Biosynthesis and identification of an N-oxide/N-glucuronide metabolite and first synthesis of an N-O-glucuronide metabolite of Lu AA21004

Article abstract of DOI:10.1124/dmd.111.040428

This article describes the biosynthesis and identification of a new class of metabolites, a piperazine N-oxide/N-glucuronide metabolite 4-[2-(2,4-dimethyl-phenylsulfanyl)-phenyl]-1-β-D-glucuronic acid-piperazine 1-oxide (4). The metabolite was found in urine and plasma from humans and animals dosed with 1-[2-(2,4-dimethylphenylsulfanyl)-phenyl]-piperazine hydrobromide (Lu AA21004, 1), as a novel multimodal antidepressant under development for treatment of depression. Human liver microsomes in combination with uridine 5′-diphosphoglucuronic acid were used as an in vitro system to generate enough material of 4 to perform one- and two-dimensional ¹H and ¹³C NMR experiments for structure elucidation. Based on rotating frame Overhauser enhancement spectroscopy NMR experiments, the distance correlation between a piperazine proton and the anomeric proton of the glucuronic acid moiety is of a magnitude similar to that of the H-3′ and H-5′ protons and can only be explained by proximity in space and the postulated structure (4). The structural analog, the N-O-glucuronic acid conjugate 6-{4-[2-(2,4-dimethyl-phenylsulfanyl)-phenyl]-piperazin-1-yloxy}-1- β-D-glucuronic acid (3) was also observed in biological samples from humans and animals and the first organic synthesis and structural identification of this metabolite is also reported. Treatment of the glucuronide metabolites 3 and 4 with β-glucuronidase gave mainly the expected hydrolysis product, the hydroxyl amine 4-[2-(2,4-dimethyl-phenylsulfanyl)-phenyl]-piperazin-1-ol (2). Copyright

Full text of DOI:10.1124/dmd.111.040428

0090-9556/11/3912-2264–2274$25.00
DRUG METABOLISM AND DISPOSITION
Copyright © 2011 by The American Society for Pharmacology and Experimental Therapeutics
DMD 39:2264–2274, 2011
Vol. 39, No. 12
40428/3730051
Printed in U.S.A.
Biosynthesis and Identification of an N-Oxide/N-Glucuronide
Metabolite and First Synthesis of an N-O-Glucuronide
Metabolite of Lu AA21004
Henriette Kold Uldam, Martin Juhl, Henrik Pedersen, and Lars Dalgaard
H. Lundbeck A/S, Copenhagen, Denmark
Received May 6, 2011; accepted September 6, 2011
ABSTRACT:
This article describes the biosynthesis and identification of a new spectroscopy NMR experiments, the distance correlation between
class of metabolites, a piperazine N-oxide/N-glucuronide metabo- a piperazine proton and the anomeric proton of the glucuronic acid
lite 4-[2-(2,4-dimethyl-phenylsulfanyl)-phenyl]-1-␤-D-glucuronic moiety is of a magnitude similar to that of the H-3؅ and H-5؅ protons
acid-piperazine 1-oxide (4). The metabolite was found in urine and and can only be explained by proximity in space and the postulated
plasma from humans and animals dosed with 1-[2-(2,4-dimethyl- structure (4). The structural analog, the N-O-glucuronic acid con-
phenylsulfanyl)-phenyl]-piperazine hydrobromide (Lu AA21004, 1), jugate 6-{4-[2-(2,4-dimethyl-phenylsulfanyl)-phenyl]-piperazin-1-
as a novel multimodal antidepressant under development for treat-
yloxy}-1-␤-D-glucuronic acid (3) was also observed in biological
ment of depression. Human liver microsomes in combination with samples from humans and animals and the first organic synthesis
uridine 5؅-diphosphoglucuronic acid were used as an in vitro sys- and structural identification of this metabolite is also reported. Treat-
tem to generate enough material of 4 to perform one- and two-
ment of the glucuronide metabolites 3 and 4 with ␤-glucuronidase
dimensional 1H and 13C NMR experiments for structure elucida- gave mainly the expected hydrolysis product, the hydroxyl amine
tion. Based on rotating frame Overhauser enhancement
4-[2-(2,4-dimethyl-phenylsulfanyl)-phenyl]-piperazin-1-ol (2).
Introduction
tertiary piperazines resulting in quaternary structures seems to occur
mainly in primates (Chiu and Huskey, 1998).
Drugs that are piperazine derivatives typically generate piperazine
ring carbon alcohols or metabolize at the piperazine nitrogen to
generate hydroxylamines/N-oxides as phase I metabolites and N-
glucuronides or N-O-glucuronides as phase II metabolites (Delbress-
ine et al., 1992; Kassahun et al., 1996; Schaber et al., 2001; Miller et
al., 2004). N-Glucuronides are common metabolic products of drugs
containing amino groups or aromatic nitrogen heterocycles and in
some cases tertiary, secondary, and primary amines form N-glucu-
ronides from a drug and its phase I metabolites (Dalgaard and Larsen,
1999). Porter et al. (1975) were the first to report the formation of
quaternary ammonium-linked glucuronide from a tertiary amine in
humans. The enzyme involved was later shown to be UGT1A4 (Green
and Tephly, 1996). In other cases with tetrazoles and triazoles,
UGT1A6 (Huskey et al., 1994a,b) or UGT1A9 (Omura et al., 2007)
was involved. In addition, N-glucuronides of pyridine (synthesis)
(Dalgaard, 1983)) and imidazoles (biosynthesis) (Vashishtha et al.,
2002)) are also known. It appears that liver microsomes from pri-
mates, guinea pigs, and rabbits form imidazole glucuronides more
readily than microsomes from rats and dogs. N-Glucuronidation of
Nϩ-Glucuronidation as a metabolic pathway and the stability and
spectroscopy of the glucuronides have been reviewed by Hawes
(1998). A convenient route of synthesis of quaternary ammonium
glucuronides has recently been described previously (Iddon et al.,
2010). In some cases it can be a challenge to distinguish whether
N- and O-glucuronidation has taken place, e.g., of a hydroxy
isoxazole (Andersen et al., 1989) and requires more extensive
investigation. The present study describes such a challenge. During
investigation of the disposition of 1-[2-(2,4-dimethyl-phenylsulfa-
nyl)-phenyl]-piperazine hydrobromide (Lu AA21004, 1) as a novel
multimodal antidepressant agent under development for treatment
of depression (Bang-Andersen et al., 2011) in humans, two major
metabolites were observed in plasma and urine. The molecular
weight and mass spectra of these metabolites indicate addition of
oxygen and glucuronic acid to the piperazine ring of the parent
drug. However, HPLC-MS data alone failed to provide sufficient
evidence of the exact structures. The present study describes the identification
of the two metabolites (2S,3S,4S,5R,6S)-6-{4-[2-(2,4-dimethyl-phenylsulfa-
nyl)-phenyl]-piperazin-1-yloxy}-3,4,5-trihydroxy-tetrahydro-pyran-2-
carboxylic acid (3) and 4-[2-(2,4-dimethyl-phenylsulfanyl)-phenyl]-1-␤-D-
glucuronic acid-piperazine 1-oxide (4) (Fig. 1) using a novel and
unambiguous organic synthesis and biosynthetic approach, respec-
Article, publication date, and citation information can be found at
http://dmd.aspetjournals.org.
doi:10.1124/dmd.111.040428.
ABBREVIATIONS: NMP, N-methyl-2-pyrrolidone; UGT, UDP-glucuronosyltransferase; Lu AA21004, 1-[2-(2,4-dimethyl-phenylsulfanyl)-phenyl]-
piperazine hydrobromide; HPLC, high-performance liquid chromatography; UDPGA, UDP-diphosphoglucuronic acid; MS, mass spectrometry;
DMSO, dimethyl sulfoxide; THF, tetrahydrofuran; AcOEt, ethyl acetate; CDCl3, chloroform-d; HRMS, high-resolution mass spectrometry; MeOH,
methanol; AcOH, acetic acid; TOCSY, total correlation spectroscopy; HSQC, heteronuclear single quantum coherence; HMBC, heteronuclear
multiple bond correlation; ROESY, rotating frame Overhauser enhancement spectroscopy.
2264

IDENTIFICATION OF AN N-OXIDE/N-GLUCURONIDE METABOLITE
2265
mobile phase B (acetonitrile-water, 90:10, 0.05% formic acid). Fractions
containing the glucuronide metabolite 4 with m/z 491 (electrospray-positive)
and retention time of approximately 11 min were collected and evaporated at
35°C under a stream of nitrogen. The final yield from the four incubations and
subsequent HPLC purifications was approximately 10 mg of glucuronide
metabolite 4 (yield 25%, needle-shaped crystalline material) from 25 mg of
starting material 2 (see Fig. 10).
Analytical HPLC-MS. Ion trap. The HPLC-MS system consisted of a
G1367 autosampler, a G1312A binary pump, a G1316A thermostatted column
compartment, a G1322A degasser (Agilent Technologies), and an LQC Deca
XP mass spectrometer (Thermo Fisher Scientific). Injections (10 ␮l) were
made on a Symmetry C18 column (2.1 ϫ 100 mm; Waters) eluted at 0.25
ml/min with a linear gradient applied (0–25 min) from 100 to 50% mobile
phase A (acetonitrile-50 mM ammonium acetate, pH 5.0, 20:80) and 0 to 50%
mobile phase B (acetonitrile-50 mM ammonium acetate, pH 5.0, 80:20)
followed by a washout (100% B, 5 min) and reequilibration (100% A, 6 min).
MicrOTOF. Accurate masses were obtained on an HPLC-MS system that
consisted of a G1313A autosampler, a G1311A quaternary pump, a G1316A
column oven, a G1322A degasser (Agilent Technologies) and a MicrOTOF
system (Bruker Daltonics, Bremen, Germany) using an electrospray ionization or
atmosphere pressure photoionization interface. Injections (5 ␮l) were made on a
SunFire C18 column (4.5 ϫ 50 mm, 3.5 ␮m; Waters) eluted at 2.5 ml/min with
a linear gradient applied (0–2 min) from 95 to 0% mobile phase A (water ϩ 0.05%
trifluoroacetic acid) and 5 to 100% mobile phase B (water-acetonitrile, 5:95, ϩ
0.03% trifluoroacetic acid) followed by washout (100% B, 0.5 min).
NMR Spectroscopy. NMR spectra were recorded at 11.75 or 14.10 T on
Bruker Instruments (Fa¨llanden, Switzerland). The instruments were an Avance I
500 equipped with a 5-mm four-nucleus probe (1H, 19F, 13C, 15N) with a Z-axis
gradient coil, an Avance DRX-500 equipped with a 1-mm triple resonance inverse
probe (1H, 13C, 15N) with a Z-axis gradient coil, or an Avance III 600 equipped
with a 5-mm triple resonance (1H, 13C, 15N) CryoProbe with a Z-axis gradient. The
experiments were run at room temperature (25°C). Tetramethyl silane or the
residual signal from deuterated solvent was used for internal reference. The
compounds were dissolved in deuterated solvents. A mixture of deuterium oxide,
acetonitrile-d3, and DMSO-d6 was used to dissolve 4.
Enzymatic Hydrolysis of N-O-glucuronide (3) and N-oxide/N-glucuro-
nide (4). The glucuronide metabolites 3 and 4 (10 ␮M) were each dissolved in
100 mM potassium dihydrogen phosphate buffer solution, pH 6.8, and added
to a ␤-glucuronidase solution (E. coli, 1000 IU/ml). The suspensions were
incubated at 37°C for 30 min for complete hydrolysis to mainly 2 and
secondarily 1. Hydrolysis of 3 and 4 (10 ␮M) spiked to human plasma required
higher amounts of ␤-glucuronidase (up to 5000 IU/ml) and/or longer incuba-
tion times (up to 4 h).
Synthesis of N-O-Glucuronide (3): Experimental. {[2-(2,4-Dimethyl-phe-
nylsulfanyl)-phenyl]-(2-hydroxy-ethyl)-amino}-ethanol (6). Aniline (5) (2.75
g, 12.0 mmol) was dissolved in N-methylpyrrolidinone (20 ml), and then
potassium iodide (4.13 g, 24.9 mmol) and ethyl bromoacetate (8.28 ml, 74.7
mmol) were added. The mixture was warmed to 120°C and stirred overnight
(22 h). Additional ethyl bromoacetate (2 ml, 18 mmol) was added, and the
mixture was stirred for an additional 4 h. The mixture was partitioned between
water (100 ml) and tert-butylmethyl ether (100 ml), and the organic phase was
washed with water (two 50-ml washes) and brine (50 ml). The combined
aqueous phases were extracted with tert-butylmethyl ether (50 ml), and the
combined organic phases were dried (Na2SO4), filtered and evaporated to
dryness in vacuo, affording a crude mixture of the corresponding diester,
which was taken directly to the next step.
FIG. 1. Structures of aryl piperazine antidepressant agent 1, hydroxylamine 2,
O-glucuronide of the hydroxylamine 3, and N-glucuronide of the N-oxide 4.
tively. Biosynthesis was performed using UDPGA and human liver
microsomes followed by semipreparative HPLC purification and suc-
cessful application of NMR spectroscopy. A sufficient amount of 4
was obtained to run NMR experiments in the carbon-observe mode,
resulting in NMR data that led to the unambiguous identification of
the structure of 4.
Materials and Methods
Chemicals. 1, 4-[2-(2,4-dimethyl-phenylsulfanyl)-phenyl]-piperazin-1-ol
(2) and 3 were obtained from H. Lundbeck A/S (Copenhagen, Denmark). The
purities of 1 and 2 were Ͼ99%. The purity of 3 was 81.6% (UV) and 100%
(evaporative light scattering). HPLC-grade solvents were from Thermo Fisher
Scientific (Waltham, MA) and Sigma-Aldrich (St. Louis, MO). Uridine 5Ј-
diphosphoglucuronic acid trisodium salt, alamethicin, D-saccharic acid 1,4-
lactone monohydrate, magnesium chloride, Tris-HCl, ␤-glucuronidase (type
VII-A from Escherichia coli), ammonium acetate, deuterium oxide, and
DMSO-d6 were purchased from Sigma-Aldrich. Potassium dihydrogen phos-
phate, sodium hydroxide, acetic acid, hydrochloride acid (2 N), ammonium
hydroxide solution (25% aqueous), potassium hydroxide, ammonium acetate,
and formic acid were purchased from Merck (Darmstadt, Germany). Acetoni-
trile-d3 was purchased from Cambridge Isotope Laboratories (Andover, MA).
Biological Preparations. Human liver microsomes, pool of mixed gender
(20 mg protein/ml), was purchased from XenoTech, LLC (Lenexa, KS).
Preparative Scale Biosynthesis of N-Oxide/N-Glucuronide (4). Pooled
human liver microsomes (mixed gender, final concentration, 0.5 mg of protein/
ml) were added to 100 ml of 100 mM Tris-HCl buffer, pH 7.4, containing 5
mM MgCl2, 2.5 mM UDPGA, 12.5 ␮g/ml alamethicin, 5 mM D-saccharic acid
1,4-lactone, and 200 ␮M hydroxyl amine 2. This suspension was incubated for
4 h at 37°C, and the reaction was stopped by the addition of ice-cold (Ϫ20°C)
acetonitrile (100 ml). To obtain enough glucuronide metabolite 4 for further
investigation, the biosynthesis was repeated four times. The ice-cold incuba-
tion suspension (200 ml) containing the glucuronide metabolite 4 was centri-
fuged (16,000g, 4°C, 10 min) and evaporated at 35°C under nitrogen in a
TurboVap LV evaporator (Zymark Corp., Hopkinton, MA) until concentrated
approximately 17-fold.
Lithium borohydride in THF (2 M, 50 ml) was added to the diester
at 0°C, and the mixture was stirred overnight (18 h). Then, 2 M NaOH
(150 ml) was added, and the mixture was stirred for 1ch and parti-
tioned between water (100 ml) and CH2Cl2 (150 ml). The aqueous
phase was extracted with CH2Cl2 (50 ml), and the combined organic
phases were dried (Na2SO4), filtered, and evaporated to dryness in
vacuo, affording a slightly yellow oil. The residue was purified using
silica gel column chromatography (AcOEt-heptane, 1:19–1:1) afford-
ing diol {[2-(2,4-dimethyl-phenylsulfanyl)-phenyl]-(2-hydroxy-ethyl)-
amino}-ethanol (6) (1.8 g, 46% over the two-step sequence) as a
Semipreparative HPLC-MS Purification. The HPLC-MS system con-
sisted of a G2258A dual-loop autosampler PS, a G1311A quaternary pump, a
G1316A thermostatted column compartment, a G1364B fraction collector prep
scale, and a G1956A LC/MSD Quad VL system (Agilent Technologies, Santa
Clara, CA). The concentrate (approximately 12 ml from the biosynthesis) was
diluted with 12 ml of mobile phase A and purified by five successive 5-ml
injections on a XBridge C18 column (10 ϫ 250 mm; Waters, Milford, MA)
eluted at 4 ml/min with a linear gradient applied (0–20 min) from 70 to 40%
mobile phase A (acetonitrile-water, 10:90, 0.05% formic acid) and 30 to 60% colorless oil.

2266
ULDAM ET AL.
FIG. 2. First synthesis of the N-O-glucuronide
metabolite (3). Conditions: (a) 1, BrCH2CO2Et,
KI, N-methyl-2-pyrrolidone; 2, LiBH4, THF
(56%, two steps); (b) SO3 ⅐ pyridine, DMSO,
CH2Cl2; (c) (CO)2Cl2, DMSO, NEt3, CH2Cl2
(d) N2H4, MeOH; (e) NaBH3CN, THF, AcOH
(11: 8%; 2: 31%, two steps from 6); (f) 1,
NaOMe, MeOH; 2, NaOH, acetone, THF (42%,
two steps).
Data for 6: 1H NMR (CDCl3, 500 M Hz) 7.44 (d, 1H, J ϭ 7.5 Hz),
was stirred under nitrogen for 30 min, triethylamine (3.95 ml, 28.4 mmol) was
added, and the mixture was slowly warmed to Ϫ20°C over 1 h. Then the
cooling bath was removed, and the mixture was stirred for 5 min at room
temperature. In a separate vial, MeOH (10 ml, 250 mmol) was added to
(2S,3S,4S,5R,6S)-3,4,5-triacetoxy-6-(1,3-dioxo-1,3-dihydro-isoindol-2-yloxy)-
tetrahydro-pyran-2-carboxylic acid methyl ester (10) (990 mg, 2.0 mmol),
followed by hydrazine monohydrate (1:1, hydrazine-water, 115.1 ␮l). After 5
min, the mixture was transferred to the vial containing the above-described
Swern oxidation mixture, followed by the addition of AcOH (5.8 ml) and
sodium cyanoborohydride (0.75 g, 12 mmol), and the mixture was stirred for
2 h. The mixture was concentrated to half-volume, AcOH (6 ml) and sodium
7.26 (d, 1H, J ϭ 8.0 Hz), 7.19 (s, 1H), 7.13 (t, 1H, J ϭ 7.5 Hz), 7.08
(d, 1H, J ϭ 8.0 Hz), 7.00 (t, 1H, J ϭ 7.5 Hz), 6.47 (d, 1H, J ϭ 7.5
Hz), 3.62 (br s, 4H), 3.24 (br s, 4H), 3.09 (br s, 2H), 2.39 (br s, 4H),
2.32 (br s, 4H). 13C NMR (CDCl3, 125 M Hz) 146.3, 142.8, 140.2,
138.7, 137.0, 132.0, 128.0, 127.0, 126.0, 125.1, 125.0, 60.1, 59.0,
21.2, 20.4. HRMS: [M ϩ H]ϩ ϭ 318.1522; found 318.1528.
(2S,3S,4S,5R,6S)-3,4,5-Triacetoxy-6-{4-[2-(2,4-dimethyl-phenylsulfanyl)-
phenyl]-piperazin-1-yloxy}-tetrahydro-pyran-2-carboxylic acid methyl ester
(11). Oxalyl chloride in CH2Cl2 (2 M, 3.54 ml) was added to CH2Cl2 (10 ml)
and cooled to Ϫ78°C, followed by the addition of DMSO (1.00 ml, 14.2
mmol), and the mixture was stirred for 15 min, and then diol 6 (900 mg, 2.835
mmol) in CH2Cl2 (1.0 ml ϩ 0.5 ml rinse) was added dropwise. The mixture
FIG. 3. Human urine from a subject dosed with 1. Metabolites 3 and 4 observed in
chromatogram with m/z values of 491.
FIG. 4. Ion trap MS2 spectra of metabolites 3 and 4.

IDENTIFICATION OF AN N-OXIDE/N-GLUCURONIDE METABOLITE
2267
FIG. 5. 1H NMR spectrum of hydroxyl amine 2.
cyanoborohydride (0.35g) were added, and the mixture was stirred at 45°C
overnight (19 h). The mixture was partitioned between AcOEt (20 ml) and
affording an off-white foam (245 mg). The product was precipitated from diethyl
ether (1 ml), and the white solid was a 2:1 mixture of by-product (from the
saturated aqueous NaHCO3 (20 ml), and the organic phase was washed with phthalimide) and desired product, whereas the filtrate was an impure mixture of
saturated aqueous NaHCO3 (two 10-ml washes). The combined aqueous desired product 11, hydroxyl amine 2, and unknown phthalimide by-products. The
phases were extracted with AcOEt (10 ml), and the combined organic phases
were dried (Na2SO4), filtered, and evaporated to dryness in vacuo. The residue
was subjected to silica gel column chromatography (AcOEt-heptane, 1:9–1:0)
filtrate was evaporated to dryness and subjected to silica gel column chromatog-
raphy (MeOH-CH2Cl2, 1:99–1:9), which removed the hydroxylamine 2 (200.3
mg, 31%), but not the other by-products. The residue was then subjected to
FIG. 6. 13C NMR spectrum of hydroxyl amine 2.

2268
ULDAM ET AL.
1 h. Then Amberlyst 15 (approximately 10 beads) was added, and the mixture
TABLE 1
1H NMR chemical shift assignments of 2, 3, and 4 (␦, multiplicity, integration)
was concentrated and purified directly using silica gel column chromatography
(MeOH-CH2Cl2, 2:98–10:90), affording the corresponding methyl ester (52
mg, 77%). The methyl ester (25 mg, 0.050 mmol) was dissolved in acetone
(2.5 ml, 34 mmol), 1.0 M sodium hydroxide in water (0.248 ml) was added,
resulting in a white precipitate (presumably the sodium salt), and the reaction
mixture was stirred for 5 min. Subsequently, the solvent was decanted off. The
residue was dissolved in MeOH-DMSO-H2O (1:1:4) and filtered. The clear
solution was subjected to preparative HPLC, affording N-O-glucuronide 3
(13.2 mg, 54%, solid) (Fig. 2).
A solvent mixture of D2O/CD3CN/DMSO-d6 (2/3.5/6) was used because of low solubility
of 4. Assignment of H and C atoms was based on 1H, decoupled 13C, TOCSY, ROESY,
HSQC, and HMBC experiments.
Position
2
3
4
Aromatic ring
H-3
H-5
H-6
H-8
H-9
H-10
H-11
C-2-CH3
7.17 (s, 1H)
7.03 (d, 1H)
7.27 (d, 1H)
6.39 (d, 1H)
6.84 (t, 1H)
7.08 (t, 1H)
7.07 (d, 1H)
2.19 (s, 3H)
2.27 (s, 3H)
7.12 (s, 1H)
6.97 (d, 1H)
7.21 (d, 1H)
6.36 (d, 1H)
6.80 (t, 1H)
7.04 (t, 1H)
7.03 (d, 1H)
2.19(s, 3H)
2.25 (s, 3H)
7.20 (s, 1H)
7.06 (d, 1H)
7.30 (d, 1H)
6.40 (d, 1H)
6.92 (t, 1H)
7.12 (t, 1H)
7.21 (d, 1H)
2.20 (s, 3H)
2.28 (s, 3H)
Results
Dosing of humans orally with 1 afforded a number of metabolites
in plasma and urine, including two compounds verified to be 3 and 4,
both with (M ϩ H)ϩ ϭ m/z 491 using HPLC-MS (Fig. 3). Fragmen-
tation of the m/z 491 ions in the ion trap mass spectrometer resulted
in the MS2 spectra shown in Fig. 4. The ions at m/z 315, 299, 298, and
240 for 4 and 299, 298, 256, and 240 for 3 correspond to the loss of
a glucuronic acid moiety (m/z 315), loss of a glucuronic acid moiety
and oxygen (m/z 299), and loss of an O-glucuronic acid moiety and
fragmentation of the piperazine ring (m/z 240 and 256). This pattern
is consistent with the fragmentation of glucuronic acid conjugate
metabolites of a monooxygenated piperazine derivative (Delbressine
et al., 1992; Schaber et al., 2001; Miller et al., 2004). Analysis of the
(bio)synthesized standards containing 3 and 4 using MicrOTOF MS
showed the compounds to have exact mass values of 491.18608 and
491.18325, respectively. The observed values are within 2.9 and 2.8
ppm (for 3 and 4, respectively) of a theoretical exact mass of
C24H31N2O7S, corresponding to a monooxygenated glucuronide me-
tabolite of 1. Analysis of the starting material 2, the hydroxylamine,
C-4-CH3
Glucuronic acid
1Ј
2Ј
3Ј
4Ј
5Ј
4.57a (d, 1H)
3.04 (m, 1H)
3.22 (m, 1H)
3.33 (m, 1H)
3.65 (d, 1H)
4.93b (d, 1H)
3.81 (m, 1H)
3.44 (m, 1H)
3.43 (m, 1H)
4.02 (d, 1H)
Piperazine ring
2Љeq
2Љax
3Љeq
3Љax
5Љeq
5Љax
6Љeq
6Љax
3.12 (br d, 1H)
2.66 (br t, 1H)
3.19 (br d, 1H)
2.85 (br t, 1H)
3.19 (br d, 1H)
2.85 (br t, 1H)
3.12 (br d, 1H)
2.66 (br t, 1H)
2.80 (m, 1H)
3.17 (m, 1H)
2.80 (m, 1H)
3.28 (m, 1H)
2.80 (m, 1H)
3.28 (m, 1H)
2.80 (m, 1H)
3.17 (m, 1H)
3.29 (m, 1H)
3.42 (m, 1H)
3.79 (m, 1H)
3.68 (m, 1H)
3.64 (m, 1H)
3.65 (m, 1H)
3.29 (m, 1H)
3.42 (m, 1H)
a JH-1Ј, H-2Ј of 8.5 Hz.
b JH-1Ј, H-2Ј of 6.6 Hz.
preparative thin-layer chromatography (MeOH-CH2Cl2, 3:97), which removed the
phthalimide by-product. Finally, the mixture was precipitated in diethyl ether (0.5 for the biosynthesis of 4, shows the compound to have an exact mass
ml at Ϫ20°C), affording glucuronic methyl ester 11 (105 mg, yield ϭ 8.1%).
Data for 2: 1H NMR (CDCl3, 500 M Hz) 8.11 (s, 1H), 7.35 (s, 1H, J ϭ 7.5
Hz), 7.25 (s, 1H), 7.16 (d, 1H, J ϭ 7.5 Hz), 7.13–7.09 (m, 2H), 6.92 (t, 1H, J ϭ
7.5 Hz), 6.41 (d, 1H, J ϭ 7.5 Hz), 3.26 (d, 2H, J ϭ 10.5 Hz), 3.14 (d, 2H, J ϭ
10.5 Hz), 2.89 (t, 2H, J ϭ 11.0 Hz), 2.68 (t, 2H, J ϭ 9.5 Hz), 2.34 (s, 3H), 2.26
(s, 3H). 13C NMR (CDCl3, 125 M Hz) 150.3, 143.4, 140.3, 137.4, 135.0,
133.4, 129.8, 129.0, 127.6, 127.4, 126.1, 121.9, 59.8, 51.8, 22.5, 21.8. HRMS:
[M ϩ H]ϩ ϭ 315.15256; found 315.15177.
value of 315.15177. The observed value is within 2.5 ppm of the
TABLE 2
13C NMR chemical shift assignments of 2, 3, and 4 (␦)
A
solvent mixture of D2O/CD3CN/DMSO-d6 (2/3.5/6) was used because of the low
solubility of 4. Assignment of H and C atoms was based on 1H, decoupled 13C, TOCSY,
ROESY, HSQC, and HMBC experiments. The 13C signals at 45.6 and 45.9 (C-2Љ and C-6Љ)
as well as the signals at 60.1 and 60.6 (C-3Љ and C-5Љ) could be interchangeable.
1
Data for 11: H NMR (CDCl3, 600 M Hz) 7.35 (d, 1H, J ϭ 8.5 Hz), 7.14
Position
2
3
4
(s, 1H), 7.07–7.01 (m, 3H), 6.86 (dt, 1H, J ϭ 1.5, 7.5 Hz), 6.40 (dd, 1H, J ϭ
1.5, 8.0 Hz), 5.28 (t, 1H, J ϭ 9.5 Hz), 5.19 (t, 1H, J ϭ 9.5 Hz), 5.06 (dd, 1H,
J ϭ 8.5, 9.5 Hz), 4.96 (d, 1H, J ϭ 8.5 Hz), 4.09 (d, 1H, J ϭ 10.0 Hz), 3.76
(s, 3H) 3.47 (br. d, 1H, J ϭ 9.0 Hz), 3.34–3.27 (m, 3H), 3.14 (br. s, 1H),
3.03–2.95 (m, 3H), 2.36 (s, 3H), 2.30 (s, 3H), 2.06 (s, 3H), 2.04 (s, 3H), 2.03
(s, 3H). 13C NMR (CDCl3, 150 M Hz) 170.2, 169.5, 169.1, 167.3, 142.4,
139.3, 136.2, 134.7, 131.7, 127.8, 126.3, 125.5, 124.6, 119.9, 102.8, 72.5, 72.3,
69.7, 69.5, 52.9, 21.2, 20.7, 20.6, 20.5. HRMS: [M ϩ H]ϩ ϭ 631.2320; found
631.2321.
4-[2-(2,4-Dimethyl-phenylsulfanyl)-phenyl]-morpholin-2-one (7). Diol 6
(70.1 mg, 0.221 mmol) was dissolved in CH2Cl2 (0.80 ml), DMSO (0.150 ml, 2.11
mmol) and N,N-diisopropylethylamine (115 ␮l, 0.662 mmol) were added, and the
reaction mixture was stirred at Ϫ10°C under nitrogen. Sulfur trioxide ⅐ pyridine (87.9
mg) was dissolved in DMSO (0.300 ml, 4.23 mmol) and added to the reaction mixture.
The mixture was stirred for 30 min, while warmed to 0°C. The mixture was purified
by directly pouring it onto a silica gel column (AcOEt-heptane, 1:4) affording the
undesired lactone 7 (51 mg, 74%) as a colorless oil.
Aromatic ring
C-1
C-2
C-3
C-4
C-5
C-6
C-7
C-8
C-9
C-10
C-11
C-12
C-2-HH3
C-4-HH3
Glucuronic acid
C-1Ј
C-2Ј
C-3Ј
C-4Ј
C-5Ј
COOH
Piperazine ring
C-2Љ
C-3Љ
C-5Љ
C-6Љ
128.1
142.6
132.4
140.2
128.7
136.4
134.1
126.8
125.2
126.7
120.9
149.4
20.6
128.1
142.5
132.5
140.1
128.9
136.3
134.0
126.9
125.2
126.9
121.1
149.2
20.6
127.5
142.9
132.6
140.5
128.9
136.7
134.3
126.6
126.1
127.0
121.3
148.0
20.8
21.2
21.2
21.4
105.7
72.2
76.6
72.0
75.9
171.3
101.5
69.1
74.9
71.3
81.0
174.0
Data for 7: 1H NMR (CDCl3, 500 M Hz) 7.34 (d, 1H, J ϭ 8.0 Hz), 7.16 (s,
1H), 7.12 (t, 1H, J ϭ 7.5 Hz), 7.05–7.02 (m, 2H), 6.95 (t, 1H, J ϭ 8.0 Hz), 6.59
(d, 1H, J ϭ 8.0 Hz), 4.47 (t, 2H, J ϭ 4.5 Hz), 4.00 (s, 2H) 3.36 (t, 2H, J ϭ 4.5
Hz), 2.37 (impurity), 2.31 (impurity). 13C NMR (CDCl3, 125 M Hz) 167.5,
145.6, 142.3, 139.5, 136.0, 134.8, 131.9, 128.0, 126.9, 125.8, 125.7, 119.8,
68.9, 53.6, 48.5.
N-O-Glucuronide 3. The tetra acetate 11 (84.1 mg, 0.133 mmol) was
dissolved in MeOH (3.0 ml, 74 mmol), 5.4 M sodium methoxide in MeOH
(24.69 ␮l) was added, and the reaction mixture was stirred under nitrogen for
50.8
58.5
58.5
50.8
50.6
56.4
58.0
50.6
45.6a
60.1b
60.6b
45.9a
a JH-1, H-2Ј of 8.5 Hz.
b JH-1Ј, H-2 of 6.6 Hz or reverse.

IDENTIFICATION OF AN N-OXIDE/N-GLUCURONIDE METABOLITE
2269
theoretical exact mass of C18H23N2OS, corresponding to a monooxy- and two-dimensional techniques (TOCSY) (Bax and Davis, 1985) for
1
genated metabolite of 1. H and 13C NMR spectra were obtained for
1H-1H correlations in coupling networks, gradient selected HSQC
(Kay et al., 1992; Willker et al., 1993) for direct correlations between
1H and 13C, gradient selected HMBC (Bax and Davis, 1985;
Willker et al., 1993; Cicero et al., 2001) for long-range correlations
between 1H and 13C, and ROESY (Bax and Davis, 1985; Hwang
and Shaka, 1992) for correlating protons close in space). See 1H
and 13C NMR spectra in Figs. 11 and 12. The anomeric proton of
the glucuronic acid moiety was observed at 4.93 ppm and with a
coupling constant of JH-1Ј, H-2Ј of 6.6 Hz. See Tables 1 and 2 for
chemical shift values for all hydrogen and carbon atoms in com-
pounds 2, 3, and 4.
2 for comparative reasons (Figs. 5 and 6).
Metabolite 3 was unambiguously prepared using a novel organic
synthesis, and various NMR experiments were performed for struc-
tural identification purposes. See Tables 1 and 2 for chemical shift
values and Figs. 7 and 8 for H and 13C NMR spectra.
1
Treatment of 3 and 4 with ␤-glucuronidase resulted in hydrolysis
mainly to a compound with (M ϩ H)ϩ ϭ m/z 315 and to a lesser
extent to a compound with (M ϩ H)ϩ ϭ m/z 299. The retention times
and MS spectra were consistent with those obtained for the hydroxyl
amine 2 and the parent drug 1.
Organic synthesis of 4 was unsuccessful, but in vitro biosynthesis
using human liver microsomes, UDPGA, alamethicin (antibiotic, precau-
tionary measure) and saccharic acid lactone (␤-glucuronidase inhibitor,
precautionary measure) (Miller et al., 2004; Picard et al., 2005; Bowal-
gaha et al., 2007; Wen et al., 2007a,b; Hintikka et al., 2008) and subse-
quent semipreparative HPLC-MS resulted in the production of approxi-
mately 10 mg of the metabolite 4. During evaporation of the fractions
collected during semipreparative HPLC-MS, 4 started precipitating and
was collected in sufficiently pure form for one- and two-dimensional
NMR experiments. The isolated metabolite was confirmed to be identical
to the metabolite (M ϩ H)ϩ ϭ 491 observed in various in vitro and in
vivo samples, including human urine, based on identical retention times
and MS fragmentation pattern (Figs. 9 and 10).
Discussion
Hydroxyl amines are the intermediate step toward N-O-glucu-
ronides, and 2 was the obvious starting point of the biosynthesis of 3
and 4 because they both were hydrolyzed to the hydroxyl amine with
␤-glucuronidase. The further reaction of this substrate with human
liver microsomes and UDPGA resulted in two glucuronides, both with
(M ϩ H)ϩ ϭ m/z 491 observed in electrospray-positive mode in the
mass spectrometer. One glucuronide was identified to be the N-O-
glucuronide 3 on the basis of identical retention times and MS
spectrum in the HPLC MS analysis of the standard and the biosyn-
thetic product.
The second metabolite with (M ϩ H)ϩ ϭ m/z 491 was speculated to be
The structure of 4, including assignment of all protons, was deter-
mined using a combination of one-dimensional (1H, decoupled 13C) the N-O,N-glucuronide 4. Other research groups have speculated about the
FIG. 7. 1H NMR spectrum of N-O-glucuronide metabolite 3.

2270
ULDAM ET AL.
FIG. 8. 13C NMR spectrum of N-O-glucuro-
nide metabolite 3.
possibility of an N-oxide/N-glucuronide metabolite, but none have described 8.4–9 Hz) (Straub et al., 1988; Delbressine et al., 1992; Turgeon et al.,
such a metabolite (Straub et al., 1988; Miller et al., 2004). 1992; Schaber et al., 2001; and Miller et al., 2004). The N-O-
The exact mass data obtained for 3 and 4 (corresponding to glucuronide 3 fits the published range. For quaternary N-glucuronides
C24H31N2O7S), and their fragmentation patterns are consistent with chemical shift values for the anomeric protons of 4.39 to 4.6 ppm with
monooxygenated glucuronide metabolites of the piperazine 1 (Del- coupling constants of approximately 9 Hz have been reported (Breyer-
bressine et al., 1992; Schaber et al., 2001; and Miller et al., 2004).
Pfaff et al., 1990; Kaku et al., 2004). The proposed N-oxide/N-
The anomeric protons of 3 and 4 were observed at 4.57 and 4.93 glucuronide 4 does not completely fit any of the ranges. The final
ppm and with coupling constants JH-1Ј, H-2Ј of 8.5 and 6.6 Hz, respec- evidence for the proposed structure can be found in the distance
tively. Chemical shifts and coupling constants of tertiary N-glucu- correlations observed in the ROESY spectra (Hwang and Shaka,
ronides and ␤-anomers are found at 5.33 to 6.31 ppm and 8.3 to 9.5 1992) of 3 and 4. See Figs. 13 and 14 and Table 1 for chemical shift
Hz, respectively (Andersen et al., 1989; Keating et al., 2006; Naka- values.
zawa et al., 2006; Yan et al., 2006), which are at lower field (higher
The lower graph in each spectrum shows the intensity of the
parts per million) than the anomeric proton of N-O-glucuronides distance correlations between the anomeric proton and the protons
(range of chemical shift 4.54–4.84 ppm and coupling constants of close in space. For 3, the correlation peaks of H-1Ј (4.57 ppm), H-3Ј
FIG. 9. Biosynthesis of 4, incubating 2 (200 ␮M)
with human liver microsomes (0.5 mg of protein/
ml) and UDPGA (2.5 mM) at 37°C for 4 h.

IDENTIFICATION OF AN N-OXIDE/N-GLUCURONIDE METABOLITE
2271
FIG. 10. Product
4
after purification using
semipreparative HPLC-MS.
(3.22 ppm), and H-5Ј (3.65 ppm) in the glucuronide moiety are N-O-hydroxylamine moiety as early as possible from starting mate-
equivalent in size, whereas a weaker correlation is seen to H-2Ј. As rials known from the published literature. Therefore, phthalimide (10)
expected, no correlations to the piperazine protons are observed for was chosen as the starting material (Mitchell and Whitcombe, 2000).
the anomeric proton in 3. However, for 4, apart from the expected The phthalimide hydroxylamine-protecting group was removed via
correlations between H-1Ј (4.93 ppm), H-3Ј (3.44 ppm), and H-5Ј brief exposure to hydrazine in MeOH and immediately transferred to
(4.02 ppm) of the glucuronide moiety, an additional distance corre- the next reaction because of its inherent instability under the reaction
lation is observed between H-1Ј and the axial piperazine proton conditions. The free hydroxylamine 9 was coupled with dialdehyde 8,
H-3Љax (3.68 ppm). This latter correlation shows that the glucuronide using sodium cyanoborohydride, thereby affording 11 via a double
and piperazine moieties have a spatial distance that can only be reductive amination. The major product of the reaction was the
explained by the structure shown in Fig. 11. The N-oxide/N-glucuro- hydroxylamine 2, which was used in the biosynthesis of 4. Several
nide metabolite 4 represents a new class of quaternary Nϩ-glucu- conditions were screened to oxidize diol 6 to the corresponding
ronides. It should be taken into account in future studies of conjugated dialdehyde 8. Surprisingly, when the oxidation was attempted using
hydroxylamines that unsymmetrically substituted amines will afford Parikh-Doering conditions (i.e., SO3 ⅐ pyridine, DMSO, and N,N-
diastereomeric conjugates.
diisopropylethylamine) (Parikh and Doering, 1967), lactone 7 was
There is also indirect evidence for the structure of 4 based on afforded instead of the desired dialdehyde 8. This problem was
different theoretical structures. In particular, differentiating between 3 overcome by using a Swern oxidation (Tidwell, 1990). One could
and 4 was deemed difficult; therefore, a structurally unambiguous possibly envision epimerization at the anomeric position under the
synthesis of 3 was developed. The strategy was to incorporate the slightly acidic conditions during the reductive amination, but because
FIG. 11. 1H NMR spectrum of N-oxide/N-
glucuronide metabolite 4.

2272
ULDAM ET AL.
FIG. 12. 13C NMR spectrum of N-oxide/N-glucuronide metabolite 4.
of anchimeric assistance from the neighboring acetyl moiety, the be linked to the N-hydroxylamine moiety. A distance correlation
␤-anomer is favored. Coupling product 11 was then deprotected using would not have been observed between the anomeric proton of the
a two-step protocol affording the N-O-glucuronide 3. The synthesis glucuronide and the piperazine ring protons in this structure. How-
unambiguously afforded the N-O-glucuronide, thus providing direct ever, a correlation is observed and the structure can thus be rejected.
identification of metabolite 3 as well as indirect identification of
metabolite 4.
A second alternative structure could theoretically have been the
␣-anomer of 3. However, in this case the coupling constant of the
One could speculate that 4 might be a “3 carboxylic ester alterna- anomeric proton in the glucuronide moiety would be only 2 to 3 Hz
tive.” In this case, the carboxylic moiety of the glucuronic acid would (Smith and Benet, 1986), and the actual observed coupling constant is
FIG. 13. Slice of ROESY spectrum of 3 show-
ing distance correlations between anomeric pro-
ton (H-1Ј) at ␦ 4.57 ppm and H-3Ј (␦ 3.22 ppm),
H-5Ј (␦ 3.65 ppm), and H-2Ј (␦ 3.04 ppm).

IDENTIFICATION OF AN N-OXIDE/N-GLUCURONIDE METABOLITE
2273
FIG. 14. Slice of ROESY spectrum of 4 show-
ing distance correlations between anomeric pro-
ton (H-1Ј) at ␦ 4.93 ppm and H-3Ј (␦ 3.44 ppm),
H-5Ј (␦ 4.02 ppm), and piperazine proton
H-3Љax (␦ 3.68 ppm) (bottom) and the normal
1H spectrum (top).
Performed data analysis: Uldam, Juhl, and Pedersen.
Wrote or contributed to the writing of the manuscript: Uldam, Juhl, Ped-
ersen, and Dalgaard.
7 to 8 Hz, indicative of the ␤-anomer, together with the fact that
strong distance correlations from H-1Ј to H-3Ј and H-5Ј are seen.
A third alternative structure is the N-1 piperazine glucuronide
instead of the observed N-4 substituted version. If this were the case,
the resulting quaternary N would result in a downfield shift of adja-
cent carbons of the piperazine ring (Miller et al., 2004). This is not
seen. In addition, distance correlations between the glucuronide pro-
tons (anomeric proton), piperazine protons H-2Љ and/or H-6Љ, and an
adjacent aromatic ring would have been observed if 4 had an N-1
glucuronide structure. None of this was observed.
References
Andersen JV, Dalgaard L, and Hansen SH (1989) Enzymic synthesis of two glucuronides of the
hydroxyisoxazole GABA-agonist, THIP, and the in vivo glucuronidation of THIP in rat.
Xenobiotica 19:1399–1406.
Bang-Andersen B, Ruhland T, Jørgensen M, Smith G, Frederiksen K, Jensen KG, Zhong H,
Nielsen SM, Hogg S, Mørk A, et al. (2011) Discovery of 1-[2-(2,4-dimethylphenylsulfanyl)
phenyl]piperazine (Lu AA21004): a novel multimodal compound for the treatment of major
depressive disorder. J Med Chem 54:3206–3221.
Bax A and Davis DG (1985) MLEV-17-based two-dimensional homonuclear magnetization
transfer spectroscopy. J Magn Reson 65:355–360.
The last speculated alternative structure is the N-1 oxide N-4
glucuronide, but the shift considerations of the third alternative also
apply here.
Bowalgaha K, Elliot DJ, Mackenzie PI, Knights KM, and Miners JO (2007) The glucuronidation
of ⌬4-3-keto C19- and C21-hydroxysteroids by human liver microsomal and recombinant
UDP-glucuronosyltransferases (UGTs): 6␣- and 21-hydroxyprogesterone are selective sub-
strates for UGT2B7. Drug Metab Dispos 35:363–370.
Breyer-Pfaff U, Becher B, Nusser E, Nill K, Baier-Weber B, Zaunbrecher D, Wachsmuth H, and
Prox A (1990) Quaternary N-glucuronides of 10-hydroxylated amitriptyline metabolites in
human urine. Xenobiotica 20:727–738.
During the biosynthesis, 1 was observed in the reaction mixture,
suggesting that either the substrate 2 or one of the products 3 or 4 is
hydrolyzed to give 1. It is most likely to arise from the nonenzymatic
hydrolysis of either the N–O bond in the glucuronide metabolite 3 or
the C–O bond in 2 in the reaction mixture (pH 6.8) or during the acidic
HPLC-MS conditions (Turgeon et al., 1992; Yan et al., 2006). Enzy-
matic and chemical hydrolysis of quaternary ammonium-linked gluc-
uronide metabolites, such as 4, is generally reported to be mediated
only by E. coli (pH 6.5–7.4) and nonenzymatically only at extreme pH
(Kowalczyk et al., 2000), in line with the suggestion above that the
parent compound 1 originates from hydrolysis of 2 or 3.
In conclusion, it is worth noting that during the analysis of samples
from humans dosed with 1 and from the biosynthesis previously
described, both the N-O-glucuronide 3 and the N-oxide/N-glucuronide
4 were formed. Furthermore, the N-oxide/N-glucuronide 4 is a new
class of metabolite that will possibly be observed in future metabolism
studies of other drugs.
Chiu SH and Huskey SW (1998) Species differences in N-glucuronidation. Drug Metab Dispos
26:838–847.
Cicero DO, Barbato G, and Bazzo R (2001) Sensitivity enhancement of a two-dimensional
experiment for the measurement of heteronuclear long-range coupling constants, by a new
scheme of coherence selection by gradients. J Magn Reson 148:209–213.
Dalgaard L (1983) Preparation and hydrolysis of N-glucuronides of pyridine and piperidine. Acta
Chem Scand B 37:923–928.
Dalgaard L and Larsen C (1999) Metabolism and excretion of citalopram in man: identification
of O-acyl- and N-glucuronides. Xenobiotica 29:1033–1041.
Delbressine LP, Moonen ME, Kaspersen FM, Jacobs PL, and Wagenaars GL (1992) Biotrans-
formation of mianserin in laboratory animals and man. Xenobiotica 22:227–236.
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.
Hawes EM (1998) Nϩ-glucuronidation, a common pathway in human metabolism of drugs with
a tertiary amine group. Drug Metab Dispos 26:830–837.
Hintikka L, Kuuranne T, Aitio O, Thevis M, Scha¨nzer W, and Kostiainen R (2008) Enzyme-
assisted synthesis and structure characterization of glucuronide conjugates of eleven anabolic
steroid metabolites. Steroids 73:257–265.
Huskey SW, Doss GA, Miller RR, Schoen WR, and Chiu SH (1994a) N-glucuronidation
reactions. II. Relative N-glucuronidation reactivity of methylbiphenyl tetrazole, methylbiphe-
nyl triazole, and methylbiphenyl imidazole in rat, monkey, and human hepatic microsomes.
Drug Metab Dispos 22:651–658.
Huskey SE, Magdalou J, Ouzzine M, Siest G, and Chiu SH (1994b) N-glucuronidation reactions.
III. Regioselectivity of N-glucuronidation of methylbiphenyl tetrazole, methylbiphenyl tria-
zole, and methylbiphenyl imidazole using human and rat recombinant UDP-glucuronosyl-
transferases stably expressed in V79 cells. Drug Metab Dispos 22:659–662.
Hwang TL and Shaka AJ (1992) Cross-relaxation without TOCSY: transverse rotating-frame
Overhauser effect spectroscopy. J Am Chem Soc 114:3157–3159.
Acknowledgments
We thank M. R. Willadsen for technical assistance with microsome incu-
bations.
Authorship Contributions
Iddon L, Bragg RA, Harding JR, and Stachulski AV (2010) A convenient new synthesis of
quaternary ammonium glucuronides of drug molecules. Tetrahedron 66:537–541.
Kaku T, Ogura K, Nishiyama T, Ohnuma T, Muro K, and Hiratsuka A (2004) Quaternary
Participated in research design: Uldam, Juhl, and Dalgaard.
Conducted experiments: Uldam, Juhl, and Pedersen.

2274
ULDAM ET AL.
ammonium-linked glucuronidation of tamoxifen by human liver microsomes and UDP-
Isolation and identification of clozapine metabolites in patient urine. Drug Metab Dispos
29:923–931.
glucuronosyltransferase 1A4. Biochem Pharmacol 67:2093–2102.
Smith PC and Benet LZ (1986) Characterization of the isomeric esters of zomepirac glucuronide
by proton NMR. Drug Metab Dispos 14:503–505.
Straub K, Davis M, and Hwang B (1988) Benzazepine metabolism revisited. Evidence for the
formation of novel amine conjugates. Drug Metab Dispos 16:359–366.
Tidwell TT (1990) Oxidation of alcohols to carbonyl compounds via alkoxysulfonium ylides: the
Kassahun K, Mattiuz E, Franklin R, and Gillespie T (1996) Olanzapine 10-N-glucuronide. A
tertiary N-glucuronide unique to humans. Drug Metab Dispos 26:848–855.
Kay LE, Keifer P, and Saarinen T (1992) Pure absorption gradient enhanced heteronuclear single
quantum correlation spectroscopy with improved sensitivity. J Am Chem Soc 114:10663–
10665.
Keating KA, McConnell O, Zhang Y, Shen L, Demaio W, Mallis L, Elmarakby S, and
Chandrasekaran A (2006) NMR characterization of an S-linked glucuronide metabolite of the
potent, novel, nonsteroidal progesterone agonist tanaproget. Drug Metab Dispos 34:1283–
1287.
Kowalczyk I, Hawes EM, and McKay G (2000) Stability and enzymatic hydrolysis of quaternary
ammonium-linked glucuronide metabolites of drugs with an aliphatic tertiary amine-
implications for analysis. J Pharm Biomed Anal 22:803–811.
Miller RR, Doss GA, and Stearns RA (2004) Identification of a hydroxylamine glucuronide
metabolite of an oral hypoglycemic agent. Drug Metab Dispos 32:178–185.
Mitchell MB and Whitcombe IWA (2000) The synthesis of the glucuronide adduct of Troca-
deTM. Tetrahedron Lett 41:8829–8834.
Nakazawa T, Miyata K, Omura K, Iwanaga T, and Nagata O (2006) Metabolic profile of
FYX-051 (4-(5-pyridin-4-yl-1H-[1,2,4]triazol-3-yl)pyridine-2-carbonitrile) in the rat, dog,
monkey, and human: identification of N-glucuronides and N-glucosides. Drug Metab Dispos
34:1880–1886.
Omura K, Nakazawa T, Sato T, Iwanaga T, and Nagata O (2007) Characterization of N-
glucuronidation of 4-(5-pyridin-4-yl-1H-[1,2,4]triazol-3-yl) pyridine-2-carbonitrile (FYX-
051): a new xanthine oxidoreductase inhibitor. Drug Metab Dispos 35:2143–2148.
Parikh JR and Doering WvE (1967) Sulfur trioxide in the oxidation of alcohols by dimethyl
sulfoxide. J Am Chem Soc 89:5505–5507.
Moffat, Swern, and related oxidations. Org React 39:297–572.
Turgeon J, Pare´ JR, Lalande M, Grech-Be´langer O, and Be´langer PM (1992) Isolation and
structural characterization by spectroscopic methods of two glucuronide metabolites of mexi-
letine after N-oxidation and deamination. Drug Metab Dispos 20:762–769.
Vashishtha SC, Hawes EM, McCann DJ, Ghosheh O, and Hogg L (2002) Quaternary ammoni-
um-linked glucuronidation of 1-substituted imidazoles by liver microsomes: interspecies
differences and structure-metabolism relationships. Drug Metab Dispos 30:1070–1076.
Wen Z, Martin DE, Bullock P, Lee KH, and Smith PC (2007a) Glucuronidation of anti-HIV drug
candidate bevirimat: identification of human UDP-glucuronosyltransferases and species dif-
ferences. Drug Metab Dispos 35:440–448.
Wen Z, Tallman MN, Ali SY, and Smith PC (2007b) UDP-glucuronosyltransferase 1A1 is the
principal enzyme responsible for etoposide glucuronidation in human liver and intestinal
microsomes: structural characterization of phenolic and alcoholic glucuronides of etoposide
and estimation of enzyme kinetics. Drug Metab Dispos 35:371–380.
Willker W, Leibfritz D, Kerssebaum R, and Bermel W (1993) Gradient selection in inverse
heteronuclear correlation spectroscopy. Magn Reson Chem 31:287–292.
Yan Z, Caldwell GW, Gauthier D, Leo GC, Mei J, Ho CY, Jones WJ, Masucci JA, Tuman RW,
Galemmo RA Jr, et al. (2006) N-glucuronidation of the platelet-derived growth factor receptor
tyrosine kinase inhibitor 6,7-(dimethoxy-2,4-dihydroindeno[1,2-C]pyrazol-3-yl)-(3-fluoro-
phenyl)-amine by human UDP-glucuronosyltransferases. Drug Metab Dispos 34:748–755.
Picard N, Ratanasavanh D, Pre´maud A, Le Meur Y, and Marquet P (2005) Identification of the
UDP-glucuronosyltransferase isoforms involved in mycophenolic acid phase II metabolism.
Drug Metab Dispos 33:139–146.
Porter CC, Arison BH, Gruber VF, Titus DC, and Vandenheuvel WJ (1975) Human metabolism
of cyproheptadine. Drug Metab Dispos 3:189–197.
Address correspondence to: Dr. Henriette Kold Uldam, Department of Drug
Metabolism, H. Lundbeck A/S, Ottiliavej 9, DK-2500 Valby, Denmark. E-mail:
hk@lundbeck.com
Schaber G, Wiatr G, Wachsmuth H, Dachtler M, Albert K, Gaertner I, and Breyer-Pfaff U (2001)