Targeting the entry region of Hsp90's ATP binding pocket with a novel 6,7-dihydrothieno[3,2-c]pyridin-5(4H)-yl amide

Article abstract of DOI:10.1016/j.ejmech.2016.10.038

The molecular chaperone Hsp90 plays an important role in cancer cell survival and proliferation by regulating the maturation and stabilization of numerous oncogenic proteins. Due to its potential to simultaneously disable multiple signaling pathways, Hsp90 has emerged as an attractive therapeutic target for cancer treatment. In this study, the design, synthesis, and biological evaluation of a series of Hsp90 inhibitors are described. Among the synthetic compounds, 6,7-dihydrothieno [3,2-c]pyridin-5(4H)-yl amide 19 exhibits a remarkable binding affinity to the N-terminus of Hsp90 in a fluorescence polarization (FP) binding assay (IC₅₀= 50.3 nM). Furthermore, it effectively inhibits the proliferation of H1975 non-small cell lung cancer (NSCLC) and Skbr3 breast cancer cell lines with GI₅₀values of 0.31 μM and 0.11 μM, respectively. Compound 19 induces the degradation of the Hsp90 client proteins including EGFR, Her2, Met, c-Raf, and Akt, and consequently promotes apoptotic cancer cell death. Compound 19 also inhibits the growth of H1975 xenografts in NOD-scid IL2R gamma^nullmice without any apparent body-weight loss. The immunohistologic evaluation indicates that compound 19 decreases the expression of Akt in xenograft tumor tissue via an inhibition of the Hsp90 chaperon function. Additionally, the cytochrome P450 assay indicates that compound 19 has no effect on the activities of five major P450 isoforms (IC₅₀> 50 μM for 1A2, 2C9, 2C19, 2D6, and 3A), suggesting that clinical interactions between compound 19 and the substrate drugs of the five major P450 isoforms are not expected. Overall, compound 19 represents a new class of Hsp90 inhibitor with its 6,7-dihydrothieno[3,2-c]pyridin-5(4H)-yl-amide structure, and it has the therapeutic potential to overcome drug resistance in cancer chemotherapy.

Full text of DOI:10.1016/j.ejmech.2016.10.038

Accepted Manuscript
Targeting the entry region of Hsp90's ATP binding pocket with a novel 6,7-
dihydrothieno[3,2-c]pyridin-5(4H)-yl amide
Ju Hui Jeong, Yong Jin Oh, Yunmee Lho, Sun You Park, Kwang-Hyeon Liu,
Eunyoung Ha, Young Ho Seo
PII:
S0223-5234(16)30900-X
10.1016/j.ejmech.2016.10.038
EJMECH 9003
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 20 July 2016
Revised Date: 16 October 2016
Accepted Date: 17 October 2016
Please cite this article as: J.H. Jeong, Y.J. Oh, Y. Lho, S.Y. Park, K.-H. Liu, E. Ha, Y.H. Seo, Targeting
the entry region of Hsp90's ATP binding pocket with a novel 6,7-dihydrothieno[3,2-c]pyridin-5(4H)-yl
amide, European Journal of Medicinal Chemistry (2016), doi: 10.1016/j.ejmech.2016.10.038.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Title: Targeting the entry region of Hsp90’s ATP binding pocket with a novel 6,7-
dihydrothieno[3,2-c]pyridin-5(4H)-yl amide
Ju Hui Jeong a, 1, Yong Jin Oh a, 1, Yunmee Lho b, Sun You Park a, Kwang-Hyeon Liu c,
Eunyoung Ha b, **, Young Ho Seo a, *
a College of Pharmacy, Keimyung University, Daegu 704-701, South Korea.
b Department of Biochemistry, School of Medicine, Keimyung University, Daegu 704-701,
South Korea.
c BK21 Plus KNU Multi-Omics based Creative Drug Research Team, College of Pharmacy
and Research Institute of Pharmaceutical Sciences, Kyungpook National University, Daegu
41566, South Korea.
* Corresponding author
** Corresponding author
E-mail addresses: seoyho@kmu.ac.kr (Y. H. Seo), hanne.md@gmail.com (E. Ha)
1 Authors are equal contributors

ACCEPTED MANUSCRIPT
Graphical Abstract
Highlights
ó Design, synthesis, and biological evaluation of a series of Hsp90 inhibitors are presented.
ó Compound 19 exhibits a remarkable anticancer activity against H1975 non-small cell lung
cancer and Skbr3 breast cancer cell lines.
ó Compound 19 inhibits the growth of H1975 xenografts in NOD-scid IL2Rgammanull mice
through Hsp90 inhibition.
ó Compound 19 has no effect on the activities of five major P450 isoforms (IC50 > 50 µM
for 1A2, 2C9, 2C19, 2D6, and 3A), suggesting that clinical interactions between
compound 19 and the substrate drugs of the five major P450 isoforms are not expected.

ACCEPTED MANUSCRIPT
Abstract
The molecular chaperone Hsp90 plays an important role in cancer cell survival and
proliferation by regulating the maturation and stabilization of numerous oncogenic proteins.
Due to its potential to simultaneously disable multiple signaling pathways, Hsp90 has
emerged as an attractive therapeutic target for cancer treatment. In this study, the design,
synthesis, and biological evaluation of a series of Hsp90 inhibitors are described. Among the
synthetic compounds, 6,7-dihydrothieno[3,2-c]pyridin-5(4H)-yl amide 19 exhibits a
remarkable binding affinity to the N-terminus of Hsp90 in a fluorescence polarization (FP)
binding assay (IC50 = 50.3 nM). Furthermore, it effectively inhibits the proliferation of H1975
non-small cell lung cancer (NSCLC) and Skbr3 breast cancer cell lines with GI50 values of
0.31 µM and 0.11 µM, respectively. Compound 19 induces the degradation of the Hsp90
client proteins including EGFR, Her2, Met, c-Raf, and Akt, and consequently promotes
apoptotic cancer cell death. Compound 19 also inhibits the growth of H1975 xenografts in
NOD-scid IL2R gammanull mice without any apparent body-weight loss. The
immunohistologic evaluation indicates that compound 19 decreases the expression of Akt in
xenograft tumor tissue via an inhibition of the Hsp90 chaperon function. Additionally, the
cytochrome P450 assay indicates that compound 19 has no effect on the activities of five
major P450 isoforms (IC50 > 50 µM for 1A2, 2C9, 2C19, 2D6, and 3A), suggesting that
clinical interactions between compound 19 and the substrate drugs of the five major P450
isoforms are not expected. Overall, compound 19 represents a new class of Hsp90 inhibitor
with its 6,7-dihydrothieno[3,2-c]pyridin-5(4H)-yl-amide structure, and it has the therapeutic
potential to overcome drug resistance in cancer chemotherapy.

ACCEPTED MANUSCRIPT
1. Introduction
Heat shock protein 90 (Hsp90) is an ATP-dependent molecular chaperone that is
responsible for the correct folding, stability and function of its substrate proteins, referred to
as “client” proteins. Many Hsp90 client proteins are oncogenic proteins that are involved in
malignant behavior and cancer progression [1,2]. These client proteins include epidermal
growth factor receptor (EGFR/ErbB1), human epidermal growth factor receptor 2
(Her2/ErbB2), mesenchymal-epithelial transition factor (Met), anaplastic lymphoma kinase
(Alk), protein kinase B (Akt/PKB), cellular rapidly accelerated fibrosarcoma (c-Raf), cyclin-
dependent kinase 4 (Cdk4), hypoxia-inducible factor 1 (Hif-1α), matrix metalloproteinase 2
(MMP2), mutant P53, and wee1 [3]. Due to the hostile environmental factors of cancer such
as hypoxia, low pH, and poor nutrition, cancer cells depend more on the chaperone function
of Hsp90 than normal cells, which explains 2-10 fold higher expression level of Hsp90 in
cancer cells than normal cells [3-5]. Accordingly, the inhibition of Hsp90 chaperone function
induces the degradation of the client proteins via the ubiquitin-proteasome pathway, and
subsequently results in a simultaneous disruption of the multiple signaling pathways in cancer.
Hsp90 consists of the following four major homologs; Hsp90
regulated protein (Grp94), and TNF receptor-associated protein 1 (Trap1) [6,7]. Hsp90
inducible form that is overexpressed in many cancer cells, and Hsp90 is the constitutive
form. Both Hsp90 and Hsp90 are mainly localized in the cytoplasm, while Grp94 and
α
, Hsp90
β
, 94-kDa glucose-
α
is an
β
α
β
Trap1 reside in the endoplasmic reticulum and the mitochondria, respectively. Hsp90 has
been extensively pursued as a promising target for overcoming drug resistance. The potential
therapeutic benefits associated with Hsp90 modulation emphasize the importance of
identifying novel Hsp90 inhibitors [8-11].

ACCEPTED MANUSCRIPT
Figure 1. Structures of known Hsp90 inhibitors
Geldanamycin (GA, 1), a benzoquinone ansamycin antibiotic was first identified as an
Hsp90 inhibitor in 1994 [12]. Since then, a number of geldanamycin analogues such as
tanespimycin (17-AAG) and alvespimycin (17-DMAG) have been developed and entered
into clinical study [13]. However, these first-generation Hsp90 inhibitors engendered several
drawbacks in the clinical applications such as a poor solubility and toxicity [14]. The most
clinically significant toxicity of the geldanamycin analogues is liver toxicity. The liver
toxicity was a major limitation in the clinical development, and it was speculated that the
benzoquinone structure of geldanamycin analogues largely contribute to the observed liver
toxicity [15].
Benefited from the X-ray crystal structures of Hsp90 and fragment-based drug discovery,
investigators discovered the second-generation Hsp90 inhibitors, which are classified into
two major cores, purine and resorcinol. The purine class of Hsp90 inhibitors are designed
according to the structural homology of ATP and include PU-H71 [16], BIIB021 [17], and

ACCEPTED MANUSCRIPT
CUDC-305 [18]. The resorcinol scaffold is another important class of Hsp90 inhibitors that
includes NVP-AUY922 [19], KW-2478 [20], AT13387 [21] and STA-9090 [22]. Based on
the wealth of compelling data from the preclinical studies, many of the second-generation
Hsp90 inhibitors have entered into the clinical phase, and display an improved potency with a
lesser liver toxicity [23]. Despite an enormous effort and promise, none of the Hsp90
inhibitors are clinically approved, and the full potential of this inhibitor clasee is yet to be
realized. Accordingly, the discovery of Hsp90 inhibitors with different chemotypes is still a
demanding task in this area.
2. Results and discussion
2.1. Analysis of ATP-binding site
Figure 2. Analysis of ATP-binding site in the N-terminal domain of Hsp90 protein and the

ACCEPTED MANUSCRIPT
binding poses of Hsp90 inhibitors in the ATP-binding site. (A) Superimposed crystal
structures of three apo-Hsp90 proteins (PDB codes: 2XJX, orange; 2VCI, cyan; and 2TUH,
yellow) and conserved waters (red-colored spheres), where the side chains of the binding site
are colored by atom types (oxygen, red; nitrogen, blue). (B) Overlaid binding modes of
AT13387 (orange), NVP-AUY922 (cyan), and STA-9090 (yellow) from co-crystal structures
of Hsp90 (PDB codes: 2XJX, 2VCI, and 2TUH), where the oxygen and nitrogen atoms of
AT13387, NVP-AUY922, and STA-9090 are shown in red and blue, respectively.
To rationally design a potent Hsp90 inhibitors, we first investigated crystal structures of
Hsp90. To do so, we aligned three Hsp90 proteins (PDB codes: 2XJX, orange; 2VCI, cyan;
2TUH, yellow) and analyzed the ATP-binding site of the Hsp90 proteins with their Hsp90
inhibitors (AT13387, orange; NVP-AUY922, cyan; and STA-9090, yellow), all of which had
been studied in clinical trials. Comparative structural analysis of the three Hsp90 proteins
indicated that the ATP-binding site in the N-terminal domain consists of a hydrophilic region,
a hydrophobic region, and an entry region (Figure 2A). The hydrophilic region is composed
of Asn51, Asp93, and three conserved water molecules and the hydrophobic region consists
of lipophilic amino acid residues, Leu107, Phe138, Trp162 and Val186. The hydrophilic and
hydrophobic regions form a deep ATP-binding pocket in the N-terminal domain of Hsp90. In
particular, the structure of the hydrophilic region is characterized by highly ordered amino
acid residues and conserved water molecules in the pocket. The three conserved water
molecules, along with the carboxylic acid terminus of Asp93 residue, form a hydrogen
bonding network with the resorcinol ring of Hsp90 inhibitors. The deep cavity formed by the
hydrophilic and hydrophobic regions plays a crucial role in the binding of the Hsp90
inhibitors. Enormous effort had therefore been directed toward the targeting of this deep
cavity, and it was found that 4-isopropyl resorcinol scaffold tightly fits into the hydrophilic

ACCEPTED MANUSCRIPT
and hydrophobic regions as a key binder. Unlike the hydrophilic and hydrophobic regions,
the entry region is relatively flexible, and diverse scaffolds such as 4,5-diarylisoxazole, 1,3-
dihydroisoindole, 4-indolyl-triazolone, and 5-aryl-1,2,3-triazole, have been studied to target
the entry region. A structural analysis of AT13387 (orange), NVP-AUY922 (cyan), and STA-
9090 (yellow) revealed that the binding poses of 4-isopropyl resorcinol ring in the ATP-
binding pocket are highly ordered in the hydrophilic and hydrophobic regions, while the
binding poses of the entry region binders are relatively disparate (Figure 2B).
To develop the Hsp90 inhibitors with a novel scaffold, an exploration of the diverse
modifications of the entry region binder, while 4-chloro or 4-isopropylresorcinol ring was
maintained as a key binder in the hydrophilic and hydrophobic regions, was conducted.
2.2. Chemistry
Scheme 1. Synthesis of compound 11a-b.
We first decided to synthesize compounds 11a-b, which had chloro group at C-5 of phenyl
ring (Scheme 1). To do so, we pursued the synthesis of carboxylic acid 9 that was a key

ACCEPTED MANUSCRIPT
intermediate in the synthetic plan. Methyl 2,4-dihydroxybenzoate (6) was synthesized from
commercially available 2,4-dihydroxybenzoic acid (5) with a slight modification of the
reported procedure [24]. Briefly, 5 was esterified in the presence of sulfuric acid in methanol
to provide methyl 2,4-dihydroxybenzoate (6) in 86 % yield. Subsequent chlorination of 6
with sulfuryl chloride in dichloromethane furnished methyl 5-chloro-2,4-dihydroxybenzoate
(7) in 45% yield, which was then protected with allyl bromide in the presence of potassium
carbonate to give compound 8 in 100 % yield. Compound 8 was hydrolyzed to furnish
carboxylic acid 9 in 80 % yield. With carboxylic acid 9 in hand, we next performed amide
coupling reaction of carboxylic acid 9 with 1,2,3,4-tetrahydroisoquinoline and 4,5,6,7-
tetrahydrothieno[3,2-c] pyridine and the reaction provided compound 10a-b in 63 and 58 %
yield, respectively. Finally, the removal of allyl-protecting groups, using PdCl2(PPh3)2 and
ammonium formate under microwave irradiation [25] furnished the resulting 5-chloro-2,4-
hydroxybenzamides (11a-b) in 66 and 27 % yield. In this synthetic route, the yields of
deprotection reactions were disappointingly low and the impurity of triphenylphosphine was
also difficult to remove from the reaction mixture by flash column. Besides, the protection-
deprotection event introduced two additional steps in the reaction sequence, and it needs to be
minimized to achieve a green and atom-economic chemistry. Therefore, we investigated
direct amide coupling reaction of 5-chloro-2,4-dihydroxybenzoic acid (12) without any
protecting group (Scheme 2).
Scheme 2. Synthesis of compound 11c-e.

ACCEPTED MANUSCRIPT
To prepare carboxylic acid 12, we carried out the hydrolysis reaction of ester 7 in the
presence of sodium hydroxide in methanol and water to provide compound 12 in quantitative
yield. With carboxylic acid 12 in hand, amide coupling reactions of carboxylic acid 12 with
various amines were carried out using EDC under microwave irradiation to provide
compound 11c-e in 14-34% yield. Although the amide coupling reactions still provided
unsatisfactory % yields, the synthetic route without the protection-deprotection steps resulted
in an improvement of the overall yields.
Scheme 3. Synthesis of compound 15.
We next synthesized benzoxazole-containing compound 15. Benzoxazole is an important
class of heterocyclic compounds found in a number of anticancer agents. The synthesis of
compound 15 is described in Scheme 3. The reaction of 9 with 2-aminophenol (13) was
carried out in the presence of thionyl chloride, dimethyl formaldehyde, and pyridine to
furnish 14 in 59% yield. The subsequent removal of the allyl-protecting groups, using
PdCl2(PPh3)2 and ammonium formate under microwave irradiation furnished compound 15 in
20% yield.

ACCEPTED MANUSCRIPT
2.3. In vitro assays of compounds 11a-e and 15
Table 1. Inhibitory activity against Hsp90
α and cell growth inhibition against cancer cell
lines
Hsp90
α
H1975b
(GI50; µM)
Skbr3c
(GI50; µM)
(FP)a
Compound
R2
(IC50; nM)
183.3
135.9
480.8
>1000
>1000
>1000
19.9
8.7
7.1
5.4
11a
11b
11c
11d
11e
15
19.1
> 50
> 50
> 50
9.3
H
N
NA
NA
> 50
H
N
OH

ACCEPTED MANUSCRIPT
a
Binding to N-terminal domain of Hsp90
α was determined by a fluorescence
b
polarization (FP) assay (14 h). Cytotoxicity on H1975 non-small cell lung cancer
cell (NSCLC) line. c Cytotoxicity on Skbr3 breast cancer cell line.
We next investigated the binding affinity of the resulting compounds (11a-e and 15) to
human recombinant Hsp90
α and their anti-proliferative activity on two human cancer cell
lines, H1975 and Skbr3, which are a gefitinib-resistant non-small cell lung cancer cell line
and a Her2-overexpressing breast cancer cell line, respectively. As shown in Table 1,
compounds 11a-c, which had secondary amide groups, exhibited an appreciable binding
affinity to human Hsp90
efficacy against H1975 and Skbr3. Notably, analogue 11b afforded the highest binding
affinity (IC50 = 135.9 nM) to human Hsp90 in the FP assay compared with other analogues.
α in a fluorescence polarization (FP) assay as well as good cellular
α
Consistent with the result of Hsp90 binding assay, analogue 11b also displayed the most
potent anti-proliferative activity against the two cancer cell lines, H1975 and Skbr3.
Interestingly, analogues 11d-e, which had secondary amide groups, were unable to bind to
human Hsp90α, so they did not show any cytotoxic effect on H1975 and Skbr3. Additionally,
benzoxazole analogue 15 did not exhibit any activity in the FP assay or the cell growth
inhibition assay. The resulting data indicated that the scaffold of the entry region binder was
critical for the improvement of the binding affinity to the ATP-binding pocket of Hsp90.
Nonetheless, compound 11b afforded the highest binding affinity to human Hsp90α, and the
most potent cellular efficacy in the two cancer cell lines, H1975 and Skbr3.
2.4. Synthesis of compound 19

ACCEPTED MANUSCRIPT
Scheme 4. Synthesis of compound 19.
Encouraged by the potency conferred by compound 11b, the synthesis of compound 19,
which contains an isopropyl group at C-5 of the phenyl ring, was embarked upon next.
Isopropyl moiety is a common hydrophobic binder that is employed in the clinical
development of resorcinol-based Hsp90 inhibitors. It has been reported that the replacement
of the 5-chloro substituent with an isopropyl group in the resorcinol ring will increase the
number of additional hydrophobic interactions in the hydrophobic region [21]. Accordingly,
we pursued the synthesis of compound 19. The synthetic route of compound 19 is illustrated
in Scheme 4. The Friedel-Crafts alkylation of 6 with isopropyl bromide was carried out in the
presence of aluminum chloride to provide methyl 1,4-dihydroxy-5-isopropylbenzoate (16) in
46% yield. A subsequent hydrolysis of 16 with lithium hydroxide in methanol and water
provided carboxylic acid 17 in 100% yield. Lastly, the amide coupling reaction of 17 with
commercially available secondary amine 18 using EDC afforded compound 19 in 52% yield.
2.5. Comparative in vitro assays of compound 19 with geldanamycin and 11b
Table 2. Inhibitory activity against Hsp90
α and cell growth inhibition against cancer cell
lines
H1975b
(GI50; µM)
Skbr3c
(GI50; µM)
Hsp90
(IC50; nM)
α
(FP)a
Compound
MW
cLogP

ACCEPTED MANUSCRIPT
GA (1)
11b
14.0
135.9
50.3
0.56
8.7
0.43
5.4
560
309
317
2.06
2.61
3.32
0.31
0.11
19
a
Binding to N-terminal domain of Hsp90
α was determined by a fluorescence
b
polarization (FP) assay (14 h). Cytotoxicity on H1975 non-small cell lung cancer
cell(NSCLC) line. c Cytotoxicity on Skbr3 breast cancer cell line.
With compound 19 in hand, we investigated the binding affinity of compound 19 to human
Hsp90
compound 19 afforded a significantly improved binding affinity to recombinant human
Hsp90 protein in the FP assay with an IC50 value of 50.3 nM. It is noteworthy that the
α and its cell growth inhibition activity on H1975 and Skbr3. As shown in Table 2,
α
binding affinity of 19 is 2.7-fold higher than that of compound 11b that has a chloro group at
C-5 of the phenyl ring as a hydrophobic binder. As expected, isopropyl moiety retained a
stronger binding affinity in the hydrophobic region than chloro moiety. The cell growth
inhibitory effect of 19 was measured with H1975 and Skbr3 cancer cell lines using MTS
colorimetric assay and it provided submicromolar GI50 values of 0.11 µM and 0.3 µM against
H1975 and Skbr3 cell lines, respectively. Interestingly, according to the cell growth inhibition
assay, the cellular efficacy of compound 19 (H1975 GI50 = 0.31 µM, Skbr3 GI50 = 0.11 µM)
was 30 to 50 fold more active than 11b (H1975 GI50 = 8.7 µM, Skbr3 GI50 = 5.4 µM) against
H1975 and Skbr3 cancer cell lines. Considering that the binding affinity of 19 (IC50 = 50.3
nM) to Hsp90
α protein in the FP assay was only 2.7 fold higher than that of 11b (IC50 =
135.9 nM), it was assumed that compound 19 is more capable of penetrating into the cells
than 11b, affording highly active cellular efficacy against cancer cell lines. Furthermore, the
difference of the hydrophobicity values between compound 19 (clogP = 3.32) and 11b (clogP
= 2.61) also supported this assumption. Importantly, compound 19 is more potent against

ACCEPTED MANUSCRIPT
H1975 and Skbr3 than the reference compound GA in the cell growth inhibition assay. The
ligand efficacy (LE), which refers to the potency per unit of the molecular weight, is
considered to be an important criteria for judging the potential of the improvement in drug
discovery. Through a comparison of the molecular weight of 19 (MW = 317) with that of GA
(MW = 560), compound 19 has a high improvement potential in preclinical and clinical
studies. Collectively, the advantage of isopropyl moiety over chloro moiety became apparent
when the protein binding affinity, cellular growth inhibitory effect, and cell permeability were
all considered.
2.6. Molecular modeling of 19
Figure 3. Molecular docking model of 19 bound in the ATP-binding pocket of human Hsp90

ACCEPTED MANUSCRIPT
(PDB code: 2XJX). The carbon, oxygen, nitrogen, and sulfur atoms of 19 are shown in lime, red,
blue, and yellow, respectively. The side chains of the binding site are colored according to the atom
types (carbon, gray; oxygen, red; nitrogen, red) and are labeled with their residue name. The water
molecules are shown as red spheres, and the hydrogen bonds are shown as dashed red lines. The
docking poses were visualized using PyMOL1.3.
To assess the precise binding pose of compound 19 in the ATP-binding pocket of Hsp90
we performed docking simulations of compound 19 using the N-terminal domain of Hsp90
(PDB code: 2XJX) as a template. The crystal structure of the ATP-binding pocket in the N-
terminal domain of Hsp90 is characterized by a network of highly ordered water molecules
α,
α
α
in the hydrophilic region of the ATP-binding pocket. The docking study indicated that the
resorcinol ring of 19 was located in the hydrophilic region as expected (Figure 3A). The
carbonyl group of 19 formed two hydrogen bonding interactions with the hydroxyl group of
Thr184 and a conserved water (Figure 3B). The hydroxyl group at C-2 of the phenyl ring also
formed bidentate hydrogen bonding interactions with the carboxylate side chain of Asp93 as
well as a conserved water and the hydroxyl group at C-4 interacted with one of the remaining
conserved water molecule through hydrogen bonding. As the isopropyl group at C-5 of 19
had been designed as a hydrophobic binder, the isopropyl group could project into the
hydrophobic region, and it efficiently formed proximal Van der Waals interactions with
lipophilic residues of Val186, Val150, Leu107, and Phe138. The 6,7-Dihydrothieno[3,2-
c]pyridin-5(4H)-yl moiety of compound 19, which was selected as the most potent entry
region binder among analogues in Tables 1 and 2, is nicely positioned in the entry region of
the ATP-binding pocket (Figure 3A). This entry region binder formed proximal Van der
Waals contacts with the hydrophobic side chains of Ala55 and Ile96 as well as the β-carbon
of Asp54.

ACCEPTED MANUSCRIPT
2.6. Biological evaluation of 19
Figure 4. (A) Effect of compound 19 on the cell proliferation of H1975. Cells were
incubated with the indicated concentration of 19 for 0, 1, 2, and 3 days and cell proliferation
was measured using the MTS assay. Data are presented as mean ± SD (n = 4). (B) Effect of
compound 19 on the expression of EGFR, Her2, Met, c-Raf, Akt, Hsp70 and Hsp90. H1975
cells were incubated with the indicated concentrations of 19 for 24 h and the expression of
the Hsp90 client proteins were analyzed using the western blot. Geldanamycin (GA, 1 µM)
and DMSO (D) were employed as a positive and a negative control, respectively.
To investigate the dose- and time-dependent anti-proliferative effect of compound 19,
H1975 cells were treated with compound 19 at various concentrations (0, 0.01, 0.1, 1, 10, 20,
30, 50 µM) for 0, 1, 2, and 3 days. Expectedly, the data indicated that compound 19 afforded a
potent growth-inhibitory effect of H1975 cells in a dose- and time-dependent manner (Figure
4A). Compound 19 almost completely impaired the growth of H1975 cells at the concentration

ACCEPTED MANUSCRIPT
of 1 µM.
To add support to the premise that compound 19 exerts its anti-proliferative effect via
Hsp90 inhibition, we next studied the cellular biomarkers of Hsp90 inhibition. Hsp90 is
responsible for maintaining the stability of EGFR, Her2, Met, c-Raf, and Akt. Hence, an
inhibition of the Hsp90 function will promote the degradation of these client proteins via the
ubiquitin-proteasome pathway. As shown in Figure 4B, the exposure of H1975 cells to 19
caused a significant reduction in the cellular protein levels of EGFR, Her2, Met, c-Raf, and
Akt, while compound 19 upregulated the cellular protein level of Hsp70, which is considered
a cellular hallmark of Hsp90 inhibition. The expression level of non-Hsp90-dependent
protein, β-actin remained unchanged. Collectively, the results indicated that compound 19
disrupted the Hsp90 chaperone function and circumvented the gefitinib-resistance in H1975
cells through a destabilizing of oncogenic proteins such as EGFR, Her2, Met, c-Raf, and Akt.
Figure 5. Compound 19 induces apoptosis in H1975 cells. (A) H1975 cells were treated with
compound 19 (0.5 and 1 µM) for 24 h, and they were then analyzed using flow cytometric
analysis with Annexin V and propidium iodide staining. (B) H1975 cells were incubated with

ACCEPTED MANUSCRIPT
compound 19 (0, 0.05, 0.1, 0.5, 1, and 2 µM) for 24 h and the expression of the apoptotic
biomarker proteins (PARP, cleaved caspase 8, caspase 3, cleaved caspase 3, Bcl-2, and Bax)
was analyzed using the western blot. Geldanamycin (GA, 1 µM) and DMSO (D) were
employed as a positive and a negative control, respectively.
The proliferation of cancer cells can be suppressed by the activation of apoptotic signals,
as it is regulated by the avoidance of the apoptosis mechanism. To determine whether the
anti-proliferative effect of compound 19 on H1975 cells is related to the induction of
apoptosis, we treated H1975 cells with compound 19 (0.5 and 1 µM) for 24 h. The cells were
then stained with Annexin V and propidium iodide, and they were analyzed using
fluorescence-activated cell sorting (FACS). As shown in Figure 5A, the treatment of H1975
cells with 19 remarkably induced early and late apoptosis at the level of 8.5 and 8.3%,
respectively. As an additional approach for the assessment of apoptosis, the effect of
compound 19 on apoptotic-biomarker proteins such as PARP, caspase8, caspase3, B-cell
lymphoma 2 (Bcl-2), and Bcl-2-associated X protein (Bax) was investigated, as shown in
Figure 5B. The exposure of H1975 cell to 0.5 µM of 19 induced the cleavage of PARP,
caspase 8, and caspase 3 that indicated that compound 19 promoted apoptotic cell death even
at the concentration of 0.5 µM. Interestingly, compound 19 did not alter the expression levels
of Bcl-2 and Bax. It is probably because compound 19 activates the apoptosis of H1975 cells
through the extrinsic apoptotic pathway. Nonetheless, the results clearly suggest that
compound 19 efficiently inhibits the growth of H1975 cells through inducing apoptosis.
2.7. Effect of 19 on cytochrome P450s

ACCEPTED MANUSCRIPT
Table 3. Inhibitory potency of compound 19 on specific P450
activities in human liver microsomes
Enzyme activity
P450
IC50 (µM)
Phenacetin O-deethylation
Tolbutamide 4-methylhydroxylation
Omepraozle hydroxylation
1A2
2C9
2C19
2D6
3A
> 50
> 50
> 50
> 50
> 50
Dextromethorphan O-demethylation
Midazolam 1’-hydroxylation
Cytochrome P450 (P450) enzymes are essential for the metabolism of many drugs and
drug metabolism via the cytochrome P450 system has emerged as an important determinant
in the occurrence of drug interactions that can lead to drug toxicities, reduced
pharmacological effect, and adverse drug reactions. Therefore, we investigated the effect of
compound 19 on the catalytic activities of clinically significant human P450s such as 1A2,
2C9, 2C19, 2D6, and 3A in human liver microsomes [26]. To do so, the inhibitory potency of
compound 19 was determined with cytochrome P450 assays in the absence and presence of
compound 19 up to 50 µM final concentration using pooled human liver microsomes. The
assay indicated that compound 19 had no effect on the activities of five major P450 isoforms
and all of the IC50 values for these enzymes were higher than 50 µM. These findings suggest
that clinical interactions between compound 19 and the substrate drugs of these P450
isoforms are not expected.
2.8. In vivo anticancer effect of 19

ACCEPTED MANUSCRIPT
Figure 6. Antitumor activity of compound 19 in a mouse xenograft model. (A) The average
volumes of the H1975-derived tumors from the vehicle and compound 19 treated mice were
plotted 10 days after tumor inoculations. Statistical significance was determined by the
Student’s t-test (*p<0.05, **p<0.01, ***p<0.001). (B) Body weight changes were recorded
during the treatment of vehicle and compound 19. (C) Immunostaining of H&E and akt in
H1975 xenograft tissue sections.
To confirm that in vitro anticancer effect of 19 translates into in vivo model, we
subcutaneously injected NOD-scid IL2Rgammanull mice (NSG) mice with H1975 human
non-small lung cancer cells and compound 19 or vehicle control was intraperitoneally

ACCEPTED MANUSCRIPT
administered to mice bearing subcutaneous tumors daily followed by weekend rest for 2
weeks at 10 mg/kg. Mice injected with vehicle control rapidly developed visible tumors that
grew continuously throughout the course of the treatment period. In contrast, mice injected
with compound 19 exhibited a delayed growth of the tumors that are significantly smaller in
size compared with those of vehicle control (Figure 6A). On the basis of body weight
measurements, compound 19 was well tolerated at the dose level of 10 mg/kg during the
course of the treatment period and no treatment-related deaths were observed. The
morphology of xenotransplanted tumor defined epitheloid tumor features similar to those of
human epithelial lung cancer. In agreement of in vitro results, we also observed decreased
Akt expression in tumor samples treated with compound 19 (Figure 6C).
3. Conclusions
In summary, the discovery and characterization of the novel Hsp90 inhibitor 19 has been
described. Systematic analysis of multiple X-ray co-crystal structures enabled the design and
optimization of the novel class of Hsp90 inhibitor 19. The overall synthetic schemes
developed in this study allowed for an easy preparation of the intermediates and the diverse
analogues. In addition, its practical synthetic route made it readily amenable to large scale
synthesis. Compound 19 demonstrated an exceptionally potent binding affinity to the N-
terminus of Hsp90
that compound 19 tightly occupied the ATP-binding pocket in the N-terminal domain of
Hsp90 , hereby supporting the strong binding affinity of 19 to Hsp90 that was observed in
α in the FP assay (IC50 = 50.3 nM). The docking simulation demonstrated
α
α
the FP assay. The strong enzyme affinity of compound 19 agreeably translated into good
cellular activity, whereby compound 19 inhibited the proliferation of H1975 NSCLC and
Skbr3 breast cancer cell lines with GI50 values of 0.31 µM and 0.11 µM, respectively. The
exposure of cancer cells to compound 19 displayed the characteristic molecular biomarkers of

ACCEPTED MANUSCRIPT
Hsp90 inhibition such as the downregulation of oncogenic proteins (EGFR, Her2, Met, c-Raf,
and Akt) and the upregulation of Hsp70. The results of FACS and western blot analysis
indicated that compound 19 inhibited the growth of H1975 cells through the inducement of
apoptosis. Cytochrome P450 assay indicated that compound 19 had no effect on the activities
of five major P450 isoforms (IC50 > 50 µM for 1A2, 2C9, 2C19, 2D6, and 3A), suggesting
that clinical interactions between compound 19 and these P450 isoforms are not expected.
Moreover, compound 19 displayed a significant in vivo efficacy in a mouse xenograft model
bearing subcutaneous H1975 tumors and the immunostaining of H&E and Akt in xenograft
tissue sections is in a good agreement with the in vitro results. Overall, the attractive
attributes of 19 warranted further studies on ADMET profile and its therapeutic evaluation
for various types of cancer. These studies will be reported in due course.
4. Experimental
4.1. Chemistry
4.1.1. Methyl 2,4-dihydroxybenzoate (6)
2,4-Dihydroxybenzoic acid (10 g, 64.88 mmol) and sulfuric acid (5 mL) in MeOH (40 mL)
was stirred at 100℃ for 24 h, equipped with a refluxing condenser under argon. After
cooling at room temperature, the mixture was concentrated under reduced pressure and added
o
water at 0 C. To filter the resulting white solid was dissolved in ethyl acetate washed with
saturated NaHCO3 solution. The organic layer was dried over Na2SO4, concentrated under
reduced pressure to afford compound 6 in 86% yield. 1H NMR (500 MHz, CD3OD) 7.56 (d, J
13
= 9 Hz, 1H), 6.24 (dd, J = 8.8 Hz, 2.0 Hz, 1H), 6.21 (d, J = 2.5 Hz, 1H), 3.78 (s, 3H). C
NMR (125 MHz, CD3OD) δ 171.7, 165.7, 164.9, 132.7, 109.1, 105.5, 103.4, 52.3.
4.1.2. Methyl 5-chloro-2,4-dihydroxybenzoate (7)

ACCEPTED MANUSCRIPT
A mixture of compound 6 (10.0 g, 59.5 mmol) and sulfuryl chloride (4.30 mL, 59.5 mmol) in
o
CH2Cl2 was stirred at 0 C for 24 h under argon. The mixture was neutralized with 10 %
NaOH to pH 5, concentrated under reduced pressure, and then extracted with ethyl acetate.
The organic layer was washed with saturated NaHCO3 solution three times, dried over
Na2SO4, concentrated under reduced pressure, and purified by column to afford compound 7
1
in 45% yield. Rf = 0.26 (2:8 ethyl acetate: hexane). H NMR(400 MHz, CDCl3) δ 10.83 (s,
13
1H), 7.82 (s, 1H), 6.61 (s, 1H), 5.94 (s, 1H), 3.92 (s, 3H). C NMR (100 MHz, CDCl3) δ
169.6, 162.5, 157.3, 130.3, 111.5, 106.8, 104.3, 52.6.
4.1.3. Methyl 2,4-bis(allyloxy)-5-chlorobenzoate (8)
A mixture of compound 7 (6.09 g, 30.00 mmol), allyl bromide (6.75 ml, 78.02 mmol) and
potassium carbonate (10.78 mL, 78.02 mmol) in DMF was stirred at room temperature for 24
h under argon. The mixture was concentrated under reduced pressure, and then extracted with
ethyl acetate. The organic layer was washed with saturated NaHCO3 solution three times,
dried over Na2SO4 and concentrated under reduced pressure to afford compound 8 in 100%
1
yield. H NMR (400 MHz, CDCl3) δ 7.85 (s, 1H), 7.25 (s, 1H), 6.44 (s, 1H), 6.44-5.95 (m,
13
2H), 5.52-5.40 (m, 2H), 5.31-5.29 (m, 2H), 4.58-4.54 (m, 4H), 3.82 (s, 3H). C NMR (100
MHz, CDCl3) δ 164.7, 158.7, 157.7, 133.0, 132.2, 131.7, 118.1, 117.4, 113.9, 112.6, 99.4,
69.7, 51.7.
4.1.4. 2,4-Bis(allyloxy)-5-chlorobenzoic acid (9)
A mixture of compound 8 (9.43 g, 37.99 mmol) and sodium hydroxide (5 g) in methanol (25
mL) - H2O (25 mL) was stirred at room temperature for 30 h. The mixture was neutralized
with 1N HCl to pH 6 and then extracted with ethyl aceatate three times. The organic layer
was dried over Na2SO4, concentrated under reduced pressure to afford compound 9 in 80%

ACCEPTED MANUSCRIPT
yield. 1H NMR (400 MHz, CDCl3) δ 8.16 (s, 1H), 6.55 (s, 1H), 6.10-5.99 (m, 2H), 5.51-5.35
(m, 2H), 4.77-4.66 (m, 4H). 13C NMR (125 MHz, CDCl3) δ 164.5, 158.9, 157.3, 134.7, 131.7,
130.8, 120.9, 118.9, 116.9, 111.3, 98.9, 71.4, 70.2..
4.1.5. (5-Chloro-2,4-dihydroxyphenyl)(3,4-dihydroisoquinolin-2(1H)-yl)methanone (11a)
A mixture of compound 9 (0.3 g, 1.12 mmol), 1,2,3,4-tetrahydroisoquinoline (0.21 mL, 1.68
mmol), N,N’-dicyclohexylcarboiimide (0.46 g, 2.23 mmol), hydroxybenzotriazole (0.15 g,
1.12 mmol), N,N-diisopropylethylamine (0.20 mL, 1.12 mmol) in 3 mL of DMF was stirred
under microwave irradiation for 3h at 80 oC. The mixture was dissolved in ethyl acetate. The
organic layer was washed with water, dried over Na2SO4, concentrated under reduced
pressure and purified by MPLC to afford intermediate compound. Rf = 0.30 (3:7 ethyl acetate:
hexane). The resulting intermediate compound was stirred under microwave irradiation for 1
o
h at 120 C in the presence of Pd Cl2(PPh3)2 (37 mg) and ammonium formate (300 mg) in 3
mL of THF. The reaction mixture was diluted with ethyl acetate. The organic layer was
washed with water, dried over Na2SO4,concentrated under reduced pressure and purified by
MPLC to afford compound 11a in 66% yield in two steps. Rf = 0.20 (2:3 ethyl acetate:
hexane). 1H NMR (500 MHz, CDCl3) δ 7.35 (s, 1H), 7.23-7.18 (m, 3H), 7.13-7.12 (m, 1H),
13
6.67 (s, 1H), 4.82 (s, 2H), 3.90 (t, J = 6.0 Hz, 2H), 3.00 (t, J = 6.0 Hz, 2H). C NMR (125
MHz, CDCl3) δ 170.7, 160.9, 155.0, 134.3, 132.5, 128.9, 128.5, 127.2, 126.8, 126.5, 110.7,
110.1, 105.3, 48.6, 44.4, 28.9. ESI MS (m/e) 304.07 [M+1]+.
4.1.6. (5-Chloro-2,4-dihydroxyphenyl)(4,5-dihydrothieno[2,3-c]pyridin-6(7H)-yl)methanone
(11b)
A mixture of compound 9 (0.3 g, 1.12 mmol), 4,5,6,7-terahydrothieno[3,2-c] pyridine
hydrochloride (0.29 g, 1.68 mmol), N,N’-dicyclohexylcarboiimide (0.46 g, 2.23 mmol),

ACCEPTED MANUSCRIPT
hydroxybenzotriazole (0.15 g, 1.12 mmol), N,N-diisopropylethylamine (0.20 mL, 1.12 mmol)
o
in 3 mL of DMF was stirred under microwave irradiation for 3h at 80 C. The mixture was
dissolved in ethyl acetate. The organic layer was washed with water, dried over Na2SO4,
concentrated under reduced pressure and purified by MPLC to afford intermediate compound.
Rf = 0.23 (1:3 ethyl acetate: hexane). The resulting intermediate compound was stirred under
o
microwave irradiation for 1 h at 120 C in the presence of Pd Cl2(PPh3)2 (50 mg) and
ammonium formate (500 mg) in 3 mL of THF. The reaction mixture was diluted with ethyl
acetate. The organic layer was washed with water, dried over Na2SO4,concentrated under
reduced pressure and purified by MPLC to afford compound 11b in 27% yield in two steps.
Rf = 0.26 (2:3 ethyl acetate: hexane). 1H NMR (400 MHz, CDCl3) δ 10.26 (s, 1H), 7.32 (s,
1H), 7.16 (d, J = 5.2 Hz, 1H), 6.78 (d, J = 5.2 Hz, 1H), 6.65 (s, 1H), 6.13 (s, 1H), 4.76 (s, 2H),
3.95 (t, J = 6.0 Hz, 2H), 3.03 (t, J = 5.6 Hz, 2H). 13C NMR (125 MHz, CDCl3) δ 171.1, 161.0,
155.1, 133.4, 131.5, 128.4, 124.8, 124.1, 110.7, 110.2, 105.4, 47.5, 44.2, 25.1. ESI MS (m/e)
310.88 [M+1]+.
4.1.7. (5-Chloro-2,4-dihydroxyphenyl)(4-phenylpiperazin-1-yl)methanone (11c)
A mixture of compound 12 (0.17 g, 0.90 mmol), 1-phenylpiperazine (0.21 mL, 1.50 mmol),
1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (0.35 g, 1.80 mmol), hydroxybenzotriazole
(0.12 g, 0.90 mmol), N,N-diisopropylethylamine (0.16 mL, 0.90 mmol) in 3 mL of DMF was
o
stirred under microwave irradiation for 3h at 120 C. The mixture was dissolved in ethyl
acetate. The organic layer was washed with 1N-HCl solution, dried over Na2SO4,
concentrated under reduced pressure and purified by MPLC to afford compound 11c in 14%
yield in two steps. Rf = 0.23 (2:3 ethyl acetate: hexane). 1H NMR (500 MHz, CDCl3) δ
10.13(s, 1H), 7.30-7.26(m, 2H), 6.94-6.93(br, 3H), 6.66(s, 1H), 5.77(s, 1H), 3.88(s, 4H),
3.24(t, J = 5.0 Hz, 4H). 13C NMR (150 MHz, CDCl3) δ 170.3, 160.9, 155.0, 150.7, 129.4,

ACCEPTED MANUSCRIPT
128.6, 120.9, 116.7, 110.2, 110.1, 105.3, 59.5, 49.7, 46.0, 31.0. ESI MS (m/e) 333.10 [M+1]+.
4.1.8. 5-Chloro-2,4-dihydroxy-N-phenylbenzamide (11d)
A mixture of compound 12 (0.30 g, 1.59 mmol), aniline (0.22 mL, 2.39 mmol), 1-ethyl-3-(3-
dimethylaminopropyl)carbodiimide (0.61 g, 3.18 mmol), hydroxybenzotriazole (0.21 g, 1.59
mmol), N,N-diisopropylethylamine (0.28 mL, 1.59 mmol) in 3 mL of DMF was stirred under
o
microwave irradiation for 3h at 120 C. The mixture was dissolved in ethyl acetate. The
organic layer was washed with 1N-HCl solution, dried over Na2SO4, concentrated under
reduced pressure and purified by column to afford compound 11d in 37% yield in two steps.
Rf = 0.23 (3:7 ethyl acetate: hexane). 1H NMR (400 MHz, CD3OD) δ 7.88 (s, 1H), 7.53 (d, J
= 8.4 Hz, 2H), 7.22 (t, J = 7.6 Hz, 2H), 7.01 (t, J = 7.4 Hz, 1H), 6.36 (s, 1H). 13C NMR (100
MHz, CD3OD) δ 166.7, 159.9, 157.8, 137.7, 129.4, 128.4, 128.0, 124.2, 121.5, 121.1, 111.8,
109.0, 103.6. ESI MS (m/e) 286.82 [M+Na]+.
4.1.9. 5-Chloro-2,4-dihydroxy-N-(4-hydroxyphenyl)benzamide (11e)
A mixture of compound 12 (0.30 g, 1.59 mmol), 4-aminophenol (0.26 g, 2.39 mmol), 1-ethyl-
3-(3-dimethylaminopropyl)carbodiimide (0.61 g, 3.18 mmol), hydroxybenzotriazole (0.21 g,
1.59 mmol), N,N-diisopropylethylamine (0.28 mL, 1.59 mmol) in 3 mL of DMF was stirred
o
under microwave irradiation for 3h at 120 C. The mixture was dissolved in ethyl acetate.
The organic layer was washed with 1N-HCl solution, dried over Na2SO4, concentrated under
reduced pressure and purified by MPLC to afford compound 11e in 34% yield in two steps.
Rf = 0.16 (3:7 ethyl acetate: hexane). 1H NMR (400 MHz, CD3OD) δ 7.87 (s, 1H), 7.32 (d, J
= 8.8 Hz, 2H), 6.71 (d, J = 8.8 Hz, 2H), 6.41 (s, 1H). 13C NMR (100 MHz, CD3OD) δ 168.2,
161.4, 159.0, 155.8, 130.7, 130.5, 124.9, 124.8, 116.3, 113.1, 110.3, 105.1. ESI MS (m/e)
280.04 [M+1]+.

ACCEPTED MANUSCRIPT
4.1.10. 5-chloro-2,4-dihydroxybenzoic acid (12)
A mixture of compound 9 (1.98 g, 9.76 mmol) and sodium hydroxide (6 g) in methanol (30
mL) - H2O (30 mL) was stirred at room temperature for 24 h. The mixture was neutralized
with 3N HCl to pH 6 and then extracted with ethyl aceatate three times. The organic layer
was dried over Na2SO4, concentrated under reduced pressure to afford compound 12 in 100%
yield. 1H NMR (400 MHz, CDCl3) δ 7.75 (s, 1H), 6.40 (s, 1H). 13C NMR (100 MHz, CD3OD)
δ 171.5, 163.6, 160.6, 131.3, 113.0, 106.6, 104.4.
4.1.11. 4-(Benzo[d]oxazol-2-yl)-6-chlorobenzene-1,3-diol (15)
A mixture of thionyl chloride (0.2 mL, 2.69 mmol) and DMF (0.21 mL, 2.69 mmol) in DCM
(30 ml) and added compound 9 (0.33 g, 1.22 mmol) in DCM (4mL) was stirred under argon
for 20 min. After added 2-aminophenol (0.13g, 1.22mmol) in DCM, pyridine (0.59 mL, 7.32
mmol) was added dropwise and stirred for 12h. The mixture was dissolved in DCM. The
organic layer was washed with 1N-HCl solution, dried over Na2SO4, concentrated under
reduced pressure and purified by MPLC to afford intermediate compound. Rf = 0.33 (3:7
ethyl acetate: hexane). The resulting intermediate compound was stirred under microwave
o
irradiation for 30 min at 120 C in the presence of Pd Cl2(PPh3)2 (20 mg) and ammonium
formate (200 mg) in 3 mL of THF. The reaction mixture was diluted with ethyl acetate. The
organic layer was washed with water, dried over Na2SO4,concentrated under reduced pressure
and purified by MPLC to afford compound 15 in 20% yield in two steps. Rf = 0.23 (3:7 ethyl
acetate: hexane). 1H NMR (400 MHz, CDCl3) δ 7.98 (s, 1H), 7.70-7.68 (m, 1H), 7.59-7.56 (m,
1H), 7.37-7.24 (m 2H), 6.76 (s, 1H). 13C NMR (100 MHz, CDCl3) δ 161.9, 159.4, 155.5,
148.9, 139.9, 127.0, 125.3, 125.1, 119.1, 111.7, 110.6, 105.0, 104.5. ESI MS (m/e) 284.19
[M+Na]+.