Pyrimido-oxazepine as a versatile template for the development of inhibitors of specific kinases

Article abstract of DOI:10.1016/j.bmcl.2005.08.098

Pyrimido-oxazepine based sub-micromolar inhibitors (2-4) for Aurora and FLT-3 were designed and synthesized. These preliminary results supported the potential use of pyrimido-oxazepines as a versatile template for developing specific kinase inhibitors.

Full text of DOI:10.1016/j.bmcl.2005.08.098

Bioorganic & Medicinal Chemistry Letters 15 (2005) 5474–5477
Pyrimido-oxazepine as a versatile template for the development
of inhibitors of specific kinases
Weitao Pan,a Hu Liu,a,* Yong-Jiang Xu,a Xin Chen,a Ki Hwan Kim,a Daniel L. Milligan,a
John Columbus,a Yaron R. Hadari,b Paul Kussie,b Wai C. Wonga and Marc Labellea
aDepartment of Chemistry, ImClone Systems Incorporated, 710 Parkside Avenue, Brooklyn, NY 11226, USA
bDepartment of Protein Science, ImClone Systems Incorporated, 180 Varick Street, NY 10014, USA
Received 1 August 2005; accepted 29 August 2005
Available online 28 September 2005
Abstract—Pyrimido-oxazepine based sub-micromolar inhibitors (2–4) for Aurora and FLT-3 were designed and synthesized. These
preliminary results supported the potential use of pyrimido-oxazepines as a versatile template for developing specific kinase
inhibitors.
ꢀ 2005 Elsevier Ltd. All rights reserved.
In the past decade, the kinase targeted cancer therapy1
has gained increasing momentum, leading to the launch
4
A
of small molecule kinase inhibitors such as Gleevec,TM,
2
N
2
7
R1
Irresa,TM, and Tarceva.TM, Both Iressa and Tarceva
are quinazoline based chemical entities that target the
epidermal growth factor receptor (EGFR). The quina-
zolines, like most of the known kinase inhibitors, are
ATP competitive inhibitors. Due to the highly conserved
ATP binding pockets, and their less conserved neighbor-
ing sites, it is feasible to modify the substituents on a
common scaffold to target specific kinases. For example,
quinazoline as a structural class has been reported as
potent inhibitors of not only EGFR but also other ki-
nases.5 Recently, we disclosed the pyrimido-oxazepine
1 as potent EGFR inhibitors.6 We therefore envisioned
the possibility of extending this scaffold to inhibitors of
other selected kinases. In this letter, we want to report
our preliminary results that will serve to validate pyrim-
ido-oxazepine, such as 1–4 (Fig. 1), as a new versatile
scaffold for selected kinase inhibitors, upon modifica-
tions of its substitution pattern (e.g., R1 and R2).
3
4
8
R2
N
O
10
9
1. A: -NH-CH2-; R1 = 4-NHAr;
R2 = H or alkoxy;
; R2 = H or alkoxy;
NHCOAr
N
2. A: -NH-CH2-, R1 = 4-
HN
N
3. A: -NH-CH2-; R1 = Cl;
4. A: -N=CH-; R1 = Cl;
R2 = 8-NHCONHAr;
R2 = 8-NHCONHAr
Figure 1.
Rational drug design and computer modeling generated
oxazepines 2 as potential inhibitor for Aurora, as well as
3 and 4 for FLT-3.
A convergent synthetic route toward 2 was developed,9
featuring a Buchwald-type coupling of weakly nucleo-
philic amines, such as 6,10 with 4-chloro oxazepine 59
(Scheme 1). Coupling of independently prepared 5 and
6 expedited SAR studies aimed at examining the influ-
ence of substitutions on either aryl ring. The Aurora
kinase inhibitory activity of 2 was determined with an
Aurora A kit (Table 1).11
Aurora7 and FLT-38 were chosen as the biological tar-
gets based on an overall evaluation of their clinical rel-
evance in cancer treatment and room for improvement
among the known inhibitors currently in development.
Results in Table 1 indicate that as expected, variations
on R2 had dramatic effects on the Aurora A inhibitory
activity. When Ar was phenyl (2a–e), a methoxy
Keywords: Kinase inhibitor; Oxazepine; FLT-3; Aurora.
*
Corresponding author. Tel.: +1 917 755 5856; fax: +1 718 633
0856; e-mail: huliu205@yahoo.com
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.08.098

W. Pan et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5474–5477
5475
H
N
Ar
N
H
N
N
O
N
Cl
N
N
NH2
H
N
NH
N
H
N
Ar
R2
N
O
R2
8
N
7
6
O
O
9
10
5
2
Scheme 1. Reagents and conditions: 0.25 equiv of Pd2(dba)3, 0.75 equiv of BINAP, t-BuONa, dioxane, 90 ꢁC.
Table 1. In vitro inhibition of Aurora A by 2
H
N
H
N
4
4
a
Compound R2
Ar
Aurora A inhibition IC50 (lM)
a
N
N
Cl
Cl
NO2
2a
2b
2c
2d
2e
2f
H
Ph
Ph
Ph
Ph
1.9
2.2
NH2
2
N
2
O
N
O
7-MeO
8-MeO
9-MeO
5 (R2 =8-NO2): 4-Cl
7: 2-Cl
8: 4-Cl
9: 2-Cl
0.97
0.60
>10
0.47
10-MeO Ph
OCN
Cl
CF3
8-MeO
9-MeO
9-Et2N
H
Ph
b
10
H
N
2g
2h
2i
Ph
5.0
3.7
3.5
1.2
4
p-MePh
p-MePh
p-MePh
N
Cl
8-MeO
8-MeO
9-MeO
8-MeO
9-MeO
8-MeO
9-MeO
9-MeO
NH
O
2
N
O
2j
NH
2k
2l
m-F-Ph
m-F-Ph
m-F-Ph
1.1
3a: 4-Cl
3b: 2-Cl
0.32
1.1
Cl
F3C
2m
Scheme 2. Reagents and conditions: (a) In, NH4Cl, EtOH, H2O,
reflux, 50–70%; (b) 1-chloro-4-isocyanato-2-trifluoromethyl-benzene
(10), 80%.
2n
p-F-Ph
0.33
a Average of three experiments with SD < 10%.
substituent at C-9 (2d) or C-8 (2c) was preferred. Com-
pound 2f, with 8,9-dimethoxy substitution, had an IC50
of 0.47 lM. Switching methoxy to a diethylamino sub-
stituent resulted in diminished activity (2d vs 2g). When
R2 was 9-methoxy, a fluoro substituent at the meta posi-
tion of the Ar group of the amide moiety improved the
Aurora A inhibitory activity by about 1-fold (2l vs 2d).
A similar observation was made when Ar was substitut-
ed with a para-fluoro group (2n vs 2d). In contrast, when
R2 was 8,9-dimethoxy, a meta-fluoro substituent result-
ed in a loss of activity (2m vs 2f).
nitro group of 12 was then reduced under hydrogenation
conditions, followed by condensation with the isocya-
nate 10 to give urea 13 in good overall yield. Addition
of 13 to dichloro-pyrimidine 14 under basic conditions
provided 15 in a 50% yield. The protected aldehyde moi-
ety in 15 was then released in the presence of catalytic
amount of p-toluene sulfonic acid (p-TsOH), followed
by cyclization at 120 ꢁC under vacuum, to afford imine
4a. Although this route worked well for 4a, which had
a 4-Cl substituent, it failed to provide the corresponding
2-Cl isomer.
Encouraged by the results, and to further validate
pyrimido-oxazepine as a versatile scaffold for kinase
inhibition, we shifted our effort to the preparation of
the oxazepine based FLT-3 inhibitors 3 and 4. The prep-
aration of 3 is illustrated in Scheme 2. The nitro groups
in oxazepine 5 and 79 were reduced with indium in the
presence of ammonium chloride to afford anilines 8
and 9, respectively, in moderate to good yields.12 Since
the early phases of SAR studies indicated that the pre-
ferred substituents on the terminal Ar moiety were
para-chloro and meta-trifluoromethyl, 8 and 9 were then
reacted with 1-chloro-4-isocyanato-2-trifluoromethyl-
benzene (10) to afford 3a and 3b in good yields.
Oxazepines 3 and 4 were tested in a homogeneous time-
resolved fluorescence (HTRF) assay,14 and the results
are summarized in Table 2. As expected, both 3b and
4a showed sub-micromolar activities against FLT-3,
with 4a having an IC50 of 0.45 lM. On the pyrimidine
ring, the preferred substitution position was C-2 instead
of C-4 (3a vs 3b), indicating a possible future SAR direc-
tion. In a further phase of the SAR study, we also re-
placed the chlorides in 3a and 4a with various primary
and secondary amines. However, this resulted in dimin-
ished FLT-3 inhibitory activity.15
Compound 4a was also screened against other kinases
including EGFR, KDR, ETK, FLT-1, IGF-1R, IR, c-
Met, and FGF-1R. It showed modest activities against
IR and KDR (IC50s of around 9.5 lM), and no activities
against the rest of the kinases even at 25 lM. Time
course studies also excluded the possibility of 4a acting
In order to obtain imine 4, an alternative route as shown
in Scheme 3 was developed. The aldehyde functionality
of 2-hydroxy-5-nitro-benzaldehyde (11) was first pro-
tected as its 1,3-dioxane derivative to afford 12.13 The

5476
W. Pan et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5474–5477
O
O
O
H
N
H
N
NO2
NO2
O
b, c
a
O
HO
O
HO
HO
Cl
CF3
11
12
13
Cl
N
NH2
Cl
N
d
14
Cl
O
O
N
H
N
H
N
4
N
N
e
O
O
2
NH
N
O
Cl
NH
H2N
Cl
CF3
O
N
15
4a
2
4
Cl
F3C
Scheme 3. Reagents and conditions: (a) HO(CH2)3OH, TsOH, 95%; (b) H2/Pd(C), 80%; (c) 1-chloro-4-isocyanato-2-trifluoromethyl-benzene (10),
79%; (d) 4,6-dichloro-pyrimidin-5-ylamine (14), K2CO3, DMF, 50%. (e) 1—p-TsOH, H2O, DME, 45 ꢁC; 2—120 ꢁC, vacuum, 60%.
Table 2. In vitro inhibition of FLT-3
2. Zvelebil, M. J. Curr. Opin. Oncol., Endocr. Metab. Invest.
Drugs 2000, 2, 74.
3. Baselga, J.; Averbuch, S. D. Drugs 2000, 60(Suppl. 1), 33.
4. de Bono, J. S.; Rowinsky, E. K. Trends Mol. Med. 2002,
8(4, Suppl.), S19.
4
A
N
2
Cl
NH
O
N
O
NH
5. (a) Wedge, S. R.; Kendrew, J.; Hennequin, L. F.;
Valentine, P. J.; Barry, S. T.; Brave, S. R.; Smith, N. R.;
James, N. H.; Dukes, M.; Curwen, J. O.; Chester, R.;
Jackson, J. A.; Boffey, S. J.; Kilburn, L. L.; Barnett, S.;
Richmond, G. H. P.; Wadsworth, P. F.; Walker, M.;
Bigley, A. L.; Taylor, S. T.; Cooper, L.; Beck, S.;
Juergensmeier, J. M.; Ogilvie, D. J. Cancer Res. 2005,
65, 4389; (b) Mortlock, A.; Jung, F. PCT Int. Appl.
WO2002000649, 2002; Chem. Abstr. 136, 85826; (c)
Hennequin, L. F. A.; Crawley, G. C.; McKerrecher, D.;
Ple, P.; Poyser, J. P.; Lambert, C. M. P. PCT Int. Appl.
WO2001066099, 2001; Chem. Abstr. 135, 227013; (d) Berg,
S.; Bhat, R.; Edwards, P.; Hellberg, S. PCT Int. Appl.
WO2003055492, 2003; Chem. Abstr. 139, 101024.
6. Smith, II, L. M.; Hadari, Y. PCT Int. Appl
WO2005009384, 2005; Chem. Abstr. 142, 172179.
7. For biology and small molecule inhibitors of Aurora A,
see: (a) Carmena, M.; Earnshaw, W. C. Nat. Rev. Mol.
Cell Biol. 2003, 4, 842; (b) Mahadevan, D.; Bearss, D. J.;
Vankayalapati, H. Curr. Med. Chem. Anti-Cancer Agents
2003, 3, 25; (c) Keen, N.; Taylor, S. Nat. Rev. Cancer 2004,
4, 927.
3 and 4
Cl
FLT-3 inhibition IC50 (lM)
F3C
a
Compound Cl
A
3a
3b
4a
4-Cl –NHCH2– 13.7
2-Cl –NHCH2–
0.9
0.45
4-Cl –N@CH–
a Average of three experiments with SD <10%.
as an irreversible FLT-3 inhibitor via covalent addition
to its imine bond.
In summary, pyrimido-oxazepine based sub-micromolar
inhibitors for Aurora and FLT-3 were designed and syn-
thesized (2–4). These preliminary results supported the
potential application of pyrimido-oxazepine as a versa-
tile template for the development of inhibitors of specific
kinases.
8. For biology and small molecule inhibitors of FLT-3, see:
(a) Sawyers, C. L. Cancer Cell 2002, 1, 413; (b) Brown, P.;
Small, D. Eur. J. Cancer 2004, 707; (c) Levis, M.; Small,
D. Curr. Pharm. Des. 2004, 10, 1183.
9. Xu, Y.-J.; Liu, H.; Pan, W.; Chen, X.; Wong, W. C.;
Labelle, M., Tetrahedron Lett, in press.
Acknowledgment
We are grateful to Dr. Peter Bohlen for his support and
helpful discussion.
10. Lock, A. A.; Keen, N. J. PCT Int. Appl. WO0121597,
2001; Chem. Abstr. 134, 266318.
11. In vitro assay for Aurora A kinase inhibition: Aurora A
kinase inhibition is determined with a CycLexꢂ Aurora A
kinase Assay/Inhibitor Screening Kit from MBL Interna-
tional. This kit uses recombinant Lats2 as a specific
Aurora A substrate. A detector antibody specifically
recognizes only the phosphorylated form of the serine83
residue on Lats2 in an ELISA format. The kit was used as
described by the manufacturer, but the ATP concentration
References and notes
1. (a) Fabbro, D.; Garcia-Echeverria, C. G. Curr. Opin. Drug
Discov. Devel. 2002, 5, 701; (b) Bridges, A. J. Chem. Rev.
2001, 101, 2541; (c) Smith, J. K.; Mamoon, N. M.; Duhe,
R. J. Oncol. Res. 2004, 14, 175.

W. Pan et al. / Bioorg. Med. Chem. Lett. 15 (2005) 5474–5477
5477
in the kinase reaction was lowered from the suggested 50–
2 lM.
12. Banik, B. K.; Nanik, I.; Becker, F. F. Organic Syntheses
2004, 81, 188.
13. Toyota, E.; Sekizaki, H.; Itoh, K.; Tanizawa, K. Chem.
Pharm. Bull. 2003, 5, 625.
14. In vitro assay for FLT-3 kinase inhibition: FLT-3 tyrosine
kinase inhibition is determined by measuring the phos-
phorylation of poly-Glu-Tyr-biotin (pGT-biotin, CIS Bio
International) peptide in a Homogeneous Time-Resolved
Fluorescence (HTRF) assay. Into a black 96-well Costar
plate is added 2 ll/well of 25· compound in DMSO (final
compound concentration in the 50-ll kinase reaction is
typically 1 nM to 10 lM). Next, 38 ll of reaction buffer
(25 mM Hepes, pH 7.5, 5 mM MgCl2, 5 mM MnCl2,
0.1 mg/ml BSA, and 2 mM DTT) containing 2 nmol pGT-
biotin and 5 ng FLT-3 enzyme is added to each well. After
5–10 min pre-incubation, the kinase reaction is initiated by
the addition of 10 ll of 10 lM ATP in reaction buffer,
after which the plate is incubated at room temperature for
45 min. The reaction is stopped by the addition of 50 ll
KF EDTA buffer (50 mM Hepes, pH 7.5, 0.5 M KF, 1 mg/
ml BSA, and 100 mM EDTA) containing 0.23 lg/ml
PY20K (Eu-cryptate labeled anti-phosphotyrosine anti-
body, CIS Bio International). After 30 min, 100 ll of
10 nM SV-XL (modified-APC-labeled Streptavidin, CIS
Bio International) in KF buffer is added, and after an
additional 2-h incubation at room temperature, the plate is
read in a RUBYstar HTRF Reader.
15. Unpublished results.