* Institut National de la Santé et de la Recherche Médicale U408, Faculté de Médecine Xavier Bichat, and
Service de Pneumologie, Hôpital Bichat, Paris, France
Correspondence: Marina Pretolani, Ph.D., INSERM U408, Faculté de Médecine Xavier Bichat, 16, rue Henri Huchard, 75018 Paris, France. E-mail: mpretol{at}bichat.inserm.fr
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
m) and with
caspase-8 and caspase-3 activation. Interleukin (IL)-5 and interferon
(IFN)-
, two cytokines known to prolong eosinophil survival,
inhibited Fas-mediated apoptosis and caspase activation but poorly
affected the decrease in m. Eosinophil incubation
with bongkrekic acid, an inhibitor of the mitochondrial permeability
transition pore (MPTP) opening, failed to modify Fas-mediated loss in
m, caspase activation, and apoptosis. In contrast,
caspase inhibitors markedly reduced eosinophil apoptosis without
significantly affecting m dissipation. We conclude
that caspase-8 and caspase-3 activation, but not MPTP opening, mediate
Fas-induced eosinophil apoptosis and are the main targets for the
protective effect of IL-5 and IFN-.
Key Words: cell death • asthma • granulocytes • m
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Factors that extend the life span of tissue eosinophils by favoring
their survival and inhibiting their apoptosis include T helper-1 and T
helper-2–derived cytokines such as interferon (IFN)-
[6
, 7
] and interleukin (IL)-5
[8
, 9
], respectively. Enhanced expression
and production of IL-5 in asthmatic airways has widely been documented
[10
]. Although the role of IFN- in tissue
eosinophilia and in allergic and inflammatory diseases is less defined
than that of IL-5, it has been suggested that this cytokine is involved
in the exacerbation of asthma observed in respiratory virus–infected
patients [11
].
Pro-apoptotic signals include the activation of Fas antigen, a transmembrane protein belonging to the tumor necrosis factor receptor family [12 ]. Fas expression by human peripheral blood and tissue eosinophils [13 14 15 16 17 ], as well as its ability to facilitate the elimination of airway eosinophils in vivo [15 , 18 ], has been extensively reported.
Fas ligation in several cell types involves the activation of caspases, a family of cysteine proteases with aspartate substrate specificity. Caspases are produced as inactive zymogens and activated by proteolytic cleavage [19 ]. Once activated, caspases cleave cytoplasmic and nuclear components, leading to cell dismantling, DNA degradation, and ultimately cell death [19 ].
Two different intracellular pathways have been described during
Fas-mediated apoptosis. These depend on the requirement of mitochondria
in the propagation of the death signal [20
].
Accordingly, Fas cross-linking by its natural ligand, Fas-ligand, or by
an agonistic antibody activates caspase-8, which, in turn, initiates a
caspase cascade, either by directly cleaving caspase-3 or by altering
mitochondrial homeostasis [20
, 21
]. This
latter phenomenon is characterized by the disruption of transmembrane
potential, defined as m, and by the opening of the
mitochondrial permeability transition pore (MPTP), which is thought to
be responsible for releasing apoptogenic factors, such as cytochrome
c [19
, 21
, 22
],
into the cytosol. Cytochrome c, in turn, activates cytosolic
factors that cleave caspase-3 [21
, 22
].
Scaffidi et al. [20
] suggested that activation of the
mitochondria-independent or mitochondria-dependent pathway during the
apoptotic process relates to the cell type. Although this concept has
been established for numerous cell lines, little is known about the
precise mechanisms that regulate Fas-induced apoptosis in eosinophils.
We wanted to determine the relative importance of caspase- and
mitochondria-dependent pathways in mediating Fas-induced eosinophil
apoptosis. In addition, we investigated the ability of IL-5 and
IFN-, two cytokines known to prolong eosinophil survival and inhibit
their apoptotic death [6
7
8
9
], to interfere with these
phenomena.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
500 eosinophils/µL). Each patient
provided a written consent, and each experiment was performed with
cells obtained from a single donor. Eosinophils were isolated by
immunomagnetic negative selection using a modification of the method of
Hansel et al. [23
]. Briefly, phosphate-buffered saline
(PBS)-diluted blood was centrifuged on Lymphocytes Separation Medium
(Eurobio, Les Ulis, France; density 1.077 g/mL) to remove mononuclear
cells. We lysed the remaining erythrocytes by incubating the
granulocyte pellet with an ice-cold isotonic ammonium-chloride
solution. In some experiments, a further elimination of mononuclear
cells with Percoll (Pharmacia Biotech, Orsay, France; density 1.082
g/mL) was required. Mixed granulocytes were then incubated with
magnetic microbead-conjugated anti-CD16 monoclonal antibody (mAb)
(Miltenyi Biotec, Paris, France) before unlabeled eosinophils were
eluted through a magnetic separation column (MACS system, Miltenyi
Biotec). The purity of the resulting cell population (>98%
eosinophils) was evaluated after cytospin preparations were stained
with Diff-Quik® dye (Merz Dade, Baxter Dade AG, Duedingen,
Switzerland).
Cell culture
Eosinophils were resuspended in RPMI 1640 (Life Technologies,
Cergy Pontoise, France), supplemented with antibiotic–antimycotic
solution (Life Technologies) and 10% fetal calf serum (Hyclone, Logan,
UT), and incubated at 37°C with 5% CO2 in a humidified
atmosphere for 0–24 h.
Eosinophils were stimulated with either 3–3000 ng/mL of agonistic
mouse anti-human Fas antigen mAb (clone 7C11, Beckman Coulter,
Villepinte, France) or 3000 ng/mL of its isotype matched mouse IgM
(Dako, Trappes, France), in the absence or presence of 0.0005–5 ng/mL
of recombinant human IL-5 (PeproTech Inc., Rocky Hill, NJ) or of
0.1–1,000 U/mL recombinant human IFN- (R&D Systems Europe Ltd.,
Oxon, U.K.). In a separate series of experiments, eosinophils were
preincubated for 2 h with 30 µM of the following permeant and
irreversible caspase inhibitors: Z-Ile-Glu-Thr-Asp-fluoromethylketone
(Z-IETD-fmk, caspase-8 inhibitor), Z-Asp-Glu-Val-Asp-fmk (Z-DEVD-fmk,
caspase-3 inhibitor), or Z-Val-Ala-Asp-fmk (Z-VAD-fmk, broad
spectrum caspase inhibitor, all from Calbiochem, Bad Soden, Germany),
or with their vehicle (i.e., a 0.4% solution of dimethyl sulfoxide
[DMSO]). Alternatively, 62 µM of bongkrekic acid (Calbiochem) or
its vehicle (i.e., a 0.02 N NH4OH solution [Sigma, St.
Quentin Fallavier, France]) was added to eosinophil cultures 2 h
before stimulation. This concentration of bongkrekic acid was selected
on the basis of previous studies showing its ability to prevent the
loss in m in neutrophilic granulocytes
[24
].
Viability and apoptosis determination by flow cytometry
For eosinophil viability, cells (0.15x106) were
washed twice in PBS and resuspended in PBS containing 2.5 µg/mL
propidium iodide (Sigma) immediately before analysis. For apoptosis
determination, eosinophils (0.15x106) were resuspended in
a hypotonic solution containing 0.1% (w/v) sodium citrate, 0.1% (v/v)
Triton X-100, and 50 µg/mL propidium iodide, as described
[25
]. In both cases, cells were applied to an Epics XL
flow cytometer (FL-3 photomultiplier, Beckman Coulter). A total of 5000
cells were analyzed with Expo 32 software (Beckman Coulter). The
percent of propidium iodide-unstained (viable) cells or propidium
iodide-low stained apoptotic nuclei, respectively, was determined.
Determination of m by flow cytometry
Changes in 
m were evaluated by detecting the
uptake of the lipophilic cationic dye, 3,3'-dihexyloxacarbocyanine
iodide (DiOC6(3), Molecular Probes Inc., Eugene, OR), which
accumulates into polarized mitochondria under the influence of
m [26
]. Eosinophils (0.3 x
106) were washed in PBS and incubated at 37°C for 30 min
in the presence of 40 nM of DiOC6(3) in RPMI
[27
]. Propidium iodide, at the final concentration of
2.5 µM, was added immediately before cytofluorimetric analysis to
exclude necrotic cells. Negative control experiments were performed by
preincubating cells (10 min, 37°C) with 200 µM of the mitochondrial
uncoupler, carbonyl cyanide m-chlorophenylhydrazone (CCCP, Sigma),
which collapses the m [27
,
28
]. A total of 5000 events were analyzed with an Epics
XL flow cytometer (DiOC6(3) emission in FL-1 and
propidium iodide in FL-3) using Expo 32 software. The percent of
DiOC6(3)-high stained cells was determined.
Apoptosis determination by electron microscopy
Ultrastructural changes induced by Fas stimulation were
determined by electron microscopy. Briefly, IgM- or anti-Fas (3000
ng/mL)-stimulated cytokine-untreated or cytokine-treated eosinophils
were fixed in 1.6% glutaraldehyde and postfixed in 2% osmic acid
(both from Sigma), as previously described [13
]. After
acetone dehydration, pellets were embedded into epoxy resin. Ultrathin
sections were contrasted with uranyl acetate and lead citrate and
examined with a JEM-100 CX II transmission electron microscope (Jeol,
Tokyo, Japan).
Preparation of eosinophil extracts
Eosinophils were lysed by sonication in 10 mM
N-(2-hydroxyethyl)piperazine-N'-(2-ethane-sulfonic acid), pH 8.0, 1%
nonidet P-40, 150 mM NaCl, 500 mM sucrose, 1 mM Na2EDTA, 2
mM phenylmethylsulfonyl fluoride, 4% ß-mercaptoethanol, 10 µg/mL
aprotinin, 10 µM leupeptin, and 10 µM pepstatin A (all from Sigma)
and subsequently centrifuged for 15 min at 12,000 g. Protein contents
in clarified supernatants were determined by comparison with an
ovalbumin standard curve (ICN Biomedical, Costa-Mesa, CA), using the
Biorad protein assay (Biorad, München, Germany).
Evaluation of caspase-8 and caspase-3 expression by Western blot
Protein extracts (25 µg) were resolved by 13% sodium dodecyl
sulfate-polyacrylamide gel electrophoresis, electroblotted on
polyvinylidene difluoride membranes (Biorad), and next reacted with a
1:10 dilution of mouse anti-human caspase-8 Ab (clone C15, a generous
gift from Dr. M. E. Peter, German Cancer Research Center,
Heidelberg, Germany), a 1:1000 dilution of rabbit anti-human caspase-3
Ab (BD Biosciences, Le Pont de Claix, France), or with the mouse
anti-ß-actin mAb (clone AC-74, Sigma) at a 1:4000 dilution.
Immunoblots were detected with peroxidase-conjugated donkey
anti-rabbit or sheep anti-mouse Abs, both at a 1:4000
dilution, using the ECL Western blotting detection system (all from
Amersham, Les Ulis, France). The intensities of the expression of
zymogen caspase-8, zymogen and cleaved-caspase-3, and ß-actin were
quantified using a densitometer (CCD-COHU, Japan) and Gel Analyst
software (Claravision, Orsay, France). Results are expressed as a
ratio, defined as the optic density (OD) values of specific caspase
bands/OD values of the corresponding ß-actin bands.
In vitro fluorogenic caspase-8 and caspase-3-like cleavage assay
The following protocol was adapted from Nicholson et al.
[29
]. Briefly, cleavage of 50 µM of the caspase-8
substrate, Z-IETD-7-amino-4-(trifluoromethyl)coumarin (Z-IETD-AFC,
Calbiochem), and the caspase-3 substrate,
acetyl-DEVD-7-amino-4-methyl coumarin (Ac-DEVD-AMC, Promega,
Charbonnieres, France), were measured in a caspase reaction
buffer consisting of 10 mM
N-(2-hydroxyethyl)piperazine-N'-(2-ethane-sulfonic acid), pH 7.5, 10%
sucrose, 0.1%
3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate
(Sigma). Aliquots (10 and 20 µg, for caspase-3 and caspase-8,
respectively) of eosinophil lysates were preincubated with 50 µM of
the caspase-8 inhibitor Ac-IETD-aldehyde (Ac-IETD-CHO, Calbiochem) or
the caspase-3 inhibitor Ac-DEVD-CHO (Promega), or with their vehicle
(i.e., a 2% DMSO solution) for 30 min at 37°C before the addition of
50 µM of Z-IETD-AFC or Ac-DEVD-AMC. AFC and AMC releases were
monitored using a Fluostar II spectrofluorimeter with filter settings
at 380 or 355 nm, respectively, for excitation and 510 or 460 nm,
respectively, for emission using the Biolise software (both from BMG
Labtechnologies, Champigny-sur-Marne, France). Results are expressed as
picomoles of free AFC or AMC, by comparison with standard curves of AFC
(Calbiochem) and AMC (Promega).
Statistical analysis
Data were analyzed statistically using the StatView SE+Graphics
program for Macintosh (Abacus Concepts, Berkeley, CA). If ANOVA was
significant, a Student’s t test for paired values was used
to assess comparability between the means (p values of 0.05
or less were considered significant). The results are expressed as
means ± SEM of the indicated number of experiments.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
inhibit Fas ligation-induced loss in viability and
apoptosis in human eosinophils
 View larger version (26K): [in a new window] |
Figure 1. Kinetics and dose–response study of inhibition by IL-5 and IFN- of
anti-Fas-induced apoptosis in human peripheral blood eosinophils. (A)
Eosinophils were incubated for 0–24 h with 3000 ng/mL of
isotype-matched mouse IgM or with 3–3000 ng/mL of anti-Fas mAb.
Apoptotic nuclei were quantified by flow cytometry after cell
permeabilization and DNA staining with propidium iodide. (B)
Eosinophils were treated for 12 and 24 h with 3000 ng/mL of mouse
IgM or of anti-Fas mAb in the absence or presence of 0.0005–5 ng/mL
IL-5 or of 0.1–1000 U/mL IFN-. The percent inhibition of
Fas-induced apoptosis by the indicated concentrations of cytokines was
determined. Results are the means ± SEM of 3–4
experiments. *P < 0.05, as compared with IgM-treated
eosinophils.
|
interfered with anti-Fas (3000 ng/mL)-induced
eosinophil apoptosis. Eosinophil incubation with 0.0005–5 ng/mL IL-5
or with 0.1–1000 U/mL IFN- resulted in dose-dependent inhibition of
Fas-mediated apoptosis at 12 and 24 h. The maximum effect was
observed with the highest concentrations of both cytokines (Fig. 1B)
.
In view of these results, the concentrations of 3000 ng/mL of anti-Fas
mAb, 5 ng/mL IL-5, and 1000 U/mL IFN- were selected for further
study.
Using flow cytometry, we also established that the increase in the
number of apoptotic nuclei in IgM (and, to higher extent, in
anti-Fas-treated eosinophils) paralleled a decrease in their viability
(Fig. 2A
and B
). The addition of IL-5 (5 ng/mL) or IFN- (1000 U/mL) to
the culture medium reduced the loss in cell viability and the increase
in the percent of apoptotic nuclei in both IgM- and anti-Fas-treated
eosinophils at 12 and, to a lesser extent, at 24 h (Fig. 2A
and 2B)
.
 View larger version (31K): [in a new window] |
Figure 2. Effect of IL-5 and IFN- on IgM- and anti-Fas-induced changes in
viability (A) and apoptosis (B) in cultured human blood eosinophils.
Eosinophils were stimulated for 0–24 h with 3000 ng/mL of
isotype-matched mouse IgM or agonistic anti-Fas mAb in the absence or
presence of 5 ng/mL IL-5 or of 1000 U/mL IFN-. Viable cells and
apoptotic nuclei were quantified by flow cytometry in intact cells or
after permeabilization and staining of DNA with propidium iodide,
respectively. Results are the means ± SEM of 8 (for
cell viability) and 26 (for cell apoptosis)
experiments.*P < 0.05, between IgM-treated cells and
values obtained at time 0; P < 0.05, as compared
with IgM-stimulated cells;  P < 0.05, as compared
with anti-Fas-treated cells.
|
restored a normal phenotype in most eosinophils (Fig. 3C
and 3D)
.
 View larger version (182K): [in a new window] |
Figure 3. Electron transmission microphotographs showing purified human
eosinophils incubated for 12 h with 3000 ng/mL isotype matched
mouse IgM (A) or with 3000 ng/mL anti-Fas mAb, in the absence (B) or
presence of 5 ng/mL IL-5 (C) or of 1000 U/mL IFN- (D). The
eosinophils indicated by the arrows show morphologic features of
apoptosis. Magnification x 1500.
|
was unrelated
to changes in Fas antigen expression at the cell surface at any time
point (data not shown).
The effect of IL-5 and IFN- was specific since a loss in their
inhibitory activity against spontaneous and Fas-induced-eosinophil
apoptosis was noted when they were heated at 100°C for 1 h (data
not shown).
Effect of IL-5 and IFN- on Fas ligation-induced
m alterations in human eosinophils
To determine whether Fas-induced apoptotic death involved
m alterations, we used flow cytometry to assess the
ability of eosinophils to incorporate the lipophilic cationic
fluorochrome DiOC6(3) into their mitochondria.
Freshly purified human eosinophils exhibited high fluorescence staining
with DiOC6(3) (Fig. 4A
and B
). A moderate and not significant reduction in
m was noted when eosinophils were incubated for 3 and
12 h with control IgM (Fig. 4A
and 4B)
. In contrast, the addition
of anti-Fas mAb induced a significant loss in
DiOC6(3) uptake, which occurred as early as 3 h
of stimulation and further decreased at 12 h (Fig. 4A
and 4B)
.
Specific staining of polarized mitochondria was confirmed by the
absence of accumulated DiOC6(3) in eosinophils
previously treated with the oxidative phosphorylation uncoupling agent
CCCP (Fig. 4A)
[28
].
 View larger version (38K): [in a new window] |
Figure 4. Effect of IL-5 and IFN- on Fas ligation-induced eosinophil
m dissipation. (A) Eosinophils were stimulated for 12
h with 3000 ng/mL of mouse IgM or of agonistic anti-Fas mAb, in the
absence or presence of 5 ng/mL IL-5 or 1000 U/mL IFN-. Cells were
incubated with 40 nM of DiOC6(3) and 2.5 µM of
propidium iodide. Negative control experiments were performed by
treating the cells with 200 µM of the mitochondrial uncoupler, CCCP.
Events were recorded on gated propidium iodide negative population.
These flow cytometry profiles are representative of 3–4 different
experiments. (B) Eosinophils were stimulated and incubated with
DiOC6(3) and propidium iodide, as described above
and DiOC6(3) high stained cells were quantified.
Results are the means ± SEM of 3–4 experiments.*
P < 0.05, as compared with IgM-treated cells;
P < 0.05, as compared with anti-Fas-treated cells.
|
to the culture medium significantly reversed
Fas-induced loss in m at 3 h (Fig. 4A
and 4B)
.
Whereas IL-5 still attenuated by 50% anti-Fas induced mitochondrial
depolarization at 12 h, IFN- had only a minor effect
(approximately 12%; Fig. 4A
and 4B ). Neither IL-5 nor IFN- modified
DiOC6(3) uptake in IgM-treated eosinophils at 3 h, although they slightly increased that observed at 12 h (Fig. 4A
and 4B)
.
Effect of IL-5 and IFN- on Fas-induced caspase-8 and caspase-3
activation
Next, we investigated the expression and activation of caspase-8
and caspase-3 following Fas engagement in human eosinophils by Western
blot analysis. Freshly purified eosinophils displayed detectable
amounts of the zymogen forms of caspase-8 and caspase-3 of 53–55
(p53–55) and 32 (p32) kDa, respectively, but not the
caspase-3-processed form of 17 kDa (p17) (Fig. 5A
and B
). Eosinophil incubation with control IgM failed to modify
the levels of p53–55 and p32 at 12 h but allowed us to detect the
p17 active subunit (Fig. 5A
and 5B)
. Cross-linking with the anti-Fas mAb
was followed by a marked reduction in the expression of both caspase-8
and caspase-3 inactive precursors and an increase in the amounts of p17
at 12 h (Fig. 5A
and 5B)
. The caspase-8-processed form of 18 kDa
was undetectable under our experimental conditions.
 View larger version (18K): [in a new window] |
Figure 5. Effect of IL-5 and IFN- on Fas ligation-induced caspase-8 and
caspase-3 proteolysis in human eosinophils. Eosinophils were cultured
for 0–12 h with 3000 ng/mL of isotype-matched mouse IgM or of anti-Fas
mAb, in the absence or presence of 5 ng/mL IL-5 or of 1000 U/mL
IFN-. Panels A and B show representative Western blot analysis for
caspase-8 and caspase-3 expression, respectively, from 3–4 different
cell preparations. Blots were probed with mouse anti-caspase-8, rabbit
anti-caspase-3, and anti-ß-actin Abs. The intensities of the
expression of the caspase-8 zymogen form of 53–55 kDa and the
caspase-3 zymogen (32 kDa) and processed (17 kDa) forms were determined
by densitometry and expressed as the ratio OD p53–55, or p32 or p17
corresponding band/OD ß-actin band values. Numbers under blots are
the means of OD ratios of 3–4 different experiments.
|
attenuated Fas-induced caspase-8 and caspase-3
proteolysis (Fig. 5A
and 5B)
, a finding that was confirmed by
densitometric analysis (Fig. 5A
and 5B)
. Caspase-8 and caspase-3
activation during anti-Fas-mediated eosinophil apoptosis was further
supported by the ability of protein extracts to cleave the specific
caspase-8 and caspase-3 substrates, Z-IETD-AFC and Z-DEVD-AMC,
respectively (Fig. 6
). These proteolytic activities augmented markedly 12 h after
anti-Fas stimulation and declined at 24 h (Fig. 6A
and 6B) . Lysates
from IgM-treated eosinophils also displayed moderate IETD-ase and
DEVD-ase activities, possibly as a reflection of spontaneous apoptosis
(Fig. 5A
and 5B)
. IL-5 and IFN- inhibited both spontaneous and
Fas-induced increases in caspase-8- and caspase-3-like activities in
eosinophil lysates at 12 h (Fig. 6) .
 View larger version (37K): [in a new window] |
Figure 6. Kinetics of Fas-induced caspase-8 and caspase-3 activities in
eosinophils (A and B) and their modulation by IL-5 and IFN- (C and
D). Eosinophils were stimulated for 0, 3, 12, or 24 h with 3000
ng/mL of mouse IgM, or of agonistic anti-Fas mAb in the absence or
presence of 5 ng/mL IL-5, or 1000 U/mL IFN-. Whole-cell lysates were
tested for their ability to cleave the caspase-8 and caspase-3
substrates, Ac-IETD-AFC (A and C) and Ac-DEVD-AMC (B and D),
respectively. Results are expressed as picomoles of free AFC and AMC
and represent the means ± SEM of 3–10 experiments. *
P < 0.05, between IgM-treated cells and values
obtained at time 0; P < 0.05, as compared with
IgM-treated cells; P < 0.05, as compared with
anti-Fas-stimulated cells.
|
Role of mitochondria and caspases during Fas ligation-induced human
eosinophil apoptosis
To determine the role of mitochondrial disruption in Fas
ligation-induced eosinophil apoptosis, cells were pretreated with
bongkrekic acid, an inhibitor of the MPTP component, adenine nucleotide
translocator (ANT) [30
]. At a concentration of 62 µM,
bongkrekic acid did not modify the loss in m in
either IgM- or anti-Fas-treated eosinophils at 3 and 12 h
(Fig. 7A
). Under these conditions, bongkrekic acid failed to reduce
eosinophil apoptotic death after 24 h of IgM or anti-Fas
stimulation (Fig. 7B)
, although a moderate but not significant
inhibitory effect was observed at 12 h (Fig. 7B)
. Bongkrekic acid
was also ineffective against Fas-induced loss in eosinophil viability
at 12 and 24 h (data not shown) and caspase-8 and caspase-3
activation at 12 h (Fig. 8A
and B
).
 View larger version (28K): [in a new window] |
Figure 7. Effect of bongkrekic acid on Fas ligation-induced loss in
m and apoptosis in human eosinophils. Eosinophils
were incubated for 2 h with 62 µM bongkrekic acid (BA) or with
its vehicle (0.02 N NH4OH) prior to stimulation with 3000
ng/mL of mouse IgM or of anti-Fas mAb for 0–24 h. In panel
A, m was determined by flow
cytometry, as described in the legend of Fig. 3
and the population of
DiOC6(3)-high stained cells was quantified. In panel
B, apoptotic nuclei were determined by flow cytometry after cell
permeabilization and propidium iodide DNA staining. Results are the
means ± SEM of 3 experiments. * P <
0.05, as compared with IgM-treated cells.
|
 View larger version (37K): [in a new window] |
Figure 8. Effect of bongkrekic acid on Fas ligation-induced caspase-8 and
caspase-3 activation. Eosinophils were incubated for 2 h with 62
µM bongkrekic acid (BA) or with its vehicle (0.02 N
NH4OH) prior to stimulation with 3000 ng/mL of mouse IgM,
or of anti-Fas mAb for 12 h. Whole-cell lysates were tested for
their ability to cleave the caspase-8 (A) or caspase-3 (B) substrates.
Results are expressed as picomoles of free AFC and AMC and represent
the means ± SEM of 3 experiments. * p < 0.05, as compared with IgM-stimulated BA-treated cells.
|
m alteration and apoptosis by preincubating the
eosinophils with the cell-permeant and irreversible inhibitors of
caspase-8 and caspase-3, Z-IETD-fmk, and Z-DEVD-fmk, respectively, or
with the broad spectrum caspase inhibitor, Z-VAD-fmk. At a
concentration of 30 µM, Z-VAD-fmk, Z-IETD-fmk, and, to a
lesser extent, Z-DEVD-fmk significantly inhibited anti-Fas mAb-induced
mitochondrial disruption at 3 h but not at 12 h (Fig. 9A
). This was accompanied by a marked reduction in Fas-mediated
apoptosis at 12 h, and, to a lesser extent, at 24 h (Fig. 9B)
.
 View larger version (34K): [in a new window] |
Figure 9. Effect of caspase inhibitors on Fas ligation-induced loss in
m and apoptosis in human eosinophils. Eosinophils
were incubated for 2 h with 30 µM of Z-VAD-fmk (VAD), Z-IETD-fmk
(IETD) or Z-DEVD-fmk (DEVD) or their vehicle, 0.4% DMSO, and then
stimulated for 0–24 h with 3000 ng/mL of mouse IgM or anti-Fas mAb.
Flow cytometric analysis was performed to evaluate (A) mitochondrial
disruption using DiOC6(3)/propidium iodide staining
and (B) apoptosis, after permeabilization of the cells and propidium
iodide DNA staining. Results are the means ± SEM of
3–11 experiments. * P < 0.05, as compared with
IgM-treated cells; P < 0.05, as compared with
anti-Fas-treated cells.
|
Similar results were obtained when the effect of caspase inhibitors was evaluated against Fas-induced loss in eosinophil viability at 12 and 24 h (data not shown).
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
To identify the intracellular signals involved in Fas-mediated apoptosis in human eosinophils, we investigated the effect of Fas ligation on mitochondrial transmembrane potential and caspase activation.
In this study, we show that activation of the Fas receptor with an agonistic antibody induced a rapid decrease in eosinophil mitochondrial potential, which occurred before the loss in cell viability and the induction of DNA fragmentation. These latter phenomena were associated with a cleavage of the zymogen forms of caspase-8 and caspase-3 and their functional activation, as ascertained by the ability of eosinophil lysates to digest specific substrates. The fact that caspase-like activities augmented at 12 h and decreased thereafter suggests that apoptotic death induced by prolonged Fas stimulation may result in a nonspecific degradation of cellular constituents.
To define the role of mitochondria in Fas-induced eosinophil apoptosis,
we investigated how MPTP contributed to this process. This channel is
formed by a complex containing ANT [22
,
30
], a protein involved in the induction of apoptotic
death [31
]. Inhibition of ANT by bongkrekic acid blocks
MPTP opening [22
, 30
]. We show here that
bongkrekic acid, at a concentration sufficient to inhibit
anti-Fas-induced mitochondrial disruption and apoptosis in human
neutrophils [24
], fails to stabilize mitochondrial
transmembrane potential and to modify the extent of apoptosis in
Fas-stimulated eosinophils. These observations suggest that the Fas
ligation-induced loss in m is unrelated to the
opening of the MPTP, similar to findings previously reported by
Marchetti et al. in murine thymocytes [32
]. Very
recently, bongkrekic acid has been shown to decrease
m in eosinophils [33
], a conflicting
finding that may result from the higher and probably toxic
concentration (100 µM) used in this study.
Our kinetics studies disclosed that caspase-8 and caspase-3 activation preceded and/or coincided with Fas-induced DNA fragmentation, which suggests that Fas activation promoted apoptosis most likely by activating these caspases. To prove this causality, we tested the ability of the potent irreversible and cell-permeant inhibitors of caspase-8 and caspase-3, Z-IETD-fmk and Z-DEVD-fmk, respectively, to interfere with Fas-induced effects. Although both compounds prevented eosinophil apoptosis induced by Fas ligation at early time points, the broad spectrum caspase inhibitor, Z-VAD-fmk, extended its protective effect over the whole time course of Fas stimulation. This result indicates that caspases other than caspase-8 and caspase-3 are presumably activated upon Fas cross-linking in human eosinophils.
We then attempted to determine the hierarchical implication and the functional connections between MPTP opening and caspase activation. Adding bongkrekic acid during anti-Fas stimulation failed to modulate caspase activation, which suggests that opening of the MPTP is not a prerequisite for generating the caspase cascade in eosinophils. In contrast, caspase-8, and, to a lesser extent, caspase-3 inhibitors preserved mitochondria from anti-Fas-induced loss in transmembrane potential at early stages of eosinophil stimulation, while they became ineffective at later time points. This observation suggests that activation of caspases, in particular caspase-8, may probably account for early, but not late, mitochondrial depolarization.
Next, we investigated the potential modulatory properties of IL-5 and
IFN- on Fas-induced apoptosis and the associated mitochondrial
disruption and caspase activation. Exposure of Fas-stimulated
eosinophils to IL-5 inhibited the loss in viability and the increase in
apoptosis at 12 h, and, to a lesser extent, at 24 h. These
findings are in agreement with those previously reported, which showed
that IL-5 failed to modify the degree of apoptosis of murine lung and
human peripheral blood eosinophils after 48–72 h of Fas stimulation
[14
, 18
, 34
]. These
observations, together with those presented in this study, suggest that
eosinophil exposure to Fas cross-linking for more than 24 h may
overcome the protective effect of IL-5.
Similar to IL-5, IFN- inhibited Fas-induced loss in eosinophil
viability and increase in apoptosis. This result extends previous
findings showing that IFN- enhanced human eosinophil survival
[6
, 7
, 35
] and decreased
spontaneous apoptotic death [7
, 35
], and
they further identify IFN- as an eosinophil-protecting cytokine.
The decreased sensitivity of IL-5- or IFN-- treated eosinophils in
response to anti-Fas triggering in the presence of IL-5 or IFN- was
unrelated to changes in Fas receptor expression. Our results are thus
consistent with the failure of IL-5 to modify the expression of this
antigen on eosinophil surface [14
, 16
], but
they differ from those previously reported showing a moderate
increasing effect of IFN- on Fas expression after 24 h of
stimulation [16
].
IL-5-mediated attenuation of Fas-induced apoptosis was associated with
the preservation of mitochondrial potential at 3 h, and, to a
lesser extent, at 12 h. Similar results were obtained in human
neutrophils, where GM-CSF was shown to inhibit Fas-induced apoptosis by
preventing the loss in m [24
]. Unlike
IL-5, the inhibition by IFN- of Fas-mediated eosinophil
apoptosis paralleled a limited effectiveness in stabilizing
mitochondria, particularly after 12 h of stimulation. Together,
these findings indicate that attenuation of eosinophil apoptotic death
by surviving stimuli, such as IL-5 and IFN-, is not necessarily
associated with a similar degree of protection against mitochondrial
depolarization.
Our kinetics studies disclosed the ability of IL-5 and IFN- to
inhibit Fas ligation-induced caspase-8 and caspase-3 activation at
12 h. Zangrilli et al. [34
] recently reported a
limited inhibition by IL-5 of Fas-induced caspase-8 and caspase-3
cleavage after 24 h of eosinophil stimulation. As for apoptosis,
this discrepancy may result from a reduced effectiveness of IL-5 after
long-term Fas stimulation. Inhibition of Fas-induced caspase-8 and
caspase-3 activation in neutrophils has previously been associated with
the apoptosis-rescuing effect of GM-CSF [24
], which
supports a role for caspases as potential targets for stimuli that
influence granulocyte apoptosis.
In conclusion, our results demonstrate the requirement of caspase
activation, but not of MPTP opening, in the propagation of apoptotic
signals downstream of Fas activation in human peripheral blood
eosinophils. Nevertheless, mechanisms involving mitochondrial
permeabilization, but independent from the MPTP opening, may operate in
amplifying caspase activation and in inducing DNA fragmentation. These
may involve Bcl-2 family proteins, which have been shown to tightly
regulate mitochondrial homeostasis [22
,
30
]. In particular, cleavage of the Bcl-2 family member,
Bid, and subsequent mitochondrial depolarization and/or release of
apoptogenic factors induced by activated caspase-8, establishes a link
between caspases and mitochondrial alterations during Fas-induced
apoptosis [21
, 30
]. Whether Bid is
expressed by human eosinophils and is a target for stimuli that
modulate mitochondrial permeabilization, such as IL-5 and IFN-, is
an interesting area for future research.
By identifying caspases as the predominant intracellular effector molecules involved in Fas-mediated apoptosis, we have helped elucidate the molecular mechanisms governing the cell death program in human eosinophils. Our work may have important implications for therapeutic considerations, particularly in the search for new ways to treat more selectively chronic inflammatory diseases in which the presence of excessive numbers of eosinophils in blood and tissues plays a critical role.
Received April 5, 2001; revised July 3, 2001; accepted July 9, 2001.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Am. J. Respir. Cell Mol. Biol. 18,315-322This article has been cited by other articles:
 |
|
 |
|
  M. Maret, C. Ruffie, S. Letuve, A. Phelep, O. Thibaudeau, J. Marchal, M. Pretolani, and A. Druilhe A Role for Bid in Eosinophil Apoptosis and in Allergic Airway Reaction J. Immunol., May 1, 2009; 182(9): 5740 - 5747. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Didichenko, N. Spiegl, T. Brunner, and C. A. Dahinden IL-3 induces a Pim1-dependent antiapoptotic pathway in primary human basophils Blood, November 15, 2008; 112(10): 3949 - 3958. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Dolgachev, M. Thomas, A. Berlin, and N. W. Lukacs Stem cell factor-mediated activation pathways promote murine eosinophil CCL6 production and survival J. Leukoc. Biol., April 1, 2007; 81(4): 1111 - 1119. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Rosas, P. F. Dijkers, C. L. Lindemans, J.-W. J. Lammers, L. Koenderman, and P. J. Coffer IL-5-mediated eosinophil survival requires inhibition of GSK-3 and correlates with {beta}-catenin relocalization J. Leukoc. Biol., July 1, 2006; 80(1): 186 - 195. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. I. Berro, G. A. Perry, and D. K. Agrawal Increased Expression and Activation of CD30 Induce Apoptosis in Human Blood Eosinophils J. Immunol., August 1, 2004; 173(3): 2174 - 2183. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. W. Al-Rabia, M. G. Blaylock, D. W. Sexton, and G. M. Walsh Membrane receptor-mediated apoptosis and caspase activation in the differentiated EoL-1 eosinophilic cell line J. Leukoc. Biol., June 1, 2004; 75(6): 1045 - 1055. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. G. Zangrilli Regulation of Eosinophil Viability by Cytokines Am. J. Respir. Cell Mol. Biol., April 1, 2002; 26(4): 388 - 390. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
