reactive metabolite trapping studies on imidazo- and 2...

DMD #51839

1

Reactive Metabolite Trapping Studies on Imidazo- and 2-Methylimidazo[2,1-

b]thiazole-based Inverse Agonists of the Ghrelin Receptor

Amit S. Kalgutkar, Tim Ryder, Gregory S. Walker, Suvi T. M. Orr, Shawn Cabral, Theunis C.

Goosen, Kimberly Lapham and Heather Eng

Pharmacokinetics, Dynamics and Metabolism–New Chemical Entities, Groton, CT (T.R.,

G.S.W., T.C.G., K.L., H.E.) and Cambridge MA (A.S.K.)

Worldwide Medicinal Chemistry, La Jolla, CA (S.T.M.O.) and Groton, CT (S.C.)

Pfizer Inc.

DMD Fast Forward. Published on April 22, 2013 as doi:10.1124/dmd.113.051839

Copyright 2013 by the American Society for Pharmacology and Experimental Therapeutics.

This article has not been copyedited and formatted. The final version may differ from this version.DMD Fast Forward. Published on April 22, 2013 as DOI: 10.1124/dmd.113.051839

at ASPE

T Journals on February 22, 2020

dmd.aspetjournals.org

Dow

nloaded from

http://dmd.aspetjournals.org/

DMD #51839

2

Running Title: Bioactivation of imidazo[2,1-b]thiazole derivatives

Address correspondence to: Amit S. Kalgutkar, Pharmacokinetics, Dynamics, and

Metabolism-New Chemical Entities, Pfizer Worldwide Research and Development,

Cambridge, MA 02139. Phone: (617)-551-3336. E-mail: [email protected]

Text Pages (including references): 35

Tables: 0

Figures: 16

References: 45

Abstract: 221

Introduction: 697

Discussion: 1550


at ASPE



Dow

nloaded from


DMD #51839

3

Abbreviations used are: GHS-R1a, growth hormone secretagogue receptor type 1a; SAR,

structure-activity relationship; compound 1, 1-(2-(2-chloro-4-(2H-1,2,3-triazol-2-yl)benzyl)-2,7-

diazaspiro[3.5]nonan-7-yl)-2-(imidazo[2,1-b]thiazol-6-yl)ethanone; compound 2, 1-(2-(2-chloro-

4-(2H-1,2,3-triazol-2-yl)benzyl)-2,7-diazaspiro[3.5]nonan-7-yl)-2-(2-methylimidazo[2,1-

b]thiazol-6-yl)ethanone; CYP, cytochrome P450; GSH, reduced glutathione; L-766,112, 5-(4-

(methylsulfonyl)phenyl)-6-phenylimidazo[2,1-b]thiazole; HLM, human liver microsomes; 3, 1-

(2-(2-chloro-4-(2H-1,2,3-triazol-2-yl)benzyl)-2,7-diazaspiro[3.5]nonan-7-yl)-2-(2-

methylimidazo[2,1-b][1,3,4]thiadiazol-6-yl)ethanone; NADPH, nicotinamide adenine

dinucleotide phosphate; DMSO-d6, dimethyl sulfoxide-d6; LC-MS/MS, liquid chromatography

tandem mass spectrometry; TCI, triple resonance probe; COSY, correlation spectroscopy;

TOCSCY, total correlation spectroscopy; HSQC, multiplicity edited heteronuclear single

quantum coherence; HMBC, heteronuclear multiple bond correlation; tR, retention time; CID,

collision-induced dissociation.


at ASPE



Dow

nloaded from


DMD #51839

4

Abstract

The current study examined the bioactivation potential of ghrelin receptor inverse agonists, 1-(2-

(2-chloro-4-(2H-1,2,3-triazol-2-yl)benzyl)-2,7-diazaspiro[3.5]nonan-7-yl)-2-(imidazo[2,1-

b]thiazol-6-yl)ethanone (1) and 1-(2-(2-chloro-4-(2H-1,2,3-triazol-2-yl)benzyl)-2,7-

diazaspiro[3.5]nonan-7-yl)-2-(2-methylimidazo[2,1-b]thiazol-6-yl)ethanone (2), containing a

fused imidazo[2,1-b]thiazole motif in the core structure. Both compounds underwent oxidative

metabolism in NADPH- and glutathione-supplemented human liver microsomes to yield

glutathione conjugates, which was consistent with their bioactivation to reactive species. Mass

spectral fragmentation and NMR analysis indicated that the site of attachment of the glutathionyl

moiety in the thiol conjugates was on the thiazole ring within the bicycle. Two glutathione

conjugates were discerned with the imidazo[2,1-b]thiazole derivative 1. One adduct was derived

from the Michael addition of glutathione to a putative S-oxide metabolite of 1, whereas, the

second adduct was formed via the reaction of a second glutathione molecule with the initial

glutathione-S-oxide adduct. In the case of the 2-methylimidazo[2,1-b]thiazole analog 2,

glutathione conjugation occurred via an oxidative desulfation mechanism, possibly involving

thiazole ring epoxidation as the rate-limiting step. Additional insights into the mechanism were

obtained via 18O exchange and trapping studies with potassium cyanide. The mechanistic

insights into the bioactivation pathways of 1 and 2 allowed the deployment of a rational chemical

intervention strategy that involved replacement of the thiazole ring with a 1,2,4-thiadiazole group

to yield -(2-(2-chloro-4-(2H-1,2,3-triazol-2-yl)benzyl)-2,7-diazaspiro[3.5]nonan-7-yl)-2-(2-

methylimidazo[2,1-b][1,3,4]thiadiazol-6-yl)ethanone (3). These structural changes not only


at ASPE



Dow

nloaded from


DMD #51839

5

abrogated the bioactivation liability but also retained the attractive pharmacological attributes of

the prototype agents.


at ASPE



Dow

nloaded from


DMD #51839

6

Introduction

Ghrelin is a 28-amino acid hunger-stimulating peptide that is produced in the stomach, pancreas

and hypothalamus (Kojima et al., 1999; Inui et al., 2004). Ghrelin is a ligand for the growth

hormone secretagogue receptor type 1a (GHS-R1a), a G-protein-coupled receptor expressed

primarily in the pituitary gland, brain, and to a lesser extent in the periphery. Through its action

on GHS-R1a, ghrelin exerts a variety of metabolic functions including stimulation of growth

hormone release, stimulation of appetite and weight gain, and suppression of insulin secretion in

rodents and humans (Wren et al., 2001; Nakazato et al., 2001; Dezaki et al., 2008; Cardona Cano

et al., 2012; Heppner et al., 2012). Thus, antagonizing GHSR-1a with small molecule

antagonists or inverse agonists is anticipated to improve glucose homeostasis and insulin

sensitivity, while eliciting beneficial effects on body weight (Serby et al., 2006; Rudolph et al.,

2007; Esler et al., 2008; Soares et al., 2008; Costantino and Barlocco, 2009).

A recent report from our laboratory disclosed a proprietary series of spirocyclic piperidine-

azetidine derivatives as potent and orally active inverse agonists of GHSR-1a (Kung et al.,

2012). Structure-activity relationship (SAR) studies revealed that analogs with distributed

polarity in the form of the fused imidazo[2,1-b]thiazole acetamide and phenyl triazole

functionalities (e.g., 1-(2-(2-chloro-4-(2H-1,2,3-triazol-2-yl)benzyl)-2,7-diazaspiro[3.5]nonan-7-

yl)-2-(imidazo[2,1-b]thiazol-6-yl)ethanone (1) and 1-(2-(2-chloro-4-(2H-1,2,3-triazol-2-

yl)benzyl)-2,7-diazaspiro[3.5]nonan-7-yl)-2-(2-methylimidazo[2,1-b]thiazol-6-yl)ethanone (2)

depicted in Figure 1) demonstrated significant improvements in binding affinity (IC50 < 10 nM)

to the ghrelin receptor, while consistently displaying inverse agonism in a ghrelin functional

assay. In addition, these structural changes also led to an improved pharmacokinetic profile (↓


at ASPE



Dow

nloaded from


DMD #51839

7

clearance, ↑oral bioavailability) in animals, which warranted a further examination of relevant

preclinical safety endpoints as a prelude to candidate nomination.

From a drug metabolism perspective, the presence of the imidazo[2,1-b]thiazole scaffold in 1

and 2 raised a significant cause for concern, given the possibility of a cytochrome P450 (CYP)-

mediated S-oxidation on the thiazole sulfur (within the imidazo[2,1-b]thiazole framework) to

yield an electrophilic S-oxide metabolite capable of reacting with proteins and/or the endogenous

antioxidant glutathione (GSH). Indeed, the characterization of a GSH conjugate derived from

Michael addition to an S-oxide metabolite of the imidazo[2,1-b]thiazole derivative, 5-(4-

(methylsulfonyl)phenyl)-6-phenylimidazo[2,1-b]thiazole inhibitor (L-766,112) (Figure 1)

provides precedence for this bioactivation pathway in liver microsomes (Thérien et al., 1997;

Trimble et al., 1997). Furthermore, a successful medicinal chemistry strategy involving

internalization of the thiazole motif, resulting in the elimination of reactive metabolite liability of

L-766,112, while retaining the primary pharmacology (selective inhibition of cyclooxygenase-2)

has been presented in subsequent work (Figure 1) (Roy et al., 1997). Besides thiazole-S-

oxidation, an additional bioactivation pathway can arise through thiazole ring scission (via the

C4–C5 epoxidation → diol pathway) (Chatfield and Hunter, 1973; Mizutani et al., 1993; Yabuki

et al., 1997; Yoon et al., 1998; Obach et al., 2008). The corresponding thioamide metabolites,

which are the by-products of thiazole ring cleavage, can undergo further S-oxidation to

electrophilic sulfenic acid derivatives (Mansuy and Dansette, 2011), capable of oxidizing or

forming mixed disulfide adducts with proteins or GSH and eliciting toxicity (Ziegler-Skylakakis

et al., 1998; Hajovsky et al., 2012).


at ASPE



Dow

nloaded from


DMD #51839

8

In this report, we summarize our findings on the in vitro bioactivation of 1 and 2 in human

liver microsomes (HLM) utilizing exogenously added nucleophiles to trap reactive species. In

the case of the imidazo[2,1-b]thiazole derivative 1, two GSH adducts were formed. One adduct

was derived from the addition of the thiol to a putative S-oxide metabolite of 1, whereas, the

formation of the second adduct involved the attachment of a second GSH molecule to the initial

GSH-S-oxide adduct with concomitant liberation of the oxidized thiazole sulfur. With respect to

the corresponding 2-methylimidazo[2,1-b]thiazole analog 2, GSH conjugation occurred via an

oxidative desulfation mechanism, possibly involving thiazole ring epoxidation as the rate-

limiting step. Characterization of the GSH conjugate structure and additional trapping studies

with cyanide ion proved to be useful in deciphering the overall bioactivation mechanism

associated with 2. The in vitro metabolism information was used to design the corresponding

thiadiazole analog, 1-(2-(2-chloro-4-(2H-1,2,3-triazol-2-yl)benzyl)-2,7-diazaspiro[3.5]nonan-7-

yl)-2-(2-methylimidazo[2,1-b][1,3,4]thiadiazol-6-yl)ethanone analog 3 (see Figure 1) that would

be devoid of reactive metabolite formation. Indeed, 3 was resistant to bioactivation in HLM

(inferred from the lack of GSH conjugate formation), while retaining inverse agonist potency for

the ghrelin receptor and HLM stability noted with 1 and 2.

Materials and Methods

Materials. Reduced nicotinamide adenine dinucleotide phosphate (NADPH), GSH and

potassium cyanide were obtained from Sigma-Aldrich (St. Louis, MO). HLM were prepared

from a mixed gender pool of 50 donors (provided by BD Biosciences, Woburn, MA). H218O

(97%) and dimethyl sulfoxide (DMSO)-d6 “100%” were obtained from Cambridge Isotope

Laboratories Inc. (Andover, MA). All other commercially available reagents and solvents were


at ASPE



Dow

nloaded from


DMD #51839

9

of either analytical or HPLC grade. The preparation of title compounds 1–3 is provided in the

supplementary section of this paper (see Supplemental Methods and Supplemental Figure 1).

Metabolite Identification and Reactive Metabolite Trapping Studies in HLM. Liver

microsomal incubations were conducted at 37 °C for 60 minutes in a shaking water bath. The

incubation volume was 1.0 ml and consisted of 0.1 M potassium phosphate buffer (pH 7.4),

MgCl2 (3.3 mM), HLM (protein concentration = 1.0 mg/ml, CYP concentration = 0.5 μM), test

compound (10 µM) and NADPH (1.3 mM). Additional microsomal incubations also included

GSH (10 mM), N-acetylcysteine (10 mM) and/or potassium cyanide (5 mM) to trap reactive

metabolites. Reactions were terminated by the addition of acetonitrile (3.0 ml). The solutions

were centrifuged (3000g, 15 min), and the supernatants were transferred to clean 15 ml glass

centrifuge tubes and concentrated to dryness. The residue was reconstituted with 200 μl 10%

acetonitrile in water and analyzed for metabolite formation using liquid chromatography tandem

mass spectrometry (LC-MS/MS).

18O Incorporation Studies in HLM. For H218O studies, 15 ml glass centrifuge tubes containing

1.0 ml of 0.1 M phosphate buffer (pH 7.4) were concentrated to dryness using an evaporative

centrifuge. H218O (840 μl, 97%) was added to the dried centrifuge containing phosphate buffer.

Potassium phosphate (pH 7.4) buffered H218O was used to create 1.0 ml incubations containing 2

(10 μM), and 3.3 mM MgCl2 (20 μl), in HLM (protein concentration = 1.0 mg/ml, CYP

concentration = 0.5 μM). The isotopic enrichment of H218O was ~ 89.7 %. Identical incubations

were also performed in the presence of GSH (10 mM). Samples were placed in a shaking water

bath set to 37 °C. After 60 minutes, the reactions were quenched with 3.0 ml of acetonitrile and

centrifuged at 3000g for 5 minutes. The supernatants were transferred to clean 15 ml centrifuge


at ASPE



Dow

nloaded from


DMD #51839

10

tubes and concentrated to dryness. The samples were analyzed for 18O incorporation by LC-

MS/MS.

LC-MS/MS Methodology for Metabolite Identification Studies. The HPLC system consisted

of an Accela quaternary solvent delivery pump, an Acella autoinjector, and a Surveyor PDA Plus

photodiode array detector (Thermo Electron Corporation, Waltham, MA). Chromatography was

performed on a Phenomenex Hydro RP column (4 μm 4.6 mm x 150 mm) (Phenomenex,

Torrance, CA). The mobile phase was composed of 5 mM ammonium formate buffer (pH = 3.0)

(solvent A) and acetonitrile (solvent B). The flow rate was 1.0 ml/min. The LC gradient started

at 10% B for 5 minutes, was ramped linearly to 50% B over 35 minutes, ramped linearly to 90%

B over 10 minutes, returned to the initial condition over 1.0 minutes, and allowed to equilibrate

for 4.0 minutes. Post-column flow passed through the PDA detector to provide UV (λ= 254 nm)

detection prior to being split to the mass spectrometer such that mobile phase was introduced into

the electrospray source at a rate of 100 μl/min. The LC system was interfaced to a Thermo

Orbitrap mass spectrometer operating in positive ion electrospray mode. Xcalibur software,

version 2.0, was used to control the HPLC/MS system. Full scan data were collected at 15,000

resolution. Data dependent product ion scans of the two most intense ions found in the full scan

were obtained at 15,000 resolution. The dynamic exclusion function was used with a 1.0 minute

exclusion duration after 3 successive product ion scans with an early exclusion if the precursor

ion falls below a signal to noise of 20.

Biosynthesis of GSH Conjugates. GSH conjugates M3-1 and M5-2 detected in the course of

reactive metabolite trapping studies with 1 and 2 were biosynthesized via a scale up of the

human liver microsomal incubations. Six incubations of 1 or 2 were carried out in 50 ml


at ASPE



Dow

nloaded from


DMD #51839

11

polypropylene centrifuge tubes at 37 °C for 60 min in a shaking water bath. The incubation

volume was 10 ml and consisted of 0.1 M potassium phosphate buffer (pH 7.4), MgCl2 (3.3

mM), HLM (protein concentration = 1.0 mg/ml), test compound (60 µM), GSH (10 mM) and

NADPH (1.3 mM). Reactions were terminated by the addition of acetonitrile (30 ml). The

solutions were centrifuged (3000g, 15 min), and the supernatants were transferred to clean 50 ml

polypropylene centrifuge tubes, and concentrated to dryness. The residue in each tube was

reconstituted with 400 μl 10% acetonitrile in water. The reconstituted samples were combined in

a 2 ml HPLC vial and were onto an HPLC semipreparative system in 8 3 μl injections. This

system consisted of a Shimadzu SiL-HTC autosampler, two LC-20AD solvent pumps, an SPD-

M20A diode array detector, and a FRC-10 A fraction collector (Shimadzu USA, Columbia,

MD). Separation was performed on a Zorbax RX C8, 9.4 x 250mm, 5 μM semipreparative

HPLC column. The mobile phase was composed of 0.1% formic acid (solvent A) and

acetonitrile (solvent B). The flow rate was 4.0 ml/min. The LC gradient started at 10% B for 5

minutes, ramped linearly to 50% B over 35 minutes, ramped linearly to 90% B over 10 minutes,

returned to the initial condition over 1.0 minutes and allowed to equilibrate for 4.0 minutes.

HPLC fractions were collected throughout the run at 1.0 minute intervals. Aliquots (100 μl) of

fractions at the retention time (tR) of the UV peak corresponding to the GSH conjugate (M3-1 or

M5-2) were analyzed by LC/MS for verification. Fractions containing the peak were combined

into a single 15 ml glass centrifuge tube and concentrated to dryness. In the case of M3-1, it was

difficult to obtain a pure isolate due to poor chromatographic resolution between M3-1 and M2-

1. Hence, a crude isolate of M3-1 was used in NMR characterization. The residues were

evacuated in a drybox for ~ 2 hours prior to reconstitution in DMSO-d6 for NMR analysis.


at ASPE



Dow

nloaded from


DMD #51839

12

NMR Analysis of Metabolites. NMR spectra for 1 and M3-1 were recorded on a Bruker

Avance 600 MHz instrument (Bruker BioSpin Corporation, Billerica, MA) controlled by

TOPSPIN V3.0 and equipped with a 1.7 mm cryo-triple resonance probe (TCI). NMR spectra

for 2 and M5-2 were recorded on a Bruker Avance 600 MHz instrument (Bruker BioSpin

Corporation, Billerica, MA) controlled by TOPSPIN V2.1 and equipped with a 5 mm cryo-TCI

probe. Synthetic samples and isolated materials were dissolved in 0.15 ml (5 mm TCI probe) or

0.05 ml (1.7 mm TCI probe) of DMSO-d6. All spectra were referenced using residual DMSO-d6

(δ = 2.49 ppm relative to tetramethylsilane, δ = 0.00 for 1H and δ = 39.5 ppm relative to

tetramethylsilane, δ = 0.00 for 13C). One dimensional spectra were typically recorded using a

sweep width of 8000 Hz and a total recycle time of approximately 7 seconds. The resulting

time-averaged free induction decays were transformed using an exponential line broadening of

1.0 Hz to enhance signal to noise. The two-dimensional data (correlation spectroscopy [COSY],

total correlation spectroscopy [TOCSY], multiplicity edited heteronuclear single quantum

coherence [HSQC], and heteronuclear multiple bond correlation [HMBC]) were recorded using

the standard pulse sequences provided by Bruker. A 1K x 128 data matrix was acquired using a

minimum of four scans and 16 dummy scans. The data was zero-filled to a size of 1K x 1K. A

mixing time of 80 ms was used in the TOCSY experiments.

Results

In Vitro Metabolism of Imidazo[2,1-b]thiazole Derivative 1 in HLM. In HLM supplemented

with NADPH and GSH, three metabolites of 1 (denoted as M1-1, M2-1, and M3-1) were

observed (Figure 2, panel A). Collision-induced dissociation (CID) spectra of 1 and its

metabolites M1-1, M2-1 and M3-1 are shown in Figures 3–5. In the case of 1 (tR = 18.82


at ASPE



Dow

nloaded from


DMD #51839

13

minutes; MH+ = 482.1524), the major fragment ions present at m/z 318.1476, 301.1210,

262.1004, 192.0318, and 165.0112 were consistent with the structure (see CID spectrum in

Figure 3, panel A). Metabolite M1-1 eluted at tR = 17.70 minutes, and possessed MH+ =

516.1579, which is an addition of 34.0055 Da to the molecular weight of 1 (Figure 3, panel B).

The presence of the fragment ions at m/z 318.1476 and 192.0318 in the CID spectrum of M1-1

indicated that the 2-(2-chloro-4-(2H-1,2,3-triazol-2-yl)benzyl-2,7-diazaspiro[3,5]-nonane)

portion was unaltered. A proposed structure of M1-1 that is consistent with the observed

molecular weight and mass spectrum is shown in Figure 3, panel B.

Metabolites M2-1 (tR = 17.08 minutes) and M3-1 (tR = 16.47 minutes) possessed MH+ at

1096.3200 m/z and 805.2311 m/z, respectively. The CID spectra of M2-1 and M3-1 are shown

in Figures 4 and 5, respectively. M2-1 contained two molecules of GSH in its structure as

evident from the facile loss of two pyroglutamic acid components (m/z 967.2753 and m/z

838.2335) in its CID spectrum. Furthermore, the presence of the fragment ion at m/z 318.1468

implied that the 2-(2-chloro-4-(2H-1,2,3-triazol-2-yl)benzyl-2,7-diazaspiro[3,5]-nonane) portion

in M2-1 was unchanged. A structure of M2-1 that is compatible with the observed molecular

weight and fragmentation pattern is shown in Figure 4 and a mechanistic hypothesis that

rationalizes its formation is provided in Figure 15 in the discussion section.

The molecular weight of M3-1 (MH+ = 805.2312) suggested that the metabolite was derived

from the addition of GSH to an oxidized metabolite of 1. The fragment ions at m/z 730.1973 and

m/z 676.1872 in the CID spectrum of M3-1 were derived from the characteristic losses of the

glycine (75 Da) and glutamic acid components (129 Da) of GSH (Baillie and Davis, 1993). The

diagnostic fragment ions at m/z 488.0904, 359.0471, and 318.1476 indicated that the 2-(2-chloro-


at ASPE



Dow

nloaded from


DMD #51839

14

4-(2H-1,2,3-triazol-2-yl)benzyl-2,7-diazaspiro[3,5]-nonane) scaffold in 1 was unmodified.

Based on the fragmentation pattern, we inferred that oxidation and subsequent GSH conjugation

had occurred on the imidazo[2,1-b]thiazole ring. There are two obvious pathways through which

this can arise, either via a CYP mediated epoxidation at the C4-C5 position of the thiazole ring

with addition of GSH or through a CYP mediated S-oxidation of the thiazole sulfur followed by

addition of GSH to the S-oxide metabolite (structure depicted in Figure 5), similar to one

discerned with the imidazo[2,1-b]-thiazole derivative L-766,112 (Trimble et al., 1996). In order

to distinguish between the two possibilities, a crude isolate of M3-1 was obtained from a large-

scale HLM incubation of 1 in the presence of NADPH and GSH, and subsequently characterized

by NMR spectroscopy.

Observed in the 1H-13C HSQC spectrum of the M3-1 isolate (Figure 6, panel B) is a methine

cross peak with a 1H chemical shift of δ 6.05 ppm that correlates to a 13C atom with a chemical

shift of δ 58.6 ppm. Furthermore, there is also set of cross peaks that indicate a pair of

inequivalent methylene resonances with 1H shifts of δ 3.56/4.65 ppm and a correlation to a 13C at

δ 62.9 ppm in the HSQC data. If the bioactivation of 1 is mediated via the epoxidation pathway

the resulting molecule would have two adjacent methine protons with chemical shifts in the

range of δ 5.0–6.0 ppm with corresponding 13C shifts in the range δ 80 ppm (for HCOH) and δ

60 ppm (for HCSG). Alternatively, if the bioactivation of 1 is mediated via the S-oxidation

pathway the resulting molecule would have one methine 1H adjacent to a methylene, again each

with 1H chemical shifts in the range of δ 5.0–6.0 ppm. Unlike the GSH-epoxide addition

product, the 13C chemical shifts of the GSH-S-oxide conjugate would have 13C shifts for both

carbons in the range of δ 50–60 ppm. The chemical shifts observed in the 1H-13C HSQC


at ASPE



Dow

nloaded from


DMD #51839

15

spectrum of M3-1 are more consistent with the structure resulting from the S-oxidation pathway,

and are also closely aligned with the NMR data reported for the GSH conjugate of L-766,112 S-

oxide (Trimble et al., 1996). Furthermore, the TOCSY data set (Figure 6, panel A) for the M3-1

isolate contains cross peaks between the δ 6.05 ppm and the δ 3.56/4.65 ppm resonances

indicating a direct coupling between all of these protons. Based on these data the M3-1 isolate is

tentatively assigned as the thiazole-S-oxide with a GSH molecule attached at the C3 position of

the thiazole ring. This structure strongly infers that the route of formation of M3-1 is through a

biotransformation step involving S-oxidation .

The potential for S-oxide formation was then specifically examined in NADPH-

supplemented HLM incubations of 1 (10 μM) in the absence and presence of GSH (10 mM)

(Figure 7, panels A and B, respectively). Two monohydroxylated metabolites (MH+ = 498.1473)

designated as M4-1 (minor) and M5-1 (major) were observed in HLM incubations in the absence

of GSH. The CID spectrum of the minor metabolite M4-1 is shown in Supplemental Figure 2.

The detection of a fragment ion at m/z 334.1420 (addition of oxygen to m/z 318.1473) indicated

that M4-1 was derived from a monohydroxylation on the spirocyclic-azetidine ring. In contrast,

the CID spectrum of the major metabolite M5-1 (Figure 8) contained fragment ions which

suggested that the imidazo[2,1-b]thiazole motif was the site of oxidation. Furthermore the

fragment ion at m/z 452.1401 (MH+ - SO) implied that M5-1 was the S-oxide metabolite of 1.

Addition of GSH to the NADPH-supplemented HLM incubation of 1 significantly diminished

the levels of M5-1 (Figure 7, panel B), which is consistent with the adduction of M5-1 to GSH

resulting in the formation of M3-1.


at ASPE



Dow

nloaded from


DMD #51839

16

In Vitro Metabolism of 2-Methylmidazo[2,1-b]thiazole Derivative 2 in HLM. Figure 2

(panel B) depicts an extracted ion chromatogram of an incubation mixture comprising of

NADPH, GSH and 2 conducted at 37 °C for 60 minutes. A total of five metabolites were

observed in a NADPH-dependent fashion. The CID spectrum of 2 (tR = 19.47 minutes, MH+ =

496.1680), and proposed structures of diagnostic fragment ions (m/z 318.1473, 276.1159 and

192.0317) is shown in Figure 9, panel A. Metabolites M1-2 (tR = 19.0 minutes) and M2-2 (tR =

21.46 minutes) possessed MH+ at 512.1630, suggesting that they were isomers derived from a

monohydroxylation in 2. The CID spectra of both metabolites (Supplemental Figures 3 and 4)

revealed fragment ions at m/z 494.1510 (loss of a water molecule) and m/z 334.1420 (addition of

oxygen to m/z 318.1473) indicative of monohydroxylation on the spirocyclic-azetidine ring.

Postulated structures of M1-2 and M2-2 are depicted in Supplemental Figures 3 and 4,

respectively. Considering that M2-2 elutes after 2 under the reversed phase HPLC conditions,

we speculate that M2-2 is an N-oxide metabolite of 2 by inference to related tertiary basic amine

drugs such as loperamide that undergo N-oxidation (Kalgutkar and Nguyen, 2004).

Metabolite M3-2 (tR = 18.0 minutes) displayed a MH+ at 530.1735. The addition of 34.0055

Da to the MH+ of 2 indicated that M3-2 was the corresponding diol metabolite, derived from

hydrolysis of an intermediate thiazole ring epoxide. The fragment ions at m/z 318.1473,

213.0322 and 192.0317 were consistent with the proposed structure (Figure 9, panel B).

Metabolite M4-2 (tR = 17.1 minutes) also possessed a MH+ at 512.1630, consistent with a

monohydroxylation in 2. In contrast with M1-2 and M2-2, metabolite M4-2 appeared to be

derived from an oxidation on the 2-methylimidazo[2,1b]thiazole ring system as indicated by the


at ASPE



Dow

nloaded from


DMD #51839

17

fragment ion m/z 318.1473 in the CID spectrum of M4-2. A proposed structure for M4-2 that is

compatible with the observed mass spectrum is shown in Supplemental Figure 5.

Metabolite M5-2 (tR = 15.89 minutes) possessed a MH+ at 787.2747, and demonstrated a

fragment ion at m/z 658.2308, which was consistent with the loss of a glutamic acid component

(129 Da) of GSH (Figure 10, panel A). This observation suggested that M5-2 was a GSH

conjugate of 2. The presence of the fragment ion at m/z 318.1474 implied that the site of

attachment of GSH was on the 2-methylimidazo[2,1-b]thiazole ring system. Furthermore, the

fragment ion at m/z 514.1776 was assigned as a cleavage adjacent to the cysteinyl thioether

moiety with charge retention on the imidazole residue. The occurrence of the fragment ion at

m/z 514.1776 is consistent with the presence of an aromatic thioether motif in M5-2 (Baillie and

Davis, 1993). A proposed structure for M5-2 that is consistent with its mass spectrum is shown

in Figure 10, panel A. Replacement of GSH with N-acetylcysteine as a trapping agent in

NADPH-supplemented HLM incubations of 2 led to the formation of M6-2 (tR = 18.2 minutes,

MH+ = 643.2212), an analogous conjugate of M5-2. The CID spectrum of M6-2 and

interpretation of key fragment ions is depicted in Figure 10, panel B. As such, elucidation of the

structure of the GSH conjugate (M5-2) was largely aided by the isolation of the thiol adduct from

large-scale HLM incubations of 2 in the presence of NADPH and GSH, and subsequent

characterization by NMR spectroscopy.

NMR Characteristics of GSH Conjugate M5-2. The 1H spectrum M5-2 contains several

critical differences from that of a similarly acquired spectrum of 2. Most notably the change in

both the 1H and 13C chemical shifts of the methyl of the imidazothiazole (δ = 2.37 ppm for 1H

and δ = 13.1 ppm for 2 and δ = 2.15 ppm for 1H and δ = 26.7 ppm for M5-2), and the appearance


at ASPE



Dow

nloaded from


DMD #51839

18

of a new methylene resonance in the 1H spectrum of M5-2 at δ = 4.93 ppm (Figure 11, top

panel). In the HSQC data set this new 1H resonance correlates with a 13C resonance at δ = 55.1

ppm (data not shown). Additionally, the 1H resonances for GSH as well as those from the

remaining 1-{2-[2-chloro-4-(2H-1,2,3-triazol-2-yl)benzyl]-2,7-diazaspiro[3.5]non-7-yl}-2-(2-

methylimidazo[2,1-b][1,3]thiazol-6-yl)ethanone can be rationalized. The critical data for the

structural determination of M5-2 is the 1H-13C HMBC data (Figure 12). In this data set there are

correlations from both the new methylene resonance (δ = 4.93 ppm) and the methyl resonance (δ

= 2.37 ppm) to a 13C with a chemical shift of δ = 201.7 ppm. A 13C chemical shift in the range of

190-220 ppm is unique and indicates a carbonyl with no adjacent hetero-atom, either a ketone or

aldehyde. A structure that satisfies all these data is shown in Figure 10 (panel A), the

biosynthetic pathway of which will be revealed in the discussion section.

18O Incorporation Studies. The source of the carbonyl oxygen in GSH conjugate M5-2 was

investigated in NADPH- and GSH-supplemented HLM incubations of 2, conducted in H218O.

As expected, the diol metabolite M3-2 incorporated 18O (MH+ = 532.1778), consistent with

epoxide ring opening by H218O (Supplemental Figure 6). MS/MS analysis of the molecular ions

of 16O- and 18O-incorporated M3-2 (MH+ = 530.1725 and MH+ = 532.1778, respectively) clearly

indicated the 2-atomic mass unit shift in the parent ion and in the acylium ion fragment m/z

213.0322 (16O M3-2) to m/z 215.0369 (18O M3-2) (compare Figure 13 with Figure 9, panel B).

In contrast, 18O incorporation was not observed in the GSH adduct M5-2. This observation

suggested that the source of the carbonyl oxygen in M5-2 is derived from molecular oxygen.

Trapping Studies with Potassium Cyanide. Attempts to trap the putative iminium ion

intermediate involved in the formation of GSH conjugate M5-2 (see discussion section and


at ASPE



Dow

nloaded from


DMD #51839

19

Figure 16 on the proposed mechanism for the formation of M5-2) were initiated in NADPH-

supplemented HLM incubations of 2 in the presence of potassium cyanide (5 mM). Figure 14

indicates the CID spectrum of an adduct M7-2 (tR = 18.4 minutes, MH+ = 539.1739) formed in

these incubations in a NADPH- and cyanide-dependent fashion. The proposed structure of the

cyano adduct M7-2 and corresponding fragment ions that led to the structural elucidation are also

shown in Figure 14. A mechanism for the formation of M7-2 is shown in Figure 16 (discussion

section).

In Vitro Bioactivation of 2-Methylimidazo[2,1-b]thiadiazole Derivative 3 in Human Liver

Microsomes. LC-MS/MS analysis of an incubation mixture of 3 in NADPH- and GSH-

supplemented HLM did not reveal the presence of any GSH conjugates of 3, suggesting that 3 is

devoid of the bioactivation liability associated with 1 and 2.

Comparison of the In Vitro Ghrelin Receptor Pharmacology and HLM Stability of

Compounds 1–3. Potency against the ghrelin receptor was determined in the previously

described GHSR-1a binding assay using [125I]-ghrelin (Kung et al., 2012). As shown in Figure

1, the IC50 values for 1, 2 and 3 were 8.0 nM, 4.7 nM, and 2.6 nM, respectively. The microsomal

stability of compounds 1–3 (final concentration = 1 μM) was assessed by monitoring substrate

consumption after incubation with HLM in the presence of NADPH co-factor for 30 min at 37

°C. Microsomal half-lives, reflecting depletion of test compounds, were scaled to the

corresponding intrinsic clearance values using the well-stirred model (Obach, 1999). Under

these experimental conditions, the half-lives of 1, 2 and 3 in HLM were 19.1 minutes, 28.6

minutes, and 38.8 minutes, which translated into intrinsic clearance values (ml/min/kg) of 48, 32

and 24, respectively.


at ASPE



Dow

nloaded from


DMD #51839

20

Discussion

The otherwise attractive pharmacological and pharmacokinetic attributes of ghrelin inverse

agonists 1 and 2 were offset by the presence of the imidazo[2,1-b]thiazole bicycle as part of the

core structure, especially in light of previous studies by Trimble et al. (1997) who demonstrated

the microsomal S-oxidation of the thiazole portion (within the bicycle) to an electrophilic

sulfoxide metabolite that underwent a 1,4-Michael addition reaction with GSH. In addition,

evidence linking thiazole ring bioactivation with preclinical (or clinical) toxicity has been

presented for several thiazole-based xenobiotics and drugs (Kalgutkar et al., 2005). For

example, the clinical hepatotoxicity noted with the nonsteroidal anti-inflammatory drug

sudoxicam (supplementary Figure S-6) has been attributed to a metabolic process involving

thiazole ring scission to a toxic thiourea metabolite (Obach et al., 2008). The structurally-related

anti-inflammatory agent (and marketed drug) meloxicam does not possess the hepatotoxic

liability associated with sudoxicam. Although introduction of a methyl group at the C5 position

of the thiazole ring in meloxicam is the only structural difference, the change dramatically alters

the metabolic profile such that oxidation of the C5 methyl group to the alcohol metabolite (see

Supplemental Figure 7) constitutes the principal metabolic fate of meloxicam in humans with

virtually no detectable thiazole ring opening (Chesné et al., 1998; Obach et al., 2008).

Taking into consideration the above mentioned structure-toxicity relationship and the

growing evidence linking immune-mediated drug toxicity with reactive metabolite formation

(Stepan et al., 2011; Park et al., 2011; Thompson et al., 2012; Sakatis et al., 2012), we decided to

examine the bioactivation potential of 1 and 2 in HLM. Incubation of 1 and 2 in NADPH- and

GSH-supplemented HLM led to the formation of reactive species that were trapped with GSH.


at ASPE



Dow

nloaded from


DMD #51839

21

Metabolism (including bioactivation) of both 1 and 2 was blocked when the HLM incubations

were conducted in the presence of ketoconazole (a selective CYP34 inhibitor), which implicated

a role for CYP3A4 in the metabolism of the two imidazo[2,1-b]thiazole derivatives. LC-MS/MS

characterization of the GSH conjugates suggested that the site of bioactivation in both

compounds was on the imidazo[2,1-b]thiazole bicycle. In the case of the imidazo[2,1-b]thiazole

derivative 1, the formation of the diol metabolite M1-1 was consistent with a bioactivation

pathway involving thiazole ring oxidation by CYP enzyme(s) at the C2-C3 position to yield the

electrophilic epoxide intermediate 12, which is hydrolyzed to the diol M1-1 or can potentially

react with GSH to yield the sulfydryl conjugate 13 (Figure 15, pathway A). An alternate

pathway that leads to M3-1 involves CYP-mediated oxidation on the thiazole sulfur to the S-

oxide metabolite M5-1, followed by the Michael addition of GSH to yield adduct M3-1 (see

Figure 15, pathway B) as outlined previously with the imidazo[2,1-b]thiazole analog L-766,112

(Trimble et al., 1997). NMR analysis of the crude M3-1 isolate indicated that the GSH adduct

was derived from addition of the thiol nucleophile to the electrophilic S-oxide M5-1 and not

from the ring opening of epoxide 12. The hypothesis is further strengthened by the observation

that addition of GSH to NADPH-supplemented HLM incubations of 1 significantly diminished

the levels of M5-1 formed in the incubations. Reason(s) for the lack of formation of the GSH

adduct 13 shown in Figure 15, pathway A are not apparent from this analysis. It is possible that

hydrolytic ring opening of 12 proceeds at a faster rate than the corresponding reaction with GSH.

Apart from the mono-GSH adduct M3-1, we also noted the formation of a novel bis-GSH

conjugate M2-1 with two molecules of GSH attached to the imidazo[2,1-b]thiazole framework.

A plausible mechanism outlining the formation of M2-1 is presented in Figure 15 (pathway C),


at ASPE



Dow

nloaded from


DMD #51839

22

and involves the initial addition of a GSH molecule to the putative S-oxide metabolite of 1 (i.e.,

M5-1) to generate M3-1 followed by addition of a second molecule of GSH across the electron

deficient C5 position on the imidazole ring system in M3-1, which results in imidazo[2,1-

b]thiazole-S-oxide ring scission. The outlined mechanism is certainly not beyond the realm of

possibilities considering the well known chemical (and glutathione transferase-mediated)

displacement reaction of GSH with electron deficient heteroaromatic rings (e.g., pyridine,

pyrimidine, etc) that contain alkylsulfoxide and/or alkylsulfone as leaving groups (Clapp, 1956;

Colucci and Buyske, 1965; Conroy et al., 1984; Graham et al., 1989; Yang et al., 2012). A

literature example that bears much commonality to the present situation is evident with the

proton pump inhibitor pantoprazole wherein, the benzimidazole-2-sulfoxide motif undergoes

nucleophilic attack by GSH at the electron-deficient C2 position on the benzimidazole ring

(Zhong et al., 2005). Certainly, one can view M5-1 (in Figure 15, pathway C) as an electron-

deficient heterocyle (imidazole ring) with an excellent leaving group (alkylsulfoxide) at the C5

position, which is poised for nucleophilic displacement by GSH.

In the case of 2, LC-MS/MS data on the GSH and N-acetylcysteinyl conjugates M5-2 and

M6-2, respectively, were also consistent with thiazole ring bioactivation. From a SAR

standpoint, the C-2 methyl group on the imidazo[2,1-b]thiazole ring in 2 did not prevent thiazole

ring bioactivation to reactive metabolite(s), which contrasts the structure-toxicity relationship

noted for sudoxicam and meloxicam. From a mechanistic perspective, we speculate that the rate-

limiting step in the formation of M5-2 and M6-2 proceeds via a CYP-mediated oxidation on the

thiazole ring in 2 to generate epoxide 14, which upon hydrolysis would yield the diol metabolite

M3-2 or undergo ring scission to a iminium species 15 (Figure 16). A two-electron reduction of


at ASPE



Dow

nloaded from


DMD #51839

23

the iminium bond in 15, possibly mediated by GSH, would lead to the 2-mercaptoimidazole 16,

which may be trapped by GSH or N-acetylcysteine via a developing ring-opened isothiocyanate

17 to afford sulfydryl conjugates M5-2 and M6-2, respectively. Overall the mechanistic

possibility is compatible with the results of the 18O labeling studies, wherein, lack of

incorporation of 18O in M5-2 indicates that the source of the carbonyl oxygen in M5-2 is from

molecular oxygen (i.e., from a CYP-mediated oxidation).

To examine whether iminium species 15 is indeed formed as an intermediate in the pathway

leading to M5-2, 2 was incubated in HLM in the presence of NADPH and excess potassium

cyanide, which is typically used to trap electrophilic iminium ions generated via a two-electron

oxidation of amines (Rose and Castagnoli, 1983; Argoti et al., 2005). The accurate mass of the

metabolite M7-2 detected in these incubations was consistent with the addition of cyanide across

15, which provided additional support for our overall mechanistic hypothesis. One can

rationalize the formation of M7-2 to occur via the initial addition of cyanide across the iminium

bond in 15 to yield the cyano adduct 18, which upon cyclization would lead to M7-2 (see Figure

16). With reference to the step involving the reduction of the iminium bond in 15 (to yield 2-

mercaptoimidazole 16), we invoked the involvement of GSH as a reducing agent since, two-

electron reduction of imines by GSH is a well precedented reaction. For example, N-acetyl-p-

benzoquinone imine, the oxidation product of acetaminophen, can react with GSH in a 1,4-

Michael fashion to yield a stable GSH adduct or undergo a two-electron reduction to the parent

drug in the presence of the thiol (Rosen et al., 1984; Potter and Hinson, 1986). In the present

situation, we anticipate that the thioaminal conjugation product obtained through addition of

GSH across the imine double bond in 15 will be highly unstable and spontaneously decompose


at ASPE



Dow

nloaded from


DMD #51839

24

back to 15. Clearly, a more rigorous experimental analysis of these individual mechanistic steps

will be needed to fully validate our hypothesis on the oxidative metabolism/bioactivation of

these unique bicyclo-heterocylic compounds. Towards this end, the imidazo[2,1-b]thiazole

intermediates used in the synthesis of the title compounds 1 and 2 (see supplementary section)

could provide much utility in testing certain mechanistic hypothesis (e.g., the use of deuterium

exchange to distinguish S-oxidation from mono-hydroxylation).

Overall, given this information, it appeared reasonable to attempt to eliminate bioactivation

liability in 1 and 2 via small structural modifications of the thiazole ring that would preserve the

physicochemical properties (e.g., molecular weight, lipophilicity) largely responsible for

pharmacologic potency and HLM stability. Consequently, the 2-methylimidazo[2,1-

b][1,3,4]thiadiazole analog 3 was synthesized to prevent thiazole ring epoxidation and/or

addition of GSH to an S-oxide metabolite. An identical medicinal chemistry strategy had been

successfully utilized to abrogate thiazole ring bioactivation with nonpeptidyl thrombopoietin

receptor agonists containing the 2-aminothiazole motif (Kalgutkar et al., 2007). As such, the

absence of the GSH conjugate formation in NADPH- and GSH-supplemented HLM incubations

with 3 indicates that the thiadiazole derivative is devoid of the reactive metabolite liability

associated with 1 and 2, and demonstrates the success of this medicinal chemistry tactic.

Compound 3 also retained the potent in vitro ghrelin inverse agonism and HLM stability

characteristics discerned with lead compounds 1 and 2. In summary, we have demonstrated the

human liver microsomal bioactivation of the imidazo[2,1-b]thiazole functionality present as part

of the core structure in novel ghrelin inverse agonists. Characterization of the GSH adducts

using LC-MS/MS and NMR techniques provided indirect information on the structures of the


at ASPE



Dow

nloaded from


DMD #51839

25

reactive species, thereby providing insight into the bioactivation mechanism. The information

gained from these studies was used to direct medicinal chemistry efforts in the design of a

compound with the attractive pharmacology and disposition characteristics observed with the

prototypes while avoiding potential safety concerns associated with the bioactivation of the

imidazo[2,1-b]thiazole functionality. Further lead optimization efforts on 3, which culminated in

the discovery of the clinical candidate will be reported in due course.


at ASPE



Dow

nloaded from


DMD #51839

26

Authorship Contributions

Participated in research design: Kalgutkar, Goosen, Lapham, Ryder, Walker, and Eng

Conducted in vitro experiments: Ryder, Walker, and Eng

Contributed new reagents or analytic tools: Orr and Cabral

Performed data analysis: Kalgutkar, Ryder, Walker, Orr, Eng, and Goosen

Wrote or contributed to the writing of the manuscript: Kalgutkar, Ryder, Walker, Orr,

and Eng


at ASPE



Dow

nloaded from


DMD #51839

27

References

Argoti D, Liang L, Conteh A, Chen L, Bershas D, Yu CP, Vouros P, and Yang E. (2005)

Cyanide trapping of iminium ion reactive intermediates followed by detection and structure

identification using liquid chromatography-tandem mass spectrometry (LC-MS/MS) Chem

Res Toxicol 18:1537-1544.

Baillie TA and Davis MR (1993) Mass spectrometry in the analysis of glutathione conjugates.

Biol Mass Spectrom 22:319-325.

Cardona Cano S, Merkestein M, Skibicka KP, Dickson SL, and Adan RA (2012) Role of ghrelin

in the pathophysiology of eating disorders: implications for pharmacotherapy. CNS Drugs

26:281-296.

Chatfield DH and Hunter WH (1973) The metabolism of acetamidothiazoles in the rat: 2-

acetamino-, 2-acetamido-4-methyl and a acetamino-4-phenylthiazole. Biochem J 134:869-

878.

Chesné C, Guyomard C, Guillouzo A, Schmid J, Ludwig E, and Sauter T (1998) Metabolism of

meloxicam in human liver involves cytochrome P4502C9 and 3A4. Xenobiotica 28:1-13.

Clap JW (1956) A new metabolic pathway for a sulfonamide group. J Biol Chem 233:207-214.

Colucci DF and Buyske DA (1965) The biotransformation of a sulfonamide to a mercaptan and

to a mercapturic acid and glucuronide conjugate. Biochem Pharmacol 14:457-466.

Conroy CW, Schwam H, and Maren TH (1984) The nonenzymatic displacement of the

sulfamoyl group from different classes of aromatic compounds by glutathione and cysteine.

Drug Metab Dispos 12:614-618.


at ASPE



Dow

nloaded from


DMD #51839

28

Costantino L and Barlocco D (2009) Ghrelin receptor modulators and their therapeutic potential.

Future Med Chem 1:157-177.

Dansette PM, Bertho G, and Mansuy D (2005) First evidence that cytochrome P450 may

catalyze both S-oxidation and epoxidation of thiophene derivatives. Biochem Biophys Res

Commun 338:450-455.

Dezaki K, Sone H, and Yada T (2008) Ghrelin is a physiological regulator of insulin release in

pancreatic islets and glucose homeostasis. Pharmacol Ther 118:239-249.

Esler WP, Rudolph J, Claus TH, Tang W, Barucci N, Brown SE, Bullock W, Daly M, Decarr L,

Li Y, Milardo L, Molstad D, Zhu J, Gardell SJ, Livingston JN, and Sweet LJ (2007) Small-

molecule ghrelin receptor antagonists improve glucose tolerance, suppress appetite, and

promote weight loss. Endocrinology 148:5175-5185.

Graham SL, Shepard KL, Anderson PS, Baldwin JJ, Best DB, Christy ME, Freedman MB,

Gautheron P, Habecker CN, Hoffman JM, lyle PA, Michelson SR, Ponticello GS, Robb CM,

Schwam H, Smith AM, Smith RL, Sondey JM, Strohmaier KM, Sugrue MF, and Varga SL

(1989) Topically active carbonic anhydrase inhibitors. 2. Benzo[b]thiophenesulfonamide

derivatives with ocular hypotensive activity. J Med Chem 32:2548-2554.

Hajovsky H, Hu G, Koen Y, Sarma D, Cui W, Moore DS, Staudinger JL, and Hanzlik RP (2012)

Metabolism and toxicity of thioacetamide and thioacetamide-S-oxide in rat hepatocytes.

Chem Res Toxicol 25:1955-1963.

Heppner KM, Müller TD, Tong J, and Tschöp MH (2012) Ghrelin in the control of energy, lipid,

and glucose metabolism. Methods Enzymol 514:249-260.


at ASPE



Dow

nloaded from


DMD #51839

29

Inui A, Asakawa A, Bowers CY, Mantovani G, Laviano A, Meguid MM, and Fujiyama M

(2004) Ghrelin, appetite, and gastric motility: the emerging role of the stomach as an

endocrine organ. FASEB J 18:439-456.

Kalgutkar AS and Nguyen HT (2004) Identification of an N-methyl-4-phenylpyridinium-like

metabolite of the antidiarrheal agent loperamide in human liver microsomes: Underlying

reason(s) for the lack of neurotoxicity despite the bioactivation event. Drug Metab Dispos

32:943-952.

Kalgutkar AS, Gardner I, Obach RS, Shaffer CL, Callegari E, Henne KR, Mutlib AE, Dalvie

DK, Lee JS, Nakai Y, O’Donnell JP, Boer J, and Harriman SP (2005) A comprehensive

listing of bioactivation pathways of organic functional groups. Curr Drug Metab 6:161-225.

Kalgutkar AS, Driscoll J, Zhao SX, Walker GS, Shepard RM, Soglia JR, Atherton J, Yu L,

Mutlib AE, Munchhof MJ, Reiter LA, Jones CS, Doty JL, Trevena KA, Shaffer CL, and

Ripp SL (2007) A rational chemical intervention strategy to circumvent bioactivation

liabilities associated with a nonpeptidyl thrombopoietin receptor agonist containing a 2-

amino-4-arylthiazole motif. Chem Res Toxicol 20:1954-1965.

Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, and Kangawa K (1999) Ghrelin is a

growth hormone-releasing acylated peptide from stomach. Nature 402:656-660.

Kung DW, Coffey SB, Jones RM, Cabral S, Jiao W, Fichtner M, Carpino PA, Rose CR, Hank

RF, Lopaze MG, Swartz R, Chen HT, Hendsch Z, Posner B, Wielis CF, Manning B, Dubins

J, Stock IA, Varma S, Campbell M, DeBartola D, Kosa-Maines R, Steyn SJ, and McClure


at ASPE



Dow

nloaded from


DMD #51839

30

KF (2012) Identification of spirocyclic piperazine-azetidine inverse agonists. Bioorg Med

Chem Lett 22:4281-4287.

Mansuy D and Dansette PM (2011) Sulfenic acids as reactive intermediates in xenobiotic

metabolism. Arch Biochem Biophys 507:174-185.

Mizutani T, Yoshida K, and Kawazoe S (1993) Possible role of thioformamide as a proximate

toxicant in the nephrotoxicity of thiabendazole and related thiazoles in glutathione-depleted

mice: structure-toxicity and metabolic studies. Chem Res Toxicol 6:174-179.

Nakazato M, Murakami N, Date Y, Kojima M, Matsuo H, Kangawa K, and Matsukura S (2001)

A role for ghrelin in the central regulation of feeding. Nature 409:194-198.

Obach RS (1999) An examination of in vitro half-life approach and nonspecific binding to

microsomes. Drug Metab Dispos 27:1350-1359.

Obach RS, Kalgutkar AS, Ryder TF, and Walker GS (2008) In vitro metabolism and covalent

binding of enol-carboxamide derivatives and anti-inflammatory agents sudoxicam and

meloxicam: insights into the hepatotoxicity of sudoxicam. Chem Res Toxicol 21:1890-1899.

Park BK, Boobis A, Clarke S, Goldring CE, Jones D, Kenna JG, Lambert C, Laverty HG,

Naisbitt DJ, Nelson S, Nicoll-Griffith DA, Obach RS, Routledge P, Smith DA, Tweedie DJ,

Vermeulen N, Williams DP, Wilson ID, and Baillie TA (2011) Managing the challenge of

chemically reactive metabolites in drug development. Nat Rev Drug Discov 10:292-306.

Potter DW and Hinson JA (1986) Reactions of N-acetyl-p-benzoquinone imine with reduced

glutathione, acetaminophen, and NADPH. Mol Pharmacol 30:33-41.

Rose J and Castagnoli N, Jr. (1983) The metabolism of tertiary amines. Med Res Rev 3:73-88.


at ASPE



Dow

nloaded from


DMD #51839

31

Rosen GM, Rauckman EJ, Ellington SP, Dahlin DC, Christie JL, and Nelson SD (1984)

Reduction and glutathione conjugation reactions of N-acetyl-p-benzoquinone imine and two

dimethylated analogues. Mol Pharmacol 25:151-157.

Roy P, Leblanc Y, Ball RG, Brideau C, Chan CC, Chauret N, Cromlish W, Ethier D, Gauthier

JY, Gordon R, Greig G, Guay J, Kargman S, Lau CK, O’Neill G, Silva J, Thérien M, van

Staden C, Wong E, Xu L, and Prasit P (1997) A new series of selective COX-2 inhibitors:

5,6-diarylthiazolo[3,2-b][1,2,4]-triazoles. Bioorg Med Chem Lett 7:57-62.

Rudolph J, Esler WP, O'connor S, Coish PD, Wickens PL, Brands M, Bierer DE, Bloomquist

BT, Bondar G, Chen L, Chuang CY, Claus TH, Fathi Z, Fu W, Khire UR, Kristie JA, Liu

XG, Lowe DB, McClure AC, Michels M, Ortiz AA, Ramsden PD, Schoenleber RW,

Shelekhin TE, Vakalopoulos A, Tang W, Wang L, Yi L, Gardell SJ, Livingston JN, Sweet

LJ, Bullock WH (2007) Quinazolinone derivatives as orally available ghrelin receptor

antagonists for the treatment of diabetes and obesity. J Med Chem 50:5202-5216.

Sakatis MZ, Reese MJ, Harrell AW, Taylor MA, Baines IA, Chen L, Bloomer JC, Yang EY,

Ellens HM, Ambroso JL, Lovatt CA, Ayrton AD, and Clarke SE (2012) Preclinical strategy

to reduce clinical hepatotoxicity using in vitro bioactivation data for > 200 compounds. Chem

Res Toxicol 25:2067-2082.

Serby MD, Zhao H, Szczepankiewicz BG, Kosogof C, Xin Z, Liu B, Liu M, Nelson LT,

Kaszubska W, Falls HD, Schaefer V, Bush EN, Shapiro R, Droz BA, Knourek-Segel VE,

Fey TA, Brune ME, Beno DW, Turner TM, Collins CA, Jacobson PB, Sham HL and Liu G

(2006) 2,4-diaminopyrimidine derivatives as potent growth hormone secretagogue receptor

antagonists. J Med Chem 49:2568-2578.


at ASPE



Dow

nloaded from


DMD #51839

32

Stepan AF, Walker DP, Bauman J, Price DA, Baillie TA, Kalgutkar AS, and Aleo MD (2011)

Structural alert/reactive metabolite concept as applied in medicinal chemistry to mitigate the

risk of idiosyncratic drug toxicity: a perspective based on the critical examination of trends in

the top 200 drugs marketed in the United States. Chem Res Toxicol 24:1345-1410.

Soares JB, Roncon-Albuquerque R Jr, and Leite-Moreira A (2008) Ghrelin and ghrelin receptor

inhibitors: agents for the treatment of obesity. Expert Opin Ther Targets 12:1177-1189.

Thérien M, Brideau C, Chan CC, Cromlish WA, Gauthier JY, Gordon R, Greig G, Kargman S,

Lau CK, Leblanc Y, Li CS, O’Neill GP, Riendeau D, Roy P, Wang Z, Xu L, and Prasit P

(1997) Synthesis and biological evaluation of 5,6-diarylimidazo[2.1-b]thiazole as selective

COX-2 inhibitors. Bioorg Med Chem Lett 7:47-52.

Thompson RA, Isin EM, Li Y, Weidolf L, Page K, Wilson I, Swallow S, Middleton B, Stahl S,

Foster AJ, Dolgos H, Weaver R, and Kenna JG (2012) In vitro approach to assess the

potential for risk of idiosyncratic adverse reactions caused by candidate drugs. Chem Res

Toxicol 25:1616-1632.

Trimble LA, Chauret N, Silva JM, Nicoll-Griffith DA, Li CS, and Yergey JA (1997)

Characterization of the in vitro oxidative metabolites of the COX-2 selective inhibitor L-

766,112. Bioorg Med Chem Lett 7:53-56.

Wren AM, Seal LJ, Cohen MA, Brynes AE, Frost GS, Murphy KG, Dhillo WS, Ghatei MA, and

Bloom SR (2001) Ghrelin enhances appetite and increases food intake in humans. J Clin

Endocrinol Metab 86:5992.


at ASPE



Dow

nloaded from


DMD #51839

33

Yang Y, Qiu F, Josephs JL, Griffith Humphreys W, and Shu Y-Z (2012) Bioactivation of 2-

(alkylthio)-1,3,4-thiadiazoles and 2-(alkylthio)-1,3-benzothiazoles. Chem Res Toxicol

25:2770-2779.

Yabuki M, Shimakura IM, Kanamura H, Iba K, and Nakatasuka O (1997) Metabolism of 4-[1-

(2-fluoro-4-biphenyl)-ethyl]-2-methylaminothiazole (SM-8849), in rats. Eur J Drug Metab

Dispos Pharmacokinet 22:25-33.

Yoon WH, Yoo JK, Lee JW, Shim CD, and Lee MG (1998) Species differences in

pharmacokinetics of a hepatoprotective agent, YH-349, and its metabolites M4, M5 and M7,

after intravenous and oral administration to rats, rabbits and dogs. Drug Metab Dispos

26:152-163.

Ziegler-Skylakakis K, Nill S, Pan JF, and Andrae U (1998) S-oxygenation of thiourea results in

the formation of genotoxic products. Environ Mol Mutagen 31:362-373.

Zhong D, Xie Z, and Chen X (2005) Metabolism of pantoprazole involving conjugation with

glutathione in rats. J Pharm Pharmacol 57:341-349.


at ASPE



Dow

nloaded from


DMD #51839

34

Figure Legends

FIG. 1. Structures of ghrelin receptor inverse agonists 1–3 from the spirocyclic piperidine-

azetidine series, and oxidative bioactivation of the imidazo[2,1-b]thiazole motif in L-766,112 to

an electrophilic S-oxide species.

FIG. 2. Extracted ion chromatograms of incubation mixtures of 1 (panel A) and 2 (panel B) in

NADPH- and GSH-supplemented HLM conducted at 37 °C for 60 minutes.

FIG. 3. CID spectra of 1 (panel A) and M1-1 (panel B).

FIG. 4. CID spectrum of bis-GSH conjugate M2-1. Structures of diagnostic fragment ions (and

corresponding theoretical exact masses) that allow metabolite identification are depicted. The

assigned regiochemistry is arbitrary.

FIG. 5. CID spectrum of GSH conjugate M3-1. Structures of diagnostic fragment ions (and

corresponding theoretical exact masses) that allow an insight into the structure are depicted.

FIG. 6. Abbreviated 1H-1H TOCSY (panel A) and 1H-13C HSQC spectrum (panel B) of M3-1.

Red cross peaks in panel B indicate methine, whereas the blue cross peaks indicate the

methylene resonances.

FIG. 7. Extracted ion chromatograms of monohydroxylated metabolites (MH+ = 498.1473) of 1

in NADPH-supplemented HLM in the absence (panel A) and presence (panel B) of GSH.

FIG. 8. CID spectrum of M5-1 derived from an NADPH-supplemented HLM incubation of 1.

FIG. 9. CID spectra of 2 (panel A) and M3-2 (panel B).


at ASPE



Dow

nloaded from


DMD #51839

35

FIG. 10. CID spectra of the GSH conjugate M5-2 (panel A) and N-acetylcysteine conjugate M6-

2 (panel B).

FIG. 11. Full 1H spectrum of M5-2 and 2.

FIG. 12. Abbreviated 1H-13C HMBC spectrum of M5-2 (panels A and B represent specific 1H

correlations highlighted in the text).

FIG. 13. MS/MS data reflecting 18O incorporation in the diol metabolite M3-2 generated in an

NADPH-and H218O-supplemented HLM incubation of 2.

FIG. 14. CID spectrum of the cyanide conjugate M7-2 derived from NADPH- and potassium

cyanide-supplemented HLM incubations of 2.

FIG. 15. Proposed bioactivation pathways for imidazo[2,1-b]thiazole derivative 1.

FIG. 16. Proposed bioactivation pathway for 2-methylimidazo[2,1-b]thiazole derivative 2.


at ASPE



Dow

nloaded from


Figure 1

R N

O

N

ClNN

N

1

S

N N

S

N N

H3C2

NS

N N

3H3C

Human GHS R1a IC50 = 8.0 nM Human GHS R1a IC50 = 4.7 nM Human GHS R1a IC50 = 2.6 nMHLM CLint = 24 ml/min/kgHLM CLint = 48 ml/min/kg HLM CLint = 32 ml/min/kg

R = R = R =

N

NS

SO2CH3

L-766,112

CYP N

NS

SO2CH3

O

GSH N

NS

SO2CH3

O

GSN

S

N

N

L-768,277 (devoid of bioactivation)

SO2CH3

S-Oxide

M.W. = 482, clogD = 1.48 M.W. = 497, clogD = 0.81M.W. = 496, clogD = 1.26

This article has not been copyedited and form

atted. The final version m

ay differ from this version.

DM

D Fast Forw

ard. Published on April 22, 2013 as D

OI: 10.1124/dm

d.113.051839 at ASPET Journals on February 22, 2020 dmd.aspetjournals.org Downloaded from


Figure 2

Time (min)

6 8 10 12 14 16 18 20 220

20

40

60

80

100

Re

lati

ve A

bu

nd

anc

e

18.82

16.4717.70

1

M1-1M3-1 M2-1

5 10 15 20 25 30 35 400

20

40

60

80

10019.47

17.1921.46

15.89

2

M2-2

M5-2

M3-2M4-2 M1-2

(A)

(B)


at ASPE



Dow

nloaded from


Figure 3

150 200 250 300 350 400 4500

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bund

ance

318.1476

262.1004192.0318 464.1407301.1210165.0112

150 200 250 300 350 400 450 500m/z

0

10

20

30

40

50

60

70

80

90

100318.1476

192.0318 309.1578

N

NS

NO

N Cl

N NN

MH+ = 516.1579

318.1480

HO

HO

M1-1

(A)

(B)


at ASPE



Dow

nloaded from


Figure 4

300 400 500 600 700 800 900 1000 1100m/z

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bund

ance

967.2753

838.23357

1078.3061650.1351

823.22137949.2654694.1805

318.1468 458.1514 805.2108506.0823414.1077855.09903

N

N

NO

NH Cl

N NN

HS

HS

S

HN

O

NHO

HO

O

NH2

O

OH


at ASPE



Dow

nloaded from


Figure 5

250 300 350 400 450 500 550 600 650 700 750 800

m/z

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bund

ance

676.1872

787.2189318.1476

472.1308532.1342

658.1761359.0471730.1973


at ASPE



Dow

nloaded from


Figure 6

SG15 14

N13

2021

16

17

N18

19

11

O12

1022

2328

27

26

25

24

7

8

N6

5

N4

3

2

S1

Cl29

N30

N31

32

33

N34

O9

3 2b

3

2b

A B

2a

2a


at ASPE



Dow

nloaded from


Figure 7

M5-1

M4-1

M4-1

(A)

5 10 15 20 25 30 35 40

Time (min)

0

10

20

30

40

50

60

70

80

90

1000

10

20

30

40

50

60

70

80

90

100

Re

lati

veA

bu

nd

ance

20.26

18.8720.74

18.83

20.2316.96

(B)

M5-1

Normalized levels to base peak: 1.63 x 107

Normalized levels to base peak: 5.33 x 105


at ASPE



Dow

nloaded from


Figure 8

150 200 250 300 350 400 450 500m/z

0

20

40

60

80

100

Rel

ativ

e A

bund

ance

318.1470

192.0317

278.0949

437.1295210.0326 395.1369138.0003181.0060 452.1401221.0582

372.0309


at ASPE



Dow

nloaded from


Figure 9

150 200 250 300 350 400 450 5000

10

20

30

40

50

60

70

80

90

100 318.1473

276.1159192.0317 478.1560301.1207 342.0561

Rel

ativ

e A

bund

ance

200 250 300 350 400 450 500m/z

0

10

20

30

40

50

60

70

80

90

100 318.1474

350.1371213.0322192.0317 416.1479 510.1446

(A)

(B)


at ASPE



Dow

nloaded from


Figure 10 R

elat

ive

Abu

ndan

ce

200 250 300 350 400 450 500 550 600 650m/z

01020

3040

5060708090

100 514.1792

318.1483 606.0683189.3719 561.5912407.892

250 300 350 400 450 500 550 600 650 700 750 8000

10

20

30

40

50

60

70

80

90

100 658.2308

341.0908 514.1776318.1474 769.2621

470.1331640.2197452.1230395.1015308.0692 695.7148542.1427

(B)

(A)


at ASPE



Dow

nloaded from


Figure 11

1H spectrum of M5-2

1H spectrum of 2

2


at ASPE



Dow

nloaded from


Figure 12

ppm

6.87.0 ppm

200

150

100

50

ppm

4.85.0 ppm

200

150

100

50

ppm

2.02.2 ppm

200

150

100

50

8 4 6

87 82

48

42

45

64

65

A B C

11

O12

N13

20

21

16

17N18

22

23

28

27

26

25

24

Cl29

N30N

31

32

33

N34

1915

14

2N3

N

8

4

7

5

CH36

O

SGR


at ASPE



Dow

nloaded from


Figure 13

150 200 250 300 350 400 450 500m/z

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bund

ance

318.1480

320.1450215.0369 352.1426192.0322 418.1516 512.1449276.1160

294.5069

N NN

Cl


at ASPE



Dow

nloaded from


Figure 14

150 200 250 300 350 400 450 500 550

m/z

0

10

20

30

40

50

60

70

80

90

100

Rel

ativ

e A

bund

ance

318.1473

222.0326

194.0376

357.6116262.7421287.8470

236.7467

N N

N

O

N

ClNN

N

M7-2MH+ = 539.1739

318.1480

SH2N

O


at ASPE



Dow

nloaded from


Figure 15

N

NS

RO

1 (MH+ = 482.1524)

CYPN

NS

ORO

GSH

H2O

N

NS

ROHO

HO

12

M1-1 (MH+ = 516.1579)

N

NS

ROS

HO

13 (MH+ = 805.2312)

HN

ONH

OH

O

O

H2N

OHO

CYP

N

NS

OR

M5-1O

GSH

N

NS

RO

O

S

HN NH

O

O

OHOH2N

OOH

M3-1 (MH+ = 805.2312)(B)

(A)

GSHH

GSHN

N

ROS

HN NH

O

O

OHOH2N

OOH

HS

S

HN

O

NHO

HO

O

NH2

O

OH

M2-1 (MH+ = 1096.3200)

H2O

(C)

N N Cl

N NN

R =

GSH




DM

D Fast Forw


OI: 10.1124/dm



Figure 16

N

NS

RO

2 (MH+ = 496.1680)

CYP N

NS

ORO

14

H3C H3CO

NHN

R

S

OCH3

N N Cl

N NN

R =

H2O

N

NS

ROHO

H3CHO

M3-2

KCN

CN

O

NHN

R

S

OCH3

CN

18

H

H

O

NN

R

SHN

H3CO

H

H

O

NN

R

SH2N

H3CO

M7-2 (MH+ = 539.1739)

2GSH GSSG

O

NHN

R

S

OCH3

15 16

O

NN

R

OCH3

S

HNR2

O

M5-2: R1 = CO(CH2)2CHNH2COOHR2 = NHCH2COOH

RSH

RSH

M6-2: R1 = COCH3; R2 = OH

R1

(MH+ = 530.1735)

(MH+ = 787.2747)(MH+ = 643.2212)

O

HN

R

OCH3

17

NCS

O

HN

R

OCH3

N

RS

SH H

- H2S




DM

D Fast Forw


OI: 10.1124/dm



reactive metabolite trapping studies on imidazo- and 2...

Documents