Discovery and optimization of 1-(1H-indol-1-yl)ethanone derivatives as CBP/EP300 bromodomain inhibitors for the treatment of castration-resistant prostate cancer

Article abstract of DOI:10.1016/j.ejmech.2018.01.087

The CREB (cAMP responsive element binding protein) binding protein (CBP) and its homolog EP300 have emerged as new therapeutic targets for the treatment of cancer and inflammatory diseases. Here we report the identification, optimization and evaluation of 1-(1H-indol-1-yl)ethanone derivatives as CBP/EP300 inhibitors starting from fragment-based virtual screening (FBVS). A cocrystal structure of the inhibitor (22e) in complex with CBP provides a solid structural basis for further optimization. The most potent compound 32h binds to the CBP bromodomain and has an IC₅₀ value of 0.037 μM in the AlphaScreen assay which was 2 times more potent than the reported CBP bromodomain inhibitor SGC-CBP30 in our hands. 32h also exhibit high selectivity for CBP/EP300 over other bromodomain-containing proteins. Notably, the ester derivative (29h) of compound 32h markedly inhibits cell growth in several prostate cancer cell lines including LNCaP, 22Rv1 and LNCaP derived C4-2B. Compound 29h suppresses the mRNA expression of full length AR (AR-FL), AR target genes and other oncogene in LNCaP cells. 29h also reduces the expression of PSA, the biomarker of prostate cancer. CBP/EP300 inhibitor 29h represents a promising lead compound for the development of new therapeutics for the treatment of castration-resistant prostate cancer.

Full text of DOI:10.1016/j.ejmech.2018.01.087

Accepted Manuscript
Discovery and optimization of 1-(1H-indol-1-yl)ethanone derivatives as CBP/EP300
bromodomain inhibitors for the treatment of castration-resistant prostate cancer
Qiuping Xiang, Chao Wang, Yan Zhang, Xiaoqian Xue, Ming Song, Cheng Zhang,
Chenchang Li, Chun Wu, Kuai Li, Xiaoyan Hui, Yulai Zhou, Jeff B. Smaill, Adam V.
Patterson, Donghai Wu, Ke Ding, Yong Xu
PII:
S0223-5234(18)30109-0
10.1016/j.ejmech.2018.01.087
EJMECH 10164
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 8 September 2017
Revised Date: 11 January 2018
Accepted Date: 26 January 2018
Please cite this article as: Q. Xiang, C. Wang, Y. Zhang, X. Xue, M. Song, C. Zhang, C. Li, C. Wu,
K. Li, X. Hui, Y. Zhou, J.B. Smaill, A.V. Patterson, D. Wu, K. Ding, Y. Xu, Discovery and optimization
of 1-(1H-indol-1-yl)ethanone derivatives as CBP/EP300 bromodomain inhibitors for the treatment of
castration-resistant prostate cancer, European Journal of Medicinal Chemistry (2018), doi: 10.1016/
j.ejmech.2018.01.087.
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
Graphical abstract
1.5
1.0
0.5
0.0
DMSO
29h 1 µM
29h 5 µM
R1173
AR-FL
PSA
C-MYC
ZA
BC
Channel
Channel
32h
IC50 = 37 nM
Tm = 8.7 ℃
∆
Structure Based Discovery
Co-crystal structure of 22e−CBP
CBP represents a potential target of PCa

ACCEPTED MANUSCRIPT
Discovery and optimization of 1-(1H-indol-1-yl)ethanone derivatives as CBP/EP300
bromodomain inhibitors for the treatment of castration-resistant prostate cancer
†
,‡,#
†,§,#
†,‡
†,‡
†
Qiuping Xiang,
Chao Wang,
Yan Zhang, Xiaoqian Xue, Ming Song, Cheng
†
,§
†,§
†
†
¶
§
Zhang, Chenchang Li, Chun Wu, Kuai Li, Xiaoyan Hui, Yulai Zhou, Jeff B.
⊥
⊥
†
†,*
∇
Smaill, Adam V. Patterson, Donghai Wu, Ke Ding, and Yong Xu
†
Guangdong Provincial Key Laboratory of Biocomputing, Joint School of Life Sciences,
Guangzhou Institutes of Biomedicine and Health, Chinese Academy of Sciences;
Guangzhou Medical University, Guangzhou, China.
‡
University of Chinese Academy of Sciences, No. 19 Yuquan Road, Beijing 100049,
China.
§
School of Pharmaceutical Sciences, Jilin University, Changchun, China, No.1266
Fujin Road, Chaoyang District, Changchun, Jilin 130021, China.
¶
State Key Laboratory of Pharmaceutical Biotechnology, Department of Medicine, The
University of Hong Kong, Hong Kong SAR.
⊥
University of Auckland, Auckland Cancer Society Research Centre, School of Medical
Sciences, Private Bag 92019, Auckland, NZ.
∇
School of Pharmacy, Jinan University, 601 Huangpu Avenue West, Guangzhou 510632,
China.
*
Corresponding Author: Yong Xu, PhD, E-mail: xu_yong@gibh.ac.cn
#
These authors contributed equally to this work.

ACCEPTED MANUSCRIPT
ABSTRACT:
The CREB (cAMP responsive element binding protein) binding protein (CBP) and its
homolog EP300 have emerged as new therapeutic targets for the treatment of cancer and
inflammatory diseases. Here we report the identification, optimization and evaluation of
1
-(1H-indol-1-yl)ethanone derivatives as CBP/EP300 inhibitors starting from
fragment-based virtual screening (FBVS). A cocrystal structure of the inhibitor (22e) in
complex with CBP provides a solid structural basis for further optimization. The most
potent compound 32h binds to the CBP bromodomain and has an IC50 value of 0.037 µM
in the AlphaScreen assay which was 2 times more potent than the reported CBP
bromodomain inhibitor SGC-CBP30 in our hands. 32h also exhibit high selectivity for
CBP/EP300 over other bromodomain-containing proteins. Notably, the ester derivative
(29h) of compound 32h markedly inhibit cell growth in several prostate cancer cell lines
including LNCaP, 22Rv1 and LNCaP derived C4-2B. Compound 29h suppresses the
mRNA expression of full length AR (AR-FL), AR target genes and other oncogene in
LNCaP cells. 29h also reduces the expression of PSA, the biomarker of prostate cancer.
CBP/EP300 inhibitor 29h represents a promising lead compound for the development of
new therapeutics for the treatment of castration-resistant prostate cancer.

ACCEPTED MANUSCRIPT
Key words:
CBP; Prostate cancer; Bromodomain inhibitor
Abbreviations
CBP, CREB binding protein; PCa, prostate cancer; CRPC, castration-resistant prostate
cancer; AR, androgen receptor; BET, Bromodomain and extra terminal protein; FBVS,
fragment-based virtual screening; SAR, structure-activity relationship; TSA, thermal
stability shift assay

ACCEPTED MANUSCRIPT
1
. Introduction
Prostate cancer (PCa) is the most common malignancy in men and remains the second
leading cause of cancer mortality in men [1,2]. Androgen-deprivation therapy (ADT)
represents the gold standard treatment for PCa. Despite the initial benefit, the majority of
patients will progress to an incurable stage known as castration-resistant prostate cancer
(CRPC) within 2−3 years. CRPC remains androgen receptor (AR) signaling dependent
and responds to the recently approved androgen-signaling inhibitor enzalutamide and the
androgen-synthesis inhibitor abiraterone acetate. However, most patients eventually
develop resistance to these second-generation therapies as well. Therefore, an alternative
strategy for the development of new therapeutic agents is urgently needed. One potential
strategy is to disrupt AR upstream and/or downstream signaling without directly targeting
the AR ligand binding domain (LBD). Recently, Chinnaiyan and others have shown that
bromodomain-containing protein 4 (BRD4) inhibitors could disrupt the interaction
between the N-terminal domain (NTD) of AR and BRD4 and block metastatic CRPC
growth in animal models [3-5]. Sadar and others have shown that EPI-001 covalently
binds to the AR NTD, blocking the transcriptional activity of AR and its splice variants
and reducing the growth of CRPC xenografts [6-8]. We have shown that retinoic acid
receptor-related orphan receptor γ (RORγ) acts upstream of AR as a key determinant of
AR gene expression and that the RORγ antagonist XY018 can potently inhibit the growth
of various AR-positive xenograft tumors in mice [9,10].
CREB (cAMP responsive element binding protein) binding protein (CBP) and its
homolog EP300 are two closely related transcriptional regulators bearing LXXLL motifs.

ACCEPTED MANUSCRIPT
They act as steroid receptor co-activators for various nuclear receptors, including the
retinoid X receptor (RXR), estrogen receptor (ER) and AR [11-15]. CBP and EP300 have
been implicated in the development of various cancers, inflammatory conditions, and
other diseases [16-23]. In the advanced prostate cancer, CBP is highly expressed, and
its expression level has been found positively linked with cancer patient survival [24,25].
Mechanistically, CBP and EP300 interact with the AR NTD to potentially stabilize both
N/C interactions and AR binding to AREs (AR response elements) [13]. Therefore,
interrupting the CBP-AR NTD interaction through targeting the bromodomain of CBP
may be an efficient strategy to block AR signaling, and the CBP bromodomain may
represent a potential therapeutic target for prostate cancer.
There has been significant progress in the development of small molecule inhibitors
targeting the CBP bromodomain [20,25-35]. Among these inhibitors, compounds 1 and 2
were reported by Zhou using NMR spectroscopy [25]. Compound 1 was discovered by
screening a focused library of 200 compounds and was the first compound that was
shown to inhibit the CBP-p53 interaction at 50 µM (Figure 1). Unzue and co-workers
have developed acetyl benzene compound 3 (Figure 1) as an inhibitor of CBP by
fragment-based docking and structure-guided optimization. Compound 3 showed
moderate potency against CBP, with an IC of 8.7 µM as determined by TR-FRET [34].
5
0
Isoxazole derivatives were initially developed for targeting BET bromodomains and later
optimized as CBP bromodomain inhibitor (4, SGC-CBP30, Figure 1) [20]. Recently,
compound 5 (I-CBP112, Figure 1), with a 1,4-benzoxazepine scaffold, was reported as a

ACCEPTED MANUSCRIPT
CBP chemical probe and displayed good selectivity for CBP over other bromodomains
and good cellular activity [29].
These studies enhance our understanding of the role of CBP in oncology, which could
potentially pave the way for the development of therapeutics for various diseases.
However, more chemotypes are still urgently needed in order to exploit the therapeutic
potential of CBP inhibition in different human diseases such as prostate cancer.
In the present study, we report the fragment-based virtual screening, structure-guided
medicinal optimization and biological evaluation of 1-(1H-indol-1-yl)ethanone
derivatives as potent and selective CBP/EP300 bromodomain inhibitors for the treatment
of castration-resistant prostate cancer.

ACCEPTED MANUSCRIPT
2
2
1
. Results and discussion
.1 Identification and
-(1H-indol-3-yl)ethanone scaffold.
validation
of
1-(1H-indol-1-yl)ethanone
and
To obtain new scaffold of CBP inhibitors, a common virtual screening strategy [10,36-38]
was used as shown in Figure 2A. The SPECS database was first filtered with a molecular
weight threshold of less than 300. After that, the resulting set of fragments was screened
through molecular docking-based evaluation. After cluster analysis and visual inspection,
1
3 representative compounds were selected and purchased for biological evaluation. The
inhibitory activity and structures of the 13 compounds were shown in Figure 2B,
Supporting Information Table S1 and Figure S2. Using an AlphaScreen assay, 3 of the 13
compounds showed micromolar-level activity, with IC50 values less than 50 µM.
Compounds
1-(4,6-dinitro-1H-indazol-1-yl)ethanone) showed IC50 values of 16.73, and 9.11 µM,
respectively (Figure 2C). Not surprisingly, the 6 has a similar scaffold with the ligand of
TS8. The corresponding ligand efficiencies were approximately 0.4. These two
6
(3-(3-acetyl-1H-indol-1-yl)propanoic
acid)
and
7
(
4
compounds bear different scaffolds and represent good starting points for further
optimization. To further evaluate the selectivity, 3 more bromodomain-containing
proteins, BRD4(1), BAZ2B, and ATAD2, were tested to assess the activity of these two
fragments by the AlphaScreen assay. The result showed that 7 displayed equipotent
activities for all the bromodomain-containing proteins tested. However, 6 showed
excellent selectivity for the CBP bromodomain over other bromodomain-containing
proteins (Table S2). This result prompted us to choose 6 as a starting point for
optimization. To design more potent derivatives, we also compared the binding modes of

ACCEPTED MANUSCRIPT
6
and 7 bound to CBP (Figure 3A,B). The predicted binding mode of 7 bound to CBP
showed that 7 interacts with the conserved residues Asn1168 and Tyr1125 through direct
and water-mediated hydrogen bonds (Figure 3B). The binding model of 6 is characterized
by the presence of a hydrogen bond between the carbonyl group of the
1
-(1H-indol-3-yl)ethanone ring and the Asn1168 side chain of CBP (Figure 3A). The
carbonyl group also interacts with Tyr1125 through an indirect hydrogen bond via a
conserved water molecule present in the ZA channel. In addition, the propionic acid
group forms favorable hydrogen bonds with Arg1173. From the predicted binding mode
analysis, we can see that the 1-(1H-indol-3-yl)ethanone moiety of 6 could serve as an
acetylated lysine mimic and that three regions could be explored for further optimization
of this lead compound (Figure 3A). The first area is the LPF shelf, the second area is the
ZA channel, and the third area is the BC channel.
2
.2 Chemistry.
The derivatives of 1-(1H-indol-1-yl)ethanone and 1-(1H-indol-3-yl)ethanone were
designed and synthesized as shown in Schemes 1 3.
−
Briefly, intermediates 21a−21b were prepared by acylation of secondary amines 20a−20b
followed by reaction with 1-(1H-indol-3-yl)ethanone to produce
-(1H-indol-1-yl)ethanone derivatives 22b and 22c in good yields. Compounds 22d and
1
2
2
2e were obtained by hydrolysis of compounds 22b and 22c, respectively. Compound
2a was readily synthesized using a procedure similar to the one used to prepare 22b.

ACCEPTED MANUSCRIPT
The synthesis of 25a−25e is illustrated in Scheme 2. 1-acetyl-1H-indole-3-carboxylic
acid (24a) was first prepared by acetylization of 1H-indole-3-carboxylic acid (23a) with
acetic anhydride, followed by condensation with various amines to yield the desired
products. The ester 25c was hydrolyzed under basic conditions to yield the corresponding
carboxylic acid (26), which was acetylated to afford compound 25e.
Intermediates 27a−27l were prepared according to Scheme 3. The corresponding
carboxylic acids 30a−l were prepared by deprotection of 27a−27l with TFA. Compound
2
3
9a was directly demethylated with BBr to give 32b, which was subsequently etherified
to generate final products 32c and 32d.
Compounds 28a−28q and 29a−29h were prepared using the procedure outlined in
Scheme 3, which is similar to the procedure used in the preparation of 27a−27l.
Compounds 31a−31q, 32a and 32e−32h were prepared by a procedure similar to what
was used for the synthesis of 30a−30l. The synthesis of compounds 32b−32d is also
outlined in Scheme 3.
a
Scheme 1. Synthesis of 1-(1H-indol-3-yl)ethanone Derivatives 22a−22e

ACCEPTED MANUSCRIPT
a
Reagents and conditions: (a) acetic anhydride, K CO , DCM, 5 h, rt, 60−80%; (b)
2
3
1
5
-(1H-indol-3-yl)ethanone, K CO , KI, 5 h, reflux, 70−80%; (c) NaOH, MeOH, rt, 2 h,
2 3
7−80%.
a
Scheme 2. Synthesis of 1-(1H-indol-1-yl)ethanone derivatives 25a−25e
a
Reagents and conditions: (a) acetic anhydride, Et N, DMAP, DCE, 2 h, 60 ºC, 38−63%;
3
(
b) aliphatic amines, HATU, DIPEA, DCM, overnight, rt, 60−71%; or aryl amines,
n-Bu N, 2-chloro-1-methylpyridinium iodide, toluene, 16 h, 90 ºC, 30−68%; (c) NaOH,
MeOH, overnight, rt, 83%; (d) Et N, DMAP, DCE, 12 h, 60 ºC, 35%.
3
3
a
Scheme 3. Synthesis of compounds 29a−29h, 30a−30l, 31a−31q, 32a−32h

ACCEPTED MANUSCRIPT
a
Reagents and conditions: (a) acetic anhydride, Et
3
N, DMAP, DCE, 2 h, 60 ºC, 38−63%;
(
b) aryl amines, n-Bu N, 2-chloro-1-methylpyridinium iodide, toluene, 16 h, 90 ºC,
3
3
5−68%; (c) TFA, DCM, 6 h, rt, 58−90%; (d) BBr , DCM, 15 h, rt, 60% (for reactant
3
2
9a); (e) bromoethane or iodopropane, K CO , KI, acetone, 12 h, 70 ºC，38−43%.
2
3
2
.3. Structure−activity relationship studies of the LPF shelf.
To discuss the structure−activity relationships of the LPF shelf, we designed various
compounds with substituents on the head group, 1-(1H-indol-3-yl)ethanone. The binding
affinities of these compounds to CBP were determined and summarized in Table 1. To
ensure that these compounds can reach the LPF shelf and form further interactions with
CBP, the propionic acid group originally present in 6 was replaced with larger groups. As
shown in Table 1, replacing the propionic acid group in 6 with the corresponding ethyl

ACCEPTED MANUSCRIPT
ester yielded 22a, which was found to have a very weak binding affinity to CBP.
Considering the flexibility of 22a, we designed and synthesized conformational
constrained analogs 22b−22e. Compounds 22b and 22d exhibited potencies similar to 6.
2
2c and 22e were 2-fold more potent than 6, and they exhibited IC values of 7.09 and
50
6
.80 µM, respectively. The results indicated that adding a substituent at the meta-position
of the piperidine can improve ligand binding affinity.
To understand the structural basis for the protein-ligand interactions, we determined the
cocrystal structure of the 22e CBP complex. As shown in Figure 4A and 4B, the binding
−
mode of 22e is clearly defined by the electron density map. 22e plugs into the narrow
pocket with its 1-(1H-indol-3-yl)ethanone group interacting with the conserved residues
Asn1168 and Tyr1125. The 1-propionylpiperidine-3-carboxylic acid group can protrude
out of the pocket and reach the LPF shelf, forming further interactions with CBP. The
carboxyl group was involved in a canonical hydrogen bond with Arg1173. The crystal
structure also revealed that the carbonyl of the 1-propionylpiperidine-3-carboxylic acid
group interacts with Arg1173 through a water-mediated hydrogen bond. There are five
water molecules located in the ZA channel. To obtain more potent compounds through
water-mediated hydrogen bonds, the head group 1-(1H-indol-3-yl)ethanone was replaced
with 1-(1H-indol-1-yl)ethanone with an amide linker attached.
We synthesized compounds 25a and 25b, which showed weaker affinities than 22e (Table
1
). It seems that the flexibility of the attached group may be detrimental to CBP affinity.
Replacing the alkyl group with an aromatic group led to 25c, which was found to bind to

ACCEPTED MANUSCRIPT
CBP protein with an IC50 value of 3.39 µM and was twice as potent as 22e. The
corresponding acid form of 25c was also synthesized (25e) and evaluated. Compound 25e
showed a small increase in activity with IC value of 1.69 µM, while the acetyl analog
5
0
2
5d showed a sharp decrease in potency (IC = 25.37 µM) relative to compound 25e.
50
This finding demonstrated that the carboxyl group of 25e, also seen in 22e, was critical
for compound activity.
Analysis of the predicted binding mode of compound 25e in complex with the CBP
bromodomain suggested that the NH group of the amide forms a new hydrogen bond
with one of the conserved water molecule near the ZA channel. The phenyl group of 25e
extends to the LPF shelf and forms VDW interactions with the hydrophobic residue
Leu1109. The carboxyl group of compound 25e had the optimal orientation to form a
very favorable hydrogen bond with the guanidinium group of Arg1173 of the CBP
bromodomain (Figure 4C). It was indicated that a carboxyl group at the meta-position of
phenyl ring was more favorable than other groups.
2
.4. Structure−activity relationship studies of the ZA channel.
To investigate the SAR in the direction of the ZA channel, the preferred carboxyl group
was maintained. The substitutions on the ortho- or meta-position (R or R in Table 2)
3
4
were explored to improve the hydrophobic interactions with the ZA channel. First,
short-chain alkyl groups were attached at the R position and yielded 30a−30c.
3
Compound 30b, with a longer-chain substitution, showed a significant loss of activity
compared with the parent compound 25e. The relatively small groups are well tolerated,

ACCEPTED MANUSCRIPT
and the corresponding compounds showed activity similar to that of 25e. When the
hydrophobic cycloalkyl groups (30d−30f) were used to occupy the hydrophobic pocket, a
small increase in inhibition was observed. Analogs 30e and 30f demonstrated IC values
5
0
of 1.89 and 1.67 µM, respectively. To determine the importance of hydrophobic
interactions, polar atoms were introduced to the cycloalkyl group. If the substituent atom
is an O, as in 30g, a 4-fold decrease in inhibition was observed. This result demonstrated
that incorporation of a polar atom causes a reduction in the inhibitory activity. A flexible
linker of one methylene carbon was used to attach polar substituents in hopes of
improving activity (30h−30k). However, these compounds did not show improved
activity. When a halogen atom (F) was introduced to the R -position, significantly
3
improved potency was achieved; Compound 30l was found to display an IC value of
5
0
0
.77 µM. As shown in Figure 5A, the predicted binding mode of compound 30l was
similar to that of compound 25e. The size of F atom is critical and may contribute to the
improved activity. This finding suggested that small substituents at the R position were
3
favorable, while the larger groups may interfere with other residues around this pocket.
For the meta-position (R , Table 2), various substitutions were introduced to increase the
4
binding affinity through improving close contact with the protein surface. The results are
shown in Table 2. Compounds with alkyl or cycloalkyl groups did not improve the
activity. Among these compounds, analog 31f was one of the best compounds with
moderate biochemical activity (IC = 2.48 µM). Flexible linkers with 1 or 2 atoms (31g
5
0
and 31h) were also used to attach polar substituents for improved activity. Not
surprisingly, compound 31g demonstrated high potency, with an IC50 value of 1.0 µM.

ACCEPTED MANUSCRIPT
To find more potent analogs based on the modeled structure of 25e in complex with CBP
bromodomain, we designed compounds with various aromatic groups at the R position
4
to explore the chemical space for affinity improvement (Table 2). The phenyl ring analog
(
31i) showed similar potency to 25e against CBP, with an IC value of 1.86 µM. The
50
mono-substituted phenyl analogs were synthesized, leading to compounds 31j−31m. No
significant potency improvement was obtained. When a chlorine atom was substituted at
the meta-position (31k), the activity matched that of 31e.
To investigate the importance of polar atoms, we synthesized compounds 31n−31q. 31n
exhibited a slightly increased potency compared to compound 31i. Analog 31o was one
of the most promising compounds with strong biochemical activity (IC50 = 0.65 µM).
Compound 31p binds to CBP with IC value of 2.81 µM, which is one-fourth as strong
5
0
as 31o. This finding may be due to the hydrophobic phenyl group of compound 31p
being exposed to the solvent environment. Compound 31q showed comparable activity,
with an IC50 value of 1.00 µM. Analysis of the predicted binding mode for compound 31o
in complex with CBP protein (Figure 5B) suggested that the furan group orient along the
ZA channel, increasing binding affinity through close contact with the protein surface.
The O atom of the furan ring is exposed to the solvent environment, which is favorable
for ligand binding. The results demonstrated that the furan group of 31o is a favorable
group in the ZA channel.
2
.5. Structure−activity relationship studies of the BC channel.

ACCEPTED MANUSCRIPT
Analysis of the predicted binding mode for compound 31o in complex with CBP
bromodomain showed that there is still some space near the BC channel for further
optimization. As shown in Table 2, we designed various substituents at the R position to
5
explore the chemical space around the BC channel for affinity improvements. We
synthesized compound 32b with a hydroxy group at the R position, which gave an IC50
5
value of 0.17 µM, making it 4 times more potent than 31o. Expanding the size of alkoxy
group, compounds 32a, 32c and 32d were prepared to improve the activity. Compound
3
2a showed activity (IC50 = 0.14 µM) similar to that of 32b. However, compounds 32c
and 32d have IC50 values of 3.89 and 26.19 µM, respectively. From the predicted binding
mode, we can see that the substituent from this direction should also be exposed to
solvent, so a larger alkyl group may decrease the potency. These results suggested that the
optimal length of the substitution is approximately 1−2 heavy atoms. We next synthesized
compounds 32e and 32f, which were found to bind to CBP with IC values of 0.29 and
5
0
0
.69 µM, respectively. From the data shown in Table 2, we can see that heterocyclic
substituents at the R position are favorable. Changing the furan group in 32a to a
-methyl-1H-pyrazole led to 32g, which showed an activity (IC50 = 0.16 µM) similar to
that of 32a (IC = 0.14 µM). The results demonstrated that the BC channel is also a
4
1
5
0
critical region for potency (Figure 5B,C). Taken together, our modifications at the BC
channel have yielded compounds 32a and 32g with much improved activities over the
initial compounds 30l and 31o.
Combining all the substituents at R
3 4 5
, R and R positions, we synthesized compound 32h.
3
2h demonstrated significantly improved activity with an IC50 of 0.037 µM, which was

ACCEPTED MANUSCRIPT
approximately 4 times more potent than 32g and 2 times more potent than compound 4
The phenyl group with fluorine, carbonyl, and 1-methyl-1H-pyrazole substituents fit
snugly in the pocket as shown in Figure 5D. The hydrogen bonding and polar interactions
may contribute to the improved potency.
2
.6. Evaluation of bromodomains selectivity.
To investigate the selectivity profile, representative compounds were tested against 8
bromodomain-containing proteins from different subgroups of the bromodomain family
using an AlphaScreen assay (Table 3). Most of the tested compounds showed excellent
selectivity for the CBP bromodomain and its paralog EP300 over other
bromodomain-containing proteins. Compounds 31o and 32a demonstrated high potency,
with IC values of 0.65 and 0.14 µM for CBP, respectively; however, they also showed
5
0
weak potency for some other bromodomain-containing proteins. Compounds 6, 25e, 30l,
and 31g showed excellent selectivity. Among all the tested compounds, 32g and 32h
showed best activity and selectivity profiles. To further confirm the selectivity over other
bromodomain-containing proteins, compound 32g and 32h were evaluated against 8
bromodomain-containing proteins using a thermal stability shift (TSA) assay (Figure 6).
Similar to the AlphaScreen assay, the TSA assay confirmed that compounds 32g and 32h
showed promising selectivity profiles.
2
.7. Inhibitory effect of cell growth, colony formation and gene expression in
prostate cancer cells.

ACCEPTED MANUSCRIPT
CBP and EP300 have been linked to the development of various diseases. CBP inhibitors
have been evaluated for their therapeutic potential for the treatment of acute leukemia
[28,39,40]. In this study, we tested our selective CBP inhibitors to determine if they
possess any growth inhibition effects on prostate cancer cell lines. We selected
AR-dependent prostate cancer cell lines such as LNCaP and 22Rv1 cells to test the
cellular proliferation inhibition effects (Table 4). Enzalutamide, a second generation of
AR antagonist, was chosen as positive control. The most potent compounds 32a, 32g, and
3
2h did not show inhibition effect against LNCaP and 22Rv1 cells, which may be due to
the highly polar carboxyl group found in 32a, 32g and 32h restricting cellular
permeability. Therefore, the polarity of the acid group was masked with ester [41,42]. In
this case, the tert-butyl ester derivatives 29a, 29g and 29h were obtained for
corresponding carboxyl substituted compounds 32a, 32g, and 32h. In the AlphaScreen
assay, the esters derivatives 29a, 29g and 29h were found to bind to CBP with IC values
5
0
of 1.05, 1.04 and 0.42 µM, respectively. However, the esters derivatives 29a, 29g and 29h
showed reasonable potency in prostate cells LNCaP and 22Rv1, with IC50 values of
approximately 2.0 µM (Figure 7A,B). These antiproliferative activities were
appoximately 17-fold potent than that of the positive control enzalutamide (Table 4).
The antiproliferative activity of the compound 29h was also evaluated against a wide
range of cancer cell lines (Table S4). Not surprisingly, the compound did not show
obvious inhibition on the growth of AR-negative prostate cancer cells PC-3 (IC > 20
5
0
µΜ). 29h also displayed very weak inhibitory activities against breast cancer cell lines
MCF-7 (ER positive), Hs578T (triple negative) and MDA-MB-231 (triple negative) with

ACCEPTED MANUSCRIPT
IC50 value of 7.46, 15.09 and 35.99 µΜ, respectively. The growth inhibitory activities of
the compound against lung fibroblast HFL-1 cells and human normal liver HL7702 cells
were also evaluated to monitor the potential toxic effects. As show in Table S4,
compound 29h exhibited less cytotoxic effect against these two normal cells with IC50
values of 100 µM and 40 µM, respectively, indicating that the cytotoxic effects of 29h
was minimal.
To further investigate the growth inhibition effects, colony formation assays were
performed for representative compounds 32g and 32h and their tert-butyl ester
derivatives 29g and 29h. Similar to the cell viability assay, compounds 32g and 32h show
no inhibitory activity for colony formation. However, the ester derivatives 29g and 29h
reduced colony formation in a dose-dependent manner (Figure 7C,D). Colony formations
were reduced to less than 10% in 22Rv1 and C4-2B cells at 4 µM. Thus, 29g and 29h
significantly inhibit growth of prostate cancer cells. Overall, 32g, 32h and their ester
derivatives 29g and 29h have good profiles for further optimization.
To investigate whether CBP inhibitors suppress the expression of full length AR, AR
spliced variants, and AR-regulated target genes, qRT-PCR was performed in LNCaP cells.
As shown in Figure 8A, compound 29h strongly inhibited mRNA expression of full
length AR. The AR-regulated genes KLK2, PSA (also known as KLK3) and TMPRSS2
were also inhibited upon 29h treatment. Myc, a known oncogenic driver in prostate
cancer, was also strongly inhibited by 29h. Furthermore, western blot analysis indicated

ACCEPTED MANUSCRIPT
that treatment of 22Rv1 cells with compound 29h resulted in down regulation of AR-FL
and AR-V7 level (Figure S2).
In a PSA luciferase reporter gene assay in LNCaP cells, both 29g and 29h significantly
reduced the expression level of PSA in a dose-dependent manner (Figure 8B). These
results indicated that targeting CBP may represent an alternative strategy for the
treatment of CRPC.
3
. Conclusions
In this study, we report the discovery of the 1-(1H-indol-1-yl)ethanone derivatives as
CBP/EP300 bromodomain inhibitors for the treatment of CRPC through structure-based
virtual screening, medicinal chemistry optimization, and biological evaluation.
Determination of the high-resolution crystal structure of the inhibitor with the CBP
provided guidance for structure-based optimization. The ester derivative 29h potently
inhibited cell growth or colony formation in the prostate cancer cell lines LNCaP, 22Rv1
and/or C4-2B. 29h also reduced the expression of AR, c-Myc and AR target genes.
Further optimization of 29h may ultimately provide a new class of therapeutics for the
treatment of CRPC.

ACCEPTED MANUSCRIPT
4
4
. Experimental section
.1. Computational methods.
The crystal structure of CBP bromodomain (PDB ID: 4TS8.pdb) [34] was used for the
molecular docking study. All of the ligand and protein preparation were performed with
in Maestro (version 9.4, Schrödinger, LLC, New York, NY, 2013) implemented in the
Schrödinger program (http://www. Schrödinger.com). The SPECS database
(http://www.specs.net), containing more than 270,000 molecules, was filtrated with a
molecular weight threshold of less than 300 Dalton. The resultant fragment dataset was
submitted to molecular docking study. The molecular docking study was performed with
the Glide program using the standard precision (SP) score mode. Top 5000 compounds
were selected for cluster analysis and visual inspection. The molecules, which bear a
carbon bond and form hydrogen bond with conserved Asn1168, were selected and
submitted for subsequent biological evaluation.
4
.2 General chemistry.
All commercial reagents and anhydrous solvents were purchased and used without
purification, unless otherwise specified. Flash chromatography was performed using
silica gel (300−400 mesh). All reactions were monitored by TLC, using silica gel plates
with fluorescence F254 and UV light visualization. All products were characterized by
1
their NMR and MS spectra. H NMR spectra were recorded using a Bruker AV-400
1
3
spectrometer at 400 MHz or 500 MHz and C NMR spectra were recorded using a
Bruker AV-500 spectrometer at 500 MHz. Coupling constants (J) are expressed in hertz
(Hz). Chemical shifts (δ), which relative to internal control (TMS), are reported in parts

ACCEPTED MANUSCRIPT
per million (ppm) units. The low resolution of ESI-MS was recorded on an Agilent 1200
HPLC-MSD mass spectrometer. The purities of compounds were determined to be over
9
5% by reverse-phase high performance liquid chromatography (HPLC) analysis. HPLC
instrument: Dionex Summit HPLC (column: Inertsil ODS-SP, 5.0 µm, 4.6 mm × 250 mm
GL Sciences Inc.); detector, UVD170U; injector, manual injector; pump, P680; detection
wavelength, 254 nm; flow rate, 1.0 mL/min.
(
4
.2.1 Procedure a of scheme 1.
A mixture of methyl piperidine-4-carboxylate (2 g, 14 mmol) (20a), 2-chloroacetyl
chloride (1.16 mL, 15.4 mmol) and K CO (5.8 g, 42 mmol) in DCM was stirred for 5 h
2
3
at rt. Upon completion, the solvent was removed in vacuo. Water was added and the
mixture was extracted with ethyl acetate (20 mL × 3). The organic layer was washed with
water, 1 N HCl and brine, dried over Na SO , and evaporated. The compound 21a
2
4
(methyl piperidine-4-carboxylate) was obtained as a yellow oil (2.072 g, 67.4% yield).
This oil was used in the next step without further purification. MS (APCI), m/z for
+
C
9
H
14ClNO
3
([M + H] ): Calcd 219.67, found 220.1.
4
.2.2 Procedure b of scheme 1.
To a solution of compound 21a (437.48 mg, 2 mmol) in acetone (50 mL) was added
-(1H-indol-3-yl)ethanone (264 mg, 1.66 mmol), K CO (688 mg, 4.98 mmol) and KI
1
2
3
(
42 mg, 0.252 mmol). The mixture was heated under reflux for 5 h. The aqueous layer
was extracted with ethyl acetate (50 mL × 3), and the organic layer was washed with
water and brine, dried over Na SO and concentrated under reduced pressure. The residue
2
4

ACCEPTED MANUSCRIPT
was purified by chromatography on
silica gel to afford methyl
1
-(2-(3-acetyl-1H-indol-1-yl)acetyl)piperidine-4-carboxylate (22b) (397 mg, 70% yield)
1
as a white solid. H NMR (400 MHz, DMSO-d ) δ 8.24 (s, 1H), 8.18 (d, J = 6.9 Hz, 1H),
6
7
3
1
.44 (d, J = 7.2 Hz, 1H), 7.20 (s, 2H), 5.29 (d, J = 4.7 Hz, 2H), 4.17 (d, J = 12.5 Hz, 1H),
.94 (d, J = 12.9 Hz, 1H), 3.64 (s, 3H), 3.24 (t, J = 12.1 Hz, 1H), 2.80 (t, J = 11.7 Hz,
H), 2.69 (t, J = 10.7 Hz, 1H), 2.42 (s, 3H), 1.94 (d, J = 11.6 Hz, 1H), 1.86 (d, J = 12.6
13
Hz, 1H), 1.71 (d, J = 10.5 Hz, 1H), 1.44 (d, J = 9.9 Hz, 1H). C NMR (125 MHz,
DMSO-d ) δ 192.05, 174.28, 164.91, 138.48, 137.57, 125.49, 122.55, 121.73, 121.27,
15.98, 110.75, 51.50, 47.40, 43.48, 40.86, 28.02, 27.50, 27.16, 18.48. MS (APCI), m/z
6
1
+
for C H N O ([M + H] ): Calcd 342.39, found 343.1. HPLC analysis: MeOH − H2O
1
9
22
2
4
(80:20), 5.20 min, 99.68% purity. The synthesis of 22a and 22c can refer to 22b (for
details, see Supporting Information).
4
.2.3 Procedure c of scheme 1.
Methyl 1-(2-(3-acetyl-1H-indol-1-yl)acetyl)piperidine-4-carboxylate (284 mg, 0.83 mmol)
in a mixture of MeOH (8 mL) was treated with 1 N NaOH (4 mL, 4.15 mmol) and stirred
for 2 h at room temperature. Then concentrated to remove MeOH and cooled to room
temperature in an ice bath and neutralized with 1 N HCl. The solid precipitate was
collected
by
filtration,
washed
with
water
and
dried.
1
-(2-(3-acetyl-1H-indol-1-yl)acetyl)piperidine-4-carboxylic acid (22d) (156 mg, 57.2%
1
yield) was white solid. H NMR (400 MHz, DMSO-d ) δ 12.32 (s, 1H), 8.23 (s, 1H), 8.20
6
–
4
8.09 (m, 1H), 7.46 (dd, J = 21.2, 7.4 Hz, 1H), 7.27 – 7.14 (m, 2H), 5.36 – 5.22 (m, 2H),
.16 (d, J = 13.0 Hz, 1H), 3.94 (d, J = 13.4 Hz, 1H), 3.27 – 3.19 (m, 1H), 2.81 (t, J = 11.1

ACCEPTED MANUSCRIPT
Hz, 1H), 2.61 – 2.55 (m, 1H), 2.43 (s, 3H), 1.98 – 1.89 (m, 1H), 1.85 (d, J = 11.2 Hz, 1H),
1
3
1
.69 (d, J = 10.6 Hz, 1H), 1.43 (d, J = 9.9 Hz, 1H), 1.24 (s, 1H). C NMR (125 MHz,
DMSO-d ) δ 192.12, 175.49, 164.92, 138.58, 137.59, 125.50, 122.58, 121.75, 121.28,
6
1
15.97, 110.80, 47.45, 43.63, 41.02, 41.02, 27.18. MS (APCI), m/z for C H N O ([M
18 20 2 4
+
+
H] ): Calcd 328.36, found 329.1. HPLC analysis: MeOH (1‰ TFA) − H O (90:10),
2
4
.15 min, 96.64% purity. The synthesis of 22e can refer to 22d (for details, see
Supporting Information).
4
.2.4 Procedure a of scheme 2.
1
H-indole-3-carboxylic acid (23a) (1.5 g, 9.3 mmol), Et N (4 mL, 27.9 mmol) and
3
DMAP (0.114 g, 0.93 mmol) were dissolved in DCE (12 mL). Then acetic anhydride (2.8
mL, 27.9 mmol) was added and the mixture was stirred for 2 h at 60 ºC. Upon completion,
the solvent was evaporated. The residue was dissolved in EtOAc, extracted with saturated
NaHCO solution (20 mL × 3), the aqueous phase was cooled to 0 °C and then acidified
3
with 1 N HCl. The precipitated solid was collected by filtration, washed with water and
dried in vacuum to obtain the desired compound 1-acetyl-1H-indole-3-carboxylic acid
1
(
24a) (1.2 g, 63% yield) as white solid. H NMR (400 MHz, DMSO-d6) δ 8.42 (s, 1H),
8
3
.35 (dd, J = 6.9, 1.8 Hz, 1H), 8.08 (dd, J = 6.5, 2.1 Hz, 1H), 7.44 – 7.28 (m, 2H), 2.73 (s,
H).
4
.2.5 Procedure b of scheme 2.
The N-(2-aminoethyl)acetamide (153 mg, 0.74mmol), HATU (218 mg, 0.74 mmol) and
DIPEA (191 mg, 1.48 mmol) were dissolved in DCM. The mixture was stirred at rt for

ACCEPTED MANUSCRIPT
3
0 min. The 1-acetyl-1H-indole-3-carboxylic acid (24a) (100 mg, 0.49 mmol) was then
added, the reaction mixture was stirred overnight at rt. The aqueous layer was extracted
with EtOAc (30 mL × 3), and the organic layer was washed with water and brine, dried
with Na SO , and evaporated. The residue was purified by silica gel to yield
2
4
N-(2-acetamidoethyl)-1-acetyl-1H-indole-3-carboxamide (25a) (100 mg, 71% yield) as
1
white solid. H NMR (400 MHz, CDCl ) δ 8.51 – 8.43 (m, 1H), 8.12 – 8.05 (m, 1H), 8.02
3
(s, 1H), 7.46 – 7.37 (m, 2H), 7.17 (s, 1H), 6.34 (s, 1H), 3.64 (dd, J = 10.7, 5.0 Hz, 2H),
1
3
3
.55 (dd, J = 10.9, 5.5 Hz, 2H), 2.70 (s, 3H), 2.03 (s, 3H). C NMR (125 MHz,
DMSO-d
6
) δ 170.68, 170.39, 164.24, 135.96, 129.65, 129.04, 126.03, 124.87, 122.46,
+
1
16.61, 116.59, 39.42, 39.36, 24.70, 23.58. MS (APCI), m/z for C H N O ([M + H] ):
1
5
17
3
3
Calcd 287.32, found 288.1. HPLC analysis: MeOH − H O (80:20), 5.26 min, 98.69%
2
purity.
The synthesis of 25b can refer to 25a (for details, see Supporting Information).
4
.2.6 Another procedure b of scheme 2.
The methyl 3-aminobenzoate (288 mg, 1.55 mmol), 1-acetyl-1H-indole-3-carboxylic acid
300 mg, 1.86 mmol) (24a), 3-chloro-1-iodo-2-methylpyridin-1-ium (950 mg, 3.72 mmol)
(
and n-Bu N (1379 mg, 7.44 mmol) were dissolved in toluene. Under nitrogen, the
3
reaction mixture was stirred overnight at 90 °C. It was concentrated and redissolved in
EtOAc. The organic phase was extracted with saturated NH Cl solution, saturated
4
NaHCO
3
solution, 1 N HCl and brine. The combined organic phase was dried over
Na SO
2
4
and concentrated under reduced pressure. The resulting crude product was

ACCEPTED MANUSCRIPT
purified
by
silica
gel
chromatography
to
yield
methyl
3
-(1-acetyl-1H-indole-3-carboxamido)benzoate (25c) (185 mg, 30% yield) as white solid.
1
H NMR (400 MHz, DMSO-d ) δ 10.28 (s, 1H), 8.83 (s, 1H), 8.39 (s, 1H), 8.36 (s, 1H),
6
8
.27 (d, J = 7.1 Hz, 1H), 8.14 (d, J = 7.9 Hz, 1H), 7.70 (d, J = 7.7 Hz, 1H), 7.54 (t, J =
13
7
.9 Hz, 1H), 7.40 (dq, J = 13.8, 6.8 Hz, 2H), 3.89 (s, 3H), 2.76 (s, 3H). C NMR (125
MHz, DMSO-d ) δ 170.80, 167.04, 163.01, 140.40, 136.03, 131.02, 130.77, 130.16,
6
1
29.03, 126.31, 125.17, 125.15, 124.84, 122.46, 121.20, 116.73, 116.03, 53.11, 24.82.
−
MS (APCI), m/z for C19
H
16
N
2
O
4
([M − H] ): Calcd 336.35, found 334.6. HPLC analysis:
MeOH − H
2
O (80:20), 10.71 min, 99.45% purity. The synthesis of 22d can refer to 25c
(for details, see Supporting Information).
4
.2.7 Procedure c of scheme 2.
Methyl 3-(1-acetyl-1H-indole-3-carboxamido)benzoate (25c) (100 mg, 0.3 mmol) in a
mixture of MeOH (10 mL) was treated with 1 N NaOH (1.5 mL, 1.5 mmol) and stirred
for overnight at room temperature. Then concentrated to remove MeOH and cooled to
room temperature and acidified with 1N HCl. The solid was collected by filtration, rinsed
with water and dried, the 3-(1H-indole-3-carboxamido)benzoic acid was obtained (25e).
1
H NMR (400 MHz, DMSO-d ) δ 12.91 (s, 1H), 11.75 (s, 1H), 9.89 (s, 1H), 8.39 (s, 1H),
6
8
7
.34 (d, J = 3.0 Hz, 1H), 8.21 (d, J = 7.4 Hz, 1H), 8.07 (d, J = 8.1 Hz, 1H), 7.62 (d, J =
.7 Hz, 1H), 7.47 (d, J = 7.3 Hz, 1H), 7.44 (d, J = 7.9 Hz, 1H), 7.26 – 7.08 (m, 2H).
4
.2.8 Procedure d of scheme 2.

ACCEPTED MANUSCRIPT
3
A mixture of 3-(1H-indole-3-carboxamido)benzoic acid (26) (67 mg, 0.23 mmol), Et N
(68.68 mg, 0.68 mmol) and DMAP (2.8 mg, 0.023 mmol) in DCE (10 mL). Acetic
anhydride (69.36 mL, 0.68 mmol) was added and stirred for 12 h at 60 ºC. Then
concentrated to remove DCE and diluted with EtOAc, the organic phase was extracted
with saturated NaHCO solution, 1 N HCl and brine. The combined organic phases were
3
dried over Na SO and concentrated under reduced pressure. The expected compound
2
4
3
-(1-acetyl-1H-indole-3-carboxamido)benzoic acid (25e) was obtained (26 mg, 35%
1
yield) as white solid; H NMR (400 MHz, DMSO-d
6
) δ 12.99 (s, 1H), 10.24 (s, 1H), 8.84
(s, 1H), 8.37 (dd, J = 9.7, 1.5 Hz, 2H), 8.27 (dd, J = 6.5, 1.3 Hz, 1H), 8.11 (dd, J = 8.1,
1
2
1
1
.1 Hz, 1H), 7.68 (d, J = 7.8 Hz, 1H), 7.51 (t, J = 7.9 Hz, 1H), 7.40 (pd, J = 7.2, 1.4 Hz,
1
3
H), 2.76 (s, 3H). C NMR (125 MHz, DMSO-d ) δ 169.74, 167.09, 161.96, 139.21,
6
35.05, 131.30, 129.66, 128.86, 128.07, 125.26, 124.10, 123.85, 121.46, 120.59, 115.70,
+
15.18, 23.76. MS (APCI), m/z for C H N O ([M + H] ): Calcd 322.32, found 323.0.
1
8
14
2
4
HPLC analysis: MeOH (1‰ TFA) − H O (90:10), 4.95 min, 99.12% purity.
2
4
.2.9 Procedure a of scheme 3.
The synthesis of compound 1-acetyl-5-methoxy-1H-indole-3-carboxylic acid (24b) (38%
1
yield) can refer to 1-acetyl-1H-indole-3-carboxylic acid (24a). The solid was white. H
NMR (400 MHz, DMSO-d ) δ 8.39 (s, 1H), 8.24 (d, J = 9.1 Hz, 1H), 7.55 (d, J = 2.6 Hz,
6
1
H), 7.00 (dd, J = 9.1, 2.6 Hz, 1H), 3.81 (s, 3H), 2.71 (s, 3H).
4
.2.10 Procedure b of scheme 3.

ACCEPTED MANUSCRIPT
The
tert-butyl
3-amino-5-(furan-2-yl)benzoate
(450
mg,
1.74
mmol),
1
3
-acetyl-5-methoxy-1H-indole-3-carboxylic acid (485 mg, 2.08 mmol) (24b),
-chloro-1-iodo-2-methylpyridin-1-ium (1.07 g, 4.17 mmol) and n-Bu N (1.54 g, 8.4
3
mmol) were dissolved in toluene. The reaction mixture was stirred under N atmosphere
2
for 16 h at 90 °C. Afterwards the solvent was evaporated, water was added and extracted
with EtOAc (3 x 50 mL). The combined organic layer was dried over Na SO and
2
4
concentrated under reduced pressure. The residue was purified by silica gel
chromatography to provide the desired compound tert-butyl
-(1-acetyl-5-methoxy-1H-indole-3-carboxamido)-5-(furan-2-yl)benzoate (29a) (370 mg,
3
4
1
5.1%) as white solid. H NMR (400 MHz, DMSO-d ) δ 10.31 (s, 1H), 8.82 (s, 1H), 8.45
6
(s, 1H), 8.25 (d, J = 9.0 Hz, 1H), 8.19 (s, 1H), 7.92 (s, 1H), 7.84 (d, J = 1.2 Hz, 1H), 7.78
(d, J = 2.6 Hz, 1H), 7.01 (m, 2H), 6.66 (dd, J = 3.3, 1.8 Hz, 1H), 3.84 (s, 3H), 2.74 (s,
+
3
4
H), 1.60 (s, 9H). HRMS (ESI) for C H N O [M + H] , calcd: 474.1790, found:
2
7
26
2
6
75.1828.
2
7a−27l, 28a−28q, 29b−29h were prepared by the method similar to that of 29a (for
details, see Supporting Information).
4
.2.11 Procedure c of scheme 3.
Tert-butyl 3-(1-acetyl-5-methoxy-1H-indole-3-carboxamido)-5-(furan-2-yl)benzoate (29a)
50 mg, 0.011 mmol) was dissolved in DCM (10 mL) and trifluoroacetic acid (1 mL).
(
The mixture was stirred at rt for 5 h. The solvent was removed and the reside was
recrystallized with petroleum/ethyl acetate to afford the desired compound