http://informahealthcare.com/xen ISSN: 0049-8254 (print), 1366-5928 (electronic) Xenobiotica, 2014; 44(8): 734–742 ! 2014 Informa UK Ltd. DOI: 10.3109/00498254.2014.880201
RESEARCH ARTICLE
Structure identification and elucidation of mosapride metabolites in human urine, feces and plasma by ultra performance liquid chromatography-tandem mass spectrometry method Xiaohong Sun*,y, Longshan Zhaoy, Lili Niu, Feng Qin, Xiumei Lu, Zhili Xiong, and Famei Li
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
School of Pharmacy, Shenyang Pharmaceutical University, Shenyang, PR China
Abstract
Keywords
1. Mosapride citrate (mosapride) is a potent gastroprokinetic agent. The only previous study on mosapride metabolism in human reported one phase I oxidative metabolite, des-p-fluorobenzyl mosapride, in human plasma and urine using HPLC method. Our aim was to identify mosapride phase I and phase II metabolites in human urine, feces and plasma using UPLC-ESI-MS/MS. 2. A total of 16 metabolites were detected. To the best of our knowledge, 15 metabolites have not been reported previously in human. 3. Two new metabolites, morpholine ring-opened mosapride (M15) and mosapride N-oxide (M16), alone with one known major metabolite, des-p-fluorobenzyl mosapride (M3), were identified by comparison with the reference standards prepared by our group. The chemical structures of seven phase I and six phase II metabolites of mosapride were elucidated based on UPLC–MS/MS analyses. 4. There were two major phase I reactions, dealkylation and morpholine ring cleavage. Phase II reactions included glucuronide, glucose and sulfate conjugation. The comprehensive metabolic pathway of mosapride in human was proposed for the first time. 5. The metabolites in humans were compared with those in rats reported previously. In addition to M10, the other 15 metabolites in humans were also found in rats. This result suggested that there was little qualitative species difference in the metabolism of mosapride between rats and humans. 6. In all, 16 mosapride metabolites including 15 new metabolites were reported. These results allow a better understanding of mosapride disposition in human.
Human, metabolism, mosapride, UPLC-ESI-MS/MS
Introduction Mosapride citrate (mosapride), 4-amino-5-chloro-2-ethoxyN-[[4-(4-fluorobenzyl)-2-morpholinyl]methyl] benzamide citrate, is a selective serotonin 5-HT4 receptor agonist with no affinity for dopamine D2 receptor (Yoshida, 1999). It is a potent gastroprokinetic agent used to improve gastrointestinal (GI) symptoms in patients with gastroesophageal reflux disease and to ameliorate constipation and response fluctuations in Parkinsonian patients (Asai et al., 2005; Liu et al., 2005). Due to little arrhythmic effect associated with the QT
*Current address: National Center for Plant Gene Research (Beijing), Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing 100101, China. yThese authors contributed equally to the article. Address for correspondence: Longshan Zhao and Famei Li, School of Pharmacy, Shenyang Pharmaceutical University, 103 Wenhua Road, Shenyang 110016, PR China. Tel: +86 24 23986289. Fax: +86 24 23986289. E-mail: [email protected] (L. Zhao); fameili@163. com (F. Li)
History Received 18 November 2013 Revised 31 December 2013 Accepted 1 January 2014 Published online 13 January 2014
prolongation, mosapride has been used increasingly in Japan and some other countries (Curran & Robinson, 2008). Few attempts have been made to elucidate the exact metabolic profile of mosapride in human, although it has been marketed for about 20 years (Matsumoto et al., 1993a,b; Sakashita et al., 1993). The first study can be traced to Matsumoto et al. (1993b), where mosapride metabolism in rat was reported using radioisotope labeling-TLC assay. Although it was sensitive, the selectivity of the method was poor and it could not give any information on metabolite structures. Four phase I metabolites were isolated from rat urine and identified as des-p-fluorobenzyl mosapride, 50 -oxo-des-p-fluorobenzyl mosapride, 3-hydroxy des-p-fluorobenzyl and 3-hydroxy 50 -oxo-des-p-fluorobenzyl mosapride. An HPLC method was used for the metabolism study of mosapride in human and only one phase I metabolite, des-p-fluorobenzyl mosapride, was identified in plasma and urine (Sakashita et al., 1993). The advanced tandem mass spectrometry (MS/MS) has become an indispensable tool in metabolite detection and identification (Amir et al., 2013; Clarke et al., 2001; Hsieh & Korfmacher, 2006; Nassar et al., 2006; Oliveira & Watson, 2000; Pedraglio et al., 2007) and it
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
DOI: 10.3109/00498254.2014.880201
can be utilized for improved mosapride metabolism study. Previously in our group, UPLC-MS/MS method was used to study mosapride metabolism in rat before initiating its human metabolism described herein. A total of 18 metabolites were observed in rat, with the chemical structures of two new metabolites identified and at least 13 new ones proposed (Sun et al., 2009a). According to the Guidance for Industry Safety Testing of Drug Metabolites (U.S. Food and Drug Administration, 2008), it is encouraged that the identification of differences in drug metabolism between animal test species and humans as early as possible. However, there is no study on the overall metabolic profile of mosapride in human. Drug metabolism can influence a drug’s distribution, rate or route of excretion and production of new and possibly active or toxic species (Bailey et al., 2013). Characterizing the metabolic profile of mosapride is critical in understanding its pharmacokinetics as well as its adverse effect profile by identifying potential reactive intermediates or adducts. Hence, our aim is to identify and elucidate mosapride phase I and phase II metabolites in human with UPLC–MS/MS method, characterize possible metabolic pathways and compare the metabolites in human with those previously reported in rat (Sun et al., 2009a).
Experimental procedures Chemicals The reference standard of mosapride citrate (Figure 1, 99.2% purity) was purchased from the National Institutes for Food and Drug Control (Beijing, China). Mosapride citrate tablets, containing 5 mg of mosapride, was supplied by Dainippon ¯ saka-shi, Osaka-fu, Japan). Sumitomo Pharma Co., Ltd. (O
Structure identification of mosapride metabolites
735
Des-p-fluorobenzyl mosapride, mosapride N-oxide and morpholine ring-opened mosapride were isolated and purified from the preparative-scale microbial transformation of mosapride and identified as pure compounds by UV, MS, NMR with the purity above 98% by our laboratory (Sun et al., 2009b). Methanol, acetonitrile and formic acid were of HPLC grade and purchased from Dikma (Richmond Hill, USA). All other chemicals were of analytical grade. Apparatus and UPLC–MS/MS conditions UPLC–MS/MS system consisted of an ACQUITYTM UPLC system (Milford, MA) and Waters MicromassÕ Quattro microTM API mass spectrometer (Manchester, UK) equipped with an ESI source. The chromatographic separation was achieved on an ACQUITY UPLCTM BEH C18 column (100 mm 2.1 mm; i.d. 1.7 mm; Waters Corp., Milford, MA) with the column temperature set at 40 C. The flow rate was 0.25 ml/min with a linear gradient running from 84% to 78% A (solvent A, 0.2% formic acid aqueous; solvent B, acetonitrile) in 4 min, 78% to 76% A in the next 8 min, then to 10% A for 1 min, and returned to initial condition lasting 2 min for reequilibration. The total run time per sample was 15 min. As for the MS performed in the positive ESI mode, the optimal MS parameters optimized using mosapride standard were as follows: capillary voltage 3.0 kV, cone voltage 30 kV, source temperature 100 C and desolvation temperature 350 C. Nitrogen was used as the desolvation and cone gas with a flow rate of 500 and 50 l/h, respectively. For MS/MS analyses performed in product scan and MRM modes, argon was used as the collision gas set at 2.5 e–3 mbar, and the
Figure 1. (A) Full scan MS spectrum of mosapride; (B) MS/MS product ion spectrum of mosapride; (C) MRM chromatogram of mosapride by UPLC– MS/MS analysis, the monitored MRM transition was m/z 422 ! 198; (D) Predominant fragmentation patterns of mosapride under ESI–MS/MS analysis.
736
X. Sun et al.
collision energy ranged from 18 to 40 eV for all metabolites. Data was acquired and processed using MassLynxTM NT 4.1 software (Waters Corp., Milford, MA). Human subjects and sample collection
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
Healthy volunteers (25–28 years old), who gave their signed written informed consent to the study, fasted overnight and were orally given 10 mg mosapride citrated dispersible tablet (two tablets). No food was allowed for at least 2 h after drug administration. Urine samples were collected before (0 h) and at 6, 12, 24 and 30 h time intervals. Feces samples were collected before (0 h) and at 24 h post-dosing. Blood samples were collected and plasma separated before (0 h) and at 0.13, 0.25, 0.33, 0.42, 0.50, 0.75, 1.0, 1.5, 2.0, 3.5, 5.0, 7.0 and 10.0 h post-dosing. All the biological samples were collected into silicone coated glass tubes and stored at 20 C until analysis. Sample preparation Oasis MCXÕ mixed-mode cation-exchange cartridges (1 ml volume, Waters Corp.) were conditioned with 1 ml of methanol, 1 ml of water, each twice. Plasma and urine samples were centrifuged at 2000 g for 10 min, the supernatant were removed for use. The samples were filtered through 0.45 mm membrane and 1 ml of plasma and urine samples were loaded onto the SPE cartridges. Then, the SPE cartridges were washed with 1 ml of 1% formic acid aqueous, 1 ml of methanol and eluted with 1 ml of methanol containing 5% ammonium hydroxide. The eluate was evaporated to dryness under a gentle stream of nitrogen at 40 C and the residue was dissolved in 200 ml of the mixture of acetonitrile – 0.2% formic acid aqueous (3:7, v/v). A small aliquot of 10 ml of the solution was injected into the UPLC–MS/MS system. Feces samples (0.5 g) were weighted and supplemented with an appropriate volume (6 ml/g) of the mixture of methanol-water (8:2, v/v). After ultrasonic extraction for 20 min, the samples were centrifuged at 2000 g for 10 min. The supernatant was transferred to a clean tube and 6 ml of ethyl acetate was added. After vortexing and centrifugation at 2000 g for 10 min, the supernatant was transferred to a clean tube and dried under a flow of nitrogen at 35 C. The residue was reconstituted in 300 ml of methanol. After filtered through 0.45 mm membrane, an aliquot of 10 ml was injected into the UPLC–MS/MS system for analysis.
Results Fragmentation patterns of mosapride The fragmentation patterns of mosapride served as templates for interpreting the structures of the metabolites. Under ESI in positive ion mode, mosapride easily formed the protonated molecular ion [M + H]+ at m/z 422 in full scan MS with the isotopic peak obserbed at m/z 424 for the 37Cl parent ion (Figure 1A). The product ion spectrum of the protonated molecular ion of mosapride is shown in Figure 1(B). Mosapride was eluted at 8.12 min under MRM mode (Figure 1C). Fragmentation of [M + H]+ of mosapride led to three product ions and the fragmentation patterns
Xenobiotica, 2014; 44(8): 734–742
are illustrated in Figure 1(D). The diagnostic fragment ions at m/z 198 and 170 were characteristic of the benzoyl part, while the one at m/z 109 was characteristic of the p-fluorobenzyl part. These characteristic product ions were the sound bases to identify the metabolites of mosapride in human. Identification and structure elucidation of metabolites In order to find the metabolites, possible structures of metabolites were speculated according to the metabolism rule of drugs firstly. Great attention had been paid to the 18 metabolites of mosapride characterized in rat (Sun et al., 2009a) to facilitate the study on metabolism of mosapride in human (urine, feces and plasma). Then, the TICs of samples after administration were compared with those of the blank samples so as to find the protonated molecular ion of potential metabolites. After that, MS/MS product ion spectra of [M + H]+ of metabolites were obtained and the fragmentation patterns were characterized (Figure 2). Finally, possible metabolites were detected by UPLC–MS/MS under MRM mode. Representative MRM chromatograms of mosapride and its 16 metabolites in blank human matrix (including urine, feces and plasma) and human matrix samples after oral administration of 10 mg of mosapride citrate dispersible tablets are shown in Figure 3. The chromatographic retention times, MS/MS data and relative abundance of mosapride and the metabolites in humans are summarized in Table 1. Relative abundance was semi-quantitative. It was obtained by dividing the peak area of each analyte by the sum of peak area of mosapride and 16 metabolites in every matrix (Dai et al., 2008; Sun et al., 2009a). It is not possible to compare absolute formation of the different metabolites because not all synthetic metabolite standards are available to compensate for the differences in metabolite ionization efficiency. Based on the method mentioned above, mosapride and 16 metabolites (M1–M16, according to the retention time) were found in human matrix (including urine, feces and plasma) after oral administration of mosapirde citrate dispersible tablets. By comparison with reference standards obtained by our group (Sun et al., 2009b), the chromatographic and mass spectrometric data of the protonated molecular ions at m/z 422 (M0), 314 (M3), 396 (M15) and 438 (M16) with retention times at 8.12, 3.20, 7.26 and 9.14 min were the same with those of mosapride, desp-fluorobenzyl mosapride, mosapride N-oxide and morpholine ring-opened mosapride reference standards, respectively. Therefore, M0 was confirmed as the unchanged parent drug, while M3, M15 and M16 were identified as desp-fluorobenzyl mosapride, mosapride N-oxide and morpholine ring-opened mosapride. Structure elucidation of other 13 metabolites was through comparing the changes in retention times, observed mass (DM) and mass spectral patterns of product ions of metabolites with those of mosapride. To the best of our knowledge, M3 was the only known metabolite of mosapride in human (plasma and urine) and the other 15 metabolites were reported in human for the first time. The structure identification and elucidation of metabolites is explained in detail below.
Figure 2. MS/MS product ion spectra and corresponding fragmentation patterns of mosapride metabolites in human. (A) M1; (B) M2; (C) M3; (D) M4; (E) M5; (F) M6; (G) M7; (H) M8; (I) M9; (J) M10; (K) M11; (L) M12; (M) M13; (N) M14; (O) M15; (P) M16.
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
DOI: 10.3109/00498254.2014.880201
Structure identification of mosapride metabolites 737
X. Sun et al.
Xenobiotica, 2014; 44(8): 734–742
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
738
Figure 3. MRM chromatograms of mosapride and its 16 metabolites in (A) blank human matrix; (B) human samples after oral administration of 10 mg of mosapride citrate dispersible tablets by UPLC–ESI–MS/MS. The MRM transitions of mosapride and its metabolites were listed at the left side of each chromatogram.
Urine metabolites In human urine, 10 metabolites, M1-M3, M5-M7, M9 and M14-M16, were observed. Metabolite M1 and M2, eluted at 1.69 and 2.34 min, respectively, showed the same protonated molecular ion [M + H]+ at m/z 330 and fragment ions at m/z 214 and 186, indicating that they were isomers formed by hydroxylation at C-3 or C-6 position of benzamide moiety (Figure 2A and B). Their [M + H]+ molecular ions were 108 Da less than that of hydroxyl mosapride, 438, indicative of a loss of 4-fluorobenzyl moiety. Therefore, M1 and M2 were proposed as des-p-fluorobenzyl hydroxymosapride. It was reported that 3-hydroxyl
des-p-fluorobenzyl mosapride was isolated and purified from rat urine after oral administration of mosapride (Matsumoto et al., 1993b). Due to the lack of reference standard, we could not define the exact hydroxylation positions of M1 and M2. Metabolite M3, eluted at 3.20 min, showed a protonated molecular ion [M + H]+ at m/z 314 which was 108 Da less than mosapride and formed by a loss of 4-fluorobenzyl moiety. Its product ion spectrum and fragmentation patterns are shown in Figure 2(C). M3 was identified as desp-fluorobenzyl mosapride by confirmation with reference standard. It has been reported to be the only known major metabolite in human urine and plasma (Matsumoto et al., 1993b).
Structure identification of mosapride metabolites
DOI: 10.3109/00498254.2014.880201
739
Table 1. Chromatographic retention times, mass spectrometric data and relative abundance of mosapride and its metabolites in male healthy volunteers after an oral administration of 10 mg of mosapride citrate dispersible tablets. Precursor ion [M + H]+
Product ion [M + H]+
Rt (min)
MS/MS data (% base peak)
M0 M1 M2 M3 M4 M5 M6
422 330 330 314 584 570 614
198 214 214 198 360 225 214
8.12 1.69 2.34 3.20 3.25 3.98 4.02
M7 M8 M9 M10 M11 M12 M13 M14 M15 M16
614 502 215 598 328 394 438 438 396 438
214 198 198 374 198 225 214 214 198 198
4.29 4.68 4.69 5.09 5.39 5.85 6.19 6.28 7.26 9.41
198 (100), 170 (43), 109 (3) 214 (100), 186 (10) 214 (100), 186 (10) 198 (100), 170 (4) 360 (100), 198 (43), 170 (2) 394 (21), 278 (6), 225 (100), 208 (6), 170 (19), 109 (3) 438 (7), 422 (3), 390 (9), 225 (9), 214 (100), 198 (38), 186 (15), 170 (4) 390 (8), 214 (100), 186 (13) 422 (20), 278 (12), 225 (7), 198 (100), 170 (5), 109 (4) 198 (100), 187 (5), 170 (5) 374 (100), 198 (75), 170 (3) 198 (100), 170 (5) 225 (100), 208 (25), 170 (73), 109 (38) 214 (100), 186 (9), 109 (4) 214 (100), 186 (14) 378 (5), 271 (9), 215 (2), 198 (100), 181 (3), 170 (4) 329 (2), 280 (66), 252 (5), 215 (19), 198 (100), 170 (14), 159 (5), 109 (3)
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
Metabolites
Relative abundance (%) Urine
Feces
Plasma
0.14 0.01 0.02 98.40 – 0.02 0.01
17.29 – 0.58 40.67 0.02 – –
55.56 – – 13.33 – – –
– 1.45 0.05 0.05 2.39 0.58 1.76 2.37 15.79 2.35
– – – – – – – – 31.11 –
0.005 – 0.003 – – – – 0.06 0.53 0.81
–: Not found; Rt: retention time.
Metabolite M5, eluted at 3.98 min, gave [M + H]+ at m/z 570 with MS/MS spectrum and fragmentation patterns shown in Figure 2(E). The ion at m/z 394 was 176 Da less than the protonated molecular ion, indicative of glucuronide conjugation. The fragment ion at m/z 394 was 28 Da less than that of mosapride, suggesting a loss of C2H4 via O-dealkylation to form 2-phenolic hydroxyl group. It further cleaved at benzamide C–N bond and gave rise to the ions at m/z 170 and 225. The ion at m/z 225 further lost a neutral molecular of NH3 to produce the ion at m/z 208. The ion at m/z 109 was the unchanged 4-fluorobenzyl moiety. Thus, M5 was tentatively assumed to be glucuronide conjugate of O-deethyl mosapride. Metabolite M6 and M7, eluted at 4.02 min and 4.29 min, respectively, both gave [M + H]+ at m/z 614, 192 Da more than that of mosapride, which indicated that they were isomers formed by hydroxylation and further glucuronide conjugation. The MS/MS spectra and fragmentation patterns are illustrated in Figure 2(F) and (G). The ion at m/z 390 was the glucuronide conjugated hydroxyl benzamide moiety. It further produced ions at m/z 214 and 186 by neutral losses of glucuronide and CO. Hence, M6 and M7 were assumed to be O-glucuronide conjugate of 3-hydroxymosapride or 6-hydroxymosapride. However, the exact phenolic hydroxyl positions could not be fully ascertained from these data. Metabolite M9, eluted at 4.69 min, gave [M + H]+ at m/z 215 with MS/MS spectrum and fragmentation patterns shown in Figure 2(I). All of the information allowed to postulating on a loss of NH3 to produce the fragment ion at m/z 198, which further lost a molecular of CO to form the one at m/z 170. The ion at m/z 187 was produced by a loss of C2H4 from the protonated molecular ion. Therefore, M9 was proposed to be 4-amino-5-chloro-2-ethoxybenzamide. Metabolites M14, eluted at 6.28 min, had a protonated molecular ion [M + H]+ at m/z 438 in the full scan MS. The MS/MS spectra showed fragment ions at m/z 214 and
186, which were consistent with the hydroxyl benzoyl moiety (Figure 2N). Hence, M14 was assumed to be 3-hydroxymosapride or 6-hydroxymosapride. Metabolite M15 was eluted at 7.26 min and gave [M + H]+ at m/z 396, which was 26 Da less than that of mosapride. The MS/MS spectrum and fragmentation patterns are shown in Figure 2(O). The fragment ion at m/z 378 was produced by a loss of water, which indicated aliphatic hydroxylation. Thus, M15 was assumed to be the morpholinyl ring-opened metabolite produced by a loss of C2H4 from the morpholinyl ring. The diagnostic ion at m/z 271 was the benzamide side formed by cleave at C–30 –NH bond. The fragment ion at m/z 181 was the 4-fluorobenzyl side formed by cleavage of C–7–NH bond along with a loss of water. M15 was finally identified as morpholine ring-opened mosapride by comparison with the reference standard. M16, with a retention time of 9.41 min, gave [M + H]+ at m/z 418. The presence of fragment ions at m/z 198, 170 and 109 suggested that the addition of O was neither in the benzoyl moiety nor in the 4-fluorobenzyl moiety. The ion at m/z 329 resulted from a loss of 4-fluorobenzyl moiety. It further produced the fragment ion at m/z 159 by cleavage at C–1 and C–7 bond (Figure 2P). Confirmed with reference standard, M16 was identified as mosapride N-oxide. Feces metabolites in human In human feces, 12 metabolites were observed. M2, M3, M9 and M14–M16 were found in human urine (as described above) as well as in human feces, while M4, M8 and M10M13 were only found in human feces. Metabolite M4, eluted at 3.25 min, had a protonated molecular ion [M + H]+ at m/z 584, a mass shift of 162 Da compared with the parent mosapride, indicative of glucoside conjugation. The fragment ion at m/z 360 was the glucoside conjugated benzoyl moiety formed by cleavage at the C–7–N bond (Figure 2D). All these data were consistent with
X. Sun et al.
Xenobiotica, 2014; 44(8): 734–742
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
740
Figure 4. Proposed metabolic profile of mosapride in humans.
glucoside conjugate of mosapride at 4-NH2. Thus, M9 was tentatively characterized as 4-glucoside-mosapride. Metabolite M8 was eluted at 4.68 min and had a protonated molecular ion [M + H]+ at m/z 502. The fragment ion at m/z 422 was formed by a neutral loss of sulfate. Resulted from cleavage of benzamide C–N bond, the fragment ions at m/z 278 and 225 were the 4-sulfate conjugated benzoyl moiety and the rest moiety, respectively (Figure 2H). With confirmation of the characteristic ions of parent mosapride at m/z 198, 170 and 109, M8 was elucidated as mosapride-4-sulfate.
Metabolite M10, eluted at 5.09 min, had a protonated molecular ion [M + H]+ at m/z 598, a mass shift of 176 Da compared with the parent mosapride. The fragment ion at m/z 374 was the glucuronide conjugated benzoyl moiety and it further produced the ions at m/z 198 and 170 by neutral losses of glucuronide and CO, respectively (Figure 2J). Thus, M10 was proposed as 4-glucuronide-mosapride. M11, eluted at 5.39 min, gave [M + H]+ at m/z 328 and fragment ions at m/z 198 and 170 in MS/MS spectra, which indicated unchanged benzoyl moiety. Its molecular weight
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
DOI: 10.3109/00498254.2014.880201
was 14 Da more than that of M3 (313), suggesting the formation of ketone next to the N atom of morpholine ring according to the drug metabolism rules (Figure 2K). Hence, M11 was presumed to be 50 -oxo-des-p-fluorobenzyl mosapride or 30 -oxo-des-p-fluorobenzyl mosapride. Metabolite M12 was eluted at 5.85 min and had a protonated molecular ion [M + H]+ at m/z 394. Its molecular weight was 28 Da less than that of mosapride, which indicated a loss of C2H4 via O-dealkylation to form 2-phenolic hydroxyl group. Cleavage of benzamide C–N bond produced the ions at m/z 170 and 225, representing the benzoyl moiety (C2H4 less than that of parent mosapride) and the rest morpholine ring moiety, respectively. The latter further lost a neutral molecular of NH3 to produce the ion at m/z 208 (Figure 2L). According to the above analyses, M12 was elucidated as Odeethyl mosapride. Metabolites M13, eluted at 6.19 min, had a protonated molecular ions [M + H]+ at m/z 438. The MS/MS spectrum showed fragment ions at m/z 214, 186 and 109, which were consistent with the hydroxyl benzoyl moiety and the unaltered 4-fluorobenzyl moiety, respectively (Figure 2M). Hence, M13 was assumed to be 3-hydroxymosapride or 6-hydroxymosapride. Plasma metabolites in human In human plasma, in addition to parent mosapride, only two phase I metabolites, M3 and M15, were detected and identified with reference standards. Elucidation of the possible metabolic pathway The proposed metabolic pathway of mosapride in human is shown in Figure 4. There were 10 metabolism pathways including 9 reported for the first time in human. Phase I metabolism pathways included: N-dealkylation (N-des-pfluorobenzyl and N-dealkylation of the long alkyl chain), Odeethylation, morpholinyl ring cleavage, hydroxylation, N-oxidation and oxidation to form ketone. Phase II metabolism pathways included: glucose, glucuronide and sulfated conjugation. A metabolite might go through more than two kinds of the above metabolism pathways. N-des-pfluorobenzyl, the only known metabolism pathway, was the most important metabolic pathway.
Discussion Previously, microbial transformation of mosapride by Caenorhabditis elegans AS 3.156 was studied in our group with M3, M15 and M16 isolated, purified and identified as pure compounds (Sun et al., 2009b). They were used as reference standards in the subsequent studies on the metabolism of mosapride in rat (Sun et al., 2009a) and in human described herein. Another 4 metabolites, M4, M8, M9 and M14 were also found in rat and C. elegans AS 3.156 model. The results obtained served well as an example of microbial transformation being an in vitro model to mimic mammalian drug metabolism. However, there were obvious differences in the metabolic profile of mosapride between human and C. elegans AS 3.156 model: (1) five phase I metabolites (M1, M2 and M11–M13) and four phase II glucuronide conjugates (M5–M7 and M10) were only found in human;
Structure identification of mosapride metabolites
741
(2) three phase II metabolites including two formylated and one acetylated metabolites were found in the microbial model only. There was only slight difference in the kinds of metabolites between rat and human. In addition to M10, the other 15 metabolites in human were found in rat. According to the Guidance for Industry Safety Testing of Drug Metabolites (U.S. Food and Drug Administration, 2008), our studies suggested that there was little qualitative species difference in the metabolism of mosapride between rat and human. However, based on the semi-quantitative analyses results of mosapride and its metabolites, there were some quantitative species differences in the metabolism of mosapride: (1) M3 was the major metabolite in human urine and feces, while M15 was the major metabolite in human plasma; (2) M3 was the major metabolite in rat urine, plasma and bile, while M12 was the major metabolite in rat feces; (3) M16 was found to be a major metabolite in rat bile (Sun et al., 2009a).
Conclusions The metabolic profile of mosapride in human was characterized by UPLC–ESI–MS/MS. A total of 16 metabolites were reported. To the best of our knowledge, 15 metabolites have not been reported previously in human. The structures of three phase I metabolites including two new ones (M15 and M16) were identified with reference standards and those of other 13 (seven phase I and six phase II metabolites) were elucidated. The phase I metabolites were mainly transformed by two main metabolism routes, dealkylation and morpholine ring cleavage, with dealkylation as the predominant metabolic pathway. The phase II metabolites were mainly formed by glucuronidation. The possible metabolic pathway of mosapride in human was proposed for the first time. There was little qualitative species difference in the metabolism of mosapride between rat and human. These results allow a better understanding of mosapride disposition in human.
Declaration of interest This study was supported by the National Key Scientific Project for New Drug Discovery and Development (No. 2009ZX09301-012). The authors report no other declaration of interest that could influence the results presented.
References Amir SY, Upendra AA, Murali P, et al. (2013). Identification of domperidone metabolites in plasma and urine of gastroparesis patients with LC–ESI-MS/MS. Xenobiotica 43:1073–83. Asai H, Udaka F, Hirano M, et al. (2005). Increased gastric motility during 5-HT4 agonist therapy reduces response fluctuations in Parkinson’s disease. Parkinsonism Relat Disord 11:499–502. Bailey J, Thew M, Balls M. (2013). An analysis of the use of dogs in predicting human toxicology and dog safety. Altern Lab Anim 41: 335–50. Clarke NJ, Rindgen D, Korfmacher WA, et al. (2001). Systematic LC/MS metabolite identification in drug discovery. Anal Chem 73: 430A–9A. Curran MP, Robinson DM. (2008). Mosapride in gastrointestinal disorders. Drugs 68:981–91. Dai HX, Wang MM, Li XR, et al. (2008). Structural elucidation of in vitro and in vivo metabolites of cryptotanshinone by HPLC– DAD–ESI–MSn. J Pharm Biomed Anal 50:27–34.
742
X. Sun et al.
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
Hsieh Y, Korfmacher WA. (2006). Increasing speed and throughput when using HPLC-MS/MS systems for drug metabolism and pharmacokinetic screening. Curr Drug Metab 7:479–89. Liu Z, Sakakibara R, Odaka T, et al. (2005). Mosapride citrate, a novel 5-HT4 agonist and partial 5-HT3 antagonist, ameliorates constipation in parkinsonian patients. Mov Disord 20:680–6. Matsumoto S, Tagawa M, Hatoyama T, et al. (1993a). Absorption, distribution, metabolism and excretion of [carbonyl-14C]mosapride citrate after repeated oral administration in rats. Arzneimittelforschung 43:1103–8. Matsumoto S, Yoshida K, Itogawa A, et al. (1993b). Metabolism of [carbonyl-14C]mosapride citrate after a single oral administration in rats, dogs and monkeys. Arzneimittelforschung 43:1095–102. Nassar AEF, Talaat RE, Kamel AM. (2006). The impact of recent innovations in the use of liquid chromatography-mass spectrometry in support of drug metabolism studies: are we all the way there yet? Curr Opin Drug Discov Devel 9:61–74. Oliveira EJ, Watson DG. (2000). Liquid chromatography-mass spectrometry in the study of the metabolism of drugs and other xenobiotics. Biomed Chromatogr 14:351–72. Pedraglio S, Rozio MG, Misiano P, et al. (2007). New perspectives in bio-analytical techniques for preclinical characterization of a drug
Xenobiotica, 2014; 44(8): 734–742
candidate: UPLC-MS/MS in in vitro metabolism and pharmacokinetic studies. J Pharm Biomed Anal 44:665–73. Sakashita M, Yamaguchi T, Miyazaki H, et al. (1993). Pharmacokinetics of the gastrokinetic agent mosapride citrate after single and multiple oral administrations in healthy subjects. Arzneimittelforschung 43: 867–72. Sun XH, Niu LL, Li XQ, et al. (2009a). Characterization of metabolic profile of mosapride citrate in rat and identification of two new metabolites: Mosapride N-oxide and morpholine ringopened mosapride by UPLC-ESI-MS/MS. J Pharm Biomed Anal 50: 27–34. Sun XH, Man F, Pang LY, et al. (2009b). Fungal biotransformation of mosapride by Cunninghamella elegans. J Mol Catal B: Enzym 59:82–9. U.S. Food and Drug Administration. (2008). Guidance for industry safety testing of drug metabolites. U.S. Food and Drug Administration, Center for Drug Evaluation and Research (CDER). Available from: http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryi nformation/guidances/ucm079266.pdf [last accessed February 2008]. Yoshida N. (1999). Pharmacological effects of the gastroprokinetic agent mosapride citrate. Nippon Yakurigaku Zasshi 113: 299–307.
RESEARCH ARTICLE
Structure identification and elucidation of mosapride metabolites in human urine, feces and plasma by ultra performance liquid chromatography-tandem mass spectrometry method Xiaohong Sun*,y, Longshan Zhaoy, Lili Niu, Feng Qin, Xiumei Lu, Zhili Xiong, and Famei Li
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
School of Pharmacy, Shenyang Pharmaceutical University, Shenyang, PR China
Abstract
Keywords
1. Mosapride citrate (mosapride) is a potent gastroprokinetic agent. The only previous study on mosapride metabolism in human reported one phase I oxidative metabolite, des-p-fluorobenzyl mosapride, in human plasma and urine using HPLC method. Our aim was to identify mosapride phase I and phase II metabolites in human urine, feces and plasma using UPLC-ESI-MS/MS. 2. A total of 16 metabolites were detected. To the best of our knowledge, 15 metabolites have not been reported previously in human. 3. Two new metabolites, morpholine ring-opened mosapride (M15) and mosapride N-oxide (M16), alone with one known major metabolite, des-p-fluorobenzyl mosapride (M3), were identified by comparison with the reference standards prepared by our group. The chemical structures of seven phase I and six phase II metabolites of mosapride were elucidated based on UPLC–MS/MS analyses. 4. There were two major phase I reactions, dealkylation and morpholine ring cleavage. Phase II reactions included glucuronide, glucose and sulfate conjugation. The comprehensive metabolic pathway of mosapride in human was proposed for the first time. 5. The metabolites in humans were compared with those in rats reported previously. In addition to M10, the other 15 metabolites in humans were also found in rats. This result suggested that there was little qualitative species difference in the metabolism of mosapride between rats and humans. 6. In all, 16 mosapride metabolites including 15 new metabolites were reported. These results allow a better understanding of mosapride disposition in human.
Human, metabolism, mosapride, UPLC-ESI-MS/MS
Introduction Mosapride citrate (mosapride), 4-amino-5-chloro-2-ethoxyN-[[4-(4-fluorobenzyl)-2-morpholinyl]methyl] benzamide citrate, is a selective serotonin 5-HT4 receptor agonist with no affinity for dopamine D2 receptor (Yoshida, 1999). It is a potent gastroprokinetic agent used to improve gastrointestinal (GI) symptoms in patients with gastroesophageal reflux disease and to ameliorate constipation and response fluctuations in Parkinsonian patients (Asai et al., 2005; Liu et al., 2005). Due to little arrhythmic effect associated with the QT
*Current address: National Center for Plant Gene Research (Beijing), Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing 100101, China. yThese authors contributed equally to the article. Address for correspondence: Longshan Zhao and Famei Li, School of Pharmacy, Shenyang Pharmaceutical University, 103 Wenhua Road, Shenyang 110016, PR China. Tel: +86 24 23986289. Fax: +86 24 23986289. E-mail: [email protected] (L. Zhao); fameili@163. com (F. Li)
History Received 18 November 2013 Revised 31 December 2013 Accepted 1 January 2014 Published online 13 January 2014
prolongation, mosapride has been used increasingly in Japan and some other countries (Curran & Robinson, 2008). Few attempts have been made to elucidate the exact metabolic profile of mosapride in human, although it has been marketed for about 20 years (Matsumoto et al., 1993a,b; Sakashita et al., 1993). The first study can be traced to Matsumoto et al. (1993b), where mosapride metabolism in rat was reported using radioisotope labeling-TLC assay. Although it was sensitive, the selectivity of the method was poor and it could not give any information on metabolite structures. Four phase I metabolites were isolated from rat urine and identified as des-p-fluorobenzyl mosapride, 50 -oxo-des-p-fluorobenzyl mosapride, 3-hydroxy des-p-fluorobenzyl and 3-hydroxy 50 -oxo-des-p-fluorobenzyl mosapride. An HPLC method was used for the metabolism study of mosapride in human and only one phase I metabolite, des-p-fluorobenzyl mosapride, was identified in plasma and urine (Sakashita et al., 1993). The advanced tandem mass spectrometry (MS/MS) has become an indispensable tool in metabolite detection and identification (Amir et al., 2013; Clarke et al., 2001; Hsieh & Korfmacher, 2006; Nassar et al., 2006; Oliveira & Watson, 2000; Pedraglio et al., 2007) and it
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
DOI: 10.3109/00498254.2014.880201
can be utilized for improved mosapride metabolism study. Previously in our group, UPLC-MS/MS method was used to study mosapride metabolism in rat before initiating its human metabolism described herein. A total of 18 metabolites were observed in rat, with the chemical structures of two new metabolites identified and at least 13 new ones proposed (Sun et al., 2009a). According to the Guidance for Industry Safety Testing of Drug Metabolites (U.S. Food and Drug Administration, 2008), it is encouraged that the identification of differences in drug metabolism between animal test species and humans as early as possible. However, there is no study on the overall metabolic profile of mosapride in human. Drug metabolism can influence a drug’s distribution, rate or route of excretion and production of new and possibly active or toxic species (Bailey et al., 2013). Characterizing the metabolic profile of mosapride is critical in understanding its pharmacokinetics as well as its adverse effect profile by identifying potential reactive intermediates or adducts. Hence, our aim is to identify and elucidate mosapride phase I and phase II metabolites in human with UPLC–MS/MS method, characterize possible metabolic pathways and compare the metabolites in human with those previously reported in rat (Sun et al., 2009a).
Experimental procedures Chemicals The reference standard of mosapride citrate (Figure 1, 99.2% purity) was purchased from the National Institutes for Food and Drug Control (Beijing, China). Mosapride citrate tablets, containing 5 mg of mosapride, was supplied by Dainippon ¯ saka-shi, Osaka-fu, Japan). Sumitomo Pharma Co., Ltd. (O
Structure identification of mosapride metabolites
735
Des-p-fluorobenzyl mosapride, mosapride N-oxide and morpholine ring-opened mosapride were isolated and purified from the preparative-scale microbial transformation of mosapride and identified as pure compounds by UV, MS, NMR with the purity above 98% by our laboratory (Sun et al., 2009b). Methanol, acetonitrile and formic acid were of HPLC grade and purchased from Dikma (Richmond Hill, USA). All other chemicals were of analytical grade. Apparatus and UPLC–MS/MS conditions UPLC–MS/MS system consisted of an ACQUITYTM UPLC system (Milford, MA) and Waters MicromassÕ Quattro microTM API mass spectrometer (Manchester, UK) equipped with an ESI source. The chromatographic separation was achieved on an ACQUITY UPLCTM BEH C18 column (100 mm 2.1 mm; i.d. 1.7 mm; Waters Corp., Milford, MA) with the column temperature set at 40 C. The flow rate was 0.25 ml/min with a linear gradient running from 84% to 78% A (solvent A, 0.2% formic acid aqueous; solvent B, acetonitrile) in 4 min, 78% to 76% A in the next 8 min, then to 10% A for 1 min, and returned to initial condition lasting 2 min for reequilibration. The total run time per sample was 15 min. As for the MS performed in the positive ESI mode, the optimal MS parameters optimized using mosapride standard were as follows: capillary voltage 3.0 kV, cone voltage 30 kV, source temperature 100 C and desolvation temperature 350 C. Nitrogen was used as the desolvation and cone gas with a flow rate of 500 and 50 l/h, respectively. For MS/MS analyses performed in product scan and MRM modes, argon was used as the collision gas set at 2.5 e–3 mbar, and the
Figure 1. (A) Full scan MS spectrum of mosapride; (B) MS/MS product ion spectrum of mosapride; (C) MRM chromatogram of mosapride by UPLC– MS/MS analysis, the monitored MRM transition was m/z 422 ! 198; (D) Predominant fragmentation patterns of mosapride under ESI–MS/MS analysis.
736
X. Sun et al.
collision energy ranged from 18 to 40 eV for all metabolites. Data was acquired and processed using MassLynxTM NT 4.1 software (Waters Corp., Milford, MA). Human subjects and sample collection
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
Healthy volunteers (25–28 years old), who gave their signed written informed consent to the study, fasted overnight and were orally given 10 mg mosapride citrated dispersible tablet (two tablets). No food was allowed for at least 2 h after drug administration. Urine samples were collected before (0 h) and at 6, 12, 24 and 30 h time intervals. Feces samples were collected before (0 h) and at 24 h post-dosing. Blood samples were collected and plasma separated before (0 h) and at 0.13, 0.25, 0.33, 0.42, 0.50, 0.75, 1.0, 1.5, 2.0, 3.5, 5.0, 7.0 and 10.0 h post-dosing. All the biological samples were collected into silicone coated glass tubes and stored at 20 C until analysis. Sample preparation Oasis MCXÕ mixed-mode cation-exchange cartridges (1 ml volume, Waters Corp.) were conditioned with 1 ml of methanol, 1 ml of water, each twice. Plasma and urine samples were centrifuged at 2000 g for 10 min, the supernatant were removed for use. The samples were filtered through 0.45 mm membrane and 1 ml of plasma and urine samples were loaded onto the SPE cartridges. Then, the SPE cartridges were washed with 1 ml of 1% formic acid aqueous, 1 ml of methanol and eluted with 1 ml of methanol containing 5% ammonium hydroxide. The eluate was evaporated to dryness under a gentle stream of nitrogen at 40 C and the residue was dissolved in 200 ml of the mixture of acetonitrile – 0.2% formic acid aqueous (3:7, v/v). A small aliquot of 10 ml of the solution was injected into the UPLC–MS/MS system. Feces samples (0.5 g) were weighted and supplemented with an appropriate volume (6 ml/g) of the mixture of methanol-water (8:2, v/v). After ultrasonic extraction for 20 min, the samples were centrifuged at 2000 g for 10 min. The supernatant was transferred to a clean tube and 6 ml of ethyl acetate was added. After vortexing and centrifugation at 2000 g for 10 min, the supernatant was transferred to a clean tube and dried under a flow of nitrogen at 35 C. The residue was reconstituted in 300 ml of methanol. After filtered through 0.45 mm membrane, an aliquot of 10 ml was injected into the UPLC–MS/MS system for analysis.
Results Fragmentation patterns of mosapride The fragmentation patterns of mosapride served as templates for interpreting the structures of the metabolites. Under ESI in positive ion mode, mosapride easily formed the protonated molecular ion [M + H]+ at m/z 422 in full scan MS with the isotopic peak obserbed at m/z 424 for the 37Cl parent ion (Figure 1A). The product ion spectrum of the protonated molecular ion of mosapride is shown in Figure 1(B). Mosapride was eluted at 8.12 min under MRM mode (Figure 1C). Fragmentation of [M + H]+ of mosapride led to three product ions and the fragmentation patterns
Xenobiotica, 2014; 44(8): 734–742
are illustrated in Figure 1(D). The diagnostic fragment ions at m/z 198 and 170 were characteristic of the benzoyl part, while the one at m/z 109 was characteristic of the p-fluorobenzyl part. These characteristic product ions were the sound bases to identify the metabolites of mosapride in human. Identification and structure elucidation of metabolites In order to find the metabolites, possible structures of metabolites were speculated according to the metabolism rule of drugs firstly. Great attention had been paid to the 18 metabolites of mosapride characterized in rat (Sun et al., 2009a) to facilitate the study on metabolism of mosapride in human (urine, feces and plasma). Then, the TICs of samples after administration were compared with those of the blank samples so as to find the protonated molecular ion of potential metabolites. After that, MS/MS product ion spectra of [M + H]+ of metabolites were obtained and the fragmentation patterns were characterized (Figure 2). Finally, possible metabolites were detected by UPLC–MS/MS under MRM mode. Representative MRM chromatograms of mosapride and its 16 metabolites in blank human matrix (including urine, feces and plasma) and human matrix samples after oral administration of 10 mg of mosapride citrate dispersible tablets are shown in Figure 3. The chromatographic retention times, MS/MS data and relative abundance of mosapride and the metabolites in humans are summarized in Table 1. Relative abundance was semi-quantitative. It was obtained by dividing the peak area of each analyte by the sum of peak area of mosapride and 16 metabolites in every matrix (Dai et al., 2008; Sun et al., 2009a). It is not possible to compare absolute formation of the different metabolites because not all synthetic metabolite standards are available to compensate for the differences in metabolite ionization efficiency. Based on the method mentioned above, mosapride and 16 metabolites (M1–M16, according to the retention time) were found in human matrix (including urine, feces and plasma) after oral administration of mosapirde citrate dispersible tablets. By comparison with reference standards obtained by our group (Sun et al., 2009b), the chromatographic and mass spectrometric data of the protonated molecular ions at m/z 422 (M0), 314 (M3), 396 (M15) and 438 (M16) with retention times at 8.12, 3.20, 7.26 and 9.14 min were the same with those of mosapride, desp-fluorobenzyl mosapride, mosapride N-oxide and morpholine ring-opened mosapride reference standards, respectively. Therefore, M0 was confirmed as the unchanged parent drug, while M3, M15 and M16 were identified as desp-fluorobenzyl mosapride, mosapride N-oxide and morpholine ring-opened mosapride. Structure elucidation of other 13 metabolites was through comparing the changes in retention times, observed mass (DM) and mass spectral patterns of product ions of metabolites with those of mosapride. To the best of our knowledge, M3 was the only known metabolite of mosapride in human (plasma and urine) and the other 15 metabolites were reported in human for the first time. The structure identification and elucidation of metabolites is explained in detail below.
Figure 2. MS/MS product ion spectra and corresponding fragmentation patterns of mosapride metabolites in human. (A) M1; (B) M2; (C) M3; (D) M4; (E) M5; (F) M6; (G) M7; (H) M8; (I) M9; (J) M10; (K) M11; (L) M12; (M) M13; (N) M14; (O) M15; (P) M16.
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
DOI: 10.3109/00498254.2014.880201
Structure identification of mosapride metabolites 737
X. Sun et al.
Xenobiotica, 2014; 44(8): 734–742
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
738
Figure 3. MRM chromatograms of mosapride and its 16 metabolites in (A) blank human matrix; (B) human samples after oral administration of 10 mg of mosapride citrate dispersible tablets by UPLC–ESI–MS/MS. The MRM transitions of mosapride and its metabolites were listed at the left side of each chromatogram.
Urine metabolites In human urine, 10 metabolites, M1-M3, M5-M7, M9 and M14-M16, were observed. Metabolite M1 and M2, eluted at 1.69 and 2.34 min, respectively, showed the same protonated molecular ion [M + H]+ at m/z 330 and fragment ions at m/z 214 and 186, indicating that they were isomers formed by hydroxylation at C-3 or C-6 position of benzamide moiety (Figure 2A and B). Their [M + H]+ molecular ions were 108 Da less than that of hydroxyl mosapride, 438, indicative of a loss of 4-fluorobenzyl moiety. Therefore, M1 and M2 were proposed as des-p-fluorobenzyl hydroxymosapride. It was reported that 3-hydroxyl
des-p-fluorobenzyl mosapride was isolated and purified from rat urine after oral administration of mosapride (Matsumoto et al., 1993b). Due to the lack of reference standard, we could not define the exact hydroxylation positions of M1 and M2. Metabolite M3, eluted at 3.20 min, showed a protonated molecular ion [M + H]+ at m/z 314 which was 108 Da less than mosapride and formed by a loss of 4-fluorobenzyl moiety. Its product ion spectrum and fragmentation patterns are shown in Figure 2(C). M3 was identified as desp-fluorobenzyl mosapride by confirmation with reference standard. It has been reported to be the only known major metabolite in human urine and plasma (Matsumoto et al., 1993b).
Structure identification of mosapride metabolites
DOI: 10.3109/00498254.2014.880201
739
Table 1. Chromatographic retention times, mass spectrometric data and relative abundance of mosapride and its metabolites in male healthy volunteers after an oral administration of 10 mg of mosapride citrate dispersible tablets. Precursor ion [M + H]+
Product ion [M + H]+
Rt (min)
MS/MS data (% base peak)
M0 M1 M2 M3 M4 M5 M6
422 330 330 314 584 570 614
198 214 214 198 360 225 214
8.12 1.69 2.34 3.20 3.25 3.98 4.02
M7 M8 M9 M10 M11 M12 M13 M14 M15 M16
614 502 215 598 328 394 438 438 396 438
214 198 198 374 198 225 214 214 198 198
4.29 4.68 4.69 5.09 5.39 5.85 6.19 6.28 7.26 9.41
198 (100), 170 (43), 109 (3) 214 (100), 186 (10) 214 (100), 186 (10) 198 (100), 170 (4) 360 (100), 198 (43), 170 (2) 394 (21), 278 (6), 225 (100), 208 (6), 170 (19), 109 (3) 438 (7), 422 (3), 390 (9), 225 (9), 214 (100), 198 (38), 186 (15), 170 (4) 390 (8), 214 (100), 186 (13) 422 (20), 278 (12), 225 (7), 198 (100), 170 (5), 109 (4) 198 (100), 187 (5), 170 (5) 374 (100), 198 (75), 170 (3) 198 (100), 170 (5) 225 (100), 208 (25), 170 (73), 109 (38) 214 (100), 186 (9), 109 (4) 214 (100), 186 (14) 378 (5), 271 (9), 215 (2), 198 (100), 181 (3), 170 (4) 329 (2), 280 (66), 252 (5), 215 (19), 198 (100), 170 (14), 159 (5), 109 (3)
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
Metabolites
Relative abundance (%) Urine
Feces
Plasma
0.14 0.01 0.02 98.40 – 0.02 0.01
17.29 – 0.58 40.67 0.02 – –
55.56 – – 13.33 – – –
– 1.45 0.05 0.05 2.39 0.58 1.76 2.37 15.79 2.35
– – – – – – – – 31.11 –
0.005 – 0.003 – – – – 0.06 0.53 0.81
–: Not found; Rt: retention time.
Metabolite M5, eluted at 3.98 min, gave [M + H]+ at m/z 570 with MS/MS spectrum and fragmentation patterns shown in Figure 2(E). The ion at m/z 394 was 176 Da less than the protonated molecular ion, indicative of glucuronide conjugation. The fragment ion at m/z 394 was 28 Da less than that of mosapride, suggesting a loss of C2H4 via O-dealkylation to form 2-phenolic hydroxyl group. It further cleaved at benzamide C–N bond and gave rise to the ions at m/z 170 and 225. The ion at m/z 225 further lost a neutral molecular of NH3 to produce the ion at m/z 208. The ion at m/z 109 was the unchanged 4-fluorobenzyl moiety. Thus, M5 was tentatively assumed to be glucuronide conjugate of O-deethyl mosapride. Metabolite M6 and M7, eluted at 4.02 min and 4.29 min, respectively, both gave [M + H]+ at m/z 614, 192 Da more than that of mosapride, which indicated that they were isomers formed by hydroxylation and further glucuronide conjugation. The MS/MS spectra and fragmentation patterns are illustrated in Figure 2(F) and (G). The ion at m/z 390 was the glucuronide conjugated hydroxyl benzamide moiety. It further produced ions at m/z 214 and 186 by neutral losses of glucuronide and CO. Hence, M6 and M7 were assumed to be O-glucuronide conjugate of 3-hydroxymosapride or 6-hydroxymosapride. However, the exact phenolic hydroxyl positions could not be fully ascertained from these data. Metabolite M9, eluted at 4.69 min, gave [M + H]+ at m/z 215 with MS/MS spectrum and fragmentation patterns shown in Figure 2(I). All of the information allowed to postulating on a loss of NH3 to produce the fragment ion at m/z 198, which further lost a molecular of CO to form the one at m/z 170. The ion at m/z 187 was produced by a loss of C2H4 from the protonated molecular ion. Therefore, M9 was proposed to be 4-amino-5-chloro-2-ethoxybenzamide. Metabolites M14, eluted at 6.28 min, had a protonated molecular ion [M + H]+ at m/z 438 in the full scan MS. The MS/MS spectra showed fragment ions at m/z 214 and
186, which were consistent with the hydroxyl benzoyl moiety (Figure 2N). Hence, M14 was assumed to be 3-hydroxymosapride or 6-hydroxymosapride. Metabolite M15 was eluted at 7.26 min and gave [M + H]+ at m/z 396, which was 26 Da less than that of mosapride. The MS/MS spectrum and fragmentation patterns are shown in Figure 2(O). The fragment ion at m/z 378 was produced by a loss of water, which indicated aliphatic hydroxylation. Thus, M15 was assumed to be the morpholinyl ring-opened metabolite produced by a loss of C2H4 from the morpholinyl ring. The diagnostic ion at m/z 271 was the benzamide side formed by cleave at C–30 –NH bond. The fragment ion at m/z 181 was the 4-fluorobenzyl side formed by cleavage of C–7–NH bond along with a loss of water. M15 was finally identified as morpholine ring-opened mosapride by comparison with the reference standard. M16, with a retention time of 9.41 min, gave [M + H]+ at m/z 418. The presence of fragment ions at m/z 198, 170 and 109 suggested that the addition of O was neither in the benzoyl moiety nor in the 4-fluorobenzyl moiety. The ion at m/z 329 resulted from a loss of 4-fluorobenzyl moiety. It further produced the fragment ion at m/z 159 by cleavage at C–1 and C–7 bond (Figure 2P). Confirmed with reference standard, M16 was identified as mosapride N-oxide. Feces metabolites in human In human feces, 12 metabolites were observed. M2, M3, M9 and M14–M16 were found in human urine (as described above) as well as in human feces, while M4, M8 and M10M13 were only found in human feces. Metabolite M4, eluted at 3.25 min, had a protonated molecular ion [M + H]+ at m/z 584, a mass shift of 162 Da compared with the parent mosapride, indicative of glucoside conjugation. The fragment ion at m/z 360 was the glucoside conjugated benzoyl moiety formed by cleavage at the C–7–N bond (Figure 2D). All these data were consistent with
X. Sun et al.
Xenobiotica, 2014; 44(8): 734–742
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
740
Figure 4. Proposed metabolic profile of mosapride in humans.
glucoside conjugate of mosapride at 4-NH2. Thus, M9 was tentatively characterized as 4-glucoside-mosapride. Metabolite M8 was eluted at 4.68 min and had a protonated molecular ion [M + H]+ at m/z 502. The fragment ion at m/z 422 was formed by a neutral loss of sulfate. Resulted from cleavage of benzamide C–N bond, the fragment ions at m/z 278 and 225 were the 4-sulfate conjugated benzoyl moiety and the rest moiety, respectively (Figure 2H). With confirmation of the characteristic ions of parent mosapride at m/z 198, 170 and 109, M8 was elucidated as mosapride-4-sulfate.
Metabolite M10, eluted at 5.09 min, had a protonated molecular ion [M + H]+ at m/z 598, a mass shift of 176 Da compared with the parent mosapride. The fragment ion at m/z 374 was the glucuronide conjugated benzoyl moiety and it further produced the ions at m/z 198 and 170 by neutral losses of glucuronide and CO, respectively (Figure 2J). Thus, M10 was proposed as 4-glucuronide-mosapride. M11, eluted at 5.39 min, gave [M + H]+ at m/z 328 and fragment ions at m/z 198 and 170 in MS/MS spectra, which indicated unchanged benzoyl moiety. Its molecular weight
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
DOI: 10.3109/00498254.2014.880201
was 14 Da more than that of M3 (313), suggesting the formation of ketone next to the N atom of morpholine ring according to the drug metabolism rules (Figure 2K). Hence, M11 was presumed to be 50 -oxo-des-p-fluorobenzyl mosapride or 30 -oxo-des-p-fluorobenzyl mosapride. Metabolite M12 was eluted at 5.85 min and had a protonated molecular ion [M + H]+ at m/z 394. Its molecular weight was 28 Da less than that of mosapride, which indicated a loss of C2H4 via O-dealkylation to form 2-phenolic hydroxyl group. Cleavage of benzamide C–N bond produced the ions at m/z 170 and 225, representing the benzoyl moiety (C2H4 less than that of parent mosapride) and the rest morpholine ring moiety, respectively. The latter further lost a neutral molecular of NH3 to produce the ion at m/z 208 (Figure 2L). According to the above analyses, M12 was elucidated as Odeethyl mosapride. Metabolites M13, eluted at 6.19 min, had a protonated molecular ions [M + H]+ at m/z 438. The MS/MS spectrum showed fragment ions at m/z 214, 186 and 109, which were consistent with the hydroxyl benzoyl moiety and the unaltered 4-fluorobenzyl moiety, respectively (Figure 2M). Hence, M13 was assumed to be 3-hydroxymosapride or 6-hydroxymosapride. Plasma metabolites in human In human plasma, in addition to parent mosapride, only two phase I metabolites, M3 and M15, were detected and identified with reference standards. Elucidation of the possible metabolic pathway The proposed metabolic pathway of mosapride in human is shown in Figure 4. There were 10 metabolism pathways including 9 reported for the first time in human. Phase I metabolism pathways included: N-dealkylation (N-des-pfluorobenzyl and N-dealkylation of the long alkyl chain), Odeethylation, morpholinyl ring cleavage, hydroxylation, N-oxidation and oxidation to form ketone. Phase II metabolism pathways included: glucose, glucuronide and sulfated conjugation. A metabolite might go through more than two kinds of the above metabolism pathways. N-des-pfluorobenzyl, the only known metabolism pathway, was the most important metabolic pathway.
Discussion Previously, microbial transformation of mosapride by Caenorhabditis elegans AS 3.156 was studied in our group with M3, M15 and M16 isolated, purified and identified as pure compounds (Sun et al., 2009b). They were used as reference standards in the subsequent studies on the metabolism of mosapride in rat (Sun et al., 2009a) and in human described herein. Another 4 metabolites, M4, M8, M9 and M14 were also found in rat and C. elegans AS 3.156 model. The results obtained served well as an example of microbial transformation being an in vitro model to mimic mammalian drug metabolism. However, there were obvious differences in the metabolic profile of mosapride between human and C. elegans AS 3.156 model: (1) five phase I metabolites (M1, M2 and M11–M13) and four phase II glucuronide conjugates (M5–M7 and M10) were only found in human;
Structure identification of mosapride metabolites
741
(2) three phase II metabolites including two formylated and one acetylated metabolites were found in the microbial model only. There was only slight difference in the kinds of metabolites between rat and human. In addition to M10, the other 15 metabolites in human were found in rat. According to the Guidance for Industry Safety Testing of Drug Metabolites (U.S. Food and Drug Administration, 2008), our studies suggested that there was little qualitative species difference in the metabolism of mosapride between rat and human. However, based on the semi-quantitative analyses results of mosapride and its metabolites, there were some quantitative species differences in the metabolism of mosapride: (1) M3 was the major metabolite in human urine and feces, while M15 was the major metabolite in human plasma; (2) M3 was the major metabolite in rat urine, plasma and bile, while M12 was the major metabolite in rat feces; (3) M16 was found to be a major metabolite in rat bile (Sun et al., 2009a).
Conclusions The metabolic profile of mosapride in human was characterized by UPLC–ESI–MS/MS. A total of 16 metabolites were reported. To the best of our knowledge, 15 metabolites have not been reported previously in human. The structures of three phase I metabolites including two new ones (M15 and M16) were identified with reference standards and those of other 13 (seven phase I and six phase II metabolites) were elucidated. The phase I metabolites were mainly transformed by two main metabolism routes, dealkylation and morpholine ring cleavage, with dealkylation as the predominant metabolic pathway. The phase II metabolites were mainly formed by glucuronidation. The possible metabolic pathway of mosapride in human was proposed for the first time. There was little qualitative species difference in the metabolism of mosapride between rat and human. These results allow a better understanding of mosapride disposition in human.
Declaration of interest This study was supported by the National Key Scientific Project for New Drug Discovery and Development (No. 2009ZX09301-012). The authors report no other declaration of interest that could influence the results presented.
References Amir SY, Upendra AA, Murali P, et al. (2013). Identification of domperidone metabolites in plasma and urine of gastroparesis patients with LC–ESI-MS/MS. Xenobiotica 43:1073–83. Asai H, Udaka F, Hirano M, et al. (2005). Increased gastric motility during 5-HT4 agonist therapy reduces response fluctuations in Parkinson’s disease. Parkinsonism Relat Disord 11:499–502. Bailey J, Thew M, Balls M. (2013). An analysis of the use of dogs in predicting human toxicology and dog safety. Altern Lab Anim 41: 335–50. Clarke NJ, Rindgen D, Korfmacher WA, et al. (2001). Systematic LC/MS metabolite identification in drug discovery. Anal Chem 73: 430A–9A. Curran MP, Robinson DM. (2008). Mosapride in gastrointestinal disorders. Drugs 68:981–91. Dai HX, Wang MM, Li XR, et al. (2008). Structural elucidation of in vitro and in vivo metabolites of cryptotanshinone by HPLC– DAD–ESI–MSn. J Pharm Biomed Anal 50:27–34.
742
X. Sun et al.
Xenobiotica Downloaded from informahealthcare.com by Korea University on 01/01/15 For personal use only.
Hsieh Y, Korfmacher WA. (2006). Increasing speed and throughput when using HPLC-MS/MS systems for drug metabolism and pharmacokinetic screening. Curr Drug Metab 7:479–89. Liu Z, Sakakibara R, Odaka T, et al. (2005). Mosapride citrate, a novel 5-HT4 agonist and partial 5-HT3 antagonist, ameliorates constipation in parkinsonian patients. Mov Disord 20:680–6. Matsumoto S, Tagawa M, Hatoyama T, et al. (1993a). Absorption, distribution, metabolism and excretion of [carbonyl-14C]mosapride citrate after repeated oral administration in rats. Arzneimittelforschung 43:1103–8. Matsumoto S, Yoshida K, Itogawa A, et al. (1993b). Metabolism of [carbonyl-14C]mosapride citrate after a single oral administration in rats, dogs and monkeys. Arzneimittelforschung 43:1095–102. Nassar AEF, Talaat RE, Kamel AM. (2006). The impact of recent innovations in the use of liquid chromatography-mass spectrometry in support of drug metabolism studies: are we all the way there yet? Curr Opin Drug Discov Devel 9:61–74. Oliveira EJ, Watson DG. (2000). Liquid chromatography-mass spectrometry in the study of the metabolism of drugs and other xenobiotics. Biomed Chromatogr 14:351–72. Pedraglio S, Rozio MG, Misiano P, et al. (2007). New perspectives in bio-analytical techniques for preclinical characterization of a drug
Xenobiotica, 2014; 44(8): 734–742
candidate: UPLC-MS/MS in in vitro metabolism and pharmacokinetic studies. J Pharm Biomed Anal 44:665–73. Sakashita M, Yamaguchi T, Miyazaki H, et al. (1993). Pharmacokinetics of the gastrokinetic agent mosapride citrate after single and multiple oral administrations in healthy subjects. Arzneimittelforschung 43: 867–72. Sun XH, Niu LL, Li XQ, et al. (2009a). Characterization of metabolic profile of mosapride citrate in rat and identification of two new metabolites: Mosapride N-oxide and morpholine ringopened mosapride by UPLC-ESI-MS/MS. J Pharm Biomed Anal 50: 27–34. Sun XH, Man F, Pang LY, et al. (2009b). Fungal biotransformation of mosapride by Cunninghamella elegans. J Mol Catal B: Enzym 59:82–9. U.S. Food and Drug Administration. (2008). Guidance for industry safety testing of drug metabolites. U.S. Food and Drug Administration, Center for Drug Evaluation and Research (CDER). Available from: http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryi nformation/guidances/ucm079266.pdf [last accessed February 2008]. Yoshida N. (1999). Pharmacological effects of the gastroprokinetic agent mosapride citrate. Nippon Yakurigaku Zasshi 113: 299–307.
