Glucuronidation of a sarpogrelate active metabolite is mediated by udp-glucuronosyltransferases 1a4, 1a9, and 2b4

Article abstract of DOI:10.1124/dmd.113.051862

Sarpogrelate is a selective serotonin 5-HT2A-receptor antagonist used to treat patients with peripheral arterial disease. This drug is rapidly hydrolyzed to its main metabolite (R,S)-1-[2-[2-(3-methoxyphenyl)ethyl]phenoxy]-3- (dimethylamino)-2-propanol (M-1), which is mainly excreted as a glucuronide conjugate. Sarpogrelate was also directly glucuronidated to an O-Acyl glucuronide and a Nglucuronide by UDP-glucuronosyltransferases (UGTs) in human liver microsomes (HLMs). Since M-1 is pharmacologically more active than sarpogrelate, we examined glucuronidation of this metabolite in HLMs and characterized the UGTs responsible for M-1 glucuronidation. Diastereomers of O-glucuronide (SMG1 and SMG3) and a N-glucuronide (SMG2) were identified by incubation of M-1 with HLMs in the presence of uridine 59-diphosphoglucuronic acid (UDPGA), and their structures were confirmed by nuclear magnetic resonance and mass spectrometry analyses. Two O-glucuronides were identified as chiral isomers: SMG1 as R-isomer and SMG3 as S-isomer. Using recombinant UGT enzymes, we determined that SMG1 and SMG3 were predominantly catalyzed by UGT1A9 and UGT2B4, respectively, whereas SMG2 was generated by UGT1A4. In addition, significant correlations were noted between the SMG1 formation rate and propofol glucuronidation (a marker reaction of UGT1A9; r = 0.6269, P < 0.0031), and between the SMG2 formation rate and trifluoperazine glucuronidation (a marker reaction of UGT1A4; r = 0.6623, P < 0.0015) in a panel of HLMs. Inhibition of SMG1, SMG2, and SMG3 formation by niflumic acid, hecogenin, and fluconazole further substantiated the involvement of UGT1A9, UGT1A4, and UGT2B4, respectively. These findings collectively indicate that UGT1A4, UGT1A9, and UGT2B4 are the major UGT isoforms responsible for glucuronidation of M-1, an active metabolite of sarpogrelate. Copyright

Full text of DOI:10.1124/dmd.113.051862

Supplemental material to this article can be found at:
http://dmd.aspetjournals.org/content/suppl/2013/05/23/dmd.113.051862.DC1.html
1521-009X/41/8/1529–1537$25.00
DRUG METABOLISM AND DISPOSITION
http://dx.doi.org/10.1124/dmd.113.051862
Drug Metab Dispos 41:1529–1537, August 2013
Copyright ª 2013 by The American Society for Pharmacology and Experimental Therapeutics
Glucuronidation of a Sarpogrelate Active Metabolite Is Mediated
by UDP-Glucuronosyltransferases 1A4, 1A9, and 2B4s
Hyo-Ji Kim, Eun Sook Jeong, Kyung-Ah Seo, Kye Jung Shin, Yeon Jae Choi, Su-Jun Lee,
Jong Lyul Ghim, Dong-Ryul Sohn, Jae-Gook Shin, and Dong-Hyun Kim
Department of Pharmacology and PharmacoGenomics Research Center, Inje University College of Medicine, Busan, South Korea
(H.-J.K., E.S.J., K.-A.S., S.-J.L., J.L.G., J.-G.S., and D.-H.K.); Integrated Research Institute of Pharmaceutical Sciences, College of
Pharmacy, The Catholic University of Korea, Bucheon, South Korea (K.J.S. and Y.J.C.); and Department of Clinical Pharmacology,
College of Medicine, Soonchunhyang University, Chonan, South Korea (D.-R.S.)
Received March 11, 2013; accepted May 23, 2013
ABSTRACT
Sarpogrelate is a selective serotonin 5-HT2A–receptor antagonist O-glucuronides were identified as chiral isomers: SMG1 as
used to treat patients with peripheral arterial disease. This drug
R-isomer and SMG3 as S-isomer. Using recombinant UGT
is rapidly hydrolyzed to its main metabolite (R,S)-1-[2-[2-(3– enzymes, we determined that SMG1 and SMG3 were predomi-
methoxyphenyl)ethyl]phenoxy]-3-(dimethylamino)-2-propanol (M-1), nantly catalyzed by UGT1A9 and UGT2B4, respectively, whereas
which is mainly excreted as a glucuronide conjugate. Sarpogrelate
was also directly glucuronidated to an O-acyl glucuronide and a N-
SMG2 was generated by UGT1A4. In addition, significant correla-
tions were noted between the SMG1 formation rate and propofol
glucuronide by UDP-glucuronosyltransferases (UGTs) in human glucuronidation (a marker reaction of UGT1A9; r = 0.6269, P <
liver microsomes (HLMs). Since M-1 is pharmacologically more 0.0031), and between the SMG2 formation rate and trifluoperazine
active than sarpogrelate, we examined glucuronidation of this glucuronidation (a marker reaction of UGT1A4; r = 0.6623, P <
metabolite in HLMs and characterized the UGTs responsible for 0.0015) in a panel of HLMs. Inhibition of SMG1, SMG2, and SMG3
M-1 glucuronidation. Diastereomers of O-glucuronide (SMG1 and formation by niflumic acid, hecogenin, and fluconazole further
SMG3) and a N-glucuronide (SMG2) were identified by incubation of substantiated the involvement of UGT1A9, UGT1A4, and UGT2B4,
M-1 with HLMs in the presence of uridine 59-diphosphoglucuronic respectively. These findings collectively indicate that UGT1A4,
acid (UDPGA), and their structures were confirmed by nuclear
magnetic resonance and mass spectrometry analyses. Two
UGT1A9, and UGT2B4 are the major UGT isoforms responsible for
glucuronidation of M-1, an active metabolite of sarpogrelate.
Introduction
arterial disease (Furukawa et al., 1991) and is equivalent to aspirin in
preventing recurrence of cerebral infarction with significantly fewer
bleeding events (Shinohara et al., 2008). Sarpogrelate is rapidly
metabolized to its active metabolite (M-1) via enzymatic hydrolysis
(Saini et al., 2004) and disappears from the plasma more rapidly in
comparison with the duration of its pharmacological effect. Although
plasma concentrations of M-1 are ,1/10 of those of sarpogrelate,
pharmacokinetic-pharmacodynamic modeling has demonstrated that
M-1 is a more effective inhibitor of platelet aggregation than
sarpogrelate (Shimizu et al., 1999). Sarpogrelate and its metabolites
are primarily excreted into the bile as glucuronide conjugates,
suggesting that glucuronidation is a critical step governing overall
clearance of sarpogrelate and M-1 in humans. However, the isoforms
involved in the glucuronidation of these compounds have not yet been
characterized.
In this study, we identified the human UGTs responsible for the
metabolism of sarpogrelate and M-1 into their glucuronide metabolites
using in vitro systems. The structures of metabolites were tentatively
characterized on the basis of mass and nuclear magnetic resonance
(NMR) spectrometry, and metabolic screening was performed with
a battery of recombinant human UGTs to identify the specific UGT
isoforms involved in N- and O-glucuronidation. Kinetic enzymatic
Glucuronidation is one of the major phase II conjugation
reactions catalyzed by a family of uridine 59-diphospho-(UDP)-
glucuronosyltransferases (UGTs), which are expressed primarily in
the liver and in extrahepatic tissues such as the intestine and kidney
(Liston et al., 2001). UGTs are involved in the metabolism of a large
number of drugs and endogenous substances by transferring glucuronic
acid from its cofactor uridine 59-diphosphoglucuronic acid (UDPGA) to
substrates, thereby transforming them into hydrophilic glucuronides, which
can be eliminated easily via biliary and renal routes (Guillemette, 2003).
Sarpogrelate, (R,S)-1-[2-[2-(3-methoxyphenyl)ethyl] phenoxy]-3
(dimethylamino)-2-propyl hydrogen succinate, is a competitive 5-
hydroxytryptamine (5-HT) receptor antagonist that inhibits responses
to 5-HT mediated by 5-HT2A receptors, including platelet aggregation
and vasoconstriction (Hara et al., 1991; Rashid et al., 2003; Nishihira
et al., 2006). This drug has been used to treat patients with peripheral
This work was supported by a National Research Foundation of Korea (NRF)
grant funded by the Korean Government (MEST) [No. R13-2007-023-00000-0].
dx.doi.org/10.1124/dmd.113.051862.
s This article has supplemental material available at dmd.aspetjournals.org.
ABBREVIATIONS: HLM, human liver microsome; HPLC, high-performance liquid chromatography; M-1, (R,S)-1-[2-[2-(3–methoxyphenyl)eth-
yl]phenoxy]-3-(dimethylamino)-2-propanol; MS/MS, tandem mass spectrometry; NMR, nuclear magnetic resonance; RLM, rat liver microsomes;
SG, sarpogrelate glucuronide; SMG, M-1 glucuronide; UDPGA, uridine 59-diphosphoglucuronic acid; UGT, UDP-glucuronosyltransferase.
1529

1530
Kim et al.
(R)-M-1 was also prepared using the same procedure except for the use of
(S)-epichlorohydrin.
studies were performed with human liver microsomes (HLMs) and
with those UGT enzymes that catalyze glucuronidation of sarpogrelate
active metabolites. Inhibition of UGT-glucuronidation activity and
correlations with known marker activities in HLMs were determined
using known high-affinity UGT substrates to facilitate identification of
the UGT isoforms involved in M-1 glucuronidation. In addition, the
feasibility of using M-1 as an isoform-specific probe substrate was
evaluated because UGT isoform-specific substrates are not well
characterized.
Preparation of Microsomes. HLMs were prepared from human liver tissues
(20 different donors) from patients undergoing partial hepatectomy for removal
of metastatic tumors at the Department of General Surgery, Busan Paik
Hospital (Busan, Korea). The samples were of nontumor-bearing parenchymal
tissue and were confirmed to be histopathologically normal. The liver tissues
and their clinical information were obtained from Inje Biobank at the Inje
University Busan Paik Hospital and were in accordance with the policies of the
institutional review board of the hospital. For the preparation of rat liver
microsomes, male Sprague-Dawley rats (6–7 weeks of age, 200–220 g body
weight) were obtained from Orient Bio (Seongnam, Korea) and were
acclimated for one week. The animal quarters were maintained at 23 6 3°C
and at 50% 6 10% relative humidity. A 12-hour light and dark cycle was used
with an intensity of 150–300 Lux. The use of animals was approved by
institutional animal care and use committees at the Inje University School of
Medicine. HLMs were individually prepared by differential centrifugation of
the liver homogenate, as described previously (Shin et al., 1999). Ten different
HLMs were pooled and used for the experiments. In the case of rats, liver
tissues from six rats were pooled for the preparation of liver microsomes.
Glucuronidation of Sarpogrelate and M-1 in HLMs or Recombinant
UGT Isoforms. The incubation mixtures consisted of 0.25 mg/ml HLMs or
Materials and Methods
Chemicals and Reagents. Racemic sarpogrelate and racemic M-1 (Fig. 1)
were kindly provided by Yuyu Pharma (Seoul, Korea). Alamethicin from
Trichoderma viride, UDPGA, niflumic acid, hecogenin, propofol, and
trifluoperzine were purchased from Sigma-Aldrich (St. Louis, MO). Flucona-
zole was obtained from DaeWoo Pharma (Seoul, Korea). Recombinant human
UGT isoforms (UGT1A1, 1A3, 1A4, 1A6, 1A7, 1A8, 1A9, 1A10, 2B4, 2B7,
2B15, and 2B17) were purchased from BD Gentest (Woburn, MA). All
solvents were of high-performance liquid chromatography (HPLC) grade and
were obtained from Fisher Scientific (Pittsburgh, PA).
Synthesis of M-1 Enantiomers. (S)-M-1 was prepared using the following recombinant UGTs, 25 mg/ml alamethicin, Tris-HCl buffer (50 mM, pH 7.5),
procedure. 2-[2-(3-Methoxyphenyl)ethyl]phenol (200 mg, 0.876 mmol) was 10 mM MgCl2, and substrate (sarpogrelate or M-1) in a total volume of 125 ml.
dissolved in 5 ml of ethanol and NaOH (42 mg, 1.051 mmol) was added. After
stirring for 20 minutes at room temperature, (R)-epichlorohydrin (0.2 ml, 2.628 After preincubation on ice for 15 minutes, the reaction was initiated by the
mmol) was added dropwise. The resulting mixture was stirred for 9 hours,
addition of 5 mM UDPGA and incubated for 20 minutes at 37°C in a shaking
solvent was evaporated, and then the residue was dissolved in ethylacetate. water bath. The reaction was terminated by the addition of 125-ml acetonitrile
The final volume of the organic solvents in the incubation mixture was 1% (v/v).
The resulting solution was washed with 1 N NaOH, dried over MgSO4, and
concentrated in vacuo. The residue was purified by column chromatography
containing 10 mM 7-hydroxycoumarin glucuronide as an internal standard and
centrifuged at 10,000g for 5 minutes at 4°C. Aliquots of the supernatant were
(n-hexane/ethylacetate = 10: 1) to afford (S)-2-[2-(2-(3-methoxy)phenylethyl) analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS)
phenoxy]methyloxirane (190 mg). This compound (34 mg, 0.119 mmol) was
dissolved in acetonitrile. LiClO4 (19.0 mg, 0.179 mmol) and dimethylamine
for the analysis of glucuronide conjugates of sarpogrelate and M-1.
LC-MS/MS Analysis of Sarpogrelate and M-1 Glucuronides. LC-MS/
hydrochloride (14.6 mg, 0.179 mmol) were added gradually and the reaction MS was performed on an API 4000 LC-MS/MS system (Applied Biosystems,
mixture was heated for 20 hours at 65°C. After solvent evaporation, the
Foster City, CA), coupled with an 1100 series HPLC system (Agilent, Santa
residue was extracted with ethylacetate, washed with saturated NaHCO3, Clara, CA). The separation was performed on a Synergi Fusion-RP 80A
dried over MgSO4, and concentrated. The residue was purified by column
column (150 Â 2.0 mm, 4 mm; Phenomenex, Torrance, CA) using a mobile
chromatography (ethylacetate) to afford 15 mg (38%) (S)-1-[2-[2-(3– phase of acetonitrile and 5 mM ammonium formate (pH 3.0) (40:60, v/v) at
methoxyphenyl)ethyl]phenoxy]-3-(dimethylamino)-2-propanol [(S)-M-1]. a flow rate of 0.2 ml/min. Electrospray ionization (ESI) was performed in the
Fig. 1. Postulated in vitro metabolic pathways of sarpogrelate and its active metabolite M-1 in human liver microsomes. An asterisk denotes a chiral carbon.

UGT Isoforms Involved in Sarpogrelate Metabolite
1531
positive ion mode with nitrogen as the nebulizing, turbo spray, and curtain Characterization of Hydrolysis Products of Sarpogrelate Glucuronides.
gases with the optimum values set at 50, 50, and 25 (arbitrary units). The turbo To characterize the hydrolysis products of sarpogrelate glucuronides, two SGs
ion spray interface was operated at 5000 V and 500°C. The multiple-reaction
were first isolated using a Synergi Fusion-RP 80A column (150 mm  2.0 mm,
monitoring mode using specific precursor/product ion transitions was employed i.d., 4 mm; Phenomenex). The mobile phase consisted of eluent (A), 0.1%
for quantification. Detection of the ions was performed by monitoring the m/z formic acid, and eluent (B), acetonitrile containing 0.1% formic acid at a flow
transitions of 606→430 for sarpogrelate glucuronide (SG), 506→330 for M-1 rate of 0.2 ml/min. The gradient program (B) was as follows: 30% (0–4
glucuronide (SMG), 339→163 for the internal standard, 7-hydroxycoumarin minutes), 38% (4–9 minutes), 30% (9–10 minutes), 30% (10–13 minutes). The
glucuronide. The collision energies were 30, 40, and 20 eV for SGs, SMGs, peaks were monitored by UV detector at 254 nm. The fractions containing SG1
and the internal standard, respectively. Peak areas for all compounds were or SG2 were collected, dried, and reconstituted in 500 mM Tris-HCl buffer (pH
integrated automatically using the Analyst software (v. 1.3; Applied Biosystems). 7.5). The samples were then kept at room temperature for 48 hours and
Isolation and NMR Analysis of M-1 Glucuronides. M-1 glucuronides
were isolated from large microsomal incubations for NMR analysis. The
incubation mixture (total volume 80 ml) consisted of 50 mM Tris-HCl buffer
analyzed by LC-MS/MS.
Chemical Inhibition. The inhibitory effects of known UGT isoform-
selective inhibitors on the formation of M-1 glucuronides were evaluated to
(pH 7.5), 25 mg/ml alamethicin, 10 mM MgCl2, 200 mM M-1, and 0.5 mg/ml identify the UGT isoforms involved in the metabolic pathway. Inhibitors used
HLMs. After preincubation for 15 minutes, the reaction was initiated by the in this study were as follows: hecogenin (1 and 10 mM) for UGT1A4, niflumic
addition of 5 mM UDPGA and incubated for 2 hours at 37°C. The reaction acid (0.1 and 1 mM) for UGT1A9, and fluconazole (1 and 3 mM) for UGT2B4
mixtures (2 ml each) were then loaded through preactivated Sep-Pac C18
cartridges (Waters, Milford, MA). Each cartridge was washed with 1 ml
distilled water and M-1 glucuronides were eluted with 1 ml methanol. The final
and 2B7. The formation rates of M-1 glucuronides from M-1 (50 mM) were
determined from reaction mixtures incubated in the presence or absence of
inhibitors. These inhibitors included niflumic acid for UGT1A9, hecogenin for
methanol eluents were pooled and evaporated under a N2 evaporator. The UGT1A4, and fluconazole for UGT2B4/2B7. With the exception of adding
residue was then reconstituted in 50% acetonitrile. M-1 glucuronides were UGT-isoform-specific inhibitors, all other incubation conditions were similar to
separated by an HPLC column (Symmetry C8, 150 Â 4.6 mm, 5 mm; Waters) those described above.
using a mobile phase of 25% acetonitrile containing 0.1% formic acid at
a flow rate of 1.2 ml/min. The fractions containing M-1 glucuronides were
collected and the organic solvent was evaporated under nitrogen stream.
The remaining fractions were lyophilized to obtain white powder. The final
amounts of M-1 glucuronides obtained after purification were 0.5 mg of
SMG1, 0.18 mg of SMG2, and 1.06 mg of SMG3. Overall M-1 glucuronide
yield was 21.5% after purification. The 1H-NMR spectra were acquired using
a JNM ECP 400-MHz spectrometer (JEOL, Tokyo, Japan) with CD3OD as
the solvent. Chemical shifts are expressed in parts per million downfield from
tetramethylsilane.
Correlation Experiments. M-1 (50 mM) was incubated with microsomes
from 20 different human livers to examine the formation of SMG1, SMG2, and
SMG3 from M-1, relative to specific UGT isoform activity. The activity of each
UGT isoform was assessed by LC-MS/MS as described previously for propofol
glucuronidation (UGT1A9) (Court, 2005) and trifluoperazine glucuronidation
(UGT1A4) (Uchaipichat et al., 2006a) with slight modifications; 10 mM
concentrations were used for both compounds and the incubation time was 30
minutes. The correlation coefficients between the formation rates of M-1
glucuronides and the activities of each UGT isoform in a panel of 20 HLMs
were calculated by Spearman’s rank correlation coefficient analysis using the
Fig. 2. LC-MS/MS chromatograms of sarpogrelate glucuronides (A, B) and M-1 glucuronides (C, D) after incubation with rat liver microsomes (RLM) and human liver
microsomes (HLM) in the presence of UDPGA. Chromatograms for parent sarpogrelate and M-1 obtained after microsomal incubation are shown as an inserted
chromatogram.

1532
Kim et al.
SAS software (SAS Institute, Cary, NC). P values ,0.05 were considered to be
statistically significant.
Data Analysis. In the microsomal incubation studies, the apparent kinetic
parameters of M-1 biotransformation (Km and Vmax) were determined by fitting
a one-enzyme Michaelis–Menten equation or a Hill equation. The calculated
parameters included the maximum rate of formation (Vmax), the Michaelis–
UDPGA, whereas one metabolite (SG1) was generated with rat liver
microsomes (RLMs) (Fig. 2). These metabolites were identified as
glucuronide conjugates by full-scan mass analysis, which showed the
addition of a 176-Da glucuronic acid to the parent sarpogrelate (m/z
430). Incubation of M-1 with HLMs in the presence of UDPGA
resulted in the formation of three metabolites (SMG1, SMG2, and
SMG3). However, only two metabolites (SMG1 and SMG3) were
identified when M-1 was incubated with RLMs (Fig. 2). M-1 metabolites
were also confirmed as glucuronide conjugates as judged by addition
of 176 Da to the parent M-1 (m/z 506). The subsequent product ion
mass spectra of the metabolites generated identical fragment ion
patterns, which showed the major fragment ion at m/z 330, indicating
loss of a glucuronide moiety (Supplemental Fig. 1).
Menten constant (apparent Km), and the intrinsic clearance (Clint = Vmax
/
apparent Km). Calculations were performed using the WinNonlin software
(Pharsight, Mountain View, CA). The percentages of inhibition were calculated
by means of the ratio of the quantities of metabolites formed in the absence and
presence of the appropriate specific inhibitor.
Results
Glucuronidation of Sarpogrelate and M-1 in Human and Rat
Structure of M-1 Glucuronides. SMG1, SMG2, and SMG3 were
Liver Microsomes. Two metabolites (SG1 and SG2) were generated isolated and their structures determined by comparing their proton
after incubation of sarpogrelate with HLMs in the presence of NMR spectra to that of M-1 (Fig. 3). SMG1 and SMG3 were tentatively
Fig. 3. 1H NMR spectra of M-1 (A) and its glucuronide metabolites, SMG1 (B), SMG2 (C), and SMG3 (D) in CD3Cl.

UGT Isoforms Involved in Sarpogrelate Metabolite
1533
assigned as O-glucuronides and SMG2 was postulated to be a quaternary at room temperature for 48 hours and reinjected into an HPLC column.
N-glucuronide. The 1H chemical shifts of the anomeric proton of SMG1 Only SMG2 was detected after partial hydrolysis of SG2, whereas
and SMG3 were 4.45 and 4.64 ppm, respectively. The 1H chemical SMG1 and SMG3 were not detected in the hydrolysis products of SG1
shifts of the anomeric proton of SMG2 were farther downfield at 4.89 or SG2 (Supplemental Fig. 2).
ppm, as evidenced by other quaternary N-glucuronides (Alonen et al.,
Identification of UGT Isoforms Involved in Glucuronidation of
2008; Zhu et al., 2008). The racemic mixture of M-1 and the similarity M-1. The relative activities of 12 recombinant human UGT isoforms
of 1H anomeric proton chemical shifts of SMG1 and SMG3 suggest that (UGT1A1, 1A3, 1A4, 1A6, 1A7, 1A8, 1A9, 1A10, 2B4, 2B7, 2B15,
these compounds are diastereomers of O-glucuronide. To assign the and 2B17) expressed in insect cells were determined in terms of the
chirality of SMG1 and SMG3, an LC-MS/MS analysis was performed formation of M-1 glucuronides at 15 and 150 mM. The formation rates
after incubation of individual enantiomer of M-1 with HLMs. The of M-1 glucuronides after incubation of M-1 with UGT isoforms
SMG1 peak was only shown in (R)-M-1, whereas the SMG3 peak was are shown in Fig. 5. SMG1 formation was catalyzed primarily by
observed in (S)-M-1, suggesting that SMG1 was (R)-M-1 O-glucuronide UGT1A9 with minor contributions by UGT2B4 and 2B7, whereas
and SMG3 was (S)-M-1 O-glucuronide (Fig. 4). Sarpogrelate can be SMG3 formation was mediated predominantly by UGT2B4 and 2B7.
enzymatically transformed to M-1 by ester hydrolysis of succinic ester. In contrast, N-glucuronide SMG2 was generated only by UGT1A4.
Furthermore, sarpogrelate can be spontaneously hydrolyzed to M-1 in The formation of SMG1, SMG2, and SMG3 from M-1 was assessed
the aqueous phase (Kamatsu et al., 1992). To further characterize the using UGT-isoform-selective inhibitors or substrates. SMG1 forma-
nature of SMG2, the hydrolysis products of sarpogrelate glucuronides tion was inhibited by the UGT1A9 inhibitor niflumic acid, while
were characterized. It is reasonable to assume that N-glucuronide of M-1 formation of SMG2 and SMG3 were not affected (Fig. 6). The formation
can be generated only from a sarpogrelate N-glucuronide via hydrolysis of SMG2 was preferentially inhibited by hecogenin, a known UGT1A4
of the succinic ester. HPLC fractions containing SG1 and SG2 were left inhibitor, whereas SMG3 formation was inhibited by fluconazole,
a known inhibitor of UGT2B4 and 2B7. Correlation analyses with
activities of known marker UGT isoforms were performed to confirm
the involvement of UGT1A4 and 1A9 in the formation of M-1
glucuronides. As shown in Fig. 7, significant correlations were found
between SMG1 formation rates and UGT1A9-mediated propofol
glucuronidation (r = 0.6269, P , 0.0031) and between SMG2
formation rates and UGT1A4-medated trifluoperazine glucuronidation
(r = 0.6623, P , 0.0015).
Enzyme Kinetic Analysis. Kinetic analyses of formation of M-1
glucuronides were performed with pooled HLMs using 10–1000 mM
M-1 (Fig. 8A). The formation of M-1 glucuronides by the HLMs and
recombinant UGT isoforms was fitted by Michaelis–Menten one-
enzyme model. Fitting of the data to a two-enzyme model did not
significantly improve the regression compared with fitting of the data
to one-enzyme model. The kinetic parameters estimated from the
pooled HLMs and recombinant UGT isoforms are shown in Table 1.
Intrinsic clearance (Clint) of SMG1, SMG2, and SMG3 in HLMs were
4.80, 1.16, and 8.72 ml/min∙mg protein, respectively. Substrate
inhibition of O-glucuronides (SMG1 and SMG 3) formation was noted
at the highest concentration (1000 mM) used. Next, the kinetic parameters
of formation of M-1 glucuronides were determined using 10–1000
mM M-1 after incubation with recombinant UGT1A4, 1A9, 2B4, and
2B7 (Fig. 8, B–D). UGT1A9-mediated formation of SMG1 showed
17-fold higher Vmax than did UGT2B4-mediated formation of SMG1.
The Clint of SMG1 by UGT1A9 was 3-fold greater than that by
UGT2B4. However, the Clint of SMG3 by UGT2B4 was similar to
that of SMG1 formation by UGT2B4. Based on the Clint data, UGT2B7
was considered to make a greater contribution to SMG3 formation did
than UGT2B4.
Discussion
The glucuronidation of sarpogrelate and its active metabolite M-1 is
an important part of sarpogrelate disposition in humans. In experimental
rats, the majority of circulating M-1 in the plasma after oral administration
was determined to be glucuronide conjugates (Takada et al., 1997). In
addition, metabolic analysis demonstrated that ;76% of M-1 was
present as a glucuronide, whereas only 24% was present as free M-1 in
bile from rats treated with sarpogrelate. Our results also suggest M-1
glucuronidation to be more efficient than that of sarpogrelate. Intrinsic
clearance rates of M-1 O-glucuronides (SMG1 and SMG3) and N-
Fig. 4. LC-MS/MS chromatograms of M-1 glucuronides after incubation of racemic
M-1 (A), R-enantiomer (B), and S-enantiomer (C) of M-1 with human liver
microsomes in the presence of UDPGA.
glucuronide (SMG2) were 10-fold higher than those of the sarpogrelate

1534
Kim et al.
Fig. 5. Formation of the metabolites SMG1 (A), SMG2 (B), and SMG3 (C) by recombinant human UGT isoforms. M-1 at two different concentrations of 15 (u) and 150
mM (j) was incubated with each recombinant UGT isoform (UGT1A1, UGT1A3, UGT1A4, UGT1A6, UGT1A7, UGT1A8, UGT1A9, UGT1A10, UGT2B4, UGT2B7,
UGT2B15, and UGT2B17) at a protein concentration of 0.25 mg/ml, and other conditions are the same with HLM. Each bar represents the mean 6 S.D. of triplicate
determinations.
O-glucuronide (SG1) and N-glucuronide (SG2) in HLMs. Un- O-glucuronide. The structure of SMG2 was further supported by
derstanding the nature of M-1 glucuronidation is of interest because species differences in the formation of SMG2; SMG2 was generated
M-1 is known to be more pharmacologically active than its parent only by HLMs and not by RLMs (Fig. 2). The formation of quaternary
compound sarpogrelate.
ammonium glucuronides from tertiary amines is a reaction largely
In the present investigation, we demonstrated that sarpogrelate restricted to humans and generally not observed in animal species,
is metabolized to O-glucuronide (SG1) and N-glucuronide (SG2), including rats, mice, dogs, and monkeys (Soars et al., 2001; Kaku
whereas M-1 is transformed into two O-glucuronides (SMG1 and et al., 2004; Kaivosaari et al., 2011). The structure of sarpogrelate
SMG3) and a N-glucuronide in the presence of UDPGA in HLMs glucuronides was confirmed by analysis of their hydrolysis products.
(Fig. 1). The structures of M-1 glucuronides were characterized by Sarpogrelate can be hydrolyzed at the ester moiety in its chemical
1
mass and H-NMR spectroscopy. No structural information could be structure, leading to formation of M-1. The conversion of SG1 to M-1
obtained from product ion mass spectra of M-1 glucuronides because in an aqueous solution indicated that SG1 was sarpogrelate O-
1
all were almost identical. NMR spectra showed that the H chemical glucuronide and the hydrolysis of the succinic glucuronide moiety
shifts of the anomeric proton of SMG1 and SMG3 were similar (4.45 from SG1 resulted in the formation of M-1. However, SG2 was
and 4.64 ppm), whereas that of SMG2 was farther downfield, at 4.89 hydrolyzed to SMG2 but not to M-1, suggesting that SG2 was a
ppm. In general, the chemical shifts of anomeric protons in O- quaternary ammonium glucuronide of sarpogrelate and that a glucu-
glucuronides were 4.4–4.6 ppm, whereas those in N-glucuronides ronide moiety remained after the hydrolysis of succinic ester, leading
were 5.1–5.5 ppm (Alonen et al., 2008; Zhu et al., 2008; Pallmann to the formation of the N-glucuronide of M-1. The structure of SMG2
et al., 2010). Accordingly, SMG1 and SMG3 were tentatively was further supported by the fact that SG2 formation occurred in
assigned as O-glucuronides and SMG2 as a quaternary ammonium N- HLMs but not in RMLs.
glucuronide of M-1. The structure of M-1 suggested that O-
Identification of UGT isoforms responsible for the formation of
glucuronidation can occur only at the hydroxyl group located on the M-1 glucuronides was demonstrated with several approaches. First,
dimethylamino-2-hydroxy-propyl moiety. Considering the racemic incubation with 12 recombinant human UGT isoforms showed that
mixture of sarpogrelate and M-1, SMG1 and SMG3 were postulated to only UGT1A4 catalyzed N-glucuronidation of M-1. UGT1A4 is the
be diastereomers of M-1 O-glucuronide. The stereochemistry of M-1 major isoform involved in the glucuronidation of tertiary amines,
O-glucuronides (SMG1 and SMG3) was confirmed by the incubation including 19-hydroxymidazolam, imipramine, and tamoxifen (Green
of individual enantiomer of M-1 with HLMs. SMG1 was identified to and Tephly, 1996; King et al., 2000; Zhu et al., 2008). No UGT1A4
be (R)-M-1 O-glucuronide and SMG3 was characterized as (S)-M-1 ortholog is present in rat tissues (King et al., 2000). Incubation of

UGT Isoforms Involved in Sarpogrelate Metabolite
1535
Fig. 6. Effects of UGT isoform-selective inhibitors on
the formation of the metabolites SMG1, SMG2, and
SMG3 in HLMs. Niflumic acid (0.1 and 1 mM, UGT1A9)
(A), hecogenin (1 and 10 mM, UGT1A4) (B), and
fluconazole (1 and 3 mM, UGT2B4/2B7) (C) were used
as inhibitors of each UGT isoform. Each bar represents
the mean 6 S.D. of triplicate determinations.
sarpogrelate or M-1 with RMLs in the presence of UDPGA generated a selective inhibitor of UGT1A9 (Gaganis et al., 2007; Miners et al.,
O-glucuronides (SG1, SMG1, and SMG3) but not N-glucuronides 2010), whereas the formation of SMG3 was inhibited by fluconazole
(SG2 and SMG2), consistent with previous reports that only the (3 mM, ;90%), a selective inhibitor of UGT2B7 (Uchaipichat et al.,
UGT1A4 isoform catalyzes N-glucuronidation of tertiary amines. SMG1 2006b; Raungrut et al., 2010) (Fig. 7). SMG2 formation was markedly
formation was catalyzed mainly by UGT1A9 with minor contributions inhibited by hecogenin (10 mM, ;60%), a potent UGT1A4 inhibitor
by UGT 2B4 and 2B7, whereas SMG3 was generated by both (Uchaipichat et al., 2006a). Thirdly, correlations between M-1
UGT2B4 and 2B7, suggesting considerable stereo-selectivity in M-1 glucuronidation and known UGT isoform activities were assessed
O-glucuronidation. UGT1A9 played a major role in glucuronidation of using a panel of HLMs. Significant correlations were observed
R-enantiomer whereas UGT2B4 and 2B7 were preferentially involved between the SMG1 formation and UGT1A9-mediated propofol
in glucuronidation of S-enantiomer of M-1.
glucuronidation rates (Court, 2005) (P = 0.0031), and between the
Secondly, inhibition of M-1 glucuronide formation was evalu- SMG2 formation and UGT1A4-mediated trifluoperazine glucuroni-
ated using UGT-isoform-specific substrates and/or inhibitors. SMG1 dation rates (Uchaipichat et al., 2006a) (P = 0.0015). Our results
formation was strongly inhibited by niflumic acid (1 mM, ;80%), collectively suggested that UGT1A4 catalyzed only N-glucuronidation
Fig. 7. Correlations between formation of the SMG1 and propofol glucuronidation (A) and between formation of the SMG2 and trifluoperazine glucuronidation (B) in 20
different HLMs. P , 0.05 indicates statistical significance. Each data point represents the mean of duplicate determinations.

1536
Kim et al.
Fig. 8. Michaelis–Menten plots for formation of the SMG1, SMG2, and SMG3 in HLM (A), SMG1 by UGT1A9 and 2B4 (B), SMG2 by UGT1A4 (C), and SMG3 by
UGT2B4 and 2B7 (D). HLMs or recombinant UGTs were incubated with 0–1000 mM M-1 at 37°C for 20 minutes in the presence of UDPGA (5 mM). Each data point
represents the mean 6 S.D. of triplicate determinations.
TABLE 1
both the variability in M-1 pharmacokinetics and the magnitude of
Kinetic parameters of M1 glucuronidation by human liver microsomes and
drug-drug interactions.
recombinant uridine 59-diphospho-glucuronosyltransferases
Acknowledgments
Kinetic Parameters
The authors thank Dr. Yohan Park (College of Pharmacy, Inje University)
for helpful comments on the interpretation of 1H-NMR spectra.
Km
Vmax
CLint (Vmax/Km
ml/min per
)
nmol/min per
milligram protein
mM
Authorship Contributions
milligram protein
Participated in research design: Lee, J.-G. Shin, D.-H. Kim.
Conducted experiments: H.-J. Kim, Jeong, Seo.
Contributed new reagents or analytic tools: K.J. Shin, Choi.
Performed data analysis: H.-J. Kim, Jeong, Seo, K.J. Shin, Ghim, Sohn,
J.-G. Shin, D.-H. Kim.
O-Glucuronide
I (SMG1)
HLM
UGT1A9
UGT2B4
234.6
564.1
105.1
1.126
0.731
0.043
4.80
1.30
0.42
N-Glucuronide
(SMG2)
HLM
UGT1A4
534.6
476.2
0.621
0.298
1.16
0.63
Wrote or contributed to the writing of the manuscript: H.-J. Kim, D.-H.
Kim.
O-Glucuronide II
(SMG3)
HLM
UGT2B4
UGT2B7
203.1
186.6
96.3
1.772
0.097
0.031
8.72
0.52
0.32
References
Alonen A, Jansson J, Kallonen S, Kiriazis A, Aitio O, Finel M, and Kostiainen R (2008) Enzyme-
assisted synthesis and structure characterization of glucuronic acid conjugates of losartan,
candesartan, and zolarsartan. Bioorg Chem 36:148–155.
Court MH (2005) Isoform-selective probe substrates for in vitro studies of human UDP-glucur-
onosyltransferases. Methods Enzymol 400:104–116.
of M-1, whereas O-glucuronidation was mediated by UGT1A9, UGT2B4,
and UGT2B7.
Furukawa K, Tanabe T, Hoshino S, Mishima Y, Shionoya S, Yoshizaki S, Katsumura T,
and Nakashima M (1991) Therapeutic effects of sarpogrelate hydrochloride (MCI-9042) on
chronic arterial occlusive disease: a double-blind comparison with ticlopidine hydrochloride.
Jpn J Clin Pharmacol Ther 7:1747–1770.
Gaganis P, Miners JO, and Knights KM (2007) Glucuronidation of fenamates: kinetic studies
using human kidney cortical microsomes and recombinant UDP-glucuronosyltransferase
(UGT) 1A9 and 2B7. Biochem Pharmacol 73:1683–1691.
In summary, our results demonstrated that the sarpogrelate active
metabolite M-1 was transformed to SMG1, SMG2, and SMG3 in
HLMs. Additionally, UGT1A9, UGT1A4 and UGT2B4 are the major
isoforms responsible for SMG1, SMG2, and SMG3 formation,
respectively. Identification of the UGT1A9, UGT1A4, and UGT2B4
isoforms as responsible for M-1 glucuronidation will prove valuable
for determination of the impact of these isoform polymorphisms on
Green MD and Tephly TR (1996) Glucuronidation of amines and hydroxylated xenobiotics and
endobiotics catalyzed by expressed human UGT1.4 protein. Drug Metab Dispos 24:356–363.
Guillemette
C (2003) Pharmacogenomics of human UDP-glucuronosyltransferase enzymes.
Pharmacogenomics J 3:136–158.

UGT Isoforms Involved in Sarpogrelate Metabolite
1537
Hara H, Osakabe M, Kitajima A, Tamao Y, and Kikumoto R (1991) MCI-9042, a new antiplatelet
agent is a selective S2-serotonergic receptor antagonist. Thromb Haemost 65:415–420.
Kaivosaari S, Finel M, and Koskinen M (2011) N-glucuronidation of drugs and other xenobiotics
by human and animal UDP-glucuronosyltransferases. Xenobiotica 41:652–669.
Kaku T, Ogura K, Nishiyama T, Ohnuma T, Muro K, and Hiratsuka A (2004) Quaternary
ammonium-linked glucuronidation of tamoxifen by human liver microsomes and UDP-
glucuronosyltransferase 1A4. Biochem Pharmacol 67:2093–2102.
Kamatsu T, Kitazawa K, Izawa O, Inoue S, Ohshima M, and Iida S (1992) Study of the metabolic
fate of (6)-2-(dimethyl-amino)-1-[[o-(m-methoxyphenethyl)-phenoxy]methyl]ethyl hydrogen
succinate hydrochloride (MCI-9042). (I). Metabolism of MCI-9042 in rats, dogs, monkeys and
human. Yakubutsu Dotai 6:353–375.
Shimizu T, Yamada Y, Tashita A, Yamamoto K, Kotaki H, Sawada Y, and Iga T (1999)
[Pharmacokinetic analysis of antiplatelet effect of sarpogrelate hydrochloride and its applica-
tion to drug dosage regimen—modeling based on reversible inhibition of 5-HT2 serotonergic
receptor in the platelet membrane by sarpogrelate hydrochloride and its metabolite]. Yakugaku
Zasshi 119:850–860.
Shin JG, Soukhova N, and Flockhart DA (1999) Effect of antipsychotic drugs on human liver
cytochrome P-450 (CYP) isoforms in vitro: preferential inhibition of CYP2D6. Drug Metab
Dispos 27:1078–1084.
Shinohara Y, Nishimaru K, Sawada T, Terashi A, Handa S, Hirai S, Hayashi K, Tohgi H,
Fukuuchi Y, and Uchiyama S, et al.; S-ACCESS Study Group (2008) Sarpogrelate-Aspirin
Comparative Clinical Study for Efficacy and Safety in Secondary Prevention of Cerebral
King CD, Rios GR, Green MD, and Tephly TR (2000) UDP-glucuronosyltransferases. Curr Drug
Infarction (S-ACCESS): A randomized, double-blind, aspirin-controlled trial. Stroke 39:
Metab 1:143–161.
1827–1833.
Liston HL, Markowitz JS, and DeVane CL (2001) Drug glucuronidation in clinical psycho-
pharmacology. J Clin Psychopharmacol 21:500–515.
Soars MG, Riley RJ, and Burchell B (2001) Evaluation of the marmoset as a model species for
drug glucuronidation. Xenobiotica 31:849–860.
Miners JO, Mackenzie PI, and Knights KM (2010) The prediction of drug-glucuronidation
parameters in humans: UDP-glucuronosyltransferase enzyme-selective substrate and inhibitor
probes for reaction phenotyping and in vitro-in vivo extrapolation of drug clearance and drug-
drug interaction potential. Drug Metab Rev 42:196–208.
Nishihira K, Yamashita A, Tanaka N, Kawamoto R, Imamura T, Yamamoto R, Eto T, and Asada
Y (2006) Inhibition of 5-hydroxytryptamine receptor prevents occlusive thrombus formation on
neointima of the rabbit femoral artery. J Thromb Haemost 4:247–255.
Pallmann T, Jonas U, Wagner M, Thevis M, Kaeferstein H, Rothschild MA, and Bender K (2010)
Enzyme-assisted synthesis and structural characterization of pure benzodiazepine glucuronide
epimers. Eur J Pharm Sci 39:233–240.
Takada Y, Takada A, and Urano T (1997) MCI-9042, the new selective antagonist of serotonergic
(5-HT2A) receptors. Cardiovasc Drug Rev 15:101–121.
Uchaipichat V, Mackenzie PI, Elliot DJ, and Miners JO (2006a) Selectivity of substrate (tri-
fluoperazine) and inhibitor (amitriptyline, androsterone, canrenoic acid, hecogenin, phenylbutazone,
quinidine, quinine, and sulfinpyrazone) “probes” for human udp-glucuronosyltransferases. Drug
Metab Dispos 34:449–456.
Uchaipichat V, Winner LK, Mackenzie PI, Elliot DJ, Williams JA, and Miners JO (2006b)
Quantitative prediction of in vivo inhibitory interactions involving glucuronidated drugs from
in vitro data: the effect of fluconazole on zidovudine glucuronidation. Br J Clin Pharmacol 61:
427–439.
Rashid M, Manivet P, Nishio H, Pratuangdejkul J, Rajab M, Ishiguro M, Launay JM,
and Nagatomo T (2003) Identification of the binding sites and selectivity of sarpogrelate,
a novel 5-HT2 antagonist, to human 5-HT2A, 5-HT2B and 5-HT2C receptor subtypes by
molecular modeling. Life Sci 73:193–207.
Zhu B, Bush D, Doss GA, Vincent S, Franklin RB, and Xu S (2008) Characterization of 19-
hydroxymidazolam glucuronidation in human liver microsomes. Drug Metab Dispos 36:
331–338.
Raungrut P, Uchaipichat V, Elliot DJ, Janchawee B, Somogyi AA, and Miners JO (2010) In vitro-
in vivo extrapolation predicts drug-drug interactions arising from inhibition of codeine glu-
curonidation by dextropropoxyphene, fluconazole, ketoconazole, and methadone in humans.
J Pharmacol Exp Ther 334:609–618.
Address correspondence to: Dr. Dong-Hyun Kim, Department of Pharmacology
and PharmacoGenomics Research Center, Inje University College of Medicine,
#633-165 Gaegum-Dong, Jin-Gu, Busan 614-735, Korea. E-mail: dhkim@inje.ac.kr
Saini HK, Takeda N, Goyal RK, Kumamoto H, Arneja AS, and Dhalla NS (2004) Therapeutic
potentials of sarpogrelate in cardiovascular disease. Cardiovasc Drug Rev 22:27–54.