Novel farnesol and geranylgeraniol analogues: A potential new class of anticancer agents directed against protein prenylation

Article abstract of DOI:10.1021/jm9902786

Protein farnesyltransferase (FTase), the enzyme responsible for protein farnesylation, has become a key target for the rational design of cancer cheraotherapeutic agents. Herein it is shown that certain novel prenyl diphosphate analogues are potent inhibitors of mammalian FTase. Furthermore, the alcohol precursors of two of these compounds are able to bloch anchorage- independent growth of ras-transformed cells. While 3-allylfarnesol inhibits protein farnesylation, 3-vinylfarnesol instead leads to abnormal prenylation of proteins with the 3-vinylfarnesyl group. In a similar manner, 3- allylgeranylgeraniol acts as a highly specific inhibitor of protein geranylgeranylation, while 3-vinylgeranylgeraniol restores protein geranylgeranylation in cells. This study indicates that certain prenyl alcohol analogues can act as prenyltransferase inhibitors in situ, via a novel prodrug mechanism. These analogues may prove to be valuable tools for investigating the therapeutic consequences of inhibiting geranylgeranylation relative to farnesylation. Furthermore, the 3-vinyl alcohol analogues can inhibit transformed cell growth through a mechanism not involving prenyltransferase inhibition.

Full text of DOI:10.1021/jm9902786

3800
J . Med. Chem. 1999, 42, 3800-3808
Novel F a r n esol a n d Ger a n ylger a n iol An a logu es: A P oten tia l New Cla ss of
An tica n cer Agen ts Dir ected a ga in st P r otein P r en yla tion
Barbara S. Gibbs,‡ Todd J . Zahn,† YongQi Mu,†,§ J udith S. Sebolt-Leopold,‡ and Richard A. Gibbs*,†
Department of Cell Biology, Parke-Davis Pharmaceutical Research, 2800 Plymouth Road, Ann Arbor, Michigan 48105, and
Department of Pharmaceutical Sciences, College of Pharmacy and Allied Health Professions, Wayne State University,
528 Shapero Hall, Detroit, Michigan 48202
Received J une 4, 1999
Protein farnesyltransferase (FTase), the enzyme responsible for protein farnesylation, has
become a key target for the rational design of cancer chemotherapeutic agents. Herein it is
shown that certain novel prenyl diphosphate analogues are potent inhibitors of mammalian
FTase. Furthermore, the alcohol precursors of two of these compounds are able to block
anchorage-independent growth of ras-transformed cells. While 3-allylfarnesol inhibits protein
farnesylation, 3-vinylfarnesol instead leads to abnormal prenylation of proteins with the
3-vinylfarnesyl group. In a similar manner, 3-allylgeranylgeraniol acts as a highly specific
inhibitor of protein geranylgeranylation, while 3-vinylgeranylgeraniol restores protein gera-
nylgeranylation in cells. This study indicates that certain prenyl alcohol analogues can act as
prenyltransferase inhibitors in situ, via a novel prodrug mechanism. These analogues may
prove to be valuable tools for investigating the therapeutic consequences of inhibiting
geranylgeranylation relative to farnesylation. Furthermore, the 3-vinyl alcohol analogues can
inhibit transformed cell growth through a mechanism not involving prenyltransferase inhibition.
In tr od u ction
The initial evidence that proteins are modified with
a mevalonate pathway intermediate was presented
scarcely over a decade ago. Preliminary biochemical
studies quickly demonstrated that there are three
different protein prenylation motifssfarnesylation, ger-
anylgeranylation, and bis-geranylgeranylation.1 The
first modification is carried out by an enzyme, protein
farnesyltransferase (FTase), which recognizes the CAAX
box (where A ) aliphatic and X ) Ser or Met) at the
carboxyl terminus of the protein substrate and then
attaches the farnesyl group from farnesyl diphosphate
(FPP, 1a ) to the free sulfhydryl of the cysteine residue
(Figure 1). The second, closely related enzyme protein
geranylgeranyltransferase I (GGTase I) attaches a
F igu r e 1. Reactions catalyzed by FTase and GGTase I.
geranylgeranyl moiety from geranylgeranyl diphosphate
(GGPP, 2a ) to a cysteine in a similar CAAX box, where
Significant progress has been made in the development
leucine is the carboxyl terminal residue. The third
of peptide-based FTase inhibitors, and some of these
enzyme, GGTase II, attaches two geranylgeranyl resi-
compounds have shown great promise in vivo as poten-
dues to two cysteine residues at the carboxyl terminus
tial anticancer agents.6 However, less work has been
done on FPP-based FTase inhibitors, and thus less is
known regarding the specificity of FTase for its iso-
prenoid substrate. We have therefore synthesized novel
FPP analogues as probes of the FPP-binding site of
FTase and characterized their interaction with recom-
binant yeast FTase (yFTase). The vinyl analogue 3-
vFPP (1b, Chart 1) was designed as a potential mech-
anism-based inhibitor but was instead a poor alternative
substrate for yeast FTase.7 In contrast, the sterically
encumbered analogue 3-tbFPP (1c) is an exceptionally
poor substrate and a potent competitive inhibitor of this
enzyme.8 The results seen with 3-tbFPP led us to
evaluate it and 3-vFPP against mammalian FTase
(mFTase), the clinically relevant variant of the enzyme.
of Rab proteins.2 Since initial studies have demonstrated
that Ras, a key protein in signal transduction, is
farnesylated, inhibition of FTase has become the subject
of intense research interest. While inhibitors of this
enzyme are thought to operate by multiple mechanisms
of action, in some tumors they block the action of mutant
Ras protein thereby halting the growth of ras-trans-
formed cells. FTase inhibitors are therefore being
developed as potentially novel anticancer drugs.3-5
* To whom correspondence should be addressed. Telephone: 313-
577-3761. Fax: 313-577-2033. E-mail: rag@wizard.pharm.wayne.edu.
‡ Parke-Davis Pharmaceutical Research.
† Wayne State University.
§ Current address: Advanced Medicine, 280 Utah Avenue, South
San Francisco, CA 94080.
10.1021/jm9902786 CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/31/1999

Novel Farnesol and Geranylgeraniol Analogues
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 19 3801
Ch a r t 1
Ta ble 1. Inhibition Constants for FPP and GGPP Analoguesa
Sch em e 1
IC50
IC50
Ki
FTase/
Km
analogue FTase GGTase I
FTase
GGTase FTase krel
3-vFPP
173 ∼100 000 96 (2700)b ∼600
156 0.524
3-tbFPP
3-alFPP
3-etFPP
3-phFPP
3-vGGPP
31
∼50 000 8.0 (310)b ∼1600
189 >100 000 31
215 >100 000
299 >100 000
>530
>460
>340
703 0.0711
715
3 050
3 380
4.3
7.5
3-alGGPP 453
a
Conditions: All inhibition constants and Km values are in
nanomolar amounts. IC50 and Ki values were determined using
recombinant mFTase9 or recombinant mGGTase I in a scintillation
proximity assay as appropriate (see Experimental Procedures). Km
and krel (relative Vmax compared to that obtained with FPP) values
were determined using a continuous spectrofluorimetric assay20,21
(see Experimental Procedures). Under the conditions of the
fluorimetric FTase assay, the Km determined for FPP itself was
b
107 nM. Values in parentheses are those previously determined
for 3-vFPP7 and 3-tbFPP8 with yeast FTase.
pounds inhibited mFTase, albeit not as potently as
3-tbFPP. The three most potent inhibitors of mFTase
(1b, 1c, and 1d ) were further characterized and were
determined to all be competitive inhibitors of the
enzyme versus FPP. The increasing interest in GGTase
I inhibitors as potential cancer chemotherapeutic
agents11-14 and our need to evaluate in vivo selectivity
(vide infra) prompted us to prepare and evaluate two
of the corresponding analogues, 3-vGGPP and 3-alGGPP
(2b and 2d , Scheme 2). Surprisingly, both 2b and 2d
bind more tightly to mFTase than to mammalian
GGTase I (Table 1). This underscores the striking
difference in diphosphate binding selectivity between
these two enzymes that are highly similar in amino acid
sequence9 and in fact share an identical R subunit.2 It
also underscores the highly selective nature of GGTase
I and the difficulty in obtaining either peptide- or
isoprenoid-based inhibitors of this enzyme that do not
also inhibit mFTase.14
Resu lts
The mammalian and yeast variants of FTase are quite
similar, but they differ sharply in their affinity for the
isoprenoid substrate FPP, with mFTase9 binding FPP
30-fold more tightly than yFTase.10 As reported previ-
ously, 3-vFPP (1b) and 3-tbFPP (1c) inhibit yFTase with
Ki values of 2.7 µM7 and 0.31 µM,8 respectively. It is
striking, but perhaps not surprising that 3-vFPP and
3-tbFPP are much more potent inhibitors of mFTase
than yFTase (Table 1). The selectivity observed for
mFTase versus the closely related enzyme mGGTase I
is also noteworthy, and is in accord with the 330-fold
selectivity that mGGTase I exhibits for its proper
isoprenoid GGPP over FPP itself.9 These findings
prompted us to synthesize three additional 3-substituted
FPP analogues: 3-alFPP (1d , Scheme 1), 3-etFPP (1e,
Chart 1), and 3-phFPP (1f). All three of these com-

3802 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 19
Gibbs et al.
Sch em e 2
may be due to the inability of the putative kinase to
accept the bulky tert-butyl-substituted alcohol. The
selectivity of the biologically active compounds was
addressed by comparing their relative degree of activity,
and that of their geranylgeraniol analogues (4b and 4d ,
Chart 1), against an isogenic set of transformed cell
lines, comprised of NIH3T3 fibroblasts transfected with
either H-Ras(61L) [H-Ras-F] or H-Ras(61L)CVLL [H-
Ras-GG].18 The most striking selectivity was observed
for 3-allylgeranylgeraniol (4d ), which exhibited an IC50
of 4.6 µM against H-Ras-GG cells but was totally
ineffective (IC50 > 25 µM) against H-Ras-F cells. Con-
versely, 3-allylfarnesol exhibited selectivity against
H-Ras-F cells relative to H-Ras-GG cells. Upon com-
parison of the biological activities of the various vinyl
analogues, H-Ras-F cells were significantly more sus-
ceptible to 3-vinylfarnesol (3b) than 3-vinylgeranyl-
geraniol (4b), whereas H-Ras-GG cells were equally
inhibited by both compounds.
The soft agar data established farnesol analogues 3b
and 3d as potent inhibitors of soft agar colony formation.
However, the mechanism for this effect has not been
firmly established. Preliminary data indicating that raf-
transfected cells are also susceptible to treatment with
the vinyl prenyl alcohols suggests that inhibitor of Ras
prenylation may not be the sole or primary mechanism
of action for these compounds (data not shown). Studies
are in progress to investigate the effects of these
compounds on normal cell growth, the growth of cul-
tured human cancer cells, and their therapeutic efficacy
in vivo. To further explore whether the inhibitory
actions of the vinyl analogues are due to abnormal
prenylation, a series of subcellular fractionation experi-
ments were carried out.19 H-Ras-F cells were treated
with lovastatin in order to block the mevalonate path-
way, thereby inhibiting protein prenylation by prevent-
ing the formation of FPP and GGPP. This inhibition is
evidenced on a Western blot as a shift of the majority
of H-Ras from the membrane fraction (DMSO control
cells) to the cytosol (lovastatin-treated cells) (Figure 2a).
Subsequent treatment of the cells with FPP significantly
reversed this effect. Dosing of the lovastatin-treated
cells with 3-alFPP results in virtually all of the H-Ras
being found in the cytosol. This is consistent with
3-alFPP acting as an FTase inhibitor rather than a
substrate. In sharp contrast, dosing of the lovastatin-
treated cells with 3-vFPP results in virtually complete
localization of the H-Ras protein in the membrane
fraction. Thus it appears that 3-vFPP acts as an
alternative substrate for FTase, leading to the formation
of 3-vinylfarnesylated Ras. As the scintillation proximity
assay used initially to evaluate 3-vFPP and 3-alFPP
could not provide evidence for their ability to act as
mFTase substrates, these analogues were tested as
alternative substrates using a continuous spectrofluo-
rimetric assay.20,21 Consistent with the subcellular
localization results, 3-vFPP is an effective mFTase
alternative substrate in vitro, and 3-alFPP is a very poor
one (Table 1). Further confirmation was provided by
treatment of H-Ras-F cells with tritium-labeled 3-vi-
nylfarnesol. As shown in Figure 3, the radiolabel
migrates with the same protein bands as seen when
cells are treated with tritiated farnesol and FPP, verify-
ing the farnesylation of these proteins by 3-vFPP.
Ta ble 2. Anchorage-Independent Growth Inhibition by
Farnesol and Geranylgeraniol Analoguesa
mean IC50 ((SE) (µM)
analogue
H-Ras-F
H-Ras-GG
3-vinylfarnesol (3b)
3-allylfarnesol (3d )
3-ethylfarnesol (3e)
3-phenylfarnesol (3f)
3-vinylgeranylgeraniol (4b)
3-allylgeranylgeraniol (4d )
10.9 ( 2.6 (8)
14.1 ( 1.4 (5)
10.2 ( 3.5 (3) >25 (4)
>25 (2)
>25 (2)
18.0 ( 4.1 (5)
>25 (4)
>25 (2)
>25 (2)
13.9 ( 2.3 (4)
4.6 ( 1.9 (3)
a
Conditions: Cells were assayed for inhibition of anchorage-
independent growth as described under Experimental Procedures.
The numbers in parentheses indicate the number of tests per-
formed.
The promising results obtained with these prenyl
analogues against mFTase suggested that they should
be evaluated as inhibitors of the growth of transformed
cells. Unfortunately, isoprenoid diphosphate analogues
have poor characteristics as potential drugs, in that they
are both unstable and unlikely to penetrate cell mem-
branes unaided, though the natural isoprenoids are
apparently taken up by cells through an active transport
system.15 However, it has been demonstrated that
mammalian cells can utilize farnesol and geranyl-
geraniol for the prenylation of proteins.16 Presumably,
the nonpolar alcohols pass through the cell membrane
and are then diphosphorylated by a kinase, or sequen-
tially by two kinases, to FPP or GGPP.17 Therefore, we
investigated the ability of the 3-substituted farnesol
analogues (3b-f, Chart 1) to inhibit the anchorage-
independent growth of transformed NIH3T3 fibroblasts
in soft agar. Of the four farnesol analogues tested, only
the vinyl (3b) and allyl (3d ) compounds exhibited
cellular activity (Table 2). 3-tert-Butylfarnesol (3c) also
proved to be inactive in cells (data not shown), which

Novel Farnesol and Geranylgeraniol Analogues
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 19 3803
F igu r e 3. Incorporation of radiolabel into total cellular
proteins. H-Ras-F cells were treated with 25 µM lovastatin
for 24 h after which the following were added directly to the
cell media: (1) untreated, (2) 1-[3H]-3-vinylfarnesol (5 µM, 1.34
µCi/mL), (3) 1-[3H]-FPP (3 µM, 50 µCi/mL; Amersham), and
(4) 1-[3H]-farnesol (3 µM, 50 µCi/mL; American Radiolabeled
Chemicals). Following an 18 h incubation period, cells were
lysed40 and proteins separated by SDS-PAGE on a 14% gel
and transferred to an Immobilon-P PVDF membrane (Milli-
pore). After drying, the membranes were sprayed with En3-
Hance (Amersham) and exposed to film (Hyperfilm MP,
Amersham) for 4 days (1-[3H]-FPP, 1-[3H]-farnesol) or 28 days
(1-[3H]-3-vinylfarnesol) before developing.
transformed 12V cells with either 3-vFPP or 3-vGGPP
results in partial restoration of the membrane localiza-
tion of K-Ras (data not shown). Preliminary cell prolif-
eration experiments have been performed to character-
ize the ability of 3-allylfarnesol and 3-allylgeranylgeraniol
to block the growth of 12V cells. Surprisingly, both
compounds were more effective in blocking the growth
of the K-ras-transformed 12V cells than the H-ras-
transformed H61 (H-Ras-F) cell line. After 72 h treat-
ment with 100 µM 3-allylfarnesol, the number of
H-Ras-F cells was decreased to 16.6 ( 4.1% of control,
while the number of 12V cells was reduced to 8.6 ( 1.5%
of the control value.
F igu r e 2. Subcellular fractionation of H-Ras in H-Ras-F and
H-Ras-GG cells. H-Ras-F (A) or H-Ras-GG (B) cells were
treated with 25 µM lovastatin for 24 h after which the
indicated FPP or GGPP analogues (resuspended in media)
were added directly to the cell media. Following an additional
24 h incubation period, the cells were harvested and lysed,
and the membranes (M) were separated from the cytosol (C).19
After solubilization of the membrane fraction, Ras protein was
immunoprecipitated from both the membrane and cytosolic
fractions by the addition of the Y13-259 antibody (OP04 from
Oncogene Science).40 The presence of Ras protein in each
fraction was analyzed by Western blotting (see Experimental
Procedures).40
Discu ssion
However, it appears that the relative levels of radiolabel
incorporation into different proteins is not the same with
3-vinylfarnesol as with farnesol itself, although quan-
titative comparisons cannot be made.
In conclusion, our results demonstrate that (a) certain
FPP analogues can act as potent inhibitors of mam-
malian FTase, (b) farnesol and geranylgeraniol ana-
logues can be prodrugs for the corresponding FPP and
GGPP derivatives, and (c) these prenyl alcohol deriva-
tives potently inhibit the growth of ras-transformed
cells. The selectivity of 3-allylfarnesol and 3-allylgera-
nylgeraniol in soft agar assays (Table 2) and their
behavior in the subcellular fractionation experiments
(Figure 2) are in accord with previous studies on FTase
and GGTase I inhibitors. That is, they appear to block
the growth of ras-transformed cells by preventing the
prenylation of the Ras protein. It is striking and
surprising that 3-allylgeranylgeraniol exhibits such
selectivity in cells, while 3-alGGPP exhibits no selectiv-
ity in vitro (Table 1). Perhaps the intracellular level of
GGPP is in the low nanomolar level in contrast to the
much higher intracellular FPP concentration.22 If this
is true, then 3-alGGPP could compete effectively with
the natural substrate for GGTase I but not FTase.
Nevertheless, 3-allylgeranylgeraniol is a highly specific
cellular inhibitor of protein geranylgeranylation and
thus may be a valuable tool to investigate the relative
biological importance of geranylgeranylation versus
The selectivity and mechanism of the observed cell
growth inhibition was further probed through the effects
of the 3-vinyl and 3-allyl GGPP analogues on the
subcellular distribution of the geranylgeranylated pro-
tein variant in H-Ras-GG cells (Figure 2b). With H-Ras-
GG cells, as with H-Ras-F cells, blockage of the meva-
lonate pathway results in a shift in the subcellular
location of the Ras protein from the membrane to the
cytosol. In accord with the results described above,
dosing of lovastatin-treated H-Ras-GG cells with 3-
vGGPP, but not 3-alGGPP, results in restoration of the
membrane localization of H-Ras-GG. The issue of K-Ras
prenylation is of significant interest, as K-Ras is the
most commonly found mutant Ras protein in human
tumors. Recent studies have demonstrated that K-Ras-
transformed cells are more resistant to treatment with
FTase inhibitors, which is presumably a result of the
ability of the normally farnesylated K-Ras protein to
also be geranylgeranylated by GGTase I.11 In accord
with this result, dosing of lovastatin-treated, K-ras-

3804 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 19
Gibbs et al.
farnesylation.14 Recent evidence has suggested that
protein geranylgeranylation, and not farnesylation,
plays an important role in regulation of endothelial NO
synthase and in cell cycle regulation.23,24
Found: 234.1982. Detailed procedures for the synthesis of 3e
and 3f, the alcohol precursors to 3-etFPP and 3-phFPP, have
been published previously,36 and they were converted into the
corresponding pyrophosphates using the general procedure of
Poulter and co-workers (vide infra).37 The experimental details
for the syntheses of compounds 1d , 2b, and 2d are presented
below.
In sharp contrast, 3-vinylfarnesol is converted to
3-vFPP, which acts as an alternative substrate for
FTase and serves as a prenyl donor both in vitro and
apparently in vivo. The corresponding geranylgeranyl
analogue 3-vGGPP appears to act in the same manner.
Thus the observed biological activity of these compounds
is not due to FTase or GGTase I inhibition. It could be
due to inhibition of squalene synthase, cis-prenyltrans-
ferase, or trans-prenyltransferase, which utilize FPP to
make cholesterol, dolichol, and ubiquinone, respectively.
However, the lower affinities of these enzymes for FPP
(Km ) 6 µM for squalene synthase;25 Km ) 24 µM for
cis-prenyltransferase26) make it less likely that micro-
molar levels of 3-vinylfarnesol or 3-vinylgeranylgeraniol
could effectively block their activity in cell culture.
Relatively high concentrations of the natural products
farnesol and geranylgeraniol have antiproliferative ef-
fects on cultured tumor cells,27,28 possibly in part
through inhibition of phosphatidylcholine biosynthe-
sis.29 However, in control experiments, 30 µM farnesol
exhibited little effect on the proliferation of H-Ras-F
cells (data not shown), in contrast to the complete
inhibition of growth seen with 3-vinylfarnesol.
Evidence has been presented that the prenyl group
plays an active role in mediating protein-protein
interactions,30-34 although this has been a controversial
issue.1,35 While there are many potential reasons that
the 3-vinyl analogues would inhibit cell growth, includ-
ing those described above, we suggest that the incor-
poration of the 3-vinylfarnesyl or 3-vinylgeranylgeranyl
group into a prenylated protein may interfere with its
interaction with various activator or acceptor proteins.36
It is unlikely that Ras protein prenylation is involved,
due to the lack of selectivity between H-Ras-F and
H-Ras-GG cells. However, the RhoB protein, which is
involved in actin cytoskeleton rearrangement, has been
implicated as a key target for farnesyltransferase
inhibitors,4,33 and its 3-vinylprenylation may lead to
growth inhibition. Prendergast and co-workers have
demonstrated that treatment of cells with FTase inhibi-
tors leads to a decrease in farnesylated RhoB and an
increase in geranylgeranylated RhoB, with a concomi-
tant suppression of RhoB-dependent cell growth, con-
sistent with RhoB-GG possessing decreased growth-
stimulatory activity relative to RhoB-F.4,33 Alternatively,
the 3-vinylprenylation of a “protein X” target,3c such as
the prenylated protein tyrosine phosphatases, may be
responsible for the observed growth inhibition effects.
While this proposal is clearly speculative, it suggests
that certain prenyl analogues may interfere with the
function of prenylated proteins in a manner distinct
from that of traditional FTase inhibitors.
Eth yl 3-Allyl-7,11-dim eth yldodeca-2(Z),6(E),10-tr ien oate
(6). Triflate 5 (317 mg; 0.794 mmol),7 triphenylarsine (25 mg;
0.082 mmol), bis(benzonitrile)palladium(II)chloride (15.3 mg;
0.039 mmol), and copper iodide (15.3 mg; 0.085 mmol) were
all placed in an argon-flushed flask and dissolved in 1.0 mL
of NMP (N-methylpyrrolidone; anhydrous, 99.5%). Allyltribu-
tyltin (534 mg; 0.5 mL; 1.61 mmol) was added dropwise, and
the reaction was stirred at 100 °C for ∼24 h. The mixture was
then taken up in a 1:1 solution of hexanes/EtOAc (100 mL),
washed with a 10% KF solution (2 × 30 mL) and water (20
mL), then dried over MgSO4, filtered, and concentrated.
Purification by flash chromatography using a 98:2 hexanes/
EtOAc solvent system yielded 118 mg (50%) of the desired allyl
ester 6. 1H NMR (300 MHz, CDCl3): δ 1.21 (t, 3H, -CH2-CH3),
1.53 (2 s, 6H, two -CH3), 1.61 (s, 3H, -CH3), 1.91-2.04 (m, 8H,
-CH2-), 3.34 (d, 2H, allyl -CH2-), 4.09 (q, 2H, -OCH2CH3), 5.10
(m, 4H), 5.63 (s, 1H, vinylic -CH-), 5.75 (m, 1H). 13C NMR (75.4
MHz, CDCl3): δ 14.3, 16.0, 17.6, 26.6, 36.7, 37.9, 39.6, 59.6,
116.0, 123.9, 124.8, 128.6, 131.3, 133.7, 135.2, 136.1, 160.6,
166.4.
3-Allyl 7,11-Dim eth yldodeca-2(Z),6(E),10-tr ien -1-ol (3d).
Allyl ester 6 (100 mg; 0.343 mmol) was dissolved in toluene
(2.4 mL; anhydrous) at -78 °C under an argon atmosphere.
Addition of diisobutylaluminum hydride (1.0 M in toluene; 1.5
mL; 9.04 mmol) followed, and the reaction was stirred at -78
°C for 1 h. The reaction was quenched by addition to saturated
aqueous potassium sodium tartrate (30 mL), the organic phase
was separated, and the aqueous layer was extracted with ethyl
acetate (3 × 20 mL). The combined organic layers were then
washed with water (10 mL) and NaCl (10 mL), dried over
MgSO4, filtered, and concentrated. Purification by flash chro-
matography using a 9:1 hexanes/EtOAc solvent system gave
1
68 mg (76%) of the desired allyl alcohol 3d as an oil. H NMR
(300 MHz, CDCl3): δ 1.54 (2s, 6H, two vinylic -CH3), 1.68 (s,
3H, vinylic CH3), 1.9-2.05 (m, 8H), 2.83 (d, J ) 6.3 Hz, 2H,
allyl -CH2-), 4.15 (d, J ) 6.8 Hz, 2H, -CH2OH), 5.0-5.08 (m,
4H), 5.51 (t, J ) 6.8 Hz, 1H, H2), 5.77 (m, 1H). 13C NMR (75.4
MHz, CDCl3): δ 16.0, 18.2, 26.0, 27.0, 35.5, 37.4, 40.1, 59.6,
115.9, 124.1, 124.7, 125.0, 131.8, 135.8, 136.5, 141.7. Calcd for
C17H28O: C, 82.2%; H, 11.4%. Found: C, 81.8%; H, 11.5%.
Gen er a l P r oced u r e for P r ep a r a tion of Dip h osp h a tes.37
N-Chlorosuccinimide (1.2 equiv) was dissolved in CH2Cl2
(distilled from CaH2). The solution was cooled to -30 °C in an
acetonitrile/dry ice bath. Methyl sulfide (1.5 equiv) was added
dropwise, and the resulting milky white mixture was warmed
to 0 °C for 5 min and recooled to -30 °C. A solution of 1 equiv
of the alcohol in 1 mL of dichloromethane was added dropwise
to the mixture at -30 °C. The reaction was slowly warmed to
0 °C and stirred for an additional hour at that temperature.
The resulting clear, colorless solution was stirred at room
temperature for 20 min and poured into 10 mL of cold brine
solution. The aqueous layer was extracted with 2 × 15 mL of
hexanes, and the combined organic layers were washed with
10 mL of cold brine solution and dried (MgSO4). Concentration
afforded the chlorides as colorless or pale yellow oils which
were used directly for the next reaction. Tris(tetra-n-butylam-
monium)hydrogen pyrophosphate (2 equiv) was dissolved in
1.0 mL of acetonitrile (distilled from P2O5). The mixture was
cooled to 0 °C, and 1 equiv of chloride in 0.5 mL of acetonitrile
was added dropwise. The reaction was stirred at room tem-
perature for 2 h, and the solvent was removed by rotary
evaporation at room temperature. The residue was dissolved
in 1-2 mL of ion exchange buffer (1:49 v/v isopropyl alcohol
and 25 mM NH4HCO3) and was passed through a column
containing 3-10 mL cation exchange resin (DOWEX AG 50W-
Exp er im en ta l P r oced u r es
F a r n esyl a n d Ger a n ylger a n yl An a logu es. Compounds
1b and 3b,7 along with 1c and 3c,8 were prepared as previously
described. Since 3b had not been submitted for analysis
previously, it was for this study. Analysis: Calcd for C16H26O:
C, 81.98%; H, 11.10%. Found: C, 80.63%; H, 10.97%. Since
the carbon analysis was off by >1%, we also confirmed the
structure of 3b by HRMS: Calcd for C16H26O: 234.1984.
+
X8, NH4 form). The column was eluted with two column
volumes of ion exchange buffer at a flow rate of ∼1 mL/min.

Novel Farnesol and Geranylgeraniol Analogues
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 19 3805
The eluent was dried by lyophilization, and a pale yellow solid
was obtained. The crude product was dissolved in 1-3 mL of
25 mM NH4HCO3 and purified by reversed phase HPLC using
a program of 5 min of 100% A followed by a linear gradient of
100% A to 100% B over 30 min (A: 25 mM aqueous NH4HCO3,
pH 8.0; B: CH3CN; Vydac pH-stable C8 4.6 mm × 250 mm
columm; flow rate: 1.0 mL; UV monitoring at 214 and 254
nm). The fractions were collected, pooled, and dried by lyo-
philization, and the diphosphates were obtained as white fluffy
solids. Due to the hygroscopic and amorphous nature of the
diphosphates and the limited amounts available in some cases,
these compounds were not characterized by elemental analysis.
However, these compounds were always purified by reversed
phase HPLC, and their purity and identity were confirmed
39.59, 39.68, 61.23, 112.01, 120.55, 123,76, 124.26, 131.32,
135.27, 138.21, 158.46, 162.47. MS-EI: 464 (M+). Calcd for C23
-
H
35F3O5S: C, 57.5%; H, 7.3%. Found: C, 58.0%; H, 7.5%.
Eth yl 3-Allyl-7,11,15-tr im eth ylh exa d eca -2Z,6E,10E,14-
tetr a en oa te 11. In a flame-dried, argon-flushed flask were
placed triflate 10 (562 mg, 1.2 mmol), Pd(PhCN)2Cl2 (23 mg,
0.061 mmol), AsPh3 (38 mg, 0.12 mmol), CuI (23 mg, 0.12
mmol), and 1.5 mL of NMP (99.5%, anhydrous). The flask was
heated to 100 °C, and to this mixture was added allyltribu-
tyltin (0.76 mL, 2.4 mmol). After 15 h at 100 °C the reaction
mixture was cooled, taken up in 100 mL of ethyl acetate, and
washed with aqueous KF (3 × 30 mL). The aqueous layer was
back-extracted with ethyl acetate (2 × 15 mL), and the
combined organic layers were dried over MgSO4. Concentration
followed by purification by flash chromatography (hexanes/
ethyl acetate 98:2) gave 11 as a colorless oil (222 mg, 52%).
1H NMR (300 MHz, CDCl3): δ 1.25 (t, J ) 6.8 Hz, 3H, -CH2-
CH3), 1.56 (app s, 9H, three -CH3), 1.68 (s, 3H, -CH3), 1.95-
2.2 (m, 12H, -CH2-), 3.42 (d, J ) 6.0 Hz, 2H, allyl -CH2-), 4.14
(q, J ) 6.8 Hz, 2H, -OCH2CH3), 5.06 (m, 5H), 5.70 (s, 1H,
vinylic -CH-), 5.79 (complex m, 1H).
3-Allyl-7,11,15-tr im eth ylh exadeca-2Z,6E,10E,14-tetr aen -
1-ol 4d . To the solution of ester 11 (191 mg, 0.53 mmol) in 3.0
mL of toluene was added diisobutyl aluminum hydride (1.0 M
solution in toluene, 1.5 mL, 1.5 mmol) under argon at -78
°C. The reaction was stirred at -78 °C for 1 h and warmed to
room temperature. The reaction was quenched by adding 30
mL of saturated aqueous potassium sodium tartrate. The
aqueous solution was extracted with ethyl acetate (2 × 20 mL).
The combined organic layers were washed with saturated NaCl
(2 × 20 mL) and dried over MgSO4. Concentration followed
by flash chromatography (hexanes/ethyl acetate ) 9:1) afforded
alcohol 4d (128 mg, 76%) as a colorless oil. 1H NMR (300 MHz,
CDCl3): 1.60 (app s, 9H, three vinylic -CH3), 1.76 (s, 3H, vinylic
CH3), 1.95-2.1 (m, 12H), 2.83 (d, J ) 6.3 Hz, 2H, allyl -CH2-),
4.15 (d, J ) 7.0 Hz, 2H, -CH2OH), 5.0-5.08 (m, 5H), 5.50 (t,
J ) 7.0 Hz, 1H, H2), 5.75 (complex m, 1H). 13C NMR (75.4
MHz, CDCl3): δ 16.5, 18.1, 26.0, 26.8, 27.2, 30.9, 35.6, 37.5,
40.1, 59.6, 115.9, 116.7, 124.2, 124.5, 124.9, 125.2, 131.6, 135.4,
135.9, 136.6. Calcd for C22H36O: C, 83.5%; H, 11.5%. Found:
C, 83.4%; H, 11.7%.
1
by analytical reversed phase HPLC, H NMR, 31P NMR, and
in some cases quantitative phosphate analysis.21
3-Allyl-7,11-dim eth yldodeca-2(Z),6(E),10-tr ien e 1-Diph os-
p h a te (3-a lF P P ) (1d ). Allyl alcohol 3d (68 mg; 0.261 mmol)
was treated with N-chlorosuccinimide (60 mg; 0.42 mmol) and
dimethyl sulfide (27 mg; 0.03 mL; 0.45 mmol) in 5.0 mL of
CH2Cl2. Following the general procedure for the preparation
of chlorides described above, 24 mg (33%) of chloride 7 was
obtained as a pale yellow oil that was used directly in the next
step. Compound 7 (24 mg, 0.087 mmol) was then treated with
tris(tetra-n-butylammonium) hydrogen pyrophosphate (365
mg; 0.40 mmol) in 3.0 mL of acetonitrile for 2 h. The resulting
material was converted to ammonium form by treatment with
3 mL of resin and 8 mL of ion exchange buffer. Following the
general reversed phase HPLC purification procedure described
above (retention time of 3-alFPP: 17 min), 32 mg (90%) of
3-alFPP 1d was obtained as a white fluffy solid. 1H NMR (300
MHz, D2O): δ 1.57 (s, 6H), 1.63 (s, 3H), 1.95-2.10 (m, 8H),
2.85 (d, 2H), 4.44 (b, 2H), 5.10-5.33 (m, 4H), 5.50 (t, 1H), 5.8
(complex m, 1H). 31P NMR (121 MHz, D2O): -5.11, -9.15.
Eth yl 7,11,16-Tr im eth yl-3-oxoh exa d eca -6E,10E,14-tr i-
en oa te 9.38 Monosodium ethyl acetoacetate 8 (3.04 g, 20.0
mmol) was dissolved in 30.0 mL of THF under argon. The
solution was cooled to 0 °C and treated with n-butyllithium
(2.0 M in cyclohexane, 10.6 mL, 21.0 mmol). After 20 min at
0 °C, neat farnesyl bromide (1.98 mL, 2.1 g, 7.3 mmol) was
added to the resulting dianion solution, and stirring was
continued for additional 30 min. The reaction mixture was
poured into a cold saturated solution of potassium hydrogen
phosphate (∼25 mL) and extracted with ether (3 × 25 mL).
The organic layers were combined, washed with water (20 mL),
and dried over MgSO4. After purification by flash chromatog-
raphy (hexanes/ethyl acetate 9:1), 1.33 g (54%) of the product
9 was obtained as a pale yellow oil. 1H NMR (300 MHz,
CDCl3): δ 1.28 (t, J ) 7.0 Hz, 3H), 1.59 (s, 3H), 1.60 (s, 6H),
1.68 (s, 3H), 2.00 (m, 8H), 2.28 (m, 2H), 2.54 (t, J ) 7.2 Hz,
C4-CH2, 2H), 3.43 (s, 2H), 4.20 (q, J ) 7.0 Hz, 2H), 5.08 (m,
3H). 13C NMR (75.4 MHz, CDCl3): δ 14.11, 15.97, 17.66 22.14,
25.66, 25.71, 26.52, 27.74, 35.21, 39.67, 43.05, 49.31, 49.39,
61.27, 61.34, 122.04, 124.33, 131.28, 135.07, 136.79, 167.19,
202.61. MS-EI: 334 (M+).
3-Allyl-7,11,15-tr im eth ylh exa d eca -2Z,6E,10E,14-tetr a -
en e Dip h osp h a te 2d . Allyl alcohol 4d (78 mg, 0.26 mmol)
was treated with N-chlorosuccinimide (46 mg, 0.34 mmol) and
dimethyl sulfide (0.029 mL, 0.39 mmol) in 3.0 mL of CH2Cl2.
Following the general procedure for the preparation of chloride
described previously, 62 mg (74%) of the corresponding allyl
chloride was obtained as a pale yellow oil that was used
directly in the next step. The chloride (62 mg, 0.19 mmol) was
treated with tris(tetra-n-butylammonium)hydrogen pyrophos-
phate (350 mg, 0.38 mmol) in 3.0 mL of acetonitrile for 2 h.
The resulting material was converted to ammonium form by
treatment with 3 mL of resin and 8 mL of ion exchange buffer.
Following the general reversed phase HPLC purification
procedure described above, 62 mg (64%) of 3-alGGPP 2d was
1
obtained as a white fluffy solid. H NMR (300 MHz, D2O): δ
Eth yl 3-(Tr iflu or om eth ylsu lfon yloxy)h exa d eca -7,11,-
15-tr im eth yl-2Z,6E,10E,14-tetr a en oa te 10.38 In an argon-
flushed flask, â-ketoester 9 (700 mg, 2.1 mmol) was dissolved
in 6.0 mL of THF. The solution was cooled to -78 °C and
potassium bis(trimethylsilyl)amide (KHMDS; 0.5 M in toluene,
2.4 mmol, 4.8 mL) was added dropwise. After 1.5 h, a slurry
of 2-[N,N-bis(trifluoromethylsulfonyl)amide]-5-chloropyridine
(946 mg, 2.4 mmol) in 2.0 mL of THF was added to the
resulting enolate solution. The reaction was allowed to warm
from -78 °C to room temperature over 3 h. It was then taken
up in 30 mL of ether, washed with 10% aqueous citric acid
(2 × 15 mL) and water (20 mL), dried over MgSO4, and
concentrated. Purification by flash chromatography (20:1
hexanes/ethyl acetate) gave 562 mg (57%) of triflate 10 as a
1.5-1.7 (3s, 12H), 1.85-2.10 (m, 12H), 2.85 (b, 2H), 4.45 (b,
2H), 4.95-5.15 (m, 5H), 5.50 (t, 1H), 5.7-5.8 (b, 1H). 31P NMR
(121 MHz, D2O): -5.27, -9.15.
Eth yl 3-Vin yl-7,11,15-tr im eth ylh exa d eca -2Z,6E,10E,14-
tetr a en oa te 12. In a flame-dried, argon-flushed flask were
placed triflate 16 (180 mg, 0.39 mmol), Pd(PhCN)2Cl2 (7.7 mg,
0.02 mmol), AsPh3 (24 mg, 0.08 mmol), CuI (7.4 mg, 0.04
mmol), and 0.5 mL of NMP (99.5%, anhydrous). To this
mixture was added vinyltributyltin (0.14 mL, 0.46 mmol), and
the reaction was stirred for 15 h at room temperature. The
reaction was taken up with 100 mL of ethyl acetate and
washed with aqueous KF (3 × 30 mL). The aqueous layer was
back-extracted with ethyl acetate (2 × 15 mL), and the
combined organic layers were dried over MgSO4. Concentration
followed by purification by flash chromatography (hexanes/
ethyl acetate 20:1) gave 12 as a colorless oil (98 mg, 73%). The
identity, and in particular the stereochemistry, of this ester
was confirmed by the similarity of its 1H NMR spectrum to
1
pale yellow oil. H NMR (300 MHz, CDCl3): δ 1.28 (t, J ) 7.2
Hz, 3H), 1.56 (s, 3H), 1.60 (s, 6H), 1.62 (s, 3H), 2.17 (m, 8H),
2.39 (m, 2H), 2.43 (t, J ) 7.2 Hz, 2H), 4.25 (q, J ) 7.2 Hz,
2H), 5.09 (m, 3H), 5.74 (s, 1H). 13C NMR (75.4 MHz, CDCl3):
δ 14.01, 15.98, 16.04, 17.66, 24.39, 25.68, 26.49, 26.72, 34.58,

3806 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 19
Gibbs et al.
the previously prepared 3-vinyl-3-desmethylfarnesyl ester. 1H
NMR (300 MHz, CDCl3): δ 1.28 (t, J ) 6.9 Hz, 3H), 1.60 (s,
9H), 1.68 (s, 3H), 1.99-2.06 (m, 8H), 2.21 (m, 2H), 2.37 (m,
2H), 4.20 (q, J ) 6.9 Hz, 2H), 5.11 (m, 3H), 5.44 (dd, 2H, J )
1.2, J ) 11.1 Hz), 5.62 (d, J ) 17.7 Hz, 1H), 5.70 (s, 1H), 7.73
American Radiolabeled Chemicals) was added as a solution
in methanol (0.30 mL; distilled from Mg). After 4 h at room
temperature, ∼4 mg of sodium borodeuteride was added to
complete the reaction, and stirring was continued overnight.
The reaction was quenched by addition of saturated NaCl (1
mL), and then the mixture was extracted with ether (3 × 1
mL). Each fraction of ether was passed through a pipet
containing MgSO4 over glass wool, and the combined ether
layers were then concentrated. The oil was then taken up in
additional ether (∼5 mL) and passed through a second plug
of MgSO4 to remove residual water. Following concentration,
19 mg (89%) of the desired alcohol was obtained as an oil
(radiochemical yield of 26% (2.17 mCi; specific activity ) 27)).
Its identity was confirmed by TLC comparison to the unlabeled
material and by conversion to 1-[3H]-3-VFPP 1b (using the
general procedure described above) that exhibited HPLC
mobility identical to the unlabeled compound and a proton
NMR spectrum similar to it except for the decreased size of
the C1 proton signal due to deuterium incorporation.
3
(dd, 1H, J ) 11.1, 17.7 Hz). 13C NMR (75.4 MHz, CDCl3): δ
14.30, 15.97, 16.08, 17.69, 25.68, 26.75, 27.46, 33.59, 39.68,
59.84, 117.37, 119.87, 123.01, 124.33, 124.08, 131.29, 133.07,
135.04, 136.30, 154.35, 166.32.
3-Vin yl-7,11,15-tr im eth ylh exadeca-2Z,6E,10E,14-tetr aen -
1-ol 4b. To the solution of ester 12 (95 mg, 0.28 mmol) in 2.0
mL of toluene was added diisobutyl aluminum hydride (1.0 M
solution in toluene, 0.7 mL, 0.7 mmol) under argon at -78
°C. The reaction was stirred at -78 °C for 1 h and warmed to
room temperature. The reaction was quenched by adding it
to 30 mL of saturated aqueous potassium sodium tartrate. The
aqueous solution was extracted with ethyl acetate (2 × 20 mL).
The combined organic layers were washed with saturated NaCl
(2 × 20 mL) and dried over MgSO4. Concentration followed
by flash chromatography (hexanes/ethyl acetate ) 4:1) afforded
alcohol 4b (52 mg, 62%) as a colorless oil. 1H NMR (300 MHz,
CDCl3): δ 1.60 (s, 9H), 1.69 (s, 3H), 2.00-2.05 (m, 8H), 2.18-
2.22 (m, 4H), 4.31 (d, J ) 6.6 Hz, 2H), 5.12 (m, 3H), 5.19 (d,
J ) 9.6 Hz), 5.33 (d, J ) 17.4 Hz), 5.59 (t, J ) 6.6 Hz, 1H),
6.64 (dd, J ) 10.8, 17.3 Hz). 13C NMR (75.4 MHz, CDCl3): δ
16.01, 17.69, 25.68, 26.62, 26.75, 27.13, 33.28, 39.69, 58.61,
11.34, 123.79, 124.14, 124.37, 127.86, 131.29, 132,15, 135.01,
135.56, 139.44. MS-EI: 302 (M+). HRMS: Calcd for C21H34O:
302.2610. Found: 302.2613. Analysis: Calcd for C21H34O: C,
83.4%; H, 11.3%. Found: C, 81.4%; H, 11.3%. Due to the
unsatisfactory elemental analysis for 4b, it was subjected to
GC/MS analysis of purity (HP5988A GC/MS, DB5ms columm,
100-250 °C at 10 degrees/min), which demonstrated that it
was >99% pure (retention time: 13.34 min). A second sample
of 4b that was incubated at 30 °C for 24 h under an air
atmosphere was 94.3% pure, by GC/MS analysis.
P r en yltr a n sfer a se IC50 a n d Ki Assa ys. FTase IC50 values
were determined using recombinant mFTase9 in a scintillation
proximity assay (using streptavidin beads from Amersham)
with tritiated FPP (specific activity 15-30 Ci/mmol, final
concentration 0.12 µM) and the peptide Biotin-Aha-Thr-Lys-
Cys-Val-Ile-Met-OH (final concentration 0.1 µM) as substrates,
in the same manner as previously described.39 The Ki values
were determined using the same assay system with varying
concentrations of tritiated FPP. GGTase I values were deter-
mined in a similar manner using recombinant mGGTase I,
tritiated GGPP, and the peptide Biotin-Aha-Thr-Lys-Cys-Val-
Ile-Leu-OH in a scintillation proximity assay.
F a r n esyltr a n sfer a se Km a n d kr el Assa ys. The kinetic
constants for the FPP analogues were determined following a
continuous spectrofluorimetric assay originally developed by
Pompliano et al. and modified by Poulter and co-workers.20,21
Utilizing dansyl-GCVLS as a peptide substrate, the linear
portion of the increase in fluorescence versus time was
measured with a Spex FluoroMax2 spectrofluorimeter (excita-
tion wavelength ) 350 nm; emission wavelength ) 486 nm).
The assay components [585 µL of assay buffer (52 mM Tris-
HCl, pH 7.0, 5.8 mM DTT, 12 mM MgCl2, 12 µM ZnCl2), 75
µL of detergent solution (0.4% n-dodecyl-â-D-maltoside in 52
mM Tris-HCl, pH 7.0), 45 µL of dansyl-GCVLS solution (12
µM in 20 mM Tris-HCl, pH 7.0, 10 mM EDTA)] were as-
sembled in a 1.5 mL Eppendorf tube in the order indicated
above and were incubated at 30 °C for a period of 5 min. FPP,
3-vFPP, or 3-alFPP (∼15 mM stock solution in 25 mM
ammonium bicarbonate, pH 7.5; final concentration 0.10 to 5
µM) was added to the assay buffer solution. The reaction was
then initiated with mFTase, the resulting solution was im-
mediately pipetted into a 1.0 mL quartz cuvette, and fluores-
cence was detected using a time-based scan at 30 °C for a
period of 300 s. Determination of the velocity was determined
by converting the rate of increase in fluorescence intensity (cps/
s) to µM/s using the formula that has been described previ-
ously.21 The fluorescence enhancement value e was determined
individually using FPP, 3v-FPP, and 3-alFPP as substrates.
3-Vin yl-7,11,15-t r im et h ylh exa d eca -2Z,6E,10E,14-t et -
r a en e d ip h osp h a te 2b. Vinyl alcohol 4b (78 mg, 0.26 mmol)
was treated with N-chlorosuccinimide (46 mg, 0.34 mmol) and
dimethyl sulfide (0.029 mL, 0.39 mmol) in 3.0 mL of CH2Cl2.
Following the general procedure for the preparation of chloride
described previously, 62 mg (74%) of the desired vinyl chloride
was obtained as a pale yellow oil that was used directly in the
next step. The chloride (62 mg, 0.19 mmol) was then treated
with tris(tetra-n-butylammonium)hydrogen pyrophosphate (350
mg, 0.38 mmol) in 3.0 mL of acetonitrile for 2 h. The resulting
material was converted to ammonium form by treatment with
3 mL of resin and 8 mL of ion exchange buffer. Following the
general reversed phase HPLC purification procedure described
above, 62 mg (64%) of 3-vGGPP 2b was obtained as a white
1
fluffy soild. H NMR (300 MHz, D2O): δ 1.57 (s, 9H), 1.64 (s,
3H), 1.97-2.18 (m, 12H), 4.59 (b, 2H), 5.10-5.33 (m, 5H), 5.62
(b, 1H), 6.74 (dd, J ) 11.4, 11.1 Hz, 1H). 31P NMR (121 MHz,
D2O): -6.72, -9.95.
1-[3H]-3-Vin yl 7,11-d im eth yld od eca -2(Z),6(E),10-tr ien -
1-ol. Unlabeled 3-vinylfarnesol 3b (20 mg; 0.085 mmol) was
dissolved in hexanes (1.9 mL; anhydrous) in the dark under
an argon atmosphere. Manganese dioxide (110 mg; 1.27 mmol)
was added in one portion, and the reaction was stirred at room
temperature for 90 min [the reaction was followed by TLC (9:1
hexanes/EtOAc)]. The reaction mixture was filtered through
a 2.5 cm silica gel plug, which was then washed with ∼5 mL
of ether. Concentration afforded 21 mg of the desired aldehyde,
which was used directly in the next step without further
purification. Note that the aldehyde was stable to storage at
-78 °C under argon for 72 h, but was unstable to storage
overnight at higher temperatures. 1H NMR (300 MHz,
CDCl3): δ 1.60 (s, 6H, two vinylic -CH3), 1.70 (s, 3H, vinylic
CH3), 2.0-2.2 (m, 4H), 2.30 (app q, 2H), 2.45 (app t, 2H), 5.15
(m, 2H), 5.60 (d, J ) 10.8 Hz, 1H), 5.69 (d, J ) 17.4 Hz, 1H),
5.93 (d, J ) 7.8 Hz, 1H, H2), 7.20 (dd, 1H, J ) 10.8, 17.4 Hz),
10.2 (d, J ) 7.8 Hz, 1H, H1). The crude aldehyde (21 mg) was
placed in a 3 mL Ace reaction vial equipped with spin vane.
Sodium borotritide (8.3 mCi; 150 mCi/mmol; 0.055 mmol;
Wester n An a lysis. Detection of Ras proteins by Western
blotting was performed as previously described from this
group40 using pan-Ras Ab-2 (Oncogene Science) and an anti-
mouse HRP conjugate secondary antibody (Amersham). Blots
were developed using ECL techniques (Amersham).
Soft Aga r Assa ys. The H-Ras-F, H-Ras-GG, and raf cell
lines have been previously described19 and were grown in
Dulbecco’s modified Eagle medium supplemented with 10%
calf serum and 1% antibiotic/antimycotic at 37 °C and 10%
CO2. Experiments were carried out in 6-well dishes in a two-
layer agar system (0.6% bottom layer and 0.3% top layer). Cells
were incorporated into the top layer along with varying
concentrations of the compound prepared in ethanol. Com-
pound addition only occurred at the time cells were seeded.
Subsequent incubation was at 37 °C with 10% CO2 for 2 weeks.
Colonies were stained with 0.5 mL/well of 1 mg/mL p-

Novel Farnesol and Geranylgeraniol Analogues
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 19 3807
(13) Macchia, M.; J annitti, N.; Gervasi, G.; Danesi, R. Geranylgeranyl
Diphosphate-Based Inhibitors of Posttranslational Geranylgera-
nylation of Cellular Proteins. J . Med. Chem. 1996, 39, 1352-
1356.
(14) For a very recent report of a highly selective peptide-based
GGTase I inhibitor, see: Vasudevan, A.; Qian, Y.; Vogt, A.;
Blaskovich, M. A.; Ohkanda, J .; Sebti, S. M.; Hamilton, A. D.
Disruption of Oncogenic K-Ras4B Processing and Signaling by
a Potent Geranylgeranyltransferase I Inhibitor. J . Med. Chem.
1999, 42, 1333-1340.
(15) Danesi, R.; McLellan, C. A.; Meyers, C. E. Specific Labeling of
Isoprenylated Proteins: Application to Study Inhibitors of the
Posttranslational Farnesylation and Geranylgeranylation. Bio-
chem. Biophys. Res. Commun. 1995, 206, 637-643.
(16) Crick, D. C.; Andres, D. A.; Waechter, C. J . Novel Salvage
Pathway Utilizing Farnesol and Geranylgeraniol for Protein
Isoprenylation. Biochem. Biophys. Res. Commun. 1997, 237,
483-487.
(17) Bentinger, M.; Grunler, J .; Peterson, E.; Swiezewska, E.; Dallner,
G. Phosphorylation of Farnesol in Rat Liver Microsomes: Prop-
erties of Farnesol Kinase and Farnesyl Phosphate Kinase. Arch.
Biochem. Biophys. 1998, 353, 191-198.
(18) Cox, A. D.; Hisaka, M. M.; Buss, J . E.; Der, C. J . Specific
Isoprenoid Modification is Required for Function of Normal, but
Not Oncogenic, Ras Protein. Mol. Cell. Biol. 1992, 12, 2606-
2615.
(19) Cox, A. D.; Solski, P. A.; J ordan, J . D.; Der, C. J . Analysis of
Ras Protein Expression in Mammalian Cells. Methods Enzymol.
1995, 255, 195-220.
(20) Pompliano, D. L.; Gomez, R. P.; Anthony, N. J . Intramolecular
Fluorescence Enhancement: A Continuous Assay of Ras Far-
nesyl: Protein Transferase. J . Am. Chem. Soc. 1992, 114, 7945-
7946.
(21) Cassidy, P. B.; Dolence, J . M.; Poulter, C. D. Continuous
Fluorescence Assay for Protein Prenyltransferases. Methods
Enzymol. 1995, 250, 30-43.
(22) Bruenger, E.; Rilling, H. C. Determination of Isopentenyl
Diphosphate and Farnesyl Diphosphate in Tissue Samples with
a Comment on Secondary Regulation of Polyisoprenoid Biosyn-
thesis. Anal. Biochem. 1988, 173, 321-327. Note that the only
known biological role of GGPP in mammalian cells is as a
substrate for GGTase I and GGTase II, and thus it is reasonable
to propose that its intracellular levels could be very low.
(23) Laufs, U.; Liao, J . K. Posttranscriptional Regulation of Endo-
thelial Nitric Oxide Synthase mRNA Stability by Rho GTPase.
J . Biol. Chem. 1998, 273, 24266-24271.
(24) Sun, J .; Qian, Y.; Chen, Z.; Marfurt, J .; Hamilton, A. D.; Sebti,
S. M. The Geranylgeranyltransferase I Inhibitor GGTI-298
Induces Hypophosphorylation of Retinoblastoma and Partner
Switching of Cyclin-dependent Kinase Inhibitors. J . Biol. Chem.
1999, 274, 6930-6934.
(25) Lindsey, S.; Harwood, H. J ., J r. Inhibition of Mammalian
Squalene Synthetase Activity by Zaragozic Acid A is a Result of
Competitive Inhibition Followed by Mechanism-based Irrevers-
ible Inactivation. J . Biol. Chem. 1995, 270, 9038-9096.
(26) Ericsson, J .; Thelin, A.; Chojnacki, T.; Dallner, G. Substrate
Specificity of cis-Prenyltransferase in Rat Liver Microsomes. J .
Biol. Chem. 1992, 267, 19730-19735.
(27) Voziyan, P. A.; Haug, J . S.; Melnykovych, G. Mechanism of
Farnesol Cytotoxicity: Further Evidence for the Role of PKC-
Dependent Signal Transduction in Farnesol-Induced Apoptotic
Cell Death. Biochem. Biophys. Res. Commun. 1995, 212, 479-
486.
(28) Burke, Y. D.; Stark, M. J .; Roach, S. L.; Sen, S. E.; Crowell, P.
L. Inhibition of Pancreatic Cancer Cell Growth by the Dietary
Isoprenoids Farnesol and Geraniol. Lipids 1997, 32, 151-156.
(29) Miquel, K.; Pardines, A.; Terece, F.; Selmi, S.; Favre, G.
Competitive Inhibition of Choline Phosphotransferase by Gera-
nylgeraniol and Farnesol Inhibits Phosphatidylcholine Synthesis
and Induces Apoptosis in Human Lung Adenocarcinoma A549
Cells. J . Biol. Chem. 1998, 273, 26179-26186.
(30) McGeady, P.; Kuroda, S.; Shimizu, K.; Takai, Y.; Gelb, M. H.
The Farnesyl Group of H-Ras Facilitates the Activation of a
Soluble Upstream Activator of Mitogen-activated Protein Kinase.
J . Biol. Chem. 1995, 270, 26347-26351.
(31) Okada, T.; Masuda, T.; Shinkai, M.; Kariya, K.; Kataoka, T.
Posttranslational Modification of H-Ras Is Required for Activa-
tion of, but Not for Association with, B-Raf. J . Biol. Chem. 1996,
271, 4671-4678.
(32) Siddiqui, A. A.; Garland, J . R.; Dalton, M. B.; Sinensky, M.
Evidence for a High Affinity, Saturable, Prenylation-dependent
p21Ha-ras Binding Site in Plasma Membranes. J . Biol. Chem.
1998, 273, 3712-3717.
iodonitrotetrazolium violet (Sigma) 24 h prior to quantitation
by image analysis.
Ack n ow led gm en t. This research was supported in
part by NIH Grants CA-67292 and CA-78819 (to R.A.G.).
T.J .Z. was supported in part by a WSU GRA award.
Y.Q.M. was the recipient of a Rumble Fellowship from
WSU, and R.A.G. was the recipient of an American
Cancer Society J unior Faculty Research Award (J FRA-
609). We are grateful for the excellent technical support
of R. Gowan, M. Latash, and M. Kirzhner (Parke-Davis).
Refer en ces
(1) Zhang, F. L.; Casey, P. J . Protein Prenylation: Molecular
Mechanisms and Functional Consequences. Annu. Rev. Biochem.
1996, 65, 241-269.
(2) Casey, P. J .; Seabra, M. C. Protein Prenyltransferases. J . Biol.
Chem. 1996, 271, 5289-5292.
(3) (a) Leonard, D. M. Ras Farnesyltransferase: A New Therapeutic
Target. J . Med. Chem. 1997, 40, 2971-2990. (b) Gibbs, J . B.;
Oliff, A. The Potential of Farnesyltransferase Inhibitors as
Cancer Chemotherapeutics. Annu. Rev. Pharmacol. Toxicol.
1997, 37, 143-166. (c) Cox, A. D.; Der, C. J . Farnesyltransferase
Inhibitors and Cancer Treatment: Targeting Simply Ras?
Biochim. Biophys. Acta 1997, 1333, F51-F71.
(4) Du, W.; Lebowitz, P. F.; Prendergast, G. C. Cell Growth
Inhibition by Farnesyltransferase Inhibitors Is Mediated by Gain
of Geranylgeranylated RhoB. Mol. Cell. Biol. 1999, 19, 1831-
1840.
(5) Sepp-Lorenzino, L.; Rosen, N. A farnesyl-protein transferase
inhibitor induces p21 expression and G1 block in p53 wild-type
tumor cells. J . Biol. Chem. 1998, 273, 20243-20251.
(6) Kohl, N. E.; Omer, C. A.; Conner, M. W.; Anthony, N. J .; Davide,
J . P.; DeSolms, S. J .; Giuliani, E. A.; Gomez, R. P.; Graham, S.
L.; Hamilton, K.; Handt, L. K.; Hartman, G. D.; Koblan, K. S.;
Kral, A. M.; Miller, P. J .; Mosser, S. D.; O’Neill, T. J .; Rands,
E.; Schaber, M. D.; Gibbs, J . B.; Oliff, A. Inhibition of Farne-
syltransferase Induces Regression of Mammary and Salivary
Carcinomas in ras Transgenic Mice. Nature (Medicine) 1995, 1,
792-797.
(7) Gibbs, R. A.; Krishnan, U.; Dolence, J . M.; Poulter, C. D. A Ste-
reoselective Palladium/Copper-Catalyzed Route to Isoprenoids:
Synthesis and Biological Evaluation of 13-Methylidenefarnesyl
Diphosphate. J . Org. Chem. 1995, 60, 7821-7829.
(8) (a) Mu, Y. Q.; Gibbs, R. A.; Eubanks, L. M.; Poulter, C. D.
Cuprate Mediated Synthesis and Biological Evaluation of tert-
Butyl and Cyclopropyl Farnesyl Diphosphate Analogues. J . Org.
Chem. 1996, 61, 8010-8015. In addition to our compounds and
other related compounds8b that were designed as FPP-competi-
tive inhibitors, several natural products have been found that
are potent FPP-competitive mFTase inhibitors.8c (b) For ex-
ample, R-(hydroxyfarnesyl)phosphonic acid: Gibbs, J . B.; Pom-
pliano, D. L.; Mosser, S. D.; Rands, E.; Lingham, R. B.; Singh,
S. B.; Scolnick, E. M.; Kohl, N. E.; Oliff, A. Selective Inhibition
of Farnesyl-Protein Transferase Blocks Ras Processing in Vivo.
J . Biol. Chem. 1993, 268, 7617-7620. (c) For example, the
chaetomellic acids: Singh, S. B.; Zink, D. L.; Liesch, J . M.; Goetz,
M. A.; J enking, R. G.; Nallin-Omstead, M.; Silverman, K. C.;
Bills, G. F.; Mosley, R. T.; Gibbs, J . B.; Albers-Schonberg, G.;
Lingham, R. B. Isolation and Structure of Chaetomellic Acids A
and B from Chaetomella acutiseta: Farnesyl Pyrophosphate
Mimic Inhibitors of Ras Farnesyl-Protein Transferase. Tetra-
hedron 1993, 49, 5917-5926.
(9) Yokoyama, K.; Zimmerman, K.; Scholten, J .; Gelb, M. H.
Differential Prenyl Pyrophosphate Binding to Mammalian Pro-
tein Geranylgeranyltransferase I and Protein Farnesyltrans-
ferase and Its Consequence on the Specificity of Protein Pren-
ylation. J . Biol. Chem. 1997, 272, 3944-3952.
(10) Dolence, J . M.; Cassidy, P. B.; Mathis, J . R. Yeast Protein
Farnesyltransferase: Steady-State Kinetic Studies of Substrate
Binding. Biochemistry 1995, 34, 16687-16694.
(11) (a) 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. (b)
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.
(12) Lerner, E. C.; Qian, Y.; Hamilton, A. D.; Sebti, S. M. Disruption
of Oncogenic K-Ras4B Processing and Signaling by a Potent
Geranylgeranyltransferase I Inhibitor. J . Biol. Chem. 1995, 270,
26770-26773.
(33) Lebowitz, P. F.; Casey, P. J .; Prendergast, G. C.; Thissen, J . A.
Farnesyltransferase Inhibitors Alter the Prenylation and Growth-
stimulating Function of RhoB. J . Biol. Chem. 1997, 272, 15591-
15594.

3808 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 19
Gibbs et al.
(34) Myung, C.-S.; Yasuda, H.; Liu, W. W.; Harden, T. K.; Garrison,
J . C. Role of Isoprenoid Lipids on the Heterotrimeric G Protein
γ Subunit in Determining Effector Activation. J . Biol. Chem.
1999, 274, 16595-16603.
(38) For a preliminary report on the synthesis of 9 and 10 via a
slightly different method, see: Mu, Y. Q.; Gibbs, R. A. Coupling
of Isoprenoid Triflates with Organoboron Nucleophiles: Syn-
thesis of all-trans-Geranylgeraniol. Tetrahedron Lett. 1995, 36,
5669-5672.
(39) Leonard, D. M.; Shuler, K. R.; Poulter, C. J .; Eaton, S. R.;
Sawyer, T. K.; Hodges, J . C.; Su, T. Z.; Scholten, J . D.; Gowan,
R. C.; Sebolt-Leopold, J . S.; Doherty, A. M. Structure-Activity
Relationships of Cysteine-Lacking Pentapeptide Derivatives
That Inhibit ras Farnesyltransferase. J . Med. Chem. 1997, 40,
192-200.
(35) Parish, C. A.; Rando, R. R. Isoprenylation/Methylation of
Proteins Enhances Membrane Association by a Hydrophobic
Mechanism. Biochemistry 1996, 35, 8473-8477.
(36) In this regard, we have recently found in a different system that
3-vinylfarnesylation of the yeast a -factor lipopeptide pheromone
significantly alters its biological activity, relative to the native
farnesylated a -factor: Dawe, A. L.; Becker, J . M.; J iang, Y.;
Naider, F.; Eummer, J . T.; Mu, Y. Q.; Gibbs, R. A. Novel
Modifications to the Farnesyl Moiety of the a -Factor Lipopeptide
Pheromone from Saccharomyces cerevisiae - a Role for Isoprene
Modifications in Ligand Presentation. Biochemistry 1997, 36,
12036-12044.
(40) Scholten, J . D.; Zimmerman, K.; Oxender, M.; Sebolt-Leopold,
J .; Gowan, R.; Leonard, D.; Hupe, D. J . Inhibitors of Farnesyl:
Protein Transferase-A Possible Cancer Chemotherapeutic. Bioorg.
Med. Chem. 1996, 4, 1537-1543.
(37) Davisson, V. J .; Woodside, A. B.; Neal, T. R.; Stremler, K. E.;
Muehlbacher, M.; Poulter, C. D. Phosphorylation of Isoprenoid
Alcohols. J . Org. Chem. 1986, 51, 4768-4779.
J M9902786