Stereoselective synthesis and structure-activity relationship of novel ceramide trafficking inhibitors. (1R,3R)-N-(3-hydroxy-1-hydroxymethyl-3-phenylpropyl)dodecanamide and its analogues

Article abstract of DOI:10.1021/jm0300779

New ceramide trafficking inhibitors, (1R,3R)-N-(3-hydroxy-1-hydroxymethyl-3-phenylpropyl)dodecanamide (HPA-12) and a series of its analogues, were synthesized in diastereomerically and enantiomerically pure forms, and the structure - activity relationship

Full text of DOI:10.1021/jm0300779

3688
J . Med. Chem. 2003, 46, 3688-3695
Ster eoselective Syn th esis a n d Str u ctu r e-Activity Rela tion sh ip of Novel
Cer a m id e Tr a ffick in g In h ibitor s.
(1R,3R)-N-(3-Hyd r oxy-1-h yd r oxym eth yl-3-p h en ylp r op yl)d od eca n a m id e a n d Its
An a logu es
Yoshitaka Nakamura,† Ryosuke Matsubara,† Hidetoshi Kitagawa,† Shuj Kobayashi,*,† Keigo Kumagai,‡
Satoshi Yasuda,‡ and Kentaro Hanada*,‡
Graduate School of Pharmaceutical Sciences, The University of Tokyo, Hongo, Bunkyou-ku, Tokyo, 113-0033, J apan, and
Department of Biochemistry & Cell Biology, National Institute of Infectious Diseases, 1-23-1, Toyama, Shinjuku-ku,
Tokyo 162-8640, J apan
Received February 13, 2003
New ceramide trafficking inhibitors, (1R,3R)-N-(3-hydroxy-1-hydroxymethyl-3-phenylpropyl)-
dodecanamide (HPA-12) and a series of its analogues, were synthesized in diastereomerically
and enantiomerically pure forms, and the structure-activity relationship was investigated.
These analogues were stereoselectively synthesized via catalytic enantioselective Mannich-
type reactions using a Cu(II)-chiral diamine 4 complex. Analysis of HPA-12 analogues having
various lengths of the amide side chain showed that the optimal chain length for the inhibition
of sphingomyelin biosynthesis is 13 with an IC50 of ∼50 nM. Masking of the hydroxy group at
the 2′- or 3-position of HPA-12 was carried out by methylation, and it was revealed that these
hydroxy groups were essential for the activity. Installation of another hydroxy group onto HPA-
12 at the same position as that in the natural ceramide was also conducted, but no enhancement
of the activity was observed.
In tr od u ction
Sphingolipids are ubiquitous constituents of the plasma
membrane in eukaryotic cells. Sphingolipids play im-
portant roles in various biological events including cell
proliferation, differentiation, and apotosis, etc.1-4 Thus,
F igu r e 1. Novel inhibitor of ceramide trafficking, HPA-12.
inhibitors of sphingolipid biosynthesis are useful tools
for investigating the functions of sphingolipids and for
creating new types of drugs.1a Our group has also
focused on inhibitors such as sphingofungin B and F,5
khafrefungine,6 etc., and the total synthesis and clari-
fication of the structural relationship of the biological
activity of these inhibitors have been accomplished.
In mammalian cells, sphingolipid biosynthesis is
initiated by the condensation of serine and palmitoyl
coenzyme A to generate ketodihydrosphingosine, which
is, after three different enzymatic steps, converted to
ceramide in the endoplasmic reticulum (ER).7 Ceramide
is converted to sphingomyelin (SM) or glycosphingolip-
ids in the Golgi apparatus.7 The main pathway of
ceramide transport from the ER to the site of SM
biosynthesis has been shown to be an ATP- and cytosol-
dependent pathway, which is different from the ER-to-
Golgi transport pathway of glycoproteins.8,9 In our
search for novel molecules that influence sphingolipid
biosynthesis, we have found that (1R,3R)-N-(3-hydroxy-
1-hydroxymethyl-3-phenylpropyl)dodecanamide (HPA-
12 (1), Figure 1) inhibits the main pathway of ceramide
transport from the ER to the site of SM synthesis,
thereby inhibiting de novo biosynthesis of SM.10 We also
showed that the (1R,3R)-configuration of HPA-12 is the
most active among the four stereoisomers.10 In addition,
we have established an efficient route for enantioselec-
tive synthesis of (1R,3R)-HPA-12 by using a chiral Cu-
(II)-diamine complex as a chiral catalyst.11-13
In the present work, we synthesized several HPA-12
analogues in diastereomerically and enantiomerically
pure forms by the chiral Cu(II)-diamine catalyzed
Mannich-type reactions. Bioassays with the HPA-12
analogues determined the optimal length of the amide
side chain of the drug and showed the vital roles of the
hydroxy groups of HPA-12 for the bioactivity.
Resu lts a n d Discu ssion
We have recently developed an efficient synthetic
method for HPA-12 using chiral Cu(II)-catalyzed asym-
metric Mannich-type reactions of N-acylimino esters.12
In the present study, we applied this methodology to
the synthesis of various HPA analogues.
First, we prepared HPA analogues that have amide
side chains of various lengths (HPA-10 to -18). Asym-
metric Mannich-type reactions of R-N-acylimino esters
2a -d proceeded in high yields with high enantioselec-
tivities using a Cu(II)-chiral diamine 4 complex (Scheme
1). The Mannich adducts 5a -d were reduced by a
combination of K-selectride and LiBEt3H to afford HPA-
11 to -14 (6a -d ) with high anti selectivities (syn/anti
* To whom correspondence should be addressed. For chemistry,
contact S.K.: e-mail, skobayas@mol.f.u-tokyo.ac.jp; fax, 81 35 684 0634;
phone 81 35 841 4790.For biology,contact K.H.: e-mail,hanak@nih.go.jp).
† The University of Tokyo.
‡ National Institute of Infectious Diseases.
10.1021/jm0300779 CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/23/2003

Ceramide Trafficking Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3689
Ta ble 1. Effect of HPA-12 (1) and Its Analogues Having
Various Chain Lengths (6a -i) on de Novo Synthesis of SM in
CHO Cells
Sch em e 1. Synthesis of HPA Analogues Using
Cu(II)-Diamine Catalyzed Mannich-Type Reactions
a
a
IC50
IC50
(nM)
compd
(nM)
compd
C9H19 (HPA-10, 6e)
300
110
70
50
70
C14H29 (HPA-15, 6f)
90
C
C
C
C
10H21 (HPA-11, 6a )
11H23 (HPA-12, 6b ) 1)
12H25 (HPA-13, 6c)
13H27 (HPA-14, 6d )
C15H31 (HPA-16, 6g) >1000
C16H33 (HPA-17, 6h ) >1000
C17H35 (HPA-18, 6i) >1000
a
The effect of the compounds at various concentrations (0 to 1
µM) on de novo synthesis of SM was examined by metabolic
labeling of lipids in CHO cells with radioactive serine. The IC50
values were determined by Scatchard plot analysis of the data
obtained from a single series of assays for each compound.
Sch em e 3. Selective Methylation of Hydroxy Groups of
HPA-12
Sch em e 2. Synthesis of HPA Analogues by the Amide
Exchange of HPA-12
We previously demonstrated that HPA-12 inhibits
intracellular ceramide transport, but not the enzyme
activity of SM synthase, thereby inhibiting SM biosyn-
thesis in intact mammalian cells.10 To determine the
activity of the HPA-12 analogues as inhibitors of cera-
mide transport, we examined the effect of these com-
pounds on SM biosynthesis by metabolic labeling of
lipids in Chinese hamster ovary (CHO) cells with
radioactive L-serine. Among these analogues, HPA-13
(6c) most effectively inhibited SM biosynthesis with ∼50
nM, causing 50% inhibition (IC50, Table 1). The chain
lengths of 11, 12, 14, and 15 (6a , 6b ()1), 6d , 6f,
respectively) were similarly effective with their IC50 of
70-110 nM, while other chain lengths (6e,g-i) were
ineffective or far less effective (Table 1). Thus, middle
chain lengths (C11-C15) were shown to be important
for the potent bioactivity of HPA compounds.
) 16/84 to 9/91) in excellent yields.12 In the reduction
step, a very small amount of moisture was found to
decrease the diastereoselectivity (syn/anti ≈ 4/6) because
of partial epimerization of the anti isomer to the syn
isomer. To avoid this epimerization, the solvents THF
and 1,2-dimethoxyethane were distilled from ketyl
before use. Finally, these HPA analogues 6a -d were
easily purified by recrystallization from dichloromethane/
n-hexane to give diastereomerically and enantiomeri-
cally pure HPA-11 to -14.
These HPA analogues were also prepared by amide
exchange of HPA-12 (Scheme 2). The N-acyl group of
HPA-12 was removed by refluxing with hydrogen chlo-
ride in methanol, and then acyl groups of various
lengths were reintroduced selectively on the nitrogen
by p-nitrophenol esters.14 It is noted that no epimer-
ization occurred during these transformations.
To investigate the effect of the hydroxy groups of
HPA-12 on the bioactivity, we synthesized various
methylated forms of HPA-12 (Scheme 3). 2′-Methoxy-
and 2′,3-dimethoxy-HPA-12 (8 and 9) were prepared by
direct methylation of HPA-12 with MeI and Ag2O.
Selective methylation at the 3-hydroxy group was
carried out via benzoyl protection/deprotection at the
2′-position to give 3-methoxy-HPA-12 (10) in good
yield.15

3690 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Nakamura et al.
Sch em e 5. Synthesis of 2-Hydroxy Analogues of
Ta ble 2. Effect of 1 ()6b) and 8-10 on de Novo Synthesis of
SM in CHO Cells
HPA-12 from L-Serine
SM biosynthesisa
compd
(% of vehicle control)
none (vehicle control)
1 ) 6b
8
9
10
100 ( 23
4.0 ( 0.3
79 ( 4
130 ( 6
119 ( 9
a
The effect of the compounds at 1 µM on de novo synthesis of
SM was examined by metabolic labeling of lipids in CHO cells with
radioactive serine. The data (mean ( SD, n ) 3) are represented
as the percentage of the mean value of the vehicle control, in which
cells were treated with 0.01% (v/v) DMSO in the absence of the
HPA derivatives.
Sch em e 4. Synthesis of 2-Hydroxy Analogues of
HPA-12 Using Cu(II)-Diamine Catalyzed Mannich-Type
Reactions
Ta ble 3. Effect of 1 ()6b), (1S,2R,3R)-14, (1S,2S,3S)-14,
(1S,2R,3S)-14, and (1S,2S,3R)-14 on de Novo Synthesis of SM
in CHO Cells
SM biosynthesisa
compd
(% of vehicle control)
none (vehicle control)
1 ) 6b
(1S,2R,3R)-14
(1S,2S,3S)-14
(1S,2R,3S)-14
(1S,2S,3R)-14
100 ( 10
8.2 ( 1.0
40 ( 3
28 ( 2
57 ( 2
17 ( 1
a
The effect of the compounds at 1 µM on de novo synthesis of
Bioassays showed that none of these methoxy forms
are able to inhibit SM biosynthesis in intact cells even
at 1 µM (Table 2). Therefore, both the 2′- and 3- hydroxy
groups of HPA-12 are essential for the potent activity
of HPA-12.
SM was examined by metabolic labeling of lipids in CHO cells with
radioactive serine. The data (mean ( SD, n ) 3) are represented
as the percentage of the mean value of the vehicle control, in which
cells were treated with 0.01% (v/v) DMSO in the absence of the
HPA derivatives.
We also synthesized the 2-hydroxy-installed HPA
analogue 14 because natural ceramide has a hydroxy
group at the corresponding position. Four diastereomers
of14 wereprepared stereoselectivelyas follows.(1S,2R,3R)-
14 was synthesized using the Cu(OTf)2-diamine cata-
lyzed Mannich-type reaction of 2b with 2-benzyloxy-1-
phenyl-1-trimethylsiloxyethene (11) (Scheme 4). The
Mannich-type reaction proceeded with high syn selec-
tivity (syn/anti ) 89:11) and high enantioselectivity (syn
isomer: 92% ee), followed by reduction and debenzyla-
tion to give the desired (1S,2R,3R)-14. While hydro-
genolysis of the benzyloxy group of 13 using Pd/C was
found to be very sluggish, Pd(OH)2/C gave the desired
triol in moderate yield along with overreduction product
15. The other three isomers of 14 were synthesized from
L-serine (Scheme 5). Diol 16 was prepared from L-serine
according to the literature procedure,16 and then both
the Boc and acetonide groups were removed under acidic
conditions followed by N-acylation to give (1S,2S,3S)-
14. (1S,2R,3S)- and (1S,2S,3R)-14 were also synthesized
in a similar way from L-serine-derived 17.16 Dihydroxy-
lation of 17 was conveniently carried out using microen-
capsulated osmium tetraoxide17 to give triols, 18 and
19, in good yields. These diastereomers, 18 and 19, were
separated by chromatography on silica gel, and then
they were converted to the desired (1S,2R,3S)- and
(1S,2S,3R)-14 in the same manner as described above.
Bioassay showed that these triol derivatives were less
effective than HPA-12 itself as inhibitors of SM biosyn-
thesis (Table 3). Hence, installation of another hydroxy
group onto HPA-12 at the position where natural
ceramide has a hydroxy group did not enhance the
activity of HPA-12.
The (1R,3R)-configuration of HPA-12 is the most
active among the four stereoisomers, and this configu-
ration of HPA-12 resembles the stereochemistry of
natural ceramide.10 On the basis of these results, we
hypothesized that HPA-12 interacts antagonistically
with a ceramide-recognition factor involved in intrac-
ellular ceramide trafficking, although the factor remains
unidentified.10 As shown in the present study, HPA
having a C10 chain is less effective than HPAs having
C11-C15 chains. In addition, HPA having a C8 or C3
chain is far less effective or is ineffective compared to

Ceramide Trafficking Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3691
HPA-12, as shown previously.10 Thus, shortening of the
amide acyl chain below C11 may considerably weaken
the interaction of HPA derivatives with the hypothetical
factor targeted by HPA-12. Although natural ceramide
has C16 or longer acyl chains, HPA derivatives having
such long chains are ineffective as inhibitors of SM
biosynthesis. The aqueous insolubility of these HPA
derivatives might prevent them from accessing the
cytoplasm of cells, and this nature could be ameliorated
by using a middle chain length (C11-C15) as the amide
acyl chain. The requirement of both 2′- and 3-hydroxy
groups of HPA12 for its bioactivity might also indicate
that both of the hydroxyl groups are important for the
interaction of HPA-12 with its targeting factor.
Gen er a l P r oced u r e for Ma n n ich -Typ e Rea ction s Us-
in g a Ch ir a l Cop p er Ca ta lyst P r ep a r ed fr om Cu (OTf)2
a n d Ch ir a l Dia m in e Liga n d 4. Cu(OTf)2 (9.0 mg, 0.025
mmol) was dried for 2 h under vacuum at 100 °C. Ligand 4
(13.5 mg, 0.028 mmol) was added to the Cu(OTf)2 flask under
an argon atmosphere, and then CH2Cl2 (1.0 mL) was added.
The light-blue solution was stirred for over 2 h until the color
changed to dark-green. CH2Cl2 (3.2 mL) was added, and then
the solution was cooled to 0 °C. Silyl enol ether 3 (0.375 mmol)
in CH2Cl2 (0.8 mL) was added to this mixture. N-Acylimino
ester 2 (0.25 mmol) in CH2Cl2 (2.5 mL) was then added slowly
over 20 min, and the reaction mixture was further stirred at
0 °C for 12 h. The reaction was quenched by addition of
saturated aqueous NaHCO3, and the reaction mixture was
allowed to warm to room temperature and stirred until the
color of the aqueous layer changed to blue. The mixture was
extracted with CH2Cl2. The organic layer was washed with
brine and dried over anhydrous Mg2SO4, and the solvent was
evaporated. The residue was dissolved in CH2Cl2 (3.0 mL), and
4 N HCl in AcOEt (0.1 mL) was added to the solution. The
mixture was stirred at room temperature for 10 min and then
evaporated to dryness. The crude product was purified by
chromatography on silica gel to afford the desired Mannich
adduct.
Con clu sion
HPA-12 is an inhibitor of intracellular transport of
ceramide from the ER to the Golgi compartment for SM
biosynthesis.10 In the present study, we synthesized a
series of HPA-12 analogues via catalytic asymmetric
Mannich-type reactions using a Cu(II)-chiral diamine
4 complex. This methodology provides an efficient and
concise approach to a series of HPA analogues in a
diastereoselective and enantioselective manner. Bioas-
says with these analogues showed that the optimal
length of the amide side chain is 13 and that the
hydroxy groups at the 2′- and 3-position are essential
for the inhibition of SM biosynthesis. Installation of
another hydroxy group at the position where natural
ceramide has a hydroxy group did not enhance the
bioactivity. Collectively, we conclude that the structure
of (1R,3R)-HPA-12 is almost optimal among phenylpro-
pyl derivatives in terms of the inhibition of intracellular
ceramide trafficking.
(2R)-4-Oxo-4-p h en yl-2-u n d eca n oyla m in obu tyr ic a cid
29
eth yl ester (5a ): [R]
-58.7 (94% ee, c 1.00, CHCl3); mp
D
1
41-42 °C; H NMR (CDCl3) δ 0.87 (t, 3H, J ) 6.8 Hz), 1.15-
1.35 (m, 17H), 1.55-1.70 (m, 2H), 2.20 (dt, 2H, J ) 3.3, 7.6
Hz), 3.59 (dd, 1H, J ) 4.1, 18.3 Hz), 3.75 (dd, 1H, J ) 4.0,
18.3 Hz), 4.21 (q, 2H, J ) 7.1 Hz), 4.96 (dt, 1H, J ) 8.0, 3.9
Hz), 6.60 (brd, 1H, J ) 8.0 Hz), 7.48 (apparent t, 2H, J ) 7.7
Hz), 7.60 (apparent t, 1H, J ) 7.4 Hz), 7.94 (apparent d, 2H,
J ) 7.3 Hz); 13C NMR (CDCl3) δ 14.0, 14.1, 22.6, 25.5, 29.1,
29.2, 29.3, 29.4, 29.5, 31.8, 36.5, 40.5, 48.2, 61.7, 128.1, 128.7,
133.7, 136.0, 171.2, 172.9, 198.0; IR (KBr) 3334, 2918, 2850,
1732, 1684, 1649, 1543 cm-1; LRMS (FAB) m/z 389 [M]+. Anal.
(C23H35NO4) C, H, N. Chiral HPLC: Daicel Chiralpak AD,
hexane/iPrOH ) 19/1, flow rate ) 1.0 mL/min; tR ) 28.8 min
(R), tR ) 23.6 min (S).
(2R)-2-Dod eca n oyla m in o-4-oxo-4-p h en ylbu tyr ic a cid
Exp er im en ta l Section
23
eth yl ester (5b): [R]
-58.5 (94% ee, c 1.02, CHCl3); mp
D
Gen er a l. Melting points are uncorrected. 1H and 13C NMR
spectra were recorded on a J EOL J NM-LA300, J NM-LA400,
or J NM-LA500 spectrometer. Tetramethylsilane (TMS) served
1
50-51 °C; H NMR (CDCl3) δ 0.88 (t, 3H, J ) 6.8 Hz), 1.2-
1.4 (m, 19H), 1.5-1.7 (m, 2H), 2.21 (dt, 2H, J ) 2.4, 7.7 Hz),
3.60 (dd, 1H, J ) 4.1, 18.3), 3.75 (dd, 1H, J ) 4.1, 18.3), 4.21
(q, 2H, J ) 7.1 Hz), 4.96 (dt, 1H, J ) 4.1, 8.0 Hz), 6.61 (brd,
1H, J ) 8.0 Hz), 7.47 (apparent t, 2H, J ) 7.6 Hz), 7.60
(apparent t, 1H, J ) 7.3 Hz), 7.94 (apparent d, 2H, J ) 7.5
Hz); 13C NMR (CDCl3) δ 14.0, 14.1, 22.6, 25.5, 29.1, 29.3, 29.4,
29.5, 31.9, 36.5, 40.5, 48.2, 61.7, 128.1, 128.7, 133.7, 136.0,
171.2, 172.9, 198.0; IR (neat) 3310, 2925, 2855, 1742, 1682,
1657, 1529 cm-1; LRMS (EI) m/z 403 [M]+. Anal. (C24H37NO4)
C, H, N. HPLC: Daicel Chiralpak AD, hexane/iPrOH ) 19/1,
flow rate ) 1.0 mL/min; tR ) 21.9 min (S), tR ) 26.6 min (R).
1
as an internal standard (δ 0) for H NMR, and CDCl3 was used
as an internal standard (δ 77.0) for 13C NMR. IR spectra were
measured with a J ASCO FT/IR-610 spectrometer. Optical
rotations were measured with a J ASCO P-1010 polarimeter.
Mass spectrometry analysis was carried out with a J EOL J MS-
700. HPLC analyses were carried out using the following
apparatuses: Shimadzu LC-10AT (liquid chromatograph),
Shimadzu SPD-10A (UV detector), and Shimadzu C-R6A or
C-R8A Chromatopac. Elemental analyses were performed on
a Yanaco CHN Corder MT-6. Column chromatography was
conducted on silica gel 60 (Merck), and preparative thin-layer
chromatography was carried out using Wakogel B-5F. All
reactions were carried out under an argon atmosphere in dried
glassware unless otherwise indicated. All reaction solvents
were dried or distilled by standard procedures.
(2R)-4-Oxo-4-p h en yl-2-tr id eca n oyla m in obu tyr ic a cid
29
eth yl ester (5c): [R]
-53.1 (92% ee, c 0.960, CHCl3); mp
D
1
51 °C; H NMR (CDCl3) δ 0.87 (t, 3H, J ) 6.8 Hz), 1.15-1.35
(m, 21H), 1.55-1.70 (m, 2H), 2.19 (dt, 2H, J ) 3.2, 7.6 Hz),
3.58 (dd, 1H, J ) 4.0, 18.3 Hz), 3.74 (dd, 1H, J ) 4.2, 18.3
Hz), 4.20 (q, 2H, J ) 7.1 Hz), 4.95 (dt, 1H, J ) 8.0, 3.9 Hz),
6.60 (brd, 1H, J ) 8.0 Hz), 7.46 (apparent t, 2H, J ) 7.8 Hz),
7.58 (apparent t, 1H, J ) 7.4 Hz), 7.93 (apparent d, 2H, J )
7.3 Hz); 13C NMR (CDCl3) δ 14.0, 14.1, 22.6, 25.5, 29.1, 29.3,
29.4, 29.5, 29.6, 31.9, 36.5, 40.5, 48.2, 61.7, 128.1, 128.7, 133.7,
136.0, 171.2, 172.9, 198.0; IR (KBr) 3326, 2917, 2849, 1731,
P r ep a r a tion of r-Im in o Ester s (2a -d ). To a suspension
of R-hydroxyglycine ethyl ester (13.3 mmol) in CH2Cl2 (40 mL)
was added thionyl chloride (9.7 mL, 133 mmol). The reaction
mixture was refluxed for 1 h and cooled to room temperature
(rt). The solvent was removed in vacuo to give the almost pure
R-chloro glycine ester in quantitative yield. Further purifica-
tion was done by washing the white precipitate with petroleum
ether under argon atmosphere three times. To a mixture of
this product (0.45 mmol) and 3.5 mmol/g piperidinometh-
ylpolystyrene (257 mg, 0.90 mmol) was added CH2Cl2 (4.5 mL).
The reaction mixture was stirred at room temperature for 10
min, and then the stirring was stopped. The polymer floated
on the surface, and the clear part of the solution (2.5 mL, 0.25
mmol of imine 2a -d ) was taken with a gastight syringe. This
solution was used immediately in the following Mannich-type
reactions.
1684, 1645, 1544 cm-1; LRMS (EI) m/z 417 [M]+. Anal. (C25H39
-
NO4) C, H, N. Chiral HPLC: Daicel Chiralpak AD, hexane/
iPrOH ) 19/1, flow rate ) 1.0 mL/min; tR ) 24.3 min (R), tR
20.0 min (S).
)
(2R)-4-Oxo-4-ph en yl-2-tetr adecan oylam in obu tyr ic acid
28
eth yl ester (5d ): [R]
-52.3 (92% ee, c 0.995, CHCl3); mp
D
1
54-55 °C; H NMR (CDCl3) δ 0.88 (t, 3H, J ) 6.8 Hz), 1.15-
1.35 (m, 23H), 1.55-1.70 (m, 2H), 2.20 (dt, 2H, J ) 3.4, 7.6
Hz), 3.59 (dd, 1H, J ) 3.9, 18.3 Hz), 3.75 (dd, 1H, J ) 4.2,
18.3 Hz), 4.21 (q, 2H, J ) 7.2 Hz), 4.96 (dt, 1H, J ) 8.1, 3.9

3692 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Nakamura et al.
Hz), 6.60 (brd, 1H, J ) 8.1 Hz), 7.48 (apparent t, 2H, J ) 7.7
Hz), 7.60 (apparent t, 1H, J ) 7.4 Hz), 7.94 (apparent d, 2H,
J ) 7.3 Hz); 13C NMR (CDCl3) δ 14.0, 14.1, 22.6, 25.5, 29.1,
29.3, 29.3, 29.4, 29.6, 29.6, 31.9, 36.5, 40.5, 48.2, 61.7, 128.1,
128.7, 133.7, 136.0, 171.2, 172.9, 198.0; IR (KBr) 3338, 2920,
2851, 1737, 1683, 1648, 1542 cm-1; LRMS (EI) m/z 432 [M +
H]+. Anal. (C26H41NO4) C, H, N. Chiral HPLC: Daicel Chiral-
J ) 6.8, 8.8, 14.6 Hz), 2.04 (ddd, 1H, J ) 3.4, 5.6, 14.6 Hz),
2.17 (t, 2H, J ) 7.6 Hz), 2.93 (brs, 2H), 3.6-3.7 (m, 2H), 4.0-
4.1 (m, 1H), 4.81 (dd, 1H, J ) 3.4, 9.0 Hz), 6.48 (brd, 1H, J )
5.9 Hz), 7.25-7.4 (m, 5H); 13C NMR (CDCl3) δ 14.1, 22.7, 25.7,
29.3, 29.3, 29.5, 29.6, 29.6, 31.9, 36.8, 40.7, 50.6, 65.8, 72.0,
125.5, 127.7, 128.6, 144.2, 174.3; IR (KBr) 3292, 2919, 2849,
1642, 1551 cm-1. HRMS (FAB): exact mass calcd for C23H39
-
-
pak AD, hexane/iPrOH ) 19/1, flow rate ) 1.0 mL/min; tR
20.1 min (R), tR ) 26.0 min (S).
)
NO3Na [M + Na]+, 400.2828. Found: 400.2832. Anal. (C23H39
NO3) C, H, N. HPLC: Daicel Chiralpak, hexane/iPrOH ) 19/
1, flow rate ) 1.0 mL/min; tR ) 12.4 min (1R,3R), tR ) 16.5
min (1S,3S).
Syn th esis of HP A-11 to -14 fr om 5a -d . To a solution of
5 (0.2 mmol) in THF (3 mL, freshly distilled from ketyl) or
DME (3 mL, freshly distilled from ketyl) was added 1 M
K-Selectride in THF (0.32 mL, 0.32 mmol) at -78 °C (THF)
or -45 °C (DME). The mixture was stirred at the same
temperature for 15 min, and then 1 M LiBEt3H in THF (0.66
mL, 0.66 mmol) was added. The reaction mixture was allowed
to warm to room temperature and further stirred for 1 h. The
reaction was quenched by careful addition of water and 30%
aqueous H2O2. The mixture was extracted with AcOEt, and
the extract was washed with saturated aqueous NaHCO3. The
combined aqueous layer was extracted with AcOEt, and the
combined organic layer was washed with brine and dried over
anhydrous Na2SO4. The solvents were evaporated, and the
residue was purified by chromatography on silica gel to give
HPA analogues (87-99% yield). These HPA analogues were
purified by recrystallization from CH2Cl2/hexane to afford
enantiomerically and diastereomerically pure products.
(1R,3R)-N-(3-Hyd r oxy-1-h yd r oxym eth yl-3-p h en ylp r o-
p yl)tetr a d eca n a m id e (HP A-14, 6d ): [R]19D -36.2 (>99% ee,
1
c 0.36, CHCl3); mp 94-95 °C; H NMR (CDCl3) δ 0.88 (t, 3H,
J ) 6.8 Hz), 1.2-1.4 (m, 20H), 1.55-1.7 (m, 2H), 1.9-2.0 (m,
1H), 2.05 (ddd, 1H, J ) 3.1, 5.0, 14.6 Hz), 2.16 (t, 2H, J ) 7.6
Hz), 2.72 (brs, 2H), 3.6-3.7 (m, 2H), 4.0-4.1 (m, 1H), 4.81 (dd,
1H, J ) 3.2, 8.8 Hz), 6.41 (brd, 1H, J ) 5.6 Hz), 7.25-7.4 (m,
5H); 13C NMR (CDCl3) δ 14.1, 22.7, 25.7, 29.3, 29.3, 29.5, 29.6,
29.6, 31.9, 36.8, 40.7, 50.6, 65.7, 72.0, 125.5, 127.7, 128.6, 144.2,
174.3; IR (KBr) 3294, 2919, 2849, 1643, 1550 cm-1. HRMS
(FAB): exact mass calcd for C24H42NO3 [M + H]+, 392.3165.
Found: 392.3174. Anal. (C24H41NO3) C, H, N. HPLC: Daicel
Chiralpak AD + guard column, hexane/iPrOH ) 19/1, flow rate
) 1.0 mL/min; tR ) 14.5 min (1R,3R), tR ) 18.9 min (1S,2S).
(1R,3R)-N-(3-Hyd r oxy-1-h yd r oxym eth yl-3-p h en ylp r o-
p yl)p en ta d eca n a m id e (HP A-15, 6f): [R]22D -31.5 (>99% ee,
1
(1R,3R)-N-(3-Hyd r oxy-1-h yd r oxym eth yl-3-p h en ylp r o-
c 0.350, CHCl3); mp 96-97.5 °C; H NMR (CDCl3) δ 0.88 (t,
p yl)d eca n a m id e (HP A-10, 6e): [R]19 -34.1 (>99% ee, c
3H, J ) 6.2 Hz), 1.2-1.4 (m, 22H), 1.55-1.65 (m, 2H), 1.85-
2.1 (m, 2H), 2.17 (t, 2H, J ) 7.3 Hz), 2.94 (brs, 2H), 3.6-3.75
(m, 2H), 4.0-4.1 (m, 1H), 4.81 (dd, 1H, J ) 3.5, 8.8 Hz), 6.48
(d, 1H, J ) 6.2 Hz), 7.2-7.4 (m, 5H); 13C NMR (CDCl3) δ 14.1,
22.7, 25.7, 29.3, 29.3, 29.5, 29.6, 31.9, 36.8, 40.7, 50.6, 65.7,
72.0, 125.5, 127.7, 128.6, 144.2, 174.4; IR (neat) 3292, 2920,
2850, 1643, 1550 cm-1. HRMS (FAB): exact mass calcd for
D
0.62, CHCl3); mp 82-83 °C; 1H NMR (CDCl3) δ 0.88 (t, 3H, J
) 6.8 Hz), 1.2-1.4 (m, 12H), 1.55-1.7 (m, 2H), 1.93 (ddd, 1H,
J ) 6.8, 8.8, 14.6 Hz), 2.05 (ddd, 1H, J ) 3.4, 5.6, 14.6 Hz),
2.16 (t, 2H, J ) 7.7 Hz), 3.21 (brs, 1H), 3.60-3.70 (m, 2H),
3.79 (brs, 1H), 4.0-4.1 (m, 1H), 4.81 (dd, 1H, J ) 3.4, 8.9 Hz),
6.38 (brd, 1H, J ) 6.3 Hz), 7.25-7.4 (m, 5H); 13C NMR (CDCl3)
δ 14.1, 22.6, 25.7, 29.3, 29.3, 29.5, 31.8, 36.8, 40.7, 50.6, 65.8,
72.0, 125.5, 127.7, 128.6, 144.2, 174.3; IR (KBr) 3292, 2920,
2850, 1643, 1551 cm-1. HRMS (FAB): exact mass calcd for
C
25H44NO3 [M + H]+, 406.3321. Found: 406.3322. Anal.
(C25H43NO3) C, H, N.
(1R,3R)-N-(3-Hyd r oxy-1-h yd r oxym eth yl-3-p h en ylp r o-
p yl)h exa d eca n a m id e (HP A-16, 6g): [R]19D -33.0 (>99% ee,
c 0.57, CHCl3); mp 99-100 °C; 1H NMR (CDCl3) δ 0.88 (t, 3H,
J ) 6.6 Hz), 1.2-1.4 (m, 24H), 1.55-1.7 (m, 2H), 1.94 (ddd,
1H, J ) 6.9, 8.8, 14.9 Hz), 2.06 (ddd, 1H, J ) 3.4, 5.1, 14.9
Hz), 2.17 (t, 2H, J ) 7.6 Hz), 2.96 (brs, 1H), 3.55-3.8 (brs,
1H), 3.6-3.75 (m, 2H), 4.0-4.1 (m, 1H), 4.83 (dd, 1H, J ) 3.4,
8.8 Hz), 6.36 (brd, 1H, J ) 6.3 Hz), 7.25-7.4 (m, 5H); 13C NMR
(CDCl3) δ 14.1, 22.7, 25.7, 29.3, 29.4, 29.5, 29.7, 29.7, 29.7,
31.9, 36.9, 40.6, 50.7, 66.0, 72.2, 125.6, 127.8, 128.7, 144.2,
174.3; IR (KBr) 3294, 2919, 2849, 1643, 1550 cm-1. HRMS
(FAB): exact mass calcd for C26H46NO3 [M + H]+, 420.3477.
Found: 420.3473. Anal. (C26H45NO3) C, H, N.
C
20H34NO3 [M + H]+, 336.2538. Found: 335.2538. Anal.
(C20H33NO3) C, H, N.
(1R,3R)-N-(3-Hyd r oxy-1-h yd r oxym eth yl-3-p h en ylp r o-
p yl)u n d eca n a m id e (HP A-11, 6a ): [R]19 -37.6 (>99% ee, c
D
0.61, CHCl3); mp 85-86 °C; 1H NMR (CDCl3) δ 0.88 (t, 3H, J
) 6.8 Hz), 1.2-1.4 (m, 14H), 1.55-1.7 (m, 2H), 1.93 (ddd, 1H,
J ) 6.8, 8.8, 14.6 Hz), 2.04 (ddd, 1H, J ) 3.4, 5.6, 14.6 Hz),
2.16 (t, 2H, J ) 7.7 Hz), 3.00 (brs, 2H), 3.60-3.70 (m, 2H),
4.00-4.10 (m, 1H), 4.80 (dd, 1H, J ) 3.4, 9.0 Hz), 6.44 (brd,
1H, J ) 6.4 Hz), 7.25-7.4 (m, 5H); 13C NMR (CDCl3) δ 14.1,
22.7, 25.7, 29.3, 29.3, 29.5, 29.6, 31.9, 36.8, 40.7, 50.6, 65.7,
72.0, 125.5, 127.7, 128.6, 144.2, 174.4; IR (KBr) 3292, 2921,
2851, 1641, 1552 cm-1. HRMS (FAB): exact mass calcd for
C21H35NO3Na [M + Na]+, 372.2514. Found: 372.2531. Anal.
(C21H35NO3) C, H, N. HPLC: Daicel Chiralpak, hexane/iPrOH
(1R,3R)-N-(3-Hyd r oxy-1-h yd r oxym eth yl-3-p h en ylp r o-
p yl)h ep ta d eca n a m id e (HP A-17, 6h ): [R]19 -33.2 (>99%
D
1
ee, c 0.205, CHCl3); mp 101-102 °C; H NMR (CDCl3) δ 0.88
) 19/1, flow rate ) 1.0 mL/min; tR ) 13.4 min (1R,3R), tR
17.9 min (1S,3S).
)
(t, 3H, J ) 6.7 Hz), 1.2-1.4 (m, 26H), 1.55-1.7 (m, 2H), 1.95
(ddd, 1H, J ) 7.6, 8.9, 14.6 Hz), 2.07 (ddd, 1H, J ) 3.4, 5.1,
14.6 Hz), 2.18 (t, 2H, J ) 7.6 Hz), 2.81 (brs, 1H), 3.5-3.8 (brs,
1H), 3.65-3.75 (m, 2H), 4.0-4.1 (m, 1H), 4.84 (dd, 1H, J )
(1R,3R)-N-(3-Hyd r oxy-1-h yd r oxym eth yl-3-p h en ylp r o-
p yl)d od eca n a m id e (HP A-12, 1 ) 6b): [R]15 -36.3 (>99%
D
ee, c 0.505, CHCl3); mp 90-91 °C; 1H NMR (CDCl3) δ 0.88 (t,
3H, J ) 6.8 Hz), 1.20-1.40 (m, 16H), 1.55-1.7 (m, 2H), 1.93
(ddd, 1H, J ) 6.8, 8.8, 14.8 Hz), 2.03 (ddd, 1H, J ) 3.4, 5.4,
14.8 Hz), 2.16 (t, 2H, J ) 7.7 Hz), 3.21 (brs, 1H), 3.65-3.7 (m,
1H), 3.87 (brs, 1H), 4.0-4.1 (m, 1H), 4.81 (dd, 1H, J ) 3.4, 8.3
Hz), 6.41 (brd, 1H, J ) 6.5 Hz), 7.25-7.4 (m, 5H); 13C NMR
(CDCl3) δ 14.1, 22.7, 25.7, 29.3, 29.3, 29.3, 29.5, 29.6, 29.6,
31.9, 36.8, 40.7, 50.4, 65.5, 71.8, 125.5, 127.7, 128.5, 144.2,
174.3; IR (neat) 3293, 2919, 2849, 1643, 1551 cm-1. HRMS
(EI): exact mass calcd for C22H37NO3 [M]+, 363.2273. Found:
363.2279. Anal. (C22H37NO3) C, H, N. HPLC: Daicel Chiralpak
AD, hexane/iPrOH ) 19/1, flow rate ) 1.0 mL/min; tR ) 11.4
min (1R,3R), tR ) 15.0 min (1S,3S).
3.4, 8.9 Hz), 6.35 (brd, 1H, J ) 6.1 Hz), 7.25-7.4 (m, 5H); 13
C
NMR (CDCl3) δ 14.1, 22.7, 25.7, 29.3, 29.3, 29.4, 29.5, 29.6,
29.6, 29.7, 31.9, 36.8, 40.6, 50.7, 65.9, 72.1, 125.5, 127.8, 128.6,
144.2, 174.3; IR (KBr) 3294, 2919, 2849, 1643, 1550 cm-1
.
HRMS (FAB): exact mass calcd for C27H48NO3 [M + H]+,
434.3634. Found: 434.3631.
(1R,3R)-N-(3-Hyd r oxy-1-h yd r oxym eth yl-3-p h en ylp r o-
p yl)octa d eca n a m id e (HP A-18, 6i): [R]19 -34.0 (>99% ee,
D
c 0.21, CHCl3); mp 101-103 °C; 1H NMR (CDCl3) δ 0.88 (t,
3H, J ) 6.8 Hz), 1.2-1.4 (m, 28H), 1.55-1.7 (m, 2H), 1.95 (ddd,
1H, J ) 7.1, 9.0, 14.8 Hz), 2.07 (ddd, 1H, J ) 3.4, 5.0, 14.8
Hz), 2.17 (t, 2H, J ) 7.6 Hz), 2.82 (brs, 1H), 3.58 (brs, 1H),
3.65-3.75 (m, 2H), 4.0-4.1 (m, 1H), 4.83 (dd, 1H, J ) 3.4, 9.0
Hz), 6.33 (brd, 1H, J ) 5.8 Hz), 7.25-7.4 (m, 5H); 13C NMR
(CDCl3) δ 14.1, 22.7, 25.7, 29.3, 29.4, 29.5, 29.6, 29.7, 29.7,
31.9, 36.9, 40.6, 50.8, 66.0, 72.3, 125.5, 127.9, 128.7, 144.2,
174.3; IR (KBr) 3294, 2919, 2849, 1643, 1550 cm-1. HRMS
(1R,3R)-N-(3-Hyd r oxy-1-h yd r oxym eth yl-3-p h en ylp r o-
p yl)tr id eca n a m id e (HP A-13, 6c): [R]19 -34.1 (>99% ee, c
D
0.54, CHCl3); mp 91-92 °C; 1H NMR (CDCl3) δ 0.88 (t, 3H, J
) 6.8 Hz), 1.2-1.4 (m, 18H), 1.55-1.7 (m, 2H), 1.93 (ddd, 1H,

Ceramide Trafficking Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3693
(FAB): exact mass calcd for C28H50NO3 [M + H]+, 448.3790.
(CDCl3) δ 14.1, 22.6, 25.7, 29.3, 29.3, 29.4, 29.5, 29.6, 31.9,
37.0, 39.9, 47.0, 56.4, 66.3, 81.6, 126.6, 127.9, 128.4, 128.6,
129.6, 129.6, 133.1, 141.3, 166.5, 172.9.
Found: 448.3789. Anal. (C28H50NO3) C, H, N.
Meth yla tion of HP A-12. To a mixture of HPΑ-12 (36.4 mg,
0.1 mmol) in iodomethane (1 mL) was added Ag2O (70 mg, 0.3
mmol). The reaction mixture was stirred at room temperature
for 12 h. The resultant mixture was filtered through a Celite
pad, and the filtrate was concentrated in vacuo. The crude
product obained was purified by chromatography on silica gel
to give the desired 8 (20.1 mg) and 9 (9.0 mg) in 53% and 23%
yield, respectively. The starting material, HPA-12, was also
recovered in 13% yield.
To a solution of benzoic acid 2-dodecanoylamino-4-methoxy-
4-phenylbutyl ester (24.6 mg, 0.051 mmol) in ethanol was
added sodium ethoxide (3.5 mg, 0.051 mmol) at room temper-
ature. The reaction mixture was stirred at room temperature
for 10 h, and then the reaction was quenched with saturated
ammonium chloride solution. The resulting mixture was
extracted with ethyl acetate three times, and the combined
organic layer was washed with brine and dried over Na2SO4.
The crude solution was concentrated in vacuo and purified by
chromatography on silica gel to afford (1R,3R)- (3-methoxy-
(1R,3R)-(3-H yd r oxy-1-m e t h oxym e t h yl-3-p h e n ylp r o-
p yl)d od eca n a m id e (8): [R]19 -38.2 (>99% ee, c 0.78,
D
CHCl3); mp 40-41 °C; 1H NMR (CDCl3) δ 0.88 (t, 3H, J ) 6.8
Hz), 1.2-1.4 (m, 16H), 1.5-1.65 (m, 2H), 1.95-2.05 (m, 2H),
2.10 (dt, 2H, J ) 2.0, 7.3 Hz), 3.36 (s, 3H), 3.42 (dd, 1H, J )
4.1, 9.5 Hz), 3.48 (dd, 1H, J ) 3.7, 9.5 Hz), 3.62 (brs, 1H), 4.15-
4.25 (m, 1H), 4.77 (dd, 1H, J ) 4.8, 7.7 Hz), 5.95 (brd, 1H, J
) 7.8 Hz), 7.2-7.4 (m, 5H); 13C NMR (CDCl3) δ 14.1, 22.6, 25.6,
29.2, 29.3, 29.3, 29.5, 29.6, 31.9, 36.8, 41.8, 47.1, 59.1, 71.8,
74.4, 125.6, 127.3, 128.3, 144.6, 173.3; IR (KBr) 3315, 2918,
2848, 1641, 1542 cm-1. HRMS (FAB): exact mass calcd for
1-hydroxymethyl-3-phenyl-propyl)dodecanamide (10) (18.6 mg)
1
in 96% yield: [R]19 -38.6 (>99% ee, c 0.68, CHCl3); H NMR
D
(CDCl3) δ 0.88 (t, 3H, J ) 6.8 Hz), 1.2-1.4 (m, 16H), 1.6-1.7
(m, 2H), 1.9-2.0 (m, 2H), 2.22 (t, 2H, J ) 7.7 Hz), 3.20 (s,
3H), 3.55-3.65 (m, 2H), 3.95-4.1 (m, 2H), 4.24 (dd, 1H, J )
5.4, 7.1 Hz), 6.49 (brd, 1H, J ) 5.6 Hz), 7.25-7.4 (m, 5H); 13
C
NMR (CDCl3) δ 14.1, 22.7, 25.8, 29.3, 29.4, 29.5, 29.6, 31.9,
36.9, 39.6, 51.8, 56.4, 66.7, 82.1, 126.4, 128.0, 128.7, 141.2,
174.4; IR (neat) 3292, 2925, 2854, 1647, 1542 cm-1. HRMS
(FAB): exact mass calcd for C23H40NO3 [M + H]+, 378.3008.
Found: 378.3023. Anal. (C23H39NO3) C, H, N.
C
23H40NO3 [M + H]+, 378.3008. Found: 378.3010. Anal.
(C23H39NO3) C, H, N.
(2R ,3R )-3-B e n zy lo x y -2-d o d e c a n o y la m in o -4-o x o -4-
p h en ylbu tyr ic Acid Eth yl Ester (12). Compound 12 was
prepared from 2b and 11 by the Mannich-type reaction using
the Cu(II)-chiral diamine 4 as mentioned above: [R]22D +12.2
(1R,3R)-(3-Me t h oxy-1-m e t h oxym e t h yl-3-p h e n ylp r o-
p yl)d od eca n a m id e (9): [R]19 -37.3 (>99% ee, c 0.41,
D
1
CHCl3); H NMR (CDCl3) δ 0.88 (t, 3H, J ) 6.8 Hz), 1.2-1.4
(m, 16H), 1.55-1.7 (m, 2H), 1.88 (ddd, 1H, J ) 5.2, 5.2, 14.4
Hz), 2.04 (ddd, 1H, J ) 8.3, 8.3, 14.4 Hz), 2.16 (t, 2H, J ) 7.7
Hz), 3.16 (s, 3H), 3.30 (s, 3H), 3.33 (dd, 1H, J ) 4.4, 9.5 Hz),
3.38 (dd, 1H, J ) 3.7, 9.5 Hz), 4.1-4.2 (m, 1H), 4.18 (dd, 1H,
J ) 5.1, 7.6 Hz), 5.87 (brd, 1H, J ) 8.1 Hz), 7.25-7.4 (m, 5H);
13C NMR (CDCl3) δ 14.1, 22.6, 25.7, 29.3, 29.3, 29.5, 29.6, 29.6,
31.9, 37.0, 40.0, 47.3, 56.4, 59.0, 74.2, 82.0, 126.7, 127.7, 128.5,
1
(90% ee, c 1.58, CHCl3); H NMR (CDCl3) δ 0.88 (t, 3H, J )
6.6 Hz), 1.15-1.35 (m, 19H), 1.50-1.60 (m, 2H), 2.15 (t, 2H, J
) 7.3 Hz), 4.0-4.3 (m, 2H), 4.37 (d, 1H, J ) 12.0 Hz), 4.81 (d,
1H, J ) 12.0 H), 5.10 (dd, 1H, J ) 2.0, 9.0 Hz), 5.43 (d, 1H, J
) 2.0 Hz), 6.23 (d, 1H, J ) 8.8 Hz), 7.2-7.4 (m, 5H), 7.47 (t,
2H, J ) 7.8 Hz), 7.59 (t, 1H, J ) 7.3 Hz), 7.90 (d, 2H, J ) 7.3
Hz); 13C NMR (CDCl3) δ 14.0, 14.1, 22.6, 25.4, 29.0, 29.2, 29.3,
29.4, 29.5, 31.8, 36.2, 53.9, 62.0, 72.2, 78.7, 128.1, 128.2, 128.3,
128.4, 128.9, 133.8, 134.7, 136.6, 169.1, 173.1, 195.5; IR (neat)
3365, 2925, 2854, 1751, 1682, 1507 cm-1. HRMS (EI): exact
141.8, 172.7; IR (neat) 3286, 2924, 2854, 1643, 1543 cm-1
.
HRMS (FAB): exact mass calcd for C24H42NO3 [M + H]+,
392.3164. Found: 392.3173. Anal. (C24H41NO3) C, H, N.
(1R ,3R )-(3-Me t h oxy-1-h yd r oxym e t h yl-3-p h e n ylp r o-
p yl)d od eca n a m id e (10). To a solution of HPA-12 (75 mg,
0.206 mmol) and triethylamine (42 mg, 0.416 mmol) in
dichloromethane (2.1 mL) was addedd benzoyl chloride (29 mg,
0.206 mmol) at 0 °C. The reaction mixture was stirred at 0 °C
for 12 h, and then the reaction was quenched by NaHCO3. The
resultant mixture was extracted with dichloromethane three
times, and the combined organic layer was washed with brine
and dried over Na2SO4. The crude solution was concentrated
in vacuo and purified by chromatography on silica gel to afford
benzoic acid (2R,4R)-2-dodecanoylamino-4-hydroxy-4-phenyl-
butyl ester (87.3 mg) in 91% yield: mp 71-72 °C; 1H NMR
(CDCl3) δ 0.88 (t, 3H, J ) 6.8 Hz), 1.2-1.4 (m, 16H), 1.5-1.65
(m, 2H), 2.05 (apparent t, 2H, J ) 6.0 Hz), 2.11 (t, 2H, J ) 7.4
Hz), 3.10 (brs, 1H), 4.3-4.35 (m, 1H), 4.4-4.55 (m, 2H), 4.86
(apparent t, 1H, J ) 6.3 Hz), 6.00 (brd, 1H, J ) 5.1 Hz), 7.25-
7.4 (m, 5H), 7.44 (apparent t, 2H, J ) 7.7 Hz), 7.58 (apparent
t, 1H, J ) 7.4 Hz), 8.01 (apparent d, 2H, J ) 7.3 Hz); 13C NMR
(CDCl3) δ 14.1, 22.7, 25.6, 29.2, 29.3, 29.4, 29.6, 29.6, 31.9,
36.9, 41.1, 47.3, 66.4, 72.1, 125.7, 127.7, 128.5, 128.6, 129.6,
129.7, 133.3, 144.2, 166.8, 173.6.
To a mixture of benzoic acid 2-dodecanoylamino-4-hydroxy-
4-phenylbutyl ester (70.0 mg, 0.15 mmol) in iodomethane (5
mL) was added Ag2O (348 mg, 1.5 mmol) at room temperature.
The reaction mixture was stirred at room temperature for 2
days. The resultant mixture was filtered through a Celite pad,
and the filtrate was concentrated in vacuo. The crude product
obtained was purified by chromatography on silica gel to give
the desired benzoic acid (2R,4R)-2-dodecanoylamino-4-meth-
oxy-4-phenylbutyl ester (49.2 mg) in 68% yield: 1H NMR
(CDCl3) δ 0.88 (t, 3H, J ) 6.8 Hz), 1.2-1.4 (m, 16H), 1.55-1.7
(m, 2H), 1.94 (ddd, 1H, J ) 4.9, 4.9, 14.4 Hz), 2.08 (ddd, 1H,
J ) 7.9, 7.9, 14.4 Hz), 2.17 (dt, 2H, J ) 2.7, 7.7 Hz), 3.18 (s,
3H), 4.24 (dd, 2H, J ) 4.9, 7.9 Hz), 4.30 (dd, 2H, J ) 4.0, 10.9
Hz), 4.35-4.5 (m, 2H), 5.92 (brd, 1H, J ) 7.6 Hz), 7.25-7.4
(m, 5H), 7.44 (apparent t, 2H, J ) 7.7 Hz), 7.58 (apparent t,
1H, J ) 7.4 Hz), 8.01 (apparent d, 2H, J ) 7.3 Hz); 13C NMR
mass calcd for
C
24H36NO5 [M - Bn]+, 418.2593. Found:
418.2592. HPLC: Daicel Chiralpak AD + AD, hexane/iPrOH
) 19/1, flow rate ) 0.95 mL/min; tR ) 34.0 min (2R,3R), tR
36.2 min (2S,3S).
)
(1S,2R,3R)-(2,3-Dih yd r oxy-1-h yd r oxym eth yl-3-p h en yl-
p r op yl)d od eca n a m id e ((1S,2R,3R)-14). To a solution of 12
(95.3 mg, 0.187 mmol) in THF (2.8 mL) was added 1 M
LiBEt3H (0.93 mL, 0.93 mmol) at 0 °C. The mixture was
allowed to warm to room temperature and stirred for 1.5 h.
The reaction was quenched by careful addition of water and
30% aqueous H2O2. The mixture was extracted with AcOEt,
and the extract was washed with saturated aqueous NaHCO3.
The combined aqueous layer was extracted with AcOEt, and
the combined organic layer was washed with brine and dried
over anhydrous Na2SO4. The solvents were evaporated, and
the residue was purified by chromatography on silica gel to
give 13. To a solution of 13 (43.2 mg, 0.092 mmol) in EtOH
(1.8 mL) was added 20% Pd(OH)2/C (43.2 mg). After replace-
ment of argon by hydrogen, the mixture was stirred for 20 min
at room temperature. Pd(OH)2/C was filtered off, and the
filtrate was concentrated in vacuo. The crude product was
purified by chromatography on silica gel to afford (1S,2R,3R)-
14 (35% yield), (1S,2R,3S)-14 (3% yield), and 15 (35% yield).
The title compound, (1S,2R,3R)-14, could be also prepared
from L-serine-derived diol ((1S,2R,3R)-16)16 by the same
method as the procedure described for (1S,2S,3S)-14 (see
below): [R]26 -5.2 (>99% ee, c 0.615, CHCl3); mp 54-55 °C;
D
1H NMR (CD3OD) δ 0.90 (t, 3H, J ) 6.4 Hz), 1.2-1.4 (m, 16H),
1.55-1.65 (m, 2H), 2.20-2.30 (m, 2H), 3.20-3.40 (m, 1H),
3.40-3.60 (m, 2H), 3.92 (dd, 1H, J ) 1.8, 7.7 Hz), 4.51 (d, 1H,
J ) 7.9 Hz), 7.2-7.5 (m, 5H); 13C NMR (CD3OD) δ 15.3, 24.6,
27.9, 31.2, 31.3, 31.4, 31.5, 31.6, 33.9, 38.0, 53.6, 63.7, 75.4,
76.8, 129.2, 129.6, 130.1, 143.6, 177.0; IR (neat) 3735, 2925,
2854, 1646, 1541, 1456, 1199, 1086, 1086, 1051, 843, 763, 701
cm-1; LRMS (EI) m/z 380 [M + H]+. Anal. (C22H37NO4) C, H,
N.

3694 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Nakamura et al.
(1S,2S,3S)-(2,3-Dih yd r oxy-1-h yd r oxym eth yl-3-p h en yl-
p r op yl)d od eca n a m id e ((1S,2S,3S)-14). Compound 16 was
prepared according to the literature procedure.16 To a solution
of 16 (109.0 mg, 0.32 mmol) in MeOH (1.5 mL) was added 4 N
HCl-AcOEt (1.5 mL). The reaction mixture was refluxed for
1 h and cooled to room temperature. The mixture was
concentrated in vacuo to afford crude product 3-amino-1-
phenylbutane-1,2,4-triol hydrochloride. To this flask was added
THF (5.0 mL), Et3N (65.4 mg, 0.64 mmol), and dodecanoic acid
p-nitrophenol ester (207.6 mg, 0.64 mmol), successively. The
reaction mixture was stirred at room temperature for 8 h, and
the reaction was quenched by adding saturated aqueous
NaHCO3 and 1 N aqueous NaOH. The mixture was extracted
with CH2Cl2 five times, and the organic layer was washed with
brine and dried over anhydrous Na2SO4. The solvent was
evaporated and the residue was purified by chromatography
Biologica l Assa ys. The effect of HPA-12 and its derivatives
on SM biosynthesis was examined by metabolic labeling with
radioactive serine in CHO-K1 cells as described previously.10
In brief, the subconfluent CHO cell monolayers in 60 mm
dishes were preincubated in 1.5 mL of serum-free medium
containing HPA-12 or its derivatives at various concentrations
up to 1 µM at 4 °C for 15 min. The concentration of DMSO,
the vehicle of the drug stock solutions, during the preincuba-
tion was adjusted to 0.01% (v/v) in each dish. After addition
of L-[14C]serine (0.75 µCi), the cells were incubated at 33 °C
for 2 h. As the drug-minus control, cells were incubated with
0.01% (v/v) DMSO in the absence of the HPA derivatives.
Then, lipids were extracted from the cells and separated by
TLC. Radioactive lipids separated on the plates were analyzed
with a BAS 2000 image analyzer (Fuji Film, Inc., Tokyo). The
radioactivity of metabolically labeled SM was corrected to the
protein amount of the cell lysate used for lipid extraction. The
corrected radioactivity was used as a measure of de novo SM
synthesis in cells. The level of SM synthesis in the drug-minus
control was taken as 100%.
on silica gel to afford (1S,2S,3S)-14 (77.4 mg, 63% yield in two
21
steps): [R]
+16.3 (>99% ee, c 1.005, CHCl3); mp 122-123
D
1
°C; H NMR (CD3 OD) δ 0.93 (t, 3H, J ) 6.6 Hz), 1.30-1.50
(m, 16H), 1.60-1.70 (m, 2H), 2.27 (t, 2H, J ) 7.5 Hz), 3.75
(dd, 1H, J ) 2.9, 7.9 Hz), 3.81 (d, 2H, J ) 4.9 Hz), 4.03 (dd,
1H, J ) 4.9, 7.9 Hz), 4.73 (d, 1H, J ) 2.9 Hz), 7.20-7.50 (m,
5H); 13C NMR (CD3 OD) δ 15.3, 24.6, 27.9, 31.2, 31.3, 31.4,
31.5, 31.6, 33.9, 38.0, 55.4, 63.1, 74.9, 76.5, 129.4, 128.6, 129.0,
129.9, 144.4, 177.7; IR (neat) 3303, 2920, 2850, 1631, 1541
cm-1. HRMS (FAB): exact mass calcd for C22H38NO4 [M + H]+,
380.2801. Found: 380.2795. Anal. (C22H37NO4) C, H, N
Ack n ow led gm en t. This work was partially sup-
ported by CREST and SORST, J apan Corporation of
Science and Technology, the Ministry of Education,
Culture, Sports, Science and Technology of J apan,
J apan Society of the Promotion of Science (J SPS), and
Terumo Life Science Foundation.
(2,3-Dih yd r oxy-1-h yd r oxym eth yl-3-p h en ylp r op yl)ca r -
ba m ic Acid ter t-Bu tyl Ester (18 a n d 19). Compound 17 was
prepared according to the literature procedure.16 To a solution
of 17 (70.9 mg, 0.27 mmol) and 50% aqueous solution of NMO
(81.0 mg, 0.35 mmol) in H2O/acetone/CH3CN (1:1:1, 2.7 mL)
was added microencapsulated OsO4 (5 mol %).17 The reaction
mixture was stirred at room temperature for 36 h. After the
reaction was completed, methanol was added to the mixture,
and the catalyst was separated by filtration and washed with
methanol. The filtrate was concentrated in vacuo, and satu-
rated aqueous Na2SO3 was added to the residue. The mixture
was extracted with CH2Cl2 four times, and the combined
organic layer was washed with brine and dried over anhydrous
Na2SO4. The solvent was evaporated, and the residue was
purified by chromatography on silica gel to afford 18 (50.6 mg,
63% yield) and 19 (23.3 mg, 29% yield). 1H and 13C NMR
spectra of 18 and 19 were identical with the reported ones.16
Refer en ces
(1) Reviews: (a) Kolter, T.; Sandhoff, K. SphingolipidssTheir
Metabolic Pathways and the Pathobiochemistry of Neurodegen-
erative Diseases. Angew. Chem., Int. Ed. 1999, 38, 1532-1568.
(b) Dickson, R. C. Sphingolipid Functions in Saccharomyces
Cerevisiae: Comparison to Mammals. Annu. Rev. Biochem.
1998, 67, 27-48.
(2) Hanada, K.; Nishijima, M.; Kiso, M.; Hasegawa, A.; Fujita, S.;
Ogawa, T.; Akamatsu, Y. Sphingolipids Are Essential for the
Growth of Chinese Hamster Ovary Cells. J . Biol. Chem. 1992,
267, 23527-23533.
(3) Spiegel, S.; Merrill, A. H., J r. Sphingolipids Metabolism and Cell
Growth Regulation. FASEB J . 1996, 10, 1388-1397.
(4) Hannun, Y. A. Function of Ceramide in Coordinating Cellular
Responses to Stress. Science 1996, 274, 1855-1859.
(5) (a) Kobayashi, S.; Furuta, T.; Hayashi, T.; Nishijima, M.;
Hanada, K. Catalytic Asymmetric Syntheses of Antifungal
Sphingofungins and Their Biological Activity as Potent Inhibi-
tors of Serine Palmitoyltransferase (SPT). J . Am. Chem. Soc.
1998, 120, 908-919. (b) Kobayashi, S.; Furuta, T. Use of
Heterocycles as Chiral Ligands and Auxiliaries in Asymmetric
Syntheses of Sphingosine, Shingofungins B and F. Tetrahedron
1998, 54, 10275-10294.
(1S,2R,3S)-(2,3-Dih yd r oxy-1-h yd r oxym eth yl-3-p h en yl-
p r op yl)d od eca n a m id e ((1S,2R,3S)-14). Preparation of
(1S,2R,3S)-14 from 18 was accomplished by the same method
as described for the synthesis of (1S,2S,3S)-14 from 16. The
(6) (a) Wakabayashi, T.; Mori, K.; Kobayashi, S. Total Synthesis
and Structural Elucidation of Khafrefungin. J . Am. Chem. Soc.
2001, 123, 1372-1375. (b) Kobayashi, S.; Mori, K.; Wakabayashi,
T.; Yasuda, S.; Hanada, K. Convergent Total Synthesis of
Khafrefungin and Its Inhibitory Activity of Fungal Sphingolipid
Syntheses. J . Org. Chem. 2001, 66, 5580-5584. (c) Mori, Y.;
Nakamura, M.; Wakabayashi, T.; Mori, K.; Kobayashi, S. Ef-
ficient Total Synthesis of Khafrefungin: Convergent Approach
Using Suzuki Coupling under Thallium-Free Conditions toward
Multigram-Scale Synthesis. Synlett 2002, 601-603.
yield was 74% in two steps: [R]26 -32.5 (>99% ee, c 1.145,
D
1
CHCl3); mp 53-54 °C; H NMR (CD3 OD) δ 0.93 (t, 3H, J )
6.2 Hz), 1.30-1.45 (m, 16H), 1.60-1.75 (m, 2H), 2.30 (t, 2H, J
) 7.1 Hz), 3.60-3.75 (m, 2H), 3.87 (dd, 1H, J ) 1.3, 8.4 Hz),
4.30-4.40 (m, 2H), 7.25-7.45 (m, 5H); 13C NMR (CD3 OD) δ
15.3, 24.6, 27.9, 31.2, 31.4, 31.5, 31.6, 33.9, 37.9, 54.2, 63.8,
75.3, 76.4, 129.4, 129.4, 129.9, 144.4, 178.2; IR (neat) 3335,
2925, 2851, 1635, 1535 cm-1. HRMS (EI): exact mass calcd
(7) For review: Merrill, A. H., J r. De novo sphingolipid biosynthe-
sis: A necessary, but dangerous, pathway. J . Biol. Chem. 2002,
277, 25843-25846.
for C22H37NO4 [M]+, 379.2723. Found: 379.2729. Anal. (C22H37
NO4) C, H, N.
-
(8) Fukasawa, M.; Nishijima, M.; Hanada, K. Genetic evidence for
ATP-dependent endoplasmic reticulum-to-Golgi apparatus traf-
ficking of ceramide for sphingomyelin synthesis in Chinese
hamster ovary cells. J . Cell Biol. 1999, 144, 673-685.
(9) Funakoshi, T.; Yasuda, S.; Fukasawa, M.; Nishijima, M.; Han-
ada, K. Reconstitution of ATP- and cytosol-dependent transport
of de novo synthesized ceramide to the site of sphingomyelin
synthesis in semi-intact cells. J . Biol. Chem. 2000, 275, 29938-
29945.
(1S,2S,3R)-(2,3-Dih yd r oxy-1-h yd r oxym eth yl-3-p h en yl-
p r op yl)d od eca n a m id e ((1S,2S,3R)-14). Preparation of
(1S,2S,3R)-14 from 19 was accomplished by the same method
as described for the synthesis of (1S,2S,3S)-14 from 16. The
yield was 70% in two steps: [R]22 -27.4 (>99% ee, c 0.250,
D
1
CHCl3); mp 82-83 °C; H NMR (CD3 OD) δ 0.93 (t, 3H, J )
6.5 Hz), 1.30-1.50 (m, 16H), 1.60-1.70 (m, 2H), 2.22 (t, 2H, J
) 7.3 Hz), 3.75-3.85 (m, 2H), 3.93 (apparent t, 1H, J ) 5.7
Hz), 4.10 (apparent q, 1H, J ) 5.1 Hz), 4.68 (d, 1H, J ) 6.2
Hz), 7.20-7.50 (m, 5H); 13C NMR (CD3OD) δ 15.3, 24.6, 27.8,
31.2, 31.3, 31.4, 31.5, 31.6, 33.9, 38.1, 54.5, 62.8, 76.8, 77.3,
129.3, 129.9, 144.1, 177.0; IR (neat) 3305, 2920, 2850, 1643,
1543 cm-1. HRMS (FAB): exact mass calcd for C22H38NO4 [M
+ H]+, 380.2801. Found: 380.2787. Anal. (C22H37NO4) C, H,
N.
(10) Yasuda, S.; Kitagawa, H.; Ueno, M.; Ishitani, H.; Fukasawa, M.;
Nishijima, M.; Kobayashi, S.; Hanada, K. A Novel Inhibitor of
Ceramide Trafficking from the Endoplasmic Recticulum to the
Site of Sphingomyelin Synthesis. J . Biol. Chem. 2001, 276,
43994-44002.
(11) Ueno, M.; Kitagawa, M.; Ishitani, H.; Yasuda, S.; Hanada, K.;
Kobayashi, S. Catalytic Enantioselective Synthesis of a Novel
Inhibitor of Ceramide Trafficking, (1R,3R)-N-(3-Hydroxy-1-
hydroxymethyl-3-phenylpropyl)dodecanamide (HPA-12). Tetra-
hedron Lett. 2001, 42, 7863-7865.

Ceramide Trafficking Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3695
(12) (a) Kobayashi, S.; Matsubara, R.; Kitagawa, H. Catalytic,
Asymmetric Mannich-Type Reactions of N-Acylimino Esters for
Direct Formation of N-Acylated Amino Acid Derivatives. Ef-
ficient Synthesis of a Novel Inhibitor of Ceramide Trafficking,
HPA-12. Org. Lett. 2002, 4, 143-146. (b) Kobayashi, S.; Mat-
subara, R.; Nakamura, Y.; Kitagawa, H.; Sugiura, M. Catalytic,
Asymmetric Mannich-Type Reactions of N-Acylimino Esters.
Reactivity, Diastereo- and Enantioselectivity, and Application
to N-Acylated Amino Acid Derivatives. J . Am. Chem. Soc. 2003,
125, 2507.
(14) Bodanszky, M. Synthesis of Peptides by Aminolysis of Nitro-
phenyl Esters. Nature 1955, 175, 685.
(15) In the initial benzoation, monobenzoate was obtained in 91%
yield along with dibenzoate (7%).
(16) Imashiro, R.; Sakurai, O.; Yamashita, T.; Horikawa, H. A Short
and Efficient Synthesis of Phytosphingosines Using Asymmetric
Dihydroxylation. Tetrahedron 1998, 54, 10657-10670.
(17) Nagayama, S.; Endo, M.; Kobayashi, S. Microencapsulated
Osmium Tetraoxide. A New Recoverable and Reusable Polymer-
Supported Osmium Catalyst for Dihydroxylation of Olefins. J .
Org. Chem. 1998, 63, 6094-6095.
(13) Kobayashi, S.; Hamada, T.; Manabe, K. The Catalytic Asym-
metric Mannich-Type Reactions in Aqueous Media. J . Am. Chem.
Soc. 2002, 124, 5640-5641.
J M0300779