A soluble bis-chelated gold(I) diphosphine compound with strong anticancer activity and low toxicity

Article abstract of DOI:10.1021/jm3009822

Gold-containing compounds have shown anticancer potential, but their clinical applications have been severely limited by poor stability and high toxicity in vivo. Here, we report a novel soluble bis-chelated gold(I)-diphosphine compound (GC20) with strong anticancer activity and low toxicity. GC20 shows strong antiproliferation potency against a broad spectrum of cancer cell lines including cisplatin-resistant cancer cells (IC₅₀ ≈ 0.5 μM) and significantly reduces tumor growth in several tumor xenografts in mouse models at doses as low as 2 mg/kg. Studies of its mechanism revealed that GC20 specifically inhibits the enzymatic activity of thioredoxin reductase by binding to selenocysteine residue, without targeting other well-known selenol and thiol groups contained in biomolecules. Remarkably, in animal studies GC20 was shown to be well tolerated even at the high dose of 8 mg/kg. Our results strongly suggest that GC20 represents a promising candidate for the development of novel anticancer drugs.

Full text of DOI:10.1021/jm3009822

Article
pubs.acs.org/jmc
A Soluble Bis-Chelated Gold(I) Diphosphine Compound with Strong
Anticancer Activity and Low Toxicity
‡
,∥
†,§,∥
‡
†
†
§
,§
Yanli Wang, Minyu Liu,
Ran Cao, Wanbin Zhang, Ming Yin, Xuhua Xiao, Quanhai Liu,*
,
‡
and Niu Huang*
†
School of Pharmacy, Shanghai Jiao Tong University, 800 Dong Chuan Road, Shanghai 200240, China
‡
National Institute of Biological Sciences, Beijing, 7 Science Park Road, Zhongguancun Life Science Park, Beijing 102206, China
§Department of Pharmacology, Shanghai Institute of Pharmaceutical Industry, 1111 Zhong Shan Bei Yi Road, Shanghai 200437,
China
*
S Supporting Information
ABSTRACT: Gold-containing compounds have shown anticancer potential, but their clinical applications have been severely
limited by poor stability and high toxicity in vivo. Here, we report a novel soluble bis-chelated gold(I)−diphosphine compound
(
GC20) with strong anticancer activity and low toxicity. GC20 shows strong antiproliferation potency against a broad spectrum
of cancer cell lines including cisplatin-resistant cancer cells (IC50 ≈ 0.5 μM) and significantly reduces tumor growth in several
tumor xenografts in mouse models at doses as low as 2 mg/kg. Studies of its mechanism revealed that GC20 specifically inhibits
the enzymatic activity of thioredoxin reductase by binding to selenocysteine residue, without targeting other well-known selenol
and thiol groups contained in biomolecules. Remarkably, in animal studies GC20 was shown to be well tolerated even at the high
dose of 8 mg/kg. Our results strongly suggest that GC20 represents a promising candidate for the development of novel
anticancer drugs.
INTRODUCTION
some cysteine (Cys) and selenocysteine (Sec) -dependent
enzymes might be their therapeutic targets. For example,
auranofin was reported to strongly inhibit two Sec-containing
■
1
Metal-containing compounds such as cisplatin and carboplatin
have long been used clinically as potential cancer chemotherapy
1
6
2
−4
enzymes, thioredoxin reductase (TrxR) and glutathione
agents.
However, the therapeutic effects of these agents are
1
7,18
peroxidase (GPx).
Auranofin was also shown to possess
not universal and their side effects, most noticeably systemic
toxicity, have limited their widespread application. Moreover,
acquired resistance in long-term treatment has frequently been
observed, which might be partially attributed to irreversible
only modest inhibitory effect on glutathione reductase (GR),
despite the similarity of GR and TrxR in their structure and
19
function. This was attributed to the higher affinity of auranofin
to the Sec residue in selenoproteins than the cysteine in GR.
However, this selectivity may have been much less expressed
because of the widespread existence of serum albumin in vivo,
which was found to react with auranofin by forming the putative
1
,4−6
consumption by enhanced glutathione (GSH) content.
Nevertheless, the success of cisplatin has stimulated a broad
interest in the exploration of compounds containing metals, such
as nickel, ruthenium, and gold, as potential anticancer agents with
20
4
Au(I)−S adduct. In fact, many gold(I) compounds were
unique pharmacologic, kinetic, and geometric properties.
discovered as binders of multiple S-donor proteins. It was
Gold-containing compounds, exemplified by the first-gen-
eration gold-based drug auranofin, have been used in the7
treatment of rheumatoid arthritis (RA) for many years.
Recently, gold complexes have gained increasing attention in
cancer therapy for their strong antiproliferation potency against
reported that [Au(d2pypp) ]Cl can inhibit the activities of TrxR
2
and its substrate, thioredoxin (Trx) by binding to the Cys pair at
Received: July 7, 2012
4
,7−15
cancer cells.
Studies of their mechanism indicated that
©
XXXX American Chemical Society
A
dx.doi.org/10.1021/jm3009822 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. In vitro inhibition of TrxR activity by GC20 via binding to the C-terminal redox-active site. DTNB reduction assay was used to examine
enzyme activities. (A) Structures of GC20, phosphinoquinoxaline, and cisplatin. (B) In vitro inhibition ability of GC20 and cisplatin. (C) Inhibitory
ability of GC20 on the activity of hGR, determined with GSSG and insulin as substrates. Measurement of GC20 inhibition on hTrxRΔ8 and hTrxRΔ16
was performed via DTNB reduction assay.
2
1
the active site. It was also shown that [1-phenyl-2,5-di(2-
pyridyl)phosphole]AuCl can interact with GR in addition to
TrxR. This was confirmed from the crystal structure of covalently
complex, GC20, was identified as the most promising one in
terms of both its structure and activity. Specifically, we were able
to show that GC20 has a strong growth inhibition property for a
broad spectrum of cancer cell lines, including cisplatin-resistant
ones. We report here our studies about the molecular target and
the antiproliferation and anticancer potential of GC20. More
importantly, we have found that GC20 shows a remarkable
selectivity between cancer cells and normal cells as well as lower
toxicity in animal models as compared to cisplatin. Our results
suggest that GC20 has advantages over cisplatin and other metal-
containing compounds in overcoming resistance, severe toxicity,
and overreactivity and could potentially become a promising
candidate for a new generation of gold(I)-based anticancer drugs.
22
modified GR with S−Au(I)−S coordination. The high affinity
of gold(I) compounds to thiols can lead to high toxicity and low
efficacy in vivo due to the “off-target” effect. The quest for novel
gold(I) compounds as therapeutically useful agents has received
very limited success and is restricted only to animal studies. This
is further borne out by the fact that no alternatives of gold(I)7-
containing compounds to cisplatin have been marketed to date.
Among the targets mentioned above, TrxR has many
advantages for the development of selective anticancer agents.
TrxR plays important roles in many intracellular processes, such
as regulation of redox metabolism, DNA synthesis, cell
23,24
RESULTS
proliferation, signaling, and apoptosis.
evidence that TrxR is overexpressed and constitutively active in
a variety of human tumors, and it is associated with tumor growth
There is ample
■
Discovery of GC20. We originally performed antiprolifera-
tion screening of 16 in-house gold complexes against two human
cancer cell lines: non-small-cell lung cancer (NSCLC) cells,
NCI-H460, and gastric cancer cells, BGC-823 (Table S1 in
Supporting Information). Among them, three compounds
showed strong inhibition of proliferation of the two cancer cell
lines. Subsequently we measured their anticancer potential and
toxicity in vivo with sarcoma S180 xenograft-bearing Kunming
mice. GC20, a bis-chelated gold(I) diphosphine compound
(Figure 1A), showed less toxicity and strong anticancer potential
compared with the other two compounds (Table S2 in
Supporting Information). GC20 was shown to inhibit 50% of
cell proliferation of both cell lines at a concentration of 0.5 μM
and to inhibit 80.38% of tumor growth of sarcoma S180
xenografts in vivo at a dose of 8 mg/kg without loss of body
weight at the same time.
23−27
and angiogenesis.
Knockdown of TrxR in mouse lung
carcinoma cells was reported to result in the recovery of both cell
morphology and anchorage-independent growth properties
similar to normal cells. When TrxR knockdown cells were
injected into mice, tumor progression and metastases were
dramatically reduced. These results strongly suggest that TrxR
is a promising anticancer target. More importantly, there is a Sec
residue in the exposed and accessible C-terminal active site of
28
TrxR. The pK value (∼5.3) of the functional Sec residue is
a
29
significantly lower than that of Cys (pK = 8.4), rendering a
a
potentially higher reactivity for TrxR than the cysteine-
dependent targets. It is therefore highly desirable to develop
new anticancer agents by specifically targeting the Sec residue of
TrxR.
Motivated by the great opportunities in the development of
gold compounds as potential anticancer agents, we screened 16
new gold compounds on two human cancer cell lines. Among
these compounds, a novel bis-chelated gold(I) diphosphine
GC20 Inhibits TrxR in Vitro by Binding to the C-
Terminal Redox Site. To explore potential molecular targets
for GC20, TrxR was chosen in our tests since it has been widely
30,31
reported as the target of gold(I) compounds.
We tested the
B
dx.doi.org/10.1021/jm3009822 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 2. (A) Inhibition effects of free Au(I) in the form of AuCl and phosphinoquinoxaline ligand. (B) Inhibition potential of GC20 and AuCl on
hTrxR U498C. (C, D) Ackermann−Potter plots of enzyme velocities in the presence of different concentrations of (C) GC20 and (D) AuCl. The
product TNB increased over time in a linear manner, and the velocity was calculated as the slope (increase in absorbance per second).
enzymatic inhibition of TrxR by GC20 and cisplatin (Figure 1A).
Inhibitory activity was examined by the 5,5′-dithiobis(2-
nitrobenzoic acid) (DTNB) reduction assay. The in vitro IC50
of GC20 for TrxR is about 247.6 ± 42.1 nM, whereas the IC50
value of cisplatin is 74.9 ± 16.3 nM (Figure 1B). Previous studies
have shown that TrxR has two redox-active sites, CVNCGC
concentration of 10 μM, while free Au(I) showed strong
inhibition with an IC50 value of 128.3 ± 15.7 nM (Figure 2B).
Thus, it is not likely that GC20 could dissociate into free Au(I) in
solution and subsequently inhibit TrxR.
GC20 Inhibits TrxR by a Reversible Mode of Action. We
determined the catalytic velocities of TrxR in the presence of
different concentrations of GC20 or AuCl, individually. From the
Ackermann−Potter plot results, GC20 inhibits TrxR in a
titration-dependent manner (Figure 2C), which is dramatically
different than the tight binding mode of AuCl (Figure 2D).
These results further indicate that GC20 is stable in solution and
the inhibition of TrxR is attributed to the property of GC20 as a
whole, rather than its ligand or the free Au(I).
(
59−64) in the N-terminus and GCUG (496−499) in the C-
32
terminus. To determine the potential binding site of GC20 in
TrxR, we prepared two C-terminal mutants with the last 16
(
hTrxRΔ16) or the last eight (hTrxRΔ8) amino acids truncated.
We also tested hGR because some gold(I) compounds have been
reported to covalently modify hGR with S−Au(I)−-S coordina-
2
2,33
tion.
GC20 showed no inhibition of either of the two C-
terminal mutants or of hGR (Figure 1C), suggesting that its
binding site is located at the C-terminal redox-active site in TrxR
and also suggesting that its inhibitory mechanism is different than
some reported Au(I) containing compounds.
Next, we wondered whether TrxR is reversibly inhibited by
GC20. Therefore, we studied this inhibition mechanism using
rapid dilution assay. It was shown that the incubation of TrxR
36
with an irreversible inhibitor, curcumin, resulted in formation
of an enzyme−inhibitor complex that is resistant to dilution of
the assay mixture (Figure 3A). In contrast, rapid dilution of the
GC20/TrxR mixture leads to complete recovery of TrxR
catalytic activity (Figure 3A), which suggests a reversible
inhibition mechanism of GC20. Consistently, time-course
experiments showed that the addition of GC20 resulted in an
almost immediate inhibition, and no significant difference in the
inhibition was observed during a nearly 2 h time course (Figure
3B).
To gain deeper insight into the GC20 inhibitory mechanism,
we tested the inhibition activities of the coordinating ligand of
GC20, {2,3-bis[tert-butyl(methyl)phosphino]quinoxaline}, and
the free Au(I) in the form of AuCl. The phosphinoquinoxaline
ligand alone has no inhibitory effect, while the free Au(I) shows
extremely potent inhibition (IC = 1.7 ± 0.4 nM) (Figure 2A).
5
0
Clearly, Au(I) is essential for the TrxR potency of GC20.
Previous studies have shown that free Au(I) is not selective
34,35
among different nucleophilic biomolecules.
Therefore, we
constructed a mutant of TrxR (hTrxR U498C) with a mutation
of Sec to Cys in the C-terminal active site. Interestingly, GC20
was inactive on the hTrxR U498C mutant even at high
GC20 Shows Selectivity between TrxR and Other
Potential Targets. We have shown that GC20 can inhibit
TrxR in vitro by selectively binding to Sec residue on the C-
C
dx.doi.org/10.1021/jm3009822 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
unchanged after 24 h of incubation (Figure 4E). In comparison,
absorbance of the GSH−Pt complex increases with both
increasing concentration of GSH and incubation time (Figure
4
D,F), which is consistent with previous reports that cisplatin can
40
interact with GSH.
GC20 Inhibits TrxR in Intact Cancer Cells. To determine
whether GC20 would inhibit TrxR in intact cells, we treated
human A549 tumor cell line with 2.0 μM GC20 for 6, 12, 24, and
4
8 h. Compared to the dimethyl sulfoxide (DMSO) control,
GC20 was found to reduce the cellular enzymatic activity of TrxR
by more than 40% after 6 h. Strikingly, cisplatin shows no
inhibition at all compared to the phosphate-buffered saline
(
PBS) control despite its high inhibition activity in vitro. As
expected, phosphinoquinoxaline ligand did not show any effect
on cellular TrxR (Figure 5A). We also examined the effect of
GC20 on the expression of TrxR by Western blot analysis. No
significant changes were observed in the content of TrxR in cells
treated by GC20 at different time points (Figure 5B). Clearly, the
reduction of TrxR’s activity is due to direct inhibition by GC20,
rather than the downregulation of protein expression. It has been
reported previously that 65−98% of the platinum in blood1
plasma was in protein-bound form after 1 day of administration.
The low efficacy of cisplatin on TrxR in cells that we observed
here may result from its off-target binding with thiol-containing
proteins such as bovine serum albumin (BSA) present in the
complete culture medium. In addition, the antiproliferation
potency of a variety of gold(I) compounds, including auranofin,
is largely reduced by the fetal bovine serum (FBS) of culture
medium in cell assays. This explanation was further supported by
the observation that cisplatin shows strong TrxR inhibition in cell
lysate, a FBS-free system (Figure 5C). All these results indicate
that GC20 has the advantage of being stable in the cell culture
system due to the absence of interaction, or a minimal if any
interaction, with the thiol groups.
Figure 3. (A) Rapid dilution assays of TrxR in the presence of GC20 or
curcumin (positive control). (B) Time course of TrxR inhibition by
GC20 treatment. TrxR (3.5 nM) from rat liver was incubated with
GC20 at 200 nM.
terminal active site. To confirm its ability to distinguish thiol and
selenol groups, we also tested the effects of GC20 on Trx, since
the Cys pair in the exposed active site can be readily targeted by
GC20 Shows Antiproliferation Potency on both Wild-
Type and Cisplatin-Resistant Cancer Cells. The cellular
activity of GC20 against a panel of different cancer cell lines was
also assessed with 3-[4,5-dimethyl-2-thiazolyl]-2,5-diphenyl-2H-
tetrazolium bromide (MTT) assay. With increasing concen-
tration of GC20, cell viability was found to be significantly
reduced after 48 h in the presence of FBS. GC20 showed broad
growth inhibition on these cancer cells, with IC50 ranging from
0.1 to 2 μM. This makes GC20 quite different from other gold(I)
compounds whose antiproliferation potency can be heavily
21
gold(I) compounds such as [Au(d2pypp) ]Cl. As expected,
2
GC20 did not show inhibition on Trx (Figure 4A), which
confirms that GC20 interacts with Sec instead of Cys. In an
attempt to further corroborate our findings, we examined the
affinity of GC20 to another selenoprotein, human glutathione
17
peroxidase (hGPx), reported to be inhibited by auranofin.
However, our result shows that GC20 has no effect on hGPx
even at 50 μM (Figure 4B). The remarkable specificity of GC20
in targeting only TrxR but not hGPx can be attributed to the
inaccessibility of the binding pocket of hGPx. Compared to
hGPx, the active site of TrxR is much more accessible and can be
easily approached by the structurally bulky and rigid GC20.
These results indicate that both the superior reactivity of Sec and
the highly exposed binding site contribute to the specificity of
GC20 for TrxR.
41,42
reduced or even eliminated altogether by FBS.
It is shown
that GC20 is more efficient than cisplatin, especially for the
cisplatin-insensitive cell lines (Table 1). It is also evident that the
phosphinoquinoxaline ligand of GC20 itself does not have any
inhibitory effect on the cancer cells tested, including QGY-7701,
HT-29, and BGC-823. In addition, the potential effect of GC20
on cisplatin-resistant cancer cells was also investigated, and our
results show that GC20 can inhibit both cisplatin-resistant and
wild-type cell lines (K111 and A549) with comparable potency
(Table 1), while in comparison, cisplatin showed inhibition only
in the two wild-type cell lines. In addition, the antiproliferation
potency of GC20 in longer treatment time (3 and 5 days) was
also measured with selected cancer cell lines. Similarly, GC20
inhibits cell proliferation even more strongly against these cancer
cell lines (Table S3 in Supporting Information).
GC20 Does Not Interact with Glutathione. Previous
studies revealed that GSH is a major contributor to the resistance
3
7,38
of cisplatin,
most likely as a result of an irreversible binding of
the thiol functionality with platinum metal center. The level of
GSH was shown to be much higher in cisplatin-resistant cancer
cells. It conjugates with and hence consumes a majority of
3
7,39
cisplatin in cytoplasm.
To evaluate its stability toward GSH,
GC20 was incubated with various concentrations of GSH for 2 h
and its UV absorption spectra were measured. It became
immediately evident from the data (Figure 4C) that GC20 is
remarkably stable even in the presence of a large excess (up to 14
equiv) of GSH. Furthermore, the UV spectra remain essentially
GC20 Induces Cell Cycle Arrest at G0/G1 Phase. We
further studied the potential mechanistic pathway responsible for
the strong antiproliferation potency of GC20. A549 cancer cells
were treated with GC20 for 48 h at three concentrations around
D
dx.doi.org/10.1021/jm3009822 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 4. Selectivity of GC20 among other potential targets. (A) Inhibition of hTrx by GC20, evaluated with insulin turbidity assay. (B) Inhibitory
ability of GC20 on hGPx, measured indirectly by a coupled reaction with GR. (C, D) Effect of different concentrations of GSH on the absorption spectra
of (C) GC20 and (D) cisplatin. GC20 or cisplatin (0.035 mM) was incubated with varying concentrations of GSH in PBS at 37 °C for 2 h. The
absorption spectrum of the sample was determined by scanning spectrophotometry over a wavelength range of 220−400 nm. (E, F) Effect of GSH on
the absorption spectra of (E) GC20 and (F) cisplatin at different incubation times. GSH (3.33 mM) was incubated with 1.67 mM GC20 or cisplatin at 37
°C. A 10 μL aliquot of the reaction mixture was withdrawn at different time points and diluted with 490 μL of PBS. The absorption spectrum of the
diluted sample was measured by scanning spectrophotometry over a wavelength range of 220−400 nm.
its IC value (0.25, 0.5, and 1 μM), and cell death was analyzed
observed in human renal cancer cells (SN12C) and human
epithelial carcinoma cells (HeLa) after treatment with GC20
Figure S5 in Supporting Information). These results further
50
with propidium iodide (PI) staining (Figure S4 in Supporting
Information). Very few stained cells were detected even at the
high concentration of 1 μM, indicating that GC20 treatment at
this concentration window does not induce significant cell death.
Nevertheless, cell proliferation was found to be strongly inhibited
by GC20 when compared to the DMSO control. We then
investigated the cell cycle progression and found significant
accumulation of cells at G0/G1 phase, with the reduction
occurring at S phase (Figure 6A,B). Similar results were also
(
confirm the antiproliferation potency of GC20 in a broad
spectrum of different cancer cell lines, most likely as a result of
G1 arrest-induced cell growth inhibition. Previous studies have
shown that cyclin D plays an important role in G1 to S phase
progression and cyclin-dependent kinase (CDK) inhibitors
contribute to G1 arrest, all of which function by regulating
E
dx.doi.org/10.1021/jm3009822 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
against a broad spectrum of cancer cells, we subsequently studied
its selectivity between NSCLC A549 cells and normal mouse
embryonic fibroblasts (MEF) cells (gifts from Dr. Liang Chen,
National Institute of Biology Sciences, Beijing, China). After
treatment with GC20 for 48 h, A549 cells were found to be
strongly inhibited (by more than 50%) at low concentrations
(
7
0.4−2 μM), whereas no effect was observed for MEF (Figure
A). In contrast, cisplatin exhibited comparable effects on both
cells at various concentrations. In other words, there was no
difference in the response of A549 and MEF cells to cisplatin
(
Figure 7B). In addition, we measured the effects of GC20
against B-cells, T-cells, and peripheral blood mononuclear cells
PBMC) isolated from human peripheral blood in comparison
(
with some human leukemia cell lines (Table S4 in Supporting
Information). Clearly, human normal cells (PBMC, CD4+ T-
cells, CD19+ B-cells) are resistant to the cytotoxic effects of
GC20, which is in contrast to the effects seen in human leukemia
cells. These results strongly support the high selectivity of GC20
in targeting cancer cells but not normal cells.
GC20 Shows Potential Anticancer Potency in Vivo with
Low Toxicity. Encouraged by the remarkable selectivity of
GC20 between cancer cells and normal cells as compared with
cisplatin, we further investigated its potential as an anticancer
agent by examining its activity and toxicity in animals. A549
tumor-bearing mice were treated with various doses of drugs or
saline vehicle for 7 days. Nine days after withdrawal of the last
3
treatment, tumor volume increased to 1850 ± 280 mm in the
saline control group. Remarkably, tumor volumes in the GC20
treatment group were limited to 360 ± 30, 510 ± 40, and 1040 ±
3
9
0 mm at doses of 8, 4, and 2 mg/kg, respectively. The tumor
growth inhibition ability of GC20 at 4 mg/kg is comparable to
that of cisplatin at 2 mg/kg, for which the tumor volume
3
increased to 530 ± 40 mm (Figure 7C). The administration of
GC20 was not shown to reduce the body weight of nude mice
bearing A549 tumors even at doses as high as 8 mg/kg (Figure S7
in Supporting Information). Similar results were observed in
several other xenografts, including murine hepatocellular
carcinoma H22 cells, Lewis lung cancer cells, and LS174T
human colon carcinoma cells (Figure S8 in Supporting
Information).
Figure 5. Inhibition of TrxR activity by GC20 in intact cancer cells. (A)
TrxR activity inhibition in intact A549 cancer cells. Cells treated with 2.0
μM compounds for the indicated time intervals were collected and lysed
before enzyme activity was determined. (B) Western blot results of
TrxR expression in A549 tumor cells treated with 2 μM GC20 or 0.1%
DMSO at different time points. (C) TrxR activity inhibition by GC20
and cisplatin in cell lysis. After A549 cells were collected and lysed, the
inhibition activity of the two compounds in cell lysate was examined.
To further establish the safety of GC20 in the mouse model,
we examined its toxicity at extreme doses and long durations.
Male and female ICR mice were selected and treated with a single
dose of GC20 or cisplatin. The mortality in mice was monitored
for 14 days after administration. The results (Table S5 in
Supporting Information) indicated that GC20 has a LD50 of
42.31 ± 1.63 mg/kg for male mice and 41.04 ± 1.61 mg/kg for
female mice, pointing to a much lower toxicity than that of
cisplatin (LD50 = 12.27 ± 2.25 mg/kg for male, LD50 = 11.27 ±
2.44 mg/kg for female). In separate experiments, Sprague-
Dawley rats were injected daily with either GC20 at doses of 2, 4,
and 8 mg/kg or cisplatin at 1 and 2 mg/kg for 28 days, and the
mortalities were measured. The results show that no rat died in
the GC20-treated group and there was little loss of body weight.
In contrast, all rats treated with 2 mg/kg cisplatin died within 7−
10 days. Note that at the dose of 1 mg/kg for cisplatin, eight out
of 10 rats died in 28 days (survival rate = 20%) (Figure 8A).
Many anticancer agents cause hematologic toxicity. Thus, we also
examined the changes of the levels of red and white blood cells
(Figure 8B,C) as well as leukocytes and granulocytes (Table S6
in Supporting Information) in the peripheral blood of SD rats for
28 days after treatment with GC20. No statistically significant
changes occur after GC20 treatment (p > 0.05).
phosphorylation of downstream effector retinoblastoma (Rb)
4
3
protein.
These observations prompted us to examine the variance of
G1-related targets by Western blot. Our result shows that 0.5 μM
GC20 can induce significant reduction of cyclin D as well as the
phosphorylation of Rb protein in A549 cancer cells at different
time points (12, 24, and 48 h) while 0.1% DMSO has no effects
on these targets (Figure 6C,D). However, no change was found
in the expression of CDK inhibitors, including P15 INK4b, P16
INK4a, P21 Cip1, and P27 Kip1 (Figure S6 in Supporting
Information). These results suggest that G1 arrest induced by
GC20 results from the reduction of cyclin D expression and
subsequent decrease in Rb phosphorylation rather than from
inhibition of CDK activity by CDK inhibitors.
GC20 Exhibits High Selectivity between Cancer Cells
and Normal Cells. The clinical utility of many anticancer agents
possessing strong antiproliferation potency is hampered by the
lack of their selectivity between cancer cells and normal cells.
After we established the antiproliferation potency of GC20
F
dx.doi.org/10.1021/jm3009822 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 1. Antiproliferation Potency of GC20 against Various Cancer Cell Lines and Cisplatin-Resistant Cell Lines
IC50 (mean ± SD, μM)
cell line
NCI-H460
cell type
GC20
cisplatin
phosphinoquinoxaline
non-small-cell lung cancer
non-small-cell lung cancer
liver cancer
0.3 ± 0.3
0.4 ± 0.4
1.2 ± 0.4
0.3 ± 0.2
0.7 ± 0.2
0.3 ± 0.2
0.3 ± 0.2
0.6 ± 0.7
0.8 ± 0.2
0.5 ± 0.50
0.7 ± 0.2
0.4 ± 0.2
9.2 ± 0.6
1.6 ± 0.1
1.8 ± 0.7
0.3 ± 0.2
0.3 ± 0.4
1.0 ± 0.7
0.1 ± 0.3
0.11 ± 0.03
8.4 ± 3.1
7.9 ± 1.9
18.7 ± 13.3
9.1 ± 4.6
7.6 ± 0.5
35.6 ± 12.1
74.7 ± 0.1
5.5 ± 3.9
2.5 ± 0.9
2.1 ± 21.4
4.8 ± 4.5
12.7 ± 8.8
44.6 ± 0.2
4.6 ± 2.4
15.5 ± 13.8
9.0 ± 7.4
2.4 ± 0.7
4.9 ± 1.6
237.2 ± 30.0
>333.3
A549
HepG2
QGY-7701
Hep-3B
liver cancer
>300.0
>300.0
>300.0
liver cancer
HT-29
colon cancer
SW620
colon cancer
BGC-823
SGC-7901
SK-OV-3
TC-1
gastric adenocarcinoma
gastric adenocarcinoma
ovarian cancer
ovarian cancer
breast cancer
MCF7
MDA-MB-231
A375
breast cancer
melanoma
K111
melanoma
K-562
leukemia
L1210
leukemia
KB
nasopharyngeal epidermoid
melanoma
K111 (cisplatin-resistant)
A549 (cisplatin-resistant)
non-small-cell lung cancer
Figure 6. Cell cycle arrest at G0/G1 phase with reduction in cyclin D expression and Rb protein phosphorylation induced by GC20. A549 cells were
treated with 0.5 μM GC20 for 12, 24, and 48 h, and the cell cycle distribution was measured by flow cytometry. Protein expression and phosphorylation
levels were analyzed by Western blots. (A) Representative plots of one time point in one set of triplicate experiments. G0/G1, S, and G2/M of the cell
cycle were sorted on the basis of DNA content after being stained with PI. (B) Bar graph of cell distributions. Means and SD were calculated from
triplicate independent experiments. (C, D) Expression of cyclins D1 and D3 and the phosphorylation level of Rb protein. Samples were probed with
anti-β-actin as loading control.
1
,40
DISCUSSION AND CONCLUSION
were considered to contribute to the resistance.
Extensive
■
study has shown in the past that many gold-containing
compounds are reactive toward Cys-containing proteins,
resulting in reduced efficacy and severe toxicity. Thus, the
development of novel gold compounds with synergistic reactivity
and selectivity is highly desirable.
Despite its historical success as a clinically useful chemotherapy
drug, cisplatin suffers from some major drawbacks such as
acquired resistance in long-term treatment, severe toxicity, and
occurrence of cisplatin-insensitive tumors, just to name a few.
Among them, enhanced GSH level and increased DNA repair
20
G
dx.doi.org/10.1021/jm3009822 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 7. Selectivity of GC20 between normal cells and cancer cells in
vitro and antitumor ability in vivo. (A) Antiproliferation potential of
GC20 on MEF and A549. (B) Antiproliferation potential of cisplatin on
MEF and A549. A549 or MEF cells were seeded into the 96-well plate
and different concentrations of GC20 or cisplatin were then added to the
cells after 12 h. The cell viability was measured with MTT assay after the
cells were treated with cisplatin for 48 h. (C) Inhibitory ability of GC20
on the growth of A549 tumor on nude mice. After establishment of solid
tumor, the mice were randomized and treated with saline control,
cisplatin, or GC20 once daily intravenously at the indicated doses for 7
days (indicated with arrows); the tumor volume was calculated
according to the width and length of tumors every 3 days during the
whole process.
Figure 8. (A) Kaplan−Meyer plot of survival of SD rats treated with
GC20 and cisplatin. (B, C) Numbers of (B) red blood cells (RBC) (M/
μL) and (C) white blood cells (WBC) (K/μL) in the peripheral blood of
SD rats were measured before and after the treatment with GC20 for 28
days (n = 10/group).
In this study, we have discovered that a novel gold(I)
compound, GC20, can selectively inhibit TrxR both in vitro and
in cancer cells, without reacting with other thiol- or selenol-
dependent targets, including GR, Trx, and GPx as well as TrxR
functional mutants and deletions (hTrxRU498C, hTrxRΔ8, and
hTrxRΔ16,). In addition, we observed that GSH does not react
with GC20, and the antiproliferation potency of GC20 is not
impeded by the presence of FBS in medium. According to
previous reports, GSH is considered to be a major contributor to
inhibit both K111 and A549 resistant cells by 50% at
concentrations as low as 0.1 μM, whereas their IC50 with
cisplatin is >200 μM. The positive performance that we observed
with GC20 helps further strengthen the notion that GC20 may
have the potential to be a useful therapeutic agent in the
treatment of a broad range of tumors.
40
cisplatin resistance. Serum albumin was shown to react with
many well-studied gold(I) compounds and to reduce their
2
0,41,42
efficacy in vivo.
The high reactivity toward thiols can also
lead to “off-target” binding or even severe side effects. Our
findings show that GC20 has the advantage of being highly
selective while maintaining a balanced and synergistic reactivity.
Further evaluation of its cellular performance revealed that
GC20 can strongly inhibit the proliferation of a broad spectrum
of cancer cell lines, including NSCLC, liver, colon, gastric,
ovarian, breast, and melanoma cancers as well as leukemia, with
IC50 ranging from 0.1 to 2 μM. Some of these cell lines were
shown to be insensitive to cisplatin, with IC higher than 20 μM.
Our results strongly suggest that GC20 may provide an
alternative way for the treatment of cisplatin-insensitive cancers.
For example, GC20 was shown to have potent inhibition of two
cisplatin-resistant cell lines. Specifically, GC20 was found to
In our study on its molecular mechanism, we have found that
0.5 μM GC20 could induce significant cell cycle arrest of A549
tumor cells by more than 20% in G1 phase after 24 h. Western
blot further revealed consistent variance in G1-related targets,
including reduction of cyclin D expression and decrease of Rb
protein phosphorylation. Previous studies have indicated that the
knockdown of TrxR in cancer cells may induce defects in G1/S
44
transition, and recovery was noticed in both cell morphology
and anchorage-independent growth property close to normal
5
0
28
cells. Similarly, Trx knockdown can result in significant G1
arrest with reduced cyclin D expression in A549 cancer cells,
45
which is consistent with our observations. Several possible
mechanisms that might be responsible for the TrxR inhibition-
H
dx.doi.org/10.1021/jm3009822 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 9. Synthetic route to GC20.
induced antiproliferation potency have been proposed. In the
most prevailing mechanism, it was believed that TrxR inhibition
could alter the permeability of mitochondrial membrane and
induce mitochondria swelling and loss of membrane poten-
g) with yield of 85%. Before the synthesis of 3, we prepared the reactant
acetylacetone sodium. Acetylacetone (309 mL, 3 mol) was slowly added
dropwise to a solution of NaOH (80 g, 2 mol) in methanol/water (4:1;
5
00 mL). During the whole process, the mixture was stirred at RT so
4
6−48
that the temperature of the reaction system remained constant. Once the
reaction was finished, the products were washed three times with ice-
cooled methanol after filtration in vacuum and then placed in a vacuum
drying oven to give the product as a white solid (191.9 g) in 94% yield.
The reactant acetylacetone sodium (7.9 g, 0.065 mol) was added and
dissolved in ethanol (180 mL) with reflux in an oil bath. Once the
desolvation was complete, the ethanol solution of acetylacetonate
sodium was transferred to a 250 mL dropping funnel and slowly added
to 2 (30.0 g, 0.05 mol) in acetone solution (180 mL) dropwise. The
reaction was stopped after reflux for 12 h in the oil bath. The mixture was
filtered to remove undissolved substance and was concentrated to get a
white solid, which was further recrystallized with ethanol to give white
needlelike crystals 3 (24.5 g), in 93% yield. Compound 3 (20.0 g, 0.039
mol) was added and dissolved in dichloromethane (100 mL). On the
other hand, Quinox P (26.3 g, 0.078 mol) was also dissolved in
dichloromethane (100 mL) and transferred to a 200 mL dropping
funnel to be added to the solution of 3 dropwise. The reaction
proceeded for 12 h at RT, and then the mixture was extracted with NaCl
solution (0.9%, 160 mL) twice. After drying with anhydrous sodium
sulfate as well as filtration and concentration, 40.8 g of 1 as a dark red
solid was obtained. The pure product 1 was obtained from
recrystallization with mixed organic solvent in 84% yield. The purity
of target compound was determined by HPLC (>99% purity), MS
tial.
According to a second mechanism, inhibition of TrxR
could increase the level of reactive oxygen species (ROS) and
lead to the overactivation of mitogen-activated protein kinase
49
(
MAPK) pathway. In a third mechanism that has been
proposed, inhibition of TrxR was believed to induce the
constitutive activation of apoptosis signal-regulating kinase
(
ASK1), which can be inhibited by Trx only in the reduced
50
state. These processes can all induce cell cycle arrest and/or cell
3
1
death. At this time, a clear distinction among these mechanisms
with regard to the action of GC20 is still premature, based on the
above-mentioned observation, and our ongoing research is
directed toward this goal.
We further examined the toxicity of GC20 both in cells and in
animals. In cells, GC20 was found to possess strong inhibition
against A549 but not MEF at concentrations as low as 0.4 μM,
whereas cisplatin did not show any effects on A549 at low
concentrations. At high concentrations (10 μM), cisplatin
inhibited both A549 and MEF cells by 50%. In addition,
human leukemic and lymphoblast cells were shown to be much
more sensitive to GC20 treatment than human normal cells
isolated from peripheral blood. In mouse xenograft models,
GC20 has comparable antitumor potential with that of cisplatin,
yet it has a better safety profile than cisplatin, either at extreme
doses (GC20, LD = 42.31 ± 1.63 mg/kg for male and 41.04 ±
1
spectra were recorded on a Micromass Q-Tof micro, and H NMR was
recorded on a Varian INOVA-400 NMR system (Figures S1−S3 in
Supporting Information).
Cells and Animals. All cancer cell lines were obtained from the
Shanghai Institute of Biochemistry and Cell Biology (SIBCB) and
cultured in RPMI 1640 medium (Gibco) containing 10% fetal bovine
5
0
1
.61 mg/kg for female; cisplatin, LD = 12.27 ± 2.25 mg/kg for
50
male and 11.27 ± 2.44 mg/kg for female) or in long duration
experiments (GC20, survival rate = 100% for as high as 4 mg/kg;
cisplatin, survival rate = 20% for as low as 1 mg/kg).
Furthermore, no obvious hematologic toxicity was observed in
animals after GC20 treatment for 28 days, which is consistent
with our safety studies at the cellular level. These results all
strongly suggest that GC20 is less toxic than cisplatin but has very
comparable efficacy in vivo. When these results are taken
together, GC20 has the potential to become a promising drug
candidate in cancer therapy.
serum (FBS) at 37 °C in a humidified incubator with 5% CO . Nude and
2
5
1
ICR mice and Sprague-Dawley rats were purchased from Shanghai
Laboratory Animal Center (SLAC, Shanghai, China). Animals were
housed under standard laboratory conditions with free access to food
and water.
Expression and Purification of hTrxRU498C, hTrxRΔ8, and
hTrxRΔ16. The hTrxRU498C construct in pET28a(+) is a gift from
Katja Becker, Interdisciplinary Research Center, Justus Liebig
University, Germany. The hTrxR gene lacking the last 24 or 48 bp
was cloned into the pET28a(+) expression vector to produce hTrxR
truncated mutants hTrxRΔ8 and hTrxRΔ16. Protein expression and
5
2
EXPERIMENTAL SECTION
Synthesis of GC20. The synthesis of GC20 (Figure 9) began with
gold powder of purity greater than 99.99% as starting material. Both gold
purification were performed as described previously. The purity of the
product was checked with sodium dodecyl sulfate−polyacrylamide gel
electrophoresis (SDS−PAGE).
■
powder (30.0 g, 0.15 mol) and (Bu) NCl (63.5 g, 0.23 mol) were added
In Vitro Enzymatic Activity Assays. For inhibition of rat liver
TrxR (Sigma−Aldrich), hTrxRU498C, hTrxRΔ8, and hTrxRΔ16, the
5,5′-dithiobis(2-nitrobenzoic acid) (DTNB) reduction assay was
performed according to manufacturer’s instructions (Sigma−Aldrich
product information sheet T9698). Typical reaction mixture (200 μL)
4
in ethanol (1600 mL) simultaneously in the presence of chlorine with
stirring at room temperature (RT). Once the oxidation was complete,
the reaction mixture was cooled overnight and followed with filtration
and drying to give product 2 as orange-colored needlelike crystals (75.2
I
dx.doi.org/10.1021/jm3009822 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
consists of 3.5 nM rat liver TrxR (or 140 nM TrxRU498C, or 200 nM
hTrxRΔ8, or 200 nM hTrxRΔ16). The rate of formation for 5-
thionitrobenzol (TNB) was determined by measuring the increase in
absorption (Δ412/min) with a Beckman Coulter paradigm detection
platform. The enzyme activities were calculated as the slopes (increase
in absorbance per second). The IC50 values were calculated as the
concentration of compound that decreased the enzymatic activity of the
untreated control by 50%.
For inhibition studies on hGR, the GSSG reduction assay was carried
out according to the manufacturer’s instructions (Sigma−Aldrich
product information sheet GRSA). Typically, 70 nM hGR (Sigma−
Aldrich) was used in the 200 μL reaction cocktail. Measurement of
increase of TNB and calculation of the enzymatic activity were done in
the same way as for the TrxR assay.
performed to determine the cell cycle distribution on a FACSCalibur
flow cytometer (Becton Dickinson). Data were analyzed with FlowJo
software (Tree Star, Ashland, OR).
Antibodies and Immunoblotting. Anti-β-actin was purchased
from Abcam. All other primary antibodies were from Cell Signaling
Technology. Cells were harvested and lysed with mammalian cell lysis/
extraction reagent (Sigma−Aldrich). The protein concentration was
determined with Bradford reagent (Sigma−Aldrich) according to
manufacturer’s instructions. Western blot analysis was conducted
according to standard procedures.
A549 Tumor Xenograft Studies. Subcutaneous flank tumors were
established by inoculating small A549 tumor pieces into female nude
mice. Tumor volumes (cubic millimeters) were calculated from
measurement of the length (L) and width (W) of tumors by the
2
3
For inhibition studies on hGPx, a commercially available kit
formula 0.5 × L × W . When flank tumors had grown to about 50 mm ,
the mice were treated once daily intravenously with GC20 (2, 4, and 8
mg/kg), cisplatin (2 mg/kg), or saline vehicle for 7 days. During the
whole process, we examined the volume of tumor every 3 days. The mice
were sacrificed and tumors were dissected and weighed 9 days after
withdrawal of drug treatment.
(
glutathione peroxidase activity kit; Cayman Chemical Co.) was used
as directed in the manual. The activity of hGPx was measured indirectly
by the coupled reaction with GR. The 200 μL reaction cocktail contains
3
0 ng of hGPx (Sigma−Aldrich). The activity of hGPx was calculated by
measuring the decrease in absorbance at 340 nm.
For inhibition studies on hTrx, the insulin turbidity assay was
Animal Toxicity Studies. ICR mice (18−20 g) were used to
evaluate the toxicity of GC20. The compounds tested were administered
in single intravenous doses (for GC20, 55.0, 46.8, 39.7, 33.8, and 28.7
mg/kg; for cisplatin, 20.0, 16.0, 12.8, 10.2, and 8.2 mg/kg) with saline
vehicle as control. Each concentration was applied to 10 female mice and
10 male mice. Changes in the appearance, behavior, and mortality in
mice were monitored for 14 days after administration. The median lethal
dose (LD50) values were determined by the Bliss method. In addition,
Sprague-Dawley rats (120−140 g) were injected intravenously with
GC20 (2, 4, and 8 mg/kg) and cisplatin (1 and 2 mg/kg) daily for 28
days to measure the long-term toxicity. The mortality of rats was
monitored and recorded daily.
53
performed as described previously. Briefly, 0.47 μM hTrx (Sigma−
Aldrich) was mixed with the reaction cocktail (100 mM phosphate
buffer, pH 7.0, 0.15 mM bovine insulin, 1 mM ethylenediaminetetra-
acetic acid (EDTA), and 0.1% Triton X-100). The reaction was initiated
by adding 330 μM dithiothreitol (DTT) to the mixture. The turbidity of
the solution was then continuously monitored at 650 nm.
Rapid Dilution Assay. Rapid dilution assay was performed as
5
4,55
described previously.
Briefly, 500 nM TrxR is incubated with 2.5 μM
GC20 or 50 μM curcumin for 30 min. After 100-fold rapid dilution of
the mixture, enzyme activity was measured as an increase in absorbance
at 412 nm for 35 min.
UV Spectral Analysis. To determine the interaction between GSH
Measurement of Hematologic Effects. Following treatment of
GC20 at doses of 8 and 2 mg/kg for 28 days, the numbers of white blood
cells (WBC) and red blood cells (RBC) as well as leukocytes and
granulocytes were measured with a multispecies hematology system
(HEMAVET 950, Drew Scientific Inc.).
Data Analysis and Statistics. All experiments were performed
three times with 10 samples/group. All results were derived from at least
three independent experiments. Data are shown as mean values ± SD.
Statistical analyses were performed by use of Student’s t test.
and GC20, a well-established UV absorbance spectrum analysis was
4
0
performed. Briefly, compounds were incubated with different
concentrations of GSH for 2 h or incubated with GSH for different
time intervals at 37 °C. The absorption spectrum was recorded by
wavelength scanning over the range 220−400 nm on a UV−visible
spectrophotometer (Shimadzu UV-2450).
Cellular TrxR Activity Inhibition Assay. A549 cells were treated
with 2.0 μM compounds in complete Dulbecco’s modified Eagle’s
medium (DMEM) (Invitrogen) for 6, 12, 24, or 48 h. The cells were
washed with PBS three times and lysed with mammalian cell lysis/
extraction reagent (Sigma−Aldrich) according to the instruction
manual. The protein concentration in the supernatant was determined
by use of Bradford reagent (Sigma−Aldrich). Examination and
calculation of TrxR activity in cells were done in the same way as for
the in vitro assays.
ASSOCIATED CONTENT
Supporting Information
■
*
S
Eight figures and six tables showing HPLC, MS, and NMR
spectral data of GC20, PI fluorescence and cell morphology of
GC20-treated A549 cells, GC20-induced G1 cell cycle arrest,
Western blot results of CDK inhibitors in GC20-treated A549
cells, antitumor ability of GC20 on various xenografts,
antiproliferative activity of screened gold-containing compounds
and anticancer potential of selected compounds, IC of GC20
Establishment of Cisplatin-Resistant Cell Lines. Cisplatin-
resistant sublines from K111 and A549 were established by continuous
56
exposure to cisplatin, as previously described. Both cell lines were
exposed to stepwise increases in concentrations of cisplatin, beginning
with 0.7 μM and doubly increased to 180 μM after the cells had regained
their exponential growth rate.
5
0
against different cancer cell lines at different treatment times,
IC of GC20 against human leukemic and lymphoblast cells and
50
Cellular Antiproliferation Assay. Cellular antiproliferative assays
normal cells isolated from peripheral blood, and acute toxicity.
This material is available free of charge via the Internet at http://
pubs.acs.org.
were carried out by the MTT (3-[4,5-dimethyl-2-thiazolyl]-2,5-
5
7
diphenyl-2H-tetrazolium bromide) method as previously described.
Briefly, cells (100 μL) were incubated with various concentrations of
compounds in a 96-well plate for 48 h at 37 °C. Then 10 μL of MTT (5
mg/mL) solution was added to each well. After further incubation at 37
AUTHOR INFORMATION
Corresponding Author
(N.H.) phone 86-10-80720645, fax 86-10-80720813, e-mail
■
*
°
C for 4 h, formazan formed from MTT was extracted by adding 100 μL
of DMSO for 12 h. Absorbance at 492 nm was then determined on a
microplate reader, and the IC50 values were calculated from the
inhibition ratios.
Cell Cycle Distribution Analysis. The effect of GC20 on cell cycle
distribution was measured with flow cytometry. A549 cells were
incubated in complete culture medium with 0.5 μM GC20 for 12, 24,
and 48 h. Cells were harvested and fixed with 70% ethanol overnight at
huangniu@nibs.ac.cn; (Q.L.) phone 86-21-65179533, fax 86-21-
65449361, e-mail liuquanhai_lqh@163.com.
Author Contributions
∥
These authors contributed equally to this work.
Notes
−
20 °C. Then cellular DNA was stained with 100 mg/mL propidium
iodide (PI) and 100 mg/mL RNase A in PBS. Flow cytometry was
The authors declare no competing financial interest.
J
dx.doi.org/10.1021/jm3009822 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
thioredoxin 1 and thioredoxin 2 oxidation as potential anticancer agents.
ACKNOWLEDGMENTS
■
J. Med. Chem. 2012, 55, 5518−5528.
We thank Professor Katja Becker (Interdisciplinary Research
Center, Justus Liebig University, Germany) for generously
providing us the hTrxR construct. We also thank Dr. Liang Chen
(15) Craig, S.; Gao, L.; Lee, I.; Gray, T.; Berdis, A. J. Gold-containing
indoles as anticancer agents that potentiate the cytotoxic effects of
ionizing radiation. J. Med. Chem. 2012, 55, 2437−2451.
(
National Institute of Biology Sciences, Beijing, China) for
(16) Marzano, C.; Gandin, V.; Folda, A.; Scutari, G.; Bindoli, A.;
providing MEF cells. We thank Nippon Chemical Industrial Co.,
Ltd., for providing the compound sample of GC20. Financial
support from the Chinese Ministry of Science and Technology
Rigobello, M. P. Inhibition of thioredoxin reductase by auranofin
induces apoptosis in cisplatin-resistant human ovarian cancer cells. Free
Radical Biol. Med. 2007, 42, 872−881.
“973” Grant 2011CB812402 (to N.H.), the Key New Drug
(17) Bhabak, K. P.; Mugesh, G. A synthetic model for the inhibition of
glutathione peroxidase by antiarthritic gold compounds. Inorg. Chem.
2009, 48, 2449−2455.
Creation and Manufacturing Program 2009ZX09120-055 and
Shanghai Committee of Science and Technology 08410703600
(18) Chiellini, C.; Casini, A.; Cochet, O.; Gabbiani, C.; Ailhaud, G.;
(to Q.H.L.) is gratefully acknowledged.
Dani, C.; Messori, L.; Amri, E. Z. The influence of auranofin, a clinically
established antiarthritic gold drug, on bone metabolism: analysis of its
effects on human multipotent adipose-derived stem cells, taken as a
model. Chem. Biodiversity 2008, 5, 1513−1520.
ABBREVIATIONS USED
■
RA, rheumatoid arthritis; TrxR, thioredoxin reductase; GPx,
glutathione peroxidase; GR, glutathione reductase; NSCLC,
non-small-cell lung cancer; GSH, glutathione; BSA, bovine
serum albumin; RB, retinoblastoma; MEF, mouse embryonic
fibroblasts; ROS, reactive oxygen species; MAPK, mitogen-
activated protein kinase; ASK, apoptosis signal-regulating kinase;
DTNB, 5,5′-dithiobis(2-nitrobenzoic acid); TNB, 5-thionitro-
benzol; RBC, red blood cells; WBC, white blood cells
(
19) Gromer, S.; Arscott, L. D.; Williams, C. H., Jr.; Schirmer, R. H.;
Becker, K. Human placenta thioredoxin reductase. Isolation of the
selenoenzyme, steady state kinetics, and inhibition by therapeutic gold
compounds. J. Biol. Chem. 1998, 273, 20096−20101.
(20) Carlock, M. T.; Shaw, C. F., III; Eidsness, M. K.; Watkins, J. W., II;
Elder, R. C. Reactions of auranofin and chloro(triethylphosphine)gold
with bovine serum albumin. Inorg. Chem. 1986, 25, 333−339.
(21) Rackham, O.; Nichols, S. J.; Leedman, P. J.; Berners-Price, S. J.;
Filipovska, A. A gold(I) phosphine complex selectively induces
apoptosis in breast cancer cells: implications for anticancer therapeutics
targeted to mitochondria. Biochem. Pharmacol. 2007, 74, 992−1002.
REFERENCES
■
(
1) Alderden, R. A.; Hall, M. D.; Hambley, T. W. The discovery and
development of cisplatin. J. Chem. Educ. 2006, 83, 728−734.
(22) Urig, S.; Fritz-Wolf, K.; Reau, R.; Herold-Mende, C.; Toth, K.;
(
2) Antman, K. H. Introduction: the history of arsenic trioxide in
cancer therapy. Oncologist 2001, 6 (Suppl. 2), 1−2.
3) Lebwohl, D.; Canetta, R. Clinical development of platinum
Davioud-Charvet, E.; Becker, K. Undressing of phosphine gold(I)
complexes as irreversible inhibitors of human disulfide reductases.
Angew. Chem., Int. Ed. 2006, 45, 1881−1886.
(
complexes in cancer therapy: an historical perspective and an update.
Eur. J. Cancer 1998, 34, 1522−1534.
4) Bruijnincx, P. C.; Sadler, P. J. New trends for metal complexes with
anticancer activity. Curr. Opin. Chem. Biol. 2008, 12, 197−206.
5) Giaccone, G. Clinical perspectives on platinum resistance. Drugs
000, 59 (Suppl. 4), 9−17.
6) Kelland, L. R. Preclinical perspectives on platinum resistance. Drugs
000, 59 (Suppl. 4), 1−8.
7) Nobili, S.; Mini, E.; Landini, I.; Gabbiani, C.; Casini, A.; Messori, L.
Gold compounds as anticancer agents: chemistry, cellular pharmacol-
ogy, and preclinical studies. Med. Res. Rev. 2010, 30, 550−580.
8) Gandin, V.; Fernandes, A. P.; Rigobello, M. P.; Dani, B.;
Sorrentino, F.; Tisato, F.; Bjornstedt, M.; Bindoli, A.; Sturaro, A.;
Rella, R.; Marzano, C. Cancer cell death induced by phosphine gold(I)
compounds targeting thioredoxin reductase. Biochem. Pharmacol. 2010,
9, 90−101.
9) Berners-Price, S. J.; Fau-Filipovska, A.; Filipovska, A. Gold
compounds as therapeutic agents for human diseases. Metallomics 2011,
, 863−873.
10) Gabbiani, C.; Messori, L. Protein targets for anticancer gold
compounds: mechanistic inferences. Anticancer Agents Med. Chem.
011, 11, 929−939.
11) Lima, J. C.; Rodriguez, L. Phosphine-gold(I) compounds as
(
23) Arner, E. S.; Holmgren, A. Physiological functions of thioredoxin
and thioredoxin reductase. Eur. J. Biochem. 2000, 267, 6102−6109.
24) Lillig, C. H.; Holmgren, A. Thioredoxin and related molecules:
(
(
from biology to health and disease. Antioxid. Redox Signaling 2007, 9,
25−47.
(
2
(
2
(25) Urig, S.; Becker, K. On the potential of thioredoxin reductase
inhibitors for cancer therapy. Semin. Cancer Biol. 2006, 16, 452−465.
(26) Powis, G.; Kirkpatrick, D. L. Thioredoxin signaling as a target for
cancer therapy. Curr. Opin. Pharmacol. 2007, 7, 392−397.
(27) Nguyen, P.; Awwad, R. T.; Smart, D. D.; Spitz, D. R.; Gius, D.
Thioredoxin reductase as a novel molecular target for cancer therapy.
Cancer Lett. 2006, 236, 164−174.
(
(
(28) Yoo, M. H.; Xu, X. M.; Carlson, B. A.; Gladyshev, V. N.; Hatfield,
D. L. Thioredoxin reductase 1 deficiency reverses tumor phenotype and
tumorigenicity of lung carcinoma cells. J. Biol. Chem. 2006, 281, 13005−
7
(
1
(
3008.
29) Nauser, T.; Steinmann, D.; Koppenol, W. H. Why do proteins use
selenocysteine instead of cysteine? Amino Acids 2012, 42, 39−44.
30) Ott, I.; Qian, X.; Xu, Y.; Vlecken, D. H.; Marques, I. J.; Kubutat,
3
(
(
D.; Will, J.; Sheldrick, W. S.; Jesse, P.; Prokop, A.; Bagowski, C. P. A
gold(I) phosphine complex containing a naphthalimide ligand functions
as a TrxR inhibiting antiproliferative agent and angiogenesis inhibitor. J.
Med. Chem. 2009, 52, 763−770.
2
(
anticancer agents: general description and mechanisms of action.
Anticancer Agents Med. Chem. 2011, 11, 921−928.
12) Rubbiani, R.; Kitanovic, I.; Alborzinia, H.; Can, S.; Kitanovic, A.;
(31) Bindoli, A.; Rigobello, M. P.; Scutari, G.; Gabbiani, C.; Casini, A.;
(
Messori, L. Thioredoxin reductase: A target for gold compounds acting
Onambele, L. A.; Stefanopoulou, M.; Geldmacher, Y.; Sheldrick, W. S.;
Wolber, G.; Prokop, A.; Wolfl, S.; Ott, I. Benzimidazol-2-ylidene gold(I)
complexes are thioredoxin reductase inhibitors with multiple antitumor
properties. J. Med. Chem. 2010, 53, 8608−8618.
as potential anticancer drugs. Coord. Chem. Rev. 2009, 253, 1692−1707.
(32) Fritz-Wolf, K.; Urig, S.; Becker, K. The structure of human
thioredoxin reductase 1 provides insights into C-terminal rearrange-
ments during catalysis. J. Mol. Biol. 2007, 370, 116−127.
(
13) Rubbiani, R.; Can, S.; Kitanovic, I.; Alborzinia, H.; Stefanopoulou,
(33) Viry, E.; Battaglia, E.; Deborde, V.; Muller, T.; Reau, R.; Davioud-
M.; Kokoschka, M.; Monchgesang, S.; Sheldrick, W. S.; Wolfl, S.; Ott, I.
Comparative in vitro evaluation of N-heterocyclic carbene gold(I)
complexes of the benzimidazolylidene type. J. Med. Chem. 2011, 54,
646−8657.
14) Schuh, E.; Pfluger, C.; Citta, A.; Folda, A.; Rigobello, M. P.;
Bindoli, A.; Casini, A.; Mohr, F. Gold(I) carbene complexes causing
Charvet, E.; Bagrel, D. A sugar-modified phosphole gold complex with
antiproliferative properties acting as a thioredoxin reductase inhibitor in
MCF-7 cells. ChemMedChem. 2008, 3, 1667−1670.
(34) Ott, I. On the medicinal chemistry of gold complexes as
anticancer drugs. Coord. Chem. Rev. 2009, 253, 1670−1681.
8
(
K
dx.doi.org/10.1021/jm3009822 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(
35) Bhabak, K. P.; Bhuyan, B. J.; Mugesh, G. Bioinorganic and
(56) Kikuchi, Y.; Kizawa, I.; Oomori, K.; Miyauchi, M.; Kita, T.; Sugita,
M.; Tenjin, Y.; Kato, K. Establishment of a human ovarian cancer cell
line capable of forming ascites in nude mice and effects of tranexamic
acid on cell proliferation and ascites formation. Cancer Res. 1987, 47,
592−596.
(57) Zhang, Y. F., J.; Jia, Y.; Wang, X.; Zhang, L.; Liu, C.; Fang, H.; Xu,
W. Development of tetrahydroisoquinoline-based hydroxamic acid
derivatives: potent histone deacetylase inhibitors with marked in vitro
and in vivo antitumor activitie. J. Med. Chem. 2011, 54, 2823−2838.
medicinal chemistry: aspects of gold(I)-protein complexes. Dalton
Trans. 2011, 40, 2099−2111.
(
modified by curcumin: a novel molecular mechanism for its anticancer
activity. J. Biol. Chem. 2005, 280, 25284−25290.
(
T. Augmentation of transport for cisplatin-glutathione adduct in
cisplatin-resistant cancer cells. Cancer Res. 1995, 55, 4297−4301.
(
36) Fang, J.; Lu, J.; Holmgren, A. Thioredoxin reductase is irreversibly
37) Goto, S.; Yoshida, K.; Morikawa, T.; Urata, Y.; Suzuki, K.; Kondo,
38) Seve, P.; Dumontet, C. Chemoresistance in non-small cell lung
cancer. Curr. Med. Chem. Anticancer Agents 2005, 5, 73−88.
39) Wang, X.; Guo, Z. The role of sulfur in platinum anticancer
chemotherapy. Anticancer Agents Med. Chem. 2007, 7, 19−34.
40) Ishikawa, T.; Ali-Osman, F. Glutathione-associated cis-
(
(
diamminedichloroplatinum(II) metabolism and ATP-dependent efflux
from leukemia cells. Molecular characterization of glutathione-platinum
complex and its biological significance. J. Biol. Chem. 1993, 268, 20116−
20125.
(
41) Berners-Price, S. J.; Girard, G. R.; Hill, D. T.; Sutton, B. M.; Jarrett,
P. S.; Faucette, L. F.; Johnson, R. K.; Mirabelli, C. K.; Sadler, P. J.
Cytotoxicity and antitumor activity of some tetrahedral bis-
(
diphosphino)gold(I) chelates. J. Med. Chem. 1990, 33, 1386−1392.
(
42) Mirabelli, C. K.; Johnson, R. K.; Sung, C. M.; Faucette, L.;
Muirhead, K.; Crooke, S. T. Evaluation of the in vivo antitumor activity
and in vitro cytotoxic properties of auranofin, a coordinated gold
compound, in murine tumor models. Cancer Res. 1985, 45, 32−39.
(
43) Massague, J. G1 cell-cycle control and cancer. Nature 2004, 432,
2
(
98−306.
44) Yoo, M. H.; Xu, X. M.; Carlson, B. A.; Patterson, A. D.; Gladyshev,
V. N.; Hatfield, D. L. Targeting thioredoxin reductase 1 reduction in
cancer cells inhibits self-sufficient growth and DNA replication. PLoS
One 2007, 2, e1112.
(
45) Mochizuki, M.; Kwon, Y. W.; Yodoi, J.; Masutani, H. Thioredoxin
regulates cell cycle via the ERK1/2-cyclin D1 pathway. Antioxid. Redox
Signaling 2009, 11, 2957−2971.
(
46) Cox, A. G.; Brown, K. K.; Arner, E. S. J.; Hampton, M. B. The
thioredoxin reductase inhibitor auranofin triggers apoptosis through a
Bax/Bak-dependent process that involves peroxiredoxin 3 oxidation.
Biochem. Pharmacol. 2008, 76, 1097−1109.
(
47) Rigobello, M. P.; Scutari, G.; Boscolo, R.; Bindoli, A. Induction of
mitochondrial permeability transition by auranofin, a gold(I)-phosphine
derivative. Br. J. Pharmacol. 2002, 136, 1162−1168.
(
48) Rigobello, M. P.; Scutari, G.; Folda, A.; Bindoli, A. Mitochondrial
thioredoxin reductase inhibition by gold(I) compounds and concurrent
stimulation of permeability transition and release of cytochrome c.
Biochem. Pharmacol. 2004, 67, 689−696.
(
49) Park, S. J.; Kim, I. S. The role of p38 MAPK activation in
auranofin-induced apoptosis of human promyelocytic leukaemia HL-60
cells. Br. J. Pharmacol. 2005, 146, 506−513.
(
50) Saitoh, M.; Nishitoh, H.; Fujii, M.; Takeda, K.; Tobiume, K.;
Sawada, Y.; Kawabata, M.; Miyazono, K.; Ichijo, H. Mammalian
thioredoxin is a direct inhibitor of apoptosis signal-regulating kinase
(
ASK) 1. EMBO J. 1998, 17, 2596−2606.
(
51) Berggren, M.; Gallegos, A.; Gasdaska, J. R.; Gasdaska, P. Y.;
Warneke, J.; Powis, G. Thioredoxin and thioredoxin reductase gene
expression in human tumors and cell lines, and the effects of serum
stimulation and hypoxia. Anticancer Res. 1996, 16, 3459−3466.
(
52) Urig, S.; Lieske, J.; Fritz-Wolf, K.; Irmler, A.; Becker, K. Truncated
mutants of human thioredoxin reductase 1 do not exhibit glutathione
reductase activity. FEBS Lett. 2006, 580, 3595−3600.
(
53) Hosoya-Matsuda, N.; Inoue, K.; Hisabori, T. Roles of
thioredoxins in the obligate anaerobic green sulfur photosynthetic
bacterium Chlorobaculum tepidum. Mol. Plant. 2009, 2, 336−343.
(
54) King, A. R.; Dotsey, E. Y.; Lodola, A.; Jung, K. M.; Ghomian, A.;
Qiu, Y.; Fu, J.; Mor, M.; Piomelli, D. Discovery of potent and reversible
monoacylglycerol lipase inhibitors. Chem. Biol. 2009, 16, 1045−1052.
(
55) Liu, T.; Toriyabe, Y.; Kazak, M.; Berkman, C. E. Pseudoirrever-
sible inhibition of prostate-specific membrane antigen by phosphor-
amidate peptidomimetics. Biochemistry 2008, 47, 12658−12660.
L
dx.doi.org/10.1021/jm3009822 | J. Med. Chem. XXXX, XXX, XXX−XXX