Discovery of Novel Celastrol Derivatives as Hsp90-Cdc37 Interaction Disruptors with Antitumor Activity

Article abstract of DOI:10.1021/acs.jmedchem.9b01290

To develop novel and efficient heat shock protein 90-cell division cycle 37 (Hsp90-Cdc37) interaction disruptors, several lipophilic fragments were introduced into celastrol (CEL) to synthesize 48 new CEL derivatives. Among all the target compounds, 41 wa

Full text of DOI:10.1021/acs.jmedchem.9b01290

Subscriber access provided by - Access paid by the | UCSF Library
Article
Discovery of novel celastrol derivatives as Hsp90-
Cdc37 interaction disruptors with antitumour activity
Na Li, Manyi Xu, Bing Wang, Zhixian Shi, Zihao Zhao,
Yunqing Tang, Xinyue Wang, Jianbo Sun, and Li Chen
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01290 • Publication Date (Web): 14 Nov 2019
Downloaded from pubs.acs.org on November 14, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of novel celastrol derivatives as Hsp90-Cdc37
interaction disruptors with antitumour activity
δ
δ
Na Li , Manyi Xu , Bing Wang, Zhixian Shi, Zihao Zhao, Yunqing Tang, Xinyue
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Wang, Jianbo Sun*, and Li Chen*
State Key Laboratory of Natural Medicines, Department of Natural Medicinal Chemistry, School
of Traditional Chinese Pharmacy, China Pharmaceutical University, 24 Tong Jia Xiang, Nanjing
2
10009, People's Republic of China
ABSTRACT
To develop novel and efficient heat shock protein 90-cell division cycle 37
(Hsp90-Cdc37) interaction disruptors, several lipophilic fragments were introduced
into celastrol (CEL) to synthesize 48 new CEL derivatives. Among all the target
compounds, 41 was screened with superior anti-proliferative activity on related cancer
cells (IC : 0.41 ~ 0.94 µM), and 41 could decrease the level of the Hsp90-Cdc37
5
0
complex in A549 cells. The capability to disrupt the Hsp90-Cdc37 interaction was
stronger than that of CEL. Furthermore, pull-down assay, UV assay and molecular
docking analysis all showed that 41 might disrupt the interaction of the Hsp90-Cdc37
complex by preferentially binding to Cdc37 in cancer cells. Further studies showed
that 41 could significantly regulate the levels of Hsp90-Cdc37 clients, thereby
inducing the apoptosis of cancer cells. Together, 41 is a novel Hsp90-Cdc37 disruptor
by binding to Cdc37 (hydrogen bond and/or covalent bond). Our results may provide
reference for the discovery of effective Hsp90-Cdc37 disruptors.
INTRODUCTION
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Heat shock protein 90 (Hsp90, 90-kDa) is a protein chaperone that is considered as
1
a promising anticancer target. Generally, Hsp90 needs to form a complex with
cochaperones to ensure the proper folding and maturation of the client proteins. These
cochaperones include cell division cycle 37 (Cdc37), heat shock protein 70 (Hsp70),
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
heat shock protein 40 (Hsp40), homeodomain only protein (Hop), p23, etc. Among
them, Cdc37 targets many protein kinases related to the occurrence and development
of cancer, such as protein kinase B (Akt), cyclin-dependent kinases 4 (Cdk4),
epidermal growth factor receptor (EGFR), human epidermal growth factor receptor-2
(Her-2), and tyrosine-protein kinase Met (c-Met), etc.3,4 Cdc37 can stabilize the
catalytic domains of these kinases, while Hsp90 can maintain their correct
5
conformation. Thus, the functions of these oncogenic kinases depend on the
Hsp90-Cdc37 complex. Disruption of on the tHsp90-Cdc37 interaction could regulate
the expression of kinases to inhibit the proliferation of cancer cells.6
Celastrol (CEL) has been reported to inhibit the proliferation of various tumours in
vitro and in vivo, such as gliomas, hepatocellular carcinomas and prostate cancer.7-9
The anti-proliferative activity of CEL has been indicated to result from disruption of
the interaction between Hsp90 and Cdc37, and this interaction is mainly located at the
N-terminal domain of Hsp90 (Hsp90 ) and the middle domain of Cdc37
N
(Cdc37M).10-12 In the structure of the human Hsp90 -Cdc37 complex, an important
N
M
hydrogen bond between Q133 (Hsp90 ) and R166/R167 (Cdc37 ) was identified as
N
M
1
2
the key interaction for the Hsp90 -Cdc37 . CEL was reported to disrupt this key
N
M
hydrogen bond due to its location in the hydrophobic pocket.10,11 However, as a
ACS Paragon Plus Environment

Page 3 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
pentacyclic triterpenoid, the spatial structure of CEL is too large to embed the
1
1
intersection angle of R166 and R167 in the structure of human Cdc37. In order to
disrupt the key hydrogen bond more efficiently, Xu et al recently reported a series of
CEL hybrids incorporated with methyl ferulate and its derivatives (hydrophobic
structures) and the antiproliferative activity of these hybrids was higher than that of
CEL. The immunoprecipitation experiment showed that compound 29 (Xu’s study)
disrupted the Hsp90-Cdc37 interaction more efficiently than CEL.13 Thus,
introduction of a hydrophobic structure to CEL may be a promising strategy in the
discovery of CEL derivatives as Hsp90-Cdc37 interaction disruptors.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To elucidate the structure-activity relationship of CEL derivatives as Hsp90-Cdc37
disruptors and to improve the druglike properties, Jiang’s study reported a series of
1
4
CEL derivatives and found a more potent compound CEL20. Compared to CEL,
although the solubility of CEL20 incorporated a polar group was greatly improved
(51.2 vs 1.44 μg/ml), the Hsp90-Cdc37 disruption activity [IC (4.71 vs 6.49 μM)]
5
0
was only slightly improved. And the antiproliferative activity of CEL20 was similar
to those of the most compounds incorporated hydrophobic groups [IC (1.12 vs
5
0
1
.20-1.97 μM]. Moreover, Jiang still emphasized the importance of the hydrophobic
interaction and a covalent bond between CEL and Cdc37 in the Hsp90-Cdc37
disruption activity in their design idea. Based on the previous studies, we designed a
series of CEL derivatives in which 29-carboxyl group was modified with lipophilic
fragments via different linkers and hope that these new compounds disrupted the
interaction of the Hsp90-Cdc37 complex by the non-covalent and/or covalent
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
combination with the target protein.
The structure of cinnamic acid (CA), which consists of a phenyl ring substituted
with an acrylic acid, contains highly conjugated π-electron with high hydrophobicity.
This structural characteristic allows it to easily bind to hydrophobic pockets.15,16
Recent studies have reported that CA exhibits potential anti-tumor activity in several
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
7
cancer cells. In terms of the modification of natural products, CA is widely used to
enhance the anti-tumor activity of the parent compounds.18-20 Thus, CA and its
analogues were first introduced to CEL to improve the capability of disrupting the
Hsp90-Cdc37 complex and inhibiting cancer cells.
At present, the structural modifications of CEL mainly focus on the C-20
carboxylic acid functionality and alterations of the A, B rings.21,22 However, many
derivatives modified at the A, B rings were unstable.22-23 These derivatives were
oxidized to CEL quickly upon purification, storage and incubation with culture
2
1
medium. Therefore, considering the stability of the target compounds, CA and its
analogues were introduced to the 20-carboxylic acid of CEL by different linkers to
afford in a total of 48 new compounds (1-48). Their anti-proliferative activities were
screened in vitro. The activity of 41 (IC : 0.41 ~ 0.94 µM) was comparable with that
5
0
of bardoxolone methyl (CDDO-Me). Furthermore, the capability of 41 to disrupt the
Hsp90-Cdc37 interaction was stronger than that of CEL. In addition, in-depth
mechanistic studies of 41 were evaluated to explain the anti-proliferative activity.
RESULTS AND DISCUSSION
Chemistry. The synthetic route of compounds 1-48 was shown in Scheme 1.
ACS Paragon Plus Environment

Page 5 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Trans-cinnamic acid (1a) or 4-methylcinnamic acid (1b) or 4-methoxycinnamic acid
(1c) or 4-Fluorocinnamic acid (1d) or 4-(trifluoromethyl) cinnamic acid (1e) was first
treated
by
the
indicated
bromohydrins
hydrochloride (EDCI)
using
and
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1-(3-dimethylaminopropyl)-3-ethylcarbodiimide
dimethylaminopyridine (DMAP) as catalysts in dichloromethane (CH Cl ) at room
2
2
2
4
temperature to get compounds 2f-t. In a similar way, compounds 6f-t and 10f-t were
obtained under the treatment of 5a-e and 9a-e, respectively. Besides, intermediates
1
0u-w were prepared by trans-1,4-dibromo-2-butene in acetone, catalysed by
+
-
potassium carbonate (K CO ) and tetrabutylammonium bromide (Bu N Br ) at 50
2
3
4
oC.25 α-cyano substituted cinnamic acids (α-CN CA) 5a-e were furnished by the
condensation of aryl formaldehydes 3a-e and cyanoacetic acid 4a, while β-cyano
substituted cinnamic acids (β-CN CA) 9a-e were synthesized by arylacetonitriles 7a-e
24,25
and glyoxylic acid 8a with K CO in methanol (MeOH).
Subsequently, the target
2
3
compounds 1-48 were accomplished through connecting 2f-t or 6f-t or 10f-w to CEL
at C-20 position in the presence of sodium bicarbonate (NaHCO ) in
3
1
3
1
3
N,N-dimethylformamide (DMF). The substituent groups R -R of 1-48 were shown
in Table 1.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
R1
H
O
O
1
/3/5/7/9a
X
a
O
Br
1/3/5/7/9b CH3
/3/5/7/9c OCH3
OH
1
R1
R1
2f-t
1a-e
1/3/5/7/9d
F
1/3/5/7/9e
CF3
O
O
O
H
O
X
b
a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
OH
O
Br
+
NC
OH
OH
CN
CN
6f-t
CN
R1
R1
R1
3a-e
4a
5a-e
O
O
O
a/d
CN
c
X
+
OH
O
Br
OHC
R1
CN
9a-e
R1
R1
7a-e
8a
10f-w
R1
H
X
R1
CH3
OCH3
R1
F
X
X
R1
X
2
/6/10f
/6/10g
/6/10h
/6/10i
(CH )
2/6/10k
2/6/10l
(CH2)4
(CH2)2
2/6/10p
2/6/10q
2/6/10r
2/6/10s
2/6/10t
2 3
(CH )
2 2
10u
H
CH CH=CHCH
2
2
2
2
2
H
(CH2)
F
2 4
(CH )
3
4
10v OCH3 CH CH=CHCH
2
2
H
(CH2)
2/6/10m OCH3 (CH2)3
CF3
2 2
(CH )
10w CF3
CH CH=CHCH
2
2
CH3
CH3
(CH )
2 2
2/6/10n
2/6/10o
OCH3
F
(CH2)4
(CH2)2
CF3 (CH2)3
CF3 (CH2)4
2/6/10j
2 3
(CH )
O
O
O
R2

X
OH
O
O

R3
R2
O
R1
X
e
H
H
O
Br
O
+
O
R3
R1
HO
HO
1-48
2
f-t, 6f-t, 10f-w
Scheme 1. Reagents and Conditions: (a) various bromohydrin, EDCI, DMAP, CH Cl ,
2
2
o
o
rt, 7 h; (b) NH OAc, tolune, 100 C, reflux, 12 h; (c) CH OH, K CO , 60 C, reflux, 6
4
3
2
3
h; (d) bromohydrin, EDCI, DMAP, CH Cl , rt, 7 h or trans-1,4-dibromo-2-butene,
2
2
+
-
o
o
K CO , Bu N Br , acetone, 50 C, reflux,7 h; (e) celastrol, NaHCO , DMF, 60 C, 7 h.
2
3
4
3
1
3
Table 1 The substituent groups (R -R ) of 1-48.
Cpd.
X
R1
H
R2
H
H
H
H
R3
H
H
H
H
Cpd.
25
X
R1
F
R2
H
H
H
H
R3
1
2
3
4
(CH2)2
(CH )
CN
CN
CN
CN
2 2
(CH )
H
26
(CH )
F
2 3
2 3
(CH2)4
(CH2)2
H
27
(CH )
F
2 4
CH3
28
(CH )
CF3
2 2
ACS Paragon Plus Environment

Page 7 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
6
(CH2)3
(CH2)4
(CH2)2
CH3
CH3
OCH3
OCH3
OCH3
F
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
(CH )
CF3
CF3
H
H
CN
CN
H
2 3
(CH )
H
2 4
7
H
(CH )
CN
CN
CN
CN
CN
CN
CN
2 2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
(CH )
H
(CH )
H
H
2 3
2 3
9
(CH2)4
(CH2)2
(CH2)3
(CH2)4
H
(CH )
H
H
2 4
10
H
(CH )
CH3
CH3
CH3
H
2 2
1
1
1
1
1
2
3
4
F
H
(CH )
H
2 3
F
H
(CH )
H
2 4
(CH )
CF3
H
(CH )
OCH3
H
2 2
2 2
(CH2)3
(CH2)4
(CH2)2
(CH2)3
CF3
H
(CH )
OCH3 CN
OCH3 CN
H
2 3
15
CF3
H
H
(CH )
H
2 4
1
1
1
1
6
7
8
9
CN
CN
CN
CN
CN
CN
CN
(CH )
F
F
CN
CN
CN
CN
CN
CN
CN
H
2 2
H
(CH )
H
2 3
(CH )
H
(CH )
F
H
2 4
2 4
(CH2)2
(CH2)3
(CH2)4
(CH2)2
CH3
CH3
CH3
OCH3
(CH )
CF3
CF3
CF3
H
H
2 2
20
(CH )
H
2 3
2
2
1
2
(CH )
H
2 4
CH CH=CHCH
H
2
2
2
2
3
4
(CH2)3
(CH2)4
OCH3
OCH3
H
H
CN
CN
47
48
CH CH=CHCH OCH3 CN
H
H
2
2
CH CH=CHCH
F
CN
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
Biological Activity.
In Vitro Anti-proliferative Activity of CEL Derivatives
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We first synthesized compounds 1-45 with saturated linear alkyl chains as linkers
in this study. The inhibition rates of 1-45 were determined in A549 cells at
concentration of 1.5 µM (Table 2). CEL was used as a positive control. As shown in
Figure 1A, the inhibition rates of the CA-CEL hybrids 1-15 (blue, 6.19%-37.61%)
and α-CN substituted CA-CEL hybrids 16-30 (red, 3.07%-34.68%) were lower than
that of CEL (purple, 47.98%). These results indicated that the introduction of CA and
α-CN-substituted CA decreased the anti-proliferative activities of CEL. Surprisingly,
when β-CN-substituted CA was introduced into CEL, the activities of several
derivatives increased (green, Figure 1A). For example, the anti-proliferative activities
of 31 (59.65%), 32 (53.69%), 37 (48.11%), 38 (53.38%), 40 (65.87%) and 41
(78.45%) were higher than that of CEL (47.98%). In the β-CN-substituted CA-CEL
hybrids (31-45, Figure 1B), the substituent on the benzene ring could affect the
activity following the order of F > H > OCH > CF > CH . These results indicated
3
3
3
that the introduction of an electron-withdrawing group (F, OCH and CF ) onto the
3
3
benzene ring might enhance anti-proliferative activity. However, as the presence of
electron withdrawing substituent increases, the activity would decrease (F> OCH and
3
CF ). Furthermore, we found that the length of the linker also significantly regulated
3
the activity (Figure 1D). For example, the inhibition rates of 40 with a 2-carbon linker
(
65.87%) and 41 with a 3-carbon linker (78.45%) were much higher than that of 42
ACS Paragon Plus Environment

Page 9 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
with a 4-carbon linker (32.15%). This phenomenon occurred with all target
compounds in this study (Figure 1D). These results suggested that once the length of
linker reached four carbons, the inhibitory effects were reduced sharply (black arrows)
(Figure 1D).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A
90
80
70
CA
C
CA-CEL (1-15)
A
6
5
0
0
β-CN CA
α-CN CA-CEL (16-30)
Superior activity
β-CN CA-CEL (30-45)
40
40
30
30
20
20
10
10
CEL
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0
0
1
1
3
2
3
4
5
6
7
8
9
10 11 12 13 14 15
6
1
17 18 19 20 21 22 23 24 25 26 27 28 29 30
32 33 34 35 36 37 38 39 40 41 42 43 44 45
Target compounds
C
H
B
O
O
CN

O
( )nO

R1
IR (%)
CF3
CH3
R1
H
O
HO
31-45 n = 2, 3, 4
1
R = H, CH , OCH , F, CF
F
OCH3
3
3
3
D
8
7
6
0
0
0
50
4
3
2
1
0
0
0
0
0
-CN CA-CEL
CN
O
(
)n
O
R1
n=2
n=3
n=4

CN CA-CEL
O
H
CH3
OCH3
F
CF3
(
)n
30
O
CN
R1
2
1
0
0
0
n=2
n=3
n=4
H
0
CH3
OCH3
F
CF3
4
CA-CEL
30
O
CN
(
)n
O
2
0
0
0
n=2
n=3
n=4
R1
1
R1
H
H
CHC3 H3 OCHO3 CH3
F
F
CF3
CF3
Figure 1. Comparison of inhibition rates of target compounds. (A) Inhibition rates of
ACS Paragon Plus Environment

Page 11 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1-45 and CEL at 1.5 μM in A549 cells. IR: inhibition rate. (B) The average inhibition
rates of 31-45 classified by the different groups in benzene. (C) Inhibition rate of
3
1-45 (D) Inhibition rate of 1-45 classified by lengths of linker.
.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Preliminary inhibitory effects of 1-45 and CEL in A549 cells
Cpd.
IRa
Cpd.
IR
Cpd.
IR
Cpd.
IR
1
3
25.97 ± 1.45
25
26.15 ± 1.82
1
2
3
4
5
6
7
8
9
22.97 ± 1.65
33.21 ± 2.33
13.27 ± 1.25
28.65 ± 1.99
29.06 ± 2.17
19.87 ± 1.52
35.32 ± 2.48
34.11 ± 2.76
20.65 ± 1.39
27.74 ± 1.88
37
38
48.11 ± 3.32
53.38 ± 3.11
26
24.99 ± 1.95
14
15
16
17
18
19
20
21
22
23
37.61 ± 1.77
6.19 ± 1.02
24.29 ± 1.34
19.09 ± 1.38
16.19 ± 1.43
29.87 ± 1.54
34.68 ± 2.75
16.49 ± 1.22
16.59 ± 1.32
18.64 ± 1.05
3.07 ± 0.65
3
9
9.57 ± 1.08
27
28
29
30
31
32
33
34
35
36
19.00 ± 1.67
16.03 ± 1.06
19.40 ± 1.54
7.85 ± 0.34
59.65 ± 2.87
53.69 ± 3.05
3.19 ± 0.12
40
41
42
43
44
45
65.87 ± 3.43
78.45 ± 3.59
32.15 ± 2.67
33.30 ± 1.98
24.51 ± 1.26
18.33 ± 1.38
1
1
0
1
20.40 ± 2.43 CELb 47.98 ± 3.29
21.59 ± 1.55
37.63 ± 2.01
14.71 ± 1.04
12
2
4
9.74 ± 1.03
a
IR: Inhibition rate (%) at 1.5 µM.
Positive control.
b
According what mentioned above, a F/H/OCH group substituted onto the benzene
3
ring of β-CN-substituted CA-CEL hybrids showed higher activity than that of the
CF /CH groups. To study the effects of the unsaturated alkyl group on the
3
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
anti-proliferative activity, we replaced the linkers of the F/H/OCH substituted β-CN
3
-
CA-CEL hybrids with unsaturated alkyl chains to obtain 46, 47 and 48, respectively.
Then, compounds 31, 32, 37, 38, 40 and 41, with stronger inhibitory activities than
CEL, along with 46-48 were evaluated to determine their IC values against four
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
tumour cell lines (A549, HOS, MCF-7 and HepG2) with the MTT method (Table 3).
As a triterpenoid derivative with multiple-target, CDDO-Me was used as a positive
control. As shown in Table 3, the anti-proliferative activities of 46 (IC : 2.86 ~ 3.54
5
0
µM), 47 (IC : 1.34 ~ 2.71 µM) and 48 (IC : 1.11 ~ 2.74 µM) had not improved
5
0
50
significantly, which suggested that the unsaturated alkyl linker showed little effect on
the activity in this study. Among all the compounds, 41, which showed the highest
activity (IC : 0.41 ~ 0.94 µM) against cancer cells, attracted our attention. Therefore,
5
0
4
1 was selected for further investigation.
Table 3. Anti-proliferative activity of the selected compounds against four cancer cell
lines.
Cpd.
IC50
A549
MCF-7
HOS
HepG2
3
3
1
2
1.02 ± 0.12
1.41 ± 0.17
1.79 ± 0.15
1.09 ± 0.13
1.02 ± 0.15
0.41 ± 0.11
0.70 ± 0.08
0.98 ± 0.05
0.97 ± 0.09
1.10 ± 0.08
2.35 ± 0.12
0.64 ± 0.07
0.54 ± 0.05
0.45 ± 0.07
1.14 ± 0.07
1.75 ± 0.05
2.18 ± 0.14
0.90 ± 0.05
0.65 ± 0.15
0.76 ± 0.13
0.97 ± 0.10
1.27 ± 0.09
3.89 ± 0.28
0.94 ± 0.23
37
3
4
4
8
0
1
ACS Paragon Plus Environment

Page 13 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
4
4
6
7
8
3.54 ± 0.11
1.34 ± 0.14
1.84 ± 0.17
1.48 ± 0.18
0.52 ± 0.09
2.86 ± 0.16
1.44 ± 0.14
1.11 ± 0.13
2.04 ± 0.15
0.85 ± 0.13
4.78 ± 0.19
3.45 ± 0.16
2.71 ± 0.18
2.74 ± 0.19
3.18 ± 0.32
0.52 ± 0.08
1.79 ± 0.17
1.33 ± 0.15
1.46 ± 0.14
0.66 ± 0.11
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CEL
CDDO-Mea
a
Positive control.
1 Disrupted the Hsp90-Cdc37 Interaction In Vitro
Many studies have reported that CEL and its derivatives could disrupt the
4
interaction of Hsp90-Cdc37 interaction. To verify whether 41 decreased the level of
the Hsp90-Cdc37 complex in vitro, co-immunoprecipitation was carried out in the
following experiments. First, we needed to confirm that 41 could not inhibit ATP
binding to Hsp90. As shown in Figure 2A, Hsp90α was pulled down by
γ-phosphate-linked ATP-sepharose. It has been reported that CEL could not block the
ATP-binding pocket of Hsp90. Similar to CEL, 41 had no effect on ATP binding.
Next, immunoblot analysis of Cdc37 was carried out after immunoprecipitation of
Hsp90α. Compared with the DMSO group, 41 could decrease the amount of Cdc37
detected when Hsp90α was pulled down with 41 treatment for 6 h (Figure 2B). These
results indicated that 41 could disrupt the Hsp90-Cdc37 interaction in A549 cells.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Effects of 41 on the level of Hsp90-Cdc37 complex in vitro. (A) Effects of
4
1 on the binding of ATP and Hsp90. Purified Hsp90α was pulled down by
γ-phosphate-linked ATP-Sepharose in the presence of 41 or CEL. Hsp90α was
detected by Western blot. (B) Effects of 41 on the amount of Cdc37 associated with
Hsp90. A549 cells were treated with DMSO or 41 for 6 h. Cell lysate were
immunoprecipitated by Hsp90α antibody. Western blot was done for detection of
Hsp90 and Cdc37. (C) Quantitative analysis of co-IP. Data were expressed as the
mean ± SD (n = 3). **P < 0.01 vs. DMSO group.
Comparison of the Regulations of 41 and CEL on Hsp90-Cdc37 Clients
Furthermore, the levels of the Hsp90-Cdc37 clients induced by 41 and CEL were
compared. As shown in Figure 3A, 41 and CEL both inhibited the expression of p-Akt
and Cdk4 at a concentration of 5 μM. The inhibitory activity of 41 was higher than
that of CEL at the same concentration (Figure 3B). These results indicated that the
capability of 41 to disrupt the Hsp90-Cdc37 interaction was stronger than that of
CEL.
ACS Paragon Plus Environment

Page 15 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A
B
CEL
μM
4
1
0
0
.5
.4
5
5
DMSO
*
*
Hsp90
p-Akt
DMSO
1 (5 μM)
CEL (5 μM)
*
*
0.3
41
4
0
0
.2
.1
0
*
*
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
**
Cdk-4
Cdk4
p-Akt
Cdk4
Figure 3. Effects of 41 and CEL on Hsp90-Cdc37clients. A549 cells were treated
with DMSO, 41 and CEL for 12 h, and the relative levels of p-Akt and Cdk4
expression were determined by Western blot assays using Hsp90 as a control. Data
are representative images and expressed as the means ± SD of each group of cells
from three separate experiments. (A) Western blot analysis of the relative levels of
p-AKT and Cdk4 expression. (B) Quantitative analysis. **P < 0.01 vs. the DMSO
group.
4
1 Combined with Cdc37
A previous study reported that CEL might disrupt the Hsp90-Cdc37 interaction
by binding to Cdc37.11,12 To determine whether 41 could play the same role, we first
designed and synthesized an affinity probe (41-Bio) based on the structure of 41 to
verify the binding capacity of 41 to Cdc37. The structure and synthetic route of
4
1-Bio is shown in Scheme 2. As shown in Figure 4, Cdc37 was pulled down by
41-Bio, and 41 could compete with 41-Bio in this assay. These results indicated that
1 could bind to Cdc37 directly.
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
O
CN
a
CN
O
b
OH
HO
CN
HO
O
+
OHC
OH
1
1a
HO
O
1
1b
8a
11c
CN
O
O
O
O
Br
c
O
Br
d
S
O
O
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CN
HO
O
HN
11d
11e
O
N
H
O
O
CN
O
O
O
S
e
O
O
NH
H
O
N
O
H
HO
4
1-Bio
Scheme 2. Reagents and Conditions: (a) 3-bromo-1-propanol, K CO , acetone, reflux,
2
3
o
overnight; (b) K CO , CH OH, 70 C, 6 h; (c) 1,3-dibromopropane, K CO acetone,
2
3
3
2
3,
o
reflux, overnight; (D) Biotin, DCC, DMAP, DMF, 30 C, 24 h. (e) CEL, NaHCO ,
3
o
DMF, 60 C, 12 h.
Figure 4. 41-Bio binds to Cdc37. The recombinant Cdc37 protein (5 mg) were
incubated with different concentrations of 41 for 2 h before or after incubated with
4
1-Bio (0.5 μM) for 2 h, and then pulled down using streptavidin beads. The total
proteins (input), bound proteins (Bind P) were immunoblotted. Data were expressed
as the mean ± SD (n = 3).**P < 0.01 vs. 41 (0.5 μM) group.
Compound 41 Possessed a Higher Affinity for Cdc37 than Hsp90
To further evaluate the protein binding ability of 41, the combination of 41,
ACS Paragon Plus Environment

Page 17 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Cdc37 and Hsp90 was detected by a UV method. This method was carried out by the
characteristics that bound thiol groups in proteins cause the UV absorption at 440 nm
to disappear, as would be the case upon binding to 41 (A, B rings). First, we needed to
confirm whether the CA moiety of 41 could interfere with the reaction. The addition
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
of 41 to dithiothreitol (DTT) was monitored by H-NMR. As shown in Figure 5 (Fig.
S151, 152), the unsaturated hydrogen signal from the CA moiety was retained (green),
indicating that the CA moiety did not affect the addition of the thiol group to 41.
1
Figure 5. Overlay of H NMR spectrum of 41 with and without DL-Dithiothreitol
(
DTT) in DMSO-d . 41 (0.02 mmol) and DTT (0.04 mmol) were dissolved in
6
1
DMSO-d . Then H NMR spectrum of the mixture was detected after reaction in
6
ultrasound for 20 min.
Based on the results proposed above, 41 was added to the purified Hsp90α (full
length) and Cdc37 (full length) respectively, and the UV spectrum of 41 (200-500 nm)
was detected vs. time. Glutathione (GSH) was used as a positive control. As shown in
Figure 6A, both GSH, Cdc37 and Hsp90 reduced the UV absorption of 41 at 440 nm,
indicating that all of three molecules could bind to 41. Analysis of the UV assay
(Figure 6B) showed that the binding of 41 to Cdc37 reached equilibrium within 5 min,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
while the combination of 41 and Hsp90 needed 15 min to equilibrate, indicating that
4
1 possessed a higher affinity to Cdc37 than Hsp90. These results suggested that the
combination of 41 and cysteine (Cys) possessed poorly selectivity in PBS. However,
under physiological conditions, 41 might reach the target protein finally as the
reversibility of the reaction.28 Therefore, 41 might disrupt the interaction of the
Hsp90-Cdc37 complex by preferentially binding to Cdc37 in cancer cells.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
A
41+GSH
41+Cdc37
41+Hsp90
0.987
0.826
0.799
0
min
0
0
0
.700
5 min
15 min
0
0
.800
0
.600
.600
.500
30 min
.600
0.400
0.400
0.325
0.400
0.284
0.176
3
68.00
400.00
450.00
nm.
500.00 520.92
373.83
400.00
450.00
nm.
515.23
373.58
400.00
450.00
nm.
500.00
529.04
B
1.05
1
*
*
*
*
**
**
0
0
0
.95
**
0.9
*
*
41
4
1 + GSH
1 + Cdc37
41
41 + Hsp90
.85
0.8
.75
41
4
*
*
**
0
.7
0.65
.6
0
0
5
15
30
T (min)
Figure 6. UV spectrum of 41 monitored after addition of GSH, Cdc37 and Hsp90.
4
1(100 μM) and GSH (100 μM) or Cdc37 (1.5 μM) or Hsp90 (1.5 μM) were
o
dissolved in PBS (1% DMSO) respectively. Then the solution was incubated at 37 C
for 0, 5, 15 and 30 min. The UV spectrum of the solution (200-500 nm) was
monitored. (A) UV spectrum of 41 was monitored after incubation with GSH, Cdc37
and Hsp90. (B) Analysis of the UV assay. A0, absorbance of the solution at 440 nm
with 0 min. A, absorbance of the solution at 440 nm with the corresponding time.
Data were expressed as the mean ± SD (n = 3).**P < 0.01 vs. 0 min group.
Molecular Docking of 41 to the Hsp90 -Cdc37 Complex
N
M
ACS Paragon Plus Environment

Page 19 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
To further investigate the mechanism of 41 inhibiting Hsp90-Cdc37 interaction,
molecular docking was first carried out to mimic the interaction between 41 (Figure
7
A) and the hydrophobic pocket of the human Hsp90 -Cdc37 protein complex
N M
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
(PDB ID: 2K5B). As shown in Figure 7B, 41 was docked in the hydrophobic pocket
of Hsp90 -Cdc37 . Four H-bonds might be formed, including between the β-CN
N
M
group of the CA moiety of 41 and R166 of Cdc37 (yellow, Figure 7B), the carbonyl
group of the CEL moiety and W168 of Cdc37 (purple, Figure 7B) and the carbonyl
group of the CA moiety (Figure 7B) and R166 (H-bond×2). These hydrogen bonds
might allow 41 efficiently disrupt the key interaction (hydrogen bonds between
Q133Hsp90 and R166/R167Cdc37) of the Hsp90 -Cdc37 by binding to Cdc37
N
M
(non-covalent bond), thereby affecting the stability of Hsp90-Cdc37.
Figure 7. Molecular docking of 41 to Hsp90 -Cdc37 . (A) Chemical structure of 41.
N
M
(
B) Binding mode of 41 with the Hsp90 -Cdc37 complex (PDB ID 2K5B). The
N M
complex was shown in a cartoon representation. The side chain R166 (yellow), R167
green) and W168 (purple) of Cdc37 and 41 (gray) were shown in a stick type. The
(
M
expected hydrogen bonds were indicated by the dashed yellow lines.
4
1 might form a covalent bond with C203 of Cdc37M
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The reaction between quinone methide (A, B rings) and cysteine (Cys) has been
considered as a key factor for the high activity of CEL derivatives. As Jiang. et al
report, CEL derivatives might form a covalent bond with C203 of Cdc37 to disrupt
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
the interaction of the Hsp90-Cdc37. In order to investigate the potential mechanism
of 41 inhibiting the Hsp90-Cdc37 interaction, the A,B rings were further modified
(41-H, Figure 8, Scheme S1). As shown in Figure 8, the inhibition rate of 41-H (20.5
%) was decreased compared to that of 41 (65.1 %) and the mixture of 41 and Methyl
thioglycolate (68.3 %) in A549 at 0.5 μM. These results indicated that the A, B rings
were critical for the activity of 41. Therefore, 41 might inhibit the activity of Hsp90
by the combination between A, B rings and Cys in the Hsp90-Cdc37 complex. This
complex depends on a hydrophobic interaction which is mainly located in a
hydrophobic core between Hsp90 and Cdc37 . And Hsp90 does not contain any
N
M
N
Cys residues, while Cdc37 shows two Cys residues (C183 and C203) on the protein
M
surface. As shown in Figure 9A, C183 is inaccessible as it is located in the depression
on the protein surface. However, the position of C203 closes to the hydrophobic core
(cyans) in Cdc37 and is more easily approached by the compounds (Figure 9B).
Therefore, 41 most likely to combine with C203Cdc37 by covalent bond to disrupt the
interaction of the Hsp90-Cdc37 complex.
ACS Paragon Plus Environment

Page 21 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
F
O
O
O
O
H
O
HS
O
O
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
HO
41
Methyl thioglycolate
No activity
Inhibition rate: 65.1 ± 4.8 %
A549
F
O
O
O
4
1 + Methyl thioglycolate = 1:1
Inhibition rate: 68.3 ± 5.6 %
O
CN
H
O
O
O
O
S
41-H
Inhibition rate: 20.5 ± 2.6 %
O
O
Figure 8. Anti-proliferative activity of 41-H in comparison with that of 41, methyl
thioglycolate and the mixture of 41 and methyl thioglycolate in A549 cells.
Figure 9. The structure of Cdc37 (PDB ID 2W0G). (A) The location of Cys183. (B)
M
The location of Cys203.
Compound 41 Regulated the Hsp90-Cdc37 Clients and Apoptosis-related Proteins
Based on prophase research, 41 could be considered sa an Hsp90-Cdc37
disruptor. Disruption of Hsp90 and Cdc37 will down-regulates the levels of
2
9
Hsp90-Cdc37 clients, resulting in the induction of apoptosis-related proteins. Thus,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the effects of 41 on the levels of Hsp90-Cdc37 clients and apoptosis-related proteins
were detected. As important clients of Hsp90-Cdc37, Cdk4 and p-Akt were selected
to measure by Western blotting. These proteins were tested after treatment with 41 in
A549 cells. As shown in Figure 10A, 41 down-regulated the levels of p-Akt and Cdk4
in a dose-dependent manner. The levels of Bax and cleaved caspase-3 were
significantly increased. In contrast, as an anti-apoptotic protein, Bcl-2 expression was
inhibited (Figure 10B). These results indicated that 41 regulated the levels of
Hsp90-Cdc37 clients and the apoptosis-related proteins to induce cell apoptosis.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
41
B
A
DMSO 1.25 2.5
5
μM
1.8
Hsp90
p-Akt
Cdk-4
1.6
1
1
.4
.2
1
DMSO
41
41 (1.25 μM)
**
*
*
*
4
41
1
(2.5 μM)
0
0
.8
.6
**
**
**
Bcl-2
**
41
41 (5.0 μM)
*
*
*
Bax
**
*
*
0.4
0.2
0
Caspase-3
β-actin
*
*
**
Hsp90
p-Akt
Cdk4
Bcl-2
Bax
Caspase-3
Figure 10. Effects of 41 on clients of Hsp90-Cdc37 and apoptosis-related proteins in
A549 cells. A549 cells were treated with DMSO or 41 for 12 h, and the relative levels
of Hsp90, p-AKT, Cdk4, Bcl-2, Bax and Caspase-3 expression were determined by
Western blot assays using β-actin as a control. Data are representative images and
expressed as the means ± SD of each group of cells from three separate experiments.
(A) Western blot analysis of the relative levels of Hsp90, p-Akt, Cdk4, Bcl-2, Bax and
Caspase-3 expression. (B) Quantitative analysis. *P < 0.05, **P < 0.01, vs the
DMSO-treated control group.
Compound 41 Decreased the Mitochondrial Membrane Potential (Δψm)
ACS Paragon Plus Environment

Page 23 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
As mentioned above, 41 could regulate the levels of Hsp90 clients and
apoptosis-related proteins to induce cell apoptosis. Changes in the Δψ are hallmark
m
events of early apoptosis, regardless of whatever the apoptosisit is extrinsic apotosis
or intrinsic apoptosis. Thus, to investigate the influence of 41 on the Δψm, JC-1
staining was used in A549 cells. As shown in Figure 11 A & B, an increase in the cell
number with J-monomers (green fluorescence, depolarized mitochondria) and a
concurrent decrease in the cell number with Jaggregates (red fluorescence, indicating
hyperpolarized mitochondria) were exhibited in this assay. In addition, a significant
loss in the Δψm was induced by treatment of 41 at 0.2 µM, which indicated that 41
could significantly induce apoptosis in A549 cells at a low concentration.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
DMSO
41
1
1
20
00
**
**
8
6
0
0
**
4
1
41
40
**
**
20
**
0
DMSO
4
4
1
1 (0.2 µM)
4
4
1
1 (0.4 µM)
4
4
1
1 (0.8 µM)
Cells with red aggregate
Cells with green aggregate
Figure 11. Effects on Δψ of 41 in A549 cells. A549 cells were treated with DMSO
m
or 41 for 24 h and then analyzed by fluorescence microscopy and flow cytometry after
JC-1 staining. (A) Flow cytometry analysis of A549 cells treated with DMSO and 41.
(
B) The quantification of cells (%) with red and green aggregates. The values are
presented as mean ± SD (n = 3). **P < 0.01 vs. the DMSO group.
The Morphological Apoptosis Induced by 41
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Next, the morphological apoptosis induced by 41 was detected by Hoechst 33342
and propidium iodide (PI) staining. Hoechst 33342 stained the apoptotic cells bright
blue, and PI stained the dead cells red. As shown in Figure 12, the cell density
decreased significantly as drug concentration increased. Simultaneously, the number
of apoptotic cells were increased. The characteristics of apoptotic cells were observed
in apoptotic cells (white arrows), such as nuclear fragmentation and chromatin
condensation.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 12. The morphological apoptosis induced by 41 in A549. A549 cells were
treated with 41 or DMSO for 48 h. Fluorescence microscopy images of A549 cells
stained by Hoechst 33342 and PI. The apoptotic cells were stained by Hoechst 33342
in bright blue and the dead cells were stained by PI in red.
Effects of 41 on Cell Apoptosis
To examine whether apoptosis was induced by 41 , an annexin V-FITC/PI dual
staining assay was performed in A549 cells. As shown in Figure 13A, after 48 h of
treatment with 41, significant apoptosis was induced in a dose-manner, especially in
late apoptotic cells (9.3%, 16.9% and 25.2% at the concentrations). Furthermore, at a
concentration of 1.2 μM, the apoptotic cells induced by treatment with 41 accounted
for 32.6% of the cells, which was higher than that of CEL (23.1%) (Figure 13B).
These results suggested that the superior activity of 41 might be mainly through an
ACS Paragon Plus Environment

Page 25 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
apoptotic pathway.
A
41
41
41
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
120
B
1
20
00
80
60
40
100
PI
1
*
*
8
6
4
2
0
DMSO
*
*
**
DMSO
*
*
41 (1.2 μM)
41
41 (1.2 μM)
0
0
0
0
CEL (1.2 μM)
CEL (1.2 μM)
* ** *
**
**
2
0
0
Annexin V
Living cells
Living cells
Apoptotic cells
Apoptotic cells
Figure 13. Apoptotic effects of 41 and CEL in A549 cells. Treatment with DMSO or
1 or CEL for 48 h, A549 cells were collected and stained with Annexin V/PI,
4
followed by flow cytometric analysis. (A) Flow cytometry analysis of A549 cells
treated with DMSO, 41 and CEL. (B) Representative histograms for the numbers of
living cells and apoptotic cells (%). The values were presented as the mean ± SD (n =
3
). **P < 0.05 vs. the DMSO group.
Effects of 41 on the Cell Cycle
Cdk4 is a key regulator of the cell cycle. Inhibition of Cdk4 expression will arrest
the cell cycle in the G /G phase. Western blotting verified that 41 could decrease the
0
1
amount of Cdk4. Thus, we further investigated the effects of 41 on cell cycle
progression. As shown in Figure 14A & B, 41 arrested the cell cycle in the G /G
0
1
phase in a dose-dependent manner in A549 cells. These results indicated that the
effects of 41 on the cell cycle were another pathway that induced the apoptosis in
cancer cells.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
DMSO
41
2
2
50
00
Freq. G1 = 60.14
Freq. S = 27.59
Freq. G2 = 19.37
300
200
Freq. G1 = 64.36
Freq. S = 26.39
Freq. G2 = 16.78
120%
100%
80%
1
50
1
00
1
00
0
5
0
G2
S
**
**
*
6
4
2
0%
0%
0%
0
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
00K 400K 700K 1M
100K 400K 700K 1M
G1
4
1
41
Freq. G1 = 67.51
Freq. S = 23.34
Freq. G2 = 16.00
Freq. G1 = 70.49
Freq. S = 23.85
Freq. G2 = 15.55
3
00
300
0
%
2
00
DMSO
41 41
41 (0.2 µM) 41 (0.4 µM) 41 (0.8 µM)
41
2
00
100
0
100
0
100K 400K 700K 1M
100K 400K 700K 1M
PI
Figure 14. Effect of 41 on cell cycle progression of A549 cells. Treatment of A549
cells with DMSO and 41 (0.2, 0.4 and 0.8 μM) for 24 h, intracellular DNA was
stained with PI. Cell cycle distribution was analyzed by flow cytometry. (A) Flow
cytometry analysis of cell cycle distribution of 41 on A549. (B) The quantification of
cells (%) in each cell phase. The values are presented as mean ± SD (n = 3). *P < 0.01,
*
*P < 0.05 vs. the DMSO group.
Conclusion
Based on the analysis of the Hsp90 -Cdc37 structure, lipophilic fragments with
N
M
small molecules (CA and α- or β-CN substituted CA) were introduced into CEL to
synthesize 48 new derivatives. Their anti-proliferative activities were evaluated
against four cancer cell lines (A549, HOS, MCF-7 and HepG2) and the SAR of these
derivatives was discussed. Further study showed that 41 disrupted the interaction of
the Hsp90-Cdc37 complex in A549 cells. Molecular docking analysis indicated that
41 might inhibit the interaction of Hsp90-Cdc37 by binding to Cdc37, and 41 bound
to Cdc37 more efficiently than Hsp90 in UV assay. Pharmacological studies showed
ACS Paragon Plus Environment

Page 27 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
that 41 could regulate the levels of Hsp90-Cdc37 clients and the apoptosis-related
proteins to induce the cell cycle arrest and the apoptosis in A549 cells. Jiang’s study
indicated that introduction of the polar groups in C29-carboxyl of CEL were favorable
on Hsp90-Cdc37 disruption. However, a suitable hydrophobic group could also
enhance Hsp90-Cdc37 disruption activity when introduced to this site in our and
previous study. These results suggested that mechanism of disrupting the
Hsp90-Cdc37 interaction by small molecules was complex. The binding capability is
not only related to the polar of the compound’s structure, but also the spatial structure,
flexibility and electrical properties, etc. Overall, 41 displayed superior
anti-proliferative activity against cancer cells by disrupting the interaction of
Hsp90-Cdc37 by preferentially binding to Cdc37 (hydrogen bond and/or covalent
bond). The current finding may provide new insight into the design of CEL
derivatives as novel Hsp90-Cdc37 disruptors.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
EXPERIMENTAL PROTOCOLS
(1) Chemical Analysis. Celastrol was purchased from Tuochukangyuan company
(Hubei, China). Synthetic reagents were purchased from chemical and biological
company. The following antibodies used for immunoprecipitation and western
blotting (Hsp90α, Cdc37; Hsp90, p-Akt, Cdk4, Bcl-2, Bax and Caspase 3) were
purchased from Abcam. The purity of all tested compounds was characterized by
HPLC analysis (LC-2030C HPLC system consisting of LC-2030C pumps and an
SPD-2030C UV detector). Individual compounds with a purity of >95% were used for
subsequent experiments (see Supporting Information).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 83
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(2) General Procedures for the Preparation of Compounds 2f-t. 1a/1b/1c/1d/1e
(
1.2 mmol, 1.2 eq), EDCI (288.0 mg, 1.5 mmol, 1.5 eq) and DMAP (61.0 mg, 1.0
mmol, 0.5 eq) were successively added to a solution of bromohydrin (1.0 mmol, 1.0
eq) in CH Cl (6 ml) and this reaction mixture was stirred for 7 h at room temperature.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
The reaction mixture was concentrated under vacuum and the residue was dissolved
in CH Cl . The organic layer was washed with 5% aqueous NaHCO solution (60 ml)
2
2
3
and aqueous saturated NaCl solution (30 ml). After drying over anhydrous Na SO ,
2
4
the CH Cl was removed in vacuum. The crude materials was purified by column
2
2
chromatography on silica gel [petroleum ether (PE)/ethyl acetate (EA) = 16:1 (v/v)] to
afford pure compounds 2f-t.
(3) General Procedures for the Preparation of Compounds 5a-e. NH OAc
4
(
577.5 mg, 7.5 mmol, 0.5 eq) and different aryl aldehydes 3a-e (13.5 mmol, 0.9 eq)
were dissolved in a solution of 4a (15.0 mmol, 1 eq, 1.28g) in tolune (45 ml). The
mixture was heated at 100℃ for 12 h. The reaction mixture was filtered to afford the
crude product 5a-5e.
(4) General Procedures for the Preparation of Compounds 6f-t. 6f-6t were
synthesized from 5a-5e by using the methods employed in the case of 2f-2t. The
crude materials was purified by column chromatography on silica gel [PE/EA=16:1
(v/v)] to achieve 6f-t.
(
5) General Procedures for the Preparation of Compounds 9a-e. K CO (3.11 g,
2 3
22.5 mmol, 1.5 eq) and different aryl acetonitriles 7a-e (15.0 mmol, 1.0 eq) were
added to a solution of 8a (1.11 g, 15.0 mmol, 1.0 eq) in CH OH (50 ml). The mixture
3
ACS Paragon Plus Environment

Page 29 of 83
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
o
was refluxed at 60 C for 6 h. The reaction mixture was filtered and the filter cake was
washed by CH Cl and then dissolved in water. The solution was acidified using 1 M
2
2
hydrochloric acid solution to pH = 4 and filtered. The filter cake was the crude
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
product 9a-e.
(
6) General procedures for synthesizing compounds 10f-t. 10f-10t were
synthesized from 9a-e by using the methods employed in the case of 2f-2t. The crude
materials were separated by column chromatography on silica gel [PE/EA=16:1 (v/v)]
to achieve 10f-t.
(7) General procedures for synthesizing compounds 10u-w. 9a/9c/9d (0.5 mmol,
+
-
1
.0 eq), K CO (138.0 mg, 1.0 mmol, 2.0 eq), Bu N Br (9.7 mg, 0.06 eq, 0.03 mmol)
2 3 4
and trans-1,4-dibromo-2-butene (321.0 mg, 1.5 mmol, 3.0 eq) were added into
o
anhydrous acetone. The reaction mixture was refluxed for 7 h at 60 C. The acetone
was removed under reduced pressure and the residue was extracted with ethyl acetate
(60 ml). The combined organic layers were dried over sodium sulfate, filtered, and the
solvent was evaporated. The crude materials were purified by column
chromatography on silica gel [PE/EA = 16:1 (v/v)] to afford 10u-w.
(
8) General Procedures for the Preparation of Target Compounds 1-48. A
mixture of CEL (45.1 mg, 0.1 mmol, 1.0 eq), 2a-o or 6a-o or 10a-r (0.3 mmol, 3.0 eq)
and NaHCO (42.0 mg, 0.5 mmol, 5.0 eq) in anhydrous DMF (6 ml) was stirred at 60
3
o
C for 7 h. The reaction solution was washed with CH Cl (100 ml) for four times.
2
2
The combined organic layers were washed by aqueous saturated NaCl solution (30
ml), and then dried over sodium sulfate, filtered, and the solvent was evaporated. The
ACS Paragon Plus Environment