A Comparative Pharmacokinetic Study by UHPLC-MS/MS of Main Active Compounds after Oral Administration of Zushima-Gancao Extract in Normal and Adjuvant-Induced Arthritis Rats

A sensitive and rapid ultra high-performance liquid-chromatography tandem mass spectrometry (UHPLC-MS/MS) method has been applied to investigate the influence of rheumatoid arthritis (RA) on the pharmacokinetics of nine analytes (daphnetin, daphnoretin, 7-hydroxycoumarin, liquiritin, isoliquiritin, liquiritigenin, isoliquiritigenin, glycyrrhizin, and glycyrrhetinic acid), which are major active components in Zushima-Gancao extract. The analytes and internal standard (IS) were separated in a Hypersil Gold C18 column and detected on a triple-stage quadrupole mass spectrometer using the validated method. All analytes exhibited good linearities (R2 > 0.98), and the lower limit of quantification (LLOQs) were sufficient for quantitative analysis. Intra- and inter-batch precision were all within 14.96% while the accuracy of nine analytes ranged from −17.99 to 14.48%, and these results were all within acceptance criteria. The extraction recoveries, matrix effects, and stabilities were all satisfactory. Main pharmacokinetic parameters of each compound were compared, and significant differences were found in parameters of daphnetin, daphnoretin, liquiritin, isoliquiritin, isoliquiritigenin, glycyrrhizin, and glycyrrhetinic acid, especially the last one, between the two groups. Therefore, adjuvant-induced arthritis has different effects on the pharmacokinetics of ingredients in Zushima-Gancao extract. The comparative pharmacokinetic study between normal and adjuvant-induced arthritis rats might provide more comprehensive information to guide the clinical usage of Zushima-Gancao extract for treating RA.


Introduction
Rheumatoid arthritis (RA) is a chronic systemic disease accompanied by the destruction of joints, but its etiology has remained unknown [1]. Western medicine, including nonsteroidal anti-inflammatory drugs (NSAIDS) [2] and biologics [3], has greatly improved the clinical efficacy of medicine acting against RA. However, the adverse reactions and toxicities associated with these drugs [4][5][6] have promoted the development of traditional Chinese herbal medicine (TCHM), such as

Linearity and LLOQ
The results of the calibration curve, linear range, correlation coefficient (R 2 ), and LLOQ of each compound are shown in Table 1, and all analytes exhibited good linearity within the test ranges.

Accuracy and Precision
The accuracy and precision results are summarized in Table 2. The intra-batch and inter-batch precision of the nine analytes were all in the range of 2.12-14.96% and 1.92-13.63%. The accuracy values ranged from −14.96 to 14.48% for intra-batch and from −17.99 to 14.44% for inter-batch. The results demonstrated that the accuracy and precision of the nine analytes were well within the acceptable limit.

Recovery and Matrix Effect
The results showed that the recoveries of each ingredient at low, medium, and high concentrations were over 60%, even if the recovery of IS was slightly lower. It was proved that the liquid-liquid extraction with ethyl acetate could be applied for extracting measured components.
Matrix effect values of nine analytes ranged from 85.54 to 114.66%, and the matrix effect of IS was 96.23%. The results indicated that the endogenous matrix peaks did not affect the quantification of all the analytes and IS. Recovery and matrix effect data of nine analytes and IS in rat plasma are shown in Table 3.

Stability
The results of stability are shown in Table 4, and all ingredients were stable at 4 • C for 24 h, at room temperature for 4 h, at −80 • C for a week, and three freeze-thaw cycles.

Pharmacokinetics
The validated method was applied to the pharmacokinetic study of DPH, DR, HDC, LQ, ILQ, LG, ILG, GL, and GA in normal and AA rats after oral administration of Zushima-Gancao extract. The plasma concentration-time curves (C-T curves) of nine analytes are presented in Figure 2, and the corresponding pharmacokinetic parameters, including C max , T max , AUC (0-t), AUC (0-∞) (AUC: area under concentration-time curve), and t 1/2 , were obtained using DAS 3.0 pharmacokinetic software (Chinese Pharmacological Association, China). Statistical differences between groups were analyzed using Student's t-tests, and p-values less than 0.05 were considered statistically significant. All data are summarized in Table 5.
According to the pharmacokinetic results, it was found that the plasma profile of LG and ILG presented bimodal (double peak) phenomenon after oral administration of Zushima-Gancao extract in both normal and AA rats. This phenomenon was because LG, as well as ILG, were conjugated with glucuronic acid in the liver, and the conjugations were hydrolyzed in the intestinal tract and then re-absorbed into the blood [32]. Interestingly, there was a similar tendency in the pharmacokinetic profiles of three pairs of structurally similar compounds (i.e., DPH and HDC, LQ and ILQ, and LG and ILG), which possibly exhibited the same absorbing and metabolic features.  Furthermore, it was noticed that the pharmacokinetic parameters of components in AA rats were significantly different from those in normal rats, especially the AUC (0-∞) of DPH in Zushima and four compounds (LQ, ILQ, GL, and GA) in Gancao. Compared with the normal group, the AUC (0-∞) of DPH, LQ, ILQ, and GL obviously decreased by 34.56%, 24.57%, 39.02%, and 29.89%,  Furthermore, it was noticed that the pharmacokinetic parameters of components in AA rats were significantly different from those in normal rats, especially the AUC (0-∞) of DPH in Zushima and four compounds (LQ, ILQ, GL, and GA) in Gancao. Compared with the normal group, the AUC (0-∞) of DPH, LQ, ILQ, and GL obviously decreased by 34.56%, 24.57%, 39.02%, and 29.89%,  Table 5. The pharmacokinetic parameters of DPH, DR, HDC, LQ, ILQ, LG, ILG, GL, and GA in normal and AA rats after oral administration of Zushima-Gancao extract to rats (n = 6, mean ± SD). Furthermore, it was noticed that the pharmacokinetic parameters of components in AA rats were significantly different from those in normal rats, especially the AUC (0-∞) of DPH in Zushima and four compounds (LQ, ILQ, GL, and GA) in Gancao. Compared with the normal group, the AUC (0-∞) of DPH, LQ, ILQ, and GL obviously decreased by 34.56%, 24.57%, 39.02%, and 29.89%, respectively. The parameters also showed that the C max of DR, LQ and ILQ decreased remarkably (p < 0.05). On the contrary, the bioavailability of LG and GA increased; the AUC(0-t) and AUC (0-∞) of GA especially were both increased 2.2-fold in the model rats compared to those in normal rats. In our previous investigation, the in vitro concentration of DPH in Zushima-Gancao extract was 6-fold compared to DR and 22-fold compared to HDC, but DPH had the worst in vivo utilization according to the AUC of the three coumarins. That was probably because DPH was partially metabolized into HDC. In terms of GA, some previous published papers have reported that GL was hydrolyzed to GA by Ruminococcus sp.POI-3, Eubacterium sp., and β-D-glucuronidase in the intestinal tract [33]. The possible explanation for the difference between two groups could be that the characteristics of intestinal flora have changed in AA rats, and more GL was metabolized into GA.

DPH
The t 1/2 value of DPH exhibited a remarkable decrease in AA rats; meanwhile, an obvious increase of T max values for ILQ, ILG, and GL were observed in AA rats compared with normal rats. This phenomenon revealed that the disease had different effects on the pharmacokinetic behaviors of different compounds.
Mass spectrometry detection was operated on TSQ Vantage triple-stage quadrupole mass spectrometer (Thermo Fisher, San Jose, CA, USA) equipped with an electrospray ionization (ESI) source. Quantification of the analytes was performed using selective reaction monitor (SRM) in the negative ionization mode. Spray voltage, sheath gas pressure, auxiliary gas pressure, vaporizer temperature, and capillary temperature were set as 3.0 kV, 45 arb, 25 arb, 450 °C, and 350 °C, respectively. The optimized SRM parameters of nine analytes and IS are listed in Table 6.
Mass spectrometry detection was operated on TSQ Vantage triple-stage quadrupole mass spectrometer (Thermo Fisher, San Jose, CA, USA) equipped with an electrospray ionization (ESI) source. Quantification of the analytes was performed using selective reaction monitor (SRM) in the negative ionization mode. Spray voltage, sheath gas pressure, auxiliary gas pressure, vaporizer temperature, and capillary temperature were set as 3.0 kV, 45 arb, 25 arb, 450 • C, and 350 • C, respectively. The optimized SRM parameters of nine analytes and IS are listed in Table 6.

Preparation of Zushima-Gancao Extract
For preparing Zushima-Gancao extract, the herbal materials composed of Zushima (162 g) and Gancao (108 g) were crushed into small pieces and mixed, and the mixture was then soaked in water (1:10, w/v) for 0.5 h before decocting for 2 h. The filtrates were collected and the residues were then refluxed in water (1:5, w/v) for 1 h. Zushima-Gancao extract (equal to 0.162 g/mL of Zushima and 0.108 g/mL of Gancao) could be obtained by mixing the two-stage filtrates and concentrating the volume to 1000 mL. The concentrations of DPH, DR, HDC, LQ, ILQ, LG, ILG, and GL in prescription were 220.60, 36.56, 10.25, 507.30, 56.48, 42.03, 3.39, and 1006.00 µg/mL respectively. GA was not detected in the extract.

Preparation of Standard Solutions and Quality Control (QC) Samples
Stock solutions were separately dissolved in methanol to concentrations of 1.27 mg/mL for DPH, 1.56 mg/mL for DR, 1.03 mg/mL for HDC, 1.20 mg/mL for LQ, 1.02 mg/mL for ILQ, 1.10 mg/mL for LG, 1.05 mg/mL for ILG, 1.08 mg/mL for GL, and 1.12 mg/mL for GA, and they were then serially diluted with methanol to prepare working solutions with 10 standards of desired concentrations. Hesperidin(IS) was dissolved in 0.5 mL of DMSO and diluted with methanol to achieve a concentration of 130 ng/mL. All the solutions were stored at 4 • C.
Quality control (QC) samples were obtained in the same way as calibration standards with blank plasma in low, middle, and high concentrations (3.29,
Rats were randomly divided into normal and AA groups with six rats in each group. The arthritic model rats were prepared as our previous research [19]. AA rats were intracutaneously injected with 0.1 mL of FAC in the left hind footpad and normal rats were injected with sterile saline. The FAC solution was prepared as follows: bacillus calmette-guerin (BCG) was bathed in water at 80 • C for 30 min, and then dissolved in a liquid paraffin-lanolin mixture (3:2, v/v). Immediately prior to use, the FAC solution were sterilized for 30 min at 120 • C.

Drug Administration and Blood Sampling
After model construction, the rats were fed for 20 days. All rats were fasted for 12 h before drug administration, with free access to water. Each rat was administered with Zushima-Gancao extract at a single dose of 10 mL/kg. Blood samples (approximately 0.3 mL) were collected into heparinized micro-centrifuge tubes from the postorbital venous plexus veins of rats at specific time points of 0, 0.08, 0.17, 0.33, 0.5, 1, 2, 3, 5, 8, 12, 24, and 36 h. All blood samples were centrifuged immediately at 5000 rpm for 6 min, and the plasma were obtained and stored at −80 • C until analysis.

Plasma Sample Preparation
A liquid-liquid extraction (LLE) method was applied to extract the compounds from thawed rat plasma samples, aliquots of 100 µL plasma sample were transferred into a 1.5 mL Eppendorf tube, and 10 µL of methanol, 10 µL of hydrochloric acid (2 mol/L), and 10 µL of IS (130 ng/mL) were added separately. The mixture was vortexed for 30 s, and 500 µL of ethyl acetate was added and mixed by vortexing for 10 min. The supernatant (about 450 µL) was transferred into another tube after centrifugation at 16,000 rpm for 10 min and then evaporated to dryness at 45 • C, 15 kPa in a Thermo SPD1010-230 SpeedVac Concentrator (Thermo Fisher, San Jose, CA, USA). The residue was dissolved in 100 µL of initial mobile phase by vortexing for 5 min, and a further centrifugation at 17,000 rpm for 10 min was applied after the residue was re-dissolved. The supernatant was injected into the UHPLC-MS/MS system for analysis.

Method Validation
The method was validated in accordance with The Guidance for Industry-Bioanalytical Method Validation for specificity, lower limit of quantification (LLOQ), linearity, accuracy, precision, recovery, matrix effect, and stability.
Specificity. The specificity was assessed to ensure no endogenous interference. It was carried out by comparing the chromatograms of blank plasma samples, blank plasma samples spiked with mixed standards and IS, plasma samples obtained from normal rats, and plasma samples obtained from AA rats collected from different rats (six in each group) after administration of Zushima-Gancao extract.
Linearity and LLOQ. The samples for calibration curves were prepared by spiking the blank rat plasma (100 µL) with 10 µL of the working solutions at different concentrations before extracting the analytes. The linearities of the calibration curves were determined by plotting the peak-area ratios of analytes/IS (Y) against concentration (X) with a weighted (W = 1/X) linear least squares regression model. The lower limit of quantification (LLOQ) was defined as the lowest concentration point of the standard curve with a signal-to-noise (S/N) ratio >10 and a precision and accuracy within ±20%.
Precision and accuracy. The QC samples and the sample at the concentration of LLOQ were analyzed with six replicates at each concentration to determine intra-batch and inter-batch accuracy (expressed as relative error, RE%) and precision (expressed as relative standard deviation, RSD%) in the same day or over three consecutive days.
Extraction recovery and matrix effect. Extraction recovery tests of nine analytes and IS were performed by comparing the peak areas of QC samples and IS added prior to extraction of the plasma samples with the peak areas of pure standards at the same concentration spiked with extracted plasma. The matrix effect was evaluated by comparing the peak areas of post-extracted blank samples spiked with analytes at three concentration levels, as well as IS with the same concentration of analytes, and IS dissolved in initial mobile phase.
Stability. To evaluate the stability of each component during analysis, QC samples at low, middle, and high concentrations (three samples for each concentration) were tested under different conditions, including placing QC samples in the autosampler at 4 • C for 24 h, at room temperature for 4 h, at −80 • C for a week, and three freeze-thaw cycles.

Conclusions
Based on a sensitive and rapid UHPLC-MS/MS method, nine analytes, three coumarins in Zushima and six compounds in Gancao, were simultaneously determined in normal and AA rat plasma samples after oral administration of Zushima-Gancao extract. The pharmacokinetic parameters between the two groups were compared and the results suggested that the pathological state could influence the pharmacokinetic characteristics of compounds in prescription. The bioavailabilities of DPH, LQ, ILQ, and GL were significantly decreased and GA increased when the rats were in a pathological state. In general, AA is a suitable model for simulating organisms with rheumatoid arthritis, and the different pharmacokinetics of nine ingredients between normal and AA rats revealed that the bioavailability of drugs is influenced by the internal environment when organisms are in a pathological state.