Substituted benzimidazoles: A novel chemotype for small molecule hKSP inhibitors

Article abstract of DOI:10.1016/j.bmcl.2009.05.040

Substituted benzimidazoles were profiled as inhibitors of kinesin spindle protein (KSP), an increasingly important target for the development of anticancer drugs. This series demonstrated the monoastral phenotypic response and was found to be active in bo

Full text of DOI:10.1016/j.bmcl.2009.05.040

Bioorganic & Medicinal Chemistry Letters 19 (2009) 3405–3409
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Substituted benzimidazoles: A novel chemotype for small molecule
hKSP inhibitors
a
a
b
b
Brian R. Lahue a, , Yao Ma , Gerald W. Shipps Jr. , Wolfgang Seghezzi , Ronald Herbst
*
a Schering-Plough Research Institute, 320 Bent Street, Cambridge, MA 02141, USA
b Schering-Plough Biopharma, 901 California Avenue, Palo Alto, CA 94304, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 31 March 2009
Revised 7 May 2009
Accepted 11 May 2009
Available online 18 May 2009
Substituted benzimidazoles were profiled as inhibitors of kinesin spindle protein (KSP), an increasingly
important target for the development of anticancer drugs. This series demonstrated the monoastral phe-
notypic response and was found to be active in both enzymatic and cellular-based assays.
Ó 2009 Elsevier Ltd. All rights reserved.
Keywords:
KSP
hKSP
Cancer
Benzimidazole
Cytokinetics
Ispinesib
Cooperative
Monoaster
Kinesin spindle protein, known as KSP or Hs Eg5, plays a critical
role in mitotic spindle assembly and function. A small-molecule
KSP inhibitor would potentially arrest mitosis by preventing normal
bipolar spindle formation,1 inducing apoptosis or cell death follow-
ing the well-documented monoastral phenotype2 visible under
microscopic magnification. As a result, KSP has become a significant
target for the development of cancer therapeutic drugs3 with the po-
tential to overcome the mechanism-based side effects of the micro-
tubule-targeting taxanes and vinca alkaloids, such as neuropathy.4
Herein, we describe the use of substituted benzimidazoles as a novel
chemotype for the inhibition of hKSP which demonstrated potency
in both in vitro (endpoint) and cell-based (Alamar blue cellular pro-
liferation) assays,5 while exhibiting positive cooperative binding6a
with Cytokinetics’ clinical quinazolinone series.6b
Hit optimization of a 2-aminoimidazole identified through
high-throughput screening (HTS) was commenced using both mix-
ture-based libraries as well as single, purified compounds. Utilizing
this twofold approach along with our proprietary affinity-based
selection mass spectrometric automated ligand identification sys-
tem (ALIS),7 mixtures of compounds (hundreds per mixture) con-
taining a diverse set of building blocks were screened for KSP
binding, the most active of which were subjected individually to
enzymatic and cell-based assays. A solid-phase synthetic strategy
adapted from the well-documented polymer-supported routes in
the literature8 was utilized for the mixture-based and the majority
of parallel discrete compound preparation (Scheme 1). This versa-
tile approach to benzimidazoles 6 allowed for rapid diversification
at three positions (6, R1, R2, and R3). Beginning with either Wang
(for X = O) or Rink (for X = NH) resin, nitrobenzoic acid 1 was
bound to the solid-phase support to give 2 as the corresponding es-
ter or amide, respectively. SNAr displacement of fluoride by pri-
mary amines provided nitroanilines 3, which were reduced with
tin(II) chloride to give 1,2-diamines 4. The benzimidazole core
was subsequently formed by cyclization with isothiocyanates to
generate 5, the resin cleavage of which provided 6 in high purity.
Methyl ester variants (6, R1 = –CO2Me) were obtained by esterifi-
cation of the corresponding carboxylic acid with trimethylsilyldia-
zomethane following cleavage from the resin. Synthesis of
secondary and tertiary amides at the R1 position, the former of
which had been demonstrated on solid phase beginning with
reductive amination of an aldehyde-based resin with various ami-
nes,8c were prepared by amidation of the carboxylic acids post-
cleavage.
The commercial sources of isothiocyanates used to diversify the
R3 position of 6 were rapidly exhausted, thus necessitating the
need for a facile synthetic approach to these building blocks. Penso
and Albanese demonstrated the conversion of trifluoromethylace-
tamides to isothiocyanates,9 which was adapted to a one-pot pro-
cedure to fit our needs (Scheme 2).
* Corresponding author. Tel.: +1 617 499 3532; fax: +1 617 499 3535.
E-mail address: brian.lahue@spcorp.com (B.R. Lahue).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.05.040

3406
B. R. Lahue et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3405–3409
Scheme 3. Synthesis of 2-aminobenzimidazoles containing aliphatic R3 substitu-
ents. (a) i-Pr2NEt; (b) H2, Pd–C; (c) CDI, Et3N, 55 °C; (d) POCl3, neat; (e) DMSO, i-
Pr2NEt, 100 °C; (f) 6 N HCl, reflux; (g) EDC, HOBt, i-Pr2NEt, HNR0 or PS-CDI, HOBt, i-
Pr2NEt, HNR0.
Scheme 1. Solid-phase synthesis of benzimidazoles. (a) X = O Wang resin, N,N0-
diisopropylcarbodiimide, DMAP; (b) X = NH (COCl)2, DMF then Rink resin, i-Pr2NEt;
(c) i-Pr2NEt; (d) SnCl2; (e) N,N0-diisopropylcarbodiimide; (f) TFA, H2O; (g) TMS–
CHN2; (h) PS-CDI resin, HOBt, i-Pr2NEt, HNR0.
Scheme 2. One-pot synthesis of isothiocyanates from anilines.9 (a) TFAA, pyr; (b)
CS2, NaOH, K2CO3.
While the conversion of anilines to arylisothiocyanates follow-
ing this route was successful in most cases, the synthesis of ali-
phatic isothiocyanates failed and therefore the introduction of
aliphatic substituents on the exocyclic nitrogen of the 2-amino-
benzimidazole scaffold required an alternative methodology. Ali-
Scheme 4. Synthesis of regioisomeric 2-aminobenzimidazoles.
(R1 = CO2Et) and the SNAr step was performed at 0 °C which selec-
tively led to 6-fluorobenzimidazoles 19.
phatic R3 substituents on compounds
6 which were not
commercially available as the corresponding isothiocyanates were
necessarily introduced through SNAr of 2-chlorobenzimidazoles 14
to give the corresponding 2-aminobenzimidazoles 15 (Scheme 3).
Beginning with 4-fluoro-3-nitrobenzoic acid ethyl ester (10,
R1 = CO2Et), SNAr displacement of fluoride with primary amines
gave 11 in analogous fashion to the solid phase approach previ-
ously illustrated. 1,2-Diamines 12, formed by hydrogenation of
nitroanilines 11, were converted to 2-chlorobenzimidazoles 14 in
a two-step procedure.10 Thermal chloride displacement10b with
amines provided 15, in which the R1 ethyl ester could be converted
to the corresponding carboxylic acid by HCl-promoted hydrolysis
and to various amides by subsequent amidation with EDC/HOBt
or PS-CDI/HOBt.
One of the initial screening hits, compound 6a, was used as a
starting point for the optimization beginning at the R2 position
(Table 1). With an IC50 = 7.4
lM in the in vitro endpoint assay, this
compound displayed monoaster formation at 100
lM concentra-
tion. The benzylic functionality was required for enzymatic activity
and substitution at the benzylic methylene was not tolerated—sim-
ple introduction of an additional methyl group as in 6b resulted in
a total loss of activity. The ortho substituent on the aromatic ring
proved to be important for potency as well. Replacing the trifluoro-
methyl moiety with a ring-nitrogen or a methoxy substituent, as in
compounds 6d and 6e, respectively, resulted in a substantial loss in
potency. However, the 1,2-dimethoxy variant 6f proved to be well-
tolerated with an IC50 = 5.0 lM and also showed monoaster forma-
Regioisomeric variants of these 2-aminoimidazoles, in which
the amide carbonyl is shifted to C6 (Scheme 4, 17) were prepared
under identical conditions to those described in Scheme 1, begin-
ning with the corresponding 3-fluoro-4-nitrobenzoic acid (16, –
H). While 2,5-difluoro-4-nitrobenzoic acid (16, –F) underwent
selective displacement of the 5-fluoro substituent by the R2
amines en route to C5 fluorobenzimidazoles 17, with 2,4-di-
fluoro-5-nitrobenzoic acid (18, R1 = CO2H) it was necessary to
modify the reaction conditions to avoid bis-fluoride displacement.
Consequently, the sequence was transitioned to solution-phase
Figure 1. 2-Aminobenzimidazole with R2 substituents (see Table 1).

B. R. Lahue et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3405–3409
3407
Table 1
In vitro endpoint assay data and SAR at the R2 position (refer to Fig. 1).
Compound
R2
IC50
7.4
(l
m)5
6a
6b
6c
6d
6e
6f
>20
>20
>20
20
Figure 2. 2-Aminobenzimidazoles in Table 2.
Table 2
In vitro data and SAR at X and R3 positions (refer to Fig. 2)
Compound
X
R30
R300
IC50 (l
M)5
6a
6j
6k
6l
6m
6n
6o
6p
6q
6r
–NH2
–NH2
–NH2
–NH2
–SO2NH2
–SO2Me
–OMe
H
H
H
F
7.4
4.8
2.5
1.3
0.52
>20
6.9
>20
2.3
1.4
2.2
13.4
33.4
5.2
–OMe
–NH2
–OH
–OH
–OH
–SMe
H
H
H
H
F
–SO2NH2
–SO2Me
–OMe
–OH
–OMe
5.0
–OH
–OMe
–OMe
–OMe
–NHMe
–NHMe
–NMe2
–N(CH2)4
–NH–Ph(o-OMe)
–NH–Ph(o–OMe)
–SMe
–SO2NH2
–OMe
–SMe
–OMe
–SO2NH2
–SO2NH2
–SO2NH2
–SO2NH2
–OMe
H
H
H
H
H
H
H
H
H
H
6s
6t
6u
6v
6w
6x
6y
6z
6aa
6g
6h
6i
>20
19.0
2.1
9.2
16.3
>20
2.4
6.0
least potent and the thiomethylether the most potent. The reverse
was true in the methyl ester series with sulfonamide 6s and thio-
ether 6u having IC50 values of 2.2 and 33 lM respectively. The
majority of amides synthesized which were larger than the pri-
mary amide (X = NH2, ꢀ30 compounds) were found to be less po-
tent than the corresponding primary amides, as illustrated by
tion in cells at a concentration of 100 lM. Benzothiophenes linked
through C3, when properly substituted, were found to be inter-
comparing compounds 6a (7.4
lM) with 6x and 6y (16.3 and
changeable groups with the ortho-trifluoromethylphenyl of 6a,
>20 M, respectively). However, two substitutions were tolerated
at this position. N-Methyl amides 6v and 6w demonstrated small
loses in potency compared to the analogous primary amides 6k
and 6a (2.5–5.2 and 7.4–9.2 lM, respectively). The ortho-methoxy-
l
resulting in IC50 = 19
IC50 = 2.1 M for unsubstituted 6i. Benzothiazole 6g, while struc-
turally similar to benzothiophene 6i but linked through C2, was
not tolerated with an IC50 > 20 M.
l
M
for chloro-substituted 6h and
l
l
aniline-based amides 6z and 6aa showed mixed results compared
to the corresponding primary amides, with 6aa losing activity ver-
While keeping the R2 ortho-trifluoromethylbenzyl substituent
constant, the left- and right-hand portions of the molecule were
profiled (Table 2). Modifying the para-substituent (R30) on the ani-
lino-ring led to significant gains in potency. Replacing the sulfon-
amide (6a) with methyl sulfone (6j) improved the in vitro
sus 6k (2.5 to 6.0
6a (7.4–2.4 M).
lM) and 6z showing a modest improvement over
l
With the apparent need for a polar group on the distal side of
the R3 ring identified, the spacer between the 2-aminobenzimidaz-
ole core and this group was probed (Fig. 3, Table 3). Small, non-aro-
matic heterocycles such as aminopyrrolidine 15a failed to display
any in vitro activity. Similarly, adding an additional one (15b,
15c, and 15d) or two (15e) methylene spacer while keeping the
active para-substituted phenyl ring in place also resulted in
significant potency loss. The synthesis of more than 50 related
compounds illustrated that the anilino ring was a requirement
for both enzymatic and cellular potency.
activity from 7.4 to 4.8
and thioether (6m) proved even more potent at 2.5
0.5 M, respectively. The addition of a fluoro substituent ortho to
the p-methoxy group improved the activity comparing 6k to 6l
(2.5 M vs 1.3 M) and 6p to 6q (>20 M vs 2.3 M).
lM. The corresponding methyl ether (6k)
lM and
l
l
l
l
l
The left hand portion of the 2-aminoimidazole core proved less
amenable to modification. In general, the carboxylic acids (X = OH,
6n–r) were found to be less potent than the corresponding primary
amides discussed above (X = NH2, 6a, 6j–m) with R30 as SMe again
With the periphery of the benzimidazole template profiled, fur-
ther substitution of the bicycle was examined (Fig. 4). Moving the
primary amide from C5 in 6a to C6 in 17a and 17b resulted in com-
plete loss of in vitro activity. 6-Fluoro derivative 19a, a regioisomer
of inactive 17b, retained the potency of the parent compound (6a)
at 7.9 lM, indicating that C6 may be amenable to further substitu-
tion provided that the C5 carbonyl remains intact. The C2 amino
functionality found in all compounds discussed thus far proved
being the most potent at 1.4
activity with IC50’s of 19 and 16
mar blue assay as well as displaying the monoastral phenotype
at 25 M. Not only were the methyl esters (X = OMe, 6s–u) in gen-
eral less potent than the corresponding carboxylic acids and pri-
mary amides, these derivatives also failed to follow the SAR
trend at the R30 position. In both the primary amide (see 6a) and
the carboxylic acid (see 6n) series, the R30 sulfonamide was the
l
M. Both 6k and 6m showed cellular
l
M, respectively, in the 48 h Ala-
l

3408
B. R. Lahue et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3405–3409
Figure 3. 2-Aminobenzimidazole with aliphatic R3 substituents (see Table 3).
Table 3
In vitro endpoint assay data for aliphatic R3 substituents
Compound
R3
IC50 (
l
M)5
15a
>20
15b
15c
15d
15e
15f
>20
>20
>20
>20
>20
Figure 4. Benzimidazoles with varying substitution patterns and their in vitro
endpoint assay IC50 values.5
Table 4
DMPK profile of selected compounds
Compound Rat AUC
(nM h)a
P450 3A4
(co/pre)b
P450 2D6
(co/pre)b
P450 2C9
(co/pre)b
Caco-2
(nm/s)c
6j
6k
6m
0
600
150
>30/>20
>30/>15
>30/<15
>30/>15
>30/>15
>30/>15
15/>8
8/>4
14/>7
18
670
530
a
b
c
Rat oral exposure over 6 h at a dose of 10 mpk.
Reported as IC50 in M.
Caco-2 permeability.
l
to be vital as well since replacing it with a methylene linker in 20a
and 20b led to potency loss.
Supplementary data
A DMPK analysis of selected compounds from this series
showed poor oral exposure in the rat, but a promising profile in
cytochrome P450 inhibition (Table 4). Compounds 6k and 6m
demonstrated good permeability in the Caco-2 assay.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.05.040.
References and notes
As compounds identified from mixture-based, solid-phase
benzimidazole libraries as hKSP inhibitors, this series appears to
be non-competitive with ADP and monastrol.3a Conclusive evi-
dence of multiple compounds in this series showing binding affin-
ity enhancement of quinazolinone-based Ispinesib analogs to KSP
through positive cooperative allosteric binding has been discov-
ered and will be reported in a separate communication. Cell-based
activity has been observed in 48 h and 7-day Alamar blue assays
and monoaster formation was observed for the most active com-
pounds. Kinesin counterscreens indicated all compounds discussed
herein are selective for hKSP over Kif3B and nKHC (with in vitro
1. Heald, R. Cell 2000, 102, 399.
2. (a) Mayer, T. U.; Kapoor, T. M.; Haggarty, S. J.; King, R. W.; Schreiber, S. L.;
Mithison, T. J. Science 1999, 286, 971; (b) Kapoor, T. M.; Mayer, T. U.; Coughlin,
M. L.; Mitchison, T. J. J. Cell Biol. 2000, 150, 1975; (c) Sakowicz, R.; Finer, J. T.;
Beraud, C.; Crompton, A.; Lewis, E.; Fritsch, A.; Lee, Y.; Mak, J.; Moody, R.;
Turincio, R.; Chabala, J. C.; Gonzales, P.; Roth, S.; Weitman, S.; Wood, K. W.
Cancer Res. 2004, 64, 3276.
3. (a) Wood, K. W.; Bergnes, G. Ann. Rep. Med. Chem. 2004, 39, 173; (b) Cox, C. D.;
Breslin, M. J.; Mariano, B. J.; Coleman, P. J.; Buser, C. A.; Walsh, E. S.; Hamilton,
K.; Huber, H. E.; Kohl, N. E.; Torrent, M.; Yan, Y.; Kuo, L. C.; Hartman, G. D.
Bioorg. Med. Chem. Lett. 2005, 15, 2041; (c) Cox, C. D.; Torrent, M.; Breslin, M. J.;
Mariano, B. J.; Whitman, D. B.; Coleman, P. J.; Buser, C. A.; Walsh, E. S.;
Hamilton, K.; Schaber, M. D.; Lobell, R. B.; Tao, W.; South, V. J.; Kohl, N. E.; Yan,
Y.; Kuo, L. C.; Prueksaritanont, T.; Slaughter, D. E.; Li, C.; Mahan, E.; Lu, B.;
Hartman, G. D. Bioorg. Med. Chem. Lett. 2006, 16, 3175; For recent reviews, see:
(d) Zhang, Y.; Xu, W. Anti-Cancer Agents Med. Chem. 2008, 8, 698; (e) Sarli, V.;
Giannis, A. Clin. Cancer Res. 2008, 14, 7583; (f) Knight, S. D.; Parrish, C. A. Curr.
Top. Med. Chem. 2008, 8, 888.
4. (a) Cox, C. D.; Coleman, P. J.; Breslin, M. J.; Whitman, D. B.; Garbaccio, R. M.;
Fraley, M. E.; Buser, C. A.; Walsh, E. S.; Hamilton, K.; Schaber, M. D.; Lobell, R. B.;
Tao, W.; Davide, J. P.; Diehl, R. E.; Abrams, M. T.; South, V. J.; Huber, H. E.;
Torrent, M.; Prueksaritanont, T.; Li, C.; Slaughter, D. E.; Mahan, E.; Fernandez-
Metzler, C.; Yan, Y.; Kuo, L. C.; Kohl, N. E.; Hartman, G. D. J. Med. Chem. 2008, 51,
4239; (b) Yan, Y.; Sardana, V.; Xu, B.; Homnick, C.; Halczenko, W.; Buser, C. A.;
Schaber, M.; Hartman, G. D.; Huber, H. E.; Kuo, L. C. J. Mol. Biol. 2004, 335, 547.
5. Each compound assayed in the in vitro endpoint assay was tested in duplicate
(n = 2) with %CV’s of <10. Further details of the in vitro and cellular
proliferation assays can be found in Supplementary data, as well as the
IC50 values >50 lM). Selectivity over HSET proved to be more of
a challenge but several compounds demonstrated moderate to
good selectivity in the corresponding biochemical assays, such as
6a (fivefold), 6o (sevenfold), 6m (eightfold), 6l (11-fold), and 6i
(>24-fold). Each benzimidazole combinatorial position has been
explored, and intuitive compounds are currently under
investigation.
Acknowledgments
The authors acknowledge Drs. Noriyuki Kawahata and Tong
Wang for their early work on this project.

B. R. Lahue et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3405–3409
3409
corresponding patent application: Shipps, G. W., Jr.; Ma, Y.; Lahue, B. R.;
Seghezzi, W.; Herbst, R.; Chuang, C.-C.; Annis, D. A.; Kirt-Ley, M.
WO2008153701(A1).
5419; (c) Mazurov, A. Bioorg. Med. Chem. Lett. 2000, 10, 67; (d) Tumelty, D.; Cao,
K.; Holmes, C. P. Org. Lett. 2001, 3, 83; (e) Huang, K.-T.; Sun, C.-M. Bioorg. Med.
Chem. Lett. 2002, 12, 1001.
6. (a) Jiang, C.; Chen, Y.; Wang, X.; You, Q. J. Mol. Model. 2007, 13, 987; (b) Lad, L.;
Luo, L.; Carson, J. D.; Wood, K. W.; Hartman, J. J.; Copeland, R. A.; Sakowicz, R.
Biochemistry 2008, 47, 3576.
7. Annis, D. A.; Chuang, C.-C.; Naim, N. Methods Prin. Med. Chem. 2007, 36, 121.
8. (a) Smith, J. M.; Gard, J.; Cummings, W.; Kanizsai, A.; Krchnak, V. J. Comb. Chem.
1999, 1, 368; (b) Kilburn, J. P.; Lau, J.; Jones, R. C. F. Tetrahedron Lett. 2000, 41,
9. Albanese, D.; Penso, M. Synthesis 1991, 1001.
10. (a) Crank, G.; Mursyidi, A. Aust. J. Chem. 1982, 35, 775; (b) Shao, B.; Huang, J.;
Sun, Q.; Valenzano, K. J.; Schmid, L.; Nolan, S. Bioorg. Med. Chem. Lett. 2005, 15,
719.