Polyisoprenylated methylated protein methyl esterase: A putative biomarker and therapeutic target for pancreatic cancer

Article abstract of DOI:10.1016/j.ejmech.2014.05.018

Pancreatic cancer is the most deadly neoplasm with a 5-year survival rate of less than 6%. Over 90% of cases harbor K-Ras mutations, which are the most challenging to treat due to lack of effective therapies. Here, we reveal that polyisoprenylated methyla

Full text of DOI:10.1016/j.ejmech.2014.05.018

European Journal of Medicinal Chemistry 81 (2014) 323e333
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Polyisoprenylated methylated protein methyl esterase: A putative
biomarker and therapeutic target for pancreatic cancer
Byron J. Aguilar, Augustine T. Nkembo, Randolph Duverna, Rosemary A. Poku,
*
Felix Amissah, Seth Y. Ablordeppey, Nazarius S. Lamango
College of Pharmacy and Pharmaceutical Sciences, Florida A&M University, Tallahassee, FL 32307, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Pancreatic cancer is the most deadly neoplasm with a 5-year survival rate of less than 6%. Over 90% of
cases harbor K-Ras mutations, which are the most challenging to treat due to lack of effective therapies.
Here, we reveal that polyisoprenylated methylated protein methyl esterase (PMPMEase) is overexpressed
in 93% of pancreatic ductal adenocarcinoma. We further present polyisoprenylated cysteinyl amide in-
hibitors (PCAIs) as novel compounds designed with structural elements for optimal in vivo activities and
selective disruption of polyisoprenylation-mediated protein functions. The PCAIs inhibited PMPMEase
Received 6 March 2014
Received in revised form
30 April 2014
Accepted 4 May 2014
Available online 9 May 2014
with Ki values ranging from 3.7 to 20
BxPC-3 cell lines were as low as 1.9 M while salirasib and farnesylthiosalicylamide were ineffective at
20 M. The PCAIs thus have the potential to serve as effective therapies for pancreatic and other cancers
with hyperactive growth signaling pathways mediated by Ras and related G-proteins.
Ó 2014 Elsevier Masson SAS. All rights reserved.
mM. The 48 h EC50 values for pancreatic cancer Mia PaCa-2 and
Keywords:
PMPMEase
m
m
Pancreatic cancer
Polyisoprenylation
Prenylation
Polyisoprenylated cysteinyl amides
PCAIs
1. Introduction
constitutively active state [5]. K-Ras is metabolized via the poly-
isoprenylation pathway (PP) and efforts to inhibit its maturation
and localization have shown anticancer benefit in PC [6]. The PP
encompasses a final reversible step that may possibly regulate the
proteins’ function. Polyisoprenylated protein methyl transferase
(PPMTase) esterifies the carboxyl terminus whereas poly-
isoprenylated methylated protein methyl esterase (PMPMEase)
hydrolyzes the ester bond. Substrate analysis of PMPMEase [7,8] led
to the synthesis of high affinity, irreversible sulfonyl fluoride in-
hibitors (Fig. 1) [9]. Meanwhile, polyunsaturated fatty acids
(PUFAs), which are known to exhibit anticancer activity, have been
shown to inhibit PMPMEase [10]. The decrease in cancer cell
viability due to inhibition of PMPMEase with PUFAs, synthetic, and
other chemopreventive agents highlights its potential as an anti-
cancer target.
Pancreatic ductal adenocarcinoma (PDA) accounts for approxi-
mately 95% of all pancreatic cancers (PC) and remains one of the
deadliest neoplasms as the number of new diagnoses closely mir-
rors deaths. The poor prognosis is attributed to lack of early
symptoms, with >50% of patients diagnosed at a late stage of the
disease [1]. Epidermal growth factor receptor (EGFR) plays a vital
role in the progression of PC [2] and is associated with poor prog-
nosis [3]. Research to uncover the molecular aberrations in PC has
revealed gene mutations in important oncogenic proteins such as
K-Ras, which functions downstream of EGFR. Mutations in K-Ras
are involved in >90% of PC cases [4]. Point mutations on this
oncogene diminish or abolish its GTPase activity leaving it in the
The study aims were three-fold. The first aim was to determine
the expression of PMPMEase in PDA and in normal pancreatic tis-
sue in view of the potential to utilize PMPMEase as a possible
biomarker and therapeutic target. Second, to design and synthesize
polyisoprenylated cysteinyl amide inhibitors (PCAIs), a first-in-class
group of compounds to either inhibit PMPMEase and/or disrupt the
polyisoprenylation-dependent interactions of the proteins it me-
tabolizes. The design strategy incorporated three key elements to
maximize selectivity, biochemical stability and bioavailability.
These included, (1) the farnesyl group for high affinity/selective
Abbreviations: EGFR, epidermal growth factor receptor; FTS, S-t, t farnesylth-
iosalicylic acid (Salirasib); FTSA, S-t, t farnesylthiosalicylamide; NT, normal tissue;
NAT, normal adjacent tissue; PC, pancreatic cancer; PCAIs, polyisoprenylated cys-
teinyl amide inhibitors; PDA, pancreatic ductal adenocarcinoma; PMPMEase, pol-
yisoprenylated methylated protein methyl esterase; PMSF, phenylmethylsulfonyl
fluoride; PP, polyisoprenylation pathway; PPMTase, polyisoprenylated protein
methyl transferase; PUFAs, polyunsaturated fatty acids; RD-PNB, L-N-(4-
nitrobenzoyl)-S-t, t-farnesyl-cysteine methyl ester; TMA, tissue microarray.
* Corresponding author.
E-mail address: Nazarius.lamango@famu.edu (N.S. Lamango).
http://dx.doi.org/10.1016/j.ejmech.2014.05.018
0223-5234/Ó 2014 Elsevier Masson SAS. All rights reserved.

324
B.J. Aguilar et al. / European Journal of Medicinal Chemistry 81 (2014) 323e333
adding an ionizable functional group (pyrrolidine or N-methyl-
piperazine) to diminish the excessive hydrophobicity. The calcu-
lated LogD values ranged from 3.15 to 5.88 at the physiological pH
of 7.4 whereas the calculated LogD value for the C15 poly-
isoprenylated sulfonyl fluoride was 5.04. In addition, the unstable,
highly reactive sulfonyl fluoride group that irreversibly inhibits
PMPMEase was substituted with an amide group that functions as a
more stable bioisostere of the substrate ester bond. The resulting
compounds, unlike the sulfonyl fluorides, are reversible PMPMEase
inhibitors.
The PCAIs were prepared according to Scheme 1. Briefly, the
starting alcohol is first converted to the bromide (2) using phos-
phorous tribromide. The resulting bromide (2) was used to alkylate
Fig. 1. Structure of PMPMEase (PMSF, L-28) and Ras (FTS and FTSA) inhibitors.
L-cysteine methyl ester to form the alpha amino ester which un-
derwent alkylation with bromoalkynoyl chloride to produce com-
pound 4. Compound 4 then served as the alkylating agent for the
selected amines (pyrrolidine and 1-methylpiperazine) designed to
impart solubility, prior to the amidation of the ester function
overnight as a neat reaction at 90 ꢀC. However, several low boiling
amines and di-substituted amines showed no reaction and required
the use of the corresponding acyl chlorides. The acyl chlorides were
obtained by hydrolyzing the ester bonds and converting the
products to the required acyl chlorides before reaction with the low
boiling point amines. Reactions with the amines resulted in com-
pounds 7aeg (pyrrolidines) and 8aeg (1-methylpiperazines).
interactions, (2) a substituted amide bioisostere of the scissile ester
bond found in the endogenous substrates, and (3) an ionizable
appendage group designed to mitigate the excessive hydropho-
bicity of the farnesyl cysteinyl amide that constitutes the phar-
macophore. Finally, to examine the effect of PMPMEase inhibition
by PCAIs in PC cell lines.
The potential beneficial impact of these compounds was further
investigated by examining the expression of PMPMEase in PDA
relative to normal pancreatic tissues that might substantiate the
use of the PCAIs as effective targeted therapies. PMPMEase over-
expression in PDA would not only contribute to its validation as a
putative drug target but also as a biomarker for potential early/
companion diagnosis and prognosis. The results strongly suggest
possible beneficial outcomes for the continuous development of
the PCAIs as a novel class of targeted therapies for PC.
2.2. Docking analysis
The in silico analysis of the PCAIs revealed compounds with
AScore docking energies ranging from ꢁ17.21 to ꢁ13.35 kcal/mol
(Table 1). A representative of the PCAIs, 8a is shown in the active
site of PMPMEase in Fig. 2. The pyrrolidine derivatives showed
lower docking energies (ꢁ17.21 to ꢁ14.03 kcal/mol) compared to
the N-methylpiperazine derivatives (ꢁ15.08 to ꢁ13.35 kcal/mol).
Compound 7d had the lowest AScore docking energy
of ꢁ17.21 kcal/mol versus ꢁ14.26 kcal/mol for its N-methylpiper-
azinyl derivative (8d).
2. Results and discussion
2.1. Chemistry
The design and synthesis of polyisoprenylated sulfonyl fluorides
was the first successful attempt at developing inhibitors of
PMPMEase [9]. Incorporation of a polyisoprenyl tail increased the
inhibitory potency determined using pseudo-first order kinetics by
330-fold compared to the lead compound PMSF. Although these
compounds are potent inhibitors, lack of polar groups means that
they are very hydrophobic (ClogP ranged from 2.8 to 4.3). Addi-
tionally, the highly reactive sulfonyl fluoride functional group is
moisture-sensitive. PCAIs address these former limitations by
2.3. Biological evaluation
The clinicopathologic data of the PC cases are summarized in
Table 2. The age range of the donors was 1 month to 78 years with a
mean age of 53 years. Intense PMPMEase immunoreactivity was
observed in pancreatic tumors. Fig. 3 is a PC TMA showing varying
Scheme 1. Synthesis of polyisoprenylated cysteinyl amide inhibitors (PCAIs) of PMPMease. Reagents: (a) ꢁ10 ꢀC, PBr3, Et2O (b) TEA, L-cysteine methyl ester, Anh. MeOH (c) 6-
bromohexanoyl chloride, TEA, DCM, (d) K2CO3, amine, toluene, (e) amine, 90 ꢀC, ON.

B.J. Aguilar et al. / European Journal of Medicinal Chemistry 81 (2014) 323e333
325
Table 1
Physicochemical, docking, and inhibition kinetics data for polyisoprenylated cysteinyl amide inhibitors of PMPMEase.
Cpd
R1
R2
LogD
3.15
AScore (kcal/mol)
Ki (
mM)
MIA PaCa-2
BxPC-3
48 h EC50 (mM)
48 h EC50
(
m
M)
7a
ꢁ14.03
17.5
>20
>20
7b
7c
7d
4.04
4.48
4.93
ꢁ16.93
ꢁ15.78
ꢁ17.21
3.7
3.2
5.6
6.8
6.8
12.7
5.8
11.2
11.6
7e
5.37
ꢁ16.73
11.9
1.9
4.0
7f
3.31
3.76
ꢁ16.10
ꢁ14.70
9.7
10.8
12.4
6.5
8.7
7g
11.0
8a
3.66
ꢁ14.24
13.8
10.4
6.3
8b
8c
8d
4.55
4.99
5.44
ꢁ13.66
ꢁ13.35
ꢁ14.26
10.1
16.9
17.6
10.9
5.9
>20
12
2.7
6.2
8e
5.88
ꢁ15.08
15.2
2.5
7.5
8f
3.82
4.27
ꢁ14.53
ꢁ14.32
20.0
13.0
6.0
3.2
8.8
8g
14.3
FTS
FTSA
e
e
e
e
3.43
6.03
ND
ND
ND
ND
>20
>20
>20
>20
degrees of inter- and intra-slice brown staining indicative of the
varying tissue expression levels of PMPMEase in various PC cases.
The intra-tissue slice distributions appear to be concentrated in
tumors as the areas of intense brown staining, indicative of high
levels of PMPMEase, are accompanied by larger numbers of blue-
stained nuclei. The scoring of PMPMEase expression in various
pancreatic tissues is summarized in Table 3. PC normal adjacent
tissue (NAT) and normal pancreatic tissue (NT) showed trace to
weak PMPMEase staining with mean scores of 145.0 ꢂ 8.6 and
115 ꢂ 7.6, respectively. With the exception of only three PDA cores,
all the cores (93%) showed intermediate to very strong PMPMEase
staining with a mean score of 393.6 ꢂ 14.4. The positive expression

326
B.J. Aguilar et al. / European Journal of Medicinal Chemistry 81 (2014) 323e333
Fig. 2. Docking of PCAIs in PMPMEase. A) A representative PCAI, compound 8a, is shown in the active site of PMPMEase. The amide bond (X) is in close proximity to the catalytic
serine (SER221, Y) and the methylpiperazine ring (Z) is located outside of the active site of the protein. B) Docking pose of compound 8a is shown surrounded by the active site
hydrophobic residues.
of PMPMEase in the PDA cores was significantly higher than both
NAT and NT (p < 0.0001). PMPMEase expression compared to tu-
mor grade, tumor status, and metastasis (Table 3) was also signif-
icantly higher compared to both NAT and NT (p < 0.0001).
dependent manner (Fig. 4). The EC50 values for the effect of the
PCAIs against MIA PaCa-2 cells ranged from 1.9 to >20
EC50 values against BxPC-3 cells ranged from 3.2 to >20
m
m
M and the
M for 48 h
(Table 1). On the other hand, the Ras inhibitors, Salirasib (FTS) and
farnesylthiosalicylamide (FTSA, Fig. 1), showed no observable ef-
PMPMEase expression was increased in donors with chronic
inflammation. Mild chronic inflammation (6 cases) showed a mean
score of 204.2 ꢂ 48.5 which is statistically significant compared to
NT (p ¼ 0.0342), but was not significant compared to NAT (Table 3).
Chronic inflammation (2 cases) showed a statistically higher mean
score of 412.5 ꢂ 12.5 compared to both NAT and NT (p < 0.0001).
Malignant tumor tissues also showed high levels of PMPMEase
immunoreactivity. Other tumor types (benign, hyperplasia, and
inflammation) showed a 1.5- to 2-fold increase in staining while
malignant tumor tissues (median score ¼ 374.5 ꢂ 55.2) showed
fects on the viability of the cells at 20
same treatment conditions (Fig. 5).
mM concentrations under the
In order to determine the mechanism of cell death, the PC cells
were treated with 8e for 48 h in the presence of 5% FBS and stained
with acridine orange/ethidium bromide (AO/EB) according to a
previously described method [12,13]. As shown in Fig. 5, signs of
apoptosis were observed in MIA PaCa-2 cells treated with 0.5, 1 and
2
m
M concentrations of compound 8e. AO permeated the live
control cells as well as cells exposed to the lower concentration of
8e (1 M) turning their nuclei green. Cells treated with concen-
trations of 2 M 8e fluoresced red due to EB uptake. Early apoptotic
about
a
3-fold increase in PMPMEase staining intensity as
m
compared to both controls (Table 3).
m
All of the PCAIs inhibited PMPMEase in a concentration-
cells with bright green fragmented nuclei were observed after 48 h
treatment of MIA PaCa-2 cells with 8e.
dependent manner with Ki values ranging from 3.7 to 20
mM
(Table 3). Of the secondary amides, ring size affected inhibitor po-
tency. In the 5-membered pyrrolidinyl series (7aee), cyclo-
The design, synthesis, and in vitro analysis of the current batch
of PCAIs demonstrate that polyisoprenylated small molecules that
effectively disrupt polyisoprenylated protein metabolism and/or
functional interactions to address cancers with hyperactive G-
proteins is entirely possible. Although the exact mechanism(s) of
action of the PCAIs against the cancer cell lines remains to be fully
explored, their effectiveness against the MIA PaCa-2 cells indicates
that they would be effective against cancers with the oncogenic K-
Ras mutations. Over 90% of PC cases [4] have this oncogenic
transformation, which is the highest proportion that has been
observed in any other type of cancer. The strategy successfully
accomplished three key objectives to obtain (1) aqueous-stable
polyisoprenylated small molecules that are (2) soluble in aqueous
buffers and can (3) be effective against tumors with not only the K-
Ras oncogene but also those with hyperactive EGFR signaling of
which Ras is a key mediator.
pentylamide analog (7b, Ki ¼ 3.7
mM) was observed to be the most
potent inhibitor of PMPMEase. Substituting the cyclopentyl moiety
with a cyclopropyl group decreased the potency by 4.7-fold (7a,
Ki ¼ 17.5
m
M) while increasing the ring size to a 6 (7c), 7 (7d) or 8-
M. A similar trend
membered ring (7e) yielded Ki values of 11e12
m
was also observed with the N-methylpiperazine derivatives (8aee).
The cyclopentylamide derivative, compound 8b, was the most
potent (Ki ¼ 10.1
mM). The Ki values for the 3-, 6-, 7-, and 8-
membered rings (8a, 8c, 8d, and 8e, respectively) were 13.8, 16.9,
17.6, and 15.2, respectively. Tertiary amide derivatives (7f, 7g, 8f,
and 8g) had Ki values ranging from 9.7 to 20 mM and here again the
pyrrolidinyl analogs showing a higher potency (Table 1).
The PCAIs were screened against MIA PaCa-2 pancreatic cancer
cells that harbor the mutant K-Ras oncogene and the pancreatic
cancer BxPC-3 cells that lack the mutation [11]. The cancer cells
were cultured and treated in 5% FBS and treated with PCAIs at
MIA PaCa-2 and BxPC-3 PC cells treated with PCAIs resulted in
significant cell degeneration. However, there is a lack of correlation
between the Ki values for PMPMEase inhibition and EC50 values
against the cell lines. This suggests that the compounds’ effects on
concentrations ranging from 0 to 20
mM for 48 h. The results show
that the PCAIs induce cell degeneration in a concentration-

B.J. Aguilar et al. / European Journal of Medicinal Chemistry 81 (2014) 323e333
327
Table 2
current drug therapies. The PCAIs are a new class of agents that not
only target a preponderant genetic aberration in PC but combines
physico-chemical elements for their effective delivery. These have
the potential to address the oncogenic Ras therapeutic void.
Clinicopathologic data for the 98 pancreatic tissue donors. Grade, tumor status,
nodal status, and metastasis are taken from the 46 carcinoma samples.
Characteristics
Donors
Characteristics
Donors
(%)
N
(%)
N
Age
Grade
1
2
3
Not determined
Tumor status
4. Experimental section
ꢄ53
41
57
41.8
58.2
8
17.4
32.6
28.3
21.7
>53
15
13
10
4.1. Synthetic procedure
Sex
Male
Female
57
41
58.2
41.8
Reagents were purchased from SigmaeAldrich (Milwaukee,
WI), Acros (Morris Plains, NJ), TCI America (Portland, OR), Alfa Aesar
(Ward Hill, MA), ChemImpex International (Wood Dale, IL) and
Santa Cruz Biotechnology (Santa Cruz, CA) and used without
further purification. Flash column chromatography was carried on
Histology
1
2
3
4
2
27
14
2
4.3
58.7
30.4
4.3
Adenosquamous carcinoma
Chronic inflammation
Chronic pancreatitis
Duct adenocarcinoma
Hyperplasia
Islet cell carcinoma
Islet cell tumor
Mild chronic inflammation
Type
Normal
NAT
Benign
Hyperplasia
Inflammation
Malignant
3
2
2
42
2
1
3.1
2.0
2.0
42.9
2.0
1.0
Not determined
Nodal status
0
1
Not determined
Metastasis
0
1
1
2.2
40e63
300 (300 MHz) in CDCl3, unless otherwise stated. Chemical shifts
) are given in ppm relative to the signal for the deuterated solvent
m SiO2. NMR spectra were obtained from a Varian Mercury
36
8
2
78.3
17.4
4.3
10
6
10.2
6.1
(d
and are reported consecutively as position (dH), relative integral,
multiplicity (s ¼ singlet, d ¼ doublet, t ¼ triplet, q ¼ quartet,
dd ¼ doublet of doublets, m ¼ multiplet and where br ¼ broad),
coupling constant (J/Hz) and assignment. Salirasib (farnesylth-
iosalicylic acid or FTS) and farnesylthiosalicylamide (FTSA) were
synthesized as previously described [15].
10
20
10
2
10
46
10.2
20.4
10.2
2.0
10.2
46.9
42
3
1
91.3
6.5
2.2
Not determined
cell viability may include mechanisms that involve PMPMEase in-
hibition and/or other disruptions of polyisoprenylation-mediated
proteineprotein interactions. Like FTS, PCAIs share the farnesyl
moiety essential for proteineprotein interactions [14]. The PCAIs
may function in a similar manner to FTS, which is known to disrupt
Ras function by dislodging it from the membrane [15]. For example,
galectin-3 is a selective binding partner of activated, GTP-bound K-
Ras [16] that forms Ras nanoclusters essential for cell signaling.
Signaling is largely dependent on these nanoclusters and can be
blocked if Ras localization is altered [17]. Relative biochemical in-
stabilities between the PCAIs in the cancer cell lines may also ac-
count for some of the differences between the inhibition kinetics
against the enzyme activity and the cell viability.
4.2. Preparative purification and MS analysis
Reaction mixtures were purified via reversed-phase high per-
formance liquid chromatography (RP-HPLC) using a Shimadzu
Prominence HPLC system with UV detection at 214 nm. Samples
were loaded onto a Hamilton PRP-1 (21.5 ꢃ150 mm ID) or Hamilton
PRP-1 (10 ꢃ 250 mm ID) column and eluted with a linear gradient
of acetonitrile in 0.1% trifluoroacetic acid at 5 mL/min or 3 mL/min,
respectively. MS analysis of the fractions was achieved using 4800
MALDI/TOF on linear low mass mode. Fractions containing only the
target compound were combined and the solvents removed under
vacuum and dried at 40 ꢀC and 3 mm Hg.
K-Ras mutations are common in PC and affects cell signaling
downstream of EGFR. Mutations lead to constitutively active Ras
and subsequent activation of the Ras/MAPK signaling pathway,
which has been associated with anti-EGFR resistance [18,19]. It has
been reported that K-Ras mutations are indicative of resistance to
anti-EGFR monoclonal antibodies and TKIs in PC [20,21] and that K-
Ras may serve as a plausible target for cancer treatment. K-Ras
requires polyisoprenylation for proper localization, but potential
drugs that inhibit its polyisoprenylation produced significant
adverse side effects [22]. PMPMEase, which plays a vital role in
regulating the function of polyisoprenylated proteins such as K-Ras,
may serve as a viable alternative drug target for inhibiting or
reversing PC tumor growth. It has recently been shown that siRNA
targeting hCE1/PMPMEase disrupt F-actin organization by inhibit-
ing the demethylation of RhoA leading to altered cell morphology
[23]. This is consistent with our own studies showing the effects of
PMPMEase inhibition-induced changes on cell morphology [9,10].
This is further indication that the PCAIs may be effective not only at
abrogating cell proliferation and tumor growth but also cell
migration and invasion that promote metastasis, the most devas-
tating feature of malignant tumors.
4.3. Synthesis of trans, trans-farnesyl bromide (2)
A solution of PBr3 (0.75 mL, 8 mmol) in anhydrous ether (5 mL)
was added drop-wise to trans, trans-farnesol (1, 5 g, 22.9 mmol) and
anhydrous ether (10 mL) at ꢁ15 to ꢁ10 ꢀC. When the addition was
complete, the mixture was kept at 0 ꢀC for a further 10 min and
then at 20 ꢀC for 30 min. The organic phase was decanted, washed
with cold 5% NaHCO3, brine, dried over MgSO4 and the solvent
removed at 40 ꢀC/3 mm Hg, yielding 2 as a pale-yellow liquid (6.4 g,
98%). 1H NMR:
d 1.60 (s, 6H), 1.67 (s, 3H), 1.72 (s, 3H), 1.95e2.15 (m,
8H), 4.16 (d, J ¼ 8.4 Hz, 2H), 5.09 (m, 2H), 5.54 (m, 1H).
4.4. Synthesis of L-S-(trans, trans-farnesyl) cysteine methyl ester
(3)
A mixture of trans, trans-farnesyl bromide (2, 4.56 g, 16 mmol),
L-cysteine methyl ester hydrochloride (2.75 g, 16 mmol) in anhy-
drous methanol (28 mL) and 7 N anhydrous NH3 solution in
methanol (92 mL) at ꢁ5 ꢀC was stirred at this temperature for 2 h.
The solvent was evaporated under vacuum and the residue was
then purified by flash chromatography, eluting with hexaneeEtOAc
3. Conclusions
1:1 (1.69 g, 31%). Rf 0.2 (hexaneeEtOAc 1:1). 1H NMR:
d 1.48 (s, 3H),
Targeting PMPMEase offers a potentially effective strategy for
treating cancers with activating mutations and/or overexpression
of both EGFR and monomeric G-proteins that confer resistance to
1.51 (s, 3H), 1.59 (s, 3H), 1.66 (s, 3H), 1.65 (s, 2H), 1.80e2.10 (m, 8H),
2.61e2.77 (dd, 1H, J ¼ 7.5 Hz), 3.09 (d, J ¼ 7.2 Hz, 2H), 3.55 (dd,
J ¼ 3.0 Hz, 1H), 3.65 (s, 3H), 5.04 (br, 2H), 5.14 (t, J ¼ 1.2 Hz, 1H).

328
B.J. Aguilar et al. / European Journal of Medicinal Chemistry 81 (2014) 323e333
Fig. 3. PMPMEase immunoreactivity in pancreatic tumor and normal tissue. I) Selected pancreatic cancer TMA cores of ductal adenocarcinoma (BeD), adenosquamous cell car-
cinoma (E and F) and benign islet cell carcinoma (G and H) showing varying degrees of brown staining indicating varying expression levels of PMPMEase. A represents an image of
normal pancreatic tissue slice. II) Magnification of selected pancreatic cancer TMA cores of duct adenocarcinoma (A2 to B1), adenosquamous cell carcinoma (B2 and B3) and benign
islet cell carcinoma (C1 to C3) showing varying degrees of brown staining indicating varying expression levels of PMPMEase. A1 represents an image of normal pancreatic tissue.
(For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
4.5. Synthesis of L-((pyrrolidinyl) hexanoyl)-S-(trans, trans-
4.6. Synthesis of L-((4-methylpiperazinyl) hexanoyl)-S-(trans,
farnesyl) cysteine methyl ester (5)
trans-farnesyl) cysteine methyl ester (6)
6-Bromohexanoyl chloride (1.4 mL, 9.2 mmol) was added drop-
wise to a stirred solution of L-S-(trans, trans-farnesyl) cysteine
methyl ester (3, 3.0 g, 8.8 mmol) and TEA (1.5 mL, 10.5 mmol) at
6-Bromohexanoyl chloride (1.4 mL, 9.2 mmol) was added drop-
wise to a stirred solution of L-S-(trans, trans-farnesyl) cysteine
methyl ester (3, 3.0 g, 8.8 mmol) and TEA (1.5 mL,10.5 mmol) at 0 ꢀC
and the mixture stirred for 3 h at RT until completion of the reac-
tion (monitored by TLC). The mixture was then washed with cold
5% NaHCO3, brine, dried over MgSO4, and the solvent removed
0
ꢀC and the mixture stirred for 3 h at RT until completion of the
reaction as determined by TLC. The mixture was then washed with
cold 5% NaHCO3, brine, dried over MgSO4, and the solvent
removed under vacuum resulting in a pale yellow oil (4). L-(6-
bromohexanoyl)-S-(trans, trans-farnesyl) cysteine methyl ester
(4) was added to a solution of pyrrolidine (0.82 mL, 10 mmol) and
K2CO3 (1.1 g, 10 mmol) in 25 mL toluene and refluxed for 3 h. The
mixture was filtered, solvent was removed and the residue was
purified by HPLC to afford the product (5, 3.5 g, 80%). 1H NMR:
under vacuum resulting in
a pale yellow oil (4). L-(6-
bromohexanoyl)-S-(trans, trans-farnesyl) cysteine methyl ester
(4) was added to a solution of 1-methylpiperazine (1.1 mL,
10 mmol) and K2CO3 (1.1 g, 10 mmol) in 25 mL toluene and refluxed
for 3 h. The mixture was filtered, solvent was removed and the
residue was purified by HPLC to afford the product (6, 4 g, 85%). 1H
d
1.18e1.51 (m, 3H), 1.51e1.86 (m, 17H), 1.90e2.17 (m, 12H), 2.19e
NMR: d 1.17e1.44 (m, 2H), 1.55e1.76 (m, 17H), 1.89e2.15 (m, 10H),
2.37 (m, 2H), 2.84 (dt, J ¼ 9.9, 8.5 Hz, 3H), 3.02e3.26 (m, 5H),
3.73e3.85 (m, 3H), 4.72 (ddd, J ¼ 29.2, 15.6, 5.1 Hz, 1H), 5.09 (t,
J ¼ 6.0 Hz, 2H), 5.20 (t, J ¼ 7.6 Hz, 1H), 6.77 (t, J ¼ 18.1 Hz, 1H).
2.26 (t, J ¼ 7.3 Hz, 2H), 2.43e2.57 (m, 3H), 2.58e2.70 (m, 2H), 2.70e
3.04 (m, 7H), 3.04e3.28 (m, 2H), 3.66e3.80 (m, 3H), 4.78 (td, J ¼ 6.3,
4.9 Hz, 1H), 5.09 (t, J ¼ 6.0 Hz, 2H), 5.20 (t, J ¼ 7.8 Hz, 1H), 6.44 (d,
J ¼ 7.6 Hz, 1H). MALDI/TOF-MS m/z 536.67 [Mþ1] (calcd. for
MALDI/TOF-MS
m/z
¼
507.61
[Mþ1]
(calcd.
for
C
29H50N2O3S ¼ 506.78).
C30H53N3O3S ¼ 535.83).

B.J. Aguilar et al. / European Journal of Medicinal Chemistry 81 (2014) 323e333
329
Table 3
PMPMEase expression in various pancreatic tissue. Mean ꢂ SEM.
Characteristic
PMPMEase staining intensity, N (%)
Total cases
Mean score
Normal
p-value
t-test
NAT
1e100
101e200
201e300
301e400
401e500
p-value
t-test
Trace
Weak
Intermediate
Strong
Very strong
Histology
Normal pancreatic tissue
Normal adjacent tissue
Adenosquamous carcinoma
Chronic inflammation
Chronic pancreatitis
Duct adenocarcinoma
Hyperplasia
Islet cell carcinoma
7 (70.0)
6 (30.0)
0 (0.0)
0 (0.0)
1 (50.0)
0 (0.0)
0 (0.0)
0 (0.0)
4 (40.0)
2 (33.3)
3 (30.0)
14 (70.0)
2 (66.7)
0 (0.0)
1 (50.0)
3 (7.1)
1 (50.0)
0 (0.0)
1 (10.0)
1 (16.7)
0 (0.0)
0 (0.0)
1 (33.3)
0 (0.0)
0 (0.0)
0 (0.0)
0 (0.0)
0 (0.0)
1 (50.0)
0 (0.0)
17 (40.5)
0 (0.0)
0 (0.0)
1 (10.0)
1 (16.7)
0 (0.0)
0 (0.0)
0 (0.0)
1 (50.0)
0 (0.0)
19 (45.2)
0 (0.0)
0 (0.0)
1 (10.0)
0 (0.0)
10
20
3
2
2
42
2
1
10
6
115.0 ꢂ 7.6
145.0 ꢂ 8.6
183.3 ꢂ 46.4
412.5 ꢂ 12.5
112.5 ꢂ 12.5
393.6 ꢂ 14.4
212.5 ꢂ 62.5
250.0 ꢂ 0.0
225.0 ꢂ 48.3
204.2 ꢂ 48.5
0.0268
<0.0001
ns
ns
<0.0001
ns
3 (7.1)
<0.0001
0.0059
e
<0.0001
0.0446
e
1 (50.0)
1 (100.0)
3 (30.0)
2 (33.3)
Islet cell tumor
Mild chronic inflammation
Type
0.0372
0.0342
0.0333
ns
Benign
Hyperplasia
Inflammation
Malignant
4 (40.0)
0 (0.0)
3 (30.0)
0 (0.0)
1 (10.0)
1 (50.0)
2 (20.0)
5 (10.9)
3 (30.0)
1 (50.0)
2 (20.0)
5 (10.9)
1 (10.0)
0 (0.0)
2 (20.0)
18 (39.1)
1 (10.0)
0 (0.0)
1 (10.0)
18 (39.1)
10
2
10
46
225.0 ꢂ 48.3
212.5 ꢂ 62.5
227.5 ꢂ 43.4
374.5 ꢂ 55.2
0.0372
0.0059
0.0199
<0.0001
0.0333
0.0446
0.0171
<0.0001
Grade
1
2
3
0 (0.0)
0 (0.0)
0 (0.0)
2 (25.0)
0 (0.0)
0 (0.0)
1 (12.5)
1 (6.7)
1 (7.1)
3 (37.5)
6 (40.0)
7 (50.0)
2 (25.0)
8 (53.3)
6 (42.9)
8
15
14
328.1 ꢂ 39.9
411.7 ꢂ 19.6
408.9 ꢂ 22.2
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
Tumor status
1
2
3
4
0 (0.0)
0 (0.0)
0 (0.0)
0 (0.0)
0 (0.0)
2 (7.4)
3 (20.0)
0 (0.0)
0 (0.0)
3 (11.1)
1 (6.7)
0 (0.0)
0 (0.0)
11 (40.7)
5 (33.3)
1 (50.0)
2 (100.0)
11 (40.7)
6 (40.0)
2
27
15
2
425.0 ꢂ 25.0
384.3 ꢂ 19.2
356.7 ꢂ 31.9
450.0 ꢂ 50.0
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
<0.0001
1 (50.0)
Nodal status
0
1
0 (0.0)
0 (0.0)
2 (5.4)
3 (37.5)
3 (8.1)
1 (12.5)
16 (43.2)
2 (25.0)
16 (43.2)
2 (25.0)
37
8
390.5 ꢂ 15.6
325.0 ꢂ 51.1
<0.0001
0.0003
<0.0001
<0.0001
Metastasis
0
1
0 (0.0)
0 (0.0)
5 (11.6)
0 (0.0)
4 (9.3)
0 (0.0)
18 (41.9)
0 (0.0)
16 (37.2)
3 (100.0)
43
3
374.4 ꢂ 16.3
458.3 ꢂ 22.0
<0.0001
<0.0001
<0.0001
<0.0001
4.7. General procedure for the synthesis of compounds 7aeg
6.39 (s, 1H), 6.92 (s, 1H). MALDI/TOF-MS m/z 574.66 [Mþ1] (calcd.
for C34H59N3O3S ¼ 573.92).
Compound 5 (100 mg, 0.20 mmol) and the respective amine
(0.4 mL, 3.14e5.80 mmol) were heated as a neat reaction at 90 ꢀC
overnight. The mixture was then purified on a HPLC system to
afford the respective products.
4.7.4. L-((pyrrolidinyl) hexanoyl)-S-(trans, trans-farnesyl) cysteine
cycloheptylamide (7d, NSL-BA-040)
Yielded 43 mg, 37%. 1H NMR
d 1.30e1.84 (m, 3H), 1.85e2.22 (m,
12H), 2.26 (dd, J ¼ 13.3, 6.3 Hz, 2H), 2.78 (d, J ¼ 6.7 Hz, 4H), 3.09 (s,
2H), 3.26e3.17 (m, 2H), 3.85 (d, J ¼ 24.2 Hz, 3H), 4.44 (d, J ¼ 6.7 Hz,
1H), 5.06e5.13 (m, 2H), 5.23 (t, J ¼ 7.3 Hz, 1H), 6.64 (d, J ¼ 7.7 Hz,
1H), 7.13 (d, J ¼ 7.0 Hz, 1H). MALDI/TOF-MS m/z 588.75 [Mþ1]
(calcd. for C35H61N3O2S for 587.94).
4.7.1. L-((pyrrolidinyl) hexanoyl)-S-(trans, trans-farnesyl) cysteine
cyclopropylamide (7a, NSL-BA-037)
Yielded 30 mg, 28%. 1H NMR
d 0.54 (s, 2H), 0.75 (s, 2H), 0.82e
1.06 (m, 2H), 1.07e1.55 (m, 18H), 1.55e1.85 (m, 5H), 1.99e2.29 (m,
5H), 2.80 (s, 5H), 3.12 (d, J ¼ 32.0 Hz, 4H), 3.80 (s, 4H), 4.45 (s, 2H),
5.09 (s, 2H), 5.22 (s, 1H), 6.78 (s, 1H). MALDI/TOF-MS m/z 532.58
[Mþ1] (calcd. for C31H53N3O2S ¼ 531.84).
4.7.5. L-((pyrrolidinyl) hexanoyl)-S-(trans, trans-farnesyl) cysteine
cyclooctylamide (7e, NSL-BA-055)
Yielded 45 mg, 37%. 1H NMR
d 1.24e1.47 (m, 2H), 1.57 (d,
4.7.2. L-((pyrrolidinyl) hexanoyl)-S-(trans, trans-farnesyl) cysteine
J ¼ 16.7 Hz, 17H), 1.66 (d, J ¼ 13.4 Hz, 8H), 1.80 (t, J ¼ 20.0 Hz, 3H),
1.89e2.20 (m, 12H), 2.24 (q, J ¼ 6.9 Hz, 2H), 2.80 (s, 4H), 3.07 (s, 3H),
3.22 (s, 2H), 3.81 (s, 24H), 3.95 (s, 2H), 4.43 (d, J ¼ 6.8 Hz, 2H), 5.09
(dd, J ¼ 6.7, 5.5 Hz, 2H), 5.24 (t, J ¼ 7.6 Hz, 1H), 6.54 (d, J ¼ 8.0 Hz,
1H), 7.00 (d, J ¼ 6.9 Hz, 1H). MALDI/TOF-MS m/z 602.51 [Mþ1]
(calcd. for C36H63N3O2S ¼ 601.97).
cyclopentylamide (7b, NSL-BA-038)
Yielded 35 mg, 31%. 1H NMR
d 1.34e1.52 (m, 4H), 1.52e1.87 (m,
20H), 2.03 (ddd, J ¼ 24.6, 13.7, 5.9 Hz, 14H), 2.26 (t, J ¼ 7.2 Hz, 2H),
2.79 (d, J ¼ 6.7 Hz, 4H), 3.10 (dd, J ¼ 14.5, 9.8 Hz, 2H), 3.16e3.29 (m,
2H), 3.81 (d, J ¼ 4.9 Hz, 2H), 4.09e4.22 (m, 1H), 4.45 (q, J ¼ 7.0 Hz,
1H), 5.05e5.50 (m, 2H), 5.23 (t, J ¼ 7.4 Hz, 1H), 6.67 (d, J ¼ 7.1 Hz,
1H), 7.09 (d, J ¼ 7.2 Hz, 1H). MALDI/TOF-MS m/z 560.53 [Mþ1]
(calcd. for C33H57N3O2S ¼ 559.89).
4.7.6. N-((R)-1-oxo-1-(pyrrolidin-1-yl)-3-((trans, trans farnesyl)
thio)propan-2-yl)-6-(pyrrolidin-1-yl)hexanamide (7f, NSL-BA-041)
4.7.3. L-((pyrrolidinyl) hexanoyl)-S-(trans, trans-farnesyl) cysteine
Yielded 27 mg, 25%. 1H NMR
d 0.83e1.48 (m, 6H), 1.70 (dd,
cyclohexylamide (7c, NSL-BA-039)
J ¼ 37.3, 18.3 Hz, 14H), 1.81e2.29 (m, 16H), 2.80 (s, 4H), 3.12 (d,
J ¼ 40.5 Hz, 4H), 3.41 (d, J ¼ 35.0 Hz, 2H), 3.54e3.93 (m, 4H), 4.88 (s,
1H), 5.09 (s, 2H), 5.22 (s, 1H), 6.82 (s, 1H). MALDI/TOF-MS m/z
546.34 [Mþ1] (calcd. for C32H55N3O2S ¼ 545.86).
Yielded 34 mg, 30%. 1H NMR
d 1.13e1.44 (m, 12H), 1.71 (t,
J ¼ 37.1 Hz, 17H), 2.16 (d, J ¼ 58.3 Hz, 12H), 2.79 (s, 4H), 3.14 (d,
J ¼ 47.6 Hz, 4H), 3.80 (s, 4H), 4.44 (s, 1H), 5.09 (s, 2H), 5.24 (s, 1H),

330
B.J. Aguilar et al. / European Journal of Medicinal Chemistry 81 (2014) 323e333
Fig. 4. Concentration-response curves of pancreatic cancer BxPC-3 (A) and MIA PaCa-2 (B) cells treated with 0e20
A) and FTSA (,) were ineffective at 20 M.
mM of PCAIs for 48 h in media containing 5% FBS. Salirasib (FTS,
m
4.7.7. N-((R)-1-oxo-1-(piperidin-1-yl)-3-((trans, trans farnesyl)
6.70 (d, J ¼ 7.4 Hz, 1H), 7.20 (s, 1H). MALDI/TOF-MS m/z 589.94
[Mþ1] (calcd. for C32H56N4O2S ¼ 588.93).
thio)propan-2-yl)-6-(pyrrolidin-1-yl)hexanamide (7g, NSL-BA-048)
Yielded 29 mg, 26%. 1H NMR
d 0.75e1.49 (m, 4H), 1.64 (d,
J ¼ 22.9 Hz, 19H), 1.88e2.36 (m, 14H), 2.79 (s, 3H), 3.06 (s, 7H), 3.54
4.8.3. L-((4-methylpiperazinyl) hexanoyl)-S-(trans, trans-farnesyl)
(d, J ¼ 18.5 Hz, 4H), 3.82 (s, 2H), 5.09 (s, 2H), 5.21 (s,1H), 6.73 (s,1H).
cysteine cyclohexylamide (8c, NSL-BA-058/NSL-RD-035)
Yielded 25 mg, 21%. 1H NMR
d 1.09e1.52 (m, 12H), 1.67 (dd,
MALDI/TOF-MS
m/z
560.361
[Mþ1]
(calcd.
for
C
33H57N3O2S ¼ 559.89).
J ¼ 23.1, 19.4 Hz, 19H), 1.91e2.13 (m, 8H), 2.26 (s, 2H), 2.76 (d,
J ¼ 5.1 Hz, 1H), 2.92 (s, 2H), 3.08e3.26 (m, 4H), 3.68 (s, 8H), 4.44 (d,
J ¼ 6.7 Hz,1H), 5.09 (t, J ¼ 6.5 Hz, 2H), 5.23 (t, J ¼ 7.5 Hz,1H), 6.55 (d,
J ¼ 7.9 Hz, 1H), 7.05 (d, J ¼ 6.9 Hz, 1H). MALDI/TOF-MS m/z 603.58
[Mþ1] (calcd. for C32H56N4O2S ¼ 602.96).
4.8. General procedure for the synthesis of compounds 8aeg
Compound 6 (100 mg, 0.20 mmol) and the respective amine
(0.4 mL, 3.14e5.80 mmol) were heated as a neat reaction at 90 ꢀC
overnight. The mixture was then purified on a HPLC system to
afford the respective products.
4.8.4. L-((4-methylpiperazinyl) hexanoyl)-S-(trans, trans-farnesyl)
cysteine cycloheptylamide (8d, NSL-BA-4-57/NSL-RD-036)
Yielded 45 mg, 36%. 1H NMR
d 1.11e1.30 (m, 2H), 1.34e1794 (m,
4.8.1. L-((4-methylpiperazinyl) hexanoyl)-S-(trans, trans-farnesyl)
29H), 1.86e2.15 (m, 8H), 2.28 (s, 2H), 2.75 (s, 2H), 2.94 (s, 3H), 3.18
(d, J ¼ 10.8 Hz, 4H), 3.71 (s, 7H), 3.88 (s, 1H), 4.45 (d, J ¼ 6.4 Hz, H),
5.09 (t, J ¼ 6.2 Hz, 2H), 5.18e5.27 (m, 1H), 6.70 (d, J ¼ 7.6 Hz, 1H),
7.20 (s, 1H). MALDI/TOF-MS m/z 617.64 [Mþ1] (calcd. for
cysteine cyclopropylamide (8a, NSL-BA-045)
Yielded 30 mg, 28%. 1H NMR
d 7.02e6.79 (m, 1H), 5.22 (s, 1H),
5.09 (s, 2H), 4.56e4.35 (m, 1H), 3.64 (s, 8H), 3.13 (d, J ¼ 22.9 Hz, 4H),
2.88 (s, 3H), 2.82e2.64 (m, 3H), 2.25 (s, 2H), 2.17e1.85 (m, 7H),
1.81e1.12 (m, 19H), 0.76 (d, J ¼ 6.9 Hz, 2H), 0.52 (s, 2H). MALDI/TOF-
MS m/z 561.29 [Mþ1] (calcd. for C32H56N4O2S ¼ 560.88).
C
32H56N4O2S ¼ 616.98).
4.8.5. L-((4-methylpiperazinyl) hexanoyl)-S-(trans, trans-farnesyl)
cysteine cyclooctylamide (8e, NSL-BA-056)
4.8.2. L-((4-methylpiperazinyl) hexanoyl)-S-(trans, trans-farnesyl)
Yielded 35 mg, 29%. 1H NMR
d 0.76e1.02 (m, 2H), 1.12e1.31 (m,
cysteine cyclopentylamide (8b, NSL-BA-049)
2H), 1.37e1.83 (m, 32H), 2.02 (dd, J ¼ 25.3, 6.5 Hz, 6H), 2.21 (d,
J ¼ 23.8 Hz, 2H), 2.83 (d, J ¼ 39.2 Hz, 5H), 3.04e3.27 (m, 4H), 3.66 (s,
6H), 3.92 (s, 1H), 4.42 (s, 1H), 5.08 (d, J ¼ 6.6 Hz, 2H), 5.24 (s, 2H),
6.55 (d, J ¼ 7.4 Hz, 1H), 6.99 (s, 1H). MALDI/TOF-MS m/z 631.62
[Mþ1] (calcd. for C32H56N4O2S ¼ 631.01).
Yielded 32 mg, 29%. 1H NMR
d 0.76e0.96 (m, 2H), 1.17e1.35 (m,
5H), 1.43 (s, 3H), 1.67 (dd, J ¼ 23.9, 18.5 Hz, 20H), 1.86e2.17 (m, 7H),
2.28 (s, 2H), 2.75 (s,1H), 2.94 (s, 3H), 3.19 (s, 4H), 3.71 (s, 6H), 4.13 (s,
1H), 4.45 (m, 1H), 5.08 (d, J ¼ 6.8 Hz, 2H), 5.21 (d, J ¼ 7.1 Hz, 1H),

B.J. Aguilar et al. / European Journal of Medicinal Chemistry 81 (2014) 323e333
331
Fig. 5. PCAIs induce cell death by apoptosis. MIA PaCa-2 cells were treated with 8e (0.5e2
mM) for 48 h and stained with AO/EB 10
mg/mL. Viable cells do no absorb EB while AO is
absorbed turning their intact nuclei green. Apoptotic cells pick up EB and are orange to red with highly condensed nuclei. The results presented are representative of triplicate
experiments (N ¼ 4). BF: Bright Field; AO: Acridine Orange; EB: Ethidium Bromide. OL: overlay of AO and EB. (For interpretation of the references to color in this figure legend, the
reader is referred to the web version of this article.)
4.8.6. N-((R)-1-oxo-1-(piperidin-1-yl)-3-((trans, trans farnesyl)
thio)propan-2-yl)-6-(4-methylpiperazin-1-yl)hexanamide (8f, NSL-
BA-007)
gradient of 0.1% ethanolamine and increasing acetonitrile. Aceto-
nitrile increased to 35% in 1.5 min and maintained for 1 min before
95% acetonitrile for 1 min and re-equilibrated with 0% acetonitrile
for 1 min.
Yielded 27 mg, 25%. 1H NMR
d
1.26 (dd, J ¼ 46.4, 31.1 Hz, 3H),
1.57e1.83 (m, 15H), 1.98 (ddd, J ¼ 21.9, 13.6, 6.6 Hz, 12H), 2.25 (t,
J ¼ 6.9 Hz, 2H), 2.65e2.97 (m, 5H), 3.03e3.25 (m, 4H), 3.45 (t,
J ¼ 6.5 Hz, 2H), 3.53e3.81 (m, 10H), 4.87 (q, J ¼ 7.4 Hz, 1H), 5.08 (dd,
J ¼ 9.5, 3.8 Hz, 2H), 5.21 (t, J ¼ 7.2 Hz, 1H), 7.05 (d, J ¼ 7.2 Hz, 1H).
4.10. Statistical analysis
The data were expressed as mean ꢂ S.E.M. Nonlinear regression
analysis of the inhibitoreresponse curves were calculated using
GraphPad Prism version 5.0 for Mac (San Diego, CA) and the con-
centrations that inhibit 50% of enzymatic activity (IC50) was
determined. The inhibition constant (Ki) was then calculated from
the previously reported kinetics analysis of RD-PNB substrate [7,8].
MALDI/TOF-MS
m/z
575.43
[Mþ1]
(calcd.
for
C
33H58N4O2S ¼ 574.90).
4.8.7. N-((R)-1-oxo-1-(piperidin-1-yl)-3-((trans, trans farnesyl)
thio)propan-2-yl)-6-(4-methylpiperazin-1-yl)hexanamide (8g, NSL-
BA-048)
Yielded 28 mg, 26%. 1H NMR
d
1.09 (dd, J ¼ 83.9, 27.4 Hz, 4H),
4.11. Effect of PCAIs on pancreatic cancer cells
1.43 (s, 5H), 1.67 (dd, J ¼ 22.9, 18.3 Hz, 22H), 1.90e2.18 (m, 5H), 2.27
(s, 2H), 2.67 (s, 1H), 2.92 (s, 5H), 3.13 (s, 4H), 3.53 (s, 3H), 3.68 (s,
6H), 5.08 (s, 2H), 5.21 (s, 1H), 6.97 (s, 2H). MALDI/TOF-MS m/z
590.31 [Mþ1] (calcd. for C34H60N4O2S ¼ 588.93).
The pancreatic adenocarcinoma cell lines BxPC-3 and MIA PaCa-
2 were obtained from the American Type Culture Collection
(Manassas, VA). BxPC-3 and MIA PaCa-2 were routinely grown in
RPMI 1640 and DMEM 1ꢃ growth medium (Invitrogen, Carlsbad,
CA) respectively. These growth media were each supplemented
4.9. Enzyme assays
with 10% fetal bovine serum, 100 units/mL penicillin and 100 mg/mL
PMPMEase enzyme inhibition analysis was conducted as pre-
viously described [9]. PCAIs were dissolved in water and RD-PNB
substrate was dissolved in DMSO. After pre-incubation of the
streptomycin (Atlanta Biological, Atlanta, GA). Cells were constantly
maintained in a humidified incubator containing 5% CO2 at 37 ꢀC
until about 80% confluent. Cells were trypsinized, counted and
enzyme (1
for 15 min, RD-PNB (1 mM) was added and incubated for 2 h at
37 ꢀC in 100 mM TriseHCl, pH 7.4 in a total volume of 100
L. The
L methanol and stored
mg) and varying concentration of inhibitor (1e1000
mM)
plated in 96-well plate at 1 ꢃ 104 cells/well in 100
mL experimental
media containing 5% FBS. The cells were then treated with PCAIs
(0e20 mM) for 24, 48 and 72 h. To determine cell viability, 20 mL of
m
incubation was stopped by adding 200
m
0.02% resazurin reagent prepared in Dulbecco PBS was added into
each well. This method is based on the fact that resazurin, a non-
toxic, water soluble, redox-sensitive dye, changes from its blue/
non-fluorescent state to a pink/highly-fluorescent product upon
reduction by viable cells. Fluorescence was measured at an
at ꢁ20 ꢀC for 5 min followed by centrifugation at 5000 ꢃ g for
5 min. The supernatants were analyzed by RP-HPLC on a Hamilton
PRP-1 column (5
detection at 260 nm. The mobile phase consisted of a linear
m
m particles, 4.1 mm ID ꢃ 50 mm) with UV

332
B.J. Aguilar et al. / European Journal of Medicinal Chemistry 81 (2014) 323e333
excitation wavelength of 560 nm and an emission wavelength of
590 nm using a Microplate Fluorescence Reader FLx 800 after 2 h.
The concentration of the inhibitor that inhibited 50% of the viability
of each cell line (EC50) was determined for each inhibitor using the
non-linear regression curve of concentration against response us-
ing the GraphPad Prism 5 software (San Diego, CA).
strong. The scoring was conducted by RD, FA, BJA and RP without
prior knowledge of the diagnosis of the individual samples on the
TMAs.
Author contributions
The manuscript was written through contributions of all au-
thors. All authors have given approval to the final version of the
manuscript.
4.12. Apoptosis assay
The mode of cell death induced by the PCAIs was determined by
the modified EB/AO [13] staining method [12,24]. 5 ꢃ104 MIA PaCa-
2 cells were cultured in 8 well Lab-Tek II chamber slide and treated
Acknowledgment
with 8e (0.5, 1, and 2
determined by staining cells with AO/EB (10
were visualized and captured by an Olympus fluorescent micro-
scope (ꢃ40 objective) using 480/30 nm excitation filter fitted with
an Olympus DP70 camera.
m
M) for 48 h and mechanisms of cell death
The research reported in this publication was supported by the
National Institute On Minority Health and Health Disparities
(NIMHD) of the National Institutes of Health (NIH) under Award
Number G12 MD007582 (previous award NIH/NCRR/RCMI G12
RR03020). The content is solely the responsibility of the authors
and does not necessarily represent the official views of the National
Institutes of Health.
mg/mL) and cell images
4.13. Docking analysis of PCAIs
Docking analysis of the PCAIs was conducted as previously
described [8,9] with an increased potential binding area of
20 ꢃ 20 ꢃ 24 points. Poses were selected based on the interaction
energy and the plausibility of the interactions between the PCAIs
and the active site residues. For example, poses with the pyrrolidine
and N-methylpiperazine ring inside the active site were deemed
implausible because these moieties would be ionized at physio-
logical pH and therefore be unlikely to be interacting with the
hydrophobic amino acids lining the active site. On the other hand,
poses with the polyisoprenyl tail outside of the active site would be
thermodynamically unlikely in the aqueous environment, espe-
cially given that the ionizable polar groups would then be in the
hydrophobic active site. Reported poses had (1) the farnesyl tail
located in the hydrophobic pocket of the active site, (2) the charged
ring structure located either outside or at the entrance of the
binding site and (3) the amide linkage preferably in close proximity
to the catalytic SER 221.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.05.018. These data include MOL
files and InChiKeys of the most important compounds described in
this article.
References
[1] R. Siegel, D. Naishadham, A. Jemal, Cancer statistics, 2012, CA Cancer J. Clin. 62
(2012) 10e29.
[2] M.J. Wieduwilt, M.M. Moasser, The epidermal growth factor receptor family:
biology driving targeted therapeutics, Cell. Mol. Life Sci. 65 (2008) 1566e
1584.
[3] S. Ueda, S. Ogata, H. Tsuda, N. Kawarabayashi, M. Kimura, Y. Sugiura, S. Tamai,
O. Matsubara, K. Hatsuse, H. Mochizuki, The correlation between cytoplasmic
overexpression of epidermal growth factor receptor and tumor aggressive-
ness: poor prognosis in patients with pancreatic ductal adenocarcinoma,
Pancreas 29 (2004) e1ee8.
[4] M. Macaluso, G. Russo, C. Cinti, V. Bazan, N. Gebbia, A. Russo, Ras family genes:
an interesting link between cell cycle and cancer, J. Cell Physiol. 192 (2002)
125e130.
4.14. Pancreatic cancer tissue microarrays
The PC tissue microarray (TMA) consisted of 196 duplicate cores
from 57 males and 41 females (Table 2). The 196 duplicate cores
included 12 mild chronic inflammation (6 cases), 4 chronic
inflammation (2 cases), 4 chronic pancreatitis (2 cases), 4 hyper-
plasia (2 cases), 84 duct adenocarcinoma (42 cases), 6 adenosqu-
amous carcinoma (3 cases), 2 islet cell carcinoma (1 case), 20 islet
cell tumor (10 cases), 40 cancer normal adjacent time (20 cases),
and 20 normal tissue (10 cases).
[5] M. Malumbres, M. Barbacid, RAS oncogenes: the first 30 years, Nat. Rev.
Cancer 3 (2003) 459e465.
[6] B.B. Friday, A.A. Adjei, K-ras as a target for cancer therapy, Biochim. Biophys.
Acta 1756 (2005) 127e144.
[7] N.S. Lamango, R. Duverna, W. Zhang, S.Y. Ablordeppey, Porcine liver carbox-
ylesterase requires polyisoprenylation for high affinity binding to cysteinyl
substrates, Open Enzym. Inhib. J. 2 (2009) 12e27.
[8] R. Duverna, S.Y. Ablordeppey, N.S. Lamango, Biochemical and docking analysis
of substrate interactions with polyisoprenylated methylated protein methyl
esterase, Curr. Cancer Drug Targets 10 (2010) 634e648.
[9] B. Aguilar, F. Amissah, R. Duverna, N.S. Lamango, Polyisoprenylation potenti-
ates the inhibition of polyisoprenylated methylated protein methyl esterase
and the cell degenerative effects of sulfonyl fluorides, Curr. Cancer Drug
Targets 11 (2011) 752e762.
[10] F. Amissah, S. Taylor, R. Duverna, L. Ayuk-Takem, N.S. Lamango, Regulation of
polyisoprenylated methylated protein methyl esterase by polyunsaturated
fatty acids and prostaglandins, Eur. J. Lipid Sci. Technol. 113 (2011) 1321e
1331.
4.15. Immunohistochemistry
The TMA and immunohistochemical staining were provided by
US Biomax (Rockville, MD). These were processed and analyzed as
previously described [25].
[11] E.L. Deer, J. Gonzalez-Hernandez, J.D. Coursen, J.E. Shea, J. Ngatia, C.L. Scaife,
M.A. Firpo, S.J. Mulvihill, Phenotype and genotype of pancreatic cancer cell
lines, Pancreas 39 (2010) 425e435.
4.16. Immunohistochemistry scoring
[12] L. Ayuk-Takem, F. Amissah, B.J. Aguilar, N.S. Lamango, Inhibition of poly-
isoprenylated methylated protein methyl esterase by synthetic musks induces
cell degeneration, Environ. Toxicol. 29 (2014) 466e477.
The method used to score the PMPMEase immunoreactivity was
adapted from that of [26]. The intensity of the staining was given
scores of 0 (no staining), 1 (trace), 2 (weak), 3 (intermediate), 4
(strong) and 5 (very strong). The score of the staining intensity was
then multiplied by the percentage of the immunoreactive tumor
cells. The overall scores ranged between 0 and 500 with those
between 1 and 100 described as trace, 101 to 200 as weak, 201 to
300 as intermediate, 301 to 400 as strong and 401 to 500 as very
[13] D. Ribble, N.B. Goldstein, D.A. Norris, Y.G. Shellman, A simple technique for
quantifying apoptosis in 96-well plates, BMC Biotechnol. 5 (2005) 12.
[14] I.M. Ahearn, K. Haigis, D. Bar-Sagi, M.R. Philips, Regulating the regulator: post-
translational modification of RAS, Nat. Rev. Mol. Cell Biol. 13 (2012) 39e51.
[15] L. Goldberg, R. Haklai, V. Bauer, A. Heiss, Y. Kloog, New derivatives of farne-
sylthiosalicylic acid (salirasib) for cancer treatment: farnesylthiosalicylamide
inhibits tumor growth in nude mice models, J. Med. Chem. 52 (2009) 197e
205.

B.J. Aguilar et al. / European Journal of Medicinal Chemistry 81 (2014) 323e333
333
[16] G. Elad-Sfadia, R. Haklai, E. Balan, Y. Kloog, Galectin-3 augments K-Ras acti-
vation and triggers a Ras signal that attenuates ERK but not phosphoinositide
3-kinase activity, J. Biol. Chem. 279 (2004) 34922e34930.
agents: a systematic review and meta-analysis of studies in advanced non-
small-cell lung cancer and metastatic colorectal cancer, Lancet Oncol.
9
(2008) 962e972.
[17] T. Tian, A. Harding, K. Inder, S. Plowman, R.G. Parton, J.F. Hancock, Plasma
membrane nanoswitches generate high-fidelity Ras signal transduction, Nat.
Cell Biol. 9 (2007) 905e914.
[18] M.L. Janmaat, J.A. Rodriguez, M. Gallegos-Ruiz, F.A. Kruyt, G. Giaccone,
Enhanced cytotoxicity induced by gefitinib and specific inhibitors of the Ras
or phosphatidyl inositol-3 kinase pathways in non-small cell lung cancer cells,
Int. J. Cancer 118 (2006) 209e214.
[19] C.Q. Zhu, G. da Cunha Santos, K. Ding, A. Sakurada, J.C. Cutz, N. Liu, T. Zhang,
P. Marrano, M. Whitehead, J.A. Squire, S. Kamel-Reid, L. Seymour,
F.A. Shepherd, M.S. Tsao, Role of KRAS and EGFR as biomarkers of response to
erlotinib in National Cancer Institute of Canada Clinical Trials Group Study
BR.21, J. Clin. Oncol. 26 (2008) 4268e4275.
[20] P. Laurent-Puig, J. Taieb, Lessons from Tarceva in pancreatic cancer: where are
we now, and how should future trials be designed in pancreatic cancer? Curr.
Opin. Oncol. 20 (2008) 454e458.
[22] J.J. Yeh, C.J. Der, Targeting signal transduction in pancreatic cancer treatment,
Expert Opin. Ther. Targets 11 (2007) 673e694.
[23] I. Cushman, S.M. Cushman, P.M. Potter, P.J. Casey, Control of RhoA methylation
by carboxylesterase I, J. Biol. Chem. 288 (2013) 19177e19183.
[24] F. Amissah, R.A. Poku, B.J. Aguilar, R. Duverna, B.O. Abonyo, N.S. Lamango,
Polyisoprenylated methylated protein methylesterase: the fulcrum of che-
mopreventive effects of NSAIDs and PUFAs, in: L.V. Berhardt (Ed.), Advances in
Medicine and Biology, Nova Biomedical, New York, 2013, pp. 119e144.
[25] F. Amissah, R. Duverna, B.J. Aguilar, R.A. Poku, N.S. Lamango, Poly-
isoprenylated methylated protein methyl esterase is both sensitive to cur-
cumin and overexpressed in colorectal cancer: implications for
chemoprevention and treatment, Biomed. Res. Int. 2013 (2013) 416534.
[26] R.M. Bremnes, R. Veve, E. Gabrielson, F.R. Hirsch, A. Baron, L. Bemis,
R.M. Gemmill, H.A. Drabkin, W.A. Franklin, High-throughput tissue microarray
analysis used to evaluate biology and prognostic significance of the E-cad-
herin pathway in non-small-cell lung cancer, J. Clin. Oncol. 20 (2002) 2417e
2428.
[21] H. Linardou, I.J. Dahabreh, D. Kanaloupiti, F. Siannis, D. Bafaloukos,
P. Kosmidis, C.A. Papadimitriou, S. Murray, Assessment of somatic k-RAS
mutations as
a mechanism associated with resistance to EGFR-targeted