An Evaluation Of Pharmacokinetic Interactions Between Characteristic Active Constituents In Si-Wu Decoction

Si-Wu Decoction is a well-known traditional Chinese medicine(TCM) formula，whichis composing of Radix Rehmanniae Preparata (the monarch herb), Radix AngelicaeSinensis (the minister herb), Radix Paeoniae Alba (the assistant herb) and RhizomaChuanxiong (the servant herb). It has been widely used in clinic for the treatment ofblood deficiency and gynecologic diseases. It was reported that fructose (from RadixRehmanniae Preparata), ferulic acid (from Radix Angelicae Sinensis and RhizomaChuanxiong), paeoniflorin (from Radix Paeoniae Alba) and tetramethylpyrazine(from Rhizoma Chuanxiong) contributed to the promoting effect of Si-Wu Decoctionon hematopoiesis in mice, which were subjected to60Co γ-rays irradiation for a modelof anaemia. In addition, the effect of weight loss of paeoniflorin and ferulic aciddisappeared when the four effective components were co-administrated to micesimultaneously.The purpose of our study was to elucidate the pharmacokinetic interactions ofcharacteristic active constituents in four hers of Si-Wu Decoction in rats, which wasexpected to show relevance to the principle of “Jun-Chen-Zuo-Shi” theory in TCMand to provide valuable information for the study of active constituent prescription ofSi-Wu Decoction. Firstly, according to data mining from literatures and previousstudy,1－3active constituents were selected from Radix Rehmanniae Preparata,Radix Angelicae Sinensis, Radix Paeoniae Alba and Rhizoma Chuanxiong,respectively. Characteristic active constituents from the four herbs were confirmed bythe evidence of the content of active constituents in herbs and the absorption in vivoof active constituents after oral administration of Si-Wu Decoction and Siwu Heji.Finally, the study of pharmacokinetic interaction between characteristic activeconstituents after oral administration to rats was conducted.The First Part: HPLC for the determination of9active constituents in fourherbs of Si-Wu DecoctionAccording to the previous study and data mining from the available Chinese andEnglish literatures,9active constituents were selected for the present study, i.e.catalpol, verbascoside and5-Hydroxymethyl furfural (5-HMF), from Radix Rehmanniae Preparata; ligustilide, ferulic acid and butylphthalide, from RadixAngelicae Sinensis and Rhizoma Chuanxiong; paeoniflorin and albiflorin, from RadixPaeoniae Alba; tetramethylpyrazine, from Rhizoma Chuanxiong. HPLC methodswere developed for the determination of9active constituents in herb extractions.50%(v/v) methanol was employed for the ultrasonic extraction of Radix RehmanniaePreparata and Rhizoma Chuanxiong, as well as70%(v/v) methanol for RadixAngelicae Sinensis and50%(v/v) ethanol for Radix Paeoniae Alba. The separation of9active constituents was performed by Agilent HC-C18（4.6×250mm；5μm）withmobile phase consisting of acetonitrile and water (containing0.5%H3PO4).The results indicated that the content of verbascoside and5－HMF in RadixRehmanniae Preparata was0.14%and0.07%, respectively, as well as0.01%,2.05%and0.04%for butylphthalide, ligustilide and ferulic acid in Radix Angelicae Sinensis,respectively;2.27%and0.41%for paeoniflorin and albiflorin in Radix Paeoniae Alba,respectively;0.02%,0.02%and0.06%for5－HMF, tetramethylpyrazine and ferulicacid in Rhizoma Chuanxiong, respectively.Combining all, we have concluded that the content of9active constituents in herbswas of significant difference. Ligustilide, peoniflorin and albiflorin should deservemore attention for their high content in herbs.The Second Part: Pharmacokinetics study of the absorbed active constituentsafter oral administration of Si-Wu Decoction and Siwu Heji to ratsThe objective of this part was to establish liquid chromatography-tandem massspectrometry (LC-MS/MS) methods for detection of9active components in ratplasma after oral administration of Si-Wu Decoction and Siwu Heji to rats and theirpharmacokinetic properties.The content of9active constituents in Si-Wu Decoction and Siwu Heji wasdetermined by LC-MS/MS after freeze-drying of the two TCM formulae. Rats wererandomly divided into Si-Wu Decoction group and Siwu Heji group. Plasma sampleswere collected before and after dosage, and treated with acetonitrile for proteinprecipitation before LC-MS/MS analysis. Pharmacokinetic parameters werecalculated by WinNonlin5.2. Statistical analysis of pharmacokinetic parametersbetween Si-Wu Decoction group and Siwu Heji group was performed on SAS9.2.It was indicated that the content of catalpol, ligustilide, butylphthalide, paeoniflorinand albiflorin in Si-Wu Decoction and Siwu Heji was2.94,244.67,4.24,1411.65,261.83mg L-1and3.64、255.50,4.68,1456.65,297.00mg L-1, respectively, while verbascoside, ferulic acid and tetrapyrazine was undetectable. Catalpol, paeoniflorinand albiflorin were detected in rat plasma after oral administration of Si-WuDecoction and Siwu Heji to rats, but catalpol could not be determined accurately.Paeoniflorin and albiflorin were absorbed and eliminated quickly in rats. Nodifference of main pharmacokinetic parameters (T1/2, Tmax, Cmax, AUC and MRT) ofPaeoniflorin and albiflorin was found between Si-Wu Decoction group and Siwu Hejigroup, which meant that the change of compatibility of Si-Wu Decoction exert littleinfluence on the pharmacokinetics of paeoniflorin and albiflorin.Combining all, catalpol, paeoniflorin and albiflorin could be selected as characteristicactive constituents of Si-Wu Decoction.The Third Part: The development of LC-MS/MS methods for determination ofcatalpol, ligustilide, ferulic acid, paeoniflorin and tetramethylpyrazine inbiological specimenCatalpol, ligustilide, ferulic acid, paeoniflorin and tetramethylpyrazine were selectedas characteristic active constituents of Si-Wu Decoction. The purpose of this part wasto develop LC-MS/MS methods for determination of catalpol, ligustilide, ferulic acid,paeoniflorin and tetramethylpyrazine in rat plasma. These methods would be used todetermine the content of the five active constituents in human serum after oraladministration of Si-Wu Decoction and Siwu Heji, and to study the pharmacokineticinteractions between the five active constituents after oral administration of theircombinations to rats.The chromatographic separation of the five active constituents was achieved on aDiamonsil C18(2.1×150mm；5μm) for catalpol with a mobile phase consisting ofmethanol and water (containing5mM ammonium formate), a Thermo SyncronisC18(2.1×150mm；3.7μm) for ligustilide and tetramethylpyrazine with a mobile phaseconsisting of methanol and water (0.1%formic acid) and for ferulic acid andpaeoniflorin with a mobile phase consisting of acetonitrile and water (0.1%formicacid). The internal standards (IS) were aucubin for catalpol, diazepam for ligustilideand tetramethylpyrazine, and puerarin for ferulic acid. Catalpol, ligustilide andtetramethylpyrazine, including the ISs, were detected by means of Multiple ReactionMonitor (MRM) mode operated in the positive ionization mode, while ferulic acidand paeoniflorin (including their IS) were operated in the negative ionization mode.Plasma samples were pretreated with acetonitrile for protein precipitation. All theLC-MS/MS methods were subjected to fully assay, including selectivity, precision and accuracy, dilution integrity, matrix effect (ME), recovery and stability.The results indicated that no endogenous interference was observed under the currentinstrumental conditions. The precision and accuracy of methods and diluted sampleswere within±15%. No dilution integrity was found after5times dilution of theplasma samples of catalpol and ligustilide. The ME and recovery of all analytes werein the range of85.73%－107.19%and90.93%－108.63%, respectively, which meantthat there was no significant matrix effect in the present methods and the extraction offour compounds were unanimous and reproducible. After being placed at ambienttemperature for4h, after being stored at20C for30days or subjected to threefreeze–thaw cycles, the values of RSD%and RE%of all analytes were within±15%,which indicated that all analytes were stable over all steps of determination.Combining all, three LC-MS/MS methods were developed in this part, which wereselective, sensitive and reproducible enough for the in vivo study of catalpol,ligustilide, ferulic acid, paeoniflorin and tetramethylpyrazine.The Fourth Part: The analysis of characteristic active constituents in humanserum after oral administration of Si-Wu Decoction and Siwu HejiThe purpose of this part was to study the absorbed characteristic active constituents inhuman serum after oral administration of Si-Wu Decoction and Siwu Heji by threewell-established LC-MS/MS methods, results of which would provide evidence forrational selection of characteristic active constituents in Si-Wu Decoction.35subjects were randomly divided into two groups: Si-Wu Decoction group (n=17)and Siwu Heji group (n=18). Serum samples were collected before dosage. After oraldosing for consecutive21days, serum samples at1h and2h were collected afteradministration in the morning of the22th day. Serum samples were pretreated withacetonitrile for protein precipitation and then subjected to the analysis of LC-MS/MS.The results indicated that paeoniflorin and ligustilide were detected and quantitated inhuman serum in Si-Wu Decoction group and Siwu Heji group, concentrations ofwhich were in the range of2.12－18.44ng·mL-1and2.18－7.20ng·mL-1,respectively. While catalpol, ferulic acid and tetrapyrazine were both undetectable.Meanwhile, the number of patients that paeoniflorin and ligustilide cound be detectedin serum in Si-Wu Decoction group was more than that of Siwu Heji group, whichsuggested that Si-Wu Decoction facilitated the absorption o active constitutents invivo. For the difference described above that will or will not result in the difference foefficacy between Si-Wu Decoction and Siwu Heji group, our penetrating research is still in progress by using high resolution mass spectrometry and metabonomics.Combining all, paeoniflorin and ligustilide could be selected as characteristic activeconstituents of Si-Wu Decoction.The Fifth Part: Pharmacokinetic interactions between five characteristic activeconstituents of Si-Wu Decoction in ratsThe purpose of this part was to evaluate the pharmacokinetic interactions betweenfive characteristic active constituents of Si-Wu Decoction in rats.Rats were randomly divided into catalpol group, ligustilide group, ferulic acid group,paeoniflorin group, tetramethylpyrazine group,(catalpol+ligustilide+ferulic acid)group,(catalpol+paeoniflorin) group,(catalpol+tetramethylpyrazine) group,(catalpol+ligustilide+ferulic acid+paeoniflorin) group,(catalpol+ligustilide+ferulic acid+tetramethylpyrazine) group,(catalpol+paeoniflorin+tetramethylpyrazine) group and (catalpol+ligustilide+ferulic acid+paeoniflorin+tetramethylpyrazine) group. The dosages were3.5mg·kg-1for catalpol,100mg·kg-1for ligustilide,3mg·kg-1for ferulic acid,100mg·kg-1for paeoniflorin and1mg·kg-1for tetramethylpyrazine. Plasma samples were collected before dosing and at differenttime after oral administration. Pharmacokinetic parameters and were calculated byWinNonlin5.2, and pharmacokinetic parameters difference between groups wasachieved by one-way analysis of variances (ANOVA, SAS9.2).For catalpol, the Cmaxand AUC of catalpol in all the co-administration groups wereboth smaller than that of catalpol group (P<0.05). The MRT in (catalpol+ligustilide+ferulic acid+paeoniflorin+tetramethylpyrazine) group was longer than that of(catalpol+ligustilide+ferulic acid+tetramethylpyrazine) group and (catalpol+paeoniflorin+tetramethylpyrazine) group (P<0.05). While the T1/2and Tmaxremainedunchanged.For ligustilide, the T1/2and Tmaxof ligustilide in the (catalpol+ligustilide+ferulicacid) group were shorter than that of ligustilide group and other co-administrationgroups (P<0.05). But the AUC of ligustilide in (catalpol+ligustilide+ferulic acid+paeoniflorin+tetramethylpyrazine) group was bigger than that of ligustilide group(P<0.05). The MRT in all the co-administration groups was shorter than of ligustilidegroup, while the Cmaxwas unchanged.For ferulic acid, the Cmaxin all the co-administration groups was smaller than that offerulic acid alone (P<0.05). Apart from (catalpol+ligustilide+ferulic acid) group,the AUC in co-administration groups was smaller than that of ferulic acid alone (P<0.05). The Tmaxin (catalpol+ligustilide+ferulic acid+paeoniflorin+tetramethylpyrazine) group was longer than that of ferulic acid alone (P<0.05).For paeoniflorin, the T1/2, Tmax, Cmaxand AUC remained unchanged before and afterco-administration with other characteristic active constituents. But the MRT in(catalpol+ligustilide+ferulic acid+paeoniflorin+tetramethylpyrazine) group waslonger than that of (catalpol+ligustilide+ferulic acid+paeoniflorin) group(P<0.05).For tetramethylpyrazine, the Cmaxand AUC in all the co-administration groups wereboth bigger than that of ferulic acid group (P<0.05). The Cmaxand AUC in (catalpol+ligustilide+ferulic acid+paeoniflorin+tetramethylpyrazine) group were also biggerthan of (catalpol+ligustilide+ferulic acid+tetramethylpyrazine) group and(catalpol+paeoniflorin+tetramethylpyrazine) group (P<0.05). Meanwhile, the AUCin (catalpol+ligustilide+ferulic acid+tetramethylpyrazine) group was bigger thanof (catalpol+paeoniflorin+tetramethylpyrazine) group (P<0.05).In conclusion, complicated pharmacokinetic interactions were found between5characteristic constituents from4herbs of Si-Wu Decoction, the underlyingmechanism of which was still unknown and required further study. These resultscould provide intriguing and valuable information significant guidance for the studyand development of active constituent prescription of Si-Wu Decoction.