0090-9556/04/3202-178–185$20.00
DRUG METABOLISM AND DISPOSITION
Copyright © 2004 by The American Society for Pharmacology and Experimental Therapeutics
DMD 32:178–185, 2004
Vol. 32, No. 2
1156/1116829
Printed in U.S.A.
IDENTIFICATION OF A HYDROXYLAMINE GLUCURONIDE METABOLITE OF AN ORAL
HYPOGLYCEMIC AGENT
Randall R. Miller, George A. Doss, and Ralph A. Stearns
Department of Drug Metabolism, Merck Research Laboratories, Rahway, New Jersey
(Received April 18, 2003; accepted September 17, 2003)
This article is available online at http://dmd.aspetjournals.org
ABSTRACT:
or the monooxygenated product provided insufficient evidence to
unambiguously determine their structures. Incubation of 1 with pig
liver microsomes resulted in formation of the same monooxygenated
derivative derived from ␤-glucuronidase treatment of the glucuronide
metabolite. This in vitro system was used to generate sufficient material for analysis by 13C NMR, and the metabolite was identified as a
hydroxylamine derivative 2. Incubation of the hydroxylamine with
monkey liver microsomes and uridine diphospho-5ⴕ-glucuronic acid
gave the same glucuronic acid conjugate as that observed in monkey
urine. 13C NMR analysis of this biosynthetic product led to its unequivocal structure assignment as the O-glucuronic acid conjugate
of the hydroxylamine 3.
The oxidative metabolism of N-aryl or N-alkyl piperazines is typically characterized by carbon oxidation or oxidation of the tertiary
amine to give the N-oxide. Metabolites, typified in the in vivo metabolism of ketoconazole, ciprofloxacin, the inotropic agent OPC8212 (vesnarinone), and the phenothiazine derivative perazine, include carbinolamine, amide, ring-opened, and N-oxidation products
(Breyer, 1972; Gau et al., 1986; Miyamoto et al., 1988; Whitehouse et
al., 1994). The hydroxylamine intermediate, formed by oxidation of
the secondary amine, if present, is rarely observed in vivo for piperazines or other nitrogen-containing alicycles. An explanation for this
may be found in reports that this intermediate is inherently unstable
(Beckett and Salami, 1972; Beckett et al., 1977; Franklin et al., 1977;
Ziegler, 1987); however, identifications of stable alicyclic hydroxylamines have been made (Beckett and al-Sarraj, 1972; Beckett et al.,
1972; Achari and Beckett, 1983; Rodriguez and Acosta, 1997; Zhang
et al., 2000). There are even fewer examples of glucuronic acid
conjugation of alicyclic hydroxylamines in the literature (Straub et al.,
1988; Delbressine et al., 1992; Schaber et al., 2001), and the structures
of these metabolites have been largely deduced from mass spectrom-
etry and in some cases 1H NMR analysis. In the course of investigating the disposition of the oral hypoglycemic agent 11 (Leibowitz et al.,
1995) in rhesus monkeys, one major metabolite was observed in urine
from monkeys dosed orally. The molecular weight of this metabolite
corresponded to addition of an oxygen atom and glucuronic acid to the
piperazine ring of the parent drug. Both LC/MS/MS and 1H NMR
analysis failed to provide sufficient information to determine the exact
structure. This communication describes the unambiguous identification of this metabolite through its biosynthesis, isolation, and structural identification by 13C NMR spectroscopy.
1
Abbreviations used are: 1, 9-[(1S,2R)-2-fluoro-1-methylpropyl]-2-methoxy-6(1-piperazinyl) purine hydrochloride; LC/MS, liquid chromatography/mass spectrometry; MS/MS, tandem mass spectrometry; [14C]1, 9-[(1S,2R)-2-fluoro-1methylpropyl]-2-methoxy-6-(1-piperazinyl)-2-[14C]purine hydrochloride; HPLC,
high-performance liquid chromatography; NH4OAc, ammonium acetate; TFA,
trifluoroacetic acid; MeOH, methanol; UDPGA, uridine diphospho-5⬘-glucuronic
acid.
Address correspondence to: Randall R. Miller, Department of Drug Metabolism, Merck Research Laboratories, P.O. Box 2000, RY800B211, Rahway, NJ
07065. E-mail: [email protected]
Materials and Methods
Chemicals. 9-[(1S,2R)-2-Fluoro-1-methylpropyl]-2-methoxy-6-(1-piperazinyl) purine hydrochloride 1 was obtained from Merck Research Laboratories
(West Point, PA). 9-[(1S,2R)-2-Fluoro-1-methylpropyl]-2-methoxy-6-(1piperazinyl)-2-[14C]purine hydrochloride 1 (28.54 ␮Ci/mg) was synthesized
by the Labeled Compound Synthesis Group (Merck Research Laboratories).
The purity of the labeled and unlabeled compounds was ⬎98.5%. HPLC grade
solvents were from Fisher Scientific Co. (Fair Lawn, NJ). Glucose 6-phosphate, glucose-6-phosphate dehydrogenase (type V from baker’s yeast), ␤-glucuronidase (type HA-4 from Helix aspersa), glycerol, polyoxyethylene 9 lauryl
ether (Lubrol PX), and NADP⫹ were purchased from Sigma-Aldrich (St.
Louis, MO). EDTA, ammonium acetate, and TFA were purchased from
Aldrich Chemical Co. (Milwaukee, WI). Dibasic potassium phosphate, monobasic sodium phosphate, potassium chloride, and magnesium chloride were
obtained from Mallinckrodt (St. Louis, MO).
Animal Studies. Animal studies were conducted with approval of the
Institutional Animal Care and Use Committee. One male rhesus monkey (3 kg)
was dosed orally with 14C-labeled 1 (5 ␮Ci/mg; 5 mg/kg at 1.5 mg/ml)
dissolved in isotonic saline solution. Two additional monkeys (7.6 and 8 kg)
were dosed orally with nonradiolabeled 1 (5 mg/kg; 2.5 mg/ml in isotonic
saline). Oral doses were administered via a nasogastric tube. All animals were
178
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
Glucuronides of piperazine hydroxylamines are rarely reported in the
literature, and even more rarely are their structures unambiguously
identified. One major metabolite was detected by liquid chromatography/mass spectrometry-radioactivity in urine from monkeys
treated with the aryl piperazine oral hypoglycemic agent 9-[(1S,2R)2-fluoro-1-methylpropyl]-2-methoxy-6-(1-piperazinyl) purine hydrochloride (1). The mass spectrum of this metabolite indicated that it
was both monooxygenated and glucuronidated on the piperazine
ring. Possible structures included the N- or O-glucuronic acid conjugates of a carbinolamine, hydroxylamine, or N-oxide. Treatment with
␤-glucuronidase gave a monooxygenated derivative of the parent
compound. 1H NMR analysis of either the glucuronic acid conjugate
IDENTIFICATION OF A HYDROXYLAMINE GLUCURONIDE
FIG. 1. Structures of the aryl piperazine hypoglycemic agent 1, hydroxylamine 2,
O-glucuronide of the hydroxylamine 3, and possible N-glucuronide of the Noxide 4 structure that was not observed.
mobile phase consisting of 15% CH3CN/30% CH3OH/100 mM NH4OAc/1%
acetic acid. Fractions with absorbance at 280 nm were collected. To obtain
enough hydroxylamine metabolite for further investigation, this procedure was
repeated on an 8-times scale. The final yield of hydroxylamine metabolite was
7 mg from 24 mg of starting material 1.
Glucuronidation. Liver microsomes from male monkeys (final concentration, 3.39 mg protein/ml) were added to 100 ml of 50 mM potassium phosphate
TABLE 1
1
H NMR chemical shifts for 1, hydroxylamine 2, and hydroxylamine glucuronide 3 (in CD3OD at 400 MHz)
7.97 (d, 1H, JHF ⫽ 1.1, H-8), 4.95 (ddq, 1H, JHF ⫽ 48.2; JHH ⫽ 3.9, 6.3 Hz, FCHCH3), 4.68 (ddq, 1H,
JHF ⫽ 21.1; JHH ⫽ 4.2, 7.1 Hz, NCHCH3), 4.48 (br m, 4H, piperazine), 3.95 (s, 3H, OMe), 3.29 (br
m, 4H, piperazine), 1.62 (dd, 3H, JHF ⫽ 0.9; JHH ⫽ 7.1 Hz, NCHCH3), 1.34 (d d, 3H, JHF ⫽ 24.0;
JHH ⫽ 6.3, FCHCH3).
7.97 (d, 1H, JHF ⫽ 1.1 Hz, H-8), 5.23 (br m, 2H, piperazine eq), 4.93 (ddq, 1H, JHF ⫽ 48.7; JHH ⫽ 4.2,
6.3 Hz, FCHCH3), 4.65 (ddq, 1H, J HF ⫽ 21.0; JHH ⫽ 4.4, 7.3 Hz, NCHCH3), 3.93 (s, 3H, OMe), 3.40
(br m, 2H, piperazine ax), 3.27 (br m, 2H, piperazine eq), 2.63 (br m, 2H, piperazine ax), 1.61 (dd,
3H, J HF ⫽ 1.0; JHH ⫽ 7.1 Hz, NCHCH3), 1.33 (dd, 3H, JHF ⫽ 24.0; JHH ⫽ 6.3, FCHCH3).
7.97 (d, 1H, JHF ⫽ 1.1 Hz, H-8), 5.23 (br m, 2H, piperazine eq), 4.93 (ddq, 1H, JHF ⫽ 48.7; JHH ⫽ 4.2,
6.3 Hz, FCHCH3), 4.68 (d, 1H, J ⫽ 8.4 Hz, anomeric proton) 4.65 (ddq, 1H, JHF ⫽ 21.0; JHH ⫽ 4.4,
7.3 Hz, NCHCH3), 3.93 (s, 3H, OMe), 3.40 (br m, 2H, piperazine ax), 3.33, 3.39, 3.41, 3.61 (m, 4H,
sugar protons), 3.27 (br m, 2H, piperazine eq), 2.63 (br m, 2H, piperazine ax), 1.61 (dd, 3H, JHF ⫽
1.0; JHH ⫽ 7.1 Hz, NCHCH3), 1.33 (dd, 3H, JHF ⫽ 24.0; JHH ⫽ 6.3, FCHCH3).
1 Aryl piperazine
2 Hydroxylamine
3 Glucuronide
TABLE 2
13
1 Aryl piperazine
2 Hydroxylamine
3 Glucuronide
C NMR chemical shifts for 1, hydroxylamine 2, and hydroxylamine glucuronide 3 (in CD3OD at 125 MHz)
162.9 (C-2), 155.5 and 154.3 (C-4 and C-6), 139.0 (C-8), 117.1 (C-5), 92.5 (d, JCF ⫽ 172.9 Hz, CHF),
55.3 (d, JCF ⫽ 24.0 Hz), 55.2 (OMe), 44.8 (piperazine, C-3⬘,5⬘) , 43.9 (br, piperazine, C-2⬘,6⬘), 17.7
(d, JCF ⫽ 22.2 Hz, FCHCH3), 14.1 (d, JCF ⫽ 4.6 Hz, NCHCH3).
163.0 (C-2), 155.6 and 154.0 (C-4 and C-6), 138.5 (C-8), 117.0 (C-5), 92.3 (d, JCF ⫽ 177 Hz, CHF),
58.8 (piperazine, C-3⬘,5⬘), 55.2 (d, JCF ⫽ 23 Hz), 55.1 (OMe), 44.8 (br, piperazine, C-2⬘,6⬘), 17.8 (d,
JCF ⫽ 26 Hz, FCHCH3), 14.1 (d, JCF ⫽ 4.7 Hz, NCHCH3)
176.7 (Glu, CO2H), 162.9 (C-2), 155.4 and 153.9 (C-4 and C-6), 138.3 (C-8), 116.9 (C-5), 106.6 (Glu,
anomeric carbon), 92.3 (d, JCF ⫽ 172.9 Hz, CHF); 78.3, 76.4, 73.7, 73.4 (4 carbons, Glu); 58.3 and
56.8 (piperazine, C-3⬘,5⬘), 55.2 (d, JCF ⫽ 23 Hz, NCHCH3), 55.1 (OMe), 44.9 (br, piperazine, C2⬘,6⬘), 17.8 (d, JCF ⫽ 22.4 Hz, FCHCH3), 14.1 (d, JCF ⫽ 4.7 Hz, NCHCH3).
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
fasted for 16 h before dosing. Urine and feces were collected for 48 h post
dose.
Biological Preparations. Pig liver microsomes were prepared as described
previously (Ziegler and Poulsen, 1978). Microsomes were prepared as described previously (Levin et al., 1972) from monkey livers that were removed
within 30 min of death, frozen in liquid nitrogen, and stored at ⫺80°C. Protein
concentrations were determined by using the Lowry method (Lowry et al.,
1951). The hydroxylamine glucuronide of 1 (100 ␮g, 40 ␮M) was incubated
for 18 h at 37°C with 25,000 units of ␤-glucuronidase in 5 ml of 20 mM
sodium acetate buffer, pH 5.
Metabolite Extraction. Typically, aliquots of the incubation medium
(0.5–1 ml) were passed through a 3-ml Analytichem BondElut C18 solid-phase
extraction cartridge (Varian, Inc., Palo Alto, CA) that was prewashed with 5 ml
of methanol followed by 5 ml of water. After loading, the cartridge was
washed with 3 ml of water and eluted with 5 ml of methanol. When radiolabeled substrate was used, 50 to 100 ␮l of the diluted incubation medium,
cartridge wash, and methanol eluate was counted by liquid scintillation to
account for recovery. Methanol was evaporated at 35°C under nitrogen in a
Turbovap LV evaporator (Zymark Corp., Hopkinton, MA). The residue was
reconstituted in 100 to 200 ␮l of mobile phase A (see below), and 25 to 50 ␮l
was analyzed by HPLC.
Preparative Scale Biosynthesis of Hydroxylamine Glucuronide (3). Oxidation. Pig liver microsomes (final concentration 2.6 mg protein/ml; 0.042
nmol cytochrome P450/ml) were suspended in 50 ml of 0.1 M glycine buffer,
pH 8.5, containing 1 mM EDTA in a 250-ml Erlenmeyer flask. Glucose
6-phosphate (3.3 mM), magnesium chloride (3.3 mM), and NADP⫹ (1.3 mM)
were added to the suspension along with 3 mg of 14C-labeled 1 (0.4 ␮Ci/mg).
The flask was placed in a reciprocating water bath at 37°C and preincubated
for 5 min. Metabolism was initiated by addition of glucose-6-phosphate
dehydrogenase (1 unit/ml). After 1-h incubation, the medium was passed
through a 60-ml C18 BondElut cartridge that had been pre-equilibrated by
treatment with 20 ml of MeOH followed by 20 ml of water. The cartridge was
then washed with 50 ml of water, and retained metabolites were eluted with 70
ml of MeOH. Methanol was evaporated in a 250-ml round-bottom flask on a
rotary evaporator. The residue was reconstituted in 10 ml of MeOH to give a
suspension that was separated by centrifugation. The organic solvent was
transferred by pipette to a 16 ⫻ 125 mm glass test tube and evaporated under
a stream of nitrogen gas at 35°C. The residue was reconstituted in 300 ␮l of
mobile phase and purified by three successive 100-␮l injections on a 4.6 ⫻ 250
mm Zorbax C8 column (DuPont, Wilmington, DE) eluted at 1 ml/min with a
179
180
MILLER ET AL.
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
FIG. 2. HPLC radiohistogram (A) and LC/MS reconstructed ion current chromatograms for m/z 309 (B) and m/z 501 (C) of 0- to 24-h monkey urine.
buffer, pH 7.7, containing 10 mM MgCl2, 3 mM UDPGA, 2% glycerol, 0.02%
Lubrol PX, and 7 mg of hydroxylamine 2 (Huskey et al., 1993). This suspension was incubated for 1 h at 37°C, and the reaction was stopped by freezing.
The hydroxylamine glucuronide metabolite 3 (4.6 mg) was purified as described above for the purification of the hydroxylamine (retention time ⬃5
min). Before NMR analysis, this metabolite was repurified by HPLC on a 4.6⫻ 250-mm Zorbax RX C8 column eluted at 1 ml/min with a mobile phase
consisting of 10% CH3CN/15% CH3OH/100 mM NH4OAc/1% acetic acid.
Under these conditions, the glucuronic acid conjugate eluted at approximately
20 min.
HPLC Analysis. The HPLC system consisted of two Shimadzu LC-600
pumps (Kyoto, Japan), a Shimadzu static-bed mixer, and a Rheodyne 7125
injector (Cotati, CA). Chromatography was performed on a DuPont Zorbax
RX C8 column (4.6 ⫻ 250 mm) eluted at 1 ml/min with a linear gradient from
(A), 20% CH3CN/10 mM NH4OAc/1% TFA to (B), 40% CH3CN/10 mM
NH4OAc/1% TFA in 30 min and held at 100% (B) for 10 min. UV detection
was performed using a Spectra Physics SpectraFocus scanning UV detector
(Spectra Physics, San Jose, CA). Fractions were collected using an Amersham
Biosciences fraction collector (Amersham Biosciences, Uppsala, Sweden).
NMR Spectroscopy. NMR spectra were obtained on a Varian Unity 400 or
500 MHz spectrometer. Chemical shifts are given in ppm downfield from
tetramethylsilane. Residual solvent signals were used as an internal reference
(CD3OD, 3.30 ppm for proton; 49.0 ppm for 13C). Coupling constants (J) are
given in hertz.
Mass Spectrometry. Mass spectral analysis was performed by LC/MS on
an MDS Sciex (Concord, ON, Canada) API III tandem mass spectrometer
using the ionspray interface. Spectra were acquired using positive ion detection, scanning from 150 to 600 Da with a 5- to 10-ms dwell time (orifice
potential ⫽ 65 V). Argon was used as the collision gas for LC/MS/MS
experiments.
Radioactivity Measurement. Radioactivity in incubation extracts or HPLC
fractions was determined by counting 0.1- to 1-ml aliquots in glass vials
containing 6 ml of liquid scintillation cocktail (Insta-Gel XF; PerkinElmer Life
and Analytical Sciences, Boston, MA). Measurement of radioactivity was
IDENTIFICATION OF A HYDROXYLAMINE GLUCURONIDE
181
performed by liquid scintillation counting using a Beckman LS5000TD liquid
scintillation spectrometer (Beckman Coulter, Fullerton, CA).
Results
In a rhesus monkey dosed orally with 14C-labeled 1 (Fig. 1),
approximately 70% of the administered radioactivity was excreted in
the urine within 24 h of oral dosing. The HPLC radiohistogram and
LC/MS-reconstructed ion current chromatograms (m/z 309 and 501)
of an extract of urine collected from 0 to 24 h postdose are shown in
Fig. 2. Parent drug and a metabolite of similar polarity (retention
time ⫽ 29 –30 s accounted for ⬃25 and ⬃50%, respectively, of the
total radioactivity recovered in urine. Both parent drug and metabolite
were observed in urine from two other monkeys dosed with unlabeled
drug, and the metabolite ratio, determined by comparison of the parent
drug ion current to that of the metabolite, was similar in all three
monkeys. Based on its mass spectrum, the metabolite gave an
(M⫹H)⫹ ⫽ m/z 501, an addition of 192 Da to the mass of the parent
drug. This is equivalent to the addition of an oxygen atom and
conjugation with glucuronic acid. Collisional activation of this ion
gave the MS/MS spectrum shown in Fig. 3. The product ions at 325,
309, 307, 266, 253, and 240 Da corresponded to fragments of the
piperazine group and were consistent with a metabolite that was a
glucuronic acid conjugate of a monooxygenated piperazine derivative
(Delbressine et al., 1992; Schaber et al., 2001) but was not conclusive.
The 1H NMR spectrum of this metabolite revealed the presence of
signals characteristic of a ␤-glucuronic acid moiety (anomeric proton,
doublet at 4.68 ppm, J ⫽ 8.4 Hz). The absence of perturbation of the
NMR signals of the fluorinated side chain or aromatic protons supported the characterization that the site of metabolism was on the
piperazine group.
Treatment of this metabolite with ␤-glucuronidase resulted in hydrolysis to a compound with (M⫹H)⫹ ⫽ m/z 325, corresponding to an
oxygenated derivative of the parent drug. This material was less polar
than the parent compound based on retention time by reversed-phase
HPLC. The MS/MS spectrum of this derivative is shown in Fig. 4.
The product ions at 307, 266, 253, and 240 Da were consistent with
monooxidation of the piperazine group. The 1H NMR spectrum of this
material was consistent with this characterization, since the eight
piperazine protons were present as four broad resonances at 2.63,
3.27, 3.40, and 5.23 ppm.
In vitro biosynthesis of these metabolites was pursued to obtain
sufficient material for further characterization. Compound 1 was
incubated with pig liver microsomes, and the HPLC-UV chromatogram (280 nm) and associated LC/MS-reconstructed ion current chromatogram of the extracts of this incubation are shown in Fig. 5. The
metabolite that eluted at 18 min with an (M⫹H)⫹ ⫽ m/z 325 was
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
FIG. 3. MS/MS spectrum obtained from collision-induced dissociation of the MH⫹ ion at m/z 509 of the hydroxylamine glucuronide metabolite 3 isolated from
monkey urine.
182
MILLER ET AL.
isolated. It coeluted by HPLC and had 1H NMR (Fig. 6) and MS/MS
spectra identical to those of the product derived from ␤-glucuronidase
treatment of the major metabolite from monkey urine. From a largescale incubation of 1 with pig liver microsomes, approximately 7 mg
of this metabolite was isolated and purified for 13C NMR analysis
(Fig. 7). Similar to the parent compound, two 13C signals for the
piperazine ring were observed in the spectrum of this metabolite.
However, in the metabolite, there was a significant downfield shift of
the C3/5 methylene signals of the piperazine ring from 44.8 to 58.8
ppm. The C2/6 methylene signals were shifted only marginally from
43.9 to 44.8 ppm.
The biosynthetic monooxygenated piperazine metabolite was then
incubated for 1 h with monkey liver microsomes that contained
Lubrol PX and UDPGA to generate glucuronides for 13C NMR
analysis. The reaction was virtually quantitative to one product, and
approximately 4.6 mg of this was isolated preparatively. This material
coeluted by HPLC and had 1H NMR (Fig. 8) and MS/MS spectra
identical to those of the major metabolite identified in the urine from
monkeys treated with 1. Comparison of the chemical shifts in the 13C
NMR spectrum of the glucuronide (Fig. 9) with those of the related
hydroxylamine metabolite revealed a slight upfield shift in the piperazine C3/5 carbons, which were now split into two peaks at 58.3 and
56.8 ppm. There was virtually no change in the C2/6 signals.
Discussion
The major metabolite of the aryl piperazine hypoglycemic agent 1
in rhesus monkeys was identified by LC/MS/MS and 1H NMR as
being both oxygenated and glucuronidated on the piperazine functional group. Assignment of an exact structure, however, was not
trivial due in part to extensive line broadening of the proton signals of
the piperazine ring observed in its 1H NMR spectrum. This broadening was not resolved by changes in either solvent or temperature,
rendering them relatively uninformative. This broadening is commonly seen with piperazine rings and is attributed to flipping of the
piperazine ring in the time frame of NMR analysis (Straub et al.,
1988). To unambiguously determine the glucuronide metabolite structure, sufficient material was needed for analysis by 13C NMR.
Also of interest and crucial to the structural identification of the
glucuronide was the structure of the aglycone. Treatment of the
glucuronide metabolite with ␤-glucuronidase liberated a derivative
that could be partially identified by LC/MS/MS and 1H NMR as
monooxygenated on the piperazine, although, as with the parent
compound, the 1H NMR spectrum of the aglycone suffered from
broad signals arising from the piperazine protons, precluding its exact
structure determination by this technique. However, due to the symmetry of the observed signals, it was clear that the additional oxygen
atom was attached to one of the two piperazine nitrogen atoms to give
either an N-oxide or a hydroxylamine. This metabolite was stable
under the conditions used to isolate as well as characterize it. To
assign the structure of this metabolite unequivocally, it was again
necessary to obtain a sufficient quantity for 13C NMR analysis.
To produce sufficient amounts of both the glucuronide and the
aglycone and obtain them with adequate purity, methods of in vitro
biosynthesis were investigated. The metabolism of 1 was evaluated in
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
FIG. 4. MS/MS spectrum obtained from collision-induced dissociation of the MH⫹ ion at m/z 325 of the hydroxylamine 2 resulting from B-glucuronidase treatment of
the glucuronide metabolite isolated from monkey urine.
IDENTIFICATION OF A HYDROXYLAMINE GLUCURONIDE
pig liver microsomes because of the expected catalytic capacity of this
in vitro system to provide a biosynthetic route to the monooxygenated
piperazine compound. Using this in vitro system, sufficient quantities
of a metabolite with identical HPLC, MS/MS, and 1H NMR characteristics to that of the aglycone were isolated for 13C NMR analysis.
Two signals were attributed to the piperazine carbons in the 13C NMR
spectrum of the aglycone. This corroborated the observation from 1H
NMR analysis that the symmetry of the piperazine was still intact and
limited the site of oxidation to one of the two nitrogens. Additionally,
there was a large (14 ppm) downfield shift of only the C3/5 methylene
signals of the piperazine, which can be attributed to the ␤-effect of the
N-OH group, although there was only a marginal shift of the C2/6
methylene signals (0.9 ppm). The significant shift of the C3/5 methylene indicated that the site of oxygen attachment must therefore be
on the secondary nitrogen. Using this rationale, the exact structure of
the aglycone was established as the hydroxylamine derivative of 1 and
not the N-oxide of the tertiary piperazine nitrogen.
Incubation of this biosynthetic hydroxylamine with monkey liver
microsomes and UDPGA produced a glucuronide that had HPLC,
MS/MS, and 1H NMR characteristics identical to those of the major
metabolite of 1 isolated from monkey urine. With the structure of the
hydroxylamine definitively established, these observations allowed
for the otherwise nontrivial distinction between the two possible sites
of glucuronide attachment, which were either at the oxygen 3 or the
nitrogen 4. These 13C NMR data were consistent with the O-glucuronide structure 3 where the C3/5 methylenes experienced an (upfield)
␥-effect and the C2/6 methylenes experience a (negligible) ␦-effect by
the glucuronic acid moiety. Also, the C3/5 methylenes displayed
larger splitting, more so than C2/6, due to close proximity to the chiral
glucuronic acid moiety. In contrast, if the structure had been the
N-glucuronide 4, the C3/5 methylenes would have exhibited a significant downfield shift due to the ␤-effect of the glucuronic acid moiety,
and this was not observed. The culmination of all of these results led
to the unambiguous structural determination of the major metabolite
found in the urine of monkeys treated with the hypoglycemic agent 1
as the O-glucuronide of the hydroxylamine 3.
FIG. 6. 1H NMR spectrum of the hydroxylamine metabolite 2.
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
FIG. 5. HPLC-UV chromatogram (A) and LC/MS reconstructed ion current
chromatograms of m/z 309 (B) and m/z 325 (C) of an incubation of 1 with pig
liver microsomes.
183
184
MILLER ET AL.
13
C NMR spectrum of the hydroxylamine metabolite 2.
FIG. 8. 1H NMR spectrum of the glucuronide metabolite 3.
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
FIG. 7.
IDENTIFICATION OF A HYDROXYLAMINE GLUCURONIDE
13
C NMR spectrum of the glucuronide metabolite 3.
References
Achari R and Beckett AH (1983) In-vitro metabolism of two mono-substituted piperazines using
liver homogenates of rats and rabbits. J Pharm Pharmacol 35:615– 616.
Beckett AH and al-Sarraj S (1972) N-oxidation of primary and secondary amines to give
hydroxylamines—a general metabolic route. J Pharm Pharmacol 24:916 –917.
Beckett AH, Purkaystha AR, and Morgan PH (1977) Oxidation of aliphatic hydroxylamines in
aqueous solution. J Pharm Pharmacol 29:15–21.
Beckett AH and Salami MA (1972) A note on the identification of N-hydroxyphen-metrazine as
a metabolic product of phendimetrazine and phenmetrazine. J Pharm Pharmacol 24:900 –902.
Breyer U (1972) Accumulation and elimination of a novel metabolite during chronic administration of the phenothiazine drug perazine to rats. Biochem Pharmacol 21:1419 –1429.
Delbressine LPC, Moonen MEG, Kaspersen FM, Jacobs PL, and Wagenaars GL (1992) Biotransformation of mianserin in laboratory animals and man. Xenobiotica 22:227–236.
Franklin RB, Dring LG, and Williams RT (1977) The metabolism of phenmetrazine in man and
laboratory animals. Drug Metab Dispos 5:223–233.
Gau W, Kurz J, Petersen U, Ploschke HJ, and Wuensche C (1986) Isolation and structural
elucidation of urinary metabolites of ciprofloxacin. Arzneim-Forsch 36:1545–1549.
Huskey SW, Miller RR, and Chiu SHL (1993) N-glucuronidation reactions. Tetrazole Nglucuronidation of selected angiotensin II receptor antagonists in hepatic microsomes from
rats, dogs, monkeys, and humans. Drug Metab Dispos 21:792–799.
Leibowitz MD, Biswas C, Brady EJ, Conti M, Cullinan CA, Hayes NS, Manganiello VC,
Saperstein R, Wang L, Zafian PT, and Berger J (1995) A novel insulin secretagogue is a
phosphodiesterase inhibitor. Diabetes 44:67–74.
Levin W, Sernatinger E, Jacobson M, and Kuntzman R (1972) Destruction of cytochrome P450 by
secobarbital and other barbiturates containing allyl groups. Science (Wash DC) 176:1341–1343.
Lowry OH, Rosenbrough NJ, Lewis A, and Randall RJ (1951) Protein measurement with the
folin phenol reagent. J Biol Chem 193:265–275.
Miyamoto G, Sasabe H, Tominaga N, Uegaki N, Tominaga M, and Shimizu T (1988) Metabolism of a new inotropic agent, 3,4-dihydro-6-[4-(3,4-dimethoxybenzoyl)-1-piperazinyl]-2(1H)-quinolinone (OPC-8212), in the rat, mouse, dog, monkey and human. Xenobiotica
18:1143–1155.
Rodriguez RJ and Acosta D (1997) Metabolism of ketoconazole and deacetylated ketoconazole
by rat hepatic microsomes and flavin-containing monooxygenases. Drug Metab Dispos
25:772–777.
Schaber G, Wiatr G, Wachsmuth H, Dachtler M, Albert K, Gaertner I, and Breyer-Pfaff U (2001)
Isolation and identification of clozapine metabolites in patient urine. Drug Metab Dispos
29:923–931.
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.
Whitehouse LW, Menzies A, Dawson B, Cyr TD, By AW, Black DB, and Zamecnik J (1994)
Mouse hepatic metabolites of ketoconazole: isolation and structure elucidation. J Pharm
Biomed Anal 12:1425–1441.
Zhang KE, Kari PH, Davis MR, Doss G, Baillie T, and Vyas K (2000) Metabolism of a dopamine
D4-selective antagonist in rat, monkey and humans: formation of a novel mercapturic acid
adduct. Drug Metab Dispos 28:633– 642.
Ziegler DM (1987) Functional groups activated via flavin-containing monooxygenases. Proceedings of the VII International Symposium: Microsomes and Drug-Oxidations; Adelaide,
Australia, pp. 297–304.
Ziegler DM and Poulsen LL (1978) Hepatic microsomal mixed-function amine oxidase. Methods
Enzymol 52:145–151.
Downloaded from dmd.aspetjournals.org at ASPET Journals on June 15, 2017
FIG. 9.
185