Antiglioma activity of GoPI-sugar, a novel gold(I)-phosphole inhibitor: Chemical synthesis, mechanistic studies, and effectiveness in vivo

Article abstract of DOI:10.1016/j.bbapap.2014.01.006

Glioblastoma, an aggressive brain tumor, has a poor prognosis and a high risk of recurrence. An improved chemotherapeutic approach is required to complement radiation therapy. Gold(I) complexes bearing phosphole ligands are promising agents in the treatment of cancer and disturb the redox balance and proliferation of cancer cells by inhibiting disulfide reductases. Here, we report on the antitumor properties of the gold(I) complex 1-phenyl-bis(2- pyridyl)phosphole gold chloride thio-β-d-glucose tetraacetate (GoPI-sugar), which exhibits antiproliferative effects on human (NCH82, NCH89) and rat (C6) glioma cell lines. Compared to carmustine (BCNU), an established nitrosourea compound for the treatment of glioblastomas that inhibits the proliferation of these glioma cell lines with an IC₅₀ of 430 μM, GoPI-sugar is more effective by two orders of magnitude. Moreover, GoPI-sugar inhibits malignant glioma growth in vivo in a C6 glioma rat model and significantly reduces tumor volume while being well tolerated. Both the gold(I) chloro- and thiosugar-substituted phospholes interact with DNA albeit more weakly for the latter. Furthermore, GoPI-sugar irreversibly and potently inhibits thioredoxin reductase (IC₅₀ 4.3 nM) and human glutathione reductase (IC ₅₀ 88.5 nM). However, treatment with GoPI-sugar did not significantly alter redox parameters in the brain tissue of treated animals. This might be due to compensatory upregulation of redox-related enzymes but might also indicate that the antiproliferative effects of GoPI-sugar in vivo are rather based on DNA interaction and inhibition of topoisomerase I than on the disturbance of redox equilibrium. Since GoPI-sugar is highly effective against glioblastomas and well tolerated, it represents a most promising lead for drug development. This article is part of a Special Issue entitled: Thiol-Based Redox Processes.

Full text of DOI:10.1016/j.bbapap.2014.01.006

BBAPAP-39278; No. of pages: 12; 4C: 5,8,9
Biochimica et Biophysica Acta xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Biochimica et Biophysica Acta
journal homepage: www.elsevier.com/locate/bbapap
Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor:
☆
Chemical synthesis, mechanistic studies, and effectiveness in vivo
E. Jortzik a,1, M. Farhadi b,1, R. Ahmadi b, K. Tóth c, J. Lohr b, B.M. Helmke d,e, S. Kehr a, A. Unterberg b, I. Ott f,g
,
a,
b,
R. Gust f,h, V. Deborde i, E. Davioud-Charvet j,k, R. Réau i, , K. Becker , C. Herold-Mende
⁎
⁎
⁎
a
Biochemistry and Molecular Biology, Justus Liebig University Giessen, Germany
Division of Experimental Neurosurgery, Department of Neurosurgery, University of Heidelberg, Germany
Division Biophysics of Macromolecules, German Cancer Research Center Heidelberg, Germany
Institute of Pathology, Elbe Klinikum Stade, Germany
Department of General Pathology, University of Heidelberg, Germany
Institute of Pharmacy, Freie Universität Berlin, Germany
Institute of Medicinal and Pharmaceutical Chemistry, Technische Universität Braunschweig, Germany
Department of Pharmaceutical Chemistry, Institute of Pharmacy, University of Innsbruck, Austria
b
c
d
e
f
g
h
i
UMR 6509 Institut de Chimie, CNRS Université de Rennes, France
UMR7509 CNRS and University of Strasbourg, European School of Chemistry, Polymers and Materials (ECPM), France
Center of Biochemistry, University of Heidelberg, Germany
j
k
a r t i c l e i n f o
a b s t r a c t
Article history:
Glioblastoma, an aggressive brain tumor, has a poor prognosis and a high risk of recurrence. An improved chemo-
therapeutic approach is required to complement radiation therapy. Gold(I) complexes bearing phosphole ligands
are promising agents in the treatment of cancer and disturb the redox balance and proliferation of cancer cells by
inhibiting disulfide reductases. Here, we report on the antitumor properties of the gold(I) complex 1-phenyl-
bis(2-pyridyl)phosphole gold chloride thio-β-D-glucose tetraacetate (GoPI-sugar), which exhibits antiproliferative
effects on human (NCH82, NCH89) and rat (C6) glioma cell lines. Compared to carmustine (BCNU), an established
nitrosourea compound for the treatment of glioblastomas that inhibits the proliferation of these glioma cell lines
with an IC50 of 430 μM, GoPI-sugar is more effective by two orders of magnitude. Moreover, GoPI-sugar inhibits
malignant glioma growth in vivo in a C6 glioma rat model and significantly reduces tumor volume while being
well tolerated. Both the gold(I) chloro- and thiosugar-substituted phospholes interact with DNA albeit more
weakly for the latter. Furthermore, GoPI-sugar irreversibly and potently inhibits thioredoxin reductase (IC50
4.3 nM) and human glutathione reductase (IC50 88.5 nM). However, treatment with GoPI-sugar did not significant-
ly alter redox parameters in the brain tissue of treated animals. This might be due to compensatory upregulation of
redox-related enzymes but might also indicate that the antiproliferative effects of GoPI-sugar in vivo are rather
based on DNA interaction and inhibition of topoisomerase I than on the disturbance of redox equilibrium.
Since GoPI-sugar is highly effective against glioblastomas and well tolerated, it represents a most promising lead
for drug development. This article is part of a Special Issue entitled: Thiol-Based Redox Processes.
Received 30 September 2013
Received in revised form 6 December 2013
Accepted 8 January 2014
Available online xxxx
Keywords:
Gold(I)–phosphole complex
Glioblastoma
Redox system
Thioredoxin reductase
Glutathione reductase
Cancer
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
Metal-based inhibitor complexes have a high potential as anticancer
agents, with platinum-based compounds such as the chemotherapeutic
Abbreviations: AP, alkaline phosphatase; BCNU, carmustine; BUN, blood urea nitrogen;
CHE, cholinesterase; Cr, creatinine; DMEM, Dulbeco's Modified Eagle Medium; GoPI, 1-
phenyl-bis(2-pyridyl)phosphole gold chloride; GoPI-sugar, 1-phenyl-bis(2-pyridyl)
phosphole gold chloride thio-β-D-glucose tetraacetate; GR, glutathione reductase; GSH,
reduced glutathione; GSSG, oxidized glutathione; GST, glutathione S-transferase; Hb,
hemoglobin; Hk, hematocrit; RBC, red blood cell count; SGOT, serum glutamate oxaloace-
tate transaminase; SGPT, serum glutamate pyruvate transaminase; TGR, thioredoxin
glutathione reductase; TMZ, temozolomide; Trx, thioredoxin; TrxR, thioredoxin reductase;
TUNEL, terminal dUTP nick end labeling; WBC, white blood cell count
agent cisplatin and its derivatives being the most prominent examples.
Although cisplatin acts in a highly effective manner on malignant tumor
cells by forming DNA–platinum adducts, its clinical use is limited by a
high systemic toxicity and fast development of cisplatin resistance
in human cancer cells [1]. The success of platinum-based drugs is an
arousing interest among researchers to study other metallodrugs
such as ruthenium and gold complexes in the treatment of cancer [2].
Different platinum and gold-based compounds have been reported to
target disulfide reductases such as thioredoxin reductase (TrxR) and/
or glutathione reductase (GR), which, at least partially, mediate their
☆
This article is part of a Special Issue entitled: Thiol-Based Redox Processes.
Corresponding authors.
⁎
1
The first two authors contributed equally to this work.
1570-9639/$ – see front matter © 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.bbapap.2014.01.006
Please cite this article as: E. Jortzik, et al., Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor: Chemical synthesis, mechanistic
studies, and effectiveness in vivo, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbapap.2014.01.006

2
E. Jortzik et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx
anti-tumor properties by inducing cytotoxic or antiproliferative effects
[3–8]. TrxR and its substrate thioredoxin (Trx) play a major role in the
homeostasis of the cellular redox state and are involved in cell prolifer-
ation by providing electrons for DNA synthesis via ribonucleotide reduc-
tase [9]. Moreover, the Trx system is closely involved in regulating
apoptosis by interacting with p53, an apoptosis-regulating protein
[10]. Expression levels of TrxR and Trx are often increased in aggressive
tumors [11,12], and upregulation of the TrxR–Trx system has been asso-
ciated with resistance to chemotherapy [13–15]. Accordingly, inhibition
of the TrxR–Trx system is known to inhibit cell proliferation and to an-
tagonize drug resistance [9,16]. GR reduces oxidized glutathione
(GSSG) and thereby maintains high concentrations of reduced glutathi-
one (GSH), which is crucial for maintaining a reducing cellular milieu.
Increased levels of glutathione and upregulated GR expression have pre-
viously been observed in tumor cells [17,18]. Tumor chemoresistance
has been linked to a dysregulation of the GSH system, resulting in an
enhanced conjugation of different anticancer agents to GSH, which are
then less toxic and exported from the cell [19]. Inhibition of GR leads
to disturbances of the cellular glutathione redox potential and an
increased efflux of GSSG, thereby potentially affecting the redox bal-
ance of tumor cells and decreasing resistance towards anticancer
agents by reducing the formation of GSH-drug conjugates. As poten-
tial target sites for anticancer drugs, human GR has an active site
with a cysteine–cysteine motif, while hTrxR has two redox-active
centers with a cysteine–cysteine and a cysteine–selenocysteine
pair [3,20,21]. Due to their limited aromatic character and the pres-
ence of a central nucleophilic phosphorous atom, phospholes are suit-
able for chemical modifications of thiols, a fact that drew our interest
towards studying them as disulfide reductase inhibitors [22]. In contrast
to reversible inhibitors, metal–ligand complexes are able to target the
catalytic cysteine–selenocysteine pair of hTrxR and additionally interact
with DNA [3]. Gold compounds exhibit a high affinity for sulfur- and
selenium-containing ligands such as proteins with selenocysteine as ex-
posed active site residues. Interestingly, gold(I)-containing complexes
are at least two orders of magnitude more active on hGR than the
corresponding platinum(II) or palladium(II)-containing complexes [3].
Furthermore, the gold(I) complexes are similarly active on both
the seleno-hTrxR and the non-seleno-hGR. Auranofin, a phosphine-
conjugated gold(I) thiosugar complex that has been developed to
treat rheumatoid arthritis, inhibits TrxR from humans, bacteria, and
parasites [23,24]. Auranofin and several hundred gold(I) phosphine
complexes were screened as part of an antineoplastic drug
development program in the '80s, which led to a large array of
pharmacokinetically different series, including the chlorophosphine
gold(I) complexes – from the simplest Et3PAuCl to the lipophilic
Ph3PAuCl – the various gold(I) thiolates; and alkyl, aryl, alkoxy,
and amino-substituted phosphine–gold complexes [25]. The structure–
activity relationships have shown that the structural changes within
ligand types coordinated to the gold atom, which govern the lipophilic
character, distribution, and partition coefficients as part of the pharma-
cokinetics, can profoundly alter the cytotoxicity of the gold(I) complexes
both in vitro and in vivo. Furthermore, as previously reported, both
auranofin 4 and its gold(I) chlorotriethyl phosphine were shown to act
as prodrugs in vivo, and function via a bioactivation process involving
the transient displacement of the thiosugar moiety or the chlorine
atom by (redox) protein(s) or low molecular weight thiols before irre-
versibly binding to the common final DNA target. Both the chlorine
atom and the thiosugar part are known leaving groups displaced by
cysteine and histidine residues of albumin [26–29]. Following different
reversible exchanges with thiol groups or proteins in the blood, these
gold(I) complexes are known to bind DNA: interaction of auranofin 4
or gold(I) chlorotriethyl phosphine led to the formation of the same
gold(I)–DNA complex [30].
penultimate C-terminal part [31–33]. As for hGR alkylated by the gold–
phosphole complex GoPI [3], the X-ray structure of wild type TGR incu-
bated with auranofin showed an almost linear S–Au(I)–S-coordination
in the active site following the release of both the leaving group and
the phosphine moiety [31]. “Auranofin-like” gold(I) complexes contain-
ing thiocyanate or xanthate affect TrxR activity in a low nanomolar
range and inhibit GR, but they do so less effectively with IC50 values in
the micromolar range. Furthermore, they have antiproliferative effects
on several human cancer cell lines [6].
Currently, temozolomide (TMZ) is the treatment of choice for
glioblastomas, a highly malignant primary brain tumor, and has mostly
replaced the currently used nitrosourea-based drugs such as carmustine
because it has a low toxicity and is easy to administer [34]. However, the
prognosis for patients with a glioblastoma remains poor, with a high
recurrence rate and a 5-year chance of survival of 9.8% [35]. A range of
structurally diverse gold(I) complexes has been studied as agents for
the treatment of cancer [36,37]. Among them are gold(I) phosphine
complexes [5,6], gold(I) carbene complexes [8,38], alkynyl gold(I) com-
plexes [39], gold(I) complexes of imidazole and thiazole-based diphos
type ligands [40], and azolate gold(I) phosphane complexes [7]. All of
them demonstrate an anticancer activity by exhibiting antiproliferative
or cytotoxic effects, and many of them were shown to target human
thioredoxin reductase. We previously have shown that the complex 1-
phenyl-bis(2-pyridyl)phosphole gold chloride (GoPI) is a highly potent
inhibitor of human GR and TrxR, with IC50 values in the low nanomolar
and micromolar range, respectively. Gold(I)-containing phosphole
complexes competitively inhibit hGR by competing with GSSG. (Phos-
phine)gold chloride complexes thereby act as prodrugs with stepwise
ligand displacement resulting in a covalently bound gold atom between
the two active site cysteines of hGR (S–Au–S coordination) [3]. The
binding mode of GoPI to hTrxR appears to be more complex due
to the presence of two redox-active sites, with a selenocysteine
in the C-terminal active site being mainly, but not exclusively,
targeted (S–Au–Se coordination) [3,31]. It has been suggested that
the selenocysteine likely mediates the transfer of gold from its
phosphine ligands in GoPI or auranofin complexes to the redox-
active Cys pair of TrxR or TGR. In addition to disulfide reductase in-
hibition, the antiproliferative effects of GoPI on glioblastoma cells
can be mediated by binding to DNA [3]. However, GoPI is unstable
due to a release of the chloride ligand in aqueous solutions. A
sugar-linked form of GoPI, the complex 1-phenyl-bis(2-pyridyl)
phosphole gold chloride thio-β-D-glucose tetraacetate (GoPI-sugar)
was synthesized based on the structures of GoPI and auranofin and
has improved solubility, stability, and bioavailability. GoPI-sugar
inhibits the human breast cancer cell line MCF-7 with IC50 values of
around 2 μM by triggering nuclear fragmentation and chromatin
condensation, which are indicative of apoptosis, and perturbing
the cell cycle. Interestingly, GoPI-sugar irreversibly inhibits hTrxR in
MCF-7 cells (IC50 = 1.9 μM) but does not affect hGR activity [41].
Thus, GoPI-sugar shows cytotoxic and cytostatic effects on human
breast cancer cells, which seem to be at least partially mediated by inhi-
bition of hTrxR [41]. This motivated us to study the effects of GoPI-sugar
on glioblastomas, the most common and lethal type of brain tumors.
Glioblastomas, which typically harbor mutations of p53, usually show
increased levels of TrxR, indicating that the Trx system is involved in
glial neoplastic diseases [11]. As delineated below, GoPI-sugar indeed
inhibits the proliferation of glioma cells in vitro and in vivo with tolera-
ble side effects.
2. Material and methods
2.1. GoPI-sugar synthesis
Interestingly, auranofin was recently reported to inhibit the
worm protein thioredoxin–glutathione reductase (TGR) from Schistosoma
Sodium hydride (37.7 mg, 1.5 mmol) was added to a solution of 1-
thio-β-D-glucose tetraacetate (272 mg, 0.75 mmol) in tetrahydrofuran
(25 mL). The heterogeneous mixture was stirred for 1 h at room
mansoni parasites, which also contains
a selenocysteine at the
Please cite this article as: E. Jortzik, et al., Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor: Chemical synthesis, mechanistic
studies, and effectiveness in vivo, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbapap.2014.01.006

E. Jortzik et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx
3
temperature. The solution was filtered and added to a tetrahydrofuran
2.4. Cell culture conditions
solution (20 mL) of 1-phenyl-2,5-bis(2-pyridyl)phosphole gold
complex 1 (408 mg, 0.68 mmol) [3]. The solution was stirred for 90 min
at room temperature, and all volatile materials were removed under
vacuum. After purification by column chromatography on silica gel
(diethyl ether/ethyl acetate: 7/3), complex 2 was obtained as a yellow
powder (436 mg, yield 70%). A patent for GoPI-sugar was registered
(EP1771181B1).
The malignant rat glioma cell line C6 (ATCC, Rockville, MD) as well
as NCH82 and NCH89 cells established from human glioblastoma tis-
sues [42] were cultured in Dulbeco's Modified Eagle Medium (DMEM)
supplemented with 10% fetal calf serum and antibiotics (100 units/mL
penicillin and 100 μg/mL streptomycin) at 37 °C in a humidified
atmosphere with 5% CO2 and 95% air. Medium changes were performed
twice a week. After reaching confluence, cells were harvested by a brief
incubation with a trypsin/EDTA solution (Viralex, PAA, Linz, Austria)
and seeded into a fresh 75 cm2 plastic tissue culture flask.
1H NMR (CDCl3, 300.1 MHz): δ = 1.81 (s, 3H, CH3), 1.85 (m,
4H, _CCH2CH2), 2.02 (s, 3H, CH3), 2.03 (s, 3H, CH3), 2.10 (s, 3H,
CH3), 3.01 (m, 2H, _CCH2CH2), 3.32 (m, 2H, _CCH2CH2), 3.69
(ddd, 1H, JHH = 2.1 Hz, JHH = 4.5 Hz, JHH = 9.9 Hz, CH), 4.00 (dd,
1H, JHH = 2.1 Hz, JHH = 12.3 Hz, CH), 4.10 (dd, 1H, JHH = 2.1 Hz,
JHH = 12.3 Hz, CH), 4.94 (t, 1H, JHH = 9.9 Hz, CH), 5.04 (m, 3H,
CH2, CH), 7.09 (2dd, 2H, JHH = 4.8 Hz, JHH = 7.8 Hz, H5 Py), 7.29
(m, 3H, m-Ph, p-Ph), 7.63 (2dd, 2H, JHH = 7.8 Hz, JHH = 7.8 Hz, H4
Py), 7.75 (2d, 2H, JHH = 7.8 Hz, H3 Py), 7.79 (d, 2H, JHH = 8.1 Hz,
o-Ph), 8.56 (2d, 2H, JHH = 4.8 Hz, H6 Py) 13C{1H}NMR (CDCl3,
75.5 MHz): δ = 20.6 (s, CH3), 20.7 (s, CH3), 20.8 (s, CH3),
2.5. Proliferation assay
Quantification of DNA synthesis was performed using the BrdU
Labeling and Detection Kit III (Roche Diagnostics, Mannheim, Germany)
as described [43]. Tumor cells were seeded in 8 replicates on 96-well
plates. For the proliferation assay after chemotherapy, GoPI-sugar
was administered in different concentrations 24 h after seeding and
incubated for 72 h. Optical density was determined, and the means of
control samples were defined as 100% proliferation rate. Values are
means of at least two independent experiments.
21.2 (s, CH3), 22.4 (d, JPC = 3.1 Hz = CCH2CH2), 29.6 (d, JPC
=
9.0 Hz, _CCH2), 29.7 (d, JPC = 9.1 Hz, _CCH2), 62.8 (s, CH2 sucre),
68.9 (s, CH sucre), 74.3 (s, CH sucre), 75.6 (s, CH sucre), 77.7 (s,
CH sucre), 83.1 (s, CH sucre), 122.0 (s, C5 Py), 122.2 (s, C5 Py), 123.9
(d, JPC = 6.5 Hz, C3 Py), 124.1 (d, JPC = 6.3 Hz, C3 Py), 126.5 (d, JPC
=
2.6. Orthotopic glioma model
56.9 Hz, i-Ph), 128.8 (d, JPC = 12.1 Hz, m-Ph), 131.4 (d, JPC = 2.6 Hz,
p-Ph), 134.2 (d, JPC = 14.0 Hz, o-Ph), 134.6 (d, JPC = 58.0 Hz, _CP),
135.1 (d, JPC = 58.0 Hz, _CP), 136.3 (s, C4 Py), 149.3 (d, JPC = 6.0 Hz,
Male 6 to 8-week-old Wistar rats were used (Charles River, Wiga,
Germany). Institutional guidelines for animal welfare and experimental
conduct were followed. After an intracranial injection of 2 × 105 C6
glioma cells [44], animals were repeatedly (on days 3, 5, 7, and 9 post-
implantation) treated intravenously with two different doses of GoPI-
sugar (22 mg/kg body weight or 30 mg/kg body weight). Control
animals received 500 μL of the vehicle intravenously (10% DMF and
10% Chremophor (1:1), adjusted to 100% with NaCl). Analysis of
tumor growth and response to therapeutic interventions was per-
formed via repetitive MRI scanning using a 2.35 T small bore MRI unit
(Biospec 24/40, BRUKER Medizintechnik, Ettlingen, Germany) as de-
scribed [44]. Body weight and neurological symptoms were recorded
every two days. All rats were euthanized via decapitation on day 16
after tumor cell implantation, and the vital organs including contralater-
al brain, liver, kidney, as well as muscles were immediately snap frozen
in liquid nitrogen and stored at −80 °C until processing for further his-
tological and biochemical analyses. Prior to euthanization, whole blood
samples were collected in order to assess potential treatment-induced
drug toxicity by measuring hemoglobin (Hb), platelet count, white
blood cell (WBC) count, red blood cell (RBC) count, creatinine (Cr),
serum glutamate oxaloacetate transaminase (SGOT), serum glutamate
pyruvate transaminase (SGPT), blood urea nitrogen (BUN), alkaline
phosphatase (AP), and bilirubin. Furthermore, the oxidative status of
the blood samples was analyzed as described below.
C
6 Py), 149.4 (d, JPC = 5.6 Hz, C6 Py), 152.3 (d, JPC = 15.8 Hz, C2 Py),
152.5 (d, JPC = 16.1 Hz, C2 Py), 152.9 (d, JPC = 14.0 Hz, C = CP), 153.5
(d, JPC = 14.0 Hz, C = CP), 169.5 (s, C_O), 169.6 (s, C_O),
170.2 (s, C_O), 170.8 (s, C_O). 31P{1H} NMR (CDCl3, 121.5 MHz)
δ
= +46.7 ppm (s). HR-MS (ESI, CH2Cl2): m/z 951.1768
(M++Na+: calculated 951.1755).
2.2. DNA melting curves
Linearized PUC 18 plasmid DNA was incubated with 1–20 μM 1
in degassed 10 mM phosphate pH 7.0 containing 1% dioxan (ratio
of 1 to number of basepairs 1–4:20–1:1). Thermal denaturation
curves were recorded between 50 and 97 °C (0.5 °C/min) at 260
nm (hyperchromic effect) on a Cary 4E spectrophotometer (Varian,
Mulgrave, Australia). After normalization to absorbance at room tem-
perature and smoothing in five-point intervals, derivative melting
curves of the samples were drawn (KaleidaGraph™).
2.3. Gel motility studies
Superhelical DNA (pUC18) was incubated with topoisomerase I
(Invitrogen) and varied concentrations of complexes 1, 3, and 4. Incuba-
tion conditions were optimized by varying incubation time (1 to 2 h),
the amount of topoisomerase I (calf thymus from Invitrogen or wheat
germ from Promega), and the reaction buffer. All incubations were
performed at 37 °C without shaking. After incubation, an equal volume
of phenol/chloroform/isoamylalcohol was added to the sample, mixed
thoroughly, and centrifuged for 5 min at 35,000 g at room temperature.
The upper (aqueous) phase was collected, and an equal volume
of phenol/chloroform/isoamylalcohol was added and centrifuged
as described above. The upper phase was collected. No further
chloroform and ether extractions were done, since phenol does
not interfere with gel electrophoresis. 1% agarose gels were
run in TBA buffer for 200 min at constant 8 V/cm and 200 mA.
0.1 μg DNA sample and a 1 kb marker from Bioron was loaded
onto the gel. For visualization the gels were stained with ethidium
bromide.
For this purpose the tissues of treated and untreated animals were
taken, weighed, homogenized using a dounce homogenizer, and ana-
lyzed according to a previously published atomic absorption spectros-
copy method [45]. Gold levels are indicated as ng gold per g tissue.
2.7. Ki-67 staining and in situ apoptosis assay
To distinguish between proliferating cells and those undergoing
apoptosis, 5 μm cryostat sections were prepared from tumor xenografts.
In order to assess proliferation in tumor xenografts, cryostat sections
were incubated with a Ki-67 antibody (mouse monoclonal antibody,
10 μg/mL, BD Pharmingen, Hamburg, Germany) for 1 h. Detection was
performed as described [42]. To measure apoptotic cells, a terminal
dUTP nick-end labeling (TUNEL) assay was performed using an in situ
apoptosis detection kit according to the manufacturer's instructions
(Boehringer Mannheim, Mannheim, Germany). The Ki-67-positive
and apoptotic cells were counted per visual field on 3 hotspots at ×20
Please cite this article as: E. Jortzik, et al., Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor: Chemical synthesis, mechanistic
studies, and effectiveness in vivo, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbapap.2014.01.006

4
E. Jortzik et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx
magnification using the AnalySIS image analysis software (Soft Imaging
System, Muenster, Germany). Counting was performed in a blinded
manner. Finally, mean values were calculated.
143 mM sodium phosphate, 6 mM EDTA, pH 7.5, 0.6 mM DTNB, 0.5 U/
mL GR, and 0.3 mM NADPH. The increase in absorbance due to reduc-
tion of DTNB was monitored at 412 nm and 25 °C. The glutathione con-
centration was calculated using a standard curve. The concentration of
glutathione in erythrocytes was estimated on the basis of whole blood
glutathione and the hematocrit ([GSH]Ery = [GSH]blood · 100 / Hk%)
[44].
2.8. Tissue preparation
Tissues from control and GoPI-sugar-treated rats were kept frozen in
liquid nitrogen, pulverized in a mortar and solved in 100 mM potassium
phosphate buffer, 2 mM EDTA, pH 7.4. After freezing and thawing three
times, the homogenized tissues were lysed by sonication and centri-
fuged at 35,000 g and 4 °C for 30 min. The clear supernatant was directly
used in enzyme assays.
2.12. Inhibition studies
For the inhibition studies, hGR and hTrxR were produced as de-
scribed previously (recombinant hGR [47] with a specific activity
of 60–120 U/mg, wild type hTrxR isolated from human placenta
[48] with 16 U/mg activity as determined in the DTNB assay, and
recombinant hTrxRU498C [49] with 0.46 U/mg as determined in the
DTNB assay). In order to test the inhibition of recombinant enzymes,
both compounds GoPI and GoPI-sugar were dissolved in DMF. hGR
inactivation was measured in the GSSG reduction assay as described
above in Section 2.9. For hTrxR, inactivation of the wild type and the
selenocysteine-lacking mutant hTrxRSec498Cys was measured with
the DTNB and Trx reduction assays (see Section 2.9.). The following
final concentrations of redox enzymes were employed in the assays
for IC50 determinations: 1.4–2.8 nM hGR, 3.7 nM wild type hTrxR
(DTNB assay), 30 nM wild type hTrxR (thioredoxin reduction assay),
2.9. Tissue samples, protein and enzyme assays
The protein concentration of the tissue samples was determined
photometrically at 595 nm using the Bradford assay and a bovine
serum albumin calibration curve. All activity assays with tissue samples
were carried out at 37 °C with pre-warmed substrates. All measurements
were done at least in duplicate. Data and mean values SD were calcu-
lated for each therapeutic group.
As the TrxR activity was supposed to be low in some of the tissues,
the sensitive DTNB-reducing activity of the enzyme was determined.
TrxR activity was measured in 100 mM potassium phosphate, 2 mM
EDTA, pH 7.4 with 3 mM DTNB as the substrate (100 mM stock solution
in DMSO). TNB− production was monitored at 412 nm, and the activity
was calculated using εTNB = 13.6 mM−1 cm−1. Non-NADPH-dependent
reduction of DTNB was determined for each sample and subtracted.
The NADPH-dependent Trx-reducing activity of hTrxR was measured
by following the oxidation of NADPH spectrophotometrically at
340 nm as described elsewhere [46].
The assay for measuring the activity of GR contained 47 mM potassi-
um phosphate, 1 mM EDTA, 200 mM KCl, pH 6.9, 100 μM NADPH, and
2 μM FAD. After adding tissue extract, the reaction was started with
1 mM GSSG. The oxidation of NADPH was monitored via a decrease in
absorbance at 340 nm. Adding FAD to the assay mixture allows one to
determine total GR (holoGR + apoGR) [44].
Glutathione peroxidase (GPx) assays were performed in 100 mM
Tris, 1 mM EDTA, pH 8.0, 2 mM GSH, 1.0 U/mL PfGR, and 100 μM
NADPH [44]. After adding tissue extract and incubating for 5–10 min,
the assay was started by adding 70 μM tert-butylhydroperoxide. A refer-
ence cuvette containing all components without tissue extract was used
as a control.
140–280 nM hTrxRU498C (DTNB assay), and
3
μM hTrxRU498C
(thioredoxin reduction assay). Half-maximal inhibition of the enzymes
(IC50, analyzed with GraphPad Prism, GraphPad software, USA) was
determined by incubating the enzymes (hGR, hTrxR and hTrxRU498C
)
for 10 min with NADPH (100 or 200 μM) and inhibitor (0.5–500 nM
and 1–40 μM) at 25 °C. The enzyme activity measurement was started
by adding GSSG (100 μM), DTNB (3 mM), or hTrx (20 μM) to the
mixtures.
2.13. Statistics
Data were analyzed for statistical significance using an unpaired
Student's t-test. P-values b0.05 were considered to be statistically
significant.
3. Results and discussion
3.1. Synthesis and crystal structure of GoPI-sugar
Glutathione S-transferase (GST) activity was determined in 100 mM
HEPES, 1 mM EDTA, pH 6.5 with 1 mM GSH and 500 μM CDNB [44]. A
reference cuvette containing all components without a tissue sample
was used as a control.
In order to increase the stability and solubility of GoPI, a sugar-linked
form of GoPI (1-phenyl-bis(2-pyridyl)phosphole gold chloride thio-β-
D-glucose tetraacetate, GoPI-sugar) was synthesized. The modified
gold complex 2 (Fig. 1) bearing the 1-phenyl-bis(2-pyridyl)phosphole
ligand was readily prepared by reacting the precursor 1 with 1-thio-β-
D-glucose tetraacetate in the presence of sodium hydride (Fig. 1B). The
novel derivative 2 was isolated as an air-stable yellow powder with
70% yield following purification by column chromatography. Its multi-
nuclear NMR data and mass spectrum support the proposed structure.
For example, its 31P{1H} NMR chemical shift at +47.3 ppm is typical
for Au(I)–phosphole complexes (e.g. 1, +39.9 ppm). It is worth noting
that replacing the Cl ligand with the thio-sugar donor (1 → 2, Fig. 1B)
results in improved air and moisture stability along with an increase
in the solubility properties of the gold complexes. For example, complex
2 is stable in CH2Cl2 solution in air for weeks, whereas derivative 1
decomposes within a few hours under the same conditions. Likewise,
complex 2 is highly soluble in pentane, whereas 1 is not soluble in this
apolar aliphatic solvent. Complex 2 was also characterized by X-ray dif-
fraction studies performed on single crystals (Fig. 1C). The metric data
of the 1-phenyl-2,5-bis(2-pyridyl)phosphole ligand are unremarkable
and compare well to those of complex 1. The P–Au distance (2.253(2)
Å) as well as the P–Au–S angle (178.28(10)°) are similar to those
encountered in phosphole–Au(I) complexes. The Au–S bond distance
2.10. Blood samples, protein and enzyme assays
To lyse erythrocytes in whole blood samples, digitonin was
added at a final concentration of 40 μg/mL to the different assay
buffers. The hemoglobin concentration in the whole blood samples
was estimated as follows: concentration (Hb) = (A340nm of sample
Hb − A340nm of lysis assay buffer) × D / 18.5; with D = dilution fac-
tor for the blood sample, factor 18.5 = ε340 nm (g/dL)−1 cm−1 [44]. The
activities of GR, GPx, and GST in whole blood samples had already been
determined in Section 2.9 for the tissue samples.
2.11. Measurement of glutathione concentrations
Total glutathione concentrations in whole blood as well as in differ-
ent tissue samples were measured as described previously [44]. Briefly,
one volume of fresh EDTA blood was mixed with two volumes of 5%
sulfosalicylic acid (w/v). After centrifugation for 10 min at 10,000 g,
10 μL of supernatant was transferred in a total volume of 500 μL
Please cite this article as: E. Jortzik, et al., Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor: Chemical synthesis, mechanistic
studies, and effectiveness in vivo, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbapap.2014.01.006

E. Jortzik et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx
5
Fig. 1. Structure (A) and synthesis (B) of gold-based compounds. Part (C) shows the crystal structure of complex 2.
(2.291(2) Å) lies in the range of classic values. Overall, this solid state
confirms the proposed structure and reveals no unexpected features.
denaturation of the complex via UV absorption. The DNA melting curves
of related metallophospholes were compared to those of related
phosphines: GoPI 1, GoPI-sugar 2, auranofin 4, and the commercially
available lipophilic Ph3PAuCl 3 at concentrations between 0 and
10 μM (Fig. 2A). In the drug–DNA interaction study, we mainly inves-
tigated the effect of the leaving group attached to the gold phosphine
moiety of the molecule. Hence, the comparison between GoPI and
GoPI-sugar was obviously guided by the evaluation of the impact of
the presence of a good leaving group, i.e. the chlorine atom, versus
3.2. GoPI-sugar shows less direct interactions with DNA than GoPI
GoPI was shown to interact with DNA in previous experiments
thereby potentially interfering with DNA replication and transcription
[3]. To study the DNA-binding behavior of GoPI-sugar, we treated a
linearized plasmid DNA with GoPI-sugar and recorded the thermal
Fig. 2. (A) Effect of gold-based compounds on thermal stability of linearized plasmid DNA. The derivative melting curves are shown for untreated DNA as a control (black) and for DNA in
the presence of 5 μM (dashed line) or 10 μM (continuous line) of the compounds (complex 1: magenta; complex 2: blue; complex 3: red; complex 4: green). (B) Gel-shift assay for testing
DNA intercalation of complex 1. Lane content: DNA size-markers: 1; increasing complex 1 concentration (0–20 μM) in the absence of topoisomerase I: 2–4; in the presence of calf thymus
topoisomerase I: 5–7; in the presence of wheat germ topoisomerase I: 8–9. A gradual upwards shift in the topoisomer distribution of the bands was observed at increasing concentrations
of complex 1 in the presence of either topoisomerases.
Please cite this article as: E. Jortzik, et al., Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor: Chemical synthesis, mechanistic
studies, and effectiveness in vivo, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbapap.2014.01.006

6
E. Jortzik et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx
the thiosugar moiety. Previous studies [25] have demonstrated that
(i) both gold coordination complexes, auranofin 4 and the related
gold(I) chlorotriethyl phosphine, did not induce any observable
changes in the electrophoretic mobility of any forms of DNA, when
tested with superhelical or relaxed DNA [50,51]; (ii) interaction of
auranofin 4 or gold(I) chlorotriethyl phosphine led to the formation
of the same gold(I)–DNA complex [30] thereby proving a prodrug effect
for auranofin since bioactivation of both compounds leads to the same
DNA adduct; (iii) the phosphine-coordinated gold(I) thiosugar com-
plexes displayed higher cytotoxicity in vitro and higher antitumor activ-
ity in vivo when compared to the phosphine-coordinated chloro gold(I)
complexes [25]; (iv) the phosphine-coordinated chloro gold(I) com-
plexes Ph3PAuCl and Et3PAuCl displayed similar antitumor properties
in vivo with the same increase in median life span (ILS) relative to the
control P388 leukemia tumor-bearing mice (ILS = 36% for Ph3PAuCl
or Et3PAuCl versus 70% for auranofin). Therefore, we conducted DNA
binding experiments with four drug representatives: both gold(I)
chloro- and thio-substituted phosphines. In addition, we selected the
commercially available Ph3PAuCl as a representative of the unavailable
Et3PAuCl in order to compare it to Et3PAu–thiosugar (auranofin) in our
DNA binding studies.
3.3. GoPI-sugar exhibits an antiproliferative effect on glioma cells and
reduces in vivo tumor growth of malignant gliomas
In order to analyze the influence of GoPI-sugar on cell prolifera-
tion and compare it to GoPI, we measured changes in DNA synthesis
by incorporating 5-bromo-2′-deoxy-uridine (BrdU) into human
(NCH89, NCH82) and rat (C6) glioma cell lines after treatment
with GoPI-sugar. In all three of these cell lines, GoPI-sugar induced
a pronounced decrease of tumor cell growth compared to solvent-
treated controls. GoPI-sugar inhibited tumor cell growth with EC50
values of 0.9 μM (NCH82), 1.5 μM (NCH89) and 1.2 μM (C6), respec-
tively (Fig. 3). Thus, GoPI-sugar inhibits not only human breast
cancer cells [41] but also the growth of a glioma cell line with EC50
values in a low micromolar range. Interestingly, when compared
to GoPI [3], antiproliferative effect on glioma cell lines was in-
creased by 2–12 times by the introduction of a thiosugar, thereby
representing a major improvement of the compound. For compari-
son, it should be mentioned that carmustine (BCNU), an established
nitrosourea compound for the treatment of glioblastomas, inhibits
the proliferation of these human glioblastoma cell lines with an
EC50 of 430 μM [3].
The melting derivative curve of this DNA was composed of two
regions: the opening of an AT-richer domain at 75 °C and a GC-richer
domain at 79 °C. Contrary to earlier studies with GoPi [3], GoPI-sugar
has no significant effect on DNA stability at these concentrations,
which is similar to the related compound auranofin. This observation
was expected, because GoPI-sugar and auranofin are more stable
prodrugs when compared to GoPI due to the nature of their leaving
group (thiosugar versus chlorine atom). A similar behavior was previ-
ously reported for auranofin and its chloro analog Et3PAuCl towards
DNA interaction, i.e. the absence of DNA interaction for the thiosugar
derivative [51]. Melting Ph3PAuCl-treated DNA shows a concentration-
dependent stabilization with the effect being somewhat different
from that of GoPI: the first melting derivative peak (around 75 °C) related
to the denaturation of the AT-rich regions is more affected. In the case of
GoPI, both domains were stabilized proportionally to the drug concentra-
tion (Fig. 2A).
In order to study the influence of GoPI-sugar on malignant
glioma growth in vivo, we employed the C6 glioma rat model
that shares important characteristics with human glioblastomas.
Animals were treated on day 3, 5, 7, and 9 post-implantation with
either a low dose (22 mg/kg body weight), a high dose (30 mg/kg) or
a vehicle alone. Tumor size and localization were assessed by represen-
tative coronal T2- and T1-weighted contrast-enhanced MRI images.
Treatment with GoPI-sugar significantly reduced mean tumor volumes
by 65% (22 mg/kg body weight) and 88% (30 mg/kg body weight) on
day 16 when compared to the control (Fig. 4). Thus, treatment with
GoPI-sugar leads to a considerable decrease in the growth of malignant
gliomas in rats, demonstrating its in vivo efficacy.
3.4. Gold detection in tumor tissue
According to thermal denaturation measurements, some gold
complexes stabilize the DNA double helix; however, GoPI-sugar
and auranofin only do so to a lesser extent. Various DNA-drug
binding modes, which can be distinguished by different tech-
niques, may cause such stabilization. In order to investigate the
mode of DNA interaction by the four gold(I) complexes 1–4, we
used a gel-shift assay based on topoisomerase I treatment of super-
helical DNA in the presence of the drugs (Fig. 2B). This technique
relies on the fact that, upon intercalation, the number of super-
helical turns in a plasmid changes, and the variations can be monitored
by topoisomerase I relaxation. As expected, complexes 2 and 4 did not
show any significant interaction, in agreement with previous observa-
tions for auranofin [30]. The absence of significant changes with
the thio-gold complex 2 in the electrophoretic mobility of any forms
of DNA is due to the fact that both, the slow reacting GoPI-sugar 2
and the fast reacting GoPI 1, are prodrugs and require a bioactivation
process to form the same final DNA adducts in vivo, as it was previously
evidenced for auranofin [30]. Interestingly, we observed that all investi-
gated gold complexes inhibited topoisomerase I. Fig. 2B presents this
inhibition for compound 1 depending on the drug concentration. Con-
sequently, DNA intercalation could not be measured unambiguously.
We observed this inhibition effect with topoisomerase I from two
different sources (animal and plant). Noteworthy, inhibition of human
topoisomerase I and II has also been reported for gold(III) complexes
of pyridyl- and isoquinolylamido ligands [52] and cyclometalated
gold(III) complexes, which also inhibit the growth of different human
cell lines [53]. Since this property is interesting in further optimization
of GoPI-sugar, it will be studied by topoisomerase inhibition assays in
future experiments.
Using atomic absorption spectroscopy we exemplarily deter-
mined the concentrations of gold in the tumor tissue of rats after
glioma cell implantation and treatment with 30 mg/kg body weight
of GoPI-sugar. In untreated control rats, gold was not detectable in
brain tumor tissue. The tumor tissue of rats treated with 30 mg/kg
of GoPI-sugar contained 658.5 ng/g gold in the contralateral
hemisphere and 804.7 ng/g gold in the tumor-bearing hemisphere.
This indicates that GoPI-sugar, or at least the gold moiety, indeed
crossed the blood–brain barrier and reached the brain tissue at relevant
levels with some preference for the glioblastoma tumor cells.
Fig. 3. Antiproliferative effects of GoPI-sugar and GoPI on two human (NCH89, NCH82)
and a rat (C6) glioma cell line by GoPI-sugar and GoPI. Corresponding EC50 values are
given in the table.
Please cite this article as: E. Jortzik, et al., Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor: Chemical synthesis, mechanistic
studies, and effectiveness in vivo, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbapap.2014.01.006

E. Jortzik et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx
7
Fig. 4. Effect of GoPI-sugar on a C6 glioma model. Part (A) shows a representative Magnevist enhanced T1-weighted MRI image from each treatment group on days 6 and 16 after cell
implantation. Part (B) shows the tumor volume of the treatment groups compared to the controls on days 6 and 16.
3.5. GoPI-sugar displays a low toxicity towards C6 rats
3.6. Reduced tumor growth is caused by downregulation of tumor cell
proliferation but not due to an induction of apoptosis
GoPI-sugar treatment was generally well tolerated and did not
lead to histopathological alterations in the vital organs including
the liver, lung, kidney, skin, gut, heart, muscle, pancreas, and spleen.
The body weight of C6 glioma-bearing rats showed no significant
differences between either of the treatment groups and the control
group, with a mean weight of 237 25 g on day 0 and 282 g 27 g
on day 16 (treated animals) compared to 251 33 g and 281 26 g
(control animals). Within a few hours after the first application, two
animals of the treatment group and one animal of the control group
died. The absence of histopathological alterations with a clinically
acute occurrence of cardiopulmonary decompensation most likely
indicates a toxic shock with anaphylaxis. However, a pulmonary
embolism cannot be excluded. Analysis of different blood parame-
ters showed no significant changes in both groups. There were mild
but insignificant differences in hepatic function parameters with
elevated serum glutamate oxaloacetate transaminase (GOT) and
decreased cholinesterase (CHE) levels in treated animals. Significant
changes in renal function parameters and blood counts were not
observed (Table 1).
Tumor sections from treated and control animals were employed to
assess cell proliferation via Ki-67 staining. In animals treated with a high
dose (30 mg/kg body weight) of GoPI-sugar, we observed a significantly
decreased tumor cell proliferation by 59% (Fig. 5), while 20 mg/kg did
show an antiproliferative effect. This was, however, not significant and
corresponds to the fact that no gold could be detected in the brain
tumor tissue of rats treated with a low dose as reported in Section 3.4.
To further explore whether reduced tumor cell proliferation is associat-
ed with increased apoptosis, we applied TUNEL staining on neighboring
tumor sections. Interestingly, we did not detect a significant increase in
the number of apoptotic cells in tumor tissue of GoPI-sugar-treated rats
(Fig. 6), indicating that GoPI-sugar does not induce apoptosis at the
concentrations examined. However, it cannot be excluded that anti-
apoptotic effects were not anymore detectable, because treatment
ended on d9 while the analysis was performed on post-mortem brains
from d16. Contrarily, GoPI-sugar treatment increased nuclear fragmen-
tation and chromatin condensation in human breast cancer cells (MCF-
7) [41]. This might point towards a different mode of cell death in brain
Table 1
Stability of metabolic parameters in the blood of rats treated with GoPI-sugar (d16). The values indicate very low acute in vivo drug toxicity of the compound.
Control
0.2 0.1
29.3 10.4
65.6 3.7
37 6.2
121.6 70.8
1.3 3.0
0.4 0.3
GoPI-sugar (22 mg/kg)
GoPI-sugar (30 mg/kg)
Cr
0.3 0.0
33.7 5.3
143.5 73.3
51.2 7.6
128.2 62.8
2.2 0.5
0.3 0.1
30.8 4.8
117 85.2
51.2 11.7
170.8 47.3
1.4 1.9
0.2 0.0
0.1 0.0
12.7 1.2
6.1 0.2
mg/dL
mg/dL
U/L
U/L
U/L
BUN
GOT
GPT
AP
GGT
CHE
Bilirubin
Hb
U/L
0.4 0.4
kU/L
mg/dL
g/dL
/pL
0.2
0
0.2
0
14.6 1.3
6.8 0.4
11.8 0.3
6.4 0.2
RBC
Hk
0.4 0.0
0.3 0.0
0.3 0.0
l/L
WBC
Platelet
5.9 2.1
1264 500.6
10.8 0.7
1297 33.2
7.9 4.6
1227 230.7
/nL
/nL
CHE, cholinesterase; Cr, creatinine; BUN, blood urea nitrogen; GGT, gamma-glutamyl transpeptidase; GOT, serum glutamate oxaloacetate transaminase; GPT, serum glutamate pyruvate
transaminase; AP, alkaline phosphatase; RBC, red blood cells; Hb, hemoglobin; Hk, hematocrit; WBC, white blood cells.
Please cite this article as: E. Jortzik, et al., Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor: Chemical synthesis, mechanistic
studies, and effectiveness in vivo, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbapap.2014.01.006

8
E. Jortzik et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx
Fig. 5. Tumor cell proliferation after treatment with GoPI-sugar (d16). (A) Tumor sections from each of the groups were examined by immunohistochemistry for cell proliferation by using
a Ki-67 antibody. (B) Quantitative analyses of the percentage of proliferating cells in GoPI-sugar-treated tumors and controls. Positive cells were counted in three hotspots per tumor
sample. A comparison of Ki-67-positive cells in tumor sections (5 μm) shows that treatment with 30 mg GoPI-sugar/kg, but not with 22 mg GoPI-sugar/kg, significantly inhibits primary
tumor cell proliferation. *P b 0.005.
tumors and breast tumors or is more likely due to different compound
concentrations and consequently different effects in tumor tissue
in vivo and in cell culture. Antiproliferative effects with induction of
apoptosis by gold(I) compounds have been demonstrated in different
cancer cell lines: Gold(I) complexes of imidazole and thiazole-based
diphos type ligands induced DNA fragmentation in the hepatic tumor
cell line H4IIE [40], and different phosphine gold(I) complexes were
shown to activate caspase-3 and capase-9 and induce apoptosis in
cancer cells [6,54]. Therefore, the effect of GoPI-sugar on cell death has
to be studied systematically.
improved hTrxR inhibition, which is reflected by an IC50 of 4.3 nM for
GoPI-sugar compared to 7 nM for GoPI as determined in the Trx-assay
(Table 3, Fig. 7). This result might be reflected in the improved anti-
proliferative activity of GoPI-sugar as discussed in Section 3.3. Since
the Sec→Cys mutant of hTrxR (hTrxRU498C) is much less affected
by GoPI-sugar than the wild type enzyme (more than two orders of mag-
nitude in both the DTNB and the Trx-assay), the major primary site of
GoPI-sugar action seems to be the reactive selenocysteine residue at
the C-terminal active site of hTrxR. In hTrxR (or S. mansoni TGR) the
selenocysteine residue has been shown to be involved in the transfer of
gold from phosphine ligands in GoPI complexes to the redox-active Cys
pair of the enzymes. In TGR catalysis, a transient intermediate bearing
an Au(I) coordinated with Sec597 and triethylphosphine, upon the
prior release of acetoxythioglucose, has been proposed. Subsequently,
the intermediate species releases triethylphosphine to form a Sec-gold-
Cys complex with any of the available cysteinyl residues, which
can rearrange to yield the more stable short-distance Cys–gold–Cys
complexes when equilibrium is reached [31]. As studied in detail
by Urig et al. 2006 [3], the inhibition pattern of disulfide reductases
by goldphospholes seems to be very complex and can be described
by an initial competition with the oxidizing substrate followed by an
irreversible reaction leading to covalently modified enzyme. Also the
substoichiometric inhibition of hTrxR which became very evident in
our studies on GoPI-sugar had already been reported in Urig et al. for
GoPI [3]. Although the underlying mechanism is not fully understood,
we hypothesize that this substoichiometric inhibition is based on various
overlaying effects which include: (i) interaction of substrates and inhib-
itors with both N-terminal and C-terminal active sites; (ii) competitive
(reversible) and covalent (irreversible) inhibition; and (iii) most of all,
enhanced instability of hTrxR in the presence of NADPH and inhibitor
which becomes critical over the incubation time of 10 min.
3.7. Inhibition of recombinant redox enzymes by GoPI-sugar
To test the interaction between recombinant redox enzymes
and GoPI-sugar, we measured the half-maximal inhibition induced
by 10 min incubation of the prereduced enzymes with GoPI-sugar on
recombinant human GR, human TrxR isolated from placenta, and the re-
combinant selenocysteine-lacking mutant hTrxRU498C in vitro (Table 3,
Fig. 7). The data can be directly compared to the values previously
reported for GoPI (Table 3, [3]). Under the conditions chosen, GoPI-
sugar inhibits human GR with an IC50 of 88.5 nM, and is thus less efficient
than GoPI which inactivated the enzyme at an almost stoichiometric
concentration of 2 nM [3]. In good accordance, auranofin, a phosphine-
coordinated gold(I) thiosugar complex, does not show a pronounced in-
hibition of hGR [55], indicating that the presence of a thiosugar moiety
decreases the affinity of the inhibitor for hGR. Eventually, the access of
the enlarged inhibitor to the active site of GR and the subsequent attack
of residue Cys58 leading to the formation of a (Cys58–S)–Au–phosphole
intermediate might be hindered by the thiosugar. After successful access,
the thioglucose might be replaced by the dithiol target as reported
for auranofin [41]. However, combining GoPI with a thiosugar slightly
Please cite this article as: E. Jortzik, et al., Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor: Chemical synthesis, mechanistic
studies, and effectiveness in vivo, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbapap.2014.01.006

E. Jortzik et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx
9
Fig. 6. Apoptosis in tumor sections from C6 glioma rats (d16) treated with GoPI-sugar and untreated controls. (A) Tumor sections from each of the groups were examined using
TUNEL staining. (B) Quantitative analysis of apoptotic cells per mm2 in treated tumors and controls. Positive cells were counted in three hotspots per tumor sample. A comparison of
TUNEL-positive cells per mm2 shows no significant difference between treated and untreated tumors.
3.8. Tissue-specific variation of redox parameters after
GoPI-sugar treatment
GST π, was reported to be elevated in malignant tumors [57]. A high ex-
pression of GST π is associated with a shorter survival time of glioma pa-
tients and is related to more aggressive and/or chemoresistant tumors
[58]. Upon treatment with certain anticancer agents, the expression of
different GST isoforms in human brain tumors is induced [19]. However,
GoPI-sugar treatment did not influence the levels of GST activity in brain
tumor tissue in our rat glioma model (Table 2), pointing towards a dif-
ferent mode of action and a lower probability of resistance development
against GoPI-sugar, an aspect that is worth studying in detail.
The activities of the selenoproteins TrxR and GPx were reduced
in most tissues after treatment with GoPI-sugar, however, not sig-
nificantly. Significantly higher TrxR activities in GoPI-sugar-treated
rats compared to untreated controls were observed in the kidney and
liver, which might be due to compensatory upregulation of the enzymes
in these organs involved in detoxification, metabolization, and excre-
tion of xenobiotics. Interestingly, contrary to the fact that GoPI-sugar
inhibits hTrxR with an IC50 of 4.3 nM in vitro (Section 3.7), brain
tumor tissue did not show any decrease in TrxR activity after treatment.
Again, this result might be due to a stimulation of TrxR activity under
drug treatment leading to a compensatory upregulation of transcription
in parallel to the inhibition. A similar phenomenon has been observed
previously when TrxR activity was determined in glioblastoma cells
treated with terpyridine platinum complexes as TrxR inhibitors [43].
Notably, in our experiments, TrxR activity was significantly lower in
untreated rat brain tumors when compared to untreated healthy brains
(Table 2). This is a somewhat unexpected finding, since different human
tumor cell lines have elevated expression levels and/or increased
activity of TrxR and Trx such as human lung adenocarcinoma cells
[59], human colorectal tumors, breast cancer cells [60], and thyroid
cancer cells [61], promoting tumor growth and progression. Similarly,
Different gold(I) complexes including GoPI were shown to affect the
redox system of cancer cells by inhibiting thioredoxin reductase and/or
glutathione reductase via their active site thiols [3,4,6,8]. In order to
systematically address the effect of GoPI-sugar treatment on the redox
status of C6 glioma rats, we measured the activities of central redox en-
zymes such as TrxR, GPx, GR, and GST, as well as concentrations of total
glutathione in the blood, five different rat tissues including the brain of
C6 glioma rats treated with GoPI-sugar, and untreated control animals.
Data are summarized in Table 2. Interestingly, total glutathione concen-
trations were only significantly reduced in the kidney, while the other
organs including the brain showed no significant alterations in total
glutathione concentration. Moreover, healthy brain and brain tumor
tissue from rats contain the same levels of glutathione. On the contrary,
Kudo et al. [56] determined significantly decreased glutathione levels in
glioblastoma compared to normal human brain tissue (195 μg/g and
444 μg/g, respectively). However, total glutathione concentrations do
not give any information on the glutathione redox potential. Examining
changes in the ratio of reduced to oxidized glutathione (e.g. by genetically
encoded probes) will allow a more precise statement on the redox state of
the cells following GoPI-sugar treatment.
GST activity was significantly increased in the blood of rats treated
with GoPI-sugar. The observed upregulation of GST most likely occurs
due to increased drug pressure. GSTs are phase II enzymes that detoxify
xenobiotics including anticancer drugs by conjugating the molecules
with GSH, which are subsequently exported from the cell. GSTs have
been implicated in resistance against cancer therapeutics and the lack-
ing clinical response of brain tumors. The major isoform in brain tissue,
Please cite this article as: E. Jortzik, et al., Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor: Chemical synthesis, mechanistic
studies, and effectiveness in vivo, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbapap.2014.01.006

10
E. Jortzik et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx
Fig. 7. Inhibition studies with GoPI-sugar on hGR (A), hTrxR wild type (B) and hTrxRU498C (C). For hTrxR, the inhibition has been tested by using both thioredoxin and DTNB as a substrate.
One representative graph of each IC50 determination is shown.
TrxR activity in glioblastoma tissue from human patients was found
to be significantly higher than in the control group (74.5 U/g wet tissue
and 14.8 U/g wet tissue, respectively) [62]. The comparatively low TrxR
activity in the tumor cells studied here might indicate a higher variabil-
ity of glioblastoma cells than originally expected. However, concerning
its therapeutic effects, the GoPI-sugar with its high affinity for TrxR
is likely to be even more effective in tumor cells with upregulated
TrxR activity.
In our in vivo study, the investigated redox parameters in brain
tissue were hardly affected by the treatment. This might partially be
explained by compensatory upregulation of redox-related enzymes.
However, based on these results, it can be assumed that the inhibition
of brain tumor growth by GoPI-sugar is based on its increased stability
with respect to nucleophiles and on its DNA intercalating properties
rather than on the inhibition of redox enzymes. As previously investi-
gated [63], the high affinity of gold(I) for sulfur and selenium ligands
Table 2
Redox parameters in different tissues of C6 glioma rats after treatment with GoPI-sugar (30 mg/kg, d16).
Group
Brain tissue
0.3
Brain tumor
0.3
Kidney
Muscle
Liver
Blood
Glutathione (mM)
Control
Treated
Control
Treated
Control
Treated
Control
Treated
Control
Treated
0
0
0.3 0.1
0.1
28.7 1.3
11.1
30.6 17
36.1 6.7
630 57.9
568 187
31.1 8.8
26.4 5.6
0.4 0.1
0.5 0.1
2.9 0.5
2.2 0.7
1.1 0.4
1.1 0.4
n.d.
n.d.
16.1 12.2
⁎
0.4 0.1
0.3 0.1
18.3 1.2
19.9 4.1
79.9 8.9
83.2 21
96.5 11
104 55.7
131 20.6
195 216
0
TrxR (mU/mg protein)
GR (mU/mg protein)
GPx (mU/mg protein)
GST (mU/mg protein)
26.7
0
17.4 7.0
13.6 1.5
17.5 6.3
22.5 7.5
117 57.3
86.7 10
39.5 17
23.3 2.2
26.5 5.9
17.1 1.8⁎
115 20.7
103 14.4
869 369
485 70.5
317 51
⁎
24.6 4.9
4
59.6
61.9 10
74.2
71.1 5.8
131
122 25.1
0
16.9
0
0
381 59.6
468 199
6.9 1.5
0
⁎
353 20.3
10.5 0.9
c — untreated control: n = 3 except in healthy brain (n = 1).
t — treated with 30 mg GoPI-sugar/kg body weight: n = 5 except in healthy brain (n = 4) and muscle (n = 3).
n.d. — not determined.
⁎
Significant differences between untreated and treated animals (p b 0.05).
Please cite this article as: E. Jortzik, et al., Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor: Chemical synthesis, mechanistic
studies, and effectiveness in vivo, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbapap.2014.01.006

E. Jortzik et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx
11
[3] S. Urig, K. Fritz-Wolf, R. Reau, C. Herold-Mende, K. Toth, E. Davioud-Charvet, K.
Becker, Undressing of phosphine gold(I) complexes as irreversible inhibitors
of human disulfide reductases, Angew. Chem. Int. Ed. Engl. 45 (2006)
1881–1886.
Table 3
Half-maximal inhibition of disulfide reductases by GoPI and GoPI-sugar after 10 min
preincubation of the NADPH-reduced enzymes with inhibitor.
[4] M. Deponte, S. Urig, L.D. Arscott, K. Fritz-Wolf, R. Reau, C. Herold-Mende, S.
Koncarevic, M. Meyer, E. Davioud-Charvet, D.P. Ballou, C.H. Williams Jr., K. Becker,
Mechanistic studies on a novel, highly potent gold-phosphole inhibitor of human
glutathione reductase, J. Biol. Chem. 280 (2005) 20628–20637.
[5] E. Vergara, A. Casini, F. Sorrentino, O. Zava, E. Cerrada, M.P. Rigobello, A. Bindoli, M.
Laguna, P.J. Dyson, Anticancer therapeutics that target selenoenzymes: synthesis,
characterization, in vitro cytotoxicity, and thioredoxin reductase inhibition
Enzyme
Oxidizing substrate
IC0.5 [nM]
GoPIa
GoPI-sugar
hGR
hTrxR
GSSG
2
7
1
88.5 28
4.3 1.6
0.49 0.04
834.9 74.8
66.8 5.8
hTrxC72S
DTNB
hTrxRU498C
hTrxC72S
DTNB
1900
3000
of
a series of gold(I) complexes containing hydrophilic phosphine ligands,
ChemMedChem 5 (2010) 96–102.
[6] V. Gandin, A.P. Fernandes, M.P. Rigobello, B. Dani, F. Sorrentino, F. Tisato, M.
Bjornstedt, A. Bindoli, A. Sturaro, R. Rella, C. Marzano, Cancer cell death induced
by phosphine gold(I) compounds targeting thioredoxin reductase, Biochem.
Pharmacol. 79 (2010) 90–101.
a
Values as reported in Ref. [3].
[7] R. Galassi, A. Burini, S. Ricci, M. Pellei, M.P. Rigobello, A. Citta, A. Dolmella, V. Gandin,
C. Marzano, Synthesis and characterization of azolate gold(I) phosphane complexes
as thioredoxin reductase inhibiting antitumor agents, Dalton Trans. 41 (2012)
5307–5318.
[8] E. Schuh, C. Pfluger, A. Citta, A. Folda, M.P. Rigobello, A. Bindoli, A. Casini, F. Mohr,
Gold(I) carbene complexes causing thioredoxin 1 and thioredoxin 2 oxidation as
potential anticancer agents, J. Med. Chem. 55 (2012) 5518–5528.
suggests that gold thiosugar will be displaced by proteins in the blood,
including redox enzymes and transport proteins, and be primarily
bound to proteins as part of the mechanism of protein-mediated trans-
port of gold to the final targets during therapy.
[9] A. Holmgren, J. Lu, Thioredoxin and thioredoxin reductase: current research with
special reference to human disease, Biochem. Biophys. Res. Commun. 396 (2010)
120–124.
4. Conclusion
[10] H. Hafsi, P. Hainaut, Redox control and interplay between p53 isoforms: roles in the
regulation of basal p53 levels, cell fate, and senescence, Antioxid. Redox Signal. 15
(2011) 1655–1667.
[11] H. Haapasalo, M. Kylaniemi, N. Paunu, V.L. Kinnula, Y. Soini, Expression of
antioxidant enzymes in astrocytic brain tumors, Brain Pathol. 13 (2003) 155–164.
[12] G. Powis, D. Mustacich, A. Coon, The role of the redox protein thioredoxin in cell
growth and cancer, Free Radic. Biol. Med. 29 (2000) 312–322.
[13] A. Yokomizo, M. Ono, H. Nanri, Y. Makino, T. Ohga, M. Wada, T. Okamoto, J. Yodoi, M.
Kuwano, K. Kohno, Cellular levels of thioredoxin associated with drug sensitivity
to cisplatin, mitomycin C, doxorubicin, and etoposide, Cancer Res. 55 (1995)
4293–4296.
[14] R.B. Penney, D. Roy, Thioredoxin-mediated redox regulation of resistance to
endocrine therapy in breast cancer, Biochim. Biophys. Acta 1836 (2013) 60–79.
[15] M. Selenius, M. Hedman, D. Brodin, V. Gandin, M.P. Rigobello, J. Flygare, C. Marzano,
A. Bindoli, O. Brodin, M. Bjornstedt, A.P. Fernandes, Effects of redox modulation by
inhibition of thioredoxin reductase on radiosensitivity and gene expression, J. Cell.
Mol. Med. 16 (2012) 1593–1605.
Gold compounds exhibit a high affinity for sulfur- and selenium-
containing ligands such as proteins with selenocysteine as exposed
active site residues. By combining gold(I) with a nucleophilic phosphole
ligand with limited aromatic character, the reactivity with thiol-
containing proteins is even more pronounced [3,4]. Here, we show
that a (phosphole)gold chloride complex linked to a thiosugar inhibits
the growth of a glioma cell line with IC50 values in a low micromolar
range. Interestingly, introducing the sugar residue did improve the sol-
ubility of GoPI-sugar and its antiproliferative effect when compared to
GoPI, thereby representing a major improvement of the compound.
Treatment of C6 rats bearing malignant glioblastomas with GoPI-sugar
was well tolerated and led to a drastic decrease in tumor growth. In con-
trast to previous studies on gold(I) complexes, targeting the thioredoxin
system and disturbing the redox system in cancer cells does not seem to
be the major mode of action of GoPI-sugar, since the activities of hTrxR
and other redox parameters were only slightly altered in the brains of
the treated animals. Our observations rather point towards inhibition
of DNA replication via the interaction of GoPI-sugar with DNA, and/or
inhibition of topoisomerase I, features that deserve to be studied in
more detail. Due to its high efficacy against glioblastomas in vivo and
its low toxicity, GoPI-sugar is a promising lead compound for future
anticancer therapies.
[16] S. Urig, K. Becker, On the potential of thioredoxin reductase inhibitors for cancer
therapy, Semin. Cancer Biol. 16 (2006) 452–465.
[17] H.W. Lim, S. Hong, W. Jin, S. Lim, S.J. Kim, H.J. Kang, E.H. Park, K. Ahn, C.J. Lim,
Up-regulation of defense enzymes is responsible for low reactive oxygen species
in malignant prostate cancer cells, Exp. Mol. Med. 37 (2005) 497–506.
[18] J.M. Estrela, A. Ortega, E. Obrador, Glutathione in cancer biology and therapy,
Crit. Rev. Clin. Lab. Sci. 43 (2006) 143–181.
[19] D.S. Backos, C.C. Franklin, P. Reigan, The role of glutathione in brain tumor drug
resistance, Biochem. Pharmacol. 83 (2012) 1005–1012.
[20] K. Fritz-Wolf, S. Urig, K. Becker, The structure of human thioredoxin reductase 1
provides insights into C-terminal rearrangements during catalysis, J. Mol. Biol. 370
(2007) 116–127.
[21] S.N. Savvides, M. Scheiwein, C.C. Bohme, G.E. Arteel, P.A. Karplus, K. Becker, R.H.
Schirmer, Crystal structure of the antioxidant enzyme glutathione reductase
inactivated by peroxynitrite, J. Biol. Chem. 277 (2002) 2779–2784.
[22] C. Hay, M. Hissler, C. Fischmeister, J. Rault-Berthelot, L. Toupet, L. Nyulaszi, R.
Reau, Phosphole-containing pi-conjugated systems: from model molecules to
polymer films on electrodes, Chemistry 7 (2001) 4222–4236.
[23] J.M. Madeira, D.L. Gibson, W.F. Kean, A. Klegeris, The biological activity of auranofin:
implications for novel treatment of diseases, Inflammopharmacology 20 (2012)
297–306.
[24] A. Debnath, D. Parsonage, R.M. Andrade, C. He, E.R. Cobo, K. Hirata, S. Chen, G.
Garcia-Rivera, E. Orozco, M.B. Martinez, S.S. Gunatilleke, A.M. Barrios, M.R. Arkin,
L.B. Poole, J.H. McKerrow, S.L. Reed, A high-throughput drug screen for Entamoeba
histolytica identifies a new lead and target, Nat. Med. 18 (2012) 956–960.
[25] C.K. Mirabelli, R.K. Johnson, D.T. Hill, L.F. Faucette, G.R. Girard, G.Y. Kuo, C.M. Sung,
S.T. Crooke, Correlation of the in vitro cytotoxic and in vivo antitumor activities of
gold(I) coordination complexes, J. Med. Chem. 29 (1986) 218–223.
[26] M.T. Coffer, C.F. Shaw, M.K. Eidsness, J.W. Watkins, R.C. Elder, Reactions of auranofin
and Et3paucl with bovine serum-albumin, Inorg. Chem. 25 (1986) 333–339.
[27] M.T. Coffer, C.F. Shaw, A.L. Hormann, C.K. Mirabelli, S.T. Crooke, Thiol competition
for Et3paus-albumin — a nonenzymatic mechanism for Et3po formation, J. Inorg.
Biochem. 30 (1987) 177–187.
Acknowledgements
The authors thank Radoslav Enchev for the DNA binding studies and
Marina Fischer for her excellent technical assistance. Parts of this work
were supported by the Deutsche Forschungsgemeinschaft (BE 1540/
11-2 to K.B. and JO 1085/1-2 to E.J.), by the Centre National de la
Recherche Scientifique, France (CNRS, France, UMR 7509 [E.D.–C.]),
and the Laboratoire d'Excellence (LabEx) ParaFrap (grant LabEx ParaFrap
ANR-11-LABX-0024 [E.D.–C.]).
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.bbapap.2014.01.006.
[28] J. Christodoulou, P.J. Sadler, A. Tucker, H-1-Nmr of albumin in human blood-plasma —
drug-binding and redox reactions at Cys(34), FEBS Lett. 376 (1995) 1–5.
[29] J. Zou, P. Taylor, J. Dornan, S.P. Robinson, M.D. Walkinshaw, P.J. Sadler, First crystal
structure of a medicinally relevant gold protein complex: unexpected binding of
[Au(PEt3)](+) to histidine, Angew. Chem. Int. Ed. 39 (2000) 2931–2934.
[30] C.E. Blank, J.C. Dabrowiak, Absorption and circular dichroism studies of a gold (I)–
DNA complex, J. Inorg. Biochem. 21 (1984) 21–29.
References
[1] P.C. Bruijnincx, P.J. Sadler, New trends for metal complexes with anticancer activity,
Curr. Opin. Chem. Biol. 12 (2008) 197–206.
[2] S. Komeda, A. Casini, Next-generation anticancer metallodrugs, Curr. Top. Med.
Chem. 12 (2012) 219–235.
[31] F. Angelucci, A.A. Sayed, D.L. Williams, G. Boumis, M. Brunori, D. Dimastrogiovanni,
A.E. Miele, F. Pauly, A. Bellelli, Inhibition of Schistosoma mansoni thioredoxin–
Please cite this article as: E. Jortzik, et al., Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor: Chemical synthesis, mechanistic
studies, and effectiveness in vivo, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbapap.2014.01.006

12
E. Jortzik et al. / Biochimica et Biophysica Acta xxx (2014) xxx–xxx
glutathione reductase by auranofin: structural and kinetic aspects, J. Biol. Chem. 284
(2009) 28977–28985.
[32] F. Saccoccia, F. Angelucci, G. Boumis, M. Brunori, A.E. Miele, D.L. Williams, A. Bellelli,
On the mechanism and rate of gold incorporation into thiol-dependent flavoreductases,
J. Inorg. Biochem. 108 (2012) 105–111.
[47] A. Nordhoff, U.S. Bucheler, D. Werner, R.H. Schirmer, Folding of the four domains
and dimerization are impaired by the Gly446–NGlu exchange in human glutathione
reductase. Implications for the design of antiparasitic drugs, Biochemistry 32 (1993)
4060–4066.
[48] S. Gromer, L.D. Arscott, C.H. Williams Jr., R.H. Schirmer, K. Becker, Human
placenta thioredoxin reductase. Isolation of the selenoenzyme, steady state
kinetics, and inhibition by therapeutic gold compounds, J. Biol. Chem. 273 (1998)
20096–20101.
[49] S. Urig, J. Lieske, K. Fritz-Wolf, A. Irmler, K. Becker, Truncated mutants of
human thioredoxin reductase 1 do not exhibit glutathione reductase activity, FEBS
Lett. 580 (2006) 3595–3600.
[50] S.T. Crooke, C.K. Mirabelli, Molecular mechanisms of action of auranofin and
other gold complexes as related to their biologic activities, Am. J. Med. 75
(1983) 109–113.
[51] C.K. Mirabelli, C.M. Sung, J.P. Zimmerman, D.T. Hill, S. Mong, S.T. Crooke,
Interactions of gold coordination complexes with DNA, Biochem. Pharmacol.
35 (1986) 1427–1433.
[52] C.R. Wilson, A.M. Fagenson, W. Ruangpradit, M.T. Muller, O.Q. Munro, Gold(III)
complexes of pyridyl- and isoquinolylamido ligands: structural, spectroscopic, and
biological studies of a new class of dual topoisomerase I and II inhibitors, Inorg.
Chem. 52 (2013) 7889–7906.
[33] A.N. Kuntz, E. Davioud-Charvet, A.A. Sayed, L.L. Califf, J. Dessolin, E.S. Arner, D.L.
Williams, Thioredoxin glutathione reductase from Schistosoma mansoni: an
essential parasite enzyme and a key drug target, PLoS Med. 4 (2007) e206.
[34] Y. Wang, Y. Wang, X. Chen, Z. Zhang, S. Li, B. Chen, C. Wu, L. Wang, X. Zhang, J. Wang,
L. Chen, T. Jiang, Comparison of the clinical efficacy of temozolomide (TMZ) versus
nimustine (ACNU)-based chemotherapy in newly diagnosed glioblastoma,
Neurosurg. Rev. 47 (2014) 73–78.
[35] R. Stupp, M.E. Hegi, W.P. Mason, M.J. van den Bent, M.J. Taphoorn, R.C. Janzer, S.K.
Ludwin, A. Allgeier, B. Fisher, K. Belanger, P. Hau, A.A. Brandes, J. Gijtenbeek, C.
Marosi, C.J. Vecht, K. Mokhtari, P. Wesseling, S. Villa, E. Eisenhauer, T. Gorlia, M.
Weller, D. Lacombe, J.G. Cairncross, R.O. Mirimanoff, Effects of radiotherapy with
concomitant and adjuvant temozolomide versus radiotherapy alone on survival in
glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC
trial, Lancet Oncol. 10 (2009) 459–466.
[36] S.J. Berners-Price, A. Filipovska, Gold compounds as therapeutic agents for human
diseases, Metallomics 3 (2011) 863–873.
[37] A. Casini, L. Messori, Molecular mechanisms and proposed targets for selected anti-
cancer gold compounds, Curr. Top. Med. Chem. 11 (2011) 2647–2660.
[38] R. Rubbiani, S. Can, I. Kitanovic, H. Alborzinia, M. Stefanopoulou, M. Kokoschka, S.
Monchgesang, W.S. Sheldrick, S. Wolfl, I. Ott, Comparative in vitro evaluation of
N-heterocyclic carbene gold(I) complexes of the benzimidazolylidene type, J. Med.
Chem. 54 (2011) 8646–8657.
[39] A. Meyer, C.P. Bagowski, M. Kokoschka, M. Stefanopoulou, H. Alborzinia, S. Can, D.H.
Vlecken, W.S. Sheldrick, S. Wolfl, I. Ott, On the biological properties of alkynyl
phosphine gold(I) complexes, Angew. Chem. Int. Ed. Engl. 51 (2012) 8895–8899.
[40] C. Wetzel, P.C. Kunz, M.U. Kassack, A. Hamacher, P. Bohler, W. Watjen, I. Ott, R.
Rubbiani, B. Spingler, Gold(I) complexes of water-soluble diphos-type ligands:
synthesis, anticancer activity, apoptosis and thioredoxin reductase inhibition,
Dalton Trans. 40 (2011) 9212–9220.
[53] J.J. Yan, A.L. Chow, C.H. Leung, R.W. Sun, D.L. Ma, C.M. Che, Cyclometalated gold(III)
complexes with N-heterocyclic carbene ligands as topoisomerase I poisons, Chem.
Commun. (Camb.) 46 (2010) 3893–3895.
[54] O. Rackham, S.J. Nichols, P.J. Leedman, S.J. Berners-Price, A. Filipovska, A gold(I)
phosphine complex selectively induces apoptosis in breast cancer cells: implications
for anticancer therapeutics targeted to mitochondria, Biochem. Pharmacol. 74
(2007) 992–1002.
[55] S. Prast-Nielsen, M. Cebula, I. Pader, E.S. Arner, Noble metal targeting
of thioredoxin reductase–covalent complexes with thioredoxin and thioredoxin-
related protein of 14 kDa triggered by cisplatin, Free Radic. Biol. Med. 49 (2010)
1765–1778.
[56] H. Kudo, T. Mio, T. Kokunai, N. Tamaki, K. Sumino, S. Matsumoto, Quantitative
analysis of glutathione in human brain tumors, J. Neurosurg. 72 (1990) 610–615.
[57] R.C. Strange, A.A. Fryer, B. Matharoo, L. Zhao, J. Broome, D.A. Campbell, P. Jones, I.C.
Pastor, R.V. Singh, The human glutathione S-transferases: comparison of isoenzyme
expression in normal and astrocytoma brain, Biochim. Biophys. Acta 1139 (1992)
222–228.
[41] E. Viry, E. Battaglia, V. Deborde, T. Muller, R. Reau, E. Davioud-Charvet, D. Bagrel,
A sugar-modified phosphole gold complex with antiproliferative properties acting
as a thioredoxin reductase inhibitor in MCF-7 cells, ChemMedChem 3 (2008)
1667–1670.
[42] S. Karcher, H.H. Steiner, R. Ahmadi, S. Zoubaa, G. Vasvari, H. Bauer, A. Unterberg, C.
Herold-Mende, Different angiogenic phenotypes in primary and secondary glioblas-
tomas, Int. J. Cancer 118 (2006) 2182–2189.
[58] A. Korshunov, A. Golanov, R. Sycheva, I. Pronin, Prognostic value of tumour associated
antigen immunoreactivity and apoptosis in cerebral glioblastomas: an analysis of 168
cases, J. Clin. Pathol. 52 (1999) 574–580.
[43] K. Becker, C. Herold-Mende, J.J. Park, G. Lowe, R.H. Schirmer, Human thioredoxin
reductase is efficiently inhibited by (2,2′:6′,2″ -terpyridine)platinum(II) complexes.
Possible implications for a novel antitumor strategy, J. Med. Chem. 44 (2001)
2784–2792.
[44] R. Ahmadi, S. Urig, M. Hartmann, B.M. Helmke, S. Koncarevic, B. Allenberger, C.
Kienhoefer, M. Neher, H.H. Steiner, A. Unterberg, C. Herold-Mende, K. Becker,
Antiglioma activity of 2,2′:6′,2″-terpyridineplatinum(II) complexes in a rat
model—effects on cellular redox metabolism, Free Radic. Biol. Med. 40 (2006)
763–778.
[45] I. Ott, H. Scheffler, R. Gust, Development of a method for the quantification of
the molar gold concentration in tumour cells exposed to gold-containing drugs,
ChemMedChem 2 (2007) 702–707.
[46] S.M. Kanzok, S. Rahlfs, K. Becker, R.H. Schirmer, Thioredoxin, thioredoxin reductase,
and thioredoxin peroxidase of malaria parasite Plasmodium falciparum, Methods
Enzymol. 347 (2002) 370–381.
[59] T. Tamura, T.C. Stadtman, A new selenoprotein from human lung adenocarcinoma
cells: purification, properties, and thioredoxin reductase activity, Proc. Natl. Acad.
Sci. U. S. A. 93 (1996) 1006–1011.
[60] M. Berggren, A. Gallegos, J.R. Gasdaska, P.Y. Gasdaska, J. Warneke, G. Powis,
Thioredoxin and thioredoxin reductase gene expression in human tumors
and cell lines, and the effects of serum stimulation and hypoxia, Anticancer
Res. 16 (1996) 3459–3466.
[61] D.T. Lincoln, F. Al-Yatama, F.M. Mohammed, A.G. Al-Banaw, M. Al-Bader, M.
Burge, F. Sinowatz, P.K. Singal, Thioredoxin and thioredoxin reductase expres-
sion in thyroid cancer depends on tumour aggressiveness, Anticancer Res. 30
(2010) 767–775.
[62] R. Kemerdere, T. Kacira, H. Hanimoglu, M. Kucur, T. Tanriverdi, B. Canbaz,
Tissue and plasma thioredoxin reductase expressions in patients with
glioblastoma multiforme, J. Neurol. Surg. A 74 (2013) 234–238.
[63] I.C. Shaw, Gold-based therapeutic agents, Chem. Rev. 99 (1999) 2589–2600.
Please cite this article as: E. Jortzik, et al., Antiglioma activity of GoPI-sugar, a novel gold(I)–phosphole inhibitor: Chemical synthesis, mechanistic
studies, and effectiveness in vivo, Biochim. Biophys. Acta (2014), http://dx.doi.org/10.1016/j.bbapap.2014.01.006