Synthesis and SAR studies of benzimidazolone derivatives as histamine H3-receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2013.08.012

A novel series of benzimidazolone-containing histamine H ₃-receptor antagonists were prepared and their structure-activity relationship was explored. These benzimidazolone analogs demonstrate potent H₃-receptor binding affinities, no P450 enzyme inhibition, and strong H₃ functional activity. Compound 1o exhibits the best overall profile with H₃K_i = 0.95 nM and rat AUC = 12.9 μM h.

Full text of DOI:10.1016/j.bmcl.2013.08.012

Bioorganic & Medicinal Chemistry Letters xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and SAR studies of benzimidazolone derivatives
as histamine H3-receptor antagonists
Qingbei Zeng, Stuart B. Rosenblum, Zhaoxia Yang, Yueheng Jiang, Kevin D. McCormick,
Robert G. Aslanian, Luli Duguma, Joseph A. Kozlowski, Neng-Yang Shih, John A. Hey, Robert E. West Jr.,
⇑
Walter A. Korfmacher, Michael Berlin, Christopher W. Boyce
Merck Research Laboratories, 126 E. Lincoln Avenue, Rahway, NJ 07065, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online xxxx
A novel series of benzimidazolone-containing histamine H3-receptor antagonists were prepared and their
structure–activity relationship was explored. These benzimidazolone analogs demonstrate potent
H3-receptor binding affinities, no P450 enzyme inhibition, and strong H3 functional activity. Compound
Keywords:
1o exhibits the best overall profile with H3Ki = 0.95 nM and rat AUC = 12.9 lM h.
H3-antagonists
Benzimidazolone
P450 inhibition
Ó 2013 Elsevier Ltd. All rights reserved.
The H3 receptor is a G-protein coupled receptor in the histamine
receptor family. It was originally characterized as an autoreceptor
to regulate the synthesis and release of histamine from histaminer-
gic neurons.1 Thereafter it was also shown to mediate other neuro-
transmitters such as dopamine,2 serotonin,3 noradrenaline,4
GABA,5 and acetylcholine.6 Therefore, histamine H3 receptor li-
gands have been proposed as potential therapeutics for several
central nervous system (CNS) indications such as obesity, attention
deficit/hyperactivity disorder (ADHD), Alzheimer’s disease, and
schizophrenia.7
Known histamine H3 antagonists can be structurally divided
into imidazole (GT-2227, and SCH 79687) and nonimidazole-based
analogs (UCL 1972, Fig. 1).8 Although some of the imidazole-based
antagonists show good potency and acceptable exposure, they
have not advanced in the clinic because of their significant cyto-
chrome P450 (CYP) enzyme inhibition liability. SCH 79687 was
the first imidazole-based H3 antagonist lead from our laboratories.
While SCH 79687 showed excellent binding affinity, strong H3
functional activity, and good oral bioavailability,8c it also, unsur-
prisingly, significantly inhibited P450 enzymes and lowered cho-
lesterol levels in rat toxicology studies, and as a result, was not
advanced for development.
acceptable oral exposure in rats. In addition, many of them exhib-
ited unacceptable levels of CYP3A4 and CYP2D6 inhibition. This
communication details the structure–activity relationship (SAR)
of compounds of general structure 1 with significantly improved
enzyme profiles and pharmacokinetic properties.
The pharmacophore consists of four cyclic elements, that is, a
benzimidazol-2-one moiety, two piperidine rings, and a pyridine
heterocycle. The survey of a number of potential structural scaf-
folds on the right-hand side indicated that pyridine moiety is
essential to binding affinity. The key change was the introduction
of the 2-amino group in the pyridine which significantly improves
the P450 enzyme profile. With this discovery, we focused on the
modifications in the benzimidazol-2-one region to improve phar-
macokinetic profile.
The preparation of benzimidazol-2-one-based H3 antagonists of
type 1 is illustrated in Scheme 1. Commercially available amino-
pyridine 3 was protected as the di-N-tert-butoxycarbonyl (BOC)
derivative 4. Compound 4 was brominated under standard N-bro-
mosuccinimide (NBS) conditions to produce bromide 5, which, in
turn, was reacted with methyl isonipecotate to generate ester 6.
The hydrolysis of 6 with lithium hydroxide in tetrahydrofuran
(THF) and water provided lithium salt 7 for the further coupling
reactions without purification.
Diamines 8 were prepared conveniently from a substituted
nitrobenzene and ethyl 4-amino-1-piperidinecarboxylate accord-
ing to the literature method.10 When the diamines 8 were treated
with triphosgene in dichloromethane in the presence of triethyl-
amine at 0 °C, benzimidazolones 9 were obtained in good yields.
The available N-positions of benzimidazolones 9 were alkylated
under two different conditions. One method was the biphasic
In order to minimize the P450 enzyme liability, we decided to
explore a new class of nonimidazole analogs. In the previous com-
munication, a novel series of nonimidazole H3 antagonists (2,
Fig. 2) was reported.9 Although many of these antagonists showed
good in vitro activity, none of these compounds achieved an
⇑
Corresponding author.
E-mail address: christopher.boyce2@merck.com (C.W. Boyce).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.08.012
Please cite this article in press as: Zeng, Q.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.08.012

2
Q. Zeng et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
copper(II) chloride and triethylamine. Both methods gave accept-
able yields. The subsequent deprotection of 10 was completed with
sodium hydroxide in water under reflux to produce 11, which was
coupled with lithium salt 7 to produce 12. The coupling reaction
was conducted with HATU and 4-dimethylaminopyridine (DMAP)
N
H
N
H
N
N
l
C
N
H
N
H
O
GT-2227
SCH 79687
l
C
N
O
N
O2
Table 1
UCL 1972
H3 binding affinity (Ki), enzyme (3A4 and 2D6) inhibition and functional activity (pA2)
of compounds 111
Figure 1. Representative H3 antagonists.
O
Y
N
N
N
O
NH2
N
O
N
N
N
N
N
R
1
N
N
N
O
c
Compound
R
Y
H3
IC50
3A4
IC50
2D6
pA2
N
N
2
R
R
Ki
(nM)a
(l
M)b
(l
M)b
O
1a
1b
H
H
H
0.6
4.7
>30
>30
>30
>30
N/A
8.1
Y
F
3C
NH2
N
O
l
C
C
1c
H
77
N/A
N/A
N/A
1
l
Figure 2.
1d
H
420
N/A
N/A
N/A
reaction of 9 and an appropriate alkyl chloride in toluene in the
presence of a base, such as a mixture of NaOH, potassium carbon-
ate and phase-transfer reagent, such as tetra-n-butyl ammonium
bisulfate. The other method involved using an aryl boronic acid
to prepare N-aryl substituted analogs. This reaction was conve-
niently carried out at room temperature in the presence of
F
O
1e
1f
H
H
1.5
0.7
>30
>30
>30
8.0
8.7
O
O
ꢀ30
N
1g
F
2.3
>30
>30
N/A
N
F
F
3C
3C
1h
1i
F
F
3.8
7.6
ꢀ30
>30
>30
8.9
8.6
( )
2
N
( )
3
( )
N
1
>30
N
Br
O
N(BOC)2
N
NH2
N
N(BOC)2
1j
F
F
F
1.5
2.2
5.1
>30
ꢀ30
22
>30
>30
>30
8.7
8.7
8.2
5
4
3
1k
1l
O
( )
4
O
N
LiO
N
N(BOC)2
N(BOC)2
1m
1n
F
F
19
>30
>30
>30
>30
7.9
8.3
7
6
F
F
O
O
2.8
Y
Y
O
O
( )
6
N
N
( )
5
F
N
H
N
O
HN
Y
1o
F
0.95
>30
>30
8.7
NH2
8
9
F
O
1p
1q
F
F
2.5
4.0
>30
>30
>30
>30
7.8
8.0
Y
N
( )
8
O
NH
N
( )
7
N
N
O
N
O
N
N
N
N
N
10
R
R
11
O
1r
F
0.5
>30
>30
N/A
Y
N
N
N
( )
9
N
N(BOC)2
Not determined.
N
1
Inhibition of [3H]-N-
a-methylhistamine binding to guinea pig brain receptor.
H3 binding Ki values are the average of at least two independent determinations.
Variation from the mean is generally within 50%.
Human liver microsome assay. IC50 values are the average of at least three
independent determinations. The assay-to-assay variability is 15%.
a
R
O
12
b
Scheme 1. Reagents and conditions: (1) (BOC)2O, Et3N, DMAP, CH2Cl2; (2) NBS,
benzoyl peroxide, CCl4; (3) methyl isonipecotate, triethylamine, DMF; (4) LiOH,
THF, H2O; (5) triphosgene, Et3N, CH2Cl2, 0 °C; (6) alkyl chloride, (n-Bu)4NHSO4,
NaOH–K2CO3, totuene, or aryl boronic acid, CuCl2, triethylamine, CH2Cl2. (7) NaOH,
water, reflux; (8) 7, HATU, DMAP, DMF; (9) HCl, 1,4-dioxane.
c
Antagonist potency in an electrically stimulated guinea pig ileum. pA2 values
are the average of at least four independent determinations. The assay-to-assay
variability is 0.2.15
Please cite this article in press as: Zeng, Q.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.08.012

Q. Zeng et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
3
Table 2
AUC of 12.9 lM h. In order to examine if a more polar heteroaryl
Rat oral AUC data of selected compounds16
substituent was favorable to H3 affinity, compounds 1p and 1r
were prepared. Interestingly, these compounds maintained high
affinity (0.5–4.0 nM). Compound 1p also displayed excellent plas-
ma concentration, while 1r exhibited slightly decreased oral PK
compared to 1o.
In summary, a novel series of H3 receptor antagonists were opti-
mized by modification of the benzimidazolone N-position. Lipo-
philicity of the substituent played an important role in the
plasma levels of the molecule. As the best compounds in the series,
both 1o and 1p showed good H3 affinity. In addition, they dis-
played significantly improved rat oral pharmacokinetics, which
supports our hypothesis that optimum lipophilicity could help
the absorption. Furthermore, the new version of the series showed
no inhibition of P450 enzymes and good potency in the functional
assay.
Compound
Rat AUC (
l
M h)
Concentration @ 6 hour (lM)
1e
1f
ig
1h
1j
1k
1l
1n
1o
1p
1r
0.1
0
0
0
0
0
4.1
1.2
3.5
1.8
5.1
12.9
10.9
3.7
0.51
0.13
0.26
0.20
0.68
1.85
1.2
0.50
in dimethylformamide (DMF). The final deprotections were real-
ized using HCl in 1,4-dioxane at room temperature to afford the
target compounds 1.11
References and notes
The histamine H3 receptor binding affinity (Ki) of the com-
pounds was evaluated in a radioligand binding assay in guinea-
pig brain homogenates utilizing the selective H3 histamine agonist
1. (a) Arrang, J.-M.; Garbarg, M.; Schwartz, J.-C. Nature 1983, 302, 832; (b) Arrang,
J.-M.; Garbarg, M.; Lancelot, J.-C.; Lecomte, J.-M.; Pollard, H.; Robba, M.;
Schunack, W.; Schwartz, J.-C. Nature 1987, 327, 117.
2. Schlicker, E.; Fink, K.; Detzner, M.; Goethert, M. J. Neural Transm. 1993, 93, 1
[Gen. Sect.].
[3H]-N- -methyl histamine as the radioligand.12,1b We began the
a
studies from the parent compound 1a (Table 1). We were pleased
to see that 1a showed a Ki value of 0.6 nM. Next, we rapidly exam-
ined the tolerability of an alkyl group at the distal N-position.
When 4,4,4-trifluorobutyl group was incorporated, the resulted
antagonist 1b showed slightly reduced affinity. The increase in
bulkiness (1c and 1d) in this region substantially reduced affinity.
In particular, the affinity was very sensitive to the steric effect from
the carbon connected to the N-position.13 In contrast, the external
alkyl chain with a more polar substituent in 1e or 1f helped the
compound to maintain good affinity. These results suggested that
a polar substituent in the remote position favored binding affinity.
The most active compounds 1e and 1f were then dosed orally to
Sprague–Dawley rats; plasma levels of both were negligible
(Table 2). We reasoned that the poor pharmacokinetic profiles of
these compounds could be due to low lipophilicity (ClogP = 0.41
and 0.33, respectively),14 which limited their absorption. Further
SAR study was undertaken with a more lipophilic alkyl group on
the left-hand side. Additionally, a small and electron-withdrawing
group, such as fluoro, was introduced to the benzimidazolone core
to reduce electron density in this region (1g–r). Not surprisingly,
compound 1g (ClogP = 0.60) displayed low plasma levels due to
its low lipophilicity, while the more lipophilic 1h (ClogP = 1.51)
did show a significantly improved pharmacokinetic (PK) profile
(Table 2). We were encouraged by this result and prepared a few
structurally similar compounds (1i–m). As shown in the Table 1,
other alkyl groups such as 2,2,2-trifluoroethyl (1j), cyclopropym-
ethyl (1k), and cyclopentyl (1l) helped maintain good activity.
The larger cyclohexylmethyl (1m) was not tolerated at this posi-
tion. Compounds 1k and 1l displayed moderate plasma concentra-
tions when dosed to rats (Table 2).16
3. Fink, K.; Schlicker, E.; Neise, A.; Goethert, M. Naunyn Schmiedebergs Arch.
Pharmacol. 1990, 342, 513.
4. (a) Schlicker, E.; Fink, K.; Hinterthaner, M.; Goethert, M. Naunyn Schmiedebergs
Arch. Pharmacol. 1989, 340, 633; (b) Molderings, G. J.; Weibenborn, G.;
Schlicker, E.; Likungu, J.; Goethert, M. Naunyn Schmiedebergs Arch. Pharmacol.
1992, 346, 46.
5. Garcia, M.; Floran, B.; Arias-Montano, J. A.; Young, J. M.; Aceves, J. Neuroscience
1997, 80, 241.
6. Clapham, J.; Kilpatrick, G. J. Br. J. Pharmacol. 1992, 107, 919.
7. (a) Stark, H.; Schlicker, E.; Schunack, W. Drugs Future 1996, 21, 507; (b) Leurs,
R.; Blandina, P.; Tedford, C.; Timmerman, H. Trends Pharmacol. Sci. 1998, 19,
177; (c) Mcleod, R. L.; Egan, R. W.; Cuss, F. M.; Bolser, D. C.; Hey, J. A. Prog.
Respir. Res. 2001, 31, 133 (New Drugs for Asthma, allergy and COPD); (d) Berlin,
M.; Boyce, C. W.; De Lera Ruiz, M. J. Med. Chem. 2011, 54, 26.
8. (a) Tedford, C. E.; Hoffmann, M.; Seyedi, N.; Maruyama, R.; Levi, R.; Yates, S. L.;
Ali, S. M.; Phillips, J. G. Eur. J. Pharmacol. 1998, 351, 397; (b) Ganellin, C. R.;
Leurquin, F.; Piripitsi, A.; Arrang, J. M.; Garbarg, M.; Ligneau, X.; Schunack, W.;
Schwartz, J. C. Arch. Pharm. 1998, 331, 395 (Weinhein, Ger.); (c) McLeod, R. L.;
Rizzo, C. A.; West, R. E., Jr.; Aslanian, R.; McCormick, K.; Bryant, M.; Hsieh, Y.;
Korfmacher, W.; Mingo, G. G.; Varty, L.; Williams, S. M.; Shih, N.-Y.; Egan, R. W.;
Hey, J. A. J. Pharmacol. Exp. Ther. 2003, 305, 1037.
9. Ting, P. C.; Lee, J. F.; Albanese, M. M.; Wu, J.; Aslanian, R.; Favreau, L.; Nardo, C.;
Korfmacher, W. A.; West, R. E.; Williams, S. M.; Anthes, J. C.; Rivelli, M. A.;
Corboz, M. R.; Hey, J. A. Bioorg. Med. Chem. Lett. 2010, 20, 5004.
10. Henning, R.; Lattrell, R.; Gerhards, H. J.; Leven, M. J. Med. Chem. 1987, 30, 814.
11. 1H and MS data was consistent with the chemical structures shown. For 1o,
MS: [M+H]+m/z 565. 1H NMR (300 MHz, CDCl3) d 8.00 (d, J = 5.5 Hz, 1H), 7.12
(dd, J = 2.2, 7.7 Hz, 2H), 7.05 (dd, J = 4.4, 8.8 Hz, 1H), 6.91 (m, 3H), 6.63 (dd,
J = 1.1, 5.5 Hz, 1H), 6.55 (s, 1H), 4.93 (m, 1H), 4.57 (m, 1H), 4.42 (s, 2H), 4.13 (m,
1H), 3.42 (s, 2H), 3.23 (m, 1H), 2.95 (m, 2H), 2.69 (m, 1H), 2.56 (m, 1H), 2.34 (m,
2H), 2.03 (m, 8H).
12. West, R. E., Jr.; Zweig, A.; Shih, N. Y.; Siegel, M. I.; Egan, R. W.; Clark, M. A. Mol.
Pharmacol. 1990, 38, 610.
13. Removal of the benzylic methyl in
R resulted in increased affinity (e.g.,
H3Ki = 12 nM for R = pCl–benzyl, Y = H and H3Ki = 6 nM for R = benzyl, Y = F;
compounds not shown).
14. ClogP values were calculated in Chemdraw (algorithm developed by BioByte).
15. (a) Rizzo, C. A.; Tozzi, S.; Monahan, E. M.; Hey, J. A. Eur. J. Pharmacol. 1995, 294,
329; (b) Valentine, A. F.; Rizzo, C. A.; Rivelli, M. A.; Hey, J. A. Eur. J. Pharmacol.
1999, 366, 73.
16. The oral pharmacokinetics evaluations were done using the ‘cassette-
accelerated rapid rat screen’ (CARRS) approach (Sprague–Dawley rats dosed
with 10 mg/kg of compound in 20% HPBCD) and their oral AUCs were
calculated for the 0–6 h intervals. See (a) Korfmacher, W. A.; Cox, K. A.; Ng,
K. J.; Veals, J.; Hsieh, Y.; Wainhaus, S.; Broske, L.; Prelusky, D.; Nomeir, A.;
White, R. E. Rapid Commun. Mass Spectrom. 2001, 15, 335; (b) Mei, H.;
Korfmacher, W.; Morrison, R. AAPS J. 2006, 8, E493.
The lipophilicity of analogs could be further increased by an
N-aryl substituent. Thus, we decided to continue the SAR studies
by incorporating a 3,4-difluorophenyl group in this region. We
were very satisfied to find that 1n displayed very good affinity
(Ki = 2.8 nM) and rat PK (rat AUC = 5.1 lM h at a dose of 10 mg/
kg). The binding affinity and PK were even better when a 3,5-
difluorophenyl substituent was attached to the N-position (1o).
This compound showed binding affinity of 0.95 nM and rat oral
Please cite this article in press as: Zeng, Q.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.08.012