2.3.1 Sample and standard preparation

Take 2 bags from each of the 5 batches of samples and mix them thoroughly. The volatile oil was extracted according to the method outlined in the 2020 edition of the Chinese Pharmacopoeia (general rule 2204 A method). The sample was weighed to 80.0 g and placed in a round-bottom flask containing 400 mL of water. Then, 2 mL of cyclohexane was added to the water surface in the volatile oil detector. The mixture was slowly heated to boiling and maintained at a gentle boil for 3 h before being cooled. The cyclohexane layer was transferred to a 2 mL volumetric flask. It was then diluted with additional cyclohexane up to the calibration mark, thoroughly shaken, and dehydrated using an appropriate amount of anhydrous sodium sulfate. The sample solution for GC–MS analysis was obtained by filtering the sample through a 0.22 μm microporous membrane.

Take 1 bag from each of the 5 batches of samples, remove the packaging, grind them finely, and mix thoroughly. Approximately 2.0 g of the sample powder was accurately weighed and placed into a stoppered 100 mL conical flask. Then, 100 mL of 70 % methanol solution was added. Subject the mixture to ultrasonic extraction for 30 min. After cooling, reweigh the flask and compensate for any weight loss by adding 70 % methanol solution. The sample was filtered using a 0.22 μm microporous membrane to obtain the filtrate, which was then utilized as the test solution for UPLC-Q-Orbitrap HRMS analysis.

A solution of mixed standard substances was created by dissolving an appropriate amount of each standard substance in methanol, resulting in a mass concentration of about 60 μg/mL. The solution was subsequently filtered through a 0.22 μm microporous membrane to obtain the filtrate, which served as the standard solution for UPLC-Q-Orbitrap HRMS analysis.

2.3.2 GC–MS analysis

The volatile components were analyzed by 7890A-5975C series gas chromatography-mass spectrometry (Agilent, USA).

An Agilent 19091S-433 capillary chromatographic column (30 m × 250 μm × 0.25 μm) was utilized for the analysis. The column temperature box initially held a temperature of 110 °C for 7 min, followed by a ramp to 170 °C at a rate of 12 °C/min and held at this temperature for 2 min. Subsequently, the temperature was raised to 230 °C at a rate of 20 °C/min and held for 2 min. Helium was employed as the carrier gas at a flow rate of 1 mL/min with an inlet temperature set at 230 °C. The split ratio was 20: 1, and the injection volume was 1 μL. Ionization was conducted using the EI mode, with an electron energy of 70 eV, an ion source temperature of 200 °C, and a scanning mass range of 30–400. The analysis utilized the NIST mass spectrometry retrieval standard library.

2.3.3 UPLC-Q-Orbitrap HRMS analysis

The samples were analyzed using ultra-high performance liquid chromatography-quadrupole-electrostatic field orbitrap high-resolution mass spectrometry (Thermo Fisher Scientific, USA). The chromatographic separation was carried out using a Thermo Scientific Accucore^TM C₁₈ column (3 mm × 100 mm, 2.6 μm). The mobile phase consisted of a 0.1 % formic acid aqueous solution (A) and methanol (B). The gradient elution method was as follows: 0–11 min, 5 %-35 % B; 11–18 min, 35 %-55 % B; 18–22 min, 55 %-75 % B; 22–24 min, 75 %-90 % B; 24–25 min, 90 %-100 % B; 25–30 min, 100 % B. The column temperature was maintained at 50 °C. The flow rate was set at 0.3 mL/min, and the injection volume was 5 μL.

The mass spectrometry conditions for the analysis involved electrospray ion source (ESI) in both positive and negative ion modes of detection. The spray voltage was set to 3.5 kV (+) and 3.0 kV (-). The auxiliary gas heating temperature was maintained at 350 °C, with a sheath gas flow rate of 35 arb and an auxiliary gas flow rate of 10 arb. The ion transport tube temperature was set at 320 °C. The scanning mode employed was a full scan of first-order mass spectrometry combined with an automatic trigger second-order mass spectrometry scanning mode (Full MS/dd-MS²). The first-order resolution was 35,000, while the second-order resolution was 17, 500. The ion scanning range extended from m/z 50 to 1,500, with collision energy gradients of 20, 40, and 60 eV being employed.

The raw data collected were imported into Compound Discoverer 3.2 software. Using the software's wizard setting and method template, an identification process for unknown compounds was established, and peak alignment and extraction were performed on the original data. To determine possible molecular formulas, the extracted molecular ion chromatographic peaks and isotope peaks were fitted. The measured spectra of secondary fragments were then compared with the mz Cloud network database and the local Chinese medicine component database, OTCML. The filtering parameters for the matching results were set as follows: first-order and second-order quality deviations of 5 ppm, and a matching score greater than 80. The filtered ions were compared to the compound information and reference substances in the database. Subsequently, the chemical constituents were further analyzed and identified using relevant literature and online databases such as PubChem, Human Metabolome Database, PubMed, and CNKI. Finally, the source of each component was attributed (Fu et al., 2021, Xing et al., 2023).

2.4.1 Gavage sample preparation

According to the drug standard issued by the State Food and Drug Administration of China, the recommended clinical oral dosage of AG is 12 g/d, while the intragastric dose for rats is determined based on the conversion coefficient between adult and rat dosages. Appropriate samples of AG were collected and dissolved in distilled water to obtain a concentration of 1 g/mL gastric juice. The gavage dose for rats was set at 10 times the equivalent daily dose of AG, corresponding to 10 g/kg. The control group received an equal volume of distilled water.

2.4.2 Preparation of the sample and standard solutions

The rats received the prescribed dose of liquid medicine or water orally twice a day over 3 consecutive days. Fasting was enforced for 12 h before the last administration, with access to drinking water was permitted. Subsequently, At 5, 15, 30, 60, 120, 240, and 360 min after the last administration, 0.5 mL of blood was collected from the orbital region and transferred into an EP tube containing heparin sodium. The blood samples were then centrifuged at 4 °C and 4000 rpm for 15 min. The resulting supernatant was collected, aliquoted, and stored at −80 °C. Prior to analysis, take 200 μL of serum from each rat at 7 time points and mix them together. Additionally, three times the amount of acetonitrile was added, followed by vortexing for 3 min to ensure thorough mixing. Centrifuge the mixture at 4 °C and 12000 rpm for 15 min. Transfer the supernatant to a clean EP tube, evaporate it with nitrogen gas at room temperature, and reconstitute the residue with 200 μL of 70 % methanol. Following vortexing for 3 min and sonication for 5 min, the mixture was again centrifuged at 4 °C and 12000 rpm for 15 min. The obtained supernatant was transferred to a glass insert in the liquid phase to yield the serum pharmacochemical test solution. The detection and analysis were performed using the UPLC-Q-Orbitrap HRMS analysis method detailed in section 2.3.3.

2.7.1 Sample and standard preparation

Each batch of AG was finely ground, and the resulting powder was accurately weighed to 2.0 g. The weighed powder was then transferred into a stoppered conical bottle. Next, 50 mL of a 50 % methanol solution was added to the conical bottle. The conical bottle was then weighed again before being subjected to ultrasonic extraction for 30 min. After cooling, any weight loss was compensated by adding additional 50 % methanol solution. The resulting solution was filtered using a 0.22 μm microporous membrane. The filtered solution served as the test solution for UHPLC-MS/MS analysis. The amounts of FTA, PO, VER, MG, HMP, CUL, LG, KOR, BI, RT and BER were accurately weighed and individually dissolved in methanol to prepare a single reference stock solution with a concentration of 0.197, 0.417, 0.426, 0.441, 0.075, 0.259, 0.102, 0.101, 0.122, 0.139 and 0.234 mg/mL, respectively.

2.7.2 Negative sample solution preparation

Negative control samples lacking Fructus Forsythiae, Rhizoma Anemarrhenae, Rhizoma Acori Tatarinowii, Radix Curcumae, Radix Rehmanniae, Herba Pogostemonis and Radix Rehmanniae, Herba Pogostemonis, Radix Isatidis, Angelica dahurica Tincture were prepared according to the preparation method of AG. The preparation method is as follows: The Rhizoma Anemarrhenae was extracted with 80 % ethanol by heating reflux for 3 times, 3 h each time, resulting in a concentrated paste. The remaining 8 herbs were extracted with water by heating reflux for 3 times (with collection of volatile oil), the first 1 h, the second and third 0.5 h each, resulting in a concentrated paste. The above two kinds of paste were mixed, and appropriate amount of dextrin and sodium cyclohexyl sulfamate were added to make granules. The volatile oil, patchouli oil (0.8 mL), peppermint oil (0.4 mL), and angelica dahurica Tincture (8 mL) were then mixed in. Finally, Negative control test solution were prepared according to 2.7.1.

2.7.3 Chromatographic conditions for UHPLC-MS/MS analysis

UHPLC-MS/MS analysis was performed using a Nexera UHPLC LC-30A/QTRAP 5500 + MS/MS system (Kyoto, Japan), equipped with an electrospray ionization (ESI) source (AB Sciex).

The chromatographic column used in the study was ACQUITY UPLC BEH C18 (2.1 × 100 mm, 1.7 μm; Waters, Massachusetts, USA), with a temperature of 35 °C. The mobile phase consisted of 0.1 % acetic acid in water (A) and acetonitrile (B). The gradient elution conditions were as follows: 0–1 min, 5 % B; 1–2 min, 50–15 % B; 2–4 min, 15–20 % B; 4–12 min, 20–80 % B; 12–13 min, 80 % B; 13–15 min, 80–5 % B; 15–16 min, 5 % B. The injection volume and flow rate were 5 μL and 0.4 mL/min, respectively.

The samples underwent positive and negative ion analysis in sMRM mode. Nitrogen served as the curtain, nebulizer, and heater gas, with pressures set at 35, 40, and 40 psi, respectively. The ion spray voltage was optimized to 5500 V, and the source temperature was set at 450 °C. Table 1 displays the optimal MRM parameters for the four analytes and IS. The data collection and analysis were performed using Analyst™ 1.6.3 software.

2.7.4 Method validation

The analysis was conducted in accordance with the guidelines outlined in the Chinese Pharmacopoeia 2020 and previous studies that specifically addressed the quantification of multi-component content in TCM (Commission, 2020, Song et al., 2023).

The mixed reference solution, test sample solution and each negative control sample solution were injected under the specified chromatographic and mass spectrometry conditions. The specificity was verified by comparing the chromatograms of mixed reference solution, test sample solution and negative control sample. The reference solutions were evaluated using the specified UHPLC-MS/MS conditions, which involved measuring concentration gradients. A standard curve was constructed using the mass concentration of the target compound (ng/mL) as the X-axis and the corresponding peak area as the Y-axis. The regression equation, correlation coefficient, and linear range were determined. The limit of quantification (S/N = 10) and the limit of detection (S/N = 3) for each analyte were also determined. The accuracy was assessed by performing six consecutive analyses of a sample solution. The reproducibility was evaluated through the analysis of six parallel sample solutions. The stability was estimated by analyzing repeated points (0, 2, 4, 8, 12, 24 h) from the same sample solution at different time intervals. The recovery rate was determined by weighing six parallel samples from the same batch with known content of each component and adding a reference solution with a similar mass concentration. The identical batch of sample solutions were injected and analyzed using various chromatographic columns (ACQUITY UPLC BEH C₁₈; UltiMate UHPLC AQ C₁₈; ACQUITY UPLC HSS C₁₈), column temperatures (31.5; 35; 38.5 °C), and acetic acid concentrations (0.09 %; 0.1 %; 0.11 %). The robustness was investigated by calculating the RSD value for each component's content tested under varied conditions.

2.8.1 Calculation of ECI

Based on the results of the NA inhibitor screening experiment, the reciprocals of the IC₅₀ values of each component were normalized to calculate the pharmacological weights of each component. The method for calculation is outlined in Formula (2).

In the formula, $W_i$ is the contribution of each component, ${\mathit{IC}}_{50i}$ is the IC₅₀ value of the active component, i is the name of the component, and n is the number of components.

The ECI is assessed using the comprehensive index evaluation method, as per the calculation method outlined in Formula (3).

In the formula, $Z_m$ is ECI of different batches of samples, $W_i$ is the efficacy weight of each component, and $X_i$ is the amount of each component in the sample.

2.8.2 Verification of ECI of AG

The NA inhibitory activity of 20 sample batches was assessed. The relative inhibition rate for each batch of AG samples were determined by comparing them to the reference inhibitory activity of FTA. To investigate the correlation between ECI and efficacy, Pearson's correlation coefficient (R) was used to calculate the correlation between ECI and the content of each component with the relative inhibition rate.

3.4.1 Optimization of the analysis conditions for UHPLC-MS/MS

The mass spectrometry parameters and chromatographic conditions of the target compounds were optimized in the early stage of this study to achieve the highest resolution chromatogram within a shorter timeframe. The precursor ions (Q1) and product ions (Q3) were selected based on the compound characteristics, while the collision energy (CE) and clustering voltage (DP) parameters were optimized based on the response values of the target ions. The detailed information is shown in Table 1. However, during the methodological investigation, it was found that the content of BC in the samples was extremely low, resulting in insufficient recovery and quantitative limits. Considering the clear dose–response relationship of NA inhibition, the extremely low content of BC would barely affect the overall inhibitory effect of the samples. Therefore, taking into account the measurability requirements of the Q-markers, it was determined to quantitate the other 11 Q-markers excluding BC. In addition, there is isomerism in the target compounds, and the detection of the target compounds mentioned above can be achieved by the retention time in chromatographic column separation. To enhance the separation process within a reasonable running time, various mobile phases (water/acetonitrile, water/methanol, 0.1 % formic acid water/acetonitrile, 0.1 % acetic acid water/acetonitrile) were used for analysis. The final elution conditions were 0.1 % acetic acid concentration in water/acetonitrile, with a flow rate of 0.4 mL/min gradient elution. Simultaneously, the optimal gradient elution time was adjusted to achieve optimal separation, symmetrical peak shapes, and ideal response for each target compound, aiming to reduce overall analysis time. Finally, 50 % methanol, 70 % methanol, methanol, 50 % ethanol, 70 % ethanol and ethanol were used as extraction solvents. The results showed that the extraction rate of Q-markers was the highest and the peak shape was better when 50 % methanol was used as extraction solvent.

3.4.2 Methodological validation

Following optimization of the detection conditions, specificity, the linearity, detection limit, quantification limit, precision, reproducibility, stability, recovery and robustness of the UHPLC-MS/MS method were evaluated. The chromatograms of the mixed reference solution and the sample solution in positive and negative ion modes are shown in Fig. 5. The chromatograms of each negative control sample solution are shown in Supplementary Fig.1. The peak shape of the 11 components tested in the sample solution was satisfactory and consistent with the retention time of the mixed reference substance. Each negative sample have no interference at the peak of the component to be tested, indicating that the method had good specificity. The correlation coefficients for each analyte within their respective concentration ranges were all greater than 0.992, indicating a strong linear relationship. The relative standard deviation (RSD) of the peak area for each component ranged from 1.16 % to 3.96 % after six repeated injections, demonstrating good instrument precision. Reference substance was added to the sample, and the recovery for each component ranged from 96.66 % to 105.04 %, validating the effectiveness of the sample preparation method. Six samples were prepared simultaneously, and the RSD of peak area for each component ranged from 0.42 % to 3.90 %, demonstrating the good repeatability of the preparation method. The results, as shown in Table 3, indicated that the peak area RSD for each component ranged from 2.33 % to 4.61 % at 0, 4, 8, 12, and 24 h post-preparation, indicating good stability of each component. The results of robustness test indicated that the RSD values for 11 components under various chromatographic columns, column temperatures and acetic acid concentrations were in the range of 1.84 % to 4.92 %, 1.62 to 4.62 % and 1.34 % to 4.24 %, respectively. This suggests that the method demonstrated good robustness within a specific range (Cui et al., 2023) (Table 3). The results of the methodological investigation are compliant with the provisions of the Chinese Pharmacopoeia, indicating that the method is sensitive, reliable, and capable of concurrently determining 11 Q-markers (Commission, 2020).

Close

Fig. 5

Multi-reaction detection chromatograms of 11 Q-markers standard (I) and sample (II) in positive (A) negative (B) ion mode.

Export to PPT

3.4.3 Analysis of the sample

In this study, UHPLC-MS/MS was employed to quantitatively analyze 11 Q-markers in 20 batches of AG. The results indicated that the samples had the highest contents of FTA and VER, ranging from 562.97 to 1829.26 μg/g. FTA is derived from Forsythia suspensa, a principal drug in the prescription. The content of this component showed similar results in a quantitative analysis of Forsythia suspensa (Chen et al., 2017b). VER is simultaneously derived from Rehmannia glutinosa and Pogostemon cablin, which could be one of the reasons for the high content of this component (Li et al., 2022a, Xie et al., 2022). Additionally, the sample also exhibits relatively high content of MG and BI. It is worth noting that MG is the primary active ingredient in Anemarrhena asphodeloides (Zhong et al., 2023), while BI is the active constituent found in Forsythia suspensa (Qi et al., 2021). The tested samples exhibited the lowest contents of BER and KOR (Table 4). The results of this study indicate variations in the levels of different Q-markers among the samples, which may potentially lead to diverse effects on the inhibition of NA activity. At the same time, the content of the same Q-marker in AG was also different among different batches, suggesting the necessity of improving its quality control level. Therefore, this quantitative analysis method is expected to provide a reference for the quality control of AG, ensuring its clinical efficacy and safety. However, further research is required to investigate the contribution of these components to the inhibitory effect of NA. Fig. 5 displayed the multi-reaction detection chromatograms of both the standard and sample.