1521-009X/44/8/1262–1269$25.00 http://dx.doi.org/10.1124/dmd.116.070185DRUG METABOLISM AND DISPOSITION Drug Metab Dispos 44:1262–1269, August 2016Copyright ª 2016 by The American Society for Pharmacology and Experimental Therapeutics

Special Section on Emerging Novel Enzyme Pathways inDrug Metabolism

Species Differences in the Oxidative Desulfurization of aThiouracil-Based Irreversible Myeloperoxidase Inactivator by

Flavin-Containing Monooxygenase Enzymes s

Heather Eng, Raman Sharma, Angela Wolford, Li Di, Roger B. Ruggeri, Leonard Buckbinder,Edward L. Conn, Deepak K. Dalvie, and Amit S. Kalgutkar

Pharmacokinetics, Pharmacodynamics, and Metabolism Department, Pfizer Inc., Groton, Connecticut (H.E., R.S., A.W., L.D.);Pharmacokinetics, Pharmacodynamics, and Metabolism Department, Pfizer Inc., La Jolla, California (D.K.D.); Pharmacokinetics,Pharmacodynamics, and Metabolism Department (A.S.K.), Worldwide Medicinal Chemistry (E.L.C., R.B.R.), and Cardiovascular and

Metabolic Research Unit (L.B.), Pfizer Inc., Cambridge, Massachusetts

Received February 22, 2016; accepted April 13, 2016

ABSTRACT

N1-Substituted-6-arylthiouracils, represented by compound 1 [6-(2,4-dimethoxyphenyl)-1-(2-hydroxyethyl)-2-thioxo-2,3-dihydropyrimidin-4(1H)-one], are a novel class of selective irreversible inhibitors ofhumanmyeloperoxidase. The present account is a summary of our invitro studies on the facile oxidative desulfurization in compound 1 to acyclic ether metabolite M1 [5-(2,4-dimethoxyphenyl)-2,3-dihydro-7H-oxazolo[3,2-a]pyrimidin-7-one] in NADPH-supplemented rats (t1/2[half-life = mean6 S.D.] = 8.66 0.4 minutes) and dog liver microsomes(t1/2 = 11.2 6 0.4 minutes), but not in human liver microsomes (t1/2 >120 minutes). The in vitro metabolic instability also manifested inmoderate-to-high plasma clearances of the parent compound in ratsanddogswith significant concentrations ofM1detected in circulation.Mild heat deactivation of liver microsomes or coincubation with theflavin-containing monooxygenase (FMO) inhibitor imipramine signifi-cantly diminished M1 formation. In contrast, oxidative metabolism ofcompound 1 to M1 was not inhibited by the pan cytochrome P450

inactivator 1-aminobenzotriazole. Incubations with recombinantFMO isoforms (FMO1, FMO3, and FMO5) revealed that FMO1 princi-pally catalyzed the conversion of compound 1 to M1. FMO1 is notexpressed in adult human liver, which rationalizes the speciesdifference in oxidative desulfurization. Oxidation by FMO1 followedMichaelis-Menten kinetics with Michaelis-Menten constant, maxi-mum rate of oxidative desulfurization, and intrinsic clearance valuesof 209 mM, 20.4 nmol/min/mg protein, and 82.7 ml/min/mg protein,respectively. Addition of excess glutathione essentially eliminated theconversion of compound 1 to M1 in NADPH-supplemented rat anddog liver microsomes, which suggests that the initial FMO1-mediatedS-oxygenation of compound 1 yields a sulfenic acid intermediatecapable of redox cycling to the parent compound in a glutathione-dependent fashion or undergoing further oxidation to a more electro-philic sulfinic acid species that is trapped intramolecularly by thependant alcohol motif in compound 1.

Introduction

We recently reported structure-activity relationship studieson N1-substituted-6-aryl-2-thiouracil derivatives as irreversible,mechanism-based inactivators of the hemoprotein myeloperoxidase(MPO, EC 1.11.2.2) with a high degree of selectivity for MPO relative toperoxidases such as thyroid peroxidase (TPO) and cytochrome P450

(P450) enzymes (Ruggeri et al., 2015). The thiouracil analogs behave assuicide substrates of MPO and covalently adduct to the heme prostheticgroup through an oxidized sulfur species (presumably a thiol radical)generated during catalysis (Tidén et al., 2011; Ruggeri et al., 2015). Theantithyroid drug propylthiouracil (PTU, Fig. 1), which irreversibly inhibitsMPO and TPO in a nonselective fashion (Lee et al., 1990; Ruggeri et al.,2015), was used as a starting point in our structure-activity relationshipwork to identify selective MPO inhibitors. Introduction of polar N1substituents and replacement of theC6 propyl group in PTUwith electron-rich aromatic functionalities resulted in significant improvements in MPO

dx.doi.org/10.1124/dmd.116.070185.s This article has supplemental material available at dmd.aspetjournals.org.

ABBREVIATIONS: ABT-107, 5-[6-[[(3R)-1-azabicyclo[2.2.2]octan-3-yl]oxy]pyridazin-3-yl]-1H-indole; AUC0–‘, area under the plasma concentration–time curve from zero to infinity; CID, collision-induced dissociation; CLint, intrinsic clearance; CLp, plasma clearance; compound 1, 6-(2,4-dimethoxyphenyl)-1-(2-hydroxyethyl)-2-thioxo-2,3-dihydropyrimidin-4(1H)-one; DMSO, dimethylsulfoxide; FMO, flavin-containing monooxygenase;GSH, glutathione; HPLC, high-performance liquid chromatography; kel, terminal rate constant; KM, Michaelis-Menten constant; LC-MS/MS, liquidchromatography tandem mass spectrometry; M1, 5-(2,4-dimethoxyphenyl)-2,3-dihydro-7H-oxazolo[3,2-a]pyrimidin-7-one; MPO, myeloperoxidase;P450, cytochrome P450; PTU, propylthiouracil; t1/2, half-life; tR, retention time; TPO, thyroid peroxidase; Vdss, steady-state distribution volume;Vmax, maximum rate of oxidative desulfurization.

1262

http://dmd.aspetjournals.org/content/suppl/2016/04/14/dmd.116.070185.DC1Supplemental material to this article can be found at:

at ASPE

T Journals on M

arch 14, 2020dm

d.aspetjournals.orgD

ownloaded from

inhibitory activity (inferred from the inactivation kinetics parameters [themaximal rate constant for inactivation and the inhibition constant] andpartition ratio) and virtually abolished TPO inhibition with the resultantcompounds.Concern over the risk of immune-mediated toxicity (e.g., agranulo-

cytosis and hepatotoxicity) associated with chronic PTU treatment(Futcher and Massie, 1950; Ichiki et al., 1998; Cooper, 2005) viaoxidative bioactivation of the thiouracil motif to protein- and thiol-reactive intermediates (Lee et al., 1988, 1990; Waldhauser and Uetrecht,1991; Jiang et al., 1994) was principally mitigated by tethering thependant nucleophilic groups to the N1-substituent, which could poten-tially quench reactive species in an intramolecular fashion. Out of thisexercise emerged the lead compound 6-(2,4-dimethoxyphenyl)-1-(2-hydroxyethyl)-2-thioxo-2,3-dihydropyrimidin-4(1H)-one (compound 1, Fig.1), with significant improvements noted in MPO inactivation potencyand selectivity relative to PTU. Intramolecular trapping of reactivespecies was demonstrated by reacting compound 1 with excess hydrogenperoxide (H2O2) (Kitamura, 1934; Kalm, 1961), which quantitativelyconverted compound 1 to the pharmacologically inactive cyclic ether 5-(2,4-dimethoxyphenyl)-2,3-dihydro-7H-oxazolo[3,2-a]pyrimidin-7-one(M1, Fig. 1), presumably via an unstable oxidized sulfur intermediate(Ruggeri et al., 2015).Importantly, no thiol conjugates of compound 1 were generated upon

addition of reduced glutathione (GSH) to the H2O2 and MPO/H2O2

incubations. Moreover, compound 1 was resistant toward metabolicturnover in reduced NADPH-supplemented human liver microsomes(half-life [t1/2] . 120 minutes) and cryopreserved human hepatocytes(t1/2 . 240 minutes), which was generally consistent with its physico-chemical properties (molecular weight = 308, lipophilicity (logD7.4 =1.1), and topologic polar surface area = 71.03 Å2). In contrast, a highmetabolic turnover of compound 1 was noted in NADPH-supplementedrat and dog liver microsomes, which translated in moderate to highplasma clearance (CLp) in these preclinical species.In vitro mechanistic studies were initiated to rationalize the species

differences in metabolism and revealed a facile conversion of compound 1principally to the cyclic ether metabolite M1 by rat and dog liver flavin-containing monooxygenase (FMO) 1, which is not expressed in adulthuman liver. The collective findings from these studies are reported herein.

Materials and Methods

Materials. The synthesis of compound 1 (chemical purity . 99% by high-performance liquid chromatography [HPLC] and NMR) has been previously reportedelsewhere (Ruggeri et al., 2015). The preparation of the M1metabolite is described inour Supplemental Data. NADPH, 1-aminobenzotriazole, reduced GSH, and imipra-mine were purchased from Sigma-Aldrich (St. Louis, MO). Pooled male Wistar-Hanrat and male and female human liver microsomes (pool of 126 livers, aged 8–10weeks) were purchased from BD Gentest (Woburn, MA), male beagle dog livermicrosomes (pool of 14 livers, aged 1–4 years) and pooled male and female humankidney microsomes from XenoTech (Kansas City, KS). Recombinant human FMO1,FMO3, and FMO5 supersomes were purchased from Corning (Oneonta, NY).

Liver and Kidney Microsomal Stability. Stock solutions of compound 1were prepared in dimethylsulfoxide (DMSO) then diluted with methanol andacetonitrile. The final concentration of solvents in the incubation mixture were0.025% DMSO, 0.5% methanol, and 0.475% acetonitrile. Assessment of t1/2 inliver or kidney microsomes was determined in triplicate with microsomes (1 mg/mlmicrosomal protein for rat, dog, and human) in 0.1 M potassium phosphate buffer(pH 7.4) containing 3.3 mM MgCl2 at 37�C.

The reaction mixture was prewarmed with 1.3 mM NADPH at 37�C for5 minutes before initiating the reaction with the addition of compound 1 (1 mM).Aliquots of the reaction mixture at 0.25, 5.0, 10, 20, 30, 40, and 60 minuteswere added to acetonitrile containing 0.1% formic acid and the internalstandard terfenadine (2 ng/ml). The samples were centrifuged before dilution of

supernatant with an equal volume of water containing 0.1% formic acid and liquidchromatography/tandem mass spectrometry (LC-MS/MS) analysis of the disap-pearance of compound 1 and the formation of cyclized metabolite M1. For thecontrol experiments, NADPH was omitted from these incubations.

A parallel incubation of liver microsomes from rat and dog containing compound1 (1 mM), GSH (5 mM), and NADPH (1.3 mM) was conducted to evaluate redoxcycling of the initial S-oxygenation product of compound 1 formed during the processof enzymatic oxidation. To test the involvement of the thermally unstable FMO in theoxidation, we preincubated rat and dog liver microsomes at 50�C for 5 minutes in theabsence of NADPH cofactor and then cooled them on ice followed by incubationsdescribed previously. For the purposes of metabolite identification studies, theconcentration of compound 1 in the microsomal incubations was raised to 10 mM.The duration of the incubation time was 60 minutes.

Separate incubations of compound 1 (10 mM) were also conducted in hu-man liver microsomes containing NADPH (1.3 mM) and GSH (1 mM) for thepurposes of trapping potentially reactive species arising from the oxidativemetabolism of compound 1. To determine the effects of P450 or FMO inhibitionon the metabolic conversion of compound 1 to M1, 1-aminobenzotriazole(P450 inactivator; 1 mM final concentration) or imipramine (FMO competitiveinhibitor; 250 mM final concentration) were preincubated with rat and dogliver microsomes in the presence of NADPH for 20 or 2 minutes, respectively,before initiating the reaction with compound 1.

Incubations of Compound 1 with Recombinant FMO Isoforms. Potassiumphosphate buffer (0.1M, pH 7.4), magnesium chloride (3.3 mM), NADPH (1.3 mM),and recombinant FMO (0.5 mg/ml) were combined and prewarmed at 37�C for2 minutes. To initiate the reaction, we added compound 1 at a final concentration of1 mM (final 0.025% DMSO, 0.5% methanol, and 0.475% acetonitrile). At each timepoint (0.25, 5.0, 10, 20, 30, 40, and 60 minutes) a 50-ml aliquot of reaction mixturewas transferred to 200 ml of acetonitrile containing 0.1% formic acid and the in-ternal standard terfenadine (2 ng/ml). After centrifugation at 2000g, equal vol-umes of supernatant and water containing 0.1% formic acid were mixed, and thedisappearance of compound 1 and formation of M1 were examined by LC-MS/MS.

To evaluate the linearity of the product (i.e., M1) formation by human FMO1and FMO3 isoforms, recombinant FMO1 or FMO3 supersomes (0.1–1.0 mg/ml)andNADPH (3.3 mM) were preincubated in 0.1M phosphate buffer (pH 7.4) for 2minutes at 37�C. Reactions (5000ml) were initiated by the addition of compound 1(1 mM), and were allowed to continue at 37�C for 0.25 to 60 minutes. Aliquots(50ml) of the reactions were quenched with 200 ml of acetonitrile containing 0.1%formic acid and the internal standard terfenadine (2 ng/ml). After centrifugation at2000g, the supernatants were combined with an equal volume of water containing0.1% formic acid, and the formation of M1 was examined by LC-MS/MS.

For determination of the Michaelis-Menten constant (KM), the maximum rateof oxidative desulfurization (Vmax), and the intrinsic clearance (CLint, Vmax/KM)for compound 1, incubations were repeated at a single protein concentration andtime point determined to be in the linear range of metabolite formation (0.1 mg/mlFMO1, 1 mg/ml FMO3, 60 minutes), containing 12 concentrations of compound1 (1–300 mM). The kinetic parameters were obtained for FMO1 using theMichaelis-Menten nonlinear regression in GraphPad Prism (GraphPad Software,La Jolla, CA). As the FMO3 reaction was not saturated within the range ofsubstrate concentrations tested, the slope of formation rate of M1 versus substrateconcentration (CLint) was calculated using linear regression.

Animal Pharmacokinetics. Dog experiments were conducted in our Associ-ation for Assessment and Accreditation of Laboratory Animal Care International–accredited facilities and were reviewed and approved by the Pfizer InstitutionalAnimal Care and Use Committee. Rat studies were performed at BioDuro,Pharmaceutical Product Development Inc. (Shanghai, People’s Republic ofChina), with animal care and in vivo procedures conducted according toguidelines from the BioDuro Institutional Animal Care and Use Committee.

Jugular vein cannulated male Wistar-Han rats (;250 g), obtained from VitalRiver (Beijing, People’s Republic of China), and male Beagle dogs (;8–11 kg)were used for these studies. Rats were provided ad libitum access to water andfood. Dogs were fasted overnight and fed 4 hours after dosing. Compound 1 wasadministered intravenously (i.v.) in 5% DMSO/95% of 30% 2-hydroxypropyl-b-cyclodextrin or 25%2-hydroxypropyl-b-cyclodextrin/75% 100mMTris buffer(pH 8.0) via the tail vein of rats (n = 3) or saphenous vein of dogs (n = 3) at a doseof 1.0 mg/kg in a dosing volume of 1 (rat) or 0.5 (dog) ml/kg.

Serial blood samples were collected before dosing and 0.033 (rat only), 0.083,0.25, 0.5, 1, 2, 4, 7, and 24 hours after dosing. Blood samples from the

Oxidative Desulfurization by FMOs 1263

at ASPE

T Journals on M

arch 14, 2020dm

d.aspetjournals.orgD

ownloaded from

pharmacokinetic studies were centrifuged to generate plasma. All plasma sampleswere kept frozen until analysis. Urine samples (0–7.0 and 7.0–24 hours) were alsocollected after i.v. administration. Aliquots of plasma or urine (10–50 ml) weretransferred to 96-well blocks and methanol/acetonitrile (1:1, v/v, 200 ml)containing an internal standard was added to each well. The supernatant wasdiluted 20-fold with methanol/water (1:1, v/v) containing 0.1% formic acid. Thesamples were then analyzed by LC-MS/MS, and the concentrations of compound1 and M1 in plasma and urine were determined by interpolation from a standardcurve. The range of quantitation was 1–2000 ng/ml for rats (linear R2 0.991compound 1 and linear R2 0.996M1) and 1–5000 ng/ml for dogs (linear R2 0.995compound 1 and quadratic R2 0.978 M1).

Determination of Pharmacokinetic Parameters. The pharmacokineticparameters were determined using noncompartmental analysis (Watson v.7.4;Thermo Scientific, Waltham, MA). The area under the plasma concentration-timecurve from t = 0 to 24 hours (AUC0–24) and t = 0 to infinity (AUC0–‘) wasestimated using the linear trapezoidal rule, and CLp was calculated as theintravenous dose divided by AUC0–‘

i.v.. The terminal rate constant (kel) wascalculated by a linear regression of the log-linear concentration-time curve, andthe terminal elimination t1/2 was calculated as 0.693 divided by kel. The apparentsteady-state distribution volume (Vdss) was determined as the i.v. dose divided bythe product of AUC0–‘ and kel. The percentage of unchanged compound 1excreted in urine over 24 hours was calculated using the following equation:Amount (in mg) of compound 1 in urine over the 24-hour interval after the dose/Actual amount of the dose of compound 1 administered (mg) � 100%. The renalclearance was derived as the ratio of amount (in mg) of compound 1 in urine overthe 24-hour interval postdose/AUC0–24.

LC-MS/MS Analysis for Quantitation of Compound 1 and M1.Concentrations of analytes from in vitro and in vivo studies were determinedon a Sciex 5500 LC-MS/MS triple quadrupole mass spectrometer (Sciex,Framingham, MA). Analytes were chromatographically separated using Agilent1290 (Santa Clara, CA) or Shimadzu LC-20AD (Shimadzu Scientific Instru-ments, Columbia, MD) pumps. A CTC PAL autosampler was programmed toinject 1 or 10ml on a Phenomenex Kinetex C18 30� 3mmHPLC (Phenomenex,Torrance, CA) or Mac Mod Halo C18 50 � 2.1 mm UPLC column (Mac ModAnalytical, Chadds Ford, PA) using amobile phase consisting of water containing0.1% (v/v) formic acid (solvent A) and acetonitrile containing 0.1% formic(solvent B) at a flow rate of 0.5 ml/min.

Compounds 1 andM1were detected using electrospray ionization (positive ionmode) in the multiple reaction monitoring mode monitoring for mass-to-charge(m/z) transition 309.1 → 164.2 or 291.1 and 275.1 →189.1, respectively.Compounds 1 and M1 standards were fit by least-squares regression of theirareas to a weighted linear equation, from which the unknown concentrationswere calculated. The dynamic range of the assay was 1.0–2000 ng/ml. Assayperformance was monitored by the inclusion of quality control samples withacceptance criteria of 630% target values.

Bioanalytical Methodology for Metabolite Identification. Qualitativeassessment of the metabolism of compound 1 was conducted using a ThermoFinnegan Surveyor photodiode array plus detector, Thermo Acela pump, and aThermo Acela Autosampler (Thermo Scientific, West Palm Beach, FL). Themonitoring wavelength (l) was 280 nm. Chromatography was performed on aPhenomenex Hydro RP C18 (4.6 mm � 150 mm, 3.5 mm) column. The mobilephase was composed of 5 mM ammonium formate buffer with 0.1% formic acid(pH 3.0) (solvent A) and acetonitrile (solvent B) at a flow rate of 1 ml/min.The binary gradient was as follows: the solvent A to solvent B ratio was held at95:5 (v/v) for 3 minutes and then adjusted to 55:45 (v/v) from 0 to 35 minutes,

30:70 (v/v) from 35 to 45 minutes, and 5:95 (v/v) from 45 to 52minutes, where itwas held for 3 minutes and then returned to 95:5 (v/v) for 6 minutes before thenext analytic run.

Identification of the metabolites was performed on a Thermo Orbitrap massspectrometer operating in positive ion electrospray mode. The spray potential was4 V, and heated capillary was at 275�C. Xcalibur software version 2.0 (ThermoScientific) was used to control the HPLC-MS system. Product ion spectra wereacquired at a normalized collision energy of 65 eVwith an isolationwidth of 2 amu.Metabolites from liver microsomes were identified in the full-scan mode (fromm/z 100 to 850) by comparing t = 0 sampleswith t = 60minutes samples or throughcomparison with synthetic standard(s), and structural information was generatedfrom collision-induced dissociation (CID) spectra of protonated molecular ions.

Results

Microsomal Stability. To examine microsomal stability, compound1 was incubated in rat, dog, and human liver microsomes or humankidney microsomes in the presence and absence of NADPH cofactor;periodically, aliquots of the incubation mixture were examined for thedepletion of compound 1 (Supplemental Fig. 1). The t1/2 (mean6 S.D.)for depletion of compound 1 in NADPH-supplemented rat, dog, andhuman liver microsomes was 8.66 0.4, 11.26 0.4 minutes, and. 120minutes, respectively. No substrate depletion (t1/2 . 120 minutes) wasnoted in rat, dog, and human incubations that lacked NADPH cofactoror liver microsomes. Figure 2 depicts the extracted ion chromatogramsof incubation mixtures of compound 1 in NADPH-supplemented livermicrosomes from rats, dogs, and humans. The major metabolite M1 wasdetected in rat and dog liver microsomes in a NADPH-dependent

Fig. 2. HPLC-UV (l = 280 nm) chromatogram of an incubation mixture ofcompound 1 (10 mM) in NADPH-supplemented (A) rat, (B) dog, and (C) humanliver microsomes.

Fig. 1. Oxidative desulfurization of the N1-substituted-6-arylthiouracil compound 1 by MPO or H2O2.

1264 Eng et al.

at ASPE

T Journals on M

arch 14, 2020dm

d.aspetjournals.orgD

ownloaded from

fashion. The MS2/MS3 spectra of compound 1 [retention time (tR) =11.25 minutes, exact mass (M + H)+ = 309.0904] and M1 [tR = 9.61minutes, exact mass (M + H)+ = 275.1026] are shown in SupplementalFigs. 2 and 3, respectively.The theoretical exact masses for the proposed fragment ion structures

in the CID spectrum of compound 1 and M1 were consistent with theobserved accurate masses (,2 ppm difference). The tR and massspectrum of M1 were identical to those discerned with an authenticstandard, which was chemically synthesized via S-methylation ofcompound 1 to the corresponding thioether derivative 2 followed byperoxide-mediated oxidative desulfurization/intramolecular cyclizationpresumably via an electrophilic sulfoxide intermediate (see the Supple-mental Data for the detailed synthetic protocol). Metabolites M2 (tR =9.19 minutes) and M3 (tR = 8.75 minutes) were rat-specific metaboliteswith an identical exact mass [295.0747 (M + H)+] and CID spectra (seeSupplemental Fig. 4 for a representative CID spectra of M3), implyingthat these metabolites were isomeric phenols derived from P450-mediated O-demethylations in compound 1.Consistent with the metabolic stability results, M1–M3 were only

detected in trace quantities in human liver microsomal incubationsof compound 1 in the presence of NADPH. Compound 1 (1 mM)appeared to be stable (t1/2 . 120 minutes) toward metabolic turnover inNADPH-supplemented human kidneymicrosomes, withminimal amountof M1 (80 nM) formed during the course of the 60-minute incubation.Compound 1 was devoid of reactive metabolite formation in NADPH-

supplemented rat, dog, and human liver microsomes as inferred from thelack of GSH conjugates formed when GSH (5 mM) was included in themicrosomal incubations (data not shown). Incidentally, inclusion ofexcess GSH significantly attenuated oxidative desulfurization ofcompound 1 (to M1) in liver microsomal incubations from rats anddogs (Supplemental Fig. 5). In the presence of GSH, compound 1 wasvirtually resistant to metabolic turnover in NADPH-supplemented dogliver microsomes [t1/2 (2 GSH) = 11.2 6 0.4 minutes; t1/2 (+ GSH) .120 minutes]. In contrast, the impact of GSH on the overall metabolicstability of compound 1 in NADPH-supplemented rat liver microsomeswas less severe [t1/2 (2 GSH) = 8.6 6 0.4; t1/2 (+ GSH) 49.7 6 1.7minutes].Qualitative examination of metabolite formation in NADPH- and

GSH-supplemented rat liver microsomal incubations of compound 1(10 mM) revealed that the formation of the O-demethylated metabolitesM2 and M3 was not impacted in the presence of the thiol nucleophile,which was in contrast to the complete disappearance of M1 (Fig. 3).Identification of Enzymes Responsible for Oxidative Desulfur-

ation of Compound 1 to M1. Incubations of compound 1 (1 mM) in

NADPH-supplemented rat or dog liver microsomes, which had beensubjected to heat treatment (50�C) for 5 minutes in the absence ofNADPH cofactor, induced metabolic resistance in compound 1 andvirtually abrogated the formation of M1, as shown in a representativeplot of an incubation mixture of compound 1 in heat-inactivated rat livermicrosomes (Fig. 4). This implies a potential role for a FMO isoform(s)in oxidative desulfurization.Incubations of compound 1 (1 mM) in 0.5 mg/ml human recombinant

FMO1, FMO3, and FMO5 in the presence of NADPH revealed thatFMO1 was principally responsible for the formation of M1, with a minorcontribution from FMO3 (Fig. 5). The contribution of FMO5 toward M1formation was insignificant in this analysis; as such, the enzyme kineticsexperiments were not pursued. The reactions with FMO1 and FMO3werelinear as a function of incubation time (5–60 minutes) and proteinconcentration up to 1.0 mg/ml of microsomal protein (results not shown).The effects of substrate concentrations on oxidative desulfurization

of compound 1 by recombinant human FMO1 and FMO3 wereinvestigated, and the results are shown in Fig. 6. Oxidation of compound1 to M1 by FMO1 followed Michaelis-Menten kinetics (Fig. 6A), withKm, Vmax, and CLint6 standard error values of 2096 12mM, 20.46 0.6nmol/min/mg protein, and 97.7 ml/min/mg protein, respectively. In thecase of FMO3, the Michaelis-Menten plot (see Fig. 6B) showed thatconversion of compound 1 to M1 was linear up to the highest substrateconcentration of 300 mM, indicating that the apparent KM value was.300 mM. Therefore, the KM and Vmax values for FMO3 could not bedetermined. The corresponding CLint value calculated from the slope ofthe Michaelis-Menten plot was 0.54 ml/min/mg protein.Consistent with these observations, conversion of compound 1 to M1

in NADPH-supplemented rat and dog liver microsomes was stronglyinhibited upon coincubation with imipramine (250 mM), a selectiveinhibitor of FMO1 (Dixit and Roche, 1984; Lee et al., 2009; Yamazakiet al., 2014), as reflected from changes in t1/2 from 8.6 to ;90 minutes(rats) and 11 to 81 minutes (dogs). In contrast, the effect of thenonselective P450 inactivator 1-aminobenzotriazole (1 mM) (Caldwellet al., 2005; Strelevitz et al., 2006; Boily et al., 2015; Parrish et al., 2015)on oxidative desulfurization was less severe (rat: t1/2 from 8.6 to 25minutes; dog: t1/2 from 11.0 to 12 minutes (Fig. 7).Intravenous Pharmacokinetics of Compound 1 after Single

Doses to Rats and Dogs. The pharmacokinetic parameters describingthe disposition of compound 1 andM1 after administration of compound1 to Wistar-Han rats and Beagle dogs are shown in Table 1 andSupplemental Fig. 6. Compound 1 demonstrated moderate to high CLp(rat CLp = 73 6 13 ml/min/kg; dog CLp = 12 6 3 ml/min/kg), and amoderate Vdss (rat Vdss = 1.4 6 0.5 l/kg; dog Vdss = 1.3 6 0.3 l/kg)resulting in terminal elimination t1/2 values of 0.4 6 0.2 and 5.3 6 0.8hours, respectively, in rats and dogs. M1 was also detected in thecirculation after i.v. administration of compound 1 to rats and dogs.The corresponding AUC0–‘ values of compound 1 and M1 in rats were

2336 37 and 49.86 11.2 ng.h/ml, respectively, whereas the correspond-ing AUC0–‘ values of compound 1 and M1 in dogs were 14506 310 and7696 136 ng.h/ml, respectively. M1 had a slightly longer elimination t1/2(relative to compound 1) in rats and dogs. Renal excretion of unchangedcompound 1 (,1% in rats and ;1.1% in dogs) and unchanged M1(;4.7% in rats and ;14.2% in dogs) was relatively low.

Discussion

Concerns over the liberation of indiscriminate electrophilic speciesduring MPO-mediated oxidation of N1-substituted-6-aryl-2-thiouracilswere minimized by ensuring a high partition ratio for MPO inactivationand by tethering nucleophilic functional groups in proximity of thethiouracil sulfur. Our medicinal chemistry strategy was also weighted

Fig. 3. HPLC-UV chromatogram of an incubation mixture of compound 1 (10 mM)in NADPH-supplemented rat liver microsomes in the absence or presence of GSH.

Oxidative Desulfurization by FMOs 1265

at ASPE

T Journals on M

arch 14, 2020dm

d.aspetjournals.orgD

ownloaded from

toward the design of compounds in the lower range of lipophilicity(logD , 1.5, topologic polar surface area , 100 Å2) to minimize thepotential for oxidative metabolism/bioactivation of the thiouracil motifin human liver.Apart from peroxidases, enzymatic bioactivation of thioureas and

related analogs (e.g., thiones, thiocarbamides, etc.) to electrophilic inter-mediates by mammalian P450 and/or FMO isoforms can also leadto toxicity (Poulsen et al., 1979; Neal and Halpert, 1982; Decker andDoerge, 1992; Onderwater et al., 1999, 2004; Smith and Crespi, 2002;Henderson et al., 2004; Ji et al., 2007). For instance, the cases of clinicalhepatotoxicity and/or nephrotoxicity that were noted with the antithy-roid drug methimazole (Martinez-Lopez et al., 1962) and the antipar-asitic agent thiabendazole (Manivel et al., 1987) have been causallylinked with their metabolism to the proximal toxicantsN-methylthioureaand thioformamide, respectively, via an initial P450-catalyzed oxidativering scission of the 2-mercaptobenzimidazole and thiazole motifs presentin these drugs. S-oxidation of the N-methylthiourea and thioformamidemetabolites to reactive metabolites by FMO enzymes is believed torepresent the key step resulting in toxicity (Mizutani et al., 1993, 2000).Consistent with our design philosophy, the N1-substituted-6-aryl-2-

thiouracil class of MPO inhibitors (represented in our present study bycompound 1) were stable toward metabolism in NADPH-supplementedhuman liver microsomes and/or cryopreserved human hepatocytes(Ruggeri et al., 2015), and they were latent to the formation of reactivespecies, as judged from the absence of GSH conjugates in humanrecombinant MPO and human liver microsomes supplemented with anexcess of the thiol nucleophile. Compound 1 was also devoid ofreversible and time-dependent inhibitory effects against major humanP450 enzymes (Pfizer data on file), whichmade it an attractive candidatefor advancement in preclinical toxicity studies.In contrast with the metabolic resistance in human hepatic tissue,

compound 1was converted to cyclic etherM1 inNADPH-supplemented

rat and dog liver microsomes. Heat inactivation, which abolishes FMOactivity while preserving P450 activity (Ziegler, 1980), provided cir-cumstantial evidence for the involvement of an FMO isoform(s) in theformation of M1. Consistent with this initial finding, the nonselectiveP450 inactivator 1-aminobenzotriazole had little effect on the conversionof compound 1 to M1 in rat and dog liver microsomes. In contrast, theFMO1 competitive inhibitor imipramine dramatically reduced theoxidative desulfurization in rat and dog liver microsomes, respectively,implying that the conversion of compound 1 to M1 in rat and dog livermicrosomes is facilitated primarily by FMO1 rather than P450 isoforms.The possibility of imipramine’s inhibitory effects occurring throughinhibition of rat P450 isoforms (Murray and Field, 1992; Masubuchiet al., 1995) can be ruled out on the basis of the results obtained with1-aminobenzotriazole.Mammalian FMOs (E.C.1.14.12.8) comprise a group of flavin adenine

dinucleotide-containing enzymes that use NADPH and molecular oxygento generate a 4a-hydroperoxyflavin intermediate, whichmediates the two-electron oxidation of soft, highly polarizable nucleophilic heteroatom(nitrogen, sulfur, and phosphorus)–containing xenobiotics (Hines et al.,1994; Cashman, 1995; Cashman et al., 1995; Ziegler, 2002; Krueger andWilliams, 2005; Phillips and Shephard, 2008). Our findings on the faciledecomposition of compound 1 to M1 in the presence of H2O2 andFMO1 are consistent with the notion that FMO1 will generallyoxygenate any nucleophilic heteroatom-containing compound that canbe oxidized by H2O2 and/or peracids (Bruice et al., 1983).To date, five distinct forms of FMO (i.e., FMO1–5) have been

identified (Lawton et al., 1994; Hernandez et al., 2004). Examinations ofadult human liver mRNA indicate high FMO3 (and FMO5) expressionbut low FMO1 expression (Dolphin et al., 1996; Koukouritaki et al.,2002; Koukouritaki and Hines, 2005; Hines, 2006; Zhang and Cashman,2006; Shimizu et al., 2011; Chen et al., 2016). In contrast, rat livers havebeen shown to express high levels of FMO1 protein (Itoh et al., 1993;Cherrington et al., 1998; Lattard et al., 2002a; Yamazaki et al., 2014).Expression of FMO1 and FMO3 in dog liver has also been reported with84%–89% amino acid sequence identity to the corresponding orthologsfrom rat and human (Ripp et al., 1999; Lattard et al., 2002b; Stevens et al.,2003). As such, dog FMO1 and dog FMO3 exhibit only 56% identitiesin primary amino acid sequence (Lattard et al., 2002b).To identify the specific FMO isoform responsible for the oxidative

desulfurization of compound 1, we conducted studies using recombinanthuman FMO isoforms. FMO3 and FMO5 were used because theyrepresent the isoforms most abundant in human liver, and FMO1 wasused because it is the ortholog of the form most abundant in adult ratand dog liver. The formation of M1 was principally mediated byrecombinant FMO1 with little to no contribution from FMO3 orFMO5 (apparent CLint for M1 formation by FMO1 was 154-foldhigher than that by FMO3), which is consistent with the inhibitoryeffects of imipramine on oxidative desulfurization of compound 1 in ratand dog liver microsomes.

Fig. 4. Oxidative desulfurization of compound 1(1 mM, d) to M1 (u) in NADPH-supplemented ratliver microsomes in the (A) absence or (B) presence ofheat inactivation (5 minutes, 50�C). Symbols depictmean and error bars for standard deviation.

Fig. 5. Oxidative desulfurization of compound 1 (1 mM) to M1 in humanrecombinant FMO1, FMO3, and FMO5. Incubations were conducted using 0.5 mg/ml supersomes in the presence of NADPH (1.3 mM) for 5 minutes at 37�C. Symbolsdepict mean and error bars for standard deviation.

1266 Eng et al.

at ASPE

T Journals on M

arch 14, 2020dm

d.aspetjournals.orgD

ownloaded from

Overall, these results suggest that liver microsomal FMO1 couldcontribute to the relatively high FMO-mediated oxidative desulfurizationof compound 1 in rat and dog liver microsomes and that lower expressionof FMO1 in human livers is a major determinant of oxidation potentialin livers from preclinical species and humans. The in vitro metabolicinstability of compound 1 also manifested in moderate to high CLp in ratsand dogs with significant circulating M1 concentrations measured in bothspecies, which provided an in vivo context for the in vitro findings,especially when considering that renal excretion of compound 1 in un-changed form was negligible in rats and dogs.Additional case studies on species differences in FMO-mediated

metabolism have also appeared in the literature, which strengthen ourobservations on the selective nature of FMO1-mediated oxidativedesulfurization in compound 1. For instance, a recent report from Liuet al., (2013) demonstrated that quinuclidine ring N-oxidation in aselective a7 neuronal acetylcholine receptor agonist ABT-107 (5-[6-[[(3R)-1-azabicyclo[2.2.2]octan-3-yl]oxy]pyridazin-3-yl]-1H-indole) oc-curred primarily in liver microsomes from rats and dogs (but not inhumans), and was also principally mediated by FMO1.

Because mRNA expression levels for FMO1 are higher in the humankidney (relative to the liver) (Dolphin et al., 1996; Zhang and Cashman,2006; Chen et al., 2016), we also examined the oxidative desulfurizationof compound 1 in NADPH-supplemented human kidney microsomes.Compared with recombinant FMO1, compound 1 was relatively stablein kidney microsomes with a minimal amount of M1 (;80 nM) formedin a NADPH-dependent fashion. Because of the unknown amount ofFMO in the commercial preparation of the human kidney microsomes,no rate comparisons can be made between recombinant and microsomalpreparations at the present time.Mammalian FMOs typically display high activity toward S-oxidation

in thioureas, thiones, and thiocarbamides. The initial oxygenation of thesulfur atom produces the electrophilic sulfenic acid (R-SOH) species thatis capable of reacting with nucleophiles, including GSH (Poulsen et al.,1979; Neal and Halpert, 1982; Krieter et al., 1984; Decker and Doerge,1992; Onderwater et al., 1999; Kim and Ziegler, 2000; Smith and Crespi,2002; Henderson et al., 2004). The sulfenic acid derivatives can undergoredox cycling in the presence of GSH coupled with the oxidation ofGSH to GSSG. The sulfenic acid metabolite can also undergo a second

Fig. 6. Kinetics of oxidative desulfurization ofcompound 1 to M1 by (A) human recombinantFMO1 (0.1 mg/ml microsomal protein) and (B)FMO3 (1 mg/ml). Incubations were conductedin the presence of NADPH (1.3 mM) at 37�Cfor 60 minutes in triplicate. Symbols depictmean and error bars for standard deviation.

Fig. 7. Oxidative desulfurization of compound 1 (1 mM) to M1 in NADPH-supplemented (A) rat and (B) dog liver microsomes in the absence or presence of FMO1 andP450 inhibitors, imipramine (250 mM) and 1-aminobenzotriazole (1000 mM), respectively. Symbols depict mean and error bars for standard deviation.

Oxidative Desulfurization by FMOs 1267

at ASPE

T Journals on M

arch 14, 2020dm

d.aspetjournals.orgD

ownloaded from

oxidation by FMO to the unstable sulfinic acid (R-SO2H), which is morereactive than the sulfenic acid metabolite and can damage the cell directlyor alkylate proteins (Onderwater et al., 1999, 2004; Ji et al., 2007).Our investigations on the oxidative desulfurization of compound 1 toM1

(Fig. 8) largely parallel the mechanistic insights noted in the literature. Forexample, addition of excess GSH essentially eliminated the conversion ofcompound 1 toM1 inNADPH-supplemented rat and dog livermicrosomes,suggesting that the initial FMO1-mediated S-oxygenation of compound 1leads to the corresponding sulfenic acid derivative 3, which undergoes redoxcycling to the parent compound 1 in a GSH-dependent fashion (presumablyvia oxidation of GSH to GSSG). A second oxidation of the sulfenic acidderivative 3 yields the more electrophilic sulfinic acid species 4, which istrapped intramolecularly (perhaps in a diffusion-controlled fashion) by thependant alcohol motif on the N1-substitutent in compound 1.In conclusion, our studies underscore one of the limitations of rats and

dogs as surrogates of adult human FMO-dependent drug metabolism

studies. The conclusions from previous animal studies that lackedsignificant amounts of liver FMO3 (i.e., rats and dogs) may need to bereconsidered. From a drug discovery perspective, our findings pro-vide a cautionary note against the use of allometric scaling of clearancefrom animals to human without a thorough knowledge of the overalldisposition/metabolic elimination mechanism of the molecule(s) underconsideration.

Authorship ContributionsParticipated in research design: Eng, Sharma, Wolford, Kalgutkar.Conducted experiments: Eng, Sharma, Conn.Contributed new reagents or analytic tools: Conn, Ruggeri.Performed data analysis: Eng, Sharma, Wolford, Dalvie, Kalgutkar.Wrote or contributed to the writing of the manuscript: Eng, Dalvie,

Buckbinder, Ruggeri, Di, Kalgutkar.

TABLE 1

Mean pharmacokinetic parameters of compound 1 and cyclic ether metabolite M1 in after intravenous (1 mg/kg)administration of compound 1 to Wistar-Han rats and beagle dogs

Pharmacokinetic parameters are expressed as mean 6 S.D.

Compound Species Dose CLp Vdss t1/2 AUC(0–‘)

mg/kg ml/min/kg l/kg h ng.h/ml

1 Rata 1.0 (n = 3) 73 6 13 1.4 6 0.5 0.4 6 0.2 233 6 37M1b NA NA 0.8 6 0.3 49.8 6 11.21 Dogc 1.0 (n = 3) 12 6 3.0 1.3 6 0.3 5.3 6 0.8 1450 6 310M1b NA NA 6.3 6 0.8 769 6 136

NA, not applicable.aIntravenous administration of compound 1 in 5% DMSO/95% of 30% 2-hydroxypropyl-b-cyclodextrin.bConcentrations of M1were estimated in animals that were administered compound 1 by i.v. route.cIntravenous administration of compound 1 as a solution in 25% 2-hydroxypropyl-b-cyclodextrin/Tris buffer (100 mM) (pH = 8.0).

Fig. 8. Proposed mechanism of oxidative de-sulfurization of compound 1 to M1 in NADPH-supplemented rat and dog liver microsomes.

1268 Eng et al.

at ASPE

T Journals on M

arch 14, 2020dm

d.aspetjournals.orgD

ownloaded from

References

Boily MO, Chauret N, Laterreur J, Leblond FA, Boudreau C, Duquet MC, Lévesque JF, Ste-MarieL, and Pichette V (2015) In vitro and in vivo mechanistic studies toward understanding the roleof 1-aminobenzotriazole in rat drug-drug interactions. Drug Metab Dispos 43:1960–1965.

Bruice TC, Noar JB, Ball SS, and Venkataram UV (1983) Monooxygen donation potential of4a-hydroperoxyflavins as compared with those of percarboxylic acid and other hydroperoxides.Monooxygen donation to olefin, tertiary amine, alkyl sulfide, and iodide ion. J Am Chem Soc 105:2452–2463.

Caldwell GW, Ritchie DM, Masucci JA, Hageman W, Cotto C, Hall J, Hasting B, and Jones W(2005) The use of the suicide CYP450 inhibitor ABT for distinguishing absorption and metab-olism processes in in-vivo pharmacokinetic screens. Eur J Drug Metab Pharmacokinet 30:75–83.

Cashman JR (1995) Structural and catalytic properties of the mammalian flavin-containing mon-ooxygenase. Chem Res Toxicol 8:166–181.

Cashman JR, Park SB, Berkman CE, and Cashman LE (1995) Role of hepatic flavin-containingmonooxygenase 3 in drug and chemical metabolism in adult humans. Chem Biol Interact 96:33–46.

Chen Y, Zane NR, Thakker DR, and Wang MZ (2016) Quantification of flavin-containing mon-ooxygenases 1, 3, and 5 in human liver microsomes by UPLC-MRM-based targeted quantitativeproteomics and its application to the study of ontogeny. Drug Metab Dispos DOI: 10.1124/dmd.115.067538 [published ahead of print].

Cherrington NJ, Cao Y, Cherrington JW, Rose RL, and Hodgson E (1998) Physiological factorsaffecting protein expression of flavin-containing monooxygenases 1, 3 and 5. Xenobiotica 28:673–682.

Cooper DS (2005) Antithyroid drugs. N Engl J Med 352:905–917.Decker CJ and Doerge DR (1992) Covalent binding of 14C- and 35S-labeled thiocarbamides in rathepatic microsomes. Biochem Pharmacol 43:881–888.

Dixit A and Roche TE (1984) Spectrophotometric assay of the flavin-containing monooxygenaseand changes in its activity in female mouse liver with nutritional and diurnal conditions. ArchBiochem Biophys 233:50–63.

Dolphin CT, Cullingford TE, Shephard EA, Smith RL, and Phillips IR (1996) Differential de-velopmental and tissue-specific regulation of expression of the genes encoding three members ofthe flavin-containing monooxygenase family of man, FMO1, FMO3 and FM04. Eur J Biochem235:683–689.

Futcher PH and Massie E (1950) Agranulocytosis due to propylthiouracil. Ann Intern Med 32:137.Henderson MC, Krueger SK, Stevens JF, and Williams DE (2004) Human flavin-containingmonooxygenase form 2 S-oxygenation: sulfenic acid formation from thioureas and oxidation ofglutathione. Chem Res Toxicol 17:633–640.

Hernandez D, Janmohamed A, Chandan P, Phillips IR, and Shephard EA (2004) Organization andevolution of the flavin-containing monooxygenase genes of human and mouse: identification ofnovel gene and pseudogene clusters. Pharmacogenetics 14:117–130.

Hines RN, Cashman JR, Philpot RM, Williams DE, and Ziegler DM (1994) The mammalianflavin-containing monooxygenases: molecular characterization and regulation of expression.Toxicol Appl Pharmacol 125:1–6.

Hines RN (2006) Developmental and tissue-specific expression of human flavin-containingmonooxygenases 1 and 3. Expert Opin Drug Metab Toxicol 2:41–49.

Ichiki Y, Akahoshi M, Yamashita N, Morita C, Maruyama T, Horiuchi T, Hayashida K, IshibashiH, and Niho Y (1998) Propylthiouracil-induced severe hepatitis: a case report and review of theliterature. J Gastroenterol 33:747–750.

Itoh K, Kimura T, Yokoi T, Itoh S, and Kamataki T (1993) Rat liver flavin-containing mono-oxygenase (FMO): cDNA cloning and expression in yeast. Biochim Biophys Acta 1173:165–171.

Ji T, Ikehata K, Koen YM, Esch SW, Williams TD, and Hanzlik RP (2007) Covalent modificationof microsomal lipids by thiobenzamide metabolites in vivo. Chem Res Toxicol 20:701–708.

Jiang X, Khursigara G, and Rubin RL (1994) Transformation of lupus-inducing drugs to cytotoxicproducts by activated neutrophils. Science 266:810–813.

Kalm MJ (1961) Peroxide desulfurization of thioureas. J Org Chem 26:2925–2929.Kim YM and Ziegler DM (2000) Size limits of thiocarbamides accepted as substrates by humanflavin-containing monooxygenase 1. Drug Metab Dispos 28:1003–1006.

Kitamura IR (1934) The action of hydrogen peroxide on organic sulfur compounds. YakugakuZasshi 54:1–13.

Koukouritaki SB and Hines RN (2005) Flavin-containing monooxygenase genetic polymorphism:impact on chemical metabolism and drug development. Pharmacogenomics 6:807–822.

Koukouritaki SB, Simpson P, Yeung CK, Rettie AE, and Hines RN (2002) Human hepatic flavin-containing monooxygenases 1 (FMO1) and 3 (FMO3) developmental expression. Pediatr Res51:236–243.

Krieter PA, Ziegler DM, Hill KE, and Burk RF (1984) Increased biliary GSSG efflux from ratlivers perfused with thiocarbamide substrates for the flavin-containing monooxygenase. MolPharmacol 26:122–127.

Krueger SK and Williams DE (2005) Mammalian flavin-containing monooxygenases: structure/function, genetic polymorphisms and role in drug metabolism. Pharmacol Ther 106:357–387.

Lattard V, Lachuer J, Buronfosse T, Garnier F, and Benoit E (2002a) Physiological factors affectingthe expression of FMO1 and FMO3 in the rat liver and kidney. Biochem Pharmacol 63:1453–1464.

Lattard V, Longin-Sauvageon C, Lachuer J, Delatour P, and Benoit E (2002b) Cloning, se-quencing, and tissue-dependent expression of flavin-containing monooxygenase (FMO) 1 andFMO3 in the dog. Drug Metab Dispos 30:119–128.

Lawton MP, Cashman JR, Cresteil T, Dolphin CT, Elfarra AA, Hines RN, Hodgson E, Kimura T,Ozols J, and Phillips IR, et al. (1994) A nomenclature for the mammalian flavin-containingmonooxygenase gene family based on amino acid sequence identities. Arch Biochem Biophys308:254–257.

Lee E, Hirouchi M, Hosokawa M, Sayo H, Kohno M, and Kariya K (1988) Inactivation ofperoxidases of rat bone marrow by repeated administration of propylthiouracil is accompaniedby a change in the heme structure. Biochem Pharmacol 37:2151–2153.

Lee E, Miki Y, Katsura H, and Kariya K (1990) Mechanism of inactivation of myeloperoxidase bypropylthiouracil. Biochem Pharmacol 39:1467–1471.

Lee SK, Kang MJ, Jin C, In MK, Kim DH, and Yoo HH (2009) Flavin-containing monooxygenase1-catalysed N,N-dimethylamphetamine N-oxidation. Xenobiotica 39:680–686.

Liu H, Deng X, Liu J, Liu N, Stuart P, Xu H, Guan Z, Marsh KC, and De Morais SM (2013)Species-dependent metabolism of a novel selective a7 neuronal acetylcholine receptor agonistABT-107. Xenobiotica 43:803–816.

Manivel JC, Bloomer JR, and Snover DC (1987) Progressive bile duct injury after thiabendazoleadministration. Gastroenterology 93:245–249.

Martinez-Lopez JI, Greenberg SE, and Kling RR (1962) Drug-induced hepatic injury duringmethimazole therapy. Gastroenterology 43:84–87.

Masubuchi Y, Takahashii C, Fujio N, Horie T, Suzuki T, Imaoka S, Funae Y, and Narimatsu S(1995) Inhibition and induction of cytochrome P450 isozymes after repetitive administration ofimipramine in rats. Drug Metab Dispos 23:999–1003.

Mizutani T, Yoshida K, and Kawazoe S (1993) Possible role of thioformamide as a proximatetoxicant in the nephrotoxicity of thiabendazole and related thiazoles in glutathione-depletedmice: structure-toxicity and metabolic studies. Chem Res Toxicol 6:174–179.

Mizutani T, Yoshida K, Murakami M, Shirai M, and Kawazoe S (2000) Evidence for the in-volvement of N-methylthiourea, a ring cleavage metabolite, in the hepatotoxicity of methimazolein glutathione-depleted mice: structure-toxicity and metabolic studies. Chem Res Toxicol 13:170–176.

Murray M and Field SL (1992) Inhibition and metabolite complexation of rat hepatic microsomalcytochrome P450 by tricyclic antidepressants. Biochem Pharmacol 43:2065–2071.

Neal RA and Halpert J (1982) Toxicology of thiono-sulfur compounds. Annu Rev PharmacolToxicol 22:321–339.

Onderwater RC, Commandeur JN, Menge WM, and Vermeulen NP (1999) Activation of micro-somal glutathione S-transferase and inhibition of cytochrome P450 1A1 activity as a modelsystem for detecting protein alkylation by thiourea-containing compounds in rat liver micro-somes. Chem Res Toxicol 12:396–402.

Onderwater RC, Commandeur JN, and Vermeulen NP (2004) Comparative cytotoxicity ofN-substituted N9-(4-imidazole-ethyl)thiourea in precision-cut rat liver slices. Toxicology197:81–91.

Parrish KE, Mao J, Chen J, Jaochico A, Ly J, Ho Q, Mukadam S, and Wright M (2015) In vitro andin vivo characterization of CYP inhibition by 1-aminobenzotriazole in rats. Biopharm DrugDispos DOI: 10.1002/bdd.2000 [published ahead of print].

Phillips IR and Shephard EA (2008) Flavin-containing monooxygenases: mutations, disease anddrug response. Trends Pharmacol Sci 29:294–301.

Poulsen LL, Hyslop RM, and Ziegler DM (1979) S-oxygenation of N-substituted thioureas cata-lyzed by the pig liver microsomal FAD-containing monooxygenase. Arch Biochem Biophys 198:78–88.

Ripp SL, Itagaki K, Philpot RM, and Elfarra AA (1999) Species and sex differences in expressionof flavin-containing monooxygenase form 3 in liver and kidney microsomes. Drug MetabDispos 27:46–52.

Ruggeri RB, Buckbinder L, Bagley SW, Carpino PA, Conn EL, Dowling MS, Fernando DP, JiaoW, Kung DW, and Orr ST, et al. (2015) Discovery of 2-(6-(5-chloro-2-methoxyphenyl)-4-oxo-2-thioxo-3,4-dihydropyrimidin-1(2H)-yl)acetamide (PF-06282999): a highly selective mechanism-based myeloperoxidase inhibitor for the treatment of cardiovascular diseases. J Med Chem 58:8513–8528.

Shimizu M, Denton T, Kozono M, Cashman JR, Leeder JS, and Yamazaki H (2011) De-velopmental variations in metabolic capacity of flavin-containing mono-oxygenase 3 in child-hood. Br J Clin Pharmacol 71:585–591.

Smith PB and Crespi C (2002) Thiourea toxicity in mouse C3H/10T1/2 cells expressing humanflavin-dependent monooxygenase 3. Biochem Pharmacol 63:1941–1948.

Stevens JC, Melton RJ, Zaya MJ, and Engel LC (2003) Expression and characterization of func-tional dog flavin-containing monooxygenase 1. Mol Pharmacol 63:271–275.

Strelevitz TJ, Foti RS, and Fisher MB (2006) In vivo use of the P450 inactivator 1-amino-benzotriazole in the rat: varied dosing route to elucidate gut and liver contributions to first-passand systemic clearance. J Pharm Sci 95:1334–1341.

Tidén AK, Sjögren T, Svensson M, Bernlind A, Senthilmohan R, Auchère F, Norman H, MarkgrenPO, Gustavsson S, and Schmidt S, et al. (2011) 2-Thioxanthines are mechanism-based inacti-vators of myeloperoxidase that block oxidative stress during inflammation. J Biol Chem 286:37578–37589.

Waldhauser L and Uetrecht J (1991) Oxidation of propylthiouracil to reactive metabolites byactivated neutrophils. Implications for agranulocytosis. Drug Metab Dispos 19:354–359.

Yamazaki M, Shimizu M, Uno Y, and Yamazaki H (2014) Drug oxygenation activities mediatedby liver microsomal flavin-containing monooxygenases 1 and 3 in humans, monkeys, rats, andminipigs. Biochem Pharmacol 90:159–165.

Zhang J and Cashman JR (2006) Quantitative analysis of FMO gene mRNA levels in humantissues. Drug Metab Dispos 34:19–26.

Ziegler DM (1980) Microsomal flavin-containing monooxygenase: oxygenation of nucleophilicnitrogen and sulfur compounds, in Enzymatic Basis of Detoxication (Jakoby WB, ed) vol 1, pp201–227 (Academic Press, New York).

Ziegler DM (2002) An overview of the mechanism, substrate specificities, and structure of FMOs.Drug Metab Rev 34:503–511.

Address correspondence to: Dr. Amit S. Kalgutkar, Pharmacokinetics, Dynamics,and Metabolism–New Chemical Entities, Pfizer Worldwide Research and Devel-opment, 610 Main Street, Cambridge, MA 02139. E-mail: amit.kalgutkar@pfizer.com

Oxidative Desulfurization by FMOs 1269

at ASPE

T Journals on M

arch 14, 2020dm

d.aspetjournals.orgD

ownloaded from