4-Aminopyrimidine tetrahydronaphthols: A series of novel vanilloid receptor-1 antagonists with improved solubility properties

Article abstract of DOI:10.1016/j.bmcl.2008.02.022

8-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-ylamino)-1,2,3,4-tetrahydronaphthalen-2-ol (4) and analogs (5-10) were shown to be potent inhibitors of human and rat TRPV1 in vitro with increased solubility over our previous series. Synthesis, SAR, and improv

Full text of DOI:10.1016/j.bmcl.2008.02.022

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 1830–1834
4-Aminopyrimidine tetrahydronaphthols: A series of novel
vanilloid receptor-1 antagonists with improved solubility properties
Elizabeth M. Doherty,a,* Daniel Retz,a Narender R. Gavva,b Rami Tamir,b
James J. S. Treanorc and Mark H. Normana
aDepartment of Chemistry Research and Discovery, One Amgen Center Drive, Thousand Oaks, CA 91320-1799, USA
bDepartment of Neuroscience, Amgen Inc., One Amgen Center Drive, Thousand Oaks, CA 91320-1799, USA
cDepartment of Neuroscience, Amgen Inc., 1120 Veterans Boulevard, South San Francisco, CA 94080, USA
Received 19 October 2007; revised 7 February 2008; accepted 8 February 2008
Available online 13 February 2008
Abstract—8-(6-(4-(Trifluoromethyl)phenyl)pyrimidin-4-ylamino)-1,2,3,4-tetrahydronaphthalen-2-ol (4) and analogs (5–10) were
shown to be potent inhibitors of human and rat TRPV1 in vitro with increased solubility over our previous series. Synthesis,
SAR, and improvements in metabolic stability and absorption of these compounds are described herein.
ꢀ 2008 Elsevier Ltd. All rights reserved.
Agonism of the ion channel vanilloid receptor-11
(TRPV1 or VR1) by capsaicin, the pungent component
naturally occurring in chili peppers, causes a rapid influx
of cations into the nerve cell setting off a cascade of
events that result in a burning sensation. This sensation
should be familiar to anyone who has enjoyed a spicy
curry. Stimulation of the neuron by this route leads ulti-
mately to desensitization to subsequent capsaicin appli-
cation as well as a decreased sensitivity to noxious
thermal stimuli. Conversely, mice lacking the TRPV1
gene, while grossly normal, exhibit a reduced response
to painful thermal stimuli in inflammatory pain models.2
These observations in knock-out mice have provided a
rationale for the pursuit of small molecule TRPV1
antagonists as a treatment for inflammatory pain.3
incorporation of groups containing a basic amine at the
2- or 6-position of the pyrimidine core of 1 resulted in
improved solubility while potency was maintained.5
We also pursued an alternative twofold approach to
improving solubility: (1) increase the basicity of the core
and (2) reduce crystallinity via partial saturation. Herein
we describe a set of tetrahydronaphthol analogs of 1
(compounds 4–10, Figs. 1 and 2) that achieved our goal
of improved solubility following this alternative
strategy.
Previous SAR studies provided us with an understand-
ing of the three key features required for optimal inter-
action of antagonists like 1 with the TRPV1 receptor: (1)
a 2-oxo or 2-amino- pyrimidine or pyridine core, (2) a
lipophilic region on one end of the molecule, and (3)
an aromatic group with hydrogen-bonding elements at
the other end. Early on in our high-throughput screen-
ing (HTS) efforts we had uncovered another series of
analogs with low nM potency, the naphthol ureas6
We have reported on a series of pyrimidine-based
TRPV1 antagonists with excellent in vitro and in vivo
properties.4 From this series, AMG 517 (1, Fig. 1) was
selected for evaluation in human clinical trials. We then
initiated a program aiming to further improve the pro-
file of the series. Specifically, due to the poor solubility
of compound 1 we encountered difficulties in formula-
tion and observed solubility-limited absorption at higher
doses in animal studies. Previously, we have shown that
exemplified by compound
2
(Fig. 1) with an
IC50 = 13 nM [as measured by inhibition of capsaicin-in-
duced 45Ca2+ influx into human TRPV1-expressing Chi-
nese hamster ovary (CHO) cells].7 This urea series
provided the same requisite key interactions with the
receptor, that is, a urea mimicking the oxopyrimidine
core flanked by a lipophilic aromatic group and a hydro-
gen bond-donating naphthol. However, the urea series
suffered from low solubility and poor pharmacokinetic
properties. Nevertheless, we felt that the naphthol
Keywords: TRPV1; VR1.
*
Corresponding author. Tel.: +1 805 447 8977; fax: +1 805 480
1337; e-mail: edoherty@amgen.com
0960-894X/$ - see front matter ꢀ 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.02.022

E. M. Doherty et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1830–1834
1831
Figure 1. Clinical candidate 1 (AMG 517), naphthol urea HTS hit 2, hybrid analog 3, tetrahydronaphthol analog 4.
Figure 2. Putative sites of metabolism on compound 4 and targets 5–10 designed to block the metabolic soft spot.
moiety provided a new direction in SAR development
for the pyrimidine series. Combining the trifluorometh-
ylphenylpyrimidine of compound 1 with the 1-amino-
naphthol of urea 2 resulted in 38 (Fig. 1), with an
IC50 = 0.65 nM. This combination of features resulted
in an increase in potency over the urea analog 2; how-
ever, we observed only a modest improvement in solu-
bility over compound 1 (Table 1). The slight but
measurable improvement in solubility, particularly in
aqueous 0.01 N HCl (0.003 mg/mL), we attributed to
the increased basicity of the aminopyrimidine versus
the oxopyrimidine core.9 By partially saturating the
naphthol ring, affording tetrahydronaphthol analog 4
(Fig. 1), a much more significant improvement in solu-
bility was gained and excellent potency was maintained.
Thus, compound 4 blocked capsaicin-induced calcium
influx into hTRPV1-expressing CHO cells with an
IC50 = 0.65 nM, in rTRPV1 with an IC50 = 3.1 nM,
and had solubilities of 0.19 mg/mL in 0.01 N HCl,
0.023 mg/mL in pH 7.4 phosphate buffered saline, and
0.060 mg/mL in simulated intestinal fluid (Table 1).10
blood flow. To determine whether the absolute configu-
ration at the chiral center had any influence on potency
or PK properties, compound 4 was resolved by chiral re-
verse-phase HPLC11 to provide the antipodes (+)-4 and
(À)-4. We observed no significant difference in potency,
solubility, or in vivo clearance for racemic 4 in compar-
ison to the single enantiomers (Table 1). For this reason,
we simplified our efforts by building the remaining SAR
with racemates.
To account for the high clearance, we conducted metab-
olite identification studies on compound 4 in isolated rat
hepatocytes.12 These studies showed that compound 4
underwent Phase I metabolism restricted to the tetrahy-
dronaphthol moiety with the formation of at least six
metabolites [M+14 and M+16]. No significant direct
Phase II conjugation was observed. We prepared a lim-
ited set of derivatives (compounds 5–10, Fig. 2), selected
based in part on synthetic accessibility, designed to
block some of the possible sites of metabolism.
The syntheses of compounds 4–10 are illustrated in
Schemes 1–3. As shown in Scheme 1, the tetrahydro-
naphthol analog 4 was prepared by heating 8-amino-
1,2,3,4-tetrahydronaphthol6b (11) with 4-chloro-6-[4-
(trifluoromethyl)phenyl]pyrimidine4a (12) in EtOH by
microwave irradiation. Reaction of 4 with N-bromosuc-
Encouraged by the improvement in solubility, we evalu-
ated the pharmacokinetic profile of compound 4 in rats.
Upon intravenous (i.v.) administration, tetrahydro-
naphthol 4 exhibited a very high rate of clearance
(CL = 3900 mL/h/kg), close to the rate of rat hepatic

1832
E. M. Doherty et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1830–1834
Scheme 1. Reagents and conditions: (a) EtOH, 160 ꢁC microwave (40%); (b) NBS, DMF (82%); (c) NCS, DMF: 6 (16%); 7 (4%).
Scheme 2. Reagents and conditions: (a) SOCl2, DCM; ethylene, AlCl3, DCM, À5 ꢁC (54%); (b) NaBH4, MeOH (77%); (c) H2NCO2(t-Bu),
Pd2(dba)3, X-Phos, Cs2CO3, dioxane, 90 ꢁC (18%); (d) 4 M HCl in dioxane (99%); (e) 4-chloro-6-[4-trifluoromethyl)phenyl]pyrimidine (12), EtOH,
160 ꢁC microwave (13%).
Scheme 3. Reagents and conditions: (a) allyl bromide, K2CO3, acetone, reflux (97%); (b) SnCl2, EtOH, 70 ꢁC (100%); (c) 1,2-dichlorobenzene, 190 ꢁC
(92%); (d) Boc2O, THF; (e) m-CPBA (21% over steps d and e); (f) NaI, acetone, reflux (29%); (g) 4 M HCl in dioxane (100%); (h) 4-chloro-6-[4-
trifluoromethyl)phenyl]pyrimidine (12), EtOH, 160 ꢁC microwave (35%); (i) H2, 10% Pd/C, MeOH (100%).
cinimide in DMF at room temperature afforded pre-
dominately the 5-bromo analog 5 in 82% yield. Alterna-
The chromanyl derivatives 9 and 10 were prepared
according to Scheme 3. Formation of the allyl ether of
2-chloro-5-nitrophenol (17) followed by tin chloride
reduction of the nitro group provided aniline 18. The
Claisen-rearranged product 19 was di-protected with
Boc anhydride to afford 20, and then oxidized with m-
CPBA to epoxide 21. NaI-catalyzed cyclization of epox-
ide 21, as described by Otani,15 afforded chroman 22.
After global deprotection with HCl in dioxane, the ani-
line was reacted with 4-chloro-6-[4-trifluoromethyl)-
phenyl]pyrimidine (12) to afford 9. Dehalogenation of
9 with hydrogen over Pd on carbon provided the final
analog 10.
tively, treatment of
4
with N-chlorosuccinimide
provided a mixture of mono- and di-chlorinated prod-
ucts that were separated by silica gel chromatography
followed by reversed-phase HPLC. In this way, modest
yields of 6 (16%) and 7 (4%) were obtained.
According to Scheme 2, following a procedure described
by Wikstrom et al.,13 Friedel-Crafts acylation of ethyl-
ene with the acid chloride of 13 followed by ring closure
provided 8-bromo-6-fluoro-2-tetralone (14). The tetra-
lone was reduced to alcohol 15 with sodium borohy-
dride. Palladium-mediated amidation14 of bromide 15
with tert-butylcarbamate provided N-Boc-6-fluoro-8-
aminotetrahydronaphthol (16). The aniline was revealed
by deprotection with HCl in dioxane and reacted with 4-
chloro-6-[4-trifluoro-methyl)phenyl]-pyrimidine (12) to
afford 8.
The compounds were tested for their ability to block the
capsaicin-induced uptake of 45Ca2+ in human or rat
TRPV1-expressing (hTRPV1 or rTRPV1) CHO cells.7
Functional activity reported as mean IC50 SEM
(nM) is summarized in Table 1. Results are the average

E. M. Doherty et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1830–1834
1833
Table 1. In-vitro inhibition of capsaicin-induced 45Ca2+ influx in hTRPV1 or rTRPV1-expressing CHO cells, solubility data, and in-vivo
pharmacokinetic profile for compounds 1 and 3–10
Compound
hTRPV1
a
IC50 (nM)
rTRPV1
a
IC50 (nM)
Solubilityb (mg/mL)
CL, i.v.
(mL/h/kg)
t1/2,
i.v. (h)
F,
p.o. (%)
AUC, p.o.
(ng h/mL)
0.01 N HCl
PBSc
SIFd
1
0.76 0.04
0.65 0.08
0.65 0.13
0.58 0.04
0.62 0.01
2.1 0.5
0.9 0.8
4.8 0.4
3.1 0.2
5.7 0.5
5.5 1.4
9.6 1.2
5.6 0.5
11 1.5
<0.0001
0.003
0.19
<0.0001
<0.0001
0.023
0.00066
0.010
0.060
0.060
0.085
0.073
0.032
0.030
—
190
> 4000
3900
4100
5800
390
500
—
6.3
0.7
1.2
1.2
0.9
5.5
6.6
—
32e
—
5300e
—
3
4
(+)-4
—
—
0.18
0.16
0.0056
0.0091
0.0077
0.027
(À)-4
5
0.0072
0.022
0.071
—
60f
34f
—
—
—
—
7600f
3400f
—
6
1.4 0.1
1.1 0.3
7
0.0125
—
8
1.1 0.6
2.9 0.6
43
28 11
8
—
—
—
—
—
—
9
10
0.136
0.200
0.0056
0.015
0.017
0.063
10
1
190 50
—
—
—
a Values are means of at least three measurements, standard error of the mean (SEM).
b Thermodynamic solubility measured in a high-throughput automated format (Ref. 10).
c Phosphate buffered saline, pH 7.4.
d Simulated intestinal fluid, pH 6.8.
e 3 mg/kg dose in 5% Tween 80/OraPlus.
f mg/kg dose in 5% Tween 80/Oraplus.
of at least three measurements. All compounds were
tested in a separate assay for agonist activity; the com-
pounds showed no agonist activity. Solubility was mea-
sured using a high-throughput SYMYX platform10 in
three aqueous media: 0.01 N HCl, pH 7.4 phosphate
buffered saline (PBS), and pH 6.8 simulated intestinal
fluid (SIF).
ity in aqueous 0.01 N HCl (0.136 and 0.200 mg/mL,
respectively). Generally, the structural modifications
made to 4 were better tolerated by the human receptor
than the rat receptor. The combined properties of com-
pounds 5 and 6 (potency, solubility, clearance and oral
absorption) made these two compounds the best in the
series.
Each of the modifications made to compound 4 (e.g.
compounds 5–10) was tolerated to some extent, antago-
nist activity was maintained in the series and, for those
compounds analyzed, solubility in all three media was
significantly increased in comparison to compounds 1
and 3. The 5-bromotetrahydronaphthol analog 5 was
only threefold less potent than 4 in the human and rat
TRPV1 assays (IC50 = 2.1 and 9.6 nM, respectively).
Introduction of the bromine atom resulted in the desired
reduction in clearance [t1/2 = 5.5 h; CL = 390 mL/h/kg].
Compound 5, dosed orally (p.o.) in rats as a 5 mg/kg
suspension in 5% Tween 80/Oraplus, demonstrated good
absorption [Foral = 60%; AUC = 7600 ng h/mL]. In
comparison to compound 4, however, the solubility of
5 in 0.01 N HCl and PBS decreased (0.0072 and
0.0077 mg/mL, respectively). This result can be ex-
plained by increased lipophilicity imparted by the bro-
mine atom. The 5-chloro analog 6 showed similar
potencies to 4 and 5, with IC50 = 1.4 nM in hTRPV1
and IC50 = 5.6 nM in rTRPV1. Like compound 5, the
pharmacokinetic profile of 6 [t1/2= 6.6 h; CL = 500 mL/
h/kg; Foral = 34%] was an improvement over 4. In addi-
tion, the solubility of 6 in 0.01 N HCl and PBS (0.022
and 0.027 mg/mL, respectively) was improved over com-
pound 5. The isomeric 7-chlorotetrahydronaphthol ana-
log 7 demonstrated no significant advantage over 6 with
regard to potency or solubility. The 6-fluoro analog 8
was more than 10-fold less potent in the rat TRPV1 as-
say relative to 4. The chromanyl analogs 9 and 10 were
also over one to two orders of magnitude less potent in
the rTRPV1 assay (IC50 = 28 and 190 nM, respectively)
although these analogs demonstrated very good solubil-
In conclusion, by increasing the pKa of 1 via the basic
2-aminopyrimidine core, incorporating tetrahydro-
naphthol as a replacement for N-acylamino-benzothia-
zole, and adding halogen atoms to block metabolism,
we have designed 4-aminotetrahydronaphthol pyrimi-
dines (e.g., 5 and 6) with excellent in-vitro potencies,
good pharmacokinetic properties, and improved solubil-
ities over clinical candidate AMG 517 (1).
Acknowledgments
We thank Sekhar Surapaneni and the PKDM team for
supplying the in-vitro and in-vivo pharmacokinetic data.
We also thank Maggie Reed and Helming Tan for gen-
erating the solubility data.
References and notes
1. Szallasi, A.; Blumberg, P. M. Pharmacol. Rev. 1999, 51,
159.
2. (a) 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, 306; (b)
Davis, J. B.; Gray, J.; Gunthrope, M. J.; Hatcher, J. P.;
Davey, P. T.; Overend, P.; Harries, M. H.; Latcham, J.;
Clapham, C.; Atkinson, K.; Hughes, S. A.; Rance, K.;
Grau, E.; Harper, A. J.; Pugh, P. L.; Rogers, D. C.;
Bingham, S.; Randall, A.; Sheardown, S. A. Nature 2000,
405, 183.
3. (a) Immke, D. C.; Gavva, N. R. Semin. Cell Dev. Biol.
2006, 17, 582; (b) Szallasi, A.; Cortright, D. N.; Blum, C.
A.; Eid, S. R. Nat. Rev. Drug Discov. 2007, 6, 357; (c)

1834
E. M. Doherty et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1830–1834
Westaway, S. M. J. Med. Chem. 2007, 50, 2589; (d) Rami,
H. K.; Gunthorpe, M. J. Drug Discov. Today: Ther.
Strategies 2004, 1, 97.
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. J.
Med. Chem. 2005, 48, 71.
4. (a) 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. WO 04014871 2004; (b) Doherty, E. M.; Bannon, A. W.;
Bo, Y.; Chen, N.; Dominguez, C.; Falsey, J.; Fotsch, C.;
Gavva, N. R.; Katon, J.; Nixey, T.; Ognyanov, V. I.;
Pettus, L.; Rzasa, R.; Stec, M.; Surapaneni, S.; Tamir, R.;
Zhu, J.; Treanor, J. J. S.; Norman, M. H. J. Med. Chem.
2007, 50, 3515.
5. (a) Wang, H.-L.; Katon, J.; Balan, C.; Bannon, A. W.;
Bernard, C.; Doherty, E. M.; Dominguez, C.; Gavva, N.
R.; Gore, V.; Ma, V.; Nishimura, N.; Surapaneni, S.;
Tang, P.; Tamir, R.; Thiel, O.; Treanor, J. J. S.; Norman,
M. H. J. Med. Chem. 2007, 50, 3528; (b) Wang, X.;
Chakrabarti, P. P.; Ognyanov, V. I.; Pettus, L. H.; Tamire,
R.; Tan, H.; Tang, P.; Treanor, J. J. S.; Gavva, N. R.;
Norman, M. H. Bioorg. Med. Chem. Lett. 2007, 17, 6539.
6. Subsequent to our identification of naphthol ureas from
HTS, examples of naphthol and tetrahydronaphthol ureas
as TRPV1 antagonists appeared in the patent literature:
(a) Yura, T.; Mogi, M.; Ikegami, Y.; Masuda, T.;
Kokubo, T.; Urbahns, K.; Lowinger, T. B.; Yoshida, N.;
Freitag, J.; Meier, H.; Wittka-Nopper, R.; Marumo, M.;
Shiroo, M.; Tajimi, M.; Takeshita, K.; Moriwaki, T.;
Tsukimi, Y. WO 014064 2003; (b) Tajimi, M.; Kokubo,
T.; Shiroo, M.; Tsukimi, Y.; Yura, T.; Urbahns, K.;
Yamamoto, N.; Mogi, M.; Fujishima, H.; Masuda, T.;
Yoshida, N.; Moriwaki, T. WO 052846 2004.
8. Compound 3 was prepared as described in Ref. 4a.
9. For comparison, the reported pKa values for 4-methoxy-
pyrimidine and 4-aminomethylpyrimidine are 2.5 and 6.1,
respectively. Brown, D. J.; Short, L. N. J. Chem. Soc.
1953, 331.
10. Tan, H.; Semin, D.; Wacker, M.; Cheetham, J. JALA
2005, 10, 364.
11. The enantiomers were separated by preparative chiral
phase HPLC on a ChiralcelTM AD column; mobile phase
25% EtOH in hexane; flow rate 30 mL/min; (+)-4 was the
first to elute. The isolated antipodes were >99% enantio-
meric purity as determined by chiral analytical LC.
12. In vitro metabolism was examined by incubating com-
pound 4 at a concentration of 10 lM with rat 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 · 250 mm, 5 lm) 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.
13. Stjernlof, P.; Ennis, M. D.; Hansson, L. O.; Hoffman, R. L.;
Ghazal, N. B.; Sundell, S.; Smith, M. W.; Svensson, K.;
Carlsson, A.; Wikstrom, H. J. Med. Chem. 1995, 38, 2202.
14. Yin, J.; Buchwald, S. L. Org. Lett. 2000, 2, 1101.
15. Kawamura, K.; Ohta, T.; Otani, G. Chem. Pharm. Bull.
1990, 38, 2088.
7. For detailed assay conditions see Doherty, E. M.;
Fotsch, C.; Bo, Y.; Chakrabarti, P. P.; Chen, N.;