|Home | About | Journals | Submit | Contact Us | Français|
A rapid and simple ultra-performance liquid chromatography–electrospray ionization–mass spectrometry (UPLC–ESI–MS) method for the determination of astragaloside III was developed and used in a pharmacokinetic and tissue distribution study in rats following the oral administration 95% ethanol extraction of Zhenqi Fuzheng capsules. Although astragaloside III and astragaloside IV have the same molecular weight and very similar structures, they were successfully separated using this method. Quantification was performed using low-energy collision tandem mass spectrometry (CID–MS-MS) with the multiple reaction monitoring scan mode of the following precursor ion → product ion at m/z 807.61→335.22 for astragaloside III and at m/z 633.18→331.18 for the internal standard (hesperidin). Both astragaloside III and astragaloside IV in rat plasma were best fit to a two-compartment model. The tissue distribution study showed the overall trend of disposition of astragaloside III were Cthymus > Cspleen > Cstomach > Cliver > Cheart > Ckidney > Clung > Ctesticle. The high levels of astragaloside III in thymus and spleen indicated an accumulation in organs involved in immune responses and showed that these organs are major target sites in vivo. The results in the article will provide valuable information for use in clinical applications of astragaloside III.
The traditional Chinese medicine Zhenqi Fuzheng capsule (ZFC) is commonly used in clinical practice to improve immunity and as an adjunct to surgical operations to promote the recovery of normal functions (1–11). ZFC classical formulation is a (2 : 1, w/w) mixture of Radix Astragali (Huangqi in Chinese) and Fructus Ligustri Lucidi (Nvzhenzi in Chinese). ZFC has been used for thousands of years in China and has played an important role in complementary therapies. Complementary therapies are increasingly used in Western medicine, especially for patients with cancer. A recent European survey showed that herbal medicine use among patients with cancer is the most common complementary therapy in use across Europe (12).
The triterpene saponins are the primary bioactive components in Radix Astragali and ZFC (13, 14) and have recently been suggested for use in extending existing multidrug anti-cancer regimens (15, 16). These widely distributed compounds composed of one or more hydrophilic glycoside moieties bound to a lipophilic triterpene derivative were originally shown to possess anti-inflammatory, vasoprotective and expectorant properties (17). However, triterpene saponins were also found to interfere with basic processes crucial for cancer promotion and progression, such as cell proliferation, apoptosis, invasion and angiogenesis (18). The multifaceted effects of saponosides, along with their excellent safety and tolerability profile after chronic and prolonged exposure, warrant research for potential application in treatment of advanced prostate cancers (19). Astragaloside III is an important triterpene saponin, demonstrating the strongest anti-gastric ulcer activity out of 16 other compounds in Radix Astragali (20).
Four saponins including astragaloside I, II, III and IV were detected in Radix Astragali preparations (21–25), and the article built the UPLC–ESI–MS method can separate the four saponins. The experiment contents about the pharmacokinetics and tissue distribution studies of astragaloside II and IV have been published by Liu et al. (22, 25). This article focused on the pharmacokinetics and tissue distribution profiles of astragaloside III in rats. Astragaloside I has not been detected in most parts of rat tissues. However, there has been a published report to examine astragaloside III using liquid chromatography–mass spectrometry (LC–MS) or describe its pharmacokinetics (26), but no published report about its tissue distribution. Zhai et al. (26) only gave the monomer compound of astragaloside III to the rats, but there are many saponin mixtures including astragaloside I, II, III and IV in plants. The quantitation and pharmacokinetic profile of astragaloside III in biological fluids and tissues will not only evaluate potential clinical applications of ZFC but also will more fully elucidate its pharmacological mechanisms of action to explain events related to the efficacy and toxicity of ZFC. Tissue distribution studies are vital for understanding the major target sites and interpreting the in vivo disposition of astragaloside III. Considering the growing significance of a potential beneficial role of phytochemicals in human health, detailed in vivo disposition studies of phytochemicals used in traditional Chinese medicine are required (27). Thus, studies examining the pharmacokinetic profile of astragaloside III from complex samples of TCM are warranted. This is the first tissue distribution study of astragaloside III in rats. The aim of this study was to develop and validate a sensitive ultra-performance liquid chromatography–electrospray ionization–mass spectrometry (UPLC–ESI–MS) method for rapid quantitation of astragaloside III in rat plasma and tissues after oral administration of 95% ethanol extraction of ZFC.
All experiments were conducted with a Quattro Premier XE Micro mass spectrometer (Waters, Milford, MA, USA) equipped with an electrospray ionization (ESI) source and interfaced to an ACQuity UPLC system (Waters).
Chromatographic separations were performed using a Waters ACQuity CSHTM equipping with a Fluoro-Phenyl column (2.1 × 50 mm, 1.7 µm). The mobile phases consisted of 0.3% (v/v) formic acid in water (A) and acetonitrile (B) using a gradient elution of 20–40% (v/v) B at 0–3 min, 40–64% B at 3–6 min and 64–82% B at 6–8 min. An aliquot of 10 μL of each sample was injected, and the run time was 8 min.
The mass detection was set in a multiple reaction monitoring (MRM) mode and the ESI interface in the positive ionization modes was used. The ESI parameters were set as follows: capillary voltage, 3.0 kV; dwell, 0.050 s; extractor voltage, 3 V; RF lens voltage, 0.1 V. The ion source and desolvation temperatures were 120 and 300°C, respectively. The desolvation and cone gas flows were set at 500 L/h. The cone voltage of compounds was 90 V. The collision energy for fragmentation of the precursor ions was set at 50 EV. The data were acquired and analyzed using Mass Lynx version 4.1 data software.
ZFC (No. J20130301) was obtained from Gansu Fuzheng Pharmaceutical Sci & Tech Co., Ltd. (China). Astragaloside III was purchased from Shanghai Sunzo Biotech Co., Ltd. Hesperidin (No. 110721, internal standard, IS) and astragaloside IV (No. 110781) was purchased from the Chinese National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China). The purity of each compound was determined to be higher than 98% using the normalization of the peak area detected with high-performance liquid chromatography (HPLC). The chemical structures of the reference compounds are shown in Figure 1. HPLC-grade acetonitrile was purchased from Merck (Darmstadt, Germany). Distilled water was further purified using a Milli-Q system. All other chemicals, such as methanol, were of analytical grade.
Sprague-Dawley rats (220–250 g body weight) were obtained from the Laboratory Animal Center of Lanzhou University (Lanzhou, China) and housed with unlimited access to food and water, except for fasting for 12 h before the experiment. The animals were maintained on a 12-h light–dark cycle (lights on at 8 : 00 AM) at an ambient temperature of 22–25°C and a relative humidity of 60%. All animal experiments were conducted in accordance with the Guidelines for Animal Experimentation of Lanzhou University (Lanzhou, China), and the protocol was approved by the Animal Ethics Committee of that institution. The concentration of astragaloside III obtained from a 95% ethanol extraction of ZFC was 0.1410 mg/g (21).
The method was validated for selectivity, linearity, precision and accuracy, matrix effects and extraction recovery according to the FDA guidelines for a bio-analytical method (28).
Appropriate amounts of astragaloside III and IS were dissolved in methanol to prepare separate stock solutions of 0.1424 and 0.2950 mg/mL, respectively. The stock solution of astragaloside III was further diluted with methanol to a concentration of 7.12 μg/mL, and the IS solution was maintained at ~0.59 μg/mL for each working solution and sample.
Plasma calibration standards and quality control samples were prepared by spiking blank plasma with the appropriate amount of working standard solutions. Calibration standards were prepared at eight concentrations ranging from 5.6 to 7,120 ng/mL, and plasma quality control (QC) samples were prepared at three concentrations of 5.6, 560 and 7,120 ng/mL. Calibration standards for various tissues, including heart, liver, spleen, lung, kidney, thymus, testicle and stomach, were prepared by spiking 200 mg of blank tissue, and QC samples at three concentrations were prepared in the same fashion. Standard calibration and QC samples were stored at −80°C until used in further analyses.
For plasma samples, 20-μL aliquots of working IS solution (0.59 μg/mL) and 0.8 mL of methanol were added to 200 μL of the plasma sample. After vortex-mixing for 2 min and ultrasonic treatment for 10 min, the samples were centrifuged at 14,000 × g for 10 min. The supernatant was transferred to another tube and evaporated to dryness in a water bath at 45°C under a nitrogen stream. The residues were reconstituted in 200-μL aliquots of HPLC mobile phase and centrifuged at 14,000×g 4°C for 10 min. Supernatants (10 μL) were injected into the UPLC system.
For tissue samples, 20-μL aliquots of working IS solution and 5.0 mL of methanol were added to small slices of tissue and processed tissue homogenates (using an IKA T25 homogenizer). The tissue homogenates were centrifuged at 14,000 × g for 10 min and then treated in the same manner as that for the plasma samples. The supernatant (10 μL) was injected into the UPLC–MS-MS system for analysis.
The specificity was assessed by analyzing a blank spleen tissue homogenate sample spiked with IS, a blank spleen tissue homogenate sample spiked with astragaloside III and IS and an experimental spleen tissue homogenate sample spiked with IS from rats after administration of ZFC extract.
The regression equation of astragaloside III could be described as y = axb. The lower limits of quantification (LLOQ) was defined as the lowest drug concentration that could be detected with a relative error and precision (relative standard deviation, RSD) of no more than 20%.
Intra-day assay accuracy and precision were established by analyzing nine replicates of the QC samples at the three concentrations described. Inter-day assay accuracy and precision were established through the performance of three consecutive days. The accuracy was determined as the percentage of deviation (relative error, RE%) between the measured and nominal concentrations. The precision was evaluated from the RSD of the concentration measurements. Intra- and inter-run accuracies and precisions for QC concentrations of ≤10% were deemed acceptable.
The extraction recovery in rat matrices for astragaloside III was determined at the three QC concentrations and calculated as the ratio of analyte peak area from extracted QC samples to that from extracted blank matrices spiked with astragaloside III standard solution. The matrix effect referred to the literature (22).
The stability of the method was determined under different conditions using nine sets of low, middle and high concentrations. The short-term stability was tested after storing the samples at room temperature for 12 h or at 4°C for 24 h. The freeze–thaw stability was tested by freezing the samples at −80°C overnight and then thawing them at 20°C, for a total of three cycles. The long-term stability was tested after storing the samples at −80°C for 30 days.
Blood samples were serially withdrawn from each animal at 0.25, 0.5, 1, 1.5, 2, 2.5, 3, 4, 5, 6, 8, 10, 12 and 24 h after oral administration of 95% ethanol extraction of ZFC (1.022 mg of astragaloside III in 220 g rat). Each sample was centrifuged, and the separated plasma samples were frozen in polypropylene tubes at −80°C prior to analysis.
Various tissue samples, including liver, lung, spleen, kidney, heart, stomach, testicles and thymus, of a given weight (200 mg) were collected from rats at 0.5, 1, 1.5, 2, 2.5, 3 and 4 h after oral administration of 95% ethanol extraction of ZFC. The tissues were rinsed with normal saline solution to remove the blood or other contamination, blotted with paper towels and stored at −80°C until treatment.
Pharmacokinetic parameters, including the area under the plasma concentration–time curve (AUC), maximum plasma concentration (Cmax), time to reach the maximum concentration (Tmax), apparent plasma clearance after oral administration (CL/F) and mean residence time (MRT), were estimated with a non-compartmental analysis method using Drug and Statistics 2.0 (DAS 2.0) software (Mathematical Pharmacology Professional Committee of China, Lanzhou, China).
The chromatography conditions were optimized by varying the mobile phase modifier (0.1, 0.2 and 0.3% (v/v) for both formic acid and acetic acid), gradient elution type (isocratic elution and gradient elution) and flow rate (0.2, 0.3, 0.4 and 0.5 min/mL). The best mobile phases composition were 0.3% (v/v) formic acid in water (A) and acetonitrile (B) using a gradient elution of 20–40% (v/v) B at 0–3 min, 40–64% B at 3–6 min and 64–82% B at 6–8 min.
The mass detection was set in a MRM mode and the ESI interface in the positive and negative ionization modes. The positive ionization modes have good separation effect. The cone voltage of compounds was selected as follows: 50, 60, 70, 80, 90 and 95 V. The collision energy for fragmentation of the precursor ions was selected at 30, 40, 50, 60 and 70 eV. The cone voltage and collision energy were determined to be 90 V and 50 eV.
Typical chromatograms of blank plasma spiked with IS, blank matrices spiked with astragaloside I, astragaloside II, astragaloside III, astragaloside IV and IS and experimental samples after oral administration of ZFC extract are presented in Figure 2. The retention times were ~2.12 min for IS, 3.20 min for astragaloside IV, 3.30 min for astragaloside III, 3.64 min for astragaloside II and 4.34 min for astragaloside I. Thus, due to the high selectivity and the use of different channels to collect the chromatograms with the UPLC–MS method, the four analytes and IS showed good specificity.
Significant differences in recoveries of astragaloside III were found for the different tissues spiked at the same concentration as that for astragaloside III. Consequently, the quantitation process was affected by differences in tissue matrices, and the method calibration could be performed by spiking different tissue samples. The linear regression analysis of astragaloside III in the various tissues of the rat were obtained by plotting the peak area ratio of astragaloside III to IS (y-axis) versus the analyte concentration (in ng/mL; x-axis) in spiked tissue samples. The regression equation of the curves and the correlation coefficients (r2) for the plasma, heart, liver, spleen, lung, kidney, thymus, testicles and stomach are listed in Table I. All calibration curves for astragaloside III displayed good linearity over the selected concentration ranges (all correlation coefficients >0.99). The LLOQ, defined as the lowest concentration at which both precision and accuracy were less than or equal to 20%, was 5.6 ng/mL in plasma and in heart, liver, spleen, lung, kidney, thymus, testicle and stomach tissue samples.
Intra-day precision, inter-day precision and accuracy for astragaloside III are exhibited in Table II. All results for all samples tested ranged from 4.2 to 9.1%, which were within the acceptable criterion of ±15%, suggesting that the method for the determination of astragaloside III in biological matrices was accurate and reproducible.
The extraction recoveries of astragaloside III from different tissue samples were obtained from nine replicate analyses of the QC sample at high, medium and low concentrations. The results showed that the mean extraction recoveries ranged from 86.4 to 95.6% (Table II). The RSDs for the extraction recoveries ranged from 5.6 to 9.6%. Therefore, it was concluded that precipitating protein directly with the methanol procedure was simple, rapid and successful, with acceptable recoveries of astragaloside III and IS from different tissue samples at all tested concentrations.
The potential for a matrix effect, caused by ionization competition, occurring between the analyte and the endogenous co-eluents was examined at three QC concentrations. For astragaloside III, the matrix effects in plasma, heart, liver, spleen, lung, kidney, thymus, testicle and stomach ranged from 86.9 to 97.4% at high, medium and low concentrations and with RSDs lower than 10% (Table II). These results indicated that ion suppression or enhancement from the rat tissue matrices for astragaloside III was negligible in the established method.
The stability of the analytes in rat tissue homogenates and plasma under the different temperatures and timing conditions was evaluated. The results for astragaloside III shown in Table III indicate that it has an acceptable stability under the tested conditions.
Astragaloside III and astragaloside IV have very similar structures and the same molecular weight, differing only in the location of a single glucose, their separation using LC–MS is difficult; thus, there are no previously published reports. However, in this study, these compounds were examined with LC–MS using different channels to collect the chromatograms and display the two compounds separately. The results indicated that the spread of the peaks was reduced and the sensitivity was improved using the Fluoro-Phenyl column (2.1 × 50 mm, 1.7 µm), whereas the UPLC BEH C18 column (2.1 × 50 mm, 1.7 µm) could not separate astragaloside III and astragaloside IV. In addition, using the mobile phase described in the methods led to good separation of astragaloside I, astragaloside II, astragaloside III and astragaloside IV in rat tissues and plasma (Figure 2).
In this study, astragaloside III and IS were detected using ESI and MRM with positive scanning switch. Under the optimized chromatographic conditions, satisfactory retention time was obtained from astragaloside III, and astragaloside IV and IS. They were successfully separated within 8 min in the standard mixture as well as the samples, exhibiting high detection sensitivity and analysis efficiency. Figure 2 shows the total ions of MRM chromatograms to quantification of astragaloside III and IS.
Many fragment ions such as astragaloside III, and astragaloside IV and IS (hesperidin) were known from the reference (20). The cone voltages were chosen in order to let the precursor ion to get the strongest response. Then collision energy was chosen to make the fragment ion to the strongest response. To avoid other material interference, they need at least two product ions to qualitatively identify the compound. Figure 3 shows MRM chromatograms to qualitatively identify astragaloside III, and astragaloside IV and IS. The precursor ion of astragaloside III and astragaloside IV was [M+Na]+ at m/z 807.61 and 807.40, respectively. The molecular ion [M + H]+ of hesperidin was at m/z 633.18. The product ions of astragaloside III of qualitative identification were at m/z 807.61 → 335.22 and 807.61 → 495.50. The product ions of astragaloside IV were qualitatively identified at m/z 807.40 → 627.50 and 807.40 → 202.94. The product ions of IS were qualitatively identified at m/z 633.18 → 331.18 and 633.18 → 483.26. We chose at m/z 807.61 → 335.22 for astragaloside III and at m/z 633.18 → 331.18 for IS of quantification because they showed the stronger response than at m/z 807.61 → 495.50 and 633.18 → 483.26, respectively. Therefore, quantification was performed using low-energy collision tandem mass spectrometry (CID–MS-MS) using the MRM scan mode of the following precursor ion → product ion at m/z 807.61 → 335.22 for astragaloside III and at m/z 633.18 → 331.18 for IS.
The concentration of astragaloside III in plasma samples was determined. The mean plasma concentration versus time curve is presented in Figure 4, and the major pharmacokinetic parameters calculated using a non-compartmental model is listed in Table IV. As shown in Figure 1, the pharmacokinetics of astragaloside III was best fit to the two-compartment model. Astragaloside IV was a two-compartment model from publishing reports (23–25). The two compounds have the same models, because they can have very similar structures, differing only in the location of a single glucose. Astragaloside III was absorbed and eliminated with tmax at 1.0 h and t1/2 at 1.085 h, which are reasonable pharmacokinetic parameters.
As shown in Figure 5, astragaloside III was widely distributed in all tissues examined after oral administration of ZFC extract. However, astragaloside III was undetectable 4 h after oral administration of ZFC extract. The highest astragaloside III levels were observed in the thymus at 2.0 h, with the next highest levels in the spleen at 1.5 h, followed by the stomach, liver, heart, kidney, lung and testicles after oral administration of ZFC extract. These high levels in the thymus and spleen indicated that astragaloside III accumulated in organs involved in the immune response, which is consistent with the immunity effects of ZFC extract. The high levels in the stomach indicated that ZFC extract was orally administered. The results which show the CLiver > CHeart > CKidney indicated that astragaloside III was distributed mainly in tissues with an abundant blood supply, suggesting that distribution depended on the blood flow and perfusion rate of the organ. The fourth highest concentration was found in the liver, suggesting that astragaloside III may be susceptible to hepatic metabolism.
The tmax concentration of astragaloside III in plasma was reached 1 h after oral administration of ZFC extract, with the liver, heart and kidney achieving their highest concentrations at the same time. The concentration of astragaloside III peaked in the stomach, spleen and testicles at 1.5 h, and in the thymus and lung at 2 h.
A simple and rapid UPLC–ESI–MS method for detection of astragaloside III in plasma and tissue was developed and validated. Because astragaloside III has a structure very similar to that of astragaloside IV, differing only in the location of a single glucose, their separation by LC–MS is difficult. Thus, this study was the first to examine astragaloside III using UPLC–ESI–MS. A Fluoro-Phenyl column was used, and different channels were set to separately collect and display the chromatograms of the two compounds.
This LC–MS method successfully determined the pharmacokinetics and tissue distribution of astragaloside III in rats after the oral administration of ZFC extract. Both astragaloside III and astragaloside IV in rat plasma were best fit to a two-compartment model. The results showed that in general the tissue concentrations for astragaloside III were in the following order: thymus > spleen > stomach > liver> heart > kidney> lung > testicle. The high levels of astragaloside III in thymus and spleen indicated an accumulation in organs involved in immune responses and showed that these organs are major in vivo target sites. The results of the pharmacokinetic and tissue distribution studies obtained here in provided valuable information for use in clinical applications of astragaloside III.
The authors greatly appreciate the financial support from the Great Research Projects of the Ministry of Education (lzujbky-2015-59).
The authors thank Professor Wei Zhou for providing the UPLC–ESI–MS facilities at the Gansu Entry-Exit Inspection and Quarantine Bureau.