Amide-modified prenylcysteine based Icmt inhibitors: Structure-activity relationships, kinetic analysis and cellular characterization

Article abstract of DOI:10.1016/j.bmc.2011.10.087

Human protein isoprenylcysteine carboxyl methyltransferase (hIcmt) is the enzyme responsible for the α-carboxyl methylation of the C-terminal isoprenylated cysteine of CaaX proteins, including Ras proteins. This specific posttranslational methylation event has been shown to be important for cellular transformation by oncogenic Ras isoforms. This finding led to interest in hIcmt inhibitors as potential anti-cancer agents. Previous analog studies based on N-acetyl-S-farnesylcysteine identified two prenylcysteine-based low micromolar inhibitors (1a and 1b) of hIcmt, each bearing a phenoxyphenyl amide modification. In this study, a focused library of analogs of 1a and 1b was synthesized and screened versus hIcmt, delineating structural features important for inhibition. Kinetic characterization of the most potent analogs 1a and 1b established that both inhibitors exhibited mixed-mode inhibition and that the competitive component predominated. Using the Cheng-Prusoff method, the K _i values were determined from the IC₅₀ values. Analog 1a has a K_IC of 1.4 ± 0.2 μM and a K_IU of 4.8 ± 0.5 μM while 1b has a K_IC of 0.5 ± 0.07 μM and a K_IU of 1.9 ± 0.2 μM. Cellular evaluation of 1b revealed that it alters the subcellular localization of GFP-KRas, and also inhibits both Ras activation and Erk phosphorylation in Jurkat cells.

Full text of DOI:10.1016/j.bmc.2011.10.087

Bioorganic & Medicinal Chemistry 20 (2012) 283–295
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Amide-modified prenylcysteine based Icmt inhibitors: Structure–activity
relationships, kinetic analysis and cellular characterization
Jaimeen D. Majmudar a,b, , Heather B. Hodges-Loaiza b,c, , Kalub Hahne b,c, James L. Donelson a,c
,
a,c,
Jiao Song a,c, Liza Shrestha a,c, Marietta L. Harrison a,c, Christine A. Hrycyna b,c, , Richard A. Gibbs
⇑
⇑
a Department of Medicinal Chemistry and Molecular Pharmacology, Purdue University, West Lafayette, IN 47907, United States
b Department of Chemistry, Purdue University, West Lafayette, IN 47907, United States
c The Purdue University Center for Cancer Research, Purdue University, West Lafayette, IN 47907, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Human protein isoprenylcysteine carboxyl methyltransferase (hIcmt) is the enzyme responsible for the
-carboxyl methylation of the C-terminal isoprenylated cysteine of CaaX proteins, including Ras
Received 18 August 2011
Revised 27 October 2011
Accepted 30 October 2011
Available online 6 November 2011
a
proteins. This specific posttranslational methylation event has been shown to be important for cellular
transformation by oncogenic Ras isoforms. This finding led to interest in hIcmt inhibitors as potential
anti-cancer agents. Previous analog studies based on N-acetyl-S-farnesylcysteine identified two prenyl-
cysteine-based low micromolar inhibitors (1a and 1b) of hIcmt, each bearing a phenoxyphenyl amide
modification. In this study, a focused library of analogs of 1a and 1b was synthesized and screened versus
hIcmt, delineating structural features important for inhibition. Kinetic characterization of the most
potent analogs 1a and 1b established that both inhibitors exhibited mixed-mode inhibition and that
the competitive component predominated. Using the Cheng–Prusoff method, the Ki values were deter-
Keywords:
Isoprenylcysteine carboxyl methyltransferase
Methylesterification
Icmt
Enzyme inhibition
Ras proteins
mined from the IC50 values. Analog 1a has a KIC of 1.4 0.2
lM and a KIU of 4.8 0.5 lM while 1b has
Anti-cancer agents
Prenylation
Methyl transferase
a KIC of 0.5 0.07 M and a KIU of 1.9 0.2 M. Cellular evaluation of 1b revealed that it alters the sub-
cellular localization of GFP-KRas, and also inhibits both Ras activation and Erk phosphorylation in Jurkat
cells.
l
l
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
followed by proteolytic removal of the –aaX residues by the endo-
protease Ras converting enzyme-1 (Rce1).10 Icmt then carries out
A target of growing interest in the field of cancer chemothera-
peutics is the human protein isoprenylcysteine carboxyl methyl-
transferase (hIcmt).1–4 Icmt is the enzyme responsible for the last
common step of the posttranslational processing pathway of CaaX
proteins.5,6 The family of CaaX proteins includes many key eukary-
otic regulatory proteins, most notably proteins in the Ras family.
These proteins terminate with a C-terminal CaaX sequence that
is composed of four amino acids in the order of cysteine, followed
by two generally aliphatic residues (aa), and a final residue (X) that
varies in its identity from protein to protein. The first step of the
CaaX protein posttranslational processing pathway is mediated
by either protein farnesyl transferase (FTase) or protein geranyl-
geranyl transferase type I (GGTase-I).7–9 FTase and GGTase-I
respectively attach either a 15-carbon or 20-carbon isoprenoid
group to the CaaX cysteine. The type of lipid group added is direc-
ted, in part, by the nature of the terminal X residue. Lipidation is
the final carboxyl methylation step. Disruption of this pathway
via enzyme inhibition prevents Ras proteins from localizing
properly in the cell, thereby hindering cellular functionality and
obstructing cellular growth.2,11,12 Therefore, the development of
inhibitors of the enzymes in this pathway is desirable for the treat-
ment of cancer driven by oncogenic Ras.
Mutant forms of K-Ras are a frequent cause of several human
malignancies, including pancreatic cancers. Research on farnesyl
transferase inhibitors (FTIs), once promising cancer chemothera-
peutic agents, showed that K-Ras tumors are often resistant to
FTI treatment due to alternative geranylgeranylation.13–16 The
reason for this difference lies in the fact that K-Ras can be alterna-
tively geranylgeranylated by GGTase I in the presence of FTIs.17–19
Although, Icmt recognizes both farnesylated and geranylgeranylat-
ed proteins,20,21 previous studies suggest that the methylation is
more important for the proper localization of farnesylated CaaX
proteins than for many geranylgeranylated proteins.20,22
These data further support that Icmt may prove to be an impor-
tant target in preventing the function of farnesylated oncogenic
Ras proteins, but may not be detrimental to the function of all CaaX
proteins. Furthermore, recent studies in cells and mouse models
⇑
Corresponding authors.
E-mail addresses: hrycyna@purdue.edu (C.A. Hrycyna), rgibbs@purdue.edu
(R.A. Gibbs).
These authors contributed equally to this work.
0968-0896/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.10.087

284
J. D. Majmudar et al. / Bioorg. Med. Chem. 20 (2012) 283–295
Figure 1. Phenoxyphenyl-farnesylcysteine (a, POP-FC), phenoxy-phenyl-3-methylbutenyl-biphenyl farnesylcysteine (b, POP-3 MB-FC) and anthranilic farnesylcysteine (c).
also support Icmt as a promising chemotherapeutic target.23,24
Wang et al. have shown that small molecule Icmt inhibitors can in-
duce autophagic cell death in PC3 prostate cancer cells.23 In mice,
inactivation of Icmt has also been shown to inhibit transformation
by oncogenic K-Ras and B-Raf.2 These key findings further support
the development of hIcmt inhibitors as chemotherapeutic agents
for Ras-driven cancers.
phenoxy-phenyl motif plays in human Icmt (hIcmt) inhibition, a
library of compounds was designed and synthesized based on the
lead POP-FC (1a). Results from this study will help us better under-
stand the role of the phenoxy-phenyl motif in enzyme recognition
with the goal of enhancing the potency of these small molecules.
2. Chemistry
Our laboratories have recently identified 2-phenoxy-phenyl-
farnesylcysteine (POP-FC, Fig. 1a) and 2-phenoxyphenyl-3-meth-
ylbutenyl-farnesylcysteine (POP-3-MB-FC, Fig. 1b) as lead Icmt
inhibitors.25 These findings prompted us to examine the role of the
phenoxy-phenyl motif in achieving Icmt inhibition. The simplest
and most widely used amino acid-based Icmt substrate is an
amide-modified farnesylated cysteine, N-acetyl-S-farnesylcysteine
(AFC).26–32 The important substrate recognition features of AFC,
which mimics the–aaX proteolyzed C-terminus of protein sub-
The analogs described herein were synthesized using previously
established methods in our laboratory.25,34 The general synthetic
design involved utilizing a known synthesis of farnesylcysteine
methyl ester34 and coupling it with the appropriate carboxylic acid
using HOBT and HBTU followed by saponification using lithium
hydroxide (as shown schematically in Scheme 2).34 In cases where
the carboxylic acid was not commercially available, the appropri-
ate aromatic carboxylic acid was synthesized and coupled to farne-
sylcysteine methylester.
The structures of the compounds synthesized utilizing this ap-
proach are shown in Figure 2. Carboxylic acids needed for the syn-
thesis for compounds 2a–c and 2e were commercially available.
For compounds 2d and 2f–h, each carboxylic acid was synthesized
as shown in Scheme 1. The general method involved using an alk-
oxide anion as the nucleophile to displace a benzylic halide to ob-
tain the respective phenoxy-phenyl analog. The aromatic bromide
was then converted into a Grignard reagent using magnesium
turnings in THF and quenching the formed Grignard with dry ice.
Carboxylic acid 5 used to prepare analog 2d was synthesized using
an Ullmann-type ether coupling as reported previously.35
strates, are the presence of an S-farnesyl moiety, a free
a-carboxyl
group and an amide modified
a
-amino group.26,28 Note that the pep-
tide Icmt substrate LARYK-S-farnesyl-C was found to have a 10-fold
lower KM for mammalian Icmt than AFC,32 suggesting that the
upstream sequence plays a role in recognition and methylation
efficiency.
Previous studies from our laboratory have indicated that the
human homolog of Icmt may possess an active site that can accom-
modate substrates or inhibitors with bulky, hydrophobic amide
moieties and that certain bulky amide-modified farnesylcysteine
(AMFC) analogs can act as low micromolar inhibitors of human
Icmt.25,33,34In order to further our understanding of the role the
Figure 2. Structures of the phenoxy-phenyl modified analogs synthesized and screened as inhibitors of hIcmt.

J. D. Majmudar et al. / Bioorg. Med. Chem. 20 (2012) 283–295
285
Scheme 1. Synthesis of various carboxylic acids. Reagents and conditions: (a) Synthesis of 2-(phenylthio)benzoic acid: (a) Potassium hydroxide, 5 equiv; Cu powder,
0.1 equiv; water (86%); (b) Synthesis of 2-(phenoxymethyl)benzoic acid: (a) NCS, dimethylsulfide, DCM, 73%, (b) NaH, DMF, 0 °C to rt, 4 h, 62%, (c) Mg turnings, 1,2
dibromoethane, CO2 quench, then 10% HCl, 68%.; (c) Synthesis of 2-(benzyloxy)benzoic acid: (a) NaH, DMF, 0 °C to rt, 4 h, 67% (b) Mg turnings, 1,2 dibromoethane, CO2
quench, then 10% HCl 78%.; (d) Synthesis of 2-((benzyloxy)methyl)benzoic acid: (a) NaH, DMF, 0 °C to rt, 4 h, 77% (b) Mg turnings, 1,2 dibromoethane, CO2 quench, then 10%
HCl, 45%.
Scheme 2. General synthetic scheme showing synthesis of analogs 2a–2h. (a) 7N NH3 in methanol, 0 °C, 3 h, 80%; (b) HOBT, HBTU, DIEA, DMF, 0 °C to rt, 6–10 h, 48–93% (c)
LiOH in methanol, 2 h, 60–85%.
We also designed and synthesized molecules in which the
phenoxyphenyl amide scaffold was maintained but the prenyl
chain was modified. To this end, we synthesized analogs 17–21
shown in Figure 3. Their synthesis is depicted in Schemes 3 and
4. Analog 17 is the enantiomer of our lead 1a, while analogs 18
and 19 contain a shorter geranyl (10 carbon) and a longer geranyl-
geranyl (20 carbon) chain, respectively. Analog 20 (synthesis
shown in Scheme 4) was specifically designed to investigate the
role of the amide proton in hIcmt inhibition and analog 21, with
an undecyl moiety was designed to probe the importance of the
prenyl chain in achieving hIcmt inhibition. We have recently re-
ported that a similar change from prenyl to straight chain alkyl is
detrimental to hIcmt inhibition when the amide linkage was re-
placed with a sulfonamide bond.36 We wanted to investigate
whether such a change is consistent across different classes of
inhibitors. The N-methyl analog 20 was synthesized using a so-
lid-phase methodology using the advanced intermediate 23, which
was synthesized as reported previously.37 Compound 23 was
loaded on the 2-chlorotrityl-chloride resin, which is essential to
this synthesis due to the acid sensitivity of the prenyl chain. After

286
J. D. Majmudar et al. / Bioorg. Med. Chem. 20 (2012) 283–295
Figure 3. Structures of the phenoxy-phenyl prenyl modified analogs synthesized and screened as inhibitors of hIcmt.
Scheme 3. Synthesis of prenyl-modified phenoxy-phenyl analogs. Reagents and conditions: (a) 7 N NH3 in Methanol, 0 °C, 3–12 h, 66–80% (b) HOBT, HBTU, DIEA, DMF, 0 °C to
rt, 6–10 h, 48–93% (c) LiOH in methanol, 2 h, 60–85%.
loading, the disulfide-protecting group was removed using DTT
and the farnesyl group was attached to the free thiol using farnesyl
chloride. The 2-phenoxyphenyl carboxylic acid was then coupled
to the secondary amine using standard coupling procedures previ-
ously used in our laboratory.25 Compound 20 was obtained upon
cleavage from the resin using 0.2% TFA in DCM. Following their
syntheses, the compounds were evaluated as substrates and inhib-
itors of hIcmt using a vapor diffusion assay.5,38,39
We have previously reported the synthesis of the anthranilic
acid analog (Fig. 1c), where the phenoxy-phenyl oxygen is replaced
by a nitrogen atom.25 This compound has an IC50 of 7.1
lM against
hIcmt. Although less potent than the parent compound 1a, the less
electronegative nitrogen does not result in a significant loss of
activity. We next wanted to evaluate the effect of changing the size
and electronegativity of the central atom connecting the ortho-
substituted phenyl ring. To this end, we synthesized analogs 2c
and 2d. The decreased electronegativity of the carbon and sulfur
atom linkers resulted in a roughly fourfold loss of inhibitory activ-
ity as compared to compound 1a, suggesting that the electronega-
tivity of the connecting atom between the two-phenyl rings may
be a determining factor for inhibitory activity. More electronega-
tive atoms such as oxygen and nitrogen present in the structure
resulted in better inhibitors than those containing less electroneg-
ative atoms such as carbon and sulfur.
3. Results and discussion
3.1. In vitro biochemical evaluation of analogs
Analogs 2a–h and 17–21 were first tested as substrates for
hIcmt at 25 lM as described in the experimental section. Upon
evaluation, none of the analogs showed substrate activity (data
The biochemical evaluation described above suggested that the
ortho substituted phenoxy-phenyl motif was important for hIcmt
inhibition. To further investigate the effect of the spatial orienta-
tion of the oxygen connector atom and the second phenyl ring
on hIcmt inhibition, we synthesized and evaluated analogs
2e–2h. Conformational restriction of the two phenyl rings through
a dibenzofuran scaffold (compound 2e) diminished inhibitory
activity significantly, as did replacing the central one oxygen linker
with a two-atom linker in analogs 2f and 2g. It is worth noting that
analog 2g, which retains the position of the oxygen atom relative
to the amide bond, displays much greater inhibition as compared
to analog 2f. The inhibitory effect of analog 1a could result from
the two oxygen atoms’ (the amide carbonyl oxygen and the linker
oxygen), involvement in a critical interaction because increasing
not shown). The compounds were subsequently tested as inhibi-
tors at 10 lM in the presence of 25 lM of the substrate AFC and
all the analogs were inhibitors of hIcmt to varying degrees. Com-
pounds 2a and 2b were specifically synthesized and tested to eval-
uate the importance of the positioning of the phenoxy-phenyl
motif in hIcmt inhibition. Both of these analogs were poor inhibi-
tors of hIcmt, both inhibiting hIcmt by less than 30% at 10
We experimentally determined the IC50 of compound 2b to be
22.6 1.2 M. This value reflects an approximately five-fold loss
lM.
l
in activity between the ortho and para isomers of the phenoxy-
phenyl motif, suggesting that the ortho-phenoxyphenyl motif is
important for hIcmt inhibition. Note that the para regio-isomer is
a significantly poorer inhibitor compared to both the meta and
the ortho isomer.

J. D. Majmudar et al. / Bioorg. Med. Chem. 20 (2012) 283–295
287
Scheme 4. Synthesis of N-methyl POP-FC. (a) Benzene, (HCHo)n, PTSA, reflux, 10 h; (b) TES,TFA.
the distance between those two oxygen atoms reduced the inhib-
itory potency of the molecule. Analog 2h, where the central oxygen
linker is more flexible compared to 1a is a poor inhibitor compared
to compound 1a.
To evaluate the effect of the stereochemistry of the amino acid
derivative POP (1a) on hIcmt inhibition, we synthesized its enan-
tiomer (17). We have recently demonstrated that the stereochem-
istry at the alpha carbon is not critical for hIcmt inhibition in a
sulfonamide series36 and consistent with this result, analog 17
Overall, our biochemical data suggest that the ortho-phenoxy-
phenyl motif is important and replacement with other regioisom-
ers leads to a significant decrease in potency. The presence and the
position of the oxygen atom connecting the two phenyl rings are
also important features, as any change in the atom itself or its spa-
tial positioning reduces hIcmt inhibition. Stereochemistry about
the alpha carbon of cysteine and the amide proton both appear
inconsequential to hIcmt inhibition. The most important element
in hIcmt inhibition with this class of compounds is the farnesyl
chain itself, as any deviation from the 15-carbon isoprene results
in a loss of inhibition, which is particularly evident with the unde-
cyl analog 21, a very poor hIcmt inhibitor.
inhibited hIcmt with an IC50 of 7.8 0.4
to 1a, which demonstrated an IC50 of 6.2 0.7
l
M, nearly equivalent
M. Next, to inter-
l
rogate the importance of the farnesyl chain in analog 1a, we syn-
thesized analogs 18, 19 and 21. The farnesyl group appears to be
very important to hIcmt inhibition because the short and longer
prenyl analogs 18 and 19 exhibited significant loss of hIcmt inhi-
3.2. Kinetic mode of inhibition for compounds 1a and 1b
bition at 10
l
M. The undecyl analog, 21, was a particularly poor
In order to determine the mode of inhibition of hIcmt by com-
pounds 1a and 1b, the most potent inhibitors in this series of com-
pounds, kinetic assays were performed in which the small
molecule substrate (AFC) concentration was varied while the
inhibitor concentration remained fixed. The data for each inhibitor
concentration were averaged from three trials, and plotted as dou-
ble reciprocal plots ( Fig. 4 and 5). These inhibitors displayed
mixed-type inhibition in which the competitive component domi-
inhibitor of hIcmt at the test concentration. These data suggest
that the presence of a farnesyl chain on the cysteine sulfur is a
critical pharmacophore for hIcmt inhibition. These data corrobo-
rate our previous findings about the importance of the farnesyl
chain for hIcmt inhibition.33,36 Finally, to interrogate the effect
of the amide proton on hIcmt inhibition, we synthesized analog
20 using the solid-phase strategy described above. This analog
was equivalent to our previous lead, 1a, in terms of percent inhi-
nated, as demonstrated by an
a
value greater than 1.52 Ki Values for
bition of hIcmt activity and has a similar IC50 of 6.7 0.4
l
M. The
compounds 1a and 1b were then determined from the IC50 values
using the Cheng–Prusoff method as adapted by Cer et al.53 For
compound 1a, the competitive (KIC) and uncompetitive (KIU) values
presence of a methyl group on the amide nitrogen appears to
have minimal effect on the inhibition potential of the analog, sug-
gesting that the amide region of the molecule may be no more
important than serving as a linker between the phenoxy-phenyl
and the farnesyl moieties (Table 1).
were 1.4 0.2
1b, the KIC and KIU values were 0.5 0.07
1.9 0.2 M, respectively.
lM and 4.8 0.5
l
M, respectively. For compound
lM
and KIU
l

288
J. D. Majmudar et al. / Bioorg. Med. Chem. 20 (2012) 283–295
Table 1
Inhibition profiles of the analogs
a,b
Structure
Analog
IC50
NA
(lM)
AFC (negative control)
1a (POP-FC–pos. control)
6.2 0.7a
1b (POP-3MB-FC)–pos. control
2.5 0.3a
2a
2b
2c
2d
2e
63b
22.6 1.2a
38b
14.7 0.8a
58b
2f
75b
2g
2h
13.4 0.6a
49b
17
18
19
7.8 0.4a
149b
38b

J. D. Majmudar et al. / Bioorg. Med. Chem. 20 (2012) 283–295
289
Table 1 (continued)
a,b
Structure
Analog
IC50
(lM)
20
6.7 0.4a
21
210b
a
IC50 values are mean of three experiments, were determined using the vapor diffusion assay and calculated using GRAPH-PAD PRISM 5.0. Concentration of the substrate
M.
used (AFC) was 25
l
b
Estimated IC50 values obtained using an in-house equation (y = À26.813x + 77.293; where y = percent inhibition at 10
lM and x = log[IC50]) that correlates percent
inhibition value at 10
2 = 0.85.
lM inhibitor concentration (in presence of 25
lM substrate, AFC, concentration) to experimental IC50 values. This logarithmic correlation has a
R
0.08
0.06
0.04
0.02
-0.2
-0.1
0.1
0.2
0.3
0.4
-0.02
Figure 5. Mode of inhibition and Ki determination of compound 1b. The phenoxy-
1/[S] (1/µM)
phenyl analog, 1a, demonstrated mixed-type inhibition with a KIC of 1.2 0.2
and a KIU of 11.2 2.4 M. Lineweaver–Burk plots were obtained by the same
method as that described for Figure 4.
lM
l
Figure 4. Mode of inhibition and Ki determination of compound 1a. 1a Displays
mixed-type inhibition. Lineweaver–Burk plots illustrate the mode (mixed) of
inhibition of hIcmt by 1a. These are double reciprocal plots from AFC substrate
curves with hIcmt, in the presence of the indicated concentration of inhibitor,
which were obtained as described in Section 6.2. The KIC is 2.6 0.5
is 10.5 1.4 M.
lM and the KIU
inhibitor, preventing K-Ras membrane localization. Surprisingly,
the data indicate that significant cellular mislocalization of GFP-
K-Ras is seen at compound levels that are significantly lower than
the biochemical IC50 value for 1b. While we are unclear as to the
reason for this result, there are at least two possibilities. First, this
compound may preferentially accumulate in Jurkat cells. Alterna-
tively, this compound may, in addition to inhibiting Icmt, block
GFP-K-Ras through another mechanism, possibly related to that
proposed for Salirasib by Kloog and co-workers.45–47
l
4. Cellular effects of Icmt inhibitor 1b
4.1. Pop-3MB (1b) alters the subcellular localization of GFP-K-
Ras in Jurkat T cells
Proper posttranslational processing is a prerequisite for Ras
membrane localization and function. To determine the cellular ef-
fects of our most potent inhibitor 1b, the membrane localization of
fluorescently tagged K-Ras was visualized in the presence and ab-
sence of 1b.40,41 We transfected Jurkat T cells with GFP-K-Ras and
treated the cells with delivery vehicle alone, the statin drug simva-
statin (which blocks K-Ras prenylation), or 1b. The cells were fixed
and GFP-K-Ras localization was visualized using fluorescent
microscopy. After 24 h of treatment, GFP-K-ras localization was
categorized into three groups: normal plasma membrane localiza-
tion, partial mislocalization, and complete mislocalization. This
was determined via visual inspection of randomly selected fields
containing 100 cells. Representative examples of normal, partial
and complete mislocalization are shown in Figure 6b (determined
as described in the Section 6). The results demonstrated that
4.2. Pop-3MB (1b) inhibits Erk phosphorylation and Ras
activation in Jurkat T cells
Low-grade stimulation through the T-cell receptor (TCR) in Jur-
kat cells results in Ras dependent Erk activation.42 We therefore
utilized signaling assays with low grade TCR stimulation to evalu-
ate the effects of 1b on Erk and Ras activation. Cells were treated
with delivery vehicle (DMSO), simvastatin, or 1b. After 24-hour
vehicle or drug treatment, Jurkat cells were low-grade stimulated
through the TCR with anti-CD-3e to activate the Ras-MAPK-Erk
pathway. To test Ras activation, the stimulated cells were lysed
and active GTP-bound Ras was precipitated from whole cell lysates
using a GST-RBD (Ras Binding Domain) pulldown assay.43 As
shown in Figure 7b, compound 1b decreased Erk activation
approximately three-fold compared to the control. Furthermore,
1b also inhibited Ras activation significantly as seen in Figure 7c.
These data suggest that the alterations in TCR-dependent Erk
0.1
l
M 1b prevented approximately 35% protein membrane local-
M compound 1b prevented 75%
ization while treatment with 10
l
of membrane localization (Fig. 6a and b). These data suggest that
1b is taken up into mammalian cells and functions as an Icmt

290
J. D. Majmudar et al. / Bioorg. Med. Chem. 20 (2012) 283–295
Figure 6. Compound 1b alters the subcellular localization of GFP K-Ras in Jurkat T cells. Jurkat T cells were transiently transfected with GFP-K-Ras and treated with DMSO or
1b at indicated concentrations for 24 h. (A) Histograms depict the differences in GFP-K-Ras localization after indicated treatments. Differences were defined as partial or
complete loss of normal localization, as defined by the images shown in (B). Results are representative of three independent experiments.
phosphorylation and Ras activation in Jurkat T cells are due to the
inhibition of Ras methylation by compound 1b.
Icmt inhibitors reduced cell viability only modestly in several can-
cer cell lines.48 These reports, coupled with our results published
here suggest that Icmt inhibitors have utility in treating Ras-driven
cancers. Further studies with substrate-based hIcmt inhibitors,
such as those developed in our laboratory, will help to clarify the
therapeutic potential of hIcmt inhibition. Analog 1b is a promising
lead and suggests that appropriate prenyl mimics can be developed
to achieve potent hIcmt inhibition and efforts are currently under-
way to exploit these findings.
5. Conclusion
We have demonstrated that low-micromolar inhibitors of hIcmt
alter the sub-cellular location of GFP-Ras and also have an effect in
the downstream pathway. Inhibition of anti-CD-3e dependent Erk
phosphorylation and Ras activation in Jurkat T cells by compound
1b suggests that hIcmt inhibitors hold promise in treatment of
K-Ras driven cancers. Our SAR data with the entire series of com-
pounds reported here have provided insight into the potential
pharmacophoric requirements for hIcmt inhibition. As there is
yet no three-dimensional structural information for hIcmt, given
that it is an integral membrane protein, these data generated from
our compounds are a small step forward in understanding the inhi-
bition requirements for hIcmt. Stereochemistry at the alpha carbon
has no significant bearing on the outcome for inhibition. Our data
further suggest that the farnesyl chain is the most important
scaffold for substrate-based Icmt inhibitors and any change in
length or overall hydrophobicity decreases inhibition. Cysmethy-
nil, a small molecule inhibitor of Icmt44 has demonstrated induc-
tion of autophagic cell death in PC3 prostate cancer cells,23 and
thus has shown promise for cancer therapy. Salirasib, a partial
inhibitor of Icmt has also shown promise in phase I clinical tri-
als.45–47 In contrast, a recent report demonstrated that very potent
6. Experimental
6.1. Synthesis and spectral analysis
General Procedure for the synthesis of analogs 2a–2h
6.1.1. Synthesis of farnesylcysteine methylester
L-Cysteine methylester hydrochloride (1.0 equiv) (obtained
from Sigma–Aldrich) was charged to a dry round bottom flask
and the atmosphere inside the flask was replaced with argon.
The reaction mixture was maintained at 0 °C and 5 mL of 7 N
ammonia in methanol (purchased from Sigma–Aldrich) was added
to the L-cysteine methylester hydrochloride. The contents were al-
lowed to stir for 15 min following which, trans–trans-farnesylchlo-
ride (0.95 equiv) (purchased from Sigma–Aldrich) was syringed
into the flask. The reactants were allowed to react for 3 h at 0 °C.

J. D. Majmudar et al. / Bioorg. Med. Chem. 20 (2012) 283–295
291
Figure 7. Inhibition of anti-CD-3
simvastatin (36.7
e
dependent Erk phosphorylation and Ras activation in Jurkat T cells following 1b treatment (25
l
M). Jurkat T cells were treated with vehicle,
l
M) or 1b for 24 h followed by stimulation with anti-CD-3e (20 lg/mL) (A) The phosphorylation of Erk was detected by immunoblotting the cell lysate with
an anti-phospho-Erk antibody (top). The active form of Ras was precipitated from the cell lysate with a GST-RBD fusion protein. Precipitated Ras was detected by
immunoblotting using an anti-pan-Ras antibody (bottom). Immunodetection of GAPDH was used as the loading control. Erk phosphorylation (B) and Ras activation (C) were
quantified using ImageJ (NIH). The results are representative of three independent experiments.
The reaction was monitored by TLC analysis. The contents of the
round bottom flask were evaporated to dryness and purified by
column chromatography via isocratic elution using a mixture of
methanol and dichloromethane (1:9). This yielded farnesylcysteine
methylester in 80% yield. The spectral data matched the reported
data.34,49
was performed under the influence of gravity using a 1:9 mixture
of methanol and methylenechloride.
6.1.3.1. Compound 2a: (R)-2-(4-Phenoxybenzamido)-3-(((2E,
6E)-3,7,11-trimethyldodeca-2,6,10-trien-1-yl)thio)propanoic
acid. 1H NMR (300 MHz, CDCl3) d 7.80 (m, 3H), 7.33 (dd, J = 18.3,
10.6 Hz, 2H), 7.06 (m, 4H), 5.41–4.74 (m, 4H), 3.13 (m, 4H), 2.10–
1.87 (m, 8H), 1.67 (d, J = 6.5 Hz, 3H), 1.59 (d, J = 4.5 Hz, 3H), 1.56
(s, 3H). 13C NMR (75 MHz, CDCl3) d 176.81, 135.09, 134.77,
131.17, 131.06, 129.86, 124.31, 124.03, 123.81, 119.74, 117.51,
39.66, 39.56, 26.69, 26.48, 25.70, 25.66, 25.63, 17.65, 17.59,
16.08, 15.95, 15.79.
6.1.2. Synthesis of the amide-modified farnesylcysteine
methylester intermediate
The appropriate carboxylic acid (1.0 equiv) was charged to a dry
round bottom flask and was dissolved in dimethylformamide
(DMF). Diisopropylethylamine (1.1 equiv), HOBT (1.1 equiv) and
HBTU (1.1 equiv) were added to this and the contents were al-
lowed to react at 0 °C for 30 min. This was followed by addition
of a solution of farnesylcysteine methylester in DMF. The contents
were allowed to react for 2 h. The reaction was monitored by TLC
analysis. On completion of the reaction, 10% aqueous citric acid
was added to the reaction mixture and the product was extracted
using ethylacetate (20 mL) as the solvent. The solvent was re-
moved under vacuum utilizing a rotary evaporator and the residue
was loaded on a short (2–3 inches) silica gel plug. The silica gel
plug was flushed with 3% methanol in dichloromethane to obtain
the crude coupling product.
6.1.3.2. Compound 2b: (R)-2-(3-Phenoxybenzamido)-3-(((2E,
6E)-3,7,11-trimethyldodeca-2,6,10-trien-1-yl)thio)propanoic
acid. 1H NMR (300 MHz, CDCl3) d 8.03–7.68 (m, 4H), 7.67–7.46
(m, 3H), 7.37 (t, J = 7.2 Hz, 1H), 7.20 (d, J = 18.4 Hz, 1H), 5.38 (dd,
J = 13.7, 7.3 Hz, 3H), 4.98 (d, J = 29.8 Hz, 1H), 3.61–2.94 (m, 4H),
2.48–2.08 (m, 8H), 1.94 (d, J = 14.1 Hz, 3H), 1.88 (s, 3H), 1.85 (s,
6H). 13C NMR (75 MHz, CDCl3) d 176.38, 167.17, 157.37, 156.40,
139.73, 135.21, 135.05, 131.12, 129.73, 124.26, 124.22, 123.76,
123.57, 121.85, 121.50, 119.32, 119.07, 117.94, 53.49, 39.61,
39.51, 29.98, 26.64, 26.43, 25.62, 25.59, 17.61, 16.01, 15.91.
6.1.3.3. Compound 2c: (R)-2-(2-Benzylbenzamido)-3-(((2E,6E)-
3,7,11-trimethyldodeca-2,6,10-trien-1-yl)thio)propanoic
acid. 1H NMR (300 MHz, MeOH) d 7.32 (d, J = 6.7 Hz, 1H), 7.18–
6.89 (m, 8H), 5.05 (t, J = 7.4 Hz, 1H), 4.92 (t, J = 6.3 Hz, 2H), 4.51
(s, 1H), 4.13–3.93 (m, 2H), 3.11–2.99 (m, 2H), 2.78 (ddd, J = 22.1,
13.6, 5.6 Hz, 2H), 1.98–1.72 (m, 8H), 1.50 (d, J = 4.1 Hz, 6H), 1.42
(s, 6H). 13C NMR (75 MHz, MeOH) d 172.60, 160.82, 142.20,
140.57, 140.30, 137.60, 136.13, 132.03, 131.45, 131.01, 130.28,
129.36, 128.46, 127.10, 127.01, 125.43, 125.11, 121.51, 40.83,
40.71, 27.77, 27.42, 25.98, 17.85, 16.35, 16.21.
6.1.3. Synthesis of the amide-modified farnesylcysteine (free
acid)
The crude product obtained from the previous step was dis-
solved in methanol and was transferred to a round-bottom flask.
The contents were allowed to cool to 0 °C. Lithium hydroxide
(1.1 equiv) was added to the mixture and the contents were al-
lowed to react for 2 h. Following the completion of the reaction,
the solvent was removed under reduced pressure and the mixture
was loaded on a silica gel column. Chromatographic separation

292
J. D. Majmudar et al. / Bioorg. Med. Chem. 20 (2012) 283–295
6.1.3.4. Compound
2d: (R)-2-(2-(Phenylthio)benzamido)-3-
172.50, 145.62, 136.84, 134.13, 133.11, 133.00, 130.81, 130.30,
128.12, 126.19, 125.24.
(((2E,6E)-3,7,11-trimethyldodeca-2,6,10-trien-1-yl)thio)propa-
noic acid. 1H NMR (300 MHz, CDCl3) d 7.61 (d, J = 32.1 Hz, 1H),
7.37 (s, 3H), 7.28 (d, J = 3.6 Hz, 3H), 7.14 (s, 2H), 7.01 (s, 1H),
5.23–4.93 (m, 3H), 4.80 (s, 1H), 3.12 (s, 2H), 2.93 (t, J = 28.2 Hz,
2H), 1.99 (dd, J = 17.7, 6.9 Hz, 8H), 1.67 (s, 3H), 1.59 (s, 5H), 1.55
(s, 3H). 13C NMR (75 MHz, CDCl3) d 171.57, 168.04, 139.63,
136.60, 135.04, 133.98, 133.09, 131.15, 130.88, 129.32, 128.84,
127.92, 126.13, 124.27, 123.81, 119.51, 77.42, 77.00, 76.57, 58.58,
39.62, 39.53, 33.17, 30.02, 26.65, 26.46, 25.65, 17.64, 16.13, 15.95.
6.1.4.2. General synthetic methodology for bromides 9, 13 and
15. In general, phenol (1.0 equiv) (or the corresponding benzyl
alcohol) was charged to a dry round-bottom flask and anhydrous
DMF was added to it. The atmosphere in the flask was replaced
with argon. Upon complete dissolution of the phenol (or the benzyl
alcohol), the contents were cooled to 0 °C and allowed to stir for
15 min. This was followed by addition of sodium hydride
(1.1 equiv) and the compounds were allowed to react for one hour.
Following this time period, a solution of the appropriate benzyl
bromide (1.2 equiv) was added as a solution in anhydrous DMF
over a period of 10 min. The contents were allowed to react for
3 h. The reaction was quenched with ice and the product was ex-
tracted with ether, followed by washing with 10% citric acid
(3 Â 20 mL), then washed with 2 M NaOH (to remove the unre-
acted phenol) and brine. The organic fractions were pooled and
concentrated under vacuum to yield product. A chromatographic
separation was performed using 2:8 ethylacetate and hexanes if
it was necessary.
6.1.3.5. Compound 2e: (R)-2-(Dibenzo[b,d]furan-4-carboxa-
mido)-3-(((2E,6E)-3,7,11-trimethyldodeca-2,6,10-trien-1-yl)thio)-
propanoic acid. 1H NMR (300 MHz, CDCl3) d 8.56 (s, 1H), 8.09 (d,
J = 17.2 Hz, 1H), 7.85–7.42 (m, 3H), 7.35 (s, 1H), 7.22 (s, 2H), 5.15 (s,
1H), 5.09–4.81 (m, 3H), 3.38–3.04 (m, 4H), 1.94 (dd, J = 19.0, 7.0 Hz,
8H), 1.64 (s, 3H), 1.56 (s, 3H), 1.47 (d, J = 12.5 Hz, 5H). 13C NMR
(75 MHz, CDCl3) d 176.57, 168.75, 156.31, 152.11, 140.64, 136.04,
132.16, 129.75, 128.50, 125.29, 124.77, 124.21, 123.82, 121.35,
120.57, 112.87, 61.28, 40.60, 40.49, 34.29, 31.15, 27.65, 27.39,
26.64, 18.63, 17.00, 16.89, 12.33.
6.1.3.6. Compound 2f: (R)-2-(2-(Phenoxymethyl)benzamido)-3-
(((2E,6E)-3,7,11-trimethyldodeca-2,6,10-trien-1-yl)thio)prop-
anoic acid. 1H NMR (300 MHz, CDCl3) d 7.61 (s, 1H), 7.47 (d,
J = 6.6 Hz, 2H), 7.30 (d, J = 17.7 Hz, 1H), 7.21 (t, J = 7.0 Hz, 3H),
6.91 (dd, J = 20.5, 7.4 Hz, 3H), 5.34–4.92 (m, 5H), 4.61 (s, 1H),
2.97 (s, 3H), 2.62 (s, 1H), 2.10–1.79 (m, 8H), 1.68 (s, 3H), 1.60 (s,
3H), 1.55 (s, 3H), 1.51 (s, 3H).
6.1.4.3. General procedure for synthesis of carboxylic acids 10,
14 and 16. Magnesium turnings (4.0 equiv) were placed in a dry
round bottom flask and flame-dried under a steady stream of ar-
gon. Upon cooling of the flask to room temperature, anhydrous
THF and one drop of 1,2-dibromoethane were added to the flask.
The contents were allowed to react for 5 min followed by addition
of the appropriate bromide (as a solution in anhydrous THF) over a
period of 5 min. The Grignard reagent was allowed to form for
2.5 h and was quenched with dry ice. This was followed by addi-
tion of 10% hydrochloric acid (10 mL). The contents from the reac-
tion quench were transferred to a separatory funnel, followed by
addition of ethylacetate (20 mL). The aqueous phase was extracted
with ethylacetate (3 Â 20 mL). The combined organic extracts were
treated with 5% sodium hydroxide (3 Â 5 mL). Concentrated
hydrochloric acid was added to the combined aqueous phase until
the pH of the solution was 2.5, which resulted in the precipitation
of the product. Filtration of the precipitate yielded the desired car-
boxylic acid in 45–78% yields.
6.1.3.7. Compound 2g. 1H NMR (300 MHz, CDCl3) d 8.72 (d,
J = 7.2 Hz, 1H), 8.22 (dd, J = 7.8, 1.6 Hz, 1H), 7.50 (d, J = 6.5 Hz,
2H), 7.46–7.29 (m, 4H), 7.13–6.98 (m, 2H), 5.26 (d, J = 10.5 Hz,
2H), 5.10 (d, J = 18.0 Hz, 3H), 4.94 (dd, J = 12.1, 6.7 Hz, 1H), 3.08
(ddd, J = 20.4, 13.2, 7.8 Hz, 3H), 2.79 (ddd, J = 20.5, 13.9, 7.8 Hz,
2H), 2.05–1.90 (m, 8H), 1.67 (s, 3H), 1.62–1.57 (m, 9H). 13C NMR
(75 MHz, CDCl3) d 175.33, 164.65, 156.96, 139.67, 135.43, 135.19,
133.04, 132.37, 131.20, 129.17, 128.49, 127.99, 124.33, 123.68,
121.39, 120.94, 119.57, 112.68, 77.43, 77.00, 76.58, 71.22, 52.44,
52.29, 39.61, 39.53, 32.90, 29.68, 26.63, 26.34, 25.62, 17.62,
16.01, 15.93.
6.1.4.4. Compound 9: 1-Bromo-2-(phenoxymethyl)benzene. 1H
NMR (300 MHz, CDCl3) d 7.71 (d, J = 8.0 Hz, 2H), 7.45 (dt, J = 7.6,
6.4 Hz, 3H), 7.29 (dd, J = 12.1, 4.4 Hz, 1H), 7.14 (t, J = 8.1 Hz, 3H),
5.27 (s, 2H). 13C NMR (75 MHz, CDCl3) d 159.27, 137.20, 133.40,
130.41, 130.01, 129.69, 128.38, 123.11, 122.05, 115.71, 70.09.
6.1.3.8. Compound
amido)-3-(((2E,6E)-3,7,11-trimethyldodeca-2,6,10-trien-1-yl)-
thio)propanoic acid. 1H NMR (300 MHz, CDCl3)
8.10 (d,
2h: (R)-2-(2-((Benzyloxy)methyl)benz-
d
J = 6.8 Hz, 1H), 7.71 (t, J = 8.2 Hz, 1H), 7.36 (dd, J = 11.4, 4.5 Hz,
2H), 7.33–7.29 (m, 4H), 7.27 (d, J = 3.7 Hz, 1H), 5.19–4.99 (m,
3H), 4.66 (d, J = 11.3 Hz, 1H), 3.36 (s, 4H), 3.07 (dd, J = 12.5,
7.4 Hz, 2H), 2.87–2.71 (m, 2H), 2.11–1.88 (m, 8H), 1.65 (d,
J = 12.1 Hz, 3H), 1.58 (s, 6H), 1.56 (s, 3H). 13C NMR (75 MHz, CDCl3)
d 175.72, 170.17, 140.64, 138.58, 136.17, 136.09, 132.14, 131.54,
131.21, 130.05, 129.28, 129.22, 128.65, 128.59, 125.21, 124.69,
120.37, 73.07, 71.12, 51.07, 40.57, 40.47, 34.25, 30.78, 27.60,
27.36, 26.58, 18.57, 16.97, 16.87.
6.1.4.5. Compound 10: 2-(Phenoxymethyl)benzoic acid. 1H
NMR (300 MHz, CDCl3) d 11.72 (s, 1H), 8.23 (d, J = 7.8 Hz, 1H),
7.87 (d, J = 7.8 Hz, 1H), 7.65 (s, 1H), 7.45 (s, 1H), 7.38–7.25 (m,
2H), 7.11–6.93 (m, 3H), 5.61 (s, 2H). 13C NMR (75 MHz, CDCl3) d
173.87, 159.53, 141.79, 134.68, 132.73, 130.52, 130.48, 128.25,
127.24, 122.00, 115.93, 115.90, 69.02.
6.1.4.6. Compound 13: 1-(Benzyloxy)-2-bromobenzene. 1H
NMR (300 MHz, CDCl3) d 7.42 (dd, J = 7.8, 1.6 Hz, 1H), 7.33 (d,
J = 7.3 Hz, 2H), 7.27–7.10 (m, 3H), 7.02 (dd, J = 7.8, 1.1 Hz, 1H),
6.76–6.61 (m, 2H), 4.90 (s, 2H). 13C NMR (75 MHz, CDCl3) d
155.71, 137.26, 134.12, 129.29, 129.20, 128.63, 127.73, 127.71,
122.85, 114.49, 71.29.
6.1.4. Synthesis of compound 5
To a solution of potassium hydroxide (5.0 equiv) in water were
added 2-iodobenzoic acid (1.0 equiv), thiophenol (1.0 equiv) and
Copper powder (0.1 equiv). The contents were refluxed for 12 h,
cooled to room temperature and acidified with 4 N hydrochloric
acid. Acidification resulted in precipitation of product.
6.1.4.7. Compound 14: 2-(Benzyloxy)benzoic acid. 1H NMR
(300 MHz, CDCl3) d 10.52 (s, 1H), 7.77 (dt, J = 27.4, 13.7 Hz, 2H),
7.36 (d, J = 4.1 Hz, 1H), 7.32 (s, 2H), 7.00 (t, J = 7.5 Hz, 1H), 6.79
(ddd, J = 19.3, 17.3, 8.3 Hz, 3H), 5.15 (s, 2H).
6.1.4.1. Compound 5: 2-(Phenylthio)benzoic acid. 1H NMR
(300 MHz, CDCl3) d 7.85 (dd, J = 7.8, 1.4 Hz, 1H), 7.37–7.27 (m,
2H), 7.24–7.12 (m, 3H), 7.02–6.95 (m, 1H), 6.87 (dd, J = 10.9,
4.2 Hz, 1H), 6.52 (d, J = 8.1 Hz, 1H). 13C NMR (75 MHz, CDCl3) d

J. D. Majmudar et al. / Bioorg. Med. Chem. 20 (2012) 283–295
293
6.1.4.8. Compound 15: 1-((Benzyloxy)methyl)-2-bromoben-
zene. 1H NMR (300 MHz, CDCl3) d 7.62–7.49 (m, 2H), 7.49–7.27
(m, 6H), 7.23–7.08 (m, 1H), 4.69 (s, 4H). 13C NMR (75 MHz, CDCl3)
d 138.97, 138.57, 133.43, 130.06, 129.84, 129.41, 129.37, 128.67,
128.33, 123.68, 73.67, 72.50.
(d, J = 17.7 Hz, 16H), 0.78 (t, J = 6.7 Hz, 3H). 13C NMR (75 MHz,
MeOH) d 178.93, 174.66, 166.37, 157.14, 134.02, 133.69, 132.52,
130.86, 125.56, 124.19, 121.04, 119.56, 37.67, 35.60, 33.33, 33.05,
30.97, 30.81, 30.74, 30.71, 30.49, 23.49, 14.39, 5.36.
6.2. Materials and methods
6.1.4.9. Compound 16. 1H NMR (300 MHz, CDCl3) d7.72–7.49 (m,
3H), 7.49–7.31 (m, 5H), 7.25–7.11 (m, 1H), 4.73 (s, 4H). 13C NMR
(75 MHz, CDCl3) d 171.87, 137.97, 135.77, 133.43, 130.06, 129.84,
129.41, 129.37, 128.67, 127.45, 123.68, 72.58, 71.65.
6.2.1. Materials
N-2-Phenoxybenzoyl-S-3-(3-methyl-but-2-enyl)-5-(4-biphenyl)-
pent-2-enyl-L-cysteine (POP-3MB, compound 1b) was synthesized in
our laboratory25 via established solution-phase methods and charac-
terized to confirm identity and purity. Stock solutions were prepared
at 50 mM in DMSO and stored at À80 °C. EGFP-C3 vectors containing
GFP K-Ras was a generous gift from Mark Philips (New York Univer-
sity, School of Medicine). pGEX vector containing GST-RBD was a kind
gift from Lawrence Quilliam (Indiana University School of Medicine).
The following antibodies were used: Pan Ras (Oncogene, OP40), Phos-
pho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) (197G2) Rabbit mAb
6.1.4.10. Compound
17: (S)-2-(2-Phenoxybenzamido)-3-
(((2E,6E)-3,7,11-trimethyldodeca-2,6,10-trien-1-yl)thio)propa-
noic acid. 1H NMR (300 MHz, CDCl3) d 10.02 (bs, 1H), 8.77 (d,
J = 6.7 Hz, 1H), 8.22 (d, J = 7.1 Hz, 1H), 7.35 (t, J = 7.9 Hz, 3H),
7.23–7.04 (m, 4H), 6.83 (d, J = 8.2 Hz, 1H), 5.07 (dd, J = 8.1, 3.9 Hz,
3H), 4.95 (d, J = 5.5 Hz, 1H), 3.18–2.86 (m, 4H), 1.99 (tt, J = 27.6,
13.8 Hz, 8H), 1.64 (d, J = 14.5 Hz, 3H), 1.58 (d, J = 4.3 Hz, 6H), 1.55
(s, 3H). 13C NMR (75 MHz, CDCl3) d 174.65, 165.04, 155.79,
155.01, 139.55, 134.99, 132.96, 132.15, 131.03, 129.85, 124.49,
124.17, 123.67, 123.26, 122.50, 119.66, 119.41, 117.77, 52.82,
39.52, 39.40, 32.71, 29.74, 26.54, 26.27, 25.55, 17.54, 15.85.
(Cell Signaling Technology, 4377L), GAPDH (Ambion, 4300), CD-3
e
(UCHT1) (BD Pharmingen, 555329), ImmunoPure Goat Anti-Rabbit
IgG, Peroxidase Conjugated (GARP, Pierce Biotechnology, 31462),
ImmunoPure Goat Anti-Mouse IgG, Peroxidase Conjugated (GAMP,
Pierce Biotechnology, 31432). Formaldehyde 37% solution was pur-
chased from Mallinckrodt.
6.1.4.11. Compound 18: (R,E)-3-((3,7-Dimethylocta-2,6-dien-1-
yl)thio)-2-(2-phenoxybenzamido)propanoic
acid. 1H
NMR
(300 MHz, CDCl3) d 10.28 (s, 1H), 8.69 (d, J = 6.6 Hz, 1H), 8.20
(dd, J = 7.8, 1.7 Hz, 1H), 7.36 (dd, J = 10.7, 4.8 Hz, 3H), 7.16 (t,
J = 7.4 Hz, 2H), 7.12–7.05 (m, 2H), 6.84 (d, J = 8.2 Hz, 1H), 5.10–
4.98 (m, 2H), 4.91 (dd, J = 11.4, 6.0 Hz, 1H), 3.17–2.85 (m, 4H),
2.07–1.88 (m, 4H), 1.64 (s, 3H), 1.54 (d, J = 4.7 Hz, 6H). 13C NMR
(75 MHz, CDCl3) d 174.74, 165.43, 156.12, 155.24, 139.95, 133.33,
132.39, 131.75, 130.15, 124.84, 123.98, 123.55, 122.66, 119.94,
119.64, 118.08, 52.89, 39.63, 32.63, 29.93, 26.51, 25.78, 17.79,
16.09.
6.2.2. Cell culture
Jurkat T cells (E 6.1) were obtained and maintained as described
previously.41 For all experiments, cells were treated using a vehicle
(DMSO) or drug (diluted from a 50 mM DMSO stock solution) at
indicated concentrations for 24 h.
6.2.3. Fluorescence microscopy
Jurkat T cells (1.5 Â 107 cells/500
ted with 20 g of either GFP-K-Ras construct (a kind gift from M.
Phillips) using the Cell Porator (Life Technologies, 800 F, 250 V,
low ohms), were treated with delivery group (DMSO), simvastatin
alone (20 M), or various concentrations of POP-3 MB (0.1, 1 or
10 M) for 24 h in an incubator at 37 °C. The cells were harvested,
ll RPMI), transiently transfec-
l
l
6.1.4.12. Compound
19: (R)-2-(2-Phenoxybenzamido)-3-
(((2E,6E,10E)-3,7,11,15-tetramethylhexadeca-2,6,10,14-tetraen-
1-yl)thio)propanoic acid. 1H NMR (500 MHz, CDCl3) d 9.73–9.72
(m, 1H), 8.85 (d, J = 6.8 Hz, 1H), 8.03 (dd, J = 7.8, 1.6 Hz, 1H),
7.46–7.32 (m, 3H), 7.18 (td, J = 7.6, 1.0 Hz, 2H), 7.12–7.05 (m,
2H), 6.89–6.82 (m, 1H), 5.16–4.98 (m, 4H), 4.68 (d, J = 4.8 Hz,
1H), 3.16–2.89 (m, 5H), 2.12–1.86 (m, 12H), 1.63 (d, J = 0.9 Hz,
3H), 1.61–1.49 (m, 12H). 13C NMR (126 MHz, CDCl3) d 177.26,
166.97, 157.50, 157.26, 140.42, 136.37, 136.08, 134.39, 132.73,
132.29, 131.40, 125.96, 125.70, 125.44, 125.36, 124.70, 121.87,
121.20, 119.63, 114.89, 50.10, 41.12, 41.05, 40.93, 34.66, 31.04,
28.07, 27.83, 27.68, 26.19, 18.06, 16.47, 16.43, 16.40.
l
l
washed with PBS three times, and plated on polylysine coverslips
for 10 min. The coverslips were washed with PBS for 10 min (3
times) and the cells were fixed with 3% formaldehyde for 10 min.
The coverslips were washed again for 10 min with PBS (3 times).
After mounting the coverslips onto the slides using FluorSave™
Reagent (Calbiochem), the slides were dried for 45 min at room
temperature. Images of cells were viewed using an Olympus BH-
2 microscope and captured with a QImaging Microimager II. The
captured images were viewed using Northern Eclipse software.
Cells were classified as either full (fully mislocalized GFP-K-Ras
construct with diffuse cellular staining), partial (partial proper
localization), and normal (normal, primarily plasma membrane
localization of GFP-K-Ras) by visual inspection.40,41 The subcellular
localization of GFP K-Ras was quantified using fluorescence
microscopy (performed on an Olympus BH-2RFCA) as previously
described.23,50
6.1.4.13. Compound 20: (R)-2-(N-Methyl-2-phenoxybenzami-
do)-3-(((2E,6E)-3,7,11-trimethyldodeca-2,6,10-trien-1-
yl)thio)propanoic acid. 1H NMR (300 MHz, MeOH) d 7.33–7.11
(m, 4H), 7.10–6.93 (m, 3H), 6.92–6.81 (m, 2H), 6.80–6.56 (m,
1H), 4.94 (d, J = 1.3 Hz, 3H), 4.30–4.13 (m, 1H), 3.12–2.98 (m,
1H), 2.94 (d, J = 7.6 Hz, 1H), 2.87 (d, J = 5.8 Hz, 1H), 2.81 (s, 2H),
2.72 (dd, J = 13.9, 4.2 Hz, 1H), 2.03–1.97 (m, 2H), 1.97–1.75 (m,
8H), 1.51 (s, 4H), 1.44 (s, 6H). 13C NMR (75 MHz, MeOH) d
171.94, 157.82, 136.14, 132.05, 131.93, 130.91, 130.87, 125.44,
125.38, 125.13, 124.55, 121.60, 121.09, 120.98, 120.36, 119.12,
62.48, 40.84, 40.70, 30.69, 29.84, 27.78, 27.46, 25.97, 17.84,
16.31, 16.24, 16.20.
6.2.4. Isolation of membrane fraction from yeast cells
Membrane fractions from yeast cells were isolated as previously
described.5 Crude membrane protein concentration was deter-
mined using Coomassie Plus Protein Assay Reagent (Pierce) accord-
ing to the manufacturer’s instructions, and compared with a
bovine serum albumin standard curve.
6.1.4.14. Compound 21: ((R)-2-(2-Phenoxybenzamido)-3-(unde-
cylthio)propanoic acid). 1H NMR (300 MHz, MeOH) d 7.93 (dd,
J = 7.8, 1.7 Hz, 1H), 7.37–7.18 (m, 3H), 7.18–6.93 (m, 4H), 6.88–
6.65 (m, 1H), 4.48 (t, J = 5.0 Hz, 1H), 2.94 (ddd, J = 33.3, 13.6,
5.0 Hz, 2H), 2.34 (dd, J = 10.9, 3.9 Hz, 2H), 1.39–1.23 (m, 2H), 1.12
6.2.5. Substrate and inhibition assays byin vitro vapor diffusion
methyltransferase assay
Methyltransferase assays were performed as previously de-
scribed5,38,39 with minor modifications. The assay mixture con-

294
J. D. Majmudar et al. / Bioorg. Med. Chem. 20 (2012) 283–295
tained a total volume of 60
l
L and a final Tris–HCl buffer concen-
M S-adeno-
g of membrane
Acknowledgments
tration of 100 mM, pH 7.5. All reactions contained 20
l
syl-[14C-methyl]methionine (14C-SAM) and 5
l
We thank Professors Mark Philips (NYU School of Medicine) and
Lawrence Quilliam (Indiana University School of Medicine) for the
generous gifts of the GFP-KRas and the GST-RBD constructs,
respectively. Funding was provided by the NCI [R01CA112483
(R.A.G.) and P30CA21328 (Purdue University Center for Cancer Re-
search CCSG)].
protein from His-hIcmt-overexpressing strain CH 2766.5 Initial
substrate screening was performed at a single concentration of
25
and compound (10
lowed by the addition of 25
tions were incubated at 30 °C for 30 min and stopped by the
addition of 50 L of 1 M NaOH/1% SDS. 100 L of this mixture
l
M in duplicate, three times. For inhibition screening, protein
M) were pre-incubated on ice in buffer, fol-
M AFC substrate and 14C-SAM. Reac-
l
l
l
l
Supplementary data
was spotted on folded filter paper (6 cm  2 cm) and lodged in
the neck of a scintillation vial containing 10 mL of scintillation
fluid, and capped to allow for diffusion of the released [14C]meth-
anol into the scintillant. The filters were pulled out after 2.5 h,
the vials tightly recapped and shaken, and then counted in a liquid
scintillation analyzer.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2011.10.087.
References and notes
1. Gelb, M. H.; Brunsveld, L.; Hrycyna, C. A.; Michaelis, S.; Tamanoi, F.; Van
Voorhis, W. C.; Waldmann, H. Nat. Chem. Biol. 2006, 2, 518.
2. Bergo, M. O.; Gavino, B. J.; Hong, C.; Beigneux, A. P.; McMahon, M.; Casey, P. J.;
Young, S. G. J. Clin. Invest. 2004, 113, 539.
3. Winter-Vann, A. M.; Casey, P. J. Nat. Rev. Cancer 2005, 5, 405.
4. Young, S. G.; Clarke, S. G.; Bergo, M. O.; Phillips, M.; Fong, L. G. The Enzymes
2006, 24, 273.
5. Anderson, J. L.; Frase, H.; Michaelis, S.; Hrycyna, C. A. J. Biol. Chem. 2005, 280,
7336.
6. Dai, Q.; Choy, E.; Chiu, V.; Romano, J.; Slivka, S. R.; Steitz, S. A.; Michaelis, S.;
Philips, M. R. J. Biol. Chem. 1998, 273, 15030.
6.2.6. Immunoblotting
Jurkat T cells (3–5 Â 105 cells mLÀ1) were treated with vehicle
(DMSO), simvastatin, or POP-3 MB (1a). Cells were lysed in 1%
Triton X-100 lysis buffer (25 mM Hepes, pH 7.2, 150 mM NaCl,
1% NP40, 5 mM EDTA, 10 l
g mLÀ1 of leupeptin and aprotinin,
and 1 mM sodium vanadate), resolved on SDS–PAGE gels, resolved
by SDS–PAGE on 12% gels and transferred to polyvinylidene difluo-
ride membranes (Immun-Blot PVDF; Bio-Rad). Membranes were
blocked with 5% goat serum and incubated with primary antibody
(anti-Pan Ras (Oncogene OP40) for N/K Ras; or anti-GAPDH
(Ambion 4300) for GAPDH) at room temperature for 1 h, followed
by incubation with horseradish peroxidase-conjugated goat
anti-mouse or goat anti-rabbit IgG. Membranes were developed
using enhanced chemiluminescence reagents (ECL; Amersham
Pharmacia Biotech) and autoradiography film. Quantitative
western blotting was performed using ImageJ system.
7. Clarke, S. Ann. Rev. Biochem. 1992, 61, 355.
8. Zhang, F. L.; Casey, P. J. Ann. Rev. Biochem. 1996, 65, 241.
9. Basso, A. D.; Kirschmeier, P.; Bishop, W. R. J. Lipid Res. 2006, 47, 15.
10. Otto, J. C.; Kim, E.; Young, S. G.; Casey, P. J. J. Biol. Chem. 1999, 274, 8379.
11. Bergo, M. O.; Leung, G. K.; Ambroziak, P.; Otto, J. C.; Casey, P. J.; Gomes, A. Q.;
Seabra, M. C.; Young, S. G. J. Biol. Chem. 2001, 276, 5841.
12. Bergo, M. O.; Leung, G. K.; Ambroziak, P.; Otto, J. C.; Casey, P. J.; Young, S. G. J.
Biol. Chem. 2000, 275, 17605.
13. Cohen, S. J.; Ho, L.; Ranganathan, S.; Abbruzzese, J. L.; Alpaugh, R. K.; Beard, M.;
Lewis, N. L.; McLaughlin, S.; Rogatko, A.; Perez-Ruixo, J. J.; Thistle, A. M.;
Verhaeghe, T.; Wang, H.; Weiner, L. M.; Wright, J. J.; Hudes, G. R.; Meropol, N. J.
J. Clin. Oncol. 2003, 21, 1301.
14. Macdonald, J. S.; McCoy, S.; Whitehead, R. P.; Iqbal, S.; Wade, J. L., 3rd; Giguere,
J. K.; Abbruzzese, J. L. Invest. New Drugs 2005, 23, 485.
15. Van Cutsem, E.; van de Velde, H.; Karasek, P.; Oettle, H.; Vervenne, W. L.;
Szawlowski, A.; Schoffski, P.; Post, S.; Verslype, C.; Neumann, H.; Safran,
H.; Humblet, Y.; Perez Ruixo, J.; Ma, Y.; Von Hoff, D. J. Clin. Oncol. 2004,
22, 1430.
16. Whitehead, R. P.; McCoy, S.; Macdonald, J. S.; Rivkin, S. E.; Neubauer, M. A.;
Dakhil, S. R.; Lenz, H. J.; Tanaka, M. S.; Abbruzzese, J. L. Invest. New Drugs 2006,
24, 335.
6.2.7. Signaling assay
Jurkat T cells (1 Â 107 cells/1 ml) were washed in PBS, incubated
with anti-CD-3e (20 lg/ml) for 15 min on ice, and stimulated at
37 °C for 10 min. Cells were washed and lysed in lysis buffer con-
taining 1% Triton X-100, 20 mM Tris/HCl, pH 7.6, 150 mM NaCl,
20 lg/ml each of leupeptin and aprotinin, 1 mM pyrophosphate,
1 mM glycerol phosphate, and 1 mM sodium orthovanadate. Sam-
ples were separated on 6–15% gradient SDS–polyacrylamide gels
and analyzed by immunoblotting with anti-phosphoERK antibod-
ies (at 4 °C overnight with anti-p-Erk (Cell Signaling Technology
4377L)). Membranes were developed using enhanced chemilumi-
nescence reagents (ECL; Amersham Pharmacia Biotech) and auto-
radiography film. Quantitative western blotting was performed
using ImageJ system.
17. Whyte, D. B.; Kirschmeier, P.; Hockenberry, T. N.; Nunez-Oliva, I.; James, L.;
Catino, J. J.; Bishop, W. R.; Pai, J. K. J. Biol. Chem. 1997, 272, 14459.
18. Adjei, A. A. J. Natl. Cancer Inst. 2001, 93, 1062.
19. Rowell, C. A.; Kowalczyk, J. J.; Lewis, M. D.; Garcia, A. M. J. Biol. Chem. 1997, 272,
14093.
20. Michaelson, D.; Ali, W.; Chiu, V. K.; Bergo, M.; Silletti, J.; Wright, L.; Young, S. G.;
Philips, M. Mol. Biol. Cell 2005, 16, 1606.
21. Volker, C.; Lane, P.; Kwee, C.; Johnson, M.; Stock, J. FEBS Lett. 1991, 295, 189.
22. Roberts, P. J.; Mitin, N.; Keller, P. J.; Chenette, E. J.; Madigan, J. P.; Currin, R. O.;
Cox, A. D.; Wilson, O.; Kirschmeier, P.; Der, C. J. J. Biol. Chem. 2008, 283, 25150–
25163.
23. Wang, M.; Tan, W.; Zhou, J.; Leow, J.; Go, M.; Lee, H. S.; Casey, P. J. J. Biol. Chem.
2008, 283, 18678–18684.
24. Wahlstrom, A. M.; Cutts, B. A.; Liu, M.; Lindskog, A.; Karlsson, C.; Sjogren, A. K.;
Andersson, K. M.; Young, S. G.; Bergo, M. O. Blood 2008, 112, 1357–1365.
25. Donelson, J. L.; Hodges-Loaiza, H. B.; Henriksen, B. S.; Hrycyna, C. A.; Gibbs, R.
A. J. Org. Chem. 2009, 74, 2975.
26. Tan, E. W.; Perez-Sala, D.; Canada, F. J.; Rando, R. R. J. Biol. Chem. 1991, 266,
10719.
27. Volker, C.; Lane, P.; Kwee, C.; Johnson, M.; Stock, J. FEBS Lett. 1991, 295, 189.
28. Volker, C.; Miller, R. A.; McCleary, W. R.; Rao, A.; Poenie, M.; Backer, J. M.; Stock,
J. B. J. Biol. Chem. 1991, 266, 21515.
29. Perez-Sala, D.; Gilbert, B. A.; Tan, E. W.; Rando, R. R. Biochem. J. 1992, 284(Pt 3),
835.
30. Henriksen, B. S.; Anderson, J. L.; Hrycyna, C. A.; Gibbs, R. A. Bioorg. Med. Chem.
Lett. 2005, 15, 5080.
31. Tan, E. W.; Rando, R. R. Biochemistry 1992, 31, 5572.
32. Stephenson, R. C.; Clarke, S. J. Biol. Chem. 1990, 265, 16248.
33. Anderson, J. L.; Henriksen, B. S.; Gibbs, R. A.; Hrycyna, C. A. J. Biol. Chem. 2005,
280, 29454.
34. Donelson, J. L.; Hodges, H. B.; MacDougall, D. D.; Henriksen, B. S.; Hrycyna, C.
A.; Gibbs, R. A. Bioorg. Med. Chem. Lett. 2006, 16, 4420.
6.2.8. GST-RBD pull down assay
GST-RBD fusion protein was expressed in bacteria as previously
described.51 After stimulation with anti CD3 antibody, Jurkat T
cells (2 Â 107) were lysed in 1% NP40 lysis buffer (1% Nonidet
P-40, 25 mM HEPES, 150 mM NaCl, 0.25% sodium deoxycholate,
10% glycerol, 10 mM MgCl2, 1 mM EDTA, 1 mM sodium vanadate,
10 mM NaF, 20 g/ml aprotinin and leupeptin, pH 7.4) and centri-
fuged at 464 Â g for 3 min at 4 °C. The lysate was incubated with
GST-RBD fusion protein conjugated glutathione-sepharose 4B
beads for 30 min at 4 °C. The resin was washed three times with
magnesium lysis buffer (25 mM HEPES, pH 7.5, 150 mM NaCl, 1%
NP-40, 0.25% sodium deoxycholate 10% glycerol, 10 mM MgCl2),
solubilized in 2Â PSB (protein sample buffer, protein sample
buffer = 100 mM Tris–Hcl, pH 6.8, 20% glycerol, 4%SDS, 0.2% bro-
mophenol blue, 200 mM dithiothreitol) and the presence of acti-
vated Ras was detected by immunoblot analysis and quantified
using the ImageJ system as mentioned above.

J. D. Majmudar et al. / Bioorg. Med. Chem. 20 (2012) 283–295
295
35. Bonnet, B.; Soullez, D.; Girault, S.; Maes, L.; Landry, V.; Davioud-Charvet, E.;
Sergheraert, C. Bioorg. Med. Chem. 2000, 8, 95.
46. Haklai, R.; Elad-Sfadia, G.; Egozi, Y.; Kloog, Y. Cancer Chemother. Pharmacol.
2008, 61, 89.
36. Majmudar, J. D.; Hahne, K.; Hrycyna, C. A.; Gibbs, R. A. Bioorg. Med. Chem. Lett.
2011, 21, 2616.
47. Jansen, B.; Schlagbauer-Wadl, H.; Kahr, H.; Heere-Ress, E.; Mayer, B. X.; Eichler,
H. G.; Pehamberger, H.; Gana-Weisz, M.; Ben-David, E.; Kloog, Y.; Wolff, K. Proc.
Natl. Acad. Sci. U.S.A. 1999, 96, 14019.
48. Judd, W. R.; Slattum, P. M.; Hoang, K. C.; Bhoite, L.; Valppu, L.; Alberts, G.;
Brown, B.; Roth, B.; Ostanin, K.; Huang, L.; Wettstein, D.; Richards, B.;
Willardsen, J. A. J. Med. Chem. 2011, 54, 5031.
37. Ruggles, E. L.; Flemer, S.; Hondal, R. J. Biopolymers 2008, 90, 61.
38. Hrycyna, C. A.; Yang, M. C.; Clarke, S. Biochemistry 1994, 33, 9806.
39. Hrycyna, C. A.; Clarke, S. Mol. Cell Biol. 1990, 10, 5071.
40. Choy, E.; Philips, M. Methods in Enzymology, Part F 2001, 332, 50.
41. Krzysiak, A. J.; Rawat, D. S.; Scott, S. A.; Pais, J. E.; Handley, M.; Harrison, M. L.;
Fierke, C. A.; Gibbs, R. A. ACS Chem. Biol. 2007, 2, 385.
49. Brown, M. J.; Milano, P. D.; Lever, D. C.; Epstein, W. W.; Poulter, C. D. J. Am.
Chem. Soc. 1991, 113, 3176.
42. Perez de Castro, I.; Bivona, T. G.; Philips, M. R.; Pellicer, A. Mol. Cell Biol. 2004,
24, 3485.
50. McCabe, J. B.; Berthiaume, L. G. Mol. Biol. Cell 1999, 10, 3771.
51. Castro, A. F.; Rebhun, J. F.; Quilliam, L. A. Methods (San Diego, Calif) 2005, 37, 190.
43. Castro, A. F.; Rebhun, J. F.; Quilliam, L. A. Methods 2005, 37, 190.
44. Winter-Vann, A. M.; Baron, R. A.; Wong, W. H.; dela Cruz, J.; York, J. D.; Gooden,
D. M.; Bergo, M. O.; Young, S. G.; Toone, E. J.; Casey, P. J. Proc. Natl. Acad. Sci.
U.S.A. 2005, 102, 4336.
52. Robert, A. Evaluation of Enzyme Inhibitors in Drug Discovery: A Guide for
Medicinal Chemists and Pharmacologists; Wiley Press: Copeland, 2005. 131.
53. Cer, R. Z.; Mudunuri, U.; Stephens, R.; Lebeda, F. J. Nucleic Acids Res. 2009, 37,
W441–W445.
45. Tsimberidou, A. M.; Rudek, M. A.; Hong, D.; Ng, C. S.; Blair, J.; Goldsweig, H.;
Kurzrock, R. Cancer Chemother. Pharmacol. 2010, 65, 235.