Discovery of small molecule benzimidazole antagonists of the chemokine receptor CXCR3

Article abstract of DOI:10.1016/j.bmcl.2008.01.074

High-throughput screening identified a low molecular weight antagonist of CXCR3 displaying micromolar activity in a membrane filtration-binding assay. Systematic modification of the benzimidazole core and tethered acetophenone moiety established tractable SAR of analogs with improved physicochemical properties and sub-micromolar activity across both human and murine receptors.

Full text of DOI:10.1016/j.bmcl.2008.01.074

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 1573–1576
Discovery of small molecule benzimidazole antagonists
of the chemokine receptor CXCR3
Martin E. Hayes,* Grier A. Wallace, Pintipa Grongsaard, Agnieszka Bischoff,
Dawn M. George, Wenyan Miao, Michael J. McPherson, Robert H. Stoffel,
David W. Greenà and Gregory P. Roth§
Abbott Bioresearch Center, 381 Plantation Street, Worcester, MA 01605, USA
Received 17 December 2007; revised 18 January 2008; accepted 18 January 2008
Available online 26 January 2008
Abstract—High-throughput screening identified a low molecular weight antagonist of CXCR3 displaying micromolar activity in a
membrane filtration-binding assay. Systematic modification of the benzimidazole core and tethered acetophenone moiety estab-
lished tractable SAR of analogs with improved physicochemical properties and sub-micromolar activity across both human and
murine receptors.
Ó 2008 Elsevier Ltd. All rights reserved.
Chemokines are a subfamily of chemotactic cytokines
that bind to cell surface G-protein coupled receptors
(GPCRs) and are responsible for modulating leukocyte
migration.1 A number of these receptors and their
endogenous ligands are thought to play a key role as
pro-inflammatory mediators of autoimmune diseases2
such as multiple sclerosis3 and rheumatoid arthritis.4
The chemokine IP-10 (CXCL10) and its receptor,
CXCR3, are highly expressed within the CNS at sites
of demyelination in multiple sclerosis patients.5 Studies
using neutralizing antibodies for CXCL10 have demon-
strated efficacy at preventing disease onset in a rodent
EAE model of multiple sclerosis.6 Therefore, small mol-
ecule antagonists of CXCR3 would be of interest as a
potential therapy for the treatment of autoimmune dis-
orders such as multiple sclerosis. Several small molecule
antagonists of CXCR3 have been reported including 4-
N-aryl-1,4-diazepine ureas,7 1-aryl-3-piperidin-4-yl-ur-
eas, and 2-amino(4-piperidinyl)azoles.8 Another small
molecule, AMG-487, has been progressed into clinical
trials for both psoriasis and rheumatoid arthritis.9
A high-throughput screen of the Abbott corporate com-
pound collection identified 1 as a moderately potent
(IC50 = 3 lM), functional antagonist of CXCR3. The
moderate molecular weight (323) and cLogP (2.7) of
the compound made it an attractive starting point for
our hit-to-lead effort (Fig. 1). Three different synthetic
approaches were utilized for the synthesis of analogs
as summarized in Scheme 1. 2-Acylbenzimidazoles such
as 4 and 2-alkylbenzimidazoles such as 9 were prepared
by condensation of aryldiamines of the type 2 with an
appropriate carboxylic acid at elevated temperatures
followed by oxidation to prepare compounds of the type
3.10 Alkylation of the resulting benzimidazoles was
Keywords: Chemokine; Chemokine antagonist; CXCR3; IP-10; CXC-
L10; Multiple sclerosis; Benzimidazole; 2-Iminobenzimidazole; Partial
solubility.
*
Corresponding author. Tel.: +1 508 688 8081; fax: +1 508 688
8100; e-mail: martin.hayes@abbott.com
Present address: Wyeth Research, Cambridge, MA 02140, USA.
Present address: Amgen Inc., Cambridge, MA 02139, USA.
Present address: Burnham Institute for Medical Research at Lake
Nona, Orlando, FL 32819, USA.
à
§
Figure 1. CXCR3 antagonist hit from HTS.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.01.074

1574
M. E. Hayes et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1573–1576
We initially examined the impact of modifications to the
acetophenone moiety as shown in Table 1. Replacement
of the 4-nitro substituent with isosteres such as the cya-
no group (entry 4a) was not tolerated but halogen
groups (entries 4b and 4c) provided binding potencies
similar to that of the lead. Removing the para-substitu-
ent also resulted in compounds with significantly re-
duced potency (entry 4d). Substitution at the C-5 and
C-6 position of the benzimidazole core with groups such
as methoxy (entries 4f and 4g) and phenyl (entries 4h
and 4i) resulted in weakly potent compounds while sub-
stitution at C-4 with methoxy (entry 4e; IC50 = 3 lM)
showed no change in the binding assay. Several modifi-
cations of the 2-acyl group, including replacement of the
carbonyl with a sulfoxide (entry 7a) and reduction of the
carbonyl to the corresponding carbinol (entry 9a), re-
sulted in compounds having reduced potency.
Analogs in the 2-acyl series, as well as many analogs
containing acyl isosteres, were poorly soluble in aqueous
buffer at the higher concentrations of the assay. This re-
sulted in an incomplete dose response curve and an
apparent partial antagonism that complicated SAR
interpretation for these compounds. However, substitu-
tion of the 2-acyl for an imino group provided com-
pounds with improved solubility19 which allowed for
complete dose response curves that typically provided
>80% inhibition relative to the endogenous ligand
(CXCL10). All future analogs incorporated this 2-imino
group because of both the improved potency and in-
creased solubility it afforded. Extended substitution on
this 2-imino group invariably resulted in a significant
reduction in potency as shown in Table 2 (entries 12a–
c). Only small aliphatic substituents were tolerated with
the 2-methylimino analog (entry 12c) being equipotent
to the unsubstituted 2-imino compound (entry 12d).
Scheme 1. Synthesis of 2-substituted benzimidazoles 4a–j, 7a, 9a, b9,
and 12a–q. Yields given in the scheme are for compounds 4b, 7a, 9a,
and 12e. Reagents and conditions: (a) R2CH(OH)CO2H, 4 N HCl,
90 °C, 16 h; (b) Dess–Martin Periodinane, CH2Cl2, rt, 16 h; (c)
ArC(O)CH2Br, K2CO3, Acetone, rt, 4 h; (d) R2I, K2CO3, MeCN, rt,
16 h, then mCPBA, CH2Cl2, rt, 16 h; (e) R2CO2H, 4 N HCl, 90 °C,
16 h; (f) R2NH2, EtOH, reflux, 16 h; (g) Pd/C, NH4CO2H, EtOH, rt,
16 h; or Fe, HCl, EtOH, reflux, 16–48 h; (h) BrCN, MeCN, rt, 8–16 h;
(i) ArC(O)CH2Br, DMF, rt, 4–12 h.
Having successfully improved solubility by incorporat-
ing the 2-imino substituent on the core, we next investi-
gated substitution on the acetophenone group. As
shown in Table 2, both bromine and chlorine were suit-
effected by treatment with an a-bromoacetophenone in
the presence of potassium carbonate.11 Sulfoxide
analogs 7 were prepared from the corresponding
benzothiourea via S-alkylation and oxidation.12,13 Prep-
aration of 2-iminobenzimidazoles 12 was accomplished
by utilizing an addition–elimination protocol. Thus,
treatment of an o-fluoronitroarene with an appropri-
ately substituted amine followed by reduction of the
nitro group using catalytic hydrogenation14 or iron in
protic acid15 and subsequent reaction of the resulting
diamines 11 with cyanogen bromide provided 2-imino-
benzimidazoles.16 Further elaboration by alkylation
with an a-bromoacetophenone gave the desired
product 12.
Table 1. Binding potency of analogs
N
R1
R2
N
R3
R3
Entry R1
R2
huRLB
a
IC50 (lM)
4a
4b
4c
4d
4e
4f
H
H
H
H
C(O)Me
C(O)Me
C(O)Me
C(O)Me
CH2C(O)Ph(4-CN) >100
CH2C(O)Ph(4-Br)
CH2C(O)Ph(4-Cl)
CH2C(O)Ph
2
8
>100
3
4-OMe C(O)Me
5-OMe C(O)Me
6-OMe C(O)Me
CH2C(O)Ph(4-Br)
CH2C(O)Ph(4-Br)
CH2C(O)Ph(4-Br)
CH2C(O)Ph(4-Br)
CH2C(O)Ph(4-Br)
>50
>50
>50
>50
4g
4h
4i
5-Ph
6-Ph
H
C(O)Me
C(O)Me
S(O)Me
Compounds were initially evaluated for their ability to
inhibit [125I]-labeled CXCL10 binding to membranes
of CHO cells stably expressing human CXCR3.17 Func-
tional antagonism was also measured in CHO cells using
a FLIPR-based calcium mobilization assay.18
7a
9a
CH2C(O)Ph(4-NO2) 11
H
CH(OH)Me CH2C(O)Ph(4-NO2) 20
a IC50 values are an average of two runs.

M. E. Hayes et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1573–1576
Table 2. Binding and functional antagonism of 2-iminobenzimidazole analogs
1575
Me
N
R3
NR1
N
R2
R1
R2
R3
huRLB IC50 (lM)
huFLIPR IC50 (lM)
a
a
Entry
12a
12b
12c
12d
12e
12f
C(O)Me
CH2CH2OMe
CH2C(O)Ph(4-NO2)
CH2C(O)Ph(4-Br)
CH2C(O)Ph(4-Cl)
CH2C(O)Ph(4-Cl)
CH2C(O)Ph(4-Br)
CH2C(O)Ph(3,4-diCl)
CH2C(O)Ph(3-Cl)
CH2C(O)Ph(2-Cl)
CH2S(O)Ph(4-Cl)
CH2CH2Ph(4-Br)
H
50
1
H
CH3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
0.8
0.8
0.8
4
H
9
H
H
12g
12h
12i
H
16
22
3
H
H
12j
H
11
6
12k
12l
CH2CH2CH2C(O)Ph(4-Br)
CH2CH(OH)Ph(4-Br)
CH2C(O)Ph(4-Cl)
CH2C(O)Ph(4-Cl)
CH2C(O)Ph(4-Cl)
CH2C(O)Ph(4-Cl)
CH2C(O)Ph(4-Cl)
CH2C(O)Ph(4-Cl)
CH2C(O)Ph(4-Cl)
CH2C(O)Ph(4-Br)
H
H
40
8
12m
12n
12o
12p
12q
12r
12s
12t
6-Me
5-Me
4-Me
4-Et
4-nPr
4-Cl
4-CF3
4-OMe
1
5
0.1
0.03
0.5
0.3
0.7
0.8
0.08
0.07
0.4
a IC50 values are an average of two runs.
able replacements for the 4-nitro group (entries 12d and
12e). Substitution at the 2- or 3-positions resulted in re-
duced potency (entries 12f–h). Isosteric replacement of
the ketone with a sulfoxide (entry 12i) also resulted in
a notable decrease in potency. Efforts to reduce the elec-
trophilic ketone to a carbinol (entry 12l) or methylene
(entry 12j) resulted in a loss of potency. Chain extension
of the acetophenone up to two atoms (entry 12k) indi-
cated that one methylene unit was preferred.
along with limited stability in rat liver microsomes
(49% parent remaining after 30 min) and modest protein
binding (85%) in rat plasma. Compound 12o also dem-
onstrated consistent activity across species in the bind-
ing assay with good alignment in the functional assay
(FLIPR) as shown in Table 2.
In summary, from an internal screen of our corporate
compound collection we identified a low micromolar
antagonist of CXCR3. Switching to a 2-iminobenzimi-
dazole core resolved an apparent partial antagonism
due to limited solubility and also afforded compounds
with improved potency that demonstrated tractable
SAR. Incorporation of small, non-polar groups at the
C-4 position of the core resulted in notable potency
gains and improved functional antagonism along with
We next re-evaluated the impact of substitution on the
benzimidazole core as shown in Table 2. Substitution
with a methyl group at the C-6 position (entry 12m)
was not tolerated while compounds possessing substitu-
tion at C-5 (entry 12n) were nearly equipotent with com-
pound 12d. A substantial boost in potency was observed
with methyl substitution at the C-4 position (entry 12o).
Other small groups also resulted in improved potency
(entry 12r) including the ethyl-substituted compound
12p that showed a 25-fold increase in potency relative
to 12d. In addition, an increase in functional antagonism
(FLIPR) was observed with compounds incorporating
small aliphatic groups at the C-4 position (compare en-
try 12o with 12d). Larger groups, such as propyl (entry
12q), and more polar substituents (entries 12s and 12t)
did not show a significant improvement in potency sug-
gesting that substituents at the C-4 position may poten-
tially be binding within a small lipophilic pocket of the
receptor.
Compound 12o was further profiled for its pharmacoki-
netic properties in mouse to establish a benchmark for
the chemotype (Fig. 2). Compound 12o exhibited good
bioavailability and half-life upon oral administration20
Figure 2. Pharmacokinetic properties and species selectivity of com-
pound 12o.

1576
M. E. Hayes et al. / Bioorg. Med. Chem. Lett. 18 (2008) 1573–1576
Society, Atlanta, GA, March 26–30, 2006; (c) Johnson,
M.; Li, A.-R.; Liu, J.; Fu, Z.; Zhu, L.; Miao, S.; Wang, X.;
Xu, Q.; Huang, A.; Matcus, A.; Xu, F.; Ebsworth, K.;
Sablan, E.; Danao, J.; Kumer, J.; Dairaghi, D.; Lawrence,
C.; Sullivan, T.; Tonn, G.; Schall, T.; Collins, T.; Medina,
J. Bioorg. Med. Chem. Lett. 2007, 17, 3339; (d) Du, X.;
Chen, X.; Mihalic, J. T.; Deignan, J.; Duquette, J.; Li, A.-
R.; Lemon, B.; Ma, J.; Miao, S.; Ebsworth, K.; Sullivan,
T. J.; Tonn, G.; Collins, T. L.; Medina, J. C. Bioorg. Med.
Chem. Lett. 2008, 18, 608; (e) Li, A.-R.; Johnson, M. G.;
Liu, J.; Chen, X.; Du, X.; Mihalic, J. T.; Deignan, J.;
Gustin, D. J.; Duquette, J.; Fu, Z.; Zhu, L.; Marcus, A. P.;
Bergeron, P.; McGee, L. R.; Danao, J.; Lemon, B.;
Carabeo, T.; Sullivan, T.; Ma, J.; Tang, L.; Tonn, G.;
Collins, T. L.; Medina, J. C. Bioorg. Med. Chem. Lett.
2008, 18, 688.
favorable pharmacokinetic properties to support further
optimization efforts.
Acknowledgments
The authors acknowledge Elizabeth Alder and Lorraine
McGuinness for generating radio-ligand binding and
FLIPR assay data, Linda E. Chovan and Candace L.
Black-Schaeffer for generating HT-ADME reports,
and Brian Bettencourt for acquiring in-vivo pharmaco-
kinetics data.
References and notes
10. Ramaiah, K.; Grossert, J. S.; Hooper, D. L.; Dubey, P.
K.; Ramanatham, J. J. Indian Chem. Soc. 1999, 76, 140.
11. Demirayak, S.; Mohsen, U. A.; Karaburun, A. C. Eur. J.
Med. Chem. Chim. Ther. 2002, 37, 255.
1. D’Ambrosio, D.; Panina-Bordignon, P.; Sigigaglia, F.
J. Immunol. Methods 2003, 273, 3.
2. Godessart, N. Ann. N.Y. Acad. Sci 2005, 1051, 647.
3. Sorensen, T. L.; Tani, M.; Jensen, J.; Pierce, V.; Lucchi-
netti, C.; Folcik, V. A.; Qin, S.; Rottman, J.; Selleberjerg,
F.; Strieter, R. M.; Frederiksen, J. L.; Ransohoff, R. M.
J. Clin. Invest. 1999, 103, 807.
4. Qin, S.; Rottman, J. B.; Myers, P.; Kassam, N.; Weinblatt,
M.; Loetscher, M.; Koch, A. E.; Moser, B.; Mackay, C. R.
J. Clin. Invest. 1998, 101, 746.
12. Gardiner, J. M.; Loyns, C. R. Tetrahedron 1995, 51,
11515.
13. Graber, D. R.; Morge, R. A.; Sih, J. C. J. Org. Chem.
1987, 52, 4620.
14. Li, Q.; Li, T.; Woods, K. W.; Gu, W.; Cohen, J.; Stoll,
V. S.; Galicia, T.; Hutchins, C.; Frost, D.; Rosenberg,
S. H.; Sham, H. L. Bioorg. Med. Chem. Lett. 2005, 15,
2918.
5. Simpson, J. E.; Newcombe, J.; Cuzner, M. L.; Woodroofe,
M. N. Neuropathol. Appl. Neurobiol. 2000, 26, 133.
6. (a) Tsunoda, I.; Lane, T. E.; Blackett, J.; Fujinami, R. S.
Multiple Sclerosis 2004, 10, 26; (b) Liu, L.; Callahan, M.
K.; Huang, D.; Ransohoff, R. M. Curr. Topics Dev. Biol.
2005, 68, 149.
15. Shepherd, T.; Smith, D. M. J. Chem. Soc., Perkin Trans. 1
1987, 501.
16. Valdez, J.; Cedillo, R.; Hernandez-Campos, A.; Yepez, L.;
Hernandez-Luis, F.; Navarrete-Vazquez, G.; Tapia, A.;
Cortes, R.; Hernandez, M.; Castillo, R. Bioorg. Med.
Chem. Lett. 2002, 12, 2221.
7. Cole, A. G.; Stroke, I. L.; Brescia, M-R.; Simhadri, S.;
Zhang, J. J.; Hussain, Z.; Snider, M.; Haskell, C.; Ribeiro,
S.; Appell, K. C.; Henderson, I.; Webb, M. L. Bioorg.
Med. Chem. Lett. 2006, 16, 200.
17. Radioligand binding assays were performed in CHO cells
expressing human or murine CXCR3. All compounds
were dissolved in DMSO and assays run at a final DMSO
concentration of 1% (v/v). [125I]-labeled CXCL10 was
purchased from Perkin-Elmer and used at 50 pM per
assay.
18. Calcium flux assays were performed in CHO cells
expressing human CXCR3 and the Ga16 coupling protein
using a FLIPR instrument (Molecular Devices). All
compounds were dissolved in DMSO and assays run at
a final DMSO concentration of 1% (v/v). Human CXCL10
was purchased from Peprotech and used at a final assay
concentration of 30 nM.
19. For example, equilibrium solubility as measured by
CLND detection of a 100-mM aqueous solution of
compound 1 is 4.5 lM as compared with 84 lM for
compound 12d and 79 lM for compound 12o.
20. Female SJL mice were obtained from Jackson Labs and
compound 12o was dosed in a vehicle of D5W with 5%
8. (a) Allen, D. R.; Bolt, A.; Chapman, G. A.; Knight, R. L.;
Meissner, J. W. G.; Owen, D. A.; Watson, R. J. Bioorg.
Med. Chem. Lett. 2007, 17, 697; (b) Watson, R. J.; Allen,
D. R.; Birch, H. L.; Chapman, G. A.; Hannah, D. R.;
Knight, R. L.; Meissner, J. W. G.; Owen, D. A.; Thomas,
E. J. Bioorg. Med. Chem. Lett. 2007, 17, 6806; (c) Watson,
R. J.; Allen, D. R.; Birch, H. L.; Chapman, G. A.; Galvin,
F. C.; Jopling, L. A.; Knight, R. L.; Meier, D.; Oliver, K.;
Meissner, J. W. G.; Owen, D. A.; Thomas, E. J.; Tremayne,
N.; Williams, S. C. Bioorg. Med. Chem. Lett. 2008, 18, 147;
(d) Knight, R. L.; Allen, D. R.; Birch, H. L.; Chapman, G.
A.; Galvin, F. C.; Jopling, L. A.; Lock, C. J.; Meissner, J.
W. G.; Owen, D. A.; Raphy, G.; Watson, R. J.; Williams,
S. C. Bioorg. Med. Chem. Lett. 2008, 18, 629.
9. (a) Berry, K., et al. Inflamm. Res. 2004, Suppl. 3, S197–
S234 (abstract A089); (b) Johnson, M. G.; Li, A.; Liu, J.;
Marcus, A. P.; Huang, A. X.; Medina, J. C. Abstracts of
Papers, 231st National Meeting of the American Chemical
Ethanol, 50% PEG 200, and
1 equivalent of 2 M
HCl:10 mpk (po) by gavage and 5 mpk (iv) as a slow
bolus in a tail vein.