Functionalized indoleamines as potent, drug-like inhibitors of isoprenylcysteine carboxyl methyltransferase (Icmt)

Article abstract of DOI:10.1016/j.ejmech.2013.02.007

The enzyme isoprenylcysteine carboxyl methyltransferase (Icmt) plays an important role in the post-translational modification of proteins involved in the regulation of cell growth and oncogenesis. The biological consequences of Icmt inhibition strongly implicate the enzyme as a potential therapeutic target for cancer and provide a compelling rationale for developing specific Icmt inhibitors as anti-cancer agents. We report here the systematic modification of the known Icmt inhibitor cysmethynil to give an analog 15 with greatly improved solubility and PAMPA permeability which was achieved with concurrent gains in Icmt inhibitory and cell-based antiproliferative activities. The modifications involved replacing the amide side chain of cysmethynil with a tertiary amine, and introducing an aminopyrimidine ring in place of m-tolyl. The presence of the weakly basic and polar aminopyrimidine ring contributed significantly to the potency and drug-like profile of the final compound.

Full text of DOI:10.1016/j.ejmech.2013.02.007

European Journal of Medicinal Chemistry 63 (2013) 378e386
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Functionalized indoleamines as potent, drug-like inhibitors of
isoprenylcysteine carboxyl methyltransferase (Icmt)
Pondy M. Ramanujulu a, Tianming Yang a, Siew-Qi Yap b, Fui-Chung Wong a,
,
Patrick J. Casey c, Mei Wang c, Mei-Lin Go a
*
a Department of Pharmacy, National University of Singapore, 18 Science Drive 4, Singapore 117543, Singapore
b Drug Development Unit, Life Sciences Institute, National University of Singapore, 28 Medical Drive, Singapore 117456, Singapore
c Program in Cancer and Stem Cell Biology, Duke-NUS Graduate Medical School, Singapore 169857, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history:
The enzyme isoprenylcysteine carboxyl methyltransferase (Icmt) plays an important role in the post-
translational modification of proteins involved in the regulation of cell growth and oncogenesis. The
biological consequences of Icmt inhibition strongly implicate the enzyme as a potential therapeutic
target for cancer and provide a compelling rationale for developing specific Icmt inhibitors as anti-cancer
agents. We report here the systematic modification of the known Icmt inhibitor cysmethynil to give an
analog 15 with greatly improved solubility and PAMPA permeability which was achieved with concur-
rent gains in Icmt inhibitory and cell-based antiproliferative activities. The modifications involved
replacing the amide side chain of cysmethynil with a tertiary amine, and introducing an amino-
pyrimidine ring in place of m-tolyl. The presence of the weakly basic and polar aminopyrimidine ring
contributed significantly to the potency and drug-like profile of the final compound.
Received 12 December 2012
Received in revised form
6 February 2013
Accepted 8 February 2013
Available online 21 February 2013
Keywords:
Isoprenylcysteine carboxyl
methyltransferase (Icmt) inhibitors
Indoleamines
Ó 2013 Elsevier Masson SAS. All rights reserved.
Antiproliferative activity
Solubility
PAMPA permeability
Drug-like properties
Metabolic stability
1. Introduction
for Icmt in the oncogenic transformation of K-Ras [5,8,9] which
has in turn stimulated interest in its potential as a putative target
Several proteins involved in oncogenesis and tumor progres-
sion, notably the Ras family of G proteins, possess a C-terminal
CaaX motif (where C is cysteine, a is usually an aliphatic amino acid
and X denotes any amino acid) which is modified by prenylation in
a three-step post-translational process. The process starts with the
attachment of a C15 farnesyl or C20-geranylgeranyl isoprenoid res-
idue to the thiol of cysteine by protein farnesyl transferase or
protein geranylgeranyl transferase type 1 [1,2]. This is followed by
the removal of the -aaX residue by a prenyl-protein specific pro-
tease Rce-1 [3] and then methylation of the newly exposed car-
boxylic acid residue of prenylcysteine by isoprenylcysteine carboxyl
methyltransferase (Icmt) [4,5]. Investigations have shown that
disrupting the activities of prenylation enzymes led to defective Ras
processing, its mislocalization within cells and aberrant cellular
function [6,7]. There is compelling evidence to support a key role
for cancer and consequently, the anti-cancer properties of Icmt
inhibitors.
Several Icmt inhibitors have been reported to date [5,10e12], of
which many are based on “minimal substrates” of Icmt, namely
N-acetyl-S-farnesyl-L-cysteine (AFC) and farnesyl thiopropionic
acid (FTPA). Well-designed medicinal chemistry approaches have
succeeded in reducing the substrate profile of these scaffolds
while enhancing inhibitory activity [11]. Some of the more
promising substrate-based Icmt inhibitors are shown in Fig. 1.
Besides substrate-based inhibitors, a few small molecule Icmt in-
hibitors have been reported. Judd and co-workers observed
nanomolar Icmt inhibitory activity for a tetrahydropyranyl deriv-
ative [12] (Fig. 1). The indole derivative cysmethynil (1, Fig. 1) is the
earliest small molecule Icmt inhibitor to be reported and remains
the most widely investigated to date [5,13,14]. It has been shown
to specifically target Icmt, elicit cell death and inhibit growth of
malignant cell-derived xenograft tumors in mice. It is however a
highly lipophilic molecule (Clog P 7) with poor water solubility
and these properties are likely to hinder its delivery. While these
* Corresponding author. Tel.: þ65 65162654; fax: þ65 68721610.
E-mail address: phagoml@nus.edu.sg (M.-L. Go).
0223-5234/$ e see front matter Ó 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2013.02.007

P.M. Ramanujulu et al. / European Journal of Medicinal Chemistry 63 (2013) 378e386
379
Fig. 1. Structures of Icmt inhibitors: substrate-based inhibitors based on AFC (a, b) and FTPA (c, d) scaffolds; Tetrahydropyranyl derivative (e) and cysmethynil (f).
constraints may be overcome by appropriate formulations, they
can also be addressed by structural modifications aimed at
enhancing potency and drug-likeness. Our earlier investigations
have shown that replacing the acetamide side chain of cysme-
thynil with diethylaminomethyl, as in 2 (Fig. 2), greatly improved
cell-based growth inhibition while retaining the same level of Icmt
inhibitory activity [10]. In spite of its protonated state, the esti-
mated lipophilicity of 2 at pH 7.4 (log D7.4) remained high at 6.4, as
compared to 6.9 for cysmethynil. Its solubility at pH 7.4 was esti-
substituents or nitrogen containing heteroaromatic rings in place of
the m-tolyl ring position 5. The estimated log D7.4 values of these
compounds suggest that the lipophilicity of 2 may be favorably
moderated by these structural replacements (Table 1). Here we
identified compound 15 as a promising drug-like substitute to 2
(Fig. 2). Compound 15 which was derived from 2 by replacing the
m-tolyl at position 5 by a heteroaromatic ring (2-aminopyridimin-
5-yl), retained good Icmt and growth inhibitory activities on a panel
of malignant cells and exceeded 2 in terms of solubility and PAMPA
permeability.
mated by ACD/Labs (Version 12.0) to be 35
exceeded that of cysmethynil (0.13 g/mL or 0.3
modest by drug-like standards where a value of ca 50
m
g/mL (87
M), but is still
g/mL has
mM) which
m
m
m
2. Results
been cited for a compound with “average” potency and perme-
ability [15].
2.1. Synthesis of target compounds
In an earlier assessment of the structure activity relationship of
cysmethynil, we showed that of the three functionalities (m-tolyl at
position 5, n-octyl at position 1, acetamide at position 3) attached to
the indole ring of cysmethynil, the acetamide side chain at position
3 was the most critical for Icmt inhibition. Removal of the side chain
yielded a compound that was devoid of Icmt inhibitory activity as
well as growth inhibitory activity on the breast cancer cell line
MDA-MB-231 [10]. Of the groups at positions 1 and 5, the m-tolyl
ring ranked behind the n-octyl side chain in terms of its impact on
Icmt inhibition. Hence we proposed that position 5 may be usefully
exploited in lead optimization as a focal point for the introduction
of groups with more favorable physicochemical profiles [10]. Based
on this premise, we designed a series of compounds structurally
related to 2 but with fluoro or phenyl rings bearing polar
The structures of the target compounds are given in Table 1. The
syntheses of these compounds were undertaken following re-
ported methods [10]. Briefly, the indole NH of 5-bromoindole was
N-alkylated in the presence of sodium hydride to give 5-bromo-1-
octyl-1H-indole (20). The diethylaminoemethyl side chain was
introduced at position 3 of 20 by the Mannich reaction. The
resulting compound (21) was then reacted in a Suzuki reaction with
the relevant boronic acid and palladium catalyst tetrakis(-
triphenylphosphine) palladium (0), to give compounds 2e15
(Scheme 1).
The same reaction sequence was followed for the fluorinated
analogs 16e19, starting from commercially available 4-fluoro,
5-fluoro, 6-fluoro and 7-fluoroindoles (Scheme 2).
Fig. 2. Modification of cysmethynil 1 to compound 2.

380
P.M. Ramanujulu et al. / European Journal of Medicinal Chemistry 63 (2013) 378e386
Table 1
2.2. Evaluation of Icmt inhibitory activity
5-Substituted analogs of compound 2: estimated log D (pH 7.4), IC50 for Icmt inhi-
bition and antiproliferative IC50 on human breast cancer MDA-MB-231 and human
prostate cancer PC3 cells.
Icmt activity was determined by measuring the rate at which the
enzyme catalyzed the transfer of a methyl group from the methyl
donor S-adenosylmethionine (SAM) to the substrate, biotin-S-far-
nesyl-L-cysteine (BFC). BFC is a farnesylated, Rce-1-protetolyzed
substrate of Icmt. The biotin in BFC facilitates the isolation of the
methylated substrate from the reaction mixture due to its affinity for
streptavidin. Briefly, the enzyme and test compound are incubated,
followed by addition of a cocktail containing BFC, SAM and tritiated
SAM. In the presence of an inhibitor, methylation of BFC is reduced
leading to diminished light emission. Inhibition is quantified by
comparing the reduction in light emission relative to the control and
expressed as IC50 of test compound (concentration required to
reduce basal Icmt activity by 50%). The method is convenient as it
does not require the separation of the bound and free tritiated SAM,
which was necessary in the previous protocol [10].
Cpd
no
R
log D
IC50 Icmt
(m
M)b
Antiproliferative IC50 (m
M)c
(pH 7.4)a
MDA-MB-231
PC3
2
3
6.37
5.23
2.18
5.43
7.90 (1.72)
6.31 (0.20)
18.1 (1.9)
14.9 (0.9)
As seen from Table 1, Icmt inhibitory activity varied within a
narrow 6-fold range. This was largely anticipated as the compounds
carry optimal substituents at positions 1 and 3 (as established from
our earlier investigations [10]) and modifications are restricted to
position 5 which is known to have minimal impact on activity. In
spite of the narrow range of Icmt inhibitory activity, there were
some discernible trends. Notably, replacing m-tolyl with a polar
heteroaromatic ring at position 5 was clearly a superior approach
than the other modifications. The most active Icmt inhibitors
(12,15) identified here were substituted pyridine and pyrimidine
analogs that had submicromolar IC50 values. Among the substituted
phenyl analogs (2e9), there was a broad preference for analogs
bearing the polar sulfonyl containing functionalities (methyl-
sulfonyl 7, aminosulfonyl 8) as well as the more lipophilic fluori-
nated analog 6. Intriguingly, the weakest inhibitor 3 was also
substituted with a polar group (hydroxyl). Clearly, the lipophilic
character of the substituent played a small role in affecting Icmt
inhibitory activity. The size of the group at position 5 was also not
important, as seen from the fluorinated analogs 16e19 which had
inhibitory activities that were comparable to the other ring
substituted analogs.
4
5
5.32
5.93
2.45
2.56
10.85
6.11 (0.16)
7.28 (2.57)
7.03 (0.81)
6
5.83
1.61
6.71 (0.21)
6.25 (0.67)
7
8
4.63
4.81
1.72
1.71
5.56 (0.58)
10.24
4.74 (1.31)
6.48 (0.17)
9
6.44
4.58
2.61
1.75
6.53 (0.49)
5.50 (0.81)
6.81 (0.53)
5.84 (0.31)
10
11
12
4.16
4.63
1.63
0.96
2.52 (0.77)
5.14 (0.67)
1.92 (0.22)
5.88 (0.44)
2.3. Evaluation of growth inhibitory activities on malignant cell
lines
13
14
15
3.84
3.51
3.54
1.58
2.26
11.05
8.47 (1.31)
5.74 (0.32)
2.55 (0.46)
3.85 (0.19)
The test compounds were investigated for their effects on the
viability of two human cancer cell lines, breast MDA-MB-231 and
prostate PC3. Growth inhibitory IC50 values (Table 1) ranged from
0.86 (0.09) 2.63 (0.43)
1.9
m
M to 18
m
M, with most compounds exhibiting equal activity on
16
17
18
19
F
4.00
4.20
4.13
5.60
6.90
4.83
4.42
2.22
2.27
7.50 (1.09)
7.36 (1.10)
9.14 (0.07)
8.87
ND
ND
ND
ND
both cell lines. Setting 5
mM as an arbitrary endpoint for anti-
F (position 4)
F (position 6)
F (position 7)
proliferative activity, it is noted that only compounds with
substituted heteroaromatic rings at position 5 complied with this
criterion, namely 11 (2-dimethylamino-pyridin-5-yl), 14 (2-cyano-
Cysmethynil (1)
1.9 (0.28)
27.4 (1.1)
25.2 (0.8)
a
Estimated with ACD/Labs 12.0.
Concentration required to reduce basal Icmt activity by 50%, mean of 2 separate
pyrimidin-5-yl, IC50 < 5 mM only on MDA-MB-231 cells) and 15
b
(2-amino-pyrimidin-5-yl). Compound 15 also demonstrated
outstanding Icmt inhibitory activity but in general, there was only
moderate correlation between the two activities for the other
compounds. As mentioned earlier, the compounds investigated
determinations or mean (ꢀSD) of 3 determinations. Actual IC50 values for are given
in Supplementary data.
c
Concentration required to reduce cell viability by 50% as determined by the MTS
assay. Conditions were 72 h incubation at 37 ꢁC, 5% CO2.
Scheme 1. Reagents and conditions: (a) 1-Bromooctane, NaH, DMF, 0 ꢁC to rt; (b) HCHO, diethylamine, acetic acid, rt, stirring overnight; (c) Boronic acid, Pd(PPh3)4, K2CO3, dioxane,
microwave 20 min 100 ꢁC.

P.M. Ramanujulu et al. / European Journal of Medicinal Chemistry 63 (2013) 378e386
381
Scheme 2. Reagents and conditions: (a) 1-Bromooctane, NaH, DMF, 0 ꢁC to rt; (b)
HCHO, diethylamine, acetic acid, rt, stirring overnight.
here were modified from a known hit compound (2) and already
incorporate features known to promote activity. Thus, they are
expected to be reasonably active and this is borne out by their
largely low micromolar IC50 values which span a narrow range for
both antiproliferative and Icmt inhibitory activities.
Fig. 3. Dose response curves of compound 15 on a panel of malignant cell lines.
2.4. Characterization of 15 for drug-like properties
Compound 15 was also assessed for antiproliferative activity on
non-malignant human lung fibroblast cells IMR90. Unfortunately,
the results showed that 15 did not selectively curtail the prolifer-
ation of malignant cells compared to normal cells to a significant
extent, and this was also true for cysmethynil and the other com-
pounds (2, 10 and 13). Notwithstanding the lack of discriminatory
activity on cell-based anti-proliferative assays, cysmethynil was
well tolerated in a dose escalation toxicity trail in mice at a dose of
300 mg/kg [13] and effectively controlled tumor growth in xeno-
graft mouse models of prostrate (PC3) and liver (HepG2) cells
[13,14]. Our unpublished results showed that 15 was also well
tolerated in mice (50 mg/kg) and able to reduce tumor growth in
HepG2-induced xenografts (unpublished results). While the
absence of selective antiproliferative activity in cell-based assays is
not desirable, it need not necessarily translate to in vivo toxicities,
as evident from cysmethynil and 15.
The aminopyrimidinyl analog 15 is the most promising com-
pound identified in our study. Its Icmt and cell-based inhibitory
profiles are superior to that of the original lead 2, and its lip-
ophilicity (log D7.4) is estimated to be lower by approximately 102
times. To further appropriate the potential of 15 as an advanced hit
candidate, we proceeded to characterize its effects on the viability
of additional cancer cell lines (pancreatic MIA Paca II and liver
HepG2) and to explore its drug like properties by determining its
kinetic solubility, PAMPA permeability, aggregation potential and
stability to metabolism by rat microsomes. The results are pre-
sented in Table 2, alongside those obtained for cysmethynil, 2, the
pyridine analog 10 and pyrimidine analog 13 for selected
properties.
Compound 15 reduced the viability of HepG2 and MIA Paca II
cells in a concentration dependent manner, like that observed for
MDA-MB-231 and PC3 cells (Fig. 3). The overlapping dose response
curves showed that 15 maintained the same level of potent activity
on all the cell lines. Its average IC50 on the 4 malignant cell lines
Compound 15 was estimated to be more soluble (0.37 mM, pH
7.4) than cysmethynil (3 ꢂ 10ꢃ4 mM, pH 7.4) and 2 (0.087 mM, pH
7.4). To verify these trends, the solubilities of cysmethynil, 2 and 15
were determined on MultiscreenÒ Solubility filter plates in phos-
phate buffer pH 7.4 after 24 h of shaking at 25 ꢁC. Cysmethynil was
was 2.07
mM, significantly surpassing that of cysmethynil and 2
(20.9 M and 5.45
m
mM respectively on 3 cell lines) in terms of
potency.
Table 2
Antiproliferative activity, physicochemical properties and in vitro metabolic data of 15 compared to cysmethynil (1), 2, 10 and 13.
Cysmethynil 1
2
10
13
15
Antiproliferative activitiesa
IC50 MDA-MB-231 M)
IC50 PC3 M)
IC50 HepG2 M)
IC50 MIA Paca II
IC50 IMR90 M)
(
m
22.1 (0.4)
21.3 (0.8)
19.3 (0.5)
eb
5.14 (0.12)
6.31 (0.20)
4.90 (0.39)
eb
6.01 (0.24)
5.84 (0.31)
eb
13.25 (0.16)
8.47 (1.31)
eb
2.48 (0.07)
2.55 (0.46)
1.65 (0.23)
1.60
(m
(
m
(
m
M)
eb
eb
(
m
29.2 (1.9)
5.26 (0.14)
5.52 (0.09)
9.84 (0.94)
2.65 (0.15)
Physicochemical data
Solubility (
m
M)c
1.14 (0.1)
Nild
1.10 (0.18)
Nild
244.8 (9.7)
271.56 (6.63)
155.9 (6.4)
14.2 (1.4)
PAMPA effective permeability (Pe) (ꢂ10ꢃ6 cm/s)c
eb
eb
Dynamic light scattering count rate (kcps)e
ꢃ10
m
M
M
164.0
22.6
366.5
24.8
16.5
16.8
28.9
20.4
53.3
24.5
ꢃ1
m
Deduced pharmacokinetic parameters from in vitro incubation with rat hepatic microsomes
Deduced half-life (min)
44.8 (8.0)
53.6 (9.0)
26.9 (0.7)
85.8 (2.1)
eb
eb
eb
eb
11.3 (0.4)
204.0 (9.0)
Deduced intrinsic clearance (
m
L/min/mg)
a
Mean (ꢀSD) of at least 3 determinations or mean of 2 separate determinations. IC50 values on MDA MB 231 and PC3 cells (only cysmethynil) were re-determined and
differed slightly from those listed in Table 1.
b
Not attempted.
c
Determinations were made at pH 7.4, 24 h agitation (solubility) or 16 h agitation (PAMPA Pe). Mean (ꢀSD) of 3 separate determinations.
d
Test compound could not be detected in acceptor wells by LCMS.
e
Mean count rates (kilocount per sec) from n ¼ 3 separate determinations, made at 2 concentrations 10
mM and 1 mM (1% DMSO, potassium phosphate buffer 5 mM pH 7.4).

382
P.M. Ramanujulu et al. / European Journal of Medicinal Chemistry 63 (2013) 378e386
indeed poorly soluble (1.14
mM) and replacing its primary amide
midazolam while 2 and cysmethynil were more stable to micro-
with an ionizable diethylamino side chain to give 2 did not improve
solubility, notwithstanding the presence of the ionizable basic
center in 2. Presumably the polarity associated with the charged
center in 2 is nullified by the presence of the other non-polar res-
idues (n-octyl, m-tolyl) in the molecule. This is borne out by the
observation that replacing the m-tolyl ring of 2 with polar hetero-
aromatic rings pyridine (10) and pyrimidine (13) significantly
improved solubility. Both pyridine and pyrimidine are weakly basic
rings that are not protonated at pH 7.4. Thus, their contribution to
the solubilities of 10 and 13 are likely due to their hydrogen bond
acceptor capabilities. When the pyrimidine ring of 13 is substituted
somal metabolism. The deduced half-lives of cysmethynil, 2 and 15
were of the order cysmethynil (45 min) > 2 (27 min) > 15 (11 min).
The shorter the half-life, the more rapid the clearance and
accordingly, 15 was predicted to be more rapidly cleared than 2 and
cysmethynil. These trends may be explained by the greater polar-
ity/lower lipophilicity of 15. A more accurate assessment of its
pharmacokinetic profile would require in vivo animal studies.
3. Discussion
The purpose of the present investigation is to explore design
strategies that would enhance the drug-like and potency profiles of
cysmethynil. Our initial approach was to replace the acetamide of
cysmethynil with a tertiary amino side chain [10]. This modification
greatly improved cell-based antiproliferative activity while main-
taining Icmt inhibitory activity for most compounds but as seen from
2, failed to overcome the solubility and permeability liabilities
associated with cysmethynil. These limitations were successfully
addressed by replacing the m-tolyl ring of 2 with a polar hetero-
aromatic ring, of which the 2-aminopyrimidine was found to be the
most promising. Other modifications like replacing m-tolyl with
substituted phenyl rings did not meet the desired activity endpoints.
Replacing the m-tolyl ring with the smaller flouro atom proved to be
disappointing although the losses in Icmt and antiproliferative ac-
tivities among the regioisomeric fluoro analogs (16e19) were sur-
prisingly limited in spite of the substantial size and electronic
differences between these two functionalities.
with an amino group to give 15, solubility decreased from 271
mM
(13) to 156 M (15). Interestingly, 15 is a solid unlike 10 and 13
m
which are viscous liquids. The change in state may be attributed to
stronger intermolecular interactions and closer packing of the
molecules in 15, presumably due to the inclusion of the hydrogen
bonding amino substituent. Tighter packing within the crystal
lattice of 15 would invariably lead to some loss in solubility [16].
Having demonstrated good solubility for 15, we proceeded to
assess it in vitro permeability using the parallel artificial membrane
permeation assay (PAMPA) with 1% lecithin in dodecane as the
simulated membrane barrier. The effective permeability (Pe) of 15
was 14.2 ꢂ 10ꢃ6 cm/s, surpassing that of cysmethynil and 2, which
could not be determined due to their poor solubilities. The Pe of 15
compared favorably to that of verapamil (10.2 ꢂ 10ꢃ6 cm/s) deter-
mined under similar conditions. Verapamil is considered as a highly
permeable compound [17] and the comparable Pe values of 15 and
verapamil is encouraging.
The 2-aminopyrimidine ring of 15 is weakly basic and intrinsi-
cally hydrophilic with pronounced hydrogen bonding capacities. Its
presence in 15 undoubtedly contributed to the remarkable im-
provements in solubility (pH 7.4) and PAMPA permeability. The
limited light scattering at pharmacologically relevant concentra-
tions of 15, which inferred a low liability as a non-specific inhibitor,
may also be attributed to the aminopyrimidine ring. Notably, these
drug-like properties were achieved with concurrent gains in Icmt
inhibitory and antiproliferative activities. On the other hand, the
polar character imparted by the aminopyrimidine ring affected
stability to rat microsomal degradation and 15 was deduced to have
a shorter half life and rapid clearance compared to 2 and cysme-
thynil. An accurate pharmacokinetic profile of 15 would have to
await animal experimentation. It may be that the in vivo half life of
15 will be moderated by its charged state (protonated diethylamino
side chain) which would predispose the compound to favorable
interactions with phospholipid membranes, thus leading to its
retention in tissues and a larger volume of distribution [21].
Notwithstanding the promising activity and physicochemical
profiles of 15, there are some concerns about the toxic potential of
the indoleamine chemotype present in 15 and the other target
compounds. Structurally, 15 is a cationic amphiphile due to the
combined presence of the lipophilic n-octyl side chain and the
Many reports have cautioned against non-specific inhibition of
enzymes due to the ability of some compounds to form aggregates
in aqueous media [18,19]. Compounds that form soluble or colloidal
aggregates normally exist as particles of 30e1000 nm diameter
which are detectable by light scattering [20]. The phenomenon is
concentration dependent and significant increases in aggregate
formation are observed with incremental increases (2e3 fold) in
concentration [20]. Here, we examined the light scattering
behavior of cysmethynil, 2, 10, 13 and 15 at 1
their IC50 Icmt values (1e2 M), and at a higher concentration of
10 M. Concentrations greater than 10 M were not investigated
due to the limited solubilities of some compounds. We found
minimal light scattering at 1 M, thus dispelling concerns that
these compounds may exhibit non-specific inhibition of Icmt. Light
scattering remained negligible at 10 M of 10, 13 and 15 but in-
mM which is close to
m
m
m
m
m
creases were observed for cysmethynil and 2. Thus some caution
should be exercised if pharmacological characterization of these
compounds are carried out at concentrations exceeding 10 mM.
The metabolic stabilities of cysmethynil, 2 and 15 were inves-
tigated by monitoring the time-dependent loss of the parent
compound when incubated with rat (male) liver microsomes
(Fig. 4). Compound 15 was cleared as rapidly as the positive control
Fig. 4. Percentages of test compound (-) and positive control midazolam ( ) relative to initial amounts (t ¼ 0) in rat liver microsomes on incubation at 37 ꢁC for 5, 15, 30 and
45 min. The percentages of remaining compound are expressed as the mean ꢀ SD (n ¼ 3).

P.M. Ramanujulu et al. / European Journal of Medicinal Chemistry 63 (2013) 378e386
383
ionizable tertiary amino functionality in the same molecule. These
entities may potentially induce hERG channel blockade [22] and a
lysosomal storage disorder called phospholipidosis [23]. As these
toxic liabilities will be more prominent in compounds with greater
amphiphilic character, we assessed the amphiphilicity of selected
compounds by an in silico approach which involved determination
of the free energy of amphiphilicity (DDGAM) [23]. A lower value of
ꢃ11.52 kJ/mol was obtained for 15 as compared to 2 (ꢃ13.04 kJ/
mol). As anticipated, cysmethynil did not possess an amphiphilic
moment. Although the amphiphilicity of 15 is less than 2, it is
admittedly still substantial and warrants further investigation as to
how amphiphilicity may be moderated by structural modifications.
(20 min, 100 ꢁC) with stirring. The resulting dark colored solution
was extracted with ethyl acetate (25 mL ꢂ 2), the combined organic
layers were washed consecutively with water and brine, and dried
(anhydrous Na2SO4). The organic solvent was removed in vacuo
and the residue purified by column chromatography with CHCl3e
methanol as eluting solvents.
5.3. General method for the syntheses of final compounds 16e19
To a stirred suspension of sodium hydride (60% dispersion in
mineral oil, 2 mmol) in anhydrous DMF (5 mL) in an ice bath was
added dropwise a solution of the fluoroindole (1.5 mmol) in
anhydrous DMF (10 mL) over a period of 10 min. 1-Bromooctane
(1.8 mmol) was then added slowly over 5 min and reaction
mixture allowed to warm to room temperature and stirred over-
night. The reaction was quenched by pouring over ice, followed by
extraction (2ꢂ) with ethyl acetate (EtOAc, 25 mL). The combined
organic layers were washed with brine, dried (Na2SO4), filtered and
concentrated under reduced pressure to give the crude residue
which was purified by flash silica gel column chromatography
(EtOAcehexane as eluting solvents). The product (22a, 22b, 22c or
22d) was obtained as an oil (Yield: 80e90%) and used as such for
the next step of the reaction where a stirred solution of 22a, 22b,
22c or 22d (2 mmol) in acetic acid (10 mL) was reacted with
HCHO (0.5 mL) and diethylamine (4 mmol). The reaction mixture
was stirred at room temperature overnight, after which it was
quenched by DCM (20 mL) and water (15 mL) with pH adjusted
to 9 using 1 M NaOH. Extraction with DCM was repeated. The
combined organic layer was washed with brine, dried (Na2SO4)
and concentrated under reduced pressure to obtain the crude
residue. The crude material was purified by column
chromatography (CHCl3eMeOH) to give the product as an oil
(yield 40e60%).
4. Conclusion
Taken together, we have identified a potent Icmt inhibitor 15
which has low micromolar antiproliferative activity on a panel of
human malignant cells. Compound 15 has good solubility, accept-
able PAMPA permeability and limited tendency to form light scat-
tering aggregates. The improvements in the drug-like profile of 15
is attributed to the presence of the polar amino-pyrimidinyl ring at
position 5 of the indoleamine scaffold. There are however unre-
solved concerns over the toxic potential of the indoleamine che-
motype that is present in 15. Nonetheless, the structureeactivity
relationships established in the course of deriving 15 may be use-
fully applied for future design strategies of potent and drug-like
Icmt inhibitors.
5. Experimental
5.1. General experimental conditions for organic synthesis
Reagents (synthetic grade or better) were obtained from com-
mercial suppliers (SigmaeAldrich Chemical Company Inc,
Singapore; Alfa Aesar, MA, USA; BioSynth AG, Switzerland; Com-
5.4. Determination of Icmt inhibition
biBlocks, SD, USA) and used without further purification. 1H and 13
C
NMR spectra were recorded on a Bruker DRX 400 spectrometer.
Chemical shifts were reported in parts per million (ppm, ) and
referenced to residual solvents, deuterated CHCl3 ( 7.26) or CD3OD
c 77.16) and CD3OD
c 49.05) for proton decoupled 13C NMR. Coupling constants (J)
Sf9 (Spodoptera frugiperda ovarian) membranes containing re-
combinant Icmt were provided by Prof PJ Casey (Duke University,
NC and Duke-NUS Graduate Medical School). S-Adenosylmethio-
nine (SAM), magnesium chloride hexahydrate, tartrazine, 4-(2-
hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) and
dithiotreitol (DTT) were purchased from Sigma (Mo, USA). [3H] S-
Adenosylmethionine, [3H] biotin and streptavidin PVT SPA Beads
were purchased from PerkinElmer (Waltham, MA). The synthesis of
biotinylated farnesylcysteine is described in the Supplementary
data. Assays were performed in black 384-well microtitre plates
from PerkinElmer.
d
d
(
(
d
d
3.31) for 1H spectra and deuterated CHCl3
(d
were reported in Hertz (Hz). Proton coupling patterns were
described as singlet (s), doublet (d), double doublet (dd), triplet (t),
quartet (q), multiplet (m) and broad (br). Reactions were routinely
monitored by thin layer chromatography on pre-coated Merck sil-
ica 60 F254 sheets. Column chromatography was carried out on
Merck Silica Gel (0.040e0.063 mm mesh). Nominal mass spectra
were analyzed on Waters LCeMS with electrospray ionization (ESI)
as probe or LcQ Finnigan MAT mass spectrometer with atmospheric
pressure chemical ionization (APCI) as probe. High resolution ac-
curate mass spectra were analyzed on Bruker MicroTOF-QII mass
spectrometer with electrospray ionization (ESI) as probe. Purities of
final compounds were verified by reverse phase HPLC. Microwave
reactions were carried out on the BiotageÒ Initiator Microwave
Synthesizer.
A vial of Icmt membrane protein (30
m
g/
m
L) stored at ꢃ80 ꢁC was
thawed at 37 ꢁC and placed on ice. The protein was diluted in assay
buffer (70 mM HEPES, 100 mM NaCl, 5 mM MgCl2, 3 mM DTT, pH 7.5)
to 0.03
lowed by the addition of inhibitor solution (1ml) or buffer (1
then 20 L of the assay buffer. The reaction was initiated by adding
mg/mL and aliquots were added to each well (5
mL/well), fol-
ml), and
m
5 mL of a solution containing biotinylated BFC (6 mM), SAM (3 mM) and
30
m
Ci/mL [3H] SAM. These reagents were stored at ꢃ20 ꢁC and
thawed on ice (SAM, [3H] SAM) or at room temperature (BFC). The
5.2. General method for the syntheses of final compounds 1e15
plate was incubated at 37 ꢁC for 30 min after which the reaction was
quenched by adding 15 mL of a solution comprising SAM (150 mM) and
5-Bromo-1-octyl-1H-indole (20) and (5-bromo-1-octyl-1H-
indol-3-ylmethyl)ethylamine (21) were prepared as previously
described [10]. To a suspension of 21 (0.8 mmol), the substituted
phenyl, pyridine or pyrimidineboronic acid (1.25 mmol) and tet-
rakis(triphenylphosphine) palladium(0) (0.05 mmol) in dry 1,4-
dioxane (3 mL) in a microwave reaction tube (20 mL) was added
K2CO3 solution (2 mL, 1.5 mM) and subjected to microwave heating
streptavidin SPA beads (20 mg/mL). After overnight incubation at
ambient temperature, radioactivity was counted (Microbeta TriluxÔ,
PerkinElmer, Turku, Finland). The degree of inhibition was assessed
from the radioactivity obtained in presence of test compound
compared to that obtained from the control sample with without test
compound. Two independent determinations were made foreachtest
compound. The % Icmt activity was plotted against log concentration

384
P.M. Ramanujulu et al. / European Journal of Medicinal Chemistry 63 (2013) 378e386
on GraphPad Prism (Version 5.0, GraphPad Software, San Diego, CA)
for the determination of IC50 (concentration required to inhibit
enzyme activity by 50% compared to control/untreated cells).
calibration curve. The concentrations of cysmethynil and 2 were
determined by LCeMS because of their low solubilities. LCMS de-
terminations were made on an Agilent 1200 Series HPLC linked to a
AB Sciex Instruments 3200 Q TRAP LC/MS/MS. Separations were
carried out on a Phenomenex Luna column [3u, C18(2), 100 A,
5 ꢂ 4.6 mm]. The internal standard was (2-(1-octyl-5-pyridin-3-
yl)-1H-indol-3-yl)acetamide for cysmethynil. The internal stan-
dard for 2 was cysmethynil. Quantification was based on the ratio of
peak areas of the daughter ion and the mother ion (M þ H),
normalized against the same ratio obtained for the internal stan-
dard. The solubility determinations were carried out in triplicates
using two different stock solutions.
5.5. Determination of cell-based growth inhibitory activity
Human breast cancer MDA-MB-231, prostate cancer PC3, he-
patocellular carcinoma HepG2, pancreatic cancer MIA Paca II and
human lung fibroblast IMR90 cells were purchased from ATCC
(Rockville, MD). All cells were grown in DMEM (Invitrogen, High
Glucose) at 37 ꢁC, 5% CO2. DMEM was supplemented with 10%
fetal bovine serum (Invitrogen, heat treated at 65 ꢁC, 30 min),
50 units/L penicillin-G and 50
mg/mL streptomycin. Cells were
subcultured at 80e90% confluency and used within 7e15 pas-
sages. Cell viability was assessed using CellTitre 96Ò Aqueous One
Solution (Promega, Madison, WI) containing the tetrazolium
salt 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-
(4-sulfophenyl)-2H-tetrazolium (MTS). Seeding densities were
2500 cells/well. Cells were grown on 96-well plates for 24 h before
aliquots of test compounds (stock solution in DMSO, diluted with
media containing 5% FBS) were added to each well and the plates
were incubated for 72 h. Final concentration of DMSO per well was
5.7. Determination of permeability
The Parallel Artificial Membrane Permeability Assay (PAMPA)
was used to determine the effective permeabilities (Pe) of cysme-
thynil, 2 and 15. Determinations were carried out on MultiScreen-IP
PAMPA assay (donor) plates (MAIPNTR10) and MultiScreen
Receiver Plates (MATRNPS50) from Millipore Corporation (USA)
with 1% lecithin (
of egg yolk, Sigma Aldrich, USA) in dodecane (ReagentPlusÒ, Sigma
Aldrich, USA) as lipid barrier. 5 L of 1% lecithin in dodecane was
dispensed into the wells of the donor plates. Aliquots (300 L) of
test compound (5 M cysmethynil and 2; 120 M 15, prepared in
L-a-phosphatidylcholine from lyophilized powder
kept at 0.5% v/v. At the end of the incubation period, 20 mL of the
m
MTS solution was added to each well, the plates were incubated
for 3 h in the dark after which absorbance readings were taken at
490 nm on a Tecan Infinite M200 Microplate reader. Cell viability
was determined from the expression:
m
m
m
0.1ꢂ PBS with 1% DMSO) were dispensed into the donor wells and
equal volumes of the buffer solution (0.1ꢂ PBS with 1% DMSO) were
added to the corresponding acceptor wells. The donor and acceptor
plates were assembled and the unit was gently agitated on a mini
shaker at room temperature (25 ꢁC) for 16 h. After this time, ali-
h
i
Absorbancecellsþcpd ꢃ Absorbancecpd
Cell Viability ¼
ꢂ 100%
½Absorbancecellsþvc ꢃ Absorbancevc
ꢄ
quots (250 mL/well from donor and acceptor plates) were trans-
ferred to wells in a UV compatible plate (Costar-3635, Corning) for
quantification at lmax of 262 nm for 15 on a microplate reader
(Tecan InfiniteÔ M200).
where Absorbancecellsþcpd ¼ absorbance of wells containing cells in
vehicle (media þ 0.5% DMSO) in the presence of test compound;
Absorbancecellsþvc ¼ absorbance of wells containing untreated cells
and vehicle (vc); Absorbancevc ¼ absorbance of wells containing
only vehicle (vc); Absorbancecpd ¼ absorbance of wells containing
test compound (cpd).
The % viability readings were plotted against log concentration on
GraphPad Prism (Version 5.0, GraphPad Software, San Diego, CA) to
give a sigmoidal curve from which IC50 (concentration required to
reduce viability by 50% compared to control/untreated cells) was
obtained. The plot was constrained to ꢅ0 and ꢆ100%. At least 2
independent determinations of IC50 were made for each compound.
For cysmethynil and 2, aliquots (250
and acceptor plates were transferred to separate glass inserts in
HPLC vials. 20 L of the internal standard (2-(1-octyl-5-pyridin-3-
yl)-1H-indol-3-yl)acetamide, 20.25 M in acetonitrile was added
mL/well) from the donor
m
m
to the sample which was then measured by LCMS (Shimadzu LC 20
series HPLC and AB Sciex Instruments 3200 Q TRAP LC/MS/MS).
Measurements were based on the ratio of peak areas of the
daughter ion and mother ion (M þ H), normalized against the same
ratio obtained for the internal standard. The mobile phase was
Milli-Q water (0.1% formic acid) and acetonitrile (0.1% formic acid),
run on gradient. The column was Phenomenex Luna column [3u,
C18(2), 100 A, 5 ꢂ 4.6 mm] and injections were made at a volume of
5.6. Determination of aqueous solubility
2 mL with flow rate of 0.6 mL/min. Calibration plots of test com-
Determination of aqueous solubility was carried out on Multi-
screenÒ Solubility filter plates (Millipore-MSSLBPC10) from Milli-
pore Corporation (MA, USA). The protocol (PC2445EN00, Millipore
Corporation) was followed. Briefly, various concentrations of the
test compound were prepared in Universal buffer (pH 7.4)/aceto-
nitrile/DMSO. The UV absorbance of these solutions were obtained
at a pre-determined wavelength and used to construct a calibration
curve for the test compound. Next, a stock solution of the test
compound in DMSO was prepared at a known concentration,
diluted with Universal buffer (pH 7.4), dispensed into wells in the
Multiscreen Solubility filter plate, and agitated for 24 h at room
temperature (25 ꢁC). Final concentration of DMSO per well was 1%
v/v except for 10 and 13 where it was 1.5% v/v. The suspension was
filtered, the filtrate diluted with acetonitrile to give the same sol-
vent composition used to prepare the calibration solutions. The
absorbance of the diluted filtrate was read at the predetermined
wavelength and the concentration of the filtrate (equivalent to
the solubility of the test compound) was determined from the
pounds were obtained under similar analytical conditions.
Pe of three standards (caffeine, quinidine and verapamil) were
determined under similar conditions. 500 mM stock solutions were
prepared and dispensed to the donor wells as described earlier.
Quantification was by UV at lmax of 272 nm (caffeine), 280 nm
(verapamil) and 330 nm (quinidine). Calibration plots of reference
compounds were determined under similar analytical conditions.
The Pe of these compounds were in the sequence verapamil
[24,25] > quinidine [26] > caffeine [25], as reported by others.
Pe was obtained from Equation (1):
Pe ¼ ꢃ 2:303 ꢂ fVAVD=½ðVA þ VDÞ ꢂ A ꢂ tꢄg
n
o
(1)
ꢂ log 1 ꢃ ðVA þ VDÞ=ðVD,SÞ ꢂ CAðtÞ=CDð0Þ
where VA and VD are the volumes of acceptor (cm3) and donor (cm3)
wells respectively, A is the area of the surface area of the membrane
(0.24 cm2), t is the permeation time (s); S is the fraction of sample

P.M. Ramanujulu et al. / European Journal of Medicinal Chemistry 63 (2013) 378e386
385
remaining in the donor and acceptor wells after permeation time
and is determined from Equation (2). CA and CD in Equation (2) refer
triple quadrupole linear ion trap mass spectrometer (Applied Bio-
systems/MDS Sciex, Concord, Ontario, Canada) for the other com-
pounds. Separations were made on a C18 column Luna 3u C18(2)
to the concentrations (
wells respectively.
mM) of compound in acceptor and donor
100 A column (2.0 ꢂ 50 mm, i.d., 3
m
M, Phenomenex, Aschaffenburg,
Germany or Eclipse Plus C18 column, 4.6 ꢂ 150 mm, i.d., 3.5
mM,
h
ꢀ
ꢁi
ꢀ
ꢁ
Agilent Technologies, Palo Alto, CA, USA) with a security guard car-
tridge (3.0 ꢂ 4 mm, Agilent Technologies, Palo Alto, CA, USA). Mobile
phase was 0.1% formic acid in acetonitrile e water as mobile phase.
Flow rate was set at 0.2 or 0.6 mL/min and the column temperature
S ¼ ðVA=VDÞ ꢂ
C
AðtÞ=CDð0Þ
þ
C
DðtÞ=CAð0Þ
(2)
The Pe of each compound was obtained from at least 3 separate
experiments using 2 different stock solutions. For each indepen-
dent determination, triplicates (3 wells) were run for each
compound.
was 30ꢁ or 40 ꢁC. 2 or 5
mL full loop sample injection was used.
Data processing was performed with AnalystTM 1.4.2 software
package (Applied Biosystems, MA, USA). The corresponding mul-
tiple reaction monitoring (MRM) transition of the test compound
was selected and used for peak configuration in Analyst 1.4.2 for
semi-quantification. The peak areas of test compound at different
time points were expressed as a % of the peak area of test com-
pound at time ¼ 0 min (¼100%). The resulting % test compound
(average of 3 measurements with SD) was plotted against incuba-
tion time drug (Fig. 1). In vitro half life (T1/2 min) was calculated
from Equation (3) where k is the slope of the line obtained from a
plot of loge (% Test compound) versus time.
5.8. Assessment of aggregation tendency by dynamic light
scattering (DLS)
Stock solutions (10 mM) of test compounds prepared in DMSO,
diluted to 1 mM with DMSO and then serially diluted with potassium
phosphate buffer (5 mM, pH 7.4, prefiltered before use) to give final
concentrations of 1 mM and 10 mM. Final concentration of DMSO was
1% v/v. Measurements were carried out on the Malvern Instrument
Zetasizer Nano ZS system equipped with a 4 mW HeeNe laser at
633 nm and detector angle of 90ꢁ. Three or more determinations of
derived count rates (kilocounts per second, kcps) were obtained
from each concentration of test compound, using two separately
prepared stock solutions. Data collection was carried out using the
software supplied with the instrument. Results are represented as
mean ꢀ standard deviation. The positive control was benzyl ben-
T1=2 ¼ 0:693=k ðminÞ
(3)
where k is the slope of the plot.
Estimated in vitro clearance was determined from Equation (4):
CLint; in vitro ¼ V ꢂ 0:693=T1=2
where V (
(4)
zoate (250
m
M) which gave a count rate of 1180 kcps (ꢀ35). The
vehicle (phosphate buffer, 1% DMSO) gave a reading of 14.9 ꢀ 0.4.
m
L/mg) ¼ Volume of incubation mixture/amount of
microsomal protein in the incubation mixture.
5.9. Determination of in vitro stability of cysmethynil, 2 and 15 in
the presence of rat liver microsomes
5.10. Estimation of log D (pH 7.4) and solubilities
The test compound was incubated with pooled male rat liver
microsomes (BD Gentest Corp, Woburn, MA) in a mixture (total
volume 500 uL) comprising the following: rat liver microsomes
These were estimated using ACD Labs Release 12 (Toronto,
Canada).
(0.3 mg microsome protein/mL), test compound (6
mM, except
cysmethynil which was evaluated at 3 M) and phosphate buffer
m
Acknowledgments
(0.1 M, pH 7.4, containing 1 mM EDTA). The mixture was pre-
incubated for 5 min at 37 ꢁC in a shaking water bath, after which the
The work is supported by a grant from the Biomedical Research
Council (Grant number: 10/1/21/19/664) to MW and MLG. Support
from the Drug Development Unit for Physiochemical Character-
ization is gratefully acknowledged. The authors thank Dr Holger
reaction was started by adding 50
prepared in phosphate buffer) to give a final concentration of 1 mM
NADPH in the mixture. Aliquots of 50 L were withdrawn imme-
mL of 10 mM NADPH (freshly
m
diately on addition of NADPH (time 0) and then at 5, 15, 30 and
45 min. On removal of the sample, reaction was quenched by
Fischer for determining the free energies of amphiphilicity (DDGAM
of cysmethynil, 2 and 15.
)
addition of chilled methanol (100
mL) which also contained the
internal standard (N-ethyl-N-{[5-(4-methylsulfonylphenyl)-1-
octyl-1H-indol-3-yl]methyl} ethanamine, compound 7, Table 1) at
Appendix A. Supplementary data
2
m
M. The mixture was then centrifuged at 10,000ꢂ g to remove the
Supplementary data related to this article can be found at
http://dx.doi.org/10.1016/j.ejmech.2013.02.007.
protein and the content of the test compound in the supernatant
was measured by LCMS.
For each test compound, the metabolic stability of a positive
control, midazolam (Yichang Humanwell Pharmaceutical Co. Ltd,
Yichang, Hubei) which is a known cytochrome P450 substrate, was
concurrently determined to evaluate the adequacy of the experi-
mental conditions. The internal standard used for the LCMS quan-
tification of midazolam was also compound 7 at 2 mM. The stability
of the test compound to microsomal degradation in the absence of
NADPH was also monitored.
Analysis was carried out by LCeMSeMS on a 1200 HPLC instru-
ment (Agilent Technologies, Palo Alto, CA, USA) coupled to a Q
TrapTM 3200 hybrid triple quadrupole linear ion trap mass spec-
trometer (Applied Biosystems/MDS Sciex, Concord, Ontario, Canada)
for cysmethynil, or a Shimadzu UFLC system (Shimadzu Scientific
Instruments, Columbia, MD) coupled to a Q TrapTM 3200 hybrid
References
[1] P.J. Casey, M.C. Seabra, Protein prenyltransferases, J. Biol. Chem. 271 (1996)
5289e5292.
[2] F.L. Zhang, P.J. Casey, Protein prenylation: molecular mechanisms and func-
tional consequences, Annu. Rev. Biochem. 65 (1996) 241e269.
[3] M.N. Ashby, CaaX converting enzymes, Curr. Opin. Lipidol. 9 (1998) 99e102.
[4] Q. Dai, E. Choy, V. Chiu, J. Romano, S.R. Slivka, S.A. Steitz, S. Michaelis,
M.R. Philips, Mammalian prenylcysteine carboxyl methyltransferase is in the
endoplasmic reticulum, J. Biol. Chem. 273 (1998) 15030e15034.
[5] A.M. Winter-Vann, R.A. Baron, W. Wong, J. dela Cruz, J.D. York, D.M. Gooden,
M.O. Bergo, S.G. Young, E.J. Toone, P.J. Casey, A small-molecule inhibitor of
isoprenylcysteine carboxyl methyltransferase with antitumor activity in
cancer cells, Proc. Natl. Acad. Sci. U S A 102 (2005) 4336e4341.
[6] F. Tamonoi, C.L. Gau, C. Jiang, H. Edamatus, K. Kato-Stankiewicz, Protein
farnesylation in mammalian cells: effects of farnesyltransferase inhibitors on
cancer cells, Cell. Mol. Life Sci. 58 (2001) 1636e1649.

386
P.M. Ramanujulu et al. / European Journal of Medicinal Chemistry 63 (2013) 378e386
[7] E. Kim, P. Ambroziak, J.C. Otto, B. Taylor, M. Ashby, K. Shannon, P.J. Casey,
[14] M. Wang, M.S. Hossain, W. Tan, B. Coolman, J. Zhou, S. Liu, P.J. Casey, Inhibition
of isoprenylcysteine carboxymethyltransferase induces autophagic-
dependent apoptosis and impairs tumor growth, Oncogene 29 (2010)
4959e4970.
S.G. Young, Disruption of the mouse Rce1 gene results in defective Ras pro-
cessing and mislocalization of Ras within cells, J. Biol. Chem. 274 (1999)
8383e8390.
[8] M.O. Bergo, B.J. Gavino, C. Hong, A.P. Beigneux, M. McMahon, P.J. Casey,
S.G. Young, Inactivation of Icmt inhibits transformation of oncogenic K-Ras
and B-Raf, J. Clin. Invest. 113 (2004) 539e550.
[15] C.A. Lipinski, Compound properties and drug quality, in: C.G. Wermuth (Ed.),
The Practice of Medicinal Chemistry, third ed., Academic Press, New York,
2008, pp. 481e490.
[9] A.W. Svensson, P.J. Casey, S.G. Young, M.O. Bergo, Genetic and pharmacologic
analyses of the role of Icmt in Ras membrane association and function,
Methods Enzymol. 407 (2006) 144e159.
[16] M. Ishikawa, Y. Hashimoto, Improvement in aqueous solubility in small
molecule drug discovery programs by disruption of molecular planarity and
symmetry, J. Med. Chem. 54 (2011) 1539e1554.
[10] M.L. Go, J.L. Leow, S.K. Gorla, A.P. Schuller, M. Wang, P.J. Casey, Amino de-
rivatives of indole as potent inhibitors of isoprenylcysteine carboxyl meth-
yltransferase, J. Med. Chem. 53 (2010) 6838e6850.
[11] (a) J.A. Bergman, K. Hahne, J. Song, C.A. Hrycyna, R.A. Gibbs, S-Farnesyl-thio-
propionic acid triazoles as potent inhibitors of isoprenylcysteine carboxyl
methyltransferase, ACS Med. Chem. Lett. 3 (2012) 15e19;
[17] E.H. Kerns, L. Di, S. Petusky, M. Farris, R. Ley, P. Jupp, Combined application of
parallel artificial membrane permeability assay and Caco-2 permeability as-
says in drug discovery, J. Pharm. Sci. 93 (2004) 1440e1453.
[18] S.L. McGovern, B.T. Helfand, B. Feng, B.K. Shoichet, A specific mechanism of
nonspecific inhibition, J. Med. Chem. 46 (2003) 4265e4272.
[19] J. Seidler, S.L. McGovern, T.M. Doman, B.K. Shoichet, Identification and pre-
diction of promiscuous aggregating inhibitors among known drugs, J. Med.
Chem. 46 (2003) 4477e4486.
[20] A.M. Gianetti, B.K. Koch, M.F. Browner, Surface plasmon resonance based
assay for the detection and characterization of promiscuous inhibitors, J. Med.
Chem. 51 (2008) 574e580.
[21] H. van de Waterbeemd, D.A. Smith, K. Beaumont, D.K. Walker, Property-based
design: optimization of drug absorption and pharmacokinetics, J. Med. Chem.
46 (2003) 4477e4486.
[22] C. Jamieson, E.M. Moir, Z. Rankovic, G. Wishart, Medicinal chemistry of hERG
optimization: highlights and hang-ups, J. Med. Chem. 49 (2006) 5029e5046.
[23] F. Holger, E.A. Atzpodien, M. Csato, L. Doessegger, B. Lenz, G. Schmitt, T. Singer,
In silico assay for assessing phospholipidosis potential of small druglike
molecules: training, validation and refinement using several data sets, J. Med.
Chem. 52( (2012) 126e139.
(b) J.D. Majmudar, K. Hahne, C.A. Hrycyna, R.A. Gibbs, Probing the iso-
prenylcysteine carboxyl methyltransferase (Icmt) binding pocket: sulfon-
amide modified farnesyl cysteine (SMFC) analogs as Icmt inhibitors, Bioorg.
Med. Chem. Lett. 21 (2011) 2616e2620;
(c) J.D. Majmudar, H.B. Hodges-Loaiza, K. Hahne, J.L. Donelson, J. Song,
L. Shrestha, M.L. Harrison, C.A. Hrycyna, R.A. Gibbs, Amide-modified pre-
nylcysteine based Icmt inhibitors: structureeactivity relationships, kinetic
analysis and cellular characterization, Bioorg. Med. Chem. 20 (2012) 283e295;
(d) J.A. Bergman, K. Hahne, C.A. Hrycyna, R.A. Gibbs, Lipid and sulfur substituted
prenylcysteine analogs as human Icmt inhibitors, Bioorg. Med. Chem. 21 (2011)
283e295;
(e) J.D. Majmudar, A. Morrison-Logue, J. Song, C.A. Hrycyna, R.A. Gibbs, Identi-
fication of a novel nanomolar inhibitor of hIcmt via a carboxylate replacement
approach, MedChemComm 3 (2012) 1125e1137.
[12] W.R. Judd, P.M. Slattum, K.C. Hoang, L. Bhoite, L. Valppu, G. Alberts, B. Brown,
B. Roth, K. Ostanin, L. Huang, D. Wettstein, B. Richards, J.A. Willardsen, Dis-
covery and SAR of emthylated tetrahydropyranyl derivatives as inhibitors of
isoprenylcysteine carboxyl methyltransferase (Icmt), J. Med. Chem. 54 (2011)
5031e5047.
[24] E.H. Kerns, L. Di, Drug-like Properties: Concepts, Structure, Design and
Methods, Academic Press, New York, 2008, pp. 287e298.
[25] K. Sugano, H. Hamada, M. Machida, H. Ushio, High throughput prediction of
oral absorption: improvement of the composition of the lipid solution used in
PAMPA, J. Biomol. Screen. 6 (2001) 189e196.
[13] M. Wang, W. Tan, J. Zhou, J. Leow, M. Go, H.S. Lee, P.J. Casey, A small molecule
inhibitor of isoprenylcysteine carboxymethyltransferase induces autophagic
cell death in PC3 prostate cancer cells, J. Biol. Chem. 283 (2008) 18678e18684.
[26] L. Di, E.H. Kerns, K. Fan, O.J. McConnell, G.T. Carter, High throughput artificial
membrane permeability assay for bloodebrain barrier, Eur. J. Med. Chem. 38
(2003) 223e232.