N-arlalkyl pseudopeptide inhibitors of farnesyltransferase

Article abstract of DOI:10.1021/jm9800907

Inhibitors of Ras protein farnesyltransferrase are described which are reduced pseudopeptides related to the C-terminal tetrapeptide of the Ras protein that signals farneslation. Reduction of the carbonyl groups linking the first three residues of the tetrapeptide leads to active inhibitors which are chemically unstable. Stability can be restored by alkylating the central amine of the tetrapeptide. Studies of the SAR of these alkylated pseudopeptides with concominant modification of the side of the third residue led to 2(S)-(2(S)-{[2(S)-2(R)-amino-3-mercaptopropylamino)-3-(S)- methylpentyl]napthalen-1-ylmethylamino}acetylamino)-4-methylsulfanylbutyric acid (11), a subnanomolar inhibitor. The methyl ester (10) of this compound exhibited submicromolar activity in the processing assay and selectively inhibited anchorage-independent growth of Rat1 cells transformed by v-ras at 2.5 - 5 μ.

Full text of DOI:10.1021/jm9800907

J . Med. Chem. 1998, 41, 2651-2656
2651
N-Ar yla lk yl P seu d op ep tid e In h ibitor s of F a r n esyltr a n sfer a se
S. J ane deSolms,* Elizabeth A. Giuliani, Samuel L. Graham, Kenneth S. Koblan,† Nancy E. Kohl,†
Scott D. Mosser,† Allen I. Oliff,† David L. Pompliano,† Elaine Rands,† Thomas H. Scholz, Catherine M. Wiscount,
J ackson B. Gibbs,† and Robert L. Smith
Departments of Medicinal Chemistry and Cancer Research, Merck Research Laboratories, West Point, Pennsylvania 19486
Received February 11, 1998
Inhibitors of Ras protein farnesyltransferase are described which are reduced pseudopeptides
related to the C-terminal tetrapeptide of the Ras protein that signals farnesylation. Reduction
of the carbonyl groups linking the first three residues of the tetrapeptide leads to active
inhibitors which are chemically unstable. Stability can be restored by alkylating the central
amine of the tetrapeptide. Studies of the SAR of these alkylated pseudopeptides with
concomitant modification of the side chain of the third residue led to 2(S)-(2(S)-{[2(S)-(2(R)-
amino-3-mercaptopropylamino)-3(S)-methylpentyl]naphthalen-1-ylmethylamino}acetylamino)-
4-methylsulfanylbutyric acid (11), a subnanomolar inhibitor. The methyl ester (10) of this
compound exhibited submicromolar activity in the processing assay and selectively inhibited
anchorage-independent growth of Rat1 cells transformed by v-ras at 2.5-5 µM.
The ras oncogene product Ras p21 is a key player in
controlling cell proliferation. Mutations in Ras proteins
can lead to unregulated cell growth and are implicated
in many human cancers.1 Ras is initially synthesized
in vivo as an inactive cytosolic protein that requires a
series of posttranslational modifications to bind to the
cell membrane and express its biological activity. The
key modification is S-farnesylation of a cysteine residue
that is the fourth amino acid from the carboxy terminus
of the protein. This transformation is catalyzed by
farnesyl transferase (FTase) and uses farnesyl pyro-
phosphate (FPP) as cosubstrate.2 The carboxy-terminal
tetrapeptide of Ras and other farnesylated proteins,
known as the CAAX box, is required for recognition by
FTase. Furthermore, the CAAX tetrapeptide alone is
able to serve as a FTase substrate.3
Several types of FTase inhibitors have been reported
which mimic this tetrapeptide.4 CAAX tetrapeptides
themselves are inhibitors of protein farnesylation, while
serving as alternate substrates for FTase. However,
these peptides do not inhibit farnesylation in whole cells
in culture due to their proteolytic lability. Recently we
described a class of CAAX-like pseudotetrapeptides
which contained reduced amide bonds between the
cysteine and second amino acid and between the second
and third amino acids.5 The prototype, L-731,734 (1),
inhibited the growth of ras-transformed cells in soft agar
and completely inhibited Ras processing in intact cells
at 100 mM.6 While L-731,734 is itself a very weak
inhibitor of FTase, intracellular hydrolysis of the ho-
moserine lactone generates the corresponding acid 2, a
20 nM inhibitor of the enzyme in vitro. Since the acid
2 was inactive in these cell-based assays, we concluded
that temporarily masking the carboxylic acid as the
lactone was required to impart adequate cell membrane
permeability to observe antiproliferative activity in this
series of CAAX analogues.
prompted us to explore the chemical stability of this
compound. In pH 7.4 buffer, 1 readily cyclizes to the
diketopiperazine 3 at 25 °C with first-order kinetics (t1/2
) 20 h) (Figure 1). The diketopiperazine (3, FTase IC50
) 5 µM) was 250-fold less active than 2 against FTase.
Since the time scale of the growth assay in soft agar is
2 weeks with feeding twice weekly, the rate of destruc-
tion of 1 could clearly play a major role in limiting its
activity in cell culture. In this paper we present the
results of a test of the hypothesis that masking the
central amine of compounds related to 1 through alky-
lation or acylation to prevent this cyclization would
result in inhibitors that exhibit a better profile in cell
culture with respect to their intrinsic potency. We were
concerned that the intrinsic potency of these alkylated
derivatives might be compromised since we had found
that N-methylation of the tetrapeptide CVFM in any
position led to weaker inhibitors.7 This is consistent
with recently published results.8
Ch em istr y
The syntheses of 1 and 2 have been described previ-
ously.5 N-Alkylated analogues were prepared by a
series of reductive alkylations and deprotections as
illustrated in Scheme 1 for the synthesis of 10 and 11.
All of the compounds (12 through 26) were synthesized
in a similar manner by substituting the requisite
aldehydes or amino acid intermediates.
Str u ctu r e-Activity Rela tion sh ip s: In Vitr o In -
h ibition of F Ta se. Compounds were characterized as
inhibitors of FTase in vitro using either FTase purified
from bovine brain or homogeneous human recombinant
FTase, [3H]FPP, and recombinant Ha-Ras protein.9-11
Total protein was precipitated with acid, and protein-
bound FPP was quantitated by scintillation counting.
The activity of the compounds is reported as an IC50
value, the concentration at which radiolabel incorpora-
tion into Ras was reduced by 50% compared to an
experiment in which no inhibitor was present. To more
The poor activity of 1 in cell culture in relation to the
intrinsic inhibitory potency of its corresponding acid (2)
† Department of Cancer Research.
S0022-2623(98)00090-9 CCC: $15.00 © 1998 American Chemical Society
Published on Web 06/16/1998

2652 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 14
Notes
F igu r e 1.
Sch em e 1. Representative Synthesis of a Farnesyltransferase Inhibitor
accurately determine the potency of the more active
compounds, they were further tested in a similar
protocol which contained less enzyme (10 pM). Repre-
sentative compounds were evaluated as inhibitors of the
closely related enzyme protein geranylgeranyl trans-
ferase (GGTase-1).9,12 Details of these assays are found
in the cited literature.
N-Methylation of 2 to give 12 led to a 4-fold loss in
FTase activity (Table 1). Replacing the sec-butyl group
of 1 with an n-propyl group gave 13 of equal potency.
N-Methylation of 13 to give 14, however, did not alter
the inhibitory activity. This suggests that the reduction
in potency upon methylation of 2 is due to a conforma-
tional bias imposed by the interaction of the N-methyl
group and the sec-butyl group of the isoleucine residue.
If that is the case, the effect must be peculiar to the
inhibitor bearing the homoserine residue at the carboxy
terminus of the inhibitor. In the nonmethylated series,
replacement of the homoserine carboxy terminus with
methionine in 2 gave 17 of comparable potency. N-
Methylation of 17 to give 18 did not alter potency. It is
possible that different bound conformations of the
inhibitors are required to correctly present the hydro-
philic side chain of homoserine as opposed to the
hydrophobic methionine side chain, perhaps in unique
binding sites.
Similar results were obtained in the CIFM series. We
previously reported that replacing the sec-butyl group
of 2 with benzyl provided a more potent inhibitor (19,
IC50 ) 4.7 nM).5 N-Methylation of 19 to give 20 led to
a 9-fold loss of activity. In the CIF-HSer series, the
effect of N-methylation was much more dramatic, lead-
ing to a 64-fold loss in potency upon N-methylation of
15 to give 16. Incorporating benzyl groups at both R1
and R2 (21) led to a 70-fold drop in FTase activity,
indicating that only one aromatic binding site is avail-
able.
Removing the substituent at R1 gave the glycine
analogue (22), a potent FTase inhibitor (IC50 ) 1.2 nM).
This shifting of the amino acid side chain from the
R-position to nitrogen has provided potent inhibitors.
Recently, this same ploy has been reported to give
potent inhibitors in the tetrapeptide CVFM series.13
Replacement of benzyl with 1-naphthylmethyl (11) gave
the most potent inhibitor in this series. All compounds
tested were 10-1000-fold selective for FTase versus
GGTase-1. No definitive trend of structure vs GGTase-1
activity could be established.

Notes
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 14 2653
Ta ble 1. FTase Inhibition and Selectivity of Pseudopeptides in Vitroa
IC50 (nM)
compd
R1
R2
R3
FPTaseb
20 ( 6 (3)
0.123* ( 0.057 (4)
83+ ( 15 (3)
27+ (1)
GGTase-1c
2
(S)CH(CH3)C2H5
H
H
CH2CH2OH
CH2CH2OH
CH2CH2OH
CH2CH2OH
CH2CH2OH
CH2CH2OH
CH2CH2OH
CH2CH2SCH3
CH2CH2SCH3
CH2CH2SCH3
CH2CH2SCH3
CH2CH2SCH3
CH2CH2SCH3
CH2CH2SCH3
CH2CH2SCH3
45000 (1)
10 (1)
11
12
13
14
15
16
17
18
19
20
21
22
23
24
1-naphthyl CH2
H
CH3
H
CH3
H
CH3
H
CH3
H
10000 (1)
10000 (1)
43000 (1)
10000 (1)
nd
CH2CH2CH3
CH2CH2CH3
CH2Ph
21 (1)
26+ ( 1 (3)
1600 (1)
CH2Ph
(S)CH(CH3)C2H5
(S)CH(CH3)C2H5
CH2Ph
30 (1)
nd
25 ( 10 (5)
4.8 ( 0.4
42 (1)
1400 (1)
1.2* ( 0.8 (2)
2.2* (1)
4700 (1)
5750 ( 50 (2)
nd
CH2Ph
CH2Ph
H
CH3
CH2Ph
CH2Ph
CH2Ph
CH2Ph
nd
450 (1)
39000 (1)
nd
H, R ) HS(CH2)3
H, R ) HS(CH2)2CO
110 (1)
a
nd, not determined. b Concentration of compound required to inhibit 50% of FPTase-catalyzed incorporation of [3H]FPP into recombinant
+
human Ha-Ras protein by 50%. Those values marked with
were obtained using enzyme from bovine brain at a concentration of
approximately 1 nM, and those marked with * used 10 pM enzyme concentration. c Inhibition of bovine type-I geranylgeranyltransferase.8
Assay results are reported as concentration ( SEM for the number of determinations shown in parentheses. With one determination, the
values are estimated to be reliable within 2-fold.
Deletion of the primary amino group of the cysteine
residue has a deleterious effect on binding to FTase:
compound 23 is at least 20-fold less potent an inhibitor
than 11. The mercaptopropionyl analogue of 23, com-
pound 24, is even less potent (IC50 ) 110 nM). Thus,
the presence of a basic group, preferably a primary
amine, in the vicinity of the cysteine residue is an
element critical for the binding of this family of inhibi-
tors. In this sense, the SAR in the naphthylmethyl
glycine series parallels the SAR that we and others
observed in analogues of CVFM5,14 and is distinct from
the SAR of tetrapeptide analogues based on CVIM.
Thus, it appears that the CVFM and naphthylmethyl
glycine series bind to FTase in a conformation distinct
from the nonaromatic inhibitors (i.e., reduced analogues
of CVIM), with the former presenting the cysteine
residue from a conformation that allows productive
interaction between the cationic amine and a comple-
mentary binding element on the enzyme. Indeed, this
observation has been exploited in the preparation of
novel FTase inhibitors.15
nesylated protein was observed with SDS-polyacryl-
amide gel electrophoresis. The HMG-CoA reductase
inhibitor lovastatin was used as a positive control.
Lovastatin at 15 µM gave >90% inhibition of Ras
processing.
In addition, the most active ester in the processing
assay was evaluated in a second cell assay which
measured its ability to inhibit the anchorage-indepen-
dent growth of Rat1 cells transformed by v-ras or v-raf
oncogene. A minimum inhibitory concentration (MIC)
required to achieve a reduction in size and number of
colonies of transformed cells in soft agar relative to
vehicle-treated controls was determined. The Raf-
transformed cells, which do not require farnesylation
for biological activity, were included to evaluate the
specificity of these inhibitors for Ras-induced cell trans-
formation. Details of these assays are described else-
where.6
Although the N-methyl analogue 12 was 4-fold less
active against FTase than 2, its corresponding ester
prodrug 25 was 4-fold more active than 1 in the
processing assay (Table 2). In pH 7.4 buffer, at 25 °C,
25 remained intact with the hydrolyzed lactone 12 (t1/2
) 70 h) as the only product. This suggests that our
hypothesis was correct: masking the central amine of
these molecules results in inhibitors that exhibit a
better profile in cell culture with respect to their
intrinsic potency.
In h ibition of Ra s F a r n esyla tion in Cell Cu ltu r e
The cytotoxicity of the prodrug esters of the more
active FTase inhibitors was assessed using a viable
staining method with MTT (3-(4,5-dimethylthiazol-2-yl)-
2,5-diphenyl-2H-tetrazolium bromide). The cytotoxic
endpoint is the highest compound concentration toler-
ated by NIH 3T3 cells in a 48-h assay. NIH 3T3 cells
transformed by v-ras were used to evaluate the effect
of the prodrug esters of the more active inhibitors on
the posttranslational processing of Ras in intact cells.
The cells were incubated in the presence of the indicated
compound or solvent control for 24 h and were labeled
with [35S]methionine during the final 20 h. Ras was
immunoprecipitated from detergent lysates of cell ex-
tracts, and the mobility of farnesylated versus nonfar-
Furthermore, the transposition of the arylalkyl group
from the R-position of the amino acid residue in A2 of
the CA1A2X box to its amino group, to give compounds
such as 11 and 22, leads to considerably more potent
FTase inhibitors. The prodrug esters 10 and 26 of the
most potent inhibitors 11 and 22, respectively, exhibited
submicromolar activity in the processing assay, 10-fold
lower than their cytotoxic endpoints. Compound 10 also
inhibited anchorage-independent growth of Rat1 cells

2654 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 14
Ta ble 2. Effect of FTase Inhibitors in Cell Culturea
Notes
a
d
See footnote a of Table 1. Highest nontoxic concentration for cultured NIH3T3 cells as assessed by MTT staining. e Inhibition of
posttranslational processing of v-Ras protein in cultured NIH3T3 cells.6 f Minimum inhibitory concentration (MIC) required to achieve
a reduction in size and number of colonies of RAT1-v-ras- or RAT1-v-raf-transformed cells in soft agar relative to vehicle-treated control.6
with NaOH under Ar with stirring for 72 h. The reaction
mixture was treated with dithiothreitol (0.008 g, 0.052 mmol)
dissolved in MeOH (0.5 mL) and prepped on a RPLC using a
VYDAC column eluting with 0.1%TFA/H2O:0.1%TFA/CH3CN
transformed by v-ras with a MIC of 2.5-5 µM. This
inhibition was selective for v-ras transformed cells since
the v-raf transformed cells were not inhibited at 5 µM.
(95:5 to 5:95 gradient over 1 h). Like fractions were combined
Con clu sion
to give 0.0045 g (28%) of 3 after lyophilization: 1H NMR (CD3-
CAAX-like pseudotetrapeptides with reduced amide
linkages between cysteine and the second amino acid
and between the second and third amino acids can be
stabilized by alkylation of the central amine of the
pseudotetrapeptide. Concomitant modification of the
akyl group on the amine as well as the alkyl side chain
of the third amino acid has led to a series of subnano-
molar FTase inhibitors. The ester prodrugs of these
potent inhibitors inhibit Ras processing at submicro-
molar levels and selectively inhibit anchorage-indepen-
dent growth of Rat1 cells transformed by v-ras. These
results suggest that the N-arylalkyl pseudopeptides
represent a useful template for further development as
cancer chemotherapeutics.
OD) δ 4.21-4.28 (m, 1H), 4.1-4.18 (m, 1H), 3.71 (d, 1H, J )
6 Hz), 3.7-3.82 (m, 2H), 3.4-3.5 (m, 1H), 2.75-3.21 (m, 6H),
2.18-2.3 (m, 1H), 1.83-2.01 (m, 1H), 1.64-1.82 (m, 2H), 1.4-
1.58 (m, 2H), 1.21-1.4 (m, 1H), 0.92-1.11 (m, 12H); MS (M +
1) 403.
P r ep a r a t ion of 2(S)-(2(S)-{[2(S)-(2(R)-Am in o-3-m er -
ca p top r op yla m in o)-3(S)-m eth ylp en tyl]n a p h th a len -1-yl-
m eth ylam in o}acetylam in o)-4-m eth ylsu lfan ylbu tyr ic Acid
Meth yl Ester (10). Glycine methyl ester hydrochloride (5)
(4.41 g, 0.035 mol) was dissolved in 1,2-dichloroethane (50 mL)
and DMF (5 mL) and treated with 3A molecular sieves (10 g)
and N-tert-butoxycarbonylisoleucinal (4) (6.3 g, 0.029 mol) with
stirring at 0 °C. Sodium triacetoxyborohydride (9.27 g, 0.044
mol) was added, and the pH of the mixture was adjusted to 6
with Et3N (3 mL, 0.022 mol). After stirring for 18 h, the
mixture was filtered, concentrated to a small volume, and
partitioned between EtOAc and water. Standard workup gave
3.88 g (54%) of compound 6 after purification by chromatog-
raphy (SiO2, EtOAc:hexane, 1:3): 1H NMR (CDCl3) δ 4.69 (m,
1H), 3.72 (s, 3H), 3.48-3.62 (m, 1H), 3.42 (ABq, 2H), 2.65 (d,
2H, J ) 6 Hz), 1.4-1.6 (m, 2H), 1.48 (s, 9H), 1.04-1.2 (m, 1H),
0.85-0.95 (m, 6H).
Compound 6 (2.00 g, 6.97 mmol) was dissolved in 1,2-
dichloroethane (56 mL), and 3A molecular sieves were added
followed by 1-naphthaldehyde (1.89 mL, 13.9 mmol) and
sodium triacetoxyborohydride (6.65 g, 31.4 mmol). The mix-
ture was stirred at ambient temperature for 16 h, filtered
through glass fiber paper, and concentrated. The residue was
partitioned between EtOAc and saturated NaHCO3 (100 mL/
25 mL). Standard workup gave 5.0 g of crude product which
was purified by chromatography (SiO2, 1:6 to 1:3 EtOAc:
hexane) to give 3.8 g of compound 7: 1H NMR (CD3OD) δ
8.44-8.38 (d, 1H, J ) 6 Hz), 7.88-7.77 (m, 2H), 7.55-7.35
(m, 4H), 6.34-6.27 (m, 1H), 4.25 (ABq, 2H), 3.66 (s, 3H), 3.40-
3.23 (m, 1H), 2.95-2.85 (dd, 1H, J ) 6, 15 Hz), 2.68-2.57 (dd,
1H, J ) 6, 15 Hz), 1.57-1.46 (m, 1H), 1.43 (s, 9H), 1.34-1.18
(m, 2H), 1.06-0.85 (m, 1H), 0.85-0.71 (m, 6H).
Exp er im en ta l Section
Solvents and reagents were obtained from commercial
suppliers and were used as received. Simple peptides used
as synthetic intermediates were prepared using standard
solution-phase methods and are not described in detail.
Reactions were generally conducted under an argon atmo-
sphere using magnetic stirring. Standard workup consisted
of extraction with an organic solvent and washing as appropri-
ate with saturated sodium bicarbonate solution and brine. The
organic solutions were dried over sodium sulfate, and the
solvent was removed on a rotary evaporator. Chromatography
was performed on silica gel (230-400 mesh) at approximately
5 psig. Preparative reverse phase HPLC was performed on a
Waters 3000 instrument using either C-18 Vydac or PrepPak
columns. Products and intermediates were characterized by
300-MHz 1H NMR. Final products were also characterized by
mass spectrometry and combustion analyses. Observed values
were within 0.4% of calculated values for the compound
formulas shown. The following examples are illustrative of
the procedures employed.
1-[2(S )-(2(R )-Am in o-3-m e r ca p t op r op yla m in o)-3(S )-
m et h ylp en t yl]-6(S)-sec-b u t yl-3(S)-(2-h yd r oxyet h yl)p ip -
er a zin e-2,5-d ion e (3). 2-{[2-(2-Amino-3-mercaptopropylami-
no)-3-methylpentyl]amino}-3-methylpentanoic acid (2-oxotet-
rahydrofuran-3-yl)amide (1)5 (0.020 g, 0.026 mmol) was dis-
solved in degassed 0.1 M KHSO4 solution buffered to pH 7.36
Compound 7 (2.61 g, 6.10 mmol) was dissolved in CH3OH
(50 mL), and 1 N NaOH (24.4 mL, 24.4 mmol) was added. The
mixture was stirred at ambient temperature for 4 h and
concentrated. The resulting residue was dissolved in H2O (25
mL) and neutralized with 1 N HCl (24.4 mL). The aqueous
layer was washed with EtOAc (3 × 50 mL). The organic layers

Notes
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 14 2655
were combined, dried with Na2SO4, filtered, and concentrated
to give 2.29 g of the corresponding acid: 1H NMR (CD3OD) δ
8.48-8.39 (d, 1H, J ) 6 Hz), 8.03-7.91 (t, 2H, J ) 6 Hz) 7.75-
7.48 (m, 4H), 5.00-4.93 (d, 1H, J ) 12 Hz), 4.78-4.66 (d, 1H,
J ) 12 Hz), 3.80-3.58 (m, 3H), 3.49-3.40 (dd, 1H, J ) 3, 12
Hz), 3.09-2.98 (dd, 1H, J ) 3, 12 Hz), 1.42 (s, 9H), 1.37-1.28
(m, 2H), 1.80-1.00 (m, 1H), 0.94-0.78 (m, 6H).
This acid was dissolved in DMF (20 mL) and treated with
HOBT (0.822 g, 6.08 mmol), EDC (1.17 g, 6.08 mmol), and
methionine methyl ester hydrochloride (1.21 g, 6.08 mmol).
The pH was adjusted to 7.5 with Et3N (1.7 mL, 12 mmol), and
the mixture was stirred at ambient temperature for 24 h. The
mixture was concentrated, and the residue was partitioned
between EtOAc (50 mL) and saturated NaHCO3 solution (25
mL). Standard workup gave 3.2 g of crude product which was
purified by chromatography (SiO2 eluting with 1:3 to 1:2
EtOAc:hexane) to give 2.82 g of compound 8: 1H NMR (CD3-
OD) δ 8.36-8.29 (d, 1H, J ) 6 Hz), 7.93-7.86 (d, 1H, J ) 6
Hz), 7.85-7.80 (d, 1H, J ) 6 Hz), 7.61-7.39 (m, 4H), 6.60-
6.52 (m, 1H), 4.32-4.06 (m, 2H), 3.90-3.69 (m, 1H), 3.65 (s,
3H), 3.27-3.14 (m, 2H), 2.93-2.70 (m, 2H), 2.19-1.78 (m, 6H,
1.63-1.30 (m, 13H), 1.19-1.05 (m, 1H), 0.95-0.81 (m, 6H).
Compound 8 (2.82 g, 5.04 mmol) was dissolved in EtOAc
(50 mL) and cooled to -25 °C. HCl was bubbled through the
mixture until TLC (95:5 CH2Cl2:CH3OH) indicated complete
reaction. Nitrogen was bubbled through the mixture to remove
excess HCl, and the mixture was then concentrated to give
2.68 g of the deprotected amine: 1H NMR (CD3OD); d 8.34-
8.28 (d, 1H, J ) 6 Hz), 8.00-7.92 (d, 2H, J ) 6 Hz), 7.83-7.71
(m, 1H), 7.68-7.49 (m, 3H), 4.76-4.55 (m, 4H), 3.84-3.75 (m,
2H), 3.71 (s, 3H), 3.59-3.70 (m, 1H), 3.21-3.00 (m, 2H), 2.57-
2.38 (m, 3H), 2.17-2.04 (m, 4H), 1.97-1.81 (m, 1H), 1.63-
1.50 (m, 1H), 1.39-1.20 (m, 1H), 1.19-1.00 (m, 1H), 0.95-
0.79 (m, 6H).
This amine (0.200 g, 0.376 mmol) was dissolved in CH3OH
(5 mL), treated with KOAc (0.074 g, 0.752 mmol), 3A molecular
sieves (0.5 g), and N-(tert-butoxycarbonylamino)-S-triphenyl-
methylcysteine aldehyde5 (0.219 g, 0.489 mmol) followed by
NaCNBH3 (0.038 g, 0.602 mmol), and stirred at ambient
temperature for 18 h. The reaction mixture was filtered and
partitioned between EtOAc (20 mL) and aqueous saturated
NaHCO3 solution. The organic layer was washed with brine
and dried (Na2SO4). Filtration and concentration to dryness
gave crude product which was chromatographed (SiO2, EtOAc:
hexane, 1:3 to 1:1) to give 0.178 g (53%) of compound 9: 1H
NMR (CD3OD) δ 8.31 (d, 1H, J ) 9 Hz), 7.75-7.89 (m, 2H),
7.15-7.5 (m, 19H), 4.25-4.40 (m, 2H), 4.03 (d, 1H, J ) 12 Hz),
3.5-3.7 (m, 1H), 3.66 (s, 3H), 3.2-3.4 (m, 3H), 2.78-2.9 (m,
1H), 2.45-2.66 (m, 3H), 2.2-2.4 (m, 3H), 2.05 (s, 3H), 1.95 (s,
3H), 1.95-2.2 (m, 2H), 1.75-1.9 (m, 1H), 1.3-1.6 (m, 2H), 1.36
(s, 9H), 0.95-1.1 (m, 1H), 0.75-0.93 (m, 6H).
triethylsilane (0.052 mL, 0.324 mmol). After 2 h the reaction
mixture was concentrated to dryness and partitioned between
hexane and 0.1%TFA/H2O, and the aqueous layer was chro-
matographed by preparative RP HPLC using a Waters Prep-
Pak. Pure fractions were combined and lyophilized to give
0.047 g (71%) of compound 11: 1H NMR (CD3OD) δ 8.3 (d,
1H, J ) 9 Hz), 7.99 (d, 2H, J ) 9 Hz), 7.5-7.65 (m, 4H), 4.48-
4.8 (m, 2H), 4.2-4.36 (m, 1H), 3.55-3.95 (m, 2H), 2.82-3.55
(m, 3H), 2.45-2.82 (m, 8H), 2.10 (s, 3H), 2.14-2.28 (m, 1H),
1.6-2.1 (m, 4H), 1.08-1.35 (m, 2H), 0.75-0.9 (m, 6H). Anal.
(C27H42N4O3S2‚2.5 CF3CO2H) C, H, N.
2(S)-(2(S)-{[2(S)-(3-Mer ca p top r op yla m in o)-3(S)-m eth -
ylp en tyl]n a p h th a len -1-ylm eth yla m in o}a cetyla m in o)-4-
m eth ylsu lfa n ylbu tyr ic Acid (23). The amine hydrochloride
(0.200 g, 0.376 mmol) from the deprotection of compound 8
described above was dissolved in CH3OH (10 mL), treated with
KOAc (0.074 g, 0.752 mmol), 3A molecular sieves (0.5 g), and
S-triphenylmethylpropionaldehyde (0.150 g, 0.451 mmol) fol-
lowed by NaCNBH3 (0.036 g, 0.564 mmol), and stirred at
ambient temperature for 18 h. The reaction mixture was
filtered and partitioned between EtOAc (20 mL) and aqueous
saturated NaHCO3 solution. The organic layer was washed
with brine and dried (Na2SO4). Filtration and concentration
to dryness gave crude product which was chromatographed
(SiO2, CH2Cl2:CH3OH, 98:2 to 94:6) to give 0.230 g (79%) of
fully protected 23. Following the procedures described above,
basic hydrolysis followed by detritylation (TFA/triethylsilane)
gave compound 23 in 27% overall yield: 1H NMR (CD3OD) δ
8.25 (d, 1H, J ) 9 Hz), 7.86-7.98 (m, 2H), 7.44-7.66 (m, 4H),
4.58-4.64 (m, 1H), 4.43 (d, 1H, J ) 12 Hz), 4.18 (d, 1H, J )
12 Hz), 3.71 (d, 1H, J ) 13 Hz), 3.56 (d, 1H, J ) 13 Hz), 3.03
(dd, 1H, J ) 3, 13 Hz), 2.44-2.71 (m, 4H), 2.27-2.40 (m, 2H),
2.12-2.28 (m, 2H), 2.10 (s, 3H), 1.85-2.1 (m, 2H), 1.46-1.8
(m, 3H), 1.06-1.2 (m, 2H), 0.82 (d, 3H, J ) 6 Hz), 0.76 (t, 3H,
J ) 6 Hz). Anal. (C27H41N3O3S2‚3.75CF3CO2H) C, H, N.
2(S )-(2(S )-{[2(S )-(3-Me r ca p t op r op ion yla m in o)-3(S )-
m eth ylpen tyl]n aph th alen -1-ylm eth ylam in o}acetylam in o)-
4-m eth ylsu lfa n ylbu tyr ic Acid (24). The amine hydrochlo-
ride (0.200 g, 0.376 mmol) from the deprotection of compound
8 described above was dissolved in DMF (5 mL) and treated
with S-triphenylmethylpropionic acid (0.157 g, 0.451 mmol),
EDC (0.086 g, 0.451 mmol), and HOBT (0.061 g, 0.451 mmol).
The pH was adjusted to 8 with Et3N (0.180 mL, 1.3 mmol)
and the reaction mixture stirred for 48 h at ambient temper-
ature. Standard workup followed by flash chromatography
(SiO2, CH2Cl2:CH3OH, 99:1 to 95:5) gave 0.230 g (77%) of the
fully protected 24. Following the procedures described above,
basic hydrolysis followed by detritylation (TFA/triethylsilane)
gave compound 24 in 48% overall yield: 1H NMR (CD3OD) δ
8.36 (d, 1H, J ) 9 Hz), 7.94-8.05 (m, 2H), 7.48-7.79 (m, 4H),
4.41 4.9 (m, 2H), 3.71-4.2 (m, 3H), 3.4-3.7 (m, 1H), 3.05-3.3
(m, 1H), 2.1-2.9 (m, 6H), 2.02 (s, 3H), 1.75-2.1 (m, 3H), 1.3-
Compound 9 (0.178 g, 0.200 mmol) dissolved in CH2Cl2 (4
mL) and TFA (2 mL) at ambient temperature was treated with
triethylsilane (0.128 mL, 0.800 mmol). After 2 h the reaction
mixture was concentrated to dryness and partitioned between
hexane and 0.1% TFA/H2O, and the aqueous layer was
chromatographed by preparative RP HPLC using a Waters
PrepPak. Pure fractions were combined and lyophilized to give
compound 10: 1H NMR (CD3OD) δ 8.29 (d, 1H, J ) 9 Hz), 8.0
(d, 2H, J ) 9 Hz), 7.5-7.75 (m, 4H), 4.6-4.8 (m, 2H), 4.2-4.4
(m, 1H), 3.74 (s, 3H), 3.6-4.0 (m, 3H), 2.95-3.4 (m, 3H), 2.5-
2.85 (m, 8H), 2.08 (s, 3H), 1.92-2.23 (m, 2H), 1.6-1.9 (m, 2H),
1.1-1.36 (m, 2H), 0.78-0.92 (m, 6H), MS m/e 563 (M + 1).
Anal. (C28H44N4O3S2‚2CF3CO2H‚H2O) C, H, N.
1.65 (m, 2H), 1.05-1.3 (m, 1H), 0.7-1.0 (m, 6H). Anal. (C27
39N3O4S2‚CF3CO2H‚.75H2O) C, H, N.
-
H
Ack n ow led gm en t. The authors thank Ms. J oy
Hartzell for preparation of the manuscript.
Refer en ces
(1) Bos, J . L. Ras Oncogenes In Human Cancer: A Review. Cancer
Res. 1989, 49, 4682-4689.
(2) Lowy, D. R.; Willumsen, B. M. Function and Regulation of Ras.
Annu. Rev. Biochem. 1993, 62, 851-891.
(3) Goldstein, J . L.; Brown, M. S.; Reiss, Y.; Stradley, S. J .; Gierasch,
L. M. Nonfarnesylated Tetrapeptide Inhibitors of Protein Far-
nesyltransferase. J . Biol. Chem. 1991, 266, 15575-15578.
(4) (a) Nigam, M.; Seong, C.-M.; Qian, Y.; Hamilton, A. D.; Sebti,
S. M. Potent Inhibition of Human Tumor p21ras Farnesyltrans-
ferase by A1A2-Lacking p21ras CA1A2X Peptidomimetics. J . Biol.
Chem. 1993, 268, 20695-20698. (b) Qian, Y.; Blaskovich, M. A.;
Saleem, M.; Seong, C.-M.; Wathen, S. P.; Hamilton, A. D.; Sebti,
S. M. Design and Structural Requirements of Potent Peptido-
mimetic Inhibitors of p21ras Farnesyltransferase. J . Biol. Chem.
1994, 269, 12410-12413. (c) Wai, J . S.; Bamberger, D. L.; Fisher,
T. E.; Graham, S. L.; Smith, R. L.; Kohl, N. E.; Mosser, S. D.;
Compound 9 (0.097 g, 0.109 mmol) was dissolved in CH3-
OH (5 mL), cooled to 0 °C, and treated with 1 N NaOH solution
(0.436 mL, 0.436 mmol). After stirring for 5 h, the reaction
was neutralized with 1 N HCl (0.436 mL, 0.436 mmol) and
extracted with EtOAc (3 × 20 mL). The organic layers were
combined, washed with brine, dried (Na2SO4), filtered, and
concentrated to give the protected acid (0.071 g) (74%) which
was used without further purification.
The acid (0.071 g, 0.081 mmol) dissolved in CH2Cl2 (2 mL)
and TFA (1 mL) at ambient temperature was treated with

2656 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 14
Notes
Oliff, A. I.; Pompliano, D. L.; Rands, E.; Gibbs, J . B. Synthesis
and Biological Activity of Ras Farnesyl-Protein Transferase
Inhibitors. Tetrapeptide Analogues with Aminomethyl and
Carbon Linkages. Biorg. Med. Chem. 1994, 2, 939-947.
(5) Graham, S. L.; deSolms, S. J .; Giuliani, E. A.; Kohl, N. E.;
Mosser, S. D.; Oliff, A. I.; Pompliano, D. L.; Rands, E.; Breslin,
M. J .; Deana, A. A.; Garsky, V. M.; Scholz, T. H.; Gibbs, J . B.;
Smith, R. L. Pseudopeptide Inhibitors of Ras Farnesyl-Protein
Transferase. J . Med. Chem. 1994, 37, 725-732.
Homology with Yeast Prenyl-Protein Transferases. Biochemistry
1993, 32, 5167-5176.
(12) (a) Yoshida, Y.; Kawata, M.; Katayama, M.; Horiuchi, H.; Kita,
Y.; Takai, Y. A Geranylgeranyltransferase for rhoA p21 Distinct
from the Farnesyltransferase for ras p21S. Biochem. Biophys.
Res. Commun. 1991, 175, 720-728. (b) Fiengold, A. A.; J ohnson,
D. I.; Farnsworth, C. C.; Gelb, M. H.; J udd, S. R.; Glomset, J .
A.; Tamanoi, F. Protein Geranylgeranyltransferase of Saccha-
romyces cerevisiae Is Specific for Cys-Xaa-Xaa-Leu Motif Proteins
and Requires the CDC43 Gene Product but not the DPR1 Gene
Product. Proc. Natl. Acad. Sci., U.S.A. 1991, 88, 4448-4452. (c)
Seabra, M. C.; Reiss, Y.; Casey, P. J .; Brown, M. S.; Goldstein,
J . L. Protein Farnesyltransferase and Geranylgeranyl-Trans-
ferase Share a Common A Subunit. Cell 1991, 65, 429-434. (d)
Yokoyama, K.; Goodwin, G. W.; Ghomashchi, F.; Glomset, J . A.;
Gelb, M. H. A Protein Geranylgeranyl-Transferase From Bovine
Brain: Implications for Protein Prenylation Specificity. Proc.
Natl. Acad. Sci., U.S.A. 1991, 88, 5302-5306. (e) Casey, P. J .;
Thissen, J . A.; Moomaw, J . F. Enzymatic Modification of Proteins
with a Geranylgeranyl Isoprenoid. Proc. Natl. Acad. Sci., U.S.A.
1991, 88, 8631-8635.
(6) Kohl, N. E.; Mosser, S. D.; deSolms, S. J .; Giuliani, E. A.;
Pompliano, D. L.; Graham, S. L.; Smith, R. L.; Scolnick, E. M.;
Oliff. A. I.; Gibbs, J . B. Selective Inhibition Of Ras- Dependent
Transformation by a Farnesyltransferase Inhibitor. Science
1993, 260, 1934-1937.
(7) Unpublished results.
(8) Leftheris, K.; Kline, T.; Natarajan, S.; DeVirgilio, M. K.; Cho,
Y. H.; Pluscec, J .; Ricca, C.; Robinson, S.; Seizinger, B. R.;
Manne, V.; Meyers, C. A.; Peptide Based P21RAS Farnesyl
Transferase Inhibitors: Systematic Modification of the Tet-
rapeptide CA1A2X Motif. Biorg. Med. Chem. Lett. 1994, 4, 887-
892.
(9) Moores, S. L.; Schaber, M. D.; Mosser, S. D.; Rands, E.; O’Hara,
M. B.; Garsky, V. M.; Marshall, M. S.; Pompliano, D. L.; Gibbs,
J . B. Sequence Dependence of Protein Isoprenylation. J . Biol.
Chem. 1991, 166, 14603-14610.
(10) Pompliano, D. L.; Rands, E.; Schaber, M. D.; Mosser, S. D.;
Anthony, N. J .; Gibbs, J . B. Steady State Kinetic Mechanism of
Ras Farnesyltransferase. Biochemistry 1992, 31, 3800-3807.
(11) Omer, C. A.; Kral, A. M.; Diehl, R. E.; Prendergast, G. C.; Powers,
S.; Allen, C. M.; Gibbs, J . B.; Kohl, N. E. Characterization of
Recombinant Human Farnesyl-Protein Transferase: Cloning,
Expression, Farnesyl Diphosphate Binding, and Functional
(13) Reuveni, H.; Gitler, A.; Poradosu, E.; Gilon, C.; Levitzki, A.
Synthesis and Biological Activity of Semipeptoid Farnesyltrans-
ferase Inhibitors. Biorg. Med. Chem. 1997, 5, 85-92.
(14) Brown, M. S.; Goldstein, J . L.; Paris, K. J .; Burnier, J . P.;
Marsters, J . C., J r. Tetrapeptide Inhibitors of Protein Farne-
syltransferase: Amino-Terminal Substitution In Phenylalanine-
Containing Tetrapeptides Restores Farnesylation. Proc. Natl.
Acad. Sci., U.S.A. 1992, 89, 8313-8316.
(15) manuscript in preparation.
J M9800907