Journal of Chromatography B, 971 (2014) 81–88
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Simultaneous quantification of Akebia saponin D and its five metabolites in human intestinal bacteria using ultra-performance liquid chromatography triple quadrupole mass spectrometry Liang Yan a , Xiaolin Yang b , Zhaoqing Meng a,c , Yongliang Yuan a , Wei Xiao c , Zhenzhong Wang c , Wenze Huang c , Zhonglin Yang a,∗ , Chunfeng Zhang a,∗∗ a
State Key Laboratory of Natural Medicines, China Pharmaceutical University, Nanjing 210009, PR China Jiangsu Provincial Center for Research and Development of Marine Drugs, Nanjing University of Traditional Chinese Medicine, Nanjing 210029, PR China c Jiangsu Kanion Pharmaceutical Co. Ltd., Lianyungang 222001, PR China b
a r t i c l e
i n f o
Article history: Received 12 July 2014 Accepted 11 September 2014 Available online 19 September 2014 Keywords: Akebia saponin D Metabolites Human intestinal bacteria UPLC-TQ/MS
a b s t r a c t A rapid and sensitive ultra-performance liquid chromatography triple quadrupole mass spectrometry (UPLC-TQ/MS) method was developed for simultaneous quantification of Akebia saponin D (ASD) and its five metabolites in intestinal mixtures of bacteria from human feces. After protein precipitation, the analytes and internal standard (IS), glycyrrhetinic acid, were determined in selected ion recording (SIR) mode with negative ion ESI source. Chromatographic separation was carried out on an ACQUITY UPLCTM BEH C18 column (100 mm × 2.1 mm, 1.7 m) using gradient elution. The mobile phase consisted of solvents A (acetonitrile) and B (0.1% aqueous formic acid) at the flow rate of 0.4 mL/min. Each sample was chromatographed within 10.5 min including equilibration time. The linearity ranged from 0.1 to 100 g/mL for ASD, and 2–1000 ng/mL for five metabolites, Dipsacus saponin A (M1), HN-saponin F (M2), hederagenin28-O--d-glucopyranoside (M3), Akebia saponin PA (M4), hederagenin (M5). The limits of detection (LOD) were 0.41, 0.59, 0.61, 0.55, 0.52 and 0.31 ng/mL for ASD, M1, M2, M3, M4 and M5, respectively. The intra- and inter-day precision was all within 11.1% and accuracy ranged from −8.33% to 12.47%. The conversion rate of five metabolites was 41.21% in 24 h. The method was validated and successfully applied to quantification of ASD and its five metabolites in human intestinal bacteria. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Akebia saponin D (3-O-␣-l-arabinopyranosyl hederagenin28--d-glucopyranoside-(1 → 6)--d-glucopyranoside, Fig. 1) is a main bioactive triterpenoid saponin isolated from the rhizome of Dipsacus asper Wall [1]. In preliminary articles, ASD has been reported to have a protective effect against cardiomyocyte apoptosis [2], neurotoxicity [3], acute myocardial ischemia injury [4], and osteoporosis [5]. In our previous study, ASD had a therapeutic effect on Alzheimer’s disease, but its bioavailability was very low after oral administration in rats [6], which may be due to the
∗ Corresponding author at: State Key Laboratory of Natural Medicines, China Pharmaceutical University, No. 24 Tongjia Lane, Nanjing 210009, PR China. Tel.: +86 25 83271426; fax: +86 25 83271426. ∗∗ Corresponding author. Tel.: +86 25 86 185129; fax: +86 25 86 185129. E-mail addresses: [email protected] (Z. Yang), [email protected] (C. Zhang). http://dx.doi.org/10.1016/j.jchromb.2014.09.013 1570-0232/© 2014 Elsevier B.V. All rights reserved.
metabolism of ASD in intestinal. Therefore, it is necessary to study its metabolic profile. Hydrolyzed reactions were found to be the major metabolic processes of ASD. In rats’ blood, ASD hydrolyzed to be hederagenin [7], which could be absorbed easily by intestinal mucosa. Moreover, ASD’s five hydrolyzed metabolites (M1, M2, M3, M4 and M5, Fig. 1) were identified by HPLC–MSn in rats’ feces [8]. And also there were three metabolites (M2, M4 and M5) of ASD simultaneously determined using LC–MS/MS in rat’s bile [9]. Although some researches on the metabolism of ASD in rats have been carried out, there are few reports about the metabolism of ASD in human. In this paper, we studied on the quantification of ASD and its five metabolites produced by human intestinal bacteria, verified the quantification method and calculated the conversion rate of five metabolites. The intestinal tract contains more than 400 bacterial species, in which Lactobacillus, Bacteroides and Streptococcus are major species, most of them to be strict anaerobes [10,11]. A lot of deconjugating enzymes in intestinal bacteria, e.g. -d-glucosidases, -d-glucuronidases and ␣-l-rhamnosidases [12,13], which release
82
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
determined to be higher than 96% by normalization of the peak area detected by HPLC. The HPLC-grade acetonitrile was purchased from TEDIA Company Inc. (Fairfield, OH, USA); formic acid was obtained from Merck KGaA (Darmstadt, Germany); ultrapure water was purified by an EPED super purification system (Nanjing, China). Other reagents were of analytical grade.
2.2. Preparation of the general anaerobic medium broth The general anaerobic medium (GAM) used for all fermentation experiments consists of tryptone 10.0 g, soya peptone 3.0 g, proteose peptone 10.0 g, digestibility serum powder 13.5 g, yeast extract 5.0 g, beef extract 2.2 g, beef liver extract powder 1.2 g, glucose 3.0 g, KH2 PO4 2.5 g, NaCl 3.0 g, soluble starch 5.0 g, l-cysteine hydrochloride 0.3 g, sodium thioglycolate 0.3 g, and distilled water 1000 mL, and then the pH was adjusted to 7.3 before autoclaving at 0.07 MPa at 121 ◦ C for 30 min.
2.3. Preparation of human intestinal bacterial mixture Fresh human fecal sample which was weighed 5 g and suspended in the GAM broth under a CO2 atmosphere and cultured in the anaerobic incubator at 37 ◦ C for 24 h was obtained from a healthy female Chinese volunteer. The obtained bacterial mixture was used as human intestinal bacterial mixture.
2.4. Incubation experiments
Fig. 1. (A) The chemical structures of ASD and its five metabolites; (B) glycyrrhetinic acid (IS).
glycones and secondary saponins from their glycosides. It is reported that there were 25 metabolites of American ginseng extract detected in human intestinal microflora [14]. The same thing happens to isoquercitrin, platycoside and soyasaponin Ab [15–17]. Due to the difficulty of getting reference standards, most studies only did qualitative analysis of metabolites. However, we established ultra-performance liquid chromatography triple quadrupole mass spectrometry (UPLC-TQ/MS) method for simultaneous quantification of ASD and its five metabolites for the first time, and calculated the conversion rate of five metabolites after ASD incubated 24 h. It described the parts of metabolic pathways for the biotransformation of ASD in human intestinal bacteria.
The standard solution of ASD was prepared by dissolving accurately weighed ASD in purified water to obtain a final concentration of 1 mg/mL. The 60 bacterial mixtures were incubated into 1 mL of GAM broth containing 0.1 mg/mL ASD in plastic tube, respectively, and the media were anaerobically incubated at 37 ◦ C. Five samples were taken out and frozen at −20 ◦ C in every 2 h. The bacterial mixture without ASD was incubated 24 h as a blank sample.
2.5. Preparation of standard and quality control (QC) samples The appropriate amounts of ASD, M1, M2, M3, M4 and M5 were separately weighed and dissolved in methanol as the stock solutions. Then, six stock solutions were mixed and diluted with methanol to give a final mixed standard solution containing 0.1 mg/mL of ASD and its five metabolites standard, and kept at 4 ◦ C when not in use. A series of working solutions of these analytes were diluting the mixed standard solutions with methanol at concentrations ranging from 100 to 100 000 ng/mL for ASD, 2 to 1000 ng/mL for five metabolites, respectively. For the validation of method, low, medium and high QC samples with concentrations of 100, 1000, 100 000 ng/mL for ASD, and 10, 100, 1000 ng/mL for five metabolites standard, were similarly prepared.
2. Experimental 2.6. Sample preparation 2.1. Materials Akebia saponin D (93.5% purity) and glycyrrhetinic acid (>98.0% purity) were purchased from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). Dipsacus saponin A, HN-saponin F, Akebia saponin PA, hederagenin-28-O--d-glucopyranoside, hederagenin were produced in our laboratory. Structures were confirmed by spectroscopic methods (MS, 1 H NMR and 13 C NMR), and their purity was
An aliquot of 1 mL of each bacterial sample was mixed with 3 mL methanol and 50 L I.S. solution (2 g/mL), vortex-mixed approximate for 1 min, and centrifuged at 17,800 × g for 10 min. The supernatant was transferred into another tube and evaporated to dryness at 50 ◦ C in a centrifugal vacuum evaporator (LABCONCO Corp., Kansas City, Missouri, USA). The residues were dissolved in 1 mL of 40% methanol, centrifuged at 17,800 × g for 10 min, and supernatant was analyzed by UPLC–MS.
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
2.7. Chromatographic conditions Chromatography was performed on a Waters Acquity UPLC system (Waters Corp., Milford, MA, USA). ACQUITY UPLCTM BEH C18 column (100 mm × 2.1 mm, 1.7 m) was used for separation. Column temperature was set at 35 ◦ C. Chromatographic separation was achieved by gradient elution at the flow rate of 0.4 mL/min. The mobile phase consisted of solvents A (acetonitrile) and B (0.1% aqueous formic acid). The gradient was as follows: 30% A at 0–4 min, 30–80% A at 4–7 min, 80–90% A at 7–7.5 min, 90% A at 7.5–8.5 min, and then an immediate reduction to 30% A at 9.5 min; 9.5–10.5 min, 30% A for equilibration of the column. Injection volume was 5 L.
83
2.9.5. Stability The stability of analytes in intestinal bacteria was evaluated by analyzing of three levels of QC samples stored at −80 ◦ C for 21 days (long term stability), at 25 ◦ C for 24 h (short term stability) and after three freeze/thaw cycles. The autosampler stability was analyzing QC samples at 4 ◦ C for 24 h. 2.10. Conversion rate of five metabolites After incubated 24 h, concentration–time curves of ASD and its five metabolites were obtained. The conversion rate of five metabolites was calculated using following equation: CM MASD × × 100% CASD MM
2.8. Mass spectrometric conditions
Conversion rate (%) =
A Waters ACQUITYTM Synapt mass spectrometer (Waters Corp., Milford, MA, USA) was connected to the UPLC system via an electrospray ionization (ESI) interface. The ion source was operated in negative ionization mode for the six compounds and IS. Quantification was performed using selected ion recording (SIR) mode of the transitions of m/z 928 → m/z 603 for ASD, m/z 795 → m/z 471 for M1, m/z 765 → m/z 603 for M2, m/z 633 → m/z 471 for M3, m/z 603 for M4, m/z 471 for M5 and m/z 469 for the internal standard. The MS parameters were as follows: source temperature 120 ◦ C, desolvation temperature 350 ◦ C, capillary voltage 3 kV, cone voltage 30 V, desolvation gas 600 L h−1 , cone gas 50 L h−1 . All data were acquired using the Waters Masslynx 4.1 software (Waters Corp., Milford, MA, USA).
where “CM ” is increased concentration of metabolites at 24 h, “CASD ” is decreased concentration of ASD in 24 h, “MASD ” is molecule weight of ASD, “MM ” is molecule weight of metabolites.
2.9. Method validation The proposed quantitative method was validated for selectivity, linearity, precision, accuracy, extraction recovery, matrix effect and stability according to the guidance of the Food and Drug Administration (FDA) for validation of bioanalytical methods [18]. 2.9.1. The selectivity The selectivity of the method was accessed by comparing the chromatograms of blank bacterial samples to those of corresponding standard samples spiked with analytes and IS and bacterial samples with ASD after incubated 18 h. 2.9.2. Linearity and lower limit of quantification Calibration curves were constructed by plotting the peak area ratios of analytes to IS v/s the nominal concentrations of standard analytes, respectively, using the weighting factor of 1/C2 . The limit of detection and quantification was determined at a signal-to-noise of 3 and 10 by analyzing the standard bacterial sample. 2.9.3. Precision and accuracy The intra- and inter-day precision and accuracy were determined by analyzing six replicates of QC samples at low, medium, high concentration levels. The relative standard deviation (RSD) was used for reporting precision. The accuracy was calculated from (measured − nominal)/nominal × 100%. The intra- and interday precision should not exceed 15% and accuracy should be within ±15% for the QC samples. 2.9.4. Extraction recovery and matrix effect The extraction recoveries of analytes were determined at three different concentrations of QC samples by comparing area ratios of analytes in the post-extraction spiked samples to that acquired from pre-extraction spiked samples. The matrix effect was evaluated by determining the peak area ratios of the analytes in post-extracted spiked samples to those acquired from unextracted spiked samples.
3. Results 3.1. Optimization of mass spectrometry In this experiment, MS/MS operation parameters were carefully optimized for determination of ASD and its five metabolites. The mass spectrometer was tuned in both positive and negative ionization mode with ESI source for optimum response of each analyte. ASD and its five metabolites in negative ion mode had better response, so we used the negative ion as the detection mode. In negative ion mode, ASD, M1, M2, M3, M4, M5 and I.S. formed the deprotonated molecular ions of m/z 928, m/z 795, m/z 765, m/z 633, m/z 603, m/z 471 and m/z 469, respectively. The detection of target analytes was firstly done in multiple reaction monitoring (MRM) mode, but we could not obtain good ion response either high energy or low energy. When we changed to use selected ion recording (SIR) mode, we obtained the good ion response of ASD and its metabolites and I.S. Quantification was performed using SIR mode of the transitions of m/z 928 → m/z 603 for ASD, m/z 795 → m/z 471 for M1, m/z 765 → m/z 603 for M2, m/z 633 → m/z 471 for M3, m/z 603 for M4, m/z 471 for M5 and m/z 469 for the I.S. Parameters such as source temperature, desolvation temperature, capillary and cone voltage, flow rate of desolvation gas and cone gas were optimized to obtain optimum response of analytes. 3.2. Optimization of LC To optimize chromatographic conditions, a suitable analytical column was at first selected to obtain high sensitivity and good separation. An ACQUITY UPLCTM BEH C18 column was the most universal column for UPLC separations. It improved the retentions of analytes and achieved the good separation. The composition of the mobile phase had an effect on the separation and ionization of analytes. Acetonitrile and methanol were both attempted as the organic modifier of the mobile phase. It was found that signalto-noise of analytes was higher with acetonitrile than those with methanol. The addition of 0.1% formic acid in aqueous portion of mobile phase produced more symmetrical peaks for M4, M5 and I.S. because of the carboxyl group of their structures. The flow rate of mobile phase, column temperature and injection volume on separation behavior were also investigated. The flow rate of 0.3 mL/min was too slow that prolonged retention time, and we obtained the good separation at the flow rate of 0.4 mL/min; There were no significant effects on separation between the column temperature 30 ◦ C and 35 ◦ C, and we selected the column temperature 35 ◦ C; The injection volume was dependent on the concentrations of analytes
84
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
and their response in the instrument, and we selected moderate injection volume 5 L in this experiment. In order to optimize peak shape, retention time and separation, several trails to screen the ratio of acetonitrile to water were carried out. Finally, a gradient elution system (acetonitrile–0.1% aqueous formic acid) as previously described was chosen. Under this chromatographic condition, the effective separation was achieved for ASD and its metabolites and IS. 3.3. Selection of internal standard According to FDA guidance, an internal standard in the analysis of biological sample could be a structurally similar analog of analyte or a stable labeled compound [18]. Glycyrrhetinic acid was adopted as I.S. due to the similarity of structure and extraction efficiency with the analytes, and its efficient ionization in the negative ionization mode. 3.4. Sample preparation The most widely employed biological sample preparation techniques are protein precipitation (PPT), liquid–liquid extraction (LLE) and solid-phase extraction (SPE). The recovery and matrix effect of SPE are sufficient but it requires much time and money. The different polar character of ASD and its five metabolites need different extraction solvent, which makes sample preparation very complicated. Thus, the PPT method was used to extract the analytes from the intestinal bacteria. Methanol and acetonitrile were tested to use as the precipitation solvents. The test results showed that there were little difference between the recoveries of ASD, M1 and M2, but methanol gave much higher recoveries of M3, M4 and M5 than acetonitrile. Thus, methanol was finally chosen as the protein precipitation reagent in this experiment. 3.5. Method validation 3.5.1. The selectivity Fig. 2 shows representative extract ions SIR chromatograms of ASD, five metabolites of ASD and IS. There is no significant interference in the determination of the analytes and IS. 3.5.2. Linearity and lower limit of quantification The regression equations for calibration curves and LLOQs of ASD and five metabolites are listed in Table 1. The results demonstrated that this method is sensitive enough to quantitative detection of ASD and its metabolites. 3.5.3. Precision and accuracy In this assay, the intra- and inter-day precision and accuracy were summarized in Table 2. At each QC level, the intra- and interday precisions (RSD) of those analytes were within 10.89%. The accuracy of those analytes was within 12.47%. The results show that this method has a satisfactory precision and accuracy. 3.5.4. Extraction recovery and matrix effect The extraction recoveries and matrix effects for ASD and its five metabolites are shown in Table 3. The extraction recoveries of ASD and five metabolites were in the range from 73.3% to 95.4%, and the extraction recovery of IS was 90.7%. The matrix effects of ASD and five metabolites were in the range from 84.53% to 101.13%, and the matrix effects of IS was 92.1%. The results showed that there is no significant matrix effect for analytes and IS. 3.5.5. Stability The stability of ASD and its five metabolites in intestinal bacteria are summarized in Table 4, which demonstrated the good stability
Fig. 2. Representative extract ion SIR chromatograms of ASD, M1, M2, M3, M4, M5 and glycyrrhetinic acid (IS): (A) blank intestinal bacteria; (B) blank intestinal bacteria spiked with the six analytes and IS; (C) ASD incubated in intestinal bacteria 18 h.
of ASD and its five metabolites at −80 ◦ C for 21 days, at 25 ◦ C for 24 h, and after three freeze/thaw cycles, as well as in post-prepared samples at 4 ◦ C for 24 h. 3.6. Conversion rate of five metabolites After incubated 24 h, the concentration–time curves of ASD and its five metabolites are shown as Fig. 3. The concentration of ASD was decreased from 99.394 g/mL to 96.850 g/mL, while the
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
85
Table 1 The regression equations, linear ranges, LOQs, and LODs of ASD and its five metabolites (n = 3). Analytes
Regression equation
R2
Linear range (ng/mL)
LOD (ng/mL)
LOQ (ng/mL)
ASD M1 M2 M3 M4 M5
y = 0.00106x + 0.00409 y = 0.00277x + 0.01160 y = 0.0018x +0.00416 y = 0.00245x + 0.00549 y = 0.00344x + 0.00801 y = 0.00394x + 0.12540
0.9998 0.9994 0.9999 0.9998 0.9987 0.9983
100–100,000 2–1000 2–1000 2–1000 2–1000 2–1000
0.41 0.59 0.61 0.55 0.52 0.31
1.32 1.91 1.95 2.00 1.87 1.05
Table 2 The inter-day and intra-day accuracies and precisions for ASD and five metabolites (n = 5). Analytes
Concentration (ng/Ml)
ASD
100 1000 100000 10 100 1000 10 100 1000 10 100 1000 10 100 1000 10 100 1000
M1
M2
M3
M4
M5
Inter-day (n = 6)
Intra-day (n = 6)
Accuracy (%)
Precision (%)
Accuracy (%)
Precision (%)
9.38 −4.91 0.51 −7.07 5.63 −0.05 −2.92 1.86 −0.14 −6.06 −0.6 1.15 9.56 −4.87 0.1 5.54 9.46 −0.05
4.7 10.89 1.57 10.12 8.86 8.32 11.1 7.65 5.14 9.74 6.06 6.09 7.85 8.1 4.87 10.23 6.23 7.51
5.42 −1.34 1.21 −8.33 4.85 0.14 −2.52 0.85 0.47 −4.87 −2.36 0.26 7.93 −6.58 −0.24 5.74 12.47 0.13
7.07 10 3.25 6.45 6.69 7.8 9.6 8.74 6.88 9.82 7.21 5.94 8.95 8.05 5.32 6.99 9.09 7.83
concentration of M1, M2, M3, M4 and M5 were 94.519, 755.066, 15.905, 183.954 and 5.424 ng/mL. The molecular weight of ASD and five metabolites are 929, 796, 766, 634, 604 and 472, respectively. Therefore, the conversion rate of these metabolites are 3.72%, 29.68%, 0.63%, 7.23% and 0.21%, respectively, total of 41.21%. 3.7. Metabolic pathways The parts of metabolic pathways of ASD in human intestinal bacteria are shown in Fig. 4. It is the same as metabolic pathways of ASD in rats’ feces [8].
4. Discussion The results show that the concentration of ASD is decreased with the reaction time prolonged. As to five metabolites, M1, M2 and M4 can be detected at 2 h, but M3 and M5 are found at 10 h and 6 h. It demonstrated that ASD’s desugar metabolites were produced sequentially. Within 24 h, the concentration of M2 and M4 increased with the reaction time prolonged, while M1, M3 and M5 increased at first but then decreased. It suggested that the production rate of the former is faster than their metabolism speed; the latter would decrease due to the faster metabolism speed.
Table 3 Extraction recoveries and matrix effects of ASD and five metabolites (n = 5). Analytes
ASD
M1
M2
M3
M4
M5
Concentration (ng/mL)
100 1000 100000 10 100 1000 10 100 1000 10 100 1000 10 100 1000 10 100 1000
Recovery (%)
Matrix effect (%)
Average (%)
RSD (%)
Average (%)
RSD (%)
82.76 79.75 94.37 81.6 79.3 86.6 86.8 83.4 90.1 75.2 87.6 85.4 74.1 78.5 93.2 73.3 89.6 95.4
9.8 7.6 5.7 4.9 9.3 6.9 7.3 6.5 7.7 8.9 4.3 5.8 7.8 8.4 5.6 10.1 3.4 6.1
87.83 90.65 96.98 91.6 95.3 92.5 93.93 91.5 94.83 90.93 101.13 97.4 84.5 90.8 97.2 88.2 91.9 98.3
7.8 8.2 4.8 7.6 8.8 6.8 7.2 5.7 7.1 5.5 6.4 3.4 8.9 7.1 6.3 9.5 2.8 4.2
86
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
Table 4 Stability of ASD and five metabolites under different condition; percentage difference from fresh controls. Analytes/storage condition
Conc. (ng/mL)
Three freeze–thaw cycles
Long-term stability −20 ◦ C for 21 d
Short-term stability 25 ◦ C for 24 h
Post-prepared sample 4 ◦ C for 24 h
ASD
100 1000 100000 10 100 1000 10 100 1000 10 100 1000 10 100 1000 10 100 1000
−3.55 9.8 4.09 −8.62 4.12 −0.65 −3.62 4.07 2.14 −1.78 5.18 2.55 −0.06 2.72 −1.98 −2.63 6.35 2.91
−6.55 14.8 8.28 −7.46 2.21 −1.74 −5.13 2.55 −1.69 −2.67 3.69 −3 −4.21 −3.04 −1.84 −1.66 3.84 −4.14
−2.32 4.8 9.09 −6.2 4.18 −0.77 −1.61 1.56 −0.82 −3.06 6.34 3.1 −7.46 −2.91 1.16 8.78 8.71 −5
−5.55 10.5 7.23 −7.26 −2.66 −0.79 −2.92 5.45 0.53 1.58 −1.07 −2.6 −4.97 5.03 −0.57 7.9 −3.47 −1.22
M1
M2
M3
M4
M5
Fig. 3. Concentration–time curves of ASD and its five metabolites in 24 h.
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
87
Fig. 4. Parts of metabolic pathways of ASD in human intestinal bacteria.
The total conversion rate of five metabolites is 41.21%. It suggests that other metabolites were produced in this reaction. It is reported that hydrolysis, hydroxylation, acetylation were the major metabolic pathways of flavonoids in A. manihot extract in intestinal bacteria [19]. Besides, acetylated, hydroxymethylated, dehydroxylated, hydroxylated metabolites could be found after isoquercitrin incubated in human intestinal bacteria [13]. In this process of desugar reaction, ASD can combine with hydroxyl, hydroxymethyl and acetyl group. At the same time, it can occur dehydroxylation, demethylation, dehydroxymethyl reactions. Moreover, five metabolites may occur secondary metabolism. In view of the above points, we might explain the reasons for the low bioavailability of ASD. Not only did ASD exist in blood as prototype, but also a large number of its metabolites may exist. When we do research on the bioavailability of ASD, its metabolites must be considered. Meanwhile, our research suggested that low bioavailability of saponin compounds may be caused by drug metabolism. Such as ginsenosides Rb1, Rb2, Rb3, DT-13 and astragaloside IV [20–22]. 5. Conclusion A selective and sensitive UPLC-TQ/MS method was developed for simultaneous quantification of ASD and its five metabolites with LLOQs of 1.32, 1.91, 1.95, 2.00, 1.87 and 1.05 ng/mL for ASD, M1, M2, M3, M4 and M5, respectively. The total conversion rate of five metabolites is 41.21%. The parts of metabolic pathways for the biotransformation of ASD in human intestinal bacteria were
described. The method was validated and successfully applied to quantification of ASD and its five metabolites in human intestinal bacteria. Authors’ contribution L. Yan and X. Yang contributed equally to this work and should be regarded as cofirst authors. Acknowledgements This work was supported by A Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions. The assistance of the Jiangsu Key Laboratory for High Technology Research of TCM Formulae is gratefully acknowledged. References [1] K.Y. Jung, K.H. Son, J.C. Do, Arch. Pharmacol. Res. 16 (1993) 32–35. [2] C.M. Li, J.W. Tian, G.S. Li, W.L. Jiang, Y.L. Xing, J. Hou, H.B. Zhu, H. Xu, G.B. Zhang, Z.F. Liu, Z.G. Ye, Eur. J. Pharmacol. 649 (2010) 100–107. [3] H.W. Suh, D.K. Song, S.O. Huh, K.H. Son, Y.H. Kim, J. Ethnopharmacol. 71 (2000) 211–218. [4] C.M. Li, Z.F. Liu, J.W. Tian, G.S. Li, W.L. Jiang, Y.L. Xing, J. Hou, H.B. Zhu, H. Xu, G.B. Zhang, Z.F. Liu, Z.G. Ye, Eur. J. Pharmacol. 627 (2010) 235–241. [5] Y.B. Niu, Y.H. Li, H.T. Huang, X.H. Kong, R. Zhang, L. Liu, Y. Sun, T.M. Wang, Q.B. Mei, Phytother. Res. 25 (2011) 1700–1706. [6] K. Li, L. Ding, Z.L. Yang, E.H. Liu, L.W. Qi, P. Li, Y.Z. Hu, Biomed. Chromatogr. 24 (2010) 550–555. [7] H. Zhu, L. Ding, S. Shakya, X.M. Qi, L. Hu, X.L. Yang, Z.L. Yang, J. Chromatogr. B 879 (2011) 3407–3414.
88 [8] [9] [10] [11] [12] [13] [14]
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
K. Li, X.L. Yang, C.F. Zhang, Z.L. Yang, Chin. J. Nat. Med. 7 (2009) 440–443. L. Wang, E.W. Liu, L.F. Han, Tianjin J. Tradit. Chin. Med. 30 (2013) 681–684. J.E. Lee, S. Lee, J. Sung, G. Kol, ISME J. 5 (2011) 362–365. J.C. Rambaud, Ann. Gastroenterol. Hepatol. (Paris) 28 (1992) 263–266. I.A. Macdonald, J.A. Mader, R.G. Bussard, Mutat. Res. Lett. 122 (1983) 95–102. V.D. Bokkenheuser, C.H. Shackleton, J. Winter, Biochem. J. 248 (1987) 953–956. J.Y. Wan, P. Liu, H.Y. Wang, L.W. Qi, C.Z. Wang, P. Li, C.S. Yuan, J. Chromatogr. A 1286 (2013) 83–92. [15] L.L. Lu, D.W. Qian, J. Yang, S. Jiang, J.M. Guo, E.X. Shang, J.A. Duan, Biomed. Chromatogr. 27 (2013) 509–514. [16] Y.W. Ha, Y.C. Na, I.J. Ha, D.H. Kim, Y.S. Kim, J. Pharm. Biomed. Anal. 51 (2010) 202–209.
[17] S.Y. Chang, M.J. Han, E.H. Joh, D.H. Kim, J. Pharm. Biomed. Anal. 52 (2010) 752–756. [18] USFDA, 2001. http://www.fda.gov/downloads/Drugs/Guidance/Compliance Regulatory Information/Guidances/UCM070107.pdf [19] C.F. Xue, S. Jiang, J.M. Guo, D.W. Qian, J.A. Duan, E.X. Shang, J. Chromatogr. B 879 (2011) 3901–3908. [20] J. Zhao, C. Su, C.P. Yang, M.H. Liu, L. Tang, W.W. Su, Z.Q. Liu, J. Pharm. Biomed. Anal. 64 (2012) 94–97. [21] L.L. Chen, W.W. Yuan, Z.F. Hu, J. Qi, D.N. Zhu, B.Y. Yu, J. Pharm. Biomed. Anal. 56 (2011) 650–654. [22] W.D. Zhang, C. Zhang, R.H. Liu, H.L. Li, J.T. Zhang, C. Mao, S. Moran, C.L. Chen, Life Sci. 79 (2006) 808–815.
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Simultaneous quantification of Akebia saponin D and its five metabolites in human intestinal bacteria using ultra-performance liquid chromatography triple quadrupole mass spectrometry Liang Yan a , Xiaolin Yang b , Zhaoqing Meng a,c , Yongliang Yuan a , Wei Xiao c , Zhenzhong Wang c , Wenze Huang c , Zhonglin Yang a,∗ , Chunfeng Zhang a,∗∗ a
State Key Laboratory of Natural Medicines, China Pharmaceutical University, Nanjing 210009, PR China Jiangsu Provincial Center for Research and Development of Marine Drugs, Nanjing University of Traditional Chinese Medicine, Nanjing 210029, PR China c Jiangsu Kanion Pharmaceutical Co. Ltd., Lianyungang 222001, PR China b
a r t i c l e
i n f o
Article history: Received 12 July 2014 Accepted 11 September 2014 Available online 19 September 2014 Keywords: Akebia saponin D Metabolites Human intestinal bacteria UPLC-TQ/MS
a b s t r a c t A rapid and sensitive ultra-performance liquid chromatography triple quadrupole mass spectrometry (UPLC-TQ/MS) method was developed for simultaneous quantification of Akebia saponin D (ASD) and its five metabolites in intestinal mixtures of bacteria from human feces. After protein precipitation, the analytes and internal standard (IS), glycyrrhetinic acid, were determined in selected ion recording (SIR) mode with negative ion ESI source. Chromatographic separation was carried out on an ACQUITY UPLCTM BEH C18 column (100 mm × 2.1 mm, 1.7 m) using gradient elution. The mobile phase consisted of solvents A (acetonitrile) and B (0.1% aqueous formic acid) at the flow rate of 0.4 mL/min. Each sample was chromatographed within 10.5 min including equilibration time. The linearity ranged from 0.1 to 100 g/mL for ASD, and 2–1000 ng/mL for five metabolites, Dipsacus saponin A (M1), HN-saponin F (M2), hederagenin28-O--d-glucopyranoside (M3), Akebia saponin PA (M4), hederagenin (M5). The limits of detection (LOD) were 0.41, 0.59, 0.61, 0.55, 0.52 and 0.31 ng/mL for ASD, M1, M2, M3, M4 and M5, respectively. The intra- and inter-day precision was all within 11.1% and accuracy ranged from −8.33% to 12.47%. The conversion rate of five metabolites was 41.21% in 24 h. The method was validated and successfully applied to quantification of ASD and its five metabolites in human intestinal bacteria. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Akebia saponin D (3-O-␣-l-arabinopyranosyl hederagenin28--d-glucopyranoside-(1 → 6)--d-glucopyranoside, Fig. 1) is a main bioactive triterpenoid saponin isolated from the rhizome of Dipsacus asper Wall [1]. In preliminary articles, ASD has been reported to have a protective effect against cardiomyocyte apoptosis [2], neurotoxicity [3], acute myocardial ischemia injury [4], and osteoporosis [5]. In our previous study, ASD had a therapeutic effect on Alzheimer’s disease, but its bioavailability was very low after oral administration in rats [6], which may be due to the
∗ Corresponding author at: State Key Laboratory of Natural Medicines, China Pharmaceutical University, No. 24 Tongjia Lane, Nanjing 210009, PR China. Tel.: +86 25 83271426; fax: +86 25 83271426. ∗∗ Corresponding author. Tel.: +86 25 86 185129; fax: +86 25 86 185129. E-mail addresses: [email protected] (Z. Yang), [email protected] (C. Zhang). http://dx.doi.org/10.1016/j.jchromb.2014.09.013 1570-0232/© 2014 Elsevier B.V. All rights reserved.
metabolism of ASD in intestinal. Therefore, it is necessary to study its metabolic profile. Hydrolyzed reactions were found to be the major metabolic processes of ASD. In rats’ blood, ASD hydrolyzed to be hederagenin [7], which could be absorbed easily by intestinal mucosa. Moreover, ASD’s five hydrolyzed metabolites (M1, M2, M3, M4 and M5, Fig. 1) were identified by HPLC–MSn in rats’ feces [8]. And also there were three metabolites (M2, M4 and M5) of ASD simultaneously determined using LC–MS/MS in rat’s bile [9]. Although some researches on the metabolism of ASD in rats have been carried out, there are few reports about the metabolism of ASD in human. In this paper, we studied on the quantification of ASD and its five metabolites produced by human intestinal bacteria, verified the quantification method and calculated the conversion rate of five metabolites. The intestinal tract contains more than 400 bacterial species, in which Lactobacillus, Bacteroides and Streptococcus are major species, most of them to be strict anaerobes [10,11]. A lot of deconjugating enzymes in intestinal bacteria, e.g. -d-glucosidases, -d-glucuronidases and ␣-l-rhamnosidases [12,13], which release
82
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
determined to be higher than 96% by normalization of the peak area detected by HPLC. The HPLC-grade acetonitrile was purchased from TEDIA Company Inc. (Fairfield, OH, USA); formic acid was obtained from Merck KGaA (Darmstadt, Germany); ultrapure water was purified by an EPED super purification system (Nanjing, China). Other reagents were of analytical grade.
2.2. Preparation of the general anaerobic medium broth The general anaerobic medium (GAM) used for all fermentation experiments consists of tryptone 10.0 g, soya peptone 3.0 g, proteose peptone 10.0 g, digestibility serum powder 13.5 g, yeast extract 5.0 g, beef extract 2.2 g, beef liver extract powder 1.2 g, glucose 3.0 g, KH2 PO4 2.5 g, NaCl 3.0 g, soluble starch 5.0 g, l-cysteine hydrochloride 0.3 g, sodium thioglycolate 0.3 g, and distilled water 1000 mL, and then the pH was adjusted to 7.3 before autoclaving at 0.07 MPa at 121 ◦ C for 30 min.
2.3. Preparation of human intestinal bacterial mixture Fresh human fecal sample which was weighed 5 g and suspended in the GAM broth under a CO2 atmosphere and cultured in the anaerobic incubator at 37 ◦ C for 24 h was obtained from a healthy female Chinese volunteer. The obtained bacterial mixture was used as human intestinal bacterial mixture.
2.4. Incubation experiments
Fig. 1. (A) The chemical structures of ASD and its five metabolites; (B) glycyrrhetinic acid (IS).
glycones and secondary saponins from their glycosides. It is reported that there were 25 metabolites of American ginseng extract detected in human intestinal microflora [14]. The same thing happens to isoquercitrin, platycoside and soyasaponin Ab [15–17]. Due to the difficulty of getting reference standards, most studies only did qualitative analysis of metabolites. However, we established ultra-performance liquid chromatography triple quadrupole mass spectrometry (UPLC-TQ/MS) method for simultaneous quantification of ASD and its five metabolites for the first time, and calculated the conversion rate of five metabolites after ASD incubated 24 h. It described the parts of metabolic pathways for the biotransformation of ASD in human intestinal bacteria.
The standard solution of ASD was prepared by dissolving accurately weighed ASD in purified water to obtain a final concentration of 1 mg/mL. The 60 bacterial mixtures were incubated into 1 mL of GAM broth containing 0.1 mg/mL ASD in plastic tube, respectively, and the media were anaerobically incubated at 37 ◦ C. Five samples were taken out and frozen at −20 ◦ C in every 2 h. The bacterial mixture without ASD was incubated 24 h as a blank sample.
2.5. Preparation of standard and quality control (QC) samples The appropriate amounts of ASD, M1, M2, M3, M4 and M5 were separately weighed and dissolved in methanol as the stock solutions. Then, six stock solutions were mixed and diluted with methanol to give a final mixed standard solution containing 0.1 mg/mL of ASD and its five metabolites standard, and kept at 4 ◦ C when not in use. A series of working solutions of these analytes were diluting the mixed standard solutions with methanol at concentrations ranging from 100 to 100 000 ng/mL for ASD, 2 to 1000 ng/mL for five metabolites, respectively. For the validation of method, low, medium and high QC samples with concentrations of 100, 1000, 100 000 ng/mL for ASD, and 10, 100, 1000 ng/mL for five metabolites standard, were similarly prepared.
2. Experimental 2.6. Sample preparation 2.1. Materials Akebia saponin D (93.5% purity) and glycyrrhetinic acid (>98.0% purity) were purchased from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). Dipsacus saponin A, HN-saponin F, Akebia saponin PA, hederagenin-28-O--d-glucopyranoside, hederagenin were produced in our laboratory. Structures were confirmed by spectroscopic methods (MS, 1 H NMR and 13 C NMR), and their purity was
An aliquot of 1 mL of each bacterial sample was mixed with 3 mL methanol and 50 L I.S. solution (2 g/mL), vortex-mixed approximate for 1 min, and centrifuged at 17,800 × g for 10 min. The supernatant was transferred into another tube and evaporated to dryness at 50 ◦ C in a centrifugal vacuum evaporator (LABCONCO Corp., Kansas City, Missouri, USA). The residues were dissolved in 1 mL of 40% methanol, centrifuged at 17,800 × g for 10 min, and supernatant was analyzed by UPLC–MS.
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
2.7. Chromatographic conditions Chromatography was performed on a Waters Acquity UPLC system (Waters Corp., Milford, MA, USA). ACQUITY UPLCTM BEH C18 column (100 mm × 2.1 mm, 1.7 m) was used for separation. Column temperature was set at 35 ◦ C. Chromatographic separation was achieved by gradient elution at the flow rate of 0.4 mL/min. The mobile phase consisted of solvents A (acetonitrile) and B (0.1% aqueous formic acid). The gradient was as follows: 30% A at 0–4 min, 30–80% A at 4–7 min, 80–90% A at 7–7.5 min, 90% A at 7.5–8.5 min, and then an immediate reduction to 30% A at 9.5 min; 9.5–10.5 min, 30% A for equilibration of the column. Injection volume was 5 L.
83
2.9.5. Stability The stability of analytes in intestinal bacteria was evaluated by analyzing of three levels of QC samples stored at −80 ◦ C for 21 days (long term stability), at 25 ◦ C for 24 h (short term stability) and after three freeze/thaw cycles. The autosampler stability was analyzing QC samples at 4 ◦ C for 24 h. 2.10. Conversion rate of five metabolites After incubated 24 h, concentration–time curves of ASD and its five metabolites were obtained. The conversion rate of five metabolites was calculated using following equation: CM MASD × × 100% CASD MM
2.8. Mass spectrometric conditions
Conversion rate (%) =
A Waters ACQUITYTM Synapt mass spectrometer (Waters Corp., Milford, MA, USA) was connected to the UPLC system via an electrospray ionization (ESI) interface. The ion source was operated in negative ionization mode for the six compounds and IS. Quantification was performed using selected ion recording (SIR) mode of the transitions of m/z 928 → m/z 603 for ASD, m/z 795 → m/z 471 for M1, m/z 765 → m/z 603 for M2, m/z 633 → m/z 471 for M3, m/z 603 for M4, m/z 471 for M5 and m/z 469 for the internal standard. The MS parameters were as follows: source temperature 120 ◦ C, desolvation temperature 350 ◦ C, capillary voltage 3 kV, cone voltage 30 V, desolvation gas 600 L h−1 , cone gas 50 L h−1 . All data were acquired using the Waters Masslynx 4.1 software (Waters Corp., Milford, MA, USA).
where “CM ” is increased concentration of metabolites at 24 h, “CASD ” is decreased concentration of ASD in 24 h, “MASD ” is molecule weight of ASD, “MM ” is molecule weight of metabolites.
2.9. Method validation The proposed quantitative method was validated for selectivity, linearity, precision, accuracy, extraction recovery, matrix effect and stability according to the guidance of the Food and Drug Administration (FDA) for validation of bioanalytical methods [18]. 2.9.1. The selectivity The selectivity of the method was accessed by comparing the chromatograms of blank bacterial samples to those of corresponding standard samples spiked with analytes and IS and bacterial samples with ASD after incubated 18 h. 2.9.2. Linearity and lower limit of quantification Calibration curves were constructed by plotting the peak area ratios of analytes to IS v/s the nominal concentrations of standard analytes, respectively, using the weighting factor of 1/C2 . The limit of detection and quantification was determined at a signal-to-noise of 3 and 10 by analyzing the standard bacterial sample. 2.9.3. Precision and accuracy The intra- and inter-day precision and accuracy were determined by analyzing six replicates of QC samples at low, medium, high concentration levels. The relative standard deviation (RSD) was used for reporting precision. The accuracy was calculated from (measured − nominal)/nominal × 100%. The intra- and interday precision should not exceed 15% and accuracy should be within ±15% for the QC samples. 2.9.4. Extraction recovery and matrix effect The extraction recoveries of analytes were determined at three different concentrations of QC samples by comparing area ratios of analytes in the post-extraction spiked samples to that acquired from pre-extraction spiked samples. The matrix effect was evaluated by determining the peak area ratios of the analytes in post-extracted spiked samples to those acquired from unextracted spiked samples.
3. Results 3.1. Optimization of mass spectrometry In this experiment, MS/MS operation parameters were carefully optimized for determination of ASD and its five metabolites. The mass spectrometer was tuned in both positive and negative ionization mode with ESI source for optimum response of each analyte. ASD and its five metabolites in negative ion mode had better response, so we used the negative ion as the detection mode. In negative ion mode, ASD, M1, M2, M3, M4, M5 and I.S. formed the deprotonated molecular ions of m/z 928, m/z 795, m/z 765, m/z 633, m/z 603, m/z 471 and m/z 469, respectively. The detection of target analytes was firstly done in multiple reaction monitoring (MRM) mode, but we could not obtain good ion response either high energy or low energy. When we changed to use selected ion recording (SIR) mode, we obtained the good ion response of ASD and its metabolites and I.S. Quantification was performed using SIR mode of the transitions of m/z 928 → m/z 603 for ASD, m/z 795 → m/z 471 for M1, m/z 765 → m/z 603 for M2, m/z 633 → m/z 471 for M3, m/z 603 for M4, m/z 471 for M5 and m/z 469 for the I.S. Parameters such as source temperature, desolvation temperature, capillary and cone voltage, flow rate of desolvation gas and cone gas were optimized to obtain optimum response of analytes. 3.2. Optimization of LC To optimize chromatographic conditions, a suitable analytical column was at first selected to obtain high sensitivity and good separation. An ACQUITY UPLCTM BEH C18 column was the most universal column for UPLC separations. It improved the retentions of analytes and achieved the good separation. The composition of the mobile phase had an effect on the separation and ionization of analytes. Acetonitrile and methanol were both attempted as the organic modifier of the mobile phase. It was found that signalto-noise of analytes was higher with acetonitrile than those with methanol. The addition of 0.1% formic acid in aqueous portion of mobile phase produced more symmetrical peaks for M4, M5 and I.S. because of the carboxyl group of their structures. The flow rate of mobile phase, column temperature and injection volume on separation behavior were also investigated. The flow rate of 0.3 mL/min was too slow that prolonged retention time, and we obtained the good separation at the flow rate of 0.4 mL/min; There were no significant effects on separation between the column temperature 30 ◦ C and 35 ◦ C, and we selected the column temperature 35 ◦ C; The injection volume was dependent on the concentrations of analytes
84
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
and their response in the instrument, and we selected moderate injection volume 5 L in this experiment. In order to optimize peak shape, retention time and separation, several trails to screen the ratio of acetonitrile to water were carried out. Finally, a gradient elution system (acetonitrile–0.1% aqueous formic acid) as previously described was chosen. Under this chromatographic condition, the effective separation was achieved for ASD and its metabolites and IS. 3.3. Selection of internal standard According to FDA guidance, an internal standard in the analysis of biological sample could be a structurally similar analog of analyte or a stable labeled compound [18]. Glycyrrhetinic acid was adopted as I.S. due to the similarity of structure and extraction efficiency with the analytes, and its efficient ionization in the negative ionization mode. 3.4. Sample preparation The most widely employed biological sample preparation techniques are protein precipitation (PPT), liquid–liquid extraction (LLE) and solid-phase extraction (SPE). The recovery and matrix effect of SPE are sufficient but it requires much time and money. The different polar character of ASD and its five metabolites need different extraction solvent, which makes sample preparation very complicated. Thus, the PPT method was used to extract the analytes from the intestinal bacteria. Methanol and acetonitrile were tested to use as the precipitation solvents. The test results showed that there were little difference between the recoveries of ASD, M1 and M2, but methanol gave much higher recoveries of M3, M4 and M5 than acetonitrile. Thus, methanol was finally chosen as the protein precipitation reagent in this experiment. 3.5. Method validation 3.5.1. The selectivity Fig. 2 shows representative extract ions SIR chromatograms of ASD, five metabolites of ASD and IS. There is no significant interference in the determination of the analytes and IS. 3.5.2. Linearity and lower limit of quantification The regression equations for calibration curves and LLOQs of ASD and five metabolites are listed in Table 1. The results demonstrated that this method is sensitive enough to quantitative detection of ASD and its metabolites. 3.5.3. Precision and accuracy In this assay, the intra- and inter-day precision and accuracy were summarized in Table 2. At each QC level, the intra- and interday precisions (RSD) of those analytes were within 10.89%. The accuracy of those analytes was within 12.47%. The results show that this method has a satisfactory precision and accuracy. 3.5.4. Extraction recovery and matrix effect The extraction recoveries and matrix effects for ASD and its five metabolites are shown in Table 3. The extraction recoveries of ASD and five metabolites were in the range from 73.3% to 95.4%, and the extraction recovery of IS was 90.7%. The matrix effects of ASD and five metabolites were in the range from 84.53% to 101.13%, and the matrix effects of IS was 92.1%. The results showed that there is no significant matrix effect for analytes and IS. 3.5.5. Stability The stability of ASD and its five metabolites in intestinal bacteria are summarized in Table 4, which demonstrated the good stability
Fig. 2. Representative extract ion SIR chromatograms of ASD, M1, M2, M3, M4, M5 and glycyrrhetinic acid (IS): (A) blank intestinal bacteria; (B) blank intestinal bacteria spiked with the six analytes and IS; (C) ASD incubated in intestinal bacteria 18 h.
of ASD and its five metabolites at −80 ◦ C for 21 days, at 25 ◦ C for 24 h, and after three freeze/thaw cycles, as well as in post-prepared samples at 4 ◦ C for 24 h. 3.6. Conversion rate of five metabolites After incubated 24 h, the concentration–time curves of ASD and its five metabolites are shown as Fig. 3. The concentration of ASD was decreased from 99.394 g/mL to 96.850 g/mL, while the
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
85
Table 1 The regression equations, linear ranges, LOQs, and LODs of ASD and its five metabolites (n = 3). Analytes
Regression equation
R2
Linear range (ng/mL)
LOD (ng/mL)
LOQ (ng/mL)
ASD M1 M2 M3 M4 M5
y = 0.00106x + 0.00409 y = 0.00277x + 0.01160 y = 0.0018x +0.00416 y = 0.00245x + 0.00549 y = 0.00344x + 0.00801 y = 0.00394x + 0.12540
0.9998 0.9994 0.9999 0.9998 0.9987 0.9983
100–100,000 2–1000 2–1000 2–1000 2–1000 2–1000
0.41 0.59 0.61 0.55 0.52 0.31
1.32 1.91 1.95 2.00 1.87 1.05
Table 2 The inter-day and intra-day accuracies and precisions for ASD and five metabolites (n = 5). Analytes
Concentration (ng/Ml)
ASD
100 1000 100000 10 100 1000 10 100 1000 10 100 1000 10 100 1000 10 100 1000
M1
M2
M3
M4
M5
Inter-day (n = 6)
Intra-day (n = 6)
Accuracy (%)
Precision (%)
Accuracy (%)
Precision (%)
9.38 −4.91 0.51 −7.07 5.63 −0.05 −2.92 1.86 −0.14 −6.06 −0.6 1.15 9.56 −4.87 0.1 5.54 9.46 −0.05
4.7 10.89 1.57 10.12 8.86 8.32 11.1 7.65 5.14 9.74 6.06 6.09 7.85 8.1 4.87 10.23 6.23 7.51
5.42 −1.34 1.21 −8.33 4.85 0.14 −2.52 0.85 0.47 −4.87 −2.36 0.26 7.93 −6.58 −0.24 5.74 12.47 0.13
7.07 10 3.25 6.45 6.69 7.8 9.6 8.74 6.88 9.82 7.21 5.94 8.95 8.05 5.32 6.99 9.09 7.83
concentration of M1, M2, M3, M4 and M5 were 94.519, 755.066, 15.905, 183.954 and 5.424 ng/mL. The molecular weight of ASD and five metabolites are 929, 796, 766, 634, 604 and 472, respectively. Therefore, the conversion rate of these metabolites are 3.72%, 29.68%, 0.63%, 7.23% and 0.21%, respectively, total of 41.21%. 3.7. Metabolic pathways The parts of metabolic pathways of ASD in human intestinal bacteria are shown in Fig. 4. It is the same as metabolic pathways of ASD in rats’ feces [8].
4. Discussion The results show that the concentration of ASD is decreased with the reaction time prolonged. As to five metabolites, M1, M2 and M4 can be detected at 2 h, but M3 and M5 are found at 10 h and 6 h. It demonstrated that ASD’s desugar metabolites were produced sequentially. Within 24 h, the concentration of M2 and M4 increased with the reaction time prolonged, while M1, M3 and M5 increased at first but then decreased. It suggested that the production rate of the former is faster than their metabolism speed; the latter would decrease due to the faster metabolism speed.
Table 3 Extraction recoveries and matrix effects of ASD and five metabolites (n = 5). Analytes
ASD
M1
M2
M3
M4
M5
Concentration (ng/mL)
100 1000 100000 10 100 1000 10 100 1000 10 100 1000 10 100 1000 10 100 1000
Recovery (%)
Matrix effect (%)
Average (%)
RSD (%)
Average (%)
RSD (%)
82.76 79.75 94.37 81.6 79.3 86.6 86.8 83.4 90.1 75.2 87.6 85.4 74.1 78.5 93.2 73.3 89.6 95.4
9.8 7.6 5.7 4.9 9.3 6.9 7.3 6.5 7.7 8.9 4.3 5.8 7.8 8.4 5.6 10.1 3.4 6.1
87.83 90.65 96.98 91.6 95.3 92.5 93.93 91.5 94.83 90.93 101.13 97.4 84.5 90.8 97.2 88.2 91.9 98.3
7.8 8.2 4.8 7.6 8.8 6.8 7.2 5.7 7.1 5.5 6.4 3.4 8.9 7.1 6.3 9.5 2.8 4.2
86
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
Table 4 Stability of ASD and five metabolites under different condition; percentage difference from fresh controls. Analytes/storage condition
Conc. (ng/mL)
Three freeze–thaw cycles
Long-term stability −20 ◦ C for 21 d
Short-term stability 25 ◦ C for 24 h
Post-prepared sample 4 ◦ C for 24 h
ASD
100 1000 100000 10 100 1000 10 100 1000 10 100 1000 10 100 1000 10 100 1000
−3.55 9.8 4.09 −8.62 4.12 −0.65 −3.62 4.07 2.14 −1.78 5.18 2.55 −0.06 2.72 −1.98 −2.63 6.35 2.91
−6.55 14.8 8.28 −7.46 2.21 −1.74 −5.13 2.55 −1.69 −2.67 3.69 −3 −4.21 −3.04 −1.84 −1.66 3.84 −4.14
−2.32 4.8 9.09 −6.2 4.18 −0.77 −1.61 1.56 −0.82 −3.06 6.34 3.1 −7.46 −2.91 1.16 8.78 8.71 −5
−5.55 10.5 7.23 −7.26 −2.66 −0.79 −2.92 5.45 0.53 1.58 −1.07 −2.6 −4.97 5.03 −0.57 7.9 −3.47 −1.22
M1
M2
M3
M4
M5
Fig. 3. Concentration–time curves of ASD and its five metabolites in 24 h.
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
87
Fig. 4. Parts of metabolic pathways of ASD in human intestinal bacteria.
The total conversion rate of five metabolites is 41.21%. It suggests that other metabolites were produced in this reaction. It is reported that hydrolysis, hydroxylation, acetylation were the major metabolic pathways of flavonoids in A. manihot extract in intestinal bacteria [19]. Besides, acetylated, hydroxymethylated, dehydroxylated, hydroxylated metabolites could be found after isoquercitrin incubated in human intestinal bacteria [13]. In this process of desugar reaction, ASD can combine with hydroxyl, hydroxymethyl and acetyl group. At the same time, it can occur dehydroxylation, demethylation, dehydroxymethyl reactions. Moreover, five metabolites may occur secondary metabolism. In view of the above points, we might explain the reasons for the low bioavailability of ASD. Not only did ASD exist in blood as prototype, but also a large number of its metabolites may exist. When we do research on the bioavailability of ASD, its metabolites must be considered. Meanwhile, our research suggested that low bioavailability of saponin compounds may be caused by drug metabolism. Such as ginsenosides Rb1, Rb2, Rb3, DT-13 and astragaloside IV [20–22]. 5. Conclusion A selective and sensitive UPLC-TQ/MS method was developed for simultaneous quantification of ASD and its five metabolites with LLOQs of 1.32, 1.91, 1.95, 2.00, 1.87 and 1.05 ng/mL for ASD, M1, M2, M3, M4 and M5, respectively. The total conversion rate of five metabolites is 41.21%. The parts of metabolic pathways for the biotransformation of ASD in human intestinal bacteria were
described. The method was validated and successfully applied to quantification of ASD and its five metabolites in human intestinal bacteria. Authors’ contribution L. Yan and X. Yang contributed equally to this work and should be regarded as cofirst authors. Acknowledgements This work was supported by A Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions. The assistance of the Jiangsu Key Laboratory for High Technology Research of TCM Formulae is gratefully acknowledged. References [1] K.Y. Jung, K.H. Son, J.C. Do, Arch. Pharmacol. Res. 16 (1993) 32–35. [2] C.M. Li, J.W. Tian, G.S. Li, W.L. Jiang, Y.L. Xing, J. Hou, H.B. Zhu, H. Xu, G.B. Zhang, Z.F. Liu, Z.G. Ye, Eur. J. Pharmacol. 649 (2010) 100–107. [3] H.W. Suh, D.K. Song, S.O. Huh, K.H. Son, Y.H. Kim, J. Ethnopharmacol. 71 (2000) 211–218. [4] C.M. Li, Z.F. Liu, J.W. Tian, G.S. Li, W.L. Jiang, Y.L. Xing, J. Hou, H.B. Zhu, H. Xu, G.B. Zhang, Z.F. Liu, Z.G. Ye, Eur. J. Pharmacol. 627 (2010) 235–241. [5] Y.B. Niu, Y.H. Li, H.T. Huang, X.H. Kong, R. Zhang, L. Liu, Y. Sun, T.M. Wang, Q.B. Mei, Phytother. Res. 25 (2011) 1700–1706. [6] K. Li, L. Ding, Z.L. Yang, E.H. Liu, L.W. Qi, P. Li, Y.Z. Hu, Biomed. Chromatogr. 24 (2010) 550–555. [7] H. Zhu, L. Ding, S. Shakya, X.M. Qi, L. Hu, X.L. Yang, Z.L. Yang, J. Chromatogr. B 879 (2011) 3407–3414.
88 [8] [9] [10] [11] [12] [13] [14]
L. Yan et al. / J. Chromatogr. B 971 (2014) 81–88
K. Li, X.L. Yang, C.F. Zhang, Z.L. Yang, Chin. J. Nat. Med. 7 (2009) 440–443. L. Wang, E.W. Liu, L.F. Han, Tianjin J. Tradit. Chin. Med. 30 (2013) 681–684. J.E. Lee, S. Lee, J. Sung, G. Kol, ISME J. 5 (2011) 362–365. J.C. Rambaud, Ann. Gastroenterol. Hepatol. (Paris) 28 (1992) 263–266. I.A. Macdonald, J.A. Mader, R.G. Bussard, Mutat. Res. Lett. 122 (1983) 95–102. V.D. Bokkenheuser, C.H. Shackleton, J. Winter, Biochem. J. 248 (1987) 953–956. J.Y. Wan, P. Liu, H.Y. Wang, L.W. Qi, C.Z. Wang, P. Li, C.S. Yuan, J. Chromatogr. A 1286 (2013) 83–92. [15] L.L. Lu, D.W. Qian, J. Yang, S. Jiang, J.M. Guo, E.X. Shang, J.A. Duan, Biomed. Chromatogr. 27 (2013) 509–514. [16] Y.W. Ha, Y.C. Na, I.J. Ha, D.H. Kim, Y.S. Kim, J. Pharm. Biomed. Anal. 51 (2010) 202–209.
[17] S.Y. Chang, M.J. Han, E.H. Joh, D.H. Kim, J. Pharm. Biomed. Anal. 52 (2010) 752–756. [18] USFDA, 2001. http://www.fda.gov/downloads/Drugs/Guidance/Compliance Regulatory Information/Guidances/UCM070107.pdf [19] C.F. Xue, S. Jiang, J.M. Guo, D.W. Qian, J.A. Duan, E.X. Shang, J. Chromatogr. B 879 (2011) 3901–3908. [20] J. Zhao, C. Su, C.P. Yang, M.H. Liu, L. Tang, W.W. Su, Z.Q. Liu, J. Pharm. Biomed. Anal. 64 (2012) 94–97. [21] L.L. Chen, W.W. Yuan, Z.F. Hu, J. Qi, D.N. Zhu, B.Y. Yu, J. Pharm. Biomed. Anal. 56 (2011) 650–654. [22] W.D. Zhang, C. Zhang, R.H. Liu, H.L. Li, J.T. Zhang, C. Mao, S. Moran, C.L. Chen, Life Sci. 79 (2006) 808–815.
