Synthesis and biological properties of novel sphingosine derivatives

Article abstract of DOI:10.1016/j.bmcl.2004.12.010

Sphingosine-1-phosphate (S-1P) derivatives such as threo-(2S,3S)-analogues, which are C-3 stereoisomers of natural erythro-(2S,3R)-S-1P, have been synthesized starting from l-serine or (1S,2S)-2-amino-1-aryl-1,3-propanediols (6). threo-(1S,2R)-2-Amino-1-aryl-3-bromopropanols (HBr salt) have also been prepared from 6. The threo-S-1Ps and the threo-amino-bromide derivatives have shown potent inhibitory activity against Ca²⁺ ion mobilization in HL60 cells induced by erythro-S-1P, suggesting that these compounds would compete with cell surface EDG/S1P receptors.

Full text of DOI:10.1016/j.bmcl.2004.12.010

Bioorganic & Medicinal Chemistry Letters 15 (2005) 1115–1119
Synthesis and biological properties of novel sphingosine derivatives
Teiichi Murakami,a,* Kiyotaka Furusawa,a Tadakazu Tamai,b,*
Kazuyoshi Yoshikaib and Masazumi Nishikawab
aNational Institute of Advanced Industrial Science and Technology (AIST), Tsukuba Central 5, Tsukuba, Ibaraki 305-8565, Japan
bCentral Research Institute, Maruha Corporation, 16-2 Wadai, Tsukuba, Ibaraki 305-4295, Japan
Received 25 October 2004; revised 2 December 2004; accepted 6 December 2004
Available online 28 December 2004
Abstract—Sphingosine-1-phosphate (S-1P) derivatives such as threo-(2S,3S)-analogues, which are C-3 stereoisomers of natural
erythro-(2S,3R)-S-1P, have been synthesized starting from L-serine or (1S,2S)-2-amino-1-aryl-1,3-propanediols (6). threo-(1S,2R)-
2-Amino-1-aryl-3-bromopropanols (HBr salt) have also been prepared from 6. The threo-S-1Ps and the threo-amino-bromide deriv-
atives have shown potent inhibitory activity against Ca2+ ion mobilization in HL60cells induced by erythro-S-1P, suggesting that
these compounds would compete with cell surface EDG/S1P receptors.
ꢀ 2004 Elsevier Ltd. All rights reserved.
Sphingolipids, for example, sphingomyelin, cerebro-
sides, and gangliosides, are ubiquitous cell membrane
components and are involved in many essential biologi-
cal processes such as cell growth, cell differentiation, and
adhesion.1 Sphingolipid metabolites such as sphingosine
and ceramide are emerging as a novel class of lipid sec-
ond messengers.2 Recently sphingosine-1-phosphate (S-
1P), one of the metabolites, has attracted considerable
attention3 both as an intracellular second messenger4
and as an intercellular mediator. It has been reported
that S-1P binds to cell surface receptor EDG (endothe-
lial differentiation gene)/S1P family (five subtypes:
EDG-1/S1P1, EDG-3/S1P3, EDG-5/S1P2, EDG-6/S1P4,
and EDG-8/S1P5 have been identified3), which are cou-
pled via plasma membrane G-protein to multiple effec-
tor systems.5 Physiological significance of S-1P seems
very important in the vascular system because blood
platelets store S-1P abundantly and release this bioac-
tive lipid extracellularly upon stimulation to bind sur-
face receptors on vascular endothelial cells.6 The
receptors bound to S-1P would affect various biological
responses, including mitogenesis, differentiation, prolif-
eration, and apoptosis, and thus are supposed to be in-
volved in a variety of pathological conditions such as
angiogenesis, inflammation, and cardiovascular dis-
eases.7 S-1P is also suggested to be a central component
of a complex network of cytokines and chemokines,
which influence the responses of cells including immuno-
suppression.8 Therefore, the search for agonists and
antagonists toward EDG/S1P receptors would provide
the basis for development of novel therapeutic agents
for such diseases.9
Sphingoid bases are 2-amino-1,3-diols with a long-chain
alkyl tail at C3. The alkyl tail may vary in chain length
(from 16 to 24 carbon atoms), unsaturation (usually
C4,5-trans olefin), and hydroxylation. The most com-
mon sphingoid base in mammalian tissues is D-erythro-
C18-sphingosine [(2S,3R,4E)-2-aminooctadec-4-ene-1,3-
diol]. Recently, Parrill and co-workers proposed that
both the C1 phosphate group and C2 ammonium moi-
ety of D-erythro-S-1P are critical for its specific binding
to EDG-1/S1P1 receptor based on their homology mod-
eling and point mutation studies. However, the role of
the C3 hydroxy group remains to be solved.10 Chung
and co-workers reported the synthesis of four stereoi-
somers of S-1P and the analogues, and their binding
affinities to EDG-1, -3, and -5.11a Their results have sug-
gested that (1) the D-erythro configuration of S-1P is
important for a high affinity binding, (2) the phosphate
group of S-1P is essential for ligand recognition by the
receptors, (3) besides the C1 phosphate group and the
C2 ammonium moiety, the presence and configuration
of the C3 hydroxy group of S-1P appears to be very
important for specific binding.
Keywords: Sphingosine-1-phosphate; Amino alcohols; EDG/S1P
receptors; Antagonists.
*
Encouraged by these studies, we planned to investigate
agonist/antagosist activities of novel S-1P analogues
Corresponding authors. Tel.: +81 29 861 9394; fax: +81 29 861
4549; e-mail: t-murakami@aist.go.jp
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.12.010

1116
T. Murakami et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1115–1119
Table 1. Inhibition concentration (50%) of S-1P derivatives (threo-
amino alcohols) against Ca2+ ion increase in HL60induced by natural
S-1P
toward EDG/S1P receptors. Herein we report the syn-
thesis of several analogues and their effects on Ca2+
ion mobilization in HL60leukemia cells expressing these
receptors.12 In addition, their actions on the growth of
vascular smooth muscle cells and on an inflammation
model are reported.
NH2
X
R
OH
We recently reported13 a highly diastereoselective syn-
thesis of both D-erythro- and L-threo-sphingosines from
L-serinal derivative (GarnerÕs aldehyde14) with 1-alkenyl
nucleophiles prepared via hydrozirconation15 of termi-
nal alkynes. Thus, as shown in Scheme 1, N-Boc-sphin-
gosine derivatives erythro-3a–c and threo-3a–c (a:
natural length C18; b: shorter-chain homologue; c: sty-
ryl analogue) were prepared13 from GarnerÕs aldehyde
(1) in a stereocontrolled manner via the N,O-isopropyl-
idene acetal derivatives erythro-2a–c and threo-2a–c,
respectively. Selective phosphorylation of C1 alcohol
of 3 was achieved by a procedure of Bielawska and
co-workers16 with (MeO)3P and CBr4 in pyridine.9d,11
Treatment of the dimethyl phosphate 4 with trimethylsi-
lyl bromide (TMSBr) followed by addition of water
resulted in deprotection of both the Boc group and
the phosphate dimethyl ester to afford S-1P derivative
(erythro-5a–c and threo-5a–c). The spectral and physical
data of these S-1P derivatives were identical with
those reported11a,16,17 and/or consistent with assigned
structures.18
Compound
R
X
IC50 (lM)a
threo-5a
threo-5b
threo-5c
9d
(E)-n-Pentadec-1-enyl
(E)-n-Dodec-1-enyl
(E)-Styryl
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
Br
0.031
0.015
0.037
0.179
0.056
0.015
0.071
0.018
Phenyl
9e
p-Nitrophenyl
p-(Methylthio)phenyl
Phenyl
9f
11db
11fb
p-(Methylthio)phenyl
Br
a The values are the mean of duplicate experiments.
b HBr salt was used.
threo-5a–c inhibit the natural S-1P induced-Ca2+ ion in-
crease at rather low concentrations (IC50 = 0.015ꢀ
0.037 lM).20 Thus the threo-(2S,3S)-analogues might
be recognized as a ligand by EDG/S1P receptors to
show potent antagonist-like activities. Since subtype
specific receptors were not used in this preliminary as-
say, these observations may reflect total affairs concern-
ing with the receptors.
We then evaluated biological activities of the S-1P ana-
logues by measuring Ca2+ ion mobilization in HL60
cells.19 The bioassays have indicated that erythro-5a–c
show the Ca2+ ion increasing activity comparable to nat-
ural S-1P, whereas threo-5a–c do not show that activity
(data not shown). Interestingly, as shown in Table 1,
Since the threo-S-1Ps showed inhibitory effect, our
attention was turned to readily available threo-amino
alcohols. Commercially available threo-2-amino-1-aryl-
1,3-propanediols (6d–f; d: phenyl; e: p-nitrophenyl; f:
p-methylthiophenyl) are suitable for our purpose. Phos-
phorylation of the primary alcohol was carried out in a
AcOH-H2O
NHBoc
Boc
R
(9:1)
N
HO
HO
R
Cp2Zr
O
(R)
Cl
R
OH
50 °C, 5 h
ZnBr2 (50 mol%)
OH
Boc
(65-78% yield)
(85-90% de)
erythro-3a-c
N
erythro-2a-c
THF, rt, 24 h
O
(80-90%)
H
(S)
(a: R = n-C13H27, b: R = n-C10H21, c: R = Ph)
O
AcOH-H2O
Boc
NHBoc
R
1
R
(9:1)
N
Et Zn
O
(S)
R
OH
50 °C, 5 h
CH2Cl2, -30 to 0 °C
OH
threo-3a-c
(82-92% yield)
(85-88% de)
threo-2a-c
(80-90%)
(MeO)3P
(2 equiv.)
Me3SiBr
O
P
HO
NH2
O
P
MeO
NHBoc
OH
H2O
(5 equiv.)
3
HO
O
R
MeO
O
R
threo-3a-c
2
1
CH2Cl2
rt, 3 h
dioxane
CBr4 (2 equiv.)
pyridine
OH
threo-4a-c
(65-75%)
threo-5a-c
(40-70%)
5 °C to rt
erythro-5a-c were prepared similarly.
Scheme 1. Synthesis of S-1P derivatives.

T. Murakami et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1115–1119
1117
(MeO)3P
(2 equiv.)
Me3SiBr
(5 equiv.)
Boc
OH
Z
Z
Z
NHR
O
P
MeO
HN
O
P
HO
NH2
H2O
1
HO
MeO
O
HO
O
2
3
CH2Cl2
rt, 3 h
CBr4 (2 equiv.)
pyridine
dioxane
OH
OH
8d-f
9d-f
(45-70%)
5 °C to rt
6d-f (R = H)
(75-85%)
(d: Z = H; e: Z = NO2; f: Z = SMe)
Boc2O
THF
7d-f (R = Boc)
(93-97%)
Me3SiBr
HBr
NH2
NBS (1.5 equiv.)
Ph3P (1.5 equiv.)
Boc
HN
Boc
OH
Z
Z
Z
(5 equiv.)
HN
MeOH
1
Br
Br
HO
2
pyridine (1 equiv.)
CH2Cl2
rt, 3 h
3
OH
OH
THF
5 °C to rt
11d,f
(75-80%)
10d,f
(70-80%)
7d,f
Scheme 2. Synthesis of threo-amino alcohols.
similar manner as above. Thus, as shown in Scheme 2,
treatment of 6 with di-tert-butyl dicarbonate gave N-
Boc derivative 7, which was treated with (MeO)3P and
CBr4 in pyridine to give the phosphate ester of the pri-
mary alcohol 8. Deprotection with TMSBr afforded
the amino-phosphate 9d–f, which was purified by recrys-
tallization from aqueous THF or by reversed phase col-
umn chromatography.21 During the final step, a small
amount of the primary bromide was formed. The bro-
mide was soluble in water and polar organic solvents,
and easily purified. Thus the bromides were prepared
to evaluate their biological activities. Treatment of
Boc-amino diol 7 with N-bromosuccinimide and triphen-
ylphosphine gave the primary bromide 10 in good yield,
which was treated with TMSBr to give the threo-
(1S,2R)-amino-bromide HBr salt 11d,f. These products
were purified by silica gel chromatography eluting with
CH2Cl2–MeOH.22
Next we evaluated the actions of the S-1P analogues on
the growth of vascular smooth muscle cells (SMCs). It is
hypothesized that, with the progression of arteriosclero-
sis, vascular SMCs are transformed from the contractile
type to the synthetic type, and while secreting inflamma-
tory cytokines, the cells proliferate to cause arterioscle-
rotic lesions (RothÕs hypothesis). SMCs are reported to
proliferate in response to natural S-1P with acts on
EDG/S1P receptors expressed on the cell surface.23 Thus
the actions of threo-S-1P analogues 9f and 5b on the
growth of vascular SMCs were measured in the follow-
ing manner. Along with natural S-1P (1.0 lM) as the
growth factor, 9f or 5b was added to cultured SMCs,24
and 24 h later, the cell density was measured by BrdU
assay.25 The results shown in Figure 2 have indicated
that compounds 9f and 5b inhibit the cell growth dose
dependently in concentrations from 0.01 to 0.03 lM
and from 0.01 to 0.1 lM, respectively.
Bioassays were carried out as mentioned above. As
shown in Table 1, not only the (S,S)-amino phosphate
9d–f but also the (S,R)-amino bromide HBr salts 11d,f
inhibited the natural S-1P-mediated Ca2+ increase at
low concentrations (IC50 = 0.015ꢀ0.179 lM). None of
these analogues induced Ca2+ ion increase.
We also evaluated the actions on inflammation. When
vascular endothelia are injured, inflammatory cytokines
such as PDGF (platelet-derived growth factor) are re-
leased from the aggregated and activated platelets,
and promote inflammation. Severe inflammation is
110
Other analogues were also examined. As depicted in Fig-
ure 1, the enantiomer of 9d (ent-9d) and the deoxy ana-
logues of 11d (12) were prepared from commercially
available (R,R)-2-amino-1-phenyl-1,3-propanediol and
(R)-2-amino-3-phenyl-1-propanol, respectively, by the
aforementioned methods. These analogues showed nei-
ther Ca2+ increasing activity nor its inhibitory effect.
100
90
80
70
60
50
40
5b
9f
HBr
O
P
NH2
NH2
0.001
0.01
0.1
1
HO
O
Br
Dose (µm)
HO
OH
ent-9d
Figure 2. Effect of (S,S)-S-1P (5b and 9f) on the growth of rat smooth
muscle cells activated by natural S-1P. (Data are the mean standard
error from at least two experiments.)
12
Figure 1.

1118
T. Murakami et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1115–1119
tion; Chapman & Hall: New York, 1997; (c) Hannun, Y.
A.; Obeid, L. M. J. Biol. Chem. 2002, 277, 25847–25850.
3. For reviews, see: (a) Yatomi, Y.; Ozaki, Y.; Ohmori, T.;
Igarashi, Y. Prostaglandins Other Lipid Mediat. 2001, 64,
107–122; (b) Spiegel, S.; Milstien, S. J. Biol. Chem. 2002,
277, 25851–25854.
4. (a) Olivera, A.; Spiegel, S. Nature 1993, 365, 557–560; (b)
Mattie, M.; Brooker, G.; Spiegel, S. J. Biol. Chem. 1994,
269, 3181–3188.
5. (a) Lee, M.-J.; Van Brocklyn, J. R.; Thangada, S.; Liu, C.
H.; Hand, A. R.; Menzeleev, R.; Spiegel, S.; Hla, T.
Science 1998, 279, 1552–1555; (b) Sato, K.; Murata, N.;
Kon, J.; Tomura, H.; Nochi, H.; Tamoto, K.; Osada, M.;
Ohta, H.; Tokumitsu, Y.; Ui, M.; Okajima, F. Biochem.
Biophys. Res. Commun. 1998, 253, 253–256.
6. (a) Yatomi, Y.; Yamamura, S.; Ruan, F.; Igarashi, Y. J.
Biol. Chem. 1997, 272, 5291–5297; (b) An, S.; Zheng, Y.;
Bleu, T. J. Biol. Chem. 2000, 275, 288–296.
presumed to destroy homeostasis of cardiovascular or-
gans and progress arteriosclerosis. Since natural S-1P
is considered to have the same action as PDGF, S-1P
is employed as an inflammation-inducing agent to estab-
lish a pseudo-blood vessel in vitro model. By using this
model,26 the actions of the threo-(S,R)-amino-bromide
11f were examined. In the presence of 11f (0.3–
3.0 lM), the number of neutrophils transmigrating
through bovine endothelial cell layer and that adhering
to the cell layer decreased to 74–82% and 36–42% of
the control (without 11f), respectively. Thus the (S,R)-
bromide 11f has shown inhibitory effect on the S-1P-in-
duced inflammation, thereby having possibility for
maintaining the homeostasis of cardiovascular organs.
Receptor subtypes EDG-1/S1P1 and EDG-3/S1P3 are
expressed in vascular endotherial cells and vascular
SMCs,7b whereas the expression of EDG-5/S1P2 is more
abundant in SMCs than in endotherial cells.9b,27 Owing
to the slight difference in receptor subtypes,28 the modes
of actions of the S-1P analogues should be further
elucidated.
7. (a) Ancellin, N.; Hla, T. J. Biol. Chem. 1999, 274, 18997–
19002; (b) Lee, M. J.; Thangada, S.; Claffey, K. P.;
Ancellin, N.; Liu, C. H.; Kluk, M.; Volpi, M.; ShaÕafi, R.
I.; Hla, T. Cell 1999, 99, 301–312; (c) Lee, O.-H.; Kim, Y.-
M.; Lee, Y.-M.; Moon, E.-J.; Lee, D.-J.; Kim, J.-H.; Kim,
K.-W.; Kwon, Y.-K. Biochem. Biophys. Res. Commun.
1999, 264, 743–750.
8. For a review: Payne, S. G.; Milstien, S.; Barbour, S. E.;
Spiegel, S. Sem. Cell Dev. Biol. 2004, 15, 467–
476.
In summary, we have synthesized sphingosine-1-phos-
phate derivatives such as threo-(S,S)-analogues starting
from L-serine or (1S,2S)-2-amino-1-aryl-1,3-propanedi-
ols. Bioassays of the S-1P analogues using HL60cells
have indicated that (1) D-threo-(2S,3S)-S-1P homo-
logues (threo-5a,b) inhibit the natural erythro-S-1P-medi-
9. For recent studies, see: (a) Mandala, S.; Hajdu, R.;
Bergstrom, J.; Xie, J.; Milligan, J.; Thornton, R.; Shei, G.-
J.; Card, D.; Keohane, C.; Rosenbach, M.; Hale, J.;
Lynch, C. L.; Rupprecht, K.; Parsons, W.; Rosen, H.
Science 2002, 296, 346–349; (b) Osada, M.; Yatomi, Y.;
Ohmori, T.; Ikeda, H.; Ozaki, Y. Biochem. Biophys. Res.
Commun. 2002, 299, 483–487; (c) Koide, Y.; Hasegawa,
T.; Takahashi, A.; Endo, A.; Mochizuki, N.; Nakagawa,
M.; Nishida, A. J. Med. Chem. 2002, 45, 4629–4638; (d)
Hakogi, T.; Shigenari, T.; Katsumura, S.; Sano, T.;
Kohno, T.; Igarashi, Y. Bioorg. Med. Chem. Lett. 2003,
13, 661–664; (e) Clemens, J.; Davis, M. D.; Lynch, K. R.;
Macdonald, T. L. Bioorg. Med. Chem. Lett. 2003, 13,
3401–3404.
10. Parrill, A. L.; Wang, D.; Bautiska, D. L.; Brocklyn, J. R.
V.; Lorincz, Z.; Fischer, D. J.; Baker, D. L.; Liliom, K.;
Spiegel, S.; Tigyi, G. J. Biol. Chem. 2000, 275, 39379–
39384.
11. (a) Lim, H.-S.; Oh, Y.-S.; Suh, P.-G.; Chung, S.-K.
Bioorg. Med. Chem. Lett. 2003, 13, 237–240; (b) Lim, H.-
S.; Park, J.-J.; Ko, K.; Lee, M.-H.; Chung, S.-K. Bioorg.
Med. Chem. Lett. 2004, 14, 2499–2503.
12. It has been reported in Ref. 5b that, when the EDG/S1P
receptor on HL60cell (prolymphoblastoma) surface is
bound to S-1P, G-protein would be phosphorylated to
activate IP3 kinase, whereafter the intracellular Ca2+
concentration is increased.
13. Murakami, T.; Furusawa, K. Tetrahedron 2002, 58, 9257–
9263.
14. For a review, see: Liang, X.; Andersch, J.; Bols, M. J.
Chem., Soc. Perkin Trans. 1 2001, 2136–2157.
ated Ca2+ ion increase at low concentrations (IC50
=
0.015ꢀ0.031 lM), (2) the long-chain alkenyl group of
threo-S-1P is replaced by aryl (9d–f) or styryl (threo-
5c) group without significant loss of the inhibitory activ-
ity, (3) the C1 phosphate group is replaced by bromide
(11d,f) without loss of the inhibitory activity, (4) Enan-
tiomeric threo-(R,R)-amino alcohol (ent-9d) and deoxy-
amino-bromide (12) do not show the activity. Therefore
the presence and configuration of the amino alcohol
moiety appear to be very important for the antagonist-
like inhibitory effect.
Although we did not use recombinant cells overexpress-
ing exogenous EDG/S1P receptors abundantly and fur-
ther experiments are required to elucidate the
relationship between the S-1P analogues and the recep-
tors, the results described here would substantially con-
tribute to the development of potent and selective
antagonist for EDG/S1P receptors. In addition, some
analogues exhibited inhibitory effects against natural
S-1P-mediated responses: growth of vascular smooth
muscle cells and neutrophils adhesion. Thus they might
contribute therapeutic tools against vascular diseases
such as arteriosclerosis.
15. For a review, see: Wipf, P.; Jahn, H. Tetrahedron 1996, 52,
12853–12909.
16. Szulc, Z. M.; Hannun, Y. A.; Bielawska, A. Tetrahedron
Lett. 2000, 41, 7821–7824.
References and notes
1. For reviews, see: (a) Hannun, Y. A.; Bell, R. M. Science
1989, 243, 500–507; (b) Kolter, T.; Sandhoff, K. Angew.
Chem., Int. Ed. 1999, 38, 1532–1568; (c) Merrill, A. H., Jr.
J. Biol. Chem. 2002, 277, 25843–25846.
2. (a) Hannun, Y. A. Science 1996, 274, 1855–1859; (b)
Hannun, Y. A. Sphingolipid-Mediated Signal Transduc-
17. (a) Ruan, F.; Sadahira, Y.; Hakomori, S.; Igarashi, Y.
Bioorg. Med. Chem. Lett. 1992, 2, 973–978; (b) Kratzer,
B.; Schmidt, R. R. Liebigs Ann. 1995, 957–963; (c)
Boumendjel, A.; Miller, S. P. F. J. Lipid Res. 1994, 35,
2305–2311; (d) Li, S.; Wilson, W. K.; Schroepfer, G. J., Jr.
J. Lipid Res. 1999, 40, 117–125.

T. Murakami et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1115–1119
1119
18. Compound
threo-5c:
1H
NMR
(CD3OD–
C10H15NOSBr2: C, 33.63; H, 4.23; N, 3.92; S, 8.98; Br,
44.75. Found: C, 34.29; H, 4.09; N, 3.91; S, 8.91; Br, 44.62.
23. Kluk, M. J.; Hla, T. Circulation Res. 2001, 89, 496–
502.
24. The rat carotid intima was rubbed by ballooning, and a
tissue fragment was transferred by explant culture. Two
weeks later, vascular smooth muscle cells harvested were
maintained in a DMEM culture medium (Gibco) contain-
ing 10% fetal bovine serum. The cells were subcultured
several times for stabilization, whereafter the cells were
seeded at a cell density of 5 · 103 cells/cm2 for use in
experiments.
CD3CO2D = 4:1, 270MHz) d 3.65 (m, 1H), 4.03 (m,
1H), 4.14 (m, 2H), 6.22 (dd, J = 6.0, 16.0 Hz, 1H), 6.80 (d,
J = 16.0Hz, 1H), 7.22–7.40(m, 5H); HRMS (CI) calcd for
C11H17NO5P (M+H)+ 274.0774, found 274.0790.
19. Assay protocol: Cell suspension, which includes Ca2+
chelating reagent Fura-2AM, was charged into a quartz
cell, which was then mounted on a fluorometer LS50B
(Perkin Elmer, for cell measurement). An excitation
wavelength was alternately switched between 340nm (for
Fura-2 chelating with Ca2+) and 380nm (for unreacted
Fura-2) at intervals of 0.5 ms, and fluorescence intensity at
510nm was measured.
25. (a) Gratzner, H. G. Science 1982, 218, 474–475; (b)
Allison, L.; Arndt-Jovin, D. J.; Gratzner, H.; Ternynck,
T.; Robert-Nicoud, M. Cytometry 1985, 6, 584.
26. A pseudo-blood vessel in vitro inflammation model was
constructed as follows: Transwells consisting of two
compartments separated by porous membrane were used.
A single layer of bovine endothelial cells was cultured on
the membrane at the bottom surface of the transwell upper
compartment. A suspension of fluorescence-labeled neu-
trophils was added to the upper compartment, and natural
S-1P was suspended in the lower compartment to an end
concentration of 10 lM. That is, the upper compartment
corresponds to the interior of the blood vessel, while the
lower one corresponds to the site of inflammation outside
the blood vessel. Fluorescence intensities at 530nm were
measured to determine the number of the neutrophils
passing through the endothelial layer and that adhering to
it.
20. The Ca2+ ion increase by each test substance was
measured as mentioned in Ref. 19. When fluorescence
intensity was stabilized after the addition of a test
substance (within a few minutes), natural S-1P (1.0 lM)
was added to the cell suspension and fluorescence intensity
was measured to examine the inhibitory effect of the
substance against Ca2+ increase by natural S-1P. When the
substance showed the inhibitory effect, dose dependency
was examined at 0.3, 0.1, 0.03, 0.01, 0.003, and 0.001 lM,
and the IC50 (50% inhibition concentration) value was
determined from at least two independent experiments.
21. Compound 9d: mp 244–247 ꢁC (dec); 1H NMR (CD3OD–
CD3CO2D = 4:1, 270MHz) d 3.59 (m, 1H), 3.76 (m, 1H),
4.00 (m, 1H), 4.89 (d, J = 8.5 Hz, 1H), 7.35–7.43 (m, 5H);
Anal. Calcd for C9H14NO5P: C, 43.73; H, 5.71; N, 5.67.
Found: C, 43.64; H, 5.67; N, 5.48.
24
22. Compound 11f: mp ca. 150 ꢁC; ½aꢁD ꢀ13.0( c 2.0, CHCl3–
MeOH (1:1)); 1H NMR (CDCl3–CD3OD, 270MHz) d
2.50(s, 3H), 3.27 (dd, J = 6.1, 12.7 Hz, 1H), 3.57 (dd,
J = 3.4, 12.7 Hz, 1H), 3.57 (1H, m), 4.81 (d, J = 8.5 Hz,
27. Okamoto, H.; Takuwa, N.; Yokomizo, T.; Sugimoto, N.;
Sakurada, S.; Shigematsu, H.; Takuwa, Y. Mol. Cell. Biol.
2000, 20, 9247–9261.
28. Takuwa, Y.; Okamoto, H.; Takuwa, N.; Gonda, K.;
Sugimoto, N.; Sakurada, S. Mol. Cell. Endocrinol. 2001,
177, 3–11, and references cited therein.
1H), 7.28 (d, J = 8.5 Hz, 1H), 7.38 (d, J = 8.3 Hz, 1H); 13
C
NMR (CDCl3–CD3OD, 67.8 MHz) d 14.8, 29.3, 57.0,
71.6, 126.2, 126.8, 135.3, 139.5; Anal. Calcd for