Novel Propargyl-Linked Bisubstrate Analogues as Tight-Binding Inhibitors for Nicotinamide N-Methyltransferase

Article abstract of DOI:10.1021/acs.jmedchem.9b01255

Nicotinamide N-methyltransferase (NNMT) catalyzes the methyl transfer from the cofactor S-adenosylmethionine to nicotinamide and other pyridine-containing compounds. NNMT is an important regulator for nicotinamide metabolism and methylation potential. Aberrant expression levels of NNMT have been implicated in cancer, metabolic, and neurodegenerative diseases, which makes NNMT a potential therapeutic target. Therefore, potent and selective NNMT inhibitors can serve as valuable tools to investigate the roles of NNMT in its mediated diseases. Here, we applied a rational strategy to design and synthesize the tight-binding bisubstrate inhibitor LL320 through a novel propargyl linker. LL320 demonstrates a K_i value of 1.6 ± 0.3 nM, which is the most potent inhibitor to date. The cocrystal structure of LL320 confirms its interaction with both the substrate and cofactor binding sites on NNMT. Importantly, this is the first example of using the propargyl linker to construct potent methyltransferase inhibitors, which will expand our understanding of the transition state of methyl transfer.

Full text of DOI:10.1021/acs.jmedchem.9b01255

Subscriber access provided by La Trobe University Library
Article
Novel Propargyl-linked Bisubstrate Analogs as Tight-
binding Inhibitors for Nicotinamide N-Methyltransferase
Dongxing Chen, Linjie Li, Krystal Diaz, Iredia David Iyamu, Ravi Yadav, Nicholas Noinaj, and Rong Huang
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01255 • Publication Date (Web): 14 Nov 2019
Downloaded from pubs.acs.org on November 15, 2019
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 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Novel Propargyl-linked Bisubstrate Analogs as
Tight-binding Inhibitors for Nicotinamide N-
Methyltransferase
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
†
, §
†, §
†
†
‡
‡
Dongxing Chen , Linjie Li , Krystal Diaz , Iredia D. Iyamu , Ravi Yadav , Nicholas Noinaj ,
and Rong Huang*, †
†Department of Medicinal Chemistry and Molecular Pharmacology, Center for Cancer
Research, Institute for Drug Discovery, Purdue University, West Lafayette, Indiana 47907,
United States.
‡Markey Center for Structural Biology, Department of Biological Sciences and
the Purdue Institute
of
Inflammation,
Immunology
and
Infectious
Disease, Purdue University, West Lafayette, Indiana 47907, United States.
§These authors contributed equally.
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 58
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. Nicotinamide N-methyltransferase (NNMT) catalyzes the methyl transfer
from cofactor S-adenosylmethionine (SAM) to nicotinamide and other pyridine-
containing compounds. NNMT is an important regulator for nicotinamide metabolism
and methylation potential. Aberrant expression levels of NNMT have been implicated in
cancer, metabolic and neurodegenerative diseases, which makes NNMT a potential
therapeutic target. Therefore, potent and selective NNMT inhibitors can serve as valuable
tools to investigate the roles of NNMT in its mediated diseases. Here, we applied a rational
strategy to design and synthesize the tight-binding bisubstrate inhibitor LL320 through a
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
novel propargyl linker. LL320 demonstrates a K value of 1.6±0.3 nM, which is the most
i
potent inhibitor to date. The co-crystal structure of LL320 confirms its interaction with
both the substrate and cofactor binding sites on NNMT. Importantly, this is the first
example of using propargyl linker to construct potent methyltransferase inhibitors, which
will expand our understanding of the transition state of methyl transfer.
INTRODUCTION
Nicotinamide N-methyltransferase (NNMT), a cytosolic enzyme that is
predominantly expressed in liver and adipose tissue, catalyzes the transfer of a methyl
group from cofactor S-adenosyl-L-methionine (SAM) to nicotinamide (vitamin B3) to
1
produce S-adenosyl-L-homocysteine (SAH) and N1-methylnicotinamide. NNMT can also
ACS Paragon Plus Environment
2

Page 3 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
methylate pyridines, quinoline, and other structurally related heterocyclic compounds. Its
+
substrate, nicotinamide, is a precursor for NAD , an important cofactor involved in redox
3
metabolism and energy homeostasis. NNMT is also implicated in methyl donor balance
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
since its expression level is inversely proportional to the SAM/SAH ratio both in vivo and
4
+
in vitro. Because of its dual involvement in the metabolism of both the NAD precursor
and the methyl donor SAM, NNMT is an important regulator for nicotinamide metabolism
and methylation potential. Therefore, NNMT has been linked to various diseases including
cancers, neurodegenerative diseases and obesity and diabetes.3,5–8 Increased expression
of NNMT enhances cell proliferation and disease progression in several cancers including
colorectal, gastric, breast, renal, lung, and ovarian cancers.5,7,9,10 Knockdown of NNMT by
siRNA or shRNA inhibits the cancer cell growth and induces apoptosis.11 In addition,
NNMT is up-regulated in Parkinson’s disease, atherosclerosis, obesity and Type 2
diabetes.3,12,13 Hence, NNMT has drawn emerging attention as a potential target for the
aforementioned diseases.
Potent and selective NNMT inhibitors would be valuable tools to investigate the
role of its catalytic activity in its physiological and pharmacological roles. However, current
reported NNMT inhibitors only have modest activity at µM levels (Figure 1). 5-amino-1-
MQ is a representative example that mimics product and binds to the substrate binding
site.14 5-amino-1-MQ can rejuvenate muscle stem cells to improve the regenerative
capacity of aged skeletal muscle. Compound JBSNF000088 acts as a competitive but slow-
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 58
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
5
turnover substrate analog. JBSNF000088 has been demonstrated to reduce MNA levels
15
and drives insulin sensitization in preclinical animal models of obesity. SS3 is a suicide
inhibitor and promotes the modification of Cys159 upon methylation.16 The inhibitor
HS312, containing an alpha-chloroacetamide group, inhibits NNMT by covalently
inhibiting Cys165 that is located in the SAM/SAH binding site, but loses its engagement
to NNMT in cells due to its reactivity with other cysteines in situ.17
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
Bisubstrate analogs have the potential to offer high potency and selectivity,
especially for small molecule methyltransferases like NNMT. To achieve such potential,
an optimal linker is critical to orient both substrate and SAM cofactor analogs to retain
their respective interactions to the target. Three previously reported bisubstrate analogs
VH45, MS2756 and MS2734 demonstrated moderate activities at µM range.18,19 During
the development of potent bisubstrate inhibitor for protein N-terminal methyltransferase
1
(NTMT1), a 3-C atom linker exhibited its advantage to link the substrate and SAM analog
20
to offer potent and selective inhibition at the nM level. Since the distance between the
sulfur atom and nucleophilic N atom is about 4 Å in both NNMT and NTMT1,21,22 we
hypothesize that a 3-C atom linker may serve as an optimal linker to produce NNMT
bisubstrate analogs with high potency. Here, we designed and synthesized a series of
bisubstrate analogs with variable 3-C atom linkers and a 4-C atom linker to investigate
the optimal linker length. Top inhibitor LL320 with a propargyl linker exhibits a K value at
i
1.6±0.3 nM, which is the most potent inhibitor for NNMT to date. In addition, we obtained
ACS Paragon Plus Environment
4

Page 5 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
co-crystal structures of both LL319 (1a) and LL320 (2a) in complex with human NNMT
(
hNNMT), which clearly delineate the occupancies of bisubstrate analogs at both
nicotinamide and SAM binding sites. Here, we reported the first case that propargyl linker
is able to construct potent and tight bisubstrate inhibitors for NNMT. The propargyl linker
expands the linker pool to devise bisubstrate analogs for methyltransferases and provides
a valuable approach to develop potent and selective 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
NH2
O
CF3
Cl
N
O
N
H
H2N
NH2
N
N
O
N
NO2
Cl
O
5-amino-1-MQ
IC50 = 1.2+0.2 µM
JBSNF000088
IC50 = 2.45 + 0.08 µM
HS312
SS3
kobs/[I] = 80+2.5 M-1min-1
IC50 = 0.15-0.25 µM
kobs/[I] = 285+60 M-1s-1
N
H2N
HOOC
O
N
H2N
O
HOOC
NH2
N
N
N
NH2
O
N
N
NH2
N
N
NH2
HOOC
N
N
HO
N
N
OH
HO
N
N
OH
HO
OH
MS2734
IC50=14+1.5 µM
Kd = 2.7+0.2 µM
MS2756
IC50 = 160+1 µM
Kd = 42.8+6.3 µM
H2N
VH45
IC50 = 29.2 + 4.0 µM
H2N
O
H2N
O
O
Figure 1. Structures of reported NNMT inhibitors.
RESULTS AND DISCUSSION
Design. There are two adjacent binding sites that are occupied by nicotinamide and
SAH from the ternary complex crystal structure of the hNNMT-nicotinamide-SAH (PDB
3
ROD).22 The kinetic mechanism of NNMT indicates a rapid equilibrium ordered
mechanism, where NNMT binds SAM first and is followed by binding nicotinamide to
23
form a ternary complex. The distance between the nicotinamide nitrogen atom and the
22
sulfur atom of SAH ranges from 3.5 to 4.2 Å, which is about the linear distance of 2 to 3
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 58
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
methylene groups. Using a saturated 3-C-atom linker has been shown to be a feasible
strategy to link both substrate and SAM analogs to generate a potent and selective
20
inhibitor for NTMT1. Thus, we hypothesized that bisubstrate analogs that covalently link
a SAM and a nicotinamide analog through a 3-C-atom linker would imitate the ternary
complex to offer higher potency and selectivity for NNMT. In addition, there is a very
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
narrow tunnel guiding to the S 2-type of methylation process of nicotinamide by SAM.
N
We proposed that a more rigid unsaturated 3-C-atom like a propargyl linker may
constrain the conformation to decrease the entropy loss and to yield a higher potency
than a saturated 3-C-atom linker (Figure 2).
Figure 2. Design of NNMT bisubstrate inhibitors 1a-3a.
To support our hypothesis, we docked proposed compound 1a into the SAH
binding site of the crystal structure of hNNMT −nicotinamide−SAH ternary complex
(
3ROD) (Figure 3). Comparison of our docking model with the ternary structure depicted
that the NAH moiety of 1a overlays well with SAH and the benzamide moiety inserts
deeper into the substrate binding pocket to attain two extra H-bonds than nicotinamide.
The oxygen atom of the benzamide forms two H-bonds with Ser201 and Ser213. Also, the
ACS Paragon Plus Environment
6

Page 7 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
nitrogen atom of the benzamide interacts with Ser213 and Asp197 residues through two
additional H-bonds.
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 3. Docking analysis of compound 1a in the binding sites of hNNMT. (A) Docking of
compound 1a (blue) in the hNNMT (gray) structure (PDB 3ROD). H-bond interactions are shown
in yellow dotted lines. (B) Overlay of the docking model with the hNNMT (gray)−nicotinamide
(yellow)−SAH (pink) complex (PDB 3ROD).
Synthesis. Syntheses of all bisubstrate analogs were accomplished in a convergent route
through reductive amination of NAH and aldehydes (Figure 2).24 Protected NAH was
synthesized as previously described.25 Compound S3 was synthesized according to
scheme S1. To synthesize compound 1a, allyl alcohol was coupled with 3-iodobenzonitrile
26
4
a under Heck coupling condition to yield aldehyde 5a, which was subject to reductive
amination with protected NAH to produce 6a (Scheme 1). Oxidation by H O followed by
2
2
TFA deprotection provided 1a.19 Other analogs 1e-1j were prepared using the similar
procedure from commercially available aldehydes 5e-5j and protected NAH or S3
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 58
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
(
Schemes 2 and 3). Compound 7a was synthesized from commercially available 3-
27
iodobenzonitrile through Sonogashira cross-coupling reaction (Scheme 4). Then 7a was
subjected to Dess-Martin oxidation to yield aldehyde 8a, which was then subjected to
reductive amination with protected NAH to produce 9a. Subsequent oxidation by H O
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
2
and TFA deprotection produced 2a. Likewise, Sonogashira reaction with 3-butynl-1-ol
offered 10, which underwent hydrogenation and subsequent oxidation to provide
aldehyde 11 (Scheme 5). Reductive amination of 11 with protected NAH through
reductive amination afforded 12a, which was then subject to H O oxidation and acid
2
2
deprotection to yield 3a. Ester prodrugs 1a* and 2a* were synthesized for the cell-based
study (Schemes 1 and 5). MS2756 was re-synthesized as reported before and served as a
28
positive control for inhibition study .
Scheme 1. Synthesis of bisubstrate inhibitors 1a - 1d and 1a*a
ACS Paragon Plus Environment
8

Page 9 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
tBuO2C
N
O
NC 1
R1
3
I
BocHN
O
NH2
OH
NC 1
R1
b
N
3
n
N
H
N
N
a
O
O
1
1
4
4
4
a R = H
5a R = H
1
1
NC
b R = 4-Me
5b R = 4-Me
1
1
c R = 6-Me
5c R = 6-Me
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
R1
1
6
a R = H, n = 1
1
6
b R = 4-Me, n = 1
1
6c R = 6-Me, n = 1
1
6dR = H, n = 2
c
HOOC
H2N
N
O
N
NH2
n
N
N
N
EtOOC
H2N
HO
OH
N
O
O
N
NH2
N
H N
2
N
N
R1
HO
OH
d
O
1
1a R = H, n = 1
1
1
1
1
b R = 4-Me, n = 1
H N
2
1
c R = 6-Me, n = 1
1a*
1
d R = H, n = 2
Scheme 2. Synthesis of bisubstrate inhibitors 1e - 1ia
tBuO2C
HOOC
H2N
H
O
N
O
NH2
N
BocHN
m
N
b
n
e
O
NH2
N
N
n
N
N
N
N
m
O
O
N
R
m
HO
OH
R
5
e 1-naphthol, m = 2
5f 2-naphthol, m = 2
g 1-naphthol, m = 1
h 2-naphthol, m = 1
i 2-naphthol, m =1
6e R = 1-naphthol, m = 2, n = 2
6f R = 2-naphthol, m = 2, n = 2
6g R = 1-naphthol, m = 2, n = 1
6h R = 2-naphthol, m = 2, n = 1
6i R = 2-naphthol, m =1, n =1
1e R = 1-naphthol, m = 2, n = 2
1f R = 2-naphthol, m = 2, n = 2
1g R = 1-naphthol, m = 2, n = 1
1h R = 2-naphthol, m = 2, n = 1
1i R = 2-naphthol, m =1, n =1
5
5
5
a
Scheme 3. Synthesis of bisubstrate inhibitor 1j
tBuO2C
HOOC
H2N
N
N
BocHN
O
O
N
NH2
N
NH2
O
O
N
e
N
b
N
N
N
N
O
H
O
O
HO
OH
O
O
O
O
5j
6j
1j
Scheme 4. Synthesis of bisubstrate inhibitor 2a - 2d and 2a*a
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 58
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
tBuO2C
BocHN
O
N
f
b
NC
R1
1
3
I
OH
H
O
NH2
OH
1
N
NC
R1
3
NC
R1
1
3
N
N
N
e
O
O
NC
R1
4a R1 = H
b R1 = 4-OMe
c R1 = 6-Me
a R1 = H
8a R1 = H
9a R1 = H
7
4
4
7b R1 = 4-OMe
8b R1 = 4-OMe
8c R1 = 6-Me
9b R1 = 4-OMe
9c R1 = 6-Me
c R1 = 6-Me
7
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
b
HOOC
H2N
N
N
N
O
O
NH2
O
N
N
N
NH2
NH2
HN
c
HN
N
N
N
N
N
N
N
HO
OH
OH
O
O
O
HO
O
H2N
NC
2d
9d
2a R1 = H
H2N
R1
2
2
b R1 = 4-OMe
c R1 = 6-Me
EtOOC
H2N
N
O
NH2
N
N
d
N
N
HO
OH
O
H2N
2a*
Scheme 5. Synthesis of bisubstrate inhibitors 3a - 3ba
tBuO2C
BocHN
N
O
NH2
N
OH
n
OH
h
b
N
NC
I
N
NC
H
NC
N
O
O
O
f
4a
1
1
10
c
NC
12a n = 1
2b n = 2
HOOC
H2N
1
N
O
NH2
n
N
N
N
N
HO
OH
O
NH2
3
3
a n = 1
b n = 2
a
Reagents and conditions: (a) Pd(OAc) , Bu NCl, NaHCO , DMF, 3 h, 45 - 84%; (b) protected NAH or S3,
2
4
3
o
NaBH CN, AcOH, MeOH, 2 h, 56 - 68%; (c) H O , K CO , DMSO; then TFA, CH Cl , 0 C to rt, 6 h, 49 - 61%;
3
2
2
2
3
2
2
o
o
o
(
d) SOCl22, EtOH, 0 C to 50 C, 51-59%; (e) TFA, CH Cl , 0 C to rt, 6 h, 49 - 59%; (f) Pd(PPh ) Cl , CuI, Et N,
2
2
3 2
2
3
ACS Paragon Plus Environment
10

Page 11 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
o
CH CN, 12 h, 87 – 95%; (g) DMP, CH Cl , 0 C to rt 1 h, 55 - 61%; (h) Pd/C, H , THF, 12 h; then DMP, CH Cl ,
3
2
2
2
2
2
0 oC to rt 1 h, 60%.
Biochemical Characterization. The SAH hydrolase (SAHH)-coupled fluorescence assay
was employed to evaluate the inhibitory activities of all synthesized compounds
continuously through monitoring the production of SAH.28,29 Under our SAHH-coupled
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
assay condition, the K values for nicotinamide and SAM were determined to be around
m
1
0 µM, which are comparable to reported values (Figure S1).28 Inhibition studies were
performed with both SAM and nicotinamide at their respective K values. All compounds
m
were incubated with NNMT for 10 min before initiating the reaction through addition of
nicotinamide. Inhibitory activity of each compound was first analyzed by obtaining initial
velocities at increasing concentrations and followed by nonlinear regression fit.
Compounds 1a-2a displayed IC values close to the enzyme concentration, suggesting
50
30
that they are tight binding inhibitors. Therefore, both tight binding inhibitors were re-
30,31
analyzed using Morrison’s quadratic equation to determine their apparent K values.
i
Compared to the control MS2756 (K = 10 ± 0.35 µM), 1a (K = 43 ± 5.0 nM), 2a (K = 6.8
i
i
i
±
2.3 nM), and 3a (K = 840 ± 50 nM) exhibit over 200-fold, 1,000-fold, and 10-fold
i
increased inhibition, respectively. Therefore, the 3-C-atom linker used to construct 1a and
a is superior to the 4-C-atom linker used to build 3a for development of NNMT
2
bisubstrate inhibitors. Among them, the propargyl linker is the best linker because
compound 2a showed the highest potency (Figure 4).
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 58
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. Biochemical Characterization of compound LL320 (2a). (A) SAH hydrolase
(
SAHH)-coupled fluorescence assay of compound LL320 ranging in concentrations from
µM to 10 µM; (B) Concentration–response plots fitted to Morrison’s quadratic
0
equation for compound LL320.
Structure-Activity Relationship (SAR) Studies. After the 3-C-atom linker was
established as the optimal length, SAR studies were carried out to understand the
contribution of each portion of a potent NNMT bisubstrate inhibitor (Figure 5). Analogs
of compounds 1a were initially synthesized for the SAR studies since compound 1a was
designed as a standard approach to build a potent bisubstrate analog for NNMT.
Nicotinamide analogs. We began SAR analysis with the investigation on nicotinamide
portion. Our synthetic scheme can accommodate the introduction of nicotinamide
analogs through a late-stage reductive amination reaction. Quinoline and 4-
methylnicotinamide have shown Km values comparable to nicotinamide,32 with
32
isoquinoline showing about a 3-fold lower K value than nicotinamide. Therefore,
m
substituted benzamide and naphthalene moieties were introduced as a nicotinamide
ACS Paragon Plus Environment
12

Page 13 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
analog to produce 1a-1j. Compared with 1a, introducing a methyl group to the ortho-
position of the benzamide marginally affected the inhibition as shown in compound 1c,
while the activity decreased more than 3-fold when introducing methyl group at the para-
position of benzamide as shown in compound 1b. When the benzamide moiety was
replaced by a naphthalene ring to produce compounds 1g and 1h, the inhibitory activities
dropped about 100-fold to 300-fold compared with 1a. Replacing the amide group with
an ester, compound 1j lost about 700-fold activity compared with 1a, which further
supported the importance of the amide group for the binding to NNMT.
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
Introduction of a methoxy group at the para-position of the benzamide of 2a produced
2b, resulting in a 7-fold loss of activity. When a methyl group was introduced at the ortho-
position of the benzamide of 2a to generate 2c, comparable activity was observed.
Aspartic acid portion. MS2734 showed over 10-fold higher inhibitory activity than MS
2
756 by replacing the aspartic amino acid with glutamic acid in the SAM portion.28
However, similar alteration of 1a to 1d failed to display such trends, as 1d exhibited a 70-
fold decreased activity. We also noticed that this alteration produced different effects on
parent compounds as 1e, 1f and 3b showed either comparable or 3-fold decreased
activities when compared to their respective parent compound 1g, 1h and 3a.
Deconstruction strategy. In order to understand the contributions of each moiety of
LL320, we applied a deconstruction strategy. When the aspartate portion is removed to
yield 2d, there is about a 500-fold loss of activity compared to 2a. Furthermore, the Ki
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 58
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
value of benzamide is more than 100 µM, resulting in over 10,000-fold loss of activity
compared to 2a. This suggests that each portion of the bisubstrate analog is important
for the potent inhibition.
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
14

Page 15 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
H2N
HOOC
H2N
H2N
HOOC
N
N
N
O
O
O
N
N
N
N
NH2
N
NH2
N
NH2
HOOC
N
N
N
N
HO
N
N
HO
HO
OH
1c
O
OH
O
OH
1b
O
1
a
H2N
H2N
H2N
Ki = 43 ± 5.0 nM
Ki = 0.15 ± 0.02 µM
Ki = 51 ± 1 nM
N
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
N
N
N
O
HOOC
N
NH2
O
N
HOOC
N
N
NH2
HOOC
N
NH2
NH2
N
N
HO
NH2
N
N
OH
NH2
N
N
HO
HO
OH
O
OH
1d
1e
1f
H2N
Ki = 3.0 ± 0.6 µM
Ki = 1.1 ± 0.16 µM*
Ki = 33.3 ± 5.8 µM*
H2N
HOOC
H2N
HOOC
N
N
O
H2N
HOOC
O
N
N
O
N
N
NH2
N
NH2
N
N
NH2
N
N
N
N
HO
HO
OH
N
N
OH
1i
HO
OH
1
g
1h
Ki = 11.8 ± 1.5 µM*
Ki = 4.0 ± 1.3 µM*
K = 27.3 ± 8 µM*
i
H2N
HOOC
N
H2N
HOOC
O
N
O
N
NH2
N
N
N
NH2
N
N
O
HO
N
N
OH
HO
O
OH
NH2
1
j
MS2756
O
Ki = 28 ± 6.3 µM*
H2N
Benzamide
Ki = >100 µM*
Ki = 10 ± 0.35 µM
O
H2N
HOOC
H2N
HOOC
H2N
HOOC
N
N
N
O
O
O
N
N
N
NH2
N
N
NH2
N
NH2
N
N
N
N
HO
OH
N
N
O
HO
OH
O
HO
OH
O
H2N
H2N
H2N
2a
2b
2c
O
Ki = 6.8 ± 2.3 nM
Ki = 52 ± 5 nM
Ki = 16 ± 2 nM
N
O
H
N
N
NH2
N
N
O
HO
OH
H2N
2
d
Ki = 3.0 ± 0.83 µM
H2N
HOOC
N
N
O
O
N
HOOC
N
N
NH2
N
NH2
NH2
N
N
N
N
HO
HO
OH
OH
3
a
3b
Ki = 0.54 ± 0.12 µM
H2N
H2N
Ki = 0.84 ± 0.05 µM
O
O
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 58
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 5. SARs of NNMT inhibitors. * The values of K for the compounds were determined
i
with duplicate experiments (n = 2).
Slow-binding inhibition. Tight-binding inhibitors exhibit slow binding characteristics.
Therefore, we investigated the optimal preincubation time that will ensure the equilibrium
condition reached for Morrison’s analysis. Previously, all synthesized compounds were only
preincubated with NNMT for 10 min. Therefore, different preincubation times (30 min, 60
min, 90 min, and 120 min) were examined to ensure the steady state condition can be
achieved (Figure 6). From the results (Figure 6B), the system reaches equilibrium after a
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
6
0 min preincubation, as there is a negligible difference in K values at a longer
i
preincubation time. In addition, rapid dilution study suggested the recovery of NNMT
activity at 3 nM of LL320 (Figure S2), suggesting that compound LL320 is a reversible,
tight binding inhibitor for NNMT.
Figure 6. Time-dependent inhibition study of compound LL320 (n=2). (A)
Concentration–response plots for different pre-incubation times fitted to Morrison’s
ACS Paragon Plus Environment
16

Page 17 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
quadratic equation for compound LL320; (B) Nonlinear regression of Morrison K values
i
with pre-incubation time.
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
Selectivity Studies. Bisubstrate analogs are known to display high selectivity to their
target.20 To evaluate the selectivity of 2a, we investigated its inhibitory activity over a
panel of methyltransferases including two closely-related methyltransferases to NNMT
like phenylethanolamine N-methyltransferase (PNMT) and indolethylamine N-
methyltransferase (INMT), two representative members from protein lysine
methyltransferase PKMT (G9a and SETD7) and protein arginine methyltransferase PRMT
(
PRMT1 and TbPRMT7), respectively (Figure S3). We also included NTMT1, sharing a SAM
cofactor binding pocket with NNMT. In addition, SAHH is included in the selectivity study
because it has a SAH binding site and is used in the coupled fluorescence assay. As shown
in Table 2 and Figure S3, compound 2a is more than 2,000-fold selective over SAM-
dependent protein methyltransferases including PNMT, INMT, G9a, SETD7, and PRMT1
since it barely displayed any inhibition at 100 µM. In addition, 2a displayed about 1,000-
fold selectivity for TbPRMT7 and NTMT1, and 100-fold selectivity for SAHH.
Table 2. Selectivity evaluation of compound LL320 (2a)
Enzyme
[LL320] (µM)
activity
a
(
%)
3.7
96
98
97
88
93
11
93
94
97
98
95
33
85
84
100
106
86
100
IC50 (µM)
>100
>100
>100
>100
PNMT
INMT
G9a
SETD7
PRMT1
70
78
106
78
60
>100
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 58
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
TbPRMT7 96
89
85
42
67
65
22
37
39
6
>33
>33
>3.7
NTMT1
SAHH
94
61
a
The values of enzyme activity are mean values of duplicate experiments (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
Figure 7. The structures of NNMT in complex with LL319 and LL320. (A) The structure of NNMT
gray cartoon) bound with LL319 (green stick) (PDB ID 6PVE). An 2Fo-Fc omit map contoured to
(
1.0  is shown for LL319 as a transparent green isosurface. (B) The structure of NNMT (gray
cartoon) bound with LL320 (yellow stick) (PDB ID 6PVS). An Fo-Fc omit map contoured to 3.0  is
shown for LL320 as a transparent yellow isosurface. (C) A structural alignment of LL319 (green
stick) and LL320 (yellow stick) within the binding pocket of NNMT. A ligand interaction diagram
is shown for LL319 (D) and LL320 (E) with NNMT.
Co-crystal Structure of LL319 (1a) and LL320 (2a) in Complex with NNMT. To
elucidate the molecular interactions of bisubstrate analogs with NNMT, we determined
the X-ray co-crystal structure of NNMT in complex with LL319 (PDB ID: 6PVE) and LL320
ACS Paragon Plus Environment
18

Page 19 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
PDB ID: 6PVS) (Figure 7). Both LL319 and LL320 were found to occupy both the cofactor
and nicotinamide binding sties of NNMT. Specifically, the adenine part of 1a and 2a in
the binary complex binds nearly identically with that of SAH. The inhibitor-
superimposition of our NNMT-LL319 and NNMT-LL320 with the published NNMT-
nicotinamide-SAH ternary complex (PDB ID: 3ROD) gave an RMSD value of 0.421 Å and
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
.442 Å (across all residues of chain A), respectively. The interactions of inhibitors-NNMT
are retained in a similar manner as previously observed with NNMT-SAH-nicotinamide in
the ternary complex of nicotinamide/SAH and NNMT-MS2756 binary complex (Figure
22
S4) . For example, the carboxyl group of the NAH portion forms four H-bonds with Tyr
2
0, Tyr25, Tyr69, and Thr163. The amino group of NAH forms two H-bonds with Gly63
and Thr163 (Figure 7D and E). Meanwhile, the adenine moiety of both 1a and 2a forms
two H-bonds with Asp142 and Val143, and two H-bonds with two water molecules.
Hydroxyl groups of the ribose also form H-bonds with both Asp85 and Asn90. Meanwhile,
the benzamide portion of 1a and 2a also binds very similarly to the nicotinamide through
H-bond interactions with Ser201, Ser213, and one water molecule. Extensive examination
of H-bonding interactions of 2 with NNMT indicates that additional H-bonds were formed
between the amino group of NAH with Ser64 and Thr67, and between the benzamide
portion with two surrounding Tyr residues 203 and 204. Superimposed structures also
suggested that both 1a and 2a extended a bit further to interact with NNMT (Figure S4A).
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 58
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
Cell Permeability Evaluation. We tested the cell permeability of compounds 1a-3a
through MS detection of their cellular levels.33 However, the compounds were barely
detected at 100 µM inside cells. In order to increase the cell permeability, corresponding
ethyl ester prodrugs 1a* and 2a* of compounds 1a and 2a were designed and
synthesized (Scheme 1 and 5). Unfortunately, prodrug 1a* barely exhibited enhanced cell
permeability and 2a* only displayed limited permeability at 100 µM in comparison with
TAT peptide through a cellular MS assay, respectively (Figure S5-7).
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
In summary, we designed and synthesized a series of potent bisubstrate inhibitors of
NNMT. This potency supports the optimal length of the methyl transfer tunnel. To our
knowledge, 2a is the most potent NNMT inhibitors reported so far with a K value of 1.6
i
±
0.3 nM in the SAHH-coupled fluorescence assay. 2a exhibited around 1,000-fold
selectivity for NNMT over other methyltransferases and almost 100-fold selectivity for
SAHH. Furthermore, the co-crystal structure of NNMT in complex with compound 1a and
2a clearly showed that the bisubstrate inhibitor occupied both cofactor SAM and
substrate binding sites, which is consistent with our design strategy and the docking study.
Despite their high potency and selectivity, limited cell permeability restricts them from
cell-based studies. However, these tight and slow binding bisubstrate inhibitors are
valuable probes for development of future cell-potent inhibitors for NNMT. More
importantly, 2a is the first example to adopt a propargyl linker to conjugate SAM and a
ACS Paragon Plus Environment
20

Page 21 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
substrate analog to yield a methyltransferase bisubstrate inhibitor. In addition, the
propargyl linker of 2a provides insights into the orientation of the methyl transfer of
NNMT. Furthermore, the strategy described here not only applies to target NNMT, but
also may have widespread impact on the development of potent inhibitors for other
methyltransferases.
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
Note added: During the revision of this manuscript, Shair and coworkers developed
another tight binding bisubstrate inhibitor NS1 with alkynyl linker for NNMT.41 The Ki
value of the compound was around 500 pM and its analogs showed modest effect on
cell-based study.
EXPERIMENTAL SECTION
Chemistry General Procedures. The reagents and solvents were purchased from
commercial sources (Fisher and Sigma-Aldrich) and used directly. Analytical thin-layer
chromatography was performed on ready-to-use plates with silica gel 60 (Merck, F254).
Flash column chromatography was performed over silica gel (grade 60, 230−400 mesh)
on Teledyne Isco CombiFlash purification system. Final compounds were purified on
preparative reversed phase high-pressure liquid chromatography (RP-HPLC) was
performed on Agilent 1260 Series system. The Agilent 1260 Infinity II Variable Wavelength
Detector (G7114A, UV = 254 nM) and a Waters BEH C18 (130Å, 5 µm, 10 mm X 250 mm)
at a flow rate of 4 mL/min using a solvent system of 100% water with 0.1% TFA to 50%
methanol over 30min were used to purify the final compounds. NMR spectra were
acquired on a Bruker AV500 instrument (500 MHz for 1H-NMR, 126 MHz for 13C-NMR).
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 58
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
TLC-MS were acquired using Advion CMS-L MS. Matrix-assisted laser desorption
ionization mass spectra (MALDI-MS) data were acquired in positive-ion mode using a
Sciex 4800 MALDI TOF/TOF MS. The Agilent 1260 Infinity II Variable Wavelength Detector
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
(
G7114A, UV = 254 nM) and an Agilent ZORBAX RR SB-C18 (80Å, 3.5 µm, 4.6 x 150 mm)
at a flow rate of 1 mL/min using a solvent system of 100% water with 0.1% TFA to 40 or
0% methanol over 20min were used to assess purity of final compounds. All the purity
6
of target compounds showed >95% in RP-HPLC.
General procedure A (Heck coupling). Tetrabutyl ammonium chloride (4.9 mmol) and
o
sodium bicarbonate (12.2 mmol) were taken in dry DMF (4 mL) and cooled to 0 C and 3-
iodobenzonitril (4a-c) (4.9 mmol) was added. Allyl alcohol (7.3 mmol) was added to the
reaction mixture, followed by a catalytic amount of Pd(OAc) (0.15 mmol) and the mixture
2
o
was stirred at 0 C for 30 min. The reaction mixture was slowly warmed up to room
temperature and further stirred for 2 h. The reaction mixture was diluted with water and
the organic product was extracted with ether. The combined extracts were dried over
anhydrous Na SO and concentrated with under reduced pressure to get the crude
2
4
product. The residue was purified with silica gel column to provide desired product.
General procedure B (Sonogashira cross coupling). Under nitrogen condition,
Pd(PPh ) Cl (0.1mmol), CuI (0.2 mmol) and iodine substrates (4a-c) (5 mmol) were
3 2
2
successively added to a 25 mL vial equipped with a stir bar. MeCN (5 mL) was added using
ACS Paragon Plus Environment
22

Page 23 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a syringe, then propargyl alcohol (6mmol) was added to the mixture. Et N (15 mmol) was
3
added at last. The reaction was stirred at room temperature for 8h. Solvent was removed
in vacuo to leave a crude mixture, which is purified by silica gel column chromatography
to afford pure product.
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
General procedure C (Oxidation of alcohols). To a solution of the alcohol (7a-c) (2.5
o
mmol) in DCM (5 mL) were added DMP (2.5 mmol) at 0 C. The resulting solution was
stirred for 1 h at room temperature. The reaction was quenched with saturated sodium
hyposulfite solution (5mL), extracted with DCM (3×25mL), washed with Na S O three
2
2
3
times and dried over anhydrous Na SO . After removal of solvent, the residue was purified
2
4
with silica gel column to provide the aldehydes.
General procedure D (Reductive amination). To a solution of protected NAH, S1 or S3
(
0.06 mmol) in 0.5 mL MeOH was added the aldehyde (5a-j, 8a-c and 11) (0.08 mmol)
followed by 2 drops of AcOH. The resulting mixture was stirred for 30 min before
NaBH CN (0.1 mmol) was added. After being stirred for 2 h at rt, the reaction was
3
quenched with saturated NaHCO and extracted with DCM (5×). The combined organic
3
layers were dried over Na SO , filtered, and concentrated. The residue was purified with
2
4
silica gel column to provide desired product.
General procedure E (Oxidation and deprotection of bisubstrate inhibitors). To a
solution of the cyano compound (6a-d, 9a-d, and 12a-b) (0.05 mmol) in DMSO (1 mL)
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 58
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
was added K CO (0.2 mmol). The mixture was cooled to 0 °C and treated with H O (0.2
2
3
2
2
mL). The reaction mixture was warmed to rt and stirred for 3h at rt. The reaction was
diluted with water and extracted with EtOAc (3×). The combined organic layers were dried
over Na SO , filtered, and concentrated to afford the crude compound. The crude
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
4
protected inhibitor was dissolved in DCM (0.5 mL). The resulting solution was cooled to 0
C and treated with TFA (0.2 mL). The reaction mixture was warmed to rt and stirred for 6
°
h. The solution was concentrated under reduced pressure, dissolved in ddH O and
2
purified by reverse HPLC to get the corresponding bisubstrate compound.
General procedure F (Deprotection of bisubstrate inhibitors). The protected inhibitor
(6e-j) was dissolved in DCM (0.5 mL). The resulting solution was cooled to 0 °C and treated
with TFA (0.2 mL). The reaction mixture was warmed to rt and stirred for 6 h. The solution
was concentrated under reduced pressure, dissolved in ddH O and purified by reverse
2
HPLC to get the corresponding bisubstrate compound.
General procedure G (Esterification of bisubstrate analogs). To a solution of
bisubstrate analog 1a or 2a (0.02mmol) in anhydrous EtOH (3 mL), SOCl (0.1mmol) was
2
added at 0 °C. The mixture was stirred at 50 °C for 2 h. The volatiles were removed under
reduced pressure. The residue was dissolved in ddH O and purified by RP-HPLC to provide
2
the ester prodrugs of the respective bisubstrate inhibitors.
ACS Paragon Plus Environment
24

Page 25 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
S)-2-amino-4-((((2R,3S,4R,5R)-5-(6-amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl)methyl)(3-(3-
carbamoylphenyl)propyl)amino)butanoic acid (1a). Compound 1a was prepared
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
according to the general procedure E. H NMR (500 MHz, Methanol-d ) δ 8.40 (s, 1H),
4
8
1
–
1
1
7
.32 (s, 1H), 7.71 – 7.61 (m, 2H), 7.39 – 7.21 (m, 2H), 6.11 – 5.95 (m, 1H), 4.71 – 4.61 (m,
H), 4.47 – 4.32 (m, 2H), 4.04 – 3.92 (m, 1H), 3.84 – 3.70 (m, 1H), 3.67 – 3.52 (m, 2H), 3.50
3.39 (m, 1H), 3.29 – 3.18 (m, 2H), 2.78 – 2.65 (m, 2H), 2.44 – 2.27 (m, 1H), 2.24 – 2.12 (m,
13
H), 2.12 – 1.97 (m, 2H). C NMR (126 MHz, Methanol-d ) δ 170.93, 170.21, 151.82, 148.27,
4
45.87, 142.66, 140.50, 133.91, 131.52, 128.42, 127.20, 125.26, 119.67, 90.70, 78.73, 73.19,
2.14, 54.97, 51.19, 46.92, 46.75, 31.65, 24.72, 24.55. MALDI-MS (positive) m/z: calcd for
+
C H N O [M + H] m/z 529.2523, found m/z 529.4907.
24
33
8
6
Ethyl
(S)-2-amino-4-((((2R,3S,4R,5S)-5-(6-amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl)methyl)(3-(3-
carbamoylphenyl)propyl)amino)butanoate (1a*). Compound 1a* was prepared
1
according to the general procedure G. H NMR (500 MHz, D O) δ 8.22 (s, 1H), 8.16 (s, 1H),
2
7.43 (d, J = 7.7 Hz, 1H), 7.34 (s, 1H), 7.21 (t, J = 7.5 Hz, 1H), 7.14 (d, J = 7.8 Hz, 1H), 5.92 (d,
J = 3.6 Hz, 1H), 4.58 (t, J = 4.2 Hz, 1H), 4.39 – 4.29 (m, 2H), 4.20 (q, J = 7.1 Hz, 2H), 4.15 –
4
.10 (m, 1H), 3.76 – 3.68 (m, 1H), 3.57 (d, J = 14.3 Hz, 1H), 3.52 – 3.43 (m, 1H), 3.43 – 3.34
(m, 1H), 3.24 (t, J = 8.3 Hz, 2H), 2.63 – 2.54 (m, 1H), 2.54 – 2.46 (m, 1H), 2.41 – 2.30 (m, 1H),
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 58
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
.27 – 2.18 (m, 1H), 2.00 – 1.80 (m, 2H), 1.17 (t, J = 7.1 Hz, 3H). MALDI-MS (positive) m/z:
+
calcd for C H N O Na [M + Na] m/z 579.2656, found m/z 579.1624.
26
36
8
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
(S)-2-amino-4-((((2R,3S,4R,5R)-5-(6-amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl)methyl)(3-(5-carbamoyl-2-
methylphenyl)propyl)amino)butanoic acid (1b). Compound 1b was prepared
1
according to the general procedure E. H NMR (500 MHz, Methanol-d ) δ 8.33 (d, J = 4.3
4
Hz, 1H), 8.26 (s, 1H), 7.62 (d, J = 2.0 Hz, 1H), 7.57 (dd, J = 7.8, 2.0 Hz, 1H), 7.16 (d, J = 7.8
Hz, 1H), 6.04 (t, J = 4.2 Hz, 1H), 4.70 (t, J = 4.2 Hz, 1H), 4.48 – 4.38 (m, 2H), 3.91 (dd, J =
9.6, 3.6 Hz, 1H), 3.80 – 3.70 (m, 1H), 3.62 (d, J = 13.9 Hz, 1H), 3.54 (td, J = 8.5, 8.0, 4.3 Hz,
1H), 3.50 – 3.42 (m, 1H), 3.35 (s, 3H), 2.75 – 2.60 (m, 2H), 2.39 – 2.29 (m, 1H), 2.22 (s, 2H),
13
2.12 (d, J = 6.5 Hz, 2H), 2.03 (p, J = 7.7 Hz, 2H). C NMR (126 MHz, Methanol-d ) δ 172.35,
4
155.10, 150.24, 149.89, 143.20, 141.70, 139.89, 132.71, 131.47, 128.90, 126.71, 121.09,
109.28, 92.11, 80.48, 74.33, 73.65, 56.09, 53.77, 49.85, 49.51, 30.41, 26.38, 24.85, 19.30.
(S)-2-amino-4-((((2R,3S,4R,5R)-5-(6-amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl)methyl)(3-(3-carbamoyl-4-
methylphenyl)propyl)amino)butanoic acid (1c). Compound 1c was prepared
1
according to the general procedure E. H NMR (500 MHz, Methanol-d ) δ 8.43 (s, 1H), 8.28
4
(
s, 1H), 7.22 (s, 1H), 7.12 (d, J = 7.8 Hz, 1H), 7.06 (d, J = 7.9 Hz, 1H), 6.08 (d, J = 4.1 Hz, 1H),
4.73 – 4.68 (m, 1H), 4.49 – 4.35 (m, 2H), 4.08 – 3.93 (m, 1H), 3.80 – 3.71 (m, 1H), 3.63 (d, J
ACS Paragon Plus Environment
26

Page 27 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
=
13.8 Hz, 1H), 3.58 – 3.50 (m, 1H), 3.49 – 3.41 (m, 1H), 3.29 – 3.18 (m, 2H), 2.71 – 2.56 (m,
13
2H), 2.38 (s, 3H), 2.36 – 2.29 (m, 1H), 2.23 – 2.12 (m, 1H), 2.08 – 1.96 (m, 2H). C NMR (126
MHz, Methanol-d ) δ 174.24, 170.38, 151.72, 148.30, 145.69, 142.85, 137.44, 136.31,
4
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
3
5
33.35, 130.75, 129.54, 126.54, 120.04, 90.63, 78.77, 73.18, 72.14, 54.91, 51.08, 46.87, 46.29,
+
1.17, 24.67, 24.49, 18.03. MALDI-MS (positive) m/z: calcd for C H N O [M + H] m/z
25
35
8
6
43.2680, found m/z 543.3776.
(S)-2-amino-5-((((2R,3S,4R,5R)-5-(6-amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl)methyl)(3-(3-
carbamoylphenyl)propyl)amino)pentanoic acid (1d). Compound 1d was prepared
1
according to the general procedure E. H NMR (500 MHz, Methanol-d ) δ 8.34 (d, J = 32.5
4
Hz, 2H), 7.79 – 7.58 (m, 2H), 7.41 – 7.20 (m, 2H), 6.07 (d, J = 3.8 Hz, 1H), 4.75 – 4.62 (m,
2
H), 4.50 – 4.32 (m, 2H), 3.97 (d, J = 7.0 Hz, 1H), 3.78 (dd, J = 13.9, 9.8 Hz, 1H), 3.62 (d, J =
13.3 Hz, 1H), 3.36 – 3.32 (m, 1H), 3.27 (d, J = 8.2 Hz, 3H), 2.82 – 2.60 (m, 2H), 2.04 (dt, J =
2.0, 6.0 Hz, 2H), 1.97 (d, J = 19.4 Hz, 2H), 1.93 – 1.81 (m, 2H). 13C NMR (126 MHz,
1
Methanol-d4) δ 170.92, 156.31, 148.48, 140.46, 134.01, 133.98, 133.94, 131.48, 128.45,
127.24, 126.24, 125.24, 90.80, 78.73, 73.06, 72.20, 52.59, 50.00, 46.94, 46.77, 31.67, 27.11,
24.43, 19.63.
(S)-2-amino-5-((((2R,3S,4R,5R)-5-(6-amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl)methyl)(3-(naphthalen-1-
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 58
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
yl)propyl)amino)pentanoic acid (1e). Compound 1e was prepared according to the
1
general procedure F. H NMR (500 MHz, Methanol-d ) δ 8.29 (s, 1H), 8.19 (s, 1H), 8.01 –
4
7
.87 (m, 1H), 7.87 – 7.75 (m, 1H), 7.70 (d, J = 8.2 Hz, 1H), 7.51 – 7.38 (m, 2H), 7.32 (t, J =
.6 Hz, 1H), 7.19 (d, J = 6.8 Hz, 1H), 5.99 (d, J = 3.4 Hz, 1H), 4.66 – 4.55 (m, 1H), 4.49 – 4.30
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
7
(
m, 2H), 4.00 – 3.88 (m, 1H), 3.84 – 3.73 (m, 1H), 3.59 (d, J = 13.8 Hz, 1H), 3.49 – 3.33 (m,
3
–
1
9
H), 3.29 – 3.25 (m, 1H), 3.20 – 3.08 (m, 1H), 3.08 – 2.96 (m, 1H), 2.23 – 2.04 (m, 2H), 2.04
13
1.78 (m, 4H). C NMR (126 MHz, Methanol-d ) δ 170.13, 152.86, 148.22, 145.75, 142.05,
4
35.86, 133.93, 131.38, 128.48, 126.89, 125.76, 125.64, 125.30, 125.08, 122.80, 119.70,
0.85, 78.26, 73.17, 72.12, 53.30, 52.83, 51.95, 48.45, 48.23, 28.87, 27.13, 19.76. MALDI-MS
+
(
positive) m/z: calcd for C H N O [M + H] m/z 550.2778, found m/z 550.3173.
28
36
7
5
(S)-2-amino-5-((((2R,3S,4R,5R)-5-(6-amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl)methyl)(3-(naphthalen-2-
yl)propyl)amino)pentanoic acid (1f). Compound 1f was prepared according to the
1
general procedure F. H NMR (500 MHz, Methanol-d ) δ 8.31 (s, 1H), 8.27 (s, 1H), 7.80 –
4
7.75 (m, 1H), 7.71 (d, J = 8.1 Hz, 2H), 7.54 (s, 1H), 7.46 – 7.38 (m, 2H), 7.19 (d, J = 8.2 Hz,
1H), 6.00 (d, J = 3.7 Hz, 1H), 4.70 – 4.56 (m, 1H), 4.48 – 4.33 (m, 2H), 3.93 (t, J = 6.1 Hz, 1H),
3.85 – 3.70 (m, 1H), 3.70 – 3.55 (m, 1H), 3.39 – 3.32 (m, 2H), 3.30 – 3.24 (m, 2H), 2.87 – 2.67
13
(
m, 2H), 2.19 – 2.03 (m, 2H), 2.03 – 1.74 (m, 4H). C NMR (126 MHz, Methanol-d ) δ 170.27,
4
153.68, 148.45, 141.67, 141.54, 137.33, 133.59, 132.25, 127.90, 127.22, 127.05, 126.25,
126.14, 125.79, 125.16, 119.80, 90.77, 78.44, 73.06, 72.17, 54.71, 53.04, 52.75, 48.37, 32.06,
ACS Paragon Plus Environment
28

Page 29 of 58
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
+
2
7.16, 24.53, 19.72. MALDI-MS (positive) m/z: calcd for C H N O [M + H] m/z 550.2778,
28
36
7
5
found m/z 550.3164.
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
(S)-2-amino-4-((((2R,3S,4R,5R)-5-(6-amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl)methyl)(3-(naphthalen-1-
yl)propyl)amino)butanoic acid (1g). Compound 1g was prepared according to the
1
general procedure F. H NMR (500 MHz, Methanol-d ) δ 8.26 (s, 1H), 8.12 (s, 1H), 7.87 (d,
4
J = 5.8 Hz, 1H), 7.73 (d, J = 6.9 Hz, 1H), 7.61 (d, J = 8.2 Hz, 1H), 7.35 (dd, J = 6.2, 3.2 Hz,
2
H), 7.25 (t, J = 7.6 Hz, 1H), 7.14 (d, J = 5.4 Hz, 1H), 5.93 (s, 1H), 5.42 (s, 1H), 4.55 (s, 1H),
4
.44 – 4.22 (m, 2H), 3.94 – 3.80 (m, 1H), 3.68 (s, 1H), 3.57 – 3.43 (m, 2H), 3.37 (s, 2H), 3.06
13
(d, J = 7.2 Hz, 2H), 3.01 – 2.91 (m, 1H), 2.28 (s, 1H), 2.08 (p, J = 8.1 Hz, 3H). C NMR (126
MHz, CD OD_SPE) δ 172.62, 154.64, 151.16, 149.70, 143.46, 137.36, 135.35, 132.83, 129.89,
3
1
28.26, 127.19, 127.06, 126.70, 126.48, 124.29, 92.10, 80.38, 74.57, 73.58, 56.00, 54.80,
53.55, 30.31, 28.04, 26.37, 25.70.
(S)-2-amino-4-((((2R,3S,4R,5R)-5-(6-amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl)methyl)(3-(naphthalen-2-
yl)propyl)amino)butanoic acid (1h). Compound 1h was prepared according to the
1
general procedure F. H NMR (500 MHz, Methanol-d ) δ 8.38 – 8.22 (m, 2H), 7.76 (d, J =
4
7
.6 Hz, 1H), 7.74 – 7.68 (m, 2H), 7.55 (s, 1H), 7.45 – 7.36 (m, 2H), 7.25 – 7.18 (m, 1H), 6.04
5.96 (m, 1H), 4.68 – 4.59 (m, 1H), 4.48 – 4.35 (m, 2H), 4.03 – 3.90 (m, 1H), 3.83 – 3.69 (m,
–
ACS Paragon Plus Environment
29