Synthesis of non-competitive inhibitors of sphingomyelinases with significant activity

Article abstract of DOI:10.1016/S0960-894X(02)00888-0

A series of short-chain analogues of N-palmitoylsphingosine-1-phosphate, modified by replacement of the phosphate and the long alkenyl side chain with hydrolytically stable difluoromethylene phosphonate and phenyl, respectively, were prepared to study the structure-activity relationship for inhibition of sphingomyelinase. The study revealed that inhibition is highly dependent upon the stereochemistry of the asymmetric centers of the acylamino moiety, and resulted in identification of a non-competitive inhibitor with the same level of inhibitory activity of schyphostatin, the most potent of the few known small molecular inhibitors of sphingomyelinase.

Full text of DOI:10.1016/S0960-894X(02)00888-0

Bioorganic & Medicinal Chemistry Letters 13 (2003) 229–236
Synthesis of Non-Competitive Inhibitors of Sphingomyelinases
with Significant Activity
Tsutomu Yokomatsu,a,* Tetsuo Murano,a Takeshi Akiyama,a Junichi Koizumi,a
Shiroshi Shibuya,a Yoshiaki Tsuji,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, 8-9-1 Nanakuma, Jonan-ku, Fukuoka 841-0180, Japan
Received 2 August 2002; accepted 9 October 2002
Abstract—A series of short-chain analogues of N-palmitoylsphingosine-1-phosphate, modified by replacement of the phosphate
and the longalkenyl side chain with hydrolytically stable difluoromethylene phosphonate and phenyl, respectively, were prepared to
study the structure–activity relationship for inhibition of sphingomyelinase. The study revealed that inhibition is highly dependent
upon the stereochemistry of the asymmetric centers of the acylamino moiety, and resulted in identification of a non-competitive
inhibitor with the same level of inhibitory activity of schyphostatin, the most potent of the few known small molecular inhibitors of
sphingomyelinase.
# 2002 Elsevier Science Ltd. All rights reserved.
Introduction
While several modest inhibitors are known,1 few inhibi-
Sphingomyelin is a major structural component of the
cell membrane. The sphingomyelin metabolites, cer-
amide, sphingosine, and sphingosine-1-phosphate, have
become the focus of extensive research due to their
possible role as lipid second messengers. The primary
sphingomyelin metabolite, ceramide, is generated
through the action of either a lysosomal acid sphingo-
myelinase (A-SMase) or a membrane-bound neutral
sphingomyelinase (N-SMase) in a so-called sphingo-
myeline cycle.1 The generated ceramide is believed to
play a pivotal role in regulation of cell growth and dif-
ferentiation, cell cycle arrest, and apoptosis as the lipid
messenger.1,2 However, direct links between SMases
and specific signaling systems have not yet been fully
established. 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 sphingomyelin hydrolysis might be
implicated in pathological states such as inflammation
and AIDS.1ꢀ3
tors of significant activity against SMases are reported.
Natural product schyphostatin 1, recently isolated from
a constituent of Trichopeziza mollissima, was found to
be a competitive inhibitor towards N-SMase with IC50
of 1.0 mM.4 This natural product is the most potent of
the few known small molecule inhibitors of N-SMase.5
In a search for a novel structural class of SMase inhibi-
tors, we have been interested in the synthesis and bio-
logical evaluation of a hydrololyticaly stable analogue 2
of sphingomyelin, in which the oxygen of the phos-
phoester linkage is replaced by a difluoromethylene
(CF2) unit. The particular utility of replacement of a
phosphoric ester oxygen of p-Tyr in a peptidyl frame-
work with a CF2-unit has been demonstrated in the
synthesis of useful biochemical tools for studyingthe
mechanism of tyrosine-kinase-mediated signal trans-
ductions,6,7 since the resultingphosphonate analouge
*Correspondingauthors. Fax:+81-426-76-3239; e-mail: yokomatu@
ps.toyaku.ac.jp (T. Yokomatsu); fax:+81-92-863-0389; e-mail:
shimeno@fukuoka-u.ac.jp (H. Shimeno).
0960-894X/03/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00888-0

230
T. Yokomatsu et al. / Bioorg. Med. Chem. Lett. 13 (2003) 229–236
mimics accurately the parental phosphate in their iso-
sterical and isopolar properties.8 However, the strategy
has not been applied to synthesis of a hydrolytically
stable analogue of sphingomyelin, which might be use-
ful to create SMase inhibitors valuable for studyingthe
ceramide-mediated signaling cascade.
In an attempt to identify the minimum structural
requirements for the activity of 3, we first synthesized
the desoxy derivative 12 from 6. N,O-Acetalization of 6,
followed by desilylation gave 7 in 60% yield. The alco-
hol 7 was oxidized with the Dess–Martin periodinane;
the resultingaldehyde was treated with lithium salts of
diethyl difluoromethylphosphonate (LiCF2PO3Et2) at
25
D
ꢀ78 ꢁC in THF to give the adduct 8, ½ꢀŠ ꢀ21.6 (c 1.1,
MeOH), in 60% yield (vide infra). Deoxygenation of a-
OH in 8 was achieved via the correspondingphenyl-
thionocarbonate in two steps by the Martin procedure12
25
D
to give 9, ½ꢀŠ ꢀ12.9 (c 1.0, CHCl3), in 55% yield. The
25
compounds 8 and 9 were transformed into 3, ½ꢀŠ ꢀ19.5
D
25
D
Duringthe studies directed towards synthesis and bio-
logical evaluation of 2 and the related analogues, we
have found that a short-chain analogue of N-palmitoyl
sphingosin-1-phosphate, modified by replacement of the
phosphate moiety and the longalkenyl side chain with
hydrolytically stable difluoromethylene phosphonic acid
and phenyl, respectively, as illustrated in the structure 3,
non-competitively inhibited Mg2+-dependent N-SMase
from bovine brain microsomes with IC50 of 400 mM.9,10
This compound 3 had the ability to suppress tumor
necrosis factor (TNF) a-induced apoptosis of PC-12
neurons at a low concentration of 0.1 mM.9 In our con-
tinuous research for novel SMase inhibitors of sig-
nificant activity, we have pursued the structure–activity
relationship (SAR) of 3. In this paper, we describe that
the inhibitory activity is highly dependent upon the ste-
reochemistry of the asymmetric center of the acylamino
moiety and the studies identified a non-competitive
inhibitor with the same level of inhibitory activity of
schyphostatin.
(c 1.1, MeOH), and the desoxy derivative 12, ½ꢀŠ ꢀ9.16
(c 1.1, MeOH), respectively, through removal of the
N,O-acetal and the Boc groups, and subsequent
N-palmitoylation and hydrolysis of the esters via 10 and
11 as shown in Scheme 1.
Our previous report described that the reaction between
LiCF2PO3Et2 and the aldehyde derived from 7 pro-
ceeded with low diastereoselectivity on the basis of the
multiplicity of the 19F NMR spectrum of the adduct 8.9
However, careful reexamination of the NMR (1H, 13C,
31P, 19F) spectrum of 8 as well as its ring-transformation
to the acetonide 13 under the conditions of Dondoni13
revealed that the configuration of a-OH of the adduct 8
is homogeneous, and was determined to be R on the
basis of the observed strongcorrelation between H a and
Hb in the NOESY spectrum (500 MHz, CDCl3) of the
acetonide 13 (Scheme 2). The stereochemistry of 8 is
Results and Discussion
Our SAR-study of 3 focussed on examination of the
effects of the hydroxy group (a-OH) adjacent to the
difluoromethylene as well as the stereochemistry of the
asymmetric centers of the acylamino alcohol moiety on
the inhibitory activity. Previously, the synthesis of 3 was
achieved from (S)-Gerner aldehydes through a multi-
step sequence.9 To synthesize a series of the analogous
compounds with different stereochemistry more readily,
commercially available (1S,2S)-2-amino-1-phenyl-1,3-
propanediol 411 was chosen as the startingmaterial in
the present study (Schemes 1 and 3).
Scheme 1. Reagents and conditions: (a) Boc2O, MeOH; (b) TBDPSCl,
imidazole, DMF; (c) 4-nitrobenzoic acid, diisopropylazodicarboxyl-
ate, Ph3P, THF; (d) DIBAL-H, THF. ꢀ20 ꢁC; (e) DMP, p-TsOH,
benzene; (f) TBAF, THF; (g) Dess–Martin periodinane, CH2Cl2; (h)
LiCF2PO3Et2, THF, ꢀ78 ꢁC; (i) n-BuLi, ClC(S)OPh, THF, ꢀ78 ꢁC; (j)
n-Bu3SnH, AIBN, toluene, 90 ꢁC; (k) 3M HCl, EtOAc; (l) palmitoyl
chloride, DMAP, Et3N; (m) TMSBr, CH2Cl2; (n) MeOH.
tert-Butyloxycarbonylation of 4, followed by selective
silylation
with
tert-butyldiphenylsilyl
chloride
(TBDPSCl) gave the threo-amino alcohol derivative 5 in
a virtually quantitative yield. The Mitsunobu inversion
of the secondary hydroxyl with 4-nitrobenzoic acid,
followed by reductive removal of the benzoate gave
erythro-amino alcohol derivative 6 in 77% yield. Both
amino alcohol derivatives 5 and 6 were used for synth-
esis of the analogous compounds of 3.
Scheme 2. Ring-transformation of 8.

T. Yokomatsu et al. / Bioorg. Med. Chem. Lett. 13 (2003) 229–236
231
tive 12 showed IC50=181 mM and was found to be two-
fold more potent than our initial inhibitor 3. The results
clearly demonstrated that a-OH in 3 is not critical for
inhibition of the N-SMase.
Inhibition of the N-SMase significantly increases when
the asymmetric center of the acylamino moiety was
modified to the unnatural configuration (Table 1). The
ent-3 showed IC50=99 mM and ca. 4-fold more potent
than 3. This trend was more strikingwith the desoxy
derivatives (12 vs ent-12). The ent-12 was found to be
ca. 60-fold more potent than 12. The IC50 and Ki of ent-
12 were determined to be 3.3 and 1.6 mM, respectively.
The results reveal that the stereochemistry of the asym-
metric center of the acylamino moiety is critical for
interaction with the enzyme. The mode of inhibition for
ent-12 was determined to be non-competitive by using
Lineweaver–Burk plot analysis (Fig. 1). The data are
results from one of two experiments that produced very
similar. The ent-12 is also a good inhibitor of A-SMase
from bovine brain lysosomes, showing48% inhibition
at a low concentration of 3.3 mM. This result shows that
the inhibitor has almost identical inhibitory potency for
both N- and A-SMases, while the IC50 values of schy-
phostatin for N- and A-SMases are 1.0 and 49 mM,
respectively.4
Scheme 3. Reagents and conditions are refereed to those shown in
Scheme 1.
consistent with the fact that the reaction proceeds
through the Felkin–Ahn transition state from the less-
hindered face of the carbonyl.14
To examine the influence of the asymmetric center at the
benzylic position on the activity, we next attempted to
synthesize the threo-isomer 18 (Scheme 3). threo-Amino
alcohol derivative 5 was transformed into the N,O-ace-
tal 1415 in a similar sequence to the synthesis of alcohol
7. The aldehyde derived from 14 was condensed with
LiCF2PO3Et2 in THF at ꢀ78 ꢁC to give 15 stereo-
selectively (98% de) in 74% yield.16 The adduct 15 was
deoxygenated via the corresponding phenylthiono-
carbonate to give 16 in 97% yield, which was manipu-
lated to 18 via 17 in good overall yield as shown in
Scheme 3. While the compound 18 was isolated as good
crystals, its structure was too unstable to be evaluated
for biological activity; the palmitoyl group completely
migrated to the neighboring hydroxyl to give 20 within
24 h on standingat room temperature. The facile N!O
acyl migration may proceed by the phosphonic acid-
catalyzed formation of 19, followed by selective clea-
vage of the C–N bond.17,18
In summary, we have identified a new non-competitive
inhibitor (ent-12), which is strongly active against
Our initial design for the SMase inhibitor 3 was based on
the substrate analogues.9 Therefore the asymmetric cen-
ter of the acylamino moiety of the analogues was defined
as the same stereochemistry to that of the natural sphin-
gomyelin. However, a preliminary study on the com-
pound 3 revealed that the mode of inhibition for Mg2+
-
dependent N-SMase from bovine brain microsomes was
non-competitive.9 The analysis indicates the bindingsite
of 3 not to be the catalytic site, and prompted us to
examine influence of the stereochemistry of the asym-
metric center of the acylamino moiety on the inhibition.
For this purpose, we synthesized the enantiomers (ent-3,
and ent-12) of 3 and 12 from (1R,2R)-2-amino-1-phenyl-
1,3-propanediol (ent-4)11 through exactly the same
sequence to that for synthesis of 3 and 12.
Figure 1. Lineweaver–Burk plots for inhibition of the N-SMase with
ent-12.
Table 1. Inhibitory activity of N-palumitoyl sphingosin-1-phosphate
analogues for N-SMase from bovine brain microsomes
Compda
IC50, (mM)b
Ki (mM)c
3
12
ent-3
ent-12
377
181
99
253
Ndd
Ndd
1.6
All compounds thus prepared were assessed for inhibi-
tion of N-SMase from Bacillus cereus as well as
N-SMase from bovine brain microsomes in comparison
with 3. While all compounds had no effect on the
N-SMase from Bacillus cereus, these compounds were
good inhibitors for Mg2+-dependent N-SMase from
bovine brain microsomes (Table 1). The desoxy deriva-
3.3
aAll compounds had no effect on N-SMase from Bacillus cereus at the
IC50 concentration.
bIC50 values were determined usingfour different inhibitor concentra-
tions and represent the mean of two independent experiments.
cDetermined by the Dixon-plots analysis.
dNot determined.

232
T. Yokomatsu et al. / Bioorg. Med. Chem. Lett. 13 (2003) 229–236
Mg2+-dependent N-SMase from bovine brain micro-
somes, but inactive for N-SMase from Bacillus cereus.
The N-SMase selectivity of this inhibitor will be an
important therapeutic strategy for various cytokine-
related and ischemic diseases since the activation of N-
SMase is linked to the receptors of TNF-a and IL-1b
and ischemic stress.2 However, ent-12, differently from
schyphostatin, inhibits A-SMase in the same potency.
TNF-a also activates A-SMase to produce ceramide.20
But, the role of A-SMase in apoptosis induced by var-
ious stimuli remains an open issue. It is conceivable that
stress signaling via the SM pathway is initiated by
hydrolysis of the plasma membrane SM by N-SMase
and later achieved by the activation of A-SMase.
Therefore, at present, design of the inhibitor(s) having
higher selectivity for N-SMase will be needed. But, the
utility of an agent having the ability to inhibit both
N-SMase and A-SMase in the same potency remains
not to be bypassed.
(9H, s), 1.10 (9H, s); IR (KBr) 3437, 2930, 2857, 1693,
1497, 1170, 1113 cmꢀ1; MS (EI) m/z 506 (M++1), 392.
Anal. calcd for C30H39NO4Si: C, 71.25; H, 7.77; N,
2.77. Found: 71.47; H, 7.80; N, 2.39.
tert-Butyl (1R,2R)-1-({[tert-butyl(diphenyl)silyl]oxy}me-
thyl)-2-hydroxy-2-phenylethylcarbamate (ent-5). This
compound was prepared from ent-411 in an analogous
manner for that of the preparation of 5. The spectro-
scopic data of ent-5 were identical with those of 5 except
25
D
for the specific rotation: ½ꢀŠ ꢀ26.3 (c 1.0, CHCl3).
tert-Butyl (1S,2R)-1-({[tert-butyl(diphenyl)silyl]oxy}me-
thyl-2-hydroxy-2-phenylethylcarbamate (6). To a stirred
solution of 5 (50.7 g, 100 mmol), 4-nitrobenzoic acid
(20.5 g, 120 mmol) and triphenyl phosphine (34.5 g,120
mmol) in THF (1000 mL) was added diisopropyl azo-
dicarboxylate (23.6 mL, 120 mmol) under ice cooling.
After beingstirred at room temperature for 12 h, the
volatile components of the mixture were removed in
vacuo. The residue was portioned to ether and hexane,
and then the crystallized material was filtered. The fil-
trates were evaporated to give an oil (40.6 g), which was
dissolved in CH2Cl2 (560 mL) and treated with DIBAL
(260 mL of 0.95 M hexane solution) at ꢀ20 ꢁC for 1.5 h.
The reaction was quenched with MeOH and saturated
potassium sodium tartrate and the mixture was extrac-
ted with CHCl3. The extracts were dried (MgSO4) and
evaporated. The residue was chromatographed on silica
Experimental
General. All reactions were carried out under nitrogen
atmosphere. Optical rotations were recorded on a
JASCO DIP-360 digital polarimeter under standard
conditions. NMR data were obtained on a Bruker DPX
400 or a Varian Mercury-300 BB usingCDCl
or
gel (hexane/EtOAc=1:1) to give 6 (22.0 g, 77%) as an
3
25
D
CD3OD as a solvent. 13C NMR (100 MHz) and 31P
NMR (162 MHz) were taken with broad-band 1H
decoupling. The chemical shift data for each signal on
1H NMR (400 or 300 MHz) are expressed as relative
ppm from CHCl3 (d 7.26) or CH3OH (d 3.30). The
chemical shifts of 13C are reported relative to CDCl3 (d
77.0) or CD3OD (d 49.0). The chemical shifts of 31P are
recorded relative to external 85% H3PO4. 19F NMR
spectra (376 MHz) were measured usingbenzotri-
fluoride (BTF) as an internal reference. IR spectra were
recorded on a JASCO FTIR-620 spectrometer. Mass
spectra were measured on a Finnigan TSQ-700 or a VG
Auto Spec E spectrometer.
oil. ½ꢀŠ +6.03 (c 0.9, CHCl3); 1H NMR (CDCl3,
300 MHz) d 7.73–7.28 (15H, m), 5.34–5.24 (1H, m),
4.99–4.94 (1H, m), 3.98–3.88 (1H, m), 3.79–3.64 (2H,
m), 1.44 (9H, s), 1.06 (9H, s); MS (EI) m/z 506
(M++1).
tert-Butyl (1R,2S)-1-({[tert-butyl(diphenyl)silyl]oxy}me-
thyl-2-hydroxy-2-phenylethylcarbamate (ent-6). This
compound was prepared from ent-5 in an analogous
manner for that of the preparation of 6. The spectro-
scopic data of ent-6 were identical with those of 6 except
25
D
for the specific rotation: ½ꢀŠ ꢀ5.98 (c 1.06, CHCl3).
tert - Butyl (4S,5R) - 4 - hyroxymethyl) - 2,2 - dimethyl - 5 -
phenyl-1,3-oxazolidin-3-carboxylate (7). A solution of 6
(18.1 g, 36 mmol), 2,2-dimethoxypropane (8.8 mL, 72
mmol) and p-TsOH (70 mg, 0.3 mmol) in benzene (300
mL) was heated under reflux for 30 min. Then, the most
amounts of benzene were removed under atmospherics
pressure during1 h. The residue was poured to ether
(150 mL) and saturated NaHCO3 (150 mL). The phases
were separated and the aqueous layer was extracted
with ether. The extracts were washed with brine, dried
(MgSO4), and evaporated. The residue was dissolved in
THF (100 mL) and treated with tetrabutylammonium
fluoride (38 g, 150 mmol) at 60 ꢁC for 5 h. THF was
removed in vacuo, and the residue was poured to CHCl3
and water. The phases were separated and the aqueous
layer was extracted with CHCl3. The combined extracts
were washed with brine, dried (MgSO4) and con-
centrated. The residue was chromatographed on silica
tert-Butyl (1S,2S)-1-({[tert-butyl(diphenyl)silyl]oxy}me-
thyl)-2-hydroxy-2-phenylethylcarbamate (5). To a stirred
solution of 411 (25.0 g, 150 mmol) in MeOH (75 mL)
was added Boc2O (37.9 mL, 165 mmol) under ice-cool-
ing. The mixture was stirred for 30 min, then the solvent
was evaporated. The residue was dissolved in DMF (700
mL) and cooled to 0 ꢁC. To the stirred solution, a solu-
tion of imidazole (22.5g, 330 mmol) in DMF (260 mL)
and a solution of TBDPSCl (49.5 g, 180 mmol) in DMF
(260 mL) were successively added. After beingstirred at
room temperature for 12 h, and heated at 60 ꢁC for 1.5
h, the mixture was poured onto sat. NaHCO3 and
extracted with ether. The extracts were washed with
brine, dried (MgSO4), and evaporated in vacuo. The
residue was chromatographed on silica gel (hexane/
EtOAc=10:1) to give 5 (75 g, 95%) as crystals: mp 93–
20
95 ꢁC; ½aŠ +26.8 (c 1.0, CHCl3); H NMR (400 MHz,
CDCl3) dD7.74–7.59 (4H, m), 7.48–7.28 (11H, m), 5.19–
5.09 (1H, m), 5.06–4.99 (1H, m), 3.90–3.68 (3H, m), 1.37
1
gel (hexane/EtOAc=5:1) to give 7 (6.6 g, 60%) as col-
25
D
orless crystals: mp 60–65 ꢁC; ½ꢀŠ ꢀ29.0 (c 1.0, CHCl3);

T. Yokomatsu et al. / Bioorg. Med. Chem. Lett. 13 (2003) 229–236
233
1H NMR (300 MHz, CDCl3) d 7.50–7.25 (5H, m), 5.30
(1H, d, J=6.5 Hz), 4.20–4.09 (1H, m), 3.60–3.22 (4H,
m), 1.73 (3H, s), 1.63 (3H, s), 1.53 (9H, s); MS (EI) m/z
307 (M+). Anal. calcd for C17H25NO4: C, 66.43; H,
8.20; N, 4.56. Found: C, 66.44; H, 8.19; N, 4.55.
mmol) in THF (32 mL) was added n-BuLi (4.9 mL of
1.55 M hexane solution) at ꢀ78 ꢁC. After beingstirred
at the same temperature for 15 min, PhOC(S)Cl (1.4
mL, 10.3 mmol) was added. The mixture was stirred at
ꢀ78 ꢁC for 1 h, then quenched with satd NaHCO3. The
mixture was extracted with ether and the extracts were
evaporated. The residue was dissolved in toluene (30
mL) and treated with n-Bu3SnH (2.1 mL, 7.8 mmol) and
AIBN (10 mg) at 90 ꢁC for 2 h. Toluene was removed in
vacuo and the residue was poured between acetonitrile
and hexane. The acetonitrile extracts were evaporated.
tert - Butyl (4R,5S) - 4 - hyroxymethyl) - 2,2 - dimethyl - 5 -
phenyl-1,3-oxazolidin-3-carboxylate (ent-7). This com-
pound was prepared from ent-6 in an analogous manner
to that for the preparation of 7. The spectroscopic data
of ent-7 were identical with those of 7, except for the
25
D
specific rotation: ½ꢀŠ +29.8 (c 1.06, CHCl3).
The residue was chromatographed on silica gel (hexane/
EtOAc=5:1) to give 9 (1.37 g, 55%) as an oil. ½ꢀŠ
25
D
1
1,1-Dimethylethyl-(4S,5R)-4-{(1R)-2-[bis(ethyloxy)phos-
phoryl]-2,2-difluoro-1-hydroxyethyl}-1,3-oxazolidine-3-
carboxylate (8). A solution of 7 (4.1 g, 13.2 mmol) in
CH2Cl2 (100 mL) was treated with Dess–Martin peri-
odinane (8.4 g, 19.8 mmol) at room temperature for 4
h. The mixture was poured on cold 10% Na2SO3 and
extracted with ether. The extracts were washed with
brine, dried (MgSO4) and concentrated to give an
aldehyde (3.6 g) [1H NMR (CDCl3) d 9.10 (1H, d,
J=2.0 Hz), 7.45–7.29 (5H, m), 5.41 (1H, d, J=5.5 Hz),
4.95 (1H, dd, J=5.5, 2.0 Hz), 1.51 (3H, s), 1.39 (3H,
s)]. The aldehyde was used for the next reaction with-
out purification. To a stirred solution of LiCF2PO3Et2,
prepared from HCF2PO3Et2 (3.7 g, 23.6 mmol) and
LDA (23.6 mmol) in THF (60 mL),12 was added a
solution of the aldehyde in THF (30 mL) at ꢀ78 ꢁC.
Duringthe addition, the temperature of the reaction
mixture was carefully controlled to be below ꢀ70 ꢁC.
After beingstirred at the same temperature for 4 h, the
mixture was poured on saturated NH4Cl and extracted
with ether. The extracts were washed with brine, dried
(MgSO4) and evaporated. The residue was chromato-
graphed on silica gel to give 8 (3.2 g, 60%) as an oil.
ꢀ12.9 (c 1.0, CHCl3); H NMR (400 MHz, CDCl3) d
7.43–7.21 (5H, m), 5.26 (1H, d, J=4.6 Hz), 4.58–4.47
(1H, m), 4.17–3.93 (4H, m), 2.20–1.94 (2H, m), 1.58–
1.50 (6H, m), 1.49 (9H, s), 1.35–1.12 (6H, m); 19F NMR
(376 MHz, CDCl3) d ꢀ49.0ꢂꢀ51.9 (2F, m); 31P NMR
(162 MHz, CDCl3) d 6.91 (t, JPF=108.2 Hz); IR (neat)
2981, 1699, 1381, 1254, 1176, 1022 cmꢀ1; MS (EI) m/z
478 (M++1), 376 (M+–Boc). HRMS (EI) calcd for
C17H25NO4F2P (M+–Boc): 375.1411. Found: 375.1407.
tert - Butyl (4R,5S) - 4 - [2 - (diethoxyphosphoryl) - 2,2 -
difluoroethyl]-2,2-dimethyl-5-phenyl-1,3-oxazolidine-3-
carboxylate (ent-9). This compound was prepared from
ent-8 in an analogous manner to that for the prepara-
tion of 9. The spectroscopic data of ent-9 were identical
with those of 9, except for the specific rotation: ½ꢀŠ
25
D
+13.9 (c 1.06, CHCl3).
Diethyl (2R,3S,4R)-1,1-difluoro-2,4-dihydroxy-3-(palmi-
toylamino)-4-phenylbutylphosphonate (10). The com-
pound 8 (1.0 g, 2.1 mmol) was dissolved in EtOAc (2
mL) and treated with 1 mL of 3 M HCl at room tem-
perature for 1 h. The volatile component mixture was
removed in vacuo and the residue was acylated with
palmitoyl chloride (0.5 mL, 1.65 mmol) in CH2Cl2 (7.5
mL) in the presence of Et3N (0.41 mL, 3 mmol) and
DMAP (20 mg, 0.17 mmol) at room temperature for 2
h. The mixture was poured on 1 M HCl (5 mL) and
extracted with ether. The extracts were washed with
brine, dried (MgSO4) and evaporated. The residue was
25
D
1
½ꢀŠ ꢀ21.6 (c 1.06, MeOH); H NMR (CDCl3) d 7.42–
7.26 (5H, m), 5.33 (1H, d, J=5.1 Hz), 4.60 (1H, broad
s), 4.20–3.90 (5H, m), 1.75 (3H, s), 1.69 (3H, s), 1.51
(9H, s), 1.26 (3H, t, J=7.1 Hz), 1.19 (3H, t, J=7.1
Hz); 13C NMR (CDCl3) d 134.6, 128.1, 128.0, 126.8,
117.8 (ddd, JCP=209.6 Hz, JCF=266.5, 266.8 Hz), 93.5,
81.0, 67.2, 64.2 (dd, JCF=6.4, 6.7 Hz), 57.4 (d, JCP=7.9
Hz), 28.1, 26.2, 24.2, 16.1 (dd, JCF=5.7, 5.9 Hz); 19F
NMR (376 MHz, CDCl3) d ꢀ51.8ꢂꢀ53.6 (1F, m),
ꢀ65.5ꢂꢀ67.7 (1F, m); 31P NMR (162 MHz, CDCl3) d
7.03 (dd, JCF=99.0, 98.7 Hz); IR (neat) 3370, 2981, 2935,
1698, 1391, 1257, 1179, 1097, 1057 cmꢀ1; MS (EI) m/z
494 (M++1), 493 (M+). HRMS (EI) calcd for
C30H51F2NO6P (M+): 493.2039. Found: 493.2027.
chromatographed on silica gel (hexane/EtOAc=1:1) to
25
D
give 10 (600 mg, 60%) as an oil. ½ꢀŠ +6.1 (c 1.0,
1
CHCl3); H NMR (400 MHz, CDCl3) d 7.45–7.23 (5H,
m), 6.58 (1H, d, J=8.3 Hz), 5.06 (1H, d, J=5.7 Hz),
4.48–4.40 (1H, m), 4.29–4.08 (5H, m), 2.13–2.05 (2H,
m), 1.53–1.43 (2H,m), 1.40–1.15 (31H, m), 0.90 (3H, t,
J=6.8 Hz); 13C NMR (100 MHz, CDCl3) d 117.4,
140.8, 128.2 (2 carbons), 127.6, 126.3 (2 carbons), 118.8
(dt, JCP=207.4 Hz, JCF=263.1 Hz), 74.4, 68.7-67.7 (m),
65.7, 64.9 (2 carbons, d, JCP=6.2 Hz), 60.3, 53.0, 36.5,
31.8, 29.6, 29.5, 29.4, 29.3, 29.2, 29.1, 25.4, 22.7 (2 car-
bons, d, JCP=16.1 Hz), 16.1 (2 carbons, d, JCP=4.1
Hz), 15.1, 14.0; 19F NMR (376 MHz, CDCl3) d -54.9
(1F, ddd, JFF=307.7 Hz, JHF=7.5 Hz, JPF=99.3 Hz),
ꢀ62.6 (1F, ddd, JFF=307.7 Hz, JHF=20.0 Hz,
JPF=103.4 Hz); 31P NMR (162 MHz, CDCl3) d 6.77
(dd, JPF=99.4, 100.2 Hz); IR (neat) 3328, 2924, 2853,
1644, 1522, 1455, 1259, 1027, 701 cmꢀ1; MS (EI) m/z
592 (M++1), 574 (M+–OH). HRMS (EI) calcd for
C30H53NO6F2P (MH+): 592.3578. Found: 592.3558.
1,1-Dimethylethyl-(4R,5S)-4-{(1S)-2-[bis(ethyloxy)phos-
phoryl]-2,2-difluoro-1-hydroxyethyl}-1,3-oxazolidine-3-
carboxylate (ent-8). This compound was prepared from
ent-7 in an analogous manner to that for the prepara-
tion of 8. The spectroscopic data of ent-8 were identical
25
to those of 8, except for the specific rotation: ½ꢀŠ
D
+10.2 (c 1.06, CHCl3).
tert - Butyl (4S,5R) - 4 - [2 - (diethoxyphosphoryl) - 2,2 - di-
fluoroethyl]-2,2-dimethyl-5-phenyl-1,3-oxazolidine-3-
carboxylate (9). To a stirred solution of 8 (2.55 g, 5.2

234
T. Yokomatsu et al. / Bioorg. Med. Chem. Lett. 13 (2003) 229–236
Diethyl (2S,3R,4S)-1,1-difluoro-2,4-dihydroxy-3-(palmi-
toylamino)-4-phenylbutylphosphonate (ent-10). This com-
pound was prepared from ent-9 in an analogous manner
to that for the preparation of 10. The spectroscopic data
manner to that for the preparation of 3. The spectro-
scopic data of ent-3 were identical to those of 3, except
for the specific rotation: ½ꢀŠ +21.0 (c 1.0, MeOH).
25
D
of ent-10 were identical with those of 10, except for the
specific rotation: ½ꢀŠ ꢀ5.1 (c 1.0, CHCl3).
(3S,4R)-1,1-Difluoro-4-hydoxy-3-palmitoylamino)-4-phe-
nylbutylphosphonic acid (12). This compound was
25
D
obtained as crystals (mp 85–86 ꢁC) from 11 in an ana-
25
D
Diethyl
(3S,4R)-1,1-difluoro-4-hydroxy-3-(palmitoyl-
logous manner to that for the preparation of 3. ½ꢀŠ
1
amino)-4-phenylbutylphosphonate (11). The compound 9
was transformed to 11 in 52% yield in an analogous
ꢀ9.16 (c 1.1, MeOH); H NMR (400 MHz, CD3OD) d
7.40 (2H, d, J=7.4 Hz), 7.32 (2H, t, J=7.4 Hz), 7.24
(1H, t, J=7.4 Hz), 4.71 (1H, d, J=6.0 Hz), 4.58–4.49
(1H, m), 2.62–2.27 (2H, m), 2.23–2.09 (2H, m), 1.50–
1.38 (2H, m), 1.37–1.05 (24H, m), 0.89 (3H t, J=6.7
Hz); 13C NMR (100 MHz, CD3OD) d 176.6, 142.7,
129.2, 128.7, 127.7, 121.8 (dt, JCP=211.4 Hz,
JCF=258.9 Hz), 76.3, 51.4, 36.3, 33.5 (dt, JCP=15.6,
JCF=19.5 Hz), 33.0, 30.8, 30.7, 30.5, 30.4 (2 carbons),
30.1, 27.0, 23.7, 14.5; 19F NMR (376 MHz, CD3OD)
d?ꢀ50.9 (1F, dddd, JHF=14.2, 26.7, JPF=104.5 Hz,
JFF=295.0 Hz), ꢀ51.8 (1F, dddd, JHF=15.7, 24.7 Hz,
JPF=104.5 Hz, JFF=295.0 Hz), 31P NMR (162 MHz,
CD3OD) d 4.42 (t, JPF=104.5 Hz); IR (KBr) 3280,
2924, 2853, 1751, 1657, 1457, 1175, 1023 cmꢀ1; MS
(FAB) m/z 520 (MH+), 502 (M+–OH). Anal. calcd for
C26H44NO5F2Pꢃ H2O: C, 58.09; H, 8.62; N, 2.61. Found:
C, 57.61; H, 8.32; N, 2.67.
manner to that for the preparation of 10. Mp 66–67 ꢁC;
25
D
½ꢀŠ ꢀ24.5 (c 1.0, CHCl3); 1H NMR (400 MHz, CDCl3)
d 7.40–7.23 (5H, m), 6.21 (1H, d, J=7.1 Hz), 5.06–5.02
(1H, m), 4.53 (1H, d, J=2.9 Hz), 4.31–4.23 (1H, m),
4.22–4.00 (4H, m), 2.57–2.32 (1H, m), 2.30–2.09 (3H,
m), 1.67–1.55 (2H, m), 1.34–1.17 (30H, m), 0.88 (3H, t,
J=6.8 Hz); 13C NMR (100 MHz, CDCl3) d 174.0,
140.8, 128.1, 127.3, 125.9, 120.4 (dt, JCP=214.9,
JCF=261.2 Hz), 75.5, 64.7 (d, JCP=7.5 Hz), 64.6 (d,
JCP=7.9 Hz), 51.1, 36.6, 31.8, 31.4 (dt, JCP=16.9,
JCF=19.7 Hz), 29.6, 29.4, 29.3, 29.2, 29.1, 25.5, 22.5,
16.1 (d, JCP=5.4 Hz), 14.0; 19F NMR (376 MHz,
CDCl3) d ꢀ44.7 (1F, dddd, JHF=15.9, 21.1, JPF=106.9
Hz, JFF=301.9 Hz), ꢀ46.0 (1F, ddt, JHF=19.0,
JPF=106.9 Hz, JFF=301.9 Hz); 31P NMR (162 MHz,
CDCl3) d 6.72 (t, JPF=106.9 Hz); IR (KBr) 3323, 2925,
2854, 1647, 1550, 1261, 1026 cmꢀ1; MS (EI) m/z 575
(M+), 557. Anal. calcd for C30H52NO5F2P: C, 62.59; H,
9.10; N, 2.43. Found: C, 62.52; H, 9.10; N, 2.41.
(3R,4S)-1,1-dufluoro-4-hydoxy-3-palmitoylamino)-4-phe-
nylbutylphosphonic acid (ent-12). This compound was
obtained as crystals (mp 85–86 ꢁC) from ent-11 in an
analogous manner to that for the preparation of 3. The
Diethyl
(3R,4S)-1,1-difluoro-4-hydroxy-3-(palmitoyl-
amino)-4-phenylbutylphosphonate (ent-11). This com-
pound (mp 66–67 ꢁC) was prepared from ent-10 in an
analogous manner to that for the preparation of 10. The
spectroscopic data of ent-12 were identical with those of
25
D
12, except for the specific rotation: ½ꢀŠ +9.16 (c 1.0,
MeOH).
spectroscopic data of ent-11 were identical with those of
25
D
11, except for the specific rotation ½ꢀŠ +24.3 (c 1.3,
Diethyl {(4R,5R,6R)-5-[(tert-butoxycarbonyl)amino]-2,2-
dimethyl-6-phenyl-1,3-dioxan-4-yl} (difluoro)methylphos-
phonate (13). To a stirred solution of 8 (672 mg, 1.36
mmol) in CH2Cl2 (4 mL) was added a 0.5 M solution of
trifluoroacetic acid in CH2Cl2 (45 mL). The mixture was
stirred at room temperature for 5 min, then neutralized
with satd aq NaHCO3. The phases were separated and
the aqueous layer was extracted with CHCl3. The com-
bined organic extracts were dried (Na2SO4) and con-
centrated. The residue was chromatographed on silica
gel (hexane/EtOAc=1:1) to give 13 (638 mg95%) as an
CHCl3).
(2R,3S,4R) - 1,1 - Difluoro - 2,4 - dihydoxy - 3- (palmitoyl-
amino)-4-phenylbutylphosphonic acid (3). To a stirred
solution of 10 (100 mg, 0.17 mmol) in CH2Cl2 (0.34 mL)
was added bromotrimethylsilane (0.13 mL, 1.02 mmol).
After beingstirred at room temperature for 24 h, the
volatile components of the mixture were removed in
vacuo. The residue was treated with MeOH (1 mL) at
room temperature for 2 h. Volatile components of the
mixture was removed in vacuo to give 3 (90 mg, 85%) as
1
oil. H NMR (400 MHz, CDCl3) d 7.46–7.22 (5H, m),
25
1
an amorphous powder. ½ꢀŠ ꢀ19.5 (c 1.0, MeOH); H
5.39 (1H, d, J=9.0 Hz), 5.10–5.03 (1H, m, Ha), 4.45–
4.35 (1H, m), 4.34–4.24 (4H, m), 4.23–4.13 (1H, m, Hb),
1.46–1.21 (21H, m); MS (EI) m/z 494 (M++1).
D
NMR (400 MHz, CD3OD) d 7.45–7.23 (5H, m), 5.48–5.38
(1H, m), 4.72 (1H, d, J=8.9 Hz), 4.78–4.61 (1H, m), 4.51
(1H, d, J=8.9 Hz), 2.20–1.85 (1H, m), 1.53–1.43 (1H, m),
1.40–1.15 (31H, m), 0.90 (3H, t, J=6.8 Hz); 13C NMR
(100MHz, CD3OD) d 175.9, 143.1, 129.1, 128.4, 127.6,
124.0–115.0 (m), 74.2, 54.5, 36.7, 35.3–34.4 (m), 33.1, 30.8,
30.5, 30.0, 26.7, 23.7, 14.4; 19F NMR (376MHz, CD3OD)
d ꢀ56.2 (1F, ddd, JFF=305.4 Hz, JFP=101.5 Hz,
JHF=8.5 Hz), ꢀ63.8 (1F, ddd, JFF=305.4 Hz, JFP=101.5
Hz, JFH=20.1 Hz); 31P NMR (121 MHz, CD3OD) d 6.53
(t, JPF=100.0 Hz,); MS (FAB) m/z 518 (M+–OH).
tert-Butyl (4S,5S)-4-(hydroxymethyl)-2,2-dimethyl-5-
phenyl-1,3-oxazolidine-3-carboxylate (14). A solution of
5 (26 g, 51 mmol), 2,2-dimethoxypropane (12.6 mL, 104
mmol) and p-TsOH (100 mg, 0.51 mmol) in benzene
(430 mL) was heated under reflux for 30 min. Then, the
most amounts of benzene were removed under atmos-
pheric pressure during1 h. The residue was portioned to
ether (150 mL) and sat aq NaHCO3 (150 mL). The
organic extracts were washed with brine, dried
(MgSO4), and evaporated. The residue was dissolved in
THF (100 mL) and treated with tetrabutylammonium
fluoride (235 mL of 1 M solution in THF, 235 mmol) at
(2S,3R,4S) - 1,1 - Difluoro - 2,4 - dihydoxy - 3- (palmitoyl-
amino)-4-phenylbutylphosphonic acid (ent-3). This com-
pound was obtained from ent-10 in an analogous

T. Yokomatsu et al. / Bioorg. Med. Chem. Lett. 13 (2003) 229–236
235
50 ꢁC for 12 h. The solvent was removed in vacuo, and
the residue was portioned between water and CHCl3.
The mixture was extracted with CHCl3. The extracts
were washed with brine, dried (MgSO4) and con-
centrated. The residue was chromatographed on silica
gel (hexane:EtOAc=5:1) to give 1415 (12.3 g, 78%) as
an oil. ½ꢀŠ +32.0 (c 1.1, CHCl3); H NMR (400 MHz,
CDCl3) dD7.46–7.26 (5H, m), 4.87–4.69 (1H, m), 3.91-
3.65 (3H, m), 1.70 (3H, s), 1.58 (3H, s), 1.52 (9H, s); IR
(2H, dd, J=6.7, 8.1 Hz), 1.50–1.41 (2H, m), 1.38 (3H, t,
J=7.0 Hz), 1.36 (3H, t, J=6.9 Hz), 1.32–1.08 (24H, m),
0.88 (3H, t, J=6.8 Hz); 13C NMR (100 MHz, CDCl3) d
174.0, 141.2, 128.1, 127.4, 125.9, 120.4 (dt, t,
JCP=214.3, JCF=261.1 Hz), 74.6, 64.9 (d, JCP=6.8
Hz), 50.6, 36.5, 34.7 (dt, JCP=15.2 Hz, JCF=20.1 Hz),
31.8, 29.6 (2 carbons), 29.4, 29.3, 29.0, 25.4, 22.6, 16.3
(d, JCP=5.1 Hz), 14.0; 19F NMR (376 MHz, CDCl3) d
ꢀ45.5ꢂꢀ47.5 (2F, m); 31P NMR (162 MHz, CDCl3) d
6.87 (t, JPF=107.4 Hz); IR (neat) 3308, 2924, 2853,
1650, 1550, 1261, 1027 cmꢀ1; MS (EI) 576 (M++1),
558 (M+-OH). Anal. calcd for C30H52NO5F2P: C,
62.59; H, 9.10; N, 2.43. Found: C, 62.63; H, 8.90; N,
2.38.
25
1
(neat) 3433, 2979, 2935, 1696, 1670, 1403, 1367 cmꢀ1
;
MS (EI) m/z 308 (M++1), 292 (M+–Me). Anal. calcd
for C17H25NO4: C, 66.43; H, 8.20; N, 4.56. Found: C,
66.24; H, 8.01; N, 4.44.
tert-Butyl (4S,5S)-4-[(1R)-2-(Diethoxyphosphoryl)-2,2-
difluoro-1-hydroxyethyl]-2,2-dimethyl-5-phenyl-1,3-oxa-
zolidine-3-carboxylate (15). The compound 14 (12 g, 39
mmol) was oxidized to the aldehyde [1H NMR
(300 MHz, CDCl3) d 9.52 (1H, d, J=3.6 Hz), 7.45-7.29
(5H, m), 4.98 (1H, d, J=8.4 Hz), 4.30–4.05 (1H, m),
1.71 (3H, s), 1.52 (3H, s), 1.42 (9H,s)] with Dess–Martin
periodinane (28.1 g, 66.3 mmol) in CH2Cl2 as described
for the synthesis of 8. The aldehyde was used for the
next reaction without purification. To a stirred solution
of LiCF2PO3Et2, prepared from HCF2PO3Et2 (15.5 g,
82 mmol) and LDA (23.6 mmol) in THF (280 mL),12
was added a solution of the aldehyde in THF (55 mL) at
ꢀ78 ꢁC. After beingstirred at the same temperature for
3 h, the mixture was worked up as described for the
synthesis of 8. The crude was chromatographed on silica
(3S,4S)-1,1-Difluoro-4-hydroxy-3-(palmitoylamino)-4-
phenylbutylphosphonic acid (18). This compound was
obtained as crystals (mp 85–86 ꢁC) from 17 in an ana-
20
D
logous manner to that for preparation of 3. ½ꢀŠ ꢀ8.90
1
(c 1.1, MeOH); H NMR (400 MHz, CD3OD) d 7.40–
7.19 (5H, m), 4.85 (1H, d, J=3.7 Hz, CHOH), 4.68–
4.58 (1H, m, CHNHCO ), 2.43–2.23 (2H, m), 2.17 (2H,
t, J=7.6 Hz, COCH2), 1.52–1.38 (2H, m), 1.37–1.12
(24H, m), 0.89 (3H, t, J=6.8 Hz); 13C NMR (100 MHz,
CD3OD) d 176.7, 142.8, 129.1, 128.5, 127.5, 121.7 (dt,
JCP=211.1 Hz, JCF=258.0 Hz), 75.1, 51.2, 36.3, 35.6
(dt, JCP=15.3 Hz, JCF=20.1 Hz), 33.0, 30.7, 30.4, 30.1,
27.0, 23.7, 14.5; 19F NMR (376 MHz, CD3OD)
ꢀ50.1ꢂꢀ50.8 (2F, m), 31P NMR (162 MHz, CD3OD) d
4.28 (t, JPF=104.1 Hz). This compound was unstable
for evaluation of the biological activity; the palmitoyl
group completely migrated to the neighboring hydroxyl
to give O-palmitoyl derivative 20 on standingat room
temperature within 24 h. This fact was estimated on the
gel (hexane/EtOAc=5:1 to 3:1) to give 15 (13.7 g, 74%)
25
D
as an oil. ½ꢀŠ ꢀ24.0 (c 1.0, CHCl3); 1H NMR
(400 MHz, CDCl3) d 7.47–7.26 (5H, m), 5.19 (1H, broad
s), 5.01 (1H, d, J=8.7 Hz), 4.43–4.19 (5H, m), 1.61 (3H,
s), 1.48 (9H, s), 1.44 (3H, s), 1.37 (6H, t, J=7.0 Hz): 19F
NMR (376 MHz, CDCl3) d ꢀ49.9ꢂꢀ51.6 (2F, m), 31P
NMR (162 MHz, CDCl3) d 6.53 (t, JPF=99.8 Hz); IR
(neat) 3357, 2983, 2936, 1701, 1391, 1259, 1166, 1062
cmꢀ1; MS (EI) m/z 494 (M++1). HRMS (EI) calcd for
C21H31NO7F2P (M+–Me): 478.1806. Found: 478.1812.
1
basis of the H NMR (400 MHz, CD3OD) analysis; the
signals of the methylene protons a to the carbonyl as
well as the methin proton a to OH were gradually shif-
ted to the downfield d 2.47 (2H, dt, J=3.3, 7.4 Hz) and
d 5.90 (2H, d, J=8.0 Hz), respectively.
Assay and inhibition of SMases. Microsomal or Bacillus
cereus N-SMase activity was measured as described
previously.19 Briefly, N-palmitoyl SM (20 nmol, Sigma)
in 50 mL of 20 mM Tris–HCl buffer (pH 7.4)/50 mM
MgCl2/0.1% Tween 20 were mixed with 200 mL of
enzyme source, where indicated, contained SMase inhib-
itor. Bovine brain microsomes as an enzyme source were
prepared as follows. The brain tissues were homogenized
with 5 vol. 20 mM Tris–HCl buffer (pH 7.4)/0.15 M
NaCl/0.25 M sucrose. After a centrifugation at 8000g
for 20 min at 4 ꢁC, the supernatant was further cen-
trifuged at 105,000g for 60 min at 4 ꢁC. The pellet
(microsomal fraction) was suspended in 9 vol. 20 mM
Tris–HCl buffer (pH 7.4) and used at a protein concen-
tration of 500 mg/mL. Bacillus cereus SMase was the
product of Sigma. Enzyme reaction was carried out at
37 ꢁC for 45 min (microsomal N-SMase) or for 10 min
(Bacillus cereus N-SMase) and terminated by adding4
mL of chloroform/methanol (2:1, v/v). After termina-
tion, 1 mL of water was added, and the mixture was
vortexed and centrifuged. The lower layer was collected,
and the chloroform was allowed to evaporate. The
amounts of ceramide in the residues were determined
tert - Butyl (4S,5S) - 4 - [2 - (diethoxyphosphoryl) - 2,2 -
difluoroethyl]-2,2-dimethyl-5-phenyl-1,3-oxazolidine-3-
carboxylate (16). This compound was obtained in 97%
yield as an oil from 15 in an analogous manner for that
25
1
of the preparation of 9. ½ꢀŠ ꢀ28.3 (c 1.1, CHCl3); H
D
NMR (400 MHz, CDCl3) d 7.47–7.26 (5H, m), 5.20–
5.12 (1H, m), 4.66–4.55 (1H, m), 4.33–4.18 (4H, m),
2.71–2.48 (2H, m), 1.59 (3H, s), 1.53 (3H, s), 1.48 (9H,
s), 1.39 (3H, t, J=7.0 Hz), 1.38 (3H, t, J=7.0 Hz); 19F
NMR (376 MHz, CDCl3) d ꢀ47.4ꢂꢀ51.2 (2F, m); 31P
NMR (162 MHz, CDCl3) d 6.79 (t, JPF=106.2 Hz); IR
(neat) 2982, 1703, 1377, 1259, 1168, 1023 cmꢀ1; MS (EI)
478 (M++1). Anal. calcd for C22H34NO6F2P: C, 55.34;
H, 7.18; N, 2.93. Found: C, 54.91; H, 7.20; N, 2.63.
Diethyl (3S,4S)-1,1-difluoro-4-hydroxy-3-(palmitoyl-
amino)-4-phenylbutylphosphonate (17). This compound
was obtained in 60% yield as an oil from 16 in an ana-
25
logous manner to that for the preparation of 10. ½ꢀŠ
D
1
+12.6 (c 1.4, CHCl3); H NMR (400 MHz, CDCl3) d
7.40–7.18 (5H, m), 6.00 (1H, d, J=7.6 Hz), 4.90 (1H, d,
J=4.2 Hz), 4.34–4.13 (5H, m), 2.64–2.43 (2H, m), 2.04

236
T. Yokomatsu et al. / Bioorg. Med. Chem. Lett. 13 (2003) 229–236
fluorometrically on HPLC as described by Previati et
al.21 For the assay of A-SMase, lysosome-rich fraction
was obtained by centrifugation at 8000g for 20 min at
4 ꢁC of the above brain homogenate and used as the
enzyme source. Followingassay procedures were the same
as those described in microsomal N-SMase assay, except
that 250 mM sodium acetate buffer (pH 5.6) was used for
the suspension of lysosomes and enzyme reaction.
6. (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., Jr.; den Hentog, J.; 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.
7. 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.
Acknowledgements
8. (a) Chamber, R. D.; O’Hagan, D.; Lamount, R. B.; Jain,
S. C. J. Chem. Soc., Chem. Commun. 1990, 1053. (b) O’Hagan,
D.; Rzepa, H. S. Chem. Commun. 1997, 645 and the references
cited therein..
This work was supported in part by Grant-in-Aid for
Scientific Research (C) from the Ministry of Education,
Culture, Sports, Science and Technology of Japan.
9. Yokomatsu, T.; Takechi, H.; Akiyama, T.; Shibuya, S.;
Kominato, T.; Soeda, S.; Shimeno, H. Biooorg. Med. Chem.
Lett. 2001, 11, 1277.
References and Notes
10. In the previous paper,9 the structure of N-substituent
(COC15H31) of 3 was drawn incorrectly as COC17H35.
11. Honjo, M. Yakugaku Zassi 1953, 73, 368.
12. Martin, S. M.; Dean, D. W.; Wagman, A. S. Tetrahedron
Lett. 1992, 33, 1839.
1. Kolter, T.; Sandhoff, K. Angew. Chem. Int. Ed. 1999, 38,
1532.
2. Levade, T.; Jaffrezou, J.-P. Biochim. Biophys. Acta 1999,
1438, 1.
3. Chatterjee, S. Arterioscler. Thromb. Vasc. Biol. 1998, 18,
1523.
13. Dondoni, A.; Perrone, D. Synthesis 1993, 1162.
14. Garner, P.; Park, J. M.; Malecki, E. J. Org. Chem. 1988,
53, 4395.
15. Jayaraman, M.; Deshmukh, A. R.; Bhawal, B. M. Tetra-
hedron 1996, 52, 8989.
4. (a) Tanaka, M.; Nara, F.; Suzuki-Konagai, K.; Hosoya, T.;
Ogita, T. J. Am. Chem. Soc. 1997, 119, 7871. (b) Nara, F.;
Tanaka, M.; Hosoya, T.; Suzuki-Koyanagi, K.; Ogita, T. J.
Antibiot. 1999, 52, 525. (c) Nara, F.; Tanaka, M.; Masuda-
Inoue, S.; Yamamoto, Y.; Doi-Yoshioka, H.; Suzuki-Koya-
nagi, K.; Kumakura, S.; Ogita, T. J. Antibiot. 1999, 52, 531. (f)
Saito, S.; Tanaka, N.; Fujimoto, K.; Kogen, H. Org. Lett.
2000, 2, 505.
5. Recent studies of synthesis of inhibitors for SMases (a)
Murakami, M.; Iwama, S.; Fujii, S.; Ikeda, K.; Katsumura,
S. Bioorg. Med. Chem. Lett. 1997, 7, 1725. (b) Hakogi, T.;
Monden, Y.; Iwama, S.; Katsumura, S. Org. Lett. 2000, 2,
2627. (c) Arenz, C.; Giannis, A. Angew. Chem. Int. Ed. 2000,
39, 1440. (d) Hakogi, T.; Monden, Y.; Taichi, M.; Iwama, S.;
Fujii, S.; Ikeda, K.; Katsumura, S. J. Org. Chem. 2002, 67,
4839.
16. The stereochemistry of 15 was also ascertained in a similar
manner to that for 8.
17. Welsh, L. H. J. Am. Chem. Soc. 1949, 71, 3500.
18. Partial N!O acyl migration was also observed with the
erythro isomers (3, 12, ent-3 and ent-12). However, the rate of
the migration was estimated to be slow in comparison to that
of the threo isomers.
19. Soeda, S.; Kominato, T.; Shimeno, H. Res. Commun.
Biochem. Cell Mol. Biol. 2000, 4, 11.
20. Schwander, R.; Wiegmann, K.; Bernardo, K.; Kreder, D.;
Kronke, M. J. Biol. Chem. 1998, 273, 5916.
¨
21. Previati, M.; Bertolaso, L.; Tramarin, M.; Bertagnolo, V.;
Capitani, S. Anal. Biochem. 1996, 233, 108.