Download Antiproliferative and apoptotic effects of O

Carcinogenesis vol.21 no.5 pp.943–951, 2000
Antiproliferative and apoptotic effects of O-Trensox, a new
synthetic iron chelator, on differentiated human hepatoma cell
lines
Nafissa Rakba3, Pascal Loyer, David Gilot,
Jean Guy Delcros1, Denise Glaise, Paul Baret2,
Jean Louis Pierre2, Pierre Brissot and Gérard Lescoat
INSERM U522, Régulations des Equilibres Fonctionnels du Foie Normal et
Pathologique, Hôpital Pontchaillou, 35033 Rennes, 1CNRS URA 1529,
Faculté de Médecine, Rennes and 2CNRS UMR 5616, Laboratoire de
Chimie Biomimétique, Université Joseph Fourier, Grenoble, France
3To
whom correspondence should be addressed
Email: [email protected]
We investigated the effects of a new iron chelator, OTrensox (TRX), compared with desferrioxamine (DFO),
on proliferation and apoptosis in cultures of the human
hepatoblastoma HepG2 and hepatocarcinoma HBG cell
lines. Our results show that TRX decreased DNA synthesis
in a time- and dose-dependent manner and with a higher
efficiency than DFO. Mitotic index was also strongly
decreased by TRX and, unexpectedly, DFO inhibited
mitotic activity to the same extent as TRX, thus there is a
discrepancy between the slight reduction in DNA synthesis
and a large decrease in mitotic index after DFO treatment.
In addition, we found that TRX induced accumulation of
cells in the G1 and G2 phases of the cell cycle whereas DFO
arrested cells in G1 and during progression through S
phase. These data suggest that the partial inhibition of
DNA replication observed after exposure to DFO may be
due to a lower efficiency of metal chelation and/or that it
does not inhibit the G1/S transition but arrests cells in late
S phase. The effects of both TRX and DFO on DNA
synthesis and mitotic index were reversible after removing
the chelators from the culture medium. An apoptotic effect
of TRX was strongly suggested by analysis of DNA content
by flow cytometry, nuclear fragmentation and DNA
degradation in oligonucleosomes and confirmed by the
induction of a high level of caspase 3-like activity. TRX
induced apoptosis in a dose- and time-dependent manner
in proliferating HepG2 cells. In HBG cells, TRX induced
apoptosis in proliferating and confluent cells arrested in
the G1 phase of the cell cycle, demonstrating that inhibition
of proliferation and induction of apoptosis occurred independently. DFO induced DNA alterations only at concentrations ⬎100 µM and without induction of caspase 3-like
activity, indicating that DFO is not a strong inducer of
apoptosis. Addition of Fe or Zn to the culture medium
during TRX treatment led to a complete restoration of
proliferation rate and inhibition of apoptosis, demonstrating that Fe/Zn-saturated TRX was not toxic in the absence
of metal depletion. These data show that TRX, at concentrations of 20–50 µM, strongly inhibits cell proliferation and
induces apoptosis in proliferating and non-proliferating
HepG2 and HBG cells, respectively.
Abbreviations: BrdU, 5-bromo-2⬘-deoxyuridine; DFO, desferrioxamine B;
FCS, fetal calf serum; HCC, hepatocellular carcinoma; LDH, lactate dehydrogenase; PBS, phosphate-buffered saline; RR, ribonucleotide reductase; TRX,
O-Trensox.
© Oxford University Press
Introduction
Iron, which plays a central role in the regulation of many
cellular functions, is an essential element of all living species.
However, progressive iron overload in the liver, which is the
main iron storage organ, is observed in genetic and secondary
hemochromatosis and leads to hepatic fibrosis and frequently to
hepatocellular carcinoma (HCC) (1,2). In addition, intracellular
iron content is known to affect cell proliferation rate. For
instance, we have previously demonstrated that iron overload
increased DNA synthesis and mitotic index in primary cultures
of rat hepatocytes stimulated by growth factors (3). Several
studies have also shown that iron overload is implicated in
tumor cell growth in vivo (4,5) and the risk of developing an
HCC seems to be related to the level and duration of iron
overload (6,7).
There is great interest in developing new efficient and nontoxic iron chelators in order to decrease iron overload in both
genetic and secondary hemochromatosis. Iron depletion by
chelators has been shown to inhibit proliferation of various
cell lines and normal activated lymphocytes in vitro (8–10). In
our laboratory we have also demonstrated that desferrioxamine
(DFO) and hydroxypyridin-4-ones decrease DNA synthesis in
both normal and transformed hepatocytes (11,12).
Cell cycle studies have shown that chelator-treated cells are
arrested in different phases of the cell cycle depending upon
the cell type and the concentration and time of exposure to
chelators (13–15). Moreover, several authors have reported
that iron chelators induce apoptosis in proliferating cells such
as activated T lymphocytes and promyelocytic HL60 and
murine lymphoma 38C13 cells (16,17). Thus, iron chelators
have been proposed as promising antiproliferative agents in
the treatment of human cancers.
We previously reported iron mobilization by and a protective
effect on hepatocytes against iron toxicity of a new iron
chelator, O-Trensox (TRX) (18). TRX is a synthetic, water
soluble, tripodal iron-sequestering agent constituted of three
molecules of 8-hydroxyquinoline connected together by appropriate linker groups. This chelator can bind both ferric (Fe3⫹)
and ferrous (Fe2⫹) ions and zinc but does not induce radical
damage (18–20). In order to determine whether TRX can
affect growth and survival of hepatic cells, we investigated
its effects on proliferation and cell death in the human
hepatoblastoma HepG2 and hepatocarcinoma HBG cell lines.
Materials and methods
HepG2 and HBG cell lines and iron chelator exposure
The human hepatoblastoma cell line HepG2 used in this study was obtained
from Knowles et al. (21) while the human hepatocarcinoma cell line HBG
was established in our laboratory and was recently described (22). Both cell
lines were maintained in the following medium: 75% minimum essential
medium, 25% medium 199 (Hank’s salts) supplemented with 10% fetal calf
serum (FCS) and containing, per ml, 7.5 IU penicillin, 5 µg bovine insulin,
1 mg bovine serum albumin, 50 µg streptomycin, 2.2 mg NaHCO3 and 7⫻
10–7 M hydrocortisone hemisuccinate. For the experiments the cells were
maintained in the same medium. Cultures were treated with TRX or DFO at
943
N.Rakba et al.
different concentrations (20–200 µM) for 3 days. Two types of experiments
for Fe and Zn supplementation were performed using Fe citrate and ZnCl2:
(i) Fe citrate or ZnCl2 was added simultaneously with TRX or DFO at
equimolar concentrations for 2 or 3 days; (ii) Fe citrate (12 µM) or ZnCl2
(12 µM) was added to the culture medium in the absence of DFO or TRX
(50 µM) after 2 days depletion by these chelators. As a positive control for
apoptosis and caspase 3-like activity, a mixture of Fas-activating antibody
(CD95, clone 7C11; Immunotech) and cycloheximide (10 µg/ml) was used.
DNA [3H]methylthymidine incorporation
[3H]methylthymidine (Amersham, Les Ulis, France) was added to the culture
medium at a final concentration of 1 µCi/ml for 24 h. DNA synthesis was
evaluated by measuring [3H]methylthymidine incorporation into TCA-precipitated DNA. Results were expressed as a percentage of control values or as
c.p.m./dish.
DNA 5-bromo 2⬘-deoxyuridine (BrdU) incorporation
BrdU (Amersham, UK) incorporated during DNA replication was detected using
an indirect immunocytochemistry method. BrdU was diluted in culture medium
(1:1000) and incubation was performed at 37°C for 24 h. The cells were fixed
for 30 min in a solution of ethanol (90%) and acetic acid (5%), then washed
three times with phosphate-buffered saline (PBS). Non-specific sites were saturated with PBS supplemented with 10% FCS for 20 min. BrdU was detected using
a mouse anti-BrdU specific monoclonal antibody. The anti-BrdU antibodies were
detected using a secondary anti-mouse IgG2A immunoglobulin coupled to
horseradish peroxidase. Positive cells were then revealed with 3,3⬘-diaminobenzidine tetrahydrochloride.
Mitotic index determination
Colcemid (1 µM) (Sigma, La Verpillière, France) was added to the culture
medium for 24 h. The cells were fixed with a solution of ethanol (90%)/acetic
acid (5%) for 30 min. To visualize cells blocked in mitosis, DNA was stained
with methylene blue for 3 min, followed by two washes in water (pH 7); the
cytoplasm was stained with Giemsa diluted in water (1:10) for 10–15 min. Cells
arrested in mitosis appeared as large granulous cells and DNA was visualized
as ‘scattered chromosomes’, condensed chromosomal aggregates or
micronuclei. The mitotic index was the percentage of mitotic cells in the total
cell population.
Analysis of DNA content by flow cytometry
After cell trypsinization, DNA was stained with propidium iodide using a DNA
preparation kit (Coulter, Hialeth, FL). DNA content was measured using an
EPICS Elite flow cytometer (Coultronics, Hialeah, FL) equipped with an argon
laser (488 nm). Data analysis was carried out using Multicycle software (Phoenix
Flow Systems, San Diego, CA).
DNA damage evaluated by DNA fragmentation
DNA isolation was performed using a Nucleon BACC kit for the extraction of
genomic DNA from animal cell cultures (Amersham, Les Ulis, France). Briefly,
cells were lysed, treated with RNase solution (50 µg/ml) and deproteinized with
sodium perchlorate solution. DNA was precipitated successively by Nucleon
resin and cold absolute ethanol. Finally, DNA was collected by centrifugation,
washed twice with cold 70% (v/v) ethanol, dried and resuspended in 10 mM
Tris–HCl, 1 mM EDTA, pH 8. DNA samples of 10 µg were loaded onto 1%
agarose gels containing 0.1 µg/ml ethidium bromide and electrophoresis was
performed for 1 h at 60 V. A 1 kb DNA ladder (Gibco, Cergy-Pontoise, France)
was used as a molecular weight marker. DNA was visualized under UV light.
Nuclear fragmentation visualized by Hoechst staining
This detection is based on DNA staining by Hoechst 33258, a specific
fluorochrome of A-T nucleotides. The cells were fixed with ethanol/acetic acid
solution, then incubated in a PBS/Hoechst solution (0.5 µg/ml) for 10 min. Cells
were washed twice with PBS. Nuclei were visualized under UV with a Zeiss
microscope.
Lactate dehydrogenase (LDH) measurement
TRX toxicity was evaluated by measuring LDH activity (mIU/ml) in both culture
medium and cells using a LDH kit (Bayer Diagnostics, Puteaux, France) adapted
to the Alcyon 300 analyzer (Alcyon, Saint Mathieu de Treviers, France). Experimental results are expressed as percent extracellular LDH/total LDH activity.
Caspase activity assay
Cells were lysed in DEVD-AMC caspase 3-like activity buffer containing 20
mM PIPES, pH 7.2, 100 mM NaCl, 10 mM dithiothreitol, 1 mM EDTA, 0.1%
CHAPS, 10% sucrose as previously described (23). Aliquots of 100 µg of crude
cell lysate were incubated with 100 µM DEVD-AMC caspase 3-like substrate
at 37°C for 2 h. Caspase 3-mediated cleavage of DEVD-AMC peptide (Bachem,
Voisin le Bretonneux, France) was measured by spectrofluorometry using excitation/emission wavelengths (λex/λem) of 380 and 440 nm. Caspase activity is
presented in arbitrary fluorescence units per 100 µg total protein. To validate
944
Fig. 1. (a) [3H]methylthymidine incorporation into DNA in HepG2 cell
cultures maintained for 24, 48 and 72 h under control conditions and in the
presence of 50 µM DFO and TRX. (b) [3H]methylthymidine incorporation
into DNA in HepG2 cell cultures maintained for 48 h in the presence of 50
µM DFO or TRX followed by reversion after removal of chelators at 24 (–
chelators 24 hr) and 48 h (– chelators 48 hr). c, control cultures. Each value
is the mean ⫾ SD of triplicate cultures in a typical experiment. Three
independent experiments were performed and gave similar results.
caspase activity, an inhibitor of caspase activity (DEVD-CHO, a non fluorescent
substrate) was used to perform a competitive assay with DEVD-AMC. DEVDAMC caspase activity is not totally specific for caspase 3 (23) since other
caspases can cleave the DEVD-AMC substrate, but with a much lower efficiency.
This activity is usually presented as DEVD-AMC caspase 3-like activity.
Statistics
Results were expressed as means ⫾ SD. Statistical analyses were performed
using Student’s t-test. The significance level was set at 0.05.
Results
TRX inhibits HepG2 cell proliferation
In order to determine whether TRX affects HepG2 cell proliferation, we measured [3H]methylthymidine incorporation into control and TRX- (50 µM) and DFO-treated (50 µM) cells during
24 h periods over 3 days. Figure 1a shows a significant decrease
in DNA synthesis in TRX-treated cells which is time dependent.
DFO also induced a diminution of DNA synthesis, but to a lesser
extent than TRX (P ⬍ 0.001; Figure 1a).
This inhibition of DNA synthesis was reversible, since after
removal of the chelators DNA synthesis increased to reach control levels after between 24 and 48 h in DFO-treated cells and
after 48 h in TRX-treated cells (Figure 1b).
Inhibition of DNA replication by DFO and TRX, as well as
reversion of the effect, were confirmed using BrdU incorporation
under the same culture conditions. After 48 h incubation at
50 µM concentration of both chelators, DNA replication was
O-Trensox antiproliferative and apoptotic effects
Fig. 2. (a) DNA synthesis evaluated by BrdU incorporation into DNA in HepG2 cell cultures maintained for 48 h in the presence of 50 µM DFO and TRX
and then without the chelators for 12, 24 and 48 h (– chelators 12, 24 and 48 hr). c, control cultures. (b) Mitotic index in HepG2 cell cultures maintained for
48 h in the presence of 50 µM DFO or TRX followed by reversion at 12, 24 and 48 h after removal of chelators (– chelators 12, 24 and 48 hr). c, control
cultures. Each value is the mean ⫾ SD of triplicate cultures in a typical experiment. Three independent experiments were performed and gave similar results.
reduced, however, TRX decreased BrdU incorporation to a much
greater extent than DFO (P ⬍ 0.001; Figure 2a). The effects of
TRX and DFO on DNA replication were maintained for at
least 12 h after removal of the chelators. The increase in DNA
synthesis began 24 h after DFO withdrawal and reached the
control level after 48 h, while DNA replication in cultures previously incubated with TRX remained significantly lower than the
controls during the 48 h after chelator removal (P ⬍ 0.05;
Figure 2a).
The decrease in DNA synthesis in HepG2 cells observed in
the presence of TRX was followed by a large decrease in the
mitotic index (Figure 2b). After 48 h in the presence of 50 µM
TRX, an 8-fold decrease in the number of mitoses was observed
compared with untreated cells (P ⬍ 0.001; Figure 2b). Incubation
with DFO led to a large reduction in mitotic activity, similar to
the decrease observed with TRX but in contrast to the partial
inhibition of DNA replication (Figures 1a and 2a).
Inhibition of mitotic activity by the chelators was progressively reversible following their removal from the culture
medium. The mitotic index began to increase 24 h after removal
and reached 70 and 50% of the control value with DFO and
TRX, respectively, after 48 h (P ⬍ 0.001; Figure 2b). As observed
with DNA replication (Figures 1b and 2a), reversion of mitotic
inhibition occurred more rapidly after DFO than TRX withdrawal.
TRX arrests HepG2 cells in either the G1 or G2 phase of the
cell cycle
To further analyze the inhibition of HepG2 cell proliferation
mediated by TRX and DFO, we measured the DNA content of
HepG2 cells treated or not with these chelators by flow cytometry. A typical experiment is shown in Figure 3a–c and results of
four distinct experiments are presented in Figure 3d.
Our results confirmed that a proportion of the cells in each
phase of the cell cycle were dramatically affected by chelator
treatment. Compared with the controls (Figure 3a and d), cultures
maintained for 48 h in the presence of TRX showed a significant
decrease in the cell number in S phase and an increase in cells
in the G1 and G2/M phases (Figure 3c and d). We have shown in
Figure 2b that the mitotic index in TRX-treated cultures was
very low, demonstrating that cells with a G2/M DNA content
were actually arrested in the G2 phase. After 48 h exposure to
50 µM DFO (Figure 3b and d), a significant increase in the cell
number in S phase was observed along with a large decrease in
cells in G2/M phase.
We have shown (Figures 1 and 2) that the inhibition of proliferation induced by DFO and TRX is reversible. We confirmed
reversion of these effects by analyzing DNA content by flow
cytometry (Figure 3d, reversion). Forty-eight hours after TRX
withdrawal, the number of cells in the S and G2/M phases was
increased compared with 48 h treated cultures, while the percentage of cells in G1 was decreased. In cultures which had been
incubated with DFO, 48 h after removing the chelator more than
50% of cells were in G2 and only 22% in S phase.
These results lead us to conclude that TRX and DFO strongly
inhibit proliferation of HepG2 cells but arrest these cells at
different steps in the cell cycle: TRX mainly arrests cells in the
G1 and G2 phases while DFO blocks them in the G1 phase and/
or during progression through S phase, as shown in Figure 3b
by the broad peak of cells with various DNA contents.
TRX induces apoptosis in HepG2 cells
The first evidence that TRX could trigger HepG2 cell death
was the fact that a large peak of cells of low DNA content
was detected by flow cytometry after 72 h treatment at 100
µM and to a much lesser extent at 50 and 20 µM (Figure 4a).
945
N.Rakba et al.
Fig. 3. DNA content analysis measured by flow cytometry in HepG2 cells. A typical experiment is shown with control cultures (a), cultures maintained in the
presence of 50 µM DFO (b) and 50 µM TRX (c) for 48 h and the results of four independent experiments are presented in table form (d) showing
percentages (⫾ SD) of cells in each phase of the cell cycle (G1, S and G2/M). A typical reversion experiment with TRX and DFO treatments for 2 days
followed by withdrawal of chelators and a DNA content analysis at day 4 is shown.
In DFO-treated cells, no similar peak of dead cells was observed
during the 3 days culture with comparable concentrations of
chelator (data not shown). This cell death induced by TRX
was confirmed by significant dose-dependent LDH release into
the culture medium after 72 h exposure to the chelator
compared with untreated cultures (P ⬍ 0.001; Figure 4b).
Analysis of DNA content by flow cytometry and evaluation
of LDH release, at least under our conditions, did not allow
us to definitely conclude that TRX-mediated cell death was
due to apoptosis. Therefore, we examined three criteria considered to characterize apoptosis: in situ nuclear fragmentation,
946
DNA degradation in oligonucleosomes and activation of cysteine aspartate proteases (caspases) specifically induced during
the final step of the hepatocyte apoptotic process (24).
In situ nuclear fragmentation was demonstrated in cultures
treated with TRX by staining HepG2 cells with Hoechst 33258
(Figure 5a, inset). We established the index of cells with
fragmented nuclei for 72 h treatment with several doses of
TRX (Figure 5a). No cells with a fragmented nucleus were
observed after 12 h exposure to 20, 50 or 100 µM; a few
positive cells were observed after 24 h incubation in the
presence of 50 and 100 µM TRX. The number increased with
O-Trensox antiproliferative and apoptotic effects
Fig. 4. (a) DNA content analysis measured by flow cytometry in control
HepG2 cells and cells maintained for 72 h in the presence of 20, 50 and
100 µM TRX. (b) LDH release in control HepG2 cell cultures and cultures
maintained for 24, 48 and 72 h in the presence of 20, 50 and 100 µM TRX.
Each value is the mean ⫾ SD of triplicate cultures in a typical experiment.
Three independent experiments were performed and gave similar results.
time of culture in a dose-dependent manner, the highest index
being observed after 72 h incubation with 100 µM TRX (P ⬍
0.001; Figure 5a).
The degradation of DNA was then analyzed and a ‘DNA
ladder’ (DNA degradation in oligonucleosomes) was observed
in gel electrophoresis, even with the lowest TRX concentration
(20 µM) at 72 h incubation. The fragmentation rate increased
with higher concentrations (Figure 5b) and longer times after
a 48 h incubation with 50 and 100 µM TRX, but not with
DNA fragmentation induced at 20 µM (data not shown).
In contrast, we did not obtain any evidence of HepG2
apoptosis after a 3 day DFO treatment at concentrations up to
100 µM, neither by flow cytometry nor after in situ Hoechst
staining. We tested whether a higher concentration (200 µM)
could induce HepG2 cell death. Analysis of DNA fragmentation
clearly demonstrated the appearance of DNA alterations in
genomic DNA of HepG2 cells at this high concentration
(Figure 5b).
In order to definitely conclude that cell death observed after
TRX and DFO treatment was due to apoptosis, we assayed
caspase 3-like activity, which is a specific protease activity
Fig. 5. (a) Nuclear fragmentation index evaluated by Hoechst staining at 12,
24, 48 and 72 h in control HepG2 cell cultures (c) and cultures maintained
in the presence of 20, 50 and 100 µM TRX. Each value is the mean ⫾ SD
of triplicate cultures. Three independent experiments were performed and
gave similar results. (b) Genomic DNA degradation analysis in HepG2
control cultures and cells maintained for 72 h in the presence of 20, 50, 100
and 200 µM DFO or 20, 50 and 100 µM TRX. (c) Caspase 3-like activity
in control HepG2 cells and in cells treated with DFO (50, 100 and 200 µM)
or TRX (20, 50 and 100 µM) for 2 and 3 days or Fas-activating antibody
(200 ng/ml) and cycloheximide (10 µg/ml) for 6 h. Two independent
experiments gave similar results.
induced during the apoptotic process in many eukaryotic cells,
including hepatocytes (23,24). Caspase 3-like activity was
significantly induced after TRX treatment in a dose- and
time-dependent manner (Figure 5c), demonstrating that TRXmediated cell death was due to apoptosis. In contrast, DFO
did not significantly increase caspase 3-like activity even at
200 µM after 3 days treatment. As a positive control for
apoptosis, we used a monoclonal Fas-activating antibody
(200 ng/ml) combined with cycloheximide (10 µg/ml), strong
apoptotic inducers in hepatic cells (24). This apoptotic signal,
947
N.Rakba et al.
which induced caspase 3-like activity within 6 h (Figure 5d),
led to more than 95% cell death within 24 h (data not shown).
It is interesting to note that the value of Fas-activating antibody/
cycloheximide-induced caspase 3-like activity was only twice
the activity measured in cells treated with 100 µM TRX for 3
days, demonstrating that TRX is a potent apoptosis inducer.
Cell cycle arrest and apoptosis induced by TRX are due to
metal depletion
To rule out the hypothesis that cell cycle arrest and apoptosis
induced by TRX could be due to a toxic effect of the molecule
unrelated to metal chelation, we treated HepG2 cells with
TRX in the presence of Fe citrate or ZnCl2 and measured
[3H]methylthymidine incorporation and caspase 3-like activity
(Figure 6). Two different protocols were used: (i) 50 µM TRX
simultaneously with Fe or Zn (50 µM) for 2 or 3 days (Figure
6a and b); (ii) 50 µM TRX for 2 days, then Fe or Zn (12 µM)
in the absence of TRX for 1 (Figure 6c) or 2 more days
(Figure 6b). These two experiments were performed to test
the toxicity of Fe- and Zn-saturated TRX for 3 days (TRX
toxicity in the absence of metal depletion) and to determine
whether, after 2 days metal depletion by TRX, the addition of
Fe or Zn at a low concentration was able to inhibit cell death
at day 3 (Figure 5) and restore proliferation between days 3
and 4, as observed in the reversion experiments (Figures 1–
3), respectively.
Several important conclusions could be drawn from these
experiments:
d Treatment with TRX for 2 or 3 days in the presence of Fe
or Zn did not affect [3H]methylthymidine incorporation into
HepG2 cells compared with untreated cells (Figure 6a). A
similar result was obtained when HepG2 cells were treated
with DFO in the presence of Fe while addition of Zn (Figure
6a) was not able to prevent DFO-mediated inhibition of
replication.
d Fe or Zn, added to the culture medium after 2 days depletion
by TRX alone, was able to restore HepG2 proliferation at day
4 compared with untreated cells or control cells in the presence
of Fe alone. With DFO, [3H]methylthymidine incorporation
was similar in control cells and cells supplemented with Fe
citrate, while in the presence of Zn DNA replication was
decreased compared with replication in cells cultured in
medium supplemented with Zn alone. However, replication
was higher than in HepG2 cells maintained for 4 days in the
presence of DFO.
d A 3 day treatment with TRX alone strongly induced DEVDAMC caspase 3-like activity, as shown in Figure 5, while
addition of Fe or Zn, either simultaneously or after a 2 day
treatment with TRX, completely abolished caspase 3-like
activation. Addition of DEVD-CHO to cell lysates prepared
from TRX-treated cultures totally inhibited DEVD-AMC caspase 3-like activity, indicating that the background level of
this protease activity was high in HepG2 cell line.
These results demonstrate that TRX is not toxic per se and
that inhibition of proliferation and induction of apoptosis by
TRX are due to metal depletion.
Cell cycle arrest and apoptosis are two independent consequences of TRX treatment
Our results demonstrate that metal depletion by TRX leads to
inhibition of cell proliferation and induction of apoptosis.
However, these experiments did not allow us to determine
whether apoptosis was a consequence of cell cycle arrest or
948
Fig. 6. (a) [3H]methylthymidine incorporation into DNA of control HepG2
cells and in cultures treated for 48 and 72 h with 50 µM DFO or TRX in
the absence or presence of 50 µM Fe citrate or ZnCl2. (b) [3H]methylthymidine
incorporation into DNA of control HepG2 cells and in cultures treated for
48 h with 50 µM DFO or TRX and then maintained in the presence of Fe
citrate or ZnCl2 (12 µM) for 2 more days without chelators. c, untreated
control cultures; C⫹Fe and C⫹Zn, cultures treated for 2 days with Fe or Zn
only. Each value is the mean ⫾ SD of triplicate cultures of two independent
experiments which gave similar results. (c) Caspase 3-like activities in
control HepG2 cells or in cultures treated with TRX (50 µM) and Fe citrate
or ZnCl2 (50 µM) for 3 days (simultaneous treatment) or in cultures treated
with TRX for 2 days then treated with Fe citrate or ZnCl2 (12 µM) for 2
more days in the absence of chelator. Two independent experiments gave
similar results.
if cell death occurred independently of this inhibition of
proliferation.
To address this issue, HepG2 cells could not be used since,
at confluency, they grow continuously, indicating loss of cell–
cell contact inhibition, then rapidly detach and die. Therefore,
we decided to use the HBG hepatocarcinoma cell line, which
has been established in our laboratory and was recently
described (22). HBG cells are characterized by the fact that
O-Trensox antiproliferative and apoptotic effects
and apoptosis are two independent consequences of TRX
treatment.
Discussion
Fig. 7. (a) [3H]methylthymidine incorporation into DNA of control
proliferating HBG cells, in proliferating cultures treated for 48 or 72 h with
50 µM TRX in the absence or presence of Fe citrate or ZnCl2 (50 µM) and
in confluent non-proliferating HBG cells treated or not with TRX. (b)
Caspase 3-like activities in control HBG cells or in proliferating and nonproliferating HBG cell cultures treated with TRX (50 µM) in the presence
or absence of Fe citrate or ZnCl2 (50 µM) for 3 days (simultaneous
treatment).
they are poorly differentiated during proliferation at low cell
density while they stop proliferating at confluency, remain
arrested in G1 phase and undergo progressive differentiation
leading to the expression of many hepatocyte-specific functions
(22). In addition, they can be maintained at confluency and in
a differentiated state for weeks without cell death. Therefore,
HBG cells appeared to be the most suitable cell system to
distinguish between the effects of TRX on proliferation and
apoptosis.
We confirmed that a 2 day treatment with TRX (50 µM)
significantly decreased [3H]methylthymidine incorporation
(Figure 7a) and induced apoptosis (Figure 7b) in proliferating
HBG cell cultures. Addition of Fe or Zn (50 µM) to the
culture medium simultaneously with TRX (50 µM) restored
and even increased DNA replication and totally inhibited
caspase 3-like activity, as previously observed in HepG2
cells. We then maintained HBG cell cultures at confluency for
3 weeks until [3H]methylthymidine incorporation decreased
(Figure 7a) and studied whether TRX induced apoptosis in
these non-proliferating cells. As shown in Figure 7b, a 2
day treatment with TRX strongly increased DEVD-AMC
caspase 3-like activity to a similar level as in TRX-treated
proliferating HBG cells. Moreover, simultaneous treatment
with TRX and Fe or Zn totally inhibited activation of
caspase 3-like proteases.
These results demonstrate that TRX-induced apoptosis is
not due to cell cycle arrest but that inhibition of proliferation
Iron is implicated particularly in DNA replication, and
cellular depletion of this metal is a potent way to reduce
cell proliferation (8–12). Very few data have been published
concerning the molecular basis of the inhibitory effects of
iron chelators on proliferation. However, it has been shown
that iron depletion arrests human neuroblastoma cells in late
G1 phase of the cell cycle (25) and breast cancer cells at
the G1/S transition (15) and that this effect is mediated
through inhibition of the expression of cyclin A and/or
induction of cyclin-dependent kinase complexes such as
cyclin D1/cdk4 (15).
In order to determine whether TRX can also affect the
growth of hepatoma cells, we evaluated its effects on
proliferation of the human hepatoblastoma HepG2 cell line.
Here we show that TRX is a potent inhibitor of HepG2
cell proliferation, strongly decreasing DNA replication and
mitotic activity and arresting the cell cycle in the G1 and
G2 phases.
DFO, which was used as a reference iron chelator, slightly
reduces DNA replication but induces a strong inhibition of
HepG2 cell mitotic activity, arresting cells in G1 and/or S
phase. Two hypotheses can be proposed to explain why
DFO only slightly decreases DNA replication after 48 h
treatment (Figures 1a and 2a) but markedly inhibits mitotic
activity (Figure 2b): (i) the inhibitory effect of DFO on
proliferation is maximal only at the end of the 48 h
incubation, allowing significant DNA replication between 24
and 48 h, while a large decrease in mitotic activity is
measured at 48 h; (ii) DFO does not inhibit the G1/S
transition but arrests HepG2 cells at different steps during
S phase progression, indicating that some cells partially
replicate DNA before arrest.
These hypotheses are both supported by two additional
observations; a decrease in [3H]methylthymidine incorporation in DFO-treated HepG2 cells between 48 and 72 h
(Figure 1a); the remaining [3H]methylthymidine incorporation
in HepG2 cells treated with DFO for 72 h, which is much
higher than incorporation measured in cells exposed to TRX,
while in both conditions mitotic activity is very low
(Figure 2b).
Another striking result is the difference in profiles of
DNA content in TRX- and DFO-treated cells. Our results
strongly suggest that these two metal chelators act on cell
cycle progression via distinct mechanisms and/or chelate
intracellular metal pools with different efficiency.
Distinct effects of metal chelators on cellular functions,
especially cell proliferation, have been reported previously.
The antiproliferative effect of iron chelators most likely
triggers inhibition of ribonucleotide reductase (RR), which
is iron dependent and crucial for deoxynucleotide synthesis
(27). Several lines of evidence indicate that chelators can
bind distinct iron pools. For instance, DFO inhibits RR
activity not directly by attacking the iron radical center of
the R2 protein subunit of RR but by chelating the intracellular
iron pool and progressively depleting the cells (27–30). This
selectivity of chelators in binding iron associated with
different cellular structures or molecules is due to the ability
of chelators to penetrate different compartments of the cell
949
N.Rakba et al.
and their affinity for Fe2⫹ or Fe3⫹ and for other metals,
especially Zn2⫹. TRX is a potent chelator of Fe2⫹, Fe3⫹
and Zn2⫹ with complex formation constants of 1018, 1029.5
and 1022, respectively, while DFO chelates Fe3⫹ and Zn2⫹
with constants of 1030 and 1011. DFO and TRX affinities
for Fe3⫹ are similar but TRX chelates Zn2⫹ with a much
higher efficiency, suggesting that metal depletion by TRX
is more efficient than depletion by DFO.
To further progress the characterization of TRX effects,
we studied whether the antiproliferative effect of TRX
observed in HepG2 cells, and the subsequent cell cycle
arrest, is followed by or concomitant with cell death. In
cultures of proliferating HepG2 and HBG cells exposed to
TRX, significant cell death by apoptosis takes place, as
evidenced by nuclear fragmentation, DNA degradation in
oligonucleosomes and caspase 3-like activation. In addition,
in confluent non-proliferating HBG cells arrested in the G1
phase of the cell cycle, exposure to TRX also induced
apoptosis, demonstrating that TRX-mediated inhibition of
proliferation and induction of apoptosis occur independently.
TRX appears to be a stronger inducer of apoptosis than
DFO. Indeed, a 3 day treatment with high concentrations
of DFO (⬎100 µM) induces DNA alterations while 20 µM
TRX is sufficient to trigger significant apoptosis. In addition,
DNA alterations occur without significant caspase 3-like
activation, suggesting that DFO-mediated cell death is due
to necrosis and/or apoptosis involving a caspase 3-independent
pathway. However, several authors have reported an apoptotic
effect of DFO at concentrations above 100 µM after 24 h
exposure in activated T lymphocytes or the promyelocytic
cell line HL60 after inhibition of cell proliferation (17),
reinforcing the hypothesis that DFO induces apoptosis.
Apoptosis is known to be modulated by intracellular
levels of Zn2⫹ (31) and recent data have shown that TRX
has a much higher affinity for Zn2⫹ than DFO. Therefore,
Zn2⫹ depletion in TRX-treated cells may explain the
significant induction of apoptosis in HepG2 and HBG cells
compared with the low rate of cell death observed in DFOtreated cultures.
Due to its pronounced apoptotic effect, it is likely that
the use of TRX would be better adapted to the potential
treatment of hyperproliferative states than to the treatment
of iron storage diseases. Additional experiments are required
to determine the rate of cellular penetration of TRX and
its ability to access and deplete the different intracellular
metal pools in order to further explain the effects of this
new metal chelator on molecular processes regulating
proliferation and apoptosis.
Acknowledgements
This work was supported by the Association pour la Recherche sur le
Cancer (grant no. 6075), the Association Fer et Foie and BIOMED 2 Iron
chelators (CE/No.BMH4-CT97-2149).
References
1. Deugnier,Y.M., Charalambous,P., Le Quilleuc,D., Turlin,B., Searle,J.,
Brissot,P., Powell,L.W. and Halliday,J.W. (1993) Preneoplastic
significance of hepatic iron-free foci in genetic hemochromatosis: a
study of 185 patients. Hepatology, 18, 1363–1369.
2. Hsing,A.W., McLaughlin,J.K., Olsen,J.H., Mellemkjar,L., Wacholder,S.
and
Fraumeni,J.F.
(1995)
Cancer
risk
following
primary
hemochromatosis: a population-based cohort study in Denmark. Int. J.
Cancer, 60, 160–162.
3. Chenoufi,N., Loréal,O., Drénou,B., Cariou,S., Hubert,N., Leroyer,P.,
950
Brissot,P. and Lescoat,G. (1995) Iron may induce DNA synthesis and
repair in rat hepatocytes stimulated by EGF/pyruvate. J. Hepatol., 26,
650–658.
4. Hann,H.W.L., Stahlhut,M.W. and Blumberg,B.S. (1988) Iron nutrition
and tumor growth: decreased tumor growth in iron deficient mice.
Cancer Res., 48, 4168–4170.
5. Thompson,H.J., Kennedy,K., Witt,M. and Juzefyk,J. (1991) Effect of
dietary iron deficiency or excess on the induction of mammary
carcinogenenis by 1-methyl-1-nitrosourea. Carcinogenesis, 12, 111–114.
6. Tarao,K., Shimizu,A., Ohkawa,S., Harada,M., Ito,Y., Tamai,S. and
Kuni,Y. (1992) Development of hepatocellular carcinoma associated
with increases in DNA synthesis in the surrounding cirrhosis.
Gastroenterology, 103, 595–600.
7. Ballardini,G., Groff,P., Zoli,M., Bianchi,G., Giostra,F., Francesconi,R.,
Lenzi,M., Zauli,D., Cassani,F. and Bianchi,F. (1994) Increased risk of
hepatocellular carcinoma development in patients with cirrhosis and
with high hepatocellular proliferation. J. Hepatol., 20, 218–222.
8. Soyano,A., Chinea,M. and Romano,E.L. (1984) The effect of
desferrioxamine on the proliferative response of rat lymphocytes
stimulated with various mitogens in vitro. Immunopharmacology, 8,
163–169.
9. Polson,R.J., Jenkins,R., Lombard,M., Williams,A.C., Roberts,S., NouriAria,K., Williams,R. and Bomford,A. (1990) Mechanisms of inhibition of
mononuclear cell activation by the iron-chelating agent desferrioxamine.
Immunology, 1990, 176–181.
10. Richardson,D.R. and Milnes,K. (1997) The potential of iron chelators
of the pyridoxal isonicotinoyl hydrazone class as effective antiproliferative
agents II: the mechanism of action of ligands derived from salicylaldehyde
benzoyl hydrazone and 2-hydroxy-1-naphthylaldehyde benzoyl
hydrazone. Blood, 89, 3025–3038.
11. Chenoufi,N., Baffet,G., Drénou,B., Cariou,S., Desille,M., Clément,B.,
Brissot,P., Lescoat,G. and Loréal,O. (1997) Deferoxamine arrests in vitro
the proliferation of porcine hepatocyte in G1 phase of the cell cycle.
Liver, 18, 60–67.
12. Chenoufi,N., Drénou,B., Loréal,O., Pigeon,C., Brissot,P. and Lescoat,G.
(1998) Antiproliferative effect of deferiprone on the HepG2 cell line.
Biochem. Pharmacol., 56, 431–437.
13. Reddel,R.R., Hedley,D.W. and Sutherland,R.L. (1985) Cell cycle effects
of iron depletion on T-47D human breast cancer cells. Exp. Cell Res.,
161, 277–284.
14. Hoyes,K.P., Hider,R.C. and Porter,J.B. (1992) Cell cycle synchronisation
and growth inhibition by 3-hydroxypyridin-4-one iron chelators in
leukemia cell lines. Cancer Res., 52, 4591–4599.
15. Kulp,K.S., Green,S.L. and Vulliet,P.R. (1996) Iron deprivation inhibits
cyclin-dependent kinase activity and decreases cyclin D/CDK4 protein
levels in asynchronous MDA-MB-453 human breast cancer cells. Exp.
Cell Res., 229, 60–68.
16. Hileti,D., Panayiotidis,P. and Hoffbrand,A.V. (1995) Iron chelators induce
apoptosis in proliferating cells. Br. J. Haematol., 89, 181–187.
17. Kovar,J., Stunz,L.L., Stewart,B.C., Kriegerbeckova,K., Ashman,R.F. and
Kemp,J.D. (1997) Direct evidence that iron deprivation induces apoptosis
in murine lymphoma 38C13. Pathobiology, 65, 61–68.
18. Rakba,N., Aouad,F., Henry,C., Caris,C., Morel,I., Baret,P., Pierre,J.L.,
Brissot,P., Ward,R.J., Lescoat,G. and Crichton,R. (1998) Iron mobilisation
and cellular protection by a new synthetic chelator O-Trensox. Biochem.
Pharmacol., 55, 1797–1806.
19. Baret,P., Béguin,C.G., Boukhalfa,H., Caris,C., Laulhère,J.P., Pierre,J.L.
and Serratrice,G. (1995) O-Trensox: a promising water-soluble iron
chelator (both FeIII and FeII) potentially suitable for plant nutrition and
iron chelation therapy. J. Am. Chem. Soc., 117, 9760–9761.
20. Caris,C., Baret,P., Béguin,C., Serratrice,G., Pierre,J.L. and Laulhère,J.P.
(1995) Metabolization of iron by plant cells using O-Trensox, a highaffinity abiotic iron-chelating agent. Biochem. J., 312, 879–885.
21. Knowles,B.B., Howe,C.C. and Aden,D.P. (1980) Human hepatocellular
carcinoma cell lines secrete the major proteins and hepatitis B antigen.
Science, 209, 497–499.
22. Glaise,D., Ilyin,G., Loyer,P., Cariou,S., Bilodeau,M., Lucas,J., Puisieux,A.,
Ozturk,M. and Guguen-Guillouzo,C. (1998) Cell cycle gene regulation in
reversibly differentiated new human hepatoma cell lines. Cell Growth
Differ., 60, 19–26.
23. Stennicke,H.R. and Salvesen,G.S. (1997) Biochemical characteristics of
caspases-3, -6, -7 and -8. J. Biol. Chem., 272, 25719–25723.
24. Rouquet,N., Carlier,K., Briand,P., Wiels,I.J. and Joulin,V. (1996) Multiple
pathways of Fas-induced apoptosis in primary culture of hepatocytes.
Biochem. Biophys. Res. Commun., 229, 27–35.
25. Brodie,C., Siriwardana,G., Lucas,J., Schleicher,R., Terada,N., Szepesi,A.,
Gelfand,E. and Seligman,P. (1993) Neuroblastoma sensitivity to growth
O-Trensox antiproliferative and apoptotic effects
inhibition by desferrioxamine: evidence for a block in G1 phase of the
cell cycle. Cancer Res., 53, 3968–3975.
26. Lucas,J.J., Szepesi,A., Domenico,J., Takase,K., Tordai,A., Terada,N. and
Gelfand,E.W. (1995) Effects of iron depletion on cell cycle progression
in normal human T lymphocytes: selective inhibition of the appearance
of the cyclin A-associated component of the p33cdk2 kinase. Blood, 86,
2268–2280.
27. Nyholm,S., Mann,G.J., Johansson,A.G., Bergeron,R.J., Graslund,A. and
Thelander,L. (1993) Role of ribonucleotide reductase in inhibition of
mammalian cell growth by potent iron chelators. J. Biol. Chem., 268,
26200–26205.
28. Cooper,C.E., Lynagh,G.R., Hoyes,K.P., Hider,R.C., Cammack,R. and
Porter,J.B. (1996) The relationship of intracellular iron chelation to the
inhibition and regeneration of human ribonucleotide reductase. J. Biol.
Chem., 271, 20291–20299.
29. Nocentini,G., Federici,F., Franchetti,P. and Barzi,A. (1993) 2,2⬘-Bipyridyl6-carbothioamide and its ferrous complex: their in vitro antitumoral activity
related to the inhibition of ribonucleotide reductase R2 subunit. Cancer
Res., 53, 19–26.
30. Oblender,M. and Carpentieri,U. (1991) Growth, ribonucleotide reductase
and metals in murine leukemic lymphocytes. J. Cancer Res. Clin. Oncol.,
117, 444–448.
31. Sunderman,F.W. (1995) The influence of zinc on apoptosis. Ann. Clin. Lab
Sci., 25, 134–142.
Received July 23, 1999; revised November 22, 1999;
accepted January 5, 2000
951