Conformationally constrained analogues of diacylglycerol. 12. Ultrapotent protein kinase C ligands based on a chiral 4,4-disubstituted heptono-1,4- lactone template

Article abstract of DOI:10.1021/jm950278f

Conformationally constrained analogues of diacylglycerol (DAG) built on a 5-[(acyloxy)methyl [-5-(hydroxymethyl)tetrahydro-2-furanone template (1, Chart 1) were shown previously to bind tightly to protein kinase Cα (PK- Cα) in a stereospecific manner. The

Full text of DOI:10.1021/jm950278f

36
J . Med. Chem. 1996, 39, 36-45
Con for m a tion a lly Con str a in ed An a logu es of Dia cylglycer ol. 12.1 Ultr a p oten t
P r otein Kin a se C Liga n d s Ba sed on a Ch ir a l 4,4-Disu bstitu ted
Hep ton o-1,4-la cton e Tem p la te
J eewoo Lee,† Rajiv Sharma,† Shaomeng Wang,† George W. A. Milne,† Nancy E. Lewin,‡ Zoltan Szallasi,‡
Peter M. Blumberg,‡ Clifford George,§ and Victor E. Marquez*,†
Laboratory of Medicinal Chemistry, Developmental Therapeutics Program, Division of Cancer Treatment, and Molecular
Mechanisms of Tumor Promotion Section, Laboratory of Cellular Carcinogenesis and Tumor Promotion, National Cancer
Institute, National Institutes of Health, Bethesda, Maryland 20892, and Laboratory for the Structure of Matter, Naval
Research Laboratory, Washington, D.C. 20375
Received April 12, 1995X
Conformationally constrained analogues of diacylglycerol (DAG) built on a 5-[(acyloxy)methyl]-
5-(hydroxymethyl)tetrahydro-2-furanone template (1, Chart 1) were shown previously to bind
tightly to protein kinase CR (PK-CR) in a stereospecific manner. These compounds, however,
racemized readily through rapid acyl migration and lost biological potency. In order to
circumvent this problem, the “reversed ester” analogues were designed as a new set of PK-C
ligands. This reversal of the ester function produced some new DAG mimetics that are
embedded in a C-4 doubly-branched heptono-1,4-lactone template. The reversed ester analogues
were impervious to racemization, and their chemically distinct branches facilitated the
enantiospecific syntheses of all targets. Compound 2, the simplest reversed ester analogue of
1 (Chart 1), exhibited a 3.5-fold reduction in binding affinity toward PK-CR which we attributed
to the loss of a stabilizing gauche interaction that caused the ester branch in 2 to be more
disordered than in the normal ester 1. However, conversion of the propanoyl branch of 2 into
a propenoyl branch restored binding affinity (3 versus 5). As expected, the compounds bound
to the enzyme with strict enantioselectivity (3 and 5 versus 4 and 6). Functionalization of the
propenoyl-branched compounds as R-alkylidene lactones, in a manner which proved successful
with the 5-[(acyloxy)methyl]-5-(hydroxymethyl)tetrahydro-2-furanone template (9 and 10),
produced stable compounds with equivalent ultrapotent binding affinities for PK-CR (7 and
8). The additional incorporation of the propenoyl-branched carbonyl into a γ-lactone ring was
performed (11-14) not only to derive a possible additional entropic advantage but also to confirm
the spatial disposition of this carbonyl function in the ligand-enzyme complex. Although no
additional entropic advantage was derived, the high binding affinities displayed by compounds
11 and 12 helped to establish the correct orientation of the equivalent carbonyl group in PK-
C-bound DAG. As expected, these DAG analogues activated PK-CR. The most potent agonist,
compound 8, stimulated phosphorylation of the R-pseudosubstrate peptide, and in primary
mouse keratinocytes it caused inhibition of binding of epidermal growth factor with an ED50 of
approximately 1 µM. In contrast to the phorbol esters, compound 8 did not induce acute edema
or hyperplasia in skin of CD-1 mice, and its pattern of downregulation with several PK-C
isozymes was different from that of phorbol 12-myristate 13-acetate (PMA).
In tr od u ction
Such a racemization was accompanied by a reduction
in PK-C binding affinity.1
In the two previous papers we showed that semirigid
diacylglycerol (DAG) analogues constructed on a 5-[(acyl-
oxy)methyl]-5-(hydroxymethyl)tetrahydro-2-furanone
template (1) possess unprecedented, high affinity levels
for protein kinase CR (PK-CR).1,2 It was also deter-
mined that potent binding affinity toward PK-CR was
associated with only one enantiomeric form of this
lactone.1 Unfortunately, a major limitation associated
with the use of compounds built on this chiral template
was their ease of racemization caused by the facile
migration of the acyl group under very mild conditions.1
Our desire to prepare stable enantiomers using this
newly discovered template led us to consider performing
chemical changes to help differentiate both branches of
the lactone. Since we have previously shown that the
integrity of the ester function is critical for high PK-C
binding affinity,3 the reversal of the ester function was
considered as a means of achieving the desired chemical
differentiation between both branches of the lactone.
Hence, compound 2, which represents the simplest
“reversed ester” analogue of 1, was considered as a
target (Chart 1). The impact of the reversal of the ester
function on PK-C binding was promptly assessed first
with racemic forms of 1 and 2 since such a seemingly
subtle change, as ester reversal, was expected to have
an important effect on the conformation of the molecule.
For example, we were aware that an important and
stabilizing interaction in compound 1, represented by
the attractive gauche interaction between the furan
oxygen and the ether oxygen of the ester branch,4 was
† Laboratory of Medicinal Chemistry, Developmental Therapeutics
Program, Division of Cancer Treatment, National Cancer Institute,
National Institutes of Health.
‡ Molecular Mechanisms of Tumor Promotion Section, Laboratory
of Cellular Carcinogenesis and Tumor Promotion, National Cancer
Institute, National Institutes of Health.
§ Laboratory for the Structure of Matter, Naval Research Labora-
tory.
X Abstract published in Advance ACS Abstracts, November 1, 1995.
This article not subject to U.S. Copyright. Published 1996 by the American Chemical Society

Conformationally Constrained Analogues of Diacylglycerol. 12
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 37
Ta ble 1. Apparent Ki for Racemic and Enantiomeric Ligands as Inhibitors of PDBU Binding to PK-C
Ta ble 2. Apparent Ki for Enantiomeric Ligands Having an
R-Alkylidene Chain as Inhibitors of PDBU Binding to PK-C
[R ) (Z)-CH3(CH2)7CHdCH(CH2)7]
absent in compound 2. Therefore, the reversed ester
branch in 2, which lacks the stabilizing gauche effect,
was expected to be more disordered than the corre-
sponding ester branch in 1. This expectation was borne
Ch a r t 1
out by the 3.5-fold loss in affinity observed for the
reversed ester 2 (vide infra). Supported by molecular
modeling studies and with the aid of the phorbol ester
pharmacophore model developed earlier,5 we considered
restoring some of the original conformational bias of the
ester branch induced by the gauche interaction in 1, by
the addition of a double bond to compound 2. On the
basis of these considerations, four compounds with
similar lipophilicities (3-6, Table 1) were selected as
initial targets for enantioselective syntheses. On the
basis of results obtained in our previous studies with
the active (R)-enantiomer of 1,1 the chiral lactones 3 and
5 were predicted to behave as high-affinity ligands for
PK-C, whereas lactones 4 and 6 were predicted to be
either ineffective or weakly binding. In addition, the
difference in binding affinity between lactones 3 and 5
was expected to reflect to what degree the double bond
was indeed capable of compensating for the loss of the
gauche effect.
An additional element that was contemplated in an
attempt to transform these compounds into ultrapotent
ligands consisted of downsizing the acyl function to a
two carbon acetyl chain and relocating the alkyl chain
to a more restricted position on the lactone ring.
Therefore, R-alkylidine lactones 7 and 8 (Table 2) were
selected as additional targets which feature an unsat-
uration in the middle of the alkyl chain. The antici-
pated “bend” caused by the cis double bond in the middle
of the alkyl chain in these compounds was intended to
mimic oleic acid, since oleoate esters of related lactone
ligands have been shown to have enhanced PK-C
binding affinities relative to the comparable saturated
esters.6 The new doubly-branched heptono-1,4-lactones
7 and 8 (Table 2) with an R,â-unsaturated methyl ester
branch represent the two most potent and stable semi-
rigid DAG analogues synthesized to date (vide infra).
The PK-C binding affinities of these compounds es-
sentially reproduced the values obtained previously for
the chiral 5-[(acyloxy)methyl]-5-(hydroxymethyl)tet-
rahydro-2-furanones 9 and 101 (Table 2) and confirmed
the bioisosteric equivalence of both templates.7
With the aid of the phorbol ester pharmacophore,5 a
conformational analysis of compounds 7 and 8 raised
the possibility of restricting even further the orientation
of the ester carbonyl group. This exercise was not only
intended to derive the possible benefits of an additional
entropic advantage, but it was additionally expected to
confirm the correct orientation of this carbonyl group
in the ligand-enzyme complex. To achieve this goal,
this ester function was incorporated into a second
lactone ring which provided two possible orientations
for the carbonyl relative to the double bond. Of the four
target compounds (11-14, Table 3) containing two
γ-lactone pharmacophores, compounds 11 and 12 were
predicted to have the two lactones in the correct
orientation in agreement with the phorbol ester phar-
macophore model.5 PK-C binding studies indicated that
although no additional entropic advantage was derived,
the high binding affinities displayed by compounds 11
and 12 helped to establish the correct orientation of the
equivalent carbonyl group in PKC-bound DAG.

38 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Lee et al.
Ta ble 3. Apparent Ki for Enantiomeric Ligands with Two
γ-Lactone Pharmacophores as Inhibitors of PDBU Binding to
PK-C [R ) (Z)-CH3(CH2)7CHdCH(CH2)7]
Sch em e 2
Sch em e 1
The two R-alkylidene targets 7 and 8 were also
prepared from intermediate 191 which reacted with
oleoyl aldehyde to give a mixture of geometric isomers
(21) after dehydration. A similar sequence of steps to
those employed for the transformation of 19 to 20 was
used to convert 21 to 22, which was isolated as a
mixture of geometric isomers. Reaction 22 with methyl
(triphenylphosphoranylidene)acetate gave the corre-
sponding final targets which were chromatographi-
cally separated as the E-(7) and Z-(8) isomers, respec-
tively.
Since building of compounds with the “inactive”
template was intended merely to corroborate the iden-
tity of the “active” template, our synthetic effort toward
this objective was limited to the preparation of the
simpler analogues 4 and 6 (Table 1), which correspond,
respectively, to the optical antipodes of lactones 3 and
5. For this synthesis, a similar strategy to that used
for the construction of chiral lactone 191 was performed
starting from D-arabinose (Scheme 3). Protection of this
sugar as the 5-O-(tert-butyldimethylsilyl)-1,2-O-isopro-
pylidene-â-D-arabinofuranose (23) was followed by chro-
mium trioxide oxidation to the keto sugar 24.9 Spiro-
lactonization via the diol 25 was performed using PCC
oxidation, as reported previously,1 to give the 2-keto-
1,7-dioxaspiro[4.4]nonane intermediate 26. In this
instance, the addition of the Grignard reagent to give
25 occurred stereospecifically from the less hindered
R-side, thus ensuring the desired stereochemical out-
come. As before, removal of the acetonide and meta-
periodate cleavage of the resulting glycol moiety pro-
vided lactone 27. Wittig olefination of 27 with myristyl
(triphenylphosphoranylidene)acetate gave the R,â-
unsaturated ester 28 which required shortening of one
of the branches by one carbon atom. This one-carbon
Ch em istr y
Synthesis of the racemic reverse ester lactone 2 was
straightforward and was performed according to Scheme
1. Partial hydrolysis of diethyl 4-oxopimelate (15) with
NaOH gave the half-ester 16 which was condensed with
myristyl alcohol to give the mixed diester 17. A Wittig
olefination reaction of 15 with methyltriphenylphos-
phonium bromide/t-BuOK gave the expected olefin 18
which after cis-hydroxylation with OsO4 cyclized uni-
directionally to the desired target lactone 2.
The enantioselective synthesis of lactones 3, 5, 7 and
8, all containing the “active” chiral template, started
with the same intermediate 191 that was used in the
preceding paper for the synthesis of the active enanti-
omer of 1 (Scheme 2). Conversion of 19 to 20, which
was also reported in the previous paper,1 was followed
by a reaction of the resulting aldehyde with myristyl
(triphenylphosphoranylidene)acetate to give the corre-
sponding target R,â-unsaturated ester 5. This Wittig
reagent, myristyl (triphenylphosphoranylidene)acetate,
was prepared in three steps from R-bromoacetic acid.8
Catalytic hydrogenation of the double bond in 5 pro-
duced the saturated branched target 3.

Conformationally Constrained Analogues of Diacylglycerol. 12
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 39
Sch em e 3
from each individual isomer provided the desired target
compounds.
Biologica l Resu lts
The affinities of ligands 2 (racemic) and 3-14 (enan-
tiomeric) for PK-CR, expressed in terms of their abilities
to displace bound [3H-20]phorbol 12,13-dibutyrate
(PDBU), were measured in the same manner as de-
scribed in the preceding papers.1,2 Analysis of the data
indicated that the reversed ester racemic ligand 2 (Ki
) 489 nM, Table 1) suffered a 3.5-fold loss in binding
affinity relative to racemate 1 (Ki ) 138 nM).1,2 This
loss in binding affinity was consistent with the presence
of a more disordered side chain in 2 and the associated
entropic penalty that had to be paid during its binding
to PK-C. Relative to the enantiomeric compounds
(Table 1), the predicted active ligands 3 (Ki ) 260 nM)
and 5 (Ki ) 109 nM) showed a direct structural cor-
respondence7 to the active (R)-enantiomer of 1 (Ki ) 96
nM)1 that was built on a 5-[(acyloxy)methyl]-5-(hy-
droxymethyl)tetrahydro-2-furanone template. Consis-
tent with our predictions, the corresponding optical
antipodes 4 and 6 were respectively 100-200-fold less
potent. A direct comparison between compound 5 (Ki
) 109 nM) and the active enantiomer of 1 reported
earlier (Ki ) 96 nM)1 demonstrated that within experi-
mental error, the more rigid R,â-unsaturated ester
branch in 5 was able to recover the loss in binding
affinity that resulted from the reversal of the ester
function. As we shall see in the molecular modeling
section, compounds built on a 5-[(acyloxy)methyl]-5-
(hydroxymethyl)tetrahydro-2-furanone template (e.g.,
compounds 1, 9, and 10) show, in their presumed bound
conformation, the furan oxygen and the ether oxygen
of the ester function in a gauche disposition. This
relative disposition of oxygens happens to correspond
to the most energetically favored rotamer in these
molecules. Rather fortuitously, the double bond in 5
appears capable of steering the carbonyl ester pharma-
cophore in the same direction (vide infra) as in the
stable gauche rotamer of 5-[(acyloxy)methyl]-5-(hy-
droxymethyl)tetrahydro-2-furanones. The same phe-
nomenon was observed for the more potent R-alkylidene
analogues 7 (Ki ) 20 nM) and 8 (Ki ) 11 nM) which
proved to be equivalent to the corresponding R-alkyli-
denes 91 and 101 built on a chiral 5-[(acyloxy)methyl]-
5-(hydroxymethyl)tetrahydro-2-furanone template (Table
2). Compounds 7 and 8 are to date the most potent and
stable DAG mimics.
Sch em e 4
The incorporation of the methyl ester group of com-
pounds 7 and 8 into a γ-lactone ring allowed us to
predict the correct relative disposition of both carbonyl
pharmacophores in the ligand-enzyme complex. In-
deed, compounds 11 (Ki ) 26 nM) and 12 (19 nM) were,
respectively, 30-100-fold more potent than their iso-
mers 13 and 14 (Table 3). The fact that 11 and 12 did
not show an increase in binding, relative to the more
flexible analogues 7 and 8, is probably indicative of a
slight deviation from the ideal conformation that results
from unfavorable steric contacts between the two γ-lac-
tone rings (see molecular modeling section).
shortening was accomplished, after removal of the silyl
ether protection, by a second metaperiodate cleavage of
the glycol moiety in 28, followed by sodium borohydride
reduction to give the target compounds 4 and 6.
For the syntheses of the bis-γ-lactones (Scheme 4),
aldehyde 22 was reacted with R-(triphenylphosphora-
nylidene)-γ-butyrolactone10 to give a mixture of all four
isomers (11-14). Chromatographic separation of these
isomers was possible only after protection of the primary
alcohol function as the tert-butyldimethylsilyl ether
(compounds 29-32). Removal of the silyl ether group
The biological activity of compound 8 was evaluated
further. First, we sought to determine whether the
binding of compound 8 to PK-CR led to activation of the
enzyme. We found that, as expected for a DAG ana-

40 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Lee et al.
F igu r e 1. Stimulation of PK-CR by compound 8. Points
represent the mean ( range of duplicate measurements in a
single experiment. A second experiment gave similar results.
F igu r e 2. Superposition of a low-energy conformer of 10 on
PDBU. The critical pharmacophore atoms (see text) are
highlighted in boxes.
logue, compound 8 stimulated phosphorylation of the
R-pseudosubstrate peptide (a standard substrate for
assay of PK-C) with an ED50 of 163 ( 17 nM (Figure
1). This finding suggests that 8 would function in the
intact cell or organism as either an agonist or a partial
antagonist. In primary mouse keratinocytes, compound
8 caused inhibition of binding of epidermal growth factor
with an ED50 of approximately 1 µM (data not shown),
which is a typical response to PK-C activation.11 In
contrast to the phorbol esters, compound 8 did not
induce acute edema (10-100 µg, 6 h) or hyperplasia
(10-100 µg, 72 h) in skin of CD-1 mice. Finally, the
ability of compound 8 to down-regulate PK-CR and PK-
Cδ in primary mouse keratinocytes after 24 h was
examined. We found that compound 8 down-regulated
the levels of PK-CR and -δ with an ED50 of ap-
proximately 300 nM, whereas it did not down-regulate
the levels of PK-Cꢀ or PK-Cη. It is well-known that the
pattern of down regulation differs for PK-C ligands that
induce different sets of biological responses.12 The
pattern of down regulation of compound 8 thus distin-
guishes it from the typical phorbol ester, phorbol 12-
myristate 13-acetate (PMA), which has greater potency
for the down regulation of PK-Cδ than PK-CR (0.8
versus 40 nM at 6 h),11 or from 12-deoxyphorbol 13-
(phenyl acetate), which down regulates PK-Cꢀ.13 These
results are of further interest in that activity of PMA is
lost by 24 h, reflecting metabolism. This is not found
for compound 8. We conclude that compound 8 has a
different pattern of behavior from the typical phorbol
esters and may thus have unique utility.
X-r a y An a lysis a n d Molecu la r Mod elin g. The
critical functional pharmacophores in the phorbol esters
that are responsible for PK-C recognition are thought
to be the C-20 hydroxyl, the C-3 carbonyl, and the C-9
hydroxyl (PDBU, Chart 2).5 The latter two appear to
function as hydrogen bond acceptors, while the primary
alcohol at C-20 functions as a hydrogen bond donor.
Correspondingly in DAG, as well as in the lactone
surrogates, the two carbonyls behave as hydrogen bond
acceptors and the primary alcohol as a hydrogen bond
donor (Chart 2).5 The superposition of PDBU on the
lowest energy conformer of the potent R-alkylidene
Z-isomer 10,1 which was constructed earlier on a
5-[(acyloxy)methyl]-5-(hydroxymethyl)tetrahydro-2-fura-
none template, produced in an excellent fit (Figure 2,
rms ) 0.390 Å) of the critical pharmacophores. Such
an optimal fit can be reached only when the lactone
carbonyl is made to correspond directly to the C-9
hydroxyl of phorbol, and the side chain ester carbonyl
overlays the C-3 carbonyl of phorbol (Chart 2). In this
Ch a r t 2
low-energy conformation, presumably the conformation
of 10 most likely to bind PK-C, the furan oxygen and
the ether oxygen of the ester branch are in a stable +sc
gauche disposition (Figure 3a). The fact that this
energetically preferred rotamer shows an excellent fit
to phorbol may explain the compound’s exceptional
affinity for PK-C (Table 2). Although we were unable
to crystallize 10, we succeeded in solving the X-ray
crystal structure of a related R-alkylidene racemate (()-
33 (Chart 2) synthesized earlier.2 This compound differs
from 10 in that it has a tetradecanylidene alkyl chain
instead of a (Z)-9-octadecenylidene alkyl chain. From
the unit cell, which contained both enantiomers, the (R)-
enantiomer was selected and its structure was compared
with that of the already described minimized structure
of 10 (Figure 3a,b). The crystal structure of the (R)-
isomer of 33 shows an analogous attractive gauche
interaction for the ester side chain, but corresponding,
instead, to that of the alternate ap gauche rotamer,
which was probably favored by crystal packing forces.
The rest of the atoms in both lactone rings remained
virtually unchanged. This comparison suggests that for
these molecules there is a definitive preference for either
one of these gauche rotamers, one of which directs the
essential pharmacophores in the correct orientation. The

Conformationally Constrained Analogues of Diacylglycerol. 12
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 41
F igu r e 3. (a) Low energy conformer of 10 as in Figure 2
showing the +sc gauche disposition of the oxygens at the C-5
branch. (b) X-ray structure of 10 as it exists in the unit cell of
a racemic crystal showing the ap gauche disposition of the
oxygens at the C-5 branch.
F igu r e 5. Superposition of a low energy conformer of 12 on
PDBU. The critial pharmacophore atoms (see text) are high-
lighted in boxes.
and probably help orient the hydrophilic part of the
molecule once the molecule becomes inserted into the
lipid bilayer. When the simpler compounds that lack
the R-alkylidene moiety (e.g., compounds 3 and 5) were
modeled and superimposed on PDBU (data not shown),
the different location of the lipophilic chain changes the
manner in which the pharmacophores relate to one
another. Assuming that the side chain’s role is indeed
to orient the molecule into the lipid bilayer, and assum-
ing that the orientation of the acyl chains in PDBU is
the correct one, the superposition of 3 and 5 onto PDBU
would force the long acyl ester carbonyl to correspond
to the C-9 hydroxyl of phorbol and the lactone carbonyl
to correspond to the C-3 carbonyl of phorbol. This
alternative correspondence leads to a poorer fit when
compared to the fit described above for compounds 8
and 10. Hence, the aliphatic and hydrophobic portion
of the molecule, when attached to the lactone ring,
constrains the lactone carbonyl to match exclusively
with the C-9 hydroxyl group of phorbol producing a
better fit (vide supra). This may explain the 10-fold
increase in binding affinity observed with the R-alkyli-
dene lactones. These observations also underscore the
important role of the side chain not in only providing
the required lipophilicity, but also in steering the
compound into the lipid bilayer and thus contributing
with an additional entropic advantage during the bind-
ing process.
F igu r e 4. Superposition of a low energy conformer of 8 on
PDBU. The critical pharmacophore atoms (see text) are
highlighted in boxes.
energy difference in vacuo between the two gauche
conformers was 1.37 kcal/mol in favor of the rotamer
with the correct orientation.
The biological equivalence between compounds 8 and
10 (Ki ca. 11 nM, Table 2) suggests that the double bond
in compound 8 is indeed capable of directing its side
chain ester into the equivalent space occupied by the
side chain ester of the +sc gauche rotamer of 10. This
can be appreciated in Figure 4, where the most stable
conformer of 8 shows an excellent fit to PDBU (rms )
0.350 Å). Also, the superposition of compound 12 and
PDBU (Figure 5, rms ) 0.710 Å) confirms the model’s
prediction regarding the correct position of the side
chain ester carbonyl required for effective binding.
An interesting qualitative observation regarding the
orientation of the alkyl chain in compounds 8, 10 and
12 (Figures 2, 4, and 5) is how well they match the
general direction of the acyl chains in PDBU. Although
the acyl chains in the phorbol esters and DAG are not
part of the pharmacophore units usually defined for
PK-C recognition, they are essential for PK-C binding
Exp er im en ta l Section
Gen er a l Exp er im en ta l. All chemical reagents were com-
mercially available. Melting points were determined on a Mel-
Temp II apparatus, Laboratory Devices, USA, and are uncor-
rected. Silica gel column 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
solvent 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. Elemental analyses were performed by
Atlantic Microlab, Inc., Atlanta, GA.
An a lysis of In h ibition of [3H]P DBU Bin d in g by Non -
r a d ioa ctive Liga n d s. Enzyme-ligand interactions were
analyzed by competition with [3H]PDBU binding essentially

42 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Lee et al.
as described in our previous work,14 except that the PK-C
preparation used here was the single isozyme PK-CR. This
recombinant PK-CR was expressed in the baculovirus system
and was isolated as described in ref 15. The ID50 values were
determined from the competition curves, and the correspond-
ing Ki values for the ligands were calculated from the ID50
values as described before.14,15 Values represent the mean (
standard error (three determinations).
CO), 3.95-4.10 (m, 4H, CH3(CH2)11CH2CH2OCO, CH3CH2O);
13C NMR (CDCl3) δ 14.05, 14.10, 22.63, 25.83, 27.92, 28.53,
29.19, 29.30, 29.46, 29.52, 29.59, 31.87, 37.04, 60.55, 64.81,
172.61, 172.69, 206.92. Anal. (C23H42O5) C, H.
Tetr adecyl 4-[(Eth oxycar bon yl)eth yl]-4-pen ten oate (18).
Potassium tert-butoxide (0.195 g, 1.73 mmol) was added to a
stirred suspension of methyl triphenylphosphonium bromide
(0.618 g, 1.73 mmol) in dry benzene (8 mL), which was
maintained under argon at room temperature. After stirring
for 30 min, the reaction mixture was cooled to 0 °C and a
solution of ketone 17 (0.46 g, 1.16 mmol) in dry THF was added
rapidly. The reaction was quenched after 60 min at 0 °C with
brine (∼10 mL), the mixture was extracted with EtOAc (3 ×
20 mL), and the combined organic layer was washed with
water (1 × 10 mL) and dried (Na2
Molecu la r Mod elin g. The methods used in the confor-
mational search and analysis for compounds 2-6 were the
same as those described in the preceding paper. However, for
compounds 7-14, which have fewer rotatable bonds within
their designated pharmacophore, a systematic conformational
search approach was applied with an angular increment of 60°
in order to better access the conformational space for these
compounds. The same program described in the preceding
paper1 was used to automatically determine the fit of each
resulting conformation to phorbol and to identify the low-
energy conformations with the best fit. The low-energy
conformations with the best fit were further minimized by a a
semiempirical quantum mechanics method, AM1 in MO-
PAC6.0 with a precision specified keyword GNORM ) 0.01
running on a host mainframe Convex C240 supercomputer,
in order to validate the molecular mechanics results obtained
with CHARMm. The results obtained with CHARMm and
AM1 for compounds 8, 10, and 12 were also validated by ab
initio quantum mechanics, in the level HF/3-21G, using
GAUSSIAN 90 running on a Cray-YMP supercomputer of the
NCI Biomedical Supercomputing Center. The results from
semiempirical and ab initio methods essentially agreed with
the CHARMm results.
SO4). The residue obtained
after concentration was purified by flash column chromatog-
raphy over silica gel using hexane/EtOAc (95:5), giving 18 as
a liquid (0.72 g, 71.5%): IR (neat) 1738 (CdO) and 1647 cm-1
;
1H NMR (CDCl3) δ 0.85 (distorted t, 3H, CH3(CH2)13), 1.20-
1.40 (m, 25 H, CH3(CH2)11CH2CH2OCO, CH3CH2OCO), 1.60
(m, 2 H, CH3(CH2)11CH2CH2OCO), 2.35 (m, 4 H, OCCH2CH2-
(CdCH2)CH2CH2CO), 2.45 (m, 4 H, OCCH2CH2(CdCH2)CH2-
CH2CO),4.00-4.15 (m, 4H, CH3(CH2)11CH2CH2OCO, CH3CH2O),
4.75 (s, 2 H, dCH2); 13C NMR (CDCl3) δ 14.09, 14.20, 22.66,
25.89, 28.60, 29.23, 29.33, 29.50, 29.56, 29.62, 31.00, 31.03,
31.89, 32.64, 60.35, 64.61, 109.75, 146.37, 173.08, 173.18.
Anal. (C24H44O4) C, H.
(()-5-(Hyd r oxym eth yl)-5-[2-(tetr a d eca n oxyca r bon yl)-
eth yl]tetr a h yd r o-2-fu r a n on e (2). A stirred solution of olefin
18 (0.117 g, 0.29 mmol) in t-BuOH (3 mL) and water (1 mL)
at 0 °C, containing N-methylmorpholine N-oxide (0.069 g, 0.59
mmol), was treated with 2.5% aqueous OsO4 solution (0.16
mL). The reaction mixture was brought to room temperature
slowly and stirred for 12 h; Na2SO3 (0.20 g) was added, and
after stirring for a few more minutes it was concentrated under
vacuum. The residue was extracted with EtOAc (3 × 10 mL),
and the combined extract was washed with water (10 mL) and
dried (Na2SO4). The residue obtained after evaporation of the
solvent was purified by flash column chromatography over
silica gel using hexane/EtOAc (3:2) as eluant to give the title
compound 2 (0.072 g, 65%) as white solid: mp 60-61 °C
X-r a y An a lysis. Crystal data for (()-33: C22H38O5, FW )
382.52, mp 58-59 °C. Triclinic space group, P1h, a ) 5.301(2)
Å, b ) 5.506(1) Å, c ) 43.520(9) Å, R ) 92.62(2)°, â ) 90.94-
(2)°, γ ) 113.67(2)°, V ) 1161.3 (5) Å3, Z ) 2, Dc ) 1.094 mg
mm-3, λ(Cu KR) ) 1.541 78 Å, µ ) 0.61 mm-1, F(000) ) 420,
T ) 293 K. Final residuals were R ) 0.080 for 2040 reflections
I > 2σ(I0). Tables of atomic coordinates, bond distances and
angles, and anisotropic thermal parameters have been depos-
ited.
4-Oxoh ep ta n ed ioic Acid , Mon oeth yl Ester (16).
A
stirred solution of diethyl 4-oxopimelate (4.81 g, 20.9 mmol)
in 95% EtOH (100 mL) was treated with a solution of NaOH
(0.835 g, 20.9 mmol) in water (30 mL) at room temperature.
After stirring for 4 h, the reaction mixture was concentrated
under suction, acidified to pH ≈ 2 with 1 N HCl, and extracted
with EtOAc (3 × 30 mL). The combined organic extract was
washed with water (2 × 50 mL), dried (Na2SO4), and concen-
trated. The oil obtained was purified by flash column chro-
matography over silica gel using EtOAc/hexane (1:1) as eluant
which provided 1.35 g of recovered starting material (15). Upon
further elution with EtOAc, the title compound 16 was isolated
as a white solid (1.44 g, 47%, based on recovered starting
material): mp 68.5-69.5 °C (EtOAc/hexane); IR (CH2Cl2) 1731
(EtOAc/hexane); IR (KBr) 3380 (OH) and 1728 (CdO) cm-1
;
1H NMR (CDCl3) δ 0.85 (distorted t, 3H, CH3(CH2)13), 1.10-
1.40 (m, 22 H, CH3(CH2)11CH2CH2OCO), 1.60 (m, 2 H,
CH3(CH2)11CH2CH2OCO), 1.86-2.12 (m, 4 H, H-4a,b, CH2CH2-
COOC14H29), 2.25-2.78 (multiplets, 4 H, H-3a,b, CH2CH2-
COOC14H29), 2.90 (t, J ) 6.5 Hz, 1 H, OH), 3.50 (dd, J ) 12.1,
6.5 Hz, 1 H, CHHOH), 3.65 (dd, J ) 12.1, 6.5 Hz, 1 H,
CHHOH), 4.05 (t, J ) 6.7 Hz, 2 H, CH3(CH2)11CH2CH2OCO);
13C NMR (CDCl3) δ 14.09, 22.65, 25.85, 28.05, 28.33, 28.51,
29.15, 29.21, 29.32, 29.48, 29.55, 29.61, 30.97, 31.88, 65.16,
66.41, 87.54, 173.28, 177.03; FAB MS m/ z (relative intensity)
385 (38, MH+), 171 (100, MH - C14H29OH). Anal. (C22H40O5)
C, H.
1
(CdO) and 1704 (CdO) cm-1; H NMR (CDCl3) δ 1.20 (t, J )
(S)-5-(Hyd r oxym eth yl)-5-[2-(tetr a d eca n oxyca r bon yl)-
(E)-eth en yl]tetr a h yd r o-2-fu r a n on e (5). A solution of 191
(0.214 g, 1 mmol) in THF (5 mL) was treated with 1 N HCl
solution (5 mL) and stirred at room temperature for 20 h. The
reaction mixture was neutralized with solid NaHCO3 and
diluted with EtOAc. The mixture was dried (MgSO4) and
concentrated to give the crude hemiacetal which was used for
the next step without further purification. This compound was
dissolved in a mixture of MeOH (10 mL) and water (5 mL),
and the solution was stirred with sodium metaperiodate (0.428
g, 2 mmol) for 2 h at room temperature. The reaction mixture
was filtered, and the filtrate was concentrated to dryness. The
residue was dissolved in EtOAc, dried (Na2SO4), and concen-
trated to give aldehyde 20, which was used for the next step
without further purification. Aldehyde 20 was dissolved in
benzene (30 mL) and treated with tetradecanyl (triphenylphos-
phoranylidene)acetate8 (0.775 g, 1.5 mmol). The reaction
mixture was refluxed for 2 h and concentrated. The residue
was filtered through a short pad of silica gel and washed with
ether, and the filtrate was concentrated to give a mixture of
formate ester and the free alcohol. The mixture was dissolved
in methanol (10 mL), cooled to 0 °C, and treated with
7.1 Hz, 3 H, CH3), 2.60 and 2.75 (multiplets, 8 H, HOOC(CH2)2-
CO(CH2)2COOCH2CH3), 4.10 (q, J ) 7.1 Hz, 2 H, COOCH2-
CH3); 13C NMR (CDCl3) δ 14.11, 27.65, 27.96, 36.80, 36.97,
37.05, 60.70, 172.71, 178.07, 206.80. Anal. (C9H14O5) C, H.
Tetr a d ecyl Eth yl 4-Oxop im ela te (17). A stirred solution
of 16 (0.784 g, 3.88 mmol), 1-tetradecanol (1.00 g, 4.66 mmol),
and DMAP (50 mg) in dry CH2Cl2 (25 mL) was kept under
argon at room temperature and treated with dicyclohexylcar-
bodiimide (5.82 mL, 1M solution in CH2Cl2). After stirring
for 24 h, the reaction mixture was quenched with a few drops
of CH3COOH and MeOH, stirred for a few more minutes, and
then concentrated under vacuum. The residue was diluted
with ether (50 mL), filtered, and concentrated. The residue
was purified by flash column chromatography over silica gel
using a gradient of 3-12% EtOAc in hexane, to give 17 (0.304
g, 60%) as a colorless solid: mp 42-43 °C (EtOAc/hexane); IR
1
(KBr) 1731 (CdO) and 1705 (CdO) cm-1; H NMR (CDCl3) δ
0.85 (distorted t, 3H, CH3(CH2)13), 1.20-1.40 (m, 25 H,
CH3(CH2)11CH2CH2OCO, CH3CH2OCO), 1.55 (m, 2 H, CH3-
(CH2)11CH2CH2OCO), 2.55 (distorted t, 4 H, OCCH2CH2-
COCH2CH2CO), 2.73 (distorted t, 4 H, OCCH2CH2COCH2CH2-

Conformationally Constrained Analogues of Diacylglycerol. 12
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 43
ammonium hydroxide solution (0.1 mL) with stirring for 20
min. The reaction mixture was quenched with acetic acid (0.1
mL) and concentrated. The residue was purified by flash
column chromatography over silica gel with EtOAc/hexanes
(3:2) as eluant to give 5 (0.250 g, 65% from 19) as a white
31.87, 32.25, 32.56, 51.89, 66.65, 83.83, 121.75, 125.03, 129.66,
130.01, 142.68, 145.24, 166.14, 169.73; FAB MS m/ z (relative
intensity) 449 (MH
+, 26). Anal. (C27H44O5) C, H.
Z-Isom er 8: oil; [R]22D -5.56° (c 0.36, CHCl3); IR (neat) 3440
(OH), 1761 and 1729 (CdO), 1666 cm-1 (CdC); 1H NMR
(CDCl3) δ 0.86 (distorted t, 3 H, CH3(CH2)7CHdCH(CH2)7-
CHdC<), 1.10-1.50 (m, 22 H, CH3(CH2)6CH2CHdCHCH2-
(CH2)5CH2CHdC<), 1.55 (br s, 1 H, OH), 2.00 (m, 4 H,
CH3(CH2)6CH2CHdCHCH2(CH2)5CH2CHdC<), 2.6-2.78 (m,
3 H, CH3(CH2)6CH2CHdCHCH2(CH2)5CH2CHdC<, H-4a), 3.12
(dm, J ) 16.0 Hz, 1 H, H-4b), 3.55-3.80 (m, 5 H, OCH3 and
CH2OH), 5.33 (m, 2 H, CH3(CH2)6CH2CHdCHCH2(CH2)5CH2-
CHdC<), 6.13 (d, J ) 15.6 Hz, 1 H, CH3OCOCHdCH), 6.22
(m, 1 H, CHdC<), 6.86 (d, J ) 15.7 Hz, CH3OCOCHdCH);
13C NMR (CDCl3) δ 14.09, 22.66, 27.19, 27.79, 28.98, 29.18,
29.30, 29.40, 29.50, 29.72, 29.74, 31.88, 32.58, 35.51, 51.88,
66.48, 83.30, 121.83, 123.03, 129.77, 129.95, 145.34, 146.20,
166.17, 168.44; FAB MS m/ z 449 (MH+, 10). Anal. (C27H44O5)
C, H.
solid: mp 58-59 °C; [R]22 -25.8° (c 1.0, CHCl3); IR (CHCl3)
D
3449 (OH), 1778 and 1752 cm-1 (CdO); 1H NMR (CDCl3) δ
0.86 (distorted t, 3 H, CH3(CH2)13), 1.10-1.40 (m, 22 H,
CH3(CH2)11CH2CH2OCO), 1.65 (m, 2 H, CH3(CH2)11CH2CH2-
OCO), 2.08 (m, 2 H, H-4a, OH), 2.42-2.74 (m, 3 H, H-4b, H-3a,b),
3.64 (dd, J ) 12.2, 7.1 Hz, 1 H, CHHOH), 3.79 (dd, J ) 12.2,
5.3 Hz, 1 H, CHHOH), 4.12 (t, J ) 6.7 Hz, 2 H, CH3(CH2)11-
CH2CH2OCO), 6.13 (d, J ) 15.7 Hz, 1 H, CHdCHCO), 6.85
(d, J ) 15.7 Hz, 1 H, CHdCHCO); 13C NMR (CDCl3) δ 14.09,
22.67, 25.88, 28.15, 28.40, 28.56, 29.23, 29.33, 29.49, 29.56,
29.63, 31.90, 65.16, 66.48, 86.98, 122.32, 144.47, 165.71,
176.11; FAB MS m/ z (relative intensity) 383 (MH+,18), 169
(100, MH - C14H29). Anal. (C22H38O5) C, H.
(R)-5-(Hyd r oxym eth yl)-5-[2-(tetr a d eca n oxyca r bon yl)-
eth yl]tetr a h yd r o-2-fu r a n on e (3). A solution of 5 (0.154 g,
0.4 mmol) in EtOAc (20 mL) was treated with 10% Pd-C (0.4
g) and hydrogenated under a hydrogen balloon for 2 h. The
reaction mixture was filtered, and the filtrate was concen-
trated. The residue was purified by flash column chromatog-
raphy over silica gel with EtOAc/hexanes (2:1) as eluant to
give the title compound 3 (0.151 g, 98%) as white solid: mp
5-O-(ter t-Bu tyld ip h en ylsilyl)-1,2-O-isop r op ylid en e-â-D-
th r eo-p en tofu r a n ose-3-u lose (24). This compound was
prepared in three steps from D-arabinose by the method of
Dahlman et al.9
5-O-(ter t-Bu t yld ip h en ylsilyl)-3-C-(3-h yd r oxyp r op yl)-
1,2-O-isop r op ylid en e-â-D-lyxofu r a n ose (25). A stirred so-
lution of 3-chloropropanol (2.127 g, 22.5 mmol) in dry THF
(20 mL) at -20 °C was treated dropwise with a solution of
methylmagnesium chloride (3M in THF, 7.5 mL) and stirred
for 20 min at the same temperature. The reaction mixture
was warmed to room temperature, and magnesium (0.82 g,
33.8 mmol) was added. The reaction mixture was refluxed for
2 h with periodic addition of dibromomethane (0.025 mL each)
at 0, 1, and 1.3 h intervals. After cooling to 0 °C, a solution of
ketone 24 (3.20 g, 7.5 mmol) in THF (25 mL) was added
dropwise. The reaction was quenched at 0 °C after 1 h by the
addition of saturated NH4Cl solution (25 mL). The layers were
separated, and the aqueous portion was extracted with EtOAc
(2 × 25 mL). The combined organic layer was dried (Na2SO4)
and concentrated. The residue obtained was purified by flash
chromatography over silica gel using EtOAc/hexane (1:1) to
60-61 °C; [R]22 +7.90° (c 1.0, CHCl3). The IR, 1H NMR, and
D
13C NMR were identical to those reported for the racemate 2.1
Anal. (C22H40O5) C, H.
(S)-5-(Hydr oxym eth yl)-5-[2-(m eth oxycar bon yl)-(E)-eth -
en yl]-3-{(E)-[(Z)-9-oct a d ecen ylid en e]}t et r a h yd r o-2-fu -
r a n on e (7) a n d (S)-5-(Hyd r oxym eth yl)-5-[2-(m eth oxy-
ca r bon yl)-(E)-eth en yl]-3-{(Z)-[(Z)-9-octa d ecen ylid en e]}-
tetr a h yd r o-2-fu r a n on e (8). A solution of 211 (0.463 g, 1
mmol) in THF (15 mL) was treated with 2 N HCl solution and
stirred for 5 days at room temperature. The reaction mixture
was then neutralized with solid NaHCO3 while being chilled
over ice, filtered, and concentrated. The residue was diluted
with EtOAc, dried (Na2SO4), and concentrated to give the
corresponding hemiacetal as an oil, which was used for the
next step without further purification. This compound was
dissolved in a mixture of MeOH (20 mL) and H2O (10 mL)
and treated with sodium metaperiodate (0.428 g, 2 mmol).
After stirring at room temperature for 4 h, the reaction
mixture was filtered and concentrated. The residue was
dissolved in ether, dried (Na2SO4), and concentrated to give
aldehyde 22, which was used for the next step without further
purification. Aldehyde 22 was dissolved in benzene (20 mL)
and stirred with methyl (triphenylphosphoranylidene)acetate
(0.067 g, 2.0 mmol) for 4 h. The solvent was evaporated, and
the residue was dissolved in ether and filtered through a short
pad of silica gel with additional ether. The filtrate was
concentrated to give a mixture of four products: the formate
esters and free alcohols of both E/ Z isomers. The mixture
was dissolved in methanol (5 mL), cooled to 0 °C, and treated
with ammonium hydroxide solution (0.05 mL) with stirring
for 20 min. The reaction mixture was quenched with acetic
acid (0.05 mL) and concentrated. The residue was purified
by flash column chromatography over silica gel with hexanes/
EtOAc (2:1) as eluant to give first E-isomer 7, followed by
Z-isomer 8. The combined yield was 0.316 g (70.5% from 21).
give the title compound 25 as a liquid (3.0 g, 82%): [R]D
)
25
+15.16° (c 0.89, CHCl3); IR (neat) 3499 cm-1; 1H NMR (CDCl3)
δ 1.05 (s, 9H, C(CH3)3), 1.30 and 1.40 (singlets, 6 H, CH3),
1.50-1.80 (m, 4 H, (CH2)2CH2OH), 2.5 (br s, 2 H, OH), 3.60
(m, 2 H, CH2OH),3.80-4.10 (m, 3 H, H-5a,b, H-4), 4.20 (d, J )
4.1 Hz, 1 H, H-2), 5.68 (d, J ) 4.1 Hz, 1 H, H-1), 7.30-7.40
(m, 6 Η, Ph), 7.60-7.80 (m, 4 Η, Ph); 13C NMR (CDCl3) δ 19.13,
26.56, 26.67, 26.81, 26.89, 35.01, 62.74, 63.62, 78.04, 84.17,
85.30, 104.45, 113.91, 127.70, 129.70, 133.04, 133.27, 135.62,
135.68. Anal. (C27H38O6Si) C, H.
(5S,6R,8S,9S)-6-[[(ter t-Bu tyld ip h en ylsilyl)oxy]m eth yl]-
8,9-O-isop r op ylid en e-2-k et o-1,7-d ioxa sp ir o[4.4]n on a n e
(26). A stirred suspension of powdered molecular sieves (4
Å, 7.5 g) and 25 (3.67g, 7.54 mmol) in anhydrous CH2Cl2 (100
mL) was treated with pyridinium chlorochromate (5.69 g, 26.4
mmol) at room temperature. After stirring for 1 h, the reaction
mixture was diluted with ether (200 mL) and filtered through
a short pad of silica gel. The silica gel pad was washed with
ether (ca. 100 mL), and the filtrate was concentrated and
purified by flash chromatography over silica gel using hexane/
EtOAc (7:3) to give the title compound 26 (3.12 g, 85.7%) as a
solid: mp 122-123 °C (EtOAc/hexane); [R]D25 ) -9.72° (c 1.07,
E-Isom er 7: white solid; mp 53-54 °C; [R]22 +20.32° (c
1
CHCl3); IR (KBr) 1793 cm-1; H NMR (CDCl3) δ 1.02 (s, 9 H,
D
0.62, CHCl3); IR (CHCl3) 3447 (OH), 1755 and 1722 (CdO),
C(CH3)3), 1.30 and 1.48 (singlets, 6 H, CH3), 2.10-2.22 (m, 1
H, H-4a), 2.30-2.70 (m, 3 H, H-4b, H-3a,b), 3.80-4.10 (m, 3 H,
(CH3)3SiOCH2, H-6), 4.48 (d, J ) 4.5 Hz, 1 H, H-9), 5.72 (d, J
) 4.5 Hz, 1 H, H-8), 7.30-7.50 (m, 6 H, Ph), 7.60-7.70 (m, 4
H, Ph); 13C NMR (CDCl3) δ 19.08, 26.46, 26.77, 27.50, 27.75,
30.93, 62.20, 83.82, 86.17, 86.51, 104.12, 115.81, 127.77,
127.81, 129.82, 132.89, 135.53, 175.18. Anal. (C27H34O6Si) C,
H.
1
and 1676 cm-1 (CdC); H NMR (CDCl3) δ 0.86 (distorted t, 3
H, CH3(CH2)7CHdCH(CH2)7CHdC<), 1.10-1.50 (m, 22 H,
CH3(CH2)6CH2CHdCHCH2(CH2)5CH2CHdC<), 2.00 (m, 4 H,
CH3(CH2)6CH2CHdCHCH2(CH2)5CH2CHdC<), 2.15 (m, 2 H,
CH3(CH2)6CH2CHdCHCH2(CH2)5CH2CHdC<), 2.68 (dm, J )
16.7 Hz, 1 H, H-4a), 3.05 (dm, J ) 16.7 Hz, 1 H, H-4b), 3.62
(AB d, J ) 12.0 Hz, 1 H, CHHOH), 3.74 (s, 3 H, CH3O), 3.78
(AB d, J ) 12.0 Hz, 1 H, CHHOH), 5.33 (m, 2 H, CH3(CH2)6-
CH2CHdCHCH2(CH2)5CH2CHdC<), 6.13 (d, J ) 15.7 Hz, 1
H, CH3OCOCHdCH), 6.76 (m, 1 H, CHdC<), 6.89 (d, J ) 15.7
Hz, CH3OCOCHdCH); 13C NMR (CDCl3) δ 14.07, 22.64, 27.12,
27.19, 28.00, 29.10, 29.25, 29.28, 29.48, 29.66, 29.72, 30.30,
(5R)-5-[(1R)-2-[(ter t-Bu tyldiph en ylsilyl)oxy]-1-h ydr oxy-
eth yl]-5-[2-(tetr a d eca n oxyca r bon yl)-(E)-eth en yl]tetr a h y-
d r o-2-fu r a n on e (28). A solution of 26 (3.07g, 6.36 mmol) in
THF (125 mL) was treated with a 1 N HCl solution (57 mL)
and stirred for 11 h at room temperature. The reaction

44 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1
Lee et al.
mixture was neutralized by the careful addition of solid
NaHCO3, after which it was concentrated and extracted with
EtOAc (3 × 40 mL). The combined organic extract was washed
with water (2 × 50 mL) and dried (Na2SO4). The residue
obtained after evaporation of the solvent was flash chromato-
graphed over silica gel using EtOAc/hexane (3:2), giving the
corresponding deprotected hemiacetal intermediate as a liquid
(1.92 g, 68%) mixture of anomers. Anal. (C24H30O6Si) C, H.
This hemiacetal intermediate (0.31 g, 0.7 mmol) was dissolved
in a mixture of MeOH (20 mL) and water (8 mL) and stirred
with sodium metaperiodate (0.599 g, 2.8 mmol) at room
temperature for 3.5 h. The reaction mixture was filtered, and
the filtrate was concentrated. The residue was dissolved in
EtOAc (30 mL), dried (Na2SO4), and concentrated to give the
intermediate aldehyde 27 which was used for the next reaction
without further purification. The above aldehyde was dis-
solved in benzene (15 mL) and stirred with tetradecanyl
(triphenylphosphoranylidene)acetate8 (0.726 g, 1.4 mmol) for
18 h at room temperature. The reaction mixture was filtered
through a short pad of silica gel and washed with ether (ca.
50 mL). The filtrate was concentrated, dissolved in MeOH (10
mL), and cooled to 0 °C. The solution was stirred and treated
with concentrated NH4OH (0.1 mL) for 12 h. The reaction
mixture was quenched with acetic acid (0.1 mL) and concen-
trated. The residue was purified by flash column chromatog-
raphy over silica gel using a gradient of 20-40% EtOAc in
hexane, to give the title compound 28 (0.359 g, 80.3%) as an
oil: [R]D25 ) +2.25° (c 1.51, CHCl3); IR (neat) 3473 (OH), 1786
and extracted with EtOAc (3 × 20 mL). The combined organic
extract was washed with water (2 × 10 mL) and dried (Na2-
SO4). The residue obtained after evaporation of the solvent
was purified by flash column chromatography over silica gel
using hexane/EtOAc (65:35) as eluant to give compound 6
(0.035 g, 34%), followed by 4 (0.024 g, 23%). Compound 6: mp
59-60 °C (EtOAc/hexane); [R]D ) +24.56° (c 0.81, CHCl3).
25
All the spectral properties were identical to the optical
antipode 5. Anal. (C22H38O5) C, H. Compound 4: mp 49-50
°C (EtOAc/hexane); [R]D ) -7.92° (c 1.06, CHCl3). All the
25
spectral properties were identical to the optical antipode 3 and
the racemate 2. Anal. (C22H40O5) C, H.
(S)-5-(Hyd r oxym eth yl)-5-[(3,4-d ih yd r o-2-oxo-5H-fu r a n -
3(E)-ylid en e)m et h yl]-3-{(E)-[(Z)-9-oct a d ecen ylid en e]}-
t et r a h yd r o-2-fu r a n on e (11), (S)-5-(H yd r oxym et h yl)-5-
[(3,4-dih ydr o-2-oxo-5H-fu r an -3(E)-yliden e)m eth yl]-3-{(Z)-
[(Z)-9-octadecen yliden e]}tetr ah ydr o-2-fu r an on e (12), (S)-
5-(Hyd r oxym eth yl)-5-[(3,4-d ih yd r o-2-oxo-5H-fu r a n -3(Z)-
ylid en e)m et h yl]-3-{(E)-[(Z)-9-oct a d ecen ylid en e]}t et r a -
h yd r o-2-fu r a n on e (13), a n d (S)-5-(H yd r oxym et h yl)-5-
[(3,4-dih yd r o-2-oxo-5H-fu r a n -3(Z)-yliden e)m eth yl]-3-{(Z)-
[(Z)-9-octa d ecen ylid en e]}tetr a h yd r o-2-fu r a n on e (14). Al-
dehyde 22 (1.0 mmol) was dissolved in a mixture of benzene
(20 mL) and dichloromethane (10 mL) and treated with
R-(triphenylphosphoranylidene)-γ-butyrolactone10 (0.070 g, 2.0
mmol). The reaction mixture was stirred for 14 h and
concentrated. The residue was passed through a short pad of
silica gel and eluted with ether. The filtrate was concentrated
under reduce pressure to give a mixture of formate esters and
free alcohols. The products were dissolved in methanol (5 mL),
and the solution was cooled to 0 °C and treated with am-
monium hydroxide solution (0.05 mL) with stirring for 10 min
at 0 °C. The reaction mixture was quenched with acetic acid
(0.05 mL) and concentrated to give four products of different
E/Z geometries (11-14) with very close Rf values on TLC. The
mixture of four products was dissolved in dichloromethane (40
mL), treated with tert-butyldimethylsilyl chloride (0.226 g, 1.5
mmol) and imidazole (0.272 g, 4.0 mmol), and stirred for 14
h. The reaction mixture was concentrated, and the residue
was purified by flash column chromatography on silica gel with
EtOAc/hexanes (5:1 to 3:1) as eluant. The order of elution of
products was as follows: 30 (protected 12), 29 (protected 11),
32 (protected 14), and 31 (protected 13). Each isomer was
dissolved in a mixture of acetic acid, water, and THF (3:1:1)
and heated to 60 °C for 48 h. The reaction mixture was
concentrated, and the residue was purified by flash column
chromatography with EtOAc/hexanes (1:1) to give the final
targets. The combined yield of products from 22 was 0.322
(70%).
and 1722 (CdO), and 1654 cm-1 1H NMR (CDCl3) δ 0.88
;
(distorted t, 3 H, CH3(CH2)13), 1.02 (s, 9 H, C(CH3)3), 1.20-
1.40 (m, 22 H, CH3(CH2)11CH2CH2OCO), 1.60 (m, 2 H,
CH3(CH2)11CH2CH2OCO), 2.10-2.22 (m, 1 H, H-4a), 2.40-2.60
(m, 3 H, H-4b, H-3a,b), 2.85 (br, 1H, OH), 3.62 (dd, J ) 11.7,
7.5 Hz, 1 H, SiOCHHOCH(OH)), 3.78 (m, 2 H, SiOCHHO-
CH(OH)), 4.08 (t, J ) 6.7 Hz, 2 H, CH3(CH2)11CH2CH2OCO),
6.02 (d, J ) 15.8 Hz, 1 H, CHdCHCO),7.00 (d, J ) 15.7 Hz, 1
H, CHdCHCO), 7.30-7.50 (m, 6 H, Ph), 7.60-7.70 (m, 4 H,
Ph); 13C NMR (CDCl3) δ 14.10, 19.14, 22.67, 25.88, 26.79, 27.51,
28.55, 29.25, 29.34, 29.40, 29.49, 29.58, 29.63, 29.67, 31.90,
63.44, 65.00, 75.08, 86.84, 121.43, 127.91, 130.03, 130.05,
132.38, 135.47, 144.16, 165.63, 175.71. Anal. (C39H58O6Si) C,
H.
(R)-5-(Hyd r oxym eth yl)-5-[2-(tetr a d eca n oxyca r bon yl)-
(E)-et h en yl]t et r a h yd r o-2-fu r a n on e (6) a n d (R)-5-(Hy-
d r oxym eth yl)-5-[2-(tetr a d eca n oxyca r bon yl)eth yl]tetr a -
h yd r o-2-fu r a n on e (4). A stirred solution of 28 (0.25 g, 0.39
mmol) in THF (7 mL) at 0 °C was treated with HF-pyridine
(1.5 mL). The reaction mixture was brought to room temper-
ature during the course of 1 h and stirred further for 15 h.
The reaction mixture was then diluted with a mixture of ice
and ether, and the layers were separated. The aqueous layer
was extracted with ether (3 × 15 mL), and the combine organic
extract was washed with water (2 × 10 mL) and dried (Na2-
SO4). The residue obtained after evaporation was purified by
flash column chromatography over silica gel using a gradient
from 60% to 100% EtOAc in hexane to give the intermediate,
(5R)-5-[(1R)-1,2-hydroxyethyl]-5-[2-(tetradecanoxycarbonyl)-
ethenyl]tetrahydro-2-furanone, as a white solid (0.115 g,
73%): mp 47-48 °C (EtOAc/hexane); [R]D25 ) +21.78° (c 1.01,
CHCl3); IR (CH2Cl2) 3453 (OH), 1779 and 1707 (CO), and 1654
Compound 12: oil; [R]22 -5.71° (c 0.14, CHCl3); IR (neat)
D
3421 (OH), 1756 (CdO), 1670 cm-1 (CdC); 1H NMR (CDCl3) δ
0.86 (distorted t, 3 H, CH3(CH2)7CHdCH(CH2)7CHdC<),
1.10-1.50 (m, 22 H, CH3(CH2)6CH2CHdCHCH2(CH2)5CH2-
CHdC<), 1.85 (br s, 1 H, OH), 2.00 (m, 4 H, CH3(CH2)6CH2-
CHdCHCH2(CH2)5CH2CHdC<), 2.60-2.85 (m, 3 H, CH3(CH2)6-
CH2CHdCHCH2(CH2)5CH2CHdC<, H-4a), 3.00-3.30 (m, 3 H,
H-4b, H-4′a,b), 3.70 (AB q, J ) 12.5 Hz, 2 H, CH2OH), 4.36 (t,
J ) 7.3 Hz, 2 H, H-5′a,b), 5.32 (m, 2 H, CH3(CH2)6CH2CHdCH-
(CH2)7CHdC<), 6.26 (m, 1 H, C17H33CHdC<), 6.61 (t, J ) 2.9
Hz, 1 H, >CdCHC); 13C NMR (CDCl3) δ 14.09, 22.66, 25.86,
27.16, 27.19, 27.86, 28.95, 29.18, 29.22, 29.29, 29.40, 29.50,
29.70, 29.73, 31.88, 36.23, 65.95, 66.38, 83.95, 123.02, 128.11,
129.72, 129.98, 136.35, 146.44, 168.47, 171.2; FAB MS m/ z
(relative intensity) 461 (MH+, 10). Anal. (C28H44O5) C, H.
1
cm-1; H NMR (CDCl3) δ 0.88 (distorted t, 3 H, CH3(CH2)13 ),
1.10-1.40 (m, 22 H, CH3(CH2)11CH2CH2OCO), 1.65 (m, 2 H,
CH3(CH2)11CH2CH2OCO), 2.00-2.22 (m, 1 H, H-4a), 2.40-2.70
(m, 3 H, H-4b, H-3a,b), 3.60 (dd, J ) 11.2, 7.0 Hz, 1 H,
CHHOCH(OH)), 3.72-3.90 (m, 2 H, CHHOCH(OH)), 4.10 (t,
J ) 6.7 Hz, 2 H, CH3(CH2)11CH2CH2OCO), 6.10 (d, J ) 15.7
Hz, 1 H, CHdCHCO), 6.95 (d, J ) 15.7 Hz, 1 H, CHdCHCO);
13C NMR (CDCl3) δ 14.10, 22.67, 25.88, 27.70, 28.49, 28.53,
29.24, 29.33, 29.49, 29.57, 29.63, 31.90, 62.27, 65.23, 75.21,
87.26, 122.04, 144.12, 165.75, 175.88. Anal. (C23H40O6) C, H.
A stirred solution of this intermediate (0.112 g, 0.28 mmol)
and NaHCO3 (0.047 g, 0.56 mmol) in ethanol (5 mL) and water
(2 mL) was treated with sodium metaperiodate (0.180 g, 0.84
mmol) at room temperature for 3.5 h. The reaction mixture
was cooled to 0 °C and treated with NaBH4 (0.0 21 g, 0.56
mmol) for 1.5 h. The reaction mixture was then concentrated
Compound 11: oil; [R]22 +26.85° (c 0.54, CHCl3); IR (neat)
D
3422 (OH), 1756 (CdO), 1680 cm-1 (CdC); H NMR (CDCl3)
1
1H NMR (CDCl3) δ 0.86 (distorted t, 3 H, CH3(CH2)7CHdCH-
(CH2)7CHdC<), 1.10-1.50 (m, 22 H, CH3(CH2)6CH2CHdCH-
CH2(CH2)5CH2CHdC<), 2.00 (m, 4 H, CH3(CH2)6CH2CHdCH-
CH2(CH2)5CH2CHdC<), 2.18 (m,
2 H, CH3(CH2)6CH2-
CHdCHCH2(CH2)5CH2CHdC<), 2.72 (dm, J ) 16.9 Hz, 1 H,
H-4a), 2.95-3.30 (m, 3 H, H-4b, H-4′a,b), 3.73 (AB q, J ) 12.1
Hz, 2 H, CH2OH), 4.35 (t, J ) 7.3 Hz, 2 H, H-5′a,b), 5.32 (m, 2
H, CH3(CH2)7CHdCH(CH2)7CHdC<), 6.64 (t, J ) 2.9 Hz, 1
H, >CdCHC), 6.77 (m, 1 H, C17H33CHdC<); 13C NMR (CDCl3)
δ 14.03, 22.58, 25.79, 27.05, 27.12, 27.95, 29.05, 29.22, 29.25,

Conformationally Constrained Analogues of Diacylglycerol. 12
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 1 45
Groups of the 3-O-Acyl Function in 2-Deoxy-L-Ribonolactones.
BioMed. Chem. Lett. 1994, 4, 1369-1374.
(4) Morimoto, Y.; Mikami, A.; Shirahama, H. Unprecented Stability
of δ-Lactones with Axial Substituents rather than Equatorial
Ones; Comparison with Prelog-Djerassi Lactone Derivative. J .
Chem. Soc., Chem. Commun. 1991, 1376-1378.
(5) Wang, S.; Milne, G. W. A.; Nicklaus, M. C.; Marquez, V. E.; Lee,
J .; Blumberg, P. M. Protein Kinase C. Modeling of the Binding
Site and Prediction of Binding Constants. J . Med. Chem. 1994,
4, 1326-1338.
(6) Marquez, V. E.; Lee, J .; Sharma, R.; Teng, K.; Wang, S.; Lewin,
N. E.; Bahador, A.; Kazanietz, M. G.; Blumberg, P. M. Confor-
mationally Constrained Analogues of Diacylglycerol. 6. Changes
in PK-C Binding Affinity for 3-O-Acyl-2-Deoxy-L-Ribonolactones
Bearing Different Acyl Chains. BioMed. Chem. Lett. 1994, 4,
355-360.
(7) The relative stereochemistry at C-5 is the same in both tem-
plates; however, the presence of the double bond in 5 (as well
as in 7 and 8) causes the chiral center to change from “R” to
“S”.
(8) Floch, Y. L.; Yuergnaux, F.; Toupet, L.; Gree, R. Syntheses et
Reactivite d’ω-Aldehydophosphoranes; Application a la Prepara-
tion de Lactones R,â-Insaturees. (Synthesis and Reactivity of
ω-Aldehydophosphoranes. Application for the Preparation of R,â-
Unsaturated Lactones.) Bull. Soc. Chim. Fr. 1991, 128, 742-
758.
(9) Dahlman, O.; Garegg, P. J .; Mayer, H.; Schramek, S. Synthesis
of Three 3-C-Hydroxymethylpentoses with the D-Ribo-, D-Xylo-
and L-Lyxo-Configurations. Identification of the Latter with a
Monosaccharide Isolated from Phase I Coxiella Burnetii Li-
popolysaccharide. Acta Chem. Scand. B 1986, 40, 15-20.
(10) Howie, G. A.; Manni, P. E.; Cassady, J . M. Synthesis of Alkyl-
Substituted R,â-Unsaturated γ-Lactones as Potential Antitumor
Agents. J . Med. Chem. 1974, 17, 840-843.
(11) Szallasi, Z.; Denning, M. F.; Smith, C. B.; Dlugosz, A. A.; Yuspa,
S. H.; Pettit, G. R.; Blumberg, P. M. Mol. Pharmacol. 1994, 46,
840.
(12) Blumberg, P. M.; Acs, G.; Areces, L. B.; Kazanietz, M. G.; Lewin,
N. E.; Szallasi, A. In Receptor-Mediated Biological Processes:
Implications for Evaluating Carcinogenesis; Spirer, H. L., Slaga,
T. J ., Eds.; Wiley-Liss: New York, 1994; pp 3-19.
(13) Szallasi, Z.; Kosa, K.; Smith, C. B.; Dlugosz, A. A.; Williams, E.
K.; Yuspa, S. H.; Blumberg, P. M. Differential Regulation by
Anti-Tumor-Promoting 12-Deoxyphorbol-13-phenylacetate Re-
veals Distinct Roles of the Classical and Novel Protein Kinase
C Isozymes in Biological Response of Primary Responses of
Primary Mouse Keratinocytes. Mol. Pharmacol. 1995, 47, 258-
265.
(14) Teng, K.; Marquez, V. E.; Milne, G. W. A.; Barchi, J r. J . J .;
Kazanietz, M. G.; Lewin, N. E.; Blumberg, P. M.; Abushanab,
E. Conformationally Constrained Analogues of Diacylglycerol.
Interaction of γ-Lactones with the Phorbol Ester Receptor of
Protein Kinase C. J . Am. Chem. Soc. 1992, 114, 1059-1070.
(15) Kazanietz, M. G.; Areres, L. B.; Bahador, A.; Mischak, H.;
Goodnight, J .; Mushinski, J . F.; Blumberg, P. M. Characteriza-
tion of Ligand and Substrate Specificity for the Calcium-
Dependent and Calcium-Independent Protein Kinase C Isozymes.
Mol. Pharmacol. 1993, 44, 298-307.
29.42, 29.60, 29.66, 30.33, 31.80, 32.87, 66.00, 66.42, 84.62,
125.09, 127.86, 129.57, 129.95, 136.42, 142.74, 169.91, 171.33;
FAB MS m/ z (relative intensity) 461 (MH+, 19). Anal.
(C28H44O5) C, H.
Compound 14: semisolid gum; [R]22D -30.00° (c 0.20, CHCl3);
IR (CHCl3) 3423 (OH), 1753 and 1720 (CdO), 1675 cm-1 (CdC);
1H NMR (CDCl3) δ 0.86 (distorted t, 3 H, CH3(CH2)7CHdCH-
(CH2)7CHdC<), 1.10-1.50 (m, 22 H, CH3(CH2)6CH2CHdCH-
CH2(CH2)5CH2CHdC<), 1.85 (br s, 1 H, OH), 2.00 (m, 4 H,
CH3(CH2)6CH2CHdCHCH2(CH2)5CH2CHdC<), 2.65 (m, 2 H,
CH3(CH2)6CH2CHdCHCH2(CH2)5CH2CHdC<), 3.00 (m, 3 H,
H-4a, H-4′a,b), 3.42 (dm, J ) 16.9 Hz, 1 H, H-4b), 3.90 (AB q, J
) 11.9 Hz, 2 H, CH2OH), 4.38 (m, 2 H-5′a,b), 5.32 (m, 2 H,
CH3(CH2)6CH2CHdCH(CH2)7CHdC<), 6.18 (m, 1 H, C17H33
-
CHdC<), 6.47 (t, J ) 2.4 Hz, 1 H, >CdCH-C); 13C NMR
(CDCl3) δ 14.10, 22.67, 27.20, 27.69, 29.03, 29.20, 29.23, 29.30,
29.50, 29.56, 29.69, 29.71, 31.89, 37.68, 65.71, 66.71, 85.63,
124.47, 126.49, 129.79, 129.95, 143.20, 145.05, 168.31, 168.67;
FAB MS m/ z (relative intensity) 461 (MH+, 22). Anal.
(C28H44O5) C, H.
Compound 13: semisolid gum; [R]22D -16.19° (c 0.42, CHCl3);
IR (neat) 3407 (OH), 1746 and 1722 (CdO), 1678 cm-1 (CdC);
1H NMR (CDCl3) δ 0.86 (distorted t, 3 H, CH3(CH2)7CHdCH-
(CH2)7CHdC<), 1.10-1.50 (m, 22 H, CH3(CH2)6CH2CHdCH-
CH2(CH2)5CH2CHdC<), 2.00 (m, 4 H, CH3(CH2)6CH2CHdCH-
CH2(CH2)5CH2CHdC<), 2.18 (m,
2 H, CH3(CH2)6CH2-
CHdCHCH2(CH2)5CH2CHdC<), 2.88 (dm, J ) 17.9 Hz, 1 H,
H-4a), 3.00 (m, 2 H, H-4′a,b), 3.40 (dm, J ) 17.9 Hz, 1 H, H-4b),
3.92 (AB q, J ) 12.0 Hz, 2 H, CH2OH), 4.38 (m, 2 H, H-5′a,b),
5.32 (m, 2 H, CH3(CH2)6CH2CHdCH(CH2)7CHdC<), 6.49 (t,
J ) 2.4 Hz, 1 H, >CdCH-C), 6.69 (m, 1 H, C17H33CHdC<);
13C NMR (CDCl3) δ 14.10, 22.67, 27.16, 27.21, 28.09, 29.16,
29.30, 29.51, 29.69, 29.74, 30.21, 31.89, 32.59, 34.64, 65.74,
66.75, 86.27, 126.45, 126.54, 129.71, 130.01, 141.66, 143.14,
168.33, 169.90; FAB MS m/ z (relative intensity) 461 (MH+,
26). Anal. (C28H44O5) C, H.
Su p p or tin g In for m a tion Ava ila ble: Tables of atomic
coordinates, bond distances and angles, and anisotropic ther-
mal parameters for compound (()-33 (Tables 4-8) (5 pages).
Ordering information is given on any current masthead page.
Refer en ces
(1) Lee, J .; Wang, S.; Lewin, N. E.; Blumberg, P. M.; Marquez, V.
E. J . Med. Chem. 1996, 39, 29-35.
(2) Sharma, R.; Lee, J .; Lewin, N. E.; Blumberg, P. M.; Marquez,
V. E. J . Med. Chem. 1996, 39, 19-28.
(3) Lee, J .; Sharma, R.; Teng, K.; Lewin, N. E.; Blumberg, P. M.;
Marquez, V. E. Conformationally Constrained Analogues of
DAG. 8. Changes in PK-C Binding Affinity Produced by Isosteric
J M950278F