ARTICLE
Auteur(s) : René Rodríguez1, Victor M
Campa1, José Riera1, M Teresa
Carcedo1, David S Ucker2, Sofía
Ramos1, Pedro S
Lazo1
1Instituto Universitario de Oncología del Principado
de Asturias and Departamento de Bioquímica y Biología Molecular,
Universidad de Oviedo, 33071 Oviedo, Spain
2Department of Microbiology and Immunology, University
of Illinois College of Medicine, Chicago, IL 60612, USA
accepté le 29 Octobre 2007
TNF can induce proliferation in several cell lines, mainly in
non-transformed cells [1] and this mitogenic effect is mediated
through gene expression [2, 3]. In this context, the transcription
factor nuclear factor-kappa B (NFκB) plays an important role.
TNF-induced proliferation seems to be mediated by this
transcription factor [4] and its inactivation can change the type
of response to TNF, switching from a proliferative response to the
induction of apoptosis [5]. TNF can also induce cell death by
apoptosis or necrosis. However, most cell types are resistant to
the toxic action of this cytokine unless protein
synthesis-dependent protective mechanisms, mainly mediated by NFκB,
are inhibited [6-8]. In these conditions, the caspase cascade can
be activated triggering the apoptotic response. The activation of
caspases may rely or not on the activation of mitochondrial
apoptotic events [9]. The action of several anti-apoptotic members
of the Bcl-2 family, such as Bcl-XL or Bcl-2, prevents
the apoptotic activation of the mitochondria, avoiding the death of
those cells in which the apoptotic mechanism depends on the
mitochondrial pathway.The mitogenic and cytotoxic effects of TNF
seem to be closely related. Thus, in some cell lines, both types of
effects can be observed simultaneously [5, 10]. Likewise, it has
been found that TNF-induced cell death occurs in a particular phase
of the cell cycle; however, this phase seems to change according to
the cell type or the environmental conditions. In the tumor cell
line WEHI-164/clon 2F, TNF inhibits DNA synthesis and induces
apoptosis in the G1/S border [11]. In some cell types,
TNF induces G1 arrest followed by apoptotic death [12].
In other cases however, the induction of arrest against apoptosis
[13]. In HeLa cells, TNF kills S, but not G1
phase-arrested cells [14], meanwhile in L929 cells synchronized in
the G1/S border, TNF-induced cell death is linked to the
abnormal induction of mitosis [15].All of these studies were
performed using cancer cells, which usually show deficiencies in
their cell cycle checkpoints, and this fact is most likely
responsible for the wide variety of responses observed. In the
present work, we aimed at studying the relationship between the
mitogenic and cytotoxic effects of TNF in a non-transformed cell
line. We found that TNF shows both mitogenic and toxic effects in
G0-synchronized NIH 3T3 murine fibroblasts. In this
model, cell death affects only that portion of cells that are
unable to proliferate. Likewise, TNF induces apoptosis in cells
arrested at any other point of the cell cycle (phases G1/S, S or
M). This type of sensitization against TNF-induced toxicity is
mediated through a mechanism not inhibitable by NFκB and triggers a
pattern of apoptotic mitochondrial activation different from that
observed after RNA synthesis inhibition.
Methods and materials
Synchronization and transfection of cell cultures
NIH 3T3 cells were synchronized in the G1/S, S or M phase by
treatment with 2 μg/mL aphidicolin, 200 μM hydroxyurea or 0.1 μg/mL
nocodazole respectively for 24 hours. Alternatively, cells were
synchronized in the G0 phase by incubation in medium
with a low percentage (0.5%) of fetal bovine serum (FBS) for 48
hours. Transfection of synchronized cells with 1 μg of Bcl-2,
IκB-mut (a gift from Dr D. Wallace) or empty vector plasmids, or
0.25 μg of NFκB-SEAP plasmid (a gift from Dr B. Darnay), in
addition to 0.2 μg of the transfection control plasmid EGFP-F, was
performed using Lipofectamine Plus Reagent (Gibco BRL).
Cell cycle analysis
After treatments, cells were collected, washed with PBS and fixed
in 70% ice-cold ethanol. Cells were washed twice with PBS followed
by incubation in 5 μg/mL PI (Sigma) and 100 μg/mL RNase A (Sigma)
for 30 min. Cell cycle analysis was performed by flow
cytometry using an Ortho-Cytoron (OrthoDiagnostics,
Johnsonn&Jonhson). The percentage of cells in the different
phases of the cell cycle was quantified using the data processing
program ModFit (Verity Software House).
Analysis of DNA synthesis
Cells grown under coverslips were incubated in medium with 0.5% FBS
for 48 hours. Cells were then treated in the presence of 20 μM
5-Bromo-2’-deoxyuridine (BrdU; Sigma) as indicated. Cells were
washed with PBS and fixed with ice-cold methanol for 15 min.
After washing another three times with PBS, cells were incubated
with 1.5 M HCl for 30 min and washed again three times with
PBS-T (PBS/0.1% BSA/0.2% Tween 20). The coverslips were incubated
with anti-BrdU (Becton Dickinson) diluted 1:30 in PBS-T for 1 hour
at room temperature, washed three times with PBS-T and incubated
with anti-mouse IgG-FITC (Dako) diluted to 1:100 in PBS-T for
30 min at room temperature. This incubation, as well as the
subsequent processing, were performed in the dark. Cells were then
washed with PBS, incubated for 10 min with 5 μg/mL PI and
washed again with PBS. The coverslips were mounted using
Fluoromount G (Southern Biotechnic). The FITC (Ex: 490 nm; Em: 520
nm) and the PI (Ex: 590 nm; Em: 615 nm) fluorescence present in
each cell was analyzed with a Confocal Laser Microscopy System (BIO
RAD MCR 600). Results are the mean of the analysis of six randomly
chosen fields of each sample (approximately 500 cells).
Determination of the number of cells by sulforhodamine B
dye
Synchronized cells were fixed immediately after the indicated
treatments using 10% trichloroacetic acid for 30 min, washed
with PBS and covered with 0.4% sulforhodamine B (dissolved in 1%
acetic acid) for 15 min. After that, cells were washed six
times with 1% acetic acid. Finally, cell-bound sulforhodamine B was
dissolved in 2 mL of 10 mM Tris-HCL pH 10.4. The
sulforhodamine B fluorescence signal (Ex: 530 nm; Em: 590 nm) was
measured using a Cytofluor TM 2350 fluorimeter (Millipore). A
calibration curve was used to express the fluorescence registered
as number of cells.
Western blot analysis
Cytosolic, mitochondria-free, extracts were prepared by suspending
cells in mitochondrial buffer (80 mM KCl, 10 mM Tris HCL, 3 mM
MgCl2, 1 mM EDTA, 5 mM KH2PO4, 10
mM sodium succinate, pH 7.4) and then incubated with digitonin (at
a ratio of digitonin: protein = 0.2) for 5 min.
Lysates were clarificated by centrifugation, and the supernatants
were resolved on 12% SDS polyacrylamide gels and blotted onto
nitrocellulose. Specific polyclonal antibodies were used to detect
cytochrome c (Pharmingen). Immunoreactive bands were visualized by
the luminol reaction (ECLplus, Amersham).
Mitochondrial membrane potential assay
Mitochondrial membrane potential was assessed, simultaneously or
not with the cell cycle, as previously described [16]. In
transfected cells, only EGFP-F positive cells were analyzed.
Assessment of cell shrinkage, rounding and condensation
The manifestation of typical apoptotic morphology, which includes
shrinkage and cell rounding, was visualized by phase contrast
microscopy. In addition, dying transfectants cells reliably
displayed a condensed pattern of EGFP F fluorescence [17]. More
than 500 cells in six, randomly chosen fields were counted for each
condition in every experiment.
Propidium iodide uptake
Cells were harvested, washed once with PBS, and then incubated in
PBS with 40 μg/mL of PI for 15 minutes at 4°C, in the dark. The
percentage of PI fluorescence-positive cells was assessed by flow
cytometry analysis. In transfected cells, only EGFP-F positive
cells were computed for PI uptake.
Electrophoretic mobility shift assay (EMSA)
After the indicated treatment, nuclear extracts were prepared as
described previously [18] and used in electrophoretic mobility
shift assay cells using a double-stranded oligonucleotide probe
containing the κB site from the mouse κ light chain enhancer
(5’-TGCACAGAGGGGACTTTCCGAGAGG-3’) as previously described [19].
Transcriptional activity of NFκB. Secreted alkaline phosphatase
(SEAP) assay
Cells were co-transfected with 1 μg of the indicated plasmids and
0.5 μg of the NFκB-SEAP plasmid, containing the codifying sequence
for the SEAP under the control of a NFκB response element. After
treatment, SEAP activity present in culture medium was assayed.
10-25 μL of these culture media were diluted in 100 μL of the assay
buffer (final concentration: 100 mM Tris pH 9.0; 0.1% BSA) in
96-well plates. Plates were incubated for 1 hour at 65°C in
darkness to inactivate endogenous phosphatases. Then, 50 μL of
1 mM 4-methyl-umbelipheryl-phosphate (4-MUP), freshly prepared from
a 100 mM stock (kept at -70°C), was added to each well. The plate
was incubated for 2-4 hours at 37°C in darkness and the
fluorescence emitted (excitement: 355 nm; emission 460 nm) was
measured in a Cytofluor TM 2350 (Millipore).
Results
TNF induces mitogenic and cytotoxic signals in NIH 3T3
cells
TNF induces mitogenic signals in some non-transformed cell types
[1]. This is the case for NIH 3T3 murine fibroblasts as indicated
by the induction of DNA synthesis. Thus, a 20-hour TNF treatment of
G0-arrested cells caused a large increase in the number
of cells entering the cell cycle and presenting DNA contents
corresponding to S and G2 phases (figure 1A). This effect
was better observed using low serum concentration (0.5%) but became
less evident in the presence of a high serum concentration (10%).
In this case, most of the TNF-induced mitogenic effect must have
been masked by the serum’s growth factors effect. This mitogenic
effect was also confirmed by analyzing the fraction of
G0-arrested cells that incorporated BrdU to newly
synthesized DNA after the treatment with TNF or 10% serum (figure 1B). However,
in these experiments we could also observe that, despite the great
increase in the percentage of BrdU-positive cells, there was a
systematic decrease in the number of cells in Go-arrested cultures
treated with TNF. These observations seemed to be contradictory,
since TNF induces both a mitogenic signal (induction of DNA
synthesis) together with a global anti-proliferative behaviour
(decrease of the cell number). In order to quantify this toxic
response, we used the fluorescent dye sulforhodamine B to study the
kinetics of growth of G0-synchronized NIH 3T3 cells
after TNF treatment. In the presence of 10% of serum there was no
significant difference between untreated and TNF-treated cultures
(figure 1C). On
the other hand, when we used 0.5% serum, control cells remained
arrested and the cultures treated with TNF showed a progressive
decrease in the number of cells so that after 48 hours, only 25% of
the initial number of cells remained in culture (figure 1D).
Taken together, these data suggest the coexistence of both
TNF-induced mitogenic and toxic effects in cells arrested in the
G0 phase. To better understand the relationship between
both types of TNF-induced signals, it became important to establish
which pool of cells were sensitive to the cytokine: those that
remained in the G0 phase, those that entered the cell
cycle or both types of cells equally. In order to study phase of
the cycle where cells die, we analyzed simultaneously the cell
cycle and the loss of mitochondrial potential by flow cytometry.
TNF promoted the entry of G0-synchronized cells into the
cell cycle. Thus, in untreated cultures, about 10% of cells were in
S or G2/M phases and only 6% of cells showed low mitochondrial
potential (figure
2A). In cultures treated with TNF, 46% of cells were in S
or G2/M. In agreement with the data shown above, TNF also induced a
cytotoxic effect. Thus after treatment, 54% of the cells showed low
mitochondrial potential. Notably, most of the cells that progressed
to S or G2 phases by effect of TNF conserved high
mitochondrial potential, while most of the cells that lost
mitochondrial potential remained in the G0 phase (figure 2B). Untreated
cells showed a similar proportion of cells in each phase of the
cycle in both, cells with high or low mitochondrial potential.
On the other hand, treatment of unsynchronized cells with TNF
and actinomycin D (Act-D) for 4 hours was sufficient to induce the
loss of mitochondrial potential in any phase of the cell cycle
(figure 2C,
D).
Taken together, these data suggest that TNF-induced mitogenic
signals cannot be completed in the whole population of
G0-arrested cells, and this would trigger death signals
in cells unable to proliferate.
TNF induces cell death in NIH 3T3 cells arrested at different
points of the cell cycle
We next determined whether the ability of TNF to induce cell death
associated with the impossibility to complete the mitogenic
signals, is specific to the G0 phase or, whether it is
independent of the phase as long as cell growth has been blocked.
To address this question we studied the toxicity of TNF in NIH 3T3
cells arrested at different points of the cell cycle. Specifically,
we arrested cells in the G1-S phase border
(G1/S) by treatment with the inhibitor of DNA polymerase
II, aphidicolin; in the S phase by treatment with the inhibitor of
ribonucleotide reductase, hydroxyurea or in mitosis by treatment
with nocodazole, a compound that affects the assembly of
microtubules (figure
3A). Once the arrest was achieved by treatment for 24 hours
with the relevant drug, we treated the cells with TNF in the
presence of the synchronicity agent for another 24 hours. After the
treatment, we washed out the floating dead cells and quantified the
remaining attached cells using sulforhodamine B. We found 50, 60
and 45% of TNF-induced death in the cultures arrested in the
G1/S, S and M phase respectively (figure 3B). The percentage
of TNF-induced cell death in cycling cells was less than 10%, and
cell death due to the synchronizing agent did not exceed 10% in any
case.
This TNF-induced toxicity was mediated by apoptosis as indicated
by annexin V binding assays (figure 3C). There was a
significant increase of annexin V-positive cells (both PI-positive
and negative) after TNF treatment of the arrested cells. Moreover,
these cells showed a characteristic apoptotic morphology: they
became smaller, rounded, shrunken and, when transfected with the
EGFP-F expression plasmid, showed an significant increase of the
GFP fluorescence intensity due to the condensation of the cell, as
previously observed [17] (a quantification of G1/S- and
S phase-arrested cells presenting this morphology is shown in figure 4C.
Thus, TNF induced apoptosis in NIH 3T3-arrested cells, but not
in cycling cells and this cytotoxic effect is independent of the
phase of the cell cycle in which cells are arrested. Therefore, it
seems that TNF cytotoxic signals occurs only when the mitogenic
signals cannot be completed by the blocking of the cell cycle
progression at any point.
TNF-induced apoptosis in arrested cells is not due to a
deficient activation of NFκB
The activation of NFκB has been revealed as one of the most
important survival and proliferation signal triggered by TNF [4, 7]
and its inhibition enhances TNF-induced toxicity [5, 6, 8]. We
investigated whether the increase in TNF toxicity in NIH
3T3-arrested cells was due to a deficient activation of NFκB. We
carried out EMSA assays using nuclear extracts of arrested or
cycling cells treated or not with TNF. This cytokine induced a
similar NFκB binding to DNA in cells arrested in G0, S
or M phases or in cycling cells (figure 4A). Moreover, this
TNF-induced increase in DNA-binding activity correlated with the
expression of the reporter gene SEAP under the control of NFκB, and
was prevented by overexpression of a dominant-inhibitor IκBα mutant
(IκB-mut; (figure
4B). This mutant has been reported to induce sensitivity to
TNF in cycling cells [8, 20]. When we inhibited the NFκB-protective
pathway in G1/S or S phase-arrested cells by
overexpression of IκB-mut, we found a further increase in the
percentage of apoptotic cells observed after the TNF treatment
(figure 4C),
suggesting that this pathway was functional in arrested cells.
Taken together, these data indicate that the increased toxicity of
TNF in arrested cells is not due to the loss of the anti-apoptotic
activity of NFκB.
TNF-induced apoptosis in arrested cells requires the
mitochondrial pathway
As described above, TNF induced apoptosis in NIH 3T3 cells arrested
in any phase of the cell cycle, eliminating the well-known
requirement of simultaneous inhibition of protein synthesis. This
type of sensitization to TNF has been described in other cell types
arrested in the S phase [14], but little is known about the
apoptotic signalling in these conditions. To address this issue, we
analyzed the participation of the mitochondria in TNF-induced
apoptosis in NIH 3T3 cells arrested in the G1/S phase.
The mitochondrial potential of the G1/S-arrested
cells remained unaffected in most cells after seven hours of TNF
treatment, although after 24 hours more than 50% of cells showed
low mitochondrial potential (figure 5A). The treatment
of cycling cells with TNF for 24 hours did not affect the
mitochondrial potential. This behaviour of the mitochondrial
potential contrasted with that observed for cycling cells treated
with TNF/Act-D. In this case, after four hours of treatment, more
than 50% of cells showed low mitochondrial potential and after six
hours more than 90% of cells showed uncoupled mitochondria (figure 5B).
Cytochrome c can be detected in the cytosol in
G1/S-arrested cells after one hour of treatment with
TNF, reaching the maximum level after eight hours (figure 5C), well in
advance of the loss of mitochondrial potential. The synchronization
process itself did not induce the release of cytochrome c. On the
other hand, the release of cytochrome c from the mitochondria and
the loss of mitochondrial potential after TNF/Act-D treatment of
cycling cells followed similar kinetics (figure 5D).
The activation of the mitochondrial pathway might not be
essential for the apoptotic mechanism triggered by death receptors
[9]. To study the relevance of this pathway in the apoptotic
mechanism induced by TNF in G1/S arrested cells or
cycling cells co-treated with Act-D, we ectopically expressed the
anti-apoptotic protein Bcl-2, a well-known inhibitor of the
mitochondrial pathway. The co-transfection with a green fluorescent
protein (GFP) variant allowed us to detect the transfected cells
either by flow cytometry or fluorescence microscopy. This variant,
a membrane-targeted GFP (EGFP-F; [21]), is retained in dying cells,
avoiding the leaking shown when soluble GFP is used [17]. In the
experiments carried out with arrested cells, we found a high level
of apoptotic cells (approximately 30%) in the control conditions
(arrested cells transfected with an empty vector and left
untreated). This percentage of abnormally high toxicity is probably
due to the process of transfection of arrested cells.
We first checked the ability of Bcl-2 to prevent the
mitochondrial apoptotic events. We found that Bcl-2 overexpression
prevented significantly TNF-induced loss of mitochondrial potential
in arrested cells (figure 6A). To see if
Bcl-2-mediated inhibition of the mitochondrial pathway corresponds
with an effective inhibition of the cell death, we analyzed two
late apoptotic events. We found that overexpression of Bcl-2
prevented both the appearance of cells with apoptotic morphology
(figure 6B) and
the loss of the cell membrane integrity (figure 6C) in cells
arrested in the G1/S border and treated with TNF for 24
hours. Similar results were obtained after TNF/Act-D treatment of
unsynchronized cells overexpressing Bcl-2 (figure 6D, E, F). These
data indicate that the mitochondrial pathway is an essential step
in TNF-induced apoptosis in NIH 3T3 sensitized either by cell cycle
arrest or RNA synthesis inhibition, even though the kinetics of
cytochrome c release and mitochondrial collapse appear to be quite
different in both conditions.
Discussion
TNF is a pleiotropic cytokine that is at the origin many different
types of responses including proliferation or cell death. This
variety of effects depends strongly on the cell type. Thus in many
tumoral cells, TNF can induce cell death while stimulating the
growth of normal cell types [1]. TNF can also induce both effects
in the same cell type [5, 10] indicating the influence of other
factors in the type of response to TNF. Moreover in some cases, a
relationship between the anti-proliferative effects of TNF and
progression through the cell cycle has been observed [11]. It is
already known that RNA synthesis inhibition or the repression of
NFκB activation alters the response of cells to TNF, so that cells
become sensitive to the toxic action of TNF [5-7]. In the present
study, we show that the TNF-induced mitogenic effect can be
superseded by the apoptotic signals when cell growth is inhibited
at different points of the cell cycle.
TNF treatment of NIH 3T3 cells arrested in the G0
phase induces a significant increase in the percentage of cells
that reach the S phase of the cell cycle (figure 1). However, rather
than triggering cell proliferation, the final effect of the TNF is
a significant decrease of the number of cells (figure 2). Moreover, TNF
induces cell death of G0 phase-arrested in a
dose-dependent manner (data not shown). These data agree with those
previously observed in 3T3-L1 pre-adipocites [22]. Based on these
observations, three models of the relationship between mitogenic
and cytotoxic signals induced by TNF in G0-arrested
cells may be proposed: 1) TNF-induced mitogenic signals could be
part of the toxic effect. In this model, TNF would favour entry
into the S phase and would subsequently induce cell death at some
point of the S, G2 or M phases. In this case, cell death
would affect those cells that had incorporated BrdU, while cells
remaining at the G0 phase would be protected. 2) TNF
would induce mitogenic signals. However, the shortage of some
essential growth factors required for TNF-induced cell
proliferation in the low serum condition, would not allow the whole
population to enter into the cell cycle. The inability of cells
that have received TNF mitogenic signals to grow would trigger the
cytotoxic signal. In this case, death would affect
non-proliferating cells, i.e. those that remain in the
G0 phase. 3) The mitogenic and the cytotoxic signals
would be completely independent. In this case, death would affect
equally proliferating and non-proliferating cells.
By the simultaneous analysis of the loss of mitochondrial
potential and the DNA content (figure 3), we found that
death occurs specifically in non-proliferating cells, i.e. those
that remained in the G0 phase, while for those cells
presenting at the S phase, DNA content was protected. These data
agree with the second of the proposed models. The fact that the
G0-arrested cells treated with TNF in the presence of
serum are resistant to the toxic action of TNF (figure 1) is also in
agreement with this model. In a rich medium, TNF mitogenic signals
would not be blocked by the lack of a given growth factor and the
toxic effect would not take place. Also in agreement with this
hypothesis is the fact that cells can be protected against
TNF-induced toxicity by the expression of several growth factors,
such as PDGF-B or bFGF [23, 24]. Moreover, TNF-induced
proliferation requires the synthesis of GM-CSF [25].
NIH 3T3 cells also became sensitive to TNF when arrested in
other phases of the cell cycle (G1/S, S or M) (figure 4). On the
other hand, exponentially growing cells treated with TNF/Act-D lost
mitochondrial potential in a similar proportion in all phases of
the cell cycle (figure
2B, C, D). Therefore, TNF-induced toxicity is more likely
related to the inability to grow rather than a particular phase of
the cell cycle. However, there are some contradictory data in this
regard. Human cervix carcinoma HeLa cells arrested in the S phase,
but not in G1, are more sensitive to the toxic effect of
TNF [14]. On the other hand, S phase-block does not sensitize the
murine fibrosarcoma L929 cell line to TNF [26]. These differences
could be explained by the fact that cell cycle checkpoints are
frequently disrupted in cancer cells and this could lead to a lack
of sensitivity of cells arrested in a particular phase of the cell
cycle to TNF. Non-transformed cells, such as NIH 3T3 on the other
hand, became sensitive to TNF after any kind of arrest as they
retained all the cell cycle controls.
In the present work, we describe for the first time, different
aspects of the TNF apoptotic pathway in cells sensitized by cell
cycle-arrest. In particular, we have studied cells arrested in the
G1/S border by treatment with aphidicolin. We found
interesting differences in the kinetics of the activation of the
mitochondrial apoptotic events depending of the type of
sensitization to TNF. In cells treated with TNF/Act-D, the loss of
mitochondrial potential and the release of cytochrome c are both
completed during the first six hours of treatment. On the other
hand, in arrested cells, the loss of mitochondrial potential shows
an significant delay compared to the release of cytochrome c (figure 5). These
results suggest that both forms of sensitization lead to TNF
cytotoxicity, but by different mechanisms. This observation is also
supported by the fact that the TNF-induced cell death in
G0-arrested NIH 3T3 cells is significantly accelerated
when transcription is inhibited by Act-D (data not shown).
Different types of relationship between the loss of mitochondrial
potential and the release of cytochrome c, depending on the
apoptotic system, have been reported. In some cases, the loss of
mitochondrial potential is an early event and coincides with
cytochrome c release, as we observed after TNF/Act-D treatment.
However in other models, is only a later sign of apoptosis, similar
to the case of arrested cells treated with TNF (see [27] for a
review). Moreover, the overexpression of Bcl-2 both in
G1/S-arrested cells and in cycling cells treated with
TNF/Act-D prevents not only the loss of mitochondrial potential
induced by TNF but also other apoptotic hallmarks such as the
appearance of apoptotic morphology and the loss of plasma membrane
integrity. This suggests that in these cells, the activation of the
mitochondrial pathway is indispensable for TNF-mediated induction
of apoptosis after both types of sensitization [9].
Similarly, it has been previously reported that the inhibition
of RNA synthesis and the repression of NFκB activation also caused
sensitivity to TNF by different pathways [6]. We checked if the
TNF-induced toxicity in arrested cells is due to a defective
activation of NFκB protective pathway. We found that NFκB is
equally activated by TNF in cycling cells and in cells arrested in
the different phases of the cell cycle (figure 5), in agreement
with previous reports [28]. Moreover, the inhibition of NFκB in
arrested cells further increases the extent of TNF-induced
apoptosis, suggesting that the blockage of cell cycle progression
and the inhibition of NFκB are two independent mechanisms for
sensitization to TNF.
In summary, we show here that TNF induces proliferation signals
in non-transformed NIH 3T3 cells, which can lead to apoptosis when
the cell growth is impaired at any phase of the cell cycle. These
findings should be taken in account when investigating the ability
of TNF to enhance the toxicity of cancer chemotherapeutic agents
such as doxorubicin [29, 30]. This, and other drugs, can induce
cell cycle-arrest in non-transformed cells [31, 32], which in turn
could increase the risk of unwanted toxicity in normal cells after
TNF treatment. We have also found that the apoptotic mechanism
triggered by TNF in the arrested cell is different from that
previously described for other known models of sensitization,
inhibition of RNA synthesis or repression of NFκB activation.
Acknowledgments
This work was supported by Grants SAF98-0174, SAF2001-3473 and
SAF2002-0388 from the Spanish CICYT, and by Grant PB-MED01-99 from
FICYT. VMC was supported by a fellowship from FICYT, RR was
recipient of a fellowship from the Spanish Ministry of Education
and MTC was supported by the Fundación Científica of the AECC.
References
1 Sugarman BJ, Aggarwal BB, Hass PE, Figari IS,
Palladino Jr. MA, Shepard HM. Recombinant human
tumor necrosis factor-alpha: effects on proliferation of normal and
transformed cells in vitro. Science 1985; 230: 943-5.
2 Beyaert R, Van Loo G, Heyninck K,
Vandenabeele P. Signaling to gene activation and cell death by
tumor necrosis factor receptors and Fas. Int Rev Cytol 2002; 214:
225-72.
3 Goldring MB, Goldring SR. Cytokines and cell growth
control. Crit Rev Eukaryot Gene Expr 1991; 1: 301-26.
4 Kirillova I, Chaisson M, Fausto N. Tumor
necrosis factor induces DNA replication in hepatic cells through
nuclear factor kappaB activation. Cell Growth Differ 1999; 10:
819-28.
5 Xu Y, Bialik S, Jones BE, et al. NF-kappaB
inactivation converts a hepatocyte cell line TNF-alpha response
from proliferation to apoptosis. Am J Physiol 1998; 275:
C1058-C1066.
6 Jones BE, Lo CR, Liu H, et al. Hepatocytes
sensitized to tumor necrosis factor-alpha cytotoxicity undergo
apoptosis through caspase-dependent and caspase-independent
pathways. J Biol Chem 2000; 275: 705-12.
7 Van Antwerp DJ, Martin SJ, Verma IM,
Green DR. Inhibition of TNF-induced apoptosis by NF-kappa B.
Trends Cell Biol 1998; 8: 107-11.
8 Wang CY, Mayo MW, Baldwin Jr. AS. TNF- and
cancer therapy-induced apoptosis: potentiation by inhibition of
NF-kappaB. Science 1996; 274: 784-7.
9 Scaffidi C, Fulda S, Srinivasan A, et al.
Two CD95 (APO-1/Fas) signaling pathways. EMBO J 1998; 17:
1675-87.
10 Manchester KM, Heston WD, Donner DB. Tumour
necrosis factor-induced cytotoxicity is accompanied by
intracellular mitogenic signals in ME-180 human cervical carcinoma
cells. Biochem J 1993; 290: 185-90.
11 Shih SC, Stutman O. Cell cycle-dependent tumor
necrosis factor apoptosis. Cancer Res 1996; 56: 1591-8.
12 Hu X, Bryington M, Fisher AB, et al.
Ubiquitin/Proteasome-dependent Degradation of D-type Cyclins Is
Linked to Tumor Necrosis Factor-induced Cell Cycle Arrest. J Biol
Chem 2002; 277: 16528-37.
13 Belizario JE, Dinarello CA. Interleukin 1,
interleukin 6, tumor necrosis factor, and transforming growth
factor beta increase cell resistance to tumor necrosis factor
cytotoxicity by growth arrest in the G1 phase of the cell cycle.
Cancer Res 1991; 51: 2379-85.
14 Meikrantz W, Gisselbrecht S, Tam SW,
Schlegel R. Activation of cyclin A-dependent protein kinases
during apoptosis. Proc Natl Acad Sci USA 1994; 91: 3754-8.
15 Coffman FD, Haviland DL, Green LM,
Ware CF. Cytotoxicity by tumor necrosis factor is linked with
the cell cycle but does not require DNA synthesis. Growth Factors
1989; 1: 357-64.
16 Rodriguez R, Meuth M. Chk1 and p21 cooperate to
prevent apoptosis during DNA replication fork stress. Mol Biol Cell
2006; 17: 402-12.
17 Harvey KJ, Lukovic D, Ucker DS.
Membrane-targeted green fluorescent protein reliably and uniquely
marks cells through apoptotic death. Cytometry 2001; 43: 273-8.
18 Claudio E, Segade F, Wrobel K, Ramos S,
Bravo R, Lazo PS. Molecular mechanisms of TNFalpha
cytotoxicity: activation of NF-kappaB and nuclear translocation.
Exp Cell Res 1996; 224: 63-71.
19 Campa VM, Iglesias JM, Carcedo MT, et al.
Polyinosinic acid induces TNF and NO production as well as
NF-kappaB and AP-1 transcriptional activation in the
monocytemacrophage cell line RAW 264.7. Inflamm Res 2005; 54:
328-37.
20 Van Antwerp DJ, Martin SJ, Kafri T,
Green DR, Verma IM. Suppression of TNF-alpha-induced
apoptosis by NF-kappaB. Science 1996; 274: 787-9.
21 Jiang W, Hunter T. Analysis of cell-cycle profiles
in transfected cells using a membrane-targeted GFP. Biotechniques
1998; 24: 349-50; (352, 354).
22 Lyle RE, Richon VM, McGehee Jr. RE.
TNFalpha disrupts mitotic clonal expansion and regulation of
retinoblastoma proteins p130 and p107 during 3T3-L1 adipocyte
differentiation. Biochem Biophys Res Commun 1998; 247: 373-8.
23 Aggarwal BB, Pocsik E, Totpal K. Transfection
of cells with basic fibroblast growth factor and Kaposi fibroblast
growth factor genes induce resistance to and receptor modulation of
tumor necrosis factor. FEBS Lett 1995; 372: 44-8.
24 Aggarwal BB, Pocsik E, Totpal K,
Ali-Osman F. Suppression of antiproliferative effects of tumor
necrosis factor by transfection of cells with human
platelet-derived growth factor B/c-sis gene. FEBS Lett 1995; 357:
1-6.
25 Quentmeier H, Dirks WG, Fleckenstein D,
Zaborski M, Drexler HG. Tumor necrosis
factor-alpha-induced proliferation requires synthesis of
granulocyte-macrophage colony-stimulating factor. Exp Hematol 2000;
28: 1008-15.
26 Vercammen D, Vandenabeele P, Beyaert R,
Declercq W, Fiers W. Tumour necrosis factor-induced
necrosis versus anti-Fas-induced apoptosis in L929 cells. Cytokine
1997; 9: 801-8.
27 Ly JD, Grubb DR, Lawen A. The mitochondrial
membrane potential (Δψm) in apoptosis; an update. Apoptosis 2003;
8: 115-28.
28 Duckett CS, Perkins ND, Leung K,
Agranoff AB, Nabel GJ. Cytokine induction of nuclear
factor kappa B in cycling and growth-arrested cells. Evidence for
cell cycle-independent activation. J Biol Chem 1995; 270:
18836-40.
29 Cao W, Chi WH, Wang J, Tang JJ,
Lu YJ. TNF-alpha promotes Doxorubicin-induced cell apoptosis
and anti-cancer effect through downregulation of p21 in
p53-deficient tumor cells. Biochem Biophys Res Commun 2005; 330:
1034-40.
30 Sleijfer S, Le TK, de Jong S,
Timmer-Bosscha H, Withoff S, Mulder NH. Combined
cytotoxic effects of tumor necrosis factor-alpha with various
cytotoxic agents in tumor cell lines that are drug resistant due to
mutated p53. J Immunother 1999; 22: 48-53.
31 Lee SM, Youn B, Kim CS, Kang C,
Kim J. Gamma-irradiation and doxorubicin treatment of normal
human cells cause cell cycle arrest via different pathways. Mol
Cells 2005; 20: 331-8.
32 Robles SJ, Buehler PW, Negrusz A,
Adami GR. Permanent cell cycle arrest in asynchronously
proliferating normal human fibroblasts treated with doxorubicin or
etoposide but not camptothecin. Biochem Pharmacol 1999; 58:
675-85.
|
Version imprimable
Version PDF
