Journal of Chromatography B, 967 (2014) 63–68
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Short communication
Metabolite identification of liguzinediol in dogs by ultra-flow liquid chromatography/tandem mass spectrometry Chen-xiao Shan, Wei Li, Hong-mei Wen ∗ , Xin-zhi Wang, Xiao-wen Zhu, Xiao-bin Cui School of Pharmacy, Nanjing University of Chinese Medicine, Nanjing 210029, People’s Republic of China
a r t i c l e
i n f o
Article history: Received 7 January 2014 Received in revised form 22 June 2014 Accepted 26 June 2014 Available online 5 July 2014 Keywords: Liguzinediol UFLC/Q-TOF MS Metabolites Beagle dog
a b s t r a c t Ultra-flow liquid chromatography/quadrupole-time-of-flight mass spectrometry (UFLC/Q-TOF MS) method combined with metabolitepilotMT software was used for analysis of the metabolites of liguzinediol in dogs. Urine, bile, feces and plasma samples were collected after intravenous administration of 8 mg/kg liguzinediol to healthy dogs. Besides liguzinediol, seven metabolites were detected and identified by UFLC/Q-TOF MS method. The results showed that liguzinediol had some main metabolic pathways in dogs including oxidation, sulfation, cysteine conjugation, N-acetylcysteine conjugation and glucuronidation. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Liguzinediol (LZDO) is a novel cardiotonic compounds with lower side effect, was synthesized in our laboratory, and its chemical name is 2,5-dihydroxymethyl-3,6-dimethylpyrazine (shown in Fig. 1). It has been proved that liguzinediol can activate cardiac sarcoplasmic reticulum calcium ATPase in vivo and vitro without arrhythmia risk [1–4]. Therefore, it is a promising candidate drug for cardiovascular diseases and has significantly smaller toxicity and higher bioavailability than any other positive inotropic drug in the market. Despite its therapeutic value has been proved [5] and well pharmacokinetics profiles in rats have been reported [6,7], its metabolism and probable drug-drug interactive in dogs is still unclear. The purpose of this study is to elucidate their metabolic profiles in Beagle dogs and identify its metabolites by interpreting finger-print product ions generated from tandem mass spectrometry. To clarify metabolic processes of LZDO in vivo, a rapid, sensitive and high resolution ultra-flow liquid chromatography/quadrupole time-of-flight mass spectrometry (UFLC/Q-TOF MS) method was established to identify its metabolites in dogs’ urine, bile, feces and plasma. Based on this method, LZDO and its metabolites were well separated and accurate masses of ion and valuable the MS/MS spectra were provided. In addition, metabolitepilotTM software combining with neutral loss filtration and mass defect filtration
∗ Corresponding author. E-mail address: [email protected] (H.-m. Wen). http://dx.doi.org/10.1016/j.jchromb.2014.06.029 1570-0232/© 2014 Elsevier B.V. All rights reserved.
technique made identify the multiple bioactive metabolites in vivo with short data interpretation time and high-quality structural information possible. The results probably provided helpful chemical information for further pharmacology and active mechanism. 2. Experiment 2.1. Reagent and material HPLC grade water, LC–MS grade methanol and formic acid was purchase from Merck KGaA (Darmstadt, Germany). LZDO (purity≥99.0%, by HPLC) was synthesized and separated in our laboratory. The structures were elucidated through NMR, MS, UV and IR methods. 2.2. LC condition LC separation was performed on a UFLC 20ADXR LC system in-line (Shimadzu Corporation, Kyoto, Japan) coupled with hybrid quadrupole time-of-flight tandem mass spectrometer QTOF MS (TripleTOFTM 5600 MS system, AB Sciex Corporation., Foster City, CA). The auto-sampler temperature was set at 10 ◦ C, and the injection volume was set at 2 l. LC was performed at 40 ◦ C using XR-ODS C18 column (2.0 × 100 mm, 2.2 m Shimadzu, Tokyo) and a gradient system with a gradient mobile phase of solvent A (0.1% formic acid in ultra-pure water) and solvent B (methanol) at a flow rate of 400 l/min. The following gradient program was used: 5% B for 1 min, linear gradient to 20% B in 1 min, 20% B for 1 min, linear gradient to 90% B in
64
C.-x. Shan et al. / J. Chromatogr. B 967 (2014) 63–68
N
1 min, 90% B for 2 min, B return to initial conditions, and equilibrate for 2 min before the next sample injection.
(M0 ) and seven metabolites of LZDO (M1 –M7 ) were detected after comparing with control samples by using the MetabolitepilotTM software. Metabolitepilot parameters followed as: Fragmentation ions were set at 122.0828 Da and 151.0867 Da. Neutral loss was set at 47.0144 Da. In addition, mass defect (MDF) was set at 40 mDa, which included I phase and II phase metabolic pathways, such as parent drug, oxidate, glucuronide, bis-glucuronide, glutathione, sulfate, glycine, cysteine pathways and so on. The finger-print product ions of LZDO (shown in Fig. 3) were found and selected to screen and identify its metabolites. All metabolites MS/MS spectrum were shown in supplementary Fig. 1.
2.3. Mass spectrometric condition
3.2. LC-MS/MS analysis of LZDO and metabolites
MS and MS/MS data were acquired by analyst® software using TripleTOFTM 5600 mass spectrometer with MS scan and information dependent acquisition (IDA) MS/MS scans. The mass spectrometer was operated in positive ion mode with a source temperature of 550 ◦ C by electrospray ionization (ESI). Nitrogen gas was used as ion source gases and collision gas. The instrument was optimized by using LZDO to obtain best parameters in MS and MS/MS experiments. The MS mass spectrometric analysis was conducted in the range of m/z 50–1000, with the following parameters: ion Source Gases 1 and 2 (GS1 and GS2), 50 psi; declustering potential, 60 V; entrance potential, 13 V and collision energy (CE) 10. MS/MS data were acquired using IDA methods. Most intense ten peaks for ions with m/z 50–1000 trigger IDA criteria using real time mass defect filtering (MDF) and dynamic background subtraction (DBS). The MS/MS collision energy was set at 35 with spread ±20. External mass calibration was performed automatically using calibrate delivery system.
LZDO (M0 ) showed an UFLC profile with a retention time at 3.61 min. Its ESI–MS spectrum gave a protonated molecular ion at m/z 169.0970 [M + H]+ (speculated as C8 H10 N2 O2 ). The major MS/MS product ions at high collision energy were m/z 151.0867 [M + H–H2 O]+ , 122.0843 [M + H–H2 O–CH2 NH]+ (rearrangement finger-print product ion, base peak). The finger-print product ions could well serve for screening and identifying metabolites from liguzinediol. The proposed fragmentations pattern was shown in Fig. 3. Metabolite M1 was found at a retention time at 2.64 min and its molecular formula was deduced as C14 H20 N2 O9 (m/z 361.1235 [M + H]+ ). The product ions at m/z 185.0916 [M + H–GlucA] + (loss of glucuronide group) suggested that it was a glucuronide conjugation product of LZDO. Besides, the product ion of 185.05 was 16 Da higher than M0 and the product ion of 138.0785 [M + H–GlucA–H2 O–CH2 NH]+ , also 16 Da higher than 122.0843 (a finger-print product ion of LZDO). All the evidence indicated that metabolite M1 was mono-oxidation and glucuronide conjugation of LZDO. Possible structure of M1 is 6-methyl-2,3,5trihydroxymethylpyrazine monoglucuronide. Metabolite M5 was detected at a retention time of 3.80 min with a protonated molecular ion of m/z 183.0761 [M + H]+ (14 Da higher than M0 ) and a derived formula of C8 H10 N2 O3 . MS/MS of this precursor ion afforded the product ion at m/z 165.0695 [M + H–H2 O]+ by loss of water, then product ion at m/z 139.0869 [M + H–CO2 ]+ by loss of carbon dioxide and product ion at m/z 121.0765 [M + H–CO2 –H2 O]+ by loss of water, which indicated that metabolite M5 was oxidation to carboxylic acid of LZDO and probable structure of M5 is 3,6-dimethyl-5hydroxymethylpyrazine-2-carboxylic acid. Metabolite M2 were eluted at a retention time of 2.88 min in bile and urine samples and its molecular formula was speculated as C8 H10 N2 O4 (m/z 199.0709 [M + H]+ ). M2 is 16 Dalton higher than M5 ([M + H]+ m/z 183.07) and should be oxidate of M5 . According to the MS/MS analysis of potential isomers M21 and M2-2, the proposed fragmentation pattern of M2 potential isomers M2-1 and M2-2 were shown in Fig. 4. The fragments of M2-1 should be probably m/z 199, 181, 163, 135, and the M2-2 probably m/z 199, 181, 163, 153, 135. The spectra of MS/MS with collision energy of 35 eV showed its major product ions of M2 at m/z 181.0607 [M + H–H2 O]+ , 163.0500 [M + H–2H2 O]+ , 135.0557 [M + H–2H2 O–CO]+ , but the fragment of m/z 153 was not detected (shown in supplementary). Thus, metabolite M2 was tentatively identified as 3,5-dihydroxymethyl-6-methylpyrazine-2-carboxylic acid (M2-1). Metabolite M3 was observed at a retention time of 3.03 min and produced a protonated molecular ion at m/z 272.1060 [M + H]+ . Its molecular formula was speculated a C11 H17 N3 O3 S, one cysteine more than LZDO. Its spectra of MS/MS with high collision energy showed major product ions at m/z 254.0963 [M + H–H2 O]+ , 183.0587 [M–C3 H6 NO2 ]+ , 151.0866 [M–C3 H6 NO2 S]+ , 122.0843 [M–C3 H6 NO2 S–CH2 NH]+ (one finger-print product ion of LZDO).
HO
N
OH
Fig. 1. Chemical structure of LZDO.
2.4. Sample collection Beagle dogs were fasted 12 h prior to study and then vein administrated with a clear solution of LZDO (8 mg/kg) according to pharmacological experiment. The dogs were kept individually in metabolic cages to collect urine and feces separately. For metabolite qualification, the blood samples were collected into heparin containing tubes (1000 IU/mL) at 0.5, 1, and 2 h and the urine and feces samples were collected at 0–4 h. The whole blood samples were immediately centrifuged at 4000 rpm for 10 min at 10 ◦ C, and the feces samples were homogenized after freeze-drying. For the biliary study, dogs were anaesthetized by intra-peritoneal administration of nembutal. Then the flexible plastic cannula surgically inserted into their common bile ducts to gather the bile before and at 12 h after administration. All the samples were stored at −80 ◦ C until they were analyzed. 2.5. Sample preparation 0.3 mL biological samples (or 0.3 g fecal samples) were mixed with 3.7 mL methanol and vortex-mixed for 2 min. After centrifugation for 5 min at 12,000 rpm, the supernatants were carefully transferred into other tubes and evaporated to dryness with continuous nitrogen at 35 ◦ C, respectively. The residue was reconstituted in 300 L mobile phase (water:menthol 50:50) and centrifuged at 15,000 rpm for 10 min. Supernatants were injected into UFLC/QTOF MS system used for further analysis. 3. Results and discussions 3.1. Metabolism profiles by UFLC-MS analysis Chromatographic profiles of LZDO in biological samples were analyzed by UFLC/Q-TOF MS (shown in Fig. 2). Parent compounds
C.-x. Shan et al. / J. Chromatogr. B 967 (2014) 63–68
65
Fig. 2. UFLC/QTOF MS base peak chromatograms of (A) urine, (B) feces, (C) bile and (D) plasma sample after i.v. administration of LZDO in dogs.
The results showed that metabolite M3 was cysteine conjugation of LZDO. The probable structure of M3 is S-[3,6-dimethyl-5hydroxymethylpyrazine-2-methyl]cysteine. Metabolite M4 with a retention time at 3.46 min had a protonated molecular ion at m/z 249.0534 [M + H]+ . Accurate mass measurement showed that the chemical formula of M4 was C8 H12 N2 O5 S. The MS/MS spectrum of M4 showed the product ions at m/z 169.0972 [M + H–SO3 ]+ , 151.0867 [M + H–SO3 –H2 O]+ , and 122.0839 [M + H–SO3 –H2 O–CH2 NH]+ . It was remarkable that M4
generated a product ion at m/z 169.09 by losing a sulfate unit, at m/z 151.08 and 122.08 which was considered as a finger-print product ion of LZDO. Hence, all the evidence indicated that metabolite M4 was identified as sulfate conjugation of LZDO. Probable structure of M4 is 3,6-dimethyl-5-hydroxymethylpyrazine-2-methyl sulfate. Metabolite M6 was detected a protonated molecular ion at m/z 345.1298 [M + H]+ (speculated as C14 H20 N2 O8 ) with retention time of 3.87 min. The characteristic neutral losses
66
C.-x. Shan et al. / J. Chromatogr. B 967 (2014) 63–68
Fig. 3. Q-TOF MS spectrum of LZDO with high mass collision energy and its proposed fragmentation pattern.
Fig. 4. The proposed fragmentation pattern of M2 potential isomers M2-1 and M2-2.
C.-x. Shan et al. / J. Chromatogr. B 967 (2014) 63–68
67
Table 1 LC–MS and LC–MS/MS analysis of LZDO and its metabolites in dog. No.
Metabolite name
Formula
Accurate mass detected (m/z)
ppm
R.T. (min)
Product ion (m/z)
Locationa
M0 M1 M2 M3 M4 M5 M6 M7
Parent Oxidation and Glucuronide Conjugationb Oxidation to Carboxylic Acidb Cysteine Conjugation Sulfate Conjugation Oxidation to Carboxylic Acid Glucuronide Conjugation N-Acetylcysteine Conjugation
C8 H12 N2 O2 C14 H20 N2 O9 C8 H10 N2 O4 C11 H17 N3 O3 S C8 H12 N2 O5 S C8 H10 N2 O3 C14 H20 N2 O8 C13 H19 N3 O4 S
169.0970 361.1235 199.0709 272.1060 249.0534 183.0761 345.1287 314.1168
−0.7 −1.7 −2.2 −1.3 −2.4 −1.8 −1.6 −0.4
3.61 2.64 2.88 3.03 3.46 3.80 3.87 5.72
151.09,122.08 185.09,167.09,138.08 181.06.163.05,135.05 254.09,183.05,151.09,122.08 169.10,151.08,122.08 165.07,139.09, 121.08 169.10,151.08,122.08 272.10,183.06,167.06,122.08
u,f,b,p U u,f u,f,b u,f,b,p u,f,b,p u,b,p u,f,b
a b
u the urine samples, b the bile samples, p the plasma samples, and f the feces samples. There may be several isomers existed.
Fig. 5. Proposed metabolic pathway of LZDO. 1 Oxidation, 2 Glucuronidation, 3 Sulfation, 4 N-Acetylcysteine Conjugation, 5 Cysteine Conjugation, 6 Deacetylation.
of 176 Da were found in the MS/MS spectrum of its product ion 169.0965 [M + H–GlucA]+ by loss of glucuronide group, followed by loss of water at m/z 151.0858 [M + H–GlucA–H2 O]+ and 122.0835 [M + H–GlucA–H2 O–CH2 NH]+ (one finger-print product ion of LZDO). These product ions also appeared in the product ion mass spectrum of M0 , which strongly suggested that metabolite M6 was glucuronide conjugation of LZDO and probable structure of M6 is 3,6-dimethyl-5-hydroxymethylpyrazine-2-methyl glucuronide. Metabolite M7 showed a protonated molecular ion at m/z 314.1175 [M + H]+ with a retention time of 5.72 min in urine, feces and bile samples. High collision energy analysis revealed product ions at 272.1071 [M + H–C2 H2 O]+ (as same as protonated molecular ion of M3 ), 183.0584 [M–C2 H2 O–C3 H6 NO2 ]+ (one product ion appeared in M3 ), 151.0868 [M–C5 H8 NO3 S]+ , 122.0841 [M–C5 H8 NO3 S–CH2 NH]+ (one finger-print product ion of LZDO and appeared in M3 ) and 167.0635 [M + H–H2 O–C5 H7 NO3 ]+ . It was notable that M7 generated product ion at m/z 167.0635 by losing a water and C5 H7 NO3 . It indicated that there was an N-acetylcysteine group in M7 . Hence, metabolite M7 was identified as N-acetylcysteine conjugate of LZDO probable structure of M7 is N-acetyl-S-[3,6-dimethyl-5-hydroxymethylpyrazine-2methyl]cysteine.
4. Conclusion In this study, the metabolites of LZDO administrated in Beagle dogs have been studied by ultra-flow liquid chromatography/quadrupole time-of-flight mass spectrometry combined with automated data analysis (MetabololitepilotTM ). LZDO and its seven metabolites (M1 –M7 ) were detected and identified simultaneously according to their retention times, accurate molecular masses and finger print product ions (shown in Table 1). Some reported structure of metabolites M3 (m/z 272.1066) and M7 (m/z 314.1175) [7] were corrected and identified as cysteine conjugation and Nacetylcysteine conjugation of LZDO respectively. The metabolite M5 (m/z 183.07) was further elucidated by 1 H NMR and 13 C NMR data. The results showed that LZDO had some main metabolic pathways in dogs including oxidation, sulfation, cysteine conjugation, N-acetylcysteine conjugation and glucuronidation on the basis of the chromatographic peak area. The main metabolites (M4 , M5 ) were found in all administrated samples. All the metabolites could be found in urine sample. Further oxidized M1 and M2 were found in urine and feces samples. M2 and M5 as one phase metabolites were probable active substance in vivo, need further confirm by pharmacological study. The possible structures of metabolites and predicted metabolic pathway in dogs were speculated in Fig. 5.
68
C.-x. Shan et al. / J. Chromatogr. B 967 (2014) 63–68
Based on the previous study, the major metabolites of M3 , M5 , M6 and M7 were the same as those found in rats. The metabolites of M1 , M2 and M4 were newly found in dogs. All the evidence showed that phase II metabolism (glutathione conjugation and glucuronidation) was main pathways of LZDO in rats and dogs. In addition, it seemed that LZDO was prone to oxidize in dogs rather than in rats and more oxidized metabolites were detected in dogs. Though the structures of metabolites cannot be determined conclusively by MS method alone and need further confirmed using other spectroscopic method, the present study is still very valuable and probably play important role in estimation of its drug-drug interactive and clinical application. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (81072542), Specialized Research Fund for the Doctoral Program of Higher Education of China (20123237110010), the Natural Science Foundation of Jiangsu Province (BK2011077), the Natural Science Fundation of Nanjing University of Chinese Medicine (12XZR21), and A Project Funded
by the Priority Academic Program Development of Jiangsu Higher Education Institutions. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jchromb. 2014.06.029. References [1] Z. Liu, H.M. Bian, L. Chen, W. Li, H.M. Wen, Chin. Pharm. J. 15 (2009) 1155– 1158. [2] Z. Liu, S. Zhou, W. Li, H.M. Wen, H.M. Bian, L. Chen, J. Chin, New Drugs Clin. Rem. 28 (2009) 293–296. [3] Z. Kaygisiz, H. Ozden, N. Erkasap, T. Koken, T. Gunduz, M. Ikizler, T. Kural, Acta Physiol. Hung. 97 (2010) 362–374. [4] M. Fang, W. Li, H.M. Wen, Z. Liu, L. Chen, J. Nanjing Univ. TCM 28 (2012) 279–281. [5] D.N. Zhang, X. Guo, Z.Q. Li, W. Wei, H.M. Bian, Chin. Pharmacol. Bull. 28 (2012) 572–574. [6] L. Zhang, W. Li, H.M. Wen, C.X.H.M. Bian, Shan, Chin. J. New drug 22 (2013) 1024–1028. [7] C.X. Shan, W. Li, H.M. Wen, X.Z. Wang, Y.H. Zhu, X.B. Cui, J. Pharm. Biomed. Anal. 62 (2012) 187–190.
Contents lists available at ScienceDirect
Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb
Short communication
Metabolite identification of liguzinediol in dogs by ultra-flow liquid chromatography/tandem mass spectrometry Chen-xiao Shan, Wei Li, Hong-mei Wen ∗ , Xin-zhi Wang, Xiao-wen Zhu, Xiao-bin Cui School of Pharmacy, Nanjing University of Chinese Medicine, Nanjing 210029, People’s Republic of China
a r t i c l e
i n f o
Article history: Received 7 January 2014 Received in revised form 22 June 2014 Accepted 26 June 2014 Available online 5 July 2014 Keywords: Liguzinediol UFLC/Q-TOF MS Metabolites Beagle dog
a b s t r a c t Ultra-flow liquid chromatography/quadrupole-time-of-flight mass spectrometry (UFLC/Q-TOF MS) method combined with metabolitepilotMT software was used for analysis of the metabolites of liguzinediol in dogs. Urine, bile, feces and plasma samples were collected after intravenous administration of 8 mg/kg liguzinediol to healthy dogs. Besides liguzinediol, seven metabolites were detected and identified by UFLC/Q-TOF MS method. The results showed that liguzinediol had some main metabolic pathways in dogs including oxidation, sulfation, cysteine conjugation, N-acetylcysteine conjugation and glucuronidation. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Liguzinediol (LZDO) is a novel cardiotonic compounds with lower side effect, was synthesized in our laboratory, and its chemical name is 2,5-dihydroxymethyl-3,6-dimethylpyrazine (shown in Fig. 1). It has been proved that liguzinediol can activate cardiac sarcoplasmic reticulum calcium ATPase in vivo and vitro without arrhythmia risk [1–4]. Therefore, it is a promising candidate drug for cardiovascular diseases and has significantly smaller toxicity and higher bioavailability than any other positive inotropic drug in the market. Despite its therapeutic value has been proved [5] and well pharmacokinetics profiles in rats have been reported [6,7], its metabolism and probable drug-drug interactive in dogs is still unclear. The purpose of this study is to elucidate their metabolic profiles in Beagle dogs and identify its metabolites by interpreting finger-print product ions generated from tandem mass spectrometry. To clarify metabolic processes of LZDO in vivo, a rapid, sensitive and high resolution ultra-flow liquid chromatography/quadrupole time-of-flight mass spectrometry (UFLC/Q-TOF MS) method was established to identify its metabolites in dogs’ urine, bile, feces and plasma. Based on this method, LZDO and its metabolites were well separated and accurate masses of ion and valuable the MS/MS spectra were provided. In addition, metabolitepilotTM software combining with neutral loss filtration and mass defect filtration
∗ Corresponding author. E-mail address: [email protected] (H.-m. Wen). http://dx.doi.org/10.1016/j.jchromb.2014.06.029 1570-0232/© 2014 Elsevier B.V. All rights reserved.
technique made identify the multiple bioactive metabolites in vivo with short data interpretation time and high-quality structural information possible. The results probably provided helpful chemical information for further pharmacology and active mechanism. 2. Experiment 2.1. Reagent and material HPLC grade water, LC–MS grade methanol and formic acid was purchase from Merck KGaA (Darmstadt, Germany). LZDO (purity≥99.0%, by HPLC) was synthesized and separated in our laboratory. The structures were elucidated through NMR, MS, UV and IR methods. 2.2. LC condition LC separation was performed on a UFLC 20ADXR LC system in-line (Shimadzu Corporation, Kyoto, Japan) coupled with hybrid quadrupole time-of-flight tandem mass spectrometer QTOF MS (TripleTOFTM 5600 MS system, AB Sciex Corporation., Foster City, CA). The auto-sampler temperature was set at 10 ◦ C, and the injection volume was set at 2 l. LC was performed at 40 ◦ C using XR-ODS C18 column (2.0 × 100 mm, 2.2 m Shimadzu, Tokyo) and a gradient system with a gradient mobile phase of solvent A (0.1% formic acid in ultra-pure water) and solvent B (methanol) at a flow rate of 400 l/min. The following gradient program was used: 5% B for 1 min, linear gradient to 20% B in 1 min, 20% B for 1 min, linear gradient to 90% B in
64
C.-x. Shan et al. / J. Chromatogr. B 967 (2014) 63–68
N
1 min, 90% B for 2 min, B return to initial conditions, and equilibrate for 2 min before the next sample injection.
(M0 ) and seven metabolites of LZDO (M1 –M7 ) were detected after comparing with control samples by using the MetabolitepilotTM software. Metabolitepilot parameters followed as: Fragmentation ions were set at 122.0828 Da and 151.0867 Da. Neutral loss was set at 47.0144 Da. In addition, mass defect (MDF) was set at 40 mDa, which included I phase and II phase metabolic pathways, such as parent drug, oxidate, glucuronide, bis-glucuronide, glutathione, sulfate, glycine, cysteine pathways and so on. The finger-print product ions of LZDO (shown in Fig. 3) were found and selected to screen and identify its metabolites. All metabolites MS/MS spectrum were shown in supplementary Fig. 1.
2.3. Mass spectrometric condition
3.2. LC-MS/MS analysis of LZDO and metabolites
MS and MS/MS data were acquired by analyst® software using TripleTOFTM 5600 mass spectrometer with MS scan and information dependent acquisition (IDA) MS/MS scans. The mass spectrometer was operated in positive ion mode with a source temperature of 550 ◦ C by electrospray ionization (ESI). Nitrogen gas was used as ion source gases and collision gas. The instrument was optimized by using LZDO to obtain best parameters in MS and MS/MS experiments. The MS mass spectrometric analysis was conducted in the range of m/z 50–1000, with the following parameters: ion Source Gases 1 and 2 (GS1 and GS2), 50 psi; declustering potential, 60 V; entrance potential, 13 V and collision energy (CE) 10. MS/MS data were acquired using IDA methods. Most intense ten peaks for ions with m/z 50–1000 trigger IDA criteria using real time mass defect filtering (MDF) and dynamic background subtraction (DBS). The MS/MS collision energy was set at 35 with spread ±20. External mass calibration was performed automatically using calibrate delivery system.
LZDO (M0 ) showed an UFLC profile with a retention time at 3.61 min. Its ESI–MS spectrum gave a protonated molecular ion at m/z 169.0970 [M + H]+ (speculated as C8 H10 N2 O2 ). The major MS/MS product ions at high collision energy were m/z 151.0867 [M + H–H2 O]+ , 122.0843 [M + H–H2 O–CH2 NH]+ (rearrangement finger-print product ion, base peak). The finger-print product ions could well serve for screening and identifying metabolites from liguzinediol. The proposed fragmentations pattern was shown in Fig. 3. Metabolite M1 was found at a retention time at 2.64 min and its molecular formula was deduced as C14 H20 N2 O9 (m/z 361.1235 [M + H]+ ). The product ions at m/z 185.0916 [M + H–GlucA] + (loss of glucuronide group) suggested that it was a glucuronide conjugation product of LZDO. Besides, the product ion of 185.05 was 16 Da higher than M0 and the product ion of 138.0785 [M + H–GlucA–H2 O–CH2 NH]+ , also 16 Da higher than 122.0843 (a finger-print product ion of LZDO). All the evidence indicated that metabolite M1 was mono-oxidation and glucuronide conjugation of LZDO. Possible structure of M1 is 6-methyl-2,3,5trihydroxymethylpyrazine monoglucuronide. Metabolite M5 was detected at a retention time of 3.80 min with a protonated molecular ion of m/z 183.0761 [M + H]+ (14 Da higher than M0 ) and a derived formula of C8 H10 N2 O3 . MS/MS of this precursor ion afforded the product ion at m/z 165.0695 [M + H–H2 O]+ by loss of water, then product ion at m/z 139.0869 [M + H–CO2 ]+ by loss of carbon dioxide and product ion at m/z 121.0765 [M + H–CO2 –H2 O]+ by loss of water, which indicated that metabolite M5 was oxidation to carboxylic acid of LZDO and probable structure of M5 is 3,6-dimethyl-5hydroxymethylpyrazine-2-carboxylic acid. Metabolite M2 were eluted at a retention time of 2.88 min in bile and urine samples and its molecular formula was speculated as C8 H10 N2 O4 (m/z 199.0709 [M + H]+ ). M2 is 16 Dalton higher than M5 ([M + H]+ m/z 183.07) and should be oxidate of M5 . According to the MS/MS analysis of potential isomers M21 and M2-2, the proposed fragmentation pattern of M2 potential isomers M2-1 and M2-2 were shown in Fig. 4. The fragments of M2-1 should be probably m/z 199, 181, 163, 135, and the M2-2 probably m/z 199, 181, 163, 153, 135. The spectra of MS/MS with collision energy of 35 eV showed its major product ions of M2 at m/z 181.0607 [M + H–H2 O]+ , 163.0500 [M + H–2H2 O]+ , 135.0557 [M + H–2H2 O–CO]+ , but the fragment of m/z 153 was not detected (shown in supplementary). Thus, metabolite M2 was tentatively identified as 3,5-dihydroxymethyl-6-methylpyrazine-2-carboxylic acid (M2-1). Metabolite M3 was observed at a retention time of 3.03 min and produced a protonated molecular ion at m/z 272.1060 [M + H]+ . Its molecular formula was speculated a C11 H17 N3 O3 S, one cysteine more than LZDO. Its spectra of MS/MS with high collision energy showed major product ions at m/z 254.0963 [M + H–H2 O]+ , 183.0587 [M–C3 H6 NO2 ]+ , 151.0866 [M–C3 H6 NO2 S]+ , 122.0843 [M–C3 H6 NO2 S–CH2 NH]+ (one finger-print product ion of LZDO).
HO
N
OH
Fig. 1. Chemical structure of LZDO.
2.4. Sample collection Beagle dogs were fasted 12 h prior to study and then vein administrated with a clear solution of LZDO (8 mg/kg) according to pharmacological experiment. The dogs were kept individually in metabolic cages to collect urine and feces separately. For metabolite qualification, the blood samples were collected into heparin containing tubes (1000 IU/mL) at 0.5, 1, and 2 h and the urine and feces samples were collected at 0–4 h. The whole blood samples were immediately centrifuged at 4000 rpm for 10 min at 10 ◦ C, and the feces samples were homogenized after freeze-drying. For the biliary study, dogs were anaesthetized by intra-peritoneal administration of nembutal. Then the flexible plastic cannula surgically inserted into their common bile ducts to gather the bile before and at 12 h after administration. All the samples were stored at −80 ◦ C until they were analyzed. 2.5. Sample preparation 0.3 mL biological samples (or 0.3 g fecal samples) were mixed with 3.7 mL methanol and vortex-mixed for 2 min. After centrifugation for 5 min at 12,000 rpm, the supernatants were carefully transferred into other tubes and evaporated to dryness with continuous nitrogen at 35 ◦ C, respectively. The residue was reconstituted in 300 L mobile phase (water:menthol 50:50) and centrifuged at 15,000 rpm for 10 min. Supernatants were injected into UFLC/QTOF MS system used for further analysis. 3. Results and discussions 3.1. Metabolism profiles by UFLC-MS analysis Chromatographic profiles of LZDO in biological samples were analyzed by UFLC/Q-TOF MS (shown in Fig. 2). Parent compounds
C.-x. Shan et al. / J. Chromatogr. B 967 (2014) 63–68
65
Fig. 2. UFLC/QTOF MS base peak chromatograms of (A) urine, (B) feces, (C) bile and (D) plasma sample after i.v. administration of LZDO in dogs.
The results showed that metabolite M3 was cysteine conjugation of LZDO. The probable structure of M3 is S-[3,6-dimethyl-5hydroxymethylpyrazine-2-methyl]cysteine. Metabolite M4 with a retention time at 3.46 min had a protonated molecular ion at m/z 249.0534 [M + H]+ . Accurate mass measurement showed that the chemical formula of M4 was C8 H12 N2 O5 S. The MS/MS spectrum of M4 showed the product ions at m/z 169.0972 [M + H–SO3 ]+ , 151.0867 [M + H–SO3 –H2 O]+ , and 122.0839 [M + H–SO3 –H2 O–CH2 NH]+ . It was remarkable that M4
generated a product ion at m/z 169.09 by losing a sulfate unit, at m/z 151.08 and 122.08 which was considered as a finger-print product ion of LZDO. Hence, all the evidence indicated that metabolite M4 was identified as sulfate conjugation of LZDO. Probable structure of M4 is 3,6-dimethyl-5-hydroxymethylpyrazine-2-methyl sulfate. Metabolite M6 was detected a protonated molecular ion at m/z 345.1298 [M + H]+ (speculated as C14 H20 N2 O8 ) with retention time of 3.87 min. The characteristic neutral losses
66
C.-x. Shan et al. / J. Chromatogr. B 967 (2014) 63–68
Fig. 3. Q-TOF MS spectrum of LZDO with high mass collision energy and its proposed fragmentation pattern.
Fig. 4. The proposed fragmentation pattern of M2 potential isomers M2-1 and M2-2.
C.-x. Shan et al. / J. Chromatogr. B 967 (2014) 63–68
67
Table 1 LC–MS and LC–MS/MS analysis of LZDO and its metabolites in dog. No.
Metabolite name
Formula
Accurate mass detected (m/z)
ppm
R.T. (min)
Product ion (m/z)
Locationa
M0 M1 M2 M3 M4 M5 M6 M7
Parent Oxidation and Glucuronide Conjugationb Oxidation to Carboxylic Acidb Cysteine Conjugation Sulfate Conjugation Oxidation to Carboxylic Acid Glucuronide Conjugation N-Acetylcysteine Conjugation
C8 H12 N2 O2 C14 H20 N2 O9 C8 H10 N2 O4 C11 H17 N3 O3 S C8 H12 N2 O5 S C8 H10 N2 O3 C14 H20 N2 O8 C13 H19 N3 O4 S
169.0970 361.1235 199.0709 272.1060 249.0534 183.0761 345.1287 314.1168
−0.7 −1.7 −2.2 −1.3 −2.4 −1.8 −1.6 −0.4
3.61 2.64 2.88 3.03 3.46 3.80 3.87 5.72
151.09,122.08 185.09,167.09,138.08 181.06.163.05,135.05 254.09,183.05,151.09,122.08 169.10,151.08,122.08 165.07,139.09, 121.08 169.10,151.08,122.08 272.10,183.06,167.06,122.08
u,f,b,p U u,f u,f,b u,f,b,p u,f,b,p u,b,p u,f,b
a b
u the urine samples, b the bile samples, p the plasma samples, and f the feces samples. There may be several isomers existed.
Fig. 5. Proposed metabolic pathway of LZDO. 1 Oxidation, 2 Glucuronidation, 3 Sulfation, 4 N-Acetylcysteine Conjugation, 5 Cysteine Conjugation, 6 Deacetylation.
of 176 Da were found in the MS/MS spectrum of its product ion 169.0965 [M + H–GlucA]+ by loss of glucuronide group, followed by loss of water at m/z 151.0858 [M + H–GlucA–H2 O]+ and 122.0835 [M + H–GlucA–H2 O–CH2 NH]+ (one finger-print product ion of LZDO). These product ions also appeared in the product ion mass spectrum of M0 , which strongly suggested that metabolite M6 was glucuronide conjugation of LZDO and probable structure of M6 is 3,6-dimethyl-5-hydroxymethylpyrazine-2-methyl glucuronide. Metabolite M7 showed a protonated molecular ion at m/z 314.1175 [M + H]+ with a retention time of 5.72 min in urine, feces and bile samples. High collision energy analysis revealed product ions at 272.1071 [M + H–C2 H2 O]+ (as same as protonated molecular ion of M3 ), 183.0584 [M–C2 H2 O–C3 H6 NO2 ]+ (one product ion appeared in M3 ), 151.0868 [M–C5 H8 NO3 S]+ , 122.0841 [M–C5 H8 NO3 S–CH2 NH]+ (one finger-print product ion of LZDO and appeared in M3 ) and 167.0635 [M + H–H2 O–C5 H7 NO3 ]+ . It was notable that M7 generated product ion at m/z 167.0635 by losing a water and C5 H7 NO3 . It indicated that there was an N-acetylcysteine group in M7 . Hence, metabolite M7 was identified as N-acetylcysteine conjugate of LZDO probable structure of M7 is N-acetyl-S-[3,6-dimethyl-5-hydroxymethylpyrazine-2methyl]cysteine.
4. Conclusion In this study, the metabolites of LZDO administrated in Beagle dogs have been studied by ultra-flow liquid chromatography/quadrupole time-of-flight mass spectrometry combined with automated data analysis (MetabololitepilotTM ). LZDO and its seven metabolites (M1 –M7 ) were detected and identified simultaneously according to their retention times, accurate molecular masses and finger print product ions (shown in Table 1). Some reported structure of metabolites M3 (m/z 272.1066) and M7 (m/z 314.1175) [7] were corrected and identified as cysteine conjugation and Nacetylcysteine conjugation of LZDO respectively. The metabolite M5 (m/z 183.07) was further elucidated by 1 H NMR and 13 C NMR data. The results showed that LZDO had some main metabolic pathways in dogs including oxidation, sulfation, cysteine conjugation, N-acetylcysteine conjugation and glucuronidation on the basis of the chromatographic peak area. The main metabolites (M4 , M5 ) were found in all administrated samples. All the metabolites could be found in urine sample. Further oxidized M1 and M2 were found in urine and feces samples. M2 and M5 as one phase metabolites were probable active substance in vivo, need further confirm by pharmacological study. The possible structures of metabolites and predicted metabolic pathway in dogs were speculated in Fig. 5.
68
C.-x. Shan et al. / J. Chromatogr. B 967 (2014) 63–68
Based on the previous study, the major metabolites of M3 , M5 , M6 and M7 were the same as those found in rats. The metabolites of M1 , M2 and M4 were newly found in dogs. All the evidence showed that phase II metabolism (glutathione conjugation and glucuronidation) was main pathways of LZDO in rats and dogs. In addition, it seemed that LZDO was prone to oxidize in dogs rather than in rats and more oxidized metabolites were detected in dogs. Though the structures of metabolites cannot be determined conclusively by MS method alone and need further confirmed using other spectroscopic method, the present study is still very valuable and probably play important role in estimation of its drug-drug interactive and clinical application. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (81072542), Specialized Research Fund for the Doctoral Program of Higher Education of China (20123237110010), the Natural Science Foundation of Jiangsu Province (BK2011077), the Natural Science Fundation of Nanjing University of Chinese Medicine (12XZR21), and A Project Funded
by the Priority Academic Program Development of Jiangsu Higher Education Institutions. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jchromb. 2014.06.029. References [1] Z. Liu, H.M. Bian, L. Chen, W. Li, H.M. Wen, Chin. Pharm. J. 15 (2009) 1155– 1158. [2] Z. Liu, S. Zhou, W. Li, H.M. Wen, H.M. Bian, L. Chen, J. Chin, New Drugs Clin. Rem. 28 (2009) 293–296. [3] Z. Kaygisiz, H. Ozden, N. Erkasap, T. Koken, T. Gunduz, M. Ikizler, T. Kural, Acta Physiol. Hung. 97 (2010) 362–374. [4] M. Fang, W. Li, H.M. Wen, Z. Liu, L. Chen, J. Nanjing Univ. TCM 28 (2012) 279–281. [5] D.N. Zhang, X. Guo, Z.Q. Li, W. Wei, H.M. Bian, Chin. Pharmacol. Bull. 28 (2012) 572–574. [6] L. Zhang, W. Li, H.M. Wen, C.X.H.M. Bian, Shan, Chin. J. New drug 22 (2013) 1024–1028. [7] C.X. Shan, W. Li, H.M. Wen, X.Z. Wang, Y.H. Zhu, X.B. Cui, J. Pharm. Biomed. Anal. 62 (2012) 187–190.
