Design and preparation of 2-benzamido-pyrimidines as inhibitors of IKK

Article abstract of DOI:10.1016/j.bmcl.2005.09.035

The design, synthesis, and the biological evaluation of 2-benzamido-pyrimidines as novel IKK inhibitors are described. By optimization of the lead compound 3, compounds 16 and 24 are identified as good inhibitors of IKK2 with IC₅₀ values of 40 and 25 nM, respectively. Compound 16 also demonstrated significant in vivo activity in an acute model of cytokine release.

Full text of DOI:10.1016/j.bmcl.2005.09.035

Bioorganic & Medicinal Chemistry Letters 16 (2006) 108–112
Design and preparation of 2-benzamido-pyrimidines
as inhibitors of IKK
Rudolf Waelchli,* Birgit Bollbuck, Christian Bruns, Thomas Buhl, Jo¨rg Eder,
Roland Feifel, Rene Hersperger, Philipp Janser, Laszlo Revesz,
Hans-Gunter Zerwes and Achim Schlapbach
¨
Novartis Institutes for BioMedical Research, CH-4002 Basel, Switzerland
Received 19 July 2005; revised 9 September 2005; accepted 15 September 2005
Available online 19 October 2005
Abstract—The design, synthesis, and the biological evaluation of 2-benzamido-pyrimidines as novel IKK inhibitors are
described. By optimization of the lead compound 3, compounds 16 and 24 are identified as good inhibitors of IKK2 with
IC50 values of 40 and 25 nM, respectively. Compound 16 also demonstrated significant in vivo activity in an acute model
of cytokine release.
Ó 2005 Elsevier Ltd. All rights reserved.
Over the past few years, IjB kinase (IKK) emerged as a
prime target for the development of novel anti-rheumat-
ic and anti-inflammatory drugs.1 IKK is part of a high
molecular weight protein complex which consists of
three subunits, the structurally related kinases IKK1
and IKK2, and the regulatory subunit NEMO (IKKc).
However, in studies using genetic mutants it was shown
that IKK2 is the essential kinase for the production of
pro-inflammatory cytokines such as IL-1b and TNFa.2
Recently, several classes of low-molecular weight inhib-
itors of IKK2 were disclosed, some of which demon-
strate impressive activity in animal models.3 These
findings further confirm the importance of IKK2 as a
key player in the transduction of pro-inflammatory
signals.
the most interesting hits. Both of them are active in
the micromolar range.
O
H
H
N
NH2
O
O
N
N
N
N
OH
N
S
N
2
1
N
H
OH
IC50 IKK2: 15 µM
IC50 IKK2: 30 µM
During our first attempts to develop a structure–activity
relationship (SAR) and to identify the key features that
would drive the inhibition, we quickly realized that hy-
brid molecules between 1 and 2, combining structural
features of the two parent hits, led to a dramatic
improvement in potency (Table 1).
In this paper, we present our own efforts in the discov-
ery of novel and selective inhibitors of IKK, particu-
larly of IKK2. By screening the Novartis compound
archive, several hits with typical kinase inhibitor mo-
tifs (2-anilino-pyrimidines and 2,4-disubstituted quina-
zolines) were found. Compounds 1 and 2 (depicted
below) are typical prototypes and represent two of
The para-carboxyl-substituted anilino-pyrimidine 3 was
the most potent representative of this small series. It
was slightly more active than the meta- but clearly fa-
vored over the ortho-derivative (compounds 4 and 5).
Extension by one carbon to the para-acetic acid 6 did
not improve the potency against IKK2.
Keywords: 2-Benzamido-pyrimidines; IKK inhibitors; IKK2; In vivo
activity.
*
Interestingly, compound 3 was only weakly active on
the complete IKK complex isolated from HeLa cells,4
Corresponding author. Tel.: +41 61 324 64 73; fax: +41 61 324 88
47; e-mail: rudolf.waelchli@novartis.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.09.035

R. Waelchli et al. / Bioorg. Med. Chem. Lett. 16 (2006) 108–112
109
Table 1. Structure–activity relationship of carboxyl-substituted
2-anilino-pyrimidines
tively charged residue was not mandatory since the
amide 10 completely retained activity against IKK2.
Based on this initial result we decided to prepare a small
library of 2-anilino-pyrimidines with para-substituted
amino-benzamides. The synthetic strategy is outlined
in Scheme 1.
H
X
N
N
N
S
Suzuki coupling5 of 2,4-dichloropyrimidine and 2-ben-
zothiophene boronic acid resulted in intermediate 7,
which was treated with methyl 4-amino-benzoate at high
temperature. The resulting ester 8 was saponified with
sodium hydroxide and then treated with thionyl chloride
to give key intermediate 9. Finally, reaction of com-
pound 9 with a set of aliphatic amines in the presence
of triethylamine produced the desired benzamides. Data
for selected compounds are summarized in Table 2.
Compound
X
IKK2 IC50 IKK1 IC50 IKK
(lM)
(lM)
complex
IC50 (lM)
3
4
5
6
p-CO2H
m-CO2H
o-CO2H
0.6
5
n.d.
18
n.d.
1
>100
>100
n.d.
>100
n.d.
Generally, compounds with an additional basic amino
group in the amide portion were more active than the
neutral inhibitors 10 and 13. Tertiary (and more basic)
amines were better than primary ones (cf. 12 and 14 vs
11). Finally, the replacement of the flexible ethyl chain
in 12 or 14 by the conformationally restricted piperidine
ring (cf. 15 and 16) resulted in a 10-fold increase in
potency. The tertiary amines, such as 15 and 16, were
p-CH2CO2H
2.2
despite the fact that it inhibited both isoforms of the
IKK enzymes.
When we replaced the carboxyl group in 3 by an amide
functionality (Table 2), it became apparent that a nega-
Table 2. Structure–activity relationship of para-amino benzamides
H
N
N
R
O
X
N
S
Compound
X
R
H
IKK2 IC50 (lM)
IKK1 IC50 (lM)
IKK complex IC50 (lM)
10
HN
0.9
3.8
8.5
NH2
N
11
12
13
14
15
16
17
18
H
0.6
0.2
3.7
1.2
n.d.
1.2
0.15
0.2
0.4
0.1
4.4
H2N
H
0.3
H2N
H2N
H
>100
0.4
n.d.
0.5
H
N
H2N
HN
H
0.04
0.04
0.09
0.05
0.08
0.07
0.25
0.15
N
H
HN
HN
HN
N
N
N
Cl
OMe

110
R. Waelchli et al. / Bioorg. Med. Chem. Lett. 16 (2006) 108–112
N
Cl
thionyl chloride produced the corresponding acid chlo-
ride 21, which was easily converted into amide 22.
Subsequent treatment with phosphoroxy chloride
resulted in chloro-pyrimidine 23. The desired new
pyrimidines were finally obtained by a Suzuki coupling
between 23 and a variety of substituted 2-thiophenyl
boronic acids. These novel boronic acids were synthe-
sized according to the general strategy outlined in
Scheme 3.
N
H2N
H
N
Cl
N
N
O
Cl
+
N
S
N
S
O
O
a)
O
HO
OH
b)
B
7
8
S
c),d)
H
H
N
N
R2
N
R2 R1
N
N
The SAR for this series of compounds is summarized in
Table 3. In general, replacement of the benzothiophene
group in 16 by 5-substituted thiophenes was well tolerat-
ed without loss of activity. Saturated side chains ap-
peared to be slightly favored over the more rigid
propargyl chain (cf. 28 vs 27). Hydrogen-bond donors
like hydroxyl (24) or amino groups (28) had a beneficial
effect on the potency. Both, methylation, as in 25, or
acetylation, as in 29, yielded less potent compounds.
Remarkably, the loss of activity of compound 29 was
much more dramatic for IKK1 than for IKK2, an
observation that we could not explain and we have
not followed up so far. Introduction of steric bulk in
the form of geminal dialkyl groups (cf. 26, 28 or 30)
was well tolerated, indicating that the binding pocket
provides enough space to accommodate bulky
substituents.7
N
H
N
R1
N
Cl
O
O
e)
S
S
9
Scheme 1. Synthesis of 2-benzamido-4-benzothiophenyl-pyrimidines.
Reagents and conditions: (a) Pd[P(Ph)3]4; DME, 1 N NaHCO3; (b)
neat, 200 °C; (c) 1 N NaOH; (d) SOCl2; (e) ÔamineÕ, THF, NEt3.
active in the low-nanomolar range and were the most
potent compounds of this series. Further substitution
of the phenyl ring of the benzamide moiety did not
improve activity (compounds 17 and 18).
Most of the benzamides were moderately selective for
IKK2 (over IKK1) but, in contrast to the carboxylic
acid 3, showed comparable activity against the complete
IKK complex.
In order to determine whether blockade of IKK2 result-
ed in inhibition of relevant downstream events, selected
compounds were tested in a cellular assay assessing the
functional consequence of IKK inhibition as detected
by the blockade of IjBa degradation.8 In this cellular
assay, the potency of the compounds was about 10- to
40-fold lower than in the cell-free kinase assays (see Ta-
ble 4), presumably reflecting incomplete penetration of
the compounds into the cells or different ATP concen-
trations in the cells compared to the cell-free assay
system.
To further extend the SAR of the benzamide series, we
next replaced the benzothiophene group by a variety
of structurally close, but more polar, thiophenes. The
synthesis is outlined in Scheme 2.
Treatment of 2-methylsulfanyl-pyrimidin-4-ol, 19, with
methyl para-amino-benzoate in 1,3-dimethyl-imidazoli-
din-2-one (DMEU) at 170 °C gave anilino-pyrimidine
20. Ester hydrolysis of 20 followed by treatment with
Generally, the SAR in this cellular assay correlated nice-
ly with the SAR in the IKK2 assay.
H2N
H
O
O
N
S
H
N
N
N
N
N
To determine the selectivity of our compounds on a cel-
lular level, we tested the same set of compounds (Table
4) for their ability to interfere with TNFa-stimulated
expression of the adhesion molecules E-selectin,
N
Cl
O
N
b),c)
a)
OH
OH
O
OH
O
19
20
21
d)
H
H
N
N
N
N
N
N
R
N
N
N
N
a)
e)
S
Br
S
Cl
O
X
O
OH
R
X
R
23
22
f)
R
H
b),c)
N
N
N
HO
R
d)
R
S
S
B
N
N
S
HO
XP
XP
O
R
R
X = N or O; R = H, Me, Et
R
Scheme 3. Synthesis of substituted thiophene boronic acids. Reagents
and conditions: (a) ÔSonogashira6Õ coupling, PdCl2[P(Ph)3]2, Cu(I)I,
NEt3, DMF; (b) H2/Pd/C; (c) introduction of protecting group ÔPÕ; (d)
LDA, B(OEt)3.
Scheme 2. Synthesis of 2-anilino-4-thiophenyl-pyrimidines. Reagents
and conditions: (a) DMEU, 170 °C; (b) 1 N NaOH; (c) SOCl2; (d) 4-(1-
pyrrolidinyl)-piperidine, NEt3; (e) POCl3, 4 N HCl in dioxane; (f)
Pd[P(Ph)3]4, Ô2-thiophenyl-boronic acidÕ.

R. Waelchli et al. / Bioorg. Med. Chem. Lett. 16 (2006) 108–112
Table 3. Structure–activity relationship of benzthiophene replacements
111
H
N
N
X
N
N
N
O
Compound
X
IKK2 IC50 (lM)
IKK1 IC50 (lM)
IKK complex IC50 (lM)
S
S
S
24
0.025
0.5
0.04
*
OH
25
26
27
28
29
30
0.1
1.0
0.7
0.1
*
O
0.05
0.15
0.04
0.15
0.06
0.05
0.2
*
*
OH
S
0.5
NH2
NH2
S
S
0.03
30.0
0.03
0.04
0.8
*
*
O
N
H
S
0.03
*
NH2
Table 4. Cellular profile of selected compounds
Compound
IjB degradation (lM)
E-Sel (lM)
ICAM (lM)
VCAM (lM)
b2M (lM)
HLA-DR (lM)
3
10
12
15
16
24
25
26
100
20
5
50
2
50
2
50
2
100
10
2
100
0.25
0.5
2
0.5
0.3
0.5
0.2
0.4
0.45
0.5
0.4
0.3
0.2
0.6
0.5
0.5
0.4
0.3
0.2
0.5
0.6
2
2
1
2
2
2
5
2.5
2
1
5
2
5
3
ICAM-1, and VCAM-1 in HUVEC cells.9 In addition,
the compounds were also tested for their effects on
IFNc-stimulated expression of the MHC molecules b2
microglobulin and HLA-DR, both events which are
not dependent on IKK activity.10,11 As illustrated in
Table 4, our compounds inhibited TNFa-induced adhe-
sion molecule expression in a potency range similar to
the IjBa degradation. Although compounds generally
showed activity in the IFNc-induced expression of b2
microglobulin or HLA-DR, their potency in these as-
says was 4- to 10-fold (example 16, 25, and 26) or even
10- to 15-fold (example 24) weaker. These data demon-
strate that our IKK2 inhibitors have an effect on down-
stream gene expression, however, on the cellular level
the selectivity was modest.
Compound 16 was also tested in two animal models.
First, its efficacy to inhibit TNFa release into plasma
upon LPS-challenge in the rat was determined. The
compound was dosed sc (30 mg/kg) or orally (30 mg/
kg) 1 h prior to the LPS-challenge. Four hours after
the challenge, plasma was collected and the systemic
TNFa levels were analyzed using a commercially avail-
able ELISA kit. Both routes of administration of inhib-
itor 16 at the indicated dose resulted in a significant
inhibition of 86% (sc) and 75% (p.o.).12 In a second
experiment, we could show that compound 16 was also
active in the thioglycollate-induced peritonitis model in
the mouse.13,14 The maximal inhibition of neutrophil
extravasation in this model was about 50% at a dose
of 10 mg/kg sc.15

112
R. Waelchli et al. / Bioorg. Med. Chem. Lett. 16 (2006) 108–112
7. For a possible binding mode, see: Bingham, A. H.;
Davenport, R. J.; Gowers, L.; Knight, R. L.; Lowe, Ch.;
Owen, D. A.; Parry, D. M.; Pitt, W. R. Bioorg. Med.
Chem. Lett. 2004, 14, 409.
In summary, optimization of the lead compound 3
through a combined parallel synthetic and classical
medicinal chemistry effort resulted in potent inhibitors
of IKK2. On a cellular level we could show that these com-
pounds inhibit the IjBa degradation and downstream
events, like adhesion molecule expression. However, the
selectivity of these compounds for the IKK pathway is
only moderate in cellular assays. Compound 16 is orally
bio-available in rats and mice, and we could demonstrate
significant in vivo activity in an acute model of cytokine
release. Future efforts will focus on further improvements
in potency and selectivity for this compound series.
8. IjBa degradation assay: for the assay, THP-1 cells were
transferred to 96-well plates at a density of 600,000 cells
per well and stimulated with 10 ng/ml TNFa for 12 min
at rt. For measuring the inhibition of TNFa-stimulated
IjBa degradation, the test compounds were dissolved in
DMSO and added to the cell suspension prior to
addition of TNFa. The concentration of IjBa in each
sample was analyzed by Western blotting. Each com-
pound was tested in duplicate and estimations of IC50
values were done based on densitometric scanning or
visual inspection.
References and notes
9. Zerwes, H.-G.; Peter, J. C.; Link, M.; Gubler, H.-P.;
Scheel, G. Anal. Biochem. 2002, 304, 166.
1. Burke, J. R. Curr. Opin. Drug Discov. Dev. 2003, 6, 720.
2. (a) Gosh, S.; Karin, M. Cell 2002, 109, S81; (b) Li, Q.;
Verma, I. M. Nat. Rev. Immunol. 2002, 2, 725; (c)
Mercurio, F.; Murray, B. W.; Shevchenko, A.; Bennett,
L. B.; Young, D. B.; Wu Li, J.; Pascual, G.; Motiwala, A.;
Zhu, H.; Mann, M.; Manning, A. M. Mol. Cell. Biol.
1999, 19, 1526.
10. Bach, E. A.; Aguet, M.; Schreiber, R. D. Annu. Rev.
Immunol. 1997, 15, 563.
11. Boehm, U.; Klamp, T.; Groot, M.; Howard, J. C. Annu.
Rev. Immunol. 1997, 15, 749.
12. This result correlates nicely with the ability of compound
16 to inhibit LPS induced TNFa release from hPBMC
with an IC50 of 0.3 lM.
3. (a) McIntyre, K. W.; Shuster, D. J.; Gillooly, K. M.;
Dambach, D. M.; Pattoli, M. A.; Lu, P.; Zhou, X.-D.;
Qiu, Y.; Zusi, F. C.; Burke, J. R. Arthritis Rheum. 2003,
48, 2652; (b) Podolin, P. L.; Callahan, J. F.; Bolognese, B.
J.; Li, Y. H.; Carlson, K.; Davis, T. G.; Mellor, G. W.;
Evans, C.; Roshak, A. K. J. Pharmacol. Exp. Ther. 2005,
312, 373.
4. Heilker, R.; Freuler, F.; Vanek, M.; Pulfer, R.; Kobel, T.;
Peter, J.; Zerwes, H.-G.; Hofstetter, H.; Eder, J. Biochem-
istry 1999, 38, 6231.
13. Cecconi, O.; Nelson, R. M.; Roberts, W. G.; Hana-
saki, K.; Mannori, G.; Schultz, C.; Ulich, T. R.;
Aruffo, A.; Bevilacqua, M. P. J. Biol. Chem. 1994,
269, 15060.
14. Bosse, R.; Vestweber, D. Eur. J. Immunol. 1994, 24, 3019.
15. Peritonitis was induced at t = 0 by ip injection of thiogly-
collate and allowed to develop for 4 h. The compound
dissolved at 1 mg/ml D(+)glucose/PEG400/H2O and then
further diluted in PEG400/H2O was applied sc at t = 0.
Control animals received an injection of sterile saline only.
After 4 h, the animals were sacrificed, PMNs migrated into
peritoneum were counted and their number was compared
to vehicle control.
5. Watanabe, T.; Suzuki, A.; Miyaura, N. Synlett 1992, 3, 207.
6. Tonda, Y.; Sonogashira, K.; Hagihara, N. Synthesis 1977,
11, 777.