A novel 5-[1,3,4-oxadiazol-2-yl]-N-aryl-4,6-pyrimidine diamine having dual EGFR/HER2 kinase activity: Design, synthesis, and biological activity

Article abstract of DOI:10.1016/j.bmcl.2008.07.057

A novel 5-[1,3,4-oxadiazol-2-yl]-N-aryl-4,6-pyrimidine diamine was synthesized and found to have potent dual EGFR/HER2 kinase inhibitory activity. The structure-based drug design of this molecule as well as the kinase and cellular inhibition of HER2 kinase dependent cell lines will be discussed.

Full text of DOI:10.1016/j.bmcl.2008.07.057

Bioorganic & Medicinal Chemistry Letters 18 (2008) 4896–4899
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
A novel 5-[1,3,4-oxadiazol-2-yl]-N-aryl-4,6-pyrimidine diamine having dual
EGFR/HER2 kinase activity: Design, synthesis, and biological activity
*
Terry V. Hughes , Guozhang Xu, Steven K. Wetter, Peter J. Connolly, Stuart L. Emanuel, Prabha Karnachi,
Scott R. Pollack, Niranjan Pandey, Mary Adams, Sandra Moreno-Mazza, Steven A. Middleton,
Lee M. Greenberger
Johnson & Johnson Pharmaceutical Research and Development, L.L.C., PO Box 300, 1000 Route 202, Raritan, NJ 08869, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel 5-[1,3,4-oxadiazol-2-yl]-N-aryl-4,6-pyrimidine diamine was synthesized and found to have
potent dual EGFR/HER2 kinase inhibitory activity. The structure-based drug design of this molecule as
well as the kinase and cellular inhibition of HER2 kinase dependent cell lines will be discussed.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 28 April 2008
Revised 10 July 2008
Accepted 14 July 2008
Available online 17 July 2008
Keywords:
EGFR
HER2
HER1
Kinase
Pyrimidine
Quinazoline mimic
Intramolecular hydrogen bond
Restricted rotation
Oxadiazole
There is a need to identify safe and effective compounds to re-
duce the abnormal cell proliferation that is characteristic of cancer
by targeting the genetic basis for the disease. Small molecule ki-
nase inhibitors have potential to treat specific genetic alterations
that play a role in the pathogenesis of human malignancies, and
may provide effective, low-toxicity therapies to inhibit tumor
growth. Inhibition of protein kinase activity has been a successful
approach for treatment of cancer and inflammatory disease.1
HER1 and HER2 are members of the epidermal growth factor
receptor family that are involved in the development and progres-
sion of several human cancers including lung and breast.2 These
receptors are regulated by kinase activity that can be inhibited
by small molecule inhibitors. To that end, we concentrated our ef-
forts to discover novel inhibitors of the HER kinases. The HER
receptors can be activated through homo- or heterodimerization
with other HER receptors resulting in phosphorylation events
and downstream signaling that produces excessive growth by
inducing cell proliferation and inhibiting apoptotic pathways. The
HER family of kinases can stimulate growth through heterodimer-
ization with other HER family members even when one of the
receptors, for example, HER1 (EGFR), has been inhibited by a small
molecule (or monoclonal antibody). Therefore, we focused on find-
ing a small molecule HER kinase inhibitor with activity against all
HER family members to prevent transactivation across all receptor
isoforms. Specifically, we screened our compound library for the
inhibition of EGFR and HER2 as they are hyper-activated in several
cancers, and their over-expression is frequently associated with
poor prognosis.3,4
An intermediate in a related oncology program, compound 1
(Fig. 1), was identified as a hit for HER family kinase inhibition with
O
Cl
NH
4
CHO
NH2
N
5
N
6
1
* Corresponding author. Tel.: +1 610 270 6561; fax: +1 610 270 6609.
E-mail address: hughe007@hotmail.com (T.V. Hughes).
Figure 1. Screening hit compound 1.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.07.057

T. V. Hughes et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4896–4899
4897
an IC50 value of 0.076
(63% inhibition at 100
stituent at the 5-position of the pyrimidine ring in an effort to im-
prove HER2 kinase activity in the series.
l
l
M against EGFR and modest HER2 activity
BnO
Cl
BnO
Cl
M).5 With 1 in hand we modified the sub-
NH
N
NH
R
R
The synthesis of 1 is outlined in Scheme 1. The commercially
available dichloropyrimidine 2 was treated with ammonia gas
and warmed in toluene at 60 °C, leading to the selective displace-
ment of only one of the chlorine atoms.6 The product, 4-amino-
6-chloro-pyrimidine-5-carbaldehyde (3), was reacted with com-
mercially available 4-benzyloxy-3-chloro-phenylamine in DMSO
at 100 °C in the presence of Et3N to give pyrimidine 1.
The conversion of 3 to 1 was highly dependent in the order of
addition for the reagents (Scheme 2). The conversion of 3 to 1 oc-
curred smoothly if the aniline is added last; however, if the Et3N
was added last or omitted 4 was a significant product of the reac-
tion (Table 1). Presumably, the small amount of HCl that was
formed from the desired conversion of 3 to 1 catalyzed the forma-
tion of the imine 4 as well as the formation of 5, which is the imine
of 1, unless the acid was neutralized by the addition of a base.
Remarkably, the formation of the imine 4 that occurred readily at
25 °C, in the absence of base, was not a significant product of the
reaction at 100 °C when Et3N was used as an acid scavenger.7
Initially the aldehyde functional group of 1 was used as a handle
to further explore the SAR of the series. We thought that conver-
sion of the aldehyde to a carboxylic acid derivative such as an ester
or an amide would be an efficient way to generate analogues. Our
synthesis of the corresponding esters and amides of analogue 1 is
shown in Scheme 3.
a
N
c
N
1
NH2
N
8
NH2
6 (R = CO2Me)
7 (R = CO2H)
b
Scheme 3. Reagents and conditions: (a) NaCN, MnO2, MeOH, AcOH, reflux, 80%; (b)
LiOH, THF, MeOH, H2O, 25 °C, 81%; (c) HATU, EtN(i-Pr)2, THF, R1NH(H) or R1OH,
25 °C, 60–80%.
only small amounts of amide 8 (R = CONHR1 or CONR1R2) observed.
However, utilization of HATU as the coupling reagent and diisopro-
pylethylamine in THF efficiently coupled the carboxylic acid 7 with
a series of primary and secondary amines to afford various amides
8 (R = CONHR1 or CONR1R2). Additionally, HATU was used to cou-
ple 7 with various alcohols to afford the corresponding esters 8
(R = CO2R1). The EFGR and HER2 kinase activity for a representative
number of analogues of 8 is shown in Table 2. Activity against Aur-
ora-A, CDK1, and VEGF-R2 kinases is also shown for comparison.
Although compound 1 (R = CHO) displayed good inhibition of
EFGR (IC50 = 0.076
HER2 kinase with an IC50 = 1.00
l
M), it was only modestly potent against
l
M. Therefore, our objective was
to keep the EGFR potency and try to gain potency against HER2.
However, none of the amides synthesized (8b–8g and 8i) displayed
inhibition of EGFR or HER2. The carboxylic acid 7 exhibited excel-
lent potency against EGFR (IC50 = 41 nM), and showed a slight
improvement in potency against HER2 (IC50 = 776 nM) compared
to the lead 1. The nitrile 8j displayed similar activity to the carbox-
ylic acid 7.9 The methyl ester 6 still retained potent EGFR activity
(IC50 = 54 nM), and showed a marked improvement in HER2 po-
tency (IC50 = 100 nM). However, the bulkier ester 8a, which con-
tained a 2-(morpholin-4-yl)ethylamino group, showed a loss in
potency against both EFGR and HER2 (IC50 = 150 nM and
The aldehyde 1 was converted to the methyl ester 6 by treat-
ment with sodium cyanide and MnO2 in MeOH/THF with one
equivalent of AcOH.8 The ester was readily hydrolyzed to carbox-
ylic acid 7 by treatment with LiOH in MeOH/THF/H2O. Initial at-
tempts at amide formation using EDCI were low yielding, with
BnO
1.53 lM, respectively). Both the amide and ester analogues contain
Cl
Cl
Cl
NH
two heteroatoms capable of participating in H-bonds. The inactive
amide analogues contain H-bond accepting O and a H-bond donat-
ing NH moieties. However, unlike the amide analogues of 8, both
heteroatoms of the ester analogues are capable of being H-bond
acceptors. This difference may determine the ability of the substi-
tuent at the 5-position of 8 to be held in the same plane as the
pyrimidine ring via intramolecular H-bonds. It appears that the
ability of pyrimidine ring of 8 to be relatively coplanar with the
substituent in the 5-position parallels the HER2 kinase activity ob-
served. None of the analogues showed appreciable potency against
the non-HER kinases Aurora-A, CDK1, or VEGF-R2.
In an effort to expand upon the improved activity of the ester
analogues (Table 2) and to avoid ester related metabolism we
decided to explore isosteres of esters. The oxadiazole ring has been
reported as an isostere for the ester functional group.10 Specifically,
we thought that the 1,3,4-oxadiazole 11 would be an interesting
analogue to pursue. Similar to an ester, compound 11 could also
form two pseudocycles via intramolecular hydrogen bonds but
11 would be metabolically stable to hydrolysis.
CHO
Cl
CHO
NH2
CHO
NH2
N
a
b
N
N
N
2
N
3
N
1
Scheme 1. Reagents and conditions: (a) NH3 (gas), toluene, 60 °C, 87%; (b) DMSO,
Et3N, 100 °C, 4-benzyloxy-3-chloro-phenylamine, 61%.
R
NH
Cl
Cl
R
R
CHO
NH2
a
N
N
N
N
N
1
+
+
N
5
NH2
N
4
NH2
N
3
Scheme 2. Reagent and condition: (a) DMSO, No Et3N, 25 °C, R = 4-benzyloxy-3-
chloro-phenyl.
Compound 11 was docked into a homology model of EGFR using
the software GLIDE.11–13 Molecular modeling of 11 shows two
intramolecular hydrogen bonds that hold the molecule in a rela-
tively flat pose in the ATP-site (Fig. 2).14 Compound 11 can exist
as two rotamers that are each capable of forming two intramolec-
ular H-bonds to the oxadiazole ring. While both rotamers are prob-
ably present in solution, or in the binding site of EGFR, the more
stable rotamer (ꢀ5 kcal/mol) is illustrated in Figure 2.15 The first
intramolecular hydrogen bond is between the O of the oxadiazole
Table 1
Relative ratioa of products 1, 4, and 5 without Et3N at 25 °C
Compound
6 min (%)
60 min (%)
1
4
5
12
39
2
10
46
20
a
The relative ratios of products were determined by HPLC analysis.

4898
T. V. Hughes et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4896–4899
Table 2
Kinase activity for analogues from Scheme 2
Compound
R
EGFR
% inh. 2
EGFR
IC50
HER2
Aurora-A
CDK1
VEGF-R2
a
a
a
a
a
l
M
(lM)
IC50
(
l
M)
IC50
(lM)
IC50
(
l
M)
IC50
(lM)
1
7
6
CHO
CO2H
CO2Me
79.9
75.2
83.6
ND
55.9
42.4
35.3
28.3
36.6
35.9
34.1
36.5
88.5
0.076
0.041
0.054
0.150
>100
ND
ND
ND
ND
ND
63% at 100
0.776
0.100
1.53
>100
10.0
l
M
30% at 100
73% at 100
50% at 100
55% at 100
>100
l
l
l
l
M
M
M
M
<10% at 100
<10% at 100
ND
<10% at 100
>100
>100
>100
>100
>100
>100
>100
>100
<10% at 100
l
l
M
M
<10% at 100 lM
30% at 100
ND
lM
8a
8b
8c
8d
8e
8f
8g
8h
8i
8j
CO2CH2CH2(morpholin-4- yl)
CONHC6H5
lM
11% at 100
>100
lM
CONHCH2CH2(piperidin-1-yl)
CONHCH2CH2(morpholin-4-yl)
CONHCH2CH2(4-methoxyphenyl)
CONHCH2CH2OMe
CON(CH2CH3)2
CO2CH2(3-fluorophenyl)
CONH(4-methoxyphenyl)
CN
ꢀ100
>100
>100
>100
>100
>100
>100
>100
10.0
>100
>100
>100
>100
>100
>100
0.470
>100
>100
>100
>100
>10
33% at 100
ND
ND
0.027
lM
lM
37% at 100
lM
a
IC50 data are the average of at least two separate experiments. IC50 values listed as >100 indicate no observed 50% inhibition at the highest dose tested, nor was an
inhibition maximum observed. ND indicates that the IC50 experiment was not performed.
The EGFR and HER2 kinase data for analogues 9–11 are shown
in Table 3. The kinase selectivity of hydrazide 9 favored HER2 inhi-
bition with good HER2 kinase potency (IC50 = 146 nM) and moder-
ate EGFR potency (IC50 = 458 nM). Surprisingly, the hydrazide 9
had better HER2 potency than any of the amides (Table 2, 8b–8g,
and 8i). However, acylation of the hydrazide 9 afforded compound
10 that displayed
a
significant loss in HER2 potency
(IC50 = 5.71 M) accompanied by a mild loss in EGFR potency
l
(IC50 = 620 nM). Cyclization of 10 to the oxadiazole 11 was accom-
panied by a 20-fold increase in EFGR inhibition (IC50 = 30 nM) and
70-fold increase in HER2 inhibition (IC50 = 80 nM). Notably, the
isostere 11 was observed to be more active against EGFR (5-fold)
Table 3
Kinase activity for analogues 8a, 9–11
Compound
R
EGFR
IC50
HER2
IC50
Aurora-A
IC50
CDK1
IC50
VEGF-R2
IC50
a
a
a
a
a
Figure 2. Docking of compound 11 with EGFR kinase shows intramolecular
hydrogen bonds and intermolecular hydrogen bonds with Met793.
(
l
M)
(
l
M)
(
l
M)
(
l
M)
<10% at
100
(
l
M)
8a
—
0.150
1.53
55% at
100
>10
11% at
100
>10
l
M
l
M
lM
of compound 11 and NH2 group of the amino-pyrimidine core. The
second intramolecular hydrogen bond is between the N-3 of the
oxadiazole ring and the NH of 11.
Synthesis of the 1,3,4-oxadiazole analogue 11 is shown in
Scheme 4. Methyl ester 6 was treated with hydrazine in ethanol
to give the hydrazide 9, which was coupled with 3-morpholin-4-
yl-propionic acid using EDCI in DMF to give 10. Dehydration of
10 with tosyl chloride and triethylamine in dichloromethane
formed the oxadiazole ring 11 in excellent yield.16
9
10
0.458
0.620
0.146
5.70
>100
>100
Morpholin-
4-yl-CH2CH2
Morpholin-
4-yl-CH2CH2
>100
>100
11
0.030
0.080
>100
>100
>100
a
IC50 data are the average of at least two separate experiments. IC50 values listed
as >100 indicate no observed 50% inhibition at the highest dose tested, nor was an
inhibition maximum observed. ND indicates that the IC50 experiment was not
performed.
R
NH
O
a
b
N
NHNH2
NH2
4
N
9
R
N
R
N
N
N
NH
NH
O
O
R1
NHNH R1
O
c
N
NH2
N
NH2
11
10
Scheme 4. Reagents and conditions: (a) NH2NH2, EtOH, reflux, 58%; (b) EDCI, DMF,
R1CO2H, 88%; (c) tosylchloride, Et3N, DCM, 25 °C, 85% (R = 4-benzyloxy-3-chloro-
phenyl, R1 = morpholin-4-yl-CH2CH2).
Figure 3. Compound 11 docked in to the ATP-site of EGFR1.

T. V. Hughes et al. / Bioorg. Med. Chem. Lett. 18 (2008) 4896–4899
4899
and HER2 (19-fold) than the ester 8a after which it was designed.
References and notes
Additionally, 11 was highly selective for HER family kinases as
demonstrated by its minimal activity against Aurora-A, CDK1,
and VEGF-R2.
Compound 11 is potent inhibitor of the HER2 dependent cell
lines N87 (IC50 = 411 nM), BT-474 (IC50 = 312 nM), and SK-BR3
(IC50 = 325 nM), but does not inhibit the growth of the non-HER2
1. Cohen, P. Nat. Rev. 2002, 309.
2. Mendelsohn, J.; Baselga, J. Oncogene 2000, 6550.
3. Slamon, D. J.; Clark, G. M.; Wong, S. G.; Levin, W. J.; Ullrich, A.; McGuire, R. L.
Science 1987, 177.
4. Klijn, J. G.; Berns, P. M.; Schmitz, P. I.; Foekens, J. A. Endocr. Rev. 1992, 3.
5. Xu, G.; Searle, L. L.; Hughes, T. V.; Beck, A. K.; Connolly, P. J.; Abad, M. C.;
Neeper, M. P.; Struble, G. T.; Springer, B. A.; Emanuel, S. L.; Gruninger, R. H.;
Pandey, N.; Adams, M.; Moreno-Mazza, S.; Fuetnes-Pesquera, A. R.; Middleton,
S. A.; Greenberger, L. A. Bioorg. Med. Chem. Lett. 2008, 3495.
6. Gomtsyan, A.; Didomenico, S.; Lee, C.; Matulenko, M. A.; Kim, K.; Kowaluk, E.
A.; Wismer, C. T.; Mikusa, J.; Yu, H.; Kohlhaas, K.; Jarvis, M. F.; Bhagwat, S. S. J.
Med. Chem. 2002, 3639.
7. Our process group has recently disclosed a detailed discussion of their efforts to
optimize this reaction: Choudhury, A.; Chen, H.; Nilsen, C. N.; Sorgi, K. L.
Tetrahedron Lett. 2008, 102.
8. Kumemura, T.; Choshi, T.; Hirata, A.; Sera, M.; Takahashi, Y.; Nobuhiro, J.;
Hibino, S. Heterocycles 2003, 13.
9. Compound 8j was prepared as previously described: Xu, G.; Lee, L.; Connolly, P.
J.; Middleton, S. A.; Emanuel, S. L.; Hughes, T. V.; Abad, M. C.; Karnachi, P. S.;
Wetter, S. K.; WO 2007081630, 2007.
10. Watjen, F.; Baker, R.; Engelstoff, M.; Herbert, R.; MacLeod, A.; Knight, A.;
Merchant, K.; Moseley, J.; Saunders, J.; Swain, C. J.; Wong, E.; Springer, J. P. J.
Med. Chem. 1989, 2282.
dependent HeLa cell line (IC50 > 100 lM).
When modeled in EGFR kinase, 11 sits in the ATP-binding cleft.
The amino-pyrimidine ring is hydrogen bonded to the hinge region
between the NH2- and COOH-terminal lobes of the kinase (Fig. 3).
N-1 of the pyrimidine is hydrogen bonded to the main chain NH of
Met793, whereas the amino group forms a hydrogen bond to the
main chain C@O of Met793. The 3-chloro-4-(benzyloxy)aniline
group is oriented deep in the ATP-binding site and makes predom-
inantly hydrophobic interactions with the protein. The aniline
nitrogen and the ether oxygen do not pick up any direct hydrogen
bonding interactions with the protein. The 2-(morpholin-4-yl)eth-
ylamino group on the oxadiazole ring is positioned toward the sol-
vent interface, and provides
a good handle to modify the
11. ProteinPrep module, Schrodinger, New York, 2005.
12. Glide 3.5, Schrodinger, New York, 2005.
physicochemical properties of the series.
13. Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.;
Repasky, M. P.; Knoll, E. H.; Shaw, D. E.; Shelley, M.; Perry, J. K.; Francis, P.;
Shenkin, P. S. J. Med. Chem. 2004, 1739.
14. No scaling factors were applied to the van der Waals radii. Default settings
were used for all the remaining parameters. The top 20 poses based on
Glidescore were energy-minimized with Macromodel using the OPLS-AA, and
the refined poses were reranked using Glidescore.
15. The energies were calculated using the OPLS_2005 force field with a constant
dielectric of 1.0 using the Macromodel module.
16. Choi, Y.; Ishikawa, H.; Velcicky, J.; Elliott, G. I.; Miller, M. M.; Boger, D. L. Org.
Lett. 2005, 4539.
In summary, we have described the early SAR leading to a novel
5-[1,3,4-oxadiazol-2-yl]-N-aryl-4,6-pyrimidine diamine, which
represents a new and selective kinase inhibitor scaffold. The key
modification that improved the potency of these analogues was
the introduction of the 1,3,4-oxadiazole as an isosteric replace-
ment of an ester functional group. This scaffold could be viewed
as a mimic of the ubiquitous quinazoline scaffold for kinase inhibi-
tion. Extensive exploration of the SAR of this new scaffold and its
ability to inhibit different kinases is underway.