Identification and pharmacokinetics of the major constituents of Fugan Fang in rat plasma

Abstract

Fugan Fang (FGF) is an effective traditional Chinese medicine (TCM) prescribed for the clinical treatment of hepatic diseases. No reports exist on the absorbed bioactive components of FGF and their pharmacokinetics after oral FGF administration. In this study, the FGF components absorbed into the blood were identified and their pharmacokinetic profiles were explored, with the goal of understanding the effective constituents of FGF. Ultra-fast liquid chromatography-high resolution mass spectrometry techniques with a target-directed database-dependent strategy were used to identify the constituents of FGF extract and FGF compounds in rat plasma after oral FGF administration. Ultra-performance liquid chromatography coupled to triple-quadrupole mass spectrometry was used to evaluate the pharmacokinetics of several FGF compounds in rat plasma. A total of 53 compounds were present in the FGF extract, and 14 prototype constituents with 11 potential metabolites were identified in rat plasma. The pharmacokinetic parameters of calycosin-7-O-β-D-glucoside, ononin, gentiopicroside, sweroside, ferulic acid, and p-coumaric acid in rats were measured. These findings provide useful information that will support studies aimed at clarifying the identity of bioactive FGF constituents and their biological effects, and will thus further the development of FGF.

---

Introduction

Hepatic fibrosis is a common pathological process in chronic liver diseases that results from nutritional deprivation, autoimmunisation, alcohol abuse, or infection by the hepatitis virus. The worsening of hepatic fibrosis leads to carcinoma or hepatocirrhosis, endangering the health of affected patients. Therefore, the prevention or reversal of hepatic fibrosis is of great importance in the treatment of chronic liver diseases. Few drugs effectively control hepatic fibrosis; however, recent research reports and clinical treatment results^1–4 reveal that some preparations used in traditional Chinese medicine (TCM) have good potential as treatments for hepatic fibrosis.

Fugan Fang (FGF), an effective TCM prescription for the clinical treatment of hepatic diseases, is composed of Radix Astragali, Radix Angelica sinensis, Flos Carthami, Radix Gentianae, and Caulis spatholobi. In 1976, folk doctors in the Heilong Jiang province of China created FGF based on numerous years of experience in utilising Chinese herbal medicine to treat hepatic diseases in the clinic (Drug Inspecting Institute of Heilong Jiang Province, 1976). In recent years, experimental studies have revealed that FGF exhibits good preventive and therapeutic effects against hepatic fibrosis^5–7 through decreasing serum aspartate aminotransferase and γ-glutamyl transpeptidase, reducing hydroxyproline content collagen deposition in liver, and improving hepatocellular degeneration and inflammatory necrosis.⁷ Therefore, FGF has the potential to be developed into a new drug for the treatment of hepatic fibrosis.

Understanding the profile of bioactive constituents in FGF is important for the elucidation of its clinical effectiveness and further development as a treatment for hepatic fibrosis. It is increasingly recognised that only components which are absorbed into the blood may be considered as potential bioactive constituents of medicinal preparations.^8,9 Furthermore, the plasma concentrations of bioactive constituents determine their therapeutic effects. Therefore, identification and pharmacokinetic study of the major compounds absorbed into the blood after administration of a medicinal preparation are essential to proper understanding of its effective constituents.

Studies on absorbed FGF components and their pharmacokinetics after oral administration of FGF have not been conducted. Therefore, in the present study, FGF constituents in rat plasma and their metabolites following oral administration of FGF were identified using ultra-fast liquid chromatography-high resolution mass spectrometry techniques (UFLC-HRMS) with a target-directed strategy, which was dependent on a database of chemical information derived from the literature on the individual herbs contained in FGF. In addition, pharmacokinetic studies were conducted on FGF constituents in rat plasma after oral FGF administration.

Materials and methods

Chemicals and materials

The reference standards of gentiopicroside, hydroxysafflor yellow A, astragaloside IV, ferulic acid, formononetin, protocatechuic acid, quercetin, and epicatechin were provided by the Chinese National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). p-Coumaric acid, ononin, sweroside, astragaloside I, calycosin, and calycosin-7-O-β-D-glucoside were purchased from Sichuan Weikeqi Biotech Co. Ltd. (Chengdu, China). The purity of all of the reference compounds was determined to be greater than 98% by HPLC analysis. The herbal ingredients of FGF, including Radix Astragali, Radix Angelicae sinensis, Flos Carthami, Radix Gentianae, and Caulis spatholobi, were purchased from Shanghai Tongjitang Pharmaceutical Co. Ltd. (Shanghai, China).

HPLC-grade acetonitrile and methanol were purchased from Burdick & Jackson Company (Muskegon, MI, USA). Acetic acid (HPLC grade) was purchased from Tedia Company (Fairfield, OH, USA). Formic acid (HPLC grade) was purchased from CNW Technologies GmbH (Düsseldorf, Germany). Deionised water was purified using a Milli-Q system (Millipore, Bedford, MA, USA). All other chemicals were of analytical grade.

Sprague-Dawley rats (250 ± 20 g) were obtained from the Laboratory Animal Center of the Shanghai University of Traditional Chinese Medicine (Shanghai, China). The rats were housed in an air-conditioned room with a temperature of 22 ± 2 °C and relative humidity of 50 ± 10%, with a 12 h light–dark cycle. The rats were allowed free access to food and water. The animals were acclimatised to the facilities for 7 days, and then fasted for 12 h with free access to water before the experiments. Animal studies were conducted according to the Institute's Guide for the Care and Use of Laboratory Animals.

Instrumentation and conditions

The UFLC-HRMS system consisted of an LC-20AD ultra-fast liquid chromatograph (Shimadzu, Kyoto, Japan) and a TripleTOF 5600 (AB SCIEX, Toronto, Canada) mass spectrometer. An Agilent Eclipse Plus C18 column (3.0 × 150 mm, 3.5 μm; Agilent Technologies, Santa Clara, CA, USA) was used with a mobile phase of 0.1% formic acid (A) and acetonitrile (B). The elution gradient for the extract was as follows: 10% B from 0 to 0.5 min, 10–20% B from 0.5 to 5 min, 20 to 35% B from 5 to 10 min, 35 to 50% B from 10 to 16 min, 50 to 70% B from 16 to 19 min, 70 to 90% B from 19 to 21.5 min, 90% B from 21.5 to 31.5 min, 10% B from 31.6 min, and 10% B from 31.6 to 39 min. The elution gradient for the plasma was as follows: 10% B from 0 to 0.5 min, 10–20% B from 0.5 to 5 min, 20 to 35% B from 5 to 10 min, 35 to 50% B from 10 to 18 min, 50 to 70% B from 18 to 22 min, 70 to 90% B from 22 to 25 min, 90% B from 25 to 34.5 min, 10% B at 34.6 min, and 10% B from 34.6 to 43 min. The column temperature was 30 °C, the flow rate was kept at 0.4 mL min⁻¹, and the injection volume was set at 10 μL. HRMS experiments were performed on a TripleTOF 5600 (quadrupole time-of-flight) instrument equipped with a DuoSpray™ source interface operated in the electrospray mode (AB SCIEX, Framingham, MA, USA). Data were acquired using Analyst TF 1.5.1 software (AB SCIEX, Framingham, MA, USA). The ion source temperature was set to 550 °C. The capillary voltage was set to 5.5 kV for the positive ion mode and −4.5 kV for the negative ion mode. The curtain gas pressure was 30 psi, and the nebuliser and the drying gas both had a pressure of 60 psi. The mass spectrometric data were collected in full scan mode, and the m/z ranged from 50 to 1200 in the positive and negative ion modes.

The ultra-performance liquid chromatography coupled to a triple-quadruple mass spectrometry (UPLC-MS/MS) system consisted of an ACQUITY UPLC (Waters Corporation, Milford, MA, USA) and a triple-quadruple mass spectrometer (Triple Quad 5500, Applied Biosystems, Foster City, CA, USA) equipped with an electrospray ionisation (ESI) source. Data acquisition and processing were performed using Analyst 1.5.2 software (AB SCIEX, Framingham, MA, USA). The column was a Waters Acquity UPLC C18 (2.1 × 100 mm, 1.7 μm; Waters Corporation, Milford, MA, USA), and the mobile phase consisted of 0.01% acetic acid with 4 mmol L⁻¹ ammonium acetate (A) and acetonitrile (B) using a gradient elution program of 10% B from 0 to 0.6 min, 10 to 30% B from 0.6 to 2.2 min, 30 to 50% B from 2.2 to 4.8 min, 50 to 90% B from 4.8 to 8 min, 90% B from 8 to 9 min, and 10% B from 9.1 to 11 min. The column temperature was set to 40 °C, the flow rate was maintained at 0.2 mL min⁻¹, and the injection volume was set at 10 μL. The electrospray ion source was operated with polarity switching between the positive and negative ion modes in a single run. The ion spray voltage was set to 5 kV in the positive ion mode and −4.5 kV in the negative ion mode. The entrance potential was 10 V in the positive ion mode and −10 V in the negative ion mode. The ion source temperature was set at 500 °C. Nitrogen was used as the nebuliser gas (50 psi), auxiliary gas (50 psi), and curtain gas (30 psi). Selected reaction monitoring (SRM) mode was employed for quantification.

Sample preparation

FGF extract. The powdered FGF was prepared according to the traditional method, as described in a previous study.¹⁰ Briefly, a blended mixture of the 5 constituent herbs of FGF was immersed in deionised water for 30 min and boiled twice for 1 h with 10-fold mass and 8-fold mass of water, respectively. After filtration, the 2 decoctions were mixed and evaporated to dryness to obtain powdered FGF. Using LC-MS,¹⁰ the amounts of hydroxysafflor yellow A, gentiopicroside, sweroside, calycosin-7-O-β-D-glucoside, p-coumaric acid, ferulic acid, ononin, calycosin, astragaloside IV, formononetin, and astragaloside I in the extract were determined to be 0.04, 9.73, 0.21, 0.17, 0.12, 0.01, 0.08, 0.05, 0.02, 0.04, and 0.44 mg g⁻¹, respectively.

One hundred milligrams of the FGF powder was extracted with 10 mL 50% methanol for 1 hour under ultrasonics. The methanol extract was centrifuged at 15 000 rpm for 10 min at 4 °C, the supernatant was filtered through a 0.45 μm membrane, and a 10 μL aliquot of the filtrate was injected into the UFLC-HRMS system for qualitative analysis of its constituents.

Rat plasma sample. For the qualitative analysis of FGF constituents in the rat plasma, an aliquot (150 μL) of plasma was added to 0.45 mL methanol to precipitate constituent proteins, and the mixture was mixed using a vortex mixer for 5 min. After centrifugation at 15 000g for 10 min, the supernatant was transferred to a new tube and dried under nitrogen gas at 37 °C. The residues were redissolved in 100 μL of the mobile phase (0.1% formic acid water (A) : acetonitrile (B) = 90 : 10). The supernatant (10 μL) of each sample was injected into the UFLC-HRMS system for analysis.

For the pharmacokinetic studies, an aliquot (50 μL) of plasma was added to 150 μL of methanol to precipitate constituent proteins, 10 μL of a solution of mixed internal standards (IS, 0.27 μg mL⁻¹ wogonoside, and 0.27 μg mL⁻¹ rhein) in methanol was added, and the mixture was mixed using a vortex mixer for 5 min. After centrifugation at 15 000g for 10 min, the supernatant was transferred to a new tube and dried under nitrogen gas at 37 °C. The residues were redissolved in 100 μL of 50% methanol containing 0.1% acetic acid. The supernatant (10 μL) was injected into the UPLC-MS/MS system for analysis.

Qualitative analysis of FGF constituents in rat plasma

Rats were administered a single 2.16 g kg⁻¹ (amount of crude material, 6 g kg⁻¹ body weight) dose of FGF extract through intragastric gavage. Blood was collected from the rats before and after the administration of FGF. Plasma samples were separated by centrifugation at 8000 rpm for 10 min and frozen at −80 °C.

The screening, identification, and further characterisation of the components of FGF were first performed with an HRMS technique using an UFLC-HRMS system in both the positive and negative ion modes. Peak View software 1.2 (AB SCIEX, Framingham, MA, USA) was used to find the corresponding molecular formula with the accurate mass weights of the peaks (error, <5 ppm). Some peaks were identified by referring to the standards. For those peaks without standards, a database including about 200 major compounds was established by collecting information from the literature on the 5 herbs contained in FGF, including their names, formulas, accurate molecular weights, and MS2 information. The accurate masses of the additive ions, such as [M − H]⁻, [M + Na]⁺, [M + H]⁺, [M + K]⁺, and [M + HCOO⁻]⁻ were also calculated. The peaks were identified by referring to MS/MS spectra in the database, retention time, isotope matching, and detected ions.

The metabolites of a single component absorbed in the blood were identified using Metabolite Pilot™ Software 1.5 (AB SCIEX, Framingham, MA, USA), based on the possible metabolic or fragmentation pathways of the components of the FGF extract. The most probable molecular formulas of the metabolites were determined using several criteria, including mass accuracy <5 ppm, the nitrogen rule, isotopic patterns, and modifications based on common metabolic pathways. Furthermore, the structures of the metabolites were elucidated based on product ion spectra.

LC-MS/MS method for quantitative analysis of FGF constituents in rat plasma

A quantitative method using a UPLC-MS/MS system was developed to simultaneously quantify calycosin-7-O-β-D-glucoside, ononin, ferulic acid, p-coumaric acid, gentiopicroside, sweroside, formononetin, and protocatechuic acid in rat plasma. Optimised mass parameters of all of the target analytes and internal standards for the LC-MS/MS analysis are shown in Table 1.

Table 1 Optimized mass parameters for the LC-MS/MS analysis of all the target analytes and internal standard^a

Compounds	Precursor ion (m/z)	Product ion (m/z)	CE (eV)	DP (V)	CXP (V)	Detected ions	Dwell time (ms)	Delay time (ms)
a Declustering potential (DP), collision energy (CE), and collision exit potential (CXP) for each analyte and internal standard.
Calycosin-7-O-β-D-glucoside	447.3	284.9	34	39	17	[M + H]^+	50	5
Ononin	431.0	269.0	46	17	20	[M + H]^+
Ferulic acid	193.0	133.8	−19	−75	−14	[M − H]^−
p-Coumaric acid	163.0	119.0	−17	−71	−13	[M − H]^−
Gentiopicroside	357.0	195.0	10	104	27	[M + H]^+
Sweroside	359.2	196.4	13	52	13	[M + H]^+
Formononetin	269.0	197.0	51	150	18	[M + H]^+
Protocatechuic acid	153.0	109.0	−18	−25	−16	[M − H]^−
Wogonoside (IS+)	462.0	285.0	24	62	20	[M + H]^+
Rhein (IS−)	283.0	239.0	−22	−36	−25	[M − H]^−

---

The selectivity of the method was evaluated by comparing the chromatograms of blank plasma samples obtained from 6 rats with corresponding plasma samples to which FGF compounds were added. The mixed working solution containing 8 compounds was diluted to appropriate concentration ranges for the construction of calibration curves. The linearity was assessed by assaying the calibration curves in rat plasma in 5 replicates. The precision was determined from inter- and intra-day runs using 3 QC levels and was expressed as relative standard deviation (RSD). Accuracy was calculated as the relative error of the observed and nominal concentrations of the QC samples. The recovery of the mixed standard at the 3 QC concentration levels was estimated by comparing 2 groups of control samples: one group in which the drug was added after extraction of the blank plasma (post-extraction, POST), and another group in which the drug was added to the plasma and the sample was prepared normally (pre-extraction, PRE). Extraction recovery was calculated as the response ratio of PRE/POST × 100%. The reproducibility of the extraction procedure was determined as RSD. The matrix effect of the target analytes was evaluated by comparing the peak areas of analytes and IS dissolved in blank plasma extracts with those of analytes dissolved at the same concentrations in methanol/water (50 : 50, v/v) at the QC concentrations of the analytes and the IS. Cross-interference of the analytes was evaluated by comparing the peak areas of all the analytes added to the plasma with each type (phenolic acids, flavonoids, and iridoid glycosides) of compound that was added to the plasma. The stability of the target analytes in rat plasma was evaluated by analysing the plasma samples of the QC samples. Short-term stability was determined after exposing the compound-spiked samples to 25 °C for 2 h and exposing the ready-to-inject samples (after extraction) to the autosampler rack (4 °C) for 24 h. Long-term stability was assessed after storing the standard compound-spiked plasma samples at −80 °C for 3 months. All stability test samples were analysed in triplicate, and the changes were determined relative to freshly prepared samples.

Pharmacokinetic study

Six rats (3 male and 3 female) received a single intragastric gavage administration of 1.08 g kg⁻¹ (amount of crude material, 3 g kg⁻¹ body weight) FGF extract. Blood samples (100 μL) were collected before dosing and subsequently collected at 0.033, 0.083, 0.25, 0.5, 1, 2, 4, 8, 12, 24, 36, and 48 h after FGF extract administration. After centrifugation at 15 000 rpm for 10 min, the plasma samples were frozen at −80 °C. The plasma concentrations of 8 constituent compounds of FGF were measured as described above using the standard curves run with each batch of samples.

Non-compartmental analysis (WinNonLin Pro 6.1, Certara, St. Louis, MO, USA) was utilised to obtain pharmacokinetic parameters for each rat. The maximum plasma concentration (C_max) was determined from the observed data, and the area under the plasma concentration–time curve (AUC_0–t) was calculated using the trapezoidal rule.

Results

Qualitative analysis of the components of FGF extract

Extracted ion chromatograms (XICs) of FGF in the positive and negative ion modes are shown in Fig. 1. Peaks 4, 12, 14, 15, 17, 21, 22, 23, 31, 37, 38, 42, 47, and 51 were attributed to protocatechuic acid, hydroxysafflor yellow A, epicatechin, gentiopicroside, sweroside, p-coumaric acid, calycosin-7-O-β-D-glucoside, ferulic acid, ononin, quercetin, calycosin, formononetin, astragaloside IV, and astragaloside I, respectively, by comparing the measured retention times and mass data with those of the reference compounds. Other peaks in the XICs were identified using the method described above. For example, when peak 2 (M = 504, C₁₈H₃₂O₁₆) was extracted with accurate mass (error < 5 ppm) using Peak View software, a series of peak ions with good isotope ratios (%difference < 20%) were obtained in negative and positive ion ESI mode, including 503.16176 at [M − H]⁻, 527.15826 at [M + Na]⁺, 505.17631 at [M + H]⁺, and 543.13219 at [M + K]⁺ at 2.19 min, confirming its molecular weight of 504.16904. The MS/MS fragments of [M + Na]⁺ were elucidated as 527 [M + Na]⁺, 365 [M + Na–Glc]⁺, and 203 [M + Na–Glc–Glc]⁺, which was consistent with data on gentianose.¹¹ The MS/MS fragments of [M − H]⁻ were elucidated as 503 [M − H]⁻, 341 [M − H–Glc]⁻, and 179 [M − H–Glc–Glc]⁻, confirming the peak as gentianose. The results for a total 53 identified FGF components are listed in the ESI,† and their structures are shown in Fig. 2.

Fig. 1 Extracted ion chromatogram of (A) extract of Fugan Fang in positive mode, (B) extract of Fugan Fang in negative mode, (C) rat plasma sample in positive mode, (D) rat plasma sample in negative mode, (E) blank plasma sample in positive mode, and (F) blank plasma sample in negative mode.

Fig. 2 Structures of the identical compounds in Fugan Fang.

Qualitative analysis of FGF constituents in rat plasma

The XICs of blank plasma and drug-containing plasma samples in the positive and negative ion modes are shown in Fig. 1. A total of 25 compounds, including 14 prototype components and 11 potential metabolites, were detected in the drug-containing plasma (Table 2). Among these prototype compounds were 4 phenolic acids, 4 isoflavonoids, 3 iridoid glycosides, and several other types on compounds. In addition to the 14 prototype compounds, 11 peaks were tentatively predicted to be metabolites of FGF components, and these were divided into 3 groups: isoflavonoid-related, iridoid glycoside-related, and phenolic acid-related metabolites. Most of the metabolites were phase II metabolites, including +C₆H₈O₆ (+176 Da) glucuronide conjugates and +SO₃ (+80 Da) sulphate conjugates. For example, peak 62 was confirmed as 9,10-dimethoxypterocarpan glucuronide, based on fragmentation in the spectra of 9,10-dimethoxypterocarpan in (−) ESI-MS and (+) ESI-MS modes, as well as previously published data.¹² Ions [M + Na]⁺ at m/z 499.12108, [M + H]⁺ at m/z 477.13914, and [M − H]⁻ at m/z 475.12459 were observed at 23.5 min. MS/MS fragment ions of peak 62 at m/z 323 [M + Na–Glu]⁺, 301 [M + H–Glu]⁺, and 191, 167 were identified in positive ion ESI mode, and peaks at m/z 299 [M − H–Glu]⁻, 284, 269, 175, and 113 were identified in negative ion ESI mode. Most of the metabolites formed by glucuronide conjugation and sulphate conjugation were identified based on comparison of accurate mass and MS/MS information with that of the prototype compounds. Some other phase I metabolites were identified according to the literature and common metabolic pathways, mainly consisting of gentiopicroside hydroxylation products.

Table 2 Identified compounds in rat plasma after the oral administration of Fugan Fang

No.	Identified compounds	Negative ion (m/z)		Positive ion (m/z)		Molecular		Fragment ions (m/z)
		Adduct	Error (ppm)	Adduct	Error (ppm)	Weight (Da)	Composition
a Confirmation in comparison with literature.b Confirmation in comparison with authentic standards.
Prototype components
4	Protocatechuic acid	−H	0.2			154	C7H6O4	153 [M − H]^−, 109 [M − HCOO]^−, 91 [M − HCOO–H2O]^−^b
11	Loganic acid or 8-hydroxy-10-hydrosweroside	−H	0.7	+H	1	376	C16H24O10	375 [M − H]^−, 213 [M − H–Glc]^−, 169 [M − H–Glc–COO]^−
12	Hydroxysafflor yellow A	−H	0.5	+H	−0.3	612	C27H32O16	613 [M + H]^+, 451 [M + H–Glc]^+, 433, 313, 211^a^,^b
15	Gentiopicroside	−H	−0.5	+H	−0.2	356	C16H20O9	357 [M + H]^+, 195 [M + H–Glc]^+, 177 [M + H–Glc–H2O]^+, 149, 121^+^a^,^b
16	Protocatechuic acid isomer	−H	0.2			154	C7H6O4	153 [M − H]^−, 109 [M − HCOO]^−, 91 [M − HCOO–H2O]^−^b
17	Sweroside	−H	−0.6	+H	0	358	C16H22O9	359 [M + H]^+, 197 [M + H–Glc]^+, 179 [M + H–Glc–H2O]^+, 127^a^,^b
21	p-Coumaric acid	−H	0.1			164	C9H8O3	163 [M − H]^−, 119 [M − H–COO]^−, 93^b
22	Calycosin-7-O-β-D-glucoside	−H	−0.1	+H	−0.4	446	C22H22O10	447 [M + H]^+, 285 [M + H–Glc]^+, 270 [M + H–Glc–CH3]^+^a^,^b
23	Ferulic acid	−H	1.1			194	C10H10O4	193 [M − H]^−, 178 [M − H–CH3]^−, 134 [M − H–CH3–COO]^−^a^,^b
31	Ononin			+H	−0.2	430	C22H22O9	431 [M + H]^+, 269 [M + H–Glc]^+, 254, 237, 226^a
34	Daidzein	−H	0.6	+H	−0.9	254	C15H10O4	253 [M − H]^−, 224 [M − H–CHO]^−, 208, 132^a
41	Senkyunolide F	−H	0.1	+H	−0.6	206	C12H14O3	205 [M − H]^−, 161, 131, 106
42	Formononetin	−H	1.9	+H	0.4	268	C16H12O4	269 [M + H]^+, 254 [M + H–CH3]^+, 226 [M + H–CH3–CO]^+, 197^a^,^b
46	Astragaloside IV	−H	−0.8	+Na	−0.6	784	C41H68O14	807 [M + Na]^+, 627 [M + Na–H2O–Glc]^+^a^,^b
Metabolites
54	Protocatechuic acid-sulfate	−H	0.7			233	C7H6O7S	232 [M − H]^−, 153 [M − H–SO3]^−, 109, 91
55	Protocatechuic acid-glucuronide	−H	0.5			330	C13H14O10	329 [M − H]^−, 153 [M − H–GA]^−, 109, 91
56	Ferulic acid-glucuronide	−H	0.7			370	C16H18O10	369 [M − H]^−, 193 [M − H–GA]^−, 178, 134
57	p-Coumaric acid-sulfate	−H	3.5			244	C9H8O6S	243 [M − H]^−, 163 [M − H–SO3]^−, 119, 93
58	Gentiopicroside-hydroxylate			+Na	−0.2	374	C16H22O10	397 [M + Na]^+, 335, 231, 217, 199, 173
59	Ferulic acid-sulfate	−H	0.9			274	C10H10O7S	273 [M − H]^−, 193 [M − H–SO3]^−, 178, 134
60	Calycosin-glucoside-glucuronide			+H	1	622	C28H30O16	623 [M + H]^+, 447 [M + H–GA]^+, 285 [M + H–GA–Glc]^+
61	Formononetin-glucuronide	−H	0			444	C22H20O10	443 [M − H]^−, 267 [M − H–GA]^−, 252, 223, 195
62	9,10-Dimethoxypterocarpan-glucuronide			+Na	−0.4	476	C23H24O11	499 [M + H]^+, 323 [M − H–GA]^+
63	Formononetin-sulfate	−H	2.4			348	C16H12O7S	347 [M − H]^−, 267 [M − H–SO3]^−, 252, 223, 195
64	7,2′-Hydroxy-3′,4′-dimethoxyisoflavan-glucuronide			+H	−0.5	478	C23H25O11	479 [M + H]^+, 303 [M − H–GA]^+

---

Method validation for quantitative analysis

Typical selected reaction monitoring chromatograms of 8 FGF components (sweroside, gentiopicroside, p-coumaric acid, calycosin-7-O-β-D-glucoside, ferulic acid, ononin, formononetin, and protocatechuic acid) and 2 internal standards are shown in Fig. 3. All calibration curves show good linearity (r² > 0.9985) between the peak area and concentration. The LOD, LLOQ, ULOQ, and QC levels are shown in Table 3. The accuracy of the intra- and inter-day variation of the investigated compounds was in the range of 99.14% to 103.43%, and the precision was less than 7.76%. The results of the recovery test were in the range of 90.69–95.09%, and the RSD values were less than 10.03%. The matrix effect of the analytes was found to be in the range of 89.50–101.56%, and the RSD values were below 11.89%. The results (see ESI†) show that the analytes were stable under the conditions mentioned above, and that the cross-interference between the 4 types of tested compounds was acceptable and negligible in the present conditions. The LC-MS/MS method complies with the requirements of biological sample determination.

Fig. 3 Typical selected reaction monitoring chromatograms of (I) calycosin-7-O-β-D-glucoside, (II) ononin, (III) ferulic acid, (IV) p-coumaric acid, (V) gentiopicroside, (VI) sweroside, (VII) formononetin, (VIII) protocatechuic acid, (IX) wogonoside, and (X) Rhein in rat plasma samples. (A) Blank plasma spiked with reference compounds (I) 129 ng mL⁻¹, (II) 71 ng mL⁻¹, (III) 512 ng mL⁻¹, (IV) 2580 ng mL⁻¹, (V) 463 ng mL⁻¹, (VI) 165 ng mL⁻¹, (VII) 26.4 ng mL⁻¹, (VIII) 1020 ng mL⁻¹, (IX) 135 ng mL⁻¹, and (X) 135 ng mL⁻¹; (B) Blank rat plasma; and (C) plasma sample 0.5 h after the oral administration of 1.08 g kg⁻¹ Fugan Fang in rats.

Table 3 The LOD, LLOQ and QC levels for calycosin-7-O-β-D-glucoside, ononin, ferulic acid, p-coumaric acid, gentiopicroside, sweroside, formononetin, and protocatechuic acid in plasma (n = 5)

Compounds	LOD^b (ng mL^−1)	LLOQ^a (ng mL^−1)	ULOQ^c (ng mL^−1)	QC level (ng mL^−1)
				Low	Middle	High
a The lowest limit of quantitation.b The limit of detection.c The upper limit of quantitation.
Calycosin-7-O-β-D-glucopyranoside	0.2	0.5	161.0	1.0	16.1	129.0
Ononin	0.1	0.3	88.8	0.6	8.9	71.0
Ferulic acid	1.0	2.0	640.9	4.0	64.0	512.0
p-Coumaric acid	6.0	10.1	3230.0	20.2	323.0	2580.0
Gentiopicroside	0.2	1.8	578.4	3.6	57.8	463.0
Sweroside	0.2	0.7	412.0	1.3	20.6	165.0
Formononetin	0.1	0.1	330.0	0.2	3.3	26.4
Protocatechuic acid	2.5	4.0	1280.0	8.0	128.0	1020.0

---

Pharmacokinetics

The quantitative analysis method was successfully used to determine the plasma concentrations of 8 FGF components (sweroside, gentiopicroside, p-coumaric acid, calycosin-7-O-β-D-glucoside, ferulic acid, ononin, formononetin, and protocatechuic acid) following a single oral administration of 1.08 g kg⁻¹ FGF extract to 6 rats. The mean plasma concentration–time profiles of gentiopicroside, calycosin-7-O-β-D-glucoside, ferulic acid, p-coumaric acid, ononin, and sweroside after oral administration of FGF are shown in Fig. 4. The concentrations of gentiopicroside and sweroside were relatively high, but the concentrations of ononin and calycosin-7-O-β-D-glucoside were very low. The concentrations of gentiopicroside and sweroside were lower than their respective LLOQ values after 48 h and 24 h, respectively. Both ferulic acid and p-coumaric acid could not be quantified after 4 h. The concentrations of calycosin-7-O-β-D-glucoside and ononin were lower than their respective LLOQ values after 2 h and 8 h, respectively. The plasma concentrations of protocatechuic acid and formononetin rapidly declined and were only detected at a few time points, and their concentration–time curves were not constructed.

Fig. 4 Profiles of mean plasma concentration–time of six components after a single oral dose of 3 g kg⁻¹ Fugan Fang extract in rats (n = 6, mean ± SD).

The main pharmacokinetic parameters are shown in Table 4. Maximum concentrations were reached within 20 min for calycosin-7-O-β-D-glucoside, ferulic acid, and p-coumaric acid, within 30 min for ononin, and at 1.5 h for gentiopicroside and sweroside. The elimination half-life of p-coumaric acid was less than 2 h, and the elimination half-life of gentiopicroside was greater than 13 h. There were significant differences in the pharmacokinetic parameters of the 6 tested compounds, with a 17-fold difference in T_max, 8-fold difference in t_1/2, 2231-fold difference in C_max, and 2721-fold difference in AUC.

Table 4 Pharmacokinetic parameters of calycosin-7-O-β-D-glucoside, ononin, gentiopicroside, sweroside, ferulic acid, and p-coumaric acid after the oral administration of Fugan Fang at a dose of 3 g kg⁻¹ to SD rats (n = 6; mean ± SD)

Parameters	Ononin	Calycosin-7-O-β-D-glucoside	Gentiopicroside	Sweroside	Ferulic acid	p-Coumaric acid
t1/2 (h)	8.64 ± 4.02	4.08 ± 2.22	13.97 ± 6.32	5.42 ± 1.76	5.16 ± 8.79	1.67 ± 0.31
Tmax (h)	0.29 ± 0.10	0.19 ± 0.09	1.42 ± 0.66	1.42 ± 0.66	0.08 ± 0.02	0.16 ± 0.10
Cmax (ng mL^−1)	0.84 ± 0.21	2.75 ± 0.61	1785.00 ± 608.76	1104.33 ± 378.14	54.65 ± 9.83	252.67 ± 62.46
AUC0–t (ng h L^−1)	3.84 ± 1.34	3.14 ± 1.12	8544.42 ± 3041.62	4861.53 ± 1564.82	40.94 ± 20.75	372.40 ± 114.24
CL/F (L h^−1 kg^−1)	0.01 ± 0.00	63.84 ± 27.58	2.12 ± 0.63	0.08 ± 0.02	0.01 ± 0.00	0.59 ± 0.16
MRT0–t (h)	4.16 ± 1.38	1.52 ± 0.74	1.95 ± 1.54	3.02 ± 0.32	1.60 ± 0.60	1.83 ± 0.21

---

Discussion

The complexity of constituents, background interference, and isomer interference are the major challenges to the identification of the components of TCM using HRMS. In the present study, formic acid was added to the mobile phase to increase the efficiency of ionisation and achieve satisfactory sensitivity, and fast UFLC conditions were selected to obtain chromatograms with better resolution of adjacent peaks within a given time. We accurately identified many peaks with the same mass using a TripleTOF 5600 instrument, and then identified FGF components by referring to standards or to data from the literature, including molecular mass, accurate mass, formula, and MS2 information. Our method allowed us to comprehensively analyse the complex FGF mixture and identify 53 compounds, demonstrating its utility for the analysis of complex extracts from natural sources.

FGF is a traditional Chinese medicinal preparation with good therapeutic effects against hepatic diseases. However, a lack of information on the absorbed components of FGF prevents elucidation of its effective constituents. In the present study, based on the identification of the constituents of FGF extract, 14 prototypes and 11 metabolites were identified in rat plasma after oral administration of FGF. Iridoid glycosides, flavonoids, and flavonoid glycosides were the most abundant components in FGF. In addition, some organic acids, including glycoside chalcones and astragalosides, were also detected. These results reveal that sulphation and glucuronidation were the main metabolic pathways of the flavonoids and organic acids found in FGF. FGF compounds absorbed into the plasma may be the effective components of FGF, due to their extensive biological activities, which include scavenging free radicals, inhibiting the production of TNF-α and TGF-β,¹³ inhibiting collagen synthesis and proliferation in hepatic stellate cells,¹⁴ antioxidant and anti-inflammatory effects,¹⁵ lipids-lowering effects,¹⁶ anti-apoptotic effects,¹⁷ and hepatoprotective effects.^13–18

Establishing a sensitive method for the simultaneous determination of multiple compounds in rat plasma, including compounds with different structures and large concentration differences, has been a significant obstacle to the clarification of the pharmacokinetics of TCM compounds. In this study, 8 FGF components (sweroside, gentiopicroside, p-coumaric acid, calycosin-7-O-β-D-glucoside, ferulic acid, ononin, formononetin, and protocatechuic acid) with measurable plasma concentrations were selected to establish the quantitative analysis method. Because some compounds, such as calycosin-7-O-β-D-glucoside, formononetin, ononin, gentiopicroside, sweroside, and wogonoside, exhibited favourable sensitivity to positive ion mode detection, while others, such as ferulic acid, protocatechuic acid, p-coumaric acid, and rhein, were found to be more sensitive to negative ion mode detection, SRM scanning with switching of the electrospray ion source polarity between the positive and negative modes in a single run was employed for the quantification. The most abundant fragment ions in the spectra were selected for quantification. The reference compounds were also used to optimise mass parameters in a manner that met the demands of the quantitative analysis based on the lowest interference level and the highest signal intensity. Post-column direct infusion was used to optimise liquid chromatography and mass spectrometric detection conditions because of its minimal influence on the signal response of the sample detection. Using these optimised conditions, the matrix effect was investigated. The established UPLC-MS/MS method for simultaneous determination of 8 compounds in rat plasma was used to clarify the pharmacokinetic features of several FGF compounds.

It was found that the concentrations of the measured components in the plasma were different after oral administration of FGF. The concentrations of iridoid glycosides, including gentiopicroside and sweroside, were highest, followed by organic acids, including p-coumaric acid and ferulic acid, which were present at higher concentrations than flavonoids, including ononin and calycosin-7-O-β-D-glucoside. Only small amounts of protocatechuic acid and formononetin were detected in the rat plasma after oral administration of FGF.

Pharmacokinetic results showed that the absorption and elimination processes of the components of FGF also differed among the compounds. The absorption of the most active components, such as calycosin-7-O-β-D-glucoside, ferulic acid, ononin, and p-coumaric acid, occurred rapidly, and these compounds reached their maximum concentrations within 30 min. The elimination of p-coumaric acid was very rapid, and the elimination of gentiopicroside was relatively slow. The rapidly absorbed components may be the fast acting constituents of FGF, whereas the slowly eliminated components may be the long acting constituents of FGF. These differing pharmacokinetic features of the components of FGF may be advantageous to the overall therapeutic effect of FGF on liver disease.

Prior to this research, there have been no information on the pharmacokinetics of the compounds after the oral administration of FGF, even though several pharmacokinetic studies of the individual herbs, Radix Astragali,¹⁹ Radix Angelicae sinensis,²⁰ Flos Carthami,²¹ Radix Gentianae,²² in FGF have been reported. In the present study of the Chinese medicinal formula, clarification of the constituents and their pharmacokinetics of FGF in rats will help researchers understand the effective constituents and provide essential information that will support future development of FGF constituents for clinical use. However, in this study we only examined the pharmacokinetic parameters of FGF constituents in rat plasma. Therefore, further research should be performed on the pharmacokinetics of FGF constituents in other bio-fluids, including urine and bile, and in other target organs, such as the liver.

Conclusion

In the present study, HRMS with a target-directed, database-dependent strategy was developed to analyse FGF extract and rat plasma after oral administration of FGF. A total of 53 compounds were identified in the FGF extract, and 14 prototype components and 11 potential metabolites were detected in the drug-containing plasma. A method for the simultaneous quantification of calycosin-7-O-β-D-glucoside, formononetin, ononin, gentiopicroside, sweroside, ferulic acid, protocatechuic acid, and p-coumaric acid from FGF in rat plasma by UPLC-MS/MS was established and validated. The pharmacokinetic parameters of calycosin-7-O-β-D-glucoside, ononin, gentiopicroside, sweroside, ferulic acid, and p-coumaric acid in rat plasma after oral administration of FGF were revealed.

Abbreviations

HPLC	High-performance liquid chromatography
MS	Mass spectrometry
TCM	Traditional Chinese medicine
FGF	Fugan Fang
HRMS	High resolution mass spectrometry
RSD	Relative standard deviation

Acknowledgements

The project was supported by Program for Shanghai Innovative Research Team in University (2009), “085” First-Class Discipline Construction of Science and Technology Innovation (085ZY1205) and National S&T Major Project (2009ZX09311-003, 2012ZX09303009-001).

References

1. M. H. Chen, J. C. Chen, C. C. Tsai, W. C. Wang, D. C. Chang, C. C. Lin and H. Y. Hsieh, Am. J. Chin. Med., 2004, 32, 195–207.
2. Y. Imanishi, N. Maeda, K. Otogawa, S. Seki, H. Matsui, N. Kawada and T. Arakawa, J. Hepatol., 2004, 41, 242–250.
3. T. Ochi, T. Kawakita and K. Nomoto, Immunopharmacol. Immunotoxicol., 2004, 26, 285–298.
4. T. Tamura, H. Kobayashi, A. Yamataka, G. J. Lane, H. Koga and T. Miyano, Pediatr. Surg. Int., 2007, 23, 343–347.
5. H. Cui, W. Gu and Y. Ma, Pharmacol. Clin. Chin. Mater. Med., 2012, 28, 137–141.
6. W. Zhao, Z. M. Zhao, Z. T. Wang, B. X. Zhu and C. H. Liu, Pharmacol. Clin. Chin. Mater. Med., 2012, 28, 15–18.
7. Z. Zhao, H. Yu, Y. Peng, H. Ren, Y. Tao, Z. Wang, B. Zhu, P. Liu and C. Liu, China J. Chin. Mater. Med., 2012, 37, 1804–1808.
8. X. J. Wang, Modernization of Traditional Chinese Medicine and Materia Medica-World Science and Technology, 2002, 4, 1–4.
9. Z. R. Liu, M. Liu, Y. P. Qi, Z. Y. Zhu, Y. F. Chai, C. Yuan and Y. Lin, J. Sep. Sci., 2013, 36, 1659–1666.
10. P. Jiang, R. Shi, Q. Wang, Y. M. Ma, H. Y. Cui, P. Liu and C. H. Liu, Biomed. Chromatogr., 2013, 27, 874–881.
11. Y. G. QI Jingjing and S. Fengrui, J. Chin. Mass Spectrom. Soc., 2009, 30, 334–339.
12. X. D. Wen, E. H. Liu, J. Yang, C. Y. Li, W. Gao, L. W. Qi, C. Z. Wang, C. S. Yuan and P. Li, J. Pharm. Biomed. Anal., 2012, 67–68, 114–122.
13. S. Y. Gui, W. Wei, H. Wang, L. Wu, W. Y. Sun, W. B. Chen and C. Y. Wu, J. Ethnopharmacol., 2006, 103, 154–159.
14. H. Liu, W. Wei, W. Y. Sun and X. Li, J. Ethnopharmacol., 2009, 122, 502–508.
15. H. Y. Kim, J. Park, K. H. Lee, D. U. Lee, J. H. Kwak, Y. S. Kim and S. M. Lee, Toxicology, 2011, 282, 104–111.
16. Q. Y. Wang, Q. H. Meng, Z. T. Zhang, Z. J. Tian and H. Liu, Acta Pharm. Sin., 2009, 44, 386–389.
17. L. H. Lian, Y. L. Wu, Y. Wan, X. Li, W. X. Xie and J. X. Nan, Chem.-Biol. Interact., 2010, 188, 127–133.
18. Y. D. Luo, J. Chen, J. Cao, X. D. Wen and P. Li, Chem. Pharm. Bull., 2009, 57, 79–83.
19. X. H. Liu, J. B. Zhao, L. Guo, Y. L. Yang, F. Hu, R. J. Zhu and S. L. Feng, Cell Biochem. Biophys., 2014, 70, 677–686.
20. X. D. Liu, Y. Y. Xue, L. Xie, H. F. Ji and G. J. Wang, J. China Pharm. Univ., 2003, 34, 448–451.
21. J. Tang, Z. Ouyang, X. Feng, J. Fang and X. Cao, China J. Chin. Mater. Med., 2011, 36, 2246–2249.
22. C. H. Wang, X. M. Cheng, Y. Q. He, K. N. White, S. W. Bligh, C. J. Branford-White and Z. T. Wang, Arch. Pharmacal Res., 2007, 30, 1149–1154.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra14148b

---