Sulphonamide-based bombesin prodrug analogues for glutathione transferase, useful in targeted cancer chemotherapy

Article abstract of DOI:10.1016/j.ejmech.2008.10.009

Glutathione transferases (GSTs) are enzymes involved in cellular detoxification by catalysing the nucleophilic attack of glutathione (GSH) on the electrophilic centre of a number of toxic compounds and xenobiotics, including certain chemotherapeutic drugs

Full text of DOI:10.1016/j.ejmech.2008.10.009

European Journal of Medicinal Chemistry 44 (2009) 2009–2016
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Sulphonamide-based bombesin prodrug analogues for glutathione transferase,
useful in targeted cancer chemotherapy
,
I. Axarli a, N.E. Labrou a, C. Petrou b, N. Rassias c, P. Cordopatis b, Y.D. Clonis a
*
a Laboratory of Enzyme Technology, Department of Agricultural Biotechnology, Agricultural University of Athens, 75 Iera Odos Street, GR-11855 Athens, Greece
b Laboratory of Pharmacognosy and Chemistry of Natural Products, Department of Pharmacy, University of Patras, GR-26500 Patra, Greece
c RAFARM S.A., 12 Korinthou & Kapodistriou Street, GR-154 51 Athens, Greece
a r t i c l e i n f o
a b s t r a c t
Article history:
Glutathione transferases (GSTs) are enzymes involved in cellular detoxification by catalysing the
nucleophilic attack of glutathione (GSH) on the electrophilic centre of a number of toxic compounds and
xenobiotics, including certain chemotherapeutic drugs. The encountered chemotherapeutic resistant of
tumour cells, thus, has been associated with the increase of total GST expression. GSTs, in addition to
GSH-conjugating activity, exhibit sulphonamidase activity, catalyzing the GSH-mediated hydrolysis of
sulphonamide bonds. Such reactions are of interest as potential tumour-directed prodrug activation
strategies. In the present work we report the design and synthesis of novel chimaeric sulphonamide
derivatives of bombesin, able to be activated by the model human isoenzyme GSTA1-1 (hGSTA1-1). These
derivatives bear a peptidyl-moiety (analogues of bombesin peptide: R–[Lue13]-bombesin, R–[Phe13]-
bombesin and R–[Ser3,Arg10,Phe13]-bombesin, where R ¼ C6H5SO2NH–) as molecular recognition
element for targeting the drug selectively to tumour cells. The released S-alkyl-glutathione, after
hGSTA1-1-mediated cleavage of the sulphonamide bond, provides an inhibitor of varied strength against
GSTs from different sources. These prodrugs are envisaged as a plausible means to sensitize drug-
resistant tumours that overexpress GSTs.
Received 1 June 2008
Received in revised form
26 September 2008
Accepted 2 October 2008
Available online 17 October 2008
Keywords:
Alkylating agents
Antineoplastic therapy
Drug resistance
Detoxification
Enzyme-activated prodrug
Glutathione transferase
Ó 2008 Elsevier Masson SAS. All rights reserved.
1. Introduction
cytosolic GSTs are homodimers or heterodimers [8]. Each monomer
has two domains, an
helical domain comprised of helices
a GSH-binding site (G-site) on top of the large
hydrophobic pocket (H-site) lies between the two domains in
which a generally hydrophobic substrate binds and reacts with GSH
[8].
a
/
b
domain that includes
4– 9. The former contains
domain. A
a1–a3, and a large a-
Glutathione transferases (GSTs) are enzymes involved in cellular
detoxification by catalysing the nucleophilic attack of glutathione
(GSH) on the electrophilic centre of a number of toxic compounds
and xenobiotics, including certain chemotherapeutic drugs. The
GST superfamily can be subdivided into a number of classes on the
basis of their amino acid sequence [1]. Within mammals, the
following classes have been defined: alpha, mu, pi, sigma, theta,
zeta, kappa and omega [2]. In addition, a subfamily of chloride
intracellular channel proteins has been shown to be members of
the cytosolic GST structural family but have no known enzymatic
activity [3]. Several other soluble GST classes have been reported in
insects: delta, epsilon [4]; plants: phi, tau, lambda, dehy-
droascorbate reductase [5]; and bacteria: beta [6] and chi [7]. The
a
a
a
Cancer remains the second-leading cause of death in the
industrialized world and worldwide; nevertheless it continues to
be underserved by effective therapeutic agents [9]. Many of the
available agents act systemically and therefore have side effects
that range from uncomfortable to life threatening. Recently, prod-
ucts have begun to emerge in this market that are specifically tar-
geted to cancer cells or act in collaboration with the body’s immune
response to combat the disease. This marks a dynamic change in
the way cancer is treated, and such innovative therapies will
transform the cancer market during the next decade [10].
Although GSTs’ detoxifying ability protects cell from certain
diseases, unfortunately it also reduces the effectiveness of certain
chemotherapeutic drugs against cancer cells. Indeed, one of the
classes of electrophilic compounds that are substrates for the GSTs
are certain alkylating agents used in antineoplastic therapy [11]. A
common problem encountered in cancer chemotherapy is the
Abbreviations: Boc, t-butoxycarbonyl group; But, t-butyl group; CDNB, 1-chloro-
2,4-dinitrobenzene; DIC, diisopropylcarbodiimide; DMF, dimethylformamide;
Fmoc, 9-fluorenylmethoxycarbonyl; G-site, glutathione binding site; GSH, gluta-
thione; GST, glutathione transferase; H-site, hydrophobic binding site; HOBt,
1-hydroxybenzotriazole; IPTG, isopropyl-thiogalactoside; Pbf, 2,2,4,6,7-pentam-
ethyldihydrobenzofuran-5-sulfonyl; TFA, trifluoroacetic acid; Trt, trityl group.
* Corresponding author. Tel.: þ30 (210) 5294311; fax: þ30 (210) 5294307.
E-mail address: clonis@aua.gr (Y.D. Clonis).
0223-5234/$ – see front matter Ó 2008 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2008.10.009

2010
I. Axarli et al. / European Journal of Medicinal Chemistry 44 (2009) 2009–2016
appearance of chemotherapeutic resistant tumour cells that no
longer respond appropriately to the antineoplastic agents. This
phenomenon, referred to as multi-drug resistance, has complicated
attempts towards cancer therapy [12]. A possible origin for the
problem appears to be an increase in the expression of total GST
activity in tumour cells [13,14]. A plausible mechanism by which
GSTs could contribute to drug resistance includes GST-dependent
prevention of drug-induced apoptosis via direct interaction with
signal transduction proteins, as suggested for GSTP1-1 [15,16]
which inhibits c-Jun N-terminal kinase. It has been demonstrated
that hGSTA1/A2 protein was increased in blast cells (derived from
acute myeloid leukemia patients) showing resistance to doxoru-
bicin in vitro [13], and a weak correlation was observed between
GST alpha in gastric cancer tissues and cisplatin resistance (in vitro)
[14]. However, more recently, it was shown that homozygous
hGSTA1*B breast cancer patients treated with cyclophosphamide
(plus other chemotherapeutic drugs) had a reduced death hazard
during the first 5 years following diagnosis compared with homo-
zygous hGSTA1*A individuals (hazard ratio, 0.3) [15]. This obser-
vation was attributed to the detoxifying role of hepatic hGSTA1
against therapeutic metabolites of cyclophosphamide.
The present work proposes a prodrug-design approach based on
the development of novel chimaeric synthetic sulphonamide-
derivatives, susceptible to activation by the human isoenzyme
GSTA1-1 (hGSTA1-1). These chimaeric prodrugs feature a peptidyl-
moiety (bombesin peptide analogues) that is specifically recognised
by a tumour cell specific receptor (bombesin receptor), to provide
a potential vehicle for selective drug delivery to cancer cells. Fol-
lowed by GST-mediated cleavage of the sulfonamide bond, the
prodrug releases a potent inhibitor for GSTs. During the past decade,
extensive knowledge has been accumulated on the involvement of
bombesin peptide-analogues in the mitogenesis of various tumour
cells, including small cell lung carcinoma (SCLC), cancers of the
gastrointestinal tract, such as pancreatic and colon cancer, as well as
breast cancer [17]. The putative role of bombesin-like peptides as
autocrine growth factors for these tumours [18] prompted us to
design and synthesis chimaeric bombesin-analogues which may
prove useful in the combat of certain cancers [18].
a total volume of 50
plasmid DNA, 0.2 mM of each dNTP, 5
m
l contained 6 pmol of each primer, 10 ng
m
l 10 ꢀ Pfu buffer and 2 units
of Pfu DNA polymerase. The PCR procedure comprised 30 cycles of
2 min at 95 ꢁC, 2 min at 55 ꢁC and 2 min at 72 ꢁC. A final extension
time at 72 ꢁC for 10 min was performed after the 30 cycles. The
resulting PCR amplicon was TOPO ligated into a T7 expression
vector (pET101/D-TOPOÒ). The resulting expression construct
pT7hGSTA1-1 was sequenced along both strands and was used to
transform competent BL21(DE3) Escherichia coli cells. E. coli cells,
harbouring plasmid pT7hGSTA1-1, were grown at 37 ꢁC in 1 L LB
medium containing 100 mg/mL ampicillin. The synthesis of
hGSTA1-1 was induced by the addition of 1 mM IPTG when the
absorbance at 600 nm was 0.6–0.8. Five hours after induction, cells
were harvested by centrifugation at 8000 r.p.m. and 4 ꢁC for
20 min, re-suspended in sodium phosphate buffer (5 mM, pH 7.7),
sonicated, and centrifuged at 10,000 g for 20 min. The supernatant
was collected and was applied to a column of S-hexyl-Sepharose
column (2 ml, 1.5 ꢀ1.5 cm I.D.) previously equilibrated with 5 mM
sodium phosphate buffer, pH 7.7. Non-adsorbed protein was
washed off with 100 ml equilibration buffer. Bound hGSTA1-1 was
eluted with potassium phosphate buffer (50 mM, pH 8.0, contain-
ing 10 mM GSH). The eluted fractions were dialysed against 50 mM
potassium phosphate buffer, pH 7.5 and stored at 4 ꢁC. For long
term storage the enzyme solution was stored at ꢂ20 ꢁC in glycerol/
0.1 M potassium phosphate buffer pH 7.0, 50/50 (v/v). Expression
and purification of Zea mays GST I were performed according to
Ref. [19]. Human spleen haematopoietic prostaglandin D synthase
was purified as described in Ref. [20], whereas soybean GSTU4-4
was expressed and purified as described in Ref. [21].
2.2.2. Electrophoresis
Protein purity was judged by SDS polyacrylamide gel electro-
phoresis using 12.5% (w/v) polyacrylamide (running gel) and 2.5%
(w/v) stacking gel, according to the methods of Laemmli, 1970 [24].
The protein bands were stained with Coomassie Brilliant Blue
R-250.
2.2.3. Synthesis, purification and quality assessment of
sulphonamide bombesin-analogues (prodrugs)
2. Materials and methods
Sulfonamide bombesin-analogues were synthesized by Fmoc
solid phase methodology [22] utilizing Rink Amide MBHA resin
[23] as the solid support. Fmoc-protected amino acids were used
with the t-butyl group (But) as side-chain protection group for Glu,
t-butoxycarbonyl group (Boc) for Trp, trityl (Trt) group for Asn, His,
Gln and 2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl (Pbf)
group for Arg. Stepwise synthesis of the peptide analogues was
achieved with diisopropylcarbodiimide/1-hydroxybenzotriazole
(DIC/HOBt) in dimethylformamide (DMF) as coupling agents [24].
Couplings were performed with Fmoc-amino acid, DIC and HOBt in
DMF in a 3.0, 3.3 and 4.5 molar excess, respectively, for 2.5 h at
room temperature. Completeness of the reaction was monitored by
the Kaiser test [25], and the chloranil test [26]. The Fmoc groups
were removed by treatment with 20% piperidine in DMF for 40 min.
In order to finally synthesis the sulphonamide bombesin-
analogues (prodrugs), the pGlu residue from the N-terminal of the
native bombesin sequence was replaced for a Glu which has a free
amino group available for coupling with another group (e.g. ben-
zylsulfonyl chloride). Coupling of benzenesulfonyl group with the
free aminoterminal, and formation of the respective sulfonamide
bond, was achieved using N-methylmorpholine. Benzenesulfonyl
chloride was used in 3 molar excess and N-methylmorpholine in 6
molar excess. The pH of the reaction was monitored and adjusted in
the range 10–11 with addition of N-methylmorpholine. Coupling of
the benzenesulfonyl-group was completed within 3.5 h. After
completion of the synthesis, the resin was treated with TFA solution
(15 ml/g peptide resin) in the presence of scavengers
2.1. Materials
Reduced glutathione, 1-chloro-2,4-dinitrobenzene (CDNB) and
human GSTP1-1 isoenzyme were obtained from Sigma–Aldrich Co
(USA). Molecular biology reagents, kits and enzymes were obtained
from Invitrogen (USA). Other reagents and analytical grade
chemicals were obtained from Sigma–Aldrich Co (USA). The 9-
fluorenylmethoxycarbonyl (Fmoc)-protected amino acids, Rink
Amide MBHA resin and peptide reagents were purchased from CBL
(Patras, Greece), Bachem (Bubendorf, Switzerland) and Nova-
biochem (La¨ufelfingen, Switzerland). Benzenesulfonyl chloride
(C6H5–SO2–Cl) was obtained from Acros Organics (Geel, Belgium).
All solvents and reagents used for solid phase synthesis were of
analytical quality and used without further purification.
2.2. Methods
2.2.1. Cloning, expression and purification of GSTs
PCR was used to amplify the full-length ORF of hGSTA1-1 from
pKKGTB vector (a much appreciated gift from Prof. W.M. Atkins,
Department of Medicinal Chemistry, University of Washington)
using the oligo primers synthesized to the 50 region of the gene (50-
CACCATGGCAGAGAAGCCCAAGCTCCAC-30) and to the 30 end of the
gene finishing at the TAA stop codon (50-TTAAAACCT-
GAAAATCTTCCTTGCTTC-30). The PCR reaction was carried out in

I. Axarli et al. / European Journal of Medicinal Chemistry 44 (2009) 2009–2016
2011
(TFA/1,2-ethanedithiol/triethylsilane/water/anisole, 95/1/1/1/2 v/v/
v/v/v) for 4 h to liberate the fully deprotected crude peptide
conjugate. The released peptide prodrug conjugate was precipi-
tated upon solvent concentration and addition of cold ether. The
precipitate was collected by filtration, washed twice with cold
ether, dried in vacuum over KOH and purified by gel filtration
chromatography on Sephadex G-15 using 20% acetic acid as the
eluent. Final purification was achieved by preparative high
the inhibitor and adding hydrogen atoms to fulfill unsatisfied
valencies. The sulphanilamide (Fig. 2B) was used as a ligand. Ligand
pdb file was generated using JME Molecular Editor software [30]
and energy minimized. Hydrogen atoms were added to the ligand
coordinate file prior to docking. The ArgusLab software, version 4.0
[31,32] was employed for predicting the geometry of the ligand
bound to the protein. ArgusLab makes use of either AScore or the
Lamarckian genetic algorithm [33] scoring functions to find the
¨
performance liquid chromatography (HPLC, Mod.10 AKTA, Amer-
low-energy binding modes. A distance-dependent dielectric
sham Biosciences, Piscataway, USA) on a Lichrosorb RP18 column
(
e
¼ 4r) and an energy cut-off distance of 10 Å were used in eval-
(C18 solid phase, 7
mm particle size, 250 mM ꢀ 8 mm) applying
uating the interaction energy between protein and ligand. Ligand
conformational flexibility was explored through torsion angle
sampling and minimization. PyMOL [34] was used for the
preparation of structure figures. The type of interactions in the
complex hGSTA1-1-sulphanilamide were analysed using iMolTalk
(http://i.moltalk.org/).
a linear gradient 10–70% acetonitrile (0.1% TFA) for 35 min, and
70–100% acetonitrile (0.1% TFA) for 5 min (flow rate 1.5 ml/min, UV
detection at 220 nm and 254 nm). The appropriate fractions were
pooled and lyophilized. Synthesis and purification of benzylsul-
fonyl-GSH was carried out by substituting the fluoride atom of
benzylsulfonyl fluoride by reduced GSH according to the procedure
of Katusz et al. [27]. All prodrug products were analysed by thin
layer chromatography (TLC, Merck pre-coated silica gel plates, type
G60-F254; solvent system 1-butanol/acetic acid/water/pyridine (4/1/
1/2 v/v/v/v)), before were further checked for their purity by
3. Results and discussion
3.1. Sulphonamidase activity of GSTs
¨
analytical HPLC (AKTA Purifier) using a Nucleosil 100 C18 column
(5
Glutathione transferases (GSTs) catalyze the nucleophilic attack
m
m particle size; 250 ꢀ 4.6 mm), and Electron Spray Ionization-
of glutathione on the electrophilic centre of a number of
Mass Spectrometry (ESI-MS) on
instrument.
a
Micromass-Platform LC
compounds. GSTs exhibit wide substrate specificity and act in vitro
and in vivo on several xenobiotics [35]. In addition to GSH conju-
gating activity, GSTs exhibit sulphonamidase activity and catalyze
the GSH-mediated hydrolysis of sulphonamide bonds to form the
corresponding amine (Fig. 1) [36,37].
2.2.4. Assay of enzyme activity and protein
GST assays were performed by monitoring the formation of the
conjugate of CDNB (1 mM) and GSH (2.5 mM) at 340 nm
To analyze this sulphonamidase activity, a collection of different
GST isoenzymes that belong to different classes were investigated
using sulphanilamide (4-aminobenzenesulfonamide) as a model
substrate. In particular, GSTs belonging to class alpha [human
GSTA1-1, (hGSTA1-1); mouse GSTA4-4, (mGSTA4-4)], pi [human
GSTP1-1, (hGSTP1-1)], sigma [Schistosoma japonicum GST, (SjGST);
human spleen haematopoietic prostaglandin D synthase, (PDS-
GST)], tau [Glycine max GST, (GmGSTU4-4)] and phi [Zea mays GST I
(ZmGSTI)] were examined and the results are presented in Table 1.
All but tau and phi class isoenzymes (tau and phi are plant specific
classes) show sulphonamidase activity towards sulphanilamide.
The isoforms examined exhibited different rates of sulphanilamide
cleavage, demonstrating isoenzyme selectivity. These variations are
consistent with the fact that GST isoenzymes show marked differ-
ences in their abilities to catalyze reactions between GSH and
various electrophiles. The notable sulphonamidase activity
observed for hGSTA1-1 is of particular importance since this
isoenzyme is expressed in several tumours and therefore may
represent a therapeutic molecular target in cases where tumour-
protective effects depend upon hGSTA1-1 activity. It is interesting
to note that alpha class GSTs have hardly been explored with
respect to chemotherapy response, even though a number of
alkylating chemotherapeutic agents in current use are known to be
substrates (e.g. busulphan, thiotepa, and the therapeutic metabo-
lites of cyclophosphamide) [38,39].
(
e
¼ 9.6 mMꢂ1 cmꢂ1) at 30 ꢁC according to a published method [19–
21]. Observed reaction velocities were corrected for spontaneous
reaction rates when necessary. All initial velocities were deter-
mined in triplicate in buffers at constant temperature. One unit of
enzyme is defined as the amount of enzyme that gives 1.0 mmole of
product per minute at pH 6.5 at 30 ꢁC. Determination of sulpha-
nilamide cleavage was carried out by measuring the released NH3,
using the Nessler’s reagent. Determination of sulfonamide bomb-
esin analogues cleavage by hGSTA1-1 was based on the detection of
the non-conjugated substrate GSH using Elman’s reagent (2-nitro-
5mercaptobenzoic acid) at 412 nm. Assays were carried out at 25 ꢁC
for 2–16 h in 20 mM potassium phosphate buffer, pH 7, in the
presence of 2.88 mM GSH and 2.2 mM peptide. Control incubations
in the absence of GSH, peptide or enzyme were also carried out to
correct for background rates of spontaneous chemical cleavage.
Specific activities were determined as
DA/h at 412 nm per mg
protein. Protein concentration was determined by the method of
Bradford [28] using bovine serum albumin (fraction V) as standard.
2.2.5. Kinetic inhibition studies with benzylsulfonyl-GSH
Initial velocities for the hGSTA1-1-catalysed reaction with GSH
as variable substrate were measured at 30 ꢁC in a total volume of
1 ml total mixture containing 100 mM potassium phosphate buffer,
pH 6.5, 1 mM CDNB, and different concentrations of GSH in the
presence or in the absence of 5 or 10
mM benzylsulfonyl-GSH. With
Although all mammalian cytosolic GSTs share a highly
CDNB as variable substrate, the reaction mixture contained in
a total volume of 1 ml: 100 mM potassium phosphate buffer, pH
6.5, 1 mM GSH and different concentrations of CDNB, again in the
presence or in the absence of 2 or 10 mM benzyl sulfonyl-GSH. The
apparent kinetic parameters were determined using the GraFit
(Erithacus Software, Ltd., UK) computer program.
conserved GSH-binding site within their canonical fold, the elec-
trophilic substrate binding site (H-site), varies significantly among
different isoforms [2,8], and it is reasonably expected to provide
a source of isoform specificity among substrates and inhibitors.
Therefore, molecular docking studies were carried out to provide in
silico structural information and help to locate the sulphanilamide
ligand binding site on hGSTA1-1. The model for the binding of
sulphanilamide lead-ligand (NH2SO2Benz) to hGSTA1-1 was con-
structed based on: (i) the proposed role of the active site Tyr7 in
catalysis, (ii) the fact that the GSH sulphur attacks carbon-4 of
sulphanilamide in the reaction and (iii) manual fitting to obtain
optimum interactions between the sulphanilamide’s hydrophobic
2.2.6. Docking of the sulphanilamide lead-ligand into the hGSTA1-1
binding site
The high-resolution structure of hGSTA1-1 in complex with the
inhibitor ethacrynic acid (PDB identifier 1GSF) [29] was used in the
docking studies. The protein was prepared for docking by removing

2012
I. Axarli et al. / European Journal of Medicinal Chemistry 44 (2009) 2009–2016
Fig. 1. A. Sulphonamides cleavage by GSTs. B: The general structure of the proposed chimaeric prodrug (e.g. R1 ¼ bombesin, R2 ¼ Benz). C: Interaction of sulphanilamide lead-ligand
with hGSTA1-1 from docking calculations. Side-chains of specific residues contributing to sulphanilamide binding are presented as heavy stick beneath the transparent surface and
labelled. The figure was prepared using PyMol [34].
chain and the hydrophobic residues of the H-site. The predicted
mode of interaction of the sulphanilamide with hGSTA1-1 is shown
in Fig. 1C. The binding of sulphanilamide to hGSTA1-1 may
primarily be achieved by hydrophobic interactions that provide the
driving force for ligand positioning and recognition. The bulk of the
interactions with the enzyme involve mainly hydrophobic residues
from the N-terminal domain (Tyr9, Phe10, Val55), and hydrophobic
residues from the C-terminal domain Met208, Phe220 and Phe222
(Fig. 5).
3.2. Design of sulphonamide prodrugs
The sulphonamidase activity of GSTs suggests that prodrugs
activated by GSTs might provide a strategy for targeted cancer
chemotherapy. This concept was realized by designing chimaeric
sulphonamide derivatives (bombesin analogues, Fig. 2) able to
undergo cleavage in tumour cells by intracellular GSTs where
certain GST isoenzymes, such as hGSTA1-1, are highly expressed
[11,13,14]. Selective targeting of these prodrugs to cancer tissues
might be made possible if they bore a chemical moiety (e.g.
bombesin analogues) to serve as a ligand for a cancer tissue-specific
receptor. Widely variable structures are apparently accommodated
by the H-site of GST. For example, sulphonamide derivatized
aliphatic or aromatic primary and secondary amines, amides,
azides, ureas, and hydrazines have been shown to be substrates for
GSTs [37]. Zhao et al. (1999) have shown that GST-mediated sul-
phonamide cleavage is relatively independent of the nature of the
amine derivatized and any amine, in principle, can be derivatized
by linkage with sulphonyl moieties, to confer a high or low degree
of intracellular liability [36]. This possibility is very significant
because it suggests that complex peptides having free amino
groups can be transformed to sulphonamides (prodrugs) which are
then activated intracellularly upon cleavage by GSTs. The structural
element that would target the prodrug selectively to tumour cells
may be bombesin peptide analogues which are selectively recog-
nised by tumour cell surface bombesin receptors [17,40,41]. The
released S-alkyl-glutathionyl-analogue (GSꢂR, Fig. 2B) may provide
a strong inhibitor against intracellular GSTs, thus facilitating cancer
chemotherapy. In this respect, it is likely that the observed strong
Table 1
Sulphonamidase activity of different GST isoenzymes using sulphanilamide as
substrate. The isoenzymes used were: human GSTA1-1, (hGSTA1-1); human GSTP1-
1, (hGSTP1-1); mouse GSTA4-4, (mGSTA4-4); Schistosoma japonicum GST, (SjGST);
Glycine max GSTU4-4, (GmGSTU4-4); Zea mays GST I, (ZmGST I); and human spleen
haematopoietic prostaglandin D synthase, (PDS-GST). The enzymatic reactions were
corrected for the nonenzymatic hydrolysis. Reported data are the average of three
separate experiments.
GSTs
Enzyme activity (DA/h mg)
SjGST
39.7
35.29
500.0
245.6
12.9
nda
hGSTA1-1
mGSTA4-4
PDS-GST
hGSTP1-1
GmGSTU4-4
ZmGSTI
nd
a
No detectable.

I. Axarli et al. / European Journal of Medicinal Chemistry 44 (2009) 2009–2016
2013
Fig. 2. A: The general structure of the sulphonamide bombesin analogues (prodrugs). B: Sulphonamide cleavage by hGSTA1-1 using GSH as a nucleophile. C: The primary structure
of bombesin and its sulphonamide-analogues, R ¼ C6H5SO2–.
inhibition of S-alkyl-glutathionyl-analogues against GSTs is asso-
ciated with a ‘product inhibition’ mode of action [42]. Product
inhibition has been observed with many GS–R conjugates,
including, for example, GS–estradiol and GS–aflatoxin conjugates
[43,44]. Conjugates such as S-hexyl-GSH and S-benzyl-GSH are
commonly used in vitro as biochemical probes and inhibitors.
Among them S-benzyl-GSH is usually considered as the strongest
inhibitor.
shown to exhibit antitumour effects on several cell lines such as SW-
1990 human pancreatic cancers, nitrosamine-induced pancreatic
cancers in hamsters, H69 human SCLC, MKN45 and Hs746T human
gastric cancers, HT-29 human colon cancers, PC-82, PC-3, and DU-
145 human prostate cancers, androgen independent Dunning R-
3327-AT-1 rat prostate cancers, estrogen dependent and indepen-
dent MXT mouse mammary cancers, MCF-7 MIII human breast
cancer, and U-87MG and U-373MG human glioblastomas [17,18].
On the basis of the above, bombesin analogues, coupled via
a sulphonamide linker to a benzyl group, may provide vehicles for
selective delivery of potent GST inhibitors (GS–R) to cancer cells. This
approach may be proved effective in overcoming chemotherapeutic
drug resistance. In addition, the bombesin analogues have been
3.3. Sulphonamide prodrug synthesis
Three different bombesin analogues [Leu13]-bombesin, [Phe13]-
bombesin and [Ser3,Arg10,Phe13]-bombesin (Fig. 2C) were synthe-
sized by Fmoc solid phase methodology [22] utilizing Rink Amide
MBHA resin [23] as the solid support. The amino acid substitutions
were decided on the basis of their differential potency against the
bombesin receptor [17,40,41]. In order to achieve prodrug forma-
tion, the pGlu residue from the N-terminal of the native bombesin
sequence was replaced for Glu which has a free amino group
available for coupling with another group. Coupling of benzene-
sulfonyl group with the free aminoterminal and formation of the
sulfonamide bond, was achieved using N-methylmorpholine. Other
Table 2
Physicochemical properties of the new sulphonamide bombesin analogues.
Analogue
MW
[M þ 1] Yield (%) TLC Rf1 HPLC tR (min)
R–[Leu13]-bombesin
R–[Phe13]-bombesin
1778.20 1779.8 78
1812.04 1813.1 79
0.30
0.32
0.35
17.05R–
17.32
16.17
R–[Ser3,Arg10,Phe13]-bombesin 1799.99 1801.8 80
organic
bases
such
as
triethylamine,
collidine
and
35
30
25
20
15
10
5
0
Fig. 3. SDS–polyacrylamide gel electrophoresis of hGSTA1-1 preparations. Protein
bands were stained with Coomassie Brilliant Blue R-250. Lane A, Escherichia coli crude
extract after induction with 1 mM IPTG; Lane B, eluted fraction from S-hexyl-GSH-
Sepharose.
Fig. 4. Schematic graph of the hGSTA1-1 sulphonamidase activity towards the
synthetic sulphonamide bombesin analogues (prodrugs). Control incubations in the
absence of GSH, peptide or enzyme were carried out to correct for background rates of
spontaneous chemical cleavage.

2014
I. Axarli et al. / European Journal of Medicinal Chemistry 44 (2009) 2009–2016
Fig. 5. Analysis of putative binding site of hGSTP1-1. A: The structure of analogue [Ser3,Arg10,Phe13]-bombesin (sulphonamide bombesin analogue No3). B: Molecular surfaces
around the enzyme’s binding site were created and coloured by electrostatic potential (red denotes a negative potential, white a neutral potential and blue a positive potential). (For
interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
N,N-diisopropylethylamine were tested for the coupling reaction,
but N-methylmorpholine was found to be most effective. The
peptide prodrug conjugates finally released from the solid phase
were precipitated and purified by GFC (Sephadex G-15, 20% acetic
acid liquid phase) followed by preparative HPLC (Lichrosorb RP18).
Analysis of prodrugs by TLC resulted in single spots, whereas that
by HPLC produced single peaks corresponding to at least 97% of the
total peptide peak integrals column. Electron Spray Ionization-
Mass Spectrometry (ESI-MS) data were consistent with the
expected peptide formulae (Fig. 2C). The physicochemical proper-
ties of the new analogues are summarised in Table 2.
the pKKGTB expression plasmid. The recombinant enzyme was
purified in a single-step procedure, using affinity chromatography
on immobilized S-hexyl-GSH affinity column (Fig. 3).
The three different bombesin analogues (1: [Leu13]-bombesin,
2: [Phe13]-bombesin and 3: [Ser3,Arg10,Phe13]-bombesin) were
evaluated as potential hGSTA1-1 substrates. Based on previous
studies [36,37], the reaction of GST-mediated sulfonamide cleavage
occurs as in the scheme shown in Fig. 2B, whereas the catalytic
mechanism may be described as follows: binding of GSH at the G-
site of hGSTA1-1 results in formation of the thiolate anion GSꢂ. This
nucleophile then attacks the (aromatic) carbon atom linked to the
sulphonyl moiety of the peptide substrate, which is bound at the H-
site of GST. Sulphonamide cleavage results in the formation of the
GS-conjugate, the corresponding amine (peptide), and sulphur
dioxide. Release of sulphur dioxide and the amine is either
concerted or results from subsequent hydrolysis of an intermediate,
initially formed sulphur dioxide-amine adduct. All sulphonamide
bombesin analogues were cleavaged by the recombinant enzyme
(Fig. 4). Among them, [Ser3,Arg10,Phe13]-bombesin served as
a better substrate for hGSTA1-1. This may be due to the presence of
a positively charged Arg5 in the analogue, establishing favourable
electrostatic interactions with the negative electrostatic potential
observed for the enzyme (Fig. 5).
3.4. Sulphonamide prodrug cleavage by hGSTA1-1
Before kinetic analysis of the prodrug cleavage reaction, it was
necessary to obtain large quantities of hGSTA1-1 isoenzyme. For
achieving high expression level of hGSTA1-1 the full-length gene of
the enzyme was subcloned into a T7-based expression vector
(pET101/D-TOPOÒ). The resulting recombinant plasmid was used to
transform the expression host E. coli BL21(DE3). Cell-free extract of
the E. coli transformants showed high GST activity with a specific
activity of 1.6 U/mg protein (Fig. 3), which is approximately 115-
fold higher compared to the original expression system based on

I. Axarli et al. / European Journal of Medicinal Chemistry 44 (2009) 2009–2016
2015
4
3
2
1
0
and a phosphorodiamidate, which in turn spontaneously forms
aziridinium species, the actual alkylating moieties [49]. More
recently Saavedra et al., (2004) have designed PABA/NO, a NO-
releasing GST-activated prodrug. PABA/NO, after hGSTP1-1-cata-
lyzed conjugation to GSH, releases a diazeniumdiolate ion, with
subsequent release of nitric oxide [50]. Encouraging results in both
cell and animal models of cancer have suggested that this prodrug
exhibits improved cytotoxic selectivity toward cancer cells, most
likely due to the high levels of GST in these cells. The above
approach exemplifies the one described in the present study,
nevertheless it lacks the second feature, namely, a structural
element that would target the drug selectively to tumor cells.
Acknowledgement
0
2
4
6
8
This work was partially supported by the Hellenic General
Secretariat for Research and Technology (programme: Operational
Programme for Competitiveness, grant No YB45).
1/[GSH] (1/mM)
6
4
References
[1] B. Mannervik, U.H. Danielson, CRC Crit. Rev. Biochem. 23 (1988) 283–337.
[2] C. Frova, Biomol. Eng. 23 (2006) 149–169.
[3] A. Dulhunty, P. Gage, S. Curtis, G. Chelvanayagam, P. Board, J. Biol. Chem. 276
(2001) 3319–3323.
[4] Z. Alias, A.G. Clark, Proteomics 7 (2007) 3618–3628.
[5] D.P. Dixon, B.G. Davis, R. Edwards, J. Biol. Chem. 277 (2002) 30859–30869.
[6] N. Allocati, B. Favaloro, M. Masulli, M.F. Alexeyev, C. Di Ilio, Biochem. J. 373
(2002) 305–311.
2
[7] E. Wiktelius, G. Stenberg, Biochem. J. 406 (2007) 115–123.
[8] A.J. Oakley, Curr. Opin. Struct. Biol. 15 (2005) 716–723.
[9] J.C. Cusack Jr., K.K. Tanabe, Surg. Oncol. Clin. N. Am. 11 (2002) 497–519.
[10] A. Sehgal, Curr. Opin. Drug Discov. Devel 5 (2002) 245–250.
[11] J.D. Hayes, D.J. Pulford, Crit. Rev. Biochem. Mol. Biol. 30 (1995) 445–600.
[12] G. Batist, A. Tulpule, B.K. Sinha, A.G. Katki, C.E. Myers, K.H. Cowan, J. Biol.
Chem. 261 (1986) 15544–15549.
0
-2
0
20
40
[13] J.M. Sargent, C. Williamson, A.G. Hall, A.W. Elgie, C.G. Taylor, Exp. Med. Biol.
457 (1999) 205–209.
1/[CDNB] (1/mM)
[14] Y. Kodera, K. Isobe, M. Yamauchi, K. Kondo, S. Akiyama, K. Ito, I. Nakashima,
H. Takagi, Cancer. Chemother. Pharmacol. 34 (1994) 203–208.
[15] V. Adler, Z. Yin, S.Y. Fuchs, M. Benezra, L. Rosario, K.D. Tew, M.R. Pincus,
M. Sardana, C.J. Henderson, C.R. Wolf, R.J. Davis, Z. Ronai, EMBO J. 18 (1999)
1321–1334.
[16] C. Sweeney, B.F. Coles, S. Nowell, N.P. Lang, F.F. Kadlubar, Toxicology 181–182
(1988) 83–87.
[17] J.C. Reubi, S. Wenger, J. Schmuckli-Maurer, J.C. Schaer, M. Gugger, Clin. Cancer
Res. 8 (2002) 1139–1146.
Fig. 6. Kinetic inhibition studies. A: Lineweaver–Burk plots for the inhibition of
hGSTA1-1 by benzylsulfonyl-GSH at different GSH concentrations. Enzyme assayed in
the absence (B) or in the presence of benzylsulfonyl-GSH (mM): 5 (C); 15 (,). B:
Inhibition of hGSTA1-1 by benzylsulfonyl-GSH at different CDNB concentrations.
Enzyme assayed in the absence (B) or in the presence of benzylsulfonyl-GSH (mM): 5
(C); 10 (,).
[18] A. Nagy, P. Armatis, R.Z. Cai, K. Szepeshazi, G. Halmos, A.V. Schally, Proc. Natl.
Acad. Sci. USA 94 (1997) 652–656.
[19] N.E. Labrou, L.V. Mello, Y.D. Clonis, Eur. J. Biochem. 268 (2001) 3950–3957.
[20] D.J. Meyer, M. Thomas, Biochem. J. 311 (1995) 739–742.
[21] B. McGonigle, S.J. Keeler, S.M. Lau, M.K. Koeppe, D.P. O’Keefe, Plant Physiol. 124
(2000) 1105–1120.
[22] B.G. Fields, L.R. Noble, Int. J. Pept. Prot. Res. 35 (1990) 161–214.
[23] S. Aventis, European Patent 3,22,348 and US Patent 5,124,478.
[24] W. Ko¨ning, R. Geiger, Chem. Ber. 103 (1970) 788–798.
[25] E. Kaiser, R.L. Colescott, C.D. Bossinger, P.I. Cook, Anal. Biochem. 34 (1970)
595–598.
[26] T. Vojkovsky, Detection of secondary amines on solid phase, Pept. Res. 8 (1995)
236–237.
[27] R.M. Katusz, B. Bono, R.F. Colman, Biochemistry 30 (1991) 11230–11238.
[28] M.M. Bradford, Anal. Biochem 72 (1976) 248–258.
[29] A.D. Cameron, I. Sinning, G. L‘Hermite, B. Olin, P.G. Board, B. Mannervik,
T.A. Jones, Structure 3 (1995) 717–727.
[30] ChemDraw Software [Available from: http://www.cambridgesoft.com/
software/ChemDraw/].
[31] Thompson M.A. ArgusLab 4.0.1. [Available from: <http://www.arguslab.
com>]. Planaria Software LLC, Seattle, WA.
[32] A. Erve, Y. Saoudi, S. Thirot, C. Guetta-Landras, J.C. Florent, C.H. Nguyen,
D.S. Grierson, A.V. Popov, Nucleic Acids Res. 34 (2006) (e43).
[33] G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew,
A.J. Olson, J. Comput. Chem. 19 (1998) 1639–1662.
[34] W.L. DeLano, The PyMOL Molecular Graphics System, DeLano Scientific, San
Carlos, USA, 2002.
[35] S. Mukanganyama, M. Widersten, Y.S. Naik, B. Mannervik, J.A. Hasler, Int. J.
Cancer 97 (2002) 700–705.
[36] Z. Zhao, K.A. Koeplinger, T. Peterson, R.A. Conradi, P.S. Burton, A. Suarato,
R.L. Heinrikson, A.G. Tomasselli, Drug Metab. Dispos. 27 (1999) 992–998.
[37] K.A. Koeplinger, Z. Zhao, T. Peterson, J.W. Leone, F.S. Schwende,
R.L. Heinrikson, A.G. Tomasselli, Drug Metab. Dispos. 27 (1999) 986–991.
Kinetic inhibition studies were carried out to evaluate the
inhibition potency of benzylsulphonyl-GSH. The inhibition patterns
are illustrated in Fig. 6. Benzylsulphonyl-GSH exhibited a competi-
tive-type of inhibition with respect to GSH (Ki ¼ 8.1 ꢃ0.7
mM) and
CDNB (Ki ¼ 1.7 ꢃ 0.02
m
M). This finding indicates that the released
inhibitor binds fairly strongly to the substrate binding site of
hGSTA1-1 and is expected to be an effective inhibitor to other GST
isoenzymes present in cancer cells.
In the past, modulation by inhibition of GST has been attempted
as a means to improve response to cancer drugs. Use of, for
example, ethacrynic acid, although effective during its experi-
mental behaviour with various GST isozymes, was not successful
enough in the clinic to merit continued development [45]. In
another approach the peptidomimetic inhibitor of GSTP1-1, TLK199
[
g
-glutamyl-S-(benzyl)cysteinyl-R(ꢂ)phenyl glycine diethyl ester]
was envisaged as a plausible means to sensitize drug-resistant
tumours that overexpress GST [46].
More recently GST-activated prodrugs were developed to
exploit high hGSTP1-1 levels associated with malignancy, poor
prognosis, and the development of drug resistance [11]. For
example, in TLK286 (Telcyta), the sulphydryl of a glutathione
conjugate has been oxidized to a sulphone [47,48]. The tyrosine-7
in hGSTP1-1 promotes a
b-elimination reaction that cleaves the
latter compound. The cleavage products are a glutathione analogue

2016
I. Axarli et al. / European Journal of Medicinal Chemistry 44 (2009) 2009–2016
[38] M. Czerwinski, J.P. Gibbs, J.T. Slattery, Drug Metab. Dispos. 24 (1996)
1015–1019.
[39] H.A.A.M. Dirven, B. van Ommen, P. van Bladeren, J. Chem. Res. Toxicol. 9 (1996)
351–360.
[45] P.J. O’Dwyer, F. LaCreta, S. Nash, P.W. Tinsley, R. Schilder, M.L. Clapper, K.D. Tew,
L. Panting, S. Litwin, R.L. Comis, R.F. Ozols, Cancer Res. 51 (1991) 6059–6065.
[46] A.S. Morgan, P.J. Ciaccio, K.D. Tew, L.M. Kauvar, Cancer Chemother. Pharmacol.
37 (1996) 363–370.
[40] I. Chatzistamou, A.V. Schally, K. Szepeshazi, K. Groot, F. Hebert, J.M. Arencibia,
Cancer Lett. 171 (2001) 37–45.
[41] J.G. Darker, S.J. Brough, J. Heath, D. Smart, J. Pept. Sci. 7 (2001) 598–605.
[42] D.J. Meyer, Xenobiotica 23 (1993) 823–834.
[47] M.H. Lyttle, A. Satyam, M.D. Hocker, K.E. Bauer, C.G. Caldwell, H.C. Hui,
A.S. Morgan, A. Mergia, L.M. Kauvar, J. Med. Chem. 37 (1994) 1501–1507.
[48] A. Satyam, M.D. Hocker, K.A. Kane-Maguire, A.S. Morgan, H.O. Villar,
M.H. Lyttle, J. Med. Chem. 39 (1996) 1736–1747.
[43] M. Chang, F. Zhang, L. Shen, N. Pauss, I. Alam, R.B. van Breemen, S.Y. Blond,
J.L. Bolton, Chem. Res. Toxicol. 11 (1998) 758–765.
[44] T.E. McHugh, W.M. Atkins, J.K. Racha, K.L. Kunze, D.L. Eaton, J. Biol. Chem. 271
(1996) 27470–27474.
[49] K.D. Tew, Expert. Opin. Investig. Drugs 14 (2005) 1047–1054.
[50] J.E. Saavedra, D.S. Bohle, K.N. Smith, C. George, J.R. Deschamps, D. Parrish,
J. Ivanic, Y.N. Wang, M.L. Citro, L.K. Keefer, J. Am. Chem. Soc. 126 (2004) 12880–
12887.