Synthesis of non-natural C2-homo-ceramide and its apoptotic activity against HL-60 cells

Article abstract of DOI:10.1016/S0960-894X(02)01026-0

Non-natural ceramide analogues, C2-homo-ceramide and C2-homo-dihydroceramide, were prepared from L-aspartic acid via L-homo-serine. The apoptotic activities of the synthesized ceramide analogues were examined in HL-60 human leukemia cells. C2-homo- and C2

Full text of DOI:10.1016/S0960-894X(02)01026-0

Bioorganic & Medicinal Chemistry Letters 13 (2003) 613–616
Synthesis of Non-Natural C2-Homo-ceramide and its Apoptotic
Activity Against HL-60 Cells
Keiji Shikata, Hayato Niiro, Hideki Azuma, Taro Tachibana and Kenji Ogino*
Department of Applied & Bioapplied Chemistry, Graduate School of Engineering,
Osaka City University, Sugimoto 3-3-138, Sumiyoshi-ku, Osaka 558-8585, Japan
Received 7 November 2002; revised 28 November 2002; accepted 28 November 2002
Abstract—Non-natural ceramide analogues, C2-homo-ceramide and C2-homo-dihydroceramide, were prepared from l-aspartic acid
via l-homo-serine. The apoptotic activities of the synthesized ceramide analogues were examined in HL-60human leukemia cells.
C2-homo- and C2-bishomo-ceramide indicate low but considerable apoptotic activities in comparison with C2-ceramide.
# 2003 Elsevier Science Ltd. All rights reserved.
Introduction
Although the experiments are preliminary, we found that
this synthetic C2-bishomo-ceramide considerably induced
apoptosis in HL60human leukaemia cells in vitro.
Sphingolipids are now well recognized as playing
important roles in cell recognition, differentiation, cell-
cell contact and cell growth.1 Specifically, ceramide
mediates apoptosis (cell death) through an intracellular
action.2 (Fig. 1) Consequently, in many studies, chemi-
cally prepared ceramide analogues have been used as a
very powerful tool to investigate the mechanism of
apoptosis. Previous studies demonstrated that sereral
kinds of C2- and C6-ceramides, which are cell-perme-
able analogues of ceramide, have been employed for the
investigation of structure–activity relationships in cer-
amide mediated apoptosis.3 These results suggest that
the apoptotic activity of dihydroceramide with the
saturated sphingoid backbone is inactive whereas the
introduction of a double bond at C4–C5 induced to
apoptosis. However, the mechanism of apoptosis that
ceramide induced has not been fully elucidated. There-
fore, we prepared novel ceramide analogues to obtain
more information on apoptotic phenomenon at the
molecular level.
In the present study, (3S,4R,5E)-3-acetylamino-5-non-
adecen-1,4-diol and its dihydro-ceramide, which have
one methylene spacer between the primary hydroxyl
and amino group of the sphingosine backbone, were
prepared. These analogues will allow us to elucidate
structure–activity relationships in ceramide-mediated
apoptosis.
Chemistry
The starting compound in the synthesis, homo-serine 3,
was obtained from l-aspartic acid according to a pre-
viously reported method.5 In order to protect the pri-
mary hydroxyl and amino group of 3 using a cyclic
system, such as Garner’s aldehyde,6 attempts to perform
deprotection of diBoc at the amino group have not been
successful. When a strong acid such as TFA was used to
remove the Boc group, the major reactant was g-lactone
4. From these results, the treatment of compound 3 with
t-butyldiphenylsilyl chloride in the presence of triethyl-
amine gave 5. One of the protecting groups in the amino
group could be removed with a very dilute solution of
TFA. When a concentrated solution of TFA was used, a
product that removed both Boc groups was generated.
We have already established the method for preparation
of (4S,5R,6E)-4-acylamino-6-eicosen-1,5-diol, a novel
and non-natural ceramide analogue, having two methyl-
ene spacer between the primary hydroxyl and amino
groups of sphingosine backbone from l-glutamic acid.4
In the next step, treatment of 6 with the lithium anion of
dimethyl methylphosphonate gave a Horner–Wadsworth–
*Corresponding author. Tel.: +81-6-6605-2799; fax: +81-6-6605-
2167; e-mail: ogino@bioa.eng.osaka-cu.ac.jp
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)01026-0

614
K. Shikata et al. / Bioorg. Med. Chem. Lett. 13 (2003) 613–616
Scheme 3. Reagents and conditions: (a) MsCl, pyridine, room tem-
perature, 30min; (b) TFA Ch 2Cl2, room temperature, 30min; (c)
CDI, THF, room temperature, 8 h.
Figure 1. Structure of sphingolipids.
Emmons (HWE) reagent 7.7 The HWE reaction with
tetradecanal in the presence of diisopropylethylamine
gave E-enone 8 exclusively. Based on our previous
result, the reduction of 8 with DIBAL-H in toluene gave
the aminoalchohol 9 (59%, syn/anti=1/6) and the
1,4-reduction product 10 (38%).4 (Scheme 1) The
TBDPS group of 9 was removed with tetra-n-butyl-
ammonium fluoride. The stereochemistry of the newly
induced chiral center in 11 was determined by spectral
analysis after conversion to the corresponding oxazoli-
dinone derivative 12 from 11. (Scheme 3) Consequently,
the stereochemical configuration of 11 is an erythro (8.0
Hz).8 Further, the enantiomer excess of 11 was deter-
mined after conversion to the bis-Mosher ester by the
1H NMR measurement (>90% e.e.).9 Without further
purification, the protecting group of 11 was removed
using an acid catalyst, and then the treatment of the
resulting product10 with acetyl chloride in NaOAc gave
the proposed compound, C2-homo-ceramide, 13.
(Scheme 2) The initial >90% e.e. could be easily raised
to >99% of 1311 after single crystallization from
AcOEt.
Furthermore, the C2-dihydro-homo-ceramide 15 was
prepared from 10 by the same method used to convert 8
to 13. (Scheme 4) The enantiomer excess of 14 was
determined after conversion to the bis-Mosher ester by
the 1H NMR measurement (>90% e.e.). Compound
1412 (>99% e.e.) was obtained after purification by
single recrystallization from AcOEt.
Biological Properties
This study examined the effects of C2-ceramide, C2-
4
Cer,13 and its analogues, C2-Cer-1 (13) and C2-Cer-2,
and their corresponding dihydroceramide analogues,
C2-DHCer, C2-DHCer-1 (15) and C2-DHCer-2. These
analogues are structurally and stereochemically similar
to C2-ceramide (Table 1). Structurally, both types of
compounds have a secondary hydroxyl group in the
a-position to the amino group, but have different
methylene spacers between the primary hydroxyl group
and amino group. To study structure–activity relation-
ships of these compounds, the apoptotic activity
Scheme 1. Reagents and conditions: (a) TFA, CH2Cl2, room temperature, 10min; (b) TBDPSCI, Et 3N, room temperature, 8 h; (c) dil.TFA,
CH2Cl2, 0 ꢀC, 1 h; (d) methylphosphonic acid dimethyl ester, n-BuLi, THF, À78 ꢀC, 2 h; (e) tetradecanal, LiCl, diisopropylethylamine, THF, room
temperature, 48 h; (f) DIBAL-H roluene, À78 ꢀC, 1 h.
Scheme 2. Reagents and conditions: (a) tetra-n-butylammonium fluoride, THF, room temperature, 2 h; (b) 1M HCl aq, 1,4-dioxane 100 ꢀC, 30min;
(c) AcCl, AcONa, THF, room temperature, 3 h.

K. Shikata et al. / Bioorg. Med. Chem. Lett. 13 (2003) 613–616
615
Scheme 4. Reagents and conditions: (a) DIBAL-H, toluene, À78 ꢀC, 1 h; (b) tetra-n-butylammonium fluoride, THF, room temperature, 2 h; (c) 1M
HCl aq, 1,4-dioxane, 100 ꢀC, 30min; (d) AcCl, AcONa, THF, room temperature, 3 h.
Table 1. Structure of ceramide analogues and percentage of apop-
tosis in HL-60cells after 6 h treatment with 10 mM
yet been identified, these data suggest that the location of
the primary hydroxyl group does not significantly affect
their apoptotic activities.
Summary
Compd
Apoptosisa (%)
Non-natural types of C2-homo-ceramide 13 and its
C2-dihydro-analogue 15 were prepared from commer-
cially available l-asparatic acid via l-homo-serine 3.
This practical method can be applied to the inter-
mediate product of the novel sphingolipid analogue
having the homo-sphingoid base. Biological observa-
tions suggest that our data could lead to the develop-
ment of a new class of anti-cancer agents.
Control
C2-Cer
3
62
13
48
21
52
20
(n=0, R=–CH=CHC13H27
)
)
)
C2-DHCer
C2-Cer-1
C2-DHCer-1
C2-Cer-2
C2-DHCer-2
(n=0, R=–CH2CH2C13H27
(n=1, R=–CH=CHC13H27
)
(n=1, R=–CH2CH2C13H27
(n=2, R=–CH=CHC13H27
)
(n=2, R=–CH2CH2C13H27
)
aValues are average of at least three separate experiments.
Further investigation related to this structure–activity
relationships is in progress.
induced by them against HL-60cells after 6 h of stimu-
lation was measured by MTT assay.14 (Table 1) In
addition, to confirm that the cell death is apoptosis, the
blebbing of cell membrane (data not shown) and the
DNA fragmentation (Fig. 2), that were known as
apoptotic characterization, were observed.
Acknowledgements
We thank Dr. N. Yasuhara (Osaka Univ.) for the gen-
erous gift of HL-60cells.
The apoptotic activities of the synthesized ceramide
analogues were examined with 10 mM after 6 h. C2-Cer
(a well-known inducer of apoptosis) was used as a
positive control. As can be seen in Table 1, the apopto-
tic activity of these ceramide analogues is in the order
References and Notes
1. (a) Hannun, Y.; Obeid, L. J. Biol. Chem. 2002, 277, 25847.
(b) Adam, D.; Heinrich, M.; Kablitz, D.; Schutze, S. Trends
Immunol. 2002, 23, 1. (c) Cuvillier, O.; Pirianov, G.; Kleuser,
B.; Venek, P. G.; Coso, O. A.; Gutkind, J. S.; Spiegel, S. Nat-
ure 1996, 381, 800.
C2-Cer
>C2-Cer-1=C2-Cer-2>C2-DHCer-1=C2-
DHCer-2>C2-DHCer. The DNA fragmentations of the
synthesized ceramide analogues were observed with 10
mM after 8 h. Figure 2 shows that the stimulation of cell
with C2-Cer, C2-Cer-1 and C2-Cer-2 resulted in the
DNA fragmentations in quantity, whereas those of
C2-DHCer, C2-DHCer-1 and C2-DHCer-2 resulted in
small quantity. There was a good correlation between the
DNA fragmentations and the apoptotic activities of
these ceramide analogues. It is interesting that C2-Cer-1
indicate a low but considerable apoptotic activity in
comparison with C2-Cer, but comparable that of
C2-Cer-2. Although the target molecule that induced
apoptosis by direct interaction with C2-ceramide has not
2. Hannun, Y. Science 1996, 274, 1855.
3. (a) Chang, Y. T.; Choi, J.; Ding, S.; Prieschl, E. E.; Baum-
ruker, T.; Lee, J. A.; Chung, S. K.; Schultz, P. G. J. Am.
Chem. Soc. 2002, 124, 1856. (b) Jonghe, S. D.; Overmeire, I. V.;
Gunst, J.; Brunst, A. D.; Hendrik, C.; Calenbergh, S. V.;
Busson, R.; Keukeleire, D. D.; Philippe, J.; Herdewijn, P.
Bioorg. Med. Chem. Lett. 1999, 9, 3159. (c) Kishida, E.;
Kasahara, M.; Takagi, Y.; Matsumura, M.; Hayashi, T.;
Kobayashi, S.; Matsuzawa, Y. J. Lipid Med. Cell Signal 1997,
16, 127. (d) Karasavvas, N.; Erukulla, R. K.; Bittman, R.;
Lockshin, R.; Zakeri, Z. Eur. J. Biochem. 1996, 236, 729. (e)
Obeid, L. M.; Linardic, C. M.; Karolak, L. A.; Hannun, Y. A.
Science 1993, 259, 1769. (f) Bielawska, A.; Crane, H. M.;
Liotta, D.; Obeid, L. M.; Hannum, Y. A. J. Biol. Chem. 1993,
268, 26226.
4. Shikata, K.; Azuma, H.; Tachibana, T.; Ogino, K. Tetra-
hedron 2002, 58, 5803.
5. Sutherland, A.; Caplan, J. F.; Vederas, J. C. Chem. Com-
mun. 1999, 555.
6. (a) Garner, P. Tetrahedron Lett. 1984, 25, 5855. (b) Koski-
nen, P. M.; Koskinen, A. M. P. Synlett 1990, 665. (c) Liang,
X.; Andersch, J.; Bols, M. J. Chem. Soc., Perkin Trans 1, 2001,
2136.
7. Chung, S. K.; Lee, J. M. Tetrahedron: Asymmetry 1999, 10,
1441.
Figure 2. DNA fragmentations of HL-60cells by treatment with (a)
ethanol vehicle; (b) 10 mM C2-Cer; (c) 10 mM C2-DHCer; (d) 10 mM
C2-Cer-1; (e) 10 mM C2-DHCer-1; (f) 10 mM C2-Cer-2 and (g) 10 mM
C2-Cer-2 after 8 h.

616
K. Shikata et al. / Bioorg. Med. Chem. Lett. 13 (2003) 613–616
8. (a) Robert, V. H.; Najib, M.; Francisco, L. J. Org. Chem.
2002, 67, 1045. (b) Luly, J. R.; Dellaria, J. F.; Plattner, J. J.;
Soderquist, J. L.; Yi, N. J. Org. Chem. 1987, 52, 1487.
9. Bakke, M.; Takizawa, M.; Sugai, T.; Ohta, H. J. Org.
Chem. 1998, 63, 6929.
29.6, 31.7, 31.9, 32.3, 51.4, 58.7, 74.4, 128.6, 134.1 and 171.3.
HRMS (FAB, positive), calcd for C21H42NO3:(M+H)+
356.3165; found 356.3173. Anal. calcd for C21H41NO3; C,
70.94; H, 11.62; N, 3.94. Found C, 70.63; H, 11.73; N, 3.95.
12. Compound 15: Mp 125–126 ꢀC; [a]D25 À13.0 (c 0.500,
1
10. The structure of 3-amino-5-nonadecen-1,4-diol was
appeared in Patent WO98/40349 from Takara Shuzo Co., Ltd,
1998, but its physical properties and stereochemistry were not
described.
CHCl3); IR(KBr) 3296, 2916, 2851, 1649 and 1549 cmÀ1; H
NMR (CDCl3): d 0.88 (3H, t, J=6.8 Hz), 1.21–1.87 (30H, m),
2.03 (3H, s), 3.55–3.70 (4H, m), 4.03–4.09 (1H, m) and 6.13
(1H, d, J=7.8 Hz); 13C NMR (CDCl3): d 14.0, 22.7, 23.2,
25.9, 29.3, 29.5, 29.6, 29.7, 30.6, 31.9, 34.3, 50.5, 58.5, 74.0 and
171.0.HRMS (FAB, positive), calcd for C21H44NO3(M+H)+
358.3321; found 358.3321.
13. Azuma, H.; Tamagaki, S.; Ogino, K. J. Org. Chem. 2000,
65, 3538.
14. Itakura, A.; Tanaka, A.; Aioi, A.; Tanogaito, H.; Mat-
suda, H. Experimental Hematology 2002, 30, 272.
11. Compound 13: Mp 107–10 ꢀC; [a]D25 À18.5 (c 0.508,
1
CHCl3); IR(KBr) 3294, 2922, 2851, 1663 and 1556 cmÀ1; H
NMR (CDCl3): d 0.88 (3H, t, J=6.8 Hz), 1.26–1.89 (24H, m),
2.03 (3H, s), 2.05 (2H, dt, J=6.8, 6.8 Hz), 3.55–3.69 (2H, m),
4.10(1H, m), 4.22 (1H, m), 5.47 (1H, dd, J=15.4, 6.1 Hz),
5.74 (1H, dt, J=15.4, 6.8 Hz) and 6.14 (1H, d, J=7.8 Hz); 13
C
NMR (CDCl3): d 14.0, 22.6, 23.1, 29.2, 29.2, 29.3, 29.5, 29.6,