An effective UFLC–MS/MS method used to study pharmacokinetics of major constituents of Fukeqianjin formula in rat plasma

Moghaniae Radix is the dried roots of Moghania macrophylla (Willd.) O. Kuntze and was obtained from Xishuangbanna Dai Autonomous Prefecture of Yunnan province of China. Rosae Laevigatae Radix is the dried roots of Rosa laevigata Michx. and was obtained from Yongshun county of Hunan province of China. Andrographis Herba is the dried aerial parts of Andrographis paniculata (Burm. f.) Nees and was gathered from Suixi county of Zhanjiang city in Guangdong province of China. Mahoniae Caulis is the dried stems of Mahonia bealei (Fort.) Carr. and was purchased from Yunnan province of China. Zanthoxyli Caulis is the dried stems of Zanthoxylum dissitum Hemsl. and was purchased from Yanling county of Zhuzhou city in Hunan province of China. Angelicae Sinensis Radix is the dried roots of Angelica sinensis (Oliv.) Diels and Codonopsis Radix is the dried roots of Codonopsis pilosula (Franch.) Nannf., both were purchased from Longxi county of Gansu province of China. And Spatholobi Caulis is the dried vine stem of Spatholobus suberectus Dunn and was harvested from Pu’er city, Yunnan province of China. All the crude drugs were identified by Prof. Xiu-Wei Yang at the School of Pharmaceutical Sciences, Peking University Health Science Center, Peking University. Voucher specimens were deposited at the herbarium of Zhuzhou Qianjin Pharmaceutical Co., Ltd. (Hunan, China) and school of Pharmaceutical Sciences, Peking University (Beijing, China).

Reference compounds including 3,4,5-trimethoxyphenol-1-O-β-d-glucopyranoside (1), 5-hydroxypicolinic acid methyl ester (2), acortatarin A (3), 9-epi-acortatarin A (4) were isolated from Mahoniae Caulis [14], neoandrographolide (5), trans-ferulic acid (6), genistin (7), salicylic acid (8), jatrorrhizine (9), palmatine (11), andrographolide (12), berberine (13), 14-deoxy-11,12-didehydro-andrographolide (15), panicolin (16), andrograpanin (17), Z-ligustilide (18), and 7-O-methylwogonin (19) were isolated from the extracts of Fukeqianjin Formula as described in our previous report [7], ononin (10) and naringenin (14) were isolated from Spatholobi Caulis [15]. The internal standard (I.S.) bavachin was isolated from the mature fruits of Psoralea corylifolia L. [16]. Their purity was determined to be above 98% by LC–MS and the chemical structures are shown in Fig. 1.

Heparin sodium injection (Lot# 20150602) was purchased from Tianjin Biochemical Pharmaceutical Co., Ltd. (Tianjin, China). Methanol (MeOH), acetonitrile (ACN) and formic acid of LC–MS grade were provided by Fisher Scientific Inc. (Pittsburgh, PA, USA). Deionized water (18.2 MΩ cm) was purified by Millipore Alpha-Q water purification system (Millipore, Bedford, MA, USA).

Rat experiment was carried out based on following the guideline of the Care and Use of Laboratory Animals in Beijing that was approved by Animal Care and Use Committee of Peking University (Approval No. LA 2018143, 1 March 2018). Male Sprague-Dawley (SD) rats with the weight of 220 ± 20 g, were provided by Laboratory animal center of Peking University Health Science Center. These rats were raised in a standard environment with temperature of 24 ± 2 °C, relative humidity of 60 ± 5%, and LD 12:12 cycles. They could acclimatize the environment for a week before the experiment, free drinking water and conventional feed were normally supplied during this period.

FKQJF extract was prepared and freeze-dried by using standard method as described previously [1] and obtained from Zhuzhou Qianjin pharmaceutical Co., Ltd. Chiefly, 21 g dried FKQJF extract was added into moderate amount of distilled water to dissolved completely by ultrasonic treatment, then diluted to get a suspension of 0.42 g/mL.

Aliquots of 100 μL plasma samples were mixed with 10 μL I.S. samples (1 μg/mL, in MeOH), then added with 500 μL ACN, respectively, and vortexed for 5 min, centrifugated at 10,000 rpm for 10 min, the supernatants were transferred into clean vials and dried in a vacuum centrifugal concentrator at 1200 rpm, 37 °C. After that, 100 μL ACN was added and ultrasonic (frequency 40 kHz, power 250 W) extracted for 5 min to dissolve the dry residue, vortexed for 1 min and centrifugated at 10,000 rpm for 5 min, 1 μL of each final supernatant was injected into UFLC–MS/MS for analysis.

An 8050 UFLC–MS/MS system equipped with LC-30A binary pump, a SIL-30AC autosampler, a SPD-M30A PDA detector, a CTO-20AC column oven, and a 8050 triple quadrupole mass spectrometer with an electrospray ionization source (ESI) was applied to detect and quantify FKQJF compounds in plasma samples. Data were acquired and processed with the software LabSolution (Ver. 5.6).

A Kinetex XB-C₁₈ HPLC column (100 × 2.1 mm, 2.6 μm) linked with an equivalent guard column was used for chromatographic separation. The aqueous phase contained 0.2% formic acid was used as mobile phase A, and organic phase ACN as mobile phase B. The gradient elution method using in this program was: 3–5 min, 28–31% B; 5–7 min, 31–34% B; 7–9 min, 34–65% B; 9–10 min, 65–95% B; 10–11.5 min, 95% B. The flow rate of mobile phase, column temperature and autosampler temperature was 0.4 mL/min, 30 °C and 4 °C, respectively. The injection volume of each sample was set as 1 μL. The main MS spectrum parameters were set as follows: the flow rate of drying gas (N₂) was 10.0 L/min; the atomized gas was 3.0 L/min and the heating gas flow rate was 10.0 L/min. The main temperatures of interface, desolvation and heat block was 300 °C, 250 °C and 400 °C, respectively. And, the interface and detector voltage were 3 kV and 1.8 kV. The specific parameter values corresponding to each compound were shown in Table 1.

The analysis method established in this experiment was systematically examined for specificity, linear, precision, accuracy, stability, recovery and matrix effect in accordance with guidelines for the validation of quantitative analytical methods for biological samples in Guidance for Industry: Bioanalytical Method Validation [17].

Data were analyzed using the Drug and Statistics (DAS) Software version 2.0 (DAS 2.0, Mathematical Pharmacology Professional Committee of China, Shanghai, China). The non-compartmental mode was applied to obtain the following PK parameters: area under the plasma concentration–time curve from time zero to the time of last quantifiable concentration (AUC_0→t) and area under the plasma concentration–time curve to time infinity (AUC_0→∞), the elimination half-life (t_1/2), mean retention time to last sampling time (MRT_0→t) and mean retention time to infinity (MRT_0→∞). The maximum plasma concentration (C_max) and the time to maximum concentration (T_max) were observed directly from the concentration–time (C–T) curves and measured data. All data are presented as mean ± SD.

Plasma samples were prepared according to the method in section of “Pretreatment of plasma samples” and analyzed by using the established method. The blood concentration–time curves of the compounds were shown in Fig. 3, and the pharmacokinetic (PK) parameters were shown in Table 6.

The results in Fig. 3 show that all the 19 compounds could be rapidly absorbed by the gastrointestinal tract after the extract of FKQJF were orally administered to rats at a single dose of 3.2 g/kg body weight, the concentration of which in plasma could be determined about 10 min later, and the plasma drug concentrations of most compounds reached the peak values at about 1–2 h. Two compounds, berberine (13) and andrograpanin (17), displayed peak times longer than 3 h (the T_max were 8.00 ± 0.00 and 3.33 ± 2.31, respectively), indicating the gastrointestinal absorption of these two compounds was lower than that of other ones. The plasma concentrations of other compounds, 5-hydroxypicolinic acid methyl ester (2), acortatarin A (3), 9-epi-acortatarin A (4), jatrorrhizine (9), palmatine (11), naringenin (14) and 7-O-methylwogonin (19) decreased rapidly with the time prolongation after reaching peak values but could still be determined in plasma after a period. This result could be inferred that the absorption of this kind of compounds has the characteristics of fast absorption, fast elimination and long residence time. C_max and AUC of 5-hydroxypicolinic acid methyl ester (2), salicylic acid (8), jatrorrhizine (9), palmatine (11) and berberine (13) were higher than the other compounds. However, in our previous study the oral absorption (C_max) of alkaloids (jatrorrhizine, palmatine and berberine) was relatively low, possibly due to complex drug interaction in a Chinese medicine formula [18]. By comparing the pharmacokinetic characteristics of acortatarin A (3) and 9-epi-acortatarin A (4), although the half-life of both was close (t_1/2 was 2.66 and 2.20 h respectively), the AUC and C_max of acortatarin A were much higher than 9-epi-acortatarin A which indicated the bioavailability of 9-epi-acortatarin A was lower than acortatarin A due to its relatively small exposure in the body. It can be seen that the structural characteristics of compounds itself influenced its absorption in vivo to a certain extent [19].

In addition, the concentration–time curves of 12 compounds including 3,4,5-trimethoxyphenol-1-O-β-d-glucopyranoside (1), neoandrographolide (5), trans-ferulic acid (6), genistin (7), salicylic acid (8), ononin (10), andrographolide (12), berberine (13), 14-deoxy-11,12-didehydroandrographolide (15), panicolin (16), andrograpanin (17), and Z-ligustilide (18) in this experiment showed double-peaks, which means that the plasma drug concentration reached the peak for the first time, and decreased with the time going but rose again to form the second peak. The bimodal phenomenon of these compounds may be related to the enterohepatic circulation [20‐22], distribution [23], variable gastric emptying and double-site absorption [24, 25] during the absorption process of these compounds. As shown in Fig. 3 and Table 6, owing to the very similar structures, except for andrograpanin the other three diterpenes (neoandrographolide, andrographolide and 14-deoxy-11,12-didehydroandrographolide) have similar pharmacokinetic parameters and concentration–time curves in vivo, being absorbed and eliminated with the similar rate. The first peak of the three diterpenes arises at 0–1 h, and the second peak appears 1.5–3 h post dose. As showed in Fig. 1, neoandrographolide is the derivative of andrograpanin, in which the hydroxyl group is substituted by glucose, but their pharmacokinetic parameters are remarkably different, with T_max at 1.00 h, neoandrographolide was absorbed more rapidly than andrograpanin (T_max at 3.33 h). Notably, distinct double-peaks of the four diterpenes were observed in plasma concentration–time curves of all these timosaponins, which was in conformity with the previous study [26]. However, the factors that may contribute to double peaks are not clear and the specific reasons need to be further studied. Double plasma concentration peaks were also observed in the plasma concentration curves profiles of berberine, which was consistent with results from previous studies [27]. Distribution re-absorption and enterohepatic circulation may contribute to double peaks phenomenon. It was reported that while the berberine was absorbed, it was distributed rapidly with higher concentration in tissues than that of the plasma, which is possible for the drug to transfer from tissues to plasma and causes another peak in plasma [28]. Another possible cause of bimodal phenomena is that berberine inhibited gastrointestinal peristalsis which influenced drug absorption [29]. Moreover, one glucoside (3,4,5-trimethoxyphenol-1-O-β-d-glucopyranoside) [18], a flavone (genistin) [30] and a phthalide (Z-ligustilide) [31] presented a double-peak absorption phase, which were reported. It is well known that the absorption of compounds in TCM is a very complex process that manifests itself through potential interaction with a host of physicochemical and physiological variables. Therefore, further detailed absorption studies are needed to elucidate the mechanism of the double-peak phenomenon in pharmacokinetics.