3-Amido-4-anilinocinnolines as a novel class of CSF-1R inhibitor

Article abstract of DOI:10.1016/j.bmcl.2011.01.033

3-Amido-4-anilinocinnolines have been identified as potent and highly selective inhibitors of CSF-1R. The synthesis and SAR of these compounds is reported, along with some physical property, pharmacokinetic and kinase selectivity data.

Full text of DOI:10.1016/j.bmcl.2011.01.033

Bioorganic & Medicinal Chemistry Letters 21 (2011) 1382–1384
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
3-Amido-4-anilinocinnolines as a novel class of CSF-1R inhibitor
⇑
David A. Scott , Les A. Dakin, David J. Del Valle, R. Bruce Diebold, Lisa Drew, Thomas W. Gero,
Claude A. Ogoe, Charles A. Omer, Galina Repik, Kumar Thakur, Qing Ye, Xiaolan Zheng
Cancer Research, AstraZeneca R&D Boston, 35 Gatehouse Drive, Waltham, MA 02451, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
3-Amido-4-anilinocinnolines have been identified as potent and highly selective inhibitors of CSF-1R. The
synthesis and SAR of these compounds is reported, along with some physical property, pharmacokinetic
and kinase selectivity data.
Received 10 December 2010
Revised 7 January 2011
Accepted 10 January 2011
Available online 14 January 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Kinase inhibitor
Cinnoline
CSF-1R
CSF-1R is a class III receptor tyrosine kinase, which along with
its ligand CSF-1 (also known as macrophage colony-stimulating
factor, M-CSF) regulates the growth, differentiation and survival
of monocytes and macrophages. A recent review summarizes the
biology of CSF-1R, its role in cancer and inflammation, and efforts
to develop selective small molecule inhibitors.1
F
F
N
NH
N
CONH2
EtO
N
We recently reported 3-amido-4-anilinoquinolines as potent
and highly selective inhibitors of CSF-1R, and described the optimi-
zation of the pharmacokinetic profile of that series through the
introduction of cyclic amines.2a,b Several of these compounds, such
as AZ683 (Fig. 1) achieved good activity in a mouse pharmacody-
namic model which measured inhibition of pCSF-1R. AZ683 also
demonstrated efficacy and reduced the level of tumor-associated
macrophages (TAMs) in a breast cancer xenograft model.3
We now report the identification of a new class of CSF-1R inhib-
itors, the 3-amido-4-anilinocinnolines. In contrast to quinazolines
and quinolines, kinase inhibitors incorporating a cinnoline scaffold
are relatively rare. One reported use involved 4-anilinocinnolines,
prepared as part of the program that ultimately delivered the dual
VEGFR/EGFR anilinoquinazoline inhibitor ZD6474 (vandetanib).4 In
contrast, an earlier report on EGFR inhibitors found the cinnoline
analogue of a potent anilinoquinazoline to be inactive.5 4-Anilinoc-
innolines and benzylideneamino cinnolines were described in an
application for PLK1 inhibitors.6 3-Amido-4-anilinocinnolines have
also been reported as inhibitors of PDE4.7
Figure 1. 3-Amidoquinoline AZ683.
which was trapped with pyrrolidine to give 2.8 Reaction of 2 with
diethyl carbonate gave 3, isolated as its sodium salt, with cycliza-
tion in TFA to give the dihydrocinnoline ester 4. The remaining
steps were similar to those employed in the synthesis of the
3-amido-4-anilinoquinolines—chlorination with POCl3 to give 5,
aniline displacement, and conversion of the esters 6 to the amides
7a–e with formamide and NaOMe. Yields are given for the synthe-
sis of difluoro compound 7a; comparable yields were achieved
with other anilines. Compounds from this scaffold demonstrated
excellent activity in both our enzyme (data not shown) and cell
proliferation assays;9 typically about twofold better in the cell than
the corresponding quinoline analogues (Table 1).
Aqueous solubility and plasma protein binding data for com-
pounds 7a–e are reported in Table 1. With the objective of improv-
ing the physical properties of the series, we next introduced cyclic
amines at the 6-position of the cinnoline scaffold. Our initial route
to the N-methylpiperazino compounds was similar to that used for
the 6,7-dimethoxycinnolines, but ran into difficulties at two points
(Scheme 2). First, the acylation of the 4-bromo-3-ethoxyaniline 8
to prepare the required aminoketone 9 was relatively low yielding
(42%).10 More surprisingly, the Buchwald–Hartwig coupling step to
introduce N-methylpiperazine onto bromoester 10 gave some of
the desired ester 11 (ꢀ30% isolated material at best, which could
To establish whether cinnolines were active against CSF-1R, we
prepared examples 7a–e in the 6,7-dimethoxycinnoline scaffold
(Scheme 1). Treatment of 2-amino-4,5-dimethoxyacetophenone 1
with concd HCl and sodium nitrite generated the diazonium salt,
⇑
Corresponding author.
E-mail address: david.scott@astrazeneca.com (D.A. Scott).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.01.033

D. A. Scott et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1382–1384
1383
O
O
O
N
O
N
O
N
MeO
MeO
MeO
MeO
MeO
MeO
MeO
MeO
CO2Et
a)
c)
b)
OEt
N
N
N
NH2
N
H
N
1
2
3
4
R1
R1
Cl
NH
N
NH
MeO
MeO
CO2Et
e)
f)
d)
MeO
CO2Et
MeO
CONH2
N
N
N
N
MeO
MeO
N
5
6
7a-e
Scheme 1. Reagents and conditions: (a) concd HCl, NaNO2, H2O, 0 °C then pyrrolidine, 0.4 N aq NaOH, 85%; (b) CO(OEt)2, NaH, THF, reflux; (c) TFA, 0 °C, 50% over two steps;
(d) POCl3, reflux, 88%; (e) aniline, EtOH, cat. AcOH, 75 °C, 88%; (f) NH2CHO, 0.5 M NaOMe in MeOH, DMF, 100 °C, 96%.
Table 1
Cell potency and physical property data for cinnolines 7a–e
Compound
R1
Cell (
lM)
Quinoline potency cell2a
(l
M)
Sol (l
M)
PPB (% free)
7a
7b
7c
7d
7e
2,4-F
0.16
0.11
0.20
0.08
0.05
0.40
0.25
0.21
0.12
0.09
3
4.5
3.3
3.1
2.2
1.2
2-F,4-Me
2-F,5-Me
2-F,3-Cl
2,3-Cl
<1
<1
<1
<1
Steps a) b),
c), d), e) from
Scheme 1
R
N
R
R
O
N
NH
N
O
NH
N
N
NH
N
Br
Br
a)
b)
Br
CO2Et
N
CO2Et
N
+
EtO
NH2
EtO
NH2
N
N
N
N
EtO
EtO
EtO
8
9
10
11
12
Scheme 2. Reagents and conditions: (a) 1.0 M BCl3 in DCM, CH3CN, 1.0 M TiCl4 in DCM, reflux, 42%; (b) N-methylpiperazine, Pd2(dba)3, BINAP, Cs2CO3, DMA, 90 °C.
then be converted to the primary amide) but also amide 12 as the
major product. This contrasted with our synthesis of the corre-
sponding quinolines, in which a variety of amines were routinely
introduced at the 6-position of the quinoline esters.2b
An alternative synthetic route avoided aminoketone 9, and
enabled amine couplings to the bromocinnoline nitrile 18, rather
than the ester 10 (Scheme 3).11 Alkoxybromoaniline 13 in a
suspension of concd HCl was treated with sodium nitrite, and then
2-cyanoacetamide and aqueous sodium acetate to give intermedi-
ate 14. Cyclization to the 3-amido-4-aminocinnoline 15 was
performed with TiCl4 in refluxing toluene, followed by hydrolysis
to the hydroxy acid 16. Hydroxy acid 16 was converted to chloroa-
mide 17 with thionyl chloride and then aqueous ammonia in
acetone. Aniline displacement and amide dehydration with phos-
phorus oxychloride gave bromocinnoline nitrile 18. These two
steps could also be reversed, to introduce the required aniline
slightly later in the sequence. Buchwald–Hartwig couplings intro-
duced N-methylpiperazine to give 19, and hydrolysis of the nitrile
under basic conditions gave final compounds 20a–i. Yields for
Scheme 3 are for 20g, but comparable yields were achieved for
other anilines and the 7-EtO compounds.
Cell potency and physical property data for compounds 20a–i
are shown in Table 2. The 7-ethoxy-6-N-methylpiperazinocinno-
lines 20a–e were more potent than both the corresponding
NH2
OH
N
Br
Br
Br
R1
CONH2
Br
R1
CO2H
a)
c)
b)
CN
R1
N
N
H
N
N
R1
NH2
N
CONH2
14
13
15
16
R2
R2
R2
Cl
NH
N
NH
N
N
NH
N
Br
R1
CONH2
d)
e), f)
g)
h)
CN
N
CN
N
CONH2
Br
R1
N
N
or f), e)
N
N
N
R1
R1
N
17
18
20a-i
19
Scheme 3. Reagents and conditions: (a) concd HCl, aq NaNO2, 0 °C then NCCH2CONH2, aq NaOAc, rt, 94%; (b) TiCl4, toluene, reflux, 87%; (c) KOH, EtOH/H2O, reflux, 86%; (d)
SOCl2, cat. DMF, reflux, then acetone, 0 °C, aq NH3, 94%; (e) aniline, EtOH, cat. AcOH, 75 °C; (f) POCl3, Et3N, DCM, reflux, 43–64% over two steps; (g) N-methylpiperazine,
Pd2(dba)3, BINAP, Cs2CO3 (or XPHOS, NatBuO), DMA, reflux, 69%; (h) KOH, tBuOH, reflux, 49%.

1384
D. A. Scott et al. / Bioorg. Med. Chem. Lett. 21 (2011) 1382–1384
Table 2
Cell potency and physical property data for cinnolines 20a–i
Compound
R1
R2
Cell (l
M)
Quinoline potency
Sol (lM)
PPB (% free)
hERG (lM)
cell2b
(lM)
20a
20b
20c
20d
20e
20f
20g
20h
20i
EtO
EtO
EtO
EtO
2,4-F
0.048
0.013
0.040
0.026
0.014
0.048
0.025
0.057
0.021
0.23
0.05
0.09
0.11
0.06
1.03
—
8
27
120
18
10
290
87
3.9
1.6
11
1.9
<1.7
9.6
6.3
7.1
7.8
16
13
—
12
5
20
>30
18
10
2-F,4-Me
2-F,5-Me
2-F,3-Cl
2,3-Cl
EtO
MeO
MeO
MeO
MeO
2,4-F
2-F,4-Me
2-F,5-Me
2-F,3-Cl
—
—
22
110
N-methylpiperazine at the 6-position delivers compounds with
good physical properties, and an excellent PK profile was observed
for 20b in multiple species. Examples in both the 6,7-dimethoxy
and 7-alkoxy-6-N-methylpiperazino scaffolds demonstrated en-
hanced cellular potency relative to the corresponding quinoline
compounds. Cinnoline 20a also showed an excellent kinase selec-
tivity profile. Our efforts to develop further CSF-1R inhibitors from
the cinnoline class will be reported in due course.
Table 3
Mouse, rat and dog PK data for 20b
Species
Cl (ml/min/kg)
T1/2 (h)
Vss (L/kg)
F (%)
Mouse
Rat
Dog
12
10
2
0.9
9.7
12.6
0.6
2.8
1.8
82
90
78
IV dosing at 3 mpk; oral dosing at 10 mpk for mice, rats, 5 mpk for dogs.
Table 4
Acknowledgments
Kinase selectivity of 20a at 1 mM and 0.1 mM
% Inhibition at 1
lM
% Inhibition at 0.1 lM
We thank Heather Blanchette and Dominique Custeau for PK
studies, and Michael Block, Paul Lyne and Scott Williams for assis-
tance with the manuscript.
CSF-1R
cKit
ARK5
95
90
78
CSF-1R
cKit
ARK5
100
35
23
7
Rsk1
44
Rsk1
MELK
78 Kinases
41
<35
MELK
78 Kinases
15
<30
References and notes
1. Patel, S.; Player, M. R. Curr. Top. Med. Chem. 2009, 9, 599.
2. (a) Scott, D. A.; Balliet, C. L.; Cook, D. J.; Davies, A. M.; Gero, T. W.; Omer, C. A.;
Poondru, S.; Theoclitou, M.-E.; Tyurin, B.; Zinda, M. J. Bioorg. Med. Chem. Lett.
2009, 19, 697; (b) Scott, D. A.; Bell, K. J.; Campbell, C. T.; Cook, D. J.; Dakin, L. A.;
Del Valle, D. J.; Drew, L.; Gero, T. W.; Hattersley, M. M.; Omer, C. A.; Tyurin, B.;
Zheng, X. Bioorg. Med. Chem. Lett. 2009, 19, 701.
3. Drew, L.; Bell, K.; Dakin, L.; Hattersley, M.; Lawson, D.; Repik, G.; Scott, D.; Shen,
M.; Omer, C. Abstract of Papers, 100th Annual Meeting of the American
Association for Cancer Research, Denver, CO, AACR: Philadelphia, PA, 2009;
Abstract 259.
4. Hennequin, L. F.; Thomas, A. P.; Johnstone, C.; Stokes, E. S. E.; Ple, P. A.;
Lohmann, J.-J. M.; Ogilvie, D. J.; Dukes, M.; Wedge, S. R.; Curwen, J. O.; Kendrew,
J.; Lambert-van der Brempt, C. J. Med. Chem. 1999, 42, 5369.
5. Rewcastle, G. W.; Denny, W. A.; Bridges, A. J.; Zhou, H.; Cody, D. R.; McMichael,
A.; Fry, D. W. J. Med. Chem. 1995, 38, 3482.
6,7-dimethoxycinnoline compounds 7a–e and the corresponding
quinoline analogues; several different anilines gave compounds
with a cell IC50 <0.05 lM. The 7-methoxycinnolines 20f–i were
also extremely potent in our cell assay. The introduction of
N-methylpiperazine significantly improved the physical properties
of this series relative to the 6,7-dimethoxy compounds. In particu-
lar, the 7-methoxy examples 20f–i combined excellent cell potency
with good aqueous solubility and plasma protein binding data.
Compounds were screened against the hERG ion channel,12 with
a trend towards the more lipophilic examples showing increased
levels of activity.
An excellent in vivo PK profile was achieved for some examples
of this class of cinnoline. Data for 20b, upon IV and oral dosing in
mice, rats and dogs, are shown in Table 3. Low rates of clearance
were observed in all three species, accompanied by good
bioavailability.
6. Bearss, D. J.; Vankayalapati, H.; Grand, C. L. WO 06124996 A2 20061123; Chem.
Abstr. 145, 505465.
7. Lunniss, C.; Eldred, C.; Aston, N.; Craven, A.; Gohil, K.; Judkins, B.; Keeling, S.;
Ranshaw, L.; Robinson, E.; Shipley, T.; Trivedi, N. Bioorg. Med. Chem. Lett. 2010,
20, 137.
8. Dowlatshahi, H. A. Synth. Commun. 1985, 15, 1095.
9. Scott, D. A.; Aquila, B. M.; Bebernitz, G. A.; Cook, D. J.; Dakin, L. A.; Deegan, T. L.;
Hattersley, M. M.; Ioannidis, S.; Lyne, P. D.; Omer, C. A.; Ye, M.; Zheng, X. Bioorg.
Med. Chem. Lett. 2008, 18, 4794.
10. Sugasawa, T.; Toyoda, T.; Adachi, M.; Sasakura, K. J. Am. Chem. Soc. 1978, 100,
4842.
11. Dahmen, J.; Hemmerling, M.; Klingstedt, T.; Sjoe, P. WO 04016615 A1
20040226; Chem. Abstr. 140, 217650.
12. hERG activity was assessed using the Ionworks assay.
13. KinaseProfiler Department, Millipore UK Limited, Gemini Crescent, Dundee
Technology Park, Dundee DD2 1SW, UK.
We previously reported the kinase selectivity profile of the 3-
amido-4-anilinoquinoline compound AZ683.2b Cinnoline 20a was
also highly selective for CSF-1R. In a panel of 83 kinases,13 cKit
and ARK5 were the only kinases other than CSF-1R against which
it showed significant (IC50 <1 lM) levels of activity (Table 4). Com-
pound 20a had an IC50 <4 nM in our CSF-1R enzyme assay.14
In conclusion, the reported 3-amido-4-anilinocinnoline com-
pounds are very potent inhibitors of CSF-1R. The introduction of
14. Enzyme assay conditions are described in: Dakin, L.; Ogoe, C. A.; Scott, D.;
Zheng, X. WO 2008090353 A1 20080731; Chem. Abstr. 149, 224267.