Novel limonene phosphonate and farnesyl diphosphate analogues: Design, synthesis, and evaluation as potential protein-farnesyl transferase inhibitors

Article abstract of DOI:10.1016/S0968-0896(98)00202-8

Limonene and its metabolite perillyl alcohol are naturally-occurring isoprenoids that block the growth of cancer cells both in vitro and in vivo. This cytostatic effect appears to be due, at least in part, to the fact that these compounds are weak yet sel

Full text of DOI:10.1016/S0968-0896(98)00202-8

Bioorganic & Medicinal Chemistry 7 (1999) 241±250
Novel Limonene Phosphonate and Farnesyl Diphosphate
Analogues: Design, Synthesis, and Evaluation as Potential
Protein-Farnesyl Transferase Inhibitors
Je€rey T. Eummer,a,y Barbara S. Gibbs,b Todd J. Zahn,a
Judith S. Sebolt-Leopoldb and Richard A. Gibbsa,*
aDepartment of Pharmaceutical Sciences, College of Pharmacy and Allied Health Professions, Wayne State University,
528 Shapero Hall, Detroit, MI 48202, USA
bDepartment of Cell Biology, Parke±Davis Pharmaceutical Research, 2800 Plymouth Road, Ann Arbor, MI 48105, USA
Received 18 February 1998; accepted 2 September 1998
AbstractÐLimonene and its metabolite perillyl alcohol are naturally-occurring isoprenoids that block the growth of cancer cells
both in vitro and in vivo. This cytostatic e€ect appears to be due, at least in part, to the fact that these compounds are weak yet
selective and non-toxic inhibitors of protein prenylation. Protein-farnesyl transferase (FTase), the enzyme responsible for protein
farnesylation, has become a key target for the rational design of cancer chemotherapeutic agents. Therefore, several a-hydroxy-
phosphonate derivatives of limonene were designed and synthesized as potentially more potent FTase inhibitors. A noteworthy
feature of the synthesis was the use of trimethylsilyl tri¯ate as a mild, neutral deprotection method for the preparation of sensitive
phosphonates from the corresponding tert-butyl phosphonate esters. Evaluation of these compounds demonstrates that they are
exceptionally poor FTase inhibitors in vitro (IC50ꢀ3 mM) and they have no e€ect on protein farnesylation in cells. In contrast,
farnesyl phosphonyl(methyl)phosphinate, a diphosphate-modi®ed derivative of the natural substrate farnesyl diphosphate, is a very
potent FTase inhibitor in vitro (Ki=23 nM). # 1999 Elsevier Science Ltd. All rights reserved.
Introduction
has been made in the development of peptide-based
FTase inhibitors, and some of these compounds have
shown great promise in vivo as potential anti-cancer
agents.5 However, less work has been done on FPP-
based FTase inhibitors,3 and thus less is known regard-
ing the speci®city of FTase for its isoprenoid substrate.
The development of protein farnesylation inhibitors as
potential novel cancer chemotherapeutic agents has
been remarkably rapid. The initial evidence that pro-
teins are modi®ed with a mevalonate pathway inter-
mediate was presented scarcely a decade ago. Protein
farnesylation, the best-studied protein prenylation pro-
cess, is carried out by the enzyme protein-farnesyl
transferase (FTase), which recognizes the CAAX box
(where A=aliphatic and X=Ser or Met) at the car-
boxyl terminus of a protein substrate and then attaches
the farnesyl group from farnesyl diphosphate (FPP,
Figure 1) to the free sulfhydryl of the cysteine residue.1
Since the key signal transduction protein and oncogene
product Ras is farnesylated, FTase has been the subject
of intense research interest. Inhibitors of this enzyme
block the action of mutant Ras proteins and halt the
growth of ras transformed cells.2±4 Signi®cant progress
Concurrent with the discovery of protein farnesylation
was the observation that certain isoprenoid natural
products are e€ective, at high micromolar to low milli-
molar levels, in blocking the growth of cancer cells in
vitro and in vivo.6,7 Particular interest has been focused
on the dietary monoterpenes limonene and perillyl
alcohol (Fig. 1), which are currently undergoing clinical
trials as anticancer agents. These compounds have mul-
tiple modes of action in cells. They appear to act in part
by inhibiting protein prenylation,8 but there has been
some controversy about the cellular relevance of this
inhibition.9±11 However, work by Gelb et al. clearly
indicates that limonene, perillyl alcohol, perillic acid,
and perillic acid methyl ester are low millimolar to
micromolar inhibitors of mammalian and yeast FTase
in vitro.12 In particular, it was demonstrated that the
methyl ester is a FPP-competitive inhibitor of yeast
FTase. But the potency of these compounds is modest
compared to many synthetic FTase inhibitors, which
Key words: Antineoplastics; antitumor compounds; enzyme inhibi-
tors; phosphonic acids and derivatives; terpenes and terpenoids.
*Corresponding author: Tel: +1 313-577-3761; fax: +1 313-577-2033;
e-mail: rag@wizard.pharm.wayne.edu
yTaken in part from the M.S. thesis of J.T.E. Current address: Bristol±
Meyers Squibb Research Institute, 5 Research Parkway, Dept. 203,
Wallingford, CT 06492, USA.
0968-0896/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(98)00202-8

242
J. T. Eummer et al. / Bioorg. Med. Chem. 7 (1999) 241±250
Figure 1.
are e€ective in the nanomolar or even subnanomolar
range.3 For example, the FPP derivative a-hydroxy-
farnesylphosphonate (FMP, Figure 1), exhibits a Ki of
5 nM versus mammalian FTase.13 Despite this in vitro
potency, FMP exhibits poor selectivity and high toxicity
in cellular assays for protein prenylation inhibition.14
We hypothesized that combining the a-hydroxy-
phosphonate moiety of FMP with the cyclic mono-
terpene structure of limonene would generate
compounds more potent than the natural products, yet
potentially with the same cellular potency and selectiv-
ity. Thus we have synthesized compounds 1±4 and tes-
ted them in vitro and in situ as potential FTase
inhibitors. The limonene phosphonate analogue 5,
along with the aromatic a-hydroxyphosphonate 6 were
also prepared and tested. Although a number of limo-
nene derivatives have been evaluated as inhibitors of
cancer cell growth,15 these are the ®rst phosphonate
analogues of limonene reported. Finally, we have eval-
uated the readily prepared FPP analogue FPMP 7 as an
alternative diphosphonate modi®ed FTase inhibitor.
compounds have the same con®guration as the biologi-
cally active natural products d-limonene and (R)-perillyl
alcohol. However, (R)-perillyl alcohol and (R)-perillalde-
hyde, the starting materials needed to prepare 1 and 2,
were not readily available at the time this project was
initiated. Therefore, the (S)-isomers 3 and 4 were initially
targeted for synthesis in order to test and develop the
synthetic methodology. This also provided an opportu-
nity to evaluate the e€ect of isomerization on the
biological activity of the phosphonates. If the mono-
terpenes bind to FTase as conformationally restricted
analogues of the farnesyl moiety of FPP, then this iso-
merization may have a profound e€ect on the ability of
the hydroxyphosphonates to inhibit the enzyme.
Initial attempts were made to prepare 3 and 4 from
commercially available (S)-perillaldehyde 8 with the same
procedure used by the Merck group to synthesize FMP
from farnesal.13 The triethylamine-catalyzed coupling of
dimethyl phosphite with 8 proceeded smoothly to give
4. However, deprotection of 4 to give 3 was unsuccess-
ful. Using the original conditions of Pompliano et al.
(TMSBr in the presence of triethylamine), the methyl
protecting groups were removed but the terminal double
bond of the isopropylidene side chain was also removed,
possibly via addition of HBr to give the tertiary bro-
mide. Deprotection attempts using a variety of other
bases in place of triethylamine were also unsuccessful,16
Results
Synthesis
The primary targets of this project were the two (R)-
hydroxyphosphonate analogues 1 and 2, as these

J. T. Eummer et al. / Bioorg. Med. Chem. 7 (1999) 241±250
243
as was the use of conditions that generate TMSBr in situ
from TMSCl.17 The lack of success was surprising and
unanticipated, in view of the fact that other groups have
successfully used methyl protecting groups for the
synthesis of a variety of phosphonates.18 However, it is
known that perillyl alcohol and its derivatives are quite
susceptible to acid-catalyzed rearrangement reactions.19
The tert-butyl ester protecting group has been exten-
sively employed in peptide synthesis. It has the parti-
cular advantage in that it can be removed under acidic,
but aprotic conditions, which contrast strongly with the
conditions used for the deprotection for methyl and
benzyl esters.21,22 More recently, it has also been used as
a protecting group in the synthesis of various phospho-
nates. In particular, Burke and co-workers have used
the tert-butyl protecting group in the synthesis of aro-
matic, a-hydroxyphosphonate inhibitors of tyrosine
kinases.23 We therefore investigated the synthesis of
tert-butyl limonene-derived a-hydroxyphosphonates
and their deprotection.
We next examined the synthesis and deprotection of the
dibenzyl phosphonate limonene derivatives, speci®cally
the compound leading to the limonene phosphonate 5,
as a wide variety of deprotection conditions are avail-
able to remove benzyl groups. Again, the preparation of
the protected phosphonate was quite facile. Commer-
cially available (S)-perillyl alcohol 9 was readily trans-
formed to the chloride 10 using the Corey±Kim
procedure (Fig. 2).20 Coupling of 10 with the sodium
salt of dibenzyl phosphinate a€orded the desired diben-
zyl phosphonate 11 in virtually quantitative yield. As
with the dimethyl phosphonate 4, treatment with
TMSBr leads to removal of the protecting groups but
also destruction of the isopropylidene double bond. A
variety of hydrogenolysis conditions were used in an
attempt to deprotect 11, and all of them were unsuc-
cessful, leading to reduction of one or both double
bonds in 11 as well as removal of the benzyl groups. A
very mild transfer hydrogenation protocol did a€ord
the monobenzyl derivative 12, but further treatment of
12 again led to reduction of the terminal double bond.
Deprotection was also attempted using hydrolytic con-
ditions. The dibenzyl ester was re¯uxed under basic
conditions (1 M KOH), but this resulted in only partial
hydrolysis after 5 h, and further or more vigorous
treatment led to decomposition. Thus it appeared that a
di€erent protecting group that allowed for an alter-
native deprotection protocol would be needed to pre-
pare the desired limonene phosphonates.
The synthesis of the aromatic analogue 6 via the tert-
butyl protected compound 14 was ®rst attempted
(Fig. 3). This compound was chosen as a model system
closely analogous to the a-hydroxyphosphonates pre-
pared by Burke, and also as an aromatic analogue of
the hydroxyphosphonates 1 and 3. Alumina-mediated
coupling of 4-isopropylbenzaldehyde 13 with di-tert-
butylphosphite a€orded the protected a-hydroxyphos-
phonate 14. Deprotection with tri¯uoroacetic acid/ani-
sole a€orded the desired aromatic phosphonate 6 in
excellent yield. Next, the synthesis of 3 using the same
procedure was attempted. Although preparation of 15
using alumina as a base was not successful, it was syn-
thesized in modest yield from 8 using both sodium eth-
oxide and triethylamine as bases. Next, deprotection of
15 was attempted using TFA/anisole as for 14. How-
ever, compound 3 was not obtained, and instead it
appeared that the limonene skeleton had undergone
acid-catalyzed double bond migration, possibly along
with more profound skeletal rearrangements.19 The
replacement of anisole with the alternative scavenger
triethylsilane did not change the results obtained with
TFA.19,22,24 However, the tert-butyl protecting groups
Figure 2.

244
J. T. Eummer et al. / Bioorg. Med. Chem. 7 (1999) 241±250
were successfully and selectively removed from 15 using
trimethylsilyl tri¯uoromethanesulfonate in the presence
of 2,6-lutidene.25 The limonene phosphonate 5 was also
prepared in an analogous manner via deprotection of
the tert-butyl phosphonate 16 (Fig. 3). Our results sug-
gest that this mild, non-acidic method may provide a
valuable alternative route for the preparation of sensi-
tive phosphonate derivatives. Note that the synthesis of
16 required preformation of the di-tert-butylphos-
phinate anion with sodium hydride, prior to its coupling
with allylic chloride 10.
smoothly a€orded (R)-perillaldehyde 17. The aldehyde
17 was coupled with dimethyl phosphinate, in the same
manner as described for 4, to give compound 2. The
tert-butylphosphonate 18 was prepared in the same
manner as described above for 15. Deprotection of 18
with trimethylsilyl tri¯uoromethanesulfonate a€orded
the desired hydroxyphosphonate 1.
A di€erent synthetic route was used to prepare the FPP
analogue FPMP (7, Fig. 5). This procedure was devel-
oped by Poulter and co-workers to prepare the geranyl
(C10) analogue of 7. Commercially available trans,trans-
farnesyl chloride (19) was coupled with the tris-tetra-
butylammonium salt 20 in acetonitrile. Ion-exchange
chromatography followed by reversed phase HPLC
With the synthetic procedure developed, we next turned
to the preparation of the (R)-isomers 1 and 2 (Fig. 4).
Oxidation of the allylic alcohol with manganese dioxide
Figure 3.
Figure 4.

J. T. Eummer et al. / Bioorg. Med. Chem. 7 (1999) 241±250
245
and IC50 values of 3.0 mM and ꢁ10 mM were deter-
mined. While the actual values may be lower due to
solubility problems with the limonene analogues (parti-
cularly with the more hydrophobic compound 2), it is
clear that these compounds are several orders of mag-
nitude less potent than FMP, which exhibits an IC50 of
30 nM under the same conditions. Despite their lack of
potency versus isolated mFTase, compounds 3±7 were
tested in a mobility shift assay to determine their ability
to interfere with the farnesylation of Ras proteins in
cells. At low micromolar concentrations, these com-
pounds had no e€ect on the farnesylation of Ras, and at
higher concentrations 4 and 5 exhibited nonspeci®c cel-
lular toxicity. Even at these higher concentrations
(100 mM), there was no indication of any e€ect on Ras
protein farnesylation, consistent with the lack of in vitro
potency.
Figure 5.
The in vitro biological results obtained with the dipho-
sphate-modi®ed FPP analogue FPMP were strikingly
di€erent. This compound is a very potent FPP-based
mFTase inhibitor, with an IC50 value only ®vefold
higher than that of FMP. This compound also exhibits
good selectivity for mFTase versus mGGTase I, a
potentially desirable feature for a therapeutically useful
mFTase inhibitor.3 As expected, this close structural
mimic of the natural substrate FPP is an FPP-competi-
tive inhibitor (Fig. 6). FPMP was also evaluated as a
potential inhibitor of Ras farnesylation in cells, but it
has no e€ect at 30 mM. This might be due to FPMP
acting as an alternative substrate for mFTase, leading to
farnesylation of Ras in the same manner as seen with
the natural substrate FPP. However, preliminary eva-
luation of FPMP in an HPLC mFTase assay indicates
that it is an exceptionally poor alternative substrate. No
farnesylated peptide product was seen under conditions
where ꢁ5% of the product obtained with FPP would
have been clearly observed. Note that Stremler and
Poulter have demonstrated that the corresponding ger-
anyl analogue of FPMP is a poor alternative substrate
for FPP synthase.27
puri®cation a€orded pure FPMP 7 in good yield. The
HPLC retention time of 7 was similar to that of the
natural product FPP, and its proton and phosphorous
NMR spectra were consistent with those expected for
FPMP.
Biological assays
The seven potential FTase inhibitors 1±7 shown in Fig-
ure 1 were tested for their ability to inhibit mammalian
FTase in vitro using a scintillation proximity assay as
previously described (Table 1).26 Compounds 3±7 were
also evaluated for their ability to block protein farnesyl-
ation in cells through a mobility shift assay, again using
a previously developed procedure.26 Surprisingly, 3±6
exhibited no signi®cant inhibition of mFTase at 1 mM,
and these compounds were not evaluated further in vitro.
The (R)-isomers 1 and 2 were tested more extensively,
Table 1 Biological assay of analoguesa
Analogue
IC50 FTase
IC50 GGTase Ki FTase
Mobility
shift assay
Discussion
1
2
3
4
5
6
7
3.0 mM
ꢁ10 mM
The dietary monoterpenes limonene and perillyl alcohol
(Fig. 1) are e€ective, at high micromolar to low milli-
molar levels, in blocking the growth of cancer cells in
vitro and in vivo.6,7 Particular interest has been focused
on these compounds, which are currently undergoing
clinical trials as anticancer agents. An initial report by
Gould indicated that these compounds act as anticancer
agents by inhibiting protein prenylation.8 Further work
by Gould in collaboration with Gelb, Tamanoi, and
their groups demonstrates that limonene, perillyl alco-
hol, perillic acid, and perillic acid methyl ester are low
millimolar to micromolar inhibitors of mammalian and
yeast FTase in vitro.12 In particular, these workers
showed that the methyl ester is a FPP-competitive inhi-
bitor of yeast FTase. These reports, in combination with
the demonstrated in vitro mFTase inhibition seen with
FMP, led us to synthesize and assay the phosphonate
analogues 1±5.
>1.0 mMb
>1.0 mMb
>1.0 mMb
>1.0 mMb
160 nm
>25 mMc
>100 mMd
>100 mMd
>25 mMc
>30 mMc
14.2 mm
23 nm
aConditions: FTase values were determined using recombinant
mFTase in a scintillation proximity assay with tritiated FPP and the
peptide Biotin-Aha-TKCVIM as substrates, in the same manner as
previously described (see experimental section). GGTase I values were
determined in a similar manner using recombinant mGGTase I, tri-
tiated GGPP, and the peptide Biotin-Aha-TKCVIL. The mobility
shift assays (determination of the ability of the analogue to block the
prenylation of H-Ras in NIH3T3 rat ®broblasts) were performed in
the same manner as previously described.
bAt the indicated concentration, the compound had no e€ect on the
rate of mFTase-catalyzed prenylation.
cAt the indicated concentration, the analogue had no e€ect on the
prenylation of Ras in cells.
dAt the indicated concentration, the analogue had no e€ect on the
prenylation of Ras cells, and exhibited non-speci®c toxicity.

246
J. T. Eummer et al. / Bioorg. Med. Chem. 7 (1999) 241±250
Figure 6. Lineweaver±Burk plot of competitive mFTase inhibition by FPMP (7). The values were obtained using the scintillation proximity assay as
described in the experimental section, with the exception that the peptide concentration was maintained at 0.10 mM and the FPP concentration was
varied. Enzyme rates were determined at 0, 0.02 mM, 0.05 mM, 0.10 mM, 0.20 mM, and 0.50 mM FPMP, as indicated on the ®gure.
However, it has become clear that limonene and its
derived monoterpenes such as perillyl alcohol have
multiple modes of action in cells. In particular, there has
been some controversy about the cellular relevance of
mFTase inhibition seen with limonene and its deriva-
tives. Hohl and Lewis have found that perillyl alcohol,
rather than blocking farnesylation, decreases total cel-
lular Ras levels.10 Two other groups have con®rmed
that the growth inhibition of transformed cells seen with
monoterpenes is not due to blockage of Ras prenyla-
tion, and furthermore is not dependent on the muta-
tional status of Ras in the limonene-treated cells.9,11
Very recently, Gould and co-workers also con®rmed
that perillyl alcohol does not inhibit Ras farnesylation
in cells, but reported that instead it interferes with pro-
tein geranylgeranylation in cells.28 In view of these more
recent reports, the lack of activity of 1±6 versus mFTase
in vitro and in the mobility shift assay may not be sur-
prising. Our results indicate that the inhibition of
mFTase by perillyl alcohol and perillic acid methyl ester
is not due to the ability of limonene to act as a con-
formationally restricted mimic of the farnesyl moiety of
FPP.
mFTase inhibitor. This ®nding is consistent with pre-
vious reports that FMP and other farnesyl-bearing FPP
analogues such as 2129 and 2230 are also potent inhibi-
tors of this enzyme (Fig. 7). However, note that the
oxygen at C1 of the farnesyl structure appears to be a
key recognition element that signi®cantly increases
binding of the analogue to the enzyme; compare the
IC50 values for 22 and the `all-carbon' analogue 2313 to
7. This ®nding is also consistent with an active site
model suggested by the Bristol±Meyers Squibb group,
which proposes that this oxygen interacts with an acidic
residue in the mFTase active site.29
The recently disclosed crystal structure of mFTase indi-
cates a deep, hydrophobic pocket lined with aromatic
residues, which apparently binds the farnesyl moiety of
FPP in an extended conformation.31 Such a binding site,
which would provide many contacts with the hydro-
carbon chain of FPP, could explain the relatively tight
binding of all of the compounds shown in Figure 7 to
mFTase. The structure of the putative FPP binding site
may also provide a clue to the very weak inhibition seen
with 1. If the hydroxyphosphonate moiety of this com-
pound binds to the zinc ion in the active site, then the
crystal structure model indicates that the monoterpene
portion of 1 would only be able to make limited
In sharp contrast to the results seen with the limonene
phosphonate analogues, FPMP (7) is a very potent

J. T. Eummer et al. / Bioorg. Med. Chem. 7 (1999) 241±250
247
molecular sieves. Brine refers to saturated aqueous
sodium chloride. Normal phase ¯ash column chroma-
tography was performed with silica gel 60 (230±440
mesh, E. Merck). Thin-layer chromatography was con-
ducted on precoated alumina plates (silica gel 60 F-254,
E. Merck), and was visualized by ultraviolet illumination
or treatment with 8% phosphomolybdic acid (PMA) in
ethanol, followed by heating. Reversed phase HPLC
was conducted on a Waters system with a Vydac pH-
stable C8 4.6Â250 mm column, a U6K injector, two 501
pumps and a 490E multiwavelength UV detector. Melt-
ing points were taken on a MEL-TEMP II melting
point apparatus using FLUKE 51 K/J thermometer and
are uncorrected. Nuclear magnetic resonance spectra
were recorded at either 300 MHz (1H NMR), 76.5 MHz
(13C NMR) or 121 MHz (31P NMR) on a General
Electric QE-300 or GN-300 spectrometer. Chemical
shifts are reported in parts per million (d) down®eld
from tetramethylsilane (TMS) using an internal TMS
standard. Multiplicities are indicated by the following
symbols: (br) broad, (s) singlet, (d) doublet, (t) triplet,
(q) quartet, (m) multiplet. Coupling constants (J) are
reported in Hertz (Hz). Infrared spectra were recorded
on a Nicolet 550 Magna-IRTM spectrometer and are
reported in reciprocal centimeters (cm 1). Mass spectra
were recorded on a Kratos MS 80 RFA mass spectro-
meter at 70 eV. Microanalysis were performed by Gal-
braith Laboratories, Inc., Knoxville, TN, USA.
(S)-(p-Menth-1,8-dien-7-ol-7-yl) phosphonic acid, dimethyl
ester (4). In a dry 25 mL round-bottom ¯ask, 2.0 mL of
CH3CN, (S)-( )-perillaldehyde 8 (151 mg, 0.16 mL,
1.0 mmol), Et3N (0.28 mL, 2.0 mmol) and dimethyl
phosphite (0.14 mL, 1.5 mmol) were combined. The
solution was stirred for 24 h under Ar. The solvent was
removed and the crude mixture was puri®ed by
reversed-phase silica gel ¯ash chromatography (70%
methanol/water) to give the desired compound as an oil
Figure 7.
contacts with the hydrophobic pocket. This interpreta-
tion implies that FPP-competitive inhibitors of mFTase
must have hydrocarbon moieties that closely mimic the
farnesyl chain in hydrophobic bulk, if not in actual
structure. This is in fact the case with two recently dis-
closed series of potent, aromatic-derived, FPP-competi-
tive mFTase inhibitors.32,33,37
1
(ꢁ350 mg, ꢁ100%). H NMR (CDCl3) d 1.25 (d, 1H),
1.74 (s, 3H), 3.79 (d, 6H), 4.37 (d, J=12 Hz, 1H), 4.72
(br s, 1H), 5.93 (s, 1H); 13C NMR (CDCl3) d 20.70,
25.57, 25.97, 27.36, 27.52, 30.57, 40.69, 53.54, 70.66,
71.00, 72.77, 73.10, 108.72, 125.27, 126.02, 133.00,
133.46, 149.43; IR (NaCl) 3396, 2953, 2851, 2366, 1643,
1
Experimental
General
1055 cm
m/z calcd for C12H21PO4, 260.1179 (M+); found,
260.1181.
;
31P NMR (CDCl3) d 25.40, 25.45; HREIMS
All moisture and/or sensitive reactions were conducted
in ¯ame or oven dried glassware under a dry argon
atmosphere. Moisture sensitive reagents were trans-
ferred by syringe and introduced to the reaction vessel
through rubber septum caps. All reaction contents were
stirred magnetically. Solvents, organic and inorganic
reagents were obtained from the following suppliers:
Aldrich Chemical Company, Fisher Scienti®c, and
VWR Scienti®c, and all reagents were used without
further puri®cation, unless otherwise stated. The fol-
lowing solvents were prepared anhydrous prior to use as
follows: THF and DME were distilled from Na/benzo-
phenone; CH2Cl2 was distilled from CaH2; PhH was
distilled from Na; CH3CN was freshly distilled from
phosphorous pentoxide; toluene, DMF, hexanes and
diethyl ether were dried and stored over activated 3 A
(S)-p-Menth-7-chloro-1,8-diene (10). In a 25 mL round-
bottom ¯ask, n-chlorosuccinimide (NCS; 588 mg,
4.4 mmol) was dissolved in 10 mL of CH2Cl2. The mix-
ture was cooled to 0 ꢂC and Me2S (0.352 mL; 4.8 mmol)
was added dropwise. The mixture was cooled to 20 ꢂC
and (S)-( )-perillyl alcohol (9; 0.62 mL, 3.9 mmol) was
slowly added. The solution was warmed to 0 ꢂC, stirred
for 1 h, quenched with 10 mL of cold brine, and dried
over MgSO4. The crude product was puri®ed by ¯ash
chromatography on silica gel (10% EtOAc/hexane) to
1
give a light yellow oil (654 mg, 96% yield). H NMR
(CDCl3) d 1.74 (s, 3H), 2.17 (br s, 4H), 4.01 (s, 2H), 4.72
(d, J=6 Hz, 2H), 5.84 (br s 1H); IR (NaCl) 3089, 2945,
1643 cm 1; HREIMS m/z calcd for C10H15Cl, 170.0862
(M+); found, 170.0861.

248
J. T. Eummer et al. / Bioorg. Med. Chem. 7 (1999) 241±250
(S)-(p-Menth-1,8-dien-7-yl) phosphonic acid, dibenzyl ester
(11). In a 25 mL round-bottom ¯ask, NaH (192 mg,
8.0 mmol, 60% oil dispersion) was suspended in 5 mL of
dry THF. The ¯ask was cooled to 10 ꢂC and dibenzyl
phosphite (1.44 mL, 6.5 mmol) was slowly added. After
15 min, the mixture was transferred to a ¯ask cooled to
0 ꢂC containing the chloride 10 (411 mg, 2.4 mmol) in
5 mL of THF and HMPA (0.35 mL, 2 mmol) over
10 min. The solution was stirred for 16 h at rt. THF was
removed and the residue was then diluted with water
and extracted with CCl4. The organic layer was col-
lected and washed with 5% citric acid, 5% NaHCO3
and brine. The product was dried over MgSO4 and sol-
vent was removed. Mineral oil was removed by stirring
the product in hexane at 78 ꢂC under Ar and the
supernatant was removed by pipette. The crude product
was then puri®ed by reversed-phase silica gel ¯ash
chromatography (90% methanol/water) to give a clear
solvent was removed to yield a brownish oil. The oil was
crystallized from MeOH/CHCl3 to give a brownish ¯u€y
solid (336 mg, 97%). 1H NMR (CD3OD) d 1.23 (d, 6H),
2.89 (m, 1H), 4.83 (dd, 1H), 7.21 (d, 2H), 7.42 (d, 2H);
13C NMR (CD3OD) d 24.6 (d), 35.34, 71.2 (d), 73.5 (d),
127.15, 128.80, 137.48, 149.68, 163.91; 31P NMR
((CD3)2SO) d 15.2; MSEI m/z 263.1.
(S)-(p-Menth-1,8-dien-7-ol-7-yl) phosphonic acid, di-tert-
butyl ester (15). In a dry 50 mL ¯ask, di-tert-butyl
phosphite (583 mg, 3.0 mmol), (S)-( )-perillaldehyde (8,
0.47 mL, 3 mmol), 4.0 mL of PhH and 0.40 mL of tri-
ethylamine were combined under Ar. A 0.07 M solution
of sodium ethoxide (204 mg in 5.0 mL of ethanol) was
slowly added to the mixture. The mixture was stirred for
15 min and 0.40 mL of triethylamine was added. After
an additional 30 min the solvent was removed and the
crude product was puri®ed by ¯ash chromatography
(5% MeOH/CHCl3) to yield a clear oil (362 mg, 35%
1
oil (940 mg, 98% yield). H NMR (300 MHz, CDCl3) d
1
1.71 (s, 3H), 2.52 (d, J=21 Hz, 2H), 4.68 (d, J=8 Hz,
2H), 4.98 (m, 4H), 5.56 (s, 1H), 7.34 (br s, 10H); 13C
NMR (75.4 MHz, CDCl3) d 20.77, 27.69, 29.93, 30.88,
33.15, 34.32, 34.97, 36.13, 40.46, 67.24, 67.57, 108.62,
yield). H NMR (CDCl3) d 1.50 (s, 18 H), 1.79 (s, 3H),
2.58 (br s, 1H), 4.15 (d, 1H), 4.72 (s, 2H), 5.82 (s, 1H);
13C NMR (75.4 MHz, CDCl3) d 20.81, 25.88 (d), 27.54,
30.50, 41.04, 72.9 (d), 74.95 (d), 82.9 (dd), 108.62, 123.7
(d), 124.4 (d), 134.0 (d), 149.89; 31P NMR (121 MHz,
CDCl3) d 15.68, 15.42; EIMS, 345.
126.50, 126.92, 127.89, 128.28, 128.51, 129.90; IR (NaCl)
1
3064, 3038, 2919, 2425, 1643, 1455, 1268, 1013 cm
NMR (121 MHz, CDCl3) d 29.35; MS-EI: 396.20.
;
31P
(S)-(p-Menth-1,8-dien-7-ol-7-yl) phosphonic acid (3). In
a dry 25 mL ¯ask, di-tert-butyl phosphonate 15 (307 mg,
0.89 mmol) and 2,6-lutidine (0.31 mL, 2.68 mmol) were
dissolved in 4 mL of CH2Cl2 under Ar. The ¯ask was
cooled to 0 ꢂC and trimethylsilyl tri¯uoromethane-
sulfonate (TMSOTf; 0.40 mL, 2.0 mmol) was added.
The reaction was stirred for 1 h. Solvent was then
removed and the reaction was quenched with methanol.
The crude product was washed twice with toluene and
following concentration it was puri®ed by reversed-
phase HPLC using a program of 5 min of 100% A fol-
lowed by a linear gradient of 100% A to 100% B over
30 min (A, 25 mM aqueous ammonium bicarbonate (pH
8.0); B, acetonitrile), with a ¯ow rate of 1 mL (UV
monitoring at 214 and 230 nm). The fractions containing
the product were lyophilized to yield a white ¯u€y solid
(S)-(p-Menth-1,8-dien-7-yl) phosphonic acid, monobenzyl
ester (12). In a dry 25 mL ¯ask, 403 mg of the di-benzyl
phosphonate 12 (1.02 mmol), 0.94 mL of cyclohexadiene
(10 mmol), 796 mg of 10% Pd±C catalyst and 4 mL of
dry ethanol were combined. The solution was stirred for
2 h under Ar. The solid was ®ltered through a plug of
Celite and washed three times with ethanol. The solvent
was removed to give an oily liquid (247 mg, 79% yield).
1H NMR (300 MHz, CDCl3) d 1.70 (s, 3H), 2.48 (d,
J=21 Hz, 2H), 4.21 (narrow m, 1H), 4.68 (d, J=4 Hz,
2H), 4.96 (m, 2H), 5.60 (narrow m, 1H), 7.34 (narrow
1
m, 5H); IR (NaCl) 2327, 1643, 1497, 1457, 734 cm
.
[4-(Isopropyl)phenyl]-hydroxymethyl phosphonate, di-tert-
butyl ester (14). In a 50 mL ¯ask, 4-(isopropyl)-
benzaldehyde (13, 1.52 mL, 10 mmol), di-tert-butyl
phosphite (Lancaster Chemical Co.; 1.94 g, 10 mmol),
and Woelm basic alumina oxide (4.0 g, activity I) were
combined and stirred at room temperature overnight.
The crude product was triturated with CH2Cl2/EtOH
(9/1, 4Â10 mL). The solvent was removed and the
resultant oil was crystallized from petroleum ether to
give the product as a light brown solid (mp 105±106 ꢂC;
1
(47 mg, 20% yield). H NMR (CD3OD) d 1.72 (s, 3H),
4.08 (d, 1H), 4.68 (narrow m, 2H), 5.77 (narrow m, 1H);
13C NMR (CD3OD) d 20.98, 27.45 (d), 29.30, 32.16 (d),
42.81, 53.68, 54.85, 55.9 (d), 56.91, 109.03, 124.36,
137.85 (d), 151.65; 31P NMR (CD3OD) d 20.25, 20.16.
(S)-(p-Menth-1,8-dien-7-yl) phosphonic acid, di-tert-butyl
ester (16). In a 25 mL round-bottom ¯ask, NaH
(192 mg, 8 mmol, 60% oil dispersion) was suspended in
5 mL of THF. The ¯ask was cooled to 10 ꢂC and di-
tert-butyl phosphite (777 mg, 4.0 mmol) was slowly
added. After 15 min, the mixture was transferred to a
¯ask cooled to 0 ꢂC containing the chloride 10 (525 mg,
3.1 mmol) in 5 mL of THF and 0.35 mL of HMPA
(2 mmol) over 10 min. The solution was stirred for 16 h
at room temperature, concentrated, diluted with water,
and extracted with CCl4. The organic layer was washed
with 5% citric acid, 5% NaHCO3 and brine, dried over
MgSO4, and concentrated. The crude product was pur-
i®ed by ¯ash chromatography (5% MeOH/CHCl3) to
1
1.83 g, 54% yield). H NMR (CDCl3) d 1.22 (d, 6H),
1.39 (d, 18H), 2.84 (m, 1H), 4.80 (dd, 1H), 7.18 (d, 2H),
7.34±7.37 (dd, 2H); 13C NMR (CDCl3) d 24.02, 30.31,
33.83, 71.33, 73.48, 83.26, 125.92, 127.15, 134.75,
148.33; Elemental analysis: C±53.34% (calcd), 53.04%
(exptl); H±7.22% (calcd), 6.85% (exptl); P±13.04%
(calcd), 12.51% (exptl).
[(4-Isopropyl)phenyl]hydroxymethyl] phosphonic acid (6).
In a 50 mL ¯ask, 14 (500 mg, 1.5 mmol), TFA (10 mL),
anisole (0.82 mL, 7.5 mmol), and CH2Cl2 (10 mL) were
combined and stirred at room temperature for 45 min.
The reaction was quenched with 10 mL of H2O and the
1
yield 16 as a clear oil (218 mg, 22% yield). H NMR

J. T. Eummer et al. / Bioorg. Med. Chem. 7 (1999) 241±250
249
(CDCl3) d 1.52 (s, 18H), 1.74 (s, 3H), 2.38 (d, 2H), 4.71
(s, 2H), 5.54 (s, 1H); 13C NMR (CDCl3) d 21.1 (d),
29.23, 30.90, 32.22, 39.28, 41.18, 42.27, 83.76, 109.41,
126.64, 130.63, 151.00; 31P NMR (CDCl3) d 20.31;
HREIMS calcd for C18H33O3P, 328.2167; found,
328.2177.
(152 mg, 1.0 mmol) and MnO2 (1.3 g, 15 mmol) were
stirred in 22 mL of hexane for 2 h at room temperature.
The solution was then ®ltered through Celite, which was
washed three times with ether. Concentration gave
1
crude (R)-( )-perillaldehyde 17 as a clear oil; H NMR
indicated a ꢁ5:1 mixture of the aldehyde and alcohol,
which was used directly in the next reaction. In a dry
25 mL round-bottom ¯ask, 2.0 mL of CH3CN, crude 17
(ꢁ150 mg), Et3N (0.28 mL, 2.0 mmol) and dimethyl
phosphite (0.14 mL, 1.5 mmol) were combined. The
solution was stirred for 24 h under Ar. The solvent was
removed and the crude product was puri®ed by ¯ash
chromatography (5% MeOH/CHCl3) to yield a clear oil
(S)-(p-Menth-1,8-dien-7-yl) phosphonic acid (5). In a dry
25 mL ¯ask, di-tert-butyl phosphonate 16 (106 mg,
0.410 mmol) and 2,6-lutidine (0.143 mL, 1.23 mmol)
were dissolved in 2 mL of CH2Cl2 under Ar. The ¯ask
was cooled to 0 ꢂC and trimethylsilyl tri-
¯uoromethanesulfonate (TMSOTf; 0.18 mL, 0.92 mmol)
was added. The reaction was stirred for 1 h, the solvent
was removed and the reaction was quenched with
methanol. The crude product was washed twice with
toluene, concentrated, taken up in water, and converted
to the ammonium salt by ion exchange. Lyophilization
a€orded the desired phosphonate as a white ¯u€y solid
1
(126 mg, 48% yield from (R)-perillyl alcohol). The H
NMR spectrum of 2 was identical to that of the enan-
tiomer 4.
Farnesyl phosphonyl(methyl)phosphonate (FPMP, 7).
Tris(tetrabutylammonium) hydrogen methyldiphos-
phate 20 (1.35 mmol, 548 mg; prepared as described by
Poulter and co-workers34 from methanediphosphonic
acid (Alfa Chemical Co.)) was dissolved in CH3CN
(4.5 mL) under an argon atmosphere. Commercial all-
trans-farnesyl chloride (19, 0.21 mmol, 50 mg) was
added to the resulting solution and the reaction was
stirred at rt for 2.5 h. The reaction mixture was then
concentrated (rotary evaporation), the residue was dis-
solved in 3 mL of deionized water, and the resulting
solution was passed through a 2Â8 cm Dowex AG50Â8
ion exchange column (NH4+ form). Lyophilization then
yielded a pale white solid, which was dissolved in 2 mL
of 25 mM ammonium bicarbonate. This resulting mix-
ture was puri®ed by reversed-phase HPLC using a pro-
gram of 5 min of 100% A followed by a linear gradient
of 100% A to 100% B over 30 min (A, 25 mM aqueous
ammonium bicarbonate (pH 8.0); B, acetonitrile), with
a ¯ow rate of 1 mL. Under these conditions, FPMP
exhibited a retention time of 20 min (UV monitoring at
214 and 230 nm). The fractions containing FPMP were
collected, the acetonitrile was removed via rotary eva-
poration, and lyophilization a€orded 48 mg (61%) of
1
(21 mg, 21%). H NMR (CD3OD) d 1.48 (m, 1H), 1.71
(s, 3H), 2.29 (d, 2H), 4.68 (narrow m, 2H), 5.52 (br s,
1H). 13C NMR (CD3OD) d 21.17, 29.50, 31.35, 32.35,
38.62, 40.35, 42.50, 108.96, 123.80, 133.36, 151.72; 31P
NMR (CD3OD) d 25.68.
(R)-(p-Menth-1,8-dien-7-ol-7-yl) phosphonic acid, di-tert-
butyl ester (18). In a dry 50 mL ¯ask, (R)-perillyl alco-
hol (commercial material from Aldrich; 152 mg,
1.0 mmol) and MnO2 (1.3 g, 15 mmol) were stirred in
22 mL of hexane for 2 h at room temperature. After
stirring, the solution was ®ltered through Celite, which
was then washed three times with ether. Concentration
followed by ¯ash chromatography (95% hexane/ethyl
acetate) gave (R)-( )-perillaldehyde 17 as a clear oily
1
liquid (97 mg, 64% yield). H NMR (CDCl3) d 1.76 (s,
3H), 4.74 (d, 2H), 6.83 (narrow m, 1H), 9.44 (s, 1H). In
a dry 25 mL ¯ask, di-tert-butyl phosphite (125 mg,
0.65 mmol), 17 (97 mg, 0.65 mmol), 1.5 mL of PhH and
0.15 mL of triethylamine were combined. A 0.5 M solu-
tion of sodium ethoxide (68 mg in 2.0 mL of ethanol)
was slowly added to the mixture. The mixture was stir-
red for 15 min. Then 0.15 mL of triethylamine was
added and the reaction was stirred for 30 min. The sol-
vent was removed and the crude product was puri®ed
by ¯ash chromatography (5% MeOH/CHCl3) to yield a
1
pure FPMP 7 as a white ¯u€y solid. H NMR (D2O) d
1.62 (s, 6H), 1.69 (two s, 6H), 2.02 (t, 2H, P-CH2-P,
partially buried), 2.1 (m, 8H), 4.41 (app t, 2H), 5.19 (m,
2H), 5.45 (t, 1H); 31P NMR (D2O) d 13.5 (d), 20.8 (d).
1
clear oil (201 mg, 91% yield). H NMR (CDCl3) d 1.44
(s, 18H), 1.70 (s, 3H), 3.40 (m, 1H), 4.10 (m, 1H), 4.68
(app s, 2H), 5.79 (narrow m, 1H); 13C NMR (CDCl3) d
20.70 (d), 25.85 (d), 27.56, 30.47, 41.00, 72.57 (d), 74.68
(d), 82.80, 108.55, 123.49 (d), 124.30 (d), 134.26 (d),
149.79.
Prenyltransferase assays. FTase IC50 values were deter-
mined using recombinant mFTase35 in a scintillation
proximity assay (using streptavidin beads from Amer-
sham) with tritiated FPP (speci®c activity 15±30 Ci/
mmol, ®nal concentration 0.12 mM) and the peptide
Biotin-Aha-Thr-Lys-Cys-Val-Ile-Met-OH (®nal con-
centration 0.1 mM) as substrates, in the same manner as
previously described.36 The Ki values were determined
using the same assay system with varying concentra-
tions of tritiated FPP. GGTase I values were deter-
(R)-(p-Menth-1,8-dien-7-ol-7-yl) phosphonic acid (1). The
phosphonate 18 was converted to the ammonium salt as
described above for 15 to yield 1 as a white ¯u€y solid.
This compound exhibited the same NMR data as its
enantiomer 3. Elemental analysis: C±23.48% (calcd),
23.60% (exptl). H±4.90% (calcd), 4.75% (exptl). N±
6.58% (calcd), 6.61% (exptl). P±3.93% (calcd), 4.17%
(exptl).
mined in
mGGTase I, tritiated GGPP, and the peptide Biotin-
a
similar manner using recombinant
Aha-Thr-Lys-Cys-Val-Ile-Leu-OH in
proximity assay.
a scintillation
(R)-(p-Menth-1,8-dien-7-ol-7-yl) phosphonic acid, dimethyl
ester (2). In a dry 50 mL ¯ask, (R)-perillyl alcohol
Mobility shift assay. This procedure was performed as
previously described from this group.26 Western blotting

250
J. T. Eummer et al. / Bioorg. Med. Chem. 7 (1999) 241±250
detection of the Ras protein used pan-Ras Ab-2 (Onco-
gene Science) and an anti-mouse HRP conjugate sec-
ondary antibody (Amersham). Blots were developed
using ECL techniques (Amersham).
17. Morita, T.; Okamoto, Y.; Sakurai, H. Bull. Chem. Soc.
Jpn. 1981, 54, 267.
18. Biller, S. A.; Forster, C. Tetrahedron 1990, 46, 6645.
19. Kayahara, H.; Udea, H.; Ichimoto, I.; Tatsumi, C. Agric.
Biol. Chem. 1970, 34, 1597.
20. Corey, E. J.; Kim, C. U.; Takeda, M. Tetrahedron Lett.
1972, 4339.
Acknowledgements
21. Leftheris, K.; Kline, T.; Vite, G. D.; Cho, Y. H.; Bhide, R.
S.; Patel, D. V.; Patel, M. M.; Schmidt, R. J.; Weller, H. N.;
Andahazy, M. L.; Carboni, J. M.; Gullo-Brown, J. L.; Lee, F.
Y. F.; Ricca, C.; Rose, W. C.; Yan, N.; Barbacid, M.; Hunt, J.
T.; Meyers, C. A.; Seizinger, B. R.; Zahler, R.; Manne, V. J.
Med. Chem. 1996, 39, 224.
22. Qian, Y.; Vogt, A.; Sebti, S. M.; Hamilton, A. D. J. Med.
Chem. 1996, 39, 217.
23. Burke, T. R.; Li, Z.-H.; Bolen, J. B.; Marquez, V. E. J.
Med. Chem. 1991, 34, 1577.
24. Pearson, D. A.; Blanchette, M.; Baker, M. L.; Guindon,
C. A. Tetrahedron Lett. 1989, 30, 2739.
25. Borgulya, J.; Bernauer, K. Synthesis 1980, 545.
26. Scholten, J. D.; Zimmerman, K.; Oxender, M.; Sebolt-
Leopold, J.; Gowan, R.; Leonard, D.; Hupe, D. J. Bioorg.
Med. Chem. 1996, 4, 1537.
This research was supported by the American Associa-
tion of Colleges of Pharmacy/Burroughs±Wellcome
Fund Grant Program for New Investigators, NSF grant
CHE-9213312, and by NIH grant CA-67292. R.A.G. is
the recipient of an American Cancer Society Junior
Faculty Research Award (JFRA-609). J.T.E. was sup-
ported in part by a GRA award from the WSU MBRS
program (GM-08167), and T.J.Z. was supported in part
by a GRA award from WSU. We also thank Professor
John Frost (Michigan State) and Dr. Hollis Showalter
(Parke±Davis) for helpful advice on the synthesis and
deprotection of tert-butyl phosphonates, and Michael
Latash (Parke±Davis) for technical assistance with the
in situ assays.
27. Stremler, K. E.; Poulter, C. D. J. Am. Chem. Soc. 1987,
109, 5542.
28. Ren, Z.; Elson, C. E.; Gould, M. N. Biochem. Pharmacol.
1997, 54, 113.
References and Notes
1. Zhang, F. L.; Casey, P. J. Annu. Rev. Biochem. 1996, 65,
241.
2. Cox, A. D.; Der, C. J. Biochem. Biophys. Acta 1997, 1333,
F51.
3. Leonard, D. M. J. Med. Chem. 1997, 40, 2971.
4. Gibbs, J. B.; Oli€, A. Annu. Rev. Pharmacol. Toxicol. 1997,
37, 143.
5. 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.; Oli€,
A. Nature (Medicine) 1995, 1, 792.
6. Crowell, P. L.; Gould, M. N. Crit. Rev. in Oncogenesis
1994, 5, 1.
7. Elson, C. E. J. Nutr. 1995, 125, 1666S.
8. Crowell, P. L.; Chang, R. R.; Ren, Z.; Elson, C. E.; Gould,
M. N. J. Biol. Chem. 1991, 266, 17679.
9. Ruch, R. J.; Sigler, K. Carcinogenesis 1994, 15, 787.
10. Hohl, R. J.; Lewis, K. J. Biol. Chem. 1995, 270, 17508.
11. Karlson, J.; Borg-Karlson, A.-K.; Unelius, R.; Shoshan,
M. C.; Wilking, N.; Ringborg, U.; Linder, S. Anti-Cancer
Drugs 1996, 7, 422.
12. Gelb, M. H.; Tamanoi, F.; Yokoyama, K.; Ghomashchi,
F.; Esson, K.; Gould, M. N. Cancer Lett. 1995, 91, 169.
13. Pompliano, D. L.; Rands, E.; Schaber, M. D.; Mosser,
S. D.; Anthony, N. J.; Gibbs, J. B. Biochemistry 1992, 31,
3800.
29. Patel, D. V.; Schmidt, R. J.; Biller, S. A.; Gordon, E. M.;
Robinson, S.; Manne, V. J. Med. Chem. 1995, 38, 2906.
30. Cohen, L. H.; Valentijn, A. R. P. M.; Roodenburg, L.;
Van Leeuwen, R. E. W.; Huismen, R. H.; Lutz, R. J.; Marel,
G. A. V. D.; Van Boom, J. H. Biochem. Pharmacol. 1995, 49, 839.
31. (a) Park, H. W.; Boduluri, S. R.; Moomaw, J. F.; Casey,
P. J.; Beese, L. S. Science 1997, 275, 1800. (b) Long, S. B.;
Casey, P. J.; Beese, L. S. Biochemistry 1998, 37, 9612.
32. McNamara, D. J.; Debrusin, E.; Leonard, D. M.; Shuler,
K. R.; Kaltenbronn, J. S.; III, J. Q.; Bur, S.; Thomas, C. E.;
Doherty, A. M.; Scholten, J. D.; Zimmerman, K. K.; Gibbs,
B. S.; Gowan, R. C.; Latash, M. P.; Leopold, W. R.; Przy-
branowski, S. A.; Sebolt-Leopold, J. S. J. Med. Chem. 1997,
40, 3319.
33. Aoyama, T.; Satoh, T.; Yonemoto, M.; Shibata, J.; Non-
oshita, K.; Arai, S.; Kawakami, K.; Iwasawa, Y.; Sano, H.;
Tanaka, K.; Monden, Y.; Koder, T.; Arakawa, H.; Suzuki-
Takahashi, I.; Kamei, T.; Tomimoto, K. J. Med. Chem. 1998,
41, 143.
34. Davisson, V. J.; Woodside, A. B.; Neal, T. R.; Stremler,
K. E.; Muehlbacher, M.; Poulter, C. D. J. Org. Chem. 1986,
51, 4768.
35. Yokoyama, K.; Zimmerman, K.; Scholten, J.; Gelb, M.
H. J. Biol. Chem. 1997, 272, 3944.
36. 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. J. Med.
Chem. 1997, 40, 192.
14. Gibbs, J. B.; Pompliano, D. L.; Mosser, S. D.; Rands, E.;
Lingham, R. B.; Singh, S. B.; Scolnick, E. M.; Kohl, N. E.;
Oli€, A. J. Biol. Chem. 1993, 268, 7617.
15. Crowell, P. L.; Ren, Z.; Lin, S.; Vedejs, E.; Gould, M. N.
Biochem. Pharmacol. 1994, 47, 1405.
37. Note added in proof: Recently it has been demonstrated
that a-hydroxygeranylphosphonate, the shorter C10 derivative
of FMP, is a much poorer inhibitor of FTase than FMP. This
is in accord with our proposal that hydrophobic bulk plays a
key role in the binding of farnesyl-based inhibitors to FTase:
Hohl, R. J.; Lewis, K. A.; Cermak, D. M.; Wiemer, D. F.
Lipids 1998, 33, 39.
16. McKenna, C. E.; Higa, M. T.; Cheung, N. H.; McKenna,
M. Tetrahedron Lett. 1977, 155.