Identification of small molecule sphingomyelin synthase inhibitors

Article abstract of DOI:10.1016/j.ejmech.2013.12.002

Sphingomyelin synthase (SMS), which catalyzes ceramide as one of the substrates to produce sphingomyelin, is a critical factor in the sphingolipid biosynthesis pathway. Recent studies indicated that SMS could serve as a novel potential drug target for the

Full text of DOI:10.1016/j.ejmech.2013.12.002

European Journal of Medicinal Chemistry 73 (2014) 1e7
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Identification of small molecule sphingomyelin synthase inhibitors
Xiaodong Deng a, Fu Lin b, Ya Zhang a, Yan Li b, Lu Zhou a, Bin Lou a, Yue Li a, Jibin Dong a,
**
,
a,
*
Tingbo Ding a, Xiancheng Jiang a, Renxiao Wang b , Deyong Ye
a Department of Medicinal Chemistry, School of Pharmacy, Fudan University, No.826, Zhangheng Rd., Shanghai 201203, People’s Republic of China
b State Key Lab of Bio-organic and Natural Products Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences,
People’s Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
Sphingomyelin synthase (SMS), which catalyzes ceramide as one of the substrates to produce sphin-
gomyelin, is a critical factor in the sphingolipid biosynthesis pathway. Recent studies indicated that SMS
could serve as a novel potential drug target for the treatment of various metabolic diseases such as
insulin resistance and atherosclerosis. However, very few small-molecule inhibitors of SMS are known. In
this study, we performed structure-based virtual screening in combination with chemical synthesis and
bioassay and discovered a class of small-molecule SMS inhibitors. The most potent compound exhibited
Received 2 April 2013
Received in revised form
13 June 2013
Accepted 8 December 2013
Available online 14 December 2013
an IC50 value lower than 20 mM in an in vitro enzymatic assay. To the best of our knowledge, this is the
Keywords:
first time that small-molecule SMS inhibitors with potency close to the micromolar range are publicly
revealed. The structureeactivity relationship demonstrated by this class of compounds provides insights
into the structural features that are essential for effective SMS inhibition.
Sphingomyelin synthase
Small-molecule inhibitors
Structure-based virtual screening
Atherosclerosis
Ó 2013 Elsevier Masson SAS. All rights reserved.
1. Introduction
two isoforms, i.e. SMS1 and SMS2. SM production is catalyzed by
SMS1 in the Golgi and SMS2 on plasma membrane [5]. SMS1 and
Sphingomyelin synthase (SMS) is the last enzyme on the
sphingomyelin (SM) biosynthesis pathway. As an integral mem-
brane protein, SMS transfers the phosphorylcholine moiety from
phosphatidylcholine onto the primary hydroxyl group on ceramide,
producing SM and diglyceride. SMS plays various physiological and
pathological roles [1, 2]. Recently, studies on mice indicated that
SMS deficiency prevented the development of atherosclerosis [3],
obesity [4], and insulin resistance [4], while SMS over-expression
significantly increased atherosclerosis [5]. These studies suggest a
connection between the activity of SMS and the change in SM level
in plasma or plasma membrane [1,5e7]. Thus, SMS may serve as a
valid target for developing novel therapies for metabolic diseases.
Discovery of small-molecule inhibitors of SMS is a promising
strategy for achieving this goal. Mammalian SMS is known to have
SMS2 share 57% sequence similarity and they have identical
enzymatic activities [6]. Three-dimensional structures of SMS have
not been resolved experimentally so far.
Before our work, the only publicly known small-molecule SMS
inhibitor was tricyclo[5.2.1.0(2,6)]-decan-8-yl dithiocarbonate
(D609, Fig. 1). It was originally identified as a cytotoxic anti-tumor
and anti-virus reagent [7]. Unfortunately, it is not practical to use
D609 as an effective SMS inhibitors due to its high instability (t1/
2 z 19.5 min in saline solution at 24 ꢀC) [6, 8] and weak inhibitory
activity (in vitro IC50 z 375 mM, and cellular EC50 z 90 mM) [9].
Optimizations of D609 led to improved stability [7, 8]. An inhibitor
of P-glycoprotein resistance protein-1, namely MS-209, was also
referred as a SMS inhibitor, but there was no further specific
experimental data to support this view [10, 11, 12]. We have syn-
thesized MS-209, and the result of its SMS inhibitory activity assay
(D609 was used as positive control) showed that the IC50 value was
higher than 500 mM.
Abbreviations: SMS, Sphingomyelin synthase; SM, Sphingomyelin; D609, po-
tassium tricyclo[5.2.1.0(2,6)]-decan-8-yl dithiocarbonate; 3D-hSMS1, three-dimen-
sional structure of human SMS1; i.p., intraperitoneally; i.v., intravenously; ICR mice,
Institute of Cancer Research mice; HIS, histidine; Huh-7, human hepatoma cell;
IC50, 50% inhibiting concentration; CC50, 50% cytotoxicity concentration.
* Corresponding author. Tel./fax: þ86 21 51980125.
It is thus much desired to develop new small-molecule SMS
inhibitors with higher potency and better stability. Such com-
pounds may be developed into successful drug candidates or
applied as effective chemical tools for probing the biological func-
tions of SMS. In our study, we applied structure-based virtual
screening to discover small-molecule compounds targeting at SMS.
** Corresponding author. Tel.: þ86 021 54925128.
E-mail addresses: wangrx@mail.sioc.ac.cn (R. Wang), dyye@shmu.edu.cn (D. Ye).
0223-5234/$ e see front matter Ó 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2013.12.002

2
X. Deng et al. / European Journal of Medicinal Chemistry 73 (2014) 1e7
with potency close to the micromolar range that are publicly
revealed.
The structureeactivity relationship of this class of compounds
indicates clearly that the cyano group in the -aminonitrile moiety
a
is the key factor for SMS inhibition. The binding mode of D2, which
was given by molecular docking (Fig. 3), suggests that the cyano
group points to the imidazole ring on the side chain of His285
residue, forming a possible favorable dipoleedipole interaction. In
addition, a possible hydrogen bond is observed between the amino
group connecting the Ar2 moiety and His274, which seems to be
another essential factor for SMS inhibition. The two aromatic
branches (Ar1, Ar2) reside at two hydrophobic sub-pockets
respectively. Our biological test results show that ortho-
substituted aryl at the Ar1 site is better than meta- or para-
substituted counterparts. Hydrophobic substituted aromatic moi-
ety in the ortho-position are preferred even more. As compared to
D2, the derivative compounds listed in Table 1 did not really ach-
ieve a higher level of potency. However, they collectively confirmed
the SMS inhibitory activities of this class of compounds. These
preliminary structureeactivity relationships are helpful for devel-
oping even more potent SMS inhibitors.
Fig. 1. The chemical structure of D609 (as potassium salt).
2. Results
2.1. Virtual screening
In a previous study [2], we reported a structural model of human
SMS1 (hSMS1) which was derived basically from homology
modeling and refined by molecular dynamics simulations. This
structural model provided the structural basis for the virtual
screenings conducted in this work.
The SPECS library, which contained structural information of
over 220,000 compounds, was considered in our study. A number
of compounds were selected out of a multi-step structure-based
virtual screening process. Samples of 93 compounds were pur-
chased from SPECS and evaluated in SMS inhibitory experiments.
Two compounds (D1 and D2 in Fig. 2) exhibited obvious inhibitions
of SMS activity in our assay.
2.4. In vivo SMS activity assay
In order to explore the potential of D2 as a promising lead
compound for drug development, this compound was further
tested in vivo for SMS inhibition. ICR mice were administrated with
D2 in dose of 6 mg/kg intraperitoneally (i.p.) or intravenously (i.v.).
Our results showed that upon the treatment of D2, the total amount
of ceramides was significantly increased (Fig. 4-A); while the total
amount of SM was significantly decreased in mice plasma (Fig. 4-B).
This observation is consistent with the expectation of SMS inhibi-
tion, which confirms that D2 is an effective SMS inhibitor both
in vitro and in vivo.
2.2. Structural modification of hit D2
By considering biological potency and synthetic feasibility, we
selected D2 as the lead compound for further structural modifica-
tion. Three different sites (Ar1, Ar2, and X) were explored on the
scaffold of D2 (Table 1): Substituents at the Ar1 site included phenyl
and o-, m-, p-benzyloxy phenyl groups; substituents at the Ar2 site
included nucleophilic, electrophilic and hydrophobic group-
substituted aromatic moieties and aromatic heterocyclic moieties;
whereas X were hydrogen, cyano, carbamoyl and tetrazole groups.
These chemical modifications were expected to explore the struc-
tureeactivity relationship of this class of compounds. Synthetic
2.5. Cytotoxicity assay
Toxicity of D2 was evaluated on Huh7 cells. The CC50 of D2 was
245
mM, which suggested that D2 has an efficient therapeutic
window and further supported that D2 could be used as a potential
molecular tool for SMS bio-function studies.
methods were developed for preparation of the
a
-aminonitrile
derivatives (Scheme 1). A total of 20
were synthesized (Table 1).
a-aminonitrile derivatives
3. Conclusion
We have discovered a class of small-molecule SMS inhibitors
through structure-based virtual screening in combination with
chemical synthesis and biological assays. Among them, compounds
2.3. In vitro SMS activity assay and SAR study
All obtained compounds were then evaluated in a HPLC-based
assay for measuring SMS inhibition that was developed in our
group [13]. As result, 17 compounds were observed to have higher
SMS inhibitory activities than D609. Among them, five compounds
exhibited SMS2 inhibition with IC50 values lower than 20
which are 10-fold more potent than D609. To the best of our
D2, D24, D28, D29, D30 and D31 have IC50 values lower than 20 mM
in an in vitro SMS inhibition assay. They could be considered as lead
compounds for obtaining even more potent SMS inhibitors and
may be eventually developed into new therapies for treating
various metabolic diseases.
mM,
knowledge, these compounds are the first group of SMS inhibitors
4. Experimental
4.1. Materials
4.1.1. Chemistry
The reagents (chemicals) were obtained from Alfa Aesar, Acros,
or Shanghai Chemical Reagent Co. and were used without further
purification. Low- and high-resolution mass spectra (LRMS and
HRMS) were given with electrospray ionization (ESI) produced by a
Finnigan MAT-95, LCQ-DECA spectrometer and IonSpec 4.7 T. Nu-
clear magnetic resonance (NMR) spectroscopy was performed on
Bruker AMX-500 and AMX-400 (tetramethylsilane (TMS) was used
D1 (IC50 =75 µM on SMS2)
D2 (IC50 =14 µM on SMS2)
Fig. 2. Chemical structures of two SMS inhibitors which were selected from structure-
based virtual screening outcomes and were verified in biological assay.

X. Deng et al. / European Journal of Medicinal Chemistry 73 (2014) 1e7
3
Table 1
Biological Activity of the synthetic hits and hit modification.
,b
,b
Compound
X
Ar1
Ar2
IC50
(
m
M) (SMS2)a
IC50 (m
M) (SMS)a
D2
Cyano-
Cyano-
Cyano-
Cyano-
Cyano-
Cyano-
Cyano-
Cyano-
Cyano-
Cyano-
Cyano-
Cyano-
Cyano-
Cyano-
Cyano-
Cyano-
2-Benzyloxyphenyl-
3-Benzyloxyphenyl-
4-Benzyloxyphenyl-
Phenyl-
Phenyl-
Phenyl-
Phenyl-
Phenyl-
13.5 ꢃ 0.7
50.7 ꢃ 5.2
NAc
24.5 ꢃ 2.4
NDd
D21
D22
D23
D24
D25
D26
D27
D28
D29
D30
D31
D32
D33
D34
D35
D36
D37
D38
D39
D40
D41
NDd
NAc
NDd
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Benzyloxyphenyl-
2-Methoxyphenyl-
3-Methoxyphenyl-
4-Methoxyphenyl-
2-Methyl-phenyl-
4-Methyl-phenyl-
2-Clorophenyl-
4-Clorophenyl-
2-Hydroxyphenyl-
4-Hydroxyphenyl-
2-Cyano-phenyl-
4-Carbamoyl-phenyl-
2,6-Dimethyl-phenyl-
2-Nitro-phenyl-
3-Pyridyl-
12.4 ꢃ 1.6
35.8 ꢃ 2.4
NAc
NDd
48.1 ꢃ 4.3
67.2 ꢃ 9.0
20.8 ꢃ 3.4
15.8 ꢃ 2.6
22.5 ꢃ 4.3
19.0 ꢃ 4.2
24.4 ꢃ 1.6
20.3 ꢃ 0.6
80.5 ꢃ 4.5
31.5 ꢃ 8.4
27.4 ꢃ 1.6
35.0 ꢃ 5.1
71.0 ꢃ 5.4
NAc
NDd
NDd
NDd
90.8 ꢃ 0.2
93.9 ꢃ 0.2
NDd
NDd
NDd
196.3 ꢃ 3.6
NDd
Cyano-
Cyano-
Cyano-
Hydrogen-
Carbamoyl-
1H-tetrazol-5-yl-
NDd
NDd
Furan-2-ylmethyl-
Phenyl-
Phenyl-
NDd
NDd
135.3 ꢃ 4.9
NDd
Phenyl-
NAc
NDd
a
The IC50 of D609 as a positive control compound was 224
The enzyme source of SMS was from ICR mice liver homogenate, and the enzyme source of SMS2 was from over-expression SMS2 cell lysis. IC50 value measurements using
m
M in SMS2 and 402
m
M in SMS.
b
the SMS inhibition assay (as shown in experimental). Five concentrations of inhibitor were used to determine the IC50 values. The shown values were the means from triplicate
assays. Statistical analysis were performed using Prism 5.02 (GraphPad Software, Inc.) and “Log[inhibitor] vs. response” was used as curve adjustment model for IC50 mea-
surement. Values are means ꢃ SD.
c
NA: no activity at 100
ND: not determined.
mM.
d
as the internal standard). Chemical shifts were reported in
d
(ppm)
separations were carried out on an Agilent C18 RP column
m).
Buffer 1 for storage of SMS enzyme resource: 0.25 M sucrose,
50 mM Tris$HCl, pH 7.4, 1 mM EDTA.
downfield from TMS. Solvent residual peak, and peak multiplicity
were described as singlet (s), doublet (d), triplet (t), quartet (q),
multiplet (m), and broad (br). The yield was not optimized.
(250 mm ꢂ 4.6 mm, 5
m
4.2. Virtual screening of the SPECS Library
4.1.2. Biochemistry
6-((N-(7-nitrobenz-2-oxa-1,
3-diazol-4-yl)amino)hexanoyl)-
A multi-step procedure was employed in our virtual screening
to discover potential small-molecule inhibitors of hSMS1. Firstly,
the entire SPECS library (about 220,000 compounds) was filtered
by drug-likeness rules, which required molecular weight lower
than 1000 Da, total number of hydrogen bond acceptors and donors
between 2 and 10, number of non-hydrogen atoms between 9 and
50, and no more than 10 rotatable bonds. This step selected out
approximately 80,000 compounds. Then, qualified molecules were
docked into the active site of hSMS1 using the molecular docking
program GOLD (version 3.2, released by CCDC). The structural
model of hSMS1 was obtained through homology modeling, which
was reported in one of our previous studies [2]. Key parameters
were set as follows: Island number ¼ 5; populations’ size on each
island ¼ 100; total genetic algorithm operations ¼ 30,000; muta-
tion rate ¼ 95%; crossover rate ¼ 95%; migration rate ¼ 10%; scoring
function ¼ ChemScore. Up to twenty binding poses were generated
for each docked molecule. After this docking step, 20,000 mole-
cules with the highest binding scores were selected out. These
molecules were further screened with the Glide docking program
implemented in the Schrödinger software (version 2008, released
by the Schröodinger Inc.) at the “extra-precision” (XP) mode. The
hSMS1 binding site was depicted by three key residues, i.e. H285,
H328 and D332, in the catalytic site [2]. Default parameters were
employed in molecular docking. The hSMS1 structure was kept
fixed during this process. Then, the top-ranked 4000 molecules
sphingosine (C6-NBD-Cer) and 1,2-dimyristoyl-sn-glycero-3-
phosphocholine (DMPC) were obtained from Santa Cruz (USA). N-
(N-(7-nitro-2, 1, 3-benzoxadiazol-4-yl)-epsilon-aminohexanoyl)
sphingosylphosphoryl choline (C6-NBD-SM) was purchased from
SigmaeAldrich (USA). Potassium tricyclo [5.2.1.0(2, 6)]-decan-8-yl-
dithiocarbonate (D609) was obtained from 3B Scientific Corpora-
tion (China). Laboratory-prepared water (Milli-Q) was used
throughout the study. All other reagents were analytical grade
available.
ICR (Institute of Cancer Research) mice were provided by Sino-
British S1PPR/BK Lab. Animal Ltd. All experimental procedures
were approved by the Institutional Review Board of Fudan
University.
SMS2 was kindly provided by Dr. Yanhui Xu’s lab, Institute of
Bioscience, Fudan University. The enzyme was highly expressed in
H5 insect cell. The cell pellet was weighed and then homogenized
by Glass homogenizer with homogenate buffer 1 (0.25 M sucrose,
50 mM Tris$HCl, pH 7.4, 1 mM EDTA) (50 mg/mL, weight of cell
pellet/volume of buffer 1). The homogenates were subjected to
centrifugation at 12,000 rpm for 10 min. The supernatant, as a raw
SMS2 solution, was stored at ꢁ20 ꢀC, or used directly.
An Agilent 1100 (Agilent Technologies, Palo Alto, CA, USA) HPLC
system was used for our experiments, attached with a quaternary
pump, a vacuum degasser and a FLD detector. The chromatographic

4
X. Deng et al. / European Journal of Medicinal Chemistry 73 (2014) 1e7
Scheme 1. General synthesis route for SMS inhibitors. Reagents and conditions: a, TMSCN, I2, CH3CN, rt; b, NaBH4, MeOH, rt; c, NaN3, ethanol, reflux; d, K2CO3, H2O2, CH2Cl2, rt.
were subjected to similarity analysis to remove structurally
redundant ones. Here, similarity matrix based on molecular
fingerprint ECFP4 was generated by Pipeline Plot (released by
Accelrys Inc.). A total of 500 clusters were obtained by using the
Matlab software (released by MathWorks Inc.). The compound with
the highest Glide binding score was selected from each cluster.
These 500 compounds were inspected visually to examine their
binding modes and likelihood of chemical modification. Finally, a
total of 93 compounds were selected as candidates. The samples of
these compounds were ordered from SPECS Inc. for subsequent
biological evaluation.
93 selected compounds were screened by the same method as
shown in followed section 4.3.1, which ICR mice liver homogenate
was used as enzyme source. 100
threshold concentration for initial screening. Compound D1 and D2
were discovered higher than 50% inhibitory under 100 M con-
centration. So D1 and D2 were further determined to afford their
IC50. The result showed IC50 of D1 and D2 were 75 M and 14 mM
mM of compound was used as
m
m
respectively (Fig. 2). And D2 were selected as lead for further
optimization.
4.3. Assessment of SMS inhibitory activities
4.3.1. SMS activity in vitro
The method of SMS activity was described by the previous
report [13]. Briefly, ICR mice were sacrificed and livers were
dissected, weighed and then homogenized with homogenate buffer
1 (0.25 M sucrose, 50 mM Tris$HCl, pH 7.4, 1 mM EDTA) (300 mg/
mL, weight of livers/volume of buffer 1). The homogenates were
subjected to centrifugation at 12,000 rpm for 10 min. The super-
natant, as a raw SMS solution, was stored at ꢁ20 ꢀC, or used
directly.
Inhibitory activity of SMS was assessed using a fluorescent
substrate C6-NBD-Ceramide;
were performed in Eppendorf tubes using 300
l
ex/em ¼ 475/525 nm. The assays
m
L total reaction
volumes. The samples were buffered in 10 mM HEPES (pH 7.4),
3 mM MnCl2 and 0.3% BSA, SMS (equivalent to 2 mg total protein)
was incubated for 30 min at 37 ꢀC in the presence of inhibitor
compounds and in the absence of substrates. The enzymatic reac-
tion was started by the addition of 1.16 mM of C6-NBD-Cer and
Fig. 3. Binding mode of D2 at the active site of hSMS1 as predicted by molecular
docking.

X. Deng et al. / European Journal of Medicinal Chemistry 73 (2014) 1e7
5
Control (i.v.)
D2 (i.v.)
D2 (i.p.)
Control (i.v.)
D2 (i.v.)
D2 (i.p.)
10000
8000
6000
4000
2000
0
2.5
2.0
1.5
1.0
0.5
0.0
A
B
Fig. 4. The effect of D2 in vivo on plasma ceramide (A) and SM (B). Mice as negative control were administered no D2. D2 were administered i.v. or i.p. in the test set of mice in dose
of 6 mg/kg. A: The amount of total ceramide in plasma of ICR mice were increased significantly in i.v. or i.p. groups. B: The amount of total SM in plasma of ICR mice were decreased
significantly in i.v. or i.p. groups. Value are mean ꢃ SD, n ¼ 5. Student’s t-test P < 0.05.
40 mM DMPC in 3
for an additional 2 h. The enzyme catalytic reactions were
quenched by 600 L of ethanol. After vortexing for 30 s, the reaction
mixture was centrifuged and the supernatant was used for HPLC
analysis. Extracts (20 L) from the supernatant were analyzed for
the formation of C6-NBD-SM by HPLC using the reversed-phase C18
column and an isocratic elution with methanol/water/trifluoro-
acetic acid (88:12:0.1 (v/v)). The flow rate of the mobile phase was
1.0 mL/min. Detection was accomplished with a fluorescence
spectrophotometer (lexcitation ¼ 475 nm, lemission ¼ 525 nm). Assays
were performed at least in triplicate. Statistical analysis and non-
linear regression were performed using Prism 5.02 (GraphPad
Software, Inc.).
m
L DMSO. The mixtures were incubated at 37 ꢀC
4.5. Toxicity of D2 on Huh-7 cells
m
Huh-7 cells were incubated with D2 in various concentrations
for 17 h and then examined for cell viability by MTT assay. CC50 is
the concentration of D2 that causes 50% inhibition compared to
the vehicle control (0.25% DMSO). Samples were run in quintu-
plicate, and the results are the mean of at least two independent
tests.
m
4.6. Chemical synthesis
4.6.1. General procedure
The key transformation for synthesis of the a-aminonitrile de-
rivatives was Strecker reaction[16, 17]. The synthetic route for the
preparation of compounds D2, D21eD38 was depicted in Scheme 1.
The substituted benzaldehyde was treated with the substituted
arylamine in CH3CN solution under room temperature, followed
with trimethylsilyl cyanide (TMSCN) and catalytic amount of I2. The
mixture was stirred at room temperature overnight to get the
4.3.2. SMS2 activity in vitro
Briefly, 300
mL of reaction mixture, which contained the samples
buffered in 10 mM HEPES (pH 7.4), 3 mM MnCl2, 0.3% BSA and SMS2
(equivalent to 2
mg total protein of cell homogenates of over-
expressing SMS2 in buffer 1), was incubated for 30 min at 37 ꢀC
in the presence of inhibitor compounds and in the absence of
substrates. The enzymatic reaction was started by the addition of
desired
a-aminonitrile derivatives. The typical reaction procedure
for preparation of
a-aminonitrile derivatives was shown in syn-
1.01 mM of C6-NBD-Cer and 40 mM DMPC in 3
m
L DMSO. The
thesis detail of compound D2. Compound D2 can be easily con-
verted to compound D41 by treating with NaN3/ethanol [18], or to
D40 by potassium carbonate and hydrogen peroxide in water [19].
The de-cyano compound D39 was obtained by following the
reductive amination condition [20].
mixtures were incubated at 37 ꢀC for an additional 1 h. The
resulting mixture was worked up and analyzed by the same pro-
cedure of SMS activity assay described above.
4.4. SMS activity in vivo
4.6.1.1. General procedure for the preparation of 2-(2-(benzyloxy)
phenyl)-2-(phenylamino) acetonitrile (D2). A solution of 2-(ben-
zyloxy)benzaldehyde (1.00 g, 4.71 mmol) and aniline (438.77 mg,
4.71 mmol) in acetonitrile (20 mL) was stirred at rt for 30 min.
Then TMSCN (701.13 mg, 7.07 mmol) and I2 (119.58 mg,
0.47 mmol) were added and the mixture was stirred overnight.
The solvent was evaporated in vacuo and the residual was diluted
with ethyl acetate/water. The organic layer was washed with
saturated brine, dried over MgSO4 and concentrated under vac-
uum. The residual was triturated with ethyl acetate/hexane (1/6),
filtered to give the product, 1.35 g, 91% yield. 1H NMR (500 MHz,
ICR mice (approximately 25 g) were maintained in restricted-
access rooms with a controlled temperature (23 ꢀC) and a 12 h
lightedark cycle and were allowed free access to standard labora-
tory diet and tap water. All experimental procedures were
approved by the Institutional Review Board of Fudan University.
Fifteen female ICR mice, which averagely weighed 25
(25 ꢃ 1 g), were randomly divided into 3 groups. Among them, 1e5
as control group were administered intravenously (mixture of 15
DMSO and 185 L phosphate-buffer saline (PBS; pH 7.4)); 6e10
were administered intravenously (mixture of 150 g in 15 L DMSO
and 185 L PBS), 11e15 were administered intraperitoneally
(mixture of 150 g in 15 L DMSO and 185 L PBS). The drug dose of
g
mL
m
m
m
m
CDCl3):
d
¼ 4.28 (br, 1H, NH), 5.15e5.21 (dd, J1 ¼ 11.78 Hz,
m
m
m
J2 ¼ 15.4 Hz, 2H, OCH2Ar), 5.62 (s, 1H, CH(CN)N), 6.72e6.73 (d,
D2 was equivalent to 6 mg/kg for each mouse. After 2 h absences or
presence of drug treatment, blood was sampled from mice eyes and
was directly frozen storage by liquid nitrogen for analysis. Plasma
ceramide and sphingomyelin were measured as previously
described via mass spectrometry (MS) [14,15]. Differences between
groups were tested by Student’s t-test. Data are presented as
mean ꢃ S.D.
J ¼ 7.79 Hz, 2H, HAr), 6.85e6.88 (m, 1H, HAr), 7.02e7.05 (m, 2H,
H
Ar), 7.21e7.25 (m, 2H, HAr), 7.32e7.52 (m, 6H, HAr), 7.53e7.54 (d,
J ¼ 7.83 Hz, 1H, HAr). 13C NMR (400 MHz, CDCl3):
d
¼ 46.32, 70.59,
112.75, 114.48, 120.12, 121.48, 127.35, 128.24, 128.74, 128.89,
129.47, 130.99, 136.08, 155.94. ESI-MS m/z 315.2 (MþHþ). HRMS
(ESI) m/z calcd for MþNaþ 337.1311; found 337.1319. IR:
2227.59 cmꢁ1 (CN).

6
X. Deng et al. / European Journal of Medicinal Chemistry 73 (2014) 1e7
4.6.1.2. 2-(3-(benzyloxy)
phenyl)-2-(phenylamino)
acetonitrile
for D2. Yield 251 mg (81%). 1H NMR (400 MHz,CDCl3):
d
¼ 2.28 (s,
(D21). This compound was prepared by using the procedure as
described for D2. Yield 219 mg (74%).
3H, CH3), 5.09 (s, 2H, OCH2Ar), 5.37 (s, 1H, CH(CN)N), 6.68e6.70 (d,
J ¼ 8 Hz, 2H, HAr), 7.02e7.04 (dd, J1 ¼ 8.2 Hz, J2 ¼ 2.4 Hz, 1H, HAr),
7.07e7.09 (d, J ¼ 7.8 Hz, 2H, HAr), 7.19e7.22 (m, 2H, HAr ), 7.34e7.45
(m, 6H, HAr). ESI-MS: m/z 329.2 (MþHþ). HRMS (ESI) m/z calcd for
MþHþ 329.1648; found 329.1643.
1H NMR (400 MHz,CDCl3):
d
¼ 5.10 (s, 2H, OCH2Ar), 5.41 (s, 1H,
CH(CN)N), 6.77e6.79 (d, J ¼ 7.8 Hz, 2H, HAr), 6.90e6.94 (m, 1H, HAr),
7.03e7.06 (m, 1H, HAr), 7.20e7.23 (m, 2H, HAr), 7.29e7.45 (m, 8H,
H
Ar). ESI-MS m/z 315.2 (MþHþ). HRMS (ESI) m/z calcd for MþHþ
315.1492; found 315.1490.
4.6.1.10. 2-(2-(benzyloxy)phenyl)-2-((2-chlorophenyl) amino)aceto-
nitrile (D29). This compound was prepared by using the procedure
as described for D2. Yield 240 mg (73%). 1H NMR (400 MHz, CDCl3):
4.6.1.3. 2-(4-(benzyloxy)phenyl)-2-(phenylamino) acetonitrile (D22).
This compound was prepared by using the procedure as described
for D2. Yield 243 mg (82%).
d
¼ 4.74 (br, 1H, NH), 5.10 (s, 2H, OCH2Ar), 5.45 (s, 1H, CH(CN)N),
6.82e6.87(m, 2H, HAr), 7.05e7.07(m, 1H, HAr), 7.20e7.23 (m, 3H,
1H NMR (400 MHz, CDCl3):
d
¼ 5.10 (s, 2H, OCH2Ar), 5.36 (s, 1H,
H
Ar), 7.33e7.45 (m, 7H, HAr). ESI-MS m/z 349.1 (MþHþ). HRMS (ESI)
CH(CN)N), 6.78 (d, J ¼ 7.8 Hz, 2H, HAr), 6.89e6.92 (m, 1H, HAr), 7.02-
7.05 (m, 2H, HAr), 7.28e7.35 (m, 2H, HAr), 7.36e7.46 (m, 5H, HAr),
7.49e7.52 (m, 2H, HAr). ESI-MS m/z 315.1 (MþHþ). HRMS (ESI) m/z
calcd for MþNaþ 337.1311; found 337.1318.
m/z calcd for MþNaþ 371.0922; found 371.0934.
4.6.1.11. 2-(2-(benzyloxy)phenyl)-2-((4-chlorophenyl) amino) aceto-
nitrile (D30). This compound was prepared by using the procedure
as described for D2. Yield 250 mg (76%). 1H NMR (400 MHz, CDCl3):
4.6.1.4. 2-phenyl-2-(phenylamino) acetonitrile (D23). This com-
pound was prepared by using the procedure as described for D2.
Yield 349 mg (89%).
d
¼ 5.14e5.17 (m, 2H, OCH2Ar), 5.56 (s, 1H, CH(CN)N), 6.62e6.64 (d,
J ¼ 8.8 Hz, 2H, HAr), 7.03e7.06 (m, 2H, HAr), 7.16e7.18 (d, J ¼ 8.8 Hz,
2H, HAr), 7.34e7.42 (m, 6H, HAr), 7.50e7.52 (m, 1H, HAr). 13C NMR
(400 MHz, CDCl3): 46.50, 70.65, 112.77, 115.65, 118.29, 121.52,
122.47, 124.92, 127.39, 128.34, 128.76, 128.84, 129.36, 131.16, 135.95,
143.67, 155.91. ESI-MS: m/z 349.2 (MþHþ). HRMS (ESI) m/z calcd for
MþNaþ 371.0922; found 371.0910.
1H NMR (500 MHz, CDCl3):
d
¼ 4.03 (s, 1H, NH), 5.43 (s, 1H,
CH(CN)N), 6.77e6.79 (d, J ¼ 10 Hz, 2H, HAr), 6.89e6.92 (m, 1H, HAr),
7.25e7.29 (m, 2H, HAr), 7.42e7.48 (m, 3H, HAr ), 7.60e7.61 (d,
J ¼ 5 Hz, 2H, HAr). 13C NMR (500 MHz, CDCl3):
d
¼ 50.26, 114.32,
118.16, 120.31, 127.25, 129.34, 129.54, 129.57, 133.98, 144.68. ESI-MS
m/z 209.1 (MþHþ). HRMS (ESI) m/z calcd for MþNaþ 231.0893;
found 231.0896.
4.6.1.12. 2-(2-(benzyloxy) phenyl)-2-((2-hydroxyphenyl) amino)
acetonitrile (D31). This compound was prepared by using the
procedure as described for D2. Yield 224 mg (72%). 1H NMR
4.6.1.5. 2-(2-(benzyloxy)phenyl)-2-((2-methoxyphenyl)
amino)
(400 MHz, CDCl3):
d
¼ 3.50 (s, 1H), 5.24 (s, 2H, OCH2Ar), 5.47 (s, d,
acetonitrile (D24). This compound was prepared by using the
CH(CN)N), 6.81e7.07 (m, 6H, HAr), 7.34e7.57 (m, 7H, HAr). ESI-MS
m/z 331.1 (MþHþ). HRMS (ESI) m/z calcd for MþNaþ 353.1260;
found 353.1254.
procedure as described for D2. Yield 250 mg (77%). 1H NMR
(400 MHz, CDCl3):
d
¼ 3.77 (s, 3H, CH3O), 5.21 (s, 2H, OCH2Ar), 5.69
(s, 1H, CH(CN)N), 6.96e6.80 (m, 4H, HAr), 7.06 (t, J ¼ 7.8 Hz, 2H) ,
7.48e7.31 (m, 6H, HAr), 7.58 (dd, J ¼ 7.4, 1.5 Hz, 1H, HAr). ESI-MS m/z
345.1 (MþHþ). HRMS (ESI) m/z calcd for MþHþ 345.1598; found
345.1597.
4.6.1.13. 2-(2-(benzyloxy)
phenyl)-2-((4-hydroxyphenyl)amino)
acetonitrile (D32). This compound was prepared by using the
procedure as described for D2. Yield 240 mg (77%). 1H NMR
(400 MHz, CDCl3):
d
¼ 5.09 (s, 2H, OCH2Ar), 5.30 (s, 1H, CH(CN)N),
4.6.1.6. 2-(2-(benzyloxy)phenyl)-2-((3-methoxyphenyl)
amino)
6.68e6.79 (m, 4H, HAr), 7.01e7.04 (m, 1H, HAr), 7.18e7.22 (m, 2H,
acetonitrile (D25). This compound was prepared by using the
H
Ar), 7.32e7.45 (m, 6H, HAr). ESI-MS: m/z 331.1 (MþHþ). HRMS (ESI)
procedure as described for D2. Yield 243 mg (75%). 1H NMR
m/z calcd for MþNaþ 353.1260; found 353.1260.
(400 MHz, acetone-d6):
d
¼ 3.73 (s, 3H, CH3O) 5.15 (s, 2H, OCH2Ar),
5.83e5.85 (d, J ¼ 9.2 Hz, 1H, CH(CN)N), 6.37e6.39 (m, 1H, HAr ),
6.49e6.51 (m, 2H, HAr), 7.07e7.50 (m, 10H, HAr). ESI-MS m/z 345.2
(MþHþ). HRMS (ESI) m/z calcd for MþHþ 345.1598; found
345.1594.
4.6.1.14. 2-(((2-(benzyloxy)phenyl)(cyano)methyl)amino)benzoni-
trile (D33). This compound was prepared by using the procedure as
described for D2. Yield 243 mg (76%). 1H NMR (400 MHz, CDCl3):
d
¼ 5.12-5.14 (d, J ¼ 7.8 Hz, NH), 5.20 (s, 2H, OCH2Ar), 5.70e5.72 (d,
J ¼ 7.8 Hz, CH(CN)N), 6.86e7.10 (m, 4H, HAr), 7.31e7.58 (m, 9H, HAr).
ESI-MS m/z 340.1 (MþHþ), 362.0 (MþNaþ). HRMS (ESI) m/z calcd
for MþNaþ 362.1264; found 362.1264.
4.6.1.7. 2-(2-(benzyloxy)phenyl)-2-((4
methoxyphenyl)
amino)
acetonitrile (D26). This compound was prepared by using the
procedure as described for D2. Yield 263 mg (81%). 1H NMR
(400 MHz, CDCl3):
d
¼ 3.77 (s, 3H, CH3O), 5.18e5.19 (d, J ¼ 3.5 Hz,
4.6.1.15. 2-(2-(benzyloxy) phenyl)-2-((2, 6-dimethylphenyl) amino)
acetonitrile (D35). This compound was prepared by using the
procedure as described for D2. Yield 232 mg (72%). 1H NMR
2H, OCH2Ar), 5.50 (s, 1H, CH(CN)N), 6.70e6.82 (m, 4H, HAr), 7.02e
7.06 (m, 2H, HAr), 7.34-7.52 (m, 7H, HAr). ESI-MS: m/z 345.2 (MþHþ).
HRMS (ESI) m/z calcd for MþHþ 345.1598; found 345.1599.
(400 MHz, CDCl3):
d
¼ 2.35 (s, 6H, 2ꢂ CH3) 5.10 (s, 1H, CH(CN)N),
5.11 (s, 2H, OCH2Ar), 6.96e7.09 (m, 4H, HAr), 7.24e7.29 (m, 2H, HAr),
7.35e7.47 (m, 6H, HAr). ESI-MS: m/z 343.1 (MþHþ). HRMS (ESI) m/z
calcd for MþHþ 343.1805; found 343.1810.
4.6.1.8. 2-(2-(benzyloxy)phenyl)-2-(o-tolylamino)acetonitrile (D27).
This compound was prepared by using the procedure as described
for D2. Yield 260 mg (84%).. 1H NMR (400 MHz, CDCl3):
d
¼ 1.98 (s,
3H, CH3), 5.15e5.22 (m, 2H, OCH2Ar), 5.58 (s, s,1H, CH(CN)N), 6.78e
6.81 (m, 2H, HAr), 7.02e7.07 (m, 3H, HAr), 7.14e7.18 (m, 1H, HAr),
7.31e7.40 (m, 7H, HAr). ESI-MS: m/z 329.2 (MþHþ). HRMS (ESI) m/z
calcd for MþHþ 329.1648; found 329.1651.
4.6.1.16. 2-(2-(benzyloxy)phenyl)-2-((2-nitrophenyl) amino) aceto-
nitrile (D36). This compound was prepared by using the procedure
as described for D2. Yield 274 mg (81%). 1H NMR (400 MHz, CDCl3):
d
¼ 5.10 (s, 2H, OCH2Ar), 5.55e5.57 (d, J ¼ 6.8 Hz, 1H, CH(CN)N),
6.89e6.93 (m, 1H, HAr), 6.99e7.01 (d, J ¼ 8.6 Hz, 1H, HAr), 7.06e7.08
(m, 1H, HAr), 7.18e7.20 (m, 2H, HAr), 7.34e7.44 (m, 5H, HAr), 7.56e
7.60 (m, 1H, HAr), 8.22e8.27 (m, 2H, HAr). ESI-MS: m/z 360.3
4.6.1.9. 2-(2-(benzyloxy) phenyl)-2-(p-tolylamino)acetonitrile (D28).
This compound was prepared by using the procedure as described

X. Deng et al. / European Journal of Medicinal Chemistry 73 (2014) 1e7
7
(MþHþ), 382.1 (MþNaþ). HRMS (ESI) m/z calcd for MþNaþ
triturated, washed with water, filtered, and dried under vacuum. The
crude product was recrystallized with ethyl acetate/hexane to give
the required product, 61 mg, 27% yield. 1H NMR (400 MHz, DMSO-
382.1162; found 382.1153.
4.6.1.17. 2-(2-(benzyloxy) phenyl)-2-(pyridin-3-ylamino) acetonitrile
(D37). This compound was prepared by using the procedure as
described for D2. Yield 217 mg (73%). 1H NMR (400 MHz, CDCl3):
d6):
d
¼ 5.10e5.18 (m, 2H, OCH2Ar), 6.29 (s, 1H, PhCHN), 6.55e6.59
(m, 3H, HAr), 6.92e7.38 (m, 10H, HAr). ESI-MS m/z 358.1 (MþHþ).
HRMS (ESI) m/z calcd for MþHþ 358.1662; found 358.1659.
d
¼ 4.54e4.56 (d, J ¼ 9 Hz, 1H, NH), 5.17 (s, 2H, OCH2Ar), 5.59e5.61
(d, J ¼ 8.6 Hz, CH(CN)N), 7.01e7.16 (m, 4H, HAr), 7.33e7.53 (m, 7H,
Acknowledgments
H
Ar), 8.08e8.12 (m, 2H, HAr). ESI-MS m/z 316.1 (MþHþ). HRMS (ESI)
m/z calcd for MþHþ 316.1444; found 316.1443.
We gratefully acknowledge the financial support from the Chi-
nese National Natural Science Foundation (Grants No.30973641,
20902013), Specialized Research Fund for the Doctoral Program of
Higher Education (No. 20090071110054), Chinese Ministry of Ed-
ucation, the open grant of the State Key Laboratory of Bio-organic
and Natural Products Chemistry, CAS, the open grant of Institute
of Bioscience, Fudan University, and the grant of the Science and
Technology Research Key Projects, the Science and Technology
Commission of Shanghai Municipality (No.11431920102), Fudan
University Graduate Innovation Foundation (EYF301069).
4.6.1.18. 2-(2-(benzyloxy)phenyl)-2-((furan-2-ylmethyl)
amino)
acetonitrile (D38). This compound was prepared by using the
procedure as described for D2. Yield 234 mg (78%). 1H NMR
(400 MHz, CDCl3):
d
¼ 3.90e4.02 (m, 2H, OCH2Ar), 4.93 (s, 1H,
CH(CN)N), 5.10e5.18 (m, 2H, NCH2Furan), 6.22e6.28 (m, 2H, HAr),
6.97e7.02 (m, 2H, HAr), 7.31e7.44 (m, 8H, HAr). ESI-MS m/z 319.1
(MþHþ). HRMS (ESI) m/z calcd for MþHþ 319.1441; found 319.1439.
4.6.1.19. N-(2-(benzyloxy)benzyl) aniline (D39). A solution of 2-
(benzyloxy)benzaldehyde (200.00 mg, 0.942 mmol) and aniline
(87.75 mg, 0.942 mmol) in methanol (5 mL) was stirred at rt for 2 h.
NaBH4 (178.25 mg, 4.71 mmol) was added to the reaction mixture
by portionwise, meanwhile, acetic acid was used for adjusting pH
5e6. After 2 h, the solvent was evaporated in vacuo; the residue
was diluted with ethyl acetate/water. The organic layer was washed
with saturated brine, dried over anhydrous MgSO4 and concen-
trated under vacuum. The residual was triturated with ethyl ace-
tate/hexane (1/5), filtered to give the required product, 84 mg, 31%
References
[1] J.B. Dong, J. Liu, B. Lou, Z.Q. Li, X. Ye, M.P. Wu, X.C. Jiang, J. Lipid Res. 47 (6)
(2006) 1307e1314.
[2] Y. Zhang, F. Lin, X. Deng, R. Wang, D. Ye, Chin. J. Chem. 29 (8) (2011)
1567e1575.
[3] Z. Li, Y. Fan, J. Liu, Y. Li, C. Huan, H.H. Bui, M.-S. Kuo, T.-S. Park, G. Cao, X.-
C. Jiang, Arterioscler. Thromb. Vasc. Biol. 32 (7) (2012) 1577e1584.
[4] Z. Li, H. Zhang, J. Liu, C.P. Liang, Y. Li, G. Teitelman, T. Beyer, H.H. Bui,
D.A. Peake, Y. Zhang, P.E. Sanders, M.S. Kuo, T.S. Park, G. Cao, X.C. Jiang, Mol.
Cell Biol. 31 (20) (2011) 4205e4218.
[5] X.-C. Jiang, C. Yeang, Z. Li, M. Chakraborty, J. Liu, H. Zhang, Y. Fan, Clin. Lipidol.
4 (5) (2009) 595e609.
[6] K. Huitema, J. van den Dikkenberg, J. Brouwers, J.C.M. Holthuis, EMBO J. 23 (1)
(2004) 33e44.
[7] R.M. Adibhatla, J.F. Hatcher, A. Gusain, Neurochem. Res. 37 (4) (2012)
671e679.
[8] A.P. Bai, G.P. Meier, Y. Wang, C. Luberto, Y.A. Hannun, D.H. Zhou, J. Pharmacol.
Exp. Ther. 309 (3) (2004) 1051e1059.
[9] A.M. Meng, C. Luberto, P. Meier, A.P. Bai, X.F. Yang, Y.A. Hannun, D.H. Zhou,
Exp. Cell Res. 292 (2) (2004) 385e392.
[10] Y. Kimura, J. Aoki, M. Kohno, H. Ooka, T. Tsuruo, O. Nakanishi, Cancer Che-
mother. Pharmacol. 49 (2002) 322e328.
[11] M. Tatsumi, T. Tsuruo, T. Nishimura, Eur. J. Nucl. Med. Mol. Imaging 29 (3)
(2002) 288e294.
[12] J. Robert, Curr. Opin. Invest. Drugs (Thomson Sci.) 5 (12) (2004) 1340e1347.
[13] X. Deng, H. Sun, X. Gao, H. Gong, W. Lu, Y. Chu, L. Zhou, D. Ye, Anal. Lett. 45
(12) (2012) 1581e1589.
[14] I. Tabas, K.J. Williams, J. Boren, Circulation 116 (16) (2007) 1832e1844.
[15] M.R. Hojjati, Z. Li, X.C. Jiang, BBA-Mol. Cell Biol. L. 1737 (1) (2005) 44e51.
[16] M. Takamura, Y. Hamashima, H. Usuda, M. Kanai, M. Shibasaki, Angew. Chem.
Int. Ed. 112 (9) (2000) 1716e1718.
[17] R. Olivera, R. SanMartin, E. Domínguez, X. Solans, M.K. Urtiaga,
M.a.I. Arriortua, J. Org. Chem. 65 (2000) 6398e6411.
[18] F. Himo, Z.P. Demko, L. Noodleman, K.B. Sharpless, J. Am. Chem. Soc. 124
(2002) 12210e12216.
[19] A.J. Davies, M.S. Ashwood, I.F. Cottrell, Synthetic Commun. 30 (6) (2000)
1095e1102.
[20] T. Lübbers, P. Angehrn, H. Gmünderb, S. Herziga, Bioorg. Med. Chem. Lett.
2007 (17) (2007) 4708e4714.
yield. 1H NMR (400 MHz, DMSO-d6):
d
¼ 4.64 (s, 2H, OCH2Ar), 5.17
(s, 2H, NCH2Ar), 6.51e6.57 (m, 2H, HAr), 6.85e6.89 (m, 1H, HAr),
7.03e7.46 (m, 11H, HAr). ESI-MS m/z 290.2 (MþHþ). HRMS (ESI) m/z
calcd for MþHþ 290.1539; found 290.1545.
4.6.1.20. 2-(2-(benzyloxy)phenyl)-2-(phenylamino)
acetamide
(D40). To a suspension of D2 (200 mg, 0.636 mmol), potassium car-
bonate (264 mg, 1.91 mmol) in water (5 mL), 30% hydrogen peroxide
(36 m
L, 0.318 mmol)wasadded dropwiseat0 ꢀC. Thereaction mixture
waswarmedtortand stirred continuouslyfor2h. Theprecipitatewas
filtered and washed with water, dried under vacuum; the crude
product was recrystallized with ethyl acetate/hexane to give the
required product, 91 mg, 43% yield. 1H NMR (400 MHz, CDCl3):
d
¼ 5.23e5.30 (m, 4H, OCH2Ar and NH2CO), 5.40 (s, 1H, CH(CN)N),
6.49e6.51(d, J ¼ 8 Hz, 2H, HAr), 6.64 (s, 1H, NH(CH)(CN)), 6.68e6.72
(m,1H, HAr), 6.98e7.01 (m, 1H, HAr), 7.10e7.14(m, 3H, HAr), 7.28e7.33
(m,1H, HAr), 7.37e7.54 (m, 6H, HAr). ESI-MS m/z 333.2 (MþHþ). HRMS
(ESI) m/z calcd for MþHþ 333.1598; found 333.1602.
4.6.1.21. N-((2-(benzyloxy)phenyl) (1H-tetrazol-5-yl) methyl) aniline
(D41). A solution of D2 (200 mg, 0.636 mmol) and sodium azide
(207mg, 3.18mmol)inethanol(5mL)wasstirredandrefluxedfor2h.
The solvent was evaporated in reduced pressure; the residue was