Article pubs.acs.org/jpr
Metabolomic Investigation of the Anti-Platelet Aggregation Activity of Ginsenoside Rk1 Reveals Attenuated 12-HETE Production Hyun Kyoung Ju,† Jin Gyun Lee,† Mi Kyung Park,‡ So-Jung Park,§ Chang Hoon Lee,‡ Jeong Hill Park,† and Sung Won Kwon*,† †
College of Pharmacy and Research Institute of Pharmaceutical Sciences, Seoul National University, Seoul 151-742, Korea College of Pharmacy, Dongguk University-Seoul, Goyang 410-820, Korea § School of Pharmacy, Sungkyunkwan University, Suwon 440-746, Korea ‡
S Supporting Information *
ABSTRACT: Comprehensive metabolomics analysis is an effective method of measuring metabolite levels in the body following administration of a pharmaceutical compound and can allow for monitoring of the effects of the compound or assessment of appropriate treatment options for individual patients. In the present metabolomics study, samples pretreated with antiplatelet compounds were extracted and subjected to ultraperformance liquid chromatography/quadrupole time-of-flight mass spectrometry. The acquired data were processed using peak clustering and evaluated by partial leastsquares (PLS) and orthogonal projections to latent structures discriminant analyses (OPLS-DA). As a result, meaningful endogenous metabolites, namely eicosanoids and thromboxane B2 (TXB2), were identified. TXB2, a key element in platelet aggregation, was decreased upon ginsenoside Rk1 treatment via inhibition of cyclooxygenase (COX) activity. One of the arachidonic acid (AA) metabolites, 12-hydroxy5,8,10,14-eicosatetraenoic acid (12-HETE), was decreased significantly in the ginsenoside Rk1-treated platelets compared to the AA-induced group. In the mechanism study of ginsenoside Rk1, a strong linkage to intracellular calcium levels, which induce platelet activation, was found. Additionally, the translocation of 12-LOX from cytosol to membrane, which is related with the intracellular calcium levels, was determined. Therefore, a decreased 12-HETE level induced by ginsenoside Rk1 on antiplatelet aggregation is related to 12-LOX translocation resulting from decreased Ca2+ levels. This study shows that global metabolomic analysis has potential for use in understanding the biological behavior of antiplatelet drugs. KEYWORDS: ginsenoside Rk1, antiplatelet aggregation activity, metabolomics, 12-hydroxy-5,8,10,14-eicosatetraenoic acid, arachidonic acid, thromboxane B2
■
INTRODUCTION Platelets or thrombocytes are small, nonuniformly shaped cell fragments derived from megakaryocytes. They circulate in the blood of mammals and have been implicated in hemostasis, the formation of blood clots to stop bleeding. Platelet aggregation follows many biochemical and biophysical pathways and is, therefore, a relatively late indicator of platelet activation.1,2 The function of platelets can be directly evaluated by platelet aggregometry, in which cloudiness is measured following the addition of a certain chemical compound. Metabolites involved in platelet aggregation can be measured in response to platelet agonists, such as arachidonic acid (AA), collagen, or adenosine diphosphate (ADP).3 Some reports have provided analytical methods suitable for certain platelet-aggregation-related metabolites,4,5 but global analysis of metabolomes has not been applied to the characterization of platelet function. Global © 2012 American Chemical Society
metabolomics study can provide the hundreds or thousands metabolites involved in antiplatelet aggregation, simultaneously. Thus, the identification of these metabolites using a metabolomic approach could be useful to predict the responses of agonist-induced platelet aggregation.6 Metabolomics refers to the comprehensive and detailed measurement of metabolites produced by the body to evaluate the metabolism of pharmaceutical compounds and treatment options.7−10 A complete overview of a patient’s metabolic profile after the drug administration can provide information on the metabolic pathways altered by the compounds, either intentionally as a target or unintentionally as a side effect.11 Furthermore, metabolites generated from the compound itself Received: May 18, 2012 Published: August 9, 2012 4939
dx.doi.org/10.1021/pr300454f | J. Proteome Res. 2012, 11, 4939−4946
Journal of Proteome Research
Article
aggregometer (Chrono-Log Co., Havertown, PA, USA) by the turbidimetry method.19
can be identified for determination of the drug metabolic pathway and potentially for development of more efficient and safer compounds.12 In this study, antiplatelet-aggregation activities of ginsenosides isolated from processed white ginseng were analyzed using a metabolomics approach. Panax ginseng C. A. Meyer is one of several traditional medicines that have been used for many therapeutic purposes in the Asian countries for many years. The pharmacological effects of ginsenosides isolated from white or red ginseng have been demonstrated in the cardiovascular system, the immune system, and the central nervous system, as well as antistress and antioxidant activity and antiplatelet aggregation.13−17 According to our previous report, ginsenosides isolated from processed white ginseng exhibit antiplatelet-aggregation activity in response to adenosine diphosphate (ADP)-, collagen-, AA-, and U46619-induced aggregation.15,16 Ginsenoside Rk1, in particular, exhibits a strong dose-dependent inhibitory activity against AA-induced aggregation that is 8- to 22-fold higher than that of acetylsalicylic acid (ASA: aspirin) positive control. Unlike ASA, the mechanism of action of ginsenoside Rk1 on platelets aggregation has not been determined. Hence, the aim of the present study was to better elucidate how ginsenoside Rk1 affects AA-induced platelet aggregation using a metabolomics approach. The findings can be applied to further studies of the antiplatelet activity of ginsenoside Rk1 and its other pharmacological effects as well as in the development of potential drugs or dietary supplements derived from natural products.
Sample Preparation for UPLC/Q-TOF MS
Two hundred microliters of the collected sample were transferred to a new Eppendorf tube with 800 μL of −20 °C methanol. The sample tube was vortexed and incubated for 30 min at −4 °C for efficient protein precipitation. Then, 0.5 mL of the supernatant obtained from centrifugation was transferred to a new tube, where it was completely dried under nitrogen gas. The dried samples were reconstituted in 70% methanol and filtered with a 0.2 μm PTFE membrane filter for injection into the UPLC/MS system (Supporting Information Table 1). System Stability and Quality Control
An amino acid mixture consisting of alanine, arginine, aspartic acid, methionine, phenylalanine, glutamic acid, asparagine, glutamine, histidine, leucine, tryptophan, proline, serine, tyrosine, and valine was dissolved in 70% methanol. Six concentrations of standards including the 15 amino acids and thromboxane B2 (TXB2) were injected to check the m/z tolerance and detectable range in both positive and negative ion modes. The run sequence was performed with a random sequence generated from a utility offered by the Web site http://www. randomsequences.org, and a postblank run was applied in-between every eight measurements to eliminate injection carryover effects and baseline drift. All contaminations caused by the system and extraction were removed (Supporting Information Figures 2 and 3).
■
Data Processing and Multivariate Analyses
All Q-TOF raw files were calibrated in the enhanced quadratic calibration mode and exported for data bucketing by ProfileAnalysis software. Advanced bucketing was used to identify all of the compounds defined in the buckets according to the bucketing parameters, ΔRT 0.1 min and Δm/z 0.1 Da. The obtained bucket file was exported to SIMCA-P+ (v12.0) for multivariate analysis.20 In this analysis, the missing data value was set at 60%, and the variables were centered and scaled according to Pareto Variance using their averages and standard deviations.
EXPERIMENTAL PROCEDURES Ginsenoside Rk1 was isolated from white ginseng steamed at 120 °C for 3 h using the previously described method.18 Purity was evaluated by a HPLC/ELSD system (about 98%). Acetylsalicylic acid (ASA; aspirin), sodium arachidonate (AA), and apyrase and formic acids (85%) were obtained from Sigma Aldrich (St. Louis, MO, USA), and HPLC-grade methanol was purchased from Duksan (Kyungki-Do, South Korea). Collagen was purchased from Chrono-Log (Havertown, PA, USA). Fluo3 acetoxymethyl (AM) was acquired from Invitrogen (Molecular Probes, Eugene, OR, USA). 12-HETE, 15-HETE, water, the COX inhibitor screening assay kit, and the lipoxygenase inhibitor screening assay kit were obtained from Cayman Chemical (Ann Arbor, MI, USA).
Cyclooxygenase (COX)-Inhibitory Activity
COX-inhibitory activity was measured with the COX inhibitor screening assay kit (Cayman Chemical, Ann Arbor, MI, USA), which uses a specific antiserum that binds to all of the major prostaglandin compounds. The assay was performed according to the manufacturer’s instructions.
Biological Samples
Male Sprague−Dawley rats (n = 8 for each group) were acquired from ORIENT-BIO Laboratory Animal Research Center Co., Ltd. (Gyeonggi-do, Korea). Whole blood samples were collected under ether anesthesia with syringes containing 2.2% trisodium citrate solution as an anticoagulant (6:1). The washed platelet suspension (4.6 × 108/mL) was prepared using apyrase (Supporting Information Figure 1). The platelet count was determined with a hematology analyzer (Exell TM18, Drew Scientific Inc., Dallas, TX, USA).
Calcium Measurement of Platelets
Platelets loaded with calcium indicators (fluo-3 a.m., final concentration of 5 μM) were centrifuged at 500g for 10 min, and the obtained platelet pellets were resuspended in saline solution (1.0 × 108/mL platelets). After addition of an agonist, the change in Ca2+ concentration was monitored with an FP777 fluorometer (Jasco, Tokyo, Japan) at an excitation wavelength of 488 nm and an emission wavelength of 535 nm. The fluorometer was calibrated by measuring the fluorescence intensity of known calcium concentrations. The intracellular calcium concentration was calculated using the following equation:
Platelet Aggregation Assay in Vitro
A 500 μL aliquot was placed in aggregometer cuvette chambers, was allowed to reach 37 °C, and was incubated with stirring for 3 min. The samples were pretreated with the antiplatelet compounds, ASA and ginsenoside Rk1, and then with collagen and AA (final concentrations: 1 μg/mL and 3 μM, respectively) to induce platelet aggregation. The chambers were immediately placed on ice. Platelet aggregation was measured with an optical
[Ca 2 +] = Kd(F − Fmin)/(Fmax − F )
Fmin, the signal in the absence of calcium, was measured after lysing platelets with Tris-EGTA, whereas Fmax, the signal when 4940
dx.doi.org/10.1021/pr300454f | J. Proteome Res. 2012, 11, 4939−4946
Journal of Proteome Research
Article
Figure 1. Antiplatelet aggregation activity of ginsenoside Rk1 and acetylsalicylic acid (ASA). (a) The antiplatelet aggregatory activity by ginsenosides Rk1 was dose-dependent upon arachidonic acid-induced platelet aggregation. (b) The inhibitory activity of Rk1 and aspirin were compared 50 μM to show that the Rk1 strongly inhibited platelet aggregation. The extent of platelet aggregation is expressed as the light transmission % (Y axis), and it was recorded for 5 min (X axis).
results indicate that Rk1 exhibits a stronger antiplateletaggregation activity than ASA and that the action of ginsenoside Rk1 in platelets might be related to AA metabolism.
calcium is saturated, was measured after the addition of a calcium chloride solution (final concentration: 4 mM). Preparation of Platelet Subcellular Fractions for Western Blot Analysis
System Stability of UPLC/Q-TOF MS Analysis
The washed platelet was suspended at 1 × 108/mL in HEPESTyrode buffer (pH 7.4). The ginsenoside Rk1 and vehicle were pretreated for 2 min, and then the cells were stimulated with AA (3 μM) in the presence of a near threshold of collagen (1.0 μg/mL) at 37 °C for 5 min. The reaction was terminated by putting samples into an ice bath, and the samples were freeze− thawed three times with liquid nitrogen. The homogenates were centrifuged at 100,000g for 60 min at 4 °C (Optima TLX benchtop ultracentrifuge, Beckman Coulter, Brea, CA). The supernatant was used as the cytosolic fraction, and the pellet was gently washed with HEPES-Tyrode buffer and used as the membrane fraction.21
We detected 15 amino acids and TXB2 by UPLC/Q-TOF MS (Supporting Information (SI) Figure 4). All of the measured m/z values were calibrated with a calibrant injected for each run (SI Figure 5). The method validation was validated with the intra- and interprecision linearity and limit of detection (LOD) in negative- and positive-ion modes. Intra- and interprecision were determined by triplicate injections at the concentration 60 μg/mL for 1 day and five other concentrations for 3 days. The results were calculated with reference to the retention time tolerance, the calibrated m/z, and the signal intensity. The m/z tolerance was calculated to be within ±4 mDa (SI Figure 6), and the coefficient of variation (C.V.) for the retention time was
