Synthesis and evaluation of a novel class Hsp90 inhibitors containing 1-phenylpiperazine scaffold

Article abstract of DOI:10.1016/j.bmcl.2014.01.070

Previously, we identified 1-(2-(4-bromophenoxy)ethoxy)-3-(4-(2- methoxyphenyl)piperazin-1-yl)propan-2-ol (1) as a novel Hsp90 inhibitor with moderate activity through virtual screening. In this study, we report the optimization process of 1. A series of analogues containing the 1-phenylpiperazine core scaffold were synthesized and evaluated. The structure-activity relationships (SAR) for these compounds was also discussed for further molecular design. This effort afforded the most active inhibitor 13f with improved activity in not only target-based level, but also cell-based level compared with the original hit 1.

Full text of DOI:10.1016/j.bmcl.2014.01.070

Bioorganic & Medicinal Chemistry Letters 24 (2014) 1557–1561
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and evaluation of a novel class Hsp90 inhibitors containing
1-phenylpiperazine scaffold
Jian-Min Jia a,b, Fang Liu a,b, Xiao-Li Xu a,b, Xiao-Ke Guo a,b, Fen Jiang a,b, Bahidja Cherfaoui a,b
,
Hao-Peng Sun a,b,c, , Qi-Dong You a,b,
⇑
⇑
a Jiang Su Key Laboratory of Drug Design and Optimization, China Pharmaceutical University, Nanjing 210009, China
b State Key Laboratory of Natural Medicines, China Pharmaceutical University, Nanjing 210009, China
c Department of Medicinal Chemistry, School of Pharmacy, China Pharmaceutical University, Nanjing 210009, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Previously, we identified 1-(2-(4-bromophenoxy)ethoxy)-3-(4-(2-methoxyphenyl)piperazin-1-yl)pro-
pan-2-ol (1) as a novel Hsp90 inhibitor with moderate activity through virtual screening. In this study,
we report the optimization process of 1. A series of analogues containing the 1-phenylpiperazine core
scaffold were synthesized and evaluated. The structure–activity relationships (SAR) for these compounds
was also discussed for further molecular design. This effort afforded the most active inhibitor 13f with
improved activity in not only target-based level, but also cell-based level compared with the original
hit 1.
Received 18 October 2013
Revised 12 January 2014
Accepted 25 January 2014
Available online 15 February 2014
Keywords:
Hsp90 inhibitors
Structure–activity relationships
Structure-based drug design
Ó 2014 Elsevier Ltd. All rights reserved.
The 90 kDa family of heat shock proteins (Hsp90) has become a
validated target for the treatment of cancer because of its central
role in correct folding and stabilization of proteins involved in
malignant behavior and tumor progression.1 The Hsp90 client pro-
teins include many oncogenic proteins such as Her2/ErbB2, Akt,
ited potent in vitro activity with low micromolar IC50 values in ATPase
activity and anti-proliferation assay. The morphological changes induced
by 1 were observed in MCF-7 cancer cells. In addition, a panel of the client
proteins, including Her2, Src, Akt, ERK, c-Raf and HIF-1a, were also found
to be downregulated by 1. These data above encouraged us to perform
Raf-1, Cdk4, HIF-1a, MET, hTERT, hormone receptors, mutant
the further optimization.
p53, surviving and IP6K2, which are related to the development
and progression of cancer.2–5 Many of them are involved in apop-
tosis, signal-transduction pathways and cell-cycle regulation.6
Thus, targeting Hsp90 may have the potential advantage of simul-
taneously blocking multiple oncogenic pathways. In addition,
Hsp90 is overexpressed in many types of cancers in humans and
the active Hsp90 in cancer cells has higher affinity to Hsp90 inhib-
itors than the latent form in normal cells.7,8 These evidences sup-
ported the hypothesis that Hsp90 inhibitors have shown
selectivity for cancer cells. Taken together, it is now evident that
Hsp90 is a promising cancer target, and the development of
Hsp90 inhibitors has become one of the most active areas of drug
discovery for cancer chemotherapy.
The availability of the 3D-structure of Hsp90 protein has en-
hanced opportunities for the rapid optimization of hit compounds.
In order to obtain more potent compounds with improved drugga-
bility, the binding mode of 1 in the ATP-binding pocket of Hsp90
(Fig. 1A, PDB ID: 2XJX) was predicted using GOLD program with
the procedure described previously.9 Although 1 bind well to
Hsp90, it only inserted into part of the binding site, missing the
occupation of the hydrophobic sub-pocket P1 (Fig. 1A). To solve
this problem, we tried to change the methoxy group on the A ring
(Table. 1) with different kinds of substituents, including steric,
electron-withdrawing and electron-donating groups, leading to
the first series of compounds (ranging from 7a to 7o, first series
in Table 1).
In the previous study, we identified a novel Hsp90 inhibitor 1-(2-(4-
bromophenoxy)ethoxy)-3-(4-(2-methoxyphenyl)piperazin-1-yl)propan
-2-ol (1, Table 1) by a combinatorial approach of pharmacophore
modeling, virtual screening and molecular docking.9 Compound 1 exhib-
The preparation of the first series was described in Scheme 1.
Reaction of commercially available 2 with 2-chloro-1-ethanol in
10% aqueous NaOH solution afforded 3. The intermediate 3 was
treated by epichlorohydrin in THF in the presence of NaH to
provide intermediate 4. The intermediates 6a–6k and 6n–6o were
prepared from commercially available substituted anilines 5a–5k
and 5n–5o using the method reported by Liu et al.10 Finally,
7a–7k and 7n–7o were synthesized by coupling the intermediates
⇑
Corresponding authors. Tel./fax: +86 025 83271216 (H.-P.S.); tel./fax: +86 025
83271351 (Q.-D.Y.).
E-mail addresses: sunhaopeng@163.com (H.-P. Sun), youqidong@gmail.com
(Q.-D. You).
http://dx.doi.org/10.1016/j.bmcl.2014.01.070
0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

1558
J.-M. Jia et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1557–1561
Figure 1. The binding patterns of compounds 1 (A), 7d (B), 8b (C) and 13f (D) in the active site of Hsp90.
7a: R = methyl;
7b: R = ethyl;
O
7c: R = isopropyl;
O
OH
b
OH
O
O
7d
: R = tert-butyl;
a
7e: R = pyrrolidinyl;
7f: R = piperidyl;
7g: R = acetamido-;
OH
N
O
N
R
Br
O
Br
4
Br
3
7h
d
: R = methanesulfonamide;
2
7i: R = benzyl methoxy;
7j: R = nitro;
e
Br
R
R
NH
7k
: R = nitrile;
c
7a - 7o
NH2
N
f
7l
: R = hydroxy;
7m: R = amino;
7n: R = -NMe2;
5a - 5k, 5n - 5o
6a - 6k, 6n - 6o
7o
: R = -N(Me)Piv.
Scheme 1. Reagents and conditions: (a) 10% NaOH, ethylene chlorohydrin; (b) NaH, THF, epichlorohydrin; (c) 150 °C, bis(2-chloroethyl)amine hydrochloride; (d) EtOH,
reflux; (e) BBr3, DCM, À78 °C; and (f) Fe, HCl, MeOH, reflux.
4 and 6a–6k or 6n–6o in ethanol under reflux. The compound 7l
was obtained by removing the benzyl protecting group of 7i. The
compound 7m was prepared via reduction of 7j. Detailed synthetic
procedures and analytical data of all compounds were described in
Supporting information.
ther tested for their cytotoxicity against a series of cancer cell
lines including Sk-Br-3, MCF-7 and HCT116 (Table 1). Unlike
our expectation, the results showed most of these derivatives dis-
played decreased cytotoxic activities. Only two compounds 7d
and 7f exhibited slight improvement.
The resulting compounds were tested by Discover RX ADP
Hunter™ Plus Assay kit (ADP assay) to determine the inhibitory
effects on Hsp90 described previously (Table 1).9 The steric
substituents on the phenyl ring (7a–7f and 7i) were favorable,
exhibiting more or less increase in potency when compared with
hit compound 1. In addition, the electron-donating groups on the
phenyl ring (7e, 7f, 7i and 7n) showed the similar inhibitory ef-
fects on Hsp90 when compared with compound 1. However,
the groups containing labile protons (7l and 7m) and the elec-
tron-withdrawing groups (7g, 7h, 7j, 7k and 7o) on the phenyl
ring were unfavorable, showing a reduction in potency (IC50
On the basis of these findings, compound 7d was selected for
further optimization through in-depth structural modification.
Firstly we planned to exploit the significance of the hydroxyl group
of 7d. Although it served as the H-bond donor, we thought it may
be easily oxidized in vivo, making the molecule unstable during
metabolism. Therefore, we tried to substitute the hydroxyl with
other groups, leading to compounds 8a and 8b (Scheme 2). Com-
pound 8a was prepared by methylation of 7d using methyl iodide
in the presence of sodium hydride. Reaction of the intermediate 3
with 1-bromo-3-chloropropane in THF in the presence of sodium
hydride afforded 9. Reaction of 9 with the intermediate 6d in the
presence of potassium carbonate provided the compound 8b.
Detailed synthetic procedures and analytical data of 8a and 8b
were described in Supporting information.
>20 lM). The data indicated that the steric substituents were tol-
erated on the hydrophobic sub-pocket P1. This phenomenon
could be explained by that the sub-pocket P1 was comprised of
the hydrophobic residues Trp162, Leu107 and Tyr139 (Fig. 1B).
The most potent compound 7d exhibited sixfold increase in
Unfortunately, the change of the hydroxyl group lead to the
sharp decreases in the inhibitory activities in both target-based
and cell-based assay (Table 2). From the docking model, the
hydroxyl oriented to the polar region of the active site, forming
Hsp90 inhibition potency (IC50 0.97 0.3
lM) compared with 1
(IC50 5.31 2.1 M). Subsequently, these compounds were fur-
l

J.-M. Jia et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1557–1561
1559
Table 1
Antiproliferative activity in different cell lines and Hsp90 ATPase activity inhibition of 7a–7o
A
OH
N
O
N
R
O
B
Br
Compd no.
R
Hsp90-ATPase activity inhibitiona (IC50
,
lM)
Antiproliferative activity cell linesa (IC50
, lM)
SK-Br-3
MCF-7
HCT116
1
7a
7b
7c
7d
7e
7f
7g
7h
7i
7j
7k
7l
7m
7n
7o
17-DMAG
–OMe
–Me
–Et
–iPr
5.31 2.1
5.12 1.7
2.43 0.1
1.03 0.8
0.97 0.3
5.43 3.1
4.37 1.2
11.45 4.9
13.84 0.8
3.67 1.8
30.41 1.3
25.77 2.2
36.49 0.7
22.61 1.9
6.32 0.3
15.29 1.3
0.93 0.1
4.42 0.4
13.54 2.3
12.43 1.7
14.51 1.2
3.01 2.0
14.8 3.6
6.17 1.4
27.72 1.8
11.14 0.9
8.43 2.3
20.43 0.8
15.63 5.2
34.56 3.5
31.41 4.1
9.03 1.2
23.21 2.3
1.34 2.7
5.67 1.7
11.65 0.5
17.88 4.2
27.16 2.5
4.16 0.8
7.58 4.1
2.91 3.2
22.25 3.6
22.83 2.4
17.87 2.4
23.26 4.1
29.15 2.2
27.97 0.1
23.76 0.6
11.90 2.5
25.01 1.1
0.39 0.8
6.61 4.9
12.93 0.1
15.19 0.4
12.38 1.3
2.51 0.9
–tBu
Pyrrolidin-1-yl
Piperidin-1-yl
–NHCOCH3
–NHSO2Me
–OBn
–NO2
–CN
–OH
–NH2
9.97 1.1
5.24 0.7
20.41 4.1
19.14 3.1
12.34 1.1
33.21 3.7
24.44 6.4
18.22 2.9
19.87 1.5
17.83 1.9
30.29 2.1
0.78 1.2
–NMe2
–N(Me)Piv
—
a
Values shown are mean SD (n = 3).
CH3
Br
a
N
O
7d
N
O
O
8a
Br
N
O
c
b
O
N
O
13a
: R = 4-Cl;
13b: R = 3-Cl;
13c: R = 2-Br;
OH
O
Cl
O
Br
Br
3
9
8b
13d
: R = 4-Me;
13e: R = 3-OMe;
13f: R = 4-OMe;
13g: R = 4-Pr;
O
OH
N
N
OH
O
O
O
O
OH
13h
O
: R = 4-F;
f
d
e
13i: R = 4-OCF3;
13j: R = 3-NO2;
13k: R = 2,4-F;
R
R
R
R
13l
:
R = 3-Cl, 4-Br;
13a - 13n
10a - 10n
11a - 11n
12a - 12n
13m: R = 2-OMe;
13n: R = 4-Et;
Scheme 2. Reagents and conditions: (a) NaH/THF, CH3I, 50%; (b) NaH/THF, 1-bromo-3-chloropropane, 3 h, reflux; (c) anhydrous K2CO3, acetone, reflux; (d) 10% NaOH,
ethylene chlorohydrin; (e) NaH, THF, epichlorohydrin; and (f) EtOH, intermidate 6d, reflux.
H-bond with Gly135, however, this polar interaction was missing
when compound 8a or 8b bond to Hsp90 (Fig. 1C), lowering the
binding affinity. The data proved the significance of the hydroxyl
group.
In the next design cycle, we turned our attention to the modifi-
cation of the B ring (Table 2) since we found 7d missed the inter-
action to another polar sub-pocket (P2, Fig. 1B) adjacent to the
main pocket. The –Br at the 4-position could not sufficiently occu-
py the P2 pocket. Therefore, we hypothesized that the overall
activity might be improved by gaining tight interactions with the
residues in P2 pocket if a substituent at the B ring was properly
oriented. Scheme 2 described the preparation of 13a–13n. The
synthesis methods of 13a–13n were similar with that of 7d in
the Scheme 1. Reaction of commercially available phenols
10a–10n with 2-chloride ethanol in 10% aqueous NaOH solution
provided 11a–11n. Compounds 11a–11n were reacted with epi-

1560
J.-M. Jia et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1557–1561
Table 2
Antiproliferative activity in different cell lines and Hsp90 ATPase activity inhibition of 8a, 8b and 13a–13n
A
R2
N
O
N
O
R1
B
Compd no.
R1
R2
Hsp90-ATPase activity inhibitiona (IC50
,
lM)
Antiproliferative activity cell linesa (IC50
,
l
M)
HCT116
2.51 0.9
>100
SK-Br-3
MCF-7
7d
8a
8b
4-Br
4-Br
4-Br
4-Cl
3-Cl
2-Br
4-Me
3-OMe
4-OMe
4-Pr
–OH
–OMe
–H
0.97 0.3
70.43 0.7
>100
3.01 2.0
43.34 1.9
32.38 1.7
1.01 0.8
2.22 0.1
15.2 1.6
5.12 0.4
1.91 1.5
1.04 1.8
4.13 1.8
13.13 2.2
17.63 3.1
21.13 0.5
25.35 3.1
5.21 0.4
1.53 1.1
3.22 1.4
1.34 2.7
4.16 0.8
21.19 1.5
57.48 1.1
2.13 0.4
3.06 0.3
6.98 2.8
3.14 0.2
2.03 0.2
0.83 0.9
2.73 1.4
21.67 0.1
20.22 1.2
27.02 3.8
19.96 1.3
9.85 1.9
1.65 0.7
3.31 1.5
0.39 0.8
35.10 2.7
3.23 0.3
2.01 1.9
10.27 0.4
4.86 1.9
0.98 1.5
1.01 0.6
3.41 1.3
23.88 2.5
14.49 0.9
16.42 1.7
29.43 0.2
8.72 0.2
1.21 1.3
4.17 2.1
0.78 1.2
13a
13b
13c
13d
13e
13f
13g
13h
13i
13j
13k
13l
13m
13n
17-DMAG
–OH
–OH
–OH
–OH
–OH
–OH
–OH
–OH
–OH
–OH
–OH
–OH
–OH
–OH
—
0.75 0.1
1.43 1.2
2.25 1.9
0.87 0.6
2.03 2.2
0.64 0.7
0.95 1.3
6.42 2.4
8.69 0.9
11.79 1.8
21.23 1.2
9.78 2.8
1.21 0.2
0.92 0.1
0.93 0.1
4-F
4-OCF3
3-NO2
2,4-di-F
3-Cl, 4-Br
2-OMe
4-Et
—
a
Values shown are mean SD (n = 3).
chlorohydrin in THF in the presence of NaH to provide compounds
12a–12n. Then, the intermediates 12a–12n and 6d were combined
into the desired products 13a–13n using K2CO3/EtOH conditions
(in Supporting information).
From Table 2, the chloro and electron-donating substituents at
the 4 position of the B ring (13a, 13d, 13f, 13g and 13n) enhanced
the potency in inhibitory activity against Hsp90. This phenomenon
showed better inhibition activity against Hsp90 than that of
the substituents at other position (13a > 13b, 7d > 13c,
13f > 13m > 13e). To our delight, the introduction of chloro or
methoxyl group at the B ring (13a, 13b, 13e, 13f and 13m) led to
the enhancement of this series compounds in anti-proliferation
against a series of cancer cell lines with high dependence on
Hsp90 (Table 2). The successful replacement prompted further
investigation on the substituent through synthesis of a series of
related analogs (Scheme 2). From Table 2, the electron-donating
substituents at the 4 position of B ring were appropriate to both
potency and anti-proliferation activities, but the electron-with-
drawing groups and bromine substituents (13c and 13i–13k) were
detrimental to anti-proliferation activities. Considering the
enzyme-based and cell-based evaluation data together, 13f, the
methoxyl group at the 4-position of the B ring, was most effective
was further confirmed by comparing 13f (IC50 0.64 0.7
lM)
with 13n (IC50 0.92 0.1 M). Oppositely, the strong electron-
l
withdrawing groups at this ring resulted in reduction of inhibitory
activity (13h–13k, Table 2). Meanwhile, di-substitution at ring B
also resulted in reduction of inhibitory activity (13k–13l, Table 2).
It is worth mentioning that the substituents at the 4 position
in producing the inhibitory activity (IC50 0.64 0.7
lM) against
both Hsp90 protein and the cancer cell lines (IC50 1.04 1.8,
0.83 0.9 and 1.01 0.6
respectively).
lM in SK-Br-3, MCF-7 and HCT116,
To further characterize 13f as a potential Hsp90 inhibitor, MCF-
7 cells were treated with varying concentrations of 13f for 36 h,
and equivalent amounts of protein from cell extracts were Wes-
tern blotted for Hsp90, Hsp70 and a series of client proteins of
Hsp90, including Her2, Akt, ERK and c-Raf, using b-actin as a load-
ing control. Compound 13f was observed to deplete the Hsp90-
dependent client proteins in a concentration-dependent fashion
(Fig. 2), which was in a similar manner with its antiproliferative
activity. Meanwhile, 13f dose-dependently up-regulated Hsp70,
which was considered as the feedback of the Hsp90 inhibition.
These data all confirmed that 13f inhibited the activity of Hsp90,
leading to the misfolding and subsequent degradation of the cli-
ents proteins.
Figure 2. Western-blot analysis of the expression level of Hsp90, Hsp70 and several
client proteins of Hsp90 after incubation with 13f in MCF-7 cells. Compound 13f
was incubated with MCF-7 cancer cells at indicated concentrations (lM) for 36 h.
Cell extracts were prepared and equivalent amounts of protein were separated by
SDS–PAGE and subsequently Western-blotted for the indicated proteins.
In an attempt to obtain some structural insights into the mech-
anism of inhibition by the optimized inhibitor 13f for Hsp90, its
binding modes in the ATP-binding pocket of Hsp90 was calculated
using GOLD. The predicted binding mode of 13f with Hsp90

J.-M. Jia et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1557–1561
1561
The SAR for the series I-III
the removal and substitution of the hydroxyl
group decreased the inhibitory activity both
in enzyme-based and cell-based assay; the
hydroxyl group formed H-bond with Gly135
Best potent compound
A
Gly135
X
N
O
N
R1
OH
N
O
O
O
N
R1 = steric groups > electron-donating >
electron-withdrawing. steric groups =
-tBu > -iPr > -Et > -Me
R2
B
P1
Inhibition Activities
R2 = electron-donating substituent > di-substitution > electron
withdrawing; electron-donating substituent = 4-OMe > 4-Me > 4-Et
> 4-Pr > 2-OMe > 3-OMe . substituent at the 4 position enhanced its
potency. the chloro or methoxyl group introduction improved anti-
proliferation activities.
OCH3
P2
µ
M
IC50 values 0.64 0.7
Cytotoxic Activities
IC50 values 0.83-1.01
µ
M
Figure 3. Structure–activity relationships of three series of compounds.
protein was presented in Figure 1D. The overall structural features
derived from the docking simulations indicated that the improved
activity of 13f was attributed to the occupation of P1 and P2 com-
pared to 1. The introduced tertiary butyl in the ring A inserted into
the hydrophobic sub-pocket P1 and the methoxyl group in the ring
B occupied the pocket P2. In addition, the hydroxyl group of 13f
donated a hydrogen bond to Gly135. This hydrogen bond seemed
to play a role of anchoring 13f at the ATP-binding site, which indi-
cated a key structural element for high affinity to Hsp90 (compared
13f with 8a and 8b, Table 2). These predicted interaction explained
the enhanced activity of 13f and related analogs.
gram of State Key Laboratory of Natural Medicines, China Pharma-
ceutical University (No. JKGQ201103), 2008ZX09401-001 and
2009ZX09501-003 of National Major Science and Technology Pro-
ject of China (Innovation and Development of New Drugs). The
authors declare no other conflicts of interest.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2014.01.
070.
In conclusion, a new series of potent inhibitors that were effec-
tive against Hsp90 have been developed by the structure-based
design. Through the systematic exploration of the functional
groups in the ring A and B positions of hit compound 1, we success-
fully developed the SAR profile for these compounds (summarized
in Fig. 3). The compound 13f with the most potent Hsp90
inhibition and anti-proliferation activities was a promising lead
for further structural modifications, guided by the valuable
information obtained from our SAR.
References and notes
1. Blagg, B. S.; Kerr, T. D. Med. Res. Rev. 2006, 26, 310.
2. Sharp, S.; Workman, P. Adv. Cancer Res. 2006, 95, 323.
3. Neckers, L.; Ivy, S. P. Curr. Opin. Oncol. 2003, 15, 419.
4. Fortugno, P.; Beltrami, E.; Plescia, J.; Fontana, J.; Pradhan, D.; Marchisio, P. C.;
Sessa, W. C.; Altieri, D. C. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 13791.
5. Chakraborty, A.; Koldobskiy, M. A.; Sixt, K. M.; Juluri, K. R.; Mustafa, A. K.;
Snowman, A. M.; van Rossum, D. B.; Patterson, R. L.; Snyder, S. H. Proc. Natl.
Acad. Sci. U.S.A. 2008, 105, 1134.
6. Hanahan, D.; Weinberg, R. A. Cell 2000, 100, 57.
7. Kamal, A.; Thao, L.; Sensintaffar, J.; Zhang, L.; Boehm, M. F.; Fritz, L. C.; Burrows,
F. J. Nature 2003, 425, 407.
Acknowledgments
8. Workman, P. Trends Mol. Med. 2004, 10, 47.
9. Jia, J.; Xu, X.; Liu, F.; Guo, X.; Zhang, M.; Lu, M.; Xu, L.; Wei, J.; Zhu, J.; Zhang, S.;
Zhang, S.; Sun, H.; You, Q. PLoS One 2013, 8, e59315.
This work is supported by the project 81230078 (key program),
81202463 (youth foundation) and 91129732 of National Natural
Science Foundation of China, 863 of the State Key Program, Pro-
10. Liu, K. G.; Robichaud, A. J. Tetrahedron Lett. 2005, 46, 7921.