Identification and synthesis of 2,7-diamino-thiazolo[5,4-d]pyrimidine derivatives as TRPV1 antagonists

Article abstract of DOI:10.1016/j.bmcl.2008.11.024

We have identified and synthesized a series of 2,7-diamino-thiazolo[5,4-d]pyrimidines as TRPV1 antagonists. An exploration of the structure-activity relationships at the 2-, 5-, and 7-positions of the thiazolo[5,4-d]pyrimidine led to the identification of

Full text of DOI:10.1016/j.bmcl.2008.11.024

Bioorganic & Medicinal Chemistry Letters 19 (2009) 40–46
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification and synthesis of 2,7-diamino-thiazolo[5,4-d]pyrimidine
derivatives as TRPV1 antagonists
a
a
a
b
Alec D. Lebsack a, , Bryan J. Branstetter , Michael D. Hack , Wei Xiao , Matthew L. Peterson ,
Nadia Nasser a, Michael P. Maher a, Hong Ao a, Anindya Bhattacharya a, Mena Kansagara a,
Brian P. Scott a, Lin Luo a, Raymond Rynberg a, Michele Rizzolio a, Sandra R. Chaplan a,
Alan D. Wickenden a, J. Guy Breitenbucher a
*
a Johnson & Johnson Pharmaceutical Research and Development L.L.C., 3210 Merryfield Row, San Diego, CA 92121, USA
b TransForm Pharmaceuticals, 29 Hartwell Avenue, Lexington, MA 02421, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We have identified and synthesized a series of 2,7-diamino-thiazolo[5,4-d]pyrimidines as TRPV1 antag-
onists. An exploration of the structure–activity relationships at the 2-, 5-, and 7-positions of the thiazol-
o[5,4-d]pyrimidine led to the identification of several potent TRPV1 antagonists, including 3, 29, 51, and
57. Compound 3 was orally bioavailable and afforded a significant reversal of carrageenan-induced ther-
mal hyperalgesia with an ED50 = 0.5 mg/kg in rats.
Received 26 September 2008
Revised 5 November 2008
Accepted 10 November 2008
Available online 13 November 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
TRPV1
Antagonists
Thiazolo-pyrimidine
The vanilloid receptor 1 (VR1 or TRPV1) is the best character-
ized member of the transient receptor potential family of ion chan-
nels.1,2 TRPV1 is a ligand-gated, non-selective cation channel that
is primarily expressed in nociceptive C- and Ad fibers. The TRPV1
receptor is activated by a wide range of stimuli including heat
(>43 °C), low pH, vanilloid ligands such as capsaicin and resinifera-
toxin (RTX), and a wide range of endogenous mediators such as
bradykinin and anandamide.3 In general, activation of the TRPV1
channel results in depolarization, neuronal hyper-excitability,
and ultimately the sensation of pain.4 TRPV1 knockout mice dem-
onstrate an impaired ability to develop inflammatory thermal
hyperalgesia, suggesting that TRPV1 has an important role in trans-
mitting inflammatory pain signals.5 For this reason, efforts to
discover small molecule TRPV1 antagonists have received consid-
erable attention from many pain research groups.6
In our work toward the discovery of novel TRPV1 antagonists,
we synthesized 5,6,7,8-tetrahydro-pyrido[3,4-d]pyrimidine as
exemplified by 1 (Fig. 1).7,8 In an effort to broaden the structure–
activity relationships of compound 1, we were interested in replac-
ing the 5,6,7,8-tetrahydro-pyrido[3,4-d]-pyrimidine core with
thiazolo[5,4-d]pyrimidine core (e.g., compound 2) given its conve-
nient preparation from commercially available materials in a single
step9 and ability to functionalize the 2- and 7-positions with the
requisite groups (Fig. 1). For these reasons we initially prepared
2,7-diamino-thiazolo[5,4-d]pyrimidines 2 and 3 (Table 1). While
compound 2 with its 3-trifluoromethyl-pyridin-2-yl substituent
provided a direct comparison to compound 1, we also investigated
the introduction of 2,6-dichlorophenyl as a replacement for 3-tri-
fluoromethyl-pyridin-2-yl in compound 3.10
Compounds were evaluated for their ability to inhibit capsaicin-
induced influx of Ca2+ in cells (HEK293) expressing human and rat
TRPV1.11 Inhibition is reported as IC50 SEM (nM) and the results
are the average of at least three independent experiments. In addi-
tion, none of the compounds reported herein displayed agonist
activity. While compound 2 was essentially inactive at human
5
CF3
CF3
N
N
N
N
7
N
H
N
H
S
N
N
N
2
N
NH
CF3
1
CF3
2
hTRPV1 IC50 = 15 1 nM
rTRPV1 IC50 = 71 17 nM
* Corresponding author. Tel.: +1 858 320 3379; fax: +1 858 784 3063.
E-mail address: alebsack@its.jnj.com (A.D. Lebsack).
Figure 1. 5,6,7,8-Tetrahydro-pyrido[3,4-d]pyrimidine 1 and thiazolo[5,4-d]pyrim-
idine 2.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.11.024

A. D. Lebsack et al. / Bioorg. Med. Chem. Lett. 19 (2009) 40–46
41
Table 1
TRPV1 activity for 2,7-diamino-thiazolo[5,4-d]pyrimidines 2–3, thiazolo[4,5-d]pyrim-
idine 4, and 2,4-diamino-benzothiazole 5
Compound Structure
hTRPV1
IC50 SEM
(nM)
rTRPV1
IC50 SEM
(nM)
CF3
N
N
Figure 3. Molecular alignment of compounds 1 (carbon atoms in green) and 3
(carbon atoms in blue).
N
H
S
2
>5000
>5000
N
N
using the MCMM method in MacroModel with an upper energy
limit of 12 kcal/mol and 10,000 conformer generation steps.13 Each
resulting geometry was energy-minimized in Jaguar14 with the
NH
CF3
CF3
CF3
*
B3LYP hybrid density-functional and the 6-31G basis set. Each of
N
N
N
the conformers of 2 and 3 shown in Figures 2 and 3, respectively,
were then used to search the corresponding set of energy-mini-
mized structures obtained from the quantum calculations using
ROCS with the ComboScore option to account for similarities in
overall shape as well as in pharmacophore features.15
N
H
S
Cl
3
12
3
3
1
N
S
NH
Cl
Similar conformers to the ones shown in Figures 2 and 3 were
identified for each of the compounds 2 and 3. For compound 3,
the energy of the corresponding minimized conformer was
0.2 kcal/mol relative to the lowest-energy conformer (hereafter re-
ferred to as the quantum mechanical reference conformer of com-
pound 3). However, for compound 2 the energy of the most similar
conformer to the one shown in Figure 2 was 10.3 kcal/mol above
the lowest-energy conformer. In fact, in the lowest-energy con-
former of 2 the 3-trifluoromethyl-pyridin-2-yl group is coplanar
with the thiazole ring (Fig. 4). In the lowest-energy conformer of
3, on the other hand, the 2,6-dichlorophenyl ring is orthogonal to
the thiazole ring (Fig. 4). The preferred orientation of the N2-aro-
matic ring thus provides a reasonable hypothesis as to the lack of
TRPV1 potency observed for 2 versus 3, the former being unable
to attain the proper orthogonal orientation for affinity.
N
N
H
N
Cl
4
680 279
864 259
NH
Cl
CF3
N
H
S
Cl
5
>5000
>5000
N
NH
Cl
Having discovered compound 3 as a potent TRPV1 antagonist,
we turned our attention to determining the importance of the het-
erocyclic nitrogens in the thiazole and pyrimidine rings while
holding the 2,6-dichlorophenyl and 4-trifluoromethyl-phenyl
amine substituents constant. To that end, we prepared thiazol-
o[4,5-d]pyrimidine 4 and benzothiazole 5 (Table 1). A substantial
and rat TRPV1, the potency of compound 3 was similar at human
TRPV1 and the potency increased 24-fold at rat TRPV1 when com-
pared with reference compound 1 (Table 1).
To gain insight into the ability of the thiazolo[5,4-d]pyrimidine
core to replace the 5,6,7,8-tetrahydro-pyrido[3,4-d]pyrimidine
core, and also to understand the difference in potency between
the 3-trifluoromethyl-pyridin-2-yl substituent in compound 2
and the 2,6-dichlorophenyl substituent in compound 3, we carried
out molecular alignments of compounds 1, 2, and 3. Initially a flex-
ible molecular alignment algorithm in MOE was used to identify
pairs of conformers of 1 and 2 and of 1 and 3 with good overall
alignments (Figs. 2 and 3).12 Both pairs of conformers appear quite
reasonable, with good alignment between key features of the
thiazolo[5,4-d]pyrimidine and corresponding features of the
5,6,7,8-tetrahydro-pyrido[3,4-d]pyrimidine. Unfortunately, while
such alignments rationalize the activity of 3 relative to 1, they do
not explain the inactivity of 2, which bears the same 3-trifluoro-
methyl-pyridin-2-yl group shown to be active in 1.
Figure 4. Molecular alignment of the lowest-energy conformers of compounds 2
(carbon atoms in orange) and 3 (carbon atoms in blue).
To understand the difference between 2 and 3 in more detail,
conformation searches for each of these compounds was performed
Figure 2. Molecular alignment of compounds 1 (carbon atoms in green) and 2
Figure 5. Molecular alignment of compounds 3 (carbon atoms in blue) and 4
(carbon atoms in orange).
(carbon atoms in pink).

42
A. D. Lebsack et al. / Bioorg. Med. Chem. Lett. 19 (2009) 40–46
in Table 1, we focused our synthetic efforts on broadening the
structure–activity relationships at the 2-, 5-, and 7-positions of
the thiazolo[5,4-d]pyrimidine.
The general synthesis of 2,5,7-substituted-thiazolo[5,4-d]-
pyrimidines is outlined in Scheme 1.16 For example, addition of
4,6-dichloro-5-aminopyrimidine 6a in the presence of 2,6-dichlo-
rophenyl isothiocyanate, cesium carbonate, and CH3CN at 50 °C
afforded the 7-chloro-thiazolo[5,4-d]pyrimidine 7a in 90% yield.9
N-Methylation of 7a in the presence of methyl iodide and sodium
hydride afforded 7b in 50% yield.17 Amination of 7-chloro-thiazol-
o[5,4-d]pyrimidine 7a in the presence of 4-trifluoromethyl-phenyl
amine and p-toluene sulfonic acid (TsOH) provided compound 3 in
80% yield. The structure of compound 3 was confirmed by single-
crystal X-ray analysis.18 Installation of amine functionality at the
5-position of the thiazolo[5,4-d]pyrimidine was accomplished
through a two-step oxidation and displacement sequence from
the corresponding 5-methylsulfanyl-thiazolo[5,4-d]pyrimidine 37
to provide compounds 39–51. Alternatively, the 5-methylsulfa-
nyl-thiazolo[5,4-d]pyrimidine 37 could undergo a thioether-boro-
nic acid cross coupling to afford compound 52 in 24% yield.19
Further elaboration of the thiazolo[5,4-d]pyrimidine scaffold is
summarized in Schemes 2 and 3. Scheme 2 illustrates the synthesis
of a variety of compounds through hydrolysis or reduction of ni-
trile 19 to afford compounds 53–57. Alternative linkages between
the thiazolo[5,4-d]pyrimidine and the N7-aryl substituents were
prepared from 7-chloro-thiazolo[5,4-d]pyrimidine 7a to provide
N-methylated 58, ether 59, and thioether 60 (Scheme 3).
Figure 6. Molecular alignment of compounds 3 (carbon atoms in blue) and 5
(carbon atoms in orange).
loss in potency at TRPV1 was observed for both thiazolo[4,5-
d]pyrimidine 4 and benzothiazole 5 when compared to thiazol-
o[5,4-d]pyrimidine 3 (Table 1).
To enhance our understanding of compounds 4 and 5 with
respect to compound 3, we employed molecular alignment meth-
ods previously described by using MCMM conformation sampling
followed by quantum mechanical calculations and ROCS.13–15
Figure 5 shows the best alignment of a conformer of 4 and the quan-
tum mechanical reference conformer of compound 3. Although the
conformer of 4 has a low relative energy above its global minimum
(0.2 kcal/mol), it afforded poor overlap with the conformer of 3 in
the thiazolo-pyrimidine region. This is consistent with the loss of
potency at TRPV1.
In contrast to the poor alignment observed between the quan-
tum mechanical reference conformer of compound 3 and the low
energy conformer of 4, good alignment were observed for a low en-
ergy conformer of 5 and reference conformer of 3 as shown in
Figure 6. The conformer of 5 was 0.4 kcal/mol above its global min-
imum. Although the comparison of 3 and 5 was not able to provide
a structural rationale for the loss of TRPV1 potency in compound 5,
the comparison does underscore the importance of maintaining
the pyrimidine ring for optimal TRPV1 potency. Given the findings
The regioisomeric thiazolo-pyrimidine 4 was synthesized in 5
steps usingan improvedversion of a previously published procedure
(Scheme 4).20 Formation of the 4-amino-thiazole-5-methyl ester 62
was accomplished in two steps from commercially available
2,6-dichlorophenyl isothiocyanate. Exposure of 4-amino-thiazole-
CF3
N
N
R1
R1
N
H
S
Cl
Cl
N
N
N
N
N
a
R
NH
Cl
Cl
Cl
S
NH2
N
R2 NR3
7a: R1 = H,
53: R = CH2NH2
54: R = CONH2
19: R = CN
55: R = COOCH3
56: R = COOH
57: R = CH2OH
a
b
6a: R1 = H
6b: R1 = CH3
6c: R1 = SCH3
c
d
e
R2 = 2,6-dichlorophenyl,
b
R
3 = H
7b: R1 = H,
R2 = 2,6-dichlorophenyl,
R3 = CH3
R1
N
N
c
Scheme 2. Reagents and conditions: (a) LiAlH4, THF, rt (75%); (b) 3 equiv Dibal,
THF, rt (74%); (c) 2 equiv TsOH, toluene, 120 °C, sealed tube, 12 h (40%); (d) MeOH,
H2SO4, reflux, 48 h (91%); (e) LiOH–H2O, THF, H2O, 12 h, (72%).
R4
N
H
S
N
R2 NR3
8-36
R1
37: R1 = SCH3, R2 = 2,6-dichlorophenyl,
R3 = H, R4 = 4-trifluoromethyl-phenyl
38: R1 = SO2CH3, R2 = 2,6-dichlorophenyl,
R3 = H, R4 = 4-trifluoromethyl-phenyl
39-51
N
N
N
N
a or b
Cl
e
f
X
S
Cl
Cl
S
Cl
Cl
N
N
NH
d
NH
52: R1 = Ph, R2 = 2,6-dichlorophenyl,
R3 = H, R4 = 4-trifluoromethyl-phenyl
58: X = NMe, R1 = CF3
59: X = S, R1 = CF3
7a
60: X = O, R1 = C(CH3)3
Scheme 1. Reagents and conditions: (a) R2–N@C@S, Cs2CO3, CH3CN, 50 °C, 12 h
(70–95%); (b) 1 equiv MeI, NaH, THF, rt, (50%); (c) 1 equiv H2N–R4, TsOH, toluene,
120 °C, sealed tube, 2 h (60–90%); (d) 1.1 equiv phenyl boronic acid, 1.3 equiv
copper(I)-thiophene-2-carboxylate, tri-2-furylphosphine (16 mol%), Pd2(dba)3
(4 mol%), THF, 50 °C, 24 h (24%); (e) 4 equiv Oxone, MeOH–THF–H2O (1:1:1), rt,
48 h, (90%); (f) 3 equiv amine (R1), t-Amyl-OH, 120 °C, sealed tube, 12 h (50–90%).
Scheme 3. Reagents and conditions: (a)
1 equiv methyl-(4-trifluoromethyl-
phenyl)-amine, 2.2 equiv HCl in IPA (1.25 M), IPA, reflux, 6 h (90% of 58); (b) 4-
trifluoromethyl-benzenethiol or 4-tert-butyl-phenol, K2CO3, CH3CN, 90 °C, 12 h
(90% of 59 or 20% of 60, respectively).

A. D. Lebsack et al. / Bioorg. Med. Chem. Lett. 19 (2009) 40–46
43
Table 2
S
C
N
NH2
S
TRPV1 activity for N2-phenyl ring substituted thiazolo[5,4-d]pyrimidines 3, 8–19, and
O
53–57
Cl
N
OCH3
CF3
N
N
a, b
N
c, d
Cl
Cl
NH
N
H
S
6
2
R1
N
Cl
NR2
61
62
N
e
4
Cl
Compound R1
R2
hTRPV1
rTRPV1
Cl
N
S
IC50 SEM (nM) IC50 SEM (nM)
NH
3
8
9
2,6-Dichloro
2,6-Dichloro
2,6-Dimethyl
2-Chloro-6-Methyl
2-Cl
2-SCH3
2-SO2CH3
2-CH3
2-CF3
3-CF3
4-CF3
H
2,6-Dichloro-4-CN
2,6-Dichloro-4-CH2NH2
2,6-Dichloro-4-CONH2
2,6-Dichloro-4-COOCH3
2,6-Dichloro-4-COOH
2,6-Dichloro-4-CH2OH
H
12
3
3
1
CH3 502 257
540 132
6
1.6 0.4
32
26 17
221 29
14 3.0
37 10
Cl
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
23 11
9
2
63
10
11
12
13
14
15
16
17
18
19
53
54
55
56
57
5
64 27
43 31
234 31
51 23
74 26
>5000
5
Scheme 4. Reagents and conditions: (a) NH2CN, sodium methoxide, rt; (b) chloro-
acetic acid methyl ester, 50 °C, 24 h (77% over two steps); (c) formamide, Ac2O (cat),
170 °C, sealed tube, 18 h; (d) POCl3, 90 °C, 6 h (26% over two steps); (e) 1 equiv 4-
trifluoromethyl-phenyl amine, 2.2 equiv HCl in IPA (1.25 M), IPA, 90 °C, 6 h (40%).
>5000
5-methyl ester 62 to formamide at 170 °C followed by chlorination
in the presence of POCl3 afforded the 7-chloro-thiazolo[4,5-d]-
pyrimidine 63 in 26% yield over two steps. Amination of 7-chloro-
thiazolo[4,5-d]pyrimidine 63 in the presence of HCl and 4-trifluoro-
methyl-phenyl amine provided compound 4 in 40% yield.
Scheme 5 describes the preparation of 2,4-diamino-benzothia-
zole 5 through a 3 step sequence. Treatment of 2,6-dichlorophenyl
amine with sodium hydride followed by addition of 2,4-dichloro-
benzothiazole 64 and di-tert-butyl dicarbonate (Boc2O) provided
2-amino-4-chloro-benzothiazole 65 in 90% yield. Palladium medi-
ated amination of 65 followed by removal of the tert-butoxycar-
bonyl group gave 2,4-diamino-benzothiazole 5.
>5000
>5000
1700 320
2680 690
53 19
54 24
57 13
3190 2130
>5000
0.5 0.2
95 58
125 50
69 27
820 300
>5000
1.0 0.4
Having discovered the 2,6-dichlorophenyl as one of the optimal
substituents at the N2-position, we held this group constant and
turned our attention to the 7-position of the thiazolo[5,4-d]pyrim-
idine (Table 3). Initially, we prepared compounds 58–60 to inves-
tigate the importance of the amino linker between the
To firmly establish what type of phenyl substitution at the N2-
position of compound 3 was required for optimal TRPV1 potency,
compounds 8–19 and 53–57 were evaluated (Table 2). To gain in-
sight into the importance of a hydrogen bond donor at the N2-posi-
tion, we replaced the hydrogen at the R2-position with a methyl
group to provide compound 8. Compound 8 led to a >42-fold loss
in TRPV1 potency when compared with compound 3. Holding the
hydrogen at the R2-position constant, we turned our attention to
the N2-phenyl ring. In general, 2,6-disubstitution was preferred
over 2-substitution on the N2-phenyl ring. For example, 2,6-dichlo-
rophenyl 3 was >5-fold more potent in human and rat TRPV1 than
2-chlorophenyl 11. From the structure–activity trends in com-
pounds 15–18, it was apparent that 2-substituted phenyl rings
were required for potency. For example, placement of a trifluoro-
methyl group at the 3 or 4 position of N2-phenyl ring in analogues
16 and 17 led to a complete loss of human TRPV1 potency whereas
2-trifluoromethyl analogue 15 afforded a TRPV1 IC50 less than
100 nM. In addition, compound 18 with no substitution on the
phenyl ring lacked significant TRPV1 potency. Interestingly, a com-
bination of substitution at the 2-, 4-, and 6-positions of the N2-phe-
nyl ring resulted in hydroxymethyl analogue 57, a compound with
improved TRPV1 potency as compared to compound 3.
Table 3
TRPV1 activity for N7-substituted thiazolo[5,4-d]pyrimidines 3, 20–35, and 58–60
Y
4
3
R
N
N
7
X
S
Cl
Cl
2
N
NH
Compound
X
Y
R
hTRPV1
IC50 SEM (nM)
rTRPV1
IC50 SEM (nM)
3
58
59
60
20
21
22
23
24
25
26
27
28
29
30
31
NH
CH 4-CF3
12
>5000
>5000
>5000
206 146
>5000
176 93
14 3.0
33 16
10 5.0
11 4.0
31 22
96 66
3
3
1
NCH3 CH 4-CF3
>1000
>5000
>5000
192 141
>5000
S
CH 4-CF3
O
CH 4-C(CH3)3
CH 3-CF3
CH 2-CF3
CH
N
CH 4-CH3
CH 4-CH(CH3)2
CH 4- C(CH3)3
CH 4-C(CH3)2COOH
CH 4-SCH3
CH 4-SO2CH3
CH 4-COOH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
H
170 97
4
4-CF3
1
32 13
3.0 0.6
1.0 0.3
2.0 0.2
12 1.0
b, c
a
5
5
2
3
2
Cl
Cl
S
Cl
Cl
S
>5000
>5000
>5000
>5000
N
N
CH 4-CONH2
NBoc
Cl
32
33
34
35
NH
NH
NH
NH
CH
>5000
>5000
4
CO N
65
64
CH 4-SO2NH2
CH
>5000
590 130
0.5 0.1
Scheme 5. Reagents and conditions: (a) 2,6-dichloro-phenylamine, NaH, DMF, rt,
30 min, followed by Boc2O, 12 h (90%); (b) 1.1 equiv 4-trifluoromethyl-phenyl
1.0 0.3
21 2.0
4
SO2 N
amine,
Pd(OAc)2
(10 mol%),
2-di-tert-butylphosphino-2’-(N,N-dimethyl-
amino)biphenyl (30 mol%), 1.4 equiv K3PO4, toluene, 90 °C, 12 h (20%); (c) 4 N HCl
CH 3-Cl-4-CF3
9
2
in dioxane, rt, 6 h (95%).

44
A. D. Lebsack et al. / Bioorg. Med. Chem. Lett. 19 (2009) 40–46
Table 4
compounds 30–33 led to a marked loss of TRPV1 potency, several
exceptions included acid 27, sulfone 29, and sulfonamide 34.
Finally, we explored the influence of substituents at the C5-po-
sition of the thiazolo[5,4-d]pyrimidine while holding the N2-2,6-
dichlorophenyl and N7-4-trifluoromethyl-phenyl substituents con-
stant (Table 4). Replacement of the C5 hydrogen in compound 3
with a methyl group in compound 36 resulted in a slight improve-
ment in potency (4-fold) in human TRPV1. However, installation of
phenyl, methylsulfanyl, or methylsulfonyl at the C5-position dete-
riorated human TRPV1 potency by >19-fold for compounds 52, 37,
38 when compared to compound 3. In an effort to improve the
aqueous solubility of compound 3 (fasted-state simulated intesti-
TRPV1 activity for 5-substituted thiazolo[5,4-d]pyrimidines 36–52
R
CF3
N
N
5
N
H
S
Cl
Cl
N
NH
Compound
R
hTRPV1 IC50 SEM
(nM)
rTRPV1 IC50 SEM
(nM)
3
36
52
37
38
–H
–CH3
–Phenyl
–SCH3
–SO2CH3
12
3
539 379
224 14
274 110
3
1
3
2
1
1
nal fluid (SIF) = 2 lg/mL and simulated gastric fluid (SGF) = <1 lg/
mL)21 we introduced amino substituents at the C5-position of the
thiazolo[5,4-d]pyrimidine. Introduction of a variety of tertiary
amines at the C5-position resulted in similar or decreased TRPV1
potency (1- to 5-fold) for compounds 39–42 and 45–47 and little
improvement in solubility when compared to compound 3. Instal-
lation of secondary amines 49 and 50 led to an increase in TRPV1
potency (4- to 6-fold) and little improvement in solubility relative
to compound 3. When compared to compounds 3 and 39–50, com-
pound 51 afforded best balance between TRPV1 potency and im-
104 51
113 56
77 23
39
40
41
59 32
25
7
N
N
N
12
5
36 28
33 13
63 11
proved solubility (SIF = 10 lg/mL and SGF = 116 lg/mL). These
results reveal that the 5-position of the thiazolo[5,4-d]pyrimidine
was tolerant of a variety of amino substituents and the aqueous
solubility could be improved through incorporation of amino
substituents.
42
43
30
8
16
3
N
N
O
>5000
>5000
NH
Compounds 3, 29, 51, and 57 were selected for pharmacokinetic
assessment in Sprague–Dawley rats since they exhibited excellent
TRPV1 potency and each represented a distinct change to the 2-, 5-,
or 7-position of the thiazolo[5,4-d]pyrimidine (Table 5). Although
compounds 29, 51, and 57 demonstrated high to moderate rates
of clearance (CL = 3.6 L/h/kg, CL = 1.8 L/h/kg, and CL = 1.4 L/h/kg,
respectively), we were pleased to discover that compound 3 affor-
ded a low rate of clearance (CL = 0.6 L/h/kg) and was well absorbed
with a bioavailability of 43%.
44
45
46
184 78
59 36
48 15
119 68
43 19
34 15
N
N
N
N
47
41 23
54 38
N
O
Given the rat pharmacokinetic profiles of compounds 3, 29, 51,
and 57, compound 3 was chosen for further evaluation.22 Since
TRPV1 receptors are also activated by endogenous factors such as
low pH, we evaluated the ability of compound 3 to inhibit pro-
ton-induced TRPV1 activation in HEK293 cells expressing human
and rat TRPV1 (Table 6).11 In addition, we also investigated the
ability of compound 3 to displace [3H]-RTX in HEK293 cells
expressing rat TRPV1 (Table 6).11 Compound 3 blocked proton acti-
vation of the channel to a similar extent as its ability to block cap-
saicin activation and the Ki was consistent with functional activity.
48
49
50
13
2
6
1
2
9
4
NH
1.0 0.3
NH
NH
3
3
2
O
51
N
39 22
11 5.0
N
H
Table 5
thiazolo[5,4-d]pyrimidine and the aryl substituent in compound 3.
Introduction of a variety of linkages including N7-methylated com-
pound 58, thioether 59, and ether 60 led to a substantial loss in
TRPV1 potency, suggesting possible hydrogen bond donor interac-
tions with the binding site of the receptor. Given the importance of
this interaction, we held X as an NH group and explored substitu-
tion of the N7-aryl substituent. In general, 4-substituted aryl
groups were favored over their 2- and 3-substituted counterparts.
For example, 2-trifluoromethyl-phenyl 21 was essentially inactive
and 3-trifluoromethyl-phenyl 20 was >17-fold less potent than 4-
trifluoromethyl-phenyl 3. Although compound 20 lost TRPV1 po-
tency when compared to compound 3, it did suggest that substit-
uents at the 3-position of N7-phenyl could be tolerated. An
investigation of this effect led to the discovery of 3-chloro-4-tri-
fluoromethyl-phenyl 35 which had TRPV1 potency similar to that
of compound 3. When compared to compound 3, potency was gen-
erally maintained with lipophilic functionality such as 4-alkyl ana-
logues 24–26. Although installation of polar functionality in
Mean pharmacokinetic parameters for compounds 3, 29, 51, and 57 following 0.5 mg/
kg iv and 2 mg/kg po administration in fasted Sprague–Dawley rats using 5%
pharmasolve/20% RH-40 cremophor/75% dextrose (5%) in water as a vehicle
Compound
Cl
Vss
(L/kg)
T1/2
(h)
po
Cmax
po
Tmax (h)
po AUCinf
(h à g/L)
10049
BLOQ
196
%F
(L/h/kg)
(
lM)
l
3a
29
51
57
0.6
3.6
1.8
1.4
1.6
2.2
2.8
2.9
2.2
1.2
1.8
1.8
1.1
2
—
2
43
—
17
45
BLOQ
0.07
0.2
4
638
a
1 mg/kg iv and 5 mg/kg po administration in fasted Sprague–Dawley rats.
Table 6
Additional TRPV1 activity for compound 3
Compound hTRPV1 (pH)
IC50 SEM (nM)
rTRPV1 (pH)
IC50 SEM (nM)
rRTX binding
Ki SEM (nM)
3
19 11
6
4
4
3

A. D. Lebsack et al. / Bioorg. Med. Chem. Lett. 19 (2009) 40–46
45
demonstrated the importance of a thiazolo[5,4-d]pyrimidine core
to maintain TRPV1 potency over related analogues such as thiazol-
o[4,5-d]pyrimidine 4 or a benzothiazole 5. An exploration of the
structure–activity relationships at the 2-, 5-, and 7-positions of
the thiazolo[5,4-d]pyrimidine led to the identification of several
potent TRPV1 antagonists including 3, 29, 51, and 57. Further eval-
uation of compound 3 in a rat model of carrageenan-induced ther-
mal hyperalgesia afforded
a significant reversal of thermal
hyperalgesia with an ED50 = 0.5 mg/kg.
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) Montell, C.; Birnbaumer, L.; Flockerzi, V.; Bindels, R. J.; Bruford, E. A.;
Caterina, M. J.; Clapham, D. E.; Harteneck, C.; Heller, S.; Julius, D.; Kojima, I.;
Mori, Y.; Penner, R.; Prawitt, D.; Scharenberg, A. M.; Schultz, G.; Shimizu, N.;
Zhu, M. X. Mol. Cell 2002, 9, 229; (b) Gunthorpe, M. J.; Benham, C. D.; Randall,
A.; Davis, J. B. Trends Pharmacol. Sci. 2002, 23, 183.
Figure 7. Functional effect of compound 3 (n = 4) in isolated rat bladder compared
with vehicle (n = 6). Tissue contraction is expressed as percentage of KCl (70 mM)-
induced tone.
3. (a) Szallasi, A.; Blumberg, P. M. Pharmacol. Rev. 1999, 51, 159; (b) Appendino,
G.; Munoz, E.; Fiebich, B. L. Expert Opin. Ther. Patents 2003, 13, 1825; (c)
Caterina, M. J.; Julius, D. Annu. Rev. Neurosci. 2001, 24, 487.
4. See Refs. 1,2, and 6b.
5. (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.; Gunthorpe, 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.
6. For recent TRPV1 reviews see (a) Breitenbucher, J. G.; Chaplan, S. R.; Carruthers,
N. I. Annu. Rep. Med. Chem. 2005, 40, 185; (b) Westaway, S. A. J. Med. Chem.
2007, 50, 2589; (c) Szallasi, A.; Gharat, L. A. Expert Opin. Ther. Patents 2008, 18,
159.
7. The synthesis and structure–activity relationships of 5,6,7,8-tetrahydro-
pyrido[3,4-d]pyrimidines as TRPV1 antagonists will be reported in due
course. In addition to our own efforts, preparations of 5,6,7,8-tetrahydro-
pyrido[3,4-d]pyrimidines as TRPV1 receptor modulators have been reported in
the patent literature. (a) Kelly, M. G.; Janagani, S.; Wu, G.; Kincaid, J.; Lonergan,
D.; Fang, Y.; Wei, Z. PCT Int. Appl. WO2005066171, 2005. (b) Norman, M. H.;
Ognyanov, V. I.; Pettus L. H. US Patent 0165032, 2005. (c) Lee, C.; Brown, B. S.;
Keddy, R. G.; Perner, R. J.; Koenig, J. R. PCT Int. Appl. WO2006062981, 2006.
8. For accounts of TRPV1 antagonists with other relevant heterocyclic cores see:
(a) Bakthavatchatam, R.; Blum, C. A.; Brielmann, H. L.; Caldwell, T. M.; De
Lombaert, S. PCT Int. Appl. WO2003062209, 2003. (b) Allison, B. D.; Branstetter,
B. J.; Breitenbucher, J. G.; Hack, M. D.;Hawryluk, N. A.; Lebsack, A. D.; McClure,
K. J.; Merit, J. E. PCT Int. Appl. WO2007109355, 2007.
9. For single step synthesis of 2-amino-7-chloro-thiazolo[5,4-d]pyrimidine see
Liu, J.; Patch, R. J.; Schubert, C.; Player, M. R. J. Org. Chem. 2005, 70, 10194.
10. (a) Lebsack, A. D.; Branstetter, B. J.; Xiao, W.; Hack, M. D.; Nasser, N.; Maher, M.
P.; Ao, H.; Bhattacharya, A.; Kansagara, M.; Scott, Brian P.; Luo, L.; Chaplan, S.
R.; Wickenden, A. D.; Breitenbucher, J. G. Abstracts of Papers, 236th ACS
National Meeting, Philadelphia, PA, United States, August 17–21, 2008; MEDI-
467.; Thiazole-4-carboxamides containing a 2,6-dichlorophenyl group have
been reported as TRPV1 antagonists (b) Xi, N.; Bo, Y.; Doherty, E. M.; Fotsch, C.;
Gavva, N. R.; Han, N.; Hungate, R. W.; Klionsky, L.; Liu, Q.; Tamir, R.; Xu, S.;
Treanor, J. J. S.; Norman, M. H. Bioorg. Med. Chem. Lett. 2005, 15, 5211.
11. Bhattacharya, A.; Scott, B. P.; Nasser, N.; Ao, H.; Maher, M. P.; Dubin, A. E.;
Swanson, D. M.; Shankley, N. P.; Wickenden, A. D.; Chaplan, S. R. J. Pharmacol.
Exp. Ther. 2007, 323, 665.
Figure 8. Representative data showing the effect of 3 on carrageenan-induced
thermal hyperalgesia in rats, n = 6/group (dosed po in 5% pharmasolve/20% RH-40
cremophor/75% dextrose (5%) in water, administered 1h prior to carrageenan
injection).
In addition to the in vitro data obtained for compound 3 in re-
combinant systems, we also examined compound 3 in a native tis-
sue (Fig. 7).11 Isolated longitudinal strips of rat bladder were used
to record capsaicin-induced isometric contractions. The potency
(EC50) of capsaicin shifted from 70 to 420 nM (6-fold) in the pres-
ence of 100 nM of compound 3, without suppression of the maxi-
mal tissue contractility.
When dosed orally, compound 3 was efficacious in a carra-
geenan-induced thermal hyperalgesia model in rats (Fig. 8).23
Compound 3 significantly prevented development of thermal
hyperalgesia at 1, 3, and 30 mg/kg compared with vehicle
(p < 0.001). When expressed as % maximal possible effect (%
MPE), the maximal degree of inhibition was approximately 57%
(in rats treated with 3 mg/kg at 4 h post-carrageenan injection).
The dose required for 50% of the maximum effect achieved by com-
pound 3 in the carrageenan model was approximately 0.5 mg/
kg po. Compound 3 did not affect basal paw withdraw latency of
the contralateral hind paws. Compound 3 plasma concentrations
determined at the end of each experiment (approximately 5 h
post-carrageenan injection) were 10 2 nM (n = 6), 58 17 nM
(n = 7), 428 40 nM (n = 7), and 3052 444 nM (n = 7) in animals
dosed orally with 0.3, 1, 3, and 30 mg/kg, respectively.
12. MOE (The Molecular Operating Environment) Version 2006.08, software
available from Chemical Computing Group Inc., 1010 Sherbrooke Street
West, Suite 910, Montreal, Canada H3A 2R7. http://www.chemcomp.com.
13. MacroModel, version 9.5, Schrodinger, LLC, New York, NY, 2007.
14. Jaguar, version 7.0, Schrodinger, LLC, New York, NY, 2007.
15. ROCS (Rapid Overlay of Chemical Structures), version 2.3.1, OpenEye Scientific
Software, Inc., Santa Fe, NM, USA, http://www.eyesopen.com, 2007.
16. Branstetter, B. J.; Breitenbucher, J. G.; Lebsack, A. D.; Xiao, W. US Patent Appl.
2008, 004253.
17. Spectroscopic methods including 2D NMR 1H/13C HSQC and HMBC were used
to confirm the structure of compound 8 and therefore 7b. In the HMBC, a
strong 3-bond coupling was observed between the proton on the N-methyl and
the two adjacent quaternary carbons for compound 8.
CF3
N
N
N
H
S
N
Cl
Cl
N
H
In summary, we have identified a series of 2,7-diamino-thiazol-
o[5,4-d]pyrimidines as potent TRPV1 antagonists. Initial studies
8

46
A. D. Lebsack et al. / Bioorg. Med. Chem. Lett. 19 (2009) 40–46
18. The authors have deposited coordinates for this compound with the Cambridge
Crystallographic Data Centre. The coordinates of 3 are CCDC 702151.
19. Liebeskind, L. S.; Srogl, J. Org. Lett. 2002, 4, 979.
22. When tested at 1 or 10
panel of 60 targets.
23. Under anesthesia, 100
l
M, compound 3 afforded no significant hits in a Cerep
lL of 1% carrageenan (Sigma) in saline was injected
20. Binnun, E.; Johnson, S. G.; Connolly, P. J.; Middleton, S. A.; Moreno-Mazza, S. J.;
Lin, R.; Pandey, N. B.; Wetter, S. K. PCT Int. Appl. WO 2007019191, 2007.
21. The solubility assay was conducted in a 96-well format using DMSO stock
solutions (10 mM of compound). DMSO was evaporated and residual solids
were re-suspended in fasted-state simulated intestinal fluid (SIF; pH 6.8) or
simulated gastric fluid (SGF; pH 1.2) for 3 days. The resulting mixtures were
filtered and analyzed by HPLC against external standards.
subcutaneously into the plantar surface of the hind paw. For methods and
references on the thermal hyperalgesia testing see: Swanson, D. M.; Dubin, A.
E.; Shah, C.; Nasser, N.; Chang, L.; Dax, S. L.; Jetter, M.; Breitenbucher, J. G.; Liu,
C.; Mazur, C.; Lord, B.; Gonzales, L.; Hoey, K.; Rizzolio, M.; Bogenstaetter, M.;
Codd, E. E.; Lee, D. H.; Zhang, S.; Chaplan, S. R.; Carruthers, N. I. J. Med. Chem.
2005, 48, 1857.