Discovery of a new class of 4-anilinopyrimidines as potent c-Jun N-terminal kinase inhibitors: Synthesis and SAR studies

Article abstract of DOI:10.1016/j.bmcl.2006.10.093

A new series of 4-anilinopyrimidines has been synthesized and evaluated as JNK1 inhibitors. SAR studies led to the discovery of potent JNK1 inhibitors with good enzymatic activity as well as cellular potency represented by compound 2b. Kinase selectivity

Full text of DOI:10.1016/j.bmcl.2006.10.093

Bioorganic & Medicinal Chemistry Letters 17 (2007) 668–672
Discovery of a new class of 4-anilinopyrimidines as potent c-Jun
N-terminal kinase inhibitors: Synthesis and SAR studies
Mei Liu,* Sanyi Wang, Jill E. Clampit, Rebecca J. Gum,
Deanna L. Haasch, Cristina M. Rondinone, James M. Trevillyan,
Cele Abad-Zapatero, Elizabeth H. Fry, Hing L. Sham and Gang Liu
Metabolic Disease Research, Global Pharmaceutical Research and Development, Abbott Laboratories,
Abbott Park, IL 60064-6098, USA
Received 29 September 2006; revised 27 October 2006; accepted 30 October 2006
Available online 2 November 2006
Abstract—A new series of 4-anilinopyrimidines has been synthesized and evaluated as JNK1 inhibitors. SAR studies led to the
discovery of potent JNK1 inhibitors with good enzymatic activity as well as cellular potency represented by compound 2b.
Kinase selectivity profile and the crystal structure of 2b are also described.
Ó 2006 Elsevier Ltd. All rights reserved.
The c-Jun N-terminal kinases (JNKs) are members of
the mitogen-activated protein kinase (MAPK) family
of signaling proteins, and they can be activated in re-
sponse to various stimuli such as environmental stress,
cytokines and fatty acids.1 Three JNK isoforms
(JNK1, 2, and 3) have been identified, with JNK1 and
JNK2 widely expressed in tissues, whereas JNK3 is
selectively expressed in the brain, heart, and testis.2
There is growing evidence suggesting that activation of
the JNK activity is involved in a number of human dis-
ease settings.3 Therefore, considerable efforts have been
directed toward the identification of JNK inhibitors
suitable for clinical development.4
elevated in adipocytes of type 2 diabetic patients.6 These
results imply that inhibitors of JNK1 can potentially in-
crease insulin sensitivity, and can be useful as therapeu-
tics for the treatment of type 2 diabetes.3
Recently we reported the discovery of aminopyridine7 as
well as 9-hydroxypyrazoloquinolinones8 as potent and
selective JNK inhibitors. In the course of pursuing addi-
tional chemotypes as JNK inhibitors, pyrimidine 1 was
identified from a high throughput screening (HTS)
(Fig. 1). This compound had an enzymatic IC50 of
1.9 lM against JNK1, and it was also selective against
two other closely related MAP kinases: p38 and ERK2
(IC50s >100 lM). Recent reports from Merck claimed
the use of 4-(4-pyrimidinyl)-5-phenylimidazole deriva-
tives as potent JNK3 inhibitors for targeting neurologi-
cal disorders,9 however, their compounds are also active
against p38 in contrast to the selectivity observed with
our hit compound 1. Encouraged by the preliminary
selectivity profile of compound 1 in addition to its
4-aminopyrimidine structural motif which is amenable
We were particularly interested in JNK1, an enzyme
that was believed to play a key role in linking obesity
and insulin resistance.5 JNK1 disrupts the insulin signal-
ing cascade via phosphorylation of the insulin receptor
substrate (IRS-1) at serine307, which leads to the degra-
dation of IRS-1. JNK1À/À mice show lower fasting
blood glucose levels as well as lower insulin concentra-
tions than their wild-type littermates, thus are resistant
to diet-induced obesity.5b In addition, JNK1 activity is
N
N
HN
N
N
O
N
N
H
H
Keywords: JNK inhibitors; Diabetes; Obesity; Kinase.
*
O
NH2
HO
O
Corresponding author. Tel.: +1 847 935 7106; fax: +1 847 938
1674; e-mail: mei.liu@abbott.com
1
2b
IC50 = 9 nM
HO
IC50 = 1.9 μM
Present address: Hoffman-La Roche Inc., Metabolic Diseases, 340
Kingsland Street, Nutley, NJ 07110, USA.
Figure 1. HTL optimization of 1 to potent JNK inhibitor 2b.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.10.093

M. Liu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 668–672
669
to rapid modification via parallel synthesis, we initiated
a
O
hit to lead (HTL) chemistry of compound 1, leading to
the discovery of a series of potent JNK1 inhibitors
represented by compound 2b.
a
N
N
+
OH
Cl
N
N
H
Cl
N
Cl
NH2
3
9
O
OH
10
The versatile common intermediate 5 was prepared in
one step from the substitution reaction between 2,4-
dichloropyrimidine 3 and anthranilic acid 4 (Scheme
1).10 The reaction was regioselective at the 4-position,
and can be carried out easily in gram-scale with no puri-
fication required. Further reaction (SNAr) of 5 with an
alcohol or phenol in the presence of NaOH, or with a
nucleophilic amine provided most of the 2-pyrimidine
modification analogs 6–8. Compounds 2a and 2b were
obtained via TBTU-mediated amide coupling of ammo-
nium chloride with 8.
N
11a: R = H
HN
N
N
b
11b: R = OH
H
O
OH
11
R
b
X
O
N
X
a
N
+
OH
Cl
N
N
H
Cl
N
Cl
NH2
X
O
OH
12
4
13
X = F or Br
Modification of the pyrimidine core or the aromatic ring
of the 4-anilino portion was achieved via similar fashion
as described for the synthesis of 8. As outlined in
Scheme 2a, analogs 11a and 11b were synthesized in a
two-step sequence from 3 and 2-amino-3-naphthoic acid
(9). Similarly, substituting 3 with a halogenated pyrimi-
dine core (12) yielded intermediate 13, where the 2-chlo-
ro group may be replaced with an aniline to give
compounds 14a–d (Scheme 2b). In the case of 5-bro-
mo-pyrimidine (12, X = Br), DMSO was needed as an
additional solvent to facilitate the reaction.
N
14a: R = H, X = F
14b: R = OH, X = F
14c: R = H, X = Br
HN
N
N
H
b
14d: R = OH, X = Br
O
OH
14
R
Scheme 2. Reagents and conditions: (a) cat. concd HCl, H2O (plus
DMSO when X = Br in pyrimidine 12), 50°C (2 h), then rt (2d), 85%;
(b) anilines, EtOH, microwave 130 °C, 20–40 min, 55–80%.
During our initial SAR studies, a series of ethers (6)
bearing different alkyl or aromatic substitutions were
prepared and tested against JNK1. As summarized in
Table 1, the different straight chain alkyl groups showed
only modest variation in potency (1 and 6a–d), with
R1 = pentyl (6c) being optimal with IC50 around 1 lM.
Few analogs with branching at the b-carbon (6e, 6f)
were about 3-fold more active than compound 1. Phenol
ethers provided a 6-fold increase in potency, however,
further modification of the phenol ring was detrimental
to JNK1 activity. Any additional substitution only
resulted in loss of JNK1 activity.11
O
N
a
N
+
OH
NH2
Cl
N
N
H
Cl
N
Cl
O
OH
3
4
5
b
N
R1O
N
N
c
H
O
OH
d
Guided by an NMR-assisted JNK1 active site model
(which was later confirmed by X-ray crystallography),
we anticipated that the ether portion of the molecule
interacted with the hinge region of the enzyme, and
replacement of the ether linkage with a mono-substituted
amino group should result in an increase in potency with
the NH proton acting as a H-bond donor to the carbonyl
group within the active site of JNK1. As shown in Table
2, direct replacement of the ether oxygen to NH
maintained JNK activity (compare analogs 1 and 7a).
In addition, secondary amines (7a and 7c) were more
active than tertiary amines (7d and 7e). To our delight,
about a 40-fold improvement in potency was observed
when an aniline group was introduced at the 2-position
(8a). This confirmed our hypothesis that a hydrogen
bond donor was needed here to gain JNK1 activity.
6, R1 = Aryl or Alkyl
N
R2
N
N
R3
N
N
H
HN
N
N
H
O
OH
7
O
OH
R
8
e
N
N
HN
N
N
H
HN
N
N
H
O
NH2
O
NH2
2a
R = H
2b
HO
R = 3-OH
Scheme 1. Reagents and conditions: (a) cat. concd HCl, H2O, 50 °C
(1 h), then rt (overnight), 90–95%; (b) phenols or alcohols, NaOH,
DMF, microwave 180 °C, 20 min, 50–60%; (c) amines, EtOH, micro-
wave 170 °C, 20 min, 50–60%; (d) anilines, EtOH, microwave 110 °C,
20 min, 60–80%; (e) ammonium chloride, TBTU, DIEA, rt, 6 h,
80–90%.
Encouraged by the promising results above, we decided
to probe the effects of various aniline substitution on
JNK1 activity (Table 3). In general, meta- or
para-substituted anilines are more potent than the
corresponding ortho analogs. For example, of the

670
M. Liu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 668–672
Table 1. SAR summary of JNK inhibitors 1 and 6
Table 3. SAR summary of JNK inhibitors (8 and 15) with 2-anilino-
substitutions
N
R1
N
N
O
N
N
H
HN
N
N
H
HO
HN
R
N
N
HO
O
H
6
O
HO
O
8
15
Compound
R1
JNK1 IC50 (lM)
a
R
1
Bu
Et
Pr
1.9
a
JNK1 IC50 (nM)
Compound
R
6a
6b
6c
6d
6e
6f
6g
44.7
2.3
8a
8b
8c
8d
8e
8f
H
48
76
Pentyl
Hex
i-Bu
1.1
2-OH
3-OH
4-OH
2-F
>100
0.7
0.7
25
35
CH2-c-Hex
Ph
35
28
0.3
3-F
4-F
8g
8h
8i
29
a Values are averages of two or more independent experiments.
3-Me
157
93
3-F,4-Me
2-Me,3-OH
3-CF3
8j
278
391
186
33
Table 2. SAR summary of JNK inhibitors 1, 7 and 8a
8k
8l
4-CF3
4-NO2
4-Morpholine
N
N
R2
8m
8n
N
R3
N
N
HN
N
N
37
H
H
H
HO
O
HO
O
7
8a
N
N
15a
46
R2
R3
JNK1 IC50 (lM)
a
Compound
1
—
Bu
—
H
H
H
1.9
2.6
N
7a
7b
7c
7d
7e
8a
15b
15c
55
85
HN
Bn
c-Hex
–(CH2)6–
18.2
0.4
15.4
>100
0.048
N
Et
—
Et
—
a Values are averages of two or more independent experiments.
a Values are averages of two or more independent experiments.
hydroxyaniline analogs (8b–8d), the meta-isomer (8c)
was the most potent among the three, which provided
a 2-fold increase in JNK1 activity as compared to the
unsubstituted aniline (8a). Fluoroanilines (8e–8g) exhib-
ited similar potency regardless of the substitution posi-
tion of the fluoride. All the methyl or trifluoromethyl
compounds (8h–l) were less potent than the unsubstitut-
ed aniline (8a), with the para-isomer showing somewhat
better potency than the meta-isomers. An introduction
of a nitro (8m) or morpholine (8n) group at the para-po-
sition gained slightly in JNK1 activity as well. Heterocy-
cles were also tolerated as exemplified by compounds
15a–c, which showed similar enzymatic potencies with
respect to the reference compound (8a).
Table 4. SAR summary for JNK inhibitors 11 and 14
X
N
N
HN
N
N
HN
N
N
H
H
O
OH
O
OH
11
14
R
R
a
JNK1 IC50 (nM)
Compound
R
X
11a
11b
14a
14b
14c
14d
H
—
—
F
705
204
82
OH
H
OH
H
F
21
32
Br
Br
Next, we synthesized compounds 11 and 14, aiming to
identify additional interactions with the enzyme by tar-
geting the hydrophobic pocket of the protein backbone.
As summarized in Table 4, the extra aromatic ring in
analogs 11 resulted in a loss of potency (ꢀ10-fold).
While only a marginal increase in JNK1 activity was
seen with compounds 14, where the 5-hydrogen of the
pyrimidine ring was replaced with a halogen, the bromo
group could serve as a handle for further optimization at
this position.
OH
20
a Values are averages of two or more independent experiments.
the presence of the acid group (which is ionized at phys-
iological pH) in these inhibitors may reduce their capac-
ity to penetrate the cell membrane via passive diffusion.
In fact, hit compound 1 showed only 22.3 lM cellular
activity in our cell-based c-Jun phosphorylation assay.12
To address this issue, the acid group was converted to
the primary amide (2). We were pleased to see that not
only the JNK1 enzymatic activity was slightly improved
In order to reach the intracellular target, JNK1 inhibi-
tors need to be cell-permeable. We were concerned that

M. Liu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 668–672
671
O
NH
E109
S
HN
Lys55
O
intra-
O
N
molecular
HB
L110
O
NH
NH
O
M111
N
NH
O
D169
N
O
O
HN
HN
O
Ile32
O
D112
O
N
HN
O
A113
N114
Figure 2. X-ray structure of compound 2b.
Table 5. Kinase selectivity profile of JNK1 inhibitor 2ba
(2a: IC50 = 34 nM, 2b: IC50 = 9 nM), but also a 20- to
40-fold increase in cellular potency (2a: c-Jun
IC50 = 590 nM, 2b: c-Jun IC50 = 1220 nM) was observed
with these compounds.
Kinases
2b (lM)
Kinases
2a (lM)
JNK1
P38
0.009
>50
25
PLK
CK2
0.73
0.73
6.8
ERK2
AKT1
CHK1
PAK4
MEK
CDK2
MK2
COT2
An X-ray structure of compound 2b bound into the
ATP site of JNK1 has been obtained (Fig. 2).13 In this
structure, two key hydrogen bonds with the pyrimidine
N1 acting as a H-bond acceptor and the 2-anilino group
serving as a H-bond donor provided the crucial hinge
interactions needed for JNK1 activity. This is consistent
with the complete loss in potency when the 2-amino
group is disubstituted (7e, Table 2). An intramolecular
H-bond between the carbonyl group and 4-amino group
was also evident, which oriented the pyrimidine ring as
well as the amide group toward the hydrophobic pocket.
Therefore, the extra ring in analogs 11 was actually
pointing away from the pocket which we were trying
to reach, as a result, about a 10-fold loss in potency
was observed. On the other hand, the halogen substitu-
tions of the pyrimidine ring were pointing in the correct
direction, which yielded a slight improvement in JNK1
activity. Further extension at this position could poten-
tially identify additional interactions between the ring
and the hydrophobic pocket, and thus more potent
and possibly selective JNK1 inhibitors may be obtained.
15
2.7
0.82
5.5
>50
>50
a ATP concentrations used were: 1.5 lM for CK2, 5 lM for JNK1,
p38, ERK2, PAK4, PLK1, and CHK1, 10 lM for AKT1, MK2, and
100 lM for CDK2, MEK, and COT2.
achieve highly selective inhibitors for this series of com-
pounds. However, there have been examples of selective
kinase inhibitors containing the 2-aminopyrimidine
structural motifs reported in the literature.9a,14 Utilizing
structure-based design, selectivity was achieved through
the modification of the molecule to target the hydropho-
bic pocket of the enzyme, which is known to contain res-
idues that vary among kinases. Our co-crystal structure
of the most potent analog 2b with JNK1 should guide
our future structure-based inhibitor design for obtaining
kinase selectivity and perhaps JNK isoform selectivity as
well. Additionally, evaluation of ADME/PK properties
of these inhibitors and further optimization of them
could provide valuable tool compounds for investigating
the effects of JNK inhibitors in vivo.
The selectivity data of 2b against a panel of kinases are
listed in Table 5. This compound has some degree of
selectivity over quite a few kinases with the exceptions
of CHK1, CK2, and PLK. Although compound 2b
shows submicromolar potency against these three ki-
nases, it still appears to exhibit at least 70- to 90-fold
greater potency against JNK1 than CHK1, CK2, and
PLK.
References and notes
1. Wellen, K. E.; Hotamisligil, G. S. J. Clin. Invest. 2005,
115, 1111.
2. Davis, R. J. Cell 2000, 103, 239.
3. (a) Manning, A. M.; Davis, R. J. Nat. Rev. Drug Discovery
2003, 2, 554; (b) Liu, G.; Rondinone, C. M. Curr. Opin.
Invest. Drugs 2005, 6, 979; (c) Sharma, P.; Ghoshal, N.
J. Chem. Inf. Model. 2006, 4, 1763, and references cited
therein.
In summary, we have identified a series of anilinopyrim-
idines as potent JNK inhibitors with good cellular
activity. Since 2-aminopyrimidine is a common pharma-
cophore for kinase inhibitors, it can be difficult to

672
M. Liu et al. / Bioorg. Med. Chem. Lett. 17 (2007) 668–672
4. (a) Bain, J.; McLauchlan, H.; Elliott, M.; Cohen, P.
49, 3563; (b) Zhao, H.; Serby, M. D.; Xin, Z.;
Szczepankiewicz, B. G.; Liu, M.; Kosogof, C.; Liu, B.;
Nelson, L. T. J.; Johnson, E. F.; Wang, S.; Pederson, T.;
Gum, R. J.; Clampit, J. E.; Haasch, D. L.; Abad-
Zapatero, C.; Fry, E. H.; Rondinone, C. M.; Trevillyan,
J. M.; Sham, H. L.; Liu, G. J. Med. Chem. 2006, 49, 4455;
(c) Liu, G.; Zhao, H.; Liu, B.; Xin, Z.; Liu, M.; Kosogof,
C.; Szczepankiewicz, B. G.; Wang, S.; Clampit, J. E.;
Gum, R. J.; Haasch, D. L.; Trevillyan, J. M.; Sham, H. L.
Bioorg. Med. Chem. Lett. 2006, 16, 5723.
Biochem. J. 2003, 371, 199; (b) Gaillard, P.; Jeanclaude-
Etter, I.; Ardissone, V.; Arkinstall, S.; Cambet, Y.;
Camps, M.; Chabert, C.; Church, D.; Cirillo, R.;
Gretener, D.; Halazy, S.; Nichols, A.; Szyndralewiez,
C.; Vitte, P.-A.; Gotteland, J.-P. J. Med. Chem. 2005, 48,
4596; (c) Stocks, M. J.; Barber, S.; Ford, R.; Leroux, F.;
St-Gallay, S.; Teague, S.; Xue, Y. Bioorg. Med. Chem.
Lett. 2005, 15, 3459; (d) Swahn, B.-M.; Huerta, F.;
Kallin, E.; Malmstro¨m, J.; Weigelt, T.; Viklund, J.;
¨
Womack, P.; Xue, Y.; Ohberg, L. Bioorg. Med. Chem.
8. Liu, M.; Xin, Z.; Clampit, J. E.; Wang, S.; Gum, R. J.;
Haasch, D. L.; Trevillyan, J. M.; Abad-Zapatero, C.; Fry,
E. H.; Sham, H. L.; Liu, G. Bioorg. Med. Chem. Lett.
2006, 16, 2590.
9. (a) Scapin, G.; Patel, S. B.; Lisnock, J.-M.; Becker, J. W.;
LoGrasso, P. V. Chem. Biol. 2003, 10, 705; (b) LoGrasso,
P. V.; Lisnock-Geissler, J.-M.; Xanthoudakis, S.; Tam, J.;
Bilsland, J. G.; Happer, S. J.; Young, L. WO 01/91749,
2001.
10. Karp, V. K.; Portnyagina, V. A.; Barkova, I. S. Khimiya
Geterotsiklicheskikh Soedineneii 1977, 9, 1252.
11. A library of 30 analogs with substituted phenols was
synthesized (data not shown). None of these compounds
gave any improvement in JNK1 activity compared to the
unsubstituted phenol 6g.
12. Gupta, S.; Campbell, D.; Derijard, B.; Davis, R. J. Science
1995, 267, 389, Also see Refs. 7a and 8 for detailed assay
procedures.
13. Crystallographic coordinates of JNK1 complexed with
2b have been deposited at the Protein Data Bank
(www.rcsb.org/) with Accession code 2no3.
14. (a) Scapin, G. Drug Discov. Today 2002, 7, 601; (b)
Schindler, T.; Bornmann, W.; Pellicena, P.; Miller, W. T.;
Clarkson, B.; Kuriyan, J. Science 2000, 289, 1938; (c)
Lett. 2005, 15, 5095; (e) Graczyk, P. P.; Khan, A.; Bhatia,
G. S.; Palmer, V.; Medland, D.; Numata, H.; Oinuma,
H.; Catchick, J.; Dunne, A.; Ellis, M.; Smales, C.;
Whitfield, J.; Neame, S. J.; Shah, B.; Wilton, D.; Morgan,
L.; Patel, T.; Chung, R.; Desmond, H.; Staddon, J. M.;
Sato, N.; Inoue, A. Bioorg. Med. Chem. Lett. 2005, 15,
4666; (f) Swahn, B.-M.; Xue, Y.; Arzel, E.; Kallin, E.;
Magnus, A.; Plobeck, N.; Viklund, J. Bioorg. Med. Chem.
Lett. 2006, 16, 1397.
5. (a) Aguirre, V.; Uchida, T.; Yenush, L.; Davis, R.; White,
M. F. J. Biol. Chem. 2000, 275, 9047; (b) Hirosumi, J.;
Tuncman, G.; Chang, L.; Go¨rgun, C. Z.; Uysal, K. T.;
¨
Maeda, K.; Karin, M.; Hotamisligil, G. S. Nature 2002,
420, 333; (c) Bennett, B. L.; Satoh, Y.; Lewis, A. J. Curr.
Opin. Pharmacol. 2003, 3, 420.
6. Waeber, G.; Delplanque, J.; Bonny, C.; Mooser, V.;
Steinmann, M.; Widmann, C.; Maillard, A.; Miklossy, J.;
Dina, C.; Hani, E. H.; Vionnet, N.; Nicod, P.; Boutin, P.;
Froguel, P. Nat. Genet. 2000, 24, 291.
7. (a) Szczepankiewicz, B. G.; Kosogof, C.; Nelson, L. T. J.;
Liu, G.; Liu, B.; Zhao, H.; Serby, M. D.; Xin, Z.; Liu, M.;
Gum, R. J.; Haasch, D. L.; Wang, S.; Clampit, J. E.;
Johnson, E. F.; Lubben, T. H.; Stashko, M. A.; Olejnic-
zak, E. T.; Sun, C.; Dorwin, S. A.; Haskins, K.; Abad-
Zapatero, C.; Fry, E. H.; Hutchins, C. W.; Sham, H. L.;
Rondinone, C. M.; Trevillyan, J. M. J. Med. Chem. 2006,
`
Wang, Z.; Canagarajah, B. J.; Boehm, J. C.; Kassisa, S.;
Cobb, M. H.; Young, P. R.; Abdel-Meguid, S.; Adams, J.
L.; Goldsmith, E. J. Structure 1999, 6, 1117.