Discovery of Potent Small-Molecule SIRT6 Activators: Structure-Activity Relationship and Anti-Pancreatic Ductal Adenocarcinoma Activity

Article abstract of DOI:10.1021/acs.jmedchem.0c01183

SIRT6 activation is thought to be a promising target for the treatment of many diseases, particularly cancer. Herein, we report the discovery of a series of new small-molecule SIRT6 activators. Structure-activity relationship analyses led to the identific

Full text of DOI:10.1021/acs.jmedchem.0c01183

Subscriber access provided by Stony Brook University | University Libraries
Article
Discovery of Potent Small Molecule SIRT6 Activators: Structure-Activity
Relationship and Anti-pancreatic Ductal Adenocarcinoma Activity
Xiuli Chen, Weining Sun, Shen-Zhen Huang, Hailin Zhang,
Guifeng Lin, Hui Li, Jingxin Qiao, Linli LI, and Shengyong Yang
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c01183 • Publication Date (Web): 12 Aug 2020
Downloaded from pubs.acs.org on August 12, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the
course of their duties.

Page 1 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of Potent Small Molecule SIRT6 Activators: Structure-Activity
Relationship and Anti-pancreatic Ductal Adenocarcinoma Activity
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
,‡
§,‡
†,‡
†
†
Xiuli Chen , Weining Sun , Shenzhen Huang , Hailin Zhang , Guifeng Lin , Hui
†
†
§
†,
Li , Jingxin Qiao , Linli Li , and Shengyong Yang *
†State Key Laboratory of Biotherapy and Cancer Center, West China Hospital,
Sichuan University, Chengdu, Sichuan 610041, P.R. China
§Key Laboratory of Drug Targeting and Drug Delivery System of Ministry of
Education, West China School of Pharmacy, Sichuan University, Chengdu, Sichuan
610041, P.R. China
‡
These authors contributed equally to this work.
*
To whom correspondence should be addressed. Shengyong Yang, State Key
Laboratory of Biotherapy and Cancer Center, West China Hospital, Sichuan
University, Chengdu, Sichuan 610041, China. E-mail: yangsy@scu.edu.cn.
KEYWORDS: SIRT6, deacetylation, small molecule activator, anti-pancreatic cancer
activity
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT
SIRT6 activation is thought to be a promising target for the treatment of many
diseases, particularly cancer. Herein, we report the discovery of a series of new small
molecule SIRT6 activators. Structure-activity relationship analyses led to the
identification of the most potent compound, 2-(1-benzofuran-2-yl)-N-(diphenylmethyl)
quinolone-4-carboxamide (12q), which showed an EC1.5 value of 0.58 ± 0.12 μM and
an EC50 value of 5.35 ± 0.69 μM against SIRT6-dependent peptide deacetylation in
Fluor de Lys assay. It exhibited weak or no activity against other HDAC family
members as well as 415 kinases, indicating a good selectivity for SIRT6. 12q
significantly inhibited the proliferation and migration of pancreatic ductal
adenocarcinoma (PDAC) cells in vitro. It also markedly suppressed the tumor growth
in a PDAC tumor xenograft model. This compound showed attractive
pharmacokinetic properties. Overall, 12q could be a good lead compound for the
treatment of PDAC, and it is worthy of further study.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 3 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
. INTRODUCTION
+
Sirtuins are NAD -dependent protein lysine deacetylases that participate in the
regulation of many physiological processes, such as the cell cycle, metabolism, stress
responses, and aging processes in prokaryotic and eukaryotic species.1-3 There are
seven mammalian sirtuins, SIRT1-SIRT7, and they differ in their subcellular
localization and in the substrate proteins they deacetylate.4, 5 Among these sirtuin
family members, of particular importance is SIRT6. The basic functions of SIRT6
include hydrolyzing long-chain acylated substrates, and catalyzing the deacetylation
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ε
of N -acetyl-lysines 9, 18, and 56 of histone H3 (H3K9ac, H3K18ac, and H3K56ac,
respectively), which have been associated with many physiological and pathological
phenotypes.6-14 The functional activation of SIRT6 might play an important role in
prolonging lifespan and intervening related diseases.15, 16 A recent study demonstrated
that SIRT6 is responsible for the more efficient DNA double-strand break repair in
long-lived species, and increased SIRT6 activity corresponds to longer lifespans.17
Numerous investigations have also shown that SIRT6 is involved in various diseases,
such as neurodegenerative diseases,18, 19 diabetes, cardiovascular diseases,
20
21, 22
and
cancers.23-28 Notably, SIRT6 has been demonstrated to function as a tumor-suppressor
gene in some cancers, implying that SIRT6 activation may have potential value in the
treatment of these cancers.13, 29, 30 For example, SIRT6 was found to be critical for the
suppression of pancreatic ductal adenocarcinoma (PDAC), one of the most lethal
malignancies.27 SIRT6 inactivation promotes PDAC progression and metastasis
through the upregulation of Lin28b, a negative regulator of let-7 microRNA.
Knocking out the SIRT6 gene results in histone hyperacetylation and c-Myc
recruitment at the Lin28b promoter, and pronounced induction of Lin28b and
downstream let-7 target genes, including IGF2BP1 and IGF2BP3.27 In contrast,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
SIRT6 activation might suppress PDAC progression and reduce metastasis formation
via the down regulation of Lin28b.27 SIRT6 activation is thus thought to be a
promising target in PDAC treatment.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Reported synthetic SIRT6 activators.
You et al reported the first synthetic SIRT6 activator, a pyrrolo[1,2-a]quinoxaline
derivative (UBCS039, Figure 1), which binds at a distal region of the fatty-acyl
substrate site and “C site” of SIRT6.31, 32 This compound showed an EC50 value of 38
3
1
μM against SIRT6. Recently, Huang et al disclosed a new potent SIRT6 activator
with cell activity, methyl 2-(N-(5-bromo-4-fluoro-2-methylphenyl)sulfamoyl)-5-(3,5-
dichlorophenyl sulfonamido)benzoate (MDL-800, Figure 1), which displayed an EC50
value of 10.3 μM against SIRT6.33 MDL-800 exhibited considerable activity in
decreasing H3K9ac and H3K56ac levels in human hepatocellular carcinoma (HCC)
cells.33 Obviously, the potency of these compounds is still not enough from the
perspective of a good drug candidate. Therefore, more potent and selective SIRT6
activators with new skeletons are needed.
To discover new SIRT6 activators, we performed a virtual screening against
various commercial chemical databases, which led to the discovery of several hit
compounds. Structural optimization to the most active hit compound was then carried
out, and this process offered a number of SIRT6 activators. The most potent activator
was subjected to further studies including assessment of its molecular level biological
activities, selectivity, and in vitro and in vivo anti-PDAC activities.
ACS Paragon Plus Environment

Page 5 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
2
. RESULTS AND DISCUSSION
.1 Discovery of Hit Compounds with the Aid of Molecular Docking-Based
Virtual Screening
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To obtain new activators of SIRT6, molecular docking was adopted to screen
commercial chemical databases including Specs, ChemDiv, Selleck, and
MedChemExpress, as well as an in-house database. Here, the crystal structure of the
SIRT6-UBCS039 (PDB entry 5MF6) complex was used, and GOLD was adopted for
molecular docking.31 The SIRT6-specific acyl channel pocket was defined as the
active site; this active site was very similar to the allosteric site of SIRT6 identified
3
3
later by Huang et al and was also found to accommodate SIRT6 modulators of the
32
34
35
quercetin compound family . GoldScore and ID-Score were utilized to estimate
the strength of interactions between the compound and the protein receptor. The final
ranking order of compounds was based on the consensus score of GoldScore and
ID-Score.36, 37 All the calculations were carried out with Discovery Studio (DS) 3.1
(
Accelrys Inc., San Diego, CA, USA).38
From the top ranked compounds, 40 compounds (Hit01-Hit40) were selected and
purchased for biochemical assays. We first tested the activity of each of these
compounds at a fixed concentration of 100 μM by the Fluor de Lys (FDL) assay using
a fluorogenic AMC-myristoyl peptide substrate (Ac-EALPKK(Myr)-AMC).31, 33, 39 In
this assay, four compounds (Hit05, Hit07, Hit11, and Hit20) showed clear activating
effects at 100 μM against SIRT6 (Figure 2A and 2B, and Supplementary Table S1).
We then measured the EC1.5 values of these compounds; EC1.5 is defined as the
concentration of a compound at which the compound is able to increase the enzymatic
activity by 50%.40, 41 The calculated EC1.5 values are 118.20 ± 1.33 μM, 109.80 ± 1.09
μM, 227.50 ± 1.31 μM, and 27.14 ± 1.18 μM for Hit05, Hit07, Hit11, and Hit20,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
respectively (Figure 2B and 2C). Structural optimization and structure-activity
relationship analyses were then carried out on the most active compound, Hit20.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. In vitro SIRT6 enzymatic activity assay for the screened compounds. A) Activation of
SIRT6-dependent peptide demyristoylation by 40 selected compounds at 100 μM, determined with
Ac-EALPKK(Myr)-AMC by the FDL assay. B) Chemical structures of hit compounds Hit05,
Hit07, Hit11, and Hit20. C) Dose-dependent demyristoylation activities of SIRT6 by 4 hit
compounds, determined with Ac-EALPKK(Myr)-AMC by the FDL assay. Data are presented as
the meanꢀ±ꢀs.d., n=ꢀ3 wells, from three independent experiments. D) Target regions for the
structural optimization and SAR analyses.
ACS Paragon Plus Environment

Page 7 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
.2 Structural Optimization and Structure-Activity Relationship (SAR) Analyses
of Hit20
To facilitate the structural optimization, molecular docking was used again to
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
predict the binding mode of Hit20 with SIRT6. As shown in Figure S1, Hit20 binds to
the specific acyl channel pocket of SIRT6, a similar binding site of UBCS039 and
MDL-801.31-33 A face-to-face π-π interaction between the benzene ring of the
quinoline scaffold and the phenyl ring of PHE86, and a σ-π interaction between the
amino group of ALA-7 and the benzene ring of 3,4-dichlorobenzene are formed. We
also noticed that there is some space not completely occupied around
3,4-dichlorobenzene group (the 4-position carboxamide substituent on the quinoline,
region I) and 3,4-dimethoxybenzene group (2-position on the quinoline, region II). It
is expected that modifications to these two regions, for example adding hydrogen
bond donors or receptors, or increasing hydrophobic interaction, might benefit the
improvement of bioactivity. Therefore, the subsequent structural optimization will
focus on region I and region II (Figure 2D). The SAR analyses are based on the
activation potencies of the compounds against SIRT6, which were determined by an
FDL assay with peptide Ac-EALPKK(Myr)-AMC as the substrate.
1
2
.2.1 Impact of the 4-Position Carboxamide Substituents (Region I, R ) on the
Quinolone
To examine the influence of 4-position carboxamide substituents (region I, R1)
on the quinolone on the bioactivity, we fixed region II as the original 3,4-dimethoxy
phenyl group and varied the group in region I. Here region I groups chosed include
phenyl groups with different substituents, as well as other fragments that might help
increase hydrogen bonding interaction and/or hydrophobic interaction. Forty
1
quinoline-4-carboxamide derivatives (PC01-40) with different R substituents were
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
purchased from commercial reagent companies. In addition, we also synthesized
fifteen new quinoline-4-carboxamide derivatives (5a-o). The synthetic routes of 5a-o
are depicted in Scheme 1. Condensation reactions of commercially available
indolin-2,3-dione (1) and 3,4-dimethoxyacetophenone (2) in EtOH in the presence of
an appropriate amount of KOH aqueous solution generated 2-(3,4-dimethoxyphenyl)
quinolone-4-carboxylic acid (3) as an intermediate a yield of 76%. Intermediate 3 was
then reacted with various substituted anilines to offer final products 5a-o. The final
yields were between 45%-79%.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Synthesis of compounds 5a-o.a
a
1
Reagents and conditions: a) KOH, EtOH, 85 °C , 24 h; b) i: SOCl
DMAP, DMF, RT, overnight.
2
, 107 C, 5-6 h; ii: R -NH ,
2
The chemical structures and bioactivities of compounds PC01-40 and 5a-o are
given in Table 1. Among compounds PC01-24 and 5a-k containing substituted phenyl
1
groups at the R position, eight compounds (5a, PC12, PC14, PC17, 5f-h, and 5j)
ACS Paragon Plus Environment

Page 9 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
displayed activation effects of >150% at 20 μM. EC1.5 values of these compounds are
1
0.28 ± 2.38 μM, 5.97 ± 1.99 μM, 19.02 ± 1.14 μM, 5.61 ± 1.24 μM, 7.53 ± 1.31 μM,
6.38 ± 1.98 μM, 2.84 ± 0.37 μM, and 4.77 ± 1.81 μM, respectively. For compounds
PC25-40 and 5l-o, which contain a naphthalene ring, benzyl moiety, polyunsaturated
ring or heterocycle with larger substituents, six compounds (5l, PC25-26, and
PC28-30) showed activation potencies of >150% at 20 μM. The EC1.5 values of these
compounds are 7.44 ± 2.08 μM, 5.97 ± 1.98 μM, 6.47 ± 4.40 μM, 7.78 ± 1.60 μM,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
9.53 ± 1.35 μM, and 8.14 ± 1.59 μM, respectively. Of all the fifty-five compounds,
h, which contains a 2-chloro-5-(trifluoromethyl) phenyl group, is the most active one
5
in terms of the EC1.5 values. We then fixed the R1 substituent as a
-chloro-5-(trifluoromethyl) phenyl group in the following optimization step.
2
Table 1. Chemical Structures and Effect of Compounds PC01-40 and 5a-o on
SIRT6-dependent Peptide Demyristoylation Activation.
% Effect
@20 μM
% Effect
@20 μM
R1
R1
ID
ID
(EC1.5 /μM)
(EC1.5 /μM)
1
47.60
Hit20
PC02
PC04
PC01
PC03
PC05
103.89
104.63
92.42
(27.14 ± 1.18)
98.69
104.33
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
PC06
PC08
PC10
98.32
83.32
PC07
PC09
PC11
5b
128.49
95.21
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
131.49
117.76
114.98
142.38
165.72
5a
(10.28± 2.38)
5
5
c
e
141.68
5d
223.94
124.45
107.90
101.37
PC12
PC14
PC16
PC18
5g
(
5.97 ± 1.99)
198.91
PC13
PC15
PC17
(
19.02 ± 1.14)
116.66
313.43
131.17
249.98
(
(
5.61 ± 1.24)
181.36
5
f
7.53 ± 1.31)
(6.38 ± 1.98)
364.62
5h
PC19
PC20
PC22
102.9
(
2.84 ± 0.37)
65.66
5i
96.94
PC21
PC23
95.52
98.19
100.97
PC24
87.18
ACS Paragon Plus Environment

Page 11 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
284.13
5
j
5k
88.61
(
(
(
4.77 ± 1.81)
326.96
5
l
5m
139.32
7.44 ± 2.08)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
276.15
239.45
PC25
PC26
5.97 ± 1.98)
(6.47 ± 4.40)
225.36
PC27
PC29
107.49
PC28
PC30
(
7.78 ± 1.60)
168.5
259.94
(19.53 ± 1.35)
(8.14 ± 1.59)
124.23
PC31
PC33
5o
105.48
PC32
5n
99.04
86.13
127.6
90.98
PC34
PC35
PC37
PC39
90.3
101.43
82.90
PC36
PC38
PC40
83.37
95.12
95.22
2
2
.2.2 Effect of the 2-Position (Region II, R ) on the Quinolone
2
To explore the effect of the substituent at the 2-position (region II, R ) of
quinolone, we installed 5-member ring, 6-member ring, or benzoheterocycles in
2
region II (R ) and fixed region I as the optimal 2-chloro-5-(trifluoromethyl)phenyl
group. We purchased two quinolone-4-carboxamide derivatives with different
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
substituents at R (PC41-42) and synthesized thirteen new quinolone-4-carboxamide
derivatives (8a-m). Scheme 2 illustrates the synthetic routes of compounds 8a-m.
Similar to before, the condensation reactions between indolin-2,3-dione (1) and
various substituted acetophenones (6a-m) in EtOH in the presence of an appropriate
amount of KOH aqueous solution generated 2-substituted quinolinone-4-carboxylic
acids (7a-m) as intermediates. The final products (8a-m) were then obtained from the
reactions of 7a-m with commercially available 2-chloro-5-(trifluoromethyl) aniline.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 2. Synthesis of compounds 8a-m.a
a
Reagents and conditions: a) KOH, EtOH, 85 °C , 24 h; b) i: SOCl , 107 C, 5-6 h; ii:
2
2
-chloro-5-(trifluoromethyl)aniline, DMAP, DMF, RT, overnight.
The chemical structures and bioactivities of compounds PC41-42 and 8a-m are
shown in Table 2. Compounds PC41-42 and 8a, containing a different substituted
phenyl groups, did not activate SIRT6. Compound 8b with 4-methoxyphenyl did
activate SIRT6 with an EC1.5 value of 20.72 ± 3.01 μM. Compounds 8d and 8e,
bearing saturated five- or six-membered rings, also did not activate SIRT6. In addition,
ACS Paragon Plus Environment

Page 13 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
compounds 8f and 8g, which contain a (substituted) thiophene at the R position, did
not show obvious activity against SIRT6 at 20 μM. Interestingly, compounds 8h-m,
bearing a (substituted) furan, or a benzo oxygen-containing heterocycle, displayed
obvious activation effects, and 8j is the most active one with an EC1.5 value of 0.85 ±
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0
.36 μM .
Table 2. Chemical Structures and Effect of Compounds PC41-42 and 8a-m on
SIRT6-dependent Peptide Demyristoylation Activation.
%
Effect
% Effect
@20 μM
ID
R2
@20 μM
ID
R2
(
EC1.5 /μM)
(EC1.5 /μM)
364.62
5h
PC41
8a
92.33
91.20
85.57
(
2.84 ± 0.37)
PC42
108.96
1
63.4
8b
8c
(20.72 ± 3.01)
8d
82.43
8e
8g
8i
76.38
8f
130.20
140.80
141.05
184.00
8h
(
18.61 ± 1.61)
7
62.05
321.32
8j
8k
(
0.85 ± 0.36)
(5.96 ± 1.36)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
28.42
158.86
8
l
8m
(14.86 ± 1.47)
(29.39 ± 5.37)
3
2.2.3 Re-optimization of the 4-Position (Region I, R ) on the Quinolone
In the preceding SAR studies, we determined an optimal moiety (benzofuranyl
group) for region II with region I fixed as 2-chloro-5-(trifluoromethyl)phenyl, which
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
was
also
determined
through
optimization.
To
examine
whether
2
-chloro-5-(trifluoromethyl)phenyl is still optimal when region II is fixed as a
benzofuranyl moiety, we synthesized seventeen compounds with different substituents
3
at the 4-position (region I, R ) of the quinolone (12a-q). Compounds 12a-q were
prepared following the route outlined in Scheme 3, which is very similar to that in
Scheme 1.
Scheme 3. Synthesis of compounds 12a-q.a
ACS Paragon Plus Environment

Page 15 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
3
Reagents and conditions: a) KOH, EtOH, 85 C, 24 h; b) R -NH
C, overnight.
2
, HATU, HOBt, DMAP, 50
Bioactivities of compounds 12a-q are shown in Table 3. All the compounds
showed obvious activities against SIRT6 at 20 μM. Compound 12q, which contains a
diphenylmethane group in region I, exhibited more potent activity than 8j. The EC1.5
value of compound 12q is 0.72 ± 0.25 μM, which is 38-fold more potent than Hit20.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 3. Chemical Structure and Effect of Compounds 12a-q on SIRT6-dependent
Peptide Demyristoylation Activation.
%
Effect
% Effect
@ 20μM
R3
R3
ID
@ 20μM
ID
12a
12c
12e
12g
12i
(
EC1.5 /μM)
(EC1.5 /μM)
7
62.05
8j
108.95
(
0.85 ± 0.36)
1
63.48
204.83
12b
(
20.72 ± 5.76)
(9.01 ± 3.51)
348.60
314.93
12d
(
3.47 ± 0.97)
(3.71 ± 1.67)
2
15.60
334.64
12f
(
4.86 ± 1.76)
(5.17 ± 1.90)
4
29.78
287.42
12h
(3.11 ± 1.99)
(4.96 ± 2.20)
313.96
12j
109.27
12k
(5.91 ± 1.53)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
208.02
12m
481.51
1
2l
(14.50 ± 3.94)
(2.19 ± 1.02)
321.40
1
1
2n
2p
119.23
118.25
12o
12q
(7.59 ± 0.99)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1229.81
(
0.72 ± 0.25)
2.3 Bioactivity Validation of Compound 12q by Various Biochemical and
Biophysical Assays
To further validate the bioactivity of 12q in vitro, various biochemical and
biophysical assays were performed, including FDL assays but with different substrate
peptides, differential scanning fluorimetry (DSF) assays, isothermal titration
calorimetry (ITC) assays, and surface plasmon resonance (SPR) assays. MDL-800
was used as a reference compound because it is the most potent SIRT6 activator
reported so far.
2
.3.1 FDLAssay with Different Substrate Peptides
4
2
As indicated before, deacetylation is one of most important functions of SIRT6.
We thus validated the influence of 12q on the deacetylation activity of SIRT6. To this
end, the FDL assay was used again, and acetyl peptide substrate
(
Ac-RYQK(Ac)-AMC) was used.43 As shown in Figure 3A, compound 12q
dramatically enhanced SIRT6 deacetylation activity with an EC1.5 value of 0.58 ± 0.12
μM, which is comparable to that of demyristoylation activity (0.72 ± 0.25 μM, Table
3
). Dose-dependent curves of 12q against SIRT6 deacetylation and demyristoylation
are displayed in Figure 3A and 3B, respectively. And the calculated EC50 values of
2q are 5.35 ± 0.69 μM and 8.91 ± 1.81 μM for SIRT6 deacetylation and
1
demyristoylation, respectively. Collectively, 12q is able to enhance both the
ACS Paragon Plus Environment

Page 17 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
deacetylation activity and demyristoylation activity of SIRT6. Though the positive
control MDL-800 showed excellent SIRT6-dependant deacetylation activity, it could
not enhance the demyristoylation activity of SIRT6 (Figure 3B), which is consistent
with Huang’s result.33
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
One may notice that different peptides were used in deacetylation
(Ac-RYQK(Ac)-AMC) and demyristoylation (Ac-EALPKK(Myr)-AMC) assays. To
examine possible influence of different peptides in FdL assays, we tested the activity
of 12q with Ac-RYQK(Myr)-AMC as the substrate. The measured EC1.5 and EC50
values are 0.49 ± 0.11 μM and 5.49 ± 0.57 μM (Figure S2), respectively, which are
comparable to the corresponding values in the cases of Ac-RYQK(Ac)-AMC and
Ac-EALPKK(Myr)-AMC. All the results indicate that different peptide substrates
have no significant impact on the bioactivity.
2.3.2 DSF Assay
The bioactivity of 12q was then validated by the DSF assay, which is a
4
4
spectroscopic technique used to identify the thermal stability of a protein. In this
assay, the recombinant human protein SIRT6 was used as the target protein. The
thermal stability of SIRT6 in the presence or absence of small molecule ligands was
measured. The melting temperature (T
m
) values of SIRT6 are 41.18 ± 0.10 °С and
4
2.19 ± 0.10 °С treated with DMSO and 12q at a concentration of 20 µM,
respectively, indicating a T
m
value of 1.01 °С (Figure 3C). At the same concentration,
value of 0.85 °С.
2.3.3 ITC and SPR Assays
ITC and SPR assays were adopted to measure the binding affinity of 12q with
SIRT6. In the ITC assay, 12q showed an equilibrium dissociation constant (K ) of
.85 ± 1.89 μM (Figure 3D). The measured thermodynamic binding parameters are
MDL-800 displayed a T
m
d
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
-1
-1
-1
-1
1
4.33 ± 5.60 kcal mol , 71.20 cal mol K , -6.90 ± 5.60 kcal mol and -21.23 kcal
-1
mol for ΔH, ΔS, ΔG, and -TΔS, respectively. The stoichiometry factor (N) is 0.31 ±
0.098. In the SPR assay, compound 12q displayed an equilibrium dissociation
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
constant (Kd) of 5.52 μM (Figure 3E). MDL-800 showed a K of 9.51 μM in the
d
same assay (Figure S3).
Figure 3. Bioactivities of 12q against SIRT6 by various biochemical and biophysical assays. A, B)
Dose-activity curves of SIRT6-dependant deacetylation and demyristoylation of 12q, determined
with acetyl substrate peptide Ac-RYQK(Ac)-AMC and myristoyl substrate peptide
Ac-EALPKK(Myr)-AMC. Data are presented as the meanꢀ±ꢀs.d., n=ꢀ3 wells, from three
independent experiments. C) Changes in the thermodynamic stability of SIRT6 after the binding
ACS Paragon Plus Environment

Page 19 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of 12q or MDL-800. D) ITC binding curves for SIRT6 and 12q. E) Representative SIRT6 binding
curves and fit steady-state evaluation for 12q using SPR.
2
.3.4 High Performance Liquid Chromatography (HPLC) Assay
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The HPLC assay was applied to test the SIRT6-dependant deacetylase activity of
1
2q.
Here,
both
peptide
substrates
Ac-RYQK(Ac)-AMC
and
Ac-EALPKK(Myr)-AMC were used. The results showed that 12q could enhance the
SIRT6-dependant deacetylation and demyristoylation activities (Figure S4).
2
.4 Selectivity of 12q
To test the selectivity of 12q, we measured the half maximal inhibitory
concentration (IC50) values of 12q against other HDAC family members, including
SIRT1-3, SIRT5, and HDAC1-11; SIRT4 and SIRT7 are not involved because they
are not available for us at present. As shown in Table 4, 12q potently activated SIRT6
but showed no activity toward SIRT2, SIRT3, SIRT5, and HDAC1-11 at
concentrations up to 200 µM. Although 12q exhibited activity against SIRT1, the
potency is almost 300 folds lower than that against SIRT6. We further tested the
activity of 12q against a panel of 415 recombinant human kinases, and 12q showed no
activity or very weak activity against these kinases (Table S2). These results indicated
that 12q is a selective activator of SIRT6 and is not a Pan Assay Interference
Compound (PAIN).
Table 4. Activities of 12q against HDAC Family Members.
Target
IC50 μM
Target
IC50 μM
SIRT1
SIRT2
171.20 ± 14.98
>200
HDAC4
HDAC5
>200
>200
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
SIRT3
SIRT5
>200
>200
.58 ± 0.12
EC1.5)
>200
>200
>200
HDAC6
HDAC7
HDAC8
HDAC9
HDAC10
HDAC11
>200
>200
>200
>200
>200
>200
0
SIRT6
a
(
HDAC1
HDAC2
HDAC3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
aThe activity of 12q against SIRT6 was determined by the FDL assay with acetyl substrate peptide
Ac-RYQK(Ac)-AMC.
2
.5 Predicted Binding Mode of 12q
Because we failed to obtain a crystal structure of the 12q-SIRT6 complex,
molecular docking was used again to predict the binding mode of 12q and SIRT6.
Figure 4 presents a predicted binding mode of 12q with SIRT6. 12q occupies the
3
1
33
allosteric pocket, which is very similar to that of UBCS039 and MDL-801 (also
see Figure S5, a superposition of three complex structures). The oxygen atom of the
carboxamide of 12q forms one hydrogen bond with the amide group in the side chain
of ASN4. The N-benzhydryl group of 12q fit perfectly in the allosteric pocket via
hydrophobic interactions with residues TYR5, VAL70, PHE82, PRO62, and PRO80.
The benzofuran forms hydrophobic interactions with LYS160 and MET157. Two
face-to-face π−π and one σ−π interactions are formed. One of the π−π interactions is
between the benzene ring of the N-benzhydryl group and the benzene ring of residue
PHE82, and the other π−π interaction is between the benzene ring of the quinoline and
the benzene ring of residue PHE86. The σ-π interaction is between the benzene ring in
the benzofuran group and the amino group in THR156 (Figure 4). These interactions
well explain the high potency of 12q.
ACS Paragon Plus Environment

Page 21 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Predicted binding mode of 12q with SIRT6 (the SIRT6 structure was taken from PDB
entry 5Y2F). 12q (green), ADPR (apricot ), H3K9-Myr (magenta) are show in the “zoomed-in”
image.
2
.6 In Vitro Anti-PDAC Activities of Compound 12q
2.6.1 Anti-viability and Anti-proliferation Activities of 12q Against PDAC Cells
The MTT assay was used to measure the anti-viability activity of 12q against
PDAC cells. Four human PDAC cell lines, PANC-1, BXPC-3, MIAPaCa-2, and
AsPC-1, were chosen for this study. As shown in Figure 5A, 12q exhibited
anti-viability activity against all the tested cell lines in a dose-dependent manner, with
IC50 values of 4.13 ± 0.15 μM, 8.27 ± 1.08 μM, 7.10 ± 1.85 μM, and 9.66 ± 1.13 μM,
against PANC-1, BXPC-3, MIAPaCa-2, and AsPC-1, respectively. In this assay,
MDL-800 also showed activity but with relatively weak potency (IC50 values of 35.19
±
1.21 μM, 57.96 ± 1.11 μM, 55.65 ± 1.03 μM, 50.51 ± 1.01 μM for PANC-1,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
BXPC-3, MIAPaCa-2, and AsPC-1, respectively) (Figure 5B). To examine the
tumor-specificity effect, we tested the anti-viability activity of 12q against eight cell
lines of other tumor types, including two liver cancer cell lines (HEPG2 and
BEL7402), one thyroid cancer cell line (OCUT-2C), one ovarian cancer cell line
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(SKOV-3), one colon cancer cell line (SW620), and three lung cancer cell lines (H526,
H1975, and NCl-H446). Although 12q displayed some activities against these tumor
cell lines, the potencies are weaker compared with those of PDAC cell lines (Table
S3), indicating a certain degree of selectivity for PDAC cells.
The colony formation assay was performed to investigate the anti-proliferation
activity of 12q against PDAC cells. As shown in Figure 5C and Figure S6A,
compound 12q could inhibit the colony formation of PANC-1, BXPC-3, MIAPaCa-2,
and AsPC-1 cells in a dose dependent-manner. The anti-proliferation activity of 12q
was further confirmed by an EdU incorporation assay (Figure 5D and Figure S6B).
2
.6.2 Cell Cycle Arrest and Apoptosis Induction Ability of 12q Against PDAC
Cells
Flow cytometry was adopted to detect cell cycle arrest and apoptotic cells upon
treatment with 12q. The results showed that compound 12q caused a dose-dependent
increase in the percentages of PANC-1 and BXPC-3 cells in the G2 phase (Figure S7),
indicating that 12q could cause PDAC cell arrest in the G2 phase. 12q treatment also
+
increased Annexin V populations in a concentration-dependent manner, implying
that 12q could induce apoptosis in PANC-1 and BXPC-3 cells (Figure S8).
ACS Paragon Plus Environment

Page 23 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Antiproliferative activities of 12q against pancreatic cancer cells in vitro. A, B) Cells
were treated with the indicated agents for 72 h, and cell viability was measured by an MTT assay.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Every experiment was carried out in triplicate. C) Cells were incubated with various
concentrations of the indicated agents for 18 days (BXPC-3) or 14 days (PANC-1). Then, the cells
were stained with crystal violet and quantified. D) The fluorescence microscopic appearance of
EdU and Hoechst on BXPC-3 and PANC-1 cells after treatment with the indicated agents for 24 h.
EdU-positive cells were quantified in 10 fields. Scale bars, 100 μm.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
.6.3 Anti-migration Ability of 12q
A wound-healing assay was used to evaluate the anti-migration ability of 12q in
vitro. Four human PDAC cell lines, PANC-1, BXPC-3, MIAPaCa-2, and AsPC-1,
were used for this experiment. As shown in Figure 6A and Figure S9, 12q
significantly inhibited the migration of PDAC cells in a time- and dose-dependent
manner, with migration inhibition rates reaching approximately 87.8%, 70.0%, 57.2%,
and 55.4% in the presence of 10 μmol/L 12q at 24 h, respectively.
2
.7 Bioactivities of 12q in Intact Cells
The cellular thermal shift assay (CETSA) was first used to investigate whether
12q could bind to the SIRT6 protein in intact cells. As shown in Figure 6B, 12q
treatment efficiently protected SIRT6 protein from temperature-dependent
degradation in PANC-1 and BXPC-3 cells, indicating that 12q is able to bind to
SIRT6. Western blot was used to examine the effect of 12q on the acetylation status of
H3K9ac, H3K18ac, and H3K56ac in intact cells; H3K9, H3K18, and H3K56 are the
typical substrates of SIRT6. As shown in Figure 6C, 12q decreased the levels of
H3K9ac, H3K18ac, and H3K56ac in PANC-1 and BXPC-3 cells in a dose-dependent
manner. MDL-800 showed the same effect.
We further investigated the influence of 12q on the Lin28b / let-7 signal pathway
in PDAC cells. As indicated before, it has been reported that SIRT6 functional loss
accelerates PDAC progression and metastasis via upregulation of Lin28b, and c-Myc
recruitment. As displayed in Figure 6C, 12q treatment decreased the expression of
ACS Paragon Plus Environment

Page 25 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Lin28b and c-Myc.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 6. Anti-migration ability of 12q and its bioactivities in intact cells. A) Anti-migration
activity of 12q against pancreatic cancer cells in vitro. BXPC-3 or PANC-1 cells were treated with
various concentrations of the indicated agents. Scale bars, 100 μm. B) Western blot analysis of
SIRT6 and its substrates H3K9ac, H3K18ac, and H3K56ac, and Lin28b, c-Myc, and IGF2BP3
protein expression in PANC-1 and BXPC-3 cells treated with the indicated doses of 12q and
MDL-800 for 72ꢀh. C) A cellular thermal shift assay (CETSA) was implemented in intact cells
exposed to 12q (25 μM).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
.8 In Vivo Effects of 12q on Subcutaneous PANC-1 Tumor Xenograft
To evaluate the antitumor activity of 12q in vivo, we used the human pancreatic
tumor xenograft model of PANC-1. BALB/c female nude mice were orally
administered 100 or 150 mg/kg/d of 12q for 30 days. MDL-800 (150 mg/kg/d) was
used as a positive control. In all the treated group, tumor growth was slowed by 12q
in a dose-dependent manner, and a tumor inhibition rate of 90.25% at a dose of 150
mg/kg was observed (Figure 7A and 7B). No significant weight loss or toxicity were
observed compared with the control group (Figures S10 and S11). MDL-800 also
showed considerable antitumor potency with a tumor inhibition rate of 78.03% at a
dose of 150 mg/kg/d.
To elucidate the mechanism of 12q-mediated antitumor efficacy in vivo, we
conducted immunohistochemistry (IHC) analysis in the PANC-1 tumor model. As
depicted in Figure 7C, 12q potently inhibited the expression of Lin28b and c-Myc and
led to a substantial decrease in tumor cell proliferation (Ki-67-positive cells)
compared with the control group in the PANC-1 tumor model. The level of H3K9ac
was lower in the 12q-treated groups than in the vehicle group. These results revealed
that 12q may suppress PDAC xenograft tumor growth in vivo by activating the
deacetylase activity of SIRT6.
ACS Paragon Plus Environment

Page 27 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 7. Antitumor efficacy of 12q in vivo. A) Mice bearing PANC-1 tumor xenografts were
treated with agents at the indicated dose or vehicle control alone over the designated treatment
schedule once tumors reached the determined size (n=6). Points, mean tumor volume; bars, SD. B)
Quantitative analysis of tumor volumes on the final study day. ***P < 0.001. C) Tumor tissue
from PANC-1 xenografts treated with vehicle control, 12q (150 mg/kg), or reference agents at the
dose mentioned above were evaluated by IHC. Scale bars, 50 μm.
2.9 Preliminary Pharmacokinetic Properties of 12q
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 69
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The preliminary pharmacokinetic (PK) properties of compound 12q were
determined by oral administration to male Sprague-Dawley rats. The pharmacokinetic
parameters of 12q at a dose of 10 mg/kg are summarized in Table 5. The area under
the concentration-time curve (AUC(0-∞)) is 755.57 h·ng/mL. The maximum plasma
concentration (Cmax) is 98.45 ng/mL. The bioavailability (F) and half-life (T1/2) are
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
% and 7.52 h, respectively.
Table 5. In vitro Pharmacokinetic Profiles of Compound 12q.
10 mg/kg p.o. a
2 mg/kg i.v. a
Pharmacokinetic parameter
CL (L/h/kg)
-
0.6 ± 0.08
Vss (L/kg)
-
1112.8 ± 322.84
9.06 ± 0.21
T1/2 (h)
7.52 ± 1.44
2.00 ± 0.00
98.45 ± 3.62
704.67 ± 80.47
755.57 ± 80.74
4.24 ± 0.48
Tmax (h)
0.08 ± 0.00
C
max (ng/mL)
5123.70 ± 905.5
3326.13 ± 476.4
3381.49 ± 468.48
-
AUC(0-t) (h·ng/mL)
AUC(0-∞) (h·ng/mL)
F (%)
a
Expressed as Mean ± SD, n = 3
3
. CONCLUSIONS
In this investigation, we identified
a
potent SIRT6 activator,
2-(1-benzofuran-2-yl)-N-(diphenylmethyl)quinolone-4-carboxamide (12q), through
virtual screening, chemical synthesis, and SAR analyses. This compound showed
EC1.5 value of 0.58 ± 0.12 μM (EC50 = 5.35 ± 0.69 μM) and 0.72 ± 0.25 μM (EC50
=
8
.91 ± 1.81 μM) against SIRT6 in the FDL assays with acetyl peptide and myristoyl
peptide substrates, respectively. The bioactivity of 12q was further validated by DSF,
SPR, and ITC assays. In target selectivity assays, 12q displayed excellent selectivity
for SIRT6 against other tested SIRT family members and HDACs. In various in vitro
ACS Paragon Plus Environment

Page 29 of 69
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
assays including MTT, colony formation, flow cytometry, and wound-healing assays,
1
2q exhibited considerable anti-PDAC activities. In the subcutaneous PANC-1 tumor
xenograft model, 12q showed potent anti-tumor activity. 12q also exhibited fairly
good pharmacokinetic (PK) properties. Overall, 12q could be a promising lead
compound for the treatment of pancreatic cancer and is worthy of further in-depth
studies.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
. MATERIALS AND METHODS
4.1 Virtual Screening Based on Molecular Docking
In this study, molecular docking was performed using GOLD (Genetic
Optimization of Ligand Docking) 5.0.34, 45 X-ray crystal structures of SIRT6 in
complex with UBCS039 (PDB entry 5MF6) and MDL-801 (PDB entry 5Y2F) were
used in the docking studies.33 The Discovery Studio 3.1 (Accelrys, Inc. USA)
software package was adopted to prepare the protein structure, including adding
hydrogen atoms to the protein, removing water molecules, and assigning the force
field (here, the CHARMm force field was adopted). The binding site was defined as a
sphere containing residues within 10 Å of the ligand, an area large enough to cover
the ligand-binding region. Compound libraries, including Specs, ChemDiv, Selleck,
MedChemExpress, and an in-house database, were filtered by SMARTS-PAINS and
38
Other Bad Groups as reported , and used for the virtual screening in this
investigation.
4.2 Chemistry Methods
Starting materials, reagents, and solvents used in the experiments were purchased
from commercial vendors unless otherwise noted, and they were analytically pure or
chemically pure. Anhydrous solvents were either dried by distillation under reduced
pressure or purchased from J&K Scientific. All reactions were monitored by
ACS Paragon Plus Environment