Metabolism of intravenous methylnaltrexone in mice, rats, dogs, and humans

Article abstract of DOI:10.1124/dmd.109.031179

Methylnaltrexone (MNTX), a selective μ-opioid receptor antagonist, functions as a peripherally acting receptor antagonist in tissues of the gastrointestinal tract. This report describes the metabolic fate of [ ³H]MNTX or [¹⁴C]MNTX bromide in mice, rats, dogs, and humans after intravenous administration. Separation and identification of plasma and urinary MNTX metabolites was achieved by high-performance liquid chromatography-radioactivity detection and liquid chromatography/mass spectrometry. The structures of the most abundant human metabolites were confirmed by chemical synthesis and NMR spectroscopic analysis. Analysis of radioactivity in plasma and urine showed that MNTX underwent two major pathways of metabolism in humans: sulfation of the phenolic group to MNTX-3-sulfate (M2) and reduction of the carbonyl group to two epimeric alcohols, methyl-6α-naltrexol (M4) and methyl-6β-naltrexol (M5). Neither naltrexone nor its metabolite 6α-naltrexol were detected in human plasma after administration of MNTX, confirming an earlier observation that N-demethylation was not a metabolic pathway of MNTX in humans. The urinary metabolite profiles in humans were consistent with plasma profiles. In mice, the circulating and urinary metabolites included M5, MNTX-3-glucuronide (M9), 2-hydroxy-3-O-methyl MNTX (M6), and its glucuronide (M10). M2, M5, M6, and M9 were observed in rats. Dogs produced only one metabolite, M9. In conclusion, MNTX was not extensively metabolized in humans. Conversion to methyl-6-naltrexol isomers (M4 and M5) and M2 were the primary pathways of metabolism in humans. MNTX was metabolized to a higher extent in mice than in rats, dogs, and humans. Glucuronidation was a major metabolic pathway in mice, rats, and dogs, but not in humans. Overall, the data suggested species differences in the metabolism of MNTX. Copyright

Full text of DOI:10.1124/dmd.109.031179

0090-9556/10/3804-606–616$20.00
DRUG METABOLISM AND DISPOSITION
Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics
DMD 38:606–616, 2010
Vol. 38, No. 4
31179/3568767
Printed in U.S.A.
Metabolism of Intravenous Methylnaltrexone in Mice, Rats, Dogs,
and Humans
Appavu Chandrasekaran, Zeen Tong, Hongshan Li, John C. L. Erve, William DeMaio,
Igor Goljer, Oliver McConnell, Yakov Rotshteyn, Theresa Hultin, Rasmy Talaat, and
JoAnn Scatina
Pharmacokinetics, Dynamics and Metabolism (A.C., Z.T., H.L., J.C.L.E., W.D., T.H., R.T., J.S.) and Discovery Analytical
Chemistry (I.G., O.M.), Pfizer, Collegeville, Pennsylvania; and Progenics Pharmaceuticals, Inc., Tarrytown, New York (Y.R.)
Received November 16, 2009; accepted January 5, 2010
ABSTRACT:
Methylnaltrexone (MNTX), a selective ␮-opioid receptor antago- were detected in human plasma after administration of MNTX,
nist, functions as a peripherally acting receptor antagonist in tis-
confirming an earlier observation that N-demethylation was not a
sues of the gastrointestinal tract. This report describes the meta- metabolic pathway of MNTX in humans. The urinary metabolite
bolic fate of [3H]MNTX or [14C]MNTX bromide in mice, rats, dogs, profiles in humans were consistent with plasma profiles. In mice,
and humans after intravenous administration. Separation and iden- the circulating and urinary metabolites included M5, MNTX-3-gluc-
tification of plasma and urinary MNTX metabolites was achieved by uronide (M9), 2-hydroxy-3-O-methyl MNTX (M6), and its glucuro-
high-performance liquid chromatography-radioactivity detection nide (M10). M2, M5, M6, and M9 were observed in rats. Dogs
and liquid chromatography/mass spectrometry. The structures of produced only one metabolite, M9. In conclusion, MNTX was not
the most abundant human metabolites were confirmed by chemi- extensively metabolized in humans. Conversion to methyl-6-nal-
cal synthesis and NMR spectroscopic analysis. Analysis of radio- trexol isomers (M4 and M5) and M2 were the primary pathways of
activity in plasma and urine showed that MNTX underwent two metabolism in humans. MNTX was metabolized to a higher extent
major pathways of metabolism in humans: sulfation of the phenolic in mice than in rats, dogs, and humans. Glucuronidation was a
group to MNTX-3-sulfate (M2) and reduction of the carbonyl group major metabolic pathway in mice, rats, and dogs, but not in hu-
to two epimeric alcohols, methyl-6␣-naltrexol (M4) and methyl-6␤- mans. Overall, the data suggested species differences in the me-
naltrexol (M5). Neither naltrexone nor its metabolite 6␤-naltrexol tabolism of MNTX.
Methylnaltrexone (MNTX) bromide [(5␣)-17-(cyclopropylmethyl)- ularly their inhibition of gastrointestinal motility, with no effect on
3,14-dihydroxy-17-methyl-4,5-epoxymorphinanium-17-ium-6-one] is a centrally mediated analgesia or opioid tolerance.
selective peripherally acting ␮-opioid receptor antagonist. In vitro testing
shows that MNTX binds to ␮ receptors at 8-fold higher potency than to
␬ receptors, and it does not interact with ␦ receptors. MNTX is a
quaternary derivative of the opioid antagonist naltrexone (Brown and
Goldberg, 1985). The addition of the methyl group to naltrexone at the
nitrogen atom forms MNTX, an inherently positively charged compound
with greater polarity and lower lipid solubility than naltrexone. MNTX is
restricted from crossing the blood-brain barrier in animals at pharmaco-
logically relevant doses (Brown and Goldberg, 1985; Yuan et al., 1996,
1998). MNTX prevents the inhibition of opioid-induced gut contractions
in guinea pig ileum (Chen and Rosow, 2007). Studies in a variety of
animal species have shown that peripherally or systemically administered
MNTX blocks the peripherally mediated side effects of opioids, partic-
MNTX reverses the opioid-induced delay in both gastric emptying
and transit time without affecting analgesia in human volunteers
(Yuan et al., 1996, 2002). In a randomized placebo-controlled study in
long-term methadone users, MNTX was shown to induce laxation
(Yuan et al., 2000). In a phase 3 clinical trial, subcutaneous injection
of MNTX rapidly induced laxation in patients with advanced illness
who were receiving opioid therapy for pain. In addition, MNTX
treatment did not affect central analgesia or lead to opioid withdrawal
(Thomas et al., 2008). Methylnaltrexone bromide (RELISTOR; Pro-
genics Pharmaceuticals, Inc., Tarrytown, NY) is currently approved in
the United States as an injectable medication 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, and is approved elsewhere around the world for
similar indications.
Disclosure: Progenics Pharmaceuticals, Inc. has a proprietary commercial
interest in methylnaltrexone. Y.R. is an employee and stockholder of Progenics
Pharmaceuticals, Inc.
A limited number of studies have examined the metabolism of
MNTX in animals and humans (Misra et al., 1987; Kim et al., 1989;
Kotake et al., 1989; Murphy et al., 2001). The focus of those studies
was to evaluate the potential uptake into the brain in rats and the
Article, publication date, and citation information can be found at
http://dmd.aspetjournals.org.
doi:10.1124/dmd.109.031179.
ABBREVIATIONS: MNTX, methylnaltrexone; HPLC, high-performance liquid chromatography; M2, MNTX-3-sulfate; LC/MS, liquid chromatog-
raphy/mass spectrometry; M4, methyl-6␣-naltrexol; M5, methyl-6␤-naltrexol; AUC, area under curve; MS/MS, tandem mass spectrometry; 2D,
two dimensional; M6, 2-hydroxy-3-O-methyl MNTX; M9, MNTX-3-glucuronide; M10, and 2-hydroxy-3-O-methyl MNTX glucuronide.
606

IN VIVO METABOLISM OF METHYLNALTREXONE
607
under a nitrogen stream in a TurboVap LV evaporator (Caliper Life Sciences),
and reconstituted in CD3OD for NMR spectroscopic analysis.
extent of demethylation in laboratory animals and humans. The data
showed that the penetration of MNTX into the rat brain was restricted.
MNTX was not demethylated to any significant extent in humans.
Mice and rats had a slightly higher ability to N-demethylate MNTX
than dogs or humans. The present studies were performed to deter-
mine the metabolism of radiolabeled MNTX in mice, rats, dogs, and
humans after intravenous administration.
Animal Studies. Male CD-1 mice and Sprague-Dawley rats were purchased
from Charles River Laboratories, Inc. (Wilmington, MA). The dogs were from
Wyeth in-house colony. The dose was prepared in 0.9% saline for all non-
clinical species. Twenty-four nonfasted male mice, weighing from 26.5 to
34.8 g at the time of dosing, were given a single 5 mg/kg (400 ␮Ci/kg) dose
of [14C]MNTX in a volume of 5.0 ml/kg via the tail vein. Sixteen nonfasted
male rats, weighing 294 to 336 g at the time of dosing, were given a single 5
mg/kg (300 ␮Ci/kg) dose of [3H]MNTX in a volume of 1.0 ml/kg via the tail
vein. Four nonfasted male beagle dogs, weighing 8.7 to 12.2 kg at the time of
dosing, were given a single 5 mg/kg (50 ␮Ci/kg) dose of [3H]MNTX at a
volume of 1.0 ml/kg via the saphenous vein. Mice were kept as a group for
each time point, and rats and dogs were kept individually in metabolic cages.
All animal housing and care were conducted in Association for Assessment
and Accreditation of Laboratory Animal Care-accredited facilities. Animal
care and use for this investigation was approved by the Wyeth Institutional
Animal Care and Use Committee. Animal rooms were maintained on a 12-h
light/dark cycle. Animals were provided food and water ad libitum.
Blood samples were collected at 0.25, 1, 4, and 24 h after dose adminis-
tration from mice (6/time point) and rats (4/time point) by cardiac puncture and
from the jugular vein of dogs. Potassium EDTA was used as the anticoagulant,
and plasma was immediately harvested from the blood by centrifugation at
4°C. Urine samples were collected from animals on dry ice at 0 to 24 h
postdose. All biological specimens were stored at approximately Ϫ70°C until
analysis.
Human Study. This study was an open-label, phase I, single-dose study of
the pharmacokinetics, mass balance, and disposition of intravenously admin-
istered [14C]methyltrexone in normal, healthy volunteers. Six healthy male
human subjects were administered a single 0.3 mg/kg (100 ␮Ci/subject) i.v.
dose of [14C]MNTX. Samples of expired CO2 for the determination of radio-
activity were collected every 15 min between 0 and 4 h postdosing and every
30 min between 4 and 8 h postdosing. Plasma samples were collected predose
and at 2, 5, 15, 30, and 45 min, and 1, 1.5, 2, 3, 4, 5, 6, 8, 10, 12, 16, 24, 36,
48, 72, 96, and 120 h postdose. Urine samples were collected at 0 to 24, 24 to
48, 48 to 72, 72 to 96, 96 to 120, and 120 to 144-h intervals postdose. Fecal
samples were collected at 24-h intervals through 168 h postdose. The plasma
and urine samples were stored at Ϫ70°C until analysis for metabolite profiles.
The protocol, the investigator’s brochure, and the informed consent forms were
submitted to the study site institutional review board for review and written
approval. Subsequent amendments to the protocols and/or any revisions to the
informed consent forms were submitted for institutional review board review
and written approval. This study was conducted in accordance with ethical
principles that have origins in the Declaration of Helsinki and in any amend-
ments that were in place when the study was conducted. This study were also
designed and performed in compliance with Good Clinical Practice guidelines.
Written informed consent was obtained from all subjects before their enroll-
ment. WinNonlin Enterprise (version 4.1; Pharsight, Mountain View, CA) was
used to calculate the area under the curve (AUC) values.
Determination of Radioactivity. Aliquots of plasma (50 ␮l) and urine (200
␮l) samples were analyzed for radioactivity with a Tri-Carb model 3100
TR/LL liquid scintillation counter (PerkinElmer Life and Analytical Sciences)
using 10 ml of Ultima Gold as the scintillation fluid. Radioactivity detection
for HPLC was accomplished with a TopCount microplate reader or an in-line
Flo-One radioactivity flow detector (PerkinElmer Life and Analytical Sci-
ences). For plasma samples, the eluent was collected at 10-s intervals into
96-well LumaPlates (PerkinElmer Life and Analytical Sciences). The plates
were dried overnight in an oven at 40°C and analyzed by the TopCount NXT
radiometric microplate reader. For urine samples, a Flo-One ␤ Model A525
radioactivity flow detector (PerkinElmer Life and Analytical Sciences) with a
250-␮l flow cell and a Waters model 996 photodiode array UV detector
(Waters, Milford, MA) set to monitor 280 nm were used for data acquisition.
The flow rate of Ultima Flo M scintillation fluid was 3.0 ml/min, providing a
mixing ratio of scintillation cocktail to mobile phase of approximately 5:1.
Sample Preparation. Animal plasma samples collected at 0.25, 1, and 4 h
and human plasma samples collected at 2, 5, 15, 30, and 45 min, and 1, 1.5, 2,
and 4 h postdose were pooled across individuals by combining an equal
volume for each time point. Aliquots (3.0 ml) of the pooled plasma were mixed
Materials and Methods
Materials. [3H]MNTX bromide salt (lot number SEL/1674) and [14C]MNTX
bromide salt (lot number 2108DCR005-1) were 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 129.4 ␮Ci/mg (158.4 ␮Ci/mg as the free base) for
[3H]MNTX bromide and 125.1 ␮Ci/mg (153.0 ␮Ci/mg as the free base) for
[
14C]MNTX bromide. Nonradiolabeled MNTX bromide salt (lot number
H10207) was synthesized by Mallinckrodt Baker, Inc. (Phillipsburg, NJ) and
had a chemical purity of 100%. The structure of MNTX including the sites of
3H and 14C labels is shown in Fig. 1. Unless indicated otherwise, when
referring to [14C]MNTX or MNTX, the free-base form is assumed. Sulfur
trioxide-N-triethylamine complex, sodium borohydride, naltrexone, 6␤-
naltrexol hydrate, glucose 6-phosphate, NADPϩ, glucose-6-phosphate de-
hydrogenase, and ␤-glucuronidase (10,100 units/mg, type B-10 from bo-
vine liver) were obtained from Sigma-Aldrich (Milwaukee, WI). Ultima
Gold and Ultima Flo M scintillation cocktails were purchased from
PerkinElmer Life and Analytical Sciences (Waltham, MA). Solvents used
for extraction and for chromatographic analysis were of high-performance
liquid chromatography (HPLC) or reagent grade and were purchased from
EMD Chemicals (Gibbstown, NJ). Deuterium oxide (D2O) and d4-methanol
(CD3OD) were obtained from Cambridge Isotope Laboratories, Inc.
(Andover, MA).
Synthesis of MNTX Metabolites. MNTX-3-sulfate, the reference standard
for MNTX metabolite M2, was synthesized according to the method of Jones
et al. (2005). In brief, MNTX bromide (74 mg) was dissolved in dioxane (20
ml) in a water bath (50°C). Sulfur trioxide-N-triethylamine complex (310 mg)
was added to the solution, and the mixture was maintained at 50°C for 2 h. The
dioxane was decanted, and the precipitate on the glass was dissolved in 1 ml
of a mixture of water and methanol (1:1, v/v). Purification was accomplished
by UV detection by using a HPLC method described later. The purified
fraction was analyzed by liquid chromatography/mass spectrometry (LC/MS),
dried under a nitrogen stream in a TurboVap LV evaporator (Caliper Life
Sciences, Hopkinton, MA), and reconstituted in CD3OD for NMR spectro-
scopic analysis.
Methyl-6␣-naltrexol and methyl-6␤-naltrexol, the reference standards for
MNTX metabolites M4 and M5, respectively, were synthesized by the method
of Malspeis et al. (1975) for naltrexone. In brief, MNTX (100 mg) was
dissolved in 95% ethanol (5 ml), and the alcoholic solution was cooled in an
ice bath. Sodium borohydride (25 mg) was slowly added with stirring, and the
mixture was brought to room temperature and stirred for an additional 2 h. The
solvent was evaporated under a nitrogen stream in a TurboVap LV evaporator
(Caliper Life Sciences), and the residue was dissolved in water (2 ml). After
acidification with 1N HCl, aliquots of the solution were injected onto an HPLC
column and the products were isolated after UV detection by using the HPLC
method described later. The purified fraction was analyzed by LC/MS, dried
FIG. 1. Structure of methylnaltrexone bromide. ء indicates the site of the 14C label.
T indicates the site of the 3H labels.

608
CHANDRASEKARAN ET AL.
with 3.0 ml of methanol, placed on ice for approximately 10 min and then
acetic acid in methanol (B, v/v) and was delivered at 0.6 ml/min. The linear
centrifuged at 4°C. The supernatant was transferred to a clean tube. The protein HPLC gradient started at 10% B and increased to 20% over 25 min and then
pellets were extracted twice with 3.0 ml of methanol. The supernatants from increased to 30% over 5 min. Flo-One analytical software (version 3.65;
precipitation and extraction were pooled, mixed, and evaporated at 22°C under PerkinElmer Life and Analytical Sciences) was used to integrate the radioac-
nitrogen in a TurboVap LV evaporator (Caliper Life Sciences) to approxi- tive peaks. DataFlo software utility (beta version 0.55; PerkinElmer Life and
mately 0.8 ml. The concentrated extract was centrifuged, the supernatant
volume was measured, and the extraction efficiency was determined by ana-
Analytical Sciences) was used to convert ASCII files from the TopCount NXT
microplate counter into the required format for processing in Flo-One analysis
lyzing duplicate 10-␮l aliquots for radioactivity concentrations. The extraction software.
method recovered greater than 80% of plasma radioactivity. An aliquot (200
␮l) of the supernatant was injected onto the HPLC column, and fractions were
collected for determination of radioactivity as described above. The 24-h
plasma samples collected from animals and the human plasma samples col-
lected after 4 h postdose were not analyzed for metabolite profiles due to low
Mass Spectrometry. Plasma extracts and urine samples were analyzed by
LC/MS for metabolite characterization. Potential N-demethylation of MNTX
in humans was investigated by analyzing plasma samples for naltrexone and
6␤-naltrexol. The limit of detection was 6 ng/ml for naltrexone and 3 ng/ml for
6␤-naltrexol. An Agilent model 1100 HPLC system including a binary pump
radioactivity concentrations. Urine samples were analyzed by direct injection and diode array UV detector (Agilent Technologies, Santa Clara, CA) was
into the HPLC system. used with the mass spectrometer described below. The HPLC conditions were
High-Performance Liquid Chromatography. A Waters model 2695 the same as described above except that a Luna C18(2) column (250 ϫ 2.0
HPLC system (Waters) with a built-in autosampler was used for analysis. mm, 5 ␮m) was used and the mobile phase was delivered at 0.3 ml/min. For
Separations were accomplished on a Luna C18(2) column (250 ϫ 4.6 mm, 5 hydrogen-deuterium exchange experiments, D2O was substituted for water in
␮m) (Phenomenex, Torrance, CA). The sample chamber in the autosampler
was maintained at 4°C, while the column was at 40°C. The mobile phase
consisted of 0.05% trifluoroacetic acid in water (A, v/v) and 0.05% trifluoro-
mobile phase A and CD3OD was substituted for methanol in mobile phase B.
During LC/MS sample analysis, up to 10 min of the initial flow was diverted away
from the mass spectrometer before evaluation of metabolites. For some LC/MS
FIG. 2. Metabolite profiles of MNTX in plasma of mice, rats, dogs, and humans 1 h after a single bolus intravenous administration.

IN VIVO METABOLISM OF METHYLNALTREXONE
609
administered a single 0.3 mg/kg (100 ␮Ci) dose of [14C]MNTX as a
1-min i.v. infusion, and all six subjects completed the study. Renal
excretion was the primary route of elimination of total radioactivity,
with mean total percent excretion of 53.6. Fecal recovery in the 0 to
168-h collection period was 17.3%. Exhaled 14CO2 accounted for less
than 0.06% of administered radioactivity in all subjects.
Metabolite Profiles in Plasma. Metabolite profiles of extracts of
plasma samples obtained from mice, rats, dogs, and humans at 1 h after
a single bolus intravenous dose of radiolabeled MNTX are shown in Fig. 2.
The recovery of radioactivity into methanol was greater than 80% for all
four species. The metabolites in plasma were identified based on the
retention times of the synthetic reference standards, where available, and
liquid chromatography MS/MS fragmentation patterns.
analyses, radioactivity was simultaneously monitored with a Radiomatic 150TR
radioactivity flow detector (PerkinElmer Life and Analytical Sciences) to estimate
retention times of radiolabeled metabolites in LC/MS data. Flow from the HPLC
was split between the mass spectrometer and radioactivity detector such that the
flow rate to the mass spectrometer was approximately 100 ␮l/min.
A Micromass Q-TOF API-US mass spectrometer (Waters) equipped with an
electrospray ionization source and operated in the positive ionization mode
was used for metabolite identification. The capillary voltage was 2.0 kV and
the cone voltage was 30 V. The collision offset for tandem mass spectrometry
(MS/MS) analysis was 20 to 30 eV. The source block temperature and the
desolvation gas temperature were set at 120 and 350°C, respectively. The
desolvation gas flow and the cone gas flow were 950 and 50 l/h, respectively.
The collision gas pressure setting was 13 psig. Time-of-flight mass spectrom-
etry resolution (m/⌬m) was approximately 8000, and the radio frequency
setting was 0.2. Micromass MassLynx software (version 4.0; Waters Corp.)
was used for analysis of LC/MS data. Microsoft Excel 2000 (Microsoft,
Redmond, WA) was used for calculations.
NMR Spectroscopy. Synthetic MNTX metabolites were dissolved in
CD3OD (160 ␮l) and transferred to an NMR tube (3 mm OD). For purpose of
comparison, a saturated MNTX solution was also prepared in CD3OD. Data
were collected on Varian Inova 500 MHz (Varian, Inc., Palo Alto, CA) and
Bruker Avance 600 MHz spectrometers (Bruker, Newark, DE) each equipped
with a 3-mm z-gradient indirect detection probe. One-dimensional proton (1H)
NMR data and two-dimensional (2D) gCOSY, TOSY, 1H-13C gHSQC, and
1H-13C gHMBC NMR data were acquired. All spectra were referenced to the
signal of CHD2OD at 3.32 ppm in the 1H spectra and 48.1 ppm in the 13C
spectra. All spectra were measured at room temperature.
After a single 0.3 mg/kg i.v. administration of [14C]MNTX to healthy
human volunteers, unchanged MNTX was the major drug-related com-
ponent in pooled plasma, representing 90.3, 79.2, 67.2, and 33.0% of
plasma radioactivity at 0.25, 0.5, 2, and 4 h postdose, respectively.
Plasma metabolites included M2 (0.7–25.0% of total plasma radioactiv-
ity), M4 (0.6–11.9% of total plasma radioactivity), and M5 (0.5–7.0% of
total plasma radioactivity). Other minor (each less than 6% of plasma
radioactivity) radioactive peaks observed in plasma extracts were not
characterized due to low concentrations. Profiles determined in samples
collected after 4 h are not reported due to low levels of radioactivity.
Naltrexone, the N-demethylation product, and its metabolite 6␤-naltrexol
were not detected in plasma of humans, as determined by an LC/MS
method. The limit of detection was estimated to be approximately 6
ng/ml for naltrexone and 3 ng/ml for 6␤-naltrexol.
Results
Human Study. Six male subjects were enrolled in this study to
evaluate the pharmacokinetics of MNTX, total radioactivity in
plasma, determine the urinary and fecal recovery of radioactivity, and
After a single 5 mg/kg i.v. dose of [14C]MNTX to mice, the parent
compound represented 40.8, 20.2, and 25.7% of total plasma radio-
characterize the metabolite profiles of MNTX. Each subject was activity at 0.25, 1, and 4 h postdose, respectively. MNTX metabolites
FIG. 3. Hydrolysis of 1-h mouse plasma sample by ␤-glucuronidase.

610
CHANDRASEKARAN ET AL.
in pooled mouse plasma included methyl-6␤-naltrexol (M5), 2-hy- radioactivity, respectively. Small amounts (each Յ6% of plasma
droxy-3-O-methyl MNTX (M6), MNTX-3-glucuronide (M9) and radioactivity) of M5 and M6 were also observed in plasma. Trace
2-hydroxy-3-O-methyl MNTX glucuronide (M10). These metabolites amounts of other metabolites observed in plasma were not character-
were characterized by LC/MS as described below. M9 and M10 were ized due to low concentrations.
the most abundant plasma metabolites, representing 30.2 to 43.5 and
After a single 5 mg/kg i.v. administration of [3H]MNTX to dogs,
6.6 to 24.6% of the total radioactivity, respectively, between 0.25 and MNTX was the major radioactive component in pooled plasma,
4 h postdose. After hydrolysis with ␤-glucuronidase, M9 and M10 representing 98.6, 77.2, and 66.4% of total radioactivity at 0.25, 1, and
were almost completely converted to MNTX and M6, respectively 4 h postdose, respectively. M9 was the only major metabolite in
(Fig. 3). Other metabolites observed in plasma were not characterized plasma, increasing over time, and represented 0 to 18.5% of total
due to low concentrations.
radioactivity in the 0.25 to 4-h period. Several other radioactive peaks
MNTX was metabolized to a lesser extent in rats compared with (each Յ9% of plasma radioactivity) observed in dog plasma were not
mice after a single 5 mg/kg i.v. administration of [14C]MNTX. The characterized due to low concentrations.
parent drug represented 93.2, 69.3, and 49.8% of total radioactivity at
Metabolite Profiles in Urine. An average of 49.8% of the admin-
0.25, 1, and 4 h postdose, respectively. M2 and M9 were the most istered dose was recovered in 0 to 24-h human urine. The metabolite
abundant metabolites in pooled plasma, and the percentage of radio- profile (Fig. 4) of the pooled 0 to 24-h urine was similar to the plasma
activity attributed to these metabolites increased over time between profiles in humans. MNTX represented approximately 82% of urinary
0.25 and 4 h, representing 1.1 to 18.3 and 2.5 to 16.9% of total radioactivity, whereas M2, M4, and M5 represented 4.3, 9.4 and 2.6%
FIG. 4. Metabolite profiles of MNTX in pooled 0 to 24-h urine of mice, rats, dogs, and humans.

IN VIVO METABOLISM OF METHYLNALTREXONE
611
of urinary radioactivity, respectively. Radioactivity concentrations Urinary metabolites included M5 (2.3% of total radioactivity), M6
were too low to determine accurate metabolic profiles in samples after (3.9%), M9 (6.5%), and M10 (2.7%). M9 and M10 isolated from
24 h.
mouse urine were also completely hydrolyzed by ␤-glucuronidase to
In mice, a mean of 56.4% of the administered intravenous dose was MNTX and M6, respectively (data not shown).
excreted in 0 to 24-h urine. The parent drug represented 83.4% of total
A mean of 56.8% of the administered dose was excreted in 0 to
radioactivity in the pooled 0 to 24-h urine sample, and the urinary 24-h rat urine. The parent drug represented 92.4% of total radioac-
metabolite profile (Fig. 4) was qualitatively similar to plasma profiles. tivity in the pooled 0 to 24-h urine sample, and the urinary metabolite
TABLE 1
Retention times and mass spectral properties for MNTX and its metabolites observed in mice, rats, dogs, and humans
Molecular Ion (Mϩ
)
Metabolite
tR
Relevant Product Ions
With H2O
With D2O
min
m/z
MNTX
M2
M4
M5
M6
23.0
14.0
20.5
25.5
24.5
12.0
14.1
356
436
358
358
386
532
562
358
438
361
361
388
537
567
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
368, 332, 314, 257, 229, 112, 55
356, 302, 284, 227, 199, 113, 55
386, 332, 314, 257, 229, 113, 55
M9
M10
FIG. 5. Product ion mass spectra of MNTX (a), M2 (b), and M4 (c).

612
CHANDRASEKARAN ET AL.
profile (Fig. 4) was also qualitatively similar to plasma profiles. from [M]ϩ generated m/z 338 and indicated the presence of an
Urinary metabolites included M2, M9, M5, and M6 (each Յ3% of aliphatic hydroxyl group. Loss of the three atoms of the bridging unit
urine radioactivity).
(i.e., carbon-carbon-nitrogen) produced a methyl-ethyl-cyclopropyl-
In dogs, approximately 66.0% of the administered dose was ex- methylamine ion at m/z 112. Cleavage of the cyclopropylmethylene
creted in 0 to 24-h urine. The metabolite profile (Fig. 4) of the pooled carbon-nitrogen bond with charge retention on the carbon or nitrogen
0 to 24-h urine was similar to the plasma profiles, and M9 was the gave m/z 55, representing the cyclopropylmethylene group, and m/z
only metabolite in urine, representing 3.2% of total radioactivity.
302, respectively. Subsequent loss of H2O from m/z 302 yielded m/z
Metabolite Synthesis and Characterization. MNTX. MNTX 284, which further expelled methylvinylamine (CH2 ϭ CHNHCH3) to
eluted at approximately 23 min. Mass spectral data for MNTX were give m/z 227. Loss of a molecule of CO from m/z 227 generated m/z
obtained for comparison with its metabolites, and these data are 199.
summarized in Table 1. For MNTX, a molecular ion, [M]ϩ, was
MNTX-3-Sulfate (M2). Metabolite M2 eluted at approximately 14
observed at m/z 356. LC/MS with deuterated solvents in the mobile min. The [M]ϩ for M2 was observed at m/z 436, which was 80 Da
phase generated an [M]ϩ at m/z 358 (data not shown). This finding larger than MNTX. LC/MS conducted with deuterated solvents in the
was consistent with the two exchangeable protons of MNTX present mobile phase produced [M]ϩ at m/z 438 (data not shown). This
on each of the two hydroxyl groups. The MS/MS spectrum obtained finding was consistent with two exchangeable hydrogens for M2, the
from collision-activated dissociation of m/z 356 from MNTX with the same as for MNTX. The proposed fragmentation scheme and product
proposed fragmentation scheme is shown in Fig. 5A. Loss of H2O ions of m/z 436 mass spectrum for M2 are shown in Fig. 5B. Neutral
TABLE 2
Chemical shifts of MNTX and its synthetic MNTX metabolite standards
MNTX
M2
M4
M5
Proton No.
1H
13C
1H
13C
1H
13C
1H
13C
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
18
19
20
21
22
6.76
6.77
120.23
118.92
140.8
143.6
89.09
207.72
24.57
32.45
72.44
27.53
119.40
127.60
48.9
72.20
34.25
57.47
72.39
3.50
2.165
5.51
52.91
6.76
7.31
119.31
123.62
6.69
6.78
119.68
118.92
139.77
146.65
89.32
65.52
23.07
30.30
72.06
28.71
121.48
130.00
65.88
73.08
27.10
57.20
73.23
3.93
6.75
6.77
119.5
118.2
4.871
4.80
88.72
4.72
4.30
1.26/1.85
1.71/1.71
3.94
4.46
3.57
1.69/2.75
1.57/1.79
3.91
95.11
62.80
24.73
31.2
73.40
28.05
119.4
130.0
45.8
71.5
24.53
57.6
72.56
3.59
2.40
1.79/2.97
1.79/2.11
4.023
1.69/2.81
1.62/1.95
3.89
24.2
32.1
71.39
27.4
3.64/3.67
3.09/3.55
3.21/3.63
3.16/3.61
2.22/3.05
3.04/3.41
2.92/3.96
1.23
0.49/0.87
0.687/0.96
3.79
2.07/2.80
2.85/3.24
2.74/3.88
1.14
0.311
0.531
33.87
57.02
72.01
3.05
0.69
0.80
1.74/2.79
3.15/3.32
2.86/3.99
1.25
0.49/0.85
0.66/0.94
3.70
1.69/2.75
2.97/3.32
2.81/3.96
1.23/1.28
0.45/0.83
0.66/0.94
3.72
2.70
5.68
52.84
6.24
53.58
3.62
52.67
FIG. 6. Formation of methyl-6-naltrexol epimers by reduction of MNTX with sodium borohydride.

IN VIVO METABOLISM OF METHYLNALTREXONE
613
loss of 80 Da from [M]ϩ produced m/z 356, the [M]ϩ for MNTX, [M]ϩ for MNTX. LC/MS conducted with deuterated solvents in the
indicating that M2 was an MNTX-3-sulfate. Product ions at m/z 338, mobile phase produced [M]ϩ at m/z 361 (data not shown). This
302, 284, and 55 were also observed for MNTX, which indicated an finding was consistent with three exchangeable hydrogens for M4 and
otherwise unchanged MNTX molecule. The product ion at m/z 382 M5, one more than for MNTX, and was consistent with the introduc-
was 80 Da larger than the corresponding ion at m/z 302 for MNTX, tion of an additional hydroxyl group. The proposed fragmentation
which was consistent with sulfation of either hydroxyl group of scheme and product ions of m/z 358 mass spectrum for M4 are shown
MNTX. Based on observation of neutral loss of H2O from [M]ϩ to in Fig. 5C. The fragmentation of M5 (data not shown) was identical
generate m/z 418, consistent with a free aliphatic hydroxyl group, it to M4. Product ions at m/z 112 and 55 were also present for MNTX
was proposed that M2 was an aromatic sulfate. Assignment of the and indicated unchanged methyl-ethyl amine and cyclopropylmethyl-
1
protons in the H NMR spectrum of synthetic MNTX-3-sulfate are ene groups. The product ions at m/z 304, 286, and 229 were 2 Da
summarized in Table 2. The key protons H-1 and H-2, with chemical larger than the corresponding ions for MNTX, and together with the
shifts far apart in comparison with MNTX, where they are overlapped, absence of a loss of CO from m/z 229 they indicated that the carbonyl
and a J coupling of 8.36 Hz suggest sulfation at the 3-hydroxyl group group had been reduced. In a chemical reaction, MNTX completely
instead of the 14-hydroxyl group (Table 2). The HPLC retention time disappeared after reduction by sodium borohydride, and two products
and LC/MS spectra of the synthetic MNTX-3-sulfate matched that for were formed in a ratio of approximately 90:10 (Fig. 6). 1H NMR
M2 observed in human plasma after intravenous administration. analysis indicated the reduction of the carbonyl group (Table 2).
Therefore, M2 was identified as MNTX-3-sulfate.
These two synthetic epimers were differentiated by 2D NOESY
M4 and M5. Carbonyl reduction metabolites M4 and M5 eluted at spectra, which demonstrated key cross-peaks between protons H-5
approximately 20.5 and 25.5 min, respectively. The [M]ϩ for both M4 and H-6 with a coupling constant of 4.9 Hz for the ␣-epimer (Fig. 7),
and M5 were observed at m/z 358, which was 2 Da larger than the but much weaker cross-peaks between protons H-5 and H-6 with a
FIG. 7. 2D NOESY NMR spectra of M4 with key cross-peaks between protons H-5 and H-6 (circled).

614
CHANDRASEKARAN ET AL.
coupling constant of 6.5 Hz for the ␤-epimer (data not shown). HPLC lation and methylation of a hydroxyl group had occurred on the
retention time and mass spectral data for synthetic methyl-6␣-nal- phenyl moiety. Therefore, M6 was tentatively identified as 2-hy-
trexol and methyl-6␤-naltrexol matched that for M4 and M5, respec- droxy-3-O-methyl MNTX.
tively. Therefore, M4 and M5 were identified as the ␣- and ␤-epimer
of reduced MNTX, respectively.
Metabolite M9. Metabolite M9 eluted at approximately 12 min. The
[M]ϩ for M9 was observed at m/z 532, 176 Da larger than the [M]ϩ
Metabolite M6. Metabolite M6 eluted at approximately 24.5 min. for MNTX, indicating that M9 was a conjugate. LC/MS conducted
The [M]ϩ for M6 was observed at m/z 386, 30 Da larger than the with deuterated solvents in the mobile phase produced [M]ϩ at m/z
[M]ϩ for MNTX. LC/MS conducted with deuterated solvents in the 537 (data not shown), indicating five exchangeable hydrogens for M9,
mobile phase produced [M]ϩ at m/z 388 (data not shown), indicating consistent with a glucuronide. The proposed fragmentation scheme
two exchangeable hydrogens for M6, identical to MNTX. The pro- and product ions of m/z 532 mass spectrum for M9 are shown in Fig.
posed fragmentation scheme and product ions of m/z 386 mass spec- 8B. Neutral loss of 176 Da from [M]ϩ produced m/z 356, identical to
trum for M6 are shown in Fig. 8A. Product ions at m/z 112 and 55 the [M]ϩ for MNTX, indicating that M9 was a glucuronide. The
were also present for MNTX and indicated unchanged methyl-ethyl product ions at m/z 302, 284, and 55 were also observed for MNTX
amine and cyclopropylmethylene groups. Loss of H2O from [M]ϩ and indicated an otherwise unchanged MNTX molecule. Therefore,
produced m/z 368, which was consistent with an aliphatic hydroxyl M9 was identified as MNTX-3-glucuronide.
group. The product ions at m/z 332, 314, 257, and 229 were 30 Da
Metabolite M10. Metabolite M10 eluted at approximately 14.1 min.
larger than the corresponding ions for MNTX, and together with the The [M]ϩ for M10 was observed at m/z 562, 190 Da larger than the
absence of loss of methanol from [M]ϩ they indicated that hydroxy- [M]ϩ for MNTX and also 176 Da larger than the [M]ϩ for M6,
FIG. 8. Product ion mass spectra of M6 (a), M9 (b), and M10 (c).

IN VIVO METABOLISM OF METHYLNALTREXONE
615
described above. LC/MS conducted with deuterated solvents in the bolic pathways of MNTX in mice, rats, dogs, and humans are shown
mobile phase produced [M]ϩ at m/z 567 (data not shown), indicating
in Fig. 9. The pathways are based on the metabolite profiles observed
five exchangeable hydrogens for M10. The proposed fragmentation
scheme and product ions of m/z 562 mass spectrum for M10 are
shown in Fig. 8C. Neutral loss of 176 Da from [M]ϩ produced m/z
386, identical to the [M]ϩ for M6, indicating that M10 was a gluc-
uronide conjugate. The product ion at m/z 55 was also observed for
MNTX, indicating an unchanged cyclopropylmethylene group. The
product ions at m/z 332 and 314 were 30 Da larger than the corre-
sponding ions for MNTX, consistent with methylation and hydroxy-
lation as was observed for M6. Enzymatic hydrolysis with ␤-glucu-
ronidase converted M9 and M10 to MNTX and M6, respectively,
indicating the M10 was a glucuronide of M6 (data not shown).
Therefore, M10 was tentatively identified as 2-hydroxy-3-O-methyl
MNTX glucuronide.
in plasma and urine. Metabolite profiles of MNTX appeared to be
qualitatively similar in plasma and urine for each of the species
examined. The primary pathways of metabolism in humans include
sulfation at the phenolic group and carbonyl reduction. The mean
plasma AUC(0–12) ratio of unchanged MNTX, as measured by a
validated bioanalytical method, to total radioactivity was 0.59. This
observation along with the urinary and fecal metabolite profiles has
suggested the pharmacokinetic differences between the parent com-
pound and metabolites. The overall recovery of the radioactivity in
urine and feces accounted for 70.9% of the administered radioactive
dose. Urinary recovery in the 0 to 24-h period was 43.9%. In this
collection interval, MNTX accounted for 37.1% of the administered
dose; M2, M4, and M5 represented 1.26, 4.25, and 0.74%, respec-
tively. Approximately 17.3% of the dose was recovered in the feces
(0–168 h). In this collection interval, MNTX accounted for 11.4% of
the administered dose in the feces; M4 and M5 represented 0.83 and
0.24% of the dose, respectively. M2 was not observed in feces.
N-Demethylation leading to the formation of naltrexone and its
Discussion
In this study, comparative metabolism of MNTX was established in
humans and the nonclinical safety models used for the development of
this drug, which included mice, rats, and dogs. The proposed meta-
FIG. 9. Metabolic pathways of MNTX in mice, rats, dogs, and humans.

616
CHANDRASEKARAN ET AL.
reduction product 6␤-naltrexol did not appear to be a significant M6 was observed in the circulation of mice and rats, but it was not
pathway of metabolism for MNTX in humans, because these metab- detected in the circulation of dogs and humans. M4 and M5, the
olites were not observed in plasma or urine. Trace amounts of radio- reduced MNTX isomers, were both observed in human plasma. How-
activity excreted in exhaled CO2 (Ͻ0.06% of a total dose) provided a ever, only M5 was observed in the plasma of mice and rats, and
strong indication that N-demethylation represents a negligible meta- neither isomer was present in dog plasma. N-Demethylation resulting
bolic pathway of MNTX in humans. These results confirmed an in naltrexone formation was not observed in humans. These data
earlier observation that N-demethylation was not a major metabolic suggested species differences in the metabolism of MNTX.
pathway of MNTX in cancer patients (Kotake et al., 1989). Thus,
Acknowledgments. We thank Daniel Evcic and Andre Negahban
potential impairment of analgesic effects of opioids through N-de-
for technical assistance in these studies.
methylation of MNTX in humans can be ruled out. In male rats, the
metabolic pathways of MNTX were sulfation, glucuronidation, hy-
droxylation, and methylation. MNTX was more extensively metabo-
lized in mice than in rats, dogs, and humans via glucuronidation,
reduction, hydroxylation, and methylation. MNTX was metabolized
in dogs primarily via glucuronidation of the phenol.
References
Brown DR and Goldberg LI (1985) The use of quaternary narcotic antagonists in opiate research.
Neuropharmacology 24:181–191.
Chen TY and Rosow CE (2007) Methylnaltrexone bromide. Drugs Future 32:771–775.
Cone E, Gorodetzky C, Darwin W, Carroll F, Brine G, and Welch C (1978) The identification
and measurement of two new metabolites of naltrexone in human urine. Res Commun Chem
Pathol Pharmacol 20:413–433.
There appeared to be some major differences between MNTX and
naltrexone in terms of metabolism in humans and laboratory animals. In
humans, in addition to the presence of 6␤-naltrexol in conjugated and
nonconjugated forms and conjugated naltrexone as the major metabolites,
2-hydroxy-3-O-methylnaltrexone, 3-O-methyl-6␤-naltrexol, and 2-hy-
droxy-3-O-methyl-6␤-naltrexol were reported as minor or trace
metabolites of naltrexone (Cone et al., 1978; Wall et al., 1981).
These metabolites were also detected in animal species along with 6␣-
naltrexol, 3-O-methylnaltrexone, and some dealkylation products (Misra
et al., 1976; Rodgers et al., 1980; Misra, 1981), which were not observed
in humans. A study in dogs showed that naltrexone was predominantly
excreted as the 3-glucuronide. MNTX was metabolized via glucuronida-
tion, hydroxylation, methylation, sulfation, and carbonyl reduction in
nonclinical species. Although both drugs appeared to be primarily me-
tabolized via conjugation of the phenolic group and reduction of the
ketone in humans, both 6␣- and 6␤-epimers of the reduced MNTX
isomers were observed for MNTX only in human plasma. In addition, the
extent of metabolism was much higher for naltrexone than for MNTX in
humans. For example, after a subcutaneous administration of naltrexone,
6␤-naltrexol represented approximately 40% of the administrated dose,
whereas the parent drug represented only approximately 10% of the
administrated dose (Wall et al., 1984), based on AUC values. In contrast
to naltrexone, the carbonyl reduction metabolites M4 and M5 of MNTX
represented less than 10% of total plasma radioactivity, based on AUC
values after a single intravenous administration of MNTX at 0.3 mg/kg.
In summary, MNTX was not extensively metabolized in humans
after a single intravenous administration. Conversion to methyl-6-
naltrexol isomers (M4 and M5) and M2 were the primary pathways of
metabolism. MNTX was more extensively metabolized in mice com-
pared with rats, dogs, and humans through glucuronidation, reduction,
hydroxylation, and methylation. M9 was a major metabolite in plasma
of mice, rats, and dogs, but was not observed in human plasma. M10
was a major metabolite in mouse plasma, but it was absent in circu-
lation of other species. M2 was the major metabolite in rat and human
plasma. However, this metabolite was not observed in other species.
Jones D, Jukes-Jones R, Verschoyle R, Farmer P, and Gescher A (2005) A synthetic approach
to the generation of quercetin sulfates and the detection of quercetin 3Ј-O-sulfate as a urinary
metabolite in the rat. Bioorg Med Chem 13:6727–6731.
Kotake AN, Kuwahara SK, Burton E, McCoy CE, and Goldberg LI (1989) Variations in
demethylation of N-methylnaltrexone in mice, rats, dogs, and humans. Xenobiotica 19:1247–
1254.
Kim C, Cheng R, Corrigall WA, and Coen KM (1989) Assay for methylnaltrexone in rat brain
regions and serum by high-performance liquid chromatography with coulometric electrochem-
ical detection. Chromatographia 28:359–363.
Malspeis L, Bathala M, Ludden T, Bhat H, Frank S, Sokoloski T, Morrison B, and Reuning R
(1975) Metabolic reduction of naltrexone I. Synthesis, separation and characterization of
naloxone and naltrexone reduction products and qualitative assay of urine and bile after
administration of naltrexone, alpha-naltrexol, or beta-naltrexol. Res Commun Chem Pathol
Pharmacol 12:43–65.
Misra A (1981) Current status of preclinical research on disposition, pharmacokinetics, and
metabolism of naltrexone. NIDA Res Monogr 28:132–146.
Misra AL, Pontani RB, and Vadlamani NL (1987) Intravenous kinetics and metabolism of
[15,16-3H]naltrexonium methiodide in the rat. J Pharm Pharmacol. 39:225–227.
Misra AL, Bloch R, Vardy J, Mule´ SJ, and Verebely
K (1976) Disposition of [15,16-
3H]naltrexone in the central nervous system of the rat. Drug Metab Dispos 4:276–280.
Murphy DB, El Behiery H, Chan VW, and Foss JF (2001) Pharmacikinetic profile of epidurally
administered methylnaltrexone, a novel peripheral opioid anatagonist in a rabbit model. Br J
Anaesth 86:120–122.
Rodgers RM, Taylor SM, Dickinson RG, Ilias AM, Lynn RK, and Gerber N (1980) Metabolism
and choleretic effect of naltrexone in the isolated perfused rat liver. Drug Metab Dispos
8:390–396.
Thomas J, Karver S, Cooney GA, Chamberlain BH, Watt CK, Slatkin NE, Stambler N, Kremer
AB, and Israel RJ (2008) Methylnaltrexone for opioid-induced constipation in advanced
illness. N Engl J Med 358:2332–2343.
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 C, Foss J, O’Connor M, Osinski J, Karrison T, Moss J, and Roizen M (2000) Methyl-
naltrexone for reversal of constipation due to chronic methadone use. JAMA 283:367–372.
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, 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.
Address correspondence to: Dr. Appavu Chandrasekaran, Pharmacokinet-
ics, Dynamics and Metabolism, Pfizer, Inc, 500 Arcola Road, Collegeville, PA
19426. E-mail: Chandra@wyeth.com