Novel vanilloid receptor-1 antagonists: 2. Structure-activity relationships of 4-oxopyrimidines leading to the selection of a clinical candidate

Article abstract of DOI:10.1021/jm070190p

A series of novel 4-oxopyrimidine TRPV1 antagonists was evaluated in assays measuring the blockade of capsaicin or acid-induced influx of calcium into CHO cells expressing TRPV1. The investigation of the structure-activity relationships in the heterocyclic A-region revealed the optimum pharmacophoric elements required for activity in this series and resulted in the identification of subnanomolar TRPV1 antagonists. The most potent of these antagonists were thoroughly profiled in pharmacokinetic assays. Optimization of the heterocyclic A-region led to the design and synthesis of 23, a compound that potently blocked multiple modes of TRPV1 activation. Compound 23 was shown to be effective in a rodent "on-target" biochemical challenge model (capsaicin-induced flinch, ED₅₀ = 0.33 mg/kg p.o.) and was antihyperalgesic in a model of inflammatory pain (CFA-induced thermal hyperalgesia, MED = 0.83 mg/kg, p.o.). Based on its in vivo efficacy and pharmacokinetic profile, compound 23 (N-{4-[6-(4-trifluoromethyl-phenyl)-pyrimidin-4-yloxy]-benzothiazol-2-yl} -acetamide; AMG 517) was selected for further evaluation in human clinical trials.

Full text of DOI:10.1021/jm070190p

J. Med. Chem. 2007, 50, 3515-3527
3515
Novel Vanilloid Receptor-1 Antagonists: 2. Structure-Activity Relationships of
4-Oxopyrimidines Leading to the Selection of a Clinical Candidate
Elizabeth M. Doherty,† Christopher Fotsch,† Anthony W. Bannon,‡ Yunxin Bo,† Ning Chen,† Celia Dominguez, James Falsey,†
Narender R. Gavva,‡ Jodie Katon,† Thomas Nixey,† Vassil I. Ognyanov,† Liping Pettus,† Robert M. Rzasa,† Markian Stec,†
Sekhar Surapaneni,§ Rami Tamir,‡ Jiawang Zhu,† James J. S. Treanor,‡ and Mark H. Norman*,†
Department of Chemistry Research and DiscoVery, Department of Neuroscience, and Department of Pharmacokinetics and Drug Metabolism,
Amgen Inc., One Amgen Center DriVe, Thousand Oaks, California 91320-1799
ReceiVed February 19, 2007
A series of novel 4-oxopyrimidine TRPV1 antagonists was evaluated in assays measuring the blockade of
capsaicin or acid-induced influx of calcium into CHO cells expressing TRPV1. The investigation of the
structure-activity relationships in the heterocyclic A-region revealed the optimum pharmacophoric elements
required for activity in this series and resulted in the identification of subnanomolar TRPV1 antagonists.
The most potent of these antagonists were thoroughly profiled in pharmacokinetic assays. Optimization of
the heterocyclic A-region led to the design and synthesis of 23, a compound that potently blocked multiple
modes of TRPV1 activation. Compound 23 was shown to be effective in a rodent “on-target” biochemical
challenge model (capsaicin-induced flinch, ED50 ) 0.33 mg/kg p.o.) and was antihyperalgesic in a model
of inflammatory pain (CFA-induced thermal hyperalgesia, MED ) 0.83 mg/kg, p.o.). Based on its in vivo
efficacy and pharmacokinetic profile, compound 23 (N-{4-[6-(4-trifluoromethyl-phenyl)-pyrimidin-4-yloxy]-
benzothiazol-2-yl}-acetamide; AMG 517) was selected for further evaluation in human clinical trials.
Introduction
compound 1 demonstrated acceptable metabolic stability in rat
liver microsomes (RLM CLin vitro ) 111 µL/min/mg) and good
in vivo pharmacokinetic properties in rats (t1/2 ) 2.8 h, CLin
vivo ) 1.2 L/h/kg, Foral ) 31% at 5 mg/kg), measurements of
stability in human liver microsomes (HLM CLin vitro ) 250 µL/
min/mg) suggested that this compound would be extensively
metabolized in humans. Further studies of compound 1 revealed
the region of the molecule that was most susceptible to
metabolism. Upon incubation of compound 1 with human liver
microsomes in the presence of NADPH, only two oxidative
metabolites were found (M1 and M2; Figure 1). Although the
exact sites of oxidation were not ascertained, analysis by MS/
MS indicated that both metabolites were the result of mono-
oxidation on the quinoline ring, while the center pyrimidine core
and the trifluoromethyl phenyl moieties were unaffected.
Therefore, our strategy to increase metabolic stability, as well
as enhance potency in this series, was to systematically explore
alternative heterocycles as replacements for the quinoline ring
found in compound 1 (exemplified by generic structure 2; Figure
1). This report describes the SAR investigations of the A-region,
which culminated in the identification of our first TRPV1
clinical candidate, AMG 517 (N-{4-[6-(4-trifluoromethyl-phen-
yl)-pyrimidin-4-yloxy]-benzothiazol-2-yl}-acetamide; compound
23). In the subsequent report, we will describe modifications
made to the C-region that resulted in improvements in the
pharmacokinetic and solubility profile of AMG 517.
Vanilloid receptor-1 (TRPV1 or VR1), a well-characterized
member of the transient receptor potential family of ion
channels, has been implicated in the transmission of pain
signaling. For this reason, the in vivo pharmacological effect
of small molecule antagonists of TRPV1 has been the subject
of intense investigations among many pain research groups.1
This report is the second in a series describing the in vitro and
in vivo properties of a novel class of pyrimidine TRPV1
antagonists discovered in our laboratories.2,3
Our general approach for developing the structure-activity
relationships (SAR) within this series was to dissect the
pharmacophore into three regions (A, B, and C; Figure 1) and
to optimize each region in a systematic fashion. In our first
report, we discussed modifications to region B, which led to
the identification of 4-oxopyrimidine 1, a potent inhibitor that
blocks capsaicin-induced 45Ca2+ influx in rat and human
TRPV1-expressing CHO cells with IC50 values of 7.4 nM and
3.7 nM, respectively. While compound 1 was shown to be
efficacious at blocking an on-target TRPV1-mediated response
in vivo in the capsaicin-induced hypothermia model in rats, it
did not show significant efficacy in the complete Freund’s
adjuvant (CFA)-induced inflammatory pain model. We postu-
lated that the minimal effect observed in the pain model may
be due to a combination of insufficient intrinsic TRPV1 potency
and inadequate exposure in vivo. Therefore, at this stage of our
SAR investigations, we sought to improve not only the potency
of the 4-oxopyrmidine TRPV1 antagonists, but also to enhance
their pharmacokinetic properties in vivo.
Chemistry
Analogues of compound 1 were prepared by one of four
general methods (methods A-D; Scheme 1). In method A,
4-chloro-6-[4-(trifluoromethyl)phenyl]-pyrimidine (3) was re-
acted with the alkoxide prepared by treatment of the requisite
heteroaromatic alcohol (ArOH) with sodium hydride, as previ-
ously described.2 The reactions were followed by HPLC or TLC
and stirred at room temperature or heated until complete. In
method B, the products were obtained by first converting
4-chloro-6-[4-(trifluoromethyl)phenyl]-pyrimidine (3) to the
As an initial step toward improving the overall profile of these
antagonists, we examined compound 1 in more detail. While
* To whom correspondence should be addressed. Tel.: 805-447-1552.
Fax: 805-480-3015. E-mail: markn@amgen.com.
† Department of Chemistry Research and Discovery.
‡ Department of Neuroscience.
§ Department of Pharmacokinetics and Drug Metabolism.
10.1021/jm070190p CCC: $37.00 © 2007 American Chemical Society
Published on Web 06/22/2007

3516 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 15
Doherty et al.
Figure 1. Lead compound 1, oxidative metabolites of compound 1, and generic SAR target 2.
Scheme 1. General Methods for the Synthesis of
4-Oxopyrimidines, 4a
the phenol with AlCl3. Intermediate diaminobenzene 12 was
also used to prepare the requisite isomeric quinoxalinones 18
and 19 (Scheme 2; eq IV). In this case, diaminobenzene 12
was condensed with ethoxy-imino-acetic acid ethyl ester under
conditions first described by Hermecz and co-workers4 to
provide a 1.7 to 1 mixture of isomeric quinoxalinones 16 and
17. The two isomers were separated by silica gel chromatog-
raphy and independently deprotected with AlCl3 to afford
3-amino-5-hydroxy-1H-quinoxalin-2-one (18) and 3-amino-8-
hydroxy-1H-quinoxalin-2-one (19).
a Method A: ArOH, NaH, DMF, 25 °C or heat. Method B: KF, DMSO,
120 °C; ArOH, K2CO3, DMF, 25 °C or 60 °C. Method C: ArOH, K2CO3,
DMSO, 80-90 °C. Method D: ArOH, DBU, CH3CN, reflux.
The final five compounds needed for this investigation
required additional modifications following the initial heteroaro-
matic alcohol coupling (compounds 22-26; Scheme 3). Treat-
ment of the 2-aminoquinoline derivative 20 (prepared by method
A, Scheme 1) with neat acetic anhydride at 105 °C provided
the corresponding acylated analogue 22 in good yield. Similarly,
2-aminobenzothiazole derivative 21 (prepared by method C,
Scheme 1) was acylated with either acetic anhydride, propionyl
chloride, or isobutyryl chloride to provide the corresponding
amides 23, 24 and 25. The final compound, N-methylamide 26,
was prepared by the reaction of compound 23 with methyl iodide
and sodium hydride in DMF at 0 °C.
corresponding fluoropyrimidine intermediate and then treating
this intermediate with various heteroaromatic alcohols in the
presence of solid potassium carbonate in DMSO at 25 °C or
60 °C. In the third general procedure, method C, 4-chloro-6-
[4-(trifluoromethyl)phenyl]-pyrimidine (3) was reacted directly
with the heteroaromatic alcohol in the presence of a base (solid
K2CO3) in DMSO at 80-90 °C. Finally, method D provided
the target compounds (4) by conducting the reactions in
refluxing acetonitrile, with DBU acting as the base.
With the exception of six examples described herein, the
requisite heteroaromatic alcohols were either commercially
available or readily prepared following published methods. To
obtain the six heteroaromatic alcohols needed to complete the
series (6, 7, 10, 14, 18, and 19), the reaction sequences shown
in Scheme 2 were followed. For example, treatment of com-
mercially available 2-chloroquinolin-8-ol (5) with methylamine
or dimethylamine under microwave heating provided 2-(me-
thylamino)quinolin-8-ol (6) and 2-(dimethylamino)quinolin-8-
ol (7), respectively (Scheme 2; eq I). For the preparation of
2-amino-8-hydroxyquinazoline (10), 3-methoxy-2-nitrobenzal-
dehyde (8) was reduced to 2-amino-3-methoxybenzaldehyde (9)
using Fe(0) and NH4Cl (Scheme 2; eq II). Condensation of
aminobenzaldehyde 9 with guanidine in decalin at 190 °C,
followed by deprotection of the phenol with NaSEt, afforded
compound 10 in 48% yield. Aminoquinazolin-5-ol (14) was
prepared in five steps starting from 2-amino-3-nitrophenol (11;
Scheme 2; eq III). Protection of phenol 11 by methylation,
followed by reduction of the nitro group with iron, provided
the corresponding diaminobenzene. This free base was unstable
upon standing and, therefore, was immediately converted to the
more stable hydrogen sulfate salt 12. Condensation of aniline
12 with ethyl glyoxylate provided an isomeric mixture of
quinoxalinones from which the desired 8-methoxy-1H-quinoxa-
lin-2-one (13) was isolated by silica gel chromatography.
Quinoxalinone 13 was converted to 3-aminoquinoxalin-5-ol (14)
by first generating the chloroquinoxaline intermediate using
POCl3, then introducing the amino group by treatment with NH4-
OH and CuI under microwave heating, and finally deprotecting
Results and Discussion
Structure-Activity Relationships (SAR). The compounds
described were tested for their ability to block the capsaicin-
(CAP, 500 nM) or acid- (pH 5.0) induced uptake of 45Ca2+ in
CHO cells expressing rat TRPV1 (rTRPV1), as previously
described.2 Functional activity is reported as IC50 ( SEM (nM)
in Tables 1-3. In a separate agonist assay, none of the
compounds presented showed agonist activity. Results for all
in vitro activities are the average of at least two independent
experiments, with three replicates at each concentration.
In our previous studies of the cinnamides series of TRPV1
antagonists,5 we found that the 7-quinolinyl moiety imparted
superior antagonist activity;6 however, this observation was
based on a limited set of 6,6-fused heteroaromatic groups
examined. Therefore, to more completely probe the heterocycle-
receptor interactions in the A-region, we systematically exam-
ined each position of the nitrogen by surveying all possible
quinoline and isoquinoline isomers in the 4-oxopyrimidine series
(27a-f and 28a-g, Table 1). When the nitrogen was moved
around the ring, TRPV1 inhibitory activity decreased to varying
degrees in both series. We found that the derivatives that
contained the nitrogen in the 8-position gave the best results
regardless of whether the ether linkage was attached to the 1-
or 2-position of the quinoline ring (compounds 28a and 1,
respectively).

Head: TRPV1 Antagonists: Selection of a Clinical Candidate
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 15 3517
Scheme 2. Preparation of Heteroaromatic Phenols 6, 7, 10, 14, 18, and 19 (ArOH)a
a Reagents and conditions: (a) H2NMe, dioxane, microwave, 200 °C; (b) HNMe2, dioxane, microwave, 200 °C; (c) Fe(0), H2O, MeOH, 60 °C; (d)
guanidine hydrochloride, Na2CO3, decalin, 190 °C; EtSNa, DMF, reflux; (e) MeI, K2CO3, DMF, 25 °C; Fe(0), 12 M HCl, EtOH, H2O, reflux; H2SO4; (f)
EtOH, NaHCO3, (HCO)CO2Et, reflux; (g) POCl3, concd NH4OH, CuI, microwave, 140 °C; (h) AlCl3, benzene, reflux; (i) EtOH, H2O, NaHCO3, 25 °C.
Scheme 3. Preparation of 2-Aminoquinoline and 2-Aminobenzothiazole Amide Analogues 22-26a
a Reagents and conditions: (a) Ac2O, 105 °C; (b) propionyl chloride, BEMP resin, THF, room temperature, 16 h; (c) isobutyryl chloride, BEMP resin,
THF, room temperature, 16 h; (d) NaH, CH3I, DMF, 0 °C.
While the 7-quinolinyl analogue 1 and the 8-quinolinyl
analogue 28a were approximately equipotent in the capsaicin-
mediated assay, the 8-quinolinyl analogue was significantly less
potent in the acid-mediated assay. Presumably, the binding
region on the receptor is subtly changed at pH 5.0, resulting in
the observed difference in activity.7 Despite an undesired
reduction in activity in the acid-mediated assay,8 the 8-quinolinyl
isomer 28a demonstrated an encouraging improvement in
microsomal stability over compound 1. Specifically, in isolated
rat and human liver microsomes, compound 28a was cleared
at rates of 44 and 135 µL/min/mg, respectively, compared to
111 and 250 µL/min/mg for compound 1. Therefore, we focused
our continued efforts on improving the potency and pharma-
cokinetic properties of the 8-quinolinyl analogue.
It has been our experience that subtle modifications to the
A-ring heterocycle can impact activity in the TRPV1 acid-
mediated assay. Therefore, an initial set of compounds was
prepared to investigate the effect of an additional substituent
on the 2-position of the quinoline ring of compound 28a (Table
2). The oxygen-substituted analogues (2-hydroxyquinoline, 29a,

3518 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 15
Doherty et al.
Table 1. Inhibition of CAP- (500 nM) and Acid- (pH 5.0) Induced
45Ca2+ Influx into rTRPV1-Expressing CHO Cells for Compounds 1,
27a-f, and 28a-g
From this initial set of 2-substituted quinolines, the 2-amino
analogue 20 was selected for further SAR development. To
examine this derivative in more detail, we conducted metabolism
identification studies to obtain information regarding possible
ways to further reduce clearance. As observed with compound
1, the center pyrimidine core and the trifluoromethyl phenyl
moieties of compound 20 were stable in human liver mi-
crosomes, but oxidative metabolism occurred on the quinoline
ring. Studies in isolated rat and human hepatocytes10 showed
that compound 20 underwent phase I and phase II metabolism,
resulting in the formation of at least 15 metabolites. Consistent
with the microsomal studies, the predominant site of metabolism
in hepatocytes was the aminoquinoline ring (Figure 2). Hy-
droxylation on the quinoline ring, followed by sulfation,
accounted for 70% of metabolism in rat hepatocytes.
With this information in hand, we prepared a series of
analogues designed to block the proposed sites of metabolism,
focusing initially on the 2-aminopyridyl portion of 20 (Table
3). The 2-amino-8-quinazolinyl analogue (30) was prepared to
determine if an additional nitrogen atom introduced at the
3-position of the quinoline ring would serve to block metabolic
oxidation. This modification resulted in the desired decrease in
microsomal clearance, however, compound 30 suffered from a
significant reduction in potency in the acid-mediated assay.
Introduction of a nitrogen to the 4-position to afford 2-amino-
8-quinoxalinyl analogue 31 gave the desired improvement in
microsomal clearance without the attendant reduction in potency
in the acid-mediated assay observed for compound 30. Never-
theless, compound 31 demonstrated unacceptably high clearance
nitrogen
position
rTRPV1 (CAP)
IC50 (nM)a
rTRPV1 (acid)
IC50 (nM)a
cmpd
1
8
7
6
5
4
3
1
8
7
6
5
4
3
2
7.4 ( 1.0
460 ( 50
1400 ( 200
>4000
8.0 ( 6.0
>4000
>4000
>4000
>4000
>4000
>4000
>4000
120 ( 40
>4000
>4000
>4000
>4000
>4000
27a
27b
27c
27d
27e
27f
28a
28b
28c
28d
28e
28f
28g
>4000
630 ( 30
730 ( 100
15 ( 7
150 ( 20
120 ( 10
660 ( 240
>4000
2900 ( 1000
>4000
a Each IC50 value reported represents an average of at least two
independent experiments with three replicates at each concentration
((SEM).
Table 2. Inhibition of CAP- (500 nM) and Acid- (pH 5.0) Induced
45Ca2+ Influx into rTRPV1-Expressing CHO Cells for Compounds 28a,
29a-d, 20, and 22
in vivo [CLin
) 4.3 L/h/kg], greater than the rate of rat
vivo
hepatic blood flow. Despite this result, additional studies were
conducted on 31, which provided insights leading to further
specific structural modifications. For example, compound 31,
dosed at 3 mg/kg, p.o., blocked 100% of the capsaicin-induced
hypothermic response in our on-target in vivo telemetry model2
(data not shown). This in vivo effect was surprising given the
compound’s high clearance and modest potency and led us to
postulate that the in vivo activity observed for 31 was due to
the formation of an active metabolite. Literature precedent11
indicated that hydroxylation of the 3-position of the 2-amino-
quinoxaline was likely, and for this reason, we prepared the
corresponding oxidized derivative of 31 for further evaluation.
The putative metabolite, compound 32, demonstrated not only
rTRPV1 (CAP)
IC50 (nM)a
rTRPV1 (acid)
IC50 (nM)a
cmpd
R
28a
29a
29b
20
29c
29d
22
H
15 ( 7
>4000
>4000
11 ( 6.5
35 ( 2
>4000
77 ( 65
>4000
>4000
>4000
62 ( 34
43 ( 4
>4000
15 ( 18
OH
OMe
NH2
NHMe
NMe2
NHAc
the anticipated improvement in clearance in the rat (CLin vivo
)
0.26 L/h/kg) but also a dramatic improvement in potency (IC50
) 0.64 nM in the capsaicin assay, and IC50 ) 0.57 nM in the
pH 5.0 assay).
a Each IC50 value reported represents an average of at least two
independent experiments with three replicates at each concentration
((SEM).
Based on the results for 30 and 31, it was clear that
replacement of a single aromatic CH with nitrogen in the
pyridinyl portion of the quinoline ring was insufficient to block
the metabolism of 20. Therefore, the 2-aminobenzothiazole
analogue 21 was designed to block two of the proposed
metabolic hot spots on the quinoline ring of 20, replacing the
C3-C4 atoms with sulfur. This modification resulted in a
dramatic improvement in the metabolic stability in human liver
micosomes (CLin vitro < 5 µL/min/mg). In addition, compound
and 2-methoxyquinoline, 29b) proved to be significantly less
potent in both assays. In contrast, the results for the nitrogen-
substituted derivatives were more promising. For example,
activity in the acid-mediated assay was restored with the 2-amino
(20), 2-methylamino (29c), and 2-N-acylamino (22) derivatives.
However, activity was diminished when the quinoline was
substituted in the 2-position with dimethylamino (29d). Taken
together, these results suggest that the NH groups of 20, 29c,
and 22 may act as hydrogen-bond donors to create a favorable
interaction with the receptor under the conditions of both the
capsaicin and the acid-mediated assays. This favorable interac-
tion is eliminated in the case of the 2-methoxy and 2-dimethy-
lamino substitutions (29b and 29d, respectively). Furthermore,
given that the 2-hydroxy derivative 29a is likely to exist
primarily in the quinolin-2-one tautomeric form,9 it may be less
active than 20, 29c, and 22 because it cannot provide a similar
hydrogen-bonding interaction in the 2-position.
Figure 2. Major metabolites of compound 20 in rat hepatocytes.

Head: TRPV1 Antagonists: Selection of a Clinical Candidate
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 15 3519
Table 3. Inhibition of CAP- (500 nM) and Acid- (pH 5.0) Induced 45Ca2+ Influx into rTRPV1-Expressing CHO Cells and In Vitro and In Vivo
Clearance Data for Analogues of 20
a Each IC50 value reported represents an average of at least two independent experiments with three replicates at each concentration ((SEM).
b Rat liver microsomal (RLM) and human liver microsomal (HLM) clearance. c Intravenous bolus injection, 1 mg/kg in DMSO, n ) 2 male Sprague-
Dawley rats.

3520 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 15
Doherty et al.
receptor against three modes of activation (capsaicin, low pH,
and heat). The results shown in Table 4 confirm that compounds
23 and 32 are potent inhibitors of the human TRPV1 under
every condition tested.
To select a suitable clinical candidate, we compared some
of the pharmacokinetic and physicochemical properties for the
two most potent compounds in the series, 23 and 32 (Table 5).
In our preliminary pharmacokinetic studies, the compounds were
dosed at 1 mg/kg i.v. and plasma samples were collected and
analyzed up to an 8-hour time point. In these studies, both
compounds demonstrated long half-lives (t1/2), low plasma
clearance (CL), and high volumes of distribution (Vss). However,
when administered orally as a suspension at 3 mg/kg, compound
23 proved to be the superior candidate in terms of bioavailability,
Foral ) 32%. In comparison, the oral bioavailability of 32 in
rats was lower (Foral ) 7%, 5 mg/kg). While the aqueous
solubility was low for both compounds,12 the melting point of
32 was 115 degrees higher than 23. The higher melting point
of 32 and the resulting high-energy barrier to dissolution may
help to explain the differences in oral bioavailability observed
between 32 and 23. Furthermore, compound 32 showed dose-
limiting exposure at higher doses in rats (30 and 300 mg/kg,
p.o.), presumably due to solubility-limited absorption. Con-
versely, compound 23 exhibited excellent exposure levels at
the same doses, making it the better candidate for further safety
assessment studies in rodents.
Figure 3. Binding elements required for optimum potency in the
A-region (compounds 32 and 23).
Table 4. Inhibition of CAP- (500 nM), Acid- (pH 5.0), or Heat-(45 °C)
Induced 45Ca2+ Influx into Human TRPV1-expressing CHO Cells for
Compounds 23 and 32.
hTRPV1 (CAP)
IC50 (nM)a
hTRPV1 (acid)
IC50 (nM)a
hTRPV1 (heat)
IC50 (nM)a
cmpd
23
32
0.76 ( 0.4
0.8 ( 0.28
0.62 ( 0.27
0.9 ( 0.35
1.3 ( 0.1
0.4 + 0.06
a Each IC50 value reported represents an average of at least two
independent experiments with three replicates at each concentration
((SEM).
21 showed a 10-fold improvement in potency over 20 in the
capsaicin-mediated assay (IC50 ) 1.2 nM vs 11 nM); however,
activity in the acid-mediated assay was sacrificed (IC50 > 4000
nM). In contrast, the N-acylated analogue 23 retained potency
in both assays (IC50 ) 0.9 and 0.5 nM in the capsaicin- and
acid-mediated assays, respectively), while also maintaining
excellent metabolic stability in human liver micosomes (CLin
vitro < 5 µL/min/mg).
With compound 23 selected as the lead candidate, its
pharmacokinetic properties were more extensively examined
(Table 6). For routine screening purposes, we had set an 8-hour
duration for initial in vivo pharmacokinetics studies in rats.
Because compound 23 exhibited a relatively high volume of
distribution and low clearance, the calculated half-life from these
preliminary studies was an underestimation. To capture the true
elimination phase, an additional study was conducted in rats
measuring blood levels beyond an estimated four half-lives. In
this extended 72 h study, a more accurate half-life of 31 h was
determined for compound 23.
Based on the overall SAR and the two exquisitely potent
TRPV1 antagonists identified (i.e., 23 and 32), a pharmacophore
model was constructed for the A-region of this series. The
proposed hydrogen-bond donor-acceptor interactions between
the carbonyl oxygen, NH proton, and the aromatic nitrogen of
the antagonists and the receptor for both 23 and 32 are shown
in Figure 3. The orientation of these elements seems to be critical
for optimum potency, as evidenced by the significant reduction
in potency observed for the isomeric analogue of 32 (compound
33). In addition, when the amide NH of compound 23 was
methylated (e.g., compound 26), key interactions with the
receptor were disrupted that also resulted in lower potency. The
reduction in potency for the N-methylated compound (26) may
be due to the elimination of the hydrogen-bonding interaction
of the NH or due to an unfavorable conformational change in
the acetyl group. Finally, the size of the pocket in this region
of the receptor also appears to be sensitive to steric bulk. For
example, significant decreases in TRPV1 inhibitory activity were
observed when the alkyl group of the amide was extended
further into the pocket, as in the case of the ethyl and isopropyl
derivatives (24 and 25, respectively).
The comparative pharmacokinetic profile for compound 23
was examined in rats, dogs, and monkeys (Table 6). In all spe-
cies examined, compound 23 exhibited low clearance and mod-
erately high volumes of distribution with consequently long half-
lives. Oral bioavailability among the three species tested ranged
from 23 to 52%. The mean exposures (AUC) for compound 23
increased in a dose proportional manner in the dose range of
1-10 mg/kg in rats, dogs, and monkeys (Figure 4).
Human pharmacokinetic parameters were projected based on
allometric scaling and are presented in Table 6. Based on the
allometric projection, compound 23 was predicted to have a
long half-life in humans (60-120 h). The projected half-life in
humans was expected to result in a 4-fold accumulation at
steady-state, based on a QD dosing regimen, with minimal peak-
to-trough fluctuations and low variability.
In addition to being tested at the rat TRPV1 receptor,
compounds 23 and 32 were evaluated at the human TRPV1
Table 5. Pharmacokinetic and Physicochemical Profile for 23 vs 32
intravenous dosinga
solubilityd (µg/mL)
human liver
microsomal
CL
(mL/h/kg)
t1/2
(h)
AUC0-∞
(ng‚h/mL)
Vss
(mL/kg)
Foral
(%)
stability
(µL/min/mg)
HClaq
(0.01 N)
mp
(°C)
cmpd
PBSe
SIFf
23
32
190
197
6.3
6.0
5400
5143
1556
1688
32b
7c
<5
26
<1
<1
<1
<1
6.6
9.2
219-221
334-335
a Study in fed male Sprague-Dawley rats dosed at 1 mg/kg in DMSO with sampling time up to 6 h. n ) 2 animals per study. Interanimal variability was
less than or equal to 30%. b Study in fasted male Sprague-Dawley rats dosed at 3 mg/kg as a suspension in 5% Tween 80/Oraplus with sampling time up
to 8 h. n ) 2 animals. Interanimal variability was less than or equal to 30%. c Study in fasted male Sprague-Dawley rats dosed at 5 mg/kg as a suspension
in 5% Tween 80/Oraplus with sampling time up to 8 h. n ) 2 animals. Interanimal variability was less than or equal to 30%. d Thermodynamic solubility
measured in a high-throughput automated format (ref 11). e Phosphate buffered saline, pH 7.4. f Simulated intestinal fluid, pH 6.8.

Head: TRPV1 Antagonists: Selection of a Clinical Candidate
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 15 3521
Table 6. Mean Pharmacokinetic Parameters for Compound 23
Following i.v. and p.o. Administration to Fasted Rat, Dog, and Monkey,
with Projected Values for Human
oral
intravenous dosinga
CL Vss
(ng‚h/mL) (mL/h/kg) (mL/kg)
dosingb
AUC0-∞
t1/2
(h)
F
(%)
species
rat
8800
7400
37 000
120
140
30
4000
7000
2300
4500
31
41
62
51
23
52
dog
monkey
human
(projected)
50
60-120
a 1 mg/kg in 80% PEG-400/H2O, n ) 3 animals per group. Variability
for AUC0-∞, CL, Vss, and t1/2 values ranged from 7 to 38% for all species.
b 1 mg/kg suspension in 10% Pluronic/Oraplus, n ) 3 animals per group.
Variability for Foral ranged from 26 to 50% for all species.
Figure 4. Dose linearity of 23 in rat, dog, and monkey at 1, 3, and 10
mg/kg, dosed p.o. as suspensions in 10% Pluronic/Oraplus.
Compound 23 was then evaluated for its ability to reverse
thermal hyperalgesia in the CFA model in rats (Figure 6). The
compound was dosed orally at 0.1, 0.3, 1, 3, 10, and 30 mg/kg
to rats that had been treated 21 h in advance with CFA injected
in the paw. Paw withdrawal latencies from a heat source were
measured 3 h post-dosing with 23. In this pain model, compound
23 showed dose-dependent inhibition of CFA-induced thermal
hyperalgesia with efficacy at a minimum dose (MED) of 0.83
mg/kg (p.o.). The maximal effect observed in this model was
approximately 35% of control at a dose of 10 mg/kg. In
comparison, in the on-target model we observed a 100% return
to control levels at 10 mg/kg (Figure 5), suggesting that the
target is fully covered at this dose. Therefore, the degree of
reversal of thermal hyperalgesia observed in the CFA model
probably reflects the magnitude of involvement of TRPV1 in
the CFA-induced inflammatory pain pathway in rodents.
In addition to the extensive in vivo pharmacokinetic studies,
the on-target in vivo efficacy of 23 was measured in a capsaicin-
induced flinching model in rats (Figure 5). The compound was
dosed at 0, 0.1, 1, 3, 10, and 30 mg/kg as a suspension in 5%
Tween 80 in Oraplus 1 h prior to capsaicin challenge. An ED50
) 0.35 mg/kg was calculated based on number of flinches per
minute counted after intraplantar injection of capsaicin (0.5 µg
in 5% EtOH /PBS). A significant correlation was found between
number of flinches and plasma levels of compound, that is, the
number of flinches decreased as plasma levels increased.
Furthermore, we demonstrated that a 3 mg/kg dose (p.o.) of
compound 23 significantly inhibited capsaicin-induced flinching
up to 24 h. At 48 h, the animals behavior had returned to control
levels.13 These results demonstrated that 23 was a potent
inhibitor of TRPV1 in vivo.
Figure 5. Efficacy and terminal plasma concentration of compound 23 in the capsaicin-induced flinch model (dosed p.o. in 5% Tween 80/Oraplus
1 h prior to capsaicin challenge; n ) 8 per group of Harlan SD male rats).
Figure 6. Dose-response for compound 23 in CFA-induced paw thermal hyperalgesia (dosed p.o. in 5% Tween 80/Oraplus; 21 h post-CFA
treatment; n ) 14-35 per group of Harlan SD male rats).

3522 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 15
Doherty et al.
All final compounds were purified to >95% purity, as determined
by reverse-phase high-performance liquid chromatography (rp-
HPLC). Purity was determined on an Agilent 1100 spectrometer
by method A, Phenomenex Luna C8 column (100 × 4.6 mm, 5 µ)
at 40 °C with a 1 mL/min flow rate using a gradient of 10-100%
0.1% TFA in acetonitrile in 0.1% TFA in water over 10 min; or
method B, YMC ODS-AM C18 column (100 × 2.1 mm, 5 µ) at
40 °C with a 0.5 mL/min flow rate using a gradient of 10-100%
0.1% TFA in acetonitrile in 0.1% TFA in water over 7 min. Silica
gel chromatography was performed using either glass columns
packed with silica gel (200-400 mesh, Aldrich Chemical) or
prepacked silica gel cartridges (Biotage or RediSep). Melting points
were determined on a Buchi-545 melting point apparatus and are
uncorrected. NMR spectra were determined with a Bruker 300 MHz
or DRX 400 MHz spectrometer. Chemical shifts are reported in
parts per million (ppm, δ units). Low-resolution mass spectral (MS)
data were determined on a Perkin-Elmer-SCIEX API 165 mass
spectrometer using electrospray (ES) ionization modes (positive or
negative). High-resolution mass spectral (HRMS) data were
determined on a 7T Bruker FTICR mass spectrometer using ES
ionization mode (positive). Combustion analyses were performed
by Atlantic Microlab, Inc., Norcross, GA, and were within 0.4%
of calculated values unless otherwise noted.
General Methods for the Synthesis of 4-Oxopyrimidines (4).
Method A: In an oven-dried, round-bottomed flask, a solution of
the alcohol (1.1 equiv) in DMF (0.15-0.25 M) was stirred at room
temperature and treated portionwise with sodium hydride, as a 60%
dispersion in mineral oil (1.2 equiv). Upon complete addition, the
reaction mixture was stirred for 10 min and then 4-chloro-6-(4-
(trifluoromethyl)phenyl)pyrimidine2 (3; 1.0 equiv) was added in
one portion. The reaction mixture was stirred at room temperature,
or heated, until complete (as determined by TLC or HPLC). The
reaction mixture was diluted with H2O and extracted with EtOAc.
The organic extract was washed successively with 1 N NaOH, H2O,
and satd NaCl, dried over Na2SO4, filtered, and concentrated in
vacuo to afford the crude product.
Based on its excellent overall in vitro and in vivo profile,
compound 23 was selected for further evaluation in human
clinical trials and was designated AMG 517. Additional details
of the pharmacological, toxicological, and clinical profiles of
AMG 517 will be reported elsewhere.
Summary
In this investigation, we examined the structure-activity
relationships in the heterocyclic A-region of the 4-oxopyrimidine
TRPV1 antagonists, with the aim of improving potency and
increasing metabolic stability of lead compound 1. Through
these SAR investigations, we developed several potent TRPV1
antagonists and increased our understanding of the pharma-
cophoric elements required in the A-ring. The initial set of
quinoline and isoquinoline derivatives revealed that the 8-quino-
linyl isomer (compound 28a) retained TRPV1 potency in the
capsaicin-mediated assay, while being more stable than com-
pound 1 upon incubation in rat or human liver microsomes.
Furthermore, we found that an additional potency-enhancing
hydrogen-bonding interaction with the TRPV1 receptor could
be obtained by substituting the 2-position of the quinoline ring
with an amino group (e.g., compound 20). Metabolite identifica-
tion studies showed that compound 20 was extensively me-
tabolized on the 2-aminoquinoline moiety. Attempts to block
the metabolism of the quinoline ring led to the identification of
two metabolically stable and exquisitely potent TRPV1 antago-
nists, benzothiazole 23 and quinoxalinone 32. Further evaluation
of compound 23 showed that it potently blocked multiple modes
of TRPV1 activation and was effective in a rodent on-target
biochemical challenge model (capsaicin-induced flinch, ED50
) 0.33 mg/kg p.o.). In the CFA pain model in rats, compound
23 showed a dose-dependent inhibition of thermal hyperalgesia
(MED ) 0.3 mg/kg, p.o.). Because of its excellent pharmaco-
kinetic properties and in vivo efficacy, compound 23 (AMG
517) was selected for further evaluation in human clinical trials
for the treatment of inflammatory pain. In part 3 of this series,
we will describe modifications made to the C-region that
resulted in improvements in the pharmacokinetic and solubility
profile of AMG 517.
Method B: To a solution of 4-chloro-6-(4-(trifluoromethyl)-
phenyl)pyrimidine (3; 2.0 g, 7.7 mmol) in DMSO (15 mL) was
added potassium fluoride (3.6 g, 61 mmol). The mixture was stirred
at 120 °C for 6 h. After cooling to room temperature, the mixture
was poured into H2O (200 mL) and extracted with EtOAc (2 ×
200 mL). The combined extracts were washed with H2O (3 × 200
mL) and satd NaCl (200 mL) and then dried over Na2SO4, filtered,
and concentrated in vacuo onto silica gel. Purification by silica gel
chromatography (gradient: 0 to 6% EtOAc in hexanes) afforded
4-fluoro-6-(4-(trifluoromethyl)phenyl)pyrimidine (1.7 g, 90%) as
Experimental Section
1
a white solid. MS (ESI, pos. ion) m/z: 243 (M + 1). H NMR
(400 MHz, CDCl3) δ 7.39 (t, J ) 1.0 Hz, 1 H), 7.80 (d, J ) 8.4
Hz, 2 H), 8.21 (d, J ) 8.4 Hz, 2 H), 9.02 (t, J ) 1.0 Hz, 1 H).
To an oven-dried, round-bottomed flask was added 4-fluoro-6-
(4-(trifluoromethyl)phenyl)pyrimidine (1.8 equiv), alcohol (1.0
equiv), K2CO3 (2.0 equiv), and DMF (0.08-0.15 M). The reaction
mixture was stirred at room temperature, or at 60 °C, until complete
(as determined by TLC or HPLC). The mixture was poured into
aq NaHCO3 and extracted with EtOAc. The extract was washed
with H2O and satd NaCl, dried over Na2SO4, filtered, and
concentrated in vacuo to afford the crude product.
Method C: A mixture of 3 (1.1 equiv), alcohol (1.0 equiv), and
solid K2CO3 (1.4-2.0 equiv) in anhydrous DMSO (0.1-0.5 M) or
anhydrous DMF (0.1-0.5 M) was stirred in an 80-90 °C oil bath
until complete (as determined by TLC or HPLC). The reaction
mixture was allowed to cool to room temperature and diluted with
EtOAc. The mixture was washed with H2O and satd NaCl, dried
over MgSO4, filtered, and concentrated in vacuo to afford the crude
product.
Capsaicin-Induced Flinch Model. Male Sprague-Dawley rats
(Charles River Laboratories (CRL), Wilmington, MA) weighing
250-300 g were used. Animals were allowed at least a 1 week
acclimation period in Amgen’s AAALAC-accredited animal care
facility prior to being tested. On test day, animals were dosed (p.o.)
with compound 23 at a dosing volume of 5 mL/kg 1 h prior to the
administration of a capsaicin challenge. The vehicle for compound
23 was 5% Tween 80 in Oraplus. Capsaicin was administered at a
dose of 0.5 µg into the right paw of the animal. The injection
volume for capsaicin was 25 µL, and the vehicle for capsaicin was
5% EtOH in PBS (Ca++and Mg++free). Immediately following
intraplantar injection of capsaicin, flinch responses were counted
for a 1 min period. Investigators counting flinches were blinded to
the treatment conditions. For each treatment group, n ) 8. The
ED50 value was computed using GraphPad Prism software (San
Diego, CA). Significance level was set at p < 0.05. Percent
inhibition was computed as 100 × (1 - (test value - mean vehicle/
vehicle value)/(mean vehicle/capsaicin value - mean vehicle/
vehicle value).
Method D: A suspension of the alcohol (1.2 equiv) in CH3CN
(0.05-0.1 M) was treated with 1,8-diazabicyclo[5.4.0]undec-7-ene
(1.2 equiv) followed by 3 (1.0 equiv). The reaction mixture was
stirred at reflux until complete (as determined by TLC or HPLC).
The mixture was allowed to cool to room temperature and
concentrated in vacuo to afford the crude product.
Chemistry. Unless otherwise noted, all materials were obtained
from commercial suppliers and used without further purification.
Anhydrous solvents were obtained from Aldrich or EM Science
and used directly. All reactions involving air- or moisture-sensitive
reagents were performed under a nitrogen or argon atmosphere.

Head: TRPV1 Antagonists: Selection of a Clinical Candidate
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 15 3523
8-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)quinolin-
2-amine (20). Following method A, 2-amino-8-quinolinol (0.70 g,
4.4 mmol) and 3 (1.2 g, 4.6 mmol), after purification of the crude
product by silica gel chromatography (98:2 CH2Cl2/2 M NH3 in
MeOH), followed by recrystallization from CH2Cl2 and hexane,
afforded 20 (1.0 g, 60%) as fine white crystals. Mp: 203-204 °C.
MS (ESI, pos. ion) m/z: 383 (M + 1). 1H NMR (400 MHz, DMSO-
d6) δ 6.44 (s, 2 H), 6.77 (d, J ) 8.9 Hz, 1 H), 7.20 (t, J ) 7.8 Hz,
1 H), 7.38 (dd, J ) 1.3, 7.6 Hz, 1 H), 7.60 (dd, J ) 1.3, 8.0 Hz,
1 H), 7.89 (d, J ) 1.0 Hz, 1 H), 7.93 (d, J ) 8.3 Hz, 2 H), 7.96 (d,
J ) 9.0 Hz, 1 H), 8.45 (d, J ) 8.2 Hz, 2 H), 8.74 (d, J ) 1.0 Hz,
1 H). Anal. (C20H13F3N4O): C, H, N, F.
H), 7.30 (dd, J ) 1.0, 8.0 Hz, 1 H), 7.41 (t, J ) 8.0 Hz, 1 H), 7.48
(d, J ) 1.0 Hz, 1 H), 7.77-7.79 (m, 3 H), 8.20 (d, J ) 8.0 Hz, 2
H), 8.82 (d, J ) 1.0 Hz, 1 H), 9.18 (br s, 1 H). Anal.
(C22H17F3N4O2S): C, H, N, S.
N-Methyl-N-(4-(6-(4-(trifluoromethyl)phenyl)pyrimidin-4-
yloxy)benzo[d]thiazol-2-yl)acetamide (26). A mixture of 23 (215
mg, 0.50 mmol) and iodomethane (31 µL, 0.50 mmol) in anhydrous
DMF (2 mL) was treated with NaH (24 mg, 0.60 mmol) at 0 °C.
The reaction mixture was stirred at 0 °C for 20 min, then quenched
with satd NH4Cl (10 mL) and extracted with EtOAc (2 × 30 mL).
The combined organic extract was washed with H2O (2 × 20 mL)
and satd NaCl (20 mL), dried over Na2SO4, filtered, and concen-
trated in vacuo. Purification of the crude product by silica gel
chromatography (30% EtOAc in hexane) provided 26 (140 mg,
63% yield) as a white solid. MS (ESI, pos. ion) m/e: 446 (M +
1). 1H NMR (300 MHz, CDCl3) δ 2.40 (s, 3 H), 3.57 (s, 3 H), 7.29
(dd, J ) 1.2, 7.9 Hz, 1 H), 7.39 (t, J ) 7.8 Hz, 1 H), 7.45 (d, J )
1.0 Hz, 1 H), 7.72-7.81 (m, 3 H), 8.20 (d, J ) 8.0 Hz, 2 H), 8.82
(d, J ) 1.0 Hz, 1 H). Anal. (C21H15F3N4O2S): C, H, N.
4-[6-(4-Trifluoromethyl-phenyl)-pyrimidin-4-yloxy]-benzothi-
azol-2-ylamine (21). Following method C, 2-amino-4-hydroxy-
benzothiazole (0.42 g, 2.5 mmol) and 3 (0.97 g, 3.8 mmol), after
purification of the crude product by silica gel chromatography (2:1
hexanes/EtOAc), provided 21 (0.77 g, 79%) as a white solid. Mp:
1
245-246 °C. MS (ESI, pos. ion) m/z: 389 (M + 1). H NMR
(400 MHz, CDCl3) δ 5.23 (br s, 2 H), 7.16-7.25 (m, 2 H), 7.45
(s, 1 H), 7.55 (dd, J ) 1.6, 7.8 Hz, 1 H), 7.77 (d, J ) 7.8 Hz, 2 H),
8.18 (d, J ) 7.8 Hz, 2 H), 8.84 (s, 1 H). Anal. (C18H11F3N4OS): C,
H, N, S.
7-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)isoquino-
line (27a). Following method A, isoquinolin-7-ol (0.40 g, 2.8 mmol)
and 3 (0.67 g, 2.6 mmol), after purification of the crude product
by silica gel chromatography (gradient: 1-2% 2 M NH3 in MeOH/
CH2Cl2), afforded 27a (0.35 g, 36%) as a tan solid. Mp: 167-168
N-{8-[6-(4-Trifluoromethyl-phenyl)-pyrimidin-4-yloxy]-quin-
olin-2-yl}-acetamide (22). A mixture of 20 (0.79 g, 2.1 mmol)
and acetic anhydride (2.5 mL, 2.6 mmol) was stirred in a 105 °C
oil bath for 8 h. The volatiles were removed in vacuo to afford a
solid residue. Purification of the crude product by silica gel
chromatography (1:2 EtOAc/hexane) followed by recrystallization
from EtOAc provided 22 (0.56 g, 63%) as a white solid. Mp: 204-
206 °C. MS (ESI, pos. ion) m/z: 425 (M + 1). 1H NMR (400 MHz,
CDCl3) δ 2.16 (s, 3 H), 7.49 (s, 1 H), 7.51-7.54 (m, 2 H), 7.74-
7.80 (m, 3 H), 7.94 (s, 1 H), 8.22 (dd, J ) 8.4, 10.4 Hz, 3 H), 8.77
(s, 1 H). Anal. (C22H15F3N4O2): C, H, N.
N-{4-[6-(4-Trifluoromethyl-phenyl)-pyrimidin-4-yloxy]-ben-
zothiazol-2-yl}-acetamide (23). A mixture of 21 (97 mg, 0.25
mmol) and acetic anhydride (0.24 mL, 2.5 mmol) was stirred in a
105 °C oil bath for 8 h. The volatiles were removed in vacuo, and
the solid residue was recrystallized from EtOAc in hexanes. The
solid was dried in vacuo at 60 °C to give 23 (52 mg, 48%) as a
white solid. Mp: 219-221 °C. MS (ESI, pos. ion) m/z: 431 (M +
1). 1H NMR (400 MHz, DMSO-d6) δ ppm 2.13 (s, 3 H), 7.35 (dd,
J ) 1.4, 7.8 Hz, 1 H), 7.39 (t, J ) 7.8 Hz, 1 H), 7.87-7.96 (m, 3
H), 7.97 (d, J ) 0.78 Hz, 1 H), 8.45 (d, J ) 8.0 Hz, 2 H), 8.80 (d,
J ) 0.78 Hz, 1 H), 12.42 (s, 1 H). Anal. Calcd for C20H13F3N4O2S‚
0.75H2O: C, 54.11; H, 3.29; N, 12.62; S, 7.22. Found: C, 54.12;
H, 3.07; N, 12.61; S, 7.30. HPLC analysis, method A: 98.2% at
215 nm; 99.5% at 254 nm; retention time 8.76 min.
N-{4-[6-(4-Trifluoromethyl-phenyl)-pyrimidin-4-yloxy]-ben-
zothiazol-2-yl}-propionamide (24). To a solution of 21 (0.19 g,
0.49 mmol) in THF (5 mL) was added propionyl chloride (87 µL,
1.0 mmol), followed by 2-tert-butylimino-2-diethylamino-1,3-
dimethyl-perhydro-1,3,2-diazaphosphorine, polymer-bound (BEMP
resin; 0.34 g, 0.75 mmol). The reaction mixture was stirred at room
temperature for 16 h. The solid was removed by suction filtration
and washed with CH2Cl2. The combined filtrate and washes were
concentrated in vacuo to give a light-yellow solid. Purification by
silica gel chromatography (1:4 EtOAc/hexane) provided 24 (0.047
g, 42%) as a white solid. Mp: 197-198 °C. MS (ESI, pos. ion)
m/z: 445 (M + 1). 1H NMR (400 MHz, CDCl3) δ 1.26 (t, J ) 8.0
Hz, 3 H), 2.47 (qt, J ) 8.0 Hz, 2 H), 7.29 (dd, J ) 1.0, 8.0 Hz, 1
H), 7.40 (t, J ) 8.0 Hz, 1 H), 7.45 (d, J ) 1.0 Hz, 1 H), 7.76-7.78
(m, 3 H), 8.18 (d, J ) 8.0 Hz, 2 H), 8.80 (d, J ) 1.0 Hz, 1 H),
8.90 (br s, 1 H). Anal. (C21H15F3N4O2S): C, H, N, S.
N-(4-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)benzo-
[d]thiazol-2-yl)isobutyramide (25). According to the procedure
described for the preparation of 24, compound 21 (0.19 g, 0.49
mmol) and isobutyryl chloride (0.11 mL, 1.0 mmol), after purifica-
tion of the crude product by silica gel chromatography (1:4 EtOAc/
hexane), provided 25 (0.10 g, 45%) as a white solid. Mp: 178-
181 °C. MS (ESI, pos. ion) m/z: 459 (M + 1). 1H NMR (400 MHz,
CDCl3) δ 1.28 (d, J ) 4.0 Hz, 6 H), 2.62 (dq, J ) 4.0, 8.0 Hz, 1
1
°C. MS (ESI, pos. ion) m/z: 368 (M + 1). H NMR (400 MHz,
DMSO-d6) δ 7.75 (dd, J ) 2.4, 8.8, 1 H), 7.91 (d, J ) 5.9 Hz, 1
H), 7.94 (d, J ) 8.4 Hz, 2 H), 8.02 (s, 1 H), 8.04 (d, J ) 2.2 Hz,
1 H), 8.11 (d, J ) 9.0 Hz, 1 H), 8.47 (d, J ) 8.2 Hz, 2 H), 8.55
(d, J ) 5.7 Hz, 1 H), 8.89 (s, 1 H), 9.33 (s, 1 H). Anal.
(C20H12F3N3O): C, H, N.
6-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)isoquino-
line (27b). Following method B, isoquinolin-6-ol (0.060 g, 0.41
mmol) and 3 (0.18 g, 0.74 mmol), after purification of the crude
product by silica gel chromatography (gradient: 10-50% EtOAc
in hexanes), afforded 27b (0.058 g, 39%) as an off-white solid.
Mp: 195-196 °C. MS (ESI, pos. ion) m/z: 368 (M + 1). 1H NMR
(400 MHz, DMSO-d6) δ 7.64 (dd, J ) 2.3, 8.8 Hz, 1 H), 7.85 (d,
J ) 5.9 Hz, 1 H), 7.89 (d, J ) 2.3 Hz, 1 H), 7.94 (d, J ) 8.4 Hz,
2 H), 8.03 (d, J ) 1.0 Hz, 1 H), 8.26 (d, J ) 9.0 Hz, 1 H), 8.47 (d,
J ) 8.2 Hz, 2 H), 8.54 (d, J ) 5.7 Hz, 1 H), 8.91 (d, J ) 1.0 Hz,
1 H), 9.37 (s, 1 H). Anal. Calcd for C20H12F3N3O: C, 65.40; H,
3.29; N, 11.44. Found: C, 64.87; H, 3.27; N, 11.18.
6-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)quino-
line (27c). Following method A, 6-hydroxyquinoline (0.11 g,
0.77 mmol) and 3 (0.20 g, 0.77 mmol), after purification of the
crude product by silica gel chromatography (2:1 hexane/EtOAc),
provided 27c (0.14 g, 48%) as a white solid. Mp: 165-166 °C.
MS (ESI, pos. ion) m/z: 368 (M + 1). 1H NMR (400 MHz,
CDCl3) δ 7.43 (d, J ) 1.2 Hz, 1 H), 7.47 (dd, J ) 4.3, 8.2 Hz, 1
H), 7.59 (dd, J ) 2.5, 9.2 Hz, 1 H), 7.67 (d, J ) 2.7 Hz, 1 H),
7.80 (d, J ) 8.2 Hz, 2 H), 8.18 (dd, J ) 1.6, 8.2 Hz, 1 H), 8.22 (d,
J ) 8.2 Hz, 2 H), 8.23 (d, J ) 9.0 Hz, 1 H), 8.89 (d, J ) 1.2 Hz,
1 H), 8.96 (dd, J ) 1.6, 4.3 Hz, 1 H). Anal. (C20H12F3N3O):
C, H, N.
3-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)quino-
line (27d). Following method A, quinolin-3-ol (0.15 g, 1.1 mmol)
and 3 (0.30 g, 1.2 mmol), after recrystallization of the crude product
from EtOAc in hexane, provided 27d (0.25 g, 64%) as a white
solid. Mp: 208-209 °C. MS (ESI, pos. ion) m/z: 368 (M + 1).
1H NMR (400 MHz, DMSO-d6) δ 7.69 (ddd, J ) 1.2, 7.0, 8.0 Hz,
1 H), 7.81 (ddd, J ) 1.4, 7.0, 8.5 Hz, 1 H), 7.95 (d, J ) 8.2 Hz,
2 H), 8.03 (d, J ) 7.8 Hz, 1 H), 8.08-8.12 (m, 2 H), 8.35 (d, J )
2.7 Hz, 1 H), 8.49 (d, J ) 8.2 Hz, 2 H), 8.90 (d, J ) 1.0 Hz, 1 H),
8.93 (d, J ) 2.5 Hz, 1 H). Anal. (C20H12F3N3O‚0.1H2O): C, H, N.
3-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)isoquino-
line (27e). Following method A, 3-hydroxyisoquinoline (1.6 g, 11
mmol) and 3 (3.9 g, 15 mmol), after purification of the crude
product by silica gel chromatography (gradient: 1-2.5% 2 M NH3
in MeOH/CH2Cl2), then repeat purification by silica gel chroma-
tography (gradient: 0.5-1.8% 2 M NH3 in MeOH/CH2Cl2), and
washing the product with 10% EtOAc in hexane, afforded 27e (0.10

3524 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 15
Doherty et al.
g, 2.5%) as a tan solid. Mp: 157-160 °C. MS (ESI, pos. ion) m/z:
4-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)isoquino-
line (28f). Following method B, isoquinolin-4-ol (0.060 g, 0.41
mmol) and 3 (0.18 g, 0.74 mmol), after purification of the crude
product by silica gel chromatography (gradient: 20-70% EtOAc
in hexanes), provided 28f (0.048 g, 32%) as a yellow solid. Mp:
1
368 (M + 1). H NMR (400 MHz, DMSO-d6) δ 7.70 (ddd, J )
1.0, 7.0, 8.0 Hz, 1 H), 7.80 (s, 1 H), 7.84 (ddd, J ) 1.2, 7.0, 8.2
Hz, 1 H), 7.93 (d, J ) 8.2 Hz, 2 H), 7.99 (d, J ) 0.98 Hz, 1 H),
8.03 (d, J ) 8.0 Hz, 1 H), 8.22 (d, J ) 7.6 Hz, 1 H), 8.46 (d, J )
8.0 Hz, 2 H), 8.91 (d, J ) 0.98 Hz, 1 H), 9.23 (s, 1 H). Anal.
(C20H12F3N3O‚0.1H2O): C, H, N.
1
206-211 °C. MS (ESI, pos. ion) m/z: 368 (M + 1). H NMR
(400 MHz, DMSO-d6) δ 7.76-7.84 (m, 2 H), 7.85-7.89 (m, 1
H), 7.96 (d, J ) 8.2 Hz, 2 H), 8.17 (d, J ) 1.0 Hz, 1 H), 8.27-
8.30 (m, 1 H), 8.49 (d, J ) 8.2 Hz, 2 H), 8.55 (s, 1 H), 8.80 (d, J
) 1.2 Hz, 1 H), 9.34 (s, 1 H). Anal. Calcd for C20H12F3N3O: C,
65.40; H, 3.29; N, 11.44. Found: C, 64.83; H, 3.21; N, 11.33.
1-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)isoquino-
line (28g). Following method A, isoquinolin-1-ol (0.20 g, 1.4 mmol)
and 3 (0.32 g, 1.2 mmol), after purification of the crude product
by silica gel chromatography (gradient: 50-80% EtOAc in
hexanes), afforded 28g (0.17 g, 37%) as a white solid. Mp: 186-
187 °C. MS (ESI, pos. ion) m/z: 368 (M + 1). 1H NMR (400 MHz,
DMSO-d6) δ 6.89 (d, J ) 7.8 Hz, 1 H), 7.63 (ddd, J ) 1.2, 6.8,
8.0 Hz, 1 H), 7.77 (d, J ) 7.4 Hz, 1 H), 7.84 (ddd, J ) 1.2, 6.8,
8.0 Hz, 1 H), 7.98 (d, J ) 8.2 Hz, 2 H), 8.03 (d, J ) 7.6 Hz, 1 H),
8.35 (d, J ) 7.4 Hz, 1 H), 8.44 (d, J ) 8.2 Hz, 2 H), 8.81 (d, J )
0.98 Hz, 1 H), 9.39 (s, 1 H). Anal. (C20H12F3N3O): C, H, N.
8-[6-(4-Trifluoromethyl-phenyl)-pyrimidin-4-yloxy]-1H-quin-
olin-2-one (29a). Following method D, 2,8-quinolinediol (0.075
g, 0.46 mmol) and 3 (0.10 g, 0.39 mmol) provided the crude
product. The solids were filtered, washed with EtOAc, and dried
in vacuo for 16 h to afford 29a (140 mg, 94%) as long white
needles. Mp: 312 °C. HRMS (TOF, pos. ion.) calcd for
C20H13F3N3O2+, 384.0954; found, 384.0951. 1H NMR (400 MHz,
DMSO-d6) δ 7.27 (dd, J ) 8.2, 9.0 Hz, 2 H), 7.65 (t, J ) 8.2 Hz,
1 H), 7.93 (d, J ) 8.2 Hz, 2 H), 8.00 (s, 1 H), 8.07 (s, 1 H), 8.46
(d, J ) 8.2 Hz, 2 H), 8.79 (s, 1 H), 12.64 (s, 1 H). Anal.
(C20H12F3N3O2): C, H, N, F.
2-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)quino-
line (27f). Following method A, quinolin-2-ol (0.30 g, 2.1 mmol)
and 3 (0.58 g, 2.2 mmol), after purification of the crude product
by silica gel chromatography (40% EtOAc in hexanes), followed
by recrystallization from 20% EtOAc/hexane, afforded 27f (0.20
g, 26%) as a white solid. Mp: 145-149 °C. MS (ESI, pos. ion)
m/z: 368 (M + 1). 1H NMR (400 MHz, DMSO-d6) δ 7.49 (d, J )
8.6 Hz, 1 H), 7.64 (ddd, J ) 1.2, 7.0, 8.2 Hz, 1 H), 7.78 (ddd, J )
1.4, 6.9, 8.3 Hz, 1 H), 7.86 (d, J ) 8.4 Hz, 1 H), 7.94 (d, J ) 8.4
Hz, 2 H), 8.08 (d, J ) 7.6 Hz, 1 H), 8.11 (d, J ) 0.98 Hz, 1 H),
8.47 (d, J ) 8.2 Hz, 2 H), 8.58 (d, J ) 8.6 Hz, 1 H), 8.98 (d, J )
0.78 Hz, 1 H). Anal. (C20H12F3N3O): C, H, N.
8-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)quino-
line (28a). Following method A, 8-hydroxyqunoline (0.17 g, 1.2
mmol) and 3 (0.30 g, 1.2 mmol), after purification of the crude
product by silica gel chromatography (4:1 hexanes/EtOAc), pro-
vided 28a (0.30 g, 70%) as a white solid. Mp: 155-157 °C. MS
(ESI, pos. ion) m/z: 368 (M + 1). 1H NMR (400 MHz, CDCl3) δ
7.46 (dd, J ) 4.1, 8.4 Hz, 1 H), 7.59-7.67 (m, 3 H), 7.79 (d, J )
8.2 Hz, 2 H), 7.83 (dd, J ) 1.6, 7.8 Hz, 1 H), 8.23 (d, J ) 8.2 Hz,
2 H), 8.26 (dd, J ) 1.6, 8.2 Hz, 1 H), 8.74 (d, J ) 1.2 Hz, 1 H),
8.85 (dd, J ) 1.8, 4.1 Hz, 1 H). Anal. (C20H12F3N3O): C, H, N.
8-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)isoquino-
line (28b). Following method A, isoquinolin-8-ol (0.060 g, 0.41
mmol) and 3 (0.12 g, 0.45 mmol), after purification of the crude
product by silica gel chromatography (gradient: 20-70% EtOAc
in hexane), afforded 28b (0.094 g, 63%) as a white solid. Mp: 194-
195 °C. MS (ESI, pos. ion) m/z: 368 (M + 1). 1H NMR (400 MHz,
DMSO-d6) δ 7.62 (dd, J ) 0.7, 7.5 Hz, 1 H), 7.88 (t, J ) 7.9 Hz,
1 H), 7.95-7.99 (m, 4 H), 8.16 (d, J ) 1.0 Hz, 1 H), 8.49 (d, J )
8.2 Hz, 2 H), 8.60 (d, J ) 5.7 Hz, 1 H), 8.82 (d, J ) 1.0 Hz, 1 H),
9.30 (s, 1 H). Anal. (C20H12F3N3O): C, H, N.
5-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)isoquino-
line (28c). Following method A, isoquinolin-5-ol (0.45 g, 3.1 mmol)
and 3 (0.58 g, 2.2 mmol), after purification of the crude product
by recrystallization from 40% EtOAc in hexanes, afforded 28c (0.51
g, 61%) as an off-white solid. Mp: 183-185 °C. MS (ESI, pos.
ion) m/z: 368 (M + 1). 1H NMR (400 MHz, DMSO-d6) δ 7.70 (d,
J ) 5.9 Hz, 1 H), 7.73-7.83 (m, 2 H), 7.96 (d, J ) 8.4 Hz, 2 H),
8.07-8.18 (m, 2 H), 8.49 (d, J ) 8.2 Hz, 2 H), 8.52 (d, J ) 6.1
Hz, 1 H), 8.79 (d, J ) 0.98 Hz, 1 H), 9.45 (s, 1 H). Anal.
(C20H12F3N3O): C, H, N.
5-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)quino-
line (28d). Following method A, 5-hydroxyquinoline (0.15 g, 1.0
mmol) and 3 (0.26 g, 1.0 mmol), after purification of the crude
product by silica gel chromatography (5:1 hexanes/EtOAc), pro-
vided 28d (0.26 g, 70%) as a white solid. Mp: 159-161 °C. MS
(ESI, pos. ion) m/z: 368 (M + 1). 1H NMR (400 MHz, CDCl3) δ
7.40-7.46 (m, 2 H), 7.48 (d, J ) 1.2 Hz, 1 H), 7.80 (d, J ) 8.6
Hz, 2 H), 7.83 (d, J ) 7.4 Hz, 1 H), 8.13 (d, J ) 8.6 Hz, 1 H),
8.22 (d, J ) 8.2 Hz, 2 H), 8.25 (d, J ) 8.6 Hz, 1 H), 8.83 (d, J )
1.2 Hz, 1 H), 8.99 (dd, J ) 1.8, 4.1 Hz, 1 H). Anal.
(C20H12F3N3O): C, H, N.
2-Methoxy-8- [6-(4-trifluoromethyl-phenyl)-pyrimidin-4-yloxy]-
quinoline (29b). Following method B, 2-methoxy-quinolin-8-ol
(0.090 g, 0.51 mmol) and 3 (0.19 g, 0.71 mmol), after purification
of the crude product by silica gel chromatography (1:9 EtOAc/
hexanes), provided 29b (0.17 g, 85%) as a white solid. Mp: 159-
161 °C. MS (ESI, pos. ion) m/z: 398 (M + 1). 1H NMR (400 MHz,
CDCl3) δ 3.58 (s, 3 H), 6.92 (d, J ) 8.6 Hz, 1 H), 7.45-7.51 (m,
2 H), 7.56-7.61 (m, 1 H), 7.68-7.74 (m, 1 H), 7.80 (d, J ) 8.2
Hz, 2 H), 8.06 (d, J ) 9.0 Hz, 1 H), 8.22 (d, J ) 8.2 Hz, 2 H),
8.79 (s, 1 H). Anal. (C20H11ClF3N3O): C, H, N.
Methyl-{8-[6-(4-trifluoromethyl-phenyl)-pyrimidin-4-yloxy]-
quinolin-2-yl}-amine (29c). To a solution of 2-chloroquinolin-8-
ol (5; 0.18 g, 1 mmol) in 1,4-dioxane (3 mL) was added 2 M
methylamine in THF (10 mL, 20 mmol). The reaction mixture was
stirred and heated in a microwave (Smith Synthesizer, Personal
Chemistry, Inc., Upssala, Sweden) at 200 °C for 12 min. The
reaction mixture was allowed to cool to room temperature, then
partitioned between EtOAc and 1 N NaOH. The aqueous layer was
extracted with EtOAc (3×), and the combined organic extracts were
washed with satd NaCl, dried over Na2SO4, filtered, and concen-
trated in vacuo. The crude product was recrystallized from MeOH/
H2O to provide 2-methylamino-quinolin-8-ol (6) as a light-yellow
solid (0.12 g, 71%). MS (ESI, pos. ion) m/z: 175 (M + 1).
Following method B, 2-methylamino-quinolin-8-ol (6; 0.12 g,
0.71 mmol) and 3 (0.22 g, 0.85 mmol), after purification of the
crude product by silica gel chromatography (1:5 EtOAc in hexanes)
and recrystallization from EtOAc in hexanes, provided 29c (0.038
g, 14%) as a white crystalline solid. Mp: 163-165 °C. MS (ESI,
4-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-yloxy)quino-
line (28e). Following method A, quinolin-4-ol (0.20 g, 1.4 mmol)
and 3 (0.32 g, 1.2 mmol), after purification of the crude product
by silica gel chromatography (gradient: 0-100% EtOAc in
hexanes), afforded 28e (0.31 g, 68%) as a white solid. Mp: 209-
210 °C. MS (ESI, pos. ion) m/z: 368 (M + 1). 1H NMR (400 MHz,
DMSO-d6) δ 6.31 (d, J ) 8.0 Hz, 1 H), 7.49 (ddd, J ) 0.8, 7.0,
7.8 Hz, 1 H), 7.67 (ddd, J ) 1.6, 6.8, 8.4 Hz, 1 H), 7.73 (d, J )
8.4 Hz, 1 H), 7.99 (d, J ) 8.2 Hz, 2 H), 8.24 (dd, J ) 1.5, 7.9 Hz,
1 H), 8.35 (d, J ) 7.8 Hz, 1 H), 8.54 (d, J ) 8.2 Hz, 2 H), 8.63 (s,
1 H), 9.45 (s, 1 H). Anal. (C20H12F3N3O): C, H, N.
1
pos. ion) m/z: 397 (M + 1). H NMR (400 MHz, CDCl3) δ 2.64
(d, J ) 4.7 Hz, 3 H), 6.60 (d, J ) 9.0 Hz, 1 H), 7.24-7.28 (m, 1
H), 7.44-7.48 (m, 2 H), 7.52-7.57 (m, 1 H), 7.77 (d, J ) 8.2 Hz,
2 H), 7.83 (d, J ) 9.0 Hz, 1 H), 8.18 (d, J ) 8.2 Hz, 2 H), 8.79 (s,
1 H). Anal. Calcd for C21H15F3N4O: C, 63.63; H, 3.81; N, 14.14.
Found: C, 62.98; H, 3.83; N, 13.77.
N,N-Dimethyl-8-(6-(4-(trifluoromethyl)phenyl)pyrimidin-4-
yloxy)quinolin-2-amine (29d). Following the same procedure
described for compound 6, 2-chloro-quinolin-8-ol (5; 0.18 g, 1.0
mmol) and 2 M dimethylamine in THF provided 2-(dimethylami-

Head: TRPV1 Antagonists: Selection of a Clinical Candidate
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 15 3525
no)quinolin-8-ol (7; 0.10 g, 56%) as a light-tan solid. MS (ESI,
pos. ion) m/z: 189 (M + 1).
for 30 h. The reaction mixture was diluted with H2O and extracted
with EtOAc (3×). The combined organic extract was dried over
Na2SO4, filtered, and concentrated in vacuo. The resulting dark red
solid was recrystallized from hexanes to yield 2-methoxy-6-
nitrobenzenamine (24 g, 87%) as orange needles. MS (ESI, pos.
ion) m/z: 169 (M + 1). H NMR (400 MHz, CDCl3) δ 3.94 (s, 3
H), 6.46 (s, 2 H), 6.63 (dd, J ) 7.8, 8.2 Hz, 1 H), 6.91 (d, J ) 7.8
Hz, 1 H), 7.75 (dd, J ) 1.2, 8.6 Hz, 2 H).
Following method B, 2-(dimethylamino)quinolin-8-ol (7; 0.10
g, 0.53 mmol) and 3 (0.14 g, 0.53 mmol), after purification of the
crude product by silica gel chromatography (1:5 of EtOAc in
hexanes) and recrystallization from EtOAc in hexanes, provided
29d (0.050 g, 23%) as a white solid. Mp: 170-182 °C. MS (ESI,
1
1
pos. ion) m/z: 411 (M + 1). H NMR (400 MHz, CDCl3) δ 2.87
(s, 6 H), 6.85 (d, J ) 9.2 Hz, 1 H), 7.24 (t, J ) 8.0 Hz, 1 H), 7.43
(d, J ) 1.2 Hz, 1 H), 7.47 (dd, J ) 1.2, 8.0 Hz, 1 H), 7.55 (dd, J
) 1.2, 8.0 Hz, 1 H), 7.76 (d, J ) 8.0 Hz, 2 H), 7.90 (d, J ) 9.2
Hz, 1 H), 8.16 (d, J ) 8.0 Hz, 2 H), 8.79 (d, J ) 1.0 Hz, 1 H).
Anal. (C22H17F3N4O): C, H, N.
A mixture of 2-methoxy-6-nitrobenzenamine (4.6 g, 27 mmol),
iron powder (11 g, 190 mmol), EtOH (130 mL), and H2O (10 mL)
was stirred at 50 °C. A solution of 12 M aq HCl (1.7 mL) was
added to the reaction mixture, dropwise with stirring. The reaction
mixture was stirred at reflux for 3 h then allowed to cool to room
temperature. The reaction mixture was neutralized with 1 N NaOH
and filtered through Celite. The filtrate was concentrated in vacuo
to afford a residue. The residue was partitioned between CH2Cl2
and satd aq NaHCO3. The aqueous phase was extracted with
CH2Cl2 (3×) and the combined organic layers were concentrated
in vacuo. The residue was dissolved in EtOH (30 mL) and treated
with concd H2SO4 until no more precipitate was formed. The
precipitate was collected by suction filtration, washed with EtOH,
and dried in vacuo for 20 h at room temperature to afford
3-methoxy-benzene-1,2-diamine hydrogen sulfate salt (12; 6.3 g,
97%) as an off-white powder. MS (ESI, pos. ion) m/z: 139 (M -
HSO4-). 1H NMR (400 MHz, D2O) δ 3.84 (s, 3 H), 6.82 (dd, J )
1.2, 8.2 Hz, 1 H), 6.93 (dd, J ) 1.2, 8.6 Hz, 1 H), 7.21 (t, J ) 8.2
Hz, 1 H).
8-[6-(4-Trifluoromethyl-phenyl)-pyrimidin-4-yloxy]-quinazo-
lin-2-ylamine (30). To a room temperature solution of 3-methoxy-
2-nitro-benzaldehyde (8; 15 g, 81 mmol) and NH4Cl (4.4 g, 82
mmol) in 80% aq MeOH (250 mL) was added iron dust (21 g, 370
mmol). The reaction mixture was stirred at 60 °C for 2 h. After
allowing to cool to room temperature, the reaction mixture was
filtered through a pad of Celite. The filter cake was washed with
MeOH and the combined filtrate was concentrated in vacuo to
afford an aqueous mixture. The mixture was extracted with
CH2Cl2 (3×), and the combined organic layers were washed with
satd NaCl, dried over Na2SO4, filtered, and concentrated in vacuo.
Purification of the crude product by silica gel chromatography
(gradient: 0 to 20% EtOAc in hexanes) provided 2-amino-3-
1
methoxy-benzaldehyde (9; 3.8 g, 31%) as a yellow oil. H NMR
(400 MHz, CDCl3) δ 3.88 (s, 3 H), 6.40 (br s, 2 H), 6.68 (t, J )
8.0 Hz, 1 H), 6.88 (dd, J ) 1.2, 8.0 Hz, 1 H), 7.12 (dd, J ) 1.2,
8.0 Hz, 1 H), 9.89 (s, 1 H).
A solution of 3-methoxy-benzene-1,2-diamine hydrogen sulfate
salt (12; 4.1 g, 17 mmol) in EtOH (21 mL) and H2O (48 mL) was
neutralized by careful addition of solid NaHCO3. The mixture was
treated with ethyl glyoxylate solution (50% in toluene, 3.8 mL, 19
mmol) then stirred at reflux for 1 h. The reaction was allowed to
cool to room temperature and partitioned between satd aq NH4Cl
and 25% i-PrOH/CHCl3. The aqueous layer was extracted with 25%
i-PrOH/CHCl3 (3×). The combined organic layers were dried over
Na2SO4, filtered, and concentrated in vacuo. Silica gel chromatog-
raphy of the crude residue (gradient: 0-2.5% MeOH/CH2Cl2)
afforded two products. (1) 8-Methoxy-1H-quinoxalin-2-one (13; 1.1
g, 37%) as an off-white powder. MS (ESI, pos. ion) m/z: 177 (M
A mixture of 2-amino-3-methoxy-benzaldehyde (9; 3.8 g, 25
mmol), guanidine hydrochloride (4.9 g, 51 mmol), Na2CO3 (5.4 g,
51 mmol), and decalin (55 mL) was stirred at 190 °C for 2.5 h.
The hot solution was decanted from the solids and allowed to cool
to room temperature. The resultant suspension was diluted with
hexane. The solid was collected by suction filtration, washed with
hexane and pentane, and dried in vacuo to afford 2-amino-8-
methoxyquinazoline (2.2 g, 48%) as a yellow amorphous solid. MS
1
(ESI, pos ion.) m/z: 176 (M + 1). H NMR (400 MHz, DMSO-
d6) δ 3.86 (s, 3 H), 6.87 (br s, 2 H), 7.09-7.16 (m, 2 H), 7.29-
1
+ 1). H NMR (400 MHz, CDCl3) δ 4.01 (s, 3 H), 7.03 (d, J )
7.36 (m, 1 H), 9.05 (s, 1 H).
8.2 Hz, 1 H), 7.26 (t, J ) 8.2 Hz, 1 H), 7.49 (dd, J ) 1.2, 8.2 Hz,
1 H), 8.32 (s, 1 H), 9.28 (s, 1 H). (2) 5-Methoxy-1H-quinoxalin-
2-one (0.82 g, 27%) as an off-white powder. MS (ESI, pos. ion)
To a suspension of NaH (60% dispersion in mineral oil, 1.4 g,
35 mmol) in DMF (100 mL), stirred at 0 °C, was added ethanethiol
(5.0 mL, 67 mmol). [Gas evolution was observed.] After the reaction
mixture was allowed to warm to room temperature, 2-amino-8-
methoxyquinazoline (1.5 g, 8.6 mmol) was added and the mixture
was stirred at reflux for 4 h. The reaction was allowed to cool to
room temperature and the solvent was removed in vacuo. The crude
residue was azeotroped in vacuo with H2O and purified by silica
gel chromatography (gradient: 0-50% 2 M NH3 in MeOH/
CH2Cl2) to give 2-amino-8-hydroxyquinazoline (10; 500 mg, 58%)
as a pale-green amorphous solid. MS (ESI, pos ion.) m/z: 162 (M
1
m/z: 177 (M + 1). H NMR (400 MHz, CDCl3) δ 4.06 (s, 3 H),
6.82 (d, J ) 7.4 Hz, 1 H), 6.95 (dd, J ) 0.8, 8.2 Hz, 1 H), 7.51 (t,
J ) 8.2 Hz, 1 H), 8.33 (s, 1 H), 12.22 (s, 1 H).
A mixture of 8-methoxy-1H-quinoxalin-2-one (13; 0.62 g, 3.5
mmol) and POCl3 (6.0 mL, 64 mmol) was stirred at 105 °C in an
oil bath for 4 h. The reaction mixture was allowed to cool to room
temperature and the excess POCl3 was removed by vacuum
distillation. The residue was partitioned between satd aq NaHCO3
and CH2Cl2 and stirred for 3 h to quench residual POCl3. The
aqueous layer was extracted with CH2Cl2 (3×). The combined
organic extract was dried over Na2SO4 and filtered through a pad
of silica gel, eluting with EtOAc. The filtrate was concentrated in
vacuo to afford 2-chloro-8-methoxyquinoxaline (0.66 g, 97%) as a
1
+ 1). H NMR (400 MHz, DMSO-d6) δ 6.72 (br s, 2 H), 7.02-
7.07 (m, 2 H), 7.21-7.26 (m, 1 H), 9.05 (s, 1 H), 9.19 (s, 1 H).
Following method D, 2-amino-8-hydroxyquinazoline (10; 220
mg, 1.3 mmol) and 3 (390 mg, 1.5 mmol), after purification of the
crude product by silica gel chromatography (gradient: 0-40%
EtOAc in hexanes) and a second purification by silica gel
chromatography (gradient: 0-5% 2 M NH3 in MeOH/CH2Cl2),
afforded 30 (110 mg, 22%) as a white solid. Mp: 251-253 °C.
1
tan solid. MS (ESI, pos. ion) m/z: 195 (M + 1). H NMR (400
MHz, CDCl3) δ 4.10 (s, 3 H), 7.16 (dd, J ) 3.6, 5.6 Hz, 1 H), 7.71
(d, J ) 3.6 Hz, 1 H), 7.71 (d, J ) 5.6 Hz, 1 H), 8.80 (s, 1 H).
1
A mixture of 2-chloro-8-methoxyquinoxaline (0.42 g, 2.2 mmol)
and CuI (0.21 g, 1.1 mmol) in 28-30% NH4OH (1.5 mL) was
stirred and heated in a microwave (Smith Synthesizer, Personal
Chemistry, Inc., Upssala, Sweden) at 140 °C for 10 min. The
reaction mixture was diluted with H2O and the solids were collected
by suction filtration and washed with copious amounts of H2O.
The solid was dried in vacuo for 20 h at room temperature to afford
8-methoxyquinoxalin-2-ylamine (0.27 g, 70%) as a brown powder.
MS (ESI, pos. ion) m/z: 176 (M + 1). 1H NMR (400 MHz, DMSO-
d6) δ 3.88 (s, 3 H), 6.98 (s, 2 H), 7.04 (d, J ) 8.0 Hz, 1 H), 7.24
(t, J ) 8.0 Hz, 1 H), 7.34 (d, J ) 8.4 Hz, 1 H), 8.27 (s, 1 H).
MS (ESI, pos ion.) m/z: 384.1 (M + 1). H NMR (400 MHz,
CDCl3) δ 5.17 (s, 2 H), 7.35 (t, J ) 8.0 Hz, 1 H), 7.54 (d, J ) 0.8
Hz, 1 H), 7.59 (dd, J ) 1.6, 7.6 Hz, 1 H), 7.69 (dd, J ) 1.6, 8.0
Hz, 1 H), 7.79 (d, J ) 8.0 Hz, 2 H), 8.22 (d, J ) 8.0 Hz, 2 H),
8.77 (d, J ) 0.8 Hz, 1 H), 9.10 (s, 1 H). Anal. (C19H12F3N5O):
C, H, N.
8-[6-(4-Trifluoromethyl-phenyl)-pyrimidin-4-yloxy]-quinoxa-
lin-2-ylamine (31). A mixture of 2-amino-3-nitrophenol (11; 25
g, 160 mmol) and K2CO3 (27 g, 195 mmol) in DMF (65 mL) was
stirred at room temperature for 1 h. Methyl iodide (12 mL, 195
mmol) was added, and the reaction was stirred at room temperature

3526 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 15
Doherty et al.
To a suspension of 8-methoxyquinoxalin-2-ylamine (0.12 g, 0.68
mmol) in benzene (10 mL) was added AlCl3 (0.82 g, 6.2 mmol),
and the mixture was stirred at reflux for 2 h. The reaction was
allowed to cool to room temperature and quenched by careful
addition of satd aq NaHCO3. The resulting mixture was extracted
with 25% i-PrOH/CHCl3 (5×). The combined organic extract was
dried over Na2SO4, filtered, and concentrated in vacuo. Purification
of the crude residue by silica gel chromatography (gradient:
0-7.5% MeOH/CH2Cl2) afforded 3-aminoquinoxalin-5-ol (14;
0.050 g, 46%) as a brown powder. MS (ESI, pos. ion) m/z: 162
lin-2-one (17; 0.75 g, 3.9 mmol) was demethylated to provide
3-amino-8-hydroxy-1H-quinoxalin-2-one (19). MS (ESI, pos. ion)
m/z: 178 (M + 1). H NMR (400 MHz, D2O) δ 6.71 (d, J ) 8.2
Hz, 1 H), 6.79 (d, J ) 8.2 Hz, 1 H), 7.22 (t, J ) 8.0 Hz, 1 H).
1
Following method D, 3-amino-8-hydroxy-1H-quinoxalin-2-one
(19; 0.69 g, 3.9 mmol) and 3 (1.0 g, 3.9 mmol), after purification
of the crude product by silica gel chromatography (gradient: 0-
2.5% MeOH in CH2Cl2), afforded 33 (0.42 g, 27%) as a white pow-
+
der. Mp: 288 °C. HRMS (TOF, pos. ion) calcd for C19H13F3N5O2
,
400.1016; found, 400.1015. 1H NMR (400 MHz, DMSO-d6) δ 7.03
(d, J ) 8.0 Hz, 1 H), 7.16 (t, J ) 8.0 Hz, 1 H), 7.20 (br s, 2 H),
7.24 (d, J ) 8.0 Hz, 1 H), 7.93 (s, 1 H), 7.95 (d, J ) 8.4 Hz, 2 H),
8.46 (d, J ) 8.0 Hz, 2 H), 8.81 (s, 1 H), 12.11 (s, 1 H). Anal.
(C19H12F3N5O2): C, H, N, F.
1
(M + 1). H NMR (400 MHz, DMSO-d6) δ 6.83 (s, 2 H), 6.94
(dd, J ) 1.2, 7.4 Hz, 1 H), 7.14 (t, J ) 8.0 Hz, 1 H), 7.24 (dd, J
) 1.4, 8.4 Hz, 1 H), 8.29 (s, 1 H), 9.13 (s, 1 H).
Following method D, 3-aminoquinoxalin-5-ol (14; 0.069 g, 0.43
mmol) and 3 (0.13 g, 0.51 mmol), after purification of the crude
product by silica gel chromatography (gradient: 0-75% EtOAc
in hexanes), afforded 31 (0.11 g, 66%) as an off-white powder.
Mp: 214-216 °C. HRMS (TOF, pos. ion.) calcd for C19H13F3N5O+,
Acknowledgment. The authors thank Paul Reider, Randy
Hungate, Jean-Claude Louis, and Chris Fibiger for their support
of this research program. Thanks go to James Davis, Gal Hever,
Rongzhen Kuang, Jue Wang, Dawn Zhu, and Shoushu Jiao for
conducting the in vivo pharmacology studies and to Nuria
Tamayo for helpful discussions regarding the SAR in this series.
Finally, we acknowledge all of our colleagues on the TRPV1
research team (pharmaceutics, toxicology, PKDM, process
research and development, analytical support, medical sciences)
who helped advance AMG 517 forward into development, with
special thanks to Annette Bak and David Hovland.
1
384.1067; found, 384.1063. H NMR (400 MHz, CDCl3) δ 4.95
(s, 2 H), 7.49 (m, 3 H), 7.78 (d, J ) 8.4 Hz, 2 H), 7.91 (dd, J )
2.4, 6.8 Hz, 1 H), 8.20 (d, J ) 8.4 Hz, 2 H), 8.32 (s, 1 H), 8.77 (s,
1 H). Anal. Calcd for C19H12F3N5O: C, 59.53; H, 3.16; F, 14.87;
N, 18.27. Found: C, 58.26; H, 3.24; F, 14.25; N, 17.63.
3-Amino-5-(6-(4-(trifluoromethyl)phenyl)pyrimidin-4-yloxy)-
quinoxalin-2(1H)-one (32). To a suspension of 3-methoxy-
benzene-1,2-diamine hydrogen sulfate salt (12; 2.4 g, 10 mmol) in
EtOH (15 mL) and H2O (1 mL) was added solid NaHCO3 (1.7 g,
20 mmol). When gas evolution was complete, ethoxy-imino-acetic
acid ethyl ester14 (15; 1.6 g, 11 mmol) was added, and the mixture
was stirred at room temperature for 16 h. The reaction mixture
was diluted with satd aq NaHCO3 and extracted with 25% i-PrOH/
CHCl3 (5×). The combined organic extract was dried over
Na2SO4, filtered, and concentrated in vacuo. Purification of the
residue by silica gel chromatography (gradient: 0-5% MeOH/CH2-
Cl2) afforded two products: (1) 3-Amino-5-methoxy-1H-quinoxalin-
2-one (16; 0.44 g, 23%) as a light brown powder. MS (ESI, pos.
ion) m/z: 192 (M + 1). 1H NMR (400 MHz, DMSO-d6) δ 3.81 (s,
3 H), 6.73 (d, J ) 8.6 Hz, 1 H), 6.75 (d, J ) 8.2 Hz, 1 H), 7.04
(t, J ) 8.2 Hz, 1 H), 7.04 (br s, 2 H), 12.08 (s, 1 H). (2) 3-Amino-
8-methoxy-1H-quinoxalin-2-one (17; 0.75 g, 39%) as a pale brown
Supporting Information Available: Combustion analysis
results for final compounds 21-26, 27a-f, 28a-g, 29a-d, and
30-33. This material is available free of charge via the Internet at
http://pubs.acs.org.
References
(1) (a) Appendino, G.; Munoz, E.; Fiebich, B. L. TRPV1 (Vanilloid
receptor, Capsaicin receptor) Agonists and Antagonists. Expert Opin.
Ther. Pat. 2003, 13, 1825-1837. (b) Rami, K. H.; Gunthorpe, M. J.
The Therapeutic Potential of TRPV1 (VR1) Antagonists: Clinical
Answers Await. Drug DiscoVery Today 2004, 1, 97-104. (c)
Breitenbucher, G.; Chaplan, S. R.; Carruters, N. I. The TRPV1
Vanilloid Receptor: A Target for Therapeutic Intervention. In Ann.
Rep. Med. Chem.; Doherty, A. M., Ed.; Elsevier Inc.: San Diego,
2005; Vol. 40, pp 185-198.
(2) Norman, M. H.; Fotsch, C.; Doherty, E. M.; Bo, Y.; Chen, N.;
Chakrabarti, P.; Gavva, N. R.; Nishimura, N.; Nixey, T.; Ognyanov,
V. I.; Rzasa, R.; Stec, M.; Surapaneni, S.; Tamir, R.; Viswanadhan,
V.; Zhu, J.; Treanor, J. J. S. Novel Vanilloid Receptor-1 Antago-
nists: 1. Conformationally Restricted Analogues of trans-Cinnamides.
J. Med. Chem. 2007, 50, 3497-3514.
(3) Chakrabarti, P. P.; Chen, N.; Doherty, E. M.; Dominguez, C.; Falsey,
J. R.; Fotsh, C. H.; Hulme, C.; Katon, J.; Nixey, T.; Norman, M. H.;
Ognyanov, V. I.; Pettus, L. H.; Rzasa, R. M.; Stec, M.; Wang, H.-
L.; Zhu, J. (Amgen Inc.) Preparation of (aryloxy)pyrimidine and
(aryloxy)pyridazine as vanilloid receptor ligands. WO 04014871,
2004.
(4) Csikos, E.; Gonczi, C.; Podanyi, B.; Toth, G.; Hermecz, I. Regiose-
lectivity in preparation of unsymmetrically substituted 3-aminoqui-
noxalin-2(1H)-ones. J. Chem. Soc., Perkin Trans. 1 1999, 13, 1789-
1793.
(5) Doherty, E. M.; Fotsch, C.; Bo, Y.; Chakrabarti, P. P.; Chen, N.;
Gavva, N.; Han, N.; Kelly, M. G.; Kincaid, J.; Klionsky, L.; Liu,
Q.; Ognyanov, V. I.; Tamir, R.; Wang, X.; Zhu, J.; Norman, M. H.;
Treanor, J. J. S. Discovery of Potent, Orally Available Vanilloid
Receptor-1 Antagonists. Structure-Activity Relationship of N-Aryl
Cinnamides. J. Med. Chem. 2005, 48 (1), 71-90.
(6) This observation is also consistent with the results of 6,6-fused
heteroaromatic-containing urea TRPV1 antagonists reported by
Gomtsyan, A.; Bayburt, E. K.; Schmidt, R. G.; Zheng, G. Z.; Perner,
R. J.; Didomenico, S.; Koenig, J. R.; Turner, S.; Jinkerson, T.; Drizin,
I.; Hannick, S. M.; Macri, B. S.; McDonald, H. A.; Honore, P.;
Wismer, C. T.; Marsh, K. C.; Wetter, J.; Stewart, K. D.; Oie, T.;
Jarvis, M. F.; Surowy, C. S.; Faltynek, C. R.; Lee, C.-H. Novel
Transient Receptor Potential Vanilloid 1 Receptor Antagonists for
the Treatment of Pain: Structure-Activity Relationships for Ureas
with Quinoline, Isoquinoline, Quinazoline, Phthalazine, Quinoxaline,
and Cinnoline Moieties. J. Med. Chem. 2005, 48 (3), 744-752.
1
powder. MS (ESI, pos. ion) m/z: 192 (M + 1). H NMR (400
MHz, DMSO-d6) δ 3.86 (s, 3 H), 6.79 (d, J ) 7.8 Hz, 1 H), 6.90
(d, J ) 7.8 Hz, 1 H), 7.05 (t, J ) 8.0 Hz, 1 H), 7.07 (s, 2 H), 11.51
(s, 1 H).
To a suspension of 3-amino-5-methoxy-1H-quinoxalin-2-one (16;
0.47 g, 2.5 mmol) in benzene (25 mL) was added AlCl3 (0.97 g,
7.4 mmol), and the mixture was stirred at reflux for 2 h. The reaction
was quenched by careful addition of satd aq NaHCO3 and extracted
with 25% i-PrOH/CHCl3 (5×). The combined organic extract was
dried over Na2SO4, filtered, and concentrated in vacuo to afford
3-amino-5-hydroxy-1H-quinoxalin-2-one (18; 0.34 g, 78%) as a
1
brown powder. MS (ESI, pos. ion) m/z: 178 (M + 1). H NMR
(400 MHz, DMSO-d6) δ 6.58 (dd, J ) 1.2, 7.8 Hz, 1 H), 6.60 (dd,
J ) 1.2, 8.2 Hz, 1 H), 6.89 (s, 2 H), 6.92 (dd, J ) 7.8, 8.2 Hz, 1
H), 8.81 (s, 1 H), 12.04 (s, 1 H).
Following method C, 3-amino-5-hydroxy-1H-quinoxalin-2-one
(18; 0.33 g, 1.9 mmol) and 3 (0.49 g, 1.9 mmol), after purification
of the crude product by silica gel chromatography (gradient:
0-2.5% 2 M NH3 in MeOH/CH2Cl2), afforded 32 (0.28 g, 38%)
as an off-white solid. Mp: 334-335 °C (dec). HRMS (TOF, pos.
ion) calcd for C19H13F3N5O2+, 400.1016; found, 400.1012. 1H NMR
(400 MHz, DMSO-d6) δ 7.03 (d, J ) 7.6 Hz, 1 H), 7.10 (d, J )
6.8 Hz, 1 H), 7.12 (br s, 2 H), 7.16 (t, J ) 8.0 Hz, 1 H), 7.85
(s, 1 H), 7.91 (d, J ) 8.4 Hz, 2 H), 8.43 (d, J ) 8.4 Hz, 2 H), 8.78
(s, 1 H), 12.30 (s, 1 H). Anal. Calcd for C19H12F3N5O2‚1.75H2O:
C, 52.97; H, 3.63; F, 13.23; N, 16.25. Found: C, 53.07; H, 3.61;
F, 13.00; N, 16.20. HPLC analysis, method B: 95.9% at 215 nm;
96.7% at 254 nm; retention time 5.67 min.
3-Amino-8-(6-(4-(trifluoromethyl)phenyl)pyrimidin-4-yloxy)-
quinoxalin-2(1H)-one (33). Following the method described for
the preparation of compound 18, 3-amino-8-methoxy-1H-quinoxa-

Head: TRPV1 Antagonists: Selection of a Clinical Candidate
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 15 3527
(7) The effect on in vitro potency for these antagonists as a result of
varying pH is beyond the scope of this discussion, but will be
presented in more detail in a subsequent article.
(8) Gavva, N. R.; Tamir, R.; Qu, Y.; Klionsky, L.; Zhang, T. J.; Immke,
D.; Wang, J.; Zhu, D.; Vanderah, T. W.; Porreca, F.; Doherty, E.
M.; Norman, M. H.; Wild, K. D.; Bannon, A. W.; Louis, J.-C.;
Treanor, J. J. S. AMG 9810 [(E)-3-(4-t-Butylphenyl)-N-(2,3-
dihydrobenzo[b][1,4]dioxin-6-yl)acrylamide], a Novel Vanilloid Re-
ceptor 1 (TRPV1) Antagonist with Antihyperalgesic Properties. J.
Pharmacol. Exp. Ther. 2005, 313, 474-484.
(9) Lewis, F. D.; Reddy, G. D.; Elbert, J. E.; Tillberg, B, E.; Meltzer, J.
A.; Kojima, M. Lewis Acid Catalysis of Photochemical Reactions.
10. Spectroscopy and Photochemistry of 2-Quinolones and their
Lewis Acid Complexes. J. Org. Chem. 1991, 56 (18), 5311-5318.
(10) Compound 20 was incubated at a concentration of 10 µM with rat
or human hepatocytes (1 million cells/mL) for 4 h at 37 °C. Control
incubations in the absence of cells were also carried out. Reactions
were quenched with acetonitrile containing 0.05% formic acid. The
supernatant was analyzed by reverse phase (YMC ODS-AQ, 4.6 ×
250mm, 5 µm) HPLC-MS/MS with radiometric detection and ion
trap mass spectrometry using electrospray ionization. Hydrolysis
experiments were conducted with glucoronidase and/or sulfatase to
confirm the presence of glucoronide and/or sulfate metabolites.
(11) Stevens, J. R.; Pfister, K.; Wolf, F. J. Substituted Sulfaquinoxalines.
I. The Isolation and Synthesis of 3-Hydroxy-2-sulfanilamidoqui-
noxaline and of Related Quinoxalines J. Am. Chem. Soc. 1946, 68
(6), 1035-1039.
(12) Tan, H.; Semin, D.; Wacker, M.; Cheetham, J. An Automated
Screening Assay for Determination of Aqueous Equilibrium Solubility
Enabling SPR Study During Drug Lead Optimization. JALA 2005,
10 (6), 364-373.
(13) Gavva, N. R.; Bannon, A. W.; Immke, D. C.; Tamir, R.; Klionsky,
L.; Hever, G.; Wang, J.; Davis, J.; Zajic, G.; Arik, L.; Shi, L.; Bak,
A.; Surapaneni, S.; Hovland, D. N., Jr.; Louis, J.-C.; Norman, M.
H.; Magal, E.; Treanor, J. J. S. Preclinical Pharmacology of AMG
517: A Novel TRPV1 Antagonist in the Clinic, Society for
Neuroscience annual meeting, Atlanta, GA, 2006, poster no. 51.21/
M13.
(14) McKillop, A.; Chattopadhyay, S. K.; Henderson, A.; Avendano, C.
Applications of Ethyl Carboethoxyformimidate to Heterocyclic
Synthesis. Preparation of Condensed Pyrazinones and 1,4-Oxazino-
nes. Synthesis 1997, 3, 301-304.
JM070190P