Quality control method for preparing heart failure treating medicine with tortoise shell glue

Abstract

This invention discloses a quality control method for preparing drugs for treating heart failure using tortoise shell glue, belonging to the field of traditional Chinese medicine quality evaluation and drug quality control technology. This method addresses the problem that existing tortoise shell glue quality control relies solely on physicochemical index detection, which cannot accurately reflect the efficacy of anti-heart failure drugs. It constructs a drug efficacy-oriented quality control system based on "chemical fingerprinting—pharmacodynamic quality markers—rapid bioactivity detection—comprehensive quality evaluation." This invention establishes a dual-channel evaluation system of hydrophilic and hydrophobic fingerprints, and combines this with anti-heart failure drug efficacy data to screen a group of pharmacodynamic quality markers. This solves the problem that existing tortoise shell glue quality control only detects amino acids or single chemical indicators, failing to reflect the true efficacy of anti-heart failure drugs. It achieves a shift from "chemical component detection" to "pharmacodynamic-related substance identification," improving the accuracy of quality evaluation and ensuring the consistency of efficacy across different batches of products.

Description

Technical Field

[0001] This invention relates to the field of traditional Chinese medicine quality evaluation and drug quality control technology, specifically a quality control method for preparing drugs for treating heart failure using tortoise shell glue. Background Technology

[0002] Tortoise shell glue is an animal-derived medicinal glue made from the carapace and plastron of turtles through extraction, concentration, and glue-making processes. It has high application value in tonifying the liver and kidneys, nourishing yin and suppressing yang, and regulating the cardiovascular system. Currently, the quality control of tortoise shell glue for treating heart failure mainly relies on methods such as total nitrogen determination, amino acid content determination, physicochemical index detection, and single chromatographic fingerprint analysis. Its quality evaluation is essentially still based on a chemical composition control model, which can only reflect some material basis information and is difficult to accurately evaluate the actual efficacy level of the product. With the development of the theory of quality markers (Q-Marker) for traditional Chinese medicine, quality control is gradually shifting from simple component detection to efficacy-related evaluation, emphasizing the establishment of a quality evaluation system of "material basis—quality marker—biological effect".

[0003] However, for tortoise shell glue used to treat heart failure, different raw material sources, extraction processes, and glue-making processes can all lead to changes in the composition and bioactivity of functional peptides. Even if traditional physicochemical indicators are qualified, its anti-heart failure effect may still vary significantly. Therefore, existing quality control methods suffer from problems such as a disconnect between quality evaluation and clinical efficacy, difficulty in identifying the causes of efficacy fluctuations, and inability to achieve traceability of process quality. In recent years, research on the quality control of traditional Chinese medicine has proposed constructing an evaluation model of "quality markers—biological effects—whole-process quality control" to achieve consistent efficacy evaluation and quality stability control. Summary of the Invention

[0004] The purpose of this invention is to provide a quality control method for preparing drugs for treating heart failure using tortoise shell glue, so as to solve the technical problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a quality control method for preparing a drug for treating heart failure using tortoise shell glue, the method comprising the following steps: Obtain the tortoise shell glue sample to be tested; Parallel extraction was performed using polar and non-polar solvent systems respectively, and a first chemical fingerprint spectrum reflecting the hydrophilic active component and a second chemical fingerprint spectrum reflecting the hydrophobic active component were established. The anti-heart failure activity of different batches of samples was determined and a pharmacodynamic database was constructed. A spectrum-effect correlation model was established based on the correlation between chemical fingerprint characteristic peaks and pharmacodynamic parameters; Based on the contribution of spectrum-effect correlation, a group of characteristic biomarkers for anti-heart failure was screened and a quantitative evaluation model was established; The ability of samples to regulate cell viability, oxidative stress level, and mitochondrial membrane potential was detected using a cardiomyocyte injury model. The comprehensive quality index is obtained by integrating the chemical fingerprint evaluation results, the characteristic biomarker evaluation results, and the bioactivity evaluation results. The quality of raw materials, production process, and finished product in the preparation of heart failure medication using tortoise shell glue was assessed based on a comprehensive quality index.

[0006] Preferably, the first chemical fingerprint is established using an aqueous extract at a controlled temperature of 40–80°C to obtain chromatographic characteristics of oligopeptide components, collagen degradation product components, and water-soluble amino acid components; the second chemical fingerprint is established using a 50–90% (v / v) alcohol solvent extract to obtain chromatographic characteristics of hydrophobic peptide components, lipid-soluble derivative components, and small organic molecule components; a unified characteristic peak library is established using a retention time correction algorithm, and a two-dimensional chemical fingerprint matrix is ​​obtained through peak matching, peak area normalization, and batch-to-batch drift correction; the two-dimensional chemical fingerprint matrix simultaneously includes peak position information, peak area information, and inter-peak ratio information to improve the stability and repeatability of chemical composition evaluation of different batches of tortoise shell glue samples.

[0007] Preferably, the pharmacodynamic database is obtained by establishing a heart failure-related evaluation system, including at least three of the following: cell survival rate recovery rate, lactate dehydrogenase release inhibition rate, reactive oxygen species decline rate, mitochondrial membrane potential recovery rate, and calcium ion homeostasis recovery rate; a pharmacodynamic matrix is ​​constructed from the pharmacodynamic parameters corresponding to each batch of samples, and a chemical matrix is ​​constructed from the two-dimensional chemical fingerprint matrix; a spectrum-effect correlation model is established using at least one of partial least squares regression analysis, grey relational analysis, and random forest analysis; the contribution weight and importance score corresponding to each characteristic peak are calculated, and candidate characteristic peaks that are significantly related to the anti-heart failure pharmacodynamics are screened according to a preset threshold.

[0008] Preferably, the group of biomarkers for treating heart failure is obtained by: statistically analyzing the frequency of occurrence, stability coefficient, and efficacy contribution value of each candidate characteristic peak in different batches of samples; identifying candidate characteristic peaks with an occurrence frequency greater than 90%, a contribution weight greater than a preset threshold, and an inter-batch coefficient of variation less than 15% as core biomarkers; performing structural analysis on the core biomarkers and establishing a multi-component quantitative detection model; the group of biomarkers includes one or more of collagen-derived functional peptides, hydroxyproline-derived peptides, glycine-enriched peptides, proline-enriched peptides, and hydrophobic functional peptides; and using the multi-component quantitative detection model to obtain the content evaluation results of the biomarker group.

[0009] Preferably, the cell viability evaluation uses H9c2 cardiomyocytes to construct an oxidative damage model or a drug-induced damage model; the cell survival rate is measured after the sample is applied to the damaged model cells; the cell viability recovery index (CVI) is calculated; where CVI is the recovery ratio of cell viability in the sample group relative to that in the model group; when the CVI reaches a preset standard, the sample is determined to have basic anti-heart failure activity; the evaluation result is used as a primary efficacy evaluation parameter in the comprehensive quality index; and is used to verify the consistency between the chemical fingerprint evaluation results and the actual biological effects.

[0010] Preferably, the oxidative stress evaluation uses a fluorescent probe detection system to measure intracellular reactive oxygen species (ROS) levels; establishes the fluorescence intensity difference relationship between the sample group, model group, and normal group; calculates the ROS inhibition index (ROI); the ROI is used to characterize the ability of the tortoise shell glue sample to inhibit free radical accumulation under oxidative damage; and establishes an oxidative stress quality grade standard based on the ROI value; the oxidative stress evaluation result serves as a secondary efficacy evaluation parameter in the comprehensive quality index; and is also used to identify efficacy deviations caused by differences in raw materials or fluctuations in production processes.

[0011] Preferably, the mitochondrial function evaluation is performed using a mitochondrial membrane potential detection reagent system; the ratio of red fluorescence to green fluorescence intensity is obtained for the sample group, normal group, and model group; the mitochondrial membrane potential recovery index (MPRI) is constructed; the MPRI is calculated based on the recovery degree of the red-green fluorescence ratio of the sample group relative to the normal group and the model group; the MPRI is used as the core indicator for evaluating the improvement of cardiomyocyte mitochondrial function by tortoise shell glue; and a mitochondrial function quality grade standard is established based on the MPRI; the evaluation results are used as the third-level efficacy evaluation parameter in the comprehensive quality index.

[0012] Preferably, the comprehensive quality index is obtained by fusing chemical fingerprint similarity index, characteristic marker content index, cell viability recovery index, reactive oxygen species inhibition index, and mitochondrial membrane potential recovery index; the weights of each evaluation index are determined by entropy weight method, adaptive weight method, or machine learning weight optimization method; a five-dimensional quality evaluation model is constructed; the model can output quality scores of 0 to 100; the samples are divided into excellent, qualified, and unqualified grades according to the scores; and a corresponding quality evaluation report is generated.

[0013] Preferably, the method further includes a quality anomaly tracing step; when the comprehensive quality index is lower than a preset threshold, the deviation of the two-dimensional chemical fingerprint matrix, the deviation of the biomarker group, and the deviation of the bioactivity are calculated respectively; a quality deviation correlation model is established by combining raw material source information, extraction process parameters, concentration process parameters, and glue preparation process parameters; the key process nodes that cause quality anomalies are identified through the correlation model; and corresponding process correction suggestions are output to achieve closed-loop quality management of the tortoise shell glue production process.

[0014] Preferably, the method is applied to the entire process of quality control of tortoise shell glue raw materials, intermediates, semi-finished products and finished products; the established two-dimensional chemical fingerprint database, spectrum-effect correlation database, characteristic marker database and bioactivity database are interconnected to form a drug efficacy-oriented quality control platform; the platform can realize batch quality evaluation, process stability evaluation, drug efficacy consistency evaluation and quality risk early warning; thereby establishing a drug quality control system for tortoise shell glue in the treatment of heart failure based on the hierarchical correlation of "chemical composition - drug efficacy marker - cell function - mitochondrial function".

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention establishes a dual-channel evaluation system of hydrophilic and hydrophobic fingerprint spectra, and combines anti-heart failure drug efficacy data to screen a group of drug efficacy quality markers. This solves the problem that existing tortoise shell glue quality control only detects amino acids or single chemical indicators, which cannot reflect the true anti-heart failure drug efficacy. It realizes the transformation from "chemical component detection" to "drug efficacy-related substance identification", improves the accuracy of quality evaluation, and ensures the consistency of drug efficacy of different batches of products. 2. This invention solves the problem that traditional quality control cannot verify the actual biological activity of drugs by constructing a three-level bioactivity evaluation system of cell viability detection, reactive oxygen species detection and mitochondrial membrane potential detection. It can quickly evaluate the protective ability of tortoise shell glue on damaged cardiomyocytes, realize the authenticity verification of drug efficacy, and directly link the quality control results with the anti-heart failure effect. 3. This invention establishes a quality control module at the organelle level by using the mitochondrial membrane potential recovery index as the core evaluation indicator. This solves the problem that existing quality control only stays at the chemical level and cannot reflect the ability to improve myocardial energy metabolism. It enables quality evaluation to penetrate to the mitochondrial functional level, improves the ability to identify the activity of key functional peptides, and can detect the problem of decreased drug efficacy caused by process damage in advance. 4. This invention establishes a comprehensive quality index model by integrating five dimensions: chemical fingerprint, quality markers, cell viability, oxidative stress, and mitochondrial function. This model solves the problems of scattered quality control indicators, inconsistent evaluation results, and inability to trace the source of quality abnormalities in traditional quality control. It achieves unified management of raw material quality evaluation, process stability evaluation, efficacy consistency evaluation, and abnormal quality traceability, and establishes a closed-loop control system of "raw materials - process - efficacy - quality". This improves the quality stability and industrial production level of tortoise shell glue for the treatment of heart failure. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the control method of the present invention. Detailed Implementation

[0017] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figure 1 : Example 1 This embodiment provides a quality control method for preparing tortoise shell glue as a drug for treating heart failure. With efficacy consistency evaluation as the core, it constructs a four-level linked quality control system: "chemical fingerprinting—quantitative analysis of characteristic biomarkers—rapid detection of bioactivity—comprehensive quality evaluation," realizing full-process quality evaluation of tortoise shell glue for treating heart failure from raw materials, intermediates to finished products. Traditional Chinese medicine quality markers (Q-Markers) and spectrum-efficacy correlation analysis are widely used to establish efficacy-related quality evaluation systems. This embodiment further couples efficacy markers with cell function evaluation to achieve efficacy-oriented quality control.

[0019] Specifically, 30 batches of tortoise shell glue samples from continuous production were selected as modeling samples, with 10g of each batch taken, pulverized, and passed through an 80-mesh sieve for later use. First, a two-dimensional chemical fingerprint database was established. The first dimension, the hydrophilic fingerprint, was obtained by purified water extraction. 1.000g of tortoise shell glue sample was weighed and placed in a 50mL centrifuge tube, 20mL of purified water was added, and the mixture was extracted at 60℃ with shaking for 45min, followed by centrifugation at 10000rpm for 10min. The supernatant was filtered through a 0.22μm filter membrane and then analyzed by high-performance liquid chromatography (HPLC). A C18 column (250mm × 4.6mm, 5μm) was used. Mobile phase A was 0.1% formic acid aqueous solution, mobile phase B was acetonitrile, gradient elution time was 60min, and the detection wavelength was 210nm. The first fingerprint, reflecting the distribution characteristics of collagen degradation peptides, oligopeptides, and free amino acids, was obtained. The second-dimensional hydrophobic fingerprint was obtained by extraction with 70% methanol. 1.000 g of tortoise shell glue sample was added to 20 mL of 70% methanol, and the mixture was ultrasonically extracted for 30 min. After filtration, ultra-high performance liquid chromatography (UHPLC) was used for detection at a wavelength of 280 nm to obtain the second fingerprint spectrum reflecting hydrophobic functional peptides, lipophilic derivatives, and small organic molecules. Retention time correction and dynamic time warping algorithms were used to align the chromatographic peaks of 30 batches of samples, establishing a two-dimensional chemical fingerprint database. Finally, 52 stable common peaks were obtained, including 31 hydrophilic common peaks and 21 hydrophobic common peaks.

[0020] Subsequently, a spectrum-effect correlation analysis model was established. Thirty batches of tortoise shell glue samples were extracted and prepared into sample solutions, which were then used to treat an H9c2 rat cardiomyocyte injury model. H9c2 cells were cultured in DMEM medium containing 10% fetal bovine serum at 37℃ and 5% CO2 until the logarithmic growth phase, after which they were seeded into 96-well plates at a density of 1×10⁶ cells per well. 4 Cells were cultured for 4 hours after inducing oxidative damage using 200 μmol / L hydrogen peroxide. The cells were then cultured for another 24 hours after adding tortoise shell glue extract. Cell viability recovery rate, lactate dehydrogenase release inhibition rate, reactive oxygen species (ROS) decrease rate, mitochondrial membrane potential recovery rate, and intracellular calcium ion homeostasis recovery rate were measured to construct a pharmacodynamic evaluation matrix Y. Simultaneously, the peak areas of 52 common peaks were used to construct a chemical matrix X. Partial least squares regression analysis, random forest analysis, and grey relational analysis were used to jointly model and calculate the importance projection value (VIP) and pharmacodynamic contribution coefficient of each characteristic peak. Based on the screening criteria of VIP > 1.5, contribution > 0.65, frequency of occurrence > 95%, and inter-batch coefficient of variation < 15%, eight core biomarker peaks significantly associated with anti-heart failure activity were finally obtained, and a characteristic biomarker database was established.

[0021] Furthermore, the structures of the eight core biomarkers selected were analyzed. Identification was performed using liquid chromatography-high resolution mass spectrometry, yielding two hydroxyproline-enriched peptides, two glycine-enriched peptides, two proline-enriched peptides, and two hydrophobic functional peptides. An external standard method was used to establish a quantitative detection model, and a biomarker comprehensive evaluation index (MQI) was constructed. The calculation formula is: MQI = Σ(C i ×W i ) / ΣW i Where Ci is the actual content of the i-th marker, and W i The corresponding contribution weight to drug efficacy is determined. Based on the modeling results, a biomarker is considered qualified when the MQI standard value is not lower than 0.85.

[0022] After completing the chemical evaluation, a three-level rapid biological activity evaluation system is established. The first-level evaluation adopts the CCK-8 method to detect the recovery ability of cell viability, measures the absorbance values of the normal group, the model group and the sample group, and calculates the cell viability recovery index CVI = (ODsample - ODmodel) / (ODnormal - ODmodel). When CVI ≥ 0.80, the first-level activity is judged as qualified. The second-level evaluation uses DCFH-DA fluorescent probe to detect the intracellular reactive oxygen species level, and calculates the reactive oxygen species inhibition index ROI = (Fmodel - Fsample) / (Fmodel - Fnormal), wherein F represents the fluorescence intensity value, and when ROI ≥ 0.40, the second-level activity is judged as qualified. The third-level evaluation is carried out by using JC-1 mitochondrial membrane potential detection kit. The red fluorescence intensity R and green fluorescence intensity G are measured respectively, the ratio of red to green fluorescence is calculated, and the mitochondrial membrane potential recovery index MPRI = [(R / G)sample - (R / G)model] / [(R / G)normal - (R / G)model] is established. When MPRI ≥ 0.75, the third-level activity is judged as qualified. Based on the bioactive quality marker strategy, cell function and mitochondrial function are incorporated into the quality evaluation indicators, so that the quality control results can directly reflect the drug efficacy level.

[0023] After completing the chemical evaluation and biological activity evaluation, a five-dimensional comprehensive quality evaluation model is constructed. The chemical fingerprint similarity index F, the marker content index M, the cell viability recovery index V, the reactive oxygen species inhibition index R and the mitochondrial membrane potential recovery index P are obtained respectively. The entropy weight method is used to calculate the weight of each evaluation index, and the comprehensive quality index QCI is constructed. The calculation formula is QCI = W1F + W2M + W3V + W4R + W5P, wherein W1 to W5 are the weight coefficients of each evaluation dimension respectively. The optimal weight combination obtained through training with 30 batches of modeling samples is 0.22, 0.28, 0.18, 0.14 and 0.18. A quality grade standard is established based on the statistical results: QCI ≥ 85 is judged as high-quality batch; 70 ≤ QCI < 85 is judged as qualified batch; QCI < 70 is judged as unqualified batch.

[0024] When the QCI of the test sample is lower than 70, the system automatically starts the quality abnormality traceability module, calculates the deviation degree of chemical fingerprint, the deviation degree of markers and the deviation degree of biological activity respectively, performs matching analysis with the raw material batch database, extraction temperature database, concentration process database and glue preparation process database, and identifies the source of abnormality through the Bayesian quality correlation model. When the chemical fingerprint is normal but the biological activity decreases significantly, it is judged that there may be active peptide degradation caused by high-temperature concentration; when the content of markers decreases and the fingerprint deviation increases, it is judged that there may be raw material quality fluctuation; when the mitochondrial membrane potential recovery index decreases significantly, it is judged that there may be a problem of key functional peptide inactivation, thereby realizing rapid positioning of quality problems and process optimization.

[0025] Verification has shown that the quality control system established in this embodiment can simultaneously achieve chemical consistency evaluation, efficacy consistency evaluation, and process stability evaluation. Compared with the traditional quality control mode that only uses total nitrogen, amino acid content, or a single fingerprint spectrum, it can more accurately reflect the true quality level of tortoise shell glue for treating heart failure, and improve batch stability and consistency of clinical efficacy.

[0026] Example 2 This embodiment verifies the applicability of the quality control method of the present invention to batches of tortoise shell glue produced using different processes. Tortoise shell raw materials from the same origin and batch were selected and produced using three different glue-making processes: Group A involved extraction at 95℃ for 10 hours followed by vacuum concentration at 65℃; Group B involved extraction at 105℃ for 12 hours followed by vacuum concentration at 75℃; and Group C involved extraction at 115℃ for 14 hours followed by vacuum concentration at 85℃. Ten batches of samples were produced continuously for each group. A two-dimensional chemical fingerprint database was established according to the method in Example 1. The results showed that the hydrophilic fingerprint similarity of Group A and Group B was 0.982 and 0.968, respectively, and the hydrophobic fingerprint similarity was 0.973 and 0.951, respectively, while the hydrophobic fingerprint similarity of Group C decreased to 0.867. Further detection of the content of eight anti-heart failure biomarkers revealed that the average MQI value of Group A was 0.93, the average MQI value of Group B was 0.88, and the average MQI value of Group C decreased to 0.69. Subsequently, an H9c2 cell oxidative damage model was established for bioactivity evaluation. In group A, the cell viability recovery index (CVI) was 0.89±0.03, the reactive oxygen species inhibition index (ROI) was 0.62±0.04, and the mitochondrial membrane potential recovery index (MPRI) was 0.87±0.05; in group B, these values ​​were 0.84±0.04, 0.57±0.05, and 0.79±0.04, respectively; and in group C, they decreased to 0.63±0.05, 0.38±0.06, and 0.54±0.07, respectively. A five-dimensional quality assessment model was used to calculate the quality index (QCI). The average QCI for group A was 91.4, for group B it was 84.6, and for group C it was 63.8. The results indicate that the high-temperature, long-duration gelation process leads to the degradation of some anti-heart failure functional peptides. The quality control method established in this invention can effectively identify the efficacy loss caused by process changes and achieve stability evaluation of the production process.

[0027] Example 3 This embodiment is used to verify the quality anomaly tracing function. A batch of tortoise shell glue samples exhibiting fluctuations in clinical efficacy during production was selected for analysis. First, a two-dimensional chemical fingerprint was established, showing a fingerprint similarity of 0.951, within the acceptable range. Further detection of biomarker content revealed an MQI of only 0.71. Cell activity evaluation was then performed, with CVI at 0.72, ROI at 0.41, and MPRI at 0.56. The system automatically activated the quality anomaly tracing module to analyze the deviations of the chemical fingerprint, biomarker, and biological activity. The results showed a chemical fingerprint deviation of only 5.2%, while the biomarker deviation reached 26.8%, and the biological activity deviation reached 31.4%. Further investigation of the production database revealed that the actual temperature during the concentration process of this batch reached 92℃, higher than the standard process setting of 75℃. By establishing a correlation model between process parameters and quality indicators, it was found that after the concentration temperature exceeded 85℃, hydroxyproline-enriched peptides and hydrophobic functional peptides underwent significant degradation, leading to a rapid decrease in MPRI. After process correction and re-production, the QCI of the resulting sample recovered to 88.7. The results demonstrate that this invention can achieve rapid location of quality anomalies and trace the source of process problems.

[0028] Example 4 This embodiment was used to verify the effectiveness of the five-dimensional comprehensive quality assessment model. One hundred batches of tortoise shell glue samples produced at different times were randomly selected, and the chemical fingerprint similarity index (F), biomarker content index (M), cell viability recovery index (V), reactive oxygen species inhibition index (R), and mitochondrial membrane potential recovery index (P) were measured. The entropy weight method was used to calculate the weights of each index, and a comprehensive QCI evaluation model was constructed. Statistical results showed that when QCI ≥ 85, the cardiac function improvement rate in the corresponding animal experiments reached over 90%; when 70 ≤ QCI < 85, the cardiac function improvement rate remained between 70% and 90%; and when QCI < 70, the cardiac function improvement rate was below 70%. Further validation was performed using receiver operating characteristic (ROC) curves, and the area under the curve (AUC) for predicting the efficacy grade using the QCI evaluation model reached 0.927. Compared to using total nitrogen alone, its prediction accuracy improved by 32.6%; compared to using fingerprint alone, its prediction accuracy improved by 24.3%; and compared to using biomarker content alone, its prediction accuracy improved by 18.7%. The results show that the five-dimensional comprehensive quality evaluation system established in this invention can more accurately reflect the actual efficacy level of tortoise shell glue in treating heart failure, achieving a quality control upgrade from "component consistency evaluation" to "efficacy consistency evaluation." The development trend of quality control in traditional Chinese medicine emphasizes a shift from simple chemical evaluation to efficacy-related evaluation and whole-process quality control. The results of this embodiment further verify the effectiveness of this technical approach.

[0029] Working principle: This invention first establishes hydrophilic and hydrophobic fingerprint profiles based on the distribution characteristics of hydrophilic functional peptides, collagen-degrading peptides, amino acid components, and hydrophobic active peptides in tortoise shell glue, forming a two-dimensional chemical fingerprint database covering the main active components of tortoise shell glue. Subsequently, using anti-heart failure efficacy data from different batches of samples and the two-dimensional fingerprint data, a spectral-efficacy correlation analysis is performed. Partial least squares regression, random forest, and grey relational analysis are used to establish a mapping relationship between chemical composition and efficacy, thereby screening out a group of quality biomarkers directly related to anti-heart failure effects and achieving precise identification of the material basis related to efficacy. This process moves quality evaluation beyond the level of chemical composition, establishing a quantitative correlation between chemical components and their contribution to efficacy.

[0030] After obtaining the efficacy and quality biomarkers, this invention further constructs an H9c2 cardiomyocyte injury model and establishes a three-tiered rapid bioactivity detection system through cell viability recovery evaluation, reactive oxygen species (ROS) inhibition evaluation, and mitochondrial membrane potential recovery evaluation. Cell viability evaluation reflects the protective effect of the sample on the survival of damaged cardiomyocytes; ROS evaluation reflects the sample's ability to inhibit oxidative stress damage; and mitochondrial membrane potential evaluation directly reflects the sample's ability to improve cardiomyocyte energy metabolism and mitochondrial function. This three-tiered biological validation mechanism achieves efficacy verification from the molecular level to the organelle level, thereby avoiding quality misjudgments caused by relying solely on chemical indicators.

[0031] This invention quantifies chemical fingerprint similarity, quality marker content, cell viability recovery ability, oxidative stress inhibition ability, and mitochondrial function recovery ability as independent evaluation indicators, and establishes a five-dimensional comprehensive quality evaluation model using the entropy weight method. The system automatically calculates the comprehensive quality index (QCI) based on the weights of each indicator and outputs the quality grade result in conjunction with a database standard model. When quality anomalies occur, the system automatically locates key influencing factors such as raw material fluctuations, extraction process abnormalities, or concentration and gelation process abnormalities by reverse analysis of the deviation degree of each dimension, achieving closed-loop control from quality testing to quality traceability. This principle realizes a full-chain correlation evaluation between raw materials, processes, products, and efficacy.

[0032] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A quality control method for preparing drugs for treating heart failure using tortoise shell glue, characterized in that: The method includes the following steps: Obtain the tortoise shell glue sample to be tested; Parallel extraction was performed using polar and non-polar solvent systems respectively, and a first chemical fingerprint spectrum reflecting the hydrophilic active component and a second chemical fingerprint spectrum reflecting the hydrophobic active component were established. The anti-heart failure activity of different batches of samples was determined and a pharmacodynamic database was constructed. A spectrum-effect correlation model was established based on the correlation between chemical fingerprint characteristic peaks and pharmacodynamic parameters; Based on the contribution of spectrum-effect correlation, a group of characteristic biomarkers for anti-heart failure was screened and a quantitative evaluation model was established; The ability of samples to regulate cell viability, oxidative stress level, and mitochondrial membrane potential was detected using a cardiomyocyte injury model. The comprehensive quality index is obtained by integrating the chemical fingerprint evaluation results, the characteristic biomarker evaluation results, and the bioactivity evaluation results. The quality of raw materials, production process, and finished product in the preparation of heart failure medication using tortoise shell glue was assessed based on a comprehensive quality index.

2. The quality control method for preparing a drug for treating heart failure using tortoise shell glue according to claim 1, characterized in that: The first chemical fingerprint was established using an aqueous extract at a controlled temperature of 40–80°C to obtain chromatographic characteristics of oligopeptide components, collagen degradation product components, and water-soluble amino acid components. The second chemical fingerprint was established using an alcohol solvent extract with a volume fraction of 50–90% to obtain chromatographic characteristics of hydrophobic peptide components, lipid-soluble derivative components, and small organic molecule components. A unified feature peak library was established using a retention time correction algorithm. A two-dimensional chemical fingerprint matrix was obtained through peak matching, peak area normalization, and batch-to-batch drift correction. The two-dimensional chemical fingerprint matrix simultaneously contains peak position information, peak area information, and peak ratio information to improve the stability and repeatability of chemical composition evaluation of different batches of tortoise shell glue samples.

3. The quality control method for preparing a drug for treating heart failure using tortoise shell glue according to claim 1, characterized in that: The pharmacodynamic database was obtained by establishing a heart failure-related evaluation system, including at least three of the following: cell survival rate recovery rate, lactate dehydrogenase release inhibition rate, reactive oxygen species decline rate, mitochondrial membrane potential recovery rate, and calcium ion homeostasis recovery rate. A pharmacodynamic matrix was constructed from the pharmacodynamic parameters corresponding to each batch of samples, and a chemical matrix was constructed from the two-dimensional chemical fingerprint matrix. A spectrum-effect correlation model was established using at least one of partial least squares regression analysis, grey relational analysis, and random forest analysis. The contribution weight and importance score corresponding to each characteristic peak were calculated, and candidate characteristic peaks significantly related to the anti-heart failure pharmacodynamics were screened according to a preset threshold.

4. The quality control method for preparing a drug for treating heart failure using tortoise shell glue according to claim 1, characterized in that: The group of biomarkers for treating heart failure was obtained by: statistically analyzing the frequency, stability coefficient, and efficacy contribution of each candidate peak in different batches of samples; identifying candidate peaks with a frequency greater than 90%, a contribution weight greater than a preset threshold, and an inter-batch coefficient of variation less than 15% as core biomarkers; performing structural analysis on the core biomarkers and establishing a multi-component quantitative detection model; the group of biomarkers includes one or more of collagen-derived functional peptides, hydroxyproline-derived peptides, glycine-enriched peptides, proline-enriched peptides, and hydrophobic functional peptides; and using the multi-component quantitative detection model to obtain the content evaluation results of the biomarker group.

5. The quality control method for preparing a drug for treating heart failure using tortoise shell glue according to claim 1, characterized in that: The cell viability evaluation was performed by constructing an oxidative damage model or a drug-induced damage model using H9c2 cardiomyocytes; the cell viability was measured after the sample was applied to the damaged model cells; and the cell viability recovery index (CVI) was calculated. Wherein, CVI is the recovery ratio of cell viability in the sample group relative to that in the model group; when CVI reaches a preset standard, the sample is determined to have basic anti-heart failure activity; the evaluation result is used as a primary efficacy evaluation parameter in the comprehensive quality index; and is used to verify the consistency between the chemical fingerprint evaluation results and the actual biological effects.

6. The quality control method for preparing a drug for treating heart failure using tortoise shell glue according to claim 1, characterized in that: The oxidative stress evaluation uses a fluorescent probe detection system to measure intracellular reactive oxygen species (ROS) levels; establishes the fluorescence intensity differences among the sample group, model group, and normal group; calculates the ROS inhibition index (ROI); the ROI is used to characterize the ability of the tortoise shell glue sample to inhibit free radical accumulation under oxidative damage; and establishes an oxidative stress quality grade standard based on the ROI value; the oxidative stress evaluation results serve as a secondary efficacy evaluation parameter in the comprehensive quality index; and are also used to identify efficacy deviations caused by differences in raw materials or fluctuations in production processes.

7. The quality control method for preparing a drug for treating heart failure using tortoise shell glue according to claim 1, characterized in that: The mitochondrial function evaluation was performed using a mitochondrial membrane potential detection reagent system. The ratio of red to green fluorescence intensity was obtained for the sample group, normal group, and model group. The mitochondrial membrane potential recovery index (MPRI) was constructed. The MPRI was calculated based on the recovery degree of the red-green fluorescence ratio of the sample group relative to the normal and model groups. The MPRI was used as the core indicator for evaluating the improvement of cardiomyocyte mitochondrial function by tortoise shell glue. A mitochondrial function quality grade standard was established based on the MPRI. The evaluation results were used as the third-level efficacy evaluation parameter in the comprehensive quality index.

8. The quality control method for preparing a drug for treating heart failure using tortoise shell glue according to claim 1, characterized in that: The comprehensive quality index is obtained by integrating the chemical fingerprint similarity index, the characteristic biomarker content index, the cell viability recovery index, the reactive oxygen species inhibition index, and the mitochondrial membrane potential recovery index. The weights of each evaluation index are determined by the entropy weight method, the adaptive weight method, or the machine learning weight optimization method. A five-dimensional quality evaluation model is constructed. The model can output a quality score of 0 to 100. Based on the score, the samples are divided into excellent, qualified, and unqualified grades, and a corresponding quality evaluation report is generated.

9. The quality control method for preparing a drug for treating heart failure using tortoise shell glue according to claim 1, characterized in that: The method further includes a quality anomaly tracing step; when the comprehensive quality index is lower than a preset threshold, the deviation of the two-dimensional chemical fingerprint matrix, the deviation of the biomarker group, and the deviation of the bioactivity are calculated respectively; a quality deviation correlation model is established by combining raw material source information, extraction process parameters, concentration process parameters, and gelation process parameters; and the key process nodes that lead to quality anomalies are identified through the correlation model. It also outputs corresponding process modification suggestions to achieve closed-loop quality management of the tortoise shell glue production process.

10. The quality control method for preparing a drug for treating heart failure using tortoise shell glue according to claim 1, characterized in that: The method is applied to the entire process of quality control of tortoise shell glue raw materials, intermediates, semi-finished products and finished products; the established two-dimensional chemical fingerprint database, spectrum-effect correlation database, characteristic marker database and bioactivity database are interconnected to form a drug efficacy-oriented quality control platform; the platform can realize batch quality evaluation, process stability evaluation, drug efficacy consistency evaluation and quality risk early warning; thereby establishing a drug quality control system for tortoise shell glue for the treatment of heart failure based on the hierarchical correlation of "chemical composition - drug efficacy marker - cell function - mitochondrial function".

Images