Pharmacokinetics Of Flavonoids Of Dracocephalum Moldavica L. And Their Metabolic Pathways

Dracocephalum moldavica L. (DML) is used in traditional Uygur medicine as a drug for the treatment of coronary heart disease, hypertension, bronchitis, headaches and other diseases. The DML flavone mixture, which is a standardized extract from DML, possesses with anti-oxidation, anti-ischemic treatment of coronary heart disease, angina and other effects. Tilianin (Til) is the main flavone in the DML flavone mixture, which have been identified as an active component responsible for the cardiovascular and anti-inflammatory effects of DML. Many literatures on preclinical pharmacokinetics of flavonoids show poor bioavailability of flavonoids. In general, flavonoids are rapidly absorbed, and they are rapidly and extensively glucuronidated in the intestinal tract and liver. Most preclinical studies have ignored the importance of pharmacokinetics of flavonoid metabolites after oral administration of flavonoids.There are a lot of reports on the pharmacological effects and clinical application of DML, and rarely research reports about exposure of the DML flavone mixture in vivo. In this paper, pharmacokinetics and metabolism of DML flavones were studied, including methylation of a flavone glucuronide. This paper enriched the metabolic pathway of flavonoid, and provided a new idea for pharmacokinetic researches of flavonoids in traditional Chinese medicine. This studies provided scientific data for the new drug development, and improvement formulation of the DML flavone mixture.1. Profiling of flavones in rat plasma after oral administration of the DML flavone mixture by UPLC-DAD-QTOFFlavonoids, from the DML flavone mixture, were identified using UPLC-DAD-QTOF. The main flavonoids identified in the DML mixture were Til, Lut-7-G (Lut-7-G), Api-7-G (Api-7-G), Aca 7-glucuronide(Aca-7-G), apigenin (Api), acacetin (Aca), luteolin 7-glucoside (Lut-7-Glc), luteolin (Lut) and diosmetin (Dio) at concentration of 7.3%,3.8%,1.1% 1.0%,0.089%,0.17%,0.45%,0.14% and 0.30% (w/w), respectively.The main flavonoids are flavones, its basic backbone shown in Figure a, with two or more phenolic hydroxyl groups, mainly in 5-,7-,3’- and 4’-hydroxyl groups. The most abundant of flavonoids in the DML flavone mixture were acacetin glycoside (Til and Aca-7-G), followed by luteolin glycosides (Lut-7-G and Lut-7-Glc) and apigenin glycoside (Api-7-G).In order to explain the UGT metabolite profiling in rat plasma after orally administration of the DML flavone mixture, we have established a strategy for identification sites of flavone glucuronidation by combined multiple methods The novel strategy involves accurate mass measurements of flavones glucuronides, their [Co (Ⅱ) (flavone glucuronide-H) (4,7-diphenyl-1,10-phenanthroline)2]+ complexes which were generated via post-column addition of CoBr2 and 4,7-diphenyl-1, 10-phenanthroline, and their MS/MS fragmentation by UPLC-DAD-Q-TOF and comparison of retention times with biosynthesized standards of the different isomers which were identified by analyzing shift in UV spectra compared with the spectra of their respective aglycones. We successfully generated a UGT metabolism profile of flavones in rat plasma after oral administration of a flavone mixture. In the present study,12 flavone glucuronides, which were Aca-7-G, Aca-5-G, Api-7-G, Aip-5-G, Lut-7-G, Lut-3’-G, Lut-4’-G, Dio-7-G, Dio-3’-G, Chr-4’-G, Dio/Chr-5-G and cirsimaritin 4’-glucuronide (Cir-4’-G), were detected and identified. Glucuronidation of the flavone skeleton at the 3’-/7- position was more prevalent, but Lut-4’-G levels exceeded Lut-7-G levels. The flavones that appeared in both the dosed rat plasma samples and the flavone mixture, but not in the controlled rat plasma samples, were regarded as absorbed flavones. Finally,10 compounds were identified as the absorbed flavones, including Api-7-G, Dio-7-G, Aca-7-G, Lut-7-G, Lut, Til, Api, Dio, Chr, and Aca. And 14 flavone metabolites were found in the rat plasma, which were the glucuronidated and sulfated derivatives. Based on the UGT metabolism profiling of flavones in rat plasma, the six main compounds (Til, Aca-7-G, Api-7-G, Lut-3’-G, Aca and Api) were selected as pharmacokinetic markers.2. Pharmacokinetics of the six flavones in rats after oral administration of the DML flavone mixtureA quantitative analysis method was developed to simultaneously determine Til, Aca-7-G, Lut-3’-G, Api-7-G, Api and Aca in rat plasma by ultra-performance liquid chromatography tandem mass spectrometry (UPLC-MS/MS). The ananlytes were extracted from plasma samples by methanol, separated by Acquity UPLC BEH C18 column (1.7μm,50 mm×2.1 mm), and erlotinib was used as the internal standard. The mobile phase consisted of 0.5 mM ammonium acetate in water (solvent A) and acetonitrile (solvent B). The flow rate was set at 0.3 mL/min. The mobile phase gradient was as follows:0-1 min,5-5% B; 1-1.5 min,5-10% B; 1.5-5 min,10-21% B; 5-6 min,21-21% B; 6-8 min,21-90%B; 8-10 min,90%B; 10-10.5 min,90-5%B. The quantitative method have been validated, all the results were within the accepted limits. The analytical procedure was successfully applied to pharmacokinetic study of the free and conjugated forms of the analytes after oral administration of the DML flavone mixture. The rats were randomly assigned to low dosing group (100 mg/kg), middle dosing group (200 mg/kg) and high dosing group (400 mg/kg). Individual plasma-concentration data were analyzed by non-compartmental model and the main pharmacokinetic parameters were calculated using Winnolin 3.3. Til, Aca-7-G, Api-7-G, Api and Aca were rapidly absorbed in the high dosing group and middle dosing groups, but concentrations of Api and Aca were too low to determine in low dosing group. Lut-7-G was rich in the DML flavone mixture, but was rarely in the rat plasma after oral administration of the DML flavone mixture, and Lut-3’-G was exposed more in rat plasma. The pharmacokinetic results showed that their maximal concentrations in blood were obtained within 0.4 h, except Lut-3’-G (about 9 h). The rat exposure was practically nonlinear under the studied dosages from 100 to 400 mg/kg. The possible reasons are that when the dose exceeds a certain limit, the catalytic ability of enzymes and carrier transport capacity reaches saturation, and plasma concentrations do not depend on the dosage. The absolute bioavailability after oral administration of SD rats in the high, medium and low-dosing group was an average of 1.63%,0.45%,0.24%, respectively.3 Pharmacokinetics of tilianin and its main metabolites in rats after oral administration of tilianinA quantitative analysis method was developed to simultaneously determine Til, Aca-7-G, acacetin-7-sulfate (Aca-7-S) and Aca in plasma using UPLC-MS/MS. The ananlytes were extracted from plasma samples by methanol, separated by Acquity UPLC BEH C18 column (1.7μm,50 mm×2.1 mm), and erlotinib was used as the internal standard. The mobile phase consisted of 0.5 mM ammonium acetate in water (solvent A) and acetonitrile (solvent B). The flow rate was set at 0.3 mL/min. The mobile phase gradient was as follows:0-1 min,5-5%B; 1-1.5 min,5-10%B; 1.5-5 min,10-21% B; 5-6 min,21-21% B; 6-8 min,21-90% B; 8-9 min,90% B; 9-9.5 min, 90-5% B. The quantitative method have been validated, all the results were within the accepted limits. The analytical procedure was successfully applied to pharmacokinetic study of the free and conjugated forms of the analytes after oral administration of Til. The rats were randomly assigned to low dosing group (15 mg/kg), high dosing group (30 mg/kg) and tail vein administration group (0.3 mg/kg). Individual plasma-concentration data were analyzed by non-compartmental model and the main pharmacokinetic parameters were calculated using Winnolin 3.3. Concentrations of Til and Aca were relatively lower in the plasma, and concentrations of Aca-7-G and Aca-7-S is relatively higher in plasma. Tmax for Aca-7-G (about 20 min) in DML group was significantly lower than that (about 9 h) in Til group (P<0.05). This phenomenon may be related to triple recycling (i.e., enterohepatic, enteric and local recycling) of Til, Aca-7-G and Aca in vivo. After oral administration of Aca (10 mg/kg), average time to peak plasma concentration (Tmax) for Aca-7-G was approximately 9 h, which was consistent with tmax for Aca-7-G in pharmacokinetics of Til. It was validated that a lot of Aca-7-G was excreted into lumen through the triple recycling, and was rowed to the colon with the intestinal peristalsis. Conjugate was hydrolyzed back to its aglycone form by bacterial β-glucuronidases in the colon. The released aglycone, Aca was subsequently reabsorbed in the colon and rapidly glucuronidated to Aca-7-G by UGTs.4. Methylation of Iuteolin7-glucuronideMethylation of Lut-7-G was studied by using rat liver S9 methylated incubation system. In the system, catechol-O-methyltransferases (COMTs) were provided by rat liver S9. S-adenosyl-L-methionine was methyl group donor. The whole reaction was conducted at 37℃ in the phosphate buffer (5 mM, pH 7.8). The reaction was stopped by adding cold methanol solution after a certain time. After incubation of Lut-7-G in the methylated reaction system, there were generated two methylated products, which were identified to Dio-7-G and Chr-7-G by QTOF. Experiments of methylation of Lut and Lut-7-G were conducted at three different concentrations (1,5 and 10μM) by rat liver S9. The results show that a preference of 4’-methylation in both Lut and Lut-7-G by rat liver S9. In the reaction, Lut-7-G was easier to generate Dio-7-G. The methylated rate of Lut-7-G were significantly slower than that of Lut (P <0.05), and about 1/2-2/3 of Lut methylated rate. In kinetics of Dio and Chr glucuronidation by using rat liver microsome, rates of Dio glucuronidation were determined at a substrate concentration rang of 0.3125-80μM and reaction time of 30 min; Rates of Chr glucuronidation were determined at a substrate concentration rang of 0.3125-15μM and reaction time of 30 min. The apparent kinetic parameters were fitted by Michaelis-Menten equation and other equations. Glucuronidation of Chr/Dio at 4’-/3’-position was more prevalent, respectively.After oral administration of Lut to bile duct cannulated rats (4.7 mg/mg), bile was collected through the duct. A lot of Lut-3’-G, Lut-7-G, Chr-7-G and Dio-7-G was detected in bile, and Chr-4’-G and Dio-3’-G was not detected in bile by using UPLC-MS/MS analysis. In conclusion, Chr-7-G and Dio-7-G may not be generated by glucuronidation of Chr or Dio which are generated by methylation of Lut in vivo, but could be mainly generated by methylation of Lut-7-G. Methylation may also be an important metabolic pathway of catechol glucuronides.In summary, a rapid and sensitive strategy combining Q-TOF fragmentation, post-column cobalt complexation and UV-shift was developed to identify the substitution positions of flavone glucuronides in rat plasma. We successfully generated a UGT metabolism profile of flavones in rat plasma after oral administration of the DML flavone mixture for the first time. In the present study,12 flavone glucuronides were detected and identified. Glucuronidation of the flavone skeleton at the 3’-/7-position was more prevalent, but Lut-4’-G levels exceeded Lut-7-G levels. Based on the profiling of flavones in rat plasma, the six main compounds (Til, Aca-7-G, Api-7-G, Lut-3’-G, Aca and Api) were selected as pharmacokinetic markers for the flavone mixture. Moreover, a sensitive, reliable and specific UPLC-MS/MS method for simultaneous analysis of the six flavones in rat plasma was developed and validated for the first time. The analytical procedure was successfully applied to pharmacokinetic study of the free and conjugated forms of the analytes after oral administration of the mixture. Systemic exposure to Aca-7-G and Lut-3’-G after oral administration of the extract was significantly greater than that to the other flavones glucuronides. The absolute oral bioavailability of Til in rat was low(less than 2%). Til presented nonlinear pharmacokinetics of the DML flavone mixture. Tmax for Aca-7-glucronide (about 20 min) in DML group was significantly lower than that (about 9 h) in Til group (P<0.05). This phenomenon may be related to triple recycling (i.e., enterohepatic, enteric and local recycling) of Til, Aca and Aca-7-G in vivo. It is firstly confirmed that Lut-7-G can be methylated to Chr-7-G and Dio-7-G by rat live S9. Methylation may also be an important metabolic pathway of catechol glucuronides.