| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Centre National de la Recherche Scientifique, UMR1599, Institut Gustave Roussy, F-94805 Villejuif, France;
Assistance Publique-Hôpitaux de Paris, Service de Néphrologie B, Hôpital Tenon, F-75020, France;
Case Western Reserve University, Department of Anatomy, School of Medicine, Cleveland, Ohio 44106, USA; and
The Amgen Institute and Ontario Cancer Institute, Department of Medical Biophysics and Immunology, University of Toronto, Toronto, Ontario M5G 2C1, Canada
1Correspondence: CNRS-UMR 1599, Institut Gustave Roussy, Pavillon de Recherche I, 39, rue Camille-Desmoulins, F-94805 Villejuif, France. E-mail: kroemer{at}igr.fr
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
1–100) results in the primary cytosolic
accumulation of AIF. AIF 1–100-induced cell death is suppressed by
neither Z-VAD.fmk or by Bcl-2. Thus, extramitochondrially targeted AIF
is a dominant cell death inducer.—Loeffler, M., Daugas, E., Susin,
S. A., Zamzami, N., Métivier, D., Nieminen, A.-L., Brothers,
G., Penninger, J. M., Kroemer, G. Dominant cell death induction by
extramitochondrially targeted apoptosis-inducing factor.
Key Words: AIF • apoptosis • Bcl-2 • caspases • cytochrome c
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
50 kbp) DNA
fragmentation. Under normal circumstances, transcription and
translation of the nuclear AIF gene give rise to a 67 kDa precursor
molecule that carries a putative mitochondrial localization sequence
(MLS) in its NH2 terminus. Upon import into the
mitochondrial intermembrane space, this 100 amino acid presequence is
cleaved by a local peptidase, leading to generation of the mature AIF
molecule (57 kDa, AIF 1–100), which is confined to the
mitochondrial intermembrane space (1
2
3)
. AIF is thus kept
in check by rigorous compartmentalization, preventing its interaction
with extramitochondrial targets (3
4
5)
. Upon induction of
apoptosis, mitochondria release AIF 1–100 and other soluble
proteins from the intermembrane space, including cytochrome
c (Cyt-c) (6
7
8
9
10
11)
. Cyt-c can interact with
nonmitochondrial proteins, including Apaf-1, thereby provoking its
oligomerization and/or a change in conformation that converts the
Apaf-1/Cyt-c complex into a caspase-9 activator (12)
. In
contrast to Cyt-c (which induces nuclear apoptosis via a cascade
involving caspase-9, caspase-3, and caspase-activated DNase, CAD)
(12)
, AIF-induced changes are caspase independent
(1
2
3)
. AIF 1–100 possesses several nuclear
localization sequences, allowing for its nuclear import. Inhibition of
nuclear import prevents the local effects of AIF (1)
. One
major difference between AIF-induced nuclear apoptosis and
Cyt-c/caspase/CAD-induced nuclear apoptosis resides in the morphology
of chromatin condensation. AIF causes a rather partial condensation
without shrinkage (‘stage I’) (1)
, whereas CAD can
induce a more advanced condensation pattern (stage II) without (stage
IIa) or with (stage IIb) formation of nuclear apoptotic bodies
(13
14
15)
. Another difference resides in the degree of
chromatin degradation induced by AIF and CAD. AIF triggers
(caspase-independent) large-scale (50 kbp) DNA fragmentation
(1
, 2)
, whereas CAD causes a classical mono- or
oligonucleosomal pattern of DNA fragmentation to multiples of 200 bp
(12
, 14
, 15)
. It is conceivable that AIF and the
Cyt-c/caspase/CAD pathways act in a complementary fashion to provide
several molecular links between mitochondrial membrane permeabilization
and nuclear apoptosis. The mitochondrial release of both AIF and Cyt-c
is at least in part under the common control of members of the Bcl-2
family. Thus, the anti-apoptotic protein Bcl-2, which localizes to
intracellular membranes including those of mitochondria, inhibits the
translocation of both AIF and Cyt-c (3
, 7
, 16
, 17)
. To further investigate the putative contribution of AIF to apoptosis, we generated AIF constructs fused to the green fluorescent protein (GFP), thus allowing for the continuous monitoring of the subcellular localization of AIF. As shown here, AIF constructs targeted to the mitochondrial intermembrane space rapidly redistribute to an extramitochondrial localization upon apoptosis induction. Transfection-enforced overexpression of such constructs can by itself induce apoptosis, via a process that involves spontaneous translocation of the proteins from mitochondria and is retarded by mitochondrion-targeted Bcl-2. A truncated AIF construct lacking the MLS exhibits a nonmitochondrial pattern of distribution and triggers caspase-independent apoptosis that is not inhibited by Bcl-2. Thus, extramitochondrially targeted AIF is a dominant cell death inducer.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
. To generate the vector pcDNA3.1+GFP,
pcDNA3.1+ (Invitrogen, Carlsbad, Calif.) was cut and ligated with
the KpnI/NotI fragment of pEGFP (Clontech, Palo
Alto, Calif.). PCR fragments coding for the different AIF
sequences were generated using pcDNA3.1+AIF or pcDNA3.1+AIF exB as a
template (1)
and the following primers for the
NH2 terminus of the protein: 5'-ggg gta ccc cga
aat gtt ccg gtg tg-3' (for full-length AIF), 5'-ggg gta ccc cat gtt agg
act gtc-3' (for the AIF 1–100 deletion mutant), and for the
carboxyl terminus of all AIF constructs: 5'-ggg gta ccc ctt cat gaa tgt
tga aga g-3'.The introduced KpnI-sites were then used to
ligate AIF into the KpnI-cut pcDNA3.1+GFP. The Cyt-c-GFP
construct has been described (18)
. PCR-based,
site-directed mutagenesis of the AIF cDNA was followed by verification
of the entire AIF cDNA sequence.
Cell lines and apoptosis induction
COS or Rat-1 cells stably transfected with a control vector
(CMV), with a human Bcl-2/Act A fusion protein specifically targeted to
mitochondria or a human Bcl-2/Cb5 fusion protein specifically targeted
to the endoplasmic reticulum (19)
were cultured in
Dulbecco’s modified Eagle’s medium supplemented with L-glutamine,
antibiotics, 2 mM pyruvate, and 10% fetal calf serum. Cells were
cultured in the presence of staurosporine (STS; 1 µM; Sigma, St.
Louis, Mo.), etoposide (20 µM), ganglioside GD3 (200 µM),
doxorubicin (20 µM), and/or Z-VAD.fmk (100 µM; 30 min before STS)
as indicated. If necessary, Z-VAD.fmk was added periodically in 24 h intervals. Apoptosis induction was monitored by incubation with the
DNA-specific dye Hoechst 33342 (2 µM). Nuclei with rippled contours
and partial chromatin condensation were considered to represent stage I
of chromatin condensation, nuclei with marked peripheral chromatin
condensation as stage IIa, and cells with nuclear bodies as stage IIb.
Transient transfection protocol
For transient transfections, 107 COS or
Rat-1 cells were trypsinized, centrifuged (300 g, 5 min),
resuspended in 0.2 ml of complete medium, and mixed with 5 µg of
plasmid DNA. After 10 min of incubation on ice in a 0.4 cm cuvette
(Bio-Rad, Hercules, Calif.), electroporation was performed at 960 µF
and 220V for 70–90 ms (GenePulser II, Bio-Rad). Cells were immediately
diluted with 10 ml of complete medium. For fluorescence microscopy,
105 cells/well were cultured on glass coverslips
(100 µm, 18 mm Ø, Polylabo, Strasbourg, France) in 12-well plates
(Polylabo) and washed with complete medium 3 h after transfection
to eliminate cell debris.
Microinjection experiments
For microinjection, COS cells were transiently transfected with
either AIF-GFP or Cyt-c-GFP and cultured overnight on glass coverslips.
The setup for the injection itself was as follows: 25 µM recombinant
horse Cyt-c (Sigma) or 7.5 µM recombinant AIF protein (purified as
described in ref 1
) were injected in PBS for 0.2 s
under a pressure of 150 hPa, using a Microinjector equipment
(Eppendorf, Hamburg, Germany). In some experiments, cells were
preincubated with 100 µM Z-VAD.fmk (Bachem, Basel, Switzerland)
before microinjection. Then cells were culture for 3 h in the
presence or absence of Z-VAD.fmk and fixed for fluorescence microscopy,
as described below.
Immunostaining protocols
A rabbit antiserum generated against a mixture of 3 peptides
derived from the mouse AIF amino acid sequence (residues 151–170,
166–185, 181–200; ELISA titer 10.000) was used (diluted 1/1000) on
paraformaldehyde- (4% w:v) and picric acid-fixed (0.19% v:v) cells
cultured on 100 µm coverslips, and revealed with a goat anti-rabbit
IgG conjugated to phycoerythrin (PE) (Southern Biotechnology,
Birmingham, Ala.). Control experiments performed with the preimmune
antiserum or in the presence of an excess (100 µM) of the three
immunogenic peptides confirmed that all detectable fluorescence was
specific (not shown). Cytochrome c was detected by means of
the mAb 6H2.B4 (PharMingen, San Diego, Calif.), revealed by a goat
anti-mouse IgG PE conjugate (Southern Biotechnology).
Conventional and confocal laser scanning microscopy
Conventional examination of samples was performed in a Leitz
Labolux S microscope equipped with standard filters for FITC/GFP, PE,
and Hoechst 33342, as well as a Leica camera. Confocal microscopy was
performed on a Leica TC-SP (Leica Microsystems, Heidelberg, Germany)
equipped with an ArKr laser mounted on an inverted Leica DM IFBE
microscope with an 63 x 1.32 NA oil objective. A minimum of 200
cells were monitored for these parameters for each data point.
Cell-free system of apoptosis
Purified HeLa nuclei were incubated with recombinant AIF protein
(final concentration 1 µg/ml; ref 1
), preincubated in
the absence or presence of 1 mM diphenyleneiodonium for 30 min) in CFS
buffer for 90 min (20)
, and nuclear apoptosis was
quantitated by staining with DNA-intercalating propidium iodide,
followed by cytofluorometric determination of DNA content
(20)
.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
;
Fig. 1
). These results, which were originally obtained with the murine AIF
gene, were confirmed for human AIF, yielding two very similar sequence
variants, suggesting that these sequence variations were not due to a
cloning artifact (Fig. 1B
). Comparison of the AIF and
AIF-exB sequences with the human AIF genome sequence revealed that this
difference was due to alternative exon usage (Fig. 1A
)
affecting part of the putative 100 amino acid amino-terminal MLS.
Both exons possessed significant homology in their carboxyl-terminal
moiety (Fig. 1B
). When GFP was fused to the carboxyl
terminus of AIF (AIF-GFP) or AIF-exB (AIF-exB-GFP) to generate chimeric
proteins, transient transfection of COS cells revealed that the
GFP-dependent fluorescence was targeted to mitochondria (Fig. 2
), as has been shown for a Cyt-c-GFP fusion construct (18
;
Fig. 2
). In contrast, a truncated AIF-GFP fusion protein in which the
amino-terminal MLS ( 1–100) was removed exhibited a diffuse
cytoplasmic staining comparable to that obtained with GFP alone (Fig. 2)
. These observations confirm the existence of an MLS in the
NH2 terminus of AIF and indicate that the
alternative exon 2 usage does not affect the mitochondrial import of
AIF.
|
|
Chronological relationship between AIF and Cyt-c release in
STS-induced apoptosis
Upon addition of the apoptosis inducer STS, mitochondrially
targeted AIF-GFP (and AIF-exB-GFP) translocate to an extramitochondrial
localization (Fig. 2)
. Similar results are obtained with other
apoptosis inducers including the proapoptotic second messenger
ganglioside GD3 (21
, 22)
as well as two chemotherapeutic
agents (doxorubicin and etoposide) (Fig. 3
). The translocation of AIF-GFP can be observed in cells that still lack
signs of fluorescence-detectable chromatin condensation (Fig. 3)
.
Immunostaining of Cyt-c-GFP-transfected cells (green fluorescence) with
an anti-AIF antibody (revealed by a PE conjugate, red fluorescence)
indicated that the translocation of AIF and Cyt-c occurs simultaneously
after STS addition (Fig. 4
). Similar results were obtained when AIF-GFP-transfected cells were
stained with an anti-Cyt-c antibody (not shown), thus excluding the
possibility that the fusion with GFP might affect the kinetics of AIF
or Cyt-c release. Staining of AIF-GFP (or Cyt-c) -transfected cells
with anti-AIF (or Cyt-c) antibodies yielded a similar distribution
for the GFP fusion protein and the immunodetectable (endogenous+transgene-encoded) protein (not shown; see ref
18
), further arguing against the possibility that the GFP
moiety might affect the kinetics of AIF redistribution. Cyt-c-GFP and
AIF-GFP translocation was observed in all STS-treated cells having
undergone incipient (stage I) or advanced (stage II) chromatin
condensation, without (stage IIa) or with formation of nuclear
apoptotic bodies (stage IIb) (Fig. 5A
B
). In addition, a fraction of cells (40%) not having
yet undergone Hoechst 33342-detectable chromatin condensation
manifested the mitochondrial release of AIF-GFP or Cyt-c-GFP (Fig. 5B
). The pan-caspase inhibitor Z-VAD.fmk failed to prevent
STS-induced AIF-GFP or Cyt-c-GFP release, although it did affect the
transition from stage I to stage II of chromatin condensation (Fig. 5B
). In conclusion, in COS cells, Cyt-c and AIF are released
simultaneously in a caspase-independent fashion before chromatin
condensation occurs.
|
|
|
Functional relationship between AIF and Cyt-c release
To further investigate the putative functional relationship
between the release of AIF and Cyt-c, COS cells transfected with either
AIF-GFP or Cyt-c-GFP were microinjected with recombinant AIF or Cyt-c,
followed by the determination of nuclear morphology and quantitation of
the translocation of AIF-GFP or Cyt-c-GFP (Fig. 6)
. Microinjection of both AIF and Cyt-c resulted in the induction of
nuclear apoptosis (Fig. 6A
) and in the translocation of
AIF-GFP and Cyt-c-GFP (Fig. 6B
), indicating that the ectopic
(nonmitochondrial) presence of both proteins suffices to cause
permeabilization of the outer mitochondrial membrane. Both AIF and
AIF 1–100 had a similar effect on nuclear morphology
(1)
and mitochondria (not shown), indicating that these
effects do not depend on the MLS. When introduced into the cytosol, AIF
can trigger the release of mitochondrial AIF, as well as that of Cyt-c,
in a reaction that is not affected by the pan-caspase inhibitor
Z-VAD.fmk (Fig. 6B
). AIF triggered nuclear apoptosis, both
stage I and II. Z-VAD.fmk largely prevented the occurrence of stage II
apoptosis, yet had no effect on stage I (Fig. 7
), indicating that AIF-induced stage I chromatin condensation occurs in
a caspase-independent fashion, whereas its advancement to stage II is
caspase dependent, in accord with previous observations (3
, 23)
. All Cyt-c induced effects were inhibited by Z-VAD.fmk at
both nuclear (Fig. 6A
) and mitochondrial levels (Fig. 6B
), confirming that Cyt-c is acting through the activation
of caspases. Altogether, these data suggest the existence of a
feed-forward loop in which the two mitochondrial intermembrane proteins
Cyt-c and AIF cause the release of further intermembrane proteins,
either in a caspase-independent (AIF) or caspase-dependent (Cyt-c)
fashion.
|
|
Caspase-independent cell death induction by transfection-enforced
AIF expression
The results discussed above were obtained shortly (16–24 h) after
transfection of COS cells with AIF-GFP or Cyt-c-GFP constructs, that
is, at a stage at which these proteins are largely mitochondrial and
<10% of the cells exhibit spontaneous translocation of AIF-GFP or
Cyt-c-GFP to the rest of the cell (Fig. 7A
). If cells were
cultured for a longer period (48–144 h), however, an increasing
percentage of transfected cells lost the mitochondrial pattern of
AIF-GFP, AIF-exB-GFP or Cyt-c-GFP and rather manifested a diffuse
distribution of GFP fusion proteins (Fig. 7A
).
Cytofluorometric assessment of the percentage of GFP-positive cells
revealed a progressive disappearance of viable cells expressing
AIF-GFP, AIF-exB-GFP, AIF- 1–100-GFP, or Cyt-c-GFP, compared with
cells transfected with GFP only (Fig. 7B
), suggesting that
these proteins are actually lethal. Accordingly, >50% of the cells
expressing AIF-GFP, AIF-exB-GFP, AIF- 1–100-GFP, or Cyt-c-GFP
exhibited features of nuclear apoptosis 96 h after
transfection (Fig. 7C
). This effect was only partially
inhibited by Z-VAD.fmk (Fig. 7C
), which failed to
prevent the appearance of condensed chromatin. However, Z-VAD.fmk
(added each 24 h at a dose of 100 µM) did prevent the
progression of nuclear apoptosis from stage I to stage II (Fig. 7C
), thus providing an internal control for its efficacy. An
increase in the frequency of Z-VAD.fmk administrations or an increase
in the dose of Z-VAD.fmk did not ameliorate the degree of inhibition
(not shown), underscoring that the AIF-GFP constructs used in this
study can induce a type of chromatin condensation and cell death that
does not rely on the function of Z-VAD.fmk-inhibitable caspases.
AIF induces apoptosis independently from its oxidoreductase
activity
Based on sequence comparisons with flavoproteins whose structure
has been determined by X-crystallography, computer-assisted structural
analysis (2)
, two conserved hexapeptide motifs critical
for binding of nicotine adenine dinucleotide (NAD) and flavine adenine
dinucleotide (FAD) were identified in the AIF protein: 303TVIGGG308 and
255CLIATG260, respectively. On theoretical grounds, mutations of these
NAD/FAD-binding motifs by insertion of trialanine stretches (mutants 1
and 2, respectively) should abolish binding of the prosthetic groups
required for electron transfer and hence abrogate the oxidoreductase
activity of AIF. Transfection of COS cells with wild-type AIF-GFP and
mutant AIF-GFP constructs induced a similar level of initially
mitochondrial AIF-GFP expression, followed by spontaneous AIF-GFP
translocation and apoptosis (Fig. 8A
). These data suggest that the apoptogenic effect of AIF
does not rely on its oxidoreductase activity. To confirm this
hypothesis in another experimental system, recombinant AIF protein was
preincubated with diphenyleneiodonium, an inhibitor of
flavonoid-containing enzymes covalently reacting with FAD
(24
25
26)
. Diphenyleneiodonium-pretreated AIF and untreated
AIF had a similar capacity to induce DNA loss when added to purified
HeLa nuclei (Fig. 8B
), indicating that the nuclear effects
of AIF do not require the presence of a redox-active reaction center.
|
Mitochondrion-targeted Bcl-2 delays AIF-induced apoptosis unless
AIF is targeted to an extramitochondrial localization
Bcl-2 is a multifunctional inhibitor of apoptosis that has been
suggested to act either on mitochondria or on the endoplasmic reticulum
(ER) to confer cytoprotection (19
, 27
28
29)
. We have taken
advantage of Rat-1 cells expressing Bcl-2 fusion proteins targeted
specifically to mitochondria or the ER (19)
to evaluate
its capacity to prevent cell death induced by transfection with AIF-GFP
and Cyt-c-GFP constructs. ER-targeted Bcl-2 has no effect on the
spontaneous translocation of AIF-GFP, AIF-exB-GFP, or Cyt-c-GFP
(Fig. 9B
) and does not affect the induction of nuclear chromatin
condensation (Fig. 9A
). In contrast, mitochondrion-targeted
Bcl-2 does retard the translocation of AIF-GFP, AIF-exB-GFP, or
Cyt-c-GFP (Fig. 9B
) and concomitantly reduces the frequency
of nuclear apoptosis (Fig. 9A
). However, Bcl-2 has no effect
on the frequency of apoptosis induced by extramitochondrially targeted
AIF- 1–100-GFP (Fig. 9A
). Bcl-2 only delays the
translocation of AIF-GFP or AIF-exB-GFP, and no Bcl-2-mediated
inhibition is found on prolonged culture (48 h in Rat-1 cells, Fig. 9
).
Altogether, these data underscore the functional importance of local
mitochondrial effects for Bcl-2-mediated cytoprotection.
|
|
CONCLUDING REMARKS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
, are both targeted to
the mitochondrial intermembrane space, as shown by appropriate AIF-GFP
fusion constructs (Fig. 2)
. To our knowledge, this is the first example
of a splice variant affecting an MLS yet not affecting mitochondrial
import. The alternative exon 2b (which is highly homologous to exon 2a)
contains an uncharged yet relatively hydrophilic stretch of 23 amino
acids, starting at residue 59, flanked by charged regions. This motif
could be an alternative intermembrane space bipartite signal
(30)
. Upon induction of apoptosis by STS (Fig. 2)
,
doxorubicin, etoposide, or ganglioside GD3 (Fig. 3)
, both proteins
(AIF-GFP and AIF-exB-GFP) translocate indistinguishably from
mitochondria to the rest of the cell before signs of chromatin
condensation are detectable (Fig. 3
, Fig. 5B
).
The proapoptotic effect of AIF appears independent from its
oxidoreductase activity (Fig. 8)
, similar to what has been reported for
Cyt-c (31)
. It is thus tempting to speculate that AIF may
have a dual function as an ubiquitous (32)
, perhaps vital
oxidoreductase catalyzing electron transfer between cytochrome
c and NAD (S. A. Susin and G. Kroemer, unpublished
observation) in the intermembrane space and as an apoptosis
effector/regulator molecule in its extramitochondrial localization. On
the one hand, AIF may function as a death effect involved in
caspase-independent apoptosis. On the other hand, AIF may function, at
least in some pathways, as an upstream regulator of Cyt-c release and
caspase activation. In COS cells, the translocation of AIF (endogenous
or AIF-GFP) and Cyt-c (endogenous or Cyt-c-GFP) occurs in a
caspase-independent (that is, Z-VAD.fmk noninhibited) fashion (Fig. 5A
). Extramitochondrial AIF can trigger the mitochondrial
release of AIF-GFP and that of Cyt-c-GFP, and this effect does not rely
on caspase activation (Fig. 6)
, emphasizing the possibility that AIF
participates in a positive amplification loop in which partial release
of AIF triggers the release of further AIF. Such a feed-forward system
would accelerate the process of AIF release. Indeed, intermediate
stages of AIF release (in which some AIF would be still retained in
mitochondria) has rarely been (<1% of cells, not shown) observed by
confocal microscopy, suggesting that the translocation of AIF from
mitochondria to the rest of the cell is a rapid event occurring in an
all-or-nothing fashion.
Upon prolonged culture, the transfection-enforced overexpression of AIF
or AIF-exB led to a gradual increase of the frequency of cells in which
AIF diffusely distributed to the cytosol (Fig. 7A
),
correlating with nuclear apoptosis (Fig. 7C
) and cell death
(Fig. 7B
). On theoretical grounds, this could be due to a
saturation of the mitochondrial protein import pathway by accumulating
AIF protein or, alternatively, due to local AIF-mediated damage of the
mitochondrial outer membrane, leading to the release of AIF that
previously has been imported into mitochondria. We favor this latter
possibility, based on three observations: 1) the intensity
of AIF-GFP-dependent fluorescence did not increase in COS cells between
48 and 144 h after transfection, although the percentage of cells
exhibiting translocation greatly increased (Fig. 7A
);
2) mitochondrial vs. diffuse staining patterns obtained with
AIF-GFP appeared to be mutually exclusive; and 3) Bcl-2
overexpression (which stabilizes mitochondrial membranes, yet has no
reported effect on the import of proteins; see ref 29
, 33
, 34
) retarded the accumulation of AIF in the cytosol (Fig. 9B
).
When AIF is present ectopically in the extramitochondrial compartment,
it induces chromatin condensation, the first stage of which appears to
be caspase independent. This has been demonstrated in three different
experimental settings: 1) by microinjection of the
recombinant AIF protein (Fig. 6A
), 2) by
transfection of cells with mitochondrially targeted AIF which
spontaneously translocates after prolonged culture (Fig. 7C
,
Fig. 9A
), and 3) by transfection of cells with an
AIF mutant lacking the MLS ( 1–100) (Figs. 2
, 7B, C
, 9
).
AIF 1–100 causes apoptosis in a fashion that is not affected by
Z-VAD.fmk (Fig. 7C
) nor by Bcl-2 (Fig. 9)
, presumably by
directly entering the nucleus. Altogether, these data confirm in a
genetic system that AIF can cause apoptosis in a caspase-independent
fashion. Furthermore, they demonstrate the feasibility of engineering a
dominant apoptosis inducer (AIF- 1–100) that overcomes
Bcl-2-mediated apoptosis inhibition. Such dominant apoptosis inducers
could prove useful in the gene therapy-mediated ablation of cancer
cells overexpressing caspase inhibitors and/or anti-apoptotic
Bcl-2-like proteins.
|
ACKNOWLEDGMENTS |
|---|
Received for publication June 15, 2000.
Revision received August 11, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUDING REMARKSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. Balakrishnan, W. G. Wierda, M. J. Keating, and V. Gandhi Gossypol, a BH3 mimetic, induces apoptosis in chronic lymphocytic leukemia cells Blood, September 1, 2008; 112(5): 1971 - 1980. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Ishimura, G. R. Martin, and S. L. Ackerman Loss of Apoptosis-Inducing Factor Results in Cell-Type-Specific Neurogenesis Defects J. Neurosci., May 7, 2008; 28(19): 4938 - 4948. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Y. Churbanova and I. F. Sevrioukova Redox-dependent Changes in Molecular Properties of Mitochondrial Apoptosis-inducing Factor J. Biol. Chem., February 29, 2008; 283(9): 5622 - 5631. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. S. Collingwood, E. V. Smirnova, M. Bogush, N. Carpino, R. S. Annan, and A. Y. Tsygankov T-cell Ubiquitin Ligand Affects Cell Death through a Functional Interaction with Apoptosis-inducing Factor, a Key Factor of Caspase-independent Apoptosis J. Biol. Chem., October 19, 2007; 282(42): 30920 - 30928. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Gong, S. Hay, K. R. Marshall, A. W. Munro, and N. S. Scrutton DNA Binding Suppresses Human AIF-M2 Activity and Provides a Connection between Redox Chemistry, Reactive Oxygen Species, and Apoptosis J. Biol. Chem., October 12, 2007; 282(41): 30331 - 30340. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Chen, L. Yang, and T. M. McIntyre Cytotoxic Phospholipid Oxidation Products: CELL DEATH FROM MITOCHONDRIAL DAMAGE AND THE INTRINSIC CASPASE CASCADE J. Biol. Chem., August 24, 2007; 282(34): 24842 - 24850. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Viemann, K. Barczyk, T. Vogl, U. Fischer, C. Sunderkotter, K. Schulze-Osthoff, and J. Roth MRP8/MRP14 impairs endothelial integrity and induces a caspase-dependent and -independent cell death program Blood, March 15, 2007; 109(6): 2453 - 2460. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Vega-Manriquez, Y. Lopez-Vidal, J. Moran, L. G. Adams, and J. A. Gutierrez-Pabello Apoptosis-Inducing Factor Participation in Bovine Macrophage Mycobacterium bovis-Induced Caspase-Independent Cell Death Infect. Immun., March 1, 2007; 75(3): 1223 - 1228. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Kroemer, L. Galluzzi, and C. Brenner Mitochondrial Membrane Permeabilization in Cell Death Physiol Rev, January 1, 2007; 87(1): 99 - 163. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Munoz-Pinedo, A. Guio-Carrion, J. C. Goldstein, P. Fitzgerald, D. D. Newmeyer, and D. R. Green Different mitochondrial intermembrane space proteins are released during apoptosis in a manner that is coordinately initiated but can vary in duration PNAS, August 1, 2006; 103(31): 11573 - 11578. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Delettre, V. J. Yuste, R. S. Moubarak, M. Bras, N. Robert, and S. A. Susin Identification and Characterization of AIFsh2, a Mitochondrial Apoptosis-inducing Factor (AIF) Isoform with NADH Oxidase Activity J. Biol. Chem., July 7, 2006; 281(27): 18507 - 18518. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ruchalski, H. Mao, Z. Li, Z. Wang, S. Gillers, Y. Wang, D. D. Mosser, V. Gabai, J. H. Schwartz, and S. C. Borkan Distinct hsp70 Domains Mediate Apoptosis-inducing Factor Release and Nuclear Accumulation J. Biol. Chem., March 24, 2006; 281(12): 7873 - 7880. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Delettre, V. J. Yuste, R. S. Moubarak, M. Bras, J.-C. Lesbordes-Brion, S. Petres, J. Bellalou, and S. A. Susin AIFsh, a Novel Apoptosis-inducing Factor (AIF) Pro-apoptotic Isoform with Potential Pathological Relevance in Human Cancer J. Biol. Chem., March 10, 2006; 281(10): 6413 - 6427. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Joza, G. Y. Oudit, D. Brown, P. Benit, Z. Kassiri, N. Vahsen, L. Benoit, M. M. Patel, K. Nowikovsky, A. Vassault, et al. Muscle-Specific Loss of Apoptosis-Inducing Factor Leads to Mitochondrial Dysfunction, Skeletal Muscle Atrophy, and Dilated Cardiomyopathy Mol. Cell. Biol., December 1, 2005; 25(23): 10261 - 10272. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. R. Marshall, M. Gong, L. Wodke, J. H. Lamb, D. J. L. Jones, P. B. Farmer, N. S. Scrutton, and A. W. Munro The Human Apoptosis-inducing Protein AMID Is an Oxidoreductase with a Modified Flavin Cofactor and DNA Binding Activity J. Biol. Chem., September 2, 2005; 280(35): 30735 - 30740. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
|
