New benzimidazoles as thrombopoietin receptor agonists

Article abstract of DOI:10.1016/j.bmcl.2005.11.096

A novel benzimidazole series of small-molecule thrombopoietin receptor agonists has been discovered. Herein, we discuss the preliminary exploration of structure-activity relationships within this chemotype.

Full text of DOI:10.1016/j.bmcl.2005.11.096

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1212–1216
New benzimidazoles as thrombopoietin receptor agonists
Igor G. Safonov,* Dirk A. Heerding, Richard M. Keenan, Alan T. Price,
Connie L. Erickson-Miller, Christopher B. Hopson, Jenna L. Levin, Kenneth A. Lord
and Peter M. Tapley
Medicinal Chemistry and Oncology Research Departments, Microbial, Musculoskeletal and Proliferative Diseases,
GlaxoSmithKline Pharmaceuticals, 1250 S. Collegeville Road, Collegeville, PA 19426, USA
Received 7 November 2005; accepted 22 November 2005
Available online 11 January 2006
Abstract—A novel benzimidazole series of small-molecule thrombopoietin receptor agonists has been discovered. Herein, we discuss
the preliminary exploration of structure–activity relationships within this chemotype.
ꢀ 2005 Elsevier Ltd. All rights reserved.
Thrombopoietin (TPO) is a member of an extensive
family of extracellular signaling proteins, collectively
termed cytokines.1 TPO plays a critical role in the pro-
duction of platelets by stimulating the proliferation of
megakaryocyte progenitor cells into fully mature mega-
karyocytes, which then undergo fragmentation to pro-
duce platelet bodies.2 Platelets are essential in the
process of blood clotting and repair of damaged blood
vessels. Thrombocytopenia, or low circulating platelet
levels, can be a debilitating condition associated with a
range of medical disorders, and is also a common side
effect in cancer patients receiving intensive chemothera-
py treatment. Current methods to manage thrombocy-
topenia suffer from drawbacks such as limited efficacy,
serious side effects, and inconvenient routes of adminis-
tration.3 Therefore, an orally bio-available non-peptide
TPO receptor agonist may be beneficial to patients
suffering from thrombocytopenia.
A
B
Acidic
HO3S
HO2C
Acidic
NH
NH
N
N
OH
Chelate
Chelate
HO
N
Lipophilic
Lipophilic
R
R
Figure 1. The design of naphtho[1,2-d]imidazole (A) and benzimid-
azole (B) templates.
fied classes of TPO mimics.5c Figure 1 illustrates the pro-
posed key pharmacophore elements required for potent
TPO agonist activity: (1) a lipophilic domain and (2) an
acidic moiety on the opposite end of the molecule separat-
edby(3)aheteroatom metalchelate in the centralportion.
The identification of small molecule mimics of hemato-
poietic growth factors is an active area of research4 and
several non-peptidyl agonists have been reported.5
Recently, a rationally designed series of naphtho[1,2-
d]imidazole TPO receptor agonists, exemplified by struc-
ture A (Fig. 1), have been described.5c There now exists a
large body of evidence for a shared pharmacophore
hypothesis between this series and the previously identi-
In our ongoing effort to identify additional small-mole-
cule TPO mimics based on the existing pharmacophore
model, we have initiated a study directed toward further
optimization of the naphtho[1,2-d]imidazole class, with
the major emphasis on improving potency and efficacy.
We immediately realized that deletion of the terminal
aromatic ring represents an obvious and logical simplifi-
cation of the existing scaffold. As can be seen in Figure
1, the resultant benzimidazole template (B) still incorpo-
rates the three pharmacophore requirements (acidic–
chelate–lipophilic), although the introduced structural
Keywords: Thrombopoietin; Benzimidazole; Receptor agonist; Struc-
ture–activity relationships.
*
Corresponding author. Tel.: +1 610 917 6567; fax: +1 610 917
4206; e-mail: igor.2.safonov@gsk.com
0960-894X/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.11.096

I. G. Safonov et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1212–1216
1213
CO2Me
changes inevitably necessitate migration of the acidic
moiety closer to the chelate domain and reduce the max-
imum number of chelatable heteroatoms from three to
two. Furthermore, the new framework benefits from
elimination of a potential metabolic liability of the phe-
nol and allows for ultimate refinements of developability
characteristics without a significant increase in molecu-
lar weight.
CO2Me
NHAc
23
CO2Me
N
a, b
c
N
NH2
H
N
NO2
NH2
25
24
CHO
CO2H
N
N
f, g
The 5-substituted benzimidazoles were prepared as de-
scribed in Scheme 1. Reduction of 4-acetamido-3-nitro-
benzoic acid (15) with borane–THF complex followed
by oxidation with Dess–Martin periodinane6 of the
resultant benzyl alcohol gave aldehyde 17. Horner–
Wadsworth–Emmons reaction of the latter, subsequent
base-mediated hydrolysis of the N-acetyl group, and
reduction of the nitro moiety under mild conditions
afforded di-aniline 20 without concomitant hydrogena-
tion of the trisubstituted olefin (43% from 15). Oxidative
cyclocondensation of 20 with sodium hydrogen sulfite
(NaHSO3) and 6-(4-tert-butyl-phenyl)-pyridine-2-carb-
aldehyde7 furnished benzimidazole 1. Finally, saponifi-
cation of the ethyl ester provided acid 2.
Benzimidazole 3 was prepared from aldehyde 17 via
an analogous sequence of transformations (Scheme 2).
d, e
N
N
H
H
N
N
26
7
O
NH
S
j
h, i
CO2H
CO2H
S
N
N
N
N
H
N
H
N
8
9
Scheme 3. Preparation of 7, 8, and 9. Reagents and conditions: (a)
K2CO3, MeOH, 60 ꢁC, 1 h , 80%; (b) 10% Pd/C, EtOH, 40 ꢁC, 2 h,
95%; (c) 6-(3,4-dimethyl-phenyl)-pyridine-2-carbaldehyde, EtOH,
H2O, NaHSO3, 60 ꢁC, 64%; (d) LiAlH4, THF, ꢀ78 ! 0 ꢁC, 98%; (e)
Dess–Martin periodinane, CH2Cl2, pyridine, 83%; (f) (carbethoxyeth-
yl)triphenylphosphorane, CH2Cl2, rt, 95%; (g) MeOH, NaOH, 50 ꢁC,
15 h, 71%; (h) diethyl malonate, PhH, piperidine, reflux, 15 h, 100%; (i)
THF, NaOH, 50 ꢁC, 15 h, 64%; (j) rhodanine, EtOH, piperidine,
reflux, 15 h, 84%.
Synthesis of the 4-substituted benzimidazoles com-
menced from commercially available methyl-2-acetami-
do-3-nitrobenzoate (Scheme 3). Saponification of the
CO2Et
CO2Et
OH
CHO
CO2H
a
b
c
d
NO2
NO2
NHAc
NO2
NO2
NO2
NHAc
NHAc
NHAc
NH2
16
17
18
19
15
CO2Et
CO2H
CO2Et
e
f
g
N
N
N
N
NH2
H
H
N
N
NH2
1
2
20
Scheme 1. Preparation of 1 and 2. Reagents and conditions: (a) BH3ÆTHF, THF, 0 ꢁC to rt, 57%; (b) Dess–Martin periodinane, CH2Cl2, pyridine,
83%; (c) (carbethoxyethylidene)triphenylphosphorane, CH2Cl2, rt, 95%; (d) K2CO3, EtOH, reflux, 1 h, 96%; (e) PtO2, EtOAc, H2 (1 atm), rt, 100%;
(f) 6-(4-tert-butyl-phenyl)-pyridine-2-carbaldehyde, EtOH, H2O, NaHSO3, 60 ꢁC, 73%; (g) MeOH, NaOH, 50 ꢁC, 89%.
O
H
N
O
H
O
H
N
N
S
CHO
S
S
S
S
S
a
b, c
d
NO2
NHAc
N
N
NH2
NO2
H
N
17
21
22
3
NH2
NHAc
Scheme 2. Preparation of 3. Reagents and conditions: (a) rhodanine, EtOH, piperidine, reflux, 15 h, 94%; (b) K2CO3, EtOH, reflux, 1 h, 96%;
(c) SnCl2, concd HCl, 100 ꢁC, 2.5 h, 96%; (d) 6-(4-tert-butyl-phenyl)-pyridine-2-carbaldehyde, EtOH, H2O, NaHSO3, 60 ꢁC, 47%.

1214
I. G. Safonov et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1212–1216
acetyl group, followed by catalytic hydrogenation of the
nitro moiety and oxidative cyclocondensation with
6-(3,4-dimethyl-phenyl)-pyridine-2-carbaldehyde-provided
benzimidazole 25 (49%, 3 steps). Reduction of the ester
moiety with LAH and Dess–Martin oxidation afforded
aldehyde 26, which was subjected to a Knoevenagel
reaction with diethylmalonate and rhodanine to give 8
and 9, respectively. A Horner–Wadsworth–Emmons
reaction of 26 with (carboethoxyethylidene)triphenyl-
phosphorane gave 7 after saponification of the ester.
We first investigated the effect of variations in the acidic
domain of the pharmacophore, in terms of the nature of
functionality and substitution pattern on the benzimid-
azole nucleus (Table 1). In accord with the pharmaco-
phore model, 5-substituted acrylic ester 1 totally
lacked proliferative activity while acid 2 was active, albe-
it modestly. Interestingly, the malonic acid derivative 3
was less active, suggesting that either the conformation-
al restriction provided by the double bond in 2 was a
feature required for activity or that the presence of the
second carboxylic acid group was a detrimental factor.
Incorporation of the thioxo-thiozolidinone (rhodanine)
group, which can be considered as a carboxylic acid iso-
stere,9 improved the potency (compounds 4–6). There
also appeared to exist a preference, with respect to the
level of proliferation relative to maximum TPO, for
The TPO receptor agonist activity of the benzimidazoles
was tested in vitro by their ability to induce proliferation
of the murine Ba/F3-hTpoR hematopoietic cell line.
These cells express the human TPO receptor, which
confers a proliferative response to TPO stimulation.8
Table 1. Effect of variation in the acidic domain of benzimidazoles on TPO mimetic activity
R1
R2
N
N
H
R3
N
Compound
R1
R2
H
R3
EC50 (lM)
Efficacya (% TPO)
CO2Et
CO2H
1
4-C(CH3)3
NAb
—
2
3
H
H
4-C(CH3)3
4-C(CH3)3
4.5
10
109
13
CO2H
CO2H
O
H
N
4
5
H
4-C(CH3)3
3,4-CH3
2
90
73
S
S
O
O
H
N
H
H
1.6
S
S
S
H
N
6
7
8
3,4-Cl
1.0
NA
NA
19
—
S
CO2H
H
H
3,4-CH3
3,4-CH3
CO2H
—
CO2H
O
H
N
9
H
3,4-CH3
NA
—
S
S
a Efficacy is defined as a percentage of the proliferation value induced by maximal TPO concentration [(TPO) ꢁ 0.1 ng/mL].
b NA, not active.

I. G. Safonov et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1212–1216
1215
bulkier lipophilic R3 substituents on the terminal aro-
matic ring in these analogues. As far as the 4-substituted
benzimidazoles are concerned, all lacked activity regard-
less of the chemical nature of the acidic domain (com-
pounds 7–9). This observation points to a plausible
conclusion that excessive clustering of the pharmaco-
phore elements renders such structures inactive.
TPO agonist activity, as compared to the parent naph-
tho[1,2-d]imidazole class.5c In light of the unexpected
activity of compound 14, our current understanding of
the pharmacophore model, as applied to benzimidazole
TPO agonists, requires a re-evaluation. The full poten-
tial of this chemotype with respect to thrombopoietic
activity as well as further refinements of the developabil-
ity profile are yet to be explored and will be reported in
due course.
Next, we proceeded to determine SAR trends in the
hypothesized metal-binding domain of the pharmaco-
phore. As can be seen in Table 2, a variety of hetero-
atom-containing biaryl substituents (compounds
10–13), which are weak metal chelators, were equally
well tolerated, furan 12 being the most potent analogue.
Surprisingly, compound 14, devoid of any heteroatoms
incorporated in the biaryl group (R), had similar poten-
cy and level of effect to the other benzimidazole TPO
receptor agonists described, indicating that the N-1
nitrogen of the benzimidazole ring alone can function
as the putative metal-binding domain for these TPO
receptor agonists.
References and notes
1. (a) Bazan, J. F. Proc. Natl. Acad. Sci. U.S.A. 1990, 87,
6934; (b) Arai, K.; Lee, F.; Miyajima, A.; Miyatake, S.;
Arai, N. Annu. Rev. Biochem. 1990, 783.
2. Kuter, D. J.; Hunt, P.; Sheridan, W.; Zucker-Franklin, D.
Thrombopoiesis and Thrombopoietins: Molecular, Cellular,
Preclinical, and Clinical Biology; Humana: Totowa, NJ,
1997, p 412.
3. Webb, I. J.; Anderson, K. C. Leuk. Lymphoma 1999, 34, 71.
4. Kaushansky, K. Leukemia 2001, 15, 673.
5. (a) Tian, S.-S.; Lamb, P.; King, A. C.; Miller, S. G.;
Kessler, L.; Luengo, J. I.; Averill, L.; Johnson, R. K.;
Gleason, J. G.; Pelus, L. M.; Dillon, S. B.; Rosen, J. Science
(Washington, DC) 1998, 281, 257; (b) Kimura, T.; Kabu-
raki, H.; Tsujino, T.; Ikeda, Y.; Kato, H.; Watanabe, Y.
FEBS Lett. 1998, 428, 250; (c) Duffy, K. J.; Darcy, M. G.;
Delorme, E.; Dillon, S. B.; Eppley, D. F.; Erickson-Miller,
C.; Giampa, L.; Hopson, C. B.; Huang, Y.; Keenan, R. M.;
Lamb, P.; Leong, L.; Liu, N.; Miller, S. G.; Price, A. T.;
Rosen, J.; Shah, R.; Shaw, T. N.; Smith, H.; Stark, K. C.;
Tian, S.-S.; Tyree, C.; Wiggall, K. J.; Zhang, L.; Luengo, J.
I. J. Med. Chem. 2001, 44, 3730; (d) Duffy, K. J.; Shaw, A.
N.; Delorme, E.; Dillon, S. B.; Erickson-Miller, C.;
Giampa, L.; Huang, Y.; Keenan, R. M.; Lamb, P.; Liu,
N.; Miller, S. G.; Price, A. T.; Rosen, J.; Smith, H.;
Wiggall, K. J.; Zhang, L.; Luengo, J. I. J. Med. Chem. 2002,
45, 3573; (e) Duffy, K. J.; Price, A. T.; Delorme, E.; Dillon,
S. B.; Duquenne, C.; Erickson-Miller, C.; Giampa, L.;
Huang, Y.; Keenan, R. M.; Lamb, P.; Liu, N.; Miller, S.
G.; Rosen, J.; Shaw, A. N.; Smith, H.; Wiggall, K. J.;
Zhang, L.; Luengo, J. I. J. Med. Chem. 2002, 45, 3576; (f)
Suzuki, H.; Furukawa, T.; Yamada, C.; Shibuya, I.;
Kurumi, M.; Yokoyama, T.; Murakami, Y. Heterocycles
2002, 56, 519; (g) Takemoto, H.; Takayama, M.; Shiota, T.
(Shionogi & Co., Ltd, Japan). Internat. Patent WO
2001007423 A1; (h) Takemoto, H.; Shiota, T.; Takayama,
M. (Shionogi & Co., Ltd, Japan). Internat. Patent WO
2001053267 A1.
We have discovered a new benzimidazole series of small-
molecule TPO mimetic agents. A preliminary screen of
functional group tolerances around the benzimidazole
core revealed a tight SAR pattern with only modest
Table 2. Effect of variation in the metal chelate domain of benzim-
idazoles on TPO mimetic activity
H
S
N
O
S
N
R
N
H
Compound Bi-aryl group (R)
EC50 (lM) Efficacya
(% TPO)
N
4
10
11
12
13
14
2
90
117
123
142
134
137
S
1.75
0.78
0.72
1.0
0.9
6. Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113,
7277.
7. All bi-aryl aldehydes used to prepare benzimidazoles 1–14
are either known compounds themselves or can be synthe-
sized in one step from known/commercially available
components by way of a general Suzuki coupling protocol.
8. Ba/F3-hTpoR cells, a murine B lymphocyte cell line
transfected with the human TPO receptor, were maintained
in RPMI-1640/10% FBS with 5 ng/mL rmIL-3 (R&D
Systems) and 500 lg/mL G418 (Gibco) at 37 ꢁC (5% CO2,
95% relative humidity). Cells were washed and plated at
2 · 105 cells/mL with 0–30 lM compound or rhTPO (0.1%
DMSO final concentration) in media. Proliferation was
measured by addition of 10 lCi/mL of [methyl-3H]thymi-
dine (Amersham Biosciences) after 72 h in culture. Cells
were lysed with water after a further 4 h and harvested using
a Brandel 96 Harvester onto a glass fiber filter (Wallac),
according to the manufacturer’s protocol. MeltiLex A
S
O
OH
a Efficacy is defined as a percentage of the proliferation value induced
by maximal TPO concentration [(TPO) ꢁ 0.1 ng/mL].

1216
I. G. Safonov et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1212–1216
scintillator sheets (Wallac) were melted onto the dried filters
Brackeen, M. F., ; Brown, K. K.; Cobb, J. E.; Collins, L.;
Harrington, W. W., Jr.; Hashim, M. A.; Hull-Ryde, E. A.;
Kaldor, I.; Kliewer, A.; Lake, D. H.; Leesnitzer, L. M.;
Lehmann, J. M.; Lenhard, J. M.; Orband-Miller, L. A.;
Miller, J. F.; Mook, R. A.; Noble, S. A.; Oliver, W.; Parks,
D. J.; Plunket, K. D.; Szewczyk, J. R.; Willson, T. M. J.
Med. Chem. 1998, 41, 5020; (c) Buckle, D. R.; Cantello, B.
C. C.; Cawthorne, M. A.; Coyle, P. J.; Dean, D. K.; Faller,
A.; Haigh, D.; Hindley, R. M.; Jefcott, L. J.; Lister, C. A.;
Pinto, I. L.; Rami, H. K.; Smith, D. G.; Smith, S. A.
Bioorg. Med. Chem. Lett. 1996, 6, 2121.
using a TTray Heatsealer (Wallac) and counted in a Wallac
1205 Betaplate reader. The mean proliferation value for
quadruplicate wells was plotted versus compound concen-
tration, and EC50 values were then calculated. The greatest
mean proliferation value for each compound was expressed
as a percentage of that induced by maximal TPO concen-
tration in order to calculate effect relative to TPOmax
.
9. (a) Tashima, T.; Kagechika, H.; Tsuji, M.; Fukasawa, H.;
Kawachi, E.; Hashimoto, Y.; Shudo, K. Chem. Pharm.
Bull. 1997, 45, 1805; (b) Henke, B. R.; Blanchard, S. G.;