Identification of 2-arylbenzimidazoles as potent human histamine H4 receptor ligands

Article abstract of DOI:10.1016/j.bmcl.2006.08.117

A series of 2-arylbenzimidazoles was synthesized and found to bind with high affinity to the human histamine H₄ receptor. Structure-activity relationships were investigated through library preparation and evaluation as well as traditional medicinal chemistry approaches, leading to the discovery of compounds with single-digit nanomolar affinity for the H₄ receptor.

Full text of DOI:10.1016/j.bmcl.2006.08.117

Bioorganic & Medicinal Chemistry Letters 16 (2006) 6043–6048
Identification of 2-arylbenzimidazoles as potent human
histamine H4 receptor ligands
Alice Lee-Dutra,* Kristen L. Arienti, Daniel J. Buzard, Michael D. Hack,
Haripada Khatuya, Pragnya J. Desai, Steven Nguyen, Robin L. Thurmond,
Lars Karlsson, James P. Edwards and J. Guy Breitenbucher
Johnson & Johnson Pharmaceutical Research and Development, L.L.C. 3210 Merryfield Row, San Diego, CA 92121, USA
Received 13 July 2006; revised 29 August 2006; accepted 29 August 2006
Available online 20 September 2006
Abstract—A series of 2-arylbenzimidazoles was synthesized and found to bind with high affinity to the human histamine H4 recep-
tor. Structure–activity relationships were investigated through library preparation and evaluation as well as traditional medicinal
chemistry approaches, leading to the discovery of compounds with single-digit nanomolar affinity for the H4 receptor.
Ó 2006 Elsevier Ltd. All rights reserved.
Histamine plays a critical role in a number of physiolog-
ical processes through interaction with four G-protein
coupled receptors: H1, H2, H3, and H4. The H1 receptor
is involved in allergic and inflammatory responses,1 the
H2 receptor participates in gastric acid secretion,2 and
the H3 receptor mediates neurotransmitter release in
the central nervous system.3 Most recently identified,
the H4 receptor is mainly expressed in dendritic cells,
eosinophils, and mast cells.4,5 The preferential expres-
sion of the H4 receptor in immune cells suggests that this
receptor is involved in the regulatory functions of hista-
mine during the immune response. Studies have shown
that the H4 receptor controls the release of inflammatory
mediators and facilitates leukocyte chemotaxis.6–10
Modulation of H4 receptor activity provides an oppor-
tunity for treating inflammatory and allergic conditions.
Recently, we reported indole, benzimidazole, and thie-
nopyrrole piperazine carboxamides as potent H4 recep-
tor ligands.11,12
synthetic route for the preparation of 1 enabled rapid
variation of substituents on different regions of the
molecule.
Cl
N
O
N
NCH3
N
H
1: Ki = 124 nM
Analogs of compound 1 were synthesized by the cou-
pling of commercially available phenylenediamines with
aromatic aldehydes in the presence of Na2S2O5 (Scheme
1).13 Most aldehyde inputs were synthesized using a two-
step procedure: alkylation of the hydroxybenzaldehyde
followed by chloride displacement with the desired
amine.
O
a
R1
O
O
n
b
R1
OH
Br
H
+
H
Cl
Cl
n
In this paper, we describe a series of 2-arylbenzimidaz-
oles with high affinity for the human H4 receptor. The
initial lead compound for this program, compound 1,
was identified as a HTS hit from our corporate com-
pound collection and exhibited moderate affinity
(Ki = 124 nM) for the H4 receptor. An efficient, modular
n = 1-3
R1
R3
R4
O
c
R2
R1
O
N
O
R2
N R2
N
H
n
N
H
R2
n
n = 1-3
n = 1-3
Scheme 1. Reagents and conditions: (a) K2CO3, CH3CN, 60 °C, 18–
24 h; (b) HNR2R2, Na2CO3, KI, n-BuOH, 95 °C, 18–36 h, ꢀ50% over
two steps; (c) phenylenediamine, Na2S2O5, DMA, 80 °C, 6–42%.
Keywords: Histamine H4; Arylbenzimidazole.
*
Corresponding author. Tel.: +1 858 320 3317; fax: +1 858 450
2089; e-mail: alee7@prdus.jnj.com
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.08.117

6044
A. Lee-Dutra et al. / Bioorg. Med. Chem. Lett. 16 (2006) 6043–6048
The initial analogs were designed to examine the impor-
tance of linker length and position about the central
benzene ring (Graph 1). These compounds were exam-
ined in a recombinant human histamine H4 competitive
binding assay using [3H]histamine as the radioligand.14
The SAR investigation revealed a dramatic dependence
of binding affinity upon the linker length as well as the
linker position about the central aryl ring. Of the nine
series that were synthesized and tested for activity, only
two sets (S2 and S6) demonstrated pKi > 5. This sug-
gests that the position of the terminal nitrogen relative
to the central ring is essential for good receptor
binding.15
some of the linkers is not due to lack of particular fea-
tures, but probably results from conformational consid-
erations. It was hypothesized that the benzimidazole
moiety was binding in the same way for all of the ana-
logs, and that the inactive linkers were unable to prop-
erly position and orient the positively charged
piperazine at low conformational energies. In order to
elucidate the bioactive conformation, stochastic confor-
mation searches of an example of each of the linkers S1–
17
S9 using MOE with the MMFF94 force field were car-
ried out. For the active linkers (S2 and S6), a limit of
5 kcal/mol above the global minimum was set. For the
inactive linkers, up to 10 kcal/mol energy above the
global minimum was allowed. All of the conformations
for all of the linkers were aligned using the benzimid-
azole moiety. The position of the terminal nitrogen of
the piperazine was plotted as a sphere in three-dimen-
sional space. A simplified version of the Active Analog
Approach18 was used. Figure 1 illustrates the procedure.
CATALYST16 was used to generate pharmacophores using
an example of each linker (S1–S9). Unfortunately, these
pharmacophores were unable to discriminate between
active and inactive linkers, presumably because of the
similarity of the molecules and because inactivity of
R1
O
N
N
aldehyde
n
R2
NCH3
N
H
S1 para, n = 2
S2 para, n = 3
7.0
6.5
S3 para, n = 4
diamine
S4 meta, n = 2
Cl
NH2
NH2
1
S5 meta, n = 3
pKi
6.0
5.5
5.0
S6 meta, n = 4
S7 ortho, n = 2
NH2
NH2
2
3
aldehyde
S8 ortho, n = 3
S9 ortho, n = 4
H3CO
NH2
NH2
1
2
3
diamine
Graph 1. Displacement of [3H]histamine from recombinant histamine H4 receptor.
Figure 1. Overlay of lowest-energy conformations of benzimidazoles in graph 1: active minus inactive binders.

A. Lee-Dutra et al. / Bioorg. Med. Chem. Lett. 16 (2006) 6043–6048
6045
The yellow spheres represent conformations of the S2
linker, and the red spheres represent conformations of
the S6 linker. Positions in space that were occupied by
both an S2 representative and an S6 representative are
illustrated in the second panel of Figure 1, and are can-
didates for the bioactive conformations of those linkers.
Positions that were also occupied by representatives of
inactive linkers were systematically removed (panels 3–
6). This led to a focused region that is 1.5 A in diameter
˚
and 8.3 A away from the central benzene ring as the
most likely region whose occupancy or unoccupancy
by a piperazine nitrogen could explain the data in Graph
˚
1. Therefore, it was theorized that this distance of 8.3 A
than those described in Graph 1, they all have low-ener-
gy conformations in which the terminal nitrogen of the
piperazine ring is located within the region identified
above. The cis-alkene, which does not span this distance,
was also included as a linker serving as a negative
control for the model. In addition, the aryl ether was
constrained as a benzofuran. As before, the N-methyl-
piperazine region of the molecule was not altered. Other
benzimidazoles were synthesized using aldehydes with
additional substitution on the benzene ring or with a
carbon analog of the ether linker. All necessary library
aldehyde inputs were synthesized using routes shown
in Schemes 1–4.19
˚
between the aryl ring and distal nitrogen of the terminal
piperazine is optimal for potent H4 receptor binding.
Note that there were multiple distinct conformations
of S2 and S6 that were positioned in this region,
and therefore it was not possible to further specify the
Benzimidazole formation was conducted in a library for-
mat using a Bohdan 48-well MiniBlock system. Sodium
metabisulfite was added to the individual reaction tubes
prior to the addition of the aldehyde and phenylenedi-
amine inputs as stock solutions in DMF. Upon reaction
completion, as monitored by HPLC, polymer-supported
hydrazine was added to the reaction tubes to remove ex-
cess aldehyde starting material. Filtration of the resul-
tant reaction mixtures was followed by purification
through reverse phase preparative HPLC. Select data
for these benzimidazoles are shown in Graph 2.
preferred binding
pharmacophore.
conformation
beyond
this
The identification of the likely piperazine binding region
suggested the possibility of creating constrained analogs
that would have increased potency. Several 48-mem-
bered libraries were designed to test this model through
the incorporation of rigidified linkages that would span
˚
this 8.3 A distance. For example, the alkyl linker was re-
placed with a benzene ring, an alkyne or a trans-alkene.
Although these linkers are in some cases quite different
An inspection of Graph 2 clearly suggests that the con-
formational model described above is incomplete. Fail-
ure of the rigidified linkers to improve H4 receptor
O
O
O
a
b, c
OH
H
H
H
Br
N
NCH3
d
e, f
O
O
N
N
H
H
NCH3
NCH3
Scheme 2. Reagents and conditions: (a) homopropargyl alcohol, Pd(PPh3)2Cl2, CuI, Et3N, THF, rt, quant.; (b) CH3SO2Cl, Et3N, CH2Cl2, 0 °C to rt,
quant.; (c) N-methylpiperazine (10 equiv), EtOH, rt, 43%; (d) H2 (1 atm), Lindlar catalyst, 1:1 EtOAc/EtOH, 3 h, 80%; (e) i—9-BBN, THF, 50 °C,
7 h; ii—HOAc, 60 °C, 14 h; iii—6 N NaOH/H2O2, 40 °C, 4 h, 50% conversion (1:1 alcohol/aldehyde); (f) Dess–Martin periodinane, CH2Cl2, 2 h,
quant.
O
c, d, e
OH
a, b
NC
NC
H
O
O
N
NCH3
OH
Scheme 3. Reagents and conditions: (a) I2, satd NaHCO3, 5 °C to rt, 18 h, 33%; (b) homopropargyl alcohol, Pd(PPh3)2Cl2, CuI, Et3N, THF, rt, 15 h,
quant.; (c) CH3SO2Cl, Et3N, CH2Cl2, 0 °C, 1.5 h, quant.; (d) N-methylpiperazine (10 equiv), CH3CN, rt, 16 h, 46%; (e) DIBAL, CH2Cl2, À60 °C to
rt, 6 h, 56%.
O
c
a, b
H
HO
N
HO
N
O
N
NH
NCH3
NCH3
Scheme 4. Reagents and conditions: (a) (Boc)2O, 1:1 THF/H2O, rt, 18 h, 99%; (b) LAH, Et2O, reflux, 64%; (c) 4-fluorobenzaldehyde, Cs2CO3, DMF,
90 °C, 18 h, 76%.

6046
A. Lee-Dutra et al. / Bioorg. Med. Chem. Lett. 16 (2006) 6043–6048
Graph 2. Displacement of [3H]histamine from recombinant histamine H4 receptor.
affinity suggests that some aspects of the compounds in
Graph 1 are not adequately represented in the confor-
mational model or in the rigidified compounds. The
chosen rigid linkers may be unable to accommodate
unknown structural requirements of the H4 receptor
that can be satisfied by a flexible linker, which can
more easily adapt its conformation. As mentioned
above, unambiguous determination of the bioactive
conformations of the compounds shown in Graph 1
was not possible due to the existence of multiple con-
formations satisfying the pharmacophore. The ether
oxygen, present in all of the linkages in Graph 1 but
absent in many of the subsequent linkages, may also
play a role. Interestingly, the benzofuran compound,
perhaps the most structurally similar constrained ana-
log to the compounds in Graph 1, does have modest
activity. Finally, the conformation of the linker can
be influenced by the substitution patterns on the cen-
tral aromatic ring, potentially introducing unfavorable
steric interactions between the benzene ring substituent
and the linker. This can prevent optimal positioning of
the piperazine for receptor binding, as exemplified by
the poor activity of the compounds incorporating alde-
hyde 8 (Graph 2).
The effect of substituents on the central ring is itself
quite interesting. Generally small lipophilic substituents
(e.g., chloro) maintained or improved binding efficiency
(Graph 2, aldehydes 1 and 2); however, increased substi-
tution, especially in the case of the di-methyl substituted
analog (Graph 2, aldehyde 11), led to a drastic loss in H4
receptor affinity. Modeling of the low energy conforma-
tions of substituted benzene rings elucidates a possible
reason for this observation (Fig. 2). For example, bis-
substitution (Fig. 2D) or the use of a carbon analog
for the ether linkage (Fig. 2C) appears to twist the linker
into an orthogonal position with respect to the central
benzene ring, leading to compounds with poor receptor
affinity. On the other hand, mono-substitution (Fig. 2B)
allows the alkyl ether linker to lie in the plane of the cen-
tral ring, enabling potent receptor binding.
In general, the SAR revealed in this library study does
not provide sufficient evidence to validate or invalidate
the proposed binding site of the piperazine component.
It may be that the model proposed above is sufficient to
explain the activity differences observed between the set
of closely related molecules in Graph 1, but unable to
account for the larger variations introduced by the rigid-
PhOMe
PhEt
CH3-PhOMe
(CH3)2-PhOMe
A
B
C
D
Figure 2. Modeling of low-energy conformations of simple substituted benzene rings.

A. Lee-Dutra et al. / Bioorg. Med. Chem. Lett. 16 (2006) 6043–6048
Table 1. Displacement of [3H]histamine from the recombinant histamine H4 receptor
6047
R1
R4 R5
R6
R2
R3
N
N
O
R7
N
H
Entry
R1
R2
R3
R4
R5
NR6R7a
Ki (nM)
1
2
CH3
CH3
H
CH3
H
C(CH3)3
H
Cl
Cl
Cl
Cl
H
H
A
A
A
A
B
46
22
93
28
250
26
65
26
9
CH3
H
H
3
H
4
–CH@CH–CH@CH–
H
H
H
5
C(CH3)3
H
Cl
Cl
Cl
CH3
H
6
H
C(CH3)3
C(CH3)3
C(CH3)3
CH3
H
H
C
A
A
C
C
7
H
H
H
8
H
H
H
9
10
CH3
CH3
H
CH3
Cl
Cl
H
H
1
a A = N-methylpiperazine; B = 3-dimethylaminopyrrolidine; C = N-methylhomopiperazine.
ified molecules. Empirically it can be reasonably con-
cluded that compounds comprised of a flexible linker
References and notes
1. Ash, A. S. F.; Schild, H. O. Br. J. Pharmacol. 1966, 27, 427.
2. Black, J. W.; Duncan, W. A. M.; Durant, C. J.; Ganellin,
C. R.; Parsons, E. M. Nature (London) 1972, 236, 385.
3. Arrang, J. M.; Garbarg, M.; Schwartz, J. C. Nature
(London) 1983, 302, 832.
4. de Esch, I. J. P.; Thurmond, R. L.; Jongejan, A.; Leurs, R.
Trends Pharmacol. Sci. 2005, 26, 462.
5. Fung-Leung, W.-P.; Thurmond, R. L.; Ling, P.; Karlsson,
L. Curr. Opin. Investig. Drugs 2004, 5, 1174.
6. O’Reilly, M.; Alpert, R.; Jenkinson, S.; Gladue, R. P.;
Foo, S.; Trim, S.; Peter, B.; Trevethick, M.; Fidock, M.
J. Recept. Signal Transduct. Res. 2002, 22, 431.
7. Hofstra, C.; Desai, P. J.; Thurmond, R. L.; Fung-Leung,
W.-P. J. Pharmacol. Exp. Ther. 2003, 305, 1212.
which can adopt the preferred conformation as a result
of appropriate choice of phenyl substituents will likely
result in potent H4 receptor ligands.
With these criteria in mind, additional analogs were
investigated in a non-library format. Substitution at
the 4-position of the benzimidazole, in addition to
the 5- and 6-positions, was found to be tolerated
(Table 1, entries 1 and 2). Larger groups, such as a
tert-butyl group or a fused benzene ring, could be
displayed on the benzimidazole ring without much ad-
verse effect (Table 1, entries 3 and 4). The N-methyl-
piperazine was replaced with pyrrolidine and
homopiperazine isosteres. The former demonstrated a
significant loss in activity, whereas the latter displayed
a slight increase in H4 receptor affinity with respect to
the parent compound (entries 5–7). Further analysis of
small lipophilic substitution on the central aromatic
ring revealed the methyl group as a suitable alterna-
tive to the previously identified chloro- and meth-
oxy-groups (entry 8).
8. Buckland, K. F.; Williams, T. J.; Conroy, D. M. Br. J.
Pharmacol. 2003, 140, 1117.
9. Thurmond, R. L.; Desai, P. J.; Dunford, P. J.; Fung-
Leung, W.-P.; Hofstra, C. L.; Jiang, W.; Nguyen, S.;
Riley, J. P.; Sun, S.; Williams, K. N.; Edwards, J. P.;
Karlsson, L. J. Pharmacol. Exp. Ther. 2004, 309, 404.
10. Ling, P.; Ngo, K.; Nguyen, S.; Thurmond, R. L.;
Edwards, J. P.; Karlsson, L.; Fung-Leung, W.-P. Br. J.
Pharmacol. 2004, 142, 1.
11. Venable, J. D.; Cai, H.; Chai, W.; Dvorak, C. A.; Grice,
C. A.; Jablonowski, J. A.; Shah, C. R.; Kwok, A. K.; Ly,
K. S.; Pio, B.; Wei, J.; Desai, P. J.; Jiang, W.; Nguyen, S.;
Ling, P. L.; Wilson, S. J.; Dunford, P. J.; Thurmond, R.
L.; Lovenberg, T. W.; Karlsson, L.; Carruthers, N. I.;
Edwards, J. P. J. Med. Chem. 2005, 48, 8289.
12. Jablonowski, J. A.; Grice, C. A.; Chai, W.; Dvorak, C. A.;
Venable, J. D.; Kwok, A. K.; Ly, K. S.; Wei, J.; Baker, S.
M.; Desai, P. J.; Jiang, W.; Wilson, S. J.; Thurmond, R.
L.; Karlsson, L.; Edwards, J. P.; Lovenberg, T. W.;
Carruthers, N. I. J. Med. Chem. 2003, 46, 3957.
13. Arienti, K. L.; Brunmark, A.; Axe, F. U.; McClure, K.;
Lee, A.; Blevitt, J.; Neff, D. K.; Huang, L.; Crawford, S.;
Pandit, C. R.; Karlsson, L.; Breitenbucher, J. G. J. Med.
Chem. 2005, 48, 1873.
14. Binding assay conditions are described in Ref. 9. Com-
pounds were tested in triplicate to generate Ki values.
15. In addition, a compound displaying a piperidine in place of
the N-methylpiperazine moiety, 5-chloro-2-[4-(3-piperidin-
1-yl-propoxy)-phenyl]-1H-benzoimidazole, was prepared
and found to display pKi < 5.5 for the human H4 receptor.
Based on this expanded SAR, 4-substituted benzimidaz-
oles with lipophilic central ring substitution and the
homopiperazine terminal diamine were prepared. These
compounds demonstrate single-digit nanomolar activity
(entries 9 and 10), with up to 100-fold improvement over
initial lead compound 1.
In summary, we have examined 2-arylbenzimidazoles
and identified compounds with low nanomolar affinity
for the H4 receptor. A pharmacophore was developed
to explain the sharp activity differences observed be-
tween different linker lengths and positions in otherwise
closely related compounds. Through library preparation
and analysis, this method was examined and the SAR of
three regions of the molecule was simultaneously ex-
plored. Additional analog preparation through tradi-
tional medicinal chemistry approaches facilitated
further SAR expansion.

6048
A. Lee-Dutra et al. / Bioorg. Med. Chem. Lett. 16 (2006) 6043–6048
16. Accelrys Inc., Catalyst, Release 4.7, San Diego: Accelrys
Inc., 2002.
Chemical Society: Washington D.C., 1979; Vol. 112, pp
205–226.
17. Molecular Operating Environment (MOE), version
2001.01; Chemical Computing Group Inc.: Montreal,
Canada, 2001.
18. Marshall, G. R.; Barry, C. D.; Bosshard, H. E.; Damm-
koehler, R. A.; Dunn, D. A. In Computer-Assisted Drug
Design; Olsen, E. C., Christoffersen, R. E., Eds.; American
19. Alkene isomerization was observed under the benzimid-
azole formation conditions used to access the trans-
alkene benzimidazoles. The trans- and cis-isomers were
isolated using reverse phase preparative HPLC, and the
individual products were evaluated for H4 receptor
binding.