Design and synthesis of disubstituted thiophene and thiazole based inhibitors of JNK

Article abstract of DOI:10.1016/j.bmcl.2010.10.066

From high throughput screening, we discovered compound 1, the prototype for a series of disubstituted thiophene inhibitors of JNK which is selective towards closely related MAP kinases p38 and Erk2. Herein we describe the evolution of these compounds to a novel class of thiophene and thiazole JNK inhibitors that retain favorable solubility, permeability, and P-gp properties for development as CNS agents for treatment of neurodegeneration. Compound 61 demonstrated JNK3 IC₅₀ = 77 nM and retained the excellent broad kinase selectivity observed for the series.

Full text of DOI:10.1016/j.bmcl.2010.10.066

Bioorganic & Medicinal Chemistry Letters 20 (2010) 7303–7307
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and synthesis of disubstituted thiophene and thiazole based
inhibitors of JNK
Roy K. Hom a, , Simeon Bowers a, Jennifer M. Sealy a, Anh P. Truong a, Gary D. Probst a, Martin L. Neitzel a,
⇑
R. Jeffrey Neitz a, Larry Fang a, Louis Brogley a, Jing Wu a, Andrei W. Konradi a, Hing L. Sham a, Gergely Tóth b,
Hu Pan b, Nanhua Yao b, Dean R. Artis b, Kevin Quinn c, John-Michael Sauer c, Kyle Powell d, Zhao Ren d,
Frédérique Bard d, Ted A. Yednock a,b,c,d, Irene Griswold-Prenner d
a Department of Chemical Sciences, Elan Pharmaceuticals, 180 Oyster Point Boulevard, South San Francisco, CA 94080, United States
b Department of Molecular Design, Elan Pharmaceuticals, 180 Oyster Point Boulevard, South San Francisco, CA 94080, United States
c Department of Lead Finding, Drug Disposition, and Safety Evaluation, Elan Pharmaceuticals, 800 Gateway Boulevard, South San Francisco, CA 94080, United States
d Department of Biology, Elan Pharmaceuticals, 1000 Gateway Boulevard, South San Francisco, CA 94080, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
From high throughput screening, we discovered compound 1, the prototype for a series of disubstituted
thiophene inhibitors of JNK which is selective towards closely related MAP kinases p38 and Erk2. Herein
we describe the evolution of these compounds to a novel class of thiophene and thiazole JNK inhibitors
that retain favorable solubility, permeability, and P-gp properties for development as CNS agents for
treatment of neurodegeneration. Compound 61 demonstrated JNK3 IC50 = 77 nM and retained the excel-
lent broad kinase selectivity observed for the series.
Received 21 September 2010
Revised 11 October 2010
Accepted 14 October 2010
Available online 21 October 2010
Keywords:
JNK3
Ó 2010 Elsevier Ltd. All rights reserved.
Kinase
Thiophene
c-Jun N-terminal kinase
The c-Jun N-terminal kinases (JNKs) are serine/threonine pro-
tein kinases within the mitogen-activated protein (MAP) kinase
family which are upregulated in response to external and internal
stressors.1 There are 10 splice variants of human JNK which are en-
coded by three genes: Jnk1, Jnk2, and Jnk3. The isoforms JNK1 and
JNK2 are widely expressed in the body, but JNK3 is expressed pri-
marily in the brain.2
IC50 = 2.2 lM with good solubility, excellent passive permeability,
and a lack of recognition by permeability glycoprotein (P-gp)
in vitro, which are all critical characteristics for a series of com-
pounds to be developed for a CNS target (Fig. 1). This series was po-
tent mainly on the isoforms JNK1 and JNK3, with 10-fold selectivity
over JNK2. Compound 1 exhibited remarkable selectivity over clo-
sely related kinases p38a and Erk2.
JNK3 has been further implicated as a key player in neurode-
generative disorders, including Alzheimer’s3 and Parkinson’s4
disease. To avoid potential toxic effects from broad kinase inhibi-
tion, it was critical to generate selective inhibitors toward the JNKs.
In particular, we required selectivity between the JNKs and the
most closely related MAP kinases p38 and Erk2.5,6 With this in
mind, our goal was the development of brain penetrant JNK inhib-
itors which were selective over these other kinases. As yet, it is un-
clear whether systemic inhibition of any or all of the isoforms
JNK1, JNK2, or JNK3 results in clinically observed adverse effects.7
Thus, we did not focus on isoform selective inhibitors.
The crystal structure of JNK3 in complex with compound 1 in
the presence of a peptidic portion of JNK-interacting protein
(JIP1) was determined to 2.2 Å resolution as shown in Figure 2. This
was performed in analogy to work reported with JNK1.9 Compound
1 was found to be a Type I ATP-competitive kinase inhibitor where
Compound 1
μ
M
JNK3 IC50 = 2.2
JNK2 IC50 = 23
JNK1 IC50 = 1.6
M
O
μ
S
μ
M
μ
M
N
H
S
We opted to pursue JNK inhibitors from a high throughput
screening campaign of the Elan small molecule compound library.
α
p38 IC50 > 50
OMe
O
μ
Erk2 IC50 > 50
M
μ
HBSS solubility = 72
M
Compound
1 emerged as a promising hit, exhibiting JNK3
MDR MDCK A>B = 144 nm/sec
P-gp efflux = 0.93
⇑
Corresponding author. Tel.: +1 650 877 0900; fax: +1 650 794 2458.
E-mail address: roy.hom@elan.com (R.K. Hom).
Figure 1. In vitro characteristics of screening hit compound 1.8,9
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.10.066

7304
R. K. Hom et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7303–7307
MeO
O
S
S
S
a
b, c
N
H
O2N
O2N
O
Cl
O
N
N
8
7
6
Scheme 2. Reagents and conditions: (a) 10 mol % Pd(PPh3)4, 2-tributylstannylox-
azole, DMF, 100 °C; (b) 10% Pd/C, 40 psi H2, EtOAc, 50 °C; (c) 4-methoxyphenyl-
acetic acid, POCl3, pyridine, 0 °C.
Table 1
JNK1, JNK2, and JNK3 biochemical IC50 for thiophene ester analogs13
O
S
R
N
H
OMe
O
Compds
R
JNK3
IC50
JNK2
IC50
JNK1
IC50
a
a
a
(
l
M)
2.2
2.7
(
l
M)
(lM)
Figure 2. Binding mode of the thiophenes: X-ray co-crystal structure of compound
1 with JNK3 in the presence of JIP1 peptide at 2.2 Å resolution.9 The pdb code is
3OXI.
1
9
2-Thienyl
Ph
4-Pyridyl
4-OMePh
4-(4-Pyridyl)–Ph
4-(Imidazolyl)-Ph
4-CF3–Ph
2-Naphthyl
1-Naphthyl
2-(1,2,3-Triazol-4-yl)-
phenyl
21.4
22.8
>50
14.1
>50
>50
>50
>50
14.6
35.0
1.8
0.7
6.6
0.8
0.46
1.0
2.0
0.6
0.40
7.0
10
11
12
13
14
15
3
17.2
1.0
0.69
1.6
6.9
3.9
1.5
the critical hydrogen bond formed between the inhibitor carbonyl
and the NH of Met-149. The thiophene ring of the inhibitor was
found to be within hydrophobic contact of the gatekeeper
Met-146 side chain. In addition, the inhibitor ester substituent
formed an internal hydrogen bond with the amide NH, thus creat-
ing planarity on the right half of the molecule. The left side 2-thi-
enylacetyl portion formed a 90° turn within the inhibitor, pointing
the arylacetyl group toward solvent.
Analogs of compound 1 were synthesized according to Scheme 1.
The ester inhibitor 3 was readily synthesized from amide coupling10
of the commercial amino ester 2. The methyl ester was saponified
under basic conditions, and then converted to the primary amide
4. The triazole analog 5 was in turn provided from treatment of 4
with dimethylformamide dimethyl acetal (DMFDMA) to give the
desired triazole inhibitor.11,12
Alternatively, JNK inhibitors were synthesized from the com-
mercial 2-chloro-3-nitrothiophene (6), as exemplified with com-
pound 8 (Scheme 2). Displacement or metal-mediated coupling
at the site of the activated halogen, followed by nitro reduction,
and then amide coupling under POCl3 conditions provided a num-
ber of the analogs.
The initial SAR was generated on variations of the N-arylacetyl
portion of the molecule (Table 1).13 Not surprisingly, phenyl (9)
was a viable isosteric replacement of the 2-thienyl group,14 but
the more polar 4-pyridyl analog (10) lost eightfold in potency. Pen-
dant para substituted hydrophilic phenylacetyl groups (11–13),
16
12.6
17
18
5-Isoquinolyl
5-Quinolyl
0.61
0.61
3.1
3.0
0.27
0.30
a
See Ref. 9.
but not a bulkier hydrophobic group (14), were tolerated or gave
slightly improved activity in JNK1 and JNK3. Larger bicyclic
naphthylacetyls (3 and 15) were also tolerated. Attempts to gain
another hydrogen bond interaction with the hinge through
2-substituted phenacetyls (e.g., 16) were not successful. The polar
N-termini 5-isoquinolylacetyl (17) and 5-quinolylacetyl (18) gave
a moderate fourfold increase in JNK3 activity and sevenfold
increase in JNK1 activity compared with compound 1. The only
modest improvements in activity were consistent with the X-ray
co-crystal structure, which indicated that this portion of the mole-
cule pointed toward solvent, and likely did not form significant
interactions with the enzyme.
Since the ester function did not form any hydrogen bonds with
the enzyme, we postulated that it could be replaced (Table 2). We
quickly discovered that any functionality of similar size to the
methyl ester, but without the ability to hydrogen bond with the
amide NH, such as the methyl alcohol 20 and the isopropenyl 22,
led to compounds which were not active, indicating that the pla-
narity was necessary for JNK inhibition in this series. Regardless,
even simple replacements which preserved the hydrogen bond
acceptor present in the ester, such as the ketone (19) or oxime
(21), also lost 6-fold and 11-fold activity versus JNK3, respectively.
Simple primary and secondary amides (25–27) were tolerated, but
these were also slightly less active than the ester. The tertiary
amide led to complete loss in activity. Larger esters, such as the
ethyl (23) and isopropyl (24) esters, led to slight increases in
JNK3 inhibition. With the exception of the esters and amides, we
were unable to replace the methyl ester with other acyclic
hydrogen bond accepting groups, and the rationale for that re-
mained unclear.
O
a
b, c
S
S
H2N
N
H
CO2Me
CO2Me
3
2
O
d
S
O
S
N
H
N
H
N
NH
CONH2
5
N
4
Scheme 1. Reagents and conditions: (a) 1-naphthylacetic acid, POCl3, pyridine,
0 °C; (b) NaOH, 1:1 THF/H2O, 50 °C, 20 h, 75%; (c) SOCl2, 60 °C, 30 min, then concd
NH4OH, CH3CN, 54%; (d) (OMe)2CHNMe2, 110 °C, 30 min, then H2NNH2–H2O, AcOH,
90 °C.
Concurrent with the study performed in Table 1, we also exam-
ined whether the central thiophene could be replaced in these
inhibitors. As compound 3 was discovered early, we began with

R. K. Hom et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7303–7307
7305
Table 2
variations on that analog. However, a number of heterocyclic
replacements failed to give compounds with any measureable inhi-
bition toward JNK3 (Table 3). This is likely due to the detrimental
energetics between the heterocyclic rings which presented a
hydrogen bond acceptor toward the C@O function of Glu-147,
including the five-membered ring analogs isothiazole, oxadiazole,
thiadiazole, and thiazole (28–31) as well as the six-membered ring
pyridyl and pyrazine analogs (32–34).
JNK1, JNK2, and JNK3 biochemical IC50 for thiophenes containing acyclic ester
variations
O
S
N
H
X
Compds
X
JNK3
IC50
JNK2
IC50
JNK1
IC50
a
a
a
(
l
M)
(lM)
(
lM)
Nevertheless, we discovered a few rings other than thiophene
which were tolerated by JNK3 (Table 3). Phenyl (35) was an accept-
able isostere, and yielded a comparable JNK3 inhibitor to the thio-
phene. Other compounds which yielded somewhat weaker JNK3
inhibitors included NH pyrazole and 3,4-disubstituted pyridyl ana-
logs (37 and 38). The 3,4-disubstituted thiophene (36) indicated
that location of the sulfur on the ring was unimportant for mainte-
nance of the hydrophobic interaction with the enzyme.
9
19
20
21
22
23
24
25
26
27
CO2Me
COMe
CH2OH
C(@N–OH)Me
C(@CH2)Me
CO2Et
CO2i-Pr
CONH2
2.7
17
>50
29
>50
0.7
1.4
4.4
3.8
>50
22.8
>50
na
na
na
46.6
47.7
>50
>50
>50
0.7
6.3
na
na
na
0.3
0.5
2.6
2.3
na
CONH(CH2)2OMe
CONMe(CH2)2OMe
Due to the perceived metabolic liability of the methyl ester, we
were eager to discover replacements which would be more stable
and perhaps more potent against JNK3. Despite our initial disap-
pointment with acyclic replacements in Table 2, we were able to re-
place the ester function with heteroaryl groups that could form the
internal hydrogen bond with the secondary amide of the inhibitor.
We discovered that the C-linked triazole (5) and methyltriazole
na = data not available.
a
See Ref. 9.
Table 3
Effect of the central ring variation on JNK1, JNK2, and JNK3 biochemical IC50
O
Table 4
Ar
Biochemical IC50 and in vitro oxidative stability for thiophene inhibitors
N
H
OMe
O
O
S
N
Compds
Ar
JNK3
IC50
JNK2
IC50
JNK1
IC50
H
a
a
a
Ar
(l
M)
(lM)
(
lM)
S
3
1.5
14.6
>50
0.4
Compds Ar
JNK3
IC50
JNK2
M) IC50
JNK1
M) IC50
Stabilityb,c
M) (m, h)
a
a
a
(l
(l
(l
S N
N
28
>50
>50
39
40
5.3
9.0
27.1
>50
3.2
1, 7
O
N
N
N O
N
29
30b
31
>50
>50
>50
>50
>50
>50
>50
>50
>50
N
26.9
0, 17
S N
N
N
41
3.6
>50
3.2
0, 5
N
N
S
N
H
N
N
32
33
>50
>50
>50
>50
>50
>50
5
0.92
2.1
3.0
0.53
1.8
1, 11
0, 4
N
H
N
O
42
16.3
N
N
N
N
34
35
>50
1.7
>50
13.1
9.9
na
N
43
42.5
>50
24.1
0, 0
N
Me
2.1
0.89
2.1
N
44
45
46
0.55
2.4
>50
1.5
0.25
1, 1
1, 1
1, 2
S
N
N
36c
37
0.90
5.4
H
N
NH
N
31.5
30.3
N
26.7
Me
N
N
N
9.4
3.4
38
6.5
23.6
6.4
N
H
a
na = data not available.
See Ref. 9.
a
b
See Ref. 9.
Ethyl ester.
In vitro metabolic stability measured as the percent remaining after 30 min
b
incubation with mouse or human microsomes.
c
c
Phenylacetamide N-terminus.
See Ref. 8 for protocols.

7306
R. K. Hom et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7303–7307
(44) analogs were the most active (Table 4). Many other heteroaryl
groups, including oxadiazoles (39 and 42) and pyrazoles (41 and
46), afforded inhibitors which were comparable or resulting in only
slightly attenuated activity compared with the methyl ester 3. The
imidazole (40), and N-Me triazole (45) or pyrazole (43) analogs led
to significant losses of activity. Despite removal of the ester, all of
these compounds exhibited poor metabolic stability, perhaps due
to the oxidation of the naphthyl function.
It became clear that, while a 2-substituted heteroatom capable
of acting as hydrogen bond acceptor, such as N or O, was necessary
for JNK inhibition in this series, it was not sufficient for activity in a
number of analogs (Table 5). Interestingly, the N-linked 1,2,4-tria-
zole (47) lost activity, while the symmetrical 1,2,3-triazole (48)
retained JNK3 inhibition. The basic imidazoles (52 and 53) and
non-aryl carbamate (49) lost significant activity, while triazoles
50 and 51, oxazole 8, and simple pyrazine 54 retained potency
against JNK1 and JNK3.
N-quinolylacetamide triazole 55 showed submicromolar potency
against JNK1 and JNK3, retained excellent solubility and selectivity
against Erk2, and moderate to good oxidative stability, but began
to exhibit P-gp efflux in vitro. The N-isoquinolylacetamide thiazole
56 showed similarly good potency and selectivity, as well as
solubility, but lower in vitro metabolic stability with enhanced
P-gp properties compared with 55.
We also examined the 4,5-disubstituted thiazoles, which also
were suitable replacements for the 2,3-disubstituted thiophenes
(Table 6). The NH triazoles (58 and 59), pyrazine (60), and bisthia-
zole (61) analogs provided greater inhibition versus JNK3 com-
pared with the ester (57). None of the derivatives 57–60 showed
acceptable stability in murine microsomes. Significantly, the bis-
thiazole analog 61 afforded the most potent JNK3 inhibitor thus
far with IC50 <100 nM and with good in vitro stability in the
Compound 55
Overall, the metabolic stability remained poor, with the excep-
tion of aminotriazole 51 and weakly active carbamate 49. How-
ever, the passive permeability of these compounds remained at a
level compatible with CNS penetration.
Combination of the favorable heterocyclic replacements and
N-termini in the thiophene series also led to more potent analogs
which retained favorable permeability properties ( Fig. 3). The
μ
μ
μ
JNK3 IC50 = 0.56
JNK2 IC50 = 0.71
JNK1 IC50 = 0.27
M
M
M
O
S
N
H
μ
Erk2 IC50 > 50
M
NH
N
μ
HBSS solubility > 100
M
N
Stability (m, h) = 23, 67
MDR MDCK A>B = 60 nm/sec
P-gp efflux = 4.2
N
Table 5
Compound 56
Biochemical IC50
, in vitro oxidative stability, and passive permeability data for
μ
μ
μ
M
M
JNK3 IC50 = 0.16
JNK2 IC50 = 0.94
JNK1 IC50 = 0.14
p38α IC50 > 50
M
M
M
thiophene inhibitors13
O
MeO
S
O
N
H
S
μ
μ
N
H
Erk2 IC50 > 50
N
Ar
μ
M
HBSS solubility = 90
S
N
Compds Ar
JNK3
JNK2
JNK1
Stabilityb,c Permc
Stability (m, h) = 1, 4
MDR MDCK A>B = 261 nm/sec
P-gp efflux = 1.1
a
a
a
IC50
M)
IC50
M)
IC50
M)
(m, h)
(nm/s)
(l
(l
(l
Figure 3. In vitro data for thiophenes 55 and 56.
N
N
N
47
48
45.6
1.5
>50
5.0
43.6
0.83
na
na
N
N
Table 6
1, 0
150
N
JNK biochemical IC50, oxidative stability, and passive permeability of thiazole
analogs13
N
49
50
37.0
5.9
>50
>50
3.4
26, 83
8, 71
260
na
O
O
S
N
O
N
H
N
R
15.0
N
N
CF3
NH2
N
N
H
Compds
R
JNK3
JNK2
IC50
JNK1
IC50
(lM)
Stabilityb,c
(m, h)
P-gp
N
IC50
M)
effluxc
a
a
a
51
52
1.6
8.7
0.79
23.9
44, 87
1, 49
243
70
(
l
(l
M)
N
H
57
58
CO2Me
1.6
5.3
1.7
11, 41
15, 14
0.8
5.0
N
N
31.0
>50
N
Me
0.82
2.6
0.71
N
N
H
N
Me
8
0.80
11.9
>50
0.22
23.4
9, 14
0, 17
na
N
59
60
1.6
7.6
2.0
1.4
0, 100
3, 2
1.1
O
N
N
N
53
27.0
1.5
338
Me
N
N
H
0.47
0.41
0.73
N
N
54
5.7
1.1
0, 0
222
N
61
0.077
0.47
0.048
50, 79
1.9
N
S
na = data not available.
a
a
See Ref. 9.
See Ref. 9.
b
b
In vitro metabolic stability measured as the percent remaining after 30 min
incubation with mouse or human microsomes.
In vitro metabolic stability measured as the percent remaining after 30 min
incubation with mouse or human microsomes.
c
c
See Ref. 8 for protocols.
See Ref. 8 for protocols.

R. K. Hom et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7303–7307
7307
B.; Moore, G.; Davis, R. J.; Karin, M. Genes Dev. 1994, 8, 2996; (c) Mohit, A. A.;
Martin, J. H.; Miller, C. Neuron 1995, 14, 67.
presence of murine and human liver microsomes. Furthermore, it
maintained the remarkable selectivity against a panel of 36 non-
3. (a) Zhu, X.; Raina, A. K.; Rottkamp, C. A.; Aliev, G.; Perry, G.; Boux, H.; Smith, M.
A. J. Neurochem. 2001, 76, 435; (b) Morishima, Y.; Gotoh, Y.; Zieg, Y.; Barrett, T.;
Takano, H.; Flavell, R.; Davis, R. J.; Shirasaki, Y.; Grenberg, M. E. J. Neurosci. 2001,
21, 7551; (c) Thakur, A.; Wang, X.; Siedlak, S. L.; Perry, G.; Smith, M. A.; Zhu, X. J.
Neurosci. Res. 2007, 85, 1668.
4. (a) Saporito, M. S.; Thomas, B. A.; Scott, R. W. J. Neurochem. 2000, 75, 1200; (b)
Mathiason, J. R.; McKenna, B. A. W.; Saporito, M. S.; Ghadge, G. D.; Roos, R. P.;
Holskin, B. P.; Wu, Z.-L.; Truszko, S. P.; Connors, T. C.; Maroney, A. C.; Thomas, B.
A.; Thomas, J. C.; Bozyczko-Coyne, D. Brain Res. 2004, 1003, 86; (c) Resnick, L.;
Fennell, M. Drug Discovery Today 2004, 9, 932; (d) Hunot, S.; Vila, M.; Teismann,
P.; Davis, R. J.; Hirsch, E. C.; Przedborski, S.; Rakic, P.; Flavell, R. A. Proc. Natl.
Acad. Sci. U.S.A. 2004, 101, 665; (e) Wang, W.; Shi, L.; Xie, Y.; Ma, C.; Li, W.; Su,
X.; Huang, S.; Chen, R.; Zhu, Z.; Mao, Z.; Han, Y.; Li, M. Neurosci. Res. 2004, 48,
195.
5. Recent example of potent, selective JNK inhibitors: (a) Kamenecka, T.; Jiang, R.;
Song, X.; Duckett, D.; Chen, W.; Ling, Y. Y.; Habel, J.; Laughlin, J. D.; Chambers,
J.; Figuera-Losada, M.; Cameron, M. D.; Lin, L.; Ruiz, C. H.; Lo Grasso, P. V. J. Med.
Chem. 2010, 53, 419; Reviews of JNK inhibitors: (b) Siddiqui, M. A.; Reddy, P. A.
J. Med. Chem. 2010, 53, 3005; (c) Lo Grasso, P.; Kamenecka, T. Mini-Rev. Med.
Chem. 2008, 8, 755.
6. Xie, X.; Gu, Y.; Fox, T.; Coll, J. T.; Fleming, M. A.; Markland, W.; Caron, P. R.;
Wilson, K. P.; Su, M. S.-S. Structure 1998, 6, 983.
7. (a) Han, S.-Y. Toxicol. Res. 2008, 24, 93; (b) Bonny, C.; Borsello, T.; Zine, A. Rev.
Neurosci. 2005, 16, 57.
8. For conditions of in vitro pharmacokinetic assays, see: Truong, A. P.; Aubele, D.
A.; Probst, G. D.; Neitzel, M. L.; Semko, C. M.; Bowers, S.; Dressen, D.; Hom, R. K.;
Konradi, A. W.; Sham, H. L.; Garofalo, A. W.; Keim, P. S.; Wu, J.; Dappen, M. S.;
Wong, K.; Goldbach, E.; Quinn, K. P.; Sauer, J.-M.; Brigham, E. F.; Wallace, W.;
Nguyen, L.; Hemphill, S. S.; Bova, M. P.; Basi, G. Bioorg. Med. Chem. Lett. 2009, 19,
4920.
JNK kinases,13,15 including both p38
a and Erk2 (IC50 >50 lM),
and showed only a low degree of P-gp efflux in MDR-MDCK cells.
In conclusion, we discovered a series of JNK1/3 selective thio-
phene inhibitors through high throughput screening. The ATP-
competitive thiophene and thiazole inhibitors were improved by
replacement of the ester with heterocyclic rings, which retained
the inhibitor’s internal hydrogen bonding to maintain the planarity
of much of the molecule. The best bis-thiazole inhibitor 61 showed
IC50 <100 nM for both JNK1 and JNK3, with good in vitro stability
against liver microsomes. Further investigation of these inhibitors
will be reported in due course.
Acknowledgments
We are grateful to DeCode Biostructures, Bainbridge Island, WA,
USA for the solution of the co-crystal structure of JNK3, JIP1, and
inhibitor 1. We thank Carla Ramos, Wes Zmolek, Ann Qin, Colin
Lorentzen, Karina Wong, Robert Galemmo, Mark Dreyer, and Mi-
chael Bova for their contributions to this work.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.10.066.
9. Details on the TR-FRET enzyme assays and crystallization conditions can be
found in Supplementary data.
10. Hill, G. B.; Mortlock, A. A. Synthesis 2007, 1697.
11. Gabriel, T.; Luk, K.-C. T.; McIntosh, J. WO 2008055842.
12. Sham, H. L.; Konradi, A. W.; Hom, R. K.; Probst, G. D.; Bowers, S.; Truong, A.;
Neitz, R. J.; Sealy, J.; Toth, G. WO 2010091310.
References and notes
13. Erk2 and p38
showed IC50 >50
Compound 58: Erk2 IC50 = 24
a
IC50 were measured for the following compounds, and all
for both enzymes: 17, 37, 50–54, 56–57, 59–61.
M and p38 IC50 >50 M.
1. For recent reviews on the JNK signaling pathway, see: (a) Waetzig, V.;
Herdegen, T. Trends Pharmacol. Sci. 2005, 26, 455; (b) Bogoyevitch, M. A.; Ngoei,
K. R. W.; Zhao, T. T.; Yeap, Y. Y. C.; Ng, D. C. H. Biochim. Biophys. Acta 2010, 1804,
463; (c) Bogoyevitch, M. A. BioEssays 2006, 28, 923.
l
M
l
a
l
14. Lima, L. M.; Barreiro, E. J. Curr. Med. Chem. 2005, 12, 23.
15. Full kinase panel data performed at Invitrogen are included in Supplementary
data.
2. (a) Derijard, B.; Hibi, M.; Wu, I. H.; Barrett, T.; Su, B.; Deng, T.; Karin, M.; Davis,
R. J. Cell 1994, 76, 1025; (b) Kallunki, T.; Su, B.; Tsigelny, I.; Sluss, H. K.; Derijard,