Synthesis and evaluation of a difluoromethylene analogue of sphingomyelin as an inhibitor of sphingomyelinase

Article abstract of DOI:10.1016/S0960-894X(01)00179-2

A sphingomyelin analogue 2, in which the long alkenyl chain and the phosphodiester moiety of sphingomyelin were replaced by a phenyl and an isosteric difluoromethylenephosphonic acid, was prepared to evaluate its inhibitory potency to sphingomyelinase. The analogue non-competitively inhibited the neutral sphingomyelinase in bovine brain microsomes with an IC₅₀ of 400 μM. The compound had the ability to suppress tumor necrosis factor α-induced apoptosis of PC-12 neurons at a low concentration of 0.1 μM.

Full text of DOI:10.1016/S0960-894X(01)00179-2

Bioorganic & Medicinal Chemistry Letters 11 (2001) 1277–1280
Synthesis and Evaluation of a Difluoromethylene Analogue of
Sphingomyelin as an Inhibitor of Sphingomyelinase
Tsutomu Yokomatsu,a,* Hiroaki Takechi,a Takeshi Akiyama,a Shiroishi Shibuya,a
Takaaki Kominato,b Shinji Soedab and Hiroshi Shimenob,*
aSchool of Pharmacy, Tokyo University of Pharmacy & Life Science, 1432-1 Horinouchi, Hachioji, Tokyo 192-0392, Japan
bFaculty of Pharmaceutical Sciences, Fukuoka University, 8-19-1 Nanakuma, Jonan-ku, Fukuoka 814-0180, Japan
Received 23 January 2001; accepted 10 March 2001
Abstract—A sphingomyelin analogue 2, in which the long alkenyl chain and the phosphodiester moiety of sphingomyelin were
replaced by a phenyl and an isosteric difluoromethylenephosphonic acid, was prepared to evaluate its inhibitory potency to sphin-
gomyelinase. The analogue non-competitively inhibited the neutral sphingomyelinase in bovine brain microsomes with an IC50 of
400 mM. The compound had the ability to suppress tumor necrosis factor a-induced apoptosis of PC-12 neurons at a low con-
centration of 0.1 mM. # 2001Elsevier Science Ltd. All rights reserved.
Sphingomyelinase (SMase) specifically catalyzes the
hydrolysis of the phosphoester linkage of membrane
sphingomyelin (SM) to generate ceramide and phos-
phorylcholine. There are several isoforms, distinguished
by different pH optima, cellular topology, and cation
dependence. An Mg2+-dependent neutral (N-) SMase
mainly operates at plasmamembranes,1 while acidic (A-)
SMase is localized in endosomal/lysosomal compart-
ments.2 Further, Mg2+-independent N-SMase3 and
alkaline SMase4 are found in cytosol and in the gastro-
intestinal tract, respectively. Activation of N-SMase
and/or A-SMase in cells occurs in response to growth
factors, cytokines, chemotherapeutic agents, and stress
conditions. The generated ceramide is believed to play a
pivotal role in regulation of cell growth and differentia-
tion, cell cycle arrest, apoptosis, and infammation as the
lipid messenger.5 However, direct links between SMases
and specific signaling systems have not yet been estab-
lished. To establish a clear picture of the metabolic
links, the synthesis of specific SMase inhibitors is
strongly required. Additionally, the SMase inhibitor is
expected to have some clinical values, because ceramide
generation following SM hydrolysis might be implicated
in pathological states such as AIDS.6
biological evaluation of a hydrololytically stable analo-
gue 1 of SM, in which the oxygen of the phosphoester
linkage is replaced by a difluoromethylene (CF2) unit
(Fig. 1).
The particular utility of replacement of a phosphoric
ester oxygen of p-Tyr in a peptidyl framework with a
CF2-unit has been demonstrated in the synthesis of
useful biochemical tools for studying the mechanism of
tyrosine-kinase-mediated signal transductions,7,8 since
the resulting phosphonate analogue mimics accurately
the parental phosphate in its isosterical and isopolar
properties.9 However, the strategy has not been applied
to synthesis of a hydrolytically stable analogue of SM,
which might be useful to create SMase inhibitors valuable
for studying the ceramide-mediated signaling cascade.10
In a modification of sphingolipids, shortening the
sphingoid backbone as well as introducing an aromatic
group in the backbone are recently proved to be a useful
strategy to create analogues having increased solubility
and cell permeability which are available for studies in
vitro and in vivo.11 Recent studies on initial velocities of
hydrolysis of short-chain analogues of SM by SMase
In a search for a novel structural class of SMase inhibi-
tors, we have been interested in the synthesis and
*Corresponding authors. Tel.:+81-426-76-3251; fax: +81-426-76-
3239; e-mail: yokomatu@ps.toyaku.ac.jp (T. Yokomatsu); tel.: +81-
92-871-6631; fax: +81-92-863-0389; e-mail: shimeno@fukuoka-u.ac.jp
(H. Shimeno)
Figure 1. Sphingomyelin (SM) and its difluoromethylene analogue.
0960-894X/01/$ - see front matter # 2001Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00179-2

1278
T. Yokomatsu et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1277–1280
reveal that the length of the alkenyl side chain is not
important but the d-erythro stereochemisty is critical
for interaction between the substrates and the enzyme.10
Considering these findings, we decided to synthesize and
evaluate the short-chain analogue 3 and the des coline
derivative 2, prior to the synthesis of 1. An additional
feature of our interest in the analogues is to install a
hydroxyl functionality to the a position of the CF2-unit.
A framework that is composed of proximate hydroxyl
and phosphonate functionalities is known to have the
strong chelating ability of metals.12 Therefore, the ana-
logue 3 might favorably interact with SMase because of
the Mg2+-dependency. In this paper, we describe the
synthesis of 2 and its inhibitory potency against various
types of SMases. The results show that the coline moi-
ety is not necessary to inhibit SMase. Preliminary data
on the anti-apoptotic effect of 2 on the cell death of PC-12
neurons are also presented.
phous powder) in virtually quantitative yield as a mix-
ture (1:1) of diastereoisomers. Attempted separation of
the diastereoisomers failed.17
Nara et al.18,19 recently reported an N-SMase inhibitor,
scyphostatin, from a discomycete, Trichopeziza mollis-
sima, where they used rat brain microsomes (Ms) as the
enzyme source. We chose Ms from bovine organs as the
enzyme source. Prior to the inhibition experiments with
2, we examined the Mg2+-dependency of N-SMases
isolated from the the liver and brain Ms, respectively,
since the Mg2+-dependent N-SMases are known to be
critical to induce apoptosis of cells.5 As shown in Figure
2, the activity of N-SMase in the brain Ms was Mg2+
-
dependent, while the Mg2+-dependency of N-SMase in
the liver Ms was very weak.20 Figure 3 shows the con-
centration-dependent inhibition of N-SMases by 2
which was more potent against N-SMase present in
brain Ms than that in liver Ms. IC50 value of 2 for the
brain N-SMase was found to be approx. 400 mM. The
obtained IC50 value is relatively high in comparison
with that of scyphostatin for rat brain N-SMase
(IC50=1.0 mM).19 Recently, Liu and Hannun demon-
strated that, in human leukemia cells, glutathione
(GSH) blocked the activity of N-SMase at the physio-
logical concentration range of 1–20 mM.21 Lineweaver–
Burk plot analysis indicates that GSH and oxidized
GSH (GSSG) are mixed-type inhibitors against N-
SMase in bovine brain Ms (Fig. 4A). The kinetic ana-
lysis of 2 also revealed it to be a non-competitive type of
inhibitor against the N-SMase (Fig. 4B).
Results and Discussion
The analogue 2 was synthesized from the known Garner
aldehyde13 (tert-butyl (S)-4-formyl-2,2-dimethyl-3-oxa-
zolidine-1-carboxylate) as shown in Scheme 1. Accord-
ing to the literature procedures,14 the Garner aldehyde
was stereoselectively transformed to the alcohol 4.
Deprotection of the acetonide, followed by selective
silylation with tert-butyldiphenylsilyl chloride (TBDPS-
Cl), gave 5, [a]2D5 +6.03 (c 0.9, CHCl3), in 88% yield for
two steps. N,O-Acetalization of 5, followed by desilyla-
tion gave 6, [a]2D5 À29.0 (c 0.9, CHCl3), in 81% yield.
The alcohol 6 was oxidized with the Dess–Martin peri-
odinane;15 the resulting aldehyde was treated with
lithium salts of diethyl difluoromethylphosphonate at
À78 ꢀC in THF to give the adduct 7 as an inseparable
mixture (1:1) of diastereoisomers in 60% yield.16 The
acetonide and N-Boc of 6 were removed and the result-
ing amino alcohol was acylated with palmitoyl chloride
to give 8 in 60% yield. Deprotection of diethyl ester 8
under the conventional conditions gave 2 (an amor-
We next examined the specificity of 2 for SMases
obtained from various origins and for ceramide
synthase present in bovine liver Ms (Table 1). As shown
Figure 2. Magnesium-dependency of N-SMase activity in bovine liver
and brain Ms.
Scheme 1. Reagent: (a) p-TsOH (cat.), MeOH, 25 ꢀC, 7.5 h; (b)
TBDPSCl, imidazole, DMF; (c) 2,2-dimethoxypropane, p-TsOH
(cat.), benzene, 80 ꢀC; (d) Bu4NF, THF; (e) Dess–Martin periodinane,
CH2Cl2; (f) LiCF2PO3Et2, THF, À78 ꢀC; (g) 3 M HCl, EtOAc; (h)
palmitoyl chloride, Et3N, DMAP (cat.); (i) TMSBr, CH2Cl2, followed
by MeOH.
Figure 3. Determinatoin of inhibitory potency of 2 against N-SMase
activity in bovine liver and brain Ms.

T. Yokomatsu et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1277–1280
1279
in Figure 3, the compound 2 at 400 mM inhibited 50%
of the N-SMase activity in bovine brain Ms, but had a
weak inhibitory effect on N-SMase in the liver Ms (15%
inhibition). On the other hand, the compound 2 had no
effect on N-SMase derived from Bacillus cereus at
400 mM. Of the SMases tested, interestingly, 2 at 400 mM
showed the most potent inhibitory effect on A-SMase
present in bovine brain lysosomes (80% inhibition). The
A-SMase gene gives rise to both lysosomal A-SMase
and secretory N-SMase, which is Zn2+-dependent.22
Thus, because the protein nature and catalytic property
of A-SMase are very similar to those of secrotory N-
SMases, the specificity of 2 may not be limited. We also
tested the inhibition of ceramide synthase present in
bovine liver Ms, because the enzyme catalyzes the con-
jugation of acyl-CoA and sphingosine, which is the
skeletal molecule of SM. The result indicates that 2 at
400 mM has no inhibitory effect on this enzyme.
0.1 mM of 2, PC-12 neurons maintained the intact
neuronal cell morphology for at least 24 h (panel C).
In summary, we have synthesized a difluoromethylene
analogue of SM and shown that the compound 2 non-
competitively inhibits the Mg2+-dependent N-SMase in
bovine brain Ms with an IC50 value of 400 mM. How-
ever, scyphostatin from discomycete, Trichpeziza mol-
lissima is more potent than 2 as an inhibitor of
mammalian Mg2+-dependent N-SMase.19 Similarly to
the properties of 2, scyphostatin inhibits lysosomal A-
SMase with IC50 value of 49.3 mM, but not N-SMases
from S. aureus and B. cereus.19 The inhibition of Mg2+
-
dependent N-SMase by scyphostatin was mixed-type.18
Nara et al.19 have shown the utility of scyphostatin as
an inhibitor of Mg2+-dependent N-SMase by using
carrageenin-induced paw edema in rats. During the
inflammation processes, TNF-a, interleukin(IL)-1b and
other cytokines play the role of central contributors.
The cytokines such as TNF-a and IL-1b activate the
Mg2+-dependent N-SMase on inflammatory cells and
generate the lipid messenger, ceramide.24 Therefore, the
50% suppression of carrageenin-induced paw edema by
po administration (100 mg/kg) of scyphostatin may be
due to the inhibition of N-SMase. Nara et al. also
showed that scyphostatin inhibits lipopolysaccharide-
induced IL-1b production in human peripheral monocytes
with IC50 value of 0.1 mM.19
Finally, we show preliminary data indicating that 2
prevents tumor necrosis factor (TNF)-a-induced cell
death of PC-12 neurons23 at a low concentration of
0.1 mM (Fig. 5). Panel A shows the intact PC-12 neurons
(control) cultured for 24 h in serum-free medium. The
cells treated for 24 h with 500 units/mL TNF-a appear
round and show characteristics of necrotic and/or
apoptotic cells (panel B). However, in combination with
In the present study, we show for the first time the pos-
sibility that the SMase inhibitor is also available for the
suppression of neuronal cell death induced at least by
TNF-a. The compound 2 morphologically inhibited
TNF-a-induced cell death of PC-1223 neurons at a con-
centration of 0.1 mM. We confirmed that the TNF-a-
induced cell death was apoptotic and that the cytokine-
induced activation of N-SMase on PC-12 neurons was
significantly inhibited by the co-presence of 2 (these
results will be described elsewhere). Recent studies sug-
gest that neurons die from apoptosis in oxygen-deprived
brains including stroke25 and trauma26 and also in the
brains of Alzheimer’s patients.27 Such unscheduled
apoptosis may largely contribute to the pathology of
neurodegenerative diseases. Therefore, the prevention of
ceramide-induced apoptosis by treatment with an
Figure 4. Lineweaver–Burk plot analyses of SM hydrolysis by bovine
brain Ms in the presence of glutathiones (panel A) or 2 (panel B).
Table 1. Inhibitory potency of 2 at 400 mM (IC50 value for brain N-
SMase) against SMases and ceramide synthase
Enzyme
Origin
Inhibition (%)
N-SMase
N-SMase
N-SMase
A-SMase
Ceramide synthase
Bovine brain Ms
Bovine liver Ms
Bacillus cereus
50
15
0
80
0
Bovine brain lysosomes
Bovine liver Ms
Figure 5. Effect of 2 on TNF-a-induced cell death of PC-12 neurons.

1280
T. Yokomatsu et al. / Bioorg. Med. Chem. Lett. 11 (2001) 1277–1280
SMase inhibitor seems to be an attractive therapeutic
strategy for these diseases. Further structural improve-
ment of 2 and studies on their action related to neuronal
cell death in vitro and in vivo are needed.
D.; Rzepa, H. S. J. Chem. Soc., Chem. Commun. 1997, 645 and
the references cited therein.
10. (a) Synthesis and evaluation of methylene analogues of
SM as an SMase inhibitor were recently described by Katsu-
mura: Hakogi, T.; Monden, Y.; Iwama, S.; Katsumura, S.
Org. Lett. 2000, 2, 2627. (b) Murakami, M.; Iwama, S.; Fujii,
S.; Ikeda, K.; Katsumura, S. Bioorg. Med. Chem. Lett. 1997,
7, 1725.
Acknowledgements
11. Van Overmeire, H.; Boldin, S. A.; Dumont, F.; Van
Calenbergh, S.; Slegers, G.; De Keukeleire, D.; Futerman,
A. H.; Herdewijn, P. J. Med. Chem. 1999, 42, 2697.
12. Fracis, M. D.; Martodam, R. R. In The Role of Phospho-
nates in Living Systems; Hilderbrand, R. L., Ed.; CRC Press:
Boca Raton, 1983; pp 57–96.
This work was supported in part by a Grant-in-Aid for
Scientific Research (C) from the Ministry of Education,
Science, Sports and Culture of the Japanese Government.
13. Garner, P.; Park, J. M. J. Org. Chem. 1987, 52, 2361.
14. Nishida, A.; Sorimachi, H.; Iwaida, M.; Matsumizu, M.;
Kawate, T.; Nakagawa, M. Synlett 1998, 389.
References and Notes
1. Hostetler, K. Y.; Yazaki, P. J. J. Lipid Res. 1979, 20, 20.
2. Ferlinz, K.; Hurwitz, R.; Vielhaber, G.; Suzuki, K.;
Sandhoff, K. Biochem. J. 1994, 301, 855.
3. Okazaki, T.; Bielawska, A.; Domae, N.; Bell, R. M.; Han-
nun, Y. A. J. Biol. Chem. 1994, 269, 4070.
4. Nyberg, L.; Nilsson, A. Biochim. Biophys. Acta 1995, 1259, 49.
5. Levade, T.; Jaffrezou, J.-P. Biochim. Biophys. Acta 1999,
1438, 1 .
15. Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113,
7277.
16. Martin, S. F.; Dean, D. W.; Wagman, A. S. Tetrahedron
Lett. 1992, 33, 1839.
17. All assays were carried out with a mixure (1:1) of the dia-
stereoisomers.
18. Nara, F.; Tanaka, M.; Hosoya, T.; Suzuki-Konagai, K.;
Ogita, T. J. Antibiot. 1999, 52, 525.
6. Chatterjee, S. Arterioscler. Thromb. Vasc. Biol. 1998, 18,
1523.
19. Nara, F.; Tanaka, M.; Masuda-Inoue, S.; Yamasato, Y.;
Doi-Yoshioka, H.; Suzuki-Konagai, K.; Ogita, T. J. Antibiot.
1999, 52, 531.
7. (a) Burke, T. R., Jr.; Smyth, M. S.; Otaka, A.; Nomizu, M.;
Roller, P. P.; Wolf, G.; Case, R.; Shoelson, S. E. Biochemistry
1994, 33, 6490. (b) Chen, L.; Wu, L.; Otaka, A.; Smyth, M. S.;
Roller, P. P.; Burke, T. R.; den Hentog, J., Jr.; Zhang, Z.-Y.
Biochem. Biophys. Res. Commun. 1995, 216, 976. (c) Yoko-
matsu, T.; Murano, T.; Umesue, I.; Soeda, S.; Shimeno, H.;
Shibuya, S. Bioorg. Med. Chem. Lett. 1999, 9, 529 and the
references cited therein.
8. A similar replacement strategy has been also applied to
synthesis of a hydrolytically stable analogue of phospholipids
as an inhibitor of phospholipase C from Bacillus cereus: Mar-
tin, S. F.; Wong, Y.-L.; Wagman, A. S. J. Org. Chem. 1994,
59, 4821.
20. The assay methods of SMases and ceramide synthase were
as described previously: Soeda, S.; Kominato, T.; Shimeno, H.
Res. Commun. Biochem. Cell Mol. Biol. 2000, 4, 1 1 .
21. Liu, B.; Hannun, Y. A. J. Biol. Chem. 1997, 272, 16281.
22. Schissel, S. L.; Keesler, G. A.; Schuchman, E. H.; Wil-
liams, K. J.; Tabas, I. J. Biol. Chem. 1998, 273, 18250.
23. PC-12 cells were obtained from RIKEN cell bank and
neuronally differentiated with nerve growth factor.
24. Hannun, Y. A. J. Biol. Chem. 1994, 269, 3125.
25. Crowe, M. J.; Bresnahan, I. C.; Shuman, S. L.; Masters,
J. N.; Beattie, M. S. Nat. Med. 1997, 3, 73.
26. Hill, I. E.; MacManus, J. P.; Rasquinha, I.; Tuor, U. I.
Brain Res. 1995, 676, 398.
27. Pettmann, B.; Henderson, C. E. Neuron 1998, 20, 633.
9. (a) Chamber, R. D.; O’Hagan, D.; Lamount, R. B.; Jain, S.
C. J. Chem. Soc., Chem. Commun. 1990, 1053. (b) O’Hagan,