Synthesis and evaluation of a novel series of farnesyl protein transferase inhibitors as non-peptidic CAAX tetrapeptide analogues

Article abstract of DOI:10.1016/S0960-894X(03)00170-7

A novel series of compounds, derived from 4-amino-phenyl piperazine, has been designed to selectively inhibit farnesyl protein transferase (FPTase) as CAAX tetrapeptide analogues. Certain of these compounds were shown to possess low nanomolar inhibitory a

Full text of DOI:10.1016/S0960-894X(03)00170-7

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1455–1458
Synthesis and Evaluation of a Novel Series of Farnesyl
Protein Transferase Inhibitors as Non-Peptidic CAAX
Tetrapeptide Analogues
Michel Perez,a,* Catherine Maraval,a Stephan Dumond,a Marie Lamothe,a
Philippe Schambel,b Chantal Etievantc and Bridget Hillc
aDepartment of Medicinal Chemistry 4, Centre de Recherche Pierre Fabre, 17, Avenue Jean Moulin, 81106 Castres Cedex, France
bDepartment of Molecular Modeling, Centre de Recherche Pierre Fabre, 17, Avenue Jean Moulin, 81106 Castres Cedex, France
cDepartment of Cancerology, Centre de Recherche Pierre Fabre, 17, Avenue Jean Moulin, 81106 Castres Cedex, France
Received 15 November 2002; revised 20 January 2003; accepted 4 February 2003
Abstract—A novel series of compounds, derived from 4-amino-phenyl piperazine, has been designed to selectively inhibit farnesyl
protein transferase (FPTase) as CAAX tetrapeptide analogues. Certain of these compounds were shown to possess low nanomolar
inhibitory activity both against the isolated enzyme and in cultured cells.
# 2003 Elsevier Science Ltd. All rights reserved.
FPTase catalyzes a key step in the post-translational
modification of different kind of proteins implicated in
cell proliferation. This reaction transfers a farnesyl
group from farnesyl pyrophosphate onto the cysteine
residue present in the carboxy terminal tetrapeptide
sequence of these proteins, called CAAX box with the C
referring to the prenylated cysteine, A referring to ali-
phatic amino acids and X referring to methionine, serine
or glutamine.1,2 Among the proteins farnesylated, Ras
proteins were identified as an important target since
they are key components in cellular signal transduction
pathway, controlling cell proliferation and differentia-
tion, and since mutations of the ras gene have been
observed in about 30% of human cancers.3 In order to
function, both mutated and wild type Ras must move to
the inner face of the plasma membrane. This ability is
only acquired after several post-translational modifi-
cations that create a membrane-binding domain on the
C-terminal portion of the protein. In these chemical
processes, because farnesylation alone can support Ras
transforming activity, inhibition of FPTase was envi-
sioned as a way to down regulate ras function.4,5 How-
ever, it is now generally acknowledged that other
proteins may also implicated in the anti-tumor activity
of farnesyl transferase inhibitors (FTIs). Indeed, about
20 different proteins have been identified as substrates
for FPTase in mammalian cells6 and some of them like
RhoB or centromer binding proteins represent com-
plementary or alternative targets to elaborate a new
possible mechanism of the antiproliferative activity.7
One of the milestones in the development of small FTIs
was the finding that all the major recognition elements
responsible for interaction with the enzyme were carried by
the isolated CAAX tetrapeptide.8 Since then, rational
design has been able to produce an important number of
non-peptidic CAAX analogues as potent inhibitors of
FPTase.9,10 The originality of this target is that most of the
FTIs discovered possess very diverse chemical structures
relative to each other. This diversity is probably due to the
very wide catalytic site of the enzyme and to the different
chemical approaches used to identify these inhibitors
(screening of natural products or libraries of synthetic
compounds). However, most of the FTIs described to date
display relatively poor cellular activity compared to their
inhibitory potencies on the isolated enzyme. Our own
research in this field has led to potent inhibitors of FPTase
both on the isolated enzyme and in cellular models.
Our strategy was also focused on CAAX box analogues.
To mimic the central core of the CVIM-tetrapeptide we
decided to use the amino-phenyl piperazine scaffold that
we previously worked in a CNS project. This scaffold
*Corresponding author. Fax: +33-5-63-71-42-99; e-mail: michel.
perez@pierre-fabre.com
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00170-7

1456
M. Perez et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1455–1458
displays some interesting structural analogies with
existing compounds of the literature.16 To interact with
the zinc atom we added an imidazole ring, a well-known
alternative to the cysteine thiol group, on the amino
function. From this substituted scaffold we synthesized
and screened a first combinatorial library from which
we identified a hit (compound 1, Fig. 1) that displayed
an IC50 of 600 nM on the inhibition of FPTase. Even
modest, this activity was sufficient to decide to optimize
this compound.
This intermediate can then undergo a diverse set of
reactions with either acid chlorides, sulfonyl chlorides,
isocyanates or isothiocyanates followed by cleavage of
the resin in acidic conditions to afford small libraries of
compounds (Scheme 2). Among these compounds, the
highest inhibitory activities were obtained with thiourea
derivatives, which surprisingly and inexplicably, had
lower IC50 values than the corresponding ureas (Table
1, compare 8b and 8d to 8i and 8g). However, the most
potent compounds in this series were poorly active in a
cellular Ras-processing assay13 (8a and 8b).
We then focused our attention on modification of the
left hand part of compound 1. We first decided to sub-
stitute the imidazole by a 4-cyanobenzyl group11
(Scheme 3). 4-Nitro-phenyl piperazine 2 was first trea-
ted with 5-chloro-thiophene-2-carboxylic acid to give
compound 9, which was then reduced with tin chloride
to give amine 10. Reductive coupling with 4-cyano-
benzyl imidazole carboxaldehyde afforded compound
11. From this intermediate we were able to prepare
compound 12c which corresponds to the imidazole-
substituted equivalent of compound 1. This modifi-
cation led to a major reduction in the IC50 value of 12c
as compared to 1 (2 vs 600 nM respectively). However,
this compound also demonstrated only poor cellular
activity in the Ras-processing assay (IC50=200 nM).
Figure 1. Structure of lead compound 1.
Optimization of the right hand part of compound 1 has
been realized using solid phase parallel synthesis, since
anchorage on a solid support was possible using one of
the imidazole nitrogens. Synthesis of the first inter-
mediate resin is described in Scheme 1. 4-Nitro-phenyl
piperazine 2 is first protected with a fmoc group and
then reduced by catalytic hydrogenation to give com-
pound 3. Imidazole carboxaldehyde was regio-selec-
tively anchored to a PS-chlorotrityl chloride resin and
then reductively aminated with compound 3 to provide
intermediate 6. This resin was treated with phenyl sul-
fonyl chloride in pyridine followed by removal of the
fmoc protecting group to give resin 7.
To optimize the cellular activity we decided to focus our
efforts on a modification or replacement of the phenyl
sulfonyl group. We used solution phase parallel synth-
esis with polymer-supported reagents and resin sca-
vengers in order to rapidly optimize this series of
compounds (Scheme 4). Reactions with sulfonyl chlor-
ides or acid chlorides were performed using PS-DIEA as
Scheme 2. Reagents and conditions: (a) RNCO or RNCS, Toluene,
70 ꢀC, 10 h; (b) RCOCl or RSO2Cl, DIEA, DCM, rt, 4 h; (c) TFA,
DCM, Et3SiH, 1h.
Table 1.
Compd
R
FPTasea Processingb GGPTasea
IC50 (nM) IC50 (mM) IC50 (nM)
8a
8b
8c
8d
8e
8f
8g
8h
8i
CS–NH–iBu
CS–NH–cHexyl
CS–NH–CH2–cHexyl
CS–NH–CH2–Ph
CS–NH–(CH2)2–Ph
CS–NH–Ph
CO–NH–CH2–Ph
CONHPh
CO–NH–cHexyl
6
20
30
60
100
600
600
1000
1000
0.8
1
>1000
>1000
>1000
>1000
>1000
>1000
nd
nd
nd
nd
nd
nd
nd
nd
nd
nd
Scheme 1. Reagents and conditions: (a) Fmoc-OSu, DMF, Na2CO3,
rt, 30 min; (b) H2, Pd/C, MeOH/THF; (c) DIEA, DMF; (d) NaBH
(OAc)3, AcOH, DCE, rt, 7 h; (e) PhSO2Cl, Pyridine, rt, 4 h (f) piper-
idine, DMF.
aFPTase and GGPTase I inhibitory assays were performed as described
in ref 12 (n=2).
bFarnesylation of H-Ras in DLD-1cells was monitored. 13

M. Perez et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1455–1458
1457
Table 2.
Compd
R
FPTasea
IC50 (nM)
Processingb
IC50 (mM)
GGPTasea
IC50 (nM)
12a
12b
12c
12d
12e
12f
12g
12h
12i
H
COPh
SO2Ph
70
2
2
2
2
2
2
2
2
nd
0.003
0.2
0.002
0.02
0.006
0.01400
>1
nd
>1000
nd
>1000
nd
>1000
CO–3-(Cl)Ph
CO–2-(F)Ph
CO–3-(F)Ph
CO–4-(F)Ph
CO–2-(OMe)Ph
CO–3-(OMe)Ph
CO–4-(OMe)Ph
CH2–3-(F)Ph
CH2–4-(F)Ph
0
nd
0.007
0.08
nd
700
nd
nd
nd
12j
12k
12l
2
20
70
nd
aFPTase and GGPTase I inhibitory assays were performed as described
in ref 12 (n=2).
bFarnesylation of H-Ras in DLD-1cells was monitored. 13
the methoxy substituent, depending on its position on
the phenyl group (compare 12h, 12i and 12j). The pre-
sence of a carbonyl or sulfonyl group seems necessary to
obtain good IC50 values, as demonstrated by the com-
pounds prepared by reductive amination (compare 12k
and 12l to 12f and 12g). For the most interesting com-
pounds, GGPTase inhibition was measured and the
results obtained demonstrated a very high selectivity of
these compounds for FPTase over GGPTase (Table 2).
Overall, the results reported here show a very interesting
in vitro profile for 4 compounds (12b, 12d, 12f and 12i).
Among these, 12d (Fig. 2) demonstrated equal inhibitory
activity on both the isolated enzyme and the Ras-pro-
cessing cellular assay, associated with a high selectivity
for FPTase over GGPTase.
Scheme 3. Reagents and conditions: (a) 5-Cl-thiophene-2-carboxylic
acid, HOOBT, EDC, DIEA, DCM, rt, 4 h, 84%; (b) SnCl2.2H2O,
EtOH, reflux, 24 h, 94%; (c) 4-CN-benzyl imidazole carboxaldehyde,
DCE, NaBH(OAc)3, AcOH, rt, 24 h, 84%.
base and PS-trisamine as scavenger to remove excess
reagent. Reductive aminations were also performed
using MP-BH3CN as reducing agent and PS-trisamine
as scavenger to remove excess aldehyde. About 80
compounds were synthesized using these methodologies
and data on a selection of the most interesting com-
pounds is summarized in Table 2.
From these results, it appears that replacement of the
sulfonyl substituent by a benzoyl group has no influence
on FPTase inhibitory activity but led to a real
improvement of the IC50 value on the cellular Ras-pro-
cessing assay (compare 12b and 12c). Substitution of the
benzoyl group gave very surprising results in terms of
Ras-processing with important variability, in the case of
Using the coordinates of a recently published X-ray
crystal structure of human FPTase,14,15 compound 12d
was docked into the active site using the flexible docking
Scheme 4. Reagents and conditions: (a) (i) RCOCl, DCM, PS-DIEA,
rt, 1h; (ii) PS-Trisamine, 5 h; (b) (i) RSO 2Cl, DCM, PS-DIEA, rt, 1h;
(ii) PS-Trisamine, 5 h; (c) (i) RCHO, MeOH, AcOH, rt, 3 h then MP-
BH3CN, rt, overnight, 48 h; (ii) PS-Trisamine, rt, overnight.
Figure 2. Proposed binding mode of compound 12d in human FPTase
active site. Connolly surface and lipophilic potential are represented.
For clarity, isoprenoid chain of FPP was colored in purple.

1458
M. Perez et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1455–1458
program FlexX (Tripos Inc.) (Fig. 2). On the Connolly
surface of the active site the lipophilic potential, calcu-
lated using the MOLCAD program (Tripos Inc.), has
been represented. The cyanobenzyl group was used as
starting fragment and placed in the same position as the
cyanobenzyl in the crystallized complex, making stack-
ing with FPP isoprenoid and pointing towards Arg202b
and Tyr166a. Then, FlexX has positioned the rest of the
molecule, fragment by fragment, into the active site
searching for favourable interactions between the ligand
and the amino acid residues. The docking program cal-
culated an interaction between the imidazole and the
zinc cation as anticipated by analysis of the crystal
structure. The aminophenyl piperazine group fits into a
hydrophobic pocket defined by Trp102b, Trp106b and
Tyr361b, which, in the case of CA1A2X peptide sub-
strates, hosts the aliphatic amino acid A2. The chlor-
obenzoyl group interacts with an additional pocket
constituted by Tyr361b, Trp106b and Leu96b. The
chlorothiophene moiety is placed in the hydrophobic
pocket constituted by Tyr131a and His149b interacting
usually with the methionine residue of CAAM peptide
substrates, while the carbonyl oxygen forms a hydrogen
bond with the Trp 102b.
References and Notes
1. Reiss, Y.; Stradley, S. J.; Gierasch, L. M.; Brown, M. S.;
Goldstein, J. L. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 732.
2. 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. J. Biol. Chem. 1991, 266, 14603.
3. Barbacid, M. Ann. Rev. Biochem. 1987, 56, 779.
4. Kato, K.; Cox, A. D.; Hisaka, M. M.; Graham, S. M.;
Buss, J. E.; Der, C. J. Proc. Natl. Acad. Sci. U.S.A. 1992, 89,
6403.
5. Casey, P. J.; Solski, P. A.; Der, C. J.; Buss, J. E. Proc. Natl.
Acad. Sci. U.S.A. 1989, 86, 8323.
6. Clarke, S. Ann. Rev. Biochem. 1992, 61, 355.
7. (a) Prendergast, G. C. Curr. Opin. Cell. Biol. 2000, 12,
1663. (b) Ashar, H. R.; James, L.; Gray, K.; Carr, D.; Black,
S.; Armstrong, L.; Bishop, W. R.; Kirschmeier, P. J. Biol.
Chem. 2000, 275, 31451.
8. Reiss, Y.; Goldstein, J. L.; Seabra, M. C.; Casey, P. J.;
Brown, M. S. Cell 1990, 62, 81 .
9. Lamothe, M.; Perez, M. IDrugs 2000, 3, 1336.
10. Gibbs, R. A.; Zahn, T. J.; Sebolt-Leopold, J. S. Curr.
Med. Chem. 2001, 8, 1437.
11. Ciccarone, T. M.; MacTough, S. C.; Williams, T. M.;
Dinsmore, C. J.; O’Neill, T. J.; Shah, D.; Culberson, J. C.;
Koblan, K. S.; Kohl, N. E.; Gibbs, J. B.; Oliff, A. I.; Graham,
S. L.; Hartman, G. D. Bioorg. Med. Chem. Lett. 1999, 9, 1991.
12. Pompliano, D. L.; Gomez, R. P.; Antony, N. J. J. Am.
Chem. Soc. 1992, 114, 7945.
13. Gibbs, J. B.; Pompliano, D. L.; Mosser, S. D.; Rands, E.;
Lingham, R. B.; Singh, S. B.; Scolnick, E. M.; Kohl, N. E.;
Oliff, A. J. Biol. Chem. 1993, 268, 7617.
14. Long, S. B.; Hancock, P. J.; Kral, A. M.; Hellinga, H. W.;
Beese, L. S. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 12948.
15. Bell, I. M.; Gallicchio, S. N.; Abrams, M.; Beese, L. S.;
Beshore, D. C.; Bhimnathwala, H.; Bogusky, M. J.; Carolyn,
A.; Culberson, J. C.; Davide, J.; Ellis, H.; Fernandes, C.; Gibbs,
J. B.; Graham, S. L.; Hamilton, K. A.; Hartman, G. D.; Heim-
brook, D. C.; Homnick, C.; Huber, H. E.; Huff, J. R.; Kassa-
hun, K.; Koblan, K. S.; Nancy, E.; Lobell, R. B.; Lynch, J. J.,
Jr.; Robinson, R.; Rodrigues, A. D.; Taylor, J. S.; Walsh, E. S.;
Williams, T. M.; Zartman, C. B. J. Med. Chem. 2002, 45, 2388.
16. Dinsmore, C. J.; Williams, T. M.; O’Neill, T. J.; Liu, D.;
Rands, E.; Culberson, J. C.; Lobell, R. B.; Koblan, K. S.;
Kohl, N. E.; Gibbs, J. B.; Oliff, A. I.; Graham, S. L.; Hart-
man, G. D. Bioorg. Med. Chem. Lett. 1999, 9, 3301.
In summary, 4-amino-phenyl piperazine derivatives
were found to be effective templates that give rise to a
series of potent inhibitors of FPTase displaying nano-
molar activities on cellular Ras-processing assay. Mole-
cular modelling demonstrates that these compounds act
as CAAX analogues and interact efficiently with the
different parts of the active site. The most interesting
molecules are in the process of being evaluated in-vivo
and these results will be reported in due course.
Acknowledgements
The authors wish to thank S. Gras and D. Perrin from
the Department of Cancerology and E. Amalric, M.
Pelissou and J.-P. Ribet from the Department of
Chemistry.