Conformationally constrained analogues of diacylglycerol (DAG). 16. How much structural complexity is necessary for recognition and high binding affinity to protein kinase C?

Article abstract of DOI:10.1021/jm9904607

The design of potent protein kinase C (PK-C) ligands with low nanomolar binding affinities was accomplished by the combined use of pharmacophore- and receptor-guided approaches based on the structure of the physiological enzyme activator, diacylglycerol (

Full text of DOI:10.1021/jm9904607

J . Med. Chem. 2000, 43, 921-944
921
Con for m a tion a lly Con str a in ed An a logu es of Dia cylglycer ol (DAG). 16.1 How
Mu ch Str u ctu r a l Com p lexity Is Necessa r y for Recogn ition a n d High Bin d in g
Affin ity to P r otein Kin a se C?
Kassoum Nacro,† Bruno Bienfait,† J eewoo Lee,‡ Kee-Chung Han,‡ J i-Hye Kang,‡ Samira Benzaria,†
Nancy E. Lewin,§ Dipak K. Bhattacharyya,§ Peter M. Blumberg,§ and Victor E. Marquez*,†
Laboratories of Medicinal Chemistry and of Cellular Carcinogenesis and Tumor Promotion, Division of Basic Sciences,
National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892, and Laboratory of Medicinal Chemistry,
College of Pharmacy, Seoul National University, Shinlim-Dong, Kwanak-Ku, Seoul 151-742, Korea
Received September 9, 1999
The design of potent protein kinase C (PK-C) ligands with low nanomolar binding affinities
was accomplished by the combined use of pharmacophore- and receptor-guided approaches
based on the structure of the physiological enzyme activator, diacylglycerol (DAG). Earlier
use of the former approach, which was based on the structural equivalence of DAG and phorbol
ester pharmacophores, identified a fixed template for the construction of a semirigid “recognition
domain” that contained the three principal pharmacophores of DAG constrained into a lactone
ring (DAG-lactones). In the present work, the pharmacophore-guided approach was refined to
a higher level based on the X-ray structure of the C1b domain of PK-Cδ complexed with phorbol-
13-O-acetate. A systematic search that involved modifying the DAG-lactone template with a
combination of linear or branched acyl and R-alkylidene chains, which functioned as variable
hydrophobic “affinity domains”, helped identify compounds that optimized hydrophobic contacts
with a group of conserved hydrophobic amino acids located on the top half of the C1 domain
where the phorbol binds. The hydrophilic/hydrophobic balance of the molecules was estimated
by the octanol/water partition coefficients (log P) calculated according to a fragment-based
approach. The presence of branched R-alkylidene or acyl chains was of critical importance to
reach low nanomolar binding affinities for PK-C. These branched chains appear to facilitate
important van der Waals contacts with hydrophobic segments of the protein and help promote
the activation of PK-C through critical membrane interactions. Molecular modeling of these
DAG-lactones into an empty C1b domain using the program AutoDock 2.4 suggests the existence
of competing binding modes (sn-1 and sn-2) depending on which carbonyl is directly involved
in binding to the protein. Inhibition of epidermal growth factor (EGF) binding, an indirect
PK-C mediated response, was realized with some DAG-lactones at a dose 10-fold higher than
with the standard phorbol-12,13-dibutyrate (PDBU). Through the National Cancer Institute
(NCI) 60-cell line in vitro screen, DAG-lactone 31 was identified as a very selective and potent
antitumor agent. The NCI’s computerized, pattern-recognition program COMPARE, which
analyzes the degree of similarity of mean-graph profiles produced by the screen, corroborated
our principles of drug design by matching the profile of compound 31 with that of the non-
tumor-promoting antitumor phorbol ester, prostratin. The structural simplicity and the degree
of potency achieved with some of the DAG-lactones described here should dispel the myth that
chemical complexity and pharmacological activity go hand in hand. Even as a racemate, DAG-
lactone 31 showed low namomolar binding affinity for PK-C and displayed selective antitumor
activity at equivalent nanomolar levels. Our present approach should facilitate the generation
of multiple libraries of structurally similar DAG-lactones to help exploit molecular diversity
for PK-C and other high-affinity receptors for DAG and the phorbol esters. The success of this
work suggests that substantially simpler, high-affinity structures could be identified to function
as surrogates of other complex natural products.
In tr od u ction
diacylglycerol (DAG).2,3 Typically, the isozymes are
cytosolic in the inactive state and, as part of the
activation process, translocate to the inner leaflet of the
cellular membrane.4 Individual cell types contain dif-
ferent combinations of PK-C isozymes, suggesting dis-
tinct roles for these isozymes in signaling pathways that
regulate cell cycle progression, differentiation, and
apoptosis.2 The binding of PK-C to the plasma mem-
brane is transient and regulated by the association with
diacylglycerol (DAG) in the membrane.5 DAG (Figure
Protein kinase C (PK-C) is a family of at least 11
serine/threonine-specific isozymes that, depending on
the specific isozyme, are regulated by phosphorylation,
by calcium, and by association with phospholipids and
* To whom correspondence should be addressed. Tel: 301-496-3597.
Fax: 301-402-2275. E-mail: marquezv@dc37a.nci.nih.gov.
† Laboratory of Medicinal Chemistry, National Cancer Institute.
‡ Seoul National University.
§ Laboratory of Cellular Carcinogenesis and Tumor Promotion,
National Cancer Institute.
10.1021/jm9904607 CCC: $19.00 © 2000 American Chemical Society
Published on Web 02/23/2000

922 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5
Nacro et al.
1) poses the intriguing question of whether the intricate
structure of these natural products and their biological
potency go hand-in-hand. To address this issue, we have
sought to design synthetically accessible ligands for the
C1 domain of PK-C based on the simpler DAG structure.
Initially, using a pharmacophore-guided approach that
was based on the geometry of bioequivalent pharma-
cophores present in DAG and in the phorbol esters,11,12,23
we synthesized a number of potent DAG-lactones in
which the glycerol backbone was constrained to a
lactone ring to reduce the entropic penalty associated
with DAG binding (Figure 1).24-26 In the present
investigation, the information provided by the X-ray27
and NMR28 structures of the cysteine-rich, C1 phorbol
ester binding domain has been used to further modify
these DAG-lactones to optimize their interaction with
a group of highly conserved hydrophobic amino acids
along the rim of the C1 domain.27,28 With this approach,
we sought to transform these structurally simple DAG-
lactones into ligands with affinities approaching those
of the phorbol esters. These compounds were intended
to serve both as probes to test our understanding of the
nature of ligand-C1 domain interactions and to explore
the basis for biological selectivity of PK-C-targeted
ligands with a library of easily obtainable DAG mimet-
ics.
The use of the receptor-guided approach described
here culminated with the synthesis of a series of
molecules that bind to PK-C with affinities in the low
nanomolar range. These compounds provide insight into
the basis for PK-C ligand specificity, and some of them
show potential utility as antitumor agents. Here, we
present a discussion based on this effective drug-design
strategy, report on the chemical syntheses of the new
targets, and analyze the biological profile of the new
potent ligands in terms of several important param-
eters, including binding affinity (Ki) to PK-CR, activation
of PK-CR, inhibition of epidermal growth factor (EGF)
binding as an indirect PK-C mediated response,29 and
antitumor activity in the National Cancer Institute 60-
cell line in vitro screen.30
F igu r e 1. Chemical structures of phorbol esters, bryostatin
1, DAG, and DAG-lactones showing the critical pharmaco-
phores suggested by computational and structure-activity
studies.11
1) functions as a central lipophilic second messenger
that is generated in response to activation of numerous
receptors which are coupled either through G-protein
or tyrosine kinase mechanisms to activation of phos-
pholipase C hydrolysis of phosphatidylinositol 4,5-
biphosphate.6 DAG can also arise indirectly from phos-
phatidic acid produced by phospholipase D.7 Physiologi-
cally, DAG binds to the C1 domain in members of the
classical (R, â, and γ), as well as the novel (δ, ꢀ, η, and
θ), PK-C isozymes to activate their downstream path-
ways.8 Among pharmacological agents, the phorbol
esters (Figure 1) bind directly to the same C1 domain
and function as potent and metabolically stable DAG
surrogates.9 Indeed, the phorbol esters bind to PK-C
with affinities that are 3 to 4 orders of magnitude
greater than that of DAG.10
Highlighting the important biological role of PK-C,
several other structural classes of natural products,
among them bryostatin 1 (Figure 1), have also been
characterized as potent DAG analogues.12 From studies
with these natural products, it is clear that different
specific ligands with similar pharmacophores that rec-
ognize PK-C (Figure 1) can differentially activate PK-C
subpathways and thereby induce different patterns of
biological responses. Thus, whereas the archetypical
phorbol-12-myristate-13-acetate (PMA) acts as a potent
tumor promoter,13 the macrocyclic lactone bryostatin 1,
currently in clinical trials as a cancer chemotherapeutic
agent, is a partial functional antagonist of PMA.14 The
structural complexity of these and other natural prod-
ucts has prompted several groups to synthesize simpler
synthetic analogues, some of which displayed significant
binding affinities toward PK-C.15-22 However, a marked
degree of structural complexity still remains in these
molecules, making it difficult to exploit chemical diver-
sity to the fullest extent. Thus, the availability of only
a limited number of distinct PK-C ligands represents
an obstacle to understand the structural basis for their
interactions with PK-C, as well as the origins of their
distinct biological effects.
Th e Recep tor -Gu id ed Ap p r oa ch
A. Hyd r ogen Bon d in g. The recent X-ray structure
of phorbol-13-O-acetate complexed to the C1b domain
of PK-Cδ27 confirmed the importance of hydrogen bonds
involving the key pharmacophores C-20 (OH) and C-3
(CdO), the latter in combination with C-4 (OH), as
essential elements for the recognition of PK-Cδ and
other PK-C isozymes by the phorbol esters (see Figure
1 for phorbol’s numbering). However, the presumed
critical C-9 (OH) pharmacophore did not participate in
any hydrogen bond to the protein and appeared instead
intramolecularly hydrogen-bonded to the C-13 (CdO)
ester. Since the role of the C-13 (CdO) has been shown
to be critical in recent structure-activity studies,31,32
it is possible that the intramolecular hydrogen-bonding
motif involving C-9 (OH) and C-13 (CdO) binds outside
the C1b domain, perhaps at the membrane-protein
interface, which would explain why it appears missing
in the phorbol-13-O-acetate/C1b complex.27 In a recent
study, we reported on the use of the program AutoDock
to anchor a DAG-lactone into an empty C1b domain
constructed with the same crystal coordinates as those
The stark contrast between the structurally complex
natural products and the simple DAG structure (Figure

Conformationally Constrained Analogues of DAG
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5 923
F igu r e 2. Docking results showing hydrogen-bonding interactions of 31 with the C1b domain of PK-Cδ in binding modes sn-1
(left) and sn-2 (right).33
Ta ble 1. PK-CR Binding Affinity (Ki) and Calculated log P for
R-Alkylidene and Acyl Branched Lactones (Z- and E-Isomers)
functions appears directly involved in binding to the
protein. In these two binding alternatives, it is possible
that the uninvolved CdO group would bind at the
membrane-protein interface and be functionally equiva-
lent to the intramolecularly hydrogen-bonded C-9 (OH)/
C-13 (CdO) array of the phorbol esters. The absolute
necessity for both sn-1 and sn-2 carbonyls to be simul-
taneously present in these DAG-lactones was demon-
strated earlier by the poor binding affinity shown by
compounds lacking either one of these functionalities.1
B. Hyd r op h obic In ter a ction s. Even though the role
of the mostly linear aliphatic ester groups in DAG has
been principally correlated with lipophilicity to facilitate
partitioning or transport between biological phases,34,35
the docking of the DAG-lactones to the C1b domain of
PK-Cδ revealed potentially important hydrophobic con-
tacts with the protein which could probably extend into
the protein-membrane interface. Analysis of the sn-1
and sn-2 binding modes in relation to the location of a
group of highly conserved hydrophobic amino acids on
the rim of the two loops of the C1b domain in PKCδ
(Met 239, Pro 241, Phe-243, Leu-250, Trp-252, and Leu
254)27 provided the basis to modify the DAG-lactones
in a fashion to optimize hydrophobic binding to these
amino acids’ side chains (vide infra). In the crystal
structure of the phorbol-13-O-acetate/C1b complex of
PK-Cδ, all these hydrophobic amino acids appear ex-
posed and positioned to point into the membrane,27 and
in the case of the equivalent C1b domain of PK-Cγ, an
NMR study of the amide proton signals perturbed by
lipid interactions in the presence and absence phorbol-
12,13-dibutyrate (PDBU) identified an equivalent set of
hydrophobic amino acids on the top half of the C1b
domain that is probably responsible for the insertion of
the enzyme into the lipid membrane.36
of the C1b domain complexed with phorbol-13-O-
acetate.1 The program consistently identified two en-
ergetically similar binding modes with an identical
network of hydrogen bonds to the same amino acids as
in the phorbol-13-O-acetate/C1b complex. As shown in
Figure 2 for DAG-lactone 31 (Table 1), these two binding
modes are designated as sn-1 and sn-2 (Figure 1)
depending on which of the nonequivalent carbonyl
Design a n d Syn th esis
A. Design a n d Con str u ction of th e Br a n ch ed
Alk yl Ch a in . An inspection of the environment of the
DAG-lactones docked in the C1b domain of PK-Cδ
suggested that hydrophobic contacts with the protein
might improve with the use of branched hydrocarbon

924 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5
Nacro et al.
defined by Leu 250 and Trp 252. In the alternative sn-2
binding mode, these interactions are reversed (Figure
4, right). The optimization of hydrophobic binding to the
loops of the C1b domain derived from the branched
chains of the DAG-lactones was expected to increase the
binding affinity of the ligands in either alternative
binding mode. However, since linear alkyl chains might
interact better with the membrane, a structure-activity
study with combinations of branched and linear chains
was initiated to select the best blend of R1 and R2
substituents.
B. Design a n d Con str u ction of Br a n ch ed DAG-
La cton es. 1. Ra cem ic r-Alk ylid en e E- a n d Z-La c-
ton es. The initial set of target compounds (7-32) is
listed in Table 1. The first subgroup (compounds 7-12)
included compounds where aldehyde 3 was used for the
construction of the R-alkylidene branch (R2). To reach
optimal lipophilicity and to ensure adequate partition-
ing into the membrane, a limited range (n ) 1, 3, and
5) of linear aliphatic acyl groups (R1 ) CnH2n+1) was
investigated. The second subgroup of compounds (13-
22) included targets in which aldehyde 4 served as the
precursor for the R-alkylidene branch. Since the Z-
isomers 13, 15, and 17 showed increased potency vis-
a`-vis the Z-isomers of the first subgroup (vide infra),
longer aliphatic acyl chains (R1 ) CnH2n+1, n ) 7 and
9) were investigated in this series to optimize lipophi-
licity (compounds 19-22). The third subgroup was
designed to explore the effects of switching the branch-
ing from the R-alkylidene side to the acyl side. There-
fore, each Z/E-pair in the third subgroup is isomeric and
of equal lipophilicity to the corresponding Z/E-pair of
the second subgroup. Finally, compounds 31 and 32
constitute the subgroup corresponding to doubly branched
acyl and R-alkylidene lactones.
F igu r e 3. Design of the 2,3,4-trimethylpentane chain for acyl
(R1) and R-alkylidene (R2) branching from the amino acid
leucine. The operation involves the removal of the NH2 group
and the addition of an extra i-Pr group.
Sch em e 1
chains instead of linear chains. For this purpose, a
branched chain containing isopropyl groups was chosen
to mimic the side chains of amino acids Leu 250, Leu
254, and Val 255 that are an integral part of the
hydrophobic rim of the C1b domain.27 The newly
designed branched motif was conceived to be attached
to the carbonyl group for acyl branching (R1) or con-
nected to the lactone via a methylene group for R-alkyli-
dene branching (R2) (Table 1). In designing an appropri-
ate pattern, consideration was given to maintaining a
symmetric branch capable of reaching hydrophobic
binding sites in every orientation. For that reason, and
also to increase hydrophobicity, the 2,3,4-trimethylpen-
tane chain was selected. The process of selection and
construction of this chain could be envisaged as derived
from the amino acid leucine following the addition of a
symmetrical isopropyl branch and the removal of the
amino group as depicted in Figure 3.
The branched 2,3,4-trimethylpentane chain was con-
structed from diisopropyl ketone as shown in Scheme
1. Aldehyde 3 was prepared as reported by Denmark et
al.,37 and catalytic hydrogenation of 3 afforded the target
aldehyde 4. Because of their close structural similarity,
both 3 and 4 were used as branched chains (R2) for the
R-alkylidene lactones (Table 1). Oxidation of the alde-
hyde provided the branched acid 5 which served as
precursor of 6 for the acyl branched (R1) lactones (Table
1).
The compounds were synthesized as racemates using
general alkylation and acylation procedures illustrated
in Scheme 2. The corresponding geometrical Z- and
E-isomers were separated, and the geometry of the
1
exocyclic double bond was assigned by H NMR spec-
troscopy. For the E-isomers 7, 9, and 11 (Table 1), the
doublet corresponding to the γ-vinyl proton resonated
at ca. δ 5.86, whereas the same signal for the Z-isomers
(8, 10, and 12, Table 1) appeared 1.4 ppm lower at ca.
δ 7.27. The â-cis vinyl proton of the E-isomers in the
doubly unsaturated compounds (7-12), as well as in all
the monounsaturated compounds (13-32, Table 1), ap-
peared consistently ca. 0.5 ppm downfield from the
corresponding â-trans proton of the Z-isomers.
2. Ra cem ic r-Alk yl La cton es. Two additional com-
pounds (65 and 66) corresponding to the cis and trans
isomers that would have resulted from the catalytic
hydrogenation of compounds 17/18 were synthesized
(Scheme 3). These compounds were investigated to
compare the effects of a branched R-alkyl moiety versus
an R-alkylidene chain. Since these compounds represent
a separate and limited structure-activity analysis, they
are not included in Table 1. The precursor compound
(64) could have been obtained by the catalytic hydro-
genation of several intermediates (i.e., 52-55); however,
53 turned out to be the most convenient starting
material. Monoacylation of 64 gave a pair of diastereo-
isomers (65 and 66) in which the relative orientation of
the acyl and R-alkyl branches could be either cis or
A nice complementarity between hydrophobic groups
in the branched DAG-lactone ligand and the C1b
domain of PK-Cδ can be appreciated in Figure 4 for the
doubly branched DAG-lactone 31 (Table 1). In the sn-1
binding mode (Figure 4, left), the R-alkylidene branched
(R2) motif appears to optimize hydrophobic contacts with
the region between Met 239 and Leu 254, whereas the
acyl branch interacts with the hydrophobic region

Conformationally Constrained Analogues of DAG
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5 925
F igu r e 4. sn-1 (left) and sn-2 (right) binding modes for DAG-lactone 31 (Table 1) (R1 ) R2 ) CH2CH(i-Pr)2). The 3D surface
representation was calculated with the program Insight 97, Molecular Simulation Inc. (http://www.msi.com/). The amino acid
residues are colored according to the Insight 97 hydrophobicity scale (red, lipophile; blue, hydrophile).
Sch em e 2
Sch em e 3
nated consistently ca. 1 ppm lower than the free CH2-
OH group. Therefore, when the branched R-alkyl chain
was located on the same side of either AB system, the
spread of the AB quartet increased due to the restricted
freedom of rotation. This simple observation was very
diagnostic and provided an easy way to differentiate
diastereoisomers 65 and 66.
3. En a n tiom er ic r-Alk ylid en e E- a n d Z-Rever se
Ester La cton es. We have previously shown that PK-C
only recognizes DAG-lactones having the R-configura-
tion.25 However, since these compounds are susceptible
to racemization via acyl migration, a series of “reversed
ester” analogues were developed to overcome this prob-
lem.26 This operation facilitated the synthesis of single
enantiomeric targets which showed equivalent biological
potencies to the corresponding “normal ester” enanti-
omers.25,26 The most potent ligands developed earlier
in this category were compounds 67 and 68, which have
an 18-carbon R-octadec-9-enylidene side chain selected
to mimic oleic acid.26 Compounds 67 and 68 (Table 2)
are examples of this class of “reversed ester” analogues
in which the relative position of the carbonyl ester
remains unchanged. For these compounds, however,
even though the relative stereochemistry between the
“normal ester” and the “reversed ester” analogues is the
same, the presence of the double bond in the latter group
causes the chiral center to change from R to S.26 As with
lactones 13-22 (Table 1), the R-alkylidene branched
2,3,4-trimethylpentane (R2) motif required lengthening
the chain of the aliphatic ester (R1) to improve overall
hydrophobicity. Therefore, targets 69 and 70 (log P )
5.41, Table 2) were constructed with a six-carbon alkyl
trans. The orientation of the R-alkyl branch was ascer-
1
tained by H NMR spectroscopy which was sensitive to
the relative anisotropy of the environment of the AB
quartets corresponding to the acylated hydroxymethyl
group [CH3(CH2)COCH2] and the free CH2OH group.
The corresponding acylated hydroxymethyl group reso-

926 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5
Nacro et al.
Sch em e 5a
Ta ble 2. PK-CR Binding Affinities (Ki) and log P for
R-Alkylidene “Reversed Ester” Lactones (Z- and E-Isomers)
Sch em e 4
a
Reagents and conditions: (a) (Bu3Sn)2O, BnBr, Bu4NBr,
toluene (99%); (b) PDC, AcOH, 4 Å mol. sieve, CH2Cl2 (98%); (c)
Cl(CH2)3OH, CH3MgCl, Mg, Br(CH2)2Br (85%); (d) PCC, 4 Å mol.
sieve (98%); (e) Dowex H+ resin, THF-H2O; (f) NaIO4, MeOH-
H2O; (g) NaBH4, MeOH (75% in 3 steps); (h) H2, Pd/C, MeOH
(99%); (i) p-TsOH, acetone (97%); (j) BnBr, AgO, DMF (84%); (k)
LiHMDS, THF, -78 °C; (i-Pr)2CHCH2CHO, HMPA; (l) MsCl, NEt3,
CH2Cl2, DBU (Z: 35%, E: 24%); (m) H5IO6, ether (96%); (n)
PPh3dCHCO2C6H13, CH2Cl2 (93%); (o) BCl3, CH2Cl2, -78 °C
(50%).
chain designed to approximate the hydrophilic/lipophilic
balance of compounds 19 and 20 (log P ) 5.62, Table
1).
gave the intermediate (5R)-5-[(1S)-1,2-dihydroxyethyl]-
5-(hydroxymethyl)oxolan-2-one (79), which provided a
chiral platform for the synthesis of the target com-
pounds. Protecting the glycol moiety as the acetonide
and the remaining primary alcohol as a benzyl ether
allowed for complete differentiation between the two
5-substituents for later manipulation. Using the same
general alkylation procedures illustrated in Scheme 2,
the corresponding geometrical E-and Z-isomers (82 and
83) were obtained and separated chromatographically.
The geometry of the exocyclic double bond was assigned
in the same manner as before. Direct treatment of each
individual isomer with periodic acid cleaved the ac-
etonide group and oxidized the resulting glycol moiety
to the aldehyde. The intermediate aldehyde was reacted
immediately with hexyl (triphenylphosphoranylidene)-
acetate to give the expected olefins 84 and 85. Removal
of the benzyl ether with BCl3 afforded the two target
geometrical isomers 69 and 70.
The already reported enantioselective syntheses of 67
and 68 (Table 2) started from spirolactone 71 con-
structed from the relatively expensive 1,2:3,5-di-O-
isopropylidene-R-D-threo-apiofuranose (72).26 Synthesis
of an equally useful spirolactone synthon (73), obtained
from cheaper L-xylofuranose (74), was developed for this
project as a better alternative (Scheme 4). Hence,
protected 1,2-O-isopropylidene-R-L-xylofuranose (75) was
converted into spirolactone 73 as reported in a brief
communication38 and described here with full experi-
mental details (Scheme 5). Monoprotection of the pri-
mary alcohol group in 74 as a benzyl ether (75) and
oxidation to the 3-ulose intermediate (76) was followed
by the stereoselective addition of the Grignard reagent
(ClMgO(CH2)3MgCl) to the carbonyl from the less
congested convex face of the molecule as directed by the
1,2-O-isopropylidene function. Lactone formation from
77 occurred in two steps via an intermediate lactol
formed as the first oxidation product of the primary
alcohol moiety with pyridinium chlorochromate (PCC).
Further oxidation gave the key intermediate spirolac-
tone 73. The isopropylidene group was then hydrolyzed,
and the resulting glycol moiety was cleaved with sodium
metaperiodate to the corresponding aldehyde. Without
isolation, the aldehyde was reduced with sodium boro-
hydride to the alcohol with the concomitant hydrolysis
of the formate ester. Hydrogenolysis of the benzyl group
Lip op h ilicity a n d P a r tition Coefficien ts
The octanol/water partition coefficients (log P) were
calculated according to the fragment-based program
KOWWIN 1.6339 and are listed in Tables 1 and 2. This
parameter was correlated with PK-C binding affinity
(vide infra) by plotting log(1/Ki) versus log P. The four
curves shown in Figure 5 represent the best quadratic

Conformationally Constrained Analogues of DAG
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5 927
Ta ble 3. Percent of Docking Simulations in sn-1 and sn-2
Binding Modes
F igu r e 5. Binding affinity (log 1/Ki) versus partition coef-
ficients (log P) in octanol/water. The red (Z-isomers) and green
(E-isomers) curves correspond to the R-alkylidene branched
series (13-22), and the black (Z-isomers) and blue (E-isomers)
curves correspond to the acyl branched series (23-30).
fit calculated for two subgroups (Z- and E-isomers) of
DAG-lactones selected from Table 1. For the R-alky-
lidene branched series (13-22), the curves revealed that
affinity for PK-C reaches a maximum at ca. log P ) 6
for both Z- and E-isomers. In these curves, a consistent
greater affinity for the Z-isomers can be appreciated.
For the acyl branched series (23-30), a similar plot
revealed that PK-C affinity starts to drop much earlier
at ca. log P ) 5. Moreover, the difference between
geometrical isomers appeared smaller and was reversed
relative to the R-alkylidene branched series.
The doubly branched compounds (31 and 32) have log
P values 0.72 lower than the most lipophilic compounds
in the two previous subgroups and showed no apparent
difference between geometric isomers in terms of PK-C
binding (Table 1). These compounds appear to have the
most favorable log(1/Ki)/log P ratio compared to the most
potent R-alkylidene analogues (21 and 22). From these
data it can be concluded that an optimal log P for these
DAG-lactones lies between 5 and 6.
The “reversed ester” lactones featured in Table 2
showed that the R-alkylidene branched Z-isomer (69)
was the most potent ligand with the lowest log P value
(5.41). It is significant that the set of isomers 69/70 is
almost 2 log P units less lipophilic than the 67/68 pair
(Table 2), and yet the Z-isomer 69 is 3 times more potent
than Z-isomer 67. For comparison, phorbol 12,13-
dibutyrate (PDBU, Ki ) 0.2 nM)40 and prostratin (12-
deoxyphorbol-13-acetate, Ki ) 4.8 nM)40 have calculated
log P values of 3.43 and 3.50, respectively.
temperature is gradually reduced, while the ligand
translates, rotates, and changes its conformation. Au-
toDock provides a simple molecular mechanics force
field to calculate the intra- and intermolecular energies.
The ligands do not have hard degrees of freedom, i.e.,
bond stretching and angle bending are not allowed, and
rings are kept rigid. By default, the force field does not
provide 1,4-torsional parameters and relies upon non-
bonded interactions only. van der Waals interactions are
calculated using a Lennard-Jones 12-6 potential, whereas
the hydrogen-bonding term is modeled by Lennard-
J ones directional 12-10 potential. Electrostatics are
calculated with a distance-dependent dielectric function.
To speed the calculations of intermolecular energies
between the macromolecule and the ligands, nonbonded
interactions in the space surrounding the active site are
precalculated and stored in three-dimensional grids, one
for each atom type of the ligand. Docking simulations
were performed 100 times with different random seeds
to ensure the convergence of the results. Available
within the program is a clustering utility that was used
to group together similar docked conformations.
Molecu la r Mod elin g
DAG-lactones 9 and 10 [R-alkylidene branched, R2 )
CHdC(i-Pr)2], 15 and 16 [R-alkylidene branched, R2 )
CH2CH(i-Pr)2], 25 and 26 [acyl branched, R1 ) CH2CH-
(i-Pr)2], 31 and 32 [doubly branched, R1 ) R2 ) CH2-
CH(i-Pr)2], and 69 and 70 [“reversed esters”, R1
)
CH3(CH2)5O, R2 ) CH2CH(i-Pr)2] were docked into the
empty PK-Cδ C1b domain using the program Autodock
2.441 (Table 3). This program combines a Monte Carlo
simulated annealing algorithm to search the conforma-
tional space with a fast evaluation of the interaction
energy. The search strategy employed by the program
is a random walk of the ligand on the surface of the
receptor, which is kept rigid. During the search, the
Using this procedure, the crystallographic position of
phorbol-13-O-acetate in the C1b domain was duplicated
49 times out of 100 docking runs with a root-mean-
square-deviation (RMSD) ranging from 0.59 to 0.76 Å.
The lowest energy docking with a RMSD of 0.72 Å

928 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5
Nacro et al.
reproduced well the position, orientation, and hydrogen-
bonding network of the original crystal structure.27
Except for the “reversed ester” lactones 69 and 70,
which docked almost exclusively in an sn-2 mode, the
program engendered both sn-1 and sn-2 binding modes
in different ratios depending on the size, stereochem-
istry, and shape of the R1 and R2 groups on the DAG-
lactones (Table 3). Both sn-1 and sn-2 binding modes
had a comparable AutoDock scoring energy and dis-
played a similar network of hydrogen bonds to Thr 242,
Leu 251, and Gly 253, as observed with phorbol-13-O-
acetate,27 with participation of the primary OH and one
of the carbonyl functions (Figure 2, yellow dotted lines).
The combined total for the sn-1 and sn-2 binding modes
ranged from 79% to 100% for the series of 100 docking
runs, with the remaining docking runs corresponding
to energetically less favorable complexes.
There are no exceptionally revealing trends in Table
3 that could be strictly correlated with biological activity
(vide infra). However, in the case of the R-alkylidene
branched lactones with a linear acyl chain (9, 10, 15,
and 16), the preference for a specific binding mode
appears to be more sensitive to the differences in
stereochemistry about the R-alkylidene double bond. On
the other hand, the difference in preference between Z-
and E-isomers for acyl branched lactones 25 and 26
becomes minimal. This observation agrees with the data
plotted in Figure 5, where differences between geo-
metrical isomers were consistently more pronounced for
the R-alkylidene branched series. One of the most
important conclusions from these docking experiments,
however, is that both sn-1 and sn-2 binding modes were
observed, except in the case of the “reversed ester”
lactones 69 and 70, where the sn-2 binding mode is
preferred.
are equal or better than the heretofore most potent
DAG-lactones 67 and 68 (Table 2), but with six fewer
carbon atoms and 2.82 log P units lower. This suggests
that binding affinity in this series probably increased
by improving hydrophobic interactions with the C1
domain and not just by a coarse increase in lipophilicity.
As observed with other DAG-lactones, PK-C binding
affinity for this series was also sensitive to the steric
disposition of the R-alkylidene chain, but contrary to the
more typical behavior,24-26 it was the E-isomers that
showed a slight advantage.
The DAG-lactones belonging to the second subgroup
in Table 1 (compounds 13-18), which have a branched
4-methyl-3-(methylethyl)pentylidene [(CH2CH(i-Pr)2]
chain at R2, showed similar or improved affinities vis-
a`-vis the equivalent compounds of the first subgroup (7-
12). Also, these compounds showed the more charac-
teristic behavior where the Z-isomers possessed higher
affinities (ca. > 2-fold) than the corresponding E-
isomers.22-24 Longer acyl chains (R1 ) CnH2n+1, n ) 7
and 9) were investigated to optimize partitioning into
the phospholipid phase (compounds 19-22). However,
despite the improved affinities achieved as the length
of R1 increased, the effect was relatively modest and not
commensurate with the number of carbons (Figure 5).
Swapping the position of the branched chain from R2
to R1 (compounds 23-30, third subgroup, Table 1) was
not as effective. Although some potent compounds were
discovered, it appears that branching at R1 does not
have the same favorable outcome as branching at R2.
The earlier decline in affinity observed for a similar
range of log P values (Figure 5) suggests that the acyl
branch chain at R1 is probably binding at a different
site compared to the same type of branch at R2. In
addition, differences between geometrical isomers were
less pronounced and showed instead a slight preference
(ca. 1.2) for the E-isomers.
Biologica l Resu lts a n d Discu ssion
Having both R1 and R2 with equally branched chains
gave two of the most potent compounds (31 and 32,
Table 1), comparable or slightly better than compounds
19-22 of the second subgroup, but 0.72 log P units less
lipophilic. Compounds 31 and 32 have similar lipophilic
groups at both R1 and R2 and showed virtually no
difference between geometrical isomers in terms of
binding affinity.
Conversion of the R-alkylidene branch into an R-alkyl
branch was briefly explored. Diastereoisomers 65 and
66 (Scheme 3) showed that a branched R-alkyl moiety
was not as effective as an R-alkylidene chain. The Ki
values (nM) for these compounds were, respectively, 20
( 1.7 (cis-isomer 65) and 23 ( 0.8 (trans-isomer 66),
significantly poorer than 17 (5.9 ( 0.2) and 18 (11 (
0.6).
Finally, the two homochiral DAG-lactones 69 and 70
(log P ) 5.41, Table 2), which were designed to have a
hydrophobic/hydrophilic balance similar to that of the
potent DAG-lactone racemates 19 and 20 (log P ) 5.62,
Table 1), showed the largest difference (8.5-fold) so far
observed between two geometrical isomers. In agree-
ment with the modeling data (Table 3), this difference
suggests that there is a more specific and discriminating
interaction between each isomer and the C1 domain/
lipid complex of PK-C. The close agreement between the
measured Ki values for Z-isomers 69 and 19 is in
A. P K-C Bin d in g Affin ity. The interaction of the
target DAG-lactones (7-32 and 65-70) with PK-C was
assessed in terms of the ability of the ligand to displace
bound [20-3H]phorbol 12,13-dibutyrate (PDBU) from a
recombinant single isozyme (PK-CR) in the presence of
phosphatidylserine as already described.24-26,42 The
inhibition curves obtained for all ligands were of the
type expected for competitive inhibition, and the ID50
values were determined by fit of the data points to the
theoretical noncooperative competition curve. The Ki’s
for inhibition of binding (Tables 1 and 2) were calculated
from the ID50 values. The Ki values for the first
subgroup of DAG-lactones (7-12, Table 1) which have
a 4-methyl-3-(methylethyl)pent-2-enylidene [CHdC(i-
Pr)2] branched chain at R2 showed that there is a critical
need for an optimal overall hydrophobicity to produce
a stable complex with the enzyme. Indeed, for a constant
and invariant set of polar groups on the lactone moiety
that are responsible for receptor recognition and hydro-
gen bonding, the expected interaction of the branched
chain with the hydrophobic regions of the C1 domain
became evident only after reaching an adequate level
of lipophilicity (log P ) 4.56) with a five-carbon linear
acyl chain [R1 ) (CH2)4CH3]. This level of lipophilicity
was probably necessary for effective partitioning into
the phospholipid phase. Remarkably, the two most
potent ligands of this subgroup (compounds 11 and 12)

Conformationally Constrained Analogues of DAG
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5 929
keeping with our previous report on the bioequivalence
of “reversed ester” DAG-lactones and conventional DAG-
lactones.26 Assuming that both compounds bind at a
common site, one would have expected the active
enantiomer 69 to be twice as potent as the racemate
19. However, in reality these compounds are structur-
ally distinct and, as discussed in the modeling section,
they probably bind differently (i.e., mixed sn-1/sn-2
mode for 19 versus pure sn-2 mode for 69).
uration was necessary to match the binding affinity to
that of the corresponding conventional DAG-lactones.26
Docking experiments with AutoDock showed that these
“reversed ester” DAG-lactones bind preferentially in the
sn-2 binding mode. Remarkably, for compounds 69 (Z-
isomer) and 70 (E-isomer), the difference in binding
affinity increased by more than 8-fold, relative to the
ca. 2-fold difference observed for the rest of the DAG-
lactones.
A structure-activity analysis that correlated PK-C
binding affinity to a specific binding preference, sn-1 or
sn-2, could not be formulated. As shown in Figure 3,
when the doubly branched DAG-lactone 31 is docked
into the empty C1b domain of PK-Cδ, both branched
lipophilic ends of the molecule in either binding mode
can cover effectively the rim of the C1b domain. It is
therefore possible that after van der Waals interactions
between the segments of the chain involved in hydro-
phobic binding to the protein, the remainder of the
chain, linear or branched, could protrude outside into
the membrane. Since it is logical to assume that
interactions with the protein would be more sterically
sensitive than interactions with the membrane, the
more rigid R-alkylidene chain would show greater
differences in binding between Z- or E-isomer when
directly involved with the protein. The groove between
the loops formed by Met 239 and Leu 254 in the C1b
domain is more sterically constrained than the alterna-
tive site near Leu 250 and Trp 252. Therefore, the
specific preference for a particular binding mode will
determine which end of the molecule is to occupy the
more constrained site. For DAG-lactones 13-22 (Table
1), the sn-1 binding mode directs the R-alkylidene
branch into the more sterically demanding groove
between Met 239 and Leu 254. Hence, in this binding
mode, the differences in Ki values between Z- and
E-isomers would be expected to be larger as was
experimentally observed (Figure 5). On the other hand,
the results for the acyl branched DAG-lactones (23-
30, Table 1) could be explained by both binding alterna-
tives. In an sn-1 binding mode, the linear R-alkylidene
(R2) branch would lie inside the Met 239 and Leu 254
groove, but the lack of branching would result in a less
efficient discrimination between geometric isomers as
experimentally observed (Figure 5). In the alternative
sn-2 binding mode, it would be the branched acyl (R1)
chain occupying the Met 239 and Leu 254 groove, but
since there are no geometrical restrictions, the chain is
free to accommodate into an optimal disposition. At the
other end, the linear R-alkylidene (R2) chain would
reside in the less sterically demanding shallow groove
near Leu 250 and Trp 252 where little differences
between geometrical isomers would be expected.
B. P K-C Bin d in g Affin ity ver su s Lip op h ilicity.
The most potent DAG-lactones in either branched
R-alkylidene or branched acyl series appear to have an
adequate lipophilic/hydrophilic balance to effectively
localize into lipid bilayers. It is possible that in such an
environment these amphiphilic molecules, regardless of
their structure, may engage in some nonspecific interac-
tion with PK-C. However, when the hydrophilic groups
on the lactone template are additionally capable of
fitting well into the C1 domain of the enzyme, a
significant increase in binding affinity is observed. What
should therefore be the optimal relationship between
lipophilicity and potency? Comparing the most potent
DAG-lactones from this study (21, 31, 32, and 69) with
some prototypic phorbol esters, we find that the DAG-
lactone ligands are ca. 10-fold less potent than PDBU
(Ki ) 0.2-0.8 nM) and ca. 2-fold more potent than
prostratin (12-deoxyphorbol-13-acetate, Ki ) 4.8 nM),
a non-tumor-promoting phorbol ester. The calculated log
P values for PDBU and prostratin are very close (3.43
and 3.50, respectively) and are almost two log P units
lower than the least lipophilic lactone (compound 69,
log P ) 5.41). This means that relative to the natural
phorbol esters, it is still possible to modify the structures
of the DAG-lactones with different and smaller branched
motifs that could reduce lipophilicity and maintain or
increase binding affinity.
C. P K-C Activa tion . Direct stimulation of phospho-
rylation of the standard R-pseudosubstrate peptide by
PK-CR was measured for a selected group of compounds
(Table 4). When interpreting the results, it is important
to realize that the sensitivity of this assay is signifi-
cantly less than that of the binding assay, being carried
out under conditions of limiting phospholipid. Despite
this caveat, some salient points deserved to be men-
tioned. For this assay, three Z -isomers (15, 17, 21) from
the R-alkylidene branched group, two Z-isomers (27 and
29) from the acyl branched group, and the two doubly
branched Z- and E-isomers (31 and 32) from Table 1
were selected, together with Z-isomers 67 and 69 from
the “reversed ester” group in Table 2. One striking
observation for the R-alkylidene branched group is that
the ED50’s for activation do not strictly correlate with
PK-C binding affinities which are determined under
conditions of saturating anionic phospholipid. The ED50
values for the R-alkylidene branched group, which have
a linear acyl chain (R1), are the weakest enzyme
activators with an ED50 range of 0.6 to 2.3 µM. More-
over, only small differences were observed when the
linear acyl chain R1 was increased from three to nine
carbons, although in the end the most lipophilic com-
pound (21) was the most potent of the three. With acyl
branching, on the other hand, enzyme activation im-
proved and the ED50 values were reduced to less than
0.5 µM compared to the set of equally lipophilic com-
According to the docking results and the above
arguments, both competing binding modes are reason-
able, and the final selection is probably controlled by
the lipophilic imbalance between the two chains and
how they interact with the lipid phase.
The “reversed ester” DAG-lactones 67-70 (Table 2)
are an interesting case in that they are more confor-
mationally restricted. In addition to the R-alkylidene
chain (R2), the “reversed ester” moiety also forms part
of an unsaturated chain. It is important to recall that,
according to our previous work, this additional unsat-

930 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5
Ta ble 4. PK-CR Activation by Selected DAG-Lactones
Nacro et al.
ing. In the case of the doubly branched analogues (31
and 32), since both R1 and R2 groups are of similar
structure and lipophilicity, their mode of binding may
alternate between the sn-1 and sn-2 modes since they
are probably not effectively discriminated by the mem-
brane. Also, it is possible that during PK-C activation
the balance between the two binding modes might
change when the enzyme-ligand complex moves from
the aqueous phase to the lipid phase during transloca-
tion. Finally, in the case of the “reversed ester” lactones,
which probably assume a more exclusive sn-2 binding
mode, good activation was seen albeit somewhat lower
than with compounds 27, 29, 31, and 32.
D. EGF Bin d in g In h ibition . The epidermal growth
factor (EGF) receptor binds EGF, causing dimerization
and unleashing of tyrosine-specific kinase activity which
leads to autophosphorylation and downstream re-
sponses coupled to autophosphorylation.29 Among the
well-characterized inhibitors of EGF binding are the
phorbol esters,43 which act indirectly on EGF binding
through activation of PK-C. The ED50 values for inhibi-
tion of EGF binding by the DAG-lactones paralleled
closely the results on PK-C activation (Table 4). As with
PK-C activation, the simple R-alkylidene branched
DAG-lactones with simple linear acyl chains were again
the weakest inhibitors (compounds 15, 17, and 21). Acyl
branching, on the other hand, led to lower ED50 values.
As seen with PK-C activation, an increase in the
lipophilicity of the linear R-alkylidene chain was not
accompanied by a proportional reduction in the ED50
value (compare compound 27 and 29). The doubly
branched compounds (31 and 32) were again equivalent
and performed as well as the best acyl branched
compound 27. Finally, among the single enantiomers,
compound 69 was as potent as 27 and the 31/32 pair.
Since inhibition of EGF binding is an indirect effect of
PK-C activation in whole cells, it is difficult to postulate
a direct structure-activity correlation. However, as
observed with PK-C activation, it appears that the
nature of the acyl branch becomes more dominant in
an assay that is additionally sensitive to issues such as
uptake, transport, and stability of the acyl chains to the
action of esterases.
pounds in the R-alkylidene branched group (compare 17
with 27 and 21 with 29). Within experimental error,
the doubly branched Z- and E-isomers (31 and 32) can
be considered equally potent to the acyl branched DAG-
lactones 27 and 29. Finally, enantiomers 67 and 69 from
the “reversed ester” group had similar ED50 values close
to 1 µM.
As PK-C activators, the most potent of the DAG-
lactones investigated appears to be about 1 order of
magnitude less potent than PDBU (ED50 ) 0.02 µM).
The lack of good correlation between binding affinity
and PK-C activation within the DAG-lactones them-
selves [compare compounds 17 (Ki ) 5.9 nM) and 27
(Ki ) 7.3 nM) in Table 4] may be due to the more
dominant role played by the physical properties of the
membrane in the PK-C activation assay. Although
membrane interactions were not part of the structure-
activity analysis in the receptor-guided approach, the
data seem to suggest that a branched acyl chain at R1
is preferable for enzyme activation. Therefore, under the
conditions of limiting lipid used in the PK-C activation
assay, membrane interactions may override the weaker
van der Waal interactions with the C1 domain of the
enzyme that were optimized with R-alkylidene branch-
E. An titu m or Activity. The majority of the DAG-
lactones were tested in the NCI in vitro primary screen
which consists of a panel of 60 different tumor cell
lines.30 The results corresponding to the three com-
pounds that gave reproducible results in two separate
screens (compounds 29, 31, and 32) are presented in
Table 5. These compounds have been selected by the
NCI for referral to the Biological Evaluation Commit-
tee.44 A compound rejected by the screen (compound 21)
with comparable lipophilicity to the other three was
included for comparison. All the compounds in this table
have similar lipophilicities with log P values of 5.89 (31
and 32) and 6.61 (21 and 29). The first data entry in
Table 5 represents the mean 50% growth inhibition
parameter (GI50) for the entire panel of cells (MG-MID).
The next column (the so-called Delta value)45 shows the
number of log units by which the most sensitive line of
the panel differs from the corresponding mean for the
entire cell panel. Inspection of the first two data columns
shows that compound 31 displayed excellent specificity
with a Delta value of 2.51. The following three columns

Conformationally Constrained Analogues of DAG
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5 931
Ta ble 5. Results of Comparative Testing in the NCI in Vitro Screen
a
b
MG-MID is the calculated mean panel GI50 concentrations (M). Delta is the number of log10 units by which the delta of the most
sensitive line(s) of the panel differs from the corresponding MG-MID. The individual deltas are calculated by subtracting each log10 GI50
from the panel mean. c GI50 is an interpolated value representing the concentration (M) at which percentage growth is inhibited 50%.
d
COMPARE is a computerized, pattern-recognition algorithm to determine the degree of similarity (see text). e This cell line was tested
in a different experiment (GI50 ) 5.0 × 10-7 M). The GI50 profiles of the individual compounds were used as “seeds” against the historical
NCI database.
include data for the three most sensitive cell lines from
the leukemia, colon, and breast subpanels. On the basis
of the results from the cellular assays in Table 4, it is
not surprising that the R-alkylidene DAG-lactone 21 (log
P ) 6.61) with a linear acyl chain (R1) failed to show
antitumor activity despite having one of the lowest Ki
values (2.3 nM, Table 1) in the PK-C binding assay. On
the other hand, notwithstanding being a weaker PK-C
ligand, DAG-lactone 29 (Ki ) 13 nM, log P ) 6.61)
displayed important antitumor activity and selectivity
resulting simply from the switching of the branched acyl
and R-alkylidene groups. The doubly branched com-
pound 31 was undoubtedly the best compound, demon-
strating the advantage of both branched R-alkylidene
and acyl ester groups. It is intriguing, however, that
despite having similar Ki’s in the binding assay (Table
1) and similar ED50’s in PK-C activation and inhibition
of EGF binding (Table 4), the change in the geometrical
disposition of the R-alkylidene chain from the Z-(31) to
the E-isomer (32) had such a dramatic effect in antitu-
mor activity. This change in sensitivity, reflected in the
value of Delta, serves as a reminder once more that
membrane effects, which are not accounted for by any
of the models employed to date, are very important.
Finally, the COMPARE program was utilized to probe
the historical NCI database for compounds having an
activity profile similar to the two best compounds 29
and 31. COMPARE is a computerized, pattern-recogni-
tion algorithm that determines the degree of similarity,
or lack thereof, of mean-graph profiles generated by the
screen.46 The response profile fingerprints of compounds
29 and 31 were used as “seeds” to find closely matching
profiles contained in the NCI database. One of the most
important uses of COMPARE is to identify compounds
of unrelated chemical structure that share a related
biochemical mechanism of action. The result is a rank-
ordered list of similarity which is expressed in terms of
a correlation coefficient. We were pleased to see that
one of the highest ranked compounds (the highest for
compound 31 and third highest for compound 29) was
prostratin (12-deoxyphorbol-13-acetate), a non-tumor-
promoting antitumor agent.40,47 The antitumor results
show that despite our limited understanding of how the
membrane modulates PK-C activity, the equivalent
spatial disposition of pharmacophores in the DAG-
lactones and the phorbol nucleus, which has been the
paradigm in our pharmacophore- and receptor-guided
approaches, is corroborated by a completely unbiased
analysis of the antitumor data in the NCI screen. Of
additional interest was that other phorbol-like struc-
tures were identified by COMPARE. For example,
highly correlated to 31 were 13,20-diacetyl-12-octadi-
enoyl-4-deoxyphorbol (correl. coeff. ) 0.704), 20-acetyl-
13-isobutyryl-12-(2,4-octadienoic acid)-4-deoxyphorbol
(correl. coeff. ) 0.662), merezein (correl. coeff. ) 0.650),
and an ingenol analogue (correl. coeff. ) 0.627). Fur-
thermore, a direct comparison between prostratin48 and
the best DAG-lactone 31 demonstrated that the latter
is more potent and selective (Table 5). Because of PK-
C’s central role in signal transduction, the enzyme has
been an attractive target for therapeutic intervention,49
although efforts have been directed mostly to inhibiting
the enzyme with catalytic-targeted inhibitors.50 On the
other hand, despite two reported clinical uses of phorbol
esters,51,52 their close association with tumor promotion
has prevented their development as drugs. The already
reported strong antipromoting effect of prostratin,
coupled with its antitumor potential, has allayed fears
that PK-C activators are necessarily tumor promoters.53
Furthermore, bryostatin 1, which is a potent PK-C
activator, is currently in clinical trials14 and like pros-
tratin fails to promote.54 The results achieved here with
simple DAG analogues, which are structurally unrelated
to the phorbol esters, should provide a stronger incen-
tive to explore the role of PK-C activators in the control
of neoplasia.
Con clu sion s
1. The initial pharmacophore-guided approach that
was based on the structure of the phorbol esters

932 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5
Nacro et al.
culminated with the successful design of structurally
simple and potent DAG surrogates (DAG-lactones). In
this work, the pharmacophore-guided approach was
refined to a higher level with information gathered from
the X-ray structure of the C1b domain of PK-Cδ com-
plexed with phorbol-13-O-acetate. The basis of this
refinement, characterized as the receptor-guided ap-
proach, helped identify regions among a group of highly
conserved hydrophobic amino acids along the rim of the
C1b domain that allowed the implementation of impor-
tant structural modifications on the simple DAG-lactone
structure. In essence, the minimal “recognition domain”
provided by the DAG-lactone was modified with a
lipophilic “affinity domain” constituted mainly of
branched R-alkylidene or acyl chains that drove binding
affinities into the low nanomolar range. The branched
alkyl chains appear to improve binding at two distinct
levels. They appear to facilitate important van der
Waals contacts with hydrophobic segments of the pro-
tein and help promote the activation of PK-C through
critical membrane interactions. The latter may involve
facilitating bilayer-to-hexagonal phase transitions re-
sulting from an increase in headgroup spacing.
Exp er im en ta l Section
Molecu la r Mod elin g a n d Dock in g Exp er im en ts. An in-
house modified version of the AutoDock 2.4 package was used.
The source code was improved such that all variables are
initialized, array boundary errors and memory corruption are
avoided, and the input files are better validated. Memory usage
is optimized by allocating memory dynamically for the grids.
The implementation of the program Autotors, which is used
to interactively select rotatable bonds in the ligand, was
modified to automatically deselect bonds that involve a pair
of conjugated atoms. The crystal coordinates of the C1b domain
complexed with phorbol-13-acetate were retrieved from the
Brookhaven Protein DataBank (code 1ptr). Using Quanta 97,56
the phorbol-13-acetate was removed and polar hydrogens and
template-based partial atomic charges were added. The dock-
ing grids used were relatively large cubes (22.5 × 22.5 × 22.5
Å3), centered around the binding site to provide enough space
for the ligand to move and rotate freely in and out of the
binding site. The 3D structures of the DAG-lactones were
calculated with CORINA57 and further optimized with the
semiempirical quantum mechanics program Amsol 6.1.1 using
the AM1 Hamiltonian.58 Partial atomic charges were CM1
charges calculated with Amsol. For phorbol-13-acetate, the 3D
structure was taken from the crystal structure without further
minimization, and partial charges were calculated using the
Gasteiger-Marsili method.59 Nonpolar hydrogens were united
with their covalently bonded carbon atoms. For each ligand,
the docking simulation was repeated 100 times, starting each
time from a different position and orientation chosen ran-
domly.
Grids were calculated using the default force field param-
eters generated by the script gen-gpf. The spacing is 0.25 Å,
and the number of points were 90 in each dimension. The grid
center coordinates were 9.460, 26.699, and 21.364 Å relative
to the coordinates of the PDB file 1ptr. For the dockings, we
used the default intramolecular force field parameters gener-
ated by the script gen-dpf, except for hydrogen bonds with
oxygens which were modeled explicitly by the directional
Lennard-J ones 12-10 potential. The internal energy of the
ligands included the electrostatic interactions. The initial
coordinates were set to 14.6, 20.7, and 35.5 Å (inside the
protein) to force the program to select randomly a different
configuration of the ligand for each run. The simulated
annealing RT factor had initially a value of 2.00 cal/mol and
decreased linearly to 0 during 100 Monte Carlo cycles with a
maximum of 30 000 accepted and 30 000 rejected steps per
cycle. Each new cycle began with the state of minimum energy
from the previous annealing cycle. Initial and final values per
Monte Carlo step for the maximum translation, quaternion
rotation, and torsion were, respectively, 0.2 and 0.1 Å, 5.00
and 1.00 Å, and 10.00 and 1.00 Å. Clustering of the docked
conformations was performed with a tolerance of 1 Å.
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
as described previously with the single isozyme PK-CR.24-26,42
This recombinant isozyme was expressed in the baculovirus
system and was isolated as described in ref 60. The ID50 values
were determined from the competition curves, and the corre-
sponding Ki values for the ligands were calculated from the
ID50 values as described before.42 Values represent the mean
( standard error (three determinations).
An a lysis of P K-Cr Activa tion . PK-C activation was
determined by measuring the transfer of 32P from [γ-32P]ATP
to PK-CR pseudosubstrate peptide as a function of activating
ligand. PK-CR was incubated in a total volume of 50 µL in
the presence of 50 mM Tris-Cl, pH 7.4, 7.5 mM magnesium
acetate, 0.1 mM CaCl2, 0.25 mg/mL bovine serum albumin,
50 µM ATP, 1 µCi [γ-32P]ATP (specific activity > 5000 Ci/
mmol), 200 nM PK-CR pseudosubstrate peptide, 100 µg/mg
phospholipid (80% phosphatidylcholine/20% phosphatidylserine),
and activating ligand dissolved in dimethyl sulfoxide (final
concentration of dimethyl sulfoxide was < 0.2%). Incubation
was for 10 min at 30 °C. The reaction was then stopped by
2. From the standpoint of a structure-activity analy-
sis, the most important change in the progression from
the pharmacophore-guided approach to the receptor-
guided approach was that the latter suggests the
existence of competing sn-1 and sn-2 binding modes. In
the pharmacophore-guided approach, the best correla-
tion obtained from the superposition of either DAG or
DAG-lactones to phorbol esters favored the sn-1 binding
modality. This is in direct contrast with the “reversed
ester” DAG-lactones which appear to prefer the sn-2
mode. Further distinction between these two binding
modes will require a combinatorial assembly with a
constant “recognition domain” modified with a library
of unsymmetrically branched hydrophobic “affinity do-
mains”.
3. The structural simplicity of the potent DAG-
lactones should dispel the myth that chemical complex-
ity and pharmacological potency go hand in hand. Even
as racemates, the DAG-lactones show low namomolar
binding affinities for PK-C and display selective anti-
tumor activity at equivalent nanomolar levels. This is
a welcome finding for future efforts to exploit molecular
diversity for PK-C and other high-affinity receptors for
DAG and the phorbol esters. DAG is now known to
interact with four classes of high-affinity receptors,
comprising a total of 15 family members identified so
far.55 The diverse family of PK-C isozymes, because of
their ubiquitous distribution and extensive character-
ization, represent the primary challenge. The potential
for generating massive numbers of simple DAG-lactone
agonists compares favorably with the more complex task
of simplifying the structures of natural products, such
as the bryostatins15-17 and teleocidins.18-21 Finally, a
completely unbiased approach using the COMPARE
program to analyze the antitumor profile of the DAG-
lactones has corroborated the use of our pharmacophore-
and receptor-guided approaches. This effort demon-
strates how this combined approach can produce a novel
class of potentially useful antitumor DAG-lactones from
a set of first principles.

Conformationally Constrained Analogues of DAG
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5 933
chilling to 0 °C, and a 25 µL aliquot was spotted onto Whatman
P81 ion exchange paper to bind the phosphorylated peptide.
The paper was washed three times in 0.5% phosphoric acid
and three times in water to remove unbound ATP. Bound
radioactivity was then measured in a scintillation counter.
Background activity was measured in the absence of PK-CR.
Maximal stimulation was determined in the presence of 1 µM
PMA. Under the conditions of the assay, the response was
linear with the amount of PK-CR. Dose-response curves were
determined from 8 to 10 concentrations of ligand. In each
experiment, activation at each concentration of ligand was
determined in duplicate. ED50 values (doses yielding half-
maximal stimulation) represent the mean of three independent
experiments, unless otherwise indicated.
the catalyst was filtered off with the aid of Celite, and the
filtrate was concentrated. The crude product obtained was
purified by flash column chromatography on silica gel with
hexanes:Et2O (98:2) as eluant to give 4 (5.81 g, 95%) as a clear
oil: IR (neat) 2960-2875 and 1725 cm-1 (CdO); 1H NMR
(CDCl3) δ 9.85 (m, 1 H, CHO), 2.36 (dd, 2 H, J ) 5.2, 1.83 Hz,
CH2CHO), 1.83 m (2 H, 2 × (HCMe2)2), 1.75 (m, 1 H, HC(i-
Pr)2), 0.97 (d, 6 H, J ) 6.6 Hz, HC(CH3)2), 0.88 (d, 6 H, J )
6.8 Hz, HC(CH3)2); 13C NMR (CDCl3) δ 203.43, 44.70, 42.86,
29.30, 21.40, 18.84. This compound was used directly in the
following step.
3,3-Diisop r op ylp r op ion ic Acid [4-Meth yl-3-(m eth yl-
eth yl)p en ta n oic Acid ] (5). Oxidation of this aldehyde was
achieved quantitatively by a specific methodology.61,62 A stirred
solution of compound 4 (1 g, 7.0 mmol) in t-BuOH (9 mL) was
treated with 2-methyl-2-butene (2 mL, 0.7 mmol), 1.3 equiv
of NaClO2 (1.37 g, 9.1 mmol), and an aqueous solution of
NaHPO4 (7 mL, 1.67 M). The oxidation was allowed to proceed
at room temperature overnight, quenched with water (10 mL),
acidified with concentrated HCl to pH 2, and extracted with
Et2O (3 × 30 mL). The combined organic extract was dried
(MgSO4), filtered, and concentrated to give the crude acid as
a clear oil: IR (neat) 3050 (br), 2961, and 1708 cm-1; 1H NMR
(CDCl3) δ 2.31 (d, 2 H, J ) 5.8 Hz, CH2CO2H), 1.84 (sextuplet,
2 H, 2 × HCMe2)2), 1.70 (m, 1 H, HC(i-Pr)2), 1.00 (d, 6 H, J )
6.8 Hz, HC(CH3)2), 0.92 (d, 6 H, J ) 6.6 Hz, HC(CH3)2); 13C
NMR (CDCl3) δ 181.54, 46.55, 32.83, 29.32, 21.26, 18.64. This
compound was used directly in the following step.
An a lysis of In h ibition of EGF Bin d in g. Inhibition of
binding of epidermal growth factor (EGF) to C3H10T1/2 cells
was determined as follows. Cells were plated at 2.5 × 105 cells/
well (6 plate well) on day 1 in Eagle’s minimum essential
medium containing 10% fetal bovine serum, 2 mM glutamine,
and 100 µg/mL gentamicin and incubated at 37 °C in a
humidified atmosphere containing 5% CO2. Two days later,
cells were washed with Eagle’s minimum essential medium
and then treated with Eagle’s minimum essential medium
containing 1 mg/mL bovine serum albumin, 25 mM BES buffer
(N,N-bis[2-hydroxyethyl]-2-aminoethanesulfonic acid), pH 7.2,
and dimethyl sulfoxide or compound dissolved in dimethyl
sulfoxide for 60 min at 37 °C (final dimethyl sulfoxide
concentration < 0.1%). Cells were then placed on ice and
washed three times with cold Eagle’s minimum essential
medium. Fresh Eagle’s minimum essential medium containing
bovine serum albumin, BES buffer, and compound, together
with 0.1 µCi/well 125I-EGF in the absence or presence of 1 µg/
well of nonradioactive EGF. The cells were incubated for 5 h
in the cold room. Then, 100 µL supernatant was removed and
radioactivity measured to determine total activity per well.
The cells were washed three times with cold Eagle’s minimum
essential medium. The cells were finally lysed in 1 mL of 0.2
N NaOH, and aliquots of each lysate were counted in a
scintillation counter to determine bound radioactivity. The
average protein value was determined for 6 wells receiving
the same treatment as controls without 125I-EGF. Control
activity was determined in the absence of added ligand.
Background was determined in the presence of 1 µg/well of
nonradioactive EGF. Dose-response curves were typically
determined from six concentrations of ligand. In each experi-
ment, inhibition at each concentration of ligand was deter-
mined in triplicate. ID50 values (doses yielding half-maximal
inhibition) represent the mean of three independent experi-
ments, unless otherwise indicated.
Gen er a l Exp er im en ta l P r oced u r es. All chemical re-
agents were commercially available. Melting points were
determined on a MelTemp II apparatus, 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 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. Optical rotations were
recorded on a Perkin-Elmer model 241 polarimeter. Positive-
ion fast-atom bombardment mass spectra (FABMS) were
obtained on a VG 7070E mass spectrometer at an accelerating
voltage of 6 kV and a resolution of 2000. Glycerol was used as
the sample matrix and ionization was effected by a beam of
xenon atoms. Elemental analysis were performed by Atlantic
Microlab, Inc., Atlanta, GA.
3,3-Diisop r op ylp r op ion ic Acid Ch lor id e [4-Meth yl-3-
(m eth yleth yl)p en ta n oyl Ch lor id e] (6). Treatment of a
stirred solution of the crude acid (ca. 7 mmol) in Et2O (14 mL)
with oxalyl chloride (1 mL, 11.5 mmol) and pyridine (10 µL)
at 0 °C for 1 h, and then at room temperature for 48 h, afforded
the crude acyl chloride after removal of all the volatiles: IR
(neat) 2962 and 1802 cm-1; 1H NMR (CDCl3) δ 2.86 (d, 2 H, J
) 5.1 Hz, CH2COCl), 1.84 (m, 3 H, HC(HC(Me)2)2), 1.10 (d, 6
H, J ) 6.6 Hz, HC(CH3)2 ), 0.92 (d, 6 H, J ) 6.6 Hz, HC(CH3)2;
13C NMR (CDCl3) δ 174.51, 47.32, 46.50, 29.19, 21.18, 18.67.
This compound was distilled once and used directly in all the
acylation steps.
5,5-Bis[(2,2-d im e t h yl-1,1-d ip h e n yl-1-sila p r op oxy)-
m eth oxy]oxola n -2-on e (33).The synthesis of this compound
was reported earlier by Sharma et al.24
Sta n d a r d Alk yla tion P r oced u r e. A stirred solution of
lactone 33 in THF (5 mL/mmol) was cooled to -78 °C under
argon and treated dropwise with lithium diisopropylamide (1.4
equiv, 2 M solution in heptane/THF/ ethylbenzene). The
mixture was stirred at -78 °C for 1.5-2 h, and a solution of
the aldehyde (1.4 to 4 equiv) in THF (1 mL/mmol) was added
dropwise to the lithium enolate at the same temperature.
Stirring at -78 °C continued for 2 h, and the reaction was
quenched by the slow addition of a saturated aqueous solution
of ammonium chloride (1.25 mL/mmol) and then warmed to
room temperature. The aqueous layer was extracted three
times with Et2O (4.15 mL/mmol), and the combined organic
extracts were washed three times with water (2.50 mL/mmol),
twice with brine (2.50 mL/mmol), dried (MgSO4), and filtered.
The filtrate was concentrated under vacuum to afford the
crude alkylation reaction product, which was purified by flash
column chromatography after eluting with the appropriate
solvent. The obtained mixture of diastereoisomers was typi-
cally used directly in the following step without further
purification.
Sta n d a r d Mesyla tion -Olefin a tion P r oced u r e. A solu-
tion of the alkylation product in dichloromethane (10 mL/
mmol) at 0 °C was treated with methanesulfonyl chloride (2
eq) and triethylamine (4 eq). This mixture was stirred at 0 °C
for 30 min and then for 2 h at room temperature. 1,8-
Diazabicyclo[5.4.0]undec-7-ene (DBU, 5 eq) was added at 0 °C,
and the resulting solution was stirred overnight at ambient
temperature. The reaction mixture was concentrated to a
brown syrup and filtered through a pad of silica gel. The
3,3-Diisop r op yl-2-p r op en a l [4-Meth yl-3-(m eth yleth yl)-
p en t-2-en a l] (3). This compound was prepared according to
the procedure of Denmark.37
3,3-Diisop r op ylp r op ion a ld eh yd e [4-Meth yl-3-(m eth yl-
eth yl)p en ta n a l] (4). Compound 3 (6 g, 42.8 mmol) was
dissolved in CH2Cl2 (55 mL) and hydrogenated at 41 psi in
the presence of 10% Pd/C at ambient temperature. After 12 h

934 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5
Nacro et al.
filtrate was concentrated, and the residue was purified by flash
column chromatography after eluting with the appropriate
solvent.
compound 39 (>90%) was isolated as a mixture of diastereo-
isomers after flash chromatography with hexanes:EtOAc (95:5
to 9:1).
Sta n d a r d Desilyla tion P r oced u r e. A stirred solution of
the silylated compound in pyridine (2 mL/mmol) was treated
dropwise with 4 equiv of HF/pyridine (70% HF solution in
pyridine) while under argon at room temperature. Stirring was
continued for 2 h, and the reaction was quenched by the
addition of solid NaHCO3. Filtration and concentration of the
filtrate gave the deprotected residue which was purified by
flash column chromatography after eluting with the appropri-
ate solvent.
(Z)-5,5-Bis[(2,2-d im eth yl-1,1-d ip h en yl-1-sila p r op oxy)-
m et h yl]-3-[4-m et h yl-3-(m et h ylet h yl)p en t -2-en ylid en e)-
oxola n -2-on e (40) a n d (E)-5,5-Bis[(2,2-d im eth yl-1,1-d i-
ph en yl-1-silapr opoxy)m eth yl]-3-[4-m eth yl-3-(m eth yleth yl)-
p en t -2-en ylid en e)oxola n -2-on e (41). Under the standard
mesylation-olefination conditions starting from 34 (3.5 g, 4.6
mmol), three fractions were isolated after chromatography
with hexanes:EtOAc (95:5). The first fraction corresponded to
the Z-isomer (40, 0.903 g, 26%), the second fraction was a
mixture of both E- and Z-isomers (0.164 g, 4%), and the third
fraction was the E-isomer (41, 2.06 g, 60%).
Com p ou n d 40: oil; IR (CHCl3) 2961, 1744 (CdO), 1632
cm-1; 1H NMR (CDCl3) δ 7.46-7.70 (m, 21 H, Ph, >CdCHCHd
C(i-Pr)2), 7.06 (dd, 1 H, J ) 12.2, 1.2 Hz, >CdCHCHdC(i-
Pr)2), 3.79 (m, 4 H, CH2OSi), 3.20 (heptuplet, 1 H, J ) 6.8 Hz,
CHMe2), 2.97 (d, 2 H, J ) 2.4 Hz, H-4a,b), 2.56 (heptuplet, 1
H, J ) 6.8 Hz, CHMe2), 1.17-1.19 (2 d, 12 H, J ) 6.8 Hz,
HC(CH3)2), 1.07 (s, 18 H,t-Bu); 13C NMR (CDCl3) δ 169.58 (Cd
O), 163.93, 135.62, 135.58, 133.54, 132.94, 132.73, 129.74,
127.72, 121.63, 117.18, 84.29, 66.29, 34.19, 29.68, 29.56, 29.34,
26.65, 24.17, 21.21, 19.20. Anal. (C47H60O4Si2) C, H.
Sta n d a r d Mon oa cyla tion P r oced u r e. Meth od A. A
mixture consisting of the corresponding 5,5-bis(hydroxymethyl)-
oxolan-2-one (17 mL/mmol) dissolved in toluene, molecular
sieves 4 Å, and dibutyltin oxide (1.5-2 eq) was heated to reflux
for 1-2 h and then cooled to 0 °C before the addition of the
corresponding acyl chloride (1.0 eq). The reaction mixture was
stirred for 1 h at 0 °C and then quenched by the addition of
an aqueous solution of potassium phosphate buffer (pH 7, 10
mL/mmol). After filtering off the solids, the aqueous layer was
extracted three times with CHCl3 (28 mL/mmol). The combined
organic extract was dried (MgSO4), filtered, and concentrated.
The residue was purified by flash column chromatography with
the appropriate solvent to give the monoacyl derivative as the
major product and a smaller amount of the diacyl compound.
Meth od B. A solution of the 5,5-bis(hydroxymethyl)oxolan-
2-one in CH2Cl2 (16.22 mL/mmol) at room temperature was
treated with 2 equiv of pyridine and stirred for at least 1 h.
The acyl chloride (1.0-1.5 eq) was added at 0 °C, and the
mixture was allowed to stand at room temperature for 1 h.
The reaction was then concentrated under vacuum and the
residue purified by flash column chromatography with the
appropriate solvent to give the monoacyl derivative as the
major product.
5,5-Bis[(2,2-d im e t h yl-1,1-d ip h e n yl-1-sila p r op oxy)-
m et h yl]-3-[1-h yd r oxy-4-m et h yl-3-(m et h ylet h yl)p en t -2-
en yl]oxola n -2-on e (34). Using the standard alkylation
procedure starting from lactone 33 (15 g, 24.1 mmol) and 1.4
equiv of aldehyde 3, compound 34 (17.21 g, 95%) was obtained
as a mixture of diastereoisomers after column chromatography
with hexanes:EtOAc (9:1).
5,5-Bis[(2,2-d im e t h yl-1,1-d ip h e n yl-1-sila p r op oxy)-
methyl]-3-[1-hydroxy-4-methyl-3-(methylethyl)pentyl]oxolan-
2-on e (35). Using the standard alkylation procedure starting
from lactone 33 (8 g, 12.8 mmol) and 4 equiv of the aldehyde
4, compound 35 (9.80 g, 99%) was isolated as a mixture of
diastereoisomers after column chromatography with hexanes:
EtOAc (9:1).
5,5-Bis[(2,2-d im e t h yl-1,1-d ip h e n yl-1-sila p r op oxy)-
m eth yl]-3-(h ydr oxyeth yl)oxolan -2-on e (36). Using the stan-
dard alkylation procedure starting from lactone 33 (15 g, 24.1
mmol) and 2 equiv of acetaldehyde (2.69 mL, 48.2 mmol),
compound 36 (15.16 g, 94%) was isolated as a mixture of
diastereoisomers after flash chromatography with hexanes:
EtOAc (5:1).
5,5-Bis[(2,2-d im e t h yl-1,1-d ip h e n yl-1-sila p r op oxy)-
m eth yl]-3-(h ydr oxybu tyl)oxolan -2-on e (37). Using the stan-
dard alkylation procedure starting from lactone 33 (10 g, 16.1
mmol) and 3 equiv of n-butyraldehyde (4.34 mL, 72.1 mmol),
compound 37 (11.15 g, 99%) was isolated as a mixture of
diastereoisomers after flash chromatography with hexanes:
EtOAc (9:1 to 4:1).
5,5-Bis[(2,2-d im e t h yl-1,1-d ip h e n yl-1-sila p r op oxy)-
m eth yl]-3-(h yd r oxyh exyl)oxola n -2-on e (38). Using the
standard alkylation procedure starting from lactone 33 (10 g,
16.1 mmol) and 3 equiv of n-hexanal (5.90 mL, 49.1 mmol),
compound 38 (15.97 g, >90%) was isolated as a mixture of
diastereoisomers after flash chromatography with hexanes:
EtOAc (95:5 to 9:1).
5,5-Bis[(2,2-d im e t h yl-1,1-d ip h e n yl-1-sila p r op oxy)-
m et h yl]-3-(h yd r oxyd ecyl)oxola n -2-on e (39). Using the
standard alkylation procedure starting from lactone 33 (5 g,
8.0 mmol) and 1.5 equiv of n-decanal (2.3 mL, 12.2 mmol),
Com p ou n d 41: white solid; mp 118-119 °C; IR (CHCl3)
2958, 1741 (CdO), 1639; 1H NMR (CDCl3) δ 7.46-7.69 (m, 21
H, Ph, >CdCHCHdC(i-Pr)2), 5.99 (d, 1 H, J ) 12.2 Hz, >Cd
CHCHdC(i-Pr)2), 3.84 (AB d, 2 H, J ) 10.7 Hz, CHHOSi), 3.77
(AB d, 2 H, J ) 10.7 Hz, CHHOSi), 3.28 (heptuplet, 1 H, J )
6.8 Hz, CHMe2), 2.92 (d, 2 H, J ) 2.4 Hz, H-4a,b), 2.61
(heptuplet, 1 H, J ) 6.8 Hz, CHMe2), 1.17-1.19 (2 d, 12 H,
HC(CH3)2), 0.97-1.05 (2 s, 18 H, t-Bu); 13C NMR (CDCl3) δ
171.83 (CdO), 166.94, 135.57, 132.88, 132.72, 130.58, 129.77,
127.73, 124.16, 117.55 85.36, 66.31 (CH2OSi), 30.78, 30.26 (C-
4), 29.65, 26.66, 24.18, 21.11, 19.20. Anal. (C47H60O4Si2) C, H.
(Z)-5,5-Bis[(2,2-d im eth yl-1,1-d ip h en yl-1-sila p r op oxy)-
m eth yl]-3-[4-m eth yl-3-(m eth yleth yl)p en tylid en e]oxola n -
2-on e (42) a n d (E)-5,5-Bis[(2,2-d im eth yl-1,1-d ip h en yl-1-
sila p r op oxy)m eth yl]-3-[4-m eth yl-3-(m eth yleth yl)p en tyli-
d en e]oxola n -2-on e (43). Under the standard mesylation-
olefination conditions starting from 35 (6.65 g. 8.7 mmol), two
fractions were isolated after chromatography with mixtures
of hexane:petroleum ether:Et2O (80:16:4) and (70:24:6) as
eluants. The first fraction corresponded to the Z-isomer (42,
2.21 g, 34%) followed by the E-isomer (43, 2.99 g, 46%).
Com p ou n d 42: solid; mp 66-67 °C; IR (CHCl3) 3071-2958,
1757 (CdO), 1666, 1114 cm-1 1H NMR (CDCl3) δ 7.70 and
;
7.47 (m, 20 H, phenyl), 6.23 (pseudo t, 1 H, >CdCHCH2CH-
(i-Pr)2), 3.78 (s, 4 H, CH2OSi), 2.91 (narrow m, 2 H, H-4a,b),
2.80 (m, 2 H, >CdCHCH2CH(i-Pr)2), 1.86 (m, 2 H, CHMe2),
1.60 (m, 1 H, HC(i-Pr)2), 1.09 (s, 18 H, t-Bu), 0.99 (d, 6 H, J )
6.8 Hz, HC(CH3)2), 0.94 (d, 6 H, J ) 6.8 Hz, HC(CH3)2); 13C
NMR (CDCl3) δ 169.34, 145.55, 135.63, 135.58, 129.78, 127.75,
132.94, 132.76, 124.30, 84.28, 65.93, 51.15, 33.03, 29.31, 26.71,
26.07, 21.64, 19.50, 19.24. Anal. (C47H62O4Si2) C, H.
Com p ou n d 43: solid; mp 68-69 °C; IR (CHCl3) 3071-2858,
1761 (CdO), 1676, 1114 cm-1 1H NMR (CDCl3) δ 7.69 and
;
7.47 (m, 20 H, phenyl), 6.85 (m, 1 H, >CdCHCH2-CH(i-Pr)2),
3.82 (AB d, 2 H, J ) 10.7 Hz, CHHOSi), 3.76 (AB d, 2 H, J )
10.7 Hz, CHHOSi), 2.86 (br s, 2H, H-4a,b), 2.18 (pseudo t, 2 H,
J ) 7.1, 6.1 Hz, >CdCHCH2CH(i-Pr)2), 1.88 (m, 2 H, CHMe2),
1.30 (m, 1 H, HC(i-Pr)2), 1.27 (s, 18 H, t-Bu), 0.99 (d, 6 H, J )
6.6 Hz, HC(CH3)2), 0.94 (d, 6 H, J ) 6.8 Hz, HC(CH3)2); 13C
NMR δ 170.58, 141.59, 135.63, 135.57, 129.82, 127.77, 132.91,
132.93, 126.62, 85.18, 66.10, 50.21, 29.66, 29.15, 28.44, 26.68,
21.59, 19.43, 19.22. Anal. (C47H62O4Si2) C, H.
(Z)-5,5-Bis[(2,2-d im eth yl-1,1-d ip h en yl-1-sila p r op oxy)-
m eth yl]-3-eth ylid en eoxola n -2-on e (44) a n d (E)-5,5-Bis-
[(2,2-d im eth yl-1,1-d ip h en yl-1-sila p r op oxy)m eth yl]-3-eth -
ylid en eoxola n -2-on e (45). Under the standard mesylation-
olefination conditions starting from 36 (13.26 g, 19.9 mmol),
three fractions were isolated after chromatography with
mixtures of hexane:EtOAc (95:5 to 3:1) as eluant. The first
fraction corresponded to the Z-isomer (44, 3.71 g, 29%) followed

Conformationally Constrained Analogues of DAG
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5 935
by a mixture of Z- and E-isomers (1.97 g, 15%) and a third
fraction of the E-isomer (45, 6.72 g, 52%).
CH3), 1.56 and 1.40 (multiplets, 6 H, >CdCHCH2(CH2)3CH3),
1.08 (s, 18 H, t-Bu), 0.98 (br t, 3 H, >CdCH(CH2)4CH3); 13C
NMR (CDCl3) δ 170.40, 139.62, 135.46, 135.42, 132.74, 132.48,
129.67, 127.62, 127.35, 85.11, 66.08, 31.46, 30.06, 29.61, 27.74,
26.62, 22.41, 19.17, 13.91. Anal. (C44H56O4Si2) C, H.
(Z)-5,5-Bis[(2,2-d im eth yl-1,1-d ip h en yl-1-sila p r op oxy)-
m eth yl]-3-d ecylid en eoxola n -2-on e (50) a n d (E)-5,5-Bis-
[(2,2-d im eth yl-1,1-d ip h en yl-1-sila p r op oxy)m eth yl]-3-d e-
cylid en eoxola n -2-on e (51). Under the standard mesylation-
olefination conditions starting from 39 (5.33 g, 6.8 mmol), two
fractions were isolated after chromatography with a mixture
of hexanes:EtOAc (97:3 and 95:5) as eluant. The first fraction
corresponded to the Z-isomer (50, 1.02 g, 20%) followed by the
E-isomer (51, 3.39 g, 65%).
Com p ou n d 44: solid; mp 113-114 °C; IR (CHCl3) 1750 (Cd
O), 1674 cm-1; 1H NMR (CDCl3) δ 7.69 and 7.48 (m, 20 H, Ph),
6.28 (qt, 1H, J ) 7.3, 2.2 Hz, >CdCHCH3), 3.81 (AB d, 2 H, J
) 10.6 Hz, CHHOSi), 3.75 (AB d, 2H, J ) 10.6 Hz, CHHOSi),
2.90 (m, 2 H, H-4a,b), 2.25 (dt, 3 H, J ) 7.3, 2.2 Hz, >Cd
CHCH3), 1.09 (s, 18 H, t-Bu); 13C NMR (CDCl3) δ 169.36,
137.22, 135.50, 135.46, 132.77, 132.59, 129.69, 127.66, 126.41,
84.35, 66.10, 33.32, 26.66, 19.21, 13.93. Anal. (C40H48O4Si2) C,
H.
Com p ou n d 45: solid; mp 128-129 °C; IR (CHCl3) 1751 (Cd
O), 1684 cm-1; 1H NMR (CDCl3) δ 7.68 and 7.48 (m, 20 H, Ph),
6.84 (m, 1H, >CdCHCH3), 3.83 (AB d, 2 H, J ) 10.7 Hz,
CHHOSi), 3.77 (AB d, 2H, J ) 10.7 Hz, CHHOSi), 2.83 (br s,
2 H, H-4a,b), 2.25 (br d, 3 H, J ) 7.1 Hz, >CdCHCH3), 1.08 (s,
18 H, t-Bu); 13C NMR (CDCl3) δ 170.16, 135.47, 135.43, 134.09,
132.74, 132.51, 129.69, 128.59, 127.64, 85.11, 66.12, 29.64,
26.62, 19.18, 15.53. Anal. (C40H48O4Si2) C, H.
(Z)-5,5-Bis[(2,2-d im eth yl-1,1-d ip h en yl-1-sila p r op oxy)-
m et h yl]-3-bu t ylid en eoxola n -2-on e (46) a n d (E)-5,5-Bis-
[(2,2-d im eth yl-1,1-d ip h en yl-1-sila p r op oxy)m eth yl]-3-bu -
tylid en eoxola n -2-on e (47). Under the standard mesylation-
olefination conditions starting from 37 (10.43 g, 15.0 mmol),
two fractions were isolated after chromatography with a
mixture of hexane:petroleum ether:Et2O (8):16:4) as eluant.
The first fraction corresponded to the Z-isomer (46, 2.44 g,
24%) followed by the E-isomer (47, 5.78 g, 57%).
Com p ou n d 50: oil; IR (neat) 3071-2856, 1759 (CdO) cm-1
;
1H NMR (CDCl3) δ 7.71 and 7.41 (m, 20 H, Ph), 6.20 (br t, 1H,
J ≈ 7.5 Hz, >CdCH(CH2)8CH3), 3.78 (AB q, 4 H, J ) 10.7 Hz,
2 × CHHOSi), 2.91 (br d, 2 H, J ≈ 1.7 Hz, H-4a,b), 2.77 (br q,
2 H, J ≈ 7.0 Hz, >CdCHCH2(CH2)7CH3), 1.35-1.50 (m, 14 H,
>CdCHCH2(CH2)7CH3), 1.09 (s, 18 H, 2 × t-Bu), 0.97 (br t, 3
H, >CdCH(CH2)8CH3). Anal. (C48H64O4Si2) C, H.
Com p ou n d 51: oil, IR (neat) 3071-2857, 1763 (CdO) cm-1
;
1H NMR (CDCl3) δ 7.71 and 7.41 (m, 20 H, Ph), 6.79 (m, 1 H,
>CdCH(CH2)8CH3), 3.79 (AB q, 4 H, J ) 10.7 Hz, 2 × CH2-
OSi), 2.83 (br s, 2 H, H-4a,b), 2.23 (br q, 2 H, J ≈ 7.0 Hz, >Cd
CHCH2(CH2)7CH3), 1.60 and 1.35 (multiplets, 14 H, >Cd
CHCH2(CH2)7CH3), 1.08 (s, 18 H, 2 × t-Bu), 0.97 (br t, 3 H,
>CdCH(CH2)8CH3). Anal. (C48H64O4Si2) C, H.
Com p ou n d 46: solid; mp 70-75 °C; IR (neat) 3061-2990,
1744 (CdO), 1678 cm-1; 1H NMR (CDCl3) δ 7.70 and 7.47 (m,
20 H, Ph), 6.20 (m, 1H, >CdCH(CH2)2CH3), 3.79 (AB d, 2 H,
J ) 10.8 Hz, CHHOSi), 3.76 (AB d, 2H, J ) 10.8 Hz, CHHOSi),
2.91 (br d, 2 H, J ) 2.0 Hz, H-4a,b), 2.77 (br q, 2 H, J ) 7.3,
>CdCHCH2CH2CH3), 1.53 (multiplet, 2 H, >CdCHCH2CH2-
CH3), 1.09 (s, 18 H, t-Bu), 1.02 (br t, 3 H, >CdCH(CH2)2CH3);
13C NMR (CDCl3) δ 169.19, 142.89, 135.49, 135.45, 135.37,
132.77, 132.59, 129.68, 127.76, 127.74, 127.64, 125.49, 84.30,
66.01, 33.13, 29.47, 26.68, 22.34, 19.21, 13.79. Anal. (C42H52O4-
Si2‚0.5H2O) C, H.
Com p ou n d 47: solid; mp 72-73 °C; IR (neat) 3116-2906,
1740 (CdO), 1632 cm-1; 1H NMR (CDCl3) δ 7.70 and 7.46 (m,
20 H, Ph), 6.79 (m, 1H, >CdCH(CH2)2CH3), 3.82 (AB d, 2 H,
J ) 10.8 Hz, CHHOSi), 3.76 (AB d, 2H, J ) 10.8 Hz, CHHOSi),
2.84 (br s, 2 H, H-4a,b), 2.22 (q, 2 H, J ) 7.3 Hz, >CdCHCH2-
CH2CH3), 1.59 (multiplet, 2 H, >CdCHCH2CH2CH3), 1.07 (br
s, 18 H, t-Bu), 1.04 (br t, 3 H, >CdCH(CH2)2CH3); 13C NMR
(CDCl3) δ 170.38, 139.36, 135.48, 135.44, 132.75, 132.52,
129.69, 127.65, 127.58, 85.12, 66.11, 32.09, 29.67, 26.65, 21.43,
19.19, 13.89. Anal. (C42H52O4Si2‚0.29H2O) C, H.
(Z)-5,5-Bis[(2,2-d im eth yl-1,1-d ip h en yl-1-sila p r op oxy)-
m eth yl]-3-h exylid en eoxola n -2-on e (48) a n d (E)-5,5-Bis-
[(2,2-d im eth yl-1,1-d ip h en yl-1-sila p r op oxy)m eth yl]-3-h ex-
ylid en eoxola n -2-on e (49). Under the standard mesylation-
olefination conditions starting from 38 (11.60 g, 16.1 mmol),
two fractions were isolated after chromatography with a
mixture of hexanes:EtOAc (98:2 and 95:5) as eluant. The first
fraction corresponded to the Z-isomer (48, 2.69 g, 24%) followed
by the E-isomer (49, 8.26 g, 73%).
Com p ou n d 48: solid; mp 90-92 °C; IR (CHCl3) 3019-2862,
1744 (CdO), 1668 cm-1; 1H NMR (CDCl3) δ 7.71 and 7.46 (m,
20 H, Ph), 6.20 (br t, 1H, J ≈ 7.6 Hz, >CdCH(CH2)4CH3), 3.79
(AB d, 2 H, J ) 10.8 Hz, CHHOSi), 3.76 (AB d, 2H, J ) 10.8
Hz, CHHOSi), 2.91 (br d, 2 H, J ≈ 2.0 Hz, H-4a,b), 2.72 (br q,
2 H, J ≈ 7.3 Hz, >CdCHCH2(CH2)3CH3), 1.41 (m, 6 H, >Cd
CHCH2(CH2)3CH3), 1.09 (s, 18 H, t-Bu), 1.02 (br t, 3 H, >Cd
CH(CH2)4CH3); 13C NMR (CDCl3) δ 169.19, 143.18, 135.47,
135.42, 135.34, 132.75, 132.56, 129.64, 127.60, 125.23, 84.28,
65.98, 33.09, 31.41, 28.77, 27.48, 26.64, 22.46, 19.18, 13.96.
Anal. (C44H56O4Si2‚0.7H2O) C, H.
(Z)-5,5-Bis(h yd r oxym eth yl)-3-[4-m eth yl-3-(m eth yleth -
yl)p en t-2-en ylid en e]oxola n -2-on e (52). Starting from 40
(0.795 g, 1.2 mmol), the standard desilylation conditions were
applied. After chromatography with hexanes:EtOAc (1:1),
compound 52 (0.256 g, 80%) was obtained as white flakes: mp
132-133 °C; IR (CHCl3) 3599 (OH), 3414 (br OH), 3015-2966,
1
1740 (CdO), 1629 cm-1; H NMR (CDCl3) δ 7.36 (broad d, 1
H, J ) 11.76 Hz, >CdCHCH)C(i-Pr)2), 7.15 (d, 1 H, J ) 12.2
Hz, >CdCHCHdC(i-Pr)2), 3.85 (AB d, 2 H, J ) 12.2 Hz,
CHHOH), 3.79 (AB d, 2H, J ) 12.2 Hz, CHHOH), 3.20
(heptuplet, 1 H, J ) 6.8 Hz, CHMe2), 2.93 (d, 2H, J ) 1.7 Hz,
H-4a,b), 2.59 (broad s, 2 H, OH), 2.56 (heptuplet, 1 H, J ) 6.7
Hz, CHMe2) 1.18-1.15 (2 d, 12 H, J ) 6.8 Hz, HC(CH3)2); 13
C
NMR (CDCl3) δ 169.56 (CdO), 166.09, 135.94, 119.61, 116.89,
84.22, 65.13 (CH2OH), 33.04 (C-4), 29.62, 29.55, 29.48, 24.18,
21.17; FABMS (m/z, relative intensity) 269 (MH+, 100). Anal.
(C15H24O4) C, H.
(E)-5,5-Bis(h yd r oxym eth yl)-3-[4-m eth yl-3-(m eth yleth -
yl)p en t-2-en ylid en e]oxola n -2-on e (53). Starting from 41
(1.33 g, 2.0 mmol) the standard desilylation conditions were
applied. After chromatography with hexanes:EtOAc (1:1),
compound 53 (0.554 g, 100%) was obtained as white needles:
mp 131-132 °C; IR (CHCl3) 3599 (OH), 3414 (br OH), 3015,
1
1742 (CdO), 1634 cm-1; H NMR (CDCl3) δ 7.61 (broad d, 1
H, J ) 12.2 Hz, >CdCHCHdC(i-Pr)2), 5.99 (d, 1 H, J ) 12.2
Hz, >CdCHCH)C(i-Pr)2), 3.89 (AB d, 2 H, J ) 12.2 Hz,
CHHOH), 3.81 (AB d, 2H, J ) 12.2 Hz, CHHOH), 3.26
(heptuplet, 1 H, J ) 6.8 Hz, CHMe2), 2.92 (d, 2H, J ) 2.2 Hz,
H4a,b), 2.59 (heptuplet, 1 H, J ) 6.8 Hz, CHMe2), 2.56 (broad
s, 2 H, OH), 1.16 (d, 12 H, J ) 6.8 Hz, HC(CH3)2); 13C NMR
(CDCl3) δ 172.04 (CdO), 168.70, 132.71, 122.35, 117.48, 85.36
(C-5), 65.13 (CH2OH), 30.51, 29.68, 29.72 (C-4), 24.09, 21.07;
FABMS (m/z, relative intensity) 269 (MH+, 100). Anal.
(C15H24O4‚0.3H2O) C, H.
(Z)-5,5-Bis(h yd r oxym eth yl)-3-[4-m eth yl-3-(m eth yleth -
yl)p en tylid en e]oxola n -2-on e (54). Starting from 42 (4.05
g, 5.4 mmol) the standard desilylation conditions were applied.
After chromatography with mixtures hexanes:EtOAc (1:1) and
(1:2), compound 54 (1.37 g, 93%) was obtained as a clear oil:
1
IR (neat) 3432 (OH), 2958-2871, 1725 (CdO), 1669 cm-1; H
NMR (CDCl3) δ 6.34 (tt, 1 H, J ) 7.3, 1.9 Hz, >CdCHCH2-
CH(i-Pr)2), 3.83 (AB d, 2 H, J ) 12.1 Hz, CHHOH), 3.77 (AB
d, 2 H, J ) 12.1 Hz, CHHOH), 2.85 (br d, 2H, J ) 1.9 Hz,
H-4a,b), 2.78 (ddd, 1 H, J ) 7.3, 5.9, 1.9 Hz, >CdCHCH2-CH-
(i-Pr)2), 2.50 (br s, 2 H, OH), 1.85 (heptuplet, 2 H, CHMe2),
1.19 (quintuplet, 1 H, CH(i-Pr)2), 0.98-0.91 (2d, 12 H, J ) 6.8
Com p ou n d 49: oil; IR (neat) 3061-2862, 1761 (CdO), 1680
cm-1; 1H NMR (CDCl3) δ 7.70 and 7.46 (m, 20 H, Ph), 6.79 (m,
1H, >CdCH(CH2)4CH3), 3.82 (AB d, 2 H, J ) 10.7 Hz,
CHHOSi), 3.75 (AB d, 2H, J ) 10.7 Hz, CHHOSi), 2.83 (br s,
2 H, H-4a,b), 2.23 (br q, 2 H, J ≈ 7.3 Hz, >CdCHCH2(CH2)3-

936 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5
Nacro et al.
Hz, HC(CH3)2); 13C NMR (CDCl3) δ 169.40, 146.91, 122.72,
84.22, 63.96, 50.50, 32.34, 28.73, 25.72, 21.05, 18.86; FABMS
(m/z, relative intensity) 271 (MH+, 100). Anal. (C15H26O4‚
0.16H2O) C, H.
) 12.0, 6.1 Hz, 2 × CHHOH), 3.78 (dd, 2 H, J ) 12.0, 5.3 Hz,
2 × CHHOH), 2.86 (m, 2 H, >CdCHCH2(CH2)3CH3), 2.80 (m,
2 H, H-4a,b), 1.40-1.50 (m, 6 H, >CdCHCH2(CH2)3CH3), 0.96
(m, 3 H, >CdCH(CH2)4CH3); 13C NMR (CDCl3) δ 171.12,
141.83, 126.38, 85.72, 64.62, 31.40, 30.13, 29.26, 27.64, 22.34,
13.88; FABMS (m/z, relative intensity) 229 (MH+, 100). This
compound was used without further purification in the next
acylation step.
(E)-5,5-Bis(h yd r oxym eth yl)-3-[4-m eth yl-3-(m eth yleth -
yl)p en tylid en e]oxola n -2-on e (55). Starting from 43 (2.06
g, 7.6 mmol) the standard desilylation conditions were applied.
After chromatography with mixtures hexanes:EtOAc (1:1),
(1:2), and (1:3), compound 55 (1.97 g, 96%) was obtained as a
clear oil: IR (CHCl3) 3406 (OH), 2958-2873, 1739 (CdO), 1674
cm-1; 1H NMR (CDCl3) δ 6.84 (m, 1 H, >CdCHCH2CH(i-Pr)2),
3.83 (AB d, 2 H, J ) 12.1 Hz, CHHOH), 3.76 (AB d, 2 H, J )
12.1 Hz, CHHOH), 3.65 (m, 2 H, OH), 2.77 (br s, 2H, H-4a,b),
2.19 (pseudo t, 2 H, >CdCHCH2CH(i-Pr)2), 1.83 (heptuplet, 2
H, CHMe2), 1.27 (quintuplet, 1 H, HC(i-Pr)2), 0.97 (d, 6 H, J
) 6.8 Hz, HC(CH3)2), 0.91 (d, 6H, J ) 6.6 Hz, HC(CH3)2); 13C
NMR (CDCl3) δ 170.61, 143.13, 124.92, 85.16, 64.09, 49.69,
29.02, 28.65, 28.09, 21.03, 18.77; FABMS (m/z, relative inten-
sity) 271 (MH+, 100). Anal. (C15H26O4‚0.2H2O) C, H.
(Z)-5,5-Bis(h yd r oxym e t h yl)-3-e t h ylid e n e oxola n -2-
on e (56). Starting from 44 (2.59 g, 4.0 mmol) the standard
desilylation conditions were applied. After chromatography
with mixtures hexanes:EtOAc (1:2) and (1:5), 0.62 g (90%) of
a ca. 1:1 mixture of E- and Z-isomers (56 and 57) was obtained:
IR (Nujol) 3365 (OH), 1740 (CdO), 1764 cm-1; 1H NMR (CDCl3)
δ 6.75 and 6.39 (multiplets, 1 H, >CdCHCH3); FABMS (m/z,
relative intensity) 173 (MH+, 100).
(E )-5,5-Bis(h yd r oxym e t h yl)-3-e t h ylid e n e oxola n -2-
on e (57). Starting from 45 (5.39 g, 8.3 mmol) the standard
desilylation conditions were applied. After chromatography
with mixtures hexanes:EtOAc (1:2) and (1:5), compound 57
(1.43 g, 100%) was obtained as a white solid: mp 105-106 °C;
IR (Nujol) 3430-3354 (OH), 1732 (CdO), 1674 cm-1; 1H NMR
(CDCl3) δ 6.75 (m, 1 H, >CdCHCH3), 3.72 (AB d, 2 H, J )
12.0 Hz, 2 × CHHOH), 3.64 (AB d, 2 H, J ) 12.0 Hz, 2 ×
CHHOH), 2.83 (br m, 2 H, H-4a,b), 1.92 (dt, 3H, J ) 7.1, 1.7
Hz, >CdCHCH3); 13C NMR (CD3OD) δ 172.83, 136.34, 130.30,
87.67, 65.34, 29.84, 15.84; FABMS (m/z, relative intensity) 173
(MH+, 100). Anal. (C8H12O4) C, H.
(E )-5,5-Bis(h yd r oxym e t h yl)-3-h e xylid e n e oxola n -2-
on e (61). Starting from 49 (8.00 g, 11.3 mmol) the standard
desilylation conditions were applied. After chromatography
with mixtures hexanes:EtOAc (1:1) and (1:2), compound 61
(2.16 g, 83%) was obtained as a clear oil: IR (CHCl3) 3381 (OH),
1
2956-2860, 1747 (CdO), 1679 cm-1; H NMR (CDCl3) δ 6.80
(m, 1 H, >CdCH(CH2)4CH3), 3.85 (dd, 2 H, J ) 12.0, 5.2 Hz,
CHHOH), 3.76 (dd, 2 H, J ) 12.0, 4.7 Hz, CHHOH), 3.44 (br
t, 2 H, CH2OH), 2.77 (br s, 2 H, H-4a,b), 2.24 (br q, 2 H, J ≈ 7.3
Hz, >CdCHCH2(CH2)3CH3), 1.37-1.54 (m, 6 H, >CdCHCH2-
(CH2)3CH3), 0.96 (br t, 3 H, J ≈ 6.5 Hz, >CdCH(CH2)4CH3);
13C NMR (CDCl3) δ 171.12, 141.83, 126.38, 85.72, 64.62, 31.40,
30.13, 29.26, 27.64, 22.34, 13.88; FABMS (m/z, relative inten-
sity) 229 (MH+, 100). This compound was used without further
purification in the next acylation step.
(Z)-5,5-Bis(h yd r oxym e t h yl)-3-d e cylid e n e oxola n -2-
on e (62). Starting from 50 (4.89 g, 6.5 mmol) the standard
desilylation conditions were applied. After chromatography
with mixtures hexanes:EtOAc (2:3) and (1:4), compound 62
(1.77 g, 100%) was obtained as a clear oil: IR (CHCl3) 3396
1
(OH), 2962-2855, 1742 (CdO), 1669 cm-1; H NMR (CDCl3)
δ 6.33 (tt, 1 H, J ) 7.8, 2.2 Hz, 1 H, >CdCH(CH2)8CH3), 3.82
(d AB q, 4 H J ) 12.1, 6.4 Hz, 2 × CH2OH), 2.88 (m, 2 H,
H-4a,b), 2.78 (m, 2 H, >CdCHCH2(CH2)7CH3), 2.12 (t, 1 H, J
) 6.4 Hz, OH), 1.34-1.60 (m, 14 H, >CdCH(CH2)7CH3), 0.96
(br t, 3 H, >CdCH(CH2)8CH3); 13C NMR (CDCl3) δ 169.20,
145.76, 123.89, 84.22, 64.96, 32.60, 31.85, 31.81, 29.63, 29.44,
29.37, 29.29, 29.21, 28.99, 27.72, 22.61, 14.05; FABMS (m/z,
relative intensity) 285 (MH+, 100). Anal. (C16H28O4) C, H.
(E )-5,5-Bis(h yd r oxym e t h yl)-3-d e cylid e n e oxola n -2-
on e (63). Starting from 51 (6.58 g, 8.6 mmol) the standard
desilylation conditions were applied. After chromatography
with mixtures hexanes:EtOAc (2:3) and (1:4), compound 63
(2.29 g, 93%) was obtained as a clear oil: IR (CHCl3) 3408-
(Z)-5,5-Bis(h yd r oxym e t h yl)-3-b u t ylid e n e oxola n -2-
on e (58). Starting from 46 (1.05 g, 1.6 mmol) the standard
desilylation conditions were applied. After chromatography
with mixtures hexanes:EtOAc (1:2) and (1:3), compound 58
(0.22 g, 70%) was obtained as an oil: IR (neat) 3395 (OH),
3018 (OH), 2918-2853, 1747 (CdO), 1678 cm-1 1H NMR
;
(CDCl3) δ 6.74 (m, 1 H, >CdCH(CH2)8CH3), 3.85 (AB q, 4 H J
) 12.1 Hz, 2 × CH2OH), 2.71 (m, 2 H, H-4a,b), 2.16 (br q, 2 H,
J ≈ 7.3 Hz, >CdCHCH2(CH2)7CH3), 1.97 (br s, 1 H, OH), 1.46
(m, 2 H, >CdCHCH2CH2(CH2)6CH3), 1.24 (m, 12 H, >Cd
CHCH2(CH2)6CH3), 0.86 (br t, 3 H, >CdCH(CH2)8CH3); 13C
NMR (CDCl3) δ 170.34, 142.26, 126.11, 84.49, 65.22, 31.82,
30.27, 29.45, 29.34, 29.23, 28.03, 22.68, 14.06; FABMS (m/z,
relative intensity) 285 (MH+, 100). Anal. (C16H28O4) C, H.
1
2958-2870, 1738 (CdO), 1667 cm-1; H NMR (CDCl3) δ 6.31
(m, 1 H, >CdCH(CH2)2CH3), 3.82 (AB d, 2 H J ) 12.2 Hz,
CHHOH), 3.76 (AB d, 2H, J ) 12.1 Hz, CHHOH), 3.34 (s, 2
H, CH2OH), 2.85 (br d, 2 H, J ) 2.0 Hz, H-4a,b), 2.73 (br q, 2
H, J ∼ 7.3 Hz, >CdCHCH2CH2CH3), 1.53 (sextuplet, 2 H, J
∼ 7.3 Hz, >CdCHCH2CH2CH3), 1.01 (t, 3 H, J ∼ 7.3 Hz, >Cd
CH(CH2)2CH3); 13C NMR (CDCl3) δ 169.95, 145.02, 124.52,
84.87, 64.37, 32.54, 29.52, 22.17, 13.64; FABMS (m/z, relative
intensity) 201(MH+, 100). Anal. (C10H16O4‚0.13H2O) C, H.
5,5-Bis(h yd r om eth yl)-3-[4-m eth yl-3-(m eth yleth yl)p en -
tyl]oxola n e-2-on e (64). A solution of 53 (0.220 g, 0.8 mmol)
in 10 mL of ethanol was reduced in a Parr hydrogenator under
40 psi of hydrogen for 6 h in the presence of 10% Pd/C (0.022
g). The catalyst was filtered off, and the ethanol was evapo-
rated to give a crude oil that was purified by flash column
chromatography with hexanes:EtOAc (1:1 to 1:2) as eluant to
give 0.224 mg (100%) of 64: oil; IR (neat) 3388 (OH), 2956-
(E )-5,5-Bis(h yd r oxym e t h yl)-3-b u t ylid e n e oxola n -2-
on e (59). Starting from 47 (4.5 g, 6.7 mmol) the standard
desilylation conditions were applied. After chromatography
with mixtures hexanes:EtOAc (1:2), (1:3), and (1:5), compound
59 (1.24 g, 93%) was obtained as a white solid: mp 71-72 °C;
IR (Nujol) 3382 (OH), 3017-2873, 1743 (CdO), 1778 cm-1; 1H
NMR (CDCl3) δ 6.80 (m, 1 H, >CdCH(CH2)2CH3), 3.85 (AB d,
2 H J ) 12.1 Hz, CHHOH), 3.76 (AB d, 2H, J ) 12.1 Hz,
CHHOH), 3.16 (s, 2 H, CH2OH), 2.79 (br s, 2 H, H-4a,b), 2.23
(br q, 2 H, J ∼ 7.3 Hz, >CdCHCH2CH2CH3), 1.59 (sextuplet,
2 H, J ∼ 7.3 Hz, >CdCHCH2CH2CH3), 1.02 (t, 3 H, J ∼ 7.3
Hz, >CdCH(CH2)2CH3); 13C NMR (CDCl3) δ 171.19, 141.47,
126.69, 85.82, 64.53, 32.09, 29.26, 21.31, 13.77; FABMS (m/z,
relative intensity) 201 (MH+, 100). Anal. (C10H16O4) C, H.
1
2873 and 1747 (CdO) cm-1; H NMR (CDCl3) δ 3.84 (AB q, 2
H J ) 12.2 Hz, CHHOH), 3.87 (AB q, 2H, J ) 12.2 Hz,
CHHOH), 3.20 (br s, 2 H, CH2OH), 2.89 (ddd, 1 H, J ) 14.0,
9.5, 4.2 Hz, H-3), 2.41 (dd, 1H, J ) 12.9, 10.1 Hz, H-4a), 1.75-
2.00 (m, 5 H, H-4b, CH2CH2CH(i-Pr)2), 1.20-1.58 (m, 3 H,
CH(CHMe2)2), 0.90 (m, 12 H, 2 × CH(CH3)2); 13C NMR (CDCl3)
δ 179.36, 86.13, 64.59, 64.25, 49.50, 40.49, 31.71, 31.23, 24.85,
28.75, 28.57, 20.95, 20.81, 18.86, 18.65; FABMS (m/z, relative
intensity) 273 (MH+, 100). Anal. (C15H28O4) C, H.
(Z)-5,5-Bis(h yd r oxym e t h yl)-3-h e xylid e n e oxola n -2-
on e (60). Starting from 48 (2.29 g, 3.3 mmol) the standard
desilylation conditions were applied. After chromatography
with mixtures hexanes:EtOAc (1:1) and (1:2), compound 60
(0.66 g, 89%) was obtained as a clear oil: IR (neat) 3400 (OH),
1,4-cis-{1-(Hyd r oxym eth yl)-4-[4-m eth yl-3-(m eth yleth -
yl)p en tyl]-3-oxo-2-oxola n yl}m eth yl Hexa n oa te (65) a n d
1,4-tr a n s-{1-(Hyd r oxym eth yl)-4-[4-m eth yl-3-(m eth yleth -
yl)p en tyl]-3-oxo-2-oxola n yl}m eth yl Hexa n oa te (66). Stan-
dard monoacylation conditions (method A) starting from 64
(0.100 g, 0.37 mmol) were applied. After chromatography with
1
2928-2863, 1739 (CdO), 1671 cm-1; H NMR (CDCl3) δ 6.32
(tt, 1 H, J ) 7.7, 2.2 Hz, >CdCH(CH2)4CH3), 3.84 (dd, 2 H, J

Conformationally Constrained Analogues of DAG
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5 937
hexanes:EtOAc (3:1), some diacylated product was isolated
(0.047 g, 27%), together with the 1,4-cis-monoacylated isomer
65 (0.033 g (24%) and the 1,4-trans-monoacylated isomer 66
(0.014 g, 10%).
(Z)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
pen t-2-en yliden e]-3-oxo-2-oxolan yl}m eth yl Bu tan oate (9).
Standard monoacylation conditions (method A) starting from
52 (0.203 g, 0.8 mmol) were applied. After chromatography
with different gradients of hexanes:EtOAc (3:1), (2:1), and
(1:1), the monoacylated product 9 (0.161 g, 63%) was isolated
together with 0.016 g (21%) of the diacylated material.
Compound 9: oil, IR (CHCl3) 3458 (OH), 2964-2872, 1746 (Cd
Com p ou n d 65: oil; IR (neat) 3451 (OH), 2955-2870, 1773
and 1744 (CdO) cm-1; 1H NMR (CDCl3) δ 4.34 (AB d, 1 H J )
11.5 Hz, CHHOCO(CH2)4CH3), 4.25 (d, 1 H, J ) 11.5 Hz,
CHHOCO(CH2)4CH3), 3.85 (AB d, 1 H, J ) 12.2 Hz, CHHOH),
3.68 (AB d, 1 H, J ) 12.2 Hz, CHHOH), 2.85 (ddd, 1 H, J )
14.0, 9.8, 4.5 Hz, H-3), 2.45 (t, 2 H, J ) 7.4 Hz, COCH2(CH2)3-
CH3), 2.40 (m, 1 H, H-4a), 1.75-2.10 (m, 5 H, H-4b, CH2CH2-
CH(i-Pr)2), 1.70 (m, 2 H, COCH2CH2(CH2)4CH3), 1.25-1.60 (m,
7 H, CO(CH2)2(CH2)2CH3, CH(CHMe2)2), 0.95 (m, 15 H, CO-
(CH2)4CH3, 2 × CH(CH3)2); 13C NMR (CDCl3) δ 178.04, 173.19,
83.72, 65.28, 64.93, 49.99, 40.63, 33.98, 32.14, 31.79, 31.14,
29.25, 29.06, 25.31, 24.45, 22.21, 21.44, 21.29, 19.34, 19.14,
13.80; FABMS (m/z, relative intensity) 371 (MH+, 62), 99
(C5H11CO+, 99). Anal. (C21H38O5) C, H.
Com p ou n d 66: oil; IR (neat) 3459 (OH), 2955-2870, 1774
and 1744 (CdO) cm-1; 1H NMR (CDCl3) δ 4.43 (AB d, 1 H J )
12.2 Hz, CHHOCO(CH2)4CH3), 4.15 (d, 1 H, J ) 12.2 Hz,
CHHOCO(CH2)4CH3), 3.80 (AB d, 1 H, J ) 11.5 Hz, CHHOH),
3.70 (AB d, 1 H, J ) 11.5 Hz, CHHOH), 2.86 (ddd, 1 H, J )
14.0, 10.0, 4.5 Hz, H-3), 2.55 (dd, 1 H, J ) 12.94, 10.0 Hz,
H-4a), 2.45 (t, 2 H, J ) 7.4 Hz, COCH2(CH2)3CH3), 1.65-2.10
(m, 7 H, H-4b, CH2CH2CH(i-Pr)2, COCH2CH2(CH2)4CH3), 1.25-
1.50 (m, 7 H, CO(CH2)2(CH2)2CH3, CH(CHMe2)2), 0.95 (m, 15
H, CO(CH2)4CH3, 2 × CH(CH3)2); 13C NMR (CDCl3) δ 177.70,
172.98, 83.12, 65.31, 64.53, 50.02, 40.40, 33.92, 32.36, 32.03,
31.14, 29.25, 29.06, 25.28, 24.42, 22.16, 21.39, 21.26, 19.27,
19.08, 13.78; FABMS (m/z, relative intensity) 371 (MH+, 62),
99 (C5H11CO+, 99). Anal. (C21H38O5) C, H.
1
O), 1632 cm-1; H NMR (CDCl3) δ 7.28 (d, 1 H, J ) 12.1 Hz,
>CdCHCH)C(i-Pr)2), 7.04 (dm, 1 H, J ) 12.2, >CdCHCHd
C(i-Pr)2), 4.25 (AB d, 1 H, J ) 11.9 Hz, CHHOC(O)n-Pr), 4.17
(AB d, 1H, J ) 11.9 Hz, CHHOC(O)n-Pr), 3.68 (AB d, 1 H, J
) 12.2 Hz, CHHOH), 3.61 (AB d, 1 H, J ) 12.2 Hz, CHHOH),
3.09 (heptuplet, 1 H, J ) 6.9 Hz, HCMe2), 2.93 (dd, 1 H, J )
17.2, 2.1 Hz, H-5a), 2.75 (dd, 1 H, J ) 17.2, 1.9 Hz, H-5b), 2.46
(heptuplet, 1 H, J ) 6.8 Hz, HCMe2), 2.28 (t, 2 H, J ) 7.4 Hz,
COCH2CH2CH3), 2.19 (br s, 1 H, OH), 1.60 (sextuplet, 2 H,
COCH2CH2CH3), 1.06 (m, 12 H, HC(CH3)2), 0.91 (t, 3 H, J )
7.4 Hz, COCH2CH2CH3); 13C NMR (CDCl3) δ 173.48, 168.88,
166.24, 135.70, 119.17, 116.83, 82.25, 65.14, 64.65, 35.93,
33.62, 29.65, 29.50, 24.12, 24.07, 21.18, 18.27, 13.61; FABMS
(m/z, relative intensity) 339 (MH+, 100), 269 (MH-CH3CH2-
CHdCO+, 33). Anal. (C19H30O5) C, H.
(E)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en t -2-en ylid en e]-3-oxo-2-oxola n yl}m et h yl Bu t a n oa t e
(10). Standard monoacylation conditions (method A) starting
from 53 (0.202 g, 0.8 mmol) were applied. After chromatog-
raphy with hexanes:EtOAc (2:1) the monoacylated product 10
(0.153 g, 80%) was isolated together with 0.071 g (20%) of the
diacylated material. Compound 10: mp 71-72 °C; IR (CHCl3)
3596 (OH), 3470 (OH), 3026-2967, 1742 (CdO), 1636 cm-1
;
1H NMR (CDCl3) δ 7.50 (dm, 1 H, J ) 12.2 Hz, >CdCHCHd
C(i-Pr)2), 5.85 (d, 1 H, J ) 12.2 Hz, >CdCHCHdCi-Pr2), 4.25
(AB d, 1 H, J ) 11.9 Hz, CHHOC(O)n-Pr), 4.21 (AB d, 1H, J
) 11.9 Hz, CHHOC(O)n-Pr), 3.70 (AB d, 1H, J ) 12.2 Hz,
CHHOH), 3.64 (AB d, 1 H, J ) 12.2 Hz, CHHOH), 3.14
(heptuplet, 1 H, J ) 6.9 Hz, CHMe2), 2.90 (dd, 1 H, J ) 17.6,
2.8 Hz, H-5a), 2.72 (dd, 1 H, J ) 17.6, 2.6 Hz, H-5b), 2.48
(heptuplet, 1 H, J ) 6.9 Hz, HCMe2), 2.42 (br s, 1 H, OH),
2.28 (t, 2 H, J ) 7.4 Hz, COCH2CH2CH3), 1.59 (sextuplet, 2
H, J ) 7.4 Hz, COCH2CH2CH3) 1.05 (d, 12 H, J ) 6.9 Hz, HC-
(CH3)2), 0.89 (t, 3 H, J ) 7.4 Hz, COCH2CH2CH3); 13C NMR
(CDCl3) δ 173.50, 171.26, 168.93, 132.59, 121.81, 117.31, 83.16,
65.33, 64.77, 35.89, 30.52, 30.12, 29.74, 24.08, 24.04, 21.09,
21.05, 18.26, 13.60; FABMS (m/z, relative intensity) 339 (MH+,
100), 269 (MH-CH3CH2CHdCO+, 33). Anal. (C19H30O5) C, H.
(Z)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en t-2-en ylid en e]-3-oxo-2-oxola n yl}m eth yl Aceta te (7).
Standard monoacylation conditions (method A) starting from
52 (0.201 g, 0.8 mmol) were applied. After chromatography
with different gradients of hexanes:EtOAc (2:1) and (1:1) the
monoacetate product 7 (0.166 g, 70%) was isolated, together
with 0.051 g (18%) of the diacetate. Compound 7: oil; IR
(CHCl3) 3595 (OH), 3469 (OH), 3024-2966, 1744 (CdO), 1629
cm-1 1H NMR (CDCl3) δ 7.27 (d, 1 H, J ) 12.1 Hz, >Cd
;
CHCH)C(i-Pr)2), 7.05 (dt, 1 H, J ) 12.2, 2.1 Hz, >CdCHCHd
C(i-Pr)2), 4.23 (AB d, 1 H, J ) 11.9 Hz, CHHOAc), 4.17 (AB d,
1H, J ) 11.9 Hz, CHHOAc), 3.68 (AB d, 1 H, J ) 12.2 Hz,
CHHOH), 3.62 (AB d, 1 H, J ) 12.2 Hz, CHHOH), 3.08
(heptuplet, 1 H, J ) 6.9 Hz, HCMe2), 2.92 (dd, 1 H, J ) 17.2,
2.2 Hz, H-5a), 2.75 (dd, 1 H, J ) 17.2, 2.0 Hz, H-5b), 2.45
(heptuplet, 1 H, J ) 6.8 Hz, HCMe2), 2.25 (br s, 1H, OH), 2.05
(s, 3H, COCH3), 1.06 (m, 12 H, HC(CH3)2); 13C NMR (CDCl3)
δ 170.88, 168.89, 166.31, 135.82, 119.06, 116.82, 82.17, 65.25,
64.61, 33.56, 29.65, 29.51, 24.10, 24.07, 21.18, 20.66; FABMS
(m/z, relative intensity) 311 (MH+, 100). Anal. (C17H26O5) C,
H.
(Z)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en t-2-en ylid en e]-3-oxo-2-oxola n yl}m eth yl Hexa n oa te
(11). Standard monoacylation conditions (method A) starting
from 52 (0.205 g, 0.8 mmol) were applied. After chromatog-
raphy with different gradients of hexanes:EtOAc (4:1), (2:1),
and (1:1), the monoacylated product 11 (0.185 g, 67%) was
isolated together with 0.069 g (25%) of the diacylated material.
Compound 11: oil; IR (neat) 3446 (OH), 2961-2871, 1744 (Cd
(E)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en t-2-en ylid en e]-3-oxo-2-oxola n yl}m eth yl Aceta te (8).
Standard monoacylation conditions (method A) starting from
53 (0.202 g, 0.75 mmol) were applied. After chromatography
with different gradients of hexanes:EtOAc (2:1), (1:1), and
(1:2), the monoacetate product 8 (0.133 g, 56%) was isolated,
together with 0.079 g (30%) of the diacetate. Compound 8:
solid; mp 61-62 °C; IR (CHCl3) 3446 (OH), 2964-2830, 1748
(CdO), 1636 cm-1; 1H NMR (CDCl3) δ 7.51 (dm, 1 H, J ) 12.2
Hz, >CdCHCHdC(i-Pr)2), 5.86 (d, 1 H, J ) 12.2 Hz, >Cd
CHCH)Ci-Pr2), 4.25 (AB d, 1 H, J ) 11.9 Hz, CHHOAc), 4.19
(AB d, 1H, J ) 11.9 Hz, CHHOAc), 3.71 (AB d, 1H, J ) 12.1
Hz, CHHOH), 3.65 (AB d, 1 H, J ) 12.1 Hz, CHHOH), 3.15
(heptuplet, 1 H, J ) 6.9 Hz, CHMe2), 2.90 (dd, 1 H, J ) 17.6,
2.8 Hz, H-5a), 2.73 (dd, 1 H, J ) 17.6, 2.7 Hz, H-5b), 2.49
(heptuplet, 1 H, J ) 6.8 Hz, HCMe2), 2.30 (br s, 1 H, OH),
2.06 (s, 3 H, COCH3), 1.05 (d, 12 H, J ) 6.9 Hz, HC(CH3)2);
13C NMR (CDCl3) δ ppm: 171.20, 170.87, 169.05, 132.76,
121.64, 117.29, 82.98, 65.43, 64.76, 30.53, 30.13, 29.75, 24.08,
24.06, 21.08, 20.66; FABMS (m/z, relative intensity) 311 (MH+,
100). Anal. (C17H26O5) C, H.
1
O), 1631 cm-1; H NMR (CDCl3) δ 7.28 (d, 1 H, J ) 12.1 Hz,
>CdCHCHdC(i-Pr)2), 7.04 (dm, 1 H, J ) 12.1, >CdCHCHd
C(i-Pr)2), 4.25 (AB d, 1 H, J ) 11.8 Hz, CHHOC(O)(CH2)4CH3),
4.16 (AB d, 1H, J ) 11.8 Hz, CHHOC(O)(CH2)4CH3), 3.68 (AB
d, 1 H, J ) 12.2 Hz, CHHOH), 3.61 (AB d, 1 H, J ) 12.2 Hz,
CHHOH), 3.08 (heptuplet, 1 H, J ) 6.9 Hz, HCMe2), 2.93 (dd,
1 H, J ) 17.2, 2.1 Hz, H-5a), 2.75 (dd, 1 H, J ) 17.2, 1.9 Hz,
H-5b), 2.46 (heptuplet, 1 H, J ) 6.8 Hz, HCMe2), 2.30 (t, 2 H,
J ) 7.4 Hz, COCH2(CH2)3CH3), 2.26 (br s, 1 H, OH), 1.57
(quintet, 2 H, J ) 7.5 Hz, COCH2CH2(CH2)2CH3), 1.25 (m, 4
H, OCO(CH2)2(CH2)2CH3), 1.06 (m, 12 H, HC(CH3)2), 0.85 (t,
3 H, J ) 6.8 Hz, OCO(CH2)4CH3); 13C NMR (CDCl3) δ 173.67,
169.01, 166.12, 135.65, 119.28, 116.83, 82.37, 65.19, 64.63,
33.53, 34.00, 31.19, 29.64, 29.48, 24.45, 24.07, 21.17, 22.21,
13.83; FABMS (m/z, relative intensity) 367 (MH+, 100), 269
(MH-CH3(CH2)3CHdCO+, 30). Anal. (C21H34O5) C, H.
(E)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en t-2-en ylid en e]-3-oxo-2-oxola n yl}m eth yl Hexa n oa te
(12). Standard monoacylation conditions (method A) starting
from 53 (0.208 g, 0.8 mmol) were applied. After chromatog-

938 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5
Nacro et al.
raphy with different gradients of hexanes:EtOAc (3:1), (2:1),
and (1:1), the monoacylated product 12 (0.178 g, 62%) was
isolated together with 0.066 g (18%) of the diacylated material.
Compound 12: oil, IR (neat): 3447 (O-H), 2961-2871, 1744
(CdO), 1637 cm-1; 1H NMR (CDCl3) δ 7.51 (dm, 1 H, J ) 12.2
Hz, >CdCHCHdC(i-Pr)2), 5.86 (d, 1 H, J ) 12.2 Hz, >Cd
CHCHdCi-Pr2), 4.26 (AB d, 1 H, J ) 11.9 Hz, CHHOC(O)-
(CH2)4CH3), 4.19 (AB d, 1H, J ) 11.9 Hz, CHHOC(O)(CH2)4-
CH3), 3.70 (AB d, 1H, J ) 12.2 Hz, CHHOH), 3.67 (AB d, 1 H,
J ) 12.2 Hz, CHHOH), 3.15 (heptuplet, 1 H, J ) 6.9 Hz,
CHMe2), 2.90 (dd, 1 H, J ) 17.6, 2.7 Hz, H-5a), 2.72 (dd, 1 H,
J ) 17.6, 2.7 Hz, H-5b), 2.49 (heptuplet, 1 H, J ) 6.9 Hz,
HCMe2), 2.30 (t, 2 H, J ) 7.4 Hz, COCH2(CH2)3CH3), 2.12 (br
s, 1 H, OH), 1.55 (quintuplet, 2 H, J ) 7.4 Hz, COCH2CH2-
(CH2)2CH3), 1.26 (m, 4 H, CO(CH2)2CH2CH2CH3), 1.06 (d, 12
H, J ) 6.9 Hz, HC(CH3)2), 0.85 (t, 3 H, J ) 7.4 Hz, CO-
(CH2)4CH3); 13C NMR (CDCl3) δ 173.70, 171.17, 168.98, 132.66,
121.72, 117.29, 83.08, 65.29, 64.80, 33.99, 31.18, 30.53, 30.14,
29.74, 24.45, 24.08, 24.06, 21.07, 21.05, 22.21, 13.83; FABMS
(m/z, relative intensity) 367 (MH+, 100), 269 (MH-CH3(CH2)3-
CHdCO+, 71). Anal. (C21H34O5) C, H.
(t, 2 H, J ) 7.5 Hz, COCH2CH2CH3), 2.20 (m, 2 H, >Cd
CHCH2CH(i-Pr)2), 1.84 (heptuplet, 2 H, CHMe2), 1.72 (sextu-
plet, 2 H, COCH2CH2CH3), 1.19 (m, 1 H, CH(i-Pr)2), 1.02 (t, 3
H, J ) 7.6 Hz, COCH2CH2CH3), 0.94 and 0.98 (2d, 12 H, J )
6.84 Hz, 2 × HC(CH3)2); 13C NMR (CDCl3) δ 173.26, 168.76,
147.45, 122.67, 82.31, 65.15, 64.50, 51.06, 35.81, 33.05, 29.26,
29.24, 26.23, 21.57, 21.53, 19.37, 19.34, 18.23, 13.55; FABMS
(m/z, relative intensity) 341 (MH+, 100), 271 (MH+-C3H6CO,
44). Anal. (C19H32O5) C, H.
(E)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en tylid en e]-3-oxo-2-oxola n yl}m eth yl Bu ta n oa te (16).
Standard monoacylation conditions (method A) starting from
55 (0.230 g, 0.9 mmol) were applied. After chromatography
with different gradients of hexanes:EtOAc (4:1), (3:1), (2:1),
and (1:1), the monoacylated product 16 (0.161 g, 55%) was
isolated together with 0.076 g (22%) of the diacylated material.
Compound 16: oil; IR (neat) 3456 (OH), 2960-2875, 1744 (Cd
1
O), 1676, 1176 cm-1; H NMR (CDCl3) δ 6.88 (m, 1 H, >Cd
CHCH2CH(i-Pr)2), 4.36 (AB d, 1H, J ) 11.9 Hz, CHHOCOn-
Pr), 4.24 (AB d, 1 H, J ) 11.9 Hz, CHHOCOn-Pr), 3.80 (AB d,
1 H, J ) 12.2, CHHOH), 3.72 (AB d, 1 H, J ) 12.2 Hz,
CHHOH), 2.91 (dd, 1 H, J ) 19.2, 2.2 Hz, H-5a), 2.74 (dd, 1 H,
J ) 19.2, 2.2 Hz, H-5b), 2.38 (t, 2 H, J ) 7.5 Hz, COCH2CH2-
CH3), 2.20 (m, 2 H, >CdCHCH2CH(i-Pr)2), 1.86 (heptuplet, 2
H, CHMe2), 1.70 (sextuplet, 2 H, COCH2CH2CH3), 1.28 (m, 1
H, CH(i-Pr)2), 1.01 (t, 3 H, J ) 7.5 Hz, COCH2CH2CH3), 0.98
and 0.92 (2d, 12 H, J ) 6.84 Hz, 2 × HC(CH3)2); 13C NMR
(CDCl3) δ 172.68, 169.52, 143.11, 124.38, 82.63, 64.83, 64.12,
49.71, 35.29, 29.32, 28.66, 28.60, 28.12, 21.07, 20.98, 18.73,
18.72, 17.72, 13.05; FABMS (m/z, relative intensity) 341 (MH+,
74), 271 (MH+-C3H6CO, 67), 71 (C3H7CO+, 100). Anal.
(C19H32O5) C, H.
(Z)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en tylid en e]-3-oxo-2-oxola n yl}m eth yl Aceta te (13). Stan-
dard monoacylation conditions (method B) starting from 54
(0.102 g, 0.4 mmol) were applied. After chromatography with
different gradients of hexanes:EtOAc (4:1), (3:1), (2:1), and
(1:1), the monoacetate product 13 (0.078 g, 66%) was isolated
together with 0.034 g (25%) of the diacetate. Compound 13:
oil; IR (neat) 3451 (OH), 2958-2873, 1749 (CdO), 1667 cm-1
;
1H NMR (CDCl3) δ 6.34 (m, 1 H, >CdCHCH2-CH(i-Pr)2), 4.32
(AB d, 1 H, J ) 11.9 Hz, CHHOAc), 4.24 (AB d, 1 H, J ) 11.9
Hz, CHHOAc), 3.77 (AB d, 1 H, J ) 12.0 Hz, CHHOH), 3.71
(AB d, 1 H, J ) 12.0 Hz, CHHOH), 2.95 (m, 1 H, H-5a), 2.80
(m, 3 H, H-5b, >CdCHCH2-CH(i-Pr)2), 2.30 (br s, 1 H, OH),
2.16 (s, 3 H, COCH3), 1.85 (sextuplet, 2 H, CHMe2), 1.19 (m,
1 H, CH(i-Pr)2), 0.94 (d, 6 H, J ) 6.8 Hz, HC(CH3)2), 0.98 (d,
6H, J ) 6.8 Hz, HC(CH3)2); 13C NMR (CDCl3) δ 170.64, 168.68,
147.57, 122.56, 82.18, 65.34, 64.48, 51.08, 33.08, 29.29, 29.24,
26.26, 21.57, 21.55, 20,59, 19.39, 19.33; FABMS (m/z, relative
intensity) 313 (MH+, 100), 271 (MH+-CH2CO, 36). Anal.
(C17H28O5) C, H.
(Z)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en tylid en e]-3-oxo-2-oxola n yl}m eth yl Hexa n oa te (17).
Standard monoacylation conditions (method A) starting from
54 (0.232 g, 0.9 mmol) were applied. After chromatography
with different gradients of hexanes:EtOAc (4:1), (3:1), (2:1),
and (1:1), the monoacylated product 17 (0.183 g, 58%) was
isolated together with 0.098 g (24%) of the diacylated material.
Compound 17: oil; IR (neat) 3447 (OH), 2958-2872, 1745 (Cd
O), 1670 cm-1; 1H NMR (CDCl3) δ 6.37 (m, 1 H, >CdCHCH2-
CH(i-Pr)2), 4.34 (AB d, 1H, J ) 11.9 Hz, CHHOCO(CH2)4CH3),
4.23 (AB d, 1 H, J ) 11.9 Hz, CHHOCO(CH2)4CH3), 3.71 (m,
2 H, CH2OH), 2.99 (m, 1 H, H-5a), 2.82 (m, 3 H, H-5b,
>CdCHCH2CH(i-Pr)2), 2.41 (t, 2 H, J ) 7.5 Hz, COCH2(CH2)3-
CH3), 1.84 (m, 2 H, CHMe2), 1.69 (m, 2 H, COCH2CH2(CH2)2-
CH3), 1.37 (m, 4 H, CO(CH2)2(CH2)2CH3), 1.19 (m, 1 H, CH(i-
Pr)2), 0.96 (m, 15 H, CO(CH2)4CH3, 2 × HC(CH3)2); 13C NMR
(CDCl3) δ 173.49, 168.52, 147.65, 122.49, 82.12, 65.07, 64.57,
51.10, 33.92, 33.13, 31.16, 29.28, 26.28, 24.44, 22.20, 21.57,
19.37, 13.82; FABMS (m/z, relative intensity) 369 (MH+, 100),
271 (MH+-C5H10CO, 54). Anal. (C21H36O5) C, H.
(E)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en tylid en e]-3-oxo-2-oxola n yl}m eth yl Hexa n oa te (18).
Standard monoacylation conditions (method A) starting from
55 (0.241 g, 0.9 mmol) were applied. After chromatography
with different gradients of hexanes:EtOAc (4:1), (3:1), (2:1),
and (1:1), the monoacylated product 18 (0.187 g, 57%) was
isolated together with 0.092 g (22%) of the diacylated material.
Compound 18: oil; IR (neat) 3447 (OH), 2957-2873, 1744 (Cd
O), 1676 cm-1; 1H NMR (CDCl3) δ 6.88 (m, 1 H, >CdCHCH2-
CH(i-Pr)2), 4.36 (AB d, 1H, J ) 11.9 Hz, CHHOCO(CH2)4CH3),
4.24 (AB d, 1 H, J ) 11.9 Hz, CHHOCO(CH2)4CH3), 3.79 (AB
d, 1 H, J ) 12.1, CHHOH), 3.72 (AB d, 1 H, J ) 12.1 Hz,
CHHOH), 2.90 (dd, 1 H, J ) 17.1, 2.3 Hz, H-5a), 2.72 (dd, 1 H,
J ) 19.2, 2.2 Hz, H-5b), 2.70 (br s, 1 H, OH), 2.38 (t, 2 H, J )
7.6 Hz, COCH2(CH2)3CH3), 2.20 (m, 2 H, >CdCHCH2CH(i-
Pr)2), 1.86 (m, 2 H, CHMe2), 1.67 (m, 2 H, COCH2CH2(CH2)2-
CH3), 1.34 (m, 5 H, CO(CH2)2(CH2)2CH3, CH(i-Pr)2), 0.95 (m,
15 H, CO(CH2)4CH3, 2 × HC(CH3)2); 13C NMR (CDCl3) δ
172.88, 169.48, 143.11, 124.36, 82.61, 64.81, 64.13, 49.72,
33.39, 30.63, 28.66, 28.62, 29.33, 28.13, 23.89, 21.67, 21.06,
21.00, 18.79, 18.73, 13.29; FABMS (m/z, relative intensity) 369
(MH+, 83), 271 (MH+-C5H10CO, 66). Anal. (C21H36O5) C, H.
(E)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en tylid en e]-3-oxo-2-oxola n yl}m eth yl Aceta te (14). Stan-
dard monoacylation conditions (method A) starting from 55
(0.232 g, 0.9 mmol) were applied. After chromatography with
different gradients of hexanes:EtOAc (4:1), (3:1), (2:1), and
(1:1), the monoacetate product 14 (0.166 g, 61%) was isolated
together with 0.060 g (19%) of the diacetate. Compound 14:
oil; IR (neat) 3452 (OH), 2958-2874, 1749 (CdO), 1676 cm-1
;
1H NMR (CDCl3) δ 6.84 (tt, 1 H, J ) 7.3, 2.7 Hz, >CdCHCH2-
CH(i-Pr)2), 4.35 (AB d, 1 H, J ) 11.8 Hz, CHHOAc), 4.25 (AB
d, 1 H, J ) 11.8 Hz, CHHOAc), 3.80 (AB d, 1 H, J ) 11.8 Hz,
CHHOH), 3.72 (AB d, 1 H, J ) 11.8 Hz, CHHOH), 2.91 (dm,
1 H, J ) 17.2 Hz, H-5a), 2.75 (d, 1 H, J ) 17.2 Hz, H-5b), 2.66
(m, 1 H, OH), 2.20 (m, 2 H, >CdCHCH2-CH(i-Pr)2), 2.14 (s,
3 H, COCH3), 1.86 (sextuplet, 2 H, CHMe2), 1.29 (m, 1 H, CH(i-
Pr)2), 0.98 (d, 6 H, J ) 6.6 Hz, HC(CH3)2), 0.92 (d, 6H, J ) 6.8
Hz, HC(CH3)2); 13C NMR (CDCl3) δ 170.09, 169.53, 143.21,
124.29, 82.55, 65.04, 64.07, 49.73, 29.33, 28.68, 28.60, 28.14,
21.08, 20.99, 18.81, 18.69, 20.06; FABMS (m/z, relative inten-
sity) 313 (MH+, 100). Anal. (C17H28O5) C, H.
(Z)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en tylid en e]-3-oxo-2-oxola n yl}m eth yl Bu ta n oa te (15).
Standard monoacylation conditions (method A) starting from
54 (0.100 g, 0.4 mmol) were applied. After chromatography
with different gradients of hexanes:EtOAc (4:1), (3:1), (2:1),
and (1:1), the monoacylated product 15 (0.077 g, 61%) was
isolated together with 0.031 g (20%) of the diacylated material.
Compound 15: oil; IR (neat) 3446 (OH), 2959-2875, 1744 (Cd
O), 1670 cm-1; 1H NMR (CDCl3) δ 6.34 (m, 1 H, >CdCHCH2-
CH(i-Pr)2), 4.35 (AB d, 1H, J ) 11.8 Hz, CHHOCOn-Pr), 4.23
(AB d, 1 H, J ) 11.8 Hz, CHHOCOn-Pr), 3.76 (AB d, 1 H, J )
12.1, CHHOH), 3.70 (AB d, 1 H, J ) 12.1 Hz, CHHOH), 2.95
(m, 1 H, H-5a), 2.76 (m, 1 H, H-5b, >CdCHCH2CH(i-Pr)2), 2.39

Conformationally Constrained Analogues of DAG
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5 939
(Z)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en tylid en e]-3-oxo-2-oxola n yl}m eth yl Octa n oa te (19).
Standard monoacylation conditions (method B) starting from
54 (0.115 g, 0.4 mmol) were applied. After chromatography
with different gradients of hexanes:EtOAc (9:1), (4:1), and
(3:1), the monoacylated product 19 (0.102 g, 60%) was isolated
together with 0.026 g (12%) of the diacylated material.
Compound 19: oil, IR (neat) 3446 (OH), 2957-2855, 1744 (Cd
O), 1676 cm-1; 1H NMR (CDCl3) δ 6.34 (m, 1 H, >CdCHCH2-
CH(i-Pr)2), 4.34 (AB d, 1H, J ) 11.9 Hz, CHHOCO(CH2)6CH3),
4.23 (AB d, 1 H, J ) 11.9 Hz, CHHOCO(CH2)6CH3), 3.76 (AB
d, 1 H, J ) 12.0, CHHOH), 3.70 (AB d, 1 H, J ) 12.0 Hz,
CHHOH), 2.99 (m, 1 H, H-5a), 2.77 (m, 3 H, H-5b, >CdCHCH2-
CH(i-Pr)2), 2.41 (t, 2 H, J ) 7.5 Hz, COCH2(CH2)5CH3), 1.85
(septuplet, 2 H, CHMe2), 1.69 (m, 2 H, COCH2CH2(CH2)4CH3),
1.35 (m, 8 H, CO(CH2)2(CH2)4CH3), 1.19 (m, 1 H, CH(i-Pr)2),
0.94 and 0.99 (m, 15 H, CO(CH2)6CH3, 2 × HC(CH3)2); 13C
NMR (CDCl3) δ 173.50, 168.55, 147.63, 122.51, 82.14, 65.08,
64.54, 51.10, 33.96, 33.13, 31.56, 29.28, 28.90, 28.81, 26.27,
24.75, 22.52, 21.57, 19.37, 13.99; FABMS (m/z, relative inten-
sity) 397 (MH+, 7), 57 (C4H9+, 100). Anal. (C23H40O5) C, H.
(E)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en tylid en e]-3-oxo-2-oxola n yl}m eth yl Octa n oa te (20).
Standard monoacylation conditions (method A) starting from
55 (0.205 g, 0.8 mmol) were applied. After chromatography
with different gradients of hexanes:EtOAc (9:1), (4:1), and
(3:1), the monoacylated product 20 (0.183 g, 60%) was isolated
together with 0.090 g (22%) of the diacylated material.
Compound 20: oil; IR (neat) 3446 (OH), 2957-2872, 1744 (Cd
O), 1676 cm-1; 1H NMR (CDCl3) δ 6.90 (m, 1 H, >CdCHCH2-
CH(i-Pr)2), 4.36 (AB d, 1H, J ) 11.8 Hz, CHHOCO(CH2)6CH3),
4.24 (AB d, 1 H, J ) 11.8 Hz, CHHOCO(CH2)6CH3), 3.80 (AB
d, 1 H, J ) 12.2, CHHOH), 3.72 (AB d, 1 H, J ) 12.2 Hz,
CHHOH), 2.90 (dd, 1 H, J ) 17.3, 1.9 Hz, H-5a), 2.75 (dd, 1 H,
J ) 17.3, 1.9 Hz, H-5b), 2.80 (br s, 1 H, OH), 2.38 (t, 2 H, J )
7.6 Hz, COCH2(CH2)5CH3), 2.20 (br t, J ≈ 6.5 Hz, 2 H, >Cd
CHCH2CH(i-Pr)2), 1.86 (sextuplet, 2 H, CHMe2), 1.67 (m, 2 H,
COCH2CH2(CH2)4CH3), 1.34 (m, 9 H, CO(CH2)2(CH2)4CH3,
CH(i-Pr)2), 0.92-0.98 (m, 15 H, CO(CH2)6CH3, 2 × HC(CH3)2);
13C NMR (CDCl3) δ 173.42, 170.05, 143.70, 123.83, 83.16,
65.29, 64.64, 50.22, 33.92, 31.53, 29.82, 29.17, 29.12, 28.96,
28.79, 28.65, 24.71, 22.50, 21.55, 21.51, 19.30, 19.25, 13.98;
FABMS (m/z, relative intensity) 397 (MH+, 26), 271 (MH+-
C7H14CO). Anal. (C23H40O5) C, H.
O), 1676 cm-1; 1H NMR (CDCl3) δ 6.90 (m, 1 H, >CdCHCH2-
CH(i-Pr)2), 4.37 (AB d, 1H, J ) 11.8 Hz, CHHOCO(CH2)8CH3),
4.24 (AB d, 1 H, J ) 11.8 Hz, CHHOCO(CH2)8CH3), 3.80 (AB
d, 1 H, J ) 12.1, CHHOH), 3.72 (AB d, 1 H, J ) 12.1 Hz,
CHHOH), 2.90 (dd, 1 H, J ) 17.3, 1.9 Hz, H-5a), 2.74 (dd, 1 H,
J ) 17.3, 1.9 Hz, H-5b), 2.80 (br s, 1 H, OH), 2.39 (t, 2 H, J )
7.5 Hz, COCH2(CH2)7CH3), 2.20 (br t, 2 H, >CdCHCH2CH(i-
Pr)2), 1.85 (sextuplet, 2 H, CHMe2), 1.67 (m, 2 H, COCH2CH2-
(CH2)6CH3), 1.29-1.34 (m, 13 H, CO(CH2)2(CH2)6CH3, CH(i-
Pr)2), 0.93-0.99 (m, 15 H, CO(CH2)8CH3, 2 × HC(CH3)2); 13C
NMR (CDCl3) δ 173.43, 169.75, 143.82, 124.65, 82.88, 65.22,
64.70, 50.24, 33.95, 31.77, 29.91, 29.32, 29.16, 29.04, 28.67,
24.74, 22.59, 29.17, 21.57, 21.53, 19.30, 19.25, 14.04; FABMS
(m/z, relative intensity) 425 (MH+, 45), 271 (MH+-C9H18CO,
50). Anal. (C25H44O5) C, H.
(Z)-[4-Eth ylid en e-1-(h yd r oxym eth yl)-3-oxo-2-oxola n yl]-
m eth yl 4-Meth yl-3-(m eth yleth yl)p en ta n oa te (23). Stan-
dard monoacylation conditions (method B) starting from a
mixture of 56 and 57 (0.203 g, 1.2 mmol) were applied. After
chromatography with different gradients of hexanes:EtOAc
(9:1), (3:1), (2:1), (1:1), and (1:5), four fractions were isolated:
(1) a mixture of diacylated E- and Z-isomers (0.087 g, 16%),
(2) 0.102 g (28%) of the desired monoacylated Z-isomer 23, (3)
a mixture of E- and Z-monoacylated isomers (0.021 g, 6%), and
(4) 0.092 g (25%) of the monoacylated E-isomer 24. Compound
23: oil; IR (neat) 3447 (OH), 2961-2876, 1743 (CdO), 1673
cm-1; 1H NMR (CDCl3) δ 6.40 (qt, 1H, J ) 7.3, ca. 2.2 Hz, >Cd
CHCH3), 4.32 (AB d, 1 H, J ) 11.8 Hz, CHHOCOCH2CH(i-
Pr)2), 4.23 (AB d, 1 H, J ) 11.8 Hz, CHHOCOCH2CH(i-Pr)2),
3.78 (AB d, 1 H, J ) 12.2 Hz, CHHOH), 3.70 (AB d, 1 H, J )
12.2 Hz, CHHOH), 2.97 (dt, 1 H, J ) 16.5, ca. 2.2 Hz, H-5a),
2.79 (dt, 1 H, J ) 16.5, ca. 2.2 Hz, H-5b), 2.61 (br s, 1 H, OH),
2.27 (m, 5 H, >CdCHCH3, COCH2CH(i-Pr)2), 1.80 (sextuplet,
2 H, CHMe2), 1.64 (quintuplet, 1 H, HC(i-Pr)2), 0.97 and 0.88
(2 d, 12 H, J ) 6.8 Hz, 2 × HC(CH3)2); 13C NMR (CDCl3) δ
174.53, 168.66, 139.73, 124.67, 82.32, 65.24, 64.58, 46.80,
32.95, 32.75, 29.31, 21.27, 18.66, 14.07; FABMS (m/z, relative
intensity) 313 (MH+, 74). Anal. (C17H28O5) C, H.
(E)-[4-Eth ylid en e-1-(h yd r oxym eth yl)-3-oxo-2-oxola n yl]-
m eth yl 4-Meth yl-3-(m eth yleth yl)p en ta n oa te (24). Stan-
dard monoacylation conditions (method B) starting from 57
(0.202 g, 1.2 mmol) were applied. After chromatography with
different gradients of hexanes:EtOAc (9:1), (3:1), (1:1), and
(1:5), the monoacylated product 24 (0.217 g, 58%) was isolated
together with 0.070 g (13%) of the diacylated material.
Compound 24: oil; IR (neat) 3450 (OH), 2960, 1743 (CdO),
1683 cm-1; 1H NMR (CDCl3) δ 6.89 (m, 1H, >CdCHCH3), 4.33
(AB d, 1 H, J ) 11.8 Hz, CHHOCOCH2CH(i-Pr)2), 4.25 (AB d,
1 H, J ) 11.8 Hz, CHHOCOCH2CH(i-Pr)2), 3.81 (AB d, 1 H, J
) 12.2 Hz, CHHOH), 3.73 (AB d, 1 H, J ) 12.2 Hz, CHHOH),
2.91 (dm, 1 H, J ) 17.2 Hz, H-5a), 2.75 (dm, 1 H, J ) 17.2 Hz,
H-5b), 2.53 (br s, 1 H, OH), 2.27 (d, 2 H, J ) 5.6 Hz, COCH2-
CH(i-Pr)2), 1.93 (dt, 3 H, J ) 7.3, 2.0 Hz, >CdCHCH3), 1.80
(sextuplet, 2 H, CHMe2), 1.64 (quintuplet, 1 H, HC(i-Pr)2), 0.96
and 0.87 (2 d, 12 H, J ) 6.8 Hz, 2 × HC(CH3)2); 13C NMR
(CDCl3) δ 174.43, 169.90, 136.22, 127.23, 83.33, 65.51, 64.58,
46.79, 32.69, 29.45, 29.26, 29.22, 21.22, 18.59, 15.65; FABMS
(m/z, relative intensity) 313 (MH+, 100). Anal. (C17H28O5) C,
H.
(Z)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en tylid en e]-3-oxo-2-oxola n yl}m eth yl Deca n oa te (21).
Standard monoacylation conditions (method B) starting from
54 (0.115 g, 0.4 mmol) were applied. After chromatography
with different gradients of hexanes:EtOAc (9:1; 4:1; 3:1) the
monoacylated product 21 (0.101 g, 56%) was isolated together
with 0.029 g (12%) of the diacylated material. Compound 21:
oil; IR (neat) 3440 (OH), 2956-2855, 1745 (CdO), 1666 cm-1
;
1H NMR (CDCl3) δ 6.34 (m, 1 H, >CdCHCH2CH(i-Pr)2), 4.34
(AB d, 1H, J ) 12.0 Hz, CHHOCO(CH2)8CH3), 4.23 (AB d, 1
H, J ) 12.0 Hz, CHHOCO(CH2)8CH3), 3.76 (AB d, 1 H, J )
12.0, CHHOH), 3.70 (AB d, 1 H, J ) 12.0 Hz, CHHOH), 3.00
(m, 1 H, H-5a), 2.80 (m, 3 H, H-5b, >CdCHCH2CH(i-Pr)2), 2.41
(t, 2 H, J ) 7.6 Hz, COCH2(CH2)7CH3), 2.33 (br s, 1 H, OH),
1.90 (sextuplet, 2 H, CHMe2), 1.69 (m, 2 H, COCH2CH2(CH2)6-
CH3), 1.34 (m, 12 H, CO(CH2)2(CH2)6CH3), 1.19 (sextuplet, 2
H, CH(i-Pr)2), 0.94-0.99 (m, 15 H, CO(CH2)8CH3, 2 ×
HC(CH3)2); 13C NMR (CDCl3) δ 173.49, 168.61, 147.55, 122.57,
82.20, 65.10, 64.52, 51.09, 33.96, 33.11, 31.76, 29.33, 29.17,
29.04, 26.27, 24.75, 22.59, 29.27, 21.27, 19.37, 14.03; FABMS
(m/z, relative intensity) 425 (MH+, 100), 271 (MH+-C9H18CO,
69). Anal. (C25H44O5) C, H.
(Z)-[4-Bu tylid en e-1-(h yd r oxym eth yl)-3-oxo-2-oxola n yl]-
m eth yl 4-Meth yl-3-(m eth yleth yl)p en ta n oa te (25). Stan-
dard monoacylation conditions (method B) starting from 58
(0.10 g, 0.5 mmol) were applied. After chromatography with
different gradients of hexanes:EtOAc (9:1), (3:1), (1:1), and
(1:5), the monoacylated product 25 (0.109 g, 61%) was isolated
together with 0.029 g (12%) of the diacylated material.
Compound 25: oil; IR (neat) 3458 (OH), 2960-2874, 1742 (Cd
O), 1670 cm-1; 1H NMR (CDCl3) δ 6.32 (m, 1H, >CdCH(CH2)2-
CH3), 4.33 (AB d, 1 H, J ) 12.0 Hz, CHHOCOCH2CH(i-Pr)2),
4.23 (AB d, 1 H, J ) 12.0 Hz, CHHOCOCH2CH(i-Pr)2), 3.78
(AB d, 1 H, J ) 12.2 Hz, CHHOH), 3.71 (AB d, 1 H, J ) 12.2
Hz, CHHOH), 2.96 (m, 1 H, H-5a), 2.70-2.90 (m, 3 H, H-5b,
>CdCHCH2CH2CH3 ), 2.28 (d, 2 H, COCH2CH(i-Pr)2), 1.81
(sextuplet, 2 H, CHMe2), 1.66 (m, 1 H, HC(i-Pr)2), 1.55
(E)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en tylid en e]-3-oxo-2-oxola n yl}m eth yl Deca n oa te (22).
Standard monoacylation conditions (method A) starting from
55 (0.220 g, 0.9 mmol) were applied. After chromatography
with different gradients of hexanes:EtOAc (9:1), (4:1), and
(3:1), the monoacylated product 22 (0.228 g, 63%) was isolated
together with 0.097 g (21%) of the diacylated material.
Compound 22: oil; IR (neat) 3448 (OH), 2956-2927, 1744 (Cd

940 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5
Nacro et al.
(sextuplet, 2 H, J ≈ 7.3 Hz, >CdCHCH2CH2CH3), 1.03 (t, 3
H, J ) 7.3 Hz, >CdCH(CH2)2CH3), 0.97 and 0.88 (2 d, 12 H,
J ) 6.8 Hz, 2 × HC(CH3)2); 13C NMR (CDCl3) δ 174.51, 168.65,
145.02, 123.99, 82.38, 65.31, 64.57, 46.78, 32.92, 32.73, 29.52
(CH2), 29.29, 22.22, 21.25, 18.63, 13.66; FABMS (m/z, relative
intensity) 341 (MH+, 89). Anal. (C19H32O5) C, H.
dard monoacylation conditions (method B) starting from 62
(0.26 g, 0.9 mmol) were employed. After chromatography with
different gradients of hexanes:EtOAc (9:1), 8:2), (3:1), and
(1:1), the monoacylated product 29 (0.199 g, 51%) was isolated
together with 0.031 g (6.5%) of the diacylated material.
Compound 29: oil; IR (neat) 3450 (OH), 2925-2857, 1744 (Cd
1
O), 1671 cm-1; H NMR (CDCl3) δ 6.32 (br t, 1H, J ) 7.6 Hz,
(E)-[4-Bu tylid en e-1-(h yd r oxym eth yl)-3-oxo-2-oxola n yl]-
m eth yl 4-Meth yl-3-(m eth yleth yl)p en ta n oa te (26). Stan-
dard monoacylation conditions (method B) starting from 59
(0.20 g, 1.0 mmol) were applied. After chromatography with
different gradients of hexanes:EtOAc (9:1), (3:1), (1:1), and
(1:5), the monoacylated product 26 (0.220 g, 64%) was isolated
together with 0.061 g (13%) of the diacylated material.
Compound 26: oil; IR (neat) 3450 (OH), 2960-2874, 1742 (Cd
O), 1680 cm-1; 1H NMR (CDCl3) δ 6.88 (m, 1H, >CdCH(CH2)2-
CH3), 4.34 (AB d, 1 H, J ) 12.0 Hz, CHHOCOCH2CH(i-Pr)2),
4.23 (AB d, 1 H, J ) 12.0 Hz, CHHOCOCH2CH(i-Pr)2), 3.81
(AB d, 1 H, J ) 12.2 Hz, CHHOH), 3.73 (AB d, 1 H, J ) 12.2
Hz, CHHOH), 2.91 (dm, 1 H, J ) 17.1 Hz, H-5a), 2.74 (dm, 1
H, J ) 17.1 Hz, H-5b), 2.55 (br s, 1 H, OH), 2.27 (d, 2 H, J )
5.6 Hz, COCH2CH(i-Pr)2), 2.21 (m, 2 H, >CdCHCH2CH2CH3),
1.80 (sextuplet, 2 H, CHMe2), 1.60 (m, 3 H, HC(i-Pr)2, >Cd
CHCH2CH2CH3), 1.03 (t, 3 H, J ) 7.3 Hz, >CdCH(CH2)2CH3),
0.97 and 0.88 (2 d, 12 H, J ) 6.7 Hz, 2 × HC(CH3)2); 13C NMR
(CDCl3) δ 174.51, 169.75, 141.57, 125.97, 82.99, 65.39, 64.72,
46.81, 32.71, 32.16, 29.74, 29.29, 21.34, 21.27, 18.64, 13.76;
FABMS (m/z, relative intensity) 341 (MH+, 92). Anal. (C19H32O5)
C, H.
>CdCH(CH2)8CH3), 4.33 (AB d, 1 H, J ) 12.0 Hz, CHHO-
COCH2CH(i-Pr)2), 4.23 (AB d, 1 H, J ) 12.0 Hz, CHHOCOCH2-
CH(i-Pr)2), 3.78 (AB d, 1 H, J ) 12.1 Hz, CHHOH), 3.71 (AB
d, 1 H, J ) 12.1 Hz, CHHOH), 3.01 (dm, 1 H, J ) 15.0 Hz,
H-5a), 2.85 (m, 3 H, H-5b, >CdCHCH2(CH2)7CH3), 2.28 (d, 2
H, J ) 5.6 Hz, COCH2CH(i-Pr)2), 1.82 (sextuplet, 2 H, 2 ×
CHMe2), 1.66 (m, 1 H, HC(i-Pr)2), 1.34-1.51 (m, 14 H, >Cd
CHCH2(CH2)7CH3), 0.88-0.98 (m, 15 H, >CdCH(CH2)8CH3,
2 × HC(CH3)2); 13C NMR (CDCl3) δ 174.51, 168.58, 145.36,
123.69, 82.33, 65.30, 64.55, 46.79, 32.94, 32.73, 31.79, 29.42,
29.31, 29.21, 29.10, 29.03, 27.68, 22.59, 21.27, 18.65, 14.03;
FABMS (m/z, relative intensity) 425 (MH+, 38). Anal. (C25H44O5)
C, H.
(E)-[4-Decyliden e-1-(h ydr oxym eth yl)-3-oxo-2-oxolan yl]-
m eth yl 4-Meth yl-3-(m eth yleth yl)p en ta n oa te (30). Stan-
dard monoacylation conditions (method B) starting from 63
(0.26 g, 0.9 mmol) were employed. After chromatography with
different gradients of hexanes:EtOAc (9:1), (8:2), (3:1), and
(1:1), the monoacylated product 30 (0.193 g, 50%) was isolated
together with 0.035 g (6.5%) of the diacylated material.
Compound 30: oil; IR (neat) 3449 (OH), 2956-2855, 1742 (Cd
O), 1679 cm-1; 1H NMR (CDCl3) δ 6.85 (m, 1H, >CdCH(CH2)8-
CH3), 4.35 (AB d, 1 H, J ) 11.9 Hz, CHHOCOCH2CH(i-Pr)2),
4.23 (AB d, 1 H, J ) 11.9 Hz, CHHOCOCH2CH(i-Pr)2), 3.80
(AB d, 1 H, J ) 12.2 Hz, CHHOH), 3.73 (AB d, 1 H, J ) 12.2
Hz, CHHOH), 2.90 (AB d, 1 H, J ) 17.0 Hz, H-5a), 2.73 (AB d,
1 H, J ) 17.0 Hz, H-5b), 2.28 (d, 2 H, J ) 5.9 Hz, COCH2CH-
(i-Pr)2), 2.23 (m, 2 H, >CdCHCH2(CH2)7CH3), 1.82 (sextuplet,
2 H, 2 × CHMe2), 1.65 (m, 1 H, HC(i-Pr)2), 1.34-1.55 (m, 14
H, >CdCHCH2(CH2)7CH3), 0.98 (m, 3 H, >CdCH(CH2)8CH3),
0.97 and 0.88 (2 d, 12 H, J ) 6.6 Hz, 2 × HC(CH3)2); 13C NMR
(CDCl3) δ 174.47, 169.98, 141.71, 125.88, 83.17, 65.47, 64.65,
46.82, 32.70, 31.76, 30.21, 29.64, 29.31, 29.27, 29.19, 28.00,
22.57, 21.24, 18.64, 14.01; FABMS (m/z, relative intensity) 425
(MH+, 83). Anal. (C25H44O5), C, H.
(Z)-[4-Hexyliden e-1-(h ydr oxym eth yl)-3-oxo-2-oxolan yl]-
m eth yl 4-Meth yl-3-(m eth yleth yl)p en ta n oa te (27). Stan-
dard monoacylation conditions (method B) starting from 60
(0.120 g, 0.9 mmol) were employed. After chromatography with
different gradients of hexanes:EtOAc (3:1) and (2:1), the
monoacylated product 27 (0.134 g, 41%) was isolated together
with 0.064 g (15%) of the diacylated material. Compound 27:
oil; IR (neat) 3447 (OH), 2959-2873, 1742 (CdO), 1670 cm-1
;
1H NMR (CDCl3) δ 6.33 (tt, 1H, J ) 7.6, 2.0 Hz, >CdCH(CH2)4-
CH3), 4.33 (AB d, 1 H, J ) 11.8 Hz, CHHOCOCH2CH(i-Pr)2),
4.23 (AB d, 1 H, J ) 11.8 Hz, CHHOCOCH2CH(i-Pr)2), 3.78
(AB d, 1 H, J ) 12.2 Hz, CHHOH), 3.71 (AB d, 1 H, J ) 12.2
Hz, CHHOH), 3.01 (dm, 1 H, J ) 16.8 Hz, H-5a), 2.95 (m, 3 H,
H-5b, >CdCHCH2(CH2)3CH3 ), 2.28 (d, 2 H, J ) 5.9 Hz,
COCH2CH(i-Pr)2), 1.82 (sextuplet, 2 H, CHMe2), 1.67 (m, 1 H,
HC(i-Pr)2), 1.40-1.51 (m, 6 H, >CdCHCH2(CH2)3CH3), 0.98
(m, 9 H, >CdCH(CH2)4CH3, HC(CH3)2), 0.89 (d, J ) 6.8 Hz,
HC(CH3)2); 13C NMR (CDCl3) δ 174.54, 168.62, 145.39, 123.70,
82.35, 65.29, 64.55, 46.79, 32.93, 32.73, 31.35, 29.31, 29.29,
28.67, 27.65, 27.62, 22.38, 21.27, 18.65, 13.91; FABMS (m/z,
relative intensity) 369 (MH+, 88), 229 (MH+-C8H16CO, 35).
Anal. (C21H36O5) C, H.
(Z)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en tylid en e]-3-oxo-2-oxola n yl}m eth yl 4-Meth yl-3-(m eth -
yleth yl)p en ta n oa te (31). Standard monoacylation conditions
(method B) starting from 54 (0.100 g, 0.4 mmol) were em-
ployed. After chromatography with different gradients of
hexanes:EtOAc (4:1) and (3:1), the monoacylated product 31
(0.065 g, 43%) was isolated as an oil; IR (neat) 3448 (OH),
1
2959-2874, 1744 (CdO), 1666 cm-1; H NMR (CDCl3) δ 6.34
(m, 1 H, >CdCHCH2CH(i-Pr)2), 4.33 (AB d, 1H, J ) 11.8 Hz,
CHHOCOCH2CH(i-Pr)2), 4.21 (AB d, 1 H, J ) 12.0 Hz,
CHHOCOCH2CH(i-Pr)2), 3.73 (m, 2 H, CH2OH), 2.99 (dm, 1
H, J ) 14.4 Hz, H-5a), 2.80 (m, 3 H, H-5b, >CdCHCH2CH(i-
Pr)2), 2.35 (m, 1 H, OH), 2.28 (d, 2 H, J ) 5.6 Hz, COCH2CH-
(i-Pr)2) 1.84 (heptuplet, 4 H, 4 × CHMe2), 1.68 (m, 1 H,
COCH2CH(i-Pr)2), 1.19 (m, 1 H, >CdCHCH2CH(i-Pr)2), 0.89-
0.99 (m, 24 H, 4 × HC(CH3)2); 13C NMR (CDCl3) δ 174.58,
168.50, 147.67, 122.51, 82.14, 65.22, 64.63, 50.11, 46.79, 33.14,
32.70, 29.31, 29.27, 26.29, 21.61, 21.57, 21.29, 19.39, 19.36,
18.67, 18.65; FABMS (m/z, relative intensity) 411 (MH+, 23).
Anal. (C24H42O5) C, H.
(E)-[4-Hexyliden e-1-(h ydr oxym eth yl)-3-oxo-2-oxolan yl]-
m eth yl 4-Meth yl-3-(m eth yleth yl)p en ta n oa te (28). Stan-
dard monoacylation conditions (method B) starting from 61
(0.12 g, 0.9 mmol) were employed. After chromatography with
different gradients of hexanes:EtOAc (3:1) and (2:1), the
monoacylated product 28 (0.128 g, 40%) was isolated together
with 0.086 g (19%) of the diacylated material. Compound 28:
oil; IR (neat) 3447 (OH), 2958-2873, 1742 (CdO), 1679 cm-1
;
1H NMR (CDCl3) δ 6.86 (m, 1H, >CdCH(CH2)4CH3), 4.35 (AB
d, 1 H, J ) 12.0 Hz, CHHOCOCH2CH(i-Pr)2), 4.24 (AB d, 1 H,
J ) 12.0 Hz, CHHOCOCH2CH(i-Pr)2), 3.81 (AB d, 1 H, J )
12.1 Hz, CHHOH), 3.73 (AB d, 1 H, J ) 12.1 Hz, CHHOH),
2.90 (dm, 1 H, J ) 17.3 Hz, H-5a), 2.74 (dm, 1 H, J ) 17.3 Hz,
H-5b), 2.28 (d, 2 H, J ) 5.6 Hz, COCH2CH(i-Pr)2), 2.24 (m, 2
H, >CdCHCH2(CH2)3CH3), 1.80 (sextuplet, 2 H, CHMe2), 1.66
(m, 1 H, HC(i-Pr)2), 1.39-1.57 (m, 6 H, >CdCHCH2(CH2)3-
CH3), 0.98 (m, 3 H, >CdCH(CH2)4CH3), 0.97 and 0.88 (2 d, 12
H, J ) 6.8 Hz, 2 × HC(CH3)2); 13C NMR (CDCl3) δ 174.53,
169.69, 141.97, 125.64, 82.90, 65.36, 64.74, 46.83, 32.71, 31.40,
30.19, 29.74, 29.32, 29.28, 27.66, 22.35, 21.28, 21.26, 18.66,
18.63, 13.87; FABMS (m/z, relative intensity) 369 (MH+, 100),
229 (MH+-C8H16CO, 65). Anal. (C21H36O5) C, H.
(E)-{1-(H yd r oxym et h yl)-4-[4-m et h yl-3-(m et h ylet h yl)-
p en tylid en e]-3-oxo-2-oxola n yl}m eth yl 4-Meth yl-3-(m eth -
yleth yl)p en ta n oa te (32). Standard monoacylation conditions
(method B) starting from 55 (0.200 g, 0.7 mmol) were em-
ployed. After chromatography with different gradients of
hexanes:EtOAc (4:1) and (3:1), the monoacylated product 32
(0.120 g, 40%) was isolated as an oil: IR (neat) 3450 (OH),
1
2958-2874, 1743 (CdO), 1676 cm-1; H NMR (CDCl3) δ 6.91
(m, 1 H, >CdCHCH2CH(i-Pr)2), 4.37 (AB d, 1H, J ) 12.0 Hz,
CHHOCOCH2CH(i-Pr)2), 4.22 (AB d, 1 H, J ) 12.0 Hz,
CHHOCOCH2CH(i-Pr)2), 3.81 (AB d, 1 H, J ) 12.1, CHHOH),
3.73 (AB d, 1 H, J ) 12.1 Hz, CHHOH), 2.91 (dd, 1 H, J )
(Z)-[4-Decylid en e-1-(h yd r oxym eth yl)-3-oxo-2-oxola n yl]-
m eth yl 4-Meth yl-3-(m eth yleth yl)p en ta n oa te (29). Stan-

Conformationally Constrained Analogues of DAG
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5 941
17.5, 2.3 Hz, H-5a), 2.76 (dd, 1 H, J ) 17.5, 2.2 Hz, H-5b), 2.29
(d, 2 H, J ) 5.6 Hz, COCH2CH(i-Pr)2), 2.21 (dd, 2 H, J ) 7.3,
5.9 Hz, >CdCHCH2CH(i-Pr)2), 1.83 (heptuplet, 4 H, 4 ×
CHMe2), 1.66 (m, 1 H, COCH2CH(i-Pr)2), 1.30 (m, 1H, >Cd
CHCH2CH(i-Pr)2), 0.88-1.00 (m, 24 H, 4 × HC(CH3)2); 13C
NMR (CDCl3) δ 174.51, 169.89, 143.75, 124.79, 83.01, 65.43,
64.74, 50.22, 46.79, 32.68, 29.88, 29.30, 29.15, 28.64, 21.55,
21.27, 19.27, 18.66, 18.63; FABMS (m/z, relative intensity) 411
(MH+, 17). Anal. (C24H42O5‚0.2H2O) C, H.
(1S,2R,3S,5S)-3-(H yd r oxym et h yl)-7,7-d im et h yl-4,6,8-
tr ioxa bicyclo[3.3.0]octa n -2-ol (1,2-O-isop r op ylid en e-r-L-
xylofu r a n ose) (74). This compound was prepared according
to literature procedures.63,64
(1S,2R,3S,5S)-7,7-Dimethyl-4,6,8-trioxa-3-[(phenylmethoxy)-
m eth yl]bicyclo[3.3.0]octa n -2-ol (75). A stirred solution of
1,2-O-isopropylidene-R-L-xylofuranose (74, 20 g, 105 mmol) in
toluene (500 mL) was treated with bis(tributyltin) oxide and
refluxed for 5 h in a flask equipped with a Dean-Stark
apparatus. After the reaction reached room temperature, it
was treated with benzyl bromide (15 mL, 126 mmol) and
tetrabutylammonium bromide (17 g, 52.5 mmol). The temper-
ature was raised to 90 °C, and stirring was continued for 60
h. The reaction mixture was then allowed to cool and was
concentrated. The residue was purified by flash column
chromatography on silica gel with EtOAc:hexanes (1:5) and
(1:1) as eluant to give 75 as a clear oil (29 g, 98.4%): 1H NMR
(CDCl3) δ 7.20-7.40 (m, 5 H, Ph), 5.96 (d, 1 H, J ) 3.6 Hz,
H-5), 4.63 (d of AB, 1 H, J ) 11.9 Hz, PhCHHO), 4.55 (d of
AB, 1 H, J ) 11.9 Hz, PhCHHO), 4.49 (d, 1 H, J ) 3.6 Hz,
H-1), 4.20-4.30 (m, 2 H, H-3, H-2), 3.92 (m, 2 H, PhCH2OCH2),
3.60 (br s, 1 H, OH), 1.46 (s, 3 H, CH3), 1.30 (s, 3 H, CH3).
This compound was used without further purification in the
next step.
-3.340 (c 2.4, CHCl3); IR (neat) 3450 and 3281 (OH) cm-1; 1H
NMR (CDCl3) δ 7.20-7.40 (m, 5 H, Ph), 5.78 (d, 1 H, J ) 3.9
Hz, H-5), 4.62 (d of AB, 1 H, J ) 12.2 Hz, PhCHHO), 4.52 (d
of AB, 1 H, J ) 12.2 Hz, PhCHHO), 4.29 (d, 1 H, J ) 3.9 Hz,
H-1), 4.01 (dd, 1 H, J ) 3.3, 7.4 Hz, H-3), 3.50-3.72 (m, 4 H,
PhCH2OCH2 and CH2OH), 1.40-1.80 (m, 4 H, CH2CH2CH2-
OH), 1.56 (s, 3 H, CH3), 1.34 (s, 3 H, CH3); 13C NMR (CDCl3)
δ 137.88 (s), 128.41 (d), 127.83 (d), 127.71 (d), 112.45 (s), 103.82
(d), 81.77 (d), 80.43 (d), 78.77 (s), 73.54 (t), 67.98 (t), 62.80 (t),
27.34 (t), 26.75 (q), 26.54 (q), 25.24 (t). Anal. (C18H26O6) C, H.
(1S ,5S ,7S ,8S )-3,3-Dim e t h yl-2,4,6-t r ioxa -7-[(p h e n yl-
m eth oxy)m eth yl]spir o[bicyclo[3.3.0]octan e-8,5′-oxolan e]-
10-on e (73). A solution of 77 (2.84 g, 8.4 mmol) in CH2Cl2
(100 mL) was treated with pyridinium chlorochromate (7.24
g, 33.6 mmol) and 4 Å molecular sieve (8.4 g). After being
stirred for 1 h at room temperature, the reaction mixture was
quenched with ether and Celite. It was stirred further for 30
min more and filtered through a pad of silica gel. The filtrate
and EtOAc washings were collected and reduced to dryness
before purification by silica gel with EtOAc:hexanes (2:1) as
eluant to give 73 (2.75 g, 98%) as a colorless oil: [R]D -44.50
(c 1.50, CHCl3); IR (neat) 1785 (CdO) cm-1; 1H NMR (CDCl3)
δ 7.20-7.40 (m, 5 H, Ph), 5.80 (d, 1 H, J ) 3.7 Hz, H-5), 4.53
(s, 2 H, PhCH2O), 4.45 (t, 1 H, J ) 5.0 Hz, H-7), 4.37 (d, 1 H,
J ) 3.7 Hz, H-1), 3.65 (m, 2 H, PhCH2OCH2), 2.55 (m, 2 H,
H-11a,b), 2.40 (m, 1 H, H-12a), 1.91 (m, 1 H, H-12b), 1.59 (s, 3
H, CH3), 1.34 (s, 3 H, CH3); 13C NMR (CDCl3) δ 175.12 (s),
137.28 (s), 128.51 (d), 127.95 (d), 127.85 (d), 114.32 (s), 103.54
(d), 87.94 (s), 84.12 (d), 77.37 (d), 73.84 (t), 67.40 (t), 27.90 (t),
26.75 (q), 26.65 (q), 25.83 (t). Anal. (C18H22O6) C, H.
(5R)-5-(Hyd r oxym et h yl)-5-[(1S)-1-h yd r oxy-2-(p h en yl-
m eth oxy)eth yl]oxola n -2-on e (78). A solution of 73 (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 in EtOAc
and treated with NaHCO3 and MgSO4. The suspension was
filtered, and the filtrate was concentrated in vacuo to give the
intermediate hemiacetal which was used for the next step
without further purification. The resulting compound was
dissolved in MeOH (150 mL) and H2O (75 mL) and treated
with sodium periodate (10.7 g, 50 mmol). After being stirred
for 1 h at room temperature, the reaction mixture was filtered
and concentrated in vacuo. The residue was dissolved in
EtOAc, dried (MgSO4), and concentrated in vacuo to give the
intermediate aldehyde which was reduced immediately with-
out further purificaiton. After it was dissolved in MeOH (200
mL) and cooled to 0 °C, sodium borohydride was added
portionwise until the starting material was consumed accord-
ing to TLC. The reaction mixture was first quenched with
acetone and then 20 min after with acetic acid before concen-
trating in vacuo. The residue was diluted with EtOAc and
filtered, and the filtrate was concentrated in vacuo. Purifica-
tion by flash column chromatography on silica gel with a
gradient of EtOAc:hexanes (3:1) to EtOAc:MeOH (9:1) as
eluant afforded 78 (5.0 g, 75% in 3 steps) as a syrup: [R]D
(1R,3S,5S)-7,7-Dim eth yl-4,6,8-tr ioxa-3-[(ph en ylm eth oxy)-
m eth yl)bicyclo[3.3.0]octa n -2-on e (76). A solution of 75 (14
g, 50 mmol) and acetic acid (5.2 mL, 90 mmol) in CH2Cl2 (100
mL) was added dropwise to a stirred suspension of pyridinium
dichromate (56.43 g, 150 mmol) and 4 Å molecular sieve (30
g) in CH2Cl2 (100 mL). Note: due to the strong exothermic
nature of this reaction the addition must be done very slowly
and with great care. After the additon was finished, stirring
continued for 1 h. The reaction mixture was then treated with
Celite and ether (400 mL) and stirred for 20 min. Following
filtration through a short pad of silica gel and further elution
with ether and EtOAc, the combined filtrate was concentrated.
The residue was purified by flash column chromatography on
silica gel with EtOAc:hexanes (1:2) as eluant to give 76 as an
oil (13.62 g, 98%): [R]D ) 147.30 (c 4.25, CHCl3); IR (neat) 1774
1
(CdO) cm-1; H NMR (CDCl3) δ 7.20-7.40 (m, 5 H, Ph), 6.12
(d, 1 H, J ) 4.5 Hz, H-5), 4.40-4.55 (m, 3 H, H-4, PhCH2O),
4.33 (d, 1 H, J ) 4.5 Hz, H-1), 3.70 (d of AB, 2 H, J ) 2.4 Hz,
PhCH2OCH2), 1.43 (s, 3 H, CH3), 1.40 (s, 3 H, CH3); 13C NMR
(CDCl3) δ 209.94 (s), 137.39 (s), 128.47 (d), 127.85 (d), 127.51
(d), 114.11 (s), 103.57 (d), 79.91 (d), 76.81 (d), 73.69 (t), 70.08
(t), 27.61 (q), 27.17 (q). Anal. (C15H18O5) C, H.
(1S,2S,3S,5S)-2-(3-Hyd r oxyp r op yl)-7,7-d im eth yl-4,6,8-
tr ioxa -3[(p h en ylm eth oxy)m eth yl)bicyclo[3.3.0]octa n -2-
ol (77). A solution of 3-chloropropanol (2.84 g, 30 mmol) in
THF (10 mL) was cooled to -20 °C and treated dropwise while
stirring with a solution of methylmagnesium chloride (10 mL,
3 M in THF). The reaction mixture was allowed to warm to
room temperature, and magnesium (1.10 g, 45 mmol) was
added. The resulting suspension was refluxed and treated at
0, 1, and 2 h intervals with dibromoethane (0.02 mL). After 3
h, the solution was cooled to room temperature. A solution of
76 (2.78 g, 10 mmol) in THF (10 mL) was added dropwise to
the above Grignard reagent, and after stirring for 1 h the
reaction was cooled over ice and quenched by the addition of
a saturated solution of NH4Cl (20 mL). The reaction mixture
was filtered, and the filtrate was extracted several times with
EtOAc. The combined organic extract was dried (MgSO4) and
concentrated. The residue was purified by flash column
chromatography on silica gel with EtOAc:hexanes (4:1) as
eluant to give 77 (2.84 g, 85%) as white solid: mp 103 °C; [R]D
-8.000 (c 1.25, CHCl3); IR (neat) 3418 (OH), 1760 (CdO) cm-1
;
1H NMR (CDCl3) δ 7.20-7.40 (m, 5 H, Ph), 4.52 (s, 2 H,
PhCH2O), 3.93 (dd, 1 H, J ) 4.1, 6.7 Hz, CH(OH)CH2OCH2-
Ph), 3.75 (d of AB, 1 H, J ) 12.1 Hz, CH(OH)CH2OCHHPh),
3.61 (d of AB, 1 H, J ) 12.1 Hz, CH(OH)CH2OCHHPh), 3.50-
3.66 (m, 2 H, CH2OH), 2.88 (br s, 2 H, OH), 2.50-2.63 (m, 2
H, H-3), 2.08-2.35 (m, 2 H, H-4); 13C NMR (CDCl3) δ 177.73,
137. 25, 128.47, 127.94, 127.77, 88.64, 73.61, 72.37, 69.78,
64.84, 29.00, 25.08. Anal. (C14H18O5) C, H.
(5R)-5-[(1S)-1,2-Dih yd r oxyeth yl]-5-(h yd r oxym eth yl)ox-
ola n -2-on e (79). A solution of 78 (3.0 g, 11.3 mmol) in MeOH
(100 mL) was treated with acetic acid (0.1 mL) and hydroge-
nated for 5 h under a balloon of hydrogen in the presence of
10% palladium on carbon (3.0 g). The reaction mixture was
filtered, and the filtrate was concentrated in vacuo. The
residue was purified by flash column chromatography on silica
gel with CHCl3:MeOH (5:1) as eluant to give the triol as a
syrup which crystallized after cooling in the refrigerator to give

942 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5
Nacro et al.
79 as a white solid (1.965 g, 99%): mp 78-79 °C; [R]D -4.20 (c
3.68, MeOH); IR (KBr) 3385 (OH), 1758 (CdO) cm-1; 1H NMR
(CD3OD) δ 3.48-3.84 (m, 5 H), 2.54-2.68 (m, 2 H), 2.10-2.40
(m, 2 H); 13C NMR (CD3OD) δ 180.3, 91.3, 75.2, 65.3, 63.0,
30.2, 26.1. Anal. (C7H12O5) C, H.
Z-Isom er 83: [R]D +2.620 (c 1.0, CHCl3); IR (neat) 1760 (Cd
1
O), 1678 (CdC) cm-1; H NMR (CDCl3) δ 7.25-7.38 (m, 5 H,
Ph), 6.18 (m, 1 H, >CdCH), 4.55 (s, 2 H, PhCH2O), 4.23 (t, 1
H, J 1′,5′a ) J 1′,5′b ) 6.8 Hz, OCHCH2O), 3.98 (dd, 1 H, J 5′a,5′b
8.7 Hz, J 5′a,1′ ) 6.8 Hz, OCHCHHO), 3.94 (dd, 1 H, J 5′b,5′a
)
)
8.7 Hz, J 5′b,1′ ) 6.8 Hz, OCHCHHO), 3.51 (s, 2 H, PhCH2OCH2),
2.95 (m, 1 H, H-4a), 2.60-2.85 (m, 3 H, H-4b and >CdCH-
CH2), 1.77 (m, 2 H, 2 × CH(CH3)2), 1.37 (s, 3 H, CH3), 1.33 (s,
3 H, CH3), 1.10 (m, 1 H, CH(i-Pr)2), 0.83-0.92 (m, 12 H, CH3);
13C NMR (CDCl3) δ 169.04, 145.89, 137.44, 128.33, 127.69,
127.43, 123.36, 109.40, 82.08, 77.45, 73.58, 72.88, 64.60, 51.08,
33.29, 29.23, 29.17, 26.03, 25.79, 25.17, 21.60, 21.51, 19.39,
19.24. Anal. (C26H38O5) C, H.
(5R)-5-[(1S)-3,3-Dim eth yl(2,4-d ioxola n yl)]-5-(h yd r oxy-
m eth yl)oxola n -2-on e (80). A solution of 79 (1.94 g, 11 mol)
in anhydrous acetone (120 mL) was treated with p-toluene-
sulfonic acid (0.13 g, 0.68 mmol) and stirred for 4 h at room
temperature. The reaction mixture was neutralized with solid
NaHCO3, filtered, and concentrated in vacuo. The residue was
purified by flash column chromatography on silica gel with
EtOAc:hexanes (3:1) as eluant to give 80 as a white solid (2.31
g, 97%): mp 69-71 °C; [R]D -30.840 (c 0.83, CHCl3); IR (KBr)
3445 (OH), 1769 (CdO) cm-1; 1H NMR (CDCl3) δ 4.16 (t, 1 H,
J 1′,5′a ) J 1′,5′b ) 6.8 Hz, OCHCH2O), 4.06 (dd, 1 H, J 5′a,5′b ) 8.5
Hz, J 5′a,1′ ) 6.8 Hz, OCHCHHO), 3.93 (dd, 1 H, J 5′b,5′a ) 8.5
Hz, J 5′b,1′ ) 6.8 Hz, OCHCHHO), 3.67 (d of AB, 1 H, J ) 12.4
Hz, CHHOH), 3.61 (d of AB, 1 H, J ) 12.4 Hz, CHHOH), 2.48-
2.80 (m, 2 H, H-3), 2.05-2.35 (m, 2 H, H-4), 1.36 (s, 3 H, CH3),
1.34 (s, 3 H, CH3); 13C NMR (CDCl3) δ 178.01, 108.89, 87.34,
78.35, 65.11, 64.73, 29.06, 25.55, 25.50, 25.03. Anal. (C10H16O5)
C, H.
E-Isom er 82: [R]D -5.160 (c 1.0, CHCl3); IR (neat) 1760 (Cd
1
O), 1678 (CdC) cm-1; H NMR (CDCl3) δ 7.25-7.38 (m, 5 H,
phenyl), 6.74 (m, 1 H, >CdCH), 4.55 (s, 2 H, PhCH2O), 4.26
(t, 1 H, J 1′,5′a ) J 1′,5′b ) 6.8 Hz, OCHCH2O), 4.02 (dd, 1 H, J 5′a,5′b
) 8.7 Hz, J 5′a,1′ ) 6.8 Hz, OCHCHHO), 3.92 (dd, 1 H, J 5′b,5′a
)
8.7 Hz, J 5′b,1′ ) 7.0 Hz, OCHCHHO), 3.54 (s, 2 H, PhCH2OCH2),
2.92 (m, 1 H, H-4a), 2.68 (m, 1 H, H-4b), 2.11 (m, 2 H, >Cd
CH-CH2), 1.78 (m, 2 H, 2 × CH(CH3)2), 1.36 (s, 3 H, CH3),
1.33 (s, 3 H, CH3), 1.10 (1 H, CH(i-Pr)2), 0.82-0.92 (m, 12 H);
13C NMR (CDCl3) δ 170.20, 142.00, 137.35, 128.29, 127.66,
127.40, 125.49, 109.36, 82.06, 77.38, 73.56, 72.83, 64.57, 50.20,
30.07, 29.05, 28.48, 25.73, 25.09, 21.51, 21.48, 19.24, 19.09.
Anal. (C26H38O5) C, H.
(5R)-5-[(1S)-3,3-Dim eth yl(2,4-d ioxola n yl)]-5-[(p h en yl-
m eth oxy)m eth yl]oxola n -2-on e (81). A stirred solution of 80
(2.16 g, 10 mmol) in DMF (20 mL) was allowed to react with
benzyl bromide (3.60 mL, 30 mmol) in the presence of silver-
(I) oxide (3.48 g, 15 mmol) for 4 days at room temperature.
The reaction mixture was filtered, and the filter cake was
washed with ether. Water was added to the combined filtrate,
and the ether layer was separated. The aqueous layer was then
extracted with ether three times, and the combined ether
extract was washed with H2O, dried (MgSO4), 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 81 as an oil (2.573 g, 84%): IR (neat) 1777 (Cd
O) cm-1; [R]D -17.090 (c 2.13, CHCl3); 1H NMR (CDCl3) δ 7.25-
Hexyl (E)-3-{(1S)-(Hyd r oxym eth yl)-4-[(Z)-4-m eth yl-3-
(m e t h yle t h yl)p e n t ylid e n e ]-3-oxo-2-oxa la n yl}p r op -2-
en oa te (69). A stirred solution of 83 (0.1 g, 0.2 mmol) in ether
(10 mL) was treated with periodic acid (0.27 g, 1.2 mmol) for
24 h at room temperature. The reaction mixture was filtered,
and the filtrate was concentrated in vacuo. The residue was
purified by flash column chromatography over silica gel with
EtOAc:hexanes (1:1) as eluant to give the correponding alde-
hyde (0.08 g, 0.227 mmol, 96%). The compound was dissolved
in CH2Cl2 (8 mL) and treated with hexyl (triphenylphospho-
ranylidene)acetate (0.19 g, 0.5 mmol). The solution was stirred
for 24 h at room temperature and concentrated in vacuo. The
residue was purified by flash column chromatography over
silica gel with EtOAc:hexanes (1:9) as eluant to give interme-
diate 85 (0.09 g, 0.2 mmol, 93%) as an oil: 1H NMR (CDCl3) δ
7.25-7.38 (m, 5 H, Ph), 6.92 (d, 1 H, J ) 15.6 Hz, CHd
CHCO2C6H13), 6.19 (m, 1 H, >CdCH), 6.14 (d, 1 H, J ) 15.6
Hz, CHdCHCO2C6H13), 4.57 (AB t, J ) 12.9 Hz, 2 H, PhCH2O),
4.13 (t, 1 H, J ) 6.6 Hz, CO2CH2C5H11), 3.53 (s, 2 H, PhCH2-
OCH2), 3.10 (m, 1 H, H-5a), 2.66-2.76 (m, 3 H, H-5b, >Cd
CHCH2), 1.77 (m, 2 H, 2 × CH(CH3)2), 1.63 (m, 2 H, CO2-
CH2CH2C4H9), 1.25-1.40 (m, 6 H, CO2(CH2)2(CH2)3CH3), 1.08
(1 H, CH(i-Pr)2), 0.80-0.92 (m, 15 H). A solution of this
compound (0.09 g, 0.2 mmol) in CH2Cl2 (15 mL) was cooled to
-78 °C, treated with boron trichloride (1 M in CH2Cl2, 0.85
mL, 0.85 mmol), and stirred for 2 h at -78 °C. The reaction
was quenched with saturated NaHCO3 solution and im-
mediately partitioned between ice-cold ether and water. The
organic layer was washed with water, dried (MgSO4), and
concentrated in vacuo. The residue was purified by flash
column chromatography on silica gel with EtOAc:hexanes
(1:2) as eluant to give 69 (0.042 g, 50%) as an oil: [R]D +11.910
7.35 (m, 5 H, Ph), 4.53 (s, 2 H, PhCH2O), 4.18 (t, 1 H, J 1′,5′a
)
J 1′,5′b ) 6.9 Hz, OCHCH2O), 4.02 (dd, 1 H, J 5′a,5′b ) 8.7 Hz,
J 5′a,1′ ) 6.9 Hz, OCHCHHO), 3.93 (dd, 1 H, J 5′b,5′a ) 8.7 Hz,
J 5′b,1′ ) 6.9 Hz, OCHCHHO), 3.50 (s, 2 H, PhCH2OCH2), 2.48-
2.75 (m, 2 H, H-3), 2.0-2.38 (m, 2 H, H-4), 1.35 (s, 3 H, CH3),
1.33 (s, 3 H, CH3); 13C NMR (CDCl3) δ 177.02, 137.25, 128.37,
127.77, 127.44, 109.18, 85.61, 78.42, 73.58, 73.11, 64.77, 28.81,
26.61, 25.64, 25.29. Anal. (C17H22O5) C, H.
(5R)-5-[(1S)-3,3-Dim eth yl(2,4-dioxolan yl)]-3-[(Z)-4-m eth -
yl-3-(m e t h yle t h yl)p e n t ylid e n e )-5-[(p h e n ylm e t h oxy)-
m eth yl]oxola n -2-on e (82) a n d (5R)-5-[(1S)-3,3-Dim eth yl-
(2,4-d ioxola n yl)]-3-[(E)-4-m eth yl-3-(m eth yleth yl)p en tyli-
d en e)-5-[(p h en ylm eth oxy)m eth yl]oxola n -2-on e (83). A
stirred solution of 81 (0.613 g, 2.0 mmol) in THF (8 mL) was
cooled to -78 °C and treated slowly with lithium bis(trimeth-
ylsilyl)amide (1 M in THF, 2.4 mL, 2.4 mmol). After 30 min, a
solution of 4-methyl-3-(methylethyl)pentanal (0.370 g, 2.6
mmol) in hexamethyl-phosphoramide (0.42 mL, 2.4 mmol) and
THF (1 mL) was added, and stirring was continued for 2 h at
-78 °C. The reaction was quenched with
a solution of
saturated ammonium chloride and diluted with ether. The
organic layer was washed with H2O, dried (MgSO4), and
concentrated in vacuo. The residue was purified by flash
column chromatography on silica gel with EtOAc:hexanes
(1:4) as eluant to give the intermediate â-hydroxy lactone (0.79
g, 89%) as an oil. The compound was dissolved in CH2Cl2 (20
mL), cooled to 0 °C, and treated with triethylamine (1.0 mL)
and methanesulfonyl chloride (0.29 mL) with stirring for 2 h
at room temperature. The reaction mixture was cooled to 0
°C, and 1,8-diazabicyclo[5,4,0]undec-7-ene (1.43 mL) was
added. After stirring for 1 h at room temperature, the mixture
was concentrated in vacuo. The residue was diluted with ether,
washed with 1 N HCl and H2O, dried (MgSO4), and concen-
trated in vacuo. The residue was purified by flash column
chromatography on silica gel with EtOAc:hexanes (1:5) as
eluant to give 83 (Z-isomer, 0.296 g, 35%) and 82 (E-isomer:
0.203 g, 24%) as oils.
(c 1.0, CHCl3); IR (KBr) 1760, 1716 (CdO), 1660 (CdC) cm-1
;
1H NMR (CDCl3) δ 6.86 (d, 1 H, J ) 15.6 Hz, CH)CHCO2-
C6H13), 6.26 (m, 1 H, >CdCH), 6.14 (d, 1 H, J ) 15.6 Hz, CHd
CHCO2C6H13), 4.14 (t, 1 H, J ) 6.6 Hz, CO2CH2C5H11), 3.76
(dd, 1 H, J ) 12.1, 6.3 Hz, CHHOH), 3.64 (dd, 1 H, J ) 12.2,
6.3 Hz, CHHOH), 3.13 (m, 1 H, H-5a), 2.66-2.76 (m, 3 H, H-5b
and >CdCHCH2), 2.03 (t, 1 H, OH), 1.77 (m, 2 H, 2 × CH
(CH3)2), 1.65 (m, 2 H, CO2CH2CH2C4H9), 1.25-1.40 (m, 6 H,
CO2(CH2)2(CH2)3CH3), 1.10 (m, 1 H, CH(i-Pr)2), 0.80-0.92 (m,
15 H); 13C NMR (CDCl3) δ 168.64, 165.86, 148.23, 145.05,
122.24, 122.13, 83.40, 66.39, 65.04, 51.10, 35.74, 31.37, 29.32,
29.26, 28.50, 26.37, 25.53, 22.48, 21.62, 21.53, 19.39, 19.33,
13.98. Anal. (C23H38O5) C, H.
Hexyl (E)-3-{(1S)-(Hyd r oxym eth yl)-4-[(E)-4-m eth yl-3-
(m e t h yle t h yl)p e n t ylid e n e ]-3-oxo-2-oxa la n yl}p r op -2-

Conformationally Constrained Analogues of DAG
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5 943
(16) Wender, P. A.; DeBrabander, J .; Harran, P. G.; J imenez, J . M.;
Koehler, M. F. T.; Lippa, B.; Park, C. M.; Shiozaki, M. Synthesis
of the First Members of a New Class of Biologically Active
Bryostatin Analogues. J . Am. Chem. Soc. 1998, 120, 4534-4535.
(17) Wender, P. A.; DeBrabander, J .; Harran, P. G.; J imenez, J . M.;
Koehler, M. F. T.; Lippa, B.; Park, C. M.; Siedenbiedel, C.; Pettit,
G. R. The Design, Computer Modeling, Solution Structure, and
Biological Evaluation of Synthetic Analogues of Bryostatin 1.
Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 6624-6629.
(18) Endo, Y.; Ohno, M.; Hirano, M.; Itai, A.; Shudo, K. Synthesis,
Conformation, and Biological Activity of Teleocidin Mimics,
Benzolactams. A Clarification of the Conformational Flexibility
Problem in Structure-Activity Studies of Teleocidins. J . Am.
Chem. Soc. 1996, 118, 1841-1855.
(19) Endo, Y.; Hirano, M.; Driedger, P. E.; Stabel, S.; Shudo, K. Novel
Conformationally Constrained Analogues of Diacylglycerol. Pro-
tein Kinase C Binding Affinity of Simplified Compounds Based
on a six-membered Lactam Moiety Bioorg. Med. Chem. Lett.
1997, 7, 2997-3000.
(20) Endo, Y.; Shimazu, M.; Fukasawa, H.; Driedger, P. E.; Kimura,
K.; Tomioka, N.; Itai, A.; Shudo, K. Synthesis, Computer
Modeling and Biological Evaluation of Novel Protein Kinase C
Agonists Based on a seven-membered Lactam Moiety. Bioorg.
Med. Chem. Lett. 1999, 9, 173-178.
en oa te (70). This compound was prepared from 82 in 44%
yield following the same procedure as for the synthesis of 69.
Com p ou n d 84: 1H NMR (CDCl3) δ 7.25-7.38 (m, 5 H, Ph),
6.95 (d, 1 H, J ) 15.8 Hz, CH)CHCO2C6H13), 6.80 (m, 1 H,
>CdCH), 6.15 (d, 1 H, J ) 15.8 Hz, CHdCHCO2C6H13), 4.57
(dd of AB, J ) 11.9 Hz, 2 H, PhCH2O), 4.13 (t, 1 H, J ) 6.6
Hz, CO2CH2C5H11), 3.54 (s, 2 H, PhCH2OCH2), 3.10 (dm, 1 H,
J ≈ 19 Hz, H-5a), 2.68 (dm, 1 H, J ≈ 19 Hz, H-5b), 2.08 (m, 2
H, >CdCHCH2), 1.76 (m, 2 H, 2 × CH(CH3)2), 1.65 (m, 2 H,
CO2CH2CH2C4H9), 1.25-1.40 (m, 6 H, CO2(CH2)2(CH2)3CH3),
1.19 (1 H, CH(i-Pr)2), 0.80-0.92 (m, 15 H).
Com p ou n d 70: [R]D +24.670 (c 0.5, CHCl3); IR (KBr) 1760,
1
1723 (CdO), 1681 (CdC) cm-1; H NMR (CDCl3) δ 6.90 (d, 1
H, J ) 15.6 Hz, CHdCHCO2C6H13), 6.82 (m, 1 H, >CdCH),
6.14 (d, 1 H, J ) 15.6 Hz, CHdCHCO2C6H13), 4.14 (t, 1 H, J
) 6.6 Hz, CO2CH2C5H11), 3.80 (d, 1 H, J ) 12.2 Hz, CHHOH),
3.66 (d, 1 H, J ) 12.2 Hz, CHHOH), 3.09 (dm, 1 H, J ≈ 18 Hz,
H-5a), 2.72 (dm, 1 H, J ≈ 18 Hz, H-5b), 2.12 (m, 2 H, >Cd
CHCH2), 1.78 (m, 2 H, 2 × CH(CH3)2), 1.65 (m, 2 H, CO2-
CH2CH2C4H9), 1.25-1.40 (m, 6 H, CO2(CH2)2(CH2)3CH3), 1.10
(1 H, CH(i-Pr)2), 0.80-0.92 (m, 15 H); 13C NMR (CDCl3) δ
169.66, 165.78, 144.75, 144.64, 124.03, 122.33, 83.71, 66.72,
65.07, 50.30, 32.46, 31.39, 29.17, 28.73, 28.50, 25.53, 22.50,
21.56, 19.30, 13.98. Anal. (C23H38O5) C, H.
(21) Qiao, L.; Wang, S.; George, C.; Lewin, N. E.; Blumberg, P. M.;
Kozikowski, A. P. Structure-Based Design of a New Class of
Protein Kinase C Modulators. J . Am. Chem. Soc. 1998, 120,
6629-6630.
(22) Kong, F.; Kishi, Y.; Perez-Sala, D.; Rando, R. R. The Pharma-
cophore of Debromoaplysiatoxin Responsible for Protein Ki-
nase-C Activation. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 1973-
1976.
(23) 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,
37, 1326-1338.
Ack n ow led gm en t. This work was supported in part
by the Research Institute of Pharmaceutical Science,
College of Pharmacy, Seoul National University. The
authors thank Dr. J ames A. Kelley from the LMC, NCI,
for mass spectral analysis and interpretation.
(24) Sharma, R. Lee, J .; Wang, S.; Milne, G. W. A.; Lewin, N. E.;
Blumberg, P. M.; Marquez, V. E. Conformationally Constrained
Analogues of Diacylglycerol 0.10. Ultrapotent Protein Kinase C
Refer en ces
(1) For paper 15 in the series, see: Benzaria, S.; Bienfait, B.; Nacro,
K.; Wang S.; Lewin, N. E.; Beheshti, M.; Blumberg, P. M.;
Marquez, V. E. Conformationally Constrained Analogues of
Diacylglycerol. 15. The Indispensable Role of the sn-1 and sn-2
Carbonyls in the Binding of DAG-Lactones to Protein Kinase C
(PK-C). Bioorg. Med. Chem. Lett. 1998, 8, 3403-3408.
(2) Quest, A. F. G. Regulation of Protein Kinase C: A Tale of Lipids
and Proteins. Enzyme Protein 1996, 49, 231-261.
(3) Newton, A. C. Protein Kinase C: Structure, Function, and
Regulation. J . Biol. Chem. 1995, 270, 28495-28498.
(4) Sakai, N.; Sasaki, K.; Ikegaki, N.; Shirai, Y.; Ono, Y.; Saito, N.
Direct Visualization of the Translocation of the γ-Subspecies of
Protein Kinase C in Living Cells Using Fusion Proteins with
Green Fluorescent Protein. J . Cell. Biol. 1997, 139, 1465-1476.
(5) Oancea, E.; Meyer, T. Protein Kinase C as a Molecular Machine
for Decoding Calcium and Diacylglycerol Signals. Cell 1998, 95,
307-318.
(6) Rhee, S. G.; Bae, Y. S. Regulation of Phosphoinositide-Specific
Phospholipase C Isozymes. J . Biol. Chem. 1997, 272, 15045-
15048.
(7) Exton, J . H. Phospholipase D: Enzymology, Mechanisms of
Regulation, and Function. Physiol. Rev. 1997, 77, 303-320.
(8) Livneh, E.; Fishman, D. D. Linking Protein Kinase C to Cell-
Cycle Control. Eur. J . Biochem. 1997, 248, 1-9.
(9) Ono, Y.; Fujii, T.; Igarashi, K.; Kuno, T.; Tanaka, C.; Kikkawa,
U.; Nishizuka, Y. Phorbol Ester Binding to Protein Kinase-C
Requires a Cysteine-Rich Zinc-Finger-Like Sequence. Proc. Natl.
Acad. Sci. U.S.A. 1989, 86, 4868-4871.
(10) Kazanietz, M. G.; Krausz, K. W.; Blumberg, P. M. Differential
Irreversible Insertion of Protein Kinase-C into Phospholipid
Vesicles by Phorbol Esters and Related Activators. J . Biol. Chem.
1992, 267, 20878-20886.
(11) Kishi, Y.; Rando, R. R. Structural basis of protein kinase C
activation by tumor promoters. Acc. Chem. Res. 1998, 31, 163-
172.
(12) Wender, P. A.; Cribbs, C. M.; Koehler, K. F.; Sharkey, N. A.;
Herald, C. L.; Kamano, Y.; Pettit, G. R.; Blumberg, P. M.
Modeling of the Bryostatins to the Phorbol Ester Pharmacophore
on Protein Kinase C. Proc. Natl. Acad. Sci. U.S.A. 1988, 85,
7197-7201.
(13) Hecker, E. In Carcinogenesis, Vol. 2: Mechanisms of Tumor
Promotion and Cocarcinogenesis; Slaga, T. J ., Sivak, A. J .,
Boutwell, R. K., Eds.; Raven Press: New York, 1978; pp 11-49.
(14) The status of bryostatin in 32 clinical trials can be found on the
NCI Web page at http://cancernet.nci.nih.gov/prot/patsrch.shtml.
(15) Wender, P. A.; DeBrabander, J .; Harran, P. G.; Hinkle, K. W.;
Lippa, B.; Pettit, G. R. Synthesis and Biological Evaluation of
Fully Synthetic Bryostatin Analogues. Tetrahedron Lett. 1998,
39, 8625-8628.
Ligands Based on a Racemic 5-Disubstituted Tetrahydro-2-
Furanone Template. J . Med. Chem. 1996, 39, 19-28.
(25) Lee, J .; Wang, S.; Milne, G. W. A.; Sharma, R.; Lewin, N. E.;
Blumberg, P. M.; Marquez, V. E. Conformationally Constrained
Analogues of Diacylglycerol 0.11. Ultrapotent Protein Kinase C
Ligands Based on a Chiral 5-Disubstituted Tetrahydro-2-Fura-
none Template. J . Med. Chem. 1996, 39, 29-35.
(26) Lee, J .; Sharma, R.; Wang, S.; Milne, G. W. A.; Lewin, N. E.;
Szallasi, Blumberg, P. M.; George. C.; Marquez, V. E. Confor-
mationally Constrained Analogues of Diacylglycerol. 12. Ul-
trapotent Protein Kinase C Ligands Based on a Chiral 4,4-
Disubstituted Heptono-1,4-Lactone Template. J . Med. Chem.
1996, 39, 36-45.
(27) Zhang, G.; Kazanietz, M. G.; Blumberg, P. M.; Hurley, J . H.
Crystal Structure of the Cys2 Activator-Binding Domain of
Protein Kinase C Delta in Complex with Phorbol Ester. Cell
1995, 81, 917-924.
(28) Ichikawa, S.; Hatanaka, H.; Takeuchi, Y.; Ohno, S.; Inagaki, F.
Solution Structure of Cysteine-Rich Domain of Protein Kinase
C Alpha. J . Biochem. 1995, 117, 566-574.
(29) Cochet, C.; Gill, G. N.; Meisenhelder, J .; Cooper, J . A.; Hunter,
T. C-Kinase Phosphorylates the Epidermal Growth Factor
Receptor and Reduces its Epidermal Growth Factor-Stimulated
Tyrosine Protein Kinase Activity. J . Biol. Chem. 1984, 259,
2553-2558.
(30) Grever, M. R.; Schepartz, S. A.; Chabner, B. A. The National
Cancer Institute: Cancer Drug Discovery and Development
Program. Semin. Oncol. 1992, 19, 622-638. Additional informa-
tion on the NCI in vitro screen can be found at the following
Web address: http://dtp.nci.nih.gov/.
(31) Krauter, G.; Von der Lieth, C.-W.; Schmidt, R.; Hecker, E.
Structure/Activity Relationships of Polyfunctional Diterpenes of
the Tigliane Type-A Pharmacophore Model for Protein-Kinase-C
Activators Based on Structure/Activity Studies and Molecular
Modeling of the Tumor Promoters 12-O-Tetradecanoylphorbol
13-Acetate and 3-O-Tetradecanoylingenol. Eur. J . Biochem.
1996, 242, 417-427.
(32) Sugita, K.; Neville, C. F.; Sodeoka, M.; Sasai, H.; Shibasaki, M.
Stereocontrolled Syntheses of Phorbol Analogues and Evaluation
of their Binding Affinity to PKC. Tetrahedron Lett. 1995, 36,
1067-1070.
(33) The molecular graphics in this figure was generated with the
program VMD, Humphrey, W.; Dalke, W.; Schulten, K. VMD:
Visual Molecular Dynamics. J . Mol. Graphics 1996, 14, 33-38.
(34) Davis, R. J .; Ganong, B. R.; Bell, R. M.; Czech. Structural
Requirements for Diacylglycerol to Mimic Tumor-promoting
Phorbol Diester Action on the Epidermal Growth Factor Recep-
tor. J . Biol. Chem. 1985, 260, 5315-5329.

944 J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 5
Nacro et al.
(35) Ganong, B. R.; Loomis, C. R.; Bell, R. M. Specificity and
Mechanism of Protein Kinase C Activation by sn-1,2-Diacyl-
glycerols. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 1184-1188.
(36) Xu, R. X.; Pawelczyk, T.; Xia, T.-H.; Brown, S. C. NMR Structure
of a Protein Kinase C-Gamma Phorbol-Binding Domain and
Study of Protein-Lipid Micelle Interactions. Biochemistry 1997,
36, 10709-10717.
(37) Denmark, S. E.; Harmata, M. A.; White, K. S. Studies on the
Addition of Allyl Oxides to Sulfonylallenes. Preparation of Highly
Substituted Allyl Vinyl Ethers for Carbanaionic Claisen Rear-
rangements. J . Org. Chem. 1987, 52, 4031-4042.
(38) Lee, J .; Sharma, R.; Marquez, V. E. Synthesis of Chiral Doubly
C-4 Branched γ-Lactones. An Efficient Synthesis of 4-C-
Hydroxymethyl-2,3-Dideoxy-Hexono-1,4-Lactones. Carbohydr.
Lett. 1995, 1, 285-290.
(39) KOWWIN 1.63, Syracuse Research Corp., http://esc.syrres.com/.
(40) Kazanietz, M. G.; Areces, 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.
(41) Morris, G. M.; Goodsell, D. S.; Huey, R.; Olson, A. J . Distributed
Automated Docking of Flexible Ligands to Proteins: Parallel
Applications of AutoDock 2.4. J . Comput.-Aid. Mol. Des. 1996,
10, 293-304.
(42) Teng, K.; Marquez, V. E.; Milne, G. W. A.; Barchi, J r., J . J .;
Kazanietz, M. G.; Lewing, N. E.; Blumberg, P. M.; Abushanab,
E. Conformationally Constrained Analogues of Diacylglycerol -
Interaction of gamma-Lactones with the Phorbol Ester Receptor
of Protein Kinase-C. J . Am. Chem. Soc. 1992, 114, 1059-1070.
(43) Lee, L. S.; Weinstein, I. B. Tumor-Promoting Phorbol Esters
Inhibit Binding of Epidermal Growth Factor to Cellular Recep-
tors. Science 1978, 202, 313-315.
(44) Dr. Edward Sausville, NCI, personal communication.
(45) Boyd, M. R.; Paull, K. D. Some Practical Considerations and
Applications of the National Cancer Institute In Vitro Anticancer
Drug Discovery Screen. Drug Dev. Res. 1995, 34, 91-109.
(46) Paull, K. D.; Shoemaker, R. H.; Hodes, L.; Monks, A.; Scudiero,
D. A.; Rubinstein, L.; Plowman, J .; Boyd, M. R. Display and
Analysis of Patterns of Differential Activity of Drugs Against
Human Tumor Cell Lines: Development of Mean Graph and
COMPARE Algorithm. J . Natl. Cancer Inst. 1989, 81, 1088-
1092.
(51) Han, Z. T.; Zhu, X. X.; Yang, R. Y.; Sun, J . Z.; Tian, G. F.; Liu,
X. J .; Cao, G. S.; Newmark, H. L.; Conney, A. H.; Chang, R. L.
Effect of Intravenous Infusions of 12-O-Tetradecanoylphorbol-
13-Acetate (TPA) in Patients with Myelocytic Leukemia: Pre-
liminary Studies on Therapeutic Efficacy and Toxicity. Proc.
Natl. Acad. Sci. U.S.A. 1998, 95, 5357-5361.
(52) Han, T. H.; Tong, Y. K.; He, L. M.; Zhang, Y.; Sun, J . Z.; Tian,
Y. W.; Zhang H.; Cui, Y. L.; Newmark, H. L.; Conney, A. H.;
Chang, R. L. 12-O-Tetradecanoylphorbol-13-Acetate (TPA)-
Induced Increase in Depressed White Blood Cell Counts in
Patients Treated with Cytotoxic Cancer Chemotherapeutic
Drugs. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 5362-5365.
(53) Szallasi, Z.; Krsmanovic, L.; Blumberg, P. M. Nonpromoting 12-
Deoxyphorbol 13-Esters Inhibit Phorbol 12-Myristate 13-Acetate
Induced Tumor Promotion in CD-1 Mouse Skin. Cancer Res.
1993, 53, 2507-2512.
(54) Szallasi, Z.; Denning, M. F.; Smith, C. B.; Dlugosz, A. A.; Yuspa,
S. H.; Pettit, G. R.; Blumberg, P. M. Bryostatin 1 Protects Protein
Kinase C-Delta from Down-Regulation in Mouse Keratinocytes
in Parallel with its Inhibition of Phorbol Ester-Induced Dif-
ferentiation. Mol. Pharmacol. 1994, 46, 840-850.
(55) Blumberg, P. M.; Acs, P.; Bhattacharyya, D. K.; Lorenzo, P. S.
In Signal Transduction and Cell Cycle Inhibitors; Gutkind, J .
S., Ed.; Humana Press: Totawa, NJ , in press.
(56) Molecular Simulation Inc. (http://www.msi.com/).
(57) Sadowski, J .; Gasteiger, J .; Klebe, G. Comparison of Automatic
3-Dimensional Model Builders Using 639 X-ray Structures. J .
Chem. Inf. Comput. Sci. 1994, 34, 1000-1008.
(58) Giesen, D. G.; Hawkins, G. D.; Liotard, D. A.; Cramer, C. J . A
Universal Model for the Quantum Mechanical Calculation of
Free Energies of Solvation in Non-Aqueous Solvents. Theor.
Chem. Acc. 1997, 98, 85-109.
(59) Gasteiger, J .; Marsili, M. Iterative Partial Equalization of Orbital
Electronegativity. Rapid Access to Atomic Partial Charges.
Tetrahedron 1980, 36, 3219-3288.
(60) Kazanietz, M. G.; Areces, 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.
(61) Bal, D. S.; Childers, W. E.; Pinnick, H. W. Oxidation of
R,â-Unsaturated Aldehydes. Tetrahedron 1981, 37, 2091-2096.
(62) Hillis, L. R.; Ronald, R. C. Total Synthesis of (-)-Grahamimycin.
J . Org. Chem. 1985, 50, 470-473.
(63) Baker, B. R.; Schaub, R. Puromycin. Synthetic Studies. XIII.
Synthesis of 6-Dimethylamino-9-(3′-Amino-3′-Deoxy-â-D-Ara-
binofuranosyl)Purine. J . Am. Chem. Soc. 1955, 77, 5900-5905.
(64) Lee, J .; Marquez, V. E.; Lewin, N. E.; Blumberg, P. M. Synthesis
of Two Rigid Diacylglycerol Analogues Having a 1,7-Dioxaspiro-
[4.4]nonane bis-γ-Butyrolactone Skeleton. 4. Synlett 1994,
206-208.
(47) Gulakowski, R. J .; McMahon, J . B.; Buckheit, R. W.; Gustafson,
K. R.; Boyd, M. R. Antireplicative and Anticytopathic Activities
of Prostratin, a Non-Tumor-Promoting Phorbol Ester, Against
Human Immunodeficiency Virus (HIV). Antivir. Res. 1997, 33,
87-97.
(48) The anticancer screen data on prostratin can be found on the
NCI web page by submitting a query at http://dtp.nci.nih.gov/
docs/dtp search.html.
(49) Caponigro, F.; French, R. C.; Kaye, S. B. Protein kinase C: A
Worthwhile Target for Anticancer Drugs? Anticancer Drug 1997,
8, 26-33.
(50) Hu, H. Recent Discovery and Development of Selective Protein
Kinase C Inhibitors. Drug Discovery Today 1996, 1, 438-447.
J M9904607