Method for analyzing and identifying anti-inflammatory active ingredients of Xinhuang tablet based on UPLC-Tripe-TOF-MS spectrum-effect relationship

Abstract

The invention discloses a method for analyzing and identifying anti-inflammatory active ingredients of Xinhuang tablets based on a UPLC-Tripe-TOF-MS (Ultra Performance Liquid Chromatography-Tripe-TOF-Mass Spectrometry) spectrum-effect relationship. According to the method, an active extract is obtained from Xinhuang tablets through a dichloromethane liquid-liquid extraction method, a fingerprint spectrum of 10 batches of samples is established by using UPLC-Tripe-TOF-MS, a double-target evaluation system is constructed by combining an LPS-induced RAW264.7 cell inflammation model and COX-2 enzyme inhibition detection, and spectrum-effect relationship research is performed through gray correlation analysis and an OPLS-DA data mining technology. Four chemical components, namely indomethacin, 7-ketolithocholic acid, erucyl amide and pyranoapigenin, which are highly related to the anti-inflammatory activity are identified; the quality control method of the Xinhuang tablets based on the key chemical markers is established, a scientific basis is provided for quality evaluation and standardization of the traditional Chinese medicine compound Xinhuang tablets, and meanwhile a new thought is provided for development of anti-inflammatory drugs.

Description

Technical Field

[0001] The present invention relates to the field of research on the correlation between the fingerprint of traditional Chinese medicine chemical components and their activities, and particularly to a method for screening and identifying the anti-inflammatory active components of the traditional Chinese medicine compound Xinhuang Tablets by using UPLC-Triple-TOF-MS technology combined with a dual-target activity evaluation system. The anti-inflammatory active components identified by this method can be further used for the quality evaluation of Xinhuang Tablets and the development of anti-inflammatory drugs. Background Art

[0002] Xinhuang Tablets is a compound traditional Chinese medicine preparation composed of eight traditional Chinese medicine components (such as Sarcandra glabra, Panax notoginseng, artificial bezoar, pig bile powder, Urena lobata, etc.) and indomethacin, and has the effects of clearing heat and detoxifying, reducing swelling and relieving pain. Clinically, it is mainly used to treat symptoms such as sore throat, toothache, jaundice, and various unknown swollen toxins caused by heat-toxin and stasis of blood. Modern pharmacological studies have shown that Xinhuang Tablets has good anti-inflammatory and analgesic effects on orthopedic trauma diseases such as gouty arthritis, soft tissue injury, osteoarthritis, ankylosing spondylitis, and rheumatoid arthritis. However, the material basis of its anti-inflammatory and analgesic effects has not been fully elucidated.

[0003] Inflammation is a complex physiological response mechanism of the immune system to tissue damage or exogenous challenges. Although the inflammatory process contributes to wound healing, excessive or uncontrolled inflammatory responses may lead to various diseases. Pain is one of the main manifestations of the inflammatory process and is usually accompanied by actual tissue damage or potential damage. A large amount of evidence has established a clear association between the inflammatory process and the pain mechanism.

[0004] In contemporary drug research, fingerprinting has been widely adopted as an important analytical method for the quality assurance and standardization of traditional Chinese medicine. Although fingerprinting can effectively characterize the diversity of components in drugs, not all compounds are therapeutically relevant. Establishing the spectrum-activity correlation helps to identify and screen the bioactive components in traditional Chinese medicine preparations. Combining pharmacological activity data with fingerprinting results can achieve a comprehensive spectrum-activity relationship analysis. By statistically correlating the fingerprint of Xinhuang Tablets with its pharmacodynamic properties, the main bioactive components can be identified, and a more scientific and rigorous quality control method can be formulated.

[0005] In the present invention, an active extract is obtained from Xinhuang Tablets by dichloromethane extraction method, and the fingerprint of 10 batches of samples is established by using UPLC-Triple-TOF-MS (Ultra Performance Liquid Chromatography Coupled with Triple Quadrupole Time-of-Flight Mass Spectrometry) technology. A dual-target evaluation system is constructed by combining the LPS-induced RAW264.7 cell inflammation model and COX-2 enzyme inhibition detection, and spectrum-activity correlation analysis is carried out by using grey relational analysis and OPLS-DA data mining technology. Four main anti-inflammatory active components are identified, including indomethacin, 7-ketolithocholic acid, erucamide, and apigeninidin. The anti-inflammatory activities of 7-ketolithocholic acid and apigeninidin are discovered for the first time. Summary of the Invention

[0006] In view of the current problems such as the unclear mechanism of action of Xinhuang Tablets in the treatment of gouty arthritis and the exploration of anti-inflammatory active ingredients in Xinhuang Tablets, the present invention provides a method for identifying the anti-inflammatory active ingredients of Xinhuang Tablets based on UPLC-Triple-TOF-MS spectrum-effect relationship analysis.

[0007] The present invention uses the spectrum-effect relationship analysis method of UPLC-Triple-TOF-MS fingerprint and dual-target activity evaluation. Through the dual-model evaluation system of inhibiting NO release and COX-2 enzyme activity in RAW264.7 cells induced by LPS, combined with grey relational analysis and OPLS-DA data mining technology, four main anti-inflammatory active ingredients (indomethacin, 7-ketolithocholic acid, apigeninidin and erucamide) in Xinhuang Tablets have been successfully identified, providing a scientific basis for the improvement of the quality control standard of Xinhuang Tablets and the development of new anti-inflammatory drugs.

[0008] The technical solution of the present invention is as follows:

[0009] A method for identifying the anti-inflammatory active ingredients of Xinhuang Tablets based on UPLC-Triple-TOF-MS spectrum-effect relationship analysis, comprising:

[0010] S1. Take Xinhuang Tablets, crush and sieve them. The obtained powder is refluxed and extracted with methanol, filtered, and the filtrate and residue are collected separately; the filtrate is concentrated under reduced pressure to obtain a methanol extract; the residue is refluxed and extracted with water, filtered, and the water extract is collected; the water extract and the methanol extract are combined to obtain a mixed extract;

[0011] The specific operation is as follows: Take Xinhuang Tablets, crush and sieve them. Mix the medicinal powder with methanol at a solid-liquid ratio of 1:5-10 (g / mL), heat to 85 °C and reflux for 3 h, cool to room temperature and filter. The residue is extracted 1-2 times repeatedly, and the methanol extracts are combined and concentrated under reduced pressure to obtain a methanol extract; the remaining residue is added to water at a solid-liquid ratio of 1:10-20 (g / mL) and refluxed for 1.5 h, cooled to room temperature and filtered. The water extract and the methanol extract are combined to obtain a mixed extract;

[0012] S2. Perform liquid-liquid extraction of the mixed extract obtained in S1 with dichloromethane. The extract is concentrated under reduced pressure and freeze-dried to obtain a dichloromethane extract, which is stored in the dark and dried.

[0013] The specific operation is as follows: Take the mixed extract obtained in S1, add 3 times the volume of dichloromethane for liquid-liquid extraction, extract 3 times, combine the extracts, concentrate under reduced pressure, and freeze-dry to obtain a dichloromethane extract;

[0014] S3. Analyze the dichloromethane extract of 10 batches of Xinhuang tablets using UPLC-Triple-TOF-MS, establish a fingerprint, and determine 15 common characteristic peaks;

[0015] Before analysis, the sample pretreatment is as follows: Take the dichloromethane extract, dissolve it with methanol, and filter it through a 0.22 μm membrane to obtain a sample solution;

[0016] The UPLC-Triple-TOF-MS analysis conditions are as follows:

[0017] Chromatographic column: ACQUITY UPLC HSS T3 (2.1 mm × 150 mm, 1.8 μm), mobile phase A: acetonitrile containing 0.1% (volume fraction) formic acid, mobile phase B: water containing 0.1% (volume fraction) formic acid, gradient elution conditions: 0 - 15 min, 5 - 40% A; 15 - 35 min, 40 - 95% A; 35 - 37 min, 95% A; 37 - 38 min, 95 - 5% A; flow rate: 0.3 mL·min -1 , injection volume: 3 μL, column temperature: 50 °C, detection wavelength: 240 nm;

[0018] Mass spectrometry analysis uses the Triple-TOF 6600 + system, equipped with an electrospray ionization source (ESI) operating in positive and negative ion modes, mass scanning range: 100 - 2000 Da; nebulizing gas 1 (GS1): 55 psi; nebulizing gas 2 (GS2): 55 psi; curtain gas (CUR): 35 psi; ion source voltage (IS): 5500 V (positive ion mode) - 4500 V (negative ion mode); ion source temperature: 600 °C (positive ion mode) and 550 °C (negative ion mode); primary scan: declustering potential (DP) 100 V; focusing voltage (CE) 10 V; secondary mass spectrometry data uses the IDA mode, and the collision-induced dissociation (CID) energy is ±40 ± 20 eV;

[0019] The sample is introduced through a CDS pump, which is also used for mass axis calibration before analysis to ensure that the mass accuracy is within 2 ppm;

[0020] S4. Establish a dual-target evaluation system and determine the anti-inflammatory activity of the dichloromethane extract of 10 batches of Xinhuang tablets;

[0021] Specifically, the dual-target evaluation system refers to: the LPS-induced RAW264.7 cell inflammation model and the COX-2 activity inhibition model; the NO inhibition rate of the dichloromethane extract of 10 batches of Xinhuang tablets is detected through the LPS-induced RAW264.7 cell inflammation model; the IC of the dichloromethane extract of 10 batches of Xinhuang tablets is detected through the COX-2 activity inhibition model50 value, i.e., COX-2 inhibitory activity;

[0022] S5. Using grey relational analysis and OPLS-DA data mining methods, perform spectrum-effect relationship analysis on the UPLC-Triple-TOF-MS fingerprint obtained in S3 and the anti-inflammatory activity obtained in S4 to screen out anti-inflammatory active ingredients;

[0023] The specific method for spectrum-effect relationship analysis is as follows: Take the NO inhibition rate and COX-2 inhibitory activity of the dichloromethane extracts of 10 batches of Xinhuang tablets as the reference sequences, and combine the peak areas of 15 common characteristic peaks in the fingerprint as the comparison sequences. Import them into grey modeling software for correlation analysis, and calculate the correlation coefficient r of each detected compound with respect to the NO and COX-2 inhibitory abilities. A correlation coefficient r > 0.8 indicates a significant strong positive correlation between variables, showing a robust linear relationship;

[0024] Use the SIMCA-P software platform (version 14) to perform chemometric analysis on the chromatographic data of the dichloromethane extracts of 10 batches of Xinhuang tablets. First, use the unsupervised pattern recognition method: principal component analysis (PCA) for dimensionality reduction and data structure visualization. At the same time, use hierarchical cluster analysis (HCA) to generate a dendrogram to visually show the similarities and differences between different batches; Subsequently, apply grey relational analysis (GRA) and orthogonal partial least squares-discriminant analysis (OPLS-DA) to identify the chemical markers that contribute most significantly to the activity differences; In the OPLS-DA model, compounds with variable importance in projection (VIP) scores equal to or exceeding 1.0 are determined as statistically significant bioactive markers; This comprehensive chemometric method effectively reveals the chemical composition variations between batches and their relationship with biological activities;

[0025] Monomer activity verification method: Through the cross-analysis of GRA and OPLS-DA, screen out the core contributing monomers to the anti-inflammatory activity of the dichloromethane extract; Use the established LPS-induced RAW264.7 cell inflammation model and COX-2 activity inhibition model to evaluate the effects of these monomers on NO release and COX-2 activity;

[0026] The finally identified anti-inflammatory active ingredients are: indomethacin, 7-ketolithocholic acid, erucamide, and pyranapigeninidin.

[0027] Compared with the prior art, the present invention has the following characteristics:

[0028] 1) The present invention for the first time performs UPLC-Triple-TOF-MS fingerprint analysis on the dichloromethane extract of Xinhuang tablets, establishes the chemical fingerprint of 10 batches of Xinhuang tablets, and the similarity is between 0.978 and 0.997, indicating a high degree of consistency in chemical components between batches.

[0029] 2) The present invention for the first time constructs a dual-target evaluation system combining an LPS-induced RAW264.7 cell inflammation model and COX-2 enzyme inhibition detection, and finds that the dichloromethane extract significantly inhibits NO release (inhibition rate of 70.91-100.19%) at a concentration of 0.08 mg / mL, and the IC 50 values of each batch of COX-2 inhibition assays are in the range of 38.35-95.15 μg / mL.

[0030] 3) The present invention for the first time conducts a spectrum-effect relationship study on the anti-inflammatory activity of Xinhuang Tablets by means of grey relational analysis and OPLS-DA data mining methods, and accurately identifies four chemical components highly correlated with anti-inflammatory activity: indomethacin, 7-ketolithocholic acid, erucamide, and apigeninidin (correlation coefficient > 0.8).

[0031] 4) The present invention for the first time conducts monomer verification on the four main anti-inflammatory active components in Xinhuang Tablets, and finds that they exhibit differential activity characteristics in two different anti-inflammatory evaluation models, confirming the necessity of a multi-model evaluation system. Apigeninidin shows the best performance (54.79%) in the NO inhibition model, indomethacin is the most effective in the COX-2 inhibition model (IC 50 value is 38.53 ± 0.97 μg / ml), and 7-ketolithocholic acid and erucamide also show certain activities. Description of the Drawings

[0032] Figure 1 : UPLC fingerprint of 10 batches of extracted samples of Xinhuang Tablets.

[0033] Figure 2 : UPLC reference fingerprint of 10 batches of extracted samples of Xinhuang Tablets.

[0034] Figure 3 : Dendrogram of hierarchical cluster analysis of 10 batches of Xinhuang Tablet samples.

[0035] Figure 4 : Score plot (a) and loading plot (b) of principal component analysis.

[0036] Figure 5 : Screening diagram of the optimal administration concentration of the extracted part of Xinhuang Tablets.

[0037] Figure 6 : Correlation degree between the peak area of the common peak in the fingerprint and the NO inhibition rate of RAW264.7 cells.

[0038] Figure 7 : OPLS-DA score plot (a) and loading plot (b) based on the NO inhibition rate.

[0039] Figure 8: Variable importance in projection (VIP) values of the OPLS-DA model based on the NO inhibition rate.

[0040] Figure 9 : Correlation degree between the peak area of the common peaks in the fingerprint and the COX-2 inhibition rate.

[0041] Figure 10 : OPLS-DA score plot (a) and loading plot (b) based on the COX-2 inhibition rate.

[0042] Figure 11 : Variable importance in projection (VIP) values of the OPLS-DA model based on the COX-2 inhibition rate.

[0043] Figure 12 : Comparison of the inhibitory effects of four standards on the NO release of LPS-induced RAW264.7 cells.

[0044] Figure 13 : Inhibitory activities (IC 50 ) of four standards against COX-2 enzyme. Detailed implementation mode

[0045] The present invention will be fully and clearly described below through specific embodiments in combination with the accompanying drawings. Obviously, the protection scope of the present invention is not limited thereto.

[0046] The present invention established a chemical fingerprint of Xinhuang Tablet by UPLC-Triple-TOF-MS / MS and constructed a dual-target anti-inflammatory activity evaluation system. This system comprehensively evaluated the anti-inflammatory effect of Xinhuang Tablet through the inhibition rate of NO production in the LPS-induced macrophage inflammation model and the inhibitory activity of the cyclooxygenase-2 (COX-2) inhibitor screening kit. Grey relational analysis and orthogonal partial least squares discriminant analysis (OPLS-DA) were used to study the spectrum-effect relationship, and potential anti-inflammatory active components in Xinhuang Tablet were explored from different angles.

[0047] In the following examples,

[0048] Xinhuang Tablet is from Xiamen Traditional Chinese Medicine Factory Co., Ltd.

[0049] Example 1 Preparation of the dichloromethane extraction part of Xinhuang Tablet

[0050] Using 10 batches of Xinhuang Tablets as raw materials, accurately weigh them, pulverize them with a pulverizer and sieve through a 80-mesh sieve. Mix 10 g of the sieved powder with 75 mL of methanol and stir evenly, then transfer it to a round-bottom flask and heat under reflux at 85 °C for 3 h. After extraction, cool the solution and filter it. Repeat the extraction of the residue twice, combine the extracts, and concentrate under reduced pressure to obtain the total extract of the fraction. The remaining residue is extracted by water extraction method, heated under reflux for 1.5 hours, cooled and filtered. Combine this filtrate with 100 mL of the total extract of the fraction and dissolve it fully. Perform liquid-liquid extraction with dichloromethane on the mixed solution, extract with dichloromethane at a solvent-to-sample volume ratio of 3:1, and extract three times. After the extract is concentrated under reduced pressure and evaporated to dryness in a water bath, the extract of the dichloromethane fraction is obtained. The extract is freeze-dried for 48 h and stored in the dark. Before UPLC fingerprint analysis, all sample solutions are filtered through a 0.22 μm membrane. For the anti-inflammatory activity evaluation, dissolve the dichloromethane extracts of 10 batches of Xinhuang Tablets in a complete medium to prepare a stock solution of 1 mg / mL, and sequentially dilute it with the complete medium into working solutions with concentration gradients of 0.01, 0.02, 0.04, 0.06, 0.08, 0.1, 0.2, 0.4 mg / mL.

[0051] Example 2 UPLC-Triple-TOF-MS Analysis and Fingerprint Establishment

[0052] (1) Chromatographic Conditions

[0053] The chromatographic column is a Waters ACQUITY UPLC HSS T3 column (2.1 mm x 150 mm, 1.8 μm); the mobile phase is 0.1% FA aqueous solution (A) - 0.1% FA acetonitrile solution (B), with linear gradient elution (0 min, 5% B; 0 - 15 min, 5% → 40% B; 15 - 35 min, 40% → 95% B; 35 - 37 min, 95% B; 37 - 38 min, 95% → 5% B). The flow rate is 0.3 mL·min -1 , the column temperature is 50 °C, the injection volume is 3 μL, and the detection wavelength is 240 nm.

[0054] (2) Mass Spectrometry Conditions

[0055] UPLC-Triple-TOF 6600 +Time-of-flight liquid chromatography-mass spectrometry was performed in both positive and negative ESI modes. Scanning range: m / z 100 - 2000; Nebulizing gas (GS1) was 55 psi; Nebulizing gas (GS2) was 55 psi; Curtain gas (CUR) was 35 psi; Ion source temperature (TEM) was 600 °C (positive) and 550 °C (negative); Ion source voltage (IS) was 5500 V (positive) - 4500 V (negative); First-stage scan: Declustering potential (DP) was 100 V; Collision energy (CE) was 10 V; Second-stage scan: Mass spectrometry data were acquired using the TOF MS - Product Ion - IDA mode, and the collision-induced dissociation (CID) energy was ±40 ± 20 eV. Before injection, the mass axis was calibrated using a CDS pump to ensure that the mass axis error was less than 2 ppm.

[0056] (3) Establishment of fingerprint

[0057] Take 10 g of the extracted part sample of Xinhuang Tablets, accurately weigh it, add 75 mL of methanol to dissolve it, sonicate and make up the volume. After the test solution is filtered through a 0.22 μm microporous filter membrane, it is analyzed and detected under the above-mentioned chromatographic-mass spectrometric conditions. The mass spectrometry information of each batch of samples is collected, and data processing and similarity evaluation are carried out using the "Similarity Evaluation System for Traditional Chinese Medicine Fingerprints" (2012 version). The chromatographic peaks are automatically matched by methods such as multi-point calibration and time window attribution. A reference fingerprint is set, and the similarity between the extracted part samples of each batch of Xinhuang Tablets and the reference fingerprint is calculated, and then the fingerprint of Xinhuang Tablets is established.

[0058] (4) Results of fingerprint analysis

[0059] Figure 1 and Figure 2 show the UPLC fingerprints and reference fingerprints of 10 batches of extracts of Xinhuang Tablets. Table 1 compiles the detailed information of the common characteristic peaks in the fingerprint chromatogram. UPLC fingerprint analysis identified 15 different peaks that were consistently present in all samples, which were designated as common characteristic markers. Indomethacin (peak 4) was selected as the reference standard for methodological verification due to its superior stability characteristics and optimal signal intensity.

[0060] UPLC-Triple-TOF-MS analysis showed that peak 1 (7.781 min) was preliminarily identified as isofraxidin (C 11 H 10 O5), and its [M+H] + ion was at m / z 223.0602, and the MS 2 fragment was at m / z 207.0286 (corresponding to the loss of -OH), 190.0256, and 162.0308 (corresponding to the cleavage of ring A). Peaks 2 and 3 were inferred to be deoxycholic acid (C 24 H 40O4) and 7-ketolithocholic acid (C 24 H 38 O4). The [2M+H]+ dimer ion (m / z 789.5927) of deoxycholic acid and the fragment at m / z 355.2624 (corresponding to the neutral loss of water molecule) are the key evidence for its structure confirmation. Peak 4 is indomethacin (C + H 19 H 16 ClNO4), and the MS 2 fragment at m / z 174.0911 reflects the fragmentation characteristics of the indole ring. Peaks 5 and 6 are tentatively assigned as docosahexaenoic acid isomers (C 24 H 36 O2), and are confirmed by comparing their different MS 2 fragmentation patterns (Peak 5: m / z 147.1170, 95.0855; Peak 6: m / z 215.1797, 161.1323). Similarly, Peaks 7 and 9 are inferred to be eicosapentaenoic acid isomers by MS 2 fragmentation pattern analysis. The structure of Peak 8 (pyranoapigeninidin) is confirmed by its characteristic fragment ions at m / z 312.0780 (corresponding to the loss of aglycone) and m / z 138.9947 (corresponding to the cleavage of ring A). Peak 10 (retention time 30.058 min) is identified as 1-monopalmitin (C 19 H 38 O4), and its MS 2 fragments at m / z 313.2737 (loss of water), 239.2379 (loss of glycerol moiety) and 95.0856 confirm the monoacylglycerol structure. Peak 11 (31.851 min) is identified as erucamide (C 22 H 43 NO), with diagnostic MS 2 fragments at m / z 321.3147 and 303.3047, indicating successive losses of NH3 and H2O from the amide group. Peak 12 (33.262 min) corresponds to monostearin (C 21 H 42 O4), generating MS 2 fragments at m / z 341.3041 (dehydration product) and 109.1012 (glycerol fragment). Peak 13 (35.226 min) is inferred to be an erucamide isomer, with a unique MS 2 fragmentation pattern at m / z 223.0646 and 207.0329, which is different from Peak 11 due to different unsaturated positions. Peaks 14 (36.010 min) and 15 (37.352 min) are based on their molecular formulas and MS 2The fragments were preliminarily identified as polyphenolic compounds at m / z 281.0499, 147.0651 (peak 14) and m / z 355.0708, 281.0514, 207.0328 (peak 15), indicating a flavonoid glycoside structure with a characteristic glycosidic cleavage pattern.

[0061] Table 1 Details of the common characteristic peaks in the fingerprint chromatogram

[0062]

[0063] (5) Similarity evaluation

[0064] Pattern comparison evaluation was carried out by comparing with the established reference fingerprint (R), and the quantitative correlation indices are summarized in Table 2. The similarity evaluation showed that the similarity range of the 10 batches of samples to the control fingerprint was 0.978 - 0.997. These highly consistent correlation values demonstrated strong quality consistency and compositional uniformity among production batches, confirming the reliability of the manufacturing process and the stability of the chemical characteristics. Since peak 4 had good resolution and high response value, it was used as the reference peak to calculate the relative peak areas of each common peak.

[0065] Table 2 Relative peak areas and similarity evaluation of ten batches of Xinhuang tablets samples

[0066]

[0067]

[0068] (6) Hierarchical cluster analysis and principal component analysis

[0069] Based on the relative peak areas of 15 common peaks in the UPLC-Triple-TOF-MS fingerprint of 10 batches of Xinhuang tablets samples, systematic cluster analysis was performed using SIMCA 14.1 software. The Ward linkage method was selected as the clustering method and the Euclidean distance was used as the distance measure. The systematic cluster analysis adopted the default parameter settings, and the data were standardized to eliminate the influence of dimensional differences. The clustering process was visually demonstrated through a dendrogram to show the similarity relationship among the samples. From Figure 3 the dendrogram, it can be seen that the 10 batches of Xinhuang tablets samples can be divided into two main groups at a dendrogram distance of about 32: the first group includes S6, S1, S10 and S9, and the second group includes S4, S5, S2, S8, S3 and S7. In the first group, the similarity between S1 and S10 is relatively high, and the clustering distance is about 3; in the second group, the similarity between S3 and S7 is the highest, and the clustering distance is about 1. The clustering distance of sample S9 from other samples is relatively large, indicating that its chemical composition is somewhat different from that of other batches. This clustering distribution pattern may be related to the different production batches of the samples, reflecting the inherent variability of Xinhuang tablets in chemical composition.

[0070] To further explore the relationship and key differential components among 10 batches of Xinhuang Tablets samples, principal component analysis was performed on the relative peak area data of 15 common peaks. Unsupervised pattern recognition was carried out using SIMCA-P 14.1 software. Figure 4 A two-dimensional plot containing both the score (a) and the loading matrix (b) is presented. The cumulative explained variance of the first two principal components (PC1 and PC2) reaches 68.5%, which can better reflect the variation information among the samples. As can be seen from (a), the 10 batches of samples show a relatively scattered distribution in the two-dimensional space, but are overall within the 95% confidence ellipse, indicating that there are certain differences among the samples but no obvious outliers. Sample S9 shows a large negative value in the PC2 direction, which is consistent with the results of hierarchical cluster analysis, further confirming that this batch of samples has unique chemical composition characteristics. Samples S1, S10, and S6 show positive distributions in the PC1 direction, S5 is near the origin, while S7, S4, and S3 are mainly distributed in the negative direction of PC1. This distribution pattern basically conforms to the grouping results of cluster analysis. (b) reveals the contribution of each chemical component to sample classification. Variables Var_13 (corresponding to peak 13, erucamide) and Var_14 have large positive loading values in the PC2 direction and are the main components differentiating S9 from other samples; variables Var_3, Var_7, Var_8, Var_9, and Var_11 show large negative loading values in both the PC1 and PC2 directions and contribute greatly to the characteristics of samples such as S7, S3, and S4; while Var_1 (isofraxidin) mainly affects the distribution of samples in the positive direction of PC1.

[0071] Combined with the results of hierarchical cluster analysis and principal component analysis, it shows that there are certain differences in the chemical composition of different batches of Xinhuang Tablets samples, which are mainly caused by the changes in the contents of characteristic components. These findings provide an important basis for subsequent studies on the relationship between spectra and efficacy.

[0072] Establishment of NO inhibition model and evaluation of anti-inflammatory activity in Example 3

[0073] (1) Cell culture

[0074] RAW264.7 cells were placed in complete medium (89% high-glucose DMEM culture solution, 10% fetal bovine serum, and 1% double antibody) and cultured under the conditions of 37 °C and 5% CO2. When the cell confluence reached 70%-80%, subculture was carried out. Before subculture, the cell morphology was observed first to ensure good cell status.

[0075] (2) Preparation of samples to be tested

[0076] After the samples were sterilized by ultraviolet for 30 min, they were added to the complete medium at a ratio of 1 mg / mL, and ultrasonic dispersion was used to prepare a suspension, which was used as the stock solution. The stock solution was serially diluted with the complete medium into sample working solutions with different concentration gradients of 0.01, 0.02, 0.04, 0.06, 0.08, 0.1, 0.2, and 0.4 mg / mL. The prepared sample working solutions need to be prepared and used immediately to avoid long-term storage.

[0077] (3) Effects of extracts of Xinhuang Tablets at different concentrations on the proliferation viability of RAW264.7 cells

[0078] RAW264.7 cells in the logarithmic growth phase were digested with 0.25% trypsin, washed with PBS, and then cell counting was performed. The cell concentration was adjusted, and they were inoculated into 96-well plates at a density of 1×10 4 cells / well, and cultured in a 5% CO2, 37 °C constant temperature incubator for 24 h to allow the cells to adhere completely. The samples were divided into a control group and a dichloromethane fraction group. Among them, the control group was added with 100 μL / well of the complete medium; the dichloromethane fraction group was added with 100 μL / well of sample working solutions with concentrations of 0.01, 0.02, 0.04, 0.06, 0.08, 0.1, 0.2, and 0.4 mg / mL, respectively. Four replicate wells were set for each concentration, and the blank control wells were added with only medium without cells. After treatment according to the above grouping, the cells were cultured for another 24 h, and the medium in the wells was discarded. 100 μL of serum-free medium containing 10% CCK-8 was added to each well, and the cells were continued to be cultured in a 37 °C, 5% CO2 constant temperature incubator for 2 h. The absorbance value at 450 nm was measured using an enzyme-linked immunosorbent assay (ELISA) reader to calculate the cell viability, and based on the cell viability results, the safe concentration range used in the subsequent anti-inflammatory experiments was determined.

[0079] Cell viability (%) = (OD value of the experimental group - OD value of the blank group) / (OD value of the control group - OD value of the blank group) × 100%

[0080] The effects of the extracts of Xinhuang Tablets at concentrations of 0.01 - 0.4 mg / mL on the proliferation viability of RAW264.7 cells were compared. The results Figure 5 showed that when the concentration of the extract was in the range of 0.01 - 0.1 mg / mL, it could significantly promote cell proliferation, and the strongest proliferation-promoting effect was shown at a concentration of 0.08 mg / mL; while when the concentration exceeded 0.1 mg / mL, it showed a trend of inhibiting cell proliferation. Based on the above results, 0.08 mg / mL was determined as the optimal administration concentration.

[0081] (4) Effects of extracts of Xinhuang Tablets on the NO level of RAW264.7 cells induced by LPS

[0082] RAW264.7 cells in the logarithmic growth phase were counted, the cell concentration was adjusted, and they were inoculated at a density of 1×104 Inoculate into 96-well plates at a density of 4 per well with 100 μL per well, and place them in an incubator at 37 °C with 5% CO2 for 24 h to allow the cells to adhere completely. Divide the samples into a control group, an LPS group, and a dichloromethane fraction group. Among them, the control group was added with 100 μL / well of complete medium; the LPS group and the dichloromethane group were respectively added with 100 μL / well of LPS working solution with a concentration of 1 μg / mL. After incubation for 1 h, the LPS group was added with 100 μL / well of complete medium, and the dichloromethane fraction group was added with 100 μL / well of the sample working solution. Each treatment group was set with 3 replicate wells. According to the above grouping treatment, continue to culture for 24 h. React according to the instructions of the nitric oxide (NO) detection kit (Shanghai Beyotime Biotechnology Co., Ltd.), and use an enzyme-linked immunosorbent assay (ELISA) reader to detect the absorbance value at 540 nm. Calculate the NO concentration of each sample group and the standard product group through the standard curve, compare with the LPS group, and calculate the inhibition rate.

[0083] NO inhibition rate (%) = (NO concentration in the LPS group - OD concentration in the experimental group) / (NO concentration in the LPS group - NO concentration in the blank group) × 100%

[0084] Table 3 of the results shows that the dichloromethane extraction fraction of Xinhuang Tablets can significantly inhibit the NO release of RAW264.7 cells induced by LPS, showing good anti-inflammatory activity, but there are obvious differences in the pharmacological effects among samples of different batches. Among them, sample S2 showed the strongest NO inhibition ability (inhibition rate reached 100.19%), while the inhibitory effect of batch S10 was relatively weak (70.91%).

[0085] Table 3 NO inhibition rates of ten batches of Xinhuang Tablet samples on the in vitro cell inflammation model

[0086]

[0087] Example 4 Establishment of a COX-2 inhibition model and evaluation of anti-inflammatory activity

[0088] (1) Preparation of experimental samples

[0089] Take an appropriate amount of the dichloromethane extract of 10 batches of Xinhuang Tablets, dissolve it with DMSO and prepare a stock solution of 1 mg / mL. Then dilute it to 6 concentration gradients with COX-2 Assay Buffer (assay buffer): 1, 5, 10, 25, 50, 100 μg / mL for subsequent evaluation of COX-2 inhibitory activity. Prepare the reagents according to the instructions of the COX-2 inhibitor screening kit (Shanghai Beyotime Biotechnology Co., Ltd.), including the preparation of COX-2 Cofactor (cofactor) working solution, COX-2 working solution, and COX-2 Substrate (substrate) working solution. The prepared working solutions are stored at the appropriate temperature and for the specified time according to the kit instructions.

[0090] Table 4 Experimental reagents for COX-2 kit

[0091]

[0092] (2) Determination of COX-2 inhibitory activity

[0093] Set multiple groups of controls on a 96-well black plate: blank control group, 100% enzyme activity control group, positive inhibitor control group, and test sample group. Add the corresponding reagents to each well in turn and incubate at 37 °C for 10 minutes. Add 5 μL of COX-2 probe to each well, quickly add 5 μL of substrate working solution and mix well. After incubating at 37 °C in the dark for 5 minutes, perform fluorescence measurement. Set the excitation wavelength to 560 nm and the emission wavelength to 590 nm. Calculate the average fluorescence values of each sample well and the blank control well, which can be recorded as RFU blank control, RFU 100% enzyme activity control, RFU positive inhibitor control, and RFU sample respectively. Calculate the inhibition percentage of each sample. Calculate the COX-2 inhibition rate, plot the dose-effect curve with the sample concentration as the abscissa and the inhibition rate as the ordinate, and calculate the IC 50 value.

[0094] COX-2 inhibition rate (%) = (RFU 100% enzyme activity control - RFU sample) / (RFU 100% enzyme activity control - RFU blank control) × 100%

[0095] The results are shown in Table 5, indicating that there are significant differences in the inhibitory activities of different batches of samples against COX-2. Among them, the samples of batch S2 showed the strongest COX-2 inhibitory activity, with the lowest IC 50 value of 38.35 ± 3.15 μg / mL; while the inhibitory activity of the samples of batch S10 was relatively weak, with the highest IC 50 value of 95.15 ± 8.35 μg / mL, about 2.5 times that of batch S2. Overall, the IC 50The values are distributed in the range of 38.35 - 95.15 μg / mL, indicating that they all have a certain COX-2 inhibitory ability, but there are significant batch-to-batch differences.

[0096] Table 5 Inhibitory activities of ten batches of Xinhuang tablets samples on COX-2 enzyme (IC 50 ) comparison

[0097]

[0098] Example 5 Grey relational analysis and OPLS-DA analysis based on NO inhibitory activity

[0099] (1) Construction of data sequences

[0100] Taking the inhibition rates of 10 batches of different Xinhuang tablets extracts on the NO release of RAW264.7 macrophages as the reference sequence, denoted as X0 = {x0(1), x0(2),..., x0(10)}. Taking the peak areas of 15 common peaks identified in the UPLC fingerprint as the comparison sequences, denoted as X i = {x i (1), x i (2),..., x i (10)}, where i = 1, 2,..., 15 represent 15 characteristic chromatographic peaks respectively.

[0101] (2) Non-quantitative processing of original data

[0102] The original data is standardized by the mean method to eliminate the influence of dimension. This method makes the data comparable by calculating the mean of each index and performing corresponding conversions, laying a foundation for subsequent correlation analysis.

[0103] (3) Grey relational degree analysis

[0104] Taking the NO inhibition rates of 10 batches of Xinhuang tablets as the reference sequence and combining the peak areas of 15 chromatographic peaks as the comparison sequences, import them into the grey modeling software (seventh edition), and use the Deng's correlation degree analysis method. After sequence initial value processing, difference sequence calculation, range determination, and correlation coefficient calculation, finally obtain the correlation degree values of each chromatographic peak with the two anti-inflammatory activities.

[0105] By Figure 6From the grey relational analysis data of the common peaks in the extraction parts of 10 batches of Xinhuang Tablets and their NO inhibition rates on RAW264.7 cells, it can be seen that the contribution rates of 15 common peaks (peak numbers) to the drug efficacy are 4>3>11>8>14>15>1>2>12>5>9>6>7>13>10. Among them, the correlation degrees of peaks 4, 3, 11, and 8 are all greater than 0.8. This shows that the anti-inflammatory activity of Xinhuang Tablets is the combined effect of multiple compounds contained in the extraction parts. The compounds that contribute the most to the anti-inflammatory effect are peak 3 (7-ketolithocholic acid), peak 4 (indomethacin), peak 8 (pyranapigeninidin), and peak 11 (erucamide).

[0106] (4) OPLS-DA analysis

[0107] Based on the UPLC-Q-TOF-MS fingerprint data and NO inhibition rate data of 10 batches of Xinhuang Tablet samples, OPLS-DA analysis was carried out using SIMCA-P 14.1 software, and the results are as Figure 7 shown. According to the anti-inflammatory activity strength of the samples, the 10 batches of samples were divided into a high-activity group (S1, S2, S3, S4, S8) and a low-activity group (S5, S6, S7, S9, S10). As Figure 7 shown in (a), in the score scatter plot, the two groups of samples show an obvious separation trend, and all samples are located within the 95% confidence ellipse, indicating that the classification effect of the model is good and there are no outlier samples. The high-activity group samples (blue dots) are mainly distributed in the area on the right side of the coordinate axis (t[1]>0), while the low-activity group samples (green dots) are mainly distributed in the left area (t[1]<0), indicating that the first principal component (t[1]) can effectively distinguish samples with different activity levels. Figure 7 The loading scatter plot (Loading plot) in (b) reveals the contribution degree of each variable to the sample classification. It can be seen that variables such as Var_3, Var_8, Var_9, and Var_11 have relatively large load values in the positive direction of the t[1] axis, which is consistent with the distribution direction of the high-activity group samples, indicating that these components may be the main substances contributing to the anti-inflammatory activity of Xinhuang Tablets. Var_4, as the reference standard for the relative peak area, has a value of 1 in each batch of samples, so its changing trend is not reflected in the OPLS-DA model. The component represented by Var_4 makes an important contribution to the anti-inflammatory activity of Xinhuang Tablets. Variables such as Var_13 and Var_14 mainly affect the distribution in the t[2] direction and have a relatively weak relationship with the sample activity. The two blue dots in the figure are located at both ends of the horizontal coordinate axis, indicating that the established OPLS-DA model has good fitting performance and prediction ability.

[0108] Figure 8The VIP (Variable Importance in Projection) value distributions of each variable are shown. Generally, variables with VIP>1 are considered to make significant contributions to the model. The results show that the VIP values of Var_9, Var_1, Var_13, Var_3, Var_11, Var_12, and Var_8 are all greater than 1 (red bars), which are the key components for distinguishing high- and low-activity samples. These results are basically consistent with the grey relational analysis. In particular, peaks 3, 8, and 11 are screened as important active components in both analysis methods, further verifying their significant contributions to the anti-inflammatory activity of Xinhuang tablets.

[0109] Combining the grey relational analysis and OPLS-DA results, it can be confirmed that peak 3 (7-ketolithocholic acid), peak 4 (indomethacin), peak 8 (apigeninidin), and peak 11 (erucamide) are the key components for Xinhuang tablets to exert anti-inflammatory effects.

[0110] Example 6 Grey relational analysis and OPLS-DA analysis based on COX-2 inhibitory activity

[0111] (1) Construction of data sequences

[0112] Taking the reciprocal of the IC 50 value (1 / IC 50 ) of the COX-2 inhibitory activity of 10 batches of different Xinhuang tablet extracts as the reference sequence, denoted as Y0 = {y0(1), y0(2),..., y0(10)}. Taking the peak areas of 15 common peaks identified in the UPLC fingerprint as the comparison sequences, denoted as Y i = {y i (1), y i (2),..., yi(10)}, where i = 1, 2,..., 15 represent 15 characteristic chromatographic peaks respectively.

[0113] (2) Non-quantitative processing of original data

[0114] The original data is standardized by the mean method to eliminate the influence of dimensions. This method calculates the mean of each index and performs corresponding conversions to make the data comparable, laying a foundation for subsequent relational degree analysis.

[0115] (3) Grey relational analysis

[0116] Taking the reciprocal of the COX-2 inhibitory activity (IC 50 value) of 10 batches of Xinhuang tablets as the reference sequence, combined with the peak areas of 15 chromatographic peaks as the comparison sequences, imported into the grey modeling software (seventh edition), and using the Deng's relational degree analysis method, after sequence initial value processing, difference sequence calculation, range determination, and correlation coefficient calculation, the relational degree values of each chromatographic peak with the two anti-inflammatory activities are finally obtained.

[0117] From Figure 9 The grey relational grade analysis data of the common peaks and COX-2 inhibitory activity of the extraction parts of 10 batches of Xinhuang tablets show that the contribution rates of 15 common peaks (peak numbers) to the drug efficacy are 8 > 14 > 9 > 11, 3 > 4 > 15 > 5, 2 > 1 > 12 > 13 > 6 > 10. Among them, the correlation degree of peak 8 is greater than 0.8, and the correlation degrees of peaks 3, 4, 9, 11 and 14 are all greater than 0.75. This shows that the anti-inflammatory activity of Xinhuang tablets is the result of the combined action of multiple compounds contained in the extraction parts.

[0118] (4) OPLS-DA analysis

[0119] According to the anti-inflammatory activity of the samples, 10 batches of samples were divided into a high-activity group (S1, S2, S4, S6, S8) and a low-activity group (S3, S5, S7, S9, S10). As Figure 10 shown in (a) below, in the score scatter plot, the two groups of samples show an obvious separation trend, and all samples are located within the 95% confidence ellipse, indicating that the classification effect of the model is good and there are no abnormal samples. The high-activity group is mainly distributed in the area on the right side of the coordinate axis (t[1] > 0), while the low-activity group is mainly distributed on the left side (t[1] < 0), indicating that the first principal component can effectively distinguish samples with different activity levels. The loading scatter plot reveals that variables such as Var_3, Var_8, Var_9 and Var_11 have high loading values in the positive direction of the t[1] axis, which is consistent with the distribution direction of the high-activity group, indicating that these components may be the main substances contributing to the anti-inflammatory activity of Xinhuang tablets. Although Var_4, as a reference standard for the relative peak area, does not show a change trend in the OPLS-DA model, its important contribution to the anti-inflammatory activity is confirmed by other analysis methods.

[0120] Variable importance in projection plot ( Figure 11 ) analysis shows that the VIP values of Var_2, Var_9, Var_6, Var_8, Var_5 and Var_12 are all greater than 1, which are the key components for distinguishing high- and low-activity samples. These results are basically consistent with the grey relational grade analysis. In particular, peaks 3, 8 and 11 are selected as important active components in both analysis methods, further verifying their significant contribution to the anti-inflammatory activity of Xinhuang tablets.

[0121] Evaluation of the anti-inflammatory activities of four monomers in Example 7

[0122] By taking the intersection of grey relational analysis and OPLS-DA, 7-ketolithocholic acid, indomethacin, apigeninidin and erucamide may be the core contributors to the anti-inflammatory activity of the dichloromethane extract. Therefore, the NO and COX-2 inhibitory activities of these four monomers were evaluated separately.

[0123] By measuring the NO results of four monomers, Figure 12 it was shown that all four monomers exhibited an obvious dose-dependent NO inhibitory effect, but there were differences in the intensity of the effect. Apigeninidin showed strong activity (54.79%) in the NO inhibition experiment, while the traditional COX-2 inhibitor indomethacin had limited effects (25.03%) in the NO inhibition experiment. Erucamide had almost no significant inhibitory effect on NO release, and the NO levels in each concentration group were similar to those in the LPS-stimulated group. 7-Ketolithocholic acid showed a slight inhibitory effect only at the highest concentration (200 μM).

[0124] Through the study of the dose-effect relationship, all four standards showed different degrees of inhibitory effects on COX-2. As the concentration increased, the inhibition rate of the standards on COX-2 gradually increased. The IC 50 values of each standard were calculated by curve fitting as Figure 13 shown. Compared with the results of the NO inhibition model, indomethacin showed the strongest inhibitory activity in the COX2 inhibition model, with the lowest IC 50 value, which was consistent with its characteristics as a widely used cyclooxygenase inhibitor in clinical practice. The COX-2 inhibitory activity of apigeninidin was at a medium level, with an IC 50 of 69.69 μM. 7-Ketolithocholic acid also had a certain inhibitory effect on COX-2, with an IC 50 of 60.85 μM. While erucamide had the weakest activity in the COX-2 inhibition model, with an IC 50 as high as 155.9 μM. This result indicates that the anti-inflammatory activity evaluation should be comprehensively evaluated using multiple models to fully reveal the action characteristics of drugs.

Claims

1. A method for identifying the anti-inflammatory active components of Xinhuang tablets based on the spectrum-effect relationship analysis of UPLC-Triple-TOF-MS, characterized in that, The method includes: S1. Take Niuhuang Jiedu Tablets, crush and sieve them. The obtained powder is reflux-extracted with methanol, filtered, and the filtrate and residue are collected separately. The filtrate is concentrated under reduced pressure to obtain a methanol extract. The residue is reflux-extracted with water, filtered, and the water extract is collected. The water extract and the methanol extract are combined to obtain a mixed extract; S2. Perform liquid-liquid extraction of the mixed extract obtained in S1 with dichloromethane. The extract is concentrated under reduced pressure and freeze-dried to obtain a dichloromethane extract, which is stored in the dark and dried; S3. Analyze the dichloromethane extracts of 10 batches of Niuhuang Jiedu Tablets using UPLC-Triple-TOF-MS, establish a fingerprint, and determine 15 common characteristic peaks; S4. Establish a dual-target evaluation system and measure the anti-inflammatory activity of the dichloromethane extracts of 10 batches of Niuhuang Jiedu Tablets; S5. Adopt grey relational analysis and OPLS-DA data mining methods to analyze the spectrum-effect relationship between the UPLC-Triple-TOF-MS fingerprint obtained in S3 and the anti-inflammatory activity obtained in S4, and screen out the anti-inflammatory active ingredients.

2. The method for identifying the anti-inflammatory active ingredients of Xinhuang Tablets based on the spectrum-effect relationship analysis of UPLC-Triple-TOF-MS according to claim 1, characterized in that, The operation of S1 is as follows: Take Niuhuang Jiedu Tablets, crush and sieve them. Mix the medicinal material powder with methanol at a solid-liquid ratio of 1:5 - 10, heat to 85°C for reflux extraction for 3 h, cool to room temperature and filter. The residue is extracted 1 - 2 times repeatedly, and the methanol extracts are combined and concentrated under reduced pressure to obtain a methanol extract. The remaining residue is added to water at a solid-liquid ratio of 1:10 - 20 and refluxed for 1.5 h, cooled to room temperature and filtered. The water extract and the methanol extract are combined to obtain a mixed extract.

3. The method for identifying the anti-inflammatory active ingredients of Xinhuang Tablets based on the spectrum-effect relationship analysis of UPLC-Triple-TOF-MS according to claim 1, characterized in that, The operation of S2 is as follows: Take the mixed extract obtained in S1, add 3 times the volume of dichloromethane for liquid-liquid extraction, extract 3 times, combine the extracts, concentrate under reduced pressure, and freeze-dry to obtain a dichloromethane extract.

4. The method for identifying the anti-inflammatory active ingredients of Xinhuang Tablets based on the spectrum-effect relationship analysis of UPLC-Triple-TOF-MS according to claim 1, characterized in that In S3, before analysis, the sample pretreatment is: Take the dichloromethane extract, dissolve it with methanol, and filter through a 0.22 μm membrane to obtain a sample solution; The UPLC-Triple-TOF-MS analysis conditions are as follows: Chromatographic column: ACQUITY UPLC HSS T3, mobile phase A: acetonitrile containing 0.1% formic acid, mobile phase B: water containing 0.1% formic acid, gradient elution conditions: 0 - 15 min, 5 - 40% A; 15 - 35 min, 40 - 95% A; 35 - 37 min, 95% A; 37 - 38 min, 95 - 5% A; flow rate: 0.3 mL·min -1 , injection volume: 3 μL, column temperature: 50 °C, detection wavelength: 240 nm; Mass spectrometry was performed using a Triple-TOF 6600 + system, equipped with an electrospray ionization source (ESI) operating in positive and negative ion modes, mass scan range: 100 - 2000 Da; nebulizing gas 1 (GS1): 55 psi; nebulizing gas 2 (GS2): 55 psi; curtain gas (CUR): 35 psi; ion source voltage (IS): 5500 V (positive ion mode) - 4500 V (negative ion mode); ion source temperature: 600 °C (positive ion mode) and 550 °C (negative ion mode); first-stage scan: declustering potential (DP) 100 V; focusing voltage (CE) 10 V; second-stage mass spectrometry data was acquired in IDA mode, and the collision-induced dissociation (CID) energy was ±40 ± 20 eV; The sample is introduced through a CDS pump, which is also used for mass axis calibration before analysis to ensure that the mass accuracy is within 2 ppm.

5. The method for identifying the anti-inflammatory active ingredients of Xinhuang Tablets based on the spectrum-effect relationship analysis of UPLC-Triple-TOF-MS according to claim 1, wherein In S4, the dual-target evaluation system refers to: the LPS-induced RAW264.7 cell inflammation model and the COX-2 activity inhibition model; the NO inhibition rate of the dichloromethane extract of 10 batches of Xinhuang tablets was detected through the LPS-induced RAW264.7 cell inflammation model; the IC 50 value of the dichloromethane extract of 10 batches of Xinhuang tablets was detected through the COX-2 activity inhibition model, that is, the COX-2 inhibitory activity.

6. The method for identifying the anti-inflammatory active components of Xinhuang tablets based on the spectrum-effect relationship analysis of UPLC-Triple-TOF-MS according to claim 5, wherein, In S5, the method for spectrum-effect relationship analysis is as follows: Take the NO inhibition rate and COX-2 inhibition activity of the dichloromethane extracts of 10 batches of Niuhuang Jiedu Tablets as the reference sequences, and combine the peak areas of 15 common characteristic peaks in the fingerprint as the comparison sequences, import them into grey modeling software for correlation analysis, calculate the correlation coefficient r of each detected compound relative to the NO and COX-2 inhibition abilities. A correlation coefficient r > 0.8 indicates a significant strong positive correlation between variables, showing a robust linear relationship; Use the SIMCA-P software platform (version 14) to perform chemometric analysis on the chromatographic data of the dichloromethane extracts of 10 batches of Niuhuang Jiedu Tablets. First, use the unsupervised pattern recognition method: principal component analysis (PCA) for dimension reduction and data structure visualization. At the same time, use hierarchical cluster analysis (HCA) to generate a dendrogram to visually show the similarities and differences between different batches; Subsequently, grey relational analysis (GRA) and orthogonal partial least squares-discriminant analysis (OPLS-DA) were applied to identify the chemical markers that contributed most significantly to the activity differences; in the OPLS-DA model, compounds with variable importance in projection (VIP) scores equal to or exceeding 1.0 were determined as statistically significant bioactive markers.

7. The method for identifying the anti-inflammatory active components of Xinhuang tablets based on the spectrum-effect relationship analysis of UPLC-Triple-TOF-MS according to claim 6, characterized in that, The finally identified anti-inflammatory active ingredients were: indomethacin, 7-ketolithocholic acid, erucamide, and apigeninidin.

Images