From methylene bridged diindole to carbonyl linked benzimidazoleindole: Development of potent and metabolically stable PCSK9 modulators

Article abstract of DOI:10.1016/j.ejmech.2020.112678

Proprotein convertase subtilisin/kexin type 9 (PCSK9) is a recently validated therapeutic target for lowering low-density lipoprotein cholesterol (LDL-C). Through phenotypic screening, we previously discovered a class of small-molecules with a 2,3′-diindolymethane (DIM) skeleton that can decrease the expression of PCSK9. But these compounds have low potency and low metabolically stability. After performing structure-activity relationship (SAR) optimization by nitrogen scan, deuterium substitution and fluorine scan, we identified a series of much more potent and metabolically stable PCSK9 modulators. A preliminary in vivo pharmacokinetic study was performed for representative analogues difluorodiindolyketone (DFDIK) 12 and difluorobenzoimidazolylindolylketone (DFBIIK-1) 13. The in vitro metabolic stability correlate well with the in vivo data. The most potent compound 21 has the EC₅₀ of 0.15 nM. Our SAR studies also indicated that the NH on the indole ring of 21 can tolerate more function groups, which may facilitate the mechanism of action studies and also allow further improvement of the pharmacological properties.

Full text of DOI:10.1016/j.ejmech.2020.112678

Journal Pre-proof
From Methylene Bridged Diindole to Carbonyl Linked Benzimidazoleindole:
Development of Potent and Metabolically Stable PCSK9 Modulators
Haibo Xie, Ka Yang, Gabrielle N. Winston-McPherson, Donnie S. Stapleton, Mark P.
Keller, Alan D. Attie, Kerry A. Smith, Weiping Tang
PII:
S0223-5234(20)30650-4
DOI:
https://doi.org/10.1016/j.ejmech.2020.112678
EJMECH 112678
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 13 June 2020
Revised Date: 16 July 2020
Accepted Date: 18 July 2020
Please cite this article as: H. Xie, K. Yang, G.N. Winston-McPherson, D.S. Stapleton, M.P. Keller, A.D.
Attie, K.A. Smith, W. Tang, From Methylene Bridged Diindole to Carbonyl Linked Benzimidazoleindole:
Development of Potent and Metabolically Stable PCSK9 Modulators, European Journal of Medicinal
Chemistry, https://doi.org/10.1016/j.ejmech.2020.112678.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. 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.
© 2020 Elsevier Masson SAS. All rights reserved.

Graphical Abstract

Highlights
•
•
•
•
Novel PCSK9 modulators were synthesized and evaluated in cell-based assays.
Compound 21 can potently reduce PCSK9 protein in cell-based assays.
Compound 21 has improved metabolic stability.
Compound 21 could be a potential candidate for the treatment of hyperlipidemia
.

From
Methylene
Bridged
Diindole
to
Carbonyl
Linked
Benzimidazoleindole: Development of Potent and Metabolically Stable
PCSK9 Modulators
Haibo Xie,1‡ Ka Yang 1‡ Gabrielle N. Winston-McPherson,1 Donnie S. Stapleton, 2 Mark P. Keller, 2 Alan D. Attie,
2 Kerry A. Smith, 1 and Weiping Tang1,3*
1 School of Pharmacy, University of Wisconsin-Madison, Madison, Wisconsin 53705, United States
2 Department of Biochemistry, University of Wisconsin-Madison, Madison, Wisconsin 53706, United
States
3
Department of Chemistry, University of Wisconsin-Madison, Madison, Wisconsin 53706, United
States
*Corresponding author: School of Pharmacy, University of Wisconsin-Madison, Madison, WI 53705,
USA. E-mail: weiping.tang@wisc.edu (Weiping Tang)
‡These authors (H. X. and K. Y.) contributed equally.
Keywords
PCSK9, Phenotypic screening, Fluorine scan, Metabolic stability, Indole
Abstract
Proprotein convertase subtilisin/kexin type 9 (PCSK9) is a recently validated therapeutic target for
lowering low-density lipoprotein cholesterol (LDL-C). Through phenotypic screening, we previously
discovered a class of small-molecules with a 2,3’-diindolymethane (DIM) skeleton that can decrease the
expression of PCSK9. But these compounds have low potency and low metabolically stability. After
performing structure-activity relationship (SAR) optimization by nitrogen scan, deuterium substitution
and fluorine scan, we identified a series of much more potent and metabolically stable PCSK9
modulators. A preliminary in vivo pharmacokinetic study was performed for representative analogues
1

difluorodiindolyketone (DFDIK) 12 and difluorobenzoimidazolylindolylketone (DFBIIK-1) 13. The in
vitro metabolic stability correlate well with the in vivo data. The most potent compound 21 has the EC50
of 0.15 nM. Our SAR studies also indicated that the NH on the indole ring of 21 can tolerate more
function groups, which may facilitate the mechanism of action studies and also allow further
improvement of the pharmacological properties.
Introduction
Cardiovascular diseases (CVD) is the number one cause of death world-wide. High levels of low-density
lipoprotein (LDL), clinically known as hypercholesterolemia or hyperlipidemia, are strongly associated
with CVD. Although statin therapy has been very successful for many patients, about 20% of them are
not able to achieve target LDL levels due to the adverse effects of statin therapy. Proprotein convertase
subtilisin/kexin type 9 (PCSK9) is secreted from the liver, binds to the LDL receptor (LDLR), and
causes its lysosomal degradation (Figure 1).[1,2] Inhibition of this interaction upregulates the LDLR and
leads to a drastic lowering of LDL and the risk of heart disease, making it the most attractive new target
for lowering LDL-cholesterol. Two anti-PCSK9 monoclonal antibodies (mAb), Evolocumab and
Alirocumab, were recently approved by FDA.[3,4] By disrupting the PCSK9-LDLR interaction, these
mAbs can reduce over 60% LDL-cholesterol in patients. However, antibodies require injection and they
are much more expensive than small molecules. Peptides that mimic the epidermal growth factor
precursor homology domain
A
(EGF-A) domain of the LDLR[5–13], PCSK9 antisense
oligonucleotides[14–18] and RNAi of PCSK9[19–21] are also under development, though they have the
same limitations as antibodies with respect to cost and oral availability. Small molecules have the
potential to overcome these limitations and have the flexibility to be tuned for desired pharmacological
properties[22,23].
2

Figure 1. PCSK9 mediated lysosomal degradation of LDLR
Phenotypic screening and target-based screening are two of the major approaches for drug discovery.
Target based screening allows for more rational design as results in biochemical assays are easier to
interpret, especially after the target-ligand interactions are fully characterized.[24] Phenotypic screening
led to the discovery of a number of first-in-class drugs in history.[25] It has attracted more and more
interest from both industry and academic in recent years[26,27] across many areas of diseases.[28–31]
Phenotypic screening aimed at inhibition of PCSK9 protein production is a promising approach for the
discovery of small molecule PCSK9 inhibitors or modulators. Berberine (BBR),[32,33]
tetrahydroprotoberberine derivatives (R)-22,[34] difluoro-2,3’-diindolymethane (DFDIM, 1),[35]
7030B-C5[36] and BRD8518[37] were discovered with the activity of reduction of secreted PCSK9
protein (Figure 2). R-IMPP showed a novel molecular mechanism of action (MoA) that targeted the 80S
ribosome and inhibited PCSK9 protein translation, and was discovered by phenotypic screening in a
CHO-K1 cell line overexpressing recombinant ProLabel-tagged PCSK9.[38] PF-06446846, an analogue
of R-IMPP, was subsequently developed with improved pharmaceutical properties.[39] By directly
targeting PCSK9 protein, a number of small molecules were developed, including LDLL-1dlnr[40] and
3f[41] that disrupts the interaction of PCSK9 and LDLR, Heparin oligosaccharide (e.g. Heparin I) that
prevents the recruitment of PCSK9 to LDLR via heparan sulfate proteoglycans (HSPG) on cell
membrane,[42] and a PCSK9 degrader using CMPD9 as the ligand.[43]
3

Figure 2. Small-molecule PCSK9 modulators derived from phenotypic screening: BBR, (R)-22,
DFDIM, 7030B-C5, BRD8518 and R-IMPP; Small-molecule PCSK9 inhibitors or binders derived from
targeted-based approach: LDLL-1dlnr, 3f, Heparin I and CMPD9.
Results and Discussion
Previously, we have discovered a class of small-molecules with a DIM skeleton that can decrease the
expression of PCSK9 with sub-micromolar IC50s in HepG2 cells.[35] After performing preliminary
structure–activity relationship (SAR) optimization, the PCSK9 inhibitor DFDIM 1 with 202 ± 78 nM
IC50 was developed. In HepG2 cells, compound 1 decreased PCSK9 protein levels in both cell lysate and
media as measured by sandwich ELISA. While working on the elucidation of the mechanism of action
(MoA) and target identification of 1, we also performed a systematic SAR investigation based on the
structure of 1 to obtain more potent, soluble, and metabolically stable analogues for in vivo
pharmacokinetics and future proof-of-concept efficacy studies. However, unlike target-based SAR
studies, phenotype-based SAR studies are more challenging due to multiple factors that can impact the
activity of the compounds. In addition to the potency for decreasing PCSK9 protein level, we also
optimized the liver microsomal stability of the compounds as liver is the primary tissue we need to
target and compound 1 has obvious metabolic liability.
4

Figure 3. SAR of N-substitutions of benzimidazole. HepG2 cells were treated with compounds at
various concentrations for 48 h. Medium sample were analyzed by human PCSK9 ELISA. Data was
normalized to vehicle (DMSO) treated group and bar graph represented as mean of relative PCSK9
protein level (n = 3) with ± SD as error bar. Statistical significance was analyzed by one-way ANOVA
in comparison with vehicle (DMSO) treated group. Not significant (ns), *p ≤ 0.05, ***p ≤ 0.001,
****p ≤ 0.0001.
Replacement of a CH motif by its isostere N often retains the activity while improving the metabolic
stability and solubility.[44,45] We initiated our optimization by the design of
difluorobenzoimidazolylindolylmethane (DFBIIM) 2 because of its ease of synthesis and also the
flexibility to introduce various R groups to the nitrogen of the benzoimidazole ring, including both
small/bulky and non-polar/polar substituents. The substituted group on the N atom of benzimidazole was
systematically explored (Figure 3). Among the diverse range of substituents, the addition of a methyl
group to the parent compound 2 significantly improved the potency of compound 8, a classical “methyl
effect”.[46,47]
We suspected that the methylene group between the two indoles in 1, or between the benzoimidazole
and indole in 2 or 8, would be one of the most metabolically labile sites, because they are activated by
two adjacent aryl rings towards potential oxidative reactions. Replacement of one or more hydrogens by
isotopic deuterium can often increase the metabolic stability, while retaining the biological activity. In
2017, the first deuterated drug, deutetrabenazine, was approved by US FDA.[48–51] Longer half-life
and the reduction of toxic metabolites are the major advantages of deuterated drugs.[52,53] Analogue 11
was designed and synthesized by deuterating the potential metabolically labile methylene group. We
then thoroughly examined the metabolic stability of PCSK9 modulators 1 and 2 across three different
species to establish the correlation among mice, rat, and human (Table 1). Both compounds 1 and 2 are
5

indeed relatively metabolically labile, compared to their analogues in the same series. The half-life in
human is generally 2-4 times longer than those in mice and rats. The metabolic stability of 11 was
increased over two times compared to compound 2 in both mice and rats. For example, the half-life of
11 in mice was increased from 7.2 to 16.3 minutes, suggesting that the methylene bridge is indeed a
metabolically labile sites as we suspected. From these results, we further converted the methylene group
of compounds 1 and 2 to carbonyl group by SeO2. The metabolic stability of ketones DFDIK 12 and
DFBIIK-1 13 in both mice and rats was increased significantly over compounds 1 and 2, respectively.
Table 1. Microsomal stability of potential PCSK9 modulators
Human Liver Microsomal Mice Liver Microsomal Rat Liver Microsomal
Stability
Stability
Stability
CLint
CLint
CLint
Compound
T1/2 (min)
T1/2 (min)
T1/2 (min)
(µL/min/mg)
(µL/min/mg)
(µL/min/mg)
63.9±15.0 23.0±5.5
16.5±0.2
7.1±0.3
83.9±1.1
194.9±9.5
85.2±1.8
34.6±1.8
46.5±4.0
54.6±6.6
23.2±2.8 60.7±7.8
9.5±0.1 145.6±1.3
1
17.5±0.2
79.4±1.0
2
-
-
-
-
-
-
-
-
16.3±0.3
40.1±2.2
30.0±2.5
25.8±2.9
23.5±1.1 59.1±3.0
76.9±5.8 18.1±1.4
47.8±5.8 29.4±3.3
42.4±2.9 33.1±3.6
11
12
13
14
It is not clear if the change of a methylene bridge in 1 and 2 to a ketone bridge in 12 and 13,
respectively, would have any effect on the activity towards lowering PCSK9 protein level. We then
compared the activity of 12 and 13 with 1 and 2 for their ability to decreased PCKS9 secretion from
HepG2 cells (Figure 4). We were pleased that compounds 12 and 13 have much higher potency than
their counterparts 1 and 2 at 1
μM concentration. Nearly 90% reduction of PCSK9 protein was observed
with 1 M of PCSK9 modulator 13.
μ
6

Figure 4. SAR of compounds with methylene or ketone linkages. HepG2 cells were treated with
compounds at various concentrations for 48 h. Medium sample were analyzed by human PCSK9
ELISA. Data was normalized to vehicle (DMSO) treated group and bar graph represented as mean of
relative PCSK9 protein level (n = 3) with ± SD as error bar. Statistical significance was analyzed by
one-way ANOVA in comparison with compound 1 treated group at same concentration. Not significant
(ns), *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.
Having two potent PCSK9 modulators with much higher metabolic stability, we decided to establish the
correlation of in vivo pharmacokinetics with the in vitro metabolic stability (Table 2). The preliminary in
vivo pharmacokinetic study for compounds 12 and 13 were performed in mice. Mice were
intraperitoneally injected with 6 mg/kg of 12 or 13. Plasma samples were collected and analyzed by LC-
MS/MS. Both compounds reached their peak concentrations of about 1.0 µM near 15 minutes after
injection (Table 2). We performed one-phase-decay nonlinear fitting of all biological and technical
replicates and obtained the plasma half-life (T1/2) and other parameters of these two compounds. The
T1/2 of 12 and 13 were 1.31 and 1.64 hours, respectively. The trend of the in vivo data for compounds 12
and 13 correlate well with their in vitro metabolic stability. These results thus provide valuable
information for further optimization and in vivo studies.
Table 2. Preliminary pharmacokinetic study DFDIK 12 and DFBIIK-1 13 in micea
12
13
Route of administration
Dose (mg/kg)
IP
IP
6.0
610.3
6.0
695.6
AUC0-∞ (µg/L•h)b
7

Cmax (µg/L)c
T1/2 (h)d
322.9 ± 23.7
1.31
294.3 ± 8.9
1.64
a. Compounds were formulated in DMSO (10 %) and Tween 80 (10 %) in PBS (80 %).
b. AUC0-∞, total exposure following a single dose.
c. Cmax, maximum plasma concentration following a single dose.
d. T1/2, half-life.
Incorporation of fluorine is a common strategy in drug discovery. About 20% of FDA approved drugs
contain at least one fluorine atom across many different dug classes.[54–56] Fluorine substitutions not
only modulate the ADME profiles but also impact protein-ligand binding affinity.[57–61] A systematic
fluorine scan on ligands has become a routine approach in hit-to-lead optimization strategies.[62,63] We
previously performed a fluorine scan for the parent diindolylmethane 1.[35] In order to examine the
effect of the two fluorine atoms on the A-ring and D-ring of compound 13, we prepared their parent
analogue 14, analogues 15 and 16, which lack one of the two fluorine atoms, and analogue 17, which
has the fluorine on the 4-position instead of 5-position. The potency of those analogues was compared at
three different concentrations (Figure 5A). Each fluorine atom on either A-ring or D-ring slightly
improved the potency and two fluorine atoms cooperatively improved the potency further. The position
of fluorine on A-ring does not significantly impact the activity for PCSK9 modulation. We then
compared the microsomal stability of compounds 13 and 14 (Table 1). The two fluorine atoms in
compound 13 improved its metabolic stability in both mice and rats over the parent compound 14
without two fluorine atoms.
We then conducted a systematic fluorine scan SAR studies for analogues with the ketone linkage and an
additional methyl group on the benzoimidazole nitrogen (Figures 6B and 6C), because of the “methyl
effect” we observed for the improved potency of compound 8 over its parent compound 2. No
significant difference was observed for the PCSK9 modulation activity among the five compounds,
including compound 18 without the fluorine in the benzoimidazole ring and compounds 19-22 with the
fluorine at different positions (Figure 5B). On the other hand, the position of the fluorine atom on the
indole ring significantly impacted activity (Figure 5C). Compared to compound 23 without the fluorine
on the D-ring, fluorine substitution on the 5 and 4 positions decreases the potency, whereas fluorine
8

substitution on the 6 and 7 positions increases the potency. The SAR trend of the fluorine scan on the D-
ring of compound 2 is very similar to the trend of the fluorine scan observed for the original hit
compound 1,[35] suggesting that these two classes of compounds have the same mechanism of action.
Compound 21 has the most significant PCSK9 modulation activity at both 10 and 100 nM
concentrations (35.0% ± 4.4% and 12.5% ± 2.1% reduction of PCSK9, respectively), though compound
20 showed similar potency (34.4% ± 2.7% and 11.9% ± 0.3% reduction of PCSK9 at 10 and 100 nM,
respectively).
Figure 5. Fluorine scan SAR of carbonyl linked analogues. (A) Fluorine substitution on A-ring and D-
ring of carbonyl linked benzimidazoleindole; (B) ‘F-scan’ SAR of A-ring carbonyl linked N-methyl-
9

benzimidazoleindole (C) ‘F-scan’ SAR of D-ring of N-methyl-benzimidazoleindole. HepG2 cells were
treated with compounds at various concentrations for 48 h. Medium sample were analyzed by human
PCSK9 ELISA. Data was normalized to vehicle (DMSO) treated group and bar graph represented as
mean of relative PCSK9 protein level (n = 3) with
analyzed by one-way ANOVA in comparison with compound 14- (A), 18- (B) or 21- (C) treated group
at same concentration. Not significant (ns), *p 0.05, **p 0.01, ***p 0.001, ****p 0.0001.
± SD as error bar. Statistical significance was
≤
≤
≤
≤
We then further compared the microsomal stability of compounds 13 and 21 (Figure 6). The methyl
group in compound 21 clearly improved its metabolic stability in both mice and rats over the parent
compound 13 without the methyl group.
We further examined the full dose-response of compounds 2, 13 and 21 for PCSK9 modulation and
compared their potency with metabolic stability in Figure 6. The effect of N-methylation and carbonyl
linker replacement on activity is dramatic. The potency was improved 139-fold by the carbonyl linker
replacement in compound 13 and 17-fold by N-methylation on compound 21. A total of 2,320-fold
increased potency was achieved by two relatively simple modifications. The metabolic stability was also
increased by the two modifications as summarized in Figure 6.
Figure 6. The effect of N-methylation and carbonyl linker replacement on potency and microsomal
stability. HepG2 cells were treated with compounds at various concentrations for 48 h. Medium sample
were analyzed by human PCSK9 ELISA. Data was normalized to vehicle (DMSO) treated group and dot
plot represented as mean of relative PCSK9 protein level (n = 3) with
± SD as error bar. The dose
response curve was fitted using “log(inhibitor) vs. response (three parameters)” by GraphPad Prism.
To facilitate the mechanism of action studies, it is important to identify a position within the compound
that can be modified without significantly interfering with the biological activity. Such a position may
also help to further improve other pharmacological properties. Analogues with an N-substituent on either
the benzoimidazole ring or the indole ring are more synthetically accessible than those with substituents
10

on the benzene ring. We first prepared compounds 27-29 with different N-alkyl substituents on the
benzoimidazole ring or the indole ring and compared their activity for PCSK9 modulation at three
different concentrations (Figure 7). Clearly, the R1 position cannot tolerate a large substituent. This is
what we expected based on SAR studies shown in Figure 3. Conversely, a large substituent on the R2
position can be well tolerated as shown by the activity of compound 28, though the activity was
decreased at lower concentrations.
Figure 7. Investigation of the effect of N-substituent on the benzoimidazole and indole. HepG2 cells
were treated with compounds at various concentrations for 48 h. Medium sample were analyzed by
human PCSK9 ELISA. Data was normalized to vehicle (DMSO) treated group and bar graph
represented as mean of relative PCSK9 protein level (n = 3) with
significance was analyzed by one-way ANOVA in comparison with vehicle (DMSO) treated group. Not
significant (ns), *p 0.05, ***p 0.001, ****p 0.0001.
± SD as error bar. Statistical
≤
≤
≤
We then further conducted a dose response studies for analogues with various R-group on the indole
nitrogen (Table 3). Although all these analogues are less potent than the parent compound 21,
compounds 32 and 33 showed very promising activity with EC50 values less than 1 nM. The activity of
compounds 28, 30, and 31 was lower than the parent compound 21, when the morpholine was
introduced with an alkyl linker. Interestingly, most of the activity could be retained in compound 33,
when the morpholine was introduced with a sulfonamide alkyl linker. Replacement of the morpholine
ring in 33 by a piperidine ring in 34 significantly reduced the activity. Derivatives of compounds 32 and
33 will be prepared in the near future for target identification and the studies of mechanism of action. At
the same time, an appropriate animal model will be identified for the investigation of the full
pharmacokinetic profile and in vivo efficacy of this class PCSK9 modulators.
Table 3. SAR of substitutions on the nitrogen of indolea
11

Compound
R
EC50 (nM)
0.15±0.03
1.21±0.34
11.0±3.5
21
27
28
30
31
32
33
39.9±19.0
88.0±25.0
0.28±0.06
0.69±0.21
O
O
S
N
O
161±32
34
a. Dose response curve (See Figure S1).
During our previous studies,[35] we checked the cytotoxicity of the DIM compounds to ensure the
reduction of PCSK9 was not due to cell death. No compounds showed any obvious cytotoxicity at
concentrations of 10
μM or lower. We also analyzed the cytotoxicity of all the improved analogues by
MTT assay at 1.0 M (See Figure S2). At this concentration, the maximal PCSK9 reduction could be
μ
achieved and most compounds did not show any obvious cytotoxicity. We can conclude that the
reduction of PCSK9 by these new analogues with much higher potency is not due to cytotoxicity.
The syntheses of compounds 2-34 are summarized in Schemes 1-4. Commercially available 7-
fluoroindole was acylated by Friedel–Crafts reaction using oxalyl chloride. After hydrolysis and Wolff-
Kishner-Huang reduction, the 7-fluoroindole-3-acetic acid product was obtained. Based on the same
12

procedure, 4, 5 and 6-F substituted isomers were prepared. The 2-fluorine of 2,4-difluoronitrobenzene
was substituted by a variety of primary amines through a SNAr reaction. After reduction, a series of
desired N-substituted o-phenylenediamines were prepared. According to a similar procedure, 5-fluoro
and 3-fluoro N1-methylbenzene-1,2-diamines were prepared from 2,4-difluoronitrobenzene and 2,6-
difluoronitrobenzene, respectively. The other two isomeric 4-fluoro and 6-fluoro N1-methylbenzene-1,2-
diamines were prepared from 2,4-difluoronitrobenzene and 2,6-difluoronitrobenzene, respectively,
through additional Boc protection, N-methylation and deprotection steps. The two intermediates, indole-
3-acetic acids and benzene-1,2-diamines, were then coupled by EDCl and HOBt, followed by
intramolecular condensation, to afford the methylene linked analogues.
The deuterated analogue 11 was easily prepared from compound 2 by H-D exchange in DMSO-d6/D2O.
The linker methylene group can also be oxidized to carbonyl by SeO2 and afforded the corresponding
carbonyl bridged analogues. The additional substituents on the nitrogen of the indole ring were
introduced by straightforward alkylation reactions.
Reagents and conditions: (a) oxalyl chloride (1.2 equiv), NaHCO3, Et2O; (b) hydrazine hydrate (5.0
equiv), NaOH, ethylene glycol, 150 °C (c) DIPEA (2.0 equiv), DMSO, 0 °C to rt, 15hrs, >95%; (d)
Pd/C, H2 (1.0 atm), MeOH, rt, 80%; (e) EDCl, HOBt, DMF, rt; (f) Toluene/AcOH (2/1), Microwave,
150 °C, d, e total yield 70%.
Scheme 1. Synthesis of Compounds 2-10
13

Reagents and conditions: (a) Microwave, DMSO-d6/D2O, 160 °C, yield > 95%; (b) SeO2, 1,4-dioxane,
100 °C, yield 71% (c) SOCl2, DMF (cat.), Et2O, rt; (d) SnCl4 (1.3 equiv), DCM/MeNO2, rt; d, e total
yield 31%.
Scheme 2. Synthesis of Compounds 11, 12 and 13
Reagents and conditions: (a) MeNH2, DIPEA, DMSO, 0 °C to rt, yield 95%; (b) Fe, NH4Cl, EtOH/H2O,
80 °C, yield 80%; (c) NH3•H2O, DIPEA, DMSO, 0 °C to rt, yield 95%; (d) 1) DMAP (10 mol %),
Boc2O (2.2 equiv), DMF, rt, 2) TFA 3.2% in DCM, yield 95%; (e) Pd/C 10%, H2 (1.0 atm), MeOH, rt,
yield 80% (f) MeI (1.0 equiv), Na2CO3, DMF, rt, yield 40%; (g) MeB(OH)2, Cu(OAc)2, Pyridine, 1,4-
14

dioxanes, 100 °C, yield 60%; (h) TFA 50 vol. % in DCM, rt, yield 90% (i) HOBt, EDCl, DMF, rt; (j)
AcOH/Toluene, Microwave, 150 °C; (k) SeO2 (2.0 equiv), 1,4-dioxanes, 100 °C; total yield of (i), (j)
and (k) 50%.
Scheme 3. Synthesis of Compounds 14-26 and 29
Reagents and conditions: (a) MeI (3.0 equiv) or n-BuI (3.0 equiv), Cs2CO3 (1.5 equiv), Acetonitrile, rt,
yield >95%; (b) K2CO3 (4.0 equiv), Acetone, 70 °C, yield 85%; (c) Morpholine, TBAI (0.2 equiv), 100
°C, yield 80%; (d) NaH (1.2 equiv), ClSO2CH2CH2Cl (1.5 equiv), DMF, 0 °C to rt, yield 70%; (e)
Morpholine (10.0 equiv) or Piperazine (10.0 equiv), TBAI (0.2 equiv), DMSO, 110 °C, yield 70%.
Scheme 4. Synthesis of Compounds 27, 28 and 30-34
Conclusions
In summary, based on our previously developed PCSK9 modulator DFDIM 1 (EC50 = 200 nM), a series
of analogues were designed and synthesized to further improve the potency and metabolic stability. Our
initial SAR studies for the substituents in DFBIIM 2 indicated that the addition of a methyl substituent
to the nitrogen of the benzimidazole ring in 8 greatly improved its potency. After investigating the liver
microsomal stability for compounds 1 and 2 in different species, we hypothesized that the methylene
bridge between the two heterocycles in DFDIM or DFBIIM may be a metabolically labile site. In order
to validate this hypothesis, deuterated compound 11 was prepared. The improved metabolic stability of
11 over 2 supports the above hypothesis. We then prepared carbonyl linked bisindole DFDIK 12 and
benzoimidazole-indole DFBIIK-1 13. We found that these two compounds had not only much better
metabolic stability, but also nearly 100-fold higher potency. These data prompted us to initiate the
15

preliminary PK study in mice to establish a correlation between in vitro metabolic stability and in vivo
PK performance of this class of PCSK9 modulators. During subsequent optimizations, we introduced
the methyl substituent to the benzimidazole nitrogen of DFBIIK-1 13 and prepared DFBIIK-2 21, which
yielded an additional 17-fold increase of potency. To systematically investigate the effect of the two
fluorine substituents on the ketone linked benzoimidazole-indoles, an ‘F-scan’ was also conducted on
both heterocycles. In addition, we also investigated the effect of different substituents on each of the two
nitrogen atoms in the benzoimidazole and indole heterocycles, respectively. We found that the nitrogen
of indole can tolerate various substituents, which may be useful for future target identification and
mechanism of action studies, which will be reported in due course.
Experimental Sections
General Information in Synthetic Chemistry
All reactions were conducted under a positive pressure of dry argon in a glassware that had been oven-
dried prior to use. Anhydrous solutions of reaction mixtures were transferred via an oven-dried syringe
or cannula. All solvents were dried prior to use unless noted otherwise. Thin-layer chromatography
(TLC) was performed using precoated silica gel plates. Flash column chromatography was performed
with the silica gel. 1H and 13C nuclear magnetic resonance (NMR) spectra were recorded on Bruker 400,
500, 600 MHz and Varian 500 MHz spectrometers. 1H NMR spectra were reported in parts per million
(ppm) referenced to 7.26 ppm of CDCl3 or referenced to the center line of a septet at 2.50 ppm of
DMSO-d6. Signal splitting patterns were described as singlet (s), doublet (d), triplet (t), quartet (q),
quintet (quint), or multiplet (m), with coupling constants (J) in hertz. High-resolution mass spectra
(HRMS) were performed on an electron spray injection (ESI) TOF mass spectrometer. The liquid
chromatography–mass spectrometry (LC–MS) analysis of final products was processed on an Agilent
1290 Infinity II LC system using a Poroshell 120 EC-C18 column (5 cm × 2.1 mm, 1.9
μm) for
chromatographic separation. Agilent 6120 Quadrupole LC/MS with multimode electrospray ionization
plus atmospheric pressure chemical ionization was used for detection. The mobile phases were 5.0%
methanol and 0.1% formic acid in purified water (A) and 0.1% formic acid in methanol (B). The
gradient was held at 5% (0–0.2 min), increased to 100% at 2.5 min, then held at isocratic 100% B for 0.4
min, and then immediately stepped back down to 5% for 0.1 min re-equilibration. The flow rate was set
at 0.8 mL/min. The column temperature was set at 40 °C. The purities of all of the final compounds
16

were determined to be over 95% by LC–MS. See the Supporting Information for 1H and 13C NMR
spectra and LC–MS purity analysis of all compounds.
General procedure for the preparation of compounds 2-10.
In a 250 mL flask with a magnetic stirring bar, 7-fluoroindole (2.0 g, 14.8 mmol) was dissolved in 50
mL ethyl ether and cooled by ice-water bath. Oxalyl chloride (2N DCM solution) (8.9 mL, 17.8 mmol)
was added by dropping funnel. Half of an hour later, the empty dropping funnel and ice-water bath were
removed, and the reaction mixture was stirred at room temperature for another 5 h. Then the reaction
mixture was quenched by NaHCO3 saturated solution and stirred overnight. After carefully adjusting to
pH 1.0 by 1N HCl, the organic phase was separated, and the aqueous phase was extracted by 100 mL
ethyl acetate. The organic phase was combined and dried by Na2SO4, then the solvent was removed by
rotavapor to yield 3.06 g of solid, which was directly used for the next step reaction.
In a 250 mL flask with a magnetic stirring bar, the 3.06 g of solids, hydrazine hydrate (3.7 g, 74.0 mmol)
and sodium hydroxide (6.5 g, 162.8 mmol) were suspended in 35 mL ethylene glycol. The reaction
mixture was heated to 60 ºC for half an hour, then heated to 150 ºC for another 4 h. After cooling to
room temperature, the reaction mixture was adjusted to pH 1.0 by 1N HCl, extracted by 100 mL ethyl
acetate twice. The organic phase was combined and dried by Na2SO4, the solvent was then removed by
rotavapor, the residue was purified by silica gel column chromatography (eluted with methanol in DCM
from 0.5% to 5%) and 2.0 g of 7-fluoroindole-3-acetic acid was prepared (two steps, total yield of 70%).
In a 150 mL flask with a magnetic stirring bar, 2,4-difluoronitrobenzene (1.0 g, 6.29 mmol) and DIPEA
(1.6 g, 12.58 mmol) were suspended in 10 mL DMSO and cooled in ice-water bath, then benzyl amine
(0.67 g, 6.29 mmol) was added to the reaction mixture. After stirring overnight, the reaction mixture was
quenched by 100 mL NaHCO3 saturated solution. The yellow precipitate was collected by filtration and
washed by 50 mL water three times and 50 mL hexane twice. After drying by high vacuum, 1.5 g of
yellow solid was generated.
In a 100 mL flask with a magnetic stirring bar, N-benzyl-5-fluoro-2-nitroaniline (0.5 g, 2.03 mmol) and
Pd/C (100 mg) were suspended in 15 mL methanol. After sealing by rubber stopper, a 2.0 L hydrogen
balloon was connected with the reaction flask by a syringe needle. The reaction mixture was then stirred
overnight at room temperature. After removing the Pd/C by filtration, the solvent was removed by
rotavapor. The residue was then purified by silica gel column chromatography to afford 352 mg of N1-
benzyl-5-fluorobenzene-1,2-diamine (80% yield).
17

In a 10 mL flask with a magnetic stirring bar, a mixture of N1-benzyl-5-fluorobenzene-1,2-diamine (118
mg, 0.54 mmol), 7-fluoroindole-3-acetic acid (100 mg, 0.52 mmol), HOBt hydrate (79 mg, 0.52 mmol),
EDCl (149 mg, 0.78 mmol) and DMF (2.5 mL) were stirred at room temperature for 5.0 h. The reaction
mixture was then poured to 100 mL ethyl acetate and washed by 100 mL NaHCO3 saturated aqueous
solution three times. The organic phase was dried by Na2SO4, the solvent was then removed by
rotavapor. The residue was suspended in toluene /AcOH (4 mL/2 mL) and heated at 150 ºC by
microwave for 1.0 h. After cooling to room temperature, the reaction mixture was then poured to 100
mL ethyl acetate, washed by 100 mL NaHCO3 saturated aqueous solution three times. The organic
phase was dried by Na2SO4, the solvent was then removed by rotavapor. The residue was then purified
by silica gel column chromatography to yield 148 mg of compound 3 (76% yield).
6-fluoro-2-((7-fluoro-1H-indol-3-yl)methyl)-1H-benzo[d]imidazole(
d) 8.72 (s, 1H), 7.35 (dd, J = 8.8, 4.6 Hz, 1H), 7.21 – 7.16 (m, 1H), 7.15 – 7.09 (m, 2H), 6.97 – 6.87
2
) 1H NMR (400 MHz, Chloroform-
δ
(m, 3H), 4.34 (s, 2H). HRMS (ESI) calcd for C16H11F2N3 (M + Na)+ 306.0813, found 306.0816. LC-MS
Purity: 98.4 %, Rt = 2.06 min.
1
1-benzyl-6-fluoro-2-((7-fluoro-1H-indol-3-yl)methyl)-1H-benzo[d]imidazole (
Chloroform-d)
3H), 7.03 – 6.94 (m, 3H), 6.91 – 6.83 (m, 4H), 5.20 (s, 2H), 4.35 (d, J = 1.1 Hz, 2H). C NMR (101
MHz, Acetone-d6) 160.0 (d, J = 236.4 Hz), 155.5 (d, J = 3.2 Hz), 150.5 (d, J = 242.5 Hz), 140.3,
3
) H NMR (400 MHz,
δ
8.37 (s, 1H), 7.71 (ddd, J = 8.9, 4.8, 0.5 Hz, 1H), 7.30 – 7.27 (m, 1H), 7.25 – 7.19 (m,
13
δ
137.3, 137.0 (d, J = 13.4 Hz), 132.2 (d, J = 5.6 Hz), 129.4, 128.3, 127.2, 125.4, 125.2, 120.7 (d, J = 10.2
Hz), 112.0 (d, J = 6.2 Hz), 115.9 (d, J = 3.4 Hz), 111.7 (d, J = 2.4 Hz), 110.2 (d, J = 25.1 Hz), 107.0 (d,
J = 16.2 Hz), 97.6 (d, J = 27.9 Hz), 47.8, 25.5. HRMS (ESI) calcd for C23H17F2N3 (M + H)+ 374.1463,
found 374.1460. LC-MS Purity: 98.6 %, Rt = 2.74 min.
6-fluoro-2-((7-fluoro-1H-indol-3-yl)methyl)-1-phenethyl-1H-benzo[d]imidazole (
Chloroform-d) 8.36 (s, 1H), 7.68 (ddd, J = 8.7, 4.8, 0.5 Hz, 1H), 7.27 – 7.23 (m, 4H), 7.04 – 6.87 (m,
5H), 6.85 – 6.80 (m, 2H), 4.17 (t, J = 7.2 Hz, 2H), 3.99 (d, J = 1.1 Hz, 2H), 2.79 (t, J = 7.2 Hz, 2H). 13C
NMR (101 MHz, Chloroform-d) 159.5 (d, J = 239.1 Hz), 154.2 (d, J = 2.9 Hz), 149.6 (d, J = 244.5
4
) 1H NMR (400 MHz,
δ
δ
Hz), 138.8, 137.5, 135.2 (d, J = 13.0 Hz), 130.6 (d, J = 5.5 Hz), 128.9, 128.7, 127.1, 124.7 (d, J = 13.5
Hz), 123.4, 120.1 (d, J = 10.1 Hz), 120.0 (d, J = 6.1 Hz), 114.5 (d, J = 3.4 Hz), 111.3 (d, J = 2.3 Hz),
110.2 (d, J = 24.9 Hz), 107.1 (d, J = 16.0 Hz), 96.2 (d, J = 27.5 Hz), 45.9, 35.5, 24.5. HRMS (ESI) calcd
for C24H19F2N3 (M + H)+: 388.1620, found 388.1618. LC-MS Purity: 96.3 %, Rt = 2.73 min.
18

1
6-fluoro-2-((7-fluoro-1H-indol-3-yl)methyl)-1-(pyridin-4-ylmethyl)-1H-benzo[d]imidazole (
(400 MHz, Methanol-d4) 8.18 (s, 1H), 8.07 – 8.03 (m, 2H), 7.72 – 7.67 (m, 1H), 7.20 – 7.17 (m, 1H),
7.10 – 7.00 (m, 3H), 6.89 – 6.82 (m, 1H), 6.78 – 6.72 (m, 1H), 6.54 – 6.50 (m, 2H), 5.48 (s, 2H), 4.47 (s,
2H). 13C NMR (101 MHz, Methanol-d4)
161.3 (d, J = 239.2 Hz), 156.7, 151.0 (d, J = 243.6 Hz),
5) H NMR
δ
δ
149.5, 147.3, 139.1, 137.2 (d, J = 13.2 Hz), 132.2 (d, J = 5.8 Hz), 126.1, 125.9, 122.2, 120.6 (d, J = 10.2
Hz), 120.5 (d, J = 6.2 Hz), 115.4 (d, J = 3.5 Hz), 111.9 (d, J = 25.5 Hz), 110.6 (d, J = 2.4 Hz), 107.3 (d,
J = 16.4 Hz), 97.8 (d, J = 28.2 Hz), 47.2, 25.8. HRMS (ESI) calcd for C22H16F2N4 (M + H)+: 375.1416,
found 375.1417. LC-MS Purity: 96.2 %, Rt = 1.95 min.
1
6-fluoro-2-((7-fluoro-1H-indol-3-yl)methyl)-1-(pyridin-3-ylmethyl)-1H-benzo[d]imidazole (
(400 MHz, Methanol-d4)
8.0 Hz, 1H), 7.10 – 7.02 (m, 3H), 6.99 – 6.93 (m, 1H), 6.89 – 6.82 (m, 2H), 6.75 (ddd, J = 11.4, 7.8 Hz,
1H), 5.47 (s, 2H), 4.48 (s, 2H). 13C NMR (101 MHz, Methanol-d4)
161.2 (d, J = 239.2 Hz), 156.5 (d, J
6
) H NMR
δ
8.23 – 8.18 (m, 1H), 7.94 – 7.90 (m, 1H), 7.70 – 7.65 (m, 1H), 7.19 (d, J =
δ
= 2.9 Hz), 151.0 (d, J = 243.5 Hz), 148.7, 147.7, 139.1 (d, J = 1.1 Hz), 137.0 (d, J = 13.2 Hz), 135.3,
133.3, 132.1 (d, J = 5.7 Hz), 126.0 (d, J = 13.7 Hz), 125.7, 124.5, 120.6 (d, J = 10.3 Hz), 120.5 (d, J =
6.4 Hz), 115.3 (d, J = 3.4 Hz), 111.8 (d, J = 25.3 Hz), 110.6 (d, J = 2.4 Hz), 107.3 (d, J = 16.3 Hz), 97.9
(d, J = 28.1 Hz), 45.9, 25.7. HRMS (ESI) calcd for C22H16F2N4 (M + H)+: 375.1416, found 375.1411.
LC-MS Purity: 99.2 %, Rt = 2.27 min.
1
2-(6-fluoro-2-((7-fluoro-1H-indol-3-yl)methyl)-1H-benzo[d]imidazol-1-yl)ethan-1-ol (
7
) H NMR (400
MHz, Methanol-d4)
δ
7.57 (dd, J = 8.8, 4.8 Hz, 1H), 7.30 – 7.22 (m, 2H), 7.17 – 7.13 (m, 1H), 7.04 –
6.97 (m, 1H), 6.93 – 6.87 (m, 1H), 6.85 – 6.78 (m, 1H), 4.50 (d, J = 1.0 Hz, 2H), 4.24 (t, J = 5.4 Hz,
2H), 3.67 (t, J = 5.4 Hz, 2H). HRMS (ESI) calcd for C18H15F2N3O (M + H)+: 328.1256, found 328.1265.
LC-MS Purity: 96.9 %, Rt = 1.96 min.
1
6-fluoro-2-((7-fluoro-1H-indol-3-yl)methyl)-1-methyl-1H-benzo[d]imidazole (
8
) H NMR (400 MHz,
Chloroform-d)
δ
8.41 (s, 1H), 7.67 (ddd, J = 8.7, 4.9, 0.5 Hz, 1H), 7.40 – 7.34 (m, 1H), 7.04 – 6.96 (m,
3H), 6.95 (ddd, J = 8.7, 2.5, 0.5 Hz, 1H), 6.93 – 6.87 (m, 1H), 4.40 (d, J = 1.1 Hz, 2H), 3.59 (s, 3H).
HRMS (ESI) calcd for C17H13F2N3 (M + H)+: 298.1150, found 298.1152. LC-MS Purity: 96.8 %, Rt =
2.02 min.
1
6-fluoro-2-((7-fluoro-1H-indol-3-yl)methyl)-1-isobutyl-1H-benzo[d]imidazole (
9
) H NMR (400 MHz,
Chloroform-d)
δ
8.50 (s, 1H), 7.66 (ddd, J = 8.2, 4.9, 1.0 Hz, 1H), 7.36 – 7.30 (m, 1H), 7.03 – 6.94 (m,
4H), 6.92 – 6.86 (m, 1H), 4.39 (d, J = 1.1 Hz, 2H), 3.79 (d, J = 7.7 Hz, 2H), 2.16 (m, 1H), 0.89 (d, J =
19

6.7 Hz, 6H). HRMS (ESI) calcd for C20H19F2N3 (M + H)+: 340.1620, found 340.1636. LC-MS Purity:
99.2 %, Rt = 2.47 min.
6-fluoro-2-((7-fluoro-1H-indol-3-yl)methyl)-1-neopentyl-1H-benzo[d]imidazole (10
MHz, DMSO-d6) 11.43 (s, 1H), 7.52 (dd, J = 8.7, 5.0 Hz, 1H), 7.44 (dd, J = 9.8, 2.5 Hz, 1H), 7.36 (s,
)
1H NMR (400
δ
1H), 7.31 – 7.25 (m, 1H), 7.00 – 6.93 (m, 1H), 6.92 – 6.84 (m, 2H), 4.36 (s, 2H), 4.05 (s, 2H), 1.00 (s,
9H). HRMS (ESI) calcd for C21H21F2N3 (M + H)+: 354.1776, found 354.1784. LC-MS Purity: 98.2 %,
Rt = 2.53 min.
6-fluoro-2-((7-fluoro-1H-indol-3-yl)methyl-d2)-1H-benzo[d]imidazole (11). Compound 2 (145 mg, 0.51
mmol) was suspended in DMSO-d6/D2O (2.5 mL/7.5 mL). The reaction mixture was then heated at 160
ºC by microwave for 5.0 h. After cooling to room temperature, the reaction mixture was then poured to
50 mL of ethyl acetate, washed by 50 mL of water three times. The organic phase was dried by Na2SO4,
the solvent was then removed by rotavapor. The residue was then purified by silica gel flash column
chromatography to yield 138 mg of compound 11 (95% yield).
6-fluoro-2-((7-fluoro-1H-indol-3-yl)methyl-d2)-1H-benzo[d]imidazole (11
)
1H NMR (400 MHz,
DMSO-d6) 12.22 – 12.16 (m, 1H), 11.47 (s, 1H), 7.51 (m, 0.5 H), 7.39 – 7.28 (m, 3H), 7.16 (m, 0.5
δ
H), 6.99 – 6.87 (m, 3H). HRMS (ESI) calcd for C16H9D2F2N3 (M + H)+: 286.1119, found 286.1123. LC-
MS Purity: 98.8 %, Rt = 2.36 min.
(5-fluoro-1H-indol-2-yl)(7-fluoro-1H-indol-3-yl)methanone (12). In a 250 mL flask with a magnetic
stirring bar, 5-fluoroindole-2-carboxylic acid (1.0 g, 5.58 mmol), SOCl2 (2.0 mL) and DMF (30 µL)
were dissolved in 30 mL of ethyl ether. The reaction mixture was stirred at room temperature for 4.0 h.
After removing ethyl ether and SOCl2 by rotavapor, the residue and 7-fluoroindole (905 mg, 6.7 mmol)
were dissolved in 20 mL of DCM and 15 mL of nitromethane. To the reaction mixture, was added SnCl4
(1N in DCM) (7.3 mL, 7.3 mmol) by syringe. The mixture was stirred at room temperature for overnight.
After quenching by 150 mL of ice-water, the mixture was extracted by 150 mL of ethyl acetate. The
organic phase was dried by Na2SO4, the solvent was removed by rotavapor. The residue was purified by
silica gel column chromatography to yield 510 mg of compound 12 with a yield of 31%.
1
(5-fluoro-1H-indol-2-yl)(7-fluoro-1H-indol-3-yl)methanone (12) H NMR (400 MHz, DMSO-d6)
δ
12.69 (s, 1H), 11.91 (s, 1H), 8.47 (s, 1H), 8.10 (d, J = 8.0 Hz, 1H), 7.52 – 7.48 (m, 1H), 7.47 – 7.43 (m,
1H), 7.38 (dd, J = 2.3, 0.9 Hz, 1H), 7.26 – 7.19 (m, 1H), 7.17 – 7.09 (m, 2H). 13C NMR (101 MHz,
DMSO-d6)
δ
180.2 (d, J = 1.2 Hz), 157.2 (d, J = 233.2 Hz), 149.2 (d, J = 244.8 Hz), 137.5, 134.5, 134.0,
20

130.1 (d, J = 4.9 Hz), 127.3 (d, J = 10.6 Hz), 124.4 (d, J = 13.4 Hz), 122.4 (d, J = 6.1 Hz), 117.6 (d, J =
3.5 Hz), 115.8 (d, J = 1.8 Hz), 113.8 (d, J = 9.6 Hz), 113.4 (d, J = 26.7 Hz), 108.1 (d, J = 15.6 Hz),
107.6 (d, J = 5.4 Hz), 106.3 (d, J = 22.8 Hz). HRMS (ESI) calcd for C17H10F2N2O (M + Na)+: 319.0653,
found: 319.0666. LC-MS Purity: 98.2 %, Rt = 2.73 min.
General procedure for SeO2 oxidation:
In a 100 mL flask with a magnetic stirring bar, compound 2 (1.2 g, 4.2 mmol) and SeO2 (932 mg, 8.4
mmol) were suspended in 20 mL of 1,4-dioxane. The reaction mixture was then heated at 100 ºC by oil
bath for 4.0 h. After cooling to room temperature, the reaction mixture was then poured to 200 mL of
ethyl acetate, washed by 100 mL of NaHCO3 saturated aqueous solution three times. The organic phase
was dried by Na2SO4 and the solvent was removed by rotavapor. The residue was then purified by silica
gel column chromatography to yield 890 mg of compound 13 (71% yield).
(6-fluoro-1H-benzo[d]imidazol-2-yl)(7-fluoro-1H-indol-3-yl)methanone (13
DMSO-d6) 13.47 (s, 0.5H), 13.44 (s, 0.5H), 12.85 (s, 1H), 9.34 (d, J = 5.2 Hz, 1H), 8.20 (d, J = 7.9
Hz, 1H), 7.91 (dd, J = 8.9, 4.9 Hz, 0.5H), 7.69 (dd, J = 9.7, 2.5 Hz, 0.5H), 7.60 (dd, J = 8.9, 4.9 Hz,
0.5H), 7.38 – 7.11 (m, 3.5H). 13C NMR (101 MHz, DMSO-d6)
177.2, 161.3, 160.1, 158.9, 157.8,
)
1H NMR (400 MHz,
δ
δ
150.7, 150.5, 150.2, 148.1, 143.4, 143.3, 139.8, 138.7, 138.6, 134.3, 134.2, 130.8, 130.1, 130.1, 124.2,
124.1, 123.2, 123.1, 122.3, 122.2, 117.7, 117.7, 114.5, 114.4, 113.7, 113.6, 113.5, 111.8, 111.6, 108.6,
108.4, 106.0, 105.7, 98.6, 98.3. HRMS (ESI) calcd for C16H9F2N3O (M + Na)+: 320.0606, found:
320.0620. LC-MS Purity: 99.8 %, Rt = 2.68 min.
1
(1H-benzo[d]imidazol-2-yl)(1H-indol-3-yl)methanone (14) H NMR (400 MHz, DMSO-d6)
δ 13.28 (s,
1H), 12.24 (s, 1H), 9.36 (d, J = 3.2 Hz, 1H), 8.42 – 8.37 (m, 1H), 7.86 (d, J = 8.3 Hz, 1H), 7.62 – 7.55
(m, 2H), 7.40 – 7.25 (m, 4H). HRMS (ESI) calcd for C16H11N3O (M + H)+: 262.0975, found 262.0978.
LC-MS Purity: 99.2 %, Rt = 2.46 min.
1
(1H-benzo[d]imidazol-2-yl)(7-fluoro-1H-indol-3-yl)methanone (15) H NMR (400 MHz, DMSO-d6)
δ
13.34 (s, 1H), 12.83 (s, 1H), 9.39 (d, J = 3.2 Hz, 1H), 8.22 (d, J = 7.5 Hz, 1H), 7.88 (d, J = 8.1 Hz, 1H),
7.61 (d, J = 8.0 Hz, 1H), 7.43 – 7.23 (m, 3H), 7.20 – 7.09 (m, 1H). HRMS (ESI) calcd for C16H10FN3O
(M + H)+: 280.0881, found 280.0893. LC-MS Purity: 99.6 %, Rt = 2.59 min.
1
(6-fluoro-1H-benzo[d]imidazol-2-yl)(1H-indol-3-yl)methanone (16) H NMR (400 MHz, DMSO-d6)
δ
13.39 (s, 1H), 12.26 (s, 1H), 9.32 (s, 1H), 8.44 – 8.35 (m, 1H), 7.88 (s, 0.5H), 7.75 – 7.51 (m, 2H), 7.37
21

– 7.17 (m, 3.5H). HRMS (ESI) calcd for C16H10FN3O (M + H)+: 280.0881, found 280.0886. LC-MS
Purity: 99.6 %, Rt = 2.53 min.
Procedure for the preparation of 4-fluoro-N1-methylbenzene-1,2-diamine and 6-fluoro-N1-
methylbenzene-1,2-diamine
tert-butyl (3-fluoro-2-nitrophenyl)carbamate
In a 100 mL flask with a magnetic stirring bar, 3-fluoro-2-nitroaniline (1.0 g, 6.41 mmol), Boc2O (3.1 g,
14.1 mmol) and DMAP (200 mg) were dissolved in 20 mL of DMF. The reaction mixture was stirred
overnight at room temperature and then diluted by 100 mL of ethyl acetate. The ethyl acetate solution
was washed by 1N HCl (50 mL) three times, dried by Na2SO4, and the solvent was removed by
rotavapor. The residue was dissolved in 16 mL of DCM with 3.0 vol % TFA (0.5 mL, 9.6 mmol) (1.5
equiv). After stirring at room temperature for 2.0 h, the reaction mixture was diluted by 100 mL of ethyl
acetate, washed by 50 mL of NaHCO3 saturated aqueous solution three times. The organic phase was
dried by Na2SO4, the solvent was removed by rotavapor. The residue was purified by silica gel column
chromatography to yield 1.58 g of tert-butyl (3-fluoro-2-nitrophenyl)carbamate (96 % yield).
The tert-butyl (3-fluoro-2-(methylamino)phenyl)carbamate was obtain by monomethylating of tert-butyl
(2-amino-3-fluorophenyl)carbamate following procedures in literature.[66]
tert-butyl (5-fluoro-2-(methylamino)phenyl)carbamate
In a 100 mL flask with a magnetic stirring bar, tert-butyl (2-amino-5-fluorophenyl)carbamate (0.6 g,
2.65 mmol), CH3I (376 mg, 2.65 mmol) and Na2CO3 (310 mg, 2.92 mmol) were suspended in 10 mL of
DMF. The reaction mixture was stirred overnight at room temperature and then diluted by 100 mL of
ethyl acetate. The ethyl acetate solution was washed by water (100 mL) three times, dried by Na2SO4,
the solvent was removed by rotavapor. The residue was purified by silica gel column chromatography to
yield 255 mg of tert-butyl (5-fluoro-2-(methylamino)phenyl)carbamate with a yield of 40 %.
(4-fluoro-1H-benzo[d]imidazol-2-yl)(7-fluoro-1H-indol-3-yl)methanone (17
)
1H NMR (400 MHz,
DMSO-d6) 13.56 (s, 1H), 12.98 (s, 1H), 9.33 (s, 1H), 8.20 (d, J = 7.9 Hz, 1H), 7.61 – 7.40 (m, 1H),
δ
7.32 (m, 2H), 7.15 (m, 2H). HRMS (ESI) calcd for C16H9F2N3O (M + H)+: 298.0786, found 298.0794.
LC-MS Purity: 99.8 %, Rt = 2.71 min.
22

(7-fluoro-1H-indol-3-yl)(1-methyl-1H-benzo[d]imidazol-2-yl)methanone (18
DMSO-d6) 12.78 (s, 1H), 8.97 (s, 1H), 8.20 (d, J = 7.9 Hz, 1H), 7.88 (d, J = 8.1 Hz, 1H), 7.75 (d, J =
)
1H NMR (500 MHz,
δ
8.1 Hz, 1H), 7.45 (ddd, J = 8.1, 7.0, 1.1 Hz, 1H), 7.37 (ddd, J = 8.1, 7.1, 1.1 Hz, 1H), 7.30 – 7.24 (m,
1H), 7.18 – 7.12 (m, 1H), 4.14 (s, 3H). HRMS (ESI) calcd for C17H12FN3O (M + H)+: 294.1037, found
294.1048. LC-MS Purity: 99.3 %, Rt = 2.68 min.
1
(4-fluoro-1-methyl-1H-benzo[d]imidazol-2-yl)(7-fluoro-1H-indol-3-yl)methanone (19) H NMR (400
MHz, DMSO-d6)
δ 12.82 (s, 1H), 8.92 (s, 1H), 8.18 (d, J = 7.9 Hz, 1H), 7.59 (d, J = 8.3 Hz, 1H), 7.48 –
7.39 (m, 1H), 7.31 – 7.24 (m, 1H), 7.22 – 7.09 (m, 2H), 4.14 (s, 3H). HRMS (ESI) calcd for
C17H11F2N3O (M + H)+: 312.0943, found 312.0956. LC-MS Purity: 99.6 %, Rt = 2.80 min.
1
(5-fluoro-1-methyl-1H-benzo[d]imidazol-2-yl)(7-fluoro-1H-indol-3-yl)methanone (20) H NMR (400
MHz, DMSO-d6)
δ 12.80 (s, 1H), 8.94 (d, J = 2.2 Hz, 1H), 8.18 (dd, J = 7.9, 2.2 Hz, 1H), 7.81 – 7.72
(m, 1H), 7.72 – 7.63 (m, 1H), 7.37 – 7.21 (m, 2H), 7.14 (ddd, J = 10.6, 7.9, 2.2 Hz, 1H), 4.13 (s, 3H).
HRMS (ESI) calcd for C17H11F2N3O (M + H)+: 312.0943, found 312.0953. LC-MS Purity: 99.8 %, Rt =
2.79 min.
1
(6-fluoro-1-methyl-1H-benzo[d]imidazol-2-yl)(7-fluoro-1H-indol-3-yl)methanone (21) H NMR (400
MHz, DMSO-d6)
δ
12.77 (s, 1H), 8.93 (s, 1H), 8.17 (d, J = 7.9 Hz, 1H), 7.88 (dd, J = 8.9, 5.0 Hz, 1H),
13
7.61 (dd, J = 9.4, 2.5 Hz, 1H), 7.28 – 7.17 (m, 2H), 7.16 – 7.08 (m, 1H), 4.09 (s, 3H). C NMR (101
MHz, DMSO-d6)
δ 179.1 (d, J = 1.5 Hz), 160.1 (d, J = 239.8 Hz), 149.2 (d, J = 244.9 Hz), 148.4 (d, J =
3.1 Hz), 139.0, 137.7, 136.7 (d, J = 14.0 Hz), 130.0 (d, J = 4.7 Hz), 124.1 (d, J = 13.3 Hz), 123.1 (d, J =
6.1 Hz), 122.2 (d, J = 10.4 Hz), 117.6 (d, J = 3.4 Hz), 115.5 (d, J = 2.0 Hz), 111.9 (d, J = 25.7 Hz),
108.4 (d, J = 15.7 Hz), 97.6 (d, J = 27.9 Hz), 32.4. HRMS (ESI) calcd for C17H11F2N3O (M + Na)+:
334.0762, found: 334.0765. LC-MS Purity: 96.3 %, Rt = 2.78 min.
1
(7-fluoro-1-methyl-1H-benzo[d]imidazol-2-yl)(7-fluoro-1H-indol-3-yl)methanone (22) H NMR (400
MHz, DMSO-d6)
δ 12.82 (s, 1H), 8.84 (d, J = 2.5 Hz, 1H), 8.17 (dd, J = 8.0, 2.4 Hz, 1H), 7.69 (dd, J =
8.1, 2.4 Hz, 1H), 7.34 – 7.19 (m, 3H), 7.19 – 7.08 (m, 1H), 4.25 (s, 3H). HRMS (ESI) calcd for
C17H11F2N3O (M + H)+: 312.0943, found 312.0956. LC-MS Purity: 99.7 %, Rt = 2.99 min.
(6-fluoro-1-methyl-1H-benzo[d]imidazol-2-yl)(1H-indol-3-yl)methanone (23
DMSO-d6) 12.21 (s, 1H), 8.88 (s, 1H), 8.41 – 8.32 (m, 1H), 7.88 (dd, J = 8.8, 4.9 Hz, 1H), 7.66 (dd, J
)
1H NMR (500 MHz,
δ
= 9.4, 2.5 Hz, 1H), 7.61 – 7.54 (m, 1H), 7.32 – 7.26 (m, 2H), 7.22 (ddd, J = 9.9, 8.9, 2.5 Hz, 1H), 4.10
23

(s, 3H). HRMS (ESI) calcd for C17H12FN3O (M + H)+: 294.1037, found 294.1051. LC-MS Purity: 99.2
%, Rt = 2.71 min.
1
(6-fluoro-1-methyl-1H-benzo[d]imidazol-2-yl)(4-fluoro-1H-indol-3-yl)methanone (24) H NMR (500
MHz, DMSO-d6)
δ 12.42 (s, 1H), 8.67 (s, 1H), 7.85 (dd, J = 8.9, 5.0 Hz, 1H), 7.66 (dd, J = 9.4, 2.5 Hz,
1H), 7.39 (d, J = 8.1 Hz, 1H), 7.30 – 7.25 (m, 1H), 7.21 (ddd, J = 9.8, 8.8, 2.5 Hz, 1H), 7.00 (dd, J =
11.1, 7.8 Hz, 1H), 4.05 (s, 3H). HRMS (ESI) calcd for C17H11F2N3O (M + H)+: 312.0943, found
312.0954. LC-MS Purity: 99.2 %, Rt = 2.59 min.
1
(6-fluoro-1-methyl-1H-benzo[d]imidazol-2-yl)(5-fluoro-1H-indol-3-yl)methanone (25) H NMR (500
MHz, DMSO-d6)
δ 12.31 (s, 1H), 8.96 (s, 1H), 8.04 (dd, J = 9.9, 2.6 Hz, 1H), 7.88 (dd, J = 8.9, 5.0 Hz,
1H), 7.67 (dd, J = 9.4, 2.5 Hz, 1H), 7.59 (dd, J = 8.8, 4.6 Hz, 1H), 7.23 (ddd, J = 9.8, 8.9, 2.5 Hz, 1H),
7.19 – 7.10 (m, 1H), 4.11 (s, 3H). HRMS (ESI) calcd for C17H11F2N3O (M + H)+: 312.0943, found
312.0956. LC-MS Purity: 95.8 %, Rt = 2.92 min.
1
(6-fluoro-1-methyl-1H-benzo[d]imidazol-2-yl)(6-fluoro-1H-indol-3-yl)methanone (26) H NMR (500
MHz, DMSO-d6)
δ 12.23 (s, 1H), 8.90 (s, 1H), 8.34 (dd, J = 8.7, 5.6 Hz, 1H), 7.88 (dd, J = 8.9, 4.9 Hz,
1H), 7.66 (dd, J = 9.4, 2.5 Hz, 1H), 7.37 (dd, J = 9.6, 2.4 Hz, 1H), 7.22 (ddd, J = 9.8, 8.8, 2.5 Hz, 1H),
7.14 (ddd, J = 9.8, 8.7, 2.4 Hz, 1H), 4.10 (s, 3H). HRMS (ESI) calcd for C17H11F2N3O (M + H)+:
312.0943, found 312.0955. LC-MS Purity: 99.7 %, Rt = 2.87 min.
(6-fluoro-1-methyl-1H-benzo[d]imidazol-2-yl)(7-fluoro-1-methyl-1H-indol-3-yl)methanone
(27
)
1H
NMR (400 MHz, DMSO-d6) 8.87 (s, 1H), 8.20 (dd, J = 8.0, 0.9 Hz, 1H), 7.88 (dd, J = 8.9, 4.9 Hz,
δ
1H), 7.67 (dd, J = 9.4, 2.5 Hz, 1H), 7.31 – 7.20 (m, 2H), 7.15 (ddd, J = 12.6, 7.9, 0.9 Hz, 1H), 4.11 (d, J
= 2.3 Hz, 3H), 4.09 (s, 3H). HRMS (ESI) calcd for C18H13F2N3O (M + H)+: 326.1099, found 326.1113.
LC-MS Purity: 99.1 %, Rt = 3.03 min.
General procedure for the preparation of compounds 28, 30 and 31
In a 25 mL flask with a magnetic stirring bar, compound 21 (200 mg, 0.64 mmol), 1-bromo-2-
chloroethane (368 mg, 2.56 mmol) and K2CO3 (353 mg, 2.56 mmol) were suspended in 10 mL of
acetone. After sealing by rubber stopper, the flask was heated at 70 ºC in an oil bath for 3 h. Then the
reaction mixture was diluted by 100 mL of ethyl acetate. The ethyl acetate solution was washed by water
(50 mL) three times, dried by Na2SO4, the solvent was removed by rotavapor. The residue was purified
by silica gel column chromatography to yield 203 mg of white solid with a yield of 85 %.
24

In a 10 mL flask with a magnetic stirring bar, the intermediate prepared above (203 mg, 0.54 mmol),
TBAI (40 mg, 0.11 mmol) were dissolved in 3 mL of morpholine. After sealing by a rubber stopper, the
flask was heated at 100 ºC in an oil bath for 4 h, and then diluted by 50 mL of ethyl acetate. The ethyl
acetate solution was washed by water (50 mL) three times, dried by Na2SO4, the solvent was removed
by rotavapor. The residue was purified by silica gel column chromatography to yield compound 28 (183
mg, 80% yield).
(7-fluoro-1-(2-morpholinoethyl)-1H-indol-3-yl)(6-fluoro-1-methyl-1H-benzo[d]imidazol-2-
yl)methanone (28) 1H NMR (400 MHz, DMSO-d6)
δ 9.02 (s, 1H), 8.22 (d, J = 8.0 Hz, 1H), 7.87 – 7.80
(m, 1H), 7.68 (dd, J = 9.4, 2.5 Hz, 1H), 7.30 – 7.21 (m, 2H), 7.19 – 7.11 (m, 1H), 4.54 (t, J = 6.1 Hz,
2H), 4.10 (s, 3H), 3.59 (t, J = 4.4 Hz, 4H), 2.74 (t, J = 6.1 Hz, 2H), 2.47 (t, J = 4.4 Hz, 4H). 13C NMR
(101 MHz, DMSO-d6)
δ 178.8, 160.1 (d, J = 239.6 Hz), 149.4 (d, J = 244.3 Hz), 148.5 (d, J = 3.1 Hz),
143.5, 137.8, 136.8 (d, J = 13.9 Hz), 130.8 (d, J = 4.1 Hz), 123.7 (d, J = 9.3 Hz), 123.4 (d, J = 6.6 Hz),
122.0 (d, J = 10.5 Hz), 117.9 (d, J = 3.6 Hz), 114.2, 111.9 (d, J = 25.9 Hz), 109.3 (d, J = 17.9 Hz), 97.8
(d, J = 27.8 Hz), 66.2, 57.8, 53.2, 46.1 (d, J = 5.0 Hz), 32.4. HRMS (ESI) calcd for C23H22F2N4O2 (M +
H)+: 425.1784, found: 425.1799. LC-MS Purity: 99.2 %, Rt = 2.43 min.
1
(6-fluoro-1-(2-morpholinoethyl)-1H-benzo[d]imidazol-2-yl)(7-fluoro-1H-indol-3-yl)methanone (29) H
NMR (400 MHz, DMSO-d6) δ 12.82 (s, 1H), 8.82 (s, 1H), 8.15 (d, J = 7.9 Hz, 1H), 7.88 (dd, J = 8.9, 4.9
Hz, 1H), 7.68 (dd, J = 9.4, 2.5 Hz, 1H), 7.29 – 7.18 (m, 2H), 7.14 (dd, J = 11.3, 7.9 Hz, 1H), 4.74 (t, J =
5.9 Hz, 2H), 3.17 (t, J = 4.6 Hz, 4H), 2.58 (t, J = 5.9 Hz, 2H), 2.31 (t, J = 4.5 Hz, 4H). HRMS (ESI)
calcd for C22H20F2N4O2 (M + H)+: 411.1627, found 411.1627. LC-MS Purity: 99.3 %, Rt = 2.28 min.
(7-fluoro-1-(3-morpholinopropyl)-1H-indol-3-yl)(6-fluoro-1-methyl-1H-benzo[d]imidazol-2-
yl)methanone (30) 1H NMR (400 MHz, DMSO-d6)
δ 8.93 (s, 1H), 8.23 (dd, J = 7.9, 2.8 Hz, 1H), 7.88 –
7.78 (m, 1H), 7.65 (dd, J = 9.4, 2.6 Hz, 1H), 7.31 – 7.20 (m, 2H), 7.19 – 7.11 (m, 1H), 4.47 (t, J = 6.6
Hz, 2H), 4.08 (s, 3H), 3.53 (t, J = 4.4 Hz, 4H), 2.30 (t, J = 4.4 Hz, 4H), 2.22 (t, J = 6.3 Hz, 2H), 2.02 –
1.92 (m, 2H). HRMS (ESI) calcd for C24H24F2N4O2 (M + H)+: 439.1940, found 439.1942. LC-MS
Purity: 98.6 %, Rt = 2.35 min.
(7-fluoro-1-(4-morpholinobutyl)-1H-indol-3-yl)(6-fluoro-1-methyl-1H-benzo[d]imidazol-2-
yl)methanone (31) 1H NMR (400 MHz, DMSO-d6)
δ 8.92 (s, 1H), 8.22 (d, J = 8.0 Hz, 1H), 7.92 – 7.80
(m, 1H), 7.66 (dd, J = 9.4, 2.5 Hz, 1H), 7.32 – 7.20 (m, 2H), 7.16 (dd, J = 12.7, 8.0 Hz, 1H), 4.44 (t, J =
25

7.2 Hz, 2H), 4.09 (s, 3H), 3.49 (t, J = 4.5 Hz, 4H), 2.37 – 2.16 (m, 6H), 1.85 (p, J = 7.4 Hz, 2H), 1.44 (p,
J = 7.3 Hz, 2H). HRMS (ESI) calcd for C25H26F2N4O2 (M + H)+: 453.2097, found 453.2102. LC-MS
Purity: 99.6 %, Rt = 2.50 min.
1
(1-butyl-7-fluoro-1H-indol-3-yl)(6-fluoro-1-methyl-1H-benzo[d]imidazol-2-yl)methanone (32) H NMR
(500 MHz, Chloroform-d)
δ 8.83 (s, 1H), 8.34 (d, J = 8.0 Hz, 1H), 7.83 (dd, J = 9.5, 4.8 Hz, 1H), 7.28 –
7.20 (m, 1H), 7.16 – 7.07 (m, 2H), 7.00 (dd, J = 12.5, 7.9 Hz, 1H), 4.36 (t, J = 7.3 Hz, 2H), 4.14 (s, 3H),
1.91 (p, J = 7.3 Hz, 2H), 1.46 – 1.34 (m, 2H), 0.96 (t, J = 7.4 Hz, 3H).13C NMR (126 MHz, Chloroform-
d)
δ 179.3, 161.1 (d, J = 243.0 Hz), 145.0 (d, J = 245.3 Hz), 148.8 (d, J = 3.2 Hz), 141.6, 138.3, 137.0
(d, J = 13.3 Hz), 131.6 (d, J = 4.1 Hz), 124.4 (d, J = 9.7 Hz), 123.6 (d, J = 6.5 Hz), 122.6 (d, J = 10.3
Hz), 118.5 (d, J = 3.8 Hz), 115.4 (d, J = 1.4 Hz), 112.4 (d, J = 25.7 Hz), 109.6 (d, J = 18.0 Hz), 96.7 (d,
J = 27.4 Hz), 50.1 (d, J = 5.0 Hz), 33.3 (d, J = 2.2 Hz), 32.6, 20.0, 13.7. HRMS (ESI) calcd for
C21H19F2N3O (M + Na)+: 390.1388, found: 390.1404. LC-MS Purity: 99.5 %, Rt = 3.19 min.
General procedure for the preparation of compounds 33 and 34
In a 25 mL flask with a magnetic stirring bar, compound 21 (200 mg, 0.64 mmol) was suspended in 6
mL of DMF. After cooling to 0 ºC by an ice-water bath, NaH (65 wt %) (29 mg, 0.77 mmol) was added
to the reaction mixture. After stirring at 0 ºC for 1.0 h, ClSO2CH2CH2Cl (157 mg, 0.96 mmol) was
added. The reaction mixture was stirred overnight. The reaction mixture was diluted by 60 mL of ethyl
acetate solution, and then washed by water (50 mL) three times, dried by Na2SO4, the solvent was
removed by rotavapor. The residue was purified by silica gel column chromatography to yield 195 mg
of white solid with a yield of 70 %.
In a 10 mL flask with a magnetic stirring bar, the intermediate prepared above (150 mg, 0.34 mmol),
morpholine (296 mg, 3.4 mmol) and TBAI (26 mg, 0.07 mmol) were dissolved in 3 mL of DMSO and
stirred at 110 ºC for 6 h. After cooling to room temperature, it was diluted by 60 mL of ethyl acetate.
The ethyl acetate solution was washed by water (60 mL) three times, dried by Na2SO4. The solvent was
removed by rotavapor. The residue was purified by silica gel column chromatography to yield 116 mg
of compound 33 (70 % yield).
(7-fluoro-1-((2-morpholinoethyl)sulfonyl)-1H-indol-3-yl)(6-fluoro-1-methyl-1H-benzo[d]imidazol-2-
yl)methanone (33) 1H NMR (500 MHz, Chloroform-d)
δ 8.94 (s, 1H), 8.35 (d, J = 8.0 Hz, 1H), 7.85 (dd,
J = 9.5, 4.8 Hz, 1H), 7.31 – 7.26 (m, 1H), 7.16 – 7.11 (m, 2H), 7.06 (dd, J = 12.7, 8.0 Hz, 1H), 4.83 (t, J
26

= 7.3 Hz, 2H), 4.16 (s, 3H), 3.68 – 3.63 (m, 4H), 3.52 (t, J = 7.3 Hz, 2H), 3.26 – 3.21 (m, 4H). HRMS
(ESI) calcd for C23H22F2N4O4S (M + H)+: 489.1403, found: 489.1417. LC-MS Purity: 95.9 %, Rt = 2.84
min.
(7-fluoro-1-((2-(piperazin-1-yl)ethyl)sulfonyl)-1H-indol-3-yl)(6-fluoro-1-methyl-1H-benzo[d]imidazol-
2-yl)methanone (34) 1H NMR (500 MHz, Chloroform-d)
δ 8.89 (s, 1H), 8.30 (d, J = 7.9 Hz, 1H), 7.86 –
7.79 (m, 1H), 7.26 – 7.21 (m, 1H), 7.13 – 7.07 (m, 2H), 7.02 (dd, J = 12.5, 7.9 Hz, 1H), 4.79 (t, J = 7.3
Hz, 2H), 4.12 (s, 3H), 3.49 (t, J = 7.3 Hz, 2H), 3.26 (t, J = 4.8 Hz, 4H), 2.89 (t, J = 4.8 Hz, 4H). HRMS
(ESI) calcd for C23H23F2N5O3S (M + H)+: 488.1538, found 488.1564. LC-MS Purity: 97.5 %, Rt = 2.50
min.
Cell Culture and Bioactivity Test: HepG2 cells were cultured in low-glucose DMEM media
supplemented with 10% FBS, 1% Penicillin/Streptomycin, 1% non-essential amino acid, 1% sodium
pyruvate and 1% L-glutamine. When the cells reach 90% confluence, they were harvested and plated
1X105 cells per well in 96-well plate. After overnight settle-down, the culture medium was removed.
200 µL of dose medium containing compounds or vehicle was added to each well. After 24 h, each well
was refreshed with the same medium, followed by another 24 h incubation. Finally, medium was
collected after centrifugation at 3000 rpm for 3 min and was used for ELISA.
ELISA Assay: ELISA assay was used to measure PCSK9 concentration in cell culture medium
according to manufacture protocol (R&D system, DY3888). 100 µL of 2.0 ug/mL PCSK9 capture
antibody in PBS was add into high-binding plate (R&D system, DY990). After overnight incubation at
room temperature, the solution was removed and washed with ~400 µL of washing buffer (0.05%
Tween-20 in PBS) three times. Reagent diluent (300 µL, 1% BSA in PBS) was added to each well and
incubated for at least 1 h. Each of following steps was performed after the removal of previous solution
and washing out remaining solution with washing buffer. PCSK9 standard or sample (100 µL) with or
without dilution in reagent was added to each well followed with 2 h incubation. Next, 100 µL of 100
ng/mL PCSK9 detection antibody in reagent was added to each well followed with 2 h incubation. After
detection antibody incubation, 100 µL of Streptavidin-HRP was added to each well and incubated for 20
min. Finally, 100 µL of TMB solution was added and incubated for 20 min followed with the addition of
50 µL of stop solution (2 N H2SO4). After mixing, the optical density of each well was determined
immediately using a microplate reader set to 450 nm and wavelength correction to 540 nm or 570 nm.
27