Synthesis and biological evaluation of 1-benzyl-5-(3-biphenyl-2-yl-propyl)- 1H-imidazole as novel farnesyltransferase inhibitor

Article abstract of DOI:10.1016/j.bmcl.2004.07.083

Analogs of compound 1 were synthesized and tested in vitro for farnesyltransferase inhibition activity.

Full text of DOI:10.1016/j.bmcl.2004.07.083

Bioorganic& Medicinal Chemistry Letters 14 (2004) 5057–5062
Synthesis and biological evaluation of
1-benzyl-5-(3-biphenyl-2-yl-propyl)-1H-imidazole
as novel farnesyltransferase inhibitor
Nan-Horng Lin,* Le Wang, Xilu Wang, Gary T. Wang, Jerry Cohen, Wen-Zhen Gu,
Haiying Zhang, Saul H. Rosenberg and Hing L. Sham
Cancer Research, R-47B, Global Pharmaceutical Products Division, Abbott Laboratories, 100 Abbott Park Road,
Abbott Park, IL 60064-3500, USA
Received 21 July 2004; revised 29 July 2004; accepted 30 July 2004
Available online 3 September 2004
Abstract—Farnesyltransferase inhibitors (FTIs) have emerged as a novel class of anti-cancer agents. Analogs of the potent FTI,
1-benzyl-5-(3-biphenyl-2-yl-propyl)-1H-imidazole, were synthesized and tested in vitro for their inhibitory activities. The most
promising compound identified from this series is analog 29 that possesses potent enzymatic and cellular activities.
Ó 2004 Elsevier Ltd. All rights reserved.
GTP-bound Ras proteins are responsible for initiating
an intracellular phosphorylation cascade, and conse-
quently play an important role in normal cellular phys-
iology and pathophysiology.1 Oncogenic Ras proteins
commonly found in human tumors1,2 are locked in the
activated GTP-bound state, which leads to a continu-
ously activated phosphorylation cascade. An essential
prerequisite for the function of the Ras protein is its
association with the plasma membrane. Ras proteins
are initially synthesized in the cytoplasm, where the
pre-Ras protein undergoes posttranslational farnesyl-
ation of the cysteine unit of the so-called CAAX box
(C, cysteine; A, any aliphatic amino acid; X, serine or
methionine) by the enzyme protein farnesyltransferase
(FT).3 Once the protein substrate is farnesylated, the
AAX tripeptide is cleaved and the new C-terminal cys-
teine carboxylate is methylated. The processed proteins
become relocated to the cell membrane. Here then trans-
mit extracellular signals to the nucleus that lead to cell
proliferation.4,5
ing the posttranslational prenylation and oncogenic
activity of Ras. It has become apparent that inhibition
of Ras prenylation is not necessary for these compounds
to exhibit antitumor activity. Instead, inhibition of
Rho-B and possibly other cellular proteins might also
account for the efficacy against malignant tumors.11–14
Although the mechanism of action of these agents is still
debated, FTIs have shown impressive efficacy in preclin-
ical models of human cancers.
The goal of our research is to identify additional mole-
cules within this class of FTIs that not only have good
in vitro potency but may also have different physical
properties (i.e., solubility and crystallinity). Our initial
discovery of compound 1 as a potent, nonpeptidic,
nonsulhydryl, selective inhibitor of FT prompted an
investigation of SAR centered around this compound.
We have previously discovered that modification of
the phenyl ring of the biaryl skeleton in 1 affected the
inhibitory potency of analogs.15 We have also thor-
oughly examined the effect of the linker of biaryl moeity
on the potency.16 It has been demonstrated that intro-
duction of an acetylenic or vinyicl linkage between the
two biphenyl rings has little impact on the potency. To
continue our search for more potent FTIs, we then
turned our attention to identify the new chemotype of
FTIs. After examination of the X-ray cocrystal struc-
tural17 of compound 1, we thought we might be able
to move the 4-cyanophenyl ring to imidazole moiety,
as shown in Figure 1, to give compound A. It was
The antitransforming properties of farnesyltransferase
inhibitors (FTIs), a novel class of cancer therapeutics,
have been widely investigated in the past decade.6–10
FTIs were originally developed with the aim of inhibit-
Keyword: FTI.
*
Corresponding author. Tel.: +1 8479388797; fax: +1 8479355466;
e-mail: nanhorng.lin@abbott.com
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.07.083

5058
N.-H. Lin et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5057–5062
CN
N
N
N
N
OH
OH
NC
OEt
OEt
CN
CN
1
A
N
N
N
N
HN
O
NC
NC
OEt
OEt
CN
CN
B
C
Figure 1.
anticipitated that these modifications would increase or
at least maintain the potency. We also planned to re-
place the acetylenic moeity with an amino linker to give
compound B or compound C if the compound A turns
out to be a weak inhibitor or possesses an undesired bio-
logical profile in vitro.
The ether analogs were prepared using a reaction se-
quence similar to that shown in Scheme 2. For example,
treatment of 2 with bromide 13 in the presence of potas-
sium hydroxide gave the ether analog 23.
Acetylenic, amino, and ether analogs of 4-cyanobenzyl
imidazole FTIs were tested for their inhibitory activity
against both FTase and GGTase-1 enzymes. The effects
of moving the 4-cyanopheny moiety to the imidazole
ring of the phenyl acetylene 1 is shown in Table 1 with
two representative examples. The potency of compound
10 (Table 1) was similar to that of lead compound 1 (Ta-
ble 2) in both the FTase and GGTase-1 assays, and they
also possess similar selectivity against FTase. Unfortu-
nately, compound 10 shows 167-fold decrease in Ras
processing potency compared with the lead compound
1. A similar SAR trend was also observed when the eth-
oxy moeity was replaced with a methoxy group (16 vs
17). Compared with 17, a 21-fold decrease in cellular po-
tency was observed with compound 16. Since analogs 10
and 16 provide weaker inhibitors, we then turned our
attention to the replacement of the acetylenic linker with
an aminomethyl moiety as described previously. It was
hoped that this modification would increase the cellular
potency.
An example of the preparation of the acetylenic alcohol
analogs is depicted in Scheme 1. Formation of the alde-
hyde 3 was accomplished by oxidation of 5-hydroxy-
methylimidazole18 2 with pyridinium sulfur trioxide.
Biphenyl acetylene 10 was prepared from the corre-
sponding iodophenyl 4 via a five-step reaction sequence.
Thus, coupling of 4 with phenylboronicaicd 5 under
Suzuki reaction conditions provided the biphenyl
compound 6. Treatment of phenol 6 with N-phenyl-
bis(trifluoromethanesulfonimide) in the presence of tri-
ethylamine gave the triflate 7, which was coupled with
trimethylsilyl acetylene utilizing Sonogashira reaction
conditions to give the acetylene analog 8. Removal of
the trimethylsilyl group with base provided the acetylene
9, which was easily converted into lithium acetylide by
treatment with tert-butyl lithium. Addition of lithium
acetylide to the imidazole aldehyde 3 generated the de-
sired alcohol 10.
The amino analog 15 was synthesized using the reaction
sequence shown in Scheme 2. Thus, coupling of a bro-
mo-toluene 11 with boronicacid 5 under Suzuki reac-
tion condition provided the biphenyl toluene 12.
Bromination of the toluene 12 with NBS gave the corre-
sponding bromide 13 in good yield. Treatment of bro-
mide 13 with ammonia furnished the aminomethyl
compound 14, which then underwent reductive amina-
tion reaction with imidazole aldehyde 3 to provide the
desired amino analog 15.
As indicated in Table 3, replacement of the acetylene
moiety with an amino methyl linker resulted in a 52-fold
increase in potency against FTase enzyme. In contrast,
only a small change was observed in the potency against
GGTase-1 (1 vs 18), and consequently the selectivity im-
proved 56-fold. Unfortunately, the cellular activity of
compound 18 is 13-fold less potent than compound 1.
Furthermore, compound 18 has a poor pharmacokinetic
(PK) profile, which we speculate might be derived from
oxidation of the amino group. To avoid this problem,

N.-H. Lin et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5057–5062
5059
N
N
OH
i
CHO
N
N
CN
CN
3
2
OH
CN
OH
OTf
CN
B(OH)2
I
ii
iii
OEt
OEt
+
OEt
CN
7
4
6
5
TMS
H
v
iv
OEt
OEt
8
9
CN
CN
CN
H
N
N
N
vi
CHO
N
+
OH
OEt
OEt
CN
CN
10
CN
Scheme 1. Reagents and conditions: (i) PySO3, DMSO; (ii) Pd(PPh3)4, toluene, dioxane; (iii) PhN(Tf)2, NEt3; (iv) PdCl2(PPh3)2, CuI, HCCTMS,
DMF; (v) K2CO3, MeOH; (vi) t-BuLi, 3, À78°C.
Br
B(OH)2
Br
i
ii
iii
OEt
OEt
+
OEt
CN
CN
CN
11
5
12
13
N
N
CN
N
N
H2N
H
N
H
iv
OEt
O
+
OEt
CN
CN
CN
14
3
15
Scheme 2. Reagents and conditions: (i) Pd(PPh3)4, CsF, DME; (ii) NBS, CH2Cl2; (iii) NH4OH; MeOH; (iv) NaCNBH4, HOAc.
we then introduced acetyl (20) or sulfonyl (21) groups to
the nitrogen atom. However, these changes resulted in
22- and 150-fold decreases in inhibitory potency against
FTase, respectively. In contrast, an approximately
4-fold decrease in potency against GGTase-1 was ob-
served. In general, the introduction of electron-with-
drawing groups to the nitrogen atom resulted in only
poor selectivity compared with that of compound 18.
We also found that the cellular activity of compounds
20 and 21 was much weaker than that of compound
18. Furthermore, the bioavailability of compound 21 is
zero in rat (unpublished result).
We also introduced a methyl group to the nitrogen atom
of aminomethyl series. This resulted in only a slight de-
crease in Ras processing potency compared with analog
18. Small changes in potency were also observed in the
enzymaticassays against both FTase and GGTase-1
(18 vs 19). Unfortunately, no improvement in the PK
profile of this compound was observed.
Replacement of acetylene with an aminomethyl
group resulted in increased potency against FTase
and better selectivity for FTase, but resulted in no
improvement in PK, we then turned our attention to

5060
N.-H. Lin et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5057–5062
Table 1. FTase inhibition, GGTase-1 inhibition, and Ras processing data for acetylenic analogs 10 and 16
CN
N
N
OH
R
CN
Compds
R
FTase IC50 (nM)a
GGTase IC50 (nM)b
Selectivity (GGT/FT)
Ras EC50 (nM)c
10
16
OEt
OMe
0.6
1.2
6600
12,000
11,000
10,000
16.7
57.5
a Concentration of compound required to reduce the human FTase-catalyzed incorporation of [3H] FPP into recombinant Ras CVIM by 50%.
b Concentration of compound required to reduce the human GGTase-catalyzed incorporation of [3H] GGPP into biotinylated peptide corresponding
to the C-terminal of human K-Ras by 50%.
c Compound concentration needed to reduce 50% of farnesylation in NIH-3T3 H-Ras cell line.
Table 2. FTase inhibition, GGTase-1 inhibition, and Ras processing data for acetylenic analogs 1 and 17
CN
N
OH
N
R
CN
Compds
R
Ftase IC50 (nM)a
GGTase IC50 (nM)b
Selectivity (GGT/FT)
Ras EC50 (nM)c
1
17
OEt
OMe
0.73
0.37
2400
4100
3288
11,081
0.1
2.74
a Concentration of compound required to reduce the human FTase-catalyzed incorporation of [3H] FPP into recombinant Ras CVIM by 50%.
b Concentration of compound required to reduce the human GGTase-catalyzed incorporation of [3H] GGPP into biotinylated peptide corresponding
to the C-terminal of human K-Ras by 50%.
c Compound concentration needed to reduce 50% of farnesylation in NIH-3T3 H-Ras cell line.
Table 3. FTase inhibition, GGTase-1 inhibition, and Ras processing data for acetylenic analogs 18–21
CN
N
N
R
N
R1
CN
Compds
R
R1
Ftase IC50 (nM)a
GGTase IC50 (nM)b
Selectivity (GGT/FT)
Ras EC50 (nM)c
18
19
20
21
H
Me
OEt
OEt
OEt
OEt
0.014
0.034
0.31
2.1
2600
1100
185,700
32,350
32,250
4760
1.34
1.57
16
COMe
SO2Me
10,000
10,000
100
a Concentration of compound required to reduce the human FTase-catalyzed incorporation of [3H] FPP into recombinant Ras CVIM by 50%.
b Concentration of compound required to reduce the human GGTase-catalyzed incorporation of [3H] GGPP into biotinylated peptide corresponding
to the C-terminal of human K-Ras by 50%.
c Compound concentration needed to reduce 50% of farnesylation in NIH-3T3 H-Ras cell line.
the replacement of the nitrogen atom with oxygen. Ether
analogs of 4-cyanobenzyl imidazole FTIs were tested for
their inhibitory activity against both FTase and
GGTase-1 enzymes. The effect of replacing nitrogen
atom with oxygen in the amino analog 18 is shown in
Table 4.

N.-H. Lin et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5057–5062
5061
Table 4. FTase inhibition, GGTase-1 inhibition, and Ras processing data for acetylenic analogs 22–29
CN
N
N
O
R
CN
Compds
R
FTase IC50 (nM)a
GGTase IC50 (nM)b
Selectivity (GGT/FT)
Ras EC50 (nM)c
22
23
24
25
26
27
28
29
3-OMe
3-OEt
3-Cl
0.46
0.27
0.37
0.37
1.2
10,000
3200
6800
990
21,700
11,850
18,380
2700
32
21.9
7.7
3-OCF3
4-OEt
<100
2300
1300
870
1920
2100
<100
52
>100
4-OCF3
4-t-Bu
3,4-OCH2O–
0.62
0.36
0.1
2420
32,000
3200
1.65
a Concentration of compound required to reduce the human FTase-catalyzed incorporation of [3H] FPP into recombinant Ras CVIM by 50%.
b Concentration of compound required to reduce the human GGTase-catalyzed incorporation of [3H] GGPP into biotinylated peptide corresponding
to the C-terminal of human K-Ras by 50%.
c Compound concentration needed to reduce 50% of farnesylation in NIH-3T3 H-Ras cell line.
As indicated in Table 4, replacement of the nitrogen
atom of the amino methyl analog resulted in a 19-fold
decrease in potency in the inhibition of FTase enzymes
(18 vs 23). In contrast to this result, little change in po-
tency was observed in the inhibition of GGTase-1. Con-
sequently, the selectivity is worse in the ether series than
the corresponding amino series (18 vs 23). In general,
ether analogs were less potent inhibitors of Ras
processing activity than their corresponding amino
analogs.
was further reduced by replacement of the trifluoro-
methoxy moiety with a tert-butyl group (cf. 27 and
28). The only acceptable substituent with regard to
Ras processing was found to be the methylenedioxy
moiety (29). In addition, compound 29 has oral
bioavailability of 11.3% in rat compared with the com-
plete lack of bioavailabilty observed in the compound
21.
In summary, we have varied the substituent pattern at
the imidazole ring of the imidazole acetylenic alcohol 1
and examined the resultant effects on inhibitory activity.
The results (Table 1) indicated that moving the 4-cyano-
phenyl moeity is not well tolerated. Replacement of
acetylene moiety of compound 18 with an amino methyl
linker increased the potency in inhibition of FTase en-
zyme and selectivity (Table 3), but the compound suf-
fered from poor oral bioavailability in rat. To further
improve the PK property of the inhibitors, we replaced
the nitrogen atom with oxygen. In general, compounds
having the ether linkage (Table 4) possessed potent
inhibitory activities against the FTase enzyme. The
highest selectivity for FTase inhibition over GGTase-1
was observed in compound 29. This compound is more
potent in inhibition of FTase enzyme and possesses bet-
ter selectivity. It also has reasonable bioavailability in
rat. Further SAR studies are in progress.
Since compound 23 provided only moderate enzymatic
and cellular activities, we then focused our attention
on SAR studies of the phenyl ring in the ether series.
The 3,4-methylenedioxy analog (29) was found to be
the most potent FTase inhibitor in this series so far.
Both electron-donating and withdrawing groups are tol-
erated at the C3- and C4-positions. To probe the steric
volume at the C4-position of the phenyl pocket we
replaced the trifluoromethoxy group with a bulky tert-
butyl group, and found this had little impact on the
potency (27 vs 28). The same phenomenon was also
observed at the C3-position.
With regard to activity against GGTase, the introduc-
tion of electron withdrawing groups such as trifluoro-
methyl or chloro resulted in an increase in potency. In
addition, a large group at the C4-position also caused
an increase in potency. In general, the ether series
showed only moderate selectivity with the exception of
compounds 22, 24, and 29, which have more than
15,000-fold selectivity in favor of FTase inhibition.
References and notes
1. (a) Clarke, S. Annu. Rev. Biochem. 1992, 61, 355; (b)
Sawyer, C. L.; Denny, C. T. Cell 1994, 77, 171.
2. Barbacid, M. Annu. Rev. Biochem. 1987, 56, 779.
3. Lowy, D. R.; Willumsen, B. M. Annu. Rev. Biochem. 1993,
62, 851.
4. Zhang, F. L.; Casey, P. J. Annu. Rev. Biochem. 1996, 65,
241.
5. Clarke, S.; Vogel, J. P.; Deschenes, R. J.; Stock, J. Proc.
Natl. Acad. Sci. U.S.A. 1988, 85, 4643.
The introduction of a trifluoromethoxy or methoxy
group to the C3-position of the phenyl ring in the ether
series was found to result in a decrease in Ras processing
potency compared with analog 23. Introduction of an
ethoxy or trifluoromethoxy group at the C4-position
caused a similar reduction in potency, and potency

5062
N.-H. Lin et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5057–5062
6. Gutierrez, L.; Magee, A. I.; Marshall, C. J.; Hancock, J. F.
EMBO J. 1989, 8, 1093.
14. Prendergast, G. C.; Du, W. Drug Resist. Update 1999, 2,
81.
7. 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. J.; Gibbs, J. B. Science 1993, 260,
1934.
8. James, G. L.; Goldstein, J. L.; Brown, M. S.; Rawson, T.
E.; Somers, T. C.; McDowell, R. S.; Crowley, C.; Lucas,
B. K.; Levinson, A. D.; Marster, J. C., Jr. Science 1993,
260, 1937.
15. Lin, N.-H.; Wang, L.; Cohen, J.; Gu, W.-Z.; Frost, D.;
Zhang, H.; Rosenberg, S.; Sham, H. Bioorg. Med. Chem.
Lett. 2003, 13, 1293.
16. Lin, N.-H.; Wang, L.; Cohen, J.; Gu, W.-Z.; Frost, D.;
Zhang, H.; Rosenberg, S.; Sham, H. Bioorg. Med. Chem.
Lett. 2003, 13, 3821.
17. Dinsmore, C.; Bell, I. M. Curr. Top. Med. Chem. 2003, 3,
1075–1093.
9. Nigam, M.; Seong, C.-M.; Qian, Y.; Hamilton, A. D.;
Sebti, S. M. J. Biol. Chem. 1993, 268, 20695.
10. Bell, I. M. J. Med. Chem. 2004, 47, 1869.
11. Wallace, A.; Koblan, K. S.; Hamilton, K.; Marqui-Omar,
D. J.; Miller, P. J.; Mosser, S. D.; Omer, C. A.; Schaber,
M. D.; Cortese, R.; Oliff, A. I.; Gibbs, J. B.; Pessi, A.
J. Biol. Chem. 1996, 271, 31306.
12. Prendergast, G. Curr. Opin. Cell Biol. 2000, 166.
13. Du, W.; Lebowitz, P. F.; Prendergast, G. C. Mol. Cell.
Biol. 1999, 19, 183.
18. Williams, T. M.; Bergman, J. M.; Brashear, K.; Breslin,
M. J.; Dinsmore, C. J.; Hutchinson, J. H.; MacTough, S.
C.; Stump, C. A.; Wei, D. D.; Zartman, C. B.; Bogusky,
M. J.; Culberson, J. C.; Buser-Doepner, C.; Davide, J.;
Greenberg, I. B.; Hamilton, K. A.; Koblan, K. S.; Kohl,
N. E.; Liu, D.; Lobell, R. B.; Mosser, S. D.; OÕNeill, T. J.;
Rands, E.; Schaber, M. D.; Wilson, F.; Senderak, E.;
Motzel, S. L.; Gibbs, J. B.; Graham, S. L.; Heimbrook, D.
C.; Hartman, G. D.; Oliff, A. I.; Huff, J. R., J. Med. Chem.
1999, 42, 3779–3784.