Conformationally constrained diacylglycerol (DAG) analogs: 4-C-hydroxyethyl-5-O-acyl-2,3-dideoxy-D-glyceropentono-1,4-lactone analogs as protein kinase C (PKC) ligands

Article abstract of DOI:10.1016/j.ejmech.2003.10.006

The (R)-DAG-lactones (5 and 7E/Z) are conformationally constrained diacylglycerol (DAG) analogs with high potency as protein kinase C (PKC) ligands. Here, we have prepared and characterized their one-carbon lengthened analogs (6 and 8E/Z). The target compounds were synthesized from 1,2-O-isopropylidene D-xylose through a key intermediate, 4-C-hydroxyethyl-2,3- dideoxy-D-glyceropentono-1,4-lactone (13); they were evaluated as competitive ligands to displace bound [³H]phorbol 12,13-dibutyrate (PDBU) from a recombinant single isozyme (PKC-α). The binding affinities of the synthesized compounds were K_i = 2.623 μM for 6, K_i = 1.080 μM for 8Z and K_i = 0.92 μM for 8E, which were ca. 27, 90, and 70 times less potent than the corresponding parent compounds (5, 7Z and 7E). Molecular modeling indicated that the reduced binding affinity of the representative 3-alkylidene lactone 8Z, as compared to 7Z, may be explained by its poor fit in the sn-1 binding mode as well as by its entropic loss due to the relatively flexible hydroxyethyl group.

Full text of DOI:10.1016/j.ejmech.2003.10.006

European Journal of Medicinal Chemistry 39 (2004) 69–77
www.elsevier.com/locate/ejmech
Original article
Conformationally constrained diacylglycerol (DAG) analogs:
4-C-hydroxyethyl-5-O-acyl-2,3-dideoxy-D-glyceropentono-1,4-lactone
analogs as protein kinase C (PKC) ligands
Jeewoo Lee a,*, SuYeon Kim a, Ji-Hye Kang a, Geza Acs b, Peter Acs b,
Peter M. Blumberg b, Victor E. Marquez c,*
a Laboratory of Medicinal Chemistry, Research Institute of Pharmaceutical Sciences, College of Pharmacy, Seoul National University,
Shinlim-Dong, Kwanak-Ku, Seoul 151-742, South Korea
b Laboratory of Cellular Carcinogenesis and Tumor Promotion, Center for Cancer Research, National Cancer Institute, NIH, Bethesda, MD 20892, USA
c Laboratory of Medicinal Chemistry, Center for Cancer Research, National Cancer Institute, NIH, Frederick, MD 21701, USA
Received 12 June 2003; received in revised form 27 October 2003; accepted 29 October 2003
Abstract
The (R)-DAG-lactones (5 and 7E/Z) are conformationally constrained diacylglycerol (DAG) analogs with high potency as protein kinase
C (PKC) ligands. Here, we have prepared and characterized their one-carbon lengthened analogs (6 and 8E/Z). The target compounds were
synthesized from 1,2-O-isopropylidene D-xylose through a key intermediate, 4-C-hydroxyethyl-2,3-dideoxy-D-glyceropentono-1,4-lactone
(13); they were evaluated as competitive ligands to displace bound [3H]phorbol 12,13-dibutyrate (PDBU) from a recombinant single isozyme
(PKC-a). The binding affinities of the synthesized compounds were Ki = 2.623 µM for 6, Ki = 1.080 µM for 8Z and Ki = 0.92 µM for 8E, which
were ca. 27, 90, and 70 times less potent than the corresponding parent compounds (5, 7Z and 7E). Molecular modeling indicated that the
reduced binding affinity of the representative 3-alkylidene lactone 8Z, as compared to 7Z, may be explained by its poor fit in the sn-1 binding
mode as well as by its entropic loss due to the relatively flexible hydroxyethyl group.
© 2003 Elsevier SAS. All rights reserved.
Keywords: Protein kinase C; DAG-lactone; Diacylglycerol; Phorbol ester
1. Introduction
tent PKC ligands has been to use conformationally con-
strained DAG analogs for the purpose of gaining an entropic
advantage [5].
The protein kinase C (PKC) family of phospholipid-
dependent serine/threonine kinases is activated by diacyl-
glycerol (DAG), which is released by the receptor-mediated
hydrolysis of membrane-resident phosphatidylinositol 4,5-
diphosphate (PIP2) or by the hydrolysis of phosphatidyl
choline (PC) [1–3]. DAG stereospecifically binds to the tan-
dem C1a and C1b domains in the regulatory domain of PKC.
The binding is competitive with respect to phorbol ester, a
tumor promoting diterpene and ultrapotent DAG analog [4].
One reason for the lower binding affinity of DAG compared
to phorbol ester is the conformational flexibility of the DAG,
which results in an entropic penalty upon binding to the
enzyme. Therefore, one aspect of our efforts to find ultrapo-
Over the past years, we have demonstrated that (R)-5-
acyloxy-5-hydroxymethyltetrahydro-2-furanone analogs (1
and 2) are conformationally constrained DAG analogs with
potent PKC binding affinities in the low nanomolar range,
stereospecificity, and promising anti-tumor activities toward
various tumor cell lines [6,7]. A pharmacophore-guided
modeling study indicated that the SAR of these DAG-
lactones matched well a three point pharmacophore model of
the phorbol esters, comprising the C3=O, C9–OH and
C20–OH groups of the phorbol ester [8–10]. This binding
model is somewhat different from the X-ray structure of the
complex of the PKC-d C1b domain and phorbol-13-acetate
determined in the absence of phospholipid, in which only the
C3=O, C4–OH and C20–OH were interpreted as interacting
with the enzyme [11]. On the other hand, the former pharma-
cophore model that we have used is consistent with experi-
* Corresponding authors.
E-mail addresses: jeewoo@snu.ac.kr (J. Lee),
marquezv@dc37a.nci.nih.gov (V.E. Marquez).
© 2003 Elsevier SAS. All rights reserved.
doi:10.1016/j.ejmech.2003.10.006

70
J. Lee et al. / European Journal of Medicinal Chemistry 39 (2004) 69–77
R
Template I
R
O
HO
O
O
HO
O
O
O
13
O
O
O
x 0.58
H
9
H
R2
O
O
O
O
OH
4
R1
OH
3
O
O
OH
O
O
OH
H27C13
H27C13
20
(S)-Diacylglycerol
Phorbol Ester
H-bonding acceptor
3 Ki = 1.5 µM
4 Ki = 0.88 µM
H-bonding donor
Template 2
R
R
C13H27
O
O
lipid layer
O
O
O
O
sn-2
O
sn-1
O
O
O
O
HO
HO
n
n
3
O
sn-2
5
C17H33
O
sn-1
O
O
5 n=1
6 n=2
7 n=1
8 n=2
OH
OH
enzyme
O
Fig. 2. The design of target lactones 6 and 8.
1. 5-O-acyl lactone
: favor sn-2 mode
2. 3-alkylidene lactone
: favor sn-1 mode
binding to the protein–lipid interface. This principle has been
useful for predicting the binding modes of DAG-lactones.
For instance, the 3-alkylidene lactone (2) favors the sn-1
binding mode because the sn-2 carbonyl interacts with the
phospholipids. Conversely, the 5-acyl lactones (1) bind to the
enzyme in the sn-2 mode, because the sn-1 position interacts
with the phospholipids [15].
We previously reported that 2-deoxy-3-O-tetradecyl-L-
ribonolactone (3), a conformationally constrained DAG sur-
rogate [16], bound stereospecifically with an affinity of
1.5 µM and its one-carbon lengthened homolog, 2,5-
Fig. 1. sn-1 and sn-2 Binding mode of (R)-DAG-lactones.
mental results that were determined under more physiologi-
cal conditions, which include phospholipid, which makes an
important contribution to the binding affinity. For instance,
Irie’s group reported that 4b-deoxy-PDBu is functionally
equipotent to PDBU [12], confirming the earlier conclusion
from natural product derivatives that the C4 hydroxyl group
is not necessary for PKC binding. Likewise, a role for the
C9–OH/C13–C=O motif of phorbol esters in the binding to
PKC is implied from the observation that both carbonyl
groups of the DAG lactones are critical and one of these
carbonyls corresponds to the C9–OH/C13–C=O motif [13].
The interpretation is that contribution of the second carbonyl
to PKC binding is through the interaction with the phospho-
lipid head groups associated with PKC in the active complex.
Further support for the pharmacophoric importance of the
C13–C=O group (maybe related to the C9–OH group through
hydrogen bonding) comes from phorbol ester analogs lack-
ing this functionality, as described by Sodeoka’s group [14].
In the modeling study comparing (R)-DAG-lactone and
phorbol ester, two hydrogen bonding acceptors in the lactone
template, represented by the two carbonyls of the
5-acyloxymethyl and the lactone, could be overlapped with
the C3=O and C9–OH groups of the phorbol ester in two
different ways (Fig. 1). This resulted in two predicted bind-
ing modes, a sn-1 and a sn-2 mode, in which the sn-1 or sn-2
carbonyls, respectively, in the DAG-lactone matched the
C3=O of the phorbol [5]. Our extensive analysis of these two
binding modes in DAG-lactones suggested a general prin-
ciple determining the binding mode: the carbonyl bearing the
larger lipid chain, mimicking the C9–OH of the phorbol ester,
will not be involved in binding to the enzyme but rather in
dideoxy-3-O-tetradecanoyl-D-galactono-1,4-lactone
(4)
[17], showed a slightly better affinity of 0.88 µM (Fig. 2).
This result was explained by the fact that, since 3 and 4 would
have a preference for the sn-2 mode, the improved fit, in the
sn-2 mode, of 4 to the phorbol ester more than compensated
for the entropic penalty due to the hydroxyethyl group [17].
In the present study, we decided to investigate one-carbon
elongated homologs (6 and 8) of potent (R)-DAG-lactones 5
and 7 as PKC ligands (Fig. 2). One advantage of the elon-
gated homologs is that they would be resistant to the racem-
ization that occurred at the final stage in the synthesis of
chiral 7, in which the chiral purity of 7 was readily destroyed
by facile 5′-O-acyl migration during purification [6]. Here,
we describe the syntheses, binding affinities, and the interac-
tion of chiral (R)-5-acyloxy-5-hydroxyethyltetrahydro-3-
furanone analogs with the enzyme.
2. Chemistry
The synthesis of the 5-acyl lactone analog (6) is outlined
in Fig. 3. Starting from 1,2-O-isopropylidene D-xylose,
spirolactone 9 was prepared in four steps according to a

J. Lee et al. / European Journal of Medicinal Chemistry 39 (2004) 69–77
71
Table 1
PKC-a binding affinities for one-carbon elongated analogues
Ki (µM)
Ki (µM)
C13H27
C13H27
O
O
O
O
O
O
0.096
2.623 + 0.082
O
O
HO
HO
5
6
O
O
O
O
O
O
O
O
0.012
0.013
1.080 + 0.087
0.92 + 0.14
HO
C17H33
HO
C17H33
8Z
7Z
O
O
O
O
O
O
O
O
HO
HO
C
H
17 33
C H
17 33
7E
8E
previous report [18]. The isopropylidene group of 9 was
deprotected under acidic conditions to afford the correspond-
ing diol, which was then cleaved to the aldehyde 10 by
sodium periodate. The ensuing reduction of 10 with sodium
borohydride produced the alcohol 11 concomitant with the
deprotection of the O-formyl group. The primary alcohol of
11 was selectively protected by a tert-butyldimethylsilyl
(TBS) group, and then the secondary alcohol was deoxygen-
ated by Barton’s procedure to produce 13. Deprotection of
the TBS group in 13 and acylation with tetradecanoyl chlo-
ride followed by benzyl deprotection with boron trichloride
produced the final compound 6.
For the synthesis of the 3-alkylidene lactone analogs, the
lactone 13 was initially employed for the condensation with
the branched aldehyde. However, problems in deprotection
of the benzyl group at the final stage forced us to replace it
with the 4-methoxyphenyl protected 17. The synthesis of the
3-alkylidene lactone is represented in Fig. 4. The condensa-
tion of lactone 17 with oleyl aldehyde, followed by elimina-
tion of the corresponding b-hydroxy lactone, afforded a sepa-
rable mixture of geometric Z/E isomers (18Z/18E = 1:1.5).
After separation at this stage, each isomer was hydrolyzed to
deprotect the TBS group, acetylated, and finally deprotected
to afford the final compounds, 8Z and 8E, respectively.
HO
BnO
OHC
O
O
O
a
b
bc
,
,
O
O
O
OHCO
BnO
O
O
HO
O
10
9
O
d,e
R1O
TBSO
BnO
RO
O
O
O
f,g
h-j
O
O
O
HO
R2O
14 R1=H, R2=Bn
BnO
13
11 R=H
12 R=TBS
15 R1=COC13H27, R2=Bn
6 R1=COC13H27, R2=H
Fig. 3. The synthesis of lactone 6. Reagents and conditions: (a) Ref. [15]; (b) 2 N HCl, THF; (c) NaIO4, MeOH, H2O; (d) NaBH4, MeOH, 54% in three steps;
(e) TBDMSCl, imidazole, DMAP, CH2Cl2, 84%; (f) PhOC(=S)Cl, DMAP, CH2Cl2, 99%; (g) Bu3SnH, AIBN, toluene, 95%; (h) Bu4NF, THF, 95%; (i)
C
13H27COCl, pyridine, DMAP, CH2Cl2, 95%; (j) BCl3, CH2Cl2, –78 °C, 92%.

72
J. Lee et al. / European Journal of Medicinal Chemistry 39 (2004) 69–77
TBSO
RO
TBSO
ArO
TBSO
ArO
O
O
O
c,d
a,b
O
O
O
13
+
C17H33
C17H33
16 R=H
17 R=Ar
18Z
18E
e-g
R1O
R2O
R1O
O
O
O
O
R2O
C
17H33
C17H33
19Z R1=H, R2=Ar
20Z R1=Ac, R2=Ar
8Z R1=Ac, R2=H
19E R1=H, R2=Ar
20E R1=Ac, R2=Ar
8E R1=Ac, R2=H
Fig. 4. The synthesis of lactone 8. Reagents and conditions: (a) H2, Pd–C, EtOAc, 95%; (b) 4-methoxyphenol, PPh3, DEAD, THF, 90%; (c) LiHMDS, THF,
–78 °C; C17H33CHO, HMPA, 76%; (d) MsCl, NEt3, CH2Cl2; DBU, 58% for 18E, 38% for 18Z; (e) AcOH–H2O–THF, 60 °C, 90%; (f) Ac2O, pyridine, DMAP,
CH2Cl2, 99%; (g) CAN, CH3CN–H2O, 0 °C, 92%.
3. Results and discussion
carbonyl) which presumably interacts with the protein–lipid
interface.
The interaction of the target compounds (6, 8Z and 8E)
with PKC was assessed in terms of the ability of the ligand to
displace bound [3H]phorbol 12,13-dibutyrate (PDBU) from
the recombinant single isozyme (PKC-a) in the presence of
phosphatidylserine, as previously described [6]. The inhibi-
tion curves obtained for all ligands were of the type expected
for competitive inhibition, and the ID50 values were deter-
mined by fitting the data points to the theoretical non-
cooperative competition curve. The Kis for inhibition of
binding were calculated from the ID50 values. As shown in
Table 1, the binding affinities of the synthesized compounds
were Ki = 2.623 µM for 6, Ki = 1.080 µM for 8Z and
Ki = 0.92 µM for 8E. Their affinities were estimated to be ca.
27, 90, and 70 times less potent than the corresponding
parent compounds, which were reported as Ki = 0.096 µM for
5, Ki = 0.012 µM for 7Z and Ki = 0.013 µM for 7E, respec-
tively [6]. However, the order of their binding affinities rep-
resented the same pattern as the parent compounds
(8Z =8E >6), and the Z- and E-isomers showed similar
potencies.
Since the binding mode of DAG-lactones depends on the
principle that “the carbonyl bearing the larger lipid chain,
corresponding to C9–OH in the phorbol ester–PKC-d com-
plex, will not be involved in binding to the receptor, but rather
it will bind to the protein–lipid interface” [13,15], 3 and 4 are
predicted to bind to the enzyme in the sn-2 binding mode
because the lipid is attached to the sn-1 carbonyl placed in the
3-O-acyl group. The computer-generated molecular super-
position of 3 and 4 on phorbol-12-myristate-13-acetate
(PMA) in the sn-2 mode indicated that the three-point fit for
the elongated homolog 4 to PMA was significantly better
than that of 3 on PMA (rms 0.391 vs. 1.237) and the distance
between the lactone carbonyl (sn-2 carbonyl) and the hy-
droxyl in 3 appeared to be too short to match the C20–OH and
the C3=O in PMA [17]. In the same way, the 3-alkylidene
DAG-lactones 7 and 8 would favor the sn-1 binding mode
because the fatty chain is close to the lactone carbonyl (sn-2
In order to investigate the comparison by modeling be-
tween phorbol, the sn-2 mode of 3 and 4, and the sn-1 mode
of 7 and 8 for comparison with their binding affinities, energy
minimized conformers of 3, 4, 7Z and 8Z, were obtained by
a conformational search based on a previous report, respec-
tively [11]. The pharmacophoric comparisons were exam-
ined only for the distance (5.348 Å) between the C3=O and
the C20–OH in phorbol 13-acetate obtained from the X-ray
structure of its complex with the C1b domain of PKC-d [11]
because the C9–OH of phorbol is assumed to bind to the
protein–lipid interface rather than to the enzyme as evi-
denced by the X-ray complex (Fig. 5). In the 2-deoxy-L-
ribonolactone analogs, whereas the distance (3.93 Å) be-
tween the lactone carbonyl (sn-2 carbonyl) and the hydroxyl
in the sn-2 mode of 3 was shorter than the 5.348 Å of phorbol,
the distance in 4 (5.38 Å) was very close to that of phorbol.
This improved matching of 4 overcame the entropic penalty
due to the one carbon extension and led to the enhancement
of its binding affinity. On the other hand, the distances be-
tween the acyloxy carbonyl (sn-1 carbonyl) and the hydroxyl
group in the (R)-lactone analogs were calculated as 5.06 Å in
7Z and 5.80 Å in 8Z, and in its sn-1 mode, were compared
with that of phorbol (5.348 Å) (Fig. 5). Contrary to the
2-deoxy-L-ribonolactone template, the one-carbon elonga-
tion in the (R)-lactone template caused relatively poor phar-
macophoric matching and resulted in weaker binding to the
enzyme. Although the relative difference to phorbol is not
significant (0.288 vs. 0.452 Å), the extra carbon in 8Z may
force the chain to coil somewhat into a less favorable confor-
mation, which costs an additional entropic penalty to reach
the binding site appropriately, and therefore, affords more
loss of binding affinity (Figs. 4 and 5).
The SAR results of 5 and 6 were hard to explain. We
initially expected an improvement in binding affinity of the
elongated homolog 6 since its binding would be in the sn-2
mode as a 2-deoxy-L-ribonolactone analogs. Unfortunately,
the binding affinity of 6 was weaker than that of the parent

J. Lee et al. / European Journal of Medicinal Chemistry 39 (2004) 69–77
73
O
O
H
H
OH
OH
O
OH
5.348
phorbol 13-acetate
4 atoms separation
2-Deoxy-L-Ribonolactone (sn-2 mode)
better
O
O
O
O
OH
O
OH
O
O
O
3 atoms separation
4 atoms separation
4
3
R-Lactone (sn-1 mode)
worse
O
O
O
O
O
O
OH
OH
O
O
4 atoms separation
5 atoms separation
8Z
7Z
Fig. 5. The overlay matching of DAG-lactones with C3=O and C20–OH in phorbol 13-acetate.
lactone 5. This template cannot be compared directly with
2-deoxy-L-ribonolactone analogs because the sn-1 carbonyls
in 4 and 6 are placed on different positions of the lactone. We,
therefore, reason that in this case, entropic loss due to the one
carbon extension is more significant than pharmacophoric
matching for enzyme binding. Nevertheless, the penalty in-
curred by lengthening the OH tether in the sn-2 mode in 6 is
less severe (×27 loss) than that in the sn-1 binding mode (×90
and ×70 loss) found in 8Z and 8E because of improved
pharmacophoric matching in the sn-2 mode.
In summary, we synthesized one-carbon elongated ana-
logs (6, 8Z and 8E) of the ultrapotent DAG (R)-lactones (5,
7Z and 7E) reported previously and found that they exhibited
lower binding affinities to PKC-a than did the parent com-
pounds. The weaker binding of 8Z/8E was a consequence of
their poor fit on the phorbol ester in the sn-1 mode, as
revealed by pharmacophore-guided molecular modeling, and
the entropic penalty for enzyme binding due to the relatively
flexible hydroxyethyl group. However, improved matching
of 6 in the sn-2 mode was overwhelmed by the entropic loss
upon elongation resulting in weaker binding although its
magnitude of loss is less severe than that of 8.
Laboratory Devices, USA and are uncorrected. Silica gel
chromatography was performed on silica gel 60, 230–
400 mesh (E. Merck). Proton and 13C NMR spectra were
recorded on a Bruker AC-250 instrument at 250 and
62.9 MHz, respectively. Spectra were referenced to the sol-
vent in which they were run (7.24 ppm for CDCl3). Infrared
spectra were recorded on a Perkin–Elmer 1600 Series FTIR
and specific rotations were measured in a Perkin–Elmer
Model 241 polarimeter. Positive-ion fast-atom bombardment
mass spectra (FABMS) were obtained on a VG 7070E mass
spectrometer. Elemental analyses were performed by Atlan-
tic Microlab, Inc., Atlanta, GA. Analyses indicated by the
symbols of the elements or functions were within 0.4% of
the theoretical values.
4.1.1. 5-O-Benzyl-3-C-hydroxypropyl-␣-lactone-1,2-O-iso-
propylidene-␣-D-ribofuranose (9)
The compound was prepared from 1,2-O-isopropylidene-
D-xylose by following the procedure of a previous report
[18].
4.1.2. 6-O-Benzyl-4-C-hydroxymethyl-2,3-dideoxy-D-threo-
hexono-1,4-lactone (11)
A solution of 9 (8.36 g, 25 mmol) in THF (150 ml) and
H2O (150 ml) was treated with Dowex H+ resin (15 g,
prewashed with MeOH) and refluxed for 24 h. The reaction
mixture was filtered and concentrated in vacuo. The residue
was dissolved with EtOAc and treated with NaHCO3 and
MgSO4. The mixture was filtered and concentrated in vacuo
4. Experimental protocols
4.1. Chemistry
All chemical reagents were commercially available. Melt-
ing points were determined on a MelTemp II apparatus,

74
J. Lee et al. / European Journal of Medicinal Chemistry 39 (2004) 69–77
to give the corresponding hemiacetal, which was used for the
next step without further purification. The hemiacetal was
dissolved in MeOH (150 ml) and H2O (75 ml) and treated
with sodium periodate (10.7 g, 50 mmol). After stirring for
1 h at room temperature, the reaction mixture was filtered and
concentrated in vacuo. The residue was dissolved with
EtOAc, dried over MgSO4 and concentrated in vacuo to give
aldehyde 10, which was used for the next step without further
purification. The aldehyde 10 was dissolved in MeOH
(200 ml) and cooled to 0 °C. The solution was treated with
sodium borohydride portionwise until the starting material
was consumed on TLC. The reaction mixture was quenched
with acetone and, after 20 min, acetic acid, and then concen-
trated in vacuo. The residue was diluted with EtOAc, filtered
and the filtrate was concentrated in vacuo. The residue was
purified by flash column chromatography over silica gel with
EtOAc/hexanes (3:1) to EtOAc/MeOH (9:1) as eluant to give
11 (5.0 g, 75% in three steps) as a syrup: [␣]D = +8.56 (c 1.87,
dissolved in ether, filtered, and the filtrate was concentrated
in vacuo. The residue was purified by flash column chroma-
tography over silica gel with EtOAc/hexanes (2:3) as eluant
to give the corresponding thionoester (0.70 g, 99%) as an oil.
The thionoester (0.7 g, 1.35 mmol) was dissolved in tolu-
ene (30 ml) and treated with AIBN (0.12 g, 0.677 mmol) and
tributyltin hydride (0.72 ml, 2.71 mmol). After heating at
90 °C for 1 h, the reaction mixture was cooled and concen-
trated in vacuo. The residue was purified by flash column
chromatography over silica gel with EtOAc/hexanes (1:5) as
eluant to give 13 (0.47 g, 95%) as an oil: [␣]D = +2.23 (c 2.20,
1
CHCl3); H NMR (CDCl3) d 7.2–7.4 (m, 5H, Ph), 4.46 (s,
2H, PhCH2O), 3.54–3.71 (m, 4H, J = 3.7 Hz, H-5 and
BnOCH2), 2.39–2.72 (m, 2H, H-2), 2.0–2.31 (m, 2H, H-3),
1.95 (t, 2H, J = 6.2 Hz, BnOCH2CH2), 0.86 (s, 9H), 0.03 (d,
6H, J = 1.5 Hz); 13C NMR (CDCl3) d 177.26, 137.93, 128.38,
127.66, 127.61, 87.37, 73.19, 68.61, 65.48, 36.58, 29.70,
28.70, 25.74, 23.37, 18.15, –5.60, –5.69; IR (neat) 1774
(C=O) cm–1. Anal. C20H32O4Si (C, H).
1
CHCl3); H NMR (CDCl3) d 7.2–7.4 (m, 5H, Ph), 4.52 (s,
2H, PhCH2O), 3.93 (dd, 1H, J = 4.2, 6.7 Hz, H-5), 3.75 (d of
AB, 1H, J = 12.2 Hz, H-6a), 3.61 (d of AB, 1H, J = 12.2 Hz,
H-6b), 3.50–3.66 (m, 2H, H-4′), 2.88 (bs, 2H, OH), 2.5–2.63
(m, 2H, H-2), 2.08–2.35 (m, 2H, H-3); 13C NMR (CDCl3) d
177.73, 137.25, 128.47, 127.94, 127.77, 88.64, 73.61, 72.37,
69.78, 64.84, 29.00, 25.08; IR (neat) 3418 (OH), 1760 (C=O)
cm–1. Anal. C14H18O5 (C, H).
4.1.5. 4-C-Benzyloxyethyl-2,3-dideoxy-D-glyceropentono-
1,4-lactone (14)
A solution of 13 (0.092 g, 0.25 mmol) in THF (5 ml) was
treated dropwise with tetrabutylammonium fluoride (1 M,
0.5 ml, 0.5 mmol) and stirred for 30 min at room temperature.
The reaction mixture was concentrated in vacuo and the
residue was purified by flash column chromatography over
silica gel with EtOAc/hexanes (3:1) as eluant to give 14
4.1.3. 6-O-Benzyl-4-C-tert-butyldimethylsilyloxymethyl-2,3-
dideoxy-D-threo-hexono-1,4-lactone (12)
1
(0.60 g, 95%) as an oil: [␣]D = –6.67 (c 0.66, CHCl3); H
A solution of 11 (0.466 g, 1.75 mmol) in CH2Cl2 (30 ml)
was treated with TBS chloride (0.316 g, 2.10 mmol), imida-
zole (0.48 g, 7 mmol) and 4-dimethylaminopyridine (0.02 g,
0.175 mmol). After stirring for 14 h at room temperature, the
reaction mixture was diluted with CH2Cl2 and washed with
0.5 N HCl solution, saturated NaHCO3 and brine. The or-
ganic layer was dried over MgSO4 and concentrated in
vacuo. The residue was purified by flash column chromatog-
raphy over silica gel with EtOAc/hexanes (2:3) as eluant to
give 12 (0.56 g, 84%) as a colorless oil: [␣]D = +1.60 (c 1.75,
CHCl3); 1H NMR (CDCl3) d 7.2–7.4 (m, 5H, Ph), 4.53 (t, 2H,
J = 12.5 Hz, PhCH2O), 3.92 (dd, 1H, J = 3.7, 7.5 Hz, H-5),
3.71 (d of AB, 1H, J = 10.7 Hz, H-6a), 3.66 (d of AB, 1H,
J = 10.7 Hz, H-6b), 3.50–3.65 (m, 2H, H-4′), 2.88 (bs, 2H,
OH), 2.5–2.65 (m, 2H, H-2), 2.08–2.40 (m, 2H, H-3), 0.86 (s,
9H), 0.03 (s, 6H); 13C NMR (CDCl3) d 177.12, 137.44,
128.50, 127.94, 127.76, 88.05, 73.60, 72.47, 69.65, 66.20,
29.18, 25.72, 25.45, 18.12, –5.60, –5.69; IR (neat) 3448
(OH), 1776 (C=O) cm–1. Anal. C20H32O5Si (C, H).
NMR (CDCl3) d 7.2–7.4 (m, 5H, Ph), 4.48 (s, 2H, PhCH2O),
3.52–3.70 (m, 4H, J = 3.7 Hz, H-5 and BnOCH2), 1.9–2.7
(m, 6H, H-2, H-3 and BnOCH2CH2); 13C NMR (CDCl3) d
176.96, 137.40, 128.50, 127.91, 127.77, 87.47, 73.44, 66.62,
65.51, 36.60, 28.78, 28.72; IR (neat) 3441 (OH), 1769 (C=O)
cm–1. Anal. C14H18O4 (C, H).
4.1.6. 4-C-Benzyloxyethyl-5-O-tetradecanoyl-2,3-dideoxy-
D-glyceropentono-1,4-lactone (15)
A solution of 13 (0.056 g, 0.224 mmol) in CH2Cl2 (10 ml)
was treated with pyridine (0.073 ml, 0.898 mmol), a catalytic
amount of 4-dimethylaminopyridine and tetradecanoyl chlo-
ride (0.122 ml, 0.449 mmol). After stirring for 2 h at room
temperature, the reaction mixture was concentrated in vacuo.
The residue was dissolved in cold ether, filtered, and the
filtrate was concentrated in vacuo. The residue was purified
by flash column chromatography over silica gel with
EtOAc/hexanes (1:2) as eluant to give 15 (0.098 g, 95%) as
1
an oil: [␣]D = +3.33 (c 1.62, CHCl3); H NMR (CDCl3) d
7.2–7.4 (m, 5H, Ph), 4.46 (s, 2H, PhCH2O), 4.20 (d of AB,
1H, J = 12.0 Hz, H-5a), 4.11 (d of AB, 1H, J = 12.0 Hz,
H-5b), 3.61 (t, 2H, J = 6.0 Hz, BnOCH2), 2.45–2.7 (m, 2H,
H-2), 2.30 (t, 2H, J = 7.4 Hz, CH2CO2), 2.08–2.28 (m, 2H,
H-3), 2.03 (t, 2H, J = 6.0 Hz, BnOCH2CH2), 1.1–1.7 (m,
22H), 0.86 (distorted t, 3H); 13C NMR (CDCl3) d 176.36,
173.14, 137.72, 128.44, 127.76, 127.67, 85.07, 73.33, 68.17,
65.15, 36.80, 34.08, 31.88, 29.63, 29.60, 29.55, 29.40, 29.31,
4.1.4. 4-C-Benzyloxyethyl-5-O-tert-butyldimethylsilyl-2,3-
dideoxy-D-glyceropentono-1,4-lactone (13)
A solution of 12 (0.52 g, 1.37 mmol) in CH2Cl2 (30 ml)
was treated with 4-dimethylaminopyridine (0.668 g,
5.46 mmol) and phenyl chlorothionoformate (0.38 ml,
2.73 mmol). After stirring for 3 h at room temperature, the
reaction mixture was concentrated in vacuo. The residue was

J. Lee et al. / European Journal of Medicinal Chemistry 39 (2004) 69–77
75
29.19, 29.07, 28.97, 28.93, 24.81, 22.65, 14.08. IR (neat)
1780 and 1741 (C=O) cm–1. Anal. C28H44O5 (C, H).
To remove 4-methoxyphenol, the crude 17 was dissolved in
CH2Cl2 (20 ml) and treated with acetic anhydride (0.5 ml)
and pyridine (0.5 ml). After stirring for 1 h at room tempera-
ture, the reaction mixture was concentrated in vacuo. The
residue was purified by flash column chromatography over
silica gel with EtOAc/hexanes (1:3) as eluant to give pure 17
4.1.7. 4-C-Hydroxyethyl-5-O-tetradecanoyl-2,3-dideoxy-D-
glyceropentono-1,4-lactone (6)
A solution of 15 (0.083 g, 0.18 mmol) in CH2Cl2 (10 ml)
was cooled to –78 °C and treated dropwise with boron
trichloride in CH2Cl2 (1 M, 0.9 ml, 0.9 mmol). After stirring
for 6 h at –78 °C, the reaction mixture was quenched with
saturated NaHCO3 solution (1 ml) and immediately parti-
tioned between ether and pH 7 buffer solution. The organic
layer was washed with pH 7 buffer solution five times, dried
over MgSO4, and concentrated in vacuo. The residue was
purified by flash column chromatography over silica gel with
EtOAc/hexanes (2:1) as eluant to give 6 (0.061 g, 92%) as a
1
(0.4 g, 90%) as an oil: [␣]D = +2.80 (c 1.54, CHCl3); H
NMR (CDCl3) d 6.7–6.9 (m, 4H, aromatic), 4.05 (t, 2H,
J = 6.2 Hz, ArOCH2), 3.75 (s, 3H, CH3O), 3.74 (d ofAB, 1H,
J = 10.8 Hz, H-5a), 3.63 (d of AB, 1H, J = 10.8 Hz, H-5b),
2.4–2.76 (m, 2H, H-2), 2.0–2.38 (m, 2H, H-3 and
ArOCH2CH2), 0.86 (s, 9H), 0.03 (d, 6H, J = 2.3 Hz); 13C
NMR (CDCl3) d 177.01, 153.92, 152.35, 115.23, 114.59,
87.07, 68.45, 63.65, 55.57, 36.02, 29.54, 28.65, 25.67, 18.08,
–5.66, –5.73; IR (neat) 1774 (C=O), 1508 and 1472 cm–1.
Anal. C20H32O5Si (C, H).
1
white solid: m.p. 39 °C; [␣]D = –3.75 (c 0.24, CHCl3); H
NMR (CDCl3) d 4.23 (d of AB, 1H, J = 11.9 Hz, H-5a), 4.13
(d of AB, 1H, J = 11.9 Hz, H-5b), 3.83 (t, 2H, J = 6.1 Hz,
HOCH2), 2.5–2.76 (m, 2H, H-2), 2.32 (t, 2H, J = 7.4 Hz,
CH2CO2), 2.21 (t, 2H, J = 8.5 Hz, H-3), 1.99 (t, 2H,
J = 6.1 Hz, BnOCH2CH2), 1.18–1.65 (m, 22H), 0.86 (dis-
torted t, 3H); 13C NMR (CDCl3) d 176.28, 173.23, 85.23,
67.96, 57.86, 39.15, 34.09, 31.88, 29.64, 29.60, 29.56, 29.42,
29.32, 29.20, 29.15, 29.07, 28.89, 24.81, 22.65, 14.09. IR
(KBr) 3543 (OH), 1752 and 1739 (C=O) cm–1. Anal.
C21H38O5 (C, H), MS(FAB) m/z 371 (MH+).
4.1.10. 5-O-tert-Butyldimethylsilyl-4-C-4-methoxyphenoxy-
ethyl-2-C-[(Z)-9-octadecaenylidene]-2,3-dideoxy-D-glycero-
pentono-1,4-lactone (18Z) and 5-O-tert-butyldimethylsilyl-
4-C-4-methoxyphenoxyethyl-2-C-[(E)-9-octadecaenylidene]-
2,3-dideoxy-D-glyceropentono-1,4-lactone (18E)
A stirred solution of 17 (0.38 g, 1.0 mmol) in THF (4 ml)
was cooled to –78 °C and treated slowly with lithium bis(tri-
methylsilyl)amide (1 M in THF, 1.2 ml, 1.2 mmol). After
stirring for 1 h, the reaction mixture was treated with a
mixture of oleyl aldehyde (0.32 g, 1.2 mmol) and hexameth-
ylphosphoramide (0.215 g, 1.2 mmol) and stirring was con-
tinued for 1 h at –78 °C and for 1 h at –40 °C. The mixture
was quenched with a solution of saturated ammonium chlo-
ride and diluted with ether. The organic layer was washed
with H2O, dried over MgSO4, and concentrated in vacuo. The
residue was purified by flash column chromatography over
silica gel with EtOAc/hexanes (1:5) as eluant to give the
intermediate b-hydroxy lactone (0.478 g, 74%). This com-
pound was dissolved in CH2Cl2 (20 ml) and cooled to 0 °C.
The solution was then stirred with triethylamine (0.41 ml,
3 mmol) and methanesulfonyl chloride (0.12 ml, 1.48 mmol)
for 30 min. The reaction mixture was warmed to room tem-
perature and stirred for 1 h before the addition of 1,8-
diazabicyclo[5,4,0]undec-7-ene (0.66 ml, 3 mmol). After
further stirring for 6 h at room temperature, the mixture was
concentrated and diluted with ether. The ethereal solution
was washed with diluted HCl, H2O and brine. The organic
layer was dried over MgSO4 and concentrated in vacuo. The
residue was purified by flash column chromatography over
silica gel with EtOAc/hexanes (1:8) as eluant to give 18Z
(Z-isomer, 0.20 g, 38%) and 18E (E-isomer: 0.27 g, 58%) as
colorless oils, respectively.
4.1.8.
5-O-tert-Butyldimethylsilyl-4-C-hydroxyethyl-2,3-
dideoxy-D-glyceropentono-1,4-lactone (16)
A solution of 13 (0.5 g, 1.37 mmol) in ethylacetate (30 ml)
was treated with 10% palladium on carbon (1.0 g) and acetic
acid (four drops). The suspension was hydrogenated under a
balloon of hydrogen for 6 h, filtered and concentrated in
vacuo. The residue was purified by flash column chromatog-
raphy over silica gel with EtOAc/hexanes (2:1) as eluant to
give 16 (0.358 g, 95%) as an oil: [␣]D = –0.92 (c 2.50,
CHCl3); 1H NMR (CDCl3) d 3.78 (t, 2H, J = 6.1 Hz,
HOCH2), 3.70 (d of AB, 1H, J = 10.8 Hz, H-5a), 3.59 (d of
AB, 1H, J = 10.8 Hz, H-5b), 2.43–2.74 (m, 2H, H-2), 1.98–
2.32 (m, 2H, H-3), 1.93 (t, 2H, J = 6.1 Hz, BnOCH2CH2),
0.86 (s, 9H), 0.03 (d, 6H, J = 2.3 Hz); 13C NMR (CDCl3) d
177.23, 87.57, 68.29, 57.86, 39.18, 29.38, 28.89, 25.70,
18.12, –5.60, –5.69; IR (neat) 3445 (OH), 1770 (C=O) cm–1.
Anal. C13H26O4Si (C, H).
4.1.9. 5-O-tert-Butyldimethylsilyl-4-C-4-methoxyphenoxy-
ethyl-2,3-dideoxy-D-glyceropentono-1,4-lactone (17)
To a stirred solution of 16 (0.32 g, 1.166 mmol) and
triphenylphosphine (0.428 g, 1.63 mmol) in THF (4 ml) was
slowly added over a period of 20 min a solution of diethyl
azodicarboxylate (0.284 g, 1.63 mmol) and 4-metho-
xyphenol (0.434 g, 3.5 mmol) in THF (4 ml). The reaction
mixture was stirred for 2 h at room temperature and concen-
trated in vacuo. The residue was purified by flash column
chromatography on silica gel with EtOAc/hexanes (1:2) as
eluant to give 17 along with inseparable 4-methoxyphenol.
Compound 18Z: Rf = 0.56 (EtOAc/hexanes = 1:5);
[␣]D = –0.57 (c 1.58, CHCl3); 1H NMR (CDCl3) d 6.73–6.85
(m, 4H, aromatic), 6.09 (m, 1H, >C=CH), 5.33 (m, 2H,
CH2CH=CHCH2), 4.04 (t, 2H, J = 6.3 Hz, ArOCH2), 3.75 (s,
3H, CH3O), 3.67 (d ofAB, 1H, J = 10.7 Hz, H-5a), 3.59 (d of
AB, 1H, J = 10.7 Hz, H-5b), 2.55–3.0 (m, 4H, >C=CHCH2-
and
H-3a,b),
1.9–2.2
(m,
6H,
ArOCH2CH2,

76
J. Lee et al. / European Journal of Medicinal Chemistry 39 (2004) 69–77
CH2CH=CHCH2), 1.1–1.4 (m, 22H), 0.75–0.9 (m, 12H, CH3
and (CH3)3Si), 0.03 (s, 6H, J = 3.0 Hz); 13C NMR (CDCl3) d
169.36, 153.96, 152.56, 143.46, 129.93, 129.82, 125.44,
115.36, 114.65, 83.32, 67.87, 63.70, 55.70, 36.16, 36.00,
31.90, 29.75, 29.65, 29.52, 29.38, 29.31, 29.22, 29.13, 27.61,
27.21, 25.72, 22.67, 18.15, 14.11, –5.56; IR (neat) 1757
(C=O), 1670 (C=C), 1508 and 1464 cm–1. Anal. C38H64O5Si
(C, H).
(CHCl3) 3371 (OH), 1727 (C=O), 1681 (C=C), 1511 and
1466 cm–1. Anal. C32H50O5 (C, H).
4.1.12. 5-O-Acetyl-C-4-methoxyphenoxyethyl-2-C-[(Z)-9-
octadecaenylidene]-2,3-dideoxy-D-glyceropentono-1,4-lac-
tone (20Z) and 5-O-acetyl-C-4-methoxyphenoxyethyl-2-
C-[(E)-9-octadecaenylidene]-2,3-dideoxy-D-glyceropento-
no1,4-lactone (20E
)
Compound 18E: Rf = 0.50 (EtOAc/hexanes = 1:5);
[␣]D = –1.28 (c 1.56, CHCl3); 1H NMR (CDCl3) d 6.73–6.85
(m, 4H, aromatic), 6.63 (m, 1H, >C=CH), 5.33 (m, 2H,
CH2CH=CHCH2), 4.04 (m, 2H, ArOCH2), 3.75 (s, 3H,
CH3O), 3.69 (d ofAB, 1H, J = 10.7 Hz, H-5a), 3.60 (d ofAB,
1H, J = 10.7 Hz, H-5b), 2.65–2.97 (m, 2H, H-3a,b), 1.9–2.25
(m, 8H, >C=CHCH2–, ArOCH2CH2 and CH2CH=CHCH2),
1.15–1.5 (m, 22H), 0.75–0.9 (m, 12H, CH3 and (CH3)3Si),
0.03 (s, 6H, J = 3.3 Hz); 13C NMR (CDCl3) d 170.53, 153.99,
152.50, 139.84, 129.99, 129.70, 127.52, 115.36, 114.65,
84.05, 67.97, 63.69, 55.68, 36.13, 32.64, 31.89, 30.12, 29.74,
29.71, 29.51, 29.31, 29.13, 28.10, 27.20, 27.14, 25.67, 22.66,
18.15, 14.09, –5.56; IR (neat) 1759 (C=O), 1683 (C=C),
1508 and 1464 cm–1. Anal. C38H64O5Si (C, H).
A solution of 19Z (0.098 g, 0.19 mmol) in CH2Cl2 (12 ml)
was cooled to –10 °C and treated with pyridine (0.092 ml,
1.14 mmol), acetic anhydride (0.072 ml, 0.76 mmol) and a
catalytic amount of 4-dimethylaminopyridine. After stirring
for 30 min at 0 °C, the reaction mixture was concentrated in
vacuo at 0 °C. The residue was purified by flash column
chromatography over silica gel with EtOAc/hexanes (1:1) as
eluant to give 20Z (0.105 g, 99%) as an oil: [␣]D = +5.81 (c
1
0.86, CHCl3); H NMR (CDCl3) d 6.73–6.85 (m, 4H, aro-
matic), 6.16 (m, 1H, >C=CH), 5.33 (m, 2H,
CH2CH=CHCH2), 4.22 (d of AB, 1H, J = 11.9 Hz, H-5a),
4.15 (d of AB, 1H, J = 11.9 Hz, H-5b), 4.05 (m, 2H,
ArOCH2), 3.74 (s, 3H, CH3O), 2.58–3.0 (m,
4H, >C=CHCH2- and H-3a,b), 2.03 (s, 3H, CH3COO), 1.9–
2.3 (m, 6H, ArOCH2CH2, CH2CH=CHCH2), 1.15–1.4 (m,
22H), 0.86 (distorted t, 3H, CH3); 13C NMR (CDCl3) d
170.38, 168.62, 154.03, 152.27, 144.70, 129.88, 129.71,
124.19, 115.24, 114.61, 81.24, 67.78, 63.27, 55.62, 36.58,
36.19, 31.83, 29.70, 29.67, 29.63, 29.59, 29.46, 29.29, 29.24,
29.20, 29.13, 29.03, 27.62, 27.15, 27.12, 22.61, 20.61, 14.05.
IR (neat) 1753 (C=O), 1670 (C=C), 1509 and 1466 cm–1.
Anal. C34H52O6 (C, H).
4.1.11. C-4-Methoxyphenoxyethyl-2-C-[(Z)-9-octadecaeny-
lidene]-2,3-dideoxy-D-glyceropentono-1,4-lactone (19Z)
and 4-C-4-methoxyphenoxyethyl-2-C-[(E)-9-octadecaeny-
lidene]-2,3-dideoxy-D-glyceropentono-1,4-lactone (19E)
A solution of 18Z (0.2 g, 0.318 mmol) in acetic acid
(9 ml), H2O (3 ml) and THF (3 ml) was heated at 60 °C for
48 h and concentrated in vacuo. The residue was purified by
flash column chromatography over silica gel with
EtOAc/hexanes (1:1) as eluant to give 19Z (0.147 g, 90%) as
Compound 20E was obtained from 19E, by following the
above procedure, as an oil in 99% yield: [␣]D = +0.83 (c 0.84,
1
1
an oil: [␣]D = –0.64 (c 0.78, CHCl3); H NMR (CDCl3) d
CHCl3); H NMR (CDCl3) d 6.73–6.85 (m, 4H, aromatic),
6.7–6.9 (m, 4H, aromatic), 6.18 (m, 1H, >C=CH), 5.33 (m,
2H, CH2CH=CHCH2), 4.05 (m, 2H, ArOCH2), 3.75 (s, 3H,
CH3O), 3.65 (dd of AB, 2H, H-5a,b), 2.57–3.0 (m,
4H, >C=CHCH2- and H-3a,b), 1.9–2.3 (m, 6H,
ArOCH2CH2, CH2CH=CHCH2), 1.15–1.44 (m, 22H), 0.86
(distorted t, 3H, CH3); 13C NMR (CDCl3) d 169.18, 154.08,
152.19, 145.07, 129.86, 129.72, 124.52, 115.27, 114.61,
83.63, 66.76, 63.75, 55.60, 36.20, 35.85, 32.53, 31.83, 29.68,
29.58, 29.45, 29.28, 29.24, 29.17, 29.14, 29.00, 27.65, 27.13,
22.61, 14.05. IR (neat) 3421 (OH), 1752 (C=O), 1670 (C=C),
1508 and 1465 cm–1. Anal. C32H50O5 (C, H).
6.71 (m, 1H, >C=CH), 5.33 (m, 2H, CH2CH=CHCH2), 4.24
(d of AB, 1H, J = 11.9 Hz, H-5a), 4.18 (d of AB, 1H,
J = 11.9 Hz, H-5b), 4.05 (m, 2H, ArOCH2), 3.74 (s, 3H,
CH3O), 2.7–2.94 (m, 2H, H-3a,b), 2.03 (s, 3H, CH3COO),
1.9–2.3 (m, 8H, ArOCH2CH2, >C=CHCH2 and
CH2CH=CHCH2), 1.15–1.55 (m, 22H), 0.86 (distorted t, 3H,
CH3); 13C NMR (CDCl3) d 170.31, 169.79, 154.04, 152.22,
141.15, 129.95, 129.57, 126.20, 115.26, 114.60, 81.85,
67.96, 63.29, 55.59, 36.40, 33.10, 31.81, 30.15, 29.67, 29.63,
29.56, 29.43, 29.23, 29.06, 28.02, 27.13, 27.06, 22.60, 20.55,
14.03. IR (neat) 1754 (C=O), 1680 (C=C), 1509 and
1466 cm–1. Anal. C34H52O6 (C, H).
Compound 19E was obtained from 18E, by following the
above procedure, as a low-melting solid in 90% yield: m.p.
1
40 °C; [␣]D = –7.85 (c 1.86, CHCl3); H NMR (CDCl3) d
4.1.13. 5-O-Acetyl-C-4-hydroxyethyl-2-C-[(Z)-9-octadeca-
enylidene]-2,3-dideoxy-D-glyceropentono-1,4-lactone (8Z)
6.7–6.9 (m, 4H, aromatic), 6.71 (m, 1H, >C=CH), 5.33 (m,
2H, CH2CH=CHCH2), 4.05 (m, 2H, ArOCH2), 3.75 (s, 3H,
CH3O), 3.66 (dd of AB, 2H, H-5a,b), 2.7–2.95 (m, 2H,
H-3a,b), 1.9–2.3 (m, 8H, ArOCH2CH2, >C=CHCH2– and
CH2CH=CHCH2), 1.15–1.5 (m, 22H), 0.86 (distorted t, 3H,
CH3); 13C NMR (CDCl3) d 170.48, 154.07, 152.19, 141.50,
129.94, 129.63, 126.60, 115.29, 114.61, 84.39, 67.01, 63.75,
55.60, 36.04, 32.68, 32.53, 31.82, 30.16, 29.67, 29.65, 29.57,
29.44, 29.23, 29.08, 28.01, 27.13, 27.08, 22.60, 14.05. IR
and
5-O-acetyl-C-4-hydroxyethyl-2-C-[(E)-9-octadeca-
enylidene]-2,3-dideoxy-D-glyceropentono-1,4-lactone (8E)
A solution of 20Z (0.09 g, 0.16 mmol) in acetonitrile
(4 ml) and H2O (1 ml) was cooled to 0 °C and treated with
ammonium cerium(IV) nitrate (0.18 g, 0.32 mmol) in one
portion.After stirring for 10 min at 0 °C, the reaction mixture
was partitioned between CH2Cl2 and H2O. The aqueous
layer was extracted with CH2Cl2, and the combined organic

J. Lee et al. / European Journal of Medicinal Chemistry 39 (2004) 69–77
77
layer was washed with H2O and brine. The organic layer was
dried over MgSO4 and concentrated in vacuo. The residue
was purified by flash column chromatography over silica gel
with EtOAc/hexanes (1:1) as eluant to give 8Z (0.067 g,
were performed on a Silicon Graphics O2 R10000 worksta-
tion.
1
Acknowledgements
92%) as an oil: [␣]D = +4.65 (c 0.86, CHCl3); H NMR
(CDCl3)
d 6.18 (m, 1H, >C=CH), 5.33 (m, 2H,
This research was supported by KOSEF grant R01-2000-
000-00176-0 from the Korea Science and Engineering Foun-
dation.
CH2CH=CHCH2), 4.19 (d of AB, 1H, J = 12.0 Hz, H-5a),
4.13 (d ofAB, 1H, J = 12.0 Hz, H-5b), 3.81 (t, 2H, J = 6.1 Hz,
HOCH2), 2.6–2.9 (m, 4H, >C=CHCH2- and H-3a,b), 2.04 (s,
3H, CH3COO), 1.9–2.1 (m, 6H, ArOCH2CH2,
CH2CH=CHCH2), 1.15–1.5 (m, 22H), 0.86 (distorted t, 3H,
CH3); 13C NMR (CDCl3) d 170.48, 168.71, 144.72, 129.88,
129.69, 124.21, 81.66, 67.69, 57.61, 39.02, 36.67, 32.52,
31.82, 29.67, 29.43, 29.27, 29.23, 29.19, 29.12, 29.05, 27.62,
27.13, 22.60, 20.61, 14.03. IR (neat) 3462 (OH), 1750
(C=O), 1670 (C=C) cm–1. Anal. C27H46O5 (C, H), MS(FAB)
m/z 451 (MH+).
References
[1] D.S. Lester, R.M. Espand (Eds.), Protein Kinase C. Current and
Future Perspectives, Ellis Horwood, New York, 1992.
[2] J.F. Kuo (Ed.), Protein Kinase C, Oxford University Press, NewYork,
1994.
[3] A.C. Newton, J. Biol. Chem. 270 (1995) 28495–28498.
[4] N.E. Sharkey, K.L. Leach, P.M. Blumberg, Proc. Natl.Acad. Sci. USA
81 (1984) 607–610.
[5] V.E. Marquez, K. Nacro, S. Benzaria, J. Lee, R. Sharma, K. Teng,
G.W.A. Milne, B. Bienfait, S. Wang, N.E. Lewin, P.M. Blumberg,
Pharmacol. Ther. 82 (1999) 251–261.
[6] J. Lee, R. Sharma, S. Wang, G.W.A. Milne, N.E. Lewin, Z. Szallazi,
P.M. Blumberg, C. George, V.E. Marquez, J. Med. Chem. 39 (1996)
36–45 (and preceding papers).
[7] K. Nacro, B. Bienfait, J. Lee, K.-C. Han, J.-H. Kang, S. Benzaria,
N.E. Lewin, D.K. Bhattacharyya, P.M. Blumberg, V.E. Marquez, J.
Med. Chem. 43 (2000) 921–944.
[8] H. Nakamura, Y. Kishi, M.A. Pajares, R.R. Rando, Proc. Natl. Acad.
Sci. USA 86 (1989) 9672–9676.
[9] R.R. Rando, Y. Kishi, Biochemistry 31 (1992) 2211–2218.
[10] S. Wang, G.W.A. Milne, M.C. Nicklaus, V.E. Marquez, J. Lee,
P.M. Blumberg, J. Med. Chem. 37 (1994) 1326–1338.
[11] G. Zhang, M.G. Kazanietz, P.M. Blumberg, J.H. Hurley, Cell 81
(1995) 917–924.
[12] M. Tanaka, K. Irie, Y. Nakagawa, Y. Nakamura, H. Ohigashi,
P.A. Wender, Bioorg. Med. Chem. Lett. 11 (2001) 719–722.
[13] S. Benzaria, B. Bienfait, K. Nacro, S. Wang, N.E. Lewin, M. Beheshti,
P.M. Blumberg, V.E. Marquez, Bioorg. Med. Chem. Lett. 8 (1998)
3403–3408.
Compound 8E was obtained from 20E, by following the
above procedure, as an oil in 92% yield: [␣]D = +11.62 (c
0.74, CHCl3); H NMR (CDCl3) d 6.72 (m, 1H, >C=CH),
1
5.33 (m, 2H, CH2CH=CHCH2), 4.18 (s, 2H, H-5a,b), 3.82 (t,
2H, J = 6.2 Hz, HOCH2), 2.78 (m, 2H, H-3a,b), 2.04 (s, 3H,
CH3COO), 1.9–2.2 (m, 8H, ArOCH2CH2, >C=CHCH2– and
CH2CH=CHCH2), 1.15–1.5 (m, 22H), 0.86 (distorted t, 3H,
CH3); 13C NMR (CDCl3) d 170.43, 170.02, 141.13, 129.92,
129.54, 126.29, 82.38, 67.89, 57.42, 39.19, 33.19, 32.47,
31.77, 30.12, 29.62, 29.59, 29.52, 29.39, 29.28, 29.19, 29.03,
28.00, 27.09, 27.02, 22.55, 20.51, 13.99. IR (neat) 3448
(OH), 1748 (C=O), 1679 (C=C) cm–1. Anal. C27H46O5 (C,
H), MS(FAB) m/z 451 (MH+).
4.2. Molecular modeling
The structures of 3, 4, 7Z and 8Z were built using the
Sybyl molecular modeling program (Tripos, Inc., St. Louis),
and then the geometries were fully optimized using the Tri-
pos force field with the following non-default options
(method: conjugate gradient, termination: gradient 0.01 kcal
mol Å, and max iterations: 10,000). The partial atomic
charges were calculated by the Gasteiger–Hückel method in
the Sybyl program. The Grid search method was used to
evaluate the conformational properties of these compounds.
Torsion angles for the hydroxyalkyl and acetate groups were
driven from –180° to 180° in steps of 30° and fully mini-
mized structures were obtained at each point.All calculations
[14] M. Sodeoka, K. Uotsu, M. Shibasaki, Tetrahedron Lett. 36 (1995)
8795–8798.
[15] V.E. Marquez, R. Sharma, S. Wang, N.E. Lewin, P.M. Blumberg,
I.-S. Kim, J. Lee, Bioorg. Med. Chem. Lett. 8 (1998) 1757–1762.
[16] K. Teng, V.E. Marquez, G.W.A. Milne, J.J. Barchi Jr, M.G. Kazanietz,
N.E. Lewin, P.M. Blumberg, E. Abushanab, J. Am. Chem. Soc. 114
(1992) 1059–1070.
[17] R. Sharma, V.E. Marquez, G.W.A. Milne, N.E. Lewin, P.M. Blum-
berg, Bioorg. Med. Chem. Lett. 3 (1993) 1993–1998.
[18] J. Lee, V.E. Marquez, N. Lewin, P.M. Blumberg, Synlett (1994)
206–208.