Fluoropyrimidin-2,4-dihydroxy-5-isopropylbenzamides as antitumor agents against CRC and NSCLC cancer cells

Article abstract of DOI:10.1016/j.ejmech.2020.112540

A major cause of failure of therapy in patients with non-small cell lung cancer (NSCLC) is development of acquired drug resistance leading to tumor recurrence and disease progression. In addition to the development of new generations of epidermal growth f

Full text of DOI:10.1016/j.ejmech.2020.112540

European Journal of Medicinal Chemistry 203 (2020) 112540
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Fluoropyrimidin-2,4-dihydroxy-5-isopropylbenzamides as antitumor
agents against CRC and NSCLC cancer cells
,
,
b
,
f
,
,
Wei-Cheng Wu a 1, Yi-Min Liu a
1, Yu-Hsuan Liao d, Kai-Cheng Hsu b e, Ssu-Ting Lien e,
I-Chung Chen a, Mei-Jung Lai b, Yu-Hsuan Li a, Shiow-Lin Pan a
,
, b, e
Mei-Chuan Chen a
*, Jing-Ping Liou a
,
c
,
d,
, b, **
a School of Pharmacy, College of Pharmacy, Taipei Medical University, Taiwan
b TMU Biomedical Commercialization Center, Taipei Medical University, Taiwan
c Traditional Herbal Medicine Research Center of Taipei Medical University Hospital, Taipei, Taiwan
d Ph.D. Program in Clinical Drug Development of Herbal Medicine, College of Pharmacy, Taipei Medical University, Taiwan
e Graduate Institute of Cancer Biology and Drug Discovery, College of Medical Science and Technology, Taipei Medical University, Taipei, Taiwan
f Faculty of Pharmaceutical Sciences, Hokkaido University, Sapporo, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A major cause of failure of therapy in patients with non-small cell lung cancer (NSCLC) is development of
acquired drug resistance leading to tumor recurrence and disease progression. In addition to the devel-
opment of new generations of epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs),
different molecular targets may provide opportunities to improve the therapeutic outcomes. In this study,
we utilized the core structure 5-fluorouracil (5-FU) or tegafur, a 5-FU prodrug combined through different
linkers with resorcinol to generate a series of fluoropyrimidin-2,4-dihydroxy-5-isopropylbenzamides
which inhibit potent Heat Shock Protein 90 (HSP90). These compounds were found to show significant
antiproliferative activity in colorectal cancer (CRC) HCT116 and NSCLC A549, H460, and H1975 (EGFR
L858R/T790 M double mutation) cells. Compound 12c, developed by molecular docking analysis and
enzymatic assays exhibits promising inhibitory activity of HSP90. This compound, 12c shows the most
potent HSP90 inhibitoryactivity with an IC50 value of 27.8 4.4 nM, superior tothatof reference compounds
AUY-922 (Luminespib) and BIIB021 whose IC50 values are 43.0 0.9 nM and 56.8 4.0 nM respectively. This
strong HSP90 inhibitory activity of 12c leads to rapid degradation of client proteins EGFR and Akt in NSCLC
cells. In addition, 12c induces significant accumulation of a sub-G1 phase population in parallel with
apoptosis by showing activated caspase-3, -8 and -9 and PARP induction. These results provide a new
strategy for development of novel HSP90 inhibitors for cancer treatment.
Received 28 February 2020
Received in revised form
26 May 2020
Accepted 4 June 2020
Available online 16 July 2020
Keywords:
Non-small cell lung cancer (NSCLC)
Colorectal cancer (CRC)
HSP90 inhibitors
© 2020 Elsevier Masson SAS. All rights reserved.
1. Introduction
and are classified into different families based on their molecular
weight [2]. Among HSP families, the role of HSP90 is the most
Heat shock proteins (HSP) function as molecular chaperones
that are responsible for proper folding, assembly, translocation, and
degradation of proteins during essential cellular growth and
development [1]. HSPs are widely expressed in plants and animals
characterized and well-studied. HSP90 is physically associated with
numerous co-chaperones such as HSP70, to recruit and interact
with diverse substrate or client proteins such as kinases, leading to
regulation of cellular processes [3]. In addition, HSP90 has been
widely reported to be involved in many human diseases, including
neurodegenerative diseases, senescence, cancer and infectious
diseases [4]. A number of HSP90 client proteins such as Akt, focal
adhesion kinase (FAK) and transcription factor hypoxia-inducible
*Corresponding author. Ph.D. Program in Clinical Drug Development of Herbal
Medicine, College of Pharmacy, Taipei Medical University, Taiwan.
** Corresponding author. School of Pharmacy, College of Pharmacy, Taipei Medical
University, Taiwan.
factor 1a (HIF1a) are involved in tumor proliferation and metas-
tasis [5]. Increased expression of HSP90 is easily detected in many
malignancies and is associated with poor prognosis [6]. Conse-
quently, targeting of HSP90 can be effective for cancer treatment
E-mail addresses: mcchen1250@tmu.edu.tw (M.-C. Chen), jpl@tmu.edu.tw
(J.-P. Liou).
1
Contributed equally to this work.
https://doi.org/10.1016/j.ejmech.2020.112540
0223-5234/© 2020 Elsevier Masson SAS. All rights reserved.

2
W.-C. Wu et al. / European Journal of Medicinal Chemistry 203 (2020) 112540
since HSP90 inhibition will result in interruption of many crucial
signaling pathways that are crucial to cancer cell survival. Several
HSP90 inhibitors have been developed and are divided into
different subgroups based on their structures. For example,
resorcinol-based HSP90 inhibitors can be found in a number of
clinical trials subjects such as AUY-922 (1, Luminespib), STA-9090
(2, Ganetespib) and AT-13387 (3, Onalespib) (Fig. 1) [7]. These
resorcinol-containing compounds have been reported to be active
against NSCLC, CRC and other different types of tumors [8].
5-Fluorouracil (4, 5-FU), an analogue of uracil (5) with a fluorine
atom at the C-5 position, is converted to several active metabolites
to further replace the normal uracil and interrupt the nucleotide
synthesis which results in DNA damage in cancer cells [9].
Currently, FOLFIRI a combination of folinic acid (8), 5-FU and iri-
notecan (6) and FOLFOX (folinic acid, fluorouracil, oxaliplatin (7))
are approved for clinical CRC regimens [10]. Tegafur (9) and cape-
citabine (10) are prodrugs of 5-FU which are converted intracellu-
larly to active metabolites leading to potent cytotoxicity [11].
Previous reports have suggested that combination therapy of
standard regimen FOLFIRI with cetuximab or bevacizumab may
improve outcomes in patients with metastatic colorectal cancer
[12] and several studies have shown that administration to NSCLC
patients of tegafur or capecitabine in combination with other drugs
helps to increase patient survival rates [13]. These results suggest
one-compound-multi-target agents could be effective and they
represent a promising strategy for CRC and NSCLC treatment.
The major clinical challenge in treatment of NSCLC is the
development of acquired resistance to drugs [14]. The best study of
acquired resistance to epidermal growth factor receptor (EGFR)-TKI
is a secondary mutation (T790 M) in exon 20 of EGFR in patients
with NSCLC [15]. In order to overcome the resistance to EGFR-TKIs,
different types of TKIs and other molecular targeted therapies have
been developed with a view to extending disease control [16]. MET
amplification, overexpression of hepatocyte growth factor (HGF)
and activation of the insulin-like growth factor 1 receptor (IGF1R)
have been identified as crucial mechanisms underlying acquired
Fig. 1. Reported chemotherapeutic agents and HSP90 inhibitors.

W.-C. Wu et al. / European Journal of Medicinal Chemistry 203 (2020) 112540
3
resistance to EGFR-TKIs [17]. Previous reports also indicate that
2.2. Biological evaluation
2.2.1. HSP90 inhibition
EGFR-TKIs, in combination with other targeted drugs such as VEGF,
MEK/ERK or HER2 have shown some potential ability to overcome
resistance mediated by different mechanisms [18].
We first evaluated the HSP90 inhibitory activity of synthetic
compounds 12a-12h and 13a-13c together with reference com-
pounds AUY-922 and BIIB021, as shown in Table 1. Compound 12a,
lacking substitution at the amide linkage, showed no HSP90
inhibitory activity. However, compounds 12b-12h and 13a-13c
with the N-substituted groups displayed good inhibitory activity
against HSP90. Most of synthetic compounds have IC50 values
against HSP90 of ~50 nM which is comparable to those of the
reference compounds. Compound 12c with an ethyl group showed
the strongest HSP90 inhibitory activity, suggesting the activity is
related to the length of substituent group. The N-substituted benzyl
group compounds 12f-12h, and 13a-13c, also displayed significant
activity and showed inhibitory activity comparable to that of the
reference compounds. Among the synthesized compounds, com-
pound 12c exhibited remarkable HSP90 inhibitory activity with an
IC50 ¼ 27.8 nM, which is about two-fold stronger than the reference
compounds.
Based on previous studies, inhibitors of 5-FU and HSP90 have
shown some success in patients with NSCLC [19]. Therefore, we
combined the structure of tegafur or 5-FU with resorcinol to exert
cytotoxic and HSP90 inhibitory activity and so decrease oncogenic
client proteins and enhance the anticancer activity for NSCLC
therapy (Fig. 2). A series of compounds (12e13) was synthesized
(Fig. 3) and their biological assays are discussed below.
2. Results and discussion
2.1. Chemistry
Tagafur (6) reacted with 4-nitrobenzyl bromide in the presence
of K2CO3, yielding compound 14 (Scheme 1). Compound 14 with a
nitro group was reduced with iron powder and ammonium chloride
in isopropanol and water to afford the amino compound (15).
Compound 15 underwent amide coupling with 1-ethyl-3-(3-
dimethylaminopropyl)carbodiimide hydrochloride (EDC∙HCl), 1-
hydroxybenzotriazole (HOBt), N-methylmorpholine (NMM) and
2,4-bis(benzyloxy)-5-isopropylbenzoic acid to produce compound
16. The protected compound (16) was debenzylated by Pd/C and
hydrogen affording compound 12a. Compounds 12b-12h and
compounds 13a-13c were prepared similarly from compound 16
and various substituted groups under basic conditions. The depro-
tection can be accomplished with hydrogen gas and palladium or
with boron trichloride, as is shown in Scheme 2.
2.2.2. In vitro cell growth inhibitory activity
In an attempt to evaluate the effect of the synthetic compounds
on growth inhibition of cancer cells, we examined synthesized
compounds 12a-12h and 13a-13c for anti-proliferative activity
against CRC HCT116 cells (Table 2). Compared to compound 4 (5-
FU), most of synthesized compounds displayed potent inhibitory
activities in CRC cells. In addition, the observed pattern of inhibi-
tory activity is consistent with the HSP90 inhibition shown in
Table 1. Compound 12c exhibited anticancer activity with a GI50
value of 10 1 nM which is 15 times more inhibitory than reference
Fig. 2. Rational design of target compounds.

4
W.-C. Wu et al. / European Journal of Medicinal Chemistry 203 (2020) 112540
Fig. 3. Structures of synthetic compounds (12e13).
Scheme 1. Synthetic route of compound 12aaaReagents and condition: (a) 4-nitrobenzyl bromide, K2CO3, DMF, rt; (b) Iron powder, NH4Cl, IPA/H2O, reflux; (c) EDC∙HCl, HOBt, NMM,
2,4-bis(benzyloxy)-5-isopropylbenzoic acid, DMF, rt; (d) H2, 10% Pd/C, MeOH, rt.
compound BIIB021.
compound may provide some selectivity toward H1975 cells with
the EGFR L858R/T790 M double mutation. Compound 12c, with the
best activity against CRC HCT116 cells, as shown in Table 2, also
demonstrated promising antiproliferative activity with GI50 values
2.2.3. In vitro cell growth inhibitory activity against human lung
cancer cell lines
of 0.07, 0.06, and 0.04
cells, respectively.
mM against A549, H460 and H1975 NSCLC
It has been reported that HSP90 inhibitors such as STA-9090 and
AUY-922 display good activity against lung cancer cells [20].
Therefore we also investigated the effect of synthetic compounds
against various human lung cancer cell lines, including two car-
rying wild-type EGFR (A549 and H460), and one EGFR L858R/
T790 M double mutations (H1975). As shown in Table 3, a majority
of our synthetic compounds have shown the ability to suppress cell
proliferation in these three human NSCLC cells independently of
their EGFR status. Compounds 13a-13c exhibit more potent cell
growth inhibitory activity than compounds 12b-12h, suggesting
that 5-FU plus resorcinol generates higher cytotoxicity in vitro than
tegafur-resorcinol. It is surprising that the H1975 cell line is the
most sensitive to many of the compounds among 12c-12h, sug-
gesting that if R1 is benzyl, cyanobenzyl or chlorobenzyl the
2.2.4. Colony formation in evaluation of antitumor activity in
different cancer cells
We examined the colony forming ability of compound 12c in
cells of three different cancers. Although compound 12c exhibits
better antiproliferative activity in CRC HCT116 cells than other tu-
mor cells (Tables 2 and 3), the colony-forming results showed a
better inhibitory effect toward NSCLC A549 and H1975 cells with
the same concentrations (0.009
mM and 0.018 mM) of 12c
(Fig. 4AeC). Taken together, these results show compound 12c
caused a significant inhibition of cell growth and colony-forming
ability in NSCLC cells, suggesting 12c may provide potentially

W.-C. Wu et al. / European Journal of Medicinal Chemistry 203 (2020) 112540
5
Scheme 2. Synthetic route of compounds 12b-12h and 13a-13ca.
aReagents and condition: (a) i. NaH, various substituted group, DMF, rt; ii. H2, 10% Pd/C, MeOH, rt, (b) i. NaH, various substituted group, DMF, rt; ii. BCl3, CH2Cl2, 0 ꢀC to rt.
Table 1
greatly induced the expression level of HSP70 without affecting
HSP90 levels. Furthermore, several client proteins such as protein
kinase B (akt) and EGFR were also downregulated in response to
compound 12c, indicating compound 12c achieves significant
suppression of HSP90 activity in cells.
HSP90a inhibition of synthetic compounds (12a-12h and 13a-13c) and reference
compounds.
Compounds HSP90a a
IC50(nM SDa) Compounds HSP90 IC50(nM SDa)
12a
12b
12c
12d
12e
12f
>1000
12h
13a
13b
13c
58.8 6.9
44.9 0.8
48.0 1.5
54.4 0.6
40.9 19.3
27.8 4.4
48.2 13.8
45.3 5.0
45.3 7.0
54.7 16.3
2.2.6. Evaluation of cell cycle progression and cell death response
We examined the effects of compound 12c on cell cycle pro-
gression using flow cytometry. Notably, the subG1 phase showed a
dramatic induction after 48 h of treatment (Fig. 6A and B) in parallel
with significant activation of caspase-3, -8, and -9, and poly-(ADP-
ribose) polymerase (PARP) (Fig. 7A and B). These results suggest that
compound 12c induces significant apoptotic cell death in NSCLC.
AUY-922 (1) 43.0 0.9
BIIB021 56.8 4.0
12g
a
SD: standard deviation. All experiments were independently performed at least
three times.
2.2.7. Molecular docking study
useful antitumor activity for NSCLC treatment.
Molecular docking analysis was performed to illuminate the
interactions between 12c and HSP90 (PDB ID: 5GGZ). The docking
pose of 12c in HSP90, can be separated into four distinct sites
(Fig. 8). In Site 1 (S1), the resorcinol group of 12c forms hydrogen
bonds with residues Leu48, Asp93, and Thr184, and hydrophobic
interactions occur between its isopropyl moiety and residues Leu48
and Phe138, while the benzene ring creates hydrophobic in-
teractions with residues Ala55 and Val186. Site 2 (S2) contains an
2.2.5. Evaluation of expression level of Hsp90 client proteins
Next, we examined some critical biomarkers of Hsp90 inhibition
in A549 and H1975 cells. Inhibition of HSP90 will lead to heat shock
factor 1 (HSF1) activation, consequently inducing several small heat
shock proteins to help appropriate folding of misfolded proteins
[21]. As shown in Fig. 5A and B, treatment with compound 12c
Table 2
Antiproliferative activity (GI50) of synthetic compounds (12a-12h and 13a-13c) against colorectal HCT116 cancer cell line.
Compounds
HCT116 GI50
(m
M
SDa)
Compounds
HCT116 GI50 (mM
SDa)
12a
12b
12c
12d
12e
12f
>10
13a
13b
13c
STA-9090 (2)
5-FU (4)
Tegafur (9)
BIIB021
0.030 0.001
0.080 0.001
0.040 0.006
0.150 0.020
>10
0.120 0.010
0.010 0.001
0.960 0.210
0.780 0.002
0.410 0.050
0.700 0.110
0.720 0.011
>10
0.150 0.020
12g
12h
a
SD: standard deviation. All experiments were independently performed at least three times.

6
W.-C. Wu et al. / European Journal of Medicinal Chemistry 203 (2020) 112540
Table 3
Antiproliferative activity (GI50) of compounds (12a-12h and 13a-13c) against lung A549, H460 and H1975 cancer cell line.
Compounds
A549 GI50
(m
M
SDa)
H460 GI50
(
mM
SDa)
H1975 GI50
(
m
M
SDa)
12a
>10
>10
>10
12b
12c
12d
12e
12f
12g
12h
13a
13b
13c
AUY922 (1)
5-FU (4)
Tegafur (9)
BIIB021
0.09 0.01
0.07 0.01
0.65 0.03
0.80 0.07
0.35 0.04
0.56 0.03
0.92 0.11
0.04 0.00
0.10 0.01
0.03 0.00
0.01 0.00
19.76 2.9
>50
0.07 0.03
0.06 0.01
0.52 0.12
0.40 0.19
0.20 0.08
0.46 0.08
0.39 0.16
0.03 0.01
0.07 0.01
0.07 0.02
0.03 0.01
e
0.06 0.00
0.04 0.00
0.34 0.03
0.16 0.04
0.13 0.01
0.16 0.03
0.23 0.05
0.03 0.00
0.05 0.02
0.02 0.00
0.02 0.00
e
e
e
0.26 0.04
0.21 0.10
0.20 0.06
a
SD: standard deviation. All experiments were independently performed at least three times.
N-ethyl moiety that occupies a hydrophobic pocket formed by
residues Ala55, Met98 and Tyr61. The phenyl group at Site 3 (S3)
occupies a cavity within the binding site and serves as a linker
between the ethyl and the tegafur moiety at Site 4 (S4). The tegafur
moiety forms hydrogen bonds with residues Asn51 and Phe138.
When comparing 12c with its derivatives, 12c has an increased
HSP90 inhibitory effect over that of its derivatives, such as unsub-
stituted 12a and 12b with a methyl group in place of the ethyl
moiety. These differences reduce their inhibitory effect against
HSP90 (Table 1). Consequently, it could be concluded that the ethyl
moiety in 12c is important for its activity against HSP90.
3. Conclusion
In this study,
a
series of fluoropyrimidin-2,4-dihydroxy-5-
isopropylbenzamides (12e13) were synthesized as HSP90 inhibitors
in an attempt to explore the compounds’ biological activity against
CRC and NSCLC. Among all the synthesized compounds, compound
12c displays remarkable activity with an IC50 value of 27.8 nM in in-
hibition of HSP90 activity. In in vitro cell growth inhibition, 12c not
only shows good activity against HCT116 cancer cell with a GI50 value
of 0.01
mMbutalsoexhibits goodactivitywithGI50 values of0.07, 0.06,
and 0.04
mM against A549, H460 and H1975 human NSCLC cell lines,
To further elucidate interactions between compound 12c and
HSP90, we compared the compound to the co-crystal ligand, 6 TN.
We observed an overlap at Site 1 and Site 2 (Fig. 8C). This co-crystal
ligand also contains a resorcinol group at Site 1. Hydrogen bonds
form with residues Asp93 and Thr184 at Site 1. Residues Thr184 and
Ala55 form hydrophobic interactions with the benzene ring. An
isopropyl moiety is attached to the benzene ring and forms further
hydrophobic interactions with residues Phe138 and Val150. The co-
crystal ligand also contains a pyridopyrimidine structure. A part of
this moiety occupies Site 2, forming hydrophobic interactions with
residues Ala55, Met98 and Leu107. Because of its position, the co-
crystal ligand does not align with Site 3 and Site 4. The reported
IC50 value of co-crystal ligand is 56 nM [22]. When compared to
compound 12c, the difference in potency may be due to its in-
teractions at Site 3 and Site 4.
We previously found that compound 12c has a comparable IC50
value to the HSP90 inhibitor AUY-922 (Table 1). A co-crystal
structure of HSP90 with AUY-922 has been determined previ-
ously (PDB ID: 2VCI) [23]. As a result, we compared compound 12c
to the AUY-922 co-crystal ligand. We found that both compounds
align with the four distinct sites (Fig. 8D). AUY-922 also contains a
resorcinol and isopropyl moiety that occupies Site 1, similar to
compound 12c. Hydrogen bonds are formed with residues Asn51,
Asp93 and Thr184. The isopropyl moiety occupies a similar hy-
drophobic pocket by residues Leu107 and Phe138. At Site 2, resi-
dues Lys58 and Gly97 form hydrogen bonds to the oxygen and
nitrogen of the N-ethylformamide, respectively. Further hydro-
phobic interactions by residue Ala55 occur at Site 3. However, Site 4
did not show stable interactions between AUY-922 and HSP90. The
additional hydrogen bond at Site 3 and hydrophobic interactions at
Site 4 suggests areas that can be exploited for HSP90 inhibition.
Together, the molecular docking analysis is consistent with the
potency of 12c against HSP90.
respectively. In addition, compound 12c significantly reduces the
client proteins EGFR and Akt and induces apoptotic cell death in
NSCLC cells. In this study, we have developed 12c as a potential agent
and a new strategy for development of novel HSP90 inhibitors for the
treatment of cancer.
4. Experimental section
4.1. Chemistry
Nuclear magnetic resonance (1H and 13C NMR) spectra were
obtained with Bruker DRX-500 spectrometer operating at 500 and
125 MHz and Bruker Fourier 300 and 75 MHz. Chemical shifts are
reported in parts per million (ppm, d) downfield from TMS as an
internal standard. High-resolution mass spectra (HRMS) were
measured with AB SCIE X (QSTAR® XL) High Resolution Electro-
spray (ESI) Mass Spectrometry spectrometer. Melting points were
measured with Buchi B-545 (Buchi, Switzerland). Purity of the final
compounds was achieved with a Waters Acquity UPLC system us-
ing C-18 column (Waters Acquity UPLC BEH C18, 1.7
mm,
2.1 mm ꢁ 50 mm). Flash column chromatography used silica gel:
SILICYCLE (SiliaFlash Irregular Silica Gel P60, 40e63
mm, 60 Å
(R12030B).
4.1.1. N-(4-((5-Fluoro-2,6-dioxo-3-(tetrahydrofuran-2-yl)-3,6-
dihydropyrimidin-1(2H)-yl)methyl)phenyl)-2,4-dihydroxy-5-
isopropylbenzamide (12a)
A mixture of 16 (2.05 mmol), 10% palladium on carbon (0.4 g) in
MeOH (40 ml) was stirred at room temperature (rt) under hydrogen
overnight. The organic layer was filtered and the residue was pu-
rified by flash chromatography over silica gel to afford compound
12a in 46% yield. 1H NMR (500Hz, DMSO‑d6)
d (ppm): 1.17 (d,
J ¼ 6.9Hz, 6H), 1.90e1.93 (m, 2H), 2.02e2.04 (m, 1H), 2.21e2.28 (m,
1H), 3.07e3.13 (m, 1H), 3.81 (q, J ¼ 7.5Hz, 1H), 4.22e4.26 (m, 1H),

W.-C. Wu et al. / European Journal of Medicinal Chemistry 203 (2020) 112540
7
Fig. 4. Compound 12c suppresses colony formation in colorectal (CRC) and non-small cell lung cancer (NSCLC) cells. (AeC) Compound 12c-mediated concentration-dependent
inhibition of colony formation in three different tumor cells lines (left panel) and quantification of colony numbers, expressed as a percentage of controls (right panel). Data
are expressed as means S.D. of at least three independent experiments. *, P < 0.05; **, P < 0.01; and ***, P < 0.001 compared with the control group.
4.95 (d, J ¼ 4.5Hz, 2H), 5.95e5.97 (m, 1H), 6.35 (s, 1H), 7.29 (d,
J ¼ 8.5Hz, 2H), 7.55 (d, J ¼ 8.5Hz, 2H), 7.83 (s, 1H), 7.97 (d, J ¼ 7Hz,
1H), 10.09 (s, 1H), 10.16 (s, 1H), 12.03 (s, 1H). 13C NMR (125 MHz,
138.51, 140.33, 148.84, 156.51, 156.71, 159.13, 159.70, 167.22.
mp ¼ 222.6e223.2 ꢀC. HRMS (ESI) for C25H27FN3O6 [MþH]þ: Calcd,
484.1878; Found, 484.1884.
DMSO‑d6)
d (ppm): 22.64, 23.47, 26.15, 31.65, 43.69, 69.56, 87.47,
102.57, 107.37, 121.34, 123.83, 124.10, 126.43, 128.25, 132.00, 137.43,

8
W.-C. Wu et al. / European Journal of Medicinal Chemistry 203 (2020) 112540
Fig. 5. Compound 12c reduces HSP90 client proteins. (AeB) Compound 12c suppresses protein expression level of HSP90 client proteins in NSCLC A549 (A) and H1975 (B) cells. Cells
were treated with indicated concentrations for 24 h, and cell lysates were immunoblotted using the indicated antibodies.
Fig. 6. Compound 12c induces subG1 phase accumulation in NSCLC cells. (AeB) Concentration-dependent effects of Compound 12c on subG1 phase accumulation in A549 (A) and
H1975 (B) cells. Cells were treated with DMSO or Compound 12c with indicated concentrations for the indicated times, and the cell cycle distribution was analyzed by flow
cytometry. The right panel are quantitative data based on flow cytometry histograms, and are presented as means S.D. of at least three independent experiments that yielded
similar results.

W.-C. Wu et al. / European Journal of Medicinal Chemistry 203 (2020) 112540
9
Fig. 7. Compound 12c induces apoptosis in NSCLC. (AeB) Effects of compound 12c on apoptosis in NSCLC cells. compound 12c increased levels of the cleaved (activated) forms of
PARP, H2AX, caspase-3, -8, and -9 in a concentration-dependent manner. Cells were exposed to compound 12c with indicated concentrations for 48 h, and cell lysates were
g
immunoblotted using the indicated antibodies.
4.1.2. N-(4-((5-Fluoro-2,6-dioxo-3-(tetrahydrofuran-2-yl)-3,6-
dihydropyrimidin-1(2H)-yl)methyl)phenyl)-2,4-dihydroxy-5-
isopropyl-N-methylbenzamide (12b)
J ¼ 3.1Hz, 3H), 0.61 (d, J ¼ 3Hz, 3H), 1.06 (t, J ¼ 7Hz, 3H), 1.89e1.92
(m, 2H), 1.97e2.00 (m, 1H), 2.20e2.28 (m, 1H), 2.72e2.78 (m, 1H),
3.76e3.83 (m, 3H), 4.23e4.27 (m, 1H), 4.92 (q, J ¼ 14.5Hz, 2H),
5.92e5.94 (m,1H), 6.17 (s,1H), 6.42 (s,1H), 7.12 (d, J ¼ 8Hz, 2H), 7.32
(d, J ¼ 8.5Hz, 2H), 7.95 (d, J ¼ 6.5Hz, 2H), 9.76 (s, 1H), 10.93 (s, 1H).
A
mixture of 16 (2.10mmole), NaH (2.52mmole), MeI
(2.52mmole) and DMF (3 ml) was stirred at rt for 2 h. The reaction
was quenched with water and extracted by EtOAc. The residue was
purified by flash chromatography over silica gel to afford an inter-
mediate. Then a mixture of this intermediate, 10% palladium on
carbon (0.4 g) in MeOH (40 ml) was stirred at rt under hydrogen
overnight. The organic layer was filtered and the residue was pu-
rified by flash chromatography over silica gel to afford compound
13C NMR (125 MHz, DMSO‑d6)
d (ppm): 12.52, 22.02, 23.44, 24.91,
31.67, 43.49, 45.13, 69.58, 87.44, 102.25, 123.75, 124.01, 124.54,
127.43, 127.75, 129.36, 134.81, 138.54, 140.36, 142.81, 148.77, 156.35,
156.56, 157.63, 158.44, 169.98. mp ¼ 173.4e174.6 ꢀC. HRMS (ESI) for
C
27H31FN3O6 [MþH]þ: Calcd, 512.2191; Found, 512.2195.
12b in 42% yield. 1H NMR (500 MHz, DMSO‑d6)
d (ppm): 0.63 (d,
J ¼ 3Hz, 3H), 0.65 (d, J ¼ 3Hz, 3H), 1.89e1.96 (m, 2H), 1.97e2.01 (m,
1H), 2.20e2.26 (m, 1H), 2.75e2.80 (m, 1H), 3.29 (s, 3H), 3.80 (q,
J ¼ 7.5Hz, 1H), 4.24 (q, J ¼ 6Hz, 1H), 4.91 (q, J ¼ 14.5Hz, 2H),
5.92e5.94 (m, 1H), 6.18 (s, 1H), 6.45 (s, 1H), 7.13 (d, J ¼ 8.5Hz, 2H),
7.29 (d, J ¼ 8Hz, 2H), 7.95 (d, J ¼ 6.5Hz, 2H), 10.72 (s, 1H). 13C NMR
4.1.4. N-(4-((5-Fluoro-2,6-dioxo-3-(tetrahydrofuran-2-yl)-3,6-
dihydropyrimidin-1(2H)-yl)methyl)phenyl)-2,4-dihydroxy-5-
isopropyl-N-propylbenzamide (12d)
The title compound was obtained as a solid in 42% yield from
compound 16 in a manner similar to that described for the prep-
(125 MHz, DMSO‑d6)
d (ppm): 22.07, 23.45, 24.96, 31.67, 38.17,
aration of 12b. 1H NMR (500 MHz, DMSO‑d6)
d (ppm): 0.60 (d,
43.49, 69.59, 87.45, 102.27, 109.06, 123.76, 124.03, 124.63, 126.50,
127.72, 129.22, 134.58, 138.54, 140.35, 144.56, 148.77, 156.37, 156.57,
157.55, 157.85, 170.29. mp ¼ 199.3e199.9 ꢀC. HRMS (ESI) for
J ¼ 3.4Hz, 3H), 0.62 (d, J ¼ 3.4Hz, 3H), 0.82 (t, J ¼ 7Hz, 3H),1.46e1.50
(m, 2H), 1.89e1.92 (m, 2H), 1.99e2.00 (m, 1H), 2.22e2.26 (m, 1H),
2.49e2.77 (m, 1H), 3.71 (t, J ¼ 7.5H, 2H), 3.80 (q, J ¼ 7.5Hz, 1H),
4.23e4.25 (m, 1H), 4.91 (q, J ¼ 14.5Hz, 1H), 5.92e5.94 (m, 1H), 6.17
(s, 1H), 6.41 (s, 1H), 7.12 (d, J ¼ 8Hz, 2H), 7.30 (d, J ¼ 8.5Hz, 2H), 7.95
(d, J ¼ 6.5Hz, 2H), 10.82 (s, 1H). 13C NMR (125 MHz, DMSO‑d6)
C
26H29FN3O6 [MþH]þ: Calcd, 498.2035; Found, 498.2041.
4.1.3. N-Ethyl-N-(4-((5-fluoro-2,6-dioxo-3-(tetrahydrofuran-2-yl)-
3,6-dihydropyrimidin-1(2H)-yl)methyl)phenyl)-2,4-dihydroxy-5-
isopropylbenzamide (12c)
d
(ppm): 11.13, 14.08, 20.29, 22.05, 23.46, 24.93, 31.69, 43.50, 51.50,
69.60, 87.46, 102.27, 108.94, 123.76, 124.03, 124.56, 127.30, 127.66,
129.30, 134.73, 138.55, 140.37, 143.00, 148.79, 156.37, 156.58, 157.56,
158.20, 170.11. mp ¼ 102.6e103.5 ꢀC. HRMS (ESI) for C28H33FN3O6
[MþH]þ: Calcd, 526.2348; Found, 526.2354.
The title compound was obtained as a solid in 33% yield from
compound 16 in a manner similar to that described for the prep-
aration of 12b. 1H NMR (500 MHz, DMSO‑d6)
d (ppm): 0.60 (d,

10
W.-C. Wu et al. / European Journal of Medicinal Chemistry 203 (2020) 112540
Fig. 8. Docking pose of compound 12c in HSP90. (A) 12c (blue) occupies the HSP90 (gray, PDB ID: 5GGZ) binding site. (B) A 2D representation of 12c docked in HSP90. Hydrogen
bonds are denoted by dashed lines, green lines represent areas of hydrophobic interactions. Interacting residues are labeled as shown. (C) The docking pose of 12c superimposed
with the co-crystal ligand, 6 TN. Overlapping sections are highlighted as previously described. (D) The docking pose of 12c superimposed with the co-crystal ligand AUY-922
(purple, PDB ID: 2VCI). Interacting residues are labeled and represented as sticks. Hydrogen bonds are denoted by dotted green lines. The four distinct sites of 12c are colored
as purple, red, yellow and blue, which are located at sections S1, S2, S3 and S4, respectively. (For interpretation of the references to color in this figure legend, the reader is referred
to the Web version of this article.)
4.1.5. N-Benzyl-N-(4-((5-fluoro-2,6-dioxo-3-(tetrahydrofuran-2-
yl)-3,6-dihydropyrimidin-1(2H)-yl)methyl)phenyl)-2,4-dihydroxy-
5-isopropylbenzamide (12e)
4.1.6. N-(4-Cyanobenzyl)-N-(4-((5-fluoro-2,6-dioxo-3-
(tetrahydrofuran-2-yl)-3,6-dihydropyrimidin-1(2H)-yl)methyl)
phenyl)-2,4-dihydroxy-5-isopropylbenzamide (12f)
The title compound was obtained as a solid in 32% yield from
compound 16 in a manner similar to that described for the prep-
The title compound was obtained as a solid in 35% yield from
compound 16 in a manner similar to that described for the prep-
aration of 12b. 1H NMR (500 MHz, DMSO‑d6)
d
(ppm): 0.64 (d,
aration of 12b. 1H NMR (500 MHz, DMSO‑d6)
d (ppm): 0.67 (d,
J ¼ 3Hz, 3H), 0.65 (d, J ¼ 3.5Hz, 3H), 1.88e1.92 (m, 2H), 1.95e1.98
(m, 1H), 2.18e2.24 (m, 1H), 2.75e2.80 (m, 1H), 3.77e3.82 (m, 1H),
4.21e4.25 (m, 1H), 4.85 (q, J ¼ 14Hz, 2H), 5.03 (s, 1H), 5.90e5.92 (m,
1H), 6.20 (s, 1H), 6.51 (s, 1H), 7.05 (d, J ¼ 8.5Hz, 2H), 7.20e7.21 (m,
3H), 7.26e7.29 (m, 4H), 7.93 (d, J ¼ 6.5Hz, 1H), 9.74 (s, 1H), 10.61 (s,
J ¼ 3Hz, 3H), 0.68 (d, J ¼ 3Hz, 3H), 1.88e1.92 (m, 2H), 1.97e1.98 (m,
2H), 2.18e2.24 (m, 1H), 2.76e2.80 (m, 1H), 3.77e3.82 (m, 1H),
4.21e4.25 (m, 1H), 4.86 (q, J ¼ 14.5Hz, 2H), 5.11 (s, 1H), 5.90e5.91
(m, 1H), 6.19 (s, 1H), 6.55 (s, 1H), 7.09 (d, J ¼ 8.5Hz, 2H), 7.20 (d,
J ¼ 8.5Hz, 2H), 7.49 (d, J ¼ 8Hz, 2H), 7.75 (d, J ¼ 8.5Hz, 2H), 7.93 (d,
J ¼ 7Hz,1H), 9.74 (s,1H),10.45 (s,1H). 13C NMR (125 MHz, DMSO‑d6)
1H). 13C NMR (125 MHz, DMSO‑d6)
d (ppm): 22.55, 23.91, 25.48,
32.15, 43.93, 53.44, 60.22, 70.05, 87.92, 102.78, 124.21, 124.48,
125.21, 127.45, 127.47, 128.10, 128.79, 129.44, 134.97, 137.93, 139.00,
140.81, 143.36, 149.22, 156.82, 157.02, 158.09, 158.18, 170.84.
mp ¼ 99.8e100.7 ꢀC. HRMS (ESI) for C32H33FN3O6 [MþH]þ: Calcd,
574.2348; Found, 574.2353.
d (ppm): 22.09, 23.42, 25.04, 31.65, 43.43, 52.79, 69.57, 87.43,
102.28, 109.81, 118.75, 123.74, 124.01, 124.83, 126.85, 127.56, 128.43,
128.92, 132.29, 134.61, 138.50, 140.32, 142.66, 143.47, 148.74, 156.34,
156.54, 157.27, 157.59, 170.38. mp ¼ 117.5e118.4 ꢀC. HRMS (ESI) for
C
33H32FN4O6 [MþH]þ: Calcd, 599.2300; Found, 599.2307.

W.-C. Wu et al. / European Journal of Medicinal Chemistry 203 (2020) 112540
11
4.1.7. N-(4-Chlorobenzyl)-N-(4-((5-fluoro-2,6-dioxo-3-
(tetrahydrofuran-2-yl)-3,6-dihydropyrimidin-1(2H)-yl)methyl)
phenyl)-2,4-dihydroxy-5-isopropylbenzamide (12g)
J ¼ 6.6Hz, 6H), 3.70 (q, J ¼ 7.2Hz, 2H), 4.08 (d, J ¼ 5.1Hz, 2H), 4.58 (s,
2H), 4.76e4.85 (m, 2H), 5.49e5.62 (m, 1H), 5.89 (s, 1H), 6.18 (s, 1H),
6.79 (d, J ¼ 8.1Hz, 2H), 6.96 (d, J ¼ 8.1Hz, 2H), 7.54 (t, J ¼ 5.7Hz, 1H),
The title compound was obtained as a solid in 34% yield from
compound 16 in a manner similar to that described for the prep-
9.47 (s, 1H), 10.41 (s, 1H). 13C NMR (125 MHz, DMSO‑d6)
d (ppm):
22.56, 25.46, 43.35, 53.09, 60.23, 102.74, 109.49, 117.84, 125.18,
125.52, 125.77, 127.54, 128.17, 129.48, 133.97, 135.34, 139.02, 140.83,
143.44, 150.28, 157.60, 157.80, 158.09, 158.42, 170.51.
mp ¼ 210.3e211.2 ꢀC. HRMS (ESI) for C24H25FN3O5 [MþH]þ: Calcd,
454.1773; Found, 454.1778.
aration of 12b 1H NMR (500 MHz, DMSO‑d6)
d (ppm): 0.64 (t,
J ¼ 2.1Hz, 6H), 1.89e1.94 (m, 2H), 1.95e1.97 (m, 1H), 2.20e2.24 (m,
1H), 3.79 (q, J ¼ 7.5Hz, 1H), 4.83 (d, J ¼ 14.5Hz, 1H), 4.88 (d,
J ¼ 14.5Hz, 1H), 5.01e5.03 (m, 2H), 5.89e5.92 (m, 1H), 6.20 (s, 1H),
6.50 (s, 1H), 7.05 (d, J ¼ 8Hz, 2H), 7.19e7.21 (m, 3H), 7.26e7.33 (m,
4H), 7.92 (d, J ¼ 6.5Hz, 1H). 13C NMR (125 MHz, DMSO‑d6)
d (ppm):
4.1.11. N-(4-((5-Fluoro-2,6-dioxo-3,6-dihydropyrimidin-1(2H)-yl)
methyl)phenyl)-2,4-dihydroxy-5-isopropyl-N-(prop-2-yn-1-yl)
benzamide (13c)
22.20, 23.48, 25.06, 31.74, 43.52, 53.05, 69.65, 87.51, 102.37, 123.63,
124.80, 127.05, 127.08, 127.68, 128.38, 129.03, 129.69, 131.39, 134.33,
136.33, 137.15, 138.56, 140.13, 143.21, 148.81, 156.11, 157.84, 170.44.
mp ¼ 99.8e101.1C. HRMS (ESI) for C32H32ClFN3O6 [MþH]þ: Calcd,
608.1958; Found, 608.1962.
The title compound was obtained as a solid in 47% yield from
compound 16 in a manner similar to that described for the prep-
aration of 13a. 1H NMR (300 MHz, DMSO‑d6)
d (ppm): 0.67 (d,
J ¼ 3.9 Hz, 6H), 2.62e2.81 (m, 1H), 3.12 (s, 1H), 4.55 (d, J ¼ 1.2Hz,
2H), 4.89 (s, 2H), 6.19 (s, 1H), 6.49 (s, 1H), 7.14 (d, J ¼ 4.8Hz, 2H), 7.28
(d, J ¼ 5.1Hz 2H), 7.84 (d, J ¼ 3.3Hz 1H), 10.52 (s, 1H). 13C NMR
4.1.8. N-(4-((5-Fluoro-2,6-dioxo-3-(tetrahydrofuran-2-yl)-3,6-
dihydropyrimidin-1(2H)-yl)methyl)phenyl)-2,4-dihydroxy-5-
isopropyl-N-(4-methoxybenzyl)benzamide (12h)
(125 MHz, DMSO‑d6)
d (ppm): 22.60, 25.48, 43.39, 75.05, 80.15,
The title compound was obtained as a solid in 35% yield from
compound 16 in a manner similar to that described for the prep-
102.78, 109.32, 125.60, 125.86, 127.71, 128.14, 129.42, 135.86, 139.02,
140.82, 142.71, 150.30, 157.61, 157.81, 158.21, 158.25, 170.35
mp ¼ 214.2e214.9 ꢀC. HRMS (ESI) for C24H23FN3O5 [MþH]þ: Calcd,
452.1616; Found, 452.1622.
aration of 12b. 1H NMR (500 MHz, DMSO‑d6)
d (ppm): 0.61 (d,
J ¼ 3.3Hz, 3H), 0.63 (d, J ¼ 3.3Hz, 3H), 1.89e1.94 (m, 3H), 2,.20e2.24
(m, 1H), 2.74e2.77 (m, 1H), 3.77 (s, 3H), 4.21e4.25 (m, 1H), 4.86 (q,
J ¼ 14.5, 2H), 4.95 (s, 2H), 5.90e5.91 (m, 1H), 6.19 (s, 1H), 6.47 (s,
1H), 6.81 (d, J ¼ 8.5Hz, 2H), 7.01 (d, J ¼ 8Hz, 2H), 7.16 (d, J ¼ 8.5Hz,
2H), 7.20 (d, J ¼ 8.5Hz, 2H), 7.92 (d, J ¼ 6.2Hz, 2H), 9.78 (s, 1H), 10.70
4.1.12. 5-Fluoro-3-(4-nitrobenzyl)-1-(tetrahydrofuran-2-yl)
pyrimidine-2,4(1H,3H)-dione (14)
(s, 1H). 13C NMR (125 MHz, DMSO‑d6)
d
(ppm): 22.09, 23.4, 25.02,
A solution of 5-fluoro-1-(tetrahydrofuran-2-yl)pyrimidine-
31.72, 43.52, 55.00, 59.81, 69.64, 87.50, 102.34, 113.74, 123.77,
124.04, 124.75, 127.15, 127.72, 128.31, 129.06, 129.21, 129.29, 134.55,
138.56, 142.84, 148.80, 157.63, 158.35, 170.37. mp ¼ 102.1e102.8 ꢀC.
HRMS (ESI) for C33H35FN3O7 [MþH]þ: Calcd, 604.2454; Found,
604.2466.
2,4(1H,3H)-dione (5.00 mmol), 1-(bromomethyl)-4-nitrobenzene
(6.50 mmol), K2CO3 (6.50 mmol) in DMF (10 ml) was stirred at rt
overnight. The reaction was quenched with water and extracted
with EtOAc. The residue was purified by flash chromatography over
silica gel to afford compound 15 in 70% yield. 1H NMR (300 MHz,
CDCl3)
d (ppm): 1.95 (m, 1H), 2.06 (m, 2H), 2.40 (m, 1H), 3.97 (q,
4.1.9. N-Ethyl-N-(4-((5-fluoro-2,6-dioxo-3,6-dihydropyrimidin-
1(2H)-yl)methyl)phenyl)-2,4-dihydroxy-5-isopropylbenzamide
(13a)
J ¼ 2.1 Hz, 1H), 4.22 (m, 1H), 5.18 (q, J ¼ 7.2 Hz, 2H), 5.97 (m, 1H),
7.42 (d, J ¼ 5.7 Hz,1H), 7.62 (d, J ¼ 8.7Hz, 2H), 8.16 (d, J ¼ 8.7 Hz, 2H).
A mixture of 16 (1.85 mmol), NaH (2.22 mmol), MeI (2.22 mmol)
and DMF (3 ml) was stirred at 0 ꢀC for 10 min and left at rt for 2 h.
The reaction was quenched with water and extracted with EtOAc.
The residue was purified by flash chromatography over silica gel to
afford an intermediate compound. Then a mixture of the inter-
mediate compound and boron trichloride (1 M) (17% in hexane) in
CH2Cl2 was stirred at 0 ꢀC for 3 h. The reaction was quenched with
water and extracted with CH2Cl2. The organic layer was collected
and dried over anhydrous MgSO4 and concentrated in vacuo to yield
an oily residue which was purified by flash chromatography over
silica gel to afford compound 13a in 36% yield. 1H NMR (500 MHz,
4.1.13. 3-(4-Aminobenzyl)-5-fluoro-1-(tetrahydrofuran-2-yl)
pyrimidine-2,4(1H,3H)-dione (15)
A mixture of 14 (0.69 mmol), iron powder (2.07 mmol),
ammonium chloride (1.38 mmol), water (1.4 ml) and isopropyl
alcohol (5.6 ml) was stirred reflux for 3 h. The reaction was
quenched with water and extracted with EtOAc. The residue was
purified by flash chromatography over silica gel to afford 15 in 81%
yield. 1H NMR (300 MHz CDCl3)
d (ppm): 1.98 (m, 1H), 2.02 (m, 2H),
2.38 (m,1H), 3.96, (q, J ¼ 6.6 Hz,1H), 4.20 (m,1H), 5.03 (q, J ¼ 4.2 Hz,
2H), 5.95 (m,1H), 6.94 (d, J ¼ 8.4 Hz, 2H), 7.35 (d, J ¼ 5.7 Hz,1H), 7.41
(d, J ¼ 8.4 Hz, 2H).
DMSO‑d6)
d
(ppm): 0.34 (d, J ¼ 7Hz, 6H), 0.78 (t, J ¼ 6.5 Hz, 3H), 3.51
(q, J ¼ 7Hz, 2H), 4.62 (s, 2H), 5.92 (s, 1H), 6.16 (s, 1H), 6.84 (d,
J ¼ 8.5Hz, 2H), 7.04 (d, J ¼ 8Hz, 2H), 7.58 (t, J ¼ 5.5Hz, 1H), 9.53 (s,
1H), 10.70 (s, 1H), 10.91 (d, J ¼ 6Hz, 1H). 13C NMR (125 MHz,
4.1.14. 2,4-bis(Benzyloxy)-N-(4-((5-fluoro-2,6-dioxo-3-
(tetrahydrofuran-2-yl)-3,6-dihydropyrimidin-1(2H)-yl)methyl)
phenyl)-5-isopropylbenzamide (16)
DMSO‑d6) d (ppm): 12.94, 22.52, 25.41, 43.35, 45.65, 102.73, 109.05,
125.06,125.49,125.74,127.89,128.28,129.79,135.56,139.04,140.84,
143.23, 150.28, 157.59, 157.80, 158.13, 159.01, 170.81.
mp ¼ 114.6e115.4 ꢀC. HRMS (ESI) for C23H25FN3O5 [MþH]þ: Calcd,
442.1773; Found, 442.1780.
A mixture of 2,4-bis(benzyloxy)-5-isopropylbenzoic acid
(0.90 mmol), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide
(1.13 mmol), hydroxybenzotriazole (0.90 mmol), N-methyl-
morpholine (1.13 mmol) and 15 (0.75 mmol) was stirred at rt
overnight. The reaction was quenched with water and extracted
with EtOAc. The residue was purified by flash chromatography over
silica gel to afford compound 16 in 90% yield. 1H NMR (300 MHz,
CDCl3): 1.25 (m, 6H), 1.93 (m, 1H), 2.02 (m, 1H), 2.05 (m, 1H), 2.38
(m, 1H), 3.34 (t, J ¼ 6.9 Hz, 1H), 3.95 (q, J ¼ 3.0Hz, 1H), 4.18 (m, 1H),
5.09e5.29 (m, 6H), 5,96 (m, 1H), 6.61 (m, 1H), 7.16e7.63 (m, 15H),
8.18 (s, 1H), 9.85 (s, 1H).
4.1.10. N-Allyl-N-(4-((5-fluoro-2,6-dioxo-3,6-dihydropyrimidin-
1(2H)-yl)methyl)phenyl)-2,4-dihydroxy-5-isopropylbenzamide
(13b)
The title compound was obtained as a solid in 42% yield from
compound 16 in a manner similar to that described for the prep-
aration of 13a. 1H NMR (300 MHz, DMSO‑d6)
d (ppm): 0.35 (d,

12
W.-C. Wu et al. / European Journal of Medicinal Chemistry 203 (2020) 112540
4.2. Biological activity
4.2.6. Immunoblotting and lentivirus expression system
Cells were seeded in dishes and allowed to attach overnight. The
cells were treated with drugs at indicated concentrations for indi-
cated times. After the indicated exposure time, cells were lysed and
the immunoblotting was performed as previously described [24].
4.2.1. Cell lines and reagents
A549, H1975, and HCT116 cells were obtained from the Amer-
ican Type Culture Collection (ATCC) (Manassas, VA, USA). Cells were
maintained in 10% fetal bovine serum (FBS)-supplemented RPMI
1640 medium (GIBCO, Grand Island, NY, USA) and 1%
penicillinestreptomycin (GIBCO) at 37 ꢀC in a humidified incubator
containing 5% CO2. Antibodies against various proteins were ob-
tained from the following sources: PARP (Poly-ADP-ribose poly-
merase) was obtained from Santa Cruz Biotechnology Inc. (Dallas,
4.2.7. Statistics and data analysis
Each experiment was performed at least three times, and
representative data are shown. Data in bar graphs are given as the
means S.D. Means were checked for statistical difference using
the t-test and P-values less than 0.05 were considered significant
(*P < 0.05, **P < 0.01, ***P < 0.001).
TX, USA). Caspase 8, caspase 9, and
gH2AX, were obtained from Cell
signaling (Danvers, MA, USA). -actin and GAPDH were obtained
b
from Millipore (Billerica, MA, USA). Caspase 3 was obtained from
Novous (Littleton, CO, USA). Anti-mouse and anti-rabbit IgGs were
from Jackson ImmunoResearch Laboratories (West Grove, PA, USA).
4.3. Molecular docking study
The crystal structure of HSP90 (PDB ID: 5GGZ) was obtained
from Protein Data Bank [25]. Protein preparation included removal
of water molecules from the crystal structure. Docking was per-
formed using the Knime [26] software and the FlexX node (https://
www.biosolveit.de/knime/). The co-crystal structure was used to
define the binding site. The 3D coordinates of docked compounds
were generated using the “Generate 3D Coordinate” mode. Finally,
all scoring parameters were used with the default settings.
4.2.2. HSP90 enzymatic assay
HSP90a Assay Kits (BPS Bioscience, San Diego, CA, USA) were
utilized following the instruction manual, master mixture. HSP90
assay buffer, DTT, BSA, H2O and FITC-labeled geldanamycin were
added to all 96 wells. Then test compounds were added to each
well designated “Test Inhibitor”. To the “Blank”, “Enzyme Positive
Control” and “Enzyme Negative Control” wells, the same solution
with inhibitor was added. HSP90 assay buffer was added to the
“Enzyme Negative Control” and “Blank” wells. Finally, the recom-
Declaration of competing interest
binant protein HSP90
a
was incubated in every well designated
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
“Enzyme Positive Control” and “Test Inhibitor” for 2e3 h at rt with
slow shaking. The fluorescent polarization of the samples was
determined by microtiter-plate reader, which can detect excitation
ranging from 475 to 495 nm and emission ranging from 518 to
538 nm.
Acknowledgement
This work was financially supported by the Ministry of Science
and Technology (MOST), Taiwan (Grant No. MOST 108-2320-B-038-
056, 108-2320-B-038-064, MOST108-2320-B-038-027-MY3 and
108-2113-M-038-005) and the “TMU Research Center of Cancer
Translational Medicine” from The Featured Areas Research Center
Program within the framework of the Higher Education Sprout
Project by the Ministry of Education (MOE) in Taiwan.
4.2.3. SRB (sulforhodamine B) assay
Cells were seeded in 96-well plates and cultured overnight
followed by the exposure to gradient concentrations of different
compounds for 48 h. Briefly, cells in Tz group were fixed in situ with
10% trichloroacetic acid (TCA) to represent a measurement of the
cell population at the time of drug addition (T0). After an additional
48 h incubation with or without compounds in medium with 5%
FBS, the assay was terminated by 10% TCA in CTL and treatment
groups. SRB dye purchased from Sigma (St. Louis, MO, USA) at 0.4%
(w/v) in 1% acetic acid was added to stain the cells. Unbound dye
was removed by 1% acetic acid and the plates were air dried. Bound
dye was subsequently solubilized with 10 mM trizma base, and the
absorbance was read at a wavelength of 515 nm.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.ejmech.2020.112540.
References
[1] H. Wang, M. Lu, M. Yao, W. Zhu, Effects of treatment with an Hsp90 inhibitor
in tumors based on 15 phase II clinical trials, Mol Clin Oncol 5 (3) (2016 Sep)
326e334.
[2] (a) S.C. Gupta, A. Sharma, M. Mishra, R.K. Mishra, D.K. Chowdhuri, Heat shock
proteins in toxicology: how close and how far? Life Sci. 86 (11-12) (2010 Mar
13) 377e384;
(b) C.J. Park, Y.S. Seo, Heat shock proteins: a review of the molecular chap-
erones for plant immunity, Plant Pathol. J. 31 (4) (2015 Dec) 323e333.
[3] (a) A. Hoter, M.E. El-Sabban, H.Y. Naim, The HSP90 family: structure, regula-
tion, function, and implications in health and disease, Int. J. Mol. Sci. 19 (9)
(2018 Aug 29) E2560, pii;
4.2.4. Colony formation assay
Cells were plated in 96 well plates (5 ꢁ 103/well) and exposed to
DMSO or compound 12c at indicated concentrations for 24 h. The
drugs were then washed away and cells were trypsinized and
seeded in 24-well plates for continuing growth for 10 days. The
colonies were fixed and stained with crystal violet (0.5% in 70%
EtOH) and the experiments were repeated at least twice.
(b) O. Genest, S. Wickner, S.M. Doyle, Hsp90 and Hsp70 chaperones: collab-
orators in protein remodeling, J. Biol. Chem. 294 (6) (2019 Feb 8) 2109e2120;
(c) F.H. Schopf, M.M. Biebl, J. Buchner, The HSP90 chaperone machinery, Nat.
Rev. Mol. Cell Biol. 18 (6) (2017 Jun) 345e360.
4.2.5. FACScan Flow Cytometric analysis
Cells were seeded in 6-well plates (2.5 ꢁ 105/well) and treated
with DMSO or 12c at various concentrations for indicated times.
Cells were washed with phosphate-buffered saline, fixed in ice-cold
70% EtOH at ꢂ20 ꢀC overnight, and stained with propidium iodide
[4] (a) W. Luo, W. Sun, T. Taldone, A. Rodina, G. Chiosis, Heat shock protein 90 in
neurodegenerative diseases, Mol. Neurodegener. 5 (2010 Jun 3) 24;
(b) J.J. Barrott, T.A. Haystead, Hsp90, an unlikely ally in the war on cancer, FEBS
J. 280 (6) (2013 Mar) 1381e1396;
(80 mg/ml) containing Triton X-100 (0.1%, v/v) and RNase A (100 mg/
(c) N. Roy, R.K. Nageshan, S. Ranade, U. Tatu, Heat shock protein 90 from
neglected protozoan parasites, Biochim. Biophys. Acta 1823 (3) (2012 Mar)
707e711;
ml) in phosphate-buffered saline (PBS). DNA content was analyzed
with the FACScan and CellQuest software (Becton Dickinson,
Mountain View, CA, USA).
(d) R. Geller, S. Taguwa, J. Frydman, Broad action of Hsp90 as a host chaperone

W.-C. Wu et al. / European Journal of Medicinal Chemistry 203 (2020) 112540
13
required for viral replication, Biochim. Biophys. Acta 1823 (3) (2012 Mar)
698e706;
S. Fujiwara, H. Watanabe, K. Kurashina, H. Hatanaka, M. Bando, S. Ohno,
Y. Ishikawa, H. Aburatani, T. Niki, Y. Sohara, Y. Sugiyama, H. Mano, Identifi-
cation of the transforming EML4-ALK fusion gene in non-small-cell lung
cancer, Nature 448 (7153) (2007 Aug 2) 561e566.
(e) H. Fuhrmann-Stroissnigg, Y.Y. Ling, J. Zhao, S.J. McGowan, Y. Zhu,
R.W. Brooks, D. Grassi, S.Q. Gregg, J.L. Stripay, A. Dorronsoro, L. Corbo, P. Tang,
C. Bukata, N. Ring, M. Giacca, X. Li, T. Tchkonia, J.L. Kirkland, L.J. Niedernhofer,
P.D. Robbins, Identification of HSP90 inhibitors as a novel class of senolytics,
Nat. Commun. 8 (1) (2017 Sep 4) 422.
[17] (a) L.V. Sequist, B.A. Waltman, D. Dias-Santagata, S. Digumarthy, A.B. Turke,
P. Fidias, K. Bergethon, A.T. Shaw, S. Gettinger, A.K. Cosper, S. Akhavanfard,
R.S. Heist, J. Temel, J.G. Christensen, J.C. Wain, T.J. Lynch, K. Vernovsky,
E.J. Mark, M. Lanuti, A.J. Iafrate, M. Mino-Kenudson, J.A. Engelman, Genotypic
and histological evolution of lung cancers acquiring resistance to EGFR in-
hibitors, Sci. Transl. Med. 3 (75) (2011 Mar 23), 75ra26;
[5] (a) E. Vartholomaiou, P.C. Echeverría, D. Picard, Unusual suspects in the twi-
light zone between the Hsp90 interactome and carcinogenesis, Adv. Canc. Res.
129 (2016) 1e30;
(b) P.J. Webber, C. Park, M. Qui, S.S. Ramalingam, F.R. Khuri, H. Fu, Y. Du,
Combination of heat shock protein 90 and focal adhesion kinase inhibitors
synergistically inhibits the growth of non-small cell lung cancer cells, Onco-
science 2 (9) (2015 Sep 15) 765e776;
(c) Y.V. Liu, J.H. Baek, H. Zhang, R. Diez, R.N. Cole, G.L. Semenza, RACK1
competes with HSP90 for binding to HIF-1alpha and is required for O(2)-
independent and HSP90 inhibitor-induced degradation of HIF-1alpha, Mol.
Cell. 25 (2) (2007 Jan 26) 207e217.
(b) K. Ohashi, Y.E. Maruvka, F. Michor, W. Pao, Epidermal growth factor re-
ceptor tyrosine kinase inhibitor-resistant disease, J. Clin. Oncol. 31 (8) (2013
Mar 10) 1070e1080;
(c) S. Yano, T. Yamada, S. Takeuchi, K. Tachibana, Y. Minami, Y. Yatabe,
T. Mitsudomi, H. Tanaka, T. Kimura, S. Kudoh, H. Nokihara, Y. Ohe, J. Yokota,
H. Uramoto, K. Yasumoto, K. Kiura, M. Higashiyama, M. Oda, H. Saito,
J. Yoshida, K. Kondoh, M. Noguchi, Hepatocyte growth factor expression in
EGFR mutant lung cancer with intrinsic and acquired resistance to tyrosine
kinase inhibitors in a Japanese cohort, J. Thorac. Oncol. 6 (12) (2011 Dec)
2011e2017;
[6] E. Pick, Y. Kluger, J.M. Giltnane, C. Moeder, R.L. Camp, D.L. Rimm, H.M. Kluger,
High HSP90 expression is associated with decreased survival in breast cancer,
Canc. Res. 67 (7) (2007 Apr 1) 2932e2937.
€
(d) S. Kobayashi, T.J. Boggon, T. Dayaram, P.A. Janne, O. Kocher, M. Meyerson,
[7] K. Jhaveri, S.O. Ochiana, M.P. Dunphy, J.F. Gerecitano, A.D. Corben, R.I. Peter,
Y.Y. Janjigian, E.M. Gomes-DaGama, J. Koren 3rd, S. Modi, G. Chiosis, Heat
shock protein 90 inhibitors in the treatment of cancer: current status and
future directions, Expet Opin. Invest. Drugs 23 (5) (2014 May) 611e628.
[8] R.N. Pillai, S.S. Ramalingam, Heat shock protein 90 inhibitors in non-small-cell
lung cancer, Curr. Opin. Oncol. 26 (2) (2014 Mar) 159e164.
[9] D.B. Longley, D.P. Harkin, P.G. Johnston, 5-fluorouracil: mechanisms of action
and clinical strategies, Nat. Rev. Canc. 3 (5) (2003 May) 330e338.
[10] (a) C. Bailly, Irinotecan: 25 years of cancer treatment, Pharmacol. Res. 148
(2019 Oct) 104398;
B.E. Johnson, M.J. Eck, D.G. Tenen, B. Halmos, EGFR mutation and resistance of
non-small-cell lung cancer to gefitinib, N. Engl. J. Med. 352 (8) (2005 Feb 24)
786e792.
[18] (a) T. Seto, T. Kato, M. Nishio, K. Goto, S. Atagi, Y. Hosomi, N. Yamamoto,
T. Hida, M. Maemondo, K. Nakagawa, S. Nagase, I. Okamoto, T. Yamanaka,
K. Tajima, R. Harada, M. Fukuoka, N. Yamamoto, Erlotinib alone or with
bevacizumab as first-line therapy in patients with advanced non-squamous
non-small-cell lung cancer harbouring EGFR mutations (JO25567): an open-
label, randomised, multicentre, phase 2 study, Lancet Oncol. 15 (11) (2014
Oct) 1236e1244;
(b) J. Giuliani, A. Bonetti, First-line therapies in metastatic colorectal cancer:
integrating clinical benefit with the costs of drugs, Int. J. Colorectal Dis. 33 (11)
(2018 Nov) 1505e1516.
(b) M. Reck, E.B. Garon, L. Paz-Ares, S. Ponce, J.C. Jaime, O. Juan, E. Nadal,
K. Kiura, R.C. Widau, S. He, R. Dalal, P. Lee, K. Nakagawa, Randomized, double-
blind phase ib/III study of erlotinib with ramucirumab or placebo in previ-
ously untreated EGFR-mutant metastatic non-small-cell lung cancer (RELAY):
phase ib results, Clin. Lung Canc. 19 (3) (2018 May) 213e220;
(c) P. Shi, Y.T. Oh, L. Deng, G. Zhang, G. Qian, S. Zhang, H. Ren, G. Wu,
B. Legendre Jr., E. Anderson, S.S. Ramalingam, T.K. Owonikoko, M. Chen,
S.Y. Sun, Overcoming acquired resistance to AZD9291, A third-generation
EGFR inhibitor, through modulation of MEK/ERK-dependent bim and mcl-1
degradation, Clin. Canc. Res. 23 (21) (2017 Nov 1) 6567e6579;
(d) Z. Yang, K.Y. Tam, Combination strategies using EGFR-TKi in NSCLC ther-
apy: learning from the gap between pre-clinical results and clinical outcomes,
Int. J. Biol. Sci. 14 (2) (2018 Feb 5) 204e216.
[11] B.R. Hirsch, S.Y. Zafar, Capecitabine in the management of colorectal cancer,
Canc. Manag. Res. 3 (2011) 79e89.
[12] (a) V. Heinemann, L.F. von Weikersthal, T. Decker, A. Kiani, U. Vehling-Kaiser,
S.E. Al-Batran, T. Heintges, C. Lerchenmüller, C. Kahl, G. Seipelt, F. Kullmann,
M. Stauch, W. Scheithauer, J. Hielscher, M. Scholz, S. Müller, H. Link,
€
N. Niederle, A. Rost, H.G. Hoffkes, M. Moehler, R.U. Lindig, D.P. Modest,
L. Rossius, T. Kirchner, A. Jung, S. Stintzing, FOLFIRI plus cetuximab versus
FOLFIRI plus bevacizumab as first-line treatment for patients with metastatic
colorectal cancer (FIRE-3): a randomised, open-label, phase 3 trial, Lancet
Oncol. 15 (10) (2014 Sep) 1065e1075;
(b) C. Cremolini, F. Loupakis, C. Antoniotti, C. Lupi, E. Sensi, S. Lonardi, S. Mezi,
G. Tomasello, M. Ronzoni, A. Zaniboni, G. Tonini, C. Carlomagno, G. Allegrini,
S. Chiara, M. D’Amico, C. Granetto, M. Cazzaniga, L. Boni, G. Fontanini,
A. Falcone, FOLFOXIRI plus bevacizumab versus FOLFIRI plus bevacizumab as
first-line treatment of patients with metastatic colorectal cancer: updated
overall survival and molecular subgroup analyses of the open-label, phase 3
TRIBE study, Lancet Oncol. 16 (13) (2015 Oct) 1306e1315;
[19] (a) J. Nakano, C. Huang, D. Liu, D. Masuya, T. Nakashima, H. Yokomise,
M. Ueno, H. Wada, M. Fukushima, Evaluations of biomarkers associated with
5-FU sensitivity for non-small-cell lung cancer patients postoperatively
treated with UFT, Br. J. Canc. 95 (5) (2006 Sep 4) 607e615;
(b) J.G. Zhao, K.M. Ren, J. Tang, Overcoming 5-Fu resistance in human non-
small cell lung cancer cells by the combination of 5-Fu and cisplatin
through the inhibition of glucose metabolism, Tumour Biol 35 (12) (2014 Dec)
12305e12315.
(c) S. Stintzing, D.P. Modest, L. Rossius, M.M. Lerch, L.F. von Weikersthal,
T. Decker, A. Kiani, U. Vehling-Kaiser, S.E. Al-Batran, T. Heintges,
C. Lerchenmüller, C. Kahl, G. Seipelt, F. Kullmann, M. Stauch, W. Scheithauer,
S. Held, C. Giessen-Jung, M. Moehler, A. Jagenburg, T. Kirchner, A. Jung,
V. Heinemann, FIRE-3 investigators. FOLFIRI plus cetuximab versus FOLFIRI
plus bevacizumab for metastatic colorectal cancer (FIRE-3): a post-hoc anal-
ysis of tumour dynamics in the final RAS wild-type subgroup of this rando-
mised open-label phase 3 trial, Lancet Oncol. 17 (10) (2016 Oct) 1426e1434.
[13] (a) Y. Chihara, A. Yoshimura, K. Date, Y. Takemura, N. Tamiya, Y. Kohno,
T. Imabayashi, M. Takeuchi, Y. Kaneko, T. Yamada, M. Ueda, T. Arimoto,
J. Uchino, Y. Iwasaki, K. Takayama, Phase II study of S-1 and paclitaxel com-
bination therapy in patients with previously treated non-small cell lung
cancer, Oncol. 24 (8) (2019 Aug) 1033. -e617;
(b) Q. Lin, F.J. Meng, Y. Liu, N. Wang, X.C. Ren, X.J. Chen, X.H. Ge, C.X. Liu, Jing-
Hu, D.Y. Wang, B. Cao, Y. Zhao, Phase II trial of capecitabine combined with
docetaxel in previously treated patients with non-small cell lung cancer: a
randomized controlled study, Oncol Lett 3 (4) (2012 Apr 1) 761e766.
[14] D.R. Camidge, W. Pao, L.V. Sequist, Acquired resistance to TKIs in solid tu-
mours: learning from lung cancer, Nat. Rev. Clin. Oncol. 11 (8) (2014 Aug)
473e481.
[20] (a) T. Shimamura, S.A. Perera, K.P. Foley, J. Sang, S.J. Rodig, T. Inoue, L. Chen,
D. Li, J. Carretero, Y.C. Li, P. Sinha, C.D. Carey, C.L. Borgman, J.P. Jimenez,
M. Meyerson, W. Ying, J. Barsoum, K.K. Wong, G.I. Shapiro, Ganetespib (STA-
9090), a nongeldanamycin HSP90 inhibitor, has potent antitumor activity in
in vitro and in vivo models of non-small cell lung cancer, Clin. Canc. Res. 18
(18) (2012 Sep 15) 4973e4985;
(b) E.B. Garon, R.S. Finn, H. Hamidi, J. Dering, S. Pitts, N. Kamranpour,
A.J. Desai, W. Hosmer, S. Ide, E. Avsar, M.R. Jensen, C. Quadt, M. Liu,
S.M. Dubinett, D.J. Slamon, The HSP90 inhibitor NVP-AUY922 potently inhibits
non-small cell lung cancer growth, Mol. Canc. Therapeut. 12 (6) (2013 Jun)
890e900.
[21] H. Dimant, D. Ebrahimi-Fakhari, P.J. McLean, Molecular chaperones and co-
chaperones in Parkinson disease, Neuroscientist 18 (6) (2012 Dec) 589e601.
[22] F. Jiang, H.J. Wang, Y.H. Jin, et al., Novel tetrahydropyrido[4,3-d]pyrimidines as
potent inhibitors of chaperone heat shock protein 90, J. Med. Chem. 59 (23)
(2016) 10498e10519.
[23] P.A. Brough, W. Aherne, X. Barril, et al., 4,5-diarylisoxazole Hsp90 chaperone
inhibitors: potential therapeutic agents for the treatment of cancer, J. Med.
Chem. 51 (2) (2008) 196e218.
[15] C. Ma, S. Wei, Y. Song, T790M and acquired resistance of EGFR TKI: a literature
review of clinical reports, J. Thorac. Dis. 3 (1) (2011 Mar) 10e18.
[16] (a) M. Takeda, K. Nakagawa, First- and second-generation EGFR-TKIs are all
replaced to osimertinib in chemo-naive EGFR mutation-positive non-small
cell lung cancer? Int. J. Mol. Sci. 20 (1) (2019 Jan 3) E146, pii:;
[24] C.H. Chen, C.A. Changou, T.H. Hsieh, Y.C. Lee, C.Y. Chu, K.C. Hsu, H.C. Wang,
Y.C. Lin, Y.N. Lo, Y.R. Liu, J.P. Liou, Y. Yen, Dual inhibition of PIK3C3 and FGFR as
a new therapeutic approach to treat bladder cancer, Clin. Canc. Res. 24 (5)
(2018 Mar 1) 1176e1189.
[25] H.M. Berman, et al., The protein Data Bank, Nucleic Acids Res. 28 (2000)
235e242.
(b) X. Wang, D. Goldstein, P.J. Crowe, J.L. Yang, Next-generation EGFR/HER
tyrosine kinase inhibitors for the treatment of patients with non-small-cell
lung cancer harboring EGFR mutations: a review of the evidence, Onco-
Targets Ther. 9 (2016 Sep 6) 5461e5473;
€
[26] R. Michael, N.C. Berthold, Fabian Dill, Thomas R. Gabriel, Tobias Kotter,
Thorsten meinl, peter ohl, christoph sieb, killian thiel, bernd wiswedel, in:
Studies in Classification, Data Analysis, and Knowledge Organization (GfKL
2007), Springer, 2007.
(c) M. Soda, Y.L. Choi, M. Enomoto, S. Takada, Y. Yamashita, S. Ishikawa,