|
|
||||||||
-mediated apoptosis in fibroblasts by TRAF2 recruitment to the IL-15R
chain


* Institute of Immunology,
¶ Department of Urology,
Department of Pathology, University Hospital Benjamin Franklin, Free University, Berlin, Germany;
§ Department of Dermatology, University Hospital Eppendorf, University of Hamburg, Germany; and
Department of Internal Medicine IV, Friedrich-Alexander-University, Erlangen, Germany
1Correspondence: Institute of Immunology, University Hospital Benjamin Franklin. Free University Berlin, Hindenburgdamm 30, D-12200 Berlin, Germany. E-mail: bulfone{at}zedat.fu-berlin.de
| ABSTRACT |
|---|
|
|
|---|
chain signaling. The
intracellular tail of IL-15R
shows sequence homologies to the TRAF2
binding motifs of CD30 and CD40. Most important, binding of IL-15 to
IL-15R
successfully competes with the TNFR1 complex for TRAF2
binding, which may impede assembly of key adaptor proteins to the TNFR1
complex, and induces I
B
phosphorylation. Thus, IL-15R
chain
stimulation is a powerful deflector of cell death very early in the
apoptosis signaling cascade, while TNF-
and IL-15 surface as major
opponents in apoptosis control.Bulfone-Paus, S., Bulanova, E., Pohl,
T., Budagian, V., Dürkop, H., Rückert, R., Kunzendorf, U.,
Paus, R., Krause, H. Death deflected: IL-15 inhibits TNF-
-mediated
apoptosis in fibroblasts by TRAF2 recruitment to the IL-15R
chain.
Key Words: cell death TNFR1 L929 I
B
| INTRODUCTION |
|---|
|
|
|---|
chains as
components of the trimolecular IL-15 receptor (IL-15R) complex, despite
the absence of sequence homologies with IL-2. In addition, IL-15 binds
to a distinct receptor chain (IL-15R
) (5
(IFN-
) production by NK cells (1
Recently we described an additional functional role of IL-15 that
is of general biological importance: its properties as a potent
inhibitor of apoptosis in vitro on activated human T and B
cells, as well as in vivo, where it protects mice from
Fas-induced lethal hepatic failure and multi-system apoptosis and from
chemotherapy-induced epithelial cell apoptosis (12
, 13)
.
Ordinarily apoptosis appears to be triggered by signaling via members
of the tumor necrosis factor (TNF) receptor family (Fas, TNFR1, DR3)
(14
15
16
17
18
19)
. Besides Fas, the other major signaling pathway
for inducing apoptosis in multiple cell types operates via stimulation
of the TNF-
receptor type 1 (TNFR1, p55) (18
, 20)
.
TNF-
elicits a wide spectrum of biological effects as the result of
complex signaling events that are initiated through trimerization of
two distinct transmembrane receptors: TNFR1 (p55) and TNFR2 (p75).
Apoptosis is mainly induced through TNFR1 via the intracellular `death
domain' (DD), a ~80 amino acid domain found in the cytoplasmic
region of TNFR1, Fas, and DR3. Aggregation of the receptors by the
trimeric ligand orients the DD in a conformation that recruits adaptor
proteins. The adaptor proteins also contain a DD and associate with the
receptor through a homotypic DD interaction (14
, 21
22
23)
.
Recently, downstream signal-transducing proteins and
receptor-associated proteins that couple the TNFR1 receptor to the
signaling cascade for the generation of cellular responses have been
identified: TRADD (TNFR1-associated death domain protein) (18
, 19
, 24
, 25)
, FADD (Fas-associated protein with death domain)
(26
, 27)
, RIP (receptor interacting protein) (28
, 29)
, and TRAF2 (TNFR-associated factor) (30
, 31)
.
FADD is part of the inducible Apo1/Fas death-inducing signaling complex
and is believed to represent the physical link to apoptosis-executing
proteases of the caspase family (26
, 27)
.
There are several lines of evidence that two TNFR1-signaling cascades
bifurcate at TRADD: one induces NF-
B activity, which may promote
cell survival (32
33
34)
; the other induces apoptosis via
FADD and the caspase machinery. RIP is recruited to the TNFR1 complex
with bifunctional activities, since it may promote either cell death or
NF-
B activation (21
22
23)
. TRAF2 interacts with TRADD
and RIP through homotypic TRAF domain interactions and is involved in
the TNF-dependent activation of NF-
B (25)
. TRAF2
appears to be involved in TNF-mediated NF-
B activation and in
signaling through the stress-activated protein kinase or c-Jun
amino-terminal kinase, both of which may be instrumental to rescuing
cells from programmed cell death (25
, 35)
.
Since TNF-
is recognized as a major stimulus for inducing apoptosis
in many cell types (15
, 16
, 30
, 36
, 37)
, we wanted to
study whether IL-15 and TNF-
are opposing forces in the cytokine
network with respect to apoptosis control (38
, 39)
. Also,
the by now comparatively well-defined biochemistry of TNF-
-induced
apoptosis, namely, the role of adaptor proteins in TNFR1-mediated cell
death (18
, 19
, 40)
, promised to help elucidate potential
molecular pathways of apoptosis inhibition by IL-15.
Therefore, the ability of IL-15 to block TNFR1-mediated proapoptotic
signaling in a murine fibroblast cell line (L929), which is highly
sensitive to TNF-
-induced apoptosis (41
, 42)
, was
examined. The data reported here suggest that IL-15 deflects
TNF-
-mediated apoptosis in these fibroblasts in vitro by
inhibiting the TNFR1 adaptor protein assembly and by recruiting the
TNFR1-associated protein TRAF2 to the IL-15R
chain.
| MATERIALS AND METHODS |
|---|
|
|
|---|
and IL-15 were
purchased from Genzyme (Cambridge Mass.). L929 cells were stimulated
with 10 ng/ml TNF-
, 10 ng/ml IL-15, 10 ng/ml IL-2, or a combination
of the cytokines.
Flow cytometric analysis
For fluorescein-activated cell sorter (FACS) analysis, 1 x
106 L929 cells were incubated for 30 min on ice
with the biotinylated fusion proteins IL-2-IgG2b (43)
and
IL-15-IgG2b (12)
, with biotinylated mouse IgG2b (49.2), or
with the following FITC- or PE-conjugated antibodies: anti-IL-2Rß
chain (TM-ß1), anti-IL-2R
chain (TUGm2), or control IgM (R334)
(all antibodies were purchased from Pharmingen, Hamburg, Germany). When
biotinylated fusion proteins were used, cells were further incubated
with streptavidin-PE (Pharmingen). After incubation, the cells were
washed and analyzed by a FACS-Sort (Becton Dickinson, Heidelberg,
Germany) (12)
. Flow cytometric analysis of propidium
iodide (PI) -stained L929 cells was performed as described
(44)
.
Reverse transcriptase-polymerase chain reaction (RT-PCR)
Cellular RNA was extracted by using the RNA Clean reagent (AGS
GmbH) according to the manufacturer's instructions. A 5 µg aliquot
of total cellular RNA was reverse transcribed using random
hexanucleotides as primers and SUPERSCRIPT II preamplification kit
(Gibco BRL). cDNA was amplified in a 20 µl PCR reaction mixture
containing 250 µM dNTPs, 200 nM primers, 2 µl 10-fold PCR buffer,
and 1 U Taq DNA polymerase (`Amplitaq', Perkin
Elmer/Cetus, Emeryville, Calif.). The primers used were: mIL-15R
sense 5' AACATCCACCCTGATTGAGTGT 3', antisense 5'
GTTTCCATGGTTTCCACCTCAA 3'; mIL-2R
sense 5'
GGATCCAAGATGGAGCCACGCTTGCTGACG 3', antisense 5'
AAGCTTTCAATACTCCATAGTGAGCACAAATGTCACC 3'; mIL-2Rß sense 5'
GTCGACGCTCCTCTCAGCTGTGATGGCTACCATA 3', antisense 5'
GGATCCCAGAAGACGTCTACGGGCCTCAAATTCCAA 3'; mIL-2R
sense 5'
GTCGACAGAGCAAGCACCATGTTGAAACTA 3', antisense 5'
GGATCCTGGGATCACAAGATTCTGTAGGTT 5'; ß-actin sense 5' GTGGGG
CGCCCCAGGCACCA 3', antisense 5' CTCCTTAATGTCACGCACGATTTC 3'. All
primers used were generated by and purchased from TIB Molbiol (Berlin).
Samples were amplified in a DNA Thermocycler (Perkin Elmer/Cetus) for
35 cycles. Each cycle consisted of denaturation at 94°C for 1 min,
annealing at 60°C for 2 min, and extension at 72°C for 2 min.
Aliquots of PCR products were then electrophoresed on 1.5% agarose gel
and visualized by ethidium bromide staining. ß-actin message
expression was used to normalize the cDNA amount to be used; a mock PCR
(without cDNA) was included to exclude contamination in all
experiments.
Apoptosis assay
For analysis of DNA laddering, 1 x
106 cells were used as described previously
(12)
. Briefly, cells were pelleted by centrifugation
200 x g for 10 min at 4°C. Cell pellets were
resuspended in 20 µl of lysis buffer (10 mM EDTA, 50 mM Tris-HCl, pH
8.0, 0.5% sodium lauryl sarcosinate, and 0.5 mg/ml proteinase K) and
incubated for 1 h at 50°C. 10 µl of 0.5 mg/ml RNaseA was added
to each sample, followed by incubation at 50°C for an additional
hour. A 5 µl loading buffer (50% glycerin, 1 mM EDTA, 0.04%
bromophenol blue) was mixed with 10 µl of each sample before loading
onto the dry wells of a 2% agarose gel containing 0.1 µg/ml ethidium
bromide.
Expression vectors and transfections
To construct the pcDNA3-mIL-15R
and TRAF2-expressing vectors,
RNA was prepared from mouse splenocytes after 48 h concanavalin A
stimulation (10 µg/ml), reverse transcribed with random priming, and
mIL-15R
cDNA was amplified with mIL-15R
-specific oligos with
restriction enzyme sites for cloning into pcDNA3(±) (Invitrogen, Leek,
The Netherlands): IL-15R
sense (BamHI)
5'-GGGGATCCTTGGCCATGGCC TCGCCG-3'; IL-15R
antisense
(EcoRI) 5'-CTGAATTCGTGTGGTTAGGCTCC TGT-3'; TRAF2 sense
(BamHI) 5'-GTGGGGGATCCAACTCACATGGCTGCA-3'; TRAF2
antisense (XbaI) 5'-CTTATCTAGAGTGGCTAGAGCTCTG-3'. L929
cells were transiently transfected withTRAF2 or IL-15R
in sense (+)
or antisense orientation (-) cDNA by the DEAE method as described
(42)
.
Immunoprecipitation and Western blotting
L929 cells were lysed in 100 µl Brij 96 extraction buffer (1%
Brij, 20 mM Tris-Cl, pH 7.4, 75 mM NaCl, 1 mM EDTA, 1 mM sodium
vanadate, 20 µg/ml aprotinin, 10 µg/ml pepstatin) and incubated for
15 min on ice. 100 micrograms of total proteins were mixed with 20 µl
electrophoresis sample buffer (62.5 mM Tris-HCL, pH 8.0, 1% glycerol,
2% sodium dodecyl sulfate (SDS), 5% ß-mercaptoethanol, 0.01%
bromophenol blue), separated by 1015% SDS-polyacrylamide gel
electrophoresis (SDS-PAGE), and transferred to a nitrocellulose
membrane in buffer containing 25 mM Tris, 192 mM glycine, 1% SDS, and
20% methanol at 150 V for 40 min. Blots were blocked for 1 h in
phosphate-buffered saline (PBS) with 0.05% Tween-20 (PBS-T) and 3%
BSA (Sigma), then probed for 1 h with the following antibodies:
anti-TRADD, anti-FADD, anti-RIP, anti-TRAF2, anti-IL-15R
, and
anti-I
B
(Santa Cruz Biotechnology, Calif.) diluted 1:200 in
PBS-T. After washing with PBS-T, blots were incubated for 1 h at
room temperature with the secondary antibodies: anti-goat or
anti-rabbit Ig horseradish peroxidase (Amersham International, Slough,
U.K.) diluted at 1:1000. Visualization of immune complexes was carried
out by an enhanced chemiluminescence (ECL) method using ECL Western
blotting detection regents (Amersham International) according to the
manufacturer's instructions. Stripping of blots was performed in 62.5
mM Tris-HCl buffer containing 2% SDS and 100 mM ß-mercaptoethanol at
4°C overnight. For immunoprecipitation, 500 µg of proteins was
incubated overnight at 4°C with 5 µg hamster anti-TNFR1 (Genzyme),
10 µg goat anti-IL-15R
, or 10 µg rabbit anti-TRAF2 antibodies in
500 µl lysis buffer. Hamster, goat, or rabbit IgG were used as
isotype-matched controls. Forty microliters of a 1:1 slurry of
streptavidin-agarose (Pierce Chemical Company, Rockford, Ill.) were
added to each sample and incubated for another hour at room
temperature. The agarose beads were washed twice with 1 ml lysis
buffer. Bound proteins were eluted by boiling for 5 min in
electrophoresis sample buffer, resolved by SDS-PAGE, and analyzed by
Western blot as described above.
| RESULTS |
|---|
|
|
|---|
chain
and only
marginally express the IL-2Rß component of the IL-2R complex
(Fig. 1A
chain on the cell surface of L929
fibroblasts (Fig. 1A
and the
absence of IL-2R
expression on L929 cells were confirmed by RT-PCR
analysis. In addition, RT-PCR could detect only a faint IL-2Rß
transcript band (Fig. 1B
|
IL-15 rescues L929 cells from TNF-
-induced apoptosis
To examine the ability of IL-15 to block the proapoptotic
signaling induced by human TNF-
, which binds exclusively to TNFR1 in
the murine system (45)
, and to define the temporal
relationship of any such interaction, L929 cells were treated for
12 h with 10 ng/ml TNF-
, 10 ng/ml IL-15, TNF-
plus IL-15, or
TNF-
, followed by 10 ng/ml IL-15 at different time points (30 s, 1,
2, 5, 10, 15 min). Apoptosis was tested by DNA laddering after 12 h and by PI staining after 24 h.
Figure 2
demonstrates that coincubation with IL-15 rendered L929 cells resistant
to apoptosis induction by TNF-
. However, suppression of
TNF-
-induced DNA fragmentation by IL-15 (Fig. 2A
)
occurred only when administered simultaneously, 30 s, or 1 min
after adding TNF-
to the culture. This was confirmed by PI staining
(FACS), which showed a fluorescence curve in the group treated
simultaneously with TNF-
+IL-15 that virtually corresponded to the
negative control; instead, L929 cells-treated with TNF-
alone or
with coadministration of IL-15 showed evidence of massive apoptosis
only after 15 min (Fig. 2B
). This suggests that IL-15 blocks
TNF-
-induced apoptosis in L929 cells at a very early stage of the
signaling cascade, most likely via the IL-15R
chain. These data also
indicate that isolated IL-15R
proteins, contrary to conventional
wisdom (5)
, can indeed serve as functional receptors in
the virtual absence of IL-15Rß and -
chains, at least in this
fibroblast cell line. Finally, these data provide the first evidence
that IL-15 exerts any functional effects at all on fibroblast apoptosis
in vitro.
|
IL-15 stimulation induces the release of adaptor proteins from the
TNFR1 complex
Current evidence suggests that TNF-
induces the
trimerization of TNFR1, which is thought to result in the recruitment
of TRADD, the adaptor protein that signals for apoptosis, via
interactions between homologous regions, the so-called death domains
(18
19
20
, 24
, 25)
. Subsequently, the TNFR1TRADD complex
recruits at least three additional adaptor proteins: FADD (26
, 27)
RIP (28
, 29)
, and TRAF2 (30
, 31)
.
FADD and RIP transmit the key apoptosis signal(s) by interacting with
caspases. In addition, together with TRAF2, RIP mediates NF-
B
activation, which may induce the expression of survival genes
(18
, 19
, 24
, 46)
. Overexpression of FADD and RIP induces
apoptosis (26
27
28
29)
, while RIP -/- cells fail to activate
NF-
B (46)
.
We next addressed how IL-15 might inhibit TNFR1-mediated
apoptosis, and first explored whether there is any evidence that IL-15
signaling through the IL-15R
chain directly interferes with TRADD-,
FADD-, RIP-, or TRAF2-TNFR1 interactions. For this purpose, lysates
from L929 cells that had been treated with TNF-
, IL-15, TNF-
plus
IL-15 or were left untreated were immunoprecipitated with a monoclonal
antibody directed against the extracellular domain of the TNFR1.
Coprecipitation of TRADD, FADD, RIP, or TRAF2 with TNFR1 was
investigated by immunoblot analysis, using the corresponding antisera.
TNF-
stimulation of L929 cells for 15 min induced the
coprecipitation of TRADD, FADD, RIP (Fig. 3
, left upper panel), and TRAF2 (Fig. 3
, left lower panel) with the
TNFR1. To the best of our knowledge, successful coprecipitation of
TNFR1 with its adaptor proteins had never been reported in any cells
other than transfectants with experimentally induced adaptor protein
overexpression (24
, 25
, 29)
; therefore, these
coprecipitation assays were subjected to rigorous validation (different
lysis buffer conditions, antibody source, etc.).
|
In contrast, costimulation of L929 cells with TNF-
plus IL-15 for 15
min blocked the association of the TNFR1 with its adaptor proteins
(Fig. 3)
. Thus, the simultaneous treatment of L929 cells with IL-15 and
TNF-
appears to inhibit apoptosis by blocking the proteinprotein
interactions of the TNFR1 with these adaptor proteins by inhibiting the
recruitment of cytosolic adaptor proteins to the stimulated TNFR1
and/or by dissociating already bound adaptor proteins from the TNFR1
complex.
The IL-15 receptor coimmunoprecipitates TRAF2
To corroborate this concept, to further dissect the effects of
TNF-
and IL-15 costimulation on IL-15R-mediated signaling, and to
identify proteins that might be interacting with the IL-15 receptor,
L929 cells were treated with TNF-
, TNF-
plus IL-15, or IL-15
(controls: TNF-
+IL-2, IL-2). Cell extracts were then
immunoprecipitated with anti-IL-15 receptor antibody. After SDS-PAGE
separation and transfer to nitrocellulose, the blots were probed with a
panel of adaptor protein-specific antisera against TRADD, FADD, RIP,
and TRAF-2, followed by anti-rabbit and anti-goat Ig conjugated with
horseradish peroxidase.
Figure 4
A (upper panel) shows that TRAF2 can associate with the IL-15
receptor on IL-15R
stimulation with IL-15 alone, and such an
association is even stronger after costimulation with TNF-
and
IL-15. No association of TRADD, FADD, or RIP with the IL-15 receptor
was detectable (not shown).
|
Therefore, when coadministered with TNF-
, IL-15 may interfere with
the proteinprotein association of TRAF2 with the TNFR1 complex, and
instead up-regulate TRAF2 association with the IL-15R
chain. Thus,
ligand-stimulated TNFR1 complex and IL-15R
would seem to compete for
binding to TRAF2, with IL-15R
likely showing a higher affinity for
TRAF2 than TNFR1.
To further confirm an association of TRAF2 with IL-15R
, L929 cells
were transiently transfected with plasmids that express IL-15R
or
TRAF2 cDNAs in sense or antisense orientation. Transfected and
untransfected cells were lysed and then immunoprecipitated either with
IL-15R
antibody or with anti-TRAF2 antiserum. After SDS gel
electrophoresis, immunoblotting was performed with anti-TRAF2 (Fig. 4B
, upper panel) and anti-IL-15R
(Fig. 4B
,
middle panel) antiserum.
As shown in Fig. 4B
(upper panel), lane 4, overexpression of
IL-15R
revealed the appearance of a TRAF2 immunoprecipitate, which
was not visible in untransfected controls with only constitutive
IL-15R
expression levels (lane 1) or in antisense transfectants
(lane 5). This suggests that in the absence of IL-15 stimulation, TRAF2
association with the IL-15R
chain is present only at a very low
level, but becomes visible with an experimentally induced, substantial
numeric increase in (low affinity?) TRAF2-IL-15R
associations. This
interpretation was further supported by the reverse experiment, where
TRAF2 overexpression in L929 cells also made visible IL-15R
complexing to TRAF2 (Fig. 4B
; middle panel, lane 2).
To obtain additional evidence that the TNFR1 complex and IL-15R
really compete for binding to TRAF2 upon activation by their respective
ligands, TRAF2-transfected L929 cells were incubated with medium,
TNF-
, IL-15, or TNF-
+IL-15 (controls: IL-2, TNF-
+IL-2). Cell
extracts were immunoprecipitated with anti-TRAF2 antibodies and
immunoblotting was performed with anti-IL-15R
antibodies. As shown
in Fig. 4C
(upper panel), stimulation of these TRAF2
overexpressing fibroblasts with TNF-
, IL-15, IL-2, and TNF-
+IL-2
showed detectable IL-15R
-TRAF2 complexes precipitated with
anti-IL-15R
antibody. Stimulation with TNF-
+IL-15 further
up-regulated the complexing of overexpressed TRAF2 with the
constitutive level of IL-15R
(Fig. 4C
, upper panel, lane
5).
Together, these data provide evidence for the concept that unstimulated
IL-15R
does not bind TRAF2 (or only with low affinity), just as
unstimulated monomeric TNFR1 does not bind TRAF2 (19
, 20)
.
Furthermore, the above findings suggest that the binding affinity of
both receptors for TRAF2 increases substantially after stimulation by
their respective ligands. Both receptors may then compete for TRAF2,
with ligand-stimulated IL-15R
showing the highest affinity for TRAF2
binding.
The IL-15R
intracellular tail contains a conserved amino acid
motif for TRAF binding
When the TRAF2 binding motif described for the intracellular
tail of CD30 and CD40 (47)
was next compared with the
IL-15R
chain intracellular segment, an interesting similarity in the
distribution of characteristic amino acid residues surfaced (Fig. 5
). The high absolute degree of identity of amino acid sequence between
the TRAF2 binding motif for IL-15R with the motif described for the
CD30 intracellular chain is further complemented by conservative
substitutions of amino acids with respect to their properties within
the potential binding motif (proline and glutamine for valine): short,
uncharged residues tend to alternate with acidic amino acids. The
strikingly close relatedness of CD30 and IL-15R
moieties suggests
the existence of a common binding domain for TRAF2. Thus, these amino
acid sequence homologies further support the biochemical evidence
described above, since they reveal a possible structural correlate for
TRAF2 association with the IL-15R
intracellular chain.
|
TNF-
and IL-15 costimulation induces NF-
B
activation
Those TNFR1-mediated effects not directly related to apoptosis
(e.g., proliferation, cytokine release, differentiation) are thought to
operate chiefly via modulating the activity of the transcription factor
NF-
B (24
, 48)
. The most common form of NF-
B is a
heterodimer of a 50 kDa (p50) and a 65 kDa protein (RelA or p65)
(49
50
51)
. NF-
B complexes are sequestered in the cytosol
bound to one or more inhibitor proteins, whose prototype is I
B
.
Upon stimulation, I
B
dissociates from NF-
B as a result of
phosphorylation and proteolytic degradation, thus permitting NF-
B to
translocate to the nucleus (48)
. In addition, NF-
B
activation may confer protection against TNFR1-induced apoptosis by
induction of cell death protective genes (25
, 32
, 34
, 35
, 50
, 51)
.
Therefore, it was interesting to explore whether the potent
antiapoptotic properties of IL-15 in L929 cells (Fig. 2)
, which had
also been noted in unrelated epithelial and hematopoietic systems
(12
, 13
, 52)
, might also be obtained (at least in part)
through NF-
B activation. Western blot analysis was performed on L929
cell extracts that had been stimulated for various time periods, using
I
B
-specific antisera. After TNF-
stimulation of L929 cells for
5 or 15 min, I
B
was found to be present in the unphosphorylated
form, which binds NF-
B and thus keeps the latter in an inactive form
(49
, 51)
.
As shown in Fig. 6
, stimulation of L929 cells by IL-15 alone neither induced nor inhibited
I
B
phosphorylation. However, simultaneous stimulation of L929
cells with TNF-
and IL-15 induced I
B
phosphorylation, starting
5 min after stimulation; this was completed within 15 min after
stimulation (Fig. 6)
. In light of the extremely rapid inhibition of
TNF-induced apoptosis by IL-15 costimulation (Fig. 2)
, this relatively
slow response of I
B
(and, implicitly, NF-
B activity) suggests
that TRAF2 recruitment to the IL-15 receptor is the faster, earlier,
and likely most critical event in the IL-15-mediated inhibition of
TNF-induced L929 apoptosis. However, IL-15 receptor-mediated
NF-
B activation by I
B
phosphorylation may at least contribute
to rescuing the cells from TNF-induced apoptosis.
|
| DISCUSSION |
|---|
|
|
|---|
are indeed functional antagonists with respect
to the control of apoptosis in a murine fibrosarcoma cell line.
Moreover, this study provides the first available evidence that IL-15
can modulate fibroblast apoptosis and demonstrates that the IL-15R
chain can serve as a fully functional signal-transducing receptor. The
latter finding is important, since it is a widely held dogma in IL-15
research that only the complete IL-15 receptor complex (
, ß,
chain) has signal-transducing functions (4
chain
is the high-affinity receptor for IL-15, any role that the IL-2Rß
chain might possibly play in the apoptosis-inhibitory effects of IL-15
reported here would, in all likelihood, be of minor importance.
Most important, the current study suggests a plausible molecular
scenario by which IL-15 can very rapidly and highly efficiently protect
cells from apoptosis. Though it remains to be investigated whether this
model is also valid for other cell types, at least in L929 fibroblasts
IL-15 appears to deflect TNF-
-induced apoptosis by blocking adaptor
protein recruitment to the TNFR1 (Fig. 7
). Ligand-stimulated TNFR1 complex and IL-15R
seem to compete for
binding to TRAF2, with IL-15R
probably showing a higher affinity for
TRAF2 than TNFR1. It is tempting to speculate that in analogy to what
has already been described for CD30-TRAF2 interactions
(53)
, the ligand-activated intracellular IL-15R
chain
rapidly `depletes' TRAF2 so that it becomes unavailable for assembly
of the apoptosis-signaling TNFR1 complex. However, once all three
protein interaction partners become available for study as purified
soluble proteins, systematic kinetic and comparative affinity analyses
of TNFR1 vs. the IL-15R
for binding to TRAF2 will be required before
this concept can be fully accepted.
|
Our data also implicate NF-
B in the actual apoptosis inhibition by
costimulation of L929 cells with TNF and IL-15. The time course of
TNF/IL-15 costimulation and apoptosis inhibition (Fig. 2)
suggests that
TRAF2 recruitment to the IL-15 receptor is the fastest, earliest, and
likely most critical event in the IL-15-mediated inhibition of
TNF-induced L929 cell apoptosis. However, later NF-
B activation by
IL-15R
-dependent I
B
phosphorylation may at least contribute to
rescuing the cells from apoptosis. In view of recent evidence that
TRAF2 may mediate TNF-induced NF-
B activation (25
, 46)
,
it is conceivable that the strong association of TRAF2 with the
ligand-activated IL-15R
chain is critically involved in the I
B
phosphorylation that can be observed here only upon costimulation with
IL-15 and TNF.
The present experiments do not clarify whether IL-15R stimulation by
its ligand blocks the association of cytosolic adaptor proteins with
TNFR1 or whether IL-15R stimulation dissociates already bound adaptors
proteins from the TNFR1. However, the extraordinarily fast apoptosis
inhibition by costimulation with TNF-
and IL-15 (Fig. 2)
suggests an
effect very early in the signaling cascade: that adaptor protein
recruitment to the TNFR1 was blocked. The observation that
TNFR1-mediated apoptosis was inhibited only if IL-15 was administered
at the latest 1 min after TNF stimulation (Fig. 2A
) makes it
fairly unlikely that such a short time span could have sufficed for
both adaptor protein recruitment to and subsequent
dissociation from the TNFR1. Instead, ligand binding-induced
conformational changes in protein affinity for TRAF2 might well occur
that fast. Preliminary results from our laboratory indicate that the
assembly of adaptor proteins (TRADD and FADD were tested) to the TNFR1
has already occurred by 1 min after TNF-
administration and is
stable for at least 15 min (S. Bulfone-Paus et al., unpublished data).
Taken together, our data provide evidence for the following scenario of
TNF/IL-15 interactions in the control of TNFR1-mediated fibroblast
apoptosis (Fig. 7)
. In the absence of TNF-
or IL-15, TRAF2 does not
bind to TNFR1 and does not or only minimally binds to IL-15R
, due to
a very low affinity of the unstimulated receptor protein for TRAF2.
After stimulation by the respective ligands, which causes a change in
receptor conformation, the TRAF2 affinity of IL-15R
and of the
trimerized TNR1 complex increases substantially. The efficiency and
time course of IL-15-mediated blocking of TNF-induced apoptosis in L929
cells (Fig. 2)
suggest that in this putative, high-affinity competition
of ligand-stimulated TNFR1 and IL-15R
for TRAF2 binding, IL-15R
subsequently wins if IL-15R
is stimulated by IL-15 in time to
increase its normally low affinity for TRAF2 before most TRAF2 is bound
to the TNFR1 complex. Thus, the stimulated IL-15R
attracts most
TRAF2 so that it is no longer available for initiating the proapoptotic
cascade mediated by proper adaptor protein assembly to the TNFR1.
I
B
phosphorylation upon costimulation with TNF-
and IL-15
(Fig. 6)
, which should result in NF-
B activation (25
, 32
, 34
, 35
, 50
, 51)
, may be exploited by IL-15-stimulated IL-15R
as a
second pathway for inhibiting TNFR1-mediated apoptosis in this
fibroblast line. TRAF2 may be an important regulatory element of this
pathway as well.
This scenario does not yet explain why the association of TRAF2 to the
IL-15R
chain is made stronger by costimulation with TNF-
and
IL-15. By coimmunoprecipitation experiments, we are currently testing
the hypothesis that both cognate receptors need to be physically
associated to compete for TRAF2 binding and that only ligand binding to
both receptors brings TNFR1 and IL-15R
into sufficiently close
proximity for the proposed competition for TRAF2 binding to be able to
unfold. First, as yet preliminary data from our laboratory indicate
that extracts from IL-15R
overexpressing L929 cells
immunoprecipitated with anti-TNFR1 can indeed coprecipitate the
IL-15R
(S. Bulfone-Paus and E. Bulanova, unpublished data).
The observation that IL-15 blocks TNFR1-mediated cell death very early
in, and at a critical crossroads of, the apoptosis signaling cascade
designates IL-15 as an unusually powerful apoptosis antagonist with
intriguing therapeutic potential in a large number of clinical
situations where apoptosis inhibition appears desirable (54
, 55)
. Our finding that TRAF2 recruitment to the ligand-stimulated
IL-15R
chain is crucial in this respect should also facilitate the
development of novel antiapoptotic drugs unrelated to IL-15 signaling
that directly target TRAF2 and its effects on NF-
B
activation.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Received for publication August 11, 1998. Revised for publication March 15, 1999.
| REFERENCES |
|---|
|
|
|---|
chain and close linkage of IL15RA and IL2RA genes. J. Biol. Chem. 270,29862-29869
B. Curr. Biol. 6,1669-1676[Medline]
B activation. Cell 81,495-504[Medline]
B in preventing TNF-
-induced cell death. Science 274,782-784
B. Science 274,784-787
-induced apoptosis by NF-
B. Science 274,787-789
B activation prevents cell death. Cell 87,565-576[Medline]
B signal. Immunity 8,297-303[Medline]
B and I
B proteins: new discoveries and insights. Annu. Rev. Immunol. 14,649-681[Medline]
B is required for inhibition of NF-
B by I
B. Genes Dev 3,1689-1698
B or die?. Curr. Biol. 7,R94-R96[Medline]
B and REL proteins: evolutionary conserved mediators of immune responses. Annu. Rev. Immunol. 16,225-260[Medline]
This article has been cited by other articles:
![]() |
E. H. Duitman, Z. Orinska, E. Bulanova, R. Paus, and S. Bulfone-Paus How a Cytokine Is Chaperoned through the Secretory Pathway by Complexing with Its Own Receptor: Lessons from Interleukin-15 (IL-15)/IL-15 Receptor {alpha} Mol. Cell. Biol., August 1, 2008; 28(15): 4851 - 4861. [Abstract] [Full Text] [PDF] |
||||
![]() |