CCR2B receptor antagonists: Conversion of a weak HTS hit to a potent lead compound

Article abstract of DOI:10.1016/S0960-894X(00)00347-4

A weak HTS hit at the CCR2B receptor has been converted into a potent antagonist by array SAR studies. Selectivity over the closely related CCR5 receptor is also achieved. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00347-4

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1803±1806
CCR2B Receptor Antagonists: Conversion of a Weak HTS Hit to
a Potent Lead Compound
Ian T. Forbes,* David G. Cooper, Emma K. Dodds, Deirdre M. B. Hickey,
Robert J. Ife, Malcolm Meeson, Martin Stockley, Theo A. Berkhout,
Jayneeta Gohil, Pieter H. E. Groot and Kitty Moores
Departments of Discovery Chemistry and Vascular Biology, SmithKline Beecham Pharmaceuticals,
New Frontiers Science Park, Third Avenue, Harlow, Essex CM19 5AD, UK
Received 20 April 2000; revised 9 June 2000; accepted 13 June 2000
AbstractÐA weak HTS hit at the CCR2B receptor has been converted into a potent antagonist by array SAR studies. Selectivity
over the closely related CCR5 receptor is also achieved. # 2000 Elsevier Science Ltd. All rights reserved.
Over recent years, there has been a rapid growth in the
number of isolated low molecular weight proteins called
1
chemokines (chemotactic cytokines). These proteins are
involved in a variety of in¯ammatory responses via
interaction with chemokine receptors located on the cell
surface of leukocytes followed by chemotaxis and in®l-
tration into the adjacent tissue. The chemokine proteins
can be divided into four families dependent on the
arrangement of conserved cysteine residues near the N-
1
terminus. Monocyte chemotactic protein-1 (MCP-1) is
a member of the CC class of chemokines, and has been
2
strongly implicated in various in¯ammatory diseases.
The e€ects of MCP-1 are mediated primarily via the
3
CCR2B receptor, and it has been widely recognised that
antagonists of this receptor are potential therapeutic
agents for various pathological conditions, e.g., athero-
4
5
sclerosis and rheumatoid arthritis. This hypothesis has
6
been recently validated by studies using MCP-1 and
7
CCR2 knockout mice. Two series of small molecule
CCR2B antagonists have recently been reported, exem-
8
pli®ed by the Roche compound 1 and the Takeda
9
compound 2. Both these compounds have major dis-
advantages, namely compound 1 possesses poor func-
tional activity, and compound 2 possesses mixed CCR5/
CCR2B receptor anities. We now report the identi®-
cation of a potent functional antagonist at the CCR2B
receptor with additional selectivity over CCR5.
High-throughput screening of the SmithKline Beecham
compound collection against the cloned human CCR2B
receptor identi®ed the indole derivative 3 as a weak ligand
12
(
K 5.3 mM). Herein we describe the SAR around 3 and
i
its conversion into a potent antagonist at the CCR2B
receptor.
*Corresponding author. Tel.:+44-1279-622126; fax:+44-1279-627841;
e-mail: ian_t_forbes@sbphrd.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00347-4

1
804
I. T. Forbes et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1803±1806
Structure±activity relationships
Initially, we focused on modi®cations to the aromatic
amide group in 3 (see Table 1). Unfortunately, it was
found that the majority of substituted benzamides inves-
tigated exhibited reduced potency at the CCR2B receptor
(
data not shown), with the exception of halogen sub-
stituted analogues, e.g., 8 and 9. Replacement of the
amide carbonyl in 9 by sulfonyl also resulted in reduced
potency. Introduction of a one- or two-carbon linker as
in 11 or 12 o€ered no advantage, however replacement
of the methylene group in 11 with NH to form the urea
13 provided a threefold potency increase. More interest-
ingly, replacement of the benz-amide functionality in 9
with 3,4-dichlorocinnamide to give 14 conferred a 10-
fold increase in anity.
Our initial chemical programme was divided into three
discrete areas: modi®cation to the biphenyl amide moiety,
variation of the chain length between the two side-chain
nitrogen atoms, and substitution on the indole ring.
Having identi®ed the cinnamide linker in 14 as opti-
mal, we explored the in¯uence of substitution on
the aromatic ring. As illustrated in Table 2, highest
anity was observed with a small lipophilic sub-
stituent at positions 3 and/or 4 on the phenyl ring, 14
and 19 representing preferred compounds. Larger
substituents as in 25, or polar groups as in 27, were
disfavoured.
Chemistry10
Amide linked compounds were synthesised by either of
the routes shown in Scheme 1. Alkylation of the
1
1
readily available indolopiperidines 4 with a bromo-
alkylphthalimide gave the chain extended protected
amine derivatives 5, which were deprotected to a€ord
primary amines 6. Coupling with the appropriate car-
boxylic acid provided the desired target structures 7.
Alternatively, the products 7 could be obtained via a
direct alkylation of 4 with the appropriate bromoalkyl
amides. Urea and sulfonamide analogues were prepared
by a similar process.
We then studied the e€ect of varying the carbon chain
length between the basic nitrogen and the amide nitrogen.
As can be seen from Table 3, a C linkage is preferred,
5
closely followed by C and C , with C and C being
6
4
3
2
essentially inactive.
ꢀ
Scheme 1. Synthetic route to amide derivatives. Reagents: (a) Br(CH
2
)
H, 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide, HOBT, CH
n
Nphthalimide, NaHCO
Cl
3
, DMF, 80 C (75±85%); (b) hydrazine, ethanol,
, room temp. (50±90%); (d) Br(CH NHCOR2,
re¯ux (65±90%); (c) R2CO
ꢀ
2
2
2
2 n
)
NaHCO
3
, DMF, 80 C (60±90%).

I. T. Forbes et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1803±1806
1805
Table 1. Amide variations
Table 3. Chain length variation
#
X
R
K
i
(nM)12
#
n
K
i
(nM)12
3
8
9
1
1
1
1
1
CˆO
CˆO
CˆO
4-Ph
4-Br
5400
7100
5100
3
3
3
1
3
0
2
3
4
5
6
>10,000
2500
660
420
>10,000
3,4-diCl
3,4-diCl
3,4-diCl
3,4-diCl
3,4-diCl
3,4-diCl
1
2
4
3
0
SO
2
>10,000
4400
5900
1
2
3
4
C(ˆO)CH
C(ˆO)CH CH
C(ˆO)NH
trans C(ˆO)CHˆCH
2
2
2
a
1600
420
aIn original 4-phenylbenzamide series.
Table 2. Cinnamide substitution
Table 4. Indole ring substitution
#
R
K
i
(nM)12
#
R
K
i
(nM)12
#
R
K
i
(nM)12
1
1
1
1
1
1
2
2
4
3,4-diCl
H
420
870
520
560
4400
350
460
790
22
23
24
25
26
27
28
29
4-Me
3-Me
4-Ph
4-Bu
4-NMe
4-OH
550
790
1450
1700
1200
6300
14
34
35
36
37
38
39
40
H
5-OH
5-MeSO NH
5-MeO
4-OH
6-OH
5-OH-1-Me
5-OH-2-Me
420
50
79
590
360
360
4200
234
5
6
7
8
9
0
1
4-Cl
3-Cl
2-Cl
4-Br
4-I
2
t
2
4-NHAc
4-CN
12,900
2800
3-CF
3
Finally, we undertook an SAR study on indole ring
substitution, with key ®ndings summarised in Table 4.
Signi®cantly, it was found that an H-bond donor at C-5
was bene®cial with the 5-OH indole 34 exhibiting a 10-
fold increase in anity. This e€ect was also mirrored by
26 nM (n=2). Similarly, in a functional model based on
inhibition of MCP-1 stimulated chemotaxis of human
monocytes, it was shown that 34 is a potent antagonist
with a mean K of 25 nM (n=2). Finally, the e€ects of
b
antagonist 34 at the CCR2B receptor (K 50 nM) were
i
the 5-MeSO NH analogue 35, which suggests a speci®c
2
demonstrated to be speci®c, with CCR5 receptor anity
considerably lower (K 4260 nM).
H-bonding interaction, as the corresponding 5-MeO, 4-
OH and 6-OH derivatives did not show this increase. The
indole NH appears to be crucial for high anity, since
the indole N-Me analogue 39 was ca. 80-fold less active.
This hypothesis is also supported by the lack of activity
for the corresponding benzofuran and benzothiophene
i
In summary we have identifed 34 as a potent CCR2B
receptor antagonist, which has comparable activity in
binding and functional studies, and which is selective
over the CCR5 receptor, its closest chemokine receptor
homologue. Unfortunately, 34 was not progressed due
to unwanted 5-HT receptor anities. Modi®cations to
structure 34, which reduce its propensity to interact with
5-HT receptors, will be the subject of a future publication.
1
3
analogues. C-2 substitution, as in 40, conferred a
slight reduction in anity which can be rationalised by
an adverse conformational e€ect on the active indole-
piperidine orientation.
Having identi®ed 34 as a potent ligand at the CCR2B
receptor, we then examined its e€ects in a functional
system, and also its e€ects at the closely related CCR5
receptor. Initially, 34 was tested for its ability to inhibit
MCP-1-stimulated calcium transients using fura-2-loaded
References and Notes
1. Saunders, J.; Tarby, C. M. Drug Discovery Today 1999, 4, 80.
2. Rollins, B. J. Mol. Med. Today 1996, 2, 198.
human monocytes, and was found to have a mean K of
b

1
806
I. T. Forbes et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1803±1806
3. Charo, I. F.; Myers, S. J.; Herman, A.; Franci, F.; Connolly,
A. J.; Coughlin, S. R. Proc. Natl. Acad. Sci. USA 1994, 91, 2752.
11. For example see: Freter, K.; Fuchs, V.; Barsumian, E.;
Oliver, J. T. Arzn. Forsch. 1985, 35, 272.
4
5
. Leonard, E. J. Challenges Mod. Med. 1994, 3, 25.
. Gong, J.-H.; Ratkay, L. G.; Water®eld, J. D.; Clark-Lewis,
12. Antagonists were tested using a radioligand binding
displacement assay. Compounds were dissolved in DMSO and
a concentration range (1% ®nal concentration in assay) and
were incubated in the presence of 0.14 nM 1 I-MCP-1 and
membranes (100,000 cell equivalents) from CHO cells stably
transfected with the MCP-1 receptor CCR2B. The incubation
took place at RT for 2 h in a total volume of 110 mL 50 mM
I. J. Exp. Med. 1997, 186, 131.
. Gosling, J.; Slaymaker, S.; Gu, L.; Tseng, S.; Zlot, C. H.;
Young, S. G.; Rollins, B. J.; Charo, I. F. J. Clin. Invest. 1999,
25
6
1
7
1
8
03, 773.
. Boring, L.; Gosling, J.; Cleary, M.; Charo, I. F. Nature
998, 394, 894.
. Lapierre, J. M. 1998, 26th National Medicinal Chemistry
2 2
HEPES, 1 mM CaCl , 5 mM MgCl , 0.5% bovine serum albu-
min (fatty acid free) pH 7.4 in 96 well plate format. Membranes
were harvested on GF/C ®lters and ®lters were washed with
4Â1 mL of 50 mM HEPES, 0.5 M NaCl pH 7.0. Filter bound
radioactivity was determined by liquid scintillation counting. For
Symposium, 14±16 June, Richmond, VA, USA.
. Baba, M.; Nishimura, O.; Kanzaki, N.; Okamoto, M.;
9
Sawada, H.; Iizawa, Y.; Shuraishi, M.; Aramaki, Y.; Okonogi,
K.; Ogawa, Y.; Meguro, K.; Fujino, M. Proc. Natl. Acad. Sci.
USA 1999, 96, 5698.
10. All novel compounds gave satisfactory analytical data in
full agreement with their proposed structures.
i
IC50's and K 's multiple curves were analysed by four parameter
®tting using in-house developed Excel software entitled In¯exion
2. All K values represent the mean of at least two determinations.
13. SmithKline Beecham, unpublished results.
i