Design, synthesis and biological evaluation of tetrazole-containing RXRα ligands as anticancer agents

Article abstract of DOI:10.1016/j.ejmech.2018.12.036

Nuclear receptor RXRα plays an important role in many biological and pathological processes. The nongenomic action of RXRα is implicated in many cancers. K-8008, a non-canonical RXRα ligand derived from sulindac, inhibits the TNFα-activated PI3K/AKT pathway by mediating the interaction between a truncated form of RXRα (tRXRα) and the p85α regulatory subunit of PI3K and exerts potent anticancer activity in animal model. Herein we report our studies of a novel series of K-8008 analogs as potential anticancer agents targeting RXRα. Two compounds 8b and 18a were identified to have slightly stronger binding to RXRα and improved apoptotic activities in breast cancer cells.

Full text of DOI:10.1016/j.ejmech.2018.12.036

Accepted Manuscript
Design, synthesis and biological evaluation of tetrazole-containing RXRα ligands as
anticancer agents
Zhiqiang Yan, Shuyi Chong, Huiyun Lin, Qian Yang, Xin Wang, Weidong Zhang,
Xiaokun Zhang, Zhiping Zeng, Ying Su
PII:
S0223-5234(18)31073-0
DOI:
https://doi.org/10.1016/j.ejmech.2018.12.036
EJMECH 10969
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 30 September 2018
Revised Date: 11 December 2018
Accepted Date: 14 December 2018
Please cite this article as: Z. Yan, S. Chong, H. Lin, Q. Yang, X. Wang, W. Zhang, X. Zhang, Z.
Zeng, Y. Su, Design, synthesis and biological evaluation of tetrazole-containing RXRα ligands as
anticancer agents, European Journal of Medicinal Chemistry (2019), doi: https://doi.org/10.1016/
j.ejmech.2018.12.036.
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
tetrazole-containing RXRα ligands as anticancer agents
Zhiqiang Yan a, 1, Shuyi Chong a, 1, Huiyun Lin a, 1, Qian Yanga, Xin Wang a, Weidong Zhang a,
Xiaokun Zhang a, b, Zhiping Zeng a, *, Ying Su b, *
.
a School of Pharmaceutical Science, Fujian Provincial Key Laboratory of Innovative Drug Target
Research, Xiamen University, Fujian, 361002, China
b
Sanford Burnham Prebys Medical Discovery Institute, 10901 N. Torrey Pines Road, La Jolla,
CA 92037, USA
1 These authors contributed equally to this work.
* To whom correspondence should be addressed:
Email: ysu@sbpdiscovery.org (Y. Su) or to zengzhiping@xmu.edu.cn (Z. Zeng)
1

ACCEPTED MANUSCRIPT
Abstract:
Nuclear receptor RXR
The nongenomic action of RXR
ligand derived from sulindac, inhibits the TNF
interaction between a truncated form of RXR (tRXR
and exerts potent anticancer activity in animal model. Herein we report our studies of a novel series
of K-8008 analogs as potential anticancer agents targeting RXR . Two compounds 8b and 18a
and improved apoptotic activities in
α
plays an important role in many biological and pathological processes.
α
is implicated in many cancers. K-8008, a non-canonical RXR
-activated PI3K/AKT pathway by mediating the
) and the p85 regulatory subunit of PI3K
α
α
α
α
α
α
were identified to have slightly stronger binding to RXR
breast cancer cells.
α
Key words: RXRα; RXRα modulator; PI3K/AKT pathway; Anticancer activity; bioisostere.
1. Introduction
As a unique and important member of the nuclear receptor (NR) superfamily, retinoid X
receptor- (RXR ) plays critical roles in many biological and pathological processes [1].
Dysfunctions of RXR are implicated in a number of diseases, such as cancer [2]. Targeted
disruption of RXR gene leads to prostatic preneoplastic lesions [3] and skin abnormalities [4].
RXR exhibits a modular organization structurally consisting of three main functional
α
α
α
α
α
domains: an N-terminal region, a DNA binding domain, and a ligand-binding domain (LBD). The
LBD is composed of a ligand-binding pocket (LBP) for binding small molecule ligands; a
transactivation function domain termed AF-2 composed of helix 12 (H12) of the LBD, a
co-regulator binding surface, and a dimerization surface [5, 6]. A well-studied mechanism of NR
action is the ligand-mediated transcriptional regulation. Ligand binding to the LBP induces the
conversion of the corepressor-binding site into a coactivator-binding site that triggers a cascade of
events and leads to biological activities. Thus, RXR
α is widely studied for its action as a
2

ACCEPTED MANUSCRIPT
ligand-mediated transcription factor and many molecules have been reported to bind to the LBP of
RXR
motif head, such as carboxyl group, a central polyene linker and a hydrophobic tail [7, 8]. The 9-cis
retinoic acid (9-cis-RA) is identified as an endogenous RXR ligand. Targretin, an RXR-selective
α. In general, RXRα LBD ligands are constituted of three important aspects, including a polar
α
ligand, acts as a highly effective agent for mammary carcinoma and is approved by FDA to treat
cutaneous T-cell lymphoma [9, 10].
Besides acting as a transcription factor, RXRα possesses important extracellular functions
[11-15]. We previously discovered that RXRα and its ligands can directly modulate the PI3K/AKT
survival pathway in a TNFα-dependent manner. TNFα is a multifunctional cytokine that controls
diverse cellular events such as cell survival and death that control the destiny of cancer cells [16].
PI3K/AKT activation is implicated in oncogenesis and drug resistance. Aberrations in this pathway
can potentially cause cell transformation, metabolic disorders, and neurodegenerative diseases, as
well as drug resistance [17]. The pathway has therefore been targeted extensively for therapeutic
application. However, strategies that rely on direct inhibition of PI3K/AKT activities are hindered
by toxicity or lack of selectivity [18]. Thus, identifying key molecules involved in the aberrant
PI3K/AKT activation could offer new strategies for drug development. We found that an
N-terminally truncated RXRα (tRXRα), detected in a variety of cancer cell lines, could act to
mediate TNFα activation of PI3K/AKT in a number of cancer cell lines [19]. tRXRα is
cytoplasmic and interacts with the p85α subunit of PI3K to activate the PI3K/AKT survival
pathway and induce anchorage-independent cell growth in vitro and cancer cell growth in animals.
Therefore, this nongenomic regulation of the PI3K/AKT signaling pathway by tRXRα provides
new strategies to inhibit the activation of PI3K/AKT in cancer cells by targeting RXRα.
We found that K-8008 (Fig. 1), a derivative of sulindac, can bind tRXR
interaction between tRXRα and p85α regulatory subunit of PI3K, leading to the inhibition of the
TNF -activated PI3K/AKT pathway [19-21]. In animal experiments, K-8008 effectively inhibited
α and mediate the
α
the growth of tumor, implying that K-8008 could be a potential anticancer lead. Furthermore, the
3

ACCEPTED MANUSCRIPT
crystal structure of the K-8008 bound RXR
α-LBD reveals a new binding mode, which is different
from the canonical ligand binding in the LBP. Thus, K-8008 represents a new class of RXR
α
ligands. Here we report the design, synthesis and biological evaluation of a series of K-8008
derivatives. A total of 26 compounds were synthesized, and their apoptotic activity, binding
activity and SAR were investigated. Two compounds 8b and 18a displayed improved apoptotic
activities in breast cancer cells.
2. Chemistry
The proposed K-8008 derivatives were synthesized according to Schemes 1-4. Compounds
8a-k and compound 9a were prepared using Scheme 1. Under the treatment of K2CO3 as a catalytic
base, benzaldehyde 1 reacted with propionic anhydride (2a) or acetic anhydride (2b) at 140 °C
overnight to afford 2a in 83% yield and 2b in 60% yield. 2a-b was reduced by hydrogen gas
catalyzed by Pd/C at room temperature in methanol for 16 h to afford 3a-b in about 90% yield. 4a-b
were prepared from 3a-b using polyphosphoric acid (PPA) as the condensing agent at 80 °C for 4 h.
With the help of samarium (II) iodide in THF solution also known as "Kagan's reagent", indanone
4a-b successfully coupled with acrylonitrile and the product mixtures were processed with a
solution of H2SO4 and HOAc (v/v= 10:1) to afford 5a and 5b in 20% to 55% yields. The obtained
cyano products 5a-b were reacted with sodium azide, catalyzed by triethylamine hydrochloride in
DMF to form the tetrazole products 6a (84%) and 7a (56%), respectively. For step vi. sodium
methoxide was used as the base for the condensation reaction of compound 6a or 7a with different
substituted aromatic aldehydes to afford compounds 8a-k and compound 9a. The stereochemistry
of this condensation is dictated by the presence of a R1 group [22]. When R1 = -CH3, Z-isomers are
the main products. Whereas R1 = -H, E-isomers are the key products. Compound 13a was prepared
using scheme 2. Compound 4a was treated with diethyl cyanethylphosphonate in the presence of
DBU as catalyst to afford 10a in 75% yield. Formation of 11 was easily achieved by reacting 10a
with hydroxylamine (50% wt in water) in the absence of catalyst. With the treatment of CDI and in
4

ACCEPTED MANUSCRIPT
the presence of DBU as catalyst, 11 underwent ring closure to give the 4H-1,2,4-oxadiazol product
12. Product 13a was finally obtained by the condensation of 12 with aromatic aldehyde. Scheme
3 was used to make compound 17a and 18a. Reformatsky reaction was applied to introduce a
methyl carboxylate into fluorinated indanone 14 with active zinc powder, α-bromide ester, and a
mixture of MeOH and 1N NaOH solution (v/v= 4:3) to form carboxylic acid 15 in 52% yield. 16
was synthesized by the condensation reaction with 4-isopropylbenzaldehyde. 16 was subsequently
amidated by reagent Amino-1H-tetrazole and catalyst EDCI/HOBT in CH2Cl2 at room temperature
for 12 h to give product 17a in 87% yield. To form compound 18a, compound 16 was reacted with
EtOH in the presence of catalyst H2SO4 and without further purification, and the products were kept
on reacting with hydroxylamine to give 18a. The final products (compounds 20a-i and 21a-b) were
afforded using Scheme 4 which is similar to Scheme 1. All the synthesized compounds were
1
structurally elucidated by different spectrometric methods, including H NMR (600 MHz), 13C
NMR (151 MHz) and HRMS (ESI) (see the spectra in the supporting information).
5

ACCEPTED MANUSCRIPT
Scheme 1. Reagents and conditions. (i) Acetic anhydride or propionic anhydride, K2CO3, 140 °C,
overnight; (ii) H2/Pd-C, MeOH, r. t., 16 h; (iii) PPA, 80 °C, 4 h; (iv) samarium (II) iodide,
acrylonitrile, isopropyl alcohol, THF, 0 °C, 20 min; AcOH, H2SO4, r. t., overnight; (v) sodium
azide, triethylamine hydrochloride, DMF, 100 °C, 24 h; (vi) R2-PhCHO, sodium methoxide,
MeOH, 80 °C, 4 h, 41%~95%.
Scheme 2. Reagents and conditions. (i) diethyl cyanethylphosphonate, DBU, acetonitrile, r. t., 18
h;(ii) hydroxylamine (50% wt in water), EtOH, 80 °C, 3 h; (iii) CDI, DBU, 1,4-dioxane, 100 °C, 4
h; (iv) 4-isopropylbenzaldehyde, sodium methoxide, MeOH, 80 °C, 4 h.
6

ACCEPTED MANUSCRIPT
N
O
N
N
N
NH
F
H
iii.
COOH
F
17a, 87%
COOH
ii.
O
F
i.
F
O
OH
NH
F
14
15, 52%
16, 79%
iv.
18a, 60%
Scheme 3. Reagents and conditions. (i) ethyl bromoacetate, zinc powder, iodine, THF, 60 °C, 4 h;
NaOH, MeOH, r. t., overnight; (ii) 4-isopropylbenzaldehyde, sodium methoxide, MeOH, 80 °C, 4
h; (iii) amino-1H-tetrazole, EDCI, HOBT, CH2Cl2, 0 °C - r. t., 12 h. (iv) H2SO4, EtOH, 60 °C, 2 h;
hydroxylamine, sodium methoxide, MeOH, r. t., 4 h (two steps).
Scheme 4. Reagents and conditions. (i) diethyl cyanethylphosphonate, DBU, acetonitrile, r.t., 18 h;
(ii) sodium azide, triethylamine hydrochloride, DMF, 110 °C, 40 h; (iii) R2-PhCHO, sodium
methoxide, MeOH, 80 °C, 4 h, 29%~57%.
7

ACCEPTED MANUSCRIPT
3. Results and Discussion
3.1 Design of K-8008 derivatives
Previously we determined the crystal structure of RXRα-LBD/K-8008 complex and identified
the binding mode of K-8008 in the LBD of RXR
α
(PBD code: 4N8R) [20]. In the crystal structure,
tetramer. Each K-8008 is located in
two separate K-8008 molecules bind to the dimers of the RXR
α
a region close to the dimer-dimer interface (Fig. 2A). The tetrazole motif of K-8008 plays an
important role by anchoring the ligand via making charge-dipole interaction with Helix 11. The
rigid indene portion of the ligand is tightly imbedded in a hydrophobic environment (Fig. 2B).
However, there is considerable space around the R2-benzene ring (Fig. 2B and Table 1.), offering
opportunity to introduce substituents on the R2-benzene ring to improve the binding affinity of the
ligands. Therefore, we first focused on exploring different substituents on the R2-benzene ring. 11
compounds 8a-k were designed and synthesized as series 1. Previously we also reported that an
E-isomer of K-80003 (Fig. 1) [21] compound 30 exerted enhanced binding activity compared to
，
K-80003. Thus, we also made an E-isomer 9a of K-8008 (as series 2 in Table 1.). Furthermore,
K-8008 was developed based on K-80003 using the bioisostere concept and so we asked if tetrazole
could be replaced by other bioisosteres. Three bioisosteres of K-8008 were synthesized as
compounds 13a, 17a and 18a (series 3 in Table 2, Scheme 2 and 3).
Recently we reported the crystal structure of K-80003 bound to RXR
that the binding mode of K-80003 is different from the binding mode of K-8008 (Fig. 2B). In the
complex structure of RXR -LBD with K-80003, three molecules of K-80003 bind to the same
large hydrophobic cavity where only one molecule of K-8008 is bound in the crystal structure of
RXR -LBD/K-80008 complex (Fig. 2C-D) [23]. The difference in the binding modes between
α-LBD and discovered
α
α
K-80003 and K-8008 could be due to the length of the linker between the charged group and the
indene ring or to the difference in the nature of the charged groups. To have a better understanding
8

ACCEPTED MANUSCRIPT
of the role of the linking length, compounds 20a-i (series 4, in Table 3 and Scheme 4) and 21a-b
(series 5, in Table 3 and Scheme 4) were designed and synthesized for biological evaluation.
Fig 1. The chemical structures of K-80003, K-8008 and cpd 30.
9

ACCEPTED MANUSCRIPT
Fig 2. Crystal structures of RXRα-LBD in complex with K-8008 and K-80003. (A) Crystal
Structure of RXRα-LBD in complex with K-8008 (PBD code: 4N8R) reveals that two K-8008
molecules bind to the dimers of the RXRα-LBD tetramer. The K-8008 molecules are shown in
magenta sticks and the tetramer structure is shown in ribbon drawing. (B) Interactions between
K-8008 and RXRα-LBD. K-8008 molecule is displayed in green sticks with its binding region in
surface representation. (C) and (D) Superposition of the crystal structures of RXRα-LBD/K-80003
(PBD code: 5TBP) and RXRα-LBD/K-8008 (PBD code: 4N8R). The ribbon drawing of
RXRα-LBD/K-80003 are displayed in yellow and the K-80003 molecules are in orange sticks. The
10

ACCEPTED MANUSCRIPT
ribbon drawing of RXRα-LBD/K-8008 are displayed in cyan and the K-8008 molecules are in
magenta sticks.
3.2 Antagonism activity of K-8008 derivatives
The synthesized compounds were first evaluated for their effects on RXRα transactivation
activity by employing RXRα-LBD reporter assay. The LBD of RXRα was cloned as a Gal4 fusion
and the resulting Gal4-RXRαLBD chimera and Gal4 reporter system were used to evaluate the
effect of synthesized compounds. Gal4-RXRα/LBD strongly activated the Gal4 reporter in the
presence of 9-cis-RA. Compounds were tested for their ability to inhibit the transactivation of
9-cis-RA at three different concentrations (1 µM, 10 µM and 50 µM). Compounds that
demonstrated inhibition effects in a dose-dependent manner (Table 1-3) were considered to possess
antagonism activity and were further evaluated for their anticancer activities. Among the 26
derivatives, 15 compounds displayed the dose-dependent inhibition of 9-cis-RA-induced reporter
activity of RXRα.
3.3 In vitro anticancer activity and SAR study
The 15 compounds that could inhibit the RXRα transcription activity in dose-dependent
manner were then evaluated for their growth inhibition of breast cancer cells in 2 cancer cell lines,
MBA-MD-231 and MCF-7, using K-8008 as a positive control. Results are listed in Tables 1, 2 and
3. Based on the obtained data, structure-activity relationship was analyzed and summarized in
Figure 3.
For compound series 1 (Table 1), we mainly focused on exploring different substituents on the
R2-benzene ring to study their structure-activity relationship. Firstly, larger steric hindrance group
such as naphthalene ring (8a) was introduced to fill the available space where the R2-benzene group
binds as described in figure 2A, but surprisingly the compound failed to dose-dependently inhibit
the transcriptional activity (see the supporting information S1). This result probably indicated that
11

ACCEPTED MANUSCRIPT
simultaneous modification with large steric hindrance group at the ortho and meta position of the
R2-benzene ring was not in favor. Modifications with tert-butyl group (8c) slightly larger than the
isopropyl group as in K-8008 at the para position maintained its antagonism activity (see the
supporting information S1) but did not enhance its apoptotic activity against breast cancer
cell(Table 1, 8c). Compound 8b where a slightly smaller but electron-withdrawing and high
lipophilicity group trifluoromethoxy was used at the para position had also no reduction in its
inhibition of the RXRα transcriptional activity (see the supporting information S1) but mildly
enhanced its apoptotic effect against MDA-MB-231 and MCF-7(IC50 = 8.90 ± 0.72 µM and 17.34
± 0.78 µM respectively) (Table 1, 8b). Introduction of other electron-withdrawing groups such as
trifluoromethyl and chlorine (8h and 8k) at the para position of the R2-benzene moiety could still
maintain the compounds’ antagonism activity (see the supporting information S2) but
unfortunately displayed weaker apoptotic activity. On the contrary, electron-donating groups such
as methoxy or methylthiol group (8d and 8e) failed to work, with diminished antagonism effect
(see the supporting information S1-2 and Table 1). Thus, it appears that the para position of the
R2-benzene ring prefers substituents with appropriate size (K-8008, 8c) and electron-withdrawing
groups (8b, 8h, 8k). For the meta position of R2-benzene moiety, similar biological effects like 8h
and 8k were observed when trifluoromethyl or chlorine (8j or 8g) was introduced. Compounds 8f
and 8i with R2 = Cl and CF3 respectively at the Ortho-position showed abolished activity (see the
supporting information S2), implying that the Ortho-position of R2-benzene cannot tolerate any
substituents. For series 2 (Table 1.), the E-isomer product (9a) also reserved its antagonism
activity (see the supporting information S3) but had a reduced apoptosis effect.
For series 3 (Table 2.), when the tetrazole group was transformed to oxadiazol and the length
between polar group and indene moiety was changed from ethyl to methyl, the antagonism activity
of compound 13a was reserved (see the supporting information S3) and its apoptotic effect in
MDA-MB-231 cancer cell was slightly improved from 12.34 ± 0.85 µM (K-8008) to 9.04 ± 1.06
µM (13a). The improvement in the apoptotic effect was also observed for compound 20a (8.48 ±
12

ACCEPTED MANUSCRIPT
0.68 µM in MDA-MB-231) where its oxadiazol was restored to tetrazole (20a in Table 3.). When a
longer linker was used (17a), the compound did not show any apoptotic activities (Table 3.).
although its antagonism activity was retained (see the supporting information S3). Compound 18a
resulted from substituting the tetrazole of compound 17a with a carboxamide displayed an
enhanced apoptotic effect. As evidenced from the IC50 values, compound 18a was 2.5-fold (5.00 ±
0.57 µM) more potent than K-8008 in MDA-MB-231 cancer cell line and 1.3-fold (20.77 ± 0.78
µM) more potent than K-8008 in MCF-7.
We then changed the linking length between polar group and indene moiety from ethyl to
methyl and evaluated the effect of different substituents on R2-benzene moiety (series 4, Table 3). It
is interesting to find that compounds of series 4 behaved differently from series 1. Modification at
para position was not in favor of electron-withdrawing group such as trifluoromethoxy or
trifluoromethyl (20b or 20f). In addition, introduction of electron donating or aliphatic groups such
as methoxy and tertiary butyl (20c and 20e) did not result in improved apoptotic activity. Unlike
compounds of series 1, when trifluoromethyl was introduced at para, meta or ortho position
respectively, only substitution at ortho position (20h) could maintain its antagonism and apoptotic
activity. The difference in the structure-activity relationship between series 1 and series 4
suggested that compounds of series 4 bind to RXRa differently from compounds of series 1.
Furthermore, replacing R2-benzene moiety with thiophene lost its apoptotic effect (20i). Two
E-isomers (21a and 21b) as series 5 were made and evaluated. Both 21a and 21b did not display
any obvious antagonistic effect (Table 3.).
Table 1. The growth inhibition (IC50, µM) of series 1 and 2 compounds against MDA-MB-231 and
MCF-7.
13

ACCEPTED MANUSCRIPT
Compound
R1
R2
isomer
Reporter Assaya
MDA-MB-231b
MCF-7b
K-8008
CH3
4-CH(CH3)2
Z
Dc
12.34 ± 0.85
26.99 ± 3.78
8a
Z
NDc
NTc
NTc
8b
8c
8d
8e
8f
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
H
4-OCF3
4-C(CH3)3
4-OCH3
4-SCH3
2-Cl
Z
Z
Z
Z
Z
Z
Z
Z
Z
Z
E
Dc
Dc
8.90 ± 0.72
12.49 ± 1.51
NTc
17.34 ± 0.78
31.10 ± 1.29
NTc
NDc
NDc
NDc
Dc
NTc
NTc
NTc
NTc
8g
8h
8i
3-Cl
24.73 ± 0.33
15.24 ± 1.73
NTc
45.55 ± 3.61
56.29 ± 5.00
NTc
4-Cl
Dc
2-CF3
NDc
Dc
8j
3-CF3
15.58 ± 2.42
22.08 ± 1.54
15.57 ± 1.29
28.76 ± 4.62
39.36 ± 3.67
41.78 ± 5.34
8k
9a
4-CF3
Dc
4-CH(CH3)2
Dc
aReporter assay. Cells were incubated with corresponding compounds of 3 concentrations 1 µM, 10 µM and 50 µM. Compounds were
tested for its dose-dependent antagonism activity.
bMTT methods. Cells were incubated with corresponding compounds at a concentration from 0.195 µM to 100 µM for 48 h. Compounds
were tested for its IC50 value. Values are mean of three independent experiments.
cD: Dependent; ND: Not dependent; NT: Not tested.
14

ACCEPTED MANUSCRIPT
Table 2. The growth inhibition (IC50, µM) of series 3 compounds against MDA-MB-231 and
MCF-7.
Compound
R3
R4
Reporter Assaya
MDA-MB-231b
MCF-7b
13a
17a
18a
H
F
F
Dc
Dc
Dc
9.04 ± 1.06
NTc
23.75 ± 1.67
NTc
5.00 ± 0.57
20.77 ± 0.78
aReporter assay. Cells were incubated with corresponding compounds of 3 concentrations 1 µM, 10 µM and 50 µM. Compounds were
tested its dose-dependent antagonism activity.
bMTT methods. Cells were incubated with corresponding compounds at a concentration from 0.195 µM to 100 µM for 48 h. Compounds
were tested its IC50 value. Values are mean of three independent experiments.
cD: Dependent; ND: Not dependent; NT: Not tested.
Table 3. The growth inhibition (IC50, µM) of series 4 and 5 compounds against MDA-MB-231 and
MCF-7.
15

ACCEPTED MANUSCRIPT
Compound
R1
R2
isomer
Reporter Assaya
MDA-MB-231b
MCF-7b
20a
20b
20c
20d
20e
20f
20g
20h
CH3
CH3
CH3
CH3
CH3
CH3
CH3
CH3
4-CH(CH3)2
4-OCF3
4-OCH3
4-SCH3
4-C(CH3)3
4-CF3
Z
Z
Z
Z
Z
Z
Z
Z
Dc
NDc
Dc
8.48 ± 0.68
NTc
27.61 ± 1.64
NTc
25.72 ± 0.82
NTc
104.00 ± 7.39
NTc
NDc
Dc
14.33 ± 0.49
NTc
38.39 ± 1.708
NTc
NDc
NDc
Dc
3-CF3
NTc
NTc
2-CF3
40.80 ± 3.10
62.28 ± 11.40
20i
Z
Dc
53.29 ± 2.81
124.5 ± 11.11
21a
21b
H
H
4-CH(CH3)2
4-OCH3
E
E
NDc
NDc
NTc
NTc
NTc
NTc
aReporter assay. Cells were incubated with corresponding compounds of 3 concentrations 1 µM, 10 µM and 50 µM. Compounds were
tested its dose-dependent antagonism activity.
bMTT methods. Cells were incubated with corresponding compounds at a concentration from 0.195 µM to 100 µM for 48 h. Compounds
were tested its IC50 value. Values are mean of three independent experiments.
cD: Dependent; ND: Not dependent; NT: Not tested.
16

ACCEPTED MANUSCRIPT
Fig 3. Summary of the structure-activity relationship.
3.4 The binding affinity of K-8008 derivatives
Fluorescence quenching assay was utilized to analyze the binding property of K-8008
derivatives. The compounds were titrated into the cuvette filled with a RXRα-LBD protein solution
(1 µmol/L) with an increasing concentration from 0.5 µmol/L to 35 µmol/L. The fluorescence
absorption excited by RXRα-LBD was quenched due to the interaction between the protein and
compounds. We used K-80003 as a positive control that was reported before as a unique RXRα
modulator with a dissociation constant (Kd) of 2.4 µM by means of 9-cis-RA competition binding
assay [21]. Using the fluorescence quenching assay, the dissociation constant of K-80003 was
evaluated as 2.38 ± 0.17 µM (Fig. 4A), which is consistent with the data from the 9-cis-RA
competition binding assay. We then measured the dissociation constant (Kd) of three compounds
(K-8008, 8b and 18a) that exerted enhanced apoptotic activity. Kd of 2.05 ± 0.26 µM, 1.59 ± 0.20
µM and 1.13 ± 0.14 µM were obtained for K-8008, 8b and 18a respectively (Fig. 4). The results
showed that both of 8b and 18a bind to RXRα slightly better than K-8008 which was consistent
with the apoptotic activity results from the MTT assay.
17

ACCEPTED MANUSCRIPT
Fig 4. Fluorescence quenching assay. 1 µmol/L protein solution was added in cuvette and
compounds with increasing concentration of 0.5 µmol/L to 35 µmol/L was titrated to the cuvette.
The excitation wavelength was set at 280 nm and emission was detected from 290 nm to 450 nm.
The value of the fluorescence intensity in 300 nm were recorded as F. Y-axis value was set to F-F0
and X-axis value was set to the equivalence ratio of compounds concentration to protein
concentration. Four compounds were tested: (A) K-80003 (B) K-8008 (C) compound 8b (D)
compound 18a.
3.5 K-8008 derivatives induce the cleavage of poly (ADP-ribose) polymerase (PARP)
We further evaluated the apoptotic effect of 8b and 18a at protein level and its dependency on
RXRα. Firstly, poly (ADP-ribose) polymerase (PARP) cleavage assay showed that both the
compounds (8b and 18a) exerted improved PARP cleavage activity in comparison with K-8008
18

ACCEPTED MANUSCRIPT
(Fig. 5A and 5C). Furthermore, their PARP cleavage effect was dose-dependent. Previously, we
reported that sulindac analogs could inhibit the TNFα-induced AKT activation in cancer cells and
enhanced the TNFα -induced apoptosis [19]. Thus, we further tested the PARP cleavage effect of
compound 8b or 18a in combination with TNFα. Fig. 5B and 5D showed that treatment of HeLa
cells or MCF-7 cells with compound 8b or 18a in combination with TNFα could significantly
enhance the cleavage PARP. Knocking down RXRα expression in HeLa cells by siRNA weakened
the effect of compounds 8b and 18a on inducing the PARP cleavage (Fig. 6A), implying that the
compounds’ apoptotic effect is RXRα-dependent. Similar results were observed in MCF-7 cancer
cell (Fig. 6B).
Fig 5. Induction of PARP cleavage by K-8008 (labeled as k8) and its derivatives without (A) or
with TNF-α (B) in HeLa Cells and without (C) or with TNF-α (D) in MCF-7 Cells. Cells cultured in
medium with 0% FBS were treated with compounds and/or TNF-α (20 ng/mL) for 8 hours in Hela
or 12 hours in MCF-7 and analyzed for PARP cleavage by immunoblotting.
19

ACCEPTED MANUSCRIPT
Fig 6. RXRα siRNA transfection inhibits the apoptotic effect of K-8008 (labeled as k8). (A) HeLa
cells transfected with control or RXRα siRNA for 48 h were treated with K-8008 and its derivatives
8b or 18a (20 µM) for 8 h and analyzed by immunoblotting. (B) MCF-7 cells transfected with
control or RXRα siRNA for 48 h were treated with 8b or 18a (30 µM) for 12 h and analyzed by
immunoblotting.
3.6 Molecular docking studies
Compounds 8b and 18a displayed similar binding affinities to RXRα-LBD and comparable
biological activities, though they are different structurally. Thus, we were intrigued to understand
their binding modes, and therefore perform molecular docking to study their binding
conformations.
For compound 8b, the original goal was to mimic the binding mode of K-8008 while
introducing some modification in the R2 group (Table 1) to enhance its interaction with the protein
in the lipophilic tail. Therefore, the K-8008/RXRα-LBD complex crystal structure (PDB code:
4N8R) was used for docking study. The results (Fig. 7A) showed that the binding mode of 8b was
almost identical to K-8008 [20]. Notably, trifluoromethoxy of 8b was positioned closely to Ala271
and form stronger interaction with Ala 271 compared with K-8008 through its fluorine atoms (Fig.
7B). This may explain why compound 8b exerts slightly enhanced activity over K-8008.
Compound 18a was designed based on the bioisostere concept by replacing the carboxylic
acid in K-80003 with hydroxyformamide. Thus, the K-80003/RXRα-LBD complex crystal
structure (PDB code: 5TBP) was used to carry out the docking study. As 3 K-80003 molecules can
bind to the same large hydrophobic cavity where only one K-8008 molecule is bound in the crystal
structure of RXRα-LBD/K-80008 complex [23], we assumed that like K-80003 3 18a molecules
would bind to each large hydrophobic cavity in the RXRα-LBD tetramer, and therefore docked 3
18a molecules separately to the three different regions where the 3 K-80003 molecules bind (Fig.
8). In general, 18a molecules (labeled as 18a-A, 18a-B and 18a-C) bound in a similar fashion to the
20

ACCEPTED MANUSCRIPT
K-80003 molecule, especially 18a-A and 18a-B (Fig. 8). As K-80003A molecule, 18a-A could
form hydrogen bond with Trp305 in chain A. However, the π-π interaction between Trp305 and the
ligand seems to be stronger for 18a-A because both of the two benzene rings of 18a-A could
contribute to the π-π interaction whereas only one of the two benzene rings of K-80003A could
form π-π interaction with Trp305. 18a-B was predicted to form hydrogen bond with backbone of
Arg302 in chain C, differently from K-80003B which forms a H-bond with Trp305 in chain C.
Other than that, 18a-B and K-80003B make similar hydrophobic interaction with the protein.
18a-C could make stronger interaction with the protein than K-80003C by forming 4 hydrogen
bonds with Gln275, Arg316 and backbones of Ala327 and Leu309 in chain A (Fig. 8). Overall, our
docking results showed that 18a may display an enhanced bind affinity to RXRα -LBD tetramer
than K-80003.
Fig. 7. Molecular docking study of compound 8b. (A) Compound 8b was docked to the
K-8008/RXRα-LBD complex crystal structure (PDB code: 4N8R) and superimposed with K-8008.
Compound 8b was showed in red sticks and K-8008 in green sticks. The binding pocket of 8b was showed
in yellow surface. (B) The interactions between compound 8b and RXRα in 2-D drawing.
21

ACCEPTED MANUSCRIPT
Fig. 8. Molecular docking studies of compound 18a. Compound 18a was separately docked to the regions
where the three molecules of K-80003 bind in the K-80003/RXRα crystal structure (PDB code: 5TBP). 18a
was showed in yellow sticks, K-80003 in green sticks and the amino acid residue in white sticks. The
hydrogen bonds between 18a and RXRα were showed in yellow dashed lines and K-80003 and RXRα in
orange dashed lines.
4. Conclusion
In summary, we have reported studies on a series of new K-8008 derivatives on their SAR as
anticancer agents targeting RXRα. Two new compounds, 8b and 18a, showed slightly enhanced
anticancer activity compared to K-8008. Compound 8b was 1.4-fold (IC50 = 8.90 ± 0.72 µM) more
potent than K-8008 in MDA-MB-231 cancer cell line and 1.6-fold (IC50 =17.34 ± 0.78 µM) in
MCF-7 and the 18a compound was 2.5-fold (IC50 = 5.00 ± 0.57 µM) more potent than K-8008 in
MDA-MB-231 cancer cell line and 1.3-fold (IC50 = 20.77 ± 0.78 µM) in MCF-7. The dissociation
constant (Kd) of 8b and 18a was 1.59 ± 0.20 µM and 1.13 ± 0.14 µM, respectively. Further
22

ACCEPTED MANUSCRIPT
biological evaluation revealed that 8b and 18a are more potent than K-8008 in inducing PARP
cleavage and the apoptotic effects of the compounds are RXRα-dependent. The studies also
demonstrated the feasibility of replacing the polar group in K-8008 with different carboxyl
bioisosteres as an optimization approach.
5. Experiment section
5.1 Chemistry
5.1.1 General procedure A: the synthesis of indene derivative from appropriate indene moiety.
To a solution of corresponding indene products (0.25 mmol, 1.0 equiv) in MeOH (5.0 mL) was
added 2.5 N NaOMe (0.75 mmol, 0.25 mL, 3.0 equiv) at room temperature to get an orange mixture.
After stirring for 30 min, the mixture was added with appropriate aromatic aldehyde (0.5 mmol, 2
equiv). The resulting mixture was refluxed at 80 °C for 4 h. After concentrated under reduced
pressure, the residue was acidified with a 1 N HCl solution to pH 4.0~6.0. After stirring for another
0.5 h at room temperature, the mixture was extracted with EtOAc (15 mL *3). The combined
organic layers were dried over anhydrous Na2SO4, filtered, and concentrated under reduced
pressure. The residue was purified by flash chromatography to afford indene derivative [21].
5.1.2 Compound 1-6a.
Compound 1-6a was synthesized according to the reference [19, 20].
5.1.3 Compound 7a.
Compound 7a was synthesized according to the reference [20, 21].
5.1.4 (Z)-5-(2-(2-methyl-1-(naphthalen-1-ylmethylene)-1H-inden-3-yl)ethyl)-1H-tetrazol (8a)
Compound 8a was synthesized according to the general procedure A. Yellow solid, 74 mg,
yield: 81.7%. 1H NMR (600 MHz, DMSO-d6) δ 8.03 (t, J = 7.06 Hz, 2H), 7.96 (d, J = 8.25 Hz, 1H),
23

ACCEPTED MANUSCRIPT
7.65 (s, 1H), 7.63 (d, J = 6.97 Hz, 1H), 7.61 (d, J = 7.89 Hz, 1H), 7.58 (d, J = 8.07 Hz, 1H), 7.54 (t,
J = 7.50 Hz, 1H), 7.23 (d, J = 7.52 Hz, 1H), 7.12 (t, J = 7.34 Hz, 1H), 6.73 (t, J = 7.43 Hz, 1H), 6.66
(d, J = 7.52 Hz, 1H), 3.15 (t, J = 7.61 Hz, 2H), 3.03 (t, J = 7.52 Hz, 2H), 2.12 (s, 3H); 13C NMR (151
MHz, DMSO-d6) δ 144.2, 142.8, 137.7, 134.8, 134.6, 134.3, 133.7, 131.4, 129.0, 128.9, 128.5,
128.3, 127.4, 127.0, 126.8, 126.0, 125.4, 124.7, 122.7, 118.3, 23.9, 22.7, 10.2. HRMS (ESI) calcd
for C24H19N4 [M-H]-: 363.1610; found: 363.1614.
5.1.5 (Z)-5-(2-(2-methyl-1-(4-(trifluoromethoxy)benzylidene)-1H-inden-3-yl)ethyl)-1H-tetrazole
(8b)
Compound 8b was synthesized according to the general procedure A. Yellow solid, 94 mg,
yield: 94.7%. 1H NMR (600 MHz, DMSO-d6) δ 7.65 (dd, J = 1.00, 8.44 Hz, 2H), 7.46 (d, J = 8.07
Hz, 2H), 7.23 - 7.26 (m, 2H), 7.16 - 7.21 (m, 2H), 6.92 (t, J = 7.90 Hz, 1H), 3.11 (t, J = 7.61 Hz,
2H), 2.99 (t, J = 7.52 Hz, 2H), 1.97 (s, 3H); 13C NMR (151 MHz, DMSO-d6) δ 148.4, 144.3, 141.8,
137.8, 136.2, 134.9, 134.1, 131.6, 129.1, 128.6, 125.0, 122.4, 121.5, 119.72, 118.4, 23.9, 22.6, 10.1.
HRMS (ESI) calcd for C21H16F3N4O [M-H]-: 397.1276; found: 397.1278.
5.1.6 (Z)-5-(2-(1-(4-(tert-butyl)benzylidene)-2-methyl-1H-inden-3-yl)ethyl)-1H-tetrazole (8c)
Compound 8c was synthesized according to the general procedure A. Yellow solid, 63 mg,
yield: 68.5%. 1H NMR (600 MHz, CHLOROFORM-d) δ 7.47 (d, J = 7.52 Hz, 1H), 7.38 - 7.42 (m,
4H), 7.10 (dd, J = 1.00, 7.30 Hz, 1H), 7.09 (d, J = 7.15 Hz, 1H), 7.07 (s, 1H), 6.88 (dt, J = 1.00, 7.30
Hz, 1H), 3.27 (t, J = 7.52 Hz, 2H), 3.05 (t, J = 7.61 Hz, 2H), 1.92 (s, 3H), 1.36 (s, 9H); 13C NMR
(151 MHz, CHLOROFORM-d) δ 151.4, 143.5, 140.6, 135.7, 135.6, 134.4, 133.6, 130.8, 129.2,
127.8, 125.3, 124.6, 122.9, 117.4, 34.8, 31.3, 24.0, 22.6, 10.0. HRMS (ESI) calcd for C24H25N4
[M-H]-: 369.2079; found: 369.2079.
5.1.7 (Z)-5-(2-(1-(4-methoxybenzylidene)-2-methyl-1H-inden-3-yl)ethyl)-1H-tetrazole (8d)
24

ACCEPTED MANUSCRIPT
Compound 8d was synthesized according to the general procedure A. Yellow solid, 36.29 mg,
yield: 42.2%. 1H NMR (600 MHz, DMSO-d6) δ 7.50 (d, J = 8.62 Hz, 2H), 7.43 (d, J = 7.52 Hz, 1H),
7.24 (d, J = 7.34 Hz, 1H), 7.20 (s, 1H), 7.18 (dt, J = 1.00, 7.50 Hz, 1H), 7.04 (d, J = 8.62 Hz, 2H),
6.93 (dt, J = 1.01, 7.47 Hz, 1H), 3.83 (s, 3H), 3.11 (t, J = 7.61 Hz, 2H), 2.99 (t, J = 7.61 Hz, 2H),
13
1.97 (s, 3H); C NMR (151 MHz, DMSO-d6) δ 159.8, 144.1, 140.0, 136.6, 135.1, 134.3, 131.4,
130.9, 128.9, 128.1, 124.7, 122.3, 118.2, 114.4, 55.7, 23.9, 22.8, 10.2. HRMS (ESI) calcd for
C21H19N4O [M-H]-: 343.1559; found: 343.1561.
5.1.8 (Z)-5-(2-(2-methyl-1-(4-(methylthio)benzylidene)-1H-inden-3-yl)ethyl)-1H-tetrazole (8e)
Compound 8e was synthesized according to the general procedure A. Yellow solid, 46 mg,
yield: 51.5%. 1H NMR (600 MHz, DMSO-d6) δ 7.49 (d, J = 8.07 Hz, 2H), 7.39 (d, J = 7.70 Hz, 1H),
7.35 (d, J = 7.89 Hz, 2H), 7.24 (d, J = 7.34 Hz, 1H), 7.20 (s, 1H), 7.17 (dt, J = 1.00, 7.20 Hz, 1H),
6.93 (dt, J = 1.00, 8.10 Hz, 1H), 3.11 (t, J = 7.61 Hz, 2H), 2.99 (t, J = 7.61 Hz, 2H), 2.54 (s, 3H),
13
1.97 (s, 3H); C NMR (151 MHz, DMSO-d6) δ 144.2, 140.7, 139.3, 137.1, 135.1, 134.2, 133.0,
130.4, 130.4, 128.3, 125.9, 124.8, 122.5, 118.3, 23.9, 22.7, 14.8, 10.2. HRMS (ESI) calcd for
C21H19N4S [M-H]-: 359.1331; found: 359.1334.
5.1.9 (Z)-5-(2-(1-(2-chlorobenzylidene)-2-methyl-1H-inden-3-yl)ethyl)-1H-tetrazole (8f)
Compound 8f was synthesized according to the general procedure A. Yellow solid, 60 mg,
yield: 69.3%. 1H NMR (600 MHz, DMSO-d6) δ 7.62 (dd, J = 1.00, 8.44 Hz, 1H), 7.57 (dd, J = 1.19,
7.43 Hz, 1H), 7.47 (dt, J = 1.56, 7.66 Hz, 1H), 7.44 (dt, J = 1.19, 7.47 Hz, 1H), 7.25 (d, J = 7.52 Hz,
1H), 7.19 (dt, J = 1.00, 6.40 Hz, 1H), 7.14 (s, 1H), 6.89 (t, J = 7.89 Hz, 2H), 3.15 (t, J = 7.61 Hz,
2H), 3.01 (t, J = 7.61 Hz, 2H), 2.01 (s, 3H); 13C NMR (151 MHz, DMSO-d6) δ 144.2, 142.4, 138.3,
135.3, 134.5, 134.2, 133.0, 131.7, 130.6, 130.0, 128.7, 127.6, 126.8, 125.0, 122.5, 118.5, 23.8, 22.5,
10.1. HRMS (ESI) calcd for C20H16ClN4 [M-H]-: 347.1064; found: 347.1066.
25

ACCEPTED MANUSCRIPT
5.1.10 (Z)-5-(2-(1-(3-chlorobenzylidene)-2-methyl-1H-inden-3-yl)ethyl)-1H-tetrazole (8g)
Compound 8g was synthesized according to the general procedure A. Yellow solid, 79 mg,
yield: 90.7%. 1H NMR (600 MHz, DMSO-d6) δ 7.52 (s, 1H), 7.48 (d, J = 7.15 Hz, 1H), 7.44 - 7.47
(m, 2H), 7.23 (d, J = 7.34 Hz, 1H), 7.14 - 7.20 (m, 3H), 6.90 (dt, J = 1.00, 7.90 Hz, 1H), 3.12 (t, J =
13
7.52 Hz, 2H), 2.99 (t, J = 7.61 Hz, 2H), 1.95 (s, 3H); C NMR (151 MHz, DMSO-d6) δ 144.3,
142.1, 139.1, 138.0, 134.9, 134.1, 133.7, 130.8, 129.1, 128.9, 128.6, 128.4, 128.2, 125.0, 122.5,
118.5, 23.8, 22.6, 10.1. HRMS (ESI) calcd for C20H16ClN4 [M-H]-: 347.1064; found: 347.1066.
5.1.11 (Z)-5-(2-(1-(4-chlorobenzylidene)-2-methyl-1H-inden-3-yl)ethyl)-1H-tetrazole (8h)
Compound 8h was synthesized according to the general procedure A. Yellow solid, 55 mg,
yield: 63.7%. 1H NMR (600 MHz, DMSO-d6) δ 7.49 - 7.55 (m, 4H), 7.23 (t, J = 8.30 Hz, 2H), 7.15
- 7.20 (m, 2H), 6.91 (t, J = 7.43 Hz, 1H), 3.11 (t, J = 7.52 Hz, 2H), 2.99 (t, J = 7.52 Hz, 2H), 1.96 (s,
13
3H); C NMR (151 MHz, DMSO-d6) δ 155.9, 144.2, 141.7, 137.7, 135.7, 135.0, 134.1, 133.3,
131.5, 129.3, 129.0, 128.5, 124.9, 122.5, 118.4, 23.9, 22.6, 10.1. HRMS (ESI) calcd for C20H16ClN4
[M-H]-: 347.1064; found: 347.1066.
5.1.12 (Z)-5-(2-(2-methyl-1-(2-(trifluoromethyl)benzylidene)-1H-inden-3-yl)ethyl)-1H-tetrazole
(8i)
Compound 8i was synthesized according to the general procedure A. Yellow solid, 56 mg,
yield: 58.5%. 1H NMR (600 MHz, DMSO-d6) δ 7.89 (d, J = 7.89 Hz, 1H), 7.75 (t, J = 7.43 Hz, 1H),
7.67 (t, J = 7.70 Hz, 1H), 7.59 (d, J = 7.52 Hz, 1H), 7.28 (d, J = 1.00 Hz, 1H), 7.25 (d, J = 7.34 Hz,
1H), 7.17 (dt, J = 1.00, 8.10 Hz, 1H), 6.83 (dt, J = 1.00, 7.90 Hz, 1H), 6.49 (d, J = 7.70 Hz, 1H),
3.13 (t, J = 7.70 Hz, 2H), 3.00 (t, J = 7.70 Hz, 2H), 2.00 (s, 3H); 13C NMR (151 MHz, DMSO-d6) δ
144.2, 142.9, 138.6, 135.9, 134.3, 134.2, 133.1, 131.8, 129.2, 128.6, 127.43, 126.6, 126.0, 125.0,
124.59, 122.6, 118.6, 23.8, 22.5, 10.0. HRMS (ESI) calcd for C21H16F3N4 [M-H]-: 381.1327; found:
381.1329.
26

ACCEPTED MANUSCRIPT
5.1.13 (Z)-5-(2-(2-methyl-1-(3-(trifluoromethyl)benzylidene)-1H-inden-3-yl)ethyl)-1H-tetrazole
(8j)
Compound 8j was synthesized according to the general procedure A. Yellow solid, 39 mg,
yield: 40.8%. 1H NMR (600 MHz, DMSO-d6) δ 7.85 (d, J = 6.42 Hz, 2H), 7.78 (d, J = 7.70 Hz, 1H),
7.72 (t, J = 7.98 Hz, 1H), 7.31 (s, 1H), 7.27 (d, J = 7.34 Hz, 1H), 7.20 (dt, J = 1.00, 7.90 Hz, 1H),
7.11 (d, J = 7.70 Hz, 1H), 6.91 (dt, J = 1.00, 7.90 Hz, 1H), 3.13 (t, J = 7.61 Hz, 2H), 3.01 (t, J = 7.61
Hz, 2H), 1.98 (s, 3H); 13C NMR (151 MHz, DMSO-d6) δ 144.4, 142.3, 138.1, 138.0, 135.0, 134.0,
133.6, 130.1, 129.8, 128.8, 128.8, 126.2, 125.1, 125.0, 124.6, 122.2, 118.6, 23.9, 22.6, 10.1. HRMS
(ESI) calcd for C21H16F3N4 [M-H]-: 381.1327; found: 381.1328.
5.1.14 (Z)-5-(2-(2-methyl-1-(4-(trifluoromethyl)benzylidene)-1H-inden-3-yl)ethyl)-1H-tetrazole
(8k)
Compound 8k was synthesized according to the general procedure A. Yellow solid, 57 mg,
yield: 60%. 1H NMR (600 MHz, DMSO-d6) δ 7.84 (d, J = 8.25 Hz, 2H), 7.74 (d, J = 8.07 Hz, 2H),
7.28 (s, 1H), 7.26 (d, J = 7.34 Hz, 1H), 7.20 (t, J = 7.43 Hz, 1H), 7.15 (d, J = 7.52 Hz, 1H), 6.93 (t,
J = 7.34 Hz, 1H), 3.13 (t, J = 7.52 Hz, 2H), 3.01 (t, J = 7.61 Hz, 2H), 1.98 (s, 3H); 13C NMR (151
MHz, DMSO-d6) δ 144.3, 142.4, 141.3, 138.3, 134.9, 134.0, 130.4, 128.8, 125.9, 125.8, 125.8,
125.1, 124.7, 122.5, 118.5, 23.9, 22.6, 10.1. HRMS (ESI) calcd for C21H16F3N4 [M-H]-: 381.1327;
found: 381.1329.
5.1.15 (E)-5-(2-(1-(4-isopropylbenzylidene)-1H-inden-3-yl)ethyl)-1H-tetrazole (9a)
Compound 9a was synthesized according to the general procedure A. Yellow solid, 70 mg,
yield: 84.0%. 1H NMR (600 MHz, DMSO-d6) δ 7.81 (d, J = 7.15 Hz, 1H), 7.60 (d, J = 8.25 Hz, 2H),
7.55 (s, 1H), 7.38 (d, J = 7.34 Hz, 1H), 7.35 (d, J = 8.07 Hz, 2H), 7.29 (dt, J = 1.10, 7.34 Hz, 1H),
7.26 (dt, J = 0.92, 7.70 Hz, 1H), 6.89 (s, 1H), 3.35 (t, J = 8.10 Hz, 2H), 3.11 (t, J = 7.52 Hz, 2H),
27

ACCEPTED MANUSCRIPT
2.94 (quind, J = 6.87, 13.80 Hz, 1H), 1.24 (d, J = 6.79 Hz, 6H); 13C NMR (151 MHz, DMSO-d6) δ
149.5, 146.6, 141.5, 138.5, 137.9, 134.6, 130.7, 128.0, 127.7, 127.4, 125.9, 122.1, 119.8, 119.2,
33.7, 25.7, 24.2, 22.1. HRMS (ESI) calcd for C22H21N4 [M-H]-: 341.1766; found: 341.1767.
5.1.16 2-(2-methyl-1H-inden-3-yl)acetonitrile (10a)
Diethyl cyanomethylphosphonate (13.409 g, 2 equiv) was dissolved in 65 mL acetonitrile in
0 °C, and was dropwise added with DBU (17.272 g, 3 equiv), stirring for 30 min. The mixture was
added with compound 4a (5 g, 1 equiv) and stirred at room temperature overnight. After
concentrated under reduced pressure, the mixture was extracted with EtOAc (15 mL *3) and
washed with brine (15 mL). The combined organic layers were dried over anhydrous Na2SO4,
filtered, and concentrated under reduced pressure. The residue was purified by flash chroma-
1
tography to afford compound 10a. White solid, 3.5 g, yield : 75%. H NMR (600 MHz,
CHLOROFORM-d) δ 7.39 (d, J = 7.34 Hz, 1H), 7.32 (d, J = 7.15 Hz, 1H), 7.31 (t, J = 7.52 Hz, 1H),
13
7.18 (dt, J = 1.19, 7.20 Hz, 1H), 3.54 (s, 2H), 3.35 (s, 2H), 2.15 (s, 3H); C NMR (151 MHz,
CHLOROFORM-d) δ 144.1, 143.2, 141.8, 126.6, 125.5, 124.8, 123.5, 118.0, 117.1, 42.9, 14.2,
13.9. HRMS (ESI) calcd for C12H11NNa+ [M+Na]+: 192.0789; found: 192.0784.
5.1.17 2-(1H-inden-3-yl)acetonitrile (10b)
Compound 10b was synthesized according to compound 10a. White solid, 4.3 g, yield: 60%.
1H NMR (600 MHz, CHLOROFORM-d) δ 7.53 (d, J = 7.89 Hz, 1H), 7.40 (t, J = 7.24 Hz, 1H), 7.37
13
(d, J = 7.34 Hz, 1H), 7.28 (t, J = 7.50 Hz, 1H), 5.65 (s, 1H), 3.08 - 3.13 (m, 4H); C NMR (151
MHz, CHLOROFORM-d) δ 167.4, 149.7, 138.1, 131.8, 127.2, 125.9, 121.6, 118.3, 85.8, 31.4, 29.9.
HRMS (ESI) calcd for C11H9NNa+ [M+Na]+: 178.0633; found: 178.0628.
5.1.18 N-hydroxy-2-(2-methyl-1H-inden-3-yl)acetimidamide (11)
28