Discovery of 4 H-Chromen-4-one Derivatives as a New Class of Selective Rho Kinase (ROCK) Inhibitors, which Showed Potent Activity in ex Vivo Diabetic Retinopathy Models

Article abstract of DOI:10.1021/acs.jmedchem.9b01143

Diabetic retinopathy (DR) is a major cause of blindness, and there is a lack of effective treatment at present. Rho-associated coiled-coil containing serine/threonine protein kinases (ROCKs) have recently been suggested as potential targets for the DR treatment. We herein report the discovery of 4H-chromen-4-one derivatives as a new class of ROCK inhibitors. Structure-activity relationship analyses led to the identification of the most active compound, 4-(dimethylamino)-N-(3-{2-[(4-oxo-4H-chromen-7-yl)oxy]acetamido}phenyl) (12j). This compound showed excellent kinase selectivity for ROCK I and ROCK II against 387 other kinases. In retinal explants, compound 12j protected retinal neurons from high glucose-induced oxidative stress and apoptosis-mediated cell death. Furthermore, 12j administration suppressed the improper proliferation of Müller cells and promoted the regression of vascular vessels in retinal explants cultured in a high glucose microenvironment. Collectively, our data suggest that 12j could be a potential lead compound for the treatment of DR, hence deserving further in-depth studies.

Full text of DOI:10.1021/acs.jmedchem.9b01143

Subscriber access provided by UNIV OF CAMBRIDGE
Article
Discovery of 4H-chromen-4-one Derivatives as a New Class
of Selective Rho Kinase (ROCK) Inhibitors, Which Showed
Potent Activity in Ex Vivo Diabetic Retinopathy Models
Lanying Zhao, Yueshan Li, Yujiao Wang, Zeen Qiao, Zhuang Miao, Jiao
Yang, Lu-Yi Huang, Chenyu Tian, Linli Li, Danian Chen, and Shengyong Yang
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01143 • Publication Date (Web): 06 Nov 2019
Downloaded from pubs.acs.org on November 6, 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 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of 4H-chromen-4-one Derivatives as a New Class of Selective Rho Kinase
(ROCK) Inhibitors, Which Showed Potent Activity in Ex Vivo Diabetic Retinopathy
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
Models
†,#
†,#
‡, #
†
†
†
Lanying Zhao, Yueshan Li, Yujiao Wang, Zeen Qiao, Zhuang Miao, Jiao Yang, Luyi
†
§
§,*
‡,*
†,*
Huang, , Chenyu Tian, Linli Li, Danian Chen, Shengyong Yang
†
State Key Laboratory of Biotherapy and Cancer Center, West China Hospital, Sichuan
University, Chengdu, Sichuan 610041, China.
‡
Department of Ophthalmology, West China Hospital, Sichuan University, Chengdu, China.
Key Laboratory of Drug Targeting and Drug Delivery System of Ministry of Education, West
§
China School of Pharmacy, Sichuan University, Sichuan 610041, China.
#
*
These authors contributed equally to this work.
Corresponding authors. Shengyong Yang, State Key Laboratory of Biotherapy and Cancer
Center, West China Hospital, Sichuan University, Chengdu, Sichuan 610041, China. Email:
yangsy@scu.edu.cn, Danian Chen, Department of Ophthalmology, West China Hospital,
Sichuan University, Chengdu, Sichuan 610041, China. Email: danianchen2006@gmail.com;
Linli Li, Key Laboratory of Drug Targeting and Drug Delivery System of Ministry of Education,
West China School of Pharmacy, Sichuan University, Sichuan 610041, China. Email:
ysylilinli@sina.com.cn.
KEYWORDS: ROCK kinase; selectivity; structure-activity relationship; diabetic retinopathy.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 76
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:
Diabetic retinopathy (DR) is a major cause of blindness, and there is a lack of effective treatment
at present. Rho-associated coiled-coil containing serine/threonine protein kinases (ROCKs) have
recently been suggested as potential targets for the DR treatment. We herein report the discovery
of 4H-chromen-4-one derivatives as a new class of ROCK inhibitors. Structure-activity
relationship (SAR) analyses led to the identification of the most active compound, 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
(dimethylamino)-N-(3-{2-[(4-oxo-4H-chromen-7-yl)oxy]acetamido}phenyl)
(12j).
This
compound showed excellent kinase selectivity for ROCK I and ROCK II against 387 other
kinases. In retinal explants, compound 12j protected retinal neurons from high glucose-induced
oxidative stress and apoptosis mediated-cell death. Furthermore, 12j administration suppressed
the improper proliferation of Müller cells and promoted the regression of vascular vessels in
retinal explants cultured in a high-glucose microenvironment. Collectively, our data suggest that
12j could be a potential lead compound for the treatment of diabetic retinopathy, hence
deserving further in-depth studies.
2
ACS Paragon Plus Environment

Page 3 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
. INTRODUCTION
Diabetic retinopathy (DR), one of the most prominent complications of diabetes, is the
leading cause of blindness. With the rising prevalence of diabetes, it is estimated that the number
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
of people who have signs of DR could substantially increase in the coming years. DR is mainly
characterized by abnormal changes in vascular features, such as microaneurysm, vascular
2
permeabilization, neovascularization and proliferative vitreoretinopathy. These changes have
long been regarded as central to disease pathogenesis, although the molecular mechanisms are
still largely unknown. The current main treatment strategy for DR is suppression of these
vascular phenotypes, including anti-VEGF (vascular endothelial growth factor), through the
injection of medications and panretinal photocoagulation4,5 that destroys hypoxic retinal tissue-
secreting proangiogenic factors. However, the therapeutic efficacy is still unsatisfactory. In
recent years, a growing body of evidence suggests that neuronal dysfunction and
3
6
2
neurodegeneration also exist in DR, even before the overt nonperfusion of the neuropile. It has
also been hypothesized that neuronal apoptosis and glial cell dysfunction caused by the
accumulation of extracellular glutamate, oxidative stress, and the reduction of neuroprotective
7
factor synthesis may play a central role in DR pathogenesis. Therefore, the identification of new
therapeutic agents that are able to simultaneously suppress neovascularization and protect retinal
neurons from degradation appears to be a promising strategy to treat DR effectively.
Recent studies have shown that Rho-associated coiled-coil kinases (ROCKs) play important
roles in both angiogenesis and neuronal degeneration.8, 9 The ROCK family contains two
members: ROCK I and ROCK II, which share approximately 65% amino acid identity. After
6
activation by GTP-bound Rho proteins, ROCKs phosphorylate a variety of substrates, such as
myosin phosphatase targeting subunit 1 (MYPT1),10 myosin light chain (MLC)11 and LIM
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 76
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
kinases (LIMK).12 These downstream substrates participate in the regulation of actin filament
polymerization and stress fiber formation, both of which are involved in endothelial cell
migration that is necessary for retinal microvascular angiogenesis.8,13 Deprivation of ROCK
activity in retinal endothelial cells promotes the depolymerization of actin filaments and inhibits
stress fiber formation, thus preventing endothelial migration and reducing microvascular damage.
Moreover, many studies have revealed that ROCKs are associated with the pathogenesis of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
4
neuronal damage. The abnormal activation of ROCKs was observed in various disorders of the
central nervous system, including inflammatory and demyelinating diseases, spinal cord injury,
stroke, Alzheimer's disease and neuropathic pain.14 Inactivation of ROCKs has been shown to
promote neurite growth and sprouting.14 Furthermore, hyperactivation of the Rho/ROCK
pathway was observed in the development of DR. All of these findings indicate that ROCKs
could be potential targets for DR. To date, a number of potent ROCK inhibitors have been
reported.15,16 However, most of these inhibitors showed poor kinase selectivity, implying
potential toxicity. Therefore, more selective and potent ROCK inhibitors are still required.
To discover selective and potent ROCK inhibitors, we first performed a molecular docking-
based high-throughput screen (HTS) against the SPECS commercial chemical library and our in-
house chemical database (see Supporting Information for details), which identified the
compound,
4-(dimethylamino)-N-(3-{2-[(2-oxo-1,2,3,4-tetrahydroquinolin-7-
yl)oxy]acetamido}phenyl) benzamide (1, Figure 1). Compound 1 displayed moderate inhibitory
activity against ROCK II with an IC50 value of 3.141 µM. Interestingly, 1 showed weak or no
activity against other related kinases (see Supporting Information Table S1 for details) in a
kinase profiling assay. We then carried out further structural modification of this compound.
Structure-activity relationship (SAR) studies of 1 led to the discovery of a highly selective and
4
ACS Paragon Plus Environment

Page 5 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
potent ROCK inhibitor. Finally, we evaluated the activity of this compound in an ex vivo DR
model.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Chemical structure of compound 1 with three regions (Region I, II and III) that are for
the structure-activity relationship analysis.
2
. RESULTS AND DISCUSSION
2.1. Chemical Syntheses
All compounds in this study were prepared following the general strategies in Schemes 1-7.
Most of the final derivatives were synthesized by well-established methods from hydroxy-4H-
chromen-4-one derivatives (3a-d) and 7-hydroxy-3,4-dihydro-2H-1-benzopyran-4-one (6) as the
starting materials, which were synthesized according to Scheme 1 and Scheme 2. Conversion of
the 2,4-dihydroxyphenyl ketones to corresponding 7-hydroxychromenones was accomplished
according to published procedures for some of the simpler analogs. The 6-hydroxychromenones,
7-hydroxychromenones and their derivatives (3a-d) were prepared with 60-85% yield from the
corresponding dihydroxy acetophenones by treatment with triethyl orthoformate and perchloric
acid (Scheme 1). Resorcinol (4) and 3-chloropropionic acid in trifluoromethanesulfonic acid at
80°C for 1 h gave the Friedel-Crafts product 5. After an aqueous workup, the crude product was
stirred in sodium hydroxide solution to afford chromanone 6 with 85% yield (Scheme 2).
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 1. Synthesis of compounds 3a-d.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
a
Reagents and conditions: (i) Triethyl orthoformate, Perchloric acid, rt, 1.5h, 80°C, 4h, 60-78%;
ii) Triethyl orthoformate, Perchloric acid, Diethyl ether, rt, 0.5h, 50°C, 5h, 85%;
(
Scheme 2. Synthesis of compound 6.a
a
Reagents and conditions: (i) 3-chloropropanoic acid, Trifluoromethanesulfonic acid, 80°C, 1h,
9
0%; (ii) NaOH, 5°C, rt, 85%.
Scheme 3 shows the synthetic routes for compounds 12a-m. 3-nitroaniline (7) was treated
with chloroacetyl chloride in THF at 0°C for 1 h and then heated to 40°C for 12 h to provide 8
with an excellent yield. Then, the O-alkylation of the phenol derivatives with intermediate 8 was
catalyzed by DBU at 90°C, which gave intermediates 9a-m with 63-75% yield. Finally, the 3-
aminobenzene derivatives (10a-m), which were derived from the nitro reduction of 9a-m by iron,
were treated with 4-(dimethylamino)benzoyl chloride (11) to afford compounds 12a-m with 45-
6
0% yield.
Scheme 3. Synthesis of compounds 12a-m.a
6
ACS Paragon Plus Environment

Page 7 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
Reagents and conditions: (i) K
DBU, Phenol derivatives, n-butanol, rt, 90°C, 3h, 63-75%; (iii) Fe, NH
h; (iv) TEA, Pyridine, rt, 60°C, 45-60%.
2
CO
3
, Chloroacetyl chloride, THF, 0°C, 40°C, 12h, 85%; (ii)
4
Cl, EtOH/H O=2/1, 80°C,
2
2
Next, we focused on the variation of the tail group. Thirty-one compounds were synthesized
by the methods shown in Scheme 4 (13-39) and Scheme 5 (40-43). Intermediate 10j was reacted
with either aryl sulfonyl/benzoyl chlorides or HATU-activated benzoic acids to afford
compounds 13-39 with 40-66% yield (Scheme 4). A deprotection reaction of 39 produced 40
with 75% yield. Compounds 41 and 42 were obtained by the nitro reduction of 26 and 28,
respectively. The hydrolysis of 37 produced 43 with 60% yield (Scheme 5).
Scheme 4. Synthesis of compounds 13-39.a
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 76
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
a
Reagents and conditions: (i) TEA, Pyridine, rt, 60°C, 40-66%; (ii) HATU, DIEA, DMF rt,
65°C, 12h, 60%.
Scheme 5. Synthesis of compounds 40-43.a
a
Reagents and conditions: (i) DCM, TFA, rt, 6 h, 75%; (ii) Fe, NH
4
Cl, EtOH/H O=2/1, 80°C,
2
2
h, 69-72%; (iii) NaOH, MeOH/H O=2/1, rt, 3 h, 60%.
2
8
ACS Paragon Plus Environment

Page 9 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 6 depicts the synthetic routes of compounds 47a-l. Similar to the key intermediate 8,
different nitroanilines reacted with chloroacetyl chloride to give intermediate 45a-l with 60-79%
yield. Then, the O-alkylation of intermediate 3a with the aforementioned intermediates afforded
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
6a-l with 63-75% yield. Finally, compounds 46a-l were reduced by iron followed by the
reaction with 4-(dimethylamino)benzoyl chloride (11) in pyridine at 60°C to give final
compounds 47a-l with 45-60% yield over two steps. Scheme 7 shows the synthetic route for
compound 50, which is very similar to that for 47a-l, except the reaction reagent chloroacetyl
chloride was replaced by chloromethanesulfonyl chloride.
Scheme 6. Synthesis of compounds 47a-l.a
a
Reagents and conditions: (i) K
2
CO , Chloroacetyl chloride ,THF, 0°C, 40°C, 12h, 60-79%; (ii)
3
DBU, Phenol derivatives, n-butanol, rt, 90°C, 3h, 63-75%; (iii) Fe, NH
2h; (iv) TEA, Pyridine, rt, 60°C, 45-60%.
4
Cl, EtOH/H O=2/1, 80°C,
2
Scheme 7. Synthesis of compound 50.a
a
Reagents and conditions: (i) K
2
CO , Chloromethanesulfonyl chloride, THF, 0°C, 45°C, 12h,
3
6
2%; (ii) DBU, Phenol derivatives, n-butanol, rt, 80°C, 3h, 63-75%; (iii) Fe, NH
4
Cl,
EtOH/H
2
O=2/1, 80°C, 2h; (iv) TEA, Pyridine, rt, 60°C, 40%.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 76
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
.2. SAR Analysis of the Synthesized Compounds.
The SAR analysis mainly focused on discussing the effects of three regions (Figure 1,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Regions I, II and III) of the synthesized compounds on the bioactivity. Here, kinase inhibitory
activity against both ROCK I and II was used for the SAR analysis. Given that protein kinase A
(
PKA) is a closely related kinase to ROCKs, we also tested the kinase inhibitory activity of our
compounds against PKA.
.2.1. Influence of Region I
First, we investigated the influence of different moieties in region I (R in Table 1) on the
2
1
bioactivity. Thirteen compounds (12a-m) were synthesized and their bioactivities are shown in
Table 1. Compound 12a with 6-position substitution significantly decreased the ROCKs
1
inhibitory activity (12a vs 1). Replacement of the original R group by substituted phenyl groups
led to a slight increase in potency (12b-f), with the exception of dimethoxybenzene (12d).
Compounds
12g
and
12h,
which
contain
benzo[d][1,3]dioxole
and
2,3-
1
dihydrobenzo[b][1,4]dioxine at the R position, respectively, did not show obvious activity
1
against ROCKs. In contrast, compound 12i with chroman-4-one at the R position displayed a
high potency against ROCKs, and 4H-chromen-4-one replacement further improved the potency
(12j). 12k with 6-position substitution in 4H-chromen-4-one completely lost ROCKs activity. A
methyl substitution at the 3- or 8-position in 4H-chromen-4-one also significantly decreased the
ROCK inhibitory potency (12l and 12m). It is worth mentioning that all compounds had no
obvious inhibitory activity against PKA. Overall, among the thirteen compounds, 12j, which
1
contains the 4H-chromen-4-one group at the R position, was the most potent one. We thus fixed
1
the R position as 4H-chromen-4-one in the following optimization steps.
1
0
ACS Paragon Plus Environment

Page 11 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 1. Kinase Inhibitory Activities of Compounds 1 and 12a-m against ROCK I, ROCK II and
PKA.
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
IC50 (μM)a
R1
Compd
ROCK I
>10
ROCK II
PKA
1
3.141±0.177 >10
9.974±0.109 >10
5.738±0.125 >10
1
2a
>10
>10
1
2b
1
2c
3.741±0.078 0.306±0.066 >10
12d
>10
>10
>10
1
2e
1.820±0.125 0.652±0.094 >10
0.445±0.062 0.098±0.025 >10
1
2f
12g
>10
>10
>10
>10
>10
>10
1
2h
1
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 76
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
2i
0.144±0.062 0.085±0.037 >10
0
1
2
3
4
5
6
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
12j
0.014±0.009 0.003±0.001 >10
1
2k
>10
>10
>10
>10
1
2l
2.149±0.049 >10
4.163±0110 >10
1
2m
>10
a
IC50 values were determined from Kinase Profiler of Eurofins. The data are averages of two
independent experiments reported as the mean ± SD.
2
.2.2. Impact of Region II.
To explore the possible influence of region II, we replaced 4-(dimethylamino)benzamide
with various amides, benzamides, benzsulfamides and penta-heterocyclic amide groups. Table 2
summarizes the inhibitory activities of the 30 resulting compounds against ROCKs and PKA
(13-38 and 40-43). The ROCK inhibitory activities of compounds 13 and 14, which contain a
chain amide group, were significantly reduced compared with that of compound 12j. Compounds
1
3-33 and 40-43 with a substituted benzamide or penta-heterocyclic amide group at region II
showed considerable potency against ROCKs but relatively weaker activity than 12j. We then
replaced the amide group with a sulfamide, which resulted in a decrease in ROCK inhibitory
activity. Similar to before, all the compounds here showed no activity or very weak activity
1
2
ACS Paragon Plus Environment

Page 13 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
against PKA. Collectively, 12j was still the most potent compound against ROCKs. We thus
fixed region II as the original 4-(dimethylamino)benzamide in the following structural
optimization process.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Kinase Inhibitory Activities of Compounds 13-38 and 40-43 against ROCK I, ROCK II
and PKA.
IC50 (μM)a
Compd
R2
ROCK I
ROCK II
PKA
>10
1
2j
0.014±0.009
0.003±0.001
1
3
4
5
2.502±0.081
1.431±0.101
0.136±0.050
0.176±0.068
>10
>10
1
1
0.120±0.078
0.183±0.095
0.016±0.010 3.255±0.110
16
17
0.022±0.008
>10
0.235±0.088
0.143±0.036
0.165±0.055
0.024±0.012
0.014±0.008
0.034±0.011
>10
>10
>10
1
1
8
9
2
2
0
1
0.129±0.032
0.303±0.060
0.054±0.020
0.049±019
>10
>10
1
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 76
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
22
0.123±0.080
0.015±0.006
>10
23
24
25
26
0.018±0.012
0.017±0.009
0.869±0.078
0.067±0.019
0.007±0.003
0.009±0.001
0.077±0.025
>10
>10
>10
0
1
2
3
4
5
6
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
0.015±0.004 9.345±0.139
2
2
7
8
0.072±0.013
0.528±0.110
0.029±0.005
0.041±0.025
>10
>10
2
9
0.460±0.098
0.101±0.069
0.043±0.014
0.005±0.001
>10
>10
30
3
3
3
1
2
3
0.473±0.111
0.160±0.074
0.068±0.027
0.033±0.021
0.010±0.008
0.005±0.002
>10
>10
>10
4
4
0
1
0.532±0.096
0.802±0.125
0.066±0.012
0.083±0.033
>10
>10
4
2
0.670±0.121
0.058±0.098
>10
1
4
ACS Paragon Plus Environment

Page 15 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
3
6.892±1.312
>10
2.027±1.020
1.345±0.236
>10
>10
34
35
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
0.803±0.196
0.263±0.085
>10
3
6
>10
>10
0.899±0.159
6.894±1.258
>10
>10
37
3
8
6.443±1.202
1.283±0.981
>10
a
IC50 values were determined from Kinase Profiler of Eurofins. The data are averages of two
independent experiments reported as the mean ± SD.
2
.2.3. Effect of Region III
Finally, we optimized region III. A total of 13 compounds with different substituted
benzene rings at region III were synthesized. Chemical structures and bioactivities of these
compounds are shown in Table 3. All compounds with a substituted benzene ring (47a-k)
showed decreased activity against ROCK I and II. Lengthening the region Ⅲ moiety by inserting
a methylene led to a negative effect (47l). Finally, we replaced the amide group with sulfamide,
and the test result showed that the ROCK inhibitory potency was still not improved. Again, all of
the compounds did not show activity against PKA.
In summary, we obtained a series of 4H-chromen-4-one derivatives as a new class of
ROCK inhibitors. SAR analyses led to the discovery of a number of compounds that showed
potent activity against both ROCK I and ROCK II, including 12j, 23, 24, 30, 32 and 33.
Compound 12j was the most potent one. Further in-depth studies including kinase selectivity, in
1
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 76
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
vitro and ex vivo neuronal protection bioactivity, anti-cell migration and regress vascular vessel
effects, anti-ROS and anti-Müller glia cell division were carried out with compound 12j.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 3. Kinase Inhibitory Activities of Compounds 47a-l and 50 against ROCK I, ROCK II and
PKA.
IC50 (μM)a
Compd
Region Ⅲ
ROCK I
ROCK II
PKA
>10
1
2j
0.014±0.009 0.003±0.001
1.038±0.498 0.169±0.039
47a
>10
>10
4
7b
>10
>10
3.526±1.265
>10
47c
>10
>10
4
7d
>10
>10
47e
0.049±0.009 0.022±0.010
>10
1
6
ACS Paragon Plus Environment

Page 17 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
47f
>10
>10
4.557±1.253
>10
>10
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
47g
>10
4
7h
3.951±1.001 0.374±0.197
1.052±0.166 0.302±0.065
>10
>10
>10
>10
4
7i
47j
>10
4.206±0.921
H
N
F3C
4
7k
>10
>10
7.102±1.654
0.999±0.095
O
N
H
4
7l
>10
>10
5
0
0.117±0.056 0.027±0.004
a
IC50 values were determined from Kinase Profiler of Eurofins. The data are averages of two
independent experiments reported as the mean ± SD.
2
.3. Selectivity of Compound 12j
To examine the kinome-wide selectivity of 12j, we tested the kinase inhibition profile of
this compound at 10 µM against a panel of 394 recombinant human protein kinases. The results
are shown in Table S2 (see Supporting Information for details). Kinases that had a high
inhibition rate at 10 μM were further tested, and the results are summarized in Table 4. The
overall kinase selectivity profile is shown as a human kinome dendrogram in Figure 2.
1
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 76
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
Obviously, 12j displayed excellent selectivity for ROCK I (IC50=0.014 µM) and ROCK II
(
IC50=0.003 µM). It also exhibited high potency against LIMK2 with an IC50 value of 0.046 µM;
LIMK2 is the direct downstream kinase of the ROCKs. In addition, 12j possessed moderate
activity against several other kinases including Aurora-A, Aurora-B, PKG1α and PKG1β, with
IC50 values greater than 0.500 µM. Most importantly, 12j did not show obvious activity against
the other 387 tested kinases (IC50 > 10 μM). The calculated selectivity index S(20) was 0.020.
Additionally, we tested the activity of 12j against 7 unrelated protein targets by differential
scanning fluorimetry (DSF) and examined the anti-proliferation activity of 12j against 15 tumor
cell lines, in which ROCKs are not abnormally activated. 12j exhibited no obvious activity
against these protein targets and did not show anti-proliferation potency against the 15 tumor cell
lines (see Table S3 and Table S4 for details). All of these data indicate that 12j has an ideal
selectivity for ROCKs and imply that it is not a Pan Assay Interference Compound (PAIN).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Kinase selectivity of 12j shown on the human kinome dendrogram. The inhibition
rates were determined using the Kinase Profiler of Eurofins.
1
8
ACS Paragon Plus Environment

Page 19 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 4. IC50 Values of Compound 12j against Selected Kinases a.
Kinase
ROCK Ⅰ
ROCK Ⅱ
LIMK2
IC50 (µM)
0.014
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
0.003
0.046
Aurora-A
Aurora-B
PKG1α
1.072
1.239
0.517
PKG1β
0.660
a
IC50 values were determined from Kinase Profiler of Eurofins. The ATP concentrations used
were 10 µM.
2
.4. Binding Mode of 12j with ROCK II
Molecular docking was used to predict the binding mode of 12j with ROCK II. Here,
GOLD was adopted for the molecular docking, and the receptor structure was taken from the
crystal structure of ROCK II (PDB code: 4WOT). The predicted binding mode of 12j with
ROCK II is shown in Figure 3. Obviously, 12j occupies the ATP binding site, and numerous
interactions contribute to the on-target activity. The N,N-dimethylaniline structure forms two π-
stacking interactions with Phe103 and Phe136. The intermediate phenylenediamine group forms
a π-cation interaction with Lys121, and the carbamoyl groups on both sides of the intermediate
group form three hydrogen bonds with residues Phe103, Asp232 and Lys121. These data explain
the results indicating substitution of intermediate region III led to a decrease in activity against
ROCK II. The chromone structure is located exactly in a terminal pocket and limits the size and
conformation of this region. In particular, the carbonyl of the chromone formed a hydrogen bond
with residue Met172. This bond is very important for stabilizing the orientation and
conformation of 12j, which explains why the carbonyl is critical for the inhibitory activity of 12j
1
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 76
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
against ROCK II. Together, these results revealed the mechanism by which 12j potently and
selectively inhibits the kinase activity of ROCK II.
0
1
2
3
4
5
6
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. Predicted binding mode of 12j with ROCK II. Dashed yellow lines indicate the
hydrogen bonding interactions. The crystal structure of the ROCK II was taken from the RCSB
Protein Data Bank (PDB entry: 4WOT)
2
.5. Compound 12j Inhibits ROCKs Activity in Vitro
We then examined the inhibitory effect of 12j on ROCK activity in intact cells. In these
assays, the SH-SY5Y cell line was used, which is a kind of human neuroblastoma cell line.
Western blot assay was adopted to measure the expression levels of MYPT1 and its
phosphorylated form (p-MYPT1); MYPT1 is a direct downstream signaling protein of ROCKs.
As shown in Figure 4A and 4B, 12j treatment dose-dependently (Figure 4A) and time-
dependently (Figure 4B) inhibited the phosphorylation of MYPT1 but did not impact the
MYPT1 expression. These results indicated that 12j could efficiently suppress the ROCK
signaling in intact cells. One may notice that 12j did not display a high potency in cellular assays
compared to that in enzymatic assays. Two factors might be responsible for this result. The first
2
0
ACS Paragon Plus Environment

Page 21 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
is that the water solubility and cell membrane permeability are not good for compound 12j. The
second is that the cell treatment time of cells by 12j is short in these assays (less than 4 h).
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
2
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 76
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
2
2
ACS Paragon Plus Environment

Page 23 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 4. 12j inhibited ROCK kinase activity in SH-SY5Y cells. (A) SH-SY5Y cells were
treated with various concentrations of 12j (0.1, 1 and 10 μM) for 2 h. Then, whole cell lysates
were subjected to an immunoblot assay to detect p-MYPT-1 and MYPT-1. Quantification of
immunoblots is presented in the lower panel. (B) SH-SY5Y cells were treated with 12j (10 μM),
and the levels of p-MYPT-1 and MYPT-1 were detected after 0.25, 1, 2 and 4 h. Quantification
of immunoblots is presented in the lower panel.
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
2.6. Compound 12j Inhibited Endothelial Cell Migration in Vitro
Endothelial cell migration is necessary for retinal microvascular angiogenesis. Deprivation
of ROCK activity in retinal endothelial cells promotes the depolymerization of actin filaments
and inhibits stress fiber formation, thus preventing endothelial migration and reducing
microvascular damage. We examined the inhibitory effect of 12j on endothelial migration by
transwell assay and scratch-wound assay. The transwell assay showed that the migrating cell
numbers were significantly reduced in the group treated with 12j compared with the group
treated with vehicle (Figure 5A). The scratch-wound assay showed that HUVEC treatment with
1
2j could significantly reduce the rate of wound healing at 24 h and 36 h (Figure 5B). These
results indicated that 12j inhibited endothelial cell migration in vitro.
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. 12j inhibited endothelial cell migration in vitro. (A) HUVEC cells were treated with
1
2j after 24 h, and the migrated cells were stained and detected by light microscopy at 40
×
and
2
00×. Quantification of migrated cells is presented in the right panel. (B) HUVEC cells were
treated with 12j and serum-free medium. The rate of wound healing was observed by light
microscopy at 200 after 24 h and 36 h (*** P < 0.001, t test).
×
2
.7. Compound 12j Rescues High Glucose-Induced Neuronal Cell Death and Decreases
ROS in Vitro and ex Vivo
ROCKs have been reported to be involved in the regulation of neuronal cell survival in
14
neurological disorders. We thus tested the effect of 12j on high glucose-induced neuronal cell
death; here SH-SY5Y cells were used. MTT assays showed that 12j significantly protected the
2
4
ACS Paragon Plus Environment

Page 25 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
cells from death, which was induced by 35 mM high glucose treatment for 5 days (Figure 6). We
then evaluated the ability of 12j to protect high glucose-induced retinal neuronal cell death in an
ex vivo DR model. Cell death was detected by marked cleaved caspase 3 and TUNEL. Neuronal
cells were marked by Chx10, which is a marker of bipolar cells (a kind of neuronal cells in
retina).20 To exclude the possibility that the neuronal cell death in HG medium was due to
osmotic changes, we replaced 35 mM glucose with mannose (GM) in the medium at the same
concentration. As depicted in Figure 7, neuronal cell death was caused by glucose exposure
rather than osmotic changes, and treatment with 12j significantly protected neuronal cells from
death.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. 12j protected high glucose-induced neuronal cell death. SH-SY5Y cells were cultured
for 5 days under the indicated conditions. Control: normal glucose control (7.5 mM); HG: high
glucose (35 mM); HG+12j: the media contained high glucose (35 mM) and 12j (1 μM). (** P <
0
.01 and * P < 0.05, t test).
2
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 76
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 7.12j exerted a neuroprotective effect in an ex vivo DR model. Sections from P8 retinal
explants were cultured for 5 days under the indicated conditions and stained for nuclei (DAPI,
blue), bipolar cells (Chx10, green), apoptosis (active caspase 3, red), and cell death (TUNEL,
+
+
2
green). Quantification of TUNEL cells and Chx10 /active caspase3 dying bipolar cells per mm
retinal section area is presented in the right panel. Error bars represent SD of measurements from
at least three animals and asterisks indicate significant differences between retinas of the NG
group and those of the other indicated groups (*, P < 0.05, t test). Scale bar= 50 μm. NG: normal
glucose control (7.5 mM); GM: osmotic control (7.5 mM glucose+ 27.5 mM mannitol); HG:
high glucose (35 mM); HG+12j: the media contained high glucose (35 mM) and 12j (1 μM).
ONL: outer nuclear layer; INL: inner nuclear layer; GCL: ganglion cell layer.
It is widely accepted that reactive oxygen species (ROS) leaking from mitochondria causes
oxidative damage to biomolecules (such as DNA, lipids and proteins) and cell death.21
Mitochondria-derived death signaling has been reported as an important pathway for retinal
22
ganglion cell (RGC) death. To confirm whether the ROCK inhibitor 12j could protect neurons
from oxidative stress, we examined the ROS levels by using the ROS detection probes DCF and
DHE in the ex vivo model. As shown in Figure 8, retinas exposed to high glucose displayed a
high level of ROS, while 12j significantly decreased the level of ROS. These data suggest that
1
2j could suppress high glucose-induced oxidative stress.
2
6
ACS Paragon Plus Environment

Page 27 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 8. 12j suppressed high glucose-induced ROS overproduction. Sections from P8 retinal
explants were cultured for 5 days under the indicated condition and stained for nuclei (DAPI,
blue), ROS (DCF, green; DHE, red). Quantification of fluorescence intensity (OD) in DHE (INL
2
and GCL) and DCF (through the retina) per 150×150 μm retinal section area is presented in the
lower panel. Error bars represent SD of measurements from at least three animals, and asterisks
indicate significant differences between retinas of the NG group and those of the other indicated
groups (*, P < 0.05, t test). Scale bar= 50 μm. NG: normal glucose control (7.5 mM); GM:
osmotic control (7.5 mM glucose+ 27.5 mM mannitol); HG: high glucose (35 mM); HG+12j: the
media contained high glucose (35 mM) and 12j (1 μM). INL: inner nuclear layer; GCL: ganglion
cell layer; ONL: outer nuclear layer.
2
.8. Compound 12j Inhibits High Glucose-Induced Müller Glia Division ex Vivo
Reactive gliosis plays a decisive role in the pathological course of neuronal damage. It has
2
3
been reported that high glucose can induce Müller glia cell proliferation and gliosis in DR.24
Therefore, we examined the effect of 12j on high glucose-induced gliosis. Müller glia cells were
detected by glutamine synthetase (GS) labeling, and cell proliferation was marked with phospho-
histone H3S10 (PH3), which is a marker of mitosis. The results showed that the number of retina
2
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 76
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
＋
＋
cells labeled with both PH3 and GS increased significantly in HG medium, indicating that
high glucose induced the proliferation of Müller glia cells (Figure 9). Treatment with 12j
significantly inhibited the proliferation of Müller glia cells (Figure 9). Collectively, the data
presented here indicated that compound 12j could suppress gliosis in response to a high glucose
microenvironment.
0
1
2
3
4
5
6
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 9. 12j suppressed high glucose-induced Müller glia division in retinal explants. Sections
from P8 retinal explants were cultured for 5 days under the indicated conditions and stained for
nuclei (DAPI, blue), Müller glia cells (GS, green), and phospho-histone H3S10 (PH3, red).
+
+
2
Quantification of GS /PH3 dividing Müller glia cells per mm retinal section area is presented
in the lower panel. Error bars represent the SD of measurements from at least three animals and
asterisks indicate significant differences between retinas of the NG group and those of the other
indicated groups (*, P < 0.05, t test). Scale bar= 50 μm. NG: normal glucose control (7.5 mM);
GM: osmotic control (7.5 mM glucose+27.5 mM Mannitol); HG: high glucose (35 mM);
HG+12j: the media contained high glucose (35 mM) and 12j (1 μM). ONL: outer nuclear layer;
INL: inner nuclear layer; GCL: ganglion cell layer.
2
.9. Compound 12j Suppresses High Glucose-Induced Microvascular Changes ex Vivo
2
8
ACS Paragon Plus Environment

Page 29 of 76
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
DR has nonproliferative and proliferative stages. The nonproliferative stage induces
microvascular injury, leading to retinal ischemia and hypoxia. The latter changes lead to an
increase of proangiogenic factors, disrupting angiogenic homeostasis and facilitating the
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
2
proliferation of blood vessels, which are characteristics of proliferative DR. However, these
hallmarks can never be reproduced in rodent DR in vivo models.24
2
5
In our previous study , we found that ex vivo retinal explants cultured from post natal day 8
to 15 (P8-15) under normal glucose or osmotic control conditions only had scattered IB4-
positive endothelial cells without any vascular plexus, suggesting hyperoxia induced vascular
2
6
regression; however, explants under high glucose had intact superficial primary and deep plexi,
indicating high glucose can inhibit hyperoxia-induced blood vessel regression, similar to the
).25
effects of the hypoxia-inducer cobalt chloride (CoCl
2
We thus examined the effect of 12j on vascular regression in retinal explants. Here vascular
endothelial cells were labeled by IB4. We found that the blood vessel regression of retinal
explants in HG medium was reversed by 12j treatment (Figure 10). This suggested that 12j had
an efficient suppressive effect on high glucose-induced microvascular changes.
The ability of 12j to reduce retinal ROS (Figure 8) and promote vascular regression (Figure
1
0) is similar to the effect of the ROCK inhibitor ripasudil (K-115) on retinal angiogenesis and
hypoxia; K-115 inhibited VEGF-induced human retinal microvascular endothelial cell
HRMECs) migration and proliferation and hypoxic retinal area in oxygen induced retinopathy
(
2
7
(OIR) mice. Thus, 12j has therapeutic potential in the treatment of retinal hypoxic neovascular
diseases such as DR.
2
9
ACS Paragon Plus Environment