancaoxiexin Decoction were 2, 3, 8, 9, 10, 12, 13, 17, 18, 23, 24, 25, 28, 29, 30, 32, 35, 40, 42, 44, a total of 20; the chromatographic peaks related to antibacterial in the water-soluble samples of Gancaoxiexin Decoction were 1, 9, 10, 11, 15, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 29, 30, 32, 33, 34, 36, 38, 41, 43, 44, a total of 25, and the chromatographic peaks related to both anti-inflammatory and antibacterial were 9, 10, 17, 23, 24, 25, 29, 30, 32, 44, a total of 10.

[0152] Combined with the previous UPLC-MS spectrum analysis and reference spectrum, 13 of the 20 chromatographic peaks related to anti-inflammatory in the water-soluble fraction samples were analyzed, including 5,7-dihydroxy-6-methoxyflavone-7-O-glucuronide (8), sucrose (10), liquiritigenin-4'-O-β-D-glucose-(1→6)-O-β-D-glucoside (12), liquiritin (17), nor-wogonin-7-O-β-D-glucuronide (23), baicalin (24), magnolamine (25), 5,7,8-trihydroxyflavone-7-O-glucoside (28 ), baicalin (29), coptisine (30), baicalin (32), berberine (35), ammonium glycyrrhizinate (40), 14 of the 25 chromatographic peaks related to antibacterial activity were resolved, namely sucrose (10), formononetin (11), liquiritigenin (17), cyclic adenosine monophosphate (21), palmatine hydrochloride (22), baicalin (24), magnolamine (25), 5,7,8-trihydroxyflavone-7-O-glucoside (28), baicalin (29), coptisine (30), baicalin (32), glycyrrhizin (36), and ural saponin P (38). Among them, 8 effective ingredients with both anti-inflammatory and antibacterial effects were analyzed, namely sucrose, glycyrrhizin, 5,7,8-trihydroxyflavone-7-O-glucoside, baicalin, magnolia alkaloids, baicalin, coptisine, and wogonin. These ingredients mainly belong to the types of sugars, glycosides, flavonoids, and alkaloids.

[0153] (3) Establishment of mathematical model for fat-soluble fraction sample data:

[0154] As described above, the antibacterial test data of the fat-soluble fraction samples against three bacteria were standardized using SPSS.25 software, and the final comprehensive index F value was obtained = 0.4281F1+0.3419F2.

[0155] As described above, OPLS analysis was performed on the common peaks of anti-inflammatory efficacy and fingerprints. The regression coefficients and VIP values of the common peaks of fingerprints of samples from fat-soluble parts and anti-inflammatory efficacy were as follows: Figure 14 and Figure 15 As shown, the anti-inflammatory mathematical model of fat-soluble part samples was established based on the regression coefficients:

[0156] Y5=-0.22521X1+0.00531X2-0.09749X3-0.20393X4-0.08223X5-0.18640X6+0.050

[0157] 00X7-0.06985X8+0.14447X9+0.10708X 10 -0.00508X 11 -0.03708X 12+0.08129X 13 +0.10721X 14 -0.16142X 15 +0.17915X 16 -0.15320X 17 +0.03237X 18 +0.03898X 19 +0.25279X 20 +0.12221X 21 -0.07160X 22 +0.11733X 23 -0.14821X 24 +0.04994X 25 -0.24421X 26 。

[0158] The comprehensive data indexes of antibacterial activity and the fingerprint chromatogram of the liposoluble fraction samples obtained above were analyzed by OPLS. The regression coefficients and VIP values of the common peaks in the fingerprint chromatogram of the liposoluble fraction samples and the antibacterial efficacy are as Figure 16 and Figure 17 shown. An antibacterial mathematical model of the liposoluble fraction samples was established based on the regression coefficients:

[0159] Y6 = 0.10055X1 - 0.32629X2 + 0.03078X3 - 0.21203X4 - 0.38789X5 - 0.31221X6 + 0.209

[0160] 18X7 + 0.27628X8 + 0.34261X9 + 0.96738X 10 +0.29493X 11 -0.12979X 12 +0.05665X 13 +0.57509X 14 +0.23398X 15 +0.22431X 16 +0.07620X 17 +0.39381X 18 -0.27984X 19 +0.67366X 20 -0.29447X 21 -0.09181X 22 +0.10824X 23 -0.83565X 24 +0.22208X 25 +0.22437X 26 。

[0161] The correlation between the common peaks of the lipophilic fraction samples and the anti-inflammatory efficacy was calculated as r 19 > r 22 > r 17 > r 14 > r1 > r 23 > r 13 > r5 > r 20 > r4 > r 10 > r2 > r7 > r 21 > r9 > r 11 > r 16 > r6 > r 18 > r 25 > r3 > r 12 > r 24 > r8 > r 26 > r 15 and the correlation with the antibacterial efficacy was r 22 > r 17 > r 19 > r 13 > r 14 > r1 > r 23 > r5 > r 10 > r 18 > r 20 > r9 > r2 > r 24 > r 12 > r7 > r6 > r3 > r8 > r4 > r 21 > r 11 > r 15 > r 16 > r 25 > r 26 .

[0162] Based on the data of GRA analysis, OPLS and PCA analysis of the lipophilic fraction samples, combined with the literature reports on the study of the spectrum-effect relationship, taking the grey correlation degree > 0.6, the regression coefficient of OPLS analysis being positive (positive correlation) and the VIP value > 0.7 as the criteria, the chromatographic peaks related to anti-inflammatory in the lipophilic fraction samples of Gancaoxiexin Decoction were screened as 2, 7, 9, 10, 13, 14, 16, 18, 19, 20, 21, 23, 25, a total of 13; the chromatographic peaks related to antibacterial were 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 18, 20, 23, 25, a total of 14; the chromatographic peaks related to both anti-inflammatory and antibacterial were 7, 9, 10, 13, 14, 16, 18, 20, 23, 25, a total of 10.

[0163] Combined with GC-MS mass spectrometry analysis, 10 out of 13 chromatographic peaks related to anti-inflammatory in the liposoluble fraction sample were identified as hexanal (2), vanillin lactoside (9), melezitose (10), p-hydroxycinnamic acid (13), ferulic acid (14), 1-(4-hydroxy-3-methoxyphenyl)-4-octen-3-one (16), (E)-1-(4-Hydroxy-3-methoxyphenyl)dec-3-en-5-one (18), 6-shogaol (19), 6-gingerol (20) and wogonin (23). 11 out of 14 chromatographic peaks related to antibacterial were identified as chromatographic vanillin lactoside (9), melezitose (10), vanillic acid (11), p-hydroxycinnamic acid (13), ferulic acid (14), palmitic acid (15), 1-(4-hydroxy-3-methoxyphenyl)-4-octen-3-one (16), 5-hydroxy-1-(4-hydroxy-3-methoxyphenyl)octan-3-one (17), (E)-1-(4-Hydroxy-3-methoxyphenyl)dec-3-en-5-one (18), 6-gingerol (20) and wogonin (23). Among them, 8 active ingredient groups related to both anti-inflammatory and antibacterial were identified: vanillin lactoside, melezitose, p-hydroxycinnamic acid, ferulic acid, 1-(4-hydroxy-3-methoxyphenyl)-4-octen-3-one, (E)-1-(4-Hydroxy-3-methoxyphenyl)dec-3-en-5-one, 6-gingerol, and wogonin. And these ingredients belong to flavonoids, phenolic acids, glycosides and polysaccharides.

[0164] Comparison between the calculated values of the mathematical model and the measured values:

[0165] Prepare 3 batches of samples of the whole recipe of Gancaoxiexin Decoction, liposoluble fraction samples and water-soluble fraction samples as the research objects by the same method. Inject them into GC-MS and UPLC under the same chromatographic conditions to obtain the peak areas (X) of the corresponding common chromatographic fingerprint peaks. Through the mathematical model established above, calculate the corresponding pharmacodynamic values (Y) as the calculated values. At the same time, conduct cell experiments and antibacterial experiments under the same conditions as before to obtain the specific pharmacodynamic data values of anti-inflammatory and antibacterial of the whole recipe, water-soluble and liposoluble fraction samples as the measured values. Compare the two groups of values and calculate the RSD value to verify the accuracy of the mathematical model. When the RSD value is less than 10%, it proves that the established model is feasible. The verification results are shown in Table 5.

[0166] Table 5: Verification results of the whole recipe, water-soluble and liposoluble fraction samples

[0167]

[0168] As can be seen from Table 5, the RSD values of the calculated values from the established mathematical formula and the actual measured values are both less than 10%, thus verifying the feasibility of the model. When conducting anti-inflammatory and antibacterial efficacy tests on Gancaoxiexin Decoction, the efficacy can be quickly calculated through the fingerprint of the whole formula and the mathematical model. Further, by combining the fingerprints and mathematical models of the water-soluble part and the lipid-soluble part, the corresponding efficacy can be detected more quickly and accurately. According to the rule that the influence on anti-inflammatory and antibacterial efficacy is water-soluble part > whole formula > lipid-soluble part, the anti-inflammatory and antibacterial effects of the samples to be tested can be evaluated in more detail by the proportion of each part in the sample to be tested and the proportion of specific components.

Claims

1. A quality detection method for traditional Chinese medicine compound based on fingerprint spectrum, characterized in that: The traditional Chinese medicine compound is Gancaoxiexin Decoction. First, establish the fingerprint of Gancaoxiexin Decoction, and then conduct anti-inflammatory and antibacterial activity tests. Perform principal component analysis on the antibacterial test results to obtain the comprehensive antibacterial efficacy index. Analyze the spectral-efficacy relationship between the anti-inflammatory efficacy results, antibacterial efficacy results and their fingerprint. Establish a spectral-efficacy relationship mathematical model for anti-inflammatory and antibacterial through orthogonal projection partial least squares analysis. Calculate the grey correlation degree through correlation analysis to obtain the drug components related to anti-inflammatory and antibacterial. The steps for establishing the fingerprint of Gancaoxiexin Decoction are as follows: (1)Prepare the test solution Weigh 20 g of Glycyrrhizae Radix, 15 g of Scutellariae Radix, 15 g of Zingiberis Rhizoma Recens, 12 g of Jujubae Fructus, 13 g of Pinelliae Rhizoma, and 5 g of Coptidis Rhizoma according to a single dose, that is, the total single dose is 80 g. Weigh 16 batches of the whole formula, and extract the samples by the semi-bionic extraction method. Extraction conditions: the pH of the water for the first decoction is 4.0, the pH of the water for the second decoction is 7.0, the pH of the water for the third decoction is 8.0, and the total decoction time is 2 h, among which the first decoction is 1.0 h, the second decoction is 0.5 h, and the third decoction is 0.5 h. The extraction temperature is 95 °C, the liquid-solid ratio is 1:10, and reflux extraction is carried out; the above pH water is adjusted with 0.1 mol·L -1 of HCl and 0.1 mol·L -1 of NaOH; Combine the extraction solutions of the 3 times, and concentrate them under reduced pressure at 55 °C to 1 g·mL -1 , respectively as the sample mother liquor, and store them at 4 °C for standby. At this time, the obtained sample mother liquor is the whole formula sample; According to the optimized extractant ratio and constant volume agent ratio, with 1:1 as the extraction ratio of the solvent to the extract, and 1:2 as the constant volume agent ratio, take 20 mL of the sample mother liquor and 20 mL of ethyl acetate solvent. Shake the extraction bottle for 5 min, let it stand for 10 min, take the upper layer solution as the fat-soluble part layer, and the lower layer solution as the water-soluble part layer. Extract 3 times, combine the 3 extraction solutions respectively. Evaporate the fat-soluble part to dryness on a rotary evaporator at 45 °C, and evaporate the water-soluble part to dryness on a rotary evaporator at 55 °C. Then, dissolve and fix the volumes to 20 mL with ethyl acetate and methanol solution with a volume fraction of 50% respectively. The mass concentrations of the water-soluble and fat-soluble sample solutions are 0.5 g·mL -1 , and store for future use at 4 °C; (2)Preparation of the reference solution: Weigh appropriate amounts of zingerone, coptisine hydrochloride, liquiritigenin, berberine hydrochloride, wogonoside, baicalein, 6-gingerol, ammonium glycyrrhizinate, baicalin, liquiritin, and wogonin. These 11 reference substances are used as a mixed reference substance for the whole formula sample. Dissolve them in methanol to a concentration of 1.0 mg·mL -1 , and store them in a refrigerator at 4°C as the reference substance solution; Weigh an appropriate amount of liquiritin, baicalin, wogonoside, coptisine hydrochloride, liquiritigenin, ammonium glycyrrhizinate, and berberine hydrochloride in the same method to prepare the reference solution for the water-soluble part; (3)Chromatographic conditions UPLC chromatographic conditions for the whole formula sample and the water-soluble fraction sample: Column: Shim-pack Scepter C18-200 (3μm, 4.6×250mm), detection wavelength 277nm, flow rate 0.3mL·min -1 , injection volume 3μL, column temperature 15°C; Gradient elution: A is an aqueous solution of formic acid with a volume fraction of 0.4%, B is acetonitrile. From 0 to 5 minutes, the volume percentage of mobile phase B increases from 2% to 6%; from 5 to 10 minutes, the volume percentage of mobile phase B increases from 6% to 8%; from 10 to 15 minutes, the volume percentage of mobile phase B increases from 8% to 15%; from 15 to 20 minutes, the volume percentage of mobile phase B increases from 15% to 20%; from 20 to 30 minutes, the volume percentage of mobile phase B increases from 20% to 23%; from 30 to 40 minutes, the volume percentage of mobile phase B increases from 23% to 32%; from 40 to 60 minutes, the volume percentage of mobile phase B increases from 32% to 35%; from 60 to 80 minutes, the volume percentage of mobile phase B increases from 35% to 50%; from 80 to 85 minutes, the volume percentage of mobile phase B increases from 50% to 80%; from 85 to 90 minutes, the volume percentage of mobile phase B increases from 80% to 98%; from 90 to 95 minutes, the volume percentage of mobile phase B decreases from 98% to 60%; from 95 to 100 minutes, the volume percentage of mobile phase B decreases from 60% to 35%; from 100 to 105 minutes, the volume percentage of mobile phase B decreases from 35% to 2%; GC-MS chromatographic conditions for the fat-soluble fraction sample: injection port temperature 260 °C, interface temperature 300 °C, splitless mode, carrier gas helium, flow rate 0.8 mL·min -1 , pressure 70 kPa, injection volume 0.5 μL, temperature programming: initial temperature 50 °C, hold for 20 min, at 3 °C·min -1 to 150 °C, hold for 5 min, then at 5 °C·min -1 to 250 °C, hold for 15 min, ionization method EI, full scan mode, m / z = 12 - 600, ion source temperature 230 °C, quadrupole temperature 150 °C.

2. The quality detection method of a traditional Chinese medicine compound based on fingerprint spectrum according to claim 1, characterized in that When establishing the fingerprint of the whole prescription of Gancaoxiexin Decoction, 16 batches of medicinal materials of the prescription of Gancaoxiexin Decoction were taken, pulverized, and 16 batches of whole prescription samples were prepared according to the extraction process. The samples were injected according to the chromatographic conditions, and the UPLC data were recorded. The original data were imported into the software of "Similarity Evaluation System for Chromatographic Fingerprints of Traditional Chinese Medicines", and the common peaks were matched by the multi-point calibration method to establish the fingerprint. Taking S11 as the reference fingerprint, the method for generating the control fingerprint was the median method, the time window width was 0.5, and the fingerprint interval was 400. The fingerprints of 16 batches of samples were obtained. There were 61 common chromatographic peaks in the whole prescription. From the fingerprint results, the similarities of 16 batches of samples with the control fingerprint were 0.999, 0.999, 0.999, 0.997, 0.997, 0.996, 0.996, 0.997, 0.999, 0.997, 0.996, 0.997, 0.996, 0.996, 0.997, 0.997 respectively, and the RSD value was 0.11%. Comparing the RSD of the elution time of the common peaks was 0.01 - 0.29%, indicating that the elution time of the common peaks was consistent. Comparing the RSD of the chromatographic peak area of the common peaks was 1.22 - 29.80%.

3. The quality detection method of a traditional Chinese medicine compound based on fingerprint spectrum according to claim 2, characterized in that: The anti-inflammatory test was carried out on RAW264.7 cells with the whole prescription sample of Gancaoxiexin Decoction to obtain the corresponding anti-inflammatory data. At the same time, the antibacterial tests were carried out on Escherichia coli, Salmonella choleraesuis and Proteus mirabilis to obtain three groups of antibacterial data. The PCA analysis was carried out on the three groups of antibacterial data using SPSS 25 statistical analysis software, and the comprehensive antibacterial data was obtained after standardization. Using SIMCA-P 14.1 software, with the peak area of the common peaks of the whole prescription sample as the independent variable X and the anti-inflammatory efficacy index as the dependent variable Y, OPLS analysis was carried out, and the regression equation was fitted: Y = a1X1 + a2X2 + … a 61 X 61 , where Y corresponds to the anti-inflammatory or antibacterial efficacy results, and X1~X 61 correspond to the peak areas of the common peaks, and a1~a 61 correspond to the regression coefficients of the common peaks and the efficacy. A positive regression coefficient indicates a positive correlation with the efficacy, while a negative number indicates a negative correlation with the efficacy. Then, the GRA correlation degree value was calculated through EXCEL. When the correlation degree > 0.6, it can be used as an effective chromatographic peak for anti-inflammatory or antibacterial.

4. The quality detection method of a traditional Chinese medicine compound based on fingerprint spectrum according to claim 3, wherein Taking the grey relational degree > 0.6, the regression coefficient of OPLS analysis being positive and the VIP value > 0.7 as the criteria, the chromatographic peaks related to anti-inflammatory in the whole prescription of Gancaoxiexin Decoction were screened as 5, 6, 10, 13, 15, 20, 22, 24, 25, 37, 39, 44, 46, 47, 48, 51, 53, 54, 56, 58, 59, 60, a total of 22; the chromatographic peaks related to antibacterial were 3, 5, 8, 9, 11, 12, 14, 16, 17, 18, 19, 20, 21, 23, 26, 27, 28, 29, 30, 33, 35, 36, 37, 40, 41, 42, 43, 44, 45, 49, 50, 51, 54, 57, 58, 60, a total of 36; the chromatographic peaks related to both anti-inflammatory and antibacterial were 5, 20, 37, 44, 51, 54, 58, 60, a total of 8.

5. A quality detection method for traditional Chinese medicine compound based on fingerprint spectrum according to claim 4, characterized in that: The establishment of the UPLC fingerprint of the water-soluble part was to inject the water-soluble sample solution according to the chromatographic conditions of the above-mentioned whole prescription samples, record the UPLC data, import the original data into the software of "Similarity Evaluation System for Chromatographic Fingerprints of Traditional Chinese Medicines", and perform common peak matching by the multi-point calibration method to establish the fingerprint. There were 46 common characteristic fingerprint chromatographic peaks in the samples. S1 - S16 were the water-soluble part samples from the 1st batch to the 16th batch respectively. Taking S9 as the reference fingerprint and the fingerprint interval being 400, the fingerprints of the water-soluble samples were obtained. From the fingerprint results, the RSD value of the similarities of 16 batches of samples with the control fingerprint was 0.04%. The RSD value of the elution time of the common characteristic fingerprint peaks was 0.01 - 0.35%, and the RSD value of the chromatographic peak area was 1.47 - 29.69%.

6. The quality detection method of a traditional Chinese medicine compound based on fingerprint spectrum according to claim 5, characterized in that: The water-soluble fraction samples were used to conduct antibacterial tests against Escherichia coli, Salmonella choleraesuis, and Proteus mirabilis, and at the same time, anti-inflammatory tests were carried out on RAW264.7 cells to obtain the corresponding antibacterial data and anti-inflammatory data. Through PCA analysis, OPLS analysis, and GRA correlation analysis, it was found that the chromatographic peaks related to anti-inflammatory in the water-soluble fraction samples of Gancaoxiexin Decoction were 2, 3, 8, 9, 10, 12, 13, 17, 18, 23, 24, 25, 28, 29, 30, 32, 35, 40, 42, 44, a total of 20; the chromatographic peaks related to antibacterial in the water-soluble samples of Gancaoxiexin Decoction were 1, 9, 10, 11, 15, 16, 17, 19, 20, 21, 22, 23, 24, 25, 26, 29, 30, 32, 33, 34, 36, 38, 41, 43, 44, a total of 25; the chromatographic peaks related to both anti-inflammatory and antibacterial were 9, 10, 17, 23, 24, 25, 29, 30, 32, 44, a total of 10.

7. The quality detection method of a traditional Chinese medicine compound based on fingerprint spectrum according to claim 6, characterized in that, The establishment of the GC-MS fingerprint of the liposoluble fraction samples was carried out by injecting samples according to the GC-MS chromatographic conditions, recording the data, importing the original data into the "Similarity Evaluation System for Traditional Chinese Medicine Chromatographic Fingerprints" software, and performing common peak matching by the multi-point calibration method to establish the fingerprint; there were 26 common chromatographic peaks in the fingerprint, and S1-S16 were 16 batches of liposoluble fraction samples; taking S10 as the reference fingerprint, the RSD value of the similarity between the 16 batches of samples and the control fingerprint was 0.73%, and the RSD value of the elution time of the common peaks was 0-0.04%; the RSD value of the chromatographic peak area was 4.02-9.98%.

8. The quality detection method of a traditional Chinese medicine compound based on fingerprint spectrum according to claim 7, characterized in that: The liposoluble fraction samples were used to conduct anti-inflammatory tests on RAW264.7 cells, and at the same time, antibacterial tests were carried out with Escherichia coli, Salmonella choleraesuis, and Proteus mirabilis to obtain the corresponding anti-inflammatory data and antibacterial data. Through PCA analysis, OPLS analysis, and GRA correlation analysis, the chromatographic peaks related to anti-inflammatory in the liposoluble fraction samples of Gancaoxiexin Decoction were screened as 2, 7, 9, 10, 13, 14, 16, 18, 19, 20, 21, 23, 25, a total of 13; the chromatographic peaks related to antibacterial were 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 18, 20, 23, 25, a total of 14; the chromatographic peaks related to both anti-inflammatory and antibacterial were 7, 9, 10, 13, 14, 16, 18, 20, 23, 25, a total of 10.

Images