Discovery of Novel Allosteric Inhibitors of Deoxyhypusine Synthase

Article abstract of DOI:10.1021/acs.jmedchem.9b01979

Deoxyhypusine synthase (DHPS) utilizes spermidine and NAD as cofactors to incorporate a hypusine modification into the eukaryotic translation initiation factor 5A (eIF5A). Hypusine is essential for eIF5A activation, which, in turn, plays a key role in regulating protein translation of selected mRNA that are associated with the synthesis of oncoproteins, thereby enhancing tumor cell proliferation. Therefore, inhibition of DHPS is a promising therapeutic option for the treatment of cancer. To discover novel lead compounds that target DHPS, we conducted synthetic studies with a hit obtained via high-throughput screening. Optimization of the ring structures of the amide compound (2) led to bromobenzothiophene (11g) with potent inhibitory activity against DHPS. X-ray crystallographic analysis of 11g complexed with DHPS revealed a dramatic conformational change in DHPS, which suggests the presence of a novel allosteric site. These findings provide the basis for the development of novel therapy distinct from spermidine mimetic inhibitors.

Full text of DOI:10.1021/acs.jmedchem.9b01979

Subscriber access provided by Universiteit Utrecht
Article
Discovery of Novel Allosteric Inhibitors of Deoxyhypusine Synthase
Yuta Tanaka, Osamu Kurasawa, Akihiro Yokota, Michael G. Klein, Koji Ono, Bunnai Saito,
Shigemitsu Matsumoto, Masanori Okaniwa, Geza Ambrus-Aikelin, Daisuke Morishita, Satoshi
Kitazawa, Noriko Uchiyama, Kazumasa Ogawa, Hiromichi Kimura, and Shinichi Imamura
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01979 • Publication Date (Web): 06 Mar 2020
Downloaded from pubs.acs.org on March 9, 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 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of Novel Allosteric Inhibitors of
Deoxyhypusine Synthase
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
,
†
†
†
‡
†
Yuta Tanaka,* Osamu Kurasawa, Akihiro Yokota, Michael G. Klein, Koji Ono,
†
†
†
‡
Bunnai Saito, Shigemitsu Matsumoto, Masanori Okaniwa, Geza Ambrus-Aikelin,
†
†
†
†
Daisuke Morishita, Satoshi Kitazawa, Noriko Uchiyama, Kazumasa Ogawa,
†
†
Hiromichi Kimura, and Shinichi Imamura
†Pharmaceutical Research Division, Takeda Pharmaceutical Company Limited: 26-1,
Muraoka-Higashi 2-chome, Fujisawa, Kanagawa 251-8555, Japan
‡Department of Structural Biology, Takeda California, 10410 Science Center Drive, San
Diego, California 92121, United States
1
PDB ID codes: 1 with DHPS: 6P4V, 11g with DHPS: 6PGR
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 59
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: Deoxyhypusine synthase (DHPS) utilizes spermidine and NAD as
cofactors to incorporate a hypusine modification into the eukaryotic translation initiation
factor 5A (eIF5A). Hypusine is essential for eIF5A activation, which, in turn, plays a key
role in regulating protein translation of selected mRNA that are associated with the
synthesis of oncoproteins, thereby enhancing tumor cell proliferation. Therefore,
inhibition of DHPS is a promising therapeutic option for the treatment of cancer. To
discover novel lead compounds that target DHPS, we conducted synthetic studies with
a hit obtained via high-throughput screening. Optimization of the ring structures of the
amide compound (2) led to bromobenzothiophene (11g) with potent inhibitory activity
against DHPS. X-ray crystallographic analysis of 11g complexed with DHPS revealed a
dramatic conformational change in DHPS, which suggests the presence of a novel
allosteric site. These findings provide the basis for the development of novel therapy
distinct from spermidine mimetic inhibitors.
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
ACS Paragon Plus Environment

Page 3 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Eukaryotic translation initiation factor 5A (eIF5A) is an essential protein for cell growth
and survival as it plays a role in the elongation process of protein translation. During this
process, eIF5A is modified to contain the unique polyamine-derived amino acid,
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
1–4
hypusine [N -(4-amino-2-hydroxybutyl)lysine].
Hypusine is post-translationally
introduced into the ε-amino group of one specific lysine residue of eIF5A by two
enzymatic reactions (Figure 1).5,6 Deoxyhypusine synthase (DHPS) catalyzes the first
reaction by cleaving the polyamine spermidine and transferring its 4-aminobutyl moiety
7
to the lysine residue of the eIF5A precursor to form a deoxyhypusine intermediate. In
the second reaction, deoxyhypusine hydroxylase (DOHH) converts the
deoxyhypusine-containing intermediate to the hypusine-containing mature eIF5A. This
hypusination of eIF5A is critical not only for the proliferation of eukaryotic cells, but also
for the regulation of protein translation of selected mRNA that are associated with the
synthesis of oncoproteins, and thus enhances the proliferation of tumor cells.4,8 In
addition, a recent publication has reported that eIF5A is involved in cellular respiration
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 59
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
and macrophage function by regulating mitochondrial proteins, which can be linked to
9
cancer cell metabolism. Therefore, blocking the biological activity of eIF5A by inhibiting
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 DHPS-mediated hypusine modification of the eIF5A precursor protein may be a
useful strategy that can be employed in the treatment of tumorigenic diseases.
Figure 1. DHPS-eIF5A pathway as a therapeutic target for cancer.
Polyamine analogues of the natural DHPS substrate spermidine have been explored
to examine their effect on DHPS activity, which led to the discovery of
N-1-guanyl-1,7-diaminoheptane (1) (GC-7) as a potent DHPS inhibitor (Figure 2A).10
Some other spermidine mimetic inhibitors were also reported as well as utilization of in
silico approach to design DHPS inhibitors.11–14 Co-crystal structure of 1 with DHPS
4
ACS Paragon Plus Environment

Page 5 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
determined that 1 binds to the spermidine binding site.16,17 However, the applications of
1
as a tool molecule to investigate the biology of DHPS are limited, due to its
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
nonselective binding properties and high structural similarity to spermidine. This may
cause undesired off-target pharmacology such as modulation of spermidine
biosynthesis and metabolism pathways.18,19 There is no DHPS inhibitor evaluated in
clinical study because a strategy to design molecules other than spermidine mimetic
inhibitors has been limited due to the lack of understanding of druggable pockets in the
protein. Therefore, identification of new tool molecules possessing drug-like properties
is needed to fully clarify the functions of DHPS.
(A)
NH2
N
N
HO
N
OH
O
N
OH
P
O
P
N
O
O
O
O
NH
O
OH
NH2
OH OH
2
H N
N
H
H2N
N
NH2
NH2
H
NAD+
O
Spermidine
1 (GC-7)
DHPS inhibitor
Coenzyme for deoxyhypusination
Substrate
(B)
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 59
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
(
C)
+
Figure 2. (A) NAD , spermidine, and GC-7 structures. (B) Structure of DHPS in complex
with NAD and GC-7 at a resolution of 1.65 Å (PDB ID 6P4V). The protein is colored by
chain and shown as a cartoon. NAD and GC-7 are shown in sticks. (C) Closeup view
6
ACS Paragon Plus Environment

Page 7 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
highlighting the hydrogen bonds (dash lines) between GC-7 and DHPS/NAD. Residues
possessing interactions with GC-7 are shown in sticks. For clarity, a part of protein is
omitted. The all images were prepared using PyMOL.15
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
Here, we describe the discovery of potent DHPS inhibitors from a hit generated by a
high throughput screening campaign. Furthermore, X-ray crystallographic analysis of a
representative compound bromobenzothiophene (11g) in complex with DHPS reveals a
dramatic conformational change in DHPS, suggesting the presence of a novel allosteric
site. These findings open the door toward a brand-new paradigm of allosteric DHPS
inhibitors with therapeutic potentials as an anti-cancer drug.
RESULTS AND DISCUSSION
X-ray Structure of DHPS with GC-7 and Profile of Hit Compound 2.
DHPS utilizes spermidine and NAD as cofactors to incorporate a 4-aminobutyl moiety
onto the Lys50 side chain of eIF5A.1,20 Initial in-house crystallography trials identified a
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 59
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
.65 Å tetrameric structure of the DHPS ternary complex containing the cofactor NAD
and 1 (Figure 2B, C), similar to a previously published 3.0 Å complex containing the
same ligands.17 Superposition of our 1.65 Å refined model with the published 3.0 Å
structure (PDB code 1RQD) revealed a slight discrepancy in the binding pose of 1
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
(Supporting Information Figure S1). These data suggest that 1 is located proximal to
NAD from a distinct monomer, with both ligands at the interface formed between
crystallographically related DHPS molecules that pack into a larger tetrameric assembly.
Hydrogens of the guanidine and amine groups of 1 interact with Glu323, Gly314, and
Asn292 on one monomer and Asp243 and NAD on another monomer.
Furthermore, enzymatic inhibitory activity of 1 is not affected by the two different
concentrations of NAD (14, 250 μM), suggesting that 1 and NAD molecules are not
competing each other (Figure 3).21 In contrast, compound 2, which is a hit identified
from the high throughput screening campaign, showed a NAD competitive profile;
decreased potency was observed at higher concentrations of NAD. In addition, 2
exhibited a weak spermidine competitive profile (data not shown). These results
8
ACS Paragon Plus Environment

Page 9 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
propelled studies to further optimize this molecule with the hope of discovering a novel
compound distinct from spermidine mimetic inhibitors.
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
O
NH
N
H
NH
H N
2
N
NH2
H
1
2 (HTS hit)
DHPS IC50 (M)
DHPS high NAD IC50 (M)
0.14 (0.11–0.17)
0.050 (0.043–0.055)
0.84 (0.76–0.92)
38 (24–61)
Figure 3. Inhibitory activity of 1 and HTS hit compound 2. Inhibitory activity was
determined by an enzyme assay in the presence of 14 or 250 μM of NAD. IC values
50
and 95% confidence intervals (CI) were calculated by nonlinear regression analysis of
percent inhibition data.
Identification of the Potent Inhibitor 11g by a Structure-Activity Relationship (SAR)
study.
To identify potent inhibitors, we conducted ligand-based SAR studies, given that
co-crystallization trials of 2 with DHPS failed. The inhibitory activity of derivatives
possessing various linkers and scaffolds are depicted in Table 1. Conversion of the
amide linker to sulfonamide (5a), urea (5b), and reverse amide (7) significantly
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 59
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
decreased potency. Regarding modification of the indole scaffold, indoline (9a) was not
tolerated, while indolinone (9b) exhibited inhibitory activity (IC50 = 42 μM), suggesting
that the acidic proton of 2 is preferable to enhance potency. In addition, orientation of
the proton appears to be necessary for the function of 2 based on the IC50 values of
pyrazole derivatives 9c and 9d (IC = 22 μM and >100 μM, respectively). Overall, the
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
50
amide linker and indole scaffold derived from compound 2 were the best right-hand
parts used for further optimization.
Table 1. SAR for Linkers and Scaffolds 2, 5a–b, 7, and 9a–d
linker
A
A
A
DHPS


compound linker
A
A
A
a
IC (μM)
5
0
0.84
2

O


(0.76–

N
NH
H
(HTS hit)
0.92)
10
ACS Paragon Plus Environment

Page 11 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9


O O
S



5
a
b
>100
>100

N
NH
NH
H
O


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
N
H
N
H
5
99

H
N


7
9
9

NH
NH
O
(68–144)
>100


O
O




a
b


N
H
42
NH
O
N
H
(30–59)
22

O
O




9c


N
H
NH
N
(9.8–50)
>100

9d
N
H
HN
N
a
Determined by enzyme assay. IC values and 95% confidence intervals (CI) were
50
calculated by nonlinear regression analysis of percent inhibition data.
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 59
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
Next, we attempted to convert the left-hand side of the molecule (Table 2), while
maintaining the 4-aminocarbornyl indole moiety. Since removal of the tert-butyl group
on the phenyl group of 2 decreased inhibitory activity (11a, IC >100 μM), we focused
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
50
on examining three types of bicyclic substituents: 2-naphtyl group (11b), 2-benzofuranyl
group (11c), and 2-benzothiophenyl group (12a). Among these, the benzothiophene
derivative 12a exhibited comparable potency to compound 2. Additionally, we
introduced a tert-butyl group to the benzothiophene ring to enhance the inhibitory
activity in a similar fashion as 2 from 11a, and confirmed that substitution at the
6'-position of the ring is tolerable. Although carboxamide derivative 12b showed
decreased potency, exploration of other substituents at this position (11e–g and 12c)
revealed that halogen substituents significantly improve DHPS inhibition. The most
potent compound, 6'-bromo derivative 11g, had an IC value of 0.062 μM. Conversely,
50
introduction of a bromo group at the 5'- (11h) or 7'-position (11i) was unfavorable,
indicating that the 6'-bromo substituent of 11g is necessary for its binding to DHPS.
12
ACS Paragon Plus Environment

Page 13 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Through the SAR studies, we discovered a potent non-spermidine mimetic DHPS
inhibitor 11g.
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 2. SAR for Left Hand Site 11a–i and 12a–c
O
4
R
N
NH
H
DHPS
compound R
a
IC (μM)
5
0
2
0.84

(
HTS hit)
(0.76–0.92)
>100
11a
phenyl
1
1b
1c
2-naphthyl
>100
O

1
>100
2.3
S

12a
(1.5–3.6)
S

11d
6'
5.4
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 59
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
(
2.0–15)
0
.43
S

11e
12b
11f
NC
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
(0.24–0.77)
1
1
S

O
H2N
(5.8–20)
0
.12
S

Cl
(0.093–0.17)
0.062
S

1
1g
2c
Br
(0.045–0.084)
0.092
S

1
I
(0.069–0.12)
S
S

11h
>100
>100
5'
Br
Br

7'
11i
aRefer to Table 1.
14
ACS Paragon Plus Environment

Page 15 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
X-ray Co-crystal Structural Analysis of DHPS with 11g.
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 determine the binding mode of the newly synthesized inhibitors, we solved the
X-ray co-crystal structure of compound 11g with DHPS protein at 1.95 Å resolution, as
shown in Figure 4. From the results of structural analysis, a tetramer of the binary
complex was identified similar to the structure of the ternary complex of 1 and NAD
(
Figure 4A), and 11g was bound to a novel allosteric pocket (vide infra). This cocrystal
structure well explained the results from the SAR studies (Figure 4B, C). Regarding the
indole moiety, NH-π interaction between a side chain of His288 and the indole ring, and
a hydrogen bond between Asn267 and NH of the indole were identified, confirming that
it is required for potency. The amide moiety of 11g formed a hydrogen bond network, (i)
between its carbonyl oxygen and the residue Lys329 which receives the butylamine
moiety of spermidine in the enzymatic reaction,22 and (ii) between NH proton on the
amide and Asp238 via
a
water molecule. As mentioned above, the
6
-bromobenzothiophene moiety is important for enhancing inhibitory activity. It is
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 59
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
suggested that the fused ring needs to be positioned at an angle suitable for binding to
the protein and does so by forming an intramolecular S-O interaction.23 In addition, a
unique halogen bonding between the 6-bromo group and the main chain carbonyl
oxygen of Leu281 was confirmed. This interaction occurs because the extension on the
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
24
Br side of the C-Br bond has a slightly positive charge. Although the bromo group as a
whole is a negatively charged moiety, only the site of the Br atom has the role of a
25
“
donor of hydrogen bond,” enabling the interaction with the carbonyl oxygen. The data
of the chlorine (11f) and iodine (12c) derivatives (IC50 = 0.12 μM and 0.092 μM,
respectively, Table 2) also support the presence of an effect as a result of halogen
bonding.
(A)
16
ACS Paragon Plus Environment

Page 17 of 59
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
(B)
(C)
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 59
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


Leu281
Lys329

-
Br
S
O
N
His288
H
Asn267
H O
2
N
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
H
Asp238
11g
DHPS IC50 (M) 0.062 (0.045–0.084)
DHPS IC50 at high NAD (M) 56 (9.7–325)
DHPS IC50 at high spermidine (M) 6.9 (2.8–17)
Figure 4. Co-crystal structure of compound 11g with DHPS (PDB ID 6PGR). (A)
Tetramer of the structure, with each monomer indicated by a different color. 11g is
shown as spheres. (B) The allosteric binding site of 11g. DHPS is shown at the surface.
1
1g and a part of the surrounding residues are shown in sticks. The colors are the same
as in (A). 11g has hydrogen bonding interactions with Asn267, Lys329, and, moreover,
with Asp238 via a crystal water molecule, shown as dash lines with the distances. 11g
also has an important interaction with His288 in NH–π interaction. For reference, the
distance between the nitrogen atom of His288 and a carbon atom of the indole ring is
shown. (C) Schematic diagram of interactions in stabilizing 11g at the allosteric site. The
inhibitory activity under normal conditions was determined by enzyme assay at 14 μM of
NAD and 2 μM of spermidine. NAD (250 μM) or spermidine (100 μM) were used at each
18
ACS Paragon Plus Environment

Page 19 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
condition of higher concentration, respectively. IC50 values and 95% confidence
intervals (CI) were calculated by nonlinear regression analysis of percent inhibition data.
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
Observations on Dynamic Conformational Change of DHPS
To elucidate the mechanism of inhibition, profiling of 11g inhibitory activity in the
presence of high concentrations of NAD (250 μM) and spermidine (100 μM) were
performed (Figure 4C). Compound 11g was less potent both at the higher
concentrations of NAD and spermidine relative to that at normal concentrations (14 μM
of NAD and 2 μM of spermidine), indicating that 11g competitively occupies the binding
sites of both NAD and spermidine.
However, 11g binds to the allosteric pocket distinct from NAD and the spermidine
binding pockets as shown in Figure 5. Comparing the structural analyses of the ternary
(with 1 and NAD) and binary (with 11g) complexes, binding of 11g appeared to induce
dynamic conformational changes in the secondary structure of the protein in the vicinity
of the allosteric pocket. Strikingly, the α-helix in the co-crystal structure with 1 and NAD
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 59
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
is unfolded and forms a loop structure in the 11g-bound structure. Furthermore, Gly282
on the unfolded loop partially occupies the NAD binding pocket, hindering the binding of
NAD (Figure 5B). In addition, Lys287 in the α-helix in the ternary (with 1 and NAD)
complex dramatically moves and overlaps the spermidine binding site, indicating an
interference of the binding of spermidine by 11g (Figure 5C). Therefore, the drastic
change in the secondary structure of the protein may be accompanied by the
interference of both NAD and spermidine binding pockets, and thus 11g competed with
NAD and spermidine regardless of its allosteric binding characteristics.
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
(A)
20
ACS Paragon Plus Environment

Page 21 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
B)
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
(C)
Figure 5. Superposition of the co-crystal structures of 11g-bound and GC-7-bound
forms (PDB IDs 6P4V and 6PGR). The 11g-bound protein, the GC-7-bound protein, and
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 59
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
1g molecule are shown in green, cyan, and purple, respectively. (A) Comparison of the
1g-bound and GC-7-bound forms. Ligand binding causes unfolding of an α-helix,
1
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
specifically, a flip of Gly282 located near the NAD binding site and movement of Lys287
and the side chain of Lys329, which receives the butylamine moiety of spermidine in the
enzymatic reaction. (B) A closeup view highlighting the NAD binding site. NAD in the
GC-7 bound form is shown as a white stick. The steric clash is possible between flipped
Gly282 in the 11g-bound form and NAD in the GC-7-bound form. The closest distance
between them (the oxygen atom in Gly282 and the 1' carbon in adenosine substructure
in NAD) is 1.6 Å. (C) Closeup view highlighting the spermidine binding site. GC-7 is
shown as the white stick. Lys329 in 11g-bound form is pointed away from spermidine
binding site, and Lys287 in the α-helix in GC-7 bound form dramatically moves and
overlaps the spermidine binding site instead.
The 11g binding to the allosteric pocket unfolds the α-helix to its original state, and
induces a steric clash with NAD by flipping Gly282, and spermidine by the movements
22
ACS Paragon Plus Environment

Page 23 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of Lys287 simultaneously. Although further investigations are required, our data implied
that DHPS protein exists in the equilibrium of at least two confirmations (α-helix folded
vs unfolded), which would regulate enzymatic activities (Figure 6). As the α-helix
unfolded conformation does not accommodate a cofactor NAD molecule, it is inactive
for enzymatic reaction. Binding of 11g to the DHPS protein plausibly shifts the
equilibrium to the inactive conformation allosterically. This may explain why 11g binds to
the allosteric pocket and exhibits DHPS inhibitory activity in a spermidine/NAD
competitive manner. Although additional mechanistic studies under physiological
conditions in cellular settings will be necessary, our allosteric modulator 11g sheds light
on the molecular mechanism of DHPS function.
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
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 59
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 6. Schematic diagram of conformational change by 11g binding in one dimer of a
tetramer of DHPS. The binding sites of spermidines are shown in the rectangle and
NAD as the rectangle with rounded corners. The 11g binding causes the unfolding of
the α-helix in the original state, and introduces steric clash with NAD by flips of Gly282
and spermidine by the movements of Lys287.
CHEMISTRY
24
ACS Paragon Plus Environment

Page 25 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 1 and Scheme 2 depict the synthesis of 4-tert-butylphenyl derivatives. Indole
derivatives 5a–b and 7 were synthesized as illustrated in Scheme 1. The treatment of
commercially available amine 3 with 4-tert-butylbenzenesulfonyl chloride or
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-tert-butylphenyl isocyanate followed by the removal of the Boc group under acidic
conditions generated 5a and 5b, respectively. Compound 7 was prepared by the
condensation of carboxylic acid 6 with 4-tert-butylaniline. 4-tert-Butylbenzamide
derivatives 9a–d with a fused ring were synthesized by the treatment of
4-tert-butylbenzoyl chloride with the corresponding amines 8a–d.
Scheme 1. Synthesis of Indole Derivatives 5a, b and 7a
a (for 4a)
c
X
X
N
N
b (for 4b)
NBoc
NH
H N
2
H
H
NBoc
3
4a, X = SO (89%)
5a, X = SO (90%)
2
2
4b, X = NHCO (72%)
5b, X = NHCO (72%)
H
N
d
HO
NH
NH
O
O
6
7
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 59
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
a
Reagents and conditions: (a) 4-tert-butylbenzenesulfonyl chloride, pyridine, 0 °C; (b)
4
-tert-butylphenyl isocyanate, THF; (c) TFA, MeCN; (d) 4-tert-butylaniline, EDC-HCl,
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
HOBt-H O, Et N, DMF, 64%.
2
3
26
ACS Paragon Plus Environment

Page 27 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 2. Synthesis of Compounds 9a–d a
compd
R
yield (%)
2
8a, 9a
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
O

NH
R
R
a
N
H N
2
H
8b, 9b

NH 53
O
8a–d
9a–d
8c, 9c
28

NH
N
8d, 9d
71

HN N
a
Reagents and conditions: (a) 4-tert-butylbenzoyl chloride, pyridine, 0 ºC.
The syntheses of compounds 11a–i and 12a–c were performed as shown in Scheme
3. Amidation of 3 with benzoyl chloride produced compound 11a, and amide derivatives
11b–i were similarly synthesized using acid chlorides prepared from carboxylic acids
10b–i and oxalyl chloride. 10d was synthesized from 6-bromobenzothiophene 13 via a
nickel-catalyzed Kumada coupling reaction26 and subsequent hydrolysis of the ethyl
ester. Reductive debromination of 6-bromobezothiopene derivative 11g produced
compound 12a. The 6-carboxamide derivative 12b was prepared by hydrolysis of
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 59
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
cyanide 11e. Copper-catalyzed iodination of 11g in the presence of NaI and
N,N’-dimethylethylenediamine at 120 °C in NMP afforded iodide derivative 12c.
Scheme 3. Synthesis of Compounds 11a–i and 12a–ca
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
c (from 11g to 12a)
d (from 11e to 12b)
e (from 11g to 12c)
O
a (for 11a)
O
S
N
R1
NH
4
H N
2
b (for 11b-i)
N
NBoc
NH
H
H
R2 6
1
R -CO H
2
3
10b-i
11a-i
R1
12a, R = H
2
2
12b, R = CONH
2
2
R1 yield (%) compd
12c, R = I
compd
1a
yield (%)
S


1
48
72
10f, 11f
59
77
Cl
Br

S

10b, 11b
10g, 11g
S

O

1
0c, 11c
62
10h, 11h
10i, 11i
72
69
Br
Br
S

S

10d, 11d
7
S

10e, 11e
70
NC
O
O
S
S
f
O
OH
Cy
N
Br
6
BF4
13
10d
N
14
Cy
a
Reagents and conditions: (a) (i) benzoyl chloride, THF; (ii) TFA, toluene; (b) (i) 10b–i,
oxalyl chloride, DMF, THF; (ii) TFA, toluene; (c) 10% Pd/C, H , EtOH, THF, 68%; (d)
2
K CO , 35% H O , DMSO, 94%; (e) CuI, NaI, N,N’-dimethylethylenediamine, NMP, 120
2
3
2
2
28
ACS Paragon Plus Environment

Page 29 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
°
C, 18%; (f) (i) 14, tert-butyl magnesium chloride, nickel(II) chloride hexahydrate, THF,
−
15 °C; (ii) 2 M NaOH, EtOH, 2%.
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
CONCLUSIONS
We discovered a novel allosteric DHPS inhibitor 11g with potent inhibitory activity
against DHPS (IC = 0.062 μM). The co-crystal structure of 11g with DHPS clarified its
50
binding mode, which elucidated our SAR studies. The newly found structure of DHPS
bearing 11g suggested the existence of the inactive conformation of DHPS where
cofactor NAD is not accommodated. 11g is a unique molecule that binds to the allosteric
site and inhibits DHPS in a spermidine/NAD competitive manner.
Although the
therapeutic potential of this molecule and analogs need to be further evaluated, these
findings provide a new foundation for a structure-based design of allosteric inhibitors
targeting DHPS, and provides a better understanding of the molecular mechanisms of
the hypusine synthesis pathway.
29
ACS Paragon Plus Environment