In vitro metabolism and identification of human enzymes involved in the metabolism of methylnaltrexone

Article abstract of DOI:10.1124/dmd.110.032169

Methylnaltrexone (MNTX) is a peripherally acting μ-opioid receptor antagonist and is currently indicated for the treatment of opioid-induced constipation in patients with advanced illness who are receiving palliative care, when response to laxative therapy has not been sufficient. Sulfation to MNTX-3-sulfate (M2) and carbonyl reduction to methyl-6α-naltrexol (M4) and methyl-6β-naltrexol (M5) are the primary metabolic pathways for MNTX in humans. The objectives of this study were to investigate MNTX in vitro metabolism in human and nonclinical species and to identify the human enzymes involved in MNTX metabolism. Of the five commercially available sulfotransferases investigated, only SULT2A1 and SULT1E1 catalyzed M2 formation. Formation of M4 and M5 was catalyzed by NADPH-dependent hepatic cytosolic enzymes, which were identified using selective chemical inhibitors (10 and 100 μM) for aldo-keto reductase (AKR) isoforms, short-chain dehydrogenase/ reductase including carbonyl reductase, alcohol dehydrogenase, and quinone oxidoreductase. The results were then compared with the effects of the same inhibitors on 6β-naltrexol formation from naltrexone, a structural analog of MNTX, which is catalyzed mainly by AKR1C4. The AKR1C inhibitor phenolphthalein inhibited MNTX and naltrexone reduction up to 98%. 5β-Cholanic acid 3α,7α-diol, the AKR1C2 inhibitor, and medroxyprogesterone acetate, an inhibitor of AKR1C1, AKR1C2, and AKR1C4, inhibited MNTX reduction up to 67%. Other inhibitors were less potent. In conclusion, the carbonyl reduction of MNTX to M4 and M5 in hepatic cytosol was consistent with previous in vivo observations. AKR1C4 appeared to play a major role in the carbonyl reduction of MNTX, although multiple enzymes in the AKR1C subfamily may be involved. Human SULT2A1 and SULT1E1 were involved in MNTX sulfation. Copyright

Full text of DOI:10.1124/dmd.110.032169

0
090-9556/10/3805-801–807$20.00
DRUG METABOLISM AND DISPOSITION
Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics
DMD 38:801–807, 2010
Vol. 38, No. 5
32169/3583024
Printed in U.S.A.
In Vitro Metabolism and Identification of Human Enzymes Involved
in the Metabolism of Methylnaltrexone
Zeen Tong, Appavu Chandrasekaran, Hongshan Li, Yakov Rotshteyn, John C. L. Erve,
William DeMaio, Rasmy Talaat, Theresa Hultin, and JoAnn Scatina
Pfizer Inc., Collegeville, Pennsylvania (Z.T., A.C., H.L., J.C.L.E., W.D., R.T., T.H., J.S.); and Progenics Pharmaceuticals, Inc.,
Tarrytown, New York (Y.R.)
Received January 11, 2010; accepted February 19, 2010
ABSTRACT:
Methylnaltrexone (MNTX) is a peripherally acting ␮-opioid receptor hydrogenase/reductase including carbonyl reductase, alcohol de-
antagonist and is currently indicated for the treatment of opioid- hydrogenase, and quinone oxidoreductase. The results were then
induced constipation in patients with advanced illness who are compared with the effects of the same inhibitors on 6␤-naltrexol
receiving palliative care, when response to laxative therapy has not formation from naltrexone, a structural analog of MNTX, which is
been sufficient. Sulfation to MNTX-3-sulfate (M2) and carbonyl catalyzed mainly by AKR1C4. The AKR1C inhibitor phenolphthalein
reduction to methyl-6␣-naltrexol (M4) and methyl-6␤-naltrexol inhibited MNTX and naltrexone reduction up to 98%. 5␤-Cholanic
(
M5) are the primary metabolic pathways for MNTX in humans. acid 3␣,7␣-diol, the AKR1C2 inhibitor, and medroxyprogesterone
The objectives of this study were to investigate MNTX in vitro acetate, an inhibitor of AKR1C1, AKR1C2, and AKR1C4, inhibited
metabolism in human and nonclinical species and to identify the MNTX reduction up to 67%. Other inhibitors were less potent. In
human enzymes involved in MNTX metabolism. Of the five com- conclusion, the carbonyl reduction of MNTX to M4 and M5 in
mercially available sulfotransferases investigated, only SULT2A1 hepatic cytosol was consistent with previous in vivo observations.
and SULT1E1 catalyzed M2 formation. Formation of M4 and M5 AKR1C4 appeared to play a major role in the carbonyl reduction of
was catalyzed by NADPH-dependent hepatic cytosolic enzymes, MNTX, although multiple enzymes in the AKR1C subfamily may be
which were identified using selective chemical inhibitors (10 and involved. Human SULT2A1 and SULT1E1 were involved in MNTX
100 ␮M) for aldo-keto reductase (AKR) isoforms, short-chain de- sulfation.
Methylnaltrexone (MNTX) is a quaternary derivative of the opioid pounds such as thyroid hormones, steroids, monoamine neurotrans-
antagonist naltrexone (Brown and Goldberg, 1985). Unlike naltrex- mitters, and bile acids (Gamage et al., 2006; Matsunaga et al., 2006).
one, this opioid receptor antagonist has restricted penetration across Carbonyl reduction of aldehyde and ketone moieties may be cata-
the blood-brain barrier (Brown and Goldberg, 1985; Yuan et al., 1996, lyzed by alcohol dehydrogenases, aldo-keto reductases (AKRs),
1
998). Thus, it reduces the peripheral side effect of opioids without short-chain dehydrogenases/reductases (SDRs) including carbonyl
reversal of analgesia or the induction of opioid withdrawal. In healthy reductase (CR), and quinone reductases (Rosemond and Walsh,
volunteers, intravenous MNTX effectively blocked acute morphine- 2004). Although these enzymes are involved in biological pro-
induced delay in oral-cecal transit time without affecting analgesia cesses such as steroid and prostaglandin metabolism, drug detox-
(
Yuan et al., 1996, 2002). A significant reversal of the gut transit ification, housekeeping, stress response, and neurotransmission,
delay in subjects dependent on chronic methadone treatment was also and are considered potential drug targets (Rosemond and Walsh,
observed with intravenous MNTX (Yuan et al., 1998, 1999, 2000). 2004), much less is known about their roles in drug metabolism
Unlike naltrexone, a structural analog of MNTX without the quater- than their roles in physiological processes. In the present study, in
nary amine, which is extensively metabolized in humans (Wall et al., vitro reduction of MNTX was evaluated in mice, rats, dogs, mon-
1
981, 1984), MNTX is only minimally metabolized in humans (Chan- keys, and humans, and human enzymes involved in the MNTX
drasekaran et al., 2010).
metabolism were identified.
It is well documented that sulfation is an important pathway in the
biotransformation of xenobiotics such as drugs and endogenous com-
Materials and Methods
1
4
Materials. [ C]MNTX bromide salt was synthesized by Selcia Limited
(Essex, UK). The radiochemical purity and the chemical purity (by UV
detection) of the radiolabeled MNTX bromide were greater than 97%. The
specific activity was 125.1 ␮Ci/mg (153.0 ␮Ci/mg as the free base). Nonra-
diolabeled MNTX bromide salt was synthesized by Mallinckrodt Baker, Inc.
(Phillipsburg, NJ) and had a chemical purity of 100%. Unless indicated
Progenics Pharmaceuticals, Inc. has a proprietary commercial interest in
MNTX. Y.R. is an employee and stockholder of Progenics Pharmaceuticals, Inc.
Article, publication date, and citation information can be found at
http://dmd.aspetjournals.org.
doi:10.1124/dmd.110.032169.
ABBREVIATIONS: MNTX, methylnaltrexone; AKR, aldo-keto reductase; SDR, short-chain dehydrogenase/reductase; CR, carbonyl reductase;
PAPS, 3Ј-phosphoadenosine 5Ј-phosphosulfate lithium salt; HPLC, high-performance liquid chromatography; LC/MS, liquid chromatography/
mass spectrometry; M4, methyl-6␣-naltrexol; M5, methyl-6␤-naltrexol; M2, methylnaltrexone-3-sulfate; P450, cytochrome P450.
801

8
02
TONG ET AL.
1
4
otherwise, when referring to [ C]MNTX or MNTX, the free base form is
(20 ␮l) of the supernatant were taken for determination of radioactivity by
ϩ
assumed. Naltrexone, 6␤-naltrexol hydrate, glucose 6-phosphate, NADP , liquid scintillation as described later. This extraction method recovered greater
and NADPH, 3Ј-phosphoadenosine 5Ј-phosphosulfate lithium salt (PAPS), than 85% of the radioactivity from the reaction mixture. An aliquot of the
glucose-6-phosphate dehydrogenase, phenolphthalein, flufenamic acid, me- supernatant was analyzed by HPLC with radioactivity flow detection.
droxyprogesterone acetate, indomethacin, phenobarbital, ethacrynic acid, men-
adione, quercetin, 4-methylpyrazole, and dicumarol were ordered from Sigma-
MNTX reduction in hepatic microsomal and cytosolic fractions was evaluated
in different species. [ C]MNTX (50 ␮M, 0.3–0.5 ␮Ci/ml) was incubated with
14
Aldrich (St. Louis, MO). 5␤-Cholanic acid-3␣,7␣-diol was ordered from liver microsomes or cytosol (4 mg/ml) of mice, rats, dogs, monkeys, and humans
Steraloids (Newport, RI). Recombinant human SULT1A1, SULT1A2, for 2 h as described above. Activity in the microsomal and cytosolic fractions in
SULT1A3, SULT2A1, and SULT1E1 were ordered from Invitrogen (Carlsbad, human intestine was also evaluated under the same conditions.
CA). Microsomal and cytosolic fractions of human liver and intestine and
hepatic cytosolic fractions of male mouse, rat, dog, and monkey were pur-
chased from XenoTech, LLC (Lenexa, KS). Ultima Gold and Ultima Flo M
Conditions were optimized for kinetic study in liver cytosol of mice, rats,
monkeys, and humans. Time course and protein dependence experiments were
conducted to determine the initial rate conditions. The time course study was
1
4
scintillation mixtures were purchased from PerkinElmer Life and Analytical conducted by incubating [ C]MNTX (50 ␮M, 0.3 ␮Ci/ml) in the presence of
Sciences (Waltham, MA). ZipTips packed with C18 sorbent were obtained
from Millipore Corporation (Billerica, MA). Solvents used for extraction and
the NADPH-regenerating system with liver cytosol (2.0 mg/ml) for 0, 10, 20,
30, 60, and 120 min under the conditions described above. The protein
1
4
chromatographic analysis were high-performance liquid chromatography dependence experiment was conducted by incubating [ C]MNTX (50 ␮M, 0.3
HPLC) or American Chemical Society reagent grade and were purchased ␮Ci/ml) for 1 h in the presence of the NADPH-regenerating system with liver
from EMD Chemicals (Gibbstown, NJ). Deuterium oxide and d -methanol cytosol (0, 0.1, 0.25, 0.5, 1.0, 2.0, 4.0, and 6.0 mg/ml) under the conditions
(
4
were obtained from Cambridge Isotope Laboratories, Inc. (Andover, MA).
described above.
1
4
MNTX Reduction In Vitro. [ C]MNTX was mixed with nonradiolabeled
Kinetics of MNTX reduction was determined under the initial rate condi-
1
4
14
MNTX (1:3 or 1:5) for the incubations. Incubations consisted of [ C]MNTX, tions. In brief, [ C]MNTX (5.0–2000 ␮M) was incubated in the presence of
magnesium chloride (10 mM), and liver microsomes or cytosol in 0.5 ml of
00 mM potassium phosphate buffer, pH 7.4, and were conducted in a shaking
the NADPH-regenerating system with liver cytosol (0.5–4.0 mg/ml) for 60 to
120 min. The incubation conditions are listed in Table 1. Likewise, time
1
water bath at 37°C. After preincubation for 3 min, the reactions were initiated course, protein dependence, and kinetics of naltrexone reduction to 6␤-
by the addition of an NADPH-regenerating system. Control incubations were naltrexol in human liver cytosol were also studied under the conditions listed
conducted in absence of the NADPH regenerating system. The NADPH- in Table 1 for comparison with MNTX.
regenerating system consisted of glucose 6-phosphate (2 mg/ml), glucose-6-
phosphate dehydrogenase (0.8 units/ml), and NADP (2 mg/ml). Incubations identify the sulfotransferase isoforms responsible for MNTX sulfation,
Identification of Human Enzymes Involved in MNTX Sulfation. To
ϩ
1
4
were stopped by the addition of 0.5 ml of ice-cold acetonitrile. Samples were
vortex-mixed. Denatured proteins were separated by centrifugation at 4300
rpm and 4°C for 10 min (model T21 super centrifuge; Sorvall, Newton, CT).
The supernatant was evaporated to a volume of approximately 0.4 ml under a
nitrogen stream in a TurboVap LV evaporator (Caliper Life Sciences, Hop-
kinton, MA). The concentrated sample was centrifuged, and duplicate aliquots
[ C]MNTX (10 ␮M, 0.3 ␮Ci/ml) was incubated with the recombinant
SULT1A1, SULT1A2, SULT1A3, SULT1E1, and SULT2A1 (0.06–0.23 mg/
ml) in the presence of PAPS (25 ␮M) in 0.5 ml of phosphate buffer, pH 7.4,
containing 10 mM MgCl . PAPS was added to start the reaction after prein-
2
cubation at 37°C for 3 min. Control incubations lacked PAPS. Incubations
were stopped after 1 h by the addition of 0.5 ml of ice-cold acetonitrile.
TABLE 1
Initial rate conditions and incubation conditions used to measure kinetic parameters for MNTX reduction to M4 and M5 in liver cytosol of mice, rats, monkeys,
and humans, and for naltrexone reduction to 6␤-naltrexol in human liver cytosol
Linear Range
Protein Concentration
Conditions for Kinetic Study
Substrate
Species
Incubation Time
Incubation Time
Protein Concentration
MNTX Concentration
M4
M5
M4
M5
min
min
mg/ml
mg/ml
min
mg/ml
␮M
MNTX
Mouse
Rat
Monkey
Human
Human
N.A.
N.A.
0–60
0–120
0–120
0–120
0–60
0–6.0
N.A.
0–6.0
0–4.0
0–6.0
0–6.0
0–6.0
0–4.0
120
120
60
120
30
4.0
4.0
4.0
1.0
0.5
5–2000
5–5000
5–2000
0.5–5000
5–1000
0–120
Naltrexone
0–120
0–0.5
N.A., not applicable.
TABLE 2
Selective chemical inhibitors used to identify the enzymes responsible for the carbonyl reduction of MNTX
Chemical Inhibitors
Reductases
References
Phenolphthalein
Flufenamic acid
AKR1C1, 1C2, 1C3, 1C4
AKR1C1, 1C2, 1C3, 1C4
AKR1C2
(Atalla et al., 2000; Rosemond et al., 2004; Steckelbroeck et al., 2006)
(Atalla et al., 2000; Rosemond et al., 2004; Steckelbroeck et al., 2004, 2006)
(Steckelbroeck et al., 2006)
5
␤-Cholanic acid-3␣,7␣-diol
Medroxyprogesterone acetate
Indomethacin
Phenobarbital
AKR1C1, 1C2, 1C4
AKR1C1, 1C2, 1C3, 1C4, CR
AKR1A1(aldehyde reductase)
AKR1B1 (aldose reductase)
CR/SDR
(Atalla et al., 2000)
(Gebel and Maser, 1992; Maser et al., 2000; Bauman et al., 2005)
(Gebel and Maser, 1992; Maser et al., 2000; Rosemond and Walsh, 2004)
Ethacrynic acid
Menadione
(Maser et al., 2000; Atalla and Maser, 2001)
(Atalla et al., 2000; Maser et al., 2000; Porter et al., 2000; Atalla and Maser, 2001; Rosemond et
al., 2004)
CR/SDR
Quercetin
CR/SDR
(Gebel and Maser, 1992; Atalla et al., 2000; Maser et al., 2000; Atalla and Maser, 2001;
Rosemond et al., 2004)
4
-Methylpyrazole
Alcohol dehydrogenase
Quinone oxidoreductase
(Maser et al., 2000; Atalla and Maser, 2001)
(Gebel and Maser, 1992; Maser et al., 2000)
Dicumarol

IN VITRO METABOLISM AND ENZYME IDENTIFICATION FOR MNTX
803
Identification of Human Enzymes Involved in MNTX Reduction. To (4.0 mg/ml) in the presence of the NADPH-regenerating system and a chem-
identify the enzymes responsible for the carbonyl reduction of MNTX,
ical inhibitor (10 and 100 ␮M; Table 2) for 2 h under the conditions described
1
4
[
C]MNTX (500 ␮M, 0.5 ␮Ci/ml) was incubated with human liver cytosol above. Quercetin was dissolved in 10% dimethyl sulfoxide in methanol;
1
4
FIG. 1. Radiochromatograms of [ C]MNTX (50 ␮M) incubated with liver cytosol (4 mg/ml) for 2 h in the presence of NADPH.

8
04
TONG ET AL.
4
-methylpyrazole was dissolved in 9% dimethyl sulfoxide in water; dicumarol equipped with an electrospray ionization source and operated in the posi-
was dissolved in 1% sodium hydroxide in water; and medroxyprogesterone tive ionization mode. Full-scan spectra were acquired from m/z 100 to 1500
acetate was dissolved in chloroform. Other chemical inhibitors were dissolved
in methanol or water. Organic solvent in the incubations was controlled within
with a scan time of 0.9 s. The capillary and cone voltages were 3.5 kV and
25 V, respectively. The source block and desolvation gas temperatures
2
% of the incubation volume. Control incubations were conducted with the were 120 and 350°C, respectively. The desolvation gas flow was 900 l/h,
vehicles but in absence of the chemical inhibitors. Positive control incubations
with human liver cytosol (1.0 mg/ml) were conducted for 1 h with naltrexone
and the time of flight-MS resolution was approximately 8000 (m/⌬m).
Argon was used for collision-activated dissociation experiments at a pres-
sure setting of 13 psig, and the collision offset for acquiring tandem mass
(
100 ␮M) as the substrate in place of MNTX. The samples were prepared and
analyzed by HPLC with radioactivity flow detection for MNTX and UV spectra varied between 20 and 35 eV, depending on the metabolite.
detection at 280 nm for naltrexone. Duplicate incubations were prepared for
each condition.
HPLC. A Waters (Milford, MA) model 2695 HPLC system with a built-in
Isolated metabolite M2 was also analyzed using a TriVersa NanoMate
chip-based electrospray ionization system (Advion BioSciences, Inc., Ithaca,
NY). A backpressure of approximately 25 psi was used, and the voltage to the
autosampler was used for analysis. Separations were accomplished on a Luna electrospray ionization chip was 1.7 kV. Precleaned V-bottomed 96-well
C18(2) column (250 ϫ 4.6 mm, 5 ␮m) (Phenomenex, Torrance, CA). The storage plates (Advion BioSciences, Inc.) were used to hold the sample before
sample chamber in the autosampler was maintained at 4°C, whereas the introduction to the mass spectrometer. Metabolite M2 was isolated from
column was at 40°C. The mobile phase consisted of 0.05% trifluoroacetic acid SULT2A1 and SULT1E1 incubations and desalted using C18 ZipTips (Milli-
in water (A) and 0.05% trifluoroacetic acid in methanol (B) and was delivered pore Corporation) before being introduced to the mass spectrometer by flow
at 0.6 ml/min. The linear HPLC gradient started at 10% B and increased to
0% over 25 min and then increased to 30% over 5 min. A Flo-One ␤ Model
infusion at approximately 50 to 150 nl/min.
2
A525 radioactivity flow detector (PerkinElmer Life and Analytical Sciences)
with a 250-␮l LQTR flow cell and a Waters model 996 photodiode array UV
detector set to monitor at 280 nm were used for data acquisition. The flow rate
Results
Carbonyl Reduction of MNTX In Vitro by Hepatic Cytosol of
of Ultima Flo M scintillation fluid (PerkinElmer Life and Analytical Sciences) Mice, Rats, Monkeys, and Humans. MNTX reduction metabolites
was 3.0 ml/min, providing a mixing ratio of scintillation mixture to mobile methyl-6␣-naltrexol (M4) and methyl-6␤-naltrexol (M5) were de-
phase of approximately 5:1.
Mass Spectrometry. Liquid chromatography/mass spectrometry (LC/MS)
analysis used an Agilent Technologies (Santa Clara, CA) model 1100 HPLC
system including a binary pump and diode array UV detector. The UV detector
incubated with human hepatic cytosol in the presence of NADPH,
was set to monitor ␭ between 190 and 600 nm. Separations were accomplished
under the same conditions as described above, except that a Luna C18(2)
column (250 ϫ 2.0 mm, 5 ␮m; Phenomenex) column was used at ambient
temperature of approximately 20°C, and the mobile phase was delivered at 0.3
ml/min.
tected in human hepatic cytosol but not in hepatic or intestinal
microsomes or intestinal cytosol when MNTX was incubated in the
presence of the NADPH-regenerating system. When MNTX was
slightly lower activity was observed than with the NADPH-regener-
ating system, whereas metabolites were not detected when NADH
was used as the cofactor (data not shown).
Species differences were observed in metabolite profiles in liver
For hydrogen-deuterium exchange experiments, deuterium oxide was sub- cytosol (Fig. 1) and were generally consistent with previous in vivo
stituted for water in mobile phase A, and d -methanol was substituted for metabolism observations (Chandrasekaran et al., 2010). Metabo-
4
methanol in mobile phase B. During LC/MS sample analysis, up to 10 min of
lites formed in liver cytosol were identified by LC/MS and had the
the initial flow was diverted away from the mass spectrometer before evalu-
ation of metabolites.
Mass spectral data for the metabolites were obtained with a Micromass
spectra data for MNTX and its metabolites identified in liver
quadrupole-time of flight API-US mass spectrometer (Waters). It was
cytosol and enzyme preparations are summarized in Table 3. In
same retention times and mass spectra as those observed in human
plasma as described elsewhere (Chandrasekaran et al., 2010). Mass
mouse, monkey, and human liver cytosol, both M4 and M5 were
TABLE 3
observed. However, M4 was observed in greater amounts than M5
Retention times and mass spectral properties for MNTX and its metabolites
in the monkey and human cytosol but in much lower amounts than
observed in liver cytosol and enzyme preparations
M5 in mouse liver cytosol. Only M5 was observed in rat liver
cytosol, and only a trace amount of M4 was observed in dog liver
cytosol. Because of the low activity, a kinetic study was not
ϩ
Molecular Ion (M
)
Metabolite
t
R
Relevant Product Ions
With H O With D O
2
2
conducted with dog liver cytosol. The capacity (Vmax/K ) for the
min
m/z
m/z
m/z
m
formation of M4 was much greater in the monkey and human liver
cytosol than in the mouse and rat liver cytosol, whereas the
capacity for the formation of M5 was greater in mouse liver cytosol
than in liver cytosol of other species (Table 4).
MNTX 23.0
356
436
358
358
358
438
361
361
338, 302, 284, 227, 199, 112, 55
418, 382, 356, 302, 284, 227, 199, 112, 55
340, 304, 286, 229, 201, 112, 55
340, 304, 286, 229, 201, 112, 55
M2
M4
M5
14.0
20.5
25.5
TABLE 4
Kinetic parameters for the formation of M4 and M5 from MNTX by liver cytosol of mice, rats, monkeys, and humans, and of 6␤-naltrexol from naltrexone by human
liver cytosol
K
m
V
max
Vmax/K
m
Substrate
Species
M4
M5
M4
M5
M4
M5
␮
M
␮M
pmol/min/mg
pmol/min/mg
ml/min/mg
ml/min/mg
MNTX
Mouse
Rat
Monkey
Human
Human
Human
1220
N.D.
391
714
396
266
793
11.0
N.D.
80.0
43.7
125
0.009
N.D.
0.20
0.175
0.103
0.082
0.079
40.7
21.9
62.5
361
0.12
6
␤-Naltrexol
625
12.6
Naltrexone
49.7
N.D., metabolite not detected.

IN VITRO METABOLISM AND ENZYME IDENTIFICATION FOR MNTX
805
FIG. 2. Formation of M2 by sulfotransferase isoforms SULT2A1 and SULT1E1.
Identification of Sulfotransferase Isoforms for MNTX Sulfa-
TABLE 5
Percentage of inhibition of the formation of M4 and M5 from MNTX, and of
1
4
tion. When [ C]MNTX was incubated with recombinant human
sulfotransferases under the conditions used in the present study,
MNTX-3-sulfate (M2) was observed as the only product catalyzed
by SULT2A1 and SULT1E1 (Fig. 2). The identity of the M2
metabolite peak was confirmed by MS compared with the synthetic
standard of M2 (Chandrasekaran et al., 2010). No products were
6␤-naltrexol from naltrexone by chemical inhibitors
M4
M5
6␤-Naltrexol
Inhibitor concentration (␮M)
1
0
100
10
100
10
100
Phenolphthalein
Flufenamic acid
␤-Cholanic acid-3␣,7␣-diol
Medroxyprogesterone acetate
Indomethacin
Phenobarbital
Ethacrynic acid
Menadione
Quercetin
-Methylpyrazole
Dicumarol
81
N.I.
21
28
1
N.I.
8
6
89
N.I.
39
59
N.I.
2
27
35
28
6
76
25
46
36
9
N.I.
13
N.I.
7
97
53
67
67
34
N.I.
48
43
46
15
75
2
10
31
N.I.
2
4
7
9
N.I.
1
98
11
34
47
6
N.I.
31
26
32
9
1
4
detected when [ C]MNTX was incubated with SULT1A1,
SULT1A2, and SULT1A3.
5
Characterization and Identification of the Enzymes for MNTX
Reduction. MNTX reduction to M4 and M5 and naltrexone reduction
to 6␤-naltrexone in human liver cytosol were most effectively (up to
4
4
N.I.
9
8%) inhibited by phenolphthalein in a concentration-dependent man-
4
5
4
ner (Table 5). 5␤-Cholanic acid-3␣,7␣-diol and medroxyprogesterone
moderately inhibited the MNTX reduction by up to 67%. 5␤-Cholanic
acid-3␣,7␣-diol and medroxyprogesterone also inhibited naltrexone
reduction. Moderate inhibition of the reduction of MNTX and nal-
N.I.
N.I.
11
N.I., no inhibition.
trexone was observed with ethacrynic acid, menadione, and quercetin. changed by dicumarol, formation of M4 was increased 2.1-, 1.4-,
No inhibition of the reduction of MNTX or naltrexone was observed and 1.9-fold in a concentration-dependent manner by flufenamic
with 4-methlpyrazole and phenobarbital at concentrations up to acid, indomethacin, and dicumarol, respectively (data not shown).
1
00 ␮M. Whereas formation of M5 was moderately inhibited by These three chemical inhibitors showed no inhibitory effect on the
flufenamic acid, slightly inhibited by indomethacin, and un- reduction of naltrexone.

8
06
TONG ET AL.
Pfaff and Nill, 2004) and by medroxyprogesterone acetate, which
Discussion
The primary metabolic pathways for MNTX in humans are sulfa-
inhibits AKR1C1, AKR1C2, and AKR1C4 (Atalla et al., 2000).
␤-Cholanic acid-3␣,7␣-diol and medroxyprogesterone acetate inhib-
5
tion and carbonyl reduction (Chandrasekaran et al., 2010). MNTX
was metabolized via glucuronidation, hydroxylation, methylation, sul-
fation, and carbonyl reduction in nonhuman species. The species
differences in the carbonyl reduction of MNTX by hepatic cytosol in
the presence of NADPH were consistent with the in vivo data in mice,
rat, dogs, and humans. Both M4 and M5 were formed in liver cytosol
of mice and humans. However, M4 was observed in a higher amount
than M5 in human liver cytosol, which was consistent with human
plasma profiles, whereas M4 was a minor isomer in mouse liver
cytosol. Only trace amounts of M4 were observed in dog liver cytosol.
The kinetic parameters in human liver cytosol also showed that the
capacity for carbonyl reduction was approximately 100-fold greater
for naltrexone than for MNTX. The kinetic data for the formation of
ited naltrexone reduction by approximately 34 and 47% at 100 ␮M,
respectively. Flufenamic acid, a selective inhibitor of AKR1C1,
AKR1C2, and AKR1C3 and a weak inhibitor of AKR1C4 (Steckel-
broeck et al., 2004, 2006), inhibited the formation of M5 but increased
the formation of M4 up to 2.1-fold in a concentration-dependent
manner, whereas little inhibitory effect was observed on 6␤-naltrexol
formation. Indomethacin is an unspecific inhibitor of both AKRs and
CR and showed low to moderate potency with either substrate (Gebel
and Maser, 1992; Maser et al., 2000). Indomethacin had no inhibitory
effect on 6␤-naltrexol formation, slightly inhibited M5 formation, but
increased M4 formation. The lack of inhibitory effect by phenobar-
bital (an inhibitor of AKR1A1 and AKR1B1) on the reduction of
MNTX and naltrexone confirmed that the AKRs capable of reducing
xenobiotic ketones are primarily the AKR1C subfamily in humans
6␤-naltrexol from naltrexone in human liver cytosol were also con-
sistent with a literature report by Porter et al. (2000).
(
Atalla et al., 2000). Altogether, the data indicated that multiple
Of the five recombinant human sulfotransferase enzymes exam-
ined, only SULT1E1 and SULT2A1 catalyzed MNTX sulfation.
SULT1E1 expressed in human liver and jejunum is important for
steroid homeostasis (Gamage et al., 2006) and also has high affinity
for ␤-estradiol, estrone, and a number of synthetic estrogens (Gamage
et al., 2006). SULT2A1, which is present in human liver, adrenal,
small intestine, and other organs, is responsible for the sulfation of
hydroxysteroids, including dehydroepiandrosterone, androgens,
isoforms in the AKR1C subfamily are responsible for MNTX reduc-
tion, with major contribution probably from AKR1C4, which is vir-
tually liver-specific (Penning et al., 2000).
Ethacrynic acid, menadione, and quercetin, the inhibitors of CR or
SDR (Maser et al., 2000; Rosemond et al., 2004), did not appear to
affect the reduction of either MNTX or naltrexone significantly.
Because Breyer-Pfaff and Nill (2004) have shown that CR was not
pregnenolone, and bile acids (Gamage et al., 2006). SULT1A1, involved in the reduction of naltrexone, the decreases in the reduction
SULT1A2, and SULT1A3, which are responsible for sulfation of of MNTX and naltrexone may be a result of nonspecific inhibition by
phenolic compounds and catecholamines, showed no activity toward these inhibitors. Incubations with the chemical inhibitors (4-meth-
MNTX.
ylpyrazole and dicumarol, respectively) of alcohol dehydrogenase and
Carbonyl reduction of MNTX was catalyzed by NADPH-depen- quinone oxidoreductase showed that these two enzymes are not in-
dent cytosolic enzymes in human liver. The enzymes responsible for volved in the reduction of MNTX. However, the quinone oxidoreduc-
MNTX reduction were identified by using selective chemical inhibi- tase inhibitor dicumarol increased M4 formation by 1.9-fold at 100
tors of AKR isoforms, CR/SDR, alcohol dehydrogenase, and quinone ␮M, similar to flufenamic acid.
oxidoreductase. The reduction of both MNTX and naltrexone was
Stereoselective effects on MNTX reduction were observed with
most effectively inhibited by phenolphthalein, which inhibits AKR1C flufenamic acid, indomethacin, and dicumarol. Whereas they inhibited
isoforms with better selectivity toward AKR1C4 than for AKR1C1, the formation of the 6␤-epimer, they enhanced the formation of the
AKR1C2, or AKR1C3 (Steckelbroeck et al., 2006). Naltrexone re- 6␣-epimer. Such effects of these inhibitors were not observed with
duction has been shown to be catalyzed by AKR1C4, with minor naltrexone because it was reduced only to one epimer, 6␤-epimer.
contribution from AKR1C1 and AKR1C2 (Breyer-Pfaff and Nill, Enzyme activation in vitro has been well documented for cytochrome
2
004). Formation of M4 and M5 was inhibited by up to 67% by the P450 (P450) isoforms, particularly CYP3A (Fayz and Inaba, 1998).
AKR1C2-selective inhibitor 5␤-cholanic acid-3␣,7␣-diol (Breyer- Various mechanisms have been proposed for the activation of P450
FIG. 3. Metabolic pathways of MNTX in
humans.

IN VITRO METABOLISM AND ENZYME IDENTIFICATION FOR MNTX
807
Chandrasekaran A, Tong Z, Li H, Erve J, DeMaio W, Goljer I, McConnell O, Rotshteyn Y,
Hultin T, Talaat R, et al. (2010) Metabolism of intravenous methylnaltrexone in mice, rats,
dogs and humans. Drug Metab Dispos doi: 10.1124/dmd.109.031179.
El-Kabbani O, Wilson DK, Petrash M, and Quiocho FA (1998) Structural features of the aldose
reductase and aldehyde reductase inhibitor-binding sites. Mol Vis 4:19.
Fayz S and Inaba T (1998) Zidovudine azido-reductase in human liver microsomes: activation by
ethacrynic acid, dipyridamole, and indomethacin and inhibition by human immunodeficiency
virus protease inhibitors. Antimicrob Agents Chemother 42:1654–1658.
Gamage N, Barnett A, Hempel N, Duggleby RG, Windmill KF, Martin JL, and McManus ME
(2006) Human sulfotransferases and their role in chemical metabolism. Toxicol Sci 90:5–22.
Gebel T and Maser E (1992) Characterization of carbonyl reducing activity in continuous cell
lines of human and rodent origin. Biochem Pharmacol 44:2005–2012.
enzymes. An activator such as 7,8-benzoflavone may enhance the
interaction of P450 reductase and P450 (Huang et al., 1981). Flavones
are allosteric effectors that may increase catalytic efficiency by de-
creasing the Km and increasing the Vmax (Schwab et al., 1988). Shou
et al. (1994) showed simultaneous binding of two substrates to the
P450 active site. Binding of a substrate to the active site can alter the
orientation of another substrate, thereby changing the regioselectivity
of the metabolism. Enzyme activation of P450 reductase and cyto-
chrome b reductase by ethacrynic acid, dipyridamole, and indometh-
5
Huang MT, Chang RL, Fortner JG, and Conney AH (1981) Studies on the mechanism of
activation of microsomal benzo[a]pyrene hydroxylation by flavonoids. J Biol Chem 256:
acin has also been reported for the reduction of 3Ј-azido-3Ј-deoxy-
thymidine to 3Ј-amino-3Ј-deoxythymidine (Fayz and Inaba, 1998).
However, activation of AKR enzyme activity has not been reported.
6
829–6836.
Jez JM, Bennett MJ, Schlegel BP, Lewis M, and Penning TM (1997) Comparative anatomy of
the aldo-keto reductase superfamily. Biochem J 326:625–636.
Lovering AL, Ride JP, Bunce CM, Desmond JC, Cummings SM, and White SA (2004) Crystal
structures of prostaglandin D(2) 11-ketoreductase (AKR1C3) in complex with the nonsteroidal
anti-inflammatory drugs flufenamic acid and indomethacin. Cancer Res 64:1802–1810.
Maser E, Stinner B, and Atalla A (2000) Carbonyl reduction of 4-(methylnitrosamino)-1-(3-
pyridyl)-1-butanone (NNK) by cytosolic enzymes in human liver and lung. Cancer Lett
148:135–144.
Matsunaga T, Shintani S, and Hara A (2006) Multiplicity of mammalian reductases for xeno-
biotic carbonyl compounds. Drug Metab Pharmacokinet 21:1–18.
Penning TM, Burczynski ME, Jez JM, Hung CF, Lin HK, Ma H, Moore M, Palackal N, and
Ratnam K (2000) Human 3alpha-hydroxysteroid dehydrogenase isoforms (AKR1C1-
AKR1C4) of the aldo-keto reductase superfamily: functional plasticity and tissue distribution
reveals roles in the inactivation and formation of male and female sex hormones. Biochem J
AKRs share a common (␣/␤) -barrel three-dimensional fold, with
8
highly conserved nicotinamide-cofactor-binding pocket and active
site (Jez et al., 1997; Sanli et al., 2003). Jez et al. (1997) proposed that
the three loops on the C-terminal side of the barrel play potential roles
in determining the positional and stereospecificity of the reaction.
Studies have shown that the active site of AKRs adapts itself to bind
tightly to different inhibitors (Urzhumtsev et al., 1997; El-Kabbani et
al., 1998). Lovering et al. (2004) examined the X-ray crystal struc-
tures of AKR1C3 complexed with indomethacin or flufenamic acid.
One molecule of indomethacin is bound in the active site, whereas
flufenamic acid binds to both the active site and the ␤-hairpin loop at
the opposite end of the central ␤-barrel. It is not clear how flufenamic
acid, indomethacin, and dicumarol interact with the enzymes and
change the stereochemistry to have differential effects on the forma-
tion of the 6␣- and 6␤-epimers. Because these effects were not
observed with naltrexone, the methyl group and charge in the mole-
cule of MNTX may have contributed to the observations.
3
51:67–77.
Porter SJ, Somogyi AA, and White JM (2000) Kinetics and inhibition of the formation of
beta-naltrexol from naltrexone in human liver cytosol. Br J Clin Pharmacol 50:465–471.
6
Rosemond MJ, St John-Williams L, Yamaguchi T, Fujishita T, and Walsh JS (2004) Enzymol-
ogy of a carbonyl reduction clearance pathway for the HIV integrase inhibitor, S-1360: role of
human liver cytosolic aldo-keto reductases. Chem Biol Interact 147:129–139.
Rosemond MJ and Walsh JS (2004) Human carbonyl reduction pathways and a strategy for their
study in vitro. Drug Metab Rev 36:335–361.
Sanli G, Dudley JI, and Blaber M (2003) Structural biology of the aldo-keto reductase family of
enzymes: catalysis and cofactor binding. Cell Biochem Biophys 38:79–101.
Schwab GE, Raucy JL, and Johnson EF (1988) Modulation of rabbit and human hepatic
cytochrome P-450-catalyzed steroid hydroxylations by alpha-naphthoflavone. Mol Pharmacol
33:493–499.
Shou M, Grogan J, Mancewicz JA, Krausz KW, Gonzalez FJ, Gelboin HV, and Korzekwa KR
In summary, sulfation and carbonyl reduction are the major
metabolic pathways for MNTX in humans (Fig. 3). Species differ-
ences observed in the carbonyl reduction of MNTX to M4 and M5
in hepatic cytosol were consistent with previously reported in vivo
metabolite profiles. The present study identified SULT2A1 and
SULT1E1 as the sulfotransferases involved in the sulfation of
MNTX in humans. AKR1C4 appeared to play a major role in the
carbonyl reduction of MNTX, although multiple enzymes in the
AKR1C subfamily may be involved in the reduction. Carbonyl
reductase may have no or limited contribution to MNTX reduction.
The mechanism of the activation of the formation of M4 while
inhibiting the formation of M5 by flufenamic acid, indomethacin,
and dicumarol warrants further investigation.
(
1994) Activation of CYP3A4: evidence for the simultaneous binding of two substrates in a
cytochrome P450 active site. Biochemistry 33:6450–6455.
Steckelbroeck S, Jin Y, Gopishetty S, Oyesanmi B, and Penning TM (2004) Human cytosolic
3
alpha-hydroxysteroid dehydrogenases of the aldo-keto reductase superfamily display signif-
icant 3beta-hydroxysteroid dehydrogenase activity: implications for steroid hormone metab-
olism and action. J Biol Chem 279:10784–10795.
Steckelbroeck S, Oyesanmi B, Jin Y, Lee SH, Kloosterboer HJ, and Penning TM (2006) Tibolone
metabolism in human liver is catalyzed by 3alpha/3beta-hydroxysteroid dehydrogenase activ-
ities of the four isoforms of the aldo-keto reductase (AKR)1C subfamily. J Pharmacol Exp
Ther 316:1300–1309.
Urzhumtsev A, T eˆ te-Favier F, Mitschler A, Barbanton J, Barth P, Urzhumtseva L, Biellmann JF,
Podjarny A, and Moras D (1997) A ‘specificity’ pocket inferred from the crystal structures of
the complexes of aldose reductase with the pharmaceutically important inhibitors tolrestat and
sorbinil. Structure 5:601–612.
Wall ME, Brine DR, and Perez-Reyes M (1981) Metabolism and disposition of naltrexone in
man after oral and intravenous administration. Drug Metab Dispos 9:369–375.
Wall ME, Perez-Reyes M, Brine DR, and Cook CE (1984) Naltrexone disposition in man after
subcutaneous administration. Drug Metab Dispos 12:677–682.
Yuan CS, Foss JF, O’Connor M, Moss J, and Roizen MF (1998) Gut motility and transit changes
in patients receiving long-term methadone maintenance. J Clin Pharmacol 38:931–935.
Yuan CS, Foss JF, O’Connor M, Osinski J, Karrison T, Moss J, and Roizen MF (2000)
Methylnaltrexone for reversal of constipation due to chronic methadone use: a randomized
controlled trial. JAMA 283:367–372.
Yuan CS, Foss JF, O’Connor M, Osinski J, Roizen MF, and Moss J (1999) Effects of intravenous
methylnaltrexone on opioid-induced gut motility and transit time changes in subjects receiving
chronic methadone therapy: a pilot study. Pain 83:631–635.
Yuan CS, Foss JF, O’Connor M, Toledano A, Roizen MF, and Moss J (1996) Methylnaltrexone
prevents morphine-induced delay in oral-cecal transit time without affecting analgesia: a
double-blind randomized placebo-controlled trial. Clin Pharmacol Ther 59:469–475.
Yuan CS, Wei G, Foss JF, O’Connor M, Karrison T, and Osinski J (2002) Effects of subcuta-
neous methylnaltrexone on morphine-induced peripherally mediated side effects: a double-
blind randomized placebo-controlled trial. J Pharmacol Exp Ther 300:118–123.
Acknowledgments. We thank Daniel Evcic for technical contribu-
tions to this study.
References
Atalla A, Breyer-Pfaff U, and Maser E (2000) Purification and characterization of oxidoreduc-
tases-catalyzing carbonyl reduction of the tobacco-specific nitrosamine 4-methylnitrosamino-
1
-(3-pyridyl)-1-butanone (NNK) in human liver cytosol. Xenobiotica 30:755–769.
Atalla A and Maser E (2001) Characterization of enzymes participating in carbonyl reduction of
4
1
-methylnitrosamino-1-(3-pyridyl)-1-butanone (NNK) in human placenta. Chem Biol Interact
30–132:737–748.
Bauman DR, Rudnick SI, Szewczuk LM, Jin Y, Gopishetty S, and Penning TM (2005)
Development of nonsteroidal anti-inflammatory drug analogs and steroid carboxylates selec-
tive for human aldo-keto reductase isoforms: potential antineoplastic agents that work inde-
pendently of cyclooxygenase isozymes. Mol Pharmacol 67:60–68.
Breyer-Pfaff U and Nill K (2004) Carbonyl reduction of naltrexone and dolasetron by oxi-
doreductases isolated from human liver cytosol. J Pharm Pharmacol 56:1601–1606.
Brown DR and Goldberg LI (1985) The use of quaternary narcotic antagonists in opiate research.
Neuropharmacology 24:181–191.
Address correspondence to: Appavu Chandrasekaran, Pfizer Inc., 500 Ar-
cola Road, Collegeville, PA 19426. E-mail: chandra@wyeth.com