A quality detection method for traditional Chinese medicine formula granules

Abstract

This invention provides a method for quality testing of traditional Chinese medicine (TCM) formula granules. The method includes: obtaining the TCM formula granules to be tested and dividing them into three equal parts to obtain a first sample, a second sample, and a third sample; performing acid hydrolysis on the first sample and pre-column derivatization on the acid-hydrolyzed sample to obtain a test sample; obtaining a first test result through quantitative detection; separating crude polysaccharides from the second sample and a preset reference standard to obtain a solid test sample and a solid reference standard; obtaining a second test result through molecular weight characteristic distribution detection; pre-treating the third sample to obtain a pre-treated sample; performing variety detection based on the Raman spectral data of the pre-treated sample and a trained variety prediction model to obtain a third test result; and performing quality testing on the TCM formula granules to be tested to obtain a quality test result. This invention can improve the accuracy of quality testing of TCM formula granules.

Description

Technical Field

[0001] This invention relates to the field of pharmaceutical technology, and also to a method for quality testing of traditional Chinese medicine formula granules. Background Technology

[0002] Traditional Chinese medicine (TCM) granules have become a mainstream dosage form in clinical practice due to their convenience. Polygonatum cyrtonema Hua, a classic "food and medicine homology" herb, is widely used in granule formulations. However, its quality control faces the following challenges: granulation processes can damage the original medicinal material's properties and microscopic characteristics, rendering traditional methods of identification inapplicable; to improve formulation formability, large amounts of maltodextrin (up to 50% or more) are often added to the granules. Maltodextrin and Polygonatum cyrtonema polysaccharides (mainly fructans) have highly similar physicochemical properties, making traditional phenol... Total sugar content determination methods such as the sulfuric acid method cannot distinguish between the two, and the test results become artificially high as the amount of excipients added increases, resulting in a large error in the quality test results. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method for quality testing of traditional Chinese medicine formula granules, so as to improve the accuracy of quality testing of traditional Chinese medicine formula granules.

[0004] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0005] This invention provides a method for quality testing of traditional Chinese medicine formula granules, comprising:

[0006] Obtain the Chinese herbal formula granules to be tested and divide them into three equal parts to obtain the first sample, the second sample, and the third sample;

[0007] The first sample was subjected to acid hydrolysis treatment according to the preset acid hydrolysis conditions to obtain an acid hydrolyzed sample;

[0008] The acid-hydrolyzed sample was subjected to pre-column derivatization to obtain the sample to be tested;

[0009] The sample to be tested is quantitatively analyzed to obtain the first test result;

[0010] The second sample and the preset reference standard were subjected to crude polysaccharide separation treatment to obtain the solid sample to be tested and the solid reference standard.

[0011] The molecular weight characteristic distribution of the solid sample to be tested and the solid reference standard are analyzed to obtain the second test result;

[0012] The third sample was pretreated to obtain a pretreated sample;

[0013] Based on the Raman spectral data of the pretreated sample and the trained variety prediction model, variety detection is performed to obtain a third detection result; the variety prediction model is obtained by training a preset network model based on the Raman spectral data of different varieties of traditional Chinese medicine formula granules.

[0014] The quality of the Chinese herbal medicine granules to be tested is determined based on the first test result, the second test result, and the third test result.

[0015] Optionally, the first sample is subjected to acid hydrolysis treatment according to preset acid hydrolysis conditions to obtain an acid-hydrolyzed sample, including:

[0016] The first sample was subjected to crude polysaccharide separation treatment to obtain crude polysaccharide;

[0017] The crude polysaccharide was subjected to acid hydrolysis under preset acid hydrolysis conditions to obtain an acid hydrolyzed sample. The preset acid hydrolysis conditions included: the acid hydrolysis reagent was trifluoroacetic acid with a concentration of 0.2 to 0.8 mol / L, the acid hydrolysis temperature was 100 to 115°C, and the acid hydrolysis time was 0.5 to 1.0 h.

[0018] Optionally, the first sample is subjected to crude polysaccharide separation treatment to obtain crude polysaccharide, including:

[0019] Take the first sample, add 10 times the amount of water at 90 to 100°C, stir to dissolve, centrifuge at 8000 r / min for 10 minutes to obtain the supernatant;

[0020] Anhydrous ethanol was added to the supernatant while stirring until the final alcohol concentration was 80%. After standing, the mixture was centrifuged at 8000 r / min for 10 minutes to obtain a precipitate.

[0021] The precipitate was washed with anhydrous ethanol and then freeze-dried under vacuum to obtain crude polysaccharide; wherein the temperature during vacuum freeze-drying was -50°C to -60°C, the pressure was less than 20 Pa, and the drying time was 24 to 48 hours.

[0022] Optionally, the acid-hydrolyzed sample is subjected to pre-column derivatization to obtain the sample to be tested, including:

[0023] Obtaining derivatization reagents;

[0024] According to the preset pre-column derivatization conditions and the derivatization reagent, the acid-hydrolyzed sample is subjected to pre-column derivatization to obtain the sample to be tested.

[0025] Optionally, the sample to be tested is quantitatively detected to obtain a first detection result, including:

[0026] Obtain preset chromatographic conditions; the preset chromatographic conditions include: the chromatographic column is a hydrophilic chromatographic column, the mobile phase is a buffer-buffer-methanol mixture, the chromatographic separation mode is gradient elution, the ultraviolet detection wavelength is 303nm, the excitation wavelength is 313nm, and the emission wavelength is 358nm;

[0027] The sample to be tested is subjected to chromatographic separation and fluorescence detection according to the preset chromatographic conditions to obtain a chromatogram;

[0028] The sample to be tested is quantitatively detected based on the chromatogram to obtain the first detection result.

[0029] Optionally, the second sample and the preset reference standard are subjected to crude polysaccharide separation processing to obtain the solid sample to be tested and the solid reference standard, including:

[0030] Obtain a preset reference standard; the preset reference standard includes a blank reference standard sample and an excipient reference sample; the blank reference standard sample is consistent with the medicinal material source of the Chinese herbal medicine formula granules to be tested;

[0031] The second sample, the blank reference control sample and the excipient control sample in the preset reference standard are subjected to crude polysaccharide separation processing to obtain the solid sample to be tested and the solid reference standard.

[0032] Optionally, a second detection result is obtained by performing molecular weight characteristic distribution detection on the solid sample to be tested and the solid reference standard, including:

[0033] The solid sample to be tested and the solid reference standard were subjected to chromatographic detection to obtain chromatographic data;

[0034] Based on the chromatographic data and the preset standard curve, the molecular weight characteristic distribution is detected to obtain the second detection result.

[0035] Optionally, the third sample is pretreated to obtain a pretreated sample, including:

[0036] The third sample was ground and dissolved to obtain a sample solution;

[0037] The sample solution was centrifuged to obtain crude polysaccharide powder;

[0038] The crude polysaccharide powder was pressed to obtain a pretreated sample.

[0039] Optionally, variety detection is performed based on the Raman spectral data of the pretreated sample and the trained variety prediction model to obtain a third detection result, including:

[0040] Raman spectral data of the pretreated sample were acquired according to preset data acquisition parameters and preset data acquisition method to obtain Raman spectral data.

[0041] The Raman spectral data are preprocessed to obtain preprocessed data;

[0042] The preprocessed data is input into the trained variety prediction model for variety detection, and a third detection result is obtained.

[0043] Optionally, the quality of the Chinese herbal medicine granules to be tested is performed based on the first test result, the second test result, and the third test result to obtain quality test results, including:

[0044] Obtain the preset product qualification conditions;

[0045] Quality inspection is performed based on the preset product qualification conditions, the first test result, the second test result, and the third test result to obtain the quality inspection result.

[0046] The above-described solution of the present invention has at least the following beneficial effects:

[0047] The above-described scheme of the present invention involves dividing the Chinese herbal medicine formula granules to be tested into three equal parts to obtain a first sample, a second sample, and a third sample. The first sample is then acid-hydrolyzed under preset acid hydrolysis conditions to obtain an acid-hydrolyzed sample. This acid-hydrolyzed sample undergoes pre-column derivatization to obtain a test sample. Quantitative detection is performed on the test sample to obtain a first detection result. The second sample and a preset reference standard are subjected to crude polysaccharide separation to obtain a test solid sample and a solid reference standard. Molecular weight characteristic distribution is detected based on the test solid sample and the solid reference standard to obtain a second detection result. The third sample is pre-treated to obtain a pre-treated sample. Variety detection is performed based on the Raman spectral data of the pre-treated sample and a trained variety prediction model to obtain a third detection result. Finally, the first, second, and third detection results are combined to perform quality testing on the Chinese herbal medicine formula granules to obtain a quality testing result. Through the synergy of these three detection methods, a complete chain of evidence is constructed, confirming the presence of components, structural integrity, and variety authenticity, thereby improving the accuracy, comprehensiveness, and ease of operation of the quality testing of Chinese herbal medicine formula granules. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating the method for quality testing of traditional Chinese medicine formula granules in an embodiment of the present invention;

[0049] Figure 2 This is a chromatogram of maltodextrin, crude polysaccharide of Polygonatum sibiricum, and crude polysaccharide of Polygonatum sibiricum formulation granules in the embodiments of the present invention;

[0050] Figure 3This is a chromatogram showing the molecular weight distribution of maltodextrin, crude polysaccharide from Polygonatum sibiricum, and crude polysaccharide from Polygonatum sibiricum granules in the embodiments of the present invention.

[0051] Figure 4 This is the direct compression spectrum in this embodiment;

[0052] Figure 5 This is the SERS spectrum of the alcohol-soluble components in this embodiment;

[0053] Figure 6 This is the crude polysaccharide spectrum in this embodiment;

[0054] Figure 7 This is the maltodextrin spectrum in this embodiment. Detailed Implementation

[0055] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0056] like Figure 1 As shown, an embodiment of the present invention proposes a method for quality testing of traditional Chinese medicine formula granules, including the following steps:

[0057] Step 101: Obtain the Chinese herbal formula granules to be tested and divide them into three equal parts to obtain the first sample, the second sample, and the third sample;

[0058] Step 102: Perform acid hydrolysis on the first sample according to the preset acid hydrolysis conditions to obtain an acid hydrolyzed sample;

[0059] Step 103: Perform pre-column derivatization on the acid-hydrolyzed sample to obtain the sample to be tested;

[0060] Step 104: Perform quantitative detection on the sample to be tested to obtain the first detection result;

[0061] Step 105: Perform crude polysaccharide separation on the second sample and the preset reference standard to obtain the solid sample to be tested and the solid reference standard;

[0062] Step 106: Perform molecular weight characteristic distribution detection based on the solid sample to be tested and the solid reference standard to obtain the second detection result;

[0063] Step 107: Pre-treat the third sample to obtain a pre-treated sample;

[0064] Step 108: Perform variety detection based on the Raman spectral data of the preprocessed sample and the trained variety prediction model to obtain a third detection result; the variety prediction model is obtained by training a preset network model based on the Raman spectral data of different varieties of traditional Chinese medicine formula granules.

[0065] Step 109: Perform quality testing on the Chinese herbal medicine formula granules to be tested based on the first test result, the second test result, and the third test result to obtain the quality test result.

[0066] The method for quality detection of traditional Chinese medicine (TCM) formula granules in this invention involves dividing the TCM formula granules to be tested into three equal parts to obtain a first sample, a second sample, and a third sample. The first sample is then acid-hydrolyzed under preset acid hydrolysis conditions to obtain an acid-hydrolyzed sample. This acid-hydrolyzed sample undergoes pre-column derivatization to obtain a test sample. Quantitative detection is performed on the test sample to obtain a first detection result. The second sample and a preset reference standard are subjected to crude polysaccharide separation to obtain a test solid sample and a solid reference standard. Molecular weight characteristic distribution is detected based on the test solid sample and the solid reference standard to obtain a second detection result. The third sample is pre-treated to obtain a pre-treated sample. Variety detection is performed based on the Raman spectral data of the pre-treated sample and a trained variety prediction model to obtain a third detection result. Finally, the quality of the TCM formula granules to be tested is determined by combining the first, second, and third detection results to obtain a quality detection result. Through the synergy of these three detection methods, a complete chain of evidence is constructed, confirming the presence of components, structural integrity, and variety authenticity, thus improving the accuracy, comprehensiveness, and ease of operation of TCM formula granule quality detection.

[0067] In an optional embodiment of the present invention, step 101 involves obtaining the Chinese herbal medicine formula granules to be tested and dividing them into three equal parts to obtain a first sample, a second sample, and a third sample.

[0068] Specifically, the same batch of Chinese herbal formula granules to be tested is obtained and divided into three equal parts to obtain a first sample, a second sample, and a third sample. This facilitates subsequent testing of the first, second, and third samples using different testing methods, improving the comprehensiveness of the quality testing of the Chinese herbal formula granules. In this embodiment, the Chinese herbal formula granules to be tested can be Polygonatum sibiricum formula granules. In specific implementation, the detection method of the present invention is particularly suitable for products with polysaccharides as the main active ingredient and maltodextrin added during the production process. , Specific qualitative and quantitative analysis of traditional Chinese medicine formula granules containing dextran excipients, evaluation of excipient interference, rapid identification of product authenticity, and quality testing.

[0069] In an optional embodiment of the present invention, step 102, which involves acid-hydrolyzing the first sample according to preset acid hydrolysis conditions to obtain an acid-hydrolyzed sample, may include:

[0070] Step 1021: Perform crude polysaccharide separation treatment on the first sample to obtain crude polysaccharide;

[0071] In an optional embodiment of the present invention, step 1021 includes:

[0072] Step 10211: Take the first sample, add 10 times the amount of water at 90 to 100°C, stir to dissolve, centrifuge at 8000 r / min for 10 minutes to obtain the supernatant;

[0073] Step 10212: Add anhydrous ethanol to the supernatant while stirring until the final alcohol concentration is 80%. After standing, centrifuge at 8000 r / min for 10 minutes to obtain the precipitate.

[0074] Step 10213: The precipitate is washed with anhydrous ethanol and then freeze-dried under vacuum to obtain crude polysaccharide; wherein the temperature during vacuum freeze-drying is -50℃ to -60℃, the pressure is less than 20Pa, and the drying time is 24 to 48 hours.

[0075] Specifically, take the Polygonatum odoratum formula granules, add 10 times the amount of hot water (90 to 100℃), stir thoroughly to dissolve, centrifuge at 8000 r / min for 10 minutes, and collect the supernatant. While stirring, slowly add anhydrous ethanol to the supernatant until the final ethanol concentration is 80%, let stand overnight to allow the polysaccharides to precipitate fully. Centrifuge at 8000 r / min for 10 minutes and collect the precipitate. Wash the precipitate three times with anhydrous ethanol, and freeze-dry under vacuum (-50℃ to -60℃, pressure <20 Pa, drying for 24 to 48 hours) to obtain the crude polysaccharide from the Polygonatum odoratum formula granules.

[0076] Step 1022: The crude polysaccharide is subjected to acid hydrolysis treatment according to preset acid hydrolysis conditions to obtain an acid hydrolyzed sample; the preset acid hydrolysis conditions include: the acid hydrolysis reagent is trifluoroacetic acid with a concentration of 0.2 to 0.8 mol / L, the acid hydrolysis temperature is 100 to 115℃, and the acid hydrolysis time is 0.5 to 1.0 h.

[0077] Specifically, the crude polysaccharide was taken and acid hydrolyzed (i.e., acid hydrolysis) was performed by adding trifluoroacetic acid (TFA) solution. The preset acid hydrolysis conditions were determined through single-factor experiments and orthogonal experimental system optimization.

[0078] Single-factor experiment investigation:

[0079] Crude polysaccharide concentrations: 0.2, 0.5, 1.0, 2.0, and 3.0 mg / mL. The acid hydrolysis temperature was fixed at 110℃, the TFA concentration at 0.5 mol / L, and the hydrolysis time at 0.5 h. The results showed that the crude polysaccharide concentration in the range of 0.2 to 3.0 mg / mL had no significant effect on the fructose peak area, and the optimal crude polysaccharide concentration was 1.0 mg / mL.

[0080] TFA concentrations: 0.1, 0.2, 0.5, 0.8, and 1.0 mol / L. The crude polysaccharide concentration was fixed at 1.0 mg / mL, the acid hydrolysis temperature at 110℃, and the acid hydrolysis time at 0.5 h. The results showed that the fructose peak area first increased and then decreased with increasing TFA concentration, reaching its maximum at 0.5 mol / L. Excessive TFA concentration led to fructose degradation; therefore, the optimal trifluoroacetic acid concentration was 0.5 mol / L.

[0081] Acid hydrolysis temperatures: 90, 100, 105, 110, 115, and 120℃. The crude polysaccharide concentration was fixed at 1.0 mg / mL, the TFA concentration at 0.5 mol / L, and the acid hydrolysis time at 0.5 h. The results showed that the fructose peak area was largest at 110℃, and fructose degradation intensified above 110℃. Therefore, the optimal acid hydrolysis temperature was 110℃.

[0082] Acid hydrolysis times: 0.5, 0.75, 1.0, 1.25, 1.5, and 2.0 h. The crude polysaccharide concentration was fixed at 1.0 mg / mL, the TFA concentration at 0.5 mol / L, and the acid hydrolysis temperature at 110℃. The results showed that the fructose peak area was largest at 0.5 h; extending the hydrolysis time led to continuous fructose degradation. Therefore, the optimal acid hydrolysis time was 0.5 h.

[0083] Orthogonal experimental verification: using An orthogonal array was used, with fructose peak area as the evaluation index, to investigate three factors: TFA concentration (A: 0.2, 0.5, 0.8 mol / L), hydrolysis temperature (B: 100, 110, 115℃), and hydrolysis time (C: 0.5, 0.75, 1.0 h). Range analysis showed that the influence of each factor was: TFA concentration (R=0.594) > hydrolysis temperature (R=0.386) > hydrolysis time (R=0.233). Analysis of variance showed that the effects of TFA concentration and hydrolysis temperature were both highly significant (P<0.001). The optimal conditions were determined to be: TFA concentration 0.5 mol / L, hydrolysis temperature 110℃, hydrolysis time 0.5 h, and crude polysaccharide concentration 1.0 mg / mL. These conditions ensured complete hydrolysis of the polysaccharide while minimizing fructose degradation.

[0084] In an optional embodiment of the present invention, step 103, which involves pre-column derivatization of the acid-hydrolyzed sample to obtain the sample to be tested, may include:

[0085] Step 1031, Obtain the derivatization reagent;

[0086] Specifically, p-aminobenzoic acid (p-AMBA) was used as the derivatizing reagent. The derivatizing reagent was obtained through the following steps: 0.35 g of p-aminobenzoic acid, 0.05 g of sodium cyanoborohydride, and 0.5 g of acetic acid were accurately weighed, dissolved in 5 mL of methanol, and vortexed to obtain the derivatizing reagent.

[0087] Step 1032: Perform pre-column derivatization on the acid-hydrolyzed sample according to the preset pre-column derivatization conditions and the derivatization reagent to obtain the sample to be tested.

[0088] Specifically, take 50g of the acid-hydrolyzed sample solution. Add 100 The derivatizing reagent was mixed thoroughly and placed in a metal bath at 70°C for 40 minutes. After the reaction, it was cooled to room temperature, and water was added to bring the volume to 2 mL. The solution was then transferred to a 0.22 mL container. The sample is filtered through an organic microporous membrane and transferred to a vial to obtain the sample to be tested.

[0089] In an optional embodiment of the present invention, step 104, which involves quantitatively detecting the sample to be tested to obtain a first detection result, may include:

[0090] Step 1041: Obtain preset chromatographic conditions; the preset chromatographic conditions include: the chromatographic column is a hydrophilic chromatographic column, the mobile phase is a buffer-buffered methanol mixture, the chromatographic separation mode is gradient elution, the ultraviolet detection wavelength is 303 nm, the excitation wavelength is 313 nm, and the emission wavelength is 358 nm.

[0091] Specifically, the chromatographic column was a hydrophilic AQ-C18 (250mm × 4.6mm, 5... The mobile phase is specifically phosphate buffer (containing 20 mmol / L tetrabutylammonium hydrogen sulfate, pH 2.5) - (phosphate buffer - methanol 1:1), that is, A is 0.1 mol / L phosphate buffer (pH 2.5) containing 20 mmol / L tetrabutylammonium hydrogen sulfate; B is mobile phase A and methanol mixed at a volume ratio of 1:1. The gradient elution program is as follows: For time (min) 0 to 6, the B-phase volume fraction (%) is 5 to 6.3; for time (min) 6 to 26, the B-phase volume fraction (%) is 6.3 to 6.8; for time (min) 26 to 34, the B-phase volume fraction (%) is 6.8 to 7.5; for time (min) 34 to 45, the B-phase volume fraction (%) is 7.5 to 8.3; for time (min) 45 to 55, the B-phase volume fraction (%) is 8.3 to 50; for time (min) 55 to 60, the B-phase volume fraction (%) is 50 (hold); for time (min) 60 to 70, the B-phase volume fraction (%) is 50 to 5. The preset chromatographic conditions can also include: a flow rate of 0.8 mL / min, a column temperature of 30℃, and an injection volume of 5... .

[0092] Step 1042: Perform chromatographic separation and fluorescence detection on the sample to be tested according to the preset chromatographic conditions to obtain a chromatogram;

[0093] Specifically, the instrument used was a high-performance liquid chromatograph equipped with a fluorescence detector (FLD) and a UV detector, and an AQ-C18 column was installed. During detection, 5 μL of the sample to be tested was aspirated. The sample is injected into a liquid chromatograph, and separation and detection are performed according to the preset chromatographic conditions to obtain a chromatogram.

[0094] Step 1043: Quantitatively detect the sample to be tested based on the chromatogram to obtain the first detection result.

[0095] Specifically, the peak area corresponding to fructose can be read from the chromatogram. The peak area corresponding to the sample to be tested can be substituted into the standard curve equation y=0.0008x-0.1066 to calculate the concentration of fructose in the sample to be tested, where y is the peak area and x is the concentration of fructose in the sample to be tested.

[0096] The fructose content, or first result, is calculated based on the fructose concentration in the sample. Here, the fructose content can be calculated using the dilution factor in the pretreatment and the fructose concentration in the sample, and can be obtained using the following formula: Where w is the fructose content and x is the fructose concentration in the sample being tested. This is the final volume of the sample after acid hydrolysis. This represents the final volume of the sample after pre-column derivatization. denoted as , where is the sampling volume after pre-column derivatization, and W is the mass of the first sample.

[0097] In practical implementation, the analytical methodology can be fully validated through the following tests on linearity, sensitivity, precision, repeatability, stability, and recovery rate. This demonstrates that the method in this embodiment has good linearity, suitable sensitivity, stability, reliability, and accuracy, and is suitable for determining the content of fructose and glucose in crude polysaccharides of traditional Chinese medicine formula granules such as Polygonatum sibiricum.

[0098] Linearity, Limit of Detection, and Limit of Quantification: Accurately pipette the mixed reference solution (fructose and glucose) and serially dilute to a series of concentrations (65.625, 131.25, 262.5, 525, 1050, 2100, 4200, 8400). ). In terms of mass concentration (x, Using the x-axis as the horizontal axis and the peak area (y) as the vertical axis, a linear regression was performed. The fructose regression equation (i.e., the standard curve equation) is: y = 0.0008x - 0.1066. =0.9904; Glucose regression equation: y=0.0165x+1.1807, = 0.9994. Using a signal-to-noise ratio (S / N) of 3 as the limit of detection (LOD) and S / N of 10 as the limit of quantitation (LOQ), the LOQ for both fructose and glucose was 65.625. .

[0099] Precision test: Take the same mixed control solution (fructose 1500... 3000 glucose Six consecutive injections were performed. The RSD of fructose retention time was 0.77%, and the RSD of peak area was 0.10%; the RSD of glucose retention time was 0.53%, and the RSD of peak area was 0.11%. All values ​​were less than 2%, indicating good instrument precision.

[0100] Repeatability test: Six samples of crude polysaccharide (C-PCP-G) from the same batch of Polygonatum odoratum formulation granules were hydrolyzed, derivatized, and determined under optimal conditions. The RSD of fructose retention time was 0.66%, and the RSD of peak area was 1.48%; the RSD of glucose retention time was 0.46%, and the RSD of peak area was 1.79%. Both were less than 2%, indicating good repeatability of the method.

[0101] Stability test: The same test solution was injected and measured at 0, 2, 4, 8, 12, and 24 h. Within 24 h, the RSD of fructose retention time was 0.15%, and the RSD of peak area was 1.25%; the RSD of glucose retention time was 0.10%, and the RSD of peak area was 0.64%. This indicates that the solution has good stability within 24 h.

[0102] Recovery test: Acid hydrolysis followed by spiking was used. Six portions of crude polysaccharide from Polygonatum sibiricum granules with known content were accurately weighed. After acid hydrolysis under optimal conditions, mixed reference solutions at three concentration levels (50%, 100%, and 150% of the sample content) were added, followed by derivatization and determination. The average recovery rate of fructose was 95.12% to 104.00%, with an RSD of 2.76% to 4.80%; the average recovery rate of glucose was 100.15% to 102.38%, with an RSD of 0.27% to 3.27%. All recoveries were within the range of 80% to 120%, meeting industry standard requirements.

[0103] The fructose-specific indicators were verified using the following methods:

[0104] Maltodextrin (MYHJ), crude polysaccharide from Polygonatum sibiricum (C-PCP), and crude polysaccharide from Polygonatum sibiricum formulation granules (C-PCP-G) were taken separately and subjected to acid hydrolysis, derivatization, and detection under optimal conditions.

[0105] MYHJ chromatogram: No chromatographic peaks were observed at the retention time corresponding to fructose; only a single strong peak was observed at the glucose position, with a glucose content of 3731.27%. .

[0106] C-PCP chromatogram: Fructose content 6091.15 The glucose content is 626.15. The fructose / glucose ratio is approximately 9.73:1.

[0107] C-PCP-G chromatogram: Fructose content 1539.03 The glucose content is 3179.67. The fructose / glucose ratio is approximately 0.48:1.

[0108] Conventional pre-column derivatization methods exhibit significant structure selectivity for fructose, making effective labeling impossible. This is primarily because fructose, as a ketose, exists mainly in a cyclic hemiacetal form at its reducing end, making it far less likely to undergo effective condensation reactions with conventional reagents compared to aldoses. Furthermore, it can undergo competitive reactions such as degradation under alkaline conditions, leading to low derivatization efficiency. To address this issue, this invention employs p-AMBA as the derivatizing reagent. Its reaction principle differs from conventional methods, thus overcoming the aforementioned structural limitations and achieving accurate quantification of fructose. Figure 2As shown, the upper curve represents maltodextrin, the middle curve represents crude polysaccharide from Polygonatum sibiricum, and the lower curve represents crude polysaccharide from Polygonatum sibiricum granules. After hydrolysis, maltodextrin (MYHJ) showed no chromatographic peak at the retention time corresponding to fructose, while crude polysaccharide from Polygonatum sibiricum (C-PCP) exhibited a strong fructose signal at the same position. This comparison directly proves that the p-AMBA derivatization method successfully solves the technical bottleneck of traditional derivatization methods' ineffective detection of fructose (traditional derivatization methods have extremely low efficiency for fructose derivatization, and the fructose detection limit is usually >200). Meanwhile, the recovery tests during method validation showed that the average recovery rate of fructose ranged from 95.12% to 104.00%, with a SD of 2.76% to 4.80%, indicating stable recovery. This stability is attributed to the stable analytical matrix provided by the crude polysaccharide enrichment step in step 1021, ensuring that subsequent derivatization and detection results are reliable and unaffected by significant interference from excipient co-precipitation.

[0109] Therefore, the fructose detected in the first test result can be entirely attributed to Polygonatum polysaccharides. Fructose can serve as a specific qualitative and quantitative indicator for Polygonatum polysaccharides, thereby fundamentally avoiding the interference of maltodextrin on traditional total sugar determination methods.

[0110] In an optional embodiment of the present invention, step 105, which involves separating crude polysaccharides from the second sample and the preset reference standard to obtain the solid sample to be tested and the solid reference standard, may include:

[0111] Step 1051: Obtain a preset reference standard; the preset reference standard includes a blank reference standard sample and an excipient reference sample; the blank reference standard sample is consistent with the medicinal material source of the Chinese herbal medicine formula granules to be tested;

[0112] Specifically, the blank reference control sample is a medicinal material that is consistent with the medicinal material source of the Chinese medicine formula granules to be tested. When the Chinese medicine formula granules to be tested are Polygonatum formula granules, the blank reference control sample is Polygonatum medicinal material; the excipient control sample is maltodextrin.

[0113] Step 1052: Perform crude polysaccharide separation processing on the second sample, the blank reference control sample and the excipient control sample in the preset reference standard to obtain the solid sample to be tested and the solid reference standard.

[0114] Specifically, the steps for separating crude polysaccharides from the blank reference control sample include: taking Polygonatum sibiricum medicinal materials from different origins, extracting them according to the formula granule production process (refluxing with 8 times the volume of water for 1.5 h, then refluxing with 6 times the volume of water for 1 h), combining the filtrates, concentrating, precipitating overnight with 80% ethanol, collecting the precipitate, washing with anhydrous ethanol, and freeze-drying to obtain the first reference standard. The steps in step 1021 are then used to separate crude polysaccharides from the excipient reference sample to obtain the second reference standard. Solid reference standards include the first and second reference standards. The steps in step 1021 are then used to separate crude polysaccharides from the second sample to obtain the solid sample to be tested.

[0115] In an optional embodiment of the present invention, step 106, which involves detecting the molecular weight characteristic distribution of the solid sample to be tested and the solid reference standard to obtain a second detection result, may include:

[0116] Step 1061: Perform chromatographic detection on the solid sample to be tested and the solid reference standard to obtain chromatographic data;

[0117] Specifically, take 5 mg each of the first solid reference standard (i.e., crude polysaccharide from Polygonatum sibiricum, C-PCP), the test solid sample (i.e., crude polysaccharide from Polygonatum sibiricum granules, C-PCP-G), and the second solid reference standard (i.e., maltodextrin reference sample, MYHJ), dissolve them in ultrapure water, and dilute to 5 mL to prepare a 1.0 mg / mL test solution. After vortexing, inject the sample according to the chromatographic detection conditions, and record the chromatogram and the retention time of the main chromatographic peaks, i.e., the chromatographic data. Here, the chromatographic detection conditions include: a mixed-pore size aqueous gel column (7.8 mm ID × 30 cm), and a mobile phase of 0.2 mol / L NaCl solution (after passing through 0.22... (Filter membrane filtration and ultrasonic degassing), elution method: isocratic elution, flow rate: 0.4 mL / min, injection volume: 20 μL. The elution time was 30 min, the column temperature was 40℃, the detector was a differential refractive index detector (RID), and the temperature was 40℃.

[0118] Step 1062: Detect the molecular weight characteristic distribution based on the chromatographic data and the preset standard curve to obtain the second detection result.

[0119] Specifically, the preset standard curve was determined as follows: Accurately weigh 5 mg each of dextran standards (molecular weights of 1, 4, 35, 60, and 95 kDa), dissolve each in ultrapure water, and dilute to 5 mL to prepare a 1.0 mg / mL single-standard solution. Inject the solutions under the chromatographic detection conditions described above, and record the chromatographic retention time t (min) for each standard. The standard was then analyzed using the logarithm of the dextran molecular weight. Using the vertical axis as the ordinate and the chromatographic retention time t as the horizontal axis, a linear regression was performed to obtain a pre-defined standard curve. The regression equation for the pre-defined standard curve is: =0.3686t+12.422, ,in, denoted as weight-average molecular weight, t as chromatographic retention time, and R as the coefficient of determination. This indicates that within the molecular weight range of 1 to 95 kDa, It exhibits a good linear relationship with retention time.

[0120] Substitute the chromatographic data into the pre-set standard curve regression equation to calculate the molecular weight parameter of each sample's chromatographic peak: weight-average molecular weight. Number average molecular weight and polydispersity index (PDI) Here, the peak area of ​​each elution fragment is obtained by slicing and integrating the chromatographic elution curve (i.e., integrating the chromatogram in the chromatographic data in segments). ,pass , The weight-average molecular weight was calculated. Number average molecular weight ,in, Let be the peak area of ​​the i-th chromatographic slice. The molecular weight corresponding to the i-th chromatographic slice (obtained by substituting the chromatographic retention time into the regression equation of the preset standard curve). Based on the weight-average molecular weight... The polydispersity index (PDI) was used to obtain a second detection result, which included the weight-average molecular weight. And polydispersity index (PDI).

[0121] The degree of interference from excipients can be determined based on the second test results. When <5 kDa and PDI<5, it is judged to be close to the intrinsic characteristics of the medicinal material, with minimal interference from excipients; when the second test result is... When the value is in the range of 25 to 35 kDa and the PDI is in the range of 15 to 25, it is judged to be a typical particle characteristic of the formulation, indicating the presence of excipient co-precipitation but within the expected process range; when the second test result is in If the value is >40 kDa or PDI>25, it is determined that the maltodextrin excipients have been added excessively or the process is abnormal.

[0122] The above-mentioned threshold is set based on the molecular weight distribution of the crude polysaccharide (C-PCP-G) in the Polygonatum odoratum formula granules. (kDa, PDI=19.3) is significantly different from pure medicinal polysaccharides ( kDa, PDI=2.02), which is also different from pure maltodextrin ( Instead of exhibiting the typical characteristics of a coprecipitated complex (kDa, PDI=15.9), it shows the characteristics of a typical coprecipitated complex (see [link to relevant documentation]). Figure 3 , Figure 3 The upper curve represents crude polysaccharides from the formulation granules, the middle curve represents crude polysaccharides from the processed medicinal materials, and the lower curve represents maltodextrin. This characteristic reflects both the presence of polysaccharides in the medicinal materials and quantifies the degree of interference from excipients. Therefore, molecular weight distribution is not an interference, but a quantifiable indicator of the process.

[0123] In an optional embodiment of the present invention, step 107, pre-treating the third sample to obtain a pre-treated sample, may include:

[0124] Step 1071: Grind and dissolve the third sample to obtain a sample solution;

[0125] Step 1072: Centrifuge the sample solution to obtain crude polysaccharide powder;

[0126] Step 1073: The crude polysaccharide powder is pressed to obtain a pretreated sample.

[0127] Specifically, weigh 1 g of the powder from the third sample and grind it evenly in an agate mortar. Dissolve the ground powder in 10 mL of boiling water (100℃) and stir thoroughly. Centrifuge at 8000 r / min for 5 minutes and collect the supernatant. While continuously stirring with a magnetic stirrer, slowly add anhydrous ethanol to the supernatant (dropping rate approximately 1 mL / min) until the final ethanol concentration reaches 70%. Let stand for 30 minutes to allow the polysaccharide to precipitate fully. Centrifuge at 8000 r / min for 5 minutes, discard the supernatant, and collect the precipitate (i.e., crude polysaccharide). Add 5 mL of ultrapure water to the precipitate to reconstitute it, transfer it to a lyophilizer and freeze-dry (-55℃, pressure <10 Pa, drying for 24 hours) to obtain crude polysaccharide lyophilized powder. Take 10 mg of crude polysaccharide lyophilized powder and press it into a circular slice with a diameter of approximately 5-8 mm using a glass slide for use as a pretreatment sample.

[0128] In an optional embodiment of the present invention, step 108, performing variety detection based on the Raman spectral data of the pretreated sample and the trained variety prediction model to obtain a third detection result, may include:

[0129] Step 1081: Raman spectral data is acquired on the pretreated sample according to preset data acquisition parameters and preset data acquisition method to obtain Raman spectral data;

[0130] Specifically, a confocal micro Raman spectrometer was used for spectral acquisition. Preset data acquisition parameters included: excitation source wavelength of 785 nm (near-infrared laser, which significantly reduces fluorescence background); objective magnification of 50x (for long working distance); laser power of 100% (approximately 50 to 100 mW); and spectral scanning range of 400-1800 nm. (This region contains the main characteristic peaks of polysaccharides); integration time: 10 seconds; number of iterations: 3; grating: 600 lines / mm; slit width: 100 mm. .

[0131] The preset data acquisition method was as follows: the crude polysaccharide tablets of the pretreated sample were placed on the stage, and the focus was adjusted to the optimal signal. Ten spectra were randomly collected from different sites on each tablet. For different batches of the same variety, five tablets were prepared for each batch, and a total of 150 spectra were collected. All spectral acquisitions were completed under the same environmental conditions (temperature 25±2℃, relative humidity 40%-60%).

[0132] Raman spectral data were acquired using the preset data acquisition parameters and preset data acquisition methods described above.

[0133] Step 1082: Perform data preprocessing on the Raman spectral data to obtain preprocessed data;

[0134] Specifically, noise reduction and correction are performed on Raman spectral data, including:

[0135] Smoothing and Noise Reduction: Savitzky-Golay (SG for short, a sliding window algorithm based on least squares polynomial fitting) smoothing filtering is used. Parameters: window size is 25, polynomial order is 3. SG smoothing is based on local polynomial fitting, which can effectively reduce high-frequency noise while preserving the characteristic peak shape of the spectrum.

[0136] Scattering correction: Multivariate scattering correction (MSC) is used to eliminate scattering effects caused by particle inhomogeneity and differences in surface roughness. The mean of all sample spectra is used as the reference spectrum, and linear fitting correction is performed on each spectrum.

[0137] Baseline correction: Adaptive iterative weighted penalized least squares (airPLS) is used to subtract fluorescence background. airPLS minimizes the objective function by iteratively optimizing the weights. Parameter settings: regularization parameter is 10000, and the maximum number of iterations is 25. Accurate baseline correction is achieved through iterative weight optimization.

[0138] Step 1083: Input the preprocessed data into the trained variety prediction model for variety detection to obtain the third detection result.

[0139] Specifically, the variety prediction model is obtained through the following training process:

[0140] Twelve different varieties of traditional Chinese medicine granules were selected (Angelica dahurica BZ, Codonopsis pilosula DS, Poria cocos FL, Polygonatum sibiricum HJ, Astragalus membranaceus HQ, Platycodon grandiflorus JG, Ophiopogon japonicus MD, Ganoderma lucidum LZ, Ginseng RS, Rehmannia glutinosa SDH, Gastrodia elata TM, and Pseudostellaria heterophylla TZS), with three batches for each variety. A two-tiered sampling method was used: 20 spectra were randomly selected from each sample category as an independent validation set; the remaining samples were used as the model training pool, divided into training and validation sets in a stratified ratio of 6:2:2.

[0141] The structure of the preset network model is as follows:

[0142] Input layer: Receives preprocessed one-dimensional vectors of Raman spectra (spectral data length N = 1401 data points, corresponding to wavenumber range 400 to 1800). scope).

[0143] The first convolutional module consists of a one-dimensional convolutional layer with 16 kernels of size 9 and a stride of 1, using equal-length padding (same padding) to ensure the output sequence length remains constant; followed by a batch normalization layer; a ReLU nonlinear activation function to extract nonlinear spectral features; a one-dimensional max-pooling layer with a pooling window size of 3 to compress the feature dimension and reduce computational cost; and a Dropout layer with a ratio of 0.2, which discards 20% of neurons to suppress model overfitting.

[0144] The second convolutional module consists of a one-dimensional convolutional layer with 32 kernels of size 7, a stride of 1, and identical padding; followed by a batch normalization layer; a ReLU non-linear activation function; a one-dimensional max pooling layer with a pooling window size of 3; and a Dropout layer with a scaling factor of 0.2.

[0145] Flatten layer: Flattens a multidimensional feature map into a one-dimensional vector.

[0146] Fully connected layer: a dense network layer containing 64 neurons, using the ReLU non-linear activation function; Dropout layer, scale 0.2.

[0147] Output layer: 12 output neurons are set up, corresponding to 12 categories of medicinal materials. The Softmax activation function is used to output the classification probability of each category, thus completing the multi-variety discrimination.

[0148] The model training strategy is as follows: the Adam optimizer is used, with an initial learning rate of 0.0001; the loss function is sparse classification cross-entropy, which is suitable for multi-class classification tasks; the batch size per iteration is set to 32, and the total maximum number of iterations is set to 200; an early stopping mechanism is added: training is terminated if there is no improvement in the accuracy of the validation set for 20 consecutive iterations to prevent overfitting; and an automatic learning rate decay strategy is enabled: if there is no performance improvement for 10 consecutive iterations, the current learning rate is reduced to 0.5 times the original value.

[0149] The preset network model is trained using the training set according to the above model training strategy. The preset network model is iterated repeatedly on the training set, continuously updating the convolutional kernel weights and fully connected layer parameters, learning the feature mapping relationship between Raman spectra and medicinal material categories, and completing network fitting. The training accuracy of each round is monitored, and with the early stopping strategy, the training is terminated when the performance of the validation set no longer improves after several consecutive rounds, thus obtaining the variety prediction model.

[0150] The preprocessed data is input into the trained variety prediction model, and convolution is performed sequentially to extract features, normalize, activate, pool, and regularize. After flattening, the data is connected to a 64-neuron fully connected layer, and finally the variety discrimination result is output through Softmax as the third detection result.

[0151] like Figures 4 to 7 As shown, the average Raman spectra (colored curves) and standard errors (gray shading) of different Chinese herbal medicine granules and maltodextrin after pretreatment are presented. Figure 4 For the direct compression method spectrum (400–1800) ); Figure 5 SERS spectra of alcohol-soluble components (530–1800) ); Figure 6 Raman spectra of crude polysaccharide components (400–1800) ); Figure 7 Raman spectra of maltodextrin (400–1800) Each spectrum is the average of 150 measurements. The horizontal axis in the graph represents the Raman wavenumber, and the vertical axis represents the Raman peak intensity. Raman wavenumber ( Raman peak intensity (au)

[0152] As attached Figures 4 to 7 As shown, the direct compression spectrum ( Figure 4 It mainly exhibits the characteristic peak of maltodextrin (480). and 1455 Different varieties exhibit highly similar spectra. Alcohol-soluble SERS spectra ( Figure 5 ) showed strong signals of common metabolites (1004, 1031, 1080) The crude polysaccharide spectrum ( Figure 6 ) at 480 (Pyranose ring skeletal vibration), 842 (Characteristic peak of α-pyranoside bond), 1122 Clear polysaccharide characteristic peaks were observed at locations such as (COC stretching vibration), with significant differences in peak shape and relative intensity among different varieties. Maltodextrin ( Figure 7 Although the characteristic peaks of ) still exist, they no longer dominate.

[0153] In an optional embodiment of the present invention, step 109, performing quality testing on the Chinese herbal medicine formula granules to be tested based on the first detection result, the second detection result, and the third detection result to obtain quality testing results, may include:

[0154] Step 1091: Obtain the preset product qualification conditions;

[0155] Specifically, the pre-set product qualification conditions include: the fructose content in the first test result is not less than 1500 mg / L. (Or set according to the actual internal control standards, such as average ±30%), the weight-average molecular weight in the second test result Within the range of 25 to 35 kDa and PDI within the range of 15 to 25, the variety identification result in the third test result is consistent with the preset label and the confidence level is not less than 95%.

[0156] Step 1092: Perform quality testing based on the preset product qualification conditions, the first test result, the second test result, and the third test result to obtain the quality test result.

[0157] Specifically, the first, second, and third test results are compared one by one with the preset product qualification conditions. Only when the first, second, and third test results are all consistent with the preset product qualification conditions does it mean that the three test results of the Chinese herbal medicine formula granules to be tested in terms of composition, structure, and identification meet the preset product qualification conditions. At this time, the quality test result is qualified.

[0158] This invention provides a method for quality detection of traditional Chinese medicine (TCM) formula granules. For the first time, fructose is used as a specific marker for Polygonatum sibiricum polysaccharides, effectively eliminating interference from maltodextrin. The method clearly detects the fructose signal in Polygonatum sibiricum polysaccharides without interference from maltodextrin. Method validation shows that linearity, precision, repeatability, stability, and recovery all meet the requirements. The sugar peak area exhibits a good linear relationship with sample concentration, making it suitable for batch-to-batch consistency monitoring. The degradation pattern of fructose during acid hydrolysis is clarified, and optimized conditions ensure detection accuracy. This method can serve as a basis for identifying excessive excipient addition or process abnormalities (such as improper alcohol precipitation concentration or insufficient washing). Molecular weight distribution characteristics can be used as a monitoring indicator for quality transfer during the production of formula granules.

[0159] The method for quality testing of traditional Chinese medicine formula granules according to this invention can be completed within 2 to 3 days, and is applicable to enterprise internal control and quality evaluation by pharmaceutical testing institutions. This method is not limited to Polygonatum sibiricum, but can also be reasonably extrapolated to other traditional Chinese medicine formula granules with polysaccharides as the main active ingredient (such as Codonopsis pilosula, Poria cocos, Astragalus membranaceus, Ganoderma lucidum, Lentinus edodes, etc.) by adjusting parameters (such as acid hydrolysis conditions, chromatographic gradient, etc.).

[0160] To make the technical solution of this invention clearer, the following detailed description of the specific embodiments of this invention is based on Polygonatum sibiricum formula granules and in conjunction with the accompanying drawings. Those skilled in the art should understand that the following embodiments are only for illustrating this invention and do not constitute a limitation on the scope of this invention. The process of combining crude polysaccharide enrichment with Raman spectroscopy acquisition / preprocessing established in this invention has been verified with 12 different varieties (Angelica dahurica, Codonopsis pilosula, Poria cocos, Polygonatum sibiricum, Astragalus membranaceus, Platycodon grandiflorus, Ophiopogon japonicus, Ganoderma lucidum, Panax ginseng, Rehmannia glutinosa, Gastrodia elata, and Pseudostellaria heterophylla) (see Example 4), proving its cross-variety applicability. Based on the technical concept disclosed in this embodiment, when applying it to other traditional Chinese medicine formula granules with polysaccharides as the main active ingredient, only the same crude polysaccharide enrichment process and spectral acquisition parameters need to be followed. Routine adjustments to parameters such as acid hydrolysis conditions and chromatographic gradients should still fall within the protection scope of this invention.

[0161] Example 1: Extraction, preparation, and basic characterization of crude polysaccharides from Polygonatum sibiricum granules

[0162] Preparation of crude polysaccharides

[0163] Accurately weigh 10 g of Polygonatum odoratum formulation granules, add 200 mL of hot water (100℃) and stir thoroughly to dissolve. Centrifuge at 8000 r / min for 10 min and collect the supernatant. While stirring, slowly add anhydrous ethanol to the supernatant until the final ethanol concentration reaches 80%. Let stand overnight (approximately 12 h), centrifuge at 8000 r / min for 10 min, and collect the precipitate. Wash the precipitate twice with anhydrous ethanol and freeze-dry under vacuum (-55℃, pressure <10 Pa, drying for 24 h) to obtain crude polysaccharide (C-PCP-G) powder from Polygonatum odoratum formulation granules. Weigh and calculate the yield. Maltodextrin was treated using the same method to obtain a maltodextrin control sample. The crude polysaccharide yields were: C-PCP-G average yield 40.54%, MYHJ yield 64.30%, indicating that maltodextrin has limited solubility in 80% ethanol and can co-precipitate with C-PCP.

[0164] Protein content determination (BCA method)

[0165] Preparation of BCA (dioctanoic acid) working solution: Mix BCA reagent A and reagent B at a volume ratio of 50:1.

[0166] Standard curve: Take bovine serum albumin (BSA) standard (2 mg / mL) and dilute it to 0, 25, 50, 100, 200, 400, and 800 mg / mL. Series concentrations. Take 25g of each. Add 200 to each well of a 96-well plate. The BCA working solution was incubated at 37°C for 30 min, and the absorbance at 562 nm was measured. The regression equation is: y = 0.0015x + 0.1183. .

[0167] Sample determination: Accurately weigh 10 mg of C-PCP-G, dissolve in water and bring the volume to 10 mL (1 mg / mL). Take 25 μL of the test solution, perform the same procedure, substitute into the standard curve to calculate the protein content, and convert it to protein residual rate.

[0168] The results showed that the residual rate of C-PCP-G protein (1.31%) was significantly lower than that of crude polysaccharide in Polygonatum odoratum (average 4.17%), mainly due to the dilution effect of maltodextrin.

[0169] Fourier transform infrared (FT-IR) characterization

[0170] A trace amount of crude polysaccharide was collected and scanned using a PerkinElmer Fourier transform infrared spectrometer in attenuated total reflectance (ATR) mode, with a wavenumber range of 4000 to 500 cm⁻¹ and a resolution of 4 cm⁻¹, for a total of 16 scans.

[0171] Results: All samples were within 3400. (OH stretching), 2930 (CH extension), 1640 (Combined water or C=O), 1400 (CH angle change), 1200 to 100 The region exhibits polysaccharide-characteristic absorption at the (COC and COH stretching) sites. C-PCP shows a weak absorption peak near 930 cm⁻¹ (β-glycosidic bond), while MYHJ shows a peak at 850 cm⁻¹. There is a shoulder peak nearby ( (-type glycosidic bond), C-PCP-G exhibits the superposition characteristics of the two.

[0172] Example 2: Monosaccharide composition analysis of Polygonatum sibiricum formulation granules (fructose-specific index)

[0173] Optimization of acid hydrolysis conditions

[0174] Using C-PCP-G as the model, the acid hydrolysis conditions were optimized through single-factor and orthogonal experiments. The optimal conditions were determined to be: TFA concentration 0.5 mol / L, acid hydrolysis temperature 110℃, acid hydrolysis time 0.5 h, and polysaccharide concentration 1.0 mg / mL.

[0175] Pre-column derivatization and HPLC-FLD analysis

[0176] C-PCP-G was acid-hydrolyzed under optimal conditions. 50 μL of the acid-hydrolyzed sample solution was taken. Add 100 p-AMBA derivatization reagent (0.35 g p-aminobenzoic acid + 0.05 g sodium cyanoborohydride + 0.5 g acetic acid dissolved in 5 mL methanol), reacted at 70 °C for 40 min, cooled, and then diluted to 2 mL with water. The solution was filtered through a 0.22 mL filter. Filter membrane, sample injection and analysis.

[0177] Fructose-specific indicator verification

[0178] MYHJ, C-PCP, and C-PCP-G were taken respectively, hydrolyzed, derivatized, and detected under optimal conditions, and the verification results of fructose-specific indicators are shown in Table 1.

[0179] Table 1. Validation results of fructose-specific indicators

[0180]

[0181] Conclusion: Fructose can be used as a specific qualitative and quantitative indicator for Polygonatum polysaccharides.

[0182] Example 3: Determination of molecular weight distribution of Polygonatum sibiricum granulated polysaccharides

[0183] Establishing a standard curve

[0184] The regression equation for establishing the preset standard curve is as follows, following step 1062: =0.3686t+12.422, .

[0185] Sample Measurement

[0186] Take 2 mg each of C-PCP, C-PCP-G, and MYHJ, dissolve them in 2 mL of 0.2 mol / L NaCl, and filter through 0.22 mL... The filter membrane was subjected to molecular weight characteristic distribution detection according to the steps in step 1062, and the molecular weight distribution results are shown in Table 2: weight-average molecular weight Number average molecular weight And polydispersity index (PDI).

[0187] Table 2 Molecular weight distribution results

[0188]

[0189] Conclusion: The molecular weight distribution of C-PCP-G is highly similar to that of MYHJ, confirming that the co-precipitation of excipients leads to molecular weight shift.

[0190] Example 4: Raman spectroscopy-machine learning identification method based on crude polysaccharide enrichment

[0191] To verify the effectiveness of the method of the present invention in the identification of actual samples, this embodiment uses the method and model established in Examples 1-3 to perform batch variety identification on independent validation set samples of 12 varieties, demonstrating the complete identification process and results.

[0192] Sample pretreatment

[0193] In this embodiment, 12 different varieties of traditional Chinese medicine formula granules were selected, with at least 3 batches of each variety. Crude polysaccharide enrichment pretreatment and Raman spectroscopy acquisition were performed. The specific steps are as follows:

[0194] Experimental materials: Various traditional Chinese medicine (TCM) granule formulations, all with maltodextrin as the excipient. The sample composition included three batches of granules containing Angelica dahurica, Codonopsis pilosula, Poria cocos, Polygonatum sibiricum, Astragalus membranaceus, Platycodon grandiflorus, Ophiopogon japonicus, Ganoderma lucidum, Panax ginseng, Rehmannia glutinosa, Gastrodia elata, and Pseudostellaria heterophylla, totaling 12 kinds of TCM. Polysaccharides were extracted from the above TCM granules to obtain different TCM polysaccharide samples, which were named according to the Chinese pinyin abbreviations of the above TCMs (BZ, FL, HJ, HQ, JG, MD, LZ, RS, SDH, TM, TZS, respectively). The control excipient maltodextrin was represented as MYHJ.

[0195] Extraction method of polysaccharides from traditional Chinese medicine: Weigh 1g of the traditional Chinese medicine formula granules, place them in a beaker, add 10mL of boiling water to dissolve them completely, add anhydrous ethanol until the ethanol volume reaches 70%, let stand for 30min, centrifuge at 8000r / min for 5min, collect the precipitate, wash three times with anhydrous ethanol, evaporate the residual ethanol from the precipitate, dissolve it in ultrapure water, and then freeze-dry to obtain different polysaccharide powders from traditional Chinese medicine. Take 10mg of crude polysaccharide freeze-dried powder, compress it into tablets using a tablet press, and set aside for later use.

[0196] Raman spectroscopy acquisition

[0197] Raman signals of polysaccharides from traditional Chinese medicine were acquired using a laser confocal microscopy Raman spectrometer. This instrument is equipped with a 785nm laser, a charge-coupled device (CCD) detector, an inverted microscope, and an automated moving stage. The grating size is 600 g / mm², and the spectral resolution is 2. Before testing, the instrument should be fully preheated.

[0198] The specific testing method is as follows: Take 10 mg of lyophilized powder for each polysaccharide, compress it into tablets using a tablet press, place the tablets on the loading stage, and focus the sample using a 50× objective lens. The laser power is 50 mW, the integration time is 10 seconds, and the process is repeated 3 times to obtain samples ranging from 400 to 1800. Raman spectral signals within a certain range were collected. The entire acquisition process was conducted in darkness to avoid environmental interference. At least 10 spectra were randomly collected from different sites on each tablet. For different batches of the same variety, 5 tablets were prepared for each batch, and a total of 150 spectra were collected. For comparison, the same samples were processed and spectrally acquired using both the "direct tableting method" and the "alcohol-soluble SERS method".

[0199] Spectral preprocessing

[0200] After exporting all raw Raman spectral data from the instrument, interference from noise, baseline drift, and scattering effects was eliminated. First, a smoothing algorithm was used to reduce noise, and interpolation was performed to correct missing and outlier values ​​in the spectrum, improving spectral smoothness and signal-to-noise ratio. Parameters were set to a window size of 25 and a polynomial order of 3. Next, the multivariate scattering correction (MSC) method was used, with the mean of all sample spectra as the reference spectrum, and scattering effects were eliminated through linear fitting. Then, an adaptive iterative weighted penalized least squares method was used to remove fluorescence background, with a penalty coefficient of 10000 and a maximum number of iterations of 25. Accurate baseline correction was achieved through iterative optimization of the weights.

[0201] The preprocessed spectral data is input into a pre-trained variety prediction model, which outputs the variety identification results and confidence level of the traditional Chinese medicine formula granules.

[0202] Conclusion: The trained variety prediction model can predict the variety of 12 different varieties of traditional Chinese medicine formula granules, which shows that the variety prediction model can accurately identify the variety of traditional Chinese medicine formula granules.

[0203] The method for quality testing of traditional Chinese medicine formula granules according to the present invention has the following outstanding beneficial effects:

[0204] It is highly systematic and integrated: It is the first to propose an overall strategy that combines common preprocessing, three-dimensional analysis and comprehensive judgment, and constructs a complete quality evaluation system, rather than an improvement of a single method.

[0205] Highly specific and directly addresses pain points: By defining "fructose" as a specific marker and revealing the molecular weight shift mechanism, it fundamentally avoids the interference of maltodextrin excipients and solves the industry problem of inflated content determination.

[0206] High accuracy and robust reliability: The methodology has been fully validated, the machine learning identification accuracy reaches 98.8%, and the misjudgments are chemically interpretable, which is far superior to existing rapid identification methods.

[0207] The process is convenient and universal: all steps utilize conventional laboratory equipment, and the entire evaluation process can be completed within 2 to 3 days. The crude polysaccharide enrichment combined with Raman spectroscopy acquisition / preprocessing established in this invention has been validated on 12 different varieties, demonstrating its cross-varietal applicability. For new varieties, only the same sample pretreatment and spectral acquisition procedures need to be followed to utilize the technical framework of this invention for quality identification. This method has strong versatility and universal applicability of its principles.

[0208] It should be noted that in the method of the present invention, the components or steps can obviously be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent solutions of the present invention. Furthermore, the steps of performing the above series of processes can naturally be performed in the order described and in chronological order, but are not necessarily required to be performed in chronological order. Some steps can be performed in parallel, overlapping, or independently of each other.

[0209] It should be noted that in the above embodiments, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments described above is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0210] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for quality testing of traditional Chinese medicine formula granules, characterized in that, include: Obtain the Chinese herbal formula granules to be tested and divide them into three equal parts to obtain the first sample, the second sample, and the third sample; The first sample was subjected to acid hydrolysis treatment according to the preset acid hydrolysis conditions to obtain an acid hydrolyzed sample; The acid-hydrolyzed sample was subjected to pre-column derivatization to obtain the sample to be tested; The sample to be tested is quantitatively analyzed to obtain the first test result; The second sample and the preset reference standard were subjected to crude polysaccharide separation treatment to obtain the solid sample to be tested and the solid reference standard. The molecular weight characteristic distribution of the solid sample to be tested and the solid reference standard are analyzed to obtain a second detection result; The third sample was pretreated to obtain a pretreated sample; Based on the Raman spectral data of the pretreated sample and the trained variety prediction model, variety detection is performed to obtain a third detection result; the variety prediction model is obtained by training a preset network model based on the Raman spectral data of different varieties of traditional Chinese medicine formula granules. The quality of the Chinese herbal medicine granules to be tested is determined based on the first test result, the second test result, and the third test result. The first sample is subjected to acid hydrolysis treatment according to preset acid hydrolysis conditions to obtain an acid hydrolyzed sample, including: The first sample was subjected to crude polysaccharide separation treatment to obtain crude polysaccharide; The crude polysaccharide was subjected to acid hydrolysis under preset acid hydrolysis conditions to obtain an acid hydrolyzed sample. The preset acid hydrolysis conditions included: the acid hydrolysis reagent being trifluoroacetic acid with a concentration of 0.2 to 0.8 mol / L, the acid hydrolysis temperature being 100 to 115°C, and the acid hydrolysis time being 0.5 to 1.0 h. The second sample and the preset reference standard are subjected to crude polysaccharide separation processing to obtain the solid sample to be tested and the solid reference standard, including: Obtain a preset reference standard; the preset reference standard includes a blank reference standard sample and an excipient reference sample; the blank reference standard sample is consistent with the medicinal material source of the Chinese herbal medicine formula granules to be tested; The second sample, the blank reference control sample and the excipient control sample in the preset control standard are subjected to crude polysaccharide separation processing to obtain the solid sample to be tested and the solid control standard. The quality of the Chinese herbal medicine granules to be tested is determined based on the first test result, the second test result, and the third test result, and the quality test results are obtained, including: Obtain the preset product qualification conditions; Quality inspection is performed based on the preset product qualification conditions, the first test result, the second test result, and the third test result to obtain a quality inspection result; wherein, the first test result, the second test result, and the third test result are compared one by one with the preset product qualification conditions, and the quality inspection result is qualified when the first test result, the second test result, and the third test result are all consistent with the preset product qualification conditions.

2. The method for quality testing of traditional Chinese medicine formula granules according to claim 1, characterized in that, The first sample was subjected to crude polysaccharide separation treatment to obtain crude polysaccharide, comprising: Take the first sample, add 10 times the amount of water at 90 to 100°C, stir to dissolve, centrifuge at 8000 r / min for 10 minutes to obtain the supernatant; Anhydrous ethanol was added to the supernatant while stirring until the final alcohol concentration was 80%. After standing, the mixture was centrifuged at 8000 r / min for 10 minutes to obtain a precipitate. The precipitate was washed with anhydrous ethanol and then freeze-dried under vacuum to obtain crude polysaccharide; wherein the temperature during vacuum freeze-drying was -50℃ to -60℃, the pressure was less than 20Pa, and the drying time was 24 to 48 hours.

3. The method for quality testing of traditional Chinese medicine formula granules according to claim 1, characterized in that, The acid-hydrolyzed sample is subjected to pre-column derivatization to obtain the test sample, including: Obtaining derivatization reagents; According to the preset pre-column derivatization conditions and the derivatization reagent, the acid-hydrolyzed sample is subjected to pre-column derivatization to obtain the sample to be tested.

4. The method for quality testing of traditional Chinese medicine formula granules according to claim 1, characterized in that, The sample to be tested is quantitatively analyzed to obtain a first detection result, including: Obtain preset chromatographic conditions; the preset chromatographic conditions include: the chromatographic column is a hydrophilic chromatographic column, the mobile phase is a buffer-buffer-methanol mixture, the chromatographic separation mode is gradient elution, the ultraviolet detection wavelength is 303nm, the excitation wavelength is 313nm, and the emission wavelength is 358nm; The sample to be tested is subjected to chromatographic separation and fluorescence detection according to the preset chromatographic conditions to obtain a chromatogram; The sample to be tested is quantitatively detected based on the chromatogram to obtain the first detection result.

5. The method for quality testing of traditional Chinese medicine formula granules according to claim 1, characterized in that, Based on the molecular weight characteristic distribution detection of the solid sample to be tested and the solid reference standard, a second detection result is obtained, including: The solid sample to be tested and the solid reference standard were subjected to chromatographic detection to obtain chromatographic data; Based on the chromatographic data and the preset standard curve, the molecular weight characteristic distribution is detected to obtain the second detection result.

6. The method for quality testing of traditional Chinese medicine formula granules according to claim 1, characterized in that, The third sample is pretreated to obtain a pretreated sample, including: The third sample was ground and dissolved to obtain a sample solution; The sample solution was centrifuged to obtain crude polysaccharide powder; The crude polysaccharide powder was pressed to obtain a pretreated sample.

7. The method for quality testing of traditional Chinese medicine formula granules according to claim 1, characterized in that, Based on the Raman spectral data of the pretreated sample and the trained variety prediction model, variety detection is performed to obtain a third detection result, including: Raman spectral data of the pretreated sample were acquired according to preset data acquisition parameters and preset data acquisition method to obtain Raman spectral data. The Raman spectral data are preprocessed to obtain preprocessed data; The preprocessed data is input into the trained variety prediction model for variety detection, and a third detection result is obtained.

Images