Fragment-based discovery of hydroxy-indazole-carboxamides as novel small molecule inhibitors of Hsp90

Article abstract of DOI:10.1016/j.bmcl.2012.04.121

Inhibitors of the Hsp90 molecular chaperone are showing considerable promise as potential molecular therapeutic agents for the treatment of cancer. Here we describe the identification of novel small molecular weight inhibitors of Hsp90 using a fragment based approach. Fragments were selected by docking, tested in a biochemical assay and the confirmed hits were crystallized. Information gained from X-ray structures of these fragments and other chemotypes was used to drive the fragment evolution process. Optimization of these high μM binders resulted in 3-benzylindazole derivatives with significantly improved affinity and anti-proliferative effects in different human cancer cell lines.

Full text of DOI:10.1016/j.bmcl.2012.04.121

Bioorganic & Medicinal Chemistry Letters 22 (2012) 4396–4403
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Fragment-based discovery of hydroxy-indazole-carboxamides as novel
small molecule inhibitors of Hsp90
⇑
Hans-Peter Buchstaller , Hans-Michael Eggenweiler, Christian Sirrenberg, Ulrich Grädler, Djordje Musil,
Edmund Hoppe, Astrid Zimmermann, Harry Schwartz, Joachim März, Jörg Bomke, Ansgar Wegener,
Michael Wolf
Merck Serono Research, Merck KGaA, Frankfurter Straße 250, D-64293 Darmstadt, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Inhibitors of the Hsp90 molecular chaperone are showing considerable promise as potential molecular
therapeutic agents for the treatment of cancer. Here we describe the identification of novel small molec-
ular weight inhibitors of Hsp90 using a fragment based approach. Fragments were selected by docking,
tested in a biochemical assay and the confirmed hits were crystallized. Information gained from X-ray
structures of these fragments and other chemotypes was used to drive the fragment evolution process.
Received 2 April 2012
Revised 26 April 2012
Accepted 29 April 2012
Available online 5 May 2012
Optimization of these high
improved affinity and anti-proliferative effects in different human cancer cell lines.
lM binders resulted in 3-benzylindazole derivatives with significantly
Dedicated to Professor Christian R. Noe on
the occasion of his 65th birthday
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
Hsp90
Heat shock protein 90
Anti-cancer drug
Fragments
Indazoles
Docking
X-ray structure
Molecular chaperones are proteins, which play a key role in the
conformational maturation, stability and function of other client
protein substrates within the cell.1,2 Many of the client proteins of
Heat shock protein 90 (Hsp90),3–6 which are involved in signal
transduction, cell cycle regulation and apoptosis, are well known
oncogenes and are often deregulated (over-expressed and mutated)
in tumor cells, and this dysregulation of pathways involving these
proteins are commonly associated with cancer pathology.7 The
association of Hsp90 with these client proteins maintains their abil-
ity to function in the deregulated state. Therefore, Hsp90 inhibitors
target tumor growth by multiple parallel mechanisms. Hsp90 has
solubility, limited bioavailability and hepatotoxicity. These issues
have led to significant efforts to identify small-molecule inhibitors.
Indeed, several compounds of different chemical classes have been
disclosed so far, like ganetespib (STA-9090),11 NVP-AUY922,12
AT-13387,13 Debio-0932 (CUDC-305)14 or PU-H7115 which have
entered trials in different phases of clinical development (Fig. 1).
A comprehensive overview has been published recently.16
Finding novel compounds as starting points for optimization is
a major challenge in drug discovery research. Fragment-based ap-
proaches have rapidly become a proven technique to identify such
starting points in a variety of research programs.17 Furthermore,
the optimization of fragment-like hits using structural information
from protein X-ray crystallography has been established as a valu-
able strategy in the search for new drug molecules of various tar-
gets including Hsp90.18 Herein, we describe the identification of
low affinity indazoles and their optimization to potent inhibitors
of the N-terminal ATP binding site of Hsp90.
attracted considerable interest as
a therapeutic target for
anticancer drugs since it was shown that both geldanamycin8 and
radicicol9 are able to inhibit Hsp90 function by binding to an ATP
binding pocket in the N-terminal domain of the protein. Based on
encouraging preclinical data several derivatives of these natural
product inhibitors like 17-AAG, 17-DMAG and the prodrug of
17-AAG, IPI-504 entered clinical studies.10 However, these
compounds have several potential limitations, including poor
A subset of about 64,000 compounds was preselected from our
compound library by filtering with calculated fragment-like prop-
erties in adaption to the Astex ‘rule of three’ (e.g., MW <300).19 This
subset was used for virtual screening by docking in the ATPase
binding site of Hsp90. The analysis of the X-ray structures of
Hsp90 in complex with reference compounds (phenole and purine
⇑
Corresponding author. Tel.: +49 6151724092; fax: +49 6151723129.
E-mail address: hans-peter.buchstaller@merckgroup.com (H.-P. Buchstaller).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.04.121

H.-P. Buchstaller et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4396–4403
4397
lipophilic pocket has resulted in Hsp90 inhibitors with significant
biochemical and cellular potency.24 Consequently, we used a rep-
resentative structure of the helical form (PDB code 1UYF) as rigid
protein model for docking in order to also identify fragments
potentially binding into this lipophilic pocket.
N
N
N
N
O
O
O
N
O
N
N
The composite inhibitor interactions served as pharmacophore
constraints (Fig. 2) for the subsequent docking of a preselected
subset of ꢀ64,000 compounds from our corporate library using
the program FlexX.25 The pharmacophore contraints (see Supple-
mentary data) were used as filters to select only fragments with
suitable H-bonds to either Asp93 or the conserved water molecule
in addition to potential hydrophobic contacts within the lipophilic
pocket of the helical form which resulted in 3810 hits. These hits
were ranked by a combination of two alternative docking scores
(FlexX- and PLP-scoring functions), clustered by chemical sub-
structures and the docking poses of top scored hits from each clus-
ter were finally visually inspected. A subset of 96 fragments was
finally selected for testing in a Hsp90 binding assay26 revealing five
N
NH
N
O
HO
OH
HO
OH
HO
OH
NVP-AUY922
AT-13387
ganetespib
(STA-9090)
O
O
O
O
NH2
NH2
N
N
N
N
N
N
N
I
S
N
S
N
N
Debio-0932
PU-H71
hits with an IC50-value below 100 lM (5% hit rate). Three of the
hits were structurally similar, low molecular weight indazoles
and the remaining two hits contained an aminopyrimidine or qui-
nazoline moiety (Fig. 3a). The binding mode of all hits could be elu-
cidated by X-ray crystallography using the ATPase domain of
human Hsp90. A superimposition of the crystal structures re-
vealed, that fragments III and V bind to the helical protein form
with their bulky hydrophobic substituents (cyclohexyl and p-chlo-
rophenyl respectively) oriented into the same lipophilic pocket as
the substituted phenyl ring of Hsp90-purine complexes (e.g., as
in PDB code 1UYF). In contrast, the three other fragments I, II
and VI bind to the open Hsp90 conformation as present in the gel-
danamycin complex (Fig. 3b). Interestingly, X-ray structures con-
firmed that the hydroxy-indazoles bind in two reversed
orientations in contact with Asp93 as predicted by our pharmaco-
phore docking protocol (Figs. 4 and 5). Fragment II interacts via H-
bonds of both indazole N-atoms to Asp93 and the conserved water
molecule, while an identical interaction pattern is found for the
phenolic OH-group of fragment III. The reversed orientation might
be influenced by the substitution pattern of the hydroxindazole
core as the cyclohexyl ring of fragment III is bound in the lipophilic
Figure 1. Selected structures of Hsp90 inhibitors in clinical trials.
derivatives) published at the time, when our virtual screening
campaign was conducted, revealed crucial interactions within the
ATP-site.20,21 These interactions include hydrogen bonds of the
phenolic OH-group and purine N-atoms to Asp93 and an adjacent
conserved water-molecule (Fig. 2). The assigned hydrogen bond
network of the phenolic OH-group in the Hsp90-complex sug-
gested a donor contact to Asp93 and an acceptor interaction to
the interstitial water molecule. Similarly, the purine amino group
forms an H-bond donor contact to Asp93 and the N1-atom an
acceptor contact to the conserved water-molecule. A large induced
fit characterized by the transition of an open, ‘apo-like’ form as in
the geldanamycin complex to a helical form upon binding of purine
derivatives was observed in the rear ATP binding site of Hsp90.22,23
This transition results in the presence of a large lipophilic pocket
formed by the side chains of Met98, Leu107, Phe138, Tyr139,
Val150 and Trp162 in the helical form. As demonstrated for purine
analogs, the optimization of hydrophobic interactions within this
Figure 2. H-bond interactions (dashed lines) of published Hsp90 inhibitors: purine-derivative in ‘helical form’ (PDB-code 1UYF, magenta) and resorcinol-derivative (PDB-
code 1YC1, orange) in ‘open form’ of Hsp90. Crucial contacts were used as pharmacophore filters (H-donor: blue, H-acceptor: red, lipophilic: golden) for virtual screening by
docking.

4398
H.-P. Buchstaller et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4396–4403
(a)
O
Cl
N
N
Cl
N
N
N
N
N
O
N
N
O
O
Cl
N
N
N
N
Fragment I
MW 202
Fragment II
MW 226
Fragment III
MW 269
Fragment IV
MW 270
Fragment V
MW 220
IC50: 87 µM
LE=0.37
IC50: 45 µM
LE=0.35
IC50: 85 µM
LE=0.28
IC50: 0.9 µM
LE=0.43
IC50: 21 µM
LE=0.40
BEI=20.1
BEI=19.2
BEI=15.1
BEI=22.4
BEI=18.4
(b)
Figure 3. (a) Chemical structures of identified fragment hits with calculated ligand efficiency (LE)37 and binding efficiency index (BEI).38 (b) Superimposition of Hsp90 X-ray
structures with all 5 fragment hits revealed the helical form for fragments III and V and the open conformation for fragments I, II and IV.
pocket of the helical form. In contrast, this lipophilic pocket is not
adressed by fragment II, but its phenole substituent forms -inter-
3-nitrophenylacetyl chloride followed by the conversion of the ni-
tro group of 2b via a reduction/diazotization/substitution protocol
to gain compound 2c. Cyclization to 3b and subsequent formation
of the benzyl-methyl-amide 24 were performed as described for
compounds 4.
p
actions to Met98 towards the solvent entrance of the ATP-site. This
flip in orientation confirms the observation that subtle changes in
the chemistry of a fragment can induce a change in compound ori-
entation, which has been reported recently for similar indazole
fragments.18
For all other 3-benzyl derivatives (12, 19–23, 25–28) and the 3-
phenethyl derivative 29 an alternative synthesis route was estab-
lished as depicted in Scheme 2. 2-Bromo-5-fluoro-anisole turned
out to be the reagent of choice, because its increased reactivity
for Friedl–Crafts acylations allowed the conversion to the corre-
sponding ketones 6 at low temperature (À50 °C) at which side
reactions of the acyl halide could be avoided. Only the synthesis
of ketone 7, precursor of the 3-phenethyl derivative 29, by acyla-
tion with 3-phenyl-propionyl chloride remained difficult due to
the preferred intramolecular cyclization to indane-1-one. Forma-
tion of the indazole core by cyclization of 6 and 7 with hydrazine
in dioxane afforded compounds 8 and 9. These compounds were
key intermediates for the preparation of the methoxy-substituted
derivative 12, the hydroxyl-indazoles 19–23 and 25–29, respec-
tively. Compound 12 was obtained from 8 via palladium-catalyzed
methoxy-carbonylation using Pd(dppf)2Cl2ÁCH2Cl2 as catalyst fol-
lowed by saponification of the ester 10 to the corresponding acid
The indazole analogs designed to obtain structure–activity rela-
tionships were synthesized by following the synthetic sequences
summarized in Schemes 1 and 2. Thus, compounds 4a–z were ob-
tained from 2-hydroxy-4-fluoro-benzoic acid 1a, which was ester-
ified with methyl iodide in dimethylformamide and subsequently
Friedl–Crafts acylated at position 5 with phenylacetyl chloride at
0 °C to give 2a. Cyclization of 2a with hydrazine in dioxane affor-
ded the key intermediate 3a. After saponification of methyl ester
3a to the corresponding acid, the final indazole carboxamides
4a–z were assembled using either amide coupling chemistry or
activation via the acyl chloride.27,28 Compounds with a substituent
at the benzyl moiety were not accessible via this protocol, because
the Friedl–Crafts acylation with substituted phenylacetyl chlorides
failed. Only compound 24 bearing a methoxy group at the 3-posi-
tion of the benzyl residue could be obtained through acylation with

H.-P. Buchstaller et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4396–4403
4399
Figure 4. X-ray structure of fragment II in complex with Hsp90 (magenta, PDB-code: 4EEH, PDB deposition number: RCSB071515, resolution 1.60 Å, Rwork/Rfree 0.185/0.212)
in comparison with the best docking solution (green).
Figure 5. X-ray structure of fragment III in complex with Hsp90 (magenta, PDB-code: 4EFT, PDB deposition number: RCSB071563, resolution 2.12 Å, Rwork/Rfree 0.206/0.230)
in comparison with the best docking solution (green).
R3
R2
R
O
O
N
R1
R1
O
O
O
F
R1
b
d
e,f
O
O
O
N
N
N
N
HO
HO
HO
1a: R = H
F
HO
H
H
2a: R1 = H
3a: R1 = H
3b: R1 = 3-OCH3
4a-z: R1 = H
2b: R1 = 3-NO2
23: R1 = 3-OCH3
a
c
1b: R = CH3
2c: R1 = 3-OCH3
Scheme 1. Reagents and conditions: (a) KHCO3, CH3I, DMF, 40 °C, 3 h; (b) (i) R1-PhCH2COCl, AlCl3, CH2Cl2, 0 °C?rt, over night; (ii) 1 N HCl, 0 °C; (c) (i) H2/Raney-Ni, THF, rt,
3 h; (ii) nitrosonium tetrafluoroborate, ACN, 0 °C, 2.5 h; (iii) CF3COOH, MeOH, 60 °C, 1 h; (d) (NH2)2ÁH2O, dioxane, reflux, 1–3 h; (e) NaOH, dioxane, reflux, 1 h; (f) R2R3NH,
EDCIÁHCl, HOBtÁH2O, DIPEA, DMF, rt, over night; or (i) SOCl2, THF, rt; (ii) R2R3NH, DIPEA, THF, rt, 1–3 h.

4400
H.-P. Buchstaller et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4396–4403
R1
R1
O
F
n
n
R1
Br
Br
Br
Br
c
d
a or b
n
N
N
N
N
O
HO
O
F
O
H
H
n=1: 6
n=2: 7
n=1: 8
n=2: 9
n=1: 13
n=2: 14
5
e
e
X
X
O
O
R1
R1
n
O
O
N
N
N
N
O
HO
n=1:
H
H
n=2:
15: X = CH3 17: X = CH3
10: X = CH3
f
11: X = H
f
f
16: X = H
18: X = H
g
g
R3
R2
R3
R2
N
O
N
R1
R1
n
O
O
N
N
N
H
N
H
HO
12
n=1: 19-23, 25-28
n=2: 29
Scheme 2. Reagents and conditions: (a) (i) R1-PhCH2COCl, AlCl3, CH2Cl2, À50?À10 °C–rt, 1–16 h; (ii) 1 N HCl, 0 °C; (b) (i) Ph(CH2)2COCl, AlCl3, CH2Cl2, À55?À40 °C, 30 min;
(ii) 1 N HCl, 0 °C; (c) (NH2)2ÁH2O, dioxane, reflux, 1.5–3 h; (d) BBr3, CH2Cl2, rt, 16–60 h; (e) Pd(dppf)2Cl2ÁCH2Cl2, 4 bar CO, Et3N, MeOH/toluene, 100 °C, over night; (f) NaOH,
dioxane, reflux, 1 h; (g) (i) SOCl2, THF, rt; (ii) R2R3NH, DIPEA, THF, rt, 1 h.
11 and formation of the final amide 12. Attempts to directly get
intermediates 16 from compounds 10 via cleavage of the ester
and methoxy group at once under acidic conditions failed. For this
reason, we established an alternative protocol for the preparation
of derivatives 19–23 and 25–29. Cleavage of the methylether of
compounds 8 and 9 with boron tribromide in dichloromethane
afforded the hydroxyl-indazols 13 and 14, which were further
derivatized using the established reaction sequence for compound
12.27,29
Information gained from X-ray structures of each of the inda-
zole fragments and other in-house Hsp90 inhibitors was used to
drive the fragment evolution process. Since both the NH as well
as the OH group forms hydrogen bonds to Asp93 and a network
of conserved water molecules, we expected them to be pivotal
for the binding mode and therefore these structural features were
retained. A ligand-induced lipophilic pocket was observed in the
helical form of Hsp90 upon binding of purine derivatives22,23 or
our in-house resorcinols, which bear substituted amides in the
ortho-position of one hydroxyl group.30 Based on the bioisosteric
relationship of indazoles and phenols, which has been exemplified
for other targets,31–33 we expected that substituted amides adja-
cent to the hydroxyl group would address this lipophilic pocket.
Thus, we focused on the modification at two positions of the inda-
zole core to gain SAR.
methyl-benzyl group at the amide nitrogen. Interestingly, the 3-
phenethyl derivative 29, which is a close analog of the initial frag-
ment III, is very potent in the Hsp90 binding assay (IC50 = 230 nM).
However, this compound was also less potent (about fivefold) in the
viability assay. Thus, we kept the benzyl residue at the 3-position
and further investigated the amide SAR.
Our initial variation clearly revealed a methyl-benzyl group (4e)
as most favorable amide substituent. Thus, we kept this moiety
and explored its substitution pattern with a diverse set of substit-
uents in the ortho-, meta- and para position (Table 2). Although we
could improve the biochemical potency for some derivatives such
as 4s, 4v or 4x, the unsubstituted compound 4e remained the best
compound in the viability assay using A2780 cells with an IC50 va-
lue of about 1 lM. Based on these data we already surmised that
such indazole amides may bind in a reversed orientation compared
to the resorcinol-amides.30 Next, we explored the SAR of the benzyl
moiety at the 3-position of the indazole core. These efforts clearly
showed that small, lipophilic substituents in the meta-position are
favorable resulting in several compounds with improved activity in
the Hsp90 binding assay and more importantly cellular activity
(22a, IC50 = 0.41 lM; 23b, IC50 = 0.49 lM). For all other analogs
(19–21, 24–28) bearing various substituents at different positions
of the benzyl ring the binding affinity was more or less unaffected.
We also examined the influence of the second substituent at the
amide nitrogen. A methyl group (23b) represents the optimum at
this position, while a partial decrease in activity in the Hsp90 bind-
ing assay was observed for compound 23a (R2 = H) and 23c
(R2 = ethyl). We assumed that the methyl residue is required to
keep the amide in the right orientation and fits into a narrow
sub-pocket at this area of the ATP-site. Interestingly, a total loss
of activity still occurred when the hydroxyl group of the optimized
compound 23b was changed from a donor to an acceptor through
methylation (12).
Comparison of the activities of the initial fragments and com-
pound 4a in the Hsp90 binding assay revealed that an amide resi-
due in ortho position of the hydroxy group is tolerated, in
principal. Furthermore, introduction of tertiary amides showed a
tractable enhancement of activity in the Hsp90 binding assay
(4b–4f, Table 1). Compounds 4b, 4c with n-alkyl chains have an
activity in the single-digit
lM range, which is a 45-fold improve-
ment in potency compared to fragment III. We expected a furanyl
residue to further enhance potency by forming additional contacts
as observed for PU derivatives. Indeed, for compound 4f an IC50 of
610 nM was found, however this derivative was less potent in the
viability assay26 with A2780 cells than compound 4e bearing a
A real breakthrough for further optimization work was the
elucidation of the X-ray structure of compound 23b, which was
solved at 2.0 Å resolution (Fig. 6). This crystal structure revealed

H.-P. Buchstaller et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4396–4403
4401
Table 1
Biological data of 6-hydroxy-1H-indazole-5-carboxylic acid amide derivatives 4a–4f and 29
R3 R2
R1
N
N
O
N
HO
H
4a-f, 29
R1
R2
R3
Hsp90 binding assay IC50
(
lM)
A2780 viability assay IC50 (lM)
a
a
Compound
4a
4b
4c
4d
4e
4f
Bzl
Bzl
Bzl
Bzl
Bzl
Bzl
(CH2)2Ph
H
H
61
nd
CH3
CH3
CH3
CH3
CH3
CH3
n-C3H7
n-C4H9
(CH2)2OCH3
Bzl
2-Furanyl
Bzl
1.9
1.8
4
0.66
0.61
0.23
>10
7.6
nd
0.98
2.5
4.6
29
a
For assay conditions and determination of IC50 values see Supplementary data.
Table 2
Summary of biological data of 3-benzyl-6-hydroxy-1H-indazole-5-carboxylic acid amide derivatives 4, 12 and 19–28
R2
R2
R1
R1
N
O
N
R3
R3
O
O
N
N
N
N
H
HO
H
12
R3
4, 19-28
R1
R2
Hsp90 binding assay IC50
(
lM)
A2780 viability assay IC50
(l
M)
a
a
Compound
4e
4g
4h
4i
4j
4k
4l
4m
4n
4o
4p
4q
4r
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
H
H
2-F
2-Cl
2-Br
2-CH3
2-OCH3
3-F
3-Cl
3-Br
3-CH3
3-OCH3
3-CF3
4-F
4-Cl
4-Br
4-CH3
4-C2H5
0.66
0.9
1.1
0.61
0.74
1.7
0.57
1.5
1.3
3.6
0.98
3.4
1.1
0.98
1.5
1.1
1.2
4.4
5.4
5
5.3
7.1
nd
3.8
nd
4.1
2.6
2.8
6.9
9.4
3.2
4.8
4.1
>10
nd
4s
4t
0.33
1.1
4u
4v
4w
4x
4y
0.63
0.39
0.76
0.47
2.3
0.98
>20
2.6
2.4
1.3
0.4
1.5
0.25
3.0
1.4
1.3
5.7
H
H
H
H
4-OC2H5
2,3-OCH3
3-OCH3, 4-Cl
3-F, 4-Cl
4z
H
12
19
20
21
22a
23a
23b
23c
24
25
26
27
28
3-CH3
2-Cl
2-CH3
3-F
3-Cl
3-CH3
3-CH3
3-CH3
3-OCH3
3-C2H5
4-Cl
4-CH3
4-C2H5
H
H
H
H
H
H
H
H
H
H
H
H
H
nd
nd
1.7
0.41
>10
0.49
nd
nd
nd
nd
>10
>10
CH3
C2H5
CH3
CH3
CH3
CH3
CH3
1.3
2.5
a
For assay conditions and determination of IC50 values see Supplementary data.
that such indazole amides bind according to the orientation of
fragment III rather than fragment II. The 3-methylbenzyl substitu-
ent is deeply buried in the lipophilic pocket of the helical form
with vdW-contacts to Met98, Leu103, Phe138, Val150 and
Trp162. The aromatic face on one side of the second benzyl group
of 23b is bent towards the indazole ring with an interplanar angle
of ꢀ50° and in lipophilic contact with Leu107. The aromatic face
on the rear side is located at the surface of the 23b complex and
oriented towards the solvent, which allows the introduction of di-
verse residues.

4402
H.-P. Buchstaller et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4396–4403
Figure 6. X-ray structure of compound 23b bound to ATP-site of Hsp90 (PDB-code 4EFU, PDB deposition number: RCSB071564, resolution 2.0 Å, Rwork/Rfree 0.204/0.223) with
hydrogen bonds shown as dashed lines.
Table 3
Summary of biological data of 3-benzyl-6-hydroxy-1H-indazole-5-carboxylic acid amide derivatives 22 and 23
N
R2
R1
O
N
N
HO
H
22, 23
Compound
R1
R2
Hsp90 binding assay IC50
(l
M)a
A2780 viability assay IC50 (l
M)a
22a
22b
22c
22d
22e
22f
22g
22h
23b
23d
23e
23f
23g
23h
23i
23j
23k
23l
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
H
2-Br
3-F
4-F
4-OCH3
3,4-OCH3
3-NHSO2CH3
4-CH2NHSO2CH3
H
0.4
0.41
0.13
0.44
0.41
0.39
2.4
0.59
0.71
0.8
0.8
0.73
0.41
0.06
0.25
0.78
0.37
0.58
0.46
0.17
0.68
0.58
0.5
3.2
0.57
0.49
0.31
0.44
0.59
0.54
0.93
1.6
3.7
0.53
4.0
1.6
0.71
2-Br
3-F
4-F
4-OCH3
3-F, 4-Cl
3-OCH3, 4-Cl
3,4-OCH3
3,4-F
3-NHSO2CH3
4-NHSO2CH3
4-CH2NHSO2CH3
0.2
0.35
0.15
23m
23n
a
For assay conditions and determination of IC50 values see Supplementary data.
The impact on activity of the substitution at this phenyl ring
relatively bulky bromine atom in the ortho position (22b, 23d)
slightly improved the cellular activity. Interestingly, the X-ray
structure of compound 23b indicated the presence of a sulfate
ion from the crystallization buffer in the electron density map in
ꢀ4 Å distance of the surface exposed benzyl group. This structural
information together with molecular modeling studies prompted
us to introduce sulfonamide substituents in the para-position of
the second benzyl group. The resulting 4-methylsulfonamide
derivative 22h (IC50 = 59 nM) is >1000-fold more potent than the
initial fragment and biochemically the most potent compound in
was subject of the preparation of the next set of compounds. As
compounds with a methyl or chloro substituent at the meta-posi-
tion showed the best cellular activity, we adhered to these residues
at position R1 and explored modifications at position R2. The bio-
logical data of these derivatives are reported in Table 3. In general,
the SAR for both subseries is fairly similar, but the compounds with
the chloro substituent in position R1 are more potent in the viabil-
ity assay. Small residues like fluoro or methoxy in meta or para po-
sition are well tolerated (22c–e, 23e–g). Introduction of the

H.-P. Buchstaller et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4396–4403
4403
M)
Table 4
Antiproliferative activities and Hsp70 up-regulation for selected compounds
a
a
Compound
Viability assays IC50
(l
M)
Hsp70 up-regulation IC50
A2780
(l
A2780
PC-3
MCF-7
Colo-205
22h
23n
0.57
0.71
4.3
>10
3.5
4.7
9.5
3.7
1
1.1
a
For assay conditions and determination of IC50 values see Supplementary data.
16. Messaoudi, S.; Peyrat, J.-F.; Brion, J.-D.; Alami, M. Expert Opin. Ther. Patents
2011, 21, 1501.
17. Foloppe, N. Future Med. Chem. 2011, 3, 1111.
18. Roughley, S. D.; Hubbard, R. E. J. Med. Chem. 2011, 54, 3989.
19. Congreve, M.; Carr, R.; Murray, C.; Jhoti, H. Drug Discovery Today 2003, 8, 876.
20. Barril, X.; Brough, P.; Drysdale, M.; Hubbard, R. E.; Massey, A.; Surgenor, A.;
Wright, L. Bioorg. Med. Chem. Lett. 2005, 15, 5187.
this series. However, as already observed by other groups, this did
not translate accordingly to cellular activity.34–36 Nevertheless, the
most potent compounds, 22h and 23n, showed reasonable antipro-
liferative activity in different cancer cell lines and caused the ex-
pected up-regulation of Hsp70 in A2780 cells (Table 4).
In conclusion, we have reported the identification of the inda-
zole scaffold for competitively inhibiting the ATP binding site of
Hsp90 by a fragment-based approach. Analysis of the X-ray struc-
tures of these fragments and other inhibitors bound to Hsp90 al-
lowed the design of compounds with significantly improved
potency. Compounds from this series displayed nM activities in
the Hsp90 binding assay, inhibited cell proliferation in different
human cancer cell lines and caused the expected up-regulation
of Hsp70 in A2780 cells. The promising data presented here have
prompted further evaluation of this scaffold and that work remains
ongoing.
21. Kreusch, A.; Han, S.; Brinker, A.; Zhou, V.; Choi, H. S.; He, Y.; Lesley, S. A.;
Caldwell, J.; Gu, X. J. Bioorg. Med. Chem. Lett. 2005, 15, 1475.
22. Wright, L.; Barril, X.; Dymock, B.; Sheridan, L.; Surgenor, A.; Beswick, M.;
Drysdale, M.; Collier, A.; Massey, A.; Davies, N.; Fink, A.; Fromont, C.; Aherne,
W.; Boxall, K.; Sharp, S.; Workman, P.; Hubbard, R. E. Chem. Biol. 2004, 11, 775.
23. Stebbins, C. E.; Russo, A. A.; Schneider, C.; Rosen, N.; Hartl, F. U.; Pavletich, N. P.
Cell 1997, 89, 239.
24. Taldone, T.; Sun, W.; Chiosis, G. Bioorg. Med. Chem. 2009, 15, 2225.
25. Rarey, M.; Kramer, B.; Lengauer, T.; Klebe, G. J. Mol. Biol. 1996, 261, 470.
26. For details of the binding assay, the cellular assay, crystallization and structure
determination, and virtual screening please see the Supplementary data.
27. Buchstaller, H.-P.; Eggenweiler, H.-M.; Wolf, M.; Sirrenberg, C. WO2008/
003396, 2008.
28. Analytical data (for LC–MS conditions see Supplementary data) for
a
representative compound 4e: 1H NMR (400 MHz, DMSO-d6, 90 °C) d (ppm):
12.15 (s, 1H), 9.67 (s, 1H), 7.36–7.10 (m, 11H), 6.82 (s, 1H), 4.53 (s, 2H), 4.17 (s,
2H), 2.76 (s, 3H). LC–MS: Rt = 2.20 min (polar gradient); ESI m/z 372.8/373.8
[M+H]+.
Supplementary data
29. Analytical data (for LC–MS conditions see Supplementary data) for representative
compounds 22a: 1H NMR (400 MHz, DMSO-d6, 90 °C) d (ppm): 12.25 (s, 1H),
9.75 (s, 1H), 7.44 (s, 1H), 7.39–7.21 (m, 9H), 6.88 (s, 1H), 4.58 (m, 2H), 4.23 (s,
2H), 2.82 (s, 3H). LC–MS: Rt = 1.76 min (regular gradient); ESI m/z 406.8/407.8/
408.7 [M+H]+; 22h: 1H NMR (400 MHz, DMSO-d6, 90 °C) d (ppm): 12.21 (s, 1H),
9.70 (s, 1H), 7.41 (s, 1H), 7.37–7.14 (m, 9H), 6.84 (s, 1H), 4.53 (s, 2H), 4.20 (s,
2H), 4.15 (d, 2H), 2.82 (s, 3H), 2.76 (s, 3H). LC–MS: Rt = 1.43 min (regular
gradient); ESI m/z 513.2/514.2/515.2/516.2 [M+H]+; 23b: 1H NMR (400 MHz,
DMSO-d6, 90 °C) d (ppm): 12.14 (s, 1H), 9.67 (s, 1H), 7.37–7.20 (m, 6H), 7.16–
6.92 (m, 4H), 6.82 (s, 1H), 4.53 (s, 2H), 4.12 (s, 2H), 2.76 (s, 3H), 2.22 (s, 3H). LC–
MS: Rt = 1.72 min (regular gradient); ESI m/z 386.8/387.8 [M+H]+; 23n: 1H
NMR (400 MHz, DMSO-d6, 90 °C) d (ppm): 12.14 (s, 1H), 9.67 (s, 1H), 7.39–7.19
(m, 6H), 7.17–6.93 (m, 4H), 6.82 (s, 1H), 4.52 (s, 2H), 4.19–4.09 (m, 4H), 2.82 (s,
3H), 2.75 (s, 3H), 2.23 (s, 3H). LC–MS: Rt = 1.39 min (regular gradient); ESI m/z
493.2/494.2 [M+H]+.
30. Eggenweiler, H.-M.; Sirrenberg, C.; Buchstaller, H.-P.; Wolf, M.; Grädler, U.;
März, J.; Musil, D.; Hoppe, E.; Zimmermann, A.; Amendt, C. Wegener, A.;
Schwartz, H.; Bomke, J. 240th ACS National Meeting, Boston, MA, United States,
August 22–26, 2010; 2010; MEDI-505.
31. Boehm, H. J.; Boehringer, M.; Bur, D.; Gmuender, H.; Huber, W.; Klaus, W.;
Kostrewa, D.; Kuehne, H.; Luebbers, T.; Meunier-Keller, N.; Mueller, F. J. Med.
Chem. 2000, 43, 2664.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2012.04.121.
These data include MOL files and InChiKeys of the most important
compounds described in this article.
References and notes
1. Young, J. C.; Agashe, V. R.; Siegers, K.; Hartl, F. U. Nat. Rev. Mol. Cell Biol. 2004, 5,
781.
2. Mosser, D. D.; Morimoto, R. I. Oncogene 2004, 23, 2907.
3. Whitesell, L.; Lindquist, S. L. Nat. Rev. Cancer 2005, 5, 761.
4. Maloney, A.; Workman, P. Expert Opin. Biol. Ther. 2002, 2, 3.
5. Kamal, A.; Boehm, M. F.; Burrows, F. J. Trends Mol. Med. 2004, 10, 283.
6. Chiosis, G.; Vilenchik, M.; Kim, J.; Solit, D. Drug Discovery Today 2004, 9, 881.
7. Hanahan, D.; Weinberg, R. A. Cell 2000, 100, 57.
8. Neckers, L.; Schulte, T. W.; Mimnaugh, E. Invest. New Drugs 1999, 17, 361.
9. Soga, S.; Shiotsu, Y.; Akinaga, S.; Sharma, S. V. Curr. Cancer Drug Targets 2003, 3,
359.
10. Banerji, U.; O’Donnell, A.; Scurr, M.; Pacey, S.; Stapleton, S.; Asad, Y.; Simmons,
L.; Maloney, A.; Raynaud, F.; Campbell, M.; Walton, M.; Lakhani, S.; Kaye, S.;
Workman, P.; Judson, I. J. Clin. Oncol. 2005, 23, 4152.
11. Wang, Y.; Trepel, J. B.; Neckers, L. M.; Giaccone, G. Curr. Opin. Invest. Drugs
2010, 11, 1466.
32. Bamborough, P.; Angell, R. M.; Bhamra, I.; Brown, D.; Bull, J.; Christopher, J. A.;
Cooper, A. W. J.; Fazal, L. H.; Giordano, I.; Hind, L.; Patel, V. K.; Ranshaw, L. E.;
Sims, M. J.; Skone, P. A.; Smith, K. J.; Vickerstaff, E.; Washington, M. Bioorg. Med.
Chem. Lett. 2007, 17, 4363.
12. Brough, P. A.; Aherne, W.; Barril, X.; Borgognoni, J.; Boxall, K.; Cansfield, J. E.;
Cheung, K.-M. J.; Collins, I.; Davies, N. G. M.; Drysdale, M. J.; Dymock, B.; Eccles,
S. A.; Finch, H.; Fink, A.; Hayes, A.; Howes, R.; Hubbard, R. E.; James, K.; Jordan,
A. M.; Lockie, A.; Martins, V.; Massey, A.; Matthews, T. P.; McDonald, E.;
Northfield, C. J.; Pearl, L. H.; Prodromou, C.; Ray, S.; Raynaud, F. I.; Roughley, S.
D.; Sharp, S. Y.; Surgenor, A.; Walmsley, D. L.; Webb, P.; Wood, M.; Workman,
P.; Wright, L. J. Med. Chem. 2008, 51, 196.
33. Folkes, A. J.; Ahmadi, K.; Alderton, W. K.; Alix, S.; Baker, S. J.; Box, G.;
Chuckowree, I. S.; Clarke, P. A.; Depledge, P.; Eccles, S. A.; Friedman, L. S.; Hayes,
A.; Hancox, T. C.; Kugendradas, A.; Lensun, L.; Moore, P.; Olivero, A. G.; Pang, J.;
Patel, S.; Pergl-Wilson, G. H.; Raynaud, F. I.; Robson, A.; Saghir, N.; Salphati, L.;
Sohal, S.; Ultsch, M. H.; Valenti, M.; Wallweber, H. J. A.; Wan, N. C.; Wiesmann,
C.; Workman, P.; Zhyvoloup, A.; Zvelebil, M. J.; Shuttleworth, S. J. J. Med. Chem.
2008, 51, 5522.
13. Woodhead, A. J.; Angove, H.; Carr, M. G.; Chessari, G.; Congreve, M.; Coyle, J. E.;
Cosme, J.; Graham, B.; Day, P. J.; Downham, R.; Fazal, L.; Feltell, R.; Figueroa, E.;
Frederickson, M.; Lewis, J.; McMenamin, R.; Murray, C. W.; O’Brian, M. A.;
Parra, L.; Patel, S.; Phillips, T.; Rees, D. C.; Rich, S.; Smith, D.-M.; Trewartha, G.;
Vinkovic, M.; Williams, B.; Woolford, A. J.-A. J. Med. Chem. 2010, 53, 5956.
14. Bao, R.; Lai, C.-J.; Qu, H.; Wang, D.; Yin, L.; Zifcak, B.; Atoyan, R.; Wang, J.;
Samson, M.; Forrester, J.; DellaRocca, S.; Xu, G.-X.; Xu, T.; Zhai, H.-X.; Cai, X.;
Qian, C. Clin. Cancer Res. 2009, 15, 4046.
34. Pratt, W. B.; Toft, D. O. Exp. Biol. Med. (Maywood) 2003, 228, 111.
35. Pearl, L. H.; Prodromou, C. Annu. Rev. Biochem. 2006, 75, 271.
36. Barta, T. E.; Veal, J. M.; Rice, J. W.; Partridge, J. M.; Fadden, P. R.; Ma, W.; Jenks,
M.; Geng, L.; Hanson, G. J.; Huang, K. H.; Barabasz, A. F.; Foley, B. E.; Otto, J.;
Hall, S. E. Bioorg. Med. Chem. Lett. 2008, 18, 3517.
37. Hopkins, A. L.; Groom, C. R.; Alex, A. Drug Discovery Today 2004, 9, 430.
38. Abad-Zapatero, C.; Metz, J. T. Drug Discovery Today 2005, 10, 464.
39. Sadowski, J. J. Comput. Aided Mol. Des. 1997, 11, 53.
15. Immormino, R. M.; Kang, Y.; Chiosis, G.; Gewirth, D. T. J. Med. Chem. 2006, 49,
4953.
40. SYBYL 8.0, Tripos International, 1699 South Hanley Rd., St. Louis, Missouri,
63144, USA.