Identification of a sulfonamide series of CCR2 antagonists

Article abstract of DOI:10.1016/j.bmcl.2010.04.142

A series of sulfonamide CCR2 antagonists was identified by high-throughput screening. Management of molecular weight and physical properties, in particular moderation of lipophilicity and study of pK_a, yielded highly potent CCR2 antagonists exh

Full text of DOI:10.1016/j.bmcl.2010.04.142

Bioorganic & Medicinal Chemistry Letters 20 (2010) 3961–3964
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification of a sulfonamide series of CCR2 antagonists
b
b
a
a
c
c
Simon Peace a, , Joanne Philp , Carl Brooks , Val Piercy , Kitty Moores , Chris Smethurst , Steve Watson ,
*
Simon Gaines a, Mara Zippoli a, Claudette Mookherjee c, Robert Ife a
a GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage SG1 2NY, UK
b GlaxoSmithKline, Upper Merion, 709 Swedeland Road, King of Prussia, PA 19406, USA
c GlaxoSmithKline, New Frontiers Science Park, Harlow CM19 5MW, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of sulfonamide CCR2 antagonists was identified by high-throughput screening. Management of
molecular weight and physical properties, in particular moderation of lipophilicity and study of pKa,
yielded highly potent CCR2 antagonists exhibiting good pharmacokinetic properties and improved
potency in the presence of human plasma.
Received 26 March 2010
Revised 29 April 2010
Accepted 29 April 2010
Available online 7 May 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
CCR2 antagonists
pKa
Sulfonamides
O
Chemokines are a family of endogenous pro-inflammatory pro-
teins which act primarily via the activation and recruitment of leu-
kocytes yet the inappropriate over-expression of such proteins is
implicated in a variety of disease conditions.1 CCL2, commonly re-
ferred to as monocyte chemoattractant protein 1 (MCP-1), acti-
vates several molecular targets including the G-protein coupled
seven-transmembrane receptor CCR2. This ligand/receptor pair is
overexpressed in numerous inflammatory conditions wherein
excessive monocyte recruitment is observed, such as atherosclero-
sis2 and rheumatoid arthritis.3
It is worth noting that while the anti-CCR2 monoclonal anti-
body ML-1202 was effective in phase II trials for multiple sclerosis
it exhibited no benefit in patients with rheumatoid arthritis.4
A host of patents and peer-reviewed publications have appeared
which describe the development of small-molecule CCR2 antago-
nists.4 Quaternary ammonium CCR5 antagonist TAK-7795 (1,
Fig. 1) interacts with CCR2 and several series containing a non-qua-
ternary basic center have been reported, for example, 26 and 37
(Merck) and 48 (BMS). Acidic CCR2 antagonists such as 5, described
by researchers at AstraZeneca, have also appeared.9 Herein we de-
scribe the discovery of an alternative class of compounds that does
not require a strongly ionizable group for activity.
CF3
H
N
NH2
O
H
N
MeS
N
H
NH
O
N+
O
4
1
O
F
O
R:
N
N
OH
O
CF3
2
R
N
H
H
N
CF3
Cl
Cl
O
3
5
Figure 1. Selected literature CCR2 antagonists.
N-Me inactive
Sulfonamide
critical for activity
O
O
Amides weak/
inactive
O
OH
S
HN
Lipophilic
aryl prefered
O
A series of biaryl sulfonamides, exemplified by compound 6,
emerged from a high-throughput screen of the GSK collection.
The preliminary SAR observed by early arrays are described in
Figure 2. While potency strongly correlated with lipophilicity,
Cl
Lipophilic groups
tolerated, polar
heteroaryl weak
CF3
6
4-lipophilic groups
preferred
* Corresponding author. Tel.: +44 1438768573.
E-mail address: Simon.2.Peace@gsk.com (S. Peace).
Figure 2. SAR from preliminary lead identification activities.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.04.142

3962
S. Peace et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3961–3964
Table 1
Table 2
Preliminary selectivity and chemotaxis data. CCR2, CCR1 and murine CCR2 conducted
in recombinant GTPgS (n P 6). Compounds 6 and 7 also reported below the lower
limit of the CCR1 assay (data not shown)
Representative central ring SAR
O
S
O
O
OH
HN
O
O
R1 R2
R3
CF3
H
CF3
O
O
OH
S
Cl
HN
6
H
H
R4
R3
R1
7
8
H
H
H
F
O
R1
Cl
R2
9
F
H
CF3
R2
Mouse
R3
Compound R1
R2
R3
R4 CCR2 fpKi CCR1 fpKi CTX 10
lM
Compound CCR2
CTX fpKi
CCR1
Spec. pKa
CO2H NHSO2
antagonist
%I at 1% HS
fpKi
fpKi
CCR2 fpKi
1%
1%
7
14
15
16
17
6.6
<5.5
7.2
7.4
7.7
<5.5
6.4
6.2
6.5
—
6.0
6.6
6.8
<5.5
6.3
6.2
—
—
—
52
65
—
—
—
—
FBS
HS
SO2Me
CN
F
6
7
8
9
6.1
6.6
7.5
6.9
—
—
6.3
6.5
6.9
6.9
5.8
5.7
6.6
6.0
3.2
3.3
—
7.9
8.7
—
6.4
7.2
6.8
6.4
4.9
5.0
Cl
2.4
7.7
18
19
20
21
Me
Cl
Me
F
<5.5
<5.5
examination of four closely related analogs (Table 1) provided
encouragement that activity could be modulated independently
of log P. Inclusion of fluorine adjacent either to the carboxylic acid
or biaryl ether increased potency, possibly as a consequence of pKa
modulation. Compounds 8 and 9 demonstrated encouraging po-
tency in monocyte chemotaxis (CTX)10; however, replacement of
fetal bovine serum (FBS) with human serum (HS) resulted in a dra-
matic reduction in potency. Conversely, the undecorated analog 7
maintained activity, suggesting that protein binding could be man-
aged. The general selectivity for this series was acceptable
although most analogs also interacted with CCR1.
2 summarizes some of the modifications realized in the central re-
gion of the molecule. Substitution at R3 with fluorine, chlorine or
nitrile conferred a 10-fold increase in potency and a modest in-
crease in selectivity over CCR1. Inclusion of a methyl group at R1
was not tolerated whereas inclusion of a chlorine substituent
was. Conversely addition of a weakly donating methyl group at
R2 proved inconsequential. Replacement of hydrogen with fluorine
at R4 (21) resulted in a significant loss of activity.
Selected modifications of the sulfonamide group are presented
in Table 3. Combination of the 3 chloro and 4-chorlo derivatives
provided the optimal 3,4-dichloro benzene sulfonamide 24. 5-
Chloro thienyl sulfonamide 25 was also very active although it is
worth nothing that the sulfonamide pKa was reduced (compare
25 and 17). Methane sulfonamide 26 was weakly active and larger
alkyl sulfonamides were only weakly active. Inhibition of chemo-
The risk that the observed in vitro activity resulted from non-
specific binding to proteins (CCR2 or CCL2) was discharged by
addition of detergent to the primary assay11 and assessment of
promiscuity using b-lactamase.12
The compounds described herein were prepared according to
the general Scheme 1.13 Synthesis of the previously unreported ni-
tro-pyridine 11 was achieved via dehydration of the primary amide
derived from 10, prepared by nitration of 2-hydroxy 5-carboxy
pyridine. Core 2-halo nitro arenes and aromatic alcohols were cou-
pled under basic conditions. Reduction of the nitro group was
achieved using standard hydrogenation protocols or tin chloride.
We discovered that transfer hydrogenation with ammonium for-
mate over platinum on carbon provided a reliable, facile method
for nitro group reduction without disturbing vulnerable chlorines,
nitriles and pyridines. For example, reduction of 12–13 was
achieved in 83% yield with simple filtration providing the product
in P90% purity. Standard DMAP/pyridine conditions proved to be
optimal for coupling the resulting amines with sulfonyl chlorides.
The medicinal chemistry effort focused on increasing potency
while balancing lipophilicity and limiting molecular weight. Table
taxis by 1 lM 24 in the presence of 0.1 and 1% human serum
was 54% and 21%, respectively. This confirmed earlier observations
of a significant serum-protein related effect. A strategy of reducing
lipophilicity and managing pKa was adopted to modulate protein
binding.
Polar core motifs such as pyrazine and pyrimidine (27 and 28,
Table 4) were 10-fold weaker than the related pyridines 29 and
30. Installation of chlorine para to the biaryl ether resulted in iden-
tification of compound 31. Compounds 30 and 31 possessed simi-
lar chemotaxis activity; however, the shift for compound 31
appears to be greater, possibly a consequence of reduced sulfon-
amide pKa.
Investigations of the biaryl ether are summarized in Table 5.
Compounds 31–34 revealed that the position of the acid was not
SO2Ar
NH2
HN
NO2
R
R = CO2Me
R = CO2H
R
R
O
R
O
OH
O
(ii) or (iii)
(iv)
(i)
(v)
R1
R1
R1
NO2
F, Cl
R1
NO2
NH2
NO2
NO2
HO
HO
O
Cl
(iii)
O
(vi)
(vii)
(viii)
N
N
N
OH
N
OH
N
N
N
Cl
CN
Cl
O
O
12
10
11
13
Scheme 1. Typical reagents and conditions. (i) DMF, K2CO3, 70 °C; (ii) SnCl2 or H2, Pd/C or H2, Pt/C; (iii) Pt/C, NH4ÁHCO2, DCM, reflux (83% for 13); (iv) ArSO2Cl, DMAP, pyridine
95 °C; (v) LiOH, MeOH/H2O 1:1, 30 °C; (vi) H2SO4, HNO3 (79%); (vii) CDI, NH3, 60 °C; (viii) POCl3, reflux (59%, two steps).

S. Peace et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3961–3964
3963
Table 3
Selected sulfonamide SAR
O
O
O
OH
S
HN
R
O
Cl
Compound
R
CCR2 fpKi
pKa
Compound
R
CCR2 fpKi
pKa
NHSO2
NHSO2
Cl
22
6.8
—
24
8.5
7.89
Cl
Cl
S
17
23
7.7
8.1
8.09
—
25
26
8.3
6.4
7.42
—
Cl
Cl
CH3
relevant for potency furthermore polar amides such as 35 were tol-
erated. The 3-pyridyl group introduced a weakly basic moiety thus
maintaining solubility in the absence of the carboxylic acid and the
2-methyl derivative 37 possessed the best anti-chemotaxis activity
in the presence of increasing human serum. The retention of activ-
ity in the presence of plasma proteins14 may be related to the pKa
of the sulfonamide rather than lipophilicity (e.g., compare 36
and 37).
The increase in sulfonamide pKa resulting from the installation
of the 2-methyl group was confirmed by repeated testing and may
be due to conformational change in the biaryl ether Inclusion of cy-
ano in the central ring (42) led to a reduction in chemotaxis activ-
ity, possibly due to reduced sulfonamide pKa.
Table 6 summarizes the pharmacokinetic and selectivity data
on a selection of analogs. The acidic compounds 24 and 31 both
displayed excellent adsorption and half-life typical of protein-
bound carboxylic acids.
Amide 35 displayed modest bioavailability as a consequence of
increased clearance whereas the 2-methyl pyridyl analog 37 exhib-
ited a very promising profile upon oral dosing. The series proved to
be generally selective in hERG and p450 assays although the non-
acid analogs interfered moderately with 2C9. In general the series
also interacts with the CCR1 receptor although dual activity was
Table 4
Aza 6-membered ring derivatives. CH unless otherwise indicated
O
O
O
OH
S
Cl
Cl
HN
O
Z
Y
A
X
Compound
A
X
Y
Z
CCR2 fpKi pKa
CTX 1 lM antagonist
NHSO2 %I at 0.1, 1% HS
27
28
29
30
31
N
N
N
6.1
6.4
7.3
7.5
8.9
—
—
7.54
7.32
6.58
—
—
15, 0
60, 41
67, 37
N
N
N
N
C–Cl
Table 5
Activity (CCR2 and chemotaxis) and physical properties of selected biaryl ether modifications
O
O
O
O
S
Cl
Cl
S
Cl
Cl
HN
HN
O
O
R1
N
N
N
CN
Cl
31-41
42
R1
Compound
R
CCR2 fpKi
CTX 1
l
M % I at 0.1, 1% HS
log D
pKa
NHSO2
2
31
32
33
34
2-CO2H
8.9
8.7
8.8
8.5
67, 36
—
42, 33
—
0.8
—
0.44
0.4
6.58
—
—
—
R
3
3-F, 2-CO2H
3-CO2H
4-CO2H
H
4
31-35
HO
N
O
( )2
35
7.6
31, 0
2.1
5.6
2-
H
2-Me
4
36
37
7.7
8.0
36, 5.4
72, 50
2.32
2.7
5.72
7.64
R
5
38
39
40
5-CO2H
5-Cl
6-CN
8.0
7.9
7.8
17, 0
55, 0
—
À0.67
6.04
—
—
6
2
N
3.1
2.12
36-40
41
42
—
—
6.7
7.3
—
—
—
N
N
—
43, 12
1.5
5.23

3964
S. Peace et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3961–3964
Table 6
Selected DMPK, developability and chemokine selectivity data
Compound
Rat PK summary
T 1/2 iv, po Cl (mL/min/
In vitro developability profile
p450 profile (pIC50
Chemokine selectivity
F
Vd (L/
hERGa
)
CCR2
CCR1
CCR4
CCR5
(%)
(h)
kg)
kg)
(pIC50
)
(pKi)
(pKi)
(pKi)
(pIC50)
1A2 2C19 2C9
2D6
3A4
24
31
35
37
39
70
97
37
75
58
6.5, 5.9
4.5, 4.5
1.1, —
0.8, 2.3
1.2, —
0.6
0.71
13
4.3
2.4
0.2
5.1
<4.5
<4.2
<4.5
—
4.4 4.96
4.4 4.4
<4.0 5.06
<4.0 4.65
<4.0 5.18
5.38
4.67
5.3
5.87 <4.0
5.97 <4.0
4.27 5.0
8.5
8.9
7.3
7.7
6.7
7.0
6.8
6.4
6.1
5.8
5.9
—
6.2
6.6
0.12
0.47
0.34
0.2
4.0
4.0
<4.0
4.69 7.6
4.71 8.0
6.36 7.9
<10%b
40%b
—
a
hERG fluorescence-polarization binding assay.
% inhibition at 10 lM antagonist.
b
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.04.142.
References and notes
1. Charo, I. F.; Ransohoff, R. M. N. Eng. J. Med. 2006, 354, 610.
2. (a) Boring, L.; Gosling, J.; Cleary, M.; Charo, I. F. Nature 1998, 394, 894; (b)
Buckle, D. R.; Hedgecock, C. J. R. Drug Discovery Today 1997, 2, 325; (iii) Reape,
T. J.; Groot, P. H. Atherosclerosis 1999, 147, 213; (iv) Dawson, T. C.; Kuziel, W. A.;
Osahar, T. A.; Maeda, N. Atherosclerosis 1999, 143, 205.
3. (a) Ruth, J. H.; Rottman, J. B.; Katschke, K. J., Jr.; Qin, S.; Wu, L.; LaRosa, G.;
Ponath, P.; Pope, R. M.; Koch, A. E. Arthritis Rheum. 2001, 44, 2750; (b) Carulli,
M. T.; Ong, V. H.; Ponticos, M.; Xu, S.; Abraham, D. J.; Black, C. M.; Denton, C. P.
Arthritis Rheum. 2005, 52, 3772.
4. Xia, M.; Sui, Z. Expert Opin. Ther. Pat. 2009, 19, 295.
5. Baba, M.; Osamu, N.; Kanzaki, N.; Okamoto, M.; Sawada, H.; Iizawa, Y.;
Shiraishi, M.; Aramaki, Y.; Okonogi, K.; Ogawa, Y.; Meguro, K.; Fujino, M. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 5698.
6. Butora, G.; Jiao, R.; Parsons, W. H.; Vicario, P. P.; Jin, H.; Ayala, J. M.; Cascieri, M.
A.; Yang, L. H. Bioorg. Med. Chem. Lett. 2007, 17, 3636.
Figure 3. Reversibility of CCR2 blockade at 10 lM 31 in 1% FBS following buffer
exchange.
7. Kothandaraman, S.; Donnely, K. L.; Butora, G.; Jiao, R.; Pasternak, A.; Morriello,
G. J.; Goble, S. D.; Zhou, C.; Mills, S. G.; MacCoss, M.; Vicario, P. P.; Ayala, J. M.;
DeMartino, J. A.; Struthers, M.; Cascieri, M. A.; Lihu Yang, L. Bioorg. Med. Chem.
Lett. 2009, 19, 1830.
8. Cherney, R. J.; Brogan, J. B.; Moa, R.; Lo, Y. C.; Yang, G.; Miller, P. B.; Scherle, P.
A.; Molino, B. F.; Carter, P. H.; Decicco, C. P. Bioorg. Med. Chem. Lett. 2009, 19,
597.
9. Kettle, J. G.; Faull, A. W.; Barker, A. J.; Davies, D. H.; Stone, M. A. Bioorg. Med.
Chem. Lett. 2004, 14, 405.
10. Boyden chamber (48-well), CCL2-stimulated chemotaxis of THP-1 cells; foetal
bovine serum (FBS) or human serum (HS) added to assay buffer. %I and pKi
correlated well.
considered to be of benefit given that signaling redundancy is
mooted as one of the likely reasons for the poor clinical efficacy ob-
served to date with selective chemokine antagonists.15
The time-dependent behavior of compound 31 in chemotaxis
was investigated. THP-1 cells were incubated with 10 lM 31 for
30 min and the assay buffer exchanged. Periodic stimulation of
chemotaxis provided the reversibility curve illustrated in Figure
3. The half-life for this reversibility was measured as approxi-
mately 280 min, suggesting a slow dissociation rate of the com-
pound from the receptor.
11. Saponin added to assay buffer resulting in no significant change in pKi, see:
Ryan, A. J.; Gray, N. M.; Lowe, P. N.; Chung, C.-W. J. Med. Chem. 2003, 46(16),
3448.
HTS hit 6 (CCR2 pKi 6.1, MW 471, c log P16 5.1) was optimized
through careful attention to molecular size and physical properties
to provide compound 37 (CCR2 pKi 8.0, MW 443, c log P 3.4). The
apparent influence of plasma proteins on activity was moderated
via removal of the carboxylic acid and through reduction of lipo-
philicity. Measured sulfonamide pKas were apparently related to
the activity shift observed in chemotaxis, as exemplified by the clo-
sely related analogs 36, 37 and 42 whereby the most lipophilic, but
least acidic compound proved to be the most effect anti-chemotac-
tic agent.
12. See ‘Identification and prediction of promiscuous aggregating inhibitors among
known drugs’: Seidler, J.; McGovern, S. L.; Doman, T. N.; Shoichet, B. K. J. Med.
Chem. 2003, 46(21), 4477.
13. (a) Goodman, K. B.; Sehon, C. A.; Cleary, P. A.; Philp, J.; Peace, S. PCT Int. Appl.
WO2006-US61543, 2007. (b) Brooks, C.; Peace, S.; Smethurst, C.; Watson, S. P.
PCT Int. Appl. WO2006-US28321, 2007. Representative experimental data
provided in Supplementary information.
14. Chromatographic data from an immobilised human serum albumin column
suggested high (>98%) protein binding for all examples. Detailed investigations
using ultrafiltration and Biacore methodology confirmed very high protein
binding (>99.5%) but failed to usefully differentiate compounds.
15. Horuk, R. Nat. Rev. Drug Disc. 2009, 8, 23.
16. Calculated using ACD c log P, version 11.