Novel sulfated gangliosides, high-affinity ligands for neural siglecs, inhibit NADase activity of leukocyte cell surface antigen CD38

Article abstract of DOI:10.1016/S0960-894X(03)00741-8

Three kinds of novel sulfated gangliosides structurally related to the Chol-1 (α-series) ganglioside GQ1bα were synthesized. These sulfated gangliosides were potent inhibitors of NADase activity of leukocyte cell surface antigen CD38. Among the synthetic

Full text of DOI:10.1016/S0960-894X(03)00741-8

Bioorganic & Medicinal Chemistry Letters 13 (2003) 3441–3445
Novel Sulfated Gangliosides, High-Affinity Ligands for Neural
Siglecs, Inhibit NADase Activity of Leukocyte Cell Surface
Antigen CD38§
Miki Hara-Yokoyama,a,* Hiromi Ito,b Kaori Ueno-Noto,c Keiko Takano,c
Hideharu Ishidab and Makoto Kisob,*
aBiochemistry, Department of Hard Tissue Engineering, Division of Bio-Matrix, Graduate School, Tokyo Medical and Dental
University, 1-5-45 Yusima, Bunkyo-ku, Tokyo 113-8549, Japan
bDepartment of Applied Bio-organic Chemistry, Gifu University, Gifu 501-1193, Japan
cGraduate School of Humanities and Sciences, Ochanomizu University, 2-1-1 Otsuka, Bunkyo-ku, Tokyo 112-8610, Japan
Received 7 May 2003; accepted 15 July 2003
Abstract—Three kinds of novel sulfated gangliosides structurally related to the Chol-1 (a-series) ganglioside GQ1ba were synthe-
sized. These sulfated gangliosides were potent inhibitors of NADase activity of leukocyte cell surface antigen CD38. Among the
synthetic gangliosides, GSC-338 (II3III6-disulfate of iso-GM1b) was surprisingly found to be the most potent structure in both the
NADase inhibition and MAG-binding activity. The present study indicates that the sulfated gangliosides are useful to study the
recognition of the internal tandem sialic acid residues a2-3-linked to Gal(II3) as well as the siglec-dependent recognition including a
terminal sialic acid residue.
# 2003 Elsevier Ltd. All rights reserved.
It has been widely recognized that gangliosides, sialic
acid-containing glycosphingolipids, are involved in var-
ious biological processes such as cell differentiation, cell
growth, cell adhesion, immune response, oncogenesis,
and many other receptor-mediated reactions.2 We have
succeeded3 in the systematic total syntheses of a variety
of gangliosides including their analogues and deriva-
tives, and have revealed their biological functions at the
molecular level.4 Among them, the cholinergic neuron-
specific Chol-1 (a-series) gangliosides5 (Fig. 1) have
been found6 to be high affinity ligands for neural siglecs7
(sialic acid-binding Ig-like lectins). The major determi-
nant of neural siglec is a2,3-linked sialic acid and the
internal sialic acids a2,3-linked to Gal (II3) and a2,6-
linked to GalNAc (III6) greatly enhance the siglec-
mediated cell adhesion. Especially, GQ1ba was a
remarkably high affinity ligand for myelin-associated
glycoprotein (MAG, siglec-4a).6 Recently, we reported
that the internal sialic acids could be substituted by
other anionic groups based on the structure-siglec
binding study.8,9
The leukocyte cell surface antigen CD3810 is a pre-
dominant NAD+ glycohydrolase in mammalian cells.11
In consistent with its ubiquitous expression, widespread
functional aspects of CD38 have been reported such as
§Synthetic Studies on Sialoglycoconjugayes, Part 133. For part 132,
see ref 1.
*Corresponding authors. Tel.: +81-3-5803-5567; fax: +81-3-5803-
0187; e-mail: m.yokoyama.bch@tmd.ac.jp (M. Hara-Yokoyama); Tel.:
+81-58-293-2916; fax: +81-58-293-2840; e-mail: kiso@cc.gifu-u.ac.jp
(M. Kiso);
Figure 1. Structure of Chol-1 (a-series) gangliosides and GT1b.
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00741-8

3442
M. Hara-Yokoyama et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3441–3445
Figure 2. Structures of novel sulfated gangliosides 1–3 and ganglioside
GM1b, GD1a, GT1b and GQ1b.
immunoregulation,12 cell proliferation,13 chemotaxis,14
insulin release15 or bone resorption.16 The NADase
activity is associated with the extracellular domain of
CD38. It has been found17,18 that these catalytic reac-
tions of CD38 are efficiently inhibited by gangliosides
such as GQ1ba, GQ1b and GT1b (Figs. 1 and 2). Thus,
the internal tandem sialic acid residue a2,3-linked to
Gal (II3) is important for the inhibition. In the present
study, we examined whether the crucial internal tandem
sialic acid residues can be substituted by other anionic
groups or not. We report here an efficient synthesis of
novel sulfated gangliosides 1–3 (Fig. 2) in which the
internal sialic acids of GQ1ba are substituted by the
sulfate groups, and their inhibitory activity for NADase
of CD38.
Scheme 1.
reactivity of the hydroxyl group at C-3 of the GalNAc
residue.
Hydrogenolytic removal of the benzyl groups, in 11, 12
and 15, followed by acetylation, gave 13, 14, and 16,
respectively, which upon selective cleavage22 of the
2-(trimethylsilyl)ethyl (SE) group, and treatment23 with
trichloroacetonitrile and DBU, afforded a series of
imidate derivatives 17–19 (Scheme 2).
For the selective protection of the II3 and III6 hydroxyl
groups, we selected the suitably protected ganglio-triose
derivatives 4 and 519 as the starting materials (Scheme 1),
which had served as the key intermediate for the first
total synthesis of ganglioside GQ1ba.19 Treatment of 4
or 5 with levulinic anhydride in pyridine in the presence
of 4-dimethylaminopyridine (DMAP) gave the III6-
levulinoyl (6) and II3III6-di-levulinoyl (7) derivatives in
94 and 90% yields, respectively. The 3,4-O-iso-
propylidene group in 6 or 7 was then cleaved by treat-
ment with 80% acetic acid at 40 ꢀC, to afford the desired
glycosyl acceptors 8 and 9 in high yields. Dimethyl(-
methylthio)sulfonium triflate (DMTST)-promoted gly-
cosylation20 of 8 or 9 with a sialyl-a(2!3)-galactose
donor 1021 was achieved b-stereoselectively to afford the
protected pentasaccharide 11 (94%), and 12 (61%)
accompanied by 15 (29%), respectively (Scheme 1). It is
of interest to note that the glycosylation of 8 (R=Bn)
with 10 gave 11 with high regioselectivity, while 9
(R=Lev) afforded a mixture of regioisomers (12 and
15). This result suggests that the steric hindrance of the
Lev group in the adjacent Gal residue reduced the
The direct glycosylation of the ceramide derivative 2024
with 17–19 was carried out in the presence of tri-
methylsilyl trifluoromethanesulfonate (TMSOTf) and
molecular sieves 4 A (AW-300) in dichloromethane to
yield the protected gangliosides 21, 22, and 27 in about
40% yields: most of the unreacted glycosyl donor (1-OH
form) and acceptor 20 were recovered. The recovered
hemiacetal was again transformed into the imidate
donors. This result is comparable to that obtained by an
efficient b-glycosidation procedure using the 2-O-piva-
loyl group as an auxiliary.25 Therefore, the yields of
ꢁ40% is satisfactory considering the additional steps in
the azidosphingosine procedure, which has been
employed for various ganglioside synthesis.3 However,
this incredibly high yield may be explained by the linear
structure of the glycosyl donors which have no bran-
ched sugar residues. The levulinoyl groups in 21, 22,
and 27 were selectively removed by treatment with
hydrazine monoacetate in ethanol to give 23, 24, and 28
respectively, which were then sulfated quantitatively

M. Hara-Yokoyama et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3441–3445
3443
potentiated the inhibitory effect (GSC-454). GSC-338
differs from GSC-454 in that the linkage between term-
inal Gal (IV) and GalNAc (III) is unnatural b1-4 form,
and inhibited the NADase activity more effectively than
GSC-454. GSC-338 was as potent as GT1b with and
without the preincubation step.
In GSC-454 and GSC-338, it is possible to locate the
two sulfate groups on GalNAc (III) and Gal (II) close
to each other, which can mimic the diphosphate struc-
ture of NAD just like the two carboxyl groups of the
tandem sialic acid residues of GT1b. Proposed struc-
tural models constructed using CAChe program28 are
shown in Figure 4. When the Gal (IV) residue is linked
to C-4 of GalNAc (GSC-338), the sugar residues I–IV
align so that the terminal sialic acid may contribute to
the recognition of the negative charge cluster of the
sulfate groups by MBP-CD38 (see Fig. 4). On the other
hand, when the Gal (IV) residue is linked to C-3 of Gal-
NAc (GSC-454), the terminal sialic acid is protruded
from the anion cluster. Such structural feature could
explain why GSC-338 is more potent than GSC-454.
In conclusion, the novel sulfated gangliosides (GSC-
335, GSC-454 and GSC-338) which have potent binding
activity to neural siglecs9 were systematically synthe-
sized. They were found to be potent inhibitors for
NADase activity of CD38. It is plausible that the two
sulfate groups on GalNAc (III) and Gal (II) substitute
the internal tandem sialic acid residues of b-series
gangliosides. Among the synthetic gangliosides, GSC-
338 (compound 3) was the most potent structure. Since
GSC-338 has also most potent MAG-binding activity,9
the anion cluster including two sulfate groups and the
terminal sialic acid must be important for expressing
both activities. Recently, a novel ganglio-series ganglio-
side 30-O-sulfo-GM1b [NeuAca2-3Galb1-3GalNAcb1-
4(HSO3-3)Galb1-4Glcb1-1Cer] has been isolated from
bovine cauda equina,29 suggesting the possibility of
occurrence of new sulfated gangliosides in nature.
However, the biological function has not been eluci-
dated because of insufficient availability of materials.
Scheme 2.
with a sulfur trioxide–pyridine complex in DMF.
Finally, all protective groups in 25, 26 and 29 were
removed by treatment with sodium methoxide in
methanol, followed by addition of water, to afford the
target compounds 1–3 in high yields (Scheme 2). The
final products were purified by chromatography
(5:4:0.7:0.07 CHCl3–MeOH–H2O–Et3N) on a column
of Sephadex LH-20. Selected physical data for synthetic
compounds are provided in ref 26.
b-Series gangliosides GQ1ba, GQ1b, and GT1b (Figs. 1
and 2) have a potent inhibitory activity on the NADase
of the extracellular domain of CD38 expressed as a
fusion protein with maltose-binding protein (MBP-
CD38).18 As GQ1ba and GT1b were more effective
than GT1aa and GD1a, respectively, the tandem sialic
acid [NeuAc-a(2,8)-NeuAc] residue a2,3-linked to the
internal Gal (II) residue seems to be crucial for the
inhibition. Furthermore, the lactonization of GT1b
greatly diminished the inhibitory potency, indicating
that the negative charges in the carboxyl group of
NeuAc are involved in the inhibition. We have pro-
posed that the negative charge cluster formed by the
two carboxyl groups of the tandem sialic acid residues
can mimic the diphosphate structure of NAD+, which
explains the inhibitory effect of b-series gangliosides.17
The effect of GSC-335, GSC-338, or GSC-454 on the
NADase activity of MBP-CD38 was examined as
described in ref 27. As shown in Figure 3, all the three
sulfated gangliosides inhibited the NADase activity.
However, GSC335 was less potent than the others. The
presence of sulfate group on C-3 of the internal Gal (II)
Figure 3. Inhibitory effect of GSC-335, 338, 454 on the NADase
activity of MBP-CD38.

3444
M. Hara-Yokoyama et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3441–3445
Figure 4. Proposed structural models of GSC-338, GSC-454 and GT1b. The boxed outlines enclose either the sulfate groups in GSC-338 and GSC-
454, or the carboxyl groups of the internal tandem sialic acids (NeuAc)2 in GT1b.
5. (a) Ishida, H.; Kiso, M. J. Synth. Org. Chem., Jpn. 2000,
58, 1108. (b) Ito, H.; Ishida, H.; Waki, H.; Ando, S.; Kiso, M.
Glycoconjugate J. 1999, 16, 585.
Therefore, the synthetic sulfated gangliosides may open
new perspectives on their structure–function study and
development of novel therapeutic agents.
6. (a) Yang, J.-S. L.; Zeller, B. C.; Shaper, L. N.; Kiso, M.;
Hasegawa, A.; Shapiro, R. E.; Schnaar, R. L. Proc. Natl.
Acad. Sci. U.S.A. 1996, 93, 814. (b) Collins, B. E.; Yang, L. J.-
S.; Mukhopadhyay, G.; Filbin, M. T.; Kiso, M.; Hasegawa,
A.; Schnaar, R. L. J. Biol. Chem. 1997, 272, 1248. (c) Collins,
B. E.; Kiso, M.; Hasegawa, A.; Tropak, M. B.; Roder, J. C.;
Crocker, P. R.; Schnaar, R. L. J. Biol. Chem. 1997, 272, 16889.
7. (a) Crocker, P. R.; Varki, A. Trends Immunol. 2001, 22,
337. (b) Angata, T.; Kerr, S. C.; Greaves, D. R.; Varki, N. M.;
Crocker, P. R.; Varki, A. J. Biol. Chem. 2002, 277, 24466.
8. Sawada, N.; Ishida, H.; Collins, B. E.; Schnaar, R. L.;
Kiso, M. Carbohydr. Res. 1999, 316, 1.
Acknowledgements
This study was supported in part by Grants-in-Aid (No.
12306007 and 15631009) for Scientific Research from
Japan Society for the Promotion of Science and by
CREST of JST (Japan Science and Tecnology Cor-
poration) (to M.K.). We are grateful to Dr. Takao
Ikami for the mass spectrometric analyses.
9. (a) Collins, B. E.; Ito, H.; Sawada, N.; Ishida, H.; Kiso, M.;
Schnaar, R. L. J. Biol. Chem. 1999, 274, 37637. (b) Ito, H.;
Ishida, H.; Collins, B. E.; Fromholt, S. E.; Schnaar, R. L.;
Kiso, M. Carbohydr. Res. In press.
References and Notes
1. Hori, K.; Sawada, N.; Ando, H.; Ishida, H.; Kiso, M. Eur.
J. Org. Chem. In press.
10. (a) Mehta, K.; Shahid, U.; Malavasi, F. FASEB J. 1996,
10, 1408. (b) Lund, F. E.; Cockayne, D. A.; Randall, T. D.;
Solvason, N.; Schuber, F.; Howard, M. C. Immunol. Rev.
1998, 161, 79.
11. Cockayne, D. A.; Muchamuel, T.; Grimaldi, C. G.; Mul-
ler-Steffner, H.; Randall, T. D.; Lund, F. E.; Murray, R.;
Schuber, F.; Howard, M. C. Blood 1998, 92, 1324.
12. Lund, F. E.; Cockayne, D. A.; Randall, T. D.; Solvason,
N.; Schuber, F.; Howard, M. C. Immunol. Rev. 1998, 161, 79.
13. Zocci, E.; Daga, A.; Usa, C.; Franco, L.; Guida, G.;
Bruzzone, S.; Costa, A.; Marchetti, C.; Flora, A. D. J. Biol.
Chem. 1998, 273, 8017.
14. Partida-Sanches, S.; Cockayne, D. A.; Monard, S.;
Jacobson, E. L.; Oppenheimer, N.; Garvy, B.; Kusser, K.;
Goodrich, S.; Howard, M.; Harmsen, A.; Randall, T. D.;
Lund, F. E. Nat. Med. 2001, 7, 1209.
15. Kato, I.; Yamamoto, Y.; Fujimura, M.; Noguchi, N.;
Takasawa, S.; Okamoto, H. J. Biol. Chem. 1999, 274, 1869.
16. Sun, U.; Adebanjo, O. A.; Moonga, B. S.; Corisdeo, S.;
Anandatheethavarada, H. K.; Biswas, G.; Arakawa, T.;
Hakeda, Y.; Koval, A.; Sodam, B.; Bevis, P. J. R.; Moser,
A. J.; Lai, F. A.; Epstein, S.; Troen, B. R.; Kumegawa, M.;
Zaidi, M. J. Cell Biol. 1999, 146, 1161.
17. Hara-Yokoyama, M.; Hirabayashi, Y.; Irie, F.; Shuto, B.;
Moriishi, K.; Sugiya, H.; Furuyama, S. J. Biol. Chem. 1995,
270, 8115.
2. (a) Sphingolipids as Signaling Modulators in the Nervous
System; Ledeen, R. W.; Hakomori, S.; Yates, A. J.; Schneider,
J. S.; Yu, R. K. Eds. The New York Academy of Science: New
York, 1998. Ann. N. Y. Acad. Sci. 845, 1. (b) Varki, A., Cum-
mings, R., Esko, J., Freeze, H., Hart, G., Marth, J., Eds.
Essentials of Glycobiology. Cold Spring Harbor Laboratory:
Cold Spring Harbor, 1999 p 115. (c) Carbohydrates in Chem-
istry and Biology; Ernst, B.; Hart, G. W.; Sinay, P., Eds.
Wiley-VCH: Weinheim, 2000, Vol. 4, p 533
3. (a) Hasegawa, A.; Kiso, M. In Carbohydrates—Synthetic
Methods and Applications in Medicinal Chemistry, Ogura, H.,
Hasegawa, A., Suami, T., Eds.; Kodansha/VCH: Tokyo/
Weinheim, 1992; pp 243–266 (b) Hasegawa, A.; Kiso, M.
Methods Enzymol. 1994, 242, 158. Kiso, M.; Hasegawa, A. In
Methods Enzymol. 1994, 242, 173. (c) Hasegawa, A.; Kiso, M.
In Preparative Carbohydrate Chemistry, Hanessian, S., Ed.;
Marcel Dekker: New York, 1997; pp 357–379 (d) Ikami, T.;
Ishida, H.; Kiso, M. Methods Enzymol. 2000, 311, 547.
4. (a) Ando, T.; Ando, H.; Kiso, M. Trends Glycosci. Glyco-
technol. 2001, 13, 573. (b) Komba, S.; Galustian, C.; Ishida,
H.; Feizi, T.; Kannagi, R.; Kiso, M. Angew. Chem. Int. Ed.
1999, 38, 1131. (c) Mitsuoka, C.; Ohmori, K.; Kimura, N.;
Kanamori, A.; Komba, S.; Ishida, H.; Kiso, M.; Kannagi, R.
Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 1597.

M. Hara-Yokoyama et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3441–3445
3445
18. (a) Hara-Yokoyama, M.; Kukimoto, I.; Nishina, H.;
Kontani, K.; Hirabayashi, Y.; Irie, F.; Sugiya, H.; Furuyama,
S.; Katada, T. J. Biol. Chem. 1996, 271, 12951. (b) Hara-
Yokoyama, M.; Nagashuka, Y.; Katsumata, O.; Irie, F.;
Kontani, K.; Hoshino, S.; Katada, T.; Ono, Y.; Fujita-Yoshi-
gaki, J.; Sugiya, H.; Furuyama, S.; Hirabayashi, Y. Biochem-
istry 2001, 40, 888.
CH2), 1.41 (m, 2H, COCH2CH2), 1.85 and 1.92 (2 s, 6H, 2
AcN), 2.74 (m, 1H, H-3Veq), 4.08 (m, 1H, H-1I), 4.28 (d, 1H,
J1,2=7.1 Hz, H-1II), 4.31 (m, 1H, H-1IV), 4.50(d, 1H,
J1,2=7.6 Hz, H-1III), 5.27 (m, 1H, ceramide-4), and 5.54 (m,
1H, ceramide-5). FAB-MS (negative ion mode, triethanola-
mine matrix): m/z 1726.8 [Mꢂ2Na]ꢂ, 1625.0[M ꢂ2Na–
SO3Na+H]ꢂ,
1435.8
[MꢂNa–NeuAc]ꢂ,
1333.8
19. Hotta, K.; Ishida, H.; Kiso, M.; Hasegawa, A. J. Carbo-
hydr. Chem. 1995, 14, 491.
[1435.8ꢂSO3Na+H]ꢂ, 1171.7 [1333.8ꢂGal]ꢂ, 968.7 [30-O-
sulfo lactosyl ceramide]0, 726.6 [glucosyl ceramide]ꢂ, 564.6
[ceramide]ꢂ; calcd for C73H129N3NaO37S2 (Mꢂ2Na):
1726.7644; found: 1726.80. Compound 3 (GSC-338): [a]D
ꢂ12.7ꢀ (c 0.2, 5:4:0.7 CHCl3–MeOH–H2O); 1H NMR
((CD3)2SO–D2O): d 0.85 (t, 6H, JMe,CH =7.0Hz, 2 MeCH2),
1.23 (s, 52H, 26 CH2), 1.44 (m, 2H, C2OCH2CH2), 1.87 and
1.90(2 s, 6H, 2 AcN), 2.71 (m, 1H, H-3 Vequiv), 4.17 (d, 1H,
J1,2=7.7 Hz, H-1I), 4.23 (d, 1H, J1,2=7.7 Hz, H-1II), 4.29 (d,
1H, J1,2=8.1 Hz, H-1IV), 4.42 (d, 1H, J1,2=7.7 Hz, H-1III),
5.32 (m, 1H, ceramide-4), and 5.52 (m, 1H, ceramide-5). FAB-
MS (negative ion mode, triethanolamine matrix): m/z 1726.9
[Mꢂ2Na]ꢂ, 1625.0[M ꢂ2Na–SO3Na+H]ꢂ, 1435.8 [MꢂNa–
20. (a) Kanie, O.; Kiso, M.; Hasegawa, A. J. Carbohydr.
Chem. 1988, 7, 501. (b) Murase, T.; Ishida, H.; Kiso, M.;
Hasegawa, A. Carbohydr. Res. 1988, 184, c1. (c) Kiso, M.;
Hasegawa, A. Methods Enzymol. 1994, 242, 173.
21. Kameyama, A.; Ishida, H.; Kiso, M.; Hasegawa, A. Car-
bohydr. Res. 1990, 200, 269.
22. Jansson, K.; Ahlfors, S.; Frejd, T.; Kihlberg, J.; Magnus-
son, G.; Dahmen, J.; Noori, G.; Stenvall, K. J. Org. Chem.
1988, 53, 5629.
23. Schmidt, R. R.; Jung, K.-H. In Carbohydrates in Chem-
istry and Biology; Ernst, B., Hart, G. W., Sinay, P., Eds.;
Wiley-VCH: Weinheim, 2000, Vol. 1, p 5.
24. Kiso, M.; Nakamura, A.; Hasegawa, A. J. Carbohydr.
Chem. 1987, 6, 411.
NeuAc]ꢂ,
1333.7
[1435.8ꢂSO3Na+H]ꢂ,
1171.8
[1333.7ꢂGal]ꢂ, 968.7 [30-O-sulfo lactosyl ceramide]ꢂ, 726.5
[glucosyl ceramide]ꢂ, 564.6 [ceramide]ꢂ; calcd for
C73H129N3NaO37S2 (Mꢂ2Na): 1726.7644; found: 1726.88.
27. NADase assay: GSC-335, GSC-338, or GSC-454 dis-
solved in DMSO–H2O (1:1, v/v) or GT1b in methanol were
dried up in test tubes. To measure the NADase activity of
MBP-CD38, the reaction mixture (10 mL) containing 5 nM
MBP-CD38 in Buffer A (50mM Tris–HCl, pH 8.0, 1 mM
EDTA) was put into the test tubes. The reaction was started
by the addition of 10 mL of 9.25 mM [carbonyl-14C]NAD (5
nCi/assay) in Buffer A. After 10min of incubation at 37 ꢀC,
the release of [14C]nicotinamide was measured as described
previously.18
25. Sato, S.; Nunomura, S.; Nakano, T.; Ito, Y.; Ogawa, T.
Tetrahedron Lett. 1988, 29, 4097.
26. Compound 1 (GSC-335): [a]D +14.1ꢀ (c 0.2, 5:4:0.7
CHCl3–MeOH–H2O); 1H NMR ((CD3)2SO–D2O): d 0.85 (t,
6H, JMe,CH 7.0Hz, 2 MeCH2), 1.23 (s, 52H, 26 CH2), 1.40(m,
2
2H, COCH2CH2), 1.80and 1.90(2 s, 6H, 2 AcN), 2.74 (m,
1H, H-3Veq), 4.17 (d, 1H, J1,2=7.3 Hz, H-1I), 4.19 (d, 1H,
J1,2=7.0Hz, H-1 II), 4.25 (d, 1H, J1,2=7.3 Hz, H-1IV), 4.52 (d,
1H, J1,2=8.4 Hz, H-1III), 5.33 (m, 1H, ceramide-4), and 5.52
(m, 1H, ceramide-5). FAB-MS (negative ion mode,
triethanolamine matrix): m/z 1646.9 [MꢂNa]ꢂ, 1624.9
[Mꢂ2Na]ꢂ, 1333.8 [MꢂNa–NeuAc]ꢂ, 1171.8 [1333.8ꢂGal]ꢂ,
888.7 [lactosyl ceramide]ꢂ, 726.7 [glucosyl ceramide]ꢂ, 564.6
[ceramide]ꢂ; calcd for C73H129N3NaO34S (MꢂNa): 1646.8076;
found: 1646.86. Compound 2 (GSC-454): [a]D +0.3ꢀ (c 0.6,
5:4:0.7 CHCl3–MeOH–H2O); 1H NMR ((CD3)2SO–D2O): d
28. CAChe3.2, Fujitsu Ltd, Tokyo, Japan, 1999.
29. (a) Kasama, T.; Hamasaki, H.; Tanaka, N.; Ishizuka, I.;
Handa, S. Proc. Jpn. Congr. Biochem. Lipids 2001, 43, 151. (b)
Kasama, T.; Hamasaki, H.; Iida-Tanaka, N.; Ishizuka, I.;
Handa, S. In Advances in Mass Spectrometry; Gelpi, E., Ed.;
John Wiley & Sons: Chichester, 2001; Vol. 15, p 557.
0.83 (t, 6H, JMe,CH =7.0Hz, 2 MeCH2), 1.22 (s, 52H, 26
2