N-Aryl-benzimidazolones as novel small molecule HSP90 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2010.10.010

We describe the development of a novel series of N-aryl-benzimidazolone HSP90 inhibitors (9) targeting the N-terminal ATP-ase site. SAR development was influenced by structure-based design based around X-ray structures of ligand bound HSP90 complexes. Lea

Full text of DOI:10.1016/j.bmcl.2010.10.010

Bioorganic & Medicinal Chemistry Letters 20 (2010) 7503–7506
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
N-Aryl-benzimidazolones as novel small molecule HSP90 inhibitors
Milan Bruncko a, , Stephen K. Tahir a, Xiaohong Song a, Jun Chen a, Hong Ding a, Jeffrey R. Huth b, Sha Jin a,
⇑
Russell A. Judge b, David J. Madar a, Chang H. Park b, Cheol-Min Park a, Andrew M. Petros b, Christin Tse a,
Saul H. Rosenberg a, Steven W. Elmore a
a Cancer Research, Global Pharmaceutical R&D, Abbott Laboratories, 100 Abbott Park Road, Abbott Park, IL 60064-6101, United States
b Structural Biology, Global Pharmaceutical R&D, Abbott Laboratories, 100 Abbott Park Road, Abbott Park, IL 60064-6101, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 July 2010
Revised 29 September 2010
Accepted 1 October 2010
Available online 12 October 2010
We describe the development of a novel series of N-aryl-benzimidazolone HSP90 inhibitors (9) targeting
the N-terminal ATP-ase site. SAR development was influenced by structure-based design based around X-
ray structures of ligand bound HSP90 complexes. Lead compounds exhibited high binding affinities, ATP-
ase inhibition and cellular client protein degradation.
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
HSP90
Benzimidazolone
Cancer
Heat shock protein 90 (HSP90) is a highly conserved chaperone
protein that is ubiquitously expressed, comprising approximately
one to two percent of total protein in normal cells.1 HSP90, as well
as other members of the chaperone family, are essential for normal
cellular housekeeping functions and mediate numerous aspects of
post-translational protein processing. These functions include the
facilitation of proper folding of newly synthesized proteins, pre-
vention of client protein aggregation, translocation of client pro-
teins across membranes, and the modulation of active client
protein conformation and signaling for normal protein turnover.2
Chaperone proteins are upregulated as an adaptive response to a
variety of toxic stresses and promote cell survival.3,4
HSP90 is 2–10-fold more abundant in tumor cells compared to
their normal counterparts.5 There is an additional dependence of
cancer cells on HSP90 because many of its client proteins are onco-
genes. HSP90 client proteins are known to be essential to tumori-
genesis and tumor maintenance and to the development of the six
characteristic hallmarks of cancer.6–11 Inhibition of HSP90 leads to
the degradation of client proteins and is a promising therapeutic
approach by which multiple oncogenic pathways can be targeted
simultaneously. Given that cancer is a heterogeneous disease, often
involving alterations in multiple proliferative and survival path-
ways, a multi-targeted approach is likely essential for effective
therapeutic strategy. Like most chaperone proteins, HSP90 func-
tions as a homodimer that is an integral part of a larger super-
chaperone complex made up of other chaperones, co-chaperones,
accessory and adaptor proteins (e.g., HSP70, HSP40, HIP, HOP,
CDC37, p50, p23, AHA1, and immunophilin).12–14 HSP90 contains
a Bergerat ATP binding fold in the N-terminus and a second, less
characterized nucleotide binding-site in the C-terminus.15 The
chaperone properties of the complex are energetically driven by
the binding and hydrolysis of ATP in an iterative cycle. Unlike other
ATP binding protein targets, such as kinases, HSP90 does not cova-
lently modify its substrates. Client proteins first bind HSP90 as part
of an intermediate complex. ATP binding and hydrolysis act as a
molecular switch that drives conformational changes in HSP90
and results in altered co-chaperone recruitment.16 The transition
to the ‘mature’ complex results in an active client protein state
and propagates the chaperone catalytic cycle. Inhibition of the
ATP-ase activity effectively turns off the switch and prevents tran-
sition. The complex is locked in the immature state, and ultimately
results in client protein degradation.
Geldanamycin analogs17 (including 17-AAG, 1)18 and radicicol
(2)19 are very efficient HSP90 inhibitors (Fig. 1). Both compounds
are very potent in in vitro assays but have potential limitations
in an in vivo setting (formulation difficulties, off-target toxicities,
and/or instability in plasma).20,21 Pharmaceutical companies and
academic institutions have exerted significant efforts to discover
small molecule inhibitors of HSP90 that would overcome problems
associated with natural product based analogs.22–24 One such
example is the diarylpyrazole, VER-49009 (3). Here we describe
our efforts to identify small molecule inhibitors of HSP90 that re-
sulted in the discovery of a novel series of N-arylbenzimidazolones.
To identify compounds that bind the N-terminal ATPase domain
⇑
Corresponding author. Fax: +1 847 935 5165.
E-mail address: milan.bruncko@abbott.com (M. Bruncko).
of human HSP90a, we employed several high-throughput screen-
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.10.010

7504
M. Bruncko et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7503–7506
O
O
H
N
OH
O
O
H
O
O
N
H
OH
HO
H
O
O
Cl
O
O
NH2
O
2
1
O
CF3
HO
Cl
HO
N
O
O
N
H
F3C
OH
N
NH
HN
3
9a
Figure 1. Compounds 1 17-AAG, 2 Radicicol, 3 VER-49009, and 9a NS1619.
ing methods. Techniques included NMR screening, affinity selec-
tion/mass spectrometry (ASMS) and Time Resolved-Fluorescence
Resonance Energy Transfer (TR-FRET) assays. ASMS screening pro-
vided a variety of HSP90 inhibitors including the N-arylbenzimi-
dazolone NS1619 (9a)25 that exhibited low micromolar affinity.
Compound binding affinity for the N-terminal ATPase domain of
Figure 2. X-ray crystal structure of benzoimidazolone 9c (cyan) (3OW6) overlaid
with radicicol 2 (orange) (1BGQ) bound to the N-terminal ATPase domain of human
HSP90.
ing affinity relative to 9c. Other positional combinations of hydro-
xyl and halogen substituents at the 4- and 5-positions conferred
significant decreases in activities, thus the 5-chloro-2,4-dihydroxy
substitution pattern was used for all following analogs.
HSP90a was determined by competition binding using TR-FRET.
Functional activity was assessed via the Malachite Green ATP-ase
assay.
Benzimidazolones 9 were conveniently prepared in four steps
from commercially available 2-fluoro-nitrobenzenes 4 and substi-
tuted anilines 5 (Scheme 1). Coupling of 4 and 5 under basic con-
ditions and elevated temperatures provided substituted
nitroanilines 6 in moderate yields. Hydrogenation of nitroanilines
Table 1 summarizes the binding affinity data for substituted N-
aryl-benzoimidazolones. Removal of the CF3 substituent from the
benzimidazolone ring of NS1619 (9a) had negligible effect on the
binding activities of 9b (Table 1), and we thus performed the first
round of SAR studies on a benzimidazolone core containing an
unsubstituted phenyl ring. Removal of the meta-CF3 group and
introduction of the hydroxy group at the 4-position of N-phenyl
ring (9c) afforded a nearly 10-fold increase in activity. The overlay
of X-ray crystal structures of radicicol and 9c (Fig. 2) clarifies this
significant boost in binding affinity. Similar to radicicol, 9c gathers
the water-mediated hydrogen bonding interaction with the pro-
tein back bone. Based on this X-ray data, we incorporated a 5-
chloro group to afford analog 9e, which showed a heightened bind-
R1
R2
F
O
R1
R2
a
b
O
R
NO2
H2N
NH
5a) R1 = CF3, R2 = H
4a) R = H
4b) R = CN
5b) R1 = H, R2 = H
R
NO2
5c) R1 = OMe, R2 = H
5d) R1 = H, R2 = OMe
5e) R1 = OMe, R2 = Cl
5f) R1 = Cl, R2 = OMe
6
Table 1
Binding data for substituted 1-aryl-benzimidazolones
R1
R1
R1
R1
R2
R2
R2
R2
O
HO
HO
c
d
O
N
N
N
NH
O
O
R
O
N
H
N
H
N
H
R
R
R
NH2
8
7
9
Compound
R1
H
R2
R
IC50, FRETa
3.32
(
l
M) IC50, MG ATPasea
81
(
lM)
6, 7, 8
a) R = H, R1 = H, R2 = CF3
b) R = H, R1 = OMe, R2 = H
c) R = H, R1 = H, R2 = OMe
d) R = H, R1 = OMe, R2 = Cl
e) R = H, R1 = Cl, R2 = OMe
f) R = CN, R1 = OMe, R2 = Cl
9a (NS1619)
CF3 5-
CF3
CF3
H
b) R = H, R1 = H, R2 = CF3
c) R = H, R1 = OH, R2 = H
d) R = H, R1 = H, R2 = OH
e) R = H, R1 = OH, R2 = Cl
f) R = H, R1 = Cl, R2 = OH
9b
9c
9d
9e
9f
H
OH
H
OH Cl
Cl OH
H
2.32
0.22
34.9
0.054
2.53
127
85
200
15.2
200
H
H
H
H
OH
Scheme 1. Reagents and conditions: (a) K2CO3, DMA, 120 °C, (26–46%); (b) H2, Pd/
a
All determinations are triplicate values.
C; (c) phosgene; (31–66%, two steps); (d) BBr3 (44–74%).

M. Bruncko et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7503–7506
7505
6 followed by treatment with phosgene afforded methylated benz-
imidazolone intermediates 8. Subsequent demethylation with BBr3
afforded the final benzimidazolones 9.
A comparison of the X-ray crystal structures of 9c and VER-
49009 (3),26 (Fig. 3) indicated that the 5-chloro-2,4-dihydroxy-
phenyl ring occupies the same site when bound to the N-terminal
ATPase domain of human HSP90. VER-49009 (3) takes advantage of
additional interactions of the carboxamide group with Gly97 and
Lys58 residues via hydrogen bonding.27 We next examined substi-
tution at the 5-position in order to mimic these interactions.
To probe the effects of substitution at the 5-position, we utilized
easily accessible 5-aminomethyl-benzimidazolone as a synthetic
precursor for variety of amides 12a,b, ureas 12d,e and sulfona-
mides 12c,f,g,h. Substituted benzimidazolones 12 were prepared
from the 5-cyano analog 8f. Borane–THF reduction of 8f yielded
5-aminomethyl-benzimidazolones 10 which was subsequently
coupled with isocyanates, acyl- and sulfonyl-chlorides to yield
Table 2
Binding data for 5-substituted 1-(5-chloro-2,4-dihydroxyaryl)-benzimidazolones
OH
Cl
HO
N
H
N
O
N
H
R
Compound
R
IC50
,
IC50, MG
ATPasea
(lM)
FRETa
(
lM)
1 (17-AAG)
—
0.02
9.8 (2)
2 (radicicol)
3 (VER-49009)
—
—
0.016 0.001 (33)
0.019 0.002 (3)
0.040
0.072
0.050
0.033
0.060
0.027
0.53 0.03 (34)
0.88 0.15 (3)
24.8
22.9
25.2
14.2
15.4
3.42
12a
12b
12c
12d
12e
12f
Ac
Bz
CH3SO2
EtNH(C@O)
PhNH(C@O)
PhSO2
12g
12h
1-Naphthyl-SO2
2-Naphthyl-SO2
0.030
0.047
0.55
1.14
a
Standard error (number of determinations). All determinations are triplicate
values.
Figure 3. X-ray crystal structure of benzimidazolone 9c (cyan) (3OW6) overlaid
with VER-49009 3 (orange) (3OWB) bound to the N-terminal ATPase domain of
human HSP90.
O
O
Cl
Cl
O
O
a
b
N
N
O
O
H2N
N
H
N
H
NC
8f
10
O
OH
Cl
Cl
O
HO
c
N
N
H
N
H
N
O
O
N
H
N
H
R
R
11
12
a) R = Ac
b) R = Bz
c) R = Ms
f) R = PhSO2
g) R = 1-Naphthyl-SO2
h) R = 2-Naphthyl-SO2
d) R = EtNH(C=O)
e) R = PhNH(C=O)
Scheme 2. Reagents and conditions: (a) BH3–THF (91%); (b) R-Cl or R-N@C@O,
Figure 4. X-ray structure of benzimidazolone 12g (3OWD)bound to the N-terminal
Et3N, (56–76%); (c) BBr3 (37–81%).
ATPase domain of human HSP90.

7506
M. Bruncko et al. / Bioorg. Med. Chem. Lett. 20 (2010) 7503–7506
Table 3
commercially available lead compound, NS1619 (9a). Develop-
ment of structure activity relationships was heavily influenced by
structure-based design based on X-ray structures of ligand bound
HSP90 complexes. Lead compounds exhibited binding affinities,
ATP-ase inhibition and cellular client protein degradation activities
comparable to or better than existing inhibitors. These compounds
may serve as viable leads for further optimization.
Cellular effect of HSP90 inhibitors on client protein knockdown and HSP70 up-
regulation in human tumor cell lines
Compound
In-cell western
Luciferase
reporter
Cell viability
assay
ErbB2 EC50
M)
HSP70 EC50
M)
Hif-1
M)
a
EC50
EC50, SKBR3a
a
a
a
(
l
(
l
(
l
(lM)
1 (17-AAG)
2 (VER-49009) 0.35
12g <0.10
<0.10
0.06
0.50
0.06
0.12
0.21
0.19
0.025
0.25
0.012
Acknowledgments
a
All determinations are triplicate values.
We thank Dr. Derek W. Nelson and Dr. Andrew Souers for their
help in preparing this manuscript.
References and notes
amides, sulfonamides and ureas 11. Demethylation with BBr3 pro-
vided 12 in good yields (Scheme 2).
1. Lindquist, S.; Craig, E. A. Annu. Rev. Genet. 1988, 22, 631.
2. Whitesell, L.; Lindquist, S. L. Nat. Rev. Cancer 2005, 5, 761.
3. Solit, D. B.; Chiosis, G. Drug Discovery Today 2008, 13, 38.
4. Karapanagiotou, E. M.; Syrigos, K.; Saif, M. W. Expert Opin. Invest. Drugs 2009,
18, 161.
5. Isaacs, J. S.; Xu, W.; Neckers, L. Cancer Cell 2003, 3, 213.
6. Neckers, L.; Neckers, K. Expert Opin. Emerg. Drugs 2005, 10, 137.
7. Neckers, L.; Ivy, S. P. Curr. Opin. Oncol. 2003, 15, 419.
8. O’Bryan, E.; Nimmanapalli, R.; Bhalla, K. Cancer Res. 2001, 61, 1799.
9. Pandey, P.; Saleh, A.; Nakazawa, A.; Kumar, S.; Srinivasula, S. M.; Kumar, V.;
Weichselbaum, R.; Nalin, C.; Alnemri, E. S.; Kufe, D.; Kharbanda, S. EMBO J.
2000, 19, 4310.
10. Cardone, M. H.; Roy, N.; Stennicke, H. R.; Salvesen, G. S.; Franke, T. F.;
Stanbridge, E.; Frisch, S.; Reed, J. C. Science 1998, 282, 1318.
11. Workman, P. Trends Mol. Med. 2004, 10, 47.
12. Kamal, A.; Thao, L.; Sensintaffar, J.; Zhang, L.; Boehm, M. F.; Fritz, L. C.; Burrows,
F. J. Nature 2003, 425, 407.
All compounds maintained good binding affinities and aryl sul-
fonamides displayed improved enzymatic inhibition in the Mala-
chite green assay. 1-Naphthyl sulfonamide 12g showed activity
comparable to both radicicol (2) and VER-49009 (3) in this assay
(Table 2).
The X-ray crystal structure of 12g bound to HSP90 is shown in
Figure 4. The 1-naphthylsulfonamido moiety of 12g adopts a ‘bent-
back’ conformation possibly due to hydrophobic collapse, allowing
the sulfonamide group to engage in a water-mediated hydrogen
bond with Lys58. The orientation and binding of the 5-chloro-
2,4-dihydroxyphenyl ring is maintained as in previous structures.
The cellular effects of 12g, 17-AAG (1), and VER-49009 (3) were
first evaluated in viability assays in the SKBR3 human breast can-
cer cell line.28 Compound 12g displayed equal or greater cell killing
behavior than 17-AAG (1) or VER-49009 (3) (Table 3).
13. Hartl, F. U. Nature 1996, 381, 571.
14. Prince, T.; Matts, R. L. J. Biol. Chem. 2004, 279, 39975.
15. Marcu, M.; Chadli, A.; Bouhouche, I.; Catelli, M.; Neckers, L. J. Biol. Chem. 2000,
275, 37181.
16. Wegele, H.; Müller, L.; Buchner, J. Rev. Physiol. Biochem. Pharmacol. 2004, 151, 1.
17. Whitesell, L.; Mimnaugh, E. G.; De Costa, B.; Myers, C. E.; Neckers, L. M. Proc.
Natl. Acad. Sci. USA 1994, 91, 8324.
Several assays were employed to directly examine the effects of
12g on the degradation of known HSP90 client proteins in human
18. Schulte, T. W.; Neckers, L. M. Cancer Chemother. Pharmacol. 1998, 42, 273.
19. Schulte, T. W.; Akinaga, S.; Soga, S.; Sullivan, W.; Stensgard, B.; Toft, D.;
Neckers, L. M. Cell Stress Chaperones 1998, 3, 100.
20. Neckers, L. Curr. Top. Med. Chem. 2006, 6, 1163.
21. Supko, J. G.; Hickman, R. L.; Grever, M. R.; Malspeis, L. Cancer Chemother.
Pharmacol. 1995, 36, 305.
22. Chiosis, G. Curr. Top. Med. Chem. 2006, 6, 1183.
23. McDonald, E.; Jones, K.; Brough, P. A.; Drysdale, M. J.; Workman, P. Curr. Top.
Med. Chem. 2006, 6, 1193.
24. Wang, M.; Shen, G.; Blagg, B. S. J. Bioorg. Med. Chem. Lett. 2006, 169, 2459.
25. Holland, M.; Langton, P. D.; Standen, N. B.; Boyle, J. P. Br. J. Pharmacol. 1996,
117, 119.
tumor cell lines. Hypoxia induced factor 1
genic protein regulated by HSP90.29 Inhibition of HSP90 causes
degradation of Hif-1 and abolishes Hif-1 regulated transcription.
To examine the effect of HSP90 inhibitor 12g on the degradation of
Hif-1 , we developed a luciferase reporter assay in which the eno-
lase promoter (Hif-1 regulated) was used to drive the expression
of a reporter construct containing the luciferase gene. The Hif-1
regulated luciferase reporter construct was integrated into NSCLC
cell line H1299. Inhibition of Hif-1 mediated transcription results
a (Hif-1a) is an onco-
a
a
a
a
a
a
in a decrease in fluorescent signal. As shown in Table 3 and 17-AAG
(1), VER-49009 (3), and benzimidazolone 12g efficiently inhibit
luciferase activity with EC50 values of 120, 210, and 190 nM,
respectively.
It has been shown that HSP90 inhibition also leads to ErbB2
degradation and HSP70 up-regulation.30, 31 An in-cell western ap-
proach was developed for rapidly monitoring both degradation of
ErbB2 and induction of HSP70 in the SKBR3 cell line known for
its high overexpression of ErbB2.28 The results summarized in Ta-
ble 3 shows 12g to be equipotent to 17-AAG in modulating Erb2
and HSP70 in a cellular context.
26. Sharp, S. Y.; Prodromou, Ch.; Boxall, K.; Powers, M. V.; Holmes, J. L.; Box, G.;
Matthews, T. P.; Cheung, K.-M. J.; Kalusa, A.; James, K.; Hayes, A.; Hardcastle,
A.; Dymock, B.; Brough, P. A.; Barril, X.; Cansfield, J. E.; Wright, L.; Surgenor, A.;
Foloppe, N.; Hubbard, R. E.; Aherne, W.; Pearl, L.; Jones, K.; McDonald, E.;
Raynaud, F.; Eccles, S.; Drysdale, M.; Workman, P. Mol. Cancer Ther. 2007, 64,
1198.
27. Dymock, B. W.; Barril, X.; Brough, P. A.; Cansfield, J. E.; Massey, A.; McDonald,
E.; Hubbard, R. E.; Surgenor, A.; Roughley, S. D.; Webb, P.; Workman, P.;
Wright, L.; Drysdale, M. J. J. Med. Chem. 2005, 48, 4212.
28. Xu, W.; Yuan, X.; Jung, Y. J.; Yang, Y.; Basso, A.; Rosen, N.; Chung, E. J.; Trepel, J.;
Neckers, L. Cancer Res. 2003, 63, 7777.
29. Minet, E.; Mottet, D.; Michel, G.; Roland, I.; Raes, M.; Remacle, J.; Michiels, C.
FEBS Lett. 1999, 460, 251.
30. Xu, W.; Mimnaugh, E.; Rosser, M. F. N.; Nicchitta, Ch.; Marcu, M.; Yarden, Y.;
Neckers, L. J. Biol. Chem. 2001, 276, 3702.
31. Elo, M. A.; Kaarniranta, K.; Helminen, H. J.; Lammi, M. J. Biochim. Biophys. Acta,
Mol. Cell Res. 2005, 1743, 115.
32. Chene, P.; Floersheimer, A.; Furet, P.; Schoepfer, J. WO 2006010594 A1.
In summary, we describe here the discovery of a novel series of
N-aryl-benzimidazolone HSP90 inhibitors targeting the N-terminal
ATP-ase site. Interestingly, Novartis has developed a parallel series
of benzimidazolone analogs32 that appears to be based on the same