Potent 2′-aminoanilide inhibitors of cFMS as potential anti-inflammatory agents

Article abstract of DOI:10.1016/j.bmcl.2007.09.057

A series of 2′-aminoanilides have been identified which exhibit potent and selective inhibitory activity against the cFMS tyrosine kinase. Initial SAR studies within this series are described which examine aroyl and amino group substitutions, as well as the introduction of hydrophilic substituents on the benzene core. Compound 47 inhibits the isolated enzyme (IC₅₀ = 0.027 μM) and blocks CSF-1-induced proliferation of bone marrow-derived macrophages (IC₅₀ = 0.11 μM) and as such, serves as a lead candidate for further optimization studies.

Full text of DOI:10.1016/j.bmcl.2007.09.057

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 17 (2007) 6070–6074
Potent 20-aminoanilide inhibitors of cFMS as potential
anti-inflammatory agents
Raymond J. Patch, Benjamin M. Brandt, Davoud Asgari, Nand Baindur,
Naresh K. Chadha, Taxiarchis Georgiadis, Wing S. Cheung, Ioanna P. Petrounia,
Robert R. Donatelli, Margery A. Chaikin and Mark R. Player*
Drug Discovery, Johnson and Johnson Pharmaceutical Research and Development, L.L.C., 8 Clarke Drive, Cranbury, NJ 08512, USA
Received 22 August 2007; revised 13 September 2007; accepted 14 September 2007
Available online 19 September 2007
Abstract—A series of 20-aminoanilides have been identified which exhibit potent and selective inhibitory activity against the cFMS
tyrosine kinase. Initial SAR studies within this series are described which examine aroyl and amino group substitutions, as well as
the introduction of hydrophilic substituents on the benzene core. Compound 47 inhibits the isolated enzyme (IC50 = 0.027lM) and
blocks CSF-1-induced proliferation of bone marrow-derived macrophages (IC50 = 0.11lM) and as such, serves as a lead candidate
for further optimization studies.
Ó 2007 Elsevier Ltd. All rights reserved.
cFMS is a type III receptor tyrosine kinase that is selec-
tively expressed on macrophages and their progenitor
cells and serves as the exclusive receptor for colony-stim-
ulating factor (CSF-1), the primary growth factor for
the macrophage lineage.1 Upon binding to CSF-1,
cFMS undergoes autophosphorylation and dimeriza-
tion, and ultimately induces the phosphorylation of
downstream signaling proteins, thereby driving the dif-
ferentiation and activation of these cells.2
collagen-induced arthritis (CIA) models, CSF-1 treat-
ment was shown to exacerbate the disease,5 whereas
anti-CSF-1 treatment resulted in a reduced severity of
disease.6 Furthermore, CSF-1 deficient (op/op) mice
were shown to be completely resistant to CIA.5
Thus, the prominent role of the CSF-1/cFMS signaling
in the activation and proliferation of macrophages
makes this transduction pathway an attractive target
for therapeutic intervention.7 We therefore sought to
identify selective inhibitors of cFMS as a potential treat-
ment for macrophage-related diseases.
The expression of macrophages is significantly induced
at sites of inflammation; this over-expression propagates
a chronic inflammatory response, which is in part due to
a CSF-1/FMS-driven mechanism.3 Thus, ligation of
cFMS by CSF-1 results in activation and proliferation
of macrophages and their subsequent release of inflam-
matory mediators. These in turn stimulate neighboring
fibroblasts to release additional CSF-1, thereby perpetu-
ating an inflammatory cycle. Inhibitors of cFMS would
represent a likely means of interrupting this cycle and
would therefore be expected to result in an anti-inflam-
matory response.4 Support for this hypothesis is pro-
vided by various animal studies. For example, in
Amongst the active chemotypes identified from a
focused screening effort, aroylanilide 1 was chosen for
further evaluation and hit-to-lead optimization. This re-
port describes initial structure–activity relationships
(SAR) within this series, which led to the identification
of an advanced analogue that proved to be suitable
for X-ray crystallographic studies (Fig. 1).
It was apparent from the primary screening set that an
aroyl-capped ortho-amino anilide structure represented
the basic pharmacophoric unit for cFMS inhibition.
Additionally, a preliminary examination of aroyl groups
revealed a preference for N- and O-containing heterocy-
cles. Moreover, nitroaromatics generally had greater
affinity for cFMS than their unsubstituted counterparts.
Of these, the 5-nitrofuroyl moiety was found to afford a
Keywords: cFMS; Colony-stimulating factor-1 receptor; CSF-1;
M-CSF; 20-Aminoanilides.
*
Corresponding author. Tel.: +1 610 458 6980; e-mail: mplayer@
prdus.jnj.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.09.057

R. J. Patch et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6070–6074
6071
N
Ar
NH
O
O
a, b, c
O
NO2
F
X
X
R1
NH
N
IC50= 0.4 μm
N
R2
Scheme 1. Synthesis of 20-aminoanilides. Reagents and conditions: (a)
HNR1R2, PS-morpholine, CHCl3, 50°C, 2 h; (b) H2, Pd/C, HOAc/
MeOH/EtOAc (1 atm.); (c) ArCOCl, PS-morpholine, dioxane, 70°C,
2 h.
1
Figure 1. High-throughput screening hit: aroylanilide 1.
10-fold improvement in inhibitory potency relative to
isoxazole 1. While we recognized the potential metabolic
and genotoxic liability introduced by this functionality,8
at this stage of the project, it served as a useful control
for assessing SAR of the 2-amino substituent. Our initial
follow-up strategy was therefore based on a two-point
diversity scan against a broader array of aroyl and ami-
no groups (Fig. 2).
Table 1. cFMS inhibitory activities for 20-amino-5-nitrofuroylanilides
2–21
NO2
O
O
NH
R1
2-21
N
R2
20-Aminoanilides were readily prepared by a standard
SNAr reaction of a 2-fluoronitrobenzene derivative with
various amines, followed by nitro group reduction and
capping with an aroyl chloride (Scheme 1).9
Compound NR1R2
Auto-Pi HEK
a
b
IC50
(lM)
293 IC50
(lM)
2
3
Piperidino
Morpholino
Piperazine
0.053
1.1
0.36
Compounds were initially assessed in an in vitro enzyme
assay for their abilities to inhibit ATP-induced auto-
phosphorylation of cFMS (Table 1).10 The importance
of maintaining an aliphatic cyclic amine structure at
the 2-position was apparent in that replacement of
piperidine with acyclic amines (5–7) led to a dramatic
loss of cFMS inhibitory activity.11 Ring size also af-
fected activity, with optimal inhibition being derived
from the 6-membered piperidine system of 2, and
decreasing with smaller (8, 9) and larger (10) ring sys-
tems. Activity was also diminished by the incorporation
of heteroatoms in the ring, as was observed with mor-
pholine and piperazine analogues, 3 and 4, respectively.
When evaluated in a cellular kinase assay,12 inhibitory
activities largely paralleled those obtained in the isolated
enzyme system, being roughly 5- to 10-fold less active.
4
>10
>10
>10
>10
>10
1.9
5
N-Ethyl-N-propylamino
N,N-Dipropylamino
Anilino
6
7
8
Azetidino
Pyrrolidino
Azepino
9
10
11
12
13
14
15
16
17
18
19
20
21
0.26
0.72
0.38
5.3
2.0
2.2
2-Methylpiperidino
3-Methylpiperidino
3,5-Dimethylpiperidino
4-Methylpiperidino
4-Methylpiperazino
4-Phenylpiperazino
4-Hydroxypiperidino
0.078
0.39
>10
0.14
0.13
0.3
1.4
0.37
1.4
4-Hydroxymethylpiperidino
4-(2-Hydroxyethyl)piperidino 0.043
3-Hydroxypyrrolidino >10
3-Hydroxymethylpyrrolidino >10
0.23
a Reported IC50 values are means of three experiments. Inter-assay
variance was <10%.
Substituent effects appeared to be largely influenced by
steric factors, being sensitive to both size and ring posi-
tion. Thus, incorporation of small substituents, such as
methyl, (14), hydroxyl (17), hydroxymethyl (18), and
hydroxyethyl (19), were best tolerated at the 4-position
of the piperidine ring. In contrast, hydroxy and
hydroxymethyl substituents rendered pyrrolidino homo-
logue 9 inactive (20 and 21, respectively). Whereas
methyl substitution was tolerated on the 4-piperazinyl
nitrogen (15), activity was lost upon phenyl substitution
b Dose–response data are the average of at least two replicates per
dose.
(16). Methyl group incorporation at the 3-position of
the piperidine ring was also tolerated (12), though to a
lesser extent than at the 4-position. Appending an addi-
tional methyl group at the 5-position further attenuated
inhibitory activity (13).
Stability studies with microsomal preparations revealed
a significant metabolic liability with early series ana-
logues. From metabolite ID studies, it was determined
that the 4-position of the piperidine ring was susceptible
to microsomal oxidation. However, this liability could
be overcome simply by methylation (as in 14), a tactic
that would extend to later series analogues as well.
For example, following a 10-min incubation period with
Ar
NO2
+
O
O
Ar
NH
O
O
NH
NH
R1
R1
R2
N
N
N
R2
Figure 2. Two-point diversity strategy.

6072
R. J. Patch et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6070–6074
human liver microsomes (HLMs), 41% of piperidinyl 2
remained intact; methylpiperidine 14 was completely
stable to these conditions (100% remaining).
An expanded search for additional heterocyclic replace-
ments led to the identification of nitropyrroloyl (33) and
cyanopyrroloyl (34) analogues as particularly potent
inhibitors of cFMS activity. Other isomeric azoles, how-
ever, proved much less active in these assays, demon-
strating the sensitivity of heteroatom positioning in
this ring (data not shown).
A more directed study was undertaken to identify suit-
able heteroaroyl groups that could replace the nitrofu-
royl system. At this time, the mechanism(s) by which
the nitro group enhanced inhibitory activities in this ser-
ies was unclear. And so, in order to determine whether
this might be due to a specific favorable protein interac-
tion or indirectly by way of favorable electronic modu-
lation of the aroyl rings, a number of furan derivatives
were evaluated (Table 2).
Examination of substituent effects on the benzene ring
revealed a range of functionality that could be accom-
modated at the 5-position (Table 3). With the excep-
tion of a carboxyl group (50), polar substituents
were particularly adventitious, not only affording in-
creased inhibitory potency in several instances, but
also providing a solubilizing handle for this chemo-
type. Exceptions to this were underivatized phenol
35 and guanidinylmethyl 46, whose cellular activities
were ꢀ100-fold less than their activities against the
isolated enzyme, possibly a reflection of limited cellu-
lar uptake. Hydroxy-containing analogues 37, 44,
and 47 were amongst the most effective inhibitors of
cFMS kinase activity. Again, the 4-methyl substituent
in 47 imparted significant microsomal stability (83%
remaining) to an otherwise metabolically susceptible
analogue, 37 (25% remaining after 10-min incubation
with HLMs).
From this study, it would appear that both electronic
and steric factors contribute to the activity-enhancing
properties of the nitro group. The presence of electron
withdrawing groups on the furan provided improved
activities to various degrees as compared with unsubsti-
tuted 22. In contrast, aminofuroyl analogues 23 and 24
were completely inactive. The diminished activity of car-
boxy analogue 29 relative to 2 may reflect a lack of
involvement of a specific charge interaction between
the 5-substituent and the enzyme. At first glance, this
might imply that electronic effects of the substituents ex-
ert a predominating influence on activity; however, the
inactivity of trifluoromethyl derivative 31 suggests that
this property alone does not suffice. Sterics and hybrid-
ization likely also contribute to adventitious substituent
effects, a notion that is supported by the improved activ-
ity of cyanofuroylanilide 32.
Kinase selectivity of lead analogue 47 was assessed
against a standard diverse panel of kinases (Table 4).
At a concentration of 1lM, only the neurotrophic tyro-
sine kinase receptor, type 1 (TRKA) was inhibited at
above 50%.
In order to evaluate the functional activities of cFMS
inhibitors, an assay was developed based upon CSF-1
driven proliferation of bone marrow-derived macro-
phages.13 Following deprivation of CSF-1, cultured
macrophages derived from mouse bone marrow can be
driven into S-phase upon re-stimulation with CSF-1.
Bromodeoxyuridine (BrDU) is incorporated into DNA
of S-phase cells and can be quantitated in an ELISA for-
mat.14 In this assay, 47 inhibited CSF-1 induced incor-
poration of BrDU into mouse macrophages with an
IC50 of 0.11 lM.
Table 2. Aroyl ring substituent effects upon cFMS binding/inhibitory
activities for 20-(piperidin-1-yl)anilides 2, 22–34
HN
R
R
O
O
O
NH
N
2, 22-32
NH
N
33, 34
a
Compound
R
Auto-Pi IC50
(lM)
HEK 293
b
IC50 (lM)
In addition to its lead characteristics, 47 aided in the
crystallization of a chimeric kinase domain of cFMS,
the co-crystal structure of which was recently re-
ported.15 The interaction of 47 with cFMS occurs
in the hinge region of the ATP pocket and involves
a critical hydrogen bond between the aroylamide car-
bonyl and the backbone NH of Cys666 (Fig. 3).
Interestingly, the ring oxygen of the furan is not in-
volved in direct binding to a specific cFMS residue;
rather, it shares an intramolecular hydrogen bond
with the amide NH, thereby stabilizing a flat confor-
mation of the aroylamide core. The methylpiperidine
occupies the ATP sugar pocket; the hydroxymethyl
extends away from this region, making a hydrogen
bond with the phenol of Tyr665. Finally, this crystal-
lographic information has provided the structural ba-
sis for further lead optimization studies within this
series.
22
2
H
>10
0.053
>10
>10
>10
4.4
NO2
NH2
NHPh
Ph
0.36
23
24
25
26
27
28
29
30
31
32
33
34
Br
Cl
1.9
1.1
CHO
CO2H
CONH2
CF3
>10
>10
>10
0.036
0.047
0.047
CN
0.37
0.18
0.15
NO2
CN
a Reported IC50 values are means of three experiments. Inter-assay
variance was <10%.
b Dose–response data are the average of at least two replicates per
dose.

R. J. Patch et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6070–6074
Table 3. Benzene ring substituent effects for 20-(piperidin-1-yl)anilides 35–47
6073
CN
O
O
X
NH
N
35-47
R
a
Compound
X
R
Auto-Pi IC50
(lM)
HEK 293
b
IC50 (lM)
35
36
37
38
39
40
41
42
43
44
45
46
47
HO
MeO
H
H
H
H
H
H
H
H
H
H
H
H
Me
0.017
0.09
2.7
0.96
0.19
3.1
Hydroxymethyl
Aminomethyl
CHO
0.025
0.2
0.037
>2
0.12
0.33
>2
CO2H
CN
0.75
0.076
1.01
0.17
0.31
2.6
Piperidin-1-yl
4-Methylpiperazinomethyl
0.017
0.08
2,3-Dihydroxypropoxymethyl
Methylsulfonamidomethyl
Guanidinylmethyl
0.019
0.037
0.034
0.024
Hydroxymethyl
0.25
a Reported IC50 values are means of three experiments. Inter-assay variance was <10%.
b Dose–response data are the average of at least two replicates per dose.
Table 4. Kinase selectivity screen for compound 47
Kinase
% Inha
Abl
Akt1
ꢁ15
0
Arg
BTK
Blk
ꢁ15
9
ꢁ4
ꢁ3
2
CDK1/cyclinB
CDK2/cyclinA
CHK1
CaMKIIa
EphB4
FES
2
0
ꢁ2
1
Fgr
Flt3
2
25
10
ꢁ9
0
Fyn
GSK3b
IGF-1R
IRAK4
Lck
1
5
PKA
ꢁ18
0
PKCh
Rsk2
Src
ꢁ1
12
54
5
TRKA
Yes
a Compound 47 was tested at single concentration of 1 lM; ATP
concentration was 100 lM. Values are averages of duplicate
experiments.
Figure 3. Cut-away illustration of the complex between cFMS (teal)
and furoylanilide 47 (gray). Hydrogen bonds (blue dashes) are shown
along with interatomic distances between heavy atoms.
References and notes
3. (a) Yanni, G.; Whelan, A.; Feighery, C.; Bresnihan, B.
Ann. Rheum. Dis. 1994, 53, 39; (b) Yang, N.; Isbel, N. M.;
Nikolic-Paterson, D. J.; Li, Y.; Ye, R.; Atkins, R. C.; Lan,
H. Y. Kidney Int. 1998, 54, 143; (c) Isbel, N. M.; Nikolic-
1. Sherr, C. J. Biochim. Biophys. Acta 1988, 948, 225.
2. (a) Pixley, F. J.; Stanley, E. R. Trends Cell Biol. 2004, 14,
628; (b) Sherr, C. J.; Roussel, M. F.; Rettenmier, C. W.
J. Cell Biochem. 1988, 38, 179.

6074
R. J. Patch et al. / Bioorg. Med. Chem. Lett. 17 (2007) 6070–6074
Patterson, D. J.; Hill, P. A.; Dowling, J.; Atkins, R. C.
Nephrol. Dial. Transplant. 2001, 16, 1638.
11. That this cyclic amine should be aliphatic was
determined from the initial screening set data. For
example, the N-pyrrolo analogue of 1 was found to
be inactive.
12. HEK293 phosphorylation assay Baumann, C. A.; Zeng, L.;
Donatelli, R. R.; Maroney, A. C. J. Biochem. Biophys.
Methods 2004, 60, 69.
13. Kaye, P. M. In Macrophages; Paulnock, D. M., Ed.;
Oxford University Press: Oxford, 2000; pp 93–113.
14. BMDM Proliferation Assay. Bone marrow-derived mac-
rophages (BMDM) were plated at a density of 5000 cells/
well into Costar 96-well microtiter plates and incubated
overnight in macrophage culture media without CSF-1.
The following day, wells were adjusted to contain 5 ng/ml
CSF-1, and graded concentrations of 47. After a 24-h
culture period, wells were further adjusted to contain
bromodeoxyuridine (BrDU) for additional 6 h. The incor-
poration of BrDU into the DNA of proliferating macro-
phages was quantified using a specific ELISA (Exalpha
Corp. Watertown, MA) and concentrations of 47 that
inhibited BrDU incorporation by fifty percent (IC50) were
calculated using GraphPad PrismÒ software and a four
parameter logistics equation.
4. (a) Hamilton, H. A. Trends Immunol. 2002, 23, 403; (b)
Conway, J. G.; McDonald, B.; Parham, J.; Keith, B.;
Rusnak, D. W.; Shaw, E.; Jansen, M.; Lin, P.; Payne, A.;
Crosby, R. M.; Johnson, J. H.; Frick, L.; Lin, M.-H. J.;
Depee, S.; Tadepalli, S.; Votta, B.; James, I.; Fuller, K.;
Chambers, T. J.; Kull, F. C.; Chamberlain, S. D.; Hutchins,
J. T. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 16078.
5. Campbell, I. K.; Rich, M. J.; Bischof, R. J.; Hamilton, J.
A. J. Leukocyte Biol. 2000, 68, 144.
6. Cook, A. D.; Braine, E. L.; Campbell, I. K.; Rich, M. J.;
Hamilton, J. A. Arthritis Res. 2001, 3, 293.
7. Lawlor, K. E.; Campbell, I. K.; Metcalf, D.; O’Donnell,
K.; van Nieuwenhuijze, A.; Roberts, A. W.; Wicks, I. P.
Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 11398.
8. (a) Jung, R.; Le, J. Y.; Wengenmayer, F.; Wolf, E.;
Kramer, M. Biomed. Biochim. Acta 1985, 44, 485; (b)
McCalla, D. R. Environ. Mutagen. 1983, 5, 745; (c)
Schmid, A. Tierarztl. Prax. 1985, 13, 471.
9. Polymer-bound
obtained from Sigma–Aldrich Chemical Co.
morpholine
(PS-morpholine)
was
10. cFMS autophosphorylation assay Schalk-Hihi, C.; Ma, H.
C.; Struble, G. T.; Bayoumy, S.; Williams, R.; Devine, E.;
Petrounia, I. P.; Mezzasalma, T.; Zeng, L.; Schubert, C.;
Grasberger, B.; Springer, B. A.; Deckman, I. C. J. Biol.
Chem. 2007, 282, 4085.
15. Schubert, C.; Schalk-Hihi, C.; Struble, G. T.; Ma, H. C.;
Petrounia, I. P.; Brandt, B.; Deckman, I. C.; Patch, R. J.;
Player, M. R.; Spurlino, J. C.; Springer, B. A. J. Biol.
Chem. 2007, 282, 4094, PDB: 2I0Y.