Design, synthesis, and biological evaluation of truncated deguelin derivatives as Hsp90 inhibitors

Article abstract of DOI:10.1016/j.ejmech.2019.02.014

A series of novel B[sbnd] and C-rings truncated deguelin derivatives have been designed and synthesized in the present study as heat shock protein 90 (Hsp90) inhibitors. The synthesized compounds exhibited micromolar antiproliferative potency toward a pan

Full text of DOI:10.1016/j.ejmech.2019.02.014

Accepted Manuscript
Design, synthesis, and biological evaluation of truncated deguelin derivatives as
Hsp90 inhibitors
Hong Yao, Feijie Xu, Guangyu Wang, Shaowen Xie, Wenlong Li, Hequan Yao, Cong
Ma, Zheying Zhu, Jinyi Xu, Shengtao Xu
PII:
S0223-5234(19)30124-2
DOI:
https://doi.org/10.1016/j.ejmech.2019.02.014
EJMECH 11106
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 22 November 2018
Revised Date: 24 January 2019
Accepted Date: 4 February 2019
Please cite this article as: H. Yao, F. Xu, G. Wang, S. Xie, W. Li, H. Yao, C. Ma, Z. Zhu, J. Xu, S. Xu,
Design, synthesis, and biological evaluation of truncated deguelin derivatives as Hsp90 inhibitors,
European Journal of Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.02.014.
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
Design, synthesis, and biological evaluation of truncated
deguelin derivatives as Hsp90 inhibitors
Hong Yaoa, Feijie Xua, Guangyu Wanga, Shaowen Xiea,Wenlong Lia, Hequan Yaoa,
Cong Mab, Zheying Zhuc, Jinyi Xua,*, and Shengtao Xua,*
aState Key Laboratory of Natural Medicines and Department of Medicinal Chemistry, China
Pharmaceutical University, 24 Tong Jia Xiang, Nanjing 210009, P. R. China
bState Key Laboratory of Chemical Biology and Drug Discovery, and Department of Applied
Biology and Chemical Technology, The Hong Kong Polytechnic University, Kowloon, Hong
Kong
cDivision of Molecular Therapeutics & Formulation, School of Pharmacy, The University of
Nottingham, University Park Campus, Nottingham NG7 2RD, U.K.
*Correspondence should be addressed to:
Biology: S.T. Xu (cpuxst@cpu.edu.cn)
Chemistry: J.Y. Xu (jinyixu@china.com)
Abstract:
A series of novel B- and C-rings truncated deguelin derivatives have been designed
and synthesized in the present study as heat shock protein 90 (Hsp90) inhibitors. The
synthesized compounds exhibited micromolar antiproliferative potency toward a
panel of human cancer cell lines. Their structure-activity relationships (SARs) were
investigated in a systematic manner. Compound 21c was identified to have high
Hsp90 binding potency (60 nM) and caused degradation of client proteins through
ubiquitin proteasome system. Further biological studies showed that compound 21c
induced a dose-dependent S and G2-phase cell cycle arrest on human breast cancer
MCF-7 cells. Flow cytometry and Western blot analyses confirmed that compound
21c caused apoptosis of MCF-7 cells. In addition, compound 21c showed much potent
inhibition on the migration and invasion of MCF-7 cells. Taken together, these results
suggest that 21c might be a promising lead compound for further development of
1

ACCEPTED MANUSCRIPT
Hsp90 inhibitors.
Keywords: Heat shock protein 90; deguelin; structure simplification; anticancer
1. Introduction
Heat shock protein 90 (Hsp90) is attracting increasing attention as a promising target
for cancer treatment in recent years[1]. As a constitutively expressed chaperon, Hsp90
plays a central role in assembly, folding, trafficking, and degradation processes of
various client proteins involved in pathways of cell proliferation, angiogenesis,
invasion and metastasis, and many of them are oncogenic proteins[2]. Additionally,
Hsp90 is often overexpressed 2-10 fold higher in cancer cells than that in normal
healthy cells[3]. Inhibition of Hsp90 by small molecules would result in degradation
of the client proteins and suppress the growth and proliferation of cancers[4].Besides,
targeting Hsp90 results in the regulation of multiple pathways simultaneously, and
may also overcome the notorious cancer resistance issue due to its potential for
combinatorial targeting of multiple oncogenic protein pathways[5]. These properties
of Hsp90 make it an ideal and selective target for the treatment of cancers.
Over the past decades, a number of Hsp90 inhibitors have been reported. The
earlier Hsp90 inhibitors are developed from natural products, such as geldanamycin (1)
and radicicol (2), which are good lead compounds for the development of Hsp90
inhibitors[6]. 17-AAG (3), a semi-synthetic geldanamycin analogue designed to
improve activity and reduce toxicity, once progressed to phase II clinical trials to treat
patients with metastatic melanoma [7]. Then, various totally synthesized inhibitors of
Hsp90 were developed with the aid of structure based drug design strategy[8]. To date,
approximately more than 20 different compounds are in clinical studies, however,
none Hsp90 inhibitors have been approved for use in clinical[9]. Several clinical trials
of Hsp90 inhibitors have been discontinued due to effectiveness and toxicity.
Consequently, there is an urgent need to develop new Hsp90 inhibitors.
2

ACCEPTED MANUSCRIPT
Figure 1. Chemical structures of some Hsp90 inhibitors.
Deguelin (4), a rotenoid isolated from the African plant Mundulea sericea
(Leguminosae) and other plants, has been reported as a Hsp90 inhibitor with potent
apoptotic and antiangiogenic effects on transformed cells and a variety of cancer
cells[10]. Mechanism studies revealed that deguelin competes with ATP for binding
with Hsp90, leading to the decrease of a series of Hsp90 client proteins, including p53,
CDK4, MAPK-1/2, Akt, and HIF-1α[11]. Despite of promising anticancer activity of
deguelin, the potential toxicity (Parkinson’s disease-like syndrome) of this compound
in rats impended its direct application in clinic[12]. The complex structure of deguelin
containing 5-fused rings designated as A-E rings (Figure 1), also raised a challenge
for their chemical synthesis for structural modifications[13].
Tuning the structure of natural products plays an important role in transforming
natural products to drugs[14]. For some natural products with structurally complex
frameworks, truncating the structure may lead to simpler and synthetically more
accessible fragment-like scaffolds. These core structures feature from NPs can also be
used for further exploration of small druggable molecules[15]. Ring-truncated
scaffolds of deguelin have been reported previously[16], however, the detailed
structure-activity relationships of truncated deguelin derivatives have not been
investigated intensively. In this study, twenty four structurally simpler truncated
deguelin derivatives were designed as Hsp90 inhibitors by truncating B- and C-rings
from deguelin and further modifying the A-, D- and E-rings (Figure 2). According to
the previous reports, presence of the B- and C-rings of deguelin restricted the
flexibility of deguelin, limiting the best conformation with which the deguelin
3

ACCEPTED MANUSCRIPT
interacted with Hsp90[17]. Our molecular docking study also implied the
conformational restriction of deguelin when docking into the active site of Hsp90
(Figure 4B). Besides, the chemical construction of B- and C-rings of deguelin is the
hardest step in total synthesis of this natural compound, truncating B- and C-rings of
deguelin would be in favorable of chemical synthesis to develop its derivatives as
Hsp90 inhibitors [18].
According to the previous studies, the chromene skeleton (D and E rings) of deguelin
was important for its activity, and the more extensive changes is inappropriate.
Because the previous work implied that the methoxyl group was important to the
activity of degelin. The modifications at R2 and carbonyl group were performed to
explore the SAR of substitutes of different size and length.
In this work, concise synthetic methodologies were further developed to prepare the
novel B- and C-rings truncated deguelin derivatives based on our previous studies
[19]. The antiproliferative activities of the synthesized compounds were evaluated and
the mechanism of the most potent compound 21c was also investigated.
Figure 2. Design strategy of truncated derivatives of deguelin.
2. Results and discussion
2.1. Chemistry
4

ACCEPTED MANUSCRIPT
As shown in Scheme 1, the B- and C-rings truncated deguelin derivatives bearing
phenolic hydroxy group at chromene moiety were synthesized via Friedel-Crafts
acylation in three steps, which was much more concise than previously reported
routes[12]. Briefly, acylates 7 were synthesized via Friedel-Crafts acylation using
various substituted phenylacetic acids or phenylpropionic acid and resorcinol as
starting materials with boron trifluoride diethyl etherate as the solvent. Chromene
skeleton was constructed via selective propargylation of resorcinol 7 with
3-chloro-3-methyl-1-butyne and a subsequent regioselective Claisen rearrangement.
Thus, truncated deguelin derivatives 9a-f were obtained in three steps with acceptable
total yields.
Scheme 1. Synthesis of truncated deguelin derivatives 9a-fa
aReagents and conditions: (a) BF3•Et2O, N2, 90oC, 80-90%; (b) 3-Chloro-3-methyl-1-butyne, KI,
CuI, K2CO3, CH3CN, rt, 60-75%; (c) N,N-Dimethylaniline, 150 oC, 85-95%.
With 9a-c as starting materials, target compounds bearing substitutions of different
sizes (methyl, acetyl, and benzyl) at phenolic hydroxy group were obtained. As shown
in Scheme 2, in the presence of 1 equiv of methyl iodide, the phenolic hydroxyl group
of 9a-c was methylated preferentially to afford target compounds 10a, 10c, and 11,
respectively. However, in the presence of more than 2 equiv of methyl iodide, the α
position of carbonyl group was also methylated and gave compounds 10b and 10d.
The double bond of 2H-chromene 11 was reduced in H2 atmosphere to provide
chroman12 with a 95% yield. On the other hand, the carbonyl group of 11 was
5

ACCEPTED MANUSCRIPT
converted to hydroxyl group using NaBH4 as the reductant, affording compound 13.
The following elimination of hydroxyl at 60 oC region-selectively provided
trans-alkenyl compound 14 in 90% yield. Acetylation of compound 9b with
anhydride afforded compound 15, while benzylation of compound 9b with excessive
benzyl bromide gave dibenzylated compound 16.
Scheme 2. Synthesis of truncated deguelin derivatives 10-16a
aReagents and conditions: (a) CH3I (1 equiv.), acetone, reflux, 80%; (b) H2, Pd/C, MeOH, 95%;(c)
CH3I (2-3 equiv.), acetone, reflux, 80%; (d) NaBH4, MeOH, 0oC, 95%; (e) AcOH, 60oC,90%; (f)
(CH3CO)2O, Et3N, DCM, rt, 90%; (g) BnBr, K2CO3, MeCN, reflux, 80%.
Although the preparation of these truncated deguelin derivatives using
Friedel-Crafts acylation is very effective, some target compounds could not be
6

ACCEPTED MANUSCRIPT
obtained with this methodology due to the lack of commercially available
polysubstituted phenylacetic acids. As shown in Scheme 3, key intermediate
acylchromene 20 was firstly synthesized from dihydroxyacetophenone (17) using the
Claisen rearrangement described above. Subsequently compound 20 reacted with
various substituted bromobenzene to afford target compounds 21a-e in moderate
yields. Synthesis of chroman 22 from chromene 21d was accomplished in 95% yield
by reduction with H2 using Pd/C as a catalyst. The aldol condensation of ketone 19
and benzaldehydes under the condition of NaH as base catalyst, THF as solvent
provided target compounds 23a-b in moderate yield.
Scheme 3. Synthesis of truncated deguelin derivatives 21-23a
aReagents and conditions: (a) 3-Chloro-3-methyl-1-butyne, KI, CuI, K2CO3, CH3CN, rt, 60%; (b)
N,N-Dimethylaniline,150oC, 95%; (c) CH3I, acetone, reflux, 80%; (d) Pd2(dba)3, DPE-phos,
t-BuOK, THF, 70oC, 70%; (e) H2, Pd/C, MeOH, 95%; (f) NaH, THF, rt, 65%.
7

ACCEPTED MANUSCRIPT
2.2. In vitro antiproliferative activities of truncated deguelin derivatives
All compounds were firstly evaluated for their preliminary antiproliferative activities
against human non-small cell lung cancer H1299 cells. Deguelin and Hsp90 inhibitor
AT13387 were used as the positive controls. As shown in Table 1, the number and
position of methoxy group in A-ring was vital to the antiproliferative effects of these
compounds. Replacement of methoxy groups (9b) with chlorine atoms (9d) reduced
the inhibition rate significantly at 10 µM. The inhibition rates of compounds 9a, 9c,
9b, and 9e increased gradually along with the increase of methoxy groups at A-ring.
Trimethoxybenzene seems to be the most favorable moiety for this region (9e, 21c,
21d, and 22). Modifications on the linker between A- and D-ring was unfavorable for
improving the activity of these derivatives, both increasing the length and reducing
carbonyl resulted in analogues with less potent activity. By comparing the inhibition
rates at 10 µM between 9c-e and 10c, 11, 21d, a conclusion could be draw that
hydroxyl and methoxyl were all tolerated at D-ring, but large substitutions at D-ring
decreased the activity significantly (15 and 16). In addition, reducing the double bond
of E-ring resulted in increased activity (12 and 22).Furthermore, compounds 9e, 11,
12, 21c-d and 22, which exhibited potent antiproliferative activities in preliminary
screening, were selected for IC50 evaluations against H1299, MCF-7 and A549 cell
lines. As shown in Table 2, all these compounds exhibited comparable
antiproliferative activities to deguelin against three cancer cell lines. Among these
compounds, 21c with a 2,4,5-trimethoxybenzene group was the most potent one with
an IC50 value of 2.51 µM on MCF-7 cells.
Table 1. Inhibition rates of compounds in H1299 cell line
Inhibition rate (%)a
Compounds
0.1 µM
3.9±1.9
4.8±2.9
6.1±1.8
1.0 µM
7.8±1.3
8.2±1.6
7.1±2.8
10.0 µM
15.1±3.8
45.3±4.2
16.1±1.9
9a
9b
9c
8

ACCEPTED MANUSCRIPT
8.3±2.7
9.7±2.4
3.6±0.9
5.6±1.2
4.5±1.8
3.6±0.4
7.1±1.0
5.1±1.4
7.2±1.5
6.5±0.7
5.2±2.0
4.1±2.0
2.1±1.2
7.1±1.9
4.7±1.2
15.3±1.8
9.6±1.7
3.5±0.7
11.0±1.3
5.3±0.8
3.8±0.2
9.9±1.6
94.78±10.6
9.1±2.3
19.7±1.7
8.5±0.8
9.3±1.3
6.5±1.2
7.3±1.8
10.8±1.3
10.7±1.4
12.4±1.4
10.8±2.0
8.6±1.8
6.2±0.6
2.5±0.8
10.2±2.8
10.4±1.2
40.3±7.8
27.4±2.6
6.5±1.0
30.6±4.3
6.4±2.1
8.7±2.1
28.5±2.7
96.8±7.3
11.5±2.3
49.0±6.5
30.1±4.8
11.3±1.8
15.5±2.0
10.6±1.5
13.3±1.9
47.3±4.9
49.3±5.3
19.8±2.0
14.4±1.7
19.5±2.6
9.9±0.8
9d
9e
9f
10a
10b
10c
10d
11
12
13
14
15
16
40.1±4.7
38.2±4.3
72.1±8.0
58.3±4.8
28.1±3.9
67.5±4.9
10.7±2.0
12.3±1.5
65.3±7.9
95.7±9.2
21a
21b
21c
21d
21e
22
23a
23b
Deguelin
AT13387b
aEach data represents mean±SD of three independent experiments
bHsp90 inhibitor AT13387 was used as the positive control
Table 2. Hsp90 binding activity and cell antiproliferative effects of representative
compounds.
IC50 (µM)a
Compounds
9

ACCEPTED MANUSCRIPT
Binding
H1299
MCF-7
A549
Hsp90 (FP)b
0.43
0.09
0.15
0.06
0.63
0.28
1.63
0.02
9e
11
12.03±0.95
11.83±0.96
12.61±1.12
3.40±0.51
7.20±0.87
5.72±0.67
5.67±0.35
0.37±0.06
11.18±0.63
6.88±0.72
7.86±0.84
2.51±0.28
5.08±0.34
5.22±0.48
8.17±0.68
0.29±0.08
13.05±1.14
6.00±0.53
7.20±0.92
3.46±0.33
6.15±0.71
5.33±0.57
6.05±0.59
0.44±0.03
12
21c
21d
22
Deguelin
AT13387c
aMTT methods, cells were incubated with indicated compounds for 72h, the values are the means
of three independent experiments;
bBinding to N-terminal domain of Hsp90 was determined by a fluorescence polarization (FP)
assay;
cHsp90 inhibitor AT13387 was used as the positive control.
Figure 3. Structure activity relationships of novel anti-tubulin compounds.
2.3.Compound 21c directly bound to Hsp90
10

ACCEPTED MANUSCRIPT
In order to confirm whether these compounds could bind to Hsp90, a fluorescence
polarization (FP) assay was performed firstly. As shown in Table 2, all the tested
target compounds (9e, 11, 12, 21c, 21d, and 22) showed appreciable binding affinity
(0.06-0.43 µM) to recombinant human Hsp90α protein, which were relatively less
potent to reference compound AT13387 (0.02 µM), but possessed much higher
binding affinity to Hsp90 than deguelin (1.63 µM). Although compound 21c showed
comparable affinity (0.06 µM) to Hsp90 than AT13387 (0.02 µM), the IC50 values of
21c (2.51-3.46 µM) were much higher than that of AT13387 (0.29-0.44 µM) against
three cancer cell lines. Besides, the structural similarity of compounds 11 and 21c
resulted similar binding affinity to Hsp90 (0.09 µM for 11 and 0.06 µM for 21c), but
the cytotoxicity of compound 11 was much lower than compound 21c, we presumed
that the reason of this inconsistence might be complex system in a cell.
Next, a monoclonal antibody that specifically bound to Hsp90 was used to
immunoprecipitate Hsp90 and then detected whether 21c could interfere with this
process. The results showed that 21c blocked the binding of Hsp90 with its antibody
in a dose-dependent manner (data not shown). In order to understand the more precise
binding mode of compound 21c with Hsp90, docking simulations were performed
using the complex between the N-terminal domain of Hsp90 and 21c (PDB
code:5J2X) as a template. The active site of the Hsp90 N-terminal was divided into
S1, S2, and S3 three pockets, and deguelin only occupied S1 and S3 pockets (Figure
4A). In addition, there was over contact between deguelin and S3 pocket due to
conformational restriction (Figure 4B, deguelin shown in cyan). Truncating B- and C-
rings of deguelin relieved the conformational restriction, as shown in Figure 4B, the
most potent compound 21c (shown in pink) was inserted deeply into the S2 pocket
and occupied S1, S2, and S3 pockets simultaneously. Interestingly, the SARs of these
truncated deguelin derivatives summarized above can be well explained by the
binding pattern taken by 21c in Figure 4D. One methoxyl at A-ring and carbonyl
group formed hydrogen bonds to the residue LYS58, oriented the trimethoxybenzene
moiety to occupy S2 pocket. The trimethoxybenzene group also formed a pi-cation
11

ACCEPTED MANUSCRIPT
interaction with residue LYS58. Excessive size of substitutions at phenolic hydroxyl
group of D-ring led to larger shifts, resulting in loss of the key hydrogen bond to the
residue LYS58. Taken together, these results suggested that compound 21c could bind
to Hsp90 directly, and this may provide useful information for further design of novel
truncated deguelin Hsp90 inhibitors.
Figure 4. Molecular modeling analysis of deguelin and compound 21c. (A) The
predicted binding model of deguelin to Hsp90 (PDB code: 5J2X) catalytic site. The
carbon atoms of deguelin are shown in cyan. The active site of the Hsp90 N-terminal
is divided into S1, S2, and S3 three pockets, and deguelin occupies S1 and S3 pockets.
(B) Docking poses of deguelin and 21c with Hsp90. The carbon atoms of deguelin are
shown in cyan and the carbon atoms of 21c are shown in pink. Deguelin has over
contact to S3 pocket due to conformational restriction. (C) Docking pose of
compound 21c, key amino acid residues within the binding site are rendered in cyan
capped stick. Red dashed lines are hydrogen bonding interactions. (D) The 2D
interaction diagram of 21c. Hydrogen bonds were shown as violet lines and pi-cation
interaction is indicated as magenta line.
12

ACCEPTED MANUSCRIPT
2.4. Compound 21c induced the degradation of Hsp90 client proteins.
Hsp90 inhibitors often cause rapid decline in client protein activities, and the
degradation of client proteins through ubiquitin proteasome system[20]. Protein
kinase B (PKB, also known as Akt) is the client protein of Hsp90, in order to
investigate whether compound 21c could decrease the bind of Akt to Hsp90 and
increase the ubiquitin-mediated degradation of Akt, co-immunoprecipitate (Co-IP)
assay was performed. As shown in Figure 5A, the binding of Akt to Hsp90 could be
clearly detected through Co-IP, however, in MCF-7 cells treated with 4 µM 21c, a
decline in Hsp90-binded Akt was observed. Further detection of ubiquinated Akt in
21c treated cells confirmed that 21c could bind to Hsp90 and mediate the degradation
of Hsp90 client proteins through ubiquitin proteasome system (Figure 5B). Further
Western blotting assay was performed to identify the inhibitory profile of 21c on
Hsp90 client proteins. The results showed that 21c significantly downregulated the
levels of Hsp90 clients p-mTOR, p-MAPK, p-AKT, HIF-1α, and survivin in a
dose-dependent manner (Figure 5c).
Figure 5. Compound 21c triggered degradation of multiple Hsp90 client proteins in a
concentration dependent manner. (A) Inhibition of Hsp90 and Akt binding in vitro by
compound 21c (4 µM); (B) Induced degradation of Akt protein by compound 21c (4
µM); (C) Western-blot analysis of the expression levels of several client proteins of
Hsp90 after incubation with 21c in MCF-7 cells.
13

ACCEPTED MANUSCRIPT
2.5.Compound 21c inhibited cell growth by inducing S and G2-phase cell cycle
arrest.
To determine whether inhibition of cell proliferation was associated with cell cycle
arrest, MCF-7 cells were treated with vehicle or various concentrations of 21c and the
relative DNA content of cells were then analyzed by flow cytometry. The results
showed that 21c-treated MCF-7 cells were arrested at the S and G2 phase. As shown
in Figure 6, MCF-7 cells treated with 4 µM 21c for 48 h had 29.6% cells in G2 phase,
as compared with 16.6% in the control. Conversely, the amount of cells at the G1
phase significantly decreased from 60.6% (control) to 40.7% (treated with 4 µM 21c).
Figure 6. Effects of compound 21c on cell cycle distribution of MCF-7 cells. Cells
treated with DMSO or indicated concentrations of 21c for 36 h were analyzed for
DNA content by flow cytometry.
2.6. Induction of apoptosis by compound 21c
In addition to cell cycle arrest, the overall cell growth inhibition induced by 21c could
be caused by increase of apoptosis. Two separate assays were used to investigate the
induction of apoptosis in MCF-7 cells by 21c. First, induction of apoptosis as a results
of treatment of 21c was assessed by Hoechst 33258 staining. MCF-7 cells treated with
vehicle or various concentrations of 21c for 36 h were stained with Hoechst 33258
and the cell morphology was visualized by fluorescent microscopy. Characteristic
features of apoptosis such as rounding of cells and nuclei, condensation and
14

ACCEPTED MANUSCRIPT
peripherization of chromosome were clearly observed in MCF-7 cells after treatment
with 21c (Figure 7).
Figure 7. Hoechst 33258 staining of cells treated with 21c. Cell morphological
alterations and nuclear changes associated with MCF-7 cells after incubation with 21c
were assessed by staining with Hoechst 33258 and visualized by fluorescence
microscopy, bars denote 50 µM.
To further confirm whether 21c could induce apoptosis in MCF-7 cells, vehicle- or
21c-incubated MCF-7 cells were stained with Annexin V and PI. As shown in Figure
8A, incubation with 21c induced a concentration-dependent (0.5, 1.0, 2.0 and 4.0 µM)
increase (from 13.6% to 71.8%) in both the early and late stage apoptosis of MCF-7
cells.Besides, the expression of apoptotic proteins were detected to determine whether
caspase-mediated apoptotic pathway was involved in 21c-treated cancer cells.
Consistent with the results obtained from deguelin in previous reports, a marked
15

ACCEPTED MANUSCRIPT
decrease in the level of Bcl-xL and increase in the level of cleaved caspase 3 were
detected in the cells treated with 21c[21].
Figure 8. Effect of compound 21c on apoptosis in MCF-7 cells. (A) Cells were
treated with DMSO or various concentrations of 21c for 36 h; (B) Histograms display
the percentage of cell distribution. Data were represented as mean±SD of three
independent experiments. *p<0.05, **p<0.01, and ***p<0.001 vs control group; (C)
The expression of Bcl-xL and cleaved caspase-3 were analyzed by Western-blot
analysis; (D) Histograms display the density ratios of Bcl-xL and cleaved caspase 3 to
GAPDH. Data was represented as mean±SD of three independent experiments.
*p<0.05, ***P<0.001, control compared with 21c treated cells.
2.7. Inhibitory effects of 21c on migration and invasion of MCF-7 cells.
Tumor invasion through extracellular matrix is an essential step in the tumor
metastasis[22]. PI3K-Akt signaling has been shown to regulate cancer cell growth and
invasion via increased motility and MMPs production[23]. In addition, deguelin has
been reported to inhibit the invasion of cancer cells, thus, in the present study, effects
of compound 21c on the migration and invasion of MCF-7 cells were examined[24].
16

ACCEPTED MANUSCRIPT
As shown in Figure 9A, the wound healing significantly decreased (indicative of
decreased cell migration) in 21c-treated MCF-7 cells as compared to control cells,
indicating that the migration of cells was significantly inhibited by 21c. Besides, the
mean percentage value of the invaded cells through Matrigel-coated filters without
test compound was 172.1±8.3 after 24 h incubation, compound 21c inhibited invasion
of MCF-7 cells in a concentration dependent manner (Figure 9D). Although only a
weak antiproliferative effect of 21c was observed at a concentration of 1 µM (Table 1),
it showed much stronger inhibition on the invasion of MCF-7 cells, and the invaded
cells were significantly decreased to 63.2±3.4 by 1 µM of 21c after 24 h incubation
(Figure 9D).
Figure 9. Compound 21c decreased the migration and invasion of breast cancer cells.
(A) Wound healing assay of MCF-7 cells treated with various concentrations of 21c;
17

ACCEPTED MANUSCRIPT
(B) Represent photomicrographs showing the invasive capability of MCF-7 cells was
impaired by 21c in a dose-dependent manner; (C) The wounding area was quantified;
(D) The number of migrating cells was determined by counting the stained cells and
histogram. ***P<0.001, **P<0.01 compared with control group.
3. Conclusion
In summary, twenty four structurally simpler derivatives of natural product deguelin
were obtained and identified as novel Hsp90 inhibitors by the optimization of
chemical structure-truncating strategy. Some of the truncated deguelin derivatives
displayed same or enhanced anti-proliferative efficacy on a panel of human cancer
cell lines, indicating that simplification of B- and C-rings of deguelin retained its
anticancer efficacy. The most potent compound 21c, was 3-fold more potent than
parent molecule deguelin against MCF-7 cancer cells. Docking studies explained the
high binding affinity of 21c to Hsp90 due to the relieved conformational restriction by
truncating B- and C- rings of deguelin. Compound 21c was identified to have high
Hsp90 binding potency (60 nM) and caused degradation of client proteins through
ubiquitin proteasome system. In addition, the molecular mechanism study of 21c on
the suppression of human breast cancer MCF-7 cell growth revealed that compound
21c caused cancer cells arrest at G2/M phase of the cell cycle, and it significantly
increased cell apoptosis, which was illustrated by chromatin condensation,
externalization of phosphatidylserine and a marked decrease in the level of Bcl-xL
and increase in the level of cleaved caspase 3. In additional, compound 21c showed
much potent inhibition on the migration and invasion of MCF-7 cells. Collectively,
our results demonstrated that the newly developed truncated deguelin derivative 21c
possessed significant anticancer efficacy, and has the potential to be further
development of Hsp90 inhibitors as anti-tumor agent.
4. Experimental section
4.1. General information
All commercially available reagents were used without further purification.
18

ACCEPTED MANUSCRIPT
Anhydrous solvents were dried through routine protocols. Flash column
chromatography was carried out on 200-300 mesh silica gel (Qingdao Haiyang
Chemical, China). Reactions were monitored by thin-layer chromatography (TLC) on
0.25 mm silicagel plates (GF254) and visualized under UV light. 1H NMR and 13C
NMR spectra were recorded with a Bruker AV-300 spectrometer (Bruker Company,
Germany) in the indicated solvents (CDCl3 or DMSO-d6, TMS as internal standard):
the values of the chemical shifts are expressed in δ values (ppm) and the coupling
constants (J) in Hz. Low-and high-resolution mass spectrums (LRMS and HRMS)
were measured on Finnigan MAT 95 spectrometer (Finnigan, Germany).
4.2. Synthesis of target compounds
4.2.1. 1-(2,4-Dihydroxyphenyl)-2-(3,4-dimethoxyphenyl)ethan-1-one (7b)
3,4-Dimethoxyphenylacetic acid (1.0 g, 5.1 mmol, 1.0 equiv.) and resorcinol (561 mg,
5.1 mmol, 1.0 equiv.) were added into 15 mL boron trifluoride ethyl etherate. The
mixture was stirred at 90
for 1.5 h under the protection of nitrogen. 30 mL cold
water containing 3.0 g sodium acetate was added into the reaction solution to
precipitate the product. The mixture was filtered and the solid was purified by ethyl
acetate to obtained 7b as a yellow solid (1.17 g, 80%). 1H NMR (300 MHz, DMSO-d6)
δ 12.38 (s, 1H), 10.01 (s, 1H), 8.03 (d, J = 8.8 Hz, 1H), 6.91 (d, J = 2.0 Hz, 1H), 6.89
(d, J = 8.2 Hz, 1H), 6.79 (dd, J = 8.2, 2.0 Hz, 1H), 6.75 (d, J = 2.4 Hz, 1H), 6.71 (dd,
J = 8.9, 2.4 Hz, 1H), 4.25 (s, 2H), 3.72 (s, 3H), 3.72 (s, 3H); ESI-MS m/z 311.1
[M+Na]+.
4.2.2. 2-(3,4-Dimethoxyphenyl)-1-(2-hydroxy-4-((2-methylbut-3-yn-2-yl)oxy)phenyl)
ethan-1-one (8b)
Compound 7b (1.17 g, 4.06 mmol, 1.0 equiv.), 3-chloro-3-methyl-1-butyne (537 µL,
4.87 mmol, 1.2 equiv.), copper iodide (77 mg, 0.4 mmol, 0.1 equiv.), potassium iodide
(67 mg, 0.4 mmol, 0.1 equiv.) and potassium carbonate (673 mg, 4.87 mmol, 1.2
equiv.) were added into 20 mL acetonitrile. The mixture was stirred at room
temperature for 12 h. The solvent was removed at reduced pressure, and the residue
19

ACCEPTED MANUSCRIPT
was treated with 10 mL water and extracted with ethyl acetate (3 × 30 mL). The
combined organic phases were dried (Na2SO4), and the solvent was removed at
reduced pressure. The crude residue was purified by column chromatography with
1
petroleum/ethyl acetate (10:1) to give 8b as a white solid (0.72 g, 50%). H NMR
(300 MHz, DMSO-d6) δ 12.38 (s, 1H), 8.03 (d, J = 8.8 Hz, 1H), 6.91 (d, J = 2.0 Hz,
1H), 6.89 (d, J = 8.2 Hz, 1H), 6.79 (dd, J = 8.2, 2.0 Hz, 1H), 6.75 (d, J = 2.4 Hz, 1H),
6.71 (dd, J = 8.8, 2.4 Hz, 1H), 4.25 (s, 2H), 3.83 (s, 1H), 3.72 (s, 3H), 3.72 (s, 3H),
1.66 (s, 6H); ESI-MS m/z 355.2 [M+H]+.
4.2.3.
2-(3,4-Dimethoxyphenyl)-1-(5-hydroxy-2,2-dimethyl-2H-chromen-6-yl)ethan-1-one
(9b)
Compound 8b (0.72 g) was dissolved in 5 mL N,N-dimethylaniline, and the mixture
was refluxed at 130 for 24 h. After the reaction was completed, 10 mL concentrated
HCl was added and extracted with ethyl acetate (3 × 30 mL). The combined organic
phases were dried (Na2SO4), and the solvent was removed at reduced pressure. The
crude residue was purified by column chromatography with petroleum/ethylacetate
(10:1) to give 9b as a white solid (0.68 g, 95%). 1H NMR (300 MHz, CDCl3) δ 12.92
(s, 1H), 7.61 (d, J = 8.8 Hz, 1H), 6.76 (m, 3H), 6.66 (d, J = 10.1 Hz, 1H), 6.30 (d, J =
13
8.8 Hz, 1H), 5.53 (d, J = 10.1 Hz, 1H), 4.10 (s, 2H), 3.82 (s, 6H), 1.41 (s, 6H); C
NMR (75 MHz, CDCl3) δ 201.8, 159.6, 159.3, 148.5, 147.6, 130.9, 127.7, 126.3,
121.0, 115.2, 112.5, 111.8, 110.8, 108.9, 107.9, 77.3, 55.4, 43.8, 27.9; HRESI-MS m/z
calcd for C21H22O5 355.1540 [M+H]+, found 355.1557.
4.2.4.
1-(5-Hydroxy-2,2-dimethyl-2H-chromen-6-yl)-2-(4-methoxyphenyl)ethan-1-one (9c)
Compound 9c was prepared as the same procedures of 9b with the commercially
available 3,4-dimethoxyphenylacetic acid as the starting material. 1H NMR (300 MHz,
CDCl3) δ 12.89 (s, 1H), 7.57 (d, J = 8.9 Hz, 1H), 7.14-7.06 (d, J = 8.1 Hz, 2H),
6.84-6.76 (d, J = 8.1 Hz, 2H), 6.63 (d, J = 10.1 Hz, 1H), 6.26 (d, J = 8.9 Hz, 1H), 5.50
20

ACCEPTED MANUSCRIPT
(d, J = 10.1 Hz, 1H), 4.06 (s, 2H), 3.72 (s, 3H), 1.37 (s, 6H); 13C NMR (75 MHz,
CDCl3) δ 202.0, 159.6, 159.2, 158.1, 130.9, 129.9, 127.7, 125.9, 115.2, 113.7, 112.5,
108.9, 107.9, 77.3, 54.8, 43.4, 27.85; HRESI-MS m/z calcd for C20H20O4, [M+H]+
325.1434, found 325.1488.
4.2.5.
2-(3,4-Dichlorophenyl)-1-(5-hydroxy-2,2-dimethyl-2H-chromen-6-yl)ethan-1-one
(9d)
Compound 9d was prepared as the same procedures of 9b with the commercially
available 3,4-dichlorophenylacetic acid as the starting material. 1H NMR (300 MHz,
CDCl3) δ 12.77 (s, 1H), 7.58 (d, J = 8.9 Hz, 1H), 7.42 (d, J = 8.2 Hz, 1H), 7.37 (d, J =
2.1 Hz, 1H), 7.11 (dd, J = 8.2, 2.1 Hz, 1H), 6.71 (d, J = 10.1 Hz, 1H), 6.37 (d, J = 8.9
13
Hz, 1H), 5.60 (d, J = 10.1 Hz, 1H), 4.17 (s, 2H), 1.47 (s, 6H); C NMR (75 MHz,
CDCl3) δ 200.1, 159.6, 134.0, 132.2, 130.9, 130.5, 130.1, 128.4, 127.9, 115.1, 112.3,
109.0, 108.3, 77.5, 43.0, 27.9; HRESI-MS m/z calcd for C19H16Cl2O3 363.0549
[M+H]+, found 363.0590.
4.2.6.
1-(5-Hydroxy-2,2-dimethyl-2H-chromen-6-yl)-2-(3,4,5-trimethoxyphenyl)ethan-1-on
e (9e)
Compound 9e was prepared as the same procedures of 9b with the commercially
available 3,4,5-trimethoxyphenylacetic acid as the starting material. 1H NMR (300
MHz, CDCl3) δ 12.92 (s, 1H), 7.62 (d, J = 8.8 Hz, 1H), 6.68 (d, J = 10.2 Hz, 1H),
6.45 (s, 2H), 6.32 (d, J = 8.8 Hz, 1H), 5.59 (d, J = 10.2 Hz, 1H), 4.11 (s, 2H), 3.82 (s,
13
9H), 1.43 (s, 6H); C NMR (75 MHz, CDCl3) δ 201.5, 159.6, 159.4, 152.8, 130.8,
129.5, 129.1, 127.8, 115.1, 112.5, 108.8, 108.0, 106.2, 105.9, 77.4, 60.3, 55.6, 44.4,
27.8; HRESI-MS m/z calcd for C22H24O6 385.1646 [M+H]+, found 385.1679.
4.2.7.
3-(3,4-Dimethoxyphenyl)-1-(5-hydroxy-2,2-dimethyl-2H-chromen-6-yl)propan-1-one
21

ACCEPTED MANUSCRIPT
(9f)
Compound 9f was prepared as the same procedures of 9b with the commercially
available 3,4-dimethoxyhydrocinnamic acid as the starting material. 1H NMR (300
MHz, CDCl3) δ 13.06 (s, 1H), 7.52 (d, J = 8.9 Hz, 1H), 6.78 (m, 3H), 6.72 (d, J =
10.1 Hz, 1H), 6.31 (d, J = 8.8 Hz, 1H), 5.58 (d, J = 10.1 Hz, 1H), 3.87 (d, J = 3.8 Hz,
13
6H), 3.20 (t, J = 7.5 Hz, 2H), 2.99 (t, J = 7.5 Hz, 2H), 1.45 (s, 6H); C NMR (75
MHz, CDCl3) δ 203.3, 159.2, 159.1, 148.4, 147.0, 133.0, 130.3, 127.7, 119.6, 115.3,
112.9, 111.2, 110.8, 108.8, 107.8, 77.2, 55.4, 55.3, 39.3, 29.6, 27.8; HRESI-MS m/z
calcd for C22H24O5 369.1722 [M+H]+, found 369.1739.
4.2.8.
2-(3,4-Dimethoxyphenyl)-1-(5-methoxy-2,2-dimethyl-2H-chromen-6-yl)ethan-1-one
(11)
Compound 9b (500 mg, 1.41 mmol, 1.0 equiv.), methyl iodide (131 µL, 2.12 mmol,
1.5 equiv.), and potassium carbonate (292 mg, 2.12 mmol, 1.5 equiv.) were dissolved
in acetone (20 mL).The mixture was refluxed for 12 h. After the reaction was
completed, allow it to cool to room temperature. The solvent was removed at reduced
pressure, and the residue was treated with 10 mL water and extracted with ethyl
acetate (3 × 30 mL). The combined organic phases were dried (Na2SO4), and the
solvent was removed at reduced pressure. The crude residue was purified by column
chromatography with petroleum/ethyl acetate (8:1) to give 11 as a white solid (0.36 g,
70%). 1H NMR (300 MHz, CDCl3) δ 7.41 (d, J = 8.6 Hz, 1H), 6.70 (s, 3H), 6.51 (m,
2H), 5.59 (d, J = 10.0 Hz, 1H), 4.11 (s, 2H), 3.75 (s, 6H), 3.68 (s, 3H), 1.35 (s, 6H);
13C NMR (75 MHz, CDCl3) δ 198.2, 157.3, 155.9, 148.3, 147.3, 130.7, 130.1, 127.1,
124.3, 121.2, 116.0, 114.3, 112.2, 112.2, 110.7, 76.4, 62.8, 55.3, 55.3, 47.7, 27.5;
HRESI-MS m/z calcd for C22H24O5 369.1697 [M+H]+, found 369.1712.
4.2.9. 1-(5-Methoxy-2,2-dimethyl-2H-chromen-6-yl)-2-phenylethan-1-one (10a)
Compound 10a was prepared as the same procedures of 11 with the commercially
available phenylacetic acid as the starting material. 1H NMR (300 MHz, CDCl3) δ
22

ACCEPTED MANUSCRIPT
7.56 (d, J = 8.6 Hz, 1H), 7.28 (m, 5H), 6.63 (d, J = 10 Hz, 1H), 6.60 (d, J = 8.7 Hz,
1H), 5.70 (d, J = 10.0 Hz, 1H), 4.29 (s, 2H), 3.78 (s, 3H), 1.47 (s, 6H); 13C NMR (75
MHz, CDCl3) δ 201.7, 156.7, 155.4, 140.9, 130.4, 130.0, 128.1, 127.5, 126.2, 124.9,
116.1, 114.4, 111.9, 76.3, 62.8, 50.0, 27.6, 27.5; HRESI-MS m/z calcd for C20H20O3
309.1485 [M+H]+, found 309.1505.
4.2.10. 1-(5-Methoxy-2,2-dimethyl-2H-chromen-6-yl)-2-phenylpropan-1-one (10b)
Compound 10b was prepared as the same procedures of 11 with the commercially
available phenylacetic acid and 3 equivalent of methyl iodide as the starting material.
1H NMR (300 MHz, CDCl3) δ 7.38 (d, J = 8.6 Hz, 1H), 7.28 (m, 4H), 7.20 (d, J = 8.7
Hz, 1H), 6.60 (d, J = 10 Hz, 1H), 6.54 (d, J = 8.7 Hz, 1H), 5.66 (d, J = 10.0 Hz, 1H),
4.73 (d, J = 6.9 Hz, 1H), 3.68 (s, 3H), 1.54 (d, J = 6.9 Hz, 3H), 1.43 (s, 6H); 13C NMR
(75 MHz, CDCl3) δ 201.7, 156.7, 155.4, 140.9, 130.4, 130.0, 128.1, 127.5, 126.2,
124.9, 116.1, 114.4, 111.9, 76.3, 62.8, 50.0, 27.6, 27.5, 18.8; HRESI-MS m/z calcd for
C21H22O3 323.1642 [M+H]+, found 323.1678.
4.2.11.
1-(5-Methoxy-2,2-dimethyl-2H-chromen-6-yl)-2-(4-methoxyphenyl)ethan-1-one
(10c)
Compound 10c was prepared as the same procedures of 11 with the commercially
available 3,4-dimethoxyphenylacetic acid as the starting material. 1H NMR (300 MHz,
CDCl3) δ 7.47 (d, J = 8.6 Hz, 1H), 7.13 (d, J = 8.7 Hz, 2H), 6.81 (d, J = 8.7 Hz, 2H),
6.56 (m, 2H), 5.65 (d, J = 10.0 Hz, 1H), 4.16 (s, 2H), 3.74 (s, 3H), 3.73 (s, 3H), 1.41
(s, 6H); 13C NMR (75 MHz, CDCl3) δ 202.0, 157.8, 156.6, 155.3, 132.9, 130.4, 130.0,
128.5, 124.9, 116.1, 114.3, 113.5, 111.9, 76.2, 62.9, 54.7, 49.1, 27.5; HRESI-MS m/z
calcd for C21H22O4 339.1591 [M+H]+, found 339.1612.
4.2.12.
1-(5-Methoxy-2,2-dimethyl-2H-chromen-6-yl)-2-(4-methoxyphenyl)propan-1-one
(10d)
23

ACCEPTED MANUSCRIPT
Compound 10d was prepared as the same procedures of 11 with the commercially
1
available 4-methoxyphenylacetic acid as the starting material. H NMR (300 MHz,
CDCl3) δ 7.28 (d, J = 8.6 Hz, 1H), 7.15 (d, J = 8.2 Hz, 2H), 6.76 (d, J = 8.2 Hz, 2H),
6.50 (d, J = 8.6 Hz, 1H), 6.49 (d, J = 10.0 Hz, 1H), 5.61 (d, J = 10.0 Hz, 1H), 4.60 (d,
J = 6.9 Hz, 1H), 3.71 (s, 3H), 3.63 (s, 3H), 1.45 (d, J = 6.9 Hz, 3H), 1.38 (s, 6H); 13C
NMR (75 MHz, CDCl3) δ 202.0, 157.8, 156.6, 155.3, 132.9, 130.4, 130.0, 128.5,
124.9, 116.1, 114.3, 113.5, 111.9, 76.2, 62.9, 54.7, 49.1, 27.5, 18.7; HRESI-MS m/z
calcd for C22H24O4 353.1747 [M+H]+, found 353.1781.
4.2.13.
2-(3,4-Dimethoxyphenyl)-1-(5-methoxy-2,2-dimethylchroman-6-yl)ethan-1-one (12)
To a solution of compound 11 (100 mg) in methanol (10 mL) was added catalytic
amount of 10% Pd/C. The mixture was reacted under the atmosphere of hydrogen at
room temperature for 12 h. After the reaction was completed, the mixture was filtrated
to remove Pd/C and the solvent was removed at reduced pressure to obtained 12 as a
1
white solid (95 mg, 95%). H NMR (300 MHz, CDCl3) δ 7.47 (d, J = 8.7 Hz, 1H),
6.77 (s, 3H), 6.56 (d, J = 8.7 Hz, 1H), 4.17 (s, 2H), 3.81 (s, 6H), 3.73 (s, 3H), 2.74 (t,
J = 6.6 Hz, 2H), 1.77 (t, J = 6.6 Hz, 2H), 1.31 (s, 6H); 13C NMR (75 MHz, CDCl3) δ
198.4, 158.7, 158.4, 148.6, 147.2, 128.9, 127.2, 123.1, 121.2, 114.7, 112.8, 112.2,
110.7, 74.9, 61.3, 55.3 (two carbons), 47.5, 31.5, 26.2, 16.8; HRESI-MS m/z calcd for
C22H26O5 371.1853 [M+H]+, found 371.1889.
4.2.14.
2-(3,4-Dimethoxyphenyl)-1-(5-methoxy-2,2-dimethyl-2H-chromen-6-yl)ethan-1-ol
(13)
To a solution of compound 11 (1.0 g, 2.71 mmol, 1.0 equiv.) in methanol (20 mL) was
o
added sodium borohydride (205 mg, 5.43 mmol, 2.0 equiv.) at 0 C. The reaction
mixture was stirred at this temperature for 1 h. After the reaction was completed, the
reaction was quenched with saturated ammonium chloride solution. The reaction
mixture was treated with 10 mL water and extracted with ethyl acetate (3 × 30 mL).
24

ACCEPTED MANUSCRIPT
The combined organic phases were dried (Na2SO4), and the solvent was removed at
reduced pressure. The crude residue was purified by column chromatography with
petroleum/ethyl acetate (6:1) to give 13 as a white solid (0.95 g, 95%).1H NMR (300
MHz, CDCl3) δ 7.13 (d, J = 8.4 Hz, 1H), 6.72 (s, 2H), 6.61 (s, 1H), 6.56 (d, J = 8.4
Hz, 1H), 6.48 (d, J = 9.9 Hz, 1H), 5.59 (d, J = 9.9 Hz, 1H), 5.06-4.97 (m, 1H), 3.78 (s,
3H), 3.75 (s, 3H), 3.59 (s, 3H), 2.99-2.80 (m, 2H), 2.24 (s, 1H), 1.37 (s, 6H); 13C
NMR (75 MHz, CDCl3) δ 152.9, 148.2, 147.1, 130.6, 129.9, 128.0, 126.4, 121.0,
116.6, 113.9, 112.3, 112.2, 110.6, 75.2, 69.1, 62.1, 55.4, 55.2, 43.9, 27.2, 27.1;
HRESI-MS m/z calcd for C22H26O5 371.1853 [M+H]+, found 371.1877.
4.2.15. (E)-6-(3,4-Dimethoxystyryl)-5-methoxy-2,2-dimethyl-2H-chromene (14)
To a solution of acetic acid was added compound 13 (500 mg), and the mixture was
o
heated to 60 C. After the reaction was completed, saturated sodium bicarbonate
solution was added to quench to the reaction. The reaction mixture was treated with
10 mL water and extracted with ethyl acetate (3 × 30 mL). The combined organic
phases were dried (Na2SO4), and the solvent was removed at reduced pressure. The
crude residue was purified by column chromatography with petroleum/ethyl acetate
(6:1) to give 14 as a white solid (0.45 g, 90%). 1H NMR (300 MHz, CDCl3) δ 7.39 (d,
J = 8.6 Hz, 1H), 7.18 (d, J = 16.4 Hz, 1H), 7.07 (m, 2H), 6.96 (d, J = 16.4 Hz, 1H),
6.87 (d, J = 8.6 Hz, 1H), 6.64 (m, 2H), 5.67 (d, J = 10.0 Hz, 1H), 3.96 (s, 3H), 3.91 (s,
13
3H), 3.79 (s, 3H), 1.46 (s, 6H); C NMR (75 MHz, CDCl3) δ 153.3, 153.0, 148.6,
148.1, 130.7, 130.0, 126.9, 125.7, 122.8, 120.7, 119.0, 116.5, 114.3, 112.4, 110.8,
108.3, 75.6, 61.9, 55.5, 55.4, 27.4; HRESI-MS m/z calcd for C22H24O4, 353.1747
[M+H]+, found 353.1759.
4.2.16. 6-(2-(3,4-Dimethoxyphenyl)acetyl)-2,2-dimethyl-2H-chromen-5-yl acetate
(15)
To a solution of compound 9b (100 mg, 0.28 mmol, 1.0 equiv.) in dichloromethane
(15 mL) was added triethylamine (78 µL, 0.56 mmol, 2.0 equiv.) and acetic anhydride
(80 µL, 0.84 mmol, 3.0 equiv.). The mixture was stirred at room temperature for 10 h.
25

ACCEPTED MANUSCRIPT
After the reaction was completed, saturated sodium bicarbonate solution was added to
quench the reaction. The reaction mixture was treated with 10 mL water and extracted
with ethyl acetate (3 × 30 mL). The combined organic phases were dried (Na2SO4),
and the solvent was removed at reduced pressure. The crude residue was purified by
column chromatography with petroleum/ethyl acetate (6:1) to give 15 as a white solid
(100 mg, 90%). 1H NMR (300 MHz, CDCl3) δ 7.71 (d, J = 8.6 Hz, 1H), 6.83 (d, J =
8.0 Hz, 1H), 6.75 (m, 2H), 6.70 (d, J = 8.6 Hz, 1H), 6.39 (d, J = 10.1 Hz, 1H), 5.71 (d,
J = 10.1 Hz, 1H), 4.10 (s, 2H), 3.86 (s, 6H), 2.37 (s, 3H), 1.46 (s, 6H); 13C NMR (75
MHz, CDCl3) δ 195.4, 168.9,157.1, 148.4, 147.4, 145.6, 131.3, 130.8, 126.7, 121.0,
115.1, 114.9, 113.1, 112.0, 110.7, 77.0, 55.3, 46.4, 27.8, 20.5; HRESI-MS m/z calcd
for C23H24O6 397.1646 [M+H]+, found 397.1659.
4.2.17.
1-(5-(Benzyloxy)-2,2-dimethyl-2H-chromen-6-yl)-2-(3,4-dimethoxyphenyl)-3-phenyl
propan-1-one (16)
To a solution of compound 9b (100 mg, 0.282 mmol, 1.0 equiv.) in acetonitrile was
added potassium carbonate (58.4 mg, 0.423 mmol, 1.5 equiv.), and benzyl bromide
(67 µL, 0.56 mmol, 2.0 equiv.). The mixture was refluxed for 6 h. After the reaction
was completed, the reaction mixture was treated with 10 mL water and extracted with
ethyl acetate (3 × 30 mL). The combined organic phases were dried (Na2SO4), and the
solvent was removed at reduced pressure. The crude residue was purified by column
chromatography with petroleum/ethyl acetate (6:1) to give 16 as a white solid (120
mg, 80%).1H NMR (300 MHz, CDCl3) δ 7.40-7.35 (m, 6H), 7.15 (m, 3H), 7.07 (m,
2H), 6.73 (m, 3H), 6.52 (m, 2H), 5.56 (d, J = 10.1 Hz, 1H), 4.84 (t, J = 7.4 Hz, 1H),
4.61 (q, J = 11.4 Hz, 2H), 3.81 (s, 3H), 3.72 (s, 3H), 3.51 (m, 1H), 3.01 (m, 1H), 1.39
(s, 3H), 1.38 (s, 3H); 13C NMR (75 MHz, CDCl3) δ 200.4, 156.6, 153.7, 148.3, 147.4,
139.6, 136.2, 131.0, 130.4, 130.0, 128.8, 128.0, 127.7, 127.3, 125.5, 125.4, 120.3,
116.2, 114.8, 112.0, 110.8, 110.6, 77.2, 76.2, 57.8, 55.2, 39.3, 27.6, 27.48; HRESI-MS
m/z calcd for C35H34O5 [M+H]+ 535.2479, found 535.2511.
26

ACCEPTED MANUSCRIPT
4.2.18. 1-(2-Hydroxy-4-((2-methylbut-3-yn-2-yl)oxy)phenyl)ethan-1-one (18)
2,4-Dihydroxyacetophenone (1.0 g, 6.57 mmol, 1.0 equiv.), copper iodide (125 mg,
0.65 mmol, 0.1 equiv.), potassium iodide (109 mg, 0.65 mmol, 0.1 equiv),
3-chloro-3-methyl-1-butyne (652.3 µL, 5.92 mmol, 0.9 equiv.), and potassium
carbonate (1.09 g, 7.89 mmol, 1.2 equiv) was dissolved in 20 mL acetonitrile. The
mixture was stirred at room temperature for 12 h. After the reaction was completed,
the solvent was removed at reduced pressure, and the residue was treated with 10 mL
water and extracted with ethyl acetate (3 × 30 mL). The combined organic phases
were dried (Na2SO4), and the solvent was removed at reduced pressure. The crude
residue was purified by column chromatography with petroleum/ethyl acetate (20:1)
to give 18 as a white solid (0.86 g, 60%). 1H NMR (300 MHz, CDCl3) δ 13.22 (s, 1H),
7.48 (d, J = 8.9 Hz, 1H), 6.6 (d, J = 8.9 Hz, 1H), 6.53 (s, 1H), 3.61 (s, 1H), 2.63 (s,
3H), 1.41 (s, 6H); ESI-MS m/z 219.1 [M+H]+.
4.2.19. 1-(5-Hydroxy-2,2-dimethyl-2H-chromen-6-yl)ethan-1-one (19)
To a solution of N,N-dimethylaniline (5 mL) was added compound 18 (0.858 g), and
the mixture was heated to 130 °C and stirred for 12 h. After the reaction was
completed, the solvent was removed at reduced pressure, and the residue was treated
with 10 mL 10% HCl and extracted with ethyl acetate (3 × 30 mL). The combined
organic phases were dried (Na2SO4), and the solvent was removed at reduced pressure.
The crude residue was purified by column chromatography with petroleum/ethyl
acetate (20:1) to give 19 as a white solid (0.81 g, 95%). 1H NMR (300 MHz, CDCl3) δ
13.23 (s, 1H), 7.74 (d, J = 8.8 Hz, 1H), 6.94 (d, J = 10.0 Hz, 1H), 6.56 (d, J = 8.8 Hz,
1H), 5.82 (d, J = 10.0 Hz, 1H), 2.77 (s, 3H), 1.69 (s, 6H); ESI-MS m/z 219.1 [M+H]+.
4.2.20. 1-(5-Methoxy-2,2-dimethyl-2H-chromen-6-yl)ethan-1-one (20)
Compound 19 (0.81 g, 1.41 mmol, 1.0 equiv.), potassium iodide (61 mg, 0.37 mmol,
0.1 equiv.), and potassium carbonate (615 mg, 4.45 mmol, 1.2 equiv.) were dissolved
in 20 mL acetone. The reaction was refluxed for 12 h. After the reaction was
completed, the solvent was removed at reduced pressure, and the residue was treated
27

ACCEPTED MANUSCRIPT
with 10 mL water and extracted with ethyl acetate (3 × 30 mL). The combined organic
phases were dried (Na2SO4), and the solvent was removed at reduced pressure. The
crude residue was purified by column chromatography with petroleum/ethyl acetate
1
(20:1) to give 20 as a white solid (0.60 g, 70%). H NMR (300 MHz, CDCl3) δ 7.58
(d, J = 8.6 Hz, 1H), 6.60 (d, J = 9.8 Hz, 2H), 5.69 (d, J = 10.0 Hz, 1H), 3.80 (s, 3H),
2.60 (s, 3H), 1.45 (s, 6H); ESI-MS m/z233.1 [M+H]+.
4.2.21.
2-(3,5-Dimethoxyphenyl)-1-(5-methoxy-2,2-dimethyl-2H-chromen-6-yl)ethan-1-one
(21a)
To a solution of 3,5-dimethoxybromobenzene (0.2 g, 0.92 mmol, 1.0 equiv.) in THF
(20 mL) was added compound 20 (192 mg, 0.82 mmol, 0.9 equiv.), potassium
tert-butoxide (155 mg, 1.38 mmol, 1.5 equiv.), tris(dibenzylideneacetone)dipalladium
(0.01 equiv.) and bis(2-diphenylphosphinophenyl)ether (0.01 equiv.). The mixture was
refluxed for 6 h. After the reaction was completed, the solvent was removed at
reduced pressure, and the residue was treated with 10 mL water and extracted with
ethyl acetate (3 × 30 mL). The combined organic phases were dried (Na2SO4), and the
solvent was removed at reduced pressure. The crude residue was purified by column
chromatography with petroleum/ethyl acetate (6:1) to give 21a as a pale yellow solid
(0.24 g, 70%). 1H NMR (300 MHz, CDCl3) δ 7.51 (d, J = 8.6 Hz, 1H), 6.60 (m, 2H),
6.42 (d, J = 2.1 Hz, 2H), 6.34 (s, 1H), 5.68 (d, J = 10.0 Hz, 1H), 4.20 (s, 2H), 3.76 (s,
13
3H), 3.77 (s, 6H), 1.45 (s, 6H); C NMR (75 MHz, CDCl3) δ 198.2, 157.3, 155.9,
148.3, 147.3, 130.7, 130.1, 127.1, 124.3, 121.2, 116.0, 114.3, 112.2, 112.2, 110.6, 76.4,
62.8, 55.3, 55.9, 47.7, 27.5; HRESI-MS m/z calcd for C22H24O5 369.1697 [M+H]+,
found 369.1741.
4.2.22.
2-(2,5-Dimethoxyphenyl)-1-(5-methoxy-2,2-dimethyl-2H-chromen-6-yl)ethan-1-one
(21b)
Compound 21b was prepared as the same procedures of 21a with the commercially
28