Synthesis and evaluation of sphingoid analogs as inhibitors of sphingosine kinases

Article abstract of DOI:10.1016/j.bmc.2005.02.053

Sphingosine 1-phosphate (S1P), a product of sphingosine kinases (SphK), mediates diverse biological processes such as cell differentiation, proliferation, motility, and apoptosis. In an effort to search and identify specific inhibitors of human SphK, the inhibitory effects of synthetic sphingoid analogs on kinase activity were examined. Among the analogs tested, we found two, SG12 and SG14, that have specific inhibitory effects on hSphK2. N,N-Dimethylsphingosine (DMS), a well-known SphK inhibitor, displayed inhibitory effects for both SphK1 and SphK2, as well as protein kinase C. In contrast, SG12 and SG14 exhibited selective inhibitory effects on hSphK2. Furthermore, SG14 did not affect PKC. In isolated platelets, SG14 blocked the conversion of sphingosine into sphingosine 1-phosphate significantly. This is the first report on the identification of a hSphK2-specific inhibitor, which may provide a useful tool for studying the biological functions of hSphK2.

Full text of DOI:10.1016/j.bmc.2005.02.053

Bioorganic & Medicinal Chemistry 13 (2005) 3475–3485
Synthesis and evaluation of sphingoid analogs as inhibitors
of sphingosine kinases
Jin-Wook Kim,a,c,e Yong-Woo Kim,b Yuichi Inagaki,c You-A Hwang,a
Susumu Mitsutake,c Yeon-Woo Ryu,e Won Koo Lee,b Hyun-Joon Ha,d
Chang-Seo Parka and Yasuyuki Igarashic,*
aDoosan Biotech, 39-3, Seongbok-dong, Yongin-si, Gyeonggi-do 449-795, Korea
bDepartment of Chemistry, Program of Integrated Biotechnology, Sogang University. Sinsu-dong, Mapo-ku, Seoul 121-742, Korea
cDepartment of Biomembrane and Biofunctional Chemistry, Graduate School of Pharmaceutical Science,
Hokkaido University, Kita 12, Nishi 6, Kita-ku, Sapporo 060-0812, Japan
dDepartment of Chemistry, Hankuk University of Foreign Studies, 89 Wangsan-ri Mohyun, Yongin-si, Kyounggi-do 449-791, Korea
eDepartment of Molecular Science and Technology, Ajou University, San 5, Wonchun-dong, Yeongtong-gu, Suwon 443-749, Korea
Received 2 February 2005; revised 25 February 2005; accepted 25 February 2005
Available online 2 April 2005
Abstract—Sphingosine 1-phosphate (S1P), a product of sphingosine kinases (SphK), mediates diverse biological processes such as
cell differentiation, proliferation, motility, and apoptosis. In an effort to search and identify specific inhibitors of human SphK, the
inhibitory effects of synthetic sphingoid analogs on kinase activity were examined. Among the analogs tested, we found two, SG12
and SG14, that have specific inhibitory effects on hSphK2. N,N-Dimethylsphingosine (DMS), a well-known SphK inhibitor, dis-
played inhibitory effects for both SphK1 and SphK2, as well as protein kinase C. In contrast, SG12 and SG14 exhibited selective
inhibitory effects on hSphK2. Furthermore, SG14 did not affect PKC. In isolated platelets, SG14 blocked the conversion of sphin-
gosine into sphingosine 1-phosphate significantly. This is the first report on the identification of a hSphK2-specific inhibitor, which
may provide a useful tool for studying the biological functions of hSphK2.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
to form S1P,6 has been implicated as playing a role in
determining cell fate because it not only produces anti-
apoptotic S1P but also decreases pro-apoptotic CER
and SPH.
The sphingolipid metabolites ceramide (CER), sphingo-
sine (SPH), and sphingosine 1-phosphate (S1P) play an
important role in the regulation of cell proliferation, sur-
vival, and apoptosis.1,2 S1P mediates diverse biological
processes such as cell differentiation, proliferation,
motility, and apoptosis, both extracellularly by binding
to members of the S1P receptor and intracellularly.3–5
S1P usually stimulates proliferation and suppresses
apoptosis, while CER and SPH inhibit proliferation
and promote apoptosis. Therefore, it has been proposed
that the balance between S1P and SPH/CER may deter-
mine cell fate. One sphingolipid metabolizing enzyme,
sphingosine kinase (SphK), which phosphorylates SPH
SphK is a highly conserved lipid kinase, and the distri-
bution of its activity varies among cell types and tis-
sues.7,8 SphK activity is increased by many growth
and survival factors, including platelet-derived growth
factor,9 nerve growth factor,10 epidermal growth fac-
tor,11 and vascular endothelial growth factor.12 To date,
two mammalian SphKs, SphK1 and SphK2, have been
identified,13,14 and differences between the two have
been recognized. While SphK1 is predominantly local-
ized in the cytoplasm, SphK2 is localized in the nucleus
as well as in the cytosol.15,16 SphK1 stimulates cell
growth and survival, but SphK2 enhances apoptosis
and suppresses cellular growth in various cell types.17
Additionally, each has its own substrate specificity.
SphK1 phosphorylates D-erythro-sphingosine (SPH)
and, to a lesser extent, dihydrosphingosine (DHS), but
Keywords: Sphingosine analog; Sphingosine kinase; Inhibitor;
Aziridine.
*
Corresponding author. Tel.: +81 11 706 3970; fax: +81 11 706
4986; e-mail: yigarash@pharm.hokudai.ac.jp
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.02.053

3476
J.-W. Kim et al. / Bioorg. Med. Chem. 13 (2005) 3475–3485
acts very poorly on phytosphingosine (PHS). On the
other hand, SphK2 phosphorylates both D-erythro-
sphingosine and dihydrosphingosine, with nearly the
same efficiency, and phytosphingosine, though to a les-
ser extent. Furthermore, the immunomodulatory pro-
drug FTY-720 is phosphorylated 30-fold more
efficiently by SphK2 than SphK1.18,19
In this study, we prepared various novel sphingoid ana-
logs to find specific inhibitors of SphK1 and/or SphK2.
We found SphK2-specific inhibitor that did not affect
SphK1 or protein kinase C (PKC) activity. These results
provide the first SphK2-specific inhibitor. Such an inhib-
itor will be useful not only in understanding the SphK
functions but also in designing therapeutics based on
the sphingoid structure.
SPH, DHS, and PHS, which can serve as precursors of
many biologically important compounds such as cere-
brosides, gangliosides, and sphingomyelin, are com-
posed of three structural units: a long-chain aliphatic
2-amino-1,3-diol, a fatty acid, and a polar head group.20
The inherent structural variation of sphingosine and its
derivatives attracts great interest, because of the diverse
biological activities of these lipids including the cell regu-
lation and signal transduction.21–23 We recently reported
that the key structural unit of sphingolipids, enantio-
merically pure 2-amino-1,3-propanediols, could be syn-
thesized from suitably substituted chiral aziridine.24
Various enantiomerically pure 2-acylaziridines were pre-
pared efficiently from the corresponding aziridine-2-car-
boxylate via Weinrebꢀs amide25–28 and organometallic
compounds. The carbonyl group was reduced stereo-
selectively by a suitable reducing agent, and then the
aziridine ring was regioselectively opened with AcOH,
pyrrolidine, and morpholine. Using this methodology,
we have prepared various sphingosine analogues, with
high yields.
2. Results
2.1. Synthesis of sphingoid analogs
The reaction of N,O-dimethyl hydroxylamine hydro-
chloride and a commercially available N-[(R)-(+)-a-
methylbenzyl]-2(S)-carboxylic acid menthol ester 1a
(Scheme 1), in the presence of I-PrMgCl in THF pro-
vided the Weinrebꢀs amide 2, with a 92% yield. The
Weinrebꢀs amide 2 was reacted with various organome-
tallic compounds to provide the corresponding ketones
3 in high yields. Reduction of the ketones 3 with LAH
resulted in a diastereomeric mixture of hydroxy azir-
idines 4a,b (approximately R/S = 1/1 to 2/1), which were
readily separable by flash chromatography.30,31 A regio-
selective ring opening reaction of aziridine-2-methanols
4a,b by AcOH and various amine nucleophiles provided
5a,b. TMS-I was used for the activation of the basic ring
nitrogen of the aziridine-2-methanols and was also used
as the source of the iodide nucleophiles to attack the less
substituted C(3) of the aziridine.32 Aziridine-2-methanol
readily opens up regioselectively, and these results
prompted us to investigate the preparation of enantio-
merically pure 7a,b by debenzylation and amide bond
formation. The debenzylation reaction did not proceed
via the usual procedure with Pearlmanꢀs catalyst, and
the reactivity was influenced by solvent, temperature
and pressure. N-acylation provided various ceramide
analogues, and the results are shown in Scheme 1.
Sphingolipid metabolism is often found to be abnormal
in a number of apoptosis-related pathological condi-
tions. In many cancer types, SphK inhibitors strongly
induce apoptosis and enhance sensitivity to gamma irra-
diation and anti-cancer drugs.29 Thus, SphK inhibitors
have been implicated as a novel sensitizer for cancer
therapeutics. Furthermore, understanding SphK func-
tions could uncover new targets for therapeutic
applications.
Scheme 1. Reagents and conditions: (i) Weinrebꢀs amine, i-PrMgCl, THF, 0 ꢁC; (ii) benzyl bromide, Mg, THF, 50 ꢁC; (iii) LAH, THF, 0 ꢁC, 85%;
(iv)* (a) AcOH, CH2Cl2, rt, (b) KOH, EtOH, 90% (two steps) or NaI, TMSCl, pyrrolidine, MeCN, 90%; (v) Pd(OH)2, H2, (100 psi), MeOH; (vi) Acyl
chloride, NaOH, THF, 76% (two steps).

J.-W. Kim et al. / Bioorg. Med. Chem. 13 (2005) 3475–3485
3477
Scheme 2. Reagents and conditions: (i) LAH, THF, 0 ꢁC; (ii) carbonyldiimidazole, CH2Cl2, rt; (iii) NaI, TMSCl, MeCN, rt, 75% (two steps); (iv)
anisole, MsOH, toluene, 60 ꢁC, 95%; (v) PPh3, DMF, 100 ꢁC, 90%; (vi) 4-octylbenzaldehyde, LHMDS, THF, 0 ꢁC, 80%; (vii) Pd/C, H2, MeOH, rt,
90%; (viii) LiOHÆH2O, EtOH, 70 ꢁC, 90%.
The reduction of the aziridine-2-carboxylate 1b to the
corresponding aziridine-2-methanol was fairly straight-
forward using LAH in THF. The enantiomerically pure
phenylethyl-protected 4(S)-iodomethyl-2-oxazolidinone
was prepared from the aziridine 2(S)-methanol 8
through regiospecific C(3)–N bond cleavage by iodotri-
methylsilane, followed by intramolecular cyclization
with carbonyldiimidazole (CDI).33 The phenylethyl
group on the nitrogen was successfully removed by
heating compound 9 in the presence of anisole and
methanesulfonic acid34 in toluene, to provide the
(4S)-iodomethyl-2-oxazolidinone 10 in an almost quan-
titative yield. The iodide 10 was converted to the
corresponding phosphonium salt 11 in a 90% yield, by
stirring with triphenylphosphine in toluene at 100 ꢁC
for 24 h.35 Treatment of the phosphonium salt 11 with
LiHMDS at À78 ꢁC generated the ylide, which was then
reacted with various benzaldehyde analogues to provide
alkenes in high yield. The utility of the olefination
procedure can be further demonstrated by the ready
conversion of the oxazolidinone to the corresponding
aryl-substituted amino alcohol by hydrogenation (5%
Pd/C, H2) to give the 4-substituted (4R)-phenylethyl-2-
oxazolidinone 13. This was then hydrolyzed in aqueous
EtOH by LiOH to provide (R)-2-amino-4-(4-octylphe-
nyl)butan-1-ol 14, in a 90% yield (Scheme 2).
3.1. Inhibitory effect of sphingoid analogs on in vitro SphK
activity
To evaluate the SphK inhibitory effect of sphingoid ana-
logs, in vitro SphK inhibition assays were performed.
3. Expression of hSphK1 and hSphK2
To examine the inhibitory effect of the synthetic
sphingoid analogs on the hSphK1 and SphK2, Chinese
hamster ovary cells (CHO-K1) were transfected with
pcDNA3-HA, pcDNA3-HA-hSphK1, or pcDNA3-
HA-hSphK2. Twenty-four hours after transfection, the
expression levels of the hSphK1 and hSphK2 were
determined by Western blotting (Fig. 1A). Quantifica-
tion using a densitometer (NIH image analyzer. 1.62)
revealed an expression level of the hSphK1 protein 20-
fold higher than that of hSphK2. Next, we examined
the sphingosine kinase activities of SphK1 and SphK2
Figure 1. Expression and enzymatic activities of hSphK1 and hSphK2.
CHO-K1 cells were transiently transfected with pcDNA3-HA,
pcDNA3-HA-hSphK1, or pcDNA3-HA-hSphK2. A. HA-hSphK1
and HA-hSphK2 proteins were separated by SDS-PAGE and analyzed
by Western blotting with an anti-HA antibody. An anti-tubulin blot
was performed as a loading control. B. Cell lysates were prepared and
SphKs activities were measured with 4 lM D-e-sphingosine,
2 lCi[c-32P] ATP, and 0.5 mM cold ATP, at 37 ꢁC for 30 min.
with
pcDNA3-HA-hSphK1-and
pcDNA3-HA-
hSphK2-transfected cells (Fig. 1B). Despite the much
higher expression level of the protein, the kinase activity
of the hSphK1 transfectant was only about 4-fold higher
than that of the hSphK2. Taken together, these results
imply that hSphK2 has comparable sphingosine kinase
activity to that of hSphK1.

3478
J.-W. Kim et al. / Bioorg. Med. Chem. 13 (2005) 3475–3485
N,N-Dimethylsphingosine (DMS), a well known SphK
inhibitor, was used as a positive control and exhibited
very strong inhibitory effect on both hSphK1 and
hSphK2 (Table 1). More than half of the sphingoid ana-
logs tested displayed inhibitory effects, and in general,
the sphingoid analogs showed a more inhibitory effect
on hSphK2 than hSphK1. From these data, together
with reported results of hSphK2 phosphorylating phyto-
sphingosine and FTY-720 more efficiently than
hSphK1,18,19 we can infer that hSphK2 may have broad-
er substrate availability than hSphK1. Two of the tested
sphingoid analogs, SG12 and SG14, exhibited strong
inhibitory effects against hSphK2 with no effect on
hSphK1 (Fig. 2A and B). The IC50 values of SG12
and SG14 on SphK2 were 22 and 4 lM, respectively,
while the IC50 of the positive control DMS was
16 lM. These results indicate that SG14 has a stronger
inhibitory effect than DMS on hSphK2.
nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-
tetrasolium (Fig. 3). After twenty-four hours, SG12 and
SG14 treatments induced dose-dependent cell death in
CHO-K1 cells. Interestingly, SG15, which is a stereoiso-
mer of SG14 and a weak SphK2 inhibitor, had no cyto-
toxic effect. Though we do not yet know if this cell death
was induced by SphK2 inhibition, these results indicate
that SG12 and SG14 have cytotoxic effects comparable
to those of DMS.
3.3. Inhibition of in vitro protein kinase C activity by
SphK inhibitors
In many studies, DMS is the inhibitor most widely used.
It inhibits SphK1 and SphK2 as well as protein kinase
C
36 and induces cell death in various cell types. It is dif-
ficult to say, however that DMS induces cell death via
SphK inhibition, since it also inhibits PKC. Thus,
DMS is imperfect as an experimental and/or therapeutic
tool. With this in mind, we performed in vitro PKC inhi-
bition assays to determine whether the sphingoid
analogs specifically inhibit SphK2. Our results demon-
strated that SG14 did not inhibit PKC, while DMS
inhibited PKC very strongly (Fig. 4). On the other hand,
SG12 exhibited moderate inhibitory effects on PKC. In
summary, SG14 can be used as a SphK2-specific inhib-
itor which does not affect PKC or SphK1. Thus, SG14
can overcome the imperfections of DMS and has great
advantages as an experimental and/or therapeutic tool.
Interestingly, SG7, in which the R4 –OH of SG12 is
substituted with –H, lost its hSphK2 selectivity, suggest-
ing the importance of the –OH substitution in its speci-
ficity. In addition, SG9 and SG15, stereoisomers of
SG12 and SG14, respectively, displayed weak inhibitory
effects on hSphK2. These results indicate that the stereo-
chemistry of SG12 and SG14 is essential for the inhibi-
tory effect.
3.2. Cytotoxic effects of a hSphK2-specific inhibitor in
CHO-K1 cells
3.4. Inhibition of the phosphorylation [3H] sphingosine
into [3H] sphingosine 1-phosphate in platelets
It is well understood that SphK inhibitors cause cell
death by inducing the accumulation of ceramide. In
many studies using various cell types, DMS induces cell
death very strongly. To investigate the influence of
hSphK2-specific inhibitors on cell growth, their cyto-
toxic effects were measured using 2-(2-methoxy-4-
To determine whether SG12 and SG14 can inhibit
endogenous SphK activity, we studied the effect of these
SphK inhibitors on [3H] sphingosine metabolism using
platelets. Blood platelets store S1P abundantly, possibly
Table 1. Inhibitory effects of sphingoid analogues on human sphingosine kinase 1, 2
ID
R1
R2
R3
R4
hSphK1 (%)
hSphK2 (%)
DMS
SG-1
SG-2
24
98
97
149
98
66
71
54
70
95
96
97
72
98
98
95
23
56
104
91
71
49
85
44
83
91
46
25
96
21
65
60
H
OH
OH
Tri-decanoyl
Tridecanoyl
Palmitoyl
Palmitoyl
Tridecanoyl
Palmitoyl
H
OH (S)
OH (R)
OH (S)
OH (R)
H
H
SG-3
SG-4
H
OH
OH
H
SG-5
SG-6
H
OH
H
H
OH
H
SG-7
SG-9
Octyl
Octyl
H
H
OH
H
OH (S)
OH (R)
H
SG-10
SG-11
SG-12
SG-13
SG-14
SG-15
SG-16
Morpholine
Pyrrolidine
OH
H
H
Decanoyl
H
Octyl
H
OH (R)
OH (S)
OH (S)
OH (R)
OH (S)
Pyrrolidine
Pyrrolidine
Pyrrolidine
Morpholine
Decanoyl
Stearoyl
Stearoyl
Decanoyl
H
H
H
Inhibitory effects toward hSphK1 and hSphK2 activities were compared at inhibitor concentrations of 50 lM (% untreated control); These data
represent an average of three different experiments; DMS was used as a positive control.

J.-W. Kim et al. / Bioorg. Med. Chem. 13 (2005) 3475–3485
3479
Figure 4. Effect of hSphK2-specific inhibitors on in vitro protein
kinase C activity. To monitor the influence of the hSphK2-specific
inhibitors on PKC activity, the indicated quantities of inhibitors were
supplemented into a PKC assay system containing phosphatidylserine,
diacylglycerol, [c-32P]ATP and the biotinylated peptide Neurogranin
(AAKIQAS*FRGHMARKK). The reaction was initiated by adding
5 ng PKC (an enzyme mixture which consisting of the a, b and c
isoforms with lesser amounts of the d and f isoforms). Data are shown
as the percentage of live cells relative to untreated controls from three
independent experiments (mean S.D).
the result of an SphK activity and a lack of S1P lyase
activity.4 In fact, S1P is produced very rapidly from
exogenously added SPH. Fifteen minutes after adding
[3H] SPH to platelets, about 80% was converted into
Figure 2. In vitro sphingosine kinase inhibition by sphingoid analogs.
(A) Inhibition assays of sphingoid analogs on hSphK1. 1 lg protein
from lysates of hSphK1-overexpressing CHO-K1 cells were incubated
with 4 lM D-e-sphingosine, 2 lCi[c-32P] ATP, and 0.5 mM cold ATP
at 37 ꢁC for 30 min. (B) Inhibition assays of sphingoid analogs on
hSphK2. Assays were performed as in A but in the presence of
200 mM KCl and using lysates from the hSphK2-overexpressing
CHO-K1 cells. Data are shown as the percentage of SphK activity
relative to untreated controls from three independent experiments
(mean S.D).
Figure 5. Effect of SphK inhibitors on [3H] Sphingosine metabolism.
Platelet suspensions were pretreated for 5 min with increasing concen-
trations of SphK inhibitors and then incubated with [3H] Sphingosine
for 15 min. (A) Autoradiography. (B) Quantification of the autoradi-
ography was performed. Data are shown as the percentage of
phosphorylated [3H] sphingosine (S1P) in the total of phosphorylated
and unphosphorylated forms of [3H] Sphingosine. Lane A: negative
control, B–D: DMS 5 lM, 20 lM, 50 lM; E–G: SG12 5 lM, 20 lM,
50 lM; H–J: SG14 5 lM, 20 lM, 50 lM.
Figure 3. Cytotoxic effects of hSphK inhibitors. The cytotoxic effect
was measured using tetrasolium salt. CHO cells (5 · 103 cell/100 lL)
were incubated with the indicated inhibitors at 37 ꢁC for 24 h, then the
absorbance of the culture supernatant was measured by spectropho-
tometry at a wavelength of 450 nm. Data are shown as the percentage
of live cells relative to untreated controls from three independent
experiments (mean S.D).

3480
J.-W. Kim et al. / Bioorg. Med. Chem. 13 (2005) 3475–3485
[3H] S1P (Fig. 5A and B). In the presence of the SphK
inhibitors, however, this conversion was inhibited in
a dose-dependent manner. In the presence of 50 lM
DMS, [3H] SPH conversion into [3H] S1P was reduced
to 15%. In the same concentration of SG12 or SG14,
the conversion was reduced to 40% or 60%, respectively
(Fig. 5B). The more significant inhibitory effect of the
non-selective DMS, in contrast to the more moderate ef-
fects of SphK2-specific SG12 and SG14, is probably due
to the considerable contribution in platelets of SphK1
activity.8 These results imply that SG12 and SG14 have
selective inhibitory activities on endogenous SphKs in
platelets. Although it is still unknown exactly what kind
of SphKs exist in platelets,7,8 these data suggests the
possibility that considerable SphK2 activity is present.
cifically inhibit SphK2 activity on these substrates, in vi-
tro SphK inhibition assays were performed. First, we
examined the relative phosphorylation efficiency for
each substrate (Fig. 6B). The phosphorylation activity
of SphK1 on sphingosine was 7-fold and 80-fold higher
compared to its activity on phytosphingosine and
FTY720, respectively (Fig. 6B). In contrast, the
phosphorylation activity of SphK2 on sphingosine was
0.5-fold less compared to that on phytosphingosine
and 3-fold higher compared to that on FTY720. As
expression level of the hSphK1 protein is 20-fold higher
than that of hSphK 2 (Fig. 1A), SphK2 phosphorylated
phytosphingosine more efficiently than SphK1, by a
20-fold margin. In addition, SphK2 phosphorylated
FTY720 more efficiently than SphK1, by 37-fold. These
results are consistent with previous reports.18,19
3.5. Inhibitory effects of SG14 on SphK2 activity
toward various substrate
Next, we examined the inhibitory effect of SG14 on the
phosphorylation of sphingosine (SPH), phytosphingo-
sine (PHS), and FTY720. DMS inhibited the phosphor-
ylation of all three substrates by either SphK1 or
SphK2. However SG14 inhibited only the SphK2 phos-
phorylation of the three different substrates (Fig. 6A).
Through this result, we can verify that SG14 specifically
inhibits SphK2, regardless of the substrate involved.
SphK2 can phosphorylate phytosphingosine and the
prodrug FTY720 very efficiently compared to
SphK1.18,19 To verify whether SG12 and SG14 can spe-
4. Discussion
In this study, we prepared various sphingoid analogs
using suitably substituted chiral aziridine. Various enan-
tiomerically pure 2-acylaziridines were produced from
the corresponding aziridine-2-carboxylate via Weinrebꢀs
amide and organometallic compounds. Reduction of the
carbonyl group and aziridine ring opening allows for the
synthesis of a variety of sphingosine analogues, with
high yields.
Among the analogs tested in this study, we found spe-
cific inhibitors for SphK2. Until now, the well known
SphKs inhibitor DMS was used most widely in studies
even though it lacks subtype selectivity and also inhibits
PKC. In contrast, our novel sphingoid analog SG14 spe-
cifically inhibited SphK2 and did not affect SphK1 or
PKC. Interestingly, SG15, a stereoisomer of SG14,
lacked most of the inhibitory activity, indicating that
the stereochemistry may be essential for the inhibitory
effect and its selectivity.
In an experiment using platelets, DMS, SG12, and SG14
caused the accumulation of SPH by blocking the con-
version of SPH into S1P. The inhibitory effect of the
non-selective DMS was the strongest, while that of
SG12 or SG14 was more moderate. There is consider-
able SphK1 activity in platelets,8 so this lower inhibition
likely reflects the SphK2-selectivity of the analogs. In
addition, this result provided evidence that SG12 and
SG14 are able to inhibit endogeneous SphK2 although
the presence of other kinases, which might also be inhib-
ited by these analogs, cannot be ruled out.
Figure 6. Effects of SphK inhibitors on phosphorylation of various
substrates. In vitro sphingosine kinase inhibition assays were per-
formed on a variety of substrates, including sphingosine (SPH),
phytosphingosine (PHS) and immunosuppressive prodrug FTY720.
(A) Comparative SphK activity against various substrates. Data are
shown as the percentage of SphK activity relative to the SphK1
activity on sphingosine. (B) Inhibition activity of DMS and SG14
against SphK activity toward various substrates. Inhibitors were used
at the concentration of 20 lM. Data are shown as the percentage of
SphK activity relative to untreated controls from three independent
experiments (mean S.D).
The exact mechanism and biological functions of SphK
regulation are not clear, but sphingolipid metabolism is

J.-W. Kim et al. / Bioorg. Med. Chem. 13 (2005) 3475–3485
3481
often found to be unusual in diseases like cancer. Signifi-
cantly elevated SphK expression has been observed in
solid tumors, as compared with normal tissue from same
patients.37 Also, cells overexpressing SphK are trans-
forming and tumorigenic.38 Conversely, inhibition of
SphK has been shown to suppress tumor growth.39
Moreover, S1P regulates angiogenesis during tumor
progression in an autocrine or paracrine manner.40,41
Thus, it is considered that SphK regulates tumor growth
and can act as an oncogene. Anti-cancer therapeutics
targeting SphK are very plausible. Indeed, in many stud-
ies the SphK inhibitors, N,N-dimethylsphingosine
(DMS) and D,L-threo-dihydrosphingosine (Safingol)
have been shown to induce growth arrest or cell death,
although whether this occurs through the inhibition of
SphK or PKC has not been established.42–45 In prelimin-
ary data from a phase I trial, Safingol, when combined
with doxorubicin and fenretinide, appeared to increase
the anti-cancer efficacy.46–48 As the cytotoxicity of
SG12 and SG14 were comparable to DMS, examining
their anti-cancer efficacy would be a reasonable next
step in our investigations.
extracts were dried over a drying agent, MgSO2, and
Na2SO4, and concentrated under reduced pressure with
the aid of a rotary evaporator. Analytical TLC (thin
layer chromatography) was performed with 0.2 mm
coated commercial silica gel, and the detection was car-
ried out using a UV lamp and PMA solution, p-anisal-
dehyde solution, or ninhydrine solution.
5.2. Procedure (SG-12)
5.2.1. (S)-N-Methoxy-N-methyl-1-((R)-1-phenylethyl)-
aziridine-2-carboxamide.2 To a flask charged with
100 mg of aziridine menthol ester (1) in THF (0.3 M)
and 45 mg of Weinrebꢀs amine was slowly added
0.46 mL of isopropyl magnesium chloride (2.0 M solu-
tion in THF), at 0 ꢁC under N2 atmosphere. The result-
ing mixture was warmed to room temperature for
10 min. After disappearance of the starting aziridine
menthol ester (as monitored by TLC), the reaction mix-
ture was partitioned into CH2Cl2 and H2O. The organic
layer was dried over MgSO4, concentrated in vacuo, and
purified by column chromatography with 50% EtOAc in
hexane to give 67 mg of product in a 94% yield.
In a recent study, SphK1 null mice were found to be
viable, fertile and without obvious abnormalities.49 In
that paper, the total SphK activity was reduced in most
tissue, but S1P levels were not markedly decreased
except in serum. These results indicate the presence
of multiple SphKs or the ability of SphK2 to consider-
ably compensate for the loss of SphK1 activity. Addi-
tionally, the function of the prodrug FTY-720
regulating lymphocyte trafficking and distribution was
not affected in the SphK1 -/- mice. This may suggest
the possibility that SphK2 can affect the functional
activation of the prodrug FTY-720. In this regard,
SG14 inhibited the phosphorylation of FTY-720 by
hSphK2 (Fig. 6B). Therefore, SG14 might be a potent
tool for regulating the functional activation of
FTY720. Additionally, SG12 was apparently phos-
phorylated by SphK2. This provides an interesting
basis for further investigation of possible therapeutic
properties of this compound.
5.2.2. 2-(4-Octylphenyl)-1-((S)-1-((R)-1-phenylethyl)azir-
idin-2-yl)ethanone. The solution of (S)-N-methoxy-N-
methyl-1-((R)-1-phenylethyl)aziridine-2-carboxamide (450
mg, 1.92 mmol) in THF (0.3 M) was added to 180 mg
of Mg, while mixing, and 660 mg of 1-(bromomethyl)-
4-octylbenzene at room temperature under N2 atmo-
sphere. The resulting mixture was heated to reflux at
50 ꢁC for 2–6 h. After completion, the reaction mixture
was extracted with EtOAc and H2O, dried over MgSO4,
and purified by column chromatography to give 580 mg
of product in 80% yield.
5.2.3. (R)-2-(4-Octylphenyl)-1-((S)-1-((R)-1-phenylethyl)-
aziridin-2-yl)ethanol. To a stirred solution of 2-(4-octyl-
phenyl)-1-((S)-1-((R)-1-phenylethyl)aziridin-2-yl)ethanone
(1 g, 2.64 mmol) in THF (0.3 M) was slowly added
120 mg of LAH at 0 ꢁC for 1 h. The mixture was
quenched by saturated aqueous KHSO4 and extracted
with EtOAc and H2O. The organic layer was concen-
trated and dried over MgSO4 and purified by
chromatography with 20% EtOAc in hexanes to give
(S)-2-(4-octylphenyl)-1-((S)-1-((R)-1-phenylethyl)aziri-
din-2-yl)ethanol (Rf: 0.5, 20% EtOAc in hexanes) in a
35–50% yield and (R)-2-(4-octylphenyl)-1-((S)-1-((R)-1-
phenylethyl)aziridin-2-yl)ethanol (Rf: 0.3, 20% EtOAc
in hexanes) in a 50–65% yield.
Many studies have established a relationship between
SphK1 and pathological conditions like cancer and
inflammation. However, until now, the participation of
SphK2 has not been well understood, despite significant
research recently to identify its function. A SphK2-spe-
cific inhibitor will be useful, then, not only in under-
standing the function of SphK2 but also in the design
of novel therapeutics.
5.2.4. (2S,3R)-2-((R)-1-Phenylethylamino)-3-hydroxy-4-
(4-octylphenyl)butyl acetate. A round flask was charged
with (R)-2-(4-octylphenyl)-1-((S)-1-((R)-1-phenylethyl)-
aziridin-2-yl)ethanol in CH2Cl2 (0.5 M). To the solution
was added AcOH (10 equiv of the substrate) at room
temperature, and the mixture was stirred for 12–18 h.
The reaction mixture was quenched with aqueous NaH-
CO3 and extracted with CH2Cl2. The organic solution
was dried over MgSO4 and purified by column chroma-
tography to give the ring opening product in a 90%
yield.
5. Experimental
5.1. Synthesis of sphingoid analogs
1H NMR spectra were recorded at 300 MHz and 13C
NMR spectra at 75 MHz, with either TMS (d = 0) or
the signal for residual CHCl3 in the CDCl3 solvent
(d = 7.27) as internal standards. J values are reported
in Hertz. Flash column chromatography was performed
with Kieselgel 60 Art 9385 (230–400 mesh). Organic

3482
J.-W. Kim et al. / Bioorg. Med. Chem. 13 (2005) 3475–3485
5.2.5. (2S,3R)-2-((R)-1-Phenylethylamino)-4-(4-octylphen-
yl)butane-1,3-diol. A flask was charged with 100 mg
of (2S,3R)-2-((R)-1-phenylethylamino)-3-hydroxy-4-(4-
octylphenyl)butyl acetate in MeOH (0.5 M). To the
solution was added 26 mg of KOH at room tempera-
ture. After 2 h, the reaction mixture was extracted with
CH2Cl2, dried over MgSO4, and concentrated to give
free alcohol in a quantitative yield.
1H), d 2.96–2.76 (m, 2H), d 2.2 (t, J = 7.8 Hz, 2H), d
1.7–1.5 (m, 2H), d 1.35–1.2 (m, 26H), d 0.88 (t,
J = 6.9 Hz, 3H); 13C NMR (75 MHz, CDCl3) d 173.8,
137.7, 129.2, 128.7, 126.7, 74.6, 62.2, 53.8, 40.8, 36.8,
34.0, 31.9, 29.7, 29.6, 29.5, 29.4, 29.3, 29.2, 25.7, 24.8,
22.7, 14.1; Anal. Calcd for C26H45NO3: C, 74.42; H,
10.81; N, 3.34. Found: C, 74.57; H, 10.76; N, 3.23.
5.3.5.
N-((R)-1-Hydroxy-4-phenylbutan-2-yl)tridecan-
1
5.2.6. (2S,3R)-2-Amino-4-(4-octylphenyl)butane-1,3-diol.
A mixture of (2S,3R)-2-((R)-1-phenylethylamino)-4-(4-
octylphenyl)butane-1,3-diol and Pd(OH)2 (20 wt %) in
EtOH (0.5 M) was stirred at room temperature under
H2 (100 psi). The reaction mixture was monitored by
TLC through the completion of the reaction. The mix-
ture was filtered using Celite. After evaporation of the
solvent, the crude product was purified by column chro-
matography to give the debenzylation product in an
80% yield.
amide (SG-5). H NMR (300 MHz, CDCl3) d 7.3–7.15
(m, 5H), d 5.65 (d, J = 7.8 Hz, 1H), d 4.05–3.95 (m,
1H), d 3.75–3.55 (m, 2H), d 2.83 (br s, 1H), d 2.68 (t,
J = 7.8 Hz, 2H), d 2.15 (t, J = 7.8 Hz, 2H), d 1.95–1.7
(m, 2H), d 1.65–1.55 (m, 2H), d 1.25 (br s, 20H), d
0.88 (t, J = 6.3 Hz, 3H); 13C NMR (75 MHz, CDCl3) d
174.3, 141.5, 128.6, 128.3, 126.3, 65.6, 51.5, 36.9, 32.8,
32.5, 31.9, 29.7, 29.6, 29.4, 29.2, 29.3, 25.7, 22.6, 14.3;
Anal. Calcd for C23H39NO2: C, 76.40; H, 10.87; N,
3.87. Found: C, 76.53; H, 10.95; N, 3.79.
5.3. Compound characterization data
5.3.6. N-((R)-1-Hydroxy-4-phenylbutan-2-yl)palmitamide
(SG-6). 1H NMR (300 MHz, CDCl3) d 7.3–7.1 (m, 5H),
d 5.7 (d, J = 7.8 Hz, 1H), d 4.05–3.92 (m, 1H), d 3.68
(dd, J = 11.1, 3.6 Hz, 1H), d 3.59 (dd, J = 11.1, 5.4 Hz,
1H), d 2.94 (br s, 1H), d 2.68 (t, J = 7.8 Hz, 2H), d
2.15 (t, J = 7.8 Hz, 2H), d 1.96–1.74 (m, 2H), d 1.65–
1.45 (m, 2H), d 1.25 (br s, 26H), d 0.88 (t, J = 6.6 Hz,
3H); 13C NMR (75 MHz, CDCl3) d 174.1, 141.4,
128.5, 128.3, 126.1, 65.7, 51.6, 36.9, 32.8, 32.5, 31.9,
29.7, 29.6, 29.5, 29.3, 29.3, 25.8, 22.7, 14.1; Anal. Calcd
for C26H45NO2: C, 77.37; H, 11.24; N, 3.47. Found: C,
77.42; H, 11.38; N, 3.39.
5.3.1. N-((2S,3S)-1,3-Dihydroxy-4-phenylbutan-2-yl)tri-
decanamide (SG-1). 1H NMR (300 MHz, CDCl3) d
7.13–7.22 (m, 5H), d 6.42 (d, J = 8.7 Hz, 1H), d 4.12
(td, J = 6.3, 1.5 Hz, 1H), d 3.95–3.85 (m, 1H), d 3.65
(d, J = 4.8, 2H), d 2.74 (d, J = 1.5 Hz, 1H), d 2.71 (d,
J = 3.6 Hz, 1H), d 1.69–1.58 (m, 2H), d 1.38–1.20 (m,
20H), d 0.87 (t, J = 6.9 Hz, 3H); 13C NMR (75 MHz,
CDCl3) d 174.3, 137.7, 129.3, 128.3, 126.6, 72.7, 64.2,
52.8, 40.7, 36.8, 31.854, 29.6, 29.5, 29.3, 29.2, 25.9,
22.6, 14.0; Anal. Calcd for C23H39NO3: C, 73.17; H,
10.41; N, 3.71. Found: C, 73.32; H, 10.32; N, 3.69.
5.3.7. (R)-2-Amino-4-(4-octylphenyl)butan-1-ol (SG-7).
1H NMR (300 MHz, CDCl3) d 7.1–6.95 (m, 4H), d
3.59 (dd, J = 10.5, 3.6 Hz, 1H), d 3.32 (dd, J = 10.5,
7.8 Hz, 1H), d 2.9–2.8 (m, 1H), d 2.7–2.6 (m, 2H), d
2.58 (t, J = 7.8 Hz, 2H), d 1.8–1.65 (m, 1H), d 1.6–1.5
(M, 2H), d 1.38–1.15 (m, 12H), d 0.88 (t, J = 6.0 Hz,
3H); 13C NMR (75 MHz, CDCl3) d 140.5, 138.7,
128.4, 128.1, 66.4, 52.4, 35.9, 35.5, 31.9, 31.9, 31.5,
29.4, 29.3, 29.2, 22.6, 14.0; Anal. Calcd for C18H31NO:
C, 77.92; H, 11.26; N, 5.05. Found: C, 77.84; H, 11.31;
N, 5.12.
5.3.2. N-((2S,3R)-1,3-Dihydroxy-4-phenylbutan-2-yl)tri-
decanamide (SG-2). 1H NMR (300 MHz, CDCl3) d
7.2–7.36 (m, 5H), d 6.38 (d, J = 7.2 Hz, 1H), d 3.9–
4.15 (m, 2H), d 3.76 (dd, J = 11.1, 2.7 Hz, 1H), d 2.98–
2.75 (m, 2H), d 2.21 (t, J = 7.2 Hz, 2H), d 1.9–1.54 (m,
2H), d 1.4–1.25 (m, 20H), d 0.88 (t, J = 5.7 Hz, 3H);
13C NMR (75 MHz, CDCl3) d 173.7, 137.6, 129.0,
128.5, 126.7, 74.5, 62.2, 53.8, 40.8, 36.8, 34.0, 31.9,
29.7, 29.5, 29.4, 29.4, 29.3, 29.2, 25.6, 24.8, 22.6, 14.0;
Anal. Calcd for C23H39NO3: C, 73.17; H, 10.41; N,
3.71. Found: C, 73.02; H, 10.42; N, 3.65.
5.3.8. (2S,3S)-2-Amino-4-(4-octylphenyl)butane-1,3-diol
(SG-9). 1H NMR (300 MHz, CD3OD) d 7.16 (d,
J = 7.8 Hz, 2H), d 7.07 (d, J = 7.8 Hz, 2H), d 3.85–3.75
(m, 1H), d 3.69 (dd, J = 10.8, 5.4 Hz, 1H), d 3.5 (dd,
J = 10.8, 6.6 Hz, 1H), d 2.84 (dd, J = 13.8, 5.4 Hz,
1H), d 2.75–2.62 (m, 2H), d 2.55 (dd, J = 7.8, 7.2 Hz,
2H), d 1.64–1.5 (m, 2H), d 1.28 (br s, 12H), d 0.89 (t,
J = 6.3 Hz, 3H); 13C NMR (75 MHz, CDCl3) d 140.1,
138.2, 128.6, 128.5, 52.1, 35.4, 35.2, 31.6, 31.1, 31.0,
29.7, 29.6, 29.5, 22.6, 14.0; Anal. Calcd for
C18H31NO2: C, 73.67; H, 10.65; N, 4.77. Found: C,
73.68; H, 10.71; N, 4.67.
5.3.3. N-((2S,3S)-1,3-Dihydroxy-4-phenylbutan-2-yl)pal-
mitamide (SG-3). 1H NMR (300 MHz, CDCl3) d
7.34–7.15 (m, 5H), d 6.44 (d, J = 8.7 Hz, 1H), d 4.23
(br s, 1H), d 4.14 (t, J = 6.3 Hz, 1H), d 4.0–3.9 (m,
2H), d 3.69 (d, J = 4.2 Hz, 2H), d 2.8–2.68 (m, 2H), d
2.35–2.2 (m, 2H), d 1.7–1.5 (m, 2H), d 1.4–1.24 (m,
26H), d 0.88 (t, J = 6.3 Hz, 3H); 13C NMR (75 MHz,
CDCl3) d 174.4, 137.6, 129.3, 128.6, 126.6, 73.0, 64.4,
52.8, 40.7, 36.8, 34.0, 31.9, 29.7, 29.6, 29.5, 29.4, 29.3,
29.2, 25.9, 24.8, 22.6, 14.1; Anal. Calcd for
C26H45NO3: C, 74.42; H, 10.81; N, 3.34. Found: C,
74.52; H, 10.92; N, 3.29.
5.3.9. (2R,3S)-3-Amino-4-morpholino-1-phenylbutan-2-ol
1
5.3.4. N-((2S,3R)-1,3-Dihydroxy-4-phenylbutan-2-yl)pal-
mitamide (SG-4). H NMR (300 MHz, CDCl3) d 7.35–
7.2 (m, 5H), d 6.47 (d, J = 7.8 Hz, 1H), d 4.1–3.98 (m,
2H), d 3.96–3.9 (m, 1H), d 3.75 (dd, J = 11.4, 3.0 Hz,
(SG-10). H NMR (300 MHz, CDCl3) d 7.35–7.2 (m,
1
5H), d 3.8–3.6 (m, 5H), d 2.96 (dd, J = 13.2, 2.4 Hz,
1H), d 2.92 (dd, J = 13.2, 6.6 Hz, 1H), d 2.68 (dd,
J = 13.5, 8.7 Hz, 1H), d 2.6–2.4 (m, 6H); 13C NMR

J.-W. Kim et al. / Bioorg. Med. Chem. 13 (2005) 3475–3485
3483
(75 MHz, CDCl3) d 138.9, 129.2, 128.3, 126.2, 77.8,
66.7, 63.9, 54.0, 49.9, 40.1; Anal. Calcd for
C32H56N2O2: C, 67.17; H, 8.86; N, 11.19. Found: C,
67.22; H, 8.92; N, 11.09.
173.2, 137.3, 129.2, 128.3, 126.4, 74.7, 54.8, 54.2, 50.1,
37.1, 36.8, 34.3, 31.9, 29.7, 29.6, 29.5, 29.4, 29.3, 29.2,
29.1, 29.0, 25.7, 24.8, 23.5, 22.6, 14.1; Anal. Calcd for
C32H56N2O2: C, 76.75; H, 11.27; N, 5.59. Found: C,
76.79; H, 11.32; N, 5.45.
5.3.10. N-((R)-4-Phenyl-1-(pyrrolidin-1-yl)butan-2-yl)deca-
1
namide (SG-11). H NMR (300 MHz, CDCl3) d 7.4–7.1
5.3.15. N-((2S,3S)-3-Hydroxy-1-morpholino-4-phenylbu-
tan-2-yl)decanamide (SG-16). 1H NMR (300 MHz,
CDCl3) d 7.35–7.15 (m, 5H), d 6.17 (d, J = 8.4 Hz,
1H), d 5.09 (br s, 1H), d 4.1–3.9 (m, 2H), d 3.65 (t,
J = 4.5 Hz, 4H), d 2.76 (d, J = 2.7 Hz, 1H), d 2.74 (br
s, 1H), d 2.69 (d, J = 5.4 Hz, 1H), d 2.6–2.4 (m, 5H), d
2.22 (t, J = 6.9 Hz, 2H), d 1.7–1.6 (m, 2H), d 1.4–1.2
(m, 14H), d 0.88 (t, J = 6.3 Hz, 3H); 13C NMR
(75 MHz, CDCl3) d 173.2, 137.8, 129.2, 128.5, 126.5,
73.5, 66.8, 60.8, 54.6, 49.3, 40.5, 36.8, 31.8, 29.4, 29.3,
29.2, 29.1, 25.8, 22.6, 14.0; Anal. Calcd for
C24H40N2O3: C, 71.25; H, 9.97; N, 6.92. Found: C,
71.33; H, 10.04; N, 6.83.
(m, 5H), d 5.7 (d, J = 7.5 Hz, 1H), d 4.1–4.0 (m, 1H), d
2.7–2.55 (m, 2H), d 2.54–2.4 (m, 6H), d 2.16 (t,
J = 7.2 Hz, 2H), d 2.0–1.85 (m, 1H), d 1.8–1.7 (m, 4H),
d 1.68–1.55 (m, 2H), d 1.38–1.25 (m, 14H), d 0.87 (t,
J = 6.3 Hz, 3H); 13C NMR (75 MHz, CDCl3) d 173.1,
142.0, 128.3, 125.7, 59.5, 54.5, 48.4, 36.9, 35.5, 32.2,
31.8, 29.4, 29.3, 29.2, 25.8, 23.6, 22.6, 14.0; Anal. Calcd
for C24H40N2O: C, 77.37; H, 10.82; N, 7.52. Found: C,
77.45; H, 10.94; N, 7.64.
5.3.11. (2S,3R)-2-Amino-4-(4-octylphenyl)butane-1,3-diol
(SG-12). H NMR (300 MHz, CDCl3) d 7.16 (d, J =
1
7.8 Hz, 2H), d 7.08 (d, J = 7.8 Hz, 2H), d 3.85–3.7 (m,
2H), d 3.54 (dd, J = 10.8, 7.8 Hz, 1H), d 1.64–1.5 (m,
2H), d 1.28 (br s, 12H), d 0.89 (t, J = 6.0 Hz, 3H);
13C NMR (75 MHz, CDCl3) d 139.8, 137.9, 128.3,
128.2, 51.8, 35.1, 34.9, 31.3, 29.9, 29.7, 29.4, 29.3, 29.2,
22.3, 13.9; Anal. Calcd for C18H31NO2: C, 73.67; H,
10.65; N, 4.77. Found: C, 73.75; H, 10.72; N, 4.69.
5.4. DNA construction
HA-tagged hSphK1 and HA-tagged hSphK2 were gen-
erous gifts from Dr. Wada (this laboratory). All nucleo-
tides were synthesized with BamHI and EcoRI ends.
The antisense primers were treated with T4 polynucleo-
tide kinase (Takara, Ohtsu, Japan). The PCR was car-
ried out with KOD plus (Toyobo, Osaka, Japan) in
the presence of 0.2 mM dNTPs, 1 mM MgCl2 and
0.3 lM polynucleotide kinase-treated primers, for 25
cycles. The PCR products were ligated using Ligation
High (Toyobo, Osaka, Japan). After a BamHI-EcoRI
digestion, the fragment was ligated into a BamHI-
EcoRI-digested pcDNA3-HA vector.
5.3.12. N-((2S,3S)-3-Hydroxy-4-phenyl-1-(pyrrolidin-1-
yl)butan-2-yl)decanamide (SG-13). H NMR (300 MHz,
1
CDCl3) d 7.3–7.16 (m, 5H), d 6.38 (d, J = 9.0 Hz, 1H),
d 5.3–5.2 (m, 1H), d 4.44–4.32 (m, 1H), d 2.94–2.8 (m,
2H), d 2.72 (dd, J = 12.0, 9.0 Hz, 1H), d 2.6–2.4 (m,
5H), d 2.3–2.2 (m, 4H), d 1.8–1.6 (m, 6H), d 1.58–1.42
(t, J = 6.9 Hz, 2H), d 1.36–1.2 (m, 14H), d 0.88
(t, J = 6.6 Hz, 3H); 13C NMR (75 MHz, CDCl3) d
173.2, 136.8, 129.3, 128.1, 126.3, 74.3, 56.3, 53.6, 48.9,
37.7, 36.6, 34.1, 31.7, 31.7, 29.5, 29.4, 29.3,0 29.2, 29.1,
29.0, 28.9, 25.6, 24.7, 23.2, 22.5, 13.9; Anal. Calcd for
C24H40N2O2: C, 74.18; H, 10.38; N, 7.21. Found: C,
74.32; H, 10.42; N, 7.19.
5.5. Cell culture and DNA transfection
Chinese hamster ovary (CHO-K1) cells were grown in
Hamꢀs F-12 medium (Sigma, St. Louis, MO)
supplemented with 10% (v/v) fetal bovine serum
(Iwaki, Japan), 100 U/mL penicillin, and 100 lg/mL
streptomycin (Sigma) at 37 ꢁC in a humidified atmo-
sphere of 5% carbon dioxide. Cells were transfected with
the N-terminal hamagglutinin(HA)-tagged plasmids
pcDNA3-HA-hSphK1 and pcDNA3-HA-hSphK2
using LIPOFECTAMINE-PLUS (Invitrogen) accord-
ing to the manufacturerꢀs instructions. Plasmids used
in the transfection were prepared with a Quantum Prep
Plasmid Miniprep Kit (BIO-RAD).
5.3.13. N-((2S,3S)-3-Hydroxy-4-phenyl-1-(pyrrolidin-1-
yl)butan-2-yl)stearamide (SG14). H NMR (300 MHz,
1
CDCl3) d 7.4–7.2 (m, 5H), d 6.16 (d, J = 8.1 Hz, 1H),
d 4.18 (t, J = 6.3 Hz, 1H), d 4.0–3.9 (m, 1H), d
2.87 (dd, J = 7.2, 3.6 Hz, 2H), d 2.85–2.75 (m, 2H), d
2.74–2.68 (m, 2H), d 2.62–2.54 (m, 2H), d 2.27 (dd,
J = 7.8, 6.9 Hz, 2H), d 1.8–1.65 (m, 6H), d 1.4–1.26
(m, 30H), d 0.91 (t, J = 6.3 Hz, 3H); 13C NMR
(75 MHz, CDCl3) d 173.0, 138.0, 129.3, 128.5, 126.4,
75.2, 59.4, 55.9, 50.4, 40.4, 36.9, 31.9, 29.7, 29.6, 29.5,
29.3, 25.8, 23.7, 22.6, 14.1; Anal. Calcd for
C32H56N2O2: C, 76.75; H, 11.27; N, 5.59. Found: C,
76.81; H, 11.32; N, 5.48.
5.6. Immunoblotting
Proteins were separated by SDS-PAGE and transferred
to a PVDF membrane (Millipore, Bedford, MA). After
blocking with 5% skimmed milk, the membrane was
incubated for 1 h with a 1:1000 dilution of HA-Y11anti-
bodies (Santa Cruz, CA). The blot was washed with
TBS-T (137 mM NaCl, 20 mM Tris–HCl (pH 7.5), and
0.05% Tween 20) and then incubated with a 1:10,000
dilution of anti-rabbit-IgG-HRT (Amersham Pharma-
cia Biotech.). The blot was washed again with TBS-T
and detection was performed with an Enhanced
Chemiluminescence Detection Kit (ECL, Amersham
5.3.14. N-((2S,3R)-3-Hydroxy-4-phenyl-1-(pyrrolidin-1-
yl)butan-2-yl)stearamide (SG-15). H NMR (300 MHz,
1
CDCl3) d 7.3–7.1 (m, 5H), d 5.84 (d, J = 7.5 Hz, 1H),
d 5.35–5.27 (m, 1H), d 4.3–4.2 (m, 1H), d 2.95–2.85
(m, 2H), d 2.71 (dd, J = 12.3, 9.3 Hz, 1H), d 2.58–2.42
(m, 4H), d 2.25–2.8 (m, 4H), d 1.74 (br s, 4H), d 1.65–
1.55 (m, 2H), d 1.5–1.4 (m, 2H), d 1.25 (br s, 30H), d
0,88 (t, J = 6.0 Hz, 3H); 13C NMR (75 MHz, CDCl3) d

3484
J.-W. Kim et al. / Bioorg. Med. Chem. 13 (2005) 3475–3485
Pharmacia Biotech.). An anti-tubulin (Sigma) blot was
performed as a loading control.
32P-labeled substrate was recovered from the reaction
mix with a biotin capture membrane, and placed in scin-
tillation vials.
5.7. In vitro sphingosine kinase assay
5.10. Platelet preparation
Protein (1 lg) from each transfected cell lysate was incu-
bated with 2 lCi[c-32P]ATP (10 mCi/lmol, Perkin–Elmer
Life Science, Boston, MA), 0.5 mM cold ATP, and
40 lM substrate (D-erythro-sphingosine (Sigma), phyto-
sphingosine (BIOMOL, Plymouth, PA), or FTY-720
(generous gift from Sankyo, Co., Ltd., Japan)) at
37 ꢁC for 30 min in assay buffer (20 mM Tris–HCl
(pH 7.5), 0.25 mM EDTA, 1 mM magnesium chloride,
12 mM b-glycerophosphate, 1 mM sodium pyrophos-
phate, 5 mM sodium fluoride, 5 mM sodium orthovana-
date, 2 mM dithiothreitol, and 1 · protease inhibitor
cocktail (Roche)). For SphK2 assay, 200 mM KCl was
added in assay buffer. Lipids were extracted by adding
200 lL chloroform/methanol/HCl (100:200:1) and vor-
texing for 1 min. Consequently, 65 lL of chloroform
and 65 lL of 1 M KCl were added and the mixture
was vortexed vigorously for 5 min. The mixtures were
centrifuged for 2 min at 5000 · g and the upper phase
was removed. The samples were evaporated and resus-
pended in chloroform/methanol (2:1). Lipids were sepa-
rated by TLC with 1-butanol/acetic acid/water (3:1:1). A
sphingosine 1-phosphate band was visualized and quan-
tified using a Fujix Bio-Imaging Analyzer, BAS2500
(Fuji Photo Film, Japan).
Preparation of washed platelets was performed as previ-
ously described.50 The platelets were finally resuspended
in a pH 7.4 buffer (138 mM NaCl, 3.3 mM NaH2PO4,
2.9 mM KCl, 1 mM MgSO4, 1 mg/m glucose, and
20 mM Hepes), and 1% bovine serum albumin (fatty
acid free) was added. All experiments using intact plate-
let suspensions were performed at 37 ꢁC.
5.11. Metabolism of [3H] sphingosine in platelets
100 lL (3 · 107 platelets/100 lL) platelets suspensions
were pre-incubated with inhibitors for 5 min, then
0.2 lCi[3-3H] sphingosine (Perkin–Elmer Life Science,
Boston, MA) were added and the mixture was incubated
for 15 min. The reaction was terminated by adding
400 lL of chloroform/methanol/HCl (100:200:1), and
130 lL of chloroform and 130 lL of 1 M KCl were
added, and the mixture was vortexed vigorously for
5 min. The mixtures were centrifuged for 2 min at
5000 · g and the upper phase was removed. The samples
were evaporated and resuspended in chloroform/metha-
nol (2:1). Lipids were separated by TLC with 1-butanol/
acetic acid/water (3:1:1). The sphingosine 1-phosphate
band was visualized by autoradiography.
5.8. Cytotoxicity assay
The cytotoxic effect was measured using 2-(2-Methoxy-
4-nitophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-
tetrasolium (Cell counting Kit-8, Dojindo, Japan).
CHO-K1 cells (5 · 103 cell/100 lL) were plated in 96-
well dishes and incubated at 37 ꢁC for 24 h in a humidi-
fied atmosphere of 5% CO2. Then, cells were incubated
with 10 lL of Sphingoid analogs for an additional 24–
48 h. Next, 10 lL of Cell counting kit-8 were added in
each well, and the samples were incubated for another
4 h. The absorbance at a wavelength of 450 nm was
measured to detect mitochondrial dehydrogenase enzy-
matic activity.
Acknowledgements
We thank Dr. M. Tani (this laboratory) and Dr. K. Tsu-
ji (this laboratory) for helping in the preparation of
platelets and for useful advice.
References and notes
1. Hannun, Y. M.; Obeid, L. M. Biochem. Soc. Trans. 1997,
25, 1171–1175.
2. Maceyka, M.; Payne, S. G.; Milstien, S.; Spiegel, S.
Biochem. Biophys. Acta 2002, 1585, 193–201.
3. Hla, T.; Lee, M. J.; Ancellin, N.; Paik, J. H.; Kluk, M. J.
Science 2001, 294, 1875–1878.
5.9. In vitro protein kinase C assay
4. Yatomi, Y.; Ozaki, Y.; Ohmori, T.; Igarashi, Y. Prosta-
glandins Other Lipid Mediat. 2001, 64, 107–122.
5. Takuwa, Y. Biochim. Biophys. Acta 2002, 23, 112–120.
6. Igarashi, Y. J. Biochem. 1997, 122, 1080–1087.
7. Murate, T.; Banno, Y.; Koizumi, K. T.; Watanabe, K.;
Mori, N.; Wada, A.; Igarashi, Y.; Takagi, A.; Kojima, T.;
Asano, H.; Akao, Y.; Yoshida, S.; Saito, H.; Nozawa, Y.
J. Histochem. Cytochem. 2001, 49, 845–855.
8. Fukuda, Y.; Kihara, A.; Igarashi, Y. Biochem. Biophys.
Res. Commun. 2003, 309, 155–160.
9. Olivera, A.; Spiegel, S. Nature 1993, 365, 557–560.
10. Edsall, L. C.; Pirianov, G. G.; Spiegel, S. J. Neurosci.
1997, 17, 6952–6960.
TM
PKC activity was measured by using a SignaTech PKC
assay system (Promega, Medison, WI) according to the
manufacturerꢀs instructions. This assay depends on the
transfer of 32P from [c-32P]ATP to the biotinylated pep-
tide, Neurogranin (AAKIQAS*FRGHMARKK), the
most specific substrate commercially available for
PKC activity. To monitor the influence of various syn-
thetic compounds on PKC activity, the indicated quan-
tities of the SG samples were supplemented into a PKC
assay buffer mixture [1.6 mg/mL phosphatidylserine,
0.16 mg/mL diacylglycerol, 100 mM Tris–HCL (pH
7.5), 50 mM MgCl2, 1.25 mM EGTA, 2 mM CaCl2,
0.5 mg/mL BSA and [c-32P]ATP] to a final volume of
20 mL. The reaction was initiated by adding 5 mL of
PKC (5 ng PKC enzyme, purified from rat brain) then
was incubated for 5 min at 30 ꢁC. The biotinylated,
11. Meyer zu Heringdorf, D.; Lass, H.; Kuchar, I.; Alemany,
R.; Guo, Y.; Schmidt, M.; Jakobs, K. H. FEBS Lett. 1999,
461, 217–222.
12. Shu, X.; Wu, W.; Mosteller, R. D.; Broek, D. Mol. Cell.
Biol. 2002, 22, 7758–7768.

J.-W. Kim et al. / Bioorg. Med. Chem. 13 (2005) 3475–3485
3485
13. Kohama, T.; Olivera, A.; Edsall, L.; Nagiec, M. M.;
Dickson, R.; Spiegel, S. J. Biol. Chem. 1998, 273, 23722–
23728.
32. Lee, K. D.; Suh, J. M.; Park, J. H.; Ha, H. J.; Choi, H. G.;
Park, C. S.; Chang, J. W.; Lee, W. K.; Dong, Y. K.; Yun,
H. S. Tetrahedron 2001, 57, 8267.
14. Liu, H.; Sugiura, M.; Nava, V. E.; Edsall, L. C.; Kono, K.;
Poulton, S.; Milstien, S.; Kohama, T.; Spiegel, S. J. Biol.
Chem. 2000, 275, 19513–19520.
15. Olivera, A.; Kohama, T.; Edsall, L.; Nava, V.; Cuvillier,
O.; Poulton, S.; Spiegel, S. J. Biol. Chem. 1999, 147, 545–
558.
33. Pyun, D. K.; Lee, C. H.; Ha, H. J.; Park, C. S.; Chang,
J. W.; Lee, W. K. Org. Lett. 2001, 3, 4197.
34. Sugiyama, S.; Morishita, K.; Ishii, K. Heterocycles 2001,
55, 353.
35. Sibi, M. P.; Renhowe, P. A. Tetrahedron Lett. 1990, 35,
7407.
16. Igarashi, N.; Okada, T.; Hayashi, S.; Fujita, T.; Jahan-
geer, S.; Nakamura, S. J. Biol. Chem. 2003, 278, 46832–
46839.
17. Liu, H.; Toman, R. E.; Goparaju, S. K.; Maceyka, M.;
Nava, V. E.; Sankala, H.; Payne, S. G.; Bektas, M.; Ishii,
I.; Chun, J.; Milstien, S.; Spiegel, S. J. Biol. Chem. 2003,
278, 40330–40336.
18. Paugh, S. W.; Payne, S. G.; Barbour, S. E.; Milstien, S.;
Spiegel, S. FEBS Lett. 2003, 554, 189–193.
19. Billich, A.; Bornancin, F.; Devay, P.; Mechtcheriakova,
D.; Urtz, N.; Baumruker, T. J. Biol. Chem. 2003, 278,
47408–47415.
20. Koskinen, P. M.; Koskinen, A. M. Synthesis 1998, 1075.
21. Cuvillier, O.; Pirianov, G.; Kleuser, B.; Vanek, P. G.;
Coso, O. A.; Gutkind, J. S.; Spiegel, S. Nature 1996, 381,
800–803.
22. Choi, O. H.; Kim, J. H.; Kinet, J. P. Nature 1996, 380,
634–636.
23. Hannun, Y. A.; Bell, E. M. Science 1989, 243, 500–
507.
24. Lee, W. K.; Ha, H. J. Aldrichim. Acta 2003, 36, 57–
63.
25. Nahm, S.; Weinreb, S. M. Tetrahedron Lett. 1981, 22,
3815–3818.
36. Igarashi, Y.; Hakomori, S.; Toyokuni, T.; Dean, B.;
Fujita, S.; Sugimoto, M.; Ogawa, T.; El-Ghendy, K.;
Racker, E. Biochemistry 1989, 28, 6796–6800.
37. French, K. J.; Schrecengost, R. S.; Lee, B. D.; Zhuang, Y.;
Smith, S. N.; Eberly, J. L.; Yun, J. K.; Smith, C. D.
Cancer Res. 2003, 63, 5962–5969.
38. Xia, P.; Gamble, J. R.; Wang, L.; Pitson, S. M.; Moretti,
P. A. B.; Wattenberg, B. W.; DꢀAndrea, R. J.; Vadas, M.
A. Curr. Biol. 2000, 10, 1527–1530.
39. Endo, K.; Igarashi, Y.; Nisar, M.; Zhou, Q.; Hakomori, S.
Cancer Res. 1991, 51, 1613–1618.
40. Lee, M. J.; Thangada, S.; Claffey, K. P.; Ancellin, N.; Liu,
C. H.; Kluk, M.; Volpi, M.; Shaꢀafi, R. I.; Hla, T. Cell
1999, 99, 301–312.
41. Hobson, J. P.; Rosenfeldt, H. M.; Barak, L. S.; Olivera,
A.; Poulton, S.; Caron, M. G.; Milstien, S.; Spiegel, S.
Science 2000, 291, 1800–1803.
42. Nava, V. E.; Hobson, J. P.; Murthy, S.; Milstien, S.;
Spiegel, S. Exp. Cell. Res. 2002, 281, 115–127.
43. Ohta, H.; Sweeney, E. A.; Masamune, A.; Yatomi, Y.;
Hakomori, S.; Igarashi, Y. Cancer Res. 1995, 55, 691–697.
44. Sweeney, E. A.; Inokuchi, J.; Igarashi, Y. FEBS Lett.
1998, 425, 61–65.
45. Cuvillier, O.; Edsall, L.; Spiegel, S. J. Biol. Chem. 2000,
275, 15691–15700.
26. Basha, A.; Lipton, M.; Weinreb, S. M. Tetrahedron Lett.
1977, 4171–4174.
46. Sakakura, C.; Sweeney, E. A.; Shirahama, T.; Hakomori,
S.; Igarashi, Y. FEBS Lett. 1996, 379, 177–180.
47. Maurer, B. J.; Melton, L.; Billups, C.; Cabot, M. C.;
Reynolds, C. P. J. Natl. Cancer Inst. 2000, 92(23), 1897–
1909.
48. Wang, H.; Maurer, B. J.; Reynolds, C. P.; Cabot, M. C.
Cancer Res. 2001, 61(13), 5102–5105.
49. Allende, M. L.; Sasaki, T.; Kawai, H.; Olivera, A.; Mi, Y.;
Deckert, G. E.; Hajdu, R.; Rosenbach, M.; Keohane, C.
A.; Mandala, S.; Spiegel, S.; Proia, R. L. J. Biol. Chem.
2004, 279, 52478–52492.
27. Lee, E.; Song, H. Y.; Kang, J. W.; Kim, D. S.; Jung, C. K.;
Joo, J. M. J. Am. Chem. Soc. 2002, 124, 384–385.
28. Jonghe, S. D.; Lamote, I.; Venkataraman, K.; Boldin, S.
A.; Hillaert, U.; Rozenski, J.; Hendrix, C.; Busson, R.;
Keukeleire, D. D.; Calenbergh, S. V.; Futerman, A. H.;
Herdewijn, P. J. Org. Chem. 2002, 67, 988–996.
29. Park, H. W.; Song, J. Y.; Kim, K. S.; Han, Y.; Kim, C.
W.; Yi, S. Y.; Yun, Y. S. Exp. Mol. Med. 2004, 36(5), 411–
419.
30. Kim, B. C.; Lee, W. K. Tetrahedron 1996, 52, 12117.
31. Yun, J. M.; Sim, T. B.; Hahm, H. S.; Lee, W. K.; Ha, H. J.
J. Org. Chem. 2003, 68, 7675.
50. Yatomi, Y.; Ozaki, Y.; Satoh, K.; Kume, S. Biochim.
Biophys Acta 1994, 1212, 337–344.