Biotransformation of prasugrel, a novel thienopyridine antiplatelet agent, to the pharmacologically active metabolite

Article abstract of DOI:10.1124/dmd.110.032086

Prasugrel, a novel thienopyridine antiplatelet agent, undergoes rapid hydrolysis in vivo to a thiolactone, R-95913, which is further converted to its thiol-containing, pharmacologically active metabolite, R-138727, by oxidation via cytochromes P450 (P450). We trapped a sulfenic acid metabolite as a mixed disulfide with 2-nitro-5-thiobenzoic acid in an incubation mixture containing the thiolactone R-95913, expressed CYP3A4, and NADPH. Further experiments investigated one possible mechanism for the conversion of the sulfenic acid to the active thiol metabolite in vitro. A mixed disulfide form of R-138727 with glutathione was found to be a possible precursor of R-138727 in vitro when glutathione was present. The rate constant for the reduction of the glutathione conjugate of R-138727 to R-138727 was increased by addition of human liver cytosol to the human liver microsomes. Thus, one possible mechanism for the ultimate formation of R-138727 in vitro can be through formation of a sulfenic acid mediated by P450s followed possibly by a glutathione conjugation to a mixed disulfide and reduction of the disulfide to the active metabolite R-138727. Copyright

Full text of DOI:10.1124/dmd.110.032086

0090-9556/10/3806-898–904$20.00
DRUG METABOLISM AND DISPOSITION
Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics
DMD 38:898–904, 2010
Vol. 38, No. 6
32086/3589665
Printed in U.S.A.
Short Communication
Biotransformation of Prasugrel, a Novel Thienopyridine Antiplatelet
Agent, to the Pharmacologically Active Metabolite
Received January 6, 2010; accepted February 26, 2010
ABSTRACT:
Prasugrel, a novel thienopyridine antiplatelet agent, undergoes disulfide form of R-138727 with glutathione was found to be a
rapid hydrolysis in vivo to a thiolactone, R-95913, which is further possible precursor of R-138727 in vitro when glutathione was
converted to its thiol-containing, pharmacologically active metab- present. The rate constant for the reduction of the glutathione
olite, R-138727, by oxidation via cytochromes P450 (P450). We conjugate of R-138727 to R-138727 was increased by addition of
trapped a sulfenic acid metabolite as a mixed disulfide with 2-nitro- human liver cytosol to the human liver microsomes. Thus, one
5-thiobenzoic acid in an incubation mixture containing the thiolac- possible mechanism for the ultimate formation of R-138727 in vitro
tone R-95913, expressed CYP3A4, and NADPH. Further experi- can be through formation of a sulfenic acid mediated by P450s
ments investigated one possible mechanism for the conversion of followed possibly by a glutathione conjugation to a mixed disulfide
the sulfenic acid to the active thiol metabolite in vitro. A mixed and reduction of the disulfide to the active metabolite R-138727.
Prasugrel [Effient (Eli Lilly and Company, Indianapolis, IN) in the involvement of a sulfenic acid and a glutathione conjugate of
United States and Efient (Eli Lilly and Company) in the European R-138727 in the in vitro production of prasugrel’s active metabolite
Union], clopidogrel [Plavix (Sanofi-Aventis, Paris, France)/Iscover (R-138727) from the thiolactone intermediate (R-95913) (Fig. 1).
(Bristol-Myers Squibb, New York, NY)], and ticlopidine (Ticlid;
Sanofi-Aventis) are thienopyridine antiplatelet agents. Prasugrel has
been shown to reduce the rate of thrombotic cardiovascular events and
stent thrombosis in patients with acute coronary syndrome that are
undergoing percutaneous coronary intervention (Wiviott et al., 2007)
(Effient package insert). The thienopyridines are prodrugs that are
converted in vivo to their pharmacologically active metabolites that
possess a thiol group via a corresponding thiolactone metabolite
(Farid et al., 2010). However, with the exception of an oxidation step
catalyzed by cytochrome P450 (Savi et al., 1994; Rehmel et al., 2006),
the mechanism for the active metabolite formation from thienopyri-
dines remained unknown until recently reported for ticlopidine and
clopidogrel (Dansette et al., 2009). In the case of prasugrel, the active
metabolite R-138727 was not detected when the thiolactone metabo-
lite R-95913 was incubated with liver homogenates or microsomes in
the absence of cofactors, but it was detected when NADPH and a
Materials and Methods
Materials. R-138727 (prasugrel active metabolite), glutathione conjugate of
R-138727, and R-95913 (prasugrel thiolactone metabolite) were synthesized
by Ube Industries, Ltd. (Ube, Japan). 5,5Ј-Dithio-bis(2-nitrobenzoic acid)
(DTNB) and tris(carboxyethyl)phosphine (TCEP), used as a reagent to reduce
DTNB to 2-nitro-5-thiobenzoic acid (TNB), were purchased from Sigma-
Aldrich (St. Louis, MO). Microsomes from insect cells that expressed
CYP3A4 (cDNA-expressed CYP3A4) were purchased from Gentest Corpora-
tion (Woburn, MA). All other chemicals and reagents were commercially
available and were of the highest grade. Water was purified by using a Milli-Q
purification system (Millipore Corporation, Billerica, MA).
Preparation of TNB-R-138727 Disulfide. A mixed disulfide of R-138727
with TNB (TNB-R-138727 disulfide) was prepared by reacting R-138727 with
equimolar DTNB in an aqueous solution (Fig. 2) because the synthesis using
the chemically unstable sulfenic acid-derivative of R-138727 and TNB was
difficult to perform. After dilution of the reaction mixture with acetonitrile and
reducing agent such as glutathione were added to the system (Kazui et after centrifugation, a 20-␮l aliquot of the supernatant fraction was directly
injected onto a YMC-Pack ODS-A (A-312) column (150 mm ϫ 6.0 mm i.d.,
5 ␮m; YMC Co., Ltd., Kyoto, Japan), maintained at ambient temperature. A
high-performance liquid chromatography (HPLC) analysis was performed by
using a low-pressure gradient elution system that consisted of an L-7100
Intelligent Pump, a D-7500 Chromato-Integrator, and an L-7400S UV Detector
(Hitachi, Ltd., Tokyo, Japan). The mobile phase, 30% (v/v) acetonitrile in
water that contained 0.01% (v/v) trifluoroacetic acid, was used at a flow rate
of 1.0 ml/min. Profiling of the reaction products was performed with UV
absorption detection at 220 nm. Two peaks in the chromatogram, which
corresponded to the two diastereoisomers of TNB-R-138727 disulfide [re-
tention times 19.18 (Peak A) and 21.57 (Peak B) min], were collected
separately and lyophilized for structure elucidation by liquid chromatog-
al., 2000). A sulfenic acid, which is a likely intermediate for the
activation of prasugrel, reacts readily with a thiol compound, typically
glutathione in vivo, to yield a disulfide metabolite (Decker et al.,
1991; Kassahun et al., 2001; Reddy et al., 2005; Dansette et al., 2009).
The formed disulfide can be further reduced to provide a thiol-
containing compound. The objective of this study is to investigate the
N.A.F. has retired from Eli Lilly and Company.
Article, publication date, and citation information can be found at
http://dmd.aspetjournals.org.
doi:10.1124/dmd.110.032086.
ABBREVIATIONS: prasugrel, 2-acetoxy-5-(␣-cyclopropylcarbonyl-2-fluorobenzyl)-4,5,6,7-tetrahydrothieno[3,2-c]pyridine; DTNB, 5,5Ј-dithio-
bis(2-nitrobenzoic acid); TCEP, tris(carboxyethyl)phosphine; TNB, 2-nitro-5-thiobenzoic acid; HPLC, high-performance liquid chromatography;
LC-MS/MS, liquid chromatography with tandem mass spectrometry; LC-MS, liquid chromatography-mass spectrometry; ESI, electrospray
ionization; MS, mass spectrometry; MPBr, m-methoxyphenacybromide; MS/MS, tandem mass spectrometry.
898

MECHANISM FOR PRASUGREL ACTIVE METABOLITE FORMATION
899
FIG. 1. Proposed metabolic pathway from the thiolactone of prasugrel to the pharmacologically active metabolite via a sulfenic acid intermediate. TNB was used to trap
a sulfenic acid intermediate forming TNB-R-138727 disulfide. GSSG, oxidized form of glutathione.
raphy with tandem mass spectrometry (LC-MS/MS) and NMR analysis, as
described below.
mixing equimolar DTNB with TCEP, dissolved in buffer (Han and Han, 1994).
Then, 10 ␮l of R-95913 (final concentration: 250 ␮M) was added to the
Identification of TNB-R-138727 Disulfide. Electrospray ionization (ESI) mixture, which was further incubated at 37°C for 0, 15, 30, and 60 min. At
analysis was performed by using a Q-TOF hybrid-type MS/MS spectrometer each time point, 80 ␮l of the incubation mixture was collected, added to 160
(Micromass UK, Ltd., Manchester, UK). A liquid chromatography-mass spec-
␮l of acetonitrile to stop the reaction, and centrifuged (15,000g, 3 min, 4°C).
trometry (LC-MS) analysis in the ESI mode was performed by using a For determination of R-95913, the supernatant samples were diluted with the
low-pressure gradient system that consisted of an L-6000, an L-6300 Intelli- supernatant fraction obtained by adding acetonitrile to cDNA-expressed
gent Pump, a D-2500 Chromato-Integrator, and an L-4000 UV-Detector (Hi- CYP3A4 without any other components and after centrifugation. A 10 ␮l-
tachi, Ltd.) equipped with a YMC Pack ODS-A column (50 mm ϫ 1.5 mm i.d., aliquot of each sample was subjected to a LC-MS analysis as described in the
5 ␮m). The other conditions for HPLC were the same as those described
above. The flow rate was set at 1 ml/min, and after passage through the
column, the eluent was introduced to the ESI source at a flow rate of
approximately 100 ␮l/min. The acceleration voltage of 3.3 kV and the cone
voltage of 45 to 50 V were used in this ionization mode. The source block was
maintained at 120°C. The desolvation gas was heated to 300°C. Xenon was
used as the collision gas, and the laboratory frame collision energy was 20 eV
when acquiring the product ion spectra.
following section.
Quantification of R-95913, R-138727, and TNB-R-138727 Disulfide by
LC-MS. Quantification of R-95913, R-138727, and TNB-R-138727 in the
incubation mixture of R-95913 was carried out on an Alliance HPLC system
that consisted of a 2795 Separations Module (Waters Corporation, Milford,
MA) coupled to a ZQ-2000 (Waters Corporation) with the ESI source in
positive ion mode. Each sample (10 ␮l) was injected onto an XTerra MS C18
column (50 mm ϫ 2.1 mm i.d., 3.5 ␮m; Waters Corporation), which was
maintained at 40°C. A two-component mobile phase (pumped at 0.5 ml/min)
After purification of Peaks A and B by HPLC, NMR analysis was per-
formed by using a Varian (INOVA-500; Varian, Inc., Palo Alto, CA) 500 MHz contained solvent A, which is a mixture of formic acid, 5 mM ammonium
spectrometer, and results were reported as ␦ in parts per million relative to acetate, and acetonitrile (0.2/95/5, v/v/v), and solvent B, which is a mixture of
Me4Si as the internal standard. Abbreviations of the 1H NMR peak splitting formic acid, 5 mM ammonium acetate, and acetonitrile (0.2/5/95, v/v/v). The
patterns are as follows: bs, broad singlet; s, singlet; and m, multiplet.
mobile phase initially consisted of 100% solvent A for 0 to 1 min, then solvent
Metabolism of R-95913 by cDNA-Expressed CYP3A4 Supplemented B was increased from 50 to 100% linearly from 1 to 3.75 min, and was
with NADPH in the Presence of TNB or DTNB. Duplicate mixtures (total maintained at 100% B from 3.75 to 4.5 min, before recycling back to the initial
volume of 390 ␮l each) containing potassium phosphate buffer (final concen- condition.
tration: 11.25 mM, pH 7.4), NADP (final concentration: 2.5 mM), D-glucose
6-phosphate (final concentration: 25 mM), glucose-6-phosphate dehydroge- as follows: capillary voltage, 3.5 kV; ion source temperature, 100°C; capillary
nase (final concentration: 0.5 units/ml), magnesium chloride (final concentra- temperature, 350°C; cone gas (N2) flow and desolvation gas (N2) flow, 50 l/h
tion: 10 mM), and cDNA-expressed CYP3A4 (final concentration: 800 pmol/ and 400 l/h, respectively; multiplier voltage, 650 V. Detection was performed
ml) were preincubated at 37°C for 5 min in the presence of 1 mM TNB or in the single ion monitoring mode. The cone voltages were 33, 33, and 27 V
The operating parameters of the mass spectrometry (MS) detector were set
DTNB as final concentration. TNB was prepared immediately before use by with selected reaction monitoring of m/z 332, 350, and 547 for R-95913,
FIG. 2. Thiol-exchange reaction of R-138727
with DTNB forming TNB-R-138727 disulfide.

900
HAGIHARA ET AL.
R-138727, and TNB-R-138727 disulfide, respectively. Phenacetin (Sigma-
Aldrich) was used as the internal standard. The concentrations of each analyte
in the samples were calculated by using the computer software MassLynx
(version 3.5; Micromass UK Ltd.) and are expressed to three significant
figures.
Standard stock solution that contained 1 mM TNB-R-138727 disulfide in
50% ethanol solution was prepared by adding a solution of DTNB in potassium
phosphate buffer to that of R-138727 in ethanol at a final concentration of 1
mM. This standard stock solution was diluted 2-fold sequentially with aceto-
nitrile to prepare standard solutions of a TNB-R-138727 disulfide at concen-
trations from 15.6 to 1000 nM. In the same manner as described above,
standard solutions of R-95913 or R-138727 were prepared by diluting the
standard stock solution that contained 1 mM R-95913 or 1 mM R-138727
sequentially 2-fold with acetonitrile. The calibration curves were constructed
by plotting the peak area ratios of the standard compound to the internal
standard versus the nominal concentrations.
Formation of the Glutathione Conjugate and R-138727 by cDNA-
Expressed CYP3A4. Duplicate mixtures (total volume of 990 ␮l each) that
contained potassium phosphate buffer (final concentration: 29–34 mM, pH
7.4), NADP (final concentration: 2.5 mM), D-glucose 6-phosphate (final con-
centration: 25 mM), glucose-6-phosphate dehydrogenase (final concentration:
0.5 units/ml), magnesium chloride (final concentration: 10 mM), and cDNA-
expressed CYP3A4 (final concentration: 108 pmol/ml) with or without gluta-
thione (final concentration: 5 mM) were preincubated at 37°C for 5 min. Then,
10 ␮l of R-95913, glutathione conjugate of R-138727, or R-138727 (final
concentration: 1, 0.4, or 1 ␮M, respectively) were added to the mixture, which
was further incubated at 37°C for 5, 15, 30, 45, and 60 min. At each time point,
160 ␮l of the incubation mixture was collected, added to the solution contain-
ing 320 ␮l of methanol, 3.2 ␮l of 500 mM m-methoxyphenacybromide (MPBr
solution in acetonitrile), and 160 ␮l of the internal standard solution (100 ng/ml
R-135766) to terminate the reaction. The mixture was stored at room temper-
ature for 10 min and centrifuged (15,000g, 3 min, 4°C). A 10 ␮l-aliquot of the
supernatant fraction was subjected to the LC-MS/MS system as described
below.
Formation of the Glutathione Conjugate Metabolite and R-138727
from R-95913 in Human Liver Microsomes and Cytosol. Triplicate mix-
tures (total volume of 990 ␮l each) contained 36–39 mM potassium phosphate
buffer (pH 7.4), an NADPH-generating system (NADPϩ 1.55 mM), 3.3 mM
glucose 6-phosphate, 0.4 units/ml glucose-6-phosphate dehydrogenase, 3.3
mM MgCl2, 5 mM glutathione, and 1 mg/ml human liver microsome with or
without human liver cytosol (1 mg/ml). Each mixture was preincubated at
37°C for 5 min, and 10 ␮l of R-95913 (10 ␮M as a final concentration) was
added to each respective mixture, which was incubated at 37°C for 1, 2, 5, 10,
15, and 30 min. At each time point, 100 ␮l of the incubation mixture was
collected and mixed with a double amount of 5 mM MPBr acetonitrile solution
and an equivalent amount of the internal standard solution to stop the reaction
and left for 10 min at room temperature to derivatize the thiol moiety of
R-138727. After derivatization, the mixture was centrifuged (15,000g, 3 min,
4°C). A 10 ␮l-aliquot of the supernatant fraction was subjected to LC-MS/MS
analysis.
dZ(1)
dt
ϭ Ϫ(k1 ϩ k4) ϫ Z(1)
(1)
(2)
(3)
dZ(2)
dt
ϭ k1 ϫ Z(1) Ϫ k2 ϫ Z(2) ϩ k3 ϫ Z(3)
dZ(3)
ϭ k2 ϫ Z(2) Ϫ k3 ϫ Z(3)
dt
where Z(1) (eq. 1), Z(2) (eq. 2), and Z(3) (eq. 3) represent the concentrations
of R-95913, the glutathione conjugate of R-138727, and R-138727, respec-
tively, and k1, k2, k3, and k4 represent the rate constants of the formation of
the glutathione conjugate of R-138727 from R-95913 (reflecting the rate-
limiting, likely the oxidation, step in two consecutive steps), the formation of
R-138727 from the glutathione conjugate of R-138727, the formation of the
glutathione conjugate of R-138727 from R-138727, and conversion of
R-95913 to metabolites other than the glutathione conjugate of R-138727 and
R-138727, respectively. The concentrations of the glutathione conjugate of
R-138727 and R-138727 were simultaneously fitted to the equations above by
using WinNonlin Professional software (version 4.0.1; Pharsight, Mountain
View, CA).
Quantification of R-95913, R-138727, and Glutathione Conjugate by
LC-MS/MS. The assays were performed based on the methods described in
the preceding section and a previous report by Farid et al., 2007b. Quantifi-
cation of R-95913, MPBr-derivatized R-138727, and glutathione conjugate
was carried out with an Alliance HPLC system, which consisted of 2690
Separations Module (Waters Corporation) coupled to a Quattro LC-MS/MS
system (Waters Corporation) with the ESI source in positive ion mode. The
mobile phase that contained methanol, purified water, and trifluoroacetic acid
(570/430/0.5, v/v/v) was applied on an Inertsil ODS-3 column (150 mm ϫ 2.1
mm i.d., 5 ␮m; GL Sciences Inc., Tokyo, Japan).
The concentrations of each analyte in the samples were calculated by using
MassLynx software (version 4.0; Waters Corporation). The standard stock
solution was diluted 5-fold sequentially with acetonitrile to prepare the stan-
dard solutions at concentrations from 1.6 to 1000 nM. The standard curves
were generated by using each analyte’s peak area ratio equal to that of the
internal standard versus the nominal concentration of the analytes.
Results
TNB-R-138727 Disulfide. The thiol-exchange reaction of
R-138727 with DTNB (Fig. 2) and trapping the sulfenic acid inter-
mediate of R-138727 with TNB (Fig. 1) generate a mixed disulfide
between TNB and R-138727 (TNB-R-138727 disulfide). We synthe-
sized TNB-R-138727 disulfide as the reference standard for analysis
by mixing DTNB and R-138727 (Fig. 2) because TNB and sulfenic
acid are unstable and unavailable. Major reaction products appeared
as two chromatographic peaks, Peaks A and B. The MS and tandem
mass spectrometry (MS/MS) spectra of Peaks A and B had identical
fragmentation patterns as well as exact mass and relative intensity of
the fragment ions, which demonstrated that Peaks A and B are
diastereoisomers. The presence of the diastereoisomers is consistent
with the chemical structure of TNB-R-138727 disulfide, which pos-
sesses two asymmetric carbons adjacent to the sulfur and nitrogen
atoms. The (MϩH)ϩ ions observed in MS spectra of Peaks A and B
were detected with m/z 547. The MS/MS spectrum of the (MϩH)ϩ
ion with m/z 547 gave a series of fragment ions with m/z 206, 248, and
317, and the proposed assignments of the fragment ions are shown in
Fig. 4. The ion with m/z 317 was considered to be a fragment ion of
(MϩH)ϩ produced by the elimination of the disulfanyl nitrobenzoic
acid moiety as shown in proposed fragmentation in Fig. 4.
Calculation of Rate Constants for Formation of the Glutathione Con-
jugate Metabolite and R-138727. As one possible pathway, the kinetic model
for the formation of the glutathione conjugate of R-138727 and R-138727 from
R-95913 in a closed incubation system is shown in Fig. 3. The corresponding
differential equations for this model are as follows:
FIG. 3. Schematic diagram of the model for the formation of the glutathione
conjugate of R-138727 and R-138727 from R-95913 in a closed incubation system.
k1, rate constant of the formation of the glutathione conjugate of R-138727 from
R-95913; k2, rate constant of the formation of R-138727 from the glutathione
conjugate of R-138727; k3, rate constant of the formation of the glutathione
conjugate of R-138727 from R-138727; k4, rate constant of conversion of R-95913
We determined the NMR spectra of Peaks A and B, which were
mixed due to rapid epimerization to each other. 1H NMR (pyridine-d5)
of the compounds were as follows: 0.82 ppm (2H, m), 1.09 ppm (2H,
m), 2.07 ppm (1H, m), 2.38 ppm (2H, m), 2.81 ppm (2H, m), 3.36 ppm
to the metabolites other than the glutathione conjugate of R-138727 and R-138727. (2H, m), 4.96 and 4.98 ppm (total 1H, each s), 6.00 ppm (1H, bs), 6.26

MECHANISM FOR PRASUGREL ACTIVE METABOLITE FORMATION
901
FIG. 4. MS/MS spectra of Peak A (a) and Peak B (b) and proposed fragmentation in positive ion LC-ESI/MS.
and 6.32 ppm (total 1H, each s), 7.26 ppm (2H, m), 7.37 ppm (1H, m), thione. Incubation of R-95913 with glutathione (Fig. 7a), but not
7.65 ppm (1H, m), 7.93 ppm (2H, m), 8.37 ppm (1H, s), clearly without glutathione (Fig. 7b), showed immediate formation of the
demonstrating that Peaks A and B are the mixed disulfide of glutathione conjugate of R-138727, which was gradually decreased
R-138727 with TNB (Fig. 5).
and was replaced by the formation of R-138727. Although the thio-
Metabolism of R-95913 by cDNA-Expressed CYP3A4 Supple- lactone decreased rapidly in the absence of glutathione (Fig. 7b), no
mented with NADPH in the Presence of DTNB or TNB. The active metabolite was detected. These data may indicate that other
thiolactone R-95913 was incubated with cDNA-expressed CYP3A4 metabolic pathways may have been occurring simultaneously in the in
supplemented with the NADPH-generating system. DTNB or TNB vitro system. The glutathione conjugate of R-138727 when added to
was added to the incubation mixture to trap the thiol-metabolite the incubation mixture rapidly decreased and converted to R-138727
R-138727 (Fig. 2) or the sulfenic acid metabolite of R-138727 (Fig. in the presence of glutathione, but R-138727 formed very little in the
1), respectively, in the form of TNB-R-138727 disulfide. The con- absence of glutathione (Fig. 7c). R-138727 when added to the incu-
centrations of R-95913, R-138727, and TNB-R-138727 disulfide in bation mixture was minimally converted to the glutathione conjugate
the incubation mixture at various times as determined by LC/MS are in the incubation mixture even in the presence of glutathione (Fig. 7d),
shown in Fig. 6. The concentration of R-95913 gradually decreased and it remained stable in the presence/absence of glutathione (data not
along with increasing the incubation time in the presence of DTNB or shown).
TNB (Fig. 6, a or b, respectively). Upon incubating R-95913 in the
Effect of Human Liver Cytosol on the Formation of the Gluta-
presence of DTNB, the rate of the substrate decrease was slower than thione Conjugate of R-138727 and R-138727. R-95913 (10 ␮M)
in the presence of TNB (1.7 Ϯ 1.4 and 5.6 Ϯ 1.8 ␮M/min with DTNB was incubated in human liver microsomes (1 mg/ml protein) with or
and TNB, respectively, as mean Ϯ S.D. values), and this incubation without human liver cytosol fraction (1 mg/ml protein) in the presence
produced neither TNB-R-138727 disulfide nor free R-138727, which of 5 mM glutathione and an NADPH-generating system. The apparent
indicates no production of the thiol-metabolite R-138727 (Fig. 6a). On formation of the glutathione conjugate of R-138727 was remarkably
the other hand, incubation in the presence of TNB led to a production decreased when the human liver cytosol fraction was added to human
of TNB-R-138727 disulfide, indicating the formation of the sulfenic liver microsomes with the formation rates of the glutathione conjugate
acid intermediate (Fig. 6b). TCEP seemed to have no significant of R-138727 and R-138727 of 54.2 Ϯ 6.5 and 4.49 Ϯ 1.77 pmol/
effect, except that a very low amount of R-138727 was detected (data min/mg protein, respectively, in human liver microsomes and 9.28 Ϯ
not shown).
2.05 and 31.5 Ϯ 7.1 pmol/min/mg protein, respectively, in a mixture
Formation of the Glutathione Conjugate of R-138727 and of human liver microsomes and cytosols (expressed as mean Ϯ S.D.
R-138727 by cDNA-Expressed CYP3A4. R-95913, glutathione con- values) (Fig. 8). The simultaneous fitting exhibited a good fit to the
jugate of R-138727, or R-138727 was incubated with cDNA-ex- observed data. The rate constant of the formation of R-138727 from
pressed CYP3A4 in the presence of NADPH with or without gluta- the glutathione conjugate of R-138727 (k2) was substantially in-

902
HAGIHARA ET AL.
FIG. 5. 1H NMR spectra of Peaks A and B obtained by NMR analysis.
creased by the addition of human liver cytosols (0.143 and 2.82 sϪ1 TNB, which forms a disulfide with the sulfenic acid groups (Lin et al.,
without and with human liver cytosols, respectively) as shown in Fig. 8.
1975; Boschi-Muller et al., 2000; Poole and Ellis, 2002; Poole et al.,
2004). We also used DTNB, which forms a disulfide with thiol groups
by a thiol exchange reaction (Riddles et al., 1979; Boschi-Muller et
al., 2000), for R-138727 detection. A mixed disulfide of R-138727
with TNB (TNB-R-138727 disulfide) is a key substance in the present
study because TNB-R-138727 disulfide could be produced either by a
thiol exchange reaction between R-138727 and DTNB (Fig. 2) or by
a coupling reaction between the sulfenic acid derivative of R-138727
Discussion
We hypothesized that a sulfenic acid intermediate can be produced
from the thiolactone intermediate of prasugrel, R-95913, and con-
verted to a disulfide with glutathione in the process of producing the
pharmacologically active metabolite R-138727 (Fig. 1). The basis for
this hypothesis is that R-138727, the thiol metabolite, is not the
immediate metabolite of R-95913 but the sulfenic acid formed is and TNB (Fig. 1).
actually that of R-95913 (the reactant) and not of R-138727 (which is
a product formed one or two steps later). In the present study, as a tool
for trapping and detecting the sulfenic acid intermediate, we used
The sulfenic acid intermediate of R-138727 was detected as TNB-
R-138727 disulfide after incubation of R-95913 with CYP3A4 and
NADPH without glutathione (Fig. 6b), whereas R-138727 was not
observed under this condition (Fig. 6a), indicating that the sulfenic
acid of R-138727 is likely produced as an intermediate in the metab-
olism of R-95913 before the formation of R-138727. In addition, we
showed the formation of a glutathione conjugate from R-95913 by
CYP3A4 in the presence of glutathione (Fig. 7a) but not in the
absence of glutathione (Fig. 7b). Because the sulfenic acid interme-
diate was formed even in the absence of glutathione (Fig. 6b),
R-95913 is likely first oxidized to the sulfenic acid intermediate in a
glutathione-independent manner and then possibly converted to a
glutathione conjugate. Another glutathione molecule could then re-
duce the R-138727 glutathione conjugate to form R-138727. This
concept is supported by the data showing that the glutathione conju-
gate is rapidly converted to R-138727 in the presence of glutathione
(Fig. 7c), whereas R-138727 is minimally converted to its glutathione
conjugate (Fig. 7d). Addition of cytosolic fraction to liver microsomes
FIG. 6. Concentrations of R-95913 (F), R-138727 (छ), and TNB-R-138727 disul-
fide (Œ) versus time during metabolism of R-95913 by cDNA-expressed CYP3A4
supplemented with an NADPH-generation system in the presence of DTNB (a) or
TNB (b).

MECHANISM FOR PRASUGREL ACTIVE METABOLITE FORMATION
903
FIG. 7. Formation of R-138727 and the glutathione conjugate of
R-138727 by cDNA-expressed CYP3A4 in the presence or absence
of glutathione. A 1 ␮M R-95913 with (a) or without (b) 5 mM
glutathione, or 400 nM glutathione conjugate of R-138727 with/
without 5 mM glutathione (expressed as solid/dashed line, respec-
tively) (c) or 1 ␮M R-138727 with 5 mM glutathione (d) was
incubated with cDNA-expressed CYP3A4 and NAPDH for 5, 15,
30, 45, and 60 min. The concentrations of R-95913 (F), R-138727
(छ), and glutathione conjugate of R-138727 (ϫ) were determined
by LC-MS/MS. The graphs show the mean values of duplicated
data.
resulted in much lower formation of the glutathione conjugate with from the S-configuration of R-138727 in human liver microsomes
increased formation of R-138727 (Fig. 8), indicating that the gluta- (Kazui et al., 2008).
thione conjugate generated can be immediately reduced to R-138727
in the presence of cytosol.
Very recently, Dansette et al. (2009) proposed a scheme that
included sulfenic acid and glutathione conjugation after the formation
Reddy et al. (2005) determined that the formation of a thiol me- of thiolactones from clopidogrel and ticlopidine, which is similar to
tabolite from sulfenic acid could proceed via a reduction or dispro- the one proposed here for prasugrel. Differences in the active metab-
portionation step that would not be glutathione-dependent. This pro- olite formation from thienopyridines do exist, especially in the initial
cess is another possible metabolic pathway for prasugrel’s active step to form the respective thiolactones. For both clopidogrel and
metabolite formation in vivo, and it would support the reported in vivo ticlopidine, the first step in the formation of the thiolactone is an
metabolic data in animals and humans (Farid et al., 2007a; Smith oxidative step in which reactive intermediates are likely to be formed,
et al., 2007).
as evidenced by their mechanism-based inhibitory effects on the
The conjugation with glutathione does not fully explain the ob- activities of CYP2B6 and CYP2C19 (Ha-Duong et al., 2001; Nishiya
served stereoselectivity in prasugrel’s active metabolite that was ob- et al., 2009a,b). However, prasugrel is converted to the thiolactone by
served in vivo. In humans, greater concentrations of the R-configu- hydrolysis in the intestine during the absorption process through the
ration at the 4-position of R-138727 (Wickremsinhe et al., 2007) were action of carboxylesterases (Williams et al., 2008), and the key
observed in plasma. This in vivo stereoselectivity may be explained in prasugrel metabolites are not mechanism-based inhibitors (Rehmel et
part by the observed in vitro stereoselectivity in the P450-mediated al., 2006; Hagihara et al., 2008; Nishiya et al., 2009a,b). The active
oxidation of the thiolactone to R-138727 by cDNA-expressed P450 metabolite of clopidogrel is thought to be formed primarily in the liver
isoforms (Rehmel et al., 2006; Baker et al., 2008). In addition, (Savi et al., 1992; Farid et al., 2010), whereas the active metabolite of
stereoselective S-methylation of R-138727 was observed in vitro prasugrel is thought to be partly formed by intestinal CYP3A before
where the S-methylated metabolite of R-138727 was generated only the thiolactone reaches the liver (Farid et al., 2007, 2010).
FIG. 8. Formation of the glutathione conjugate
of R-138727 and R-138727 from R-95913 in
human liver microsomes and cytosols. R-95913
(10 ␮M) was incubated in 1 mg/ml human liver
microsomes (a) or 1 mg/ml human liver micro-
somes and cytosols (b) supplemented with 5
mM glutathione and an NADPH-generating
system. The concentrations of R-138727 (छ)
and glutathione conjugate of R-138727 (ϫ)
were determined over time (mean ϩ S.D., n ϭ
3). The solid lines represent the simultaneous
fitting curves of the concentration-time profiles
of R-138727 and the glutathione conjugate of
R-138727.

904
HAGIHARA ET AL.
ticlopidine and its derivatives of human liver cytochrome p450. Mechanism-based inactivation
of CYP 2C19 by ticlopidine. Adv Exp Med Biol 500:145–148.
Hagihara K, Nishiya Y, Kurihara A, Kazui M, Farid NA, and Ikeda T (2008) Comparison of
human cytochrome P450 inhibition by the thienopyridines prasugrel, clopidogrel, and ticlo-
pidine. Drug Metab Pharmacokinet 23:412–420.
In conclusion, the results of the present in vitro study have provided
insight into the bioactivation mechanisms of prasugrel, including a
sulfenic acid intermediate and possibly a glutathione conjugate, al-
though there are marked differences in thiolactone formation and site
of active metabolite formation among thienopyridines.
Han JC and Han GY (1994)
A procedure for quantitative determination of tris(2-
carboxyethyl)phosphine, an odorless reducing agent more stable and effective than dithiothre-
itol. Anal Biochem 220:5–10.
Kassahun K, Pearson PG, Tang W, McIntosh I, Leung K, Elmore C, Dean D, Wang R, Doss G,
and Baillie TA (2001) Studies on the metabolism of troglitazone to reactive intermediates in
vitro and in vivo. Evidence for novel biotransformation pathways involving quinone methide
formation and thiazolidinedione ring scission. Chem Res Toxicol 14:62–70.
Kazui M, Hagihara K, Farid NA, and Kurihara A (2008) Stereoselectivity of thiol S-methyl
transferase responsible for the methylation of the active metabolite of prasugrel. Drug Metab
Rev 40 (Suppl 3):101.
Acknowledgments. We thank Drs. Takashi Izumi and Takahiro
Murai (Daiichi Sankyo Co., Ltd.) and Drs. Mary Pat Knadler and
Steven A. Wrighton (Eli Lilly and Company) for helpful comments
and discussion on this study.
Kazui M, Ishizuka T, Yamamura N, Iwabuchi H, Kita J, Yoneda K, Kurihara A, Hirota T, and
Ikeda T (2000) Mechanism for metabolic activation of CS-747, a new thienopyridine anti-
platelet agent. 13th International Symposium on Microsomes and Drug Oxidations; 2001 Jul
10–14; Stresa, Italy. Clinical Pharmacology and Therapeutics, Alexandria, VA.
Lin WS, Armstrong DA, and Gaucher GM (1975) Formation and repair of papain sulfenic acid.
Can J Biochem 53:298–307.
Nishiya Y, Hagihara K, Kurihara A, Okudaira N, Farid NA, Okazaki O, and Ikeda T (2009a)
Comparison of mechanism-based inhibition of human cytochrome P450 2C19 by ticlopidine,
clopidogrel, and prasugrel. Xenobiotica 39:836–843.
Nishiya Y, Hagihara K, Ito T, Tajima M, Miura S, Kurihara A, Farid NA, and Ikeda T (2009b)
Mechanism-based inhibition of human cytochrome P450 2B6 by ticlopidine, clopidogrel, and
the thiolactone metabolite of prasugrel. Drug Metab Dispos 37:589–593.
Poole LB and Ellis HR (2002) Identification of cysteine sulfenic acid in AhpC of alkyl
hydroperoxide reductase. Methods Enzymol 348:122–136.
Drug Metabolism & Pharmacokinetics
Research Laboratories (K.H., M.K.,
A.K., H.I., M.I., O.O.) and Medicinal
Chemistry Research Laboratories I
(H.K., N.T.), Daiichi Sankyo Co., Ltd.,
Tokyo, Japan; Department of Drug
Disposition, Lilly Research
Laboratories, Eli Lilly and Company,
Indianapolis, Indiana (N.A.F.); and
Yokohama College of Pharmacy,
Yokohama, Japan (T.I.)
KATSUNOBU HAGIHARA
MIHO KAZUI
ATSUSHI KURIHARA
HARUO IWABUCHI
MINORU ISHIKAWA
HIROYUKI KOBAYASHI
NAOKI TANAKA
OSAMU OKAZAKI
NAGY A. FARID
TOSHIHIKO IKEDA
Poole LB, Karplus PA, and Claiborne A (2004) Protein sulfenic acids in redox signaling. Annu
Rev Pharmacol Toxicol 44:325–347.
Riddles PW, Blakeley RL, and Zerner B (1979) Ellman’s reagent: 5,5Ј-dithiobis(2-nitrobenzoic
acid)–a reexamination. Anal Biochem 94:75–81.
Reddy VB, Karanam BV, Gruber WL, Wallace MA, Vincent SH, Franklin RB, and Baillie TA
(2005) Mechanistic studies on the metabolic scission of thiazolidinedione derivatives to
acyclic thiols. Chem Res Toxicol 18:880–888.
Rehmel JL, Eckstein JA, Farid NA, Heim JB, Kasper SC, Kurihara A, Wrighton SA, and Ring
BJ (2006) Interactions of two major metabolites of prasugrel, a thienopyridine antiplatelet
agent, with the cytochromes P450. Drug Metab Dispos 34:600–607.
References
Baker JAR, Oluyedun OA, Farid NA, Ring BJ, Wrighton SA, Kurihara A, and Guo Y (2008)
Formation of the isomers of the active metabolite of prasugrel by allelic variants of the human
cytochrome P450 isozymes. Drug Metab Rev 40 (Suppl 3):322.
Boschi-Muller S, Azza S, Sanglier-Cianferani S, Talfournier F, Van Dorsselear A, and Branlant
G (2000) A sulfenic acid enzyme intermediate is involved in the catalytic mechanism of
peptide methionine sulfoxide reductase from Escherichia coli. J Biol Chem 275:35908–35913.
Dansette PM, Libraire J, Bertho G, and Mansuy D (2009) Metabolic oxidative cleavage of
thioesters: evidence for the formation of sulfenic acid intermediates in the bioactivation of the
antithrombotic prodrugs ticlopidine and clopidogrel. Chem Res Toxicol 22:369–373.
Decker CJ, Cashman JR, Sugiyama K, Maltby D, and Correia MA (1991) Formation of
glutathionyl-spironolactone disulfide by rat liver cytochromes P450 or hog liver flavin-
containing monooxygenases: a functional probe of two-electron oxidations of the thiosteroid?
Chem Res Toxicol 4:669–677.
Farid NA, Smith RL, Gillespie TA, Rash TJ, Blair PE, Kurihara A, and Goldberg MJ (2007a)
The disposition of prasugrel, a novel thienopyridine, in humans. Drug Metab Dispos 35:1096–
1104.
Farid NA, McIntosh M, Garofolo F, Wong E, Shwajch A, Kennedy M, Young M, Sarkar P,
Kawabata K, Takahashi M, et al. (2007b) Determination of the active and inactive metabolites
of prasugrel in human plasma by liquid chromatography/tandem mass spectrometry. Rapid
Commun Mass Spectrom 21:169–179.
Farid NA, Payne CD, Small DS, Winters KJ, Ernest CS 2nd, Brandt JT, Darstein C, Jakubowski
JA, and Salazar DE (2007) Cytochrome P4503A inhibition by ketoconazole affects prasugrel
and clopidogrel pharmacokinetics and pharmacodynamics differently. Clin Pharmacol Ther
81:735–741.
Farid NA, Kurihara A, and Wrighton SA (2010) The metabolism and disposition of the
thienopyridine antiplatelet drugs ticlopidine, clopidogrel, and prasugrel in humans. J Clin
Pharmacol 50:126–142.
Savi P, Herbert JM, Pflieger AM, Dol F, Delebassee D, Combalbert J, Defreyn G, and Maffrand
JP (1992) Importance of hepatic metabolism in the antiaggregating activity of the thienopy-
ridine clopidogrel. Biochem Pharmacol 44:527–532.
Savi P, Combalbert J, Gaich C, Rouchon MC, Maffrand JP, Berger Y, and Herbert JM (1994)
The antiaggregating activity of clopidogrel is due to a metabolic activation by the hepatic
cytochrome P450–1A. Thromb Haemost 72:313–317.
Smith RL, Gillespie TA, Rash TJ, Kurihara A, and Farid NA (2007) Disposition and metabolic
fate of prasugrel in mice, rats, and dogs. Xenobiotica 37:884–901.
Wickremsinhe ER, Tian Y, Ruterbories KJ, Verburg EM, Weerakkody GJ, Kurihara A, and Farid
NA (2007) Stereoselective metabolism of prasugrel in humans using a novel chiral liquid
chromatography-tandem mass spectrometry method. Drug Metab Dispos 35:917–921.
Williams ET, Jones KO, Ponsler GD, Lowery SM, Perkins EJ, Wrighton SA, Ruterbories KJ,
Kazui M, and Farid NA (2008) The biotransformation of prasugrel, a new thienopyridine
prodrug, by the human carboxylesterases 1 and 2. Drug Metab Dispos 36:1227–1232.
Wiviott SD, Braunwald E, McCabe CH, Montalescot G, Ruzyllo W, Gottlieb S, Neumann FJ,
Ardissino D, De Servi S, Murphy SA, et al. (2007) Prasugrel versus clopidogrel in patients
with acute coronary syndromes. N Engl J Med 357:2001–2015.
Address correspondence to: Katsunobu Hagihara, Drug Metabolism & Pharma-
cokinetics Research Laboratories, Daiichi Sankyo Co., Ltd., 1-2-58 Hiromachi, Shi-
nagawa-Ku, Tokyo, 140-8710, Japan. E-mail: hagihara.katsunobu.fc@daiichisankyo.
co.jp
Ha-Duong NT, Dijols S, Macherey AC, Dansette PM, and Mansuy D (2001) Inhibition by