Structure-activity studies of cerulenin analogues as protein palmitoylation inhibitors

Article abstract of DOI:10.1021/jm980591s

Activation of ras oncogenes occurs in a high percentage of tumors, making the enzymes involved in the posttranslational processing of their encoded proteins (p21s) attractive targets for the development of new drugs. Although most effort has focused on farnesyl transferase, which catalyzes the first processing step, attachment of palmitate to p21 is required for optimal transformation by H-ras and N-ras. We have demonstrated that the natural product cerulenin ([2R,3S]-2,3-epoxy-4-oxo-7,10-trans,trans-dodecadienamide) inhibits the palmitoylation of H-ras-and N-ras-encoded p21s in parallel with inhibition of cell proliferation. More than 30 analogues of cerulenin, both aromatic and aliphatic, with various chain lengths and amide substitutions, have been synthesized for use in SAR studies. Studies on the inhibition of T24 cell proliferation indicate that the α-keto-epoxy moiety is critical for cytotoxicity, while alkyl chain length had only modest effects on potency. Several compounds inhibited the incorporation of [³H]palmitate into p21 in intact T24 cells, with the unsubstituted carboxamides being more active than N,N-dimethyl compounds. In contrast to the effects on palmitoylation, the only compounds which inhibited fatty acid synthase contained alkyl side chains of 12 carbons or fewer. Regression analyses indicated that inhibition of palmitoylation is more closely related to inhibition of proliferation than is inhibition of fatty acid synthase. Further characterization of the molecular pharmacology of these and analogous compounds may define a new class of drugs with antitumor activity.

Full text of DOI:10.1021/jm980591s

4932
J . Med. Chem. 1999, 42, 4932-4941
Articles
Str u ctu r e-Activity Stu d ies of Cer u len in An a logu es a s P r otein P a lm itoyla tion
In h ibitor s
David S. Lawrence, J ack T. Zilfou, and Charles D. Smith*
Department of Pharmacology, Pennsylvania State University, Hershey, Pennsylvania 17033
Received J uly 26, 1999
Activation of ras oncogenes occurs in a high percentage of tumors, making the enzymes involved
in the posttranslational processing of their encoded proteins (p21s) attractive targets for the
development of new drugs. Although most effort has focused on farnesyl transferase, which
catalyzes the first processing step, attachment of palmitate to p21 is required for optimal
transformation by H-ras and N-ras. We have demonstrated that the natural product cerulenin
([2R,3S]-2,3-epoxy-4-oxo-7,10-trans,trans-dodecadienamide) inhibits the palmitoylation of H-ras-
and N-ras-encoded p21s in parallel with inhibition of cell proliferation. More than 30 analogues
of cerulenin, both aromatic and aliphatic, with various chain lengths and amide substitutions,
have been synthesized for use in SAR studies. Studies on the inhibition of T24 cell proliferation
indicate that the R-keto-epoxy moiety is critical for cytotoxicity, while alkyl chain length had
only modest effects on potency. Several compounds inhibited the incorporation of [3H]palmitate
into p21 in intact T24 cells, with the unsubstituted carboxamides being more active than N,N-
dimethyl compounds. In contrast to the effects on palmitoylation, the only compounds which
inhibited fatty acid synthase contained alkyl side chains of 12 carbons or fewer. Regression
analyses indicated that inhibition of palmitoylation is more closely related to inhibition of
proliferation than is inhibition of fatty acid synthase. Further characterization of the molecular
pharmacology of these and analogous compounds may define a new class of drugs with
antitumor activity.
In tr od u ction
transferase. H-ras- and N-ras-encoded p21s are then
palmitoylated by an enzyme which forms a thioester
linkage between nearby cysteine residues and the C-16
chain of palmitate, presumably via palmitoyl-CoA. This
palmitoylation has been shown to be necessary for
anchoring the H-Ras protein to the cell membrane and
expression of the full transforming activity of the
oncogene.9,10 With the K-Ras 2B protein, the cell mem-
brane anchoring function of the palmitate group is
performed by a series of lysine residues, although the
nature of this interaction is not yet defined. Because
these posttranslational modification steps are crucial to
the transforming activity of the p21s, much effort has
gone into characterizing the enzymes involved, espe-
cially with regard to the development of inhibitors. Most
of the attention has focused on the first step catalyzed
by farnesyl protein transferase (FPTase), which has
been isolated,11 cloned,12 and used in screening assays
to identify inhibitors.13,14 Recently, it has been found
that geranylgeranylation of K-ras-encoded p21 can occur
in cells treated with FPTase inhibitors, thus circum-
venting the effects of the inhibitor.15-17 Efforts aimed
at blocking the endoprotease enzyme involved in the
second modification step have also increased.18 Although
the carboxymethyltransferase has been identified,19 no
inhibitors have been described in the literature.
The ras oncogenes play critical roles in the develop-
ment of a high percentage of human cancers. The three
human ras genes (H-ras, N-ras, and K-ras) encode low
molecular weight G-proteins, termed p21s, which act as
signal transducers between extracellular growth factors
and the nucleus in both normal and transformed
cells.1-3 While mutations in the ras oncogenes are
primarily responsible for activation of p21s, overexpres-
sion of the proteins can also promote uncontrolled cell
growth. Consequently, much attention has been focused
on the biochemical and pharmacological properties of
the ras oncogenes and the p21s they encode with the
hope that development of inhibitors of p21 function may
produce clinically useful antitumor drugs.4-6
p21 proteins undergo a series of posttranslational
modifications that result in their anchoring in the
plasma membrane, which is crucial for the transforming
activity of the oncoprotein.7-10 As shown in Figure 1,
the first modification of the H-ras-encoded p21 attaches
a farnesyl moiety to cysteine 186, which is the fourth
amino acid residue from the C-terminus of the protein.
After farnesylation, the three C-terminal amino acid
residues are cleaved by an endoprotease, and the newly
formed C-terminus is esterified by a carboxymethyl
* Address for correspondence: Pennsylvania State University Col-
lege of Medicine, Department of Pharmacology, H078, 500 University
Drive, Hershey, PA 17033-2390. Telephone: (717) 531-1672. E-mail:
cdsmith@psu.edu.
The final processing step for N- and H-Ras oncopro-
teins involves the formation of one or two thioesters with
palmitate, respectively. This palmitoylation is different
10.1021/jm980591s CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/21/1999

Cerulenin Analogues as Protein Palmitoylation Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 24 4933
Sch em e 1. Synthesis of Cerulenin Analoguesa
a
Reagents: (a) 2 equiv of EtMgBr, THF, 0 °C; (b) RCHO, 0 °C
f RT; (c) H2SO4(aq) or satd NH4Cl(aq); (d) H2(g), Lindlar’s
catalyst, EtOAc (or THF), RT; (e) 35 °C; (f) NaOCl(aq), p-dioxane
(or DMF/Et2O - 1:1), 0 °C f RT; (g) NH3(aq), MeOH, 0 °C; (h)
HN(CH3)2, THF/MeOH, 0 °C f RT; (i) pyr2‚Cr2O7, CH2Cl2, RT or
TPAP, NMO, CH2Cl2, RT.
Furthermore, exogenous fatty acids did not protect
against cerulenin-induced cytotoxicity, and metabolic
labeling with [3H]palmitate followed by immunoprecipi-
tation of p21 demonstrated that cerulenin inhibits the
acylation reaction.26,27 This is consistent with a previous
report of the ability of cerulenin to inhibit the acylation
of viral surface glycoproteins expressed on chicken
embryo fibroblasts.28
F igu r e 1. Posttranslational processing of Ras proteins.
Because of the activity of cerulenin and the potential
utility of selective inhibitors of p21 palmitoylation as
antitumor agents, we have synthesized analogues of
cerulenin for studies of the effects on both fatty acid
synthase and p21 palmitoylation. Data presented here
demonstrate that the antiproliferative effect of these
compounds is more closely related to inhibition of p21
palmitoylation than inhibition of fatty acid synthase.
F igu r e 2. Open and cyclized form of cerulenin.
Ch em istr y
from the farnesylation step in that it is reversible. While
the farnesyl group is permanently attached, the palmi-
tate group is dynamically added and removed with a
turnover half-life of 20 min, compared to the 24 h half-
life of the p21 protein in T15 cells.20 The cleavage of
the palmitate moiety is catalyzed by thioesterases that
are well-studied;21-23 however, the enzymology of the
palmitate attachment is not yet characterized. Further-
more, no selective inhibitors of p21 palmitoylation have
yet been described in the literature.
The natural product cerulenin ([2R,3S]-2,3-epoxy-4-
oxo-7,10-trans,trans-dodecadienamide) has long been
known to inhibit fatty acid synthesis by alkylation and
consequent inactivation of â-ketoacyl acyl-carrier pro-
tein synthase, a component of the fatty acid synthase
complex,24 and recent studies on its apoptic and anti-
tumor activities have assumed this mechanism of ac-
tion.25 However, we have recently shown that cerulenin
inhibits the proliferation of ras-transformed NIH/3T3
fibroblasts, T24 bladder carcinoma cells which over-
express H-ras, and MCF-7 cells which overexpress N-ras
at doses lower than those required to inhibit the
proliferation of non- or K-ras-transformed cells.26,27
Cerulenin has been studied as an antifungal and
antibacterial agent for nearly 30 years. In protic media
(such as a cellular environment), cerulenin exists in an
equilibrium between two formssopen-chain and cyclized
hydroxylactam (Figure 2). Several groups have de-
scribed the total synthesis of this and related com-
pounds, both as racemic mixtures and as stereospecific
products.29-31 Cerulenin can be considered to be com-
prised of an R-keto-epoxycarboxamide with an octadi-
enyl side chain. To explore the effects of various side
chain lengths, as well as the functionalities surrounding
the epoxide moiety, we chose to follow the general
synthetic methodology of J akubowski et al.32 This work
describes the regioselective syntheses of not only ceru-
lenin and tetrahydrocerulenin but also their N,N-
dimethyl derivatives (Scheme 1). We sought to use these
substituted amides to explore the importance of the
cyclized hydroxylactam isomer of the epoxycarboxamide
for inhibiting fatty acid synthase and p21 palmitoyla-
tion. Although the majority of the procedures described
by J akubowski et al. proceeded with little difficulty, the
critical step of epoxide formation from the butenolide

4934 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 24
Sch em e 2. Synthesis of “BOCA” Compoundsa
Lawrence et al.
Ta ble 1. Biological Activities of Cerulenin Analogues
IC50 for
inhibition of T24
cell proliferation
(µM)a
inhibition of
fatty acid
inhibition of
p21 palmitoylation
(% inhibited)c
synthase
compound
(% inhibited)b
cerulenin
2
17 ( 4
15 ( 6
29 ( 10
>240
96 ( 3
85 ( 1
3 ( 2
75 ( 13
40 ( 8
4 ( 2
0 ( 0
0 ( 0
5 ( 2
4 ( 2
5 ( 2
6 ( 0
3 ( 1
10 ( 5
0 ( 0
74 ( 11
92 ( 6
44 ( 24
0 ( 0
2 ( 1
0 ( 0
0 ( 0
0 ( 0
9 ( 6
5 ( 2
0 ( 0
3 ( 1
1 ( 1
2 ( 1
5 ( 2
2 ( 1
0 ( 0
0 ( 0
48 ( 2
22
59
32
33
19
2
35
45
16
27
29
34
40
3
42
63
40
47
67
28
9
2a
5a
5b
5c
5d
5e
5f
6a
6b
6c
6d
6e
6f
7a
7b
7c
7d
7e
7f
8a
8b
8c
8d
8e
8f
>220
>200
>180
140 ( 16
>150
>220
>200
130 ( 0
110 ( 54
190 ( 54
>140
27 ( 11
7 ( 2
20 ( 9
13 ( 2
12 ( 0
17 ( 4
>190
220 ( 0
95 ( 19
69 ( 20
62 ( 25
52 ( 19
140 ( 23
77 ( 31
49 ( 19
40 ( 11
40 ( 16
34 ( 1
a
Reagents: (a) isobutyl chloroformate, N-methyl-morpholine,
DME, -13 °C; (b) NH3(g) or (CH3)2NH, THF; (c) H2O2, NaOH, 55
°C.
failed to give any product in our hands. The more
common method of Nozoe was used instead.33 Although
this method routinely gave only 40-50% yields, the
starting material was easily recovered and could be
recycled in subsequent reactions, thereby enhancing the
overall yield. In addition, the classic oxidation of sec-
ondary alcohols to ketones with the Collins reagent was
improved by using the more recently developed TPAP
reagent.34,35 Since preliminary studies had demon-
strated that tetrahydrocerulenin possesses equivalent
cytotoxic activity to its naturally occurring precursor,
we reasoned that the double bonds on the octadienyl
side chain of cerulenin may be unimportant to the
protein acyl transferase. In fact, since the enzyme’s
function is to add a palmitate group to the farnesylated
p21 protein, it was hypothesized that a saturated 16-
carbon chain might imitate the natural substrate more
effectively than cerulenin. Therefore, tetrahydroceru-
0
0
16
0
0
9a
9b
9c
9d
9e
9f
0
17
10
0
31
21
a
Values represent the mean ( SEM concentration required to
inhibit the proliferation of T24 cells by 50% compared with
untreated control cultures. At least three experiments were
performed with each compound. IC50 values of >140 µM are
b
considered to be essentially nontoxic. Values represent the mean
( SEM percentage inhibition of fatty acid synthase activity in the
presence of 150 µM of the indicated compound. At least two
experiments were performed with each compound. c Values rep-
resent the mean percentage inhibition of p21 palmitoylation
compared with untreated control samples. Cerulenin was used at
5 µg/mL and the mean ( SEM for six experiments. Other
compounds were tested at 10 µg/mL, corresponding to 26-47 µM,
in one or two experiments.
lenin analogues with chain lengths of C8H17 to C17H35
,
possessing a free amide or N,N-dimethyl-substituted
amide, were synthesized in nonoptimized yields (∼2%
from propiolic acid). All intermediates and final com-
pounds gave spectral data that were in agreement with
the literature values for tetrahydrocerulenin and N,N-
dimethyl-tetrahydrocerulenin. Samples of all intermedi-
ates with an epoxide moiety were retained for evalua-
tion in cytotoxicity, fatty acid synthase, and p21
palmitoylation assays.
Aromatic analogues of cerulenin (e.g., 3-(benzoyl)-
oxiranecarboxamides, “BOCA”) were patented in 1978
as inhibitors of lipogenesis.36 Our syntheses of BOCA
(2) and N,N-dimethyl-BOCA (2a ) were based on the
procedures in that patent, but improved upon to some
extent (Scheme 2). Notably, formation of the benzoyl-
acrylamides from commercially available benzoylacrylic
acid was done through the use of a mixed anhydride of
the acid activated with isobutylchloroformate, followed
by quenching with either ammonia or dimethylamine.37
Epoxidation with hydrogen peroxide in basic EtOH gave
BOCA and N,N-dimethyl-BOCA in 21% and 51% yields,
respectively, with confirming spectral data. As a result
of the epoxidation conditions, these aromatic products
both possessed trans epoxides. This was confirmed by
the 1H NMR splitting coefficients of the epoxide protons
(J AB < 4 Hz). Although this synthetic strategy did not
give products in high yields, it is amenable to a variety
of aromatic groups by simple Friedel-Crafts acylation
with maleic anhydride to give the starting benzoyl-
acrylic acid and so may prove useful for the preparation
of additional aromatic analogues of cerulenin.
Resu lts a n d Discu ssion
Cytotoxicity testing of cerulenin and its analogues
was performed using concentrations up to 50 µg/mL
with T24 human bladder carcinoma cells that over-
express H-[Val12]-ras. Table 1 shows the mean IC50
values (in µM) obtained from at least three separate
experiments using the SRB assay.38 Similar results
were obtained with the MTS assays (not shown). No
inhibition of growth was observed with some compounds
at the highest concentration tested, corresponding to
140-240 µM, and these compounds are considered to
be essentially nontoxic. The data demonstrate that the
aromatic compounds BOCA (2) and N,N-dimethyl-
BOCA (2a ) are nearly as cytotoxic as cerulenin, while

Cerulenin Analogues as Protein Palmitoylation Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 24 4935
the aliphatic analogues display a range of toxicities
depending on the chain length and functionalities
surrounding the epoxide moiety. The most toxic ana-
logues are the R-keto-epoxycarboxamides (7a -f), fol-
lowed by their N,N-dimethyl derivatives (9a -f). Their
reduced precursors, possessing R-hydroxyl groups, (6a -f
and 8a -f) are noticeably less toxic, while the parent
epoxylactones (5a -f) are all essentially nontoxic. The
cytotoxicities of the unsubstituted epoxycarboxamides
appear to be largely independent of chain length, e.g.,
the IC50s for 7a -f vary only from 7 to 27 µM, with the
analogue 7b, with a 10-carbon chain (including the
carbonyl carbon), being the most potent. The N,N-
dimethyl analogues 9 show a clearer relationship be-
tween cytotoxicity and increasing chain length. Even so,
the free carboxamide of each chain length was more
cytotoxic than its corresponding N,N-dimethyl analogue.
It is interesting to note, however, that the cytotoxicity
of the cerulenin analogues is fairly structure specific.
For example, the mere possession of an epoxide moiety
is not sufficient to impart a toxic effect, although toxicity
is significantly enhanced by oxidizing the secondary
alcohol to a ketone. Furthermore, the long chain fatty
acid analogues do not exert a purely physical lysogenic
effect on the cells. The cytotoxicities of the epoxidized
compounds were associated with typical morphological
indicators of apoptosis, including cell rounding, nuclear
fragmentation, and membrane blebbing (not shown).
Consequently, the effects of prolonged exposure of cells
to these compounds are irreversible, at least at the
cellular level.
of MCF-7 cells with cerulenin also inhibits cell prolif-
eration and p21 palmitoylation, this compound reduces
the palmitoylation of both H-ras- and N-ras-encoded
27
p21.
Data in Table 1 also demonstrate that several
of the cerulenin analogues are also able to inhibit p21
palmitoylation. For example, compounds 2 and 2a both
inhibited p21 palmitoylation, demonstrating that aro-
matic analogues of cerulenin are also effective. Ad-
ditionally, several compounds of the cis-2,3-epoxy-
alkanoic acid γ-lactone series (compounds 5a , 5b, 5e,
and 5f) inhibit p21palmitoylation nearly as effectively
as cerulenin. Similarly, the longer alkyl N,N-dimethyl-
cis-2,3-epoxy-4-oxoalkanamides (compounds 9e and 9f)
also effectively inhibit p21 palmitoylation. The series
of cis-2,3-epoxy-4-oxoalkanamides (compounds 7a -f)
comprise a family of particularly effective inhibitors of
p21 palmitoylation. The greatest inhibitory activity
among the compounds was demonstrated by Compound
7e, which can be considered to be a palmitoyl analogue
of cerulenin. It is important to again note that this
compound, as with several other cerulenin analogues,
does not inhibit fatty acid synthase. Therefore, the
abilities of these compounds to inhibit tumor cell
proliferation is more closely associated with inhibition
of the palmitoylation of p21, and possibly other proteins,
than with reduction in the biosynthesis of fatty acids.
This is substantiated by regression analysis of the
biological effects of the 12 compounds 7a -f and 9a -f
that demonstrated a correlation coefficient of -0.728 (p
) 0.0048) between the IC50 for cell proliferation and the
inhibition of p21 palmitoylation. In contrast, a correla-
tion coefficient of -0.381 (p ) 0.1987) was calculated
for the association between the IC50 for cell proliferation
and the inhibition of fatty acid synthase.
Because cerulenin and certain of its analogues have
been demonstrated to inhibit fatty acid synthase, the
effects of our compounds on this enzyme were deter-
mined. Results from these determinations are also
indicated in Table 1. To allow comparison of the effica-
cies of various compounds, data indicate the percentage
by which fatty acid synthesis is reduced in samples
treated with 150 µM of the indicated compounds. As
expected, cerulenin and compound 2 very effectively
inhibited the tumor cell fatty acid synthase activity.
Significant inhibition also occurred in samples incu-
bated with compounds 5a , 7a , and 7b , while modest
inhibition occurred in samples incubated with com-
pounds 5b and 7c. Importantly, other compounds that
inhibit cell proliferation had no significant inhibitory
effects on fatty acid synthesis. For example, compounds
7d -f potently inhibit cell proliferation but are es-
sentially inactive against fatty acid synthase. Similarly,
Compounds 8a -f and 9a -f have varying inhibitory
effects on tumor cell proliferation, but do not inhibit
fatty acid synthase. Therefore, the mechanism for the
antiproliferative activities of the cerulenin analogues is
clearly distinguished from inhibition of fatty acid syn-
thesis.
The inhibitory effects of the epoxylactones (5a -f) are
interesting to note. Although not as potent as the keto-
epoxycarboxamides, these compounds do inhibit palmi-
toylation of p21 yet display very low toxicity toward the
cells. Recall that the keto-epoxycarboxamide moiety can
equilibrate in protic media (such as within a cell)
between an open, straight chain and a closed, hydroxy-
lactam configuration. Since the N,N-dimethyl analogues
cannot form the hydroxylactam conformer and have
shown reduced inhibitory behavior toward palmitoyla-
tion, it seems likely that the active form of the keto-
epoxycarboxamide is the hydroxylactam. Furthermore,
there is a structural similarity between the hydroxy-
lactam and the epoxylactone moiety of compounds 5a -
f, which also show some inhibitory activity. We hypoth-
esize that the palmitoylation inhibitory activity of
compounds 7a -f is dependent not only on chain length
but also on their conformation as open straight chains
or closed hydroxylactams.
Other structure-activity relationships have been
explored with cerulenin analogues as inhibitors of fatty
acid synthase39 and HIV protease.40,41 Although the
studies were not systematically performed, several
“desoxy” derivatives of cerulenin, lacking the R-keto
group, were tested as inhibitors of HIV protease. The
relative stereochemistry of the epoxide moiety was
claimed to be important (2S-trans-epoxydodecenoic acid
was significantly more potent than either 2S- or 2R-
cis-epoxydodecenoic acid), while increasing the length
of the hydrocarbon chain decreased activity, and the
The effects of the cerulenin analogues on the palmi-
toylation of p21 were determined, and results are
indicated in Table 1. To allow comparison of the effica-
cies of various compounds, data shown indicate the
percentage by which incorporation of [3H]palmitate into
p21 is reduced in samples treated with 10 µg/mL of the
indicated compounds, with the exception that cerulenin
was used at a dose of 5 µg/mL, which inhibited p21
palmitoylation by approximately 50%. Since treatment

4936 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 24
Lawrence et al.
functionality of the group adjacent to the epoxide also
affected activity. Blumenstein et al.40 also postulated
that the ability to cyclize into the hydroxylactam
conformation may enhance the activity against HIV
protease, although the proline methyl ester-substituted
amide of 2R-cis-epoxydodecenoic acid (which cannot
cyclize) displayed half the activity of cerulenin. The SAR
studies involving inhibition of FAS were more system-
atically performed, but they varied only the chain length
and the presence and position of the double bonds in
the chain. While cerulenin remained the most potent
inhibitor of the group (IC50 ) 4.5 µM), removal or
repositioning of the double bonds or lengthening the
chain (leaving the double bonds intact) resulted in less
potent inhibitors (IC50 ) 250-360 µM and 20-35 µM,
respectively).
These data are consistent with that of Table 1 which
clearly demonstrate that the structural requirements
for cerulenin analogues are target specific. Most impor-
tantly, the size and the saturation of the aliphatic side
chain have markedly different effects on the activities
of the analogues toward fatty acid synthase and p21
palmitoylation. Additionally, the presence of the R-keto
moiety appears to be more critical for interaction with
fatty acid synthase than with p21:palmitoyltransferase.
cells that are not expected to be sensitive to palmitoy-
lation inhibitors. Alternately, the known ability of
unprocessed H-Ras to act as a dominant negative
protein46,47 would also lead to inhibition of proliferation
in such cells. In any event, the present work describes
the first synthetic lead compounds for development of
potential therapeutic agents targeting this important
cellular activity.
Exp er im en ta l Section
Reagents and starting materials were obtained from Aldrich
Chemical (Milwaukee, MN) and used as received unless
otherwise noted. Cerulenin was obtained from Sigma at 95%
purity and stored as a stock solution of 5 mg/mL in EtOH at
4 °C. No noticeable decrease in activity occurred over several
months. Solvents were either purchased as “anhydrous” or
“ACS grade” and stored over 4 Å molecular sieves. “Flash
chromatography” refers to the method of Still et al.42 and
generally used Selecto Scientific silica gel (32-63 µm). Thin
layer chromatography was performed on EM Science silica gel
60 F-254 plates (250 µm thick). Radial chromatography was
performed on a Chromatotron model 7924T (Harrison Re-
search, Palo Alto, CA) using 1, 2, or 4 mm thick silica rotors.
HPLC analysis was performed on an ISCO gradient system
using a Whatman Partisil ODS-2 5 µm analytical column (4.6
× 250 mm). Melting points were determined in an open
capillary on a MelTemp II melting point apparatus and are
uncorrected. Infrared spectra were measured on a Nicolet 250
1
FT-IR; values are expressed in wavenumbers (cm-1). H and
Con clu sion
13C NMR spectra were obtained on a Bruker 300 WB spec-
trometer. Chemical shifts are reported in ppm (δ) using
tetramethylsilane (TMS) as reference, and coupling constants
(J ) are reported in hertz. Mass spectra were obtained on a
Finnigan LCQ ion trap mass spectrometer with an electro-
spray ion source or were from Mass Consortium (San Diego,
CA). Microanalyses were determined by Atlantic Microlab, Inc.
(Norcross, GA). All chemical yields are unoptimized and
generally represent the result of a single experiment.
We have synthesized more than 30 analogues of the
natural product cerulenin, both aromatic and aliphatic,
with varying chain lengths and amide substitutions.
Cytotoxicity and p21 palmitoylation results indicate that
the R-ketoepoxide moiety is important for activity and
that increasing the length of the aliphatic chain up to
16 carbons generally increases potency. Furthermore,
both cytotoxicity and palmitoylation assays show that
the free ketoamideswhich can cyclize to form a hydroxy-
lactamsis more potent that the N,N-dimethylated
amide which cannot cyclize. These structural require-
ments are clearly distinct from those that affect the
abilities of these agents to inhibit fatty acid synthase,
and the antiproliferative effects of these compounds are
more closely related to inhibition of p21 processing.
Because of the importance of p21 function in human
cancer, inhibition of its posttranslational modification
has become a popular therapeutic target. Although
many groups have described inhibitors of the initial
farnesylation step, the palmitoylation step has not been
characterized from either an enzymological or a phar-
macological perspective. We have now shown that
cerulenin and its analogues represent a novel class of
anti-Ras agents since no other compounds have yet been
described as selective palmitoylation inhibitors.
Cell Cu ltu r e a n d Cytotoxicity Assa ys. T24 human
bladder carcinoma cells were obtained from ATCC and grown
in RPMI-1640 with 10% fetal bovine serum and 50 µg/mL
gentamycin sulfate. To test the effects of the compounds on
cell growth, cells were seeded in 96-well tissue culture dishes
at 10% confluency and were allowed to attach and recover for
24 h. Varying concentrations of cerulenin or an analogue were
added to each well, and the cells were allowed to incubate for
an additional 72 h. The number of surviving cells was
determined by staining with sulforhodamine B as described
by Skehan et al.38 or metabolism of MTS (Promega CellTiter
Aqueosus system) following the manufacturer’s instructions.
Similar results were obtained with both of these assays.
F a tty Acid Syn th a se Assa y. Fatty acid synthase was
partially purified from HL-60 human promyelocytic leukemia
cells as described by Royer et al.43 The HL-60 cell lysate was
incubated with varying concentrations of cerulenin or one of
the synthetic compounds of this invention and then assayed
for fatty acid synthase according to Royer et al.43 which
monitors the rate of oxidation of NADPH in the presence of
acetyl-CoA and malonyl-CoA.
p 21 P a lm itoyla tion Assa y. T24 cells were seeded in 60
mm dishes at approximately 25% confluency and allowed to
attach overnight. The cells were then treated with 5 µg/mL of
cerulenin (or 10 µg/mL of cerulenin analogue) for 4 h followed
by a 2 h incubation with [3H]palmitate at 100-200 µCi/mL.
Cell lysates were made using a PBS-TDS buffer (phosphate-
buffered saline containing 1% Triton x-100, 0.5% sodium
deoxycholate, 0.1% sodium dodecyl sulfate, 50 µM phenyl-
methylsulfonylflouride, and 2 µg/mL each of pepstatin, leu-
peptin, and aprotinin), and the protein concentration was
determined for each sample using the Bio-Rad protein assay
(Bio-Rad). Normalized samples (containing 100-300 µg of
protein in different experiments) were combined with 2 µg of
Y13-259 anti-Ras antibody (Santa Cruz Biotechnology) for 1-2
It is recognized that eukaryotic cells contain several
classes of palmitoylated protein.45 While many of these
acylations appear functionally inconsequential, palmi-
toylation of several prenylated or N-myristoylated pro-
teins is required for their proper subcellular localization
and function. Interestingly, most of these known palmi-
toylated proteins are involved in signal transduction,
including several src family tyrosine kinases. Therefore,
the cellular effects of protein palmitoylation inhibitors
may reflect attenuation of signals generated by diverse
proliferative stimuli. This may underlie the activity of
these agents, at higher doses, toward K-ras-transformed

Cerulenin Analogues as Protein Palmitoylation Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 24 4937
h, followed by the addition of 20 µL of Protein G PLUS-Agarose
(Santa Cruz Biotechnology) for approximately 18 h at 40 °C.
The samples were then centrifuged and washed 5 times with
PBS-TDS, and radioactivity in the recovered pellets was
determined using a scintillation counter.
2850 (aliphatic CH), 2225 (CtC), 1660 (CdO). 1H NMR
(CDCl3): δ 5.5 (bs, 1H, OH), 4.52 (t, 1H, J ) 6.6 Hz, H4), 1.77
(t, 2H, J ) 6.6 Hz, H5), 1.27-1.47 (m, 12H, CH2), 0.87 (t, 3H,
J ) 6.7 Hz, CH3).
4-Hyd r oxytr id ec-2-yn oic Acid (3b). White plates (mp
62-63 °C, 59%). IR (KBr): 3400 cm-1 (OH), 2925-2850
(aliphatic CH), 2250 (CtC), 1675 (CdO). 1H NMR (CDCl3): δ
6.65 (bs, 1H, OH), 4.53 (t, 1H, J ) 6.6 Hz, H4), 1.77 (m, 2H,
H5), 1.26 (bs, 14H, CH2), 0.88 (t, 3H, J ) 6.7 Hz, CH3).
4-Hyd r oxyp en ta d ec-2-yn oic Acid (3c). Dodecanal was
vacuum distilled and stored under nitrogen at 0 °C before use.
White powder (mp 71-72 °C, 47%). IR (KBr): 3250 cm-1 (OH),
2925-2850 (aliphatic CH), 2325 (CtC), 1660 (CdO), 1275
3-Ben zoyla cr yla m id e (1) a n d N,N-Dim eth yl-3-ben zoyl-
a cr yla m id e (1a ). 3-Benzoylacrylic acid (Acros Chemical, 5.50
g, 31.2 mmol) was dissolved in dimethoxyethane (DME, 55 mL)
and chilled to -13 °C. After deaeration with nitrogen, isobu-
tylchloroformate (35.9 mmol, 4.90 g, 4.65 mL) was added
dropwise, followed by N-methylmorpholine (3.4 mL). Forma-
tion of the mixed anhydride was monitored by TLC (silica, 25%
EtOAc in hexanes) and, after 1 h, was complete. Gaseous
ammonia was bubbled through the reaction for 5 min and then
allowed to warm to room temperature, and the ammonia
bubbling continued for another 15 min. Alternatively, di-
methylamine (62.4 mmol, 31.2 mL of a 2.0 M solution in THF)
was added, and the reaction was stirred at -13 °C for 15 min
and then allowed to warm to room temperature and stirred
for 1 h. Reactions were quenched with H2O (10 mL), followed
by dilution with another 40 mL water and extraction with
CHCl3 (4 × 40 mL). Combined CHCl3 extracts were washed
with water and brine, then dried over Na2SO4. Evaporation
and vacuum desiccation gave 4.15 g (23.7 mmol, 75.9%) of 1
as a light yellow solid, mp 121-123 °C, or 6.34 g (31.2 mmol,
100%) of 1a as a light yellow solid, mp 57-60 °C. Products
were sufficiently pure by HPLC (C18, 0.5 mL/min, 75/25 MeOH/
H2O, 215 nm) to use in the next step. (1): IR (KBr) 3450 and
3400 (s, NH2), 3325 (aromatic CH), 1675 (s, carbonyl); 1H NMR
(CDCl3) δ 7.95-7.89 (m, 2H, H2′/H6′), 7.62-7.54 (m, 1H, H4′),
7.51-7.42 (m, 2H, H3′/H5′), 7.32 (d, 2H, J AB ) 21 Hz, H2/H3),
5.78 (bs, 2H, NH2). (1a ): IR (KBr) 3325 (aromatic CH), 1675
1
(carbonyl C-O). H NMR (CDCl3): δ 6.65 (bs, 1H, OH), 4.53
(t, 1H, J ) 6.6 Hz, H4), 1.77 (m, 2H, H5), 1.26 (bs, 18H, CH2),
0.88 (t, 3H, J ) 6.7 Hz, CH3).
4-Hyd r oxya lk -2-yn oic Acid . Meth od B (3d -f). Reaction
was run as for 3a , but aldehyde was dissolved in THF. After
aldehyde addition, reaction was stirred at 0 °C for 3 h and
then allowed to warm to room temperature overnight. When
TLC (silica, 25% EtOAc/hexanes) showed no aldehyde remain-
ing, the reaction was worked up by chilling to 0 °C, diluting
with Et2O (50 mL), and quenching with saturated NH4Cl (36
mL). Acidification with 1 N HCl to pH ∼ 2 gave a clear,
biphasic mixture which was separated, the organic phase was
extracted with fresh Et2O (2 × 50 mL), and the combined Et2O
extracts were washed with H2O and brine (100 mL each).
Drying over MgSO4 and evaporation of volatiles gave the crude
product as a yellow oil, which was crystallized from petroleum
ether at -20 °C. Compounds 3d -f were made using this
method.
4-Hyd r oxyh ep ta d ec-2-yn oic Acid (3d ). Tetradecanal was
purchased from Fluka and used as received. White powder (mp
75-77 °C, 30.2%). IR (KBr): 3425 cm-1 (OH), 2925-2850
(aliphatic CH), 2250 (CtC), 1720 (CdO), 1260 (carbonyl C-O).
1H NMR (CDCl3): δ 4.52 (t, 1H, J ) 6.5 Hz, H4), 1.77 (m, 2H,
H5), 1.26 (bs, 22H, CH2), 0.88 (t, 3H, J ) 6.7 Hz, CH3).
4-Hyd r oxyn on a d ec-2-yn oic Acid (3e). Hexadecanal was
synthesized from hexadecanol.43 White powder (mp 76-77 °C,
61.5%). IR (KBr): 3275 cm-1 (OH), 2950-2850 (aliphatic CH),
2250 (CtC), 1670 (CdO), 1275 (carbonyl C-O). 1H NMR
(CDCl3): δ 6.6 (vbs, 1H, OH), 4.52 (t, 1H, J ) 6.6 Hz, H4),
1.76 (m, 2H, H5), 1.44 (m, 2H, H6), 1.26 (bs, 24H, CH2), 0.88
(t, 3H, J ) 5.9 Hz, CH3).
4-Hyd r oxyh en eicos-2-yn oic Acid (3f). Octadecanal was
synthesized from octadecanol.43 White powder (mp 83-85 °C,
46.2%). IR (KBr): 3275 cm-1 (OH), 2950-2850 (aliphatic CH),
2250 (CtC), 1670 (CdO), 1275 (carbonyl C-O). 1H NMR
(CDCl3): δ 4.49 (t, 1H, J ) 6.6 Hz, H4), 1.70 (m, 2H, H5), 1.49
(m, 2H, H6), 1.29 (bs, 28H, CH2), 0.88 (t, 3H, J ) 5.9 Hz, CH3).
Alk en oic Acid γ-La cton es (4a -f). Hydroxy-alkynoic acids
were dissolved in either EtOAc (for compounds 4a -e, 125 mL)
or THF (compound 4f, 125 mL) and were hydrogenated at
atmospheric pressure over Lindlar catalyst as in J akubowski
et al.32 The reaction was monitored by HPLC (C18, CH3CN/
H2O - 65/35, 0.5 mL/min, 220 nm). Starting material was
consumed within 2-3 h, and the mixture was filtered through
Celite to remove the catalyst. Heating the filtrate at 35 °C for
10 min closed the lactone ring. Evaporation of volatiles,
followed by distillation under reduced pressure or flash chro-
matography (silica, 25% EtOAc/hexanes) gave pure product
as a clear oil, which solidified on standing for the longer chain
compounds. Compounds 4a -f were made using this method.
Dod ec-2-en oic Acid (4a ). Clear oil (bp 113 °C @ 0.5 mmHg,
85.5%). IR (CDCl3): 2975-2850 cm-1 (aliphatic CH), 1775 (Cd
O). 1H NMR (CDCl3): δ 7.45 (d, 1H, J AB ) 5.5 Hz, H3), 6.10
(d, 1H, J AB ) 5.8 Hz, H2), 5.04 (t, 1H, J AB ) 5.6 Hz, H4), 1.9-
1.6 (m, 2H, H5), 1.28 (bs, 12H, CH2), 0.88 (t, 3H, J AB ) 6.0 Hz,
CH3).
1
(CdO); H NMR (DMSO-d6) δ 8.03 (d, 2H, J AB ) 7.6 Hz, H2′/
H6′), 7.74 (d, 1H, J AB ) 15.2 Hz, H2), 7.69 (d, 1H, J AB ) 7.3
Hz, H4′), 7.58 (d, 2H, J AB ) 7.57 Hz, H3′/H5′), 7.425 (d, 1H, J AB
) 15.2 Hz, H3), 3.11(s, 3H, NCH3), 2.95 (s, 3H, NCH3′).
3-Ben zoyloxir a n eca r boxa m id e (2) a n d N,N-Dim eth yl-
3-Ben zoyloxir a n eca r boxa m id e (2a ). Acrylamides (1 and
1a , 5.708 mmol) were dissolved in 80% EtOH (50 mL). Next,
30% H2O2 (1 mL) was added, followed by 0.1 N NaOH (1.35
mL), and the reaction was stirred at 50 °C while monitoring
by HPLC (C18, 0.5 mL/min, 75/25 MeOH/H2O, 215 nm).
Additional H2O2 (2 mL) was added over the next 4 h period
while maintaining pH at ∼8 by addition of 0.1 N NaOH. After
reaction was complete, the solvents were evaporated then
vacuum desiccated over P2O5 to remove remaining water.
Crude BOCA (2) was recrystallized from methanol to give 300
mg of a white powder (mp 160-162 °C, 1.57 mmol, 27.5%),
while crude N,N-dimethyl-BOCA (2a ) was recrystallized from
acetone/hexanes to give 643 mg of a white powder (mp 97-
100 °C, 2.93 mmol, 51.4%). (2): IR (KBr) 3425 (bs, asymmetric
NH), 3225 (bs, symmetric NH), 3075-3025 (aromatic CH),
1
1675 (ketone CdO), 1605 (amide CdO); H NMR (DMSO-d6)
δ 8.02 (d, 2H, J AB ) 7.6 Hz, H2′/H6′), 7.75-7.71 (m, 1H, H4′),
7.62-7.57 (m, 2H, H3′/H5′), 4.62 (s, 1H, H3), 3.52 (s, 1H, H2);
13C NMR (DMSO-d6) δ 194.46 (C4), 168.64 (C1), 135.30 (C1′),
133.86 (C4′), 128.71 (C2′/C6′), 128.01 (C3′/C5′), 56.21 (C3), 55.17
(C2); MS m/z 210 (M + NH4). Anal. (C10H9NO3) C: calcd, 62.82;
found, 62.17; H: calcd, 4.74; found, 4.79; N: calcd, 7.33; found,
6.71. (2a ): IR (KBr) 3060-3025 (aromatic CH), 2925 (aliphatic
CH), 1680 (ketone CdO), 1660 (amide CdO); 1H NMR (DMSO-
d6) δ 8.06 (d, 2H, J AB ) 7.6 Hz, H2′/H6′), 7.75-7.70 (m, 1H,
H4′), 7.59 (t, 2H, J ) 7.6 Hz, H3′/H5′), 4.65 (s, 1H, H3), 4.06 (s,
1H, H2), 3.05 (s, 3H, NCH3), 2.89 (s, 3H, NCH3′); 13C NMR
(DMSO-d6) δ 192.90 (C4), 165.48 (C1), 135.17 (C1′), 134.52 (C4′),
129.12 (C2′/C6′), 128.73 (C3′/C5′), 55.89 (C3), 54.10 (C2), 36.79
(NCH3), 35.97 (NCH3′); MS m/z 220 (M + H). Anal. (C12H13
-
NO) C: calcd, 65.74; found, 64.99; H: calcd, 5.98; found, 5.72;
N: calcd, 6.39; found, 6.35.
4-Hyd r oxya lk -2-yn oic Acid s. Meth od A (3a -c). The
magnesium dianion of propiolic acid and the appropriate
aldehyde were condensed as in J akubowski et al.32
4-Hyd r oxyd od ec-2-yn oic Acid (3a ). White plates (mp
67-8 °C, 53.28%). IR (KBr): 3500-3275 cm-1 (OH), 2925-
Tr id ec-2-en oic Acid γ-La cton e (4b). Clear oil (bp 115 °C
@ 0.675 mmHg, 87.2%). IR (KBr): 3100 cm-1 (alkene CH),
2950-2850 (aliphatic CH), 1760 (CdO), 1600 (CdC), 1175
(carbonyl C-O). 1H NMR (CDCl3): δ 7.45 (d, 1H, J AB ) 5.1
Hz, H3), 6.10 (d, 1H, J AB ) 5.0 Hz, H2), 5.04 (t, 1H, J AB ) 5.6

4938 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 24
Lawrence et al.
Hz, H4), 1.65 (m, 2H, H5), 1.27 (bs, 14H, CH2), 0.88 (t, 3H, J AB
) 6.0 Hz, CH3).
cis-2,3-Ep oxyh ep ta d eca n oic Acid γ-La cton e (5d ). Pu-
rification by radial chromatography (silica, 15% EtOAc/hex-
anes) gave product as a white solid (mp 59-60 °C, 44.8%) and
recovered starting material (20.8%). IR (KBr): 2950-2850
cm-1 (aliphatic CH), 1780 (CdO), 1200 (carbonyl C-O). 1H
NMR (CDCl3): δ 4.58 (t, 1H, J AB ) 6.5 Hz, H4), 3.97 (d, 1H,
J AB ) 2.4 Hz, H3), 3.77 (d, 1H, J AB ) 2.2 Hz, H2), 1.67 (t, 2H,
J AB ) 7.0 Hz, H5), 1.26 (bs, 22H, CH2), 0.88 (t, 3H, J AB ) 6.5
Hz, CH3). Anal. (C17H30O3) C, H.
cis-2,3-Ep oxyn on a d eca n oic Acid γ-La cton e (5e). Puri-
fication by radial chromatography (silica, 15% EtOAc/hexanes)
gave product as a white solid (mp 66-67 °C, 40.8%) and
recovered starting material (30.8%). IR (KBr): 2950-2850
cm-1 (aliphatic CH), 1775 (CdO), 1200 (carbonyl C-O). 1H
NMR (CDCl3): δ 4.57 (t, 1H, J AB ) 6.5 Hz, H4), 3.95 (d, 1H,
J AB ) 2.3 Hz, H3), 3.77 (d, 1H, J AB ) 2.1 Hz, H2), 1.67 (t, 2H,
J AB ) 7.7 Hz, H5), 1.26 (bs, 26H, CH2), 0.88 (t, 3H, J AB ) 5.8
Hz, CH3). Anal. (C19H34O3) C, H.
cis-2,3-Ep oxyh en eicosa n oic Acid γ-La cton e (5f). Puri-
fication by radial chromatography (silica, 15% EtOAc/hexanes)
gave product as a white solid (mp 71-73 °C, 30.8%) and
recovered starting material (46.0%). IR (KBr): 2950-2850
cm-1 (aliphatic CH), 1775 (CdO), 1190 (carbonyl C-O). 1H
NMR (CDCl3): δ 4.57 (t, 1H, J AB ) 6.5 Hz, H4), 3.96 (d, 1H,
J AB ) 2.4 Hz, H3), 3.77 (d, 1H, J AB ) 2.2 Hz, H2), 1.67 (m, 2H,
H5), 1.26 (bs, 30H, CH2), 0.88 (t, 3H, J AB ) 6.8 Hz, CH3). Anal.
(C21H38O3) C, H.
cis-2,3-Ep oxy-4-h yd r oxya lk a n a m id es (6a -f). Epoxy lac-
tones were dissolved in methanol (∼2 mL/mmol lactone) and
chilled to 0 °C. Concentrated ammonium hydroxide (to give
10% solution) was added slowly, and the solution was stirred
at 0 °C for 30 min until precipitation occurred and TLC (silica,
5% MeOH in CH2Cl2) showed the reaction to be complete.
Dilution with CH2Cl2 (2× volume of MeOH), drying over Na2-
SO4, evaporation of volatiles, and vacuum desiccation gave
crude products as white powders.
P en ta d ec-2-en oic Acid γ-La cton e (4c). Clear oil that
solidified in the condenser (bp 113 °C @ 0.05 mmHg, mp 35-
37 °C, 87.0%). IR (KBr): 3100 cm-1 (alkene CH), 2950-2850
(aliphatic CH), 1760 (CdO), 1600 (CdC), 1175 (carbonyl C-O).
1H NMR (CDCl3): δ 7.456 (d, 1H, J AB ) 5.5 Hz, H3), 6.10 (d,
1H, J AB ) 5.2 Hz, H2), 5.03 (t, 1H, J AB ) 5.6 Hz, H4), 1.7 (m,
2H, H5), 1.26 (bs, 18H, CH2), 0.88 (t, 3H, J AB ) 6.0 Hz, CH3).
Hep ta d ec-2-en oic Acid γ-La cton e (4d ). Colorless crystals
(mp 35-37 °C, 73.2%). IR (KBr): 3075 cm-1 (alkene CH),
2950-2850 (aliphatic CH), 1740 (CdO), 1600 (CdC), 1175
(carbonyl C-O). 1H NMR (CDCl3): δ 7.45 (d, 1H, J AB ) 5.5
Hz, H3), 6.10 (d, 1H, J AB ) 5.5 Hz, H2), 5.03 (t, 1H, J AB ) 5.6
Hz, H4), 1.7 (m, 2H, H5), 1.26 (bs, 22H, CH2), 0.88 (t, 3H, J AB
) 6.0 Hz, CH3).
Non a d ec-2-en oic Acid γ-La cton e (4e). White powder (mp
54-56 °C, 53.1%). IR (KBr): 2950-2850 cm-1 (aliphatic CH),
1775 (CdO). 1H NMR (acetone-d6): δ 7.73 (d, 1H, J AB ) 5.7
Hz, H3), 6.09 (dd, 1H, 5.7 Hz, H2), 5.10 (t, 1H, J AB ) 5.6 Hz,
H4), 1.76 (m, 2H, H5), 1.60 (m, 2H, H6), 1.29 (bs, 24H, CH2),
0.88 (t, 3H, J AB ) 6.0 Hz, CH3).
Hen eicos-2-en oic Acid γ-La cton e (4f). White powder (mp
61-63 °C, 69.4%). IR (KBr): 2950-2850 cm-1 (aliphatic CH),
1740 (CdO). 1H NMR (acetone-d6): δ 7.74 (d, 1H, J AB ) 5.6
Hz, H3), 6.09 (d, 1H, 5.6 Hz, H2), 5.08 (m, 1H, H4), 1.76 (m,
2H, H5), 1.61 (m, 2H, H6), 1.29 (bs, 28H, CH2), 0.88 (bt, 3H,
CH3).
cis-2,3-Ep oxya lk a n oic Acid γ-La cton es (5a -f). Buteno-
lides (14.2 mmol) were dissolved in either p-dioxane (4a , 4b,
8.5 mL/mmol butenolide) or N,N-DMF/Et2O (4c-f, 1:1, 30 mL/
mmol butenolide), chilled to 0 °C, and treated with sodium
hypochlorite solution (∼4% aqueous, 2 equiv) added dropwise.
Reactions were stirred at 0 °C for 1 h and then allowed to
warm to room temperature over 1 h, while monitoring by TLC
(silica, 15% EtOAc/hexanes). When no further change in the
product to starting material ratio was observed, reaction
mixtures were worked up as follows: p-dioxane reactions were
partitioned between Et2O and 5% Na2SO3, the aqueous layers
were extracted with fresh Et2O (2×), and the combined Et2O
extracts were washed with brine, dried with MgSO4, and
evaporated to colorless oils. The DMF/Et2O reactions were
diluted with Et2O, the layers were separated, and the organic
layers were washed with 10% NaSO3, 5% NaHCO3, H2O, and
brine (500 mL each), followed by drying over Na2SO4 and
evaporation of volatiles.
cis-2,3-Ep oxy-4-h yd r oxyd od eca n a m id e (6a ). Purifica-
tion by radial chromatography (silica, 5% MeOH in CH2Cl2)
gave a white powder (mp 108-109 °C, 67.9%). IR (KBr):
3450-3150 cm-1 (NH/OH stretches), 2950-2850 (aliphatic
CH), 1690 (CdO), 1460 (NH bend), 1340 (CN). 1H NMR
(acetone-d6): δ 6.86 (bs, 1H, NH), 6.73 (bs, 1H, NH′), 4.04 (d,
1H, J AB ) 5.2 Hz, OH), 3.53-3.48 (m, 1H, H4), 3.41 (d, 1H,
J AB ) 4.5 Hz, H2), 3.00 (dd, 1H, J AB ) 4.5 Hz, H3), 1.64-1.52
(m, 2H, H5), 1.30 (bs, 12H, CH2), 0.88 (t, 3H, J AB ) 6.7 Hz,
CH3). Anal. (C12H23NO) C, H, N.
cis-2,3-Epoxy-4-h ydr oxytr idecan am ide (6b). Crude white
semisolid was recrystallized from CHCl3/hexanes to give a
white powder (mp 101-102 °C, 21.6%). IR (KBr): 3500-3150
cm-1 (NH/OH stretches), 3000-2800 (aliphatic CH), 1700 (Cd
O). 1H NMR (CDCl3): δ 6.20 (bs, 1H, NH), 6.11 (bs, 1H, NH′),
3.545 (d, 1H, J AB ) 4.5 Hz, H2), 3.46 (q, 1H, H4), 3.125 (q, 1H,
cis-2,3-Ep oxyd od eca n oic Acid γ-La cton e (5a ). Crude
product was purified by flash chromatography (silica, 15%
EtOAc/hexanes) to give product as a colorless oil (47.4%) as
well as recovered starting material (15.8%). IR (neat): 2975-
2850 cm-1 (aliphatic CH), 1775 (CdO). 1H NMR (CDCl3): δ
4.57 (t, 1H, J AB ) 6.5 Hz, H4), 3.97 (d, 1H, J AB ) 2.4 Hz, H3),
3.77 (d, 1 H, J AB ) 2.0 Hz, H2), 1.66 (t, 2H, J AB ) 7.6 Hz, H5),
1.27 (bs, 12H, CH2), 0.88 (t, 3H, J AB ) 5.9 Hz, CH3). Anal.
(C12H20O3) C, H.
cis-2,3-Ep oxytr id eca n oic Acid γ-La cton e (5b). Purifica-
tion by radial chromatography (silica, 25% EtOAc/hexanes)
gave product as a white solid (mp 40-42 °C, 2.15 mmol, 45.2%)
and recovered starting material (25.1%). IR (KBr): 3000-2800
cm-1 (aliphatic CH), 1780 (CdO), 1200 (carbonyl C-O). 1H
NMR (CDCl3): δ 4.57 (t, 1H, J AB ) 6.5 Hz, H4), 3.97 (d, 1H,
J AB ) 2.3 Hz, H3), 3.77 (d, 1H, J AB ) 2.1 Hz, H2), 1.67 (t, 2H,
J AB ) 7.7 Hz, H5), 1.27 (bs, 14H, CH2), 0.88 (t, 3H, J AB ) 5.9
Hz, CH3). Anal. (C13H22O3) C, H.
H3), 1.68 (t, 2H, H5), 1.27 (bs, 14H, CH2), 0.87 (t, 3H, J AB
)
6.8 Hz, CH3). Anal. (C13H25NO) C: calcd, 64.16; found, 65.49;
H: calcd, 10.35; found, 10.08; N: calcd, 5.76; found, 2.28.
cis-2,3-Ep oxy-4-h yd r oxyp en ta d eca n a m id e (6c). Purifi-
cation by radial chromatography (silica, 5% MeOH in CH2-
Cl2) gave a white powder (mp 103-105 °C, 29.7%). IR (KBr):
3500-3100 cm-1 (NH/OH stretches), 3000-2800 (aliphatic
CH), 1700 (CdO). 1H NMR (CDCl3): δ 6.18 (bs, 1H, NH), 5.99
(bs, 1H, NH′), 3.56 (d, 1H, J AB ) 4.5 Hz, H2), 3.45 (q, 1H, H4),
3.13 (m, 1H, H3), 1.68 (m, 2H, H5), 1.26 (bs, 18H, CH2), 0.88
(t, 3H, J AB ) 6.8 Hz, CH3). Anal. (C15H29NO) C: calcd, 66.38;
found, 65.95; H: calcd, 10.77; found, 10.35; N: calcd, 5.16;
found, 4.72.
cis-2,3-Ep oxy-4-h yd r oxyh ep ta d eca n a m id e (6d ). Puri-
fication by radial chromatography (silica, 5% MeOH in CH2-
Cl2) gave a white powder (mp 106-108 °C, 29.1%). IR (KBr):
3450-3175 cm-1 (NH/OH stretches), 2950-2850 (aliphatic
CH), 1680 (CdO). 1H NMR (CDCl3): δ 6.15 (bs, 1H, NH), 5.78
(bs, 1H, NH′), 4.93 (bs, 1H, OH), 3.56 (d, 1H, J AB ) 4.8 Hz,
H2), 3.47 (m, 1H, H4), 3.14 (m, 1 H, H3), 1.67 (m, 2H, H5), 1.26
cis-2,3-Ep oxyp en ta d eca n oic Acid γ-La cton e (5c). Pu-
rification by radial chromatography (silica, 15% EtOAc/hex-
anes) gave product as a white solid (mp 46-48 °C, 40.5%) and
recovered starting material (40.4%). IR (KBr): 3000-2800
cm-1 (aliphatic CH), 1800 (CdO), 1200 (carbonyl C-O), 850
1
(asym epoxide C-C). H NMR (CDCl3): δ 4.57 (t, 1H, J AB
6.5 Hz, H4), 3.97(d, 1 H, J AB ) 2.4 Hz, H3), 3.77 (d, 1H, J AB
)
)
2.1 Hz, H2), 1.65 (m, 2H, H5), 1.26 (bs, 18H, CH2), 0.88 (t, 3H,
J AB ) 6.0 Hz, CH3). Anal. (C15H26O3) C: calcd, 70.83; found,
70.26; H: calcd, 10.30; found, 10.24.
(bs, 22H, CH2), 0.88 (t, 3H, J AB ) 6.7 Hz, CH3). Anal. (C17H33
NO‚C3H6O) C, H, N.
-

Cerulenin Analogues as Protein Palmitoylation Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 24 4939
cis-2,3-Ep oxy-4-h yd r oxyn on a d eca n a m id e (6e). Purifi-
cation by radial chromatography (silica, 5% MeOH in CH2-
Cl2) gave a white powder (mp 110-112 °C, 73.8%). IR (KBr):
3450-3400 cm-1 (NH/OH stretches), 2950-2850 (aliphatic
CH), 1700 (CdO). 1H NMR (CDCl3): δ 6.75 (bs, 1H, NH), 6.55
(bs, 1H, NH′), 3.86 (bs, 1H, OH), 3.53 (m, 1H, H4), 3.40 (d, 1H,
J AB ) 4.3 Hz, H2), 3.00 (m, 1H, H3), 1.56 (m, 2H, H5), 1.30 (bs,
26H, CH2), 0.88 (t, 3H, J AB ) 6.7 Hz, CH3). Anal. (C19H37NO)
C: calcd, 69.68; found, 67.57; H: calcd, 11.39; found, 11.08;
N: calcd, 4.28; found, 3.96.
(M + 1). Anal. (C19H35NO) C: calcd, 70.11; found, 67.64; H:
calcd, 10.84; found, 10.64; N: calcd, 4.30; found, 3.99.
cis-2,3-Ep oxy-4-oxoh en eicosa n a m id e (7f). Hydroxya-
mide 6f (50 mg, 0.141 mmol) was oxidized according to the
procedures of Griffith et al.34,35 by dissolving in 10% CH3CN
in CH2Cl2 (40 mL). After addition of 4 Å molecular sieves
(powdered, 70.5 mg) and N-methyl morpholine oxide (0.212
mmol, 24.8 mg), the mixture was stirred for 10 min at room
temperature. The reaction was initiated by addition of [tetra-
(propyl)ammonium]perruthenate (5 mol %, 7.05 µmol, 2.5 mg).
After the mixture was stirred for 23 h at room temperature,
TLC (silica, 5% MeOH/CH2Cl2) showed reaction to be ∼50%
complete. Another 70 mg of powdered molecular sieves were
added, and the reaction stirred another 71 h. TLC showed that
reaction did not proceed to completion (ca. 60%). Filtration
through a plug of silica with CH2Cl2 f CH2Cl2/Et2O (1:1),
followed by flash chromatography (silica, CH2Cl2/Et2O - 1:1),
gave 29 mg of white solid (mp 99-100 °C, 0.082 mmol, 58%).
IR (KBr): 3400 cm-1 (asym NH stretch), 3150 (sym NH),
2950-2850 (aliphatic CH), 1725 (ketone CdO), 1675 (amide
CdO). 1H NMR (CDCl3): δ 6.31 (bs, 1H, NH), 5.45 (bs, 1H,
NH′), 3.88 (d, 1H, J AB ) 5.2 Hz, H3), 3.73 (d, 1H, J AB ) 5.2
Hz, H2), 2.57 (m, 2H, H5), 1.59 (bs, 2H, H6), 1.25 (bs, 28H, CH2),
0.88 (t, 3H, J AB ) 6.0 Hz, CH3). 13C NMR (CDCl3): δ 202.91
(C4), 167.56 (C1), 58.55 (C3), 55.56 (C2), 41.37 (C5), 32.15
(C19), 29.91, 29.80, 29.58, 29.49, 29.26 (C7-C18), 23.31 (C6),
22.91 (C20), 14.34 (C21). Anal. (C21H39NO) C: calcd, 71.34;
found, 70.54; H: calcd, 11.12; found, 10.79; N: calcd, 3.96;
found, 2.95.
N,N-Dim e t h yl-cis-2,3-e p oxy-4-h yd r oxya lk a n a m id e s
(8a -f). Epoxy lactones (2.36 mmol) were dissolved in MeOH
(10 mL) and chilled to 0 °C. Dimethylamine (2 equiv, 2.36 mL
of a 2.0 M solution in THF) was added dropwise, and the
mixture was stirred at 0 °C for 2 h and then allowed to warm
to room temperature overnight. When TLC (silica, 3% MeOH
in CH2Cl2) indicated that the reaction was complete, all
volatiles were removed by rotary evaporation, followed by
vacuum desiccation, to give the crude product as a yellow
semicrystalline solid.
N ,N -D im e t h y l-ci s-2,3-e p o x y -4-h y d r o x y d o d e c a n a -
m id e (8a ). Recrystallization from hexanes gave pure product
as colorless needles (mp 64-65 °C, 89.4%). IR (KBr): 3475
cm-1 (OH stretch), 2950-2850 (aliphatic CH), 1640 (CdO). 1H
NMR (CDCl3): δ 3.85 (bs, 1H, OH), 3.575 (d, 1H, J AB ) 4.1
Hz, H2), 3.19 (s, 3H, NCH3), 3.17 (m, 1H, H4), 3.08 (dd, 1H,
J AB ) 4.1 Hz, H3), 2.99 (s, 3H, NCH3′), 1.67 (t, 2H, J AB ) 7.3
Hz, H5), 1.27 (bs, 12H, CH2), 0.88 (t, 3H, J AB ) 5.9 Hz, CH3).
Anal. (C14H27NO) C, H, N.
cis-2,3-Ep oxy-4-h yd r oxyh en eicosa n a m id e (6f). Purifica-
tion by radial chromatography (silica, 5% MeOH in CH2Cl2)
gave a white powder (mp 111-113 °C, 30.9%). IR (KBr):
3450-3175 cm-1 (NH/OH stretches), 2950-2850 (aliphatic
1
CH), 1680 (CdO), 1470 (carbonyl C-N). H NMR (CDCl3): δ
6.76 (bs, 1H, NH), 6.58 (bs, 1H, NH′), 3.88 (bs, 1H, OH), 3.53
(m, 1H, H4), 3.40 (d, 1H, J AB ) 4.3 Hz, H2), 3.01 (m, 1H, H3),
1.55 (m, 2H, H5), 1.30 (bs, 30H, CH2), 0.88 (t, 3H, J AB ) 6.7
Hz, CH3). Anal. (C21H41NO) C, H, N.
cis-2,3-Ep oxy-4-oxoa lk a n a m id es (7a-f). Hydroxyamides
were oxidized with pyridinium dichromate according to
J akubowski et al.32
cis-2,3-Ep oxy-4-oxod od eca n a m id e (7a , Tetr a h yd r ocer -
u len in ). Purification by radial chromatography (silica, CH2-
Cl2/Et2O - 5:1) gave a white solid (37.4%, 1.9% from propiolic
acid). IR (KBr): 3400 cm-1 (asym NH stretch), 3200 (sym NH),
3000-2800 (aliphatic CH), 1725 (ketone CdO), 1660 (amide
CdO). 1H NMR (CDCl3): δ 6.3 (bs, 1H, NH), 5.55 (bs, 1H, NH′),
3.79 (AB q, 2H, J AB ) 5.3 Hz, H2, H3), 2.56 (m, 2H, H5), 1.6
(m, 2H, H6), 1.27 (bs, 10H, CH2), 0.87 (t, 3H, J AB ) 6.8 Hz,
CH3). Anal. (C12H21NO) C: calcd, 63.41; found, 63.82; H: calcd,
9.31; found, 8.21; N: calcd, 6.16; found, 5.62.
cis-2,3-E p oxy-4-oxot r id eca n a m id e (7b ). White solid
(6.0%). IR (KBr): 3400 cm-1 (asym NH stretch), 3100 (sym
NH), 3000-2800 (aliphatic CH), 1725 (ketone CdO), 1675
(amide CdO). 1H NMR (CDCl3): δ 6.31 (bs, 1H, NH), 5.50 (bs,
1H, NH′), 3.875 (d, 1H, J AB ) 5.2 Hz, H3), 3.73 (d, 1H, J AB
)
5.2 Hz, H2), 2.57 (m, 2H, H5), 1.61 (m, 2H, H6), 1.26 (bs, 12H,
CH2), 0.88 (t, 3H, J AB ) 5.9 Hz, CH3). Anal. (C13H23NO) C, H,
N.
cis-2,3-Ep oxy-4-oxop en ta d eca n a m id e (7c). White solid
(mp 75-80 °C, 29%). IR (KBr): 3500 cm-1 (asym NH stretch),
3150 (sym NH), 2925-2850 (aliphatic CH), 1725 (ketone Cd
O), 1660 (amide CdO). 1H NMR (CDCl3): δ 6.33 (bs, 1H, NH),
5.65 (bs, 1H, NH′), 3.80 (AB q, 2H, J AB ) 5.2 Hz, H2, H3), 2.58
(m, 2H, H5), 1.60 (m, 2H, H6), 1.26 (bs, 16H, CH2), 0.88 (t, 3H,
J AB ) 5.9 Hz, CH3). Anal. (C17H27NO) C: calcd, 66.88; found,
66.46; H: calcd, 10.10; found, 9.81; N: calcd, 5.20; found, 4.85.
N ,N -D im e t h y l-ci s-2,3-e p o x y -4-h y d r o x y t r id e c a n a -
m id e (8b). Purification by radial chromatography (silica, 3%
MeOH in CH2Cl2) gave a white solid (mp 46-48 °C, 66.0%).
IR (KBr): 3475 cm-1 (OH stretch), 2950-2850 (aliphatic CH),
cis-2,3-E p oxy-4-oxoh ep t a d eca n a m id e (7d ). A crude
sample of 6d (235 mg, 0.785 mmol) was oxidized according to
Corey and Schmidt44 with 1.5 equiv of pyr2‚Cr2O7. After 51 h
of stirring at room temperature, TLC (silica, 5% MeOH in CH2-
Cl2) showed the reaction to be complete. Filtration through
Celite with CH2Cl2, followed by filtration through a plug of
silica with CH2Cl2/Et2O (1:1), gave 127 mg of a white solid.
Partially purified product was radially chromatographed
(silica, CH2Cl2/Et2O - 1:1) to give 6 mg of a white solid (0.02
mmol, 3%). IR (KBr): 3400 cm-1 (asym NH stretch), 2950-
2850 (aliphatic CH), 1720 (ketone CdO), 1660 (amide CdO),
1375 (carbonyl CN). 1H NMR (CDCl3): δ 6.33 (bs, 1H, NH),
5.54 (bs, 1H, NH′), 3.74 (AB q, 2H, J AB ) 5.2 Hz, H2, H3), 2.57
(m, 2H, H5), 1.60 (m, 2H, H6), 1.25 (bs, 22H, CH2), 0.88 (t, 3H,
J AB ) 6.0 Hz, CH3). Anal. (C17H31NO) C: calcd, 68.65; found,
68.41; H: calcd, 10.51; found, 9.99; N: calcd, 4.71; found, 4.19.
1
1640 (CdO). H NMR (CDCl3): δ 3.575 (d, 1H, J AB ) 4.1 Hz,
H2), 3.195 (s, 3H, NCH3), 3.17 (m, 1H, H4), 3.08 (dd, 1H, J AB
) 4.1 Hz, H3), 2.99 (s, 3H, NCH3′), 1.67 (m, 2H, H5), 1.26 (bs,
14H, CH2), 0.88 (t, 3H, J AB ) 6.35 Hz, CH3). Anal. (C15H29NO)
C, H, N.
N,N-Dim et h yl-cis-2,3-ep oxy-4-h yd r oxyp en t a d eca n a -
m id e (8c). White solid (mp 54-56 °C, 75.4%). IR (KBr): 3475
cm-1 (OH stretch), 2950-2850 (aliphatic CH), 1640 (CdO). 1H
NMR (CDCl3): δ 3.575 (d, 1H, J AB ) 4.0 Hz, H2), 3.195 (s, 3H,
NCH3), 3.17 (m, 1H, H4), 3.104 (dd, 1H, J AB ) 4.1 Hz, H3),
2.99 (s, 3H, NCH3′), 1.67 (m, 2H, H5), 1.26 (s, 18H, CH2), 0.88
(t, 3H, J AB ) 6.5 Hz, CH3). Anal. (C17H33NO) C: calcd, 68.19;
found, 68.32; H: calcd, 11.11; found, 10.81; N: calcd, 4.68;
found, 4.21.
N,N-Dim et h yl-cis-2,3-ep oxy-4-h yd r oxyh ep t a d eca n a -
m id e (8d ). White solid (mp 61-63 °C, 85.2%). IR (KBr): 3500
cm-1 (OH stretch), 2925-2850 (aliphatic CH), 1640 (CdO). 1H
NMR (CDCl3): δ 3.575 (d, 1H, J AB ) 4.1 Hz, H2), 3.19 (s, 3H,
NCH3), 3.16 (m, 1H, H4), 3.08 (m, 1H, H3), 2.99 (s, 3H, NCH3′),
1.67 (m, 2H, H5), 1.25 (s, 22H, CH2), 0.88 (t, 3H, J AB ) 6.4 Hz,
CH3). Anal. (C19H37NO) C, H, N.
cis-2,3-Ep oxy-4-oxon on a d eca n a m id e (7e). Hydroxya-
mide 6e was oxidized as for 7d above. Purification by flash
chromatography (silica, CH2Cl2/Et2O - 1:1) gave 74 mg of an
off-white solid (mp 98 °C (dec), 0.227 mmol, 48.7%). IR (KBr):
3390 cm-1 (asym NH stretch), 3150 (sym NH), 2950-2850
(aliphatic CH), 1720 (ketone CdO), 1660 (amide CdO). 1H
NMR (CDCl3): δ 6.33 (bs, 1H, NH), 5.63 (bs, 1H, NH′), 3.87
(bs, 1H, H3), 3.72 (bs, 1H, H2), 2.57 (bm, 2H, H5), 1.60 (bm,
2H, H6), 1.25 (bs, 24H, CH2), 0.88 (bt, 3H, CH3). MS: m/e 326.3
N,N -Dim et h yl-cis-2,3-ep oxy-4-h yd r oxyn on a d eca n a -

4940 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 24
Lawrence et al.
m id e (8e). White solid (mp 67-69 °C, 90.2%). IR (KBr): 3475
cm-1 (OH stretch), 2950-2850 (aliphatic C-H), 1640 (CdO),
(m, 2H, H6), 1.25 (bs, 28H, CH2), 0.88 (t, 3H, J AB ) 6.6 Hz,
CH3). 13C NMR (CHCl3): δ 205.58 (C4), 164.34 (C1), 58.14 (C3),
54.72 (C2), 39.87 (C5), 36.29, 35.30 (NCH3), 31.89 (C19), 29.65,
29.43, 29.35, 29.02, (C7-C18), 22.74, 22.66 (C6, C20), 14.09
(C21). Anal. (C23H43NO) C, H, N.
1
1380 (C-N). H NMR (CDCl3): δ 3.87 (bs, 1H, OH), 3.575 (d,
1H, J AB ) 3.8 Hz, H2), 3.19 (s, 3H, NCH3), 3.15 (m, 1H, H4),
3.08 (m, 1H, H3), 2.99 (s, 3H, NCH3′), 1.67 (t, 2H, J AB ) 7.1
Hz, H5), 1.26 (bs, 26H, CH2), 0.88 (t, 3H, J AB ) 5.5 Hz, CH3).
Anal. (C21H41NO) C, H, N.
N,N-Dim et h yl-cis-2,3-ep oxy-4-h yd r oxyh en eicosa n a -
m id e (8f). White solid (mp 73-75 °C, 85.4%). IR (KBr): 3475
cm-1 (OH stretch), 2950-2850 (aliphatic C-H), 1640 (CdO).
1H NMR (CDCl3): δ 3.86 (bs, 1H, OH), 3.57 (d, 1H, J AB ) 4.1
Hz, H2), 3.19 (s, 4H, NCH3 + H4), 3.08 (dd, 1H, J AB ) 4.1 Hz,
H3), 2.99 (s, 3H, NCH3′), 1.67 (m, 2H, H5), 1.25 (bs, 30H, CH2),
0.88 (t, 3H, J AB ) 6.8 Hz, CH3). Anal. (C23H45NO) C: calcd,
72.01; found, 71.87; H: calcd, 11.82; found, 11.14; N: calcd,
3.65; found, 3.35.
N,N-Dim eth yl-cis-2,3-ep oxy-4-oxoa lk a n a m id es (9a -f).
N,N-Dimethyl-hydroxyamides were oxidized as for 7a above.
After the mixture was stirred overnight at room temperature,
TLC (silica, CH2Cl2/Et2O - 1:1) showed reaction to be com-
plete. Workup as for 7a gave a brown oil which was filtered
through Florisil and radially chromatographed to give a pale
yellow oil (36.6%, 2.4% from propiolic acid).
N,N-Dim et h yl-cis-2,3-ep oxy-4-oxod od eca n a m id e (9a ,
N,N-Dim eth yl-tetr a h yd r ocer u len in ). IR (neat): 2950-
2850 cm-1 (aliphatic CH), 1710 (ketone CdO), 1660 (amide
CdO). 1H NMR (CDCl3): δ 3.82 (d, 1H, J AB ) 4.8 Hz, H3), 3.69
(d, 1H, J AB ) 4.9 Hz, H2), 3.13 (s, 3H, NCH3), 2.93 (s, 3H,
NCH3′), 2.535 (m, 2H, H5), 1.56 (m, 2H, H6), 1.26 (bs, 10H,
CH2), 0.88 (t, 3H, J AB ) 5.9 Hz, CH3). Anal. (C14H25NO) C, H,
N.
Ack n ow led gm en t. This work was supported by
NIH grant R01 CA 75248 (C.D.S.) and NRSA fellowship
F32 CA 74549 (D.S.L.) from the National Cancer
Institute.
Refer en ces
(1) Barbacid, M. Ras genes. Annu. Rev. Biochem. 1987, 56, 779-
827.
(2) Lowy, D. R.; Willumsen, B. M. Function and regulation of ras.
Annu. Rev. Biochem. 1993, 62, 851-891.
(3) Khosravi-Far, R.; Der, C. J . The ras signal transduction pathway.
Cancer Metastasis Rev. 1994, 13, 67-89.
(4) Sigal, I. S.; Smith, G. M.; J urnak, F.; Marsico-Ahern, J . D.;
D’Alonzo, J . S.; Scholnick, E. M.; Gibbs, J . B. Molecular ap-
proaches towards an anti-ras drug. Anti-Cancer Drug Des. 2
1987, 107-115.
(5) Gibbs, J . B.; Oliff, A.; Kohl, N. E. Farnesyltransferase inhibi-
tors: ras research yields a potential cancer therapeutic. Cell
1994, 77, 175-178.
(6) Bolton, G. L.; Sebolt-Leopold, J . S.; Hodges, J . C. Ras oncogene
directed approaches in cancer chemotherapy. Ann. Rep. Med.
Chem. 1994, 29, 165-174.
(7) Willumsen, B. M.; Christensen, A. The p21 ras C-terminus is
required for transformation and membrane association. Nature
1984, 310, 583-586.
(8) Hancock, J . F.; Magee, A. I.; Childs, J . E.; Marshall, C. J . All
ras proteins are polyisoprenylated but only some are palmitoy-
lated. Cell 1989, 57, 1167-1177.
(9) Hancock, J . F.; Paterson, H.; Marshall, C. J . A polybasic domain
or palmitoylation is required in addition to the CAAX motif to
localize p21 ras to the plasma membrane. Cell 1990, 6, 133-
139.
N,N-Dim eth yl-cis-2,3-Ep oxy-4-oxotr id eca n a m id e (9b).
Pale yellow oil (52.6%). IR (neat): 2950-2850 cm-1 (aliphatic
CH), 1710 (ketone CdO), 1650 (amide CdO). 1H NMR
(CDCl3): δ 3.76 (AB q, 2H, J AB ) 5.0 Hz, H2/H3), 3.13 (s, 3H,
NCH3), 2.93 (s, 3H, NCH3′), 2.53 (m, 2H, H5), 1.56 (m, 2H, H6),
1.25 (bs, 12H, CH2), 0.87 (t, 3H, J AB ) 6.1 Hz, CH3). 13C NMR
(CDCl3): δ 205.67 (C4), 164.43 (C1), 58.28 (C3), 54.82 (C2),
39.91 (C5), 36.37, 35.35 (NCH3), 31.92 (C11), 29.48, 29.46,
29.32, 29.11 (C7-C10), 22.83, 22.72 (C6, C12), 14.16 (C13).
Anal. (C15H27NO) C, H, N.
(10) Cadwallader, K. A.; Paterson, H.; MacDonald, S. G.; Hancock,
J . F. N-terminally myristoylated ras proteins require palmitoy-
lation or a polybasic domain for plasma membrane localization.
Mol. Cell. Biol. 1994, 14, 4722-4730.
(11) Pompliano, D. L.; Rands, E.; Schaber, M. D.; Mosser, S. D.;
Anthony, N. J .; Gibbs, J . B. Steady-state kinetic mechanism of
Ras farnesyl:protein transferase. Biochemistry 1992, 31, 3800-
3807.
(12) Chen, W.-J .; Andres, D. A.; Goldstein, J . L.; Russell, D. W.;
Brown, M. S. cDNA cloning and expression of the peptide binding
subunit of rat p21 ras farnesyltransferase, the counterpart of
yeast RAM1/DPR1. Cell 1991, 66, 327-334.
N,N-Dim eth yl-cis-2,3-epoxy-4-oxopen tadecan am ide (9c).
Clear oil (52.0%). IR (neat): 2950-2850 cm-1 (aliphatic CH),
1
1710 (ketone CdO), 1660 (amide CdO). H NMR (CDCl3): δ
(13) Kohl, N. E.; Mosser, S. D.; deSolms, S. J .; Giuliani E. A.;
Pompliano, D. L.; Graham, S. L.; Smith, R. L.; Scholnick, E. M.;
Oliff, A.; Gibbs, J . B. Selective inhibition of ras-dependent
transformation by a farnesyltransferase inhibitor. Science 1993,
260, 1934-1937.
(14) J ames, G. L.; Goldstein, J . L.; Brown, M. S.; Rawson, T. E.;
Somers, T. C.; McDowell, R. S.; Crowley, C. W.; Lucas, B. K.;
Levinson, A. D.; Marsters, J . C., J r. Benzodiazepine peptidomi-
metics: potent inhibitors of ras farnesylation in animal cells.
Science 1993, 260, 1937-1942.
(15) J ames, G. L.; Goldstein, J . L.; Brown, M. S. Polylysine and CVIM
sequences of K-RasB dictate specificity of prenylation and
confer resistance to benzodiazepine peptidomimetic in vitro. J .
Biol. Chem. 1995, 270, 6221-6226.
(16) Rowell, C. A.; Kowalczyk, J . J .; Lewis, M. D.; Garcia, A. M. Direct
demonstration of geranylgeranylation and farnesylation of Ki-
ras in vivo. J . Biol. Chem. 1997, 272, 14093-14097.
(17) Whyte, D. B.; Kirschmeier, P.; Hockenberry, T. N.; Nunez-Oliva,
I.; J ames, L.; Catino, J . J .; Bishop, W. R.; Pai, J . K. K- and N-ras
are geranylgeranylated in cells treated with farnesyl protein
transferase inhibitors. J . Biol. Chem. 1997, 272, 14459-14464.
(18) Castro, A. C.; Adams, J .; Behnke, M. L.; Chen, S.; Grenier, L.;
Klunder, J . M.; McCormack, T. A.; Oalmann, C. J .; Plamondon,
L.; Soucy, F.; Wong, H. Abstracts of Papers. 216th National
Meeting of the American Chemical Society, Boston, MA; Ameri-
can Chemical Society: Washington, DC, 1998; MEDI 215.
(19) Clarke, S.; Vogel, J . P.; Deschenes, R. J .; Stock, J . Posttransla-
tional modification of the Ha-ras oncogene protein: Evidence
for a third class of protein carboxy methyltransferases. Proc.
Natl. Acad. Sci. U.S.A. 1988, 85, 4643-4647.
3.75 (AB q, 2H, J AB ) 4.9 Hz, H2/H3), 3.13 (s, 3H, NCH3), 2.93
(s, 3H, NCH3′), 2.53 (m, 2H, H5), 1.56 (m, 2H, H6), 1.25 (bs,
16H, CH2), 0.88 (t, 3H, J AB ) 5.6 Hz, CH3). 13C NMR (CDCl3):
δ 205.65 (C4), 164.44 (C1), 58.26 (C3), 54.81 (C2), 39.91 (C5),
36.36, 35.35 (NCH3), 31.96 (C13), 29.66, 29.52, 29.45, 29.39,
29.11 (C7-C12), 22.83, 22.74 (C6, C14), 14.17 (C15). Anal.
(C17H31NO) C, H, N.
N,N-Dim et h yl-cis-2,3-ep oxy-4-oxoh ep t a d eca n a m id e
(9d ). Clear oil (44.0%). IR (neat): 2950-2850 cm-1 (aliphatic
CH), 1710 (ketone CdO), 1660 (amide CdO). 1H NMR
(CDCl3): δ 3.76 (AB q, 2H, J AB ) 4.9 Hz, H2/H3), 3.13 (s, 3H,
NCH3), 2.93 (s, 3H, NCH3′), 2.53 (m, 2H, H5), 1.56 (m, 2H, H6),
1.25 (bs, 20H, CH2), 0.88 (t, 3H, J AB ) 6.1 Hz, CH3). 13C NMR
(CDCl3): δ 205.78 (C4), 164.53 (C1), 58.36 (C3), 54.93 (C2),
40.07 (C5), 36.49, 35.50 (NCH3), 32.09 (C15), 29.81, 29.78,
29.64, 29.56, 29.52, 29.22 (C7-C14), 22.95, 22.86 (C6, C16),
14.29 (C17). Anal. (C19H35NO) C, H, N.
N,N-Dim eth yl-cis-2,3-epoxy-4-oxon on adecan am ide (9e).
Off-white solid (11%). IR (KBr): 2950-2850 cm-1 (aliphatic
CH), 1720 (ketone CdO), 1645 (amide CdO), 1375 (carbonyl
CN). 1H NMR (CDCl3): δ 3.75 (AB q, 2H, J AB ) 4.5 Hz, H2/
H3), 3.13 (s, 3H, NCH3), 2.93 (s, 3H, NCH3′), 2.53 (m, 2H, H5),
1.56 (m, 2H, H6), 1.25 (bs, 24H, CH2), 0.88 (t, 3H, J AB ) 6.6
Hz, CH3). MS: m/e 354.4 (M + 1). Anal. (C21H39NO) C, H, N.
N,N-Dim eth yl-cis-2,3-epoxy-4-oxoh en eicosan am ide (9f).
Produced by TPAP oxidation (as for 7f) as a white solid (mp
53-55 °C, 93.4%). IR (KBr): 2950-2850 cm-1 (aliphatic CH),
(20) Magee, A. I.; Gutierrez, L.; McKay, I. A.; Marchall, C. J .; Hall,
A. Dynamic fatty acylation of p21 N-ras. EMBO J . 1987, 6,
3353-3357.
(21) Camp, L. A.; Hoffmann, S. L. Purification and properties of a
palmitoyl-protein thioesterase that cleaves palmitate from H-ras.
J . Biol. Chem. 1993, 268, 22566-22574.
1
1725 (ketone CdO), 1650 (amide CdO). H NMR (CDCl3): δ
3.82 (d, 1H, J AB ) 4.9 Hz, H3), 3.70 (d, 1H, J AB ) 4.9 Hz, H2),
3.13 (s, 3H, NCH3), 2.93 (s, 3H, NCH3′), 2.53 (m, 2H, H5), 1.56

Cerulenin Analogues as Protein Palmitoylation Inhibitors
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 24 4941
(35) Griffith, W. P.; Ley, S. V. TPAP: Tetra-n-propylammonium
(22) Camp, L. A.; Verkruyse, L. A.; Afendis, S. J .; Slaughter, C. A.;
Hoffmann, S. L. Molecular cloning and expression of palmitoyl-
protein thioesterase. J . Biol. Chem. 1994, 269, 23212-23219.
perruthenate,
a mild and convenient oxidant for alcohols.
Aldrichim. Acta 1990, 23, 13-19.
(23) Duncan, J . A.; Gilman, A. G.
A cytoplasmic acyl-protein
(36) Carr, J . B.; Durham, H. G. 3-(Benzoyl)oxiranecarboxamides. U.S.
Patent 4,091,221, 1978.
thioesterase that removes palmitate from G protein R subunits
and p21ras. J . Biol. Chem. 1998, 273, 15830-15837.
(24) Omura, S. The antibiotic cerulenin, a novel tool for biochemistry
as an inhibitor of fatty acid synthesis. Bacteriol. Rev. 1976, 40,
681-697.
(25) Pizer, E. S.; Wood, F. D.; Heine, H. S.; Romantsev, F. E.;
Pasternick, G. R.; Kuhajda, F. P. Inhibition of fatty acid
synthesis delays disease progression in a xenograft model of
ovarian cancer. Cancer Res. 1996, 56, 1189-1193.
(26) Smith, C. D.; Zilfou, J . T.; Laing, N.; Lawrence, D. S.; Dexter,
D. W. Inhibition of palmitoylation of ras oncoproteins by ceru-
lenin. Proc. Am. Assoc. Cancer Res. 1980, 40, 118.
(27) Smith, C. D.; Zilfou, J . T.; Laing, N.; Varner, A. S. Cerulenin
inhibits protein palmitoylation and ras function. Submitted,
Cancer Res.
(28) Schlessinger, M. J .; Malfer, C. Cerulenin blocks fatty acid
acylation of glycoproteins and inhibits vesicular stomatitis and
Sinbis virus particle formation. J . Biol. Chem. 1982, 257, 9887-
9890.
(29) Boekman, R. K., J r.; Thomas, E. W. A total synthesis of
DL-cerulenin: a novel fatty acid antibiotic and lipid synthesis
inhibitor. J . Am. Chem. Soc. 1977, 99, 2805-2806.
(30) Mani, N. S.; Townsend, C. A. A concise synthesis of (+)-cerulenin
from a chiral oxiranyllithium. J . Org. Chem. 1997, 62, 636-
640.
(31) Ohri, H.; Emoto, S. Stereoselective synthesis of (+)- and (-)-
tetrahydrocerulenin from D-glucose. The absolute configuration
of cerulenin. Tetrahedron Lett. 1978, 24, 2095-2098.
(32) J akubowski, A. A.; Guziec, F. S., J r.; Sugiera, M.; Tam, C. C.;
Tischler, M. Total synthesis of (()-cerulenin, (()-tetrahydro-
cerulenin, and related compounds. J . Org. Chem. 1982, 47,
1221-1228.
(33) Ohta, T.; Tsuchiyama, H.; Nozoe, S. Total synthesis of (()-
cerulenin and (()-tetrahydrocerulenin. Heterocycles 1986, 24,
1137-1143.
(34) Griffith, W. P. Preparation and use of tetra-n-butylammonium
per-ruthenate (TBAP reagent) and tetra-n-propylammonium
per-ruthenate (TPAP reagent) as new catalytic oxidants for
alcohols J . Chem. Soc., Chem. Commun. 1987, 1625-1627.
(37) Bodansky, M.; Bodansky, A. The Practice of Peptide Synthesis;
Springer-Verlag: New York, 1994; pp 89-90.
(38) Skehan, P.; Stoneng, R.; Scudiero, D.; Monks, A.; McMahon, J .;
Vistica, D.; Warren, J . T.; Bokesh, H.; Kenney, S.; Boyd, M. R.
New colorimetric cytotoxicity assay for anticancer-drug screen-
ing. J . Natl. Cancer Inst. U.S.A. 1990, 82, 1107-1112.
(39) Morisaki, N.; Funabashi, H.; Shimazawa, R.; Furukawa, J .;
Kawaguchi, A.; Okuda, S.; Iwsaki, S. Effect of side-chain
structure on inhibition of yeast fatty-acid synthase by cerulenin
analogues. Int. J . Biochem. 1993, 211, 111-115.
(40) Blumenstein, J . J .; Copeland, T. D.; Oroszlan, S.; Michejda, C.
J . Synthetic non-peptide inhibitors of HIV protease. Biochem.
Biophys. Res. Commun. 1989, 163, 980-987.
(41) Blumenstein, J . J .; Michejda, C. J .; Oroszlan, S.; Copeland, T.
D. U.S. Patent 5,190,969, 1993.
(42) Still, W. C.; Kahn, M.; Mitra, A. Rapid chromatographic tech-
niques for preparative separation with moderate resolution. J .
Org. Chem. 1978, 43, 2923-2924.
(43) Royer, G. P.; Townsend, C. A. Cerulenin compounds for fatty
acid synthesis inhibition. U.S. Patent 5,539,132, 1996.
(44) Corey, E. J .; Schmidt, G. Useful procedures for the oxidation of
alcohols involving pyridinium dichromate in aprotic media.
Tetrahedron Lett. 1979, 399-402.
(45) Resh, M. D. Fatty acylation of proteins: new insights into
membrane targeting of myristoylated and palmitoylated pro-
teins. Biochim. Biophys. Acta 1999, 1451, 1-16.
(46) Gibbs, J . B.; Schaber, M. D.; Schofield, T. L.; Scolnick, E. M.;
Sigal, I. S. Xenopus oocyte germinal-vesicle breakdown induced
by [Val12]Ras is inhibited by a cytosols-localized Ras mutant.
Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 6630-6634.
(47) Lerner, E. C.; Qian, Y.; Blaskovich, M. A.; Fossum, R. D.; Vogt,
A.; Sun, J .; Cox, A. D.; Der, C. J .; Hamilton, A. D.; Sebti, S. M.
Ras CAAX peptidomimetic FTI-277 selectively blocks oncogenic
ras signaling by inducing cytoplasmic accumulation of inactive
ras-raf complexes. J . Biol. Chem. 1995, 260, 26802-26806.
J M980591S