Discovery, synthesis and anti-atherosclerotic activities of a novel selective sphingomyelin synthase 2 inhibitor

Article abstract of DOI:10.1016/j.ejmech.2018.12.028

The sphingomyelin synthase 2 (SMS2) is a potential target for pharmacological intervention in atherosclerosis. However, so far, few selective SMS2 inhibitors and their pharmacological activities were reported. In this study, a class of 2-benzyloxybenzamides were discovered as novel SMS2 inhibitors through scaffold hopping and structural optimization. Among them, Ly93 as one of the most potent inhibitors exhibited IC₅₀ values of 91 nM and 133.9 μM against purified SMS2 and SMS1 respectively. The selectivity ratio of Ly93 was more than 1400-fold for purified SMS2 over SMS1. The in vitro studies indicated that Ly93 not only dose-dependently diminished apoB secretion from Huh7 cells, but also significantly reduced the SMS activity and increased cholesterol efflux from macrophages. Meanwhile, Ly93 inhibited the secretion of LPS-mediated pro-inflammatory cytokine and chemokine in macrophages. The pharmacokinetic profiles of Ly93 performed on C57BL/6J mice demonstrated that Ly93 was orally efficacious. As a potent selective SMS2 inhibitor, Ly93 significantly decreased the plasma SM levels of C57BL/6J mice. Furthermore, Ly93 was capable of dose-dependently attenuating the atherosclerotic lesions in the root and the entire aorta as well as macrophage content in lesions, in apolipoprotein E gene knockout mice treated with Ly93. In conclusion, we discovered a novel selective SMS2 inhibitor Ly93 and demonstrated its anti-atherosclerotic activities in vivo. The preliminary molecular mechanism-of-action studies revealed its function in lipid homeostasis and inflammation process, which indicated that the selective inhibition of SMS2 would be a promising treatment for atherosclerosis.

Full text of DOI:10.1016/j.ejmech.2018.12.028

Accepted Manuscript
Discovery, synthesis and anti-atherosclerotic activities of a novel selective
sphingomyelin synthase 2 inhibitor
Yali Li, Taomin Huang, Bin Lou, Deyong Ye, Xiangyu Qi, Xiaoxia Li, Shuang Hu,
Tingbo Ding, Yan Chen, Yang Cao, Mingguang Mo, Jibin Dong, Min Wei, Yong Chu,
Huiti Li, Xian-Cheng Jiang, Nengneng Cheng, Lu Zhou
PII:
S0223-5234(18)31065-1
DOI:
https://doi.org/10.1016/j.ejmech.2018.12.028
EJMECH 10961
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 31 August 2018
Revised Date: 11 December 2018
Accepted Date: 12 December 2018
Please cite this article as: Y. Li, T. Huang, B. Lou, D. Ye, X. Qi, X. Li, S. Hu, T. Ding, Y. Chen, Y. Cao,
M. Mo, J. Dong, M. Wei, Y. Chu, H. Li, X.-C. Jiang, N. Cheng, L. Zhou, Discovery, synthesis and anti-
atherosclerotic activities of a novel selective sphingomyelin synthase 2 inhibitor, European Journal of
Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2018.12.028.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Discovery, synthesis and anti-atherosclerotic activities of a novel selective
sphingomyelin synthase 2 inhibitor
a,b,#
a,c,#
a
a, #
a
a
a
a
a
a
Yali Li
, Taomin Huang
a
, Bin Lou , Deyong Ye , Xiangyu Qi , Xiaoxia Li , Shuang Hu , Tingbo Ding , Yan Chen ,
a
a
a
a
a,d,
a,
Yang Cao , Mingguang Mo , Jibin Dong , Min Wei , Yong Chu , Huiti Li , Xian-Cheng Jiang *, Nengneng Cheng *,
a,
Lu Zhou *
a
b
School of Pharmacy, Fudan University, No.826, Zhangheng Rd., Shanghai 201203, China
Institute for Agro-food Standards and Testing Technology, Shanghai Academy of Agricultural Sciences, No.1000, Jinqi Rd.,
Shanghai 201403, China
c
Eye Ear Nose Throat Hospital, Fudan University, No. 83, Fenyang Rd., Shanghai 200031, China
d
State University of New York Downstate Medical Center, Brooklyn, New York 11203, USA
Graphical abstracts
Vehicle
Low dose
High dose
Discovery of a 2-benzyloxybenzamide Ly93 as a novel selective SMS2 inhibitor and the studies on
its in vitro and in vivo biological activities
1
/ 1

ACCEPTED MANUSCRIPT
1
2
Discovery, synthesis and anti-atherosclerotic activities of a novel
selective sphingomyelin synthase 2 inhibitor
a,b,#
a,c,#
a, #
a
a
3
4
5
6
7
Yali Li
, Taomin Huang
, Bin Lou , Deyong Ye , Xiangyu Qi , Xiaoxia
a
a
a
a
a
a
Li , Shuang Hu , Tingbo Ding , Yan Chen , Yang Cao , Mingguang Mo ,
a
a
a
a
a,d,
Jibin Dong , Min Wei , Yong Chu , Huiti Li , Xian-Cheng Jiang
*,
a,
a,
Nengneng Cheng *, Lu Zhou *
a
8
9
0
1
2
School of Pharmacy, Fudan University, No.826, Zhangheng Rd., Shanghai 201203, China
b
Institute for Agro-food Standards and Testing Technology, Shanghai Academy of Agricultural
1
1
1
Sciences, No.1000, Jinqi Rd., Shanghai 201403, China
c
Eye Ear Nose Throat Hospital, Fudan University, No. 83, Fenyang Rd., Shanghai 200031, China
d
State University of New York Downstate Medical Center, Brooklyn, New York 11203, USA
1
3
1
4
1
5
1
1
6
7
ABBREVIATIONS
18
19
20
21
SM, sphingomyelin; SMS, sphingomyelin synthase; SMS2, sphingomyelin synthase
2; PC, phosphatidylcholine; SPT, serine palmitoyltransferase; AS, atherosclerosis;
IC , 50% inhibiting concentration; apoB, apolipoprotein B; apoE, apolipoprotein E;
5
0
KO, knock out; HPLC, high-performance liquid chromatography.
2
2
2
2
2
2
2
3
4
5
6
7
#
These authors contributed equally to this work. *Correspondence should be addressed to: Tel.
+1-718 270 6701. E-mail address: xjiang@downstate.edu (X Jiang); E-mail address:
chengnn@163.com (N Cheng); Tel.
zhoulu@fudan.edu.cn (L Zhou).
&
Fax: +86-21-51980125. E-mail address:
1
/ 62

ACCEPTED MANUSCRIPT
1
Abstract
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
The sphingomyelin synthase 2 (SMS2) is a potential target for pharmacological
intervention in atherosclerosis. However, so far, few selective SMS2 inhibitors and
their pharmacological activities were reported. In this study, a class of
2-benzyloxybenzamides were discovered as novel SMS2 inhibitors through scaffold
hopping and structural optimization. Among them, Ly93 as one of the most potent
inhibitors exhibited IC values of 91 nM and 133.9 µM against purified SMS2 and
5
0
SMS1 respectively. The selectivity ratio of Ly93 was more than 1400-fold for
purified SMS2 over SMS1. The in vitro studies indicated that Ly93 not only
dose-dependently diminished apoB secretion from Huh7 cells, but also significantly
reduced the SMS activity and increased cholesterol efflux from macrophages.
Meanwhile, Ly93 inhibited the secretion of LPS-mediated pro-inflammatory cytokine
and chemokine in macrophages. The pharmacokinetic profiles of Ly93 performed on
C57BL/6J mice demonstrated that Ly93 was orally efficacious. As a potent selective
SMS2 inhibitor, Ly93 significantly decreased the plasma SM levels of C57BL/6J
mice. Furthermore, Ly93 was capable of dose-dependently attenuating the
atherosclerotic lesions in the root and the entire aorta as well as macrophage content
in lesions, in apolipoprotein E gene knockout mice treated with Ly93. In conclusion,
we discovered a novel selective SMS2 inhibitor Ly93 and demonstrated its
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
anti-atherosclerotic
activities
in
vivo.
The
preliminary
molecular
mechanism-of-action studies revealed its function in lipid homeostasis and
inflammation process, which indicated that the selective inhibition of SMS2 would be
a promising treatment for atherosclerosis.
2
5
6
Keywords:
2
Sphingomyelin synthase 2; Inhibitor; ApoE KO mice; Atherosclerosis.
2
/ 62

ACCEPTED MANUSCRIPT
1
1. Introduction
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
Sphingomyelin (SM) is one of the major phospholipids in the circulation [1]. It
has been proved that human plasma SM levels are an independent risk factor for
coronary heart disease [2, 3] and prognostic in patients with acute coronary syndrome
[3]. In apoE KO mice, the plasma SM levels that are four-fold higher than those in
control mice [4]. The raised plasma SM could further induce the increased
atherosclerosis found in these animals [5]. Meanwhile, SM that are enriched in
atherogenic lipoproteins has significant effects on the metabolism of apoB-containing
lipoproteins [1, 6]. The deficiency of SM in atherogenic lipoproteins could reduce
the atherogenic properties of the mice [7].
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
Moreover, SM is also one of the major phospholipids in the plasma membrane.
Macrophages, as the most prominent cell types in atherosclerotic lesions, are
associated with two hallmarks of atherosclerosis involving the formation of foam
cells due to excessive accumulation of cholesterol by macrophages [8] and
inflammation [1, 9]. Studies on the membrane SM of macrophages revealed that
membrane SM could play an important role in the development of atherosclerosis
[10-12]. The SM level of macrophage membrane is closely related to the cholesterol
efflux and inflammatory responses in macrophages [10, 13]. Thus, the inhibition of
SM biosynthesis could directly reduce SM levels not only in atherogenic lipoproteins
but also in the membrane of macrophages.
The reduction of SM, with concomitant reduction of atherosclerosis, could be
achieved by pharmacologically inhibiting serine palmitoyltransferase (SPT) in mice
models, which is the first enzyme for SM biosynthesis [14, 15]. However, the
inhibition of SPT could influence the whole sphingolipid de novo synthesis pathway
and induce many off-target side effects. Therefore, the inhibition of sphingomyelin
synthase (SMS) is an alternative approach for the reduction of SM.
SMS is the last enzyme in the SM biosynthetic pathway. It catalyzes the
3
/ 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
conversion of ceramide to SM. The SMS gene family consists of three members
including SMS1, SMS2 and SMS related protein (SMSr). Both SMS1 and SMS2, but
not SMSr, have the SMS catalytic activity [16, 17]. In our previous studies, SMS1 and
SMS2 are proved to be the two factors that influence plasma SM levels [7, 11, 18].
Both SMS1 and SMS2 KO mice have lower SM levels in all the tested tissues [11].
However, in the SMS1 knockdown T cells, the TCR signaling was impaired through
the dysfunction of lipid rafts [19]. Meanwhile, the SMS1 KO mice exhibited
moderate neonatal lethality [11, 20], mitochondrial dysfunction and the increased
oxidative stress, which could cause lipodystrophy and impair the insulin secretion
from pancreas [20, 21]. Furthermore, the deficiency of SMS1 but not SMS2 could
cause hearing impairments in mice. As to the side effects of SMS1 deficiency, the
inhibition of SMS2 is an optimal strategy for the reduction of SM level.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
Our studies have already indicated that the SMS overexpression could promote
the accumulation of atherogenic lipoprotein and increase atherogenic potential [22].
Contrarily, the SMS deficiency reduces the SM accumulation and thereby alleviates
atherosclerosis in a mouse model [10-12]. These intriguing findings raise the
possibility that SMS2 is a potential therapeutic target of atherosclerosis. The selective
SMS2 inhibitors are likely to be developed as anti-atherosclerotic drugs. However,
the impact of selective SMS2 inhibition on the development of atherosclerosis has
not been investigated so far due to the lack of potent selective SMS2 inhibitors.
Up to date, very few selective SMS2 inhibitors have been reported in the
literature. D609 (Fig. 1), firstly discovered as an anti-tumor agent [23, 24], was
reported with weak SMS inhibitory activity in vitro (IC = 177-600 µM) [25-27]. The
5
0
α-aminonitrile derivatives were the first series of SMS inhibitors discovered through
rational drug design by our research group. The representative compound D2 (Fig. 1)
has an IC value of 24.5 µM for SMS from liver homogenate of ICR mice and an IC
5
0
50
value of 13.5 µM for SMS2 over-expressed insect cell lysis [28, 29]. However, the
latently toxigenic α-aminonitrile group of compound D2 precluded the further
4
/ 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
application in the in vivo medication of atherosclerosis. SAPA derivatives (Fig. 1)
were then discovered as a novel class of SMS1 inhibitors with micromolar inhibitory
activities against SMS1 over-expressed Hela cell lysis [30]. However, the effects of
SAPA 1j on SMS2 had not been explored. Furthermore, oxazolopyridine derivatives
(Fig. 1) were designed and firstly tested the inhibitory activity towards purified
SMS2. Among them, the compound QY10 possess an IC value of 1.7 µM against
5
0
purified SMS2 enzyme [31]. Moreover, a 2-quinolone derivative (Fig. 1) was
identified as a potent SMS2 selective inhibitor (IC50 = 16 nM for SMS2
over-expressed membrane fraction), of which the in vitro and in vivo biological
activities were not reported [32]. 4-Benzyloxybenzo[d]isoxazole-3-amine
derivatives were discovered as potent SMS2 selective inhibitors. The most
promising representative compound (Fig. 1) was found to be with IC values of
1
1
1
1
5
0
0.10 µM against SMS2 and 56 µM against SMS1 [33].
CN
O
N
H
O
-
+
O
S K
N
O
O
H
S
S
O
Br
N
H
D609
D2
SAPA 1j
IC50/SMS = 177-600
MW = 266.46
M
IC50/SMS = 24.5
IC50/SMS2 = 13.5
MW = 314.39
M
M
IC50/SMS1 = 2.1
MW = 503.41
M
CF3
N
N
O
N
O
O
O
N
N
N
N
O
CF3
H
O
F
O
Cl
N
2
-quinolone derivative
IC50/SMS2 = 16 nM
4-benzyloxybenzo[d]isoxazole
QY10
-3-amine derivative
IC50/SMS2 = 1.7
IC50/SMS1 > 100
MW = 316.36
M
M
IC50/SMS1 = 14
M
IC50/SMS2 = 0.10
IC50/SMS1 = 56
M
M
selectivity ratio = 880
MW = 625.57
selectivity ratio = 560
MW = 369.78
1
4
5
1
Fig 1. Chemical structures of reported sphingomyelin synthase inhibitors.
16
In the current study, 2-(benzyloxy)-N-arylbenzamide derivatives were
5
/ 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
discovered as a novel series of SMS2 inhibitors. The representative compound Ly93
was proved with the selectivity ratio of more than 1400-fold for purified SMS2 over
SMS1. The in vitro biological evaluations were performed on Ly93 for the exploration
of the potential anti-atherosclerotic mechanisms. Finally, Ly93 was put into the in
vivo studies and exhibited significant anti-atherosclerotic activities in apoE KO mice.
The research in this paper indicated that the selective inhibition of SMS2 would be a
promising treatment for atherosclerosis.
8
2. Experiments and Results
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
2.1 The discovery of 2-benzyloxybenzamides as novel SMS2 inhibitors.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
Although the α-aminonitrile group of D2 could play a role in the inhibition of
SMS based on the docking model of hSMS1 [28], the latent toxigenicity of
α-aminonitrile precluded the further application of D2 in the in vivo medication of
atherosclerosis. Thus, as shown in Fig. 2, scaffold hopping was firstly conducted on
the α-aminonitrile group (section A) of the previously reported SMS2 inhibitor D2
[28] to eliminate the potential toxigenic effects of the lead compound D2. The
detailed approaches of scaffold hopping were shown in Supplementary Fig. 1. The
compounds derived from D2 were then validated by the SMS2 inhibition assay
using SMS2 over-expressed insect cell lysate. As shown in Supplementary Table 1,
only the compound D2-e retained the SMS2 inhibitory activity with the IC value
5
0
of 13.4 µM, while the other compounds D2-a, b, c, d were less potent. The results
indicated that the amide group of D2-e, similar to the α-aminonitrile group, could
accommodate to the steric restriction and the electronic effects of the binding site
[28]. It was also estimated that the amide group (section A) of D2-e might be vital
for the SMS2 inhibition as well as for the improved physicochemical property.
Based on the docking model of D2-e with hSMS2 which was derived from the
previously reported model of hSMS1 [34], the further chemical modifications were
performed on the section B to investigate the structure and activity relationship
6
/ 62

ACCEPTED MANUSCRIPT
1
2
3
4
(SAR). Finally, as the benzyl group was located at the position between two helixes
with space tolerance, substitutes with different electronic and steric effects were
introduced onto the benzene ring to explore the SAR of the section C and discover
more potent and selective SMS2 inhibitors.
5
6
Fig 2. The design strategy and SAR study of 2-benzyloxybenzamides as novel
SMS2 inhibitors.
7
8
9
0
1
2
3
4
5
6
7
8
2.2 The synthesis of novel 2-benzyloxybenzamides.
2-benzyloxybenzamides 6-11 were firstly designed to investigate the SAR of
the section B of the compound D2-e. Compounds 6-11 were synthesized according
to the Scheme 1. Intermediates 3a and 3b were synthesized through the nucleophilic
substitution reaction of methyl salicylate 1 with benzyl bromide 2a and 2b
respectively. Then 3a and 3b were hydrolyzed to give 2-benzyloxybenzoic acid 4a
and 4b respectively. The 4a was reacted with thionyl chloride and amine 5a, 5b, 5d or
5e sequentially to give 6, 7, 9 and 10. Compound 8 was prepared through the
condensation of 4a with cyclohexanamine 5c. Compound 11 was obtained from the
condensation of 4b with N-boc-piperidin-3-amine 5f and subsequently the
deprotection of the Boc group.
1
1
1
1
1
1
1
1
1
1
9
0
2
Scheme 1. The synthetic route for 2-benzyloxybenzamides 6-11. Reagents and
21
conditions: (a) K CO , acetone, rt; (b) 1) 4 M NaOH (aq), CH OH; 2) 2M HCl (aq);
2
3
3
7
/ 62

ACCEPTED MANUSCRIPT
o
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
(c) 1) SOCl , Pyridine (cat.), reflux; 2) Pyridine, CH Cl , 0 C to rt; (d) 1) EDCI,
2 2 2
o
DMAP, CH Cl , 0 C to rt; 2) HCl (aq, con.), EA.
2
2
2-benzyloxybenzamides 19-45 were designed to investigate the SAR of the
section C of the compound 10. Compounds 19-45 were synthesized according to the
Scheme 2. The key intermediate 12 was prepared through the debenzylation of
compound 10 under the hydrogen and catalytic palladium. 15c and 15d were
synthesized through the nucleophilic substitution of 5-chlorosalicylic aldehyde 13
with bromoalkane. Commercially available reagents 15a and 15b coupled with 15c
and 15d were reacted with sodium borohydride to give the benzyl alcohol 17a-17d.
Meanwhile, intermediate 14a was obtained from the reduction of 13 by sodium
borohydride and then with salicylic alcohol 14b to afford benzyl alcohol 17j-17n.
Commercially available benzoic acids 16a-16e were reacted with borane in
tetrahydrofuran to give benzyl alcohol 17e-17i. Benzyl bromides 18k-18x was
prepared from the bromination of benzyl alcohols 17a-17n with phosphorus
tribromide, and subsequently with the commercially available reagents 18a-18j
provided the target compounds 19-22 (the compound 22 was named as Ly93),
24-33 and 36-45 by reacting with the key intermediate 12. Finally, benzyl alcohols
17o-17q directly provided the target compounds 23, 34 and 35 through the Mitsunobu
reaction with 12.
1
1
1
1
1
1
1
1
1
1
2
0
8
/ 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
Scheme 2. The synthetic route for 2-benzyloxybenzamides 19-45. Reagents and
o
conditions: (a) K CO , CH CN, 60 C; (b) K CO , acetone, rt; (c) K CO , DMF, 100
2
3
3
2
3
2
3
o
o
o
o
C; (d) NaBH , C H OH, 0 C to rt; (e) BH , THF, 0 C to rt; (f) PBr , CH Cl , 0 C
4
2
5
3
3
2
2
o
to rt; (g) H , 10% Pd/C, CH OH, rt; (h) DEAD, Ph P, THF, 0 C to rt.
2
3
3
The structures of all newly synthesized compounds were elucidated from their
analytical and spectroscopic data which were collected in the Material and Methods
and Supplementary Material.
2.3 The in vitro SMS2 inhibitory activities of novel 2-benzyloxybenzamides.
All the newly prepared compounds were further evaluated the SMS2 inhibitory
activities on the SMS2 over-expressed insect cell lysate. The SMS2 inhibitory
activities of 2-benzyloxybenzamides were shown in Table 1. The results revealed that
27 compounds were more potent SMS2 inhibitors than D2-e (IC = 13.4 µM).
1
1
1
1
1
1
1
1
5
0
Among them, 11 compounds exhibited robust SMS2 inhibitory activities with the
IC values lower than 1 µM. Compound 45 with the IC value of 0.43 µM was most
5
0
50
effective than other 2-benzyloxybenzamides on the SMS2 inhibition.
Table 1. The SMS2 inhibitory activities of 2-benzyloxybenzamides 6-11 and
19-45.
B
O
O
O
O
SAR
G
m
N
N
H
H
R
C
2
-benzyloxybenzamides
D2-e
6-11, 19-45
1
8
SMS2
Compd.
D2
G
m
R
a
IC (µM)
5
0
13.5
9
/ 62

ACCEPTED MANUSCRIPT
D2-e
0
1
1
0
0
0
-H
-H
-H
-H
-H
-H
13.4
34.2
49.0
43.4
11.7
3.2
6
7
8
9
1
0
11
0
2-Cl
> 100
2
2
2
2
2
2
3
-F
3.5
1
2
2
2
2
2
2
9
0
0
0
0
0
0
0
-Cl
1.1
0
-CH3
1.5
1
-C H
0.88
2.1
2
5
2 (Ly93)
-CF3
-CN
-F
3
4
5
31.8
1.6
1
0 / 62

ACCEPTED MANUSCRIPT
3
3
2
2
2
2
2
2
2
2
2
2
2
2
-Cl
1.4
26
27
28
29
30
31
32
33
34
35
36
37
38
39
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-CN
60.7
0.74
1.5
-Cl, 5-F
-Cl, 6-F
,5-di-Cl
,6-di-Cl
-F, 3-Cl
0.99
0.69
3.1
-CH , 5-F
0.74
1.1
3
-CH , 5-Cl
3
-CH , 3-Cl
3.8
3
,6-di-CH3
0.89
2.8
,5-di-OCH3
-OCH , 5-Cl
0.67
11.6
3
-O(CH ) OCH , 5-Cl
2 2
3
11 / 62

ACCEPTED MANUSCRIPT
2
2
2
2
2
2
-O(CH ) OCH , 5-Cl
4.1
4
4
4
4
4
0
1
2
3
4
5
0
0
0
0
0
0
2 3
3
-O(CH ) Cl
1.3
2 4
-O(CH ) Cl
0.67
0.47
0.52
0.43
2 5
-O(CH ) Cl
2 6
-O(CH ) CH , 5-Cl
2 5
3
-O(CH ) CH , 5-Cl
4
2 6
3
a
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
IC values are the means of three separate determinations on SMS2 over-expressed insect cell
50
lysate and were determined by more than five concentrations of each inhibitor. Statistical
calculation of IC values was performed on GraphPad Prism 5.02 (GraphPad Software, Inc.).
5
0
2.4 The SARs of 2-benzyloxybenzamides as novel SMS2 inhibitors.
In order to explore the interesting enzymatic profile, a preliminary
structure-activity relationship of 2-benzyloxybenzamides was analyzed on the basis
of in vitro enzymatic activities.
The SMS2 inhibitory activities were dramatically reduced by not only the
introduction of methylene (6 vs D2-e) but also the substitution of phenyl with a
saturated ring (8 vs D2-e). Compounds 9 and 10 had improved SMS2 inhibitory
activities due to the substitution of phenyl with pyridyl. The aromatic group G was
proved to be crucial for SMS2 inhibition by comparison of 10 with 11.
Substituent group R was also important for the SMS2 inhibitory activities of
2-benzyloxybenzamides. The SMS2 inhibitory activities were enhanced by the
introduction of a chlorine atom (20) or an alkyl group (21, 22) onto the ortho-position
as well as halogen atoms (25, 26) onto the meta-position. However, the presence of
1
1
1
1
1
1
1
1
2 / 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
ortho- or meta- cyano group (24, 27) could dramatically weaken the inhibitory
activities toward SMS2. Furthermore, the SMS2 inhibitory potency could be further
elevated by the dual substitution of halogen atoms or alkyl groups at 2,5- and 2,6-
positions. It exhibited the superimposed effects of advantaged groups. The
hydrophobic chain of C4-C7 alkoxy group on the ortho-position (41-45) were also the
advantaged groups for improving the SMS2 inhibitory potency.
The structure-activity relationships achieved by 2-benzyloxybenzamides would
be helpful for developing more potent SMS2 inhibitors in the future.
2.5 The in vitro inhibitory activities and selectivity of Ly93 against purified SMS1
and SMS2.
1
1
1
1
1
1
1
1
1
1
2
2
2
Among the 2-benzyloxybenzamides derivatives shown above, the novel small
molecule Ly93 was designed, synthesized and validated (its structural spectra were
shown in Supplementary Fig. 2), and showed to be one of the most potent SMS2
inhibitors. Due to the distinguished SMS2 inhibitory activity and optimal properties,
Ly93 was selected for the further studies. As SMS1 and SMS2 purified enzymes
were expressed (unpublished work from Prof Yu Cao), it was possible for us to
evaluate the in vitro inhibitory activities and selectivity of compound Ly93 against
purified SMS1 and SMS2. As a result, as shown in Fig. 3, Ly93 exhibited
dose-dependent SMS inhibitory activities toward purified SMS1 and SMS2. The IC50
values of Ly93 were 91 nM and 133.9 µM toward purified SMS2 and SMS1
respectively. The selectivity ratio of Ly93 was > 1400-fold for purified SMS2 over
SMS1. Therefore, Ly93 was proved to be a novel potent selective SMS2 inhibitor.
A
B
C
1
20
00
1
20
00
1
20
00
1
SMS1
SMS2
1
1
80
60
40
20
8
6
4
0
0
0
8
6
4
2
0
0
0
0
0
20
0
-4
-3
-2
-1
0
1
2
3
Vehicle 15
100
300
Vehicle 12
60
300
Ly93 (µM)
Ly93 (nM)
lg[con Ly93](µM)
2
3
1
3 / 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
Fig 3. The inhibitory activities of Ly93 against purified SMS1 and SMS2 in vitro.
(A) The inhibitory activity of Ly93 against purified SMS1 enzyme in vitro. (B) The
inhibitory activity of Ly93 against purified SMS2 enzyme in vitro. The results were
expressed as percent of SMS1 or SMS2 activities in vehicle groups in which there is
no inhibitor. (C) The dose-dependent inhibitory activities of Ly93 against purified
SMS1 and SMS2 enzyme respectively in vitro. The IC value of D2 toward purified
5
0
SMS2 was 20.9 µΜ. The inhibitory rate of D2 toward purified SMS1 was 57 % at
the concentration of 100 µΜ. The IC value of Ly93 toward SMS1 was tested with
5
0
ten concentrations. The IC50 value of Ly93 toward SMS2 was tested with eight
concentrations. Values were shown in means ± SD, n = 3.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2.6 The molecular docking study.
Molecular docking study was performed to investigate the interaction of Ly93
with hSMS2 which was derived from the previously reported model of hSMS1 [34]
and built through homology modeling. The binding modes of Ly93 with hSMS2 were
shown in Fig. 4A. As it was shown, Ly93 located at the active site of hSMS2 which
was responsible for SM production and thus blocked the entry of the substrate PC.
The key binding forces between Ly93 and hSMS2 were hydrophobic, π-π stacking
and van der Waals interactions. The central phenyl moiety formed hydrophobic
interactions with Ile278 and Leu225. It also formed a π-π stacking interaction with
His229 and van der Waals interactions with His270 and His272 respectively. The
pyridine ring formed π-π stacking and van der Waals interactions with Tyr281 and
Tyr282. Moreover, as Ly93 located close to the helix, the detrimental steric hindrance
limited the structural modification. The benzene ring of benzyl group formed a
cation-π interaction with Arg286 which could be reinforced by the electron-donating
substitution of the benzene ring. Besides, the hydrophobic substitution of the
ortho-position on the benzene ring of benzyl group could improve the hydrophobic
interaction with Ile263 and thus enhance the inhibitory potency against SMS2.
Meanwhile, as the benzene ring of benzyl group was located at the position between
1
4 / 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
two helixes and the space tolerance was improved, the long alkoxy chain at the
ortho-position of the benzene ring could reach the surface of the protein, increase the
hydrophobic property and improve the SMS2 inhibitory activities. As shown in Fig
4B, Ile278 in hSMS2 had a hydrophobic interaction with the central phenyl moiety of
Ly93, while Val334 at the same site in hSMS1 did not show similar hydrophobic
interaction. The conformation of Ly93 could be stabilized by the compact
hydrophobic pocket in hSMS2. This difference may contribute to the inhibitory
selectivity of Ly93 between hSMS1 and hSMS2. The docking studies were in
accordance with the results of enzymatic assay and gave more SAR information of
2-benzyloxybenzamides.
1
11
12
13
14
15
16
17
18
19
Fig 4. The molecular docking results of Ly93 with hSMS2 and hSMS1. (A) The
binding mode of Ly93 with hSMS2. (B) The overlay map of binding modes of Ly93
with hSMS1 (yellow) and hSMS2 (red). The ribbons in colors stand for the six
transmembrane (TM1-TM6) of hSMS2 and hSMS1. Carbon, oxygen, nitrogen, and
hydrogen atoms of the Ly93 molecule are colored green, red, blue and gray,
respectively.
In order to investigate the biological actions of selective SMS2 inhibitor, the in
vitro and in vivo biological evaluations were subsequently performed with Ly93.
1
5 / 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
2.7 The biological effects of Ly93 on the apoB secretion from Huh7 cells,
intracellular SMS activity and cholesterol efflux in macrophages.
The secretion of apolipoprotein B (apoB) and the efflux of cholesterol are
important processes of cells that are involved in the development of atherosclerosis.
ApoB which is synthesized in the liver is acknowledged as an atherogenic
lipoprotein. It is also the major protein component of VLDL and chylomicron (CM)
in which SM is enriched [1, 6, 35, 36]. In order to investigate the impact of SMS2
inhibition on apoB particle production, Huh7 cells, a human hepatoma cell line, were
firstly treated with Ly93. Then apoB secretion from the Huh7 cells was analyzed. As a
result, the treatment of Ly93 diminished apoB secretion from Huh7 cells in a
dose-dependent manner (Fig. 5A). It was suggested that the inhibition of SMS2 in
the liver could decrease the production of atherogenic lipoprotein.
1
1
1
1
1
1
1
1
1
As it is known that the cholesterol efflux from macrophages plays an important
role in reverse cholesterol transport that is an anti-atherogenic process [37]. In order
to study the effects of Ly93 on the cholesterol efflux from macrophages, the bone
marrow-derived macrophages were treated with different concentrations of Ly93. As
a result, the treatment of Ly93 not only significantly reduced SMS activity (Fig. 5B)
but also increased the cholesterol efflux in a dose-dependent fashion (Fig. 5C).
A
B
C
1
25
**
1
25
00
25
20
*
*
*
100
1
*
7
5
75
50
25
0
**
15
*
**
5
0
10
5
*
**
2
5
***
0
0
Vehicle
12.5
25
Vehicle12.5
25
50
100
6.25
12.5
25
Ly93(µM)
Ly93 (µM)
Ly93(µM)
1
9
0
2
Fig 5. The biological effects of Ly93 on the apoB secretion from Huh7 cells,
intracellular SMS activity and cholesterol efflux in macrophages. (A) The effect
of Ly93 on the apoB-containing particle secretion from Huh7 cells. Huh7 cells were
21
22
23
24
35
treated by Ly93 for 17 h and then were labeled with [ S]-methionine in the presence
3
5
of vehicle or Ly93 for 2 hours. The medium was collected to detect [ S]-apoB and
1
6 / 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
the quantitation was displayed. (B) Ly93 inhibits macrophage lysate SMS activity in
a dose-dependent manner. (C) Ly93 treatment significantly increased macrophage
cholesterol efflux toward HDL. Values are mean ± SD, n = 5. *P < 0.05, **P < 0.01,
***P < 0.001.
2.8 The in vitro inhibitory activities of Ly93 against the secretion of LPS-mediated
pro-inflammatory cytokine and chemokine in macrophages.
Our previous studies have shown that SMS deficiency in macrophages promotes
inflammatory responses and subsequently leads to the attenuation of atherosclerosis
in mouse models. The mechanism was attributed to the reduced SM levels in plasma
membrane lipid rafts [10-12]. Experiments were performed to investigate the effect
of the selective SMS2 inhibitor Ly93 on the inflammatory factors of macrophages.
Firstly, the production of IL-6 which is a well-known pro-atherogenic cytokine [38]
was analyzed when the macrophages were treated with Ly93. As shown in Fig. 6A,
the treatment of Ly93 could dose-dependently attenuate the LPS-induced IL-6
secretion in macrophages with the comparison of the control. Subsequently,
macrophages were pretreated with Ly93 and then stimulated by LPS (50 ng/ml). The
activation of NFκB which is a key inflammatory downstream of TLR4 in
LPS-induced macrophages was determined. As shown in Fig. 6B, the levels of
cytosolic IκBα in macrophages were directly increased in a dose-dependent manner
by the treatment of Ly93. Thus, the activation of NFκB was indicated to be
attenuated by Ly93. Furthermore, the production of MCP-1 which is a
pro-inflammatory chemokine was measured in the culture medium of Ly93-treated
macrophages. As shown in Fig. 6C, the levels of MCP-1 in the culture medium of
macrophages were also decreased by the Ly93 treatment.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
1
7 / 62

ACCEPTED MANUSCRIPT
A
B
C
IκB
β-actin
2
00
50
00
1
25
1
25
00
*
*
1
**
100
75
50
25
0
1
*
*
*
*
7
5
2
5
0
5
0
**
1
*
*
5
0
**
*
*
0
Vehicle 6.25
12.5
25
50
Vehicle 6.25
12.5
50
100
Vehicle 6.25
12.5
25
50
Ly93 (µM)
Ly93 (µM)
Ly93 (µM)
1
2
Fig 6. The in vitro inhibitory activities of Ly93 against the LPS-mediated
pro-inflammatory cytokine and chemokine secretion in macrophages. (A) IL-6
secreted by macrophages in culture medium after LPS (10 ng/ml for 16 h) treatment.
(B) Western blot of IκB after LPS treatment (insert) and the quantitation was
displayed. (C) MCP-1 production in macrophages induced by LPS. Values are mean ±
SD, n = 7. *P < 0.05, **P < 0.01.
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2.9 The pharmacokinetics of Ly93.
Studies on the pharmacokinetics of Ly93 were performed on C57BL/6J mice
according to the procedures shown in the part of Materials and Methods. The
representative mass spectra and chromatograms of Ly93 in plasma were shown in
Supplementary Fig. 3. The plasma concentration-time profiles of different Ly93
doses were shown in Fig. 7A. The pharmacokinetic parameters of Ly93 after
intragastric administration in C57BL/6J mice were shown in Supplementary Table
2. The AUC0~8h could amount to 6.26 µg·h/ml and the maximum concentration (Cmax)
could reach 6.12 µg/ml when mice were treated by the intragastric administration
with a single dose of Ly93 (50 mg/kg). Moreover, as shown in Fig. 7B, the AUC0-8h
1
1
1
1
1
1
1
1
1
1
2
2
2
were linearly correlated with the intragastric administration doses of Ly93 (R =
0.989). The high Vz/F indicated that compound Ly93 was with good permeability.
The pharmacokinetic results demonstrated that compound Ly93 with reasonable
physicochemical properties was suitable for the further in vivo evaluations.
1
8 / 62

ACCEPTED MANUSCRIPT
1
2
Fig 7. The pharmacokinetics of Ly93. (A) Plasma concentration-time profile of
Ly93. (B) The correlation between Ly93 AUC0-8h and dose from 25 to 100 mg/kg
after i.g. administration. Values are mean ± SD, n = 6.
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
2.10 The effects of Ly93 on plasma SM levels and liver enzymes in C57BL/6J mice
in vivo.
SM levels in the circulation are closely associated with lipoprotein metabolism
and the development of atherosclerosis [2, 7]. In order to investigate the impact of
Ly93 on the plasma SM level in vivo, C57BL/6J mice were treated with Ly93 (100
mg/kg, i.g.) once daily for 7 days. The plasma SM levels were analyzed at day 0, 3
and 7. As shown in Fig. 8A, the treatment of Ly93 significantly decreased the plasma
SM levels compared with vehicle group. At day 7, the concentration of Ly93 in livers
were also determined and compared with the concentration of Ly93 in blood. It was
found that the Ly93 concentrations in livers were five-fold higher than that in plasma
(Fig. 8B). The enriched concentration of Ly93 in the liver would favor the inhibition
of SMS activities and the attenuation of SM in plasma in vivo. Moreover, the
concentration of plasma alanine aminotransferase (ALT) and aspartate
aminotransferase (AST) in C57BL/6J mice at day 7 were determined to investigate
the impact of Ly93 on the liver enzymes. As shown in Fig. 8C, the concentration of
plasma ALT and AST in Ly93-treated C57BL/6J mice did not exhibit significant
changes in the mice of the vehicle group. As the obvious reduction of SM level in
plasma was observed on Ly93-treated C57BL/6J mice, Ly93 as a selective SMS2
1
1
1
1
1
1
1
1
1
1
2
2
2
1
9 / 62

ACCEPTED MANUSCRIPT
1
inhibitor was investigated for its atherosclerotic activities on the animal model.
2
3
Fig 8. The effects on plasma SM concentration and liver enzymes by the
treatment of Ly93 in vivo. (A) Plasma SM levels at day 0, 3 and 7 after Ly93 (i.g.,
100 mg/kg) treatment once daily for 7 days (n = 10). (B) Ly93 distribution in blood
and liver measured at day 7 after Ly93 (i.g., 100 mg/kg) treatment once daily for 7
days (n = 5). (C) AST and ALT levels at day 7 after a daily application of Ly93 (i.g.,
100 mg/kg) (n = 10). Values are mean ± SD, **P < 0.01.
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
2.11 The in vivo anti-atherosclerotic activities of Ly93 in apoE KO mice.
As it was shown in Supplementary Fig. 4, an in vivo experiment was designed
to investigate the anti-atherosclerotic activities of Ly93 in apoE KO mice which are the
acknowledged experimental animal model of spontaneous atherosclerosis. Firstly,
thirty apoE KO mice which were eight-week-old had been fed with the Western-type
diet for six weeks. Then thirty mice were equally divided into three groups and
subsequently had been treated (i.g.) daily with vehicle, low dose (12.5 mg/kg) and high
dose (40 mg/kg) of Ly93 respectively along with the Western-type diet for seven
weeks.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
During the trial, the body weights of all apoE KO mice in three groups were
monitored weekly to investigate the effect of Ly93 on the normal growth of apoE KO
mice. As it was shown in Supplementary Fig. 5, the growth curves of apoE KO mice
did not significantly change in either group. At the end of the experiment, the levels of
ALT and AST in the plasma of apoE KO mice were tested to investigate the
accumulative effect of Ly93 on the animal livers. As it was shown in Supplementary
2
0 / 62

ACCEPTED MANUSCRIPT
1
2
Fig. 6, the levels of ALT and AST in the plasma of apoE KO mice which were treated
by Ly93 did not show statistic changes when compared with the control group.
3
4
2.11.1 The effects on the liver SMS activities and plasma SM levels of apoE KO
mice by Ly93 treatment.
5
6
7
8
9
0
1
The livers of apoE KO mice were homogenized for the determination of SMS
activities. As shown in Fig. 9A, liver SMS activities in low dose and high dose groups
showed significant reductions (-35% and -39%, respectively; P < 0.05). Furthermore,
plasma SM levels of Ly93-treated mice were measured. It was found that there was a
significant and dose-dependent reduction in the SM levels of Ly93-treated mice as
compared with control animals (-25% and -41%, respectively; P < 0.05 and P < 0.01;
Fig. 9B).
1
1
1
2
3
1
Fig 9. The liver SMS activities and SM levels in Ly93-treated apoE KO mice. (A)
SMS activity in apoE KO mice liver after Ly93 treatment. (B) Plasma SM levels in
apoE KO mice after Ly93 treatment. Low dose: 12.5 mg/kg, i.g., High dose: 40 mg/kg,
i.g.. Values are mean ± SD, n = 10. *P < 0.05, **P < 0.01.
14
15
16
1
1
7
8
2.11.2 The blood cell analysis of apoE KO mice by flow cytometry after Ly93
treatment.
1
9
0
SM levels on the plasma membrane of certain blood cells, such as lymphocytes,
are closely associated with inflammation and the development of atherosclerosis. Flow
2
2
1 / 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
cytometry was utilized to evaluate the effect of Ly93 treatment on blood cells which
participate in the process of atherosclerosis. As shown in Fig. 10A and 10B, the
lymphocyte in mice of Ly93-low dose group and Ly93-high dose group were lowered
by 21% and 24% respectively compared with that of vehicle-treated group. The
monocyte (Fig. 10A and 10C) were significantly lowered in Ly93-low dose group
(-51%, P < 0.001) and Ly93-high dose group (-50%, P < 0.001), compared with control
group. CD4+/CD8+ cell ratios were reduced in Ly93-low dose group (-34%, P < 0.01)
and Ly93-high dose group (-37%, P < 0.01), compared with vehicle-treated group (Fig.
10D and 10E). Moreover, CD11b+CD11c+ subset of dendritic cells in Ly93-high dose
group was dramatically decreased by 57% (P < 0.05), while no significant changes
were found between Ly93-low dose group and control group (Fig. 10F and 10G).
CD11b+Ly6C++ and Ly6G+cells were attenuated by 16% (P < 0.05) in Ly93-low dose
group and 43% (P < 0.01) in Ly93-high dose group compared with vehicle group (Fig.
10H and 10I).
1
1
1
1
1
15
16
17
18
Fig 10. The blood cell analysis by flow cytometry in apoE KO mice after Ly93
treatment. (A) Representatives of the analysis of blood neutrophile granulocytes
(NG), lymphocytes and monocytes in vehicle and Ly93 treated animals. (B-C)
2
2 / 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
Quantitative display of gated lymphocytes and monocytes. (D) Representatives of the
analysis of blood CD4+ T cells and CD8+ T cells. (E) Quantitative display of the ratio
of CD4+ and CD8+ T cells. (F) Representatives of the analysis of blood
CD11b+CD11c+ dendritic cells. (G) Quantitative display of the percentage of the
dendritic cells. (H) Representatives of the analysis of blood CD11b+Ly6c+Ly6G+
cells. (I) Quantitative display of CD11b+Ly6c+Ly6G+ cells. Values are mean ± SD, n
= 5, *P < 0.05, **P < 0.01, ***P < 0.001.
8
9
2.11.3 The effects on the development of atherosclerosis in apoE KO mice by
Ly93 treatment.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
ApoE KO mice were sacrificed after 7 weeks of treatment with Ly93. The
proximal aortae and aortic arches of mice in each group were dissected and
photographed. The lesion areas of mice in each group were measured to evaluate the
inhibitive effect of Ly93 on atherogenesis. Aortic arches with atherosclerotic plaques
were shown in Fig. 11A. All mice were found present lesions in aortic arches. However,
those of Ly93 groups were noticeably smaller than those of vehicle-treated group. In
addition, significantly reduced lesion areas in Ly93-low dose group and Ly93-high
dose group were revealed by oil-red O in the whole aorta of apoE KO mice (-32% and
-41%, respectively; P < 0.05 and P < 0.01; Fig. 11B and 11C). Furthermore, there were
dose-dependently decreased lesions between two Ly93-treated groups. The proximal
aorta of mice in Ly93 groups also showed decreased lesion areas by hematoxylin and
eosin (H&E) staining compared with animals in vehicle-treated group (-38% and -50%,
respectively; P < 0.01; Fig. 11D and 11E). Immunostaining was then performed to
determine macrophage accumulation and investigated the affection of Ly93 treatment
on the composition of lesions in the aortic root of apoE KO mice. Although all sections
stained positively for monocyte/macrophage markers (MOMA), mice in Ly93-treated
groups had a significant reduction in macrophage content compared with that in
vehicle-treated group (-28% and -55%, respectively; P < 0.05 and P < 0.01; Fig. 11F
2
3 / 62

ACCEPTED MANUSCRIPT
1
and 11G).
2
3
Fig 11. The effects on the development of atherosclerosis in apoE KO mice with
Ly93 treatment. 8-week-old apoE KO mice were fed with a high fat and cholesterol
diet for 6 weeks, and then the mice in three groups were treated with vehicle (i.g.), low
dose Ly93 (12.5 mg/kg, i.g.) and high dose Ly93 (40 mg/kg, i.g.) respectively every
day for 7 weeks. (A) Aortic arches with atherosclerotic plaques (red arrows). (B)
Enface aortic plaque analysis after Oil Red O staining. (C) Quantitative display of en
faces aortic plaque. (D) Aortic root assay for lesion areas after Hematoxylin and eosin
staining. (E) Quantitative display of lesion areas in the aortic root. (F)
Immunohistochemical staining of macrophage accumulation in lesions (shown as
brown stained regions indicated by blue arrows). (G) Quantitative display of
macrophage content. Quantifications were done by using Image J software. Six
alternate sections (4 µm thick) sliced from paraffin-fixed aortic root tissues of each
transplanted mouse were used for the analysis. Values are mean ± SD. n = 9, *P <
0.05, **P < 0.01.
4
5
6
7
8
9
0
1
2
3
4
5
6
1
1
1
1
1
1
1
2
4 / 62

ACCEPTED MANUSCRIPT
1
3. Discussion and Conclusion
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
In this study, the 2-benzyloxybenzamide (D2-e) was firstly discovered as SMS2
inhibitor through scaffold hopping of lead compound D2. D2-e was tested with
moderate SMS2 inhibitory activity (IC50 = 13.4 µM). Then, a series of
2-benzyloxybenzamides were designed and synthesized through structural
optimization. The newly obtained compounds were put into the evaluation of the
SMS2 inhibitory activities. The results of in vitro enzymatic assays revealed that the
optimized compounds had significantly improved the SMS2 inhibitory activities
compared with D2-e. The structure-activity relationships were also studied based on
the in vitro enzymatic inhibitory activities. The 2-benzyloxybenzamides were proved
to be with more potential toward SMS2 inhibition and higher selectivity over SMS1
than the oxazolopyridine derivatives obtained from the lead compound D2 [31].
Furthermore, the 2-benzyloxybenzamide compound 10 with improved inhibitory
activity toward purified SMS2 (IC = 1.5 µM) became an optimal lead compound
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
5
0
that promoted the discovery of 4-benzyloxybenzo[d]isoxazole-3-amine derivatives
as selective SMS2 inhibitors [33].
Among the 2-benzyloxybenzamides, the novel small molecule Ly93 was one of
the most potent SMS2 inhibitors. Considering its prominent SMS2 inhibitory activity
and optimal properties, Ly93 was selected for further studies. Fortunately, it
exhibited nanomolar inhibitory activity against purified SMS2 (IC = 91 nM). The
5
0
selectivity ratio of Ly93 was more than 1400-fold for purified SMS2 over SMS1
(IC = 133.9 µM). The binding modes of Ly93 with the homology model of hSMS2
5
0
were predicted by the molecular docking study. It demonstrated that Ly93 bound with
hSMS2 at the same active site as other reported SMS2 inhibitors [28, 31, 33].
Therefore, Ly93, as a novel selective SMS2 inhibitor, was utilized as a validation
tool for the investigations of SMS2-relevant pathways.
In Huh7 cells, the treatment of Ly93 decreased the apoB secretion from
2
5 / 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
hepatocytes in a dose-dependent manner. It was in accordance with the previous
studies that SM has significant effects on the metabolism of atherogenic lipoproteins
[1]. In macrophages, Ly93 attenuated the intracellular SMS activity and
simultaneously increased the cholesterol efflux that involves in the reverse
cholesterol transport and anti-atherogenic process. Moreover, due to the decreased
SM in the macrophage membrane, Ly93 depressed the secretion of LPS-mediated
pro-inflammatory cytokine and chemokine from macrophages indirectly. Since it has
been reported that cellular SM in macrophage is one of the major factors that
modulate the atherogenic functions of macrophage [12], it is conceivable that SMS
inhibition-mediated membrane SM reduction could have consequences in terms of
macrophage inflammation and cholesterol efflux which are the critical factors of
atherogenesis. The effects of Ly93 on the cholesterol efflux and inflammatory
responses in macrophages could further contribute to the anti-atherosclerotic
activities in vivo.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
Considering the linear correlation of AUC0-8h between doses of Ly93 after
intragastric administration, Ly93 was further studied the SMS inhibitory activities in
vivo. Ly93 induced the inhibition of SMS that specifically reduced the plasma SM
levels in C57BL/6J mice without obvious effects on the liver enzymes. This could
make contributions to the reduction of atherosclerosis mediated by SM-enriched
lipoproteins.
Since the reduction of plasma SM levels was observed in Ly93-treated
C57BL/6J mice, the further in vivo studies were conducted in apoE KO mice that are
an acknowledged model of spontaneous atherosclerosis. The 7-week treatment of
Ly93 significantly decreased the SMS activities in mice livers and SM levels in
plasma in a dose-dependent manner. However, the levels of cholesterol and
triglyceride in apoE KO mice were not significantly impacted by the treatment of
Ly93 (Supplementary Fig. 7). It was indicated that the selective inhibition of SMS2
could specifically reduce the SM level.
2
6 / 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
In previous studies, it has been reported that SM levels on the plasma membrane
of lymphocytes and monocytes are closely associated with inflammation and the
development of atherosclerosis [39, 40]. In this study, the SMS inhibition with the
treatment of Ly93 could reduce the lymphocytes and monocytes in apoE KO mice
indicating the decreased inflammatory responses and atherosclerosis. It was also
demonstrated that inhibition of SMS activity could raise the ratio of CD8+T/CD4+T.
The results were in accordance with the reported studies that a subset of CD8+T cells
could reduce atherosclerotic lesions in apoE KO mice [41].
In the process of atherosclerosis development, CD11b+CD11c+ subset of
dendritic cells was rapidly increased [42-44]. Dendritic cells in the aorta intimal
rapidly ingest lipid and become foam cells when mice were fed with high fat/high
cholesterol diet. Depletion of these intimal dendritic cells reduces foam cell formation
[43]. Deficiency of CD11c+, a surface marker on dendritic cells, reduced
atherosclerotic lesions [44]. Our result indicated that SMS inhibition reduces
CD11b+CD11c+ dendritic cells in the blood (Fig. 10F and 10G). This could
contribute to the reduction of atherosclerosis in apoE KO mice. CD11b+Ly6C+ and
Ly6G+ cells (granulocytic myeloid-derived suppressor cells, granulocytic MDSCs)
are also involved in pro-inflammatory processes and chronic inflammation [45, 46].
These cells were reduced by Ly93 treatment in a dose-dependent manner (Fig. 10H
and 10I). The significance of these changes, as related to atherosclerosis, deserves
further studies.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
Finally, the body weights and liver enzymes of Ly93-treated apoE KO mice did
not exhibit significant changes with those of vehicle group. It could suggest that the
Ly93 scarcely induce side-effects on the apoE KO mice. Under the safe treatment of
Ly93 for 7 weeks, the selective SMS2 inhibitor attenuated the atherosclerotic plaque
in aortic arches and root as well as in the entire aorta. Meanwhile, the accumulation
of macrophages in atherosclerotic lesions was prevented by Ly93. As macrophages
participate in the atherogenic processes including inflammatory responses and the
2
7 / 62

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
formation of foam cells, the reduced macrophages in atherosclerotic lesions could
suggest that Ly93 directly impact the macrophages through which it indirectly
prevents the development of atherosclerosis.
In summary, 2-benzyloxybenzamides were discovered as novel SMS2 inhibitors.
Ly93 was found to be a novel orally efficacious selective SMS2 inhibitor with
nanomolar SMS2 inhibitory activity. In in vivo studies, Ly93 dose-dependently
inhibited the liver SMS activities, reduced the plasma SM levels and finally
attenuated the development of atherosclerosis. The results of in vitro assays
demonstrated that Ly93 diminished apoB secretion from Huh7 cells, reduced the
SMS activity and increased cholesterol efflux in macrophages in a dose-dependent
manner. Meanwhile, Ly93 could inhibit the secretion of LPS-mediated
pro-inflammatory cytokine and chemokine in macrophages. These preliminary
molecular mechanism-of-action studies showed that Ly93 could affect both lipid
homeostasis and inflammation process. To our knowledge, it is the first study that
selective SMS2 inhibitor is investigated the anti-atherosclerotic activities in vivo,
which would make contributions to the further studies on the SMS2-relevant
pathways and promote the development of selective SMS2 inhibitor to be a
promising treatment for atherosclerosis.
1
1
1
1
1
1
1
1
1
1
9
4. Materials and Methods
20
21
22
23
24
25
26
27
4.1 Materials
Reagents used in the chemical synthesis were purchased from commercial
sources and were used without further purification except for the special case. Flash
column chromatography was carried out at medium pressure using silica gel (200-300
mesh) purchased from Qingdao Haiyang Chemical Co. Ltd. All the reactions were
monitored by thin layer chromatography (TLC) on silica gel. Mass spectra data were
given with electrospray ionization (ESI) produced by a Finnigan MAT-95,
LCQ-DECA spectrometer and IonSpec 4.7 T. MS. High-resolution mass spectra data
2
8 / 62