Synthesis and in vitro antiproliferative activity of C5-benzyl substituted 2-amino-pyrrolo[2,3-d]pyrimidines as potent Hsp90 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2016.11.062

A novel series of heat shock protein 90 (Hsp90) inhibitors was identified by X-ray crystal analysis of complex structures at solvent-exposed exit pocket C. The 2-amino-pyrrolo[2,3-d]pyrimidine derivatives, 7-deazapurines substituted with a benzyl moiety at C5, showed potent Hsp90 inhibition and broad-spectrum antiproliferative activity against NCI-60 cancer cell lines. The most potent compound, 6a, inhibited Hsp90 with an IC₅₀of 36?nM and showed a submicromolar mean GI₅₀value against NCI-60 cell lines. The interaction of 6a at the ATP-binding pocket of Hsp90 was confirmed by X-ray crystallography and Western blot analysis.

Full text of DOI:10.1016/j.bmcl.2016.11.062

Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and in vitro antiproliferative activity of C5-benzyl substituted
2-amino-pyrrolo[2,3-d]pyrimidines as potent Hsp90 inhibitors
Ju-Hyeon Lee a,c,1, Sang Chul Shin b,d,1, Seon Hee Seo a, Young Ho Seo e, Nakcheol Jeong c, Chan-Wha Kim d,
Eunice EunKyeong Kim b, , Gyochang Keum a,
⇑
⇑
a Center for Neuro-Medicine, Korea Institute of Science and Technology (KIST), Hwarangro 14-gil 5, Seongbuk-gu, Seoul 02792, Republic of Korea
b Biomedical Research Institute, KIST, Hwarangro 14-gil 5, Seongbuk-gu, Seoul 02792, Republic of Korea
c Department of Chemistry, Korea University, Anam-ro 145, Seongbuk-gu, Seoul 02841, Republic of Korea
d School of Life Sciences and Biotechnology, Korea University, Seoul 02841, Republic of Korea
e College of Pharmacy, Keimyung University, Daegu 42601, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel series of heat shock protein 90 (Hsp90) inhibitors was identified by X-ray crystal analysis of
complex structures at solvent-exposed exit pocket C. The 2-amino-pyrrolo[2,3-d]pyrimidine derivatives,
7-deazapurines substituted with a benzyl moiety at C5, showed potent Hsp90 inhibition and broad-
spectrum antiproliferative activity against NCI-60 cancer cell lines. The most potent compound, 6a,
inhibited Hsp90 with an IC50 of 36 nM and showed a submicromolar mean GI50 value against NCI-60 cell
lines. The interaction of 6a at the ATP-binding pocket of Hsp90 was confirmed by X-ray crystallography
and Western blot analysis.
Received 21 October 2016
Revised 17 November 2016
Accepted 22 November 2016
Available online xxxx
Keywords:
Hsp90
Structure-based design
Cancer
Ó 2016 Published by Elsevier Ltd.
Antiproliferative
Pyrrolo[2,3-d]pyrimidines
Heat shock protein 90 (Hsp90) is expressed abundantly in the
cytoplasm of normal cells to maintain protein homeostasis. It is
an adenosine triphosphate (ATP)-dependent molecular chaperone
that plays a key role in folding, stabilization, and maturation of
the substrate client proteins involved in cell growth, differentia-
tion, and survival.1,2 Hsp90 is highly expressed in most tumor cells
and plays a significant role in the proliferation, malignant transfor-
mation, and progression of cancer cells. Hsp90 client proteins
include many oncoproteins necessary for tumor development, such
as receptor tyrosine kinases, signal transducers, cell-cycle regula-
tors, growth factors, and transcription factors.3,4
Inhibition of Hsp90 chaperone activity causes proteosomal
degradation of various client oncoproteins that are involved in
commonly deregulated pathways in cancer cells, including growth
factor, transcription factor, angiogenesis, apoptosis, cell cycle, tis-
sue invasion, metathesis, and oncogenic fusion protein.5,6
(Hsp90-addiction). Hsp90 is an attractive target for cancer therapy,
and many Hsp90 inhibitors have been evaluated in clinical trials
for the treatment of cancers driven by Hsp90 client proteins.7–10
Although Hsp90 inhibitors have shown promising antitumor
activity in clinical trials,11–16 they have also shown limited sin-
gle-agent activity due to low exposure, low solubility, and adverse
events.3,7 Combination therapy using Hsp90 inhibitors with other
targeted anticancer drugs, including drugs targeting Hsp90 clients,
showed synergistic effects by sensitizing cells or overcoming resis-
tance.17,18 The liver toxicity of first-generation geldanamycin ana-
logs, tanespimycin (17-AAG), and alvespimycin (17-DMAG), and
visual disorders of various second-generation Hsp90 inhibitors,
were the most common adverse events and major limitations in
the clinical studies.3,19 Also, neurological toxicities of BIIB021, such
as syncope and dizziness, were reported in a phase-I clinical trial.20
Among Hsp90 inhibitors discovered by structure-based design
using the X-ray crystal structures of Hsp90, radicicols (AT13387
and STA-9090), purine (PU-H71), and benzamide analogs (SNX-
5422, XL888 and TAS-116) are currently being investigated in clin-
ical trials (Fig. 1).19
Inhibition of Hsp90 can simultaneously regulate numerous tar-
gets or signal pathways essential for cancer cell survival and prolif-
eration. Cancer cells are highly reliant on this chaperoning activity
Taldone and coworkers categorized the Hsp90 inhibitors
using computational analysis of the cocrystal structure/docking
studies based on their interactions with three N-terminal binding
regions: the ATP-binding pocket A, lipophilic pocket B, and a
⇑
Corresponding author.
E-mail addresses: eunice@kist.re.kr (E.EunKyeong Kim), gkeum@kist.re.kr
(G. Keum).
1
These authors contributed equally.
http://dx.doi.org/10.1016/j.bmcl.2016.11.062
0960-894X/Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Lee J.-H., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.11.062

2
J.-H. Lee et al. / Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
NH
OH
N
O
O
HN
HO
N
O
N
N
N
H2N
O
HO
N
N
HN
N
OH
O
AT13387
XL888
N
STA-9090
CF3
O
HN
N
O
N
N
N
N
N
N
O
O
N
H2N
N
NH2
N
H2N
O
N
O
N
S
N
H
NH2
I
O
TAS-116
SNX-5422
PU-H71
Fig. 1. Current Hsp90 inhibitors in active clinical trials.
solvent-exposed exit pocket C.21 The 2-amino and 9-pyridyl groups
of purine BIIB021 (PDB code: 3QDD) bind to the essential pocket A
and the hydrophobic pocket B, respectively.22 The purine PU-H71
(PDB code: 2FWZ) has an additional hydrophilic substituent at
N9 for binding to exit pocket C.23 Unlike most Hsp90 inhibitor ser-
steric hindrance. The desired 7-deazapurine compounds 6a–k
were prepared via N-alkylation with various alkyl halides 5 or
2-chloromethyl-4-methoxy-3,5-dimethyl-pyridine, in moderate
to good yields.
The binding affinities of 5-benzyl-2-amino-7H-pyrrolo[2,3-d]
ies, both purine compounds have a conserved
a-helix 4 outside of
pyrimidines derivatives 6a–k to Hsp90a were tested using a
pocket B. EC144, a 7-deazapurine compound with the N7 of purine
replaced with a carbon atom, also binds to pocket C with a C5 sub-
stituent (PDB code: 3NMQ), and the outer helix is conserved in the
cocrystal structure.24
fluorescence polarization (FP) competitive binding assay with
fluorescently labeled geldanamycin (FITC-GM) at Reaction Biology
Corporation (RBC, Malvern, PA, USA).26 BIIB021, PU-H71, and
17-AAG were used as positive controls (Table 1).
We determined that the hydrophilic substituent at C5 of 7H-2-
amino-pyrrolo[2,3-d]pyrimidine, 7-deazapurine, allowed binding
at the hydrophobic channel of exit pocket C (Fig. 2). This hydropho-
bic binding in the purine series at exit pocket C enables additional
interaction by maintaining the hydrogen bonding in pocket A and
O
HN
O
R1
H2N
N
ii)
NH2
O
2
i)
HN
H2N
H
R1
p–p stacking in pocket B with a conserved outer a-helix 4, based
N
H
N
on comparison with the cocrystal structures of other purine inhibi-
tors. In this report, we describe the synthesis of 7H-2-amino-pyr-
rolo[2,3-d]pyrimidine derivatives having a benzyl moiety at C5 as
novel potent Hsp90 inhibitors and discuss their antiproliferative
activities.
3a-g
1
OMe
N
R1
Cl
Cl
N
Cl
R1
N
N
iii)
iv)
OMe
All the title compounds reported herein were prepared as
N
H
N
H2N
from
H2N
N
illustrated in Scheme 1.25 First,
a-bromination of 3-arylpropyl
4a-g
6a-g
N
aldehydes with bromine in the presence of TMSBr, and subsequent
cyclization with 2,6-diaminopyrimidin-4(3H)-one 2, afforded
3-benzyl-substituted pyrrolo[2,3-d]pyrimidin-4-ones 3a–g in
moderate yields. The para-position in the phenyl ring of
3-(30-methoxyphenyl)propanaldehyde was also brominated, and
the resulting dibrominated product was used in the next step to
give 3e. Chlorination of 3a–g using POCl3 in the presence of
N,N-dimethylaniline as the base, with prior N-acetyl protection
and final deprotection with methanolic hydrochloric acid, provided
the 4-chloro-2-amino-pyrrolo[2,3-d]pyrimidines 4a–g in good
yields via three steps. The chlorination of benzyl pyrimidin-4-one
compounds substituted with methyl and methoxy groups at the
ortho-position of the benzyl group was unsuccessful, likely due to
R2 Cl
R1 = H ( ), 3-CH3
4a
a
b c d
( ), 4-CH3 ( ), 4-CH3O ( ),
4-Br-3-CH 3O ( ), 3-CF3 ( ), 4-Cl ( )
5
iv)
e
f
g
Cl
N
X
O
O
N
R2
=
N
H2N
R2
6h-k
6h
6i 6k
X = H ( ), Br ( ), I ( )
6j
Scheme 1. Reagents and reaction conditions: (i) TMSBr, Br2 (1 equiv), DCM/1,4-
dioxane (10/1), rt; (ii) 2, NaOAc (2 equiv), CH3CN/H2O (1/1), rt, 27–75%; (iii) Ac2O,
reflux, 3 h; POCl3, PhNMe2, BnEt3N+Clꢀ, CH3CN, 100 °C, 1 h; 3 N HCl in MeOH, 50 °C,
2 h, 21–92% in three steps; (iv) 5 or 2-chloromethyl-4-methoxy-3,5-dimethyl-
pyridine, K2CO3, DMF, 40 °C, 31–63%.
Pocket C
Y
BIIB021: X-Y = N
EC144: X- Y =
Cl
X
N
O
N
OH
H2N
N
Pocket A
N
C
Pocket B
Fig. 2. Structural superimposition of complex structures in the ATP-binding pocket of mHsp90ND shows that the outer
a-helix 4 over pocket B is conserved at PU-H71 (PDB
code: 2FWZ, purple) and collapsed at 17-AAG (PDB code: 1A4H, blue). Schematic representation of purine analogs and their interactions in the binding pockets of Hsp90
(right). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
a
Please cite this article in press as: Lee J.-H., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.11.062

J.-H. Lee et al. / Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
3
Table 1
Competitive binding affinity to Hsp90
normal cell line.
a using an FP assay, and in vitro inhibitory data of compounds 6a–k against MCF-7, HCT116, and SK-BR-3 cancer cell lines and the HFF-1
Compound
FP, HSP90
% Inh. (10
a
% Inhibition at 10
MCF-7
l
Mc
l
M)a
IC50
,
l
Mb
SKBR3
HCT116
HFF-1
6a
6b
6c
6d
6e
6f
6g
6h
6i
105.1
92.1
92.4
103.7
96.4
52.4
97.0
7.7
0.036 0.0039
0.051 0.0029
0.058 0.011
0.094 0.0023
0.079 0.0060
8.70 2.71
0.048 0.013
ndd
85.61
86.53
83.20
80.06
62.01
79.09
87.97
26.4
69.9
88.52
94.18
81.04
80.89
78.70
89.77
90.09
30.57
48.75
70.41
74.49
46.74
83.02
5.74
10.05
13.20
73.11
62.83
6.30
87.15
71.33
69.43
80.24
21.58
72.59
80.54
4.32
33.2
74.8
53.8
ndd
54.03
69.94
74.04
11.15
20.68
36.23
6j
6k
0.65 0.04
0.75 0.20
40.92
63.80
BIIB021
PU-H71
17-AAG
96.1
106.2
107.0
0.029 0.009
0.051 0.014
0.056 0.012
84.81
70.55
79.74
86.20
65.34
88.89
71.94
78.43
74.44
51.28
ndd
65.49
a
b
c
Percent inhibition values at 10
The compounds were tested in a 10-dose IC50 with threefold serial dilution starting at 100
% inhibitions were obtained after incubation for 72 h, and the values are the average of at least two independent measurements, and SD are less than 20%.
Not determined.
l
M are the average of two independent experiments, and standard deviation (SD) are less than 20%.
l
M, and IC50 values are the means of two experiments SEM.
d
Table 2
Percent growth inhibition at 10 l
M and GI50 values of compound 6a and PU-H71 over the NCI-60 cell line panel.a,b
Cancer type
Leukemia
Cell line
6a
PU-H71
GI50 M)
Cancer type
Cell line
6a
PU-H71
GI50 M)
Growth%
GI50
(lM)
(l
Growth%
GI50
(lM)
(l
CCRF-CEM
HL-60(TB)
K-562
MOLT-4
RPMI-8226
SR
17.83
9.64
0.46
0.39
0.33
0.42
1.00
0.43
0.51
3.39
0.58
1.10
1.19
0.50
0.96
0.36
0.50
0.32
0.39
0.40
0.38
0.30
0.29
0.41
0.79
0.31
0.76
0.89
5.00
0.85
0.32
0.41
0.427
0.063
0.071
0.091
1.514
0.060
0.087
2.512
0.263
0.288
0.759
0.089
0.170
0.058
0.059
0.048
0.282
0.063
0.437
0.038
0.059
0.059
2.089
0.145
0.851
0.646
0.724
0.129
0.079
0.123
Melanoma
M14
28.23
16.45
39.51
26.77
32.93
22.23
1.02
0.65
0.43
2.48
0.47
0.69
0.90
0.39
0.50
0.56
6.55
1.48
0.55
0.45
0.32
1.56
3.04
0.4
0.44
0.76
0.67
0.48
0.49
0.61
0.88
0.23
0.58
6.39
0.53
0.44
–
0.234
0.091
0.355
0.417
0.457
0.309
0.372
0.089
0.589
33.113
3.020
0.155
3.631
0.045
0.513
0.437
0.295
0.525
0.759
0.174
0.162
0.324
0.132
0.048
0.046
0.479
1.288
1.514
0.155
0.083
MDA-MB-435
SK-MEL-2
SK-MEL-28
SK-MEL-5
UACC-257
UACC-62
IGROV1
OVCAR-3
OVCAR-4
OVCAR-5
OVCAR-8
ADR-RES
SK-OV-3
786-0
A498
ACHN
CAKI-1
RXF 393
SN12C
TK-10
UO-31
PC-3
DU-145
MCF7
13.29
11.56
33.85
14.26
20.74
48.04
33.07
4.32
50.54
34.71
64.51
14.5
10.58
ꢀ7.65
11.02
11.96
14.69
5.27
Non-small cell lung cancer
A549
EKVX
Ovarian cancer
Renal cancer
45.0
HOP-62
HOP-92
NCI-H226
NCI-H23
NCI-H322M
NCI-H460
NCI-H522
COLO 205
HCC-2998
HCT-116
HCT-15
HT29
KM12
SW-620
SF-268
SF-295
SF-539
35.66
53.58
63.01
31.91
31.49
13.56
44.37
53.47
25.95
18.41
64.83
30.50
26.18
25.43
31.28
53.77
16.5
Colon cancer
CNS cancer
9.30
18.53
49.78
ꢀ1.68
37.15
39.36
39.44
33.93
0.99
Prostate cancer
Breast cancer
SNB-19
SNB-75
U251
LOX IMVI
MALME-3M
MDA-MB-231
HS 578T
BT-549
T-47D
MDA-MB-468
13.55
48.94
15.0
23.85
6.48
Melanoma
18.57
a
Data were obtained from the National Cancer Institute (NCI) in vitro disease-oriented human tumor cell line screen (five-dose–response curve) for compound 6a (NSC
782636) and PU-H71 (NSC 750424).
b
Bold figures indicate % growth lower than 10% at 10 lM for 6a.
Among 5-benzyl-substituted compounds 6a–k, compounds
6a–g (except 6f) substituted with (4-methoxy-3,5-dimethylpyri-
Among the tested 5-benzyl-substituted 2-amino-7H-pyrrolo
[2,3-d]pyrimidines, compounds 6a–g substituted with
din-2-yl)methyl group at position
inhibited Hsp90 by more than 90% at 10
the potency of compounds showing more than 50% inhibition at
10 M, their IC50 values against Hsp90 were determined with
threefold serial dilution starting at 100 M.
7
(e.g., BIIB021) potently
(4-methoxy-3,5-dimethylpyridin-2-yl) methyl group (e.g., BIIB021)
at position 7 exhibited excellent potency, similar to the reference
compounds BIIB021, PU-H71, and 17-AAG (29, 51, and 56 nM,
respectively). The non-substituted benzyl compound 6a showed
the highest potency with an IC50 value of 36 nM. Among the
a
lM. To further examine
l
a
l
Please cite this article in press as: Lee J.-H., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.11.062

4
J.-H. Lee et al. / Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
substituted benzyl compounds, compounds 6b and 6g, possessing
para-chloro and meta-methyl substituents, respectively, also
showed strong potency (IC50 values of 48 and 51 nM, respectively).
Compound 6h, with the benzyl substituent at N7, showed no affin-
ity. Benzodioxolylmethyl derivatives 6j and 6k, having additional
ortho-bromo or iodo substituents at the benzodioxole ring (e.g.,
C8-substituent of PU-H71), showed decreased inhibitory activity.
The antiproliferative activities of all compounds were initially
cell lung cancer), COLO-205, HT29 and KM12 (colon cancer), SF-
295 (CNS cancer), LOX IMVI and UACC-62 (melanoma), and
MDA-MB-468 (breast cancer).
Compound 6a showed impressive antiproliferative activity in
the single-dose test and satisfied the threshold inhibition criteria
of the NCI. Therefore, it was further evaluated in five-dose testing
mode to determine the GI50 value (the molar concentration causing
half-maximal growth inhibition). The GI50 values of 6a are shown
in Table 2, and compared to those of PU-H71 (NSC 750424). Com-
pound 6a showed high potency with submicromolar GI50 values
against 49 of the 59 cell lines tested. The average GI50 value of
assessed at 10 lM against three human cancer cell lines, MCF-7
and SK-BR-3 (breast cancer cells) and HCT-116 (colorectal
carcinoma), by MTT assay. The percentage of growth inhibition
(GI) was recorded, and BIIB021, PU-H71, and 17-AAG were used
as reference compounds (Table 1).
compound 6a was lower than 1
lM (0.98 1.32
lM) and was
twice than that of PU-H71 (0.49 0.73
l
M), omitting the ovarian
All compounds (except 6h and 6i) showed potent antiprolifera-
cancer OVCAR-4 cell line.
tive activity at 10
compounds. Furthermore, the cytotoxicity of the compounds at
10 was compared against the human foreskin fibroblast
(HFF-1) normal cell line using an MTT assay. The compounds
(except 6c, 6f and 6g) showed low GI activities against HFF-1 nor-
mal cells with sufficient therapeutic selectivity and were compara-
ble to, or better than, those of BIIB021 and 17-AAG.
l
M, comparable to the potency of the reference
We determined the crystal structure of 6a bound to the
N-terminal domain of mouse Hsp90 (mHsp90ND) at 1.5 Å resolu-
tion (PDB code: 5H22).28 The complex structure confirmed that
6a binds to the ATP-binding site of mHsp90ND with all atoms well
defined in electron density maps (Fig. 3). The inhibitor binding
lM
pockets can be categorized into three pockets: pocket
A is
composed of residues 41–65; pocket B is composed of residues
128–138 and 96–100; and pocket C is composed of residues
Encouraged by the Hsp90 inhibitory and anticancer activities,
the antiproliferative activity of compound 6a was assessed at
101–113 and Lys58 (Fig. 3b). The outer
a-helix 4 over pocket B,
10
l
M against a panel of 60 human cancer cell lines at the National
including the binding pocket C regions, was flexible and thus
caused structural rearrangement to an extended helical structure.
The binding pockets A and B were somewhat rigid, whereas bind-
ing pocket C on the surface was looser. The 2-amino group in the
pyrrolo[2,3-d]pyrimidine ring of 6a binds to binding pocket A,
forming hydrogen bonds to Asp93. The pyridine ring at N7 binds
to binding pocket B, which is lined by hydrophobic residues,
Cancer Institute (NCI; Bethesda, MD, USA).27 The results are listed
in Table 2. Compound 6a showed a significant broad-spectrum
antiproliferative activity, and its mean growth inhibition for the
full panel of 60 cell lines were 73.3% 17.4. Importantly, com-
pound 6a displayed strong antiproliferative activity (>90% inhibi-
tion) over nine cell lines: HL-60 (leukemia), HOP-92 (non-small
Fig. 3. X-ray structure of 6a bound to the ATP-binding site of mHsp90ND and comparison of the binding mode with those of PU-H71 and EC144. (a) 6a binds to the ATP-
binding site of Hsp90. The 2FoꢀFc electron-density map contoured at the 1.5
r level shows unambiguous positioning of all the atoms in 6a. Hsp90 is shown in the ribbon
presentation with the side chains of the key interacting residues shown in the stick model. Binding mode of (b) 6a (PDB code: 5H22),27 (c) PU-H71 (PDB code: 2FWZ)23 and (d)
EC144 (PDB code: 3NMQ)24 with the three binding pockets highlighted. The binding pockets A and B are shown in blue and orange, respectively, while the binding pocket C is
shown in the same color as the inhibitor.
Please cite this article in press as: Lee J.-H., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.11.062

J.-H. Lee et al. / Bioorganic & Medicinal Chemistry Letters xxx (2016) xxx–xxx
5
pyrimidines. The inhibitors were identified by an FP assay. The
most potent Hsp90 inhibitor, 6a, showed potent and broad spec-
trum antiproliferative activity against clinically important NCI-60
cancer cell lines with submicromolar GI50 values. The interaction
between 6a and Hsp90 was confirmed by X-ray crystallography.
Western blot analysis of MCF7 cells showed that treatment of 6a
degraded several client proteins of Hsp90 and induced expression
of Hsp70. New Hsp90 inhibitors with appropriate hydrophobic
interactions in the cavity of pocket C that conserve the outer
a-
helix 4 of pockets B and C are promising compounds for the devel-
opment of potent broad-spectrum anticancer agents.
Acknowledgments
This research was supported by the Korea Institute of Science
and Technology (KIST) Institutional Program (2E26650 and
2E26663). We express our sincere appreciation and gratitude for
the National Cancer Institute (NCI, Bethesda, MD, USA) for per-
forming the anticancer evaluation of the new compounds. This
work was also supported by a grant from the Global Research
Laboratory Program of the Ministry of Science, ICT, Future Planning
of Korea (MISP: grant No. NRF 20110021713), and the R&D
Convergence Program of National Research Council of Science &
Technology of the Republic of Korea.
Fig. 4. Effects of compound 6a on the expression of EGFR, Her2, Akt, Cdk4, Hsp70,
and Hsp90 in the MCF-7 cell line. b-Actin was used in a loading control. Compound
6a induced proteasomal degradation of Hsp90 client proteins (EGFR, Her2, Akt and
Cdk4) and upregulated Hsp70 at various concentrations. Geldanamycin (GA, 1
and DMSO (D) were used as positive and negative controls
lM)
namely Leu103, Phe138, Tyr139, Val150, and Trp162. The benzyl
group at C5 is packed against the hydrophobic residues Leu107
and Ile110 of binding pocket C, where BIIB021 has no substituent
for the appropriate binding (PDB code: 3QDD).22 The overall
binding mode of 6a is similar to those seen in other purine-based
inhibitors possessing binding pockets A and B. However, there
are variations in the moiety occupying binding pocket C, e.g., N9
of PU-H71 does not form significant contacts (Fig. 3c), while C5
of EC144 binds to Lys58 (Fig. 3d). Therefore, the conformations of
the residues in binding pocket C were flexible, accommodating
hydrophobic moieties such as the substituted benzyl group. These
structural data indicate that the 7-benzyl group can exploit the
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2016.11.
062.
References
1. Maloney A, Workman P. Expert Opin Biol Ther. 2002;2:3.
2. Whitesell L, Lindquist SL. Nat Rev Cancer. 2005;5:761.
3. Garcia-Carbonero R, Carnero A, Paz-Ares L. Lancet Oncol. 2013;14:e358.
4. Macario AJL, de Macario ECN. Engl J Med. 2005;353:1489.
5. Powers MV, Workman P. Endocr Relat Cancer. 2006;13:S125.
6. Mahalingam D, Swords R, Carew JS, Nawrocki ST, Bhalla K, Giles FJ. Br J Cancer.
2009;100:1523.
extended hydrophobic pocket C, formed by
a4 in the structure of
6a, to develop more potent Hsp90 inhibitors.
Western blot analysis of MCF7 cells treated with 6a demon-
strated that the compound induced degradation of Hsp90 client
proteins and expression of Hsp70. To determine the effect of 6a
on the down-regulation of Hsp90 client proteins, MCF-7 cells were
7. Neckers L, Workman P. Clin Cancer Res. 2012;18:64.
8. Bhat R, Tummalapalli SR, Rotella DP. J Med Chem. 2014;57:8718.
9. Jhaveri K, Taldone T, Modi S, Chiosis G. Biochim Biophys Acta. 2012;1823:742.
10. Kim T, Keum G, Pae AN. Expert Opin Ther Patent. 2013;23:919.
11. Eccles SA, Massey A, Raynaud FI, et al. Cancer Res. 2008;68:2850.
12. Floris G, Debiec-Rychter M, Wozniak A, et al. Mol Cancer Ther. 2011;10:1897.
13. Lundgren K, Zhang H, Brekken J, et al. Mol Cancer Ther. 2009;8:921.
14. Smyth T, Van Looy T, Curry JE, et al. Mol Cancer Ther. 2012;11:1799.
15. Ying W, Du Z, Sun L, et al. Mol Cancer Ther. 2012;11:475.
16. Scaltriti M, Dawood S, Cortes J. Clin Cancer Res. 2012;18:4508.
17. Wang M, Shen A, Zhang C, et al. J Med Chem. 2016;59:5563.
18. Yin Z, Zhang H, Lundgren K, Wilson L, Burrows F, Shores CG. Int J Cancer.
2010;126:1216.
incubated with compound 6a (0, 1, 5, 10, 15 and 30 lM) for 24 h.
The expression of Hsp90 client proteins (EGFR, Her2, Akt and
Cdk4, along with Hsp70 and Hsp90) were measured by Western
blot analysis. As expected, compound 6a reduced the expression
levels of Hsp90 client proteins, including EGFR, Her2, Akd, and
Cdk4, in
a concentration-dependent manner (Fig. 4). The
expression levels of EGFR and Her2 decreased in a dose-dependent
manner upon treatment of the cells with 6a, while Akt and Cdk4
were significantly degraded. In addition, treatment with 6a
increased the protein level of Hsp70 in a dose-dependent manner.
Overall, the results strongly suggest that compound 6a disrupted
the Hsp90 chaperone mechanisms in the MCF-7 cell line.
19. Ohkubo S, Kodama Y, Muraoka H, et al. Mol Cancer Ther. 2015;14:14.
20. Saif MW, Takimoto C, Mita M, et al. Clin Cancer Res. 2014;20:445.
21. Taldone T, Patel PD, Patel M, et al. J Med Chem. 2013;56:6803.
22. Shi J, Van de Water R, Hong K, et al. J Med Chem. 2012;55:7786.
23. Immormino RM, Kang Y, Chiosis G, Gewirth DT. J Med Chem. 2006;49:4953.
24. Yun TJ, Harning EK, Giza K, et al. J Immunol. 2011;186:563.
25. Keum G, Kim EE, Seo SH et al., Korea Patent 2016-0059366, 2016.
26. Reaction Biology Corp.: www.reactionbiology.com.
In summary, a new series of pyrrolo[2.3-d]pyrimidines was
synthesized. Potent Hsp90 inhibitors were designed based on the
X-ray complex structure of Hsp90, with a modified benzyl side
chain at C5 and an N7 substituent of 2-amino-7H-pyrrolo[2,3-d]
27. DTP Data Search: https://dtp.cancer.gov/dtpstandard/dwindex/index.jsp (NSC
782636).
28. The structure of 6a in complex with mHsp90ND is deposited in the RCSB
Protein Data Bank (PDB) under PDB code 5H22.
Please cite this article in press as: Lee J.-H., et al. Bioorg. Med. Chem. Lett. (2016), http://dx.doi.org/10.1016/j.bmcl.2016.11.062