Small Molecule Inhibitors Simultaneously Targeting Cancer Metabolism and Epigenetics: Discovery of Novel Nicotinamide Phosphoribosyltransferase (NAMPT) and Histone Deacetylase (HDAC) Dual Inhibitors

Article abstract of DOI:10.1021/acs.jmedchem.7b00467

Cancer metabolism and epigenetics are among the most intensely pursued research areas in anticancer drug discovery. Here we report the first small molecules that simultaneously inhibit nicotinamide phosphoribosyltransferase (NAMPT) and histone deacetylase (HDAC), two important targets of cancer metabolism and epigenetics, respectively. Through iterative structure-based drug design, chemical synthesis, and biological assays, a highly potent dual NAMPT and HDAC inhibitor was successfully identified. Compound 35 possessed excellent and balanced activities against both NAMPT (IC₅₀ = 31 nM) and HDAC1 (IC₅₀ = 55 nM). It could effectively induce cell apoptosis and autophagy and ultimately led to cell death. Importantly, compound 35 showed excellent in vivo antitumor efficacy in the HCT116 xenograft model. This proof-of-concept study demonstrates the feasibility of discovering an inhibitor targeting cancer metabolism and epigenetics and provides an efficient strategy for multitarget antitumor drug discovery.

Full text of DOI:10.1021/acs.jmedchem.7b00467

Subscriber access provided by GRIFFITH UNIVERSITY
Article
Small Molecule Inhibitors Simultaneously Targeting Cancer Metabolism and
Epigenetics: Discovery of Novel Nicotinamide Phosphoribosyltransferase
(NAMPT) and Histone Deacetylase (HDAC) Dual Inhibitors
Guoqiang Dong, Wei Chen, Xia Wang, Xinglin Yang, Tianying Xu, Pei
Wang, Wannian Zhang, Yu Rao, Chao-Yu Miao, and Chunquan Sheng
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00467 • Publication Date (Web): 08 Sep 2017
Downloaded from http://pubs.acs.org on September 8, 2017
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 free 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 accessible to all
readers and 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.
Journal of Medicinal Chemistry 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 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Small Molecule Inhibitors Simultaneously Targeting Cancer
Metabolism and Epigenetics: Discovery of Novel Nicotinamide
Phosphoribosyltransferase (NAMPT) and Histone Deacetylase
(HDAC) Dual Inhibitors
†,1
†,1
†,1
2
1
1
Guoqiang Dong , Wei Chen , Xia Wang , Xinglin Yang , Tianying Xu , Pei Wang ,
1
,2
,1
,1
Wannian Zhang , Yu Rao* , Chaoyu Miao* , Chunquan Sheng*
1
School of Pharmacy, Second Military Medical University, 325 Guohe Road,
Shanghai 200433, People’s Republic of China
2
MOE Key Laboratory of Protein Sciences, School of Pharmaceutical Sciences,
Tsinghua University, Beijing 100084, People’s Republic of China
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 65
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
Cancer metabolism and epigenetics are among the most intensely pursued research
areas in anticancer drug discovery. Here we report the first small molecules that
simultaneously inhibit nicotinamide phosphoribosyltransferase (NAMPT) and histone
deacetylase (HDAC), two important targets of cancer metabolism and epigenetics,
respectively. Through iterative structureꢀbased drug design, chemical synthesis and
biological assays, a highly potent dual NAMPT and HDAC inhibitor was successfully
identified. Compound 35 possessed excellent and balanced activities against both
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
NAMPT (IC = 31 nM) and HDAC1 (IC = 55 nM). It could effectively induce cell
50
50
apoptosis and autophagy and ultimately led to cell death. Importantly, compound 35
showed excellent in vivo antitumor efficacy in the HCT116 xenograft model. This
proofꢀofꢀconcept study demonstrates the feasibility of discovering an inhibitor
targeting cancer metabolism and epigenetics and provides an efficient strategy for
multiꢀtarget antitumor drug discovery.
Keywords: NAMPT, HDAC, dual inhibitor, structureꢀbased design, antitumor
activity.
2
ACS Paragon Plus Environment

Page 3 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
In recent decades, drug discovery has been strongly focused on the development of
1
, 2
drugs acting against a specific biological target with high potency and selectivity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
This paradigm is based on the lock and key model proposed by Fischer and has led to
3
numerous successful drugs in the market. However, single target drugs might not
achieve the desired therapeutic effects for the treatment of complex diseases such as
cancer. The underlying reason is that the dysregulation of multiple signaling pathways
is a hallmark of cancer development and progression, and the existence of
4
ꢀ6
compensating signaling pathways has reduced the efficacy of single target drugs.
Specific drug combinations have expanded the treatment to achieve more durable
disease control. However, the employment of drug cocktails often complicates the
doses/schedule and negatively impacts patient compliance. It can also introduce the
2
unpredictable pharmacokinetic (PK) profile and drugꢀdrug interactions. In contrast, a
single multiꢀtarget molecule has the advantages of higher synergistic effect, more
1
predictable PK profile, reduced compliance difficulties and so on. However, rational
design of multiꢀtarget drugs still remains a highly challenging task, particularly for the
targets belong to different protein families. The difficulties mainly include the
identification of the most suited combination of targets, achieving the balanced or
desired activities against different targets, complex optimization of polypharmacology
1, 2, 7
profile and drugꢀlikeness to transform leads into drug candidates.
For example,
with deep understanding of tumor pathogenesis, multiꢀtarget antitumor agents have
been considered as an important strategy to overcome drug resistance and improve
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 65
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
, 2, 4, 8
therapeutic effects.
Epigenetic regulation has played an important role in the onset and progression of
9
human diseases. Specifically, histone deacetylases (HDACs) are a family of
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
9
epigenetic enzymes that has achieved great success for cancer therapeutics. HDAC
inhibitors (HDACi) have been recently developed as potent antineoplastic agents.
Four of them, 1 (SAHA, Figure 1A), FK228, PXDꢀ101 and LBHꢀ589 have been
1
0ꢀ13
approved by FDA for the treatment of hematologic cancer.
However, treatment of
solid tumors by HDACi still remains a big challenge and discovery of HDACiꢀbased
14, 15
multiꢀtarget antitumor agents may be an effective strategy.
Cancer is a disease of genetic heterogeneity, but the metabolic alteration is the
1
6, 17
common feature of malignant cells.
During the past decade, targeting cancer
metabolism has emerged as a promising strategy for the development of antitumor
18
agents. Among these targets, nicotinamide adenine dinucleotide (NAD) metabolism
has attracted broad interests because of its essential role in a number of cellular
1
9
process to sustain tumor cell growth and survival. NAD is an important cofactor that
plays a central role in cellular redox reactions and serves as a second messenger for a
number of cellular process for survival. In mammalian cells, there are three main
pathways for NAD biosynthesis (Figure S1 in supporting information): a de novo
pathway synthesized from tryptophan (Trp), the primary salvage pathway using
nicotinamide (NAM) as the precursor, and the alternative salvage pathway dependent
1
7, 20
on nicotinic acid.
In humans, the main source of cellular NAD is from primary
salvage pathway using NAM, and nicotinamide phosphoribosyltransferase (NAMPT)
4
ACS Paragon Plus Environment

Page 5 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
is the rateꢀlimiting enzyme. Cancer cells have an increased demand for ATP and
consume NAD at a higher rate than normal tissues. Inhibition of NAMPT can
dramatically impact NAD metabolism and cancer proliferation. Thus, NAMPT has
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
7, 21
emerged as an attractive target for the development of anticancer therapeutics.
To
1
7, 20, 22
date, a number of novel and potent NAMPT inhibitors have been reported.
Among them, 2 (FK866, Figure 1B) and 3 (CHS828) have been progressed to human
23ꢀ25
clinical trials.
However, their clinical development of has been hampered by
26, 27
toxicity such as doseꢀlimiting thrombocytopenia.
NAMPT inhibitor 2 was proven to synergistically enhance the inhibitory effect of
28
HDACi, suggesting that dual NAMPT/HDAC inhibitors are likely to possess
promising antitumor profiles (Table S1 in Supporting Information). Moreover, the
structural features and pharmacophore of NAMPT and HDAC inhibitors (Figure 1)
offer an opportunity to design dual NAMPT/HDAC inhibitors. Herein, the first dual
NAMPT/HDAC inhibitors were rationally designed, synthesized and assayed.
Interestingly, a highly potent inhibitor with balanced activity against NAMPT/HDAC
and excellent in vivo antitumor potency was successfully identified.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 65
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 1. Pharmacophoric model and chemical structures of the NAMPT (A) and
HDAC (B) inhibitors. (C) Identification of 6 as a NAMPT/HDAC dual inhibitor.
RESULTS AND DISCUSSION
Chemistry. The synthetic routes of the target compounds are shown in Schemes 1-5.
Compounds 12a-t were prepared by the procedure detailed in Scheme 1. Reduction of
2
commercially available starting material 7 with Raney Ni and H led to intermediate 8.
4
ꢀAminobenzoic acid was treated with 1,1'ꢀthiocarbonyldiimidazole (TCDI) or
N,N'ꢀcarbonyldiimidazole (CDI) to give isothiocyanate or isocyanate 10, which was
condensed with intermediate 8 to yield thioureas or ureas (11a-t). Condensation of
compound 11a-t with oꢀphenylenediamine afforded the target compounds 12a-t.
6
ACS Paragon Plus Environment

Page 7 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 1
R2
CN
R2
a
NH2
R1
N
R1
N
O
8
7
O
c
X
OH
d
X
N
H
O
R
N
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
NH2
H
R
N
H
OH
OH
b
X
1
1a-t
12a-t
N
H2N
9
X = O,S 10
O
O
N
F
N
H
N
H
H
12o X = S, R =
12a X = S, R =
12h X = S, R =
12i X = S, R =
F
N
N
N
H2N
2p X = S, R =
N
H
N
H
N
H
1
1
2b X = S, R =
2c X = O, R =
Cl
N
N
Br
12j X = S, R =
N
H
N
N
H
12q X = O, R =
H
1
N
N
Cl
N
H
N
H
12k X = S, R =
12r X = O, R =
12d X = O, R =
F
N
F
N
N
Cl
F
N
H
N
H
12l X = S, R =
12e X = S, R =
12s X = O, R =
12t X = S, R =
H2N
H2N
Cl
N
N
H
N
12m X = S, R =
H
12f X = O, R =
HN
F
N
N
N
Cl
2n X = S, R =
N
H
N
H
1
12g X = O, R =
H
BocHN
N
Reagents and conditions: (a) Raney Ni, H
CDI or TCDI, CH Cl , rt, 2 h, yield 85ꢀ95%; (c) THF, rt, 12 h, yield 62ꢀ81%; (d)
oꢀphenylenediamine, HBTU, Et N, DMF, rt, 3 h, yield 43ꢀ75%.
2 3
, NH /MeOH, rt, 12 h, yield 21ꢀ89%; (b)
2
2
3
Replacement of core group with various binary heterocycleꢀderived amides and
introducing a methylene between the amide and phenyl linker moiety gave
compounds 16a-g (Scheme 2). Methyl 4ꢀ(aminomethyl) benzoate was condensed
with carboxylic acids to provide intermediates 14a-g. Then, treatment of compounds
1
4a-g with LiOH followed by coupling with oꢀphenylenediamine gave target
compounds 16a-g.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 65
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
Scheme 2
O
O
O
O
OH
O
a
H
N
b
H
R
R
N
H N
2
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
O
O
1
3
14a-g
15a-g
N
O
1
1
1
1
6a R =
6b R =
6c R =
6d R =
N
1
6e R =
N
N
c
N
H
H
R
N
NH2
H
N
N
N
H
N
O
1
6f R =
N
N
1
6a-g
N
N
H
16g R =
S
N
H
N
Reagents and conditions: (a) RꢀCOOH, EDC, DMAP, DMF, rt, 3 h, yield 31ꢀ69%;
b) LiOH, THF/MeOH/H O, rt, 24 h, yield 82ꢀ97%; (c) oꢀphenylenediamine, HBTU,
Et N, DMF, rt, 4 h, yield 37ꢀ75%.
(
2
3
Compounds 24 was synthesized according to the procedures outlined in Scheme 3.
Starting from 3ꢀiodopyridine (17), intermediate 19 was prepared via two steps
2
9
according to the reported methods. Methyl 4ꢀ(2ꢀbromoethyl)benzoate was reacted
with NaN to give azide 21, which was coupled with intermediate 19 by
3
copper(I)ꢀcatalyzed click reaction at room temperature to obtain 1,4ꢀdisubstituted
30
1
,2,3ꢀtriazole 22. Hydrolysis of the methylester in 22 gave the acid 23, which were
condensed with oꢀphenylenediamine to afford target compound 24.
8
ACS Paragon Plus Environment

Page 9 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 3
I
a
b
TMS
N
N
N
17
18
19
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
O
O
O
c
d
O
O
N
N
Br
N3
N
N
2
0
21
22
O
O
OH
N
e
f
H
NH2
N
N
N
N
N
N
N
N
2
4
23
Reagents and conditions: (a) Ethynyltrimethylsilane, Pd (PPh ) , CuI, Et N, rt, 1 h;
2
3 2
3
o
(
b) K
yield 85%; (d) 19, CuSO
LiOH, THF/MeOH/H O, rt, 24 h, yield 90%; (f) oꢀphenylenediamine, HBTU, Et
DMF, rt, 4 h, yield 27%.
2
CO
3
, MeOH, rt, 0.5 h, yield 81% over two steps; (c) NaN
·5H O, VcNa, N , tꢀbutanol/H O, rt, 12 h, yield 82%; (e)
N,
3
, DMF, 80 C, 8 h,
4
2
2
2
2
3
As shown in Scheme 4, 4ꢀethynylaniline (25) was reacted with CDI to produce
isocyanate 26, which was treated with 3ꢀ(aminomethyl)pyridine to give intermediate
27. Compound 28 was obtained by click reaction of methyl 4ꢀ(azidomethyl)benzoate
with 27. Hydrolysis of 28 gave the acid 29, which were condensed with
oꢀphenylenediamine to afford target compounds 30. Compound 35 was synthesized
according the procedure described in Scheme 5. Condensation of 4ꢀethynylbenzoic
acid (31) with 3ꢀ(aminomethyl)pyridine in the present of HBTU
(
3
oꢀbenzotriazoleꢀN,N,N,Nꢀtetramethylꢀuronium ꢀhexafluorophosphate) and Et N gave
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 65
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
intermediate 32. Finally, compounds 35 was prepared using the condition similar to
that of compounds 30.
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
Reagents and conditions: (a) CDI, THF, rt, 2 h; (b) 3ꢀ(aminomethyl)pyridine, THF,
rt, 12 h, yield 74% over two steps; (c) methyl 4ꢀ(azidomethyl)benzoate, CuSO
4 2
·5H O,
VcNa, N , tꢀbutanol/H O, rt, 12 h, yield 69%; (d) LiOH, THF/MeOH/H O, rt, 24 h,
2
2
2
yield 93%; (e) HBTU, Et N, DMF, rt, 3 h, yield 68%.
3
Reagents and conditions: (a) 3ꢀ(aminomethyl)pyridine, HBTU, Et
3
N, DMF, rt, 3 h,
O, VcNa, N
2
O, rt, 12 h, yield 67%; (c) LiOH, THF/MeOH/H O, rt, 24 h, yield 81%;
yield 45%; (b) methyl 4ꢀ(azidomethyl)benzoate, CuSO
4
·5H
2
2
,
tꢀbutanol/H
2
10
ACS Paragon Plus Environment

Page 11 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
3
d) oꢀphenylenediamine, HBTU, Et N, DMF, rt, 3 h, yield 50%.
Structure-based Design of Novel NAMPT/HDAC Dual Inhibitors. Generally,
NAMPT inhibitors are linear in shape and the pharmacophore consists of a core (or
cap) group as the NAM mimetic (typically a meta or para substituted pyridine), a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
0
linker and a relatively narrow and hydrophobic tail group (Figure 1C). Previously,
our group identified a series of novel NAMPT inhibitors by high throughput screening
22, 31, 32
and structureꢀbased design.
Among them, compound 4 (MS0) is a highly active
inhibitor containing typical pharmacophoric groups, namely a core binding moiety
(pyridinylmethylthiourea), a linker (phenyl group) and a tail group (piperidinesulfonyl
22, 33
group).
Interestingly, HDAC inhibitors share similar molecular shape and
structural characteristics to those of NAMPT inhibitors (Figure 1C). Briefly, typical
HDAC inhibitors contain a zincꢀbinding group (ZBG, hydroxamic acid or
34
orthoꢀaminoanilide), a linker and a hydrophobic cap group. Based on the structural
features of 4, we envisioned that NAMPT/HDAC dual inhibitors could be designed
35
by fusion of the pharmacophore of 4 with HDAC inhibitor 5 (CIꢀ994, Figure 1C).
To validate the hypothesis, a new compound (6) was designed by merging the core
group (pyridinylmethylthiourea) of NAMPT inhibitor 4 with the ZBG motif
(Nꢀ(2ꢀaminophenyl)benzamide) in HDAC inhibitor 5 using a common phenyl linker.
Because current docking methods can accurately reproduce the binding modes of
2
2, 36
NAMPT and HDAC inhibitors (Figure S2 in supporting information),
compound
6
was subjected for molecular docking to investigate whether it can form favorite
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 65
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
interactions with the two targets. After the docking process, molecular dynamics
simulations were performed on the docked complexes to further validate the stability
and affinity of the obtained binding poses using Desmond/Maestro nonꢀcommercial
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
37
version 2016.3 (Figure S3 in supporting information). As shown in Figure 2A,
compound 6 interacted with NAMPT by a linear conformation. Its pyridinyl group
formed faceꢀtoꢀface πꢀπ interactions with Phe193 and Tyr18′, respectively. Two NH
groups in thiourea formed hydrogen bonding interactions with the carboxyl group of
Asp219. The terminal Nꢀ(2ꢀaminophenyl)benzamide was located at the outside of
active site. The terminal NH group formed hydrogen bonding interaction with the
2
backbone carbonyl group of Val350. For the binding mode of 6 with HDAC1, main
interactions were involved in the linker and ZBG (Figure 2B). The phenyl linker and
terminal phenyl group formed πꢀπ stacking interaction with His141. Besides chelating
2
+
with Zn , the terminal Nꢀ(2ꢀaminophenyl)benzamide group was engaged with a
hydrogen bonding network with Gly149, His140 and Hisl41. The terminal NH group
2
formed two hydrogen bonds with the nitrogen atom in the imidazole of His140 and
Hisl41, respectively. The amide NH group formed hydrogen bonding interaction with
the backbone carbonyl group of Gly149. Based on the above analysis, compound 6
shares a similar binding mode with typical HDAC1 or NAMPT inhibitors. The
38
sequence identity between human HDAC1 and HDAC2 is 86%. Compound 6 was
also docked into the HDAC2 (PDB: 3MAX) active pocket. The binding mode of
compound 6 with HDAC2 was very similar to that with HDAC1 (detailed description
provided in Figure S4, supporting information). Inspired by the modeling results,
12
ACS Paragon Plus Environment

Page 13 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compound 6 was synthesized (Scheme S1 in Supporting Information) and assayed for
22,
inhibitory activity against NAMPT and HDAC1 using the method described before.
3
9
In the NAMPT inhibitory assay, human recombinant NAMPT was used to catalyze
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NAM to NMN. The product NMM was quantified by reacting with acetophenone and
formic acid, with the formation of fluorescent derivative detected with an excitation
of 382 nm and an emission wavelength of 445 nm. In the HDAC1 inhibitory assays,
substrate AcꢀLeuꢀGlyꢀLys(Ac)ꢀAMC was deacetylated by recombinant HDAC1 and
followed hydrolyzed by trypsin to release 7ꢀaminoꢀ4ꢀmethylcoumarin (AMC). AMC
was detected with an excitation of 350ꢀ360 nm and an emission wavelength of
4
50ꢀ460 nm. To our delight, compound 6 showed an IC50 value of 116 nM and 1.6
ꢀM against NAMPT and HDAC1, respectively (Figure 2C and 2D).
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 65
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 2. Identification of compound 6 as a NAMPT/HDAC dual inhibitor. (A)
Proposed binding mode of compound 6 in the active site of NAMPT (PDB ID: 2GVJ).
(B) Proposed binding mode of compound 6 in the active site of HDAC1 (PDB ID:
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
BKX). The figure was generated using PyMol (http://www.pymol.org/). (C)
Concentration response curve of 6 on NAMPT activity. (D) Concentration response
curve of 6 on HDAC1 activity. Data are shown as mean ± standard deviation.
NAMPT/HDAC Dual Inhibitor 6 Exhibits Potent In vitro and In vivo
Antitumor Activity. NAMPT/HDAC1 dual inhibitor 6 was further assayed for
growth inhibitory activities toward human cancer cellꢀlines HCT116 (colon cancer)
and HepG2 (liver cancer) using the MTT assay. Two lead compounds, NAMPT
inhibitor 4 and HDAC inhibitor 5, were used as reference drugs. As shown in Figure
3A, compound 6 showed good inhibitory activity against the HCT116 cell line (IC50
=
1
.0 ꢀM), which was more potent than 4 and 5. Although the antitumor activity against
HepG2 cell line was slightly decreased, compound 6 was still more active than the
positive drugs. Therefore, compound 6 was progressed to in vivo mouse
pharmacokinetic and xenograft efficacy studies. After single oral administration at a
dose of 25 mg/kg, compound 6 demonstrated acceptable pharmacokinetic properties
(Figure S6 and Table S2 in supporting information). The terminal halfꢀlife was
approximately 1.8 h and the Cmax was 2.3 ꢁM. The in vivo antitumor potency of 6 was
investigated in HCT116 xenograft nude mice models. After oral administration for 14
consecutive days, compound 6 achieved significant tumor growth inhibition (Figure
14
ACS Paragon Plus Environment

Page 15 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
B, P < 0.05) at the dose of 25 mg/kg/day. Moreover, it was observed to be well
tolerated during the test and no significant loss of body weight was observed,
indicating that its toxicity is low (Figure 3C).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. In vitro and in vivo antitumor efficacy of NAMPT/HDAC1 dual inhibitor 6.
(A) Enzyme inhibition and in vitro antitumor activity of compound 6. (B) In vivo
antitumor efficacy of compound 6 in the HCT116 tumor xenograft model. * P < 0.05,
versus the control group, determined with Student’s t test. (C) Body weight loss as a
result of compound 6 administration in HCT116 xenograft experiment.
Structural
Optimization
and
Structure-activity
Relationship
of
NAMPT/HDAC Dual Inhibitor 6. Although compound 6 was proven to be a
NAMPT/HDAC1 dual inhibitor and showed potent antitumor activity, its inhibitory
activity against HDAC1 was only moderate and its antitumor activity remained to be
further improved. Thus, further optimization studies were focused on achieving the
balanced activity toward both targets and improving the in vivo antitumor potency. On
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 65
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 basis of the binding mode of compound 6, classical pharmacophore for
NAMPT/HDAC dual inhibitors consists of a ZBG, a hydrophobic linker and a
hydrophobic core group (Figure 4). Orthoꢀaminoanilide was chosen as the ZBG
because it is one of the most wellknown group for chelating the zinc ion. Examination
of the binding mode of 6 with NAMPT (Figure 2A) suggested that linking the
orthoꢀaminoanilide as the tail group made the ZBG point to the solvent region and
was well tolerable. The NAM mimetic (i.e. substituted pyridine) served as a core
group, which intercalated at the site between Phe193 and Tyr18′ and formed πꢀπ
interactions. Because of the presence of large hydrophobic patches at the HDAC
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
15
surface rim, it was conceivable that the core group could form new interactions.
Different linker moieties contained in reported potent NAMPT or HDAC inhibitors
were analyzed and chosen as the hydrophobic linker group. Therefore, a series of
derivatives were designed and synthesized by optimization of the core group and
linker (Figure 4). Initially, the importance of pyridinyl core group was investigated by
introducing substitutions (12a-j), pyridinyl/phenyl replacement (12k-o) and
pyridine/heterocycle (12t, 16a-g) replacement, which was supposed to form various
interactions with NAMPT and HDAC. Then, the thiourea group was optimized by
urea (12q) or α,βꢀunsaturated amide (12r, 12s) to adjust the hydrogen bonding
interaction. Finally, 1,2,3ꢀtriazole was introduced to achieve excellent and balanced
dual inhibition (24, 30 and 35) because the correponding binding pocket in NAMPT
and HDAC could accomodate flexible linkers with different length. Most core and
linker groups were selected from highly potent HDAC or NAMPT inhibitors, which
16
ACS Paragon Plus Environment

Page 17 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
will be described in the following sections. Their in vitro data for inhibition of
NAMPT/HDAC1 as well as the cytotoxicity toward HCT116 (colon cancer), HepG2
(liver cancer) and MDAꢀMBꢀ231 (breast cancer) cells are listed in Tables 1-2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compounds 1 and 2 were used as the reference drugs.
Figure 4. Design strategy of the NAMPT/HDAC1 dual inhibitors.
Initially, various substitutions were introduced on the pyridinyl group (compounds
12aꢀj). These compounds generally showed improved inhibitory activity against
HDAC1 (IC50 range: 0.021 ꢀM ꢀ 0.37 ꢀM). However, their inhibitory activities
toward NAMPT were signicantly reduced (IC > 2 ꢀM). Furthermore, the pyridinyl
50
group was replaced by substituted phenyl groups (compounds 12kꢀp) and nitrogen
17, 33
containing heterocycles
(compounds 12t and 16a-g). Unfortunately, these
derivatives did not show increased NAMPT inhibitory activity, although their good
HDAC1 inhibitory activity was retained (Table 1).
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 65
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
To explain the SAR, compounds 12a and 12k were docked and superimposed with
compound 6 in the active site of NAMPT (Figure S5). The binding mode of 12a and
1
2k was very similar to that of compound 6. However, fluorine substituted pyridinyl
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
and phenyl group could not act as the NAM mimetic because the fluorine atom
occupied the position of pyrindine nitrogen atom of compound 6 (Figure S5B), which
might be the reason why the NAMPT inhibitory activity of these compounds was
decreased. Thus, it can be confirmed that the pyridine ring is essential for the NAMPT
inhibitory activity.
Table 1. Enzyme inhibition and in vitro antitumor activity of the target compounds
IC50 (µM)
Compds
12a
R
X
NAMPT
HDAC1
HCT116 MDA-MB-231
HepG2
S
>2.0
0.37 ± 0.031
10 ± 2.6
81 ± 7.6
5.2 ± 0.55
S
O
O
1
2b
>2.0
>2.0
>2.0
0.25 ± 0.013
0.021 ± 0.001
0.28 ± 0.015
6.1 ± 0.51
5.2 ± 0.47
14.0 ± 1.7
42 ± 3.7
34 ± 3.9
91 ± 8.3
19 ± 2.7
2.4 ± 0.19
13 ± 0.10
12c
12d
S
1
2e
>2.0
>2.0
0.20 ± 0.026
27 ± 2.3
>100
>100
14 ± 1.5
O
1
2f
0.33 ± 0.047
7.3 ± 0.80
3.1 ± 0.28
18
ACS Paragon Plus Environment

Page 19 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
S
S
S
1
2g
>2.0
>2.0
>2.0
>2.0
3.3 ± 0.28
0.44 ± 0.053
0.26 ± 0.022
0.35 ± 0.031
100 ± 12
15 ± 1.9
11 ± 2.1
>100
100 ± 9.3
13 ± 1.2
12h
85 ± 9.8
63 ± 5.9
39 ± 4.1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2i
11 ± 2.1
1
2j
9.2 ± 0.79
7.5 ± 0.63
S
S
S
S
S
12k
>2.0
>2.0
>2.0
>2.0
>2.0
0.30 ± 0.029
0.17 ± 0.011
0.30 ± 0.025
0.18 ± 0.021
0.10 ± 0.009
15 ± 1.9
5.9 ± 0.62
6.3 ± 0.54
5.9 ± 0.49
8.4 ± 0.77
>100
74 ± 8.0
>100
8.8 ± 7.5
2.9 ± 0.22
5.4 ± 0.56
2.6 ± 0.23
5.1 ± 0.49
1
2l
12m
12n
92 ± 8.9
>100
12o
S
1
2p
2q
>2.0
0.22 ± 0.035
0.22 ± 0.018
2.8 ± 0.19
9.4 ± 0.88
>100
>100
3.3 ± 0.30
5.1 ± 0.48
O
1
0.55 ± 0.047
O
O
12r
0.23 ± 0.029
>2.0
0.62 ± 0.057
0.67 ± 0.054
48 ± 5.0
>100
2.7 ± 0.35
1
2s
2.4 ± 0.22
13 ± 1.4
0.56 ± 0.034
S
O
O
1
2t
>2.0
>2.0
>2.0
0.046 ± 0.005
0.087 ± 0.009
0.071 ± 0.006
22 ± 2.0
1.8 ± 0.11
3.3 ± 0.39
>100
31 ± 4.3
5.5 ± 0.47
4.7 ± 0.53
16a
46 ± 4.4
41 ± 4.4
16b
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 65
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
O
O
O
1
6c
>2.0
>2.0
>2.0
0.084 ± 0.006
0.033 ± 0.002
0.093 ± 0.008
0.24 ± 0.021
4.0 ± 0.65
10 ± 2.0
48 ± 3.8
>100
1.7 ± 0.14
6.1 ± 0.55
3.4 ± 0.42
16d
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
16e
7.4 ± 0.66
54 ± 6.5
O
O
1
6f
>2.0
>2.0
18 ± 2.1
98 ± 9.1
40 ± 5.1
23 ± 0.30
16g
0.083 ± 0.009
0.026 ± 0.003
3.0 ± 0.26
0.38 ± 0.022
a
1
2
5
6
NT
3.1 ± 0.22
1.6 ± 0.12
10 ± 2.3
>100
4.1 ± 0.40
0.89 ± 0.074
10 ± 3.2
a
0.009 ± 0.0002
NT
1.3 ± 0.17
a
a
NT
1.2 ± 0.10
1.6 ± 0.18
NT
a
0.12 ± 0.02
1.0 ± 0.11
NT
4.2 ± 0.37
a
NT = not tested
Further optimzation efforts were focused on the thiourea group and phenyl linker.
When the thiourea group of lead compound 6 was replaced by urea (12q) or
α,βꢀunsaturated amide (12r), a banlanced NAMPT/HDAC1 inhibitory activity was
40
achieved. Due to the importance of 1,2,3ꢀtriazole in the NAMPT inhibitors, a series
of 1,2,3ꢀtriazole containing derivatives were further synthesized (24, 30 and 35).
Interestingly, the insertion of 1,2,3ꢀtriazole containing linker led to the increased
activity toward both enzymes (Table 2). In particular, compound 35 (NAMPT IC50
=
0
.031 ꢀM, HDAC1 IC50 = 0.055 ꢀM) performed the best for the balanced inhibition
of NAMPT and HDAC1. Considering that orthoꢀamino anilides were reported to be
38
slowꢀbinders, HDAC1 inhibitory activity was measured after 3 h incubation with
20
ACS Paragon Plus Environment

Page 21 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compounds 5 and 35. As a result, the IC50 value of compounds 5 and 35 against
HDAC1 was 60 nM and 24 nM, respectively.
Table 2. Enzyme inhibition and in vitro antitumor activity of target compounds
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
IC50 (µM)
Compds
R
n
NAMPT
HDAC1
HCT116 MDA-MB-231
HepG2
2
24
0.19 ± 0.021
0.18 ± 0.011
5.5 ± 0.43
41 ± 3.7
3.2 ± 0.33
1
1
30
0.018 ± 0.001
0.031 ± 0.009
0.041 ± 0.002
>100
>100
8.1 ± 0.79
4.5 ± 0.35
0.055 ± 0.004
0.026 ± 0.003
2.4 ± 0.22
10 ± 2.4
35
a
1
NT
3.1 ± 0.22
1.6 ± 0.12
10 ± 2.3
>100
4.1 ± 0.40
0.89 ± 0.074
10 ± 3.2
a
2
5
6
0.009 ± 0.0002
NT
1.3 ± 0.17
a
1.2 ± 0.10
1.6 ± 0.18
a
NT
NT
a
0.12 ± 0.02
1.0 ± 0.11
NT
4.2 ± 0.37
a
NT = not tested.
As shown in Figure 5A, the binding mode of compound 35 with NAMPT was very
similar to that of compounds 2 and 6. Its pyridinyl group formed faceꢀtoꢀface πꢀπ
interactions with Phe193 and Tyr18′, respectively. The amide carbonyl group formed
hydrogen bonding interactions with the hydroxyl group of Ser275. Moreover, the
terminal Nꢀ(2ꢀaminophenyl)benzamide projected out of the active site, whose
carbonyl group and terminal NH group formed hydrogen bonding interactions with
2
the side chain of Lys189 and the backbone carbonyl group of Asp184, respectively.
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 65
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 binding mode of compound 35 with HDAC1 was shown in Figure 5B. The
phenyl linker and terminal phenyl group formed πꢀπ stacking interaction with Phe150
2
+
and His141, respectively. Besides chelating with Zn , the terminal
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
Nꢀ(2ꢀaminophenyl)benzamide group was engaged with a hydrogen bonding network
2
with His140, Hisl41 and Gly149. The terminal NH group formed two hydrogen
bonds with the nitrogen atom in the imidazole of His140 and Hisl41, respectively. The
amide NH group formed hydrogen bonding interaction with the backbone carbonyl
group of Gly149. The NH group in head formed hydrogen bonding interaction with
the backbone carbonyl group of Glu98.
Figure 5. (A) Proposed binding mode of compound 35 in the active site of NAMPT
(PDB ID: 2GVJ). (B) Proposed binding mode of compound 35 in the active site of
HDAC1 (PDB ID: 4BKX). The figure was generated using PyMol
(http://www.pymol.org/).
In vitro antitumor activity assay revealed that most target compounds showed
moderate to good inhibitory activity against the tested three cancer cell lines. In
general, compounds are more effective against HCT116 and HepG2 cells than
22
ACS Paragon Plus Environment

Page 23 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
MDAꢀMBꢀ231 cells. Most compounds (12a-p, 12s-t, and 16a-g) seems to act mainly
through the inhibition of HDAC, and their in vitro antitumor activity was generally
consistent with the HDAC inhibitory activities. For example, compound 16g
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
NAMPT IC50 > 2 ꢀM, HDAC1 IC50 = 0.083 ꢀM) showed good activity against
HCT116 (IC50 = 3.0 ꢀM) and HepG2 (IC50 = 0.38 ꢀM) cell lines. For the
NAMPT/HDAC1 dual inhibitors, compound 35 showed potent activity against all the
tested cell lines particularly for HCT116 cells (IC50 = 2.4 ꢀM). Surprisingly,
compound 30 was only active against HepG2 cells despite its good activity against
both targets. The underlying reasons remain to be further investigated. Considering
the enzyme inhibition and antitumor activity, compound 35 was selected for further
evaluations.
In vitro HDAC Isoform Selectivity of Compound 35. To obtain evidence for the
HDAC isoform selectivity, the inhibitory activity of compound 35 was examined
against recombinant HDAC1, HDAC2, HDAC3, HDAC4, HDAC6 and HDAC8
(Table 3). As depicted in Table 3, compound 35 exhibited remarkable selectivity for
HDAC1 and HDAC2 over other isoforms. It showed nanomolar activity against
HDAC1 (IC50 = 55 nM) and HDAC2 (IC50 = 75 nM). In contrast, its activity against
HDAC3 (IC50 = 1867 nM) was significantly decreased. For other isoforms, such as
HDAC6 and HDAC8, compound 35 was totally inactive.
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 65
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
Table 3. HDAC profiling of compound 35 and 1
IC50 (nM)
HDAC class
1
35
HDAC1
HDAC2
HDAC3
HDAC4
HDAC6
HDAC8
I
I
I
IIa
IIb
I
26 ± 2.1
149 ± 8.5
51 ± 4.1
> 100 ꢁM
21 ± 1.4
7300 ± 300
55 ± 3.5
75 ± 7.4
1867 ± 252
> 100 ꢁM
> 100 ꢁM
> 100 ꢁM
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
Compound 35 Induces Cell Cycle Arrest, Apoptosis and Autophagy. To
investigate whether compound 35 can lead to cell death by apoptosis, an annexin
VFITC/propidium iodide (PI) binding assay was performed. Compound 5 was used as
a reference drug. HCT116 cells were treated with vehicle alone or tested compounds
at 5 ꢁM for 72 h, stained with Annexin VꢀFITC and PI, and analyzed by flow
cytometry. As shown in Figure 6A, compound 35 caused significant induction of
apoptosis in the HCT116 cell line. The percentage of apoptotic cells for compound 35
was 43.99% (Q2 + Q4), which displayed a higher apoptosis level than compounds 4
and 5 (40.51% and 31.56%, respectively). The result revealed that compound 35
could efficiently induce apoptosis of HCT116 cells and showed synergistic effect at
the cellular level.
Western blot assay was performed to determine the autophagic activity induced by
compound 35. HCT116 cells were exposed to compound 35 at 5 ꢁM for 24ꢀ72 h. An
equivalent volume of DMSO was added as the control. As shown in Figure 6B, a
significant intratumoral increase in LC3ꢀII level was observed after drug treatment for
2
4 h compared with the control. However, this kind of increase was followed by a
significant reduction at 48 h and 72 h of incubation, suggesting that compound 35
24
ACS Paragon Plus Environment

Page 25 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
induced activation of autophagy and increased autophagosome formation at 24 h after
drug treatment. To validate the result, Pꢀ62 was also measured by Western blot. As
depicted in Figure 6B, Pꢀ62 level was significant decreased at 24 h, 48 h and 72 h of
incubation, indicating that compound 35 induced Pꢀ62 degradation. Autophagy has
been considered as a cytoprotective mechanism, which also has the capacity to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
1
promote cell death. Both NAD synthesis inhibition and HDACi could induce
42, 43
autophagy in cancer cells.
However, they worked in different ways. Compound
41
2
ꢀinduced autophagy promoted cell death , while autophagy induced by 1 functioned
as a cytoprotective mechanism. Inhibition of 1ꢀinduced autophagy increased cell
44
apoptosis in glioblastoma stem cells. To illustrate the function of autophagy in
35ꢀinduced cell death, HCT116 cells were incubated with autophagy inhibitor
wortmannin for 1 h and subsequently treated with 35 for 72 h. Cell apoptosis was
assessed using Annexin VꢀFITC and PI staining and analyzed by flow cytometry. As
shown in Figure 6A, inhibition of autophagy by wortmannin dramatically decreased
35ꢀinduced cell apoptosis (43.99% vs 27.68%), suggesting that autophagy induced by
3
5 promoted cell death in HCT116 cells.
The effect of compound 35 on various phases of cell cycle progression was tested
in HCT116 cells (Figure 6C). HCT116 cells were treated with vehicle alone or
45
compound 35 at 1 and 5 ꢁM for 48 h, and evaluated by flow cytometric method. The
ratios in G2 phase of the cell cycle were 19% at 1 ꢁM and 39% at 5 ꢁM, respectively.
In contrast, the ratios of cells untreated in G2 phase of the cell cycle were 5%. In
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 65
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
comparison to the control population, the cell cycle data clearly showed compound 35
arrested HCT116 cells mainly at the G2 phase
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 6. Compound 35 induces apoptosis and autophagy, and arrests cell cycle
at G2 phase. (A) Cell apoptosis induced by compounds 4, 5 and 35. HCT116 cells
were treated with 4, 5 or 35 at 5 ꢁM in the presence or absence of wortmannin (0.25
ꢁ
M) for 72 h. Apoptosis was examined by flow cytometer. (B) Western blot analysis
and corresponding quantification of LC3ꢀII and Pꢀ62 form in untreated control cells
ctr) and HCT116 cells treated with 35 (5 ꢁM) at different time points. (C) Cell cycle
(
analysis after 48 h of treatment. HCT116 cells treated with 35 (at 1 and 5 ꢁM) for 48
h were assayed by flow cytometry after staining with PI.
26
ACS Paragon Plus Environment

Page 27 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound 35 Inhibits NAMPT and HDAC in HCT116 Cells. The effect of the
compound 35 on the acetylation level of histone 3 and histone 4 is shown in Figure
7
A. Human HCT116 cells were exposed to compound 35 at 1 ꢁM and 5 ꢁM for 24 h.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
An equivalent volume of DMSO was added as the control. The data showed that
compound 35 caused a dramatic increase in acetylꢀhistone H3 and acetylꢀhistone H4
in a dose dependent manner. As shown in Figure 7B, compound 35 decreased the
cellular NAD level effectively after incubation with human cell line HCT116 for 24 h.
To verify whether NAMPT is the direct binding target in intact cells, cellular thermal
2
2
shift assay (CETSA) was performed in HCT116 cells. Treated cells were exposed to
compound 35 for 4 h and were harvested. The cell lysates were collected, diluted and
heated at different temperatures, which were analyzed by sodium dodecyl sulfate
polyacrylamide gel electrophoresis (SDSꢀPAGE) followed by Western blot. NAMPT
protein levels from cells incubated with 35 were more stable as compared with the
control, indicating potent binding of 35 with NAMPT (Figure 7C). As shown in
Figure 7D, CESTA melt curves in intact cell for NAMPT target with 35 moved to the
o
right obviously, with the ꢂTm value of 1.58 C. Thus, NAMPT was confirmed as the
direct binding target of 35 in living cells
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 65
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. Inhibitory effects of compound 35 against NAMPT and HDAC in
HCT116 Cells. (A) Western blot analysis of acetylated histone H3 and acetylated
histone H4 after 24 h treatment with 35 (at 1 ꢁM and 5 ꢁM) in HCT116 cells. (B)
Concentration response curve of 35 on NAD level of HCT116 cells after 24 h
treatment. (C) Western bolt of cellular thermal shift assay in intact cell for NAMPT
with compound 35 (at 10 µM). (D) CESTA melt curves in intact cells for NAMPT
with 35 (at 10 µM).
In vivo Antitumor Efficacy and Preliminary PK study of Compound 35.
Compound 35 was chosen to evaluate the in vivo antitumor efficacy duo to its
excellent and balanced inhibitory activity toward both NAMPT and HDAC1. Initially,
PK studies of compounds 1, 2 and 35 in mice were performed, which were
administered intraperitoneally (ip) at 25 mg/kg twice a day (bid). Compound 35
produced high plasma concentration, which was much higher than reference
28
ACS Paragon Plus Environment

Page 29 of 65
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compounds 1 and 2 (Figure S7 and Table S3 in supporting information). Then,
HCT116 xenograft nude mouse model was prepared according the procedure
4
5
described previously. when implanted tumors had reached a volume of 100ꢀ300
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
mm , compound 35 was administered intraperitoneally (ip) at 25 mg/kg twice a day
for 21 consecutive days. Compounds 1 (25 mg/kg, ip, bid) and 2 (15 mg/kg, ip, bid)
11, 25
were used as the positive control because of their clinical experience.
Tumor
volumes and body weights were monitored every 3 days over the course of treatment.
Tumor growth inhibition (TGI) and relative increment ratio (TꢀC) were calculated at
46
the end of treatment. As shown in Figure 8A, treatment with compound 35 caused a
dramatic reduction in tumor growth. Moreover, compound 35 demonstrated much
higher antitumor activity (TGI = 69%, TꢃC = 31%) than the reference drug 1 (TGI =
3
3%, TꢃC = 67%) and 2 (TGI = 39%, TꢃC = 61%). The final tumor tissue size
visualized in Figure 8B also explicitly showed an excellent antineoplastic activity of
compound 35. These finding demonstrated that dual NAMPT/HDAC inhibitor led to
improved in vivo antitumor potency as compared with single HDAC or NAMPT
inhibitor. The PK profile of compound 35 was further investigated. It was
administered ip at 2 mg/kg in SpragueꢀDawley (SD) rats (Figure S8 in supporting
information). The terminal halfꢀlife of 35 was approximately 1.8 h. The peak
concentration Cmax, plasma clearance (Cl) and the volume of distribution absorption
(Vss) was 745 ng/mL, 6173 mL/h/kg and 6229 mL/kg, respectively (Table S2). The
results indicated that the PK properties of compound 35 remain to be further
improved to achieve better in vivo antitumor activities.
29
ACS Paragon Plus Environment