Identification of a novel, orally bioavailable histamine H3 receptor antagonist based on the 4-benzyl-(1H-imidazol-4-yl) template

Article abstract of DOI:10.1016/S0960-894X(02)00055-0

A novel series of histamine H₃ receptor antagonists, based on the 4-benzyl-(1H-imidazole-4-yl) template, incorporating urea and carbamate linkers has been prepared. Compound 3j is a selective H₃ antagonist and demonstrates excellent

Full text of DOI:10.1016/S0960-894X(02)00055-0

Bioorganic & Medicinal Chemistry Letters 12 (2002) 937–941
Identification of a Novel, Orally Bioavailable
Histamine H3 Receptor Antagonist Based on the
4-Benzyl-(1H-imidazol-4-yl) Template
Robert Aslanian,* Mwangi W. Mutahi, Neng-Yang Shih, Kevin D. McCormick,
John J. Piwinski, Pauline C. Ting, Margaret M. Albanese, Michael Y. Berlin,
Xiaohong Zhu, Shing-Chun Wong, Stuart B. Rosenblum, Yueheng Jiang,
Robert West, Susan She, Shirley M. Williams, Matthew Bryant
and John A. Hey
The Schering Plough Research Institute, 2015 Galloping Hill Road, Kenilworth, NJ 07033, USA
Received 16 November 2001; accepted 17 January 2002
Abstract—A novel series of histamine H3 receptor antagonists, based on the 4-benzyl-(1H-imidazole-4-yl) template, incorporating
urea and carbamate linkers has been prepared. Compound 3j is a selective H3 antagonist and demonstrates excellent oral plasma
levels in the rat and monkey. # 2002 Elsevier Science Ltd. All rights reserved.
Allergic rhinitis is a debilitating disease that affects 10–
30% of the US population.1 Although the sneezing,
rhinorrhia, and pruritus associated with this disease are
adequately treated with H1 antihistamines, the nasal
congestion that often accompanies it is not.2 Current
therapies for the treatment of congestion include oral
decongestants such as pseudoephedrine, topical decon-
gestants such as oxymetazoline, and nasal steroids.
These treatments are effective, but all have drawbacks.
Therefore, new methods for the treatment of nasal con-
gestion should be an important therapeutic advance.
receptors that regulate the release of norepinephrine,
an endogenous neurotransmitter that maintains vas-
cular tone. H3 blockade should reestablish the release of
norepinephrine and result in vasoconstriction (decon-
gestion). This hypothesis has been demonstrated in a
histamine-driven cat model of nasal congestion.7 Based
on this data, we have undertaken a project to discover a
novel, selective H3 antagonist that can be used in com-
bination with an H1 antihistamine for the treatment of
the nasal congestion associated with seasonal or per-
ennial allergic rhinitis. This paper describes the syn-
thetic efforts that have led to the identification of a
novel, orally bioavailable H3 antagonist based on the
4-benzyl-(1H-imidazol-4-yl) scaffold.8
Recent evidence supports the idea that the combination
of an H1 antagonist with an H3 antagonist acts as a
nasal decongestant.3 During a nasal allergic reaction,
the actions of histamine released from mast cells are not
blocked by H1 antihistamines alone, which primarily
prevent plasma extravization and mucus secretion. In
the periphery, H3 receptors found on sympathetic
nerves modulate sympathetic neurotransmission.4À6
Mast cell derived histamine may contribute to nasal
congestion by promoting vascular engorgement (i.e.,
vasodilatation) through activation of prejunctional H3
Initial efforts from our lab identified the novel amidine 1
as a potent and selective H3 antagonist in vitro using
guinea pig brain membranes.8 Additionally, 1 was active
in vivo in a guinea pig model9 when dosed intravenously
(ED50=0.3 mg/kg).
*Corresponding author. Fax: +1-908-740-7152; e-mail: robert.
aslanian@spcorp.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00055-0

938
R. Aslanian et al. / Bioorg. Med. Chem. Lett. 12 (2002) 937–941
Table 1. H3 binding affinity of compounds 3 and 5
However, when dosed orally in the rat, the plasma AUC
of 1 was zero. We reasoned that the poor plasma levels
observed with 1 might be due to its dibasic nature,
which could potentially decrease its absorption, and
that removal of one of the basic groups might lead to
improved pharmacokinetics. Because the 4-substituted
imidazole ring is usually necessary for good H3 activ-
ity,10 we focused our efforts on replacing the amidine
moiety with neutral linkers. A survey of a number of
potential linkers and chain lengths led to the identifica-
tion of the ureas 3 and carbamates 5 as moieties that
conferred good H3 potency and novelty. Members of
the structural classes 3 and 5 were prepared and tested
for H3 binding affinity and oral pharmacokinetics.
Example no.
X
R
H
Ki (nM)a
3a
5b
NH
O
NAb
37
3c
5c
NH
O
21
25
3d
5d
NH
O
6
7
Scheme 1 outlines the synthesis of the urea and carba-
mate analogues 3 and 5. The synthesis of the key inter-
mediate
2 has been previously described (Tr is
3e
3f
NH
NH
32
18
triphenylmethyl).8 Reduction of the nitrile of 2 using
Raney-nickel gave the primary amine. Reaction of the
amine with a variety of isocyanates and deprotection
gave the ureas 3. The carbamate analogues 5 were
prepared from 2 via hydrolysis of the nitrile to the
carboxylic acid followed by esterification. The ester
was reduced to the carbinol and reacted with an iso-
cyanate to give the carbamate 5 after deprotection.
Compounds were evaluated for H3 binding affinity
using guinea pig brain membranes as described by
Korte et al.11
3g
3h
NH
NH
5
16
Table 1 gives the structures and H3 binding affinity for
urea analogues 3 and carbamate analogues 5.11 The
initial conclusion to be drawn from these data is that a
urea or carbamate linker is a suitable replacement for
the amidine group in 1. Therefore, in agreement with
the observation of others,12 a basic moiety in the side
chain of this series of H3 antagonists is not necessary for
good H3 binding. This observation is consistent with a
recently proposed pharmacophore model in which the
amidine moiety of a homologue of 1 did not interact
with two putative hydrogen-bonding sites on the H3
receptor.13 Apparently, it is not so much the basic nat-
ure of the linker that matters as the spacial orientation it
imposes on the molecule. As further corroboration of
this hypothesis, both the urea and carbamate linkers
generally display similar H3 binding profiles for the
same R group.
3i
5i
NH
O
7
18
3j
5j
NH
O
4
5
3k
NH
9
22
3l
3m
3n
NH
NH
NH
NA
90
The nature and position of the substituents around the
aromatic ring of the carbamate and urea also play a role
in the binding profile of these analogues. Compounds
that lackan aromatic group R or in which the aromatic
ring is unsubstituted or ortho substituted were less active
(e.g., 3a,e and 5b). Optimum binding was obtained with
substituents in the 3- or the 3,5-positions of the aromatic
ring for both the urea and carbamate analogues (e.g., 3d,
5d, 3g, 3i, 3j, 5j, and 3p). Spacing the aromatic ring an
additional one or two carbons from the linker decreased
potency (e.g., 3m, 3n, and 3o). Heterocyclic substitution
was also unfavorable if the heterocycle was unsub-
stituted (e.g., 3p and 3q). However, introduction of the
3,5-substitution pattern on the heterocycle reestablishes
a good binding profile (e.g., 3r).
3o
3p
NH
NH
430
NA
3q
3r
NH
NH
270
3
aH3 binding Ki values are the average of at least two independent
determinations. The assay-to-assay variation was generally Æ2-fold.
The average Ki for the standard H3 antagonist thioperamide in this
assay is 7.3Æ0.7 nM.
bNA=Less than 50% inhibition when screened at 1 mg/mL.11

R. Aslanian et al. / Bioorg. Med. Chem. Lett. 12 (2002) 937–941
939
Scheme 1. Reagents and conditions: (a) H2, Ra-Ni, MeOH/NH3, 85%; (b) RNCO, pyridine; (c) 1 N HCl/MeOH; (d) 2 N NaOH, EtOH; (e) H2SO4,
EtOH; (f) trityl chloride, Et3N, 57% for steps d, e, and f; (g) DIBAl-H, THF, 94%. See Table 1 for the definition of R.
Having established the 3-substitution or 3,5-disubstitu-
tion patterns as optimal on the aromatic ring of the urea
or carbamate, and aniline analogues as superior to the
corresponding benzyl or phenethyl derivatives, we next
explored the SAR of the core region of this structure.
We chose the potent urea analogue 3j as the template
for this investigation and prepared analogues 7, 9,14 12,
and 14 (Schemes 2–5).
Scheme 4. (a) EtMgBr, CH2Cl2, 4-NO2C6H4CHO, 75%; (b) Ac2O,
pyridine, CH2Cl2 (c) H2, Pd(OH)2/C, HOAc 50% for steps b and
c; (d) HCl/ MeOH; (e) 3,5-Cl2C6H4NCO, THF, 66% for steps d
and e.
Scheme 2. (a) NaH, THF, CH3I, 96%; (b) 1 N HCl, MeOH, 83%.
Scheme 5. (a) EtMgBr, CH2Cl2, 4-NCC6H4CHO, 66%; (b) H2, Ra-Ni,
NH3/MeOH, 99%; (c) 3,5-Cl2C6H4NCO, THF, 50%; (d) NaI,
Me2SiCl2, acetone/CH2Cl2, 54%; (e) maleic acid, MeOH/CH2Cl2,
92%.
Scheme 3. (a) H2, Ra-Ni, NH3/MeOH; (b) 3,5-Cl2C6H4NCO, CH2Cl2,
25% for two steps; (c) HCl/dioxane, 81%.
In general, the core region is not amenable to structural
change. Methylation of the urea nitrogens led to a
decrease in binding affinity (7, Ki=370 nM). Elongation
of the carbon chain between the imidazole and phenyl
rings (9) or truncation of the chain between the central
phenyl ring and the urea nitrogen (12) also led to
decreased binding affinity (Ki=120 nM and 36% inhi-
bition at 1 mg/mL respectively11). Shifting the point of
attachment on the central phenyl ring from a 1,4-con-
figuration to a 1,3-configuration also had a negative
impact on activity (14, 33% inhibition at 1 mg/mL11).
These results are consistent with those seen for similar
Scheme 6. (a) CH3I, CH2Cl2/MeOH, Et3N, 40 ꢀC, 19%.
compounds in the amidine series.8

940
R. Aslanian et al. / Bioorg. Med. Chem. Lett. 12 (2002) 937–941
Table 2. Pharmacokinetic parameters for 3j
Rat
Monkey
Dose (mg/kg)a
AUC (mg h/mL)
Cmax (mg/mL)
t1/2 (h)
Bioavailability
10
3
12.6
1.7
4.4
44%
.
18.1
1.5
—
—
aCrystalline 3j dosed in methyl cellulose.
Scheme 7. (a) AlCl3, dichloroethane, 4-NCC6H4COCl, 54%; (b) H2,
Ra-Ni, MeOH/NH3, 49%; (c) 3,5-Cl2C6H4NCO, THF, 73%; (e)
NaBH4, MeOH/THF, 99%; (e) Me2SiCl2, NaI, CH2Cl2/acetone, 72%;
(f) NaOH, MeOH/dioxane/H2O, 69%.
Based on its favorable H3 binding affinity, 3j was fur-
ther screened against other G-protein coupled receptors,
including the H1, H2, and M1À5 receptors and was inac-
tive at the highest dose tested (1 mM).17 It did however
show a2a activity (207 nM), dopamine uptake antagon-
ism (805 nM) and imidazoline I2 activity (155 nM). The
oral pharmacokinetics of 3j were ascertained in the rat
and monkey (Table 2). Unlike the amidine 1, the urea 3j
exhibited very good pharmacokinetics in both the rat
and monkey.
We next turned our attention to the imidazole moiety.
Despite some recent, notable exceptions,15 it is generally
true that a 4-substituted imidazole ring is necessary for
good H3 binding affinity and that substitution of the
ring or replacement with other heterocycles leads to loss
of activity.10 This was indeed the case with this series as
substitution of the imidazole nitrogens with methyl (15a
and b, Scheme 6)16 or replacement of the imidazole ring
by other heterocycles such as pyrrole 18 (Scheme 7),
pyrazole 21 (Scheme 8) or the furan moiety found in the
H2 receptor antagonist ranitidine, 24 (Scheme 9), abol-
ished activity.
In conclusion, exchanging the basic amidine linker of 1
with a neutral urea or carbamate linker led to the iden-
tification of a new series of H3 receptor antagonists
exemplified by the urea 3j. Compound 3j is a potent and
selective H3 receptor antagonist showing no activity
against the H1 or H2 receptors. More significantly, 3j
displays good oral pharmacokinetics in the rat and
monkey, which supports the hypothesis that the basic
amidine moiety was the source of the poor pharmaco-
kinetic profile seen in 1. The full biological profile of 3j
will be published in due course.
Acknowledgements
The authors would like to thank the Drug Metabolism
and Pharmacokinetics group for determining the phar-
macokinetic profile of compounds 1 and 3j.
Scheme 8. (a) t-BuLi, THF/Et2O, 4-NCC6H4CHO, 55%; (b) H2, Ra-
Ni, MeOH/NH3, 91%; (c) 3,5-Cl2C6H4NCO, THF, 88%; (d)
Me2SiCl2, NaI, CH2Cl2/acetone, 53%.
References and Notes
1. Galant, S. P.; Wilkinson, R. Formulary 1999, 34, 1016.
2. Spector, S. J. Aller. Clin. Immunol. 1999, 103, S386.
3. McLeod, R. L.; Egan, R. W.; Cuss, F. M.; Bolser, D. C.
Hey, J. A. In Progress in Respiratory Research; Hansel, T. T.,
Barnes, P. J., Eds.; Karger: Basel, 2001; Vol. 31, p 133.
4. Koss, M. C.; Hey, J. A. Naunyn Schmiedeberg’s Arch. Par-
macol. 1992, 346, 208.
5. McLeod, R. L.; Gertner, S. B.; Hey, J. A. Br. J. Pharmacol.
1993, 110, 553.
6. McLeod, R. L.; Gertner, S. B.; Hey, J. A. Gen. Pharmacol.
1996, 27, 1001.
7. McLeod, R. L.; Mingo, G. G.; Herczku, C.; DeGennaro-
Culver, F.; Kreutner, W.; Egan, R. W.; Hey, J. A. Am. J.
Rhinol. 1999, 13, 391.
8. Aslanian, R.; Brown, J. E.; Shih, N.-Y.; Mutahi, M. wa;
Green, M. J.; She, S.; Del Prado, M.; West, R.; Hey, J. A.
Bioorg. Med. Chem. Lett. 1998, 8, 2263.
Scheme 9. (a) HOCH2CH2OH, p-TSA, toluene, 95%; (b) Mg, THF,
5-dimethylaminomethylfuraldehyde, 19%; (c) HCl/H2O/MeOH; (d)
.
NH2OH HCl, MeOH, 36% for steps c and d; (e) H2, Ra-Ni, EtOH,
90% crude; (f) 3,5-Cl2C6H4NCO, THF, 36%; (g) TFA, Et3SiH,
CH2Cl2, 65%.

R. Aslanian et al. / Bioorg. Med. Chem. Lett. 12 (2002) 937–941
941
9. Hey, J. A.; Del Prado, M.; Egan, R. W.; Kreutner, W.;
Chapman, R. W. Br. J. Pharmacol. 1992, 107, 347.
10. De Esch, I. J.; Timmerman, H.; Menge, W. M. P. B.;
Nederkoorn, P. H. J. Arch. Pharm. Pharm. Med. Chem. 2000,
333, 254.
11. Korte, A. K.; Myers, J.; Shih, N.-Y.; Egan, R. W.; Clark,
M. A. Biochem. Biophys. Res. Commun. 1990, 168, 979. Com-
pounds are screened at a concentration of 1 mg/mL and those
demonstrating greater than 50% inhibition are submitted for a
Ki determination.
12. See for example Stark, H.; Sadek, B.; Krause, M.; Huls,
A.; Ligneau, X.; Ganellen, C. R.; Arrang, J.-M.; Schwartz, J.-
C.; Schunack, W. J. Med. Chem. 2000, 43, 3987.
13. De Esch, I. J. P.; Mills, J. E. J.; Perkins, T. D. J.; Romeo,
G.; Hoffmann, M.; Wieland, K.; Leurs, R.; Menge, W. M. P. B.;
Nederkoorn, P. H. J.; Dean, P. M.; Timmerman, H. J. Med.
Chem. 2001, 44, 1666.
14. Compounds 8 and 10 are prepared according to ref 8 and
Aslanian, R. G.; Green, M. J.; Shih, N.-Y. U.S. Patent
5,578,616 and Aslanian, R. G. U.S. 5,990,147.
15. (a) Ganellin, C. R.; Leurquin, F.; Piripitsi, A.; Arrang, J.-
M.; Garbarg, M.; Ligneau, X.; Schunack, W.; Schwartz, J.-C.
Arch. Phar. Med. Chem. 1998, 331, 395. (b) Meier, G.;
Apelt, J.; Reichert, U.; Graßmann, S.; Ligneau, X.; Elz, S.;
Leurquin, F.; Ganellin, C. R.; Schwartz, J.-C.; Schunack, W.;
Stark, H. Eur. J. Pharm. Sci. 2001, 13, 249. (c) Linney, I. D.;
Buck, I. M.; Harper, E. A.; Kalindjian, S. B.; Pether, M. J.;
Shankley, N. P.; Watt, G. F.; Wright, P. T. J. Med. Chem. 2000,
43, 2362.
16. The imidazole isomers 15a and 15b are separated using
silica gel chromatography eluting with 90% CH2Cl2/10%
MeOH saturated with NH3.
17. Screened by MDS Panlabs, PO Box 26–127, Taipei,
Taiwan (106), ROC.