A pharmacokinetic study on twenty-one compounds in rat plasma by integrating UPLC-QQQ-MS/MS with GC-MS after oral administration of Suxiao Jiuxin pill

2.1 Reagents, chemicals, and materials

LC-grade methanol and acetonitrile were purchased from Fisher Chemical (Fairlawn, OSU, USA). LC-grade formic acid was bought from Anaqua Chemicals Supply (Wilmington, DE, USA). Ultrapure water was produced by Milipore Milli-Q water purification system (Milford, MA, USA). Neochlorogenic acid, protocatechuic acid, chlorogenic acid, cryptochlorogenic acid, caffeic acid, vanillic acid, vanillin, scopoletin, isochlorogenic acid A, ferulic acid, isochlorogenic acid C, senkyunolide Ⅰ, senkyunolide H, senkyunolide A, neocnidilide, angelicide, levistolide A, 3-butylidenephthalide, astragalin (internal standard, IS₁) and α-terpineol (IS₂) were purchased from Chengdu Desite Bio-Technology Co., Ltd. (Chengdu, Sichuan, China). Ligustilide, isoborneol, and borneol were obtained from National Institutes for Food and Drug Control (Beijing, China). The purities of above standards were all more than 98 %. Suxiao Jiuxin pills (No.611323) were provided from Tianjin Pharmaceutical Da Ren Tang Group Corporation Limited NO.6 Traditional Chinese Medicine Factory (Tianjin, China).

2.2 UPLC-QQQ-MS/MS condition

Agilent 1290 UPLC system (Santa Clar, CA, USA) was employed in this work. A Waters ACQUITY UPLC®HSS T3 column (2.1 × 100 mm, 1.8 µm) was utilized to perform separation in a 23 ℃ thermostated column compartment. The mobile phase was composed of 0.1 % formic acid aqueous solution (A) and methanol (B). The gradient elution procedure was as follow: 0 ∼ 1 min, 15 ∼ 45 % B; 6 ∼ 9 min, 86 ∼ 95 % B; post run time was set as 5 min. The flow rate was set as 0.3 mL/min, injection volume was set as 5 µL.

Agilent 6470 QQQ-MS/MS system was equipped with an air jet spray electron spray ionization (AJS ESI) ion source. Positive and negative acquisition modes with using multiple reaction monitoring (MRM) were both utilized in this analysis. The ion source parameters were set as follow: drying gas, 300 ℃; drying gas flow rate, 7 L/min; nebulizer, 35 psi; sheath gas, 350 ℃; capillary voltage, 3.5 kV. The MRM ion transition parameters of eighteen targets and IS₁ were optimized (Table 1).

2.3 GC-MS condition

Shimadzu QP 2010 GC-MS (Kyoto, Japan) was equipped with an electron ionization (EI) ion source. The separation procedure was performed in a HP-5MS column (30 m × 0.25 mm × 0.25 μm). In the case of high-purity helium as carrier gas, the flow rate was set as 1.84 mL/min. The gradient heating procedure was as follow: the initial temperature was held at 110 ℃ for 4 min, and it was elevated to 140 ℃ at an increment of 40 °C/min, then elevated to 210 ℃ at an increment of 200 °C/min, thereafter, elevated to 230 ℃ at an increment of 31 ℃/min and remained for 1.2 min. The injection volume was set as 2 µL at a split ratio of 15:1.

For the mass spectrum module, the temperatures of EI ion source and the injector were set as 230℃ and 250℃, respectively. The solvent delay time was set as 5 min. The single ion monitoring (SIM) acquisition mode was applied for obtaining higher specificities of analyte (Table 2).

2.4 Preparation of standard and quality control (QC) samples

For the UPLC-QQQ-MS/MS analysis, the stock solutions of standard references were prepared by dissolving appropriate substance with methanol to a concentration of 1 mg/mL, respectively. The working solutions were prepared by mixing appropriate stock solution of each analyte and diluting into different concentrations with methanol. Calibration standard solutions were prepared by spiking with 20 µL working solution, 20 µL IS₁ (1000 ng/mL) and 20 µL formic acid into 100 µL blank plasma, subsequently, obtained mixture was extracted by the method same with “2.5. Sample Pretreatment”. The final concentrations of calibration standard solutions were at the ranges of to the concentration ranges of 0.25 ∼ 100 ng/mL (neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, caffeic acid, vanillic acid, vanillin, scopoletin, isochlorogenic acid A, isochlorogenic acid C, senkyunolide H, angelicide, levistolide A), 1.0 ∼ 400 ng/mL (protocatechuic acid), 4 ∼ 1600 ng/mL (ferulic acid, ligustilide), 2 ∼ 800 ng/mL (senkyunolide Ⅰ), 5.5 ∼ 2200 ng/mL (senkyunolide A) and 0.35 ∼ 140 ng/mL (neocnidilide), respectively. The QC samples were prepared in the same way at three concentration levels (low, medium, high). All the samples were stored in 4℃ before analysis.

For the GC-MS analysis, the stock solutions of standard references were prepared by dissolving appropriated substance with n-hexane to a concentration of 1 mg/mL, respectively. The working solutions were prepared by mixing appropriate stock solutions of each analyte and diluting into different concentration with n-hexane. Calibration standard solutions were prepared by spiking with 10 µL working solution and 10 µL IS₂ (5000 ng/mL) into 100 µL blank plasma, subsequently, obtained mixture was extracted by the method same with “2.5. Sample Pretreatment”. The final concentrations of calibration standard solutions were at the ranges of 12.5 ∼ 4000 ng/mL (isoborneol and 3-butylidenephthalide) and 25 ∼ 8000 ng/mL (borneol), respectively. The QC samples were prepared in the same way at three concentration levels (low, medium, high). All the samples were stored in 4 ℃ before analysis.

2.5 Sample preparation

For UPLC-QQQ-MS/MS analysis, 20 µL IS₁ (1000 ng/mL), 20 µL formic acid and 20 µL methanol were successively added into 100 µL plasma and vortexed for 1 min. The analytes extraction was operated by protein precipitation method with 800 µL acetonitrile. The mixture was vortexed for 5 min and centrifuged for 10 min at 14000 rpm. Thereafter, the supernatant was dried under flow nitrogen gas. The residue was dissolved with 100 µL methanol and vortexed for 5 min. The reconstituted mixture was centrifuged for 10 min at 14000 rpm. A liquid of 5 µL supernatant was injected into UPLC-QQQ-MS/MS system for analysis.

For GC-MS analysis, 10 µL IS₂ (5000 ng/mL) and 10 µL n-hexane were successively added into 100 µL plasma and vortexed for 1 min. The analytes extraction was operated by liquid–liquid extraction method with 130 µL n-hexane. The mixture was vortexed for 5 min and centrifuged for 10 min at 14000 rpm. A liquid of 2 µL n-hexane fraction was injected into GC-MS system for analysis.

2.6 Method validation

The specificity, linearity, sensitivity, precision and accuracy, recovery, matrix effect and stability were evaluated for ensuring a high reliable quantitative method according to the US Food and Drug Administration (Xu et al., 2019, Tang et al., 2020, Zhang et al., 2022). Specificity was evaluated by comparing the chromatograms among blank plasma, medium concentration level QC sample and the plasma at 5 min after oral administration of SX. The linearity of calibration curve was plotted by using the peak areas of the analytes to IS versus concentration with a 1/X² as weighting factor. The sensitivity was assessed by limit of quantitation (LOQ) that concentration at signal-to-noise (S/N) of 10. Six replicates of QC samples at each concentration level on the same day and three consecutive days were prepared and determined. The relative standard deviation (RSD) and relative error (RE) were calculated for expressing the intra- and inter-day precision and accuracy, respectively. The recovery was investigated by comparing the peak area of analytes in pre-treatment QC samples with corresponding pos-treatment spiked sample. The matrix effect was calculated by the peak area ratio of analytes extracted form QC samples to the equal concentration standard solution. The stability was evaluated by QC samples stored in room temperature for 4 h, autosampler for 12 h, −80℃ refrigerator for 7 days and three cycles of freeze and thaw, respectively.

2.7 Pharmacokinetic study

The male Sprague–Dawley rats (220 ± 10 g) were purchased from Beijing Huafukang Bio-Technology Co., Ltd. (Beijing, China). The animals were housed in 40 ∼ 60 % humidity, 23 ∼ 27 ℃, and 12 h dark-light cycle at the animal center of Traditional Chinese Medicine of Tianjin University. Free access to water and food were permitted to animals until 12 h before the experiment. All animals were randomly divided into two groups for LC-MS/MS and GC-MS analysis, respectively. The rats were received SX at the dosage of 600 mg/kg (dissolved by pure water to the concentration of 80 mg/mL) by oral administration after fasted for 12 h. The contents of analytes in SX were determined by the established UPLC-QQQ-MS/MS integrated with GC-MS method (Table 3). A liquid of 200 µL blood sample was obtained from ophthalmic venous plexus into heparinization centrifuged tube at pre-dose, 0.03, 0.08, 0.17, 0.25, 0.33, 0.5, 0.75, 1 h, 2 h, 4 h, 6 h, 10 h and 24 h post dosing, respectively. All blood samples were directly centrifuged at 7000 rpm for 10 min. The plasma layer was then transferred into clean tube and stored at −80℃ before use. All animal studies were conducted under the guidance of Laboratory Animal Ethics Committee of Tianjin University of Traditional Chinese Medicine (TCM-LAEC2022028).

2.8 Data analysis

The pharmaceutic parameters were calculated by Drug and Statistics software (DAS, version 3.0, Medical College of Wannan, China) with a non-compartment model.

3.1 Optimization of UPLC-QQQ-MS/MS conditions and sample preparation

The stationary phase was extremely crucial for getting a satisfactory separation. The main chemical regredient of most TCMs were naturally organic products which were easily absorbed by reverse-phase stationary. For this work, two commonly used reverse stationary, Waters ACQUITY UPLC®BEH C18 (2.1 × 100 mm, 1.7 μm) and Waters ACQUITY UPLC®HSS T3 (2.1 × 100 mm, 1.8 µm), were utilized to optimize the separation of eighteen analytes. Comparatively, the Waters ACQUITY UPLC®HSS T3 chromatographic column showed a better retained performance. It was able to separate analytes within a short time. Besides, the mobile phase was also a key factor affecting the peak shape and resolution. Subsequent optimization was done by changing the kinds of mobile phase A (water and 0.1 % formic acid aqueous solution) and mobile phase B (methanol and acetonitrile). After adjusting the gradient elution program, a satisfactory separation was obtained using 0.1 % formic acid aqueous solution and methanol as mobile phase (Fig. 2).

Close

Fig. 2

The MRM chromatograms of eighteen analytes and IS₁ in blank plasma (A), blank plasma spiked with medium concentration QC (B), and plasma sample at 5 min after oral administration Suxiao Jiuxin pill (C).

Export to PPT

Protein precipitation and liquid–liquid extraction methods were both investigated in this experiment. Meanwhile, in view of the carboxyl group of phenolic acid, which led to a strong absorption with plasma protein, the formic acid was applied in phenolic acid analytes extraction (Sun et al., 2013, Huang et al., 2017). The result showed acetonitrile possessed a higher extraction property on all analytes than methanol and ethyl acetate. For the ionization of ESI mode, the complex matrix of plasma could affect target compounds response (Ismaiel et al., 2008, Yadav et al., 2012). The optimization of redissolving reagent was a direct and effective method to minimize the interference from endogenous ingredients. As a result, methanol showed more pleasing result than acetonitrile and the mixture of methanol and acetonitrile (v:v, 1:1).

3.2 Method validation

The specificity has been investigated by comparing the chromatograms among blank plasma, blank plasma spiked with medium concentration QC, and plasma sample at 5 min after oral administration of SX. Obviously, there were no significant interferences at the same retention time of analytes and internal standards (Figs. 2 and 3). It indicated a good specificity of these detection methods.

Close

Fig. 3

The SIM chromatograms of three analytes and IS₂ in blank plasma (A), blank plasma spiked with medium concentration QC (B), and plasma sample at 5 min after oral administration Suxiao Jiuxin pill (C).

Export to PPT

The calibration curves, correlation coefficients (R), linear ranges and LOQs of twenty-one analytes were exhibited in Table 4. All the calibration curves of analytes possessed fine correlation coefficients more than 0.995 within a certain concentration range. The LOQs of all analytes were as low as 12.5 ng/mL.

The QC samples at three concentration levels were used to assess the extraction recovery and matrix effect (Table5). The extraction recoveries of analytes were at the range of 61.2 ∼ 107.0 % with RSDs no more than 11.7 %. It suggested that the extraction method was parallel to obtain analytes from plasma within a wide concentration range. For the matrix effect analysis, excepted neochlorogenic acid, protocatechuic acid, chlorogenic acid, vanillic acid and scopoletin, other analytes did not possess obvious endogenous interference that matrix effect ranged from 70.7 % to 114.8 % with RSDs less than 13.7 %. For the chemical ingredients with matrix interferences, their RSDs were less than 13.7 %, which indicating a stable endogenous interference in the sample solution. In addition, the acceptable LOQs (less than 1 ng/mL) suggested the interference was not sufficient to impede their highly sensitive determination.

The intra- and inter-precision and accuracy were assayed by six replicates of QC samples at three concentration levels on the same day (intra-day) and after three consecutive days (inter-day). The calibration curve constructed on the same testing day were used for assessment. The ranges of intra- and inter-precision were 0.4 ∼ 7.9 % and 0.8 ∼ 9.0 %, respectively, with accuracy within ± 12 % for all QC samples (Table 6).

The results of the stability stored in room temperature for 4 h, autosampler for 12 h, −80℃ refrigerator for 7 days, and three cycles of freeze and thaw were listed in Table 7. All values were within the acceptable range of less than 12 %.

3.3 Pharmacokinetic study

The validated UPLC-QQQ-MS/MS integrated with GC-MS method was employed to quantify twenty-one compounds in rat plasma after oral administration of SX. The mean plasma concentration – time curves were visualized by GraphPad Prism 8.0.2 software (Fig. 4). Unfortunately, there were some compounds whose plasma concentration was too low to describe an intact pharmacokinetic profile, such as neochlorogenic acid, protocatechuic acid, chlorogenic acid, etc. It might be caused by their low content in Suxiao Jiuxin Pill. Ultimately, the pharmacokinetic parameters of nine analytes were analyzed by DAS software with non-compartment model (Table 8).

Close

Fig. 4

The plasma concentration–time curves of analytes (n = 6, mean ± SD).

Export to PPT

The maximum plasma concentration (C_max) was tightly bound to the clinical application of drug. Only when the plasma concentration reached therapeutic concentration could it take the effect. The C_max from high to low were borneol (683.35 μg/L), isoborneol (233.61 μg/L), senkyunolide A (200.51 μg/L), senkyunolide Ⅰ (96.42 μg/L), ligustilide (88.74 μg/L), ferulic acid (50.07 μg/L), neocnidilide (13.79 μg/L), senkyunolide H (12.49 μg/L), and levistolide A (5.24 μg/L). Area under concentration–time curve (AUC _(0-tn) and AUC _(0-∞)) reflected the total exposure of analytes in plasma. The AUCs of senkyunolide Ⅰ, senkyunolide A, ferulic acid, ligustilide, borneol and isoborneol were greater than 47.2 μg/L*h, which suggesting they possessed high exposure in vivo. It might be due to the higher content in SX. The time about reaching C_max (T_max) of all analytes were within 0.4 h, which indicated the active ingredients in SX could be rapidly absorbed into blood and build up to therapeutic concentration. Half-life (T_1/2z) represented the rate of drug clearance in vivo. All analytes possessed short T_1/2z values ranging from 0.29 to 1.24 h. The rapid absorbed and distributed would provide a rapid effect on acute cardiovascular diseases. What’s more, it could be found that the T_max and T_1/2z values of borneol and isoborneol, ligustilide and neocnidilide were closely similar, but their doses were quite different. The reason for this phenomenon was probably due to their similar chemical structure.