Discovery of TRPM8 Antagonist (S)-6-(((3-Fluoro-4-(trifluoromethoxy)phenyl)(3-fluoropyridin-2-yl)methyl)carbamoyl)nicotinic Acid (AMG 333), a Clinical Candidate for the Treatment of Migraine

Article abstract of DOI:10.1021/acs.jmedchem.8b00518

Transient-receptor-potential melastatin 8 (TRPM8), the predominant mammalian cold-temperature thermosensor, is a nonselective cation channel expressed in a subpopulation of sensory neurons in the peripheral nervous system, including nerve circuitry implicated in migraine pathogenesis: the trigeminal and pterygopalatine ganglia. Genomewide association studies have identified an association between TRPM8 and reduced risk of migraine. This disclosure focuses on medicinal-chemistry efforts to improve the druglike properties of initial leads, particularly removal of CYP3A4-induction liability and improvement of pharmacokinetic properties. A novel series of biarylmethanamide TRPM8 antagonists was developed, and a subset of leads were evaluated in preclinical toxicology studies to identify a clinical candidate with an acceptable preclinical safety profile leading to clinical candidate AMG 333, a potent and highly selective antagonist of TRPM8 that was evaluated in human clinical trials.

Full text of DOI:10.1021/acs.jmedchem.8b00518

Subscriber access provided by UNIV OF DURHAM
Article
Discovery of TRPM8 Antagonist (S)-6-(((3-fluoro-4-(trifluoromethoxy)phenyl)
3-fluoropyridin-2-yl)methyl)carbamoyl)nicotinic acid (AMG
33), a Clinical Candidate for the Treatment of Migraine
(
3
Daniel B. Horne, Kaustav Biswas, James Brown, Michael D. Bartberger, Jeffrey Clarine, Carl D
Davis, Vijay K Gore, Scott Harried, Michelle Horner, Matthew R. Kaller, Sonya G. Lehto, Qingyian
Liu, Vu V Ma, Holger Monenschein, Thomas T Nguyen, Chester Yuan, Beth D. Youngblood,
Maosheng Zhang, Wenge Zhong, Jennifer R. Allen, Jian Jeffrey Chen, and Narender Gavva
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00518 • Publication Date (Web): 27 Aug 2018
Downloaded from http://pubs.acs.org on August 28, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 71
Journal of Medicinal Chemistry
Discovery of TRPM8 Antagonist (S)-6-(((3-fluoro-4-(trifluoromethoxy)phenyl)(3-
1
2
3
fluoropyridin-2-yl)methyl)carbamoyl)nicotinic acid (AMG 333), a Clinical Candi-
date for the Treatment of Migraine
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
†‡*
†
†
†
Daniel B. Horne, Kaustav Biswas, James Brown, Michael D. Bartberger, Jeffrey
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Ω
Ω
†
†
ꢀ
Clarine, Carl D. Davis, Vijay K. Gore, Scott Harried, Michelle Horner, Matthew R.
†
†
†
†
†
Kaller, Sonya G. Lehto, Qingyian Liu, Vu V. Ma, Holger Monenschein, Thomas T.
†
†
†
†
†
Nguyen, Chester C. Yuan, Beth D. Youngblood, Maosheng Zhang, Wenge Zhong,
†
†
†
Jennifer R. Allen, Jian Jeffrey Chen, and Narender R. Gavva
†
Ω
Departments of Amgen Discovery Research, Pharmacokinetics and Drug Metaboꢀ
ꢀ
lism, and Comparative Biology and Safety Sciences, Amgen Inc., One Amgen Center
Drive, Thousand Oaks, CA 91320 & Amgen Inc., 360 Binney Street, Cambridge, MA
0
2142
Transient receptor potential melastatin 8 (TRPM8), the predominant mammalian cold
temperature thermosensor, is a nonꢀselective cation channel expressed in a subpopulation
of sensory neurons in the peripheral nervous system, including nerve circuitry implicated
in migraine pathogenesis: the trigeminal and pterygopalatine ganglia. Genomeꢀwide asꢀ
sociation studies have identified an association between TRPM8 and reduced risk of miꢀ
graine. This disclosure focuses on medicinal chemistry efforts to improve drugꢀlike
properties of initial leads, particularly removal of CYP3A4 induction liability and imꢀ
provement of PK properties. A novel series of biarylmethanamide TRPM8 antagonists
was developed, and a subset of leads were evaluated in preclinical toxicology studies to
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 71
identify a clinical candidate with acceptable preclinical safety profile leading to clinical
candidate AMG 333, a potent and highly selective antagonist of TRPM8 which was evalꢀ
uated in human clinical trials.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
2
+
The transient receptor potential (TRP) channels are nonꢀselective, Ca permeable ionꢀ
channels, some of which function as molecular sensors. The TRP channel family is
comprised of 28 distinct members which are activated by a variety of stimuli including
voltage, pH, temperature (cold and heat), mechanical pressure, osmotic pressure, and enꢀ
dogenous and exogenous ligands. Amongst the TRP channels, a subset are implicated in
temperature sensation and thermoregulation (TRPV1ꢀ4 activated by heat/warmth;
TRPM8 activated by innocuous cold; and TRPA1 activated by noxious cold). TRPM8
ionꢀchannels are activated by cool temperatures (18ꢀ23 °C) as well as exogenous ligands
1
such as menthol and icilin. TRPM8 is expressed in the lung, bladder, artery, prostate,
and in key components of the peripheral nervous system such as the dorsal root ganglia
(DRG), pterygopalatine (SPG), and trigeminal ganglia (TG) that are involved in pain
sensation and migraine. Potential applications of modulating TRPM8 for treatment of
2
3
4
neuropathic and chemotherapyꢀinduced cold pain, prostate cancer, and overactive and
5
painful bladder syndromes have been described. Previous investigations from various
groups have resulted in identification of several smallꢀmolecule scaffolds that modulate
ACS Paragon Plus Environment

Page 3 of 71
Journal of Medicinal Chemistry
6
ꢀ11
TRPM8 activity. Herein we communicate our efforts targeting TRPM8 inhibition as a
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
potential treatment for migraine.
We previously described SAR studies leading to the identification of TRPM8 inhibitors
12a
12b
1
and 2 (Figure 1) and results in rat in vivo pharmacodynamic (PD) models of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
TRPM8 target engagement, the icilinꢀinduced wetꢀdog shake (WDS) and cold pressor
test (CPT). Compound 2 was a potent inhibitor of TRPM8 (h / rTRPM8 IC = 25 / 115
50
nM), had reasonable pharmacokinetic (PK) properties in rats, and was effective in vivo in
both the rat WDS and CPT models of TRPM8 inhibition. Initially, we evaluated potent
TRPM8 antagonists as potential therapeutics for neuropathic pain, but in our hands, 2
13
and other TRPM8 antagonists were ineffective in animal models of neuropathic pain.
However, data later emerged from genomeꢀwide association studies (GWAS) that identiꢀ
fied an association between single nucleotide polymorphs (SNPs) that are intergenic to
TRPM8 (one SNP located about 1000 bp 5’ to the start codon and a second SNP in the
14
intron), with a reduced risk for migraine. Based on this new genetic data, we pivoted
our existing TRPM8 preclinical program with the intent of seeking a therapeutic for miꢀ
graine pain.
Our lead optimization efforts described herein, seeking to improve in vitro and in vivo
properties of early leads, ultimately led to the nomination of development candidate (S)ꢀ
6ꢀ(((3ꢀfluoroꢀ4ꢀ(trifluoromethoxy)phenyl)(3ꢀfluoropyridinꢀ2ꢀ
yl)methyl)carbamoyl)nicotinic acid (35) that was subsequently advanced into human
clinical studies.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 1. Previously reported tetrahydroisoquinoline and tetrahydronaphthyridine
TRPM8 antagonists.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
RESULTS AND DISCUSSION
12b
As previously reported, optimization of a tetrahydroisoquinoline series of TRPM8
antagonists resulted in tetrahydronaphthyridine 2 which showed improved PK properties
and demonstrated potent activity in two TRPM8 engagement PD models: (i) the icilinꢀ
induced rat WDS model with complete inhibition at 10 mg/kg, C = 0.19 ꢁM; (ii) the rat
u
CPT, with complete inhibition of blood pressure increase at 10 mg/kg po, C = 0.49 ꢁM.
u
However, despite achieving plasma exposures >10ꢀfold over the exposure levels resulting
in complete inhibition of rat WDS or CPT, 2 failed to show efficacy in rodent models of
neuropathic pain. In addition, we discovered that potent induction of CYP3A4 in human
hepatocytes, regulated through the pregnane X receptor (PXR), was endemic throughout
this series. Since CYP3A4 plays an important role in human drug metabolism, upon reiꢀ
nitiating our program towards a treatment for migraine, we undertook efforts to eliminate
the CYP induction potential of this series of compounds. In addition, we desired to reꢀ
start our efforts with a novel chemical scaffold that might provide potentially improved
properties, since fairly extensive SAR investigations
ACS Paragon Plus Environment

Page 5 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Identification of potent biarylmethanamine based TRPM8 antagonists as a
starting scaffold.
had already been explored in the tetrahydroisoquinoline and tetrahydronaphthyridine
series.
In an earlier investigation into alternative syntheses of the tetrahydronaphthyridine core
of compound 2, we discovered that one of the uncyclized intermediates in that synthesis,
urea analog 3, suffered minimal loss of potency compared to the cyclized analog 2 (Figꢀ
ure 2). We viewed analog 3 as a novel starting scaffold we could use to potentially adꢀ
dress the suboptimal properties of the series. At this point we also discovered that the
preferred stereochemistry of the uncyclized analogs was opposite that of the tetrahyꢀ
dronaphthyridine series. Confirmation of stereochemistry by Xꢀray crystal and VCD or
optical rotation confirmed our stereochemical assignment of both series (Figure 3 and
SI). We postulate that the uncyclized series may be binding in an alternative binding
conformation than the cyclized series since there is some divergence in the SAR between
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 71
the series, although there is some common SAR that suggests the series share the same
binding pocket.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Initial SAR in the biarylmethanamide series began with substitution of the 3ꢀBr substitꢀ
uent of analog 3 seeking improvements in potency. The 3ꢀposition was tolerant of several
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
lipophilic groups, e.g. F (120 nM), allyl (28 nM), CF (4, 51 nM), and propynyl (48 nM),
3
and the 3ꢀCF analog was selected for further SAR exploration.
3
A small library of amides (52 analogs) and ureas (203 analogs) was made based on the
biarylmethanamine of analog 4. A variety of aryl amides from the library were well tolꢀ
erated and retained potency on TRPM8 (34 of 52, <0.5 ꢁM hTRPM8 IC ), while ureas
5
0
were less tolerated (22 of 203, <0.5 ꢁM hTRPM8 IC ), and urea 4 represented the most
5
0
potent compound among >200 analogs in the urea series. Retaining potency in the amide
series was in stark contrast to the SAR developed in the earlier fused tetrahydroisoquinoꢀ
line and naphthpyridine series, in which amides generally resulted in a significant loss of
potency (e.g. analog 5 vs. 6). Since the library showed divergence of SAR from the earꢀ
lier tetryhydroquinoline series (e.g. 1, 2, and 5), it was anticipated that an impact on reꢀ
ducing CYP3A4 induction could be developed in the amide series where we had been
unable to introduce CYP induction reducing elements while still maintaining potency in
the earlier series. From these libraries, quinoline amide 7 (Table 1) had the best overall
properties, so we concentrated subsequent SAR exploration around this scaffold.
Initial investigation of the Bꢀring demonstrated that removal of the trifluoromethyl
group, analog 8, resulted in >20ꢀfold loss of activity, but that TRPM8 inhibition could be
ACS Paragon Plus Environment

Page 7 of 71
Journal of Medicinal Chemistry
increased with incorporation of increasingly lipophilic groups at the 3ꢀ and 4ꢀpositions
9ꢀ11, 13ꢀ14) of the pendant phenyl group. In contrast, substitution at the 2ꢀposition of
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
the phenyl (12), incorporation of hetereocycles (19ꢀ20), or benzylic groups (21) were
generally not well tolerated, leading to less potent analogs. Investigation of the Aꢀring
demonstrated that removal or transposition of the pyridine nitrogen (22ꢀ23) led to losses
in potency. In addition, the changes above that led to more potent compounds, had miniꢀ
mal effects on CYP induction and in vitro stability, likely due to the inability to incorpoꢀ
rate polarity into this sector of the molecule.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
It became apparent at this stage that a handle to address CYP3A4 induction and in vitro
stability liabilities would need to come from modifications made elsewhere in the moleꢀ
cule.
Since it has been demonstrated that PXR transactivation and the resulting CYP3A4 inꢀ
duction potential can be decreased by lowering lipophilicity and/or introducing polarity
15
into the periphery of the molecule, we turned our attention to the Cꢀring of the moleꢀ
cules based on the previous SAR showing that it was unlikely that incorporation of polarꢀ
ity was going to be tolerated in the Aꢀ or Bꢀrings with respect to maintaining TRPM8 poꢀ
tency.
Table 1. SAR Summary for Biarylmethanamides
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Human Hepatoꢀ
cyte CYP3A4 Inꢀ
b
CLint
a
IC (nM)
[ꢁL/(min•
5
0
cLog
duction
mg)]
Comꢀ
Aꢀ
Bꢀ
D
c
(
mRNA POC)
pound
ring ring
(pH
POC
POC
d
hTRP rTRP mTRP HL RL
7.4)
(%) @ (%) @
M8
M8
M8
M
M
1
ꢁM
10 ꢁM
1
01± 124±1
2
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
25±9
51
<14 16
<14 <14
ꢀꢀ
ꢀꢀ
3.9
5
8
14
4
7
ꢀꢀ
145
63
75
64
74
4.2
4.9
18±8 34±6 53±18 40
28
3
89±
9
1260±
14
8
9
ꢀꢀ
ꢀꢀ
43 104
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
4.3
4.4
129
225
24
61
ACS Paragon Plus Environment

Page 9 of 71
Journal of Medicinal Chemistry
7
5±6
8
143±1
>39
9
1
2
3
4
5
6
7
8
9
10
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
152
151
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
4.8
4.8
4.8
4.3
5.3
4.9
5
>
39
11
92
321
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
>20,0
>20,00
0
1
2
146 335
57 251
0
0
9
2
5±2
1
13
224
2±1
6
1
1
4
5
67±31 20
15
51
157
218
23
1
9±0.
1
16
ꢀꢀ
ꢀꢀ
ꢀꢀ
49±40 50
48
71
59
ꢀꢀ
48
ꢀꢀ
5.0
4.2
4.7
1
7
75
119
373
51
>
39
18
301
706
ꢀꢀ
ꢀꢀ
9
>
10,0
00
>10,00
0
1
9
ꢀꢀ
<14 85
ꢀꢀ
ꢀꢀ
2.8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 71
>
10,0
>10,00
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
20
ꢀꢀ
ꢀꢀ
15 160
ꢀꢀ
ꢀꢀ
ꢀꢀ
ꢀꢀ
3.5
0
0
0
1
140
2050±
670
>39
9
2
1
78
4.2
±109
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
22
311
109
ꢀꢀ
ꢀꢀ
663
212
20
61
25
31
ꢀꢀ
ꢀꢀ
4.5
6.1
2
3
37
39
a
2+
IC values based on inhibition of icilin (1 ꢁM) induced influx of Ca into human, rat,
5
0
or mouse TRPM8ꢀexpressing CHO cells. Each IC value reported represents mean IC
50
50
calculated from three 22ꢀpoint concentrationꢀresponse plots; when replicant experiments
b
were performed, the data is reported as the mean value and standard deviation. In vitro
microsomal stability (intrinsic clearance, CL ), measured in a highꢀthroughput automatꢀ
int
c
ed format in human liver microsomes (HLM) and rat liver microsomes (RLM). Exꢀ
pressed as a percentage of the activation response with rifampicin when both compounds
d
are used at a concentration of 1 or 10 ꢁM Calculated LogD (cLogD) calculated with the
ACD/LogD Suite. www.acdlabs.com/logdsuite/ Advanced Chemistry Development, Inc.,
Toronto, ON, Canada. 24 April 2007.
Initial incorporation of polar groups into the periphery of the Cꢀring such as hydroxy
(25), nitrile (26), pyridyl (27), or methoxy pyridyl (28) functionalities were not successꢀ
ACS Paragon Plus Environment

Page 11 of 71
Journal of Medicinal Chemistry
ful in reducing PXR transactivation as compared to the unsubstituted benzamide (Table
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
2
), since these analogs all still showed potent in vitro PXR transactivation, >90% POC @
µM and was confirmed for analog 28 in the human hepatocyte CYP3A4 induction asꢀ
say. These changes also had only a minor effect on calculated LogD (cLogD) of the anaꢀ
logs. Our first breakthrough in reducing CYP induction was a change to the Cꢀring of the
analogs that drastically lowered the cLogD of the molecule, pyridone 29, which for the
first time showed lower CYP induction (17% @ 1 ꢁM) while still maintaining potency
on TRPM8 (hTRPM8 IC = 6 nM). We next attempted to reduce lipophilicity further by
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
replacement of the Aꢀring trifluoromethyl group (at position X). We demonstrated that
substitution of the trifluoromethyl group with a fluorine was tolerated, with analog 31 reꢀ
ducing CYP induction significantly (6% @ 1 µM) while still maintaining potency on
TRPM8 (hTRPM8 = 10 nM). In contrast, attempts to reduce CYP induction by replacing
the trifluoromethyl group in the early urea analog 4 with fluoro, gave fluoro urea analog
3
0, which resulted in a significant loss of potency and was also confirmed to still have
potent activation of hPXR. Further SAR studies around the pyridone 31 scaffold identiꢀ
fied several additional analogs, including pyridones 32 and 33. These pyridones were
potent on human and rodent TRPM8, had low turnover in liver microsomes, and reduced
CYP induction activity. Pyridones 31ꢀ33 were differentiated in subsequent in vivo PK
and toxicological studies.
Concurrent with the discovery of the pyridones as a Cꢀring subunit with desirable in
vitro properties, carboxylic acid 34 emerged as a functional group that was simultaneousꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 71
ly able to reduce CYP induction and maintain potency. Additional analogs of carboxylic
acid 34 identified nicotinic acid 35, which had a slight improvement in potency while
maintaining other properties in the appropriate range. Substitution of the 3ꢀfluoro of 35
with a trifluoromethyl group gave analog 36, in which CYP induction was significantly
increased. It should also be noted that the reduction in CYP induction potential, as
achieved by lowering lipophilicity with pyridone and carboxylic analogs 29ꢀ35, was also
accompanied by a significantly decreased intrinsic clearance in liver microsomes.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Optimization of CYP Induction Liability
Human
Hepatocyte
b
CLint
Co
CYP3A4 Inꢀ
duction
a
IC (nM)
[ꢁL/(min
•mg)]
m
50
1
2
po X
R
R
R
(
mRNA
un
d
c
POC)
hTRP rTRP mTRP HL RL POC POC cLog
M8 M8 M8
M
M
D
(%) @ (%)
ACS Paragon Plus Environment

Page 13 of 71
Journal of Medicinal Chemistry
1
ꢁM @ 10 (pH
1
2
3
4
5
6
7
ꢁM
7.4)
CF
3
2
2
4
5
CF3
CF3
CF3
CF3
F
H
H
H
27
ꢀꢀ
43
6±4
40
29 57
62
ꢀꢀ
45
5.1
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CF
3
13
ꢀꢀ
8
8±7 21±1
ꢀꢀ
ꢀꢀ
4.6
4.7
4.2
5.0
2.0
3.3
1.1
1.5
CF
3
26
27
35
ꢀꢀ
ꢀꢀ
ꢀꢀ
19 21
19 50
ꢀꢀ
CF
3
2
2
2
3
3
3
7
8
9
0
1
2
54
ꢀꢀ
ꢀꢀ
CF
3
CF3 H 15±5
24±11 15 18
56
17
ꢀꢀ
94
37
ꢀꢀ
CF
3
<1
CF3
CF3
H
H
6±3 12±2 27±46
<14
<14
<14
<14
4
<
1
F
F
F
119
ꢀꢀ
234
4
<
1
CF3 H 10±4 21±9 15±7
6
29
45
4
OC
F3
<1
4
F
9±3 12±3
9±2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 71
<
1
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
33
F
F
F
CF3
F
F
5±3
20
7±2
30
ꢀꢀ
ꢀꢀ
ꢀꢀ
<14
<14
<14
<14
5
5
38
34
34
51
1.5
4
OC
F3
<1
4
3
3
4
5
1.6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
OC
F3
<1
4
F 13±4 20±6
4
0.13
0.68
CF
<1
4
36
CF3 H 10±4 13±6 17±10
41
3
a-d
See table footnotes for Table 1
Evaluation of in vivo rat PK of pyridone and carboxylic acid containing analogs
demonstrated that the compounds had excellent rat PK, suitable for further profiling in
the rat WDS biochemical challenge model (Table 3). In the progression of this program,
we used the rat WDS model as a measure of in vivo TRPM8 target engagement and triꢀ
aged compounds based on their in vivo potency. Compounds 31, 32, 33, and 35 all
demonstrated robust efficacy in the rat WDS model with plasma levels resulting in 90%
inhibition of WDS (IC90,ub) corresponding to 1ꢀ10 fold coverage over the in vitro
rTRPM8 IC .
5
0
Table 3. Rat Pharmacokinetic Properties, WDS results, and Isolated Rabbit Heart
CV Assessment for Selected Compounds
ACS Paragon Plus Environment

Page 15 of 71
Journal of Medicinal Chemistry
po
Ex Vivo Isoꢀ
lated Rabbit
Heart
1
2
3
4
5
6
7
8
9
rWDS
Model
a
iv Rat PK
Rat
a
PK
Comꢀ
No NOEL
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Structure
Plasma
pound
CL
Vss
Effect
/
t½
h)
IC90
(CLub)
(L/k
F%
Levꢀ rWDS
(
IC90,ub
)
(
g
el
IC90ub
(
L/h/kg) g)
(
ꢁM)
(µM) (Fold)
0
.024
1.55
e
3
1
1.0
1.7
2.1
1.7
41
13
16
6.3
>100
0.1
1
1.0
59
b
(0.35)
(0.109)
0
.092
0.37
e
32
92
b
(
2.0)
(0.017)
0.092
0.80
e
3
3
3
5
83
3
41
c
(
0.74)
(0.074)
0
.32
0.80
f
71
>10
>420
d
(
11)
(0.024)
a
PK studies in male Sprague Dawley rats conducted with sampling time up to 24 h, n =
3
animals per study, Interanimal variability was less than or equal to 30%, iv studies
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 71
b
c
done in fed and po studies in fasted rats. 2 mg/kg iv dosing in 100% DMSO. 1 mg/kg
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
d
iv dosing in 1% Tween 80, 2% HMPC, 97% water/MSA pH 2.2. 1 mg/kg iv dosing in
e
f
100% DMSO. 1 mg/kg po in 1% Tween 80, 2% HMPC, 97% water/MSA pH 2.2. 1
g
mg/kg po dosing in 1% Tween 80, 2% HMPC, 97% water/NaOH pH 9.0. Isolated rabꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
bit heart assay run at continuous buffer infusion at concentrations of 0.1, 0.3, 1.0, 3.0,
10.0 ꢁM test article. NOEL = Maximum concentration at which no significant effect on
CV parameters (HR, Qtc, Qrs, Jtc, LV dpt) was observed.
Compound 35 exhibited statistically significant (p<0.05 by Dunnett’s Multiple Comparꢀ
isons) and doseꢀdependent prevention of icilinꢀinduced rat WDS, at 0.6, 1 and 3 mg/kg
(Figure 3). The effect at 3 mg/kg is considered to be full inhibition (unbound plasma conꢀ
centration of 120ꢁnM). The estimated ED was 1.1 (0.75 – 1.74) mg/kg, plasma IC =
90
90
0
.80 ꢁM, IC90,ub = 0.024 ꢁM.
(a)
(b)
ACS Paragon Plus Environment

Page 17 of 71
Journal of Medicinal Chemistry
Figure 3. (a) 35 inhibits icilinꢀinduced wetꢀdog shakes (WDS) in a doseꢀdependent
1
2
3
manner (b) 35 inhibition of WDS is plasma concentrationꢀdependent. ***p<0.05 by
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Dunnett’s Multiple Comparisons.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Since TRPM8 can be activated by cold in addition to icilin, we evaluated doses in the
same range of effect in rat WDS in the coldꢀinduced increase in blood pressure ꢀ cold
pressor test model. Since validated rodent models of migraine did not exist we were unꢀ
able to evaluate the potential efficacy of the compounds against migraine pain precliniꢀ
cally, thus the CPT was intended to be used as a measure of target engagement that could
also be translated to a biomarker for human TRPM8 target engagement by evaluation of
human CPT in future clinical studies. Compound 35 inhibited coldꢀinduced increase in
blood pressure in the rat CPT in a doseꢀdependent manner with 1 and 3 mg/kg producing
significant inhibition (p<0.05 by Dunnett’s Multiple Comparisons), with an ED of 1.1
9
0
mg/kg, IC = 1.1 ꢁM, IC
= 0.032 ꢁM (Figure 4). Compound 35 also fully blocked
9
0
90,ub
the cold pressor response at 3 mg/kg at average total plasma concentration of 3.3 ꢁM,
plasma unbound concentration = 0.10 ꢁM.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a)
b)
c)
Figure 4. Rat ColdꢀPressor Test (a) Increase in blood pressure during 5 minutes of exꢀ
posure to cold; 35 dose response 0.3, 1, 3 mg/kg. (b) 35 inhibited coldꢀinduced increase
in blood pressure in the Cold Pressor Test (CPT) in a doseꢀdependent manner (c) Inhibiꢀ
tion of CPT by 35 is plasma concentrationꢀdependent. ***p<0.05 by Dunnett’s Multiple
Comparisons.
Prior to initiating rat and dog toxicology studies, preliminary investigation of potential
adverse cardiovascular effects were evaluated using the ex vivo isolated rabbit heart
(
IRH) model. In the IRH model, pyridone 31 caused >20% reduction of left ventricle
contractility beginning at 0.3 ꢁM (Table 3), 31 was inactive in patch clamp assays against
ACS Paragon Plus Environment

Page 19 of 71
Journal of Medicinal Chemistry
hERG, NaV1.5, and LꢀType Ca channels, however other ionꢀchannels could also be reꢀ
sponsible for the contractility finding. The plasma exposure ratio of the IRH study no
effect level to the rat WDS IC90,ub of only 1ꢀfold was deemed a significant risk, which
was confirmed upon further profiling of 31 in an anesthetized dog CV study. Therefore,
further development of pyridone 31 was discontinued. Pyridone analogs 32 and 33
demonstrated improved IRH multiples compared to 31, and acceptable cardiovascular
profiles in dog CV studies, while carboxylic acid analog 35 showed no effect in IRH or
early in vivo dog cardiovascular safety assessments.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Based on the above results, pyridone 32 and carboxylic acid 35 were advanced to rat
and dog toxicology studies. Pyridone 32 had high safety multiples in a rat 14ꢀday toxiꢀ
cology assessment, however a finding of vasculitis was identified in a 14ꢀday dog toxiꢀ
cology study, at a dose of 10 mg/kg and above, and resulted in a low safety multiple.
Fortunately, carboxylic acid 35 (AMG 333) was well tolerated in 14ꢀday rat and dog toxꢀ
icology studies (Figure 5), and was subsequently nominated as a clinical candidate for
the treatment of migraine. In vitro toxicology profiling showed that 35 was selective
over several other TRP channels (IC TRPV1, V3, V4 > 20 ꢁM; TRPA1 > 40 ꢁM) and
5
0
had no hits in offꢀtarget activity panels; CEREP (144 targets @ 10 ꢁM; POC >45%) or
Ambit kinase (100 @ 1 ꢁM; POC >50%). 35 was well tolerated in 28ꢀday rat and dog
preclinical safety studies and advanced to Phase 1 human clinical trials.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. 14ꢀday Rat and Dog Toxicology Safety Multiples for Compound 35.
CHEMISTRY
A general, enantioselective route to analogs was developed using Ellman’s chiral tertꢀ
1
6
butylsulfinamide chemistry. Beginning with 3ꢀsubstituted picolinaldehydes 37a-c, forꢀ
mation of tertꢀbutylsulfinyl imines 38a-c was accomplished with (S)ꢀtertꢀ
butylsulfinamide and copper(II) sulfate (Scheme 1). Addition of aryl Grignard reagents
to the imines gave (S)ꢀdiarylmethyl sulfinamides 39a-n with good diastereoselectivity
(typically >9:1 dr). It should be noted that the addition of Grignard reagents to (S)ꢀtertꢀ
butylsulfiniyl imines 38a-c gave predominantly (S)ꢀstereochemistry at the benzylic
amine center as the major diastereomer. This stereochemistry is a reversal of the stereoꢀ
chemistry that is typically observed with Nꢀtertꢀbutylsulfinyl aromatic aldimines. A
ACS Paragon Plus Environment

Page 21 of 71
Journal of Medicinal Chemistry
similar reversal of stereochemistry was previously reported to occur with 2ꢀ
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
17
pyridylsulfinyl aldimine, and is likely a result of chelation of the pyridine and imine niꢀ
trogens and nucleophille addition from the less hindered, top face of intermediate A. Proꢀ
filing of both the enantiomers of urea products 44 and 46 demonstrated a significant loss
of potency for the (R)ꢀenantiomer (Figure 6). Finally HCl treatment of sulfinamide 39
gave amine hydrochlorides 40a-n. Amine intermediates 40o-u were prepared by analoꢀ
gous procedures detailed in the experimental section.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Scheme 1. Enantioselective Synthesis of Biarylmethanamines
a
Reagents and conditions: (a) (S)ꢀtertꢀbutylsulfinamide, CuSO , DCM, rt; (b) ArBr, Mg,
4
THF; 3, THF, ꢀ78 °Cꢀrt (c) HCl, MeOH/1,4ꢀdioxane, rt.
Finally, formation of analogs using biarylmethanamine 40 was accomplished with eiꢀ
ther amide forming conditions to give enantiomerically pure amides 41, or isocyanates to
give ureas 42 (Scheme 2).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
a
Scheme 2. Amide and Urea Formation
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
3
2
Reagents and conditions: (a) DIPPA, HATU, DMF, R CO H (b) CDI, DCM R NH ,
2
2
then 40a-n and DIPEA.
Assignment of absolute stereochemistry of analogs was accomplished by Xꢀray analyꢀ
sis of a key chiral intermediate, bromide 43, which lead to a more potent enantiomer of
3
ꢀpyridyl urea, 44 (Figure 3). A subset of additional analogs were confirmed to contain
(S)ꢀstereochemistry using the computational optical rotation and VCD spectra analysis
12b
method described in our previous manuscript.
Figure 3. Determination of Absolute Stereochemistry
ACS Paragon Plus Environment

Page 23 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
EXPERIMENTAL SECTION
In Vitro Human and Rat TRPM8 Functional Assay. Chinese Hamster Ovary (CHO)
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
cells stably expressing rat TRPM8 were generated using tetracycline inducible TꢀREx
expression system from Invitrogen, Inc (Carlsbad, CA). To enable a luminescence
readout based on intracellular increase in calcium, each cell line was also coꢀtransfected
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
with pcDNA3.1 plasmid containing jellyfish aequorin cDNA. Cells were seeded in 96ꢀ
well plates 24 h before the assay and TRPM8 channel expression was induced with 0.5
ꢁg/mL of tetracycline. On assay day, the growth media was removed and cells were inꢀ
cubated with assay buffer for 2 h. Cells were then exposed to test compounds (at varying
concentrations) and incubated for 2.5 min prior to adding the agonist, 1 ꢁM icilin (1ꢀ(2ꢀ
3
hydroxyphenyl)ꢀ4ꢀ(3ꢀnitrophenyl)ꢀ3,6ꢀdihydropyrimidinꢀ2ꢀone). The luminescence was
measured by a chargedꢀcouple device (CCD) cameraꢀbased FLASHꢀluminometer built
by Amgen, Inc. Compound activity was calculated using GraphPad Prism 4.01
(GraphPad Software Inc, San Diego, CA), the data reported represents mean IC calcuꢀ
5
0
lated from three 22ꢀpoint concentrationꢀresponse plots; when replicant experiments were
performed, the data is reported as the mean value and standard deviation.
Human and Rat Liver Microsomal Stability. Liver microsomal stability was measꢀ
ured at 37 °C in phosphate buffer (66.7 mM, pH 7.4). Test compounds (1 ꢂM) were incuꢀ
bated with pooled human or rat liver microsomes at 0.25 mg/mL of protein, with or withꢀ
out NADPH (1 mM). After 30 minutes the reaction was stopped by the addition of aceꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 71
tonitrile containing 0.5% formic acid and internal standard. The quenched samples were
centrifuged at 1,650 x g for 20 minutes. The supernatants were analyzed directly for unꢀ
changed test compound using liquid chromatography and tandem mass spectrometric deꢀ
tection (LCꢀMS/MS). Intrinsic clearance was calculated based on substrate disappearꢀ
ance rate assuming first order elimination of compound over the 30 minute incubation.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CYP3A4 Induction. Cryopreserved human hepatocytes were thawed and isolated usꢀ
ing CHRM™ (Lifetech, Carlsbad, CA) according to the manufacturer’s instructions.
Hepatocytes were then resuspended with plating media (1x DMEM supplemented with
0
.1 µM dexamethasone, 10% fetal bovine serum, 1x ITS, 1x PS, and 1x Lꢀglu) and platꢀ
ed on 96ꢀwell collagen I coated plates with a cell density of approximately 70,000 viable
cells/well. The plates were then placed in a 37 °C incubator under an atmosphere of 95%
air/5% CO and 90% relative humidity and allowed to attach overnight. The following
2
day the plating media was removed and replaced with culture media (1X Williams mediꢀ
um E supplemented with 0.1 µM dexamethasone, 1X ITS, 1x PS, 1x Lꢀglu) and the cells
placed back in the incubator overnight. On the third day, the medium was aspirated and
replaced with fresh culture media containing test articles dissolved in DMSO (final
DMSO concentration of 0.1% in media), this process was repeated 24 h later for a total
4
8 hour incubation with test article. On Day 5, hepatocytes were lysed in 100 µL of 1X
Quantigene® 2.0 lysis mixture and stored at ꢀ80°C prior to being assayed for CYP3A4
and bꢀActin mRNA content using the Quantigene® 2.0 branched DNA assay (Affymetꢀ
rix Fremont, CA) according to manufacturer’s instructions. Plate washing steps were perꢀ
ACS Paragon Plus Environment

Page 25 of 71
Journal of Medicinal Chemistry
formed on an Elx405 automated microplate washer (BioTek Instruments Inc., Winooski,
VT) and luminescence analyzed on a Luminoskan Ascent microplate luminometer
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(Thermo Labsystems, Helsinki, Finland).
Rat Pharmacokinetics. Male SpragueꢀDawley rats (n=3 per group) were dosed intraꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
venously (fed), with test article formulated in DMSO or by oral gavage (fasted) with test
compound formulated as a suspension in 5% Tween 80/Oraplus. Blood samples were
taken over 16 hours post dose, with plasma prepared by centrifugation and analyzed by
LCꢀMS/MS. Total plasma concentration timeꢀcourse data were analyzed by noncomꢀ
partmental pharmacokinetic methods.
In Vivo Assays. Icilin-Induced “Wet-Dog” Shaking in Rats. All experimental proꢀ
cedures for in vivo experiments were conducted in accordance with a protocol approved
by the Institutional Animal Care and Use Committee (IACUC). Male SpragueꢀDawley
rats (220–300 g, Harlan, n = 8/treatment) were first habituated to the testing room for 30
min and then placed into the automated LABORAS (Metris) apparatus with a transparent
Plexiglas® shoebox cage for 20 min for each rat. Antagonists (31, 32, 33 = 2% HPMC
1
% Tween 80 pH 2.2 with MSA; 35 = 2% HPMC 1% Tween 80 pH 9.0 with NaOH) or
vehicle control (2% HPMC 1% Tween 80 pH 2.2 with MSA or 2% HPMC 1% Tween 80
pH 9.0 with NaOH) were administered po 120 min prior to administration of icilin (3.0
mg/kg, 2%HPMC 1%Tween 80 in a injection volume of 2.5 mL/kg po) and WDS were
recorded for a duration of 30 min post icilin administration.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 71
Cold Pressor Test (CPT) in Rats. TRPM8 antagonists were evaluated in rat CPT to
determine whether TRPM8 antagonists would attenuate the increase in blood pressure
resulting from exposure to cold stimulation of the paws and ventral half of the body.
Male SpragueꢀDawley rats weighing 350ꢀ450 g were instrumented with a unilateral caꢀ
rotid arteryꢀcannula connected to a transducer for measuring blood pressure using a Digiꢀ
Med Blood Pressure Analyzer, Model 400. Animals were orally administrated with 35
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(2% HPMC 1% Tween 80 pH 9.0 with NaOH) or vehicle control (2% HPMC 1% Tween
8
0 pH 9.0 with NaOH) 120 min prior to the cold challenge and anesthetized with sodium
pentobarbital at 60 mg/kg IP 20 min prior to cold. Blood pressure was recorded for 4
min for preꢀcold baseline and additional 5 min during immersion of the paws and ventral
half of body in ice water (approximately 0 °C). Percent inhibition attributed to treatment
with test compound was then determined using the following Formula: [1ꢀ(cold evoked
change in MBP/cold evoked change in MBP postꢀvehicle)]x100. Plasma was collected
through artery catheter immediately after CPT for PK analysis.
Chemistry. Unless otherwise noted, all materials were obtained from commercial supꢀ
pliers and used without further purification. Anhydrous solvents were obtained from Alꢀ
drich or Fisher Scientific and used directly. All reactions involving airꢀ or moistureꢀ
sensitive reagents were performed under a N2 or Ar atmosphere. All final compounds
were purified to >95% purity, as determined by LC/MS obtained on Agilent 1100 and HP
1100 spectrometers. Silica gel chromatography was performed using either glass columns
packed with silica gel (100ꢀ200 or 200ꢀ400 mesh, Aldrich Chemical) or prepacked silica
ACS Paragon Plus Environment

Page 27 of 71
Journal of Medicinal Chemistry
gel cartridges (Biotage or ISCO). 1H NMR spectra were determined with a Bruker 300
MHz or DRX 400 MHz spectrometers. Chemical shifts are reported in parts per million
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
ppm, δ units). Chiral purifications by preparative supercritical fluid chromatography
SFC) were performed on an Agilent preparative SFC system using Chiralcel or Chiꢀ
(
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ralpak ODꢀH, AD, ADꢀH, or AS columns. Elution was by gradient using 5ꢀ55% MeOH
with 0.2ꢀ1% diethylamine modifier in supercritical carbon dioxide over 2ꢀ5 min at 50ꢀ70
mL/min. Trifluoromethylphenyl Grignard reagents used in the synthesis of amines have
18
been reported to undergo highly exothermic decomposition, appropriate precautions
during generation of this reagent should be undertaken.
Amine Intermediates
(
S)-(3-bromopyridin-2-yl)(4-(trifluoromethyl)phenyl)methanamine hydrochloride
(
40a). Step 1. To a solution of 3ꢀbromopicoline (25 g, 0.145 mol) in CCl was added Nꢀ
4
bromosuccinimide (51.66 g, 0.29 mol) and benzoylperoxide (2.5 g, 0.010 mol). The
mixture was then gradually heated to reflux and stirred for 30 h. The reaction mixture
was then cooled to rt, the succinimide was removed by filtration, and the filtrate was
concentrated under reduced pressure. The resulting product was purified by flash column
chromatography using silica (100ꢀ200 mesh) with 10% EtOAc in hexanes to furnish 3ꢀ
1
bromoꢀ2ꢀdibromomethylpyridine (10.0 g, 0.0303 mol, 21% yield) as a yellow oil. H
NMR (300 MHz, CDCl ) δ 8.70 (dd, J=1.46, 4.53 Hz, 1H), 7.88 (dd, J=1.46, 8.04 Hz,
3
1
H), 7.14ꢀ7.22 (m, 2H).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 71
Step 2. A suspension of 3ꢀbromoꢀ2ꢀdibromomethylꢀpyridine (10.0 g, 30.3 mmol) in
morpholine (30.0 mL) was heated to 60 °C for 1 h. The reaction mixture was cooled to rt
and diluted with EtOAc (200 mL). The pH was adjusted to pH 4 by adding citric acid
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(40.0 g). The reaction mixture was then extracted with EtOAc (3 x 200 mL) and the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
combined organic layers were dried over Na SO and concentrated under reduced presꢀ
2
4
sure. The residue was purified by column chromatography using silica (100ꢀ200 mesh)
using 3% EtOAc in hexane as eluent to give 3ꢀbromoꢀpyridineꢀ2ꢀcarbaldehyde 37a (4.06
1
g, 21.8 mmol, 72% yield) as a yellow oil. H NMR (300 MHz, CDCl ) δ 10.27 (s, 1H),
3
8
.78 (dd, J=1.32, 4.53 Hz, 1H), 8.06 (dd, J=1.32, 8.18 Hz, 1H), 7.34ꢀ7.43 (m, 1H).
Step 3. To 3ꢀbromoꢀpyridineꢀ2ꢀcarbaldehyde 37a (2.03 g, 10.9 mmol) in DCM (50
mL) was added (S)ꢀ2ꢀmethylpropaneꢀ2ꢀsulfinamide (1.98 g, 16.4 mmol), and copper(II)
sulfate (5.23 g, 32.7 mmol). After stirring for 16 h at rt, the suspenion was filtered
through a celite cartridge and the cartridge rinsed with DCM. The filtrate was concenꢀ
trated in vacuo and the residue suspended in 20% EtOAc/hexanes and again filtered
through a celite cartridge. The filtrate was adsorbed onto a plug of silica gel and chroꢀ
matographed through a preꢀpacked silica gel column (40 g), eluting with 0% to 30%
EtOAc in hexane, to provide (S)ꢀNꢀ((3ꢀbromopyridinꢀ2ꢀyl)methylene)ꢀ2ꢀmethylpropaneꢀ
2ꢀsulfinamide 38a (2.14 g, 7.40 mmol, 68% yield) as a light yellow oil.
Step 4. To a stirred suspension of magnesium (2.14 g, 88 mmol) in ether (50 mL) was
added 4ꢀbromobenzotrifluoride (5.06 mL, 36.2 mmol). The stirring was continued for 4h
(cautious! exothermic!). The solution was transferred dropwise to a stirring ꢀ78 °C soluꢀ
ACS Paragon Plus Environment

Page 29 of 71
Journal of Medicinal Chemistry
tion of (S)ꢀNꢀ((3ꢀbromopyridinꢀ2ꢀyl)methylene)ꢀ2ꢀmethylpropaneꢀ2ꢀsulfinamide 38a (5.1
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
g, 17.64 mmol) in THF (100 ml). The stirring was continued for another hour after the
addition, quenched with saturated NH Cl, extracted with Et O (3x), dried over MgSO ,
4
2
4
and concentrated in vacuo. The material was purified by ISCO (0ꢀ70% EtOAc/Hexanes)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
to
give
(S)ꢀNꢀ((S)ꢀ(3ꢀbromopyridinꢀ2ꢀyl)(4ꢀ(trifluoromethyl)phenyl)methyl)ꢀ2ꢀ
1
methylpropaneꢀ2ꢀsulfinamide 39a (6.30 g, 14.5 mmol, 82% yield) as a yellow oil. H
NMR (300 MHz, CDCl ) δ 8.57 (br d, J=4.09 Hz, 1H), 7.85 (d, J=8.04 Hz, 1H), 7.54 (s,
3
4
H), 7.14 (dd, J=4.68, 7.89 Hz, 1H), 6.58 (br d, J=7.31 Hz, 1H), 6.35 (br d, J=7.75 Hz,
1H), 1.43 (s, 9H).
Step 5.
To a stirred solution of (S)ꢀNꢀ((S)ꢀ(3ꢀbromopyridinꢀ2ꢀyl)(4ꢀ
(trifluoromethyl)phenyl)methyl)ꢀ2ꢀmethylpropaneꢀ2ꢀsulfinamide 39a (0.220 g, 0.505
mmol) in EtOH/DCM (1:1, 8 mL) was added HCl (4M in dioxane, 1.01 mL, 4.04 mmol)
at 0 °C. Stirring was continued for 1 h, then concentrated to dryness to provide (S)ꢀ(3ꢀ
bromopyridinꢀ2ꢀyl)(4ꢀ(trifluoromethyl)phenyl)methanamine hydrochloride 40a as a solid
which was used without further purification in the subsiquent step.
(
R)-(3-bromopyridin-2-yl)(4-(trifluoromethyl)phenyl)methanamine hydrochloride
(ent-40a). To a stirred solution of (R)ꢀNꢀ((R)ꢀ(3ꢀbromopyridinꢀ2ꢀyl)(4ꢀ
trifluoromethyl)phenyl)methyl)ꢀ2ꢀmethylpropaneꢀ2ꢀsulfinamide (0.96 g, 2.205 mmol,
(
prepared by analogous procedure to (S)ꢀenantiomer 40a) in EtOH (20 mL) at 0 °C was
added HCl (4M in dioxane, 5.51 mL, 22.1 mmol). Stirring was continued for 2 h. The
mixture was concentrated to the dryness to give the title compound (730 mg, 100% yield)
ACS Paragon Plus Environment