Article pubs.acs.org/JAFC
Pharmacokinetic Profile of Eight Phenolic Compounds and Their Conjugated Metabolites after Oral Administration of Rhus verniciflua Extracts in Rats Ming Ji Jin,† In Sook Kim,† Jong Suk Park,† Mi-Sook Dong,§ Chun-Soo Na,# and Hye Hyun Yoo*,† †
Institute of Pharmaceutical Science and Technology and College of Pharmacy, Hanyang University, Ansan, Gyeonggi-do 426-791, Korea § School of Life Sciences and Biotechnology, Korea University, Seoul 136-713, Korea # Lifetree Biotech Company, Ltd., Suwon, Gyeonggi-do 441-350, Korea ABSTRACT: Rhus verniciflua (Toxicodendron vernicifluum) is a medicinal tree popularly used in Asian countries such as China, Japan, and Korea as a food additive or herbal medicine because of its beneficial effects. R. vernicif lua extract (RVE) contains diverse phenolic compounds, such as flavonoids, as its major biological active constituents. In this study, the pharmacokinetic profiles of eight phenolic compounds were investigated following oral administration of RVE to rats. The eight phenolic compounds were 2,4-dihydroxybenzoic acid, 3,4-dihydroxybenzoic acid, fisetin, fustin, butin, sulfuretin, taxifolin, and garbanzol. The plasma concentrations of the eight compounds were determined by using a liquid chromatography−triple-quadrupole mass spectrometer before and after treatment with β-glucuronidase. When 1.5 g/kg RVE was administered, the eight compounds were all detected in plasma, mainly as conjugated forms. These pharmacokinetic data would be useful for understanding the pharmacological effects of RVE. KEYWORDS: Rhus verniciflua Stokes, phenolic compounds, bioanalytical method, pharmacokinetics
■
anti-inflammatory and antinociceptive effects,25 and taxifolin inhibited oxidative stress and cell apoptosis to prevent diabetic cardiomyopathy in vivo and in vitro.26 However, to our knowledge, pharmacokinetic studies of the bioactive phenolic compounds in RVE have not been performed, although pharmacokinetics is crucial for providing evidence of biological activity in vivo and for understanding the action mechanisms of functional foods. As for analytical methods for RVE, only a few published methods are available, which are high-performance liquid chromatography (HPLC)based methods.15 Furthermore, bioanalytical methods for the quantitation of phenolic compounds of RVE in biofluids or tissues have not yet been reported. In this study, a precise, accurate, and sensitive method based on liquid chromatography−tandem mass spectrometry (LCMS/MS) for the simultaneous quantitation of eight phenolic compounds of RVE (Figure 1) in rat plasma was developed. Pharmacokinetic profiling of the phenolic compounds was investigated following oral administration of RVE to rats using the developed method. Subsequently, we presented the pharmacokinetic properties of phenolic compounds of RVE as conjugated metabolites as well as parent forms.
INTRODUCTION Rhus verniciflua (Toxicodendron vernicifluum) is a medicinal tree cultivated in Asian countries, including China, Japan, and Korea. It has been traditionally used as a health food, herbal remedy, or food additive.1,2 R. vernicif lua contains medicinal ingredients with diaphoretic, cathartic, antiviral, sedative, and antirheumatic activities.3 There are many reports on the pharmacological activities of R. vernicif lua extracts (RVE). These include antioxidant activity,2,4−9 antiproliferative activity with growth of human lymphoma cells,1 anti-inflammatory activity,10 antimutagenic activity,11 apoptotic effects,12,13 and antimicrobial activity.14 These pharmacological activities of RVE are believed to be attributed to its bioactive ingredients, such as phenolic compounds. In previous chemical studies,15 diverse phenolic compounds, such as caffeic acid, 2,4-dihydroxybenzoic acid (2,4-DHBA), p-coumaric acid, chlorogenic acid, protocatechuic acid, garbanzol, fisetin, kaempferol, kaempferol-3-O-glucoside, quercetin, butein, butin, gallic acid, fustin, 2,6,3′,4′-tetrahydroxy-2-benzylcoumaran-3-one, and sulfuretin, have been reported as constituents of RVE. These phenolic compounds have been shown to exhibit various beneficial effects.8,9,15,16 For example, fisetin exhibited anti-inflammatory activity.17 Fustin exerted neuroprotection on 6-hydroxydopamine-induced cell death.18 Butin has a protective effect against hydrogen peroxide-induced apoptosis via its antioxidant properties,19 2,4-DHBA showed an antioxidative activity for free radical scavenging,20 and 3,4-DHBA exhibited antioxidant, antiinflammatory, antihyperglycemic, and anti-apoptotic activities.21 Sulfuretin exhibited anti-inflammatory activity and therapeutic value in preventing β-cell damage.22−24 Garbanzol showed © 2015 American Chemical Society
■
MATERIALS AND METHODS
Chemicals. Sulfuretin and fustin were purchased from Extrasynthese (Genay Cedex, France). Chrysin, fisetin, butin, garbanzol, Received: Revised: Accepted: Published: 5410
April 6, 2015 May 22, 2015 May 22, 2015 May 22, 2015 DOI: 10.1021/acs.jafc.5b01724 J. Agric. Food Chem. 2015, 63, 5410−5416
Article
Journal of Agricultural and Food Chemistry
Figure 1. Chemical structures of the eight phenolic compounds of Rhus vernicif lua analyzed in this study.
Figure 2. Multiple reaction monitoring chromatograms of (A) blank rat plasma, (B) plasma spiked with standards (100 ng/mL), and (C) plasma taken 1 h after oral administration of Rhus vernicif lua extracts. taxifolin, 2,4-DHBA, 3,4-DHBA, ascorbic acid, and formic acid were purchased from Sigma Chemicals (St. Louis, MO, USA). β-Glucuronidase (140 U/mL) was purchased from Roche Applied Science (Mannheim, Germany). HPLC grade water, methanol,
and acetonitrile were purchased from J. T. Baker (Philipsburg, NJ, USA). Preparation of RVE. The heartwoods of R. vernicif lua were cut into small pieces and extracted with a 10-fold volume of water for 3 h 5411
DOI: 10.1021/acs.jafc.5b01724 J. Agric. Food Chem. 2015, 63, 5410−5416
Article
Journal of Agricultural and Food Chemistry Table 1. Intra- and Interday Assay Results for Accuracy and Precision of the Developed Method intraday (n = 5)
interday (n = 3)
compound
theor concn (ng/mL)
concn found (mean ± SD)
accuracy (%)
CV (%)
concn found (mean ± SD)
accuracy (%)
CV (%)
2,4-DHBA
2 100 1000
1.8 ± 0.1 101.7 ± 4.4 979.3 ± 37.6
91.4 101.7 97.9
4.4 4.3 3.8
2.0 ± 0.3 103.7 ± 3.8 969.0 ± 47.7
100.7 103.7 96.9
15.0 3.7 4.9
3,4-DHBA
2 100 1000
2.2 ± 0.1 98.6 ± 7.9 1014.5 ± 53.5
108.4 98.6 101.4
6.5 8.0 5.3
2.0 ± 0.2 96.8 ± 11.0 952.1 ± 64.1
99.6 96.8 95.2
9.2 11.3 6.7
butin
2 100 1000
1.9 ± 0.1 101.5 ± 5.0 874.6 ± 92.4
96.2 101.5 87.5
2.8 4.9 10.6
1.8 ± 0.2 102.4 ± 6.6 850.7 ± 101.5
89.0 102.4 85.0
11.1 6.4 11.9
fisetin
2 100 1000
1.9 ± 0.2 98.7 ± 3.8 967.5 ± 117.4
96.4 98.7 96.7
7.9 3.9 12.1
2.1 ± 0.2 96.2 ± 8.1 911.5 ± 136.6
104.4 96.2 91.2
7.6 8.4 15.0
fustin
2 100 1000
1.8 ± 0.2 102.5 ± 5.1 889.0 ± 18.4
92.4 102.5 88.9
8.1 5.0 2.1
1.7 ± 0.2 103.1 ± 6.0 946.4 ± 55.0
84.2 103.1 94.6
11.7 5.8 5.8
garbanzol
2 100 1000
1.9 ± 0.1 106.0 ± 5.7 868.9 ± 40.3
92.5 106.0 86.9
4.4 5.4 4.6
1.9 ± 0.1 109.9 ± 6.9 852.9 ± 39.8
94.3 109.9 85.3
4.2 6.2 4.7
sulfuretin
2 100 1000
1.9 ± 0.2 96.6 ± 4.1 1066 ± 39.0
94.8 96.6 106.6
9.9 4.2 3.7
2.1 ± 0.3 93.6 ± 9.1 977.9 ± 144.9
106.9 93.6 97.8
15.4 9.7 14.8
taxifolin
2 100 1000
1.8 ± 0.2 96.6 ± 4.1 969.7 ± 100.7
90.3 96.6 97.0
12.5 4.3 10.4
1.9 ± 0.0 98.0 ± 5.6 974.0 ± 93.6
97.2 98.0 97.4
1.5 5.7 9.6
at 100 °C and then filtered for removal of impurities. The extract was added to 10% ethanol and 50% dextrin and concentrated through spray-drying. The final desiccated extract was used for the pharmacokinetics experiments. Calibration and Quality Control (QC) Standards. Each analyte was dissolved in dimethyl sulfoxide at a concentration of 1 mg/mL to make a stock solution. The working standard mixture was prepared by diluting and combining each stock solution with methanol. The internal standard solution (IS; 2.5 μg/mL chrysin) was prepared in methanol. The working standard mixture (50 μL) was spiked into 450 μg/mL of blank rat plasma to prepare calibration (1, 2, 5, 10, 20, 50, 100, 200, 500, and 1000 ng/mL) and QC (2, 100, and 1000 ng/mL) standards. All standard samples were kept in a freezer (−20 °C) until they were analyzed. Sample Preparation. Fifty microliters of plasma was added to 200 μL of phosphate-buffered saline (PBS; pH 7.2) with IS (5 μL) and 10% ascorbic acid (10 μL) and extracted with 1 mL of ethyl acetate. For β-glucuronidase-treated samples, 50 μL of plasma was added to 200 μL of PBS with β-glucuronidase (1.4 units), and it was incubated for 60 min at 55 °C. After incubation, the mixture was spiked with IS (5 μL) and 10% ascorbic acid (10 μL) and extracted with 1 mL of ethyl acetate. The organic layer was taken and evaporated to dryness. The residue was reconstituted in 70% MeOH (100 μL) and transferred to an LC vial. LC-MS/MS Analysis. An Agilent 6460 triple-quadrupole mass spectrometer with an Agilent 1260 Infinity HPLC system (Agilent Technologies, Palo Alto, CA, USA) was used for LC-MS/MS analysis. A Poroshell 120 EC-C18 column (50 × 3.0 mm, 5 μm; Agilent) was used, and column temperature was adjusted to 40 °C. The mobile phase was composed of (A) 0.1% formic acid and (B) 90% acetonitrile
with 0.1% formic acid. The flow rate was 0.25 mL/min, and gradient elution was used as follows: 10−90% B for 3 min; 90% B for 1 min; and re-equilibrium (10% B) for 4 min. Electrospray ionization was used in negative ion mode. Multiple reaction monitoring (MRM) was employed. The precursor/product ion pairs were m/z 153 → 109 for 3,4-DHBA and 2,4-DHBA; m/z 287 → 109 for fustin; m/z 303 → 285 for taxifolin; m/z 271 → 243 for garbanzol; m/z 285 → 135 for fisetin; m/z 271 → 135 for butin; m/z 269 → 133 for sulfuretin; and m/z 253 → 143 for chrysin (IS). Dwell time for each transition was set at 150 ms. Method Validation. The analytical method validation was conducted according to the U.S. FDA guideline for bioanalytical method validation (2001). The calibration curve was generated using eight concentration levels of calibration standards in triplicate. The intraday assay for precision and accuracy was conducted by repeated analyses (n = 5) of QC samples at three concentrations (2, 100, and 1000 ng/mL) for each analyte. The interday assay was performed by repeated analyses on five consecutive days. The precision was presented as percent coefficient of variation (%CV). The accuracy was presented as percentage recoveries to nominal concentration. Freeze-and-thaw stability was tested at three concentrations (20, 100, and 500 ng/mL). The QC samples were frozen at −20 °C for 24 h and thawed completely at room temperature. After this cycle had been repeated two times more, the samples were analyzed. The QC samples were also tested for longterm (−20 °C for 2 weeks) and short-term (room temperature for 4 h) stabilities. Postpreparative stability was tested 24 h after the prepared QC samples had been kept in an autosampler (4 °C). Animals. Male Sprague−Dawley rats (270−300 g, 8 weeks old) were purchased from NaRa Biotech Co. (Pyungtaksi, Korea). The rats 5412
DOI: 10.1021/acs.jafc.5b01724 J. Agric. Food Chem. 2015, 63, 5410−5416
Article
Journal of Agricultural and Food Chemistry Table 2. Stability of Eight Phenolic Compounds of RVE in Rat Plasma under Different Storage Conditions (n = 3) compound
spiked concn (ng/mL)
short-term stability
long-term stability
freeze−thaw stability
autosampler stability
2,4-DHBA
20 100 500
100.0 ± 2.5 97.6 ± 0.9 94.8 ± 1.8
94.5 ± 1.0 95.6 ± 2.1 87.2 ± 0.3
107.4 ± 5.8 99.4 ± 6.8 99.4 ± 6.4
106.7 ± 6.8 90.0 ± 13.3 99.4 ± 2.3
3,4-DHBA
20 100 500
101.5 ± 1.1 101.1 ± 3.3 102.4 ± 1.6
87.3 ± 1.1 94.6 ± 1.3 85.0 ± 0.5
106.7 ± 1.5 100.0 ± 5.8 98.1 ± 5.2
112.7 ± 6.9 93.7 ± 13.7 102.5 ± 2.1
butin
20 100 500
98.8 ± 1.1 102.3 ± 2.6 104.6 ± 2.3
95.6 ± 1.0 94.6 ± 2.7 87.4 ± 0.4
107.5 ± 4.9 99.8 ± 6.7 101.9 ± 4.5
111.7 ± 1.0 96.0 ± 12.7 104.7 ± 1.1
fisetin
20 100 500
104.9 ± 4.6 97.1 ± 1.4 94.4 ± 3.8
101.2 ± 0.0 92.6 ± 5.1 95.7 ± 3.0
96.0 ± 1.8 104.9 ± 1.0 105.1 ± 1.4
107.5 ± 5.0 90.0 ± 1.4 88.4 ± 4.5
fustin
20 100 500
102.1 ± 2.2 101.2 ± 2.4 101.3 ± 2.2
93.6 ± 2.7 91.7 ± 2.8 85.4 ± 0.4
109.3 ± 3.2 99.3 ± 6.4 101.1 ± 3.7
115.1 ± 1.1 96.3 ± 11.0 103.4 ± 3.2
garbanzol
20 100 500
109.0 ± 2.8 106.4 ± 2.4 105.4 ± 1.6
94.8 ± 0.9 97.9 ± 2.4 86.4 ± 0.7
110.6 ± 2.7 102.0 ± 6.0 103.1 ± 3.7
116.7 ± 2.4 99.7 ± 10.1 106.5 ± 2.2
sulfuretin
20 100 500
95.0 ± 1.0 91.9 ± 1.9 97.9 ± 0.4
95.3 ± 2.0 95.5 ± 0.5 87.0 ± 0.4
106.3 ± 4.1 101.2 ± 7.2 102.7 ± 3.9
113.6 ± 2.7 95.3 ± 12.0 105.9 ± 1.9
taxifolin
20 100 500
87.1 ± 6.3 92.0 ± 4.3 92.6 ± 1.6
99.2 ± 6.4 95.8 ± 3.9 102.0 ± 2.0
105.0 ± 10.4 91.8 ± 7.4 94.0 ± 5.7
104.5 ± 2.9 85.9 ± 1.9 85.5 ± 1.8
were housed in a temperature-controlled (23 ± 2 °C) and moisturecontrolled (55 ± 10%) room under a 12 h light/dark cycle and allowed access to food and water. The rats were fasted overnight and until 6 h after dosing. All animal care and experimental procedures were approved by the Institutional Animal Care and Use Committee of Hanyang University (2015-0019). Pharmacokinetic Experiments. RVE (1.5 g/kg) in 3 mL of water was orally administered to the rats. Whole blood samples were collected by cannulating a polyethylene tube (PE-50) into the carotid artery at intervals of 1, 2, 4, 6, 8, 10, 12, and 24 h after dosing. The collected samples were centrifuged at 15000g for 5 min to obtain plasma. The plasma samples were kept frozen at −20 °C until they were analyzed.
■
Specificity. The specificity of the method was evaluated by comparing MRM chromatograms of 2,4-DHBA, 3,4-DHBA, fisetin, fustin, garbanzol, taxifolin, butin, sulfuretin, and chrysin for blank plasma, plasma spiked with standards, and plasma obtained 1 h after oral administration of RVE. As shown in Figure 2, there was no interfering peak at the retention time for each analyte in blank matrix. Matrix Effect. The postextraction method was used to evaluate the matrix effect. The test was performed at three concentrations for all analytes: 20, 100, and 500 ng/mL. The percent recovery was >90% at all concentrations for all analytes (data not shown). This indicated that the ion suppression by matrix effect was negligible in the present method. Linearity and Lower Limit of Quantitation (LLOQ). Good linearity (R > 0.998) was observed in the calibration curves (1− 1000 ng/mL) for all analytes. The back-calculated data for all concentration points except LLOQ showed acceptable accuracy (85−115%) and precision (≤15%). The LLOQ was 1 ng/mL for all analytes as the resulting data met the criteria for accuracy within 80−120% and precision of ≤20%. Precision and Accuracy. For all QC levels, the intraday precision was