Structure-activity studies of a novel series of isoxazole-3-carboxamide derivatives as TRPV1 antagonists

Article abstract of DOI:10.1016/j.bmcl.2010.12.092

Optimisation of a screening hit incorporating both TRPV1 activity and solubility was conducted. Substitution of the isoxazole-3-carboxamide with the bespoke 1S, 3R-3-aminocyclohexanol motif afforded the requisite balance of potency and solubility. Compounds 32 and 40 were found to have antihyperalgesic effects in the rat CFA Hg assay and induce a mechanism based hyperthermia.

Full text of DOI:10.1016/j.bmcl.2010.12.092

Bioorganic & Medicinal Chemistry Letters 21 (2011) 892–898
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structure–activity studies of a novel series of isoxazole-3-carboxamide
derivatives as TRPV1 antagonists
Ronald Palin a, , Lynn Abernethy b, Nasrin Ansari d, Kenneth Cameron a, Tom Clarkson a, Maureen Dempster c,
⇑
David Dunn d, Anna-Marie Easson a, Darren Edwards a, John Maclean a, Katy Everett a, Helen Feilden a,
Koc-Kan Ho d, Steve Kultgen d, Peter Littlewood b, Duncan McArthur a, Deborah McGregor c,
Hazel McLuskey c, Irina Neagu d, Stuart Neale b, Lesley-Anne Nisbet a, Michael Ohlmeyer d, Quynhchi Pham d,
Paul Ratcliffe a, Yajing Rong d, Andrew Roughton d, Melanie Sammons b, Robert Swanson d, Heather Tracey b,
Glenn Walker c
a Department of Chemistry, MSD, Newhouse, Lanarkshire ML1 5SH, UK
b Department of Pharmacology, MSD, Newhouse, Lanarkshire ML1 5SH, UK
c Department of Molecular Pharmacology, MSD, Newhouse, Lanarkshire ML1 5SH, UK
d Pharmacopeia, PO Box 5350, Princeton, NJ, 08543, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Optimisation of a screening hit incorporating both TRPV1 activity and solubility was conducted. Substi-
tution of the isoxazole-3-carboxamide with the bespoke 1S, 3R-3-aminocyclohexanol motif afforded the
requisite balance of potency and solubility. Compounds 32 and 40 were found to have antihyperalgesic
effects in the rat CFA Hg assay and induce a mechanism based hyperthermia.
Received 29 October 2010
Revised 16 December 2010
Accepted 18 December 2010
Available online 23 December 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
TRPV1 antagonist
Antihyperalgesia
Isoxazoles
Hyperthermia
Transient receptor potential vanilloid 1 (TRPV1) is a Ca2+ per-
meant non-selective cation channel expressed in a subpopulation
of primary afferent neurons.1 TRPV1 is located both in the periph-
ery and CNS on C and Ad fibres, the afferents commonly associated
with nociception. In addition to mediating the effects of exogenous
capsaicin (the pungent component of chilli peppers) primary affer-
ent TRPV1 receptors are thought to trigger the actions of heat
(>43 °C), and protons (pH <6.8) and are modulated by a variety
of endogenous lipid mediators including anandamide and bradyki-
nin.2 Consequently, TRPV1 is believed to act as an integrator of
nociceptive responses to both chemical and thermal noxious
stimuli.3
Since the receptor was cloned and characterised by Julius and
co-workers in 19971 there has been a considerable amount of re-
search interest into this ion channel. TRPV1 knockout mice pro-
vided evidence that TRPV1 played a key role in pain, with a clear
attenuation to thermal hyperalgesia in response to proinflamma-
tory agents.4 TRPV1 also shows increased expression in pain states
in both rat and human.5 A number of small molecule ligands have
also been published that show a clear reversal of both thermal and
mechanical hyperalgesia in preclinical pain models.6 This data
clearly supports the role of TRPV1 antagonists in the management
of acute and chronic pain.7
A staggering recent investment in R&D has led to a plethora of
small molecule antagonists in the area.8 Among the earliest TRPV1
antagonists to appear in the literature were the aryl ureas, for
example, BCTC9 (1) (Fig. 1). These served only as tools to explore
pre-clinical pharmacology because of the poor aqueous solubility,
poor oral bioavailability and metabolic instability. Other research-
ers have reported efforts to improve upon the poor solubility and
metabolic instability of BCTC replacing the central piperazine ring
with a phenyl ring, maintaining TRPV1 antagonism but failing to
improve solubility.10
Over the past few years a number of small molecule TRPV1
antagonists such as 2, 3 and 4 (Fig. 1) have entered clinical trials.11
However, the sensitivity of TRPV1 to heat has suggested a role in
maintenance of body temperature, and clinical trials of at least
one TRPV1 antagonist have been stopped because of unacceptable
levels of hyperthermia.8d
These compounds have in common a flat, relatively linear shape,
with moderate to high lipophilicity and consequently low solubility.
⇑
Corresponding author. Tel.: +44 1698 736128; fax: +44 1698 736187.
E-mail address: ronnie.palin@gmail.com (R. Palin).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.12.092

R. Palin et al. / Bioorg. Med. Chem. Lett. 21 (2011) 892–898
893
F
O
F
O
H
F
N
F
F
NH
S
NH
HN
N
H
N
F
N
H
N
H
Br
O
N
N
O
N
N
N
O
N
N
N
H
Cl
2
4
1
3
Figure 1.
Table 1
After an HTS campaign we identified isoxazole-3-carboxamide
5 (Fig. 2), with moderate potency at TRPV1 (Table 1). In this report
we describe the synthesis and biological evaluation of a series of
isoxazole-3-carboxamides that modulate TRPV1. Initial optimisa-
tion goals for this series were improvements in potency and solu-
bility sufficient to demonstrate in vivo efficacy and to provide
confidence for full LO resourcing. This study culminated in the
identification of potent new TRPV1 antagonists with good oral bio-
availability and efficacy demonstrated in models of inflammatory
pain suitable for further optimisation.
The general synthesis towards isoxazole-3-carboxamide deriva-
tives is outlined in Scheme 1. The strategy involved the reaction of
various substituted acetophenones (6a–f) with diethyl oxalate in
the presence of sodium ethoxide. The resulting 2,4-diketo esters
(7a–f) on treatment with hydroxylamine hydrochloride furnished
substituted 3-isoxazole esters (8a–f) in excellent yields. The 4-po-
sition was further elaborated by bromination in the presence of
NBS in acetic acid at elevated temperatures to yield the tri-substi-
tuted isoxazoles (9a–f). The esters were then heated to high tem-
perature in the presence of cyclopentylamine and base to deliver
amides (10a–f).
SAR of the 4- and 5-position of the central isoxazole ring
R
X
O
4
1
N
3
2
H
O
N
No.
R
X
TRPV1 IC50 (nM)
10a
10b
5
10c
10d
10e
10f
14
2-Cl
3-Cl
4-Cl
4-F
Br
Br
Br
Br
Br
Br
Br
H
Inactive
4222
370
377
410
1252
46
201
112
15
4-Br
4-OCF3
4-CF3
4-CF3
4-CF3
4-CF3
15
16
F
Cl
R1
R1
R1
O
b
a
O
OEt
In order to exploit the 4-position the strategy outlined in
Scheme 2 was adopted. The previously prepared 3-isoxazole ester
(8f) was exposed to either NCS to deliver the chloro substituted
derivative (11fa) or selectfluor to install the fluoro moiety (11fb).
The latter method often required elevated temperatures and deliv-
ered moderate yields. The halogenated derivatives were then
saponified to reveal the acids (12fa and 12fb) and amides formed
via the acid chlorides (13fa and 13fb) or directly using standard
coupling conditions to give 14–40.
O
O
c
OEt
O
O
N
8a-f
7a-f
6a-f
R1
R1
Br
d
Br
O
O
O
N
N
H
O
OEt
N
The preparation of the bespoke amine 41 is outlined in Scheme 3
following modified methods of Sammes and Thetford.12 Reaction of
potassium phthalimide with 3-bromocyclohexene (42) underwent
the Gabriel reaction to give compound 43. Treatment of 43 with
NBS in ethanolic chloroform gave only the 1,3-adduct because the
phthalimide group prefers to react by a six-membered transition
state.12 Release of the alcohol was realised by treating with dilute
acid. This was readily reduced to furnish the desbromo compound
44 as a racemic cis mixture. Resolution to the enantiomerically
pure amino alcohol 41 was achieved using the lipase-catalysed
methods of Gotor and co-workers.13 Treating the N-protected
derivative 44 with lipase B from Candida antarctica (CAL-B) in THF
for 3 h in the presence of vinyl acetate gave 45 and 46 which was
easily separated using column chromatography. Hydrazinolysis of
the N-protected aminocyclohexanol 45 produced the dihydroph-
9a-f
10a-f
a R1 = 2-Cl, b R1 = 3-Cl, c R1 = 4-F, d R1 = 4-Br, e R1 = 3-OCF3, f R1 = 4-CF3
Scheme 1. Reagents and conditions: (a) diethyl oxalate, 21% wt sodium ethoxide in
EtOH, THF; (b) hydroxylamine hydrochloride, EtOH, 85 °C; (c) N-bromosuccinimide,
AcOH, 100 °C; (d) amine, N,N-diisopropylethylamine, DCM.
thalazine-1,4-dione side-product from which the chemically pure
3-aminocyclohexanol 41 was not easily isolated. The crude was
treated with benzyl chloroformate in the presence of base to give
the N-protected carbamate which allowed for easier purification
on silica gel. Hydrogenolysis of the protecting group then afforded
the desired enantiomerically pure aminocyclohexanol 41 in >99%
ee. The absolute configuration was tentatively assigned as 1S, 3R
for 41 based on the literature data.12
Compounds were tested in a TRPV1 assay.14 Initial investiga-
tions centred on scanning at the 5-position of the central isoxazole
core (Table 1). 2- or 3-chloro substituted aryl groups (10a and 10b)
leading to inactive or weakly active compounds respectively. In
extreme cases agonism could be obtained by substituting in the
2-position (data not shown). Replacing the 4-chloro substituent
with either 4-fluoro (10c) or 4-bromo (10d) led to equally potent
Br
O
Cl
N
H
O
N
5
Figure 2.

894
R. Palin et al. / Bioorg. Med. Chem. Lett. 21 (2011) 892–898
R1
R1
R1
method
A or B
R2
N
R2
N
O
O
a
O
O
OEt
O
N
OEt
O
OH
8f R1 = 4-CF3
11fa R1 = 4-CF3, R2 = Cl
11fb R1 = 4-CF3, R2 = F
12fa R1 = 4-CF3, R2 = Cl
12fb R1 = 4-CF3, R2 = F
d
b
R1
R1
R2
c
R2
N
O
O
R'
O
N
N
H
O
Cl
14- 40
13fa R1 = 4-CF3, R2 = Cl
13fb R1 = 4-CF3, R2 = F
Scheme 2. Reagents and conditions: Method A: N-chlorosuccinimide, AcOH, 100 °C; Method B: selectfluor, tetramethylene sulfone, 120 °C: (a) Lithium hydroxide, H2O, THF;
(b) oxalyl chloride, DCM, cat. DMF; (c) amine, N,N-diisopropylethylamine, DCM; (d) amine, 1-hydroxybenzotriazole, N-(3-dimethylaminopropyl)-N0-ethylcarbodiimide,
triethylamine, THF.
Br
O
O
a
b, c, d
N
N
O
43
HO
O
44
(+/-) cis
42
e
O
O
NH2
f, g, h
1
3
N
N
+
3
1
OH
AcO
HO
O
O
41
(1R,3S)
(1R,3S)
46
45
Scheme 3. Reagents and conditions: (a) potassium phthalimide, DMF, 100 °C; (b) N-bromosuccinimide, EtOH, CHCl3; (c) 2 N HCl, MeOH; (d) tributyltin hydride, AIBN, MeOH;
(e) vinyl acetate, C. antarctica; (f) hydrazine hydrate, MeOH; (g) benzyl chloroformate, Na2CO3, H2O; (h) 10% palladium on carbon, H2 (1 atm), MeOH.
compounds, whilst the 4-trifluoromethoxy group (10e) appeared
detrimental to activity when compared to the progenitor com-
pound (5). The 4-trifluoromethyl (10f) group led to a dramatic
improvement in activity and was the optimal group for potency
in this region. Exploration of the 4-position of the isoxazole core
showed halogens to be optimal. The unsubstituted compound
(14) showed a fourfold loss in potency whilst the fluoro (15) and
chloro (16) substituents in this position showed a twofold and
>10-fold increase in potency respectively. Therefore it appears that
4-trifluoromethyl on the aromatic and chloro on the 4-position of
the isoxazole are the optimal groups when the cyclopentylamine
is in place.
A variety of 3-carboxamide substituents were investigated
(Table 2) whilst retaining the 4-trifluoromethyl on the aromatic
and the chloro on the 4-position of the isoxazole. The majority of
the functionality probed in this region reflected the fact that the
physicochemical properties required to increase the solubility.
Aliphatics show a degree of activity (21 and 22) however the solu-
bility was generally poor as expected. Pendant groups containing a
strongly basic nitrogen or acidic group were poorly tolerated (data
not shown). Interestingly the tertiary amide (20) switched the
chemotype from antagonism to agonism. This was also true for a
number of other examples (data not shown) suggesting the impor-
tance of the NH functionality for antagonistic effects. This has also
been shown by a number of other groups in the area.15
Aminoalcohols were generally more potent than the aliphatic
counterparts with small cyclic systems more potent than the acy-
clic progenitors. Interestingly the free hydroxyl of 17 appeared al-
most fourfold more potent than when the hydroxyl was masked as
a methoxy group (19). The scope of the contribution of the amin-
ocycloalkanols to potency and solubility was examined further
through incorporation of a series of commercially available 3-
aminocyclopentanols. It was discovered that potency at TRPV1
was strongly influenced by the stereochemistry of the amino and
hydroxyl substituents on the cyclopentyl core. The 1S, 3R-3-amino-
cyclopentanol (27) was established with the greatest potency, two-
fold more potent that the corresponding cis enantiomer (28), and
at least 10-fold more potent than the trans variants (25 and 26).
However it was equipotent with the cyclopentylamine progenitor
(16) although the hydroxyl moiety was found to impart solubility
in all of the aminocyclopentanols. Based on these encouraging
findings a series of compounds based on cyclohexyl containing
pendant groups was investigated. At the time only mixtures of iso-
mers were available commercially and served as the de facto scan-
ning mode for the aminocyclohexanols. The mixture of four
compounds was found to be 20-fold more potent than the progen-
itor cyclohexylamine (22). The racemates of the cis (30) and trans
(33) isomers of the 3-aminocyclohexanol showed that the cis iso-
mer was preferred being 20-fold more potent. This was also true
for the 2-aminocyclohexanol (29) although this was ninefold less

R. Palin et al. / Bioorg. Med. Chem. Lett. 21 (2011) 892–898
895
Table 2
potent than the 3-aminocyclohexanol counterpart (30). It was clear
that the cis orientation was linked to the strongest TRPV1 potency,
and one particular isomer assigned as 1S, 3R-aminocyclohexanol
(32) conferred TRPV1 potency of 3 nM, 37-fold more potent than
its enantiomer.
SAR of the 3-carboxamide position of the central isoxazole ring
Cl
O
F3C
R1
N
H
O
N
Whilst the increase from five to six-membered ring size im-
parted an increase in TRPV1 potency the physiochemical proper-
ties of the cyclohexyl based system resulted in lower solubility.
This proved to be the general trend whereby cyclohexyl based
compounds were more potent but less soluble, requiring a return
to intervention elsewhere in the molecule. A survey was envisaged
in which the isoxazole 4-substituent was fixed as chloro, whilst
varying the pendant 5-isoxazole aryl and 3-carboxamide groups.
This search for moieties associated with improved solubility led
to a number of interesting patterns and identified combinations
of groups important for both solubility and potency (Table 3).
When the 4-trifluoromethyl substituent on the aryl was replaced
with 4-fluoro (37) a 19-fold weaker but much more soluble com-
pound emerged. This marked the identification of the 4-fluoro aryl
substituent as a moiety associated with improved solubility. This
was then combined with the 1S, 3R-3-aminocyclohexanol carbox-
amide substituent of the highly potent TRPV1 antagonist 32 result-
ing in compound 39. This displayed a threefold improvement in
potency over 37 but resulted in a higher solubility of 39 mg/L.
The aryl group was probed further for different substitution pat-
terns that might bring similar solubility and improved TRPV1 po-
tency. Compound 40 was identified containing a 3,4-difluoroaryl
substituent and 1S, 3R-3-aminocyclohexanol with the crucial com-
bination of both TRPV1 potency (79 nM) and solubility (17 mg/L).
Given the data it is obvious the aminoalcohol moiety in 40 plays
a pivotal role in both activity and solubility. In an attempt to ratio-
nalise the interactions we investigated the conformation of this
part of the molecule. Conformational analysis of 40 was carried
out by NMR. 1H NMR spectra show that there is a significant differ-
ence in the H-1 and H-3 signals when going from the hydrophobic
(CDCl3) to the polar (CD3OD) solvent. The large coupling constants
No. R1
TRPV1 IC50 (nM) Stereochemistry Solkin16 (mg/L)
N
OH
OH
17
18
19
595
238
—
—
—
7
H
N
H
5
N
H
O
2212
NT
N
20
21
263a
87
—
—
5
1
OH
N
H
22
23
92
43
—
—
1
7
N
H
O
N
H
OH
24
25
26
27
28
29
392
197
762
23
Trans
NT
4
N
H
OH
1R, 3R
N
*
*
*
*
H
OH
OH
OH
1S, 3S
NT
5
N
H
1S, 3R
N
H
55
1R, 3S
6
N
H
Table 3
SAR of the 3-carboxamide and 5-aryl positions of the central isoxazole ring
65
cis racemate
NT
N
R1
H
OH
Cl
O
R
R2
N
30
31
32
33
7.4
cis racemate
1R, 3S
NT
3
H
N
H
OH
ON
No.
R2
*
R1
H
R
F
TRPV1 IC50 (nM)
434
Solkin (mg/L)
19
112
3
N
H
OH
OH
OH
OH
OH
OH
37
N
H
1S, 3R
2
N
H
27
38
H
F
CF3
23
5
149
trans racemate
1
N
H
*
N
H
OH
F
404
33
N
H
*
34
120
cis racemate
2
N
H
OH
39
32
40
H
H
F
F
141
3
39
2
N
OH
H
35
36
48
cis racemate
NT
1
N
H
OH
CF3
F
N
H
OH
OH
118
—
*
N
H
OH
79
17
N
H
NT = not tested.
a
Compound shown to be an agonist.
All compounds shown to be single enantiomers.

896
R. Palin et al. / Bioorg. Med. Chem. Lett. 21 (2011) 892–898
Table 4
The observed coupling constants (J) for H-1 and H-3 to H-2 determined from the
PERCH simulation17
Proton
J (Hz) [CD3OD]
J (Hz) [CDCl3]
1?2
11.24 (ax-ax)
3.95 (ax-eq)
10.43 (ax-ax)
3.99 (ax-eq)
8.17
4.08
7.33
3.56
3?2
O
HO
R
NH
3
H
R
H1
H1
O
3
H
B
N
H
O
OH
R
HO
H
NH
A
3
1
H
C
Figure 3. The possible conformations of 3-aminocyclohexanol moiety.
Figure 4. The effect 40 (10, 100, 1000 nM) on capsaicin-evoked currents recorded
in adult rat DRGs. Reduction in the capsaicin-evoked currents is shown in a single
neurone (a) and (b) mean data from three experiments. Currents are evoked by 1 s
applications of capsaicin (500 nM) at the time illustrated by the rectangle in (A).
between H-1 and H-2 and H-3 and H-2 in CD3OD (Table 4) indicate
H-1 and H-3 are both in an axial position in this solvent and there-
fore the OH and amide groups are confirmed as being cis. The
change in coupling constants on changing solvents is associated
with the cyclohexyl ring flipping between different conformations.
The lowest energy conformation of this system is expected to be
the chair with the hydroxyl and amide groups equatorial (Fig. 3,
A). Conformation A is seen to be dominant in CD3OD, because
the coupling constants of H-1 and H-3 to H-2 are large, (Table 4)
and so these protons are axial. However, in CDCl3 there is a signif-
icant contribution from either the inverted chair conformation (B)
or the twist-boat conformations (C) and so the H-1 and H-3 cou-
pling constants are of intermediate magnitude (Table 4).
The stabilization of B and C in CDCl3 is likely to be due to hydro-
gen bonding from the OH to either the NH or the carbonyl oxygen.
The fact that these conformations could be stabilized in a hydro-
phobic environment may be important if this group sits in a hydro-
phobic binding pocket. In fact this is speculated as the preferred
conformation since literature examples suggest that a polar group
cannot be tolerated in this region.
involved in the metabolism of 32 and 40 (data not shown). This
is an excellent profile for compounds and further indicates that it
is unlikely to suffer from severe drug–drug interactions.
The TRPV1 antagonist 40 was tested for effects at the native
receptor using whole-cell patch clamp electrophysiology18 to re-
cord capsaicin (500 nM)-evoked currents in adult rat dorsal root
ganglion neurones in vitro (Fig. 4). The compound produced a con-
centration-dependent reduction in the current amplitude, with the
top concentration tested, 1000 nM, reducing the capsaicin-evoked
current by 91 3% compared to control (n = 3; P < 0.01, ANOVA,
followed by Dunnett’s ad hoc test). This effect was reversible on
washout of the compound.
Both compounds demonstrated a sufficient profile suitable to
progress for in vivo evaluation in our analgesic behavioural
models.
Together 32 and 40 were advanced into rat pharmacokinetic
experiments (Table 5). The compounds were administered to
Wistar BRL rats and showed good absorption; with Tmax ꢀ4 h
(32) and 2 h (40) post dose. However 32 had considerably better
oral bioavailability presumably due to the much lower clearance
of 32 (5.5 mL/min/kg) compared with 40 (13.4 mL/min/kg).
Both 32 and 40 showed no significant inhibition of P450 en-
zymes (data not shown) and appeared clean across the five major
isoforms and as such had no inhibition problems giving confidence
that the risk of drug–drug interactions would be low. The P450
identification assay also showed that a range of enzymes were
The TRPV1 receptor antagonists 40 and 32 were profiled in rat
models of acute inflammatory thermal hyperalgesia (complete Fre-
und’s adjuvant (CFA)) and on body temperature in telemetered
rats. In the CFA model, 40 (10, 30, 100
ment) significantly reversed the thermal hyperalgesia induced by
CFA, with an MED of 30 mol/kg (see Fig. 5a). At this dose, the
lmol/kg; po; 2 h pre-treat-
l
paw withdrawal latencies (PWLs) were significantly attenuated
from baseline values of 4.7 0.6 to 9.5 1.2 s (ꢀ59%), with an
ED50 value of 25.2 lmol/kg. 32 (10, 30, 100 lmol/kg; po; 3 h pre-
treatment) also had an MED of 30
lmol/kg in this assay, with
PWL’s significantly attenuated from baseline values of 3.6 0.5 to
Table 5
PK profile of compounds 32 and 40 in Wistar rats
No.
iv dose (mg/kg)
Cl (mL/min/kg)
Vss (L/kg)
T
½
(h)
po dose (mg/kg)
AUC0–in (ng h/mL)
Tmax (h)
Cmax (ng/mL)
F (%)
32
40
1.0
2.0
5.4
13.4
0.8
1.0
1.9
0.9
10
10
32368
3313
4.0
2.2
3210
408
100
27

R. Palin et al. / Bioorg. Med. Chem. Lett. 21 (2011) 892–898
897
(b) Compound 32: CFA Hargreave's
(a) Compound 40: CFA Hargreave's
Vehicle
30umol/kg
10umol/kg
Vehicle
30umol/kg
10umol/kg
100umol/kg
15
10
5
15
10
5
100umol/kg
*
*
*
*
0
0
-24h
0h
Time (h)
p<0.05 (One-way ANOVA, post hoc Dunn's compared to vehicle)
2h
-24h
0h
3h
Time (h)
*
(c) Compound 40: Body temperature
(d) Compound 32: Body temperature
vehicle
100mol/kg
veh
100mol/kg
38.0
38.0
#
#
#
#
#
#
#
#
#
#
37.5
37.0
36.5
36.0
#
37.5
37.0
36.5
36.0
-1
1
2
3
4
5
6
-1
1
2
3
4
5
6
Time (h)
Time (h)
#
p<0.05 (two-way ANOVA, post hoc unpaired t-test, compared to vehicle)
Figure 5. The effect of po administration of (a) 40 (10, 30, 100
lmol/kg) and (b) 32 (10, 30, 100 lmol/kg) on paw withdrawal latency (PWL) in the CFA Hg assay and (c) 40
(100 mol/kg) and (d) 32 on body temperature in telemetered animals. Data are expressed as mean sem.
l
(a) CFA Hg
(b) Body temperature
15
38.0
37.5
37.0
36.5
10
5
0
36.0
0
-1
0
1
4
7
1
4
7
Time (days)
Time (days)
Vehicle (Temp)
100µmol/kg (Temp)
Vehicle (PWL)
100µmol/kg (PWL)
Figure 6. The effect of po administration 32 (100
lmol/kg) assessed 3 h after compound administration on (a) paw withdrawal latency (PWL) in the CFA assay and (b) on
body temperature in telemetered rats. Data are expressed as mean sem.
8.7 0.8 s (ꢀ51%). The ED50 value for 32 was calculated to be
sic efficacy was maintained throughout the period whilst tolerance
to the effects on body temperature was apparent by day 4 (see
Fig. 6).
7.4 lmol/kg (see Fig. 5b).
Compounds 40 (10 and 100
100 mol/kg) were also evaluated for effects on body temperature
in rats. Compound 40 at 10 mol/kg showed a small, but signifi-
cant, increase in body temperature (data not shown); at
100 mol/kg 40 showed a larger and more sustained (1–5 h) in-
lmol/kg; po) and 32 (10, 30 and
l
In summary, we have identified a novel series of TRPV1 antag-
onists. We have shown key features of the isoxazole-3-carboxam-
ide derivatives that confer both TRPV1 potency and solubility. The
bespoke 1S, 3R-3-aminocyclohexanol unit is highlighted as a fun-
damental stereochemically defined group that aids both potency
and solubility. The trifluoromethyl moiety was shown to impart
strong potency to the molecules at TRPV1 whilst introduction of
fluoro substituents were found to make important contributions
to the compounds exhibiting the requisite balance of solubility
and potency. In particular compounds 32 and 40 were identified
as the most promising compounds from this series and progressed
l
l
crease (ꢀ0.3–0.5 °C) in body temperature (see Fig. 5c). Similarly,
32 also showed a significant increase in body temperature at
100 lmol/kg (0.6–1.0 °C) from 1 to 10 h post-drug administration
(see Fig. 5d). In an additional study with 32, the effects of daily dos-
ing of the compound, for 7 days on both the thermal hyperalgesia
and body temperature were examined 3 h after drug administra-
tion to determine if tolerance would develop. In this study, analge-

898
R. Palin et al. / Bioorg. Med. Chem. Lett. 21 (2011) 892–898
11. Gunthorpe, M.; Chizh, B. Drug Discovery Today 2009, 14, 56.
12. Sammes, P. G.; Thetford, D. J. Chem. Soc., Perkin Trans. 1989, 655.
13. Levy, L. M.; Gonzalo, G.; Gotor, V. Tetrahedron Asymmetry 2004, 15, 2051.
14. CHO cells transfected with human TRPV1 are loaded with Ca2+ sensitive dye. If
TRPV1 is activated by an agonist, extracellular Ca2+ can flow through the open
channel and stimulate intracellular fluorescence. Conversely, if the TRPV1
channel is exposed to the agonist capsaicin, in the presence of antagonism
from the test compound, the channel will remain closed and intracellular
fluorescence will not occur. In this way the same TRPV1 transfected cell line
and assay can be used to search for both agonists and antagonists. Results are
an average of at least two independent experiments with three replicates at
each concentration.
into animal studies. Both compounds were able to attenuate the
acute inflammatory thermal response in the rat CFA assay. How-
ever, dose related increases in body temperature in rats were ob-
served although this did tolerate out over several days. Most
TRPV1 antagonists described to date cause a modest increase in
body temperature in preclinical studies.19 However, the magnitude
and duration of temperature elevation appears dependant on the
PK profile and modality specific to blockade of TRPV1 activation.20
Given AMG-517 elevates temperature in humans8d and subse-
quently has been withdrawn from clinical development, it will be
interesting to see the clinical outcomes of future trials21 with other
TRPV1 antagonists to understand whether the hyperthermia liabil-
ity can be managed either with anti-pyretics or through shortening
the half-life of the compounds.
15. Cheung, W. S.; Calvo, R. R.; Tounge, B. A.; Zhang, S.-P.; Stone, D. R.; Brandt, M.
R.; Hutchinson, T.; Flores, C. M.; Player, M. R. Bioorg. Med. Chem. Lett. 2008, 18,
4569.
16. Aqueous solubilities were determined using
a
medium-throughput
adaptation of shake-flask methodology. 10 mM solution of the test
a
A
compound in DMSO was added to 0.05 M phosphate buffered saline pH 7.4
such that the final concentration of DMSO was 2%. The resultant mixture
was then vortex mixed (1500 rpm) for 24 0.5 h at 21 2 °C. After mixing,
the resultant solution/suspension was filtered under vacuum using a filter
Acknowledgements
plate (Millipore Multiscreen HTS, 0.4 lM). The concentration of the
compound in the filtrate was determined by High Performance Liquid
Chromatography (HPLC) running a generic acid gradient method with UV
detection at 230 nm. Peak areas from analysis of the diluted filtrates were
quantified by comparison to a calibration line prepared by injecting onto the
HPLC three different volumes of a 50 lM solution of the test compound in
DMSO. Solubilities were determined in duplicate for each test compound
and average values reported.
The work described here was conducted between November
2006 and December 2007. We would like to thank our colleagues
in the Analytical section for structure and purity determination
of all compounds.
17. PERCH software Version 1/2005, PERCH Solutions Ltd, Kuopio, Finland.
18. Methodology adapted from Vellani, V.; Mapplebeck, S.; Moriondo, A.; Davis, J.
B.; McNaughton, P. A. J. Physiol. 2001, I, 813–825. Primary cultures of rat dorsal
root ganglion neurones were prepared from adult Wistar rats. Standard whole-
cell patch clamp recordings were made from neurones on days 2–4 in culture.
The neurones were exposed to capsaicin (500 nM) applied through a local
perfusion system for 1 s, repeated every 30 s. Capsaicin evoked an inward
current which typically showed a brief period of runup in amplitude before
stabilizing. Once the amplitude was stable, test compounds were applied
through the local perfusion system. Compounds were applied until the
amplitude was judged to have plateaued over at least three successive
capsaicin applications.
19. (a) Gavva, N. R.; Bannon, A. W.; Hovland, D. N., Jr.; Lehto, S. G.; Klionsky, L.;
Surapaneni, S.; Immke, D. C.; Henley, C.; Arik, L.; Bak, A.; Davis, J.; Ernst, N.;
Hever, G.; Kuang, R.; Shi, L.; Tamir, R.; Wang, J.; Wang, W.; Zajic, G.; Zhu, D.;
Norman, M. H.; Louis, J.-C.; Magal, E.; Treanor, J. J. S. J. Pharmacol. Exp. Ther.
2007, 323, 128; (b) Gavva, N. R.; Bannon, A. W.; Surapaneni, S.; Hovland, D. N.,
Jr.; Lehto, S. G.; Gore, A.; Juan, T.; Deng, H.; Han, B.; Klionsky, L.; Kuang, R.; Le,
A.; Tamir, R.; Wang, J.; Youngblood, B.; Zhu, D.; Norman, M. H.; Magal, E.;
Treanor, J. J. S.; Louis, J.-C. J. Neurosci. 2007, 27, 3366; (c) Swanson, D. M.; Dubin,
A. E.; Shah, C.; Nasser, N.; Chang, L.; Dax, S. L.; Jetter, M.; Bitenbucher, J. G.; Liu,
C.; Mazur, C.; Lord, B.; Gonzales, L.; Hoey, K.; Rizzolio, M.; Bogenstaetter, M.;
Codd, E. E.; Lee, D. H.; Zhang, S.-P.; Chaplan, S. R.; Carruthers, N. I. J. Med. Chem.
2005, 48, 1857.
References and notes
1. Caterina, M. J.; Schumacher, M. A.; Tominaga, M.; Rosen, T. A.; Levine, J. D.;
Julius, D. Nature 1997, 389, 816.
2. (a) Ross, R. A. Br. J. Pharmacol. 2003, 140, 790; (b) Shin, J.; Cho, H.; Hwang, S. W.;
Jung, J.; Shin, C. Y.; Lee, S.-Y.; Kim, S. H.; Lee, M. G.; Choi, Y. H.; Kim, J.; Haber, N.
A.; Reichling, D. B.; Khasar, S.; Levine, J. D.; Oh, U. Proc. Natl. Acad. Sci. 2002, 99,
10150.
3. Willis, W. D. Exp. Brain Res. 2009, 96, 5.
4. Caterina, M. J.; Leffler, A.; Malmberg, A. B.; Martin, W. J.; Trafton, J.; Petersen-
Zeitz, K. R.; Koltzenburg, M.; Basbaum, A. I.; Julius, D. Science 2000, 288, 5464.
5. (a) Matthews, P. J.; Aziz, Q.; Facer, P.; Davis, J. B.; Thompson, D. G.; Anand, P.
Eur. J. Gastroenterol. Hepatol. 2004, 16, 897; (b) Lu, Y.; Fei, Y.; Hao, L.; Feng-Yu,
L.; Ji-Sheng, H.; Guo-Gang, X.; You, W. Mol. Pain 2008, 4, 61.
6. Szallasi, A.; Appendino, G. J. Med. Chem. 2004, 47, 2717.
7. Kym, P. R.; Kort, M. E.; Hutchins, C. W. Biochem. Pharmacol. 2009, 78, 211.
8. (a) Gunthorpe, M. J.; Hannan, S. L.; Smart, D.; Jerman, J. C.; Arpino, S.; Smith, G.
D.; Brough, S.; Wright, J.; Egerton, J.; Lappin, S. C.; Holland, V. A.; Winborn, K.;
Thompson, M.; Rami, H. K.; Randall, A.; Davis, J. B. J. Pharmacol. Exp. Ther. 2007,
321, 1183; (b) Surowy, C. S.; Neelands, T. R.; Bianchi, B. R.; McGaraughty, S.; El
Kouhen, R.; Han, P.; Chu, K. L.; McDonald, H. A.; Vos, M.; Niforatos, W.; Bayburt,
E. K.; Gomtsyan, A.; Lee, C.-H.; Honore, P.; Sullivan, J. P.; Jarvis, M. F.; Faltynek,
C. R. J. Pharmacol. Exp. Ther. 2008, 326, 879; (c) Charrua, A.; Cruz, C. D.;
Narayanan, S.; Gharat, L.; Gullapalli, S.; Cruz, F.; Avelino, A. J. Urol. 2009, 181,
379; (d) Gavva, N. R.; Treanor, J. J. S.; Garami, A.; Fang, L.; Surapaneni, S.;
Akrami, A.; Alvarez, F.; Bak, A.; Darling, M.; Gore, A.; Jang, G. R.; Kesslak, J. P.; Ni,
L.; Norman, M. H.; Palluconi, G.; Rose, M. J.; Salfi, M.; Tan, E.; Romanovsky, A.
A.; Banfield, C.; Davar, G. Pain 2008, 136, 202.
20. (a) Garami, A.; Shimansky, Y. P.; Pakai, E.; Oliveira, D. L.; Gavva, N. R.;
Romanovsky, A. A. J. Neurosci. 2010, 30, 1435; (b) Lehto, S. G.; Tamir, R.; Deng,
H.; Klionsky, L.; Kuang, R.; Le, A.; Lee, D.; Louis, J. C.; Magal, E.; Manning, B. H.;
Rubino, J.; Surapaneni, S.; Tamayo, N.; Wang, T.; Wang, J.; Wang, J.; Wang, W.;
Youngblood, B.; Zhang, M.; Zhu, D.; Norman, M. H.; Gavva, N. R. J. Pharmacol.
Exp. Ther. 2008, 326, 218.
9. Valenzano, K. J.; Grant, E. R.; Wu, G.; Hachicha, M.; Schmid, L.; Tafesse, L.; Sun,
Q.; Rotshteyn, Y.; Francis, J.; Limberis, J.; Malik, S.; Whittemore, E. R.; Hodges,
D. J. Pharmacol. Exp. Ther. 2003, 306, 377.
21. Rami, H. K.; Gunthorpe, M. J. Drug Discovery Today: Ther. Strateg. 2004, 1, 97.
10. Park, H.-G.; Choi, J.-Y.; Kim, M.-H.; Choi, S.-H.; Park, M.-K.; Lee, J.; Suh, Y.-G.;
Cho, H.; Oh, U.; Kim, H.-D.; Joo, Y. H.; Shin, S. S.; Kim, J. K.; Jeong, Y. S.; Koh, H.-
J.; Park, Y.-H.; Jew, S.-S. Bioorg. Med. Chem. Lett. 2005, 15, 631.