Structure-Activity Relationship Studies of Substituted 2-(Isoxazol-3-yl)-2-oxo-N′-phenyl-acetohydrazonoyl Cyanide Analogues: Identification of Potent Exchange Proteins Directly Activated by cAMP (EPAC) Antagonists

Article abstract of DOI:10.1021/acs.jmedchem.5b00635

Exchange proteins directly activated by cAMP (EPAC) as guanine nucleotide exchange factors mediate the effects of the pivotal second messenger cAMP, thereby regulating a wide variety of intracellular physiological and pathophysiological processes. A series of novel 2-(isoxazol-3-yl)-2-oxo-N′-phenyl-acetohydrazonoyl cyanide EPAC antagonists was synthesized and evaluated in an effort to optimize properties of the previously identified high-throughput (HTS) hit 1 (ESI-09). Structure-activity relationship (SAR) analysis led to the discovery of several more active EPAC antagonists (e.g., 22 (HJC0726), 35 (NY0123), and 47 (NY0173)) with low micromolar inhibitory activity. These inhibitors may serve as valuable pharmacological probes to facilitate our efforts in elucidating the biological functions of EPAC and developing potential novel therapeutics against human diseases. Our SAR results have also revealed that further modification at the 3-, 4-, and 5-positions of the phenyl ring as well as the 5-position of the isoxazole moiety may allow for the development of more potent EPAC antagonists.

Full text of DOI:10.1021/acs.jmedchem.5b00635

Subscriber access provided by UNIV OF MISSISSIPPI
Article
Structure-Activity Relationship Studies of Substituted 2-(Isoxazol-3-yl)-2-
Oxo-N'-Phenyl-Acetohydrazonoyl Cyanide Analogues: Identification of
Potent Exchange Proteins Directly Activated by cAMP (EPAC) Antagonists
Na Ye, Yingmin Zhu, Haijun Chen, Zhiqing Liu, Fang C. Mei,
Christopher Wild, Haiying Chen, Xiaodong Cheng, and Jia Zhou
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b00635 • Publication Date (Web): 07 Jul 2015
Downloaded from http://pubs.acs.org on July 12, 2015
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 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Structure-Activity Relationship Studies of Substituted 2-(Isoxazol-3-yl)-2-Oxo-N'-Phenyl-
Acetohydrazonoyl Cyanide Analogues: Identification of Potent Exchange Proteins Directly
Activated by cAMP (EPAC) Antagonists
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
†
,‖
‡,‖
†
†
‡
†
Na Ye, Yingmin Zhu, Haijun Chen, Zhiqing Liu, Fang C. Mei, Christopher Wild,
†
‡,
†,
Haiying Chen, Xiaodong Cheng, * and Jia Zhou *
†
‡
Chemical Biology Program, Department of Pharmacology and Toxicology, University of Texas
Medical Branch, Galveston, Texas 77555, United States
Department of Integrative Biology and Pharmacology, Texas Therapeutics Institute, The Brown
Foundation Institute of Molecular Medicine, The University of Texas Health Science Center,
Houston, Texas 77030, United States
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 63
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:
Exchange proteins directly activated by cAMP (EPAC) as guanine nucleotide exchange factors
mediate the effects of the pivotal second messenger cAMP, thereby regulating a wide variety of
intracellular physiological and pathophysiological processes. A series of novel 2-(isoxazol-3-yl)-
0
1
2
3
4
5
6
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
-oxo-N'-phenyl-acetohydrazonoyl cyanide EPAC antagonists were synthesized and evaluated in
an effort to optimize properties of the previously identified high-throughput (HTS) hit 1 (ESI-09).
Structure–activity relationship (SAR) analysis led to the discovery of several more active EPAC
antagonists (e.g. 22 (HJC0726), 35 (NY0123), and 47 (NY0173)) with low micromolar to
submicromolar inhibitory activity. These inhibitors may serve as valuable pharmacological
probes to facilitate our efforts in elucidating the biological functions of EPAC and developing
potential novel therapeutics against human diseases. Our SAR results have also revealed that
further modification at the 3-, 4-, and 5-positions of the phenyl ring as well as the 5-position of
the isoxazole moiety may allow for the development of more potent EPAC antagonists.
ACS Paragon Plus Environment

Page 3 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
The pivotal second messenger cAMP can regulate a wide range of physiological and
pathological processes via signaling pathways which are involved in many disease states
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
including cancer, diabetes, inflammation, heart failure, and neurological disorders. The
physiological effects of cAMP are mainly transduced by three different downstream effectors,
2
3, 4
including protein kinase A (PKA), cyclic nucleotide-activated ion channels,
and the most
5-10
recently discovered exchange proteins directly activated by cAMP (EPAC). After activated by
the elevation in intracellular concentration of cAMP, EPAC proteins act as guanine nucleotide
exchange factors (GEFs) to catalyze the exchange of GDP for GTP on the Ras-like small
5, 6, 11
GTPases, Rap1 and Rap2,
thereby controlling the Rap-mediated biological processes. Two
5, 6, 11
isoforms of EPAC proteins have been identified, EPAC1 and EPAC2,
which are coded by
two independent genes in both mature and developing tissues of mammals with distinct
expression patterns. EPAC1, also known as cAMP-GEF-I (RAPGEF3 in humans), is nearly
ubiquitously expressed, whereas the expression of EPAC2, also known as cAMP-GEF-II
(
RAPGEF4 in humans), is relatively restricted. At least three splice variants of human EPAC2
1
2
have been found to exhibit a confined expression pattern. Epac2A is somewhat the most
6
, 13, 14
broadly expressed, and largely found in brain, pituitary and pancreatic islets.
By contrast,
13, 15
Epac2B has so far only been detected in the adrenal and testis,
while Epac2C expression is
1
6
restricted to the liver.
Since their discovery in 1998, EPAC proteins have evoked significant interest toward
elucidating their roles in physiological processes and pathogenesis, as well as the molecular
mechanisms of EPAC activation. Accumulating evidence has demonstrated that EPAC proteins
play a critical role in insulin secretion, cardiac contraction, osteoclast differentiation, vascular
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 63
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
permeability, neurotransmitter release, Treg-mediated immune suppression, integrin-mediated
cell adhesion, cell migration and proliferation, cell exocytosis and apoptosis, as well as gene
1
7-23
transcription and chromosomal integrity.
Meanwhile, a substantial number of small-molecule
0
1
2
3
4
5
6
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
EPAC ligands including various cAMP analogues and newly discovered EPAC-specific
antagonists have been discovered and exploited for uncovering the biological functions of
10
24, 25
EPAC. Various cAMP derivatives have been reported,
such as 8-pCPT-2’-O-Me-cAMP
2
6-28
29
29-31
(007),
8-pCPT-2’-O-Me-cAMP-AM (007-AM), and Sp-8-pCPT-2’-O-Me-cAMPS,
which have been used as powerful pharmacological tools in EPAC related research. However,
these cAMP analogues as EPAC ligands are limited in their use due to a lack of selectivity
between the isoforms of EPAC and potential off-target effects. Recently, substantial effort has
been made to focus on non-nucleotide small molecules as EPAC-selective or isoform-specific
antagonists, such as 4-cyclopentyl-2-((2,5-dimethylbenzyl)thio)-6-oxo-1,6-dihydropyrimidine-5-
carbonitrile
(HJC0197)
and
4-cyclopropyl-2-((2,5-dimethylbenzyl)thio)-6-oxo-1,6-
3
2
dihydropyrimidine-5-carbonitrile (HJC0198) as EPAC antagonists, 5,7-dibromo-6-fluoro-3,4-
33, 34
dihydro-2-methyl-1(2H)-quinolinecarboxaldehyde (CE3F4) as an EPAC1 selective inhibitor,
as well as 1,3,5-trimethyl-2-tosylbenzene (ESI-05) and 1-(mesitylsulfonyl)-2,4-dimethyl-1H-
3
5
pyrrole (HJC0350) as specific EPAC2 antagonists.
To identify new pharmacological probes with high capability of specifically inhibiting
EPAC activity, our team developed an automated, robust and sensitive fluorescence-based high-
throughput screening (HTS) assay, which is based on detecting the dose-dependent change in
fluorescence intensity of a fluorescent cyclic nucleotide analog, 8-NBD-cAMP with purified full-
3
6, 37
length EPAC2 protein.
After analyzing the EPAC2 8-NBD-cAMP competition activity,
selected compounds were evaluated by their ability to inhibit cAMP-mediated EPAC1 and
ACS Paragon Plus Environment

Page 5 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
EPAC2 GEF activity using purified recombinant full-length EPAC1 and EPAC2 proteins. A
Maybridge Hitfinder compound library of 14,400 structurally diverse small molecules has been
3
2,
35,
38
screened.
Among these, 2-(5-tert-butylisoxazol-3-yl)-N'-(3-chlorophenyl)-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
oxoacetohydrazonoyl cyanide (ESI-09, 1) has been identified and characterized as a novel non-
3
8
cyclic nucleotide EPAC-specific antagonist with no effect towards PKA (Figure 1). NMR
39
spectroscopy experiments have also verified that 1 specifically interacts with the cAMP
4
0-49
binding domain (CBD) of human EPAC1 (aa. 148-318).
1 dose-dependently competes with
8-NBD-cAMP in binding to EPAC2, with an IC value of 9.0 µM. In the presence of 20 µM of
5
0
cAMP, 1 also dose-dependently displays cAMP-mediated the GEF inhibitory activity of full-
3
9
length EPAC2 and EPAC1 with apparent IC values of 4.4 μM and 10.8 μM, respectively.
50
NH2
O
N
N
NC
HN
N
N
3
4
O
N
N
5
O
O
2
HO
6
5
P
O
Cl
3
O
OH
4
1:
screen hit
high-throughput
ESI-09
cAMP
Figure 1. Structures of cAMP and HTS hit 1
as a specific EPAC inhibitor has been a useful tool compound in revealing the roles of
1
5
0
cAMP in regulation of Plasmodium falciparum merozoite invasion of human erythrocytes and
5
1
restoration of endothelial barrier after thrombin-induced hyperpermeability, uncovering the
23
functions of EPAC in osteoclast formation, cAMP dependent regulation of Schwann cell
52
53-55
proliferation and differentiation, in addition to other fields.
Further biological functional
investigations have also demonstrated that 1 can specifically block intracellular Akt
phosphorylation, EPAC-mediated Rap1 activation and insulin secretion in pancreatic β cells.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 63
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
Treatment of pancreatic ductal adenocarcinoma (PDA) cells with 1 has no obvious effect on cell
proliferation and viability, but can give rise to a significant decrease in cell migration and
3
8
invasion. Further, by employing an orthotopic metastatic PDA mouse model, 1 was found to
0
1
2
3
4
5
6
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
reduce local and distant spread of MIA PaCa-2 cells and significantly decrease metastasis to the
2
0
liver at a dose of 10 mg/kg via i.p. injection for 3 weeks. Compound 1 also enhances leptin
signaling in organotypic hypothalamic slice culture system. Administration of 1 in wild-type
56
mice at a dose of 50 mg/kg by oral gavage for 3 weeks significantly reduces plasma leptin.
Moreover, treatment of 1 at a nontoxic concentration can attenuate the formation of cytopathic
effects, significantly reduce viral yields, and effectively protect permissive cells against Middle
East respiratory syndrome coronavirus (MERS-Cov) infection by inhibiting viral RNA
replication and protein expression of MERS-CoV without affecting the expression and
57
localization of EPAC protein. It was also shown that hit 1 can completely recapitulate the
EPAC1 knockout phenotype in vivo via pharmacological inhibition of EPAC1, and significantly
5
8
block the early stage of rickettsial attachment and invasion into the nonphagocytic host cells.
Treatment with 1 at a dose of 10 mg/kg via i.p. injection for 12 days significantly protects wild-
type mice against rickettsial infection with much milder disease manifestations and dramatically
5
8
improved survival. Taken together, these findings support the proof-of-concept study that
compound 1 is a selective pharmacological probe in unraveling the in vivo functions of EPAC,
and may provide potential novel therapeutics for the prevention and treatment of various human
diseases including pancreatic cancer, diabetes, obesity, Middle East respiratory syndrome
coronavirus infections, and fatal rickettsioses.
5
8
58
Compound 1 displays excellent bioavailability, low toxicity to animals, good
1
0
membrane permeability, no significant inhibitory effects on PDEs, as well as very weak
ACS Paragon Plus Environment

Page 7 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
inhibitory activities towards hERG and CYP450 enzymes. All these combined observations
support that non-nucleotide small molecule 1 may have superior advantages in terms of off-
target effects, selectivity and toxicities over the traditional cAMP analogues. Despite a potential
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
59
concern associated with its protein denaturing properties at high concentrations, our extensive
3
9
biochemical and pharmacological study has defined its therapeutic window and validated that 1
indeed acts as an EPAC specific antagonist. Therefore, it is imperative to further optimize 1
through rational drug design approaches to develop advanced leads with enhanced activity,
specificity, and better druglike properties.
Herein, we report our chemical optimization efforts using the HTS hit 1 as the chemical
lead and the detailed structure-activity relationship (SAR) studies on a series of substituted 2-
(isoxazol-3-yl)-2-oxo-N'-phenyl-acetohydrazonoyl cyanide analogues. Several new molecules
such as 2-(5-(tert-butyl)isoxazol-3-yl)-N-(3,5-dichlorophenyl)-2-oxoacetohydrazonoyl cyanide
6
0
(22,
HJC0726),
2-(5-(tert-butyl)isoxazol-3-yl)-2-oxo-N-(3,4,5-
6
0
trichlorophenyl)acetohydrazonoyl cyanide (35, NY0123), and 2-(5-cyclohexylisoxazol-3-yl)-
60
2
-oxo-N-(3,4,5-trichlorophenyl)acetohydrazonoyl cyanide (47, NY0173) have been identified
as potent EPAC antagonists with low micromolar to submicromolar inhibitory activity.
RESULTS AND DISCUSSION
61
62
Design. Given the well characterized X-ray crystal structures of inactive and active forms of
EPAC2 proteins, we examined the predicted the binding mode via molecular docking studies of
1
into the cAMP binding domain B of active EPAC2 protein using the AutoDock Vina
10
algorithm, thereby understanding an effective interaction mode of the 2-(isoxazol-3-yl)-2-oxo-
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 63
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
N'-phenyl-acetohydrazonoyl cyanide-based chemotype. Results suggest that the binding of the
inhibitor to the EPAC2 protein may primarily occur through two terminal hydrophobic
interactions and a unique linker. In the present work we focus our new molecular design on the
optimization of the aforementioned two main terminal hydrophobic pharmacophores: the 3-
chlorophenyl moiety (P1) and the 5-tert-butyl group on the isoxazole (P2), as well as the
exploration of the linker (P3) for systematic chemical modifications as depicted in Figure 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
P3
O
NC
N
O
N
HN
Cl
P1
P2
Figure 2. Proposed structural modifications on 1
Chemistry. As outlined in Scheme 1, the key intermediates 4a-e were generated by
condensation of the commercially available pinacolone 2a-e with oxalic acid diethyl ester 3 in
the presence of NaH (60%) followed by oximation with hydroxylamine hydrochloride and
6
3
subsequent cyclization by heating in one-pot procedure in 30-87% yields. Ethyl esters 4a-e
were then converted into the corresponding ketonitriles 5a-e by the treatment with MeLi as the
6
3
base and CH CN as cyanide source under our modified Kowalsko’s protocol. Requisite
3
aromatic amines 6 were treated with sodium nitrite and 2 N hydrochloric acid to give the
corresponding aryldiazonium chlorides 7. Direct coupling of the aryldiazonium salts 7 with the
ACS Paragon Plus Environment

Page 9 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
crude cyanomethyl ketones 5a-e in the presence of NaOAc as the catalyst at 0 °C afforded new
substituted 2-(isoxazol-3-yl)-2-oxo-N'-phenyl-acetohydrazonoyl cyanide derivatives 8 – 48 in
2
4-76% yields in two steps from 4a-e (Scheme 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
Scheme 1. Synthesis of Substituted 2-(Isoxazol-3-yl)-2-Oxo-N'-Phenyl-Acetohydrazonoyl
a
Cyanide Analogues with Modification on the Phenyl Ring and the Isoxazol Ring
O
EtO C
O
2
N
NC
R2
N
c
a, b
+
OEt
O
EtO
O
O
O
R2
R2
2a-e
3
4a-e
5a-e
2
2
2
2
2
2
2
2
a: R = t-Bu
4a: R = t-Bu
5a: R = t-Bu
2
2
2
b: R = Me
4b: R = Me
5b: R = Me
2
2
2
c: R = c-Pr
4c: R = c-Pr
5c: R = c-Pr
d: R2 c-Hex
=
4d: R2 c-Hex
=
5d: R2 c-Hex
=
2
2
2
e: R = Ph
4e: R = Ph
5e: R = Ph
O
NC
N
Cl
NH2
N
e
O
d
N
N
R1
HN
R1
R2
R1
6
7
8 - 48
-
1
=
:
H; 2-Cl; 4-Cl; 3-Br; 3-F; 3-CF3 3-NO2 3-COOEt; 3-CN; 3-COCH3 3-
; ; ;
For compounds 8 48 R
OMe; 3-CH OH; 3-Me; 2,5-di-Cl; 3,5-di-Cl; 3,4-di-Cl; 2,3-di-Cl; 3,5-di-CF ; 3-F, 5-CF ; 3,5-di-F; 3-
2
3
3
,
Cl, 5-CF ; 3-F, 5-Cl; 3-F, 4-Cl; 3-Cl, 4-F; 3-CF 4-Cl; 3,4-di-F; 3-Cl, 4-CF ; 3,4,5-tri-Cl; 3,4,5-tri-
3
3
3
R2 t-Bu; Me; c-Pr; c-Hex; Ph.
=
F; 2,4,6-tri-Cl.
a
Reagents and conditions: (a) NaH, THF, 0 °C to rt; (b) NH OH•HCl, EtOH/THF, rt to reflux,
2
3
0-87% for two steps (c) CH CN, MeLi, THF, -78 °C; (d) 2N HCl, NaNO , H O, 0 °C; (e) 5a-e,
3 2 2
NaOAc, EtOH, 24-76% for three steps.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 63
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
As depicted in Schemes 2 and 3, amides 51-53 and 55, urea 54, carbohydrazide 56 and
pyrazole 57 were prepared as novel substituted 2-(isoxazol-3-yl)-2-oxo-N'-phenyl-
acetohydrazonoyl cyanide analogues bearing different linkers between isoxazole and
chlorobenzene of 1. Condensation of the commercially available isoxazol-3-ylamines 49a-b with
2-cyanoacetic acid in the presence of EDCI and DIPEA yielded the corresponding intermediates
0
1
2
3
4
5
6
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
5
0a-b, which were then converted to 51 and 52 by coupling with 3-chlorophenyldiazonium salt
upon treatment with NaOAc, in 87% and 53% yields over two steps, respectively. Amide 53 and
urea 54 were prepared by condensation of 49a with 3-chlorobenzoic acid and 1-isocyanato-4-
chlorobenzene, in 33% and 17% yields, respectively. Compound 55, the reverse amide of 53,
was prepared first by removing the protection group of the ester 4a, following by condensation
with 3-chloroaniline in the presence of HBTU and DIPEA, in 48% yield for two steps.
Hydrazinolysis of ester 4a and then condensation with 3-chlorobenzaldehyde provided
6
4
carbohydrazide 56 with a unique linker in 61% yield over two steps. To mask the ketone group,
pyrazole 57 was prepared by cyclization of the linker of 1 with hydrazine hydrate in refluxing
65
ethanol in 64% yield.
Scheme 2. Synthesis of Substituted 2-(Isoxazol-3-yl)-2-Oxo-N'-Phenyl-Acetohydrazonoyl
a
Cyanide Analogues with Modification on the Linker
ACS Paragon Plus Environment

Page 11 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
NC
N
O
O
O
N
N O
a
b
N
R3
R3
N
H
R3
NC
HN
H N
N
2
H
Cl
Cl
3
3
3
49a R = t-Bu
50b R = t-Bu
50b R = Me
51 R = t-Bu
52 R = Me
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
3
3
49b R = Me
O
N
N
O
O
c
O
N
H N
d
O
N
2
N
H
N
N
Cl
H
H
53
49a
54
O
O
N
O
N
O
O
O
e, f
g, h
N
N
NH
H
EtO
Cl
Cl
55
4a
56
a
Reagents and conditions: (a) 2-cyanoacetic acid, EDCI, DIPEA, DCM, rt; (b) 7, NaOAc, EtOH,
0 °C to rt, 53-87% yields for two steps. (c) 3-chlorobenzoic acid, (COCl) , Et N, DCM, 0 °C to rt,
2
3
3
3%; (d) 1-isocyanato-3-chlorobenzene, DCM, rt, 17%; (e) LiOH, THF/H O, rt; (f) 3-
2
chloroaniline, HBTU, DIPEA, rt, 48% for two steps (g) NH NH , EtOH, reflux; (h) 3-
2
2
chlorobenzaldehyde, EtOH, rt, 61% for two steps.
Scheme 3. Synthesis of Substituted 2-(Isoxazol-3-yl)-2-Oxo-N'-Phenyl-Acetohydrazonoyl
Cyanide Analogue 57 with Cyclization of the Linker
H
O
N
NC
N
H N
2
N
NH NH •H O
N
N
O
2
2
2
HN
N
O
N
EtOH, reflux
4%
6
Cl
Cl
1
57
In Vitro Evaluation of EPAC2 Inhibition. All newly synthesized compounds have been
evaluated for their ability to compete with 8-NBD-cAMP binding to recombinant EPAC2
36
proteins to determine IC values. The previous hit 1 and cAMP were used as the reference
5
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 63
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
10, 39
compounds and our data were almost identical to those previously reported,
with IC values
50
of 8.9 and 32.0 μM in inhibiting EPAC2, respectively.
As shown in Table 1, we initially focused on investigating the effect of various
substitutions on the phenyl ring of our previous hit 1. To evaluate the importance of substitutions,
we removed 3-chloro on the phenyl ring of 1 leading to compound 8, which has a about 7-fold
loss of binding affinity with an IC of 59.8 μM. Adjustment of 3-chloro to either 2-position or 4-
0
1
2
3
4
5
6
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
5
0
position (as in compound 9 or 10) resulted in a 3-fold decreased inhibitory activity in comparison
with 1, suggesting that a proper substitution at 3-position is beneficial to its activity. We next
explored the electronic effect of substitutions at the 3-position of the phenyl ring in detail.
Replacement of 3-chloro with its congeners bromo and fluoro (as in compounds 11 and 12) led
to a slight loss of activity compared to that of 1, while increased activity by about 3-4 fold
compared to that of non-substituted analog 8, with IC50 values of 14.0 μM and 22.4 μM,
respectively. Compounds 13 and 14 with other electron-withdrawing groups such as
trifluoromethyl and nitro led to approximately 5-fold and 2-fold increase in potency when
compared to that of 8, with IC values of 11.9 μM and 31.3 μM, respectively. Compounds 15
5
0
with ethyl carboxyl and 16 with cyano group both retained a similar inhibitory activity to that of
. However, compound 17 with acetyl group displayed a dramatic loss of activity. Introduction
8
of electron-donating groups such as methoxy and its isomer hydroxymethyl afforded compounds
18 and 19 which also resulted in a significant loss of activity. Interestingly, compound 20 with a
methyl group was about 2-fold more potent than non-substituted analog 8, with an IC of 25.1
5
0
μM. These results suggest that 3-position of phenyl ring of 1 can tolerate chemical modifications
with either electron-withdrawing or electron-donating groups.
ACS Paragon Plus Environment

Page 13 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 1. Apparent IC Values of Substituted 2-(Isoxazol-3-yl)-2-Oxo-N'-Phenyl-
Acetohydrazonoyl Cyanide Scaffolds with Modifications on the Phenyl Ring for Competing
with 8-NBD-cAMP in Binding to EPAC2
50
O
NC
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
N
HN
R1
1
a
b
compd
R
IC (μM)
Relative affinity (RA)
5
0
cAMP
32.0 ± 5.9
8.9 ± 1.2
1.0
3.6
0.5
1.1
1.2
2.3
1.4
2.7
1.0
0.4
0.5
-
1
8
9
3-Cl
H
59.8 ± 9.4
28.3 ± 4.0
26.5 ± 4.1
14.0 ± 2.4
22.4 ± 2.8
11.9 ± 1.7
31.3 ± 5.5
71.4 ± 23.8
67.8 ± 23.5
2-Cl
4-Cl
3-Br
3-F
1
1
1
0
1
2
13
3-CF
3
1
4
3-NO
2
15
3-COOEt
3-CN
1
6
c
17
3-COCH
3
NE
1
1
2
2
2
2
2
2
8
9
0
1
2
3
4
5
3-OMe
NE
-
3-CH
3-Me
OH
NE
-
2
25.1 ± 4.8
1.3
4.4
16.8
10.7
0.9
5.0
5.5
6.0
20.0
12.8
7.4
2,5-di-Cl
3,5-di-Cl
3,4-di-Cl
2,3-di-Cl
3,5-di-CF
7.3 ± 1.2
3
9
1.9 ± 0.3
3.0 ± 0.6
35.6 ± 7.9
6.4 ± 0.9
5.8 ± 0.9
5.3 ± 0.8
1.6 ± 0.3
2.5 ± 0.4
4.3 ± 0.5
3
26
3-F, 5-CF
3
2
7
3,5-di-F
28
3-Cl, 5-CF
3-F, 5-Cl
3-F, 4-Cl
3
2
9
30
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 63
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
3
3
3
3
3
3
1
2
3
4
5
6
3-Cl, 4-F
3-CF , 4-Cl
2.7 ± 0.3
2.9 ± 0.5
6.6 ± 0.8
3.6 ± 0.6
0.9 ± 0.2
1.1 ± 0.2
19.0 ± 1.7
11.9
11.0
4.8
3
3,4-di-F
3-Cl, 4-CF
8.9
3
3,4,5-tri-Cl
3,4,5-tri-F
2,4,6-tri-Cl
35.6
29.1
1.7
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
37
a
The values are the mean ± SE of at least three independent experiments.
b
c
RA = IC50,cAMP / IC50,compound
.
NE: No effect, indicating that a specific IC cannot be calculated from the data points collected.
5
0
We then explored the effect of disubstituted phenyl rings. By keeping 3-chloro and
introducing another chloro group on 2, 4, 5, or 6-position (as in compounds 21-24), 3,4- and 3,5-
dichloro analogs 22 and 23 bearing the best binding affinity, were about 5- and 3-fold more
potent than that of parent compound 1, with IC values of 1.9 μM and 3.0 μM, respectively. We
5
0
then explored fluorinated compounds given their notable successes in drug discovery with an
increasing number of fluorine-containing drugs on the market. Due to the unique nature of
fluorine, fluorine substitution is commonly used in contemporary medicinal chemistry and drug
design to alter the physicochemical and conformational properties, as well as selective reactivity
of developing candidates, thereby improving metabolic stability, solubility, bioavailability, and
66, 67
protein–ligand interactions.
Moreover, 3-fluoro and 3-trifluoromethyl analogues 12 and 13
from the above mono-substituted SAR studies also exhibited comparable potency to that of 3-
chloro (as in compound 1). Therefore, a series of compounds with fluoro, trifluoromethyl, and
chloro groups in 3,4- or 3,5-disubstituted style were designed and synthesized (as in compounds
2
5-34). All of these compounds were approximately 2-6 folds more potent than hit 1.
Compounds 28, 29, 31 and 32 displayed comparable potency to that of compounds 22 and 23,
with IC values of 1.6-3.0 μM (Figure 3). Inspired by the results from 3,4- and 3,5-disubstituted
5
0
ACS Paragon Plus Environment

Page 15 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
design, we attempted to combine their substituted effects and study 3,4,5-trisubstituted analogs.
Intriguingly, 3,4,5-trichloro substituted analog 35 and 3,4,5-trifluoro substituted analog 36
demonstrated an enhanced potency with IC values of 0.9 and 1.1 μM, respectively (Figure 3),
5
0
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
while 2,4,6-trichloro substituted analog 37 resulted in a dramatic loss of activity with an IC of
5
0
19.0 μM. These findings suggest that further optimization on the substitutions at the 3-, 4-,
and/or 5-positions of the phenyl ring may offer the potential to develop new and more potent
EPAC2 inhibitors.
Figure 3. Relative binding affinity of EPAC2 antagonists. Dose-dependent competition of
EPAC2 antagonists with 8-NBD-cAMP in binding to EPAC2.
Having identified an optimal substitution for the phenyl ring that maintains good potency,
we then directed our effort on tuning the isoxazole scaffold. Compounds 1, 22 and 35 were
chosen as reference compounds. By keeping 3-Cl, 3,5-di-Cl, and 3,4,5-tri-Cl as constant
substitutions on the phenyl ring, we embarked on exploring the steric hindrance effects of
substitutions on the 5-position of the isoxazole ring via replacing t-butyl group of the reference
compound with methyl, c-propyl, c-hexyl, or phenyl group. As shown in Table 2, compounds
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 63
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
3
8-41, 42-45, and 46-48 were three small parallel series, which mainly displayed good EPAC2
inhibitory effects in the order of t-Bu > c-Hex ≥ c-Pr > Me > Ph. Among these, compounds 43,
4, 46 and 47 were more potent than hit 1, with IC values of 3.45-8.8 μM (Figure 3). Notably,
4
50
0
1
2
3
4
5
6
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
compounds 41 and 48 with phenyl substitutions were totally inactive, indicating that more
conformationally constricted biaryl moieties may be unsuitable for fitting into the shallow
hydrophobic pockets of EPAC proteins. These results suggest that appropriate bulky alkyl
substitutions offering sufficient lipophilicity are more favorable for the ligand to interact with
EPAC2 proteins.
Table 2. Apparent IC Values of Substituted 2-(Isoxazol-3-yl)-2-Oxo-N'-Phenyl-
Acetohydrazonoyl Cyanide Scaffolds with Modifications on the Isoxazole Ring for
Competing with 8-NBD-cAMP in Binding to EPAC2
50
O
NC
N
O
R2
N
HN
R1
1
2
a
b
compd
cAMP
1
R
R
IC (μM)
Relative affinity (RA)
5
0
32.0 ± 5.9
8.9 ± 1.2
1.0
3.6
-
3-Cl
t-Bu
Me
c
38
39
40
41
22
3-Cl
NE
3-Cl
c-Pr
c-Hex
Ph
60.0 ± 20.9
11.3 ± 2.1
NE
0.5
2.8
-
3-Cl
3-Cl
3
9
3,5-di-Cl
3,5-di-Cl
3,5-di-Cl
3,5-di-Cl
3,5-di-Cl
3,4,5-tri-Cl
3,4,5-tri-Cl
t-Bu
Me
1.9 ± 0.3
16.8
1.4
5.8
6.7
1.1
35.6
3.6
42
22.3 ± 0.4
5.5 ± 1.5
4.8 ± 1.2
28.0 ± 4.0
0.9 ± 0.2
8.8 ± 2.3
4
3
c-Pr
c-Hex
Ph
44
4
5
35
t-Bu
c-Pr
4
6
ACS Paragon Plus Environment

Page 17 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
4
7
8
3,4,5-tri-Cl
3,4,5-tri-Cl
c-Hex
3.5 ± 0.8
NE
9.1
-
Ph
a
The values are the mean ± SE of at least three independent experiments.
b
c
RA = IC50,cAMP / IC50,compound
.
NE: No effect, indicating that a specific IC cannot be calculated from the data points collected.
5
0
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
Amides 51-53 and 55, urea 54, carbohydrazide 56 and pyrazole 57 represented a
subseries of novel substituted 2-(isoxazol-3-yl)-2-oxo-N'-phenyl-acetohydrazonoyl cyanide
analogues bearing different linkages between the isoxazole and chlorobenzene by modifications
of the linker length and type. Interestingly, compounds 51 and 52 with an additional NH
fragment inserted into the linker of 1 and 38 were completely inactive in binding to EPAC2
(Table 3). Other linkages including amide for 53 and 55, urea for 54, carbohydrazide for 56 as
well as pyrazole for 57 also led to an entire loss of potency towards EPAC2 proteins. These
results indicate that the length, shape, and type of the linkages are important for the ligand to
interact with EPAC2 proteins.
Table 3. Apparent IC Values of Substituted 2-(Isoxazol-3-yl)-2-Oxo-N'-Phenyl-
50
Acetohydrazonoyl Cyanide Scaffolds with Modifications on the Linker for Competing with
8
-NBD-cAMP in Binding to EPAC2
Cl
Linker
R3
N
O
3
a
compd
Linker
R
IC (μM)
50
O
N
1
N
t-Bu
8.9
H
CN
O
N
b
5
1
N
N
H
Me
NE
H
CN
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 63
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
N
52
N
H
N
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
NE
NE
NE
NE
NE
H
CN
O
5
5
3
4
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
O
N
N
H
H
O
55
N
H
O
N
5
5
6
7
N
H
H
N
H N
N
2
t-Bu
NE
N N
a
The values are the mean ± SE of at least three independent experiments.
b
NE: No effect, indicating that a specific IC cannot be calculated from the data points collected.
5
0
In Vitro Evaluation of EPAC1 Inhibition. From the biological results discussed above,
compounds 22, 25, 28-29, 31-32, 34−36, 44 and 46-47 were identified as potent EPAC2
inhibitors with IC values lower than 10 μM. Therefore, these selected compounds together with
50
hit 1 were further evaluated for their ability to inhibit EPAC1-mediated Rap1b-bGDP exchange
activity. As shown in Table 4, the previous hit 1 is slightly less potent on EPAC1 inhibition than
that on EPAC2, with an IC value of 10.8 μM. Interestingly, almost all of our selected, newly
5
0
synthesized analogues exhibited significantly enhanced binding affinity compared to that of 1.
1
0
Due to the structural discrepancy between EPAC1 and EPAC2 protein, we found that there is
no similar SAR trend towards EPAC1 as observed for EPAC2. Among these, compound 22 and
3
5 shared the best inhibitory activity as EPAC1 inhibitors, with the IC50 values of 2.4 μM
ACS Paragon Plus Environment

Page 19 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
Figure 4). Compounds 22, 35, and 47, with IC values lower than 4 μM for both EPAC1 and
5
0
EPAC2 may serve as valuable pharmacological tools to probe the functions of EPAC in disease,
or as potential drug candidates for further preclinical development.
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
Table 4. Apparent IC Values of Substituted 2-(Isoxazol-3-yl)-2-Oxo-N'-Phenyl-
Acetohydrazonoyl Cyanide Scaffolds for Inhibiting EPAC1 GEF activity
50
a
compd
EPAC1 IC (μM)
50
1
10.8 ± 1.6
2.4 ± 0.2
5.7 ± 1.7
5.7 ± 1.5
11.6 ± 3.1
7.8 ± 2.0
9.7 ± 2.3
4.3 ± 0.9
2.4 ± 0.3
8.9 ± 2.2
2.6 ± 0.6
4.1 ± 0.7
4.0 ± 0.9
2
2
25
2
8
29
3
3
3
3
3
4
1
2
4
5
6
4
46
4
7
a
The values are the mean ± SE of at least three independent experiments.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 63
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. Relative inhibitory activity for EPAC1-mediated Rap1b-bGDP exchange.
Solubility of Selected Ligands in Water and Ethanol. To examine whether the synthesized
substituted 2-(isoxazol-3-yl)-2-oxo-N'-phenyl-acetohydrazonoyl cyanide analogues have better
solubility than 1 and to define their therapeutic window, a previously reported HPLC analysis
6
8, 69
method
was employed to measure the solubility of several selected analogues such as 25, 28,
3
1, 35, 36, and 47 (Table 5). One-point calibration was performed against standards with known
concentrations of the sample compounds to determine concentrations of the indicated
compounds in samples. 1 itself has a limited aqueous solubility with a maximum concentration
around 5.95 µg/mL, while possessing an excellent solubility in ethanol (14.98 mg/mL). Addition
of another or two chloro atoms onto the phenyl ring of 1 not only enhanced their binding affinity
on EPAC1 and EPAC2, but also improved their aqueous solubility. For instance, aqueous
solubility of analogue 22 with 3,5-di-Cl on phenyl ring is enhanced with a saturated
concentration of 45.4 µg/mL, and that of the 3,4,5-tri-Cl derivatives 35 and 47 is 12.7 and 10.1
µg/mL, respectively. Meanwhile, fluorinated compounds 25, 28, 31 and 36 also exhibit improved
both potency and aqueous solubility. Particularly, compound 25 with 3,5-di-CF on the phenyl
3
ring of 1 displays a significantly improved solubility both in water and ethanol, with saturated
ACS Paragon Plus Environment

Page 21 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
concentrations of 418.5 µg/mL and more than 150 mg/mL, indicating approximately 70-fold and
1
0-fold improvement in comparison with that of 1, respectively (Table 5).
Table 5. Solubility of Selected Compounds in Water and Ethanol
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
a
a
compd
MW
Solubility in Water
Solubility in Ethanol
3
9
5.95 ± 0.83 µg/mL
14.98 ± 0.39 mg/mL
(45.3 ± 1.2 mM)
1
330.769
(18.0 ± 2.5 µM)
3
9
45.4 ± 1.4 µg/mL
6.8 ± 0.86 mg/mL
(18.6 ± 2.3 mM)
22
365.214
432.320
398.770
348.762
399.659
350.301
425.694
(124.3 ± 3.8 µM)
4
(
18.5 ± 17.3 µg/mL
968.0 ± 40.0 µM)
> 150 mg/mL
(> 347.0 mM)
2
2
3
3
3
4
5
8
1
5
6
7
2
0.1 ± 0.8 µg/mL
14.6 ± 0.9 mg/mL
(36.6 ± 2.3 mM)
(50.4 ± 2.0 µM)
8
2.6 ± 3.6 µg/mL
8.21 ± 0.9 mg/mL
(23.5 ± 2.6 mM)
(236.8 ± 10.3 µM )
1
2.7 ± 1.0 µg/mL
4.37 ± 0.52 mg/mL
(10.9 ± 1.3 mM)
(31.8 ± 2.5 µM)
75.7 ± 7.1 µg/mL
(216.0 ± 20.3 µM)
28.6 ± 1.4 mg/mL
(81.6 ± 4.0 mM)
10.1 ± 0.79 µg/mL
3.63 ± 0.46 mg/mL
(8.5 ± 1.1 mM)
(23.7 ± 1.9 µM)
a
The values are the mean ± SE of at least three independent experiments.
Predicted Docking Mode of Compound 47 with the cAMP Binding Domain (CBD) of
EPAC2 Protein.
Since our robust HTS assay is particularly sensitive for identifying compounds that directly
compete with 8-NBD-cAMP in binding to EPAC2, we predicted that our compounds may bind
to the cAMP binding domain (CBD) of EPAC2. Molecular docking studies were performed to
help us better understand the structure-activity relationships of these new compounds toward
EPAC2. Using the AutoDock 4.0 method, the docking results revealed that our new compounds
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 63
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
32, 62
fit nicely into the functional cAMP binding pocket of active EPAC2 (PDB code: 3CF6).
To
further characterize the binding pose, we selected analogue 47 as a case study for our theoretical
investigation. As shown in Figure 5A, the hydrophobic cyclohexylisoxazolyl moiety interacts
with the hydrophobic residues of Phe 367, Leu406, Ala407, and Ala415, while 3,4,5-
trichlorophenyl fragment forms hydrophobic interactions with residues Val386 and Leu397.
Moreover, this binding mode is further stabilized by the occurrence of hydrogen bonds between
the NH hydrogen atom of the linker and the oxygen atom of residue Asp402. These results are in
full agreement with our established SAR data that hydrophobic groups at the isoxazolyl moiety
and phenyl scaffold as well as the types of the linkers are crucial to target EPAC2. The molecular
0
1
2
3
4
5
6
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
0
docking studies of 47 also comply with our previous results of hit 1, through the overlay
analysis of two ligands as depicted in Figure 5B. Given the critical role of the linker, these
docking findings may also help us understand why amides 53 and 55, urea 54, carbohydrazide 56
and pyrazole 57 display no significant activity. Taken together, these studies suggest that
optimization of the substitutions at the isoxazolyl moiety and phenyl scaffold while keeping the
NH groups on the linker may provide further opportunities to improve the binding affinity and
specificity of the substituted 2-(isoxazol-3-yl)-2-oxo-N'-phenyl-acetohydrazonoyl cyanide
analogues towards EPAC proteins.
ACS Paragon Plus Environment

Page 23 of 63
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
Figure 5. (A) Predicted binding mode and molecular docking of 47 into the cAMP binding
domain B (CBD) of EPAC2 protein. Important residues are drawn in sticks. Hydrogen bonds are
shown as dashed purple lines. (B) Overlay analysis of ligands 1 and 47. 1 is shown in purple
while 47 is depicted in yellow.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 63
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
CONCLUSION
In summary, a series of substituted 2-(isoxazol-3-yl)-2-oxo-N'-phenyl-acetohydrazonoyl cyanide
derivatives have been designed, synthesized and biologically evaluated with respect to their
inhibitory capability towards EPAC proteins. The SAR results based on EPAC2 activity were
consistent with our docking studies, indicating that further modifications at the 3-, 4-, 5-positions
of the phenyl scaffold and the 5-position of the isoxazole moiety may allow us to tune the
original hit 1 to achieve more potent EPAC2 antagonists. Interestingly, various linkers explored
led to a significant loss in activity, indicating that the linker in the parent compound plays a
critical role in the binding activity towards EPAC. Furthermore, several new molecules such as
compounds 22, 35, and 47 have been identified as potent EPAC1 and EPAC2 inhibitors with
IC values in the low micromolar to submicromolar range. These compounds may hold promise
0
1
2
3
4
5
6
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
5
0
as potential drug candidates for the preclinical development to IND-enabling studies towards
novel therapeutics against human diseases. Additionally, these small-molecule inhibitors may
serve as valuable pharmacological probes to facilitate our efforts in elucidating the physiological
functions of EPAC in disease and associated signaling pathways. Further studies of selected
compounds in various human disease models and a systematic optimization based upon
identified advanced chemical leads towards EPAC sub-type selectivity are under way and will be
reported in due course.
EXPERIMENTAL SECTION
General Chemistry Information. All commercially available starting materials and
solvents were reagent grade, and used without further purification. Reactions were performed
under a nitrogen atmosphere in dry glassware with magnetic stirring. Preparative column
ACS Paragon Plus Environment

Page 25 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
chromatography was performed using silica gel 60, particle size 0.063-0.200 mm (70-230 mesh,
flash). Analytical TLC was carried out employing silica gel 60 F254 plates (Merck, Darmstadt).
Visualization of the developed chromatograms was performed with detection by UV (254 nm).
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
13
NMR spectra were recorded on a Bruker-600 or Bruker-300 ( H, 600 & 300 MHz; C, 150 & 75
1
13
MHz) spectrometer. H and C NMR spectra were recorded with TMS as an internal reference.
Chemical shifts were expressed in ppm, and J values were given in Hz. High-resolution mass
spectra (HRMS) were obtained from Thermo Fisher LTQ Orbitrap Elite mass spectrometer.
Parameters include the following: Nano ESI spray voltage was 1.8 kV; Capillary temperature
o
was 275 C and the resolution was 60,000; Ionization was achieved by positive mode. Melting
points were measured on a Thermo Scientific Electrothermal Digital Melting Point Apparatus
and uncorrected. Purities of final compounds were established by analytical HPLC, which was
carried out on a Shimadzu HPLC system (model: CBM-20A LC-20AD SPD-20A UV/VIS).
HPLC analysis conditions: Waters µBondapak C18 (300 × 3.9 mm); flow rate 0.5 mL/min; UV
detection at 270 and 254 nm; linear gradient from 10% acetonitrile in water to 100% acetonitrile
in water in 20 min followed by 30 min of the last-named solvent (0.1% TFA was added into both
acetonitrile and water). All biologically evaluated compounds are > 95% pure.
Ethyl 5-tert-butylisoxazole-3-carboxylate (4a). To a solution of pinacolone (500 mg,
5
.0 mmol) in THF (10 mL) was added NaH (60%) (220 mg, 5.5 mmol) at 0°C. The reaction
mixture was stirred at rt for about 30 min. Diethyl oxalate (730 mg, 5.0 mmol) was added at 0°C
and the mixture was then stirred at rt overnight. To the resulting solution, hydroxylamine
hydrochloride (380 mg, 5.5 mmol) in ethanol (10 mL) was added dropwise. The mixture was
heated at reflux for 16 h. After this time the sodium chloride was removed by filtration and the
filtrate was concentrated under reduced pressure. The residue was purified by short column
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 63
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
chromatography on silica gel eluting with hexane/ethyl acetate (4:1) to provide 4a as a colorless
1
oil (855 mg, 87%). H NMR (600 MHz, CDCl ) δ 6.37 (s, 1H), 4.43 (q, J = 7.2 Hz, 2H), 1.41 (t,
3
J = 7.2 Hz, 3H), 1.37 (s, 9H).
0
1
2
3
4
5
6
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
Ethyl 5-methylisoxazole-3-carboxylate (4b). Compound 4b was prepared in 55% yield
1
(two steps from acetone) by a procedure similar to that used to prepare compound 4a. H NMR
(
300 MHz, CDCl ) δ 6.36 (s, 1H), 4.38 (q, J = 7.2 Hz, 2H), 2.48 (s, 3H), 1.34 (t, J = 7.2 Hz, 3H).
3
Ethyl 5-cyclopropylisoxazole-3-carboxylate (4c). Compound 4c was prepared in 74%
yield (two steps from 1-cyclopropylethanone) by a procedure similar to that used to prepare
1
compound 4a. H NMR (300 MHz, CDCl ) δ 6.32 (s, 1H), 4.44 (q, J = 7.1 Hz, 2H), 2.07 (m, 1H),
3
1
.42 (t, J = 7.1 Hz, 3H), 1.18 – 1.10 (m, 2H), 1.06 – 0.98 (m, 2H).
Ethyl 5-cyclohexylisoxazole-3-carboxylate (4d). Compound 4d was prepared in 30%
yield (two steps from 1-cyclohexylethanone) by a procedure similar to that used to prepare
1
compound 4a. H NMR (300 MHz, CDCl ) δ 6.37 (s, 1H), 4.42 (q, J = 7.2 Hz, 2H), 2.91 – 2.78
3
(m, 1H), 2.12 – 2.04 (m, 2H), 1.91 – 1.62 (m, 3H), 1.49 – 1.25 (m, 8H).
Ethyl 5-phenylisoxazole-3-carboxylate (4e). Compound 4e was prepared in 73% yield
1
(two steps from acetophenone) by a procedure similar to that used to prepare compound 4a. H
NMR (300 MHz, CDCl ) δ 7.78 (dd, J = 6.5, 3.2 Hz, 2H), 7.45 (dd, J = 5.0, 1.8 Hz, 3H), 6.91 (s,
3
1
H), 4.45 (q, J = 7.1 Hz, 2H), 1.42 (t, J = 7.1 Hz, 3H).
2-(5-tert-Butylisoxazol-3-yl)-N'-phenyl-2-oxoacetohydrazonoyl cyanide (8). To a
solution of CH CN (410 mg, 10.0 mmol) in anhydrous THF (5 mL) was added 1.6 M methyl
3
lithium in diethyl ether (3.1 mL, 5.0 mmol) at -78 °C under nitrogen. The mixture was stirred at -
7
8 °C for 0.5 h, and 4a (500 g, 2.5 mmol) in THF (5 mL) was then added dropwise. The solution
was stirred at -78 °C for 1 h and then quenched with acetic acid (0.3 mL, 5.0 mmol). The mixture
ACS Paragon Plus Environment

Page 27 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
was warmed to 0 °C and poured onto ice/water (10 mL) and extracted with ethyl acetate (50 mL).
The organic lay was dried over Na SO , filtered and concentrated under reduced pressure. The
2
4
crude residue 5a (490 mg) was obtained as a yellow oil and directly used for next step without
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
further purification.
To a solution of aniline (22 mg, 0.24 mmol) in H O (1 mL) cooled to -5°C was added
2
0
.24 mL of 1 N HCl (aq.). To the resulting acidic aniline solution, 1 mL solution of sodium
nitrite (16 mg, 0.24 mmol) in H O was added dropwise to generate the aryldiazonium salt
2
solution. To the aryldiazonium salt solution was added sodium acetate (33 mg, 0.4 mmol),
followed by 1 mL solution of crude 5a (38 mg, 0.2 mmol) in ethanol. The reaction mixture was
o
stirred at 0 C for 5 min, and then poured onto H O (2 mL) and extracted with ethyl acetate (20
2
mL). The organic layer was dried over Na SO , filtered and concentrated under reduced pressure.
2
4
The residue was purified by short column chromatography on silica gel eluting with hexane/ethyl
acetate (2/1) to provide the desired product 8 (45 mg, 76%, two steps from 4a) as a yellow solid
o
1
(
mp 118-119 C). HPLC purity 99.4% (t = 20.50 min). H NMR (600 MHz, DMSO-d ) δ 12.60
R 6
(bs, 1H), 7.49 (d, 2H, J = 7.8 Hz), 7.42 (d, 2H, J = 7.8 Hz), 7.20-7.23 (m, 1H), 6.66 (s, 1H), 1.37
1
3
(s, 9H). C NMR (150 MHz, DMSO-d ) δ 180.8, 179.2, 160.0, 142.1, 129.3, 125.6, 116.8, 112.2,
6
+
1
10.7, 100.2, 32.4, 28.3. HRMS (ESI) calcd for C H N O 297.1346 (M + H) , found 297.1355.
1
6
17
4
2
2
-(5-tert-Butylisoxazol-3-yl)-N'-(2-chlorophenyl)-2-oxoacetohydrazonoyl cyanide (9).
Compound 9 was prepared in 76% yield (two steps from 4a) by a procedure similar to that used
o
to prepare compound 8. The title compound was obtained as a yellow solid (mp 143-144 C).
1
HPLC purity 96.6% (t = 22.77 min). H NMR (600 MHz, DMSO-d ) δ 7.53-7.60 (m, 1H), 7.47-
R
6
1
3
7
.52 (m, 1H), 7.37-7.46 (m, 1H), 7.22-7.36 (m, 1H), 6.66 (s, 1H), 1.34 (s, 9H). C NMR (150
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 63
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
MHz, DMSO-d ) δ 180.8, 178.6, 159.9, 139.3, 129.9, 128.1, 126.9, 123.5, 119.6, 114.4, 110.7,
6
+
9
9.9, 32.3, 28.2. HRMS (ESI) calcd for C H ClN O 331.0956 (M + H) , found 331.0969.
16 16
4
2
2
-(5-tert-Butylisoxazol-3-yl)-N'-(4-chlorophenyl)-2-oxoacetohydrazonoyl
10). Compound 10 was prepared in 75% yield (two steps from 4a) by a procedure similar to that
used to prepare compound 8. The title compound was obtained as a yellow solid (mp 142-143
cyanide
0
1
2
3
4
5
6
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
1
C). HPLC purity 98.1% (t = 21.74 min). H NMR (600 MHz, DMSO-d ) δ 7.46-7.60 (m, 4H),
R
6
1
3
6
.64 (s, 1H), 1.36 (s, 9H). C NMR (150 MHz, DMSO-d ) δ 181.5, 179.6, 160.8, 142.5, 130.1,
6
129.9, 119.2, 113.3, 111.6, 100.8, 33.0, 29.0. HRMS (ESI) calcd for C H ClN O 331.0956 (M
1
6
16
4
2
+
+
H) , found 331.0963.
-(5-tert-Butylisoxazol-3-yl)-N'-(3-bromophenyl)-2-oxoacetohydrazonoyl
2
cyanide
(11). Compound 11 was prepared in 75% yield (two steps from 4a) by a procedure similar to that
used to prepare compound 8. The title compound was obtained as a yellow solid (mp 162-163
o
1
C). HPLC purity 98.3% (t = 21.93 min). H NMR (600 MHz, DMSO-d ) δ 12.84 (s, 1H), 7.60
R
6
1
3
(
s, 1H), 7.48-7.52 (m, 1H), 7.36-7.40 (m, 2H), 6.69 (s, 1H), 1.38 (s, 9H). C NMR (150 MHz,
DMSO-d ) δ 181.0, 179.3, 160.0, 143.8, 131.4, 127.9, 122.3, 119.0, 116.2, 113.3, 110.5, 100.3,
6
+
32.5, 28.4. HRMS (ESI) calcd for C H BrN O 375.0451 (M + H) , found 375.0456.
1
6
16
4
2
2
-(5-tert-Butylisoxazol-3-yl)-N'-(3-fluorophenyl)-2-oxoacetohydrazonoyl
12). Compound 12 was prepared in 62% yield (two steps from 4a) by a procedure similar to that
used to prepare compound 8. The title compound was obtained as a yellow solid (mp 140-141
cyanide
(
o
1
C). HPLC purity 99.4% (t = 21.73 min). H NMR (600 MHz, DMSO-d ) δ 12.88 (s, 1H), 7.46
R
6
(td, J = 8.2, 6.5 Hz, 1H), 7.38 – 7.31 (m, 1H), 7.22 (d, J = 10.8 Hz, 1H), 7.03 (td, J = 8.4, 2.2 Hz,
1
3
1
2
H), 6.67 (s, 1H), 1.36 (s, 9H). C NMR (150 MHz, DMSO-d ) δ 180.9, 179.2, 162.6 (d, J =
6
42.3 Hz), 160.2, 131.2 (d, J = 8.85 Hz), 113.3, 112.9, 111.9 (d, J = 21.2 Hz), 110.8, 103.7,
ACS Paragon Plus Environment

Page 29 of 63
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
+
1
3
03.6, 100.3, 32.4, 28.4. HRMS (ESI) calcd for C H FN O 315.1252 (M + H) , found
16 16 4 2
15.1248.
-(5-tert-Butylisoxazol-3-yl)-N'-(3-trifluoromethylphenyl)-2-oxoacetohydrazonoyl
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
cyanide (13). Compound 13 was prepared in 33% yield (two steps from 4a) by a procedure
similar to that used to prepare compound 8. The title compound was obtained as a yellow solid
o
1
(
mp 147-148 C). HPLC purity 96.0% (t = 21.80 min). H NMR (600 MHz, DMSO-d ) δ 7.74
R 6
(d, J = 8.0 Hz, 1H), 7.70 (s, 1H), 7.64 (t, J = 7.9 Hz, 1H), 7.51 (d, J = 7.6 Hz, 1H), 6.63 (s, 1H),
1
3
1.36 (s, 9H). C NMR (150 MHz, DMSO-d ) δ 181.2, 179.7, 161.1, 131.1, 130.5 (q, J = 31.5
6
Hz), 125.5, 123.5, 122.4, 121.8, 113.7, 112.2, 100.6, 32.9, 28.9. HRMS (ESI) calcd for
+
C H F N O 365.1220 (M + H) , found 365.1230.
17 16
3
4
2
2
-(5-tert-Butylisoxazol-3-yl)-N'-(3-nitrophenyl)-2-oxoacetohydrazonoyl cyanide (13).
Compound 14 was prepared in 29% yield (two steps from 4a) by a procedure similar to that used
o
to prepare compound 8. The title compound was obtained as a yellow solid (mp 151-152 C).
1
HPLC purity 96.4% (t = 20.33 min). H NMR (600 MHz, DMSO-d ) δ 12.03 (s, 1H), 8.19 (s,
R
6
1
H), 7.97 (d, 1H, J = 8.4 Hz), 7.85 (d, 1H, J = 7.8 Hz), 7.67 (t, 1H, J = 7.8 Hz), 6.62 (s, 1H),
1
3
1.37 (s, 9H). C NMR (150 MHz, DMSO-d ) δ 180.5, 179.0, 171.9, 160.8, 148.5, 130.7, 124.6,
6
1
19.1, 113.4, 112.1, 111.2, 100.1, 32.4, 28.4. HRMS (ESI) calcd for C H N O 342.1197 (M +
16 16 5 4
+
H) , found 342.1207.
Ethyl 3-(2-(2-(5-tert-butylisoxazol-3-yl)-1-cyano-2-oxoethylidene)hydrazinyl)benzo-
ate (15). Compound 15 was prepared in 74% yield (two steps from 4a) by a procedure similar to
that used to prepare compound 8. The title compound was obtained as a yellow solid (mp 155-
o
1
1
8
56 C). HPLC purity 98.8% (t = 21.53 min). H NMR (600 MHz, DMSO-d ) δ 12.92 (s, 1H),
R 6
.16 (s, 1H), 7.76 (d, 1H, J = 7.8 Hz), 7.74 (d, 1H, J = 8.4 Hz), 7.57 (t, 1H, J = 8.4 Hz), 6.67 (s,
ACS Paragon Plus Environment